Jul 29, 1994 - FERENC D. TO´ TH,1,2 PETER MOSBORG-PETERSEN,1 JOLÃN KISS,2 GEORGE ..... aminocellulose by the method described by Moudgal and Porter (31). ...... Rabbit antiserum to HIV-1 Tat from Bryan Cullen was obtained.
JOURNAL OF VIROLOGY, Apr. 1995, p. 2223–2232 0022-538X/95/$04.0010 Copyright q 1995, American Society for Microbiology
Vol. 69, No. 4
Interactions between Human Immunodeficiency Virus Type 1 and Human Cytomegalovirus in Human Term Syncytiotrophoblast Cells Coinfected with Both Viruses ´ TH,1,2 PETER MOSBORG-PETERSEN,1 JOLA ´ N KISS,2 GEORGE ABOAGYE-MATHIESEN,1 FERENC D. TO HENRIK HAGER,1 CLAUS B. JUHL,1 LAJOS GERGELY,2 MILAN ZDRAVKOVIC,1 ´ NOS ARANYOSI,3 LA ´ SZLO ´ LAMPE ´ ,3 AND PETER EBBESEN1* JA Department of Virus and Cancer, The Danish Cancer Society, DK-8000 Aarhus C, Denmark,1 and Institute of Microbiology2 and Department of Obstetrics and Gynaecology,3 University Medical School, H-4012 Debrecen, Hungary Received 29 July 1994/Accepted 29 December 1994
Human cytomegalovirus (HCMV) and human immunodeficiency virus type 1 (HIV-1) may interact in the pathogenesis of AIDS. The placental syncytiotrophoblast layer serves as the first line of defense of the fetus against viruses. We analyzed the patterns of replication of HIV-1 and HCMV in singly an dually infected human term syncytiotrophoblast cells cultured in vitro. Syncytiotrophoblast cells exhibited restricted permissiveness for HIV-1, while HCMV replication was restricted at the level of immediate-early and early gene products in the singly infected cells. We found that the syncytiotrophoblasts as an overlapping cell population could be coinfected with HIV-1 and HCMV. HIV-1 replication was markedly upregulated by previous or simultaneous infection of the cells with HCMV, whereas prior HIV-1 infection of the cells converted HCMV infection from a nonpermissive to a permissive one. No simultaneous enhancement of HCMV and HIV-1 expression was observed in the dually infected cell cultures. Major immediate-early proteins of HCMV were necessary for enhancement of HIV-1 replication, and interleukin-6 production induced by HCMV and further increased by replicating HIV-1 synergized with these proteins to produce this effect. Permissive replication cycle of HCMV was induced by the HIV-1 tat gene product. We were unable to detect HIV-1 (HCMV) or HCMV (HIV-1) pseudotypes in supernatant fluids from dually infected cell cultures. Our results suggest that interactions between HIV-1 and HCMV in coinfected syncytiotrophoblast cells may contribute to the transplacental transmission of both viruses. tion of cytokines that then activate HIV-1 in latently infected cells (4, 9, 40). It has been observed that HCMV can reactivate during pregnancy without a corresponding increase in antibody titer (51). Since HCMV antigens and genetic material have been found in trophoblastic tissue infected with HCMV in vitro (1), it is reasonable to assume that HIV-1 and HCMV can coinfect the same ST target. In this study, we examined interactions between HIV-1 and HCMV in human term ST cells cultured in vitro. We show that the ST cells as overlapping target cell population can be coinfected with HIV-1 and HCMV. We find that the response of ST cells to HIV-1 infection is semipermissive but expression of HIV-1 is markedly upregulated by previous or simultaneous infection with HCMV. HCMV gene expression is restricted to the immediate-early (IE) and early genes in ST cells. However, previous infection of the same cellular targets with HIV-1 removes this block, and productive HCMV infection occurs. Our results suggest that interactions between HIV-1 and HCMV in ST cells may contribute to the spread of these viruses from mother to the fetus.
Vertical transmission of human immunodeficiency virus type 1 (HIV-1) from mother to child is one major problem concerning the propagation of AIDS. In addition to mother-infant transmission of HIV-1 by exposure to blood during the perinatal period (62) and via breast milk (55), it is currently believed that intrauterine transplacental infection of the fetus is the most important mechanism of vertical transmission (50). The syncytiotrophoblast (ST) layer, a fetal placental derivative, forms the critical selective barrier separating maternal blood from the underlying connective tissue (11). Thus, only after traversing the ST would HIV-1 be exposed to underlying cytotrophoblast cells and to placental macrophages, fibroblasts, and endothelial cells. Although HIV-1 can enter ST cells (59), the trophoblast is assumed to restrict the rate of virus transmission, because trophoblast cells exhibit restricted permissiveness for HIV-1 and the limited virus shedding is followed by a continuous decline in HIV-1 gene activities, resulting in sequestration of provirus (60). These findings suggest that multiple cofactors are likely involved in transplacental transmission of the virus. One possible mechanism for the stimulation of HIV-1 replication is coinfection with other viruses (12). A good candidate for this scenario is human cytomegalovirus (HCMV). HCMV encodes its own trans-acting factors, which can activate HIV-1 expression through specific sequences in the long terminal repeat of HIV-1 (5). In addition, HCMV can induce produc-
MATERIALS AND METHODS Viruses. The IIIB strain of HIV-1 was grown in H9 cells. The AD169 strain of HCMV was propagated in MRC-5 cells. Supernatants were cleared of cells by centrifugation, filtered through 0.45-mm-pore-size filters, and stored at 2808C until use. Cells and supernatants were determined to be mycoplasma free by a commercial screening system (Bionique Laboratories, Inc., Saranac Lake, N.Y.). Cell lines. HIV-1 stocks were titrated by syncytium-forming assay in CEM-SS cells (33). The CEM-SS and H9 cells were maintained in RPMI 1640 medium containing 10% fetal calf serum (FCS), 250 U of penicillin per ml, and 125 mg of streptomycin per ml. For production and titration of HCMV, MRC-5 cells were used. The MRC-5 cells were grown in Eagle’s minimum essential medium supplemented with 10% FCS and antibiotics.
