US20070009512A1

US20070009512A1 - Recombinant bivalent monospecific immunoglobulin having at least two variable fragments of heavy chains of an immunoglobulin devoid of light chains

Info

Publication number: US20070009512A1
Application number: US11/520,128
Authority: US
Inventors: Raymond Hamers; Serge Muyldermans
Original assignee: Vrije Universiteit Brussel VUB
Current assignee: Vrije Universiteit Brussel VUB
Priority date: 1995-04-25
Filing date: 2006-09-13
Publication date: 2007-01-11
Also published as: WO1996034103A1; EP0739981A1; DE69637483T2; EP1816198A1; DE69637483D1; EP0822985A1; US7655759B2; EP0822985B1; JP4394162B2; AU5647896A; JPH11503918A; ATE391179T1; US20030088074A1

Abstract

The present invention relates to fragments, especially variable fragments of immunoglobulins which are by nature devoid of light chains, these fragments being nevertheless capable of exhibiting a recognition and binding activity toward specific antigens. The present invention further relates to the use of such immunoglobulin fragments formed of at least one heavy chain variable fragment or derived therefrom, for therapeutic or veterinary purposes and especially for passive immunotherapy or serotherapy.

Description

RELATED APPLICATIONS

This application is a continuation of U.S. patent application Ser. No. 10/154,971, filed May 28, 2002, which is a continuation of U.S. patent application Ser. No. 08/945,244, filed Jan. 16, 1998, which is a national stage application under 35 U.S.C. §371 of International Application PCT/EP1996/01725 designating the United States of America, filed Apr. 25, 1996, the entire disclosures of which are incorporated herein by reference.

FIELD OF THE INVENTION

The present invention relates to fragments, especially variable fragments of immunoglobulins which are by nature devoid of light chains, these fragments being nevertheless capable of exhibiting a recognition and binding activity toward specific antigens. These fragments of immunoglobulins can for example be obtained by the expression in host cells for example in prokaryotic cells or eukaryotic cells of nucleotide sequences obtained from animals naturally expressing so-called “two-chain immunoglobulins”, for instance from animals of the camelid family.
The present invention further relates to the use of such immunoglobulin fragments formed of at least one heavy chain variable fragment or derived therefrom, for therapeutic or veterinary purposes and especially for passive immunotherapy or serotherapy.

BACKGROUND OF THE INVENTION

Functional immunoglobulins devoid of light polypeptide chains termed <<two-chain immunoglobulin>> or <<heavy-chain immunoglobulin>> have been obtained from animals of the family of camelids and have been described in an international patent application published under number WO 94/04678, together with two publications, especially Hamers-Casterman et al., 1993 and Muyldermans et al., 1994).
The isolation and characterization of these immunoglobulins, together with their cloning and sequencing have been described in the above referenced documents which are incorporated by reference in the present application.
According to WO 94/04678 it has been established that different molecules can be isolated from animals which naturally produce them, which molecules have functional properties of the well known four-chain immunoglobulins these functions being in some cases related to structural elements which are distinct from those involved in the function of four-chain immunoglobulins due for instance to the absence of light chains.
These immunoglobulins having only two chains, neither correspond to fragments obtained for instance by the degradation in particular the enzymatic degradation of a natural four-chain model immunoglobulin, nor correspond to the expression in host cells, of DNA coding for the constant or the variable regions of a natural four-chain model immunoglobulin or a part of these regions, nor correspond to antibodies produced in lymphopathies for example in mice, rats or human.
The immunoglobulins devoid of light chains are such that the variable domains of their heavy chains have properties differing from those of the four-chain immunoglobulin variable heavy chain (V_H). For clarity reasons, this variable domain according to the invention will be called V_HHin this text to distinguish it from the classical V_Hof four-chain immunoglobulins. The variable domain of a heavy-chain immunoglobulin according to the invention has no normal interaction sites with the V_Lor with the C _H1 domain which do not exist in the heavy-chain immunoglobulins. It is hence a novel fragment in many of its properties such as solubility and conformation of main chains. Indeed the V_HHof the invention can adopt a three-dimensional organization which distinguishes from the three-dimensional organization of known four-chain immunoglobulins according to the description which is given by Chothia C. and Lesk A. M. (1987-J. Mol. Biol. 197, 901-917).
According to the results presented in patent application WO 94/04678, the antigen binding sites of the isolated immunoglobulins, naturally devoid of light chains are located on the variable region of their heavy chains. In most cases, each heavy chain variable region of these two-chain immunoglobulins can comprise an antigen binding site.
A further characteristic of these two-chain immunoglobulins is that their heavy polypeptide chains contain a variable region (V_HH) and a constant region (C_H) according to the definition of Roitt et al but are devoid of the first domain of the constant region is called C _H1.
These immunoglobulins of the type described hereabove can comprise type G immunoglobulins and especially immunoglobulins which are termed immunoglobulins of class 2 (IgG2) or immunoglobulins of class 3 (IgG3), according to the classification established in patent application WO 94/04678 or in the publication of Muyldermans et al. (Protein Engineering, Vol. 7, No. 9, pp 1129-1135-1994).
The absence of the light chain and of the first constant domain lead to a modification of the nomenclature of the immunoglobulin fragments obtained by enzymatic digestion, according to Roitt et al.
The terms Fc and pFc on the one hand, Fc′ and pFc′ on the other hand corresponding respectively to the papain and pepsin digestion fragments are maintained.
The terms Fab, F(ab)₂, F(ab′)₂, Fabc, Fd and fv are no longer applicable in their original sense as these fragments have either a light chain, the variable part of the light chain or the CH₁domain.
The fragments obtained by papain digestion or by V8 digestion, composed of the V_HHdomain of the hinge region will be called FV_HHh or F(V_HHh)₂depending upon whether or not they remain linked by the disulphide bonds.
The immunoglobulins referring to the hereabove given definitions can be originating from animals especially from animals of the camelid family. These heavy-chain immunoglobulins which are present in camelids are not associated with a pathological situation which would induce the production of abnormal antibodies with respect to the four-chain immunoglobulins. On the basis of a comparative study of old world camelids (Camelus bactrianus and Camelus dromaderius) and new world camelids (for example Lama Paccos, Lama Glama, and Lama Vicugna) the inventors have shown that the immunoglobulins devoid of light polypeptide chains are found in all species. Nevertheless differences may be apparent in molecular weight of these immunoglobulins depending on the animals. Especially the molecular weight of a heavy chain contained in these immunoglobulins can be from approximately 43 kd to approximately 47 kd, in particular 45 kd.
Advantageously the heavy-chain immunoglobulins. of the invention are secreted in blood of camelids.
The variable fragments of heavy chains of immunoglobulins devoid of light chains can be prepared starting from immunoglobulins obtainable by purification from serum of camelids according to the process for the purification as described in detail in the examples of WO 94/04678. The variable fragments can also be obtained from heavy-chain immunoglobulins by digestion with papain or V8 enzymes.
These fragments can also be generated in host cells by genetic engineering or by chemical synthesis. Appropriate host cells are for instance bacteria (e.g. E. coli) eukaryotic cells including yeasts or animal cells including mammalian cells, or plant cells.
The observation by the inventors that Camelidae produce a substantial proportion of their functional immunoglobulins as a homodimer of heavy chains lacking the C _H1 domain and devoid of light chains (Hamers-Casterman et al., 1993), led to the proposal of having recourse to an immunized camel to generate and select single variable antibody fragments (V_HH) and furthermore give access to the corresponding nucleotide sequences.
Cloned camel single V_HHfragments were displayed on bacteriophages for selection and in bacteria for the large scale production of the soluble proteins, and were shown to possess a superior solubility behaviour and affinity properties compared to the mouse or human V_Hequivalents (Muyldermans et al., 1994). Following this strategy, one would obtain small ligand binding molecules (MW around 16,000 D) which are not hindered by the presence of an oligopeptide linker (Borrebaeck et al. 1992) or not inactivated by the disassembly of the VH-VL complex (Glockshuber et al., 1990). The camel V_HHfragments have the additional advantage that they are characteristic of the heavy chain antibodies which are matured in vivo in the absence of light chains.

