Cytomegalovirus Latency in Cultured Human Cells

Transcription

1 J. gen. Virol. (I973), 18, I43-I5I Printed in Great Britain I43 Cytomegalovirus Latency in Cultured Human Cells By EVA GONCZ~)L AND L. V/~CZI Department of Microbiology, Medical School of Debrecen, Debrecen, Hungary (Accepted 29 September I972 ) SUMMARY Infectious virus could not be recovered by disruption of human embryonic fibroblast cultures inoculated with cytomegalovirus and treated with cytosinearabinoside (ara-c), either during treatment or 4 to 5 days after removing ara-c. However, when intact cells were used for infection, at least I in 7o ceils was found harbouring virus. Examination by immunofluorescence revealed that 4 to 5 % of the cells contained virus-specific antigens in the form of small granules diffusely distributed in the nuclei. The presence of infectious virus could be detected even in disrupted cells 4 to 5 days after removing ara-c; intracellular antigens of every kind were also found to develop. Experiments showed that the cultures continued to harbour latent infection indefinitely in the presence of ara-c. Cultures harbouring latent virus were susceptible to superinfection with cytomegalovirus. INTRODUCTION Cytomegalovirus (CMV) infection is very common. Most of it is subclinical, but infection is followed by persistence of the virus in the body in a latent form for the rest of life. About IO % of apparently healthy blood donors carry the virus in this latent form in their white blood cells, without detectable excretion in the throat and urine, and can infect susceptible patients by blood transfusion (Diosi, Moldovfin & Tomescu, 1969; Perham et al. I971)- Immuno-suppressive therapy can also reactivate latent infection, giving rise to virus excretion during life, and histological evidence of disseminated disease at post mortem (Craighead, Hanshaw & Carpenter, I967; Munk & Runnebaum, t97o). We have developed a system in which cells, or a certain part of them, harboured latent cytomegalovirus, thus making it possible for this condition to be studied in vitro. METHODS Virus. CMV, strain RAWLES, was obtained from Dr Stern, St George Hospital Virus Laboratory, London. The strain has been maintained in our laboratory since I964 on human embryonic fibroblast cells in which titrated virus amounted to Io 5.5 to Io 6.5 TCD 5o[ o-i mi. Tissue cultures. Primary or secondary fibroblast cell cultures were prepared as described previously (G6der et al. i963). The growth and maintenance medium was Parker's no. I99 containing to % bovine serum. Experimental media contained cytosine-arabinoside (ara-c) at a concentration of to #g/ml. Establishment of virus latency. Human embryonic fibroblast cultures in tubes were infected

2 144 E. GONCZOL AND L. V,~CZI with I TDC5o of virus/cell. After I h adsorption, the inoculum was removed, cultures were washed 3 times with PBS and experimental medium containing ara-c was added to one part of the cultures. The other part of the cultures was overlaid with culture medium without ara-c. At intervals indicated under 'Results', ara-c was removed from the experimental cultures, which were then washed 3 times and ara-c-free culture medium added. The following were studied: development of cytopathic changes, production of infectious virus, number of cells harbouring virus, intraceuular antigens and, beginning with the 8th day, the susceptibility of the cultures to superinfection with CMV. Determination of infectious virus production. Three replicate samples from cultures were washed 3 times with PBS, removed from the glass with trypsin, homogenized by ultrasonic vibration (MSE ultrasonic apparatus; I'5 A for I min) in culture medium of identical volume as the original, centrifuged and the supernatant fluid titrated on human embryonic fibroblast cells. The virus titres were determined according to Reed & Muench (1938). Ascertaining the number of cells harbouring virus. Three replicate samples from cultures were washed, then removed from the glass with trypsin, resuspended in culture medium of identical volume as the original, counted, and tenfold dilutions made from the suspension. Human embryonic fibroblast monolayers were inoculated and the proportion of cells containing virus ascertained. Determination of intracellular antigens. An indirect immunofluorescent technique was used as described earlier (V~iczi & GOncz61, I972). Serum applied was obtained from a patient suffering from primary CMV infection. The anti-cmv titre of the serum amounted to 1 : z56 as measured by immunofluorescence. Susceptibility of the cultures to superinfection with CMV. On the 8th day the medium was removed from the cultures pretreated with CMV and ara-c; the cultures were washed 3 times with PBS and infected with I TCD5o of cytomegalovirus/cell; this was done immediately after removing ara-c and on the 9th, Ioth and I lth day, respectively. For purposes of control: (a) cultures were pretreated with ara-c only for 8 days, and (b) untreated human embryonic fibroblast cultures were inoculated. After I h adsorption inoculum was removed, cultures were washed 3 times with PBS and overlaid with ara-c-free culture medium. Virus yield was determined 1 h after infection and then daily until the I2th day. RESULTS Development of virus yiem and cytopathic changes after treatment of the cells with ara-c Human embryonic fibroblast cultures were infected with I TDC5o of virus/cell and medium containing ara-c was added. On the 8th day after infection the media were removed from part of the cultures, which were washed 3 times with PBS and overlaidwith medium free ofara-c. Production of infectious virus in these cultures, and in those maintained in media containing ara-c during the whole experiment, as well as that in the control cultures without the chemical compound, was determined I h after infection and at daily intervals from I to 2o days thereafter. The virus yield of the cultures was I01'75 TDC50/o.I ml at I h after infection and in the case of control cultures IO 2'25 TDC5o/o.I ml on the Ist day, culminating to 1@.5 TDC5o/ 0.I ml on the 4th day (Fig. I). Conversely, virus could not be detected, even on the I st day, from infected and ara-ctreated cells. When ara-c was removed on the 8th day after infection, the presence of

