WO2008117259A1

WO2008117259A1 - Protein domains with choline affinity for improving expression, immobilization and purification of polypeptides

Info

Publication number: WO2008117259A1
Application number: PCT/IB2008/051314
Authority: WO
Inventors: Jesús Miguel Sanz Morales; Beatriz MAESTRO GARCÍA-DONAS; Victor Manuel HERNÁNDEZ ROCAMORA; Miguel AREVALO RODRÍGUEZ; Angel Cebolla Ramirez
Original assignee: Universidad Miguel Hernandez
Priority date: 2007-01-29
Filing date: 2008-01-29
Publication date: 2008-10-02
Also published as: ES2321788B1; ES2321788A1

Abstract

The present invention concerns a polypeptide sequence derived from the choline-binding module of Streptococcus pneumoniae amidase LytA, and a set of expression vectors for obtaining traductional fusions to the codifying sequence of said polypeptide in bacterias and yeasts. The invention also concerns various structural improvements and improvements to the purification process, which enable the expression and stability of fusions, and also minimize interferences with the fused polypeptide and optimize the proteolytic separation of said module. An example of an embodiment described is the construction of a set of plasmidic vectors with the necessary operative signals for expression in bacterias or yeasts of one of these sequences, and which facilitate fusion thereof to heterologous genes in order to obtain hybrid proteins which can be immobilized in a range of mediums, and optionally purified by means of affinity chromatography.

Description

Protein domains of choline affinity to improve expression, immobilization and purification of polypeptides

Technical sector

The present invention relates to the use of a polypeptide sequence that, fused to a protein of interest, allows its immobilization on certain supports by specific affinity for them, and its subsequent purification from them in a single step. A set of plasmid vectors with the appropriate genetic elements to facilitate the recombinant expression of fused proteins to the aforementioned polypeptide sequence, both in bacteria and yeasts, as well as the procedures necessary to carry out the aforementioned immobilization and purification processes is also described. . The sector of the technique to which this invention can be applied thus comprises the production, immobilization and purification of polypeptides and proteins of biotechnological interest, whether for industrial, agronomic, therapeutic or diagnostic purposes in biomedicine, or as a research and analysis tool. in the field of life sciences.

State of the art

The immobilization of enzymes in solid supports, as well as the purification of proteins in a single chromatographic step are two processes of essential interest in Biotechnology that often share a common prior procedure: the fusion of the protein of interest with a polypeptide or "tag" affinity that enables a strong interaction with a solid support (Uhlén, M.et al (1992). Curr. Op. Biotechnol. 3, 363-369; Waugh. (2005). Trends Biotechnol. 23, 316-320). The addition to said support of a cell extract of an organism in which hybrid protein expression has occurred enables specific binding of the fusion to the support, and the elimination of the rest of proteins by means of the thorough washing of the same. At this point it is possible to choose, either to use the enzymatic activity thus retained, or to selectively elute the hybrid protein by adding a competing ligand (affinity chromatography base), with the result of a high degree of purity in the final preparation of the protein of interest.

The immobilization of an enzyme in a solid phase facilitates its recovery and reuse in a reactor (Trevan, M. D. et al (1990). In: Basic Biotechnology. Ed. Acribia, Zaragoza, Spain.). In addition, the interactions that the support maintains with the enzyme and its ability to protect it from the solvent can increase its stability, and therefore its yield. Among the immobilization methods that are used, the non-covalent bonding methods to the support have several advantages over covalent bonding procedures. In the first case, very varied supports can be used that do not need to be activated, minimizing the chances of structural distortion of the protein by not modifying it covalently. In addition, non-covalent bonding is in general a reversible process, allowing the regeneration of the supports, of which there is a wide range available commercially as any expert in the field can verify.

As for affinity chromatography, it is obvious that the reduction in the number of purification steps of a protein translates not only in a higher yield, but also in an increase in purity and a considerable saving of time and effort.

Choline-binding modules (ChBM) constitute a family of polypeptides that are part of the so-called choline-binding proteins (ChBP, "choline-binding proteins), present in a variety of microorganisms (Swiatlo, E. et al (2004). In: The Pneurnococcus. American Society for Microbiology Press, Washington DC). The ChBMs consist in turn of sequence repetition. highly conserved of approximately 20 amino acids (ChBR or "choline-binding repeats"; Pfam code PF01473: http: // www. sanger.ac .uk // cgi-bin / Pfam / getacc? PF01473.), which form structures of the type loop-hairpin-β (Fernández-Tornero, C. et al (2001). Nat. Struct. Biol. 8, 1020-1024.) Two consecutive ChBRs form a hill binding site. The affinity of ChBMs for hill and its structural analogues (Sanz, JM et al. (1988). FEBS Lett. 232, 308-312.), has allowed the development of an efficient fusion protein purification system with some of these ChBMs (Sánchez- Puelles, JM et al. (1992) Eur. J. Biochem. 203, 153-159.) Basically, the procedure consists in applying a cell extract containing the pr fusion otein on a derivatized support with tertiary or quaternary amines. The immobilized protein thus maintains its functionality, and can be easily eluted by the addition of a competing ligand, such as choline. The procedure thus described is the basis of a prior patent (ES 2 032 717).

Table 1 shows a comparative study of the characteristics of the most commonly used affinity "tags" together with the hill binding module system (Sánchez-Puelles, JM et al. (1992) Eur. J. Biochem. 203 , 153-159; Lichty, JJ et al (2005) Protein Expression and Purification 41, 98-105; Waugh, DS (2005) Trends Biotechnol. 23, 316-320). As can be seen from the study, the procedure object of this patent has a series of specific advantages that make it highly competitive, and that justifies its exploitation to become an extended protein immobilization and purification procedure. Fundamentally, the system is highly specific, you can use a large variety of supports already available commercially (resins, paper, multi-well plates, etc ...), shows very few incompatibilities, and has all the necessary powers to think of an efficient scaling for its use in bioreactors at industrial level.