SUMMARY OF THE INVENTION

The inventors have obtained evidence that variable fragments of high chains of immunoglobulins devoid of light chains can display an effective therapeutic activity when they are generated against a determined antigen.
To develop this technology of preparing and identifying useful camel V_HHfragments, it is critical (i) that camels can be immunized with a variety of antigens, (ii) that the camel V_HHgenes can be cloned and expressed on filamenteous phages and in E coli for easy selection with the immobilized antigen by panning, (iii) that the expressed camel V_HH's are properly folded, and (iv) that they have good solubility properties and possess high affinities and specificities towards their antigen.
Camel V_HHgenes derived from the heavy chain immunoglobulins lacking the light chains were previously cloned and analysed (Muyldermans et al., 1994). A comparison of the amino acid sequences of these camel V_HHclones clearly showed that the key features for preserving the characteristic immunoglobulin fold are all present. The specific amino acid replacements observed in the camel V_HHclones could correlate with the absence of the VL (variable light chains) and the functionality of the camel single V_HHdomain (Muyldermans et al., 1994).
The invention thus relates to a variable fragment (V_HH) of a heavy chain of an immunoglobulin devoid of light chains, which is encoded by a nucleotide sequence obtainable by the following process:
treating blood lymphocytes or other appropriate cells of an animal of the Camelid family previously immunized with a determined antigen, in order to give access to their mRNA,
synthesizing a first strand of cDNA starting from the obtained mRNA,
contacting the obtained cDNA with at least two different primer oligonucleotides in conditions allowing their hybridization to at least two complementary nucleotide sequences contained in the cDNA, said primers comprising a BACK primer (back p1) having nucleotide sequence (SEQ ID NO: 1): 5′-GATGTGCAGCTGCAGGCGTCTGG(A/G)GGAGG-3′ and a FOR primer (for p1) replying to the following nucleotide sequence (SEQ ID NOS: 2 and 3, respectively:

5′-CGCCATCAAGGTACCGTTGA-3′

or

5′-CGCCATCAAGGTACCAGTTGA-3′
amplifying the DNA fragment located between the nucleotide sequence hybridized with said primers and,
recovering amplified DNA corresponding to bands of different size orders including:

- a band of around 750 basepairs which is the amplified product of the variable heavy chain (V_H), C _H1, hinge and part of CH2 region of a four-chain immunoglobulin,
- a band of around 620 basepairs which is the amplified product of the variable heavy-chain (V_HH), long hinge, and part of the CH2 of the camel two-chain immunoglobulin IgG2,
- a band of around 550 basepairs which is the amplified product of the variable heavy-chain (V_HH), short hinge, and part of the CH2 of the camel two-chain immunoglobulin IgG3,
- purifying the two shortest bands of around 620 and 550 basepairs from agarose gel, for example by Gene Clean,
- recovering the amplified DNA fragments containing nucleotide sequences encoding the V_HHfragments,
- digesting the amplified products with restriction enzymes having target sites within the amplified fragments and/or in the nucleotide primers, for example with PstI and BstEII,
- recovering the digested amplified DNA fragments,
- ligating the amplified DNA fragments to a plasmid vector, for example in a pHEN4 vector, in conditions allowing the expression of the amplified fragments when the obtained recombinant vector is used to transform a host cell,
- transforming a determined bacterial host cell for example an E. coli cell with the obtained recombinant plasmid vector, and growing the cells on selective medium, to form a library,
- infecting the obtained library of recombinant host cells after culture in an appropriate selective medium, with bacteriophages, for instance M13K07 bacteriophages to obtain recombinant phagemid virions,
- incubating the recombinant host cells in conditions allowing secretion of recombinant phagemid virions particles containing the recombinant plasmid, for instance the pHEN4 plasmid packaged within the M13 virion.
- isolating and concentrating the recombinant phagemid virions,
- submitting the phagemid virions to several rounds of panning with the antigen of interest previously immobilized, in conditions allowing the adsorption of the phagemid virions on the immobilized antigen,
- eluting the adsorbed phagemid virions, and growing them on appropriate cells,
- amplifying the phagemid virions by infecting the cells with helper bacteriophage,
- recovering the virions and testing them for their binding activity against the antigen of interest, for example by ELISA,
- recovering the phagemid virions having the appropriate binding activity,
- isolating the nucleotide sequence contained in the plasmid vector and capable of being expressed on the phagemid virions as a V_HHamino acid sequence having the appropriate binding activity.

