HK40035897B - Compositions relating to a mutant clostridium difficile toxin and methods thereof - Google Patents

Description

Compositions and methods involving mutant Clostridium difficile toxins

This application is a divisional application of Chinese Patent Application No. 201611125122.9, filed on April 20, 2012, entitled “Compositions and methods relating to mutant clostridium difficile toxins”.

Cross-reference of related applications

This application claims priority to U.S. Provisional Patent Application 61/478,474, filed April 22, 2011, and U.S. Provisional Patent Application 61/478,899, filed April 25, 2011. The entire contents of the foregoing applications are incorporated herein by reference in their entirety.

Invention Field

This invention relates to compositions and methods involving mutant Clostridium difficile toxins.

Background of the Invention

Clostridium difficile is a Gram-positive anaerobic bacterium associated with gastrointestinal diseases in humans. If the native gut microbiota is reduced due to antibody treatment, Clostridium difficile colonization typically occurs in the colon. Infection can lead to antibiotic-associated diarrhea and sometimes pseudomembranous colitis through the secretion of glucosylated toxins, toxin A and toxin B (308 and 270 kDa, respectively, which are the main virulence factors of Clostridium difficile).

Toxin A and toxin B are encoded by genes tcdA and tcdB, respectively, within a 19kb pathogenicity locus (PaLoc). In non-pathogenic strains of Clostridium difficile, this locus is replaced by an additional 115-base-pair sequence.

Both toxin A and toxin B are potent cytotoxins. These proteins are homologous glucosyltransferases that inactivate small GTPases of the Rho/Rac/Ras family. The resulting disruption of signal transduction leads to cell-cell junction dysfunction, actin cytoskeleton dysregulation, and/or loss of apoptosis, resulting in deep secretory diarrhea associated with Clostridium difficile infection (CDI).

Over the past decade, the frequency and severity of Clostridium difficile outbreaks in hospitals, nursing homes, and other long-term care facilities have increased dramatically. Key factors contributing to this increase include the emergence of highly virulent pathogens, increased antibiotic use, improved detection methods, and increased exposure to airborne spores in healthcare facilities.

Metronidazole and vancomycin are currently the standard of care for antibiotic treatment of Clostridium difficile-associated disease (CDAD). However, approximately 20% of patients receiving this treatment experience a recurrence of infection after stage I CDI, and up to 50% of those patients experience additional recurrences. Treating recurrences is a significant challenge, with most recurrences typically occurring within one month of the previous stage.

Therefore, there is a need for immunogenic and/or therapeutic compositions of Clostridium difficile and methods thereof.

Invention Overview

These and other objectives are provided herein.

In one aspect, the present invention relates to an immunogenic composition comprising a mutant Clostridium difficile toxin A. The mutant Clostridium difficile toxin A, relative to the corresponding wild-type Clostridium difficile toxin A, comprises a glucosyltransferase domain having at least one mutation and a cysteine protease domain having at least one mutation. In one embodiment, at least one amino acid of the mutant Clostridium difficile toxin A is chemically cross-linked.

In one aspect, the present invention relates to an isolated polypeptide comprising the amino acid sequence shown in SEQ ID NO:4, wherein the methionine residue at position 1 is optionally absent, and wherein the polypeptide comprises at least one amino acid side chain chemically modified with 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) and N-hydroxysuccinimide (NHS).

In one embodiment, at least one amino acid of the mutant Clostridium difficile toxin is chemically cross-linked.

In one embodiment, the at least one amino acid is chemically cross-linked as follows: formaldehyde, 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC), N-hydroxysuccinate/salt, or a combination of EDC and NHS.

In one embodiment, the immunogenic composition is recognized by a respective antitoxin neutralizing antibody or its binding fragment.

In one embodiment, the immunogenic composition exhibits reduced cytotoxicity relative to the corresponding wild-type Clostridium difficile toxin.

In another aspect, the present invention relates to an immunogenic composition comprising a mutant Clostridium difficile toxin A, said mutant Clostridium difficile toxin A comprising a glucosyltransferase domain having SEQ ID NO:29 having amino acid substitutions at positions 285 and 287 (relative to the corresponding wild-type Clostridium difficile toxin A), and a cysteine protease domain having SEQ ID NO:32 having an amino acid substitution at position 158 (relative to the corresponding wild-type Clostridium difficile toxin A), wherein at least one amino acid of said mutant Clostridium difficile toxin A is chemically cross-linked.

In another aspect, the present invention relates to an immunogenic composition comprising a mutant Clostridium difficile toxin A, said mutant Clostridium difficile toxin A comprising SEQ ID NO:4, wherein at least one amino acid of said mutant Clostridium difficile toxin A is chemically cross-linked.

In another aspect, the present invention relates to immunogenic compositions comprising SEQ ID NO:4, SEQ ID NO:5, SEQ ID NO:6, SEQ ID NO:7, or SEQ ID NO:8.

In one aspect, the present invention relates to immunogenic compositions comprising mutant Clostridium difficile toxin B. The mutant Clostridium difficile toxin B, relative to the corresponding wild-type wild-type Clostridium difficile toxin B, comprises a glucose-transferase domain having at least one mutation and a cysteine protease domain having at least one mutation.

In another aspect, the present invention relates to isolated polypeptides comprising the amino acid sequence shown in SEQ ID NO:6, wherein the methionine residue at position 1 is optionally absent, and wherein said polypeptide comprises at least one amino acid side chain chemically modified with 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) and N-hydroxysuccinimide (NHS).

In another aspect, the present invention relates to an immunogenic composition comprising a mutant Clostridium difficile toxin B, said mutant Clostridium difficile toxin B comprising a glucosyltransferase domain having SEQ ID NO:31 having amino acid substitutions at positions 286 and 288 (relative to the corresponding wild-type Clostridium difficile toxin B), and a cysteine protease domain having SEQ ID NO:33 having an amino acid substitution at position 155 (relative to the corresponding wild-type Clostridium difficile toxin B), wherein at least one amino acid of said mutant Clostridium difficile toxin B is chemically cross-linked.

In another aspect, the present invention relates to an immunogenic composition comprising a mutant Clostridium difficile toxin B, said mutant Clostridium difficile toxin B comprising SEQ ID NO:6, wherein at least one amino acid of said mutant Clostridium difficile toxin B is chemically cross-linked.

In one aspect, the present invention relates to an immunogenic composition comprising mutant Clostridium difficile toxin A and mutant Clostridium difficile toxin B, wherein mutant Clostridium difficile toxin A comprises SEQ ID NO:4 and mutant Clostridium difficile toxin B comprises SEQ ID NO:6, wherein at least one amino acid of each of the mutant Clostridium difficile toxins is chemically cross-linked.

In another aspect, the present invention relates to recombinant cells comprising polynucleotides or their progeny, said polynucleotides encoding any of the aforementioned mutant Clostridium difficile toxins, wherein said cells lack endogenous polynucleotides encoding the toxins.

In another respect, the present invention relates to antibodies or antibody-binding fragments thereof, which are specific to immunogenic compositions comprising mutant Clostridium difficile toxin.

In one aspect, the present invention relates to a method for treating Clostridium difficile infection in mammals. The method comprises administering an immunogenic composition to the mammal comprising mutant Clostridium difficile toxin A and mutant Clostridium difficile toxin B, wherein mutant Clostridium difficile toxin A comprises SEQ ID NO:4, and mutant Clostridium difficile toxin B comprises SEQ ID NO:6, wherein at least one amino acid of each of the mutant Clostridium difficile toxins is cross-linked with formaldehyde.

In one aspect, the present invention relates to a method for treating Clostridium difficile infection in mammals. The method comprises administering an immunogenic composition to the mammal comprising mutant Clostridium difficile toxin A and mutant Clostridium difficile toxin B, wherein mutant Clostridium difficile toxin A comprises SEQ ID NO:4, and mutant Clostridium difficile toxin B comprises SEQ ID NO:6, wherein at least one amino acid of each of the mutant Clostridium difficile toxins is cross-linked with 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide and/or N-hydroxysuccinimide (NHS).

In one aspect, the present invention relates to a method for inducing an immune response to Clostridium difficile infection in mammals. The method comprises administering an immunogenic composition to a mammal comprising mutant Clostridium difficile toxin A and mutant Clostridium difficile toxin B, wherein mutant Clostridium difficile toxin A comprises SEQ ID NO:4, and mutant Clostridium difficile toxin B comprises SEQ ID NO:6, wherein at least one amino acid of each of the mutant Clostridium difficile toxins is cross-linked with formaldehyde.

In one aspect, the present invention relates to a method for inducing an immune response to Clostridium difficile infection in mammals. The method comprises administering to a mammal an immunogenic composition comprising mutant Clostridium difficile toxin A and mutant Clostridium difficile toxin B, wherein mutant Clostridium difficile toxin A comprises SEQ ID NO:4, and mutant Clostridium difficile toxin B comprises SEQ ID NO:6, wherein at least one amino acid of each of the mutant Clostridium difficile toxins is cross-linked with 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide and/or N-hydroxysuccinimide (NHS).

In one implementation, the treatment method or the method of inducing an immune response is performed in the mammal in need.

In one implementation, the treatment method or the method of inducing an immune response includes a mammal with a recurrent Clostridium difficile infection.

In one embodiment, the method of treatment or the method of inducing an immune response includes parenteral administration of the composition.

In one embodiment, the method of treatment or the method of inducing an immune response includes an immunogenic composition further comprising an adjuvant.

In one embodiment, the adjuvant comprises aluminum hydroxide gel and CpG oligonucleotide. In another embodiment, the adjuvant comprises ISCOMATRIX.

In one embodiment, the isolated polypeptide comprises at least one aspartic acid residue side chain of the polypeptide that has been chemically cross-linked with glycine or at least one glutamic acid residue side chain of the polypeptide.

In one embodiment, the isolated polypeptide includes at least one crosslink between an aspartic acid residue side chain and a lysine residue side chain of the polypeptide; and at least one crosslink between a glutamic acid residue side chain and a lysine residue side chain of the polypeptide.

In one embodiment, the isolated polypeptide includes a β-alanine moiety attached to at least one lysine residue side chain of the polypeptide.

In one embodiment, the isolated polypeptide includes an aspartic acid residue side chain attached to the polypeptide or a glycine moiety attached to a glutamic acid residue side chain of the polypeptide.

In one embodiment, the isolated polypeptide comprises the amino acid sequence shown in SEQ ID NO:4, wherein the methionine residue at position 1 is optionally absent, and wherein at least one lysine residue of the polypeptide is linked to a β-alanine moiety via a side chain.

In one embodiment, the isolated polypeptide comprises the amino acid sequence shown in SEQ ID NO:6, wherein the methionine residue at position 1 is optionally absent, and wherein at least one lysine residue of the polypeptide is linked to a β-alanine moiety via a side chain.

In one embodiment, the isolated polypeptide includes a side chain attached to an aspartic acid residue or a second lysine residue attached to a glutamic acid residue side chain.

In one embodiment, the isolated polypeptide includes an aspartic acid residue side chain or a glutamic acid residue side chain of the polypeptide linked to the glycine moiety.

In one embodiment, the isolated polypeptide has an EC ₅₀ of at least about 100 μg/ml.

In one aspect, the immunogenic composition comprises an isolated polypeptide having the amino acid sequence shown in SEQ ID NO:4 (wherein the methionine residue at position 1 is optionally absent), and an isolated polypeptide having the amino acid sequence shown in SEQ ID NO:6 (wherein the methionine residue at position 1 is optionally absent), and wherein said polypeptide has at least one amino acid side chain chemically modified with 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) and N-hydroxysuccinimide (NHS).

In one embodiment, the polypeptide includes at least one of the following: a) at least one β-alanine moiety attached to the lysine side chain of the polypeptide; b) at least one crosslink between an aspartic acid residue side chain and the lysine residue side chain of the polypeptide; and c) at least one crosslink between a glutamic acid residue side chain and the lysine residue side chain of the polypeptide.

In one embodiment, the isolated polypeptide has an EC ₅₀ of at least about 100 μg/ml.

In one aspect, the immunogenic composition comprises an isolated polypeptide having the amino acid sequence shown in SEQ ID NO:4 (wherein the methionine residue at position 1 is optionally absent), and an isolated polypeptide having the amino acid sequence shown in SEQ ID NO:6 (wherein the methionine residue at position 1 is optionally absent), and a) wherein at least one lysine residue of SEQ ID NO:4 is linked to a β-alanine moiety by a side chain, and b) wherein at least one lysine residue of SEQ ID NO:6 is linked to a β-alanine moiety by a side chain.

In one embodiment, the immunogenic composition comprises a second lysine residue side chain of SEQ ID NO:4 linked to an aspartic acid residue side chain or linked to a glutamate residue side chain, and wherein the second lysine residue of SEQ ID NO:6 is linked to an aspartic acid residue side chain or linked to a glutamate residue side chain.

In one embodiment, the immunogenic composition comprises an aspartic acid residue side chain or a glutamic acid residue side chain of a polypeptide having the amino acid sequence shown in SEQ ID NO:4 (wherein the methionine residue at position 1 is optionally absent) that is linked to a glycine moiety.

In one embodiment, the immunogenic composition comprises an aspartic acid residue side chain or a glutamic acid residue side chain of a polypeptide having the amino acid sequence shown in SEQ ID NO:6 (wherein the methionine residue at position 1 is optionally absent) that is linked to a glycine moiety.

In one embodiment, the isolated polypeptide has an EC ₅₀ of at least about 100 μg/ml.

In one aspect, the immunogenic composition comprises: isolated polypeptides having the amino acid sequence of SEQ ID NO:84 and isolated polypeptides having the amino acid sequence of SEQ ID NO:86, wherein each polypeptide comprises: a) at least one crosslink between an aspartic acid residue side chain and a lysine residue side chain of the polypeptide; b) at least one crosslink between a glutamic acid residue side chain and a lysine residue side chain of the polypeptide; c) a β-alanine moiety attached to at least one lysine residue side chain of the polypeptide; and d) a glycine moiety attached to at least one aspartic acid residue side chain or at least one glutamic acid residue side chain of the polypeptide.

Brief description of the attached diagram

Figure 1: Sequence alignment of wild-type Clostridium difficile toxin A from strains 630, VPI10463, R20291, and CD196 and mutant toxin A with SEQ ID NO:4, using CLUSTALW alignment with default parameters.

Figure 2: Sequence alignment of wild-type Clostridium difficile toxin A from strains 630, VPI10463, R20291, and CD196 and mutant toxin B with SEQ ID NO:6, using CLUSTALW alignment with default parameters.

Figure 3: A diagram showing the identification of wild-type toxin-negative Clostridium difficile strains. For toxin A, the culture media of 13 Clostridium difficile strains were tested by ELISA. As shown, 7 strains expressed toxin A while 6 strains did not (strains 1351, 3232, 7322, 5036, 4811, and VPI 11186).

Figures 4A and B: SDS-PAGE results show that triple mutant A (SEQ ID NO:4), double mutant B (SEQ ID NO:5), and triple mutant B (SEQ ID NO:6) do not glucosylate Rac1 or RhoA GTPase in the in vitro glucosylation assay using UDP- ¹⁴ C-glucose; while wild-type toxin B glucosylates Rac1 at concentrations from 10 μg to 1 ng.

Figure 5: Western blot showing the elimination of cysteine protease activity in mutant toxins A and B (SEQ ID NO: 4 and 6, respectively) compared to the cleavage fragments of wild-type toxins A and B (SEQ ID NO: 1 and 2, respectively). See Example 13.

Figure 6: A graph showing the residual cytotoxicity of triple mutant toxins A and B (SEQ ID NO: 4 and 6, respectively) at high concentrations (e.g., about 100 μg/ml) in an in vitro cytotoxicity test conducted in IMR-90.

Figure 7: A similar graph showing the _EC50 values of triple mutant toxin B (SEQ ID NO:6) and heptameric mutant toxin B (SEQ ID NO:8).

Figure 8: A graph representing the results of the in vitro cytotoxicity assay, where ATP levels (RLU) are plotted against the elevated concentrations of triple mutant TcdA (SEQ ID NO:4) (top part) and triple mutant TcdB (SEQ ID NO:6) (bottom part). The residual cytotoxicity of mutant toxins A and B can be completely eliminated with neutralizing antibodies specific to mutant toxin A (top part - pAb A and mAb A3-25+A60-22) and specific to mutant toxin B (bottom part - pAb B).

Figure 9: IMR-90 cell morphology 72 hours after treatment. Part A shows control cells simulating the treatment. Part B shows cell morphology after treatment with formalin-inactivated mutant TcdB (SEQ ID NO: 6). Part C shows cell morphology after treatment with EDC-inactivated mutant TcdB (SEQ ID NO: 6). Part D shows cell morphology after treatment with wild-type toxin B (SEQ ID NO: 2). Part E shows cell morphology after treatment with triple mutant TcdB (SEQ ID NO: 6). Similar results were observed in TcdA treatment.

Figure 10: A graph showing the neutralizing antibody titers described in Example 25 (Research muCdiff 2010-06).

Figure 11: A graph showing the neutralizing antibody titer described in Example 26 (Research muCdiff 2010-07).

Figure 12: A graph showing the neutralizing antibody response against toxins A and B in hamsters after four immunizations, as described in Example 27 (Study on Clostridium difficile ham 2010-02).

Figure 13: A graph showing the neutralizing antibody response in hamsters after inoculation with chemically inactivated genetic mutant toxins and List Biological toxins, as described in Example 27 (Study on Clostridium difficile ham 2010-02).

Figure 14: Survival curves of hamsters in the three immunized groups compared with the non-immunized control, as described in Example 28 (Study on Clostridium difficile ham 2010-02, continued).

Figure 15: A graph showing the relative neutralizing antibody responses in hamsters to different formulations of Clostridium difficile mutant toxin (Study on Clostridium difficile 2010-03), as described in Example 29.

Figures 16A-B: Graphs showing strong relative neutralizing antibody responses in cynomolgus monkeys to chemically inactivated genetically modified toxins A and B (SEQ ID NO: 4 and 6, respectively), as described in Example 30.

Figure 17: Amino acid sequences of the variable regions of the light chain (VL) and heavy chain (HL) of A3-25 mAb IgE. Signal peptide - highlighted; CDR - italicized and underlined; constant region - bold and underlined (complete sequence not shown).

Figure 18: A graph showing the titration of individual toxin A monoclonal antibodies in a toxin neutralization assay, where ATP levels (quantified by relative optical units - RLU) were used as an indicator of cell viability. mAbs A80-29, A65-33, A60-22, and A3-25 showed increasing concentration-dependent neutralization of toxin A compared to the toxin control (4 x _EC50 ), but not to the positive rabbit antitoxin A control. mAbs A50-10, A56-33, and A58-46 did not neutralize toxin A. The cell-only control was 1–1.5 x ^10⁶ RLU.

Figure 19: Grouping of 8 epitopes of toxin B mAb using BiaCore.

Figures 20A-C: Synergistic neutralizing activity of combinations of toxin A mAbs: Adding different dilutions of neutralizing antibodies A60-22, A65-33, and A80-29 to increasing concentrations of A3-25 mAb synergistically enhances the neutralization of toxin A (regardless of dilution). The RLU (<0.3 x ^10⁶ ) of the control with only toxin A (4 x _EC⁵⁰ ) is shown, while the RLU of the cell-only control is 2–2.5 x ^10⁶ , as shown in Figures 20B and 20C.

Figure 21: Synergistic neutralizing activity of combinations of toxin B mAbs: Neutralization of toxin B by mAbs 8-26, B60-2 and B59-3 is shown in Figure 21A. The neutralization of toxin B is synergistically enhanced when a dilution of B8-26 is combined with that of B59-3 (Figure 21B).

Figure 22: Western blot showing that Rac1 GTPase expression decreased from 24 to 96 hours in the extract of the genetically modified toxin B (SEQ ID NO: 6), but not in the extract treated with wild-type toxin B (SEQ ID NO: 2). The blot also showed that Rac1 was glycated in the toxin B-treated extract, but not in the extract treated with the genetically modified toxin B.

Figures 23A-K: Graphs representing the results of in vitro cytotoxicity assays, where ATP levels (RLU) are plotted against the following: increased concentrations of *Clostridium difficile* culture medium and hamster serum bank (■); crude toxins (culture harvest) from each strain and hamster serum bank (●); purified toxins (commercially available toxins from List Biologicals) and hamster serum bank (▲); crude toxin control; and purified toxin (◆), control. Toxins from each strain were added to cells at 4 x _EC50 values. Figure 23 shows the results for mutants TcdA (SEQ ID NO:4) and TcdB (SEQ ID NO:4). An immunogenic composition of NO:6 (in which the mutant toxin is inactivated by EDC, for example, according to Example 29 described herein, Table 15) induced neutralizing antibodies that exhibited neutralizing activity against toxins from at least 16 different CDC strains of Clostridium difficile (compared to controls containing only the respective toxin): 2007886 (Fig. 23A); 2006017 (Fig. 23B); 2007070 (Fig. 23C); 2007302 (Fig. 23D); 2007838 (Fig. 23E); 2007886 (Fig. 23F); 2009292 (Fig. 23G); 2004013 (Fig. 23H); 2009141 (Fig. 23I); 2005022 (Fig. 23J); 2006376 (Fig. 23K).

Figure 24 illustrates exemplary EDC/NHS inactivation of the mutant Clostridium difficile toxin, resulting in at least three possible types of modification: cross-linking, glycine adduct, and β-alanine adduct. Part A illustrates cross-linking. The carboxylic acid residues of the triple mutant toxin are activated via addition to EDC and NHS. The activated ester reacts with a primary amine to form a stable amide bond, leading to intramolecular and intermolecular cross-linking. Part B illustrates the formation of the glycine adduct. Following inactivation, the residual activated ester is quenched by adding glycine to form a stable amide bond. Part C illustrates the formation of the β-alanine adduct. Three moles of NHS react with one mole of EDC to form activated β-alanine. This then reacts with a primary amine to form a stable amide bond.

Figure 25 illustrates exemplary mutant Clostridium difficile toxin EDC/NHS inactivation, resulting in at least one of the following types of modifications: (A) crosslinking, (B) glycine adduct, and (C) β-alanine adduct.

Sequence Summary

SEQ ID NO:1 shows the amino acid sequence of wild-type Clostridium difficile 630 toxin A (TcdA).

SEQ ID NO:2 shows the amino acid sequence of wild-type Clostridium difficile 630 toxin B (TcdB).

SEQ ID NO:3 shows the amino acid sequence of the mutant TcdA, which has mutations at positions 285 and 287 compared to SEQ ID NO:1.

SEQ ID NO:4 shows the amino acid sequence of the mutant TcdA, which has mutations at positions 285, 287, and 700 compared to SEQ ID NO:1.

SEQ ID NO:5 shows the amino acid sequence of the mutant TcdB, which has mutations at positions 286 and 288 compared to SEQ ID NO:2.

SEQ ID NO:6 shows the amino acid sequence of the mutant TcdB, which has mutations at positions 286, 288, and 698 compared to SEQ ID NO:2.

SEQ ID NO:7 shows the amino acid sequence of the mutant TcdA with mutations at positions 269, 272, 285, 287, 460, 462 and 700 compared to SEQ ID NO:1.

SEQ ID NO:8 shows the amino acid sequence of the mutant TcdB, which has mutations at positions 270, 273, 286, 288, 461, 463 and 698 compared to SEQ ID NO:2.

SEQ ID NO:9 shows the DNA sequence encoding wild-type Clostridium difficile 630 toxin A (TcdA).

SEQ ID NO:10 shows the DNA sequence encoding wild-type Clostridium difficile 630 toxin B (TcdB).

SEQ ID NO:11 shows the DNA sequence encoding SEQ ID NO:3.

SEQ ID NO:12 shows the DNA sequence encoding SEQ ID NO:4.

SEQ ID NO:13 shows the DNA sequence encoding SEQ ID NO:5.

SEQ ID NO:14 shows the DNA sequence encoding SEQ ID NO:6.

SEQ ID NO:15 shows the amino acid sequence of wild-type Clostridium difficile R20291TcdA.

SEQ ID NO:16 shows the DNA sequence encoding SEQ ID NO:15.

SEQ ID NO:17 shows the amino acid sequence of wild-type Clostridium difficile CD196TcdA.

SEQ ID NO:18 shows the DNA sequence encoding SEQ ID NO:17.

SEQ ID NO:19 shows the amino acid sequence of wild-type Clostridium difficile VPI10463TcdA.

SEQ ID NO:20 shows the DNA sequence encoding SEQ ID NO:19.

SEQ ID NO:21 shows the amino acid sequence of wild-type Clostridium difficile R20291TcdB.

SEQ ID NO:22 shows the DNA sequence encoding SEQ ID NO:21.

SEQ ID NO:23 shows the amino acid sequence of wild-type Clostridium difficile CD196TcdB.

SEQ ID NO:24 shows the DNA sequence encoding SEQ ID NO:23.

SEQ ID NO:25 shows the amino acid sequence of wild-type Clostridium difficile VPI10463TcdB.

SEQ ID NO:26 shows the DNA sequence encoding SEQ ID NO:25.

SEQ ID NO:27 shows the DNA sequence of the pathogenic locus of wild-type Clostridium difficile VPI10463.

SEQ ID NO:28 shows the amino acid sequence of residues 101-293 of SEQ ID NO:1.

SEQ ID NO:29 shows the amino acid sequence of residues 1-542 of SEQ ID NO:1.

SEQ ID NO:30 shows the amino acid sequence of residues 101-293 of SEQ ID NO:2.

SEQ ID NO:31 shows the amino acid sequence of residues 1-543 of SEQ ID NO:2.

SEQ ID NO:32 shows the amino acid sequence of residues 543-809 of SEQ ID NO:1.

SEQ ID NO:33 shows the amino acid sequence of residues 544-767 of SEQ ID NO:2.

SEQ ID NO:34 shows the amino acid sequence of the mutant TcdA, wherein residues 101, 269, 272, 285, 287, 460, 462, 541, 542, 543, 589, 655, and 700 can be any amino acid.

SEQ ID NO:35 shows the amino acid sequence of the mutant TcdB, where 102, 270, 273, 286, 288, 384, 461, 463, 520, 543, 544, 587, 600, 653, 698 and 751 can be any amino acid.

SEQ ID NO:36 shows the amino acid sequence of the variable light chain of the neutralizing antibody against Clostridium difficile TcdA (A3-25mAb).

SEQ ID NO:37 shows the amino acid sequence of the variable heavy chain of the neutralizing antibody against Clostridium difficile TcdA (A3-25mAb).

SEQ ID NO:38 shows the amino acid sequence of CDR1, a variable light chain of a neutralizing antibody against Clostridium difficile TcdA (A3-25mAb).

SEQ ID NO:39 shows the amino acid sequence of CDR2, the variable light chain of the neutralizing antibody against Clostridium difficile TcdA (A3-25mAb).

SEQ ID NO:40 shows the amino acid sequence of CDR3, a variable light chain of a neutralizing antibody against Clostridium difficile TcdA (A3-25mAb).

SEQ ID NO:41 shows the amino acid sequence of CDR1 of the variable heavy chain of the neutralizing antibody against Clostridium difficile TcdA (A3-25mAb).

SEQ ID NO:42 shows the amino acid sequence of CDR2 of the variable heavy chain of the neutralizing antibody against Clostridium difficile TcdA (A3-25mAb).

SEQ ID NO:43 shows the amino acid sequence of CDR3 of the variable heavy chain of the neutralizing antibody against Clostridium difficile TcdA (A3-25mAb).

SEQ ID NO:44 shows the DNA sequence encoding SEQ ID NO:3.

SEQ ID NO:45 shows the DNA sequence encoding SEQ ID NO:4.

SEQ ID NO:46 shows the DNA sequence encoding SEQ ID NO:5.

SEQ ID NO:47 shows the DNA sequence encoding SEQ ID NO:6.

SEQ ID NO:48 shows the nucleotide sequence of the immunostimulatory oligonucleotide ODN CpG 24555.

SEQ ID NO:49 shows the amino acid sequence of the variable heavy chain of the Clostridium difficile TcdB neutralizing antibody (B8-26mAb). SEQ ID NO:50 shows the amino acid sequence of the signal peptide of the variable heavy chain of the Clostridium difficile TcdB neutralizing antibody (B8-26mAb).

SEQ ID NO:51 shows the amino acid sequence of CDR1 of the variable heavy chain of the Clostridium difficile TcdB neutralizing antibody (B8-26mAb).

SEQ ID NO:52 shows the amino acid sequence of CDR2 of the variable heavy chain of the Clostridium difficile TcdB neutralizing antibody (B8-26mAb).

SEQ ID NO:53 shows the amino acid sequence of CDR3 of the variable heavy chain of the Clostridium TcdB neutralizing antibody (B8-26mAb).

SEQ ID NO:54 shows the amino acid sequence of the constant region of the variable heavy chain of the Clostridium difficile TcdB neutralizing antibody (B8-26mAb).

SEQ ID NO:55 shows the amino acid sequence of the variable light chain of the Clostridium difficile TcdB neutralizing antibody (B8-26mAb).

SEQ ID NO:56 shows the amino acid sequence of the signal peptide of the variable light chain of the Clostridium difficile TcdB neutralizing antibody (B8-26mAb).

SEQ ID NO:57 shows the amino acid sequence of CDR1 of the variable light chain of the Clostridium difficile TcdB neutralizing antibody (B8-26mAb).

SEQ ID NO:58 shows the amino acid sequence of CDR2 of the variable light chain of the Clostridium difficile TcdB neutralizing antibody (B8-26mAb).

SEQ ID NO:59 shows the amino acid sequence of CDR3 of the variable light chain of the Clostridium difficile TcdB neutralizing antibody (B8-26mAb).

SEQ ID NO:60 shows the amino acid sequence of the variable heavy chain of the Clostridium difficile TcdB neutralizing antibody (B59-3mAb).

SEQ ID NO:61 shows the amino acid sequence of the signal peptide of the variable heavy chain of the Clostridium difficile TcdB neutralizing antibody (B59-3mAb).

SEQ ID NO:62 shows the amino acid sequence of CDR1 of the variable heavy chain of the Clostridium difficile TcdB neutralizing antibody (B59-3mAb).

SEQ ID NO:63 shows the amino acid sequence of CDR2 of the variable heavy chain of the Clostridium difficile TcdB neutralizing antibody (B59-3mAb).

SEQ ID NO:64 shows the amino acid sequence of CDR3 of the variable heavy chain of the Clostridium difficile TcdB neutralizing antibody (B59-3mAb).

SEQ ID NO:65 shows the amino acid sequence of the constant region of the variable heavy chain of the Clostridium difficile TcdB neutralizing antibody (B59-3mAb).

SEQ ID NO:66 shows the amino acid sequence of the variable light chain of the Clostridium difficile TcdB neutralizing antibody (B59-3mAb).

SEQ ID NO:67 shows the amino acid sequence of the signal peptide of the variable light chain of the Clostridium difficile TcdB neutralizing antibody (B59-3mAb).

SEQ ID NO:68 shows the amino acid sequence of CDR1 of the variable light chain of the Clostridium difficile TcdB neutralizing antibody (B59-3mAb).

SEQ ID NO:69 shows the amino acid sequence of CDR2 of the variable light chain of the Clostridium difficile TcdB neutralizing antibody (B59-3mAb).

SEQ ID NO:70 shows the amino acid sequence of CDR3 of the variable light chain of the Clostridium difficile TcdB neutralizing antibody (B59-3mAb).

SEQ ID NO:71 shows the amino acid sequence of the variable heavy chain of the Clostridium difficile TcdB neutralizing antibody (B9-30mAb).

SEQ ID NO:72 shows the amino acid sequence of the signal peptide of the variable heavy chain of the Clostridium difficile TcdB neutralizing antibody (B9-30mAb).

SEQ ID NO:73 shows the amino acid sequence of CDR1 of the variable heavy chain of the Clostridium difficile TcdB neutralizing antibody (B9-30mAb).

SEQ ID NO:74 shows the amino acid sequence of CDR2 of the variable heavy chain of the Clostridium difficile TcdB neutralizing antibody (B9-30mAb).

SEQ ID NO:75 shows the amino acid sequence of CDR3 of the variable heavy chain of the Clostridium difficile TcdB neutralizing antibody (B9-30mAb).

SEQ ID NO:76 shows the amino acid sequence of the constant region of the variable heavy chain of the Clostridium difficile TcdB neutralizing antibody (B9-30mAb).

SEQ ID NO:77 shows the amino acid sequence of the variable light chain of the Clostridium difficile TcdB neutralizing antibody (B9-30mAb).

SEQ ID NO:78 shows the amino acid sequence of the signal peptide of the variable light chain of the Clostridium difficile TcdB neutralizing antibody (B9-30mAb).

SEQ ID NO:79 shows the amino acid sequence of CDR1 of the variable light chain of the Clostridium difficile TcdB neutralizing antibody (B9-30mAb).

SEQ ID NO:80 shows the amino acid sequence of CDR2 of the variable light chain of the Clostridium difficile TcdB neutralizing antibody (B9-30mAb).

SEQ ID NO:81 shows the amino acid sequence of CDR3 of the variable light chain of the Clostridium difficile TcdB neutralizing antibody (B9-30mAb).

SEQ ID NO:82 shows the amino acid sequence of the mutant TcdB, wherein the residues at positions 102, 270, 273, 286, 288, 384, 461, 463, 520, 543, 544, 587, 600, 653, 698 and 751 can be any amino acid.

SEQ ID NO:83 shows the amino acid sequence of mutant TcdA with mutations at positions 269, 272, 285, 287, 460, 462 and 700 compared to SEQ ID NO:1, wherein methionine at position 1 is missing.

SEQ ID NO:84 shows the amino acid sequence of mutant Clostridium difficile toxin A with mutations at positions 285, 287 and 700 compared to SEQ ID NO:1, wherein methionine is missing at position 1.

SEQ ID NO:85 shows the amino acid sequence of mutant Clostridium difficile toxin B with mutations at positions 270, 273, 286, 288, 461, 463 and 698 compared to SEQ ID NO:2, wherein methionine is missing at position 1.

SEQ ID NO:86 shows the amino acid sequence of mutant Clostridium difficile toxin B with mutations at positions 286, 288 and 698 compared to SEQ ID NO:2, wherein methionine at position 1 is missing.

SEQ ID NO:87 shows the amino acid sequence of wild-type Clostridium difficile 2004013TcdA.

SEQ ID NO:88 shows the amino acid sequence of wild-type Clostridium difficile 2004111TcdA.

SEQ ID NO:89 shows the amino acid sequence of wild-type Clostridium difficile 2004118TcdA.

SEQ ID NO:90 shows the amino acid sequence of wild-type Clostridium difficile 2004205TcdA.

SEQ ID NO:91 shows the amino acid sequence of wild-type Clostridium difficile 2004206TcdA.

SEQ ID NO:92 shows the amino acid sequence of wild-type Clostridium difficile 2005022TcdA.

SEQ ID NO:93 shows the amino acid sequence of wild-type Clostridium difficile 2005088TcdA.

SEQ ID NO:94 shows the amino acid sequence of wild-type Clostridium difficile 2005283TcdA.

SEQ ID NO:95 shows the amino acid sequence of wild-type Clostridium difficile 2005325TcdA.

SEQ ID NO:96 shows the amino acid sequence of wild-type Clostridium difficile 2005359TcdA.

SEQ ID NO:97 shows the amino acid sequence of wild-type Clostridium difficile 2006017TcdA.

SEQ ID NO:98 shows the amino acid sequence of wild-type Clostridium difficile 2007070TcdA.

SEQ ID NO:99 shows the amino acid sequence of wild-type Clostridium difficile 2007217TcdA.

SEQ ID NO:100 shows the amino acid sequence of wild-type Clostridium difficile 2007302TcdA.

SEQ ID NO:101 shows the amino acid sequence of wild-type Clostridium difficile 2007816TcdA.

SEQ ID NO:102 shows the amino acid sequence of wild-type Clostridium difficile 2007838TcdA.

SEQ ID NO:103 shows the amino acid sequence of wild-type Clostridium difficile 2007858TcdA.

SEQ ID NO:104 shows the amino acid sequence of wild-type Clostridium difficile 2007886TcdA.

SEQ ID NO:105 shows the amino acid sequence of wild-type Clostridium difficile 2008222TcdA.

SEQ ID NO:106 shows the amino acid sequence of wild-type Clostridium difficile 2009078TcdA.

SEQ ID NO:107 shows the amino acid sequence of wild-type Clostridium difficile 2009087TcdA.

SEQ ID NO:108 shows the amino acid sequence of wild-type Clostridium difficile 2009141TcdA.

SEQ ID NO:109 shows the amino acid sequence of wild-type Clostridium difficile 2009292TcdA.

SEQ ID NO:110 shows the amino acid sequence of wild-type Clostridium difficile 2004013TcdB.

SEQ ID NO:111 shows the amino acid sequence of wild-type Clostridium difficile 2004111TcdB.

SEQ ID NO:112 shows the amino acid sequence of wild-type Clostridium difficile 2004118TcdB.

SEQ ID NO:113 shows the amino acid sequence of wild-type Clostridium difficile 2004205TcdB.

SEQ ID NO:114 shows the amino acid sequence of wild-type Clostridium difficile 2004206TcdB.

SEQ ID NO:115 shows the amino acid sequence of wild-type Clostridium difficile 2005022TcdB.

SEQ ID NO:116 shows the amino acid sequence of wild-type Clostridium difficile 2005088TcdB.

SEQ ID NO:117 shows the amino acid sequence of wild-type Clostridium difficile 2005283TcdB.

SEQ ID NO:118 shows the amino acid sequence of wild-type Clostridium difficile 2005325TcdB.

SEQ ID NO:119 shows the amino acid sequence of wild-type Clostridium difficile 2005359TcdB.

SEQ ID NO:120 shows the amino acid sequence of wild-type Clostridium difficile 2006017TcdB.

SEQ ID NO:121 shows the amino acid sequence of wild-type Clostridium difficile 2006376TcdB.

SEQ ID NO:122 shows the amino acid sequence of wild-type Clostridium difficile 2007070TcdB.

SEQ ID NO:123 shows the amino acid sequence of wild-type Clostridium difficile 2007217TcdB.

SEQ ID NO:124 shows the amino acid sequence of wild-type Clostridium difficile 2007302TcdB.

SEQ ID NO:125 shows the amino acid sequence of wild-type Clostridium difficile 2007816TcdB.

SEQ ID NO:126 shows the amino acid sequence of wild-type Clostridium difficile 2007838TcdB.

SEQ ID NO:127 shows the amino acid sequence of wild-type Clostridium difficile 2007858TcdB.

SEQ ID NO:128 shows the amino acid sequence of wild-type Clostridium difficile 2007886TcdB.

SEQ ID NO:129 shows the amino acid sequence of wild-type Clostridium difficile 2008222TcdB.

SEQ ID NO:130 shows the amino acid sequence of wild-type Clostridium difficile 2009078TcdB.

SEQ ID NO:131 shows the amino acid sequence of wild-type Clostridium difficile 2009087TcdB.

SEQ ID NO:132 shows the amino acid sequence of wild-type Clostridium difficile 2009141TcdB.

SEQ ID NO:133 shows the amino acid sequence of wild-type Clostridium difficile 2009292TcdB.

