Persistent Herpes Simplex Virus Infections Established in Two Burkitt Lymphoma Derived Cell Lines

Transcription

1 J. gen. ViroL 0976), 32, Printed in Great Britain 51 Persistent Herpes Simplex Virus Infections Established in Two Burkitt Lymphoma Derived Cell Lines By W. G. ROBEY, B. J. GRAHAM, C. L. HARRIS,* M. J. MADDEN, G. R. PEARSON-~ AND G. F. VANDE WOUDE Viral Biology Branch, National Cancer Institute, Bethesda, Maryland 2ooi4, U.S.A. (Accepted 3 March I976 ) SUMMARY Examination of P3HR-I cells (Epstein-Barr virus [EBV] producer) persistently infected with the MAL strain of herpes simplex virus type I (HSV-I) suggested that only a few cells were actively producing a virus indistinguishable from HSV- ~ (MAL) despite the presence of immunofluorescent HSV-~ antigens associated with the majority of cells. EBV-specific immunottuorescence was not altered in HSV-I persistently infected P3HR-I cells. HSV-~ persistently infected cells, labelled for 72 h with 14C-thymidine, incorporated approx. 8 % of the label into cell associated HSV-I DNA as resolved by caesium chloride gradients. Values greater than 8 % of the total were suggested by hybridization of gradient fractions with 3H-HSV- I DNA. To determine whether the establishment of HSV persistent infections in Burkitt lymphoma derived cells was a general phenomenon, six strains of HSV-I (MAL, KOS, Patton, Syn R, BF and Syn V) and two strains of type 2 (333 and MS) were used to infect the P3HR-I and Raji (EBV non-producer) cell lines derived from Burkitt lymphomas. In P3HR-I cells, persistent infections were established with all strains of HSV-~ but not with HSV-2. In Raji cells, persistent infections were established with all strains of HSV-I, except Syn V, and with both strains of HSV-2. No external support was required to maintain these infections. INTRODUCTION Studies performed by Henle, Henle & zur Hausen 0969) and Floyd et al. 097 I) with several Burkitt lymphoma (BL) derived cell lines containing Epstein-Barr virus (EBV) demonstrated marked differences in their permissiveness to infection by herpes simplex viruses type I (HSV-t) and type 2 (HSV-2). In such infected cells, virus replication varied from a typical cytopathic effect (c.p.e.) within I day after infection to a delayed cell death requiring as long as 7 weeks. In order to extend these studies further, particularly with respect to the variations in permissiveness of BL derived cell lines to herpesvirus replication, we infected the EBV producing clone, P3HR-I, of Jijoye (or P3J) cells (Hinuma & Grace, I967) and the non-producing Raji cells (Pulvertaft, I965) with six strains of HSV-I and two strains of HSV-2. We subsequently observed that persistent infections could be established with most virus strains in both cell types and the maintenance of these persistent infections required no external support. An extensive characterization of P3HR-I cells persistently infected with the MAL strain of HSV-~ suggested an atypical HSV-I replication. * Present address: Frederick Cancer Research Center, Frederick, Maryland. -~ Present address: Mayo Clinic, Rochester, Minnesota.

2 5 2 W. G. ROBEY AND OTHERS METHODS Cell cultures. P3HR-I and Raji cells were grown and maintained in RPMI ~64o medium supplemented with IO ~ heat-inactivated foetal calf serum (FCS), glutamine (2 ram), penicillin (o'o5 mg/ml) and streptomycin (o'o5 mg/ml) in a humidified 5 ~ carbon dioxide atmosphere at 37 C. Vero cells (American Type Culture Collection, CCL 8 t) were grown and maintained in the same medium supplemented with Hepes buffer (25 ram). Grand Island Biological Company (Grand Island, N.Y.) media and supplements were used throughout. P3HR-I cells were subcultured every I2 to t4 days and Raji cells every 6 to 8 days by diluting 2 Io viable cells (trypan blue exclusion) to 2o ml with medium and incubating as described. Cultures were routinely tested for mycoplasma by broth and agar methods (Microbiological Associates, Bethesda, Md.) and were found to be negative. Electron microscopic analyses for mycoplasma, as well as detection of mycoplasma DNA on caesium chloride gradients, were also negative. Virus. The MAL strain of HSV-I, cloned 3 times in Vero cells, was obtained from Dr Berge Hampar (National Cancer Institute, Bethesda, Md.). Two syncytial variants of the