In a preferred embodiment of the invention, the variable V_HHfragments are obtainable by adding to the hereabove described amplification step of the cDNA with BACK and FOR primers (p1), a further amplification step with a BACK primer corresponding to the oligonucleotide sequence which has been described hereabove (back p1) and the FOR primer (for p2) having the following nucleotide sequence (SEQ ID NO: 4): 5′-CG ACT AGT GCG GCC GCG TGA GGA GAC GGT GAC CTG-3′. Not and BstEII sites which can be used for cloning in the pHEN4 vector have been underlined. This FOR primer allows hybridization to the codon position of framework 4 (FR4) region of the V_HHnucleotide sequences (amino acid position 113-103).
According to another variant of the process described, this additional amplification step can replace the amplification step which has been described with BACK primer and a FOR primer having respectively the following nucleotide sequences (SEQ ID NOS 1-3, respectively): 5′-GATGTGCAGCTGCAGGCGTCTGG(A/G)GGAGG-3′
5′-CGCCATCAAGGTACCGTTGA-3′
or
5′-CGCCATCMGGTACCAGTTGA-3′
The restriction sites have been underlined.
In another embodiment of the invention the amplification step of the synthetized cDNA is performed with oligonucleotide primers including hereabove described BACK primer and FOR primer having the following sequences (SEQ ID NOS 5 & 6, respectively): FOR primer 3: 5′-TGT CTT GGG TTC TGA GGA GAC GGT-3′
According to this latter embodiment, the V_HHfragments of the invention are immediately and specifically amplified by a single amplification (for instance PCR reaction) step when the mixture of FOR primers is used.
These latter primers hybridize with the hinge/framework 4 and short hinge/framework 4 respectively. Each of these FOR primers allows the amplification of one IgG class according to the classification given in patent application WO 94/04678.
The variable V_HHfragments corresponding to this definition can also be obtained from other sources of animal cells, providing that these animals are capable of naturally producing immunoglobulins devoid of light chains according to those described in the previous patent application WO 94/04678.
These variable fragments (V_HH) can also be obtained by chemical synthesis or by genetic engineering starting from DNA sequences which can be obtained by the above described process.
The variable fragment of a heavy chain of an immunoglobulin devoid of light chains according to the preceding definitions is specifically directed against an antigen against which the animal has been previously immunized, either by natural contact with this antigen or by administration of this antigen in order to generate an immune response directed against it.
The process which is proposed hereabove to prepare a nucleotide sequence coding for the variable fragments of the invention contains steps of phage display library construction which allow the selection of nucleotide sequences coding for variable fragments of heavy chains having the desired specificity.
According to one preferred embodiment of the invention, the variable fragments of a heavy chain of a immunoglobulin is obtainable from an animal having been previously immunized with a toxin, especially a toxin of a bacteria or a part of this toxin sufficient to enable the production of immunoglobulins directed against this toxin and especially immunoglobulins devoid of light chains.
According to another embodiment of the invention, the variable fragments of a heavy chain of a immunoglobulin is obtainable from an animal having been previously immunized with substances contained in venom of animals.
The antigen used for immunization of the animals is usually under a non toxic form.
The variable fragments according to the invention can be derived from immunoglobulins belonging to different classes especially belonging to IgG2 or IgG3 immunoglobulin classes, according to the classification given in patent application WO/04678.
In a preferred embodiment of the invention, the variable fragment of a heavy-chain of an immunoglobulin devoid of light chains is directed against the tetanus toxin of Clostridium tetani or against a fragment thereof.
The variable fragments of heavy chains of immunoglobulins devoid of light chains can be also generated against toxins or part thereof from pathogenic organisms such as bacteria and especially can be chosen among the toxins or toxoids of the following bacteria: Clostridium, especially Clostridium Botulinum or Clostridium Perfringens, Staphylococcus, Pseudomonas, Pasteurella, Yersinia, Bacillus Anthracis, Neisseria, Vibrio, especially Vibrio cholera, enterotoxic E. coli. Salmonella, Shigella, Listeria.
Other antigens appropriate for the preparation of the V_HHfragments of the invention can be obtained from the following organism: anemonies, coral, jellyfish, spiders, bees, wasps, scorpions, snakes, including those belonging to the families of Viperidae, Crotalidae, Lapidea.
According to another embodiment of the invention, the variable fragment V_HHof a heavy chain of an immunoglobulin devoid of light chains is characterized in that it comprises the following amino acid sequences (SEQ ID NOS: 25 & 26, respectively):
(Glu/Asp) Val Gln Leu Gln Ala Ser Gly Gly Gly Ser Val Gln Ala Gly (Gly/Gln) Ser Leu Arg Leu Ser Cys Ala (Ala/Thr) Ser Gly (CDR1) Trp (Phe/Tyr) Arg Gln Ala Pro Gly Lys Glu (Arg/Cys) Glu (Gly/Leu) Val (Ser/Ala) (CDR2) Arg (Phe/Leu) Thr Ile Ser (Arg/Leu/Gln) Asp Asn Ala Lys Asn Thr (Val/Leu) Tyr Leu (Gln/Leu) Met Asn Ser Leu (Lys/Glu) Pro Glu Asp Thr Ala (Val/Met/Ile) Tyr Tyr Cys Ala Ala (CDR3) Trp Gly Gln Gly Thr Gln Val Thr Val Ser Ser or
(Glu/Asp) Val Gln Leu Gln Ala Ser Gly Gly Gly Ser Val Gln Ala Gly (Gly/Gln) Ser Leu Arg Leu Ser Cys Ala (Ala/Tlu) Ser Gly (Ala,Thr,Ser,Ser/Tyr,Thr,Ile,Gly) (CDR1) Trp (Phe/Tyr) Arg Gln Ala Pro Gly Lys Glu (Agr/Cys) Glu (Gly/Leu) Val (Ser/Ala) (CDR2) Arg (Phe/Leu) Thr Ile Ser (Arg/Leu/Gln) Asp Asn Ala Lys Asn Thr (Val/Leu) Tyr Leu (Gln/Leu) Met Asn Ser Leu (Lys/Glu) Pro Glu Asp Thr Ala (Val/Met/Ile) Tyr Tyr Cys Ala Ala (CDR3) Trp Gly Gln Gly Thr Gln Val Thr Val Ser Ser,
wherein CDR1, CDR2 and CDR3 represent variable amino acid sequences providing for the recognition of a determined epitope of the antigen used for the immunization of Camelids, CDR1, CDR2 and CDR3 sequences comprising from 5 to 25 amino acid residues preferably CDR1 contains from 7 to 12 amino acid residues, CDR2 contains from 16 to 21 amino acid residues and CDR3 contains from 7 to 25 amino acid residues.
The camel V_HHspecific amino acid residues Ser 11, Phe 37, Glu 44, Arg 45, Glu 46, Gly 47 are underlined.
One preferred variable fragment according to the invention is encoded by a nucleotide sequence present in recombinant plasmid pHEN4-αTT2(WK6) deposited at the BCCM/LMBP (Belgium) under accession number LMBP3247.
The pHEN4-αTT2 (described in FIG. 2) is a plasmid carrying a PeIB leader signal, a camel V_HHgene of which the protein binds tetanus toxoid, a decapeptide tag (frim immunoZAP H of Stratacyte) and gene IIIp of M13 in the pUC 119 polylinker between the HindIII and EcoR1 sites. This plasmid was transfused in E. coli WK6 cells.
A specific variable fragment according to the invention is for instance characterized in that it comprises the following αTT1 amino acid sequence (SEQ ID NO: 7):
Glu Val Gln Leu Gln Ala Ser Gly Gly Gly Ser Val Gln Ala Gly Gly Ser Leu Arg Leu Ser Cys Ala Ala Ser Gly Gly Gln Thr Phe Asp Ser Tyr Ala Met Ala Trp Phe Arg Gln Ala Pro Gly Lys Glu Cys Glu Leu Val Ser Ser Ile Ile Gly Asp Asp Asn Arg Asn Tyr Ala Asp Ser Val Lys Gly Arg Phe Thr Ile Ser Arg Asp Asn Ala Lys Asn Thr Val Tyr Leu Gln Met Asp Arg Leu Asn Pro Glu Asp Thr Ala Val Tyr Tyr Cys Ala Gln Leu Gly Ser Ala Arg Ser Ala Met Tyr Cys Ala Gly Gln Gly Thr Gln Val Thr Val Ser Ser
According to another preferred embodiment of the present invention, the variable fragment comprises the following αTT2 amino acid sequence (SEQ ID NO: 8):
Glu Val Gln Leu Gln Ala Ser Gly Gly Gly Ser Val Gln Ala Gly Gly Ser Leu Arg Leu Ser Cys Thr Ala Ala Asn Tyr Ala Phe Asp Ser Lys Thr Val Gly Trp Phe Arg Gln Val Pro Gly Lys Glu Arg Glu Gly Val Ala Gly Ile Ser Ser Gly Gly Ser Thr Thr Ala Tyr Ser Asp Ser Val Lys Gly Arg Tyr Thr Val Ser Leu Glu Asn Ala Lys Asn Thr Val Tyr Leu Leu Ile Asp Asn Leu Gln Pro Glu Asp Thr Ala Ile Tyr Tyr Cys Ala Gly Val Ser Gly Trp Arg Gly Arg Gln Trp Leu Leu Leu Ala Glu Thr Tyr Arg Phe Trp Gly Gln Gly Thr Gln Val Thr Val Ser Ser
In a preferred embodiment of the invention, the variable V_HHfragment of the invention is altered in order to diminish its immunogenic properties. Such a modification can lead to an alternated immunological reaction against the V_HHfragments of the invention when they are administered to a host either human or animal, for passive immunoprotection for example.
The invention further relates to a pharmaceutical composition comprising an immunoglobulin heavy chain variable fragment according to those which have been defined hereabove, in admixture with a physiologically acceptable vehicle.
Such pharmaceutical composition can be used for the treatment by passive immunisation, of infections or acute intoxications by toxins such as those of Clostridium, especially Clostridium Botulinum or Clostridium Perfringens, Staphylococcus, Pseudomonas, Pasteurella, Yersinia, Bacillus Anthracis, Neisseria, Vibrio, especially Vibrio cholera, enterotoxic E. coli, Salmonella, Shigella, Listeria or anemonies, coral, jellyfish, spiders, beas, wasps, scorpions, snakes, including those belonging to the families of Viperidae, Crotalidae, Lapidea.
The present invention further relates to nucleotide sequences coding for a variable fragment (V_HH) of a heavy chain of an immunoglobulin devoid of light chains, obtainable by the process which has been described hereabove.
Specific nucleotide sequences are those corresponding to αTT1 and αTT2 as described on FIGS. 4A and 4B.
According to an embodiment of the invention, a preferred nucleotide sequence is the sequence contained on plasmid pHEN4-αTT2 deposited at the BCCM/LMBP collection in Belgium on Jan. 31, 1995 under no. LMBP3247.
The invention further provides means for the preparation of bivalent or even multivalent monospecific DNA constructs of variable fragments of an immunoglobulin devoid of light chains and their expression products. It thus gives access to the preparation of monovalent bispecific or multispecific variable constructs obtained from sequences encoding V_HHfragments combined with a linker sequence. Bivalent monospecific constructs contain 2 nucleotide sequences coding for V_HHfragments directed against the same antigen or epitope. Monovalent bispecific constructs contain on one molecule one nucleotide sequence coding for a V_HHfragment directed against one antigen or epitope and another nucleotide sequence coding for a fragment directed against another antigen or epitope.
The corresponding expression products (protein constructs) can be obtained by genetic engineering especially by expression in host cells, like bacteria (e.g. E. coli) or eukaryotic cells, of the above DNA constructs.
Accordingly a variable fragment of the V_HHtype having a determined antigen specificity, can be linked to at least one further variable fragment V_HHhaving a determined similar or different specificity in terms of antigen and/or epitope specificity.
The obtained constructs (in terms of expression products) and especially the bivalent monospecific constructs advantageously offer means to improve the affinity for the antigen(s) against which they are obtained.
The linker sequence between the V_HHfragments can be for example a sequence corresponding to the coding sequence of the hinge domain of immunoglobulin devoid of light chains (e.g. the long hinge domain) as described by (Hamers-Casterman C. et al, 1993) or a sequence derived therefrom.
As an example, in order to ligate these two variable coding sequences of V_HHfragments to obtain monovalent bispecific construct, the sequence coding for the hinge and CH₂domains, especially coding for the long hinge and CH₂domains of an immunoglobulin devoid of light chains can be used. These domains have been described in WO 94/04678.
As another example, for instance for the preparation of bispecific or multispecific DNA constructs, the sequence used as linker between the V_HHfragments is derived from the coding sequence of the hinge and is devoid of the terminal part containing nucleotides coding for the cysteine residue, or more generally devoid of the codons enabling dimerisation of the V_HHfragment.
Preferred linkers include: the sequence starting at nucleotide 400 and ending at nucleotide 479 or between nucleotides 479 and 486 of the nucleotide sequence disclosed on FIG. 15 or the sequence starting at nucleotide 400 and ending at nucleotide 495 or between nucleotides 487 and 495 of the nucleotide sequence of FIG. 15.
The linkers can be for instance obtained by digestion of a plasmid containing the coding sequence for the V_HH, hinge and CH₂domains of an immunoglobulin devoid of light chains, with BstEII and XmnI (or KpnI) endonucleases and further amplification of the sequence with primers annealing to each end of the hinge coding sequence as described above and illustrated in the examples.
As an example, constructs (monovalent or multivalent, monospecific or multispecific) can be obtained having a specificity with respect to two or more different toxins or generally antigens of different pathogen organisms including bacteria and viruses.
The invention also relates to a process for the preparation of monovalent bispecific constructs of variable fragments of a heavy chain of an immunoglobulins which comprises the following steps:
a) ligating a nucleotide sequence coding for a variable V_HHfragment having a determined antigen- or epitope-specificity to a linker nucleotide sequence to form a V_HHlinker fragment;
b) ligating the formed nucleotide sequence coding for the V_HH-linker fragment to a nucleotide sequence coding for another V_HHfragment having a different antigen- and/or epitope-specificity,
wherein the linker sequence contains the nucleotide sequence coding for part of a hinge domain wherein the codons responsible for the dimerisation of the V_HHfragments especially by formation of a disulfide bridge between the last cysteine residues within the hinge domain are deleted.
According to a preferred embodiment, additional steps of ligation are performed with sequences coding for variable fragments (V_HHfragments) having the same specificity or a different specificity with respect to the above fragments.
In such a case the V_HHfragment coding sequences recovered from step b) must be digested so as to produce a nucleotide sequence having the following structure hinge linker-V_HH. In accordance V_HH-hinge linker)_ncoding sequences are obtained wherein n is a number higher than 2.
Preferably, the sequence encoding the hinge domain preferably the long hinge domain of the immunoglobulins devoid of light chains is the nucleotide sequence comprising or corresponding to nucleotides 400 to 479 or up to nucleotides 486 of the sequence of FIG. 15.
In a particular embodiment of the process for the preparation of bivalent or multivalent monospecific or multispecific constructs, the V_HHfragment coding sequence linked to a nucleotide sequence encoding the hinge domain has to be amplified. Oligonucleotide primers have been defined which permit the amplification of the sequence of interest. These oligonucleotides anneal respectively with their 3′ end to the beginning of the V_HHgene or coding sequence and to the terminal part of the hinge coding sequence. Appropriate primers are for instance (SEQ ID NOS: 9 & 10, respectively):
A4 (Sf I site underlined):

