Structural Proteins of Herpes Simplex Virus

Transcription

1 J. gen. Virol. (I97I), Io, I63-t7I i63 Printed in Great Britain Structural Proteins of Herpes Simplex Virus By D. J. ROBINSON ~ AND D. H. WATSON Department of Virology, University of Birmingham, Medical School, Metchley Park Road, Birmingham B I5 2 TJ (Accepted 13 October 197 ) SUMMARY Herpes simplex virus, grown in BHKzI cells, was extensively purified by fluorocarbon treatment, ultracentrifugation, sucrose gradient centrifugation and chromatography on calcium phosphate. The purity of the product was assessed by immunodiffusion, electrophoresis in polyacrylamide gels and removal of added radioactive host material from the virus. Purified virus was disrupted with sodium dodecyl sulphate, urea and dithiothreitol, and at least eight polypeptide components were separated by electrophoresis in polyacrylamide gel. INTRODUCTION Several virus specified proteins have been identified in cells infected with herpes simplex virus (Watson et al. 1966). One of these can be identified as a structural component of the virus, since 'monoprecipitin' antiserum to it neutralizes virus infectivity (Watson & Wildy, 1969). Other antigens could be classified as structural by deriving them from disrupted virus particles, as has been done for example with adenovirus (Wilcox & Ginsberg, 1963), or as non-structural by adsorbing antisera with disrupted virus particles to leave antibodies which react only with presumptive non-structural antigens, as has been done with influenza virus (Dimmock, 1969). Such immunological procedures could also be combined with methods of polyacrylamide gel electrophoresis (Summers, Maizel & Darnell, I965) by showing that the antigens precipitated by antiserum contain polypeptides differing from the structural polypeptides of the virus, as has been done with influenza virus (Dimmock & Watson, 1969). All these methods involve the preparation of highly purified virus, a goal which has proved difficult to achieve with herpes virus. Thus Roizman (1969) has noted ' that none of the published procedures.., satisfactorily separated an artificial mixture of unlabelled virus and [C]amino acid-labelled debris from uninfected cells' and this difficulty has undoubtedly led to the neglect of the chemistry of the viruses of the herpes group noted by Green (1969). We report here a method for purification of herpes virus which is satisfactory judged by a number of criteria, including that quoted by Roizman (I969). We further describe the pattern of structural polypeptides in polyacrylamide gel electrophoresis of such purified preparations for, although there have been a number of reports on such patterns for both herpes simplex (Spear & Roizman, 1968; Spear, Keller & Roizman, 197o; Olshevsky & Becker, I97O) and pseudorabies viruses (Shimono, Ben-Porat & Kaplan, I969; Stevens, Kado-Boll & Haven, 1969), none of them have presented adequate criteria for the purity of the virus preparations used. Roizman's comment makes this a mandatory requirement in work of this kind. * Present address: Scottish Horticultural Research Institute, Invergowrie, Dundee, Scotland. II~2

