By R. W. HONESS,* K. L. POWELL t, D. J. ROBINSON,~ CAROLINE SIM AND D. H. WATSON

Transcription

1 J. gen. ViroL 0974), 22, I59-~69 Printed in Great Britain I59 Type Specific and Type Common Antigens in Cells Infected with Herpes Simplex Virus Type 1 and on the Surfaces of Naked and Enveloped Particles of the Virus By R. W. HONESS,* K. L. POWELL t, D. J. ROBINSON,~ CAROLINE SIM AND D. H. WATSON Department of Virology, The Medical School, Birmingham, BI 5 2TJ, U.K. (Accepted 12 September I973) SUMMARY Of eleven virus-specific antigens detected in cells infected with herpes simplex virus type 1, six have been shown to be type specific, i.e. not present in cells infected with type 2 virus. Two of the five type common antigens are also present in cells infected with pseudorabies virus. Electron microscopic studies of immune agglutination of virus particles have shown that both naked and enveloped particles possess type specific and type common antigenic determinants on their surface. Further, one of the type common determinants on the naked particle surface interacted with antiserum to pseudorabies virus. INTRODUCTION Previous work in this laboratory has shown that there are at least twelve virus-specific antigens in BHK 21 cells infected with herpes simplex virus type t (Watson et al. 1966). One of these, designated Band II, is presumably a structural antigen since virus infectivity is neutralized by an antiserum against Band II (Watson & Wildy, 1969). This antigen is common to both type I and type 2 viruses (Sim & Watson, 1973). Another antigen presumably corresponds to the virus-specific thymidine kinase, since after absorption with an extract of cells infected with the kinaseless mutant of Dubbs & Kit (1964) an antiserum against herpes-infected RK 13 cells gives only one precipitin line in immunodiffusion tests with extracts of cells infected with wild-type virus (Buchan, Luff & Wallis, 197o). Hitherto there has been little information about the other antigens, although one antigen is apparently common to cells infected with either herpes simplex or pseudorabies viruses (Watson et al. 1967). in this paper we provide further information about some of the other antigens present in cells infected with herpes simplex type I. We have extended our previous results on antigens common to cells infected with herpes simplex and pseudorabies viruses. We have confirmed and extended the observation of Schneweis & Nahmias (I97I) that some of the antigens present in cells infected with herpes virus type 1 are type common (i.e. also present in cells infected with type 2 virus), while others are type specific (i.e. present only in cells infected with type I virus). We shall also describe the results of studies on agglutination of purified * Present address: Committee on Virology, Department of Microbiology, University of Chicago, Illinois 6o637, U.S.A. ~ Present address: Department of Virology and Epidemiology, Baylor University College of Medicine, Houston, Texas 77o25, U.S.A. :~ Present address: Scottish Horticultural Research Institute, Invergowrie, Dundee. Present address: Department of Microbiology, School of Medicine, Leeds LS2 9NL.

2 I60 R. W. HONESS AND OTHERS preparations of naked and enveloped particles of herpes simplex virus type [. These provide information about the location of some type specific and some type common structural antigens. We have also identified another type common antigen present in naked particles. This differs from Band II in that the corresponding antiserum does not neutralize virus infectivity. METHODS Cells. BHK 2t cells and RK I3 cells were grown irl Eagle's medium containing calf serum and rabbit serum, respectively (Watson et al. 1966). Viruses. Stocks of herpes simplex virus type I strain HFEM, herpes simplex virus type 2 strains Lovelace and 3345, and pseudorabies virus strain Dekking were all grown in BHK 2I cells as previously described (Watson et al. 1966, ~967). Stocks for preparation of immunizing antigen (see below) were grown in RK 13 cells. Infected cell extracts. BHK 21 cells were infected in suspension at an input multiplicity of to p.f.u./cell, and dispersed in rotating 8o oz bottles. The cells were harvested 16 h after infection and disrupted in distilled water at a concentration of lo s cells/ml by ultrasonic vibration. Concentrated extracts were prepared where necessary by vacuum dialysis of the supernatant fluid obtained after centrifuging infected cell extracts at,ooooog for I h. Antisera. General antisera (to all herpes virus-specific proteins) were prepared by repeated injections of freeze-dried extracts of RK 13 cells infected with herpes simplex virus type [ (Watson et al. z966), or pseudorabies virus (Buchan & Watson, I969). A general antiserum against herpes simplex virus type 2, kindly supplied by Dr G. R. B. Skinner and Mrs M. E. Thouless, was prepared in a similar manner. Antiserum against Band II, a structural antigen of herpes simplex type I, was prepared as previously described (Watson & Wildy, 1969) by lymph-node inoculation of a rabbit with the precipitin band obtained by reaction of general antiserum with the appropriate fractions from preparative polyacrylamide gel electrophoresis (Watson, I969). Further antisera against Band II antigen were prepared by similar inoculation of the precipitin band obtained using the original antiserum to Band II antigen in the reaction with an appropriate electrophoretic fraction. Antiserum specific for type ~ virus was prepared by absorption of a general antiserum against type z virus with an excess of extract from cells infected with type 2 virus (Sim & Watson, 1973). The absorbed serum was centrifuged at Iooooog for ] h and the antiserum in the supernatant fluid concentrated back to its original vol. by vacuum dialysis. This antiserum is described as type t specific (absorbed) antiserum. An antiserum against purified naked virus particles (see below) was prepared by lymphnode inoculation of the virus preparation after inactivation with formaldehyde. 2 x o n inactivated particles were inoculated with Freund's incomplete adjuvant using the methods described previously (Watson & Wildy, i969). A second inoculation was made 6 weeks after the first. The rabbit was bled ~o days after the second inoculation. Virus purification. Purified preparations of type z virus comprising less than I ~o enveloped particles were obtained by centrifuging in sucrose density gradients and chromatography on calcium phosphate as previously described (Robinson & Watson, I97I). These preparations contained rto diffusible antigens detectable in immunodiffusion tests with either general antiserum against type z virus or antiserum against uninfected host cells. They contained 2 to 5/zg protein/lo 10 virus particles as determined by the method of Lowry et al. (I95I). We have also now purified enveloped herpes simplex type I virus. These preparations contained less than 1% naked particles. The procedure will be described in detail and evalu-

3 Herpes virus specific antigens ated in another paper. Briefly, advantage was taken of the fact that infected HEp-2 cells release large amounts of enveloped virus particles into the medium late in infection. The virus was collected by precipitation with polyethylene glycol, and the concentrated virus material further purified by two cycles of sedimentation in sucrose velocity gradients. Electron microscopic analysis of immune agglutination of virus particles. Appropriate (about 5 x io 11 particles/ml) concentrations of purified naked and enveloped virus particles were incubated at 37 C for I h, with 5 vol. of the serum under test. These samples were then diluted and processed for particle counting by the loop-drop method of Watson (1962). Grids were examined at 4oooo x and the distribution of particles, singly and in aggregates, was counted. A particle was said to be part of an aggregate if it was separated from its neighbour by less than half a particle diam. All samples were examined at the same relative dilution of the original particle suspension and the particle recoveries (by comparison with a standard calibrated latex suspension) were comparable, except when very large aggregates were formed. In every series of determinations, the particles were also incubated with preimmune serum and a comparable concentration of bovine serum albumin. In some cases a third control was used: general antiserum against type I virus absorbed with the homologous antigen, freed of excess particles by sedimentation. Immunodiffusion tests. These were done in 3-mm-thick layers of agar ([onagar no. 2) in phosphate-buffered saline (Dulbecco & Vogt, 1954). The wells were arranged hexagonally and were 8 mm in diam., each well being separated from neighbouring wells by 3 mm of gel (see Figs. I to 5). Type specific and type common antigens RESULTS Serological tests Fig. I shows an agar gel immunodiffusion plate in which dilutions of general type I antiserum were tested against dilutions of type I infected cell extract in a pattern arranged so that antiserum/antigen ratios from 27[I to I[27 were tested on the same plate. A total of I[ precipitin bands were detected by examination of the original plate with a hand lens. Similar tests with the type specific (absorbed) antiserum showed that it gave 6 precipitin bands with an extract of cells infected with type I virus, but none with an extract of cells infected with type 2 virus. (Fig. 2 