NZ234212A

NZ234212A - Dna and nucleotide encoding modified vp1/p2a region of the rhinovirus system; peptides encoded by these; and peptides cleavable in trans by protease 2a

Info

Publication number: NZ234212A
Application number: NZ234212A
Authority: NZ
Inventors: Wolfgang Sommergruber; Peter Volkmann; Friederike Fessl; Ingrid Maurer-Fogy; Ernst Kuechler; Dieter Blaas; Timothy Skern; Manfred Zorn; Herbert Auer
Original assignee: Boehringer Ingelheim Int
Priority date: 1989-06-25
Filing date: 1990-06-25
Publication date: 1993-03-26
Also published as: CA2019661A1; IE950434L; FI903144A0; IE902270L; IL94841A0; WO1991000348A3; AU657205B2; PT94458A; AU644078B2; EP0410147B1; IE902270A1; DK0410147T3; PT94458B; AU5779690A; IE66395B1; ATE114717T1; EP0410147A1; AU3524593A; ES2064537T3; WO1991000348A1

Abstract

The present invention relates to systems which make it possible to introduce mutations in the VP1/2A and in the complete coding region of the 2A protease, to test systems with which the "cis" and "trans" activity can be investigated, to oligonucleotides which encode the mutations, and to the oligopeptides derived therefrom and to the use thereof, furthermore oligopeptides which are cleaved by the 2A protease into trans with different efficiency, and to the use thereof.

