NO326308B1 - Peptide fragments and their use, as well as fusion protein, nucleic acid fragment, vector and process for producing, purifying and detecting proteins. - Google Patents

Abstract

The invention relates to new peptide fragments, fusion proteins containing said peptide fragments, methods for producing same and the use of the peptide fragments. The invention also relates to a method for the purification of fusion proteins and a method for detecting proteins. The invention further relates to nucleic acids coding for the peptide fragments or the fusion proteins and to vectors containing said nucleic acids.

Description

The present invention relates to new peptide fragments, fusion proteins comprising the peptide fragments, methods for their production and use of the peptide fragments. The invention also relates to a method for purifying fusion proteins and a method for detecting proteins.

The invention further relates to nucleic acids which code for peptide fragments or for fusion proteins and vectors which comprise the nucleic acids.

It has become possible through modern molecular biology to produce almost any proteins, peptides or derivatives thereof in almost unlimited quantities. In this context, purification of proteins has often been shown to be limiting and most often ineffective and is thus generally the cost-determining factor.

This is why a number of techniques for purifying proteins have been developed. The techniques normally used to purify proteins from cell supernatants, crude cell extracts or cells are, for example, salt precipitation or precipitation with organic solvents, ultrafiltration, dialysis, gel electrophoresis, isoelectric focusing, chromatofocusing, ion exchange or gel chromatography, hydrophobic chromatography, immunoprecipitation or IMAC (= Immobilized Metal Affinity) Chromatography).

Simple methods can be used for purification, with a combination of different techniques usually required. Conventional purification methods are described, for example, in the textbooks Protein Purification Process Engineering (Ed. R.G. Harrison, 1994, Marcel Dekker, Inc., New York, page 209 - 316, ISBN 0-8247-9009-X), Protein Purification (Ed. R.K. Scopes, 1994, Springer Verlag New York, chapter 4 - 7, ISBN 0-387-94072-3) and Methods for Protein Analysis (Ed. R.A. Copelog, 1994 Chapman & Hall, page 59-112, ISBN 0-412-03741 -6). It is important for the purification of the proteins that the purification takes place under conditions that are as weak as possible, and that are as selective as possible and as fast as possible. Furthermore, protein losses should be kept as small as possible. Many of the methods for protein purification are not sufficiently selective and are only suitable for small amounts of protein and/or very expensive.

Methods for purification of proteins by IMAC (= Immobilized Metal Affinity Chromatography) described by Porath et al. (Nature, Vol.258, 1975: 598 - 599) is a compromise between specified requirements that must be met by optimal purification. However, this method still has some disadvantages. Not all metal ions bind equally well to the carrier material so that some ions are washed out and thus contaminate the required product. Many proteins do not bind at all to the chromatography material and therefore cannot be purified, or bind too weakly to even be eluted during the necessary steps for washing column materials. This results in unwanted loss of product. Because the selectivity is usually insufficient for a one-step purification, in contrast to purifications by biospecific affinity purification methods such as, for example, purification with antibodies, additional purification steps are necessary.

In order to purify a number of proteins using this method, various so-called tags have been developed, such as polyhistidine, His-Trp, His-Tyr or (His-Asp)n (see Sporeno et al., J. Biol. Chem. 269 (15), 1994:10991 -10995, Le Grice et al., Eur. J. Biochem., 187 (2), 1990: 307 - 314, Reece et al., Gene, 126 (1), 1993:105 -107, De Vos et al., Nucl. Acid. Res., 22 (7), 1994:1161 -1166, Feng et al., J. Biol. Chem. 269 (3), 1994: 2342 - 2348, Hochuli et al., Biotechnology, 1988:1321-1325, Patwardhan et al., J. Chromatography A, 787, 1997: 91-100, Hutchens et al., J. Chromatography, 604, 1992:133-141). These tags are linked to the protein to be purified using molecular biology at the nucleic acid level. It has been possible through these tags to further improve protein purification in some areas, although this method still has some disadvantages. It is still not possible to predict the reliability of whether a protein can be purified according to this method (see Immobilized Metal Affinity Chromatography, L. Kågedal, pages 227 - 251 in Protein Purification, Eds. J.C. Janson, L. Rydén, 1989 , VCH Publishers, Inc., New York, ISBN 0-89573-122-3), which means that this method cannot be applied even to every protein. The metal ion may be washed out or proteins may be bound so weakly that they are partially lost during the washing steps. In addition, the selectivity is still unsatisfactory. In addition, the capacity of the column material for loading with the proteins to be purified is in some cases too low so that a large amount of column material must be used in a purification. The protein yield is still unsatisfactory. This leads to unnecessary costs.

EP 282042 describes recombinant fusion proteins with peptides containing His and Asp residues which have a high affinity for immobilized metal ions.

J. Volz et al., Journal of Chromatography A, vol. 800, 1998, pp. 29-37 describes polypeptides containing specific peptides of His and Cys residues which have a high affinity for Ni and Cu ions.

Volz et al. have described the purification of ATPases from Helicobacter pylori without the use of an additional His-tag sequence. These ATPases contain natural metal binding sites that make purification by IMAC possible. Our own studies have shown that these binding studies show a binding affinity that is too low for effective purification of all desired proteins.

