Method for the affinity purification of recombinant proteins based on the lectin activity of the CRD of a galectin

Abstract

The present invention relates to a novel method for the affinity purification of proteins of interest in a single step, based on the lectin activity of the CRD (Carbohydrate Recognition Domain) of a galectin or part of said domain retaining the ability to bind β-galactosidase derivative.

Claims

1. A fusion protein, comprising: a lectin tag derived from a carbohydrate recognition domain (CRD) of a naturally occurring galectin, fused with a molecule of interest, with a site for cleavage by protease inserted between the two, wherein said lectin tag comprises the amino acid sequence of SEQ ID NO: 1.

2. The fusion protein as claimed in claim 1, wherein said lectin tag comprises the CRD.sub.SAT domain of galectin-3 set forth in amino acids 1 to 156 of SEQ ID NO: 3.

3. The fusion protein as claimed in claim 2, wherein the cleavage site is a site for cleavage by TEV protease.

4. A CRD.sub.SAT domain comprised of the amino acid sequence set forth in SEQ ID NO: 1.

5. The fusion protein as claimed in claim 1, wherein the cleavage site is a site for cleavage by TEV protease.

6. The fusion protein as claimed in claim 3, wherein said site for cleavage by TEV protease comprises amino acids 165 to 171 of SEQ ID NO:3.

7. The fusion protein as claimed in claim 3, wherein said fusion protein comprises an 8-residue spacer arm disposed between said CRD.sub.SAT domain of galectin-3 and said site for cleavage by TEV protease, wherein said 8-residue spacer arm is the amino acid sequence set forth in amino acids 157 to 164 of SEQ ID NO: 3.

8. A CRD.sub.SAT domain comprising the amino acid sequence set forth in residues 1 to 156 of SEQ ID NO: 3.

Description

DESCRIPTION OF THE FIGURES

(1) FIG. 1 depicts a circular and linear diagram of the pCARGHO vector.

(2) FIG. 2 depicts the main features of the pCARGHO vector (SEQ ID NOs: 2 and 3).

(3) FIG. 3 depicts a schematic structure of human galectin-3.

(4) FIG. 4 depicts the protein sequence of human galectin-3 and of the 3 CRDs designed (SEQ ID NO: 4).

(5) FIG. 5 depicts the 3D structure of the lectin CRD domain of the galectin-3 interacting with lactose.

(6) FIG. 6 depicts monitoring the purification of human galectin-3 by lactose affinity chromatography.

(7) FIG. 7 depicts the diagram of the purification method of the presently disclosed subject matter by lactose affinity.

(8) FIG. 8 depicts the follow-up purification of 3 forms of truncated human galectin-3 by lactose affinity chromatography. The CRD.sub.LITIL (A), CRD.sub.GGVVP (B), and CRD.sub.SAT (C) were expressed in E. coli. Rosetta 2 (DE 3) bacteria. The bacteria were lysed and purification tests were carried out on lactose-agarose column. At each purification step, a sample was taken to be run on acrylamide gel (SDS-PAGE). After electrophoretic migration, the gel was stained with Coomassie blue. SSO: sonication supernatant of bacterial extract before passage on lactose affinity column. CSO: sonication pellet of bacterial extract before passage on lactose affinity column. SPR: supernatant after passage on the lactose affinity column. E: elution fraction location of the protein of interest.

(9) FIG. 9 depicts an amino acid sequence alignment of the CRD.sub.SAT domain of galectin-3 in different mammals, and also the consensus sequence of the CRD.sub.SAT domain.

(10) FIG. 10 depicts monitoring the purification of bacterial thioredoxin (Trx1) by lactose affinity chromatography according to the process of the presently disclosed subject matter.

(11) FIG. 11 depicts monitoring the purification of extracellular domain TREM-1, residues 21 to 136, by lactose affinity chromatography according to the process of the presently disclosed subject matter.