Field of the invention

The present invention is useful for the expression, immobilization and purification of proteins of interest by fusion to the described polypeptide sequences. Therefore, it can be used to develop useful products for reagent industries and life science research tools. In addition, it can be applied to the chemical industry for the purification of useful enzymes, for the construction of enzymatic bioreactors for chemical synthesis or analysis, or processes that require biotransformation, even within the agri-food field. It is also equally applicable to the pharmaceutical and healthcare industry in general, to produce and purify recombinant proteins applicable to diagnosis and therapies, as well as for the preparation of immobilized protein matrices for application in proteomics.

Table 1. Comparative study of the affinity tags most used in the purification or immobilization of recombinant proteins.

Table 1 (Continued)

In bold, the most favorable characteristics of the comparative study are shown, while in italics the major disadvantages are shown.

Abbreviations: ChBM, choline binding modules; IMAC, immobilized metal affinity chromatography; MBP, maltose binding protein; GST, glutathione-S-transferase.

Detailed description of the invention

The object of the present invention comprises immobilization by non-covalent binding and purification of recombinant proteins based on their fusion to polypeptide sequences derived from that of the choline-binding domain of LytA lithic amidase of S. pneurnoniae (C-LytA) , through genetic systems that allow its expression in bacteria and yeasts. The present invention describes improvements to the general method described in a previous patent (ES 2 032 717), since it defines the use of a specific region of the C-LytA domain and opens up a range of possibilities of using this system under non-standard conditions of temperature and pH

The present invention comprises the use of a cropped version of a DNA sequence (SEQ ID N ² 1, coding for C-LytA, SEQ ID N ² 2), as well as a mutant variant thereof, which encode smaller polypeptides and with higher expression levels than the original full C-LytA polypeptide, maintaining or improving its properties for the immobilization and purification of recombinant fusion polypeptides on solid supports derivatized with tertiary and quaternary amines.

The trimmed polypeptide derived from C-LytA, LYTAG2 (SEQ ID

N ² 3) has 131 amino acids and a molecular weight of 14.62 kDa. Amino acids 1-104 of the LYTAG2 polypeptide correspond to residues 59-163 of SEQ ID No. 2. Amino acids 105-131 of the LYTAG2 polypeptide correspond to an artificial sequence with demonstrated ability to form an α-helix.

The LYTAG2t polypeptide (heat resistant LYTAG2, SEQ ID N ² 4) also has 131 amino acids and a molecular weight of 14.73 kDa, and differs from the LYTAG2 polypeptide only in the substitutions Asnl9 → Lys, Ser20 → Lys, Thr53 → Lys and Thr76 → Lys

Novelty and development of the invention

The novel features of the objects of the present invention can be summarized in 4 points:

1.- The minimized LYTAG2 domain improves the expression and solubility characteristics observed for the entire C-LytA domain, while its smaller size decreases the probability of interference with the functions of the protein fused to it. In addition, it has been described that the C-LytA region excluded from LYTAG2 is partially deployed in solution (Maestro and Sanz (2005) Biochem J. 387, 479-88), so its elimination minimizes the risk of aggregation and attack unspecific for proteases.

2.- In order to further reduce the possible structural interaction with the fused polypeptide, the system object of the present invention incorporates a spacer amino acid sequence capable of adopting a rigid conformation in helix D (Arai, R. et al. (2001 ) Prot. Eng. 14, 529-532), thus separating both polypeptides permanently, as opposed to other looser connectors that are used commercially. The effective separation between polypeptides could also facilitate accessibility to the various specific cut endopeptidases that are intended to be used for the proteolytic separation of the affinity tag, since it is well known that in a good number of occasions the site of recognition of the Endopeptidase is hidden due to nonspecific interactions between the fused domains. In short, this improvement of the system is especially useful when the expressed fusion protein is going to be used directly, in solution or immobilized in one of the supports Amino chromatography described above, for any type of biotechnological test or application, and also facilitates the elimination of the affinity tag when its presence in the protein of interest is incompatible with certain applications or tests.

3- An improvement in the purification procedure is proposed, consisting of the use of moderate concentrations of detergents in order to help the solubility of hydrophobic proteins. In the configuration object of the present invention, the coding sequence of the protein to be expressed can be fused following a coding sequence of the target recognizable by a specific endopeptidase, which is located next to the spacer sequence, allowing recovery of the protein of interest without the presence of unwanted amino acid residues at its amino terminal end. The addition of certain detergents such as sodium dodecyl sulfate at a given concentration allows the LYTAG2 module to be selectively deployed, thus facilitating the access of the endopeptidase to its cutting site.

4- Finally, the functionality of the minimized domain LYTAG2 can be improved by substituting certain amino acids located on the surface of the protein (variant LYTAG2t), significantly increasing the yield of immobilization / purification at high temperatures and at acidic pH.

In the described configuration, the expression of the resulting fusion protein is produced under the control of an improved mutant version of the prokaryotic promoter Pm, adjustable by the NahR / XylS2 system, inducible by salicylate (Cebolla et al. (2001) Nucleic Acids Res 29, 759-766). Another preferred configuration is to place the fusion coding sequence under the control of a promoter compatible with eukaryotic organisms such as yeasts, insect cells, mammals, etc., so that the fusion protein expressed in this type of cells can be purified.

Applications

Through the use of usual molecular biology techniques, hybrid proteins containing the LYTAG2 or LYTAG2t sequences can be designed in order to:

1.- Its heterologous expression in bacteria, yeasts, cell cultures, etc.

