Ectromelia Virus-Induced Changes in Primary Cultures of Mouse Hepatocytes

J. gen. Virol. (1985), 66, 2171-2181. Printed in Great Britain 2171

Key words: ectromelia virus/poxvirus/primary cultures/hepatocytes

Ectromelia Virus-induced Changes in Primary Cultures of Mouse Hepatocytes

By D. N. LEES AND J. STEPHEN* Department of Microbiology, University of Birmingham, PO Box 363, Birmingham B15 2TT, U.K.

(Accepted 17 July 1985)

SUMMARY Mouse hepatocytes were isolated by collagenase perfusion, maintained in non- proliferating monolayer culture and shown to retain liver cell function as judged by gluconeogenesis for 15 to 18 h. Such cells could be infected with and support the replication of a virulent strain of ectromelia virus. Virus antigen and characteristic cytoplasmic 'B'-type poxvirus inclusion bodies were demonstrated by immunofluorescence in virtually all cells. By electron microscopy it was shown that 'B'-type inclusions were the site of virus replication, and that the biogenesis of ectromelia virus and ultrastructural changes in hepatocytes were similar to those observed in infected mouse livers. Early cell rounding effects, a normal characteristic of poxvirus infections in tissue culture cells, were not seen in ectromelia-infected hepatocytes, although late degenerative changes did occur. Pulse-labelling of hepatocyte cultures with [35S]methionine showed that ectromelia virus inhibited the rise in protein synthesis seen in controls and imposed a gradual decline in host protein synthesis to an extent and at a rate significantly different from that in mouse L929 cells. Gluconeogenesis was inhibited by ectromelia virus infection of hepatocytes.

INTRODUCTION In this paper we describe the establishment of a system involving infection of primary cultures of murine hepatocytes with virulent ectromelia virus, which is potentially capable of allowing the study under controlled conditions in vitro of virus-cell interactions relevant to the pathogenesis of disease caused by this virus. To our knowledge there is no infection caused by a pathogenic animal virus in a naturally susceptible host for which the biochemical mechanisms and/or determinants responsible for direct cell injury have been elucidated in vivo or in differentiated target cells in vitro. Oldstone et al. (1982) showed in vivo that lymphocytic choriomeningitis virus induces runting in mice by suppressing the synthesis of growth hormone in the pituitary, without cell necrosis or inflammation. Fukuda et al. (1983) studied in vitro the loss of electrophysiological responsiveness in nerve cells from the dorsal root ganglia of guinea-pigs infected in vivo with herpes simplex virus type 2. Some cells overcame the initial defect and regained their electrical properties and this was interpreted in terms of the virus becoming latent. The mechanisms and determinants of these virus-induced disease-related phenomena have not been elucidated. Potentially the most meaningful studies carried out in vitro are those involving viruses and macrophages although the mechanisms and determinants involved in the survival of macrophages or the destruction of the virus are still largely unclear (Mogensen, 1979). Even these experiments require caution in their design and interpretation since many studies are carried out with activated peritoneal macrophages as opposed to say liver Kupffer cells (Kirn & Keller, 1983), or alveolar macrophages (Rodgers & Mims, 1981). The species of origin has also been shown to be profoundly inportant: influenza virus kills mouse but not human alveolar macrophages (Rodgers & Mims, 1981, 1982). Kirn et al. 0983) have identified the macrophage- destroying component(s) of frog virus 3 (FV3) which leads to the degenerative hepatitis caused

