HK1247216B - Microbial transglutaminases, substrates therefor and methods for the use thereof - Google Patents

Description

Microbial transglutaminases, substrates and methods of use thereof

CROSS-REFERENCE TO RELATED APPLICATIONS

This application is based upon, claims the benefit of, and incorporates by reference U.S. Provisional Patent Application Serial No. 62/094,495, filed on December 19, 2014, and entitled “Identification of Transglutaminase Substrates and Uses Therefor,” and U.S. Provisional Patent Application Serial No. 62/260,162, filed on November 25, 2015, and entitled “System and Method for Identification and Characterization of Transglutaminase Species.”

Statement Regarding Federally Funded Research

not applicable.

Background Art

The present invention relates generally to the identification of transglutaminases and substrates, and more specifically to the discovery and characterization of a microbial transglutaminase from Kutzneria albida .

Elucidating the details of enzyme activity and specificity is important for understanding the physiological functions of enzymes and for the biotechnological applications of reactions catalyzed by enzymes. For example, transglutaminases belong to a large family of related enzymes that includes microbial and mammalian transglutaminases. Transglutaminases catalyze the cross-linking of two polypeptides or peptide chains by forming an isopeptide bond between the γ-carboxamide group of a glutamine residue and the ε-amino group of a lysine residue. Elucidating the details of transglutaminases' activity and specificity is important for the biotechnological applications of cross-linking reactions catalyzed by transglutaminases (e.g., modification of proteins for labeling, tagging, multiprotein complex formation, etc.).

Because of its small size, robust performance, stability and calcium independence of its activity, microbial transglutaminase is the most studied transglutaminase to date. Several studies have demonstrated that a variety of long alkylamines can replace the lysine substrate of transglutaminase and that the simple dipeptide glutamine-glycine can serve as a glutamine substrate. These discoveries of lysine and glutamine substrates for transglutaminases have facilitated the development of various tests for transglutaminase activity and practical assays for modifying proteins using transglutaminases. However, several challenges may still arise in the identification and characterization of known and novel transglutaminases. One challenge is that the specificity of a particular transglutaminase for isopeptide bond formation may be too broad or too narrow for a particular application. Another challenge is that transglutaminases with the same or similar substrate specificity are not available for orthogonal labeling strategies, etc. Another challenge is the identification of substrates for uncharacterized or poorly characterized transglutaminases. Additional challenges may arise with factors related to a given transglutaminase, such as origin, specificity, activity, stability, etc. and combinations thereof.

SUMMARY OF THE INVENTION

The present invention overcomes the aforementioned disadvantages by providing systems and methods for identifying and characterizing transglutaminases, as well as substrates and uses therefor.

According to one aspect of the present invention, a substrate tag for a microbial transglutaminase comprises one of an acyl-donor tag having at least 80% sequence identity to the peptide sequence YRYRQ (SEQ ID NO: 1), and an amine-donor tag having at least 80% sequence identity to the peptide sequence RYESK (SEQ ID NO: 2).

In one aspect, the microbial transglutaminase has at least 80% sequence identity to Kuznetsova albicans microbial transglutaminase (SEQ ID NO: 6).

In another aspect, the substrate tag further comprises a detectable label.

In another aspect, the detectable label is selected from the group consisting of a biotin moiety, a fluorescent dye, a ruthenium label, a radioactive label, and a chemiluminescent label.

In another aspect, the acyl-donor tag has the peptide sequence APRYRQRAA (SEQ ID NO: 24).

According to another aspect of the invention, a method for forming an isopeptide bond in the presence of a microbial transglutaminase comprises exposing the microbial transglutaminase to a first substrate and a second substrate, wherein the first substrate comprises an acyl-donor tag having at least 80% sequence identity to the peptide sequence YRYRQ (SEQ ID NO: 1), and the second substrate comprises an amine-donor tag having at least 80% sequence identity to the peptide sequence RYESK (SEQ ID NO: 2), and cross-linking the first substrate and the second substrate, thereby forming an isopeptide bond between the acyl-donor tag and the amino-donor tag.

In one aspect, the microbial transglutaminase has at least 80% sequence identity to Kuznetsova albicans microbial transglutaminase (SEQ ID NO: 6).

In another aspect, the step of cross-linking the first substrate and the second substrate forms an isopeptide bond between the γ-carboxamide group of the acyl-donor tag and the ε-amino group of the amino-donor tag.

In another aspect, at least one of the first substrate and the second substrate comprises a detectable label.

In another aspect, the detectable label is selected from the group consisting of a biotin moiety, a fluorescent dye, a ruthenium label, a radioactive label, and a chemiluminescent label.

In another aspect, the acyl-donor tag has the peptide sequence APRYRQRAA (SEQ ID NO: 24).

In another aspect, cross-linking of the first substrate to the second substrate is achieved in a yield of at least about 70%.

In another aspect, the yield is achieved in about 30 minutes.

According to another aspect of the invention, a kit for forming isopeptide bonds in the presence of a microbial transglutaminase comprises a purified microbial transglutaminase having at least 80% sequence identity to Kuznetsova albicans microbial transglutaminase (SEQ ID NO: 6).

In one aspect, the kit further comprises one of a first substrate comprising an acyl-donor tag having at least 80% sequence identity to the peptide sequence YRYRQ (SEQ ID NO: 1) and a second substrate comprising an amine-donor tag having at least 80% sequence identity to the peptide sequence RYESK (SEQ ID NO: 2).

In another aspect, at least one of the first substrate and the second substrate comprises a detectable label.

In another aspect, the detectable label is selected from the group consisting of a biotin moiety, a fluorescent dye, a ruthenium label, a radioactive label, and a chemiluminescent label.

In another aspect, the acyl-donor tag has the peptide sequence APRYRQRAA (SEQ ID NO: 24).

In another aspect, the kit further comprises the other of the first substrate and the second substrate.

According to another aspect of the present invention, the enzyme for forming an isopeptide bond comprises a purified microbial transglutaminase having at least 80% sequence identity to Kuznetsova albicans microbial transglutaminase (SEQ ID NO: 6).

In one aspect, the isolated microbial transglutaminase is expressed and isolated in the presence of ammonium.

In another aspect, the ammonium is present at a concentration of at least about 10 μM.

According to another aspect of the invention, an acyl-donor substrate for a transglutaminase comprises an amino acid sequence having the formula _Xaa1 - _Xaa2 - _Xaa3 - _Xaa4 - _Xaa5 , wherein Xaa is any amino acid, wherein at least one of _Xaa3 , _Xaa4 and _Xaa5 is glutamine, wherein one of _Xaa4 and _Xaa5 is arginine, wherein the amino acid sequence comprises at least one arginine consecutively adjacent to glutamine, and wherein the total number of amino acids selected from the group consisting of arginine, glutamine, phenylalanine, tryptophan and tyrosine in the amino acid sequence is at least 4.

In one aspect, Xaa ₅ and at least one of Xaa ₁ , Xaa ₂ , and Xaa ₃ are arginine.

According to another aspect of the invention, an amine-donor substrate for a transglutaminase comprises an amino acid sequence having the formula _Xaa1 - _Xaa2 - _Xaa3 - _Xaa4 - _Xaa5 , wherein Xaa is any amino acid, wherein the amino acid sequence comprises at least one lysine, wherein one of _Xaa1 and _Xaa2 is selected from tyrosine and arginine, and wherein the total number of amino acids selected from arginine, serine, tyrosine and lysine in the amino acid sequence is at least 3.

In one aspect, one of Xaa ₄ and Xaa ₅ is lysine.

In another aspect, the amino acid sequence comprises no more than 2 amino acids lysine.

The foregoing and other aspects and advantages of the present invention will become apparent from the following description. In the description, reference is made to the accompanying drawings, which form a part hereof and in which preferred embodiments of the invention are shown by way of illustration. Such embodiments do not necessarily represent the entire scope of the invention, but, therefore, reference is made to the claims and this text for interpreting the scope of the invention.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram illustrating one embodiment of a method for identifying and characterizing transglutaminase species according to the present invention.

Figure 2A shows a multiple sequence alignment of the Kuznetsova albicans KALB_7456 hypothetical protein (upper row) and the Streptomyces mobaraensis microbial transglutaminase (lower row) using Clustal Omega version 1.2.1 (Sievers et al., 2011. Molecular Systems Biology 7:539). Identical amino acid residues are marked with an asterisk (*), and similar residues are marked with a colon (:). Conserved residues in the catalytic triad (Cys, Asp, His) of the S. mobaraensis microbial transglutaminase are highlighted in gray.

Figure 2B is the amino acid sequence of the putative transglutaminase KALB_7456 from K. albida, including the general cleavage site prediction determined by ProP 1.0 (Duckert et al., 2004. Protein Engineering, Design and Selection 17: 107-112), with the predicted signal peptide sequence ('s') and propeptide cleavage site ('P') indicated.

Figure 2C is a bar graph showing the propeptide cleavage potential of the putative transglutaminase KALB_7456 from Kuznetsova albicans as a function of amino acid sequence position, as determined by ProP 1.0. The dashed line indicates the propeptide cleavage potential threshold, and the dotted line indicates the predicted amino acid position of the signal peptide.

Figure 3A is an optical image of an SDS-PAGE gel showing the expression profile of Kuznetsova albicans transglutaminase (KalbTG) fusion proteins. The amino-terminal fusion partners are grouped in adjacent lanes, as indicated by lane pairs numbered 1-7 (1: 8X-His tag; 2: dsbA signal peptide; 3: ompT signal peptide; 4: Escherichia coli ( E. coli ) SlyD (EcSlyD); 5: 2xEcSlyD; 6: FkpA; 7: maltose binding protein). Lanes labeled 'L' are loaded with a standard molecular weight ladder (values shown in kDa). Lanes labeled 'P' and 'S' represent the insoluble (precipitate) and soluble (supernatant) fractions of E. coli cell lysate, respectively. The Coomassie blue-stained protein band representing His-KalbTG is marked with an asterisk (*) in lane pair 1.

Figure 3B is a schematic diagram of the expression and purification strategy of the 2xSlyD-fusion protein. N-terminal fusion with two parts of the sensitive-to-lysis D (SlyD) protein confers solubility and is cleavable by Factor Xa. Cleavage of the propeptide sequence by trypsin further matures the enzyme.

FIG3C is an image of an SDS-PAGE gel showing modular purification of KalbTG. Lanes labeled 'L' were loaded with a standard molecular weight ladder (values shown in kDa). Lane 1: Fractions containing 2xSlyD-KalbTG from the first Ni ⁺ -IMAC gradient elution (0-250 mM imidazole). Lane 2: Proenzyme of KalbTG purified by Ni ⁺ -IMAC gradient elution, on-column Factor Xa digestion, and size exclusion chromatography. Lane 3: Fractions from the second Ni ⁺ -IMAC gradient elution after sequential on-column digestion with Factor Xa and trypsin (0-250 mM imidazole). Lane 4: Concentrate of lane 3, filtered through a 50,000 molecular weight cutoff (MWCO) membrane.

Figure 4A is a log-log scatter plot showing the correlation between the fluorescence signal data generated by KalbTG against repeat elements on a 5-mer peptide array in the presence of a biotinylated amine-donor substrate. Each data point represents a pair of repeat peptides from a library of 1.4 million unique peptides synthesized in duplicate. The 22 data points showing the highest fluorescence signal are labeled with their respective 5-mer peptide sequences.

Figure 4B is a log-log scatter plot showing the correlation between the fluorescence signal data generated by KalbTG against repeat elements on a 5-mer peptide array in the presence of a biotinylated glutamine-donor substrate (Z-APRYRQRAAGGG-PEG-biotin). Each data point represents a pair of repeat peptides from a library of 1.4 million unique peptides synthesized in duplicate. The 17 data points showing the highest fluorescence signal are labeled with their respective 5-mer peptide sequences.

FIG5A is a log-log scatter plot showing the correlation between fluorescence signal data generated by Streptomyces mobaraensis MTG against repeat elements on a 5-mer peptide array in the presence of a biotinylated amine-donor substrate. Each data point represents a pair of repeat peptides from a library of 1.4 million unique peptides synthesized in duplicate. The 17 data points corresponding to the highest fluorescence signals generated by KalbTG from FIG4A are labeled with their respective 5-mer peptide sequences.