2 I64 D. J. ROBINSON AND D. H. WATSON Virus production METHODS The HrEM strain of herpes simplex virus was grown in BHK2I (C I3)cells. Confluent cell monolayers in 8o oz. Winchester bottles were each infected with 2 x lo 7 p.f.u, of virus in 20 ml. modified Eagle's medium (Vantsis & Wildy, ~962) containing Io To calf serum and IO ~o tryptose phosphate broth (ETC). After adsorption of the virus for 2 hr at 32, a further 2oo ml. ETC was added and incubation was continued until widespread cytopathic effect was observed (usually after 2 to 3 days). The cells were scraped off the glass into a small volume of ETC, sedimented at IOOO rev./min, for IO min., and washed once in phosphatebuffered saline (PBS). The resulting pellet was resuspended in distilled water and the cells disrupted with an ultrasonic probe (MSE Ltd, Crawley, Sussex). Radioactively labelled virus was prepared by infecting BHK cells in suspension with virus at a multiplicity of at least 2o p.f.u./cell. After I hr at 37, the cells were washed with aminoacid-free Eagle's medium, and 5 x lo s were dispensed into each Winchester bottle. Eagle's medium (ioo ml.) containing theusual amounts of glutamine, arginine and histidine, -3%- of the normal concentration of all other amino acids and o.6 ~ calf serum was added to each bottle together with 8oo/~c of a mixture of [3H]amino acids ([ZH]De-valine, 393 mc/m-mole; 2,3-[~H]L-phenylalanine, I c/m-mole; 4,5-[aH]L-lysine, 5IO mc/m-mole; 4,5-[~H]L-leucine, I c/m-mole, Radiochemical Centre, Amersham). Virus growth was permitted for 2o hr at 32o, and the cells were harvested and disrupted as described above. The virus from three bottles of labelled cells was mixed with the virus from L 5 unlabelled bottles after disruption of the cells. The yields and particle/infectivity ratios of labelled and unlabelled virus were equal. Purification All operations were performed at 4. Stage I. The crude virus suspension was homogenized in an MSE homogenizer for about IO sec. with ½ vol. of trifluorotrichloroethane (Arcton I~3; I.C.L Ltd.). After low-speed centrifugation, the upper aqueous layer was removed and further centrifuged for 45 min. at ~5,ooo rev./min. (I4,ooog) in the AR 6o rotor of the Christ Omega II ultracentrifuge. (All accelerating forces quoted are those at the middle of the tube). The supernatant fluid was discarded and the pellet resuspended in a small volume of o.o2 M-phosphate buffer, ph 7"4. The resuspended virus was treated briefly with the ultrasonic drill and clarified by centrifugation at low speed. The resulting suspension was adjusted with o.o2 M-phosphate buffer to contain not more than 4 x io x2 particles/ml, and incubated overnight at 4 with 50 #g./ml. ribonuclease (ex-bovine pancreas 5 x crystallized; Koch-Light), 5o/zg./ml. deoxyribonuclease (ex-bovine pancreas, salt-free; Koch-Light), and 3 mm-mgci2. Stage 2. Rate zonal centrifugation was performed on 5 to 4 ~ (w/v) sucrose gradients in 3 ml. cellulose nitrate tubes. Fourml. virus suspension was layered on to a 26 ml. gradient and centrifuged at I2,6oo rev./min. (i6,ooog) for 9 min. at 4 in the Christ SW27 rotor. Gradients were harvested using a hypodermic syringe inserted through the wall of the tube. Routinely, three fractions were taken, the top 12 ml., the middle 12 ml., and the bottom 6 ml. containing the pellet. The virus band was in the middle fraction. No advantage in terms of virus purity was gained by taking more fractions of the gradient, since only low protein concentrations were found in the middle region of the gradient. Visible bands were sometimes seen in this region, but these bore no relation to the virus peak. Up to 5o ~ of the input virus was found in the pellet, together with a large amount of protein. Omission of the nuclease treatment resulted in 9 ~ of the virus sedimenting to the pellet.

3 Proteins of herpes virus 165 After dilution to 3o ml. with distilled water, the virus was sedimented at 25,ooo rev./min. (6o,ooog) for 6o min. in the Christ SW 27 rotor. The resulting virus pellet was resuspended in a small volume of o.o2 M-phosphate buffer, ph 7"4. Stage 3- Calcium phosphate was prepared by the method of Burness 0967) and packed into a 2o ~.5 cm. column. The column was washed with o'oo5 M-phosphate buffer, ph 7"0 and 3 x io 12 to 3 ~o13 particles of virus from sucrose gradients were then applied. The column was eluted stepwise with o.oo 5 M; 0"I M; O'2 M; O" 3 M- and o. 4 M-phosphate buffer, ph 7"o. A flow-rate of I.o to 1.5 ml./min, was maintained. Fractions of approximately 6 ml. were collected and the E28o determined in a Unicam SP5oo spectrophotometer. The peak containing virus eluted with o-2 M-phosphate was pooled and concentrated by ultracentrifugation as described above. Chromatography columns, and all glass used for handling highly purified virus, were silicone-coated. Labelling of uninfected cells Eighty oz. bottles were seeded with 6 x ~o 7 BHK cells in 4oo ml. ETC and incubated for 3 days at 37. The medium was then removed and replaced with 22o ml. Eagle's medium containing ~ the concentration of amino acids, supplemented by 5 ~ calf serum and 2o #c [l~c]chlorella hydrolysate (specific activity 52 me/m-atom carbon, Radiochemical Centre, Amersham, Buckinghamshire). Incubation was continued for a further 24 hr at 37. When the cells had become fully confluent they were scraped from the glass, washed three times in PBS, resuspended in distilled water and disrupted by ultrasonic vibration. Electrophoresis in polyacrylamide gel Analytical electrophoresis was done using the method of Davis 0964) modified as previously described (Watson, r969). The modification of this method for use with sodium dodecyl sulphate (SDS), the preparation of samples, and the slicing and elution of gels containing radioactive proteins were described by Dimmock & Watson 0969). Coomassie Brilliant Blue was superior to amido black for staining gels containing SDS. Radioactivity was counted by transferring the eluates to Whatman GF/C glass fibre discs, which were dried and counted in a Packard Tri-Carb Liquid Scintillation Counter, using a toluenebased scintillation fluid. Serology Immunodiffusion was done in I ~ agar gel using the template micro-technique of Crowle 0958). The preparation of high-titre antisera against RK r3 cells infected with herpes simplex, and against uninfected BHK 2I cells was described by Watson et al. (~966.) Other methods Particle counting was done by the 'loop drop' method of Watson, Russell & Wildy (r 963). Infectivity was assayed by the method of Russell 0962). Protein concentration was determined by the method of Lowry et al. (I95r). Purification RESULTS The preliminary stages (I and 2) of purification typically achieved about 5o-fold concentration, tenfold purification and Io to 2o ~ recovery of virus particles (Table I). On the calcium phosphate columns most virus (up to 5 ~ of input) eluted with o-2 M-phosphate, although there were further minor peaks at 0"3 M- and o'4 M-phosphate. Virus in these