0000-6631 © 1985 SGM 2172 D. N. LEES AND J. STEPHEN by intravenous injection of FV3; however, this is an intoxication since the virus does not infect mice. From studies in cultured cells it is clear that cytocidal viruses cause both biochemical and morphological lesions leading to cell death. Much effort has been expended in the elucidation of the mechanisms involved in such virus-induced inhibition of 'vital' cell functions (Shatkin, 1983; Fraenkel-Conrat & Wagner, 1984) and less on loss of 'luxury' or differentiated functions (Oldstone et al., 1982; Oldstone & Fujinami, 1982). The evidence, however, that such biochemical and morphological lesions are operative in vivo and are therefore responsible for specific pathological effects is largely missing. While, ultimately, there is no substitute for in vivo studies, it is arguable that primary cultures and/or organ cultures (Porter et al., 1980) could provide systems amenable for experimentation which are potentially more relevant to events occurring in the intact animal. However, organ cultures suffer from some of the limitations imposed on in vivo experiments, namely the difficulty in establishing a synchronous virus infection, and also in directing the virus exclusively to the target cell of interest. Since it is clear that the nature of virus-induced inhibition of protein synthesis is in some cases host cell-dependent (Jen et al., 1980; Jen & Thach, 1982; Otto & Lucas-Lenard, 1980; Harper et aL, 1979), it is important that virus-induced cytopathology should be investigated wherever possible in relevant cells; primary cell cultures, once removed from the animal, could fulfil some of the necessary criteria. One such system is primary hepatic parenchymal cells (Seglen, 1976). The latter retain for a time several liver-specific functions in vitro (Yamada et al., 1980; Smith & Hammarstrom, 1979; Maslansky & Williams, 1982; Michalopoulos et al., 1978), and have been exploited in the study of several pathogens (Arnheiter et al., 1982; Taguchi et al., 1983; Isom, 1980; Arnheiter, 1980; Kawamura et al., 1983). Ectromelia virus is a poxvirus which is a natural pathogen of the mouse causing the disease mousepox. The pathogenesis of ectromelia infection involves sequential spread of virus from the primary lesion to internal target organs (mainly liver and spleen) and thence back to the epidermis to cause the characteristic external lesions (Fenner, 1949; Mims, 1964). The degree of hepatic involvement is a major factor in determining the outcome of the infection. Thus, infection of primary cultures of mouse hepatic parenchymal cells with virulent ectromelia virus would enable the cytopathogenicity of this virus to be studied in an important target cell only one step removed from the in vivo situation. In this paper, we describe the results of an initial study in which we have concentrated on establishing the cell system and looked at a small number of virus-induced effects on such cells. In particular, attention was paid to the morphological sequelae to infection, inhibition of hepatocyte protein synthesis, and inhibition of gluconeogenesis. The latter was chosen as the test for retention of liver cell function (Yamada et al., 1980). Where relevant, comparisons were made between effects in hepatocytes and mouse L929 cells, an established murine cell line.

METHODS Mice. Outbred white mice 12 to 16 weeks old (30 to 45 g) were used for the preparation of hepatocyte cultures. Hepatocyte cultures. At the outset of this work little published data existed on techniques for obtaining isolated mouse hepatocytes. Methods for obtaining rat hepatocytes (Seglen, 1976; Page & Garvey, 1979) were therefore adapted for the mouse. The basic perfusion buffer (pH 7-4) consisted of 142 mu-NaC1, 6.7 mM-KCI, 25 mM- N aHCO3, gassed with 95 ~ air/5 ~ CO,_ and kept at 37 °C. Mice were anaesthetized with pentobarbitone (May & Baker Ltd, Dagenham, Essex, UK. ; 50 ms/ks body weight) and treated with heparin (500 units intravenously). The liver was then perfused in situ via the portal vein at a flow rate of 1 ml/min. Perfusion cycles were carried out using: (i) basic perfusion buffer pH 7.4 for 15 to 20 min; (ii) basic perfusion buffer containing 0.5 mM-EGTA for 5 min; (iii) basic perfusion buffer for 5 min; (iv) basic perfusion buffer containing 5.0 mM-CaCI2, 0.07% (w/v) collagenase (Sigma, type IV) for 20 min. Following perfusion cycle (iv), livers were excised, teased apart, and incubated at 37 °C on a rotary shaker for a further 10 min in the same solution. After filtration through coarse nylon mesh, crude cell suspensions were sedimented by centrifugation (50 g, 5 min, room temperature) and resuspended in chilled washing buffer (142 mM-NaC1, 6.7 mM-KCI, 1.2 mM-CaCI2, 10 mM-HEPES, pH 7-4). This was repeated twice and cell numbers adjusted to 4 × 105 viable (i.e. trypan blue-excluding) cells/mt in Williams' medium E (Flow Laboratories) supplemented with 10~ (v/v) foetal bovine serum (IDRL, Birmingham University), 2.0 mM-glutamine, 0.01 ~ (w/v) gentamicin, 200 mU insulin per ml (Sigma) and 1 ~tM-dexamethasone Ectromel•-infected hepatocytes 2173