Description

<div class="application article clearfix" id="description"> 23 4 2 1 Priority Date(s): 2. *F.'k\. Afv?P. Complete S— . I; 6 9p,, Class: u?1 iwlo^co,- mms-Jxw, A3; £fnM?l%n\ V^99; X.qi o u, . n ^ 6 MR 1993 Publication Date: P.O. Journal. No: U.ty?.- K*J Patents Form No. 5 NEW ZEALAND PATENTS ACT 1953 COMPLETE SPECIFICATION MUTATION OF THE RHINOVIRUS SYSTEM WE, BOEHRINGER INGELHEIM INTERNATIONAL GMBH, a body corporate under the laws of THE FEDERAL REPUBLIC OF GERMANY of D-6507 Ingelheim am Rhein, FEDERAL REPUBLIC OF GERMANY hereby declare the invention, for which we pray that a patent may be granted to us, and the method by which it is to be performed, to be particularly described in and by the following statement: 2342 12 - 2 - The present invention relates to systems which make it possible to insert mutations in the VP1/2A and in the overall coding region of the protease 2A, test systems by means of which the "cis" and "trans" activity can be investigated, oligonucleotides which code these mutations, as well as the oligopeptides derived therefrom and the use thereof. It further relates to oligopeptides which are cleaved from protease 2A in trans with varying efficiency, and the use thereof. Rhinoviruses are ss(+)RNA viruses and represent a genus within the picornaviridae (Cooper, P.D. et al., 1978, Intervirology 10, 165 - 180; MacNaughton, M. R., 1982, Current Top. Microbiol. Immunol. 97, 1 - 26). They are widespread, attack the upper respiratory tract in humans and give rise to acute infections which lead to catarrh, coughs, sore throats etc. and are referred to generally as colds (Stott, E. J. and Killington, R. A., 1972, Ann. Rev. Microbiol. 26., 503 - 524). Infections caused by rhinoviruses are among the commonest diseases in man. Admittedly, the illness is usually harmless but, owing to the temporary weakening of the body, secondary infections may occur, caused by other viruses or bacteria, which may under certain circumstances have serious consequences. Of the total of about 115 different serotypes of human rhinoviruses which are known, up till now 4 serotypes (HRV IB, 2, 14 and 89) have been cloned and fully sequenced: German Patent Application P 35 05 148.5; Skern; T. et al., 1985, Nucleic Acids Res. 13, 2111-2126; Diichler, M. et al., 1987, Proc. Natl. Acad. Sci. USA 84, 2605 - 2609; Stanway, G. et al., 1984, Nucleic Acids Res. 12, 7859 - 7877; Callahan, P. L. et al., 1985, Proc. Natl. Acad. Sci. USA 82, 732 - 736; Hughes, R. et al., 1988, J. Gen Virol. 69, 49 - 58). Hardly any other viral system is as dependent on controlled limited proteolysis for the regulation of the course of the infection as the Picornaviridae. The r> 2 j - 3 - genomic single-strand (+)RNA of the rhinoviruses is modified shortly after, infection by the cleaving of the oligopeptide VPg bound to the 5' end and serves as mRNA for the synthesis of a polyprotein which includes the entire continuous reading frame of the nucleic acid sequence (Butterworth, B. E., 1973, Virology 56, 439 - 453; Mc Lean, C. and Rueckert, R. R., 1973, J. Virol. 11, 341 - 344; Mc Lean, C. et al., 1976, J. Virol. 19, 903 - 914). The muture viral proteins are formed exclusively by proteolytic splitting from this polyprotein, the effective proteases themselves being part of this polyprotein. The first step in this processing is the cleaving of the precursor of the coat proteins, which is carried out by the protease P2A. In the order of the genes, the sequence of the protease P2A comes directly after the section coding for the coat proteins. P2A is thus the first detectable enzymatic function of the virus, owing to its location in the polyprotein. It cleaves itself from the precursor of the coat proteins and is responsible for the separation of the capsid precursor PI from the remainder of the polyprotein ("cis" activity). The separation of the coat protein region from the section which is responsible for replication takes place during the -translation of the polyprotein. In polio virus, this primary cleaving at the P1-P2 region may be carried out intermolecularly in vitro, i.e. "in trans" by the mature protease 2A (Krausslich, H.-G. and Wimmer, E., 1988, J Ann. Rev. Biochem., 57, 701 - 754). This step is essential for the further continuation of viral infection (compartmentalisation of replication and virus assembly). In cardio- and aphtoviruses, t?"* ft unlike the polio viruses, this cleaving is catalysed h$r"Y 21JANI993S known that in all probability all the enzymes ! .V participating in this maturation cleaving are virally coded (Toyoda, H. et al. , 1986, Cell, 4j5, 761-770). In the polio virus there are three types of cleaving signals; there are the Q-G sites, which are used most and are recognised by the viral proteases P3C, and the Y-G site which is used by P2A as a recognition signal. Initially, protease P3C was the centre of interest in the explanation of the proteolytic processing of picorna viruses. Very early on, it became possible to describe a proteolytic activity in EMC which was equivalent to P3C (Pelham, H. R. B., 1978, Eur. J. Biochem. 85, 457 - 461; Palmenberg A. C. et al., 1979, J. Virol. 32. 770 - 778). In the course of further investigations it became apparent that the leader peptide (L) of cardioviruses (e.g. EMCV) and aphtoviruses (e.g. FMDV), which is not present in rhino- and enteroviruses, is involved in the proteolytic processing of EMCV (Palmenberg, A. C., 1987, J. Cell. Biochem. 33. 1191-1198). It then became possible, by isolating polio P3C and using immunological methods, to demonstrate that P3C excises itself autocatalytically out of the polyprotein, in order to attack all potential Q-G cleaving sites in "trans". The use of recombinant systems which, inter alia. represented the P3C region, made it possible to express P3C of some entero- and rhinoviruses (Werner, G. et al., 1986, J. Virol. 57, 1084 - 1093) and to effect accurate characterisation of the P3C of polio (Hanecak, R. et al., 1984, Cell 37, 1037 - 1073; Korant, B. D. and Towatari, T., 1986, Biomed. Biochim. Acta 45. 1529 - 1535) and the equivalent proteolytic function in FMDV (Klump, W. et al., 1984, Proc. Natl. Acad. Sic. USA 81. 3351 - 3355; Burroughs, J. N. et al., 1984, J. Virol. 50, 878 - 883). Mutagenesis studies carried out in vitro have demonstrated that the replacement of the highly conserved amino acids cysteine (position number 47) and histidine (position number 161) in P3C of poliovirus results in an inactive enzyme, whereas the 23 4 2 - 5 - mutation of the unconserved cysteine (position number 153) has no appreciable influence on the proteolytic activity of polio P3C. This mutagenesis of recombinant P3C, brought about by oligonucleotides, and the additional inhibitor studies carried out lead one to conclude that polio P3C belongs to the category of cysteine proteases (Ivanoff, L. A. et al., 1986, Proc. Natl. Acad. Sci. USA 83., 5392 - 5396). It has also been shown, by "in vitro" mutagenesis of polio P3C (i.e. by exchanging the conserved valine for alanine in position 54 of the protease) , that this mutation in a "full size" cDNA of polio after transfection into COS 1 cells, results in a polymerase-deficient virus (Dewalt, P. G. and Semler, B. L. , 1987, J. Virol. 61, 2162 - 2170). Antibodies developed against polio P3C did admittedly suppress all the cleaving carried out at Q-G but did not suppress cleaving between Y-G (Hanecak, R. et al., 1982, Proc. Natl. Acad. Sci. USA 79. 3973 - 3977). This observation lead to the conclusion that the proteolytic processing at Y-G sites requires its own protease. The seat of this second proteolytic activity could clearly be ascribed to P2A in poliovirus. It was interesting to discover P2A carries out alternative cleaving in the protease-polymerase region (3CD), which also occurs at a Y-G site and yields inactive enzyme 3C and 3D. This cleaving possibly serves to regulate the quantities of the enzymes 3C and 3D (Lee, C.K. and Wimmer, E., 1988, Virol., 166. 1435 - 1441). Since the synthesis of host protein is very rapidly stopped during infection with poliovirus in HeLa cells, but the translation of the poliovirus RNA is able to proceed unhindered, it was assumed that one of more regulating factors of the translation were changed during the infection. In fact, findings show that the eukaryotic initiation factor 4F is changed by the proteolytic cleaving of the p220 component during the poliovirus infection in HeLa cells (Etchison, D. et al., 234212 - 6 - 1984, J. Virol. 51, 832 - 837; Etchison, D. et al., 1982, J. Biol. Chem. 257. 14806 - 14810). It was subsequently demonstrated that P2A is indirectly responsible for this modification of p220 in infected cells (Krausslich, H. G. et al., 1987, J. Virol. 61, 2711 - 2718; Lloyd, R.E. et al. 1986, Virol., 150. 299 - 303; Lloyd, R. E. et al., 1987, J. Virol., 61, 2450 - 2488). It has only recently been shown that p220 is also broken down in cells infected with rhinovirus (Krausslich, H.-G., unpublished data). The question of the "trans" activity of the protease P2A has been answered in the affirmative in the poliovirus system, by making insertions and deletions in P2A to express an unprocessed P1-P2 region in E. coli which was able to be cleaved from poliovirus, starting from P2A, using cells infected with poliovirus (Nicklin, M. J. H., et al., Proc. Natl. Acad. Sci. USA, 1987, 84, 4002 - 4006), the P2A having been translated "in vitro" (Krausslich, H.-G. et al. 1987, J. Virol., 61, 2711 - 2718) or expressed in E. coli. or with a purified 2A protein from poliovirus (Konig, and Rosenwirth, B., 1988, J. Virol., 62, 1243 - 1250). By large scale production of recombinant polio 3C in E. coli and purification of this protease, the specific cleaving efficiency was demonstrated using native and synthetic substrates "in vitro" (Nicklin, M.J.H. et al., 1988, J. Virol., 62, 4586 - 4593). In addition to the "cis" activity, the protease 2A was also demonstrated to have a "trans" activity in the rhinovirus system by specific cleaving of a peptide substrate which represents the native cleaving region between VP1 and 2A (Sommergruber, W., wt al., 1989, Virol, 169. 68 - 77). Furthermore, by cloning and expression of the protease 3C of HRV14 in an E. coli maxicell system it was possible to demonstrate that rhino 3C is proteolytically active in recombinant systems, a precursor of 3C "in trans" is responsible for the release of 3C and ZnCl2 specifically inhibits the processing of 3C precursor forms (Cheah, K.C., et al., 1988, Gene, 69, 265-274). It has also been possible to demonstrate the specific cleaving of a peptide substrate in E. coli extracts containing the protease 3C of HRV14 (Libby, R.T. et al., 1988, 27, 6262 - 6268). Protease 3C, like 2A, is capable of excising itself from the polyprotein autocatalytically (Hanecak, R., et al., 1984, loc. cit.). The amino terminal cleaving site is clearly preferred to the carboxy end. Furthermore, a varying affinity for other cleaving sites can be observed. 3C is sufficient for cleaving the capsid protein precursor Pi in 1ABC and ID, whilst further cleaving into 1AB and 1C requires the complex 3CD, the protease with the carboxy-terminal replicase. In poliovirus, the substrate specificity of the protease 3C is possibly modulated by the polymerase 3D by means of structural changes (Jore, J., et al., 1988, J. Gen. Virol., 69, 1627 - 1636; Ypma-Wong, M. F., et al., 1988, Virol., 166, 265 - 270). A comparison of the amino acid sequences of the individual proteins shows that the viral enzymes are conserved to a particular degree. Thus, the homology between protease P2A of HRV89 and HRV2 is 85%; in the case of protease P3C, 75% of the amino acids are identical (Diichler, M. et al., 1987, loc. cit.). These values are substantially higher than the percentages observed on average in the protein as a whole. It can therefore be assumed that precisely the viral enzymes are particularly well conserved in evolution and their properties are very similar in different rhinoviruses. It is very interesting also to discover that two proteins similar to the picornaviral proteases P3C and P2A have been found in the plant-viral system of Comoviridae (Cowpea Mosaic Virus) (Garcia, J. A. et al., 1987, Virology, 159, 67 - 75; Verver, J. et al., 1987, EMBO, 6, 549 - 554). These two viral proteins are involved in the proteolytic processing of the two polyproteins coded by two separately packaged ss (+)RNA molecules (B and M RNA), whilst great similarity can be found between the two Cowpea mosaic proteases and the picorna viruses in their sequence and cleaving specificity. This remarkable homology of non-structural proteins between picorna and comoviruses not only indicates a genetic relationship between these two families of virus but also indicates how essential viral proteolytic processing is for these two families of viruses. The third type of viral maturation cleaving, namely that of VPO (precursor protein of VP2 and VP4) has been described in the case of mengo and rhinovirus by means of X-ray structural data. This last proteolytic event in viral maturation appears to be based on an unusual autocatalytic protease type in which basic groups of the viral RNA participate in the formation of the catalytic centre, these basic groups acting as proton acceptor (Arnold, E. et al., 1987, Proc. Natl. Acad. Sci. USA, 84., 21 - 25). In the case of FMDV (Acharya, R. , et al., 1989, Nature, 337. 709 - 716) the crystalline structure certainly gives no indication of any such mechanism in aphtoviruses. The cleaving sites of the viral proteases were -determined in the poliovirus system by N-terminal sequencing of the majority of poliovirus proteins (Pallansch, M. A. et al., 1984, J. Virol., 49. 873 - 880). The position of the cutting sites of HRV2 between VP4/VP2, VP2/VP3 and VP3/VP1 were determined by N-terminal sequencing of VP2, VP3 and VP1. The cleaving signal between VP1 and P2A was determined on the one hand by C-terminal sequencing of VP1 (Kowalski, H. et '' al., 1987, J. Gen. Virol. 86, 3197 - 3200) and on the" '! other hand by N-terminal sequence analysis of P2A ^ 2 ?JANI995 (Sommergruber, W., et al., 1989, loc. cit.). V y. Furthermore, by cloning and sequencing HRV2 (Skern, et al., 1985 Nucleic Acids Res., 13., 2111-2126) by means 234 2 - 9 - of sequence comparisons with poliovirus and HRV14, the majority of cleaving sites have been derived. Thus, five different cleaving signals were found in HRV2: Q-S, Q-G, Q-N, A-G and E-S. Cysteine proteases are widespread in nature (e.g. papain, cathepsin B, H and S) and the characterisation and inhibition thereof is of major scientific and therapeutic value (for a survey see Turk, V., 198 6, Cysteine Proteinases and their Inhibitors, Walter de Gruyter; Barrett, A. J. and Salvesen, G., 1986, Proteinase Inhibitors, Elsevier). The general importance of the inhibition of virally coded proteases has been brought back into the spotlight of possible antiviral uses not least by work on the protease of the human immunodeficiency virus 1 (HIV 1). By deletions and point mutations in the protease region of this type of retrovirus, the essential role of the protease in the maturation of this type of virus has been recognised (Katoh, I., et al., 1985, Virol., 145. 280 - 292; Kohl, N. E., et al., 1988, Proc. Natl. Acad. Sci. USA, 85, 4686 - 4690; Crowford, S. and Godd, S.P., 1985, J. Virol., 53, 899 - 907). By X-ray structural analysis and molecular biological studies it has also been shown that the protease of HIV I belongs to the Asp type, is able to process itself even on the precursor protein (and in recombinant prokaryotic systems), is capable of cleaving "in trans" specific peptides and occurs as an active protease in a homodimeric form (Navia, M. A. et al., 1989, Nature, 337. 615 - 620; Meek, T.D., et al., 1989, Proc. Natl. Acad. Sci. USA, 86, 1841 - 1845; Katoh, I. et al., 1985, loc. cit.). In the picornaviral systems, all kinds of organic and inorganic compounds as well as peptide derivatives and proteins are now known which have an inhibitory effect on the proteolytic processing of these viruses. The effect of these substances is based on the direct interaction with the proteases (Kettner, C. A. et al., 1987, US Patent No.: 4,652,552; Korant, B. D. et al., 1986, J. Cell. Biochem. 32., 91-95) and/or on the indirect route of interaction with substrates of these proteases (Geist, F. C., et al., 1987, Antimicrob. Agents Chemother. 