4 I66 D. J. ROBINSON AND D. H. WATSON Step Volume (ml.) Original sonic- 153 treated cells Stage I, after 15 Arcton treatment and ultracentrifugation Stage 2, after 3 sucrose gradient sedimentation Stage 3, after o'5 calcium phosphate chromatography Table i. Purification of herpes simplex virus and removal of 14C-labelled host-cell material Protein/ particle Recovery Virus Protein ratio of particles/ cone. (,ttg./io 1 particles* Purificaml. (mg./ml.) particles) (~) tiont 'I I00 I00 I Counts/ min./ml. 65,7oo Counts/ min./io 10 particles 4"5 x IO 12 I4" I02, '4 x IO is 5" "5 II'4 I9,48o 30 7"8 x Io 12 3"I 3" '7 * Total particles present after each stage as a percentage of those in the original material. t Calculated from protein/particle ratio. 730 Fig. I. Virus at different stages of purification tested in immunodiffusion with antiserum against herpes-infected RK I3 cells. A, antiserum; o, disrupted cells; L stage I; after Arcton treatment and ultracentrifugation; 2, stage 2; after sucrose gradient sedimentation; 3, stage 3, after calcium phosphate chromatography. peaks did not differ either in appearance in the electron microscope or in particle/infectivity ratios. Under the conditions of growth, 5 to I0 ~ of the original yield of virus particles possessed envelopes, and significantly less than ~ ~ of the particles in any of these peaks was enveloped. Only virus eluting at o.2 M-phosphate was used for further study. The concentrated virus obtained by ultracentrifugation of the 0.2 M eluate had a protein content of 3"9 #g./io1 particles. In zo experiments protein contents of 1"4 to 7"8 #g./io 1 particles were obtained. The mean value was 4"2/~g./IO 1 particles.

5 Proteins of herpes virus 167 The final preparation gave no lines in immunodiffusion tests using antisera either against RK ~ 3 cells infected with herpes virus or against uninfected BHK cells, suggesting that the virus was virtually free of diffusible antigens. The original preparation when tested at the same particle concentration gave multiple precipitin bands (Fig. *). No stained bands were observed when a sample of virus containing ioo/~g, protein was electrophoresed in polyacrylamide gel in the absence of sodium dodecyl sulphate, although ~ oo #g. of crude material showed multiple bands. The particle/infectivity ratio of purified virus was 3 to 6 times greater than that for the crude virus. The final virus yield did not therefore represent a selected class of particles of low infectivity. I I l I I I l I I A B C D E - o Fraction Fig. 2. Electrophoresis of [all]virus proteins in 7 ~ acrylamide gel. RB, refractile band of free sodium dodecyl sulphate; PR, phenol red; Q---@, labelled virus; O--O, labelled mock-infected cells. Removal of host-cell material in purification. To assess further the efficiency of the purification procedure, a crude extract of infected cells from zo 80 oz. bottles was mixed with a corresponding extract of three bottles of uninfected cells, labelled with [l~c]chlorella hydrolysate. Virus was then purified from this mixture by the above method, and at each stage samples were withdrawn for determination of radioactivity. The activity at each stage was expressed as counts/min./io 1 virus particles (last two columns of Table 0- Using this test the overall purification from crude virus was about moo-fold.