SEQ ID NO:134 shows the amino acid sequence of wild-type Clostridium difficile 014TcdA.

SEQ ID NO:135 shows the amino acid sequence of wild-type Clostridium difficile 015TcdA.

SEQ ID NO:136 shows the amino acid sequence of wild-type Clostridium difficile 020TcdA.

SEQ ID NO:137 shows the amino acid sequence of wild-type Clostridium difficile 023TcdA.

SEQ ID NO:138 shows the amino acid sequence of wild-type Clostridium difficile 027TcdA.

SEQ ID NO:139 shows the amino acid sequence of wild-type Clostridium difficile 029TcdA.

SEQ ID NO:140 shows the amino acid sequence of wild-type Clostridium difficile 046TcdA.

SEQ ID NO:141 shows the amino acid sequence of wild-type Clostridium difficile 014TcdB.

SEQ ID NO:142 shows the amino acid sequence of wild-type Clostridium difficile 015TcdB.

SEQ ID NO:143 shows the amino acid sequence of wild-type Clostridium difficile 020TcdB.

SEQ ID NO:144 shows the amino acid sequence of wild-type Clostridium difficile 023TcdB.

SEQ ID NO:145 shows the amino acid sequence of wild-type Clostridium difficile 027TcdB.

SEQ ID NO:146 shows the amino acid sequence of wild-type Clostridium difficile 029TcdB.

SEQ ID NO:147 shows the amino acid sequence of wild-type Clostridium difficile 046TcdB.

SEQ ID NO:148 shows the amino acid sequence of wild-type Clostridium difficile 001TcdA.

SEQ ID NO:149 shows the amino acid sequence of wild-type Clostridium difficile 002TcdA.

SEQ ID NO:150 shows the amino acid sequence of wild-type Clostridium difficile 003TcdA.

SEQ ID NO:151 shows the amino acid sequence of wild-type Clostridium difficile 004TcdA.

SEQ ID NO:152 shows the amino acid sequence of wild-type Clostridium difficile 070TcdA.

SEQ ID NO:153 shows the amino acid sequence of wild-type Clostridium difficile 075TcdA.

SEQ ID NO:154 shows the amino acid sequence of wild-type Clostridium difficile 077TcdA.

SEQ ID NO:155 shows the amino acid sequence of wild-type Clostridium difficile 081TcdA.

SEQ ID NO:156 shows the amino acid sequence of wild-type Clostridium difficile 117TcdA.

SEQ ID NO:157 shows the amino acid sequence of wild-type Clostridium difficile 131TcdA.

SEQ ID NO:158 shows the amino acid sequence of wild-type Clostridium difficile 001TcdB.

SEQ ID NO:159 shows the amino acid sequence of wild-type Clostridium difficile 002TcdB.

SEQ ID NO:160 shows the amino acid sequence of wild-type Clostridium difficile 003TcdB.

SEQ ID NO:161 shows the amino acid sequence of wild-type Clostridium difficile 004TcdB.

SEQ ID NO:162 shows the amino acid sequence of wild-type Clostridium difficile 070TcdB.

SEQ ID NO:163 shows the amino acid sequence of wild-type Clostridium difficile 075TcdB.

SEQ ID NO:164 shows the amino acid sequence of wild-type Clostridium difficile 077TcdB.

SEQ ID NO:165 shows the amino acid sequence of wild-type Clostridium difficile 081TcdB.

SEQ ID NO:166 shows the amino acid sequence of wild-type Clostridium difficile 117TcdB.

SEQ ID NO:167 shows the amino acid sequence of wild-type Clostridium difficile 131TcdB.

SEQ ID NO:168 shows the amino acid sequence of wild-type Clostridium difficile 053TcdA.

SEQ ID NO:169 shows the amino acid sequence of wild-type Clostridium difficile 078TcdA.

SEQ ID NO:170 shows the amino acid sequence of wild-type Clostridium difficile 087TcdA.

SEQ ID NO:171 shows the amino acid sequence of wild-type Clostridium difficile 095TcdA.

SEQ ID NO:172 shows the amino acid sequence of wild-type Clostridium difficile 126TcdA.

SEQ ID NO:173 shows the amino acid sequence of wild-type Clostridium difficile 053TcdB.

SEQ ID NO:174 shows the amino acid sequence of wild-type Clostridium difficile 078TcdB.

SEQ ID NO:175 shows the amino acid sequence of wild-type Clostridium difficile 087TcdB.

SEQ ID NO:176 shows the amino acid sequence of wild-type Clostridium difficile 095TcdB.

SEQ ID NO:177 shows the amino acid sequence of wild-type Clostridium difficile 126TcdB.

Invention Details

The inventors have surprisingly discovered (along with other matters) mutant clostridium difficile toxins A and B, and methods thereof. The mutants are characterized in that they are immunogenic and exhibit reduced cytotoxicity compared to their respective wild-type forms. The invention also relates to their immunogenic portions, their biological equivalents, and isolated polynucleotides comprising nucleic acid sequences encoding any of the aforementioned substances.

The immunogenic compositions described herein unexpectedly demonstrated the ability to elicit novel neutralizing antibodies against Clostridium difficile toxins, and these antibodies can confer active and/or passive protection against Clostridium difficile attack. The novel antibodies target various epitopes of toxins A and B. The inventors also discovered that combinations of at least two neutralizing monoclonal antibodies exhibit an unexpected synergistic effect in the in vitro neutralization of each toxin A and toxin B.

Compared to mammals not treated with the compositions of the present invention described herein, the compositions can be used to treat, prevent, reduce the risk of, reduce the occurrence of, reduce the severity of, and/or delay the onset of Clostridium difficile infection, Clostridium difficile-associated disease (CDAD), syndrome, condition, symptom, and/or its complications in mammals.

In addition, the inventors have discovered recombinant nonspore-producing Clostridium difficile cells capable of stably expressing mutant Clostridium difficile toxins A and B, as well as a method for producing them.

Immunogenic compositions

In one aspect, the present invention relates to immunogenic compositions comprising mutant Clostridium difficile toxins. The mutant Clostridium difficile toxins comprise an amino acid sequence having at least one mutation in the glucosyltransferase domain and at least one mutation in the cysteine protease domain relative to the corresponding wild-type Clostridium difficile toxin.

As used herein, the term "wild-type" refers to a form found in nature. For example, a wild-type polypeptide or polynucleotide sequence is a sequence present in an organism that can be isolated from a natural source and has not been intentionally modified by human intervention. This invention also relates to isolated polynucleotides comprising nucleic acid sequences encoding any of the aforementioned polypeptides. Furthermore, this invention relates to the use of any of the aforementioned compositions in the treatment and prevention of Clostridium difficile infection, Clostridium difficile-associated diseases, syndromes, disease conditions, symptoms, and/or complications in mammals, reducing their risk, severity, morbidity, and/or delaying their onset compared to mammals not treated with the compositions, and methods for preparing the compositions.

As used herein, “immunogenic composition” or “immunogen” refers to a composition that elicits an immune response in mammals to which the composition is administered.

"Immune response" refers to the development of a beneficial humoral (antibody-mediated) and/or cellular (antigen-specific T cell or its secreted products) response against Clostridium difficile toxin in recipient patients. Immune responses can be humoral, cellular, or both.

An immune response can be an active response induced by administration of an immunogenic composition or an immunogen. Alternatively, an immune response can be a passive response induced by administration of an antibody or sensitized T cells.

The presence of a humoral (antibody-mediated) immune response can be determined by, for example, cell-based assays known in the art such as neutralizing antibody assays, ELISA, etc.

Cellular immune responses are typically triggered by the presentation of polypeptide epitopes associated with class I or II MHC molecules to activate antigen-specific CD4+ helper T cells and/or CD8+ cytotoxic T cells. Responses may also involve the activation of monocytes, macrophages, NK cells, basophils, dendritic cells, astrocytes, microglia, eosinophils, or other components of the innate immune system. The presence of a cell-mediated immune response can be determined by proliferation assays (CD4+ T cells) or CTLs (cytotoxic T lymphocytes) known in the art.

In one embodiment, the immunogenic composition is a vaccine composition. As used herein, a "vaccine composition" is a composition that elicits an immune response in a mammal to which the composition is administered. The vaccine composition can protect an immunized mammal against subsequent attack by an immune substance or an immune cross-reactive substance. The protection may be complete or partial in terms of reduction of symptoms or infection compared to an unimmunized mammal under the same conditions.

The immunogenic compositions described herein are cross-reactive, meaning they possess the ability to elicit an effective immune response (e.g., humoral immune response) against toxins produced by another Clostridium difficile strain different from the strain from which the composition originates. For example, the immunogenic compositions described herein (e.g., derived from Clostridium difficile 630) can elicit cross-reactive antibodies capable of binding toxins produced by multiple strains of Clostridium difficile (e.g., toxins produced by Clostridium difficile R20291 and VPI10463). See, for example, Example 37. Cross-reactivity indicates the cross-protective potential of a bacterial immunogen, and vice versa.

As used herein, the term "cross-protection" refers to the ability of an immunogenic composition to induce an immune response that prevents or attenuates infections caused by different bacterial strains or species of the same genus. For example, the immunogenic compositions described herein (e.g., derived from Clostridium difficile 630) can induce an effective immune response in mammals to attenuate Clostridium difficile infections and/or attenuate Clostridium difficile diseases caused by strains other than 630 (e.g., Clostridium difficile R20291) in mammals.

Exemplary mammals from which an immunogenic composition or immunogen elicits an immune response include any mammal, such as mice, hamsters, primates, and humans. In a preferred embodiment, the immunogenic composition or immunogen elicits an immune response in a person administered the composition.

As mentioned above, toxin A (TcdA) and toxin B (TcdB) are homoglucosyltransferases that inactivate small GTPases of the Rho/Rac/Ras family. The effects of TcdA and TcdB on mammalian target cells depend on a multi-step mechanism involving receptor-mediated endocytosis, membrane translocation, autoproteolytic processing, and monosaccharidation of the GTPase. Many of these functional activities have been attributed to segregated regions in the primary sequence of the toxins, and the toxins have been imaged to show that these molecules are structurally similar.

The wild-type TcdA gene has approximately 8130 nucleotides and encodes a protein with a reduced molecular weight of approximately 308-kDa and approximately 2710 amino acids. As used herein, wild-type Clostridium difficile TcdA includes Clostridium difficile TcdA from any wild-type Clostridium difficile strain. Wild-type Clostridium difficile TcdA may include such a wild-type Clostridium difficile TcdA amino acid sequence that, when optimally aligned, for example by using the GAP or BESTFIT procedure with a predetermined gap weight, has at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, preferably about 98%, more preferably about 99%, or most preferably about 100% identity with SEQ ID NO:1 (full length).

In a preferred embodiment, the wild-type *Clostridium difficile* TcdA comprises the amino acid sequence shown in SEQ ID NO:1, which describes the wild-type amino acid sequence of TcdA from *Clostridium difficile* strain 630 (also disclosed in GenBank accession numbers YP_001087137.1 and/or CAJ67494.1). *Clostridium difficile* strain 630 is known in the art as a PCR-ribonucleic acid type 012 strain. SEQ ID NO:9 describes the wild-type gene of TcdA from *Clostridium difficile* strain 630, which is also disclosed in GenBank accession number NC_009089.1.

Another example of wild-type *Clostridium difficile* TcdA comprises the amino acid sequence shown in SEQ ID NO:15, which describes the wild-type amino acid sequence of TcdA from *Clostridium difficile* strain R20291 (also disclosed in GenBank accession number YP_003217088.1). *Clostridium difficile* strain R20291 is known in the art as a highly virulent strain and a PCR-ribonucleic acid type 027 strain. The amino acid sequence of TcdA from *Clostridium difficile* strain R20291 has approximately 98% identity with SEQ ID NO:1. SEQ ID NO:16 describes the wild-type gene from *Clostridium difficile* strain R20291, which is also disclosed in GenBank accession number NC_013316.1.

Another example of wild-type *Clostridium difficile* TcdA comprises the amino acid sequence shown in SEQ ID NO:17, which describes the wild-type amino acid sequence of TcdA from *Clostridium difficile* strain CD196 (also disclosed in GenBank accession number CBA61156.1). CD196 is a strain from a recent outbreak in Canada and is known in the art as PCR-RNA type 027 strain. The amino acid sequence of TcdA from *Clostridium difficile* strain CD196 has approximately 98% identity with SEQ ID NO:1 and approximately 100% identity with TcdA from *Clostridium difficile* strain R20291. SEQ ID NO:18 describes the wild-type gene of TcdA from *Clostridium difficile* strain CD196, which is also disclosed in GenBank accession number FN538970.1.

Further examples of the amino acid sequence of wild-type Clostridium difficile TcdA include SEQ ID NO:19, which describes the wild-type amino acid sequence of TcdA from Clostridium difficile strain VPI10463 (also disclosed in GenBank accession number CAA63564.1). The amino acid sequence of TcdA from Clostridium difficile strain VPI10463 has approximately 100% (99.8%) identity with SEQ ID NO:1. SEQ ID NO:20 describes the wild-type gene of TcdA from Clostridium difficile strain VPI10463, which is also disclosed in GenBank accession number X92982.1.

Further examples of wild-type Clostridium difficile TcdA include TcdA from wild-type Clostridium difficile strains available from the Centers for Disease Control and Prevention (CDC, Atlanta, GA). The inventors have found that, at optimal alignment, the amino acid sequence of TcdA from wild-type Clostridium difficile strains available from the CDC has at least approximately 99.3%–100% identity with amino acid residues 1–821 of SEQ ID NO:1 (TcdA from Clostridium difficile 630). See Table 1.

The inventors have also discovered that, when optimally aligned (e.g., when optimally aligned with the full-length sequence), the amino acid sequence of TcdA from wild-type Clostridium difficile strains can contain at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% to about 100% identity with SEQ ID NO:1.

Table 1: Percentage similarity of amino acid residues 1-821 of TcdA from wild-type Clostridium difficile strains obtained from CDC and from amino acid residues 1-821 of SEQ ID NO: 1 at optimal comparison.

Therefore, in one embodiment, the wild-type Clostridium difficile TcdA amino acid sequence comprises a sequence of at least about 500, 600, 700, or 800 consecutive residues, which, when optimally aligned, for example by using a predetermined gap weight procedure GAP or BESTFIT, has at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, preferably about 98%, more preferably about 99%, or most preferably about 100% identity with a sequence of the same length between residues 1-900 of SEQ ID NO:1. Examples include the strains described above (e.g., R20291, CD196, etc.) and those listed in Table 1.

In another embodiment, the wild-type Clostridium difficile TcdA amino acid sequence comprises a sequence that, upon optimal alignment, has at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, preferably about 97%, preferably about 98%, more preferably about 99%, or most preferably about 100% identity with any sequence selected from SEQ ID NO:87-109. See Table 1-a.

The wild-type TcdB gene has approximately 7098 nucleotides and encodes a protein with a reduced molecular weight of approximately 270 kDa and approximately 2366 amino acids. As used herein, wild-type Clostridium difficile TcdB comprises Clostridium difficile TcdB from any wild-type Clostridium difficile strain. Wild-type Clostridium difficile TcdB may contain a sequence that, when optimally aligned, for example by using a predetermined interval weight procedure GAP or BESTFIT, has at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, preferably about 98%, more preferably about 99%, or most preferably about 100% identity with SEQ ID NO:2. In a preferred embodiment, the wild-type *Clostridium difficile* TcdB comprises the amino acid sequence shown in SEQ ID NO:2, which describes the wild-type amino acid sequence of TcdB from *Clostridium difficile* strain 630 (also disclosed in GenBank accession numbers YP_001087135.1 and/or CAJ67492). SEQ ID NO:10 describes the wild-type gene of TcdB from *Clostridium difficile* strain 630, which is also disclosed in GenBank accession number NC_009089.1.

Another example of wild-type *Clostridium difficile* TcdB comprises the amino acid sequence shown in SEQ ID NO:21, which describes the wild-type amino acid sequence of TcdB from *Clostridium difficile* strain R20291 (also disclosed in GenBank accession numbers YP_003217086.1 and/or CBE02479.1). The amino acid sequence of TcdB from *Clostridium difficile* strain R20291 has approximately 92% identity with SEQ ID NO:2. SEQ ID NO:22 describes the wild-type gene of TcdB from *Clostridium difficile* strain R20291, which is also disclosed in GenBank accession number NC_013316.1.

Another example of wild-type *Clostridium difficile* TcdB comprises the amino acid sequence shown in SEQ ID NO:23, which describes the wild-type amino acid sequence of TcdB from *Clostridium difficile* strain CD196 (also disclosed in GenBank accession numbers YP_003213639.1 and/or CBA61153.1). SEQ ID NO:24 describes the wild-type gene of TcdB from *Clostridium difficile* strain CD196, which is also disclosed in GenBank accession number NC_013315.1. The amino acid sequence of TcdB from *Clostridium difficile* strain CD196 has approximately 92% identity with SEQ ID NO:2.

Other examples of the amino acid sequence of wild-type Clostridium difficile TcdB include SEQ ID NO:25, which describes the wild-type amino acid sequence of TcdB from Clostridium difficile strain VPI10463 (also disclosed in GenBank accession numbers P18177 and/or CAA37298). The amino acid sequence of TcdB from Clostridium difficile strain VPI10463 is 100% identical to SEQ ID NO:2. SEQ ID NO:26 describes the wild-type gene of TcdB from Clostridium difficile strain VPI10463, which is also disclosed in GenBank accession number X53138.1.

Other examples of wild-type Clostridium difficile TcdB include TcdB from wild-type Clostridium difficile strains available from the Centers for Disease Control and Prevention (CDC, Atlanta, GA). The inventors have found that, at optimal alignment, the amino acid sequence of TcdB from wild-type Clostridium difficile strains available from the CDC has at least approximately 96%–100% identity with amino acid residues 1–821 of SEQ ID NO:2 (TcdB from Clostridium difficile 630). See Table 2.

Table 2: Percentage similarity of amino acid residues 1-821 of TcdB from wild-type Clostridium difficile strains obtained from CDC and from amino acid residues 1-821 of SEQ ID NO:2 at optimal comparison.

Therefore, in one embodiment, the wild-type Clostridium difficile TcdB amino acid sequence comprises a sequence of at least about 500, 600, 700, or 800 consecutive residues, which, when optimally aligned, for example by using a predetermined gap weight procedure GAP or BESTFIT, has at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, preferably about 97%, preferably about 98%, more preferably about 99%, or most preferably about 100% identity with a sequence of the same length between residues 1-900 of SEQ ID NO:2. Examples include the strains described above (e.g., R20291, CD196, etc.) and those listed in Table 2.

In another embodiment, the wild-type Clostridium difficile TcdB amino acid sequence comprises a sequence that, upon optimal alignment, has at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, preferably about 97%, preferably about 98%, more preferably about 99%, or most preferably about 100% identity with any sequence selected from SEQ ID NO:110-133. See Table 2-a.

The genes for toxins A and B (tcdA and tcdB) are part of a 19.6-kb locus (pathogenicity locus, PaLoc), which contains three additional small open reading frames (ORFs) tcdD, tcdE, and tcdC, and can be considered useful for virulence. PaLoc is known to be stable and conserved in toxin-producing species. It is present at the same chromosomal integration site in all toxin-producing strains analyzed to date. The pathogenicity locus (PaLoc) is absent in non-toxin-producing strains. Therefore, the wild-type Clostridium difficile strain described herein is characterized by the presence of the pathogenicity locus. Another preferred feature of the wild-type Clostridium difficile strain described herein is the production of TcdA and TcdB.

In one embodiment, the wild-type *Clostridium difficile* strain is a strain possessing a pathogenic locus that is at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, preferably about 98%, more preferably about 99%, or most preferably about 100% identical to the pathogenic locus of *Clostridium difficile* 630 or VPI10463. The total pathogenic locus sequence of *Clostridium difficile* VPI10463 is registered in the EMBL database as accession number X92982 and is also shown in SEQ ID NO: 26. Strains with PaLoc identical to the reference strain VPI10463 are preferably designated as toxin type 0. Strains of toxin types I-VII, IX, XII-XV, and XVIII-XXIV produce TcdA and TcdB, regardless of variations in their toxin genes.

The glucosyltransferase domain is located at the N-terminus of the toxin. The glucosyltransferase activity of the toxin is related to its cytotoxic function. Unbound by any mechanism or theory, it is believed that the glucosyltransferase activity in both toxins catalyzes the monosaccharidation of small GTP-binding proteins in the Rho/Rac/Ras superfamily. Glycosylation of these GTP-binding proteins significantly modifies cellular physiology, causing toxin-infected host cells to lose structural integrity and disrupt important signal transduction pathways. The Asp-Xaa-Asp (DXD) motif, associated with manganese, uridine diphosphate (UDP), and glucose binding, is a typical feature of the glucosyltransferase domain. Unbound by mechanisms and theories, it is believed that residues important for catalytic activity, such as the DXD motif, remain directly unchanged in TcdB from known "historical" strains such as 630 and from highly virulent strains such as R20291. The DXD motif is located at residues 285-287 of wild-type Clostridium difficile TcdA (according to SEQ ID NO:1) and residues 286-288 of wild-type Clostridium difficile TcdB (according to SEQ ID NO:2).

Overall alignment algorithms (e.g., sequence analysis programs) are known in the art and can be used to optimally align two or more amino acid toxin sequences to determine whether the toxin contains a specific marker sequence (e.g., DXD in the glucosyltransferase domain, DHC in the cysteine protease domain, etc., as described below). The optimally aligned sequence is compared with its respective reference sequence (e.g., SEQ ID NO:1 of TcdA or SEQ ID NO:2 of TcdB) to determine the presence of the marker motif. “Optimal alignment” refers to the alignment that gives the highest percentage of similarity. Such alignments can be performed using known sequence analysis programs. In one embodiment, a suitable wild-type toxin is identified by comparing the query sequence with a reference sequence using a CLUSTAL alignment (e.g., CLUSTALW) with predetermined parameters. The relative numbering of conserved amino acid residues is based on the residue numbering of the reference amino acid sequence to indicate small insertions or deletions (e.g., 5 amino acids or less) in the aligned sequence.

As used herein, the term "according to the numbering of..." refers to the residue number of the reference sequence when comparing a given amino acid or polynucleotide sequence with a reference sequence. In other words, the number or residue position of a given polymer is specified relative to the reference sequence, rather than the actual numerical position of the residues in the given amino acid or polynucleotide sequence.

For example, a given amino acid sequence, such as that of a highly virulent wild-type Clostridium difficile strain, can be aligned with a reference sequence (e.g., the sequence of a historical wild-type Clostridium difficile strain, such as 630) if necessary to optimize the match between the two sequences. In these cases, although a gap is present, residue numbers in the given amino acid or polynucleotide are specified relative to the reference sequence being aligned. As used herein, "reference sequence" refers to the qualifying sequence used as the basis for sequence comparison.

Unless otherwise specified, all amino acid positions of TcdA in this document refer to SEQ ID NO:1. Unless otherwise specified, all amino acid positions of TcdB in this document refer to SEQ ID NO:2.

The glucosyltransferase domain of TcdA used herein may begin at exemplary residues 1, 101, or 102 of wild-type Clostridium difficile TcdA, such as SEQ ID NO:1, and may end at exemplary residues 542, 516, or 293. Any minimum residue position between residues 1 and 542 of TcdA may be combined with the maximum residue position to define the sequence of the glucosyltransferase domain, provided it contains the DXD motif region. For example, in one embodiment, the glucosyltransferase domain of TcdA comprises SEQ ID NO:27, which is identical to residues 101-293 of SEQ ID NO:1, and which contains the DXD motif region. In another embodiment, the glucosyltransferase domain of TcdA comprises SEQ ID NO:28, which is identical to residues 1-542 of SEQ ID NO:1.

The glucosyltransferase domain of TcdB used herein may begin at exemplary residues 1, 101, or 102 of wild-type Clostridium difficile TcdB, such as SEQ ID NO:2, and may end at exemplary residues 542, 516, or 293. Any minimum residue position between residues 1 and 543 of TcdB may be combined with the maximum residue position to define the sequence of the glucosyltransferase domain, provided that it contains the DXD motif region. For example, in one embodiment, the glucosyltransferase domain of TcdB comprises SEQ ID NO:29, which is identical to residues 101-293 of SEQ ID NO:2, and which contains the DXD motif region. In another embodiment, the glucosyltransferase domain of TcdB comprises SEQ ID NO:30, which is identical to residues 1-543 of SEQ ID NO:1.

Unbound by theories or mechanisms, it is believed that the N-terminus of TcdA and/or TcdB is cleaved by autoproteolytic processes, resulting in the translocation and release of the glucosyltransferase domain into the host cell cytoplasm, where it can interact with Rac/Ras/Rho GTPases. Wild-type Clostridium difficile TcdA has been shown to be cleaved between L542 and S543. Wild-type Clostridium difficile TcdB has been shown to be cleaved between L543 and G544.

The cysteine protease domain is associated with the autocatalytic proteolytic activity of the toxin. The cysteine protease domain is located downstream of the glucosyltransferase domain and is characterized by a catalytic tripartite of aspartic, histidine, and cysteine (DHC), such as D589, H655, and C700 of wild-type TcdA and D587, H653, and C698 of wild-type TcdB. Without being bound by mechanistic or theoretical constraints, the catalytic tripartite is considered conserved between toxins from "historical" strains such as 630 and TcdB from highly virulent strains such as R20291.

The cysteine protease domain of TcdA used herein may begin at exemplary residue 543 of wild-type TcdA, such as SEQ ID NO:1, and may end at exemplary residues 809, 769, 768, or 767. Any minimum residue position between residues 543 and 809 of wild-type TcdA may be combined with the maximum residue position to define the sequence of the cysteine protease domain, provided it contains a catalytic triplet DHC motif region. For example, in one embodiment, the cysteine protease domain of TcdA comprises SEQ ID NO:32, having a DHC motif region located at residues 47, 113, and 158 of SEQ ID NO:32, which correspond to D589, H655, and C700 of wild-type TcdA numbered according to SEQ ID NO:1, respectively. SEQ ID NO:32 is identical to residues 543-809 of SEQ ID NO:1, TcdA.

The cysteine protease domain of TcdB used herein may begin at exemplary residue 544 of wild-type TcdB, such as SEQ ID NO:2, and may end at exemplary residues 801, 767, 755, or 700. Any minimum residue position between residues 544 and 801 of wild-type TcdB may be combined with the maximum residue position to define the sequence of the cysteine protease domain, provided it contains a catalytic triplet DHC motif region. For example, in one embodiment, the cysteine protease domain of TcdB comprises SEQ ID NO:33, which includes a DHC motif region located at residues 44, 110, and 115 of SEQ ID NO:33, corresponding to D587, H653, and C698 of wild-type TcdB numbered according to SEQ ID NO:2, respectively. SEQ ID NO:33 is identical to residues 544-767 of SEQ ID NO:2, TcdB. In another embodiment, the cysteine protease domain of TcdB comprises residues 544-801 of SEQ ID NO:2.

In this application, the immunogenic composition comprises a mutant Clostridium difficile toxin. As used herein, the term "mutant" refers to a molecule exhibiting a structure or sequence different from the corresponding wild-type structure or sequence, for example by having cross-links compared to the corresponding wild-type structure and/or by having at least one mutation compared to the corresponding wild-type sequence when optimally aligned, for example by using a predetermined gap weight procedure GAP or BESTFIT. As used herein, the term "mutant" also includes molecules exhibiting functional properties different from the corresponding wild-type molecule (e.g., loss of glucosyltransferase and/or loss of cysteine protease activity).

Clostridium difficile toxins from any of the above-described wild-type strains can be used as a source for producing mutant Clostridium difficile toxins. Preferably, Clostridium difficile 630 is a source for producing mutant Clostridium difficile toxins.

Mutations can include substitutions, deletions, truncations, or modifications of wild-type amino acid residues that are normally located at that position. Preferably, the mutation is a non-conserved amino acid substitution. The invention also covers isolated polynucleotides comprising nucleic acid sequences encoding any of the mutant toxins described herein.

The term "non-conservative" amino acid substitution used in this article refers to the change from one type of amino acid to another, as shown in Table 3 below.

Examples of non-conservative amino acid substitutions include substitutions in which an aspartic acid residue (Asp, D) is replaced by an alanine residue (Ala, A). Other examples include substitution of an aspartic acid residue (Asp, D) with an asparagine residue (Asn, N), and substitution of arginine (Arg, R), glutamic acid (Glu, E), lysine (Lys, K), and/or histidine (His, H) residues with an alanine residue (Ala, A).

Conservative substitution refers to, for example, the exchange of amino acids between the same classes according to Table 3.

The mutant toxin of the present invention can be prepared using techniques known in the art for preparing mutants, such as site-directed mutagenesis, mutagenesis using a mutagen (e.g., UV light), etc. Site-directed mutagenesis is preferred. Alternatively, nucleic acid molecules having the target sequence can be synthesized directly. Such chemical synthesis methods are known in the art.

In this invention, the mutant Clostridium difficile toxin contains at least one mutation in the glucosyltransferase domain relative to the corresponding wild-type Clostridium difficile toxin. In one embodiment, the glucosyltransferase domain contains at least two mutations. Preferably, the mutation reduces or eliminates the glucosyltransferase activity of the toxin compared to the glucosyltransferase activity of the corresponding wild-type Clostridium difficile toxin.

Exemplary amino acid residues in the glucosyltransferase domain of the mutable TcdA include at least one or any combination of the following (according to the numbering of SEQ ID NO:1, compared to wild-type Clostridium difficile TcdA): W101, D269, R272, D285, D287, E460, R462, S541, and L542.

Exemplary mutations in the glucosyltransferase domain of TcdA include at least one or any combination of the following (compared to wild-type Clostridium difficile TcdA): W101A, D269A, R272A, D285A, D287A, E460A, R462A, S541A, and L542G. In another preferred embodiment, the glucosyltransferase domain of TcdA includes mutations of D285A and D287A (compared to wild-type Clostridium difficile TcdA).

Exemplary amino acid residues in the glucosyltransferase domain of the mutable TcdB include at least one or any combination of the following (according to the numbering of SEQ ID NO:2, compared to wild-type Clostridium difficile TcdB): W102, D270, R273, D286, D288, N384, D461, K463, W520, and L543.

Exemplary mutations in the glucosyltransferase domain of TcdB include at least one or any combination of the following (compared to wild-type Clostridium difficile TcdB): W102A, D270A, D270N, R273A, D286A, D288A, N384A, D461A, D461R, K463A, K463E, W520A, and L543A. In a preferred embodiment, the glucosyltransferase domain of TcdB comprises L543A (compared to wild-type Clostridium difficile TcdB). In another preferred embodiment, the glucosyltransferase domain of TcdB comprises D286A and D288A mutations (compared to wild-type Clostridium difficile TcdB).

Any of the mutations described above can be combined with mutations in the cysteine protease domain. In this invention, the mutant Clostridium difficile toxin contains at least one mutation in the cysteine protease domain relative to the corresponding wild-type Clostridium difficile toxin. Preferably, the mutation reduces or eliminates the cysteine protease activity of the toxin compared to the cysteine protease activity of the corresponding wild-type Clostridium difficile toxin.

Exemplary amino acid residues in the cysteine protease domain of the mutable TcdA include at least one or any combination of the following (according to the numbering of SEQ ID NO:1, compared to wild-type Clostridium difficile TcdA): S543, D589, H655, and C700. Exemplary mutations in the glucosyltransferase domain of TcdA include at least one or any combination of the following (compared to wild-type Clostridium difficile TcdA): S543A, D589A, D589N, H655A, and C700A. In a preferred embodiment, the cysteine protease domain of TcdA includes the C700A mutation (compared to wild-type Clostridium difficile TcdA).

Exemplary amino acid residues in the cysteine protease domain of the mutable TcdB include at least one or any combination of the following (according to SEQ ID NO:2, compared to wild-type Clostridium difficile TcdB): G544, D587, H653, and C698. In a preferred embodiment, the cysteine protease domain of TcdB includes the C698A mutation (compared to wild-type Clostridium difficile TcdB). Other amino acid residues in the cysteine protease domain of the mutable TcdB include K600 and/or R751 (compared to wild-type TcdB). Exemplary mutations include K600E and/or R751E.

Therefore, the mutant Clostridium difficile toxin of the present invention has a mutated glucose transferase domain relative to the corresponding wild-type Clostridium difficile toxin and a mutated cysteine protease domain relative to the corresponding wild-type Clostridium difficile toxin.

An exemplary mutant *Clostridium difficile* TcdA comprises a glucosyltransferase domain, which includes SEQ ID NO:29, having amino acid substitutions at positions 285 and 287 relative to the corresponding wild-type *Clostridium difficile* toxin A; and comprises a cysteine protease domain, which includes SEQ ID NO:32, having an amino acid substitution at position 158 relative to the corresponding wild-type *Clostridium difficile* toxin A. For example, such a mutant *Clostridium difficile* TcdA comprises the amino acid sequence shown in SEQ ID NO:4, wherein the initial methionine is optionally absent. In another embodiment, the mutant *Clostridium difficile* toxin A comprises the amino acid sequence shown in SEQ ID NO:84.

Other examples of mutant Clostridium difficile toxin A comprise the amino acid sequence shown in SEQ ID NO:7, having mutations in D269A, R272A, D285A, D287A, E460A, R462A, and C700A with SEQ ID NO:1, wherein the initiating methionine is optionally absent. In another embodiment, mutant Clostridium difficile toxin A comprises the amino acid sequence shown in SEQ ID NO:83.

Another exemplary mutant TcdA contains SEQ ID NO:34, wherein the residues at positions 101, 269, 272, 285, 287, 460, 462, 541, 542, 543, 589, 655 and 700 can be any amino acid.

In some embodiments, mutant Clostridium difficile toxins exhibit reduced or eliminated autoproteolytic processing compared to the corresponding wild-type Clostridium difficile toxins. For example, mutant Clostridium difficile TcdA may contain mutations at one or any combination of the following residues (compared to the corresponding wild-type Clostridium difficile TcdA): S541, L542, and/or S543. Preferably, mutant Clostridium difficile TcdA contains at least one or any combination of the following mutations (compared to the corresponding wild-type Clostridium difficile TcdA): S541A, L542G, and S543A.

Another exemplary mutant, Clostridium difficile TcdA, contains (compared to the corresponding wild-type Clostridium difficile TcdA) the S541A, L542, S543, and C700 mutations.

An exemplary mutant Clostridium difficile toxin B comprises a glucosyltransferase domain, which includes SEQ ID NO:31, having amino acid substitutions at positions 286 and 288 relative to the corresponding wild-type Clostridium difficile toxin B, and comprises a cysteine protease domain, which includes SEQ ID NO:33, having an amino acid substitution at position 155 relative to the corresponding wild-type Clostridium difficile toxin B. For example, such a mutant Clostridium difficile TcdB comprises the amino acid sequence shown in SEQ ID NO:6, wherein the initial methionine is optionally absent. In another embodiment, the mutant Clostridium difficile toxin A comprises the amino acid sequence shown in SEQ ID NO:86.

Other examples of the mutant Clostridium difficile TcdB comprise the amino acid sequence shown in SEQ ID NO:8, having mutations in D270A, R273A, D286A, D288A, D461A, K463A, and C698A relative to SEQ ID NO:2, wherein the initiating methionine is optionally absent. In another embodiment, the mutant Clostridium difficile toxin A comprises the amino acid sequence shown in SEQ ID NO:85.

Another exemplary mutant TcdB comprises SEQ ID NO:35, wherein the residues at positions 101, 269, 272, 285, 287, 460, 462, 541, 542, 543, 589, 655 and 700 can be any amino acid.

As another example, the mutant *Clostridium difficile* TcdB may contain mutations at positions 543 and/or 544 compared to the corresponding wild-type *Clostridium difficile* TcdB. Preferably, the mutant *Clostridium difficile* TcdB contains (compared to the corresponding wild-type *Clostridium difficile* TcdB) L543 and/or G544 mutations. More preferably, the mutant *Clostridium difficile* TcdB contains (compared to the corresponding wild-type *Clostridium difficile* TcdB) L543G and/or G544A mutations.

Another exemplary mutant Clostridium difficile TcdB contains (compared to the corresponding wild-type Clostridium difficile TcdB) the L543G, G544A, and C698 mutations.

In one aspect, the present invention relates to isolated polypeptides having mutations at any position of amino acid residues 1-1500 according to SEQ ID NO:2 to define an exemplary mutant clostridium difficile toxin B. For example, in one embodiment, the isolated polypeptide comprises a mutation between amino acid residues 830 and 990 of SEQ ID NO:2. Exemplary positions of the mutation include positions 970 and 976 according to SEQ ID NO:2. Preferably, the mutation between residues 830 and 990 is a substitution. In one embodiment, the mutation is a non-conservative substitution in which Asp(D) and/or Glu(E) amino acid residues are not replaced by neutralizing amino acid residues such as lysine (K), arginine (R), and histidine (H) when acidified. Exemplary mutations include E970K, E970R, E970H, E976K, E976R, and E976H of SEQ ID NO:2 to define the mutant clostridium difficile toxin B.

In another aspect, the present invention relates to isolated polypeptides having mutations at any position of amino acid residues 1-1500 according to SEQ ID NO:1 to define an exemplary mutant clostridium difficile toxin A. For example, in one embodiment, the isolated polypeptide comprises a mutation between amino acid residues 832 and 992 of SEQ ID NO:1. Exemplary positions of the mutation include positions 972 and 978 according to SEQ ID NO:1. Preferably, the mutation between residues 832 and 992 is a substitution. In one embodiment, the mutation is a non-conservative substitution in which Asp(D) and/or Glu(E) amino acid residues are not replaced by neutralizing amino acid residues such as lysine (K), arginine (R), and histidine (H) when acidified. Exemplary mutations include D972K, D972R, D972H, D978K, D978R, and D978H of SEQ ID NO:1 to define the mutant clostridium difficile toxin A.

The peptides of the present invention may contain an initiating methionine residue, in some cases due to a host cell-mediated process. Depending on, for example, the recombinant production method and/or the fermentation or growth conditions of the host cell, it is known in the art that the N-terminal methionine encoded by the translation start codon can be removed from the peptide post-translation in the cell, or the N-terminal methionine can be retained in the isolated peptide.

Therefore, in one aspect, the present invention relates to an isolated polypeptide comprising the amino acid sequence shown in SEQ ID NO:4, wherein the initiating methionine (located at position 1) is optionally absent. In one embodiment, the initiating methionine of SEQ ID NO:4 is absent. In another aspect, the present invention relates to an isolated polypeptide comprising the amino acid sequence shown in SEQ ID NO:84, which is identical to SEQ ID NO:4, but without the initiating methionine.

On the other hand, the isolated polypeptide comprises the amino acid sequence shown in SEQ ID NO:6, wherein the initiating methionine (located at position 1) is optionally absent. In one embodiment, the initiating methionine of SEQ ID NO:6 is absent. On one hand, the present invention relates to an isolated polypeptide comprising the amino acid sequence shown in SEQ ID NO:86, which is identical to SEQ ID NO:6, but without the initiating methionine.

