Herpes simplex virus infection of the human sensory ... - Springer Link

Archives

Vifrology

Arch Virol (1988) 101 : 87-104

© by Springer-Verlag1988

Herpes simplex virus infection of the human sensory neuron An electron microscopy study E. Lycke, B. Hamark, Maria Johansson, Antonina Krotochwil, J. Lycke, and B. Svennerholm Department of Virology, Institute of Medical Microbiology, University of G6teborg, G6teborg, Sweden Accepted April 25, 1988

Summary. Herpes simplex virus (HSV) type 1 was used to infect cultures of human embryonic dorsal root ganglion cells. Infected cultured were studied by electron microscopy. Viral nucleocapsids were observed to be internalized into neuronal cells bodies and neuritic extensions by fusion of the viral envelope and the plasma membrane. No signs of internalization by endocytosis were noted. Nucleocapsids were transported in neurites and were within 2 hrs postinfection found located near the microtubules and close to the nuclear pores in the perikaryon. A primary envelopment of nucleocapsids occurred at the inner lamina of the nuclear membrane and virions appeared between the two laminae. Presence of non-enveloped nucleocapsids outside the nuclear membrane and in close contact with the endoplasmic reticulum suggested that nucleocapsids could pass to the cytoplasmic side probably by de-envelopment at the outer nuclear membrane. A secondary envelopment occurred at the endoplasmic reticulum where the virions also became enclosed in transport vesicles. Enveloped virus appearing in the cytoplasm of neurons and neuritic extensions was always found only inside these transport vesicles. During their passage through the cytoplasm the virion-transport vesicle complexes were surrounded by smaller lysosome-like vesicles possibly derived from the Golgi apparatus. Fusion reactions between vesicles with virions and the smaller vesicles seemed to occur. We discuss if in this way the virion-transport vesicle complexes might be provided with glycosyl transferases and substrates necessary for maturation and completion of glycosylation of the viral envelope glycoproteins. The transport vesicles seemed essential for egress of virions from the infected cell by releasing virus when fusing with the plasma membrane.

Introduction The sensory neurons of the nervous system are the main target cells for the herpes simplex virus (HSV) infection, in the sense that in the natural cycle of

88

E. Lycke et al.

HSV they represent the final cellular destination for the primary infection and the site for establishment of latency [8]. In cultures of dissociated and differentiated rat sensory dorsal root ganglion (DRG) cells, HSV is taken up by neuritic extensions [33] and is transported axonally to the neuronal cell soma as nucleocapsids [17]. This axonal transport is sensitive to the action of taxol and other compounds interacting with microtubular functions suggesting that maintained microtubular function is essential for the transport of the nucleocapsids [14]. Similarly, the transport of virus from the neuronal soma in a centrifugal direction i.e. transport of the de novo formed virus seems dependent upon microtubular activities but the virus is present as enveloped particles enclosed in cytoplasmic vesicles of neuritic extensions [17]. But for these observations most of the virus-nerve cell interactions, including the morphogenesis of HSV in neuronal cells, remain unexplored. With established cell-lines of fibroblast- or epithelial origin the entry of herpesvirus after attachment has been attributed to fusion of virus envelope and plasma membrane [6, 20, 21, 25, 29] or endocytosis [2, 10]. It is still not settled if both these processes are equally important for establishment of infection, although recent observations of Fuller and Spear [6] exclusively favor fusion as the essential means for HSV penetration of the cell. The envelopment of newly formed nucleocapsids and the egress of herpesvirions from the cell are not wholly understood either. Stackpole [31] launched the hypothesis of envelopment of nucleocapsids at the inner lamella, a subsequent de-envelopment at the outer lamella of the nuclear membrane and a final envelopment of the virus in the endoplasmic reticulum. Envelopment at cytoplasmic membranes has also been observed by others [4, 28]. In addition, egress of virus has been supposed to occur either by transfer of virions via the cisternae of the endoplasmic reticulum, which in the infected cell was supposed to connect the nuclear membrane with the extracellular fluid [27], or by enclosure of virus into cytoplasmic vesicles [20] transferring virus from the endoplasmic reticulum via the Golgi apparatus to the cell surface [30]. Material and methods Virus