(Sigma). Two ml of suspension was seeded into collagen-coated 35 mm plastic Petri dishes [0-2 ml of 0-15 ~ (w/v) bovine dermal collagen (Flow Laboratories) was added to dishes which were then air-dried] with or without collagen-coated plastic coverslips (Thermanox, Flow Laboratories) and incubated at 37 °C in a humid atmosphere of 5 ~ CO., in air. After 2 to 3 h, medium containing unattached cells was aspirated and replaced with fresh culture medium. L. cells. Mouse L929 cells (provided by Dr I. Allan, Birmingham University) were grown in Eagle's MEM, supplemented with 10~o (v/v) bovine serum, 2.0 mM-glutamine and 0.01 ~ (w/v) crystamycin (Glaxo). Virus. The virulent Moscow strain of ectromelia virus (Fenner, I949) was obtained from Professor F. Fenner (Australian National University, Canberra, Australia) as freeze-dried infected mouse spleen. For virus production confluent monolayers of L929 cells in MEM were inoculated with reconstituted virus (2 p.f.u./cell) and incubated for 2 to 3 days at 35 °C. Infected cells were harvested by shaking and centrifugation (1800 g, 15 min, 4 °C) and stored at - 70 °C. Before use virus stocks were sonicated and titrated by plaque assay under a CM-cellulose overlay (Blanden, 1970). To minimize the possibility of attenuation, virus was passed no more than 5 times in L929 cells and the LDs0 values of virus stocks were shown to be below 10 p.f.u, following intraperitoneal inoculation of 12- to 14-week-old outbred white mice. Injection ofhepatocytes. After incubation of hepatocyte cultures for 4 to 5 h medium was aspirated and each dish inoculated with ectromelia (or vaccinia) virus diluted in 2 ml Williams' medium E to give either 5 or 25 p.f.u./cell. Dishes were reincubated at 35 °C in a humidified atmosphere of 5~ CO2 in air. Microscopical observations. For direct microscopical observation, monolayers on coverslips were fixed with methanol and stained with Giemsa. For immunofluorescent analysis, coverslips were fixed with ethanol and stained using an indirect sandwich technique utilizing rabbit anti-vaccinia virus immunoglobulin (Birkbeck & Stephen, 1970) and sheep anti-rabbit fluorescein immunoconjugate (Wellcome Diagnostics). For electron microscopy, coverslips were fixed with 2~ (v/v) glutaraldehyde and post-fixed with 29/0 (w/v) osmium tetroxide. After embedding in Araldite resin and removing the coverslip, horizontal ultrathin sections of the monolayer culture were made and stained with uranyl acetate and lead citrate. Specimens were examined using a Philips 201 electron microscope at 60 kV. Protein synthesis. Cell monolayers were washed twice with Dulbecco's phosphate-buffered saline (PBS) and pulse-labelled with 0.3 ml [35S]methionine (40 gCi/ml; New England Nuclear) in hepatocyte buffer (68 mlvl-NaC1, 5.3 mM-KCI, 1.2 mM-CaCI2, 6.4 mM-MgCI2, 1.1 mM-KH2PO4, 0.7 mM-Na2SO4, 30 mM-HEPES, 30 mM-TES, 36 mM-tricine, pH 7.5) for 15 min (35 °C). Following addition of 1 ml of Williams' medium E, monolayers were incubated for a further 15 min before being washed twice with PBS, scraped off into 0. l ml PBS using a rubber policeman and stored at - 70 °C. Ten gl samples were assayed for total radioactivity by TCA precipitation. SDS- PAGE was based on the method of O'Farrell (1975): 3sS-labelled cell suspensions were diluted 1:1 in lysis buffer containing 2 ~ (w/v) SDS and 2 ~ (v/v) 2-mercaptoetbanol, boiled for 5 rain and 5 to 50 gl (2 x 104 d.p.m.) layered into each sample well. Gels (l 2 ~o and 4 ~/polyacrylamide in running and stacking gels respectively) were run at a constant g mA until completion, dried and exposed to presensitized X-ray film (X-Omat RP, Kodak) at - 70 °C for 14 days. Autoradiographs were developed in DX-80 (Kodak). Wherever possible, equivalent amounts of radioactivity were loaded onto each lane. Gluconeogenesis. The assay was a modification of those reported by Assimacopoulos-Jeannet et al. (1973) and Page & Garvey (1979). Cell monolayers were washed twice with PBS and incubated in 0.5 m150 p.M-lactate solution in hepatocyte buffer containing 1.1 ~tCi/ml L-[U-~4C]lactate (Amersham) for 30 min at 35 °C before being scraped off using a rubber policeman and stored at - 20 °C. Cell suspensions were lysed by ultrasonication, deproteinized by addition of 1 ml 20 mM-ZnSO~ followed by 1 ml 20 mM-Ba(OH)2 and centrifuged (2000 g, 5 min). One ml of supernatant was mixed with 0.1 g Dowex 50W (Fluka) and 0.4 g Amberlite MB-3 (Fluka) resins for 30 rain and decanted. A 0.1 ml amount was assayed for radioactivity by counting in 10 ml scintillation fluid (ACS If, Amersham) on a Beckman LS 7500 liquid scintillation counter.

