Purified Herpes Simplex Virus Protease and Methods of Purification

This invention is directed to a novel method for purifying herpes simplex viral protease.

The Herpes Group of viruses includes many viruses which are pathogenic in humans, such as Herpes Simplex Virus (HSV), Varicella Zoster Virus (VZV), Cytomegalovirus (CMV), and Epstein-Barr Virus (EBV). In this group, B-capsids are essential intermediate structures in virion replication and are formed during the assembly of viral capsids in cell nuclei. One of the more abundant protein components of B capsids, the assembly protein, undergoes proteolytic pressing at its C-terminus and is packaged in mature, infectious virions only in its shortened, mature form (Gibson W. and Roizman, B. 1972. J. Virology 10:1044-1052).

For HSV, the assembly protein is designated ICP35. The protease which acts on the assembly protein is encoded in the same open reading frame as the ICP35 assembly protein, designated the U.sub.L 26 gene (Liu, F. and Roizman, B. 1992 Proc. Natl. Acad. Sci. USA 89:2076-2080). The catalytic domain of the protease is contained within the first 247 amino acids of a 635 amino-acid protein and is apparently released from the precursor through cleavage by the HSV protease itself (Weinheimer, S. P. et al., 1993 J. Virology 67:5813-5822; Welch, A. R. et al., 1991, Proc. Natl. Acad. Sci. USA 88: 10792-10796; Deckman, I. C. et al., 1992, J. Virol. 66: 7362-7367).

Protease activity appears to be essential for viral replication within the entire group of herpes viruses. Thus, it would be desirable to characterize the herpes group proteases as potential antiviral targets. However, the herpes viral proteases, in the form of fusion proteins, have only been partially purified. (Dilanni, C. L. et al., 1993, J. Biolog. Chem. 268:25449-25454; and Weinheimer, S. P. et al., supra). Other herpes proteases (cmv) have been purified and described (see, e.g. WO 93/01291, published Jan. 21, 1993, Applicant: The Johns Hopkins University).

This invention is directed to a method for obtaining isolated and purified herpes simplex vital protease, preferably from the Herpes Simplex 1 or Herpes Simplex 2 Viruses. The method for producing a mature viral protease comprises the steps of separating precursor vital protease from non-protease constituents of a cell culture, and subjecting the precursor protease to autoprocessing conditions so that the precursor protease undergoes autocatalytic cleavage to form mature protease.

In a preferred embodiment of this invention the separation of precursor protein from other cellular components occurs using a cation exchange column. If desired, the mature protease can be separated from the cleavage products by an additional chromatography step. The high purity of the mature protease product of this invention takes advantage of the finding that mature prom and precursor proteases elute differently, so that impurities which co-elute with precursor protease do not co-elute with mature protease.

In one embodiment of this invention the protease is from HSV-1; however due to the sequence homologics between the two vital genes and proteins, this invention applies to both Herpes Simplex Virus 1 and Herpes Simplex Virus 2.

FIG. 1 is the DNA and amino acid sequence for HSV 1 protease (SEQ. ID. NOS.:1 and 2). The amino acids confirmed by peptide sequencing are double-underlined. N-termini of the 32 kDa and 27 kDa proteins were identical, beginning with the Ala in position 2.

As used throughout the specification and claims, the following definitions shall apply:

Protease--When used without the descriptor "precursor" or "mature" refers to either form of protease.

Precursor protease--A protease from a Herpes Simplex 1 or 2 Virus which undergoes autolytic cleavage, producing as one of the cleavage products a mature viral protease.

Mature protease--The active form of viral protease which results from the autocatalytic cleavage of a precursor protease.

Autoprocessing--The autocatalytic cleavage of a precursor protease to produce a mature protease.

Autoprocessing conditions--Any condition which increases the ability of the precursor protein to undergo autocatalytic cleavage.

Substantially similar protease--Protease which has substantially the same proteolytic activity as the proteases of this invention and which is encoded by a DNA which hybridizes to the DNA of FIG. 1 under stringent conditions.

Stringent hybridizing conditions--Any of the hybridization conditions known in the an and so considered. See, for example those described in Maniatis et at., 1982, Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory pages 387-389, hereby incorporated by reference. An example of stringent hybridization conditions is incubation in 4 X SSC at 65.degree. C., followed by washing in 0.1 X SSC.