In another aspect, the isolated polypeptide comprises the amino acid sequence shown in SEQ ID NO:7, wherein the starting methionine (located at position 1) is optionally absent. In one embodiment, the present invention relates to an isolated polypeptide comprising the amino acid sequence shown in SEQ ID NO:83, which is identical to SEQ ID NO:7, but without the starting methionine. In another aspect, the isolated polypeptide comprises the amino acid sequence shown in SEQ ID NO:8, wherein the starting methionine (located at position 1) is optionally absent. In one embodiment, the isolated polypeptide comprises the amino acid sequence shown in SEQ ID NO:85, which is identical to SEQ ID NO:8, but without the starting methionine.

In one aspect, the present invention relates to an immunogenic composition comprising SEQ ID NO:4, wherein the initiating methionine (located at position 1) is optionally absent. In another aspect, the present invention relates to an immunogenic composition comprising SEQ ID NO:6, wherein the initiating methionine (located at position 1) is optionally absent. In yet another aspect, the present invention relates to an immunogenic composition comprising SEQ ID NO:7, wherein the initiating methionine (located at position 1) is optionally absent. In yet another aspect, the present invention relates to an immunogenic composition comprising SEQ ID NO:8, wherein the initiating methionine (located at position 1) is optionally absent.

In another aspect, the present invention relates to an immunogenic composition comprising SEQ ID NO:83. In another aspect, the present invention relates to an immunogenic composition comprising SEQ ID NO:84. In another aspect, the present invention relates to an immunogenic composition comprising SEQ ID NO:85. In yet another aspect, the present invention relates to an immunogenic composition comprising SEQ ID NO:86.

In addition to evoking an immune response in mammals, the immunogenic compositions described herein exhibit reduced cytotoxicity compared to the corresponding wild-type Clostridium difficile toxins. Preferably, the immunogenic compositions are safe for administration to mammals compared to their respective wild-type toxins and exhibit minimal (e.g., a reduction of about 6-8 log ₁₀ ) cytotoxicity to no cytotoxicity.

As used herein, the term cytotoxicity is the term as understood in the art and refers to apoptotic cell death and/or a state in which one or more normally biochemically or biologically abnormally impaired cells compared to the same cells under the same conditions but without a cytotoxic agent. Toxicity can be quantified, for example, by the amount of substance required to induce 50% cell death in a cell or mammal (i.e., EC ₅₀ or ED ₅₀ , respectively), or by other methods known in the art.

Assays indicating cytotoxicity are known in the art, such as cell rounding assays (see, for example, Kuehne et al. Nature. 2010 Oct 7; 467(7316):711-3). The effects of TcdA and TcdB cause cells to round (e.g., lose their shape) and die, and such phenomena can be observed by optical microscopy. See, for example, Figure 9.

Other exemplary cytotoxicity assays known in the art include glycosylation assays associated with Ras phosphorus screen imaging assays labeled with [ ^14C ] glucose (as described in Busch et al., J Biol Chem. 1998 Jul 31; 273(31): 19566-72) and cytotoxicity assays preferably described in the examples below, wherein EC ₅₀ may refer to the concentration of an immunogenic composition that exhibits at least about 50% cytopathic effect (CPE) in preferred human diploid fibroblasts (e.g., IMR90 cells (ATCC CCL-186 ^™ )) compared to the same cells under the same conditions without the toxin. In vitro cytotoxicity assays can also be used to evaluate the concentration of a composition that inhibits at least about 50% of wild-type Clostridium difficile toxin-induced cytopathic effect (CPE) in preferred human diploid fibroblasts (e.g., IMR90 cells (ATCC CCL-186 ^™ )) compared to the same cells under the same conditions without the toxin. Other exemplary cytotoxicity assays include Doern et al., J The cytotoxicity assays described in Clin Microbiol. 1992 Aug; 30(8): 2042-6. Cytotoxicity can also be determined by measuring the ATP levels in cells treated with the toxin. For example, a luciferase substrate such as Promega can be used, which emits fluorescence measured as relative light units (RLU). In such an assay, cell viability can be proportional to the amount of ATP or the RLU value in the cell.

In one embodiment, the cytotoxicity of the immunogenic composition is reduced by at least about 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, 10000, 11000, 12000, 13000-fold, 14000-fold, 15000-fold, or more compared to the corresponding wild-type Clostridium difficile toxin. See, for example, Table 20.

In another embodiment, the cytotoxicity of the immunogenic composition is reduced by at least about 2-log ₁₀ , more preferably about 3-log ₁₀ , and most preferably about 4-log ₁₀ or more compared to the corresponding wild-type toxin under the same conditions. For example, as measured by standard cytopathic effect (CPE), the EC ₅₀ of mutant Clostridium difficile TcdB can be about ^10-9 g/ml compared to an exemplary wild-type Clostridium difficile TcdB with an EC ₅₀ value of at least about ^10-12 g/ml. See, for example, Tables 7A, 7B, 8A, and 8B in the Examples section below.

In another embodiment, the _EC50 of the cytotoxicity of the mutant Clostridium difficile toxin, as measured by, for example, the in vitro cytotoxicity assay described herein, is at least about 50 μg/ml, 100 μg/ml, 200 μg/ml, 300 μg/ml, 400 μg/ml, 500 μg/ml, 600 μg/ml, 700 μg/ml, 800 μg/ml, 900 μg/ml, 1000 μg/ml, or greater. Therefore, in a preferred embodiment, the immunogenic composition and the mutant toxin are biologically safe for the administered mammal.

Unbound by mechanistic or theoretical constraints, TcdA with D285 and D287 mutations compared to wild-type TcdA, and TcdB with D286 and D288 mutations compared to wild-type TcdB, are expected to be defective in glycosyltransferase activity and therefore defective in inducing cytopathic effects. Furthermore, toxins with mutations in the DHC motif are expected to be defective in autocatalytic processing and therefore lack any cytotoxic effects.

However, the inventors unexpectedly discovered that the exemplary mutant TcdA having SEQ ID NO:4 and the exemplary mutant TcdB having SEQ ID NO:6 unexpectedly exhibited cytotoxicity (although significantly reduced compared to wild-type Clostridium difficile 630 toxin), despite exhibiting dysfunctional glucosyltransferase activity and dysfunctional cysteine protease activity. Without being bound by mechanistic or theoretical constraints, it is assumed that the mutant toxins produce cytotoxicity through a novel mechanism. However, the exemplary mutant TcdA having SEQ ID NO:4 and the exemplary mutant TcdB having SEQ ID NO:6 were unexpectedly immunogenic. See the examples below.

While chemical cross-linking of wild-type toxins may not inactivate them, the inventors have further discovered that at least one amino acid in the chemically cross-linked mutant toxin further reduces the cytotoxicity of the mutant toxin compared to the same mutant toxin lacking chemical cross-linking and compared to the corresponding wild-type toxin. Preferably, the mutant toxin is purified before contact with the chemical cross-linking agent.

Furthermore, although chemical cross-linking agents may alter useful epitopes, the inventors unexpectedly discovered that genetically modified mutant Clostridium difficile toxins having at least one chemically cross-linked amino acid yielded immunogenic compositions that elicited a variety of neutralizing antibodies or their binding fragments. Thus, epitopes associated with neutralizing antibody molecules were unexpectedly retained after chemical cross-linking.

Crosslinking (also referred to herein as “chemical deactivation” or “deactivation”) is a method of chemically linking two or more molecules through covalent bonds. The terms “crosslinking agent,” “crosslinking agent,” and “crosslinking substance” refer to molecules capable of reacting with and/or chemically linking to specific functional groups (primary amines, thiols, carboxyl groups, carbonyl groups, etc.) on peptides, polypeptides, and/or proteins. In one embodiment, the molecule may comprise two or more reactive ends capable of reacting with and/or chemically linking to specific functional groups (primary amines, thiols, carboxyl groups, carbonyl groups, etc.) on peptides, polypeptides, and/or proteins. Preferably, the chemical crosslinking agent is water-soluble. In another preferred embodiment, the chemical crosslinking agent is a heterodifunctional crosslinking substance. In another embodiment, the chemical crosslinking agent is not a bifunctional crosslinking substance. Chemical crosslinking agents are known in the art.

In a preferred embodiment, the crosslinking agent is a zero-length crosslinking agent. "Zero length" refers to a crosslinking agent that mediates or generates direct crosslinking between functional groups of two molecules. For example, in the crosslinking of two peptides, a zero-length crosslinking agent causes a bridge or crosslink between the carboxyl group forming the amino side chain of one peptide and the amino group of the other peptide without introducing exogenous substances. Zero-length crosslinking agents can catalyze, for example, the formation of ester bonds between hydroxyl and carboxyl moieties and/or amide bonds between carboxyl and primary amino moieties.

Exemplary suitable chemical crosslinking agents include formaldehyde; formalin; acetaldehyde; propionaldehyde; water-soluble carbodiimides (RN=C=NR’) including 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide (EDC), 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide hydrochloride, 3-(2-morpholino-(4-ethyl)carbodiimide N-methyl-p-toluenesulfonic acid 1-cyclohexyl ester (CMC), N,N′-dicyclohexylcarbodiimide (DCC) and N,N′-diisopropylcarbodiimide (DIC) and their derivatives; and N-hydroxysuccinimide (NHS); benzoylformaldehyde; and/or UDP-dialdehyde.

Preferably, the crosslinking agent is EDC. When the mutant Clostridium difficile toxin peptide is chemically modified with EDC (e.g., by contacting the peptide with EDC), in one embodiment, the peptide includes (a) at least one crosslink between an aspartic acid residue side chain and a glutamic acid residue side chain of the peptide. In one embodiment, the peptide includes (b) at least one crosslink between a glutamic acid residue side chain and a lysine residue side chain of the peptide. In one embodiment, the peptide includes (c) at least one crosslink between a C-terminal carboxyl group and an N-terminal amino group of the peptide. In one embodiment, the peptide includes (d) at least one crosslink between a C-terminal carboxyl group and a lysine residue side chain of the peptide. In one embodiment, the peptide includes (e) at least one crosslink between an aspartic acid residue side chain of the peptide and a lysine residue side chain of a second isolated peptide. In one embodiment, the peptide includes (f) at least one crosslink between a glutamic acid residue side chain of the peptide and a lysine residue side chain of a second isolated peptide. In one embodiment, the peptide includes (g) at least one crosslink between a C-terminal carboxyl group of the peptide and a second N-terminal amino group of the isolated peptide. In one embodiment, the polypeptide includes (h) at least one crosslink between the carboxyl group at the C-terminus of the polypeptide and the lysine residue side chain of the second isolated polypeptide. See, for example, Figures 24 and 25.

"Secondly separated polypeptide" refers to any separated polypeptide present during the reaction with EDC. In one embodiment, the second separated polypeptide is a mutant Clostridium difficile toxin polypeptide having the same sequence as the first separated polypeptide. In another embodiment, the second separated polypeptide is a mutant Clostridium difficile toxin polypeptide having a different sequence from the first separated polypeptide.

In one embodiment, the polypeptide includes at least two modifications selected from (a)-(d). In an exemplary embodiment, the polypeptide includes (a) at least one crosslink between an aspartic acid residue side chain and a glutamic acid residue side chain of the polypeptide; and (b) at least one crosslink between a glutamic acid residue side chain and a lysine residue side chain of the polypeptide. In another embodiment, the polypeptide includes at least three modifications selected from (a)-(d). In yet another embodiment, the polypeptide includes modifications (a), (b), (c), and (d).

When more than one mutant polypeptide is present during chemical modification via EDC, in one embodiment, the resulting composition contains at least one of any (a)-(h) modifications. In one embodiment, the composition contains at least two modifications selected from (a)-(h) modifications. In another embodiment, the composition contains at least three modifications selected from (a)-(h) modifications. In yet another embodiment, the composition contains at least four modifications selected from (a)-(h) modifications. In yet another embodiment, the composition contains at least one of each (a)-(h) modification.

In an exemplary embodiment, the resulting composition comprises (a) at least one crosslink between an aspartic acid residue side chain and a glutamic acid residue side chain of the polypeptide; and (b) at least one crosslink between a glutamic acid residue side chain and a lysine residue side chain of the polypeptide. In one embodiment, the composition further comprises (c) at least one crosslink between a carboxyl group at the C-terminus of the polypeptide and an amino group at the N-terminus of the polypeptide; and (d) at least one crosslink between a carboxyl group at the C-terminus of the polypeptide and a lysine residue side chain of the polypeptide.

In another exemplary embodiment, the resulting composition comprises (e) at least one crosslink between the aspartic acid residue side chain of the polypeptide and the lysine residue side chain of the second isolated polypeptide; (f) at least one crosslink between the glutamic acid residue side chain of the polypeptide and the lysine residue side chain of the second isolated polypeptide; (g) at least one crosslink between the carboxyl group at the C-terminus of the polypeptide and the amino group at the N-terminus of the second isolated polypeptide; and (h) at least one crosslink between the carboxyl group at the C-terminus of the polypeptide and the lysine residue side chain of the second isolated polypeptide.

In another exemplary embodiment, the resulting composition comprises (a) at least one crosslink between the aspartic acid residue side chain and the glutamic acid residue side chain of the polypeptide; (b) at least one crosslink between the glutamic acid residue side chain and the lysine residue side chain of the polypeptide; (e) at least one crosslink between the aspartic acid residue side chain of the polypeptide and the lysine residue side chain of the second isolated polypeptide; and (f) at least one crosslink between the glutamic acid residue side chain of the polypeptide and the lysine residue side chain of the second isolated polypeptide.

In a preferred embodiment, the chemical crosslinking agent comprises formaldehyde, and more preferably, in the absence of lysine, a reagent containing formaldehyde. Glycine, having a primary amine content, and other suitable compounds can be used as quenchers in the crosslinking reaction. Therefore, in another preferred embodiment, the chemical reagent comprises formaldehyde, and glycine is used.

In another preferred embodiment, the chemical cross-linking agent comprises EDC and NHS. As is known in the art, NHS can be included in the EDC coupling method. However, the inventors have surprisingly found that NHS can promote further reduction in the cytotoxicity of mutant Clostridium difficile toxins compared to the corresponding wild-type toxins, compared to genetically mutated toxins, and compared to genetically mutated toxins chemically cross-linked via EDC. See, for example, Example 22. Therefore, without being bound by mechanism or theory, mutant toxin peptides having a β-alanine moiety (e.g., derived from the reaction of mutant toxin peptides, EDC, and NHS) having at least one lysine residue side chain linked to the peptide can promote further reduction in the cytotoxicity of mutant toxins compared to, for example, Clostridium difficile toxins (wild-type or mutant) in which a β-alanine moiety is absent.

The use of EDC and/or NHS may also include the use of glycine or other suitable compounds having a primary amine as quenchers. Any compound having a primary amine can be used as a quencher, such as glycine methyl ester and alanine. In a preferred embodiment, the quencher compound is a nonpolymerizable hydrophilic primary amine. Examples of nonpolymerizable hydrophilic primary amines include, for example, amino sugars, amino alcohols, and amino polyols. Specific examples of nonpolymerizable hydrophilic primary amines include glycine, ethanolamine, glucosamine, amine-functionalized polyethylene glycol, and amine-functionalized polyethylene glycol oligomers.

In one aspect, the present invention relates to mutant Clostridium difficile toxin polypeptides having at least one amino acid side chain chemically modified by EDC and a non-polymeric hydrophilic primary amine, preferably glycine. Compared to the corresponding wild-type toxin, the resulting glycine adduct (e.g., from a triple mutant toxin treated with EDC and NHS and quenched with glycine) can promote a reduction in the cytotoxicity of the mutant toxin.

In one embodiment, when the mutant Clostridium difficile toxin polypeptide is chemically modified with EDC and glycine, the polypeptide comprises at least one modification of the polypeptide when modified with EDC (e.g., at least one of any of the modifications described in (a)-(h) above) and at least one of the following exemplary modifications: (i) a glycine moiety attached to a carboxyl group at the C-terminus of the polypeptide; (j) a glycine moiety attached to at least one aspartic acid residue side chain of the polypeptide; and (k) a glycine moiety attached to at least one glutamic acid residue side chain of the polypeptide. See, for example, Figures 24 and 25.

In one embodiment, at least one amino acid of the mutant *Clostridium difficile* TcdA is chemically cross-linked and/or at least one amino acid of the mutant *Clostridium difficile* TcdB is chemically cross-linked. In another embodiment, at least one amino acid of SEQ ID NO:4, SEQ ID NO:6, SEQ ID NO:7 and/or SEQ ID NO:8 is chemically cross-linked. For example, at least one amino acid can be chemically cross-linked by a reagent comprising a carbodiimide such as EDC. The carbodiimide can form a covalent bond between the free carboxyl group (e.g., from the side chain of aspartic acid and/or glutamic acid) and the amino group (e.g., in the side chain of lysine residues) to form a stable amide bond.

As another example, at least one amino acid can be chemically crosslinked using reagents including NHS. NHS ester-activated crosslinking substances can react with primary amines (e.g., at the N-terminus of each polypeptide chain and/or in the side chains of lysine residues) to generate amide bonds.

In another embodiment, at least one amino acid can be chemically crosslinked by a reagent including EDC and NHS. For example, in one embodiment, the invention relates to an isolated polypeptide having the amino acid sequence shown in SEQ ID NO:4, wherein the methionine residue at position 1 is optionally absent, wherein the polypeptide comprises at least one amino acid side chain chemically modified by EDC and NHS. In another embodiment, the invention relates to an isolated polypeptide having the amino acid sequence shown in SEQ ID NO:6, wherein the methionine residue at position 1 is optionally absent, wherein the polypeptide comprises at least one amino acid side chain chemically modified by EDC and NHS. In another embodiment, the invention relates to an isolated polypeptide having the amino acid sequence shown in SEQ ID NO:84, SEQ ID NO:86, SEQ ID NO:83, SEQ ID NO:85, SEQ ID NO:7, or SEQ ID NO:8. The polypeptide is modified by contacting the polypeptide with EDC and NHS. See, for example, Figures 24 and 25.

When the mutant Clostridium difficile toxin polypeptide is chemically modified by EDC and NHS (e.g., by contact), in one embodiment, the polypeptide comprises at least one modification of the polypeptide when it is modified by EDC (e.g., at least one of any of the modifications described above (a)-(h)) and (l) a β-alanine moiety linked to at least one lysine residue side chain of the polypeptide.

In another aspect, the present invention relates to mutant Clostridium difficile toxin polypeptides, wherein the polypeptide comprises at least one amino acid side chain chemically modified by EDC, NHS, and a nonpolymeric hydrophilic primary amine, preferably glycine. In one embodiment, the polypeptide comprises at least one modification when the polypeptide is modified by EDC (e.g., at least one of any of the modifications described above (a)-(h)), at least one modification when the polypeptide is modified by glycine (e.g., at least one of any of the modifications described above (i)-(k)), and (l) a β-alanine moiety linked to at least one lysine residue side chain of the polypeptide. See, for example, Figures 24 and 25.

In one aspect, the present invention relates to mutant Clostridium difficile toxin polypeptides, wherein at least one lysine residue of the polypeptide has a side chain linked to a β-alanine moiety. In one embodiment, a second lysine residue of the polypeptide has a side chain linked to an aspartic acid residue and/or a glutamate residue. The “second” lysine residue of the polypeptide includes lysine residues of the polypeptide that are not linked to the β-alanine moiety. The aspartic acid side chain and/or glutamate side chain to which the second lysine residue is linked may be an aspartic acid side chain and/or a glutamate side chain of the polypeptide to form an intramolecular crosslink, or may be an aspartic acid side chain and/or a glutamate side chain of the second polypeptide to form an intermolecular crosslink. In another embodiment, at least one aspartic acid residue of the polypeptide has a side chain and/or at least one glutamate residue linked to a glycine moiety. The aspartic acid residue and/or glutamate residue linked to the glycine moiety is not linked to a lysine residue.

As another example of chemically cross-linked mutant Clostridium difficile toxin peptides, at least one amino acid can be chemically cross-linked with a reagent including formaldehyde. Formaldehyde can react with N-terminal amino acid residues and the side chains of arginine, cysteine, histidine, and lysine. Formaldehyde and glycine can form a Schiff base adduct, which can be linked to N-terminal primary amino, arginine, and tyrosine residues, and to a lesser extent to asparagine, glutamine, histidine, and tryptophan residues.

Chemical cross-linking agents are believed to reduce the cytotoxicity of toxins, for example, by in vitro cytotoxicity assays or by animal toxicity measurements, if the treated toxin has lower toxicity than the untreated toxin under the same conditions (e.g., about 100%, 99%, 95%, 90%, 80%, 75%, 60%, 50%, 25%, or 10% less toxicity).

Preferably, compared to a mutant toxin under the same conditions but without a chemical cross-linking agent, the chemical cross-linking agent reduces the cytotoxicity of the mutant Clostridium difficile toxin by at least about 2-log ₁₀ , more preferably about 3-log ₁₀ , and most preferably about 4-log ₁₀ or more. Compared to the wild-type toxin, the chemical cross-linking agent preferably reduces the cytotoxicity of the mutant toxin by at least about 5-log ₁₀ , about 6-log ₁₀ , about 7-log ₁₀ , about 8-log ₁₀ or more.

In another preferred embodiment, for example by in vitro cytotoxicity assays as described herein, chemically inactivated mutant Clostridium difficile toxins exhibit EC 50 values greater than or at least about 50 μg/ml, 100 μg/ml, 200 μg/ml, 300 μg/ml, 400 μg/ml, 500 μg/ml, 600 μg/ml, 700 μg/ml, 800 μg/ml, 900 μg/ml, 1000 μg/ml, or _greater .

The reaction conditions for contacting the mutant toxin with the chemical cross-linking agent are within the experience of those skilled in the art, and these conditions can vary depending on the reagents used. However, the inventors have surprisingly discovered optimal reaction conditions for contacting the mutant Clostridium difficile toxin peptide with the chemical cross-linking agent while maintaining the functional epitope and reducing the cytotoxicity of the mutant toxin compared to the corresponding wild-type toxin.

Preferably, the conditions for contacting the mutant toxin with the cross-linking agent are selected, wherein the minimum concentration of the mutant toxin is about 0.5, 0.75, 1.0, 1.25, 1.5, 1.75, 2.0 mg/ml to the maximum concentration is about 3.0, 2.5, 2.0, 1.5 or 1.25 mg/ml. Any minimum value can be combined with any maximum value to define a suitable reactive concentration range for the mutant toxin. Most preferably, the reactive concentration of the mutant toxin is about 1.0–1.25 mg/ml.

In one embodiment, the minimum concentration of the reagents used in the reaction is about 1 mM, 2 mM, 3 mM, 4 mM, 5 mM, 10 mM, 15 mM, 20 mM, 30 mM, 40 mM, or 50 mM, and the maximum concentration is about 100 mM, 90 mM, 80 mM, 70 mM, 60 mM, or 50 mM. Any minimum value can be combined with any maximum value to define a suitable range of concentrations of the chemical reagents used in the reaction.

In a preferred embodiment where the reagent includes formaldehyde, the concentration used is preferably any concentration from about 2 mM to 80 mM, and most preferably about 40 mM. In another preferred embodiment where the reagent includes EDC, the concentration used is preferably any concentration from about 1.3 mM to about 13 mM, more preferably about 2 mM to 3 mM, and most preferably about 2.6 mM.

Exemplary reaction times for contacting the mutant toxin with the chemical cross-linking agent include minimums of about 0.5, 1, 2, 3, 4, 5, 6, 12, 24, 36, 48, or 60 hours, and maximums of about 14 days, 12 days, 10 days, 7 days, 5 days, 3 days, 2 days, 1 day, or 12 hours. Any minimum value can be combined with any maximum value to define a suitable range of reaction times.

In a preferred embodiment, the step of contacting the mutant toxin with the chemical cross-linking agent is carried out for a time sufficient to reduce the cytotoxicity of the mutant Clostridium difficile toxin to at least about ₁₀₀₀ μg/ml in suitable human cells, such as IMR-90 cells, in a standard in vitro cytotoxicity assay compared to the same mutant toxin without the cross-linking agent. More preferably, the reaction step is carried out for at least twice as long as the time required to reduce the cytotoxicity of the mutant _toxin to at least about 1000 μg/ml in suitable human cells, and most preferably at least three times or longer. In one embodiment, the reaction time does not exceed about 168 hours (or 7 days).

For example, in one embodiment where the reagent includes formaldehyde, the mutant toxin is preferably contacted with the reagent for about 12 hours, which has been shown to be an exemplary time period sufficient to reduce the cytotoxicity of the mutant Clostridium difficile toxin to at least about ₁₀₀₀ μg/ml in suitable human cells, such as IMR-90 cells, in standard in vitro cytotoxicity assays compared to the same mutant toxin without a cross-linking agent. In a more preferred embodiment, the reaction proceeds for about 48 hours, which is at least about three times the sufficient reaction time. In such an embodiment, the reaction time preferably does not exceed about 72 hours.

In another embodiment where the reagent includes EDC, the mutant toxin is preferably exposed to the reagent for about 0.5 hours, more preferably at least about 1 hour, or most preferably about 2 hours. In such an embodiment, the reaction time is preferably no more than about 6 hours.

Exemplary pH values for contacting the mutant toxin with the chemical cross-linking agent include minimum values of about pH 5.5, 6.0, 6.5, 7.0, or 7.5, and maximum values of about pH 8.5, 8.0, 7.5, 7.0, or 6.5. Any minimum value can be combined with any maximum value to define a suitable range of pH values. Preferably, the reaction is carried out at pH 6.5–7.5, and more preferably at pH 7.0.

Exemplary temperatures for contacting the mutant toxin with the chemical cross-linking agent include minimums of about 2°C, 4°C, 10°C, 20°C, 25°C, or 37°C, and maximum temperatures of about 40°C, 37°C, 30°C, 27°C, 25°C, or 20°C. Any minimum value can be combined with any maximum value to define a suitable range of reaction temperatures. Preferably, the reaction is carried out at about 20°C–30°C, and most preferably at 25°C.

The immunogenic composition described above may contain a mutant Clostridium difficile toxin (A or B). Therefore, the immunogenic composition may be included in a formulation or kit in isolation vials (e.g., isolation vials containing a composition containing mutant Clostridium difficile toxin A and isolation vials containing a composition containing mutant Clostridium difficile toxin B). The immunogenic compositions may be used simultaneously, sequentially, or separately.

In another embodiment, the immunogenic composition described above may comprise two mutant Clostridium difficile toxins (A and B). Any combination of mutant Clostridium difficile toxin A and mutant Clostridium difficile toxin B can be combined to form an immunogenic composition. Therefore, the immunogenic composition can be combined in a single vial (e.g., a single vial containing a composition containing mutant Clostridium difficile TcdA and a composition containing mutant Clostridium difficile TcdB). Preferably, the immunogenic composition comprises mutant Clostridium difficile TcdA and mutant Clostridium difficile TcdB.

For example, in one embodiment, the immunogenic composition comprises SEQ ID NO:4 and SEQ ID NO:6, wherein at least one amino acid of each of SEQ ID NO:4 and SEQ ID NO:6 is chemically cross-linked. In another embodiment, the immunogenic composition comprises mutant Clostridium difficile toxin A, which comprises SEQ ID NO:4 or SEQ ID NO:7, and mutant Clostridium difficile toxin B, which comprises SEQ ID NO:6 or SEQ ID NO:8, wherein at least one amino acid of each of the mutant Clostridium difficile toxins is chemically cross-linked.

In another embodiment, the immunogenic composition comprises any sequence selected from SEQ ID NO:4, SEQ ID NO:84, and SEQ ID NO:83, and any sequence selected from SEQ ID NO:6, SEQ ID NO:86, and SEQ ID NO:85. In another embodiment, the immunogenic composition comprises SEQ ID NO:84 and an immunogenic composition comprising SEQ ID NO:86. In another embodiment, the immunogenic composition comprises SEQ ID NO:83 and an immunogenic composition comprising SEQ ID NO:85. In another embodiment, the immunogenic composition comprises SEQ ID NO:84, SEQ ID NO:83, SEQ ID NO:86, and SEQ ID NO:85.

It should be understood that any composition of the present invention, such as an immunogenic composition comprising mutant toxin A and/or mutant toxin B, can be used in different proportions or combinations of amounts for therapeutic effect. For example, mutant Clostridium difficile TcdA and mutant Clostridium difficile TcdB may be present in the immunogenic composition in a ratio within the range of 0.1:10 to 10:0.1, A:B. In another embodiment, for example, mutant Clostridium difficile TcdB and mutant Clostridium difficile TcdA may be present in the immunogenic composition in a ratio within the range of 0.1:10 to 10:0.1, B:A. In a preferred embodiment, the proportions are such that the composition contains a larger total amount of mutant TcdB than the total amount of mutant TcdA.

In one aspect, the immunogenic composition is capable of binding a neutralizing antibody or a binding fragment thereof. Preferably, the neutralizing antibody or a binding fragment thereof is the neutralizing antibody or a binding fragment thereof described below. In an exemplary embodiment, the immunogenic composition is capable of binding an antitoxin A antibody or an antigen-binding fragment thereof, wherein the antitoxin A antibody or its binding fragment comprises a variable light chain having the amino acid sequence of SEQ ID NO:36 and a variable heavy chain having the amino acid sequence of SEQ ID NO:37. For example, the immunogenic composition may comprise a mutant Clostridium difficile TcdA, SEQ ID NO:4, or SEQ ID NO:7. As another example, the immunogenic composition may comprise SEQ ID NO:84 or SEQ ID NO:83.

In another exemplary embodiment, the immunogenic composition is capable of binding to an antitoxin B antibody or its antigen-binding fragment, wherein the antitoxin B antibody or its binding fragment comprises a variable light chain and a variable heavy chain of B8-26. For example, the immunogenic composition may comprise a mutant Clostridium difficile TcdB, SEQ ID NO:6, or SEQ ID NO:8. As another example, the immunogenic composition may comprise SEQ ID NO:86 or SEQ ID NO:85.

Recombinant cells

In another aspect, the present invention relates to recombinant cells or their progeny. In one embodiment, the cells or their progeny contain polynucleotides encoding mutant Clostridium difficile TcdA and/or mutant Clostridium difficile TcdB.

In another embodiment, the recombinant cell or its progeny contains a nucleic acid sequence that, when optimally aligned, for example by using a predetermined gap weight procedure GAP or BESTFIT, encodes a polypeptide that has at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, preferably about 98%, more preferably about 99%, or most preferably about 100% identity with any one of SEQ ID NO:4, SEQ ID NO:6, SEQ ID NO:7, or SEQ ID NO:8.

In another embodiment, the recombinant cell or its progeny contains a nucleic acid sequence that, when optimally aligned, for example by using a predetermined gap weight procedure GAP or BESTFIT, encodes a polypeptide that has at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, preferably about 98%, more preferably about 99%, or most preferably about 100% identity with any one of SEQ ID NO:84, SEQ ID NO:86, SEQ ID NO:83, or SEQ ID NO:85.

In another embodiment, the recombinant cell or its progeny contains a nucleic acid sequence that, when optimally aligned, for example by using a predetermined gap-weighted procedure GAP or BESTFIT, has at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, preferably about 98%, more preferably about 99%, or most preferably about 100% identity with any one of SEQ ID NO:11, SEQ ID NO:12, SEQ ID NO:13, SEQ ID NO:14, SEQ ID NO:44, SEQ ID NO:45, SEQ ID NO:46, or SEQ ID NO:47.

The recombinant cells can be derived from any cell suitable for recombinant production of the polypeptides of the present invention. For example, prokaryotic or eukaryotic cells. Preferably, the recombinant cells are derived from any cell suitable for expressing a heterologous nucleic acid sequence of more than about 5,000, 6,000, preferably about 7,000, and more preferably about 8,000 nucleotides or longer. The prokaryotic host cell can be any Gram-negative or Gram-positive bacterium. In an exemplary embodiment, the prokaryotic host cell lacks endogenous polynucleotides encoding toxins and/or spores.

Gram-negative bacteria include, but are not limited to, Campylobacter, Escherichia coli, Flavobacterium, Fusobacterium, Helicobacter, Ilyobacter, Neisseria, Pseudomonas, Salmonella, and Ureaplasma. For example, recombinant cells may be derived from Pseudomonas fluorescens cells, as described in paragraphs [0201]-[0230] of U.S. Patent Application Publication 2010013762, which is incorporated herein by reference.

Gram-positive bacteria include, but are not limited to, Bacillus, Clostridium, Enterococcus, Geobacillus, Lactobacillus, Lactococcus, Oceanobacillus, Staphylococcus, Streptococcus, and Streptomyces. Preferably, the cells are derived from Clostridium difficile cells.

The wild-type Clostridium difficile strains identified by the inventors lack endogenous polynucleotides encoding Clostridium difficile toxins. Strains lacking endogenous toxin A and B genes include the following strains available from the American Type Culture Collection (ATCC): Clostridium difficile 1351 (ATCC 43593 ^TM ), Clostridium difficile 3232 (ATCC BAA-1801 ^TM ), Clostridium difficile 7322 (ATCC 43601 ^TM ), Clostridium difficile 5036 (ATCC 43603 ^TM ), Clostridium difficile 4811 (ATCC 43602 ^TM ), and Clostridium difficile VPI11186 (ATCC 700057 ^TM ).

Therefore, in one embodiment, the recombinant Clostridium difficile cells are derived from the strains described herein. Preferably, the recombinant Clostridium difficile cells or their progeny are derived from strains selected from: Clostridium difficile 1351, Clostridium difficile 5036, and Clostridium difficile VPI11186. More preferably, the recombinant Clostridium difficile cells or their progeny are derived from Clostridium difficile VPI 11186 cells.

In a preferred embodiment, the sporulation gene in recombinant Clostridium difficile cells or their progeny is inactivated. The spores can be infectious, highly resistant, and promote the maintenance of Clostridium difficile in aerobic environments outside the host. The spores can also promote the survival of Clostridium difficile in the host during antimicrobial therapy. Therefore, Clostridium difficile cells lacking the sporulation gene can be used to produce safe immunogenic compositions for administration to mammals. Furthermore, using such cells promotes safety during preparation, such as protecting equipment, future products, and personnel.

Examples of sporulation genes that can be targeted for inactivation include spo0A, spoIIE, ^σE , ^σG , and ^σK . Preferably, the spo0A gene is inactivated.

Methods for inactivating Clostridium difficile sporulation genes are known in the art. For example, sporulation genes can be inactivated by targeted insertion of selective markers such as antibiotic resistance markers. See, for example, Heap et al., J Microbiol Methods. 2010 Jan; 80(1):49-55; Heap et al., J. Microbiol. Methods, 2007 Sept; 70(3):452-464; and Underwood et al., J Bacteriol. 2009 Dec; 191(23):7296-305. See also, for example, Minton et al., WO2007/148091, entitled “DNA Molecules and Methods,” pp. 33–66, which are incorporated herein by reference in their entirety, or the corresponding U.S. Publication US 20110124109 A1, paragraphs [00137]–[0227].

Methods for generating mutant Clostridium difficile toxins

In one aspect, the present invention relates to a method for producing mutant Clostridium difficile toxins. In one embodiment, the method comprises culturing any of the recombinant cells described above or their progeny under suitable conditions to express a polypeptide.

In another embodiment, the method includes culturing recombinant cells or their progeny under suitable conditions to express a polynucleotide encoding a mutant Clostridium difficile toxin, wherein the cells contain the polynucleotide encoding the mutant Clostridium difficile toxin, and wherein the mutant contains (relative to the corresponding wild-type Clostridium difficile toxin) at least one mutated glucosyltransferase domain and at least one mutated cysteine protease domain. In one embodiment, the cells lack an endogenous polynucleotide encoding the toxin.

In another embodiment, the method includes culturing recombinant Clostridium difficile cells or their progeny under suitable conditions to express a polynucleotide encoding a mutant Clostridium difficile toxin, wherein the cells contain a polynucleotide encoding the mutant Clostridium difficile toxin and the cells lack an endogenous polynucleotide encoding the toxin.

In another aspect, the present invention relates to a method for generating mutant Clostridium difficile toxin. The method comprises the steps of: (a) contacting Clostridium difficile cells with recombinant Escherichia coli cells, wherein the Clostridium difficile cells lack an endogenous polynucleotide encoding the mutant Clostridium difficile toxin and the Escherichia coli cells include a polynucleotide encoding the mutant Clostridium difficile toxin; (b) culturing the Clostridium difficile cells and the Escherichia coli cells under suitable conditions to transfer the polynucleotide from the Escherichia coli cells to the Clostridium difficile cells; (c) selecting Clostridium difficile cells containing a polynucleotide encoding the mutant Clostridium difficile toxin; (d) culturing the Clostridium difficile cells of step (c) under suitable conditions to express the polynucleotide; and (e) isolating the mutant Clostridium difficile toxin.

In the method of this invention, the recombinant *E. coli* cells comprise a heteropolynucleotide encoding the mutant *Clostridium difficile* toxin described herein. The polynucleotide may be DNA or RNA. In one exemplary embodiment, the polynucleotide encoding the mutant *Clostridium difficile* toxin has undergone codon optimization for *E. coli* codon selection. Methods for codon optimization of polynucleotides are well known in the art.

In one embodiment, the polynucleotide comprises a nucleic acid sequence that is at least about 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the polynucleotide encoding the mutant Clostridium difficile toxin A described herein. Exemplary polynucleotides encoding mutant Clostridium difficile toxin A include SEQ ID NO:11, SEQ ID NO:12, SEQ ID NO:44, and SEQ ID NO:45.

In another embodiment, the polynucleotide comprises a nucleic acid sequence that is at least about 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the polynucleotide encoding the mutant Clostridium difficile toxin B described herein. Exemplary polynucleotides encoding mutant Clostridium difficile toxin B include SEQ ID NO:13, SEQ ID NO:14, SEQ ID NO:46, and SEQ ID NO:47. In another embodiment, the polynucleotide encodes SEQ ID NO:83, SEQ ID NO:84, SEQ ID NO:85, or SEQ ID NO:86.

In one embodiment, the *E. coli* cell containing the heterologous polynucleotide is an *E. coli* cell that stably hosts the heterologous polynucleotide encoding a mutant *Clostridium difficile* toxin. Exemplary *E. coli* cells include cells selected from: MAXStbl2 ^™ *E. coli* competent cells (Invitrogen, Carlsbad, CA), One Stbl3 ^™ chemocompetent *E. coli* (Invitrogen, Carlsbad, CA), ElectroMAX ^™ Stbl4 ^™ *E. coli* competent cells (Invitrogen), and *E. coli* CA434. In a preferred embodiment, the *E. coli* clonal host cell is not DH5α. More preferably, the *E. coli* clonal host cell is a MAXStbl2 ^™ *E. coli* competent cell.

The method of the present invention further includes the step of culturing the *Clostridium difficile* cells and the *Escherichia coli* cells under suitable conditions to transfer polynucleotides from the *E. coli* cells to the *Clostridium difficile* cells, resulting in recombinant *Clostridium difficile* cells. In a preferred embodiment, the culture conditions are suitable for transferring polynucleotides from *E. coli* cells (donor cells) to *Clostridium difficile* cells (recipient cells) and resulting in genetically stable inheritance.

Most preferably, the culture conditions are suitable for bacterial conjugation, as is known in the art. "Conjugation" refers to a specific process of transferring polynucleotides, in which a unidirectional transfer of polynucleotides (e.g., bacterial plasmids) occurs from one bacterial cell (i.e., the "donor") to another (i.e., the "recipient"). The conjugation process involves contact between donor and recipient cells. Preferably, the donor *E. coli* cell is an *E. coli* CA434 cell.