RESULTS Hepatocyte cultures The hepatocyte isolation technique yielded 0.6 x 106 to 1.2 × 106 cells/g body weight with viabilities of 75 to 85 ~ as judged by trypan blue exclusion. Contamination by non-parenchymal cells was less than 5~. By 2 h post-seeding, 50~ of parenchymal cells had attached with viabilities of > 95 ~o. The mouse hepatocyte isolation and primary culture techniques reported here are similar to those published following the commencement of this study by Klaunig et al. (1981 a, b). 2174 D. N. LEES AND J. STEPHEN

Fig. 1. Ectromelia virus-infected mouse hepatocyte. Cultures were infected at a multiplicity of 25 p.f.u./ce|l and examined 18 h post-infection. Characteristic poxvirus 'B'-type inclusion (B) contained various immature virion developmental stages. Note rounded mitochondria (m), a degree of cytoplasmic vacuolation (v) and a total absence of cytoplasmic glycogen. The latter was readily demonstrable in control cells (not shown). Bar marker represents 1 ~tm. Microscopical observations Hepatocyte monolayers were inoculated with ectromelia virus at a m.o.i, of 25 p.f.u./cell and incubated at 35 °C for varying periods of time. At 8 to 10 h post-infection, Giemsa staining revealed small eosinophilic cytoplasmic inclusion bodies in over 90~ of cells. By 18 h post- infection, these virus inclusions, ranging in number from 8 to 20 per cell, were clearly delineated within the cytoplasm of virtually all hepatocytes. At this time, virus antigen was demonstrable by immunofluorescence in over 95 ~ of cells; many cells also showed focal concentrations of antigen corresponding in size and shape to the eosinophilic inclusions. Ultrastructural examination showed that inclusion bodies were the site of virus replication and can therefore be regarded as 'B'-type inclusions (Moss, 1974) (Fig. 1 and 2). 'A'-type inclusions (Ichihashi et al., Ectromelia-infected hepatocytes 2175

Fig. 2. Ectromelia virus-infected mouse hepatocytes. Cultures were infected at a multiplicity of 25 p.f.u./cell and examined 18 h post-infection. Specific cytoplasmic fluorescence was observed in virtually all cells. Inset shows focal concentrations of virus antigen corresponding to 'B'-type inclusions (arrow). Note (i) there is a complete absence of any cell rounding in infected cells and (ii) apparent fluorescence in detached rounded up cells is non-specific and was observed in control cultures.

1971) were not observed in ectromelia virus-infected hepatocyte monolayers at the light or electron microscopical level. Virus-induced ultrastructural changes in hepatocytes included the loss of all cytoplasmic glycogen, mitochondrial rounding and a degree of cytoplasmic vacuolation (Fig. 1). Similar ultrastructural changes were reported by Leduc & Bernhard (1962) in infected liver cells in mice suffering from mousepox. Poxvirus-induced c.p.e, in cultured cells include early cell rounding, cell fusion and ultimate degenerative changes (Bablanian, 1984). No evidence for cell rounding in cells infected with a m.o.i, of 25 (or 200) was observed even as late as 18 h post-infection (Fig. 2). By comparison, ectromelia virus-induced early cell rounding was clearly demonstrated in monolayers of mouse L929 cells under similar conditions of infection (Fig. 3), in agreement with the data of Ozaki & Higashi (1959). Late degenerative changes including cytoplasmic vacuolation, nuclear pyknosis and inability to exclude trypan blue became obvious in ectromelia virus-infected hepatocytes 30 to 40 h post-infection. Cell fusion was not observed in infected hepatocyte cultures.

Growth kinetics The virus growth curve showed an approximate 100-fold increase in virus titre with maximum yield occurring at 24 h post-infection (Fig. 4). This, together with the simultaneous appearance of'B'-type inclusions in virtually all cells, demonstrates that cultured hepatocytes are permissive for ectromelia virus replication; a one step synchronous growth cycle occurred under these infection conditions. Protein synthes& In control cultures incorporation of radiolabel rose slowly with time (Fig. 5). In contrast, incorporation into infected hepatocytes underwent a very gradual decline (Fig. 5). Polypeptide production in [3SS]methionine pulse-labelled cultures of hepatocytes and L929 cells was followed by SDS-PAGE (Fig. 6, 7). The similarities, differences and difficulties in interpretation are discussed below. 2176 D. N. LEES AND J. STEPHEN

Fig. 3. Ectromelia virus-infected L929 cells. Cultures were infected at a multiplicity of 25 p.f.u./celt and examined 7 h post-infection. Specific cytoplasmic fluorescence was observed in all cells; "B'-type inclusions are arrowed. Note extensive cell rounding.

