METHODS OF PRODUCING SHIGA TOXIN B-SUBUNIT (STxB) MONOMERS AND OLIGOMERS, AND USES THEREOF

Abstract

A method of producing a monomer of a Shiga toxin B-subunit (STxB) protein or of a variant thereof by peptide chemical synthesis, as well as to a method of producing a pentamer of the STxB protein or of the variant thereof. The methods are particularly advantageous as they overcome major issues typically observed in peptide chemical synthesis, including solubility and purity issues.

Claims

1.-14. (canceled)

15. A method of producing a monomer of a Shiga toxin B-subunit (STxB) protein or of a variant thereof by peptide chemical synthesis, comprising the steps of: a) stepwisely coupling amino acid residues onto a support following an amino acid sequence with at least 60% identity to the amino acid sequence set forth in SEQ ID NO: 2 or a fragment thereof, wherein said amino acid residues comprise α amino-protecting groups and optionally side chain-protecting groups, thereby obtaining a synthetic peptide; b) deprotecting the synthetic peptide obtained from step a) by removing: b1) the final α amino-protecting group, and b2) optionally, the side-chain protecting-groups, thereby obtaining a deprotected synthetic peptide; and c) cleaving the deprotected synthetic peptide obtained from step b) from the support, thereby obtaining a free monomer of the STxB protein or of the variant thereof.

16. The method according to claim 15, wherein step a) comprises the following sequence of substeps: a1) removing the α amino-protecting group from the support or from the N-terminal amino acid of the synthetic peptide; a2) optionally, washing the support; a3) coupling the next amino acid following an amino acid sequence with at least 60% identity to the amino acid sequence set forth in SEQ ID NO: 2 in a linear C- to N-terminal direction; a4) optionally, washing the support; a5) optionally, capping unreacted amino groups; and a6) optionally, washing the support.

17. The method according to claim 16, wherein: Substep a3) is reiterated more times and/or is carried out for a longer period of time to couple amino acid residues Cys 4, Thr 12, Thr 21, Asn 35, Leu 36, and Ile 45 with respect to SEQ ID NO: 2 numbering than to couple the other amino acid residues of the STxB protein or of the variant thereof; and dipeptides Val 5-Thr 6, Asp 18-Thr 19, Phe 30-Thr 31, Leu 41-Ser 42, and Val 50-Thr 51 with respect to SEQ ID NO: 2 numbering are coupled in substep a3) in a pseudoproline dipeptide form.

18. The method according to claim 16, wherein substep a3) is reiterated more times and/or carried out for a longer period of time to couple amino acid residues Cys 4, Thr 12, Thr 21, Asn 35, Leu 36, Leu 39, Ile 45, Cys 57, and Val 65 with respect to SEQ ID NO: 2 numbering than to couple the other amino acid residues of the STxB protein or of the variant thereof.

19. The method according to claim 16, wherein substep a3) is reiterated more times and/or carried out for a longer period of time to couple amino acid residues Cys 4, Thr 12, Thr 21, Asn 35, Leu 36, Leu 39, Ile 45, Thr 49, Cys 57, and Val 65 with respect to SEQ ID NO: 2 numbering than to couple the other amino acid residues of the STxB protein or of the variant thereof.

20. The method according to claim 16, wherein substep a3) is reiterated more times and/or carried out for a longer period of time to couple amino acid residues Thr 1, Cys 4, Tyr 11, Thr 12, Thr 21, Asn 35, Leu 36, Leu 39, Ile 45, Thr 46, Thr 49, Cys 57, and Val 65 with respect to SEQ ID NO: 2 numbering than to couple the other amino acid residues of the STxB protein or of the variant thereof.

21. The method according to claim 16, wherein dipeptides Val 5-Thr 6, Asp 18-Thr 19, Phe 30-Thr 31, Leu 41-Ser 42, Val 50-Thr 51, and Phe 63-Ser 64 with respect to SEQ ID NO: 2 numbering are coupled in substep a3) in a pseudoproline dipeptide form.

22. The method according to claim 15, wherein the method further comprises one or more of the steps of: d) precipitating the free monomer of the STxB protein or of the variant thereof obtained from step c), and optionally air-drying and/or lyophilizing the precipitated STxB protein or the variant thereof; e) oxidizing the free monomer of the STxB protein or of the variant thereof obtained from step c) or d) under conditions suitable for the formation of an intramolecular disulfide bond between Cys 4 and Cys 57 with respect to SEQ ID NO: 2 numbering.

23. The method according to claim 15, wherein a variant of the Shiga toxin B-subunit (STxB) protein comprises one of the peptides with amino acid sequence SEQ ID NO: 1 and 3 to 21.

24. The method according to claim 15, wherein the monomer of the STxB protein or of the variant thereof is produced by peptide chemical synthesis of at least two fragments of said STxB protein or variant thereof, and wherein said at least two fragments are ligated to obtain the monomer of the STxB protein or of the variant thereof.

25. The method according to claim 24, wherein said fragments of the STxB protein or of the variant thereof consist of the peptides with SEQ ID NO: 22 and SEQ ID NO: 23.

26. The method according to claim 15, wherein the method does not comprise a purification step by chromatography.

27. A monomer of a Shiga toxin B-subunit (STxB) protein or of a variant thereof obtained by the method according to claim 15.

28. A method of producing a pentamer of a Shiga toxin B-subunit (STxB) protein or of a variant thereof by peptide chemical synthesis, comprising the steps of: a) carrying out the method of producing a monomer of a STxB protein or of a variant thereof by peptide chemical synthesis according to claim 15, thereby obtaining a free monomer of the STxB protein or of the variant thereof; and b) refolding the free monomer of the STxB protein or of the variant thereof by dialyzing said protein or the variant thereof against at least one dialysate, thereby obtaining a pentamer of the STxB protein or of the variant thereof.

29. The method according to claim 28, wherein the method does not comprise a purification step by chromatography.

30. A pentamer of a Shiga toxin B-subunit (STxB) protein or of a variant thereof obtained by the method according to claim 28.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0799] FIG. 1 is a set of four chromatograms illustrating the UPLC-MS analysis of crude STxB peptides from two independent synthesis batches (FIG. 1A and FIG. 1B, respectively, with absorption wavelength=214 nm, and x-axis=time, in minutes) after cleavage. FIG. 1C and FIG. 1D show the mass spectrum of the portion underlined in black in FIG. 1A and FIG. 1B, respectively, with x-axis=m/z.

[0800] FIG. 2 is a chromatogram illustrating the RP-HPLC purification of crude STxB peptides from the second batch of synthesis (corresponding to FIG. 1B) on a Water XBridge BEH 300 Prep C18 5 μm 30×150 mm column. 11.4 mg of crude peptide was injected in 2 mL guanidinium chloride comprising little amount of TCEP+1 mL 1% TFA. The program consisted of 8 minutes 95% A (99.9% H.sub.2O, 0.1% TFA)+5% B (90% acetonitrile, 9.99% H.sub.2O, 0.1% TFA) to eliminate salts (not shown) followed by a 20-minute linear gradient from 5% to 100% B. “Clean” (solid black line) and “unclean” (hatched black lines) fractions were collected from several injections. Absorption wavelength: 214 nm. x-axis: time, in minutes.

[0801] FIG. 3 is a set of four chromatograms illustrating the UPLC-MS analysis of the lyophilized compound obtained from the “clean” fraction (FIG. 3A) and “unclean” fraction (FIG. 3B) with absorption wavelength=214 nm, and x-axis=time, in minutes; and mass spectra of the portion underlined in black in FIG. 3A (FIG. 3C) and in FIG. 3B (FIG. 3D), with x-axis=m/z.

[0802] FIG. 4 is a set of six chromatograms illustrating UPLC analysis upon oxidation of STxB. FIGS. 4A and 4B are control chromatograms, showing characteristic peaks of STxB in reduced form in the presence of TCEP (FIG. 4A) and in oxidized form in absence of TCEP (FIG. 4B). FIGS. 4C to 4F show oxidation analyses of STxB at different time points (FIG. 4C: t=0 h; FIG. 4D: t=4 h; FIG. 4E: t=8 h; FIG. 4F: t=24 h). Absorption wavelength: 214 nm. x-axis: time, in minutes.

[0803] FIG. 5 is a set of six chromatograms, illustrating UPLC-MS analysis of folded STxB samples. FIG. 5A shows folded STxB protein obtained from crude peptide (i.e., without HPLC purification step); FIG. 5B shows folded STxB protein obtained from the “clean” fraction; and FIG. 5C shows folded STxB protein obtained from the “unclean” fraction; with absorption wavelength=214 nm, and x-axis=time, in minutes. FIG. 5D, FIG. 5E and FIG. 5F show the mass spectrum of the portion underlined in black in FIG. 5A, FIG. 5B and FIG. 5C, respectively, with x-axis=m/z.