31, 622 - 624; Perrin, D. D. and Stiinzi, H., 1984, Viral Chemotherapy 1, 288 - 189). The problem with the majority of these substances is the relatively high concentration needed for inhibition and the toxicity of the compounds, which is considerable in some cases. As has already been explained, the course of the infection by picorna viruses is critically dependent on the viral enzymes. Since precisely these enzymes are particularly well conserved and are very similar in their properties in various rhinoviruses, they appear to be the obvious target of chemotherapeutic attack, such as for example the viral enzyme P2A. The chemotherapeutic approach is to inhibit the enzymatic activity by means of specific inhibitors. If the first proteolytic activity, the P2A activity, is inhibited, all further maturation of the viral system is prevented. In view of the marked homology of the P2A region of HRV2 not only to other rhinoviruses but also to representatives of other groups of the picornaviridae (unpublished data of A. Palmenberg) it is thoroughly possible that an inhibitor against HRV2-P2A can also be used on other picornaviruses. A test system for inhibitors of this first proteolytic activity has been established (Sommergruber, W. et al-, loc. cit.). By means of this system it has been demonstrated that: the polypeptide 2A of HRV2 is a protease which, when expressed as a substrate enzyme in E. coli consisting of a fusion section of MS2 polymerase (98 amino acids), the region for the viral coat 2 <5 4 2 - li - protein VP1 and the protease 2A, recognises its own N-terminus and cleaves itself from the precursor; E. coli extracts which contain the active protease 2A are capable of specifically cleaving a 16 amino acid long peptide which represents the native cleaving region between PI and P2 in a symmetrical manner; by deletion of the last 10 C-terminal amino acids of protease 2A, the proteolytic function is destroyed; by replacement of the highly conserved Arg 13 4 for Gin in the C-terminus of 2A the proteolytic function is again suppressed; deletions and mutations in the presumably active centre of 2A inhibit the proteolytic activity. Experimental data on the catalytic mechanism of poliovirus 30 showed that the protease 3C could be blocked by inhibitors for thiol proteases such as, for example, N-ethylmaleimide or iodoacetamide (Pelham, H.R.B., 1978, Eur. J. Biochem., 85, 457 - 462). In the light of comparisons of the 3C sequences of picornaviruses (Argos, P., et al., 1984, Nucleic Acids Res., 12, 7251 - 7267) and point mutation experiments (Ivanoff, L., et al., 1986, loc. cit.) a cysteine-histidine pair was defined as the presumed active centre for cysteine protease. By comparison with other picornaviruses and other inhibitor studies (Parks, G.D., et al., 1986, J. Virol., 60, 376 - 384; Konig, H and Rosenwirth, B., 1988, J. Virol., 62, 1243 - 1250) and direct mutagenesis of the presumed active centre of the protease 2A of HRV2 (Sommergruber, W., et al., 1989, Virol., 169, 68 - 77) it seems very probable that 2A also has a cysteine as a nucleophile. Replacement of the cysteine 106 of HRV2-2A for a serine in the presumed active centre of the enzyme destroys the activity (Sommergruber, W. et al., loc. 234 - 12 - cit) . Comparison of the last 50 amino acids with the active centre of 2A with the SWISSPROT sequence databank shows great similarities between the 2A protein and serine proteases and not with cysteine proteases. Cysteine obviously acts in the active centre of a serine protease. Replacement of the serine by a cysteine in subtilisin (Philipp, M., et al., 1971, Methods in Enzymology, 19, "Proteolytic Enzymes", Colowick, S.P. and Kaplan, N. 0.) to obtain thiosubtilisin by chemical modification achieves the esterase activity, whilst in the case of trypsin the directed mutagenesis of the serine into a cysteine reduces the activity by a factor of 100,000 (Higaki, J. N., et al., 1987, Cold Spring Harbor Symp. Quant. Biol., 52, 615 - 621). A similar, if less significant result is obtained by comparing the 3C protein with other proteases (Gorbalenya, A. E. et al., 1986, FEBS Lett., 194. 253-257). A common precursor for serine and cysteine proteases is even postulated, to which viral proteases are still very similar (Gorbalenya, A. E., et al., 1989, FEBS Lett., 243. 103-114). A clear distinction between the two viral proteases 2A and 3C is drawn by Bazan and Fletterick. They find strong structural homology between the 3C different picornaviruses and related plant viruses with the chymotrypsin family of the serine proteases when the crystalline structure of chymotrypsin is compared with secondary structural predictions of the viral enzymes. The ^-folding sheet configurations which decide the structure are conserved, insertions and deletions occur only in the connecting loops (Bazan, J. F. and Fletterick, R. J., 1988, Proc. Natl. Acad. Sci. USA, 85, 7872 - 7876). The protein 2A has structural similarities with the subtilisin family of the serine proteases. In the light of these comparisons, the amino acids His 18, Asp 35 and Cys 106 probably form the active centre of HRV2-2A. The proteins with, the greatest similarity to the last 50 amino acids of 2A were: Thaumatin I and II from "Thaumatococcus daniellii Benth.", the fruit of a wUijuuuLy ItiduiiuLUa WfciaL ATtiudxi L>u^>h. JJIcj-UL (VOLU dei. T7t*l , H., 1972, FEBS Lett., 21, 88 - 90? Iyengar, R. B., et al., 1979, Eur. J. Biochem., 90, 195 - 204), which is known in the literature for its extraordinary sweetening power. In the purification of a related protein, monellin, a proteolytic activity is described (Morris, J. A. and Caganr R. H., 1972, Biochem. Biophys. Acta, 261. 114-122; Van der Wei, H., 1972, Europ. J, Biochem-, 31. 221 - 225) which, may possibly be an integral part of these sweet proteins. Recently, the marked similarity of these proteins to a protease inhibitor was described (Richardson, M., et al., 1987, Nature, 327. 432 - 434), which belongs to a group of proteins which is induced if a plant is attacked by insects, microorganisms or viruses. Some of these proteins induced in plants also act as a proteinase inhibitor or protease (Cornelissen, b. j. c. et: al., 1986, Nature, 321. 531 - 532). An object of the present invention was to provide systems which maJce it possible to insert any desired point mutations or deletions in the overall coding . region, particularly the cleaving region of protease 2A. One aspect of the invention provides a DNA molecule coding for the VP1/P2A region of the Rhinovirus system, for example HRV2, and having, in the region of the protease 2A, at least one non-native unique recognition site for a restriction enzyme. Such DNA molecules can form part of plasm ids for example and be used in test systems for studying rhinoviruses. For- example, starting from the expression vector pEx2A, modified, plasmids having altered nucleotide sequences, whilst retaining their ajuino acid sequence were established. These modified expression systems, whilst retaining their native ' amino acid sequence, have unique, or singular, recognition sites (i.e. signal sequences) for restriction enzymes which 2 7 fa 0 *1 t") 3 T £. i £ - 14 - were introduced ty means of synthetic double-stranded oligonucleotides. The pEx2A/21 and pEx2A/II systems were obtained, the expression products of which behaved identically to those of pEx2A (see Fig. 5, traces 2 and 3). These new expression systems make it possible, by exploiting the newly generated restriction sites, to incorporate mutation oligonucleotides which were used on the one hand for fine analysis of the cleaving site and may be used on the other hand for deletion studies in order to determine the smallest region of HRV2-2A which is still proteolytically active. These systems, particularly the mutated systems, can also be used as test systems for investigating the intramolecular activity of the protease, the "cis" activity of the protease, the "cis" activity, since they enable direct measurement of the effects of the mutations on the "cis" activity. At the same time the effects of these mutations on the "trans* activity can also be determined. Our invention will be described with particular reference to HRV2, but is applicable to other members of the Rhinovirus system. For example, we also describe work on HRV89. The mutations in the PI position of the cleaving site of protease 2A of HRV2 (hereinafter referred to HRV2-2A) were carried out either in the expression vector system pExl8521 or pEx2A or the derivative pEx2A/21 thereof. First of all, the alanine in the PI site of the HRV2 was replaced by a tyrosine. This mutation produces a y/g amino acid pair at the :2A cleaving site, such as occurs in the polio virus (Toyoda, H. et al. r 1986, Cell, 4J5, 761 - 770). A PstX/Hindlil fragment containing the mutation (Ala Tyr) was used to replace the wild-type fragment of pExl8521. The pEx2A/21 was used in order to introduce further mutations in and around the cleaving site by means of double-stranded oligonucleotides (see Fig. 1). A double-stranded oligonucleotide represents, for example, those amino acids in positions Pi—4 and PI'—41 as were discovered in the serotype ERV89 (see Fig. 1}. 23 4 2 - 15 - Constructions which contain the amino acids glutamine, leucine, methionine, phenylalanine and tyrosine at the PI position were cleaved just as efficiently as the wild type, as is clear from the absence of bands of uncleaved material at 65Kd (or at 75Kd, in the case of the tyrosine mutation). The mutants which have valine or isoleucine at the Pi site were not fully processed, however. When the construction containing valine was used, 25% of the induced protein was not processed; when there was an isoleucine group at this point, up to 50% of the induced material was not cleaved. A significant reduction in the cleaving efficiency is also encountered when using those amino acids in positions Pl-4 and PI'-41 as were discovered in the serotype HRV89 (about 65-70% of the induced protein is not processed). To sum up, it can be said that the amino acids isoleucine and valine clearly do not fit very well into the active centre of HRV2-2A; this is all the more remarkable as valine is the very amino acid which is found in the PI site of HRV89. Possibly the methyl group of the /3-C atom interferes with the topography of the active centre or disrupts a structural change in the substrate which is necessary for proteolytic cleaving. It can therefore be assumed that there are differences in the structures of the active centres of 2A between serotypes HRV2 and HRV89. The alanine group was therefore replaced by a threonine group since threonine has a hydroxyl group at the /3-carbon atom. The expression product of the construction with Ala -*■ Thr exchange at the PI site was cleaved satisfactorily: the hydroxyl group at the f3-carbon atom is obviously, unlike the methyl group of valine and isoleucine, not big enough to influence the cleaving efficiency. In order to investigate the influence of mutations at the PI•, P2', P41 and P91 sites on the cleaving - 16 - efficiency, the oligonucleotides shown in Fig. 20 and Fig- 21 were introduced into the relevant sites. Constructions containing the amino acids trytophan, lysine, threonine and glutamic acid at the PI1 site were not cleaved, as is clear from the absence of the band at 50 kd corresponding to the cleaved material. A construction in which the glycine at the PI1 site had been deleted was not cleaved either. These results indicate that, unlike the PI site, changes at the Pi1 site result in a polypeptide which cannot be cleaved. The construction which has a deletion of the proline group at the P21 site was not cleaved. The replacement of the valine group at the P91 site by aspartic acid resulted in a 3 0% reduction in cleaving efficiency, whereas threonine did not affect the efficiency. Nor did the presence of a threonine group at the P41 site have any effect on the activity of the HRV2-2A protease. These results indicate that the negative charge of the aspartic acid at the P41 site is not essential for the protease activity but that a negative charge at the P91 site affects the proteolytic cleaving. A valine/threonine exchange at the P91 site does not influence the activity. This system was also used in order to investigate the influence of the sequence between PI' and P9' on the proteolytic activity of HRV2-2A. The natural sequences which occur in HRV14 and poliovirus were used as the starting basis (the sequences of poliovirus serotype 1, 2 and 3 (sabin strains are identical in this region). The sequences for the mutagenesis are shown in Figure 22. Very great variability can be found in this region; it was therefore necessary to investigate whether HRV2-2A is able to recognise such cleavage sites. The expressed protein of the construction with the PI1-P91 (P2 1 Leu -+ Pro) sequence of HRV14 was not processed. No cleaving of the expressed protein was observed either in the construction which has the PI1 to P91 sequence of 2342 12 - 17 - poliovirus or in a very similar construction which constitutes a variant of the PI1 to P9' polio sequence (P41 His -<■ Met, P6' Asn -► Tyr) . As was demonstrated, the expressed protein of a construction having valine at the PI site or with those amino acids which can be found between P^/P.,,^, in HRV89 are only cleaved to a level of 50% (Example 6). This finding was extended by replacing amino acids at other positions of the cleavage signal in order to investigate the influence of the individual amino acids in this region. Fig. 23 shows a comparison of the amino acid sequences of the two HRV serotypes. The cleaving of the expressed proteins clearly shows that the protease HRV2-2A cannot readily recognise the cleavage sites which contain the amino acids of HRV89. The presence of valine at the PI site is primarily responsible for this, as can be seen from the difference between the mutations HRV89 P2_5 and P.,^- For example, a 30% reduction in cleaving is seen as a result of the alanine -* valine exchange. In order to support this result, the sequence corresponding to the cleavage site region of human rhinovirus serotype IB was incorporated: Arg-Pro-Ile-Ile-Thr-Thr-Ala-Gly-Pro-Ser-Asp-Met-Tyr-Val-His-Val replaced by Arg-Ala-Ser-Met-Lys-Thr-Val-Gly-Pro-Ser-Asp-Leu-Tyr-Val-His-Val. The sequence has a valine group at the PI site; the expression product was only cleaved to a level of 65%, as can be seen in Fig. 26. The present invention provides as another aspect thereof, an oligonucleotide, coding for a modified VP1/P2A region, said modification occurring in the total coding region as well as peptides derived from such oligonucleotides. 23 4 2 - 18 - Preferred mutations are the amino acids valine or isoleucine at the PI site and Val-Thr-Asn-Val-Gly-Pro-Ser-Ser at the P^/P.,,.^ site. The peptidomimetics derived therefrom, which comprise ether bonds, for example, peptides provided with markers or D-amino acids, for example, are also a subject of the present invention. These compounds are particularly suitable as guide substances for finding the active substance in order to influence the "cis" activity of the protease 2A, and particularly as guide substances for inhibitors. In order to demonstrate that the 2A gene of HRV89 (EPA 261 403) also codes for a protease, a DNA construction was produced in which a DNA fragment corresponding to the region of the 2A gene was bound to the DNA fragment coding for HRV2-VP1. After expression of this hybrid in the pEx system, conclusions can be drawn, from the length of the fusion protein, as to whether the 2A protein of HRV89 acts as a protease. The differences in the 2A gene between HRV2 and HRV89 are shown in Figure 28. Two mutants were also constructed in order to test the influence of individual amino acids on the cleaving efficiency. Analysis of the expression products of the three constructions containing HRV89-2A is shown in Figure 31. The fusion protein of the mutant, which has the HRV2 cleavage site (pEx2A/S2/89) is cleaved almost completely; the HRV89-2A enzyme is therefore able to recognise the HRV2 cleavage site just as efficiently as the HRV2-2A enzyme. The uncleaved product migrates somewhat more slowly in the construction with the HRV89-2A, although the two 2A proteins have the same number of amino acids. Possibly, the 13 differences in the amino acid sequence affect the mobility of the 2A polypeptide. The interpretation of the results with the other two constructions having the HRV89-2A (pEx2A/S89/89 and pEx2A/S89'/89) were made more difficult by the fact that, in addition to the 65 kd and - 19 - 50 kd products expected, two other bands with molecular weights of 62 kd and 60 kd were found. A more thorough examination of the bands of the mutant pEx2A/S2/89 with the HRV2 cleavage site showed that these bands were also present in this construction. These bands were not recognised by anti-PC2 0 and therefore do not have the C-terminal end of the 2A protein; it therefore appears to be possible that the bands were caused by pausing of the ribosomes or a chain break in the region in front of the C-terminus of the polypeptide. Although there are only four difference in this region between the 2A polypeptides of the HRV2 and HRV89 from the point of view of proteins, the codon usage is very different. Analysis of the expression products of the intermediate construction pEx2A/S89"/C89 also showed a somewhat larger, uncleaved product and the presence of two additional bands, indicating that the reasons for this are connected with the region of the last 2 0 amino acids. In spite of the presence of two additional bands it is obvious that, in the expression system described here, the HRV89-2A is able to recognise the HRV2-2A cleavage site better than its own, as can be seen from the higher proportion of uncleaved product in the constructions pEx2A/S89 '/89 and pEx2A/S89/89. Obviously, the presence of alanine at the PI site is critical to the cleaving by HRV2-2A. In HRV89-2A, on the other hand, the situation is somewhat different: the HRV89-2A enzyme does not appear to recognise its own cleavage site particularly well. It is possible that in vivo parts of 2B are necessary for the activity of HRV89-2A and possibly they may influence the configuration of the 2A polypeptide. The configuration of the capsid proteins of HRV89 might also play a part in the cleaving. It would be conceivable for the slow cleaving to have a biological function, e.g. in virus replication, but it is a fact that the HRV89 serotype - 20 - grows significantly more slowly than the HRV2 serotype. As has already been mentioned, the P2A of the poliovirus system is responsible for one or more regulating factors of the translation in the host cell being altered during the infection. These findings can also be applied to the rhinoviral system. The "trans" activity of the P2A responsible for this change in the regulating processes has been detected in the rhinovirus system for the protease 2A. The influencing or inhibition of this process would have a therapeutically positive, though subsidiary effect on the viral events. A further object of the present invention was therefore to find the parameters necessary for the "trans" activity. Accordingly, different peptide substrates and different conditions for incubation were investigated. The experiments show that the cleaving efficiency of HRV2-2A in trans depends essentially on the length of the cleavage peptide used and the cleaving specificity is not so much determined by the cleaving signal itself: rather, the amino acid sequence of the cleaving region has a considerable influence on the "recognisability" of a cleaving site. The present invention therefore also relates to oligopeptides which are cleaved with varying degrees of efficiency by the protease 2A. Particularly preferred are those oligopeptides which have at least five amino acids from the C-terminal region of VP1 and eight amino acids from the N-terminus of the protease 2A or