Method of transpeptidation

Abstract

The invention provides methods of transpeptidation, transpeptidation products obtainable from said methods, and chimeric fusion polypeptides. An exemplary method of transpeptidation includes: ligating a donor substrate to an acceptor protein or peptide using a transpeptidase, wherein the donor substrate is a protein, peptide, amino acid amide or amino acid ester, and wherein the ligation reaction generates a ligated product and a by-product that is a substrate for the transpeptidase; and cleaving one or more amino acids from a terminus of the by-product using an exopeptidase, such that the truncated by-product is not a substrate for the transpeptidase, thereby shifting the reaction equilibrium of the transpeptidation reaction towards the ligated product.

Description

[0001] METHOD OF TRANSPEPTIDATION

[0002] Field of Invention

[0003] The present invention relates to a method of transpeptidation.

[0004] Background to the Invention

[0005] Site-specific labelling of proteins is a widely-used tool in both industrial and academic research. Transpeptidases such as sortases, butelases and asparaginyl endopeptidases, provide an effective strategy for labelling the N- and C-termini of proteins. This approach has now been developed and applied to increasing complex systems including cell-surface labelling, in vivo labelling and the production of therapeutic antibody-drug conjugates. Numerous refinements to these methods have been reported including the evolution of enzymes with enhanced and altered reactivity, the use of multiple enzymes to enable orthogonal labelling reactions and the development of approaches to the labelling of internal residues. In almost all cases however, an excess quantity of one of the reaction partners is required to enable full conversion of unlabelled to labelled protein.

[0006] Transpeptidases recognize a defined peptide motif in their substrate protein or peptide. An active-site cysteine residue reacts with this motif to generate a thioacyl intermediate which then reacts with a substrate peptide or protein to yield the product. The key challenge in such labelling reactions is that they are generally fully reversible (Figure 1A). The substrate recognition motif for the enzyme (e.g., LPXT / G for sortase) is still present in the product of the reaction, and the by-products produced during formation of the initial thioacyl intermediate are also substrates for the second step. The reversibility of the reaction results in it being under thermodynamic control, which leads to equilibrium mixtures of labelled and unlabelled products (Figure IB). Production of pure labelled proteins typically requires a large excess of the labelling peptide and / or removal of unlabelled proteins from reaction mixtures. For N-terminal labelling, the use of ester / depsipeptide substrates where the by-product, an alcohol, is no longer a substrate for the enzyme have been reported (Figure 1C). This strategy however is not applicable to C-terminal labelling since it would require incorporation of an ester linkage into an expressed protein. For C-terminal labelling, a variety of alternative strategies have therefore been developed. For example, the physical removal of by-product peptides for C-terminal labelling via the use of centrifugal filtration devices (5), the use of chemical reagents to react with product peptides (6) or chelation using high concentrations of Ni2+(7, 8). In other work, the substrate preferences of the asparaginyl endopeptidases have been exploited to enable selective reaction (9).

[0007] However, none of these previous strategies allow one-pot controlled quantitative peptide ligation using transpeptidases, that shift the reaction equilibrium towards the ligated product and are suitable for both N- and C- terminal labelling.

[0008] Summary of the Invention

[0009] In a first aspect, there is provided a method of transpeptidation comprising: ligating a donor substrate to an acceptor protein or peptide using a transpeptidase, wherein the donor substrate is a protein, peptide, amino acid amide or amino acid ester, and wherein the ligation reaction generates a ligated product and a by-product that is a substrate for the transpeptidase, then cleaving one or more amino acids from a terminus of the by-product using an exopeptidase, such that the truncated by-product is not a substrate for the transpeptidase, thereby shifting the reaction equilibrium of the transpeptidation reaction towards the ligated product.

[0010] The inventors have developed a coupled enzyme transpeptidation system involving a transpeptidase in combination with an exopeptidase. Advantageously, the by-product is degraded by the exopeptidase, removing the by-product as a substrate for the transpeptidase. As such, the reaction equilibrium is shifted in the forward direction, favouring the ligated product and preventing the reversibility of the reaction. This approach greatly enhances the efficiency of the transpeptidation reaction, facilitating its wider application in experimental and industrial processes. In addition, the coupled enzyme transpeptidation system enables both quantitative N- and C-terminal labelling of proteins or peptides, and quantitative fusion of expressed peptides or proteins.

[0011] The method of transpeptidation is suitable for various applications, including but not limited to protein or peptide labelling, conjugating protein or peptides to therapeutic agents (e.g., antibody-drug conjugates, conjugate vaccines, conjugate biopharmaceuticals), cell-surface labelling, or in vivo labelling. In some embodiments, the method of transpeptidation is a method of protein or peptide labelling. In the method, a donor substrate is ligated to an acceptor protein or peptide using a transpeptidase. The term "ligating...using a transpeptidase", as used herein, is intended to mean the donor and acceptor are conjugated together via the formation of an amide linkage between a free amino group and a carbonyl group, catalysed by the transpeptidase.

[0012] The donor substrate may be any suitable protein, peptide, amino acid amide (e.g., glycinamide) or amino acid ester (e.g., glycyl ester) that provides an acyl donor group for the transpeptidation reaction. In some embodiments, the donor substrate is a protein, a peptide, glycinamide or a glycyl ester. In some embodiments, the donor substrate is a protein, a peptide or glycinamide. In some embodiments, the donor substrate is a protein or peptide. In some embodiments, the donor substrate is a peptide. The term "protein", as used herein, refers to a macromolecule comprising one or more polypeptides, and includes natural proteins, synthetic proteins and modified proteins (e.g., polypeptides linked to non-polypeptides, such as tagged proteins, biotinylated proteins, PEGylated proteins, click-modified proteins, proteins containing non- naturally occurring amino acids, etc). The term "peptide", as used herein, refers to a molecule comprising a chain of amino acids, and includes natural peptides, synthetic peptides and modified peptides (e.g., peptides linked to non-peptides, such as tagged peptides, peptides containing non-naturally occurring amino acids, etc). The term "amino acid amide", as used herein, refers to an amide of an amino acid, and includes the amide form of naturally occurring and non-naturally occurring amino acids, which may or may not be modified. The term "amino acid ester", as used herein, refers to an ester of an amino acid, and includes the ester form of naturally occurring and non-naturally occurring amino acids, which may or may not be modified.

[0013] The acceptor protein or peptide may be any suitable protein or peptide that provides an acyl acceptor group for the transpeptidation reaction. In some embodiments, the acceptor is an acceptor protein.

[0014] The transpeptidase may be any suitable enzyme that ligates the donor substrate to the acceptor protein or peptide through a transpeptidation reaction (i.e., catalyses the formation of an amide linkage between a free amino group and a carbonyl group). Typically, the transpeptidase cleaves a recognition motif in the acceptor protein or peptide to form an acyl-enzyme intermediate; nucleophilic attack on the acyl intermediate by an N-terminal amine of the donor substrate releases the enzyme and results in the formation of the amide linkage, forming the ligated product and a by-product. For example, the transpeptidase may be a transpeptidase enzyme (e.g., sortase, butelase, asparaginyl endopeptidase, connectase) or an enzyme engineered to have transpeptidase activity (e.g., Trypsiligase, subtiligase, peptiligase, omniligase-1). In some embodiments, the transpeptidase is a cysteine transpeptidase. In some embodiments, the transpeptidase is a sortase, a butelase or an asparaginyl endopeptidase. In some embodiments, the transpeptidase is a sortase. In some embodiments, the transpeptidase is a sortase A. In some embodiments, the transpeptidase is Staphylococcus aureus sortase or an evolved functional form thereof. In some embodiments, the transpeptidase is Staphylococcus aureus sortase A or an evolved functional form thereof. Evolved sortases are known to those skilled in the art, see for example, Podracky, C.J., An, C., DeSousa, A. et al. Laboratory evolution of a sortase enzyme that modifies amyloid-0 protein. Nat Chem Biol 17, 317-325 (2021) (hereby incorporated by reference in its entirety). In some embodiments, the transpeptidase is Streptococcus pyogenes sortase A.

[0015] The transpeptidation reaction generates a ligated product and a byproduct. The term "ligated product", as used herein, is intended to mean the product from the transpeptidation reaction which comprises the newly formed amide linkage (i.e., the conjugated donor-acceptor product). The term "byproduct", as used herein, is intended to mean a product released from the transpeptidation reaction that is not the ligated product. The by-product may also be referred to as the "leaving group". In some embodiments, the by-product is a peptide, an amino acid amide (e.g., glycinamide, N-ethyl glycinamide), an amino acid ester (e.g., a glycyl ester) or a peptide-like molecule. In some embodiments, the by-product is a peptide or glycinamide. In some embodiments, the byproduct is a peptide.

[0016] The by-product is a substrate for the transpeptidase, meaning that the transpeptidase is capable of catalysing the reverse reaction, resulting in the regeneration of the donor substrate and the acceptor protein or peptide. In the theoretical reverse reaction, the ligated product binds to the transpeptidase to form an acyl-enzyme intermediate; nucleophilic attack on the acyl-enzyme intermediate by an N-terminal amine of the by-product releases the enzyme and results in the formation of a peptide bond to re-form the donor substrate and the acceptor protein or peptide.

[0017] In the method, one or more amino acids are cleaved from a terminus of the by-product using an exopeptidase. The term "cleaving...using an exopeptidase", as used herein, is intended to mean that the exopeptidase catalyses the hydrolysis of the terminal or penultimate peptide bond of the byproduct to release one or more amino acids, forming a truncated by-product.

[0018] The exopeptidase may be any suitable enzyme that cleaves one or more amino acids from a terminus of the by-product to result in a truncated by-product that is not a substrate for the transpeptidase.

[0019] In some embodiments, one or two amino acids are cleaved from a terminus of the by-product. In some embodiments, one amino acid is cleaved from a terminus of the by-product. In some embodiments, no more than one amino acid is cleaved from a terminus of the by-product. In some embodiments, no more than two amino acids are cleaved from a terminus of the by-product.

[0020] The terminus of the by-product may be the N-terminus or the C-terminus. In some embodiments, the one or more amino acids are cleaved from the N- terminus of the by-product.

[0021] The exopeptidase may be an aminopeptidase or a carboxypeptidase. In some embodiments, the exopeptidase is a carboxypeptidase, wherein the one or more amino acids are cleaved from the C-terminus of the by-product. In some embodiments, the exopeptidase is an aminopeptidase, wherein the one or more amino acids are cleaved from the N-terminus of the by-product.

[0022] Various aminopeptidases are known to the skilled person, for example, D- aminopeptidase, aminopeptidase W, XXX-prolyl aminopeptidase, XXX-prolyl dipeptidase. In some embodiments, the aminopeptidase is a D-aminopeptidase. In some embodiments, the aminopeptidase is a glycyl aminopeptidase. In some embodiments, the D-aminopeptidase is Brucella anthropi D-aminopeptidase.