Figure 5B is a graph of the fluorescence signal data generated by the Streptomyces mobaraensis MTG of Figure 5A, with the labeled data points from Figure 4A omitted. The 16 data points corresponding to the highest fluorescence signals generated by MTG are labeled with their respective 5-mer peptide sequences.

Figure 5C is a graph of the fluorescence signal data generated by KalbTG of Figure 4A, with the labeled data points from Figure 4A omitted. The 16 data points corresponding to the highest fluorescence signals generated by S. mobaraensis MTG from Figure 5B are labeled with their respective 5-mer peptide sequences.

Figure 5D is a graph of S. mobara MTG and KalbTG activities obtained by measuring the NADH oxidation rate at 340 nm and 37°C for varying concentrations of the glutamine-donor substrates Z-GGGDYALQGGGG (0-1 mM) and Z-GGGYRYRQGGGG (0-1 mM) in the presence of the amine-donor substrate cadaverine (1 mM) in a GLDH-coupled assay.

Figure 6A is a series of SDS-PAGE gel images showing an experimental time course for Cy3 labeling of a Q-labeled Thermus thermophilus SlyD moiety (left: bright field; right: Cy3 fluorescence) and control data (left: bright field; right: Cy3 fluorescence). Successful single labeling was observed by a shift in electrophoretic mobility corresponding to the 6 kDa molecular weight of the label and by the fluorescent signal of the labeled species. Data are shown for a 60-minute time course performed with a 10-fold labeling excess and an 18-hour incubation performed with a 50-fold labeling excess. Lanes labeled 'L' were loaded with a standard molecular weight ladder (values shown in kDa). Lanes labeled '-' and '+' represent control reactions with SlyD containing either the Streptomyces mobaraensis MTG Q-tag (DYALQ (SEQ ID NO: 22)) and either KalbTG (-) or Streptomyces mobaraensis MTG (+).

FIG6B is a series of SDS-PAGE gel images showing the pH profile of KalbTG labeling efficiency for pH values between 6.2 and 9.0 (left: bright field; right: Cy3 fluorescence). The highest labeling yield was observed after 15 minutes of reaction time at pH 7.4. A standard molecular weight ladder was loaded into the lane labeled 'L' (values shown in kDa).

Figure 6C is a series of SDS-PAGE gel images showing dual site-specific functionalization of the construct YRYRQ-PEG27-(Factor Xa cleavage site)-PEG27-PEG27-DYALQ with Cy3 and Cy5 fluorescent labels. Each group (labeled 1, 2, and 3) includes three images of the same gel lane arranged from left to right in the following order: i) bright field; ii) Cy3 fluorescence; and iii) Cy5 fluorescence. Group 1: mixture of peptide construct and 10-fold excess Cy3 label. Group 2: mixture of peptide construct and 10-fold excess Cy3 label after incubation with KalbTG enzyme for 30 minutes. Group 3: Streptomyces mobaraensis MTG enzyme and Cy5 label were added to the combination of Group 2 and incubated for 15 minutes without intermediate blocking or purification steps; the dual-labeled construct was obtained in near-quantitative yield.

Figure 7A is a three-dimensional alignment of the active enzyme structures of KalbTG (dark gray) and S. mobaraensis MTG (light gray), which reveals high conservation in the core and active site regions and high variability in the surrounding loop regions.

Figure 7B is a three-dimensional surface overlay of the active enzyme structures of KalbTG (dark grey) and S. mobara MTG (light grey), illustrating that the binding grooves of S. mobara MTG and the more compact KalbTG can be similarly occupied by the propeptide (ribbon structure).

The three-dimensional ribbon structure of Figure 7C illustrates the contributing factors to the formation of the KalbTG active cleft, which includes two highly charged hydrophilic loops that are thought to mediate substrate recruitment, act as substrate mimics, or a combination thereof. The two hydrophilic loops are labeled with their corresponding sequences (i.e., NHEEPR (SEQ ID NO: 3) and YRYRAR (SEQ ID NO: 4)).

Detailed Description of the Invention

As discussed above, in different situations, it may be useful to illustrate the basic understanding of the enzymes provided by the details of enzymatic activity and specificity, and for developing biotechnology applications comprising those enzymes. For example, conventional chemical strategies for modifying therapeutic and diagnostic proteins often lack site-specificity, connection stability, stoichiometric control or a combination thereof, thereby producing heterogeneous conjugates that may cause interference (for example, immunoreactivity or stability of an interfering therapeutic agent). In one aspect, it is anticipated that future industrial development of treatment and diagnostic reagents will see a large increase in the complex forms of stable and true site-specific conjugations. Therefore, it is desirable to provide attractive and cost-effective alternatives to the chemical strategies established.

Microbial transglutaminase (MTG) is a protein-glutamine γ-glutamyltransferase (EC 2.3.2.13) first described in 1989 by researchers at Ajinomoto Co., Inc., and is one of the most widely used enzymes for cross-linking proteins and peptides in many food and biotechnology applications. MTG was first discovered in the organism Streptomyces mobaraensis and later isolated from it. MTG catalyzes the formation of a stable isopeptide bond between an acyl group (e.g., glutamine side chain; acyl-donor) and an alkyl-amine (e.g., lysine side chain; amine-donor). In the absence of a reactive amine group, the enzymatic reaction with water results in the deamination of the glutamine side chain. This bacterial enzyme functions in the absence of added cofactors (such as Ca2 ⁺ or GTP) and over a wide range of pH, buffer, and temperature conditions.

In contrast to sortase A, whose natural substrate specificity is very stringent, known MTG (e.g., from Streptomyces mobara) is generally an indiscriminate enzyme with respect to substrate molecules, and the specificity of transglutaminase variants remains largely unknown. Although significant scientific efforts have been made to establish MTG as a candidate enzyme in the development of therapeutic antibody-drug conjugates, large-scale production of such MTG-mediated immunoconjugates has been hampered by the enzyme's low specificity.

Known MTG species are primarily representatives of the Streptomyces or Bacillus families. These MTG species exhibit very similar basic amino acid structures and substrate specificities. All known active MTG species exhibit a molecular weight of at least approximately 38 kDa. Being cross-linking enzymes in nature, known MTGs generally exhibit broad substrate specificity for amine-donating substrates and relatively low specificity for acyl-donating substrates. High-throughput screening for improved MTG substrates has previously been limited to phage panning or mRNA display. While a recent leading array-based high-throughput screening approach has successfully identified substrates for Streptomyces mobaraensis MTG (U.S. Provisional Patent Application Serial No. 62/094,495 to Albert et al., filed December 19, 2014), only the substrate specificity of this enzyme and homologous enzymes is known, thereby precluding any bioorthogonal conjugation approaches (e.g., simultaneous labeling of biomolecules using two or more different label-substrates and two or more transglutaminases). Therefore, there is a need for high-throughput protocols for identifying and characterizing new transglutaminases. Furthermore, there is a need for improved transglutaminases with greater activity, specificity, or a combination thereof. Furthermore, there is a need for acyl-donor tags (e.g., glutamine- or Q-tags) and amine-donor tags (e.g., lysine- or K-tags) that are specific and unique substrates for the transglutaminases of interest.

Systems and methods for identifying and characterizing transglutaminases can overcome these and other challenges. To this end, the present invention provides structural and biochemical characterizations of known and unknown candidate transglutaminases. The present invention further provides characterizations of recombinant production of candidate transglutaminases, high-throughput screening of potential transglutaminases substrates via high-density peptide arrays, and semi-orthogonal conjugation of biomolecules using newly characterized transglutaminases. In another aspect, the present invention provides acyl-donor tags (e.g., glutamine- or Q-tags) and amine-donor tags (e.g., lysine- or K-tags) that are specific and unique substrates for target transglutaminases. As used herein, the term 'tag' refers to a sequence comprising one or more amino acids or other similar molecules that can be grafted, fused, conjugated, or otherwise attached to another structure, such as a protein, peptide, small molecule, detectable label (e.g., fluorescent dye), oligonucleotide, non-amino or nucleic acid polymer (e.g., polyethylene glycol), etc. In one aspect, the nature of the attachment should allow access of the tag to the enzyme, wherein the tag is a substrate for the enzyme.

In one embodiment of the present invention, a previously unknown transglutaminase species from the organism Kuznetsova albus was identified, recombinantly expressed, purified, and characterized using a high-throughput array-based screening protocol. The Kuznetsova albus transglutaminase was determined to exhibit high selectivity and substrate specificity for its array-determined substrate sequence, but reacted only poorly or not at all with the substrate sequence of the Streptomyces mobara enzyme. Thus, it can be said that the Kuznetsova albus transglutaminase is bioorthogonal to the Streptomyces mobara enzyme. In another aspect, the Kuznetsova albus transglutaminase exhibits a surprisingly lower molecular weight (approximately 30 kDa) than all previously described MTG species (e.g., Streptomyces mobara MTG is approximately 38 kDa), thereby foreshadowing an advantage for production and enzymatic labeling purposes. Furthermore, the Kuznetsova albus transglutaminase has a significantly different basic amino acid structure compared to all currently known proteins. Taken together, these properties make Kuznetsova transglutaminase very attractive for a wide range of applications, including, but not limited to, the versatile, cost-effective, and site-specific conjugation of biomolecules to a variety of labeling molecules. Additional or alternative applications in which Kuznetsova transglutaminase may be effective include the production of therapeutic antibody-drug conjugates or chemiluminescent antibodies for in vitro diagnostic use.

In one aspect, many challenges arising from the recombinant production of transglutaminases can be overcome by implementing embodiments of the present invention. Referring to FIG1 , a method 100 for identifying and characterizing transglutaminases includes a step 102 of identifying candidate transglutaminases. Step 102 can include searching for homologs of known or suspected transglutaminases to identify candidate transglutaminases for further investigation. In an exemplary embodiment, the transglutaminases can be microbial transglutaminases (e.g., Streptoverticillium sp . transglutaminases, Kutzneria sp. transglutaminases, Streptomyces sp . transglutaminases, etc.) or mammalian transglutaminases. In embodiments wherein the enzyme is a mammalian transglutaminase, the mammalian transglutaminase can, for example, be selected from the group consisting of: human factor XIII A transglutaminase, human factor XIII B transglutaminase, factor XIII transglutaminase, keratinocyte transglutaminase, tissue-type transglutaminase, epidermal transglutaminase, prostate transglutaminase, neuronal transglutaminase, human transglutaminase 5, and human transglutaminase 7.

The search for known or suspected homologs of candidate transglutaminases can include the use of one or more search tools or databases. One suitable tool includes the Protein Basic Local Alignment Search Tool (Protein BLAST) from the National Center for Biotechnology Information (NCBI). The Protein BLAST tool can be provided with sequences of known or suspected transglutaminases, which can be obtained from various databases. An example database is the Universal Protein Catalog (UniProt). However, other databases can be used in addition to, or as an alternative to, UniProt. In one aspect, when using Protein BLAST, a threshold Expect-value (E-value) can be selected to narrow the search results. In one embodiment, it can be used to select an E-value of less than about 10^{^-8} . In another embodiment, it can be used to select an E-value of less than about 10^{^-10} . In another embodiment, it can be used to select an E-value of less than about 10^{^-12} .

Step 102 can also include performing a sequence alignment of the transglutaminase sequences using an alignment tool. An example alignment tool includes the Clustal Omega 1.2.1 tool (Sievers et al. 2011, Molecular Systems Biology 7: 539). The alignment tool can provide a percent identity matrix value, identify potential conservation of catalytically active residues (if known), or a combination thereof. Other tools that can be used in step 102 include the ProP 1.0 Server from the Technical University of Denmark (Duckert et al. 2004, Protein Engineering, Design & Selection 17(1): 107-112) to predict propeptides and signal sequences of transglutaminases. In one aspect, the candidate transglutaminase can have at least about 20%, at least about 25%, at least about 30%, at least about 35%, or more similarity to a known transglutaminase. Furthermore, candidate transglutaminases can be characterized by conservation of at least one or more active site residues with respect to known or suspected transglutaminases, indicating that enzymatic structure and function may be retained.