six amino acids from each of these regions. They can be investigated for their suitability as competitive inhibitors of the "trans" activity and may possibly be used therapeutically in rhinoviral infections. However, these compounds are particularly suitable as guide substances in discovering the active substance for influencing the "trans" activity of the protease 2A, 23 4 2 - 21 - particularly as guide substances for inhibitors. The results obtained provide valuable guidance for the structure/activity-related search for active substances which will have a therapeutically useful effect on the "cis" and/or "trans" activity of the rhinoviral proteases. When testing various conditions for incubation it was shown that, in the presence of 2.5 M NaCl in the trans activity buffers (see Table 1), the peptide substrate is completely cleaved by HRV2-2A and there is no non-specific degradation of the peptide substrates and their cleavage products (demonstrated by the example of the trans-cleaving of peptide K; see Fig. 8). Three reasons may be adduced for this salt effect: a) Influencing the solubility or stabilisation of the peptide substrates, the cleavage products thereof and the HRV2-2A in the cell extract by high salt concentrations. b) Suppression of the non-specific degradation of the peptides, their cleavage products and the HRV2-2A by inhibition of the host cell-specific E. coli proteases in the cell extract. c) Promoting a suitable structure for the HRV2-2A which will favour the "trans" activity or facilitate the substrate binding. In order to characterise the HRV2-2A further, the influence of various detergents, solvents and denaturing agents on the efficiency of the "trans" activity of HRV2-2A was investigated. It was found that all the detergents used and even glycerol have a drastically negative effect on the activity of HRV2-2A (Table 1). Even at very low dilutions a 30% inhibition was found, with the exception of DMSO. Substances such as ammonium 234212 - 22 - sulphate, guanidinium hydrochloride and urea, on the other hand, showed a comparatively much smaller effect on the "trans" activity. The use of high ion intensities did not inhibit the "trans" activity but, on the contrary, resulted in increased efficiency of the "trans" activity (100% cleaving of the peptide substrate, suppression of the non-specific degradation of the peptide cleavage products, etc.). These results may provide the first indications as to the mechanism of the "trans" activity of HRV2-2A and make it possible to speak of a possible structural quality of the protease 2A during the cleaving of peptides in trans. - On the one hand it is known that the presence of high salt concentrations (2.5 M NaCl) favours the formation of non-monomeric structures (e.g. dimeric structures), - but on the other hand mild detergents prevent this structure (e.g. Tween, NP40, glycerol etc.), which in the case described above is shown by a reduction in the "trans" activity. The possibility that HRV2-2A occurs in a multimeric form is supported by the fact that when HRV2-2A-containing E. coli cell extracts are separated on a non-denaturing protein gel and subsequently an immune blot is produced with anti-PC20, the protease can only be detected in a higher molecular mass corresponding to its dimeric form (see Fig. 9) . As has already been shown by deletion studies and in vitro mutagenesis of the C-terminal end of the protease 2A, this region is essential for the "cis" activity (Sommergruber, W. et al., loc. cit.). Anti-PC2 0, a polyclonal antiserum directed against the last 20 amino acids of the native protease 2A, however, has no influence on the "trans" activity of 2A. This means 234212 - 23 - that: a) either the C-terminus of HRV2-2A is not essential for the "trans" activity or b) the C-terminal region is not accessible for the antiserum under these conditions; i.e. for structural reasons the antibodies are unable to bind to the antigenic site. The results obtained provide valuable points of reference for the structure/activity-related search for active substances which will have a therapeutically useful effect on the "trans" activity of the viral proteases, particularly the rhinoviral proteases. The present invention relates inter alia to (a) a viral system, characterised in that it has mutations at or around the PI site of the rhinoviral protease 2A, which will influence the "cis" activity of the protease; (b) a viral system characterised in that it has mutations which result in a disruption of the active centre of the rhinoviral protease 2A; (c) a viral system in which the mutations code for the amino acids Tyr, Gin, Leu, Met, Phe, Val, lie, particularly for valine or isoleucine at or around the PI site; (d) a viral system wherein the (P4-P1-P11-P4') region of HRV2 is replaced by that of HRV89; (e) use of the viral systems for discovering therapeutically relevant compounds; and (f) use of the amino acid sequences derived from these viral systems for the preparation of pharmaceutical compositions which will negatively affect the proteolytic processing of rhinoviruses. The term "viral system" is intended to encompass recombinant expression systems and in vitro test systems which can be used on the one hand for the production of - 24 - wild-type HRV2 2A and mutants thereof (particularly mutants of the active centre and the cleaving site region) and on the other hand as test systems for discovering suitable inhibitors, whilst partial sequences of these systems can serve as a basis for the synthesis of therapeutically effective oligonucleotides and oligopeptides. The invention will now be described by way of non-limiting examples with reference to the drawings. In the drawings the single letter IUPAC code for the amino acids is as follows: A alanine, D aspartic acid; E glutamic acid; F phenylalanine; G glycine; H histidine; I isoleucine; K, lysine; L leucine; M methionine; N asparagine; P proline; Q glutamine; R arginine; S serine; T threonine; V valine; W trytophan; Y tyrosine. Unless otherwise stated, in specifying the mutations in the cleavage site region, reference is made to the wild-type of HRV2 or HRV89 (HRV2: Pl=Ala; PI1=Gly; HRV89: Pl=Val; Pl'=Gly). Figure 1 shows the sequence of oligonucleotides as used for mutagenesis of the cleavage site region of HRV2-2A. The lower case letters indicate the native sequence of the cDNA of HRV2 or the sequence of the expression plasmid pEx2A/21, a restriction enzyme-modified variant of pEx2A (see Example 2). The capital letters indicate the particular mutation. wt represents wild type; HRV89 P1-P4/P11 -P41 corresponds to the cleaving site region as found in the human rhinovirus, serotype 89. Figure 2A shows the expression products of the mutagenesis of the C-terminus of VPl; replacement of Ala by Val, Leu, Phe, Gin and Tyr in the PI position. - 25 - Picture A: Coomassie Blue-stained 12.5' polyacrylamide gel. SDS — 0: pEx2A: Tyr: non-induced pEx2A/21 induced starting plasmid pEx2A/21 Val, Leu, Phe, Gin: mutants of pEx2A/21 with an altered amino acid at position PI (Ala -► Val, Leu, Phe and Gin) mutant of pExl8521 (Ala -+ Tyr in the PI position); this vector construction contains a part of 2B which prevents recognition by anti-PC2 0 75 kd 65 kd 50 kd 24 kd 15 kd unprocessed expression product of pExl8521 unprocessed expression product of pEx2A and mutants processed expression product of pExl8521, pEx2A and mutants; corresponds to the fusion part (98 amino acids of the MS2 polymerase), the C-terminal end of VP3 and total VPl cleavage product of pExl8521, consisting of HRV2-2A and the N-terminal section of 2B HRV2-2A Figure 2B shows the expression products of the mutagenesis of the cleavage signal for 2A; replacement of Ala by Met and lie in position PI and replacement of the cleaving region of HRV2 (Ile-Thr-Thr-Ala-Gly-Pro-Ser-Asp) by the cleavage region of HRV89 (Val-Thr-Asn-Val-Gly-Pro-Ser-Ser); this relates to positions PI to P4 and PI• to P4'. a: Coomassie Blue-stained 10% SDS-polyacrylamide gel b: Western blot of an identical gel with anti-MS2-Pol c: Western blot of a 15 anti-PC20 - 26 - % SDS-polyacrylamide gel with Trace m: Trace 1: marker proteins pVP2D2 Trace 2: pEx2A/21 Trace 3: pEx2A/21 Trace 4: pEx2A/21 Trace 5: pEx2A/21 Trace 6: pEx2A/21 (MS2Pol-VP2 fusion protein as • positive control for the anti-MS2-Pol antibody) not induced induced with mutation Ala-^Met in position PI with mutation Ala->Ile in position PI with the cleaving region HRV2 replaced by HRV8 9; (ITTAGPSEHVTNVGPSS) in positions PI to P4 and PI1 to P41 (for detailed description see Figure 1 or Example 1) 65 kd 50 kd 15 kd unprocessed expression product of pEx2A/21 and mutants of pEx2A/21 processed expression product of pEx2A/21 and mutants HRV2-2A Figure 3 diagrammatically shows the new creation of recognition sites for restriction enzymes around the cleaving site of HRV2-2A and in the coding part of HRV2-2A. Non-emboldened restriction enzymes reproduce the recognition sequences which were originally present in the pEx2A; framed, emboldened ones indicate the newly introduced recognition sequences in pEx2A/21 and pEx2A/II. AS indicates amino acids. igg: ■ Figure 4 shows the doubled-stranded oligonucleotide for;' preparing pEx2A/21, a restriction-modified variant of pEx2A. , 2 3 4 2 12 - 27 - The oligonucleotide is flanked by an Acc I or a BstE II site; the emboldened restriction sites were introduced into the DNA by means of the altered nucleotide sequence of the mutation oligo whilst retaining the original amino acid sequence. Lower case letters indicate the original sequence of HRV2-CDNA; capital letters indicate the modified sequence. Nucleotides marked with a dash indicates the recognition sequence for the restriction enzyme in question. Figure 5 shows the Western blot of a 15% SDS-polyacrylamide gel using anti-PC20. Trace 1: expression product of pEx2A Trace 2: expression product of pEx2A/21, a mutant of pEx2A with new singular restriction sites in and around the cleaving region of HRV2-2A. Trace 3: expression product of pEx2A/II, a mutant of pEx2A, which has new singular restriction sites within the coding region of HRV-2A. The arrow on the left at 15 Kd indicates the expression product (HRV2-2A) which is specifically shown up by the peptide antibody. On the right the molecular size markers are shown (in Kd) . In all three cases the HRV2-2A is still proteolytically active, is recognised by anti-PC20 and exhibits an expression product of the same size. Figure 6 shows 3 double-stranded oligonucleotides for preparing pEx2A/II, a mutant with 2 new singular restriction sites inside the region coding for HRV2-2A. The 3 mutation oligonucleotides were ligated to one - 28 - another by means of overhanging ends (marked by * or #), which have no restriction sites. After ligation has been carried out the fragment about 270 bp long can be recut with the restriction enzymes BstE II and Apa I. Lower case letters indicate the original nucleotide sequence of the HRV2-CDNA; capital letters indicate the mutated sequence. Emboldened restriction sites correspond to the newly generated recognition sequences. Figure 7 shows various peptide substrates and their effect on the efficiency of the "trans" activity of HRV2-2A. + Cleaving with great efficiency and comparable kinetics. - Peptide cannot be cleaved under standard conditions (see Example 3). +/- Peptide is cleaved with only low efficiency. The peptide polio 2A/I represents the cleaving site VPl/2A for the protease 2A of poliovirus type 1 and 2A/II represents the alternative cleaving site for protease 2A in 3CD for poliovirus type 1. Figure 8 shows HPLC profiles of peptide K (TRPIITTYGPSDMYVH; tR=21.6) and the cleaving products thereof TRPIITTY (tR=18.6) and GPSDMYVH (tR=12.2) after incubation in supernatants of pEx2A; the "trans" activity test was carried out as described in Example 3, using various NaCl concentrations: 1 2.5 M NaCl 2 0.5 M NaCl 3 50 mM NaCl 4 5 mM NaCl 23 4 - 29 - 5 0.5 mM NaCl The dotted line indicates the position of the peak which represents the uncleaved peptide substrate (tR=21.6). Figure 9 shows the immune blot of a denaturing and a non-denaturing protein gel from soluble fractions of pEx2A and pEx34b. Figure 9A Immune blot of a denaturing 15% acrylamide gel using the peptide antibody PC2 0 Figure 9B Immune blot of a non-denaturing 15% acrylamide gel using the peptide antibody PC20 Trace 1 soluble part of the E. coli extract of pEx2A Trace 2 soluble part of the E. coli extract of pEx34b (negative control for PC20) The small narrow arrows indicate the position of the molecular size markers. The thick arrows at 15 kd on the denaturing protein gel (monomeric form) or at 30 kd on a non-denaturing protein gel (possible dimeric form) indicate the position of the HRV2-2A specific band on the Western blot. Figure 10 shows HPLC profiles of peptide K (TRPIITTYGPSDMYVH; tR= 21.6) and the cleaving products thereof, namely TRPIITTY (tR=l8.6) and GPSDMYVH (tR=12.2). The "trans" activity test was carried out as described in Example 3, except for 150 mM of NaCl in the "trans" activity buffer and a 30 minute pre-incubation of the cell extract of pEx2A in the presence of various concentrations of anti-PC20. The following dilution 2342 12 - 30 - stages of PC20 were used. Trace 11/1 PC2 0 Trace 2 1/10 PC20 Trace 3 1/100 PC20 Trace 4 1/1000 PC20 Trace 5 1/10,000 PC20 Figure 11: Construction of pEx34c x 18521 Figure 12: Construction of the expression plasmid pEx34c x 18521 Figure 13: Electrophoretic separation of the expression products of pEx34c x 18521 (trace 1) and pEx34c x 18731 (trace 2), and the viral coat proteins of HRV2 (trace HRV2) on a 10% SDS polyacrylamide gel and staining with Coomassie Brilliant Blue. Figure 14: Western blot of the expression products of pEx34c x 18521 or 18731 with a polyclonal anti-VPl serum; in traces 1 to 3 the expression product is separated from 3 different clones of 18521 and in trace 4 from 18731. The higher molecular bands in traces 1 to 3, which are difficult to detect, represent the still unprocessed expression product of 18521. Figure 15: Amino acid sequence of the HRV2-protease-2A-region (in heavy type). The capital letters indicate the amino acids which are identical between rhino, polio and coxsackie viruses. Double arrows indicate the position and nature of the exchange or deletion used to characterise the probably active centre and the C-terminus of the protease 2A of HRV2. Figure 16: Oligonucleotide cassette for mutagenesis of the probably active centre of protease 2A of HRV2. By 12 - 31 - combining the two double-stranded oligonucleotides WT12+WT34 the coding region for wild-type protease 2A is obtained. After the last amino acid of 2A (glutamine 14 2) two stop codons were inserted. If instead of WT12 the double-stranded oligonucleotide WT34 is combined with an oligonucleotide which has a suitable mutation or deletion owing to its altered base sequence, any desired change in the amino acid sequence in this region can be made (shown here by the example of the construction of wild-type 2A and the mutations for Cysl06-+Ser, Cysll2->Ser and Hisll4-*Gly) . Figure 17 shows the protein pattern of pEx3 4c (negative control), pExl8521, pExl8731, pEx2A and all the mutants of pEx2A in the probably active centre of 2A. Band A corresponds to the unprocessed expression product (65 K) of pEx2A or the mutants of pEx2A Band B corresponds to the expression product (58 K) of pExl8731 Band C corresponds to the processed expression product (50 K; MS2-polymerase part+C-terminal part of VP3+ total VPl) of pExl8521, pEx2A and pEx2A[Prol03-»Gly] Band D corresponds to the mature protease 2A (15 K) Fig. 17A; Anti HRV2 ; Western blot of the same gel with a polyclonal antiserum against HRV2 Fig. 17B: Anti PC20; Western blot of the same gel with a polyclonal antiserum against PC20 (peptide synthesised from the last 20 amino acids of 2A; see Example 7) Fig. 18: Coomassie Blue; protein gel stained with ^21 JAN 1993^ Coomassie Blue <>/ 2 7 f H O H £ - 32 - Fia. 19 shows the sequence of the oligonucleotides as used for the mutagenesis of the PI site of HRV2-2A. The lower case letters give the original sequence of the cDNA of HRV2 or the sequence of the expression plasmid pEx2A/21, a restriction enzyme-modified variant of pEx2A (see Example 2). Capital letters indicate the mutation whilst wt indicates wild-type. Fig. 20 shows the sequence of oligonucleotides as used for the mutagenesis of the PI' site of HRV2-2A. The lower case letters indicate the original sequence of the cDNA of HRV2 or the sequence of the expression plasmid pEx2A/21, a restriction enzyme-modified variant of pEx2A (see Example 2). Capital letters indicate the particular mutation whilst the dots indicate the deletion of an amino acid, wt denotes wild-type. Figure 21 shows the sequence of the oligonucleotides as used in the mutagenesis of the P21, P4' and P9' sites of HRV2-2A. The lower case letters indicate the original sequence of the cDNA of HRV2 or the sequence of the expression plasmid pEx2A/21, a restriction enzyme-modified variant of pEx2A (see Example 2). Capital letters indicate the particular mutation and the dots indicate the deletion of an amino acid, wt denotes wild-type. Figure 22 shows the sequence of the oligonucleotides as used for the mutagenesis of the P1 region or P and P1 regions of HRV2-2A. The lower case letters indicate the original sequence of the cDNA of HRV2 or the sequence of the expression plasmid pEx2A/21, a restriction enzyme modified variant of pEx2A (see Example 2). Capital letters indicate the particular mutation. 