[0023] In some embodiments, the aminopeptidase is sequence-specific, optionally wherein the aminopeptidase is specific for GX2-terminated peptides, wherein X2 is a small, polar or negatively charged (i.e., acidic) amino acid, or amino acid amide (e.g., glycinamide). In some embodiments, X2 is selected from the group consisting of alanine, glycine, serine, threonine, methionine, asparagine, glutamine, tyrosine, aspartic acid and glutamic acid.

[0024] The truncated by-product is not a substrate for the transpeptidase, meaning that the truncated by-product is not capable of nucleophilic attack on the acyl-enzyme intermediate. As such, the truncated by-product cannot be used in the reverse reaction. The term "truncated by-product", as used herein, is intended to mean the final product following cleavage of the by-product by the exopeptidase. For example, a by-product having GGX at its N-terminus is cleaved by a glycyl aminopeptidase to produce a GX-terminated by-product, and cleaved a second time by the aminopeptidase to produce an X-terminated by-product. In this case, the X-terminated by-product is not a substrate for the transpeptidase.

[0025] As a result of the method, the reaction equilibrium of the transpeptidation reaction is shifted towards the ligated product. In other words, truncation of the by-product prevents the transpeptidation reaction from being under thermodynamic control, favouring aminolysis over hydrolysis. In some embodiments, the method results in at least 10-fold selectivity for aminolysis over hydrolysis. In some embodiments, the method results in at least 20-fold selectivity for aminolysis over hydrolysis. In some embodiments, the method results in at least 30-fold selectivity for aminolysis over hydrolysis. In some embodiments, the method results in at least 40-fold selectivity for aminolysis over hydrolysis. In some embodiments, the method results in at least 50-fold selectivity for aminolysis over hydrolysis. In some embodiments, the method results in at least 60-fold selectivity for aminolysis over hydrolysis. In some embodiments, the method results in at least 70-fold selectivity for aminolysis over hydrolysis. In some embodiments, the method results in at least 80-fold selectivity for aminolysis over hydrolysis. In some embodiments, the method results in at least 90-fold selectivity for aminolysis over hydrolysis. In some embodiments, the method results in at least 100-fold selectivity for aminolysis over hydrolysis. In some embodiments, the method proceeds with near equimolar concentrations of the donor substrate and acceptor peptide or protein.

[0026] In some embodiments, the N-terminus of the donor substrate comprises a motif that is not a substrate for the exopeptidase. In this case, the exopeptidase does not cleave one or more amino acids from the N-terminus of the donor substrate. In some embodiments, the N-terminus of the donor substrate comprises the motif XXi, wherein Xi is a hydrophobic or positively charged amino acid. In some embodiments, Xi is selected from the group consisting of cysteine, phenylalanine, histidine, isoleucine, lysine, leucine, proline, arginine, valine or tryptophan. In some embodiments, Xi is selected from the group consisting of phenylalanine, histidine, isoleucine, lysine, leucine, proline, arginine, valine or tryptophan. In some embodiments, the N-terminus of the donor substrate comprises the motif GXi, wherein Xi is a hydrophobic or positively charged amino acid, optionally wherein the N-terminus of the donor substrate comprises the motif GV.

[0027] The term "the N-terminus ... comprises the motif", as used herein, is intended to mean that the substrate is terminated at the amino end with the motif. For example, where the N-terminus comprises the motif GXi, the donor substrate has the glycine residue at the N-terminus, followed by the Xi residue, optionally followed by a further sequence.

[0028] Typically, the acceptor protein or peptide comprises a recognition motif that is cleaved by the transpeptidase to generate the by-product. Various transpeptidases and their recognition motifs are known to the skilled person, see for example, Jacobitz AW, Kattke MD, Wereszczynski J, Clubb RT. Sortase Transpeptidases: Structural Biology and Catalytic Mechanism. Adv Protein Chem Struct Biol. 2017;109:223-264; Morgan HE, Turnbull WB, Webb ME. Challenges in the use of sortase and other peptide ligases for site-specific protein modification. Chem Soc Rev. 2022 May 23;51(10):4121-4145, which are hereby incorporated by reference in their entirety. Examples of transpeptidases include, but are not limited to:

[0029] In some embodiments, the recognition motif comprises LPXT / G or LPXT / GX2, such that the by-product comprises glycinamide or GX2. In some embodiments, the recognition motif comprises LPET / G or LPET / GX2. In some embodiments, X2 is a small, polar or negatively charged amino acid. In some embodiments, X2 is selected from the group consisting of alanine, glycine, serine, threonine, methionine, asparagine, glutamine, aspartic acid and glutamic acid. In some embodiments, X2 is alanine.

[0030] In some embodiments, the N-terminus of the donor substrate comprises a motif which is not a substrate for the exopeptidase, and wherein the acceptor protein or peptide comprises a recognition motif that is cleaved by the transpeptidase to generate the by-product.

[0031] In some embodiments, the donor substrate comprises a protein having an N-terminus which is not a substrate for the exopeptidase and is a substrate for the transpeptidase. In some embodiments, the donor substrate consists of a protein having an N-terminus which is not a substrate for the exopeptidase and is a substrate for the transpeptidase.

[0032] In some embodiments, the donor substrate comprises a protein having an N-terminal glycine which is not a substrate for the exopeptidase.

[0033] In some embodiments, the N-terminus of the donor substrate comprises the motif GXi, wherein Xi is a hydrophobic or positively charged amino acid, and wherein: (i) the acceptor protein or peptide comprises the motif X / GX2, wherein I is the cleavage site of the transpeptidase and X2 is a small, polar or negatively charged amino acid; or (ii) the acceptor protein or peptide comprises the motif X / G-amide, wherein the amide group is -CONH2, -CONR2 or -CONHR. R represents an organic group or hydrogen.

[0034] In some embodiments, the N-terminus of the donor substrate comprises the motif GXi, wherein Xi is a hydrophobic or positively charged amino acid, and wherein the acceptor protein or peptide comprises the motif LPXT / GX2, wherein I is the cleavage site of the transpeptidase and X2 is a small, polar or negatively charged amino acid or an amide group.

[0035] Positively charged or hydrophobic amino acids include C, F, H, I, K, L, P, R, V, W and Y. Small, polar or negatively charged amino acids include A, D, E, G, M, N, Q, S and T.

[0036] In some embodiments, the transpeptidase and the exopeptidase are provided by a chimeric fusion polypeptide. For example, the chimeric fusion polypeptide may comprise transpeptidase and exopeptidase domains in tandem. In some embodiments, the chimeric fusion polypeptide further comprises a purification tag. As an example, the chimeric fusion polypeptide may further comprise a His-tag and / or a chitin-binding domain.

[0037] In some embodiments, the N-terminal and / or C-terminal ends of the acceptor protein or peptide are ligated to donor substrates. For example, a first donor substrate may be ligated to the N-terminus of the acceptor protein or peptide, and a second donor may be ligated to the C-terminus of the acceptor protein or peptide. The skilled person will recognise that any suitable donor(s) may be used and the donor substrates may be the same or distinct. In some embodiments, the by-product and the donor substrate have identical N-terminal residues. In some embodiments, the by-product and the donor substrate have glycine N-terminal residues. In some embodiments, the byproduct and the donor substrate have identical N-terminal residues, and distinct residues at the second position.

[0038] In a second aspect, there is provided a transpeptidation product obtainable from the process as described herein. The transpeptidation product may be directly obtainable from the process as described herein. In some embodiments, the transpeptidation product is further modified by subsequent chemical or enzymatic reactions.

[0039] The above description in relation to the method of transpeptidation is equally applicable to this aspect. The transpeptidation product may also be referred to as the ligated product.

[0040] In a third aspect, there is provided a chimeric fusion polypeptide comprising a transpeptidase domain and an exopeptidase domain. The above description in relation to the method, and in particular the chimeric fusion polypeptide, is equally applicable to this aspect.

[0041] In some embodiments, the transpeptidase domain comprises a sortase A domain. In some embodiments, the exopeptidase domain comprises a D- aminopeptidase domain. In some embodiments, the transpeptidase domain comprises a sortase A domain, and the exopeptidase domain comprises a D- aminopeptidase domain.

[0042] In a fourth aspect, there is provided a method of transpeptidation comprising: ligating a donor protein or peptide to an acceptor protein or peptide using a sortase, wherein the donor protein or peptide comprises GXi at its N- terminus, wherein Xi is a hydrophobic or positively charged amino acid, and wherein the acceptor protein or peptide comprises a LPXTGX2 motif, wherein X2 is a small, polar or negatively charged amino acid, and wherein the ligation reaction generates a ligated product comprising a LPXTGXi motif and a by-product comprising GX2 at its N terminus, cleaving one or more glycine residues from the N-terminus of the by-product using a D-aminopeptidase, such that the truncated by-product is not a substrate for the transpeptidase, thereby shifting the reaction equilibrium of the transpeptidation reaction towards the ligated product.

[0043] The above description in relation to the method according to the first aspect is equally applicable to this aspect.

[0044] Where applicable or not specifically disclaimed, any one of the embodiments described herein are contemplated to be able to be combined with any other one or more embodiments, even though the embodiments are described under different aspects of the disclosure.

[0045] The foregoing and other objects, features and advantages of the disclosure will become more apparent from the following detailed description, which proceeds with reference to the accompanying figures.

[0046] Brief Description of the Drawings

[0047] The invention will now be described in detail, by way of example only, with reference to the figures.

[0048] Figure 1. Peptide and protein labelling reactions catalysed by sortase. A. General reaction scheme for transpeptidation. Sequence specific recognition. Attack of a catalytic cysteine residue on the peptide change yields a thioacyl intermediate which can cleaved by either aminolysis (to yield a new peptide) or hydrolysis. B. Sortase (SaSrtA) catalyses the transpeptidation of a LPETGX recognition motif N-terminal glycine-containing peptides and proteins in an equilibrium fashion. C. Strategy for perturbing equilibria of N-terminal labelling reactions by the use of depsipeptide (ester) substrates. D. Strategy investigated in this study to disrupt these equilibria and yield quantitative labelling. The addition of D-aminopeptidase leads to sequence-specific removal of N-terminal glycines enabling enhanced labelling using both sortase with broader substrate specificity as well as unmodified sortases controlled by the identity of the aminoacid residue following the N-terminal glycine.

[0049] Figure 2. Screening of reaction conditions to enable quantitative labelling.

[0050] A. Initial reaction scheme for screening of combined activity of SpSrtA with DAP.

[0051] B. Reaction scheme for screening 6 of combined activity of SrtA / 7M and DAP using peptide substrates. C. Transpeptidation of model peptides by SrtA in the presence (black) and absence (grey) of D-aminopeptidase analysed by HPLC. K* = dansyl lysine. In the presence of DAP, complete conversion to the product peptide is observed.