The information collected in step 102 using one or more of the aforementioned tools can be used to select candidate transglutaminases from predicted or known transglutaminases sequences for expression and purification in step 104. In one aspect, step 104 can include rapidly screening for expression conditions of candidate transglutaminases. One suitable method for screening includes inserting a genetic insert of a candidate transglutaminase into one or more expression vectors designed for soluble cytosolic or periplasmic expression in a host organism using a fragment exchange system (Geertsma, et al. 2011. Biochemistry 50(15): 3272-3278). Other screening methods can also be used.

Step 104 can also include a preliminary screen to identify evidence of expression of the full-length protein with the expected electrophoretic mobility. In the event that poor expression or no expression of the full-length active candidate transglutaminase protein is observed, the candidate transglutaminase sequence can be fused to a chaperone protein (e.g., SlyD) to increase the likelihood of expression of the functional enzyme. The expression of the candidate transglutaminase can be further optimized by screening different incubation times, incubation temperatures, inducer concentrations, induction times, media types, media volumes, etc., and combinations thereof.

In general, it should be understood that a variety of feasible purification and expression strategies can be employed in step 104 of method 100. In one embodiment, the candidate transglutaminase sequence can be incorporated into a modular expression construct. Example expression constructs for use in step 104 can include chaperone modules, protease cleavage site modules, purification tag modules, detection modules, and the like, and combinations thereof. For example, an expression construct can include one or more SlyD chaperone modules (arranged to produce a transglutaminase-chaperone fusion protein), one or more protease cleavage site modules (flanked by transglutaminase sequences for separation of the modules after expression), and one or more 8X-histidine tags or other purification modules (for recovery of one or more segments of the expressed protein). For expression constructs that include protease cleavage site modules, the expressed protein can be treated with one or more proteases to produce an activated protein. For example, the expressed protein can be treated with Factor Xa protease, trypsin protease, thrombin protease, or another similar protease to cleave any chaperone protein, propeptide sequence, purification tag, and the like from the expressed candidate transglutaminase.

To produce a selected transglutaminase protein, the gene sequence encoding the candidate transglutaminase (including some, zero, or all of the predicted signal and propeptide sequences) can be codon-optimized and chemically synthesized for expression in a particular host organism (e.g., E. coli). Expression can be performed according to standard molecular biology protocols as described, for example, in Green, et al., 2012, Molecular Cloning: A Laboratory Manual , Cold Spring Harbor Laboratory Press.

In the next step 106 of method 100, the candidate transglutaminases expressed and purified in step 104 are screened against a substrate library to identify potential substrates, including acyl-donor sequences, amine-donor sequences, or both. In one aspect, the substrate library can include a plurality of peptide components synthesized in an array by maskless array synthesis (U.S. Patent Publication No. 2015/0185216 to Albert et al., filed December 19, 2014). The peptide components can be prepared from natural amino acids, non-natural amino acids, other molecular building blocks, and the like, and combinations thereof. In addition, the peptides can be in linear, cyclic, or constrained (macrocyclic) form.

As used herein, the terms "peptide," "oligopeptide," or "peptide binder" refer to organic compounds composed of amino acids, which can be arranged in a linear chain (linked together by peptide bonds between the carboxyl and amino groups of adjacent amino acid residues), in a cyclic form, or in a constrained form (e.g., a "macrocycle" form). The term "peptide" or "oligopeptide" also refers to shorter polypeptides, i.e., organic compounds composed of less than 50 amino acid residues. Macrocycle (or constrained peptide) is used herein in its conventional sense to describe cyclic small molecules such as peptides of about 500 to about 2,000 daltons.

The term "natural amino acid" refers to one of the 20 amino acids commonly found in proteins and used in protein biosynthesis, as well as other amino acids (including pyrrolysine and selenocysteine) that can be incorporated into proteins during translation. The 20 natural amino acids include histidine, alanine, valine, glycine, leucine, isoleucine, aspartic acid, glutamic acid, serine, glutamine, asparagine, threonine, arginine, proline, phenylalanine, tyrosine, tryptophan, cysteine, methionine, and lysine.

The term "unnatural amino acid" refers to an organic compound that is not encoded by the standard genetic code or is not incorporated into proteins during translation. Thus, unnatural amino acids include, but are not limited to, amino acids or analogs of amino acids, including, but not limited to, D-isostereomers of amino acids, β-amino-analogs of amino acids, homocitrulline, homoarginine, hydroxyproline, homoproline, ornithine, 4-amino-phenylalanine, cyclohexylalanine, α-aminoisobutyric acid, N-methyl-alanine, N-methyl-glycine, norleucine, N-methyl-glutamate, tert-butylglycine, α-aminobutyric acid, tert-butylalanine, 2-aminoisobutyric acid, α-aminoisobutyric acid, 2-aminoindan-2-carboxylic acid, selenomethionine, dehydroalanine, lanthionine, γ-aminobutyric acid, and derivatives thereof, in which the amine nitrogen has been mono- or di-alkylated.

Continuing with reference to FIG1 , step 108 of method 100 includes identifying the optimal amine-donor substrate sequence, acyl-donor substrate sequence, or both from a substrate library. In one aspect, the activity of a candidate transglutaminase on a library substrate can be measured using one or more direct or coupled assays. Generally speaking, direct assays involve measuring reactants (e.g., substrates, cofactors, etc.) and products (e.g., isopeptide bonds, deamidated substrates, etc.) of an enzyme reaction. For example, spectrophotometry can be used to track the progress of an enzyme reaction by measuring the change in absorbance associated with a reactant or product over time. In cases where an enzyme reaction is not amenable to one or more direct assays, or in addition to or as an alternative to direct assays, coupled assays can be employed. In the case of coupled assays, the product of the target enzyme reaction can serve as a substrate for another, more easily measured, secondary reaction. Examples of coupled assays include measurements of redox reactions involving cofactors such as NADP(H) and NAD(H), which participate in the target enzyme reaction as either a product or a reactant. In the case of reactions catalyzed by transglutaminase, oxidative coupling assays relying on glutamate dehydrogenase (GLDH) can be implemented. An example of a GLDH assay includes β-casein as a cross-linking substrate and detects deamidation via GLDH oxidation of NADPH. Notably, it can be used to select assays compatible with high-throughput screening formats to allow the study of large numbers of substrates (e.g., greater than 1 million) in parallel.

Using one of the aforementioned direct or indirect assays, step 108 can include using a peptide substrate array to identify a specific sequence or motif recognized by the candidate transglutaminase. For example, the transamidation reaction between millions of unique peptides and a biotinylated amine donor can be quantified in parallel on one or more arrays, and the sequence of the peptide with the highest signal output (i.e., the best substrate) can be determined. Thereafter, in the next step 110 of method 100, the best substrate can be resynthesized in a separate (on the array or in solution) assay, scanned, and tested for transglutaminase activity. Thus, step 110 includes characterizing the candidate transglutaminase in the presence of the best substrate identified in step 108.

Characterization of a candidate transglutaminase can include determining parameters such as specificity, selectivity, affinity, activity, etc. In addition, two or more candidate transglutaminases can be characterized with respect to orthogonality. Here, the ability of the candidate transglutaminases to act on the same substrate can be identified. In the case where two different transglutaminases cannot act on the same substrate, the two different transglutaminases can be said to be orthogonal. However, in the case where two different transglutaminases can act on one or more of the same substrates, but have different degrees of activity, the two different transglutaminases can be said to be semi-orthogonal. In one aspect, a peptide substrate array can deliver readouts of all viable 5-mer peptide sequences at once. Therefore, the data collected from the substrate library can be used to identify differences in substrate specificity of each candidate transglutaminase, and to identify orthogonal, semi-orthogonal, and non-orthogonal transglutaminases.

Step 110 of method 100 can also include characterizing the ability of the candidate transglutaminase to perform site-specific labeling on a protein substrate. In one aspect assay, the optimal substrate sequence identified in step 108 can be further analyzed on an array and in solution in step 110. Experiments can also be performed to quantify the cross-reactivity of the candidate transglutaminase with different substrates. In one aspect, a protein scaffold can be used for labeling schemes because loops containing epitopes (i.e., substrate sequences) can be grafted onto the scaffold for presentation to a binding agent or enzyme. One scheme including the use of a scaffold is described in PCT Application Publication No. WO 2012/150321 to Andres et al., filed May 4, 2012. An example scaffold can advantageously include one or more FK506-binding protein (FKBP) domains as sites for grafting epitope-containing loops.

Labeling of the scaffold protein by a candidate transglutaminase can be achieved under a variety of conditions. Factors that can be varied for labeling experiments include the ratio of substrate to transglutaminase, the ratio of one substrate to another substrate, labeling time, pH, etc. It is noteworthy that substrate represents any peptide, protein or other structure, including one or more amine-donor or acyl-donor substrate sequences. Example substrates include scaffold proteins having acyl-donor or amine-donor substrate sequences grafted thereon, detectable labels conjugated to or otherwise bound to acyl-donor or amine-donor substrate sequences, isolated acyl-donor or amine-donor substrate sequences, etc., and combinations thereof. Using standard techniques, such as sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) combined with optical (e.g., bright field, fluorescence) detection, the labeling yield can be measured over time. For example, the first substrate can include one or more detectable labels. By identifying molecular weight migration on an SDS-PAGE gel and then detecting the label within the gel, the cross-linking of the first substrate with the second substrate (e.g., protein scaffold) can be analyzed.

Labels for use with embodiments of the present invention include any suitable label that can be combined with a transglutaminase substrate. Examples of suitable labels include fluorescent labels, chemiluminescent labels, radioactive labels, chemical labels (e.g., comprising "click" chemistry), haptens, toxins, etc. and combinations thereof. More generally, suitable labels are compatible with at least one substrate for transglutaminase (e.g., an acyl-donor substrate or an amine-donor substrate) in that the label does not eliminate the ability of transglutaminase to act on the labeled substrate. In addition, suitable labels can produce a detectable signal relative to an unlabeled transglutaminase substrate. Specific examples of detectable labels for use in any suitable embodiment herein can include fluorescein, rhodamine, Texas Red, phycoerythrin, Oregon Green (e.g., Oregon Green 488, Oregon Green 514, etc.), AlexaFluor 488, AlexaFluor 647 (Molecular Probes, Eugene, Oregon), Cy3, Cy5, Cy7, biotin, ruthenium, DyLight fluorescer (including, but not limited to, DyLight 680), CW 800, trans-cyclooctene, tetrazine, methyltetrazine, etc. Examples of haptens include biotin, digoxigenin, dinitrophenyl, etc. Examples of toxins include amatoxins (e.g., amanitin), maitansinoids, etc.

In certain embodiments, step 110 may also include identifying and characterizing a three-dimensional (3D) crystal structure of the transglutaminase to provide further insight into the properties of the transglutaminase. The crystal structure of the transglutaminase may be performed in the presence or absence of one or more substrates, cofactors, etc. Analysis of the crystal structure may provide insight into possible locations for site-specific mutagenesis to improve the performance of the transglutaminase. Crystallization of candidate transglutaminases with specific substrate sequences may further reveal interactions between the transglutaminase and substrate sequences to inform modifications to either or both of the transglutaminase and substrate sequences to tailor the performance of the transglutaminase for a specific application. In addition, analysis of the crystal structure may serve as an independent confirmation of the reliability of array-based substrate discovery (see Example 5).

In step 112 of method 100, the substrate sequence identified in step 108 and characterized in step 110 is selected for use in a downstream application of the selected candidate transglutaminase. In general, a particular acyl-donor or amine-donor substrate sequence can be unique to a given transglutaminase. Thus, for a given application, it can be used to first select a transglutaminase and then select one or more substrate sequences. For applications where specificity and selectivity are important (e.g., orthogonal labeling using two or more transglutaminases), step 112 can include selecting a substrate sequence that is specifically and selectively labeled with the selected transglutaminase. However, other applications may benefit from selecting a substrate sequence that can be acted upon by more than one transglutaminase. Substrates can also be selected to achieve a specific degree of transglutaminase activity, in which case it can be used to achieve faster or slower reaction times. To further customize the selected substrate sequence, after step 112, method 100 can return to step 106 for additional rounds of screening. In this case, the selected substrates can be extended, matured, etc. in subsequent rounds of screening on the peptide array. Example methods for extension and maturation of peptide sequences are described in U.S. Patent Publication No. 2015/0185216, filed December 19, 2014, to Albert et al.