shows an immunodiffusion test of type specific (absorbed) antiserum with one dilution of each infected cell extract.) Six of the I I virus-specific antigens detected in cells infected with type I virus are therefore type specific. Presumably the other five should be type common (i.e. present in cells infected with either type of virus). The accuracy of this estimate was confirmed by tests of the extract of cells infected with type I virus against general antiserum to type 2 virus (which showed 4 lines) and of the general antiserum to type I virus with an extract of cells infected with type 2 virus (which showed 5 lines). Similar results were obtained with several batches of antisera and infected cell extract. Band H antigen We have previously reported that antiserum to Band I[ gives a single precipitirl band with extracts of cells infected with either type z (Watson & Wildy, I969) or type 2 virus (Sire & Watson, I973). We have now found a second precipitin band by using the antiserum concentrated fivefold (Fig. 3). The additional precipitin band was not revealed in tests with dilutions of the concentrated antiserum. Both lines represent type common antigens since they were also detected in tests with extracts of cells infected with type 2 virus. They thus I6I

4 I62 R. W. HONESS AND OTHERS Fig. I. Agar gel immunodiffusion test of an extract of cells infected with herpes simplex virus type ~ (I) with general antiserum against type I virus (A-I). Dilutions 0/3, I/9, 1/27) of both infected cell extract and antiserum are labelled accordingly. Fig. 2. Agar gel immunodiffusion test of extracts of cells infected with herpes simplex virus type I (I) or type 2 virus (2) with type specific adsorbed antiserum (A-I/2) prepared by absorption of general antiserum against type I virus with an extract of cells infected with type 2 virus. differ f r o m the ' t h y m i d i n e kinase a n t i g e n ' which can be detected in type I infected cells (Buchan et al. I97 o) b u t n o t in type 2 infected cells (Thouless, i972), using general a n t i s e r u m to type ~ virus a b s o r b e d with excess o f an extract o f cells infected with the t h y m i d i n e - k i n a s e deficient m u t a n t o f type ~ virus. B o t h precipitin b a n d s were detected in similar tests with B a n d I I antigen f r a c t i o n p r e p a r e d b y p o l y a c r y l a m i d e gel etectrophoresis. A n t i s e r u m to B a n d I I showed no r e a c t i o n with extracts o f uninfected cells in i m m u n o d i f f u s i o n tests.

5 Herpes virus specific antigens I63 Fig. 3. Agar gel immunodiffusion test of concentrated antiserum against Band li antigen (A-IIc) with extract of cells infected with type I virus (I). Dilutions indicated appropriately. Antigens shared with extracts of cells infected with pseudorabies virus We have previously shown that general antiserum to type I virus gives a precipitin band with pseudorabies virus (Watson et al. 1967). We have now extended these results and shown the reciprocal reaction between general antiserum to pseudorabies virus and extracts of cells infected with type I virus. We have shown that two precipitin bands can be revealed in both reactions, although not necessarily at the same antigen/antiserum ratio. Eight different general antisera to type I virus have shown two precipitin bands with extracts of cells infected with pseudorabies virus. The same precipitin bands were obtained in reactions of these extracts with general antisera to type 2 virus. Similar results were obtained in tests of general antisera to pseudorabies virus with extracts of cells infected with type I virus. Two different precipitin bands were again observed. These differed from the precipitin bands given with antiserum to Band II and antiserum to 'thymidine kinase antigen' (Fig. 4). These antigens were type common, i.e. also present in cells infected with type 2 virus. The cross-reactions were confirmed by complement fixation. None of the antisera to pseudorabies virus showed any significant neutralizing activity against either type I or type 2 virus. Similarly, general antisera against either type of herpes simplex virus did not neutralize pseudorabies virus. Capsid antigen Antiserum to purified naked particles produced one precipitin band with extracts of cells infected with both type I and type 2 virus but gave no reaction with extracts of cells infected with pseudorabies virus. The precipitin band differed from Band ii antigen and thymidine kinase antigen (Fig. 5). The antiserum gave no significant neutralization of either type I or type 2 virus, in agreement with Dreesman et al. 0972) who also failed to observe neutralization of type ~ virus by antiserum to purified nucleocapsids.

