J. gen. ViroL (I975), 29, I67-I78

I67

Printed in Great Britain

Some Structural Antigens of Herpes Simplex Virus Type 1 By K. L. P O W E L L * AND D. H. W A T S O N t

Department of Virology, The Medical School, Birmingham Bt 5 2TJ, England (Accepted 16 July 1975) SUMMARY

Several of the major structural polypeptides of herpes simplex virus were obtained in purified form by polyacrylamide gel electrophoresis of purified virus particle polypeptides. Antisera made by footpad inoculation of these polypeptides into rabbits were used to study the antigenic properties of two envelope glycoproteins and of the major capsid protein.

INTRODUCTION

Studies of the antigens produced by members of the herpesvirus group are as yet rudimentary; they have been hindered by the lack of easily purified virus components such as are available from cells infected with adenovirus (Valentine & Pereira, 1965) or disrupted influenza virus particles (Laver & Valentine, 1969). Studies in this laboratory (Watson, 1969; Watson & Wildy, 1969) and more recently in another (Cohen, Ponce de Leon & Nichols, 1972) have allowed the preparation of antisera against one of the virus structural antigens, band II. A simple and reproducible method for the purification of herpesvirus particles in reasonably large quantities would facilitate the preparation of antisera to virus structural components. In the last year Powell et al. (1974) and Courtney & Benyesh-Melnick (1974) have shown that useful serological reagents for herpes simplex virus proteins can be produced using material prepared by polyacrylamide gel electrophoresis in gels containing SDS. Purified virus particles are obviously the best source of virus structural proteins. Here we report a purification method which enables reasonably large quantities of virus particles to be prepared. The use of these to prepare antisera against some of the major virus particle polypeptides is described. METHODS

Cells. BHK 21 cells were grown in Eagle's medium containing tryptose phosphate broth and calf serum (Watson et al. 1966). These cells were used for antigen production and virus assay. HEp-2 cells, used to produce virus for purification, and Vero cells, used in some of the immunofluorescence studies, were grown in a similar manner. Viruses. The HFEM strain of herpes simplex virus type I, and the 3345 strain of herpes simplex virus type 2 were grown in BHK 21 or HEp-2 cells as previously described (Watson et al. I966, I967). Purification of virus. Preliminary experiments indicated the cell line and conditions to be * Present address: Department of Virology and Epidemiology, Baylor College of Medicine, Houston, Texas 77o25, U.S.A. t Present address: Department of Microbiology, School of Medicine, Leeds LS2 9NL, England.