In other embodiments, it can be used to provide site-specific labeling with non-selective transglutaminase activity. Non-selective transglutaminases may be useful if the substrate is not produced recombinantly, if the labeling sites and labeling ratios can be controlled, or if the labeling sites and labeling ratios are not critical for the current application, etc. An example of labeling with a non-selective transglutaminase is the conjugation of a payload to a deglycosylated or glycosylated IgG. However, transglutaminases with non-specific activity may be limited to a narrow range of possible applications. Therefore, in other cases, it may be useful to provide a transglutaminase that has specific activity for only one particular substrate or a collection of similar substrates.

In summary, method 100 can be used to identify and characterize one or more candidate transglutaminases together with one or more corresponding substrates. Putative or known transglutaminases can be expressed and screened on a substrate library to identify preliminary substrate sequences that elicit the desired transglutaminase activity. The optimal substrate can then be selected and optionally refined in an iterative manner, thereby generating transglutaminases-substrates combinations that can be implemented for a variety of applications.

Example

Example 1: Identification of Kuznetsova albicans microbial transglutaminase

The need for a viable, robust, enzymatic, industrial-scale method for site-specific conjugation (e.g., antibody-drug conjugates) creates a high demand for conjugating enzymes. Among other factors, these enzymes are suited for high reaction rates, conjugation efficiency, and substrate specificity. Furthermore, they are suited for production-economical enzymes, including enzymes with low molecular weights, enzymes that are cofactor-independent, and combinations thereof. Regarding the discovery of new microbial transglutaminases, a search was conducted for homologs of this enzyme that potentially met all of the aforementioned criteria, using the amino acid sequence of the Streptomyces mobara protein-glutamine γ-glutamyltransferase as a query. This yielded the putative gene product KALB_7456 from the bacterium Kuznetsova albus DSM 43870, a spore-forming, Gram-positive bacterium sequenced in 2014 (Rebets et al., 2014. BMC Genomics 15: 885).

Use the NCBI protein BLAST tool web interface to search for sequences similar to MTG from Streptomyces mobaraensis. Enter the amino acid sequence of Streptomyces mobaraensis protein-glutamine gamma-glutamyltransferase (UniProt accession number P81453) as the query. The amino acid sequence of Streptomyces mobaraensis protein-glutamine gamma-glutamyltransferase is as follows:

Manual screening for peptide sequences with E-values less than 10 ^{^-10} and shorter than Streptomyces mobaraensis MTG yielded the putative gene product KALB_7456 (GenBank accession number AHI00814.1; UniProt accession number W5WHY8) from the bacterial strain Kuznetsova albicans DSM 43870. The amino acid sequence of the gene product KALB_7456 is as follows:

Sequence alignment of the S. mobaraensis and K. albicans sequences was performed using Clustal Omega 1.2.1, resulting in a 32% identity matrix value and identifying the conservation of catalytically active residues from S. mobaraensis MTG (C140, R331, and H350, numbered based on P81453 (SEQ ID NO: 5)). The propeptide and signal sequence of the putative K. albicans microbial transglutaminase were predicted using the ProP 1.0 Server from the Technical University of Denmark. The only predicted propeptide cleavage site was VAAPTPR/AP, which had a score exceeding the threshold (0.513), where the slash mark between the amino acids R and A indicates the predicted cleavage site.

Comparison of the primary structures of the gene products of Streptomyces mobaraensis (SEQ ID NO: 5) and Kuznetsova albus (SEQ ID NO: 6) revealed 30% similarity, with varying conservation of active site residues ( FIG. 2A ), indicating that enzyme structure and function may be preserved. The entire Kuznetsova albus gene product is significantly smaller than that of Streptomyces mobaraensis MTG, which totals a calculated molecular weight of 30.1 kDa. Since Streptomyces mobaraensis MTG is produced as an inactive zymogen and processed by extracellular proteases to produce a 38 kDa active form, a similar activation mechanism was predicted for the putative Kuznetsova albus MTG, and the signal and propeptide sequences in the N-terminal region of the protein were analyzed using the ProP 1.0 server ( FIG. 2B and 2C ). The sequence VAAPTPR/AP is the only predicted propeptide cleavage site, where cleavage occurs between the amino acids arginine and proline, as indicated by the slash. The sequence VAAPTPR/AP corresponds to the dispase site SAGPSFR/AP in Streptomyces mobara MTG, but is presumed to lack dispase reactivity, as phenylalanine is an essential residue in the enzyme recognition motif. Furthermore, a signal peptide with a high-probability cleavage site, GLPTLIA/TT, was predicted using the ProP 1.0 server. However, the predicted signal peptide cleavage site did not bear sequence similarity to the significantly longer Streptomyces mobara MTG presequence or other known signal peptides. Based on the predicted signal peptide and propeptide cleavage sites, the molecular weights of the mature Kuznetsova transglutaminase, proenzyme, and preproenzyme were calculated to be 26.4 kDa, 27.7 kDa, and 30.1 kDa, respectively.

Example 2: Evaluation of parallel constructs for recombinant production of KalbTG

To rapidly screen expression conditions for a putative Kuznetsova albicans transglutaminase (KalbTG), we used a fragment exchange system to insert synthetic genetic inserts into various expression vectors designed for soluble cytosolic or periplasmic expression in Escherichia coli (Geertsma, et al. 2011. Biochemistry 50(15): 3272-3278) .

Initial screening at a 5 ml scale clearly demonstrated that a protein with the expected electrophoretic mobility of the full-length KalbTG fusion was expressed, and that fusion with the tandem SlyD chaperone (Scholz, et al. 2005. Journal of Molecular Biology 345(5): 1229-1241) produced the highest amount of soluble protein of all the constructs tested ( FIG3A ). Expression of this construct was further optimized by screening different incubation times and temperatures, isopropyl β-D-1-thiogalactopyranoside (IPTG) inducer concentrations and induction times, media types, and volumes. Referring to FIG3B and 3C , the modular nature of the selected fusion constructs provided the SlyD fusion protein, proenzyme, and activated enzyme through a combination of sequential purification and proteolytic cleavage steps. Starting from the N-terminus, the expression construct 200 includes two consecutive SlyD chaperones 202, a Factor Xa protease cleavage site ( _C1 ) KalbTG propeptide 204, a trypsin protease cleavage site ( _C2 ) KalbTG enzyme 206, and an 8X-histidine tag 208. The SlyD chaperones 202 are cleaved from the expression construct 200 by Factor Xa protease 210. In addition, the propeptide 204 is cleaved from the KalbTG enzyme 206 by trypsin protease 212, thereby generating the activated form of the KalbTG construct 200. The purified and activated enzyme remains stable at 4°C and through multiple freeze-thaw cycles. The melting point of the purified and activated enzyme was determined to be 48.9°C using differential scanning calorimetry (DSC). It should be understood that the method described herein represents one of many possible purification strategies. In addition, the parallel cloning scheme described enables the re-evaluation of different constructs and the production process on a laboratory scale in an efficient and economical manner.

For the production of KalbTG, the gene sequence encoding the putative Kuznetsova albicans microbial transglutaminase (including the predicted signal and propeptide ( KalbTGpp ), including only the predicted propeptide ( kalbTGt3 ), excluding the predicted signal and propeptide ( kalbTGt1 ), or excluding the predicted signal and with an additional Factor Xa cleavage site inserted after the propeptide ( kalbTGt2 )) was codon-optimized for E. coli expression (Roche Sequence Analysis Web interface), chemically synthesized (GeneArt, ThermoFisher, Regensburg) and cloned via fragment exchange (Fx) cloning (Geertsma, et al. 2011, Biochemistry 50(15): 3272-3278) into a vector providing the cleavage-sensitive D chaperone protein (SlyD, UniProt entry P0A9K9, truncated after Asp165 (Scholz et al. 2005, Journal Of Molecular Biology 345(5): 1229-1241), followed by a protease factor Xa cleavage site and a C-terminal 8X-His tag. The amino acid sequence of the truncated SlyD after Asp165 is as follows:

The carrier is based on the pQE-80 series of Qiagen, comprises the IPTG-inducible protein expression of T5 promoter and provides resistance to ampicillin.The expression construct for all experiments described in this work is called EcSlyD2-Xa-KalbTGt3-8xHis.In addition, as initial expression screening, in carrier, carry out fragment exchange cloning, it gives and 8X-His tag, dsbA and ompT signal peptide, single SlyD or FkpA chaperone protein part and maltose binding protein (MBP) N-terminal fusion.According to standard molecular biology protocol (Green, et al., 2012. " Molecular Cloning: A Laboratory Manual ", Cold Spring Harbor Laboratory Press), carry out plasmid preparation and expression plasmid to the transformation of the intestinal bacteria Bl21 Tuner cells of chemical competence.

To prepare active KalbTG enzyme, 0.4 to 1 liter of Terrific Broth (TB) medium was inoculated with an overnight culture of EcSlyD2-Xa-KalbTGt3-8xHis-Tag in E. coli Bl21 Tuner at a ratio of 1:50. The amino acid sequence of the EcSlyD2-Xa-KalbTGt3-8xHis-Tag expression construct is as follows:

Cells were incubated in baffled shake flasks at 37°C, 180 rpm, and protein expression was induced with 1 mM IPTG after the cell density reached an OD _{600 nm} of 0.8-1.2. Cells were harvested by centrifugation (7878 x g for 30 min at 4°C). The supernatant was discarded, and the cell pellet was either stored at -80°C or immediately processed for nickel-immobilized metal affinity chromatography (Ni ⁺ -IMAC).

For subsequent Ni ⁺ -IMAC purification of EcSlyD2-Xa-KalbTGt3-8xHis, the cell pellet was resuspended in 30-50 ml of phosphate-buffered saline (PBS) in the presence of lysozyme and DNase I. The cells were disrupted by high-pressure homogenization at 2 kbar. To remove cell debris, the suspension was centrifuged (17,210 x g at 4°C for 30 min).

The supernatant, free of cell debris, was filtered through a 0.45 µm polyethersulfone (PES) membrane and loaded onto a 5 ml HisTrap column. The column was washed with at least 5 column volumes of PBS, and the His-tagged protein was eluted using a linear gradient of 0-250 mM imidazole in PBS (30 ml, 5 ml min ⁻¹ ). 3 ml fractions containing protein, identified by Abs _{280 nm} , were collected, diluted in PBS, and concentrated using an Amicon Ultra concentrator (10,000 MWCO; 5,000 xg for 15-30 min). The protein concentration of the fractions was determined using a Bradford assay (BioRad, according to the manufacturer's instructions). 5-10 µg of protein per sample was analyzed by SDS-PAGE (ThermoFisher Novex, according to the manufacturer's instructions). The purified protein was aliquoted into 200 µl volumes, frozen by brief incubation in liquid nitrogen, and stored at −80°C.

To cleave the SlyD chaperone and propeptide from the EcSlyD2-Xa-KalbTGt3-8xHis-tag, the protein was immobilized on a 5 ml His Trap column and digested on-column with Factor Xa followed by trypsin. 1 microgram of Factor Xa was applied per 50 µg of total protein and incubated on the column for 1.5 hours. The protease and cleaved SlyD were washed from the column with PBS.

The amino acid sequence of the EcSlyD2-Xa-KalbTGt3-8xHis-tag expression construct after Factor Xa digestion is as follows:

After trypsin digestion, the amino acid sequence of the KalbTG enzyme (KalbTGt3) is as follows:

For crystal structure analysis of KalbTG, after Factor Xa digestion, the His-tagged protein was eluted using a 0-250 mM linear imidazole gradient and a polishing step was performed using size exclusion chromatography (GE Superdex 200 μg 16/60, PBS). Alternatively, to obtain an active and pure enzyme preparation, activation was performed as follows: 200 μg ^ml trypsin was added to a His Trap column and incubated for 15-30 min. The protease and cleaved propeptide were washed from the column with PBS, and the digested KalbTG was collected in the same manner as above. To eliminate high-molecular-weight impurities from the active KalbTG, the enzyme preparation was filtered through an Amicon Ultra concentrator (50,000 MWCO). The filtrate was tested for activity using a GLDH-coupled assay and analyzed for purity by SDS-PAGE, as shown in Figure 3C. The remaining filtrate was divided into 200 μl aliquots, frozen in liquid nitrogen, and stored at -80°C.

