VIROLOGY
239, 259±268 (1997) VY978895
ARTICLE NO.
Efficient Virus Transmission from Dendritic Cells to CD41 T Cells in Response to Antigen Depends on Close Contact through Adhesion Molecules Yasuko Tsunetsugu-Yokota,*,²,1 Sachiko Yasuda,*,² Akira Sugimoto,* Takenori Yagi,³ Miyuki Azuma,§ Hideo Yagita,¶ Kiyoko Akagawa,* and Toshitada Takemori*,² *Department of Immunology, and ²AIDS Research Center, National Institute of Infectious Diseases, Tokyo, Japan; ³Department of Chest Medicine, Institute of Pulmonary Cancer Research, School of Medicine, Chiba University, Chiba, Japan; §Allergy and Immunology, National Children's Medical Research Center, Tokyo, Japan; and ¶Department of Immunology, School of Medicine, Juntendo University, Tokyo, Japan Received March 26, 1997; returned to author for revision June 17, 1997; accepted October 9, 1997 Monocyte-derived cultured dendritic cells (DCs) are potent antigen-presenting cells (APCs) and are susceptible to HIV-1Lai infection. Compared to the low level of virus production by HIV-1-infected DCs alone, a level of virus two to three orders of magnitude higher was produced by cocultivation of HIV-1-infected DCs with autologous resting CD41 T cells in the presence of a nominal antigen. In this coculture system, direct contact of HIV-1-infected DCs with T cells was crucial for efficient virus transmission and subsequent virus production. Blocking of the LFA-1/ICAM-1 or LFA-3/CD2 interaction between these cells substantially reduced virus production, without influence on IL-2 production by activated T cells. In contrast, cell±cell transmission of HIV between non-APCs and activated T cells was not blocked by an antibody against LFA-3. Since a low level of virus production by HIV-infected DCs was upregulated by cross-linking of CD40, it was suggested that not only focal adhesion, but also mutual activation of HIV-infected DCs and T cells through adhesion molecules, may potentiate virus transmission and production and that such activation signals to HIV may be distinct from signals responsible for IL-2 production in activated T cells. © 1997 Academic Press
1992a,b), it may depend on the maturation stage and the level of CD4 expression of DCs (Weissman et al., 1995). We and others recently demonstrated the susceptibility of DCs or LCs to HIV-1 infection in vitro (Ludewig et al., 1995; Patterson et al., 1995; Tsunetsugu-Yokota et al., 1995) or in vivo (McIlroy et al., 1995). A more recent study on simian immunodeficiency virus (SIV) infection as the model of mucosal transmission showed that initial targets of SIV infection were likely to be DCs just beneath the epithelium and that infected cells migrated to lymph nodes within 2 days after virus inoculation (Spira et al., 1996). Therefore, it is highly likely that DCs, although a minor subpopulation, play a crucial role in the initiation and dissemination of HIV-1 infection during immune activation in vivo. We have demonstrated previously that virus production by HIV-infected DCs was barely detectable within the sensitivity of an ELISA system prepared in-house (Tsunetsugu-Yokota et al., 1995). However, under physiologic conditions of an antigen-specific immune response, a massive amount of virus was produced from activated T cells in the culture supernatant. This marked virus production was antigen-dependent; in the absence of PPD antigen, the level of virus production was 10 to 20 times less than that in the presence of the antigen. Furthermore, treatment of DCs with zidovudine during the process of HIV infection inhibited subsequent virus
INTRODUCTION The immunopathogenicity of human immunodeficiency virus type-1 (HIV-1) infection remains largely unknown. Although CD41 T cells are one of the targets of HIV infection, most of the CD41 T cells in the peripheral blood are resting and are latently infected with HIV-1 (Schnittman et al., 1989). Cellular activation is required for such T cells to support virus replication (Bukrinsky et al., 1991; Zack et al., 1990), which probably occurs in the lymphoid tissues in vivo (Embretson et al., 1993; Pantaleo et al., 1993). The lymphoid tissues are the places where DCs migrated from a local tissue would present acquired antigens to T cells and then activate T cells with an antigen-specific manner (Steinman, 1991). Therefore, it is possible that virus replication associated with immune activation of T cells are mediated mostly by DCs. Based on the submucosal localization of one type of DCs, Langerhans cells (LCs), it has been postulated that the initial targets for sexual transmission are LCs (Lehner et al., 1991). Although the susceptibility of DCs to HIV-1 infection has been questioned (Cameron et al.,
1 To whom correspondence and reprint requests should be addressed at Department of Immunology, AIDS Research Center, National Institute of Infectious Diseases, 1-23-1, Toyama-cho, Shinjuku-ku, Tokyo 162, Japan. Fax: (81)-3-5285-1150. E-mail:
[email protected].