[0052] Figure 3. Screening of reaction conditions to enable quantitative labelling. A. Example of reaction MS analysis where nucleophilic peptide substrate is not a substrate for DAP (e.g. GV-6 terminated). In the presence of DAP (black), the reaction equilibrium shifts away from the peaks corresponding to unlabelled PanZ towards the labelled species with a small increase in the hydrolysed species. Additional peaks marked with * are mass adducts of PanZ see Figure 7 for more details. B. Screening of peptide substrates in the presence and absence of DAP enables identification of suitable reaction pairs. 200 pM PanZ-LPETGAH6 was incubated with 400 pM of a model labelling peptide with sequence GXSKYG in the presence of 10 pM Srt7M with and without 20 pM DAP. Green = labelled, blue = unlabelled, red = hydrolysis. C. Time-dependence of C-terminal labelling of the pentameric CTB using 2 equivalents of a labelling peptide. Green = I D. Peptide concentration dependence of C-terminal labelling of PanZ (200 pM) and CTB (200 pM) after 2 hours using 50 pM Srt7M, 20 pM DAP and the indicated ratio of labelling peptide. After 4 hours, the proportion of hydrolysis significantly increased (see Figure 8). Green = labelled, Blue 17 = unlabelled, Red = hydrolysis. E. MS analysis of C-terminal labelling of CTB using 1.5 and 2 equivalents of labelling peptide as shown in D.

[0053] Figure 4. Optimisation of conditions for N-terminal labelling of proteins and comparison with the use of depsipeptide substrates at 25 °C. A. ESMS analysis of time-dependent labelling of 100 pM GVG-CTB by Srt7M (2 pM) with 1.2 eq (FITC)-peptide with (black) and without (grey) DAP (10 pM). B Comparison of N-terminal labelling of GVG-CTB labelling with Srt7M and a fluorescent peptide in the presence (black) and absence (grey) of DAP after 2 hours. C Time-dependent labelling of 100 pM GVG-MBP labelling using Srt7M (5 pM) and 3 equivalents of a peptide substrate in the presence (black) and absence (grey) of DAP (10 pM), or with a depsipeptide substrate (blue). D MS analysis of MBP labelling using 3 equivalents of labelling reagents after 2 hours shows almost complete conversion to labelled product.

[0054] Figure 5. Optimization of protein labelling using chimeric SrtA-D- aminopeptidase constructs SrtH7D and SrtCH7D at 25°C and 4°C and demonstration of enhanced labelling using SrtLPXSG / DAP. A Time course of C- terminal labelling of 100 pM CTB using 25 pM SrtH7D and 200 pM peptide at 25°C B. MS analysis of labelling of 100 pM CTB using 25 pM SrtCH7D after 4 h. C Peptide substrate concentration dependence of labelling of 100 pM MBP-LPETGA construct after 4 h at 25°C. (green = labelled, blue = unlabelled, red = hydrolysis) D Application of SrtH7D to N-terminal labelling of GVG-MBP (black) is equivalent to the use of depsipeptide substrates (blue) E MS analysis of sequential SrtH7D-dependent C-terminal labelling, TEV cleavage and SrtLPXSG / depsipeptide-dependent labelling of 127. i. MS of initial construct, ii. MS analysis of product of C-terminal modification with SrtH7D and peptide GVSKYG. iii. MS analysis of TEV-cleavage, iv. MS analysis of product of N-terminal modification with SrtLPXSG and depsipeptide GABA-YLPESoGG F Time course of N-terminal labelling of 100 pM CTB using 105 pM labelling peptide and 2 pMSrt H7D at 4°C. Inset: MS analysis of labelling at 10 h. G Labelling of 100 |jM GVG- CTB using 5 |jM SrtLPXSG with (black) and without (grey) DAP at 4°C H SDS- PAGE analysis of protein fusion between an affimer (nominal 50 pM) with a C- terminal LPETGA motif and GVG-MBP (50 pM) using SrtH7D (5 pM) at 4°C.

[0055] Figure 6. Initial test reaction used for assay of Srt7M / DAP-dependent ligation to PanZ. A Reaction scheme B. Exemplar data for labelling of 200 pM PanZ using Srt7M with a GVSKYG test peptide in the presence (black) and absence of DAP after overnight incubation at 37 °C (reproduced from figure 3A). In the presence of Srt 7M the peaks corresponding to PanZ-LPETGA (c. 16200) shift to lower mass (c. 15000) corresponding to the labelled protein. In the presence of dAP a small amount of the hydrolysed species (c. 15000 Da) is observed. The additional peak at c. +100 Da is due to an unknown covalent adduct contaminating the protein used for initial screening. C. Exemplar data for labelling of 200 pM PanZ using a peptide which is also a substrate for DAP (GASKYG). In the presence of DAP only hydrolysis is observed.

[0056] Figure 7. Screening reaction of 200 pM PanZ-LPETGA with 400 pM peptide with sequence GXSKYG catalyzed by Srt7M (5 pM) in the presence and absence of D-aminopeptidase (10 pM) after overnight (~ 16 hour incubation) at 25 °C.

[0057] Figure 8. pH-dependence of C-terminal labelling of 200 pM PanZ-LPETGA with 400 pM peptide GVSKYG catalysed by 50 pM Srt7M in the presence of 20 pM D-aminopeptidase after 4 hours at 25 °C. Maximal labelling is only obtained above pH 8.0.

[0058] Figure 9. Dependence of labelling on peptide ratio. Labelling of 200 pM PanZ-LPETGA catalysed by 50 pM Srt7M, 20 pM D-aminopeptidase and the indicated ratio of a GVSKYG peptide after 4 hours at 25 °C.

[0059] Figure 10. Null effect of increasing DAP concentration. Labelling of 200 pM PanZ-LPETGA using 50 pM Srt7M and 400 pM GVSKYG after 2 hours at 37 °C and pH 8.0 with the indicated ratio of D-aminopeptidase (20 pM or 80 pM final concentration).

[0060] Figure 11. Exploration of the temperature and Srt7M concentration dependence of labelling using Srt7M at 25 °C and 37 °C. Only marginal differences in product distribution were observed with variation in temperature and catalyst concentration. (All reaction 200 pM PanZ-LPETGA and 20 pM D- aminopeptidase with indicated relative ratios of Srt7M and GVSKYG peptide substrate).

[0061] Figure 12. Comparative turnover of PanZ labelling reactions in the presence of 20% (w / v) co-solvents. All reactions consisted of 200 pM protein substrate with 400 pM labelling peptide, 50 pM Srt7M and 20 pM D- aminopeptidase. DMSO: dimethylsulfoxide, GLY glycerol PG propylene glycol, DMA: dimethyl acetamide. DMF dimethylformamide. Red - hydrolysed, Green - labelled, Blue - unlabelled.

[0062] Figure 13. C-terminal labelling of CTB-LPETGA after 4 hours shows near complete consumption of starting material and increased hydrolysis. All reactions consist of 200 pM CTB-LPETGA in the presence of the indicated molar equivalents of a GVSKYG peptide, 30 pM Srt7M and 20 pM D-aminopeptidase (red - hydrolysed, green - labelled, blue - unlabelled).

[0063] Figure 14. A Comparative time-course of labelling of 100 pM GVG-CTB using 2 pM Srt7M and 120 pM depsipeptide with sequence AYLEPToGG (blue), or 2 pM Srt7M, 10 pM DAP and 120 pM peptide with sequence AYLEPTGG (black), respectively. B Comparative labelling of 100 uM GVG-CTB using 1.05, 1.2 and 2 equivalents of peptides and depsipeptide substrates and 2 pM Srt7M. Red Peptide but no DAP, Black Peptide plus 10 pM DAP, Blue Depsipeptide substrate).

[0064] Figure 15. Time-course of labelling of 100 pM GVG-CTB using 2 pM SrtA. 10 pM DAP and 1.2 (blue), 1.5 (brown) and 2 equivalents (green) of a AYLPETGG peptide analysed by ESMS.

[0065] Figure 16. Comparative C-terminal labelling of 200 pM PanZ-LPETGA using 25 pM Srt7M + 10 pM DAP (Normal), 25 pM SrtH7D and 25 pM SrtCH7D and 200 pM labelling peptide at 25 °C and pH 8.0. Red - hydrolysed. Green - labelled, Blue - unlabelled.

[0066] Figure 17. C-terminal labelling of 100 pM MBP-LPETGA-H6 with 25 pM SrtH7D and indicated ratio of labelling peptide GVSEYG after 4 hours.

[0067] Figure 18. C-terminal labelling of 100 pM MBP-LPETGA-H6 with 25 pM SrtH7D and indicated ratio of labelling peptide GVSEYG after 4 hours.

[0068] Figure 19. Peptide substrate concentration dependence of N-terminal labelling of 95 pM GVG-CTB with the indicated concentration of peptide AYLPETGG-CO2H using 2 pM SrtH7D (Hashed), 2 pM Srt7M + 2 pM DAP* (black) and 2 pM Srt7M (grey) after 2 hours at 25 °C. (*samples for 1.7 and 2.28 equivalents use 10 pM DAP). All data points represent single experiments except where error bars shown.

[0069] Figure 20. Peptide and catalyst concentration dependence of C-terminal labelling of 127 using SrtH7D. A Exploration of SrtH7D and peptide substrate concentration of labelling 100 pM I27-LPETGA after 3 hours. B Direct comparison of labelling of 100 pM I27-LPETGA using 10 pM SrtH7D and 100-200 pM peptide GVSKYG after 3 hours. Error bars represent 3 independent measurements. Figure 21. Time course of labelling reaction of 100 pM GVG-CTB with 105 pM labelling peptide AYLPETGG-CO2H and 5 pM SrtH7D at pH 8 and indicated temperature and times.

[0070] Figure 22. Time-dependence of C-terminal labelling of 100 pM GVG-MBP using 25 pM SrtH7D and 200 pM of labelling peptide GVSEYG at 4 °C.

[0071] Figure 23. Peptide concentration (1.1 - 3 equivalents) and timedependence (24 h - 7 days) of / V-terminal labelling of 100 pM GVG-CTB using 10 pM SrtLPXSG and labelling peptide YALPESGAG with and without 10 pM DAP at 4 °C.

[0072] Figure 24. Protein fusion between Aff-LPETGA and GVG-MBP at 4 °C catalyzed by A. 1% SrtH7D, B 2% SrtH7D.

[0073] Figure 25. Protein fusion of GVG-MBP and MBP-LPETGA to form an MBP dimer catalyzed by SrtH7D. Reaction mixture consists of 50 pM MBP-LPETGA, 75 pM GVGK-MBP and 12.5 pM SrtH7D at pH 8, 4 °C.

[0074] Figure 26. Characterisation data for MBP-DAP. Upper left SDS-PAGE analysis of purification by amylose affinity chromatography Upper right Chromatogram for purification by size-exclusion chromatography. Bottom Deconvolved mass spectrum of purified protein analysed by ESMS.