3 100 ~w 8o ~ 60 e-~ Q ~ N 40 ~ 20 Cytomegalovirus latency in cultured human cells "7. 10 a o ~ L~ ~ 10~ -6 ~ 10_~ > I0 ix/a / ix / ix / o---o----o----o I I I I Time after infection (days) I Ara-C 1 Latent phase Fig, 1. Course of the disappearance and reappearance of CMV in the presence and following removal of ara-c. Replicate cultures of human embryonic fibroblast cells were infected with CMV and fed with medium containing ara-c. On the 8th day after infection, ara-c was removed from a part of them, the other part, however, was overlaid with ara-c containing medium during the whole experiment. All cultures were examined daily for the presence of virus c.p.e. Replicate samples were harvested daily and assayed for the presence of infectious virus. Infectivity: Q---Q, control cultures without ara-c; O--Q, cultures treated with ara-c; ±--±, cultures after removal of ara-c. infectious virus could be ascertained 5 days after removal of the drug; this was to 2s TDC 5o[ o. I ml on the ~ 6th day. The infectious virus was paralleled by the development of cytopathic changes in the cultures (Fig. O. The examination of virus content of the cell suspensions obtained from the cultures Cultures were trypsinized and cells counted. From half of the cell suspension tenfold dilutions were made. The remaining half was disrupted and similarly diluted. Human embryonic fibroblast cultures were inoculated with the dilutions and examined for cytopathic changes. The presence of infective virus could be detected in the cultures at every stage of the experiment when intact cells were used for inoculation (Table I). When infection was caused by diluted cells, at least I from among 5o, 6o and 7o cells proved to contain virus on the 1st, 2nd, 5th and 8th day. On the other hand, after the cells had been disrupted no

4 146 E. GONCZ(SL AND L, VACZI Table I. Comparative table for virus content of disrupted and &tact cells following treatment with ara-c* Days after Number of Virus content of Virus content of infection cells disrupted cells intact cells o o ooo t oo OO o * Ara-C was added at the time of infection. t Ara-C was removed on the 8th day after infection. infectious virus could be detected even when cell suspensions at higher concentrations were used. Despite the fact that ara-c had been removed on the 8th day, the number of cells containing virus failed to rise on the x3th, ~6th and I8th day. However, the presence of infectious virus could be ascertained in disrupted cells; on the I6th day the virus-containing dilution of disrupted cells corresponded to that of the cell suspension (Table 0. The examination of virus-induced antigens in ara-c-treated cells In the presence of ara-c, cells showed specific fluorescence in the form of very small granules diffusely distributed in the nucleus (Fig. 2 a). The percentage of cells containing antigen varied according to the time elapsed after infection. On the I st day after infection ~o to 15 ~, and from the and day 4o to 5o ~ of the cells contained virus-specific antigens. When ara-c was removed on the 8th day after infection, antigens of different types were found to develop successively in 4o to 5o ~ of the ceils, culminating on the ~ 6th day. The antigens detected were: intranuclear, large granules, intranuclear inclusion-like bodies, perinuclear, diffuse cytoplasmic and intracytoplasmic inclusion-like bodies (Fig. 2 b).