* Corresponding author. Mailing address: The Danish Cancer Society, Department of Virus and Cancer, Gustav Wieds Vej 10, DK-8000 Aarhus C, Denmark. Phone: 45-86-12-73-66. Fax: 45-86-19-54-15. 2223
2224
´ TH ET AL. TO
Isolation, culture, and differentiation of cytotrophoblast cells. Pure populations of term placental trophoblast cells were separated by the method of Douglas and King (7). As a first step, a cell preparation highly enriched for cytotrophoblasts was obtained by using a discontinuous Percoll gradient as described by Kliman et al. (23). The remaining contaminating cells were then removed by treatment with mouse monoclonal antibodies (MAbs) against class I and class II major histocompatibility complex antigens (Dako, Copenhagen, Denmark) followed by magnetic microspheres coated with goat anti-mouse immunoglobulin (IgG) (Dynal AS, Oslo, Norway). The purity of the trophoblast population was tested by indirect immunofluorescence assay (IFA) using monoclonal mouse IgG antibodies to classic epithelial cell (cytokeratin), fibroblast (vimentin), T-cell (CD3), monocyte (CD14), endothelial cell (CD31), and macrophage (CD68) markers and fluorescein isothiocyanate (FITC)-conjugated rabbit anti-mouse IgG (Dako). No vimentin-, CD3-, CD14-, CD31-, or CD68-positive cells could be detected, indicating complete removal of all possible contaminating elements. The cells isolated by this procedure were cytokeratin positive and considered to be pure cytotrophoblasts (21). For in vitro differentiation of trophoblast cells, keratinocyte growth medium (Clonetics Corporation, San Diego, Calif.) supplemented with 15% heat-inactivated FCS was used. Plating efficiency was usually about 85%, and cell numbers were adjusted to about 250,000 cells per cm2. Syncytium formation was revealed by desmoplakin staining as described elsewhere (8). After 4 days in culture, virtually all colonies consisted of aggregates of large multinucleated cells. To exclude the possibility that viral growth in various experiments is influenced by the different states of cellular differentiation, the release of chorionic gonadotropin from trophoblasts was measured at different time intervals after seeding them in culture. The enzyme-linked immunosorbent assay (ELISA) kit used for detection of human chorionic gonadotropin was purchased from Biomedica (Vienna, Austria). The release of hormone reached a maximum level at day 5, after which it showed a plateau and no further increase in hormone production could be detected. The morphologic differentiation was in a good correlation with the hormone production. We used 5-day-old ST cultures in all experiments. Cell cultures were tested routinely for mycoplasma contamination, and only cultures found to be negative were used in the experiments. Coinfection procedure. To determine what effect time of HIV-1 exposure in relation to HCMV infection has on virus replication, HIV-1 and HCMV were added at the same time to the cultures, or one virus was added to the cells earlier than the other. Control cultures received the same quantity of either virus alone. Five-day-old ST cultures were infected in 25-cm2 flasks (Nunc, Copenhagen, Denmark) with HIV-1 at a multiplicity of infection (MOI) of 1.0 syncytiumforming unit per cell. HCMV infections were performed at a MOI of 1.0 PFU per cell. The MOI was calculated according to the number of cytotrophoblast cells seeded for syncytium formation. ST cells were incubated with virus for 2 h at 378C and then washed five times with serum-free medium. The final wash was taken for detection of residual virus. After washing, fresh medium was added to the cells. Detection of virus replication. At different time intervals, clarified culture supernatants were evaluated for HIV-1 by reverse transcriptase (19) and CEM-SS syncytium-forming (33) assays. Production of infectious HCMV by ST cells was assessed by PFU assay in MRC-5 cells (6). Assay for cytokine determination. Supernatants of virus-infected and uninfected control cultures were collected at different time intervals after infection and assayed for alpha interferon (IFN-a), IFN-b, interleukin-1a (IL-1a), IL-1b, IL-2, IL-6, tumor necrosis factor alpha (TNF-a), TNF-b, and granulocyte-macrophage colony-stimulating factor (GM-CSF) activities. Cytokine concentrations were measured in each supernatant by using ELISA kits according to the manufacturers’ technical guidelines. IFN-a and IFN-b kit systems were from Serotec (Kidlington, Oxford, England). ELISA kits specific for IL-1a, IL-1b, IL-2, and TNF-b were purchased from R & D Systems (Minneapolis, Minn.). ELISA kits used for detection of IL-6, TNF-a, and GM-CSF were products of Innogenetics N.V. (Zwijnaarde, Belgium). IFAs. (i) Immune sera and conjugates. Mouse MAbs to IE (catalog no. 207-88), early (catalog no. 207-89), and late (catalog no. 207-88) antigens of HCMV as well as FITC-conjugated MAb to IE antigen 1 (IEA-1) of HCMV (catalog no. 307-83) were purchased from Biosoft (Paris, France). MAb to HIV-1 Tat protein (catalog no. 85 329) was from American BioTechnologies (Cambridge, Mass.), and MAb to HIV-1 p24 antigen (catalog no. M 857) was from Dako. Rabbit antiserum raised against HIV-1 Tat protein (catalog no. 705) was kindly supplied by the NIH AIDS Research and Reference Reagent Program (Rockville, Md.). Goat anti-mouse Ig conjugated to FITC (catalog no. GM17F) and goat anti-mouse Ig conjugated to tetramethyl rhodamine isothiocyanate (TRITC) (catalog no. GM17T) were produced by the Central Laboratorium van de Bloedtransfusiedienst van het Nederlandse Rode Kruis (Amsterdam, The Netherlands). Swine anti-rabbit Ig conjugated to FITC (catalog no. F 205) was purchased from Dako. (ii) Indirect cytoplasmic IFA. For analyses of expression of virus-induced proteins, indirect cytoplasmic IFA was used. Immunofluorescence staining was performed on cells which had been cultured in chamber slides (Nunc). After removal of the culture medium, the cells were washed three times in phosphatebuffered saline (PBS) and then fixed in 80% acetone for 10 min at 2208C. Before staining, the slides were washed again three times in PBS containing 1% FCS (PBS-FCS). Primary antibodies were diluted according to the manufacturers’
J. VIROL. instructions and incubated with cells for 30 min at 378C. Cells were then washed by three sequential washes of 10 min each in PBS-FCS. Appropriate secondary antibodies were then added, and the slides were incubated for 30 min at 378C. After incubation, the samples were washed three times in PBS-FCS and once in distilled water. Coverslips were mounted with a solution containing equal parts of glycerol and PBS, and slides were examined by fluorescence microscopy. As specific controls, we demonstrated that virus-specific antibodies did not bind to uninfected cells and that the secondary antibodies did not stain cells by itself. (iii) Two-color immunofluorescence tests. Simultaneous detection of HIV-1 Tat protein and HCMV late antigen (LA) in dually infected ST cells was carried out by double indirect cytoplasmic IFA. Cells fixed in 80% acetone were incubated for 30 min at 378C with MAb to HCMV LA diluted 1:40 and rabbit antiserum raised against HIV-1 Tat protein diluted 1:1,000. Control samples were incubated with anti-HCMV LA MAb plus normal rabbit serum or with anti-HIV-1-Tat rabbit immune serum plus normal mouse serum as well as with both normal sera. After incubation, the cells were washed three times in PBSFCS. Thereafter, TRITC-labeled goat anti-mouse Ig serum diluted 1:30 and FITC-labeled swine anti-rabbit Ig serum diluted 1:20 were added to the samples. The cells were incubated for another 30 min at 378C and then washed three times in PBS-FCS and once in distilled water before evaluation in fluorescence microscope. Coexpression of HCMV IEA and HIV-1 p24 antigen in ST cells infected with both viruses was demonstrated by a combination of direct and indirect cytoplasmic IFA. In the first phase of the reaction, indirect cytoplasmic IFA was performed with MAb to HIV-1 p24 antigen diluted 1:40 and TRITC-labeled goat anti-mouse Ig serum diluted 1:30. In the second phase of the reaction, cells were incubated with FITC-conjugated MAb to HCMV IEA-1 diluted 1:40 for 30 min at 378C and then washed four times before evaluation. Control cell samples treated with normal mouse serum and/or conjugates were included in the experiments. Covalent attachment of antibodies to the outer surface of liposomes. (i) Monospecific antibodies. Antibodies to Tat HIV-1 protein were isolated from a rabbit antiserum (NIH AIDS Research and Reference Reagent Program; catalog no. 705) by using an insoluble form of Tat protein. Full-length Tat protein was chemically synthesized by the solid-phase method, using fluorenylmethyloxycarbonyl chemistry on polystyrene resin (41) in a model 431A peptide synthesizer (Applied Biosystems, Foster City, Calif.). Tat protein was coupled to aminocellulose by the method described by Moudgal and Porter (31). Rabbit antiserum was added to a suspension of antigen cellulose, and the mixture was incubated for 1 h at room temperature and overnight at 28C. The suspension was washed with cold 1% NaCl until the washings contained no protein. For dissociation of antibodies from the antigen cellulose, 1% NaCl adjusted with HCl to pH 2 was used as the eluting agent. The pH was then readjusted to 7.2 by dialysis against PBS. MAb E-13, reacting with both IE-1- and IE-2-encoded HCMV proteins (30), was produced by Biosoft, and neutralizing MAbs to