5′-CATGCCATGACTCGCGGCCCAGCCGGCCATGGCCGA(G,T)GT(G,

C)CAGCT-3′
AM007:

5′-GGCCATTTGCGGCCGCATTCCATGGGTTCAGGTTTTGG-3′
These chosen primers contain target sequences for specific endonucleases, thus allowing the cloning of the digestion products of the amplified fragments in a suitable vector.
The obtained DNA constructs are then used to transform host cells, for instance E. coli and the expressed proteins are then isolated and purified. The expression products of these DNA constructs are within the scope of the invention.
The heavy-chain antibodies, such as those derived from camel, and their fragments present clear advantages over other antibodies or fragments thereof derived from other animals. These are linked to the distinctive features of the heavy chain antibodies and in particular the novel fragments which can be produced by proteolytic cleavage within the hinge of these heavy-chain antibodies to generate the V_HHand the (V_HHh)2 fragments. The V_HHdomain of a heavy chain has distinct genetic entities which confer properties of solubility not found in VH fragments derived from conventional antibodies. This property, in addition to its small size and to the fact that the amino acid sequence of the framework region is very homologous to that of human, ensures a minimum of immunogenicity. These properties would allow repetitive treatment with heavy chain V_HHfragments for passive immunisation or antibody therapy. As mentioned above, V_HHand the (V_HHh)2 fragments can easily be produced by proteolytic cleavage of camel immunoglobulins or via recombinant DNA technology.
The most important field of passive immunisation is intoxication due to bacterial toxins and in particular acute intoxication or intoxication due to drug resistant bacteria. Passive immunisation or treatment by antibodies is justified in those cases where vaccination is unpractical or its effects short-lived. They are particularly justified for acute intoxication which if left untreated would have lethal or crippling effects.
The following list of indications is non-exhaustive:
Tetanos due to infection by Clostridium tetani is an important post-trauma infection and current immunisations are not long lasting. It is also important in the veterinary field.
Botulism due to ingestion of toxins produced by Clostridium Botulinum and related species.
Gangrene due to infection by Clostridium.
Necrotic Enteritis and Enterotoxemia in humans and livestock due to Clostridium Perfringens ingestion.
Food poisoning due to Staphylococcal endotoxins in those cases where antibiotics are not recommended.
Pseudomonas infection refractory to antibiotic treatment and in particular ocular infections where rapid intervention is warranted.
Diphteria toxin infection
Pasteurella and Yersinia infection causing lethal outcomes in human and livestock.
Anthrax toxin produced by Bacillus Anthraxis and responsible for one of the five major livestock diseases.
Infections due to other bacterial agents such as Neisseria or viral agents.
Furthermore, the relative resistance of the V_HHfragment to proteolytic cleavage by digestive enzymes (e.g. pepsin, trypsin) offer the possibility of treatment against important gut pathogens, such as Vibrio cholera and other vibrios, enterotoxic E. coli, Salmonella species and Shigella or pathogens ingested with food such as Listeria.
Another major target for immunotherapy is in the treatment of intoxication due to bites or contact with toxic invertebrates and vertebrates. Among the invertebrates are sea anemonies, coral and jellyfish, spiders, bees and wasps, scorpions. In the vertebrates, the venemous snakes are of particular importance and in particular those belonging to the families of Viperidae, Crotalidae and Lapidea.
Passive immunisation with partially purified immunoglobulins from immunized animals are already being used. In developing countries, antitetanos and antidiphteria antisera are still produced on a very large scale, usually in horses. Anti-venom antibodies are produced, although on a much smaller scale, against venoms, especially snake venoms.
Another field of application is in combination with the therapeutic use of toxins in medical or surgical practice where neurotoxins such as botulinum toxin are increasingly used.
The invention also relates to the oligonucleotide primers described hereabove, either alone or in kits.
Other characteristics of the invention will appear from the figures and the examples which are described hereafter.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1: 1% agarose gel electrophoresis of the PstI/BstEII digested PCR amplification product of the camel V_HHgene (lanes 1 and 2) next to the 123 bp ladder of BRL used as a size marker (lane 4). The PCR product comigrates with the 3^rdband of the marker, 369 bp in length.
FIG. 2: Map of the pHEN4 with the nucleotide sequence of the V_HHcloning site shown in the lower part of the figure (SEQ ID NOS: 27-29, respectively). The PstI and BstEII sites can be used to clone the camel V_HHPCR product shown in FIG. 1.
FIG. 3: 100 individual clones were randomly selected from the original camel V_HHlibrary (O), or after the first (1), second (2), third (3) or fourth (4) round of panning. After M13 infection the virions were tested for binding activity against immobilized tetanus toxoid. The number of positive clones are shown as a function of number of pannings.
FIGS. 4A and 4B: Nucleotide sequence and the corresponding amino acid sequence of the two identified camel V_HHanti tetanus toxoid clones pHEN4-αTT1 (FIG. 4A) and pHEN4-αTT2 (FIG. 4B) (SEQ ID NOS: 13 & 14 and 15 & 16, respectively). The framework Serll, Phe37 and kg or Cys 45 characteristic for the camel V_HHheavy chain antibodies (Muyldermans et al., 1994) are double underlined.
FIG. 5: SDS-polyacrylamide gel electrophoresis of the proteins extracted from the periplasm of WK6 cultures induced with IPTG. Lane 1 & 8, protein size marker (Pharmacia) MW are (from top of to bottom) 94,000; 67,000; 43,000; 30,000; 20,100 and 14,400 D. Lanes 2 and 7 Expressed periplasmic proteins extracted from WK6 cells containing pHEN4-αTT2′ and pHENA-αTT1′ cloning vector. Lane 3 & 4, Purified V_HHdomain of pHEN4-αTT2 at 10 and 1 microgram. Lanes 5 & 6, Purified V_HHdomain of pHEN4-αTT1 at 10 and 1 migrogram. The position of the expressed soluble camel VH protein is indicated with an arrow. It is clearly absent in the second lane.
FIG. 6: The total periplasmic extract of 1 liter of culture of WK6 cells carrying the pHEN4-αTT2 was concentrated to 5 ml and fractionated by gel filtration on Superdex 75 (Pharmacia) using 150 mM NaCL, 10 mM sodium phosphate pH 7.2 as eluent. The pure V_HHis eluted at the fractions between the arrows.
FIG. 7: CD (Circular dichroism) spectrum (Absorbance versus wavelength in nm) of the purified V_HHdomain αTT2 at 3.9×10⁻⁶Min water measured in a cuvette with a path length of 0.2 cm. The negative band near 217 and 180 nm and the positive band around 195 nm are characteristic for β structures (Johnson, 1990).
FIG. 8: Specificity of antigen binding shown by competitive ELISA. The experiments were carried out in triplicate with the bacterial periplasmic extracts of pHEN4-αTT1 and pHEN4-αTT2.
FIG. 9: Number of mice surviving after I.P injection of 100 ngr tetanus toxin (10×LD50) or co-injection of tetanus toxin with the purified V_HHαTT1, αTT2 or the non-specific cVH21 (Muyldermans et al., 1994) at 4 or 40 microgram.
FIG. 10: Variability plot of the camelid V_HHsequence (CDR3 and framework 4 regions are not included).
The alignment of the V_HHamino acid sequences of camel and lama (a total of 45 sequences) was performed according to Kabat et al. The variability at each position was calculated as the number of different amino acids occurring at a given position, divided by the frequency of the most common amino acid at that position. Positions are numbered according to Kabat et al. The positions above the horizontal bar indicate the amino acids which are referred to as (CDR1) and (CDR2) in the consensus sequence.
A variability number equal to 1 indicates a perfectly conserved amino acid at that position. The higher the variability number the more likely it will be that the amino acid at this position will deviate from the consensus sequence.
FIG. 11: Nucleic acid sequence of LYS2 V_HHand translation product thereof (SEQ ID NOS: 17 & 18, respectively).
FIG. 12: Nucleic acid sequence of LYS3 V_HHand translation product thereof (SEQ ID NOS: 23 & 24, respectively).
FIG. 13A: Scheme to construct the bivalent monospecific anti-LYS3 camel V_HH.
FIG. 13B: Scheme of constructed monovalent bispecific andi-LYS3-long hinge linker-anti-LYS2-Tag and schematic diagram of same.
FIG. 14A: Scheme of constructed bispecific anti-LYS3-long hinge linker-anti-LYS2-Tag and schematic diagram of same.
FIG. 14B: Scheme of constructed monovalent bispecific anti-LYS3-long hinge linker-anti-LYS-2 Tag and schematic diagram of same.
FIGS. 15A and 15B: Nucleotide and amino acid sequence (SEQ ID NOS: 21 & 22, respectively) of the anti-LYS3-long hinge/Cys-Tag. This protein will spontaneously dimerise.
FIGS. 16A, 16B and 16C: Nucleotide and amino acid sequence (SEQ ID NOS: 23 & 24, respectively) of the anti-LYS3-long hinge linker-anti-LYS2-Tag polypeptide.