234212 - 33 - wt denotes wild-type. Poliovirus PI1-81 corresponds to the P' region as found in all three types of poliovirus (sabin strains). Poliovirus Pl'-S', P41 His-<-Met, P61 Asn-+Tyr corresponds to the modified poliovirus sequence with the two mutations. HRV14 PI1 -8 ' , P2 ' Leu-+Pro corresponds to the sequence of the P' region as found in human rhinovirus serotype 14, with a mutation. HRV1B P7-81 corresponds to the sequence of the cleavage site as is found in human rhinovirus serotype IB. Figure 23 shows the sequence of the oligonucleotides as used in the mutagenesis of the cleavage site region of HRV2-2A. The lower case letters indicate the original sequence of the cDNA of HRV2 or the sequence of the expression plasmid pEx2A/21, a restriction enzyme-modified variant of pEx2A (see Example 2). Capital letters indicate the particular mutation, wt denotes wild-type. The amino acids which correspond to the mutated sequence of HRV2-2A are specified. Figure 24 shows the expression products of the mutagenesis of the PI site of the cleavage site of HRV2-2A (i.e. the last amino acid of VPl). A (Ala) can be found in the wild-type pEx2A/21: the substitutions of Ala by Tyr, Phe, Gin, Met, Leu, lie and Val are described in Example 6 and the replacement of Ala by Thr is described in Example 8. Image A: Coomassie Blue stained 10% SDS polyacrylamide gel. 65kD unprocessed expression product of the mutants 59kD processed expression product of the mutants; corresponds to the fusion part (98 amino acids of the MS2 polymerase), the C-terminal end of VP3 and the total VPl 234212 - 34 - 15kD HRV2-2A The percentage of uncleaved product was determined by densitometry and is given between image A and B. Image B: Western Blot of a 15% polyacrylamide gel incubated with the anti-PC20 antibody. Figure 25 shows the expression products of the mutagenesis of PI', P21, P4• and P9• sites of the cleavage site of HRV2-2A (i.e. the N-terminus of 2A itself). Replacement of Gly by Glu, Lys, Thr and Trp in the PI site, of Asp by Thr in the P4' site and Val by Asp and Thr in the P9 • site. Deletions of Gly in the PI1 and Pro in the P2• site. Image A: Coomassie Blue-stained 10% SDS-polyacrylamide gel. 65kD unprocessed expression product of the mutants 50kD processed expression product of the mutants; corresponds to the fusion part (98 amino acids of the MS2-polymerase), the C-terminal end of VP3 and the total VPl 15kD HRV2-2A The percentage of uncleaved product was determined by densitometry. Image B: Western Blot of a 15% polyacrylamide gel incubated with the anti-PC20 antibody. Figure 26 shows the expression products of mutagenesis of the P' or P and P1 regions of HRV2-2A; replacement of the P' region of HRV2 by the P' region of poliovirus; 2 3 4 ^ - 35 - for the P' region of poliovirus with an exchange at the P4 ■ site (His-+Met) and an exchange at the P61 site (Asn-*Tyr) ; for the P' region of HRV14 with an exchange at the P21 site (Leu-*Pro) , and replacement of the P and P1 region by those of HRV1B. Table A: Sequence of mutations compared with the HRV2-2A cleavage site. Trace 1: pEx2A/21 with the PI'-71 region of poliovirus serotype 1 Trace 2: pEx2A/21 with the PI'-7 1 (P4 ' His-+Met, P61 Asn-^Tyr) region of poliovirus serotype l Trace 3: pEx2A/21 with the PI'-71 region (P21 Leu-+Pro) of HRV14 Trace 4: pEx2A/21 with the Pl-7/Pl'-8' region of HRV1B The amino acids underlined in variants 2 and 3 indicate that these are differences from the wild-type poliovirus serotype 1 of the wild-type sequence of HRV14. The amino acid underlined in variant 4 shows a difference at this position from the sequence of HRV2. The reduction in cleaving efficiency (as a percentage) determined by densitometry is given in each case. Image B: Coomassie Blue-stained 10% SDS-polyacrylamide gel. The percentage of uncleaved product was determined by densitometry. Image C: Western Blot of a 15% polyacrylamide gel incubated with the anti-PC20 antibody. 65kD unprocessed expression product of the mutants 50kD processed expression product of the mutants; corresponds to the fusion part (98 amino acids 234 2 12 - 36 - of the MS2 polymerase), the C-terminal end of VP3 and the total VPl 15kD HRV2-2A Figure 27 shows the expression products of the mutagenesis of the P region of HRV2-2A; replacement by those amino acids which occur in the P region of HRV89. The precise changes are shown in Figure 22. Table A: Sequence of the mutations compared with the HRV2-2A cleavage site. Trace 1: pEx2A/21 with replacement at the P2 site of Thr-+Asn Trace 2; pEx2A/21 with the P2 -5 region of HRV89 Trace 3: pEx2A/21 with the PI -5 region of HRV89 Trace 4: pEx2A/21 with the PI -8 region of HRV89 Trace 5: pEx2A/21 with the PI- -4/P1'-4' region of HRV89 (see Example 6) The reduction in cleaving efficiency (in percent) determined by densitometry is given in each case. Image B: Coomassie Blue stained 10% SDS polyacrylamide gel. The reduction in cleaving efficiency was determined by densitometry and is given in percent. Image C: Western Blot of a 15% polyacrylamide gel incubated with the anti-PC20 antibody. 65kD unprocessed expression product of the mutants 50kD processed expression product of the mutants; corresponds to the fusion part (98 amino acids - 37 - of the MS2 polymerase), the C-terminal end of VP3 and the total VPl 15kD HRV2-2A Figure 28 shows the differences in amino acids between HRV2-2A and HRV89-2A. The lower case letters show the amino acid sequence of the last 8 groups of VPl of HRV2 and the total sequence of HRV2-2A. The amino acids which are different in HRV89-2A are shown underneath the HRV2 sequence in capital letters. The cleavage site is indicated by an arrow. Figure 29 shows the sequences of the oligonucleotides which were used in Example 9. A. PvuII/Hindlll oligonucleotide cassette for substitution of the 21 amino acids at the C-terminus of pEx2A/Pl-4/Pl'-4'/89 by those of 2A of HRV serotype 89. A stop codon (marked with an asterisk) was inserted after the last amino acid of 2A. B. Mlul/Nsp (7524) I oligonucleotide cassette for introducing the 0.342kb Nsp (7524) I/Pvu II DNA fragment into the pEx2A/S89"/C89 plasmid cut with Mlu I and Pvu II. C. Oligonucleotides as used for the mutagenesis of the 2A cleavage site in the plasmid pEx2A/S89/89. The lower case letters give the original sequence of the cDNA of HRV89 or the sequence of the restriction variant pEx2A/S89/89. Capital letters indicate the particular mutation whilst the dots indicate the deletion of an amino acid and wt denotes wild-type. 23 4 2 - 38 - Figure 3 0 shows the restriction cutting sites which were used in order to construct the plasmid pEx2A/S89/89. Abbreviations in the restriction enzyme: A, Ava I; Ac, Acc I; B, Bst EII; C, Cla I; Hind III; Mlu I; N, Nsp (7524) I; P, Pst II; Pv, Pvu II. Figure 31 shows the expression products of the plasmids pEx2A/S89/89 and pEx2A/S89"/C89 or of two cleavage site mutants of plasmid pEx2A/S89/89 (pEx2A/S2/89 and pEx2A/S89'/89) . Table A: Sequence of the mutations compared with the HRV2-2A cleavage site. The cleavage site in question and the enzyme in the constructions are given. 89' and 89" indicate that the cleavage sites in these constructions contain mutations from the HRV89 wild-type. Trace 1: Trace 2: Trace 3: Trace 4: Trace 5: Fig. B: pEx2A/21. pEx2A/S2/89 with the cleavage site sequence of HRV2 and the 2A sequence of HRV89. pEx2A/S89 '/89 with replacement of Val-*Ala at the PI site. pEx2A/S89/89 with the cleavage site sequence of HRV89 and the 2A sequence of HRV89. pEx2A/S89"/C89. Coomassie Blue-stained 10% SDS-polyacrylamide gel. Fig. C: Western Blot of a 10% polyacrylamide gel incubated with the anti-MS2-Pol antibody. 234 2 1 - 39 - Image D: Western Blot of a 15% polyacrylamide gel incubated with the anti-PC20 antibody. Image E: Western Blot of a 10% polyacrylamide gel incubated with the anti-MS2-Pol antibody. O 0 o - 40 - EMPIGEN (Alkyl-dimethyl-ammonlum-betaln) 50% 5% 0,5% 0,05% 0,005% 100% 1 00% 100% 45% 35% NP40 (Octyl-phenol-ethy leneoxide) 50% 5% 0,5% 0,05% 0,005% 100% 50% 45% 45% 45% Tween 20 Polyoxyethlensorbftan Monolaurate) / 5% 0,5% 0,05% 0,005% / 55% 45% 4 0% 30% Glycerin 40% 4% 0,4% 0,04% 0,004% 80%. 35% 35% 35% 35% Ammoniurnsulfate 1,9 M 0,4 M 40 mM 4 mM 0,4 mM 10% ; 0% 0% 0% ft . 0% Guanidinium-Hydrochloride 3,5 M 0,7 M 70 mM 7 mM 0,7 mM 100%. 90% 50% 0% 0% Urea 3,5 M 0,7 M 70 mM 7 mM 0,7 mM 100% 20% 5% 0% . 0% DMSO (Dlmethy Is ulf oxide) 10% 1% 0,1% 0,01% 0,001% 25% ; : 10% . >5% >5% >5% NaCl 2,5 M 0,5 M 50 mM 5 mM 0,5 mM 0% 0% 0-5% 5% 5-20% Table 1 The influence of various denaturing and solubilising agents on the "trans" activity of HRV2-2A in the "trans" cleavage assay using peptide K (see the text) as substrate. The top box of each line gives the concentration of the substance in question (in molarity or percent v/v). The shaded lower part indicates the degree of inhibition found, based on the ratio of the peak surfaces of peptide substrate to peptide cleavage products (for the details of methods see text). 0% means that no uncleaved peptide substrate can be detected. 23 4 2 - 41 - Example 1; Preparation of an active and inactive P2A-enzyme substrate of HRV2 for expression in E. coli Starting from pUC9 and the cDNA clones of HRV2 (Fig. 11) an expression system for P2A was constructed. First of all, about 10 /xg of pUC9 were opened by double digestion with BamHI and PstI in the polylinker region. The linearised form of pUC9 was separated from traces of the uncut plasmid using Whatman DE81 paper (Dretzen, G.M., Bellard, P., Sassone-Corsi and Chambon, P.; 1981, Anal. Biochem. 112. 295 - 298). After separation of the DNA fragments on agarose gel a slot was cut in front of and behind the DNA band to be isolated and a strip of DE81 paper was inserted in each slot. Electrophoresis was continued (ensuring that the gel was not covered with buffer) until the desired DNA fragment was totally bound to the front DE81 strip. The back DE81 strip prevented contamination by larger DNA fragments. The DE81 paper with the bound DNA fragment was transferred into a 1.5 ml Eppendorf tube (with an outflow hole in the bottom of the vessel and polyallomer wool placed above it) and washed twice for 5 minutes with 400 /xl of washing buffer (0.2M NaCl, 25mM TrisHCl pH=8.0, ImM EDTA), the washing solution being caught in a second Eppendorf tube arranged below it by brief centrifuging (about 1 sec.). Then the bound DNA was washed out of the DE81 paper by incubating twice for 15 minutes in 200 fJLl of elution buffer (1M NaCl, 25mM TrisHCl pH=7.5, ImM EDTA) . The 400 /il of eluate were centrifuged for 10 minutes in an Eppendorf centrifuge (15000g) to remove any fragments of paper. The supernatant was carefully transferred into a new Eppendorf vessel, 800 /zl of 96% ethanol were added, the mixture was precipitated at -20°C (about 2 hours) , washed twice with 7 0% ethanol and dried. In parallel, the plasmid from clone 719 was 23 4 2 - 42 - digested with BglH and PstI and the plasmid from clone 107 was digested with PstI (Fig. 11). These two plasmids are pBR322 vectors and contain HRV2-cDNA fragments which had been inserted via homopolymeric G-C regions in pBR322. The Bglll/PstI fragment of clone 719 represents the HRV2-CDNA region of 2145 - 2421 (see Fig. 11); the Pstl/PstI fragment of clone 107 covers the adjacent region 2421-3100. These two fragments were inserted in the BamHI and PstI site of pUC9, with both cutting sites (BglH and BamHI) being destroyed. This construction was designated pl8 (see Fig. 11). In order to obtain competent cells for transformation, a modified method of Mandel and Higa was used (Mandel, M. and Higa, A., 1970, J. Mol. Biol. 53., 159-162). 0.5 ml of an "overnight culture" of E. coli strain HB 101 in 50 ml of LB medium (10 g/1 of trypton, 5 g/1 of yeast extract, 10 g/1 of NaCl) were over-inoculated, cultured up to an OD6OO of about 0.4 and then pelleted for 5 minutes at 5 k and 4°C. The bacteria were then carefully resuspended in 25 ml of 0.1M MgCl2 (ice-cold), put on ice for 5 minutes and centrifuged once again for 5 minutes at 5 k and 4°C. The pellet was resuspended in 25 ml of 0.1M CaCl2 (ice-cold) , put on ice for 4 hours and centrifuged for 5 minutes at 5 k and 4°C. The cells were taken up in 2.5 ml of lx storage buffer (0.1M CaCl2/glycerol = 4/1% v/v) , put on ice for 20 minutes, divided into 100 ij.1 aliquots and flash-frozen in liquid nitrogen and stored at -80°C. 5 ij.1 of the ligation mixture described above were added to 100 n1 of competent cell suspension thawed on ice, the cells were incubated for 1 hour on ice and for 2 minutes at 42°C and then placed on ice for 5 minutes. Before the cells were plated out 900 /xl of LB-medium were added, the mixture was incubated for 10 minutes at 37"C and 200 n1 batches of cell suspension were placed on LB-agar dishes (1.5% agar in LB-medium with 100 mg/1 of ampicillin) and incubated overnight. - 43 - As described above, competent JM 101 cells were transformed with the plasmid pl8. Some of the Ampr clones obtained were subjected to restriction analysis and plasmid DNA was obtained on a large scale from one of the positive clones (18/1) (T. Maniatis et al., 1982, Molecular Cloning: A Laboratory Manual; Cold Spring Harbor 86ff). The plasmid was then purified by a Sephacryl-S-1000 column (diameter 0.9 cm, length 20 cm). A TE-buffer was used as elution buffer. The eluate was divided into fractions of about 0.5 ml and the individual fractions were measured at 260 nm (usually, the plasmid peak appears between the 9th and 14th fractions). The beginning of the RNA-peak can be expected from the 17th fraction onwards (OD more than 3.0). The fractions in question were combined and lyophilised, the plasmid was taken up in 0.5 ml of TE-buffer, incubated for 5 minutes at 65 °C, extracted with phenol/chloroform and chloroform, precipitated, centrifuged, dried and dissolved in 100 fil of TE-buffer. About 10 nq of plasmid 18/1 were cut with AccI and Hindlll (the AccI site comes from the HRV2-CDNA whilst the Hindlll site comes from the polylinker region of pUC9). In parallel, the clone 521 was digested with AccI and Hindlll. The Accl/Hindlll fragment of clone 521 includes the HRV2-CDNA of nucleotide number 3075-3698 (Fig. 11). The Accl/Hindlll fragment of 521 was ligated into 18/1 (Accl/Hindlll) and competent JM 101 were transformed therewith as described above. The colonies obtained were checked by restriction analysis (EcoRI, PstI, AccI and Hindlll) and the plasmid DNA of some clones was sequenced. A clone which contained the HRV2 sequence of 2145 - 3698 and the correct reading frame, was designated 18521 and selected for expression. In construction 18521, the homopolymeric G-C regions which originate from the cloning in pBR322 are present on the 5'-side of the Hindlll site. pl8521 was cut EcoRI and Hindlll, the 23 4 2 - 44 - fragment was isolated using DE81 paper and inserted in pEx34c (EcoRI/Hindlll) by ligation. pEx34 is a 3.0 kb expression vector (a derivative of pPLc24; E. Remault, P. Stanssens and W. Fiers; 1981, Gene 15, 81-93) which contains the following sections: - the prokaryotic ribosomal binding site - part of the coding region for the first 98 N-terminal amino acids of MS2 polymerase; this fusion part has hydrophobic and basic amino acids and causes a reduction in the solubility of the fusion protein and increases the stability of the expressed product in the cell. - the fusion protein is under the control of the left-hand lambda promoter - a small polylinker region in 3 different reading frames (pExa, b and c) with cutting sites for EcoRI, BamHI and Hindlll makes it possible to insert suitable DNA fragments in phase behind the fusion protein part - the "ori" and ampicillin resistance region of pBR322. E. coli W6 (lambda) constitutively expresses the gene for the wild-type lambda repressor and is suitable for cultivating the pEx plasmids. E. coli 537, on the other hand, carries the cl 853 lambda repressor mutation (inactive at 42°C) on another plasmid which also carries a kanamycin resistance gene (K. Strebel, E. Beck, K. Strohmaier and H. Schaller; 1986, J. Virol. 57. 983-991). By insertion of the EcoRI/Hindlll fragment of pl8521 into pEx34c it was possible to obtain an expression system including the region: (VP3)-VP1-P2A-(P2B) of HRV2 (2145-3698; see Fig. 12). This expression system produces a viral polypeptide acting as substrate, which also has P2A protease activity. In order to obtain an inactive enzyme substrate for P2A the expression vector pEx34c x 18521 was cultivated in EL. 23 4 2 - 45 - coli W6(lambda). As described above, the vector was isolated from 500 ml of an overnight culture using the large-scale preparation technique. 2 jxg of pEx34c x 18521 were digested with Hindlll and purified using DE81 as described above. Then the linearised vector was digested with Bal31-nuclease as follows: about 1 /xg of the vector linearised with Hindlll were incubated with 1U Bal31 nuclease (Biolabs) in 20 mM TrisHCl pH=8, 600 mM NaCl, 12 mM MgCl2 and 1 mM EDTA. Aliquot samples were taken after 1, 2, 3, 4, 5, 6 and 8 minutes and digestion was stopped by the addition of EDTA (final concentration 30 mM). The DNA was recovered by ethanol precipitation and 100 ng of the plasmid were incubated with 100 U T4 ligase in 10 mM TrisHCl pH=7.5, 6 mM MgCl2, 6 mM BME and 1 mM ATP overnight at 15°C. The T4-DNA ligase mixture was used directly for the transformation of E. coli W6(lambda). Some of the clones were picked and the plasmid DNA in question was sequenced according to Maxam and Gilbert (Maxam, A. and Gilbert, W. , 1980, Nucleic Acids Res. 6j>, 499 - 560). A clone of which the cDNA ends with the HRV2 nucleotide number 3321 (see Skern, T. et al., 1985, Nucleic Acids Res. 13, 2111 - 2126) was designated 18731. This deletion mutant of 18521 was used as an inactive P2A enzyme substrate for expression studies. Example 2: Expression and detection of the fusion proteins The plasmid pEx34c x 18521 and plasmids of clones of the Bal31 digestion of pEx34c x 18521 were transferred into E. coli cells. The cells were cultured overnight at 28°C in LB-medium (+100 mg of ampicillin/1 and 25 mg of kanamycin/1). The cultures were then diluted 1:5 with preheated (42°C) LB-medium without antibiotics (induction of the lambda-PL promoter) and - 46 - incubated for 2 hours with vigorous shaking at 42°C. After 2 hours the cells from 1 ml of the culture were harvested (2 min. in the Eppendorf centrifuge, 4°C) and resuspended in 500 jul of cold sonication buffer (150 mM NaCl, 50 mM TrisHCl pH=8 and 1 mM EDTA). The cells were broken up using an M. S. E. Ultrasonic Power apparatus (3 bursts of 5 seconds; 1.7 Amp), with a pause of 45 seconds between the individual sonications to avoid over-heating the samples. Insoluble material was then obtained by centrifuging for 2 minutes in the Eppendorf centrifuge and the pellets were dissolved in 200 /z 1 of sample buffer (4% SDS, 125 mM TrisHCl pH=6.8, 10% BME, 10% glycerol and 0.02% bromophenyl blue). After heating to 95°C for 4 minutes, 10 /xl batches of the samples were separated on 10% SDS-PAA gel (Lammli, U. K., 1970, Nature 227. 