MAL strain were isolated in this laboratory by serial passage of this virus through Vero (Syn V) or through JLSV9 mouse cells producing Rauscher leukaemia virus (Syn R; Andrese et al. unpublished data). The KOS strain of HSV-I was obtained from Dr John J. Docherty (The Pennsylvania State University, University Park, Pa.). A fresh oral isolate of HSV (BF) was obtained from Dr Bernhard Fleckenstein (der Universitat Erlangen-Nurnberg, Erlangen, Germany) and determined to be HSV-I by neutralization kinetics. The Patton strain of HSV-T and 333 strain of HSV-z were obtained from Electronucleonics, Inc. (Silver Spring, Md.). The MS strain of HSV-z was obtained from Dr Alan Rabson (National Cancer Institute, Bethesda, Md.). Stocks of the above laboratory strains of herpesviruses, except the BF strain, were produced after multiple passages through Vero cells. A stock of the BF strain was obtained by two passages through Veto cells. Virus assay. Virus was added to freshly confluent Vero monolayers in a volume of o.a ml, incubated for 15 rain at 25 C followed by 45 min at 37 C and then re-fed with 4 ml of Vero maintenance medium (4 ~ FCS). The cultures were incubated 30 to 36 h at 37 C before staining with 0"5 ~ crystal violet in 7o ~o methanol. Virus titres are expressed as p.f.u.]ml. Herpesvirus infection. Two million viable cells from I2 to 14 day P3HR-~ or 6 to 8 day Raji cultures were collected by centrifuging at 80o g for 5 min and resuspended in a ml of growth medium containing the appropriate amount of virus for high and low multiplicities of infection (m.o.i.). Sham-infected control cells were resuspended in medium. The cultures were incubated for z h at 37 C in a shaker bath, transferred to 75 cm 2 plastic flasks (Falcon Plastics, Oxnard, Ca.), diluted to Io ml with medium and incubated at 37 C. Infected P3HR-I cultures were maintained through the initial infection and two routine passages of ~4 days each and infected Raji cultures were maintained through the initial infection and two routine passages of 7 days each. Samples of the above culture fluids and cells were diluted with complete medium, frozen at -70 C until titration, thawed, filtered through o-45 #m Swinnex filters (Millipore Corporation, Bedford, Ma.) and titred on Vero monolayers. Adsorption of HSV-I (MAL). Vero and P3HR-I cell suspensions were prepared and treated in a manner similar to that described by Docherty, Mantyjarvi & Rapp (I972) except diluted and centrifuged portions were filtered through o'45/zm Swinnex filters and stored at -70 C until titred on Vero cell monolayers. Titres are presented as unadsorbed virus in p.f.u./ml.

3 Persistent HSV infections in lymphoma cells 53 Infectious centre assay. Two million viable P3HR-I cells were infected for 2 h at 37 C in a shaker bath with HSV- I(MAL) at a m.o.i, of I-o and o.oi p.f.u./cell, washed 3 times with complete medium and adjusted to a concentration such that 0-2 ml contained either 5 Io2 (m.o.i. = I.o p.f.u./cell) or 5 x io ~ (m.o.i. = o-oi p.f.u./cell) viable cells. Replicate monolayers of Vero cells were infected at these multiplicities. After incubation for 2 h at 37 C, cells were removed by trypsinization, washed 3 times with complete medium and adjusted so o.2 ml contained either I x Io z or 5 x io ~ (m.o.i. = I.o p.f.u./cell) or I x Io ~ (m.o.i. = o'oi p.f.u./cell) viable cells. The appropriate number of infected P3HR-I or Vero cells in o-2 ml were plated on to Vero monolayers in 6o mm diam. Petri dishes (Falcon Plastics), incubated I5 min at 25 C, 45 min at 37 C and then very carefully overlayered with 4 ml of complete medium containing o'5 ~ (w/v) methyl cellulose. The plates were then incubated in a 5 ~ CO2 humidified atmosphere at 37 C. Immunofluorescence. Washed cell smears were prepared on slides, dried at 37 C, fixed in acetone for Io rain and treated for 45 min at 37 C with the appropriately diluted human serum that was either positive for EBV (virus capsid antigen and early antigen) and negative for HSV or positive for HSV and negative for EBV. Smears were washed with saline and reacted with goat anti-human IgG (heavy and light chain) conjugated with ftuorescein isothiocyanate (Hyland Laboratories, Costa Mesa, Ca.) and examined by ultraviolet microscopy. Isotopic labelling and purification of cellular and HSV DNA. Approx. 