The McIntyre strain of HSV type 1 was used throughout the study. Suspensions of virus were freshly produced on a green monkey kidney cell line (AH-1), cleared by low speed centrifugation and used directly for inoculation of DRG cell cultures. The concentration of infectious virus was estimated to 1.3 x 108 pfu per ml. Suspensions were diluted to give multiplicities of 100 to 200 pfu per cell. At 30 min postinfection the cultures were rinsed 3 times and fresh medium added. The human DRG cultures were highly permissive to HSV and 24 hour yields of 3 x 10 6 pfu, corresponding to a production of 2,000-3,000 pfu per cell, were obtained. Cell culture

Dorsal root ganglia (DRG) of human embryos of 10 to 12 weeks gestation were collected by forceps under a dissecting microscope. The cells were dissociated by treatment with 0.25 per cent trypsin in a Ca + + and Mg ++ free Hanks' buffer at pH7.2 for 30min at 37°.

HSV infection of human neurons

89

The trypsin was then removed by low speed centrifugation and the cells were resuspended in Eagle's MEM supplemented with 10 per cent fetal calf serum. 0.5 per cent (w/v) glucose, L-glutamine (150 mg/1,000 ml), nerve growth factor (50 ng/ml), gentamycin (5 mg/1,000 ml) and chick embryo extract. This medium was also used for culturing the cells. Cultures were prepared by seeding the cells on collagen coated (Vitrogen 100, Collagen Corp.) plastic dishes (5 cm, Falcon), and incubation at 37°. The standard growth medium was replaced every second day with a medium containing in addition 10-5 M of cytosine-arabinoside, fluorodeoxyuridine and uridine, respectively, to minimize growth of fibroblasts. Fourtyeight hours before inoculation of virus the cultures were thoroughly washed and standard growth medium without inhibitors of mitosis was used. Virus-inoculated cultures were incubated for 5 min to 20 hours at 37°. Neurons and neurites were identified by the presence of their neurofilaments and microtubutar structures. In addition a monoclonai antibody specifically directed against ganglioside, GD 1 b, of human brain [5] was used for identification of neuronal cells. Immunofluorescence indicated that non-neuronal cell contamination constituted less than one per cent of the cells in the cultures. The human DRG cultures were maintained for at least 2 weeks without signs of morphological degeneration.

Electron microscopy (EM) At 1, 2, 18, and 20 hours of incubation, postinfection, 6 to 10 DRG cultures were processed for EM. DRG cultures were fixed in Karnovsky's fixative for ultrastructural study, and subsequently fixed in 1 per cent osmium tetroxide, dehydrated through a series of alcohols and embedded in Agar Resin 100. Thin sections were prepared with an Ultratome Nova and placed on formvar and carbon-coated grids. Sections for electron microscopy were stained with uranyl acetate and lead citrate according to standard procedures and studied in a JEM--1200 EX microscope.

Results

Attachment, penetration, and axonal transport After 3 to l0 days o f incubation at 37 ° the large neuronal cells were well provided with extensions forming a m o r e or less complex neuritic network. In infected cultures virions were observed attached to the plasma m e m b r a n e of neurons and neurites and fusion of the viral envelope and plasma m e m b r a n e was frequently seen in cultures examined at I or 2 hours postinfection (Fig. 1). Images were obtained d e m o n s t r a t i n g the binding o f the projections on the virion envelope to the plasma m e m b r a n e . The binding seemed to occur all along the plasma m e m b r a n e and was n o t associated with coated pits. We were n o t able to detect internalization o f nucleocapsids by endocytosis, a prerequisite for subsequent endosome-involving fusion-reactions by which m a n y viruses are internalized [32]. At 2 hours postinfection naked nucleocapsids were f o u n d inside filopodia of g r o w t h cones, neuritic extensions and neuronal cell soma, suggesting a neuritic transport of HSV nucleocapsids (Fig. 2), in agreement with previously reported findings [17]. Nucleocapsids reaching the neuronal soma were often found close to the nuclear m e m b r a n e and near the nuclear pores. Several of these nucleocapsids appeared to be empty (Fig. 3). However, neither these or other viral nucleocapsids of the cytoplasm d e m o n s t r a t e d signs of degradation.