I68

K.L. POWELL AND D. H. WATSON

used for virus growth. Monolayers of HEp-2 cells were infected at high input multiplicity (50 to lOO p.Lu./cell) with strain HFEM. After allowing 2 h for adsorption at 32 °C, the cells were washed once with warm medium and incubated with replenished medium at 32 °C for 2 days. At the end of the growth period the medium was collected and used for virus purification. Virus was concentrated from the medium using polyethylene glycol (tool. wt. approx. 6000, British Drug Houses Ltd, Poole, England) at a final concentration of 8 (w/v), in the presence of 0"5 M-NaC1. These concentrations were shown by preliminary experiments to precipitate the virus infectivity and particles efficiently from the medium. The concentrated virus pellet was resuspended in a small volume of 0.02 M-phosphate buffer, p H 7"4, by ultrasonic vibration, and layered on a preformed 5 to 45 ~ 26 ml sucrose gradient. The virus was sedimented at I25OO rev/min for 1 h in the SW27 rotor of a Christ Omega II ultracentrifuge. At this stage virus could be seen as an opalescent band in the centre of the gradient. This step effectively separated virus particles from contaminating protein and a second sucrose gradient cycle rendered the virus adequately pure for our purposes. Production of labelled virus. To produce radioactively-labelled virus, cells were washed after the adsorption period with amino acid-free Eagle's medium containing the usual amounts of glutamine, arginine, histidine, methionine, threonine, tryptophan and inositol, ~/3o the normal concentration of the other amino acids, and 5 ~ calf serum. Incubation was then continued in the same medium with 200 to 500 #Ci per bottle of a mixture of radioactively-labelled amino acids ([3H]-L-leucine 4,5-T, 20 to 38 Ci/mmol; [3H]-L-lysine 4,5-T, I8-8 Ci/mmol; [3H]-L-3-phenylalanine 2,3-T, I Ci/mmol and [3H]-L-valine 2,3-T, 19 to 31"6 Ci/mmol) obtained from the Radiochemical Centre, Amersham, England. Under these conditions the yields and particle-to-infectivity ratios of labelled and unlabelled virus were equal. Virus labelled with [3H]-glucosamine or [14C]-amino acids was prepared in the same manner. Production of naked virus particles. These were purified by the method of Robinson & Watson 0971) and contained less than I ~ enveloped virus particles. Pre-labelling of cells before virus infection. Cells were labelled from 24 to 72 h after passage with Ioo/zCi per roller bottle of a mixture of [x4C]-labelled amino acids ([14C]-L-leucine CI 4 (U), 270 to 342 mCi/mmol; [14C]-L-lysine CI4 (U), 270 to 342 mCi/mmol; [14C]-L-3phenylalanine Ct4 (U), 405 to 5r3 mCi/mmol; and [14C]-L-valine Ct 4 (U), 225 to 280 mCi/ mmol) obtained from the Radiochemical Centre. This labelling period was followed by a 6 h incubation in normal medium, when practically all incorporated radioactivity was TCA-precipitable. The cells were then infected and after adsorption were labelled as for the production of labelled virus. After 48 h the virus was collected and purified in the usual manner. Electrophoresis techniques. For analytical electrophoresis the method of Davis (1964) w a s used, as modified by Dimmock & Watson (1969) for the inclusion of SDS. Slab gel electrophoresis was done in a similar manner, using identical buffers and gel compositions. However, because of technical difficulties with the slab gel method, dithiothreitol was not included in these gels, nor were they pre-run to remove catalyst. Serological methods. General antisera (to all virus-specified proteins) and antisera to band II antigen were prepared as described previously (Watson et al. 1966; Watson & Wildy, I969). Antisera to individual virus structural proteins were prepared by disrupting virus particles using SDS, urea, and dithiothreitol, and electrophoresis of the structural virus polypeptides on 7 ~ polyacrylamide gels containing these reagents. After electrophoresis, a longitudinal strip was removed from each gel and the remainder was frozen. The strips

Structural antigens o f herpes simplex type ~

I69

of gel were stained and the polypeptides located in the gel; the strip was then replaced on the rest of the gel and the segments of the gel containing polypeptides of interest removed. Segments of several gels were pooled and homogenized in Freund's incomplete adjuvant prior to injection. Rabbits were given injections of this material at 2-week intervals and were bled 2 weeks after the second injection. Complement fixation tests and neutralization tests were done using the methods described by Sim & Watson 0973) and fluorescent antibody tests by the method of Ross, Watson & Wildy (1968). Gel diffusion tests were done in 3 mm thick layers of agar (Ionagar No. 2) in phosphate buffered saline (Dulbecco & Vogt, I954). The wells were arranged hexagonally and were 8 mm in diam., each well being separated from its neighbour by 3 mm of gel. Immune agglutination tests were performed as described by Honess et al. 0974)Other methods. The 'loop-drop' method of Watson, Russell & Wildy 0963) was used for particle counting. Infectivity was assayed by the method of Russell 0962) and protein concentration by the method of Lowry et aL (195 0. Immunofluorescent staining was done by a direct method as described by Ross et al. (I968). RESULTS

Purification The purification of virus from medium surrounding infected cells is shown in Table I. This shows the particle, infectivity and purity data from a typical experiment. The behaviour of this material on a second sucrose gradient is shown in Fig. I. All protein and radioactivity in the preparation was associated with the peak of virus particles and infectivity (this was also true when particles were centrifuged on caesium chloride gradients). The protein-toparticle ratio of the purified virus was near that expected for a particle of the size and composition of the enveloped virus particle. We were unable to demonstrate the presence of any diffusible herpes virus-specific or host cell-specific antigens in the final virus preparations by either gel immunodiffusion tests or by non-SDS polyacrylamide gel electrophoresis. More direct evidence for virus purity comes from the experiment in which virus was grown in cells labelled before virus infection. In this experiment only a very small amount of 'prelabel' was found in the purified virus preparation and, as can be seen from Fig. 2, most of this label was associated with virus peaks, probably as a result of 'turnover' of pre-label into virus particle polypeptides. Finally we have compared the polypeptides revealed by staining a polyacrylamide gel after electrophoresis of purified labelled virus with those shown in a parallel gel by slicing and counting. No additional peaks were detected by staining, as would be expected for virus relatively uncontaminated with host cell material. Acrylamide gel profiles of purified virus The acrylamide gel profiles, as we have previously reported (Powell et al. I974), showed I3 clearly visible peaks. It should be noted that these gels contain the reducing agent dithiothreitol. The polypeptides are more clearly resolved by electrophoresis in slabs of polyacrylamide gel followed by autoradiography of [14C]-labelled polypeptides (Fig. 3). In addition to the I3 species already detected, we resolved 2 minor peaks running slightly ahead of peak o, z peaks at the position of peak 4, and noted that peak 5 splits into 2 widely spaced peaks. However, we believe that this last feature may represent an artefact since these two peaks can also be seen on stained cylindrical gels from which dithiothreitol is omitted. Major difficulties in nomenclature are represented by the regions marked 2/3 and 8/9, which are not clearly resolved in either stained cylindrical gels or autoradiography of