6 I64 R. W. H O N E S S A N D O T H E R S Fig. 4. Agar gel immunodiffusion test of extract of cells infected with type I virus 0 ) with concentrated antiserum against Band II antigen (A-IIc) and pseudorabies virus (A-PRc) and with antiserum specific for the virus thymidine kinase antigen ( A q / T K - ). The single precipitin line shown in the reaction with antiserum to pseudorabies virus differs from the lines given with the two other antisera. On the original plate an additional precipitin line was shown in this reaction this also differed from the precipitin lines revealed with the other antisera. Fig. 5. Agar gel immunodiffusion test of extract of cells infected with type I virus (I) with antiserum against Band II antigen (A-II) to purified naked particles (A-NV) and against the thymidine kinase antigen (A-I/TK-). The capsid antigen precipitin band revealed in the reaction with A-NV differs from the Band II and thymidine kinase antigens.

7 Herpes virus specific antigens 165 Table I. Agglutination of naked particles of herpes simplex virus type I by specific antisera Percentage agglutinated particles calculated from No. of single particles in mixture with immune serum Antiserum Percentage total no. compared with similar mixture (no. of experiments in brackets) of particles seen in aggregates with pre-immune serum* Pre-immune (2) 3"9 (o) Bovine serum albumin (z) 5 (NS)? o Type I general (2) > 99 98"9 Type 2 general (2) 98"1 89"2 Band II (2) Pseudorabies general (2) 45"7 64"2 47"7 72"4 Type ~ (absorbed) specific (2) 95' * Percentage agglutinated particles = Ioo (I no. of single particles in mixture with immune serum] serum t NS indicates agglutination not significant. All other agglutinations are significant. Table 2. Agglutination of enveloped particles of herpes simplex virus type I by specific antisera Antiserum (no. of experiments in brackets) f Percentage agglutinated particles calculated from Percentage total no. of particles seen in aggregates Pre-immune (5) 14"6 Bovine serum albumin (5) 17"3 (NS)t Type I general (5) 95"1 Type 2 general (5) 71"9 Band II (4) Pseudorabies general (2) 37"4 ~o'o (NS)t Type ~ absorbed specific (2) 68"5 Type 2 absorbed with type I 7"5 (NS)~ antigen (2) No. of single particles in mixture with immune serum compared with similar mixture with pre-immune sermun* (o) o~ 96"7 7z-o 37"3 23"5 (NS)t 86.5 o~ * Calculated as shown in footnote to Table 1. t See footnote to Table I. In these mixtures there were actually more single particles (although not significantly more) in mixture with immune serum than in mixture with pre-immune serum. Naked particles Immune agglutination of naked and enveloped particles Table I shows the results of agglutination experiments with naked particles of type I virus. The figures quoted were derived from a number of experiments with each antiserum. The particles were very efficiently agglutinated by general antisera to both types of virus. In addition the type I specific (absorbed) antiserum also agglutinated. Both type specific and type common antigens are therefore exposed on the surface of the naked particles. Antiserum to Band I[ also agglutinated naked particles. The surface components of the particles responsible for this agglutination may account for some or all of the agglutination by general antisera to the type 2 virus. However, these general antisera contain type common

8 I66 R.W. HONESS AND OTHERS antibodies reacting with extracts of cells infected with pseudorabies virus (see preceding section). Table I shows that general antiserum to pseudorabies agglutinated naked particles. It therefore seems likely that this agglutination represents interaction(s) with type common surface antigen(s) other than Band II antigen(s). General antisera to type 2 virus are presumably able to agglutinate by interaction with both Band I[ antigen(s) and type common antigen(s) shared with pseudorabies virus. Enveloped particles Table 2 shows the result of corresponding experiments for enveloped particles of type virus. Significant agglutination was again found with general antisera to both type I and type 2 viruses. Previous experiments did not show agglutination of enveloped particles by antisera to type I virus (Watson & Wildy, 1963), presumably because of the inferior antisera then available. The agglutination by general antisera to type 2 virus was abolished by absorption with Band II antigen, and we deduce that the only virus-specific type common antigens on the surface of enveloped virus reacting with general antisera to type 2 virus are those which also interact with antiserum to Band II antigen. However, our