human IL-2 (catalog no. 1 444 603) and IL-6 (catalog no. 1 271 121) were purchased from Boehringer-Mannheim GmbH (Mannheim, Germany). Normal mouse IgG (Sigma Chemical Co., St. Louis, Mo.) was used as a control. (ii) Cross-linking of lipid to the antibody carbohydrate chain. Antibodies were chemically modified to increase cellular penetration by covalent attachment of antibodies to the external surface of liposomes (13). Briefly, small unilamellar vesicles were prepared from mixtures of phosphatidylethanolamine and phosphatidylcholine by sonication of lipid. After sonication, the suspension was centrifuged at 400,000 3 g for 1 h to remove the larger liposomes. Antibodies were activated for attachment to liposomes by treatment with 1% fluorodinitrobenzene and 0.06 M NaIO4. Activated antibodies were separated from the excess reagents by desalting on Sephadex G-50 into 10 mM succinate buffer. For coupling to liposomes, the activated antibodies were mixed with liposome suspension, and 1 M carbonate buffer (pH 9.5) was then added to raise the pH to 9.5 for reaction. After 2 h, NaBH4 was added, and the solution was left at room temperature overnight. The mixture was then neutralized to pH 6.0 with HCl, and the liposomes and nonbound antibodies were separated by sedimentation at 400,000 3 g for 1 h. Small unilamellar vesicles were concentrated on an Amicon PM10 membrane and applied to Sepharose 6B column eluted with 10 mM succinate buffer. (iii) Evaluation of cellular penetration and subcellular localization of antibodies covalently linked to liposomes. The passage of chemically modified antibodies through the cytoplasm to the nucleus was evaluated by detection of radioactively labeled antibodies in the various subcellular fractions. Before linkage to liposomes, antibodies were labeled with 125I by the chloramine-T method (14). ST cells cultured in chamber slides were incubated with chemically modified, radioactively labeled antibodies at a concentration of 10 mg/ml. At different time intervals, the medium was removed, and subcellular fractions were separated as described by New et al. (34). Briefly, cells were homogenized in ice-cold 0.25 M sucrose–5 mM Tris-HCl (pH 7.4), and the various subcellular fractions were separated by differential centrifugation at 48C. Then samples of each fraction were counted for radioactivity. The rate of radioactivity in the nuclear fraction reached a peak value as early as 2 h after exposure. The presence of 22.5% relative specific activity in this cellular compartment indicated the traffic of antibodies to the nucleus. PCR. For detection of HIV-1 proviral DNA in the cells, a protocol described by Schnittman et al. (47) was used. Briefly, genomic DNA was coamplified by
VOL. 69, 1995
COINFECTION OF SYNCYTIOTROPHOBLASTS WITH HIV-1 AND HCMV
2225
using the HIV-1 gag-specific primers SK 38 and SK 39 (37) and primers PCO3 and PCO4 specific for the b-globin gene (45). Amplification was carried out by means of a thermocycler (Perkin-Elmer Cetus, Norwalk, Conn.) with denaturation at 948C, annealing at 558C, and extension at 728C for 40 cycles. Amplification products were subjected to electrophoresis in 1% agarose gel, transferred onto a Hybond NTC membrane (Amersham, Little Chalfont, England), and hybridized with a 32P-labeled probe (SK 19 gag) at 568C for 60 min. After hybridization, the membranes were exposed to Kodak XAR film at 2708C. Synthetic oligonucleotides (Pharmacia, Uppsala, Sweden) were used to amplify DNA from the major IE gene of HCMV. The sequences were (59 to 39) CCACCCGTGGTGCCAGCTCC (IE-1, upstream primer) CCCGCTCCTCCT GAGCACCC (IE-2, downstream primer), and CTGGTGTCACCCCCAGAGT CCCCTGTACCCGCGACTATCC (IE-3, probe) for the oligomer set complementary to the IE gene (52). Globin primers (PCO3 and PCO4) and probe (RS 18) corresponded to published sequences (44). The HCMV PCR was performed with Taq polymerase (Perkin-Elmer Cetus) as described by Chou (3). In brief, temperature cycling consisted of 948C for 1 min, 558C for 3 min, and 728C for 2 min, repeated for 40 cycles. The primers consisted of the HCMV pairs (IE-1 and IE-2) and the primer pair specific for b-globin (PCO3 and PCO4). Amplification was verified by gel electrophoresis, Southern blotting, and hybridization with 32 P-end-labeled IE-3 oligomer at 568C for 60 min. Autoradiography was performed at 2708C with Kodak XAR film.
RESULTS Virus replication in ST cells infected with HCMV or HIV-1 alone. Using indirect cytoplasmic IFA, we found IE and E antigens in HCMV-infected cells, but no LA was seen. Consistent with the absence of LA expression, we could not detect production of infectious virus. The IEA of HCMV was found in the nuclei of the HCMV-infected ST cells within 1 h after infection. The largest number of IEA-positive nuclei was seen 3 days after infection, after which there was a decrease in the frequency of IEA-positive cells (Fig. 1A). After day 6, no IEA expression was observed. The early gene product was detectable in the nuclei within 24 h after infection and persisted as long as 8 days in the culture (Fig. 1B). HIV-1-infected ST cells were examined for the presence of Tat protein by indirect cytoplasmic IFA. Nuclei stained positively for Tat first on day 9. The nuclear staining reached a maximum level on day 11 and persisted by day 14 (Fig. 1C). Since cells infected with HIV-1 alone did not produce enough cell-free virus to be measured directly by reverse transcriptase assay, HIV-1 yields were determined by CEM-SS syncytiumforming assay. In cultures infected with HIV-1 alone, infectious virus was first detected on day 10; the yield reached a peak value on day 12 and sharply decreased thereafter (Fig. 2). Effect of dual infection on HIV-1 and HCMV production. The interaction between HCMV and HIV-1 was examined in timed experiments in which either ST cells were infected with both viruses at the same time or one virus was added to the cells earlier than the other. Figure 2 shows that HIV-1 replication was enhanced by HCMV when both viruses were added simultaneously to the cellular targets or HCMV-infected cells were superinfected with HIV-1. In cell cultures coinfected simultaneously with HCMV and HIV-1 or infected with HIV-1 on day 1 after HCMV infection, HCMV had a dramatic stimulating effect on HIV-1 replication. HCMV was able to enhance HIV-1 replication by 1,500-fold, with peak levels at 2 to 4 days postinfection. Hence, an accelerated expression of HIV-1 in dually infected cultures was further indicated by the earlier detection of release of infectious virus. After day 3 or 4, HIV-1 production by coinfected ST cells sharply decreased, and by 18 days, HIV-1 was no longer recoverable from the supernatant of dually infected cultures. Similar observations were made when cells preinfected with HCMV were superinfected with HIV-1 on day 4 after HCMV infection. However, the rate of enhancement of HIV-1 replication was somewhat lower, and HIV-1 production by ST cells ceased by day 10. When cells prein-
FIG. 1. Detection of virus-induced, nonstructural antigens in human ST cells by indirect immunofluorescence microscopy. (A) HCMV IEA at 6 h after HCMV infection; (B) HCMV early antigen at 16 h after HCMV infection; (C) HIV-1 Tat protein at 10 days after HIV-1 infection. Magnification, 3166.
fected with HCMV were superinfected with HIV-1 on day 7 or later after HCMV infection, no stimulation of HIV-1 replication was observed (data not shown). It is noteworthy that cell cultures coinfected simultaneously with both viruses or infected with HIV-1 after HCMV infection did not produce any infectious HCMV. Heat inactivation of HCMV (1 h, 568C) resulted in total abrogation of its enhancing effect on the replication of HIV-1 (data not shown). In contrast to these findings, preinfection of ST cells with HIV-1 converted HCMV infection from a nonpermissive to a productive one (Fig. 3). However, this phenomenon proved to be time interval dependent. Only supernatants from ST cells
2226
´ TH ET AL. TO
FIG. 2. Effect of HCMV infection on HIV-1 production. ST cells were infected with HIV-1 alone (F), HCMV and HIV-1 simultaneously (E), HCMV for 1 day and then HIV-1 (Ç), or HCMV for 4 days and then HIV-1 (h). At various times postinfection, supernatants were collected and the amount of HIV-1 present was determined by syncytium-forming assay. The results of five separate experiments (mean 6 standard deviations) are presented.
infected with HCMV at days 7 and 11 after HIV-1 infection were able to produce characteristic HCMV cytopathic effect when cultured on MRC-5 monolayers; supernatant fluids from ST cells infected with HCMV earlier did not produce HCMV cytopathic effect. In cultures infected with HCMV on day 7 after HIV-1 infection, maximal HCMV production occurred on day 8 after exposure to HCMV, reaching 7.5 3 103 PFU/ml of supernatant. When ST cells were infected with HCMV on day 11 after HIV-1 infection, maximal HCMV production appeared on day 5 after HCMV infection. Then HCMV production sharply decreased, and no virus release was detected after 15 days postinfection in the cultures. Interestingly, HCMV-
FIG. 3. Induction of HCMV replication by coinfection with HIV-1. ST cells were infected with HIV-1 for 1 day (E), for 4 days (h), for 7 days (}), and for 11 days (å) and then with HCMV. At various times postinfection, supernatants were collected and the amount of HCMV present was determined by plaque assay. The results of five separate experiments (mean 6 standard deviations) are presented.