In another embodiment of the protocol for producing active KalbTG enzyme, E. coli BL21 Tuner carrying the plasmid pQE-EcSlyD2-Xa-KalbTGt1-8H (ColE1 origin; _{IPTG-inducible T5 promoter) was inoculated into 10 L of standard E. coli fermentation medium similar to TerrificBroth (yeast extract, K2HPO4, NH4Cl , glycerol} , antifoam, _MgSO4.7H2O , _H3PO4 , NaOH) and containing an additional 1 g _{of NH4Cl} per liter. The sequence of the EcSlyD2 _- Xa-KalbTGt1-8xHis construct is as follows:

The EcSlyD2-Xa-KalbTGt1-8xHis construct has a similar (but not identical) modular composition to the construct shown in FIG3B , with one notable difference being that the EcSlyD2-Xa-KalbTGt1-8xHis construct omits the KalbTG propeptide 204 and the trypsin protease cleavage site (C ₂ ).

Fermentation was carried out at 35°C for 26 h until an _OD600 of 44 _was reached. The cells were harvested and resuspended in a buffer containing 50 mM Tris-HCl pH 8.0, 1 mM EDTA, 1 mM DTT, and 10 mM ( _NH4 ) _2SO4 . The cells were disrupted by a high-pressure homogenizer at 800 bar. The resulting cell extract was pretreated with 1-3% Polymin-G20 and then loaded onto a Q-Sepharose XL column (strong anion exchange matrix; GE Healthcare LifeSciences) at a protein concentration of approximately 30 mg/ml. The bound protein was washed with 20 mM Tris-HCl pH 8.0, 1 mM EDTA, 1 mM DTT, 10 mM ( _NH4 ) _{2SO4 ,} and 150 mM NaCl, and then eluted with a 150-500 mM NaCl gradient over 30 column volumes. The eluate was dialyzed in 20 mM Tris-HCl pH 8.0, 0.1 mM EDTA, 0.1 mM DTT, 10 mM (NH ₄ ) ₂ SO ₄ , 500 mM NaCl (10 kDa molecular weight cutoff), concentrated, and loaded onto a Ni-NTA column. The bound His-tagged protein was washed with 20 mMTris-HCl pH 8.0, 0.1 mM EDTA, 0.1 mM DTT, 10 mM (NH ₄ ) ₂ SO ₄ , 500 mM NaCl, 25 mM imidazole and eluted with a 20 column volume 25-200 mM imidazole gradient. The purified protein was dialyzed (10 kDa molecular weight cutoff) against 20 mM Tris-HCl, 1 mM EDTA, 1 mM DTT, and 10 mM (NH ₄ ) ₂ SO ₄ (pH 8.0), concentrated to 1.77 mg/ml, analyzed by SDS-PAGE and GLDH activity assay, and frozen at -80°C in 10 mg aliquots. Prior to use, (NH ₄ ) ₂ SO ₄ was removed by dialysis against a 10 kDa molecular weight cutoff filter. Factor Xa digestion was performed as described herein to remove the 2xSlyD portion from the KalbTG construct, thereby generating the KalbTG enzyme (KalbTGt1) having the following sequence:

One aspect of the protocol for expressing the KalbTG enzyme derived from the EcSlyD2-Xa-KalbTGt1-8xHis construct includes the addition of a source of ammonium ions (i.e., (NH ₄ ) ₂ SO ₄ or NH ₄ Cl), a natural inhibitor of the KalbTG enzyme, to the fermentation broth and purification buffer. The use of ammonium (or ammonia) enables the production of the KalbTG fusion protein without the need for expression or downstream cleavage of the propeptide sequence. The autocatalytic activity of the KalbTG enzyme is reversibly inhibited by the presence of ammonium ions (from ammonium chloride in solution) until the final dialysis to remove ammonium ions before the use of the KalbTG enzyme. Surprisingly, the purification process including the use of ammonium chloride results in an increase in KalbTG enzyme activity of up to about 9-fold as measured by the GLDH assay, making the KalbTG enzyme very competitive with currently commercially available MTG (see Example 3, Table 4). Therefore, it can be used to express, purify, or otherwise isolate transglutaminases in the presence of ammonium chloride, another ammonium salt, or another source of ammonia.

In certain embodiments, it can be useful to select a source of ammonium where the counterion has a neutral effect in the overall process. For example, in experiments where the ammonium counterion was sulfate, a negative effect on expression was observed compared to using chloride as the counterion. However, the use of sulfate as the counterion may have minimal to no negative effect with respect to the purification steps. Therefore, it can be useful to first determine the effect of the counterion on the expression and purification of a given transglutaminase. Additionally, the concentration of ammonium ions present during the expression or purification of a given transglutaminase can be varied. In one embodiment, the ammonium may be present at a concentration of at least about 10 μM. In another embodiment, the ammonium may be present at a concentration of at least about 100 μM. In another embodiment, the ammonium may be present at a concentration of at least about 1 mM. In another embodiment, the ammonium may be present at a concentration of at least about 10 mM.

Example 3: Use of peptide arrays for transglutaminase substrate discovery

To identify potential acyl-donor and amine-donor substrates for Kuznetsov albicans transglutaminase, a high-throughput screen on a 5-mer peptide array was performed (Figures 4A and 4B). An activity of at least 1.65 U/mg for SlyD-fused and mature KalbTG enzymes was confirmed by commercial assays (Zedira MTG-ANiTA-KIT; relative to 4.3 U mg ^-1 by MTG provided with the kit and a 0.07 U/mg blank using BSA). The assay used β-casein as a cross-linked substrate and detected deamidation by glutamate dehydrogenase-dependent NADPH oxidation. KalbTG was assayed using a peptide array prepared by maskless array synthesis (U.S. Patent Publication No. 2015/0185216 to Albert et al., filed December 19, 2014) to identify specific recognition motifs. The efficiency of the transamidation reaction between 1.4 million unique 5-mer peptides and a biotinylated amine donor was quantified in parallel on two arrays, and the sequence of the peptide with the highest turnover was determined ( FIG4A ). The nine best peptides were resynthesized and tested for KalbTG activity in a separate GLDH coupled assay. The results of the GLDH coupled assay are shown in Table 1 along with the corresponding array data. It should be noted that the array signal for the sequence MLAQG (SEQ ID NO: 13) is marked as not applicable (na) because this sequence was not included on the array. Similarly, the measurement of background signal applies only to experiments performed on the array.

KalbTG activity was determined in a GLDH-coupled assay by measuring the incorporation of biotinylated amine-donors and NADH oxidation rates on the peptide array in the presence of amine-donor substrates (500 µM) at 340 nm and 37°C using 100 µM of each of the nine best-performing array-selected glutamine-containing substrates and two S. mobara MTG glutamine-containing substrates. A strong correlation was observed between the best array-selected substrates and their performance in the GLDH-coupled assay, while KalbTG showed no activity against the preferred S. mobara MTG substrates DYALQ (SEQ ID NO: 22) and MLAQG (SEQ ID NO: 13). Testing of the top sequences in Table 1 confirmed YRYRQ (SEQ ID NO: 1) and RYRQR (SEQ ID NO: 14) as the best-performing 5-mer substrates, with turnover rates of 3.52 ± 0.08 pmol ^NADHs and 3.60 ± 0.12 pmol ^NADHs , respectively. The lysine-containing substrate YKYRQ (SEQ ID NO: 20) exhibited the highest turnover rate in the GLDH assay (4.00 ± 0.18 pmol ^s ). Without being bound by theory, this may be an artifact of lysine cross-reactivity and was therefore omitted from further analysis. Surprisingly, no activity was detected using the well-known S. mobara MTG recognition motif MLAQGS (SEQ ID NO: 23) (represented by the 5-mer sequence MLAQG (SEQ ID NO: 13)) or the S. mobara MTG substrate DYALQ (SEQ ID NO: 22).

A second round of maturation on the peptide array yielded APRYRQRAA (SEQ ID NO: 24) as the best performing 9-mer substrate, which was then resynthesized as a biotinylated peptide to serve as an acyl donor for discovery of the optimized lysine recognition motif behind the 5-mer peptide array ( FIG. 4B ). Again, the six best lysine-containing peptides were resynthesized and tested in an activity assay in KalbTG solution using peptides containing the optimized glutamine recognition sequence YRYRQ (SEQ ID NO: 1) as an acyl donor (Table 2). It should be understood that since cadaverine was omitted from the array, calculations of the array signal for cadaverine were not applicable.

Referring to Table 2, KalbTG activity was determined by measuring biotinylated glutamine donor incorporation and NADH oxidation rates at 340 nm and 37°C in the presence of a glutamine donor substrate (200 µM) in a GLDH-coupled assay using 100 µM of each of the following: i) the six best-performing array-selected lysine substrates, ii) cadaverine, and iii) the preferred MTG lysine substrate, ARSKL (SEQ ID NO:30). The highest turnover in the GLDH assay was observed with the sequence RYESK (SEQ ID NO:2) (4.47 ± 0.16 pmol NADHs ^-1 ). This represents a small but significant increase relative to cadaverine (3.51 ± 0.12 pmol s ^-1 ) or ARSKL (SEQ ID NO:30) (3.87 ± 0.31 pmol s ^-1 ), peptides previously established on a peptide array as preferred MTG lysine donor motifs. Additional details regarding the S. mobara MTG substrate peptide ARSKL (SEQ ID NO: 30) can be found in U.S. Provisional Patent Application Serial No. 62/094,495 to Albert et al., filed December 19, 2014.

Regarding the construction of the peptide array, a library of 1,360,732 unique 5-mer peptides was designed using all combinations of the 18 natural amino acids except cysteine and methionine, as well as any dimer or longer repeat of the same amino acid, and any peptide containing a dipeptide selected from the sequences HR, RH, HK, KH, RK, KR, HP, and PQ. The library was synthesized in duplicate on the same array using maskless light-directed peptide array synthesis. Each 5-mer peptide was flanked by a 3-amino acid linker at the N- and C-termini, synthesized using a mixture of glycine and serine with a 3:1 ratio.

To test the specificity of KalbTG for glutamine-containing substrates, N-(biotinyl)cadaverine was used as an alternative to lysine substrates to biotinylate glutamine-peptides on the peptide array. The KalbTG labeling reaction was performed in a SecureSeal ^™ chamber (Grace Bio-Labs) at 37°C for 45 minutes in 1200 μL of 100 mM Tris-HCl (pH 8), 1 mM DTT, 50 μM N-(biotinyl)cadaverine, and 0.2 ng ^μl⁻¹ KalbTG. Following incubation, the chamber was removed and the array was washed in 20 mM Tris-HCl (pH 7.8), 0.2 M NaCl, and 1% SDS for 1 minute, followed by a 1-minute wash in 20 mM Tris-HCl. The biotin-linked array was stained with 0.3 μg ^ml Cy5-streptavidin in 10 mM Tris-HCl (pH 7.4), 1% alkali-soluble casein, and 0.05% Tween-20 for 1 hour at room temperature. Cy5 fluorescence intensity was measured using a fluorescence scanner at a resolution of 2 μm and a wavelength of 635 nm.

To test the specificity of KalbTG for lysine substrates, a chemically synthesized Z-APRYRQRAAGGG-PEG-biotin peptide (containing the sequence APRYRQRAAGGG (SEQ ID NO: 31)) was used as a glutamine-containing substrate to biotinylate lysine-containing peptides. Array biotinylation was performed as described above using 0.1 ng ^μl⁻¹ KalbTG and 0.8 μM peptide at 37°C for 15 minutes. It should be noted that "Z-" preceding a peptide or other similar construct herein represents a carboxybenzyl group unless otherwise indicated.