DETAILED DESCRIPTION

Example 1: Researching the Optimal Form of Truncated Human Galectin-3: Monitoring Purification of 3 Truncated Forms by Lactose Affinity Chromatography

(12) Galectin-3

(13) Galectin-3 is an animal lectin of 243 to 286 amino acids depending on the species (Cooper, Biochim. & Biophys. Acta, 1572(2-3): 209-231, 2002) [8]. It is approximately 30 kDa and is composed of a small N-terminal domain, a lateral chain and a C-terminal lectin domain (CRD: Carbohydrate Recognition Domain) (FIGS. 3 and 4) (Leffler et al., Glycoconj. J., 19: 433-440, 2004; Ochieng et al., Biochim. & Biophys. Acta, 1379: 97-106, 1998) [9, 10]. Formed of several beta sheets (Salomonsson et al., J. Biol. Chem., 285: 35079-35091, 2010; Seetharaman et al., J. Biol. Chem., 273: 13047-13052, 1998) [11, 12], the lectin domain enables the protein to interact with molecules containing β-galactoside residues, for example lactose (FIG. 5).

(14) Given its lectin properties, galectin-3 was able to be purified specifically in a single step by affinity chromatography using an agarose resin grafted with lactose molecules, according to the protocol described previously (FIG. 6).

(15) For this purpose, the galectin-3 (whole form) was expressed in E. coli C41(DE3) bacteria then purified on lactose-agarose column. At each purification step, a sample was taken to be run on acrylamide gel (SDS-PAGE). After electrophoretic migration, the gel was stained with Coomassie Brilliant Blue. A: bacterial extract before passage on lactose affinity column. B: bacterial extract after passage on lactose affinity column. C and D: column washes. E: elution fraction of galectin-3; elution with a solution of PBS+lactose 150 mM.

(16) The results are presented in FIG. 6.

(17) The ease with which this purification is carried out, and also the high degree of purity obtained, pointed towards the idea of developing a fusion partner (tag) intended for the purification of recombinant proteins by lactose affinity chromatography. The idea was to use a part of the lectin domain of the galectin-3 (CRD) capable of binding lactose in order to constitute this fusion partner and enable purification in a single step, during which this partner could be cleaved by TEV protease (FIG. 7).

(18) Researching the Optimal Form of Truncated Galectin-3

(19) Creation of the Optimized Nucleotide Sequence Encoding CRD.sub.SAT and Integration in an Expression Vector of pET-20b Type

(20) Starting from the nucleotide sequence of human galectin-3, 3 CRD sequences encoding 3 different CRD proteins were cloned: CRD.sub.LITIL (14 kDa), CRD.sub.GGVVP (15 kDa) and CRD.sub.SAT C17 kDa, natural form, non-synthetic, non-optimized) (figure. 4). As noted previously, the abbreviation “CRD”, as in the “CRD domain of a galectin”, refers to the highly conserved carbohydrate recognition domain of a galectin that has the ability to bind lactose. In the context of the present invention, truncated forms of the CRD domain are identified by their starting amino acid residues. Namely, the CRD.sub.SAT domain begins with the “SAT” sequence at amino acid residues 96-98 of SEQ ID NO: 4 and extends to the end at amino acid 250. Similarly, the CRD.sub.GGVVP domain begins with the “GGVVP” sequence at amino acid residues 124-128 of SEQ ID NO: 4 and extends to amino acid 250; likewise, the CRD.sub.LITIL domain begins with the “LITIL” sequence at amino acid residues 131-135 of SEQ ID NO: 4 and extends to amino acid 250. In the context of the fusion proteins of present invention, these CRD-based domains are collectively referred to as “lectin tags”.

(21) It emerges therefrom that the 3 CRDs are expressed but have varying solubility. Thus, CRD.sub.LITIL (A) was in the totally insoluble form (in the pellet) and was not located in the eluate, and therefore was impossible to purify. CRD.sub.GGVVP (B) was produced in small amounts, partly in soluble form, but lost its lectin function and therefore could not be purified. CRD.sub.SAT (C) was produced entirely in soluble form and was able to be purified (was located in the eluate) (FIG. 8).