J. VIROL.
producing ST cultures survived after 15 days. No HIV-1 production was observed in HCMV-producing dually infected ST cell cultures. Live HIV-1 was required to convert HCMV infection from a nonpermissive to a productive one, since heatinactivated (1 h, 568C) HIV-1 was unable to induce replication of HCMV (data not shown). Presence of both HIV-1 and HCMV in the same cell. Twocolor IFA with mouse MAbs to HCMV IEA and HIV-1 p24 showed the simultaneous presence of both viruses in the same cells of dually infected cultures producing high levels of HIV-1 (Fig. 4). In addition, using a double-label antibody technique with mouse MAb to HCMV LA and rabbit antiserum to HIV-1 Tat, we found cells coinfected with both viruses in HCMVproducing, dually infected ST cell cultures (Fig. 5). In both cases, the percentages of double-positive cells were above 80%. These results demonstrated that individual ST cells could serve as simultaneous targets for both HCMV and HIV-1. Effects of HCMV and HIV-1 viral infections and coinfections on cytokine synthesis. To determine which of the cytokine-modulating activities of HCMV and HIV-1 may have a role in regulation of viral expression and/or replication, we compared the cytokine secretions in cultures of ST cells infected with HCMV or HIV-1 alone or coinfected with both viruses. In some cases, before their use for stimulation, viruses were inactivated by heat (1 h, 568C). Supernatants of the virustreated cell cultures were harvested at different time intervals and assayed for IFN-a, IFN-b, IL-1a, IL-1b, IL-2, IL-6, TNF-a, TNF-b, and GM-CSF activities. No IFN-a, IFN-b, IL-1a, IL-1b, TNF-a, TNF-b, and GM-CSF activities could be detected in supernatants of ST cell cultures either before or after infection with HCMV, HIV-1, or both viruses. As shown in Fig. 6, uninfected ST cell cultures did not release any IL-6 activity, whereas live HCMV induced IL-6 production. The IL-6 induction was rapid; IL-6 was detected as early as 6 h after stimulation with HCMV, reaching a peak value of 280 pg/ml at 24 h after HCMV exposure. Release of IL-6 into the culture supernatant gradually decreased over the next 2 days. Exposure to inactivated HCMV preparation induced amounts of IL-6 production comparable to that induced by live HCMV preparation (data not shown). These observations indicate that in the case of HCMV, a structural protein seems responsible for induction of IL-6 synthesis. In contrast to HCMV, results indicated that HIV-1 alone had no IL-6inducing potential. When ST cells were superinfected with HIV-1 on day 1 after HCMV infection, an approximately 2.5fold increase in IL-6 production was detected compared with that in cells induced with HCMV alone. Live HIV-1 was required to enhance HCMV-induced IL-6 production, since heat-inactivated HIV-1 was unable to cause a further increase in IL-6 release (data not shown). In contrast to these results, previous infection of ST cells with HIV-1 did not influence HCMV-induced IL-6 production (data not shown). Uninfected ST cells constitutively released low levels (40 pg/ml) of IL-2 into the culture medium (Fig. 7). Following single viral infections, both HCMV and HIV-1 were found to enhance IL-2 production. In contrast, no increase in IL-2 production was found in response to inactivated HCMV or HIV-1. In cell cultures induced with live HCMV, maximal IL-2 release was observed at day 4 postinfection and decreased sharply thereafter. HCMV induced a 10-fold increase in IL-2 production compared with that in uninfected cells. In ST cells induced with live HIV-1, IL-6 release was maximal on day 3 following infection and decreased over the next 2 days. HIV-1-infected cells produced approximately 11 times more IL-2 than did uninfected cells. Superinfection of ST cells with HIV-1 on day 1 after HCMV infection further increased the rate of IL-2
VOL. 69, 1995
COINFECTION OF SYNCYTIOTROPHOBLASTS WITH HIV-1 AND HCMV
2227
FIG. 4. Simultaneous expression of HCMV IE and HIV-1 p24 proteins in the same ST cells. Cells were infected with HCMV for 1 day and then with HIV-1. The dually infected chamber slide culture was fixed for immunofluorescence staining on day 2 after HIV-1 infection. Cells shown in panel A were stained with FITC-conjugated mouse MAb against the IEA-1 of HCMV. The same cells shown in panel B were incubated with mouse MAb against the p24 antigen of HIV-1, and then immune complexes were visualized with TRITC-conjugated goat anti-mouse Ig serum. FITC and TRITC were excited to fluorescence at wavelengths of 488 and 555 nm, respectively. Note the nuclear localization of HCMV IEA-1 in panel A and cytoplasmic localization of HIV-1 p24 antigen in panel B. Magnification, 3200.
production induced by HCMV alone. Compared with that in uninfected cells, a 23-fold elevation in IL-2 secretion was observed in dually infected cell cultures on day 4 after HCMV exposure. In contrast, infection of ST cells with HIV-1 7 or 11 days before HCMV infection did not affect the stimulation of IL-2 by HCMV, because the amounts of IL-2 detected were comparable to those elicited by HCMV alone. HIV-1 and HCMV stock preparations used for our studies were also assayed for IL-2 and IL-6, but no such activities were found in them. Effects of antibodies to HCMV IEA, HIV-1 Tat, IL-2, and IL-6 on virus replication in dually infected cells. On the basis of the results of the previous experiments, we postulated that IEAs and virus-induced interleukin production are mediators of HCMV-induced stimulation of HIV-1 replication and conversely, HIV-1 tat stimulates HCMV gene expression and rep-
lication. To test this hypothesis, monospecific antibodies, chemically modified to increase cellular penetration, were used at a concentration of 10 mg/ml as means of blocking transactivating viral proteins and interleukin activities. For study of the inhibitory effects of MAbs on HCMVinduced enhancement of HIV-1 replication, cell cultures infected with HIV-1 on day 1 after HCMV infection were used. Treatment of ST cells with MAbs was started 24 h before exposure to HCMV. Following adsorption of HCMV and HIV-1, the viruses were removed and the cells were washed three times, after which fresh medium supplemented with 10 mg of the appropriate MAb per ml was added. The medium was changed daily, and fresh antibodies were added till day 6 after HCMV infection. As seen in Fig. 8, MAb E-13 reacting with IEA-1 and IEA-2 of HCMV could completely eliminate the HIV-1 replication-enhancing activity of HCMV. In dually
2228
´ TH ET AL. TO
J. VIROL.
FIG. 5. Simultaneous expression of HIV-1 Tat protein and a late antigen of HCMV in the same ST cells. Cells were infected with HIV-1 for 7 days and then with HCMV. The dually infected chamber slide culture was fixed for immunofluorescence staining on day 5 after HCMV infection. Dual immunofluorescence localization of HIV-1 (A) and HCMV (B) proteins was detected with a monospecific rabbit polyclonal antiserum against HIV-1 Tat protein and mouse MAb against HCMV LA. Antibody staining was followed by FITC-conjugated swine anti-rabbit and TRITC-conjugated goat anti-mouse Ig second antibodies. FITC and TRITC were excited to fluorescence at wavelengths of 488 and 555 nm, respectively. Note the nuclear localization of HIV-1 Tat in panel A and cytoplasmic localization of HCMV late antigen in panel B. Magnification, 3200.