failure to detect type common antigens other than Band II in the envelope surface cannot be taken to prove absence of such a determinant. The result may only reflect deficiencies in the antisera (cf. the failure of Watson & Wildy (I963) to observe immune agglutination of enveloped particles). Type specific antigens were also revealed on the envelope surface by the agglutination by type I specific (absorbed) antiserum. As expected, general antiserum to type z virus did not agglutinate after absorption with type I extract. The slight agglutination recorded for antiserum to pseudorabies virus was not significant. Once again, this may not prove the absence of the corresponding antigen. DISCUSSION We have extended previous reports (Schneweis & Nahmias, I97I) on type specific and type common antigens of herpes simplex virus. The total number of virus-specific antigens recognized is small compared to the coding potential of the virus. This may reflect not only the insensitivity of gel imlnunodiffusion tests but also the fact that apparently single lines may be multiple (cf. the Band II single line which can be resolved as two at certain serum antigen ratios). The recognition of type specific antigenic determinants should also provide a further basis for the selection of restricted classes of virus-specific proteins for further analysis, e.g. by polyacrylamide gel separation of polypeptides precipitated by type specific antisera. We have already proved the value of immunological selection in providing a criterion of virus specificity and in restricting the number of polypeptides to be studied (Honess & Watson, I974). Such a selection is invaluable with a virus capable of coding for a number of proteins: sufficient to challenge the resolution of the best available separation techniques. That we should have found both type specific and type common antigens is expected since Kieff et al. (1972) found that types I and 2 viruses shared 47 % of the base sequences of the DNA. Some of the other antisera studied here provide the basis for even greater selectivity. Antiserum to Band ii antigen is plainly more complex than originally reported but is still considerably more specific than general antiserum since it has a restricted neutralizing activity (Sim & Watson, I973) as well as only producing two type common precipitin bands in gel immunodiffusion. The restricted number of antigens shared with pseudorabies virus provides another selective procedure for further work. The immunological result can be

9 Herpes virus specific antigens 167 related to the more restricted genetic homology between pseudorabies and herpes simplex viruses compared to the homology between the two simplex types (Bronson et al. i972g Ludwig, Biswal & Benyesh-Melnick, 1972). One of the 'shared' antigens can also be detected in cells infected with varicella virus (G. Nankervis, E. Gold & D. H. Watson, personal communication). It is interesting to note here that Kirkwood, Geering & Old 0972) have presented evidence for a 'group specific antigen' in nucleocapsids of Luck6, Burkitt and herpes simplex viruses. The agglutination studies have provided some information on the location of type specific and type common antigens on the surfaces of the capsid and the envelope. These studies show at least one type specific determinant in the envelope together with one type common antigen which may correspond to the antigen which interacts with antiserum to Band II to cause agglutination and neutralization of infectivity. Similarly we have recognized at least one type specific antigen and two type common sites on the capsid surface. One of the type common sites presumably relates to the agglutination of naked particles by antiserum to Band II antigen and the other to agglutination by antiserum to pseudorabies viruses. There are some problems in correlating precipitin bands in this way to antigens on particle surfaces which may or may not react with neutralizing antibody. These are best illustrated by reference to antiserum to Band II antigen. This we have shown previously has type specific and type common neutralizing activities (Sim & Watson, I973), produces two type common precipitin bands and agglutinates both naked and enveloped particles. There may, of course, be more than one determinant involved in any or all of these activities. Conversely, at least three determinants must be involved because the two precipitin lines are type common and there is type specific neutralizing activity. The agglutinating and type common neutralizing activities could then relate to the two precipitin bands. Correlation