Exemplary suitable (conjugating) conditions for transferring polynucleotides from *E. coli* cells to *C. dinitrate* cells include growing *C. dinitrate* liquid cultures in brain-heart extract medium (BHI; Oxoid) or Schaedlers anaerobic medium (SAB; Oxoid). In another embodiment, solid *C. dinitrate* cultures can be grown on fresh blood agar (FBA) or BHI agar. Preferably, *C. dinitrate* is grown at 37°C in an anaerobic environment (e.g., 80% _N₂ , 10% _CO₂ , and 10% _H₂ [vol/vol]). In one embodiment, the suitable conditions include aerobic growth of *E. coli* in Luria-Bertani (LB) medium or on LB agar at 37°C. For conjugative transfer to *C. dinitrate*, exemplary suitable conditions include anaerobic growth of *E. coli* on FBA. As known in the art, the liquid or solid culture medium may include antibiotics. Examples of such antibiotics include cycloserine (250 μg/ml), cefoxitin (8 μg/ml), chloramphenicol (12.5 μg/ml), thiamphenicol (15 μg/ml), and erythromycin (5 μg/ml).

The method of the present invention further includes the step of selecting recombinant Clostridium difficile cells comprising a polynucleotide encoding a mutant Clostridium difficile toxin. In an exemplary embodiment, the recombinant Clostridium difficile cells are receptors that conjugate polynucleotides encoding a mutant Clostridium difficile toxin derived from recombinant Escherichia coli cells.

The method of the present invention includes the step of culturing recombinant cells or their progeny under conditions suitable for expressing a polynucleotide encoding a mutant Clostridium difficile toxin, resulting in the production of the mutant Clostridium difficile toxin. Suitable conditions for the expression of the polynucleotide in the recombinant cells include culture conditions suitable for the growth of Clostridium difficile cells, which are known in the art. For example, suitable conditions may include culturing Clostridium difficile transformants in brain-heart extract medium (BHI; Oxoid) or Schaedlers anaerobic medium (SAB; Oxoid). In another embodiment, solid Clostridium difficile cultures may be grown on FBA or BHI agar. Preferably, Clostridium difficile is grown at 37°C in an anaerobic environment (e.g., 80% _N₂ , 10% _CO₂ , and 10% _H₂ [vol/vol]).

In one embodiment, the method of the present invention includes the step of isolating the resulting mutant Clostridium difficile toxin. Methods for isolating proteins from Clostridium difficile are known in the art.

In another embodiment, the method includes the step of purifying the resulting mutant Clostridium difficile toxin. Methods for purifying peptides, such as chromatography, are known in the art.

In an exemplary embodiment, the method further includes the step of contacting the isolated mutant Clostridium difficile toxin with the chemical cross-linking agent described above. Preferably, the cross-linking agent comprises formaldehyde, ethyl-3-(3-dimethylaminopropyl)carbodiimide, or a combination of EDC and NHS. Exemplary reaction conditions are described above and in the Examples section below.

In another aspect, the present invention relates to immunogenic compositions comprising the mutant Clostridium difficile toxin described herein produced by any method, preferably produced by the method described above.

Antibody

Surprisingly, the immunogenic compositions of the present invention elicit novel antibodies in vivo, indicating that the immunogenic compositions contain the native structures (e.g., preserved antigenic epitopes) retained by their respective wild-type Clostridium difficile toxins and that the immunogenic compositions contain epitopes. Antibodies produced against the toxin of one strain of Clostridium difficile can be able to bind to the corresponding toxin produced by another strain of Clostridium difficile. That is, antibodies and their binding fragments can be "cross-reactive," which means the ability to react with similar antigenic sites on toxins produced by multiple strains of Clostridium difficile. Cross-reactivity also includes the ability of antibodies to react with or bind to antigens that do not stimulate their production, i.e., the reaction between antigens and antibodies produced against different but similar antigens.

In one aspect, the inventors have surprisingly discovered monoclonal antibodies with neutralizing effects against Clostridium difficile toxins and methods for preparing the same. The antibodies of this invention can neutralize Clostridium difficile toxin cytotoxicity in vitro, inhibit the binding of Clostridium difficile toxins to mammalian cells, and/or neutralize Clostridium difficile toxin enterotoxicity in vivo. This invention also relates to isolated polynucleotides comprising nucleotide sequences encoding any of the aforementioned polypeptides. Furthermore, this invention relates to the use of any of the above-described compositions in the treatment and prevention of Clostridium difficile infection, Clostridium difficile-related diseases, syndromes, disease conditions, symptoms, and/or complications in mammals, reducing their risk, severity, morbidity, and/or delaying their onset compared to mammals not treated with the compositions, and methods for preparing the compositions.

The inventors further discovered that combinations of at least two neutralizing monoclonal antibodies can exhibit unexpected synergistic effects in neutralizing TcdA or TcdB, respectively. Antitoxin antibodies or their binding fragments can be used to inhibit Clostridium difficile infection.

An "antibody" is a protein containing at least one or two heavy (H) chain variable regions (abbreviated as VH in this paper) and at least one or two light (L) chain variable regions (abbreviated as VL in this paper). VH and VL can be further subdivided into hypervariable regions called "complementarity-determining regions" ("CDRs"), which intersect with more conserved regions called "framework regions" (FRs). The extent of the framework regions and CDRs has been precisely defined (see Kabat, E.A., et al. Sequences of Proteins of Immunological Interest, Fifth Edition, U.S. Department of Health and Human Services, NIH Publication No. 91-3242, 1991, and Chothia, C. et al., J. Mol. Biol. 196:901-917, 1987). The term "antibody" includes complete immunoglobulins of the IgA, IgG, IgE, IgD, and IgM types (and their subclasses), wherein the light chain of the immunoglobulin may be of the κ or λ type.

The antibody molecule can be full-length (e.g., IgG1 or IgG4 antibody). The antibody can have various isotypes, including IgG (e.g., IgG1, IgG2, IgG3, IgG4), IgM, IgA1, IgA2, IgD, or IgE. In a preferred embodiment, the antibody is an IgG isotype, such as IgG1. In another preferred embodiment, the antibody is an IgE antibody.

In another embodiment, the antibody molecule includes an "antigen-binding fragment" or "binding fragment," as used herein, referring to a portion of an antibody that specifically binds to a Clostridium difficile toxin (e.g., toxin A). A binding fragment is, for example, a molecule in which one or more immunoglobulin chains are not full-length but specifically bind to the toxin.

Examples of binding portions covered in the term "binding fragment" of an antibody include (i) a Fab fragment, a monovalent fragment consisting of VL, VH, CL, and CH1 domains; (ii) an F(ab′) ₂ fragment, a bivalent fragment comprising two Fab fragments linked by disulfide bonds in a hinge region; (iii) an Fd fragment, consisting of VH and CH1 domains; (iv) an Fv fragment, consisting of the VL and VH domains of a single arm of the antibody; (v) a dAb fragment (Ward et al., Nature 341:544-546, 1989), consisting of a VH domain; and (vi) a separated complementarity-determining region (CDR) with sufficient framework to specifically bind antigen-binding portions, such as variable regions. The binding fragments of the light chain variable region and the binding fragments of the heavy chain variable region, such as the two domains VL and VH of the Fv fragment, can be linked by a synthetic linker that makes them single-chain proteins using recombination methods. VL and VH pair to form a monovalent molecule (called a single-chain Fv (scFv), see, for example, Bird et al. (1988) Science 242:423-426; and Hustonet et al. (1988) Proc. Natl. Acad. Sci. USA 85:5879-5883). Such single-chain antibodies are also encompassed within the term "binding fragment" of antibody. These antibody portions are obtained using techniques known in the art and are screened for use in the same manner as whole antibodies.

As used herein, an antibody that "specifically binds" to a specific polypeptide or an epitope on a specific polypeptide, or is "specific" to a specific polypeptide or an epitope on a specific polypeptide, is an antibody that binds to that specific polypeptide or epitope on a specific polypeptide but substantially does not bind to any other polypeptide or polypeptide epitope. For example, when referring to a biomolecule (e.g., a protein, nucleic acid, antibody, etc.) that "specifically" binds to a target, such as when measured under specified conditions (e.g., immunoassay conditions in the case of an antibody), the biomolecule binds its target molecule in a mixed population of molecules containing the target and does not bind other molecules in significant quantities. The binding between an antibody and its target determines the presence of the target in a mixed population of molecules. For example, "specifically binding" or "specifically binding" means that the affinity of an antibody or its binding fragment for wild-type and/or mutant toxins of Clostridium difficile is at least twice its affinity for nonspecific antigens.

In an exemplary embodiment, the antibody is a chimeric antibody. Chimeric antibodies can be produced using recombinant DNA techniques known in the art. For example, the gene encoding the Fc constant region of a mouse (or other species) monoclonal antibody molecule can be digested with a restriction enzyme to remove the region encoding the mouse Fc and replace it with an equivalent portion of the gene encoding the human Fc constant region. Chimeric antibodies can also be produced using recombinant DNA techniques, wherein DNA encoding the mouse variable region can be ligated to DNA encoding the human constant region.

In another exemplary embodiment, the antibody or its binding fragment is humanized using methods known in the art. For example, once a mouse antibody is obtained, at least a portion of the antibody's CDR can be replaced with a human CDR. Humanized antibodies can also be generated by replacing sequences of mouse Fv variable regions not directly involved in antigen binding with equivalent sequences of human Fv variable regions. General methods for generating humanized antibodies are known in the art.

For example, monoclonal antibodies against Clostridium difficile TcdA or Clostridium difficile TcdB can also be generated using labeling techniques such as hybridoma technology (see, for example, Kohler and Milstein, 1975, Nature, 256:495-497). In short, immortalized cell lines are fused with lymphocytes from mammals immunized with Clostridium difficile TcdA, Clostridium difficile TcdB, or mutant Clostridium difficile toxins described herein, and the resulting hybridoma cell culture supernatants are screened to identify hybridomas that produce monoclonal antibodies binding to Clostridium difficile TcdA or Clostridium difficile TcdB. Typically, the immortalized cell lines are derived from the same mammalian lymphocytes. Hybridoma cells producing the monoclonal antibodies of the present invention are detected by screening the hybridoma culture supernatants for antibodies binding to Clostridium difficile TcdA or Clostridium difficile TcdB using assays such as ELISA. Human hybridomas can be prepared in a similar manner.

As an alternative to antibody production through immunization and selection, the antibodies of the present invention can also be identified by screening recombinant immunoglobulin libraries using Clostridium difficile TcdA, Clostridium difficile TcdB, or mutant Clostridium difficile toxin as described herein. The recombinant antibody library can be, for example, an scFv library or a Fab library. Furthermore, the antibodies described herein can also be used in competitive binding studies to identify additional anti-TcdA or anti-TcdB antibodies and their binding fragments. For example, additional anti-TcdA or anti-TcdB antibodies and their binding fragments can be identified by screening human antibody libraries and identifying molecules in those libraries that compete with the antibodies described herein in a competitive binding assay.

Furthermore, the antibodies covered by this invention include recombinant antibodies, which can be generated using phage display methods known in the art. In phage display methods, phages can be used to display antigen-binding domains expressed from a repertoire or antibody library (e.g., human or mouse). Phages expressing antigen-binding domains that bind to immunogens described herein (e.g., mutant Clostridium difficile toxins) can be selected or identified using antigens such as labeled antigens.

Antibodies and their binding fragments are also within the scope of this invention, wherein specific amino acids have been substituted, deleted, or added. Preferably, preferred antibodies have amino acid substitutions in the frame region to improve binding to the antigen. For example, a small number of receptor frame residues selected from the immunoglobulin chain can be replaced by corresponding donor amino acids. Preferred positions of the substitutions include amino acid residues near the CDR, or those capable of interacting with the CDR. Standards for selecting amino acids from donors are described in U.S. Pat. No. 5,585,089 (e.g., columns 12-16). The receptor frame can be a mature human antibody frame sequence or its common sequence.

As used herein, "neutralizing antibody or its binding fragment" refers to the respective antibody or its binding fragment that binds to a pathogen (such as Clostridium difficile TcdA or TcdB) and reduces the infectivity and/or activity (e.g., reduced cytotoxicity) of the pathogen in mammalian and/or cell cultures compared to the pathogen under the same conditions in the absence of the neutralizing antibody or its binding fragment. In one embodiment, the neutralizing antibody or its binding fragment is capable of neutralizing at least about 70%, 75%, 80%, 85%, 90%, 95%, 99% or more of the biological activity of the pathogen compared to the pathogen under the same conditions in the absence of the neutralizing antibody or its binding fragment.

As used herein, the term "antitoxin antibody or its binding fragment" refers to an antibody or its binding fragment that binds to a respective clostridium difficile toxin (e.g., clostridium difficile toxin A or toxin B). For example, an antitoxin A antibody or its binding fragment to an antibody or its binding fragment that binds to TcdA.

The antibodies or their binding fragments described herein can be produced in any wild-type and/or transgenic mammal, including, for example, mice, humans, rabbits, and goats.

When the above-described immunogenic composition is an immunogenic composition pre-administered to a population, for example for vaccination, the antibody response generated in an individual can be used to neutralize toxins from the same strain as well as from strains that do not stimulate the production of said antibodies. See, for example, Example 37, which illustrates the cross-reactivity of the immunogenic composition between toxins from 630 strains and from various wild-type Clostridium difficile strains.

In one aspect, the present invention relates to antibodies or binding fragments thereof specific to Clostridium difficile TcdA. Monoclonal antibodies that specifically bind to TcdA include A65-33, A60-22, A80-29 and/or preferably A3-25.

In one aspect, the present invention relates to antibodies or binding fragments thereof specific to TcdA from any wild-type Clostridium difficile strain, such as those described above, for example, specific to SEQ ID NO:1. In another aspect, the present invention relates to antibodies or binding fragments thereof specific to the immunogenic compositions described above. For example, in one embodiment, the antibody or binding fragment thereof is specific to an immunogenic composition comprising SEQ ID NO:4 or SEQ ID NO:7. In another embodiment, the antibody or binding fragment thereof is specific to an immunogenic composition comprising SEQ ID NO:4 or SEQ ID NO:7, wherein at least one amino acid of SEQ ID NO:4 or SEQ ID NO:7 is cross-linked with formaldehyde, EDC, NHS, or a combination of EDC and NHS. In another embodiment, the antibody or binding fragment thereof is specific to an immunogenic composition comprising SEQ ID NO:84 or SEQ ID NO:83.

Antibodies or their binding fragments having variable heavy chain and variable light chain regions can also bind TcdA, wherein the variable heavy chain and variable light chain regions have at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, preferably about 98%, more preferably about 99%, or most preferably about 100% similarity to the variable heavy chain and light chain regions of A65-33, A60-22, A80-29 and/or preferably A3-25.

In one embodiment, the antibody or its antigen-binding fragment comprises a variable heavy chain region containing an amino acid sequence that is at least about 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of the variable heavy chain region of A3-25 shown in SEQ ID NO:37.

In another embodiment, the antibody or its antigen-binding fragment comprises a variable light chain region containing an amino acid sequence that is at least about 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of the variable light chain region of A3-25 shown in SEQ ID NO:36.

On the other hand, the antibody or its antigen-binding fragment comprises a variable heavy chain region containing an amino acid sequence that is at least about 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of the variable heavy chain region shown in SEQ ID NO:37; and comprises a variable light chain region containing an amino acid sequence that is at least about 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of the variable light chain region shown in SEQ ID NO:36.

In another embodiment, an antibody or its binding fragment having a complementarity-determining region (CDR) of a variable heavy chain and/or variable light chain having A65-33, A60-22, A80-29 and/or preferably A3-25 can bind to TcdA. The CDRs of the variable heavy chain region of A3-25 are shown in Table 4 below.

The CDR of the variable light chain region of A3-25 is shown in Table 5 below.

In one embodiment, the antibody or its binding fragment comprises the amino acid sequence of the heavy chain complementarity-determining region (CDR) shown in SEQ ID NO:41 (CDR H1), 42 (CDR H2), and 43 (CDR H3); and/or comprises the amino acid sequence of the light chain CDR shown in SEQ ID NO:38 (CDRL1), 39 (CDR L2), and 40 (CDR L3).

In one exemplary embodiment, an antibody specific to clostridial toxin A or a binding fragment thereof specifically binds to an epitope in the N-terminal region of TcdA, such as the epitope between amino acids 1-1256 of TcdA according to SEQ ID NO:1.

In a preferred embodiment, an antibody specific to Clostridium difficile toxin A, or a binding fragment thereof, specifically binds to an epitope in the C-terminal region of toxin A, such as the epitope between amino acids 1832-2710 of TcdA according to SEQ ID NO:1. Examples include A3-25, A65-33, A60-22, and A80-29.

In another embodiment, an antibody specific to Clostridium difficile toxin A or a binding fragment thereof specifically binds to an epitope in the "translocation" region of Clostridium difficile toxin A, for example preferably including epitopes of residues 956-1128 of TcdA according to SEQ ID NO:1, or epitopes between amino acids 659-1832 of TcdA according to SEQ ID NO:1.

In another aspect, the present invention relates to antibodies or binding fragments thereof specific to Clostridium difficile TcdB. For example, the antibody or binding fragment thereof may be specific to TcdB from any wild-type Clostridium difficile strain, as described above, for example, specific to SEQ ID NO:2. In another aspect, the present invention relates to antibodies or binding fragments thereof specific to the above-described immunogenic compositions. For example, in one embodiment, the antibody or binding fragment thereof is specific to an immunogenic composition comprising SEQ ID NO:6 or SEQ ID NO:8.

In another embodiment, the antibody or its binding fragment is specific to an immunogenic composition comprising SEQ ID NO:6 or SEQ ID NO:8, wherein at least one amino acid of SEQ ID NO:6 or SEQ ID NO:8 is cross-linked with formaldehyde, EDC, NHS, or a combination of EDC and NHS. In another embodiment, the antibody or its binding fragment is specific to an immunogenic composition comprising SEQ ID NO:86 or SEQ ID NO:85.

Monoclonal antibodies that specifically bind to TcdB include antibodies produced by the B2-31, B5-40, B70-2, B6-30, B9-30, B59-3, B60-2, B56-6 and/or preferably B8-26 clones described herein.

Antibodies or their binding fragments that can bind to TcdB may also include antibodies or their binding fragments having variable heavy chain and variable light chain regions that are at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, preferably about 98%, more preferably about 99%, or most preferably about 100% identical to the variable heavy chain and light chain regions of B2-31, B5-40, B70-2, B6-30, B9-30.

In one embodiment, the antibody or its antigen-binding fragment comprises a variable heavy chain region containing an amino acid sequence that is at least about 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of the variable heavy chain region of A3-25 shown in SEQ ID NO:49.

In one embodiment, the antibody or its antigen-binding fragment comprises a variable heavy chain region containing an amino acid sequence that is at least about 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of the variable heavy chain region of A3-25 shown in SEQ ID NO:60.

In one embodiment, the antibody or its antigen-binding fragment comprises a variable heavy chain region containing an amino acid sequence that is at least about 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of the variable heavy chain region of A3-25 shown in SEQ ID NO:71.

In another embodiment, the antibody or its antigen-binding fragment comprises a variable light chain region containing an amino acid sequence that is at least about 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of the variable light chain region of A3-25 shown in SEQ ID NO:55.

In another embodiment, the antibody or its antigen-binding fragment comprises a variable light chain region containing an amino acid sequence that is at least about 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of the variable light chain region of A3-25 shown in SEQ ID NO:66.

In another embodiment, the antibody or its antigen-binding fragment comprises a variable light chain region containing an amino acid sequence that is at least about 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of the variable light chain region of A3-25 shown in SEQ ID NO:77.

The amino acid sequence of the variable heavy chain of the neutralizing antibody (B8-26mAb) against Clostridium difficile TcdB is shown in SEQ ID NO:49. See Table 25-a.

The amino acid sequence of the variable light chain of the neutralizing antibody (B8-26mAb) against Clostridium difficile TcdB is shown in SEQ ID NO:55. See Table 25-b.

In one embodiment, the antibody or its binding fragment comprises the amino acid sequence of the heavy chain CDR shown in SEQ ID NO:51 (CDR H1), 52 (CDR H2) and 53 (CDR H3), and/or comprises the amino acid sequence of the light chain CDR shown in SEQ ID NO:57 (CDR L1), 58 (CDRL2) and 59 (CDR L3).

The amino acid sequence of the variable heavy chain of the neutralizing antibody (B59-3mAb) against Clostridium difficile TcdB is shown in SEQ ID NO:60. See Table 26-a.

The amino acid sequence of the variable light chain of the neutralizing antibody (B59-3mAb) against Clostridium difficile TcdB is shown in SEQ ID NO:66. See Table 26-b.

In one embodiment, the antibody or its binding fragment comprises the amino acid sequence of the heavy chain CDR shown in SEQ ID NO:62 (CDR H1), 63 (CDR H2) and 64 (CDR H3), and/or comprises the amino acid sequence of the light chain CDR shown in SEQ ID NO:68 (CDR L1), 69 (CDRL2) and 70 (CDR L3).

The amino acid sequence of the variable heavy chain of the neutralizing antibody (B9-30mAb) against Clostridium difficile TcdB is shown in SEQ ID NO:71. See Table 27-a.

The amino acid sequence of the variable light chain of the neutralizing antibody (B9-30mAb) against Clostridium difficile TcdB is shown in SEQ ID NO:77. See Table 27-b.

In one embodiment, the antibody or its binding fragment comprises the amino acid sequence of the heavy chain CDR shown in SEQ ID NO:73 (CDR H1), 74 (CDR H2) and 75 (CDR H3), and/or comprises the amino acid sequence of the light chain CDR shown in SEQ ID NO:79 (CDR L1), 80 (CDRL2) and 81 (CDR L3).

In one aspect, the present invention relates to antibodies or their binding fragments specific to wild-type Clostridium difficile TcdB from any Clostridium difficile strain, such as those described above, for example, specific to SEQ ID NO:2. In another aspect, the present invention relates to antibodies or their binding fragments specific to the immunogenic compositions described above. For example, in one embodiment, the antibody or its binding fragment is specific to an immunogenic composition comprising SEQ ID NO:6 or SEQ ID NO:8. In another embodiment, the antibody or its binding fragment is specific to an immunogenic composition comprising SEQ ID NO:6 or SEQ ID NO:8, wherein at least one amino acid of SEQ ID NO:6 or SEQ ID NO:8 is cross-linked with formaldehyde, EDC, NHS, or a combination of EDC and NHS.

Antibodies or their binding fragments having variable heavy chain and variable light chain regions can also bind TcdB, wherein the variable heavy chain and variable light chain regions have at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, preferably about 98%, more preferably about 99%, or most preferably about 100% similarity to the variable heavy chain and light chain regions of B2-31, B5-40, B70-2, B6-30, B9-30, B59-3, B60-2, B56-6 and/or preferably B8-26.

In one embodiment, the antibody or its antigen-binding fragment comprises a variable heavy chain region containing an amino acid sequence that is at least about 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of the variable heavy chain region of B8-26 (SEQ ID NO:49).

In another embodiment, the antibody or its antigen-binding fragment comprises a variable light chain region containing an amino acid sequence that is at least about 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of the variable light chain region of B8-26 (SEQ ID NO: 55).

On the other hand, the antibody or its antigen-binding fragment contains a variable heavy chain region whose amino acid sequence is at least about 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to the amino acid sequence of the variable heavy chain region of B8-26 (SEQ ID NO: 49). Or 100% identical; and containing a variable light chain region whose amino acid sequence is at least about 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of the variable light chain region of B8-26 (SEQ ID NO: 55).

In another embodiment, an antibody or its binding fragment thereof having a CDR of a variable heavy chain and/or a variable light chain having B2-31, B5-40, B70-2, B6-30, B9-30, B59-3, B60-2, B56-6 and/or preferably B8-26 may also bind to TcdB.

In one embodiment, the antibody or its binding fragment comprises the amino acid sequence of the heavy chain complementarity-determining region (CDR) of B8-26, and/or the amino acid sequence of the light chain CDR of B8-26.

In a preferred embodiment, an antibody specific to Clostridium difficile toxin B, or a binding fragment thereof, specifically binds to an epitope in the N-terminal region of toxin B, such as an epitope between amino acids 1-1256 of TcdB according to SEQ ID NO:2. Examples include B2-31, B5-40, B8-26, B70-2, B6-30, and B9-30.

In an exemplary embodiment, an antibody specific to Clostridium difficile toxin B or a binding fragment thereof specifically binds to an epitope in the C-terminal region of toxin B, such as the epitope between amino acids 1832-2710 of TcdB according to SEQ ID NO:2.

In another embodiment, an antibody specific to Clostridium difficile toxin B, or a binding fragment thereof, specifically binds to an epitope in the "translocation" region of Clostridium difficile toxin B, preferably including epitopes of residues 956-1128 of TcdB according to SEQ ID NO:2, such as epitopes between amino acids 659-1832 of TcdB. Examples include B59-3, B60-2, and B56-6.

Antibody combination

Antitoxin antibodies or their binding fragments can be administered in combination with other anti-clostridium difficile toxin antibodies (e.g., other monoclonal antibodies, polyclonal gamma globulins) or their antigen-binding fragments. Possible combinations include antitoxin A antibodies or their binding fragments, and antitoxin B antibodies or their antigen-binding fragments.

In another embodiment, the combination comprises an antitoxin A antibody or a binding fragment thereof and another antitoxin A antibody or an antigen-binding fragment thereof. Preferably, the combination comprises a neutralizing antitoxin A monoclonal antibody or a binding fragment thereof and another neutralizing antitoxin A monoclonal antibody or a binding fragment thereof. Surprisingly, the inventors have discovered a synergistic effect of such combinations in the neutralization of toxin A cytotoxicity. For example, the combination comprises at least two of the following neutralizing antitoxin A monoclonal antibodies: A3-25; A65-33; A60-22; and A80-29. More preferably, the combination comprises antibody A3-25 and at least one of the following neutralizing antitoxin A monoclonal antibodies: A65-33; A60-22; and A80-29. Most preferably, the combination comprises all four antibodies: A3-25; A65-33; A60-22; and A80-29.

In another embodiment, the combination comprises an antitoxin B antibody or a binding fragment thereof and another antitoxin B antibody or an antigen-binding fragment thereof. Preferably, the combination comprises a neutralizing antitoxin B monoclonal antibody or a binding fragment thereof and another neutralizing antitoxin B monoclonal antibody or an antigen-binding fragment thereof. Surprisingly, the inventors have discovered a synergistic effect of such a combination in the neutralization of toxin B cytotoxicity. More preferably, the combination comprises a combination of at least two of the following neutralizing antitoxin B monoclonal antibodies: B8-26; B9-30 and B59-3. Most preferably, the combination comprises all three antibodies: B8-26; B9-30 and B59-3.

In another embodiment, the combination includes an antitoxin B antibody or a binding fragment thereof and another antitoxin B antibody or an antigen-binding fragment thereof. As previously described, the inventors have discovered that, in the respective neutralization of toxin A and toxin B, a combination of at least two neutralizing monoclonal antibodies can exhibit an unexpected synergistic effect.

In another embodiment, the substances of the present invention can be formulated as mixtures, or as covalently linked antibodies (or covalently linked antibody fragments) chemically or genetically linked using techniques known in the art, thereby resulting in the binding properties of both antitoxin A and antitoxin B. The formulation of combinations can be guided by determining one or more parameters, such as affinity, affinity power, or biological potency, either individually or in combination with another substance.

Such combination therapies are preferably additive and/or synergistic in their therapeutic activity, for example, in the suppression, prevention (e.g., prevention of relapse), and/or treatment of Clostridium difficile-associated diseases or conditions. Administration of such combination therapies can reduce the dosage of the therapeutic substance (e.g., a mixture of antibodies or antibody fragments, or cross-linked or genetically fused bispecific antibodies or antibody fragments) required to achieve the desired effect.

It should be understood that any composition of the present invention, such as antitoxin A and/or antitoxin B antibodies or their respective antigen-binding fragments, can be combined in different proportions or amounts for therapeutic effects. For example, antitoxin A and antitoxin B antibodies or their respective binding fragments may be present in the composition in a ratio ranging from 0.1:10 to 10:0.1 (A:B). In another embodiment, antitoxin A and antitoxin B antibodies or their respective binding fragments may be present in the composition in a ratio ranging from 0.1:10 to 10:0.1 (B:A).

In another aspect, the present invention relates to a method for generating neutralizing antibodies against Clostridium difficile TcdA. The method comprises administering the immunogenic composition described above to a mammal and recovering the antibody from the mammal. In a preferred embodiment, the immunogenic composition comprises mutant Clostridium difficile TcdA having SEQ ID NO:4, wherein at least one amino acid of said mutant Clostridium difficile TcdA is chemically cross-linked, preferably via methyl or 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide. Exemplary neutralizing antibodies against TcdA that can be generated include A65-33; A60-22; A80-29 and/or A3-25.

In another aspect, the present invention relates to a method for generating neutralizing antibodies against Clostridium difficile TcdB. The method comprises administering the immunogenic composition described above to a mammal and recovering the antibody from the mammal. In a preferred embodiment, the immunogenic composition comprises mutant Clostridium difficile TcdB having SEQ ID NO:6, wherein at least one amino acid of said mutant Clostridium difficile TcdB is chemically cross-linked, preferably via methyl or 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide. Exemplary neutralizing antibodies against TcdB that can be generated include B2-31; B5-40, B70-2; B6-30; B9-30; B59-3; B60-2; B56-6; and/or B8-26.

preparation

The compositions of the present invention (e.g., compositions comprising mutant Clostridium difficile toxins, immunogenic compositions, antibodies, and/or antibody-binding fragments thereof as described herein) can be in a variety of forms. These forms include, for example, semi-solid and solid dosage forms, suppositories, liquid forms such as liquid solutions (e.g., injectable and perfusionable solutions), dispersions or suspensions, liposomes, and/or dried forms such as, for example, lyophilized powders, lyophilized forms, spray-dried forms, and/or foam-dried forms. For suppositories, the binder and carrier include, for example, polyalkylene glycols or triglycerides; such suppositories can be formed from mixtures containing the compositions of the present invention. In exemplary embodiments, the compositions are in the form of a solution or suspension suitable for use prior to injection in a liquid carrier. In further exemplary embodiments, the compositions are emulsified or encapsulated in liposomes or microparticles, such as polylactic acid compounds, polyglycolates, or copolymers.

In a preferred embodiment, the composition is lyophilized and immediately reconstituted before use.

In one aspect, the present invention relates to pharmaceutical compositions comprising any composition described herein formulated with a pharmaceutically acceptable carrier (such as, for example, compositions comprising mutant Clostridium difficile toxins, immunogenic compositions, antibodies, and/or antibody-binding fragments thereof described herein). “Pharmaceutically acceptable carrier” includes any physiologically suitable solvent, dispersion medium, stabilizer, diluent, and/or buffer.

Exemplary stabilizers include carbohydrates such as sorbitol, mannitol, starch, dextran, sucrose, trehalose, lactose, and/or glucose; inert proteins such as albumin and/or casein; and/or other large, slowly metabolizing macromolecules such as polysaccharides, such as chitosan, polylactic acid, polyglycolic acid, and copolymers (e.g., latex-functionalized SEPHAROSE ^™ agarose, agarose, cellulose, etc.), amino acids, polyamino acids, amino acid copolymers, and lipid aggregates (e.g., oil droplets or liposomes). Additionally, these carriers can be used as immunostimulants (i.e., adjuvants).

Preferably, the composition comprises trehalose. The preferred amount (wt%) of trehalose ranges from a minimum of about 1%, 2%, 3%, or 4% to a maximum of about 10%, 9%, 8%, 7%, 6%, or 5%. Any minimum value can be combined with any maximum value to define a suitable range. In one embodiment, the composition comprises about 3%-6% trehalose, most preferably 4.5% trehalose, for example, in a 0.5 mL dose.

Examples of suitable diluents include distilled water, saline, physiological phosphate buffered saline, glycerol, alcohols (such as ethanol), Ringer's solution, dextrose solution, Hanks' balanced salt solution, and/or lyophilized excipients.

Exemplary buffers include phosphates (such as calcium phosphate, sodium phosphate); acetates (such as sodium acetate); succinates (such as sodium succinate); glycine; histidine; carbonates, Tris (tris(hydroxymethyl)aminomethane), and/or bicarbonates (such as ammonium bicarbonate) buffers. Preferably, the composition includes a Tris buffer. Preferred amounts of Tris buffers range from a minimum of about 1 mM, 5 mM, 6 mM, 7 mM, 8 mM, 9 mM, 10 mM to a maximum of about 100 mM, 50 mM, 20 mM, 19 mM, 18 mM, 17 mM, 16 mM, 15 mM, 14 mM, 13 mM, 12 mM, or 11 mM. Any minimum value can be combined with any maximum value to define a suitable range. In one embodiment, the composition comprises approximately 8 mM to 12 mM of Tris buffer, most preferably 10 mM of Tris buffer, for example, in a dose of 0.5 mL.

In another preferred embodiment, the composition comprises a histidine buffer. Preferred amounts of the histidine buffer range from a minimum of about 1 mM, 5 mM, 6 mM, 7 mM, 8 mM, 9 mM, 10 mM to a maximum of about 100 mM, 50 mM, 20 mM, 19 mM, 18 mM, 17 mM, 16 mM, 15 mM, 14 mM, 13 mM, 12 mM, or 11 mM. Any minimum value can be combined with any maximum value to define a suitable range. In one embodiment, the composition comprises about 8 mM to 12 mM of histidine buffer, most preferably 10 mM, for example, in a 0.5 mL dose.

In another preferred embodiment, the composition comprises a phosphate buffer. Preferred amounts of the phosphate buffer range from a minimum of about 1 mM, 5 mM, 6 mM, 7 mM, 8 mM, 9 mM, 10 mM to a maximum of about 100 mM, 50 mM, 20 mM, 19 mM, 18 mM, 17 mM, 16 mM, 15 mM, 14 mM, 13 mM, 12 mM, or 11 mM. Any minimum value can be combined with any maximum value to define a suitable range. In one embodiment, the composition comprises about 8 mM to 12 mM of phosphate buffer, most preferably 10 mM, for example, in a 0.5 mL dose.

The pH of the buffer is typically chosen to stabilize the selected active material, and this pH can be determined by methods known to those skilled in the art. Preferably, the pH of the buffer is within the physiological pH range. Thus, a preferred pH range is from about 3 to about 8; more preferably, from about 6.0 to about 8.0; even more preferably, from about 6.5 to about 7.5; and most preferably, from about 7.0 to about 7.2.

In some embodiments, the pharmaceutical composition may include a surfactant. Any surfactant is suitable, whether it is amphoteric, nonionic, cationic, or anionic. Exemplary surfactants include polyoxyethylene sorbitan ester surfactants (e.g., polysorbate 20 and/or polysorbate 80); polyoxyethylene fatty ethers derived from dodecyl alcohol, hexadecyl alcohol, octadecyl alcohol, and oleyl alcohol (known as Brij surfactants), such as triethylene glycol monododecyl ether (Brij 30); Triton X 100, or tert-octylphenoxypolyethoxyethanol; and sorbitan esters (commonly referred to as SPAN), such as sorbitan trioleate (Span 85) and sorbitan monolaurate, and combinations thereof. Preferred surfactants include polysorbate 80 (polyoxyethylene sorbitan monooleate).

Preferred amounts (by weight%) of polysorbate 80 range from a minimum of about 0.001%, 0.005%, or 0.01% to a maximum of about 0.010%, 0.015%, 0.025%, or 1.0%. Any minimum value can be combined with any maximum value to define a suitable range. In one embodiment, the composition comprises about 0.005% to 0.015% of polysorbate 80, most preferably 0.01% of polysorbate 80.

In an exemplary embodiment, the immunogenic composition comprises trehalose and Phosphate 80. In another exemplary embodiment, the immunogenic composition comprises a Tris buffer and polysorbate 80. In another exemplary embodiment, the immunogenic composition comprises a histidine buffer and polysorbate 80. In another exemplary embodiment, the immunogenic composition comprises a phosphate buffer and polysorbate 80.

In one exemplary embodiment, the immunogenic composition comprises trehalose, Tris buffer, and polysorbate 80. In another exemplary embodiment, the immunogenic composition comprises trehalose, histidine buffer, and polysorbate 80. In yet another exemplary embodiment, the immunogenic composition comprises trehalose, phosphate buffer, and polysorbate 80.

The compositions described herein may also include components of petroleum, animal, plant, or synthetic origin, such as peanut oil, soybean oil, and/or mineral oil. Examples include glycols such as propylene glycol or polyethylene glycol.

In some embodiments, the pharmaceutical composition further includes formaldehyde. For example, in a preferred embodiment, the pharmaceutical composition further including formaldehyde has an immunogenic composition, wherein the mutant Clostridium difficile toxin of the immunogenic composition has been contacted with a chemical cross-linking agent including formaldehyde. The amount of formaldehyde present in the pharmaceutical composition can vary from a minimum of about 0.001%, 0.002%, 0.003%, 0.004%, 0.005%, 0.006%, 0.007%, 0.008%, 0.009%, 0.010%, 0.013%, or 0.015%, to a maximum of about 0.020%, 0.019%, 0.018%, 0.017%, 0.016%, 0.015%, 0.014%, 0.013%, 0.012%, 0.011%, or 0.010%. Any minimum value can be combined with any maximum value to define a suitable range. In one embodiment, the pharmaceutical composition comprises approximately 0.010% formaldehyde.

In other embodiments, the pharmaceutical compositions described herein do not include formaldehyde. For example, in a preferred embodiment, the formaldehyde-free pharmaceutical composition has an immunogenic composition wherein at least one amino acid of the mutant Clostridium difficile toxin is chemically cross-linked with a substance including EDC. More preferably, in such embodiments, the mutant Clostridium difficile toxin has not been contacted with a chemical cross-linking agent including formaldehyde. As another exemplary embodiment, the lyophilized form of the pharmaceutical composition does not include formaldehyde.

In another embodiment, the composition described herein may include an adjuvant as described below. Preferred adjuvants enhance the intrinsic immune response to an immunogen without causing conformational changes in the immunogen that could affect the qualitative form of the immune response.

Exemplary adjuvants include 3De-O-acylated monophospholipid A (MPL ^™ ) (see GB 2220211 (GSK)); aluminum hydroxide gels such as Alhydrogel ^™ (Brenntag Biosector, Denmark); aluminum salts (such as aluminum hydroxide, aluminum phosphate, aluminum sulfate) which may be used in the presence or absence of immunostimulants such as MPL or 3-DMP, QS-21, and polymeric or monomeric amino acids such as polyglutamic acid or polylysine.

Other exemplary adjuvants are immunostimulatory oligonucleotides such as CpG oligonucleotides (see, for example, WO1998/040100, WO2010/067262), or saponins and immunostimulatory oligonucleotides such as CpG oligonucleotides (see, for example, WO00/062800). In a preferred embodiment, the adjuvant is a CpG oligonucleotide, most preferably a CpG oligodeoxynucleotide (CpGODN). The preferred CpG ODN is class B, which preferentially activates B cells. For the present invention, the CpG ODN has the amino acid sequence 5’T*C*G*T*C*G*T*T*T*T*T*C*G*T*T*T*T*T 3’ (SEQ ID NO:48) where * denotes a phosphate thioester bond. The CpG ODN with this sequence is known as CpG 24555, which is described in WO2010/067262. In a preferred embodiment, CpG 24555 is used in conjunction with aluminum hydroxide salts such as aluminum gel.