2.- Its immobilization in derivatized supports with tertiary and quaternary amines, and whose matrices may consist of:

2.1) Chromatographic resins (dextran, agarose, cellulose, methacrylate, etc ...);

2.2) Paper or membranes;

2.3) Multiwell plates;

2.4) Micro or nanoparticles.

3.- The use of the fusion protein thus immobilized for the following processes:

3.1.) Purification of the fusion protein from the above-mentioned supports by means of specific elution with choline or its analogues;

3.2.) The construction of bioreactors to carry out biotransformations and enzymatically catalyzed chemical reactions;

3.3.) The design of biosensors;

3.4.) The construction of protein arrays for high performance assays in proteomics.

4.- The separation of the LYTAG2 or LYTAG2t polypeptides by a combination of limited proteolysis and gentle treatment with sodium dodecyl sulfate, and the resulting purification of the protein of interest. Examples of realization

Example 1: Purification analysis of reduced versions of C-LYTAG expressed E. coli under the control of an improved Pm promoter.

The present example describes how clipped fragments of the C-LytA choline binding domain can be selected and modified while maintaining and even exceeding the properties of the entire fragment for the expression and purification of fusion proteins. The studies described in this example have, on the one hand, the identification of a polypeptide sequence, derived from that of the choline affinity domain of the LytA protein (C-LytA), but smaller than this, which allows it to be used as a "tag" of purification, maintaining an affinity for the supports used to immobilize and / or purify proteins fused thereto similar to that observed for fusions to the complete C-LytA domain.

At the same time, the genetic constructions necessary to carry out the expression of the fusion proteins object of these studies have been carried out using a series of bacterial vectors constructed for this purpose, and which are based on the cascade activation system described and patented by A. Onion and collaborators (Onion, A. et al (2001) NucleicAcids Res, 29, 759-766; Onion, A., et al (2002). Appl Environ Microbiol, 68, 5034-5041; Patent WO00 / 78976). The developed plasmids incorporate significant improvements with respect to the expression and purification vectors of pALEX series proteins already marketed by Biomedal (www.biomedal.com), such as the use of a mutant Pm promoter with lower basal activity, and the presence of a more versatile multiple cloning region. Other advantages are the incorporation of a more efficient ribosome binding site and two consecutive transcription terminators, located after the region of multiple cloning and three stop codons, one in each of the three phases of translational reading.

The construction of these plasmids has been carried out as described below. In the first step an improved version of the Pm promoter was selected. To do this, from plasmid pCCD5 (Onion, A., et al (2002). Appl Environ Microbiol, 68, 5034-5041), and using as primers the oligonucleotides PMMu5 (SEQ ID N ⁰ 5) and PMNh3 (SEQ ID N ⁰ 6), a mixture of 112 bp PCR products containing mutant versions of the Pm promoter was amplified under mutagenic conditions, which is activated by the product of the xylS2 mutant allele of the xylS gene of Pseudomonas in the presence of salicylate (Onion , A. et al (2001) Nucleic Acias Res, 29, 759-766; Onion, A., et al (2002). Appl Environ Microbiol, 68, 5034-5041). The mixture of PCR products was digested with Muñí and Nhel, and the resulting fragments were cloned between the EcoRI and Xbal sites of plasmid pJBAl11 (Andersen, JB et al (1998) Appl Environ Microbiol, 64, 2240-2246), previously digested with these enzymes, thus replacing the original promoter from which a green-fluorescent protein (GFP) is expressed. From the resulting plasmid mixture, one was selected with a mutant Pm promoter that, introduced into a host E. coli strain for said cascade expression system, showed a particularly reduced level of activity (detected as GFP expression ), in basal conditions (in the absence of the inducer, salicylate), maintaining a level of activity as high as that of the wild promoter under induction conditions (presence of 2 mM salicylate). From the selected plasmid, pIZ1203, and in order to isolate the Pm promoter and the downstream region thereof from the activity of a second promoter present in the plasmid (Plac, from the pUC18Not vector from which the pJBAlll plasmid is derived), plasmid pIZ1203Rev8 was constructed, in which the 1878 bp Notl fragment containing the Pm promoter is in the reverse orientation with respect to pIZ1203. In the next step, from pIZ1203Rev8 were constructed three new plasmids, called pMABl, pMAB2 and pMAB3, obtained by replacing in pIZ1203Rev8 the 755 bp Sphl-HindIII fragment containing the GFP coding region, by synthetic duplexes, obtained respectively from the complementary oligonucleotide pairs MA0 (SEQ ID No. 7) and MAO20 (SEQ ID No. 8), MA021 (SEQ ID No. 9) and MAO22 (SEQ ID No. 10), or MAO23 (SEQ ID No. 11) and MAO24 (SEQ ID No. 12).

The inserted synthetic duplexes contain targets for various restriction enzymes, which can be used in the cloning of DNA fragments and their expression in the new vectors.

Next, and to validate the functioning of the new expression vectors, the expression of a model protein in one of these plasmids was analyzed. To this end, a 3116 bp BarnüI-HindlII fragment of the E. colí lacZ gene was cloned, with the coding sequence of the enzyme D-galactosidase, between the Barnül and HindIII sites of the pMAB3 vector described above. In the resulting plasmid, pMAB3-lacZ, the lacZ sequence is fused to the multiple cloning region of pMAB3, in the correct reading phase with respect to the first ATG start codon following the ribosome binding region, and which is part of the target for the Sphl enzyme present at the beginning of the multiple cloning region. Plasmid pMAB3-lacZ was used to transform the commercial bacterial strain REG-I (Biomedal, www.biomedal.com), host of the cascade system regulatory module (Cebolla, A., et al (2002). Appl Environ Microbiol, 68, 5034-5041), and the expression of D-galactosidase activity in cell extracts of transformant cultures under induction conditions was verified (results not shown). In the next step, pMAB3-lacZ was used as a starting point in the construction of a series of plasmids for the expression of "mini-C-LytA" fusions (reduced versions of C-LytA, represented in Figure IA) at the end amino terminal of D-galactosidase. For this, the complete coding sequence of C-LytA (SEQ ID No. 1) was used as a template.