of the neutralizing activities with precipitin bands or agglutinating antibodies is difficult because of the much greater sensitivity of the neutralization test. Neutralizing activity of this antiserum was detected using antiserum/virus particle ratios about x ooo-fold greater than those at which agglutination is observed. We can, therefore, never be entirely sure that neutralization is not due to a small population of antibodies not detectable in other tests. There are less acute problems in correlating precipitin bands with agglutination, and supporting evidence may be given by polyacrylamide gel electrophoresis of immune precipitates. We show elsewhere (Honess & Watson, ~974) that the major polypeptide of Band II immune precipitates, and therefore presumably of Band II precipitin lines, co-electrophoreses in polyacrylamide gels with a naked particle polypeptide. It therefore seems reasonable to relate agglutination of naked particles to a Band II precipitin line. Accordingly, we believe the two Band II precipitin lines are related to agglutination of naked and enveloped particles respectively. Subject to our reservations on correlations with neutralizing activity we link type common neutralizing activity of the antiserum to Band II with envelope agglutinating activity. There are at least two further type common structural antigens (see Table 3). One is the capsid antigen, related to a type common precipitin band but not to neutralizing activity, and which may or may not be on the capsid surface. The other is a pseudorabies shared antigen which we believe is represented on the capsid surface, since general pseudorabies antisera agglutinate naked particles but do not neutralize infectivity or agglutinate enveloped particles. We cannot definitely correlate this structural antigen with either of the two pseudorabies shared antigens identified in immunodiffusion tests. There are two type specific structural antigens corresponding to the two type specific neutralizing activities of general antisera (Sim & Watson, I973). One can apparently be

10 I68 Type common R. W. HONESS AND OTHERS Table 3- Summary of antigens and precipitin bands identified T structural b i structural or non-structural Function to which related Enveloped particle agglutination by antiserum to Band I1 and type common neutralizing activity of antiserum to Band II and general antisera Naked particle agglutination by antiserum to Band II Naked particle agglutination by general antisera to pseudorabies Capsid antigen Pseudorabies shaled antigen Pseudorabies shared antigen Correlated to precipitin band?? Total number correlated precipitin bands Number to be correlated 5 0 Total observed 5 Type specific T 1 t structural 1 non-structural First type specific neutralizing activity of general antisera Second type specific neutralizing activity of general antisera (identical to similar activity of antiserum to Band II) Type specific agglutination of naked particles Thymidine kinase No? Total number correlated precipitin bands Number to be correlated 2 4 Total observed 6 related to a precipitin band in immunodiffusion tests (K. L. Powell, A. Buchan, C. Sim & D. H. Watson, personal communication) and we shall return to this component in a subsequent publication. The other, corresponding to the type specific neutralizing activity of antiserum to Band II, obviously cannot relate to a precipitin band. There must be one further type specific structural antigen corresponding to type specific agglutination of naked particles which may or may not correspond to a precipitin line. It now remains for us to consider non-structural antigens. Correlation of precipitin bands with some of the virus-specific enzymes identified in cells infected with type I virus (Keir et al. 1966; Klemperer et al. 1967; Morrison & Keir, 1968) is once again difficult, although tbymidine kinase has been linked with one particular precipitin line (Buchan et al 197o ). Even here there is a slight problem; the antigen detected in immunodiffusion tests is type specific but there seems to be limited cross-neutralization of enzyme activity (Tbouless, 1972). We summarize these findings in Table 3. We have identified in all seven structural antigens. Three are type specific and we can relate one of these to a precipitin band. Four are type common and we can relate three of these to a precipitin band. Of the remaining five type specific precipitin bands we have observed, one corresponds to thymidine kinase antigen, leaving four we have been unable to correlate with any function. The remaining two type