680-685). Control experiments showed that all the expressed proteins are insoluble in the sonication buffer. The gells were subsequently stained with Coomassie Brilliant Blue as follows: - Staining: - Decolorising: - Glycerol bath: - Ethanol bath: - Drying: 30 - 60 min in 50% methanol, 10% acetic acid and 0.1% Coomassie Brilliant Blue overnight in 5% methanol and 10% _ acetic acid 30 min in 7% acetic acid and 2% glycerol 1 to 2 minutes in 96% ethanol on 3MM paper; 2-3 hours at 80°C on gel dryer (Hoefer, SE 1160) Fig. 13 shows a typical picture of a Coomassie Brilliant Blue staining of an expression of pEx34c x 18521 in 537. The deletion mutant 18731 was also separated (see 2*z * <0 L. - 47 - Fig. 13). pEx34c x 18731 ends at nucleotide number 3 321 and therefore no longer possesses the probably active centre of P2A. In order to demonstrate the antigenic specificity of these expressed forms of 18521 and 18731, a Western blot was made using a polyclonal serum against VPl (ATCC-VR-1112 AS/GP; VPl is an integral part of both expression plasmids pEx34c x 18521 and 18731). The Western blot was taken as follows: For electrotransfer of the separated proteins from the gel to the nitrocellulose, 4 layers of 3MM paper (Whatman) and 1 layer of nitrocellulose (Schleicher and Schuell, BA85, 0.45 /im) were cut precisely to the dimensions of the separating gel and pre-incubated in transfer buffer: 20 mM Tris base 150 mM glycine 20% methanol p.a. (pH=8.8; need not be titrated any longer). The composition of the "transfer sandwich" was made up according to the following plan (avoiding air bubbles!): - Pol -»• scotch brite -► 2 layers of 3 MM -» gel -► nitrocellulose -»• 2 layers of 3 MM -»• scotch brite -»• + pol The protein gel was also equilibrated in transfer buffer for 2 minutes before the sandwich was put together. The transfer buffer can also be prepared as a 10 x solution (24.2 g of Tris and 112.6 g of glycine per litre without methanol). The transfer was carried out in transfer buffer at about 1 ampere, for 2 hours in a protein blot apparatus in the presence of 0.1% Empigen BB (alkyldimethylammoniumbetain; No. 62 852, Marchon France S.A.) (R.E. Mandrell et al.; J. Immunol. Meth. 2342 1 2 - 48 - 67. S. 1 (1984)). The efficiency of the transfer was checked using the pre-stained marker proteins. The filters with the proteins bound on them were soaked overnight at ambient temperature in 50 ml of "blocking solution" i.e. PBS: 137.0 mM NaCl 2.7 mM KC1 8 . 0 mM Na2HP04 1. 5 mM KH2P04 0. 5 mM MgCl2. 6H20 1.0 mM CaCl2.2H20 with 1% BSA, 1% Tween 20 (polyoxyethylene(20)-sorbitanmonolaurate) and 10% heat-inactivated foetal calves' serum (HIFCS). The polyclonal antiserum against VPl (ATCC VR-1112 AS/GP) was pre-incubated before use with an E. coli lysate in order to remove any antibodies specific to E. coli. Equal volumes of antiserum and E. coli cell lysate were mixed together, incubated for 2 hours at ambient temperature and overnight at 4°C. The cross-reacting E. coli proteins were separated from the supernatant as an immunoprecipitate by centrifuging (5 minutes, Eppendorf centrifuge). The nitrocellulose filter was then incubated for 3 hours in "blocking solution" with the polyclonal antibody (pre-incubated polyclonal antiserum diluted 1/500 to 1/1000 in "blocking solution") in a Plexiglas box on a tilting mechanism. The filter was then rinsed thoroughly under running tap water (about 15 minutes) and washed 3 times for 15 minutes with 50 ml of PBS (+1% Tween 20). In the last step the filter was incubated in about 50 ml of "blocking solution" with the alkaline phosphatase conjugated rabbit anti-antibody (diluted 1/5000 to 1/7500 in "blocking solution") at ambient temperature for 3 hours. Finally, the filter was rinsed thoroughly once more under running tap water (15 minutes) and - 49 - washed three times with 50 ml of PBS (+1% Tween 20) as described above. Staining was carried out in 10 ml of phosphatase buffer: 100 mM TrisHCl pH=9.5 100 mM NaCl 5 mM MgCl2 in the presence of the dyes nitro-blue-tetrazolium (NBT; 165 /xg/ml) and 5-bromo-4-chloro-3-indolyl-phosphate (BCIP; 82.5 jug/ml) . The alkaline phosphatase-conjugated anti-rabbit IgG (Fc) and the dyes NBT and BCIP are obtained from Promega Biotec (Protoblot TM System). The staining reaction was also stopped by rinsing under running tap water after about 1 minute. Fig. 14 shows a typical image of a Western blot of pEx34c x 18521 or 18731. Example 3: Establishing an expression system for producing native protease 2A, and analysis of point mutants in the probably active centre of 2A. In order to investigate the role of some highly conserved amino acids in the region of the probably active centre (see Fig. 15), the in vitro mutagenesis and the deletion of individual amino acids in this region was carried out using an oligonucleotide cassette. In parallel, this method was used to express native protease 2A. Starting from pExl852l, by digestion with Apa 1 (nucleotide number 3458 of HRV2 cDNA) and Hindlll (restriction cutting site comes from the polylinker region of the vector; see Example 1) in accordance with the manufacturer's instructions (Biolabs), a 264 bp long DNA fragment was obtained which was replaced by two - 50 - double-stranded oligonucleotides WT 12 and WT 34 with Apa I/HindIII "sticky ends" (see Fig. 16). First of all, 1 /Ltg each of the single-stranded oligonucleotides WT1, WT2, WT3 and WT4 were kinased in 10 /xl (20 mM TrisHCl pH=7.5, 10 mM MgCl2, 20 mM DTT and 1 mM ATP) with 2 U T4-polynucleotide kinase (Biolabs) for 30 minutes at 37°C. Then the kinase mixtures of WT1 and WT2 as well as WT3 and WT4 were combined, incubated for 10 minutes at 68°C, 30 minutes at 45°C, 10 minutes at ambient temperature and then briefly on ice. The oligonucleotides kinased and hybridised in this way were combined, the concentration was adjusted to 1 mM with a 10 mM ATP solution and 28 U T4-DNA ligase (Boehringer Mannheim) was added for the ligation. The ligation itself was carried out first of all for 2 hours at 20°C, then a further 7 U of T4-DNA ligase were added and the ligation mixture was incubated for 40 hours at 14°C. Then the ligation mixture was incubated for 10 minutes at 70°C, adjusted to 100 mM with 1 M NaCl solution and the multimeric forms of the oligonucleotide produced were "re-cut" with 50 U Apa I and 20 U Hindlll. The purification and isolation of the correct double-stranded Apa I/HindIII fragment and the back-cloning of this fragment into pExl8521 (cut with Apa I and Hindlll) and identification by means of sequencing were carried out as described in Example l. A positive clone was selected and used for expression studies and for "trans" activity tests. In this way the clones pEx2A, pEx2A Cys 106 -+ Ser, - Cys 112 -+ Ser and - His 114 -♦ Gly were obtained. The induction of expression was carried out precisely as described in Example 2 and gave the results shown in Fig. 17. 234212 - 51 - Example 4: Production of the peptide antibody PC20 The last 20 amino acids of protease 2A constitute a potential antigenic determinant. A peptide (PC20) which had precisely this amino acid sequence was synthesised and used to induce antibodies in rabbits: 57 0 nq of this peptide were taken up in 0.4 ml of PBS solution and drawn up in a 5 ml syringe. 0.5 ml of Freund's adjuvant (CAF; GIBCO) were drawn up in a second 5 ml syringe and the two components were then mixed using a three-way stopcock valve until an emulsion was formed. The rabbit was pricked in the artery in the ear in order to obtain a preserum for the negative control. The immunisation was achieved by subcutaneous injection of the peptide/CFA mixture at 4 different sites (0.2 ml per injection site) in the back region. After 5 weeks, the immunisation was boosted with 1.2 ml of peptide solution by injecting 0.2 ml by intramuscular route into the back. Eight days later a blood sample was taken by pricking the heart. The blood was allowed to coagulate at ambient temperature, the fibrin and solid constituents were removed with a sterile rod and the blood was centrifuged at 2000 rpm. The serum was divided into aliquots and stored at -18°C. Example 5: Introduction of new singular recognition sequences for restriction enzymes in the section of the expression vector pEx2A coding for the C-terminal region of VPl and the HRV2-2A. In the course of the analysis of the cleaving specificity and in order to insert deletions in the coding region for HRV2-2A, a modified plasmid was - 52 - established, starting from the expression vector pEx2A (see Example 3), this plasmid having new singular recognition sequences for restriction enzymes in the coding region around the cleavage site and in the region for the protease 2A. This modified expression system, whilst retaining the native amino acid sequence, has a changed nucleotide sequence which was introduced by means of synthetic double-stranded oligonucleotides. 1. Starting from the expression plasmid pEx2A, an AccI fragment (about 700 bp) was isolated under standard conditions. This fragment includes the C-terminus of VPl, the entire coding region of HRV2-2A (including the stop codons after the last amino acid for the HRV2-2A) and parts of the vector DNA (see Fig. 11). The purified AccI fragment of pEx2A was then re-cut with BstEII. The larger BstEII/AccI fragment (about 63 5 bp) was separated from the smaller AccI/BstEII fragment (about 115 bp), purified and ligated under standard conditions with a synthetic double-stranded oligonucleotide (about 115 bp; see Fig. 12) which corresponds to the small AccI/BstEII fragment but contains a modified nucleic acid sequence. The newly recombined AccI fragment was re-cut with AccI in order to destroy multimeric forms and ligated into the original pEx2A vector linearised with AccI and then transformed into competent E. coli W6(l). Some 100 ampicillin-resistant clones were obtained and 3 clones were selected and sequenced on the basis of restriction enzyme analyses. One clone, pEx2A/2l, was transformed and expressed in E. coli 537 as described above, the expression product of pEx2A/21 showing identical behaviour on a Western blot as that of pEx2A (see Figure 13, trace 2). 2. In order to obtain suitable new singular restriction cutting sites in the middle region of pEx2A coding for HRV2-2A, first of all the vector DNA of pEx2A was - 53 - digested with BstEII and Apal and separated from the resulting BstEII/Apal fragment (about 270 bp) on agarose gels. Both Apal and BstEII are recognition sequences for restriction enzymes which can originally be found in the cDNA of HRV2 (position number 3458 for Apal and 3188 for BstEII, see German Patent Application P 35 05 148.5). In parallel, 3 double-stranded oligonucleotides (2A-1512AB, 2A-1600AB and 2A-1728AB), representing the region between the Apal and BstEII site of HRV2-2A (see Fig. 4) were ligated, re-cut with Apal/BstEII and incorporated in the pEx2A vector which had been linearised with Apal and BstEII. This modified pEx2A vector was used to transform competent E. coli W6(1)• Of some 100 clones, 4 were selected on the basis of the restriction pattern, then sequenced and competent E. coli 537 was transformed therewith. The expression pattern of a clone (pEx2A/II) shows identical characteristics to those of pEx2A (see Fig. 13, trace 3). As shown in Fig. 11, the newly introduced BstEII/Apal fragment has singular cutting sites for BglH and SnaBI. These two new expression vectors pEx2A/21 and pEx2A/II make it possible, by using the newly generated restriction sites, to incorporate mutation oligonucleotides which were used, on the one hand, for fine analysis of the cleavage site (see Example 6) and which can be used on the other hand for deletion studies, in order to determine the smallest proteolytically active region of HRV2-2A. Example 6: Mutation analysis of the PI site of protease 2A of HRV2. The mutations of the PI position of the cleavage site of protease 2A of HRV2 (hereinafter referred to as HRV2-2A) were carried out either in the expression - 54 - vector system pExl8 521 or in pEx2A. 1. Ala -+ Tyr mutation at the cleavage site of HRV2-2A using the expression vector system pExl8 521. First of all, the alanine at the PI site of HRV2 was exchanged for a tyrosin. This mutation leads to a Y/G amino acid pair at the 2A cleavage site, such as occurs in poliovirus (Toyoda, H. et al., 1986, Cell, 45 761 - 770). Starting from the isolated 1.3 kb Pstl/Hindlll fragment of the expression vector pExl8521 this fragment was incorporated in a BLUESCRIPT vector (Stratagene Cloning Systems). From this newly recombined plasmid, the single-stranded form was isolated by conventional methods or following the manufacturer's instructions (Amersham kit: "oligonucleotide-directed in vitro mutagenesis system") The point mutation of the single-stranded form was carried out using the oligonucleotide EBI 936 (see Fig. 1) following the manufacturer's instructions (Amersham kit: "oligonucleotide directed in vitro mutagenesis kit". Recombinant plasmids containing the mutation were verified by sequencing. The Pstl/Hindlll fragment containing the mutation (Ala -+ Tyr) was isolated and used to replace the wild-type fragment of pExl8521. 2. Mutations at the cleavage site of HRV2A-2A using the expression vector pEx2A. A derivative of the vector pEx2A, namely pEx2A/21, which contains inter alia a newly generated Mlul recognition sequence at position 3137 of the HRV2-cDNA (see Example 5) was used to introduce further mutations in and around the cleavage site by means of double-stranded oligonucleotides (see Fig. 1), using the newly created restriction site Mlul and the BstEII site which - 55 - was originally present in the HRV2-CDNA. A double-stranded oligonucleotide, for example, represents the amino acids in positions Pl-4 and PI'-4' as were found in the serotype HRV89 (see Fig. 1). The preparation of the vector DNA, the integration of the double-stranded oligonucleotide and the DNA sequencing were carried out using standard methods. The expression of the mutated proteins was carried out in E. coli 537 or AR58 as described. The expression products were analysed on SDS polyacrylamide gels (see Fig. 2A and 2B). Identical polyacrylamide gels were subjected to Western-blot analysis, using on the one hand a mouse-monoclonal antibody directed against the N-terminal part of MS2-polymerase (e.g. the anti-MS2-pol; Hansen, J. et al., Embo Journal, Vol. 7, 1785-1791 (1988)); on the other hand a rabbit-polyclonal antiserum was used which recognises a peptide consisting of the last 20 amino acids of native HRV2-2A (hereinafter referred to as anti-PC20). Constructions which contain the amino acids glutamine, leucine, methionine, phenylalanine and tyrosine at the PI position were cleaved as efficiently as the wild-type, as is clear from the absence of bands of uncleaved material at 65 Kd (or, in the case of tyrosine mutation, at 75 Kd) . The mutants which have valine or isoleucine at the PI site were not fully processed, however. When the construction containing valine was used, 25% of the induced protein was not processed; if there is an isoleucine group at this site, even 50% of the induced material was not cleaved. A significant reduction in cleaving efficiency is also found when using those amino acids in positions Pl-4 and PI*-4* which are found in the serotype HRV89 (about 65-70% of the induced protein is not processed). To sum up, it can be said that the amino acids isoleucine and valine obviously do not fit very well into the active centre of HRV2-2A; this is all the more remarkable as valine is the very amino acid which is - 56 - found in the PI site of HRV89. Possibly, the methyl group of the (3-C atom interferes with topography of the active centre or a structural changed in the substrate necessary for proteolytic cleaving. It can therefore be assumed that there are differences in the structures of the active centres of 2A between the serotypes HRV2 and HRV89. Example 7; The influence of different peptide substrates and various conditions of incubation on the efficiency of the "trans" activity of HRV2-2A in vitro. The peptides used in this Example were obtained in i some cases from the firm "Multiple Peptide Systems" (San Diego, California, USA) of synthesised by the "solid phase" method (Merrifield, R.B. 1963, J. Am. Chem. Soc., 85, 2149 - 2154). The peptides were then purified by reverse phase HPLC (0.1% trifluoroacetic acid and acetonitrile as the mobile phase) and identified by HPLC analysis, "fast atom bombardment mass spectrometry (FAB-MS) and amino acid sequencing (Hunkapiller, M.W. and Hood, L.E., 1983, Science, 219, 650 - 659). Unless otherwise stated, all the peptides used have, for reasons of stability, an N-terminus blocked by means of an acetyl group whilst the C-terminus always occurs as an acid amide. The induction of the 2A expression systems in the E. coli strain 537 was carried out was described in Example 2. After 2 hours' incubation at 42"C the cells from 1 ml of culture were harvested (30 seconds in the Eppendorf centrifuge, 4°C) and resuspended in 500 /il of "trans" activity buffer (50 mM Tris HC1 pH=8.0, 2.5 M NaCl and 1 mM EDTA). The cells were broken up by means of an M.S.E. Ultrasonic Power apparatus (30 seconds; in a bath of ice water). Any insoluble material was removed by centrifuging at 23 4 2 57 100,000 g (4°C, 30 min. in the ultracentrifuge; rotor type Ti 50, Beckmann) and 100 jul batches of this supernatant were each combined with 5 /xl of an aqueous peptide solution (4 mg/ml) and incubated for 3 0 minutes at 37°C. The reaction was stopped by the addition of an equal volume of 1M HC104. The samples were then incubated on ice for 20 minutes, centrifuged for 10 minutes at 4°C (Eppendorf centrifuge), mixed with an equal volume of 1.4 M KjHPO^., incubated for 5 minutes at ambient temperature and the precipitate formed was removed by centrifuging (5 mins in an Eppendorf centrifuge). The supernatant containing the peptide was then separated by reverse phase HPLC. The following is a list of all the important HPLC data for the peptide analysis: HPLC apparatus: Automatic injector: Waters WISP 710 A HPLC pumps: Waters Model 510 (2 Waters Temperature Controller Integrator: Gradient control: pumps, mixture at high pressure end) Absorbance Detector 214 nm Extended Module: 280 nm (filter device) (300 C) Waters QA 1 Data System (identical to Hewlett Packard: 3390 A integrator) HPLC conditions: Column: Bakerbond C18 5 mm 4.6*250 mm Photometer sensitivity: 214 nm: 0.5 AUFS 280 nm: 0.05 AUFS 23 4 2 - 58 - Recorder: Sensitivity 10 mV Advance: 1 cm per min. Flow: lml/min Temperature of column: 30°C Quantity of sample injected: 150 jul Eluant A: distilled water + 0.1% trifluoroacetic acid Eluant B: acetonitrile + 0.1% trifluoroacetic acid Gradient conditions: Isocratic 1 minute 90% A 10% B Linear gradient 27 minutes 60% A 40% B Linear gradient 1 minute 20% A 80% B Isocratic 5 minutes 20% A 80% B Integrator settings: Attenuation: 8 Threshold: 7 Chart Speed: 0.5 cm/min (from 8-30 min., otherwise 0.2 cm/min) Preparation of the HPLC samples: The samples were shaken on a vibrating finger, treated for 10 seconds in an ultrasonic bath, centrifuged (Eppendorf centrifuge) and the supernatant was injected. The identification of the peptides and their cleavage products was carried out using the peptide substrate originally 16 amino acids long (Ac-TRPIITTAGPSDMYVH-NHj, peptide H) by means of a reference peptide (GPSDMYVH-NH2) which constitutes the C-terminal cleavage product. The cleaving of peptide H, which occurs specifically between alanine and glycine, was used as a reference system in all of the investigations. When modified peptide substrates were used the C-terminal cleavage products were identified by N-terminal sequencing (Hunkapiller, M. W. and Hood, L. E. loc. n "7 / o / . ,1 / - 59 - cit.) while the N-terminal cleavage products blocked by an acetyl group at the N-terminus were identified by Beckman aminoacid analysis. 