[0804] FIG. 6 (filed in color) is a set of eight photographs, showing colocalization of recombinant STxB (rSTxB) and synthetic STxB (STxB protein obtained from crude peptide, without RP-HPLC purification [sSTxB-HPLC]; STxB protein obtained from the “clean” fraction after RP-HPLC purification [sSTxB+HPLC clean]; and STxB protein obtained from the “unclean” fraction after RP-HPLC purification [sSTxB+HPLC unclean]), with the Golgi after 50 minutes of incubation at 37° C. STxB was labelled with mouse monoclonal anti-STxB antibodies (clone 13C4) (left panels). Golgi was labelled with rabbit polyclonal anti-giantin antibodies (center panels). Merge (right panels): STxB channel in green, Golgi channel in red.

[0805] FIG. 7 is a chromatogram illustrating the HPLC-MS analysis of the STxB-Cys-amide synthetized without pseudoprolines and with double coupling, with absorption wavelength=214 nm, and x-axis=time, in minutes. The eight detection peaks are numbered in grey circles.

[0806] FIG. 8 is a set of eight chromatograms illustrating the ESI+ spectra from each of the chromatogram detection peaks numbered 1 to 8 in FIG. 7, with x-axis=m/z.

[0807] FIG. 8A: detection peak 1; FIG. 8B: detection peak 2; FIG. 8C: detection peak 3;

[0808] FIG. 8D: detection peak 4; FIG. 8E: detection peak 5; FIG. 8F: detection peak 6;

[0809] FIG. 8G: detection peak 7; FIG. 8H: detection peak 8.

[0810] FIG. 9 is a set of two chromatograms illustrating the UPLC-MS analysis of crude STxB-Cys-amide peptide (FIG. 9A) and folded STxB-Cys-amide peptide (FIG. 9B) after synthesis with 6 pseudoprolines and double or quadruple coupling. Absorption wavelength: 214 nm. x-axis: time, in minutes.

[0811] FIG. 10 (filed in color) is a set of 6 photographs, showing colocalization of recombinant [rSTxB-Cys] or synthetic [sSTxB-Cys-amide] STxB-Cys protein with the Golgi. Merge: STxB channel in green, Golgi channel in red.

EXAMPLES

[0812] The present invention is further illustrated by the following examples.

Example 1

[0813] Solid-Phase Synthesis of Shiga Toxin B-Subunit (STxB) Protein

[0814] Material and Methods

[0815] Reagents

[0816] Solid-phase synthesis of full length monomeric STxB with SEQ ID NO: 2 was performed on a Prelude Instrument (Gyros protein Technologies), at 12.5 μmol scale, using a Fmoc-Arg(Pbf)-Wang low loading resin (Novabiochem). This resin is pre-loaded with a arginine residue comprising an α amino-protecting group (fluorenylmethyloxycarbonyl; Fmoc), and a side-chain protecting-group (2,2,4,6,7-pentamethyldihydrobenzofuran-5-sulfonyl; Pbf).

TABLE-US-00013 SEQ ID NO: 2 TPDCVTGKVEYTKYNDDDTFTVKVGDKELFTNRWNLQSLLLSAQITGMTV TIKTNACHNGGGFSEVIFR

[0817] Amino acids and pseudoprolines were purchased from Novabiochem.

[0818] N-methylmorpholine (NMM), acetic anhydride (Ac.sub.2O), acetic acid, thioanisole, anisole, triisopropylsilane (TIS), sodium phosphate monobasic, sodium phosphate dibasic and dimethyl sulfoxide (DMSO) were obtained from Sigma Aldrich.

[0819] 2-(6-chloro-1H-benzotriazol-1-yl)-N,N,N′,N′-tetramethylaminium hexafluorophosphate (HCTU) was obtained from VWR.

[0820] Dichloromethane (DCM), piperidine and diethyl ether were purchased from Carlo Erba.

[0821] Dimethylformamide (DMF) was obtained from Merck Millipore.

[0822] N-methyl-2-pyrrolidone (NMP) was obtained from BDH Chemicals.

[0823] Trifluoroacetic acid (TFA) was purchased from Fisher Scientific.

[0824] Guanidine hydrochloride (GuHCl) was purchased from Calbiochem.

[0825] Synthesis

[0826] The resin was swelled twice in 3 mL DCM for 30 seconds with mixing, then once in 3 mL NMP for 5 minutes with mixing.

[0827] Standard Synthesis Cycle

[0828] The synthesis workflow was set as follows on the Prelude Instrument, with one cycle being defined as substeps (1) to (6) defined below, each cycle leading to the addition of one amino acid to the growing peptide, in a linear C- to N-terminal direction following SEQ ID NO: 2. [0829] (1) Deprotection [0830] This substep was carried out twice in a row per cycle, with 2 mL of 20% piperidine in NMP for 3 minutes each time, with mixing. [0831] (2) Washes [0832] This substep was carried out three times in a row per cycle, with 3 mL of NMP for 30 seconds each time, with mixing. [0833] (3) Coupling [0834] This substep was carried out twice in a row per cycle, with 1300 μL of Fmoc-protected amino acid (200 mM in NMP=20.8 eq.; except for cysteine residues: 200 mM in DMF=20.8 eq), 1000 μL of HCTU (250 mM in NMP=20 eq.) and 500 μL of NMM (1 M in NMP=40 eq.) for 10 minutes each time, with mixing. [0835] (4) Washes [0836] This substep was carried out twice in a row per cycle, with 3 mL of NMP for 30 seconds each time, with mixing. [0837] (5) Capping [0838] This substep was carried out once per cycle, with 2000 μL of Ac.sub.2O (250 mM in NMP) and 500 μL of NMM (1 M in NMP=40 eq.) for 5 minutes, with mixing. [0839] (6) Washes [0840] This substep was carried out three times in a row per cycle, with 3 mL of NMP for 30 seconds each time, with mixing.

[0841] This synthesis workflow therefore includes a systematic double coupling (substep (3)) per amino acid residue.

[0842] Adjusted Synthesis Cycle

[0843] In the course of the synthesis protocol setup, difficult positions were identified (i.e., amino acid positions in SEQ ID NO: 2 where one cycle with double coupling does not yield satisfactory levels of amino acid addition).

[0844] For the following positions, a quadruple coupling (substep (3)) was implemented: Cys 4, Thr 12, Thr 21, Asn 35, Leu 36, Leu 39, Ile 45, Thr 49, Cys 57, and Val 65 (with respect to SEQ ID NO: 2 numbering). The synthesis workflow was therefore adapted as follows for these difficult positions: [0845] (1) Deprotection

[0846] This substep was carried out twice in a row per cycle, with 2 mL of 20% piperidine in NMP for 3 minutes each time, with mixing. [0847] (2) Washes

[0848] This substep was carried out three times in a row per cycle, with 3 mL of NMP for 30 seconds each time, with mixing. [0849] (3) Coupling [0850] This substep was carried out four times in a row per cycle, with 1300 μL of Fmoc-protected amino acid (200 mM in NMP=20.8 eq.; except for cysteine residues: 200 mM in DMF=20.8 eq), 1000 μL of HCTU (250 mM in NMP=20 eq.) and 500 μL of NMM (1 M in NMP=40 eq.) for 10 minutes each time, with mixing. [0851] (4) Washes [0852] This substep was carried out twice in a row per cycle, with 3 mL of NMP for 30 seconds each time, with mixing. [0853] (5) Capping [0854] This substep was carried out once per cycle, with 2000 μL of Ac.sub.2O (250 mM in NMP) and 500 μL of NMM (1 M in NMP=40 eq.) for 5 minutes, with mixing. [0855] (6) Washes [0856] This substep was carried out three times in a row per cycle, with 3 mL of NMP for 30 seconds each time, with mixing.

[0857] Final Deprotection

[0858] Once the whole STxB peptide with SEQ ID NO: 2 was synthetized, the final α amino-protecting group (i.e., the Fmoc protecting group borne by the threonine residue in position 1 of SEQ ID NO: 2), were removed, according to the following substeps: [0859] (1) Deprotection [0860] This substep was carried out twice in a row, with 2 mL of 20% piperidine in NMP for 3 minutes each time, with mixing. [0861] (2) NMP Wash [0862] This substep was carried out once, with 3 mL of NMP for 30 seconds, with mixing. [0863] (3) DCM Washes [0864] This substep was carried out four times in a row, with 3 mL of DCM for 30 seconds each time, with mixing.