5 Cytomegalovirus latency in cultured human cells 147 Fig. a. Intracellular antigens: (a) in the presence, or 4 to 5 days after, removing ara-c; (b) on the 8th day after removing ara-c. The susceptibility of cultures harbouring latent virus to superinfection with CMV On the 8th day after infection ara-c was removed and one part of the cultures was immediately inoculated with CMV, while the other parts were infected on the 9th, Ioth and I Ith day. For control purposes cultures that had been maintained in medium containing ara-c from the start of the experiment and untreated human embryonic fibroblast cultures were infected. Virus yield of disrupted cells was determined I, 24, 48, 72, 96 h later. The virus yield of cultures harbouring latent virus, after having been superinfected with CMV at any time after removing ara-c, corresponded to that of cultures pretreated with ara-c only (Table 2). After untreated human embryonic fibroblast cultures had been infected with CMV, virus yield exceeded that of the experimental cultures (Table 2). The time allowed for latent harbouring of the virus In order to ascertain the time limit beyond which viruses cannot be retrieved from the cells, ara-c was removed on the I2th, I8th and 22nd day after infection. Cells were washed 3 times with PBS and overlaid with ara-c-free medium. The amount of infectious virus from the disrupted cells was determined every day. Infective virus could always be determined on the 4th or 5th day after removal of ara-c. The maximum virus yield amounted to IO2'STDC5o[o.I ml in every case (Fig. 3).

6 O0 Table 2. Susceptibility of cells harbouring latent virus to superinfection with I TDC 5 o of CMV[cell f Time of superinfection (days after beginning of expt) : I Virus yield (TDC 5o/o'1 ml) of cultures infected for (h) F Cell type i z4 48 7z 96 I z i i z 4 Latent virus harbouring I01" IO 3"73 I "5 10 1" I01' ' " " cells Ara-C-treated cells IO 1' I " I0 3'5 IO 4"25 IO 1" IO 3' "5 10 1"75 Untreated cells 101"75 IO 2'25 IO 3"75 I0 5"25 IO 5.5 IO 1"5 IO 2"25 I0 4' '75 IO 2.5 IO 3.75 IO 1' N : >.< >, N

7 Cytomegalovirus latency in cultured human cells I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I ~ tr3 A--A--& /'.--A--A /X--/X A/ A/ LX / : /,/ J o/ o/ > 0 C-., 0 0 C, t i t i l I I I I i C.,~ O Time after infection (days) Fig. 3. Course of reappearance of CMV following removal of ara-c at different intervals. Replicate cultures of human embryonic fibroblast cells were infected with CMV and fed with medium containing ara-c. On the I2th, I8th and 22nd day after infection ara-c was removed and cultures were refed with ara-c-free medium. Cultures were harvested daily and assayed for the presence of infectious virus. The arrows indicate the removal of ara-c. Infectivity: O--O, cultures treated with ara-c, ~--A, cultures after removal of ara-c. DISCUSSION In our previous experiments, ara-c at a concentration of IO #g/ml inhibited the development of cytopathic changes as well as the production of infectious virus, inasmuch as the tissue cultures were infected with an inoculum not exceeding I TDC5o of cytomegalovirus/ cell (V~iczi & GOncz61, 1972). A certain proportion of the cells were, however, found to harbour virus in a noninfectious form. When applying intact cells for infection at least I cell from among 5o, 60 or 7o cells was able to induce instantaneous virus synthesis. Virus-specific antigens in the form of diffusely distributed small granules in the nucleus could be detected in 4o to 50 % of the cells by means of the imunofluorescent technique. The fact that fewer cells harbouring virus were detectable when intact cells were used for inoculation, than when applying the immunofluorescent technique, can be attributed to the circumstance that the plating efficiency of human embryonic fibroblast cells on human embryonic fibroblast monolayers was low and only a limited number of cells was able to transfer virus. On the 4th to 5th day after removal of ara-c, cytopathic changes appeared in each of the cultures; this spread to 50 to 6o ~ of the cells on the 8th to 9th day after elimination of ara-c. Parallel with the development of c.p.e, the intracellular antigens in different form and localization were detectable and the presence of infectious virus could be observed in disrupted cells. The virus yield, however, failed to attain that of the untreated controls, which fact may be attributed to ara-c having a certain toxic effect on the host ceils. Gelderman et al. (I968) and Seto & Carver (t969) found that cells inoculated with CMV to be insusceptible to superinfection with Toxoplasma gondii and Newcastle disease virus (NDV) on the 2nd to 5th day after infection with CMV. The two latter authors described a change in the susceptibility of the cells inoculated with CMV to superinfection with NDV: according to the amounts of CMV inoculum employed, insusceptibility set in at various periods. Our experiments showed that, when using ~ TCD 5o/cell of CMV to establish latency, no insusceptibility to superinfection with CMV could be ascertained up to the 4th day after