Purified protease--Viral protease which has a minimum purity of 90% by weight.

To produce the purified mature protease of this invention, the gene for the precursor protease, either from HSV-1 or HSV-2, may be introduced into a host cell. The full length protein encoded by the HSV-1 protease gene has 635 amino acids, but the catalytic activity resides in the first 247 amino acids (the first MET is cleaved off, leaving an active 246-residue mature protease). Therefore, it is not necessary to transform a host cell with the entire DNA which encodes all 625 amino acids. It is only required that the host cell be transformed with a gene fragment which will result in the production of the active portion. One may thus transform a host cell with DNA encoding the first 247 amino acids, or one may transform a host cell with a DNA encoding a longer portion, and allow the protease to exert its autocatalytic ability to convert the longer portion into the active mature protease. In a preferred embodiment, host cells are transformed with a segment which encodes at least 306 amino acids, which can self-cleave.

A preferred host cell is a bacterial host cell such as Escherichia coli, but any cell which is capable of producing Herpes Simplex Virus protease, including insect cells may be used. The gene for the precursor protease may be part of a vector system, such as a plasmid or baculovirus, wherein the nucleic acid encoding the precursor protease is under the control of a promoter, and/or other desired regulatory elements such as enhancers or the like. Suitable vectors are well known in the art and there are numerous commercially available vectors. One preferred vector for such expression is pET 3C, described in more detail in the Examples. The protease gene (or gene fragment) is inserted into the vector and the vector is introduced into the host cells using known techniques. The host cells are cultured under conditions suitable to their expression of the introduced gene; such culture conditions are also well known in the art.

In a preferred embodiment, the expression product of the Herpes Simplex Viral 1 protease open reading frame has 306 amino acids, as seen in FIG. 1. However, in host cells such as E. coli, the initial N-formyl methionine is removed, so the first amino acid of the resulting precursor protease begins with amino acid #2. Under appropriate autoprocessing conditions, the precursor protease self-cleaves between amino acid residues 247 and 248 to produce two cleavage products: the mature protease of 246 amino acids and a 59 amino acid fragment.

The 306 amino acid protein depicted in FIG. 1 has an identical amino acid sequence to the first 247 amino acids of the corresponding region of the HSV-1 strain 17 protease (Genebank accession number X14112 locus H21CG). At codons 300 and 301, two nucleotides base changes alter two residues of the expression construct relative to strain 17 (L300S, S301P). However, after autoprocessing (below), these residues are not retained in the mature HSV protease.

The precursor form of the protease is expressed and accumulates in the host cell. In a preferred embodiment, isopropyl-.beta.,D-thiogalactopyranoside (IPTG) may be used to induce protein synthesis from the plasmid-encoded gene, particularly when the transcription of the gene is driven by a T7 promoter. In one embodiment, the expressed precursor protease is 306 amino acids long and has an apparent molecular mass of 32 kDa. Host cells are harvested, and may be lysed. One such way to lyse is treatment with hen egg white lysozyme in pH 7.5 50 mM TRIS HCl, followed by treatment with DNAse. Soluble and insoluble portions of the lysate are separated by centrifugation. While the precursor protease may be found in both the solubilized or non-solubilized fraction, it is preferred to further purify the protease from the soluble fraction in order to avoid the uncertainties encountered when enzymes are exposed to solubilizing agents such as urea or guanidine hydrochloride.

Next, the precursor protein is separated from the non-protease constituents of the host cells culture by cation exchange chromatography. In this step, the soluble fraction is applied to a cation exchange column resin, preferably to a "Fractogel-SO.sub.3 .RTM." column, commercially available from E. Merck. Fractogel-SO.sub.3 .RTM. is a tentacle cation exchanger consisting of an insoluble matrix, copolymerized from oligoethyleneglycol, glycidylmethacrylate and pentaerythroldimethacrylate, to which are grafted polymerized chains of acrylamide derivatives that are approximately 15 to 50 units in length. Most of the host cell proteins do not adhere to the cation exchange resin, but the precursor and mature forms of the protease do. It was surprising and unexpected that the protease would bind to this cation exchange resin, as the isoelectric points for the precursor and mature forms of the protease enzyme are 6.3 and 5.8, while the pH used in the Fractogel.RTM. cation exchange column step is higher than the enzyme isoelectric point. In a preferred embodiment the pH of the cation exchange column is between 6.5 and 8.5, more preferably, the pH is 7.0. The precursor protease is eluted from the column in partially purified form with a gradient of increasing salt concentration.