The "mini-C-LytAl" construct, encoding the polypeptide SEQ ID No. 13, was obtained by PCR from the sequence SEQ ID No. 1 using oligonucleotides MAO27 (SEQ ID No. 14) and MAO29 (SEQ ID N 15).

The "mini-C-LytA2" construct, encoding the polypeptide SEQ ID No. 16, was obtained by PCR from the sequence SEQ ID No. 1 using oligonucleotides MAO28 (SEQ ID No. 17) and MAO30 (SEQ ID N 18).

The "mini-C-LytA3" construct, encoding the polypeptide SEQ ID No. 19 was obtained by PCR from the sequence SEQ ID No. 1 using oligonucleotides MAO30 (already described) and MA031 (SEQ ID No. 20).

Finally, the "mini-C-LytA4" construct, encoding the polypeptide SEQ ID No. 21 was obtained by PCR from the sequence SEQ ID No. 1 using oligonucleotides MAO32 (SEQ ID No. 22) and MAO33 (SEQ ID No. 23).

In the next step, the described PCR products were fused, using an overlapping PCR strategy using "bridge" oligonucleotides (MAO38 (SEQ ID No. 24) to "mini-C-LytAl"; MAO37 (SEQ ID No. 25) for "mini-C-LytA2"; MAO39 (SEQ ID No. 26) for "mini-C-LytA3", and MAO40 (SEQ ID No. 27) for "mini-C-LytA4"), to the coding sequence of a spacer polypeptide called HL4GS (LAEAAAKEAAAKEAAAKEAAAKAAAGS, SEQ ID No. 28) fused directly at its carboxyl terminus to the specific cut-off protease target sequence NIa, (NVVVHQA, SEQ ID No. 29) (Perez-Martin, J., et al. (1997). Protein Eng, 10, 725-730), which was obtained by PCR with oligonucleotides MAO34 (SEQ ID No. 30) and MAO35 (SEQ ID No. 31), complementary at their 3 ^' ends.

Finally, the PCR products obtained, containing the modules "mini-C-LytAl-HL4GSNIa", "mini-C-LytA2-HL4GSNIa", "mini-C-LytA3-HL4GSNIa" and "mini-C-LytA4-HL4GSNIa" , were digested with Sphl and inserted in the proper orientation at the Sphl site of pMAB3-lacZ, respectively obtaining plasmids pMAB3-LlHN-LacZ-56, pMAB3-L2HN-LacZ-16, pMAB3-L3HN-LacZ-28 and pMAB3- L4HN-LacZ-34.

To analyze the expression and purification properties of the fusion proteins encoded in the described plasmids, the REG-I strain already described with each of them was transformed separately, including the REG-I transformant with the precursor plasmid in these studies pMAB3-lacZ. The transformants described were grown at 37 ° C with stirring in liquid LB medium with 100 mg / ml ampicillin (to select the presence of plasmids, which confer resistance to this antibiotic) to an optical density at 600 nanometers (D.0. ₆ oonm) of between 0.8 and 1.0, and then for 12 hours at 20 ⁰ C with shaking in the same medium, in the absence or presence of 2 mM salicylate. For each transformant, the cells from both cultures were collected by centrifugation and used by sonication in a 20 mM potassium phosphate buffer solution pH 7.0, 0.1% Triton X-100, 1.5 M NaCl. The supernatant of the crude extract, obtained after centrifuging it to remove cell debris, was applied to an affinity matrix consisting of DEAE-Sepharose (Sigma) by incubating with gentle stirring at room temperature for 1 hour, after which the resin with the same buffer used in lysis, was equilibrated with 20 mM potassium phosphate pH 7.0, 0.1% Triton X-IOO, and finally the fusion protein was eluted with a volume of 250 mM choline in 20 mM potassium phosphate pH 7.0, 0.1% Triton X-100, repeating this last step twice more. Figure IB shows the results of the protein analysis obtained by SDS-PAGE and Coomassie staining. As can be seen, all the constructions analyzed were expressed significantly under induction conditions. However, only the "mini-C-LytAl-HL4GSNIa-LacZ" and "mini-C-LytA2-HL4GSNIa-LacZ" fusions could be purified from the cell extracts under the conditions tested. Of these two, the construction "mini-C-LytA2-HL4GSNIa-LacZ" was chosen for further studies, due to the high level of expression detected for this fusion with respect to the "mini-C-LytAl-HL4GSNIa-LacZ" fusion, and its apparent greater affinity for the resin used, which could allow washing under more drastic conditions and thus reduce contamination of the eluted protein with resin-bound protein in a non-specific manner. The mini-C-LytA2 module therefore forms the basis of the LYTAG2 sequence.

Example 2

Fusion expression vectors to LYTA G2 in E. coli.

The following example shows how advanced bacterial expression vectors can be obtained using the LYTAG2 tag. The construction of a series of plasmid vectors is carried out, for the expression of translational fusions to the coding sequence of the choline affinity domain LYTAG2 described in Example 1, which can be used in the production and purification of heterologous proteins in bacteria E coli In these new plasmids, the coding sequence of the NIa endopeptidase target has been replaced by another coding sequence of that of another specific cut endopeptidase, the enterokinase, which recognizes the DDDDK sequence (SEQ ID No. 32)

(Choi, et al. (2001). Biotechnol Bioeng, 75, 718-724). To do this, from plasmid pMAB3-L2HN-lacZ-16 described in the

Example 1 and using MAO62 oligonucleotides as primers

(SEQ ID No. 33), which includes the inverse complementary sequence of the DDDDK target encoder, and the MAO49 oligonucleotide

(SEQ ID No. 34), a 469 bp product was amplified by PCR.