It is an aim according to the present invention to provide additional tags for protein purification by IMAC which do not have the above-mentioned disadvantages and which consequently make it possible, for example, to use the tags to a more extensive extent and/or to apply the column material with a higher density, and which exhibit a higher selectivity and which thus simplifies cleaning. We have discovered that that purpose is achieved by peptide fragment characterized by having the general sequence His-X<1->His-X<2->X<3->X<4->Cys-X<5-> X<6->Cys where the variables X<1> to X<6> in the sequence have the following meanings independently of each other: X<1> = Asn;

X<2>=Gln, Glu or Arg;

X3 = Gly, Thr or Tyr;

X<4>= Asn or Arg;

X<5>=Gly or Light;

X<6>=Cys.

The general sequence His-X<1->His-X2-X<3->X<4->Cys-X<5->X<6->Cys corresponds to SEQ ID No:1 wherein X<1> corresponds with the amino acids designated Xaa in position in SEQ ID No: 1, and X<2> corresponds to Xaa in position 4, X<3> corresponds to Xaa in position 5, X<4> corresponds to Xaa in position 6, X<5 > corresponds to Xaa in position 8 and X<6> corresponds to Xaa in position 9. The amino acids mentioned above for X<1> to X<6> may represent corresponding amino acids designated Xaa in SEQ ID No: 1.

Particularly preferred peptide fragments are fragments with the sequence

Specified sequences correspond in each case to the sequences SEQ ID NO: 2, SEQ ID NO: 3, SEQ ID NO: 4, and SEQ ID NO: 5.

The present invention further relates to fusion protein, characterized in that it comprises a protein fragment according to claim 1.

A nucleic acid fragment is further described, characterized in that it codes for a protein fragment according to claim 1.

The nucleic acid according to the invention comprises a nucleic acid fragment according to claim 4.

According to the invention, the nucleic acid codes for a fusion protein according to claim 3. The invention further relates to vector, characterized in that it comprises a nucleic acid fragment according to claim 4 or 6.

The above-mentioned protein fragment sequence is encoded by the nucleic acid fragments according to the invention. The degenerate genetic code must be taken into account in this context. The nucleic acid fragments according to the invention can in principle be present in any suitable nucleic acid. Nucleic acid fragments are preferably inserted into vectors in such a way that it is possible to form composite nucleic acid sequences (= gene constructs) which code for the fusion proteins according to the invention. For expression, these gene constructs can advantageously be housed in a suitable host organism which makes optimal expression of the fusion protein possible. Suitable vectors are well known to those skilled in the art and can be found, for example, in the book Cloning Vectors (Eds. Pouwels P. H. et al. Elsevier, Amsterdam-New York-Oxford, 1985, ISBN 0 444 904018). Apart from plasmids, vectors mean all other vectors known to those skilled in the art, such as, for example, phages, vims, transposons, IS elements, plasmids, cosmids, linear or circular DNA. These vectors can undergo autonomous replication or chromosomal replication in the host organisms.

The nucleic acid sequences according to the invention mean sequences that have been functionally linked to one or more regulatory signals, preferably to increase gene expression. These can be 3' and/or 5' terminal regulatory sequences to enhance expression and are chosen for optimal expression depending on the chosen host organism and gene. These regulatory sequences are, for example, sequences to which inducers or repressors are bound and which thus regulate the expression of the nucleic acids. The gene construct can also preferably contain one or more so-called enhancer sequences functionally linked to the promoter and these make increased expression of the nucleic acid sequence possible. This can, for example, take place through an approved interaction between RNA polymerase and DNA. It is also possible to insert further advantageous sequences at the 3' end of the DNA sequences such as other regulatory elements or terminators. Nucleic acid fragments from the invention are preferably inserted into the vector in such a way that they form the N-terminal region of the fusion protein. However, it can also be located at the C-terminal end, or be located within the protein, but in this case the function of the protein must not be affected, and cleavage from the fusion protein is no longer possible.

Regulatory sequences are intended to make specific expression of the genes and protein expression possible. This can mean, for example, depending on the host organism, that the gene is expressed or overexpressed only after induction, or that it is immediately expressed and/or overexpressed.

Advantageous regulatory sequences are for example present in promoters such as cos, tac, trp, tet, trp-tet, Ipp, lac, Ipp-lac, lacl<q*> T7, T5, T3, gal, tre, ara, SP6, I-Pr promoter or I-Pl promoter, which is preferably used in gram-negative bacteria. Additional advantageous regulatory sequences are present, for example, in the gram-positive promoters amy and SP02, in the yeast or fungal promoters ADC1, MFa, AC, P-60, CYC1, GAPDH, TEF, rp28, ADH or in the plant promoters CaMV/35S, SSU, OCS, Iib4, usp, STLS1, B33, nos or in the ubiquitin or phaseolin promoter. Also advantageous in this context are promoters of pyruvate decarboxylase and methanol oxidase from, for example, Hansenula. It is also possible to use artificial promoters for regulation.