infected cell cultures treated with MAb E-13, the rate and kinetics of HIV-1 replication were the same as in cell cultures infected with HIV-1 alone (Fig. 2 and 8). Anti-IL-6 antibodies were less potent inhibitors of HCMV-induced stimulation of HIV-1 replication. Dually infected cell cultures, treated with anti-IL-6 MAb, produced approximately four times less HIV-1 than did untreated cultures. The degree of inhibition could not be elevated by further increase in the concentration of antiIL-6 MAb. In this system, anti-IL-2 MAb or normal mouse IgG up to 20 mg/ml had no inhibitory effect on HCMV-induced stimulation of HIV-1 replication (data not shown). Next we studied the effects of anti-Tat, anti-IL-2, and antiIL-6 antibodies on HIV-1-induced HCMV gene expression and replication in ST cell cultures infected with HCMV on day 11 after HIV-1 infection. Antibody treatments were started 1 day before infection with HCMV and continued for another 6
days. As shown in Fig. 9, anti-Tat antibody at a concentration of 10 mg/ml was capable of totally suppressing the replication of HCMV in this system. In contrast, no inhibitory effect of anti-IL-2 or anti-IL-6 MAbs or normal mouse IgG on HCMV replication was detected at concentrations of 10 mg/ml or higher (data not shown). In control experiments, uninfected ST cells were treated with chemically modified antibodies at a concentration of 10 mg/ml for 6 days. There was no difference in cell viability between antibody-treated and control cultures as determined by the trypan blue dye exclusion test. Attempts to detect pseudotypes. When HCMV-infected ST cells were superinfected with HIV-1 or both viruses were added simultaneously to the cellular targets, enhanced HIV-1 production was observed (Fig. 2) but no infectious HCMV could be detected. When cells preinfected with HIV-1 were
VOL. 69, 1995
COINFECTION OF SYNCYTIOTROPHOBLASTS WITH HIV-1 AND HCMV
2229
FIG. 6. Kinetics of IL-6 production after exposure of ST cells to HIV-1, HCMV, or both viruses. Symbols: E, uninfected cells; h, cells infected with HIV-1 alone; Ç, cells infected with HCMV alone; å, cells infected with HCMV for 1 day and then with HIV-1. At various times postinduction, supernatants were obtained and the amount of IL-6 present was measured with an ELISA kit. Data are expressed as means 6 standard deviations (five independent experiments).
FIG. 8. Effect of treatment with antibodies to HCMV IEAs or IL-6 on HIV-1 replication in dually infected cells. Cell cultures were infected with HCMV for 1 day and then with HIV-1. For details of antibody treatment, see the text. Symbols: Ç, untreated cells; ■, cells treated with anti-HCMV IEA MAb (10 mg/ml); F, cells treated with anti-IL-6 MAb (10 mg/ml). At various times postinfection, supernatants were collected and the amount of HIV-1 released was determined by syncytium-forming assay. Each point represents mean 6 standard deviations (five separate experiments).
superinfected with HCMV, HCMV gene expression and replication were induced (Fig. 3) without any HIV-1 production by the dually infected cells. However, coinfection of cells with HCMV and HIV-1 may result in formation of hybrid virions composed of the nucleocapsid of HIV-1 and the envelope glycoproteins of HCMV, or vice versa. Progeny viruses from culture supernatant fluids of HCMV- or HIV-1-producing dually infected ST cells were tested for the presence of pseudotypes by PCR. MRC-5 cells were infected with culture supernatant fluid from HCMV-producing ST cell cultures at 2, 4, 6, 8, and 10 days after HCMV infection. On day 3 after infection, genomic
DNA was extracted from both infected and uninfected MRC-5 cells and then tested for the presence of the HIV-1 genome by PCR. DNA samples from uninfected and HIV-1-infected H9 cells were used as controls. No HIV-1 DNA was detected in any of the infected MRC-5 cell samples tested (data not shown). Supernatants of HIV-1-producing, dually infected ST cell cultures were tested for the presence of pseudotypes consisting of HCMV core and HIV-1 envelope. H9 cells were infected with culture supernatant fluid from these cultures at 1 to 6 days
FIG. 7. Kinetics of IL-2 production after exposure of ST cells to HIV-1, HCMV, or both viruses. Symbols: E, uninfected cells; h, cells infected with HIV-1 alone; Ç, cells infected with HCMV alone; å, cells infected with HCMV for 1 day and then with HIV-1. At various times postinduction, supernatants were obtained and the amount of IL-2 present was measured with an ELISA kit. Data are expressed as means 6 standard deviations (five independent experiments).
FIG. 9. Effect of treatment with antibodies to HIV-1 Tat protein on HCMV replication in dually infected cells. Cell cultures were infected with HIV-1 for 11 days and then with HCMV. For details of antibody treatment, see the text. Symbols: å, untreated cells; É, cells treated with anti-Tat antibodies (10 mg/ml). At various times postinfection, supernatants were collected and the amount of HCMV released was determined by plaque assay. Each point represents mean 6 standard deviations (five separate experiments).
2230
´ TH ET AL. TO
J. VIROL.
after HIV-1 infection. On day 3 after infection, genomic DNA was extracted from both infected and uninfected H9 cells and then tested for the presence of the HCMV genome by PCR. DNA samples from uninfected and HCMV-infected MRC-5 cells were included in each experiment. We were unable to demonstrate rescue of defective HCMV genome by infectious HIV-1 (data not shown). DISCUSSION In this study, we found that ST cells exhibited restricted permissiveness for HIV-1, but virus expression was markedly upregulated by previous or simultaneous infection with HCMV. HCMV expression was restricted to the IE and early genes in ST cells. However, previous HIV-1 infection converted HCMV infection from a nonpermissive to a permissive one. Interactions between HIV-1 and HCMV are complex and bidirectional. HIV replication can be enhanced by HCMV infection (17, 18, 49). However, HCMV may have the opposite effect and repress the HIV infection in tissue culture (25). Although the reasons for these apparent discrepancies remain to be elucidated, it seems likely that cell-specific factors, the relative permissiveness of the cell for the two viruses, and the production of cytokines can all affect the balance between stimulation or inhibition. In cells semipermissive for HCMV and/or HIV, an upregulation of HIV replication has been observed following coinfections with both viruses (17, 18, 49). Under conditions in which both HCMV and HIV-1 can undergo fully permissive infection, HCMV can repress HIV-1 gene expression (25). In cells that are permissive for the HIV but restricted for synthesis of the HCMV IE and early gene products, the presence of HCMV has no effect on HIV-1 gene expression or virus production (20). Our data are similar to those of Jault et al. (20), who, using neuroblastoma cell line SY5Y, which is slightly permissive for HIV-1 replication and fully permissive for HCMV replication, observed a transient stimulation of HIV-1 production between days 1 and 3 after HCMV infection. The initial stimulation was then followed by inhibition of HIV-1 production. However, there are some important differences: (i) we used ST cells semipermissive for HIV-1 replication and nonpermissive for HCMV replication; (ii) HIV enhancement in our study was induced by previous or simultaneous HCMV infection but not by superinfection with HCMV; and (iii) in our system, there was no simultaneous replication of both viruses. The time of coinfection may also influence the expression of coinfecting viruses. It has been shown that prior Epstein-Barr virus transformation of B lymphocytes increases the replication of HIV in these cells (38). On the other hand, HIV expression was markedly reduced when B cells were exposed to HIV either simultaneously with or prior to Epstein-Barr virus (16). Our data are in line with these observations except that in our system, simultaneous coinfection with HIV-1 and HCMV tilted the balance in favor of HIV-1 replication. The different results obtained from various experimental systems suggest that biological interaction between HIV and herpesviruses may be dependent on both cell-specific factors and the coinfecting herpesvirus. In vitro studies of HIV-1 and HCMV coinfections in various cell cultures have shown that the presence of the HIV-1 genome can change HCMV infection from an abortive to a productive one (49) or can support enhanced HCMV replication (18, 28). Our results add another cell type to those in which HCMV gene expression and replication can be augmented by coinfection with HIV-1.