1. Starting from the standard peptide H, first of all the question of the length of the peptide and its influence on the efficiency of the "trans" activity were investigated. An asymmetric shortening of the peptides at the N-terminus (see Fig. 7; peptides H, A, B, C, D and E) showed that at least 5 amino acids of the P-region (corresponding to the last C-terminal amino acids of VPl) must be present for cleaving in trans. If the cleavage peptides are symmetrically shortened (see Fig. 7; peptide G, I and J) no cleaving was observed with a peptide length of 7 and 10 amino acids (peptide J and I) whilst only slight cleaving was observed when a peptide substrate 12 amino acids long were used (peptide G) . 2. The cleaving carried out in poliovirus between VPl and 2A of polio-2A was carried out between tyrosine and glycine. Further cleaving is also carried out in poliovirus in the 3CD region between tyrosine and glycine (Toyoda, H. et al.; loc. cit.). The question as to whether HRV2-2A accepts the polio cleavage was answered using other peptide substrates (see Fig. 7; peptides F and K). If the cleavage signal Ala/Gly for the HRV2-2A is replaced by the cleavage signal for polio-2A (Tyr/Gly) without replacing all the other amino acids, these two peptides can be cleaved with the same efficiency as the native peptide substrates of comparable length by HRV2-2A. If, however, peptides are used (see Fig. 7; peptides polio 2A/I and 2A/II) which represent not only the cleavage signal but the entire native cleavage region of poliovirus, no cleaving by HRV2-2A occurs. The peptide polio 2A/I represents the cleavage site VP1/2A for the protease 2A of poliovirus 23 4 2 - 60 - type 1 and 2A/II represents the alternative cleavage site for the protease 2A in 3CD for poliovirus type 1. 3. On testing various conditions of incubation it was shown that, in the presence of 2.5 M NaCl in the "trans" activity buffer (see Table 1) total cleaving of the • peptide substrate by HRV2-2A takes place and there is no non-specific degradation of the peptide substrates and their cleavage products (shown by the Example of the "trans" cleaving of peptide K; see Fig. 8). In these tests, for reasons of improved stability or yield, peptide K was used instead of the original peptide H in the peptide synthesis. All the results shown hereinafter with peptide K may, however, be reproduced with the same efficiency by using peptide H. 4. For further characterisation of HRV2-2A, the influence of various detergents, solvents and denaturing agents on the efficiency of the "trans" activity of HRV2-2A was investigated. The "trans" activity test was carried out under the standard conditions specified hereinbefore (except that only 0.5 M NaCl was used in the "trans" activity buffer). Peptide K was used as the substrate for this test series. It was found that all the detergents used, such as Empigen (alky1-dimethyl-ammonium-betain), NP40 (octyl-phenol-ethyleneoxide), TWEEN 20 (polyoxyethylene sorbitan monolaurate), DMSO (dimethylsulphoxide) and even glycerol have a drastically negative effect on the "trans" activity of HRV2-2A (Table 1). Even at the lowest levels of dilution a 40% inhibition was found, except with DMSO. Substances such as ammonium sulphate, guanidinium hydrochloride and urea, on the other hand, demonstrated a comparatively much smaller influence on the "trans" activity. The use of high ion intensities (up to 2.5 M NaCl) did not inhibit the "trans" activity. On the contrary, the presence of high NaCl concentrations ■i i - 61 - resulted in increased efficiency of the "trans" activity (100% cleaving of the peptide substrate, suppression of the non-specific degradation of peptide cleavage products, etc.). 5. The polyclonal antiserum anti-PC20, which is directed against the last 20 amino acids of the native protease 2A, recognises HRV2-2A on Western blots (Sommergruber, W. et al., loc. cit.). In order to establish the influence of antiPC2 0 on the efficiency of the "trans" activity, increasing amounts of anti-PC20 were added to the "trans" activity tests carried out under standard conditions (except that only 150 mM of NaCl were used) (see Fig. 10). This experiment, in which peptide K was used as peptide substrate, showed that anti-PC2 0 has no effect on the "trans" activity of 2A. Example 8 Mutation analysis of the cleavage site of protease 2A of HRV2. 1. Mutations of the PI site of protease 2A of HRV2 (hereinafter referred to as HRV2-2A). The alanine group was replaced by a threonine group since threonine has a hydroxyl group at the /3-C atom. A derivative of the vector pEx2A, namely pEx2A/21, which contains inter alia a newly generated Mlu I recognition sequence at position 3137 of the HRV2-CDNA and a newly introduced Ava I site (position 3163, see Figure 4) , was used to introduce the Ala->Thr mutation into the PI site using a double-stranded oligonucleotide (see Figure 19), making use of the newly created restriction site Mlu I and Ava I. For the constructions, the oligonucleotide from Figure 19 was - 62 - ligated with the 0.45kb Mlu I/Hind III fragment and the 3.7kb Ava I/Hind III fragment of pEx2A/21. The preparation of the vector DNA, the incorporation of the double-stranded oligonucleotide and the DNA sequencing was carried out using standard methods. The mutated protein was expressed in E. coli 537 or AR58, as described. The expression products were analysed on SDS-polyacrylamide gels (see Figure 24); the effect of the mutations on the cleaving was determined by densitometry, by measuring the ratio between cleaved and uncleaved polypeptide. The PI mutants listed in Example 6 were tested according to this method (see Figure 24). For the densitometry, the polyacrylamide gel was placed between two sheets of previously boiled cellophane film (Bio-Rad) and dried. The polypeptides on the tracers were stained with Coomassie Blue and then quantitatively determined on a Beckmann DU8 Spectralphotometer or a Sebia Preference Ecran Densitometer. Identical polyacrylamide gels were subjected to Western blot analysis, using on the one hand a monoclonal mouse antibody directed against the N-terminal part of the MS2 polymerase (anti-MS2-Pol; Hansen, J. et al., EMBO Journal, Vol. 7, 1785-1791 (1988)), and on the other hand a polyclonal rabbit antiserum which recognises a peptide consisting of the last 20 amino acids of the original HRV2-2A (hereinafter designated anti-PC20; Sommergruber, W. et al., loc. cit.). The expression product of the construction with the Ala-^Thr exchange at the PI site was satisfactorily cleaved. 2. Mutations at the PI' site of the protease 2A of HRV2. A derivative of the vector pEx2A, namely pEx2A/21, - 63 - which contains inter alia a newly generated Mlu I recognition sequence at position 3137 of the HRV2-CDNA was used to introduce mutations into the PI1 site by means of double-stranded oligonucleotides (see Figure 20), using the newly created restriction site Mlu I and the Bst EII site originally present in the HRV2-CDNA. The construction and analysis of the mutations were carried out as described above; the analysis of the results on polyacrylamide gels is shown in Figure 25. Constructions containing the amino acids tryptophan, lysine, threonine and glutamic acid at the PI* site were not cleaved, as is apparent from the absence of the band at 50kd corresponding to the cleaved material. A construction in which the glycine at the PI' site had been deleted was not cleaved either. The derivative of the vector pEx2A/21 was also used in order to introduce mutations into the P21, P4' and P9• site of the cleavage site of protease 2A by means of double-stranded oligonucleotides (see Figure 21). Once again, the newly created Mlu I and the Bst EII site originally present in the HRV2-CDNA were used. The construction and analysis of the mutations were carried out as described above; the analysis of the results on polyacrylamide gels is shown in Figure 25. The construction which has the deletion of the proline group at the P2' site was not cleaved. The replacement of the valine group at the P9' site by aspartic acid resulted in a 30% reduction in the cleaving efficiency, whereas the presence of threonine at the same site did not affect the efficiency. The presence of a threonine group in the P41 site also showed no influence whatever on the activity of the HRV2-2A protease. Furthermore, this system was used to investigate the influence of the sequence between PI1 and P9• on the proteolytic activity of HRV2-2A. The natural sequences which occur in HRV14 and poliovirus were used as the starting basis (the sequences of poliovirus serotype 1, 23 4 2 - 64 - 2 and 3 (sabin strains) are identical in this region). The sequences for the mutagenesis are shown in Figure 22. The oligonucleotides in Figure 22 were introduced into the vector pEx2A/21 as described above and the expressed proteins were analysed (see Figure 26). The expressed protein of the construction with the Pi1-P9' (P21 Leu-»Pro) sequence of HRV14 was not processed. No cleaving of the expressed protein was observed either in the construction which contains the Pl'-P9' sequence of poliovirus or in a very similar construction which contains a variant of the Pl'-P9' polio sequence (P41 His-+Met, P6' Asn-»-Tyr) . 3. Cleaving of the 2A protease of HRV2 at the cleavage site of HRV89 Once again, mutants were prepared using the vector pEx2A/21, but using the newly introduced Ava I site (position 3163, see Figure 4) instead of the Bst EII site. For the constructions, the oligonucleotides in Figure 23 were ligated with the 0.45kb Mlu I/HindIII fragment and the 3.7kb Ava I/Hind III fragment of pEx2A/21. The construction and analysis of the mutations were carried out as described above; the resulting polyacrylamide gels are shown in Figure 27. The cleaving of the expressed proteins clearly shows that the protease HRV2-2A cannot satisfactorily recognise those cleavage sites which contain the amino acids of HRV89. To support this result the sequence corresponding to the cleavage site region of human rhinovirus serotype IB was incorporated: Arg-Pro-Ile-Ile-Thr-Thr-Ala-Gly-Pro-Ser-Asp-Met-Tyr-Val-His-Val instead of Arg-Ala-Ser-Met-Lys-Thr-Val-Gly-Pro-Ser-Asp-Leu-Tyr-Val-His-Val. - 65 - The sequence has a valine group at the PI site; the expression product was only cleaved to a level of 65%, as can be seen from Figure 26. Example 9: Expression of the protease 2A of HRV89 and investigation of the cleavage specificity 1. Expression of the protease 2A of HRV89 (hereinafter referred to as HRV89-2A). In order to show that the 2A gene of HRV89 (EPA 261 403) codes for a protease, a DNA construction was produced in which a DNA fragment corresponding to the region of the 2A gene was bound to the DNA fragment coding for the HRV2A-VP1. After expression of this hybrid in the pEx system, conclusions can be drawn from the length of the fusion protein as to whether the 2A protein of HRV89 acts as a protease. The differences in the 2A gene between HRV2 and HRV89 are shown in Figure 28. To obtain this construction, the following method was used. The starting plasmid used was the derivative of pEx2A/21 described in Example 6 which, in the region of the cleavage site in positions Pl-4/Pl'-4l, codes for an amino acid sequence such as occurs in the serotype HRV89 (hereinafter referred to as pEx2A/Pl-4/Pl'-4'/89) . First of all the Pvu Il/Hind III fragment at the C-terminus of the HRV2-2A (see Fig. 30) was replaced by a synthetic oligonucleotide which contains the sequence of the HRV89-2A (Figure 29). The resulting plasmid (referred to as pEx2A/S89"/C89) was then digested with Mlu I and Pvu II and ligated 2 7 / snj 0 4 c - 66 - with a synthetic Mlu I/Nsp (7524) I oligonucleotide (see Figure 29) and an Nsp (7524) I/Pvu II fragment as shown in Figure 30. The synthetic oligonucleotide codes for the cleavage site as found in the serotype HRV89, and the Nsp (7 524) I/Pvu II fragment codes for the HRV89-2A of amino acids 9 to 121. This fragment may be obtained, for example, from the clone HRV89/68 which was described in EPA 261 403. A plasmid was obtained which was designated pEx2A/S89/89 (see Figures 29 and 30). Two further mutants of pEx2A/S89/89 were constructed in order to check the influence of the individual amino acids on the cleaving efficiency. Again, the mutants were obtained by incorporating double-stranded oligonucleotides, making the use of the above-mentioned Mlu I restriction site and a Cla I restriction site newly present in the Mlu I/Nsp (7524) I oligonucleotide (see Figure 29). In one mutation (hereinafter referred to as pEx2A/S2/89) the HRV2 cleavage site was re-established, and in the second (hereinafter referred to as pEx2A/S891/89) valine was replaced by alanine at the PI site (see Figure 29). The preparation of the vector DNA, the incorporation of the double-stranded oligonucleotides and the DNA sequencing were carried out according to standard methods. The expression of the mutated protein was carried out in E. coli 537 or AR58, as described above. The expression products were separated on SDS-polyacrylamide gels and stained with Coomassie Blue (Figure 31). Identical polyacrylamide gels were subjected to Western blot analysis, using on the one hand a monoclonal mouse antibody directed 23 4 2 - 67 - against the N-terminal part of MS2-polymerase (anti-MS2-Pol; Hansen, J. et al., EMBO Journal, Vol. 7, 1785-1791 (1988)), and on the other hand a polyclonal rabbit antiserum (anti-PC20). The analysis of the expression products of the three constructions with the HRV89-2A is shown in Figure 31. The fusion protein of the mutant which contains the HRV cleavage site (pEx2A/S2/89) is cleaved almost totally; the HRV89-2A enzyme is therefore able to recognise the HRV2 cleavage site just as efficiently as the HRV2-2A enzyme. The uncleaved product migrates somewhat more slowly in the construction with the HRV89-2A, although the two 2A proteins have the same number of amino acids. Possibly, the 13 differences in the amino acid sequence affect the mobility of the 2A polypeptide. The interpretation of the results with the other two constructions with the HRV89-2A (pEx2A/S89/89 and pEx2A/S89'/89) was made more difficult by the fact that in addition to the expected 65kD and 50kD products two other bands with molecular weights of 62kD and 60kD were found. More thorough examination of the bands of the mutant pEx2A/S2/89 with the HRV2 cleavage site showed that these bands were also present in this construction. These bands were not recognised by anti-PC20 and therefore do not have the C-terminal end of the 2A protein; it would therefore appear to be possible that the bands were caused by pausing of the ribosomes or a chain break in the region in front of the C-terminus of the polypeptide. Although there are only four differences in this region between the 2A polypeptides of the HRV2 and the HRV89 from the point of view of proteins, the codon usage is very different. Analysis of the expression products of the intermediate - 68 - construction pEx2A/S89"/C89 also showed a somewhat larger, uncleaved product and the presence of two additional bands, indicating that the reasons for this are connected with the region of the last 2 0 amino acids. In spite of the presence of the two additional bands it is obvious that in the expression system described here the HRV89-2A can recognise the HRV2-2A cleavage site better than its own, as can be seen from the higher proportion of uncleaved product in the constructions pEx2A/S89*/89 and pEx2A/S89/89. Obviously, the presence of alanine at the PI site is crucial to the cleaving by HRV2-2A. In HRV89-2A, on the other hand, the situation is somewhat different: the HRV89-2A enzyme appears not to recognise its own cleavage site particularly well. It is possible that in vivo parts of 2B are necessary for the activity of HRV89-2A and they may possibly influence the configuration of the 2A polypeptide. The configuration of the capsid proteins of HRV89 might also play a part in the cleaving. It is conceivable that the slow cleaving has a biological function, e.g. in the virus replication, but it is nevertheless a fact that the HRV89 serotype grows significantly more slowly than the HRV2 serotype. </div>