50 i , ,

25 I I I I I I 40

X ~" 20 ~ 3O

o o -~ 2o ~ 10

~-" 5

I I I I 5 l0 15 20 25 30 5 10 15 20 Time post-infection (h) Time post-infection (h) Fig. 4 Fig. 5 Fig. 4. Kinetics of virus production in ectromelia virus-infected hepatocytes. Cells were infected at a multiplicity of 5 p.f.u./cell. At each time point duplicate monolayers were washed with PBS, scraped into 1 ml MEM, ultrasonicated and virus titres assayed as described in the text. Fig. 5. Incorporationof[35S]methionineinto controlandectromeliavirus-infectedmousehepatocy tes. Cultures were infected with 25 p.f.u./cell (or mock-infected) and incubated at 35 °C. At each time point duplicate monolayers were pulse-labelled with [35S]methionine for 15 min followed by a further 15 min incubation in Williams' medium E. The incorporation of radiolabel into TCA-precipitable material was then assayed. C), Control; O, infected. Gluconeogenesis Gluconeogenic rates in control cultures remained fairly constant for 15 to 16 h; thereafter, progressive deterioration occurred. In contrast, gluconeogenesis in infected hepatocyte monolayers was fairly rapidly inhibited; 50~ inhibition was reached by 6 to 7 h post-infection with subsequent progressive inhibition (Fig. 8). Ectromelia-inJected hepatocytes 2177 Control Infected 5 7 10 20111 3 5 7 10 20

Fig. 6. SDS PAGE profiles of polypeptide production in control and ectromelia virus-infected mouse hepatocytes. Cultures were infected and radiolabelled as for Fig. 5. SDS-PAGE and autoradiography were performed as described in the text. Numbers indicate times post-infection in hours. Representative virus (v) polypeptides are marked. The synthesis of hepatocyte polypeptides remained constant throughout incubation in control cultures, and was slowly inhibited in infected cultures. This compares with the rapid shut-off in infected L929 cells (Fig. 7). The virus-specific polypeptide profile was similar in both ectromelia-infected hepatocytes and L cells (Fig. 7), insofar as comparisons could be made due to the lack of shut-off and low incorporation levels in hepatocytes.

DISCUSSION Primary cultures of murine hepatocytes retained liver function as judged by gluconeogenesis for an appropriate period of time, and ectromelia virus-induced changes in isolated cells were remarkably similar to those observed in the whole animal. Gluconeogenesis is an index of cellular biochemical integrity since it represents a complex metabolic pathway with many intermediate steps for lactate utilization in glucose production (Hanson & Mehlman, 1976). Several workers have investigated gluconeogenesis in isolated hepatocyte suspensions from a variety of species (Page & Garvey, 1979; Zaleski & Bryla, 1977) but few have looked at the duration of the gluconeogenic response in hepatocyte primary cultures. Yamada et al. (1980) assayed gluconeogenic rates in primary cultures of rat hepatocytes after 24 and 48 h in culture, and showed a marked reduction by the latter time. Our data (Fig. 8) comprise a more detailed statement of the onset of this functional decline. One important point is the comparatively short duration of the retention of gluconeogenic function: approximately 15 h. While the cells retained morphological and ultrastructural integrity for much longer, our stated objectives could only be fulfilled within this 15 h period. The biogenesis of ectromelia virus (Fig. 1) was indistinguishable from that of ectromelia virus in infected mouse livers (Leduc & Bernhard, 1962) or that of other poxviruses (Moss, 1974). In 2178 D. N. LEES AND J. STEPHEN

Control Infected 11 10201 I I 3 5 7 l0 I

Fig. 7. SDS-PAGE-profiles of polypeptide production in control and ectromelia virus-infected L929 cells. Cultures were infected and radiolabelled as for Fig. 5. SDS-PAGE and autoradiography were performed as described in the text. Numbers indicate times post-infection in hours. Representative virus (v) polypeptides are marked. Synthesis of host polypeptides was rapidly inhibited; large reductions occurred by 5 h post-infection and almost complete inhibition by 7 h post-infection.