Regulatory sequences or factors can furthermore preferably have a beneficial effect on the expression of the introduced gene and thereby increase it. Reinforcement of regulatory elements can advantageously take place at the transcriptional level by using strong transcriptional signals such as the aforementioned promoters and/or enhancers. On the other hand, strengthening of translation is also possible by, for example, improving the stability of mRNA.

Nucleic acid sequences according to the invention also preferably contain signals which make it possible for the proteins to be secreted into the medium or into the cell compartments. Examples of sequences of this type that can be mentioned are typical signal sequences such as, for example, the signal sequence of ompA (E. coli membrane protein). It is additionally possible that other advantageous sequences are present, such as the sequence of the a factor or of YAC (= yeast artificial) [sic] chromosomes.

Gene constructs (=nucleic acid sequences) according to the invention additionally preferably contain sequences which make it possible to eliminate protein fragments having the above-mentioned sequence according to the invention from the N or C terminus of the fusion protein, preferably from the N-terminal end. These sequences code for example for cleavage sites for a number of proteases, for example for factor Xa, enterokinase, human renin, carboxypeptidase A, thrombin, trypsin, dipeptidyl peptidases, papain, plasmin, pepsin or other proteases. Preferred cleavage sites are, for example, Xa, human renin, dipeptidyl peptidases, carboxypeptidase A or enterokinase, because these enzymes have a high specificity and thus unwanted cleavage of the protein to be purified can be avoided. If other proteases are used, care must be taken so that none of the cleavage sites are inside the protein to be purified. The protein fragment can also be deleted by cleavage with cyanogen bromide [for example 2-(2-nitrophenylsulfenyl)-3-bromo-3'-methylindolinium, hydroxylamine etc] in the presence of formic acid. In this case, however, refolding of the protein is usually necessary and this results in this method being less preferred. Elimination of the protein fragment by exoprotease cleavage (under kinetic control) is also possible. However, this usually results in product mixes. Cleavage sites that are preferably used are those that make it possible to delete protein fragments without leaving residues of protein fragments in the protein to be purified. If the protein fragments according to the invention can be tolerated in the fusion protein without loss of function and without other disadvantages, a specific site for solution of the protein fragment can be avoided.

Suitable vectors are in principle all vectors that make expression in pro- or eukaryotic cells possible. In this context, it is possible to use vectors that replicate in only one genus or those that replicate in several genera (referred to as shuttle vectors). Examples of advantageous vectors are plasmids such as the E.coli plasmids pEGFP, pLG338, pACYC184, pBR322, pUC18, pUC19, pKC30, pRep4, pHS1, pHS2, pPLc236, pMBL24, pl_G200, pUR290, plN-lll<113->B1, Igt11 or pBdCI, preferably pEGFP, in Streptomyces plJ101, plJ364, plJ702 or plJ361, in Bacillus pUB110, pC194 or pBD214, in Corynebacterium pSA77 or pAJ667, in fungi pALS1, pll_2 or pBB116, in yeast 2um, pAG-1, YEp6, YEp13 or pEMBLYe23 or in plants pLGV23, pGHlac"1", pBIN19, pAC2004 or pDH51. Said plasmids represent a small selection of possible plasmids. Additional plasmids are well known to those skilled in the art and can be found, for example, in the above-mentioned book Cloning Vectors (Eds. Pouwels P. H. et al. Elsevier, Amsterdam-New York-Oxford, 1985, ISBN 0 444 904018).

In another embodiment of the vector, the nucleic acid sequence according to the invention can also advantageously be introduced in the form of a linear DNA into the microorganisms and be integrated by heterologous or homologous recombination into the genome of the host organism. This linear DNA can consist of a linearized vector such as a plasmid or only of the nucleic acid, i.e. the nucleic acid fragment and the gene for the protein (= fusion protein gene) and, when appropriate, other regulatory sequences.

Host organisms suitable for gene engineering are in principle all prokaryotic or eukaryotic organisms. Host organisms which are preferably used are microorganisms such as gram positive or gram negative bacteria, archaebacteria, fungi, yeast, animal or plant cells such as Drosophila, especially D. melanogaster, mice, zebra fish or tobacco. What is preferably used are gram positive or gram negative bacteria, fungi or yeast, particularly preferred is the genus Escherichia, Bacillus, Streptomyces, Aspergillus or Saccharomyces, very particularly preferred is E. coli.

Particularly preferred are the following combinations of vector and host organisms such as Escherichia coli and its plasmids and phage and promoters belonging thereto, and Bacillus and plasmids and promoters thereof, Streptomyces and plasmids and promoters thereof, Aspergillus and plasmids and promoters thereof or Saccharomyces and plasmids and promote thereof.

The present invention further relates to the method for producing fusion proteins in claim 3, characterized in that it comprises the fusion of a nucleic acid fragment as claimed in claim 6 to a gene that codes for a protein.

It further describes a method for purifying fusion proteins according to claim 3, characterized in that it comprises

a) bringing liquids containing the fusion protein into contact with immobilized metal ions in such a way that an affinity bond can be formed between

the metal ions of the fusion protein,

b) removal of unbound compounds present in the liquid,

c) elution of bound fusion protein in which the affinity binding is avoided by changing the liquid medium and

d) collection of purified fusion protein.