11 Herpes virus specific antigens 169 common precipitin bands correspond to pseudorabies shared antigens, which may or may not be structural. This work was supported by a research grant from the Medical Research Council. Two of us (R.W.H., K.L.P.) are indebted to the Science Research Council for research studentships. We gratefully acknowledge technical assistance from Mrs C. Wallis, R. Wilkinson and A. Hall. REFERENCES BRONSON, D. L., GRAHAM, B. 1., LUDWIG, H.O., BENYESH-MELNICK, M. & BISWAL, N. (I972). Studies on the relatedness of herpes viruses through DNA-RNA hybridization. Biochimica biophysica acta 259, BUCHAN, A., LUFF, S. & WALLIS, C. (I970). Failure to demonstrate interaction of subunits of thymidine kinase in cells simultaneously infected with herpes virus and a kinaseless mutant. Journal of General Virology 9, BUCHAN, A. & WATSON, D. H. (1969). The immunological specificity of thymidine kinases in cells infected by viruses of the herpes group. Journal of General Virology 4, DREESMAN, G. R., SURIANO, J. R., SWARTZ, S. K. & McCOMBS, R. M. (I972). Characteristics of the herpes virus. I. Purification and amino acid composition of nucleocapsids. Virology 50, DOBBS, D. R. & KIT, S. (I964). Mutant strains of herpes simplex deficient in thymidine kinase-inducing ability. Virology 22, 493-5o2. DULBECCO, R. & VOGT, M. (I954). Plaque formation and isolation of pure lines with poliomyelitis viruses. Journal of Experimental Medicine 99, 167-I82. HONESS, P,. W. & WATSON, D.H. (I974). Herpes virus-specific polypeptides studied by polyacrylamide gel electrophoresis of immune precipitates. Journal of General Virology 22, 171-I 8 5, KEIR, H. M., SUBAK-SHARPE, H., SHEDDEN, W. L H., WATSON, D. H. & WILDY, P. (I966). Immunological evidence for a specific DNA polymerase produced after infection by herpes simplex virus. Virology 3 o, KIEFF, E., HOYER, B., BACHENHEIMER, S. & ROIZMAN, B. (1972). Genetic relatedness of type I and type 2 herpes simplex viruses. Journal of Virology 9, KIRKWOOD, J., GEERING, G. & OLD, L. J. (I972). Demonstration of group- and type-specific antigens of herpes viruses. In Oncogenesis and Herpes Viruses, p Edited by P. M. Biggs, (3. de-th6 and L. N. Payne. Lyons: IARC. KLEMPERER, H. G., HAYNES, G. R., SHEDDEN, W. I. H. & WATSON, D. H. (I967). A virus-specific thymidine kinase in BHK Zl cells infected with herpes simplex virus. Virology 3 I, I2O-I28. LOWRY, O. H., ROSEBROUGH, N. J., FARR, A. L & RANDALL, R. J. (195I). Protein measurement with the Folin phenol reagent. Journal of Biological Chemistry 193, LUDWIG, H. O., BISWAL, N. & BENYESH-MELNICK, M. (1972). Studies on the relatedness of herpes viruses through DNA-DNA hydridization. Virology 49, 95 iot. MORRISON, J. M. & KEIR, H. M. (1968). A new DNA--exonuclease in cells infected with herpes virus: partial purification and properties of the enzyme. Journal of General Virology 3, ROBINSON, D. J. & WATSON, D. H. (I971). Structural proteins of herpes simplex virus. JournalofGeneralVirology io, 163-t7I. SCIINEWEIS, K. E. & NAHMIAS, A. J. (I97I). Antigens of herpes simplex virus type 1 and 2- immunodiffusion and inhibition passive haemagglutination studies. Zeitschriftfiir Immunitgitsforschung und experimentelle Therapie 141, SIM, C. & WATSON, D. H. (I973). The role of type specific and cross reacting structural antigens in the neuralization of herpes simplex virus types I and 2. Journal of General Virology 19, THOULESS, M. E. (I972). Serological properties of thymidine kinase produced in cells infected with type I or type 2 herpes virus. Journal of General Virology I7, 3o WATSON, D. H. (I962). Particle counts on herpes virus in phosphotungstate negatively stained preparations in electron microscopy. Sth International Congress for Electron Microscopy (Philadelphia, U.S.A.), vol. 2, X4. Edited by S. S. Breese. WATSON, D. H. (1969). The separation of herpes virus specific antigens by polyacrylamide gel electrophoresis. Journal of General Virology 4, I5I-I6I. WATSON, D. H., SHEDDEN, W. I. H., ELLIOT, A., TETSUKA, T., WILDY, P., BOURGAUX-RAMOISY, D. & GOLD, E. (1966). Virus specific antigens in mammalian cells infected with herpes simplex virus. Immunology xi, WATSON, D. I-I. & WILDY, P. (I963). Some serological properties of herpes virus particles studied with the electron microscope. Virology 2I, IOO-I I I. WATSON, D. H. & WlLDY, P. (I969). The preparation of 'monoprecipitin' antisera to herpes virus specific antigens. Journal of General Virology 4, WATSON, D.H., WILDY, P., HARVEY, B. A. M. & SHEDDEN, W. I. H. (1967). Serological relationships among viruses of the herpes group. Journal of General Virology x, (Received 5 July I973)