DETAILED DESCRIPTION OF THE INVENTION

EXAMPLE I

Generation of Specific Camel V_HHFragments Against Tetanus Toxoid
In this application, results are presented, which prove the feasibility of generating specific camel V_HHfragments with demonstrated folding and good binding affinity. This was done by generating a library of camel fragments derived from the dromedary IgG2 and IgG3 isotype, display of the V_HHlibrary on phage as fusion proteins with the gene III protein of bacteriophage M13 to allow selection of the antigen binders, and finally of expressing and extracting the soluble and functional V_HHfragments from E. coli. As antigen, we choose the tetanus toxoid was chosen because comparisons are possible with published data. In addition, the tetanus toxoid is a highly immunogenic protein that is routinely used as a vaccine in humans to elicit neutralizing antibodies. The two camel V_HHfragments that were identified were specific and of high affinity. The affinities of the two camel V_HHfragments appear to be comparable with those from the human anti-tetanus toxoid F_AB'S recently obtained by Mullinas et al. (1990) and by Persson et al. (1991).
Camel Immunization
The serum of a camel (Camelus dromedarius) was shown to be non-reacting with tetanus toxoid (RIT, Smith Kline Beecham, Rixensart, Belgium). This camel was injected with 100 μg tetanus toxoid at days 9, 30, 52, 90 and with 50 μg at days 220, 293 and 449. The blood was collected 3 days after each injection.
mRNA Purification of Camel Blood Lymphocytes
Peripheral blood lymphocytes were purified with Lymphoprep (Nycomed, Pharma) from the bleeding at day 452. Aliquots of 1×10⁶−5×10⁶cells were pelleted and frozen at −85° C. and subsequently used as an enriched source of B-cell mRNA for anti-tetanus toxoid.
The mRNA was prepared from a total of 106 peripheral blood lymphocytes either by the “Micro FastTrack” mRNA isolation kit (Invitrogen) or the “QuickPrep Micro mRNA Purification” kit of Pharmacia, following the recommendations of the manufacturer. With both protocols, up to a few μgr of mRNA was obtained which was used in the subsequent cDNA synthesis step.
cDNA Synthesis and PCR Amplification of Camel V_HHGene
The first-strand cDNA was synthesized with the Invitrogen “cDNA-cycle” or the Pharmacia “Ready-To-Go” kit. The first-strand cDNA was used immediately afterwards for the specific amplification of the camel V_HHregion by PCR. The primers used have following sequences: the BACK primer (SEQ ID NO: 1) (5′-GATGTGCAGCTGCAGGCGTCTGG(A/G)GGAGG-3′), the internal PstI site is underlined) is designed to hybridize to the framework 1 region (codons 1 to 10) of the camel V_HH, while the FOR primer (5′-CGCCATCMGGTACCAGTTGA-3′) hybridizes in the CH2 region. The PCR was carried out with the Taq polymerase from Boehringer Mannheim.
The PCR product was purified according to standard protocols (Sambrook et al., 1989) and digested with the PstI restriction enzyme of which the target site occurred in the BACK primer, and with BstEII which has a naturally occurring site in the framework 4 of the camel V_HHregions. The resulting fragments of approximately 360 bp (FIG. 1) were ligated into the pHEN4 vector cut with the same restriction enzymes. The pHEN4 vector (FIG. 2) is the pHEN1 plasmid (Hoogenboom et al., 1991)—a pUC119 based vector—where the myc-tag was replaced by the decapeptide tag present in the ImmunoZAP H vector (Stratacyte). Also the polylinker was modified to allow the cloning of the camel V_HHgene between a PstI and a BstEII site located after the PeIB leader signal and in front of the decapeptide tag and gene III of bacteriophage M13.
Construction of a Camel V_HHLibrary
The ligated DNA material was precipitated with 10 volumes and resuspended in 10 μl water and electrotransformed in E. coli XL1 Blue MRF′ cells (Stratagene). After electroporation according to the recommended protocol (Stratagene) we kept the cells for 1 hour at 37° C. in 1 ml SOC medium before plating on LB plates containing 100 μg ampicilline/ml. After an over night incubation at 37° C. the transformed cells were grown out into colonies and some 500,000 recombinant clones were obtained. About 20 colonies, randomly selected, were toothpicked and grown in selective medium (LB/Ampicilline) to prepare plasmid DNA and to check their insert by sequencing. For each clone tested, we found a different V_HHregion with the amino acid sequence and contents characteristic for a V_HHoriginating from a camel heavy chain immunoglobulin (Muyldermans et al., 1994). This indicates that a vast camel V_HHlibrary was generated.
The remaining 500,000 clones were scraped from the plates with a minimal amount of LB containing 50% glycerol and stored at −85° C. until further use.
Panning with Tetanus Toxoid
The library was screened for the presence of anti-tetanus toxoid camel V_HH's by panning. To this end, approximately 10⁹cells (=5 ml suspension of the frozen recombinant clones) were grown to midlogarithmic phase in 200 ml of LB medium supplemented with 1% glucose and 100 μg ampicilline/ml before infection with M13K07 bacteriophages. After adsorption of the bacteriophages on the E. coli cells for 30 min at room temperature, the cells were harvested by centrifugation and washed in LB medium supplemented with ampicilline and kanamycin (25 μg/ml). The cells were incubated overnight at 37° C. to secrete the recombinant pHEN plasmid packaged within the M13virion containing a camel V_HHfused to some of its M13gene III proteins (Hoogenboom et al., 1991). The phagemid virions were prepared according to the protocol described by Barbas et al. (1991). The phage pellets were resuspended in blocking solution (1% casein in phosphate buffered saline, PBS), filtered through a 0.2 μm filter into a sterile tube and used for panning. For the panning the Falcon 3046′ plates were coated overnight with 0.25 mg/ml or 2 mg/ml tetanus toxoid dissolved in PBS or hydrogencarbonate pH 9.6. The wells were subsequently washed and residual protein binding sites were blocked with blocking solution at room temperature for 2 hours. The adsorption of the phagemid virions on the immobilized antigen and the washing and elution conditions were according to Marks et al. (1991) or were taken from the protocol described by the <<Recombinant Phage Antibody System>> of Pharmacia. Four consecutive rounds of panning were performed. After the fourth round of panning the eluted phagemid virions were added to exponentially growing TGI cells (Hoogenboom et al. 1991) and plated on ampicillin containing LB plates. After overnight growth several colonies were grown individually in LB medium to midlogarithmic growing phase, and infected with M13K07 helper phage. The virions were prepared and tested for their binding activity against tetanus toxid immobilised on mitrotiter plates. The presence of the virion binding to the immobilized antigen was revealed by ELISA using a Horse RadishPeroxidase/anti-M13 conjugate (Pharmacia). The percentage of binders was increasing after each round of pannning. In the original library we found 3 clones out of 96 which showed binding with the immobilized tetanus toxoid. This number was increased to 11, 48 and 80 after the first, second and third round of panning. All of the individual clones which were tested after the fourth round of panning were capable of recognizing the antigen, as measured by ELISA (FIG. 3). Ten positive clones were grown and tested by PCR to check the presence of an insert with the proper size of the V_HHgene, and their DNA was finally sequenced. The sequencing data revealed that two different clones were present among this set of 10 clones. The plasmid DNA of these clones was named pHEN4-αTT1 and pHEN4-αTT2. Plasmid DNA was deposited at the “Belgian Coordinated Collections of Microorganisms” BCCM/LMBP on Jan. 31, 1995 under accession number LMBP3247, and it was shown that these two different clones contained a cDNA coding for a camel V_HH(FIG. 4). Comparison of the amino acids in these clones with the camel V_HHclones analysed before (Muyldermans et al., 1994) clearly indicated that the anti-tetanus camel V_HHoriginated from a heavy chain immunoglobulin lack of the CH1 domain and light chains. Especially the identity of the key residues at position 11 (Ser), 37 (Phe) and 45 (Arg or Cys) and 47 (Leu or Gly) proved this statement (Muyldermans et al., 1994).