[0075] Figure 27. Characterization data for Srt7M. Upper left Purification of Srt7M by Ni-NTA affinity chromatography. Upper right Purification of SrtH7D by size-exclusion chromatography (Superdex 75 26 / 60 column, isocratic elution with 50 mM Tris, 100 mM NaCI pH. 8.0). Bottom ESMS characterisation of purified SrtH7D confirms identity.

[0076] Figure 28. Characterization data for SrtH7D. Upper left Purification of SrtH7D by Ni-NTA affinity chromatography. Upper right Purification of SrtH7D by size-exclusion chromatography (Superdex 200 16 / 60 column, isocratic elution with 50 mM Tris, 100 mM NaCI pH. 8.0). Bottom ESMS characterisation of purified SrtH7D confirms identity.

[0077] Figure 29. Upper left Purification of SrtCH7D by Ni-NTA affinity chromatography Upper right Purification of SrtCH7D by size-exclusion chromatography (Superdex 200 16 / 60 column, isocratic elution with 50 mM Tris, 100 mM NaCI pH. 8.0). Pure SrtCH7D is obtained in the fractions eluting between 65 and 75 mL. Bottom ESMS characterisation of purified SrtCH7D confirms identity.

[0078] Figure 30. ESMS analysis of the purified PanZ-LPETGA construct showing expected mass. Figure 31. Purification of unfolded CTB-LPETGA-H6 by Ni-NTA affinity chromatography.

[0079] Figure 32. Characterization of refolded CTB-LPETGA Upper left SDS- PAGE analysis of CTB-LPETGA prior to and after size-exclusion chromatography. Unboiled samples of CTB-LPETGA maintain their pentameric structure on the SDS- PAGE gel confirming multimerization. Upper Right Size-exclusion chromatography purification of CTB-LPETGA. (Superdex 75 26 / 60 column). Bottom ESMS characterisation of purified CTB-LPETGA confirms identity.

[0080] Figure 33. Upper left Purification of MBP-LPETGA by Ni-NTA affinity chromatography. Upper right Purification of MBP-LPETGA by size-exclusion chromatography (Superdex 75 26 / 60) Bottom ESMS characterisation of purified MBP-LPETGA confirms identity.

[0081] Figure 34. Characterisation data for purification of GVG-MBP Upper left SDS-PAGE analysis of purification of protein by amylose affinity chromatography. Upper right Chromatogram for size-exclusion chromatography of protein (Superdex 75 26 / 60). Bottom ESMS analysis of protein confirms identity of GVG- MBP.

[0082] Figure 35. Characterization data for GVG-CTB Upper left SDS-PAGE analysis of purification of protein by Ni-NTA affinity chromatography. Upper right Chromatogram for size-exclusion chromatography of protein (Superdex 75 16 / 60). Bottom ESMS analysis of protein confirms identity of GVG-CTB.

[0083] Figure 36. ESMS Characterization data for Affimer-LPETGA before and after iodoacetamide capping.

[0084] Figure 37. Characterisation data for (TEV)-GVG-I27-LPETGA-H6. Upper left SDS-PAGE analysis of Ni-NTA affinity chromatography purification of protein. Upper right Chromatogram for size-exclusion chromatography of protein (Superdex 75 26 / 60). Bottom ESMS analysis of protein confirms identity of (TEV)-GVG-I27-LPETGA-H6.

[0085] Figure 38. Exemplar analytical HPLC data.

[0086] Figure 39. Unoptimised N-terminal labelling using SrtH7D at low protein substrate concentrations. Reaction mixtures consist of 10 mM GVG-CTB incubated with the indicated concentration of substrate peptide AYLPETGG at pH 8, 4 °C.

[0087] Figure 40. Unoptimised C-terminal labelling using SrtH7D at low protein substrate concentrations. Reaction mixtures consist of 10 mM CTB-H6 incubated with the indicated concentration of substrate peptide GVSKYG and SrtH7D at pH 8, 4 °C for 4 and 24 hours. Extent of labelling analysed by SDS-PAGE and ESMS. Green labelled, blue unlabelled, red hydrolysis product. Figure 41. Percentage N-terminal labelling of protein constructs determined by ES-MS. All experiments correspond to incubation of 100 mM substrate protein with 2 mM SrtH7D, 150 mM labelling peptide AYLPETGG at 4 °C for 24 hours.

[0088] Figure 42. Exemplar data for kinetic characterisation of SrtH7D using a fluorescence unquenching based peptide assay for transpeptidase activity. 50 pM Abz-GALPETGAK(DNP)-NH2, 500 nM SrtH7D, 50 mM Tris pH 8.0 (@ 25 °C), 100 mM NaCI, was incubated at 25 °C with the indicated concentration of substrate peptide. (Abz 2-aminobenzoyl, DNP 2,4-dinitrophenyl).

[0089] Figure 43. Comparison of SrtH7D to individual components in a peptide- based assay. 50 pM Abz-GALPETGAK(DNP)-NH2, 500 nM catalyst (SrtH7D, Srt7M or Srt7M+DAP), 50 mM Tris pH 8.0 (@ 25 °C), 100 mM NaCI, were incubated at 25 °C with the indicated concentration of substrate peptide. Srt7M - blue triangle, Srt7M+dAP - red circle, SrtH7D - black square. All data is fitted to the Michaelis- Menten equation for a single substrate.

[0090] Figure 44. Stability to storage of SrtH7D stored at 4 °C (M-P). -20 °C (Q- T) and -80 °C (U-X) in the presence of increasing concentrations of glycerol (0,10,20,40% respectively) determined by peptide-based assay. Similar activity is observed in all samples after 1 and 5 weeks storage except for samples stored with 40% glycerol.

[0091] Figure 45. Comparison of activity of SrtH7D after storage at -80 °C for ~1 year and ~3 years. 50 pM Abz-GALPETGAK(DNP)-NH2, 500 nM SrtH7D, 50 mM Tris pH 8.0 (@ 25 °C), 100 mM NaCI, were incubated at 25 °C with the indicated concentration of substrate peptide. Four samples of SrtH7D were compared (freshly purified - blue triangle, ~ 1 year old - red circle and black square, and ~ 3 years old - green triangle). All data is fitted to the Michaelis-Menten equation for a single substrate.

[0092] Figure 46. Catalyst, temperature, and peptide equivalent dependence of C-terminal labelling of CTB-LPETGAH6 using a combination of Srt5M and D- aminopeptidase determined by mass spectrometry after 24 (top) and 48 hours (bottom). Reactions (100 pM CTB-LPETGAH6, Catalyst Srt5M, Damp (D- aminopeptidase), peptide GVSEYG and temperature as shown. Optimal labelling is seen at the lowest concentrations of Srt5M and temperature after 48 hours with little variation as a result of peptide or D-aminopeptidase concentration.

[0093] Figure 47. Exemplar data for labelling of proteins using SrtH5D. Top Labelling of MBP-LPETGAH6 using equimolar labelling peptide. 200 pM MBP- LPETGAH6 was incubated with 200 pM GVSEYG and 20 pM SrtH5D for 30 hours at 4 °C. Bottom Initial screening data for labelling of proteins using increased equivalents of peptides. 100 pM MBP-LPETGAH6 was incubated with 2.5 LIM catalyst and the indicated ratio of labelling peptide at 4 °C for 30 hours or 65 hours

[0094] Description

[0095] The inventors have developed a coupled enzyme transpeptidation system involving a transpeptidase in combination with an exopeptidase. Advantageously, the by-product is degraded by the exopeptidase, removing the by-product as a substrate for the transpeptidase. As such, the reaction equilibrium is shifted in the forward direction, favouring the ligated product and preventing the reversibility of the reaction. This approach greatly enhances the efficiency of the transpeptidation reaction, facilitating its wider application in experimental and industrial processes. In addition, the coupled enzyme transpeptidation system enables both quantitative N- and C-terminal labelling of proteins or peptides.

[0096] In the following example, the inventors demonstrate that an aminopeptidase may be used in combination with a sortase to shift the transpeptidation reaction towards the ligated product. However, it is anticipated that the invention is more widely applicable to a combination of a transpeptidase with an exopeptidase.

[0097] Examples

[0098] Examole 1

[0099] Materials and methods

[0100] Protein overexpression and purification

[0101] MBP-DAP A synthetic oligonucleotide coding for Ochrobacterium anthropii was obtained from Genscript and subcloned into pNIC-MBP to obtain an expression construct for N-terminal His6-MBP-tagged protein under the control of the T7 promoter. Following transformation into E. coli BL21 (DE3), cells were inoculated into 400 mL LB media supplemented with kanamycin (50 pg mL-1) and the media incubated at 37°C until ODeoo = 0.6 at which point protein expression was induced with IPTG (0.5 mM final concentration). Cell pellets were isolated by centrifugation (10000g, 10 min) and resuspended in 100 mM potassium phosphate, 0.1 mM EDTA, 5 mM p-mercaptoethanol, pH 7.0 before lysis by sonication. The cleared lysate was purified by automated Ni-NTA affinity / size- exclusion chromatography using a BioRad NGC system. Briefly, the cleared lysate was applied to a 5 mL Qiagen Histrap FF column, the column washed with 50 mM potassium phosphate, 100 mM NaCI, pH 7.0 buffer supplemented with 50 mM imidazole and eluted with the same buffer supplemented with 500 mM imidazole. The eluted protein fraction (10 mL) was directly applied to a Superdex 200 (26 / 60) column and isocratically eluted with 50 mM potassium phosphate, 100 mM NaCI pH 7.0 to yield the pure protein. Protein identity was confirmed by SDS- PAGE and ESMS (expected mass 100363 Da, observed 100379 Da) and quantified by UV absorbance at 280 nm (Etheor = 147710 M'1cm ).

[0102] SrtLPXSG and SaSrtA were overexpressed and purified as described in Morgan et al. (22).

[0103] SpSrtA An expression construct for SpSrtA (23) was obtained from Addgene (51139) and transformed into E. coli BL21 (DE3). Protein was overexpressed via IPTG induction, cells isolated resuspended (50 mM Tris, 150 mM NaCI, 10 mM CaCIz, pH 7.5), lysed via mechanical disruption (Constant Systems cell disrupter) at 20 kpsi and the cleared lysate was purified via Ni-NTA affinity chromatography. Protein containing fractions were dialysed against lysis buffer and used as isolated.

[0104] Srt7M. A synthetic gene coding for Srt7M (24) was subcloned from into pETlla as previously described for SrtLPXSG (22) to obtain a C-terminal His- tagged protein. Protein was purified using the methods described for SrtLPXSG in Morgan et al. but without CaCI2 in buffers. Protein identity was confirmed by SDS-PAGE and ESMS (expected mass 17851 Da, observed 17851 Da) and quantified by UV absorbance at 280 nm (etheor = 17440 M-1cm4).