The invention also relates to the use of a protein fragment according to claim 1 or

a nucleic acid fragment according to claim 4 for the purification of proteins.

The present invention further relates to a method for the production of protein fragments which have the ability to enter into a reversible affinity bond with immobilized metal ions, characterized in that it comprises carrying out the following steps:

a) preparation of a nucleic acid library starting from any suitable nucleic acid sequence encoding a protein fragment having the sequence

where the histidine and cysteine residues of the sequence are conserved in the nucleic acid library.

b) fusing the nucleic acids from the library to a receptor gene which makes it possible to detect the fusion protein encoded by the resulting nucleic acid via

its binding to immobilized metal ions and

c) selection of nucleic acid sequences that exhibit a reversible binding to immobilized metalloins that is at least 1.5 times stronger than the sequence in natural

Helicobacter pylori ATPase-439.

Methods for detecting proteins are also described,

characterized in that it includes the detection of individual proteins that comprise a protein fragment as required in claim 1 in a protein mixture via antibodies that are directed against the protein fragment.

The invention also describes the use of a protein fragment according to claim 1 or a nucleic acid fragment according to claim 4 for the purification of proteins.

Fusion proteins according to the invention can be produced as described above in a method in which nucleic acid fragments according to the invention, which code for protein fragments having the above-mentioned sequence, are fused to a gene which codes for the proteins to be purified, and, where appropriate, other advantageous sequences such as promoter and/or enhancer sequences, cleavage sites for proteases, etc. If necessary for this purpose, a suitable restriction enzyme cleavage site is introduced between the nucleic acid fragment and the gene of the protein to be purified, and this construct is inserted via suitable cleavage sites into a vector . Methods of this type are known to those skilled in the art and can be found, for example, in books by Sambrook, J. et al. (1989) Molecular donation: A laboratory manual, Cold Spring Harbor Laboratory Press, by F.M. Ausubel et al. (1994) Current protocols in molecular biology, John Wiley and Sons or D.M. Glover et al., DNA Cloning Vol. 1, (1995), IRL Press (ISBN 019-963476-9). Additional advantageous vectors are the Pichia pastoris vectors pPic and pGap. This yeast organism is also a suitable host organism for protein expression.

Protein fragments according to the invention are suitable for the production of fusion proteins which can be easily purified, cheaply and efficiently using protein fragments. Protein fragments and fusion proteins according to the invention can thus be successfully purified, very selectively and in high yields and with high purity. Protein fragments according to the invention and thus fusion proteins produced from them are preferably distinguished by binding to immobilized metal ions at least 1.5 times stronger than the Helicobacter pylori ATPase-439 sequence.

All proteins are suitable in principle for the production of fusion proteins. Proteins that are preferably used are those that have a biological effect in humans, animals or plants or those that are of interest for organic synthesis. Examples are proteins such as enzymes, hormones or storage or binding or transport proteins. Examples that can be mentioned are proteins such as hydrolases such as lipases, esterases, amidases, nitrilases, proteases, mediators such as cytokines such as lymphokines such as MIF, MAF, TNF, interleukins such as interleukin 1, interferons such as y-interferon, tPA, hormones such as proteohormones, glycohormones, oligo- or polypeptide hormones such as vasopressin [sic], endorphins, endostatin, angiostatin, growth factors, erythropoietin, transcription factors, integrins such as GPIIb/llla or abblll, receptors such as various glutamate receptors, angiogenesis factors such such as angiotensin.

The method according to the invention for purifying fusion proteins makes it possible, for example, to purify proteins from natural sources such as plant or animal extracts, plant or animal cell lysates, from the culture medium, fermentation power or from synthesis power, to name just a few as examples.

The fusion protein is suitably expressed in a suitable host organism (see above) prior to purification to increase the yield of the fusion protein. The host organism is grown in a suitable synthetic or complex medium containing a carbon source, a nitrogen source, and, where appropriate, inorganic salts, vitamins and trace elements, at a suitable temperature and with suitable aeration.

Depending on whether or not the fusion protein is secreted from the cells, the cells are first broken up and the cells or cell detritus advantageously removed. Methods used for cell disruption are those known to those skilled in the art, such as ultrasound, French press, enzyme cleavage, osmotic shock and many others. Cells or cell detritus can be removed, for example, by centrifugation or filtration. Removal of the cells or cell detritus is absolutely necessary.