2 Io 8 Veto cells were infected with HSV-I(MAL) at a m.o.i, of Io and incubated at 37 C. Cells were labelled with io #Ci/ml of 3H-thymidine (6 Ci/mmol, New England Nuclear, Boston, Ma.) beginning 3 h post infection (p.i.) until 9o ~ c.p.e. (8 to Io h) was observed. The P3HR-I control and HSV-I infected cultures at a density of I Io 6 viable cells/ml were labelled with I/zCi/ml of 14C-thymidine (40 mci/mmol, New England Nuclear) for 72 h. The collected cell pellet from P3HR-I cells was suspended in Io vol. of phosphate-buffered saline (PBS), lysed with I ~o SDS and incubated for 8 h at 37 C with autodigested pronase (Calbiochem, San Diego, Ca.) at a final concentration of 2oo #g/ml. The DNA was deproteinized by two phenol extractions followed by two extractions with chloroform-isoamyl alcohol (96:4). The methods of virus purification, viral DNA isolation and isopycnic centrifugation of DNA were the same as those described by Simonds et al. (I975). Radioactivity was determined as trichloroacetic acid insoluble ct/min using a Packard model 338o liquid scintillation counter. DNA-DNA filter hybridization. The techniques of filter hybridization used (Gillespie & Spiegelman, 1965; Denhardt, I966; Warnaar & Cohen, I966)" contained the following modifications. Cell DNA was denatured with o-2 N-NaOH for 30 rain at room temperature. The salt concentration of the DNA solution was increased to 3 SSC (SSC = Ioo mm- NaC1, Io mm-sodium citrate) and the DNA was immobilized on to pre-soaked (3 x SSC) nitrocellulose membrane filters (Schleicher and Schuell, Keene, N.H.) and washed once with neutralizing solution (I.O M-tris-HC1, ph 7"5, 0.6 M-NaC1) and once with 3 x SSC. After drying for 8 h at 8o C, the filters were incubated for 5 h at 66 C in 5 ml of preincubation medium (Denhardt, 1966 ). 3H-HSV-I DNA (5 x io 5 ct/min//~g) was sheared by sonication (six I min intervals), heat denatured (3o min at IO5 C) and quickly chilled on ice. Three x IO ct/min of DNA in I.o ml of 3 x SSC was added to each filter. The hybridization reaction was terminated after incubation at 66 C for 22 to 24 h. The filters were washed three times with 3 x SSC on both sides and radioactivity was determined as described above.

4 54 W. G. ROBEY AND OTHERS 6-5 I I I 1! i 6-0 ~.. ~ i ~ 5"5-,s 45 "~, '/' /" N ~g/ 4.0 I I i I,, i I I (b) 7'0 o m ~mm\m/m 5.0 ~A A~A~A 2.0,\._/ 3.0 l.o i I i I I i I l Time p.i. (days) Fig. I. (a) Growth and (b) virus production kinetics of P3HR-~ cells infected with HSV-~(MAL) during the infection passage. (a) 0--0, viable control cells; (3..., trypan-blue-sensitive control cells; [], viable cells at m.o.i, of Io; []... ~, trypan-blue-sensitive cells at m.o.i. of Io;, viable cells at m.o.i, of o'ooi; A... A, trypan-blue-sensitive cells at m.o.i, of o-ooi. (b) [] [], virus titre (p.f.u./ml) at m.o.i, of lo; A A, virus titre at m.o.i, of o.ool RESULTS Cell growth and virus produetion kinetics of P3HR-I cells infected with HSV-I The kinetics of HSV-I(MAL) replication in P3HR-I cells was determined at two m.o.i. When the rate of P3HR-~ cell growth at the higher 0o) and lower (o.ooi) m.o.i, were compared to sham-infected controls, only the culture infected at the higher m.o.i, showed a growth lag between 5 and 9 days p.i. (Fig. I a) perhaps indicating slight HSV-induced c.p.e. However, by I I days after infection the three cultures were equivalent in cell growth. The virus production curve (Fig. I b) for both HSV-~ infected cultures showed a distinct 3-day lag before new virus was produced. This lag was markedly different from that observed following a similar HSV-I m.o.i, on Veto cells where maximum virus production occurred within t2 to 4 8 h after infection and resulted in total c.p.e. (data not shown). The time of maximum virus production preceded maximum cell growth by 2 days in the low m.o.i, and