This product was digested with Sphl, and the resulting 414 bp fragment was cloned in the appropriate orientation at the Sphl site of each of the plasmids pMABl, pMAB2 and pMAB3 described in Example 1, obtaining, respectively, the plasmids pMAB7, pMAB8 and pMAB9, whose scheme is depicted in Figure 2. pMAB7, pMAB8 and pMAB9 allow expression of fused polypeptides to the carboxyl end of the sequence

MLADRWRKHTDGNWYWFDNSGEMATGWKKIADKWYYFNEEGAMKTGWVKYKDT WYYLDAKEGAMVSNAFIQSADGTGWYYLKPDGTLADRPEFTVEPDGLITVKLAEAAA KEAAAKEAAAKEAAAKAAAGSDDDDK (SEQ ID N ⁰ 35).

The plasmid sequences described differ only in the presence of one, or two bases inserted immediately following the Sphl target located at the beginning of the multiple cloning region, thus facilitating translational fusion to LYTAG2, in the appropriate reading phase. , of any fragment that can be cloned into one (or between two) of the restriction targets of said multiple cloning region. The presence of a target for the blunt-cut restriction enzyme PshAI, overlapping with the sequence coding for the target of the kinase, also allows direct fusion of DNA fragments to the sequence of said target, subsequently making it possible to cleave in a manner precise the purification tag of the expressed protein.

To complete this exemplary embodiment, a new plasmid, pMAB9-lacZ, was constructed from the pMAB9 vector for the expression of a LYTAG2-D-galactosidase fusion, by inserting between the Bamñl and HindIII sites of pMAB9 the 316 bp Bamñl-HindlII fragment of the E. coli lacZ gene described in Example 1. Also, and with the aim of comparing this fusion to LYTAG2 with another identical to the complete C-LytA domain, and thus being able to confirm previously obtained data, which indicated a better expression of the "tag" trimmed LYTAG2, another plasmid was constructed as described below. A 353 bp PCR product, amplified from plasmid pMAB3-LlHN-LacZ-56 described in Example 1 with the oligonucleotides MAO49 (already described) and MAO106 (SEQ ID No. 36), and another PCR product of 572 pb, overlapping with the first sequence and amplified from plasmid pMAB7, described in this example, with oligonucleotides MAO107 (SEQ ID No. 37) and MAO50 (SEQ ID No. 38), were used in a third PCR in which obtained a product of 795 bp containing the coding sequence of the complete C-LytA domain fused to that of the spacer and the target for the enterokinase already described (SEQ ID No. 39). The product thus obtained was digested with Sphl, and the resulting 585 bp fragment was cloned, in the appropriate orientation, at the Sphl site of plasmid pMAB3-lacZ of Example 1, resulting in plasmid pMAB3-LytA-HEK-lacZ.

Both plasmids, pMAB9-lacZ and pMAB3-LytA-HEK-lacZ, were used to transform the REG-I host strain already described, and the transformants obtained were cultured in a manner similar to that described in Example 1, in the presence of different concentrations of the salicylate inducer. From the resulting cultures, soluble total protein cell extracts were obtained that were analyzed by SDS-PAGE and Coomassie staining, and in parallel, used in activity assays - | —galactosidase to more quantitatively determine the amount of protein of fusion expressed in active form, using o-nitrophenyl-β-D-galactopyranoside as substrate (Miller, J. (1972). Experiments ín Molecular Genetics. CoId Spring Harbor, New York, USA). The results of both types of analysis, shown in Figure 3, confirm that the LYTAG2-LacZ fusion is expressed, in soluble and active form, at a level considerably higher than that of the C-LytA-LacZ fusion, and therefore that the new "tag" LYTAG2 constitutes a significant improvement over the entire C-LytA domain. The differences observed could be due, among other reasons, to the fact that the messenger RNA encoding the LYTAG2-LacZ fusion has greater stability (and therefore a longer half-life), or translates into protein more efficiently than RNA messenger encoding the C-LytA-LacZ fusion. Likewise, the LYTAG2 polypeptide could have a greater resistance than the C-LytA domain to attack by cellular proteolytic machinery, the best expression of the LYTAG2-LacZ fusion being due to its greater stability once synthesized.

Example 3

Expression vector of fusions to LYTA G2 in Sαcchαromyces cerevisiαe.

This example shows how yeast vectors can be developed that use the LYTAG2 purification tag. The construction of a plasmid vector is carried out, for the expression of translational fusions to the coding sequence of the choline affinity domain LYTAG2, which can be used in the production and purification of heterologous proteins in S. cerevisiae yeast.

The construction of this eukaryotic expression vector, based on the Galactose-inducible Gal4-pGAil system (Johnston, M. (1992). In: The Molecular Biology of the Yeast Saccharomyces. EW Jones, Pringle, JR and Broach, JR CoId Spring Harbor, NY, CoId Spring Harbor Laboratory Press. 2: 193-281), has been carried out by inserting the module LYTAG2- multiple cloning region of plasmid pMAB7 described in Example 2, between the regions corresponding to the pGALl promoter and the tCYCl terminator of plasmid p416-promGALl (Mumberg, D. et al (1994) Nucleic Acids Res, 22, 5767-5768). To do this, using two independent PCR products amplified from pMAB7 with the oligonucleotide pairs MA091 (SEQ ID No. 40) and MAO92 (SEQ ID No. 41), or MAO93 (SEQ ID No. 42) and MAO94 (SEQ ID No. 43), a hybrid product with ends compatible with those generated by cutting with the restriction enzymes Xbal and Xhol, which was inserted into p416-promGALl, was obtained by denaturation and renaturation of the mixture of both PCR products. previously digested with these enzymes. The resulting vector pMAB10-5 (Figure 4) can be used similarly to the parental plasmid pMAB7, for the expression of LYTAG2 fusions in yeast.