I7O

K. L. P O W E L L

AND

D. H. W A T S O N

Table I. Purification of herpes simplex virus Volume (ml) Medium

700

p.f.u./ml

Total p.f.u,

/o°/ recovered

Particles per ml

6.0 × IO8

4"2 x IO11

.

3 6 × lO 1°

3"4 X 1011

(I00)

3"2 × 1011

3"0 × 10 TM

(I00)

78I

After t st sucrose gradient

o'75

I-O × IO 11 7"5 × 101°

22

I'4 X 1013

I'1 × IO TM

37

I9

After 2nd sucrose gradient

0"50

O"9;'(1011

13

I'25×1012

6"2×1011

20

14

I

I

.

# g protein p e r lo TM particles

9'4

TM

.

~ recovered

PEG ppt

4"5×IO

.

Total particles .

I

-

5"0

"r ×

×

2'5 d~

5

10

15

Fraction Fig. 1. Co-sedimentation o f herpes simplex virus infectivity a n d radiolabel o n a 5 to 45 ~ (w/v) sucrose velocity gradient: IS]--IS/, infectivity (p.f.u./ml); 0 - - 0 , radio/abel.

gel slabs. As shown in Fig. 4 these regions of the gels contain virus polypeptides labelled with [SH]-glucosamine. They probably represent glycoproteins and may be heterogeneous in molecular size. For this reason, until we know more about the polypeptide moieties of these glycoproteins, we prefer not to subdivide these regions further. Fig. 4 also shows that there are three glycoprotein regions in the virus particle polypeptide profile. The major polypeptides of the naked herpes virus particle are compared with those of enveloped particles in Fig. 5. Clearly, polypeptides r and I3 of the enveloped virus particle are the major components. Polypeptide 1I of the naked virus particle, and polypeptide I2 of the enveloped virus particle do not co-electrophorese, and, as previously suggested by Gibson & Roizman 0972), we believe that polypeptide II may be a precursor of polypeptide I2.

Structural antigens of herpes simplex type I

I7I

1.5 ¸

T

"~ 1 . 0 -

5

._2

.o

"

7

13

1~

f,

"~

3

0.5 !

8

9

4

12 I0

0oJ

.

.

.

.

.

50 Fraction Fig. 2. Herpes simplex virus from 'pre-labelled' cells: examination of virus particle polypeptides for pre-infection and post-infection label. G--C), Fre-infection [xaC]; @--@, post-infection [~H].

Preparation of antisera to virus structural components On the basis of the polypeptide profiles discussed above, we chose to study 4 polypeptides by making antisera against SDS-disrupted components; these were polypeptide VPo, the component of very high mol. wt., the major capsid protein VPI, and the z major glycoprotein regions VP2/3 and VP8/9. Sera obtained by footpad inoculation of rabbits were used in 5 tests: gel diffusion, complement fixation, immune agglutination, neutralization, and fluorescent antibody tests. The results with VP2/3 have been published in detail elsewhere (Powell et al. I974). Antiserum to VPo failed to react in any of the tests while the other sera reacted in all tests. In gel immunodiffusion tests, antiserum to VPz/3 gave a single precipitin line and antiserum to VPI a single precipitin line close to the antigen well; antiserum to VP8/9 gave z precipitin lines, one weak and the other strong. None of the sera tested reacted with host cell antigens or with components of the tissue culture medium. In all cases the gel diffusion lines obtained with these sera were weak. This may reflect the relatively small amount of immunogen used to inject the animals or more likely, the low solubility of the antigens used. Since the lines are difficult to reproduce photographically, we do not show these results. The results of complement fixation tests using the antipolypeptide and control sera are shown in Table z. Both antiglycoprotein sera reacted to a high level with homologous antigen homogenate (I/64o to I/I28O); they did not react with a supernatant fluid obtained from the same antigen preparation by centrifuging at I o o o o o g as previously described I2