[0865] Cleavage

[0866] The resin was cleaved in 5 mL TFA:thioanisole:anisole:TIS:H.sub.2O (82.5:5:5:2.5:5) for 2 hours under stirring. Under these conditions, side-chain protecting groups optionally borne by the amino acid residues (in particular non-aliphatic amino acid residues) were also removed.

[0867] The cleavage solution was then precipitated in 40 mL cold diethyl ether.

[0868] After 3 washes with 45 mL cold diethyl ether, the precipitate was air dried. The precipitate was then mixed in 15 mL 10% acetic acid in water/acetonitrile mix and lyophilized.

[0869] Purification

[0870] After dissolution in 6 M of guanidinium chloride (GndHCl) containing a small amount of tris(2-carboxyethyl)phosphine (TCEP) and acidification with 1% TFA solution, crude STxB peptides were purified with a Waters HPLC (comprising a Waters 2545 Quaternary Gradient Module, a Waters 2998 Photodiode Array Detector and a Waters FlexInject).

[0871] The column used was a Water XBridge BEH 300 Prep C18 5 μm 30×150 mm column, with: [0872] Solvent A: 99.9% H.sub.2O, 0.1% TFA; [0873] Solvent B: 90% acetonitrile, 9.99% H.sub.2O, 0.1% TFA; [0874] Program: [0875] 95% A+5% B for 8 minutes (to eliminate guanidinium salts), and [0876] Linear gradient from 5% to 100% B in 20 minutes.

[0877] Analyses

[0878] All samples were analyzed with a Waters UPLC-MS (comprising an ACQUITY UPLC H-Class sample manager, an ACQUITY UPLC PDA eLambda Detector, and a Single Quadrupole Detector 2 for electron spray ionization/mass spectra [ESI-MS]]).

[0879] The column used was an ACQUITY UPLC BEH C18 column, 130 Å, 1.7 μm, 2.1 mm×50 mm, with: [0880] Solvent A: 0.1% formic acid in Milli-Q water; [0881] Solvent B: 0.1% formic acid in acetonitrile; [0882] Program: [0883] 5% B for 0.2 minutes, [0884] Linear gradient from 5% to 95% B in 2.3 minutes, [0885] 100% B for 0.5 minutes, and [0886] 5% B for 1 minute.

[0887] Samples were dissolved in 6 M GndHCl, acidified with 1% formic acid aqueous solution and filtered before injection. All analyses are shown at 214 nm (characteristic absorption wavelength for amide bond of peptides).

[0888] Oxidation Analyses

[0889] Samples were acidified with 1% formic acid aqueous solution and analyzed with a Waters UPLC-MS (comprising an ACQUITY UPLC H-Class sample manager and an ACQUITY UPLC PDA eLambda Detector).

[0890] The column used was an ACQUITY UPLC BEH C18 column, 130A, 1.7 pm, 2.1 mm×50 mm, with: [0891] Solvent A: 0.1% formic acid in Milli-Q water; [0892] Solvent B: 0.1% formic acid in acetonitrile; [0893] Program: [0894] 5% B for 0.2 minutes, [0895] Isocratic flow at 33% B for 2.3 minutes, [0896] 100% B for 0.5 minutes, and [0897] 5% B for 1 minute.

[0898] Oxidation and Folding

[0899] Purified lyophilized STxB peptide was dissolved to 0.5 mg/mL in oxidation buffer (7 M GndHCl, 50 mM sodium phosphate, 2% DMSO, pH adjusted to 8).

[0900] The solution was then incubated under stirring at 37° C. for 24 hours, to form a disulfide bond between Cys 4 and Cys 57 of SEQ ID NO: 2.

[0901] The solution was then dialyzed at 4° C. with Slide-A-Lyzer™ G2 Dialysis Cassettes, 3.5 kD MWCO from Thermo Scientific against the following: [0902] 3 M GndHCl, 50 mM sodium phosphate pH 8.0, 5 mM EDTA, for 6 to 10 hours; [0903] 1 M GndHCl, 50 mM sodium phosphate pH 8.0, 1 mM EDTA, overnight; [0904] PBS for 4 hours; [0905] PBS for 4 hours; and [0906] PBS overnight.

[0907] After removal from the dialysis cassette, the solution was centrifuged to remove the precipitate. Supernatant was kept and concentrated using centrifugal filters (Amicon Ultra Centrifugal filters, 10 kD MWCO).

[0908] Concentration was measured with Nanodrop 2000, using ε=8250M.sup.−1.Math.cm.sup.−1.

[0909] Small aliquots were flash-freezed and stored at −20° C.

[0910] Intracellular Trafficking Assay by Immunofluorescence

[0911] Intracellular trafficking assays were performed on HeLa cells, cultured at 37° C. under 5% CO.sub.2 in Dulbecco's modified Eagle's medium (DMEM, Invitrogen), supplemented with 10% heat-inactivated fetal bovine serum (FBS), 0.01% penicillin-streptomycin, 4 mM glutamine and 5 mM pyruvate.

[0912] Cells were plated the day before on lamellae in 4-well plates, 60 000 cells/well.

[0913] Binding

[0914] Cells were incubated for 30 minutes at 4° C. in presence of 500 μL STxB dilution in cold complete medium (0.2 μM or 1 μM), then washed 3 times with 500 μL PBS with Ca.sup.2+ and Mg.sup.2+ (PBS.sup.++).

[0915] Internalization

[0916] 500 μL of complete medium preheated at 37° C. was added on cells. Cells were incubated for 50 minutes at 37° C., then washed 3 times with PBS.sup.++.

[0917] Fixation

[0918] Cells were treated with 500 μL of 4% paraformaldehyde (PFA) during 20 minutes, then washed once with 50 mM of NH.sub.4Cl, and incubated with 50 mM of NH.sub.4Cl for at least 30 minutes.

[0919] Permeabilization

[0920] Cells were washed 3 times with 500 μL of PBS/BSA/Saponin 1× (1×PBS/1.0% BSA/0.1% Saponin), and then incubated at room temperature for 30 minutes in presence of 500 μL of PBS/BSA/Saponin 1×.

[0921] Incubation with Antibodies

[0922] Lamellae were incubated with 30 μL of primary antibody dilution into PBS/BSA/Saponin 1× for 30 minutes at room temperature, then washed 3 times with PBS/BSA/Saponin 1×.

[0923] Primary antibodies used were the mouse monoclonal clone 13C4 anti-STxB antibody (Strockbine et al., 1985. Infect Immun. 50(3):695-700), at 1/250 dilution; and a home-made rabbit polyclonal antibody against the Golgi marker Giantin, used at 1/100 dilution.

[0924] Same was done with the secondary antibodies (anti-mouse Cy3 and anti-rabbit A488 used at 1/100 dilution each).

[0925] Slide Preparation

[0926] Lamellae were washed in water and then added on slides on 6 μL of Fluoromount-G™+Hoechst. Polymerization was allowed for 30 minutes at 37° C.

[0927] Microscope Observation

[0928] Slides were observed with the following equipment from the Biomaging Cell and Tissue Core Facility of the Institut Curie in Paris, France (PICT-IBiSA): upright Leica DM6000 microscope (with a CCD 1392×1040 CoolSnap HQ2 camera from Photometrics, pixel: 6.45 pm; and a Lumen 200 lamp illumination source from Prior Scientific). A Leica HCX PL Apo 63× oil objective was used for pictures.

[0929] Results

[0930] Synthesis

[0931] The full length STxB monomeric sequence (SEQ ID NO: 2) was synthetized on a 12.5 μmol scale, starting from a preloaded low loading Fmoc-Arg(Pbf) resin.

[0932] Coupling was carried out twice for 10 minutes with 20.8 equivalents of amino acids, 20 equivalents of HCTU coupling agent and 40 equivalents of N-methylmorpholine (NMM) in N-methyl-2-pyrrolidone (NMP). Coupling was then followed by a 5-minute capping substep with acetic anhydride and NMM in NMP.

[0933] Six different pseudoprolines were also used along the sequence to prevent R-chain formation and aggregation (Val 5-Thr 6, Asp 18-Thr 19, Phe 30-Thr 31, Leu 41-Ser 42, Val 50-Thr 51, and Phe 63-Ser 64 with respect to SEQ ID NO: 2 numbering).

[0934] For some positions which have been identified as “difficult positions”, coupling was repeated four times instead of two (Cys 4, Thr 12, Thr 21, Asn 35, Leu 36, Leu 39, Ile 45, Thr 49, Cys 57, and Val 65 with respect to SEQ ID NO: 2 numbering).