(22) Given the various tests carried out, CRD.sub.SAT was chosen to constitute the desired fusion partner. The physicochemical characteristics of this protein, determined in silico, are as follows:

(23) TABLE-US-00001 (SEQ ID NO: 1, natural form)           10          20         30 MSATGAYPA TGPYGAPAGP LIVPYNLPLP GGVVPRMLIT   40         50       60         70 ILGTVKPNAN RIALDFQRGN DVAFHFNPRF NENNRRVIVC 80          90         100        110 NTKLDNNWGR EERQSVFPFE SGKPFKIQVL VEPDHFKVAV 120        130        140        150 NDAHLLQYNH RVKKLNEISK LGISGDIDLT SASYTMI
Number of amino acids: 156
Molecular weight: 17 360.9 Da
Isoelectric point: 9.30
Total number of negatively-charged amino acids (Asp+Glu): 13
Total number of positively-charged amino acids (Arg+Lys): 17
Molar extinction coefficient: 12 950 M.sup.−1 cm′ (at 280 nm).
Abs 0.1% (=1 g/l) 0.746, with the proviso that all the cysteines are in reduced form.

(24) The sequence encoding CRD.sub.SAT was optimized in silico in order to promote expression of this heterologous protein in E. coli (removal of codon bias) and to make it possible to increase the solubility thereof (substitution of arginine 36 for lysine), and also the rigidity thereof by increasing bulk (substitution of alanine 152 for threonine), thereby making it possible for the protease to cleave 14 amino acids downstream.

(25) This optimized CRD.sub.SAT sequence was integrated into an expression vector of pET-20b type: the pCARGHO vector (FIG. 2), in order to enable the expression and purification of a protein of interest according to example 3.

Example 2: pCARGHO: Materials and Methods

(26) The pCARGHO plasmid enables the production of a fusion protein may consist, in order, of: the CRD.sub.SAT from human galectin-3, a spacer arm enabling flexibility, a site for cleavage by TEV protease and the protein of interest.

(27) The E. coli strain used should be of (DE3) type, that is to say should have the T7 RNA polymerase gene integrated into its genome.

(28) The pCARGHO plasmid is derived from pET20b(+) from Novagen and includes the following elements (FIG. 1):

(29) TABLE-US-00002 ELEMENTS POSITION Origin of replication 1944 T7 promoter 797-813 Ribosome binding site: RBS 742-747 CRD.sub.SATG sequence 242-736 TEV recognition site sequence 221-241 Multiple cloning site: MCS 159-215 T7 terminator 26-72 F1 origin 3694-4149 bla ampicillin resistance gene 2705-3562

(30) The main characteristics of the pCARGHO vector are indicated in FIG. 2 (SEQ ID NOs: 2 and 3).

(31) Cloning in the pCARGHO Plasmid

(32) The protocol below is an example of cloning of a PCR fragment of a protein of interest in the pCARGHO vector. For some experiments (enzymatic digestion, ligation reaction, bacterial transformation), reference should especially be made to the suppliers of the reagents used.