Consistent with previous reports (5, 10), we also found that HCMV IEA expression is sufficient for HIV transactivation. Studies on the mechanism of transactivation of HCMV replication by HIV-1 led to controversial results. HIV-1 tat gene expression has been reported to enhance HCMV synthesis (18). On the contrary, in human osteosarcoma cells genetically engineered to contain an envelope-deficient HIV-1 proviral construct, HIV-1 tat gene expression did not play a major role in the stimulation of HCMV replication (28). In our system, we found that antibodies to Tat inhibited HCMV replication induced by coinfection with HIV-1. Ho et al. (18) found that the HIV-1 tat gene product augmented expression of the HCMV IE gene and that the tat gene product increased titers of infectious HCMV. By contrast, Skolnik et al. (49) reported that a Jurkat cell line constitutively expressing the HIV-1 tat gene did not exhibit increased levels of HCMV DNA, nor did the HIV-1 tat gene in a vaccinia virus recombinant construct increase detectable levels of particular HCMV early proteins in human embryonic lung cells. We observed a restriction in HCMV replication in ST cells after the expression of early genes. Since HCMV utilizes multiple stage-specific promoters (for a review, see reference 54), the productive HCMV infection in dually infected ST cells may be due to the functions of tat gene product on the late genes of HCMV. It will be of interest to study whether tat gene expression can transactivate HCMV early or late promoters independently. However, it cannot be excluded that the tat-induced HCMV replication is mediated by a cellular intermediate, rather than the result of a direct interaction between the HIV-1 gene product and HCMV. It is known that human trophoblast cells are capable of producing IL-1 (39), IL-2 (2), IL-6 (22), TNF-a and -b (24), IFN-a and -b (56), and GM-CSF (35). Of these, only production of IL-2 and IL-4 was induced by HCMV or HIV-1 in ST cells. Enhancement of low-level spontaneous IL-2 production by live HCMV or HIV-1 was not causally related to augmentation of HIV-1 replication or induction of a permissive replication cycle of HCMV in dually infected cell cultures. HIV-1 alone had no IL-6-inducing potential, whereas both live and heat-inactivated HCMV induced IL-6 production in ST cell cultures. The latter observation is in agreement with the results of Clouse et al. (4), who found that heat-inactivated HCMV was able to induce cytokine secretion in monocytes. In contrast to IL-2, IL-6 production induced by HCMV and further increased by replicating HIV-1 had a role in enhanced HIV-1 replication, because inhibition of IL-6 activity led to a fourfold decrease in HIV-1 release by coinfected cells. Thus, IL-6 may synergize with HCMV IEAs in the induction of HIV-1 expression. Cytomegalovirus IEAs transactivate HIV-1 by a transcriptional mechanism (58), whereas IL-6 acts alone posttranscriptionally (42). It remains unclear if IL-6 can increase viral transcription in synergy with HCMV IEAs as was described for IL-6 and TNF-a (42). HIV-1 and HCMV stock preparations were found to possess no IL-2 or IL-6 activities. These findings are consistent with previous reports. The long-term effect of HIV-1 infection on T cells involves marked inhibition of IL-2 production (43). Human T cells do not produce spontaneously IL-6 (57), and HIV-1 does not induce the expression of IL-6 mRNA and the secretion of IL-6 in them (32). Fibroblasts do not produce IL-2, but IL-6 secretion can be induced by various kinds of stimulation, including HCMV (53). However, HCMVinduced IL-6 production increases between 4 and 24 h after infection and then sharply decreases. Consequently, no IL-6 activity can be found in the supernatant fluid of cultures at the time of virus release. The absence of IL-2 and IL-6 in the virus stocks used for our studies indicated that external IL-2 and
VOL. 69, 1995
COINFECTION OF SYNCYTIOTROPHOBLASTS WITH HIV-1 AND HCMV
IL-6 effects were not involved in cooperation between HIV-1 and HCMV in dually infected ST cells. Since no IFN was released upon induction with HCMV or HIV-1, an endogenous IFN response seems to have no role in repression of HCMV replication and diminished HIV-1 production in ST cells. It is worthy of note that HIV-1 infection induces a high level of 29,59-oligoadenylate synthetase activity in human trophoblast cells without any IFN production (61). The functional activity of the 29,59-oligoadenylate synthetase system was demonstrated by the protection of HeLa CD41 cells against HIV infection (48). The mechanisms involved in the absence of simultaneous, reciprocal enhancement of HCMV and HIV-1 replication in ST cells are not clear. HCMV and HIV make use of cellular transcription factors for the control of viral gene expression (26, 46). This has important consequences, because competition for the same cellular protein(s) may determine whether a viral infection is productive or abortive in dually infected cells. Experiments are under way to characterize the nature of the interaction among cellular transcription factors and infectious cycles of HCMV and HIV-1 in ST cells. In our studies, we were unable to detect pseudotypes of HIV-1 and HCMV. Our results do not exclude the possibility of HIV-1 HCMV pseudotypes in other systems. In fact, pseudotypes with HIV-1 nucleocapsid within HCMV envelopes may be present in urine samples obtained from HIV-1seropositive HCMV excretors (27). Viral entry into fetal circulation of the placenta is of critical importance in the development of intrauterine HIV infection. Transport of HIV-1 across the ST barrier would be particularly efficient in delivery of the virus to its first target cells. As the villous macrophages, termed Hofbauer cells, possess CD4 receptors (13), they are likely candidates for infection by the virus. Similar to other phagocytic cells, Hofbauer cells are probably motile, and they occur in sufficient numbers potentially to permit the transfer of HIV to other cells. Endothelial cells, lining placental blood vessels, appeared to express CD4 (29) and may also become infected with the virus. We have shown that dual infection of ST cells with HCMV and HIV-1 can result in enhanced HIV release as well as in permissive HCMV infection, depending on the time of coinfection. Our data suggest that in vivo dual infection of ST cells with HIV-1 and HCMV could contribute to the transplacental transmission of both viruses and may result in the development of rapid HIV disease in congenitally coinfected children (36). ACKNOWLEDGMENTS This work was supported by funds from The Danish Cancer Society (Denmark) and by research grant 1483 from the National Science Foundation (Hungary). Rabbit antiserum to HIV-1 Tat from Bryan Cullen was obtained through the AIDS Research and Reference Reagent Program, Division of AIDS, NIAID, NIH, Rockville, Md. REFERENCES 1. Amirhessami-Aghili, N., R. Lahijani, P. Manalo, S. St. Jeor, F. D. Tibbitts, M. R. Hall, and A. Afsari. 1989. Persistence of human cytomegalovirus DNS sequences without cell-virus homology in human placental explants in culture. Int. J. Fertil. 34:411–419. 2. Boehm, K. D., M. F. Kelley, J. Ilan, and J. Ilan. 1989. The interleukin 2 gene is expressed in the syncytiotrophoblast of the human placenta. Proc. Natl. Acad. Sci. USA 86:656–660. 3. Chou, S. 1990. Differentiation of cytomegalovirus strains by restriction analysis of DNA sequences amplified from clinical specimens. J. Infect. Dis. 162:738–742. 4. Clouse, K. A., P. B. Robbins, B. Fernie, J. M. Ostrove, and A. S. Fauci. 1989. Viral antigen stimulation of the production of human monokines capable of regulating HIV-1 expression. J. Immunol. 143:470–475. 5. Davis, M. G., S. C. Kenney, J. Kamine, J. S. Pagano, and E. S. Huang. 1987.