To demonstrate the usefulness of the expression and purification system described in this example, the yeast strain W303-1A (MATa ura3-l leu2-3, 112 trpl-1 canl-100 ade2-l hís3-ll, 15) was transformed with a plasmid derived from the pMABlO-5 vector, obtained by cloning between the Sphl and HindIII sites thereof a 3128 bp Sphl-HindIII fragment obtained from the plasmid pMAB3-lacZ described in Example 1, and carrying the D-galactosidase coding sequence . The resulting plasmid, pMABlO-LacZ, contains a translational fusion between the coding sequences of the LYTAG2 polypeptide and the D-galactosidase identical to that expressed by the bacterial plasmid pMAB9-lacZ described in Example 2.

The transformant cells of strain W303-1A with plasmid pMABlO-LacZ, grown in synthetic medium SD without uracil (to select the presence of the plasmid, which complements the ura3 mutation of the yeast strain), and with 2% galactose as carbon source, were collected by centrifugation, and were used by mechanical breakage with spheres of glass. The extract obtained was processed and the LYTAG2-LacZ fusion protein was purified and analyzed in a similar manner as described in Example 1. The results are shown in Figure 5.

Example 4. Use of detergents in the purification of polypeptides fused to the choline binding domain in DEAE-cellulose.

The present example shows how a choline binding sequence that includes LYTAG2 can allow purification of fused polypeptides in the presence of certain detergents useful for purifying hydrophobic proteins. In certain cases, the hydrophobicity of the protein fused to a C-LytA derivative may necessitate the use of detergents in order to increase its solubility and / or avoid non-specific interactions thereof with the affinity matrix or with other proteins. (Hjerten. S. et al (1988) Biochirn. Biophys. Acta. 939, 476-84; Moldes, C, et al (2004), Appl. Environ. Microbiol. 10, 4642-4647; Ivanov, AV et al. (2004) J. Biol. Chem. 279, 29832-29840; Yueh, SCH et al. (2006) J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. (In press)). For this reason, the effect of the presence of several of the detergents most commonly used in protein solubilization on the process of purification of a C-LytA derivative in DEAE-cellulose was verified. The C-LytA derivative used (SEQ ID No. 44) is a cropped version thereof, which contains amino acids 32 to 162 of the sequence SEQ ID No. 2, and therefore contains the sequence LYTAG2 described previously (SEQ ID No. 3), and can be expressed from plasmid pCE17 (Sanchez-Puelles, et al (1990) Gene, 89, 69-75). 200 ml cultures of strain RB791 transformed with plasmid pCE17, grown at 37 ⁰ C and induced with 2% lactose for 16 hours were centrifuged and sonicated in 20 mM potassium phosphate buffer pH 7.0, 150 mM NaCl, and containing the detergents shown in Table 2.

TABLE 2. EFFECT OF DETERGENTS ON THE PURIFICATION OF THE DERIVATIVE OF C-

LytA SEQ ID N ⁰ 44

Detergent Concentration Relative yield (%)

Without detergent ___ 100

0.2 mM SDS 85

0.8 mM SDS 60

SDS + 3.0 mM hill 1.6 mM> 95

Triton X-100 2O mM 65 n-Dodecil β-D-Maltoside 5 mM> 95

3 - [(3-Colamidopropyl) dimethylammonium] -l-propanesulfonate 16 mM> 95

(CHAPS)

16 mM 90 sodium colato

46 mM 65 sodium colato

The extracts were then added on a column of DEAE-cellulose (Sigma), equilibrated in the same buffer. The column was thoroughly washed with a buffer containing 20 mM potassium phosphate pH 7.0 plus 1.5 M NaCl and the detergent used, and finally the protein was eluted using a buffer containing 20 mM potassium phosphate pH 7.0 plus 150 mM NaCl , the detergent used and 150 mM choline chloride. As can be deduced from the results shown in Table 2, the purification procedure generally admits the presence of a moderate concentration of detergents without significant impairment of their performance, which opens the door to the immobilization and / or purification of hydrophobic proteins such as the membrane ones. Example 5. Deployment of the choline binding label by submyellar concentrations of sodium dodecyl sulfate in the absence of choline.

The present example shows how a method for separation of the purification tag can be performed so that the specific proteolysis site is more exposed to the site-specific endopeptidase chosen to separate the tag from the fused polypeptide. Figure 6 shows the circular dichroism spectrum in the far ultraviolet region of the sequence protein SEQ ID No. 44 recorded in 20 mM phosphate buffer, pH 7.0, in the absence of choline, and in the presence or absence of SDS 2 mM. It can be seen that the protein completely loses its three-dimensional structure in the presence of the aforementioned concentration of SDS, which is below its critical micellar concentration under these conditions (3.25 mM). There is no record in the scientific literature of any protein that is fully deployed in the presence of such a low concentration of SDS. On the other hand, the separation of affinity "tags" by proteolysis is usually hindered by the inaccessibility of the protease at its cutoff point due to the latter's concealment due to the folding of the fused polypeptides themselves. Therefore, our results suggest that the proteolytic cleavage of SEQ ID No. 44 in a fusion protein can be facilitated thanks to the specific deployment of the choline binding module by using submicelar concentrations of SDS, which should not affect even the structure of the fused polypeptide, or the activity of many proteases, such as factor Xa (GST Gene Fusion System Handbook, GE Healthcare), Lys-C endoproteinase, GIu-C endoproteinase (V8 protease), Arg-C endoproteinase, pronase, trypsin, chymotrypsin, elastase (see references at: http: //www.cbs.cnrs.jr/MAJ/WGICIELS/CWE/cloedoc/PROWLj>roteαsej> rofll.html), etc ... Example 6. Mutations in the choline binding domain that allow purification in DEAE-cellulose at higher temperatures