(33) The PCR fragment used should contain the NcoI restriction site at its 5′ end and another BamHI, EcoRI, SacI, SalI, HindIII, NotI and XhoI restriction site at its 3′ end. The ends may either be blunt or extended by a 3′ adenosine. It should be ensured that the composition of the PCR fragment guarantees that the reading frame is abided by from the start codon ATG. 1) Digest 1 μg of pCARGHO plasmid and 1 μg of PCR fragment in parallel in 20 μl of 1× digestion buffer (10× stock) with 10 units of NcoI and 10 units of the second endonuclease chosen (depending on the site available in the MCS), at 37° C. for 1 h to 2 h. The enzymes will then be inactivated at 65° C. for 10 minutes. 2) Verify the complete digestion of the plasmid after migration on agarose gel (5 μl). 3) Purify the digested PCR fragment and plasmid, with the aim of eliminating the MCS fragment from the plasmid and the free ends from the PCR fragment. The purification may be carried out on gel and/or using specific kits. 4) Assay the plasmid and the PCR fragment (insert). 5) Prepare the following ligation mixture: 30-50 ng of the digested plasmid 50 ng of insert in 1 μl of stock ligation buffer 10× 1 μl of T4 DNA ligase H.sub.2O q.s. to 10 μl 6) Incubate for 10 minutes to 2 h at room temperature or at 16° C. for 16 h. 7) Transform the competent cloning bacteria (TOP10, DH5a type): take off 2 μl of the ligation reaction and add 50 μl of bacteria. Incubate for 30 minutes in ice. Heat to 42° C. for 1 minute. 8) Add 200 μl of SOC (or LB) medium and incubate at 37° C. for 20 minutes to 1 h. Spread on selective agar (LB Agar, 100 μg/ml of ampicillin). Incubate overnight at 37° C. 9) Look for the presence of positive clones (verify the presence of the insert in the plasmid) by methods well known in the art such as PCR directly on colonies, minipreparation of plasmid DNA, digestion by suitable restriction enzymes and migration on agarose gel, test of overexpression of the fusion protein. 10) Sequence the plasmids extracted from the clones assumed to be positive, in order to validate the molecular cloning.
Expression of the Fusion Protein “CRD.sub.SAT Protein of Interest”
Test of Expression of the Fusion Protein 1) Transform competent expression bacteria [BL21(DE3) type] with the pCARGHO-X vector (X being the sequence encoding the protein of interest fused to the CRD.sub.SAT tag) 2) Inoculate the colonies isolated on agar in 5 ml of LB medium+ampicillin 100 μg/ml and culture until 2×10.sup.8 cells/ml (A.sub.600=0.5-0.6). 3) Divide the sample into two cultures of 2.5 ml. 4) Add IPTG into one of the cultures at a final concentration of 1 mM. Incubate both cultures at 37° C. for 2-3 h. 5) Sample 500 μl from each culture. Centrifuge at maximum speed for 1 minute, dispense with the supernatants and suspend the bacterial pellets with 100 μl of Laemmli buffer 1×. 6) Boil the samples for 1 minute. Run 10 μl of each sample and a molecular weight marker on an SDS-PAGE gel at a suitable percentage. Migrate, and reveal by staining with Coomassie Brilliant Blue for example.
Production of the Fusion Protein

(34) Once the sequencing and the expression tests have been validated, the protein of interest fused to the CRD.sub.SAT tag (CRD.sub.SAT protein of interest) may be produced according to the procedure below. 1) Transform competent expression bacteria [BL21(DE3) type] with the pCARGHO-X vector (X being the sequence encoding the protein of interest fused to the CRD.sub.SAT tag). Carry out preculture in Luria Bertani LB medium+ampicillin 100 μg/ml, at 37° C. for 16 h. 2) Inoculate this preculture to 1/100th in 1 liter of rich medium of Luria Bertani LB type+ampicillin 100 μg/ml. 3) Culture at 37° C. until 2×10.sup.8 cells/ml (A.sub.600=0.5-0.6). Add IPTG at a final concentration of 1 mM. Incubate the bacteria at 37° C. with agitation for 2-3 h. 4) sample 20 μl of the crude bacterial suspension (A), centrifuge (4000 g for 20 minutes) the culture, eliminate the supernatant and suspend the bacterial pellet in 25 ml of lysis buffer.
Different adjuvants may make it possible to avoid proteolysis (ethylenediaminetetraacetic acid (EDTA), phenylmethylsulfonyl fluoride (PMSF), etc.) or oxidation (dithiothreitol (DTT), β-mercaptoethanol). 5) Place the sample in ice and lyze the bacteria using a sonicator or a French press. Sample 20 μl of this lyzed bacterial suspension (A′). 6) Centrifuge the lyzed bacterial suspension at 20 000 g for 20 minutes. Recover the protein supernatant (Sp). Sample 20 μl of this supernatant (B). Dilute, where appropriate, with lysis buffer or column buffer.
Purification of the Fusion Protein “CRD.sub.SAT Protein of Interest”
Purification in Automatic Mode 1) After filtration over 0.45 μm membrane, inject the sample Sp on a lactose-Sepharose® resin (15 to 20 ml) equilibrated beforehand with 5 CV (column volumes) of column buffer. Sample 20 μl of the soluble fraction that has flowed through the chromatography column (C). 2) Wash the column with 10 CV of column buffer. Sample 20 μl of the washing liquid (D) at the column outlet. 3) Elute the fusion protein with 1-2 CV of column buffer and +150 mM of lactose. Collect the corresponding fractions (size of the fractions: approximately ⅕ of the volume of the column). Sample 20 μl of eluate (E). Monitor the fusion protein present in the fractions collected, by absorption at 280 nm and/or by colorimetric methods (BCA, Bradford, etc.).
If the CRD.sub.sat tag would be undesirable for the subsequent applications of the protein of interest, it is possible to cleave it after elution of the fusion protein (cf. section on cleavage of the fusion protein). 4) Bring together the fractions containing the fusion protein and concentrate if necessary. Conserve the protein of interest at −80° C. after freezing in liquid nitrogen with or without addition of 30 to 50% glycerol. 5) Evaluate the quality of the purification by SDS-PAGE+staining: dilute samples A to E to half strength in Laemmli 2× buffer then boil (95° C., 5 minutes) before migrating them on an acrylamide gel at the suitable percentage. 6) Regenerate the column with 5 CV of 2 M NaCl solution then 5 CV of water. Conserve the column in water+ethanol 20% or in Tris-Hcl 20 mM, EDTA 1 mM.
Purification in Batch Mode 1) Deposit the sample Sp on a lactose agarose resin (10-15 ml) equilibrated beforehand with 5 to 10 CV of column buffer (Tris 20 mM, EDTA 5 mM, NaCl 150 mM, pH to be adapted depending on the protein of interest) and place under weak agitation for 1 to 2 hours. Then arrange a column with fritted disk. Sample 20 μl of the soluble fraction that has flowed through the chromatography column (C). 2) Wash the column with 10 CV of column buffer then with 10 CV of PBS buffer. Sample 20 μl at the column outlet. 3) Elute the fusion protein with 2.5 CV of PBS buffer and +150 mM of lactose. 4) Evaluate the quality of the purification, assay the fusion protein then store it according to the procedures described in the section on purification in automatic mode. 5) Regenerate the column with 10 CV of 2 M NaCl solution.
Cleavage of the Fusion Protein “CRD.sub.SAT Protein of Interest”

