This application claims the benefit of priority of Singapore Patent Application No. 10201904085W filed May 7, 2019, and Singapore Patent Application No. 10201910861 S filed November 19, 2019, the contents of which being hereby incorporated by reference in its entirety for all purposes.

Field of the Invention

The present invention lies in the technical field of enzyme technology and specifically relates to enzymes having Asx-specific ligase and cyclase activity and to nucleic acids encoding those as well as methods of the manufacture of said enzymes. Further encompassed are methods and uses of these enzymes.

Background of the Invention

Head-to-tail macrocyclization of peptides and proteins has been used as a strategy to constrain structures and enhance metabolic stability against proteolytic degradation. In addition, a constrained macrocyclic conformation may also improve pharmacological activity and oral bioavailability. Although most peptides and proteins are produced as linear chains, circular peptides ranging from 6 to 78 residues occur naturally in diverse organisms. These cyclic peptides usually display high resistance to heat denaturation and proteolysis and have inspired a new trend in protein engineering, as demonstrated by recent successes in the cyclization of cytokines, histatin, ubiquitin C-terminal hydrolase, conotoxin and bradykin in-grafted cyclotides. Furthermore, cyclic peptides have been used as therapeutics, including valinomycin, gramicidin S and cyclosporine.

To date chemical methods are typically used for the cyclization of peptides. One possible strategy is native chemical ligation. This method requires an N-terminal cysteine and a C-terminal thioester, requirements that limit its application for non-cysteine-containing peptides. Furthermore, chemical methods are not always feasible, especially for large peptides and proteins.

Although enzymatic methods employing a naturally-occurring cyclase would be ideal, currently only very few peptide cyclases are known and they are for various reasons not fully exploited.

Recent discovery of a novel cyclase, butelase-1 , from cyclotide-producing plant Clitoria ternatea proved that a unique type of asparaginyl endopeptidase (AEP) is the processing enzyme to cyclize the linear precursors of cyclotides (Nguyen GKT, et al. (2014) Nat Chem Biol 10(9):732- 738). AEPs (or legumains) are cysteine proteases belonging the subfamily C13 (EC 3.4.22.34) under clan CD. Compared to AEPs which are hydrolases, butelase-1 reverses the enzymatic direction of AEPs and strongly promotes aminolysis to catalyze peptide-bond formation. Bioassays show that butelase-1 is not only a cyclase but also an efficient peptide ligase that can ligate biomolecules by forming peptide bond between Asn/Asp and any amino acids except Pro with the highest reported catalytic efficiency to date of 1 ,340,000 IVHs ^-1 . Butelase-1 is a versatile protein engineering tool for protein and peptide ligation, modification, cyclization, tagging, and live cell labeling and has been described in detail, including its uses, in international patent application WO 2015/163818 A1 . Such butelase-1 like peptide ligases, designated as peptide asparaginyl ligases (PALs), are useful biochemical and biotechnological tools for linkage-specific and site- specific protein modifications and precision biomanufacturing of biotherapeutics such as antibody-drug conjugates.

AEPs and PALs are expressed as proenzymes, generally consisting a ~10-kDa pro-domain, an active ~32-kDa core domain formed by six b-strands surrounded by five a-helices, and a 15-kDa C-terminal cap domain formed by six tightly-bound helices. Both AEPs and PALs display intrinsic protease activity at acidic pH for autolytic maturation. Their biosynthetic processing is similar, involving an autolytic activation in the acidic subcellular compartments such as lysosomes and lytic vacuoles. In vitro, activations are usually performed at pH 4 to 5. A major structural change after the acidic activation is the cleavage and dissociation of the C-terminal cap domain, and which exposes the catalytic site in the core domain. Religation of cap and core domains is reported at near neutral pH when both domains remain intact and in close proximity after cleavage.

Plant AEPs play important roles in protein degradation, maturation, programmed cell death, and host defense via their proteolytic activity triggered in the acidic environment of vacuoles. AEPs, such as butelase 2, OaAEP2 and HaAEPI (sunflower Helianthus annuus) display predominantly protease activity even at neutral pH, with a very low level of ligase activity. Certain AEPs catalyze both ligation and hydrolysis products from peptide substrates carrying AEP-recognition signals at near neutral pH (6-7.5). Very rarely, AEPs mediate peptide splicing including both peptide bond breaking and formation, such as in the maturation of concanavalin A, by mediating circular permutation. In contrast to these“bi-functional” or“predominant” AEPs, PALs like butelase 1 and OaAEPI b catalyze the formation of ligation products essentially devoid of any hydrolytic product at near neutral pH, and their ligase activity is preponderant even under mild acidic conditions (pH <6).

Currently, only a handful of such PALs have been identified. They include the prototypic PAL butelase-1 , as well subsequent discoveries butelase-1 like enzymes, OaAEPI b (Harris KS, et al. (2015) Nat Commun 6(1 ):101 99) and HeAEP3 (Jackson MA, et al. (2018) Nat Commun 9(1 ) :241 123) identified from other cyclotide-producing plants Oldenlandia affinis and Hybanthus enneaspermus, respectively. To date, the molecular mechanisms differentiating AEPs and PALs are not known. Despite the publication of several plant AEP crystal structures, including both proenzymes and active form, the structural determinants that underpin their nature as protease or ligase are still unresolved. Enzymes from both extremes share the same structure with r.m.s.d < 1 A (e.g. OaAEPI b and HaAEPI ).

Since ligase/cyclase activity is highly desirable and there is need in the art for novel ligases/cyclases that can be reliably used as molecular tools for peptide ligation and cyclization, it would be helpful to identify the determinants that control enzyme directionality of PALs and AEPs, as this would provide opportunities for deliberately tailoring these enzymes for specific needs.

Summary of the invention

The inventors of the present invention found that the enzymatic activity of AEPs and PALs is controlled by subtle differences at key positions near the catalytic center. These local alterations control the access to the S-acyl enzyme intermediate of water molecules (leading to hydrolysis) or of incoming nucleophiles (leading to ligation). By studying a series of putative AEPs and PALs from two cyclotide-producing plants Viola yedoensis (var. phillipica) and Viola canadensis and using the recombinant enzymes to investigate the molecular mechanisms responsible for ligase catalytic activity, two putative Ligase Activity Determinants (LADs) could be identified and validated by structural comparison, MD simulation and site-directed mutagenesis. These results explain the molecular mechanism allowing the conversion of AEPs into PALs and provide a useful tool for the discovery and engineering of new ligases. In the course of these studies, further useful PALs were identified that allow efficient recombinant expression and show high cyclization activities.

In a first aspect, the present invention thus relates to an isolated polypeptide having protein ligase, preferably cyclase, activity comprising or consisting of

(i) the amino acid sequence as set forth in SEQ ID NO:1 ;

(ii) an amino acid sequence that shares at least 60, preferably at least 70, more preferably at least 80, most preferably at least 90 % sequence identity with the amino acid sequence set forth in SEQ ID NO:1 over its entire length;

(iii) an amino acid sequence that shares at least 80, preferably at least 90, more preferably at least 95 % sequence homology with the amino acid sequence set forth in SEQ ID NO:1 over its entire length; or

(iv) a fragment of any one of (i)-(iii).

The polypeptide consisting of SEQ ID NO:1 is also referred to herein as“VyPAL2” or“VyPAL2 active form/domain”. In another aspect, the present invention also relates to nucleic acid molecules encoding the polypeptides described herein, as well as a vector containing such a nucleic acid, in particular a copying vector or an expression vector.

In a further aspect, the invention is also directed to a host cell, preferably a non-human host cell, containing a nucleic acid as contemplated herein or a vector as contemplated herein. The host cell may be an insect cell, such as a Sf9 (Spodoptera frugiperda) cell.

A still further aspect of the invention is a method for manufacturing a polypeptide as described herein, comprising culturing a host cell contemplated herein ; and isolating the polypeptide from the culture medium or from the host cell.

In a still further aspect, the present invention relates to the use of polypeptides described herein for protein ligation, in particular for cyclizing one or more peptide(s).

In still another aspect, the invention relates to a method for cyclizing a peptide, the method comprising incubating said peptide with the polypeptides described above in connection with the inventive uses under conditions that allow cyclization of said peptide.

In a still further aspect, the invention relates to a method for ligating at least two peptides, the method comprising incubating said peptides with the polypeptides described above in connection with the inventive uses under conditions that allow ligation of said peptides.

In another aspect, the invention relates to a solid support material onto which the isolated polypeptides of the invention are immobilized as well as the use thereof and methods that use such substrates.

In another aspect, the invention also encompasses a transgenic organism, such as a plant, comprising a nucleic acid molecule encoding a polypeptide having protein ligase and/or cyclase activity as described herein. The polypeptide is preferably not naturally present in said organism. Accordingly, the present invention also features transgenic organisms, preferably plants, that express a heterologous polypeptide according to the invention.

In still another aspect, the invention also encompasses a method for increasing the protein ligase activity of a polypeptide having asparaginyl endopeptidase (AEP) activity, the method comprising the steps of substituting the amino acid residue at the position corresponding to position 126 of SEQ ID NO:1 with either an A or a G residue. In these embodiments, the amino acid residue at the position corresponding to position 127 of SEQ ID NO:1 may be selected such that the sequence at positions 126/127 is either GA or AP or AA, preferably GA or AP. If the amino acid at the position corresponding to position 126 of SEQ ID NO:1 is G, it is preferred that the amino acid at the position corresponding to position 127 of SEQ ID NO:1 is not P.

In a still further aspect, the invention is also directed to a method for producing a polypeptide having protein ligase activity, the method comprising:

(i) providing a polypeptide having asparaginyl endopeptidase (AEP) activity; and

(ii) introducing one or more amino acid substitutions into the polypeptide having

asparaginyl endopeptidase (AEP) activity, wherein said substitutions comprise substituting the amino acid residue at the position corresponding to position 126 of SEQ ID NO:1 with an A or a G residue and optionally substituting the amino acid residue at the position corresponding to position 127 of SEQ ID NO:1 with either a P or an A residue, such that the amino acid sequence in the positions corresponding to positions 126/127 in SEQ ID NO:1 is either GA, AA or AP, preferably GA or AP.

Brief description of the drawings

Figure 1 shows the enzymatic activity of recombinant \/yPAL1 -3 and \/yAEP1 . (A) Reaction scheme of ligase-mediated cyclization of GN14-SL (SEQ ID NO:59). (B) Analytical HPLC and MALDI-TOF mass spectrometry data of \/yPAL2-mediated cyclization under different reaction pH values. ^* : racemized synthetic GN14-SL. Note that MALDI-TOF MS was more sensitive against cyclic cGN14 than the linear species. (C) Quantitative summary of product ratio and reaction yield of each enzyme analyzed using RP-HPLC. For each reaction, a molar ratio of purified active enzyme:GN14-SL = 1 :500 was mixed and reacted at 37 °C for 10 min. Average yield and error bars were calculated from experiments performed in triplicate

Figure 2 shows the substrate specificity of \/yPAL2 against (A) substrates carrying degenerated native recognition motifs derived from Vy and Ct cyclotides as set forth in SEQ ID Nos. 48 and 59-77, (B) substrates with 20 different amino acids at PT position (X = 20 AA; SEQ ID NO:78), (C) substrates with 20 different amino acids at P2’ position (SEQ ID NO:79). All reactions were performed with a molar ratio of active \/yPAL2: substrate = 1 : 500 at pH 6.5, 37 °C for 10 min. Yields were quantitatively analyzed using RP-HPLC.

Figure 3 shows the enzyme kinetics of VyPkl2 and butelase 1 . HPLC-based kinetic study using the peptide substrate: GISTKSIPPISYRNSLAN (SEQ ID NO:60). A quantity of 50 nM purified active \/yPAL2 (or butelase 1 extracted from plant (15)) was used for each reaction. The amount of cyclization product at each time point was determined using analytical RP-HPLC. Average initial rate (Vo) of three repeat experiments were used for Michaelis-Menten curve plotting. Figure 4 depicts retro-engineering experiments in the S1’ pocket of \/yPAL3. All reactions were performed in the pH range 4.5 to 8.0. MS peaks of the hydrolysis product GN14 (SEQ ID NO:48) are marked with a dashed-line. (A) MS analysis of reactions catalyzed by \/yPAL3 wild-type. (B) MS spectra of reactions catalyzed by \/yPAL3-Y175G. (C) HPLC profiles of reactions catalyzed by \/yPAL3 and \/yPAL3-Y175G. (D) Quantitative summary of product ratio and reaction yield analyzed using RP-HPLC for \/yPAL3-Y175G.

Figure 5 shows the activity of VcAEP and the \/cAEP-Y168A mutant (LAD2). MS analysis and HPLC-based quantitative summary of the reaction catalyzed by (A) \/cAEP wild type and (B) \/cAEP-Y168A mutant targeting the LAD2 region. All reactions were performed at pH values ranging from 4.5 to 8.0. MS peaks of the hydrolysis product GN14, cyclization product cGN14, and sodium ion adduct of cGN14 are marked with dashed-lines. A dramatic improvement of ligase activity for the \/cAEP-Y168A mutant is clearly visible

Figure 6 shows ligase activity determinants (LADs) of PALs and the proposed catalytic mechanism. (A) Sequence alignment of PALs and AEPs studied in this work. Catalytic triad Asn- His-Cys is shaded in black. Residues belonging to the S1 pocket are shaded in blue. Proposed LAD residues are boxed in red. Residues of LAD1 and LAD2 are indicated. The conserved disulfide bond near LAD1 is highlighted in orange. The poly-Pro loop (PPL) is in a green box and the MLA loop in purple. The nomenclature of secondary structures was adapted from Trabi et al. (Trabi M, et al. (2004) J Nat Proc/67(5):806-810) with alterations according to the crystal structure of VyPkl2 (this work). Residues and motifs crucial for activity are labelled with the same color codes as used in the sequence alignment. Residues below the dotted line correspond to the oxyanion hole and those above the dotted line correspond to the proposed activity determinants. (B) Schemes proposed for ligation and hydrolysis by \/yPAL2 and the role of LAD1 and LAD2. The first step of the mechanism is identical for hydrolysis and ligation, leading to formation of the S-acyl enzyme intermediate and is the rate limiting step. Its main determinant is LAD1 . LAD2 controls the nature of activity to either favor the nucleophilic attack by a peptide (ligation) or a water molecule (hydrolysis). Full sequences of all aligned polypeptides are set forth in SEQ ID Nos. 5-14, 1 8 and 80-87 (VyPALI = SEQ ID NO:5; VyPAL2 = SEQ ID NO:6; VyPAL3 = SEQ ID NO:7; VyPAL4 = SEQ ID NO:8; VyPAL5 = SEQ ID NO:9; VyAEPI = SEQ ID NO: 10; VyAEP2 = SEQ ID NO:1 1 ; VyAEP3 = SEQ ID NO:12; VyAEP4 = SEQ ID NO:13; VcAEP = SEQ ID NO:14; Butelase-1 = SEQ ID NO:88; Butelase 2 = SEQ ID NO:80; OaAEPI b = SEQ ID NO:81 ; OaAEP2 = SEQ ID NO:82; HeAEP3 = SEQ ID NO:83; PxAEP3b = SEQ ID NO:84; CeAEP = SEQ ID NO:85; HaAEPI = SEQ ID NO:86; AtLEGy = SEQ ID NO:87).

Figure 7 shows the immobilization of PALs, butelase-1 and VyPAL2, by non-covalent affinity binding or covalent attachment. (A) Affinity binding of glycosylated PALs with concanavalin A (ConA) agarose beads to give ConA-PAL 1 and ConA-Vy2 2. (B) Affinity binding of biotinylated PALs with NeutrAvidin agarose beads to give NA-Bu1 (b) 3 and NA-Vy2(b) 4. Biotinylated PALs were prepared by coupling of succinimidyl-6-(biotinamido)hexanoate (NHS-LC-biotin) to the amino groups of PALs. (C) Covalent attachment of PALs with active NHS-ester on agarose beads to give agarose-Bu1 5 and agarose-Vy2 6. The distance between enzymes and the agarose beads are calculated by the sizes of the spacer moieties and the pre-coupled affinity binding ligands.

Figure 8 shows peptide macrocyclization by immobilized PAL beads. For each reaction, 1 mM of immobilized PALs calculated based on their protein loading was mixed with 0.2 mM KN14-GL (SEQ ID NO:51 ). The reactions were performed at pH 6.5 at room temperature with gentle rocking for 5 min. Products were eluted out from the spin column and analyzed with MALDI-TOF MS. KN14-GL 7 (calc mass 1659.4 Da, obs. Mass 1660.0 Da). cKN14 8 (calc mass 1471 .3 Da, obs. Mass 1471 .9 Da).