20 - i 0 I I I

× 15

I I I I 5 i0 15 20 25 Time post-infection (b) Fig. 8. Gluconeogenesis in control (O) and ectromelia virus-infected (0) hepatocytes. Cultures were infected at a multiplicity of 25 p.f.u./cell (or mock-infected) and incubated at 35 °C. At each time point [~C]lactate was added to duplicate monolayers for 30 min. The de novo production of labelled glucose was then assayed.

contrast, the response of the cells was significantly different to that normally seen in a wide range of poxvirus-infected tissue culture cells, but remarkably similar to that observed in ectromelia virus-infected mouse livers. The lack of ectromelia virus-induced cell rounding in isolated hepatocyte monolayers, in Ectromelia-infected hepatocytes 2179 dramatic contrast to L929 cells (Fig. 3), makes the relevance of this phenomenon in mousepox highly questionable. This result indicates that the interaction of ectromelia virus (or its proteins) with hepatocytic cell membranes and cytoskeletal structures may be quite different to that in L929 cell and other cell lines. Limited experiments on vaccinia virus-infected hepatocytes (data not given) showed that this poxvirus behaved similarly to ectromelia virus as judged by yield from comparable inocula and the lack of early cell rounding. The main difference between the two systems was the nature of the eosinophilic 'B'-type inclusions observed at 18 h post- infection. For vaccinia virus infection these were larger and more diffuse than their ectromelia virus counterparts. The study of protein synthesis inhibition was unexpectedly complicated by the pattern of incorporation of radiolabel in uninfected cells (Fig. 5). The reasons for the slow increase with time are not known but two possible explanations can be put forward. Firstly, there may be an actual increase in the synthesis of proteins. However, no variation in polypeptide profiles was observed throughout incubation (Fig. 6); in fact other workers have reported reductions in total protein recovered from hepatocyte cultures during incubation (Burrows & Davidson, 1982). Secondly, Seglen et al. (1980) observed in hepatocytes a decline in the rate of protein catabolism and hence in internal amino acid pool size during the first 20 h in culture. This could cause a greater demand for exogenous amino acids leading to increased radiolabel incorporation for the same overall rate of protein synthesis. No such complications are observed in L929 and HeLa cells. There are at least two ways in which inhibition of protein synthesis can be interpreted in ectromelia virus-infected hepatocytes. Firstly, ectromelia virus infection of hepatocytes inhibits the increase in protein synthesis seen in controls and imposes only a gradual shallow decline in protein synthesis (Fig. 5) with little change in the profile of host polypeptide patterns throughout 10 h post-infection (Fig. 6) when subsequent time points are compared to the initial time point. This is in marked contrast to the result obtained in L929 cells (Fig. 7) in which ectromelia virus induced a dramatic cessation in host protein synthesis as frequently reported for other poxviruses in cultured cells (Bablanian, 1984). The difference was not due to inefficient infection of hepatocyte monolayers since all of the cells were demonstrably infected as judged by immunofluorescent staining (Fig. 2), or inefficient replication of virus since the increase in virus yield from either hepatocytes (Fig. 4) or L929 cells (data not shown) was approximately 100-fold. Despite the differences induced by ectromelia virus in hepatocytes and L929 cells, careful inspection of the gel profiles reveals that all the virus-specified peptides visible in the hepatocytes were present in the L929 cell profile (Fig. 6, 7). Two factors mitigate against a more critical evaluation of this point. First, the total amount of radiolabel incorporation at 10 h in the hepatocyte system was very low (Fig. 6). Second, the dramatic shut down in cell synthesis in L929 cells makes the observation of virus-specified peptides easier and increases the probability of radiolabel being directed into virus peptides; a similar problem affected the choice of cell line for the work described by Harper et al. (1979). This interpretation is favoured by us. However, a second interpretation is possible. If one expressed the radiolabel incorporation of infected cells as a percentage of the time-matched controls, then the picture is one of more dramatic shut down with a rate not significantly different to that seen for gluconeogenesis (Fig. 8). Ideally, the experiment should be done with cells in equilibrated protein synthesis, but this would not be possible if one wished to preserve as much as possible demonstrable liver cell function as judged, in this case, by gluconeogenesis. The inhibition of gluconeogenesis was unequivocably rapid (Fig. 8). At present, we do not know whether this is a selective inhibition or a non-specific consequence of virus-cell infection. It will be of interest to examine infected animals for possible hypoglycaemia during and after infection of the liver since, in addition to inhibition of gluconeogenesis, Nossal & de Burgh (1954) reported loss of glycogen in ectromelia virus-infected mouse livers, a fact also observed by us in infected hepatocytes (Fig. 1). In conclusion, the results in this paper diminish the relevance of cell rounding and dramatic shut down in protein synthesis, and draw attention to the potential significance of specific direct virus-induced biochemical effects as important aspects of ectromelia virus-induced cell injury in 2180 D. N. LEES AND J. STEPHEN vivo. Moreover, the nature of the system is such that it should be possible to examine further, interference with vital cell functions as well as gluconeogenesis and other specific liver cell functions. It will also be of interest to examine the effects on hepatocytes of an avirulent strain of ectromelia.

We thank the Medical Research Council for a studentship to D.N.L.