259
0042-6822/21900 $25.00 Copyright © 1997 by Academic Press All rights of reproduction in any form reserved.
260
TSUNETSUGU-YOKOTA ET AL.
production by 64 to 98% (Tsunetsugu-Yokota et al., 1995). Therefore, it was strongly suggested that the virus enters into DCs, but is not simply held on their surface, is transmitted to T cells, and is produced mainly by activated T cells. By use of this in vitro model system, which mimics what is presumably occurring in the lymphoid tissues of HIV-1 infected individuals, the importance of direct contact of DCs and T cells and the role of adhesion molecules during DC±T cell interaction in efficient virus transmission and production of virus were investigated further. METHODS 1. Antibodies and reagents The following mouse monoclonal antibodies (mAbs) were used: anti-CD1a (OKT6; IgG1), anti-CD8 (OKT8; IgG2a), anti-CD14 (3C10; IgG2b), anti-CD21 (THB-5; IgG2a), anti-HLA-DR (L243; IgG2a), anti-LFA-1b (Ts1/18.1.2.11.4; IgG1), anti-CD58 (LFA-3, Ts2/9.1.4.3; IgG1), anti-CD11b (OKM-1; IgG2b), anti-CD54 (ICAM-1), anti-CD16 (3G8; IgG1), and anti-CD64 (32.2; IgG1) mAbs. Anti-CD4 mAb (7-4; IgG1) and anti-CD40 mAb were kindly provided by Dr. H. Nakauchi (Tsukuba University, Tsukuba, Japan) and Dr. H. Kikutani (Osaka University, Osaka, Japan), respectively. PE-conjugated mAbs against CD3, CD14, and CD20 were purchased from Becton Dickinson (Becton Dickinson and Co., Fujisawa, Japan). Purified protein derivative (PPD) from mycobacterium tuberculosis was obtained by courtesy of Drs. T. Kataoka and M. Kinomoto (Department of Bacterial and Blood Products, NIID Japan). 2. Cell preparation Peripheral blood mononuclear cells (PBMCs) from healthy donors were separated by Ficoll±Hypaque density gradient and enriched for CD141 cells with magnetic anti-CD14 beads and a magnetic cell sorter (MACS, Miltenyi Biotec, Cologne, Germany). After removal of plastic-non-adherent cells by extensive washing, CD141 cells were cultured with RPMI 1640 supplemented with 10% fetal bovine serum (FBS), 2 mM glutamine, and antibiotics in the presence of IL-4 (100 u/ml, Genzyme Inc., Cambridge, MA) and GM-CSF (500 u/ml, Kirin Breweries Inc., Tokyo) for 1 week as described previously (Tsunetsugu-Yokota et al., 1995). These monocyte-derived dendritic cells were stained with FITC-conjugated anti-CD1a and PE-conjugated anti-CD3, -CD14, and -CD20 antibodies (Becton Dickinson). Then cells with a high FSC/SSC ratio, positive for CD1a and negative for CD3, -14, and -20, were further purified by FACSvantage (Becton Dickinson). The monocyte-derived DCs express high levels of CD1a, HLA-DR, HLA-DQ, LFA-1, ICAM-1, LFA-3, and CD40 and low to moderate levels of FcgR II (CD32), B7-1, and B7-2 (Akagawa et al., 1996).