VIR

29

I72

K. L. P O W E L L

AND

D . H. W A T S O N

~: 2::: 7;

2?

i~!: : i :! i: i:

::

~i~~i

i i/:i:i •: ':~!~iii ,~ !i!2 i!i 13 ::~:i::

~

:,~ii : v

•

• •

!iii:?~o¸ : :

rb :,:,

Fig. 3. A n a l y s i s o f h e r p e s s i m p l e x virus p o l y p e p t i d e s by slab p o l y a c r y l a m i d e gel electrophoresis and autoradiography.

Structural antigens of herpes simplex type I I 4

I

173

I

-

67

7 ×

.=,

!

~2

89

i Uot 25

10

50

75

Fraction

Fig. 4- Analysis of herpes simplex virus polypeptides labelled with [~H]-glucosamine(D--D) and [14C]-aminoacids (O--11) by polyacrylamide gel electrophoresis.

3

-

15

rb

7 -

tI21

10

.~ ,-e

-

5

Fig. 5- (a) Polyacrylamide gel analysis of [14C]-labelledenveloped (O--1t) and [aH]-labellednaked (O -- O ) herpes simplex virus particles. (Watson, 1969) or with antigen made from cells infected with heterologous virus. Antiserum to the capsid antigen reacted only with antigen homogenates; however it did show some cross reactivity with herpes simplex virus type 2-infected cell antigen. The control sera anti-band II, general antiserum to herpes simplex virus type I, and general antiserum to herpes simplex virus type 2 cross reacted, as would be expected (Sim & Watson, 1973), thus demonstrating that the antigen preparations were satisfactory. In fluorescent antibody tests the sera showed interesting reactions, especially in the localized patterns of fluorescence observed (Fig. 6). Antiserum to VP2/3, VPS/9 and VPI I2-2

174

K. L. P O W E L L T a b l e 2.

AND

D. H. W A T S O N

Reactivity of antipolypeptide sera in complement fixation tests T y p e i antigen*

Serum Anti-VPo Anti-VP2/3 Anti-VPI Anti-VP8/9 C o n t r o l sera Antiband I I General a n t i s e r u m to type ~ virus G e n e r a l a n t i s e r u m to type 2 virus

T y p e 2 antigen*

S/N

Homogenate

S/N

Homogenate

--t 20 to 4o 4o 2o

-128o 32o 64o

-20 2o to 4o 2o

-4o 8o 40

4o 2560 I280

32o 5120 256o

2o 2560 256o

12o 5120

2560

* T h e figures q u o t e d are the dilutions at the titration end points. !" N o activity detected.

Fig. 6. I m m u n o f l u o r e s c e n c e observed in Vero cells infected for 6 h with herpes simplex virus type [ a n d mock-infected cells. (a) A n t i s e r u m to VP2/3 ; (b) mock-infected cells with a n t i s e r u m to VPz/3 ; (c) p r e - i m m u n e s e r u m ; (d) a n t i s e r u m to VPI.

Structural antigens of herpes simplex type I

I75

Table 3. Agglutination of naked and enveloped particles of herpes simplex virus

type I by antisera

Antiserum Pre-immune VPI VPz/3 Band I[ Type I general Type 2 general

Percentage of the total number of naked particles seen in aggregates 4 56 4 42 > 99 98