Production of Soluble Camel V_HHwith Anti-Tetanus Toxoid Activity
The plasmid DNA of the two clones which scored positive in the tetanus toxoid ELISA were transformed into WK6 cells. These cells are unable to suppress the stopcodon present in the vector between the decapeptide tag and the gene III protein. The WK6 E. coli cells harboring the pHEN4-αTT1 or pHEN4-αTT2 plasmid were grown at 37° C. in 1 liter of TB medium with 100 mgr ampicillin/ml and 0.1% glucose. When the cells reached an OD₅₅₀of 1.0 we harvested the cells by centrifugation at 5000 rpm, 10 minutes. The cell pellet was washed once in TB medium with ampicillin, but omitting the glucose. The cells were finally resuspended in 1 liter of TB medium with ampicillin (100 μg/ml). We induced the expression of the camel V_HHdomain by the addition of 1 mM IPTG and further growth of the cells at 28° C. for 16 hours. The expressed proteins were extracted from the periplasmic space following the protocol described by Skerra and Plucthun (1988). We pelleted the E. coli cells by centrifugation at 4000 g for 10 min. (4° C.). The cells were resuspended in 10 ml TES buffer (0.2 M Tris-HCl pH 8.0, 0.5 mM EDTA, 0.5 M sucrose). The suspension was kept on ice for 2 hours. The periplasmic proteins were removed by osmotic shock by addition of 20 ml TES diluted ¼ with water. The suspension was kept on ice for 1 hour and subsequently centrifuged at 12,000 g for 30 minutes at 4° C. The supernatant contained the expressed camel V_HHdomain. The extract corresponding to 400 μl cell culture was applied under reducing conditions on a SDS/polyacrylamide protein gel. The extracted proteins were visualized in the SDS/polyacrylamide gels by Coomassie blue staining (FIG. 5). A protein band with an apparent molecular weight of 16,000 D was clearly present in the E. coli cultures containing the recombinant clones and induced with IPTG. Alternatively, the presence of the camel V_HHproteins in the extract was revealed by Western blot using a specific rabbit anti-camel V_HHor rabbit anti-dromedary IgG serum or the anti-tag antibody.
We estimate from the band intensity observed in the Coomassie stained gel that more than 10 mg of the camel V_HHprotein (non-purified) can be extracted from the periplasm of 1 liter induced E. coli cells.
For the purification of the anti-tetanus toxoid camel V_HHwe concentrated the periplasmic extract 10 times by ultrafiltration (Millipore membrane with a cut off of 5000 Da). After filtration the concentrated extract from the pHEN4-αTT2 was separated according its molecular weight by gelfiltration on Superdex-75 (Pharmacia) (FIG. 6) equilibrated with PBS (10 mM phosphate buffer pH 7.2, 150 mM NaCl). The peak containing the anti-tetanus toxoid activity eluted at the expected molecular weight of 16,000 Da indicating that the protein behaved as a monomer and does not dimerize in solution. The fractions containing the pure V_HH(as determined by SDS-PAGE) were pooled and the concentration was measured spectrophotometrically using a calculated E₂₈₀(0.1%) of 1.2 and 2.3 respectively for the αTT1 and αTT2. From the UV absorption at 280 nm of the pooled fraction we could calculate a yield of 6 mgr of purified protein per liter of bacterial culture. The purified protein could be further concentrated by ultrafiltration to 6 mgr/ml in PBS or water without any sign of aggregation, as seen on the UV spectrum.
Concerning the expression yield in E. coli it should be realized that at this stage we did not try to optimize the expression or the protein extraction conditions. However, as the yield of the purified αTT2 camel V_HHreached 6 mgr per liter of bacterial culture, and as we obtained the soluble protein at a concentration of 6 mgr/ml, it is clear that the expression is comparable or better than other scFv's or FAB's expressed in E. coli. Furthermore, the solubility of the camel V_HHαTT2 is certainly better than that obtained for the mouse V_Hfragments. The yield and solubility is certainly in the range needed for most applications.
To prove the proper folding of the purified protein, the αTT2 was brought at a concentration of 3.9×10⁻⁶M and used it for CD measurement (FIG. 7). The CD spectrum is characteristic for a polypeptide with a β-pleated sheet folding as expected for a well structured immunoglobulin fold (Johnson, 1990).
The Camel Anti-Tetanus Toxoid V_HHAffinity Measurements
The binding of the camel V_HHantibody to the tetanus toxoid immobilised on the microtiter plates was revealed by the successive incubation with firstly, the rabbit anti-camel V_HHor rabbit anti-dromedary IgG, and secondly, a goat anti-rabbit/alkaline phosphatase conjugated antibodies (Sigma). The apparent affinity of the camel V_HHproteins against tetanus toxoid was estimated by inhibition ELISA exactly as described by Persson et al. (1991) for the human anti-tetanus toxoid FAB fragments they produced in E. coli.
The specificity of the soluble camel V_HHfor the tetanus toxoid was suggested from the ELISA experiments in which we competed the binding with free antigen was competed. An apparent inhibition constant of around 10⁻⁷, 10⁻⁸M was observed for both V_HHfragments (FIG. 8). This compares favorably with the inhibition constants for the human anti-tetanus toxoid FAB fragments cloned by Persson et al. (1991) which were in the range of 10⁻⁷to 10⁻⁹M.
The measurement of the affinity constant by ELISA is however, more reliable if determined according to the procedure of Friguet et al. (1987). With this protocol we found an affinity constant of 6.10⁷M⁻¹and 2.10⁷M⁻¹for the αTT1 and αTT2 respectively. These affinities are consistent with a specific V_HH-antigen interaction (the polyspecific antibodies generally bind their antigen with affinities of 10⁶M⁻¹or less (Casali et al. 1989)).
Epitope Recognition of αTT1 and αTT2.
Tetanus toxin consists of three domains. The C fragment binds to the neuronal cells, it is said to be the neurospecific binding domain. The B domain appears to be involved in the neuronal penetration of the A domain or L chain (Montecucco & Schiavo, 1993). The L chain is responsible for the intracellular activity.
The C fragment is the most immunogenic part of the tetanus neurotoxin, and a recombinant C fragment is commercially available (Boehringer and Calbiochem). We showed by ELISA that the αTT1 bacterial extract binds equally well both to the complete tetanus toxoid and to the recombinant C fragment. Therefore the epitope of this camel V_HHis present on the C fragment. By contrast, the αTT2 extract binds to the complete tetanus toxoid, but not to the C fragment. Therefore the αTT2 recognizes an epitope located on the A or B domain.
The in vivo Neutralization of Tetanus Toxin Toxicity
The neutralizing activity of the purified camel αTT1 or αTT2 V_HHdomains against tetanus toxin was tested. As a control, eight NMRI mice of 8 to 12 weeks (80 to 100 gr) were injected I.P. with 400 ng tetanus toxin (SmithKline Beecham Biologicals) (=10 times the LD50) in 0.1 ml PBS. To test the neutralizing activity of the camel V_HHαTT1 or αTT2 we preincubated 4 or 40 mg of this purified recombinant protein with 400 ng of the tetanus toxin in 0.1 ml of PBS for 30 minutes before I.P. injection into the mice. The survival of the mice was followed over a period of 2 weeks (FIG. 9). It is clear that all mice injected with the tetanus toxin alone or in the presence of a non-specific purified camel V_HH(cVH21 of Muyldermans et al., 1994) were killed within 3 days. The survival of the mice injected with the tetanus toxin was increased significantly by the co-injection of only 4 mg of the purified camel αTT1 or αTT2. The survival was even more pronounced for the co-injection of tetanus toxin with 40 mg of camel V_HH. It appears that the αTT1 had a slightly higher neutralizing activity than the αTT2. This could originate from its intrinsic higher affinity for binding the tetanus toxin (Simpson et al., 1990). Alternatively it might result from the binding of the αTT1 V_HHto the fragment C of the tetanus toxin which inhibits more the toxic effect than the binding of the αTT2 to its epitope outside the C fragment.