Figure 9 shows the determination of efficiency of immobilized butelase-1 and VyPAL2 by comparing with the standard activity curves of their soluble forms. Free enzyme concentration used was ranged from 1 to 8 nM. Reaction rate (V) was calculated by amount of product cKN14 (per second. Normal reaction buffer refers to the 20 mM sodium phosphate buffer (pH 6.5) containing 1 mM DTT, and 0.1 M NaCI. ConA reaction buffer refers to the reaction buffer with additional 5 mM CaCL and 5 mM MgCL.

Figure 10 shows operational stability of immobilized PALs. (A) RP-HPLC monitoring of cyclization of KN14-GL into cKN14 by five immobilized PALs 1 , 3-6 in 1 00 repeated reactions. For each experiment, 100 pL of reaction mixture (pH 6.5) containing 0.1 mM KN14-GL (SEQ ID NO:51 ) was given. Amount of beads used was adjusted according to the effective concentration of each type to give the molar ratio of effective enzyme:substrate = 1 :350 ~ 1 :600. Reaction was conducted at room temperature for 3-5 min. (B) Summary of operational stability of five immobilized PALs 1 , 3-6.

Figure 11 shows the stability of five immobilized PALs 1 , 3-6, butelase-1 and VyPAL2 after storing in 4 °C for 1 , 29 and 64 days.

Figure 12 shows the macrocyclization of peptides and proteins by NA-Bu1 (b) 3 at pH 6.5 at room temperature for 10 min with gentle shaking. (A) Cyclization of SFTI(D/N)-HV (SEQ ID NO:54) 12 (0.1 mM, calc mass 1767.9 Da, obs. Mass 1767.9 Da) into SFTI(D/N) 13 (calc mass 1513.8 Da, obs. Mass 1 513.3 Da) by NA-Bu1 (b) 3 to give 95% crude yield ( ^* ) of cyclic SFTI(D/N) 13 as determined by MALDI-TOF MS. (B) Cyclization of the folded linear bacteriocin precursor AS-48K 14 (50 mM, calc mass 7783.5 Da, obs. Mass 7779.1 Da) by NA-Bu1 (b) 3 to give cyclic AS-48 15 (calc mass 7145.1 Da, obs. Mass 7148.1 Da) with 83% yield as determined by UHPLC. Figure 13 shows the cyclooligomerization of peptide RV7 16 (SEQ ID NO:55; 0.2 mM, calc mass 956.5 Da, obs. Mass 957.7 Da) by NA-Bu1 (b) 3 to give 85% cyclodimer c17 (calc mass 1404.8 Da, obs. Mass 1405.9 Da) and 8% cyclotrimer c18 (calc mass 2107.2 Da, obs. Mass 2109.3 Da) in 40 min.

Figure 14 shows continuous-flow peptide and protein ligation by NA-Vy2(b) 4. (A) Ligation of Ac- RYANGI 19 (calc mass 735.4 Da, obs. Mass 734.3 Da; SEQ ID NO:56) and GLAK(FAM)RG 20 (calc mass 958.7 Da, obs. Mass 959.6 Da; SEQ ID NO:57) in 1 :10 molar ratio by NA-Vy2(b) 4 to yield ligation product Ac-RYANGLAK(FAM)RG 21 (calc mass 1506.0 Da, obs. Mass 1507.0 Da; SEQ KID NO:58). (B) C-terminal fluorescent labeling of recombinant protein DARPin9_26-NGL 22 (calc mass 19968 Da, obs. mass 19950 Da; SEQ ID NO:49) by GLAK(FAM)RG 20 (SEQ ID NO:57) in 1 :5 molar ratio to give fluorescent protein DARPin9_26-NGLAK(FAM)RG 23 (calc mass 20728 Da, obs. mass 20703 Da). Reaction products were analyzed by MALDI-TOF MS in the positive-ion linear-mode and the crude yield was calculated by peak area.

Detailed description

The present invention is based on the inventors’ identification of novel enzymes having peptide ligase/cyclase activity isolated from Viola yedoensis and Viola canadensis. Specifically, the inventors successfully identified novel ligases from plants of the Violaceae using homology with enzymes capable of ligase activity, such as known enzyme butelase-1 (WO 2015/1 63818 B1 ). These enzymes were named peptide asparagine ligases (PALs), to highlight their specific transpeptidase activity and to differentiate them from AEPs. By purification and testing of the corresponding recombinant enzymes, it was found that solely \/yPAL2 has ligase activity at a wide range of pH values ranging from 4.5 to 8.0 with the maximum catalytic rate at pH 6.5-7.0 that is only 3.5 times less efficient than butelase 1 , and displays minimal hydrolase activity only at acidic pH (4.5), making it a recombinant PAL valuable for biotechnological applications. \/yPAL1 , despite being a good ligase, showed a promiscuous activity with some hydrolysis at acidic pH. \/yPAL3 was characterized by an overall low catalytic efficiency together with a dominant hydrolysis activity at low pH. Moreover, the \/yAEP1 protein, predicted to be a protease based on sequence homology, was indeed found to be a protease at low pH (Fig. 1 ). To reveal the molecular bases for the differences in activities between these enzymes, the crystal structure of \/yPAL2 was obtained and used as a template to model the structure of \/yPAL protein isoforms. These comparisons pointed to two areas surrounding the S1 active site pocket: the S2 and the ST pockets that show subtle but critical variations between an AEP and a PAL.

One residue of OaAEPI b located in the S2 pocket, was previously reported to be a“gate-keeper”, as it was found to play an important role in controlling enzyme efficiency (Yang, et al. (2017) JACS. Doi :10.1021 /jacs.6b12637). The inventors found that this residue is commonly a glycine, while it appears to be a hydrophobic or bulky residue in PALs, such as valine in butelase-1 and cysteine in OaAEPI b. However, using the nature of the“gate-keeper” residue as the only criterion is insufficient to explain the range of activities observed in \/yPAL1 -3 isoforms: VyPAL2 a very efficient PAL and VyPAL3, a very poor enzyme both have similar gate-keeper residues: e.g. I and V respectively (Fig. 6A). Moreover, \/cAEP that has Val (like butelase 1 ) as a gate-keeper residue is a protease (Fig. 5A). The inventors identified sequence variations in two regions of \/yPAL1 -3 that act as ligase activity determinants (LADs) : (i) the S2 pocket (LAD1 ) comprising residues W243, I244 (the gate-keeper) and T245 in \/yPAL2 and (ii) the S2’ pocket (LAD2) including residues A174 and P175 (Fig. 6). The LAD2 region of \/yPAL1 and \/yPAL2 is identical, while their Gate-keeper region (LAD1 ) bears two variations (Fig. 6): The T245A substitution was hypothesized to have little effect since the side chain of residue 245 is oriented opposite to the substrate binding area. It was therefore concluded that the difference in activity observed between \/yPAL1 and 2 is due to the other W243L substitution making the enzyme“leakier” for hydrolysis and explaining the slight shift of \/yPAL1 towards hydrolase at lower pH.

Compared to the VyPALI and 2 ligases, \/yPAL3 has variations at both LAD1 and LAD2. However, the conservative substitutions at LAD1 : V245 instead of an lie gate-keeper residue and V246 instead of Thr ( \/yPAL2) are unlikely to account for the drastic change of activity that was observed (Fig. 1 C). Rather, on the other side of the active site, in LAD2, the AP dipeptide present in \/yPAL1 and 2 is replaced by the bulkier YA dipeptide. This variation was found to be responsible for the lower ligase activity observed in \/yPAL3 compared to \/yPAL1 and \/yPAL2, as the bulky Tyr residue at this position could hamper the access of a peptidyl nucleophile to the acyl-enzyme intermediate. Confirming this, the inventors found that by inserting a smaller hydrophobic side chain such as Gly (or Ala) at the first position of LAD2 ligase efficiency could be significantly increased, as seen in the corresponding \/yPAL3 single Y1 75G mutant (Fig. 4). Importantly, the inventors could confirm the involvement of the LAD2 region in controlling the ligase activity by introducing an equivalent mutation into VcAEP from Viola Canadensis (compare Fig. 5A and B). The volume occupied by Y at this position in VcAEP is likely to cause adverse effects such as accelerating the dissociation of leaving group and slowing down the binding of an incoming peptide, that are essential steps to displace the catalytic water molecule and thus favor ligation over hydrolysis. This is in line with the importance of the interactions at the prime side that was proposed previously to favor cyclization by preventing premature thioester hydrolysis. On the other hand, the side chain of Tyr175 should not disturb the putative catalytic water molecule. This water molecule is presumably located right above Gly1 74 of VyPal3 -a strictly conserved residue immediately following the catalytic His- as observed in the cases of AtLEG-gamma and other legumains (Zauner, et al. (2018) J. Biol. Chem. Doi :10.1074/jbc.M1 17.817031 )

The inventors found that the mechanism of AEPs and PALs can be decomposed in two steps: (i) Acyl-enzyme thioester intermediate formation which is likely the rate-limiting step and (ii) Nucleophilic attack by a water molecule (hydrolysis) or nucleophilic peptide (ligation) on the acyl- enzyme intermediate. Together with the known information on gate-keeper mutagenesis performed on OaAEPI b (Yang, supra), the obtained results set forth in the examples show that hydrophobic residues such as Val/lle/Cys/Ala at this central position favor ligation, while the presence of Gly favors proteolysis (Figs. 1 and 6). The LAD1 in S2 pocket may affect substrate positioning with an impact on enzyme activity, possibly by inducing some specific conformational strain in the substrate. Since changes at the gate-keeper mainly affect substrate binding and positioning, they will have a direct impact on intermediate formation and thus on the overall reaction rate. Conversely, changes in LAD2 would affect the nature and accessibility of the nucleophile and as a consequence, be decisive on the nature of the overall reaction catalyzed.

LAD2 was found to be a crucial determinant for the nature of the activity catalyzed by VyPALs and VcAEP: A bulky residue on this side of the active site, such as Tyr at the first position of the YA dipeptide in \/yPAL3 and YP in \/cAEP, facilitates the departure of the cleaved peptide group, which thus results in recruitment of the catalytic water and exposing the acyl-enzyme thioester to nucleophilic water molecules. This mechanism is in line with earlier studies that showed that a cleaved peptide group remaining in the ST and S2’ pockets displaced the nucleophilic water molecule and thus favors ligation over hydrolysis. Moreover, a bulky residue oriented in the direction where the incoming nucleophilic peptide would bind, hampers access to the acyl-enzyme intermediate, and thus severely reduces the rate of ligation. Conversely, small hydrophobic dipeptides like GA/AA/AP in the LAD2 retain the departing group (blocking access to the thioester bond), until another peptide acts as a nucleophile, leading to ligase activity. However, it was also found that mutations of both sites of VyPkl2 to engineer an AEP did not result in an efficient and drastic conversion into a protease like OaAEP2 or butelase 2, suggesting the existence of other determinants for proteolysis, beside LAD1 and LAD2 (data not shown). One attractive possibility is that residues within LAD1 (the gatekeeper) LAD2 (this work) and MLA cooperate to determine protease vs ligase activity. In this respect, it was found that the presence of a truncated MLA alone (Chen et al. (1 998) FEBS Lett 441 (3):361 -5) does not necessarily imply a ligase activity, because VcAEP, which possesses a truncated MLA (Fig. 6) displays mainly protease activity (Fig. 5).

In summary the inventors discovered that the molecular determinants governing asparaginyl endopeptidases and ligases activity are primarily found in the amino acid composition of the substrate-binding grooves flanking the S1 pocket, in particular the LAD1 and LAD2 that are centered around the S2 and ST pockets, respectively. Combining structural analysis and mutagenesis study, it was uncovered that, for an efficient peptide asparaginyl ligase, the first position of LAD1 is preferably bulky and aromatic, such as W/Y, and the second position hydrophobic, such as V/l/C/A but not G. For LAD2, it was found that GA/AA/AP dipeptides are favored. A bulky residue such as Y is disadvantageous at the first position of LAD2, as it is likely to destabilize the acyl-enzyme intermediate, by affecting the binding affinity of substrates and controlling the accessibility of water molecules and by increasing the dissociation rate of the cleaved peptide tail after the N/D residue. Therefore, a small residue such as G or A at the first position of this dipeptide is a necessary, although not always sufficient, condition for ligase activity. As long as this condition is met, a natural AEP is amenable to become a PAL through mutations or changes at other locations such as LAD1 (gate keeper) or more remote regions like the MLA.

Based on the above findings, the invention, in a first aspect, covers polypeptides having peptide asparaginyl ligase (PAL) activity in isolated form and, more specifically, is directed to an isolated polypeptide comprising, consisting essentially of or consisting of the amino acid sequence as set forth in SEQ ID NO:1 . The polypeptide consisting of the amino acid sequence set forth in SEQ ID NO:1 is also referred to as“VyPAL2” or“VyPAL2 active form/domain” herein.“Isolated”, as used herein, relates to the polypeptide in a form where it has been at least partially separated from other cellular components it may naturally occur or associate with. The polypeptide may be a recombinant polypeptide, i.e. polypeptide produced in a genetically engineered organism that does not naturally produce said polypeptide. Both native and recombinant polypeptides are post- translationally modified by N-linked glycosylation.

A polypeptide according to the present invention exhibits protein ligation activity, i.e. it is capable of forming a peptide bond between two amino acid residues, with these two amino acid residues being located on the same or different peptides or proteins, preferably on the same peptide or protein so that said ligation activity cyclizes said peptide or protein. Accordingly, in various embodiments, the polypeptide of the invention has cyclase activity. In various embodiments, this protein ligation or cyclase activity includes an endopeptidase activity, i.e. the polypeptide form a peptide bond between two amino acid residues following cleavage of an existing peptide bond. This means that cyclization need not to occur between the termini of a given peptide but can also occur between internal amino acid residues, with the amino acids C-terminal or N-terminal to the amino acid used for cyclization being cleaved off. In a preferred embodiment, the polypeptide forms a cyclized peptide by ligating the N-terminus to an internal amino acid and cleaving the remaining C-terminal amino acids.

The polypeptide as disclosed herein is“Asx-specific” in that the amino acid C-terminal to which ligation occurs, i.e. the C-terminal end of the peptide that is ligated, is either asparagine (Asn or N) or aspartic acid (Asp or D), preferably asparagine.

“Polypeptide”, as used herein, relates to polymers made from amino acids connected by peptide bonds. The polypeptides, as defined herein, can comprise 50 or more amino acids, preferably 100 or more amino acids.“Peptides”, as used herein, relates to polymers made from amino acids connected by peptide bonds. The peptides, as defined herein, can comprise 2 or more amino acids, preferably 5 or more amino acids, more preferably 1 0 or more amino acids, for example 10 to 50 amino acids.

In various embodiments, the polypeptide comprises or consists of an amino acid sequence that is at least 60%, 65%, 70%, 75%, 76%, 77%, 78%, 79%, 80%, 81 %, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 90.5%, 91 %, 91 .5%, 92%, 92.5%, 93%, 93.5%, 94%, 94.5%, 95%, 95.5%, 96%, 96.5%, 97%, 97.5%, 98%, 98.5%, 99%, 99.25%, or 99.5% identical or homologous to the amino acid sequence set forth in SEQ ID NO:1 over its entire length. In some embodiments, it has an amino acid sequence that shares at least 60, preferably at least 70, more preferably at least 80, most preferably at least 90 % sequence identity with the amino acid sequence set forth in SEQ ID NO:1 over its entire length or has an amino acid sequence that shares at least 80, preferably at least 90, more preferably at least 95% sequence homology with the amino acid sequence set forth in SEQ ID NO:1 over its entire length.

In various embodiments, the polypeptide may be a precursor of the mature enzyme. In such embodiments, it may comprise or consist of the amino acid sequence set forth in SEQ ID NO:2 or SEQ ID NO:3. Also encompassed are polypeptides having an amino acid sequence that is at least 60%, 65%, 70%, 75%, 76%, 77%, 78%, 79%, 80%, 81 %, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 90.5%, 91 %, 91 .5%, 92%, 92.5%, 93%, 93.5%, 94%, 94.5%, 95%, 95.5%, 96%, 96.5%, 97%, 97.5%, 98%, 98.5%, 99%, 99.25%, or 99.5% identical or homologous to the amino acid sequence set forth in SEQ ID NO:2 or SEQ ID NO:3 over its entire length.

The identity of nucleic acid sequences or amino acid sequences is generally determined by means of a sequence comparison. This sequence comparison is based on the BLAST algorithm that is established in the existing art and commonly used (cf. for example Altschul et al. (1990) “Basic local alignment search tool”, J. Mol. Biol. 215:403-410, and Altschul et al. (1997):“Gapped BLAST and PSI-BLAST: a new generation of protein database search programs”; Nucleic Acids Res., 25, p. 3389-3402) and is effected in principle by mutually associating similar successions of nucleotides or amino acids in the nucleic acid sequences and amino acid sequences, respectively. A tabular association of the relevant positions is referred to as an "alignment." Sequence comparisons (alignments), in particular multiple sequence comparisons, are commonly prepared using computer programs which are available and known to those skilled in the art.