Percentage of the total number of enveloped particles seen in aggregates 15 Io 12 37 95 7z

stained herpes simplex virus-infected BHK, HEp-2 and Vero cells, but not uninfected cells. Pre-immune sera gave only weak fluorescence. Both antisera against herpes simplex virus particle glycoproteins reacted similarly; they gave bright Cytoplasmic fluorescence with herpes simplex virus-infected cells, especially around the nuclear membrane, and they did not react with the nucleus. With herpes simplex virus type z-infected cells they gave weak fluorescence indistinguishable from pre-immune sera, but this is difficult to interpret since in our hands pre-immune sera always give weak fluorescence with herpes simplex virusinfected ceils. Antiserum to VPI gave very intense nuclear staining and weak cytoplasmic staining was observed late in infection. The difference in the two patterns between glycoprotein and capsid sera is very striking. One interesting manifestation of the glycoprotein serum staining was the 'cobweb' appearance of the stained cytoplasm, as though the antigen may be related to some structural material. Neutralization tests with these sera showed only one positive anti-VP2/3; this result was remarkable in that activity was only detected against herpes simplex virus type I. This finding has already been examined in detail (Powell et al. I974). Immune agglutination tests were carried out to determine the location of the antigenic site on the virus particle. Antiserum to each glycoprotein region did not agglutinate either enveloped or naked virus particles (Table 3); however, antiserum to the major capsid protein did show small but positive reactivity to the naked virus. All the control sera reacted as one would expect, agglutinating both enveloped and naked virus particles. DISCUSSION

As stated in the Introduction, there is a need for a simple and rapid method of herpesvirus particle purification. We believe that the procedure we have outlined provides such a method. Spear & Roizman 0972) argued against the use of extracellular virus for the purification of herpes simplex virus, since they found that virus release was inefficient and only occurred on cell disintegration. Our results would appear to contradict this, since electron microscopy of virus preparations concentrated from medium shows more than 9o ~ enveloped virus, whereas if disintegration had occurred, one would expect to find naked virus particles. It is our belief that the choice of virus strain and conditions of growth are very important in this context. Although there is a considerable loss of virus particles during the purification procedure, we believe this to be random since the particle-to-infectivity ratio of the virus preparation remains reasonably constant throughout the purification procedure. This would indicate that our purification procedure is not deleterious to the infectivity of the virus particles.

I76

K. L. P O W E L L AND D. H. W A T S O N Table 4- Comparison of polypeptides of herpes shnplex virus recognized in this study and in that of Spear & Roizman (I972)

Spear & Roizman (I972) c

Peak no.

Mol. wt. ( x to -s)

I-3

260--275

0

4 5 6 7/8 9 I0 II

184 155 146 I26 I 12 98 93

Absent I Absent 2/3 Absent Absent 4a

IZ I3[14

87 78

15 16 I7-21 22a

7I 65 44-59 39

22

23 24

This study

4b 5 6 7 8/9 and IO tt 12

13 Absent

Comments There is a difference in the amounts of minor components in this region in this study -Major capsid protein -Heterogeneous glycoprotein region --Originally designated 4; extra minor band detected by autoradiography -Two peaks on gels lacking dithiothreitol; both behave similarly as determined by criteria of Gibson & Roizman (I974)* --Heterogeneous glycoprotein region Precursor to 12 ? --

---

* Gibson & Roizman (I974): both peaks exhibit pink fluorescence under intense illumination when stained with Coomassie brilliant blue. O u r criteria indicate that virus o f reasonable purity is obtained. Attempts to purify the virus particles further did not significantly alter the protein-to-particle ratio nor the acrylamide gel profile o f purified virus particles. Enveloped virus does, however, survive centrifuging in caesium chloride and potassium tartrate gradients with little loss o f infectivity, and these may be used as alternative methods to purify virus particles. Protein-to-particle ratios allow the comparison o f preparations o f virus particles f r o m various laboratories. Heine et. al. (i974) calculated that the protein mass o f the virion ranged from I6 to 19/zg protein/lo 10 particles. Our preparations give very similar values (see TabIe 0 ; hence b o t h laboratories are working with material o f comparable purity. As one would expect, the acrylamide gel profiles of our purified virus are in reasonably g o o d agreement with those o f Spear & Roizman. Table 4 gives a list o f the polypeptides observed by both groups and their likely correspondence. The differences m a y be summarized thus: in their preparations, Spear & R o i z m a n 0 9 7 2 ) find a polypeptides (4 and 6) which we do not observe. Since these are polypeptides found in large amounts in infected cells (Honess & Roizman, 1973), they may be minor contaminants o f virus particle preparations. Secondly, we detect differences in the virus particle glycoproteins which are probably due to strain-specific differences in the mobility o f these components (K. L. Powell, personal communication). The similarities are, however, more striking than the differences. Polypeptide lO o f our scheme probably resembles polypeptide 19 o f Heine et al. (I974)Their I9c represents the most mobile of their components (labelled I9) and is f o u n d in the capsid; the less mobile band I9e is f o u n d in enveloped virus particles. F r o m Fig. 5 it is clear