EXAMPLE II

Generation of Specific Camel V_HHFragments Against Lysozyme
Using the same protocol as the one described in Example I (specific steps or conditions modifying those of Example I are indicated hereafter) for the generation of specific camel V_HHfragments having a specificity and an affinity for tetanus toxoid, V_HHfragments have been obtained against lysozyme.
We chose the Hen Egg Lysozyme (HEL) as an antigen to immunize a camel (Camelus dromedarius). This protein was selected for the reason that comparisons can be made with several other mouse monoclonal antibody fragments recognizing the same antigen and of which the structure even in complex with its antigen are known.
Camel Immunization
The serum of a camel was shown to be non-reacting with lysozyme. We injected this camel with 100 μg lysozyme (Boehringer) at days 9, 30, 52, 90 and with 50 μg at days 220, 293 and 449. The blood was collected on average 3 days after each injection.
The following steps were then performed as in Example 1.
mRNA purification of camel blood lymphocytes.
cDNA synthesis and PCR amplification of camel V_HHgene.
Construction of Camel V_HHlibrary.
Panning with lysozyme (the Falcon 3046′ plates were coated with 1 mg/ml lysozyme).
96 colonies were randomly chosen and grown individually in LB medium.
The virions were prepared and tested for their binding activity against lysozyme immobilised on microtiter plates.
The percentage of binders was increasing after each round of panning. Twenty positive clones were grown and tested by PCR to check the presence of an insert with the proper size of the V_HHgene, and their DNA was finally sequenced. The sequencing data revealed that two different clones were present among this set of 10 clones. The plasmid DNA of these clones was named pHEN4-αLYS2 and pHEN4-αLYS3, and it was shown that these two different clones contained a cDNA coding for a camel V_HH(FIGS. 11, 12). Comparison of the amino acids in these clones with the camel V_HHclones we analysed before (Muyldermans et al., 1994) clearly indicated that the anti-lysozyme camel V_HHoriginated from a heavy chain immunoglobulin lacking the CH1 domain and light chains. Especially the identity of the key residues at position 11 (Ser), 37 (Phe), 44 (Glu), 45 (Arg) and 47 (Gly) proved this statement (Muyldermans et al., 1994).
Production of Soluble Camel V_HHwith Anti-Lysozyme Activity
For the purification of the anti-lysozyme camel V_HHwe concentrated the periplasmic extract 10 times by ultrafiltration (Millipore membrane with a cut off of 5000 Da). After filtration the concentrated extract from the pHEN4-αLYS2 can be purified by Protein A-Sepharose chromatography. Elution of the αLYS2 V_HHis done with 100 mM Tri-ethanol amine. The pH of eluate is immediately neutralized with 1 M Tris-HCl (pH 7.4). Unfortunately the expressed αLYS3 V_HHdoes not bind to Protein A. Therefore the purification has to be performed by affinity chromatography. The concentrated extract is applied on a column of lysozyme immobilized on CNBr-Sepharose (Pharmacia). Elution of the anti-lysozyme V_HHis obtained with 100 mM Tri-ethanolamine. The eluate has to be neutralized as described above. Further purification of both anti-lysozyme V_HH'S can obtained by gel filtration on Superdex-75 (Pharmacia) equilibrated with PBS (10 mM phosphate buffer pH 7.2, 150 mM NaCl). The peak containing the anti-lysozyme activity eluted at the expected molecular weight of 16,000 Da indicating that the protein behaved as a monomer and doesn't dimerize in solution. The fractions containing the pure V_HH(as determined by SDS-PAGE) were pooled and the concentration was measured spectrophotometrically. A yield of 5 mg of purified protein per liter of bacterial culture was calculated. The purified protein could be further concentrated by ultrafiltration to 10 mg/ml in PBS or water without any sign of aggregation, as seen on the UV spectrum.
The Camel Anti-Lysozyme V_HHAffinity Measurements
The specificity of the soluble camel V_HHfor the lysozyme was suggested from the ELISA experiments in which we competed the binding with free antigen. An apparent inhibition constant of around 5.10⁻⁷and 5.10⁻⁸M was observed for the α-LYS3 and α-LYS2 respectively. These affinities are consistent with a specific V_HH-antigen interaction (the polyspecific antibodies generally bind their antigen with affinities of 10⁶M⁻¹or less (Casali et al. 1989).
Epitope Recognition of α-LYS2 and α-LYS3
To analyse whether the two camel V_HHwith anti-lysozyme activity bind to the same or to different epitopes we used the techniques of additive binding in ELISA (Friguet et al., 1989). An addivity index of more than 40 indicates pairs of antibodies that can bind simultaneously on the antigen, while addivity indices of less than 20 is characteristic for pairs of antibodies with overlapping epitopes. Our camel α-LYS2 and α-LYS3 had an addivity index of 45. From this experiment it appears that the α-LYS2 and α-LYS3 bind to different epitopes on the lysozyme molecule.

EXAMPLE 3

Making Bivalent Monospecific or Monovalent Bispecific Binding Constructs from Camelid V_HH's
From the camel V_HH's with specificity to tetanus toxin (α-TT1 or α-TT2) or with specificity to lysozyme (α-LYS2 or α-LYS3) cloned in the pHEN4 bacterial expression vector, we made constructs with following characteristics:
1. V_HHwith ProX repeat sequences of the camel long hinge including the 3 Cys and part of the CH2 domain. These constructs can be also used as an intermediate for the next constructs.
2. V_HHwith ProX repeat sequences of the long hinge of camel with one Cys followed by a stopcodon in the pHEN4. These are bivalent constructs with monospecificity.
3. V_HHlinked with the ProX repeat sequences of the long hinge of camel (without Cys) followed by a second V_HH. These are monovalent constructs with bispecificity, or bivalent constructs with monospecificity depending on the V_HH's.
1. Camel V_HHwith Camel Long Hinge and Dart of CH2 Domain
The (pHEN4-α-LYS3) or the (pHEN4-α-TT2) plasmids were digested with BstEII and Xmn I. BstEII cuts in the framework 4 of the camel V_HH, and Xmn I cuts in the β-lactamase gene of pHEN4. The DNA fragment containing the camel V_HHwas isolated from agarose gel.
A clone containing a camel V_HHwith unknown specificity, the camel long hinge and the first part of the CH2 domain cloned in pBluescript (Statagene) was cut with the same enzymes (Bst EII and Xmn I) and the DNA resulting fragment containing the hinge and CH2 parts was isolated from agarose gel.
The two DNA fragments (one containing the V_HHof determined specificity, the other containing the coding sequence of the hinge and CH₂domains) were mixed and ligated to each other and used to transform E. coli cells. As a result a (pHEN4-α-LYS3-long hinge-CH2) plasmid and a (pHEN4-α-TT2-long hinge-CH2) plasmid have been obtained.
2. Bivalent Monospecific Constructs (FIGS. 13A, 13B and 15)
The (pHEN4-α-LYS3-long hinge-CH2) plasmid was taken as template for amplification with primers A4 and AM007 (SEQ ID NOS: 9 & 10, respectively).
A4 (Sfi I Site Underlined):