[0935] Yield

[0936] UPLC-MS analysis of the resulting product shows a dominant peak (FIGS. 1A & 1B, at 1.71 minutes) corresponding to the expected mass of the monomeric STxB peptide (FIGS. 1C & 1D); and small contaminant peaks (FIG. 1A, at 1.57 and 1.79 minutes) which are more or less present depending on the synthesis batch (see the absence of contaminant picks on FIG. 1B).

[0937] 65.7 mg of peptide were recovered from the second synthesis batch after cleavage (corresponding to FIG. 1B), representing approximately 70% yield in mass.

[0938] Purification

[0939] After cleavage, the monomeric STxB peptide could either be purified through reverse-phase high pressure liquid chromatography (RP-HPLC); or alternatively, could directly be oxidized and refolded.

[0940] RP-HPLC purification on a C18 column led to high peptide loss: out of 43.8 mg of crude peptide (from the second synthesis batch, dissolved in 2 mL of 6 M GndHCl containing little amount of TCEP and 1 mL 1% TFA) injected onto the column, only 0.8 mg of “clean” and 3.8 mg of “unclean” fractions were recovered (FIG. 2), corresponding to a 10.5% yield when considering the sum of the two fractions recovered.

[0941] Much of the peptide remained stuck on the column and had to be removed through extensive washes via DMSO followed by a short gradient from 5 to 100% of B solvent. 16 mg were recovered with these successive washes.

[0942] UPLC-MS analysis of the lyophilized STxB peptide from the “clean” fraction show a single, sharp peak at 1.71 minutes (FIG. 3A), corresponding to the expected mass of the monomeric STxB peptide (FIG. 3C). On the other hand, UPLC-MS analysis of the lyophilized STxB peptide from the “unclean” fraction show a saw-teeth peak at 1.71 minutes extending to 1.9 minutes (FIG. 3B), also corresponding to the expected mass of the monomeric STxB peptide (FIG. 3D).

[0943] Oxidation Analyses

[0944] The crude STxB peptide, either directly after synthesis and cleavage (i.e., without purification step), or the “clean” and “unclean” fractions obtained after RP-HPLC purification, could then be oxidized to form the intramolecular disulfide bond between Cys 4 and Cys 57 of SEQ ID NO: 2, and folded to yield a functional protein.

[0945] As a control, oxidized STxB peptide was analyzed by UPLC, after incubation at room temperature for 1 hour before analysis in 3 M GndHCl in presence of 100 mM TCEP (FIG. 4A) or in absence of TCEP (FIG. 4B).

[0946] In presence of TCEP, the reduced form of the STxB peptide is expected (i.e., without the intramolecular disulfide bond), corresponding to the peak seen at 1.55 minutes. In absence of TCEP, the oxidized form of the STxB peptide is expected (i.e., with the intramolecular disulfide bond), corresponding to the peak seen at 1.42 minutes.

[0947] Oxidation could be followed by UPLC analyses using an isocratic flow, at different time points (FIG. 4C: t=0 h; FIG. 4D: t=4 h; FIG. 4E: t=8 h; FIG. 4F: t=24 h).

[0948] It can be clearly seen from FIGS. 4C to 4F that the amount of oxidized form increases over time and is dominant at t=24 h (peak at 1.41 minutes on FIG. 4F), when oxidation is stopped to start the folding step.

[0949] Folding

[0950] Folding was achieved via stepwise dialyses, in buffers successively comprising 6 M, 3 M and 1 M GndHCl, and ending up with buffers comprising PBS but no GndHCl.

[0951] After centrifugation, the supernatant was recovered, concentrated and analyzed by UPLC-MS.

[0952] Folded samples obtained after oxidation from the crude STxB peptide (i.e., directly after synthesis and cleavage, without purification step), or from the “clean” and “unclean” fractions obtained after RP-HPLC purification, are very pure and comparable in quality (FIGS. 5A to 5C), and exhibit the expected mass of the monomeric STxB peptide (FIGS. 5D to 5F).

[0953] Their concentration was assessed by measuring absorbance at 280 nm and using monomeric STxB extinction coefficient ε=8250 M.sup.−1.Math.cm.sup.−1.

[0954] About 2.34 mg of folded STxB was obtained from 6 mg of crude peptide without purification, corresponding to a 39%-yield.

[0955] About 0.24 mg of folded STxB was obtained from 0.8 mg of the “clean” fraction obtained after RP-HPLC purification, corresponding to a 30%-yield.

[0956] About 0.72 mg of folded STxB was obtained from 3.8 mg of the “unclean” fraction obtained after RP-HPLC purification, corresponding to a 19%-yield.

[0957] Oligomerization and Intracellular Trafficking Assay by Immunofluorescence

[0958] Folded samples were than tested for oligomerization and functionality using an immunofluorescence assay. This assay relies on the fact that upon normal trafficking, STxB in its pentameric form should be in the Golgi after 50 minutes of incubation at 37° C.

[0959] Several STxB samples were tested and compared: [0960] “rSTxB”, corresponding to STxB obtained by recombinant expression; [0961] “sSTxB-HPLC”, corresponding to STxB obtained by chemical synthesis as described herein, without a RP-HPLC purification step; [0962] “sSTxB+HPLC clean”, corresponding to the “clean” fraction of STxB obtained by chemical synthesis as described herein, after a RP-HPLC purification step; and [0963] “sSTxB+HPLC unclean”, corresponding to the “unclean” fraction of sSTxB obtained by chemical synthesis as described herein, after a RP-HPLC purification step.

[0964] Cells were incubated with 0.1 μM of the STxB samples for 30 minutes at 4° C. for binding to their receptor Gb3 (no internalization can occur at 4° C.).

[0965] After several washings to remove unbound STxB, cells were incubated for 50 minutes at 37° C. for synchronized internalization, and then fixed and labelled with antibodies.

[0966] FIG. 6 shows colocalization of recombinant and synthetic STxB from the different samples with the Golgi (labelled with anti-giantin antibodies) after 50 minutes of incubation at 37° C., meaning that these samples contain functional STxB pentamers.

[0967] Conclusion

[0968] In conclusion, the data presented hereinabove show that a functional STxB pentamer could be obtained via solid-phase synthesis.

[0969] Upon setup of the synthesis method, some improvements were brought to the method, in order to ensure high synthesis yields and avoid incomplete syntheses of intermediate products. In that sense, some “difficult positions” have been identified in SEQ ID NO: 2, for which coupling of the amino acid was carried out 4 times in a row instead of 2. Pseudoprolines have also been used along the sequence, to prevent R-chain formation and aggregation.

[0970] Moreover, the Inventors have shown that no purification step was required to eliminate contaminants from the sample, saving thereby one step in the process and important material losses. Indeed, a functional STxB peptide could be efficiently obtained directly by oxidizing and folding the crude STxB peptide, without purification step, with a final yield close to 40%.

Example 2

[0971] Bacterial Endotoxins Levels

[0972] The level of bacterial endotoxins was assessed in various batches of STxB protein, recombinantly or synthetically produced, in “lab conditions” or “GLP/GMP-like conditions”. These measurements were carried out by Inovalys (Angers, France). When synthetically produced, the STxB protein was oxidized and folded directly from lyophilized peptide obtained after synthesis and cleavage, without purification step.

[0973] Bacterial endotoxins in these samples were measured using the chromogenic kinetic method described in the European Pharmacopoeia 9.8, section 2.6.14 (method D).

[0974] This technique is used to measure the chromophore released from a suitable chromogenic peptide by the reaction of endotoxins with the lysate. It measures either the time (onset time) needed for the reaction mixture to reach a predetermined absorbance, or the rate of color development. The test is carried out at the incubation temperature recommended by the lysate manufacturer (usually 37±1° C.).

[0975] Table 1 shows the results of these measurements, in endotoxin unit per mg of STxB protein.

TABLE-US-00014 TABLE 1 Recombinant Recombinant Synthetic Synthetic “Lab “GLP- “Lab “GMP- Samples conditions” like” conditions” like” Cc (EU/mg) 480 20 23 4

[0976] Cc: concentration; EU: endotoxin unit

Example 3

[0977] Role of Pseudoproline and Multiple Coupling in the Efficiency of the STxB Synthesis

[0978] 1) STxB-Cys-Amide Synthesis without Pseudoproline, with Double Coupling and

[0979] Capping

[0980] Material and Methods

[0981] Reagents

[0982] Solid-phase synthesis of a full length monomeric mutant of the STxB protein with SEQ ID NO: 2 comprising a cysteine residue at the C-terminus (SEQ ID NO: 21) was performed on a Prelude Instrument (Gyros protein Technologies), at 12.5 μmol scale, using a ChemMatrix rink amide matrix resin.