A comparison of this kind also allows a statement as to the similarity to one another of the sequences that are being compared. This is usually indicated as a percentage identity, i.e. the proportion of identical nucleotides or amino acid residues at the same positions or at positions corresponding to one another in an alignment. The more broadly construed term "homology", in the context of amino acid sequences, also incorporates consideration of the conserved amino acid exchanges, i.e. amino acids having a similar chemical activity, since these usually perform similar chemical activities within the protein. The similarity of the compared sequences can therefore also be indicated as a "percentage homology" or "percentage similarity." Indications of identity and/or homology can be encountered over entire polypeptides or genes, or only over individual regions. Homologous and identical regions of various nucleic acid sequences or amino acid sequences are therefore defined by way of matches in the sequences. Such regions often exhibit identical functions. They can be small, and can encompass only a few nucleotides or amino acids. Small regions of this kind often perform functions that are essential to the overall activity of the protein. It may therefore be useful to refer sequence matches only to individual, and optionally small, regions. Unless otherwise indicated, however, indications of identity and homology herein refer to the full length of the respectively indicated nucleic acid sequence or amino acid sequence.

In various embodiments, the polypeptide described herein comprises the amino acid residue N at the position corresponding to position 19 of SEQ ID NO:1 ; and/or the amino acid residue H at the position corresponding to position 124 of SEQ ID NO:1 ; and/or the amino acid residue C at the position corresponding to position 166 of SEQ ID NO:1 . In various embodiments, at least the catalytic dyad formed by the amino acid residue H at the position corresponding to position 124 of SEQ ID NO:1 ; and/or the amino acid residue C at the position corresponding to position 166 of SEQ ID NO:1 is present, preferably in combination with the amino acid residue N at the position corresponding to position 19 of SEQ ID NO:1 , thus forming the complete catalytic triad. It has been found that these amino acid residues are necessary for the catalytic activity (ligase/cyclase/endopeptidase activity) of the polypeptide. In preferred embodiments, the polypeptides thus comprise at least two, more preferably all three of the above indicated residues at the given or corresponding positions.

All amino acid residues are generally referred to herein by reference to their one letter code and, in some instances, their three letter code. This nomenclature is well known to those skilled in the art and used herein as understood in the field.

In various embodiments, the polypeptide described herein comprises the amino acid residue A at the position corresponding to position 126. In various embodiments, the polypeptide described herein comprises the amino acid residue A or P, preferably P, at the position corresponding to position 127 of SEQ ID NO:1 . Alternatively, the amino acid residue at the position corresponding to position 126 of SEQ ID NO:1 may be G. In these embodiments, the amino acid residue at the position corresponding to position 127 of SEQ ID NO:1 is preferably A. These motifs AP, AA and GA are also referred to herein as Ligase Activity Determinant 2 (LAD2), as they are critical determinants for the ligase activity and mutation of other amino acids at these positions to these motifs may convert an endopeptidase to a ligase enzyme in that its predominant enzymatic activity is switched. In various embodiments the motif at the positions corresponding to positions 126 and 127 of SEQ ID NO:1 is not GP, but either AP, AA or GA.

In various embodiments, the polypeptide described herein comprises the amino acid residue W or Y at the position corresponding to position 195, the amino acid residue I or V at the position corresponding to position 196, and the amino acid residue T, A or V at the position corresponding to position 197 of SEQ ID NO:1 . It has been found that this motif W-l/V-T/A/V, also referred to herein as Ligase Activity Determinant 1 (LAD1 ), is also a critical determinant for the ligase activity. In addition to the known gatekeeper position that corresponds to position 196 in SEQ ID NO:1 , it has been found that also positons 195 and 197, in particular 195, are relevant for determining ligase/endopeptidase activity. Again, mutation of other amino acids at these positions to these motifs may convert an endopeptidase to a ligase enzyme in that its predominant enzymatic activity is switched or increase the ligase activity of a mixed ligase/endopeptidase.

In various embodiments, the polypeptide described herein comprises the amino acid residues R at the position corresponding to position 21 , H at the position corresponding to position 22, D at the position corresponding to position 123, E at the position corresponding to position 164, S at the position corresponding to position 194, and D at the position corresponding to position 215 of SEQ ID NO:1 . These amino acid residues are also referred to herein as“S1 pocket”.

In various embodiments, the polypeptide described herein comprises the amino acid residues C at the positions corresponding to positions 199 and 212 of SEQ ID NO:1 . These two residues typically form a disulfide bridge in the mature polypeptide.

The polypeptide of the invention may, in various embodiments, comprise further more or less invariable sequence elements, such as the poly-Pro loop (PPL). Said loop has the consensus sequence P/A-G/T/S-X-X-P/E-G/D/P-V/F/A/P-P-L/P/A/E-E and comprises at least 2 and up to 5 proline residues. Typical are 2, 3, 4 or 5 proline residues at the indicated positions. The PPL occupies positions 200-208 of SEQ ID NO:1 .

Another motif that may be present in the polypeptides of the invention is the so-called MLA motif spanning residues 244-249 of SEQ ID NO:1 . This may have the sequence KKIAYA or NKIAYA (SEQ ID Nos. 15 and 1 6).

In various embodiments, the polypeptides of the invention comprise the LAD1 and LAD2 motifs as described above. In further embodiments, they additionally comprise one, two, three or all four of the S1 pocket, SS bridge, PPL and MLA motif, as defined above. In various embodiments, the isolated polypeptide of the invention can be activated by acid treatment at a pH of 5.0 or less, preferably 4.5 or less. This applies to those polypeptides that comprise a C-terminal cap sequence or activation domain. Such C-terminal domain is, for example, present in the polypeptides having the amino acid sequence set forth in SEQ ID NO:3. The concrete sequence used therein (SEQ ID NO:17) is derived from the cap sequence of VyPAU (SEQ ID NO:5)

The isolated polypeptides of the present invention preferably have enzymatic activity, in particular protein ligase, preferably cyclase activity. In various embodiments, this means that they can ligate a given peptide with an efficiency of 60 % or more, preferably 70 % or more, more preferably 80% or more. The efficiency is determined as the amount of a given peptide/polypeptide cyclized relative to the total amount of said peptide/polypeptide in %.

It is preferred that the polypeptides of the invention have at least 50 %, more preferably at least 70, most preferably at least 90 % of the protein ligase activity of the enzyme having the amino acid sequence of SEQ ID NO:1 .

In various embodiments, the isolated polypeptide of the invention is capable of cyclizing a given peptide with an efficiency of 60 % or more, preferably 80 % or more, preferably at a pH of 5.5 or higher. The cyclization activity may also be determined at pH values of 6.0, 6.5, 7.0, 7.5 or higher. This is relevant, since at low pH conditions, such as below pH 5, many ligases may exhibit a certain degree of endopeptidase activity.

In various embodiments, the polypeptides of the invention hydrolyze a given peptide with an efficiency of 20 % or less, preferably 5 % or less. The efficiency is determined as the amount of a given peptide/polypeptide hydrolyzed relative to the total amount of said peptide/polypeptide in %. Again, since the pH may influence the activity, hydrolysis activity is preferably determined at a pH of 5.5 or higher, for example at pH values of 6.0, 6.5, 7.0, 7.5 or higher

In addition to the above-described modifications, polypeptides according to the embodiments described herein can comprise amino acid modifications, in particular amino acid substitutions, insertions, or deletions. Such polypeptides are, for example, further developed by targeted genetic modification, i.e. by way of mutagenesis methods, and optimized for specific purposes or with regard to special properties (for example, with regard to their catalytic activity, stability, etc.). If such additional modifications are introduced into the polypeptides of the invention, these preferably do not affect, alter or reverse the sequence motifs detailed above, i.e. the catalytic residues, the LAD1 and LAD2 motifs. This means that the above-defined features of these residues/motifs are not changed by these additional mutations beyond that what is defined above. It can be further preferred that additionally one, two, three or all four of the S1 pocket, SS bridge, PPL and MLA motif are retained without additional modifications, i.e. modifications going beyond those detailed above. In addition, nucleic acids contemplated herein can be introduced into recombination formulations and thereby used to generate entirely novel protein ligases, cyclases or other polypeptides.

In various embodiments, the polypeptides having ligase/cyclase activity may be post- translationally modified, for example glycosylated. Such modification may be carried out by recombinant means, i.e. directly in the host cell upon production, or may be achieved chemically or enzymatically after synthesis of the polypeptide, for example in vitro.

For example, the known PAL butelase-1 (SEQ ID NO:18) is glycosylated at N94 and N286 with bulky heterogeneous glycans, which results in an increase of additional mass of about 6 kDa. The recombinant VyPAL2 (SEQ ID Nos. 1 -3) is glycosylated at positions N102, N145 and N237, using the numbering of SEQ ID NO:2, with small glycans, and which results in an additional increased mass of about 3 kDa. The polypeptides of the invention may thus be glycosylated with bulky, heterogeneous glycans, for example at positions corresponding to positions N94 and N286 of SEQ ID NO:1 8 or with small glycans at positions corresponding to positions N1 02, N145 and N237 of SEQ ID NO:2.

The objective of the described modifications may be to introduce targeted mutations, such as substitutions, insertions, or deletions, into the known molecules in order, for example, to alter substrate specificity and/or improve the catalytic activity. For this purpose, in particular, the surface charges and/or isoelectric point of the molecules, and thereby their interactions with the substrate, can be modified. Alternatively or additionally, the stability of the polypeptide can be enhanced by way of one or more corresponding mutations, and its catalytic performance thereby improved. Advantageous properties of individual mutations, e.g. individual substitutions, can supplement one another.

In various embodiments, the polypeptide may be characterized in that it is obtainable from a polypeptide as described above as an initial molecule by single or multiple conservative amino acid substitution. The term "conservative amino acid substitution" means the exchange (substitution) of one amino acid residue for another amino acid residue, where such exchange does not lead to a change in the polarity or charge at the position of the exchanged amino acid, e.g. the exchange of a nonpolar amino acid residue for another nonpolar amino acid residue. Conservative amino acid substitutions in the context of the invention encompass, for example, G=A=S, l=V=L=M, D=E, N=Q, K=R, Y=F, S=T, G=A=I=V=L=M=Y=F=W=P=S=T.

Alternatively or additionally, the polypeptide may be characterized in that it is obtainable from a polypeptide contemplated herein as an initial molecule by fragmentation or by deletion, insertion, or substitution mutagenesis, and encompasses an amino acid sequence that matches the initial molecule as set forth in SEQ ID Nos. 1 -14 over a length of at least 150, 160, 170, 180, 190, 200, 210, 220, 230, 240, 250, 260 or 265 continuously connected amino acids. It is preferred that in such embodiments, the amino acids N19, H124 and C166, as well as the above-defined LAD1 , LAD2 and optionally also any one or more of the S1 pocket, the PPL, MLA motif and disulfide bridge contained in the initial molecule are still present.

In various embodiments, the present invention thus also relates to fragments of the polypeptides described herein, with said fragments retaining enzymatic activity. It is preferred that they have at least 50 %, more preferably at least 70, most preferably at least 90 % of the protein ligase and/or cyclase activity of the initial molecule, preferably of the polypeptide having the amino acid sequence of SEQ ID NO:1 . The fragments are preferably at least 150 amino acids in length, more preferably at least 200 or 250. It is further preferred that these fragments comprise the amino acids N, H and C at positions corresponding to positions 19, 124 and 166 of SEQ ID NO:1 as well as the above-defined LAD1 , LAD2 and optionally also any one or more of the S1 pocket, the PPL, MLA motif and disulfide bridge contained in the initial molecule. Preferred fragments therefore comprise amino acids 19-197, more preferably 1 9-212, most preferably 19-249 of the amino acid sequence set forth in SEQ ID NO:1 .

The nucleic acid molecules encoding the polypeptides described herein, as well as a vector containing such a nucleic acid, in particular a copying vector or an expression vector also form part of the present invention.

These can be DNA molecules or RNA molecules. They can exist as an individual strand, as an individual strand complementary to said individual strand, or as a double strand. With DNA molecules in particular, the sequences of both complementary strands in all three possible reading frames are to be considered in each case. Also to be considered is the fact that different codons, i.e. base triplets, can code for the same amino acids, so that a specific amino acid sequence can be coded by multiple different nucleic acids. As a result of this degeneracy of the genetic code, all nucleic acid sequences that can encode one of the above-described polypeptides are included in this subject of the invention. The skilled artisan is capable of unequivocally determining these nucleic acid sequences, since despite the degeneracy of the genetic code, defined amino acids are to be associated with individual codons. The skilled artisan can therefore, proceeding from an amino acid sequence, readily ascertain nucleic acids coding for that amino acid sequence. In addition, in the context of nucleic acids according to the present invention one or more codons can be replaced by synonymous codons. This aspect refers in particular to heterologous expression of the enzymes contemplated herein. For example, every organism, e.g. a host cell of a production strain, possesses a specific codon usage. "Codon usage" is understood as the translation of the genetic code into amino acids by the respective organism. Bottlenecks in protein biosynthesis can occur if the codons located on the nucleic acid are confronted, in the organism, with a comparatively small number of loaded tRNA molecules. Also it codes for the same amino acid, the result is that a codon becomes translated in the organism less efficiently than a synonymous codon that codes for the same amino acid. Because of the presence of a larger number of tRNA molecules for the synonymous codon, the latter can be translated more efficiently in the organism.

By way of methods commonly known today such as, for example, chemical synthesis or the polymerase chain reaction (PCR) in combination with standard methods of molecular biology or protein chemistry, a skilled artisan has the ability to manufacture, on the basis of known DNA sequences and/or amino acid sequences, the corresponding nucleic acids all the way to complete genes. Such methods are known, for example, from Sambrook, J., Fritsch, E. F., and Maniatis, T, 2001 , Molecular cloning: a laboratory manual, 3rd edition, Cold Spring Laboratory Press.

"Vectors" are understood for purposes herein as elements - made up of nucleic acids - that contain a nucleic acid contemplated herein as a characterizing nucleic acid region. They enable said nucleic acid to be established as a stable genetic element in a species or a cell line over multiple generations or cell divisions. In particular when used in bacteria, vectors are special plasmids, i.e. circular genetic elements. In the context herein, a nucleic acid as contemplated herein is cloned into a vector. Included among the vectors are, for example, those whose origins are bacterial plasmids, viruses, or bacteriophages, or predominantly synthetic vectors or plasmids having elements of widely differing derivations. Using the further genetic elements present in each case, vectors are capable of establishing themselves as stable units in the relevant host cells over multiple generations. They can be present extrachromosomally as separate units, or can be integrated into a chromosome resp. into chromosomal DNA.

Expression vectors encompass nucleic acid sequences which are capable of replicating in the host cells, by preference microorganisms, particularly preferably bacteria, that contain them, and expressing therein a contained nucleic acid. In various embodiments, the vectors described herein thus also contain regulatory elements that control expression of the nucleic acids encoding a polypeptide of the invention. Expression is influenced in particular by the promoter or promoters that regulate transcription. Expression can occur in principle by means of the natural promoter originally located in front of the nucleic acid to be expressed, but also by means of a host-cell promoter furnished on the expression vector or also by means of a modified, or entirely different, promoter of another organism or of another host cell. In the present case at least one promoter for expression of a nucleic acid as contemplated herein is made available and used for expression thereof. Expression vectors can furthermore be regulated, for example by way of a change in culture conditions or when the host cells containing them reach a specific cell density, or by the addition of specific substances, in particular activators of gene expression. One example of such a substance is the galactose derivative isopropyl-beta-D-thiogalactopyranoside (IPTG), which is used as an activator of the bacterial lactose operon (lac operon). In contrast to expression vectors, the contained nucleic acid is not expressed in cloning vectors.

In a further aspect, the invention is also directed to a host cell, preferably a non-human host cell, containing a nucleic acid as contemplated herein or a vector as contemplated herein. A nucleic acid as contemplated herein or a vector containing said nucleic acid is preferably transformed into a microorganism, which then represents a host cell according to an embodiment. Methods for the transformation of cells are established in the existing art and are sufficiently known to the skilled artisan. All cells are in principle suitable as host cells, i.e. prokaryotic or eukaryotic cells. Those host cells that can be manipulated in genetically advantageous fashion, e.g. as regards transformation using the nucleic acid or vector and stable establishment thereof, are preferred, for example single-celled fungi or bacteria. In addition, preferred host cells are notable for being readily manipulated in microbiological and biotechnological terms. This refers, for example, to easy culturability, high growth rates, low demands in terms of fermentation media, and good production and secretion rates for foreign proteins. The polypeptides can furthermore be modified, after their manufacture, by the cells producing them, for example by the addition of sugar molecules, formylation, amination, etc. Post-translation modifications of this kind can functionally influence the polypeptide.