5′-CATGCCATGACTCGCGGCCCAGCCGGCCATGGCCGA(G,T)GT(G,

C)CAGCT-3′
AM007:

5′-GGCCATTTGCGGCCGCATTCCATGGGTTCAGGTTTTGG-3′
These primers anneal respectively with their 3′ end to the beginning of the V_HHand to the end of the structural upper hinge of the camel long hinge sequence. The primer AM007 will extend the 3′ end of the α-LYS3 or of the α-TT2 gene (depending on the template) with (SEQ ID NO: 12) CCCATGGAATGCGGCCGCAAATGTCC. The NcoI and NotI sites are underlined. These nucleotides up to the Not I site code for the amino acids Pro Met Glu Cys.
The PCR fragment is double digested with Sfi I and Not I, and the resulting fragments are cloned in the pHEN4 vector cleaved with the same enzymes. The ligated material is transformed in WK6 E. coli cells and selected on ampicillin. The transformed clones are checked for their insert by PCR and by sequencing. The plasmid (pHEN4-α-LYS3 long hinge/Cys) and (pHEN4-α-TT2-long hinge/Cys) were generated.
The extraction of the expressed V_HHα-LYS3-long hinge/Cys or α-TT2-long hinge/Cys proteins lead to isolation of a dimerised molecule because of the formation of the disulfide bridge between the Cys residue within the long hinge. Both camel V_HHdimer constructs (α-LYS3 long hinge/Cys)₂and α-TT2 long hinge/Cys)₂are well expressed in E coli upon induction with IPTG, and are easily obtained from the periplasm. They were quite soluble and bound the original antigen with high affinity and high specificity.
3. Monovalent Bispecific Protein Constructs (FIGS. 14A, 14B and 16)
In the previous plasmid constructs (pHEN4-α-LYS3 long hinge/Cys) and (pHEN4-α-TT2-long hinge/Cys), we have two restriction sites for Nco I. Digestion of the plasmid with this enzyme allows the isolation of the camel V_HHgene followed by the long hinge without the Cys codon. Ligation of the (α-LYS3-long hinge) fragment into the pHEN4-α-LYS2 or in the pHEN4-α-TT2 plasmids linearised with Nco I creates the plasmids (pHEN4-α-LYS3-long hinge linker-.alpha.-LYS2) or (pHEN4-.alpha.-LYS3long hinge linker-α-TT2). Expression of the gene leads to the production of the α-LYS3 V_HHlinked to the α-LYS2 V_HHor linked to the α-TT2 V_HHby the intermediate of a linker based on the structural upper hinge of the camel long hinge.
Following this protocol monovalent bispecific proteins consisting of the camel V_HHof α-LYS3 linked to the camel V_HHof α-LYS2 and that of the camel V_HHof α-LYS3 linked to the camel V_HHof α-TT2 can be isolated. Both proteins are expressed well in E coli and can be extracted from the periplasm. In ELISA the binding properties of the latter protein to the tetanus toxoid and to the lysosyme can be shown.
With these gene constructs at hand it becomes possible and straightforward to exchange either V_HHwith any other V_HHwith another specificity.
For example we can exchange the second camel V_HHby digesting the plasmid with Pst I, or with Nco I and to ligate the DNA fragment containing the V_HH-long hinge linker into the pHEN4-V_HHlinearised with either Pst I or Nco I.
Similarly, we exchanged the first camel V_HHα-LYS3 gene from the (pHEN4-α-LYS3 long hinge linker-α-LYS2) plasmid construct into (pHEN4-α-TT1-long hinge linker-α-LYS2). This was done by cutting the plasmid with BstEII and further ligating the DNA fragment containing the (long hinge linker-α-LYS2) into the (pHEN4-α-TT1) plasmid linearised with BstEII.
With a slight modification of this protocol it becomes even possible to generate multivalent constructs. In this case the (V_HH-long hinge linker-V_HH) plasmid needs to be digested with BstEII and the DNA fragment containing the (long hinge linker-V_HH) gene should be isolated from agarose gel. Because of the asymmetry in the recognition site of BstEII, it is only possible to obtain head-to-tail ligations upon self ligation. The self-ligated DNA should thereafter (with or without prior size selection) be ligated into the pHEN4-V_HHplasmid linearised with BstEII. This will create a plasmid of the type (pHEN4-[V_HH-long hinge linker]_n).

REFERENCES CITED IN THE PRECEDING EXAMPLES

Borrebaeck et al., (1992) Bio/Technology 10, 697-698.
Casali et al., (1989) Annu. Rev. Immunol. 7, 513-536.
Friguet et al., (1983) J.Immunol. Meth. 60, 351 et (1989) Protein Structure.
A practical approach (Ed. T. E. Creighton) IRL Press p. 287-310)
Glockshuber et al., (1990) Biochemistry 29, 1362-1367.
Hamers-Casterman et al., (1993) Nature 363, 446-448.
Hoogenboom et al., (1991) Nucl. Acids Res. 19, 4133-4137.
Johnson W. C., (1990) Proteins: Structure, Func & Genetics 7, 205-214.
Kabat et a. (1991) Sequences of Proteins of Immunological Interest (US Dept.Health Human Services, Washington) 5th Ed.
Marks et al., (1991) J. Mol. Biol. 222, 581-597.
Montecucco & Schiavo (1993) TIBS 18, 324-329.
Mullinax et al., (1990) Proc.Natl.Acad.Sci USA 87, 8095-8099.
Persson et al., (1991) Proc.Natl.Acad.Sci USA 88, 2432-2436.
Sambrook et al., (1989) Molecular Cloning CSHL Press
Simpson et al., (1990) J. Pharmacol. & Exp. Therap. 254, 98-103.
Skerra and Pluckthun, (1988) Science 240, 1038-1040.
Sequence CWU 1

Having thus described several aspects of at least one embodiment of this invention, it is to be appreciated various alterations, modifications, and improvements will readily occur to those skilled in the art. Such alterations, modifications, and improvements are intended to be part of this disclosure, and are intended to be within the spirit and scope of the invention. Accordingly, the foregoing description and drawings are by way of example only.

Claims

1. A method for treating a subject comprising the step of passive immunization or immunotherapy with a variable fragment of a heavy polypeptide chain (V_HH) of an immunoglobulin devoid of light chains.

2. The method according to claim 1, wherein the passive immunization is a treatment for infection or intoxication due to a bacterial toxin.

3. The method according to claim 2, wherein the intoxication is acute intoxication or intoxication due to drug resistant bacteria.

4. The method according to claim 2, wherein the infection or intoxication is due to infection by bacterial agents or viral agents.

5. The method according to claim 4, wherein the bacterial agent is Clostridium, Staphylococcus, Pseudomonas, Pasteurella, Yersinia, Bacillus anthracis, Neisseria, Vibrio, E. coli, Salmonella, Shigella, Listeria, Corynebacterium diphtheriae or a pathogen ingested with food.

6. The method according to claim 5, wherein the Clostridium is Clostridium tetani, Clostridium botulinum or Clostridium perfringens.

7. The method according to claim 5, wherein the Vibrio is Vibrio cholera.

8. The method according to claim 5, wherein the E. coli is enterotoxic E. coli.

9. The method according to claim 2, wherein the intoxication is due to bites or contact with a toxic invertebrate or toxic vertebrate.

10. The method according to claim 9, wherein the toxic invertebrate is a sea anemone, a coral, a jellyfish, a spider, a bee, a wasp or a scorpion.

11. The method according to claim 9, wherein the vertebrate is a venomous snake.

12. The method according to claim 11, wherein the venomous snake is a snake of the Viperidae family, Crotolidae family or Lapidae family.

13. The method according to claim 1, wherein the treatment is for tetanus, botulism, gangrene, necrotic enteritis, enterotoxemia, food poisoning, ocular Psuedomonas infection, diphtheria, or anthrax.

14. The method according to claim 1, wherein the passive immunization or immunotherapy is performed in combination with medical or surgical use of a neurotoxin.

15. The method according to claim 14, wherein the neurotoxin is botulinum toxin.

16. The method according to claim 1, wherein the subject is a human.