The liquid containing the fusion proteins is then brought into contact with immobilized metal ions so that an affinity bond between the fusion protein and the metal ions can be formed. The binding takes place at pH values higher than 7, preferably for example at pH 7.0 to 9.0, preferably between pH 7.5 to 8.0. Advantageous buffers are single buffers or buffer blogs such as, for example, 50 to 1000 mM buffers such as 50 mM Tris/HCl pH 8.0 + 150 mM NaCl, 100 mM sodium acetate pH 7.7 + 500 mM NaCl, 20 mM sodium phosphate pH 7.7 + 500 mM NaCl or 50 mM Tris/HCl pH 8.0 + 150 mM NH 4 Cl. These buffers make it possible to apply the fusion proteins to immobilized metal ions. In the simplest and particularly advantageous case, the fusion proteins are brought into contact with immobilized metal ions directly in the buffer used for disruption or in the incubation medium. It is advantageous that the liquids and the immobilized metal ions are brought into contact with each other on a conventional chromatography column. This facilitates the removal of unbound compounds, for example proteins, by washing the column with a suitable buffer. Suitable buffers are buffers which do not reduce binding of the protein fragments according to the invention or of the fusion proteins in the metallinones, or which have the ability to remove impurities. Buffers of this type preferably have a pH of over 7, preferably a pH of between 7.0 to 9.0, preferably between pH 7.5 to 8.0. It is also possible to purify host blogs, in which immobilized metal ions are placed in a container and then the liquids are added or vice versa in this way. Said buffers can be used in these host blogs. Blottings are preferably centrifuged or filtered between the individual washing steps.

Carrier materials suitable in principle for immobilizing the metal ions are all conventional and can be easily derivatized, show no or only some non-specific adsorption, show good physical, mechanical and chemical stability, and have a high external and internal surface area. Suitable materials can be obtained commercially, for example from Pharmacia LKB, Sweden (Sepharose™6B or Superose™), Pierce, USA (immobilized iminodiacetic acid I or II, immobilized tris(carboxymethyl)-ethylenediamine), Sigma, USA (immobilized iminodiacetic acid-agarose ), Boehringer Mannheim, Tysklog (zinc chelate-agarose) or Toyo Soda, Japan (TSKgel Chelate-5PW). EP-B-0 253 303 describes further suitable materials. Further suitable and advantageous materials are materials such as Ni-coated microtiter plates (nickel-chelate-coated Flashplate®, NEN life science products) or magnetic particles or specific metal ion-coated and binding membranes.

Different metal ions are bound in suitable materials preferably via groups such as IDA (= iminodiacetic acid), NTA (= nitrilotriacetic acid) or TED (= tris(carboxy-methyl)-ethylenediamine). Suitable metal ions are Co, Cu, Fe, Ca, Mg, Ni, Al, Cd, Hg or Zn, preferably Fe, Ni or Cu, particularly preferred is Ni or Cu, very particularly preferred is Ni. Loading of the materials with metal ions preferably takes place with 0.1 to 0.4 M solutions of the metal salts in aqueous, unbuffered solution.

After washing, the fusion protein is eluted with a suitable buffer. This buffer avoids affinity binding between the fusion protein and immobilized metal ions. The fusion proteins can be eluted via a pH gradient (low pH values < pH 7.0 elute), competitive ligands such as imidazole, organic solvents such as acetone or ethanol, chelating agents such as EDTA or NTA and/or detergents such as Tween 80. Elution via competitive ligands such as imidazole and/or detergents are preferred. Imidazole is used for elution in a range from 0.05 to 0.7 M, preferably from 0.1 to 0.5 M. Competing ligands and/or detergents are preferably used in a buffer, but can also be used in water. Advantageous buffers are buffers that correspond to the buffers used to be loaded on immobilized metal ions. This has the advantage that no unwanted interactions between the column material, bound proteins and buffer occur. Advantageous buffers preferably have a pH higher than pH 7, preferably a pH between 7.0 to 9.1, preferably between pH 7.5 to 8.0. These buffers are preferably applied via an increasing gradient. In the case where a pH gradient is used for elution, it is possible to use buffers with a pH below pH 7.0 and/or acids. Eluted fusion protein is collected and can be used immediately or else, if necessary and if desired, processed further. Suitable application and elution buffers can be found, for example, in the book Protein Purification (Eds. J.C. Janson, L. Rydén, VCH Publisher Inc., 1989, pp. 227 to 251).

Protein fragments according to the invention can be deleted using the methods described above, such as cyanogen bromide or protease cleavage. In this case, it is possible that the residues of the protein fragment remain in the molecule or are completely eliminated from the protein to be purified. The protein fragment is preferably completely removed from the protein.

The nucleic acid library based on the above sequence can be constructed according to methods of mutagenesis known to those skilled in the art. For this purpose, the sequence can be subjected to, for example, a site-directed mutagenesis as described in D.M. Glover et al., DNA Cloning Vol.1, (1995), IRL Press (ISBN 019-963476-9), ch. 6, pp. 193 et seq.

Spee et al. (Nucleic Acids Research, Vol. 21, No. 3, 1993: 777 - 778) describes a PCR method using dITP for random mutagenesis.

Application of an in vitro recombination technique for molecular development is described by Stemmer (Proe. Nati. Acad. Sei. USA, Vol. 91, 1994:10747-10751).

Moore et al. (Nature Biotechnology Vol. 14, 1996:458 - 467) describes a combination of the PCR method and the recombination method.

Application of an in vitro recombination technique for molecular development is described by Stemmer (Proe. Nati. Acad. Sei. USA, Vol. 91, 1994:10747-10751). It is also possible to use a combination of the two methods.

The use of mutant strains with defects in the DNA repair system is described by Bornscheuer et al. (Strategies, 11,1998:16-17). Rellos et al. describes a PCR method using non-equimolar amounts of nucleotides (Protein Expression and Purification, 5, 1994: 270 - 277).