As autologous resting T cells, CD142 cells were kept frozen and enriched for CD41 T cells by negative selection on the day of the experiment as described previously (Tsunetsugu-Yokota et al., 1995). The purity of CD41 T cells was .96% as assessed by FACScan (Becton Dickinson). As a nonantigen presentinc cells (non-APCs) for cell-tocell virus transmission, HOS-CD4.Fusin (human osteosarcoma) cells expressing both CD4 and CXCR4 were used (No. 3319; obtained through AIDS Research and Reference Reagent Program, Division of AIDS, NIAID, NIH). 3. HIV-1Lai infection and cell culture For HIV-1Lai infection of DCs, 1 3 105 cpm of RT activity was added to 1 3 106 cells, incubated for 2 h at 37°C, and then washed thoroughly. These DCs (designated as HIV-infected DCs throughout this article) (1 3 105 cells) and autologous resting CD41 T cells (1 3 106 cells) in 1 ml per well were placed into 24-well culture plates. Because of the limited numbers of purified primary DCs or T cells, usually the individual culture was placed in one well. To see the reproducibility, all the experiments were repeated two to three times using cells from different donors. These cells were cultured in the presence or absence of IL-4 and GM-CSF and a nominal antigen, PPD, was added to the coculture at 50 mg/ml in order to highly activate CD41 T cells. Because most Japanese receive BCG vaccination early in their life, a vigorous proliferation of T cells by PPD in vitro was observed in the presence of DCs as APCs (data not shown). In order to study the efficiency of cell-to-cell transmission and virus production in HIV-infected DC and T-cell coculture, the level of virus production in this coculture system was compared with that in the coculture of uninfected DCs and T cells, which was inoculated with serially diluted cell-free HIV-1Lai stock. To block a direct interaction between HIV-infected DCs and T cells, we cultured HIV-infected DCs (0.5 3 105) in either the upper or the lower chamber of a transwell culture plate separated by a semipermeable membrane (Coster, Cambridge, MA), always in parallel with T cells (1 3 106) and uninfected DCs (0.5 3 105) in the presence of PPD antigen in the lower chamber. In order to inhibit the interaction of HIV-infected DCs and T cells or to stimulate HIV-infected DCs, we first incubated HIV-infected DCs with various mAbs (20 mg/ml) at room temperature for 20 min and then cocultured them with CD41 T cells in the presence of PPD. When half of the culture supernatant was replaced, mAbs were added every time to maintain the initial concentration. Likewise, HOS-CD4.Fusin cells were infected with HIV-1Lai, washed thoroughly and incubated with various mAbs. Purified CD41 T cells were stimulated for 1 day with phorbol myristeric acid (PMA)
HIV TRANSMISSION AND PRODUCTION BY DC
261
at 10 mM and ionomycin (IM) at 500 mM in RPMI 1640 supplemented with 10% FBS and IL-2 at 50 units per ml. These activated T cells were washed and cocultured with HIV-infected HOS-CD4.Fusin cells. Two days later, T cells were separated from HOSCD4.Fusin cells and cultured further in the presence of the same concentration of mAb. 4. Source and detection of HIV-1 The culture supernatant of CEM cells freshly infected with HIV-1Lai was collected on the day when reverse transcriptase (RT) activity was at its peak. The supernatant containing infectious HIV-1Lai was filtered through a 0.22-mm filter, aliquotted, and then stored at 285°C. The level of HIV-gag p24 antigen in the culture supernatant was measured by using either an in-house ELISA system (Tsunetsugu-Yokota et al., 1995) or a highly sensitive HIV-1 p24 ELISA kit from Coulter (Coulter Corp., Hialeah, FL). 5. IL-2 activity IL-2 produced by activated T cells was measured by use of an IL-2-dependent murine T-cell line (MTH), which was kindly provided by Dr. Hitoshi (Institute of Medical Science, University of Tokyo, Tokyo). In brief, 5 3 103 MTH cells in 50 ml were added to 50 ml of culture supernatant in a 96-well flat-bottom microtiter plate (Sumitomo Bakelite Co., Ltd., Fujisawa, Japan) in triplicate, incubated for 18 h, and then pulsed with 0.5 mCi of [3H]thymidine for 6 h. Cells were harvested onto ready filters (Beckman Instruments Inc., Somerset, NJ), and radioactivity was measured.
FIG. 1. Cross-linking of CD40 upregulated HIV-1 production in HIV1Lai-infected DCs and in coculture of HIV-infected DCs with resting CD41 T cells in the presence of PPD antigen. Highly purified DCs were infected with HIV-1Lai for 2 h at 37°C, washed thoroughly, and then cultured alone (1 3 105 cells) or cocultured with autologous resting CD41 T cells (1 3 106 cells) plus PPD antigen at 50 mg/ml, in the absence (blank columns) or in the presence of anti-CD40 (black columns), each at 20 mg/ml. After washing of DCs, p24 antigen was usually undetectable. The culture supernatant was collected, and the level of virus production was measured with an HIV-1 p24 ELISA kit from Coulter.