Further embodiments are represented by those host cells whose activity can be regulated on the basis of genetic regulation elements that are made available, for example, on the vector, but can also be present a priori in those cells. They can be stimulated to expression, for example, by controlled addition of chemical compounds that serve as activators, by modifying the culture conditions, or when a specific cell density is reached. This makes possible economical production of the proteins contemplated herein. One example of such a compound is IPTG, as described earlier.

Preferred host cells are prokaryotic or bacterial cells, such as E. coli cells. Bacteria are notable for short generation times and few demands in terms of culturing conditions. As a result, economical culturing methods resp. manufacturing methods can be established. In addition, the skilled artisan has ample experience in the context of bacteria in fermentation technology. Gram negative or Gram-positive bacteria may be suitable for a specific production instance, for a wide variety of reasons to be ascertained experimentally in the individual case, such as nutrient sources, product formation rate, time requirement, etc. In various embodiments, the host cells may be E.coli cells.

Host cells contemplated herein can be modified in terms of their requirements for culture conditions, can comprise other or additional selection markers, or can also express other or additional proteins. They can, in particular, be those host cells that transgenically express multiple proteins or enzymes.

The host cell can, however, also be a eukaryotic cell, which is characterized in that it possesses a cell nucleus. A further embodiment is therefore represented by a host cell which is characterized in that it possesses a cell nucleus. In contrast to prokaryotic cells, eukaryotic cells are capable of post-translationally modifying the protein that is formed. Examples thereof are fungi such as Actinomycetes, or yeasts such as Saccharomyces or Kluyveromyces or insect cells, such as Sf9 cells. This may be particularly advantageous, for example, when the proteins, in connection with their synthesis, are intended to experience specific modifications made possible by such systems. Among the modifications that eukaryotic systems carry out in particular in conjunction with protein synthesis are, for example, the bonding of low-molecular-weight compounds such as membrane anchors or oligosaccharides. In various embodiments, the host cells are thus eukaryotic cells, such as insect cells, for example Sf9 cells.

The host cells contemplated herein are cultured and fermented in a usual manner, for example in discontinuous or continuous systems. In the former case a suitable nutrient medium is inoculated with the host cells, and the product is harvested from the medium after a period of time to be ascertained experimentally. Continuous fermentations are notable for the achievement of a flow equilibrium in which, over a comparatively long period of time, cells die off in part but are also in part renewed, and the protein formed can simultaneously be removed from the medium.

Host cells contemplated herein are preferably used to manufacture the polypeptides described herein.

A further aspect of the invention is therefore a method for manufacturing a polypeptide as described herein, comprising culturing a host cell contemplated herein; and isolating the polypeptide from the culture medium or from the host cell. Culture conditions and mediums can be selected by those skilled in the art based on the host organism used by resorting to general knowledge and techniques known in the art.

In a still further aspect, the present invention relates to the use of polypeptides described above for protein ligation, in particular for cyclizing one or more peptide(s).

It is understood that while the uses of the enzymes described herein are described in the following by reference to peptide substrates, they can similarly be used for the corresponding polypeptides or proteins. The invention thus also covers embodiments where polypeptides or proteins are used as substrates. These polypeptides or proteins can comprise the structural motifs as described below in the context of peptide substrates. Also encompassed are embodiments, where peptide fragments, such as fragments of human peptide hormones that retain functionality, or peptide derivatives, such as (backbone) modified peptides, including, for example, thiodepsipeptides, are utilized. Accordingly, the present invention also covers fragments and derivatives of the peptide substrates disclosed herein.

In various embodiments the peptide to be ligated or cyclized can be any peptide, typically at least 10 amino acids in length, as long as it contains a recognition and ligation sequence that is recognized, bound and ligated by the ligase/cyclase. This amino acid sequence of the peptide to be ligated or cyclized may comprise the amino acid residue N or D, preferably N. In various embodiments, the peptide to be cyclized or ligated comprises the amino acid sequence (X)oN/D(X)p, with X being any amino acid, o being an integer of 1 or more, preferably 2 or more, and p being an integer of 1 or more, preferably of 2 or more. In a preferred embodiment, (X) _p is X ³ X ⁴ (X)r, H(X) _r or HV(X)rWith X ³ being any amino acid with the exception of P, preferably H, G or S, X ⁴ being a hydrophobic or aromatic amino acid, preferably selected from L, I, V, F, C, W, Y and M, and r being 0 or an integer of 1 or more. In various embodiments, the peptide comprises the amino acid sequence(X) ₀ NH or (X) ₀ NHV or (X) ₀ NGL or (X) ₀ NSL. Said amino acid sequence is preferably located at or near the C-terminus of the peptide to be ligated or cyclized, as all amino acids C-terminal to the N will be cleaved off during ligation/cyclization. Accordingly, in all afore mentioned embodiments, p or r are preferably integers of up to 20, preferably up to 5. Particularly preferred are embodiments, where p is 2, with (X) _p preferably being X ³ X ⁴ (X) _r as defined above, and where r is optionally 0.

In alternative embodiments, the peptide to be ligated or cyclized may comprise the amino acid sequence (X) ₀ N7D ^* , wherein X is any amino acid, o is an integer of at least 2 and the C-terminal carboxy group (of the N or D residue) is replaced by a group of the formula -C(0)-N(R’)2, with R’ being any residue, such as, for example, alkyl. In such embodiments, the terminal -C(0)OH group of the N or D residue, preferably the alpha-carboxy group in case of D, is modified to form the group -C(0)-N(R’)2. These C-terminally amidated D or N residues are indicated herein by D ^* and N ^* , respectively. It has been found that the enzymes disclosed herein can cleave the amide group and ligate said N or D residue to the N-terminus of another peptide of interest or the N-terminus of the same peptide that comprises the N or D residue.

The N-terminal part of the peptide to be ligated preferably comprises the amino acid sequence X ¹ X ² (X) _q, wherein X can be any amino acid; X ¹ can be any amino acid with the exception of Pro; X ² can be any amino acid, but preferably is a hydrophobic amino acid, such as Val, lie or Leu , or Cys; and q is 0 or an integer of 1 or more. Preferred are in the X ¹ position in the following order: G = H >M = W = F = R = A = I = K = L = N = S = Q = C>T = V = Y>D = E.“=” indicates that the respective amino acids are similarly preferred, while“>” indicates a preference of the amino acids listed before the symbol over the ones listed after the symbol. Preferred in the X ² position are in the following order : L > V > I > C > T > W > A = F > Y > M > Q > S. Less preferred in the X ² position are P, D, E, G, K , R, N and H. Particularly preferred in the X ¹ position are G and H and in the X ² position L, V, I and C, such as the dipeptide sequences GL, GV, Gl, GC, HL, HV, HI and HC.

In preferred embodiments, the peptide to be ligated or cyclized thus comprises in N- to C-terminal orientation, the amino acid sequence X ¹ X ² (X) _q (X) ₀ N/D(X) _p , wherein X, X ¹ , X ² , o, p, and q are defined as above, with o preferably being at least 7. In various embodiments, (1 ) q is 0 and o is an integer of at least 7; and/or (2) X ¹ is G or H; and/or (3) X ² is L, V, I or C; and/or (4) p is at least 2 but not more than 22, preferably 2-7, more preferably H(X) _r or HV(X) _r , most preferably HX or HV. In various embodiments, (1 ) q is 0 and o is an integer of at least 7; and (2) X ¹ is G or H; and (3) X ² is L, V, I or C; and (4) p is at least 2 but not more than 22, preferably 2-7, more preferably (X) _P is X ³ X ⁴ (X)r, H(X) _r or HV(X) _r , most preferably HX or HV or XL or GL or XS or LS.

In various embodiments, the peptide to be cyclized is the linear precursor form of a cyclic cystine knot polypeptide, in particular a cyclotide. Cyclotides are a topologically unique family of plant proteins that are exceptionally stable. They comprise ~30 amino acids arranged in a head-to-tail cyclized peptide backbone that additionally is restrained by a cystine knot motif associated with six conserved cysteine residues. The cystine knot is built from two disulfide bonds and their connecting backbone segments forming an internal ring in the structure that is threaded by the third disulfide bond to form an interlocking and cross braced structure. Superimposed on this cystine knot core motif are a well-defined beta-sheet and a series of turns displaying short surface-exposed loops.

Cyclotides express a diversity of peptide sequences within their backbone loops and have a broad range of biological activities. They are thus of great interest for pharmaceutical applications. Some plants from which they are derived are used in indigenous medicines, including kalata-kalata, a tea from the plant Oldenlandia affinis that is used for accelerating childbirth in Africa that contains the prototypic cyclotide kalata B1 (kB1 ). Their exceptional stability means that they have attracted attention as potential templates in peptide-based drug design applications. In particular, the grafting of bioactive peptide sequences into a cyclotide framework offers the promise of a new approach to stabilize peptide-based therapeutics, thereby overcoming one of the major limitations on the use of peptides as drugs.

In various embodiments, the peptide to be cyclized is thus 10 or more amino acids in length, preferably up to 50 amino acids, in some embodiments about 25 to 35 amino acids in length. The peptide to be cyclized may comprise or consist of the amino acid of the precursor of cyclotide kalata B1 from Oldenlandia affinis as set forth in SEQ ID NO:20. In various embodiments, the peptide to be cyclized comprises or consists of the amino acid sequence (X)nC(X)nC(X)nC(X)nC(X)nC(X)nC(X)nNHV(X) _n , wherein each n is an integer independently selected from 1 to 6 and X can be any amino acid . Such peptides are precursors of cyclic cystine knot polypeptides that form cystine bonds between the six cysteine residues, as described above, and which can be cyclized by the enzymes described herein by cleaving off the C-terminal HV(X) _n sequence and ligating the (then C-terminal) N residue to the N-terminal residue.

The peptides to be cyclized may, in various embodiments, include the linear precursors disclosed in US2012/0244575. This document is for this purpose incorporated herein by reference in its entirety.

In various additional embodiments, the peptides to be cyclized include, but are not limited to linear precursors of peptide toxins and antimicrobial peptides, such as bacteriocins, such as bacteriocin AS-48 (SEQ ID NO:19), conotoxins, thanatins (insect antimicrobial peptides) and histatins (human saliva antimicrobial peptides). Other peptides that may be cyclized are precursors of cyclic human or animal peptide hormones, including, but not limited to neuromedin, salusin alpha, apelin and galanin. Exemplary peptides include or consist of any one of the amino acid sequences set forth in SEQ ID Nos. 21 -31 .

Further peptides that can be ligated or cyclized using the enzymes and methods disclosed herein include, without limitation, Adrenocorticotropic Hormone (ACTH), Adrenomedullin, Intermedin, Proadrenomedullin, Adropin, Agelenin, AGRP, Alarin, Insulin-Like Growth Factor-Binding Protein 5, Amylin, Amyloid b-Protein, Amphipathic Peptide Antibiotic, LAH4, Angiotensin I, Angiotensin II, A-Type (Atrial) Natriuretic Peptide (ANP), Apamin, Apelin, Bivalirudin, Bombesin, Lysyl- Bradykinin, B-Type (Brain) Natriuretic Peptide, C-Peptide (insulin precursor), Calcitonin, Cocaine- and Amphetamine-Regulated Transcript (CART), Calcitonin Gene Related Peptide (CGRP), Cholecystokinin (CCK)-33, Cytokine-Induced Neutrophil Chemoattractant-1 /growth-related oncogene (CINC), Colivelin, Corticotropin-Releasing Factor (CRF), Cortistatin, C-Type Natriuretic Peptide (CNP), Decorsin, human neutrophil peptide-1 (HNP-1 ), HNP-2, HNP-3, HNP-4, human defensin HD5, HD6, human beta defensin-1 (hbd1 ), hbd2, hbd3, hbd4, Delta Sleep-Inducing Peptide (DSIP), Dermcidin-1 L, Dynorphin A, Elafin, Endokinin C, Endokinin D, b-Lipotropin, g- Endorphin, Endothelin-1 ,Endothelin-2, Endothelin-3, Big-Endothelin-1 , Big-Endothelin-2, Big- Endothelin-3, Enfuviritide, Exendin-4, MBP, Myelin Oligodendrocyte Protein (MOG), Glu- fibrinopeptide B, Galanin, Galanin-like Peptide, Big Gastrin (Human), Gastric Inhibitory Polypeptide (GIP), Gastrin Releasing Peptide, Ghrelin , Glucagon, Glucagon-like peptide-1 (GLP- 1 ), GLP-2, Growth Hormone Releasing Factor (GRF, GHRF), Guanylin, Uroguanylin, Uroguanylin Isomer A, Uroguanylin Isomer B, Hepcidin, Liver-Expressed Antimicrobial Peptide (LEAP-2), Humanin, Joining Peptide (rJP), Kisspeptin-10, Kisspeptin-54, Liraglutide, LL-37 (Human Cathelicidine), Luteinizing Hormone Releasing Hormone (LHRH), Magainin 1 , Mastoparan, alpha-Mating Factor, Mast Cell Degranulating (MCD) Peptide, Melanin-Concentrating Hormone (MCH), alpha-Melanocyte Stimulating Hormone (alpha-MSH), Midkine, Motilin, neuroendocrine regulatory peptide 1 (NERP1 ), NERP2, Neurokinin A, Neurokinin B, Neuromedin B, Neuromedin C, Neuromedin S, Neuromedin U8, Neuronostatin-13, Neuropeptide B-29, Neuropeptide S (NPS), Neuropeptide W-30, Neuropeptide Y (NPY), Neurotensin, Nociceptin, Nocistatin, Obestatin, Orexin-A, Osteocalcin, Oxytocin, Catestatin, Chromogranin A, Parathyroid Hormone (PTH), Peptide YY, Pituitary Adenylate Cyclase Activating Polypeptide 38 (PACAP-38), Platelet Factor- 4, Plectasin, Pleiotrophin, Prolactin-Releasing Peptide, Pyroglutamylated RFamide Peptide (QRFP), RFamide-Related Peptide-1 , Secretin, Serum Thymic Factor (FTS), Sodium Potassium ATPase Inhibitor-1 (SPAM ), Somatostatin, Somatostatin-28, Stresscopin, Urocortin, Substance P, Echistatin, Enterotoxin STp, Guangxitoxin-1 E, Urotensin II, Vasoactive intestinal peptide(VIP), and Vasopressin as well as fragments and derivatives thereof. The afore-mentioned peptides may be of human or animal, such as rat, mouse, pig, origin. All of them all well-known to those skilled in the art and their amino acid sequences are readily available.

In various other embodiments, polypeptides or proteins of more than 50 amino acids length are used as cyclization substrates. In such a reaction, the polypeptide/protein may be cyclized by ligating its C- to its N-terminus.

In various embodiments, two or more peptides are ligated by the enzymes of the invention. This may include formation of macrocycles consisting of two or more peptides, preferable are macrocyclic dimers. The peptides to be ligated can be any peptides, as long as at least one of them contains a recognition and ligation sequence that is recognized, bound and ligated by the ligase/cyclase. Suitable peptides have been described above in connection with the cyclization strategy. The same peptides can also be used for ligation to another peptide that may be the same or different. One of the peptides to be ligated may for example be a polypeptide that has enzymatic activity or another biological function. The peptides to be ligated may also include marker peptides or peptides that comprise a detectable marker, such as a fluorescent marker or biotin. According to such embodiments, a polypeptide that has bioactivity can be fused to a detectable marker. In various embodiments, at least one of the peptides to be ligated has a length of 25 amino acids or more, preferably 50 amino acids or more (and thus may be a“polypeptide”, in the sense of the present invention).

The peptides to be ligated can comprise or consist of any of the amino acid sequences set forth in SEQ ID Nos. 32 to 42. Preferred peptides to be ligated to form (macrocyclic) dimers include the peptides having the amino acid sequence set forth in any one of SEQ ID Nos. 32-36. Preferred N-terminal peptides to be ligated (with one C-terminal peptide) to form a linear fusion peptide include the peptides having the amino acid sequence set forth in any one of SEQ ID Nos. 22, 25 and 32. Preferred C-terminal peptides to be ligated (with one N-terminal peptide) to form a linear fusion peptide include the peptides having the amino acid sequence set forth in any one of SEQ ID Nos. 23, 24 and 26.

The peptides to be ligated or cyclized can also be fusion peptides or polypeptides in which an Asx-containing tag has been C-terminally fused to the peptide of interest that is to be ligated or fused. The Asx-containing tag preferably has the amino acid sequence N/D(X) _P , as defined above, including the various embodiments. Alternatively, an amidated N or D (N ^* or D ^* as defined above) may be fused to the C-terminal end of the peptide or polypeptide to be ligated or fused. The other peptide to which this fusion peptide or polypeptide is ligated can be as defined above. Alternatively, the fusion peptide or polypeptide may be cyclized by forming a bond between its C- and N- terminus. In one embodiment, the fusion peptide or polypeptide may be green fluorescent protein (GFP) fused to the C-terminal tag of the amino acid sequence NHV (SEQ ID NO: 43) and the ligated peptide may be a biotinylated peptide of the amino acid sequence GIGK(biotinylated)R (SEQ ID NO: 44). Generally, polypeptides and proteins that may be ligated to peptides, such as peptides bearing signaling or detectable moieties, or cyclized using the methods and uses described herein, include, without limitation antibodies, antibody fragments, antibody-like molecules, antibody mimetics, peptide aptamers, hormones, various therapeutic proteins and the like.