Structural antigens o f herpes simplex type I

I77

that this is also true for our polypeptide Io: the component found in capsids corresponds to the fast running side of peak Io. We have not been able to show any incorporation from radioactively labelled glucosamine into this material. This may be significant in relation to the band II antigen (Watson &Wildy, 1969) which is responsible for type-common neutralization (Sim &Watson, I973). Honess &Watson (1974) showed that this antigen prepared by immune precipitation co-migrated with peak to, and was glycosylated. In addition, Honess et aL 0974) showed that antiserum to this antigen agglutinated both naked and enveloped virus. If band 1I antiserum reacts with both these components, i.e. polypeptide Io from both naked and enveloped virus, then that could explain its reactivity. We were successful in preparing antisera to 3 of the virus particle polypeptide regions, using material purified by SDS-polyacrylamide gel electrophoresis. Antiserum to both the major envelope glycoprotein regions VP2/3 and VPS/9 reacted type-specifically in gel diffusion and complement fixation tests, and antiserum to VP2/3 neutralized virus typespecifically. This may indicate that the type-specific antigenic determinants of herpesvirus, like the sub-type-specific antigenic determinants of influenza virus, are carried by the virus particle glycoproteins. Herpesvirus glycoproteins are inserted into the plasma membrane (Heine, Spear & Roizman, ~972) and presumably into the nuclear membrane where the virus obtains its envelope. In agreement with this interpretation, both these sera stain the nuclear membrane and the cytoplasm in the fluorescent antibody tests. Both antisera to glycoprotein regions reacted largely with the homogenate from infected cells and not with soluble material, suggesting that the antigens involved are antigenically active when inserted into the cell membranes. The antigenic determinant involved with antiserum to VP2/3 is clearly not modified by treatment with SDS since (a) the serum neutralizes virus, and (b) the serum reacts in complement fixation tests with purified virus particles. Antiserum to the major capsid protein reacted in gel diffusion tests as if the antigen were in a large aggregate form, and reacted in complement fixation tests only with cell homogenates and not with soluble fractions. In fluorescent antibody tests the nucleus was the major area stained. These observations suggest the presence in the nucleus of an aggregated form of capsid protein, possibly not virus particles per se. Earlier results (Watson, Wildy & Russell, I964) suggested that there may be an accumulation of capsomeres around the assembly site of naked virus particles; it may be that antiserum to VPI stains such aggregates. Immune agglutionation tests showed that the capsid protein antigenic determinant was located on the particle surface since these particles could be agglutinated by the antiserum to VPt. These results indicate that a useful method for the preparation of herpes simplex virus particles and antigens has been evolved. This has already allowed the preparation of typespecific antigenic reagents (Powell et al. t974) as well as studies of the virus particle surface antigens (Honess et al. t974). It should prove useful for many other studies. We are grateful to the Medical Research Council for financial support for this work and to the Science Research Council for a research studentship to one of us (K.L.P.).

I78

K. L. P O W E L L

A N D D. H. W A T S O N

REFERENCES COHEN, G. M., PONCE DE LEON, M. & NICHOLS, C. (1972). Isolation of a herpes simplex virus-specific antigenic