[0105] SrtH7D. Synthetic gene constructs for SrtCH7 consisting of an N-terminal His-tag and chitin-binding domain prior to the coding sequence for Srt7M with an additional C-terminal MCS were obtained from Genscript and subcloned between the Xbal and Xhol sites of pET28a. The coding sequence for the chitin-binding domain was removed via sequential double digestion with EcoRI and Mfel, and religation to obtain SrtH7 as described by Dolan et al. (25) The coding sequence for DAP was subcloned between the Ndel and BamHI sites of the SrtH7 construct. Protein was overexpressed and purified as previously described for Srt7M using either sequential or automated sequential Ni-NTA affinity / size-exclusion chromatography. Protein identity was confirmed by SDS-PAGE, ESMS (expected mass 75923 Da, observed 75923 Da) and quantified by UV absorbance (etheor = 94935 M'1cm4).

[0106] SrtCH7D. The expression construct for SrtCH7D was prepared as for SrtH7D but without removal of the Chitin Binding domain coding sequence. Protein was overexpressed and purified as for SrtH7D. Protein identity was confirmed by SDS-PAGE, ESMS (expected mass 81888 Da, observed 81888 Da) and quantified by UV absorbance (etheor = 116030 M'1cm'1)

[0107] PanZ-LPETGA. The expression construct for PanZ-LPETGAH6 was obtained via two-step cloning from pBAD24-Yhhk (26). Initially the coding sequence for PanZ (also known as YhhK) was amplified by PCR before subcloning of a Nhel / Xhol fragment into the Xbal and Xhol sites of pET28a to generate an expression construct for PanZ with a C-terminal SLPETGGLEH6 addition. Subsequently Quikchange site-directed mutagenesis was used to generate the C- terminal sequence SLPETGALEH6. Protein was overexpressed and purified as described previously for the unmodified C-terminal his-tagged protein. Protein identity was confirmed by SDS-PAGE and ESMS (expected mass 16224 Da, observed 16224 Da) and quantified by UV absorbance at 280 nm (etheor = 26470 M'1cm-1).

[0108] CTB-LPETGA-H6. The expression construct for inclusion-body-based expression of CTB-LPETGA-H6 was generated by mutagenesis of the expression construct for soluble expression of CTB-LPETGA (49) (itself a derivative of pMalp5x). Briefly, site-directed mutagenesis was used to remove the coding sequence for the LTIIb signal peptide, insert an initiator methionine codon and insert the sequence for a C-terminal Ser-His6 sequence. The resulting plasmid was transformed into E. coli BL21 (DE3). LB media was inoculated with transformed cells and incubated at 37°C until OD600 = 0.6, protein overexpression was induced (0.5 mM IPTG) and media incubated overnight (~16 hours) at 25°C). Cells were isolated by centrifugation, resuspended in 50 mM sodium phosphate 300 mM NaCI buffer, pH 8.0 and lysed via mechanical disruption. The cell debris were isolated by centrifugation and the pellet resuspended in 100 mM NaHzPC , 100 mM Tris, 8M Urea, pH 8.0 (resuspension buffer) before re-centrifugation. The cleared solution was applied to a Ni-NTA affinity column (gravity fed) and the column washed with resuspension buffer supplemented with 25 mM imidazole and 0.1% v / v Triton-X (Figure 31, Wash 1) and resuspension buffer supplemented with 25 mM imidazole (Wash 2) before elution in resuspension buffer supplemented with 400 mM imidazole.

[0109] MBP-LPETGA-H6. The expression construct for MBP-LPETGA-H6 was obtained via site-directed mutagenesis of plasmid pMalc5x to insert an oligonucleotide coding for a C-terminal SSNLPETGASH6* sequence into the linker region of the pMalc5x plasmid. Protein was overexpressed in E. coli BL21 using standard protocols for IPTG-induced overexpression. Isolated cells were resuspended in 50 mM Tris, 150 mM NaCI pH 8.0, mechanically lysed and the cleared lysate purified by either Ni-NTA (eluting with 250 mM imidazole) or amylose (eluting with 100 mM glucose) affinity chromatography prior to subsequent further purification via size exclusion chromatography (Superdex 200 26 / 60 with isocratic elution in 50 mM Tris, 150 mM NaCI pH 8.0). Protein identity was confirmed by SDS-PAGE and ESMS (expected mass 42336 Da, observed 42333 Da) and quantified by UV absorbance at 280 nm (Etheor = 66350 M'1crrrx).

[0110] GVG-MBP. The expression construct for GVG-MBP was constructed via mutagenesis of pMalc5x to insert nucleotides encoding a GVGK tetrapeptide following the initiator methionine codon and insertion of a stop codon prior to the region coding the polyasparagine linker. Following transformation into E. coli BL21, cells were cultured at 37°C and protein overexpression induced with IPTG (1 mM) at OD600 before continued growth for 3-4 hours. Cells were isolated by centrifugation, resuspended in 50 mM Tris, 150 mM NaCI pH 7.6 and mechanically lysed (constant systems cell disruptor). The cleared lysate was purified via amylose affinity chromatography (eluting with 10 mM maltose) and size-exclusion chromatography, identity confirmed by SDS-PAGE and HRMS (expected mass 40449 Da, observed 40448 Da) and protein concentration estimated using UV absorbance (etheor = 66350 M'1cm ).

[0111] GVG-CTB. The overexpression construct for GVG-CTB was generated via site-directed mutagenesis of an existing construct for overexpression of GGG-CTB proceeded by LTBII leader sequence for periplasmic export in the pMal backbone. Following transformation into E. coli BL21, cells were cultured in LB media and protein overexpression induced with IPTG. Culture media was cleared by centrifugation (10000g, 10 min) and protein isolated via precipitation with ammonium sulfate (570g / L) and centrifugation (10000g, 25 min). Following resuspension in 50 mM Tris, 150 mM NaCI pH 7.6, protein was purified by sequential Ni-NTA affinity chromatography (using native affinity of the CTB for the resin) and size-exclusion chromatography on Superdex 75 with 50 mM HEPES, 100 mM NaCI pH 8.0. Protein identity was confirmed by SDS-PAGE and ESMS (expected mass 11856 Da, observed 11856 Da) and quantified by UV absorbance at 280 nm (etheor = 11585 M'1cm4).

[0112] (TEV)-GVG-I27-LPETGA-H6. An expression construct for the Titin 127 domain with an N-terminal TEV-cleavage sequence and C-terminal LPETGA motif was obtained from Genscript subcloned into pET28a. Protein was overexpressed from E. coli BL21 (DE3) and purified by sequential Ni-NTA affinity and size- exclusion chromatography (Superdex 75 26 / 60). Protein identity was confirmed by SDS-PAGE and ESMS (expected mass 12627 Da, observed 12626 Da) and quantified by UV absorbance at 280 nm (etheor = 8605 M'1cm ).

[0113] Chemical experimental

[0114] Peptide and depsipeptide synthesis. All peptides were synthesized using standard protocols for peptide synthesis either manually or using a Liberty Blue synthesizer. Peptide synthesis was initiated on Glycine-Chlorotrityl resin or Rink Amide resin using the Oxyma Pure strategy. Peptide purity was assessed by HPLC and LCMS. Where required, peptides were purified by preparative HPLC.

[0115] Peptide labelling assay. Peptides (500 pM) were combined with SrtA (50 pM) and MBP-DAP (5 pM) in 50 mM Tris.HCI pH 8.0 supplemented with 1 mM CaCl2 and incubated at 37°C. Samples were analysed by LC-MS - curves shown represent total absorbance between 190 and 650 nm.

[0116] Protein labelling assays. Protein and peptide concentration were determined by UV absorbance at 280 nm. Unless otherwise stated standard labelling reactions were conducted in 50 mM Tris, 100 mM NaCI, pH 8.0 (corrected for the incubation temperature) incubated in an Eppendorf Thermomixer. Reaction samples were submitted for direct analysis by LC-MS or quenched with 50% MeCN:H2O and stored at 4°C prior to analysis.

[0117] Protocol for double labelling of 127 using SrtH7D and SrtLPXSG. GVSKYG (200 pM, 2 eq.) was mixed with TEV-GVG-I27-LPETGAH6 (100 pM) in PBS buffer (pH 8.0). To the reaction was added H7D (25 pM, 25 mol%) for a total volume of 500 pL. The reaction mixture was incubated at 25°C for 2.5 hr and applied to a short nickel column (~ 3 mL nickel-NTA resin) to remove excess sortase and washed with PBS buffer. The eluant and wash fractions were concentrated and washed with PBS to remove excess peptide via diafiltration in a spin filter (Amicon 3 kDa MWCO). The concentration of the C-terminally labelled protein was determined using an extinction coefficient of 10095 M-1cm4). TEV cleavage was performed by mixing C-terminally labelled 127 (TEV-GVG-I27- LPETGVSKYG) (100 pM) with his-tagged TEV protease (20 pM, 20 mol%) and shaking for 1 hour at RT. Excess protease was removed via nickel chromatography and the eluant was concentrated via diafiltration. The concentration of the TEV cleaved protein was determined using an extinction coefficient of 8605 M'1cm4. N-terminal labelling of GVG-I27-LPETGVSKYG (100 pM) was performed by adding GABA-YLPESoGG (500 pM, 5 eq.) in PBS buffer containing 1 mM CaCl2. SrtA(LPXSG) (20 pM, 20 mol%) was added and the reaction mixture was incubated at 37°C for 2 hr. Each step of the reaction was analyzed via HRMS.

[0118] Results

[0119] Identification of a transpeptidase / aminopeptidase pair

[0120] The inventors sought to identify a second enzyme which would selectively degrade a particular peptide by-product to remove it from the reaction equilibrium. The initial plan was to exploit the substrate promiscuity of sortase A from Streptococcus pyogenes (SpSrtA) which is reported to recognize both LPXT / G and LPXT / A as substrate motifs (2). The inventors postulated that inclusion of an enzyme which selectively degrades or modifies a peptide with an N-terminal glycine residue would promote labelling of proteins using a substrate peptide with an N-terminal alanine residue. For the second enzyme, the inventors identified D-aminopeptidase (DAP) from Ochrobacterium anthropii since this is known to act on both peptides containing N-terminal D-amino acid and peptides with N-terminal glycine residues whereas L-amino-acid-terminated peptides are not substrates (10-12). The product of cleavage is a free amino-acid and the truncated peptide, and the inventors have previously shown that while glycinamide and peptides terminated with a single glycyl residue are substrates for sortase, glycine is not (4).