The S. mobara MTG reaction on the peptide array was performed in 100 mM Tris-HCl (pH 8), 1 mM DTT containing 10 μM N-(biotinyl)cadaverine and 0.1 ng μl ⁻¹ S. mobara MTG at 37° C. for 15 minutes. As shown in FIG4A , the best 22 glutamine-donor substrates identified on the 5-mer peptide array using KalbTG are (in no particular order) FRQRG (SEQ ID NO: 18), YRYRQ (SEQ ID NO: 1), QRQRQ (SEQ ID NO: 19), FRQRQ (SEQ ID NO: 16), RYRQR (SEQ ID NO: 14), RQRQR (SEQ ID NO: 17), YRQSR (SEQ ID NO: 32), YKYRQ (SEQ ID NO: 20), LRYRQ (SEQ ID NO: 33), YRQRA (SEQ ID NO: 34), VRYRQ (SEQ ID NO: 35), QRQTR (SEQ ID NO: 36), YRQTR (SEQ ID NO: 37), PRYRQ (SEQ ID NO: 38), RFSQR (SEQ ID NO: 39), WQRQR (SEQ ID NO: 40), QYRQR (SEQ ID NO: 41). NO:21), VRQRQ (SEQ ID NO:41), RYTQR (SEQ ID NO:42), AYRQR (SEQ ID NO:43), YQRQR (SEQ ID NO:44) and RYSQR (SEQ ID NO:15). As shown in Figure 4B, the best 17 lysine-donor substrates identified on the 5-mer peptide array using KalbTG are (in no particular order) NYRFK (SEQ ID NO:45), YQKWK (SEQ ID NO:46), YKYKY (SEQ ID NO:47), RWKFK (SEQ ID NO:48), RFYSK (SEQ ID NO:49), YKYAK (SEQ ID NO:50), YRYAK (SEQ ID NO:51), RYSYK (SEQ ID NO:52), YKSFK (SEQ ID NO:53), YKSWK (SEQ ID NO:54), KYRYK (SEQ ID NO:55), YKYNK (SEQ ID NO:56), RYSKY (SEQ ID NO:25), RYESK (SEQ ID NO:2), PYKYK (SEQ ID NO:57), FYKYK (SEQ ID NO:58), and FYESK (SEQ ID NO:59). The 16 glutamine-donor substrates identified using MTG and shown in Figures 5B and 5C are (in no particular order) EWVAQ (SEQ ID NO:60), EWALQ (SEQ ID NO:61), DYFLQ (SEQ ID NO:62), DYALQ (SEQ ID NO:22), EYWLQ (SEQ ID NO:63), DWALQ (SEQ ID NO:64), DWYLQ (SEQ ID NO:65), DYWLQ (SEQ ID NO:66), EYVAQ (SEQ ID NO:67), DYVAQ (SEQ ID NO:68), DWVAQ (SEQ ID NO:69), EYVLQ (SEQ ID NO:70), EWIAQ (SEQ ID NO:71), WYALQ (SEQ ID NO:72), EYALQ (SEQ ID NO:73), and EYFLQ (SEQ ID NO:74).

Example 4: Specificity of KalbTG for mature glutamine substrates and application in semi-orthogonal conjugation

Since peptide arrays can deliver readouts for all available 5-mer peptides at once, a single data set is sufficient to evaluate how enzymes differ in substrate specificity. The optimal KalbTG glutamine substrate ( FIG. 4A ) was found in the middle of the signal distribution on the array performed with MTG ( FIG. 5A ). By contrast, the best-performing S. mobaraensis MTG glutamine-containing substrate ( FIG. 5B ) exhibited a relatively low signal on the KalbTG array ( FIG. 5C ). To confirm that the two transglutaminases have orthogonal glutamine-containing substrate preferences and to quantify the amount of cross-reactivity, the kinetics of the two enzymes were determined in the presence of different concentrations of the substrate peptides Z-GGGYRYRQGGGG and Z-GGGDYALQGGGG ( FIG. 5D ). Notably, the Z-conjugated substrates include the peptide sequences GGGYRYRQGGGG (SEQ ID NO: 75) and GGGDYALQGGGG (SEQ ID NO: 76). Streptomyces mobaraensis MTG exhibited similar _KM values in the range of 0.6-0.9 mM for both substrates, while the turnover _kcat was significantly higher for the Z-GGGDYALQGGGG substrate comprising the preferred DYALQ (SEQ ID NO: 22) sequence (1.39 s ^-1 versus 0.93 s- ¹ for YRYRQ (SEQ ID NO: 1)), resulting in catalytic efficiencies (kcat _KM ^-1 ) of 1.64 x ¹⁰³ [M ^-1 s ^-1 ] and 1.44 x ¹⁰³ [M ^-1 s ^-1 ], _respectively (Table 3). Relative to the engineered S. mobara MTG enzyme, KalbTG appears to have a lower substrate binding efficiency ( _KM of 2 mM), but a higher turnover ( _kcat of 1.92 s ^-1 ), resulting in a _kcat KM ^-1 _of 0.89 x ¹⁰³ [M ^-1 s ^-1 ]. KalbTG appears to be completely unreactive towards the S. mobara MTG substrate Z-GGGDYALQGGGG, and thus no kinetic parameters could be determined, as indicated by 'nd' in Table 3.

Next, the array and solution data were used to perform site-specific labeling on protein substrates. The molecular chaperone SlyD is a useful scaffold for labeling schemes because the loop containing the epitope can be transplanted onto the FKBP domain for presentation to a binding agent or enzyme (PCT Application Publication No. WO 2012/150321A1 by Andres et al.). A chimeric protein consisting of the Thermus thermophilus FKBP domain and the KalbTG recognition sequence RYRQR (SEQ ID NO: 14) was produced. Labeling was performed with a 10-fold excess of KalbTG K-tag-Cy3 and a substrate:enzyme ratio of 72:1, resulting in approximately 70% yield of labeled protein species after 15 minutes (Figure 6A). This yield remained constant over a 60-minute time course. Incubation with a 50-fold excess of labeling only slightly increased the yield of labeled material. A molecular weight shift from 13 kDa to 19 kDa was observed on the SDS-PAGE gel, which corresponds exactly to the incorporation of a single 6 kD labeled molecule. An identically constructed FKBP domain containing the S. mobara MTG sequence DYALQ (SEQ ID NO: 22) instead of RYRQR (SEQ ID NO: 14) showed no label incorporation when incubated with KalbTG ( FIG6A ), indicating that the reaction is restricted to the site of the KalbTG recognition motif and that none of the five other glutamines inherent to the FKBP domain are recognized. Furthermore, we determined the pH dependence of the labeling reaction at pH 6.2, 6.8, 7.4, 8.0, 8.5, and 9 ( FIG6B ). The highest labeling efficiency after 15 minutes was observed at pH 7.4, with activity trailing off at pH 8.5 and above. These findings correspond well with the published pH preferences of S. mobara MTG.

Returning to Figure 6C, the high sequence specificity of KalbTG was used to conjugate a 6 kDa Cy3 tag to the YRYRQ (SEQ ID NO: 1) site of a 7 kDa substrate peptide containing the glutamine-containing motifs of KalbTG and Streptomyces mobara MTG. The reaction was allowed to proceed for 30 minutes to saturate the YRYRQ (SEQ ID NO: 1) site. SDS-PAGE analysis confirmed that the tag was incorporated at a single site. The substrate peptide was then incubated with Streptomyces mobara MTG and the 6 kDa Cy5 tag for 15 minutes. This resulted in the formation of site-specifically doubly labeled conjugates, with all singly labeled species visibly converted to doubly labeled species. These results demonstrate that KalbTG and Streptomyces mobara MTG constitute a semi-orthogonal labeling system with unprecedented ease of use, yield, and efficiency. Therefore, KalbTG can be used to synthesize complex protein conjugates of interest for therapeutic or diagnostic applications on an industrial scale.

Regarding the GLDH coupled assay, to determine whether the KalbTG peptides selected in the array assay were also preferred substrates in solution reactions and to quantify the cross-reactivity of KalbTG and S. mobaraensis MTG with different substrates, a continuous glutamate dehydrogenase (GLDH) coupled assay for S. mobaraensis MTG activity was applied (see Oteng-Pabi, et al., 2013. Analytical biochemistry 441(2): 169-173).

For glutamine substrate evaluation, the cells were cultured in the presence of 500 µM α-ketoglutarate, 500 µM or 1 mM cadaverine (as an amine donor to replace lysine-containing peptides), 2 U ml ⁻¹ glutamate dehydrogenase (GLDH), 500 µM NADH, and 0-1 The assay was performed in the presence of mid-mM concentrations of a glutamine-containing substrate peptide (Z-GGGQRWRQGGGG, Z-GGGWRYRQGGGG, Z-GGGYRYRQGGGG, Z-GGGRYRQRGGGG, Z-GGGRYSQRGGGG, Z-GGGFRQRQGGGG, Z-GGGRQRQRGGGG, Z-GGGFRQRGGGGG, Z-GGGQRQRQGGGG, Z-GGGYKYRQGGGG, Z-GGGQYRQRGGGG, Z-GGGDYALQGGGG, or Z-GGGMLAQGSGGG) in 200 mM MOPS, 1 mM EDTA, pH 7.2, in a clear 96-well microtiter plate (total volume 200 µl per well). Notably, Z-conjugated peptides include the sequences GGGQRWRQGGGG (SEQ ID NO:77), GGGWRYRQGGGG (SEQ ID NO:78), GGGYRYRQGGGG (SEQ ID NO:75), GGGRYRQRGGGG (SEQ ID NO:79), GGGRYSQRGGGG (SEQ ID NO:80), GGGFRQRQGGGG (SEQ ID NO:81), GGGRQRQRGGGG (SEQ ID NO:82), GGGFRQRGGGGG (SEQ ID NO:83), GGGQRQRQGGGG (SEQ ID NO:84), GGGYKYRQGGGG (SEQ ID NO:85), GGGQYRQRGGGG (SEQ ID NO:86), GGGDYALQGGGG (SEQ ID NO:76), and GGGMLAQGSGGG (SEQ ID NO:87).

For the amine substrate evaluation, the assay conditions were the same as for the glutamine substrate evaluation, except that 100 µM of each amine substrate (Z-GGGRYSKYGGGG, Z-GGGAYRTKGGGG, Z-GGGRYRSKGGGG, Z-GGGYKGRGGGGG, Z-GGGRYGKSGGGG, Z-GGGRYESKGGGG, Z-GGGPGRYKGGGG, Z-GGGARSKLGGGG, or cadaverine) and 200 µM of a glutamine donor (Z-GGGYRYRQGGGG or Z-GGGDYALQGGGG) were used. Notably, amine substrates include the sequences GGGRYSKYGGGG (SEQ ID NO:88), GGGAYRTKGGGG (SEQ ID NO:89), GGGRYRSKGGGG (SEQ ID NO:90), GGGYKGRGGGGG (SEQ ID NO:91), GGGRYGKSGGGG (SEQ ID NO:92), GGGRYESKGGGG (SEQ ID NO:93), GGGPGRYKGGGG (SEQ ID NO:94), and GGGARSKLGGGG (SEQ ID NO:95).

Reactions were initiated by adding 5 µg ^ml⁻¹ of Streptomyces mobaraensis MTG or KalbTG, and NADH oxidation was recorded continuously at 340 nm for 60 minutes using a Biotek Synergy H4 microplate reader controlled at 37°C, with brief shaking intervals preceding each measurement. After a short delay period (in which GLDH was saturated with transglutaminase-mediated ammonia release), a linear rate of absorbance versus time corresponding to transglutaminase turnover was observed and Michaelis-Menten kinetic analysis was performed. Absorbance rates in milliospatial density units per minute ( ^mODmin⁻¹ ) were converted to molar rates of NADH turnover ( ^pmols⁻¹ ) using the formula of Equation 1 (previously determined from an NADH standard curve):

(Equation 1)

For the labeling assay, the chaperone protein SlyD from Thermus thermophilus (Universal Protein Resource (UniProt) Number Q5SLE7) was used as the labeling scaffold for KalbTG. The SlyD sequence is:

The KalbTG glutamine donor sequence (Q-tag) was recombinantly grafted onto the FKBP domain of SlyD, resulting in the following polypeptide sequence:

8X-histidine-tagged proteins were produced in E. coli Bl21 Tuner and purified by standard nickel agarose-based immobilized metal ion affinity and size exclusion chromatography (HisTrap, Superdex 200; GE Healthcare).