The present example shows how the sequence of the purification tag can be modified so that it can withstand affinity purification processes under high temperature conditions, as well as methods for doing so. The use of immobilized enzymes in bioreactors at high temperatures has a high industrial interest, since under these conditions i) the reaction rate is substantially increased; H) decreases the viscosity of reagents and products, and Hi) the bioreactor works in adverse conditions for the growth of contaminating microorganisms. Figure 7 shows the stability profile of the SEQ ID No. 44 protein bound to DEAE-cellulose as a function of temperature. In this experiment, 1 mg of protein was immobilized in 1 ml of DEAE-cellulose (Sigma). Then, the resin was heated for 30 minutes, and allowed to cool for 1 hour. After washing with 20 mM phosphate buffer, pH 7.0, the protein was eluted with 20 mM phosphate buffer, pH 7.0 plus 150 mM choline. This result indicates that the effectiveness of SEQ ID No. 44 as an affinity tag decreases sharply above 65 ⁰ C. In this sense, it has been described that thermophilic organisms proteins acquire their thermostability frequently thanks to the presence of favorable electrostatic interactions on the molecular surface (Vieille, C. and Zeikus, GJ (2001). Microbiol. Mol. Biol. Rev. 65, 1-43). For this reason, after the inspection of the three-dimensional structure described for the C-LytA protein (Fernández-Tornero, C. et al (2001). Nat. Struct. Biol. 8, 1020- 1024), we observe that Asn51 substitutions → Lys, Ser52 → Lys, Thr85 → Lys and ThrlOδ → Lys could increase the thermal stability of module SEQ ID No. 44. Consequently, based on plasmid pCE17 (Sanchez-Puelles, et al (1990) Gene, 89, 69-75) the aforementioned mutations were introduced by PCR using oligonucleotide pairs SEQ ID No. 45 and SEQ ID No. 46, SEQ ID No. 47 and SEQ ID No. 48, and SEQ ID No. 49 and SEQ ID No. 50, thus obtaining plasmid pCE17m4, which allows the expression of the mutant variant SEQ ID No. 51, which contains the module LYTAG2t (SEQ ID No. 4). Protein SEQ ID No. 51 was purified by affinity chromatography on cellulose DEAE with the same procedure described in Example 4. The protein maintains its affinity for the resin at temperatures up to 90 ⁰ C in the same tests as those performed with SEQ ID No. 44, with an acceptable purification performance (> 80%, Figure 7), and therefore could be used for the immobilization and / or purification of thermostable proteins.

Example 7. Modifications of choline binding domains and procedures for purification at more acidic pH

The following example shows how the modified purification tag in certain residues can allow its purification at more acidic pHs. The theoretical isoelectric point of polypeptide SEQ ID No. 44 is 5.1, which implies that the expected solubility for this protein acquires its minimum value at this pH value. To investigate how this circumstance could affect the purification procedure, three identical purifications were made starting from 200 ml cultures of the grown and induced strain RB791 [pCE17m4] as specified in Example 4. After centrifugation of the samples, The cell precipitate was resuspended and sonic in 20 mM sodium phosphate buffer, pH 7.0, in 20 mM sodium acetate buffer, pH 5.0, or in 20 mM glycine buffer, pH 3.0. Purifications were carried out as specified in Example 4, but maintaining the pH specified in each case, and in the absence of Triton X-100. The results are shown in the Table 3. It can be verified that the purification performance of the SEQ ID No. 44 sequence at pH 5.0 or lower is practically zero.

TABLE 3. PURIFICATION PERFORMANCE IN DEAE-CELLULOSE AT ACID pH.

On the other hand, the insertion of four plants increases the isoelectric point of SEQ ID No. 51 to 8.33, which allows a recovery of the solubility at acidic pH and therefore an evident improvement in the purification performance from strain RB791 [pCE17m4] (Table 3). These results open the possibility of using LYTAG2t for immobilization and / or purification of proteins under moderately acidic conditions.

Description of the figures

Figure 1. (A), scheme of the reduced versions of C-LytA analyzed. (B), expression and purification of the mini-C-LytA-HL4GSNIa fusions to LacZ (or LacZ, as a control). 1 total uninduced culture protein; 2, total protein of the induced culture; 3, non-soluble fraction of the induced culture extract; 4, soluble fraction of the induced culture extract; 5, protein not retained in the resin; 6, washed resin protein; 7, 8 and 9, consecutive elution fractions with choline; M, molecular weight marker (BioRad).

Figure 2. Bacterial vectors for the expression of fusions to LYTAG2 in E. coli.

Figure 3. Comparative expression of a LYTAG2-LacZ fusion against a C-LytA-LacZ fusion in E. coli, under the control of the salicylate-inducible nahR / Psal xylS2 / Pm system, in the presence of different concentrations of said inducer in The culture medium.

Figure 4. Eukaryotic vector for the expression of fusions to LYTAG2 in S. cerevisiae.

Figure 5. Expression and purification of a fusion to LYTAG2-LacZ in S. cerevisiae. 1, molecular weight marker (BioRad); 2, total soluble protein from the induced culture; 3, protein not retained in the resin; 4, washed column protein; 5, elution fraction with choline.