In various embodiments, the ligase activity is used to fuse a peptide bearing a detectable moiety, such as a fluorescent group, including fluoresceins, such as fluorescein isothiocyanate (FITC), or coumarins, such as 7-Amino-4-methylcoumarin, to a polypeptide or protein, such as those mentioned above. In various embodiments, the protein can be an antibody fragment, such as a human anti-ABL scFv, for example with the amino acid sequence set forth in SEQ ID NO:45, or an antibody mimetic, such as a darpin (designed ankyrin repeat proteins), for example a darpin specific for human ERK, for example with the amino acid sequence set forth in SEQ ID NO:46.

Use of a detectable marker such as fluorescein or derivatives thereof and/or of a peptide that can easily be radiolabeled with elements 1-125 or 1-131 , this allows using a single reagent imaging of tumors in vivo using PET or SPECT followed by fluorescent detection in organ sections or biopsies.

In still another aspect, the invention relates to a method for cyclizing a peptide, polypeptide or protein, the method comprising incubating said peptide, polypeptide, or protein with the polypeptides having ligase/cyclase activity described above in connection with the inventive uses under conditions that allow cyclization of said peptide.

In a still further aspect, the invention relates to a method for ligating at least two peptides , polypeptides or proteins, the method comprising incubating said peptides, polypeptides or proteins with the polypeptides described above in connection with the inventive uses under conditions that allow ligation of said peptides.

The peptides, polypeptides and proteins to be cyclized or ligated according to these methods are, in various embodiments, similarly defined as the peptides, polypeptides and proteins to be cyclized or ligated according to the above-described uses.

In the methods and uses described herein, the enzyme and the substrate can be used in a molar ratio of 1 :100 or higher, preferably 1 :400 or higher, more preferably at least 1 :1000.

The reaction is typically carried out in a suitable buffer system at a temperature that allows optimal enzyme activity, usually between ambient (20°C) and 40°C.

Immobilizing enzymes on solid supports has a long history with a primary goal of lowering enzyme consumption by repetitively using the same batch of enzymes. In addition, site-separation of solid- phase immobilization reduces aggregation, leading to increased stability and activity of biocatalysts, and simplifies the purification by avoiding contamination of products by enzymes. Consequently, immobilized biocatalysts have been developed for industrial uses to a billion-scale market, such as immobilized lactase in food industry and immobilized lipase in biodiesel production. Compared with conventional industrial processes using chemical catalysts, immobilized enzymes are economically attractive and environmentally friendly.

There are three main-stream immobilization technologies, including attachment to carriers either or non-covalently, physical entrapment, and self-crosslinking. For biocatalysts such as PALs with an exposed substrate-binding surface for biomolecule-based substrates, strategies based on attachment to hydrophilic porous resins by either covalent-binding and affinity-binding methods are direct, convenient, and feasible to facilitate their performance in aqueous conditions.

The thus immobilized peptide ligases are stable, reusable and highly efficient in mediating macrocyclization and site-specific ligation reactions.

The inventors compared different methods to immobilize naturally-occurring butelase-1 and recombinantly expressed asparaginyl ligases VyPAL2. It was surprisingly found that immobilization of PALs overcomes the limitations of soluble enzymes, which include aggregation and autolysis into less active forms, albeit at a very slow rate at near neutral pH. The major advantages of immobilization on a solid support provide site separation and pseudo-dilution to prevent trans-autolytic degradation and enhance stability. The inventors confirmed these major advantages of the immobilized ligases: reusable >1 00 runs with undiminished enzymatic activity, enhanced stability and prolonged shelf-life, and simpler downstream purification process. More importantly, it was found that site-separation of immobilized enzymes permits the use of high enzyme concentrations to accelerate ligation reactions to complete in minutes, such as cyclization, cyclooligomerization and ligation reactions either under one-pot conditions or in a continuous flow- reactor. These advantages bode well in reducing amount of ligases, scale-up use for industrial scale and adaptation for nanodevices.

Accordingly, in one aspect of the invention, in the above-described methods and uses the polypeptides having ligase/cyclase activity may be immobilized on a suitable support material. Suitable support materials include various resins and polymers that are used in chromatography columns and the like. The support may have the form of beads or may be the surface of larger structure, such as a microtiter plate. Immobilization allows for a very easy and simple contacting with the substrate, as well as easy separation of enzyme and substrate after the synthesis. If the polypeptide with the enzymatic function is immobilized on a solid column material, the ligation/cyclization may be a continuous process and/or the substrate/product solution may be cycled over the column.

Accordingly, the present invention, in one aspect, also covers a solid support material comprising the isolated polypeptide according to the invention immobilized thereon. The solid support material may comprise a polymer resin, preferably in particulate form, such as those mentioned above. The isolated polypeptide can be immobilized on the solid support material by covalent or non-covalent interactions. The solid support may be, for example, an agarose bead.

In exemplary embodiments, the polypeptides having ligase/cyclase activity are glycosylated and may be immobilized by means of concanavalin A (Con A), a lectin (carbohydrate-binding protein) that is isolated from Canavalia ensiformis (jack bean). It binds specifically to a-D-mannose and a-D-glucose containing biomolecules, including glycoproteins and glycolipids. Said ConA protein is used in immobilized form on affinity columns to immobilize glycoproteins and glycolipids. Accordingly, in various embodiments, the isolated polypeptide having ligase/cyclase activity is glycosylated and non-covalently bound to a carbohydrate-binding moiety, preferably concanavalin A, coupled to the solid support material surface. Embodiments of glycosylated polypeptides of the invention have been described above.

The solid support materials described above can be used for the on-column cyclization and/or ligation of at least one substrate peptide or in a method for the cyclisation or ligation of at least one substrate peptide, comprising contacting a solution comprising the at least one substrate peptide with the solid support material described above under conditions that allow cyclization and/or ligation of the at least one substrate peptide. The substrate peptides are those described above and include also the above polypeptide substrate. In various embodiments, the polypeptide having ligase or cyclase activity is glycosylated and the immobilization is facilitated by interaction with a carbohydrate-binding moiety, preferably a concanavalin A moiety or variant thereof, covalently linked to the solid support. In such embodiments, the polypeptide of the invention may be butelase-1 (comprising the amino acid sequence of SEQ ID NO:1 8 (active fragment) or SEQ ID NO:88 (full length sequence)) and the solid support may be an agarose bead.

In various other embodiment, the polypeptide having ligase or cyclase activity is biotinylated and the immobilization is facilitated by interaction with a biotin-binding moiety, preferably a streptavidin, avidin or neutravidin moiety or variant thereof, covalently linked to the solid support. Functionalization of the polypeptide with the biotin may be achieved using methods known in the art, such as functionalization with a biotin ester with N-Hydroxysuccinimide (NHS), such as succinimidyl-6-(biotinamido)hexanoate. In such embodiments, the polypeptide may be VyPAL2 having the amino acid sequence of SEQ ID NO:1 or 2 or variants thereof as defined herein. The solid support may be an agarose bead and the biotin-binding moiety may be an avidin variant, such as neutravidin (deglycosylated avidin).

In various other embodiments, the polypeptide having ligase or cyclase activity is immobilized on the solid support by reaction of free amino groups in the polypeptide, for example from lysine side chains, with an N-hydroxysuccinimide functional group on the surface of the solid support. The solid support may be agarose beads and the polypeptide may be VyPAL2 having the amino acid sequence of SEQ ID NO:1 or 2 or variants thereof as defined herein.

In various further aspects, the invention also features a method for increasing the protein ligase activity of a polypeptide having asparaginyl endopeptidase (AEP) activity, the method comprising the steps of substituting the amino acid residue at the position corresponding to position 126 of SEQ ID NO:1 with either a small hydrophobic residue or a G residue, preferably an A or a G residue. In various embodiments, the amino acid residue at the position corresponding to position 127 of SEQ ID NO:1 is A, in particular if the position corresponding to position 126 in SEQ ID NO:1 is G. In various embodiments, the amino acid residue at the position corresponding to position 127 of SEQ ID NO:1 is P, if the position corresponding to position 126 in SEQ ID NO:1 is A. In various embodiments the motif at the positions corresponding to positions 126 and 127 of SEQ ID NO:1 is not GP, but either AP, AA or GA. If the amino acid at the position corresponding to position 127 in SEQ ID NO:1 is not such that the motif AP, AA or GA is obtained, it may be substituted, too. As described above, it has been found that said position(s) within the LAD2 motif is a critical determinant for enzyme directionality, with GA and AP generally yielding enzymes with predominantly or exclusively ligase functionality. In various embodiments, said method may also be a method for producing a polypeptide having protein ligase activity, the method comprising:

(i) providing a polypeptide having asparaginyl endopeptidase (AEP) activity; and

(ii) introducing one or more amino acid substitutions into the polypeptide having

asparaginyl endopeptidase (AEP) activity, wherein said substitutions comprise substituting the amino acid residue at the position corresponding to position 126 of SEQ ID NO:1 with an A or a G residue and optionally substituting the amino acid residue at the position corresponding to position 127 of SEQ ID NO:1 with either a P or an A residue, such that the amino acid sequence in the positions corresponding to positions 126/127 in SEQ ID NO:1 is either GA, AA or AP, preferably GA or AP.

Again, in such methods, the amino acid residue at the position corresponding to position 126 of SEQ ID NO:1 is A, in particular if the position corresponding to position 126 in SEQ ID NO:1 is G, or the amino acid residue at the position corresponding to position 127 of SEQ ID NO:1 is P, if the position corresponding to position 126 in SEQ ID NO:1 is A, with the motif at the positions corresponding to positions 126 and 127 of SEQ ID NO:1 being not GP, but preferably either AP, AA or GA. If the amino acid at the position corresponding to position 127 in SEQ ID NO:1 is not such that the motif AP, AA or GA is obtained, the method may comprise substituting said position in step (ii).

The polypeptide subjected to said method for increasing its ligase/cyclase activity may be an asparaginyl endopeptidase (AEP) and may, in various embodiments, comprise or consist of an amino acid sequence that shares at least 60, preferably at least 70, more preferably at least 80, most preferably at least 90 % sequence homology or sequence identity with the amino acid sequence set forth in any one of SEQ ID Nos. 10-14 (VyAEP1 -4; VcAEP) over its entire length. In various embodiments, the to-be-mutated polypeptide has an amino acid residue at the position corresponding to position 126 of SEQ ID NO:1 that is neither G nor A, such that the positions corresponding to positions 126 and 127 of SEQ ID NO:1 do not have the sequence motif GA or AP. It may however have the motif GP, which can then be substituted by GA, AA or AP in the described methods.

The invention also encompasses a transgenic organism, such as a plant, comprising a nucleic acid molecule encoding a polypeptide having protein ligase and/or cyclase activity as described herein. The polypeptide is preferably not naturally present in said host organism or host plant. Accordingly, the present invention also features transgenic organism/plants - with the exception of human beings - that express a heterologous polypeptide according to the invention. In various embodiments such transgenic organisms/plants may further comprise at least one nucleic acid molecule encoding one or more peptides to be cyclized or one or more peptides to be ligated. These may be peptides as defined above in connection with the uses and methods of the invention. In one embodiment, the peptide to be cyclized is a linear precursor form of a cyclic cystine knot polypeptide, for example like those defined above. These precursors of peptides or polypeptides to be cyclized may be naturally present in said organism/plant but are preferably also artificially introduced, i.e. the nucleic acids encoding them are heterologous.

Such transgenic organisms/plants may, due to the co-expression of the enzyme and its substrate, therefore directly produce a cyclized peptide of interest.

All embodiments disclosed herein in relation to the polypeptides and nucleic acids are similarly applicable to the uses and methods described herein and vice versa. The invention is further illustrated by the following non-limiting examples and the appended claims.

Examples

Materials and Methods

RNA extraction and construction of Vy transcriptome and search of AEP analogs

Fresh Viola yedoensis fruits harvested in early September were subjected to RNA extraction by Trizol method and RNA samples were subjected to lllumina Hiseq sequencing (service provided by Beijing Genetic Institute). The sequenced database has been deposited to NCBI SRA database with accession no. PRJNA494974. After assembly using Trinity, data containing 14.69 GB bases was generated that gave 86,674 Unigenes. Using butelase 1 proenzyme amino acid sequence for homology search using the blastp server, eleven AEP-like mRNA sequences with E value lower than 1 e ^-103 , including six complete sequences containing starting and stop codons, three partial sequence with a full functional core domain and N- or C-terminal missing sequences and two truncated sequences with incomplete core domain were identified. Sequence alignment was performed using ClustalW in BioEdit. A search using butelase 1 proenzyme sequence resulted in over 500 hits with >60% sequence identity and >90% sequence coverage.

Cloning, recombinant expression and purification of VyAEP/PALs and VcAEP in bacteria

\/yAEP1 (Vy= Viola yedoensis), \/yPAL1 -3 and VcAEP (Vc = Viola Canadensis) cDNA sequences without the predicted signal peptides were synthesized and cloned into pET28a(+) (GenScript, Beijing, China) with Ndel/Xhol restriction enzymes in-frame with an N-terminal His6-tag (Hemu et al. (2019) PNAS, June 1 1 , 2019, vol. 1 16, no. 24, 1 1737-1 1746). Point mutations were constructed using Q5 mutagenesis kit (NEB). Plasmids were transformed into SHuffle T7 E. coli that constitutively express DsbC and was pre-transformed with Erv1 p expression plasmid pMJS9. Fresh culture of transformed cells with Oϋboo = 0.4 were treated with 0.1 % arabinose for 1 h to induce production of Ervl p followed by 0.1 mM IPTG treatment at 16 °C for 18-24 h to induce expression of the target proteins. Bacterial cells from a volume of 1 L induced cell culture were harvested by centrifugation at 6000 g for 15 min. A volume of 1 0 mL Lysis buffer (50 mM Na HEPES, 0.1 M NaCI, 1 mM EDTA, 5 mM b-mercapto-ethanol, 0.1 % TritonX-100 at pH 7.5) were added to resuspend every 1 g cell pellet. Cell lysis was conducted by sonicating at 50% amplitude with 5s/5s pulse for 20 min on ice. Clarified cell lysate containing soluble proteins was loaded into self-packed column containing 1 mL Complete nickel beads (Roche) that had been pre equilibrated with chilled binding buffer (50 mM Na HEPES, 0.1 M NaCI, 1 mM EDTA, 5 mM b-ME, pH 7.5). After washing with 20 mL washing buffer (50 mM HEPES, 50 mM imidazole, 0.1 M NaCI, 1 mM EDTA, 5 mM b-ME, pH 7.5), His6-proteins were eluted with 4 x 2 mL elution buffer (50 mM HEPES, 500 mM imidazole, 0.1 M NaCI, 1 mM EDTA, 5 mM b- mercapto-ethanol, pH 7.5). 10x dilution of eluted proteins were loaded into GE HiTrap Q 5 mL column (GE life sciences) equilibrated with ion-exchange (IEX) buffer A (20 mM sodium phosphate buffer, pH 7.5, 1 mM EDTA, 5 mM mercapto-ethanol). The proteins were eluted with a gradient of IEX buffer B (1 M NaCI in 20 mM sodium phosphate buffer, pH 7.5, 1 mM EDTA, 5 mM b- mercapto-ethanol). The fractions containing the target protein were then concentrated four times prior to injection on a size exclusion chromatography (SEC) column (S75 16/60) equilibrated in 20 mM sodium phosphate buffer pH 7.55, 0.1 M NaCI, 5% glycerol, 1 mM EDTA, 5 mM b- mercapto-ethanol. The protein was then concentrated to about 1 mg/mL (equivalent to 20 mM) and stored at 4 °C or at - 80 °C after addition of 20% sucrose and 0.1 % Tween-20.