RESULTS Stimulation of CD40 potentiated HIV-1 production in HIV-infected DCs DCs were infected with HIV-1Lai (104cpm per 105 cells) and thoroughly washed to remove unadsorbed virus (cell-free virus in the medium was not detectable after washing). On day 4, a low level (less than 100 pg/ml) of virus production became detectable by use of the highly sensitive p24-gag ELISA kit (Coulter), and this increased slightly on day 7 after infection (Fig. 1). Cross-linking of CD40 expressed on DCs is known to upregulate the expression of costimulatory molecules and the secretion of various cytokines (Caux et al., 1994; Cella et al., 1996; McLellan et al., 1996). We therefore examined whether stimulation of DCs through CD40 activates viral replication in HIV-infected DCs. HIV-infected DCs were washed thoroughly, and then CD40 cross-linking was carried out with anti-CD40 mAb at 20 mg per milliliter. In a typical experiment, virus production in DCs was increased
two to three times by CD40 cross-linking at both day 4 and day 7. As we observed previously (TsunetsuguYokota et al., 1995), coculturing of DCs infected with HIV-1Lai with resting T cells in the presence of PPD antigen resulted in extremely high levels of gag-p24 in the culture supernatant, approximately two to three orders of magnitude above the level produced by DCs alone. The level was further upregulated three to five times by cross-linking of CD40. This result indicates that a stimulatory signal from CD40 may activate HIV-1 in DCs. Direct contact between HIV-infected DCs and T cells is required for efficient virus transmission and production To assess whether infection of T cells with cell-free HIV-1Lai could also cause remarkable virus production as observed in the coculture of HIV-infected DCs and T cells, we inoculated the serially diluted cell-free virus
262
TSUNETSUGU-YOKOTA ET AL.
FIG. 2. Infection of T cells by HIV-1Lai-infected DCs is more efficient than cell-free virus infection. (A) The kinetics of virus production in the coculture of HIV-infected DCs and CD41 T cells as described under Materials and Methods was compared with that in the coculture of uninfected DCs and T cells, which is inoculated with serially diluted cell-free HIV-1Lai stock. The level of HIV-gag p24 was determined at day 7. The representative result of three experiments with reproducibility was shown. (B) HIV-infected and washed DCs (0.5 3 105) were cultured in either the upper or the lower chamber of a transwell culture plate separated by a semipermeable membrane. In the lower chamber, T cells (1 3 106) and uninfected DCs (0.5 3 105) were always cultured in the presence of PPD antigen. For comparison, cell-free virus was inoculated into the lower chamber, where T cells and uninfected DCs were cultured in the presence of PPD antigen as above or into upper chamber separated from DC±T cell coculture by the membrane. The representative result of three experiments was shown. Because of the limitation of the number of purified cells available, each column usually represent the level of p24 antigen at day 7 in one well of 1 ml culture. The result of duplicate cultures were consistent as small error bars shown in the right two columns.
into a coculture composed of uninfected DCs and T cells in the presence of PPD antigen. As shown in Fig. 2, prominent virus production, comparable in magnitude and in kinetics to that produced in HIV-infected DCs and T cell coculture (closed circle), was detected by inoculation of cell-free virus with 1000 cpm of RT activity (open circle) into the coculture, but not by that of cell-free virus with 100 cpm of RT activity (open square). According to our standardization, 100 cpm of RT activity was equivalent to 100 pg of p24-gag antigen (data not shown). Since HIV-infected DCs (initially undetectable cell-free virus in the supernatant) can produce virus at most equivalent to 100 cpm of RT activity during culture for 7 days (see Fig. 1), we can assume that, even if 100 pg of virus is produced by HIV-infected DCs at the initiation of DC±T-cell coculture, this amount of cell-free virus infection would not result in as prominent a virus production as in the coculture of T cells and DCs infected with HIV-1 in the presence of antigen. To investigate further whether direct contact between HIV-infected DC and T cells is important for efficient virus transmission and production, we added HIV-infected DCs (5 3 104) either into an upper chamber of a transwell
plate separated by a 5-mm pore-size membrane from a lower chamber or into a lower chamber. In the lower chamber, 1 3 106 CD41 T cells and 0.5 3 105 uninfected DCs were always plated in the presence of PPD antigen. A representative result is shown in Fig. 2B. Inoculation of cell-free virus (1000 cpm) into either the upper or the lower chamber caused a comparable level of virus production in DC±T cell coculture plated in the lower chamber (Fig. 2B, left two columns), suggesting that cell-free virion is able to penetrate freely through the transmembrane. In contrast, when HIV-infected DCs were separated by a transmembrane from CD41 T cells activated by uninfected DCs, the level of virus production at day 7 after cultivation was reduced approximately 50 times below that in the culture of CD41 T cells in direct contact with HIV-infected DCs. The number of DCs as antigenpresenting cells at 0.5 3 105 (uninfected DCs only) or 1 3 105 (HIV-infected and uninfected DCs) did not affect the level of T-cell activation significantly (data not shown). Taken together, these results indicate that direct contact of HIV-infected DCs with T cells upon activation by antigen is crucial for efficient virus transmission and production.