6 I68 D. J. ROBINSON AND D. H. WATSON ~ kl i~;i,.. : + : Fig. 3. Electrophoresis of virus proteins in 7 % acrylamide gel; stained pattern. Scale of fractions corresponds to those of Fig. 2. If it is assumed that host-cell protein from infected cells was removed as efficiently as the labelled uninfected cell protein, the residual activity gives an indication of the total amount of contaminating protein in the final purified product. Thus, if we assume that the Ioo #g. protein/to l particles in the crude preparation was wholly host protein, 1_ of this amount or o.~/~g, will have been associated with Io 1 particles in the purified preparation, that is about 2"5 % of the 3"9 #g- protein/ml purified virus particles. Analysis of structural polypeptides Samples of purified virus labelled with [3H]amino acids for electrophoresis on SDSpolyacrylamide gels contained at least 2o,ooo counts/min, representing ~ to z x ~o n virus particles. The basic pattern of labelled bands found in 7 % acrylamide gels was obtained five times, using three separate batches of virus (Fig. 2). The complex band A always contained about 5o % of the recovered radioactivity, and appeared to consist of at least two major components. In addition, the shoulder on the anodal side of band A could occasionally be resolved as a minor peak. Similarly, peak D often appeared to contain two

7 Proteins of herpes virus poorly resolved components. The same pattern was observed in stained gels (Fig. 3). Virtually the same pattern was obtained using a continuous buffer system (Summers et al. I965), although the bands were less sharp and resolution consequently lower. To check whether any of the observed peaks could have been due to contamination with a host-cell protein, mock-infected cells were labelled with [3H]amino acids, mixed with unlabelled infected cells, and the virus purified in the usual way. A sample containing I-4 io 11 particles and I I48 counts/min, was electrophoresed on a 7 % gel (Fig. 2). The result I I I I I I I I I I c E g u RB PR I Fraction Fig. 4- Electrophoresis of [3H]virus proteins in 14 ~oo acrylamide gel. RB, refractile band; PR, phenol red. excluded the possibility that any of the components observed were host-cell proteins specifically retained during the purification. However, this does not imply that all the proteins observed were specified by the virus DNA. Dimmock & Watson (I969) found that the front running peak coincident with the refractile band of free sodium dodecyl sulphate was an artifact of the system. To test whether this was also true for herpes virus proteins, electrophoresis was carried out in I4 % gels (Fig. 4)- Bovine serum albumin and ovalbumin, run as markers in parallel gels of the same batch, permitted peaks in the two gel strengths to be correlated. While the peaks usually observed in 7 % gels were all retained in the first 15 fractions, and all the counts at refractile band were dispersed, two new minor peaks were observed (G and H in Fig. 4). DISCUSSION It is clear that criteria for purity can only be negative; that is, they can only demonstrate impurity rather than purity. The criteria we have used here do not disprove that virus of high purity was obtained. The protein/particle ratio approximates to that calculated from

8 I7O D.J. ROBINSON AND D. H. WATSON the dimensions of the particle assuming it to be composed principally of protein. The protein assay may not give absolute values but is useful in giving an order of magnitude. We feel it is valuable to list our criteria fully, if only to serve as a yardstick for future work. We have shown by electron microscopy that our preparations of highly purified virus contained a proportion of enveloped particles too small to be accurately estimated, but certainly less than I ~. (We are unable to trace the exact cause of the selective removal of enveloped particles. There was no separation of naked and enveloped particles in the sucrose gradients, although recentrifugation of stage 2 virus on higher-resolution gradients showed some enrichment of enveloped particles on the faster running side of the virus peak. Combination of data from many experiments suggests a gradual loss of enveloped particles at all stages but a more pronounced loss on calcium phosphate chromatography. We have been unable to elute enveloped particles in any molarity of eluting buffer up to M. A higher proportion of enveloped particles in the original preparation would be required to investigate this more thoroughly). The patterns obtained in polyacrylamide electrophoresis in the presence of SDS and urea thus reflect the polypeptide composition of naked particles. The results suggest nucleocapsids of herpes simplex virus are composed of at least eight polypeptides. Estimates of the number of structural virus polypeptides by counting peaks in polyacrylamide electrophoresis are subject to a number of possible errors. The purity of the virus is critical and small amounts of an impurity will appear as a peak on electrophoresis. Thus, when virus purified only to stage 2 was used in electrophoresis, the main features of the pattern were similar, although a high and irregular level of counts between the peaks obscured the details. Spurious peaks could also be due to protein aggregation. On the other hand, two polypeptides of closely similar molecular weight might be only poorly resolved, if at all, by this technique, and virus components might not be detected. There may therefore be up to three components in peak A and two in peak D, making it possible that there are as many as ~ i polypeptides in the naked particle. The virus obtained should be sufficiently pure to allow progress in the discrimination of structural and non-structural antigens along the lines indicated in the Introduction. We wish to thank Mrs Christine Wallis and Messrs R. Wilkinson and P. Shepherd for their skilful assistance with the tissue culture. We are also grateful to Dr A. Buchan and Professor P. Wildy for many valuable discussions. Thanks are due to Messrs G. D. Searle and Co. Ltd. for the loan of the spectrophotometer. D. J. Robinson was supported by a grant from the Medical Research Council, and during part of this work D. H. Watson was a member of the Medical Research Council Virus Research Group. REFERENCES BURNESS, g. V. ~. (I967). Separation of plaque-type variants of encephalomyocarditis virus by chromatography on calcium phosphate. Journal of Virology x, 3o8. CROWEr, A. J. 0958). A simplified micro double-diffusion agar precipitin technique. Journal of Laboratory and Clinical Medicine 5z, 784. DAVIS, B. J. (I964). Disc electrophoresis. II. Method and application to serum proteins. Annals of the New York Academy of Sciences xzx, 4o4. DIMMOCK, N. J. 0969). New virus-specific antigens in ceils infected with influenza virus. Virology 39, 224. DIMMOCK, N. J. & WATSON, O. n. 0969). Proteins specified by influenza virus in infected cells: analysis by polyacrylamide gel electrophoresis of antigens not present in the virus particle. Journal of General Virology 5, 499. CREEN, M. (I969). Chemical composition of animal viruses. In The Biochemistry of Viruses. Ed. by H. B. Levy. New York: Marcel Dekker.