5 Persistent HSV infections in lymphoma cells t I I t i.6 E.E >- f/3 lo s \ 10 4 I I I I Time p.i. (rain) Fig. 2. Adsorption of HSV-x(MAL) by P3HR-I (0 O) and Vero (O... ) cells at m.o.i, of (a) I.o and (b) 0'I. Titres represent HSV-I remaining in solution. by 6 days in the high m.o.i, which indicated that virus production appeared to be related to m.o.i, and, more importantly, to the rate of cell growth. Since a typical HSV-I c.p.e, was not observed in these cultures, the HSV-I infected cells were passaged and examined for continued HSV-I production. After the first passage, no differences were detected between the growth rates and number of viable or trypan-bluesensitive cells in the control and the two HSV infected P3HR-I cultures; further, infected cultures were indistinguishable with respect to the rate and level of HSV production (data not shown) and, in effect, had become persistently infected. P3HR-I cells infected with HSV-I(MAL) in an analogous manner have now undergone 44 routine passages without losing the ability to produce HSV-I. I 120 Kinetics of HSV-I adsorption to P3HR-I and Vero cells To determine whether the establishment of a persistent infection in P3HR-I cells and not Vero cells was related to differing virus adsorption characteristics, the kinetics of HSV-I adsorption were compared using both cell types (Fig. 2). At the m.o.i, tested, Vero cells adsorbed HSV-I at a rate greater than P3HR-I cells during the first I5 rain but the extent of adsorption for both cell types was the same after 2 h. Both cell types adsorbed approx. 7o ~ of the input p.f.u. More than 9o % of the input virus was recovered in the virus controls after z h. Infectious centre assay of infected P3HR-I and infected Vero cells Since the extent of adsorption of HSV-I to P3HR-~ and Vero cells was shown to be equivalent, an infectious centre assay was used to determine the number of each cell type that became productively infected (Table I). At both m.o.i., infected Vero cells produced plaques within 48 h and total c.p.e, by 72 h whereas infected P3HR-I cells did not produce any plaques prior to 7z h and then only < 1% of the cells plated were productively infected. Immunofluorescence of HSV- I superinf ected P3HR- t cells To determine the extent of HSV antigenic expression as a function of passage level, a P3HR-I culture persistently infected with HSV-I(MAL) was examined by indirect

6 56 W. G. ROBEY AND OTHERS Table I. Infectious centre assay of HSV-I(MAL) infected Vero and P3HR-I cells HSV-I infected cells Plaques* plated on to Vero HSV c - - ~ ' monolayers m.o.i. 48 h 72 h 90 h I IO ~ Vero I'O 62 c.p.e.t -- 5 x io s Vero I-o 236 c.p.e x Io 2 P3HR-I I'O o 3 45 I x IO 4 Vero o'oi I93 c.p.e X IO n` P3HR.-I o'oi o 2 35 * Average of 4 plates. t Total c.p.e. J; Microscopically visible secondary plaques detected. Table 2. Immunofluorescence of HSV-I(MAL) in infected and control P3HR-I cells Immunofluorescence (% of positive cells) y, P3HR-I culture EBV HSV Control (all passages) io o Infected passage 3 IO 3o Infected passage 6 I o 8o Infected passage 24 IO 80 immunoftuorescence (Table 2). The percentage of cells showing HSV-specific fluorescence increased from 3o % at passage 3 to 8o % by passage 6. After 24 routine passages, no further changes were observed in the percentage of cells positive for HSV. In cells from the same culture, HSV infection had no effect on EBV expression since Io % of the cells were constantly positive for EBV antigens. Similar results were obtained by Zajac, Henle & Henle (1969) and Henle et al. (I969). The effect of multiple medium changes on virus production in HSV-I persistently infected P3HR-I cells" The effect of multiple medium changes on a HSV-I(MAL) persistently infected P3HR-I culture (passage level 6) was studied to determine whether c.p.e, could be induced by continuous cell growth (Henle et al. I969) or ifa cyclical relationship existed between cell growth and virus production (Hampar & Copeland, I965). Cell culture fluids were harvested, replaced with fresh medium at 3- to 4-day intervals and cell counts and virus titres were determined (Fig. 3)- Under these conditions, the HSV-I infected ceils continued to grow and produced an average of 5 los p.f.u./ml. At times when high HSV titres were obtained (day 16, I9 and 4I) no significant variations in the numbers of viable and trypan-blue-sensitive cells were observed. Furthermore, throughout the experiment, the percentage of cells showing HSV and EBV-specific fluorescence remained constant at 80 % and IO % respectively. Thus, after 48 days of continuous cell growth, no evidence for increased cell death was obtained nor did the production of HSV by these cells appear to be cyclical in a manner analogous to the Chinese hamster system studied by Hampar & Copeland (I965). Characterization of virus DNA produced by HSF-I persistently infected P3HR-I cells The DNA of HSV-I produced by persistently infected P3HR-I cells was compared to DNA of the parental HSV-I(MAL) produced in Vero cells to determine whether the two DNA species had related densities in caesium chloride. Extracellular virions were purified