(35) The fusion protein is cleaved by TEV protease, the cleavage site of which is located at the C-terminus of the CRD.sub.SAT tag. The cleavage occurs after the elution of the fusion protein and should be followed by a step of either cation exchange or size exclusion chromatography in order to separate the CRD.sub.SAT tag (pI=8.7, MW=18.8 kDa) and the TEV protease (pI=8.8, MW=27 kDa) from the protein of interest. 1) Dilute the concentrated solution of fusion protein to a concentration of 1-2 mg/ml in the cleavage buffer: 25 mM Tris-HCl, pH 8, 150-500 mM NaCl, 15 mM β-mercaptoethanol. Sample 20 μl of sample (A). 2) Add the TEV protease at a ratio of 1:100, i.e. 1 mg (10 000 units) of protease per 100 mg of fusion protein. The “ideal” ratio may be optimized. 3) Incubate the mixture at 4° C. overnight or else at room temperature or at 30-37° C. for shorter periods of time, the limiting factor being the stability of the protein of interest. Sample 20 μl of sample after cleavage (B). 4) Run samples A and B on SDS-PAGE gel in order to verify the effectiveness of the cleavage. 5) Eliminate the CRD.sub.SAT tag and also the TEV protease by: a) cation exchange chromatography: i. Dialyze or dilute the cleavage reaction mixture with the aim of lowering the ionic strength with a 20 mM Tris-HCl, 25 mM NaCl, pH 7 buffer. The salt concentration should not exceed 25 mM. ii. Equilibrate the SP-Sepharose® or equivalent column with 5 CV of column buffer (according to the manufacturer's recommendations). iii. Deposit the mixture on the column and immediately collect the fraction not retained, which contains the protein of interest (as long as the protein of interest has an acid pI (<7)). iv. Concentrate the protein of interest, where appropriate, and conserve it at 80° C. after freezing in liquid nitrogen with or without addition of 30 to 50% glycerol. v. Regenerate the column with 5 CV of 2 M NaCl buffer (elution of the CRD.sub.SAT tag and of the TEV protease), 5 CV of water. Conserve in water+20% ethanol. b) size exclusion chromatography i. Equilibrate the Superdex 75® or Superdex 200® or equivalent column with 2 CV of column buffer (according to the manufacturer's recommendations). ii. Inject the mixture on the column and collect the fractions corresponding to the protein of interest without tag [as a function of the molecular weight (size) thereof]. iii. Concentrate the protein of interest, where appropriate, and conserve it at 80° C. after freezing in liquid nitrogen or addition of 30 to 50% glycerol.
If the tag has a molecular weight (MW) that is very different from that of the protein of interest, detection by staining the gel is sufficient. If, however, the MW thereof is very similar, more specific detection by western blot is possible using anti-CRD.sub.sat antibodies.