Figure 6. Circular dichroism spectra at 20 ⁰ C in the far ultraviolet of LYTAG2 protein in 20 mM phosphate buffer, pH 7.0 (solid line) and in the same buffer with 2.0 mM SDS (dotted line).

Figure 7. Purification performance of LYTAG2 (dark circuits) and LYTAG2t (light circuits) in DEAE-cellulose.

Claims

1. Isolated nucleotide sequence, derived from SEQl ID N ⁰ 1, comprising a sequence encoding a polypeptide derived from the choline affinity domain of the S. pneumoniae LytA protein, characterized in that:

to. It has an affinity for solid supports with tertiary or quaternary amines on its surface. b. the interaction with the supports can be interrupted by the addition of choline in solution at a concentration greater than 30 mM, c. it has a maximum of 162 amino acids, and a maximum of 5 of the choline binding repeats (ChBRs) present in the polypeptide of the sequence SEQl ID N ⁰ 2, encoded in the sequence SEQl ID N ⁰ 1. d. fused to another polypeptide may allow its affinity purification to the supports described in

(a) under conditions that include (b). and. encodes a polypeptide whose levels of expression, solubility and / or stability, and affinity for the supports described in (a) under extreme conditions are superior to those of the complete sequence SEQ ID N ⁰ 2

2. Nucleotide sequence according to claim 1 characterized in that, fused to the coding sequence of another polypeptide, it allows the expression of it as a hybrid polypeptide, at a level higher than that of the hybrid polypeptide resulting from an equivalent fusion with the coding sequence of SEQ ID N ⁰ 2 complete.

3. Nucleotide sequence according to claims 1 and 2 characterized in that it contains a region, added to one of its ends, and which codes for a polypeptide sequence with helix structure

4. Nucleotide sequence according to claims 1 to 3 characterized in that it comprises a sequence coding for the sequence polypeptide SEQ ID N ⁰ 3

5. Mutants of the nucleotide sequence encoding the polypeptide SEQ ID N ⁰ 44, and characterized in that:

to. the properties of the SEQ ID N ⁰ 44 polypeptide described in claim 1, at elevated temperatures, and / or b improve. the properties of the SEQ ID N ⁰ 44 polypeptide described in claim 1 improve at extreme pH values.

6. Nucleotide sequence of claim 5 with mutations determining one, some or all of the following substitutions with respect to the sequence SEQ ID N ⁰ 44: Asn51 to Lys, Ser52 to Lys, Thr85 to Lys and ThrlOδ to Lys.

7. Nucleotide sequence of claim 6 characterized in that it comprises a sequence coding for the sequence polypeptide SEQ ID N ⁰ 51.

8. Nucleotide sequence derived from that of claim 4 characterized in that it comprises a sequence coding for the sequence polypeptide SEQ ID N ⁰ 4.

9. Nucleotide sequence of claims 1 to 3 which is at least 50% identical to the sequences encoding the polypeptides SEQ ID N ⁰ 3 OR SEQ ID N ⁰ 4.

10. Nucleotide sequence derived from that of claim 4, characterized in that it incorporates a coding region for an amino acid sequence recognizable by a specific cutting endopeptidase, and which is located at one end, corresponding to the coding sequence of the polypeptide with helix structure

11. DNA vectors containing one of the sequences of claims 1 to 10.

12. Vectors according to claim 11, characterized in that the sequence of claims 1 to 10 is located under the control of a transcription promoter, and adjacent to a region with targets for restriction enzymes, or a specific recombination site , which allows the translational fusion of said sequence to that of the polypeptide to be immobilized or purified, to its amino or carboxyl ends.

13. Vectors according to claim 12, characterized in that the promoter is active in cells of prokaryotic or eukaryotic organisms such as yeasts, filamentous fungi, plants, or animals.

14. Vectors according to claim 13, characterized in that the promoter is the Pm promoter of Pseudomonas putida, or derivatives thereof.

15. Vectors according to claim 14, characterized in that the promoter is derived from the promoter region located between the GALI and GALIO genes of S. cerevisiae.

16. Transformed cells containing vectors according to claims 11 to 15, in episomic form or integrated into genomic DNA of said cells.

17. Method of immobilization of polypeptides by fusion to a polypeptide sequence according to claims 1 to 10 comprising:

to. the use of a solid substrate with the surface covered by tertiary or quaternary amines to which the polypeptide of claims 1 to 10 binds affinity. b. a solution where the above polypeptide fusion polypeptide is solubilized, which can have a concentration of 0 mM to 1500 mM NaCl, up to 3 mM choline, a detergent such as SDS (up to 2 mM) or Triton X-IOO (up to 1.5 %) and a pH of 3 to 9. c. a wash solution with a concentration of NaCl equal to or more concentrated than in (b) and which can be applied to the resin with the fusion polypeptide at a temperature of 4 ⁰ C to 95 ⁰ C. d. a solution to eventually elute choline containing fusion protein at a concentration greater than 30 mM.

18. Method according to claims 1 and 17 wherein the tertiary or quaternary amine is diethylaminoethanol (DEAE), trimethylammonium (QAE), choline, or any similar compound.

19. Method according to claims 1, 17 and 18 wherein the substrate is a polymer of the type of agarose, dextran, cellulose, silicones, methacrylate, polystyrenes, or other types of polymers which may have on its surface tertiary or quaternary amines from which they join the nucleotide sequence of claims 1 to 10.

20. Method of over-expression and purification of polypeptides using the vectors of claims 11 to 15, the heterologous sequence fused to the sequence of the type of claims 1 to 10, and the method of immobilization of fusion proteins of claims 17 to 19.

21. Use of the sequence of claims 1 to 10 and the method of claims 17 to 20 for the immobilization and purification of polypeptides for therapeutic, diagnostic and any type of industry where recombinant proteins are needed.