90

E. Lycke et al.

Fig. 1 a-d Virions attaching to (a, b) and fusing (b, e, d) with the plasma membrane at 2 h postinfection. Bars: a-e: 100nm; d: 200nm

Nucteocapsid envelopment and egress of virus Infected neurons observed at 20 h postinfection revealed nucleocapsids at the nuclear membrane in different stages of envelopment. Segments of the nuclear membrane of infected cells suggested a state of high activity with folding and formation of additional membranous structures associated with the inner lamina of the nuclear membrane (Fig. 4). Often small granular aggregates of electron dense material were observed inside foldings of the nuclear membrane. The membranous loops generally involved the inner lamina, often leaving the cy-


91

toplasma-facing counterpart of the membrane unfolded. Nucleocapsids in budding-like reactions were seen and enveloped virus appeared between the inner and outer laminae of the nuclear membrane, suggesting that the inner lamina was a site for primary envelopment (Fig. 5). Virions in such a position seemed sometimes attached to the outer but free from the inner lamina. However, nucleocapsids seemingly in a stage of envelopment also occurred at sites situated on the cytoplasmic side of the nuclear membrane. These findings were interpreted as indications of a transfer of nucleocapsids through the two laminae of the nuclear membrane by an envelopment--de-envelopment process. The cytoplasmic nucleocapsids were associated with membranes probably representing portions of the endoplasmic reticulum. Different degrees of enclosure of these nucleocapsids suggested that the endoplasmic reticulum represented another, second site for virus envelopment (Fig. 6).

Fig. 2. a-c Viral nucleocapsids inside long filopodium extending from a neuritic growth cone, at 2 h postinfection. The sites exhibiting the viral nucleocapsids are further enlarged in b and c. Note the many vesicles enmeshed between tubular cisternae of the smooth endoplasmic reticulum and mitochondria in the varicosity of the filopodium. Bars: 500, 100, and 200 nm, respectively, d Nucleocapsid inside neuritic extension close to microtubules. Bar: 100nm

E. Lycke et al. : HSV infection of human neurons

93

Fig. 3. a Empty HSV nucleocapsid close to the nuclear membrane of a neuronal cell at 18 h postinfection. On the upper side the section grazes the nuclear membrane so that the nuclear pores appear as dark circles in the membrane. In the middle of some of the pores, the central granule of the pores can be seen as a dense dot. Bar: 200 nm. b Viral nucteocapsid facing a nuclear pore at 18 h postinfection. Mitochondria, endoplasmatic reticulum, microtubules and neurofilaments are seen in the cytoplasm. Bar: 100nm In the cytoplasm of b o t h nerve cell bodies and neuritic extensions newly formed, enveloped virus appeared inside an additional m e m b r a n o u s structure, forming the wall of a transport vesicle. A n interesting finding was that sections o f these virion-vesMe complexes often revealed protrusions in a way suggesting that they might represent virion-vesicle complexes merging with smaller memb r a n o u s sacs (Fig. 7). Finally, fusion o f the wall o f the transport vesicles with the plasma connecting the vesicle with the extracetlular e n v i r o n m e n t seemed to be a way o f release o f virions from the infected cells (Fig. 8). In accord with this a s s u m p t i o n virions were d e m o n s t r a t e d extracellularly immediately outside the open vesicle.