The nucleic acid library can advantageously be produced according to a PCR technique using two complementary, degenerate oligonucleotides (called wobble primers), as described in the examples. It is important that the histidine and cysteine residues present in this sequence are conserved.

For sequencing for protein fragments that have improved metal-binding affinity, the nucleic acid fragment according to the invention which codes for the protein fragment is fused to a reporter gene. Advantageous reporter genes make it easy to detect the binding to immobilized metal ions via, for example, the binding of antibodies that are labeled with a fluorescent dye and are directed against the reporter gene, or via self-fluorescent proteins such as the advantageous and preferred egf protein from E.coli (= green fluorescence protein, see Prasher et al., Gene 111 (2), 1992: 229 - 233) or preferably bioluminescence protein from Aequoria Victoria or via light-generating proteins such as the luciferin/luciferase system. Particularly preferably, a gfp protein mutant (= egfp = enhanced green fluorescence protein) is used with a 35-fold higher fluorescence activity caused by two point mutations in position 64, replacement of Phe by Leu, and position 65, replacement of Ser by Thr . This protein mutant has the advantage over the wild-type protein that it is soluble and does not form any inclusion bodies. Application of the egfo protein makes it possible to localize and quantify the protein concentration in each phase of the purification of the proteins without interfering with the purification and without using other cofactors or substrates (Poppenborg et al.,

J. Biotechnol., 58 (2), 1997, 77-88). The egfp protein is also distinguished by a high stability against a wide pH range (pH 5.5 to 12), bleaching with photooxidation, oxidizing and weak reducing agents such as 2% mercaptoethanol. The protein shows a decrease in fluorescence above 37°C. Similarly, gfp-uv (blue fluorescence) and eyfp (yellow fluorescence) proteins are also suitable as reporter genes.

Suitable sequences are selected by comparison with the binding affinity of immobilized metal ions from the following natural Helicobacter pylori ATPase-439 sequence His-lle-His-Asn-Leu-Asp-Cys-Pro-Asp-Cys. Protein fragment sequences according to the invention show a reversible binding to immobilized metal ions which is at least 1.5 times stronger, preferably at least 2 times and especially preferably at least 3 times stronger. Advantageous sequences make it possible for the protein yield after purification to be at least 20%, preferably at least 30%, particularly preferably at least 40% and very preferably at least 50%.

The method according to the invention for screening the nucleic acid library is particularly advantageous for automation. This method can easily be used for testing a large number of nucleic acid fragments and protein fragments for their metal ion-binding affinity in so-called high-throughput screening.

Proteins can be easily detected using the protein fragments according to the invention. In the method according to the invention for detecting proteins, individual proteins which comprise a protein fragment which has the above-mentioned protein fragments according to the invention in a protein blogging are detected via antibodies which are directed against the protein skeleton. Detection of these fusion proteins preferably takes place via mono- or polyclonal antibodies directed against the protein fragment (= tag). The protein blog can preferably be fractionated by chromatography or electrophoresis before detection and then transferred (= blotted) to a suitable membrane (for example PVDF or nitro-cellulose) by conventional methods (see Sambrook et al.). This membrane is then incubated with an antibody directed against tag. It is preferred to wash the membrane several times and then to detect bound antibodies via a specific reaction with a second antibody that is, for example, enzyme-conjugated (for example, alkaline phosphatase, peroxidase, etc.) and is directed to a constant region of the first, in a Western blot or immunoblot. Corresponding antibodies are commercially available. When magnetic particles are used, washing can be omitted and antibody-coated magnetic particles can be purified by wiping with magnets.

The protein fragment according to the invention has the advantage over conventional His tags for protein detection that they have a strong antigenic effect and are thus more suitable for the production of antibodies against the fog.

The invention is further illustrated in the following examples.

Examples:

Chelating Sepharose Fast-Flow from Pharmacia LKB, Uppsala, Sweden, was used for testing the binding to metal chelate columns (= immobilized metal ions). Ampicillin, imidazole, EDTA and all other reagents were obtained from Fluka (Buchs, Switzerlog). DNA gel extraction kit, Midi Plasmid-Kit and Prepspin Plasmid Kit originated from Qiagen (Hilden, Tysklog), restriction enzymes, DNA-modifying enzymes, T4-DNA ligase and Taq polymerase originated from MBI Fermentas (St. Leon-Rot, Tysklog) . The Taq Dye Cycle Sequencing Kit was obtained from Applied Biosystems (Weiterstadt, Germany).

E. coli strain DH5a (F" endA1 hsdR17 [rk', mk<+>] supE44 thil XgyrA96 relA1 A (argF laczya) U169) was used in the cloning experiments. Plasmids containing the gene for the egf protein were obtained from Clonetech USA. E. coli strains were grown in Luria-Bertani medium (= LB) with 100 µg/ml Ampicillin at 37 °C to select clones for transformation with the egfp vector.The lysis buffer contained 50 mM sodium phosphate buffer pH 8.0, 300 mM NaCl, 1 mg/ml Lysozyme and 1 mM PMSF (= phenylmethane sulfonyl fluoride = specific trypsin and chymotrypsin inhibitor).