Another class of exemplary adjuvants includes saponin adjuvants, such as Stimulon ^™ (QS-21, which is a triterpenoid glycoside or saponin, Aquila, Framingham, Mass.) or particles derived therefrom, such as ISCOM (immunostimulatory complexes), and adjuvants. Therefore, the compositions of the present invention can be provided as ISCOM, CTB-containing ISCMS, liposomes, or encapsulated compounds such as acrylates or poly(DL-lactide-co-glycosides) to form microspheres suitable for adsorption size. Generally, the term "ISCOM" refers to an immunogenic complex formed between glycosides, such as triterpenoid saponins (especially Quil A), and an antigen containing a hydrophobic region. In a preferred embodiment, the adjuvant is the ISCMATRIX adjuvant.

Other exemplary adjuvants include RC-529, GM-CSF, and complete Freund's adjuvant (CFA) and incomplete Freund's adjuvant (IFA).

Another class of exemplary adjuvants are glycolipid analogues including N-glycosylamides, N-glycosylurea, and N-glycosylcarbamates/esters, each of which has an amino acid substituted in the sugar residue.

Optionally, the pharmaceutical composition comprises two or more different adjuvants. Preferred adjuvant combinations include, for example, any combination containing at least two of the following adjuvants: alum, MPL, QS-21, ISCOMATRIX, CpG, and aluminum gel. An exemplary adjuvant combination includes a combination of CpG and aluminum gel.

Alternatively, in one embodiment, the composition is administered to mammals without adjuvant.

The compositions described herein can be administered via any route of administration, such as, for example, parenteral, topical, intravenous, mucosal, oral, subcutaneous, intraarterial, intracranial, intradural, intraperitoneal, intranasal, intramuscular, intradermal, infusion, rectal, and/or percutaneous routes for prophylactic and/or therapeutic applications. In a preferred embodiment, the composition is administered via parenteral administration, more preferably intramuscular administration. Typical intramuscular administration is performed in the muscles of the arm or leg.

The compositions described herein can be administered in combination with other therapies that are at least partially effective in the prevention and/or treatment of Clostridium difficile infection. For example, the compositions of the present invention can be administered before, concurrently with, or after the following therapies: biotherapy; probiotic therapy; anal implantation; immunotherapy (such as intravenous immunoglobulin); and/or accepted standards of care for antibiotic treatment of Clostridium difficile-associated disease (CDAD), such as metronidazole and/or vancomycin.

The compositions of the present invention involving toxin A and toxin B can be administered to mammals in any combination. For example, an immunogenic composition comprising mutant Clostridium difficile TcdA can be administered before, simultaneously with, or after an immunogenic composition comprising mutant Clostridium difficile TcdB. Conversely, an immunogenic composition comprising mutant Clostridium difficile TcdB can be administered before, simultaneously with, or after an immunogenic composition comprising mutant Clostridium difficile TcdA.

In another embodiment, an immunogenic composition comprising an antitoxin A antibody or a binding fragment thereof may be administered before, simultaneously with, or after an immunogenic composition comprising an antitoxin B antibody or a binding fragment thereof. Conversely, an immunogenic composition comprising an antitoxin B antibody or a binding fragment thereof may be administered before, simultaneously with, or after an immunogenic composition comprising an antitoxin A antibody or a binding fragment thereof.

In another embodiment, the compositions of the present invention may be administered before, simultaneously with, or after a pharmaceutically acceptable carrier. For example, an adjuvant may be administered before, simultaneously with, or after a composition comprising a mutant Clostridium difficile toxin. Therefore, the compositions of the present invention and pharmaceutically acceptable carriers may be packaged in the same vial or they may be packaged in separate vials and mixed before use. The compositions may be formulated for single-dose administration and/or multiple-dose administration.

Methods for the protection and/or treatment of Clostridium difficile infection in mammals

In one aspect, the present invention relates to a method for inducing an immune response against Clostridium difficile toxins in mammals. The method comprises administering an effective amount of the composition described herein to the mammal. For example, the method may include administering an effective amount to generate an immune response against each Clostridium difficile toxin in the mammal.

In an exemplary embodiment, the present invention relates to a method for inducing an immune response against Clostridium difficile TcdA in mammals. The method includes administering to the mammal an effective amount of an immunogenic composition comprising mutant Clostridium difficile TcdA. In another exemplary embodiment, the present invention relates to a method for inducing an immune response against Clostridium difficile TcdB in mammals. The method includes administering to the mammal an effective amount of an immunogenic composition comprising mutant Clostridium difficile TcdB.

In another embodiment, the method includes administering to a mammal an effective amount of an immunogenic composition comprising mutant Clostridium difficile TcdA and an effective amount of an immunogenic composition comprising mutant Clostridium difficile TcdB. In another aspect, the compositions described herein can be used to treat, prevent, reduce the risk of, reduce the incidence of, reduce the severity of, and/or delay the onset of Clostridium difficile infection, Clostridium difficile-associated disease (CDAD), syndromes, conditions, symptoms, and/or complications in mammals (compared to mammals not treated with the compositions). The method includes administering an effective amount of the compositions to a mammal.

Based on the severity of the infection, three clinical syndromes caused by Clostridium difficile infection were identified. The most severe form is pseudomembranous colitis (PMC), characterized by profuse diarrhea, abdominal pain, signs of systemic disease, and distinctive colonic endoscopic presentation.

Antibiotic-associated colitis (AAC) is also characterized by fulminant bowel movements, abdominal pain and tenderness, systemic signs (e.g., fever), and leukocytosis. Intestinal damage in AAC is less severe than in PMC, which lacks the characteristic colonic endoscopy features and has a lower mortality rate.

Finally, antibiotic-associated diarrhea (AAD, also known as Clostridium difficile-associated diarrhea (CDAD)) is a relatively mild syndrome characterized by mild to moderate diarrhea, lack of large bowel inflammation (characterized by, for example, abdominal pain and tenderness) and signs of systemic infection (e.g., fever).

These three different devices typically occur with increasing frequency. That is, PMC usually occurs less frequently than AAC, while AAD is usually the most common clinical manifestation of Clostridium difficile disease.

A common complication of Clostridium difficile infection is recurrent or relapsed disease, occurring in up to 20% of all patients who recover from Clostridium difficile disease. Relapses are clinically characterized by AAD, AAC, or PMC. Patients who have experienced one relapse are more likely to experience another.

As used in this article, Clostridium difficile infection status includes, for example, mild, mild to moderate, moderate, and severe Clostridium difficile infection. The status of Clostridium difficile infection can vary depending on the presentation and/or severity of symptoms.

Symptoms of Clostridium difficile infection may include physiological, biochemical, histological, and/or behavioral symptoms, such as, for example, diarrhea; colitis; spastic colitis, fever, leukocytosis in stool, and inflammation on colonic biopsy; pseudomembranous colitis; hypoalbuminemia; generalized edema; leukocytosis; sepsis; abdominal pain; asymptomatic carriers; and/or complications and intermediate pathological phenotypes that occur during the development of the infection, as well as combinations thereof. Therefore, for example, administration of an effective amount of the composition described herein can, for example, treat, prevent, reduce the risk of, reduce the occurrence of, reduce the severity of, and/or delay the onset of diarrhea; abdominal pain, cramps, fever, inflammation on colonic biopsy, hypoalbuminemia, generalized edema, leukocytosis, sepsis, and/or asymptomatic carriers (compared to mammals not treated with the composition).

Risk factors for Clostridium difficile infection may include, for example, the ongoing or immediate use of antimicrobial agents (covering any antimicrobial substance with an antimicrobial spectrum and/or activity against anaerobic bacteria, including, for example, vitamins that disrupt the normal colonic microbiota, such as clindamycin, cephalosporins, metronidazole, vancomycin, fluorescent quinolones (including levofloxacin, moxifloxacin, gatifloxacin, and ciprofloxacin), linezolid, etc.); the ongoing or immediate discontinuation of prescription metronidazole or vancomycin; and the ongoing or immediate contact with a healthcare facility. (e.g., hospitals, long-term care facilities, etc.) and nursing staff; those currently or immediately to be treated with proton pump inhibitors, H2 antagonists, and/or methotrexate, or combinations thereof; those currently suffering from or at risk of gastrointestinal diseases, such as inflammatory bowel disease; those who have previously, are currently, or are immediately to undergo gastrointestinal surgery or gastrointestinal management in mammals; those with a history of or recurrent Clostridium difficile infection and/or CDAD, for example, patients with one or more Clostridium difficile infections and/or CDAD; and those who are at least or over approximately 65 years of age.

In the methods described herein, the mammal can be any mammal, such as, for example, a mouse, a hamster, a primate, and a human. In a preferred embodiment, the mammal is a human. According to the invention, the human may include an individual who has previously exhibited Clostridium difficile infection, Clostridium difficile-associated disease, syndrome, condition, symptoms, and/or its complications; an individual currently exhibiting Clostridium difficile infection, Clostridium difficile-associated disease, syndrome, condition, symptoms, and/or its complications; and an individual at risk of Clostridium difficile infection, Clostridium difficile-associated disease, syndrome, condition, symptoms, and/or its complications.

Examples of individuals exhibiting symptoms of Clostridium difficile infection include individuals who have exhibited or are exhibiting the symptoms described above; individuals who have had or are currently having Clostridium difficile infection and/or Clostridium difficile-associated disease (CDAD); and individuals with recurrent Clostridium difficile infection and/or CDAD.

Examples of patients at risk of Clostridium difficile infection include individuals at risk of or currently undergoing planned antimicrobial use; individuals at risk of or currently undergoing discontinuation of prescription metronidazole or vancomycin; individuals at risk of or currently undergoing planned exposure to care facilities (e.g., hospitals, long-term care facilities, etc.) or care workers; and/or individuals at risk of or currently undergoing planned treatment with proton pump inhibitors, H2 antagonists, and/or methotrexate, or combinations thereof; individuals who have had or are currently experiencing gastrointestinal diseases such as inflammatory bowel disease; individuals who have undergone or are currently undergoing gastrointestinal surgery or treatment; and individuals who have had or are currently experiencing recurrent Clostridium difficile infection and/or CDAD, such as patients with one or more Clostridium difficile infections and/or CDAD; and individuals approximately 65 years of age or older. These at-risk patients may or may not currently show symptoms of Clostridium difficile infection.

In asymptomatic patients, prevention and/or treatment can be initiated at any age (e.g., around 10, 20, or 30 years old). However, in one implementation, it is not necessary to begin treatment before the patient reaches approximately 45, 55, 65, 75, or 85 years of age. For example, the composition described herein can be administered to asymptomatic individuals aged 50–85 years.

In one embodiment, a method for treating, preventing, reducing the risk of, reducing the incidence of, reducing the severity of, and/or delaying the onset of Clostridium difficile infection, Clostridium difficile-associated disease, syndrome, condition, symptoms, and/or its complications in mammals comprises administering an effective amount of the composition described herein to mammals in need, mammals at risk of Clostridium difficile infection, and/or mammals suspected of having Clostridium difficile infection. An effective amount includes, for example, an amount sufficient to treat, prevent, reduce the risk of, reduce the incidence of, reduce the severity of, and/or delay the onset of Clostridium difficile infection, Clostridium difficile-associated disease, syndrome, condition, symptoms, and/or its complications (compared to mammals not treated with the composition). Applying an effective amount of the composition of the present invention can, for example, treat, prevent, reduce the risk of, reduce the occurrence of, reduce the severity of, and/or delay the onset of diarrhea, abdominal pain, cramps, fever, inflammation on colon biopsy, hypoalbuminemia, generalized edema, leukocytosis, sepsis, and/or asymptomatic carriers (compared to mammals not treated with the composition). In a preferred embodiment, the method comprises administering an effective amount of the immunogenic composition described herein to mammals in need, mammals at risk of Clostridium difficile infection, and/or mammals suspected of having Clostridium difficile infection.

In another embodiment, methods for treating, preventing, reducing the risk of, reducing the occurrence of, reducing the severity of, and/or delaying the onset of Clostridium difficile infection, Clostridium difficile-associated disease, syndrome, condition, symptoms, and/or its complications in mammals include administering an effective amount of the composition described herein to a mammal suspected of having or experiencing a Clostridium difficile infection. An effective amount includes, for example, an amount sufficient to treat, prevent, reduce the risk of, reduce the occurrence of, reduce the severity of, and/or delay the onset of Clostridium difficile infection, Clostridium difficile-associated disease, syndrome, condition, symptoms, and/or its complications (compared to a mammal not treated with the composition).

Administration of an effective amount of the composition can improve signs or symptoms of at least one Clostridium difficile infection in a subject, such as those described below. Administration of an effective amount of the composition described herein can, for example, reduce the severity and/or recurrence of diarrhea compared to mammals not treated with the composition; reduce the severity and/or recurrence of abdominal pain, cramps, fever, inflammation on colon biopsy, hypoalbuminemia, generalized edema, leukocytosis, sepsis, and/or asymptomatic carriers, etc. Optionally, the presence of infection symptoms, signs, and/or risk factors is determined prior to treatment. In a preferred embodiment, the method comprises administering an effective amount of the antibody and/or its binding fragment described herein to a mammal suspected of having or undergoing Clostridium difficile infection.

Therefore, an effective amount of the composition refers to an amount sufficient to achieve the desired effect (e.g., preventive and/or therapeutic effect) in the method of the present invention. For example, the amount of immunogen used for administration can vary from a minimum of about 1 μg, 5 μg, 25 μg, 50 μg, 75 μg, 100 μg, 200 μg, 500 μg, or 1 mg per injection to a maximum of about 2 mg, 1 mg, 500 μg, or 200 μg. Any minimum value can be combined with any maximum value to define a suitable range. Typically, about 10, 20, 50, or 100 μg of each immunogen is used per human injection.

The amount of the composition of the invention applied to the subject can depend on the severity of the infection and/or individual characteristics, such as overall health, age, sex, weight, and tolerance to the drug. It can also depend on the degree, severity, and type of the disease. The effective amount can also vary depending on factors such as the route of administration, target site, patient's physiological state, patient's age, whether the patient is human or animal, other therapies administered, and whether the treatment is prophylactic or therapeutic. Those skilled in the art can determine the appropriate amount based on these and other factors.

The effective dose may include one or more effective doses (e.g., 2, 3, 4 or more doses) used in the methods described herein. The effective dose may need to be gradually increased to optimize safety and efficacy.

A combination of dosage and dosing frequency sufficient to achieve prophylactic and/or therapeutic use is defined as a prophylactically effective or therapeutically effective regimen. In prophylactic and/or therapeutic regimens, the composition is typically administered more than one dose until a sufficient immune response is achieved. Typically, the immune response is monitored, and repeated doses are given if the immune response diminishes.

The composition can be administered in multiple doses over a period of time. Treatment can be monitored by measuring antibody, or activated T-cell or B-cell responses (to the therapeutic substance, e.g., immunogenic compositions including mutant Clostridium difficile toxins) over time. A decreased response indicates the need for a booster dose.

Example

Example 1: Identification of toxin-negative Clostridium difficile strains

To identify Clostridium difficile strains lacking toxin (A and B) genes and those expressing toxins, 13 Clostridium difficile strains were tested. For toxin A, the culture media of the 13 Clostridium difficile strains were tested by ELISA. Seven strains expressed toxin A: Clostridium difficile 14797-2, Clostridium difficile 630, Clostridium difficile BDMS, Clostridium difficile W1194, Clostridium difficile 870, Clostridium difficile 1253, and Clostridium difficile 2149. See Figure 3.

Six strains did not express toxin A and lacked the entire pathogenic locus: *Clostridium difficile* 1351 (ATCC43593 ^™ ), *Clostridium difficile* 3232 (ATCC BAA-1801 ^™ ), *Clostridium difficile* 7322 (ATCC 43601 ^™ ), *Clostridium difficile* 5036 (ATCC43603 ^™ ), *Clostridium difficile* 4811 (ATCC 3602 ^™ ), and *Clostridium difficile* VPI 11186 (ATCC 700057 ^™ ). VPI 11186 was selected based on its effectiveness in taking up plasmid DNA through binding.

The same 13 strains (PaLoc; Braun et al., Gene. 1996 Nov 28; 181(1-2):29-38.) were tested in multiplex PCR assays using primers other than the pathogenic locus. PCR results showed that DNA from the six toxin A-negative strains (via ELISA) did not amplify any genes from PaLoc (tcdA-tcdE). Flanking sequences from PaLoc (cdd3 and cdu2) were observed (data not shown).

Example 2: Inactivation of the sporulation pathway in Clostridium difficile VPI 11186

Knocking out spore formation in Clostridium difficile producing strains facilitates large-scale fermentation in a safe production environment. The ClosTron system was used to create non-spore-producing Clostridium difficile strains. See Heap et al., J Microbiol Methods. 2009 Jul; 78(1):79-85. The ClosTron system was used to inactivate the target gene by site-directed insertional inactivation of group II introns of the spo0A1 Clostridium gene. Spore formation in the toxin-negative strain VPI11186 was inactivated using this Clostridium technique. Erythromycin-resistant mutants were selected and the presence of the insertion cassette (not shown) was confirmed by PCR. Loss of spore formation ability was confirmed in two independent clones.

Example 3: Genetic modification of toxin A and B genes to inactivate cytotoxic function

Full-length mutant toxins A and B open reading frames (ORFs) based on the genome sequence of strain 630Δ were designed for synthesis at Blue Heron Biotech. See, for example, SEQ ID NO:9-14. The active sites responsible for the cytotoxic glucosyltransferase activity were altered by two allelic substitutions: D285A/D287A for toxin A (see SEQ ID NO:3) and D286A/D288A for toxin B (see SEQ ID NO:5). Two nucleotides were mutated at each aspartic (D) codon to create an alanine (A) codon. See, for example, SEQ ID NO:9-14. Additionally, a pair of vectors expressing mutant toxins lacking cysteine residues were constructed at Blue Heron Biotech according to custom synthesis. Seven cysteine residues from mutant toxin A and nine cysteine residues from mutant toxin B were replaced with alanine. This substitution included the catalytic cysteine of the proteases catalyzed by toxins A and B. In addition, silencing mutations were introduced where needed to remove restriction enzyme sites used for vector construction.

Example 4: pMTL84121fdx expression vector

The plasmid shuttle vector used for expression of Clostridium difficile mutant toxin antigens was selected from the pMTL8000 series regulatory system developed by Minton Lab (see Heap et al., J Microbiol Methods. 2009 Jul; 78(1):79-85). The selected vector pMTL84121fdx contained the Clostridium difficile plasmid pCD6Gram+ replicon, catP (chloramphenicol/thiamphenicol) select marker, p15a Gram- replicon and tra function, as well as the Clostridium sporogenes feroxin promoter (fdx) and distal multiple cloning site (MCS). Empirical data showed that the low copy number p15a replicon conferred greater stability to Escherichia coli compared to the ColE1 select. The fdx promoter was chosen because it produces higher expression (compared to other promoters tested with the CAT report subconstruct, such as tcdA, tcdB; or heterogeneous tetR or xylR) (data not shown).

Example 5: Cloning the modified toxin ORF into pMTL84121fdx

Using standard molecular biology techniques, the pMTL84121fdx vector was used to perform multiple cloning of the NdeI and BglII sites, subcloning the full-length mutant toxins A and B open reading frames (ORFs) based on the strain 630Δ genome sequence. To facilitate cloning, the ORFs contain a proximal NdeI site with a start codon and a BglII site immediately downstream of the stop codon.

Example 6: Site-directed mutagenesis of TcdA to create triple mutants

The catalytic cysteine residues of the self-catalytic protease domain are substituted in SEQ ID NO:3 and 5 (i.e., in each "double mutant") (i.e., C700A for TcdA and C698A for TcdB). For mutagenesis of mutant toxin A, a 2.48 kb NdeI-HindIII fragment from the TcdA D285A/D287A expression plasmid was subcloned into pUC19 (after the same restriction enzyme digestion) and site-directed mutagenesis was performed on this template. Once new alleles were identified by DNA sequence analysis, the modified NdeI-HindIII fragment was reintroduced into the expression vector pMTL84121fdx to create "triple mutants", namely SEQ ID NO:4 and SEQ ID NO:6.

Example 7: Site-directed mutagenesis of TcdB to create triple mutants

For mutagenesis of mutant toxin B, a 3.29kb NdeI-EcoNI fragment from the mutant toxin B plasmid was modified and reintroduced. Since the EcoNI site was absent in the available cloning vectors, a slightly larger 3.5kb NdeI-EcoRV fragment was subcloned into pUC19 (prepared with NdeI-SmaI). After mutagenesis, the modified internal 3.3kb NdeI-EcoNI fragment was cut and used to replace the corresponding mutant toxin B expression vector pMTL84121fdx for evaluation. Because this targeted strategy was found to have low cloning efficiency, another strategy was simultaneously attempted to introduce the C698A allele (involving a 1.5kb DraIII replacement). Both strategies independently produced the desired recombinants.

Example 8: Creating additional mutant toxin variants through site-directed mutagenesis

At least 12 different Clostridium difficile mutant toxin variants were constructed. Allelic substitutions were introduced into the N-terminal mutant toxin gene fragment via site-directed mutagenesis (kit). The recombinant toxins were also modified as a reference control to evaluate the ability of this plasmid-based system to produce proteins (quantitatively equivalent in bioactivity to natural toxins purified from wild-type Clostridium difficile strains). In this case, allelic substitutions were introduced to restore the original glucosyltransferase substitution. Additionally, a pair of cysteine-free mutant toxin vectors were custom-synthesized at Blue Heron Biotech.

The 12 toxin variants include: (1) a mutant Clostridium difficile toxin A with the D285A/D287A mutation (SEQ ID NO:3); (2) a mutant Clostridium difficile toxin B with the D286A/D288A mutation (SEQ ID NO:5); (3) a mutant Clostridium difficile toxin A with the D285A/D287A C700A mutation (SEQ ID NO:4); (4) a mutant Clostridium difficile toxin B with the D286A/D288AC698A mutation (SEQ ID NO:6); (5) a recombinant toxin A with SEQ ID NO:1; (6) a recombinant toxin B with SEQ ID NO:2; (7) a mutant Clostridium difficile toxin A with the C700A mutation; (8) a mutant Clostridium difficile toxin B with the C698A mutation; (9) (10) Mutant Clostridium difficile toxin A with mutations C700A, C597S, C1169S, C1407S, C1623S, C2023S, and C2236S; (11) Mutant Clostridium difficile toxin B with mutations C698AC395S, C595S, C824S, C870S, C1167S, C1625S, C1687S, and C2232S; The mutant Clostridium difficile toxin A (SEQ ID NO:7) having mutations of D285A, D287A, C700A, D269A, R272A, E460A, and R462A; and (12) the mutant Clostridium difficile toxin B (SEQ ID NO:8) having mutations of D270A, R273A, D286A, D288A, D461A, K463A, and C698A.

Example 9: Stability of the transformant

Rearranged plasmids were obtained using a common DH5α Escherichia coli laboratory strain. Conversely, transformation with an Invitrogen Stbl2 ^™ E. coli host after 3 days of growth at 30°C on LB chloramphenicol (25 μg/ml) plates yielded slow-growing, full-length mutant toxin recombinants. Lower cloning efficiencies were obtained with the associated Stbl3 ^™ and Stbl4 ^™ E. coli strains, although these lines were found to be stable for plasmid retention. Transformants were subsequently passaged at 30°C in agar or liquid culture with chloramphenicol selection. The use of LB (Miller's) medium also improved transformant recovery and growth (compared to a medium based on animal-free tryptone soybean).

Example 10: Transformation of Clostridium difficile with pMTL84121fdx encoding wild-type or genetically mutant toxin gene

Transformation of *Clostridium difficile* using *E. coli* conjugation transfer was performed as described in Heap et al., *Journal of Microbiological Methods*, 2009, 78(1): p.79-85. *pMTL84121fdx* encoding the wild-type or mutant toxin gene was used to transform *E. coli* host CA434. *C. coli* host CA434 was an intermediate for immobilizing the expression plasmid into the *C. difficile* producing strain VPI 11186spo0A1. CA434 is a derivative of *E. coli* HB101. This strain carries the Tra+Mob+R702 conjugation plasmid, which confers resistance to Km, Tc, Su, Sm/Spe, and Hg (due to Tn1831). Chemocompetent or electrocompetent CA434 cells were prepared, and the expression vector transformants were selected on Miller's LB CAM plates at 30°C. Slow-growing colonies that appeared after 3 days were picked and amplified to mid-log phase in 3 mL LB chloramphenicol culture (~24 h, 225 rpm, orbital shaker, 30 °C). The *E. coli* culture was collected by slow (5,000 g) centrifugation to avoid damaging the pili, and the cell pellet was gently resuspended in 1 mL PBS using a wide-bore pipette. The cells were concentrated by slow centrifugation. Most of the PBS was removed by decanting, and the decanted pellet was transferred to an anaerobic incubator and resuspended in 0.2 mL of *Clostridium difficile* culture. The pellet was then spotted onto BHIS agar plates and allowed to grow for 8 h or overnight. In the case of mutant toxin A transformants, overnight conjugation yielded better results. Cell clumps were scraped into 0.5 mL PBS, and 0.1 mL was plated onto BHIS selective medium supplemented with 15 μg/mL thiamphenicol (a more potent chloramphenicol analog) and D-cycloserine/cefoxitin to kill the *E. coli* donor cells. Transformants appearing 16–24 h later were purified by re-stripering onto new BHIS (with supplements) plates, and subsequent cultures were tested for expression of the recombinant or mutant toxins. Glycerol permanent stock and seed stock were prepared from clones showing good expression. Small-scale plasmid preparations were also performed from 2 mL cultures using a modified Qiagen kit procedure (with cells pretreated with lysozyme, which was optional). The Clostridium difficile DNA was used as a template for PCR sequencing to verify clonal integrity. Alternatively, large-scale plasmid DNA was prepared from E. coli Stbl2 ^™ transformants and sequenced.

Example 11: Clostridium difficile expression analysis of toxin A and B triple mutants (SEQ ID NO:4 and 6, respectively) and heptavalent B mutant (SEQ ID NO:8).

Growing transformants in 2 mL culture (for routine analysis) or in a stopcock flask (for time-course experiments). Centrifuging (10,000 rpm, 30 s) of the sample (2 mL) to concentrate cells: discard the supernatant and concentrate 10x (Amicon-ultra 30k); dehydrate the pellet and freeze at -80°C. Thaw the cell pellet on ice, resuspend in 1 mL of hydrolysis buffer (Tris-HCl pH 7.5; 1 mM EDTA, 15% glycerol) and sonicate (impact with a microtip 1x 20 s). Centrifuging the hydrolysate at 4°C and concentrating the supernatant 5-fold. Combining the supernatant and hydrolysate with sample buffer and heating (10 min, 80°C), then loading onto duplicate 3-8% Tris-acetic acid SDS-PAGE gels (Invitrogen). One gel was stained with Coomassie Brilliant Blue, and the second gel was electroblotted for Western blotting analysis. Rabbit antisera specific to toxin A and toxin B (Fitgerald; Biodesign) were used to detect mutant toxin A and B variants. Expression of the heptafold mutant toxin B (SEQ ID NO: 8) was also confirmed by Western blot hybridization.

Example 12: Elimination of mutant toxin glucosyltransferase activity

The genetically modified (DM) toxins A and B (SEQ ID NO: 3 and 5, respectively) and the triple mutant ( ^TM ) toxins A and B (SEQ ID NO: 4 and 6, respectively) did not transfer 14 C-glucose to ¹⁰ μg of RhoA, Rac1, and Cdc42 GTPase in an in vitro glucose assay (in the presence of UDP- ¹⁴ C-glucose [30 μM], 50 mM HEPES, pH 7.2, ₁₀₀ mM KCl, ₄ mM MgCl2, 2 mM MnCl2, 1 mM DTT, and 0.1 μg/μL BSA). However, the wild-type A and B toxin controls (with SEQ ID NO: 1 and 2, respectively) effectively transferred ^14C -glucose to GTPase at low doses (and even lower doses - data not shown) of 10 and 1 ng each (Figs. 4A and 4B), even in the presence of 100 μg of the mutant toxin (Fig. 4B), indicating a reduction of at least 100,000-fold compared to their respective wild-type toxins. Similar results were detected for Cdc42 GTPase (data not shown).

Specifically, in Fig. 4B, wild-type toxin A and toxin B (1 ng) or triple mutant toxin A and triple mutant toxin B (100 μg) were incubated with hoA GTPase at 30 °C for 2 hours (in the presence of UDP- ¹⁴ C-glucose). As shown, 1 ng of wild-type TcdA and TcdB transferred ¹⁴ C-glucose to RhoA, but 100 μg of triple mutant toxin A and triple mutant toxin B did not. When 1 ng of wild-type TcdA or TcdB was added to the reaction of 100 μg of each of the triple mutant toxin A or triple mutant toxin B, glucosylation of RhoA was detected, indicating a lack of glucosylation inhibitors. Sensitive data for the detection of glucosylation activity was assessed as 1 ng of wild-type toxin (1:100,000 ratio) against a background of 100 μg of mutant toxin. The results showed that mutations in the glucosyltransferase active sites of triple mutant toxin A and triple mutant toxin B reduced any measurable glucosyltransferase activity (less than 100,000-fold lower than the activity of the respective wild-type toxins). Similar assays were also developed to qualitatively assess glucosyltransferase activity via TCA precipitation of glucosylated GTPase.

Example 13: Elimination of autocatalytic cysteine protease activity

When the cysteine protease cleavage site is mutated, the autocatalytic cleavage function in triple mutants A and (TM) (SEQ ID NO:4 and 6, respectively) is eliminated. As shown in Figure 5, wild-type (wt) toxins A and B (SEQ ID NO:3 and 5, respectively) are cleaved in the presence of inositol-6-phosphate. Double mutant toxins A and B (SEQ ID NO:3 and 5, respectively) are also cleaved in the presence of inositol-6-phosphate (data not shown), similar to the wild type. Toxin A (SEQ ID NO:3) is cleaved from 308 kDa into two fragments of 245 and 60 kDa. Toxin B (SEQ ID NO:5) is cleaved from 270 kDa into two fragments of 207 and 63 kDa. Triple mutants A and B (TM) (SEQ ID NO:4 and 6, respectively) remain unaffected at 308 and 270 kDa, respectively, even in the presence of inositol-6-phosphate. See Figure 5. Therefore, genetic modification inactivates cysteine protease activity.

More specifically, in Figure 5, 1 μg of triple mutant A and triple mutant B were incubated at room temperature (21 ± 5 °C) in parallel with wild-type TcdA and TcdB (from List Biologicals) for 90 min. The cleavage reaction was performed in 20 μL of Tris-HCl, pH 7.5, 2 mM DTT (with or without inositol-6-phosphate (TcdA 10 mM and TcdB 0.1 mM)). The entire reaction volume was then loaded into 3–8% SDS/PAGE; protein bands were made visible by silver staining. As shown, in the presence of inositol-6-phosphate, wt TcdA and cdB were cleaved into two protein bands of 245 kD and 60 kD, and 207 kD and 63 kD, respectively. The triple mutant toxin A and triple mutant toxin B were not cleaved, thus confirming that the C700A mutation in triple mutant toxin A and the C698A mutation in triple mutant toxin B blocked the cleavage.

Example 14: Residual cytotoxicity of triple mutant toxins A and B (SEQ ID NO: 4 and 6, respectively)

The cytotoxicity of the genetically modified toxins was assessed using an in vitro cytotoxicity assay (IMR90 cells, a human diploid lung fibroblast cell line). These cells were sensitive to both toxins A and B. As an additional preferred embodiment, Vero normal kidney cells from African green monkeys (Cercopithecus aethiops) were used for the cytotoxicity assay because they were observed to have reasonable sensitivity to both toxins A and B. Preferably, HT-29 human colorectal adenocarcinoma cells were not used for the cytotoxicity assay because they showed significantly reduced sensitivity to the toxins (compared to the Ver and IMR90 cell lines). See, for example, Table 6 below.

Serially diluted samples of the mutant toxin or wt toxin were added to monolayer cultures of cells grown in 96-well tissue culture plates. After incubation at 37°C for 72 h, the metabolic activity of the cells on the plates was assessed by measuring cellular ATP levels, expressed as relative light units (RLUs), generated by adding a luciferase-based reagent (Promega, Madison, WI). High RLUs indicated viable cells, while low RLUs indicated metabolically inactive cells that were rapidly dying. The level of cytotoxicity (expressed as _EC50 ) was defined as the amount of wt toxin or mutant toxin that elicited a 50% reduction in RLU (compared to the level in the cell culture control) (details of this assay are described below). As shown in Figures 6, Tables 7A and 8A, the _EC50 values for TcdA and TcdB were approximately 0.92 ng/mL and 0.009 ng/mL, respectively. The _EC50 values for triple mutant toxin A and triple mutant toxin B were approximately 8600 ng/mL and 74 ng/mL, respectively. Despite a cytotoxicity reduction of approximately 10,000-fold compared to the wt toxin, both genetically modified toxins demonstrated low levels of residual cytotoxicity. This residual cytotoxicity could be blocked with neutralizing antitoxin monoclonal antibodies, indicating specificity to the triple mutant toxins but unlikely to be associated with the enzymatic activity of known wt toxin enzymes (glucosylation or autoproteolysis).

Both wt toxins exhibited potent in vitro cytotoxicity; small amounts were sufficient to induce various effects on mammalian cells, such as cell rounding (cytopathic effect or CPE) and a lack of metabolic activity (measured by ATP levels). Therefore, two in vitro assays (CPE or cell rounding assay and ATP assay) were developed to verify that no residual cytotoxicity was retained in the mutant toxin drug substance. Results are expressed as _EC50 , which is the amount by which the toxin or mutant toxin induces either 1) 50% of cell development at CPE or 2) a 50% reduction in ATP levels (measured in relative light units).

In the CPE assay, samples of the drug substance were serially diluted and incubated with IMR90 cells to observe its potential cytopathic effect. The CPE assay was scored from 0 (normal cells) to 4 (~100% cell rounding), with a score of 2 (~50% cell rounding) defined as the _EC50 value of the tested sample. This method was used to test mutant toxin drugs at a concentration of 1000 μg/mL, which is the maximum tolerable concentration without matrix interference that can be tested in this assay. Therefore, no detectable cytotoxicity was reported as _EC50 >1000 μg/mL.

The ATP assay is based on the measurement of the amount of fluorescence signal generated from ATP, which is proportional to the amount of metabolically active cells. The maximum tolerable concentration (without assay interference) that can be tested in this assay is approximately 200 μg/mL. Therefore, no detectable cytotoxicity has been reported as EC ₅₀ >200 μg/mL.

Different concentrations of mutant toxins A and B were added to cells in parallel (with a toxin control). The endpoint was cell viability determined by cellular ATP levels (using Promega). The intensity of fluorescence was proportional to the ATP level or the number of surviving cells.

The in vitro cytotoxicity (EC ₅₀ ) of wild-type (wt) toxin A was 920 pg/mL, while that of toxin B was 9 pg/mL. The in vitro cytotoxicity (EC ₅₀ ) of mutant toxin A (SEQ ID NO:4) was 8600 ng/mL, while that of mutant toxin B (SEQ ID NO:6) was 74 ng/mL. Although these values represent a 9348-fold and 8222-fold reduction, respectively, residual cytotoxicity was detected in both mutant toxins.

In other words, in in vitro cytotoxicity assays in IMR-90 cells, the triple mutant toxins A and B (SEQ ID NO:4 and 6, respectively) showed significantly reduced cytotoxicity compared to wt toxins A and B (SEQ ID NO:1 and 2, respectively). As shown in Figure 6, although both triple mutant toxins exhibited a significant reduction in cytotoxicity ( ^10⁴- fold) relative to the wt toxins, residual cytotoxicity was observed at higher concentrations for both triple mutant toxins.

Additionally, the residual cytotoxicity of each triple mutant toxin can be completely neutralized (e.g., a reduction in toxicity of at least 6-8 log ₁₀ relative to the wild-type toxin) using toxin-specific antibodies. See Example 16 below.

Cell culture assays are more sensitive for detecting cytotoxicity than in vivo animal models. When delivered via intraperitoneal (ip) or intravenous (iv) to a lethal challenge mouse model, wt TcdA had an _LD50 of ~50 ng/mouse, while wt TcdB was more potent with an _LD50 of ~5 ng/mouse. Conversely, the in vitro assays based on cell cultures described above had _EC50 values of 100 pg/well (wt TcdA) and 2 pg/well (wt TcdB).

Example 15: Residual cytotoxicity of the genetically modified heptasome toxin B (SEQ ID NO: 8)

As shown in Figure 7, the _EC50 values of the triple mutant toxin B (SEQ ID NO:6) (20.78 ng/mL) and the heptameric mutant toxin B (SEQ ID NO:8) (35.9 ng/mL) were similar, indicating that the four additional mutations further modifying the glucosyltransferase active site and GTPase substrate binding site did not further reduce the cytotoxicity of the genetically modified toxins. The _EC50 value of the double mutant toxin B (SEQ ID NO:5) was also similar to that of the triple and heptameric mutant toxins (data not shown). This observation suggests that the cytotoxicity of the mutant toxins is surprisingly independent of glucosyltransferase and substrate recognition mechanisms.

Example 16: Residual cytotoxicity of triple mutant toxins A and B (SEQ ID NO: 4 and 6, respectively)

To further assess the nature of the residual cytotoxicity, the mutant toxins (SEQ ID NO: 4 and 6) were mixed and incubated with their respective neutralizing antibodies before being added to an IMR90 cell monolayer culture.

The results (Figure 8) showed that the residual cytotoxicity of mutant toxins A and B (SEQ ID NO: 4 and 6, respectively) could be completely eliminated by neutralizing antibodies specific to mutant toxin A (top, Figure 8) and mutant toxin B (bottom, Figure 8). Progressive concentrations of mutant toxins A (top) and B (bottom) were incubated with rabbit antitoxin monoclonal antibodies (pAb, 1:10 dilution) or mouse monoclonal antibodies (1:50 dilution from a stock solution containing 3.0 mg IgG/mL) before being added to IMR90 cells. After incubation with IMR90 cells at 37°C for 72 hours, the substrate was added, and relative optical units (RLU) were measured using a fluorescence program in a photometer to measure ATP levels. Lower ATP levels indicated higher toxicity. Controls included Tcd and TcdB (4 times their corresponding _EC50 values).

Publicly available reports indicate that mutations in the glucosyltransferase or self-catalyzing protease domains of the toxin lead to complete inactivation of the toxicity. However, our data do not align with these publications, which can be attributed to the fact that our study tested highly purified mutant toxins with increased cleavage concentrations, whereas published reports used crude culture hydrolysates; the increased time points at which cell rounding was observed in mutant toxin-treated cells (e.g., 24, 48, 72, or 96 hours), whereas reported observations were made only within less than 12 hours; the use of cell lines exhibiting significantly higher sensitivity to the toxin in this cytotoxicity assay, compared to HT-29 human colorectal adenocarcinoma cells used in published cytotoxicity assays; and/or to unknown activities or processes beyond glucosylation that could contribute to the residual toxicity of the mutant toxin.