7 Persistent HSV infections in lymphoma cells I,, I I,, I/,~,, ~lr~ 6-5 '~O"" ~...o/ ~ 6-0 / t / /,o, o ~j/o 5.0 # o" I/ 4.5 l I I i I i I i i i i i 7.0 ii-i / ~o- ". an- m/ \=~,,, nn-nn'nn/ \.=~'>- 6.0 (b) l I i i i i i i I I '~ n--i t Time (days) Fig. 3. Effect of multiple medium changes on (a) growth rate and (b) virus production of P3HR-T cells persistently infected with HSV-I(MAL):, viable cells; (3... O, trypan-bluesensitive cells; n, virus titre of cell-free medium (p.f.u./ml). and virus DNA from HSV-T persistently infected and control P3HR-I cultures and HSV-I(MAL) infected Vero cultures were prepared as described (Simonds et al I975). Virions from the HSV-r infected P3HR-I culture contained one major species of DNA which had a buoyant density of I"727 g/rnl which co-banded with DNA isolated from HSV-T(MAL) infected Vero cells and one minor species which had a density of ~'7oo g/ml (Fig. 4)- Pre-treatment of virus with deoxyribonuclease before extraction eliminated the lighter density cellular DNA species while 90 % of the g/ml peak was resistant to digestion (data not shown). EBV DNA in the density region of g/ml (Schulte- Holthausen & zur Hausen, ~97o) was not detected in either the control or HSV-I infected P3HR-~ cultures as the labelling conditions used here were not optimal for EBV production (Hinuma et al. ~967). Thus, HSV-I persistently infected P3HR-I cells produced a virus DNA physically indistinguishable from the parental HSV-I (MAL). Detection of viral DNA in HSV-I persistently infected P3HR-T cells The relative amount of HSV-T DNA synthesized in persistently infected P3HR-t cells was determined by DNA-DNA filter hybridization (Fig. 5). Cellular DNA from HSV-I infected cells, fractionated by isopycnic banding in caesium chloride, contained a minor peak (fractions 36 to 47) of high guanine + cytosine virus DNA which represented approx. 4 % of the total 14C-thymidine incorporated and a major peak of cell DNA (fractions 6o to 80) of high adenine + thymidine content (Fig. 5 a). A more accurate estimate of the amount of high-density DNA labelled, after adjusting for the differential uptake of x~c-thymidine into molecules with different base content (e.g. minor peak = 68 % G + C; cellular peak =

8 58 w. G. ROBEY AND OTHERS 18- I I I l 1 HSV-I i X e-, 12.=_? O 3o 4O 50 Fraction number Fig. 4. Caesium chloride gradient characterization of DNA from extracellular virions purified from HSV-[ persistently infected (Q 0) and control ( (3) P3HR-I cultures. The arrow designates the fraction at which the peak of radioactivity of 8H-labelled HSV-I virion DNA from Vero cells was detected when mixed with 14C-labelled HS -I virion DNA isolated from HSV-I persistently infected P3HR- I cells. 39 ~ G+C), would be 8 ~. When representative gradient fractions were hybridized to 3H-thymidine labelled HSV-I DNA, significant amounts of ~H label were detected in the density region where HSV-I DNA banded and in a region of the gradient more dense than the median density of cell DNA. However, the relative amount of SH bound with respect to the amount of 14C-DNA immobilized on the filter was greater in the high density HSV-I virus DNA region (Fig. 5a). Fractionated control P3HR-I cell DNA bound only small amounts of HSV-I DNA and then only in the low-density cell DNA region (Fig. 5b). Thus, both density species of DNA present in the persistently infected P3HR-I cells contained sequences homologous to virus DNA; and therefore a minimum of 8 ~ of the label incorporated in 72 h was taken to be virus-specific. The response of P3HR-I and Raft cells to infection with two types and several strains of HSV Six strains of HSV-I and two strains of HSV-2 were used to infect P3HR-I and Raji cells at a high and a low m.o.i, to see if other persistent infections could be established. These m.o.i, produced > 80 ~ c.p.e, in Vero cells with all virus strains tested within I2 to 48 h p.i. Differences in response to HSV infection were observed with the P3HR-I and Raji cells as demonstrated by cell growth patterns and virus production (Table 3). Only infected cultures with normal growth rates and those that produced > I x io 2 p.f.u./ml by the end of passage 2 were defined as persistently infected. Some cultures exhibited decreased relative growth