6. 7. 8. 9. 10.
11. 12.
13. 14. 15. 16. 17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
2231
Immediate-early gene region of human cytomegalovirus trans-activates the promoter of human immunodeficiency virus. Proc. Natl. Acad. Sci. USA 84:8642–8646. DeMarchi, J. M., and A. S. Kaplan. 1976. Replication of human cytomegalovirus DNA: lack of dependence on cell DNA synthesis. J. Virol. 18:1063– 1070. Douglas, G. C., and B. F. King. 1989. Isolation of pure villous cytotrophoblast from term human placenta using immunomagnetic microspheres. J. Immunol. Methods 119:259–268. Douglas, G. C., and B. F. King. 1990. Differentiation of human trophoblast cells in vitro as revealed by immunocytochemical staining of desmoplakin and nuclei. J. Cell Sci. 96:131–141. Dudding, L., S. Haskill, B. D. Clark, P. E. Aurons, S. Sporn, and E. S. Huang. 1989. Cytomegalovirus infection stimulates expression of monocyteassociated mediator genes. J. Immunol. 143:3343–3352. Elfassi, E., S. Michelson, F. Bachelerie, F. Arenzana-Seisdedos, and J. L. Virelizier. 1987. Transactivation of human immunodeficiency virus long terminal repeat during the early phases of human cytomegalovirus infection. Ann. Inst. Pasteur/Virol. 138:461–470. Enders, A. C. 1989. Trophoblast differentiation during the transition from trophoblast plate to lacunar stage of implantation in the rhesus monkey and human. Am. J. Anat. 186:85–92. Gendelman, H. E., W. Phelps, L. Feigenbaum, J. M. Ostrove, A. Adachi, P. M. Howley, G. Khoury, H. S. Ginsberg, and M. A. Martin. 1986. Transactivation of the human immunodeficiency virus long terminal repeat sequence by DNA viruses. Proc. Natl. Acad. Sci. USA 83:9759–9763. Goldstein, J., M. Braverman, C. Salafia, and P. Buckley. 1988. The phenotype of human placental macrophages and its variation with gestational age. Am. J. Pathol. 133:648–659. Greenwood, F. C., W. M. Hunter, and J. S. Glover. 1963. The preparation of 131 I-labelled human growth hormone of high specific activity. Biochem. J. 89:114–123. Heath, T. D., D. Robertson, M. S. C. Birbeck, and A. J. S. Davies. 1980. Covalent attachment of horseradish peroxidase to the outer surface of liposomes. Biochim. Biophys. Acta 599:42–62. Henderson, E. E., J. Y. Yang, R. D. Zhang, and M. Bealer. 1991. Altered HIV expression and EBV-induced transformation in coinfected PBLs and PBL subpopulations. Virology 182:186–198. Hirka, G., K. Prakash, H. Kawashima, S. A. Plotkin, P. W. Andrews, and E. Go ¨nczo¨l. 1991. Differentiation of human embryonal carcinoma cells induces human immunodeficiency virus permissiveness which is stimulated by human cytomegalovirus coinfection. J. Virol. 65:2732–2735. Ho, W., J. M. Harrouse, R. F. Rando, E. Go¨nczo¨l, A. Srinivasan, and S. A. Plotkin. 1990. Reciprocal enhancement of gene expression and viral replication between human cytomegalovirus and human immunodeficiency virus type 1. J. Gen. Virol. 71:97–103. Hoffman, A. D., B. Banapour, and J. A. Levy. 1985. Characterization of the AIDS-associated retrovirus reverse transcriptase and optimal conditions for its detection in virions. Virology 147:326–335. Jault, F. M., S. A. Spector, and D. H. Spector. 1994. The effects of cytomegalovirus on human immunodeficiency virus replication in brain-derived cells correlate with permissiveness of the cells for each virus. J. Virol. 68:959–973. Jie, Z., S. I. Fey, H. Hager, P. Ho ¨llsberg, P. Ebbesen, and P. M. Larsen. 1990. Markers for human placental trophoblasts in two-dimensional gel electrophoresis. In Vitro 26:937–948. Kameda, T., N. Matsuzaki, K. Sawai, T. Okada, F. Saji, T. Matsuda, T. Hirano, T. Kishimoto, and O. Tanizawa. 1990. Production of interleukin-6 by normal human trophoblast. Placenta 11:205–213. Kliman, J. J., J. E. Nestler, E. Sermasi, J. M. Sanger, and J. F. Strauss. 1986. Purification, characterization and in vitro differentiation of cytotrophoblasts from human term placentae. Endocrinology 118:1567–1582. Kohchi, C., D. I. Mizuno, and G. I. Soma. 1991. Expression of tumor necrosis factor-a and -b transcripts in embryonal carcinoma and trophoblast cell lines: inflammation-like state as possible regulatory mechanism for ontogenesis. Eur. Cytokine Net. 2:245–255. Koval, V., C. Clark, M. Vaishnav, S. A. Spector, and D. H. Spector. 1991. Human cytomegalovirus inhibits human immunodeficiency virus replication in cells productively infected by both viruses. J. Virol. 65:6969–6978. Lu, Y., M. Stenzel, J. G. Sodroski, and W. A. Haseltine. 1989. Effects of long terminal repeat mutations on human immunodeficiency virus type 1 replication. J. Virol. 63:4115–4119. Margalith, M., R. T. D’Aquila, L. J. Bechtel, B. R. Smith, J. C. Kaplan, and M. S. Hirsch. 1992. Detection of HIV-1 DNA in fibroblast cultures infected with urine from HIV-1 seropositive cytomegalovirus (CMV) excretors, abstr. POB-3883, p. B 328. In Abstracts of the VIII International Conference on AIDS, vol. 2. Margalith, M., D. J. Medina, G. D. Hsiung, B. R. Smith, R. T. D’Aquila, J. C. Kaplan, L. Bechtel, M. Z. Wang, P. R. Skolnik, and M. S. Hirsch. 1993. Interactions between HIV-1 and cytomegalovirus in human osteosarcoma cells carrying both viruses. AIDS Res. Hum. Retroviruses 9:519–527. Maury, W., B. Potts, and A. B. Rabson. 1989. HIV-1 infection of firsttrimester and term placental tissue: a possible mode of maternal-fetal trans-
2232
´ TH ET AL. TO
mission. J. Infect. Dis. 160:583–588. 30. Mazeron, M. C., G. Jahn, and B. Plachter. 1992. Monoclonal antibody E-13 (M-810) to human cytomegalovirus recognizes an epitope encoded by exon 2 of the major immediate early gene. J. Gen. Virol. 73:2699–2703. 31. Moudgal, N. R., and R. R. Porter. 1963. The use of antigen-cellulose suspensions for the isolation of specific antibodies. Biochim. Biophys. Acta 71:185–187. 32. Nakajima, K., O. Martı´nez-Maza, T. Hirano, E. C. Breen, P. G. Nishanian, J. F. Salazar-Gonzalez, J. L. Fahey, and T. Kishimoto. 1989. Induction of IL-6 (B cell stimulatory factor-2/IFN-b2) production by HIV. J. Immunol. 142:531–536. 33. Nara, P. L., W. C. Hatch, N. M. Dunlop, W. G. Robey, L. O. Arthur, M. A. Gonda, and P. J. Fischinger. 1987. Simple, rapid, quantitative, syncytiumforming microassay for the detection of human immunodeficiency virus neutralizing antibody. AIDS Res. Hum. Retroviruses 3:283–302. 34. New, R. R. C., C. D. V. Black, R. J. Parker, A. Puri, and G. L. Scherphof. 