DNA methods, such as fusions, restrictions, PCR or transformations, etc. were performed as described in Sambrook, J. et al. (1989) Molecular donation: A laboratory manual, Cold Spring Harbor Laboratory Press or F.M. Ausubel et al. (1994) Current protocols in molecular biology, John Wiley and Sons. Fluorescence-labeled dideoxy-DNA sequencing method was used for the sequencing. DNA sequencing was performed using the Taq Dye Deoxy™ Cycle Sequencing Kit (Applied Biosystems) and the 373A DNA Sequencing System (Applied Biosystems) according to the manufacturer's instructions.

Example 1: Preparation of randomly mutagenized N-terminal metal-binding sites that were bound to the egfp gene and his6-egfp

For the PCR reaction, plasmid egfp and the two following complementary oligonucleotides were used.

In the case of his6-egfp the following two complementary primers were used.

Conditions used for the PCR reaction were as follows:

Mixture:

PCR program 95°C 7min

95°C 1 min

56°C 1min 30x

72°C 3 min

72°C 7 min

The PCR products were each cleaved with NcoI and Noti and ligated into the egfp vector which had been cleaved with the same enzyme, to exclude mutations in the vector (see figure 1). PCR-vector ligations were used to transform E. coli. Transformants were plated on LB agar with 100 ug/ml Ampicillin and incubated at 37°C.

Example 2: Cultivation conditions and preparation of cell lysates.

Transformed colonies showing fluorescence and some not showing fluorescence were selected and grown in 50 ml LB medium containing 100 µg ampicillin. Colonies showing fluorescence were selected for high-throughput screening and were incubated in sterile microtiter dishes containing 250 µl LB medium with 100 µg/ml Ampicillin. After incubation, the cultures were centrifuged. The pellets were resuspended in 2 ml of lysis buffer, incubated on ice for 20 min and then disrupted by ultrasound (2 times, 5 min with a Branson Sonifier 250). After centrifugation (15 min, 4°C, 20000 x g), the different egfp mutants were obtained in the supernatant. All screened clones were sequenced (see Tables I and II). Clones that did not fluoresce contained stop codons in the sequence so that no functional proteins were expressed (see Table I, A8, A13, M16a, Z4, Z11 and Z13). To quantify bound proteins, a fluorescence measurement was carried out in all experiments and correlated to a large extent with the gfp concentration (see figure 2).

Example 3: Low-throughput screening with Ni-NTA columns from Qiagen

600 µl of lysed cells was applied to a Ni-NTA column, washed twice with 600 µl of a washing solution (50 mM sodium phosphate buffer, pH 8.0, 250 mM NaCl) and then eluted with a 0.7 M imidazole solution.

Example 4: High-throughput screening with membrane felt dishes

Membrane filter microtiter dishes (MultiScreen 5 µm; supplied by Millipore, Molsheim, Tysklog) were used for high-throughput screening. 250 µl of a stirred chelate Sepharose suspension was placed in each well of the membrane filter dishes three times. After each addition, the Sepharose was centrifuged down (2 min, 23°C, 350 rpm). All further steps were performed in a Beckmann Biomek 2000 robot. Mini columns in the wells were washed twice with 250 µl of water. Sepharose was then applied with 250 µl of a metal salt solution and equilibrated three times with 200 µl buffer (50 mM sodium phosphate, pH 8.0, 250 mM NaCl). 0.3 M NiCI2, 0.3 M CuSO4 or 0.3 M ZnCI2 solutions were used in each of the different bloggings (applied with metal ions). Vague non-buffered metal-salt solutions were used. 250 µl of cell lysate supernatants were placed on each of these mini-columns. The columns were then washed twice with 250 µl of equilibration buffer. Bound proteins were eluted with 2 x 100 µl 0.5 M imidazole in equilibration buffer. Chelate-Sepharose in the filters can be regenerated with 250 µl 50 mM EDTA, 1 M NaCl in water and used in further screening experiments.

Example 5: IMAC experiments

A conventional chromatography system consisting of a glass column, two peristaltic pumps for applying the solutions, a UV detector (LKB UV-MII), a printer (LKB

RIC 102) and a fraction collector (LKB FRAC-200) were selected in the experiment. All equipment was from Pharmacia. The column was packed with chelating Sepharose Fast-Flow Gel (Pharmacia), washed with 7 bed volumes of deionized water and added metal ions with 7 bed volumes of 0.3 M NiCI2 solution. The column was then washed and equilibrated with 7 bed volumes of IMAC buffer (50 mM sodium phosphate, pH 8.0, 250 mM NaCl). 1 ml samples of the cell lysates were applied to the column at a flow rate of 1.5 ml/min and washed with 10 bed volumes of IMAC buffer. Bound proteins are eluted via an increasing gradient with 0.5% of a 0.5 M imidazole solution per ml elution solution and finally 5 bed volumes of a 5 M imidazole/water solution. The protein fractions were identified by UV detection and pooled. When complete, the column was washed and regenerated with 50 mM EDTA/1 M NaCl solution. This washing step decoupled metal, bound cell debris and proteins from the column. Eluted fractions were examined both optically and spectroscopically. Some of the examined clones showed no affinity for the matrix (see tab. I, M15, M16), while others showed good binding (see tab. I, M13, Z5 and Z7). Clones M13 and Z5 eluted from the column in a sharp band, while there was smearing of clone Z7 on the column.