8 I50 E. (3ONCZOL AND L. VACZI removing ara-c. Beyond this we did not examine the possible development of insusceptibility, seeing that even in cultures not exposed to superinfection the presence of infective CMV was detectable. It would therefore seem that insusceptibility of CMV-infected cells to NDV has a specific character. This has been confirmed by these authors, who were also unable to reproduce this phenomenon by means of Sindbis virus. The cells and virus employed by us would warrant the assumption that the cells harbouring virus were resistant to superinfection, while the remainder were not. It was possible, however, that cells harbouring latent virus were also susceptible to superinfection. Both cases were observed when other cell-virus systems were applied (Jensen & Koprowski, 1969; Sauer & Hahn, 197o). The reason for the 4 to 5 days delay in the reappearance of infectious CMV and the formation of new virus-induced intracellular antigens after removing ara-c has not been accounted for. It is possible that this delay is required for the repair of macromolecules of the host cells, disturbed by ara-c, and consequently for the synthesis of virus DNA to take place. This is contradicted by the experiments in which superinfecting virus is able to replicate both in cultures pretreated with ara-c and in latent virus-harbouring cultures. It became evident from our experiments that cells continue to harbour viruses in the course of the whole experiment if ara-c is not eliminated from the cultures. It thus would seem that the duration of the latent phase is unlimited in the presence of ara-c. In the case of herpes simplex, cell systems were developed in vitro, where no productive virus cycle was enacted in cells otherwise susceptible to the virus. Virus synthesis was blocked at some stage of replication by either adding immune-serum to the culture medium (Hinze & Walker, i96i; Szanto, 1963) or by inactivating the virus with u.v. irridation (Duff & Rapp, i97i ) or else by treating the infected cells with ara-c (O'Neill, Goldberg & Rapp, 1972). The CMV-system we used was exactly comparable to the HSV-system in which ara-c was used to develop latency. This similarity suggests that in the phenomenon of the latency of CMV and HSV so often observed in humans identical or similar mechanism is involved. We trust that the study of cell systems harbouring latent virus might possibly bring us nearer to the understanding of this mechanism. We are grateful to Mrs Maria Rozgonyi for her excellent technical assistance. DISCUSSION CRAIGHEAD, J. E., HANSHAW, J. B. & CARPENTER, C. B. (I967). Cytomegalovirus infection after renal allotransplantation. The Journal of the American Medical Association 2oi, DIOSI, v., MOLDOVAN, E. & TOMESCU, N. (I969). Latent cytomegalovirus infection in blood donors. British Medical Journal 4, 66o-662. DUFF, a. & RAPP, F. (1971). Properties of hamster embryo fibroblast transformed in vitro after exposure to ultraviolet-irradiated herpes simplex virus type 2. Journal of Virology 8, GEDER, L., KOLLER, M., GONCZOL, E., JENEY, E. & Gt)NCZ6L, I. 0963). Isolation of herpes zoster virus strain. Acta microbiologica Academiae scientiarum hungaricae xo, I55-I61. GELDERMAN, A. n., GRIMLEY, P. M., LUNDE, M. N. & RABSON, A. S. (I968). Toxoplasma gondii and cytomegalovirus: mixed infection by a parasite and a virus. Science, New York x6o, I I3O-I 132. HINZE, It. C. & WALKER, D. L. (I96I). Variation of herpes simplex virus in persistently infected tissue cultures. Journal of Bacteriology 82, 498-5o4. JENSEN, V. C. & KOPROWSKI, H. (1969). Absence of repressor in SV40-transformed cells. Virology 37, MUNK, K. & RUNNEBAUM, H. (I970). Virusinfectionen nach Organtransplantationen bei immunsuppressiver Therapie. Deutsche medizinische Wochenschrift 95, O'NEIL, E. J., GOLDBERG, R. Y. & RAPP, F. (I972). Herpes simplex virus latency in cultured human cells following treatment with cytosine-arabinoside. The Journal of General Virology x4, I89-I97.

9 Cytomegalovirus latency in cultured human cells 15 I PEP, HAM, T. G. M., AUL, E.O., CONEWAY, P.J. & MOTT, M.G. (I97I). Cytomegalovirus infection in blood donors - a prospective study. British Journal of Haematology 20, 3o7-32o. REED, L. S. & MUENCH, H. (1938). A simple method of estimating fifty per cent endpoints. American Journal of Hygiene 27, SAOER, 6. & HAHN, E. C. (I970). The superinfection of abortively infected monkey kidney cells with SV40. Arehiv fiir die gesamte Virusforschung 30, o. SETO, D. S. Y. & CARVER, D. H. (1969). Interaction between cytomegalovirus and Newcastle disease virus as mediated by intrinsic interference. Journal of Virology 4, SZANTO, J. (I963). Course of persistent infection of HeLa and Detroit 6 cells with herpes simplex virus. Aeta virologica 7, v~,czi, L. & GONCZ6L, E. (1972). The effect of cytosine-arabinoside on the multiplication of cytomegalovirus and on the formation of virus induced intracellular antigens. Acta virologica (in the press). (Received 17 July 1972 ) II VIR 18