9 Persistent HSV infections in lymphoma cells 59 I I I I I I 3 (a)?, " ", / l \ 1 7 X X e-, 4 (b) 4 ~Z S'/ 0 _ ~ ~o..,...~ t~," I Fraction number Fig. 5. Filter hybridization of HSV-I DNA produced in Vero cells (O... O) with cellular DNA (Q Q) isolated from HSV-I (a) persistently infected and (b) control P3HR-I ceils after caesium chloride density-gradient purification. rates during the infection and/or the first passage (Table 3). The MS strain of HSV-2 at the high m.o.i, was lethal to P3HR-I cells and the Syn V strain of HSV-I was lethal to Raji ceils at both m.o.i. Thus, the surviving cultures were either persistently infected (2o of 3 2 cultures tested) or resistant to infection since virus was not detected after two routine passages. DISCUSSION We observed that persistent infections were established when P3HR-~ cells were infected with the MAL strain of HSV-I instead of the typical herpesvirus replication expected. The characteristics of this particular infection indicated that HSV-I(MAL) adsorbed equally well to P3HR-I cells and to lethally permissive Vero cells but less than 1% of the P3HR-t cells became productively infected compared to the majority of Veto cells. Also, the parent HSV-I(MAL) replicated more slowly in P3HR-I cells than in Veto cells, but the progeny

10 - (o) 60 W.G. ROBEY AND OTHERS Table 3. HSV infection of P3HR-I and Raft cells Growth rate* t Virus Strain m.o.i. Infection Passage i Passage z Log10 titre'~ - HSV-I MAL io.o -- (-) o (o) o (o) 5"8 (6'4) o.ooi o (o) o (o) o (o) 5'9 (5"0) KOS io.o o (-) o (o) o (o) 4"4 (6"3) o.ooi o (o) o (o) o (o) 4"7 (6-7) Patton 2o'o -- (o) o (o) o (o) 5"4 (4"3) 0"o0I o (o) o (o) < 2"0 (5'7) Syn R 0'5 o (o) o (-) o (o) 5"9 (5"5) O'OOI 0 (13) 0 (--) 0 (0) <2"0 (5"5) BF o-i6 o (-) o (-) o (o) 5'5 (6-0) O'OOI O (0) 0 (13) 0 (0) <2"0 (<2"0) SynV IO-O -- (-) o (-) 0 ( ) 6-o (<2.o) O'OOI 0(O) O(--) O(><) <2"0(<2'0) HSV t'3 -- (-) o (o) o (o) <2.o (6'5) 0"002 o (o) o (o) o (o) < 2"o (3"3) MS IO'O o (-) o (-) x(o) <2"o (4'3) o(o) o(o) o(o) <2.0(<2.0) * Growth rates relative to sham-infected controls; P3HR-I cell data presented first and Raji data in parentheses. (o), control rate; (-), decreased rate, > 5o ~ below control; ( ), death. t Virus titre at the end of passage z. At this time, P3HR.-I cell counts ranged from 2"4 IO 6 to 4"2 ~o viable cells/ml and Raji cell counts ranged from z'i IO 6 to 3'5 x [o viable cells/ml. virus from each cell type replicated at the same rate in Vero cells. The majority of persistently infected P3HR-I cells were positive for immunofluorescent HSV antigens by the sixth passage and have remained at equivalent levels through 24 passages. The cells positive for EBV-specific antigens by immunofluorescence remained constant in agreement with other similar observations (Henle et al. ~969; Zajac et al. 1969, Floyd et al. I97 0. To date the HSV-I persistently infected P3HR-I cells have undergone 44 routine passages without losing the ability to produce HSV and as such this infection did not appear to be self-limiting (Blackmore & Morgan, I967). We were unable to significantly accelerate virus production or cell death with multiple medium changes nor did we detect cyclical fluctuations in cell growth and virus production (Hampar, I966). The analyses of radioisotopical!y labelled virus and cellular DNA indicated that the HSV-I persistently infected culture produced a virus DNA physically and genetically related to the parental HSV-t(MAL). A minimum of 8 ~ of the total label incorporated by infected cells in 72 h appeared to be virus DNA. The relative percentage was actually somewhat greater as the cellular DNA fraction also contained HSV DNA sequences as determined by hybridization. Furthermore, when the cellular DNA fraction was recycled on caesium chloride, additional high density virus DNA was detected (data not shown). Analogous observations have been described for EBV DNA associated with Raji cell DNA (Tanaka & Nonoyama, ~974; Adams & Lindahl, I975). For the P3HR-I culture persistently infected with HSV-L we conclude that only about I ~ of the cells are actively producing a virus that is indistinguishable from HSV-I by nucleic acid density and hybridization, although a majority of cells are positive for HSV-~ immunofluorescent antigens. Together with the isotopic DNA studies, these data suggest that the P3HR-I cell exerts control over the replication of HSV-~ which may be analogous