TABLE-US-00015 SEQ ID NO: 21 TPDCVTGKVEYTKYNDDDTFTVKVGDKELFTNRWNLQSLLLSAQITGMTV TIKTNACHNGGGFSEVIFRC

[0983] Synthesis

[0984] The resin was swelled twice in 3 mL DCM for 30 seconds with mixing, then once in 3 mL NMP for 5 minutes with mixing.

[0985] Synthesis Cycle

[0986] The synthesis workflow was set as follows on the Prelude Instrument, with one cycle being defined as substeps (1) to (6) defined below, each cycle leading to the addition of one amino acid to the growing peptide, in a linear C- to N-terminal direction following SEQ ID NO: 21. [0987] (1) Deprotection [0988] This substep was carried out twice in a row per cycle, with 2 mL of 20% piperidine in NMP for 3 minutes each time, with mixing. [0989] (2) Washes [0990] This substep was carried out three times in a row per cycle, with 3 mL of NMP for 30 seconds each time, with mixing. [0991] (3) Coupling [0992] This substep was carried out twice in a row per cycle, with 1300 μL of Fmoc-protected amino acid (200 mM in NMP=20.8 eq.; except for cysteine residues: 200 mM in DMF=20.8 eq), 1000 μL of HCTU (250 mM in NMP=20 eq.) and 500 μL of NMM (1 M in NMP=40 eq.) for 10 minutes each time, with mixing. [0993] (4) Washes [0994] This substep was carried out twice in a row per cycle, with 3 mL of NMP for 30 seconds each time, with mixing. [0995] (5) Capping [0996] This substep was carried out once per cycle, with 2000 μL of Ac.sub.2O (250 mM in NMP) and 500 μL of NMM (1 M in NMP=40 eq.) for 5 minutes, with mixing. [0997] (6) Washes [0998] This substep was carried out three times in a row per cycle, with 3 mL of NMP for 30 seconds each time, with mixing.

[0999] This synthesis workflow therefore includes a systematic double coupling (substep (3)) per amino acid residue.

[1000] Final Deprotection

[1001] Once the whole STxB peptide with SEQ ID NO: 21 was synthetized, the final α amino-protecting group (i.e., the Fmoc protecting group borne by the threonine residue in position 1 of SEQ ID NO: 21), were removed, according to the following substeps: [1002] (1) Deprotection [1003] This substep was carried out twice in a row, with 2 mL of 20% piperidine in NMP for 3 minutes each time, with mixing. [1004] (2) NMP Wash [1005] This substep was carried out once, with 3 mL of NMP for 30 seconds, with mixing. [1006] (3) DCM Washes [1007] This substep was carried out four times in a row, with 3 mL of DCM for 30 seconds each time, with mixing.

[1008] Cleavage

[1009] The resin was cleaved in 5 mL TFA:thioanisole:anisole:TIS:H.sub.2O (82.5:5:5:2.5:5) for 2 hours under stirring. Under these conditions, side-chain protecting groups optionally borne by the amino acid residues (in particular non-aliphatic amino acid residues) were also removed.

[1010] The cleavage solution was then precipitated in 40 mL cold diethyl ether.

[1011] After 3 washes with 45 mL cold diethyl ether, the precipitate was air dried. The precipitate was then mixed in 15 mL 10% acetic acid in water/acetonitrile mix and lyophilized.

[1012] Analyses

[1013] The crude peptide was analysed using Agilent 1100 series HPLC (Santa Clara, USA), with a Vydac C.sub.4 (214TP) column (300 Å, 5 μm, 4.6 mm i.d.×250 mm; Cat. No. 214TP54 S/N E980716-1-4), coupled online to an Esquire ion trap mass spectrometer equipped with an AP-ESI source (Bruker, Germany). The gradient used was from 5 to 100% acetonitrile with 0.1% TFA for 60 minutes at 1 mL/minute.

[1014] Results

[1015] Table 2 and FIG. 7 show the truncated products identified from a STxB-Cys-amide synthesis without pseudoprolines and with double coupling.

TABLE-US-00016 TABLE 2 Detection Truncated peak m/z Calculated Sequence identification product (Figure) detected mass (theoretical mass) A 8 1041.5 1+ 1040.5 Ac-FSEVIFRC-NH.sub.2 (FIG. 8H) Residues 63-70 of SEQ ID NO: 21 (m=1040.51) B 7 1098.6 1+ 1097.6 Ac-GFSEVIFRC-NH.sub.2 (FIG. 8G) Residues 62-70 of SEQ ID NO: 21 (m=1097.53) C 6 1155.6 1+ 1154.6 Ac-GGFSEVIFRC-NH.sub.2 (FIG. 8F) Residues 61-70 of SEQ ID NO: 21 (m=1154.55) D 5 1212.6 1+ 1211.6 Ac-GGGFSEVIFRC-NH.sub.2 (FIG. 8E) 606.9 2+ Residues 60-70 of SEQ ID NO: 21 (m=1211.58) E 3 1326.7 1+ 1325.7 Ac-NGGGFSEVIFRC-NH.sub.2 (FIG. 8C) 663.9 2+ Residues 59-70 of SEQ ID NO: 21 (m=1325.62) F 1 876.5 2+ 1751.0 Ac-NACHNGGGFSEVIFRC-NH.sub.2 (FIG. 8A) Residues 55-70 of SEQ ID NO: 21 (m=1750.77) G 2 1048.0 2+ 2093.7 Ac-IKTNACHNGGGFSEVIFRC-NH.sub.2 (FIG. 8B) 698.9 3+ Residues 52-70 of SEQ ID NO: 21 (m=2092.99) H 1 1098.5 2+ 2195.0 Ac-TIKTNACHNGGGFSEVIFRC-NH.sub.2 (FIG. 8A) 732.7 3+ Residues 51-70 of SEQ ID NO: 21 (m=2194.04) I 3 1148.1 2+ 2294.2 Ac-VTIKTNACHNGGGFSEVIFRC-NH.sub.2 (FIG. 8C) 765.8 3+ Residues 50-70 of SEQ ID NO: 21 4 (m=2293.11) (FIG. 8D) J 3 1198.6 2+ 2395.2 Ac-TVTIKTNACHNGGGFSEVIFRC-NH.sub.2 (FIG. 8C) 799.43 3+ Residues 49-70 of SEQ ID NO: 21 (m=2394.16) K 5 1292.6 2+ 2583.2 Ac-GMTVTIKTNACHNGGGFSEVIFRC-NH.sub.2 (FIG. 8E) 862.2 3+ Residues 47-70 of SEQ ID NO: 21 (m=2582.22)

[1016] The synthesis without pseudoprolines and with double coupling is therefore not working (chemical collapse observed at Thr46). The full length product is not detected. Many truncated products are formed at the beginning of the synthesis.

[1017] 2) STxB Wild Type Synthesis: With 6 Pseudoprolines, Single Coupling for all Residues Except Extra Double Coupling for all Thr Residues, and Capping

[1018] Material and Methods

[1019] Reagents

[1020] Solid-phase synthesis of a full length monomeric mutant of the STxB protein with SEQ ID NO: 2 was performed on a Prelude Instrument (Gyros protein Technologies), at 12.5 μmol scale, using a Fmoc-Arg (pbf)-Wang LL resin.

TABLE-US-00017 SEQ ID NO: 2 TPDC(VT)GKVEYTKYNDD(DT)FTVKVGDKEL(FT)NRWNLQSLL(LS) TAQITGM(VT)IKTNACHNGGG(FS)EVIFR

[1021] Parenthesis correspond to amino acid residues coupled in a pseudoproline dipeptide form (positions 5-6, 18-19, 30-31, 41-42, 50-51, and 63-64 of SEQ ID NO: 2).

[1022] Synthesis

[1023] The resin was swelled twice in 3 mL DCM for 30 seconds with mixing, then once in 3 mL NMP for 5 minutes with mixing.

[1024] Standard Synthesis Cycle

[1025] The synthesis workflow was set as follows on the Prelude Instrument, with one cycle being defined as substeps (1) to (6) defined below, each cycle leading to the addition of one amino acid to the growing peptide, in a linear C- to N-terminal direction following SEQ ID NO: 2. [1026] (1) Deprotection

[1027] This substep was carried out twice in a row per cycle, with 2 mL of 20% piperidine in NMP for 3 minutes each time, with mixing. [1028] (2) Washes [1029] This substep was carried out three times in a row per cycle, with 3 mL of NMP for 30 seconds each time, with mixing. [1030] (3) Coupling [1031] This substep was carried out once per cycle, with 1300 μL of Fmoc-protected amino acid (200 mM in NMP=20.8 eq.; except for cysteine residues: 200 mM in DMF=20.8 eq), 1000 μL of HCTU (250 mM in NMP=20 eq.) and 500 μL of NMM (1 M in NMP=40 eq.) for 10 minutes, with mixing. [1032] (4) Washes [1033] This substep was carried out twice in a row per cycle, with 3 mL of NMP for 30 seconds each time, with mixing. [1034] (5) Capping [1035] This substep was carried out once per cycle, with 2000 μL of Ac.sub.2O (250 mM in NMP) and 500 μL of NMM (1 M in NMP=40 eq.) for 5 minutes, with mixing. [1036] (6) Washes [1037] This substep was carried out three times in a row per cycle, with 3 mL of NMP for 30 seconds each time, with mixing.