Discussion We have studied by electron microscopy the HSV infection o f h u m a n sensory neurons. The ganglionic nerve cells are the final targets for the primary HSV infection and the source o f virus in recurrent infections. We found that in m a n y respects the morphogenesis o f HSV in the neuronal cells featured events similar to those observed in cells of n o n - n e u r o n a l origin. However, we also noted a n u m b e r of details n o t discussed previously, and which provide us with a novel

Fig. 4. a Folding of the nuclear membrane adjacent to viral nucleocapsids at 20 h, postinfection. Additional membranous structures on the nuclear side of the nuclear membrane and areas with a granular matrix. In the lower left part of the section an enveloped virus is seen in a cytoplasmic vesicle. Bar: 200 nm. b Nucleocapsids inside a loop of the nuclear membrane. Close connection between nuclear membrane and endoplasmic reticulum. Bar: 100 nm. e Multifolded inner lamina of the nuclear membrane. Note that the outer lamina is markedly less involved in the enwrapping process. In the lower right part of the figure is a virion enclosed in a transport vesicle. Bar: 200 nm. d Additional nuclear membranous structures enwrapping particulate matrix inside the nucleus, close to the nucleolus. Two intranuctear nucleocapsids are seen. Bar: 200 nm


95

hypothesis for the acquisition of envelope and the glycosylation of the viral peplomers during virus maturation. Several observations indicate that the viral glycoproteins building up to projections on the HSV envelope in various ways determine both the attachment and the fusogenic activities by which the virus is internalized [1, 7, 15, 25]. Fusion of HSV envelope and plasma membrane at the surface of the cell was first described by Morgan et al. [20]. Subsequently it has been "a tacit acceptance" that HSV penetration is effected by fusion rather than by endocytosis [23]. Recently Fuller and Spear [6] have presented further evidence for the relevance of fusion in the initiation of the HSV infection. Anti-glycoprotein D neutralizing antibodies inhibited penetration of HSV by blocking fusion of

96

E. Lycke et al.

Fig. 5. a Nucleocapsid budding at the nuclear membrane at 20 h, postinfection. Bar: 200 nm. b Virion between the laminae of the nuclear membrane. Note the virus envelope disengaged from the inner but attached to the outer lamina. Inset: Magnification x 160,000. Bar: 200 nm


97

Fig. 6. a-d. Endoplasmic reticulum partly enclosing nucleaocapsids at 20 h, postinfection. The reticulum in close contact with the nuclear membrane, b The section cuts a folding of the nuclear membrane so that a cytoplasmic part seems surrounded by the nucleus, c, d Nucleocapsids almost completely enclosed by membranes of the endoplasmic reticulum. Completely enwrapped virus has received a transport vesicle in addition to the envelope. Bars: 200 nm

98

E. Lycke et al.

Fig. 7. Two virions in transport vesicles surrounded by smaller membranous sacks, some seemingly fusing with the wall of the transport vesicles. At 20 h, postinfection. Bar: 100nm

virion envelope and plasma membrane. It has also been shown that agents that block endocytosis or raise the pH of endosomes do not interfere with initiation of HSV infection in Vero cells [13]. In accord, we have observed that treatment with chloroquine and NH4C1 to block pH-dependant endocytosis does not inhibit uptake of HSV into neurites of DRG cells [Svennerholm et al., unpublished]. Moreover, in cells with phagocytic activity the phagocytized virus generally remains within phagosomes and appears to be degraded [29]. Entry of HSV by endocytosis does not result in productive infection [Campadelli-Fiume et al., results presented at the Herpesvirus Workshop in Philadelphia, 1987]. The electron micrographs of HSV infected human DRG nerve cells of the present report are in complete agreement with the hypothesis of fusion as the important reaction for internalization of HSV nucleocapsids. We found no signs of incorporation of virus by endocytosis and our images produced with cultures fixed within the first 2 hours of infection demonstrated virions with their projections attached to the plasma membranes and viral envelopes which fused with the plasma membrane of both neurites and perikaryon. Internalized HSV nucleocapsids, and often empty capsids, were frequently observed in close connection with the nuclear membrane facing nuclear pores. Since DNA released in the cytoplasm will be degraded [9] it is probably essential that the nucleocapsid is structurally maintained in the cytoplasm until the DNAprotein complex is released into the nucleus. Observations that cytoplasmic nucleocapsids become rounded or irregular have been interpreted as signs of