Cloning, recombinant expression and purification of VyAEP/PALs in insect cells

cDNAs were cloned into pFB-Sec-NH (Amp ⁺ ) donor vector in frame with a N-terminal His6-TEV tag and transformed into E. coli DHI OBac competent cells (Invitrogen) (Shrestha Bet al. (2008) Methods in Molecular Biology (Clifton, N.J.), pp 269-289). White colonies after X-gal blue/white selection (37 °C, 48 h) were picked for colony-PCR using M13/FBAC2 primer mix and sequencing. Positive colonies were amplified for bacmid production using resuspension, lysis and neutralization buffers from QIAprep kit (Qiagen) followed by isopropanol precipitation. Extracted bacmids were transfected into Sf9 ( Spodoptera frugiperda) insect cells for virus packaging using cellfectin and Grace’s insect media (Gibco, Thermo Fisher Scientific). After 72 h, supernatant containing P0 viruses were harvested for infection. After three rounds of infection and amplification of viruses, 1 L SF9 insect cells at a concentration of 2.5 x10 ⁶ cells/mL were infected with 25 rriL P3 virus and cultured for 72 hours at 27°C, 120 rpm. Media containing secreted proteins were collected by centrifugation at 4000 g for 20 min. The pH of the supernatant was then set at 7.5, before injection on a GE excel affinity purification column (GE life sciences). After binding, the beads were washed using buffer A (20 mM Na HEPES pH 7.5, 1 50 mM NaCI and 5 mM b- mercapto-ethanol). Elution of the target protein was achieved with buffer A supplemented with 500 mM imidazole and the fractions containing the protein were diluted 10 times and subjected to IEX and SEC purification as described above.

Acid-induced auto-activation

Activation was performed under various conditions with a pH buffers ranging from 4 to 7 at interval of 0.5, 50 mM sodium citrate buffer or 50 mM sodium phosphate buffer, with 1 mM EDTA and 5 mM b- mercapto-ethanol), at four temperatures (4, 16, 25 and 37 °C), times (15 min to 16 h) and using several surfactant additives (Tween-20, TritonX-100, /V-lauroylsarcosine, and Brij35, at concentrations of 0.05 mM to 1 mM). The activated samples were analyzed with both SDS-PAGE and activity tests (amount of activated enzyme solution equivalent to 50 nM proenzyme, 20 mM GN14-SL, 20 mM sodium phosphate buffer, pH 6.5, 1 mM DTT, 1 mM EDTA, incubated at 37 °C for 5 min, followed by product formation by MALDI-TOF mass spectrometry). This allowed to determine that the optimal activation condition was acidification at pH 4.5 (50 mM sodium citrate buffer, 1 mM DTT, 1 mM EDTA, 0.1 M NaCI) at 4 °C for 12-16 h with 0.5 mM /V-lauroylsarcosine. Subsequently, active enzymes were purified on a size exclusion chromatography column (S100 16/60) pre-equilibrated at pH 4.0 in the SEC buffer (20 mM sodium citrate buffer, 1 mM EDTA, 5 mM b- mercapto-ethanol, 5% glycerol, 0.1 M NaCI). Fractions containing the target proteins were neutralized to pH 5.0-6.5 after elution and stored at 4 °C or at -80 °C after addition of 20% sucrose until further use.

Determination of auto-activation sites

Activated enzymes were subjected to SDS-PAGE and gel bands containing the active protein (migrating around 33-35 kDa) were cut into thin slice for in-gel digestion. The reduction and alkylation of disulfide bonds was performed in one-pot by addition of 5 mM DTT and 10 mM bromoethylamine in a buffer containing 1 M Tris-HCI at pH 8.6, by heating at 55 °C for 30 min. Tryptic digestion was performed with a quantity of 10 pg/mL trypsin (Pierce, MS grade, Thermo Scientific) at pH 7.8 at 37 °C overnight, which resulted in peptide bond cleavage mainly after Arg, Lys and Cys-ethylamine. Digested peptides were extracted from gel pieces with 50% acetonitrile (0.1 % formic acid) and solvents were removed by Speedvac. Digested peptides were redissolved in 1 % formic acid and subjected to LC-MS/MS sequencing on a Dionex UltiMate 3000 UHPLC system (Thermo Scientific Inc., Bremen, Germany) linked to Orbitrap Elite mass spectrometer (Thermo Scientific Inc., Bremen, Germany) as earlier described (Hemu X, et al. (2018) Methods Mol Biol. 2018;1719:379-393; Serra A, et al. (2016) Sci Rep 6(1 ): 23005). Peptides were fragmented using higher-energy collisional dissociation (HCD). Resultant spectra from tryptic digestions were analyzed using PEAKS studio (version 7.5, Bioinformatics Solutions, Waterloo, Canada) where 10 ppm MS and 0.05 Da MS/MS tolerances were applied. Quality of peptide spectra was evaluated manually.

Characterization of enzyme activity at various pH values

Enzyme activity was examined using purified active enzymes with protein concentration determined by A ^280nm absorbance (NanoDrop™ 2000 Spectrophotometer, Thermo Fisher Scientific). Reaction mixture containing 40 nM active enzyme, 20 mM substrate GN14-SL in the reaction buffers with pH values ranging from 4.5-8.0 (20 mM sodium citrate buffer or 20 mM sodium phosphate buffer, with 1 mM EDTA, 5 mM b- mercapto-ethanol) were incubated at 37 °C for 10 min and the reaction was quenched by adding 10x vol. of 0.2% trifluoroacetic acid (TFA) to reduce pH to a value < 2. Reaction results were checked preliminarily using MALDI-TOF mass spectrometry and reaction products were quantitated by RP-HPLC on a C4 analytical column (Aries widepore 150 c 4.6 mm, Phenomenex). Peak areas were obtained in LC solution postrun analysis software (Shimadzu).

Substrate specificity and enzyme kinetics

Peptide library 1 contains synthetic peptides GN14-X< _n) S and GD14-X< _n) (GN14 = SEQ ID NO:48) of which X(n) (n = 0-4 residues) were derived from natural cyclotide precursors from Violaceae and Fabaceae species. Peptide library 2 contain 20 synthetic peptide GN12-XL (GN12 = GLYRRGRLYRRN; SEQ ID NO:47) and Peptide library 3 contain 20 synthetic peptides GN12- GX (X for each of the 20 natural amino acids). \/yPAL2-mediated cyclization reactions were performed with a fixed molar ratio of active enzyme: substrate (1 :500) at pH 6.5 at 37 °C for 10 min and the reaction quenched with 0.2% TFA. Each substrate was tested in triplicate and quantitatively analyzed using RP-HPLC.

For kinetic studies, the cyclization reactions were conducted at pH 6.5 at 37 °C with a fixed concentration of active enzymes (10 nM) and various concentrations (2-20 mM) of the substrate GN14-SLAN (SEQ ID NO:48 + SLAN). The yield of cyclization product cGN14 was quantified by RP-HPLC at every 20 s intervals and the initial rate V ₀ (mM/s) was plotted against substrate concentration [S] (mM) to obtain the Michael-Menten curve in order to analyze the kinetic parameters (k _cat and KM) of each enzyme (GraphPad Prism).

Crystallization, data collection and structure determination of VyPAL2

VyPAL2 at a concentration of 10 mg/ml was screened for crystallization. Crystals suitable for X- ray crystallography appeared after 3 to 7 days in 20% PEG 3350, and 0.2 M magnesium formate dihydrate. Crystals were then mounted on a cryo-loop and flash frozen in liquid nitrogen. Diffraction data were collected at 100K on the MX2 Beamline at the Australian Synchrotron. Data processing was performed using the XDS software (Kabsch W (2010) Xds. Acta Crystallogr Sect D Biol Crystallogr 66(2) :125-132). Data collection statistics are shown in Table S1 below. The structure was solved by the molecular replacement method, using OaAEP-C247A (PDB access code: 5H0I (Yang R, et al. (2017) J Am Chem Soc 139(15):5351 -5358) monomer structure as a search probe. A clear solution containing two independent molecules in the asymmetrical unit was obtained using program Molrep (from the CCP4 suite of programs). Refinement was performed using Buster/TNT (GlobalPhasing Ltd) and manual corrections of the model were performed using the Coot program for molecular graphics (CCP4). Structure analysis and figure production were realized using PyMol (Schrodinger). Refinement statistics are presented in Table S1.

Table S1. Data collection and refinement statistics of VyPAL2.

PDB code: 6IDV

Crystallization condition 20% PEG 3350, 0.2M Mg ²⁺ formate trihydrate

Wavelength (A) 0.953723

Resolution (A) 50-2.4 (2.54-2.4)

Space group C 2

156.8/69.8/104.4

Unit cell

90/110.2/90

Measured reflections 159400 (15614) Unique reflections 41528 (4039)

Multiplicity 3.8 (3.8)

Completeness (%) 99.54 (97.63)

Mean l/signa I (/) 5.78 (1 .26)

Rmerge (%) ^a 21.9 (124.5)

CCl/2 (%) ^b 98.3 (48.4)

R-work ^c 19.50 (29.87)

R-free ^d 23.63 (37.81)

Number of non-hydrogen atoms

Macromolecule 7199

Ligands 177

water 427

Protein residues

RMS (bonds, A) 0.009

RMS (angles, °) 1.16

Ramachandran favored (%) 99.5

Ramachandran outliers (%) 0.5

Average B-factor (A ² )

Macromolecules 45.9

Ligands 47.3

solvent 49.1

Values in parenthesis are those for the last shell

a Rmerge = å| /j - < / > |/å/j , where /j is the intensity of an individual reflection, and < / > is the average intensity of that reflection. ^b CC1/2= percentage of correlation between intensities from random half-dataset (P. A. Karplus, K. Diederich, Science 2012, 336, 1030-1033). ^c Rwork = å||F ₀ | - |F _C ||/å|F _C |, where F ₀ denotes the observed structure factor amplitude, and F _c the structure factor amplitude calculated from the model. ^d Rtree is as for Rwor but calculated with 5% of randomly chosen reflections omitted from the refinement. Molecular dynamics (MD) simulation

To obtain the equilibrated position of the modeled peptide substrate bound to \/yPAL2, an initial \/yPAL2-peptide complex modeled from reference (Schechter I, Berger A (1967) Biochem Biophys Res Commun 27(2):157-162) was subjected to an all-atom, explicit-solvent molecular dynamics simulation using NAMD 2.12 (Phillips JC, et al. (2005) J Comput Chem 26(16):1781— 1802). The cyclized aspartic acid of \/yPAL2 was replaced with a regular aspartic acid. The complex was simulated in a water box, where the minimal distance between the solute and the box boundary was 10 A along all three axes. The charges of the solvated system were neutralized with counter-ions, and the ionic strength of the solvent was set to 150 mM NaCI using VMD (Humphrey W, Dalke A, Schulten K (1996) J Mol Graph 14(1 ) :33— 8, 27-8). The fully-solvated system was subjected to conjugate-gradient minimization for 10,000 steps, subsequently heated to 310 K in steps of 5 ps. The system was simulated for a total of 20 ns with the backbone atoms of the protein ligase, as well as the Ca atom of N343 of the peptide constrained, using a harmonic potential of the form U(x) = k (x-Xret) ² , where k is 1 kcal mol ^-1 k ² and x _re t is the initial atom coordinates. Such constraints allow the side chains of \/yPAL2 and the rest of the peptide substrate to move freely. All simulations were performed under the NPT ensemble assuming the CHARMM36 force field for the protein (Best RB, et al. (2012) J Chem Theory Comput 8(9) :3257- 3273) and assuming the TIP3P model for water molecules. Enzyme, beads and substrates

Butelase-1 was extracted from plant material of Clitoria ternatea grown in the local herb garden. After a few rounds of size-exclusion and anion exchange chromatography on HPLC (Shimadzu) as described in previously protocol (Nguyen et al., Nat Protoc 2016, 1 1 (10), 1977-1988), purified butelase-1 was obtained and store in a pH 6.0 buffer containing 20 mM sodium phosphate, 0.15 M NaCI, 5 mM b-mercaptoethanol (b-ME) and 20% sucrose at 4 °C or -80 °C. Recombinant VyPAL2 was expressed using Bac-to-Bac® Baculovirus system (Thermo Fisher Scientific) in Sf9 insect cells via a secretory pathway governed by the N-terminal GP64 signal peptide as previously described (Hemu et al., Proc. Natl. Acad. Sci. U. S. A. 2019, 116 (24), 1 1 737-1 1746). Expressed proenzymes were purified by Nickel-affinity binding on HisTrap Excel column (GE Healthcare), ion-exchange chromatography on HiTrap Q column (GE Healthcare) and size exclusion chromatography on HiLoad Superdex 75 column (GE Healthcare) using an NGC-FPLC System (Bio-Rad). Acid-induced autoactivation was performed at pH 4.5 at 4 °C overnight in the presence of 1 mM Dithiothreitol (DTT) and 0.5 mM N-lauroylsarcosine. Activated enzymes with molecular weight about 35 kDa was purified again by size-exclusion chromatography with a pH 4.0 citrate buffer. Purified active enzymes was stored in a pH 6.5 buffer containing 20 mM sodium phosphate, 0.1 M NaCI, 5 mM b-ME and 20% sucrose at 4 °C or -80 °C.

All beads are from commercial sources. Pierce™ NHS-Activated Agarose beads (Thermo Fisher Scientific) have protein loading of 1 -20 mg protein per ml_. Pierce™ NeutrAvidin™ Agarose beads (Thermo Fisher Scientific) have protein loading of >8 mg biotinylated protein per ml_. Concanavalin A (ConA) Agarose beads (G-Biosciences) have protein loading of 15-30 mg ConA per mL.

All peptide substrates used in the activity assays, including KN14-GL, GN14-HV, GN14-GL, GN14-SLAN, SFTI(D/N)-HV, RV7, and GLAK(FAM)RG (FAM, Fluorescein amidite) were synthesized by Fmoc chemistry on a Liberty-1 microwave synthesizer (CEM) using the protocol described before (Hemu, X. ; Zhang, X.; Tam, J. P., Org. Lett. 2019). The protein substrate AS- 48K (SEQ ID NO:1 9) was also synthesized chemically. Purified AS-48K was dissolved in 8 M urea and underwent refolding by dialysis (Hemu et al. J. Am. Chem. Soc. Comm. 2016, 138 (22), 6968-71 ) The protein substrate DARPin9_26-NGL was cloned into pET28a(+) vector with N- terminal His6-TEV-GLGSG sequence and C-terminal GSGSNGL tail (SEQ ID NO:49). Recombinant expression was done in Shuffle® T7 E. coll (New England Biolabs) after a 24 h induction by 0.1 mM IPTG at 16 °C. Soluble proteins were extracted from cell lysate and purified by Ni-NTA affinity chromatography on a HisTrap HP 5 mL column (GE Life Sciences) and ion- exchange chromatography on a HiTrap Q 5 mL column (GE Life Sciences) using a NGC-FPLC System (Bio-Rad). Thermostability of PALs

1 pg enzyme was mixed with 8x SYPRO orange fluorescent dye (Thermo Fisher Scientific) and diluted to a final volume of 25 mI_ with a series of buffers (50 mM sodium phosphate, 0.1 M NaCI, 5 mM b-ME) with pH ranged from 5 to 8 in a 96-well plate. pH buffers are ThermoFluor Assay was conducted in a Real-Time PCR Detection System (Bio-Rad) with temperature increased from 25 to 85 °C. The melting temperature was calculated by plotting the change of RFU per degree against temperature.

Reactions mediated by soluble and immobilized PALs

All reactions by soluble or immobilized PALs were conducted using a phosphate reaction buffer (20 mM sodium phosphate, pH 6.5, 0.1 M NaCI, 1 mM DTT) with the exception of ConA-Bu1 , of which the ConA-reaction buffer contained additional 5 mM CaCL and 5 mM MgCL. The reaction pH was kept at 6.5 to maximize the enzyme activity. The reducing reagent DTT was added freshly before use. Reactions were performed at room temperature without heating to prevent degradation of reused enzymes. Reaction mixtures were analyzed by MALDI-TOF mass spectrometry or reversed-phase (RP) HPLC.

Immobilization on NHS-activated agarose beads

Enzyme solutions were prepared with cold PBS (pH 7.4) to the concentration of 1 0 mM and added to NHS-Activated Agarose Dry Resin (75 mg requires 1 mL solution) and the preparation was shake slowly at 4 °C for 3 h. The mixture was then loaded into a chilled spin column and the flow through was collected. Beads were wash with 2x bead-volume of wash buffer (20 mM sodium phosphate, 1 mM DTT, 5% glycerol, pH 6.0 or 6.5) and the flow through was collected after binding and washing. Excessive NHS groups were blocked by socking beads in quenching buffer (1 M Tris-HCI, pH 7.4) for 1 h with gentle shaking at 4 °C. Beads were washed again with reaction buffer, followed by the addition of 2x bead-volume of reaction buffer with 20% ethanol and keeping the slurry at 4 °C.