1990. Liposomes in biological systems, p. 243–251. In R. R. C. New (ed.), Liposomes: a practical approach. IRL Press, Oxford. 35. Nicola, N. A., D. Metcalf, G. R. Johnson, and A. W. Burgess. 1979. Separation of functionally distinct human granulocyte-macrophage colony-stimulating factors. Blood 54:614–627. 36. Nigro, G., G. Castelli, and S. Mattra. 1991. Progression towards AIDS in children with congenital HIV-cytomegalovirus coinfection, abstr. MC-3208, p. 350. In Abstracts of the VII. International Conference on AIDS, vol. 1. 37. Ou, C. Y., S. Kwok, S. W. Mitchell, D. H. Mack, J. J. Sninsky, J. W. Krebs, P. Feorino, D. Warfield, and G. Schochetman. 1988. DNA amplification for direct detection of HIV-1 in DNA of peripheral blood mononuclear cells. Science 239:295–297. 38. Pagano, J. S., S. Kenney, D. Markovitz, and J. Kamine. 1988. Epstein-Barr virus and interactions with human retroviruses. J. Virol. Methods 21:229– 239. 39. Paulescu, L., A. King, Y. W. Loke, M. Cintorino, E. Bellizzi, and D. Boraschi. 1991. Immunohistochemical localization of IL-1a and IL-1b in normal human placenta. Lymphokine Cytokine Res. 10:443–448. 40. Peterson, P. K., G. Gekker, C. C. Chao, S. Hu, C. Edelman, H. H. Balfour, and J. Verhoef. 1992. Human cytomegalovirus-stimulated peripheral blood mononuclear cells induce HIV-1 replication via a tumor necrosis factor-amediated mechanism. J. Clin. Invest. 89:574–580. 41. Plaue´, S., and P. Briand. 1990. Solid-phase peptide synthesis, p. 74–84. In M. H. V. Van Regenmortel, J. P. Briand, S. Muller, and S. Plaue´ (ed.), Synthetic peptides as antigens. Elsevier, Amsterdam. 42. Poli, G., P. Bressler, A. Kinter, E. Duh, W. C. Timmer, A. Rabson, J. S. Justement, S. Stanley, and A. S. Fauci. 1990. Interleukin 6 induces human immunodeficiency virus expression in infected monocytic cells alone and in synergy with tumor necrosis factor a by transcriptional and post-transcriptional mechanisms. J. Exp. Med. 172:151–158. 43. Purvis, S. F., D. L. Georges, T. M. Williams, and M. M. Lederman. 1992. Suppression of interleukin-2 and interleukin-2 receptor expression in Jurkat cells stably expressing the human immunodeficiency virus Tat protein. Cell. Immunol. 144:32–42. 44. Saiki, R. K., T. L. Bugawan, G. T. Horn, K. B. Mullis, and H. A. Erlich. 1986. Analysis of enzimatically amplified b-globin and HLA-DQa DNA with allele-specific oligonucleotide probes. Nature (London) 324:163–166. 45. Saiki, R. K., D. H. Gelfand, S. Stoffel, S. J. Scharf, R. Higuchi, G. T. Horn, K. B. Mullis, and H. A. Erlich. 1988. Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239:487–491. 46. Sambucetti, L. C., J. M. Cherrington, G. W. G. Wilkinson, and E. S. Mocarski. 1989. NF-kB activation of the cytomegalovirus enhancer is mediated
J. VIROL. by a viral transactivator and by T cell stimulation. EMBO J. 8:4251–4258. 47. Schnittman, S. M., M. C. Psallidopoulos, H. C. Lane, L. Thompson, M. Baseler, F. Massari, C. H. Fox, N. P. Salzman, and A. S. Fauci. 1989. The reservoir for HIV-1 in human peripheral blood is a T cell that maintains expression of CD4. Science 245:305–308. 48. Schro¨der, H. C., D. Ugakovic, H. Merz, Y. Kuchino, T. Okamoto, and W. E. G. Mu ¨ller. 1990. Protection of HeLa-T41 cells against human immunodeficiency virus (HIV) infection after stable transfection with HIV LTR29,59-oligoadenylate synthetase hybrid gene. FASEB J. 4:3124–3130. 49. Skolnik, P. R., B. R. Kosloff, and M. S. Hirsch. 1988. Bidirectional interactions between human immunodeficiency virus type 1 and cytomegalovirus. J. Infect. Dis. 157:508–514. 50. Sprecher, S., G. Soumenkoff, F. Puissant, and M. Degueldre. 1986. Vertical transmission of HIV in 15-week fetus. Lancet ii:288–289. 51. Stagno, S., D. Reynolds, A. Tsiantos, D. A. Fuccillo, R. Smith, M. Tiller, and C. A. Alford. 1975. Cervical cytomegalovirus excretion in pregnant and nonpregnant women: suppression in early gestation. J. Infect. Dis. 131:522–527. 52. Stenberg, R. M., D. R. Thomsen, and M. F. Stinski. 1984. Structural analysis of the major immediate-early gene of human cytomegalovirus. J. Virol. 49:190–199. 53. St. Jeor, S., J. Admiraud, E. Bruening, and J. Riolo. 1993. Induction of cytokines by human cytomegalovirus, p. 123–126. In S. Michelson and S. A. Plotkin (ed.), Multidisciplinary approach to understanding cytomegalovirus disease. Elsevier, Amsterdam. 54. Stinski, M. F. 1990. Cytomegalovirus and its replication, p. 1959–1980. In B. N. Fields and D. M. Knipe (ed.), Virology. Raven Press, New York. 55. Thiry, L., S. Sprecher-Goldberger, T. Jonckheer, J. Levy, J. CogniauxLeclerc, and N. Clumek. 1985. Isolation of AIDS virus from cell free breast milk of three healthy virus carriers. Lancet ii:891–892. 56. To ´th, F. D., M. No¨rskov-Lauritsen, C. B. Juhl, G. Aboagye-Mathiesen, and P. Ebbesen. 1990. Interferon production by cultured human trophoblasts and choriocarcinoma cell lines induced by Sendai virus. J. Gen. Virol. 71:3067– 3069. 57. Villiger, P. M., M. T. Cronin, T. Amenomori, W. Wachsman, and M. Lotz. 1991. IL-6 production by human T lymphocytes. Expression in HTLV-infected but not in normal cells. J. Immunol. 146:550–559. 58. Walker, S., C. Hagemeier, J. G. P. Sissons, and J. H. Sinclair. 1992. A 10-base-pair element of the human immunodeficiency virus type 1 long terminal repeat (LTR) is an absolute requirement for transactivation by the human cytomegalovirus 72-kilodalton IE1 protein but can be compensated for by other LTR regions in transactivation by the 80-kilodalton IE2 protein. J. Virol. 66:1543–1550. 59. Zachar, V., N. No ¨rskov-Lauritsen, C. Juhl, B. Spire, J. C. Chermann, and P. Ebbesen. 1991. Susceptibility of cultured human trophoblasts to infection with human immunodeficiency virus type 1. J. Gen. Virol. 72:1253–1260. 60. Zachar, V., B. Spire, I. Hirsch, J. C. Chermann, and P. Ebbesen. 1991. Human transformed trophoblast-derived cells lacking CD4 receptor exhibit restricted permissiveness for human immunodeficiency virus type 1. J. Virol. 65:2102–2107. 61. Zhang, G. Y., B. Beltchev, A. Fournier, Y. H. Zhang, A. Malassine´, C. Bisbal, B. Ehresmann, C. Ehresmann, J. L. Darlix, and M. N. Thang. 1993. High levels of 29,59-oligoadenylate synthetase and 29,59-oligoadenylate-dependent endonuclease is human trophoblast. AIDS Res. Hum. Retroviruses 9:189– 196. 62. Ziegler, J. B., D. A. Cooper, R. O. Johnson, and J. Gold. 1985. Postnatal transmission of AIDS-associated retrovirus from mother to infant. Lancet i:896–898.