99

Fig. 8. Exocytosis. Cytoplasmic vesicles surrounding virion, close to the plasma membrane (a), connected with the plasma membrane of a neuronal cell (be). The vesicle is open towards the extracellular environment (be) with virions immediately outside the cell at 20h, postinfection. Bars: a, b, d: 100nm; c: 50nm; e: 20nm

capsid disassembly [18, 29], but more probably the DNA-protein complex is released into the nucleus without gross structural changes of the capsid. The envelopment of the nucleocapsid is a complex process implicating the laminae of the nuclear membrane [3, 22] as well as cytoplasmic membranes [22, 31]. The HSV infection of human D R G cells exhibited several electron microscopical features previously demonstrated with infected cells of non-neuronal origin. Thus, we observed aggregation in the nucleus of particulate material in association with viral nucleocapsids appearing near the nuclear membrane, remarkable enlargements and foldings of especially the inner lamina of the

t O0

E. Lycke et al.


101

nuclear membrane with enclosure of nucleocapsids often without significant involvement of the outer lamina, and presence of virions between the nuclear laminae. There are at least three different ways by which de novo formed nucleocapsids can reach the cytoplasmic side of the nuclear membrane. Of these, active passage of the nucleocapsids through nuclear pores seems hardly possible, since the capsids are far too large. Rupture of the nuclear membrane has been observed [24] but it seems doubtful if a cell thus damaged would allow subsequent envelopment. A more probable process would be an envelopment at the inner lamina of the nuclear membrane and passage out by fusion of envelope and the outer membrane as originally suggested by Morgan et al. [19a] and later also by Stackpole [31]. In fact we observed virions between the nuclear membranes apparently disengaged from the inner lamella but closely attached to the inner side of the outer membrane. Thus, since we observed that nucleocapsids frequently were present outside the nuclear membrane but saw no transport of virions in endoplasmic cisternae we believe that an essential step in the morphogenesis of HSV involves de-envelopment of virions by fusion of envelope and the outer nuclear membrane and that in this way the nucleocapsids might gain entrance to the cytoplasm. Our findings suggest that a second, final envelopment of the virus occurs at the membranes of the endoplasmic reticulum. There the virus concomitantly receives a further enclosure, the transport vesicle. Results presented by Johnson and Spear [11] have suggested that virus envelopes acquired at the inner nuclear lamina contain immature glycoproteins and that the mature glycosylated forms of the viral glycoproteins are achieved when the virions pass the Golgi apparatus. We would, hypothetically, suggest on the basis of our electron microscopy image that the biosynthesis of the N-glycosidically linked oligosaccharides is confined to the endoplasmic reticulum [12] and a late post-translation modification with addition of O-linked sugars, a process which has been demonstrated to be Golgi dependent [11], takes place inside the virion-transport vesicle complexes and involves transport of glycosyl transferases from the Golgi to these complexes by primary-lysosome-like, enzyme containing vesicles pinched off from the Golgi apparatus. The lack of visible virus enveloped as well as absence of virion-transport vesicle complexes at the Golgi membranes may be support for this assumption. The O-glycosidic glycosylation of viral proteins is considered to be a result of cellular transferase activities [16] and all three subcompartments of the Golgi apparatus are supposedly involved in formation of O-glycosyl oligosaccharides [26]. A similar hypothesis, i.e., that herpesvirus maturation and completion of viral glycosylation is confined to cytoplasmic vacuoles and distinct from membranes of the nucleus, endoplasmic reticulum and the Golgi apparatus has been formulated by Montalvo et al. [19]. Observing the morphogenesis of varicellazoster virus in infected human melanoma cells and using the periodate-thiocarbohydrazide silver proteinate method they failed to detect viral glycocon-