REFERENCES ARNHEITER, H. (1980). Primary monolayer culture of adult mouse hepatocytes - a model for the study of hepatotropic viruses. Archives of Virology 63, 11-22. ARNHEITER, H., BAECHI, T. & HALEER, O. (1982). Adult mouse hepatocytes in primary monolayer culture express genetic resistance to mouse hepatitis virus type 3. Journal of Immunology 129, 1275-1281. ASSIMACOPOULOS-JEANNET,F., EXTON, J. H. & JEANRENAUD,B. (1973). Control of gluconeogenesis and glycogenolysis in perfused livers of normal mice. American Journal of Physiology 225, 25 32. BABLANIAN,R. (1984). Pox virus cytopathogenicity: effects on cellular macromolecular synthesis. In Comprehensive Virology, vol. 19, pp. 391 429. Edited by H. Fraenkel-Conrat & R. R. Wagner. New York: Plenum Press. BIRKBECK, T. H. & STEPHEN, J. (1970). Specific removal of host-cell or vaccinia-virus antigens from extracts of infected cells by polyvalent disulphide-linked immunosorbents. Journal of General Virology 8, 133-142. BLANDEN, R. V. (1970). Mechanisms of recovery from a generalized viral infection: mousepox. I. The effects of anti-thymocyte serum. Journal of Experimental Medicine 132, 1035-1054. BURROWS, R. B. & DAVIDSON, P. F. (1982). Protein catabolism in cultures of hepatocytes derived from mice of various ages. Mechanisms of Aging and Development 19, 85 94. rUNNER, F. (1949). Mouse-pox (infectious ectromelia of mice): a review. Journal of Immunology 63, 341-373. FRAENKEL-CONRAT,H. & WAGNER, R. R. (editors) (1984). Virus Cytopathology. Comprehensive Virology, vol. 19. New York: Plenum Press. FUKUDAJ., KURATE,T., YAMAMOTO,A. & YAMAGUCH,K. (1983). Morphological and physiological studies on cultured nerve cells from guinea pigs infected with herpes simplex virus in vivo. Brain Research 262, 79 89. 14ANSON, g. W. & ME14LMAN,M. A. (1976). Gluconeogenesis : Its Regulation in Mammalian Species. New York: John Wiley. HARPER, L., BEDSON, H. S. & BUCHAN, A. (1979). Identification of orthopox-virus by polyacrylamide gel electrophoresis of intracellular polypeptides. Virology" 93, 435-444. IC141HASHI, Y., MATSUMOTO,S. & DALES, S. (1971). Biogenesis of poxviruses; role of A-type inclusions and host cell membranes in virus dissemination. Virology 46, 507-532. ISOM, 14. C. (1980). DNA synthesis in isolated hepatocytes infected with herpes viruses. Virology 103, 199-216. JEN, G. & THAC14,R. E. (1982). Inhibition of host translation in encephalomyocarditis virus-infected L cells: a novel mechanism. Journal of Virology 43, 250 261. JEN, G., DETJEN, B. M. & THACH, R. E. (1980). Shut offof HeLa cell protein synthesis by encephalomyocarditis virus and poliovirus: a comparative study. Journal of Virology 35, 150 156. KAWAMURA,S., TAGUCHI, F., ISHIDA, T., NAKAYAMA,M. & FUJIWARA, K. (1983). Growth of Tyzzer's organism in primary cultures of adult mouse hepatocytes. Journal of General Microbiology 129, 277-283. KIRN, A. & KELLER, F. (1983). How viruses may overcome non-specific defences in the host. Philosophical Transactions of the Royal Society B 303, 115 122. KIRN, A., GUT, J.-P., BINGEN, A. & STEFFAN, A.-M. (1983). Murine hepatitis induced by frog virus 3: a model for studying the effect of Sinusoidal cell damage on the liver. Hepatology 3, 105-111. KLAUNIG, J. E., GOLDBLATT,P. J., HINTON, D. E., LIPSKY, M. M., CHACKO,J. & TRUMP, B. F. (1981 a). Mouse liver cell culture. I. Hepatocyte isolation. In Vitro 17, 913-925. KLAUNIG, J. E., GOLDBLATT,P. J., HINTON, D. E., LIPSKY, M. M., CHACKO, J. & TRUMP, B. F. (198l b). Mouse liver cell culture. II. Primary culture. In Vitro 17, 926 934. LEDUC, F. 14. & BERNHARD, W. (1962). Electron microscopy study of mouse liver infected by ectromelia virus. Journal of Ultrastructure Research 6, 466488. MASLANSKY,C. J. & WILLIAMS,G. H. (1982). Primary cultures and the levels of cytochrome P450 in hepatocytes from mouse, rat, hamster and rabbit liver. In Vitro 18, 683 693. MICHALOPOULOS,G., SATTLER, G. L. & PITOT, 14. C. (1978). Hormonal regulation and the effects of glucose on tyrosine aminotransferase activity in adult rat hepatocytes cultured on floating collagen membranes. Cancer Research 38, 1550-1555. MIMS, C. A. (1964). Aspects of the pathogenesis of virus disease. Bacteriological Reviews 28, 30-71. MOGENSEN, S. C. (1979). Role of macrophages in natural resistance to virus infections. Microbiological Reviews 43, 1 26. MOSS, B. (1974). Reproduction of poxviruses. In Comprehensive Virology, vol. 3, pp. 405 474. Edited by H. Fraenkel-Conrat & R. R. Wagner. New York: Plenum Press. NOSSAL, G. J. V. & DE BURGH, P. M. (1954). Ectromelia virus multiplication in mouse liver. Journal of General Microbiology 10, 345-352. O'FARRELL, P. H. (1975). High resolution two-dimensional electrophoresis of proteins. Journal of Biological Chemistry 250, 4007-4021. Ectromelia-inJected hepatoeytes 2181