[1038] This synthesis workflow therefore includes a single coupling (substep (3)) per amino acid residue, except for all threonine residues for which an adjusted synthesis cycle was carried out as outlined below.

[1039] Adjusted Synthesis Cycle

[1040] As the coupling of threonine residues has been reported in the art to be difficult, possibly due to steric hindrance, we have tested an alternative protocol where the substep of coupling threonine residues was repeated twice, and was extended to 20 minutes instead of 10. The synthesis workflow was therefore adapted as follows for threonine residues: [1041] (1) Deprotection [1042] This substep was carried out twice in a row per cycle, with 2 mL of 20% piperidine in NMP for 3 minutes each time, with mixing. [1043] (2) Washes [1044] This substep was carried out three times in a row per cycle, with 3 mL of NMP for 30 seconds each time, with mixing. [1045] (3) Coupling [1046] This substep was carried out twice in a row per cycle, with 1300 μL of Fmoc-protected threonine (200 mM in NMP=20.8 eq.), 1000 μL of HCTU (250 mM in NMP=20 eq.) and 500 μL of NMM (1 M in NMP=40 eq.) for 20 minutes each time, with mixing. [1047] (4) Washes [1048] This substep was carried out twice in a row per cycle, with 3 mL of NMP for 30 seconds each time, with mixing. [1049] (5) Capping [1050] This substep was carried out once per cycle, with 2000 μL of Ac.sub.2O (250 mM in NMP) and 500 μL of NMM (1 M in NMP=40 eq.) for 5 minutes, with mixing. [1051] (6) Washes [1052] This substep was carried out three times in a row per cycle, with 3 mL of NMP for 30 seconds each time, with mixing.

[1053] Final Deprotection

[1054] Once the whole STxB peptide with SEQ ID NO: 2 was synthetized, the final α amino-protecting group (i.e., the Fmoc protecting group borne by the threonine residue in position 1 of SEQ ID NO: 2), were removed, according to the following substeps: [1055] (1) Deprotection [1056] This substep was carried out twice in a row, with 2 mL of 20% piperidine in NMP for 3 minutes each time, with mixing. [1057] (2) NMP Wash [1058] This substep was carried out once, with 3 mL of NMP for 30 seconds, with mixing. [1059] (3) DCM Washes [1060] This substep was carried out four times in a row, with 3 mL of DCM for 30 seconds each time, with mixing.

[1061] Cleavage

[1062] The resin was cleaved in 5 mL TFA:thioanisole:anisole:TIS:H.sub.2O (82.5:5:5:2.5:5) for 2 hours under stirring. Under these conditions, side-chain protecting groups optionally borne by the amino acid residues (in particular non-aliphatic amino acid residues) were also removed.

[1063] The cleavage solution was then precipitated in 40 mL cold diethyl ether.

[1064] After 3 washes with 45 mL cold diethyl ether, the precipitate was air dried. The precipitate was then mixed in 15 mL 10% acetic acid in water/acetonitrile mix and lyophilized.

[1065] Analyses

[1066] The crude peptide was analysed using Agilent 1100 series HPLC (Santa Clara, USA), with a Vydac C.sub.4 (214TP) column (300 Å, 5 μm, 4.6 mm i.d.×250 mm; Cat. No. 214TP54 S/N E980716-1-4), coupled online to an Esquire ion trap mass spectrometer equipped with an AP-ESI source (Bruker, Germany). The gradient used was from 5 to 100% acetonitrile with 0.1% TFA for 60 minutes at 1 mL/minute.

[1067] Results

[1068] Despite the use of pseudoprolines and a long double coupling for all threonine residues, we observed the presence of deletion peptides (reported in Table 3) corresponding to: [1069] (i) the absence of threonine residue incorporation at positions 49 and 54 of SEQ ID NO: 2; [1070] (ii) the lack of incorporation of the pseudoproline FT at position 30-31 of SEQ ID NO: 2, and [1071] (iii) the lack of incorporation of cysteine at position 57 and isoleucine at position 52 of SEQ ID NO: 2.

TABLE-US-00018 TABLE 3 Calculated Products mass Sequence identification A 1361 Ac-HNGGGFSEVIFR No coupling of Cys 57 Residues 58-69 of SEQ ID NO: 2 B 1648.7 Ac-NACHNGGGFSEVIFR No coupling of Thr 54 Residues 55-69 of SEQ ID NO: 2 C 1878.2 Ac-KTNACHNGGGFSEVIFR No coupling of Ile 52 Residues 53-69 of SEQ ID NO: 2 D 2192.1 Ac-VTIKTNACHNGGGFSEVIFR No coupling of Thr 49 Residues 50-69 of SEQ ID NO: 2 E 4220.5 Ac-NRWNLQSLLLSAQITGMTVTIKTNACHNGGGFSEVIFR No coupling of pseudoproline (FT) 30-31 Residues 32-69 of SEQ ID NO: 2 F 7690.3 Ac-TPDCVTGKVEYTKYNDDDTFTVKVGDKELFTNRWNLQSLLL SAQITGMTVTIKTNACHNGGGFSEVIFR Residues 1-69 of SEQ ID NO: 2

[1072] 3) STxB-Cys-Amide Synthesis with 6 Pseudoprolines, Double or Quadruple Coupling and Capping

[1073] Material and Methods

[1074] Reagents

[1075] Solid-phase synthesis of a full length monomeric mutant of the STxB protein with SEQ ID NO: 2 comprising a cysteine residue at the C-terminus (SEQ ID NO: 21) was performed on a Prelude Instrument (Gyros protein Technologies), at 12.5 μmol scale, using a ChemMatrix rink amide matrix resin.

TABLE-US-00019 SEQ ID NO: 21 TPDC(VT)GKVEYTKYNDD(DT)FTVKVGDKEL(FT)NRWNLQSLL(LS) AQITGMT(VT)IKTNACHNGGG(FS)EVIFRC

[1076] Parenthesis correspond to amino acid residues coupled in a pseudoproline dipeptide form (positions 5-6, 18-19, 30-31, 41-42, 50-51, and 63-64 of SEQ ID NO: 21).

[1077] Synthesis

[1078] The resin was swelled twice in 3 mL DCM for 30 seconds with mixing, then once in 3 mL NMP for 5 minutes with mixing.

[1079] Standard Synthesis Cycle

[1080] The synthesis workflow was set as follows on the Prelude Instrument, with one cycle being defined as substeps (1) to (6) defined below, each cycle leading to the addition of one amino acid to the growing peptide, in a linear C- to N-terminal direction following SEQ ID NO: 21. [1081] (1) Deprotection [1082] This substep was carried out twice in a row per cycle, with 2 mL of 20% piperidine in NMP for 3 minutes each time, with mixing. [1083] (2) Washes [1084] This substep was carried out three times in a row per cycle, with 3 mL of NMP for 30 seconds each time, with mixing. [1085] (3) Coupling [1086] This substep was carried out twice in a row per cycle, with 1300 μL of Fmoc-protected amino acid (200 mM in NMP=20.8 eq.; except for cysteine residues: 200 mM in DMF=20.8 eq), 1000 μL of HCTU (250 mM in NMP=20 eq.) and 500 μL of NMM (1 M in NMP=40 eq.) for 10 minutes each time, with mixing. [1087] (4) Washes [1088] This substep was carried out twice in a row per cycle, with 3 mL of NMP for 30 seconds each time, with mixing. [1089] (5) Capping [1090] This substep was carried out once per cycle, with 2000 μL of Ac.sub.2O (250 mM in NMP) and 500 μL of NMM (1 M in NMP=40 eq.) for 5 minutes, with mixing. [1091] (6) Washes [1092] This substep was carried out three times in a row per cycle, with 3 mL of NMP for 30 seconds each time, with mixing.

[1093] This synthesis workflow therefore includes a systematic double coupling (substep (3)) per amino acid residue.

[1094] Adjusted Synthesis Cycle

[1095] Although it was known in the art that threonine residues were difficult to couple, a repeating the coupling step twice in a row for an extended time of 20 minutes each time was not sufficient to synthesis STxB with no truncated products, as shown above.