[0121] The inventors obtained recombinant DAP via overexpression as an N- terminal MBP-His-tagged fusion protein and initially tested ligation using peptide substrates with recognition sequences LPETGGE and H2N-AAS and a combination of DAP and SpSrtA (Figure 2A). While they observed formation of the desired LPETAA-containing product, the conjugation reaction was slow, and they consistently observed hydrolysis of the substrate LPETGGE peptide to yield LPET at a similar rate as aminolysis to form the desired labelled product. This is presumably due to the low concentrations of the AAS co-substrate peptide used in this approach compared to previous reports of the use of SpSrtA (2, 13-15). While the hydrolysis activity of SpSrtA prevented efficient labelling, the inventors also noted an unanticipated by-product (LPETGE) in which a single glycine residue had been removed from the initial substrate, presumably following release of GGE by the sortase, removal of glycine by DAP and religation to the sortase-acyl enzyme intermediate. This observation suggested that DAP activity might be affected by the succeeding residues since the partially degraded product peptide must be present at a sufficient concentration to enable the reverse reaction. Such selectivity by DAP has previously been observed for D-Ala-terminated peptide substrates but not fully characterised for all amino-acids (10). The inventors therefore incubated DAP (10 mol%) with a library of glycine-terminated peptides with the sequence GXSKYG at 25°C for 2 hours. Broadly, the inventors observed partial or complete hydrolysis of glycine residues from peptides where the second residue was small, polar or negatively charged (i.e., A, D, E, G, M, N, Q, S or T) but not from those peptides where the second residue was positively charged or hydrophobic (i.e., C, F, H, I, K, L, P, R, V, W or Y). A subsequent screen to investigate non-natural amino-acids indicated that the same patterns were observed - in this case the inventors were able to detect hydrolysis of the Gly residue with a variety of D-amino acids (a, f, I, m or v), linear non-natural amino acids such as norvaline and aminobutyric acid and norleucine but observed no hydrolysis with ornithine, o-methylalanine and N-methyl alanine.

[0122] The inventors hypothesized that they would be able to use the selectivity of DAP for the second residue in its substrate peptide to direct ligation of glycine- terminated substrate peptides. In other words, they could use peptide sequences LPETGXi and GX2 such that, while the GXi product peptide is a substrate for DAP, GX2 is not and the equilibrium product will therefore be a peptide with sequence LPETGX2 (Figure ID). They therefore investigated ligation of available peptide substrates with the C-terminal sequence LPETGGS and N-terminal sequence H2N- GV but this time using both SaSrtA and the Ca-independent SaSrtA 7M heptamutant (Srt7M) as a catalyst (1, 16) (Figure 2B). In this case, addition of DAP to the reaction mixture led to a shift towards formation of the desired product after 3 hours (Figure 2C). This demonstrated that the system does catalyse selective product formation as desired.

[0123] Rather than optimising the reaction at high peptide substrate concentrations, the inventors immediately moved to investigate labelling of a model protein and used an available construct for the PanD regulatory protein, PanZ (17). This protein was produced with a LPETGA motif inserted prior to a C- terminal His-tag. In the absence of DAP, using 2 equivalents of a GV-terminated co-substrate, the inventors initially observed ~50% turnover to product after overnight incubation at 25°C (Figure 3A grey). Addition of DAP increased product formation to 80% (Figure 3A black) with the remaining protein evenly distributed between the unlabelled substrate and a hydrolysis product. In contrast, when a peptide which is a substrate for DAP (e.g., GA-terminated) was used, the inventors only observed either unlabelled protein or the hydrolysis product formed from the thioester intermediate (Figure 6C). They therefore used labelling of PanZ to screen potential peptide substrates by incubating the peptides and PanZ with Srt7M in the presence and absence of DAP (Figure 3B and 7). When the labelling peptide was a substrate for DAP, the inventors only observed the hydrolysed protein with trace unlabelled substrate but in cases where the labelling peptide is not a substrate for DAP, the inventors observed a shift in the product distribution from the unlabelled substrate towards the labelled product with variable proportions of the hydrolysis product. From this screen, the inventors selected the GV42 terminated peptide together with the LPETGA motif as their preferred substrates for future experimentation.

[0124] Optimization of C-terminal labelling

[0125] The next aim was to optimize the conditions of the labelling reaction. Increasing the pH to above pH 8 (Figure 8) led to >95% conversion to the product, presumably resulting from favouring aminolysis over the hydrolysis reaction at pH above the pKa of the N-terminal amine, consistent with previous results on the pH optimum of sortase labelling (18). The proportion of the desired product also increased with the number of equivalents of peptide used for labelling, but only marginal enhancement was observed above two equivalents of the labelling peptide with the PanZ substrate (Figure 3D and 9). Variation in the amount of DAP added to the reaction did not affect overall reaction turnover (Figure 10) and while an increase in temperature from 25°C to 37°C allowed a reduction in the amount of catalyst, it did not significantly change conversion levels relative to hydrolysis (Figure 11). Finally, the inventors investigated the effect of common co-solvents on the reaction (Figure 12). Addition of DMSO, glycerol and propylene glycol had no effect on the reaction whereas addition of DMA and DMF inhibited, but did not stop, the reaction.

[0126] The inventors next investigated modification of the C-terminus of the pentameric cholera toxin B-4 subunit (CTB) which is widely used as a tool in cell biology and neuroscience (19). The inventors have previously reported a CTB construct with a C-terminal sortase recognition site, however periplasmic expression yielded heterogenous substrates in which only 80-90% of the protomer chains included the desired extension (20). The inventors therefore generated a new C-terminally His-tagged CTB construct incorporating a LPETGA labelling motif. Cytosolic expression of the unfolded protein into inclusion bodies, followed by dissolution in 8M urea, affinity purification and subsequent refolding provided homogeneous material for the labelling experiments. In this case, the inventors used automated sampling to monitor the labelling reaction in real time using 2 equivalents of a labelling peptide and observed 12 maximal (>95%) labelling after approximately 2 hours (Figure 3D and 13). Reduction in the number of equivalents of the labelling peptide to only 1.5 equivalents yielded essentially identical results (Figure 3D, 3E). In this case the overall level of labelling was marginally higher than that observed for PanZ, which the inventors attribute to the longer linker between the globular protein domain and the LPETG recognition motif for the CTB protein.

[0127] Optimization of N-terminal labelling

[0128] Having demonstrated effective C-terminal labelling using this coupling enzyme approach, the inventors investigated its application to N-terminal labelling and compared it to the use of depsipeptide substrates. They tested the enzyme combination using protein constructs in which a GlyVal-labelling motif was added to the N-termini of the expressed proteins. The inventors initially investigated labelling of a pentameric GVG-CTB using mass spectrometry to follow the reaction using a single equivalent of the labelling reagent (Figure 4A and 4B) and 2 mol% Srt7M in the presence (black) and absence (grey) of 10 mol% DAP. Addition of DAP to the reaction mixture shifted the reaction equilibrium from a statistical 1: 1 distribution between labelled and unlabelled products towards only the labelled product. The inventors next compared labelling using depsipeptides to labelling using the combination of peptides and DAP (Figure 14A) and saw essentially identical behavior using the two approaches. Variation in the concentration of the labelling reagent (Figures 14B and 15) showed that as little as 1.2 equivalents of the peptide label were sufficient to enable ~95% labelling of this protein at 25°C. Both the depsipeptide and new coupled enzyme approach led to near-quantitative labelling within 2 hours. However, the new approach is more robust due to the stability of the peptide substrates in solution in comparison to depsipeptides which are subject to non-enzymatic hydrolysis over time. Labelling of the model maltose-binding protein construct GVG-MBP (3) revealed a similar pattern (Figure 40 and 4D), with the addition of DAP again enabling near quantitative N-terminal labelling using an unactivated peptide, equivalent to that observed using depsipeptide substrates.

[0129] Creating a chimeric bifunctional catalyst

[0130] Given that addition and removal of two enzymes increases the chance for product contamination and that a single reagent is simpler to handle and monitor, the inventors investigated the combination of both enzymatic activities into a single chimeric polypeptide. Their approaches to N- and C-terminal labelling used variable ratios of the two enzymes but DAP activity never appeared to be a limiting factor - they therefore anticipated that fusing together equimolar quantities of the two enzymes would not significantly affect either class of reaction. The inventors generated fusion proteins consisting of both the sortase and DAP domains in tandem with his-tag (SrtH7D) or both his-tag and a chitinbinding domain (SrtCH7D) affinity purification tags. SrtH7D and SrtCH7D both expressed at high levels and after purification the chimeric proteins proved stable to short-term storage at 4 °C as well as storage at -20 and -80 °C.

[0131] The inventors initially compared the activity of SrtH7D and SrtCH7D to the separate component proteins for C-terminal labelling of PanZ and observed successful labelling with a slight increase in the level of hydrolysis (Figure 16). The inventors therefore tested C-terminal labelling of CTB using SrtH7D, and in this case observed essentially identical kinetic behaviour to the use of the separate components (Figure 5A). SrtCH7D was also highly effective, and they were able to isolate the modified CTB with high levels of labelling using just 2 equivalents of the peptide without further optimisation (Figure 5B). The inventors next used this approach to C-terminally modify MBP, creating a new C-terminal modified MBP-LPETGAHe construct for that purpose. Once again, they saw essentially quantitative labelling using a small excess of the labelling peptides (Fig 5C and 17) with only a small difference in labelling efficiency between 1.5 and 3 equivalents of the labelling peptide.

[0132] For N-terminal labelling, the inventors initially compared the activity of the SrtH7D / peptide combination with the Srt7M / depsipeptide strategy to label MBP - and saw comparable results using the two approaches and 3 equivalents of a labelling reagent (Figures 5D and 18). They next used GVG-CTB to investigate the concentration dependence of the labelling reaction, comparing the labelling reaction between the combined SrtH7D construct and the separate components at equimolar concentrations (Figure 19). In this case, the inventors observed near quantitative labelling at 25°C using 1.5-2 equivalents of the labelling peptide, which is consistent with the results for the independent components.

[0133] Finally, the inventors generated a Titin 127 domain (a model protein commonly used in multidomain constructs for mechanical unfolding of proteins) suitable for dual N- and C-terminal labelling. This incorporated a C-terminal sortase motif and a latent N-terminal GVG sequence that could be revealed by action of TEV protease. It was possible to label this protein quantitatively on its C- terminus using 2-3 equivalents of the labelling peptide and as little as 10 mol% SrtH7D over a 2-hour time-period (Figure 5E and 20). Then, following TEV 1 cleavage to create a new GVG N-terminus, the protein was labelled using an orthogonal LPESoG depsipeptide substrate (3) and Srt4S-9 (SrtLPXSG) to yield a double labelled protein (Figure 5E).

[0134] Further investigation of the scope of the coupled enzyme labelling system

[0135] Having demonstrated that they can achieve high level N- and C-terminal labelling, the inventors wanted to address three further questions. Can they further enhance the efficiency of labelling, can they use a different sortase in combination with D-aminopeptidase and can they use this method to achieve protein fusion? Protein labelling using the combination of Srt7M and D- aminopeptidase at 25°C is already rapid and nearly equimolar in terms of protein and peptide substrate. For N-terminal labelling, a small excess of peptide is sufficient to enable quantitative labelling however for C-terminal labelling, peptide hydrolysis remains a challenge even at relatively high ratios of label to protein. In practice, even though the sortase selectively catalyzes aminolysis over hydrolysis, since the latter reaction is irreversible, a small amount of the hydrolysis product is still seen. This problem can be counteracted by carrying out reactions at sufficiently high substrate concentration, however this is not practical for the majority of proteins. The inventors therefore sought to determine whether we could use temperature to control this side reaction.