The labeled peptide is chemically synthesized to have (in order from N-terminus to C-terminus) a "Z-" group (i.e., carboxybenzyl), a transglutaminase lysine donor sequence (K-tag), 8-amino-3,6-dioxaoctanoic acid (O2Oc), a peptide, and a Cy3 or Cy5 fluorescent dye. The basic chemical structure of the labeled peptide is:

KalbTG K-label-Cy3:

MTG K-Tag-Cy5:

wherein E is glutamate, U is β-alanine, and C(sCy3-MH) and C(Cy5-MH) represent C-terminal cysteine modified with sulfo-Cy3 maleimide or Cy5 maleimide, respectively, after synthesis.

For orthogonal labeling experiments, molecules containing KalbTG and MTG Q-tags were chemically synthesized with the basic chemical structure:

.

All peptides were synthesized using commercially available building blocks on a 0.25 mmol scale via standard solid phase peptide synthesis based on fluorenylmethoxycarbonyl (FMOC). After solid phase synthesis, the peptide was cleaved with a solution of 95% TFA, 2.5% triisopropylsilane, and 2.5% water. The peptide was then precipitated with diisopropyl ether and purified by high performance liquid chromatography (RP18-HPLC) based on reversed phase C18 using a water/TFA acetonitrile gradient. Dye labeling was achieved by reacting the peptide with sulfo-Cy3 maleimide (Lumiprobe) and Cy5 maleimide (GE Healthcare). Purification of dye-labeled peptides was achieved using a water/TFA acetonitrile gradient by RP18-HPLC. The identity of the peptide was confirmed by liquid chromatography-mass spectrometry (LC-MS) (Thermo Scientific RSLC-MSQplus system) using a Kinetex C18 2.6µm, 50 x 3 mm column (Phenomenex).

Unless otherwise indicated, labeling reactions were performed in 200 mM MOPS, pH 7.2, and 1 mM EDTA in the presence of 72 µM substrate protein, 720 µM labeled peptide, and 1 µM transglutaminase at 37°C for 15 minutes. For pH-dependent labeling profiles, experiments were performed in 200 mM MOPS (for each of pH 6.2, 6.8, and 7.4) or in 200 mM Tris (for each of pH 8.0, 8.5, and 9.0). For orthogonal labeling experiments, 1.5 µM EcSlyD2-Xa-KalbTGt3-8xHis was added to a 20 µl volume containing 100 µM substrate peptide and 1 mM KalbTG K-tag-Cy3. After incubation at 37°C for 30 minutes, 1 mM MTG K-tag-Cy5 and 1.5 µM Streptomyces mobaraensis MTG were added, and the mixture was incubated at 37°C for an additional 15 minutes. The reaction was stopped by adding 50 mM TCA. Samples were taken between incubation steps and analyzed by fluorescence in SDS-PAGE gels (BioRad ChemiDoc gel documentation system, Cy3 and Cy5 LEDs and filter sets). In another orthogonal labeling experiment, 4.65 µM EcSlyD2-Xa-KalbTGt3-8xHis was added to a 50 µl volume containing 302 µM substrate peptide and 3.02 mM KalbTG K-tag-Cy3 (i.e., Cy3-labeled lysine substrate for KalbTG). After incubation at 37°C for 30 minutes, the mixture was diluted by adding 400 µl of buffer and then concentrated to 50 µl using a 10 kDa molecular weight cutoff spin filter. Next, 3.02 mM MTG K-Tag-Cy5 (i.e., Cy5-labeled lysine substrate for Streptomyces mobaraensis MTG) and 4.65 µM Streptomyces mobaraensis MTG were added and incubated at 37°C for an additional 15 minutes. The reaction was stopped by the addition of 50 µM (NH ₄ ) ₂ SO ₄ . Then, 2 µg of Factor Xa was added, and the mixture was incubated at room temperature for 2 hours. The reaction was stopped by the addition of 50 mM TCA. The mixture was filtered (0.2 µm spin filter) and analyzed by LC-MS with UV detection at 214 nm and 305 nm.

Another experiment was performed to determine the activity of KalbTG enzymes prepared from two different constructs and stored at two different temperatures. The KalbTGt3 (SEQ ID NO: 10) and KalbTGt1 (SEQ ID NO: 12) enzymes tested were obtained from the EcSlyD2-Xa-KalbTGt3-8xHis-tag construct (SEQ ID NO: 8) and EcSlyD2-Xa-KalbTGt1-8xHis-tag construct (SEQ ID NO: 11) described in Example 3, respectively. The activity of the two KalbTG enzymes was tested after storage at 4°C or -20°C and compared to the activity of commercially available Streptomyces mobaraensis MTG according to a published protocol (Oteng-Pabi, et al., 2013. Analytical biochemistry 441(2): 169-173). Specifically, the assay was performed at 37°C in a total volume of 200 μl. The assay mixture consisted of 200 mM MOPS, 1 mM EDTA, and 1 mM DTT (at pH 7.4). α-ketoglutarate and NADH were also added at a concentration of 500 µM, along with 2 U/mL GLDH and 0.1 µg to 1.0 µg of one of the transglutaminases. The glutamine substrate was 200 µM Z-GGGYRYRQGGGG-OH, and the amine donor was 10 mM cadaverine, where Z represents a carboxybenzyl group. All data were collected in triplicate (average of three wells in a microtiter plate), and baseline activity (enzyme-free buffer) was subtracted. The resulting data are shown in Table 4.

Example 5: Identification and characterization of the 3D crystal structure of KalbTG

As shown in Figures 7A-7C, alignment of the complete structure of MTG from Kuznetsova albicans with MTG from Streptomyces mobara provides insight into the properties of KalbTG. In one aspect, the first 19 N-terminal amino acids and the C-terminal artificial GGGSG-8X-His tag are disrupted and therefore do not participate in the structure. The overall structure of KalbTG is similar to the previously described Streptomyces mobara MTG structure (Kashiwagi, et al. 2002. J BioChem 277(46): 44252-44260.), forming a disc-like shape of the α+β pleated sheet type with two polycyclic protrusions forming the active site cleft. However, the structure differs in two α-helices ( _α4 and _α5 in Kashiwagi numbering) that are absent from the KalbTG structure and two small β-strands ( _β2 and _β4 ) that contain less hydrophobic residues in KalbTG, replacing AF with SF and LV with QV, respectively, bringing the total elements in KalbTG to only nine α-helices and six β-strands. The catalytic triad (Cys64, Asp255, His274) is structurally conserved (Cys82, Asp211, His224 numbered from the beginning of the KalbTG open reading frame). However, the sulfhydryl side chain of Cys82 in KalbTG is embedded 2.6 Å deeper in the active cleft than in its S. mobaraensis MTG counterpart. The crystal structure of the S. mobara MTG zymogen (Yang et al. 2011. J Bio Chem 286(9): 7301-7307) shows that the active cleft is tightly occupied by the L-shaped propeptide. The binding groove of KalbTG can be aligned with the propeptide of S. mobara MTG without steric hindrance, indicating that a similar zymogen mechanism may exist in KalbTG (Figure 7B). Surprisingly, one of the loops forming the active cleft near Cys82 has the amino acid sequence YRYRAR (SEQ ID NO: 4), which, except for the glutamine side chain, is identical to the preferred KalbTG substrate found on the peptide array (i.e., the best two 5-mer peptides are YRYRQ (SEQ ID NO: 1) and RYRQR (SEQ ID NO: 14); Figure 7C). Therefore, the crystal structure analysis serves as an independent verification of the reliability of the peptide array with respect to the identification of substrate sequences.

For crystallization and structural characterization of KalbTG, KalbTG was crystallized in PBS at 22°C using the sitting drop (200 nL) vapor diffusion method as follows: 8 mg/mL of protein was mixed 1:1 with an unbuffered reservoir solution consisting of 0.2 M ammonium tartrate and 20% PEG3350. Crystals were cryoprotected in a reservoir solution containing 20% ethylene glycol and then rapidly cooled in liquid nitrogen. Data were collected at 100 K using a Pilatus 6M detector at SLS beamline PX-II and integrated and scaled in space group P3 using XDS (PMID 20124692). The l = 3n reflection had an I/σ > 9, making the presence of a helical axis unlikely. Self-Patterson and pairwise analyses revealed no suspected pathologies in the data. The unit cell volume is consistent with two or three KalbTG molecules in the asymmetric unit, with Matthews parameters of 3.5 ^ų /Da and 2.3 ^ų /Da, respectively. Data collection statistics are summarized in Table 5.

The structure of KalbTG (226 residues) was determined by molecular replacement using Streptomyces mobaraensis transglutaminase (354 residues, RCSB Protein Data Band (PDB) ID No. 3iu0) as a search model. Initial attempts using the full Streptomyces mobaraensis TG were generally unsuccessful, not to be limited by theory, because the enzymes have very different sizes. The two transglutaminases share 28.2% sequence identity and 38.9% sequence similarity over the entire length of KalbTG. When searching for both molecules in the asymmetric unit in space group P3 with a log-likelihood gain (LLG) of 213, a variant of Streptomyces mobaraensis TG lacking the loop region and trimmed to the hydrophobic core yielded a potential solution using Phaser crystallization software (McCoy et al., 2007. J Appl Crystallogr. Aug 1;40(Pt 4):658-674). Trigonal space groups _P31 and _P32 did not yield a solution, consistent with the high intensity of the l = 3n reflection. The model was refined to an R _freedom of 46% using BUSTER crystallization software (Blanc et al., 2004. Acta Cryst . D60, 2210-2221). Some secondary structure elements were visible in the electron density map and included in the model, which was then subjected to 10 rounds of automatic model building and refinement in CBUCCANEER and REFMAC5 (Winn et al., 2011. Acta Cryst . D67, 235-242). The resulting model contained all protein residues and had an R _freedom of 30%. The structure was refined in COOT (Emsley et al., 2010. Acta Cryst . D66, 486-501) and refined with PHENIX (Adams et al., 2010. Acta Cryst. D66, 213-221). Model refinement statistics are summarized in Table 5.

Referring to Table 5, the statistics of the highest resolution shells are shown in parentheses.

Dynamic scanning calorimetry (DSC) measurements were performed in a VP-Capillary DSC instrument (MicroCal/GE Healthcare) in the temperature range of 20°C to 90°C and a scan rate of 90°C h ^-1 using PBS as a reference.

Example 6: Analysis of the characteristics of the identified KalbTG peptide substrate

With reference to Tables 6 and 7, analysis of the KalbTG peptide substrates identified herein revealed a set of features shared by those substrates.

Turning first to Table 6, 22 5-mer peptide sequences identified as acyl-donor substrates for KalbTG are listed using single-letter amino acid codes, along with information including amino acid position (numbered from N-terminus to C-terminus) and amino acid counts for individual amino acids (R and Q) and amino acid groups (R+Q, F+W+Y, and R+Q+F+W+Y). In one aspect, for KalbTG, the data revealed that the acyl-donor substrates (including 5-mer amino acid sequences having the formula _Xaa1 - _Xaa2 - _Xaa3 - _Xaa4 - _Xaa5 , where Xaa is any amino acid) generally adhered to several design principles. First, each 5-mer sequence included at least one glutamine (Q). More specifically, at least one of the third, fourth, and fifth positions (i.e., _Xaa3 , _Xaa4 , and _Xaa5 ) of the 5-mer sequence was a glutamine. Notably, sequences with two or more adjacent glutamines were generally not observed. However, several sequences were observed that included a glutamine at each of the third and fifth positions. In another aspect, it was observed that each 5-mer sequence included at least one arginine (R). More specifically, at least one of the fourth and fifth positions (i.e., Xaa ₄ and Xaa ₅ ) of the 5-mer sequence was an arginine. For example, for each of the 22 sequences analyzed, an arginine was found at either (but not both) the fourth or fifth position. Furthermore, it was observed that when the fifth position was an arginine, at least one additional arginine was located at one of the first, second, and third positions (i.e., Xaa ₁ , Xaa ₂ , and Xaa ₃ ) of the 5-mer sequence.