HIV TRANSMISSION AND PRODUCTION BY DC
263
virus production by anti-CD4 mAb may also reflect the blocking effect of this antibody on HIV-1 entry via CD4. The blocking effect of anti-ICAM-1, anti-LFA-1, or antiLFA-3 suggests that DC±T-cell interaction through the ICAM-1/LFA-1 or LFA-3/CD2 pathways plays an important role in the initial virus transmission to T cells and, probably, in the subsequent activation of virus-infected T cells. Disparity of virus production and IL-2 production in DC±T cell coculture by blocking interaction of adhesion molecules FIG. 3. Blocking antibody against adhesion molecules inhibits virus production in coculture of HIV-1Lai-infected DCs and T cells stimulated with PPD antigen. DCs infected with HIV-1Lai and washed were first incubated with various mAbs (20 mg/ml) at room temperature for 20 min and then cocultured with CD41 T cells in the presence of PPD. Half of the culture supernatant was replaced every 2 days, and mAbs were added to maintain the initial concentration in each well. The level of virus production without mAb was considered to be 100%. The summary of two to three experiments at day 5 were shown. Each dot represents an individual experiment and the average of each dot was depicted as a column.
Interaction of DCs and T cells via adhesion molecules is important for efficient virus production Optimal T-cell activation requires not only the processed antigenic peptide on MHC molecules, but also costimulatory signals provided by antigen-presenting cells (reviewed in (Mueller et al., 1989)). The monocytederived DCs express a variety of cell adhesion molecules, including LFA-1 (CD11a/CD18), ICAM-1 (CD54), LFA-3 (CD58), B7-1 (CD80), B7-2 (CD86), and CD40 (Akagawa et al., 1996; Fagnoni et al., 1995; Sallusto and Lanzavecchia, 1994), which deliver costimulatory signals into T cells as well as mediating antigen-dependent or antigen-independent DC±T-cell conjugate formation (Hauss et al., 1995). Therefore, the roles of some of the adhesion molecules in virus transmission from HIV-infected DCs to T cells and subsequent massive virus production in activated T cells were evaluated. We first incubated HIV-1-infected DCs for 20 min with an excess amount of mAb (20 mg/ml) against CD4, HLADR, LFA-1, ICAM-1, LFA-3, or CD64 (FcgR I) as a control, and we cocultured them with T cells in the presence of PPD. Five days after culture in the continuous presence of each mAb, the level of p24 gag antigen in the culture supernatant was determined by ELISA. The results based on 2 to 3 experiments are summarized in Fig. 3. Virus production in the culture was significantly diminished by anti-CD4, anti-HLA-DR, anti-ICAM-1, anti-LFA-1, or anti-LFA-3 mAb, but not by anti-CD64 mAb. The blocking effect of anti-HLA-DR or anti-CD4 on virus production may reflect the inhibition of T-cell activation through PPD antigen presentation by DCs. The potent inhibition of
Because anti-LFA-1 or anti-LFA-3 mAbs showed a potent inhibitory effect on virus production, we investigated whether this inhibitory effect coincided with the inhibition of the T-cell activation process. To analyze the effect of these antibodies on both virus production and T-cell activation, we cocultured HIV-infected DCs and T cells in the absence of IL-4 and GM-CSF, which fully support the survival of monocyte-derived DCs in culture. In the absence of these cytokines, a relatively low, but still significant amount of virus was produced in the coculture of HIV-infected DCs and T cells upon stimulation with PPD (Tsunetsugu-Yokota et al., 1995). Virus production in the culture supernatant was estimated at day 7 postinfection, and IL-2 production was measured at day 1 as an early indicator of T-cell activation. Representative result of 2 to 3 experiments are shown in Fig. 4. Addition of either anti-LFA-1 or anti-LFA-3 mAb into the coculture diminished virus production substantially (Fig. 4A), but did not affect IL-2 production by activated T cells (Fig. 4B). Addition of two mAbs in combination abrogated the virus production almost entirely; this was accompanied by a prominent inhibition of IL-2 