[0136] As they had observed more hydrolysis at 37°C than at 25°C, the inventors decided to investigate the reaction at 4°C, with the hypothesis that the selectivity for aminolysis over hydrolysis would be higher as the temperature decreased. They tested N-terminal labelling of the GVG-CTB construct using 0.02 eq SrtH7D and 1.1 eq peptide with overnight incubation at 4, 16 and 25°C (Figure 21). Although slower, reaction at 4°C eventually led to the same level of labelling as observed for higher temperatures. Following overnight incubation, the inventors observed >99% protein labelling at 4 °C, whereas at higher temperatures hydrolysis-mediated unlabelling had occurred. Varying the concentration of peptide revealed that only 1.05 equivalents were required to yield essentially quantitative labelling, and a time course experiment revealed that the level of labelling remained stable under these conditions between 10 and 24 hours (Figure 5F) before the peptide label started to be removed by hydrolysis. In practice, these concentrations are within the error of the method used to determine the concentration of reagents suggesting that equimolar N-terminal labelling with sortase has been achieved for the first time. They similarly reinvestigated the C- terminal labelling of both MBP-LPETGA under the same conditions but this time using 0.25 eq of the SrtH7D catalyst. In this case, the inventors saw quantitative labelling of protein with no detectable hydrolysis using 2 equivalents of the labelling peptide (Figure 22) but were not able to reproducibly produce the same success with near 1 equimolar peptide substrates as for the N-terminus.

[0137] The inventors have previously demonstrated the orthogonality of the Srt4S-9 (SrtLPXSG) variant to WTSrtA for depsipeptide-mediated labelling (3) and used those reagents in the dual labelling approach discussed above. They therefore investigated whether this Ca-dependent variant sortase could be combined with DAP to enable quantitative labelling with peptides. The inventors screened labelling of GVG-CTB at 4°C in the presence and absence of DAP using a variety of peptide label ratios (Figures 5G and 23). They observed enhanced labelling relative to SrtLPXSG alone in all cases and observed >95% labelling for both 2 and 3 equivalents of peptide label after 24 h incubation. This suggests that the method should be readily portable to other Ca-dependent and independent variants of sortase with different substrate specificities or to other transpeptidases.

[0138] Finally, the inventors considered if SrtH7D could be applied to fusion of two globular proteins. We have previously attempted to use sortase variants to enable protein fusion, however at 25°C they often observed hydrolysis of the N- terminal protein fragment containing the LPETG motif. The inventors therefore investigated fusion at 4°C using a small excess of the C-terminal fragment. Using these conditions, they observed quantitative fusion of an N-terminal fragment (in this case an Affimer protein, Aff-LPETGA (21) with a model C-terminal fragment (GVG-MBP)) over a 7-day incubation period at 4°C (Figure 5H) using 10 mol% (5 pM) SrtH7D. Slower ligation was observed at 1 and 2.5 pM SrtH7D (Figure 24). Similar trends in successful ligation were observed with dimerization of MBP (Figure 25) and formation of other fusions from their constituent parts.

[0139] Discussion

[0140] The ideal transpeptidation reaction would be one in which two substrates, either peptides or proteins, were mixed in equimolar concentrations in the presence of a catalyst and a quantitative yield of a product is produced. In this work the inventors have demonstrated that combination of a sequence-specific aminopeptidase with a sequence-specific transpeptidase is an effective strategy towards such a reaction. As shown above, the combined broad substrate tolerance of the sortase in terms of the second residue in the nucleophilic substrate and the sequence specificity of the D-aminopeptidase used, enables selective transpeptidation. Given that the two peptidases can selectively form one peptide product preferentially, the key parameter that remains to be optimised is the chemoselectivity of the transpeptidase for aminolysis over hydrolysis.

[0141] The established approaches to transpeptidation usually employ a large excess of label to ensure transformation which makes the hydrolysis side-reaction less important since hydrolysis can be outcompeted by both the substrate and byproduct peptides. The critical benefit of the coupled enzyme approach is that it allows this excess to be minimized, however avoiding hydrolysis is critical to its success. For N-terminal labelling, irreversible hydrolysis of the peptide substrate is of less concern than the hydrolysis of protein is for C-terminal labelling since, loss of peptide label to hydrolysis can be readily counteracted using a small excess of the peptide; however, for C-terminal labelling any hydrolysis leads to an irreversible reduction in the overall conversion.

[0142] At the concentrations used for protein labelling, the inventors observe approximately 100-fold selectivity for aminolysis over hydrolysis at 25°C. Given the relative concentrations of labelling peptide (200 pM) and water (55 M) this corresponds approximately to a 2.5 x 106-fold selectivity for aminolysis over hydrolysis at this temperature, which is approximately 40 kJ mol'1difference in the energy barrier between the two catalyzed pathways. Given this difference, dropping the temperature to 4°C would be expected to improve the selectivity of reaction 3-4-fold. This estimate is largely consistent with the small enhancement in selectivity observed. Further improvement in this ratio will require use of a different catalyst. The inventors anticipate that other natural or evolved enzymes could exhibit such selectivity at higher temperatures and that reducing Kmfor the aminolytic substrate will be a key parameter to be optimized to maximize reaction turnover.

[0143] As stated above, for N-terminal labelling, a moderate (1.5-2-fold) excess of peptide is sufficient to enable near quantitative labelling using a small (2 mol%) amount of catalyst at 25°C. Overall the approach is broadly equivalent to the inventors previous depsipeptide approach but with the advantage of using standard peptides which are stable to storage in buffer. At 4°C, although the reaction is slower, the relative stability of the product to hydrolysis means that the reaction is more efficient and requires a smaller excess of peptide. Although longer incubation times for C-terminal labelling lead to irreversible hydrolysis, in the presence of a small excess of peptide, reaction mixtures show stable product distributions for 10-20 hours enabling flexibility in protein labelling workflows. In this regard, the success of protein fusion experiments is particularly notable, although in these cases the inventors suspect that steric hindrance may disfavour rebinding of the catalyst to the product, thereby disfavouring the hydrolysis reaction and allowing a lower concentration of catalyst to be used.

[0144] In this study, the inventors have demonstrated combined application of D- aminopeptidase with two different sortases, however they anticipate that the same strategy could be readily applied to other variant sortases or transpeptidases. For example, the inventors would predict that combination of D- aminopeptidase with OaAEPl and -NGM, and GL- or GV-terminated substrates would work effectively.

[0145] Examole 2

[0146] The inventors further considered using this method at lower protein concentrations. Using 10 pM protein, it was possible to achieve quantitative N- terminal labelling using only 1.1 or 1.5 equivalent of peptide (Figure 39). However, for C-terminal labelling any hydrolysis leads to an irreversible reduction in overall conversion and an excess of peptide is required to ensure that the peptide / water concentration ratio is maintained. For example, using 10 pM protein, while complete consumption of unlabelled protein is observed at all peptide concentrations, 400 pM peptide is required to fully suppress hydrolysis— this is the concentration of peptide used in labelling higher concentrations of protein (Figures 3E and 40).

[0147] The inventors considered that their earlier optimisation of N-terminal protein labelling (4) was dependent upon labelling of proteins with a single N- terminal glycine, and had for example initially generated constructs for labelling the cholera toxin B subunit with N-terminal triglycine motifs (27). Given the observed sequence specificity for degradation of hexapeptides by D- aminopeptidase and the sequence requirements for C-terminal labelling peptides they initially anticipated that these proteins would not be substrates for N- terminal labelling using the combination of sortases and a D-aminopeptidase. To test this hypothesis, the inventors investigated labelling of the GGG-CTB construct using the SrtH7D combined construct. They used the optimised conditions for labelling of GVG-CTB and incubated 100 pM of either GGG-CTB or GVG-CTB at 4 °C with 2 pM SrtH7D and 150 pM of the AYLPETGG labelling peptide. In both cases, they saw essentially quantitative labelling, with no detectable unlabelled protein in the case of GGG-CTB (figure 41). They hypothesized that this successful labelling was due to steric hindrance of binding of D-aminopeptidase as a result of the short linker region and therefore investigated labelling of a much longer protein - the N-terminal glycine revealed on the N-terminal his-tag in the pET28a-encoded vectors as the result of action of methionine aminopeptidase, with the N-terminal sequence GSSH... They once again incubated 100 pM of a model protein (in this case human ATIC) in the same conditions for 24 hours. They once again observed complete labelling of the protein in these conditions and therefore repeated the labelling reaction with the individual labelling proteins (Srt7M and DAP) in combination and individually. They observed complete labelling in the presence of both enzymes, ~66% labelling in the presence of Srt7M alone (c.f. 60% expected) and no evidence of protein N-terminal degradation in the presence of D-aminopeptidase. To confirm the generality of this labelling, they labelled two additional available proteins expressed in the same backbone (LgtC and DsbA) observing 99 and 97% labelling respectively under the same conditions. This suggests that this approach should be effect for labelling a wider range of proteins on the N-terminus.

[0148] The inventors further characterised the behaviour of the SrtH7D fusion using a fluorescence unquenching-based peptide assay for transpeptidase activity (figure 42). They used this approach to further compare the behaviour of the separate components of the chimera compared to the combined chimera and saw a small decrease in maximal activity compared to the separate components (figure 43). Systematic investigation of storage at 4, -20 and -80 °C showed little difference in residual activity for storage at all temperatures over a 5 week period (figure 44) The same assay was used to compare samples of SrtH7D prepared at different times and stored at -80 °C for between 1 and 3 years showing I reduction in activity over this prolonged storage time, (figure 45). Finally, the specificity of the system for the N-terminus of model peptides was re-evaluated using the fluorescence dequenching assay (Table 1). While some variation in activity was observed between substrates these spanned only a approximately 2- fold range in specific activity.

[0149] Table 1. Kinetic parameters for SrtH7D-mediated ligation of peptides determined by a peptide based fluorescence-unquenching assay. A similar value of Vmax / Km is observed for all substrates tested.

[0150] The inventors investigated the combined application of D-aminopeptidase with the Ca-dependent pentamutant Srt5M (28). Combined application of the two enzymes to C-terminal labelling of CTB-H6 was investigated by systematic variation of multiple variables (figure 46). Reaction samples were quenched with EGTA prior to analysis using electrospray mass spectrometry. Optimal labelling was observed at 4 °C using 1% Srt5M with little variation in the effect of increasing the concentration of labelling peptide from 2 to 3 equivalents or in the amount of D-aminopeptidase used. A chimeric protein analogous to SrtH7D consisting of Srt5M linked to DAP (termed SrtH5D or H5D) was constructed in an analogous manner. Assays of this protein indicated that is behaved similarly to SrtH7D with good labelling of MBP-LPETGAH6 observed relative to Srt5M alone (Figure 47). In view of the many possible embodiments to which the principles of the disclosed invention may be applied, it should be recognised that the illustrated embodiments are only preferred examples of the invention and should not be taken as limiting the scope of the invention. Rather, the scope of the invention is defined by the following claims.