fraction which stimulates the production of neutralizing antibody. Journal of Virology xo, IO2I-lO3O. COURTNEY, R . J. & BENYESH-MELNICK, M. (I974). Isolation and characterization of a large molecular-weight polypeptide of herpes simplex virus type I. Virology 62, 539-55 I. DAVIS, B. J. (I964). Disc electrophoresis. II. Method and application to human serum proteins. Annals o f the New York Academy of Sciences x2x, 404-427. DIMMOCK, N. J. & WATSON, D. H. (I969). 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, 499-509. DULBECCO, R. ~, VOGT, M. (1954). Plaque formation and isolation of pure lines with poliomyelitis viruses. Journal of Experimental Medicine 99, 167-182. GIBSON, W. & ROIZMAN, B. (1972). Proteins specified by herpes simplex virus. VIII. Characterization and composition of multiple capsid forms of subtypes I and 2. Journal of Virology xo, Io44-Io52. GIBSON, W. & ROIZMAN, B. (I974). Proteins specified by herpes simplex virus. X. Staining and radiolabeling properties of B capsid and virion proteins in polyacrylamide gels. Journal of Virology x3, I55-I65. HEINE, J. W., HONESS, R. W., CASSAt,E. & ROlZMAN, B. (I974)- Proteins specified by herpes simplex virus. XII. The virion polypeptides of type 1 strains. Journal of Virology x4, 640-65 I. HEINE, J. W., SPEAR,P. G. & ROIZMAN, B. (1972). Proteins specified by herpes simplex virus. VI. Viral proteins in the plasma membrane. Journal of Virology 9, 431-439. HONESS, R. W., POWELL, K. L., ROBINSON, D. J., SIM, C. & WATSGN, D. H. (I974). Type specific and type common antigens in cells infected with herpes simplex virus type I and on the surfaces of naked and enveloped particles of the virus. Journal of General Virology 22, 159-169. HONESS, R. W. & ROIZMAN,B. (I973). Proteins specified by herpes simplex virus. XI. Identification and relative molar rates of synthesis of structural and non-structural herpes virus polypeptides in the infected cell. Journal of Virology x2, I347-I365. HONESS, R.W. & WATSON, D.H. (1974). Herpes virus-specific polypeptides studied by polyacrylamide gel electrophoresis of immune precipitates. Journal of General Virology 22, 17I-I 85. LAVER, W.G. & VALEt,trINE, R.C. (I969). Morphology of the isolated hemagglutinin and neuraminidase subunits of influenza. Virology 38, IO5-119. LOWRY, O. H., ROSEBROUGH, N. J., EARR, A. L. & RANDALL, R. S. ( I 9 5 I ) . P r o t e i n m e a s u r e m e n t w i t h t h e F o l i n

phenol reagent. Journal of Biological Chemistry x93 , 265-275. POWELL, K. L., BUCHAN, A., SIM, C. & WATSON, n . H . (1974). T y p e - s p e c i f i c protein in herpes simplex virus

envelope reacts with neutralising antibody. Nature, London 249, 36o-361. ROBINSON, D.J. & WATSON, D.H. (I97I). Structural proteins of herpes simplex virus. Journal of General Virology xo, 163-171. ROSS, L. J. N., WATSON, D.H. & WILDY, P. (I968). D e v e l o p m e n t a n d l o c a l i z a t i o n o f virus-specific a n t i g e n s

during the multiplication of herpes simplex virus in B H K 21 cells. Journalof General Virology 2, I 15-122. RUSSELL, W. C. (1962). A sensitive and precise plaque assay for herpes virus. Nature, London x95, IO28-IO29. SIM, C. & WATSON,D. H. (I973). The role of type specific and cross reacting structural antigens in the neutralization of herpes simplex virus types I and 2. Journal of General Virology 19, 217-233. SPEAR, P. G. & ROIZMAN, B. (I972). Proteins specified by herpes simplex virus. V. Purification and structural proteins of the herpesvirion. Journal of Virology 9, I43-I59. VALENTINE, R. C. & PEREIRA, H. G. (1965). Antigens and structure of the adenovirus. Journal of Molecular Biology x3, 13-2o. WATSON, D. H. (1969). The separation of herpes virus-specific antigens by polyacrylamide gel electrophoresis.

Journal of General Virology 4, 151-16 I. WATSON, D. H., RUSSELL, W. C. & WILDY, P. (1963). E l e c t r o n m i c r o s c o p i c p a r t i c l e c o u n t s o n h e r p e s v i r u s using the phosphotungstate negative staining technique. Virology 19, 2 5 0 - 2 6 0 . 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 II, 399-4o8. WATSON, n. H. & WILDY, P. (1969). The preparation of 'monoprecipitin' antisera to herpes virus specific antigens. Journal of General Virology 4, 163-168. WATSON, D. H., WILDY, P., HARVEY, B. A. M. & SHEDDEN, W. I. H. (I967). S e r o l o g i c a l r e l a t i o n s h i p s a m o n g

viruses of the herpes group. Journal of General Virology I, 139-14I. WATSON, D. H., WlLDY, P. & RUSSELL, W. C. (I964). Q u a n t i t a t i v e e l e c t r o n m i c r o s c o p i c s t u d i e s o n the g r o w t h

of herpes virus using the techniques of negative staining and ultramicrotomy. Virology 24, 523-538.

(Received 6 M a y I 9 7 5 )