9 Proteins of herpes virus 17 I LOWRY, O. H., ROSEBROUGH, N. J., FARR, A. L. & RANDALL, R. J. (I951). Protein measurement with the Folin phenol reagent. Journal of Biological Chemistry x93, 265. OLSHEVSKY, U. & BECKER, Y. (I970). Synthesis of herpes simplex virus structural proteins in arginine deprived cells. Nature, London 226, 851. ROIZMAN, B. (I 969). The herpes viruses--a biochemical definition of the group. Current Topics in Microbiology and Immunology 49, I. RUSSELL, W. C. (I962). A sensitive and precise plaque assay for herpes virus. Nature, London x95, lo28. SHIMONO, H., BEN-PORAT, T. & KAPLAN, A. S. (t969)- Synthesis of proteins in cells infected with herpesvirus. I. Structural viral proteins. Virology 37, 49. SPEAR, P. G., KELLER, J. M. & ROIZMAN, B. (I970). Proteins specified by herpes simplex virus. II. Viral glycoproteins associated with cellular membranes. Journal of Virology 5, 123. SPEAR, P. G. & ROIZMAN, B. 0968). The proteins specified by herpes simplex virus. I. Time of synthesis, transfer into nuclei, and properties of proteins made in productively infected cells. Virology 36, 545. STEVENS, J. G., KADO-BOLE, G. J. & HAVEN, C. B. (t969). Changes in nuclear basic proteins during pseudorabies virus infection. Journal of Virology 3, 490. SUMMERS, D. F., MAIZEL, J.V. & DARNEEL, J.E. (1965). Evidence for virus-specific noncapsid proteins in poliovirus-infected HeLP cells. Proceedings of the National Academy of Sciences of the United States of America 54, 5o5. VANTSIS, J. T. & WILDY, P. (I962). Interaction of herpes virus and HeLP cells: comparison of cell killing and infective center formation. Virology 17, 225. WATSON, O. H. (I969). The separation of herpes virus-specific antigens by polyacrylamide gel electrophoresis. Journal of General Virology 4, I51. WATSON, D. H., RUSSELL, W. C. & WlLDY, P. (I963). Electron microscopic particle counts on herpes virus using the phosphotungstate negative staining technique. Virology 19, 25o. WATSON, D. H., SHEDDON, W. I. H., ELLIOTT, A., TETSUKA, T., WILDY, P., BOURGAUX-RAMOISY, D. & GOLD, E. (t966). Virus specific antigens in mammalian cells infected with herpes simplex virus. Immunology ix, 399. WATSON, D. H. & W~LDY, P. (I969). The preparation of 'monoprecipitin' antisera to herpes virus specific antigens. Journal of General Virology 4, 163. WILCOX, W. C. & ONSBER6, U. S. 0963). Structure of type 5 adenovirus. I. Antigenic relationship of virusstructural proteins to virus-specific soluble antigens from infected cells. Journal of Experimental Medicine II8, 2. (Received 29 September 197o)