Example 6: Experiment comparing egfp-type and his-tag

Clone M13 was compared with egfp wild-type protein and normal histags in a comparable experiment. The Egfp wild-type protein does not bind to metal chelate columns. The fluorescence was no longer detectable on the column after washing the column. Clone M13 binds to the column in a sharp band, but the his tag proteins bind over the entire column. This is due to the lower affinity which ultimately leads to a lower capacity in the column. The protein yield in the case of M13 is 56%, which is higher than 48% with his mists.

Clones A6, A10, M13, Z5 and Z7 bind well to the metal chelate column, while clones M14, M15 and M16 show no binding.

Example 7: Metal binding affinity

Many of the clones showed a preferred binding to Ni<2+> or Cu<2+>. As for M13, no binding to Zn<2+> was observed. When using Ni chelate columns, clone M13 clearly showed better purification of the proteins compared to the His mists. Likewise, the latter resulted in a cleaner product when compared to M13 using Cu chelate columns, but, due to the fact that the bond to the column material is very strong in both cases, the Cu ions were washed out by drastic elution conditions. This leads to contamination of the products.

Example 8: Experiment comparing ATPase-439 sequence and

protein fragments according to the invention

The ATPase-439 comparative clone was performed analogously to examples 1 and 6. The primer used was the following:

The other primers and PCR conditions were as described in example 1.

The experiment was performed with a Qiagen Ni-NTA Spin set under native conditions, the lysis was performed as described under these conditions; the columns that had already been loaded were equilibrated with 600 µl of 50 mM sodium phosphate buffer, pH 8.0, 300 mM NaCl and centrifuged at 2000 rpm (= 420 x g) for 2 min. Then 600 ul of the lysate was applied and centrifuged for 2 min. Two washes were performed with 600 µl of a 50 mM sodium phosphate buffer, pH 8.0, 300 mM NaCl each time. Then 600 µl of 50 mM sodium phosphate buffer, pH 8.0, 300 mM NaCl, 0.5 M imidazole was used in two elutions. The yield of pure protein was 1.5-fold higher with clone M13 than the metal-binding site of the ATPase-439 comparative clone. Clone M13 therefore binds better and can also be better eluted.

Claims (14)

1. Peptide fragment, characterized in that it has the general sequence His-X<1->His-X<2>-X3-X<4->Cys-X<5->X<6->Cys where the variables X< 1> to X<6> in the sequence have the following meanings independently of each other: X<1> = Asn; X<2>= Gin, Glu or Arg; X3=Gly, Thr or Tyr; X<4>= Asn or Arg; X<5>=Gly or Lys; X<6>=Cys. 2. Peptide fragment, characterized in that it has the sequence 3. Fusion protein, characterized in that it comprises a protein fragment according to claim 1. 4. Nucleic acid fragment, characterized in that it codes for a protein fragment according to claim 1. 5. Nucleic acid, characterized in that it comprises a nucleic acid fragment according to claim 4. 6. Nucleic acid, characterized in that it codes for a fusion protein according to claim 3. 7. Vector, characterized in that it comprises a nucleic acid fragment according to claim 4 or 6. 8. Process for producing fusion proteins in claim 3, characterized in that it comprises fusing a nucleic acid fragment as claimed in claim 6 to a gene that codes for a protein. 9. Method for purifying fusion proteins according to claim 3, characterized in that it comprises a) bringing liquids containing the fusion protein into contact with immobilized metal ions in such a way that an affinity bond can be formed between the metal ions of the fusion protein, b) removing unbound compounds present in the liquid, c) elution of bound fusion protein in which the affinity binding is avoided by changing the liquid medium and d) collection of purified fusion protein. 10. Use of a protein fragment according to claim 1 or a nucleic acid fragment according to claim 4 for purification of proteins. 11. Method for the production of protein fragments that have the ability to enter into a reversible affinity bond with immobilized metal ions, characterized in that it comprises carrying out the following steps: a) production of a nucleic acid library starting from any suitable nucleic acid sequence that codes for a protein fragment with the sequence where the histidine and cysteine residues of the sequence are conserved in the nucleic acid library. b) fusing the nucleic acids from the library to a receptor gene that makes it possible to detect the fusion protein encoded by the resulting nucleic acid via its binding to immobilized metal ions and c) selecting nucleic acid sequences that exhibit a reversible binding to immobilized metal ions of at least 1.5 times stronger than the sequence in native Helicobacter pylori ATPase-439. 12. Method according to claim 11, characterized in that the ategf protein from Aequoria Victoria is used as a reporter gene. 13. Method for detecting proteins, characterized in that it comprises detection of individual proteins comprising a protein fragment as required in claim 1 in a protein mixture via antibodies directed against the protein fragment. 14. Use of a protein fragment according to claim 1 or a nucleic acid fragment according to claim 4 for purification of proteins.