[0151] References

[0152] 1. Q. Wu, H. L. Ploegh, M. C. Truttmann, Hepta-Mutant Staphylococcus aureus Sortase A (SrtA(7m)) as a Tool for in Vivo Protein Labeling in Caenorhabditis elegans. ACS Chemical Biology 12, 664-673 (2017).

[0153] 2. J. M. Antos et al., Site-specific N- and C-terminal labeling of a single polypeptide using sortases of different specificity. Journal of the American Chemical Society 131, 10800-10801 (2009).

[0154] 3. H. E. Morgan, Z. L. P. Arnott, T. P. Kaminski, W. B. Turnbull, M. E. Webb, Combined Application of Orthogonal Sortases and Depsipeptide Substrates for Dual Protein Labeling. Bioconjug Chem 33, 2341-2347 (2022).

[0155] 4. D. J. Williamson, M. A. Fascione, M. E. Webb, W. B. Turnbull, Efficient N- terminal labeling of proteins by use of sortase. Angew. Chem. Int. Ed. 51, 9377- 9380 (2012). 5. L. Freiburger et al., Efficient segmental isotope labeling of multi-domain proteins using Sortase A. J. Biomol. NMR 63, 1-8 (2015).

[0156] 6. T. M. S. Tang et al., Use of an asparaginyl endopeptidase for chemo-enzymatic peptide and protein labeling. Chemical Science 11, 5881-5888 (2020).

[0157] 7. S. A. Reed, D. A. Brzovic, S. S. Takasaki, K. V. Boyko, J. M. Antos, Efficient Sortase-Mediated Ligation Using a Common C-Terminal Fusion Tag. Bioconjugate Chemistry 31, 1463-1473 (2020).

[0158] 8. D. Row, T. J. Roark, M. C. Philip, L. L. Perkins, J. M. Antos, Enhancing the efficiency of sortase-mediated ligations through nickel-peptide complex formation. Chem Commun (Camb) 51, 12548-12551 (2015).

[0159] 9. F. B. H. Rehm et al., Site-Specific Sequential Protein Labeling Catalyzed by a Single 1 Recombinant Ligase. J Am Chem Soc 141, 17388-17393 (2019).

[0160] 10. Y. Asano, A. Nakazawa, Y. Kato, K. Kondo, Properties of a novel D- stereospecific aminopeptidase from Ochrobactrum anthropi. The Journal of biological chemistry 264, 14233-14239 (1989).

[0161] 11. L. Fanuel et al., The DmpA aminopeptidase from Ochrobactrum anthropi LMG7991 is the prototype of a new terminal nucleophile hydrolase family. The Biochemical journal 341 ( Pt 1), 147-155 (1999).

[0162] 12. C. Bompard-Gilles et al., Crystal structure of a D-aminopeptidase from Ochrobactrum anthropi, a new member of the 'penicillin-recognizing enzyme' family. Structure (London, England : 1993) 8, 971-980 (2000).

[0163] 13. Z. Zou, M. Noth, F. Jakob, U. Schwaneberg, Designed Streptococcus pyogenes Sortase A Accepts Branched Amines as Nucleophiles in Sortagging. Bioconjugate Chemistry 31, 2476-2481 (2020).

[0164] 14. M. Wojcik, K. Szala, R. van Merkerk, W. J. Quax, Y. L. Boersma, Engineering the specificity of Streptococcus pyogenes sortase A by loop grafting. Proteins 88, 1394-1400 (2020).

[0165] 15. K. Strijbis, E. Spooner, H. L. Ploegh, Protein Ligation in Living Cells Using Sortase. Traffic 13, 780-789 (2012).

[0166] 16. I. Wuethrich et al., Site-specific chemoenzymatic labeling of aerolysin enables the identification of new aerolysin receptors. PLoS ONE 9 (2014).

[0167] 17. D. C. F. Monteiro et al., The structure of the PanD / PanZ protein complex reveals negative feedback regulation of pantothenate biosynthesis by coenzyme A. Chemistry & biology 22, 492-503 (2015).

[0168] 18. Z. Wu, H. Hong, X. Zhao, X. Wang, Efficient expression of sortase A from Staphylococcus aureus in Escherichia coli and its enzymatic characterizations. Bioresour Bioprocess 4, 13 (2017).

[0169] 19. A. K. Kenworthy et al., Cholera Toxin as a Probe for Membrane Biology. Toxins 13, 543 (2021). 20. J. P. Dolan et al., Synthesis of cholera toxin B subunit glycoconjugates using site-specific orthogonal oxime and sortase ligation reactions. Frontiers in Chemistry 10, 958272 (2022).

[0170] 21. M. R. Balmforth et al., Piggybacking on the Cholera Toxin: Identification of a CTB-Binding Protein as an Approach for Targeted Delivery of Proteins to Motor Neurons. Bioconjug Chem 32, 2205-2212 (2021).

[0171] 22. H. E. Morgan, Z. L. P. Arnott, T. P. Kaminski, W. B. Turnbull, M. E. Webb, Combined Application of Orthogonal Sortases and Depsipeptide Substrates for Dual Protein Labeling. Bioconjug Chem 33, 2341-2347 (2022).

[0172] 23. C. P. Guimaraes et al., Site-specific C-terminal and internal loop labeling of proteins using sortase-mediated reactions. Nat Protoc 8, 1787-1799 (2013).

[0173] 24. M. D. Witte et al., Site-specific protein modification using immobilized sortase in batch and continuous-flow systems. Nat Protoc 10, 508-516 (2015).

[0174] 25. J. P. Dolan et al., Synthesis of cholera toxin B subunit glycoconjugates using site-specific orthogonal oxime and sortase ligation reactions. Frontiers in Chemistry 10, 958272 (2022).

[0175] 26. S. Nozaki, M. E. Webb, H. Niki, An activator for pyruvoyl-dependent I- aspartate o-decarboxylase is conserved in a small group of the y-proteobacteria including Escherichia coli. MicrobiologyOpen 1, 298-310 (2012).

[0176] 27. D.C. Machin et al., Sortase-Modified Cholera Toxoids Show Specific Golgi Localization. Toxins 16, 194. (2024).

[0177] 28. L. Chen et al., Improved variants of SrtA for site-specific conjugation on antibodies and proteins with high efficiency. Sci. Rep. 6, 31899 (2016)

Claims

Claims1. A method of transpeptidation comprising: ligating a donor substrate to an acceptor protein or peptide using a transpeptidase, wherein the donor substrate is a protein, peptide, amino acid amide or amino acid ester, and wherein the ligation reaction generates a ligated product and a by-product that is a substrate for the transpeptidase, cleaving one or more amino acids from a terminus of the by-product using an exopeptidase, such that the truncated by-product is not a substrate for the transpeptidase, thereby shifting the reaction equilibrium of the transpeptidation reaction towards the ligated product.

2. A method according to claim 1, wherein the exopeptidase is an aminopeptidase, and wherein the one or more amino acids are cleaved from the N-terminus of the by-product.

3. A method according to claim 2, wherein the aminopeptidase is a D- aminopeptidase.

4. A method according to claim 2 or claim 3, wherein the aminopeptidase is a glycyl aminopeptidase.

5. A method according to claim 3, wherein the D-aminopeptidase is Brucella anthropi D-aminopeptidase.

6. A method according to any one of claims 2-5, wherein the aminopeptidase is sequence-specific, optionally wherein the aminopeptidase is specific for GX2- terminated peptides, wherein X2 is a small, polar or negatively charged amino acid, or other glycineamide derivative.

7. A method according to any preceding claim, wherein the N-terminus of the donor substrate comprises a motif that is not a substrate for the exopeptidase.

8. A method according to claim 7, wherein the N-terminus of the donor substrate comprises the motif XXi, wherein Xi is a hydrophobic or positively charged amino acid.

9. A method according to claim 8, wherein the N-terminus of the donor substrate comprises the motif GXi, wherein Xi is a hydrophobic or positively charged amino acid, optionally wherein the N-terminus of the donor substrate comprises the motif GV.

10. A method according to any preceding claim, wherein the acceptor protein or peptide comprises a recognition motif that is cleaved by the transpeptidase to generate the by-product.

11. A method according to claim 10, wherein the recognition motif comprises LPXT / G or LPXT / GX2, such that the by-product comprises glycinamide or GX2, optionally wherein the recognition motif comprises LPET / G or LPET / GX2.

12. A method according to 11, wherein X2 is a small, polar or negatively charged amino acid, optionally wherein X2 is A.

13. A method according to any preceding claim, wherein the N-terminus of the donor substrate comprises a motif which is not a substrate for the exopeptidase, and wherein the acceptor protein or peptide comprises a recognition motif that is cleaved by the transpeptidase to generate the by-product.

14. A method according to claim 13, wherein the N-terminus of the donor substrate comprises the motif GXi, wherein Xi is a hydrophobic or positively charged amino acid, and wherein:(i) the acceptor protein or peptide comprises the motif X / GX2, wherein I is the cleavage site of the transpeptidase and X2 is a small, polar or negatively charged amino acid; or(ii) the acceptor protein or peptide comprises the motif X / G-amide, wherein the amide group is -CONH2, -CONR2 or -CONHR.

15. A method according to claim 14, wherein the N-terminus of the donor substrate comprises the motif GXi, wherein Xi is a hydrophobic or positively charged amino acid, and wherein the acceptor protein or peptide comprises the motif LPXT / GX2, wherein I is the cleavage site of the transpeptidase and X2 is a small, polar or negatively charged amino acid or an amide group.

16. A method according to any preceding claim, wherein the transpeptidase is a cysteine transpeptidase.

17. A method according to any preceding claim, wherein the transpeptidase is a sortase, a butelase or an asparaginyl endopeptidase.

18. A method according to claim 17, wherein the transpeptidase is a sortase.

19. A method according to claim 18, wherein the transpeptidase is a sortase A, optionally wherein the sortase A is Streptococcus pyogenes sortase A or an evolved form thereof.

20. A method according to any preceding claim, wherein the transpeptidase and the exopeptidase are provided by a chimeric fusion polypeptide.

21. A method according to any preceding claim, wherein the N-terminal and / or C-terminal ends of the acceptor protein or peptide are ligated to donor substrates.

22. A method according to any preceding claim, wherein the by-product and the donor substrate have identical N-terminal residues, optionally wherein the byproduct and the donor substrate have glycine N-terminal residues.

23. A transpeptidation product obtainable from the method according to any one of claims 1 to 22.

24. A chimeric fusion polypeptide comprising a transpeptidase domain and an exopeptidase domain.

25. A chimeric fusion polypeptide according to claim 24, wherein the transpeptidase domain comprises a sortase A domain, and wherein the exopeptidase domain comprises a D-aminopeptidase domain.

Images