In one embodiment of this invention, the protease is further purified by hydrophobic interaction chromatography (HIC) after the addition of sodium chloride to 2M and application to the HIC column. Elution from the HIC column is achieved by application of a decreasing salt gradient to the HIC column. The majority of the protease obtained at this stage of purification is precursor protease, although a small amount of mature protease and trace amounts of host proteins may also be present. This mixture, which is essentially free from associated viral proteins forms another aspect of this invention.

The precursor protease is then subjected to autoprocessing conditions which favor its autocatalytic conversion to mature form. One preferred autoprocessing condition includes concentrating the precursor protease to a concentration of at least about 1 mg/ml, although in some instances somewhat lower concentrations will also exhibit autocatalytic activity. Another preferred condition for autoprocessing includes changing the solvent in order to increase the activity of the precursor protease or itself. One such change to the solvent includes the addition of salts. Suitable salts include the potassium, sodium and ammonium salts of: sulfate, citrate, succinate, isocitrate, cis-aconitate, fluoride, and any other anion with multiple charges. A preferred salt is sodium citrate. The concentration of the salt may be between about 0.5M to 2M, more preferably 1M.

After autoprocessing the 32 kDa protein band observed on SDS page gels disappears and two new bands, with apparent molecular masses of 27 kDa and 6 kDa appear.

These proteins may be reintroduced into a HIC column and washed with salt to recover essentially pure 27 kDa protease. The autoprocessing procedure results in a change in elution position of the enzyme on the second HIC step, freeing it from impurities which still elute at the previous position.

Alternatively, in another embodiment of this invention, the precursor protease may be autoprocessed after the first chromatography step (using Fractogel.RTM.-SO.sub.3). The resulting mature form of the protease is re-chromatographed using Fractogel.RTM.-SO.sub.3 and elutes at a different position than the precursor protease. The change in elution position allows the separation of the enzyme from impurities which co-elute with the precursor protease.

The N-termini of the 32 kDa protease and its 27 kDa and 6 kDa cleavage product were sequenced and found to be in agreement with the amino acid sequence of FIG. 1 (SEQ. ID. NO.:2) beginning with the Ala at position 2. The N-terminal sequence of the 59 amino acid 6 kDa sequence was found to be consistent with autocleavage between amino acids 247 and 248 of the precursor.

HSV-1 protease has been shown to cleave specific alanineserine bonds in its substrate protein (DiIanni et al., 1993, J. Biol. Chem. 268:2048-2051). Therefore, the cloned and purified protease was tested to determine it it also showed this activity. Peptides having the two known cleavage sites were made and incubated with the purified 27 kDa protein. These peptides were: ##STR1##

After incubation with the enzyme, two new peptide peaks were obtained in the HPLC analysis for each substrate incubated with the enzyme. The peptide representing the cleavage site proximal to the N-terminus of the 635 residue HSV-1 protease His-Thr-Tyr-Leu-Gln-Ala-Ser-Glu-Lys-Phe-Lys-Met-amide (SEQ. ID. No.:3) was cleaved by the 27 kDa mature protein at the expected site between the alanine and serine to yield two six amino acid peptides. Likewise, the peptide with the sequence of the cleavage site proximal to the C-terminus of the 635 residue HSV-1 preprotease, Ala-Leu-Val-Asn-Ala-Ser-Ser-Ala-Ala-His-Val-Asp-Val-Asp-Thr-Ala-Arg-Ala-Al a-Asp (SEQ. ID. NO.:4) was cleaved between the Alanine-Serine to yield a five amino acid peptide and a 15 amino acid peptide. The identities of the peptide products were confirmed to be as predicted with amino acid analysis. These products also co-eluted with synthesized peptide product standards. No other cleavage products were observed in the HPLC chromatograms.