[1096] Extensive experiments were thereof carried out to identify so-called “difficult positions”, for which a quadruple coupling (substep (3)) was implemented: Cys 4, Thr 12, Thr 21, Asn 35, Leu 36, Leu 39, Ile 45, Thr 49, Cys 57, and Val 65 (with respect to SEQ ID NO: 21 numbering). The synthesis workflow was therefore adapted as follows for these difficult positions: [1097] (1) Deprotection [1098] This substep was carried out twice in a row per cycle, with 2 mL of 20% piperidine in NMP for 3 minutes each time, with mixing. [1099] (2) Washes [1100] This substep was carried out three times in a row per cycle, with 3 mL of NMP for 30 seconds each time, with mixing. [1101] (3) Coupling [1102] This substep was carried out four times in a row per cycle, with 1300 μL of Fmoc-protected amino acid (200 mM in NMP=20.8 eq.; except for cysteine residues: 200 mM in DMF=20.8 eq), 1000 μL of HCTU (250 mM in NMP=20 eq.) and 500 μL of NMM (1 M in NMP=40 eq.) for 10 minutes each time, with mixing. [1103] (4) Washes [1104] This substep was carried out twice in a row per cycle, with 3 mL of NMP for 30 seconds each time, with mixing. [1105] (5) Capping [1106] This substep was carried out once per cycle, with 2000 μL of Ac.sub.2O (250 mM in NMP) and 500 μL of NMM (1 M in NMP=40 eq.) for 5 minutes, with mixing. [1107] (6) Washes [1108] This substep was carried out three times in a row per cycle, with 3 mL of NMP for 30 seconds each time, with mixing.

[1109] Final Deprotection

[1110] Once the whole STxB peptide with SEQ ID NO: 21 was synthetized, the final α amino-protecting group (i.e., the Fmoc protecting group borne by the threonine residue in position 1 of SEQ ID NO: 21), were removed, according to the following substeps: [1111] (1) Deprotection [1112] This substep was carried out twice in a row, with 2 mL of 20% piperidine in NMP for 3 minutes each time, with mixing. [1113] (2) NMP Wash [1114] This substep was carried out once, with 3 mL of NMP for 30 seconds, with mixing. [1115] (3) DCM Washes [1116] This substep was carried out four times in a row, with 3 mL of DCM for 30 seconds each time, with mixing.

[1117] Cleavage

[1118] The resin was cleaved in 5 mL TFA:thioanisole:anisole:TIS:H.sub.2O (82.5:5:5:2.5:5) for 2 hours under stirring. Under these conditions, side-chain protecting groups optionally borne by the amino acid residues (in particular non-aliphatic amino acid residues) were also removed.

[1119] The cleavage solution was then precipitated in 40 mL cold diethyl ether.

[1120] After 3 washes with 45 mL cold diethyl ether, the precipitate was air dried. The precipitate was then mixed in 15 mL 10% acetic acid in water/acetonitrile mix and lyophilized.

[1121] Analyses

[1122] The STxB peptide was analyzed with a Waters UPLC-MS (comprising an ACQUITY UPLC H-Class sample manager, an ACQUITY UPLC PDA eLambda Detector, and a Single Quadrupole Detector 2 for electron spray ionization/mass spectra [ESI-MS]]).

[1123] The column used was an ACQUITY UPLC BEH C18 column, 130A, 1.7 μm, 2.1 mm×50 mm, with: [1124] Solvent A: 0.1% formic acid in Milli-Q water; [1125] Solvent B: 0.1% formic acid in acetonitrile; [1126] Program: [1127] 5% B for 0.2 minutes, [1128] Linear gradient from 5% to 95% B in 2.3 minutes, [1129] 100% B for 0.5 minutes, and [1130] 5% B for 1 minute.

[1131] The lyophilized STxB peptide was dissolved in 6 M GndHCl, acidified with 1% formic acid aqueous solution and filtered before injection. Analysis is shown at 214 nm (characteristic absorption wavelength for amide bond of peptides).

[1132] Oxidation Analyses

[1133] The STxB peptide was acidified with 1% formic acid aqueous solution and analyzed with a Waters UPLC-MS (comprising an ACQUITY UPLC H-Class sample manager and an ACQUITY UPLC PDA eLambda Detector).

[1134] The column used was an ACQUITY UPLC BEH C18 column, 130A, 1.7 μm, 2.1 mm×50 mm, with: [1135] Solvent A: 0.1% formic acid in Milli-Q water; [1136] Solvent B: 0.1% formic acid in acetonitrile; [1137] Program: [1138] 5% B for 0.2 minutes, [1139] Isocratic flow at 33% B for 2.3 minutes, [1140] 100% B for 0.5 minutes, and [1141] 5% B for 1 minute.

[1142] Oxidation and Folding

[1143] Lyophilized STxB peptide was dissolved to 0.5 mg/mL in oxidation buffer (7 M GndHCl, 50 mM sodium phosphate, 2% DMSO, pH adjusted to 8).

[1144] The solution was then incubated under stirring at 37° C. for 24 hours, to form a disulfide bond between Cys 4 and Cys 57 of SEQ ID NO: 21.

[1145] The solution was then dialyzed at 4° C. with Slide-A-Lyzer™ G2 Dialysis Cassettes, 3.5 kD MWCO from Thermo Scientific against the following: [1146] 3 M GndHCl, 50 mM sodium phosphate pH 8.0, 5 mM EDTA, for 6 to 10 hours; [1147] 1 M GndHCl, 50 mM sodium phosphate pH 8.0, 1 mM EDTA, overnight; [1148] PBS for 4 hours; [1149] PBS for 4 hours; and [1150] PBS overnight.

[1151] After removal from the dialysis cassette, the solution was centrifuged to remove the precipitate. Supernatant was kept and concentrated using centrifugal filters (Amicon Ultra Centrifugal filters, 10 kD MWCO).

[1152] Concentration was measured with Nanodrop 2000, using ε=8250M.sup.−1.Math.cm.sup.−1.

[1153] Small aliquots were flash-freezed and stored at −20° C.

[1154] Intracellular Trafficking Assay by Immunofluorescence

[1155] Intracellular trafficking assays were performed on HeLa cells, cultured at 37° C. under 5% CO.sub.2 in Dulbecco's modified Eagle's medium (DMEM, Invitrogen), supplemented with 10% heat-inactivated fetal bovine serum (FBS), 0.01% penicillin-streptomycin, 4 mM glutamine and 5 mM pyruvate.

[1156] Cells were plated the day before on lamellae in 4-well plates, 60 000 cells/well.

[1157] Binding and Internalization

[1158] Cells were incubated for 45 minutes at 37° C. in presence of 0.2 μM STxB in complete medium, then washed 3 times with 500 μL PBS with Ca.sup.2+ and Mg.sup.2+ (PBS.sup.++).

[1159] Fixation

[1160] Cells were treated with 500 μL of 4% paraformaldehyde (PFA) during 20 minutes, then washed once with 50 mM of NH.sub.4Cl, and incubated with 50 mM of NH.sub.4Cl for at least 30 minutes.

[1161] Permeabilization

[1162] Cells were washed 3 times with 500 μL of PBS/BSA/Saponin 1× (1×PBS/1.0% BSA/0.1% Saponin), and then incubated at room temperature for 30 minutes in presence of 500 μL of PBS/BSA/Saponin 1×.

[1163] Incubation with Antibodies

[1164] Lamellae were incubated with 30 μL of primary antibody dilution into PBS/BSA/Saponin 1× for 30 minutes at room temperature, then washed 3 times with PBS/BSA/Saponin 1×.

[1165] Primary antibodies used were the mouse monoclonal clone 13C4 anti-STxB antibody (Strockbine et al., 1985. Infect Immun. 50(3):695-700), at 1/250 dilution; and a home-made rabbit polyclonal antibody against the Golgi marker Giantin, used at 1/100 dilution.

[1166] Same was done with the secondary antibodies (anti-mouse Cy3 and anti-rabbit A488 used at 1/100 dilution each).

[1167] Slide Preparation

[1168] Lamellae were washed in water and then added on slides on 6 μL of Fluoromount-G™+Hoechst. Polymerization was allowed for 30 minutes at 37° C.

[1169] Microscope Observation

[1170] Slides were observed with the following equipment from the Biomaging Cell and Tissue Core Facility of the Institut Curie in Paris, France (PICT-IBiSA): upright Leica DM6000 microscope (with a CCD 1392×1040 CoolSnap HQ2 camera from Photometrics, pixel: 6.45 μm; and a Lumen 200 lamp illumination source from Prior Scientific). A Leica HCX PL Apo 63× oil objective was used for pictures.