(36) iv. Conserve the column in water+20% ethanol.

(37) Addendum: If TEV digestion is not carried out, and if lactose would be inappropriate for the subsequent applications of the purified protein, dialysis may be carried out or a gel filtration column may be carried out.

Example 3: Monitoring Purification of Bacterial Thioredoxin (Trx1) by Lactose Affinity Chromatography

(38) Trx1 fused to CRD.sub.SAT was expressed in E. coli Rosetta2 (DE3) bacteria according to the protocol described above.

(39) The bacteria were lyzed and purification was carried out on lactose-agarose column according to the protocol described above.

(40) At different purification steps, a sample was taken to be run on acrylamide gel (SDS-PAGE).

(41) After electrophoretic migration, the gel was stained with Coomassie Brilliant Blue.

(42) The results are presented in FIG. 10.

(43) MW: molecular weights.

(44) Lac: purified fraction of Trx1-CRD.sub.SAT fusion protein on lactose-agarose resin, elution with 150 mM of lactose.

(45) TEV: fraction after cleavage with TEV protease, to 1/100, overnight at ambient temperature.

(46) SP flow through: fraction not retained after injection on SP Sepharose (cation exchange) column of the digestion reaction medium (TEV).

(47) SP 100%: fraction eluted with 1 M of NaCl (SP Sepharose).

Example 4: Monitoring Purification of Human Membrane Receptor TREM-1 (Extracellular Domain) by Lactose Affinity Chromatography

(48) TREM-1 (21-136) fused to CRD.sub.SAT was expressed in E. coli C41 (DE3) bacteria according to the protocol described above.

(49) The bacteria were lyzed and purification was carried out on lactose-agarose column according to the protocol described above.

(50) At different purification steps, a sample was taken to be run on acrylamide gel (SDS-PAGE).

(51) After electrophoretic migration, the gel was stained with Coomassie Brilliant Blue.

(52) The results are presented in FIG. 11.

(53) MW: molecular weights.

(54) Lac: purified fraction of CRD-TREM-1(21-136) fusion protein on lactose-agarose resin, elution with 150 mM of lactose.

(55) TEV: fraction after cleavage with TEV protease, to 1/100, overnight at ambient temperature.

(56) Phe1: fraction not retained after injection on Phenyl Sepharose (hydrophobic interaction) column of the digestion reaction medium (TEV), containing the CRD tag.

(57) Phe2: fraction, eluted with 300 mM of ammonium sulfate, containing TREM-1, 13.7 kDa, residues 21 to 136 with a yield of 4 mg/l of bacterial culture.

REFERENCE LIST

(58) 1. Stadtman et al., Free Radic Biol Med., 9(4):315-25, 1990 2. Chen et al., Mol Cell Biol. May; 26(10): 3728-3737, 2006 3. Andberg et al., Protein Science, 16:1751-1761, 2007 4. Ortiz-Salmeron et al., Eur. J. Biochem., 268, 4307-4314, 2001 5. International application WO 2009/121994 6. International application WO 9966053 7. Tielker et al., Biotechniques, 41(3): 327-332, 2006 8. Cooper, Biochim. & Biophys. Acta, 1572(2-3): 209-231, 2002 9. Leffler et al., Glycoconj. J., 19: 433-440, 2004 10. Ochieng et al., Biochim. Biophys. Acta, 1379: 97-106, 1998 11. Salomonsson et al., J. Biol. Chem., 285: 35079-35091, 2010 12. Seetharaman et al., J. Biol. Chem., 273: 13047-13052, 1998 13. Taylor et al., Biochem. J., 274: 575-580, 1991 14. Kelker et al., J. Mol. Biol. 342, 1237-1248, 2004 15. Radaev et al., Structure 11, 1527-1535, 2003