While the peptide cleavage reactions observed were consistent with the specificity of HSV-1 protease toward protein substrates, the possibility of a small amount of E. coli protease contributing to the reaction needed to be considered. The hydrolytic activity of the partially purified 306-residue precursor with the His-Thr-Tyr-Leu-Gln-Ala-Ser-Glu-Lys-Phe-Lys-Met-amide (SEQ. ID. NO.:3) substrate was found to be 5 nmol/hr/mg and the corresponding value for the mature 246-residue protease was 105 nmol/hr/mg under assay conditions detailed in the Examples. Thus, conversion to the mature form increased the activity of the protease and was unlikely to be due to contaminating bacterial proteases. The fractions from the HIC columns that contained peptide cleavage activity also contained either the 32 kDa protein or the 27 kDa protein. The shift in the elution position of the protease upon conversion of the 32 kDa protein to the 27 kDa form coincided with the shift in elution position of the peak peptide cleavage activity. Therefore, cleavage was not due to bacterial contamination.

The mature protease may be used to identify potential viral protease inhibitors.

This invention may be further illustrated by the following non-limiting Examples.

The bacterial expression plasmid coding for the expression of the first 306 amino acids of the HSV ICP35 open reading frame was obtained by polymerase chain reaction with a purified restriction fragment of viral genomic DNA (KOS strain) as a template. Primers were designed on the basis of the HSV-1 strain 17 sequence (McGeoch, D. J. et al., 1988 J. Gen. Virol. 69:1531-1574) which is hereby incorporated by reference. The amplification employed 7-deaza-GTP.

The DNA encoding the 306 amino acid fragment was inserted into the NdeI and BamHI sites of the expression vector pET 3C, a known and available vector, described in Studier, et al., 1990 Methods Enzyomol. 185:60-89, which uses the T7 promoter. This vector was designated pT7-306. Expression clones were obtained by transformation of BL21(DE3)pLysS cells. Cells were grown at 37.degree. C. in shaker flasks in Luria-Bertani medium with 75 .mu.g/ml ampicillin and 50 .mu.g/ml chloramphenicol to an optical density of 0.5 (600 nm). HSV protease synthesis was induced by the addition of isopropyl-.beta.,D-thiogalacto-pyranoside (IPTG) to 0.4 mM, and incubation continued for 2 hours before harvesting cells by centrifugation.

Double stranded DNA was denatured with 0.2M NaOH and sequenced with the dideoxy chain termination method and Sequenase v. 2.0 (United States Biochemical, Cleveland, Ohio). Primers were based upon the Glasglow Herpes Simplex Virus-1 strain 17 sequence and were obtained from Midland Certified Reagent Co., Midland, Tex. Compression of products on gel electrophoresis were resolved by the substitution of dGTP with either 7-deaza-dGTP or dITP. All nucleotide assignments were made on the basis of agreement from both strands of DNA. In most cases the assignment was also confirmed with multiple overlaps from different primers.

Initial detection of the expressed HSV protease in both cell lysates and preliminary fractionates was by immunoblot detection of the antigen in sodium dodecylsulfate polyacrylamide electrophoresis (SDS PAGE) gels. Samples were transferred to PVDF membranes and probed with rabbit antisera raised against the peptide Arg-His-Leu-Ser-Pro-Ala-Ser-Arg-Glu-Gly-Ala-Arg-Arg-Leu (SEQ. ID. NO.:5), corresponding to amino acid residues 176-189 in FIG. 1. Antisera was kindly provided by Dr. A. Conley, Merck & Co., Inc. Following the identification of a Coomassie brilliant blue staining band unique to induced cells as the protein of interest, Coomassie brilliant blue staining of SDS gels was used to identify the 306 amino acid precursor in subsequent purification steps.

Cell pellets were suspended in cold 50 mM Tris HCl, pH 7.5 plus 10% glycerol and 1 mM dithiothreitol (DTT). Hen egg white lysozyme was added to a final concentration of 0.1%. Following the viscosity increase accompanying lysis, MgCl.sub.2 and DNase (bovine spleen) were added to final concentrations of 5 mM and 0.01%, respectively. After 30 minutes on ice, the lysate was centrifuged 50 minutes at 4.degree. C. and 15,000 rpm in a Sorvall SS34 rotor.