Example 17: A novel mechanism of cytotoxicity of genetically modified toxins

To investigate the mechanism of residual cytotoxicity of mutant toxins, IMR-90 cells were treated with either wt toxin B (SEQ ID NO:2) or mutant toxin B (SEQ ID NO:6), and the glucosylation of Rac1 GTPase was studied over treatment time. Samples were collected from 24 to 96 hours, and cell extracts were prepared. Glucosylated Rac1 was distinguished from un-glucosylated Rac1 by Western blotting using two Rac1 antibodies. One antibody (23A8) recognized both forms of Rac1, while the other (102) recognized only un-glucosylated Rac1. As shown in Figure 22, for toxin B, the total Rac1 level did not change over time, and most of the Rac1 was glucosylated. On the other hand, treatment with mutant toxin B (SEQ ID NO:6) resulted in a significant decrease in total Rac1, however, this Rac1 was un-glucosylated at all time points. This shows that mutant toxin treatment has a negative effect on Rac1 levels, but wt toxin does not. As shown in Figure 22, at the indicated time points, actin levels were similar in cells treated with the toxin and the mutant toxin B, and similar to those in the simulated treatment. This demonstrates that the mutant toxin exerts cytotoxicity through a mechanism that drives a different glucosylation pathway than the wild-type toxin.

Example 18: Chemical treatment of genetic mutant toxins

Although a 4-log reduction in cytotoxic activity was preferred for genetically modified mutant toxins, further reductions in cytotoxic activity (2-4 log) were considered. Two chemical inactivation strategies were evaluated.

The first method uses formaldehyde and glycine to inactivate the mutant toxin. Formaldehyde inactivation occurs through the formation of a Schiff base (imine) between formaldehyde and a primary amine on the protein. The Schiff base can then react with many amino acid residues (Arg, His, Trp, Tyr, Gln, Asn) to form intramolecular or intermolecular crosslinks. This crosslink fixes the protein structure, leading to its inactivation. Additionally, formaldehyde can react with glycine to form a Schiff base. This glycyl Schiff base can then react with amino acid residues to form an intermolecular protein-glycine crosslink. Formaldehyde reduces the cytotoxic activity of the mutant toxin to below the detection limit (>8 log ₁₀ for triple mutant B (SEQ ID NO:6) and >6 log ₁₀ for triple mutant A (SEQ ID NO:4)). However, recovery over time is observed when the formaldehyde-inactivated (FI) triple mutant toxin is incubated at 25°C. Cytotoxicity recovery can be prevented by adding a small amount of formaldehyde (0.01-0.02%) to the FI-triple mutant toxin storage solution. See Example 23.

Another method uses 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC)/N-hydroxysuccinimide (NHS) treatment to generate inactivated mutant toxins. In this method, EDC/NHS reacts with the carboxyl group on the protein to form an activated ester. This activated ester can then react with a primary amine on the protein to form a stable amide bond. Similar to the reaction with formaldehyde, this reaction leads to intramolecular and intermolecular crosslinks. The amide bonds formed by EDC/NHS treatment are more stable and irreversible than the fragile imine bonds formed by formalin inactivation. In addition to the crosslinks formed by the reaction of the activated ester with the primary amine on the peptide, glycine and β-alanine adducts can also be formed. Glycine adducts are formed upon the annealing of unreacted activated esters with glycine, without mechanistic or theoretical limitations. The amine of glycine reacts with the activated ester on the peptide to form a stable amide bond. β-alanine adducts are formed by the reaction of activated β-alanine with a primary amine on the peptide, without mechanistic or theoretical limitations. This reaction leads to a stable amide bond. Activated β-alanine is produced by the reaction of 3 moles of NHS with 1 mole of EDC.

To achieve a 2-4 log reduction in cytotoxic activity relative to genetically modified mutant toxins (6-8 log relative to natural toxins), the chemically inactivated mutant toxin should have an _EC50 value ≥1000 μg/mL. In addition to the reduction in cytotoxic activity, retention of key epitopes identified by dot blot hybridization analysis is advantageous. Currently, numerous reaction conditions have been identified that satisfy both cytotoxicity reduction and epitope recognition requirements. Several batches of inactivated mutant toxins were prepared for animal studies, and analytical data from several representative batches are shown in Tables 7A and 7B, and Tables 8A and 8B.

List = List Biologicals; CPE = Cytopathic Effect Assay; EC ₅₀ = Lowest concentration at which 50% of cells show cytotoxicity; mAb = Monoclonal antibody; neutral = Neutralization; ND = Not performed; TM = Activation site and cleavage mutant (“triple mutant”).

List = List Biologicals; CPE = Cytopathic Effect Assay; EC ₅₀ = Lowest concentration at which 50% of cells show cytotoxicity; mAb = Monoclonal antibody; neutral = Neutralization; ND = Not performed; TM = Activation site and cleavage mutant (“triple mutant”).

Example 19: Purification

At the end of fermentation, the fermenter is cooled. The cytoplasm is recovered by continuous centrifugation and resuspension in a suitable buffer. Cell lysis of the cell suspension is achieved by high-pressure homogenization. For mutant toxin A, the homogenate is flocculent, and the flocculent solution is continuously centrifuged. This solution is filtered and then transferred to subsequent processing. For mutant toxin B, the homogenate is clarified by continuous centrifugation and then transferred to subsequent processing.

The mutant toxin A (SEQ ID NO:4) was purified using two chromatographic steps followed by a final buffer replacement. The clarified lysis buffer was loaded onto a hydrophobic interaction chromatography (HIC) column and eluted with a sodium citrate gradient. The product from the HIC column was then loaded onto a cation exchange (CEX) column and eluted with a sodium chloride gradient. The CEX collection containing purified mutant toxin A was replaced with the final buffer via dialysis filtration. The purified mutant toxin A was then replaced with the final drug substance intermediate buffer via dialysis filtration. After dialysis filtration, the residue was filtered through a 0.2-micron filter and then chemically inactivated to the final drug substance. The protein concentration target was 1–3 mg/mL.

The mutant toxin B (SEQ ID NO: 6) was purified using two chromatographic steps followed by a final buffer replacement. The clarified lysis buffer was loaded onto an anion exchange (AEX) column and eluted with a sodium chloride gradient to remove the bound mutant toxin. Sodium citrate was added to the product collection from the AEX column and loaded onto a hydrophobic interaction chromatography (HIC) column. The bound mutant toxin was eluted with a sodium citrate gradient. The HIC collection containing the purified mutant toxin peptide (SEQ ID NO: 6) was replaced with the final buffer via dialysis filtration. The purified mutant toxin B was replaced with the final drug intermediate buffer via dialysis filtration. After dialysis filtration, the residue was filtered through a 0.2 μm filter and then chemically inactivated to the final drug substance. The protein concentration target was 1–3 mg/mL.

Example 20: Formaldehyde/Glycine Inactivation

After purification, the mutant toxins A and B (SEQ ID NO: 4 and 6, respectively) were inactivated at 25°C for 48 hours with 40 mM (1.2 mg/mL) formaldehyde. This inactivation was performed at pH 7.0 ± 0.5 in 10 mM phosphate buffer and 150 mM sodium chloride buffer (containing 40 mM (3 mg/mL) glycine). The inactivation time was set to be more than three times the time required to reduce the _EC50 to 1000 μg/mL in IMR90 cells. After 48 hours, the biological activity was reduced by 7 to 8 log ₁₀ relative to the native toxin. After 48 hours of incubation, the inactivated mutant toxins were replaced with a drug substance buffer via dialysis filtration. For example, the inactivated toxins were concentrated to 1–2 mg/mL using a 100 kD regenerated cellulose acetate ultrafiltration cartridge, and the buffer was replaced.

Example 21: Inactivation of N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide (EDC)/N-hydroxysuccinimide (NHS)

After purification, the mutant toxins (SEQ ID NO:4 and SEQ ID NO:6) were inactivated at 25°C for 2 hours with 0.5 mg of EDC and 0.5 mg of NHS per mg of purified mutant toxin A and B (approximately 2.6 mM and 4.4 mM, respectively). This reaction was further inactivated by adding glycine to a final concentration of 100 mM and incubating at 25°C for an additional 2 hours. The inactivation was carried out at pH 7.0 ± 0.5 in 10 mM phosphate and 150 mM sodium chloride buffer. The inactivation time was set to be more than three times the time required to reduce the _EC50 to 1000 μg/mL in IMR90 cells. After 2 hours, the biological activity was reduced by 7 to 8 log ₁₀ relative to the native toxin. After 4 hours of incubation, the inactivated mutant toxins were replaced with drug substance buffer by dialysis filtration. For example, using a 100kD regenerated cellulose acetate ultrafiltration cartridge, the inactivated toxins can be concentrated to 1-2 mg/mL and the buffer replaced.

Unless otherwise stated, the following terms, as used in the Examples section, refer to the composition produced according to Example 21 of this specification: "EDC/NHS-treated triple mutant toxin"; "EDC-inactivated mutant toxin"; "Mutant toxin [A/B] drug substance"; "EI-mutant toxin"; "EDC/NHS-triple mutant toxin". For example, the following terms are synonymous: "EDC/NHS-treated triple mutant toxin A"; "EDC-inactivated mutant toxin A"; "Mutant toxin A drug substance"; "EI-mutant toxin A"; "EDC/NHS-triple mutant toxin A". As another example, the following terms are synonymous: "EDC/NHS-treated triple mutant toxin B"; "EDC-inactivated mutant toxin B"; "Mutant toxin B drug substance"; "EI-mutant toxin B"; "EDC/NHS-triple mutant toxin B".

Both mutant toxin A and mutant toxin B were prepared in a batch manner, comprising (1) fermentation of a toxin-negative Clostridium difficile strain (VPI 11186) containing plasmids encoding the respective genetic triple mutant toxin polypeptides, the fermentation being carried out in a medium containing soybean hydrolysate, yeast extract HY YEST ^™ 412 (Sheffield Bioscience), glucose, and thiamphenicol, (2) purification of the genetic mutant toxin (the “drug intermediate”) from cell-free lysate to a purity of at least above 95%, wherein ion exchange and hydrophobic interaction chromatography processes were used, (3) chemical inactivation by treatment with EDC/NHS followed by glycine annealing/blocking reaction, and (4) replacement with a final buffer matrix.

Example 22: Study on support for inactivation and preparation conditions

To optimize the chemical inactivation of the mutant toxin, a design of statistical experiment (DOE) was employed. Factors detected in the DOE included temperature, formaldehyde/glycine concentration, EDC/NHS concentration, and time (Tables 9 and 10). _EC50 values in IMR90 cells were measured to monitor loss of biological activity. Additionally, cell morphology of IMR-90 cells was observed at various time points after treatment. See Figure 9, which shows the morphology at 72 hours post-treatment. To determine the effect on protein structure, dot blot hybridization analysis was used to monitor epitope recognition, employing a set of monoclonal antibodies targeting different domains of the toxin.

In the formaldehyde/glycine inactivation of Clostridium difficile mutant toxins, the final reaction conditions were selected to achieve the desired level of cytotoxic activity reduction (7 to 8 log ₁₀ ) while maximizing epitope recognition. See Example 20 above.

In the EDC/NHS inactivation of Clostridium difficile mutant toxins, the final reaction conditions were selected to achieve the desired level of cytotoxic activity reduction (7 to 8 log ₁₀ ) while maximizing epitope recognition. See Example 21 above.

In another embodiment, the EDC-NHS reaction is annealed by adding sufficient alanine to anneal the reaction. The use of alanine can result in modifications on the mutant toxin protein, similar to modifications made when annealing with glycine. For example, annealing with alanine can result in an alanine moiety on the side chains of glutamic acid and/or aspartic acid residues of the mutant toxin. In another embodiment, the EDC-NHS reaction is annealed by adding sufficient glycine methyl ester to anneal the reaction.

Under optimal conditions, the chemically inactivated triple mutant Clostridium difficile toxins A and B resulted in a further reduction in residual cytotoxicity to undetectable levels (>1000 μg/mL – the highest concentration tested by CPE assay), while preserving antigenicity (measured by its reaction with toxin-specific neutralizing antibodies). The results presented in Table 28 demonstrate a progressive reduction in cytotoxicity from wt toxins to EDC/NHS-treated triple mutant toxins. Immunofluorescence staining confirmed that the triple mutant toxins (SEQ ID NO: 4 and 6) and the mutant toxin drug substance exhibited comparable binding to IMR-90 cells, indicating that the loss of cytotoxicity was not due to reduced cell binding (data not shown). Compared to mutant toxin A drug substance, mutant toxin B drug substance achieved a higher fold reduction in cytotoxicity, consistent with the observed ~600-fold higher potency of TcdB compared to TcdA.

Table 28. Summary of Cytotoxicity

The results of cytotoxicity assays for mutant toxin B modified with EDC alone, or mutant toxin B modified with both EDC and sulfonyl-NHS, were also evaluated. See Table 29.

Table 29

Conditions: Triple mutant toxin B (“TM TcdB”) (SEQ ID NO: 6) was modified in a weight ratio of mutant toxin B:EDC:sulfonated-NHS = 1:0.5:0.94. This molar ratio is equivalent to (corresponding to a higher MW of sulfonated-NHS) the standard EDC/NHS reaction described in Example 21. To determine the effect of sulfonated-NHS, the sulfonated-NHS ratio was varied from 0.5x to 2x the standard ratio. A duplicate reaction was carried out in 1x PBS pH 7.0 at 25°C and initiated by adding EDC solution. After 2 hours, the reaction was annealed by adding 1M glycine pH 7.0 (0.1M final concentration) and incubated for an additional 2 hours. The annealed reaction was desalted using a Vivaspin 20 apparatus, and the mutant toxin B drug substance (“TM TcdB-EDC”) was concentrated, sterilely filtered into sterile vials, and used for evaluation in cytotoxicity assays.

At the same molar ratio, sulfonated NHS reduced _EC50 to approximately 250 ug/mL (compared to >1000 ug/mL for NHS). Even at twice the molar ratio, sulfonated NHS did not show the same effectiveness as NHS in reducing cytotoxicity. See Table 30.

Table 30

To determine the number and type of modifications, peptide mapping was performed on triple mutant toxin B samples inactivated by EDC/NHS and EDC/sulfonyl-NHS. Similar amounts of glycine adducts, crosslinking, and dehydroalanine modifications were observed in both samples. However, β-alanine was not observed in the sulfonyl-NHS sample.

Wild-type toxin B (SEQ ID NO:2) was inactivated using a standard protocol (see Example 21); toxin B:EDC:NHS1:0.5:0.5 was incubated at 25°C for 2 hours, followed by annealing in 1x PBS pH 7.0 with 1M glycine (0.1M final concentration) and incubation for an additional 2 hours. The sample was desalted, and the concentration was determined for cytotoxicity assays. The EC ₅₀ of this sample was <244 ng/mL.

Example 23: Restoration Research

To determine whether Clostridium difficile mutant toxins inactivated by formaldehyde/glycine or EDC/NHS were reactivated, inactivated mutant toxin samples (1 mg/mL) were incubated at 25°C for 5–6 weeks. Aliquots were taken weekly, and the _EC50 value in IMR90 cells was determined. One formaldehyde/glycine-inactivated sample contained no formaldehyde, while another sample contained 0.01% formaldehyde. The _EC50 was determined by CPE assay.

Partial recovery was observed at 25°C in the absence of residual formaldehyde (Table 11). Cytotoxic activity increased approximately 3-fold after 5 weeks. Although cytotoxic activity increased, it still showed a 7 log ₁₀ decrease relative to the native toxin after 5 weeks. Recovery was completely prevented by the presence of 0.010% formalin. No recovery was observed in the EDC/NHS-inactivated samples. Throughout the 6-week incubation period, the EC50 values for all four sets of EDC/NHS-treated triple mutant toxin A (SEQ ID NO:4) and EDC/NHS-treated triple mutant toxin B (SEQ ID NO:6) remained at an initial level >1000 μg/mL. Conversely, the _EC50 values for FI-treated triple mutant toxin A (SEQ ID NO:4) and FI-treated _triple mutant toxin B (SEQ ID NO:6) were unstable and decreased to unacceptably low _EC50 values, indicating either increased cytotoxicity or recovery from inactivation. See Table 11.

Beyond consistently reducing cytotoxicity to undetectable levels (>1000 μg/mL, measured by CPE assay), the mutant toxins inactivated by EDC/NHS retained important epitopes (which are targets for toxin-neutralizing mAbs). See Table 31. The FI mutant toxins, however, showed loss of the same antigenic determinants.

Table 31. EDC/NHS inactivation reduced the cytotoxicity of genetically modified toxins while preserving important antigenic determinants.

Cytotoxicity was measured in IMR90 cells using the CPE assay.

b is the value determined by Biacore ^™ analysis using multiple neutralizing mAbs (targeting each non-overlapping toxin epitope).

c is the average of the two experiments.

For the first three rows, neutralizing mAbs "1", "2", and "3" refer to mAbs A60-22, A80-29, and A65-33 for toxin A, respectively. For the last three rows, neutralizing mAbs "1", "2", and "3" refer to mAbs B8-26, B59-3, and B-56-15 for toxin B, respectively.

Example 24: Preclinical Immunogenicity Study

Key preclinical objectives include testing compositions containing Clostridium difficile mutant toxins A and B in small animals and nonhuman primates (NHPs). Mice and hamsters are immunized to determine whether the Clostridium difficile composition can elicit neutralizing antibodies against mutant toxins A and B. Induction of serum neutralizing antibody responses to the antigens is tested after a series of immunizations in mice, hamsters, and cynomolgus monkeys. In some embodiments, the genetically and/or chemically inactivated mutant toxins are formulated in neutral buffers, aluminum phosphate buffers, or buffers containing ISCMATRIX as an adjuvant. Neutralizing antibody responses are typically tested 2–4 weeks after each booster or final dose.

The toxin neutralization assay demonstrated the ability of the antiserum to neutralize the cytotoxic effects mediated by Clostridium difficile TcdA or TcdB, and thus could be used to measure the functional activity of antibodies present in samples. The toxin neutralization assay was performed on the human lung fibroblast cell line IMR-90, which is sensitive to both TcdA and TcdB. Briefly, 96-well microtiter plates seeded with IMR-90 cells served as targets for toxin-mediated cytotoxicity. The ability of each test serum sample to neutralize TcdA and TcdB was analyzed separately. Appropriate serial dilutions of the test antiserum were mixed with fixed concentrations of TcdA or TcdB and incubated at 37°C for 90 minutes in a humidified incubator (37°C/5% _CO2 ) to allow toxin neutralization. To control quality, all plates included a reference standard and control containing known titers of antitoxin antibodies. Ninety minutes later, the toxin-antiserum mixture was added to IMR-90 cell monolayer cultures, and the plates were incubated for an additional 72 hours. Subsequently, the substrate was added to assay plates to determine the level of adenosine triphosphate (ATP) present in metabolically active cells, which was measured as relative optical units (RLU). High ATP levels indicated high cell viability, and the level was directly proportional to the amount of toxin neutralized by the antibody present in the sample. For preclinical data, the RLU data were plotted against the dilution of the tested antiserum sample to generate a four-parameter logistic (4-PL) regression response fit curve. Neutralization titers are expressed as the sample dilution exhibiting a 50% reduction in cytotoxicity.

Example 25: Immunogenicity study in mice: *Clostridium difficile* 2010-06

The aim of this study was to evaluate the immunogenicity of two forms of mutant Clostridium difficile toxin B (SEQ ID NO:6) (each chemically inactivated by different methods). In this study, untreated mutant toxin B (SEQ ID NO:6) (genetically inactivated but not chemically inactivated) with or without adjuvant was used as a control.

The group of 10 mice was immunized intramuscularly with 10 μg of the immunogen according to Table 12.

Table 12. Testing of chemically inactivated mutant toxin B (SEQ ID NO: 6) in mice

Results: No adverse effects were observed after each administration of the vaccine candidate. As shown in Figure 10, mice in each group developed significant and potent antitoxin B neutralizing antibodies after three administrations of their respective mutant toxins.

Based on the low levels observed at week 12, it appears that EDC-inactivated mutant toxin B (group 2) and formalin-inactivated mutant toxin (group 1) produced robust neutralizing responses in mice.

In the absence of chemical inactivation, the genetically modified toxin B (SEQ ID NO: 6) produced a neutralizing response after two administrations (groups 3-4, week 8), which was enhanced after a third administration (groups 3-4, week 12).

Example 26: Immunogenicity study in mice: *Clostridium difficile* (mu) 2010-07:

The aim of this study was to evaluate the immunogenicity of chemically inactivated Clostridium difficile mutant toxins A and B (SEQ ID NO: 4 and 6, respectively), alone or in combination. Immunogens for all groups were formulated using aluminum phosphate as an adjuvant.

Five mice were immunized intramuscularly with 10 μg of the immunogen according to Table 13.

Table 13. Testing of chemically inactivated genetically A and B mutant toxins in mice (SEQ ID NO: 4 and 6, respectively).

Table 13. Testing of chemically inactivated genetically A and B mutant toxins in mice (SEQ ID NO: 4 and 6, respectively).

Results: No adverse effects were observed after each administration of the vaccine candidate. As shown in Figure 11, after two administrations of the chemically inactivated mutant toxin, antitoxin A neutralizing antibodies (groups 3–5) were boosted to titers between 3 and 4 log ₁₀ , while antitoxin B neutralizing antibodies (groups 1–2, 5) remained low to undetectable, consistent with data from the mouse study described above (Figure 10). Antitoxin B neutralizing antibodies in groups 1, 2, and 5 boosted to 2–3 log ₁₀ after the third administration (week 12 titer) and reached their peak after the fourth administration (week 14 titer). Antitoxin A neutralizing antibody titers in groups 3–5 increased slightly after the third (week 12 titer) and fourth immunization (week 14 titer).

Example 27: Hamster Immunogenicity Study: Clostridium difficile 2010-02

The aim of this study was to evaluate the immunogenicity and protective potential of Clostridium difficile triple mutants and chemically inactivated mutant toxins A and B in a Syrian golden hamster model. The Syrian golden hamster model represents the best challenge model simulating human CDAD. Mutant toxins A and B from the same batches used in mouse studies of Clostridium difficile 2010-07 were also used in this study. As a control, a group was given mutant toxins without aluminum adjuvant.

Five Syrian golden hamsters were immunized intramuscularly with 10 μg of the immunogen according to Table 14.

Table 14. Chemically inactivated mutant toxins A and B (SEQ ID NO: 4 and 6, respectively) tested in hamsters (Clostridium difficile 2010-02).

Results: No adverse effects were observed after immunization with the mutant toxin. As shown in Figure 12, the neutralizing response to antitoxin A was between 2–3 log ₁₀ (formalin-inactivated mutant toxin (groups 1–2)) and 3–4 log ₁₀ (EDC-inactivated mutant toxin (group 3)) after a single dose of the mutant toxin. Antitoxin A antibody levels increased in all three groups after the second dose. No increase in antitoxin A antibody levels was observed in any of the three groups after the third dose. Similar results were observed after the fourth immunization, with an increase in titer observed in the aluminum-adjuvant-free formalin-inactivated group (group 2).

Following a single dose, the antitoxin B neutralizing response was undetectable in the formalin-inactivated mutant toxin groups (groups 1-2) and only exceeded 2 log ₁₀ in the EDC-inactivated mutant toxin group (group 3). After the second dose, the antitoxin B neutralizing antibody titer increased to 3-4 log ₁₀ in both formalin-inactivated groups (groups 1-2) and to 4-5 log ₁₀ in the EDC-inactivated group (group 3). For all three groups, an increase in antitoxin B neutralizing antibody titers was observed after the third and/or fourth dose, with peak titers reached at week 16 (after the last dose). See Figure 12.

In Figure 13, the level of neutralizing antibody response to chemically inactivated genetic mutant toxin (Figure 12) is compared with the level induced by toxins from List Biological Laboratories, Inc. (Campbell, California) (i.e., wild-type toxins prepared by formalin inactivation of toxins purchased from List Biological Laboratories; a control reagent used to establish a hamster attack model).

As used herein, “FI” in figures and tables refers to formalin/glycine treatment of the toxin for 2 days at 25°C, unless otherwise stated. “EI” in figures and tables refers to EDC/NHS treatment for 4 hours at 30°C, unless otherwise stated. In Figure 13, 5 hamsters were treated with various mutant toxin compositions, while 11 hamsters were treated with toxoids purchased from ListBiological.

The data in Figure 13 show that, in hamsters administered according to Table 14, the neutralizing antibody titers against toxin A (Figure 13A) and toxin B (Figure 13B) induced by two doses of the immunogenic composition containing EDC-inactivated mutant toxin were higher than the neutralizing antibody titers induced by toxoids from List Biologicals.

Example 28: Hamster Immunogenicity Study: Clostridium difficile 2010-02 (Continued)

To assess the protective efficacy of the mutant toxin, immunized hamsters, along with a control group of unimmunized animals, were first administered orally with clindamycin antibiotics (30 mg/kg) to disrupt the normal gut microbiota. Five days later, the hamsters were challenged with orally administered wild-type Clostridium difficile spores (630 strains, 100 CFU/animal). Signs of CDAD (wet tail) in hamsters were monitored daily for 11 days post-challenge. Animals identified as having severe CDAD were euthanized using a clinical system that scored numerous different parameters, including post-stimulation activity, dehydration, excrement, temperature, and body weight, which are known in the art.

On day 11, the study was terminated and all animals were euthanized. Figure 14 shows the survival curves for each of the three immunized groups (Groups 1-3 according to Table 14) compared to the unimmunized control. As shown therein, all unimmunized animals developed severe CDAD and required euthanasia 1-3 days after the attack (0% survival). Both groups administered formalin-inactivated mutant toxin had 60% survival curves, and the animals did not require euthanasia until day 3 (Group 1) or day 4 (Group 2). The group administered EDC-inactivated mutant toxin had an 80% survival curve, with one animal (out of five) requiring euthanasia on day 7. Therefore, Clostridium difficile spores can protect hamsters from lethal attacks.

Example 29: Hamster Immunogenicity Study: Immunogenicity of Clostridium difficile 2010-03: Genetically and Chemically Inactivated Clostridium difficile Mutant Toxin

The aim of this study was to evaluate the immunogenicity of unadjuvanted Clostridium difficile triple mutants and chemically inactivated mutant toxins A and B (SEQ ID NO: 4 and 6, respectively) in a Syrian golden hamster model. The same batches of mutant toxins A and B (SEQ ID NO: 4 and 6, respectively) used in the mouse study of Clostridium difficile 2010-07 were also used in this study. As a control, one group (Group 1) was given phosphate-buffered saline as a placebo.

Groups of 5 or 10 Syrian golden hamsters were immunized according to the immunization guidelines in Table 15. Animals were given the drug three times. Additionally, the drug was administered to the animals every two weeks.

Results: See Figure 15. No antitoxin A or B antibodies were observed in the placebo control group. Following a single dose, antitoxin A neutralizing antibodies of 2–3 log ₁₀ were observed in the formalin-inactivated group (group 2) and the mutant toxin group (group 4), and 3–4 log ₁₀ were observed in the EDC-inactivated group (group 3). Following a second immunization with the relevant mutant toxin, an increase in antitoxin A neutralizing antibody levels was observed in each of these groups (2–4) (compare titers at weeks 2 and 3 in Figure 15). Following a third dose of the mutant toxin (administered at week 4), antitoxin A neutralizing antibody titers in groups 2–4 were higher than their week 4 titers.

Antitoxin B neutralizing antibodies were detectable after the second dose, with increases of 3-4 log ₁₀ in the formalin-inactivated group (Group 2) and the EDC-inactivated group (Group 3), and 2-3 log ₁₀ in the genetic triple mutant group (Group 4). Following the third immunization (week 4), antitoxin B neutralizing antibody titers increased to 3-4 log ₁₀ in the formalin-inactivated mutant toxin group (Group 2) and the genetic triple mutant toxin group (Group 4), and to 4-5 log ₁₀ in the EDC-inactivated mutant toxin group (Group 3).

For antitoxin A and antitoxin B neutralizing antibodies, peak titers were observed at week 6 (after the third dose) in all vaccination groups (groups 2–4).

Evaluation of immunogenic compositions using aluminum gel/CpG or ISCOMATRIX as adjuvants

Hamsters immunized with immunogenic compositions containing chemically inactivated mutant toxins (prepared with aluminum gel, ISCMATRIX, or aluminum gel/CpG24555 (Alh/CpG)) developed strongly neutralizing antitoxin antisera. Peak antitoxin A and antitoxin B responses were observed to be 2-3 times higher and statistically significant in groups immunized with the mutant toxins (prepared in Alh/CpG or ISCMATRIX) than in groups immunized with the aluminum gel-prepared vaccine alone. See Table 32, which shows the 50% neutralizing titer. Hamsters (n = 10 hamsters/group) were immunized at weeks 0, 2, and 4 with 10 μg of each mutant toxin A drug substance and mutant toxin B drug substance (prepared with 100 μg aluminum gel, or 200 μg CpG 24555 + 100 μg aluminum gel, or 10 U ISCMATRIX). Serum was collected at each time point and the functional antitoxin activity was analyzed in a toxin neutralization assay. The geometric mean titers are presented in Table 32. An asterisk (*) indicates statistical significance when comparing titers with the aluminum gel group (p<0.05).

The protective potency of immunogenic compositions containing mutant toxin drug substances formulated with these adjuvants was tested. Hamsters were immunized at week 5 and orally administered clindamycin (30 mg/kg), and challenged according to the method described above. A control group of unimmunized hamsters (n=5) was included. Increased potency was observed in hamsters immunized with mutant toxin drug substances (with Alh/CpG or ISCMATRIX as adjuvants) (100% survival) compared to those immunized with aluminum glue alone (70% survival). Therefore, Clostridium difficile spores can protect hamsters from lethal attacks.

Example 30: Inoculation of Clostridium difficile into cynomolgus monkeys

The aim of this study was to test the immunogenicity of low- and high-dose EDC-inactivated and formalin-inactivated Clostridium difficile mutant toxins in cynomolgus monkeys. All mutant toxins were formulated with adjuvants, except for one group which served as an unadjuvanted control (group 5).

Results: Figure 16 shows the antitoxin neutralizing antibody response in these animals at weeks 0, 2, 3, 4, 6, 8, and 12. Antitoxin A titers in all five groups were between 2 and 3 log ₁₀ after a single dose (week 2 titer). These titers increased with each subsequent dose to each group. Titers did not decrease between weeks 3 and 4 in these animals. Peak values were between 4 and 5 log ₁₀ in all groups. At all time points, the group without ISCMATRIX adjuvant (group 5) had the lowest titers, demonstrating the effectiveness of ISCMATRIX in enhancing the immune response. The unadjuvanted control group (group 5) peaked at week 12, as did the group immunized with a high dose of EDC-inactivated mutant toxin (group 4); all other groups peaked at week 6, two weeks after the last dose. Titers increased after the second dose in all groups (week 3 time point). Similar to the antitoxin A response, the antitoxin B response did not decrease from week 3 to week 4. Following the third dose (week 6), antitoxin B neutralizing antibody titers were between 3 and 4 log ₁₀ in all groups, except for the low-dose formalin-inactivation group (group 1) and the high-dose EDC-inactivation group (group 4), both of which had titers >4 log _10. Peak titers were observed in all groups at week 12, except for the low-dose EDC-inactivation group (group 3), which had a peak titer at week 8. All groups had peak titers >4 log ₁₀ .

Example 31: Production of Monoclonal Antibodies

Although toxins A and B share many structural homologs, the neutralizing activity of antibodies has been found to be toxin-specific. In this invention, several antibodies specific to individual toxins and targeting various epitopes and functional domains were identified, exhibiting high affinity and strong neutralizing activity against natural toxins. Antibodies were isolated from mice immunized with commercially available formalin-inactivated (FI)-mutant toxins or recombinant fully-mutant toxins (SEQ ID NO: 4 and 6, rendered non-toxic by introducing specific mutations at their catalytic sites) to produce toxin A and B mAbs, respectively. Epitope mapping of the antibodies showed that most of the mAbs targeting toxin A (49 out of 52) targeted the non-catalytic C-terminal domain of the toxin.

Monoclonal antibodies against toxin B target three domains of the protein. Of the total 17 toxin B-specific mAbs, 6 are specific to the N-terminus (e.g., amino acids 1-543 of wild-type Clostridium difficile TcdB, e.g., 630), 6 are specific to the C-terminus (e.g., amino acids 1834-2366 of wild-type Clostridium difficile TcdB, e.g., 630), and 5 are specific to the intermediate translocation domain (e.g., amino acids 799-1833 of wild-type Clostridium difficile TcdB, e.g., 630). The use of mutant Clostridium difficile toxins (e.g., SEQ ID NO: 4 and 6) as immunoantigens thus provides a key advantage over formalin inactivation methods for presenting most (if not all) of the antigenic epitopes, rather than formalin inactivation methods which adversely affect the antigenic structure of mutant toxins.

Example 32: Characterization of mAb A3-25 of toxin A, comprising a variable light chain having the amino acid sequence of SEQ ID NO:36 and a variable heavy chain having the amino acid sequence of SEQ ID NO:37.

Of particular interest is mAb A3-25 because this antibody does not conform to all attempts to define its immunoglobulin (Ig) isotype (using isotype kits commonly used for IgG, IgM, and IgA). Further analysis using Western blotting with Ig H-chain-specific antiserum revealed that A3-25 is an IgE isotype, which is rare in mAb production. This was confirmed by nucleotide sequencing of mRNA isolated from A3-25 hybridoma cells. The amino acid sequence deduced from the variable region nucleotide sequences of the H- and L-chains of A3-25 is shown in Figure 17.

To further evaluate A3-25mAb in Clostridium difficile infection and disease models, the Ig isotype of A3-25mAb was changed to mouse IgG1 by grafting the variable region molecule of the εH chain onto the mouse γ heavy chain according to a published method.

Example 33: Neutralizing activity and epitope mapping of toxin-specific antibodies

In addition, in attempts to identify functional/neutralizing antibodies, the ability of all monoclonal antibodies to neutralize wild-type toxins was evaluated in standard cytopathic effect (CPE) assays or in more rigorous and quantitative assays (based on measurements of ATP as an indicator of cell viability).

Of the 52 toxin A-specific antibodies tested, four mAbs (A3-25, A65-33, A60-22, and A80-29 (Table 17 and Figure 18) exhibited varying levels of neutralizing activity. BiaCore competitive binding assays and hemagglutination inhibition (HI) assays were performed to map the antibody epitopes. The results showed that these antibodies target different epitopes of the toxin A protein (Table 17). To further confirm the binding sites on this protein, the antibodies were individually evaluated using toxin fragments with known sequences in Western blotting or dot blot hybridization assays. All four neutralizing antibodies were found to target the C-terminal region of the toxin.

Of the 17 toxin B-specific antibodies identified, 9 were found to be neutralizing. Of these 9 neutralizing mAbs, 6 targeted the N-terminus of the B toxin, while the other three targeted the translocation domain of the B toxin (Table 18). Based on the Biacore competitive binding assay, these 9 neutralizing monoclonal antibodies were grouped into 4 epitope groups, as shown in Figure 19.

Table 17: Characterization of mAbs of selected toxin A

Table 18: Characterization of mAbs of selected toxin B

Example 34: Identification of a novel toxin A antibody combination with significantly enhanced neutralizing activity:

When tested individually in an ATP-based neutralization assay, the four toxin A mAbs (A3-25, A65-33, A60-22, and A80-29) showed incomplete or partial toxin A neutralizing activity. mAb A3-25 was the most potent antibody, while the other three exhibited lower neutralizing activity, with A80-29 barely breaking the background (Figure 18). However, when A3-25 was combined with any of the other three mAbs, a synergistic neutralizing effect was observed in all three combinations, significantly greater than the sum of the neutralization by the individual antibodies, as shown in Figures 20A-C. Furthermore, all three combinations exhibited the complete neutralizing capacity typically observed in antitoxin A polyclonal antibodies.

Example 35: Identification of a novel toxin B antibody combination exhibiting significantly enhanced neutralizing activity:

We also observed the synergistic neutralization of mAbs of toxin B from different epitope groups (identified by BiaCore analysis). mAb B8-26 of toxin B (the most dominant mAb from group 1) was combined with multiple mAbs from group 3. These combinations were evaluated in a toxin B-specific neutralization assay, and the results are shown in Figure 21 and Table 19.

A synergistic neutralizing effect was observed when B8-26 was combined with mAbs from epitope group 3 (Fig. 21B), but not when combined with any other mAbs (data not shown).

Example 36: In vitro screening of safe and effective mutant toxin compositions using mAbs:

Genetically modified Clostridium difficile mutant toxins A and B (e.g., SEQ ID NO: 4 and 6) showed residual cytotoxicity (as determined by in vitro cytotoxicity assays). Although we have achieved ~4 log reductions in cytotoxicity of each mutant toxin (Table 20), further chemical inactivation of the mutant toxins, such as treatment with formalin, is still preferred. However, chemical inactivation treatment can be drastic and can adversely affect key antigenic epitopes of the toxin or mutant toxin.

To optimize the biological process, a statistical design (DOE) was employed for the chemical inactivation (treatment with formalin and EDC/NHS) of triple mutants TcdA and B (1 mg/mL). To optimize formalin inactivation of triple mutant TcdA, we varied the formalin/glycine concentration (20–40 mM), pH (6.5–7.5), and temperature (25–40 °C). For triple mutant TcdB, we varied the formalin/glycine concentration from 2 to 80 mM, with temperatures and pH at 25 °C and 7.0, respectively. All formalin treatments were incubated for 24 hours. For formalin inactivation, “40/40” in Tables 21 and 23 represents the concentrations of formalin and glycine used in the reaction. For EDC/NHS treatment, we varied the concentrations of the triple mutant TcdA from 0.25 to 2.5 mg/mg and the triple mutant TcdB from 0.125 to 2.5 mg/mg, and incubated at 25°C for 4 hours. At the end of the reaction, all samples were desalted in 10 mM phosphate, pH 7.0. After purification, the residual cytotoxicity of the treated Tcd and the mAb recognition of the epitopes were analyzed by dot blot hybridization. The goal was to identify treatment conditions that reduced cytotoxicity to desired levels (EC ₅₀ > 1000 μg/mL) without adversely affecting the epitopes recognized by a set of neutralizing antibodies (++++ or +++). The treatment conditions (marked with checkmarks in Tables 21-24) produced potentially safe and effective immunogenic compositions that retained reactivity to at least four neutralizing mAbs and exhibited a 6-8 log ₁₀ reduction in cytotoxicity (relative to the respective wild-type toxin cytotoxicity). The selection results are shown in Tables 21-24. Additional data from different treatment conditions of the triple mutant toxin, derived from in vitro cytotoxicity and toxin neutralization assays, are shown in Tables 33 and 34. See also, for example, Examples 20 and 21 above, which provide further details on preferred crosslinking treatment conditions for the mutant toxin.