Claims

<div class="application article clearfix printTableText" id="claims"> - 69 - WHAT^WE CLAIM IS:-

1. A DNA molecule, coding for the VP1/P2A region of the Rhinovirus system, and having, in the region of the protease 2A, at least one non-native unique recognition site for a restriction enzyme.

2. A DNA molecule according to claim 1, comprising at least one non-native recognition site recognised by a restriction enzyme selected from Xmal, Smal, Mlul, Eagl Bgixi and SnaBI.

3. A DNA molecule according to claim 2, comprising four non-native recognition sites recognised, by respective restriction enzymes Xmal, Smal, Mlul and Eagl.

4. A DNA molecule according to claim 2 comprising two non-native recognition sites recognised by respective restriction enzymes BglH ajid SnaBI.

5. A DNA molecule according to any of the preceding claims which codes; for the "7P1/P2A region of HRV2.

6. An expression plasmid herein designated pEx2A/21 comprising a DNA molecule according to claim 3.

7. An expression plasmid herein designated pEx2A/n comprising a DNA molecule according to claim 4.

8. A test system, comprising a DNA molecule according to any one of claims l to 5 incorporated in a suitable expression system.,

9. A test system according to claim 8 wherein the expression plasmicl pEx2A/21 is used. - 70 -

10. A test system according to claim 8 wherein the expression plasmid pEx2A/II is used.

11. An oligonucleotide, coding for a modified VP1/P2A region, said modification occurring in the total coding region.

12. An oligonucleotide according to claim 11 wherein said modification is at the cleavage site or in the cleavage region of protease 2A, and affecting the "cis" activity thereof.

13. An oligonucleotide according to claim 11 or claim 12 wherein the modification leads to a disruption of the active centre of the rhinoviral protease 2A.

14. An oligonucleotide according to any one of claims 11 to 13, wherein said modification includes substitutions coding for at least one of the amino acids Val, lie, Tyr, Gin, Leu, Met or Phe at or around the PI site of the protease 2A.

15. An oligonucleotide according to claim 14 wherein said modification includes a substitution coding foi? valine or isoleucine.

16. An oligonucleotide according to claim 11, coding for the modified amino acid sequence Val-Thr-Asn-Val-Gly-Pro-Ser-Ser at the Pl-4/Pl'-4' site of the protease 2A.

17. A peptide derived from an oligonucleotide according to any one of claims 11 to 16, and peptidomimetics derived therefrom.

18. A peptide cleavable in trans by the protease 2A.

19. A peptide comprising at least five amino acids of the C-terminal region of VPl and at least eight amino acids from the N-terminus of protease 2A or at least six amino acids from each of these regions.

20. A peptide according to claim 18, selected from peptides H, A, B, C, G, F and K as herein defined.

21. A test system according to claim 9 or claim 10 comprising an oligonucleotide according to any one of claims 11 to 16.

22. A DNA molecule according to claim 1 substantially as hereinbefore described.

23. A test system according to claim 8 substantially as hereinbefore described.

24. An oligonucleotide according to claim 11 substantially as hereinbefore described.

25. A peptide according to claim 17 or claim 18 and peptidomimetics thereof substantially as hereinbefore described. BOEHRINGER INGELHEIM International GmbH by their Attorneys BALDWIN SON & CAREY </div>