The supernatant from the centrifuged mixture (10 ml) was diluted 5-fold with 25 mM sodium phosphate, pH 7.0, containing 10% glycerol and 1 mM DTT and applied to a 10.times.100 mm column of E. Merck Fractogel.RTM.-SO.sub.3 which was equilibrated in the above buffer. All fractionation procedures took place at 4.degree. C. Following a 5 ml wash with equilibration buffer, a gradient to 1M NaCl in equilibration buffer was applied and ten 5 ml fractions were collected. Following analysis of fractions with SDS PAGE (14% acrylamide), the most highly purified fractions were combined with 5 ml of 50 mM TrisHCl, pH 7.5 10% glycerol, 1 mM DTT. NaCl was added to a final concentration of 2M. The sample was applied to a 7.75.times.100 Baker Hi-propyl HIC column equilibrated with 50 mM Tris-HCl, pH 7.5, 10% glycerol, 2M NaCl, and 1 mM DTT. Following wash with the HIC equilibrium buffer, a gradient from equilibrium buffer to the same buffer without the added NaCl was run (50 ml), and fractions were collected. Fractions containing the majority of the HSV 32 kDa band were pooled, and concentrated on Centricon-10 filters (Amicon, Inc., Beverly, Mass.). The salt concentration of the sample was reduced 5-fold by dilution with HIC buffer lacking NaCl. The final sample was reconcentrated and left for 16 hr at 4.degree. C. for autoprocessing. NaCl was added to a final concentration of 2M and the sample was reapplied and eluted from the Baker HIC column as described above.

The protocol in Example 3 was followed except that autoprocessing was induced after the first chromatography step. The first Fractogel.RTM.-SO.sub.3 column was loaded as in Example 3, and eluted with a linear gradient of 0 to 0.5M sodium citrate. Peak fractions from the elution of the Fractogel.RTM.-SO.sub.3 were pooled and sodium citrate was added to a final concentration of 0.5M. The sample was allowed to autoprocess for 3 hr at 4.degree. C. The salt concentration was then lowered by passing the sample over PD10 columns (Bio-Rad Inc., Hercules, Calif.) which had been equilibrated in 25 mM sodium phosphate, pH 7.0 containing 10% glycerol, 1 mM EDTA, 1 mM DTT. The desalted sample was applied to a fresh Fractogel.RTM.-SO.sub.3 column as before, and eluted with a linear gradient of sodium citrate from 0 to 0.5M, and fractions collected. Analysis with Coomassie brilliant blue staining of SDS gels (14% acrylamide), revealed that the fractions containing the peak of the mature HSV protease contained a single 27 kDa band, the HSV protease.

The amino acid sequences surrounding the known cleavage sites of the 635 amino acid precursor (at residues 247/248 and 610/611) were used for the peptide substrates His-Thr-Tyr-Leu-Gln-Ala-Ser-Glu-Lys-Phe-Lys-Met-amide (SEQ. ID. NO.:3) and Ala-Leu-Val-Asn-Ala-Ser-Ser-Ala-Ala-His-Val-Asp-Val-Asp-Thr-Ala-Arg-Ala-Al a-Asp (SEQ. ID. NO.:4). Reactions were initiated by the addition of enzyme aliquots to substrate at 30.degree. C, incubated for the times indicated and quenched by the addition of 5% H.sub.3 PO.sub.4. Separation of cleaved peptides front substrate was achieved by HPLC on either 15 cm or 5 cm Vydac C-18 reverse phase columns with 0.1% H.sub.3 PO.sub.4 and acetonitrile as the aqueous and organic phases, respectively. Routine assays during purification were performed with 4 mg/ml His-Thr-Tyr-Leu-Gln-Ala-Ser-Glu-Lys-Phe-Lys-Met-amide (SEQ. ID. NO.:3) in pH 7.5 Tris HCl, 10% glycerol and 1 mM EDTA for 0.5 to 4 hrs. Additional assays for kinetic characterization of peptide cleavage were performed in a mixture of 52 mM MES, 52 mM TAPSO and 100 mM diethanolamine, pH 7.5 with 25% glycerol, 1 mM EDTA, 1 mM dithiothreitol and 0.1% bovine serum albumin added.

The protein and peptide samples were hydrolyzed with 6N constant boiling HCl in sealed, evacuated vials at 100.degree. C. for 20 hrs. Amino acids were analyzed and quantified on a Beckman 6300 analyzer. N-terminal sequence analysis were performed with Edman degradation on an Applied Biosystems Model 470A gas-phase sequencer equipped with a Model 120A PTH analyzer.