102

E. Lycke et al.

jugates in the viral envelope at the nuclear membranes but got a positive glycoprotein staining of virus which was inside cytoplasmic vacuoles. In our study the egress of virions seemed to result from fusion of the vesicle wall with the plasma membrane. Enveloped virus in the cytoplasm was always enclosed in a transport vesicle whether present in the neuronal cytoplasma or in neuritic extensions. The vesicles, we believe, are of importance not only for protection and transport but also for egress of virus from the infected cell by fusion of the transport vesicle wall with the plasma membrane. Supplementary to these observations we noted that early in infection when nucleocapsids seemed to be transported in neurites in somatopetal direction they appeared closely related to the microtubules. The possible association of microtubular functions and axonal transport of HSV has been reported previously [14]. It is reasonable to assume that microtubular gliding and movement of organelles of the uninfected neuron might be the mechanisms utilized for the axonal transport of virus.

Acknowledgement The constructive discussions with Dr. H. A. Hanson, University of G6teborg, are acknowledged. Also we would like to acknowledge the proficiency of the photographic work done by Mr. Hans Jansson. Financial support was obtained from the Swedish Medical Research Council, grant 4514.

References 1. Claesson-Welsh L, Spear PG (1986) Oligomerization of herpes simplex virus glycoprotein B. J Virol 60:803-806 2. Dales S, Silverberg H (1969) Viropexis of herpes simplex virus by HeLa cells. Virology 37:475-480 3. Darlington RW, Moss LH (1968) III. Herpesvirus envelopment. J Virol 2:48-55 4. Epstein MA (1962) Observations on mode of release of herpes virus from infected HeLa cells. J Cell Biol 12:589-597 5. Fredman P, Jeansson S, Lycke E, Svennerhotm L (1985) A monoclonal antibody reacting specifically with ganglioside GD 1 b in human brain. FEBS Lett 189:23-26 6. Fuller AO, Spear PG (1987) Anti-glycoprotein D antibodies that permit adsorption but block infection by herpes simplex virus 1 prevent virion-cell fusion at the cell surface. Proc Natl Acad Sci USA 84:5454-5458 7. Gompels U, Minson A (1986) The properties and sequence of glycoprotein H of herpes simplex virus type 1. Virology 153:230-247 8. Hill TJ (1985) Herpes simplex virus latency. In: Roizman B (ed) The herpesviruses, rot 3. Plenum Press, New York London, pp 175-240 9. Hochberg E, Becker Y (1968) Adsorption, penetration and uncoating of herpes simplex virus. J Gen Virol 2:231-241 10. Hummeler K, Tommassini N, Zajac B (1969) Early events in herpes simplex virus infection: a radioautographic study. J Virol 4:67-74 11. Johnson DC, Spear PG (1983) O-linked oligosaccharides are acquired by herpes simplex virus glycoproteins in the Golgi apparatus. Cell 32:987-997 12. Kornfeld R, Kornfeld S (1985) Assembly of asparagine-linked oligosaccharides. Ann Rev Biochem 54:631-664