11 Persistent HSV infections in lymphoma cells to the control exerted over the expression of EBV (Hinuma & Grace, 1967; Maurer, Imamura & Wilbert, I97o). The establishment of persistent HSV infections in Burkitt lymphoma derived cells is in contrast to other reports (Henle et al. I969; Floyd et al. 197I). However, it should be noted that although the cells and viruses studied herein were similar to, they were not identical to, those studied by the above investigators. Under the conditions reported here, no external support was required to maintain these infections. To our knowledge, only the HSV-1 persistently infected Chinese hamster cells described and characterized by Hampar & Copeland (I965), Hampar (I966) and Hampar & Burroughs (1969) and the Earle's L cells described by Nii (1969)fit this classification. Persistent HSV-1 infections in cell culture have usually required external support such as the presence of antiserum to HSV (Wheeler & Canby, ~959; Fernandez, 196o; Wheeler, I96o; Hinze & Walker, 196I; Szanto, 1963) decreased (Coleman & Jawetz, 1962 ) or increased (Floyd et al incubation temperatures or co-infection with polyoma virus (Glasgow & Habel, I963). The production of interferon did not appear to be a factor in establishing the persistent infections reported here since virus-free concentrates of tissue culture medium from infected and control P3HR-I cells did not reduce either the rate of appearance or number of vesicular stomatitis virus plaques detected on human foreskin cells (Dr Maneth Gravell, personal communication). Moreover, Henle et al. (1969) were unable to detect interferon in Jijoye and Raji cells following infection with the H strain of HSV-I. Furthermore, since we have been unable to detect mycoplasmas in our cultures, we assumed that this contamination was not a contributing factor in the establishment of persistent infections. We are not suggesting that the observations reported herein for P3HR-I cells persistently infected with HSV-I(MAL) necessarily apply to the other cells and viruses studied (Table 3) since P3HR-I and Raji cells responded to HSV infection in various ways: 0) the cultures were resistant to infection (9 of 32 cultures tested); (2) the cultures were lethally permissive to infection (3 of 32 cultures tested); or the cultures became persistently infected (2o of 32 cultures tested). Whether the resistance to infection is at the adsorption level or due to a true cellular block is under investigation. The establishment of persistent infections was related to the initial virus m.o.i., as some cultures infected at the higher m.o.i, became persistently infected but the same cultures at the lower m.o.i, did not. However, in cultures where persistent infections were established [similar to HSV-I(MAL) in P3HR-I cells], the virus titres at the end of the second passage for the individual cell type were generally equivalent and independent of the initial m.o.i. In all persistent infections established, maximum virus titres were obtained slightly before or at the same time maximum cell population levels were obtained indicating virus replication in these cultures may be dependent on cell growth. Taken together, the variety of responses observed to HSV infection by the two cell lines suggests the presence of complicated interactions involving cell type, virus type and virus strain. We wish to thank Dr Robert Manaker, Dr Richard Ascione, Dr Robert Goldberg, Dr Angelo Andresse and Dr Berge Hampar for their helpful discussions. We thank Dr Maneth Gravell for performing the interferon analyses and the assistance of Mr Tom Orr in the immunofluorescence studies is greatly appreciated. 6I 5 VIR 32