Biotinylation and Immobilization on NeutrAvidin agarose beads

Enzyme solutions were prepared with cold PBS (pH 7.4) containing 5 mM b-ME to the concentration of 10 mM and mixed with 20-fold molar equivalent of Ezlink® NHS-LC-biotin (succinimidyl-6-(biotinamido)hexanoate, spacer length ~2.2 nm, Thermo Fisher Scientific) dissolved in DMF (stock concentration 10 mM). Biotinylation was conducted 4 °C overnight. Excessive NHS-LC-biotin was removed by buffer exchange with PBS (pH 7.4) using Vivaspin 10 kDa MWCO centrifuge concentrators (Sartorius, Germany). After buffer exchange, activity of biotinylated enzymes was compared with the untreated enzymes to show no activity loss. NA agarose beads was equilibrated with pH 6.5 reaction buffer. A mixture of 0.2 mL equilibrated beads with 1 mL biotinylated enzymes (5 mM) was prepared by gentle shaking at 4 °C for 3 h, followed by loading into a chilled spin column and collection of flow through. Beads were washed with 20x bead-volume of reaction buffer, and kept as a slurry at 4 °C in the presence of 2x bead- volume of reaction buffer with 5 mM b-ME and 20% ethanol.

Immobilization on concanavalin A agarose beads

ConA beads were equilibrated with 10 bead-volume of cold equilibration buffer (1 M NaCI, 5 mM MgCl2, 5 mM CaCL, pH 7.2, Mg ions were used here to substitute Mn ions) (Young, N. M., FEBS Lett 1983, 161, 247-250). Enzyme solutions were prepared with ConA reaction buffer to the concentration of 5 mM. One milliliter of enzyme solution was mixed with 1 ml_ of equilibrated ConA beads with slowly shaking at 4 °C for 3 h. The mixture was loaded into a chilled spin column and the flow through was collected. Beads were wash with 20x bead-volume of ConA reaction buffer and kept as a slurry at 4 °C in 2x bead-volume of ConA reaction buffer with 20% ethanol. Activity of immobilized enzymes stored with or without ethanol showed no significant difference.

Determination of immobilization yield

The concentration of unbound proteins in the flow through after binding and washing were determined using Nanodrop 2000 spectrophotometer (Thermo Fisher Scientific) by measuring the UV absorbance at 280 nm. For the flow through after washing, a concentration step using centrifugal filter (Vivaspin 10 kDa MWCO, Sartorius) was performed to bring up the protein concentration readable by spectrophotometer which has sensitivity threshold of 0.008 at 280 nm.

Determination of activity of immobilized PALs

Free butelase-1 (SEQ ID NO:18) and VyPAL2 (SEQ ID NO:2) were prepared in stock concentrations ranged from 1 to 8 mM. In each reaction, 1 mI_ enzyme stock was added to 100 mI_ 0.2 mM KN14-GL so the final enzyme concentrations ranged from 10 to 80 nM. After 5 min incubation at room temperature, the reaction was quenched by adding 0.5% TFA to bring down the pH to 2 and all the reaction solution was injected into analytical RP-HPLC (Aries-C18, 150x4.6 mm, 3 u, Shimadzu). The amount of cKN14 was calculated by the peak area at 220 nm in the HPLC profile. The initial reaction rate V was then calculated by increase of cKN14 concentration per second. A standard curve of reaction rate against enzyme concentration was plotted for each PAL, which reflects the turnover rate of free enzymes in the tested system. For the ease of calculation, the units of enzyme concentration and the corresponding reaction rate were converted based on the stock concentrations of free enzymes, as (mM) and (mM/s), respectively. Activity of immobilized PALs were examined using 10 pL beads in a 1 mL system with the same experimental setting. 100 pL out of 1 mL reaction solution was injected into RP-HPLC for product quantification. The reaction rate of each immobilized PAL was also converted into (pM/s) and compared with the stand curve to calculate the actual effective concentration. The activity ratio was calculated by the rate of immobilized PALs against rate of soluble PALs. Accession codes. The nucleotide sequence for butelase 1 has been deposited in the GenBank database under the accession number KF918345.

Example 1 : Mining AEPs in the Violaceae transcriptomes and initial classification using the “gate-keeper” residue

Violaceae is one of the major cyclotide-producing plant families, suggesting the presence of PALs in their genomes. With the hope of identifying PALs, data mining on two plants from this family, Viola yedoensis (Vy) and Viola canadensis ( Vc ) was performed.

To obtain the transcriptome of V. yedoensis, total RNA was extracted from fresh fruits and sequenced followed by assembly of the database (NCBI SRA accession no. PRJNA494974). Precursor sequences of butelase 1 and OaAEPI b were used to search for sequences homologous to AEPs. A total of eleven AEP precursors were found from the V. yedoensis transcriptome, including six complete sequences, three partial sequences containing an intact core domain and two truncated sequences having an incomplete core domain that were discarded. The transcriptome of Vc is readily available in 1 KP database and an AEP homolog (NJLF- 2006002) named VcAEP was obtained by BLASTp using butelase 1 sequence. In order to cluster the nine Vy sequences and VcAEP, it was chosen to use the nature of the gate-keeper residue as a criterion: It was previously observed that mutation of a Cys residue (Cys-247) near the active site of OaAEPI b (PDB access code: 5H0I), to larger amino acids (Thr, Met, Val, Leu, lie) reduced ligation catalytic efficiency, while mutations to smaller residues such as Ala resulted in over a hundred-fold improved ligation efficiency (Yang R, et al. (2017) J Am Chem Soc 139(15):5351— 5358). Moreover, mutation of this“gate-keeper” residue into Gly resulted in an increased amount of hydrolysis product, suggesting that this site, located in the S2 substrate-binding pocket, plays an important role in modulating the enzyme function. Using butelase 1 amino-acids sequence to search for homologues in the NCBI data-bank returned more than 500 hits that share over 60% sequence identity, with 90% sequence coverage. Among them, more than 95% sequences carry Gly at the gate-keeper site, including both proteases and“dual-functional” ligases, which agreed with the fact that PALs are rare in plant AEPs.

Using this criterion, four V. yedoensis sequences were classified as putative VyAEPs due to the presence of Gly as gate-keeper and designated VyAEP1 -4 (SEQ ID Nos. 10-13). The other five, designated VyPAL1 -5 (SEQ ID Nos. 5-9) as well as VcAEP from V. Canadensis (SEQ ID NO:14), were classified as putative VyPALs as they contain Val (like butelase 1 ) or lie as gate-keeper residues.

Example 2: Production of Active Recombinant VyAEP and VyPALs

Based on sequence identity, these putative AEPs and PALs could be partitioned into four groups: VyAEP 1 and 2 (98.9%), VyAEP3 and 4 (96.2%), VyPALI ,2,4 and 5 (>99%) and VyPAL3. VyPAL3 only shares <70% core sequence identity with other putative \/yPALs but is 94% identical to \/cAEP. \/yAEP1 , \/yPAL1 -3 and \/cAEP were expressed for further studies. Recombinant expression was performed using both bacterial and insect cell systems, and the genes encoding complete amino acid sequences were cloned into the expression vectors, with the signal peptide substituted by a His-tag for affinity purification. Following metal-affinity, ion-exchange, and size- exclusion chromatography (see methods), bacterial and insect cell systems yielded -0.5 mg/L and 10-20 mg/L of purified proenzymes, respectively.

Following purification, proenzymes were subjected to 12-16 h activation at 4 °C, pH 4.5 in the presence of 0.5 mM /V-lauroylsarcosine, 5 mM b-mercaptoethanol, and 1 mM EDTA. Such mild but prolonged treatment allows cleavage and degradation of the cap domain, preventing cap domain re-ligation. Activated enzymes were further purified using size-exclusion chromatography. The auto-activation sites of purified active \/yPAL2 were determined by LC-MS/MS sequencing of the tryptic digested active forms. The Asn/Asp cleavages sites at both ends of the core domain were found to be N43/N46/D48 in the N-terminal pro-domain region, and D320/N333 in the linker region. This confirmed the complete removal of the inhibitory cap domain and the production of a mixture containing active forms, through protein processing at multiple sites.

Example 3: Ligase vs Protease Activity of VyAEPI and VyPAL1 -3

To determine the activity of \/yAEP/PALs, a model peptide substrate termed “GN14-SL” GISTKSIPPISYRNSL (SEQ ID NO:59) with a MW of 1733 Da was prepared. GN14-SL contains the tripeptide recognition motif “NSL” at its C-terminus derived from the precursors of Vy cyclotides and analog of SFTI-1 (Fig. 1A). A fixed enzyme: substrate molar ratio (1 :500) was used in all ligation reactions, which were performed at 37 °C for 10 min, at pH values ranging from 4.5 to 8.0 (at 0.5 intervals). The cyclization of GN14-SL was monitored using MALDI-TOF mass spectrometry. The yields of cyclic product cGN14 (MW: 151 5 Da) and linear product GN14 (MW: 1533 Da) were quantified using RP-HPLC (Fig. 1 B).

Among four PAL enzymes tested, \/yPAL2 exhibited the best ligase activity, and did not produce any hydrolytic product at pH 5.5-8.0. At the optimal pH of 6.5, over 80% cyclization yield was observed (Fig. 1 C). \/yPAL1 also resulted in pure cyclization at pH 6-8 and, at the optimal pH of 7.0, about 80% cyclization yield is obtained. \/yPAL3 displayed dominant hydrolysis activity at pH 4.5-5.5 and dominant ligase activity at pH 6.0-7.0. Its catalytic efficiency was the lowest among the three putative \/yPALs, as only 20% substrate was converted into cyclized product at the optimal pH 7.0 in 10 min. As anticipated, the putative protease \/yAEP1 displayed hydrolysis activity in the tested pH range of 4.5-8, although at near neutral and basic pH of 6.5-8, cyclization became noticeable. All four enzymes displayed varying degrees (2-40%, Fig. 1 C) of protease activity at pH less than 5.0, reflecting the intrinsic proteolytic activity needed for acid-induced auto activation. Next, substrate specificity of \/yPAL2 was studied, using three sets of peptide libraries (Fig. 2). Efficient cyclization required a minimum of three residues as the C-terminal recognition signal Asn-P1’-P2’ (using the Schechter and Berger nomenclature (49). At P1’, small amino acids especially Gly and Ser are favored, but not Pro. The P2’ position favors the presence of hydrophobic or aromatic residues, such as Leu/lle/Phe. The catalytic efficiency of \/yPAL2 was examined using substrate GN14-SLAN (GISTKSIPPISYRNSLAN; SEQ ID NO:60) that gave 274,325 M- ¹ S ¹ when performing at pH 6.5 at 37 °C, which was 3.5-fold less than butelase 1 (971 ,936 M- ¹ s- ¹ ) (Fig. 3).

Example 4: Crystal structure of VyPAL2

To understand the molecular mechanisms responsible for the differences in nature and efficiency between PALs and AEPs identified here, the crystal structure of the VyPkl2 proenzyme at a resolution of 2.4 A was obtained. As expected, the structure displays the pro-legumain fold with the active domain on the N-terminus (residues 51 to 320) and the cap domain on the C-terminus (residues 344 to 483). These two domains are connected by a flexible linker (residues 321 to 343). The asymmetric unit contains two monomers of \/yPAL2, forming a homodimer. In solution, this oligomeric form of VyPfkL is present only at high protein concentrations (>5 mg/mL) as inferred from gel filtration results. As the protein was expressed in insect cells, several asparagine residues on the surface of the protein are glycosylated with one to three N-linked sugars (one N- acetylglucosamine (GlcNac), two GlcNac or two GlcNac and one Fucose) on Asn102, Asn145 and Asn237 respectively. Members of the C13 subfamily share a conserved a-b-a sandwich structure and a His172-Cys214 catalytic dyad located in a well-defined oxyanion hole. Peptide bond cleavage is catalyzed by the Cys thiol, which mediates an N-to-S acyl transfer to give the Asn-(S)-Cys thioester intermediate. The imidazole ring of His acts as a general base to accept a proton from the catalytic Cys.

The structure is similar to other PALs and AEPs such as OaAEPI b (PDB code: 5H0I), AfLEGy (5NIJ, 50BT), Ha AEP1 (6AZT), or butelase 1 (6DHI), with an average root-mean-square deviation of atomic positions (r.m.s.d.) of 1 .0 A. Moreover, comparing the active domain alone returns r.m.s.d. values closer to an average 0.7 A, showing that the core domain structure is strongly conserved. This further indicates that enzyme specificity is due to subtle variations in the substrate binding pockets that influence the stability of the S-acyl intermediate and accessibility of the catalytic water molecule. In the present pro-enzyme form, helix a6 (the first helix in the cap domain) makes an angle of about 90° with the linker peptide. At the junction between the linker region and the a6 helix, Gln343 is anchored inside the oxyanion hole (or S1 pocket). In recent structures of active forms of HaAEPI and AtLEGy, the bound substrate or inhibitor are shifted by a distance of about 2.5 A compared to the linker region and covalently linked to the catalytic cysteine via a thioester bond. Example 5: Modeling the substrate-enzyme interactions using energy minimization

Structures of ligand-bound active forms of both HaAEPI (PDB access code: 50BT) and AtLEGy (6AZT) indicated that only small conformational changes occur after activation of the protein and cap release. Therefore, the active form of the \/yPAL2 ligase using the present crystal structure of \/yPAL2 was modeled and residues Gly52 to Asn326, which are clearly visible in the electron density were included. This is also in line with the boundaries of the VyPkl2 active form determined using LC-MS, namely N43/N46/D48 and D320/N333. To obtain an initial model of a peptide substrate bound to the active form of the \/yPAL2, the structure of the complex between AtLEGy and a peptide inhibitor (Zauner et al. (2018) J Biol Chem 293(23) :8934-8946) having the sequence NH2-LKVIH-NSL-COOH (SEQ ID NO:50) was used. The N-terminal sequence of this peptide corresponds to the original linker sequence and the C-terminal dipeptide is based on substrate specificity studies presented in Fig. 2. Energy minimization of the resulting complex with the peptide was then performed, constraining only the Ca atoms of the active protein. The alpha-carbon atom of the P1 Asn residue was fixed at the position found in AtLEGy and used as an anchor to maintain the substrate in the S1 pocket. Upon MD equilibration of the system for 20 ns, the N-terminal portion of the substrate“LKVIHN” (part of SEQ ID NO:50) was shifted due to repulsion between I244 from VyPAL2 and the substrate. As a result, the alpha-carbon atom of the lie at the substrate P3 position is displaced by 3 A. The C-terminal“SL” dipeptide on the other hand, becomes more extended, leading to a better fit of the peptide into the substrate binding pockets. This more stable and energetically favorable position for the modelled substrate was used to map the ST and S2’ pockets that define the recognition motifs both for protease and ligase activities. By analyzing the interface with the model substrate, residues of the active form of \/yPAL2, that are lining the S4-S2' pockets were defined. The composition of S4 agreed with earlier works on AfLEGy and involved residues from both the disulfide-clamped poly-Pro loop (PPL) equivalent to c341 loop in caspase-1 , and the MLA region (equivalent to c381 loop in caspase-1). On the other side of the S1 pocket, the ST pocket is shaped by the amide groups of H172, G173 and A174 that accommodates the backbone atoms of PT and P2’ residues of the peptide. The S2’ pocket is lined by Y185 and backbone atoms of G179 and M180, which favors binding of hydrophobic residues at the P’2 position. MD simulation shows that the interaction between hydrophobic Leu side-chain of the peptide and the phenol ring of Y1 85 is favored, which is in agreement with the preference for lle/Val/Phe at P2’ observed in the specificity study (Fig. 2C).

Example 6: Identification of ligase-activity determinants in the S2 and ST pockets

Although classified and confirmed as PALs, TyPAL1 -3 displayed various levels of ligase activity in terms of both cyclization/hydrolysis ratio and catalytic efficiency. Thus, the structures of \/yPAL1 and TyPAL3 were modeled using the experimental crystal structure of TyPAL2 as template. The resulting models are likely to be accurate given the sequence identity between these three proteins. Mapping the polymorphic residues on \/yPAL1 -3 structures indicates variations in the substrate-interacting surface located in the S2 and S1’ pockets. One variation lies in the first residue of S2: Leu243 in \/yPAL1 in lieu of the aromatic and bulky Trp present in both VyPAL2 and \/yPAL3. In the same region, position 244 of VyPAL2 is either lie or Val, introducing little variation in local hydrophobicity. Finally, the side-chain of residue at position 245 is facing a direction opposite from the S1 pocket (and the backbone atoms of \/yPAL1 -3 completely overlap), suggesting that this residue has little impact on catalysis. However, on the other side of the S1 pocket, a more drastic difference is observed in the vicinity of S’1 and S’2: Alai 74-Pro175 in both \/yPAL1 and 2 is replaced by Tyr175-Ala176 in \/yPAL3.

Example 7: Selectively improving the ligase activity of VyPAL3 and 1/cAEP

To validate experimentally these structural observations, \/yPAL3 was first targeted: the ΎA” dipeptide in the ST region was mutated into“GA” as found in the butelase 1 sequence. As anticipated, this Y175G point mutation resulted in a strong and selective increase of ligation activity observed at lower pH (4.5-6), when compared with the wild-type \/yPAL3 (Fig. 4). In addition, the catalytic efficiency was also significantly improved, with the maximum cyclization yield increasing from 20% to 80% (compare Fig. 4C and 1 C).