OLDSTONE, M. B. A. & FUJINAMI, R. S. (1982). Virus persistence and avoidance of immune surveillance: how measles can be induced to persist in cells, escape immune assault and injure tissues. In Virus Persistence, pp. 185-202. Symposia of the Society]or General Microbiology, vol. 33. Edited by B. W. J. Mahy, A. C. Minson & G. K. Darby. Cambridge: Cambridge University Press. OLDSTONE, M. B. A., SINHA, Y. N., BLOUNT, P., TISHON, A., RODRIGUEZ, M., VON WEDEL, R. & LAMPERT, P. W. (1982). Virus-induced alterations in homeostasis: alterations in differentiated functions of infected cells in vivo. Science 218, 1125-1127. OTTO, M. J. & LUCAS-LENARD, J. (1980). The influence of host cell on the inhibition of virus protein synthesis in cells doubly infected with vesicular stomatitis virus and mengovirus. Journal of General Virology 50, 293-307. OZAKI, ¥. & HIGASHI,N. (1959). 3-a. Studies on the growth of viruses of ectromelia and vaccinia in strain L cells and HeLa cells. Annual Report, Institute of Virus Research, Kyoto University, Series B, 2, 65-113. PAGE, D. T. & GARVEY, J. S. (1979). Isolation and characterization of hepatocytes and Kupffer cells. Journal of Immunological Methods 27, 159-173. PORTER, D. D., MUCK, K. B. & PRINCE, G. A. (1980). The age dependence of respiratory syncytial virus growth in ferret lung can be shown in organ and monolayer culture. Clinical Immunology and lmmunopathology 15, 415-423. RODGERS, B. C. & MIMS, C. A. (198 l). Interaction of influenza virus with mouse macrophages. InJbction and Immunity 31, 751-757, RODGERS, B. C. & MIMS, C. A. (1982). Influenza virus replication in human alveolar macrophages. JournalofMedical Virology 9, 177-184. SEGLEN, P. O. (1976). Preparation of isolated rat liver cells. Methods in Cell Biology 13, 29-83. SEGLEN, P. O., SOLHEIRN, A. E., GRINDE, B., GORDON, P. B., SCHWARZE, P. E., GJESSING, R. & POLl, A. (1980). Amino acid control of protein synthesis and degradation in isolated rat hepatocytes. Annals of the New York Academy of Sciences 349, 1 17. SHATKIN, A. J. (1983). Molecular mechanisms of virus-mediated cytopathology. Philosophical Transactions of the Royal Society B 303, 167-178. SMITH, C. I. & HAMMARSTROM, L. (1979). Detection of albumin-secreting hepatocytes at the single cell level by a protein A plaque assay. European Journal of Immunology 9, 570-571. TAGUCHI, r., gAWAMURA, S. & FUJIWARA, K. (1983). Replication of mouse hepatitis viruses with high and low virulence in cultured hepatocytes. Injection and Immunity 39, 955-959. YAMADA, S., OTTO, S. P., KENNEDY, L. D. & WHAYNE, T. F. (1980). The effects of dexamethasone on metabolic activity of hepatocytes in primary monolayer culture. In Vitro 16, 559 570. ZALESKI, J. & BRYLA, J. (1977). Effects of oleate, palmitate and octanoate on gluconeogenesis in isolated rabbit liver cells. Archives of Biochemistry and Biophysics 183, 553-562.

(Received 1 April 1985)