12 62 W. G. ROBEY AND OTHERS REFERENCES ADAMS, A. & LINDAHL, T. (I975). Epstein-Barr virus genomes with properties of circular DNA molecules in carrier cells. Proceedings of the National Academy of Sciences of the United States of America 72, I477- I48I. RLACKMORE, a. V. & MORGAN, H. g. (I967). Self-limiting infection with herpes simplex virus in cell culture. Acta Virologiea (Praha) H, COLEUAN, V. & JAWETZ, E. (I96I). A persistent herpes simplex infection in antibody-free cell culture. Virology I3, DENnARDT, O. T. (I 966). A membrane-filter technique for the detection of complementary DNA. Biochemical and Biophysical Research Communications 23, 64 t-646. DOCHERTY, J. J., MANTYJARVI, R. A. & RAPP, F. 0972). Mechanism of the restricted growth of herpes simplex virus type 2 in a hamster cell line. JournalofGeneral Virology x6, FERNANDEZ, C. G. (I960). Persistence of herpes simplex virus in HeLa cells. Nature, London I85, 268. FLOYD~ R., GLASSER, R., VONKA, V. & BENYESit-MELNICK, M. (1971)- Studies on the growth of herpes simplex virus in lymphoblastoid cells. Acta Virologica (Praha) I5, I33-I42. GILL~SPIE, D. & SPIEGELMAN, S. (1965)- A quantitative assay for DNA-RNA hybrids with DNA immobilized on a membrane. Journal of Molecular Biology I2, ~LASGOW, r. A. & HABEL, K. 0963). Role of polyoma virus and interferon in a herpes simplex virus infection in vitro. Virology x9, HAMPAR, R. (I966). Persistent cyclic herpes simplex virus infections in vitro. III. Asynchrony in the progression of infection and cell regrowth. Journal of Bacteriology 9t, 1965-I97o. HAMPAR, B. & BURROUGHS, M. a. K. 0969)- Mechanism of persistent herpes simplex virus infection in vitro. Journal of the National Cancer Institute 43, HAMPAR, B. & COPELAND, M. g. 0965). Persistent herpes simplex virus infection in vitro with cycles of cell destruction and regrowth. Journal of Bacteriology 9 o, 2o I-IENLE, W., HENLE, G. & ZUR HAUSEN, H. (I969). Effect of herpes simplex virus on cultured Burkitt tumor cells and its failure to influence the Epstein-Barr virus carrier state. Cancer Research 29, IJINUMA, V. & GRACE, J. T. JOY. (1967). Cloning of immunoglobulin-producing human leukemic and lymphoma cells in long-term cultures. Proceedings of the Society for Experimental Biology and Medicine I24, IO7-III. HINUMA, Y., KONN, M., VAMAGUCHI, J. & GRACE, J. W. JUN. 0967)- Replication of herpes-type virus in a Burkitt lymphoma cell line. Journal of Virology i, I2o3-~2e6. HINZE, H. & WALKER, D. L Variation of herpes simplex virus in persistently infected tissue cultures. Journal of Bacteriology 82, NAORER, ~. A., IMAMURA, L & WlLBERT, S.M. 0970)- Incidence of Epstein-Barr virus-containing cells in primary and secondary clones of several Burkitt lymphoma cell lines. Cancer Research 3o, 287o NIl, S. (I969) Persistent infection with herpes simplex virus in vitro. I. Establishment and characteristics of persistent herpes simplex virus infection in Earle's L cells. Biken's Journal I2, PULVERTAFT, g. S. V. 0965). A study of malignant tumours in Nigeria by short-term tissue culture. Journal of Clinical Pathology i8, 26~-273. SCHULTE-ROLTHAUSEN, n. & ZUR HAUSEN, H. 097O). Partial purification of the Epstein-Barr virus and some properties of its DNA. Virology 4 o, SIMONDS, J. A., ROBEY, W. G., GRAHAM, B. J., OIE, H. & VANDE WOL1DE, G. F. 0975). Purification of herpesvirus saimiri and properties of the viral DNA. Archives of Virology 49, SZANTO, s. 0963). Course of persistent inflections of HeLa and Detroit 6 cells with herpes simplex virus. Acta Virologica (Praha) I, TANAKA, A. & NONOYAMA, M. 0974). Latent DNA of Epstein-Barr virus: separation from high-molecularweight cell DNA in a neutral glycerol gradient. Proceedings of the National Academy of Sciences of the United States of America 7I, t. WAR~AAR, S. O. & COnE~, J. A. 0966). A quantitative assay for DNA-DNA hybrids using membrane filters. Biochemical and Biophysical Research Communications 24, WHEELER, C.E. 0960). Further studies on the effect of neutralizing antibody upon the course of herpes simplex infections in tissue culture. Journal of Immunology 84, WHEELER, C. E. & eanby, C. M. 0959)- Influence of specific antibody on herpes simplex infections in tissue culture. American Medical Association Archives of Dermatology 79, ZAJAC, B. A., I-IENLE, W. & HENLE, G. (1969). The effect of mumps virus on the resistance of Burkitt lymphoma cell lines to various viruses. Cancer Research 29, ~ 467-~ 475. (Received 27 October I975)