[1171] Results

[1172] Synthesis

[1173] This strategy of synthesis is very effective. Only few truncated synthesis products were observed (minor peak at 1.63 min in FIG. 9A), the main UPLC peak mass corresponding to the expected product (peak at 1.79 min in FIG. 9A). 42.7 mg of STxB peptide were recovered after cleavage, representing approximately 44% yield in mass.

[1174] Oxidation and Folding

[1175] Directly after synthesis and cleavage, the crude STxB peptide could then be oxidized to form the intramolecular disulfide bond between Cys 4 and Cys 57 of SEQ ID NO: 21, and folded to yield a functional protein.

[1176] Folding was achieved via stepwise dialyses, in buffers successively comprising 6 M, 3 M and 1 M GndHCl, and ending up with buffers comprising PBS but no GndHCl.

[1177] After centrifugation, the supernatant was recovered, concentrated and analyzed by UPLC-MS. Folded STxB peptide obtained after oxidation is very pure, as seen on FIG. 9B, and represents a 15%-yield (on average from 4 separate experiments).

[1178] Oligomerization and Intracellular Trafficking Assay by Immunofluorescence

[1179] Folded STxB was than tested for oligomerization and functionality using an immunofluorescence assay. This assay relies on the fact that upon normal trafficking, STxB in its pentameric form should be in the Golgi after 50 minutes of incubation at 37° C.

[1180] Two STxB samples were tested and compared: [1181] “rSTxB-Cys”, corresponding to STxB with SEQ ID NO: 21 obtained by recombinant expression; and [1182] “sSTxB-Cys-amide”, corresponding to STxB with SEQ ID NO: 21 obtained by chemical synthesis as described herein.

[1183] Cells were incubated with 0.2 μM of the STxB samples for 45 minutes at 37° C. for binding to their receptor Gb3 and internalization; then fixed and labelled with antibodies.

[1184] FIG. 10 shows colocalization of recombinant and synthetic STxB with the Golgi (labelled with anti-giantin antibodies) after 45 minutes of incubation at 37° C., meaning that both STxB samples contain functional STxB pentamers. Note that synthetic STxB-Cys [sSTxB-Cys-amide] was transported as efficiently as recombinant STxB-Cys [rSTxB-Cys] by retrograde trafficking from the plasma membrane to the Golgi apparatus.

[1185] Conclusion

[1186] In conclusion, this synthesis strategy with 6 pseudoprolines, systematic double coupling and quadruple coupling at difficult positions, is very effective, and the STxB protein that was obtained after refolding is biologically active in a highly selective intracellular trafficking assay.

[1187] Moreover, it is conceivable to reduce the number of coupling substep iterations in favour of an equally extended time of coupling (e.g., double coupling for twice 10 minutes can be replaced by a single coupling for 20 minutes once only).

Example 4

[1188] STxB Synthesis by Native Chemical Ligation (NCL) with 4 or 5 Pseudoprolines, Single/Double Coupling and No Capping

[1189] As an alternative to the full-length synthesis procedure described above, STxB synthesis was also performed using a native chemical ligation (NCL) approach.

[1190] For this, STxB was divided into two segments (“STxB-N” with SEQ ID NO: 22; and “STxB-C” with SEQ ID NO: 23), and several pseudoprolines were used in the synthesis of the longer STxB-N segment.

TABLE-US-00020 SEQ ID NO: 22 TPDCVTGKVEYTKYNDDDTFTVKVGDKELFTNRWNLQSLLLSAQITGMTV TIKTNA SEQ ID NO: 23 CHNGGGFSEVIFR

[1191] STxB-N with SEQ ID NO: 22 was synthesized as a C-terminal hydrazide ( . . . Thr-Asn-Ala-NH—NH.sub.2) for NCL. STxB-C was synthesized with a C-terminal acid using standard Fmoc-SPPS.

[1192] 1) STxB-N Synthesis with 4 Pseudoprolines and Simple/Double Coupling STxB-N with SEQ ID NO: 22 was synthesized as described above, with amino acid residues coupled in a pseudoproline dipeptide form at positions 5-6, 18-19, 30-31, and 41-42 of SEQ ID NO: 22 (in parenthesis below).

TABLE-US-00021 SEQ ID NO: 22 TPDC(VT)GKVEYTKYNDD(DT)FTVKVGDKEL(FT)NRWNLQSLL(LS) AQITGMTVTIKTNA

[1193] A 10 minutes single coupling was performed for each amino acid residue, except for positions Thr 1, Cys 4, Tyr 11, Thr 12, Thr 21, Asn 35, Leu 36, Ile 45 and Thr 46 of SEQ ID NO: 22, where a double coupling (twice 10 minutes) was performed.

[1194] Fmoc deprotection was monitored using the Gyros-PTI Prelude X peptide synthesizer, which allows UV monitoring to measure Fmoc deprotection.

[1195] The UV images show incomplete Fmoc deprotection at the amino acids . . . LLLSAQIT . . . (positions 39-46 of SEQ ID NO: 22), leading to presumed deletions at those positions based on MS data: [1196] 6300 Da: STxB-N with SEQ ID NO: 22, no deletion [1197] 6187 Da (Δ 113 Da): isoleucine deletion [1198] 6172 Da (Δ 128 Da): glutamine deletion [1199] 6100 Da (Δ 199 Da): LS pseudoproline deletion

[1200] 2) STxB-N Synthesis with 5 Pseudoprolines and Simple/Double Coupling

[1201] The same strategy was implemented, but STxB-N with SEQ ID NO: 22 was synthesized with amino acid residues coupled in a pseudoproline dipeptide form at positions 5-6, 18-19, 30-31, 41-42, and 50-51 of SEQ ID NO: 22 (in parenthesis below).

TABLE-US-00022 SEQ ID NO: 22 TPDC(VT)GKVEYTKYNDD(DT)FTVKVGDKEL(FT)NRWNLQSLL(LS) AQITGMT(VT)IKTNA

[1202] A 10 minutes single coupling was performed for each amino acid residue, except for positions 1, 4, 11, 12, 21, 35, 36, 45 and 46 of SEQ ID NO: 22, where a double coupling (twice 10 minutes) was performed.

[1203] Again, Fmoc deprotection was monitored using the Gyros-PTI Prelude X peptide synthesizer, which allows UV monitoring to measure Fmoc deprotection.

[1204] The addition of a fifth VT pseudoproline in the STxB-N sequence resolves the problem of deletions, and highlights the crucial role of this additional modification in the efficiency of the STxB-N synthesis. Mass spectroscopy analysis showed 5 peaks: [1205] 6300 Da: STxB-N with SEQ ID NO: 22, no deletion [1206] 6356 Da (Δ 56 Da): tBu protecting group [1207] 6400 Da (Δ 100 Da): unknown—commonly seen with hydrazide resin peptides [1208] 6414 Da (Δ 114 Da): unknown [1209] 6453 Da (Δ 153 Da): dimethoxybenzyl modification

[1210] 3) Native Chemical Ligation of STxB-N and STxB-C

[1211] Native chemical ligation was performed according to standard methods (Dawson et al., 1994. Science. 266(5186):776-779; Johnson & Kent, 2006. J Am Chem Soc. 128(20):6640-6646) using 4-mercaptophenylacetic acid (MPAA) as thiol catalyst and TCEP as reducing agent. The hydrazide method was used to generate C-terminal thioesters (Zheng J S, Tang S, Qi Y K, Wang Z P, Liu L. Nat Protoc. 2013; 8(12):2483-2495; Zheng J S, Tang S, Guo Y, Chang H N, Liu L. Chembiochem. 2012; 13(4):542-546; Fang G M, Li Y M, Shen F, et al. Angew Chem Int Ed Engl. 2011; 50(33):7645-7649).

[1212] One equivalent of StxB-N (0.7 μmol) was dissolved in 675 μL of “activation buffer” (6 M GuHCl, 100 mM NaPO.sub.4, pH 3) and activated (conversion of hydrazide to acyl azide) with freshly prepared sodium nitrite (20 mM final concentration NaNO.sub.2) for 20 minutes at −20° C.

[1213] In parallel, 3 equivalents of StxB-C (2.1 μmol) were dissolved in 675 μL of a solution containing 200 mM MPAA in “ligation buffer” (6 M GuHCl, 200 mM phosphate, pH 7).

[1214] Both solutions were combined, and the final pH was adjusted to 7.0-7.2 to initiate the ligation after activation of the hydrazide. After 15 minutes, 150 μL of 0.5 M TCEP in pH 7 ligation buffer was added to the reaction to a final concentration of 50 mM.

[1215] Finally, 3 mL of 10% AcOH was added. The solution was centrifuged at 5000 g to remove any aggregates. The supernatant was then filtered with 0.2 μm nylon filter before HPLC purification and subsequent analysis.