103

13. Koyama AH, Uchida T (1984) Inhibition of multiplication of herpes simplex virus type 1 by ammonium chloride and chloroquine. Virology 138:332-335 14. Kristensson K, Lycke E, R6ytt/i M, Svennerholm B, Vahlne A (1986) Neuritic transport of herpes simplex virus in rat sensory neurons in vitro. Effects of substances interacting with microtubular function and axonal flow (nocodazole, taxol and erythro-9-3-(2hydroxynonyl) adenine). J Gen Virol 67:2023-2028 15. Little SP, Jofre JT, Courtney R J, Schaffer PA (1981) A virion-associated glycoprotein essential for infectivity of herpes simplex virus type 1. Virology 115:149-160 16. Lundstr6m M, Jeansson S, Olofsson S (1987) Host cell induced differences in the Oglycosylation of herpes simplex virus gC- 1. II. Demonstration of ubiquitous cell-specific galactosyl transferases. Virology 161:395-402 17. Lycke E, Kristensson K, Svennerholm B, Vahlne A, Ziegler RJ (1984) Uptake and transport of herpes simplex virus in neurites of rat dorsal root ganglia cells in culture. J Gen Virology 65:55-64 18. Miyamoto K, Morgan C (1974) Structure and development of viruses as observed in the electron microscope. XI. Entry and uncoating of herpes simplex virus. J Virol 8: 910-918 19. Montalvo EA, Parmley RT, Grose C (1986) Varicella-zoster viral glycoprotein envelopment: ultrastructural cytochemical localization. J Histochem Cytochem 34: 281284 19a. Morgan C, Rose HM, Holden M, Jones EP (1959) Electron microscopic observations on the development of herpes simplex virus. J Exp Med 110:643-656 20. Morgan C, Rose HM, Mednis B (1968) Electron microscopy of herpes simplex virus. I. Entry. J Virol 2:507-516 21. Nemerow GR, Cooper NR (1984) Early events in the infection of human B lymphocytes by Epstein-Barr virus: the internalization process. Virology 132:186-198 22. Nii S, Morgan C, Rose HM (1968) Electron microscopy of herpes simplex virus. II. Sequence of development. J Virol 2:517-553 23. Roizman B, Batterson W (1985) Herpesviruses and their replication. In: Fields BN, et al (eds) Virology. Raven Press, New York, pp 497-526 24. Roizman B, Spear PG (1973) Herpesviruses. In: Dalton AJ, Haguenau F (eds) U1trastructure of animal viruses and bacteriophages, an atlas. Academic Press, New York, pp 1M13 25. Sarmiento M, Haffey M, Spear PG (1979) Membrane proteins specified by herpes simplex viruses. III. Role of glycoprotein (VP 7 (B 2) in virion infectivity. J Virol 29: 1149-1158

26. Schachter H (1986) Biosynthetic controls that determine the branching and the microheterogeneity of protein-bound oligosaccharides. Biochem Cell Biol 64:163-180 27. Schwartz J, Roizman B (1969) Concerning the egress of herpes simplex virus from infected cells. Electron and light microscopy observations. Virology 38:42~49 28. Siminoff P, Menefee MG (1966) Normal and 5-bromodeoxyuridine-inhibited development of herpes simplex virus: an electron microscopic study. Exp Cell Res 44: 241255 29. Smith JD, de Harven E (1974) Herpes simplex virus and human cytomegalovirus replication in Wl-38 cells. II. An ultrastructural study of viral penetration. J Virol 14: 945-956 30. Spear PG (1985) Glycoproteins specified by herpes simplex viruses. In: Roizman B (ed) The Herpesviruses. Plenum Press, New York London, pp 315-356 31. Stackpole CW (1969) Herpes-type virus of the frog renal adenocarcinoma. I. Virus development in tumor transplants maintained at low temperature. J Virol 4:75-93 32. White J, Kielian M, Helenius A (1983) Membrane fusion proteins of enveloped animal viruses. Q Rev Biophys 16:15t-195

104


33. Ziegler RJ, Herman RE (1980) Peripheral infection in culture of rat sensory neurons by herpes simplex virus. Infect Immun 28:620-623 Authors' address: Dr. E. Lycke, Department of Clinical Virology, University of G6teborg, Guldhedsgaten 10 B, S-413 46 G6teborg, Sweden. Received January 15, 1988