In another aspect, the 5-mer sequences each include at least one arginine adjacent to a glutamine. For example, the 5-mer sequence FRQRG (SEQ ID NO: 8) includes a glutamine at the third position, flanked by arginines at both the second and fourth positions, while the 5-mer sequence YRYRQ (SEQ ID NO: 1) includes an arginine at the fourth position, followed by a glutamine at the fifth position. In addition to the presence of at least one arginine and at least one glutamine in each 5-mer sequence, amino acids with aromatic side chains (i.e., phenylalanine, tryptophan, and tyrosine) were found in many 5-mer sequences. Specifically, it was observed that the total number of positions occupied by arginine, glutamine, phenylalanine, tryptophan, or tyrosine in each 5-mer sequence was at least 4 (see the last column in Table 6 for amino acid counts of R+Q+F+W+Y). For example, the 5-mer sequence FRQRG (SEQ ID NO: 8) includes one glutamine, two arginines, and one phenylalanine for a total of four positions occupied by amino acids selected from arginine, glutamine, phenylalanine, tryptophan, and tyrosine. In another embodiment, the 5-mer sequence YRYRQ (SEQ ID NO: 1) includes two arginines, two phenylalanines, and one glutamine for a total of five positions occupied by amino acids selected from arginine, glutamine, phenylalanine, tryptophan, and tyrosine. Notably, the 5-mer sequence lacking aromatic amino acids includes a total of at least four amino acids selected from arginine and glutamine (e.g., QRQRQ (SEQ ID NO: 19), QRQTR (SEQ ID NO: 36)).

Referring to Table 7, 22 5-mer peptide sequences identified as amine-donor substrates for KalbTG are listed using a single-letter amino acid code, along with information including amino acid position (numbered from N-terminus to C-terminus) and amino acid counts for individual amino acids (K, Y, R, S) and amino acid groups (K+Y+R+S). Similar to the data obtained for the acyl-donor sequences in Table 6, for KalbTG, amine-donor substrates (including 5-mer amino acid sequences having the formula _Xaa1 - _Xaa2 - _Xaa3 - _Xaa4 - _Xaa5 , where Xaa is any amino acid) generally adhere to several design principles. First, each 5-mer sequence includes at least one lysine (K). More specifically, at least one of the fourth and fifth positions of each 5-mer sequence is a lysine, with the sole exception of the 5-mer sequence YKGRG (SEQ ID NO: 29). Notably, sequences with two or more adjacent lysines were not generally observed. Although several 5-mer sequences were observed that included two total lysines, no 5-mer sequences with more than two lysines were found among the 5-mer sequences analyzed. On the other hand, it was observed that the majority of 5-mer sequences included at least one tyrosine (Y), with the only two exceptions among the 22 sequences in Table 7 being RWKFK (SEQ ID NO: 48) and ARSKL (SEQ ID NO: 30).

After the amino acids lysine and tyrosine, the next most common amino acids occurring in the 5-mer sequences of Table 7 are arginine and serine, with at least one arginine or serine present in many 5-mer sequences. Specifically, it was observed that the total number of positions occupied by lysine, tyrosine, arginine, or serine in each 5-mer sequence was at least 3 (see the last column in Table 7 for amino acid counts of K+Y+R+S). For example, the 5-mer sequence NYRFK (SEQ ID NO: 45) includes tyrosine, arginine, and lysine for a total of 3 positions occupied by amino acids selected from lysine, tyrosine, arginine, and serine. In another embodiment, the 5-mer sequence RYRSK (SEQ ID NO: 27) includes 2 arginines, 1 tyrosine, 1 serine, and 1 lysine for a total of 5 positions occupied by amino acids selected from lysine, tyrosine, arginine, and serine.

It is noteworthy that all 5-mer sequences include a total of at least two amino acids selected from lysine, tyrosine, and arginine or serine. For example, ARSKL (SEQ ID NO:30) includes one lysine, one arginine, one serine, and zero tyrosines. Thus, the total number of amino acids selected from lysine, tyrosine, and arginine is two, and the total number of amino acids selected from lysine, tyrosine, and serine is also two. Of course, as discussed above, the total number of positions occupied by lysine, tyrosine, arginine, or serine in the 5-mer sequence ARSKL (SEQ ID NO:30) is at least three. In a related aspect, 5-mer sequences having one lysine and at least one of the amino acids tyrosine and arginine include a total of at least two amino acids selected from lysine, tyrosine, and arginine (e.g., ARSKL (SEQ ID NO:30), FYESK (SEQ ID NO:59)). In addition, each 5-mer sequence includes at least one tyrosine or arginine at one of positions 1 and 2 (ie, Xaa ₁ and Xaa ₂ ).

The schematic flow charts shown in the figures are generally described as logic flow chart diagrams. Thus, the order of description and the steps marked indicate one embodiment of the method presented. Other steps and methods that are equivalent in function, logic or effect to one or more steps of the method explained or parts thereof can be envisioned. In addition, the forms and symbols used in the figures are provided to explain the logical steps of the method and are understood not to limit the scope of the method. Although multiple arrow types and line types can be used, they are understood not to limit the scope of the corresponding method. In fact, some arrows or other connectors can be used to simply indicate the logical flow of the method. For example, an arrow can indicate a waiting or monitoring phase of unspecified duration between the enumerated steps of the described method. In addition, the order in which a particular method occurs may or may not be strictly in accordance with the order of the corresponding steps shown.

In the following description, the present invention is presented in several different embodiments with reference to the accompanying drawings, in which like reference numerals represent like or similar elements. Reference throughout this specification to "one embodiment," "an embodiment," or similar language means that a particular component, structure, or feature described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, the appearance of the phrases "in one embodiment," "in an embodiment," and similar language in this specification may, but do not necessarily, all refer to the same embodiment.

The components, structures or features of the present invention can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are listed to provide a thorough understanding of the embodiments of the system. However, those skilled in the relevant art will recognize that the systems and methods can be practiced without one or more specific details, or with other methods, components, materials, etc. In other cases, well-known structures, materials or operations are not shown or described in detail to avoid obscuring aspects of the present invention. Therefore, the foregoing description is intended to be exemplary and does not limit the scope of the inventive concept.

Each reference cited in this application is incorporated herein by reference in its entirety.

Claims (26)

1. A substrate tag for microbial transglutaminase of *Kuznerella whiteis* as shown in SEQ ID NO:6, said substrate tag comprising one of an acyl-donor tag peptide sequence YRYRQ (SEQ ID NO:1) and an amine donor tag peptide sequence RYESK (SEQ ID NO:2). 2. The substrate label of claim 1, further comprising a detectable marker. 3. The substrate label of claim 2, wherein the detectable marker is selected from biotinylate, radioactive markers, and chemiluminescent markers. 4. The substrate label of claim 3, wherein the radioactive marker is a ruthenium marker. 5. The substrate label of claim 3, wherein the chemiluminescent label is a fluorescent dye. 6. The substrate tag of claim 1, wherein the acyl-donor tag has the peptide sequence APRYRQRAA (SEQ ID NO: 24). 7. A method for forming heteropeptide bonds in the presence of *Kuznerella whiteis* microbial transglutaminase as shown in SEQ ID NO:6, the method comprising: Microbial transglutaminase was exposed to a first substrate and a second substrate, the first substrate comprising the acyl-donor-tagged peptide sequence YRYRQ (SEQ ID NO:1) and the second substrate comprising the amine-donor-tagged peptide sequence RYESK (SEQ ID NO:2). The first substrate and the second substrate are crosslinked to form an isopeptide bond between the acyl donor tag and the amine donor tag. 8. The method of claim 7, wherein the step of crosslinking the first substrate and the second substrate forms an isopeptide bond between the γ-formamide group of the acyl-donor tag and the ε-amino group of the amine-donor tag. 9. The method of claim 7, wherein at least one of the first substrate and the second substrate comprises a detectable marker. 10. The method of claim 9, wherein the detectable marker is selected from biotinylate, radioactive markers, and chemiluminescent markers. 11. The method of claim 10, wherein the radioactive label is a ruthenium label. 12. The method of claim 10, wherein the chemiluminescent label is a fluorescent dye. 13. The method of claim 7, wherein the acyl-donor tag has the peptide sequence APRYRQRAA (SEQ ID NO:24). 14. The method of claim 7, wherein the crosslinking of the first substrate with the second substrate is achieved in a yield of at least 70%. 15. The method of claim 14, wherein the yield is achieved within approximately 30 minutes. 16. A kit for forming heteropeptide bonds in the presence of microbial transglutaminase, the kit comprising purified *Kuznerella alba* microbial transglutaminase as shown in SEQ ID NO:6, the kit further comprising one of a first substrate and a second substrate, the first substrate comprising an acyl-donor-tagged peptide sequence YRYRQ (SEQ ID NO:1), and the second substrate comprising an amine-donor-tagged peptide sequence RYESK (SEQ ID NO:2). 17. The kit of claim 16, wherein at least one of the first substrate and the second substrate comprises a detectable marker. 18. The kit of claim 17, wherein the detectable label is selected from biotinylate, radiolabels, and chemiluminescent labels. 19. The kit of claim 18, wherein the radioactive label is ruthenium labeling. 20. The kit of claim 18, wherein the chemiluminescent label is a fluorescent dye. 21. The kit according to claim 16, wherein the acyl-donor tag has the peptide sequence APRYRQRAA (SEQ ID NO: 24). 22. The kit of claim 16, wherein the kit further comprises another of the first substrate and the second substrate. 23. An enzyme product for forming isopeptide bonds, said enzyme product comprising isolated Kuznerella white microbial transglutaminase as shown in SEQ ID NO:6 and ammonium. 24. The enzyme product according to claim 23, wherein the ammonium is present at a concentration of at least about 10 μM. 25. An acyl-donor substrate for transglutaminase, said acyl-donor substrate being selected from YRYRQ (SEQ ID NO:1), FRQRG (SEQ ID NO:8), RYRQR (SEQ ID NO:14), RYSQR (SEQ ID NO:15), FRQRQ (SEQ ID NO:16), RQRQR (SEQ ID NO:17), QRQRQ (SEQ ID NO:19), YKYRQ (SEQ ID NO:20), QYRQR (SEQ ID NO:21), YRQSR (SEQ ID NO:32), LRYRQ (SEQ ID NO:33), YRQRA (SEQ ID NO:34), VRYRQ (SEQ ID NO:35), QRQTR (SEQ ID NO:36), YRQTR (SEQ ID NO:37), PRYRQ (SEQ ID NO:38), RFSQR (SEQ ID NO:39), WQRQR (SEQ ID NO:38), LRYRQ (SEQ ID NO:39), RFSQR (SEQ ID NO:39), and WQRQR (SEQ ID NO:30). NO:40), VRQRQ (SEQ ID NO:41), RYTQR (SEQ ID NO:42), AYRQR (SEQ ID NO:43) or YQRQR (SEQ ID NO:44). 26. An amine-donor substrate for transglutaminase, said amine-donor substrate being selected from RYESK (SEQ ID NO:2), RYSKY (SEQ ID NO:25), AYRTK (SEQ ID NO:26), RYRSK (SEQ ID NO:27), RYGKS (SEQ ID NO:28), YKGRG (SEQ ID NO:29), ARSKL (SEQ ID NO:30), NYRFK (SEQ ID NO:45), YQKWK (SEQ ID NO:46), YKYKY (SEQ ID NO:47), RWKFK (SEQ ID NO:48), RFYSK (SEQ ID NO:49), YKYAK (SEQ ID NO:50), YRYAK (SEQ ID NO:51), RYSYK (SEQ ID NO:52), YKSFK (SEQ ID NO:53), YKSWK (SEQ ID NO:54), KYRYK (SEQ ID NO:55), YKYNK (SEQ ID NO:56), RYSKY (SEQ ID NO:57), RYSKY (SEQ ID NO:58), RYK (SEQ ID NO:59), RYK (SEQ ID NO:50), RYK (SEQ ID NO:51), RYSYK (SEQ ID NO:52), YKSFK (SEQ ID NO:53), YKSWK (SEQ ID NO:54), KYRYK (SEQ ID NO:55), YKYNK (SEQ ID NO:56), RYSKY (SEQ ID NO:57), RYK (SEQ ID NO:58), RYK (SEQ ID NO:59), RYK (SEQ ID NO:50), RYK (SEQ ID NO:51), RYK (SEQ ID NO:52), RYK (SEQ ID NO:53), RYK (SEQ ID NO:54), RYK (SEQ ID NO:55), RYK (SEQ ID NO:56), RYK (SEQ ID NO:57), RYK (SEQ ID NO:58), RYK (SEQ ID NO:5 NO:56), PYKYK (SEQ ID NO:57), FYKYK (SEQ ID NO:58) or FYESK (SEQ ID NO:59).