Table 33

Chemical cross-linking reaction conditions for the triple mutant toxin A (SEQ ID NO:4) sample mentioned in Table 33

Samples 1-4 were modified with EDC/NHS. Conditions: 30℃, 20mM MES/150mM NaCl, pH 6.5. The reaction was initiated by adding EDC. After 2 hours of reaction, 1M glycine was added to samples A, B, and C to a final glycine concentration of 50mM. No glycine was added to sample D. The reaction was set up with different weight ratios of mutant toxin A (SEQ ID NO:4):EDC:NHS, as shown below.

1 L44166-157A 1:0.25:0.25w:w:w

2 L44166-157B 1:1.25:1.25

3 L44166-157C 1:2.5:2.5

4 L44166-157D 1:2.5:2.5

Sample 5L44905-160A was incubated with 80mM HCHO, 80mM glycine, 80mM NaPO4 (pH 7) and 1 mg/mL mutant toxin A (SEQ ID NO:4) protein at 25°C for 48 hours.

Sample 6L44166-166, modified with EDC/NHS, was incubated at 25°C in 20mM MES/150mM NaCl at pH 6.5. The ratio of mutant toxin A (SEQ ID NO:4):EDC:NHS was 1:0.5:0.5. The reaction was initiated by adding EDC. After 2 hours of reaction, 1M glycine was added to a final concentration of 0.1M glycine, and the mixture was incubated for another 2 hours. After this time, the reaction buffer was replaced with 1X PBS on a Sephadex G25 reactor.

Sample 7L44905-170A was prepared with 80mM HCHO, 80mM glycine, 80mM NaPO4, pH ₇ , and 1 mg/mL mutant toxin A (SEQ ID NO:4) protein, and reacted at 35°C for 48 hours. This formalin reaction was intended to generate excessive cross-linking, thereby strictly reducing antigen binding.

Sample 8L44897-61, containing 32mM HCHO/80mM glycine, was reacted at 25℃ for 72 hours.

Sample 9L44897-63, containing 80mM HCHO/80mM glycine, was reacted at 25℃ for 72 hours.

The following reactions all have a reaction time of 24 hours:

Sample 10L44897-72 tube #1, 25℃, 80mM NaPi, pH 6.5, 20mM HCHO/20mM glycine

Sample 11L44897-72 tube #2, 25℃, 80mM NaPi, pH 6.5, 40mM HCHO/40mM glycine

Sample 12L44897-72 tube #3, 32.5℃, 80mM NaPi, pH 7.0, 30mM HCHO/30mM glycine

Sample 13L44897-72 tube #4, 32.5℃, 80mM NaPi, pH 7.0, 30mM HCHO/30mM glycine

Sample 14L44897-72 tube #5, 32.5℃, 80mM NaPi, pH 7.0, 30mM HCHO/30mM glycine

Sample 15L44897-75 tube #6, 25℃, 80mM NaPi, pH 7.5, 20mM HCHO/20mM glycine

Sample 16L44897-75 tube #7, 25℃, 80mM NaPi, pH 7.5, 40mM HCHO/40mM glycine

Sample 17L44897-75 tube #8, 40℃, 80mM NaPi, pH 6.5, 20mM HCHO/20mM glycine

Sample 18L44897-75 tube #9, 40℃, 80mM NaPi, pH 6.5, 40mM HCHO/40mM glycine

Sample 19L44897-75 tube #10, 40℃, 80mM NaPi, pH 7.5, 20mM HCHO/20mM glycine

Sample 20L44897-75 tube #11, 40℃, 80mM NaPi, pH 7.5, 40mM HCHO/40mM glycine

The following eight samples were reacted at 25°C for the indicated times in 80 mM NaPi at pH 7.0 containing 78 mM HCHO and 76 mM glycine:

Sample 21L44897-101 (pre-modified) TxA control time zero control sample, unmodified or exposed to HCHO/glycine

Sample 22L44897-101, 2hr

Sample 23L44897-101, 4hr

Sample 24L44897-101, 6hr

Sample 25L44897 102, 24hr

Sample 26L44897-103, 51hr

Sample 27L44897-104, 74hr

Sample 28L44897-105, 120hr

Sample 29 (L44980-004) is an EDC/NHS modified mutant toxin A (SEQ ID NO:4) (triple mutant toxin A (SEQ ID NO:4)-EDC). The reaction conditions were: 25°C, buffer 20mM MES/150mM NaCl, pH 6.6. The ratio of triple mutant toxin A (SEQ ID NO:4):EDC:NHS was 1:0.5:0.5 w:w:w. The reaction was initiated by adding EDC. After 2 hours of reaction, a senser was added to a final concentration of 0.1M, and the reaction was continued at 25°C for another 2 hours. The reaction was terminated by removing salts using a Sephadex G25 reactor.

The following 12 samples and 2 controls are from the recovery experiment, with the samples incubated at 25℃ and 37℃:

Reaction 1 = 25℃, 80mM NaPi, pH 7.0, only 40mM HCHO (without glycine), reaction time 24 hours.

Reaction 2 = 25℃, 80mM NaPi, pH 7.0, 40mM HCHO/40mM glycine, reaction time 24 hours.

The following four samples were produced by reacting at 25°C with 80 mM NaPi at pH 7.0, 40 mM HCHO/40 mM glycine for the indicated times:

44 L44897-116-6 29.5hr

45 L44897-116-7 57.5hr

46 L44897-116-8 79.5hr

47 L44897-116-9 123.5hr

Sample 48L44897-139 was reacted at 25℃ for 48 hours with 80mM NaPi, pH 7.0, and 40mM HCHO/40mM glycine.

Sample 49L44166-204: EDC/NHS modified mutant toxin A (SEQ ID NO:4). 25°C, buffer 1xPBS, pH 7. Mutant toxin A (SEQ ID NO:4):EDC:NHS = 1:0.5:0.5 w:w:w. Reacted with EDC/NHS for 2 hours, then 1M glycine was added to a final concentration of 0.1M and reacted for another 2 hours. The buffer was then replaced on a Sephadex G25 with 20 mL of histidine/100 mM NaCl at pH 6.5.

Chemical cross-linking reaction conditions for the mutant toxin B sample mentioned in Table 34

The triple mutant toxin B (SEQ ID NO:6) was chemically cross-linked and tested under the following reaction conditions. Sample L44905-86 was tested in experiments involving three formalin reaction variations and two incubation temperatures. Six samples were taken daily, for a total of 18 samples. The first sample in the list is the untreated control (making a total of 19 samples). The untreated control consisted of the untreated triple mutant toxin B peptide (SEQ ID NO:6).

Reaction 1 (“Rxn1”) = 80 mM HCHO, 80 mM glycine, 80 mM NaPO4 pH 7, 1 mg/mL triple mutant toxin B (SEQ ID NO: 6) protein

Reaction 2 (“Rxn2”) = 80 mM HCHO, glycine-free, 80 mM NaPO4 pH 7, 1 mg/mL triple mutant toxin B (SEQ ID NO: 6) protein

Reaction 3 (“Rxn3”) = 80 mM HCHO, glycine-free, 80 mM NaPO4 pH 7, 1 mg/mL triple mutant toxin B (SEQ ID NO:6) protein + cyanoborohydride capping. Cyanohydride capping involves adding 80 mM _CNBrH4 to the final desalted reaction and incubating at 36°C for 24 hours.

For sample L34346-30A, 0.5 g EDC and NHS per gram of triple mutant toxin B (SEQ ID NO: 6), in 20 mM MES, 150 mM NaCl, pH 6.5 at 30 °C for 4 hours.

For sample L34346-30B, 0.5 g of EDC and NHS per gram of triple mutant toxin B (SEQ ID NO: 6) was incubated at 30°C for 2 hours, followed by the addition of glycine (final concentration g/L) and incubation at 30°C for another 2 hours in 20 mM MES, 150 mM NaCl, pH 6.5. The only difference between the two reactions for L34346-30A and L34346-30B is the addition of glycine to reaction L34346-30B.

Example 37: Antibodies induced by the immunogenic composition are able to neutralize toxins from various Clostridium difficile strains.

To assess whether antibodies induced by an immunogenic composition containing a mutant toxin drug substance can neutralize a broad spectrum of diverse toxin sequences, strains representing diverse ribonucleotypes and toxin types were sequenced to identify the degree of genetic diversity among the strains compared to the mutant toxin drug substance. Subsequently, in an in vitro neutralization assay, culture supernatants containing toxins secreted from various strains were tested using serum from immunized hamsters to determine the coverage of the immunogenic composition and its protective ability against diverse toxins from prevalent clinical strains.

HT-29 cells (a colon cancer cell line) and IMR-90 cells were both used to test for neutralization of toxins expressed by CDC strains. HT-29 cells were more sensitive to TcdA; the _EC50 of purified TcdA in these cells was 100 pg/mL (compared to 3.3 ng/mL for TcdB). On the other hand, IMR-90 cells were more sensitive to TcdB; the _EC50 of purified TcdB in these cells ranged from 9 to 30 pg/mL (compared to 0.92 to 1.5 ng/mL for TcdA). The specificity of the assays for TcdA and TcdB in these cell lines was confirmed by using polyclonal and monoclonal toxin-specific antibodies. For normalization of the assays, culture filters of 24 CDC isolates were tested at concentrations four times their respective _EC50 values. Three of the strains had toxin levels too low to be tested in the neutralization assay.

Twenty-four strains representing diverse ribonucleotypes/toxin types (covering over 95% of prevalent Clostridium difficile strains in the US and Canada) were obtained from the CDC. These isolates included strains representing ribonucleotypes 027, 001, and 078, and three CDAD-prevalent strains in the US, Canada, and the UK. Strains 2004013 and 2004118 represent ribonucleotype 027; strain 2004111 represents ribonucleotype 001; and strains 2005088, 2005325, and 2007816 represent ribonucleotype 078. To determine the degree of genetic diversity among the disease-causing clinical isolates and the 630 strains, the toxin genes (TcdA and TcdB) from these clinical strains were fully sequenced. See Table 35. The amino acid sequences of the toxins were aligned using ClustalW in the Megalign ^™ program, and sequence identity was analyzed. For tcdA, genome alignment analysis showed that all clinical isolates and strain 630 shared approximately 98–100% amino acid sequence identity. The C-terminal portion of the tcdA gene showed slightly more diversity. The same analysis was performed on the tcdB gene, which exhibited even greater sequence diversity. Notably, strains 2007838/NAP7/126 and 2007858/NAP1/unk5 showed the greatest diversity patterns compared to strain 630 in both the N-terminal (79–100%) and C-terminal (88–100%; data not shown) domains.

Hamster serum banks (HS) were collected from Syrian golden hamsters immunized with immunogens including mutant TcdA (SEQ ID NO:4) and mutant TcdB (SEQ ID NO:6), wherein the mutant toxins were inactivated using EDC according to, for example, Example 29 above, Table 15, and prepared with aluminum phosphate. The results in Table 35 show that, in in vitro neutralization assays, toxin B from at least the individual culture supernatants was neutralized by serum from the immunized hamsters.

Table 35. Description of Clostridium difficile strains from the CDC and their ability to neutralize various toxins in immunized hamster serum.

The level of beta-toxin was too low to allow for neutralization testing.

Figure 23 shows the results of a neutralization assay on an IMR-90 using toxin preparations from various Clostridium difficile strains. This data shows that the TcdB neutralizing antibody in hamster antiserum was able to neutralize toxins from all 21 isolates tested, including highly toxic strains and TcdA-negative, TcdB-positive strains. At least 16 different strains of Clostridium difficile were obtained from CDC (Atlanta, GA) (previously described) and cultured in Clostridium difficile media under suitable conditions (as known in the art and as described above). Culture supernatants containing secreted toxins were analyzed to determine their cytotoxicity ( _EC50 ) on IMR-90 monolayer cultures and subsequently tested at 4-fold _EC50 using various dilutions of hamster serum, wherein the hamsters were immunized with mutant toxin A and mutant toxin B (prepared with aluminum phosphate). Using the in vitro cytotoxicity assays described above, the cytotoxicity of crude toxins and purified toxins (commercially available toxins from List Biologicals) derived from the culture supernatants of each strain (not purified from the individual supernatants) against IMR-90 cells was tested.

In Figures 23A-K, the graphs show results from in vitro cytotoxicity assays (previously described), where ATP levels (RLU) are plotted against increasing concentrations for: *Clostridium difficile* culture medium and hamster serum bank (■); crude toxin and hamster serum bank (●); purified toxin and hamster serum bank (▲); crude toxin control; and purified toxin (◆), control. Toxins from each strain were added to the cells at 4 x _EC50 values.

As shown in Figures 23A-K, hamsters that received the immunogen surprisingly developed neutralizing antibodies exhibiting neutralizing activity against toxins from at least 16 different Clostridium difficile CDC strains (compared to controls containing only the individual toxins): 2007886 (Figure 23A); 2006017 (Figure 23B); 2007070 (Figure 23C); 2007302 (Figure 23D); 2007838 (Figure 23E); 2007886 (Figure 23F); 2009292 (Figure 23G); 2004013 (Figure 23H); 2009141 (Figure 23I); 2005022 (Figure 23J); 2006376 (Figure 23K). See also Table 35 for other Clostridium difficile strains from which toxins were tested and neutralized by an immunogenic composition containing mutant toxin A and mutant toxin B (formulated in aluminum phosphate).

In another study, serum from hamsters was used to test culture supernatants containing toxins secreted by various Clostridium difficile strains (obtained from CDC and Leeds Hospital, UK) in an in vitro neutralization assay. The hamsters were administered mutant toxin A and mutant toxin B (prepared with aluminum gel). The experimental design is shown in Table 36. Results are presented in Tables 37 and 38.

Example 38: Peptide mapping of EDC/NHS triple mutant toxin

To characterize the EDC/NHS-inactivated triple mutant toxins, peptide mapping experiments were performed on four sets of EDC/NHS-treated triple mutant toxin A (SEQ ID NO:4) and four sets of EDC/NHS-treated triple mutant toxin B (SEQ ID NO:6). After digestion of the mutant toxins with trypsin, the resulting peptide fragments were separated by reversed-phase HPLC. Mass spectrometry was used to identify modifications resulting from the inactivation process. For both mutant toxin A and mutant toxin B, more than 95% of the theoretical trypsin peptides were identified. Cross-linking and glycine adducts (glycine was used as a capping agent) were identified. β-alanine adducts were also observed in both mutant toxin A and mutant toxin B. Without being mechanistically or theoretically confined, the β-alanine adducts appear to derive from the reaction of three moles of NHS with one mole of EDC, in which NHS-activated β-alanine is formed. This molecule can then react with a lysine group to form a β-alanine adduct (+70 Da). Low levels (0.07 mol/mol protein) of dehydroalanine (-34 Da) were also observed in the EDC/NHS-treated triple mutant toxin B sample. Dehydroalanine is a result of desulfonation of cysteine residues. The same type and extent of modification was observed in all four batches of each mutant toxin, indicating that the method produces a consistent product. Peptide mapping (with over 95% sequence coverage) confirmed the presence of modification. A summary of the modifications is shown in Table 39. See also Figures 24-25. Furthermore, the size and charge heterogeneity of the triple mutant toxin A and triple mutant toxin B drug substances was increased compared to the individual triple mutant toxins A and B, which did not exhibit chemical inactivation. Consequently, size exclusion chromatography (SEC) and anion exchange chromatography (AEX) spectra showed relatively broad peaks (data not shown).

Table 39. Summary of modifications observed in mutant toxin drugs

The degree of modification was calculated by dividing the HPLC region of the modified peptide by the HPLC region of the natural + modified peptide.

Example 39: Production of pharmaceutical products

The Clostridium difficile immunogenic composition (pharmaceutical product) contains two active pharmaceutical ingredients (mutant toxin A and mutant toxin B). An exemplary pharmaceutical product is a lyophilized formulation containing 10 mM Tris buffer pH 7.4, 4.5% (w/w) trehalose dihydrate, and 0.01% (w/v) polysorbate 80, comprising each of mutant toxin A and mutant toxin B. See Table 40. The immunogenic composition is prepared for injection by resuspending the lyophilized vaccine in a diluent or an aluminum-based diluent. The placebo will contain a sterile normal saline solution (0.9% sodium chloride) for injection.

Table 40

Buffer preparation

Add water for injection (WFI) to the mixing tube. While mixing, add the excipient and dissolve it in the solution. Measure the pH. If necessary, adjust the pH to 7.4 ± 0.1 with HCl. Dilute the solution to the final weight with WFI, then filter using a 0.22 μm Millipore Express SHC XL150 filter. Take a pre-filtration bioburden reduction sample prior to filtration. Sample the filtered buffer for osmolality and pH.

Formulation preparation

Based on pre-calculated amounts, collect the thawed mutant toxin drug substance in preparation tubes according to the following sequence of operations: First, add 50% of the target dilution buffer volume required to achieve 0.6 mg/mL to the tube, then add mutant toxin A and mix at 100 rpm for 5 minutes. Next, add mutant toxin B to the tube and further dilute the solution to the 0.6 mg/mL dilution point, then mix again at 100 rpm for 5 minutes. Sample and test the total mutant toxin concentration. Based on the mutant toxin concentration value in the treatment, dilute the solution to 100% volume and mix at 100 rpm for 15 minutes. Sample the prepared drug product for pH and bioburden pre-filtration. Then filter the prepared drug product using a Millipore Express SHCXL150 for overnight storage, or transfer to a delivery tube for sterile filtration.

Transfer the prepared fraction to the filling area, sample for bioburden level, and then sterilize and filter using two tandem Millipore Express SHC XL150 filters. Fill the prepared fraction into pyrogen-free glass vials at a target volume of 0.73 mL. Seal the filled vials and load them into the freeze dryer. Perform lyophilization cycles as shown in Table 41. Upon completion of the cycle, backfill the freeze-drying chamber with nitrogen to 0.8 atm to fully seal the seal. Unload the chamber and cap the vials with a flip-off seal.

Table 41. Freeze-drying cycle setup points for Clostridium difficile drug products

The drug product stability data are summarized in Table 42. This data indicates that the drug product is physically and chemically stable when stored at 2–8°C for at least 3 months or at 25°C or 40°C for at least 1 month. Under both storage conditions, the impurity levels detected by size exclusion chromatography (SEC) remained unchanged, and the in vitro antigenicity also remained unchanged up to the last time point tested.

Table 42: Stability of Lyophilized Pharmaceutical Products ^a

Table 42: Stability of Lyophilized Pharmaceutical Products ^a

For these tests, the lyophilized DP was reconstituted using 60 mM NaCl diluent.

Example 40: Vaccine diluent

For saline solutions, 60 mM NaCl is used as a diluent for lyophilized adjuvant-free pharmaceutical products to ensure isotonic solutions upon reconstitution.

Aluminum glue: Aluminum glue "85" 2% (Brenntag) is a commercially available Good Manufacturing Practice (GMP) grade product consisting of octahedral crystalline sheets of aluminum hydroxide. Exemplary aluminum glue diluent formulations are shown in Table 43. This exemplary formulation may be used in combination with the pharmaceutical products described above.

Studies using aluminum gel adjuvant formulations demonstrated 100% binding of mutant toxin A and mutant toxin B to 1 mg Al/mL aluminum gel (pH 6.0 to 7.5). Maximum binding of both drugs was observed at the highest protein concentrations tested (300 μg/mL each).

Protein binding to aluminum gel was also tested using lyophilized drug product formulations (containing 200 μg/mL of each drug substance and aluminum gel ranging from 0.25 to 1.5 mg/mL). The drug products were reconstituted with diluents containing varying concentrations of aluminum gel, and the percentage of each mutant toxin bound was measured. All tested concentrations of aluminum gel demonstrated 100% binding to the antigen.

The binding kinetics of the protein to aluminum gel were also evaluated using target doses of mutant toxin A and mutant toxin B (200 μg/mL each). Results showed that 100% of the mutant toxins bound to the aluminum gel throughout the 24-hour response time (RT).

CpG 24555 and Aluminum Gel: CpG 24555 is a synthetic 21-meric oligodeoxynucleotide (ODN) having the sequence 5-TCG TCG TTTTTC GGT GCT TTT-3 (SEQ ID NO: 48). Exemplary formulations combining CpG 24555 and aluminum gel diluent are shown in Table 44. This exemplary formulation may be used in combination with the pharmaceutical products described above.

The adjuvant is a saponin-based adjuvant known in the art. Exemplary formulations of the adjuvant are shown in Table 45. This exemplary formulation may be used in combination with the pharmaceutical products described above.

Example 41: Immunogenicity and Proof of Concept of Mutant Toxin Drug Composition with Aluminum Gel as Adjuvant in NHP Model

The immunogenicity of mutant toxin drug substance compositions with aluminum gel as an adjuvant was evaluated in an NHP model (specifically, cynomolgus monkeys). NHP immunized with a combination of mutant toxin A and mutant toxin B drug substance (formulated with aluminum gel) at two-week intervals (weeks 0, 2, and 4) at doses of 10 μg each, developed strong neutralizing antitoxin responses. See Table 46. After the third immunization, both antitoxin A and antitoxin B neutralizing responses reached and remained within or above the protective range until at least week 33 (the last time point studied).

At weeks 0, 2, and 4, cynomolgus monkeys (n = 8) were immunized with 10 μg of each mutant toxin A drug substance and mutant toxin B drug substance (prepared with 250 μg of aluminum gel) IM. Serum was collected at each time point, and functional antitoxin activity was analyzed in a toxin neutralization assay. GMTs are provided in Table 46. The protective titer ranges provided in this table illustrate the range of neutralizing antibody titers corresponding to a significant reduction in Clostridium difficile infection relapse in the Merck monoclonal antibody treatment trial.

Correspondence between the human protective antibody titers in the Merck mAb treatment trial and the titers induced by Pfizer’s vaccine candidate in NHP.

A phase II efficacy study using Merck/Medarex mAbs (Lowy et al., N Engl J Med. 2010 Jan 21; 362(3): 197-205) appears to demonstrate a correlation between serum levels of neutralizing antitoxin mAbs and CDAD relapse prevention. Following administration of toxin-specific mAbs to humans, serum antibody levels in human receptors in the range of 10–100 μg/mL showed protection against relapse (70% reduction in CDAD relapse).

Immunogenic compositions containing mutant toxin drug substances were tested to determine whether they could induce a robust neutralizing antibody response in humans, by comparing data published in the Merck/Medarex Phase 2 study with antibody levels induced by the immunogenic compositions in an NHP model. This was achieved by using the characteristics of previously published Merck/Medarex mAbs to convert the range (10–100 μg/mL) of these mAbs obtained from the serum of subjects showing no signs of relapse to a 50% neutralizing titer, and comparing these titers (“protective titer range”) with those observed in the preclinical model described herein. As shown in Table 46, after the third dose, the immunogenic compositions containing mutant toxin A and mutant toxin B drug substances (with aluminum gel as an adjuvant) generated an immune response in NHP that reached the “protective titer range” and remained within or above this range until week 33. The levels of toxin-neutralizing antibodies induced in NHP by the Clostridium difficile immunogenic composition of the present invention are comparable to the serum antibody levels in the Merck/Medarex test subjects (who showed protection against CDAD relapse).

Example 42: Immunogenicity of mutant toxin drug composition with ISCOMATRIX or aluminum gel/CpG 24555 (Alh/CpG) as adjuvant in NHP model

In NHP, both ISCOMATRIX and Alh/CpG statistically significantly enhanced the neutralizing titers of antitoxins A and B (compared to vaccines administered with aluminum gel alone) (Table 47). Vaccines administered with Alh/CpG or ISCOMATRIX elicited an above-ground antitoxin response at an earlier time point (weeks 2–4) (compared to aluminum gel alone (weeks 4–6), which can play an important role in protection against CDAD relapse in humans). Immunogenic compositions with Alh/CpG or ISCOMATRIX as adjuvants generated antitoxin neutralizing titers reaching the protective range more rapidly than aluminum gel (see also Example 41) and remained within or above this range until week 33.

As shown in Table 47, cynomolgus monkeys were immunized at weeks 0, 2, and 4 with 10 μg of each mutant toxin A drug substance and mutant toxin B drug substance IM, wherein the drug substance was prepared with 250 μg of aluminum gel (n=8), or with 500 μg of CpG + 250 μg of aluminum gel (n=10), or with 45 U of ISCANATRIX (n=10). Serum was collected at each time point, and functional antitoxin activity was analyzed in the toxin neutralization assay described above. GMT is listed in the table. An asterisk (*) indicates statistical significance when comparing titers with the aluminum gel group (p<0.05). The protective titer range represents the range of neutralizing antibody titers corresponding to a significant reduction in Clostridium difficile infection relapse in the Merck/Medarex mAb treatment trial.

The doses of mutant toxin A and mutant toxin B administered (in the presence of ISCMATRIX or Alh/CpG adjuvant) were also evaluated to assess the neutralizing antitoxin antibody titers generated in NHP. In one study, NHP was treated with either a low dose (10 μg) or a high dose (100 μg) of each mutant toxin (prepared with ISCMATRIX). Responses were compared at each time point post-immunization. As shown in Table 48, antitoxin neutralizing titers were strong in both treatment groups. At most time points, antitoxin A titers were nearly equal between the low and high dose groups, while antitoxin B titers tended to be higher in the high-dose group.

As shown in Table 48, cynomolgus monkeys (n=5) were immunized at weeks 0, 2, and 4 with 10 μg or 100 μg of each mutant toxin A drug substance and mutant toxin B drug substance (prepared in 45 U of ISCANATRIX) IM. Serum was collected at each time point, and functional antitoxin activity was analyzed in a toxin neutralization assay. GMT is listed in the table. The protective titer range represents the range of neutralizing antibody titers corresponding to a significant reduction in Clostridium difficile infection relapse in the Merck/Medarex mAb treatment trial.

In attempts to enhance the kinetics of the antitoxin B response, NHP was immunized with a constant dose of mutant toxin A (10 μg) mixed with elevated doses of mutant toxin B (10, 50, or 100 μg) in the presence of either ISCMATRIX or Alh/CpG adjuvant. Regardless of the adjuvant, the group receiving higher doses of mutant toxin B (50 or 100 μg) tended to induce a higher antitoxin B neutralizing response (compared to the 10 μg dose of mutant toxin B) (Table 50, marked with * to indicate statistically significant improvement). This trend was observed at most time points after the final immunization. However, in some cases, the antitoxin A neutralizing response showed a statistically significant decrease as the amount of mutant toxin B increased (marked with ^ in Table 49).

As shown in Tables 49 and 50, NHP was immunized at weeks 0, 2, and 4 with different ratios of mutant toxin A and mutant toxin B (in ISCMATRIX (45 U per dose) or Alh/CpG (250 μg/500 μg per dose)) IM (10 per group), where 10 μg of mutant toxin A was used plus 10, 50, or 100 μg of mutant toxin B (named 10A:10 in Tables 49 and 50, respectively). B, 10A:50B, and 10A:100B. Table 49 shows the antitoxin A titer. Table 50 shows the antitoxin B titer. GMT is listed in the tables. The protective titer range represents the range of neutralizing antibody titers corresponding to a significant reduction in Clostridium difficile infection relapse in the Merck mAb treatment trial. The symbol ^ indicates a statistically significant decrease in neutralizing titer compared to the 10A:10B group (p<0.05). The asterisk * indicates a statistically significant increase in neutralizing titer compared to the 10A:10B group (p<0.05).

Example 43: Toxicity study of five-week repeated-dose administration of the immunogenic composition in cynomolgus monkeys, including a 4-week recovery period.

A five-week repeated-dose toxicity study was conducted in cynomolgus monkeys using PF-06425095 (an immunogenic composition comprising triple mutant toxin A and triple mutant toxin B, combined with adjuvant aluminum hydroxide and CpG 24555) to evaluate the potential toxicity and immunogenicity of Clostridium difficile triple mutant toxin A and triple mutant toxin B combined with aluminum hydroxide and CpG 24555 adjuvant (PF-06425095). The following were administered as initial immunization to cynomolgus monkeys (6 animals/sex/group): PF-06425095 at 0.2 or 0.4 mg/dose of triple mutant toxin A and triple mutant toxin B (low and high dose immunogenic composition groups, respectively), 0.5 mg aluminum as aluminum hydroxide, and 1 mg CpG 24555, as well as adjuvant-only combinations (aluminum hydroxide + CpG 24555; PF-06376915), followed by three booster doses (days 1, 8, 22, and 36). A separate group of animals (6 animals/sex) received 0.9% isotonic saline (pH approximately 7.0). The immunogenic composition carrier consisted of 10 mM Tris buffer (pH 7.4), 4.5% trehalose dihydrate, and 0.1% polysorbate 80. The adjuvant control carrier consisted of 10 mM histidine buffer (with 60 nM NaCl, pH 6.5). The total administration volume was 0.5 mL per injection. All administrations were given to the left and/or right quadriceps femoris muscle. The selected animals underwent a 4-week dosing-free observation period to assess the reversibility of any effects observed during the dosing period of this study.

No adverse findings were observed in this study. PF-06425095 was well tolerated and produced only local inflammatory responses with no evidence of systemic toxicity. During the dosing period, the following increases (dose-dependent) were observed in the immunogenic composition-treated group from pretests: fibrinogen (23.1% to 2.3x) on days 4 and 38, C-reactive protein (2.1x to 27.5x) on day 4 and 38 (2.3x to 101.5x), and globulin (11.1% to 24.1%) on day 36 and/or 38; and were consistent with the expected inflammatory response to administration of the adjuvanted immunogenic composition.

The increases in fibrinogen and C-reactive protein observed on day 4 partially recovered on day 8, with increases in fibrinogen (25.6% to 65.5%) and C-reactive protein (4.5x and 5.6x) occurring only in the high-dose immunogenic composition group. Increases in interleukin (IL)-6 were observed on day 1, hour 3 (8.3x to 127.2x individual values at hour 0 of day 1, administration response) and on day 36, hour 3 (9.4x to 39.5x individual values at hour 0 of day 36) in both the low-dose and high-dose immunogenic composition groups. No changes were observed in other cytokines (IL-10, IL-12, interferon-induced protein (IP-10), and tumor necrosis factor-α (TNF-α). Elevations in these acute-phase proteins and cytokines are partly expected normal physiological responses to the administration of exogenous antigens. During the recovery phase (when cytokines were not assessed), these clinicopathological parameters showed no PF 06425095-related or adjuvant-related changes. Additionally, local changes were observed at the injection site, with similar incidence and severity in the adjuvant control group and in the low- and high-dose immunogenic composition groups; therefore, they are not directly related to PF-06425095. During the administration period, changes included mild to moderate chronic active inflammation characterized by myofibril dissociation resulting from macrophage infiltration, typically containing basophilic granules (interpreted as aluminum-containing adjuvant), lymphocytes, and other abnormalities. Lymphocytes, plasma cells, neutrophils, eosinophils, necrotic debris, and edema were observed. Extracellular basophilic material was also present within these lesions of chronic active inflammation. At the end of the recovery period, mild to moderate chronic inflammation and mononuclear cell infiltration, along with mild fibrosis, were observed. These findings at the injection site represent a local inflammatory response to the adjuvant. Other microscopically visible changes included a mild to moderate increase in lymphocyte structure at the iliac (draining) lymph nodes and a mild increase in cellular structure in the splenic germinal centers, which were observed during the administration period in the adjuvant control group and in the low- and high-dose immunogenic combination groups. These microscopically visible findings were of low severity at the end of the recovery period. These effects represent an immune response to antigenic stimulation and a pharmaceutical response to the adjuvant or PF-06425095. No test-related increases were observed in anti-DNA antibodies.

Given the absence of any adverse findings, the no significant adverse effect level (NOAEL) in this study was four doses administered in two 0.5 mL injections in the high-dose immunogenic combination group (0.4 mg of triple mutant toxin A and triple mutant toxin B/dose as PF-06425095).

Example 44: Efficacy of passively transferred seropositive NHP serum to hamsters

Five Syrian golden hamsters were administered oral clindamycin antibiotics (30 mg/kg) to disrupt the normal gut flora. Five days later, the hamsters were challenged with oral wild-type Clostridium difficile spores (630 strains, 100 CFU/animal) and intraperitoneally (IP) NHP serum according to Table 51. Regardless of the mechanism or theory, disease symptoms following spore challenge typically began to appear approximately 30–48 hours after the challenge.

Following three immunizations with mutant toxin A and mutant toxin B (A:B ratios of 10:10, 10:50, and 10:100, formulated with ISCANATRIX, see Example 42, Tables 49 and 50), NHP serum samples exhibiting the highest titers (antitoxin A serum and antitoxin B serum) were collected and administered to hamsters. NHP serum was collected at time points of 5, 6, and 8 weeks (immunizations were performed at 0, 2, and 4 weeks), as described in Example 42. The results are shown in Tables 52-54 below. The symbol “+” indicates the geometric mean (GM) excluding animals #3 nonresponders. “*TB” represents the day the final blood collection was performed and the animal was euthanized; not all animals were euthanized on the same day.

In another study, Syrian golden hamsters were administered oral clindamycin antibiotics (30 mg/kg) to disrupt the normal gut flora. Five days later, the hamsters were attacked with orally administered wild-type Clostridium difficile spores (630 strains, 100 CFU/animal) and NHP serum was administered intraperitoneally (IP) according to Table 55. Regardless of the mechanism or theory, disease symptoms following spore attack typically begin to appear approximately 30–48 hours after the attack.

Following three immunizations with mutant toxin A and mutant toxin B (in A:B ratios of 10:10, 10:50, and 10:100, formulated with aluminum gel and CpG 24555, see Example 42, Tables 49 and 50), NHP serum was collected from samples taken from NHP and administered to hamsters. NHP serum was collected at time points of 5, 6, 8, and 12 weeks, as described in Example 42 (immunization of NHP at weeks 0, 2, and 4). The results are shown in Tables 56-59 below. Further studies of the hamster serum were conducted to determine the inhibitory concentration ( _IC50 ) value, which was determined using the toxin neutralization assay described above. The toxin-neutralizing antibody levels induced in hamsters by the Clostridium difficile immunogenic composition of the present invention were comparable to the serum antibody levels in the Merck/Medarex test subjects (who exhibited protection against CDAD relapse).

* = Decrease on that day

Example 45: Identification of mutant toxin drug substances

The primary structure of the triple mutant toxin A is shown in SEQ ID NO:4. The NH ₂ -terminal Met residue at position 1 of SEQ ID NO:4 is derived from the start codon of SEQ ID NO:12 and is deleted in isolated proteins (e.g., see SEQ ID NO:84). Therefore, in Examples 12 to 45, “SEQ ID NO:4” refers to SEQ ID NO:4 in which the starting methionine (at position 1) is deleted. The purified triple mutant toxin A (SEQ ID NO:4) (intermediate drug substance - Lot L44993-132) and the EDC/NHS-treated triple mutant toxin A (SEQ ID NO:4) (“mutant toxin A drug substance” - Lot L44898-012) both exhibit a single NH ₂ -terminal sequence starting with SLISKEELIKLAYSI (positions 2-16 of SEQ ID NO:4).

The primary structure of the triple mutant toxin B is shown in SEQ ID NO:6. The NH ₂ -terminal Met residue at position 1 of SEQ ID NO:6 is derived from the start codon of SEQ ID NO:12 and is deleted in isolated proteins (e.g., see SEQ ID NO:86). Therefore, in Examples 12 to 45, “SEQ ID NO:6” refers to SEQ ID NO:6 in which the starting methionine (at position 1) is deleted. Both the purified triple mutant toxin B (SEQ ID NO:6) (intermediate drug substance - Lot 010) and the EDC/NHS-treated triple mutant toxin B (SEQ ID NO:6) (“Mutant Toxin B Drug Substance” - Lot L44906-153) exhibit a single NH ₂ -terminal sequence starting with SLVNRKQLEKMANVR (positions 2-16 of SEQ ID NO:6).

Circular dichroism (CD) spectroscopy was used to evaluate the secondary and tertiary structures of triple mutant A (SEQ ID NO:4) and mutant toxin A. CD spectroscopy was also used to evaluate the secondary and tertiary structures of triple mutant B (SEQ ID NO:6) and mutant toxin B. CD spectroscopy was also used to assess the potential effect of pH on the structures. The effect of EDC treatment on triple mutant toxin A was analyzed by comparing CD data obtained from mutant toxin A with data obtained from triple mutant toxin A. The effect of EDC treatment on triple mutant toxin B (SEQ ID NO:6) was analyzed by comparing CD data obtained from mutant toxin B with data obtained from triple mutant toxin B.

Far-UV CD data of the mutant toxin A drug substance were obtained at various pH values. Spectra recorded at pH 5.0–7.0 indicate a high proportion of α-helices in the secondary structure, suggesting that the protein's polypeptide backbone adopts a well-defined conformation dominated by α-helices.

Near-UV CD spectra of the mutant toxin A were also obtained. The strong negative ellipticity between 260-300 nm indicates that the aromatic side chains are in a uniquely rigid environment, suggesting that the mutant toxin A possesses a tertiary structure. In fact, characterizing features derived from individual aromatic side chain types can be distinguished in the spectrum: the shoulder at ~290 nm and the maximum negative peak at ~283 nm are due to the absorption of polarized light by the ordered tryptophan side chains, the negative peak at 276 nm originates from the tyrosine side chains, while the small shoulders at 262 and 268 nm indicate phenylalanine residues involved in three-dimensional contact. Far- and near-UV results provide evidence that the mutant toxin A maintains its fully folded structure at physiological pH. Almost identical far- and near-UV CD spectra observed at pH 5.0–7.0 indicate that no detectable structural changes occurred within this pH range. CD data could not be collected at pH 3.0 and 4.0 because the protein is insoluble at these pH points. In comparative far- and near-UV CD spectra of mutant toxin A and triple mutant toxin A, the spectra of the two proteins were essentially identical under all studied experimental conditions, indicating that EDC treatment had no detectable effect on the secondary and tertiary structures of triple mutant toxin A. This finding is consistent with the results of gel filtration and analytical ultracentrifugation, which showed no detectable changes in Stokes radius and sedimentation/friction coefficient, respectively.

The mutant toxin A drug substance (and the triple mutant toxin A) contains 25 tryptophan residues distributed throughout the primary sequence and can be used as a convenient internal fluorescent probe. Fluorescence emission spectra of the mutant toxin A drug substance as a function of temperature were obtained in the 300–400 nm range. At 6.8 °C and excited at 280 nm, the mutant toxin A drug substance exhibits a characteristic tryptophan fluorescence chromophore. The fluorescence emission maximum was observed at ~335 nm, indicating that the tryptophan residues are in a nonpolar environment, typical of the protein's internal environment rather than a polar aqueous environment. These fluorescence emission spectral results, together with the CD experimental results presented in this report, confirm that the mutant toxin A drug substance maintains a fully folded structure.

The fluorescence of the external probe 8-anilino-1-naphthalenesulfonic acid (ANS) was used to characterize the possible conformational changes in mutant toxin A and triple mutant toxin A under pH variations. The results showed that when mutant toxin A or triple mutant toxin A was titrated with this probe at pH 7.0, the ANS fluorescence density did not increase significantly, indicating that no hydrophobic surface was exposed on the protein under these conditions. Changing the pH to 2.6 caused a dramatic change in the ANS fluorescence quantum field with increasing probe concentration until the fluorescence quantum field reached apparent saturation. This increase in the ANS fluorescence quantum field indicates that at low pH (2.6), both mutant toxin A and triple mutant toxin A underwent pH-induced conformational changes to expose hydrophobic surfaces. Such conformational changes suggest that EDC-induced modification and inactivation of triple mutant toxin A do not limit the conformational plasticity of mutant toxin A (DS).