To further validate our hypothesis about the crucial role of the ST region in determining ligase activity, t/cAEP with predominantly protease activity and virtually absent ligase activity was targeted (Fig. 5A). The mutation Y168P169 -> A168P169 in its ST region (equivalent to Y175A176 in TyPAL3) was introduced. The Y168A mutation drastically affected both the type of enzymatic activity and the catalytic efficiency (Fig. 5B) towards the GN14-SLDI substrate. The reaction with the wild-type \/cAEP was performed using an enzyme to GN14-SLDI molar ratio of 1 :200 for 5 h. In contrast, for VcAEP-Y168A the ratio was 1 :2000 and reaction was quenched after 2 min incubation at 37 °C. At near neutral pH, \/cAEP-Y168A was able to convert over 60% substrate into its cyclic form, with less than 5% hydrolysis product formed (Fig. 5B).

Example 8: Preparation of active butelase-1 and VyPAL2

Two different sources of PALs were used, the naturally-occurring and activated butelase-1 isolated from plant (Nguyen et al. Nat. Chem. Biol. 2014, 10 (9), 732-738) and the insect-cell expressed VyPAL2 zymogen which requires an acid-induced step to be activated (Hemu et al. Proc. Natl. Acad. Sci. U. S. A. 2019, 1 16 (24), 1 1 737-1 1746). Butelase-1 used in this study was extracted from fresh plant tissues of Clitoria ternatea and purified via anion-exchange and size- exclusion chromatography as previously described (Nguyen et al. Nat Protoc 2016, 11 (10), 1977- 1988). Recombinant VyPAL2 was expressed in the proenzyme form by baculovirus expression system (Shrestha et al. In Genomics Protocols, Starkey, M. ; Elaswarapu, R., Eds. Humana Press: Totowa, NJ, 2008; pp 269-289) in a secretory pathways using insect cells (Hemu et al., supra) The activated forms of VyPAL2 were obtained by an acid-induced auto-activation at pH 4.5 and purified by size-exclusion chromatography using a sodium citrate buffer at pH 4. Both butelase-1 and expressed VyPAL2 zymogen were glycosylated, and their glycosylated forms appear as bold bands that are larger than the calculated protein weights in SDS-PAGE (data not shown).

Example 9: Non-covalent Immobilization of active PALs

Butelase-1 is glycosylated at N94 and N286 with bulky heterogeneous glycans based on previous studies, which results in an increase of additional mass of about 6 kDa. The recombinant VyPAL2 is glycosylated at N102, N145 and N237 with small glycans, and which results in an additional increased mass of about 3 kDa (data not shown). Thus, lectin-beads were an obvious first choice and the most direct method to immobilize these two glycosylated PALs via affinity attachment. ConA is one of the most commonly and widely used plant lectins (Saleemuddin & Husain Enzyme Microb. Technol. 1991 , 13 (4), 290-295; Rudiger & Gabius Glycoconjugate J. 2001 , 18, 589-613). ConA-attachment is reversible, allowing the recovery of glycoenzymes using elution buffers containing mannosyl and glucosyl monosaccharides (Dulaney Mol. Cell. Biochem. 1978, 21 (1 ), 43-63) (Figure 1 A). For insoluble support, 6% crosslinked agarose beads were used as they are highly porous, hydrophilic, stable, inert to chemical and physical modifications, and their relatively large pore sizes allowing free diffusion of compounds <4000 kDa (Zucca et al. Molecules 2016, 2/ (1 1 )).

Affinity binding of glycoenzymes with ConA beads was performed by mixing 1 mg of freshly prepared ligases and 1 mL beads that were pre-equilibrated with a pH 6.5 ConA-reaction buffer and gently shake at 4 °C for 3 h. The low enzyme loading of 1 mg/mL (equivalent to ~27 mM of ligase) and the gentle shaking could facilitate diffusion of solutes. Beads were washed with ConA reaction buffer after binding. Butelase-1 immobilized on ConA-beads gave ConA-Bu1 1 in 39% yield as 61 % of enzymes remained in the solution. In contrast, ConA-bound VyPAL2 was dissociated from beads quickly after a few rounds of washing. Consequently, ConA-Vy2 2 was excluded in all subsequent experiments.

The observed difference of ConA affinity to butelase-1 and VyPAL2 can be attributed to their glycosylated forms. Plant-derived butelase-1 contains the complex high-mannose /V-glycans that bind to ConA with high affinity (Wilson Curr. Opin. Struct. Biol. 2002, 12 (4), 569-577; Strasser Front Plant Sci 2014, 5, 363). In contrast, the insect cell-expressed VyPAL2 contains simple N- glycans that binds to ConA with low affinity (Shi & Jarvis, Curr Drug Targets. 2007, 8 (10), 1 1 16- 1 125) In the crystal structure of VyPAL2, confirmed glycans are not bigger than tri-saccharides. In addition, ConA-immobilized PALs are not suitable to catalyze reactions containing either soluble sugars or glycoproteins that may bind to ConA and exchange with the immobilized PALs.

For comparison, a second non-covalent immobilization was experimentally tested by exploiting the exceptionally high binding between biotin and avidin. The avidin-biotin binding is considered practically irreversible with dissociation constants in the range of 10 ^-15 M. To eliminate non-specific lectin binding, a deglycosylated form of avidin, NeutrAvidin (NA), was used which retains the strong affinity binding of amine-linked biotins as the glycosylated avidin (Figure 7B). This method required modifying some of the primary amines of PALs with biotins. The sequences of both active butelase-1 and VyPAL2 contain multiple Lys residues, which are not located close to the catalytic site or substrate binding surface. Thus, immobilization of PALs involving Lys-NFL was not expected to hinder the catalytic site of PALs.

To biotionylate the lysine side chains, succinimidyl-6-(biotinamido)hexanoate (NHS-LC-biotin) was used for biotinylation of active butelase-1 and VyPAL2. The coupling reaction of N-hydroxy succimide ester (NHS-ester) to primary amines on the ligases is generally performed in basic conditions with pH ranging from 7.2 to 9.0. Since active PALs, regardless of whether they are plant-produced or insect-cell expressed, are less thermally stable in basic conditions, we performed the biotinylation at pH 7.4 at 4 °C to minimize degradation of the ligases. It was experimentally confirmed that biotinylated enzymes do not show activity loss. Affinity binding of the biotinylated butelase-1 , Bu1 (b), and biotinylated VyPAL2, Vy2(b) with NA beads was performed at pH 6.5 at 4 °C for 3 h. After immobilization, beads were washed with chilled pH 6.5 reaction buffer. This method resulted in 49% and 45% immobilization yield to give NA-Bu1 (b) 3 and NA-Vy2(b) 4, respectively.

Example 10: Covalent Immobilization of active PALs by direct coupling

The covalent approach confers an irreversible stable immobilization. We selected a well- established covalent immobilization method by coupling the primary amines on the N-terminus or Lys-side chains of PALs to an NHS-ester (Figure 7C; Anderson et al. J Am Chem Soc 1964, 86 (9), 1839-1842; Cuatrecasas & Parikh Biochemistry 1972, 11 (12), 2291 -2299). Similar to the previously described biotinylation of ligases with NHS-LC-biotin, the direct immobilization on NHS-activated agarose beads were performed at pH 7.4 at 4 °C overnight to give agarose-Bu1 5 and agarose-Vy2 6. The beads were then washed with chilled pH 6.5 reaction buffer. The results showed that this method directly immobilized active butelase-1 and VyPAL2 with 83% and 81 % yield, respectively.

Example 11 : Activity of Immobilized PALs

The activity of immobilized PALs was determined by comparing the initial reaction rate catalyzed by immobilized PALs with the rate catalyzed by their soluble counterparts. The ligase activity of free butelase-1 or VyPAL2 was measured by the macrocyclization of a model peptide substrate KN14-GL (KLGTSPGRLRYAGN-GL; SEQ IDNO:51 ) 7, sequence derived from a natural cysteine- rich peptide bleogen pB1 ⁴⁴ with a C-terminal PAL-recognition signal tripeptide NGL to give the end-to-end cyclic product cKN14 8 (Figure 8). We used a substrate concentration of 0.2 mM to maximize the reaction rate because this concentration is much higher than the known Michaelis constant KM of butelase-1 and VyPAL2. Reaction was quenched after 5 min and the amount of cKN14 produced in each reaction was measured by RP-HPLC. The standard curves of reaction rate against the concentrations of free enzymes were plotted to calculate the turnover rate (Figure 9). The effective concentration of immobilized PALs were determined by interpolating the measured reaction rates of immobilized PALs on the standard curve.

Table 2 summaries the results which show that non-covalent attachment of ConA-Bu1 1 and the NA-linked biotinylated enzymes NA-Bu1 (b) 3 and NA-Vy2(b) 4 retained 50% and 20-30% activity of their soluble enzymes, respectively. The covalent attachment of agarose-Bu1 5 and agarose- Vy2 6 retained about 5% activity of their soluble enzymes. Direct attachment to agarose beads via a tetranoic spacer that was calculated to be about 1 nm only in agarose-Bu1 and agarose- Vy2 is perhaps too short (Figure 7). In contrast, ConA-Bu1 has a spacer longer than 8 nm (ConA tetramer + glycan) (Becker et al. J Bio Chem 1975, 250 (4), 1513-1524) and the NeutrAvidin- immobilized NA-linked biotinylated enzymes have a spacer approximately 8 nm long (NeutrAvidin tetramer + NHS-LC-biotin) (Livnah et al. Proc Natl Acad Sci U S A 1993, 90, 5076-5080). The correlation between the activity of immobilized PALs with the distance between the enzymes and solid supports suggested that short spacer may reduce enzyme mobility and accessibility of substrates. To improve the activity of immobilized enzymes via direct attachment method, a longer spacer needs to be exploited.

Table 2. Summary of immobilization yield and effective concentration of immobilized PALs

PAL Loading

Effectiv

Obs. e

Cone Yield Cone Ratio

_ (UM) (%) (uM) (%)

Non-covalent

ConA-Bu1 ϊ Ϊ05 39 Ϊ6 44

NA-Bu1 (b) 3 13.2 49 3.3 25

NA-Vy2(b) 4 12.1 _ 45 2.8 _ 23

Covalent _

Agarose-Bu1

5 22.4 83 1 .1 5

Agarose-Vy2

6 21 .8 81 0.6 3

Expected maximal PAL concentration on beads is 1 mg/mL = 27 mM.

Obs. Cone = observed protein loading of PAL on beads.

Yield = Observed concentration/expected maximal concentration.

Ratio = V(immobilized enzyme) / V(free enzyme) = Effective Conc./Obs. Cone.

ConA = concanavalin A.

NA = NeutrAvidin.

(b) = biotin. Example 12: Immobilized PALs display high operational stability and prolonged storage stability

Solid-phase immobilization of PALs would minimize self-aggregation and auto-proteolysis, and in turn, enhanced stability. To show the operational stability and reusability, each immobilized PAL was reused 100 times and their efficacy in the cyclization of linear peptide KN14-GL 7 analyzed. In each run, the same batch of immobilized-PAL agarose beads was used, and the reaction mixture was analyzed using C18 reversed-phase HPLC (Figure 10A). Figure 10B summarizes the product analysis of five immobilized PALs, all of which showed that >90% catalytic activity was retained after 100 runs.

To show the prolonged shelf-life of immobilized PALs stored at 4 °C, their ligase activity was monitored every week for a period of two months in cyclizing the peptide substrate GN14-HV (SEQ ID NO: 52; GISTKSIPPISYRN-HV, 9) or GN14-SLAN (SEQ ID NO:53; GISTKSIPPISYRN- SLAN, 10) to yield cGN14 11 by MALDI-TOF mass spectrometry. Figure 1 1 shows that immobilized PALs are more stable than their soluble counterparts in prolonged storage. All five immobilized PALs, retained >90% activity after nine weeks. In contrast, butelase-1 or VyPAL2 lost about 30% activity after a two-month storage under the same storage conditions.

It was found that addition of a reducing reagent, such as Tris(2-carboxyethyl)phosphine (TCEP), dithiothreitol (DTT) or b-mercaptoethanol (b-ME), is crucial for keeping the catalytic Cys of active PALs in the reduced form. Both soluble and immobilized PALs stored in a non-reducing buffer lost activity in two weeks due to the oxidation of the catalytic cysteinyl sulfhydryl. Once the sulfhydryl is oxidized leading to inactivation, the ligase activity can be restored, sometimes but not always, after treating with buffers containing one or more reducing reagents. It was also observed that immobilized butelase-1 is slightly more stable than immobilized VyPAL2, suggesting that plant-derived PALs may benefit from higher-level of glycosylation which enhances their molecular stability from proteolytic degradation.

Example 13: Applications of immobilized PALs for ligation reactions

The reusability of immobilized PALs allows us to use a much higher enzyme concentration than their soluble counterpart to accelerate the catalytic ligation reactions. In the following five examples NeutrAvidin-immobilized NA-Bu1 (b) 3 and NA-Vy2(b) 4 were used to showcase this advantage of immobilized PALs for cyclization and ligation, as well as their uses in a continuous flow system.

The first example was a cyclization reaction of a SFTI-substrate containing a sterically hindered Pro at the P2 position, which causes a slower ligation reaction than those substrates with a less hindered amino acid occupying the same P2 position. Figure 12A shows that butelase-1 -mediated cyclization of a 14-residue disulfide-containing peptide SFTI analog, GRCTKSIPPICFPN-HV 12 (SEQ ID NO:54), is 50% complete to yield the cyclic SFTI 13 after 30 min. In contrast, increasing the effective concentration of the NeutrAvidin-immobilized butelase-1 NA-Bu1 (b) 3 five-fold resulted in accelerating the cyclization ligation to complete within 10 min.

In the second example, the soluble and NeutrAvidin-immobilized butelase 1 was compared for cyclizing a 70-residue protein, circular bacteriocin AS-48. This circular bacteriocin is a highly sought-after food preservative produced by the lactic bacteria for its ability to kill broad-spectrum of microorganisms. AS-48 is the second largest naturally-occurring head-to-tail macrocycles known. The free butelase-1 was used to cyclize the folded AS-48K 14 (SEQ ID NO:1 9), which contains an N-terminal dipeptide and C-terminal hexapeptide sequence for butelase-1 recognition. Using an enzyme:substrate ratio of 1 :100 at 37 °C, the reaction was complete in 1 h whereas increasing effective concentration of NA-Bu1 (b) five-fold of the free butelase-1 accelerated the completion of cyclization in 1 0 min in a 83% isolated yield of cyclic AS-48 15 (Figure 12B).

The third example was PAL-mediated cyclooligomerization of peptides. This reaction involves both oligomerization and head-to-tail cyclization of the nascent oligomers. Using this approach the formation of bioactive cyclo-oligomeric peptides using a simple peptidyl monomer as building block was demonstrated. The cyclooligomerization of RV7 (RLYRNHV, 16; SEQ ID NO:55), using the NeutrAvidin-immobilized butelase NA-Bu1 (b) in an enzyme:substrate ratio of 1 :100, completed within 40 min to yield 83% cyclodimer c17 and 8% cyclotrimer c18 of RLYRN (Figure 13). In contrast, the reaction using butelase-1 with an enzyme:substrate ratio of 1 :500, in which effective concentration of the soluble form was five-fold lower the immobilized form, did not complete after 4 h (data not shown).

In the last two examples, the PAL-mediated intermolecular ligation was used in a continuous-flow system. Unlike cyclization reactions which have the advantage of high effective concentrations, intermolecular ligations would require both high concentrations of substrates and enzymes, and thus immobilized PALs can be reused in high concentration to overcome this limitation. Using a self-packed column (internal diameter 4 mm) with NA-Vy2(b) 4 beads, peptide ligation of Ac- RYANGI 19 (10 mM; SEQ ID NO:56) was performed with a synthetic fluorescent peptide GLAK(FAM)RG 20 (100 mM; SEQ ID NO:57) under different flow rate from 0.05 to 0.5 mL/min (Figure 14A). At flow rate of 0.05 mL/min, we observed completed ligation reaction to yield Ac- RYANGLAK(FAM)RG 21 (SEQ ID NO:58). Finally, we used this packed-bed column of NA-Vy2(b) to label a 193-residue recombinant protein, anti-Her2 DARPin9_26-NGL 22 (SEQ ID NO:49) with GLAK(FAM)RG 20. A reaction containing 1 mM DARPin9_26-NGL and 5 mM GLAK(FAM)RG afforded 78% yield of DARPin9_26-NGLAK(FAM)RG 23 with a flow rate of 20 pL/min (Figure 14B). The unreacted peptides were readily removed from ligation products by dialysis or centrifugal filters with molecular weight cut off >3 kDa.