The effect of EDC treatment on the hydrodynamic properties of triple mutant toxin A was evaluated using size exclusion chromatography on a G4000SWXL column. Mutant toxin A and triple mutant toxin A were injected onto a G4000SWXL column (equilibrated at pH 7.0, 6.0, and 5.0). The data showed that size exclusion chromatography failed to detect any difference in Stokes radius between the mutant toxin A and triple mutant toxin A. Therefore, EDC treatment does not drastically affect the hydrodynamic properties of triple mutant toxin A, and consequently, it does not drastically affect its overall molecular shape.

Further analysis of the triple mutant toxin A and the mutant toxin A drug substance was performed using multi-angle laser scattering (MALLS) technology. EDC treatment of the triple mutant toxin A resulted in a heterogeneous mixture composed of various polymers and monomers. This heterogeneity reflects the introduction of numerous EDC-induced intermolecular and intramolecular covalent bonds (between carboxyl groups and the primary amines of the protein).

The obtained data provide the physical and chemical characterization of triple mutant toxin A and the mutant toxin A drug substance (EDC-treated triple mutant toxin A), and describe key features in its primary, secondary, and tertiary structures. The data demonstrate that EDC treatment of triple mutant toxin A results in covalent modification of its polypeptide chain, but does not affect the protein's secondary and tertiary structures. EDC treatment leads to intramolecular and intermolecular crosslinks. The biochemical and biophysical parameters obtained for the mutant toxin A drug substance (and triple mutant toxin A) are given in Table 60.

Far-UV CD data of the mutant toxin B drug substance were obtained at various pH values. The spectra recorded at pH 5.0–7.0 indicate a high proportion of α-helices in the secondary structure, suggesting that the protein's polypeptide backbone adopts a well-defined conformation dominated by α-helices.

Near-UV CD spectra of the mutant toxin B were also obtained. The strong negative ellipticity between 260-300 nm indicates that the aromatic side chains are in a uniquely rigid environment, suggesting that the mutant toxin B possesses a tertiary structure. In fact, characterizing features derived from individual aromatic side chain types can be distinguished in the spectrum: the shoulder at ~290 nm and the maximum negative peak at ~283 nm are due to the absorption of polarized light by the ordered tryptophan side chains, the negative peak at 276 nm originates from the tyrosine side chains, while the small shoulders at 262 and 268 nm indicate phenylalanine residues involved in three-dimensional contact. Far- and near-UV results provide evidence that the mutant toxin B maintains its fully folded structure at physiological pH. Very similar far- and near-UV CD spectra observed at pH 5.0–7.0 indicate that no detectable structural changes occurred within this pH range. CD data could not be collected at pH 3.0 and 4.0 because the protein is insoluble at these pH points.

In comparative far- and near-UV CD spectra of mutant toxin B and triple mutant toxin B, the spectra of the two proteins were very similar between pH 5.0 and 7.0, indicating that EDC treatment had no detectable effect on the secondary and tertiary structures of the protein.

The triple mutant toxin B contains 16 tryptophan residues distributed throughout the primary sequence and can be used as a convenient internal fluorescent probe. Fluorescence emission spectra of the mutant toxin B drug substance as a function of temperature were obtained in the 300–400 nm range. At 7 °C and excited at 280 nm, the mutant toxin B drug substance exhibited a characteristic tryptophan fluorescence chromophore. The fluorescence emission maxima were observed at ~335 nm, indicating that the tryptophan residues are in a nonpolar environment, typical of the protein's internal environment rather than a polar aqueous environment. This result, together with the results of the CD experiment (see above), confirms that the mutant toxin B drug substance maintains a fully folded structure.

The fluorescence of the external probe 8-anilino-1-naphthalenesulfonic acid (ANS) was used to characterize the possible conformational changes in mutant toxin B and triple mutant toxin B under varying pH conditions. The results showed that when mutant toxin B or triple mutant toxin B was titrated with this probe at pH 7.0, the ANS fluorescence density did not increase substantially, indicating that no hydrophobic surface was exposed on the protein under these conditions. Changing the pH to 2.6 caused a dramatic change in the ANS fluorescence quantum field (in the presence of mutant toxin B) with increasing probe concentration until the fluorescence quantum field reached apparent saturation. This increase in the ANS fluorescence quantum field indicates that at low pH (2.6), mutant toxin B underwent a pH-induced conformational change exposing a hydrophobic surface. This type of conformational change suggests that EDC-induced modification and inactivation of triple mutant toxin B do not limit the conformational plasticity of mutant toxin B (DS).

The effect of EDC treatment on the hydrodynamic properties of triple mutant toxin B was evaluated using size exclusion chromatography on a G4000SWXL column. Mutant toxin B and triple mutant toxin B were injected into the G4000SWXL column (equilibrated at pH 7.0, 6.0, and 5.0). The data indicated that size exclusion chromatography failed to detect any difference in Stokes radius between the mutant toxin B and triple mutant toxin B; therefore, EDC treatment does not drastically affect the hydrodynamic properties of the protein, and consequently, does not drastically affect its overall molecular shape.

Further analysis of the triple mutant toxin B and the mutant toxin B drug substance was conducted using multi-angle laser scattering (MALLS) technology. EDC treatment of the triple mutant toxin B resulted in a more heterogeneous mixture composed of various polymers and monomers. This heterogeneity reflects the introduction of numerous EDC-induced intermolecular and intramolecular covalent bonds (between carboxyl groups and the primary amines of the protein).

The obtained data provide the physical and chemical characterization of triple mutant toxin B and the mutant toxin B drug substance (EDC-treated triple mutant toxin B), and describe key features in its primary, secondary, and tertiary structures. The data demonstrate that EDC treatment of triple mutant toxin B results in covalent modification of its polypeptide chain, but does not affect the protein's secondary and tertiary structures. EDC treatment leads to intramolecular and intermolecular crosslinks. The main biochemical and biophysical parameters obtained for the mutant toxin B drug substance (and triple mutant toxin B) are given in Table 61.

Aspects of the present invention

The following entries describe other embodiments of the invention:

C1. An isolated polypeptide comprising the amino acid sequence shown in SEQ ID NO:4, wherein the methionine residue at position 1 is optionally absent, and wherein the polypeptide comprises at least one amino acid side chain chemically modified with 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) and N-hydroxysuccinimide (NHS).

C2. An isolated polypeptide comprising the amino acid sequence shown in SEQ ID NO:6, wherein the methionine residue at position 1 is optionally absent, and wherein the polypeptide comprises amino acid side chains chemically modified with 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) and N-hydroxysuccinimide (NHS).

C3. The isolated polypeptide of entry C1 or C2, wherein at least one aspartic acid residue side chain of the polypeptide or at least one glutamic acid residue side chain of the polypeptide is chemically modified with glycine.

C4. The isolated polypeptide of any one of items C1-C3, wherein the polypeptide includes:

a) At least one crosslink between the aspartic acid side chain and the lysine side chain of the polypeptide; and

b) At least one crosslink between the glutamic acid residue side chain and the lysine residue side chain of the polypeptide.

C5. An isolated polypeptide of any one of entries C1-C4, wherein the polypeptide comprises a β-alanine moiety attached to at least one lysine residue side chain of the polypeptide.

C6. The isolated polypeptide of entry C4, wherein the polypeptide comprises an aspartic acid residue side chain attached to the polypeptide or a glycine moiety attached to a glutamic acid residue side chain of the polypeptide.

C7. An isolated polypeptide comprising the amino acid sequence shown in SEQ ID NO:4, wherein the methionine residue at position 1 is optionally absent, and wherein at least one lysine residue of the polypeptide is linked to a β-alanine moiety via a side chain.

C8. An isolated polypeptide comprising the amino acid sequence shown in SEQ ID NO:6, wherein the methionine residue at position 1 is optionally absent, and wherein at least one lysine residue of the polypeptide is linked to a β-alanine moiety via a side chain.

C9. A separate polypeptide of entry C7 or C8, wherein the side chain of the second lysine residue of the polypeptide is linked to the side chain of an aspartic acid residue or to the side chain of a glutamic acid residue.

C10. An isolated polypeptide of any one of items C7-C9, wherein the aspartic acid residue side chain or the glutamic acid residue side chain of the polypeptide is linked to the glycine moiety.

C11. An isolated polypeptide of any one of entries C1-C10, wherein the polypeptide has an EC ₅₀ of at least about 100 μg/ml.

C12. An immunogenic composition comprising an isolated polypeptide having the amino acid sequence shown in SEQ ID NO:4, wherein the methionine residue at position 1 is optionally absent, and an isolated polypeptide having the amino acid sequence shown in SEQ ID NO:6, wherein the methionine residue at position 1 is optionally absent, and wherein said polypeptide has at least one amino acid side chain chemically modified with 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) and N-hydroxysuccinimide (NHS).

C13. An immunogenic composition of entry C12, wherein the polypeptide comprises at least one of the following:

a) At least one β-alanine moiety is attached to the lysine side chain of the polypeptide;

b) At least one crosslink between the aspartic acid residue side chain and the lysine residue side chain of the polypeptide;

and

c) At least one crosslink between the glutamate residue side chain and the lysine residue side chain of the polypeptide.

C14. Immunogenic composition of entry C12, wherein the polypeptide has an _EC50 of at least about 100 μg/ml.

C15. An immunogenic composition comprising an isolated polypeptide having the amino acid sequence shown in SEQ ID NO:4, wherein the methionine residue at position 1 is optionally absent, and an isolated polypeptide having the amino acid sequence shown in SEQ ID NO:6, wherein the methionine residue at position 1 is optionally absent, and

a) wherein at least one lysine residue side chain of SEQ ID NO:4 is connected to the β-alanine moiety, and

b) wherein at least one lysine residue of SEQ ID NO:6 is linked to the β-alanine moiety via a side chain.

C16. The immunogenic composition of entry C15, wherein the second lysine residue side chain of SEQ ID NO:4 is linked to an aspartic acid residue side chain or a glutamate residue side chain, and wherein the second lysine residue of SEQ ID NO:6 is linked to an aspartic acid residue side chain or a glutamate residue side chain.

C17. An immunogenic composition of any one of items C12-C16, wherein the aspartic acid residue side chain or the glutamic acid residue side chain of the polypeptide is linked to the glycine moiety, wherein the polypeptide has the amino acid sequence shown in SEQ ID NO:4, wherein the methionine residue at position 1 is optionally absent.

C18. An immunogenic composition of any one of items C12-C16, wherein the aspartic acid residue side chain or the glutamic acid residue side chain of the polypeptide is linked to the glycine moiety, wherein the polypeptide has the amino acid sequence shown in SEQ ID NO:6, wherein the methionine residue at position 1 is optionally absent.

C19. An immunogenic composition of any one of entries C12-C18, wherein the polypeptide has an _EC50 of at least about 100 μg/ml.

C20. An immunogenic composition comprising an isolated polypeptide having the amino acid sequence shown in SEQ ID NO:84 and an isolated polypeptide having the amino acid sequence shown in SEQ ID NO:86, wherein each polypeptide comprises:

a) At least one crosslink between the aspartic acid residue side chain and the lysine residue side chain of the polypeptide;

b) At least one crosslink between the glutamic acid residue side chain and the lysine residue side chain of the polypeptide;

c) A β-alanine moiety attached to at least one lysine residue side chain of the polypeptide; and

d) A glycine moiety attached to at least one aspartic acid residue side chain of the polypeptide or at least one glutamic acid residue side chain of the polypeptide.

C21. An immunogenic composition comprising a mutant Clostridium difficile toxin A, relative to the corresponding wild-type Clostridium difficile toxin A, said mutant Clostridium difficile toxin A comprising at least one mutated glucosyltransferase domain and at least one mutated cysteine protease domain.

C22. The composition of entry C21, wherein the mutation is a non-conservative amino acid substitution.

C23. The composition of entry C22, wherein the substitution includes alanine substitution.

C24. A composition of any one of entries C21-C23, wherein the wild-type Clostridium difficile toxin A comprises a sequence having at least 95% identity with SEQ ID NO:1.

C25. The composition of entry C24, wherein the wild-type Clostridium difficile toxin A comprises a sequence having at least 98% identity with SEQ ID NO:1.

C26. The composition of entry C25, wherein the wild-type Clostridium difficile toxin A comprises SEQ ID NO:1.

C27. A composition of any one of entries C21-C26, wherein the glucosyltransferase domain comprises at least two mutations.

C28. The composition of entry C27, wherein, according to the numbering of SEQ ID NO:1, the at least two mutations are present at amino acid positions 101, 269, 272, 285, 287, 269, 272, 460, 462, 541 or 542.

C29. A composition of any one of entries C21-C26, wherein the glucosyltransferase domain includes SEQ ID NO:29.

The composition of entry C29, C30, wherein the glucosyltransferase domain comprises at least two nonconserved mutations present at amino acid positions 101, 269, 272, 285, 287, 269, 272, 460, 462, 541, or 542, or any combination thereof, of SEQ ID NO:29.

C31. A composition of any one of entries C21-C26, wherein, according to the number SEQ ID NO:1, the cysteine protease domain comprises a mutation present at positions 700, 589, 655, 543, or any combination thereof.

C32. A composition of any one of entries C21-C26, wherein the cysteine protease domain includes SEQ ID NO:32.

C33. The composition of entry C32, wherein the cysteine protease domain comprises a non-conserved mutation, said non-conserved mutation being present at positions 1, 47, 113, 158, or any combination thereof, of SEQ ID NO:32.

C34. The composition of entry C21, wherein the mutant Clostridium difficile toxin A comprises SEQ ID NO:4.

C35. The composition of entry C21, wherein the mutant Clostridium difficile toxin A comprises SEQ ID NO: 84.

C36. The composition of entry C21, wherein the mutant Clostridium difficile toxin A comprises SEQ ID NO:7.

C37. The composition of entry C21, wherein the mutant Clostridium difficile toxin A comprises SEQ ID NO: 83.

C38. A composition of any one of items C21-C33, wherein at least one amino acid of the mutant Clostridium difficile toxin A is chemically cross-linked.

C39. The composition of entry C38, wherein the amino acid is chemically cross-linked by formaldehyde.

The composition of entry C38, C40, wherein the amino acid is chemically cross-linked by 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide.

C41. Compositions of entries C38 or C40, wherein the amino acid is chemically cross-linked by N-hydroxysuccinimide.

C42. A composition of any one of items C21-C41, wherein the composition is recognized by an antitoxin A neutralizing antibody or a binding fragment thereof.

C43. An immunogenic composition comprising a mutant Clostridium difficile toxin A, relative to the corresponding wild-type Clostridium difficile toxin A, said mutant Clostridium difficile toxin A comprising: a glucosyltransferase domain comprising SEQ ID NO:29 having amino acid substitutions at positions 285 and 287; and a cysteine protease domain comprising SEQ ID NO:32 having amino acid substitutions at position 158; wherein at least one amino acid of said mutant Clostridium difficile toxin A is chemically cross-linked.

C44. An immunogenic composition comprising SEQ ID NO:4 or SEQ ID NO:7, wherein at least one amino acid of SEQ ID NO:4 or SEQ ID NO:7 is chemically cross-linked.

C45. A composition of entry C43 or C44, wherein at least one amino acid is cross-linked by formaldehyde.

C46. A composition of entry C43 or C44, wherein the at least one amino acid is cross-linked by 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide.

C47. A composition of entries C43, C44, or C46, wherein at least one amino acid is cross-linked by N-hydroxysuccinimide.

C48. A composition of entry C43 or C44, wherein the composition is recognized by an antitoxin A neutralizing antibody or a binding fragment thereof.

C49. An immunogenic composition comprising SEQ ID NO:4.

C50. An immunogenic composition comprising SEQ ID NO:84.

C51. An immunogenic composition comprising SEQ ID NO:7.

C52. Immunogenic compositions comprising SEQ ID NO:83.

C53. A composition of any one of items C49-C52, wherein at least one amino acid is chemically cross-linked.

C54. A composition of any one of items C21-C51, wherein the composition exhibits reduced cytotoxicity relative to the corresponding wild-type Clostridium difficile toxin A.

C55. Includes the isolated polypeptide of SEQ ID NO:84.

C56. Includes the isolated polypeptide of SEQ ID NO:86.

C57. Includes the isolated polypeptide of SEQ ID NO:83.

C58. Includes the isolated polypeptide of SEQ ID NO:85.

C59. An immunogenic composition comprising a mutant Clostridium difficile toxin B, relative to the corresponding wild-type Clostridium difficile toxin B, said mutant Clostridium difficile toxin B comprising at least one mutated glucosyltransferase domain and at least one mutated cysteine protease domain.

The composition of C60, entry C59, wherein the mutation is a non-conservative amino acid substitution.

Compositions of entry C60, wherein the substitution includes alanine substitution.

C62. A composition of any one of entries C59-C61, wherein the wild-type Clostridium difficile toxin B comprises a sequence having at least 95% identity with SEQ ID NO:2.

C63. The composition of entry C62, wherein the wild-type Clostridium difficile toxin B comprises a sequence having at least 98% identity with SEQ ID NO:2.

C64. The composition of entry C63, wherein the wild-type Clostridium difficile toxin B comprises SEQ ID NO:2.

C65. A composition of any one of entries C59-C64, wherein the glucosyltransferase domain comprises at least two mutations.

C66. The composition of entry C65, wherein, according to the numbering of SEQ ID NO:2, the at least two mutations are present at amino acid positions 102, 286, 288, 270, 273, 384, 461, 463, 520, or 543.

C67. A composition of any one of entries C59-C64, wherein the glucosyltransferase domain includes SEQ ID NO:31.

C68. The composition of entry C67, wherein the glucosyltransferase domain comprises at least two nonconserved mutations present at amino acid positions 102, 286, 288, 270, 273, 384, 461, 463, 520, or 543 of SEQ ID NO:31.

C69. A composition of any one of entries C59-C64, wherein, according to the number SEQ ID NO:2, the cysteine protease domain comprises a mutation present at positions 698, 653, 587, 544, or any combination thereof.

C70. A composition of any one of entries C59-C64, wherein the cysteine protease domain includes SEQ ID NO:33.

C71. The composition of entry C70, wherein the cysteine protease domain comprises a non-conserved mutation, said non-conserved mutation being present at positions 1, 44, 110, 155, or any combination thereof, of SEQ ID NO:33.

The composition of entry C59, C72, wherein the mutant Clostridium difficile toxin B comprises SEQ ID NO: 6.

The composition of entry C59, C73, wherein the mutant Clostridium difficile toxin B comprises SEQ ID NO: 86.

Composition of entry C59, C74, wherein the mutant Clostridium difficile toxin B comprises SEQ ID NO:8.

Composition of entry C59, C75, wherein the mutant Clostridium difficile toxin B comprises SEQ ID NO: 85.

C76. A composition of any one of items C59-C71, wherein at least one amino acid of the mutant Clostridium difficile toxin B is chemically cross-linked.

Compositions of entry C76, wherein the amino acids are chemically cross-linked by formaldehyde.

C78. The composition of entry C76, wherein the amino acid is chemically cross-linked by 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide.

C79. Compositions of entries C76 or C78, wherein at least one amino acid is chemically cross-linked by N-hydroxysuccinimide.

C80. A composition of any one of items C59-C79, wherein the composition is recognized by an antitoxin B neutralizing antibody or a binding fragment thereof.

C81. An immunogenic composition comprising a mutant Clostridium difficile toxin B, relative to the corresponding wild-type Clostridium difficile toxin B, said mutant Clostridium difficile toxin B comprising: a glucosyltransferase domain comprising SEQ ID NO:31 having amino acid substitutions at positions 286 and 288; and a cysteine protease domain comprising SEQ ID NO:33 having amino acid substitutions at position 155; wherein at least one amino acid of said mutant Clostridium difficile toxin B is chemically cross-linked.

C82. An immunogenic composition comprising SEQ ID NO:6 or SEQ ID NO:8, wherein at least one amino acid of SEQ ID NO:6 or SEQ ID NO:8 is chemically cross-linked.

C83. A composition of entry C81 or C82, wherein at least one amino acid is cross-linked by formaldehyde.

C84. A composition of entry C81 or C82, wherein the at least one amino acid is cross-linked by 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide.

C85. Compositions of entries C81, C82, or C84, wherein at least one amino acid is cross-linked by N-hydroxysuccinimide.

C86. A composition of entry C81 or C82, wherein the composition is recognized by an antitoxin B neutralizing antibody or a binding fragment thereof.

C87. An immunogenic composition comprising SEQ ID NO: 6.

C88. An immunogenic composition comprising SEQ ID NO:86.

C89. An immunogenic composition comprising SEQ ID NO:8.

C90. An immunogenic composition comprising SEQ ID NO:85.

C91. A composition of any one of items C59-C89, wherein the composition exhibits reduced cytotoxicity relative to the corresponding wild-type Clostridium difficile toxin B.

C92. An immunogenic composition comprising SEQ ID NO:4 and an immunogenic composition comprising SEQ ID NO:6, wherein each of SEQ ID NO:4 and 6 has at least one amino acid that is chemically cross-linked.

C93. An immunogenic composition comprising SEQ ID NO:84 and an immunogenic composition comprising SEQ ID NO:86, wherein each of SEQ ID NO:84 and 86 has at least one amino acid that is chemically cross-linked.

C94. A composition of entry C92 or C93, wherein at least one amino acid is cross-linked by formaldehyde.

C95. A composition of entry C92 or C93, wherein the at least one amino acid is cross-linked by 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide.

C96. A composition of entries C92, C93, or C95, wherein at least one amino acid is cross-linked by N-hydroxysuccinimide.

C97. Recombinant cells or their progeny, including SEQ ID NO:11, SEQ ID NO:12, SEQ ID NO:13, SEQ ID NO:14, SEQ ID NO:44, SEQ ID NO:45, SEQ ID NO:46, or SEQ ID NO:47.

C98. Recombinant cells or their progeny, including nucleic acid sequences encoding SEQ ID NO:4, SEQ ID NO:6, SEQ ID NO:7, or SEQ ID NO:8.

C99. Recombinant cells or their progeny, including a nucleic acid sequence encoding SEQ ID NO:84.

C100. Recombinant cells or their progeny, including the nucleic acid sequence encoding SEQ ID NO:86.

C101. Recombinant cells or their progeny, including the nucleic acid sequence encoding SEQ ID NO:83.

C102. Recombinant cells or their progeny, including the nucleic acid sequence encoding SEQ ID NO:85.

C103. Recombinant cells of entry C97 or C98, wherein the cells are derived from Gram-positive bacterial cells.

C104. Recombinant cells of entries C97, C98, or C99, wherein the cells are derived from Clostridium difficile cells.

C105. Recombinant cells of any one of entries C97-C104, wherein the cells lack endogenous polynucleotides encoding toxins.

C106. Cells of entry C104 or C105, wherein the cells are derived from Clostridium difficile cells selected from the following: Clostridium difficile 1351, Clostridium difficile 3232, Clostridium difficile 7322, Clostridium difficile 5036, Clostridium difficile 4811, and Clostridium difficile VPI11186.

C107. Cells of entry C106, wherein the cells are Clostridium difficile VPI 11186 cells.

C108. Cells of entry C106 or C107, wherein the sporulation gene of the Clostridium difficile cell is inactivated.

C109. Cells of entry C108, wherein the spore-forming gene contains the spo0A gene or the spoIIE gene.

C110. Methods for generating mutant Clostridium difficile toxins, including:

Recombinant cells or their progeny are cultured under suitable conditions to express a polynucleotide encoding a mutant Clostridium difficile toxin, wherein the cells contain the polynucleotide encoding the mutant Clostridium difficile toxin, and wherein, relative to the corresponding wild-type Clostridium difficile toxin, the mutant contains at least one mutated glucosyltransferase domain and at least one mutated cysteine protease domain.

C111. The method of entry C110, wherein the cell lacks endogenous polynucleotides encoding toxins.

C112. The method of entry C110, wherein the recombinant cell or its progeny comprises cells from any one of entries C97-C111.

C113. The method of entry C110 also includes the isolation of mutant clostridium difficile toxins.

C114. The method of entry C113 also includes contacting isolated mutant Clostridium difficile toxin with formaldehyde.

C115. The method of entry C114, wherein the contact occurs for a maximum of 14 days.

C116. The method of entry C115, wherein the contact occurs for a maximum of 48 hours.

C117. The method of entry C114, wherein the contact occurs at approximately 25°C.

The method of entry C113, C118, further includes contacting isolated mutant Clostridium difficile toxin with ethyl-3-(3-dimethylaminopropyl)carbodiimide.

C119. The method of entry C118, wherein the contact occurs for a maximum of 24 hours.

C120. The method of entry C120, wherein the contact occurs for a maximum of 4 hours.

C121. The method of entry C118, wherein the contact occurs at approximately 25°C.

The method in entry C118, C122, also includes contacting isolated mutant Clostridium difficile toxin with N-hydroxysuccinimide.

C123. An immunogenic composition produced by any of the methods in entries C110-C122.

C124. A method for generating neutralizing antibodies against Clostridium difficile toxin A, comprising: administering an immunogenic composition to a mammal, said immunogenic composition comprising SEQ ID NO:4, wherein the methionine residue at position 1 is optionally absent, wherein at least one amino acid of SEQ ID NO:4 is cross-linked by formaldehyde, 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide, N-hydroxysuccinimide, or a combination of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide and N-hydroxysuccinimide, and recovering the antibody from said mammal.

C125. A method for generating neutralizing antibodies against Clostridium difficile toxin A, comprising: administering an immunogenic composition to a mammal, the immunogenic composition comprising SEQ ID NO:84, wherein at least one amino acid of SEQ ID NO:84 is cross-linked by formaldehyde, 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide, N-hydroxysuccinimide, or a combination of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide and N-hydroxysuccinimide, and recovering the antibody from the mammal.

C126. A method for generating neutralizing antibodies against Clostridium difficile toxin B, comprising: administering an immunogenic composition to a mammal, said immunogenic composition comprising SEQ ID NO:6, wherein the methionine residue at position 1 is optionally absent, wherein at least one amino acid of SEQ ID NO:6 is cross-linked by formaldehyde, 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide, N-hydroxysuccinimide, or a combination of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide and N-hydroxysuccinimide, and recovering the antibody from said mammal.

C127. A method for generating neutralizing antibodies against Clostridium difficile toxin A, comprising: administering an immunogenic composition to a mammal, the immunogenic composition comprising SEQ ID NO:86, wherein at least one amino acid of SEQ ID NO:86 is cross-linked by formaldehyde, 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide, N-hydroxysuccinimide, or a combination of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide and N-hydroxysuccinimide, and recovering the antibody from the mammal.

C128. An antibody or antibody-binding fragment thereof specific to an immunogenic composition, said immunogenic composition comprising, optionally, the methionine residue at position 1 of SEQ ID NO:4, or optionally, the methionine residue at position 1 of SEQ ID NO:7.

C129. An antibody or antibody-binding fragment thereof of entry C128, wherein at least one amino acid of SEQ ID NO:4, which is optionally absent, or at least one amino acid of SEQ ID NO:7, which is optionally absent, at position 1, is cross-linked by formaldehyde, 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide, N-hydroxysuccinimide, or a combination of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide and N-hydroxysuccinimide.

C130. An antibody or an antibody-binding fragment thereof, comprising: the amino acid sequence of the heavy chain complementarity-determining region (CDR) shown in SEQ ID NO:41 (CDR H1), SEQ ID NO:42 (CDR H2) and SEQ ID NO:43 (CDR H3), and the amino acid sequence of the light chain CDR shown in SEQ ID NO:38 (CDR L1), SEQ ID NO:39 (CDR L2) and SEQ ID NO:40 (CDR L3).

C131. An antibody or antibody-binding fragment thereof from entries C128, C129, or C130, wherein the antibody or antibody-binding fragment thereof comprises: a heavy chain comprising the amino acid sequence shown in SEQ ID NO:37; and a light chain comprising the amino acid sequence shown in SEQ ID NO:36.

C132. A composition comprising two or more antibodies or antibody-binding fragments thereof, wherein the antibodies or antibody-binding fragments thereof are selected from any antibody or antibody-binding fragment thereof from any of entries C128-C131.

C133. An antibody or antibody-binding fragment thereof specific to an immunogenic composition, said immunogenic composition comprising: SEQ ID NO:6 wherein the methionine residue at position 1 is optionally absent, or SEQ ID NO:8 wherein the methionine residue at position 1 is optionally absent.

C134. An antibody or antibody-binding fragment thereof of entry C133, wherein at least one amino acid of SEQ ID NO: 6, which is optionally absent, or at least one amino acid of SEQ ID NO: 8, which is optionally absent, is cross-linked by formaldehyde, 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide, N-hydroxysuccinimide, or a combination of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide and N-hydroxysuccinimide.

C135. An antibody or an antibody-binding fragment thereof, comprising: the amino acid sequence of the heavy chain complementarity-determining region (CDR) shown in SEQ ID NO:51 (CDR H1), SEQ ID NO:52 (CDR H2) and SEQ ID NO:53 (CDR H3), and the amino acid sequence of the light chain CDR shown in SEQ ID NO:57 (CDR L1), SEQ ID NO:58 (CDR L2) and SEQ ID NO:59 (CDR L3).

C136. An antibody or an antibody-binding fragment thereof, comprising: the amino acid sequence of the heavy chain complementarity-determining region (CDR) shown in SEQ ID NO:61 (CDR H1), SEQ ID NO:62 (CDR H2) and SEQ ID NO:63 (CDR H3), and the amino acid sequence of the light chain CDR shown in SEQ ID NO:68 (CDR L1), SEQ ID NO:69 (CDR L2) and SEQ ID NO:70 (CDR L3).

C137. An antibody or an antibody-binding fragment thereof, comprising: the amino acid sequence of the heavy chain complementarity-determining region (CDR) shown in SEQ ID NO:73 (CDR H1), SEQ ID NO:74 (CDR H2) and SEQ ID NO:75 (CDR H3), and the amino acid sequence of the light chain CDR shown in SEQ ID NO:79 (CDR L1), SEQ ID NO:80 (CDR L2) and SEQ ID NO:81 (CDR L3).

C138. A composition comprising two or more antibodies or antibody-binding fragments thereof, wherein the antibodies or antibody-binding fragments thereof are selected from any one of entries C133-C137.

C139. A method for treating Clostridium difficile infection in mammals, comprising: administering to said mammal an immunogenic composition comprising SEQ ID NO:4, wherein the methionine residue at position 1 is optionally absent, and an immunogenic composition comprising SEQ ID NO:6, wherein each of SEQ ID NO:4 and 6 has at least one amino acid cross-linked by formaldehyde.

C140. A method for treating Clostridium difficile infection in mammals, comprising: administering to said mammal an immunogenic composition comprising SEQ ID NO:4, wherein the methionine residue at position 1 is optionally absent, and an immunogenic composition comprising SEQ ID NO:6, wherein each of SEQ ID NO:4 and SEQ ID NO:6 has at least one amino acid crosslinked by a combination of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide, N-hydroxysuccinimide, or 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide and N-hydroxysuccinimide.

C141. A method for treating Clostridium difficile infection in mammals, comprising administering to said mammals: an immunogenic composition comprising SEQ ID NO:84 and an immunogenic composition comprising SEQ ID NO:86, wherein each of SEQ ID NO:84 and SEQ ID NO:86 has at least one amino acid crosslinked by 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide and N-hydroxysuccinimide.

C142. A method for inducing an immune response against Clostridium difficile in mammals, comprising administering to said mammal: an immunogenic composition comprising SEQ ID NO:4 wherein the methionine residue at position 1 is optionally absent; and an immunogenic composition comprising SEQ ID NO:6 wherein the methionine residue at position 1 is optionally absent; wherein each of SEQ ID NO:4 and SEQ ID NO:6 has at least one amino acid crosslinked by formaldehyde.

C143. A method for inducing an immune response against Clostridium difficile in mammals, comprising administering to said mammal: an immunogenic composition comprising SEQ ID NO:4 wherein the methionine residue at position 1 is optionally absent; and an immunogenic composition comprising SEQ ID NO:6 wherein the methionine residue at position 1 is optionally absent; wherein each of SEQ ID NO:4 and SEQ ID NO:6 has at least one amino acid crosslinked by a combination of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide, N-hydroxysuccinimide, or 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide and N-hydroxysuccinimide.

C144. A method for inducing an immune response against Clostridium difficile in mammals, comprising administering to said mammals: an immunogenic composition comprising SEQ ID NO:84 and an immunogenic composition comprising SEQ ID NO:86, wherein each of SEQ ID NO:84 and SEQ ID NO:86 has at least one amino acid crosslinked by 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide and N-hydroxysuccinimide.

C145. The method of any one of entries C139-C144, wherein the mammal is a mammal for which it is needed.

C146. The method of any one of entries C139-C144, wherein the mammal suffers from recurrent Clostridium difficile infection.

C147. The method of any one of entries C139-C144, wherein the composition is applied parenterally.

C148. The method of any one of entries C139-C144, wherein the composition further comprises an adjuvant.

C149. The method of entry C148, wherein the adjuvant comprises aluminum.

The method of C150, entry C148, wherein the adjuvant comprises aluminum hydroxide gel and CpG oligonucleotide.

C151. The method of entry C148, wherein the adjuvant comprises

Claims (37)

1. A polypeptide comprising the amino acid sequence shown in SEQ ID NO: 6, with or without a methionine residue at position 1. 2. A nucleic acid encoding a polypeptide as shown in SEQ ID NO: 6, wherein a methionine residue at position 1 is present or absent. 3. An immunogenic composition for inducing neutralizing antibodies against Clostridium difficile toxin B, comprising an effective amount of the polypeptide shown in SEQ ID NO: 6, wherein the methionine residue at position 1 is present or absent. 4. The composition of claim 3, wherein the polypeptide comprises SEQ ID NO: 6. 5. The composition of claim 3, wherein the polypeptide comprises SEQ ID NO: 6, in which the methionine residue at position 1 is absent. 6. The composition of claim 3, further comprising carbohydrates. 7. The composition of claim 6, wherein the carbohydrate is selected from sorbitol, mannitol, starch, dextran, sucrose, trehalose, lactose, and glucose. 8. The composition of claim 6, wherein the carbohydrate comprises sucrose. 9. The composition of claim 6, wherein the carbohydrate comprises trehalose. 10. The composition of any one of claims 3-9, further comprising a buffer. 11. The composition of claim 10, wherein the buffer is selected from phosphate buffers; acetate buffers; succinate buffers; glycine buffers; histidine buffers; carbonate buffers; Tris (tris(hydroxymethyl)aminomethane) buffers; and bicarbonate buffers. 12. The composition of claim 11, wherein the phosphate is selected from potassium phosphate and sodium phosphate; the acetate is sodium acetate; the succinate is sodium succinate; and the bicarbonate is ammonium bicarbonate. 13. The composition of claim 10, wherein the buffer comprises a Tris buffer. 14. The composition of claim 10, wherein the buffer comprises histidine. 15. The composition of claim 10, wherein the buffer comprises a phosphate. 16. The composition of claim 3, further comprising a surfactant. 17. The composition of claim 16, wherein the surfactant is selected from polyoxyethylene sorbitan surfactant; polyoxyethylene fatty ethers derived from dodecyl alcohol, hexadecyl alcohol, octadecyl alcohol and oleyl alcohol; triethylene glycol monododecyl ether; tert-octylphenoxypolyethoxyethanol; and sorbitan ester. 18. The composition of claim 16, wherein the surfactant comprises polyoxyethylene sorbitan ester. 19. The composition of claim 16, wherein the surfactant comprises polyoxyethylene sorbitan monooleate. 20. The composition of claim 16, wherein the surfactant comprises polysorbate 80. 21. The composition of claim 3, wherein the composition does not contain formaldehyde. 22. The composition of claim 3, wherein the composition comprises an adjuvant. 23. The composition of claim 22, wherein the adjuvant is selected from aluminum hydroxide; aluminum salts; saponins and CpG oligonucleotides. 24. The composition of claim 22, wherein the adjuvant is selected from 3De-O-acylated monophosphate lipid A; aluminum phosphate; aluminum sulfate; QS-21; polyglutamic acid; polylysine; and CpG oligodeoxynucleotide. 25. The composition of claim 22, wherein the adjuvant is selected from immunostimulatory particles and CpG oligonucleotides comprising SEQ ID NO: 48. 26. The composition of claim 22, wherein the adjuvant comprises aluminum hydroxide. 27. The composition of claim 22, wherein the adjuvant comprises aluminum sulfate. 28. The composition of claim 22, wherein the adjuvant comprises a CpG oligonucleotide. 29. The composition of claim 22, wherein the adjuvant comprises QS-21. 30. The composition of claim 22, wherein the adjuvant comprises immunostimulatory particles. 31. The composition of claim 3, wherein the composition is lyophilized. 32. The composition of claim 3, wherein the composition is liquid. 33. An antibody or antigen-binding fragment thereof induced by the composition of claim 3 and having neutralizing activity against Clostridium difficile toxin B, comprising a subset selected from... SEQ ID NO: 49 and 55 SEQ ID NO: 60 and 66, or The light and heavy chains of SEQ ID NO: 71 and 77 are specific for immunogenic compositions comprising a polypeptide consisting of SEQ ID NO: 6, which is present or absent of a methionine residue at position 1. 34. An antibody or antigen-binding fragment thereof induced by the composition of claim 3 and having neutralizing activity against Clostridium difficile toxin B, comprising the amino acid sequences of the heavy chain complementarity-determining regions (CDRs) of CDR H1 shown in SEQ ID NO: 51, CDR H2 shown in SEQ ID NO: 52, and CDR H3 shown in SEQ ID NO: 53, and the amino acid sequences of the light chain CDRs of CDR L1 shown in SEQ ID NO: 57, CDR L2 shown in SEQ ID NO: 58, and CDR L3 shown in SEQ ID NO: 59. 35. An antibody or antigen-binding fragment thereof induced by the composition of claim 3 and having neutralizing activity against Clostridium difficile toxin B, comprising the amino acid sequences of the heavy chain complementarity-determining regions (CDRs) of CDR H1 (SEQ ID NO: 62), CDR H2 (SEQ ID NO: 63), and CDR H3 (SEQ ID NO: 64), and the amino acid sequences of the light chain CDRs of CDR L1 (SEQ ID NO: 68), CDR L2 (SEQ ID NO: 69), and CDR L3 (SEQ ID NO: 70). 36. An antibody or antigen-binding fragment thereof induced by the composition of claim 3 and having neutralizing activity against Clostridium difficile toxin B, comprising the amino acid sequences of the heavy chain complementarity-determining regions (CDRs) of CDR H1 (SEQ ID NO: 73), CDR H2 (SEQ ID NO: 74), and CDR H3 (SEQ ID NO: 75), and the amino acid sequences of the light chain CDRs of CDR L1 (SEQ ID NO: 79), CDR L2 (SEQ ID NO: 80), and CDR L3 (SEQ ID NO: 81). 37. A composition having neutralizing activity against Clostridium difficile toxin B, comprising an effective amount of two or more antibodies selected from any one of claims 34-36 or their antigen-binding fragments.