production by T cells. Such a synergistic effect by antiLFA-1 and anti-LFA-3 mAbs on T-cell activation is well known and indicates redundant or complementary functions of these adhesion molecules (Teunissen et al., 1994; Thomas et al., 1993). Large DC±T cell conjugates formed during coculture for 3 days were markedly inhibited only by the combination of anti-LFA-1 and anti-LFA-3 mAbs, although earlier small clusters containing a few cells were still visible (Fig. 4C). This observation indicates that the small DC±T-cell cluster formation is sufficient for virus transmission, but that the large conjugates formed later on may help to enhance T cell activation further, which results in rapid expansion of HIV. LFA-1 but not LFA-3 is required during transmission and expansion of HIV-1 in coculture of non-APC and activated T cells Whether interactions of LFA-1/ICAM-1 and LFA-3/CD2 are also important in other types of cell-to-cell transmission and expansion of HIV-1 was assessed. HOS-
264
TSUNETSUGU-YOKOTA ET AL.
FIG. 4. Blocking effect of anti-LFA-1 and anti-LFA-3 mAbs on virus production and on IL-2 production in HIV-infected DC±T-cell coculture stimulated with PPD. (A) HIV-infected DCs were treated with anti-LFA-1, anti-LFA-3, or in combination as described in the legend of Fig. 3, and cocultured with CD41 T cells in the presence of PPD. The culture supernatants at day 7 were collected, and the amount of p24 antigen was measured by ELISA. The level of virus production in the control culture without mAb was considered to be 100%. The representative result of two experiments with reproducibility was shown. (B) The level of IL-2 produced by activated T cells on day 1 in the culture supernatant of experiment shown in (A) was determined by use of an IL-2-dependent murine T-cell line, MTH. Proliferation of MTH cells was measured as described under Materials and Methods. The bar indicates the standard deviation of the [3H]thymidine uptake in triplicate cultures. (C) HIV-1Lai-infected DCs cocultured with CD41 T cells stimulated with PPD in the presence of anti-LFA-3 mAb (left) or anti-LFA-1 and anti-LFA-3 mAbs in combination (right) were observed under the microscope at 3 days after cultivation (310).
HIV TRANSMISSION AND PRODUCTION BY DC
CD4.Fusin cells were infected with HIV-1Lai, washed thoroughly, and cocultured with activated CD41 T cells. In this system, activation of CD41 T cells before coculture is necessary because, unlike DCs, osteosarcomaderived HOS-CD4.Fusin cells are not antigen-presenting cells. HOS-CD4.Fusin cells express LFA-3 but not LFA-1 or MHC class II (data not shown). After coculture for 2 days, CD41 T cells were separated from HIV-1-infected HOS-CD4.Fusin cells and cultured for a further 12 days with these antibodies continuously present. The level of virus production in the activated T cells was extremely low compared to T cells cocultured with HIV-infected DCs (peak value approximately 800 pg per ml). It was interesting to note that, as shown in Fig. 5A, only antiLFA-1 could block HIV-1 production by a substantial level despite the fact that HOS-CD4.Fusin cells do not express LFA-1. Anti-LFA-1 mAb also inhibited large conjugates of activated T cells while anti-LFA-3 mAb did not (Fig. 5B). A similar result was obtained using HeLa-CD4 as HIVinfected non-APC (data not shown). These data indicate that LFA-1 is not important for transmission from HOSCD4.Fusin cells to activated T cells, but for expansion of the virus among activated T cells. DISCUSSION When HIV-infected DCs contact resting T cells upon antigen stimulation in vitro, which presumably happens frequently in lymphoid tissue, HIV-1 would be transmitted from DCs to T cells, resulting in massive virus expansion (Tsunetsugu-Yokota et al., 1995). A similar scenario was also described by Weissman et al., (1996). Our previous study indicated that the virus entering into DCs, but not simply held on their surface, is largely responsible for efficient virus transmission to T cells and massive virus production by activated T cells (Tsunetsugu-Yokota et al., 1995). We showed in the current study that a small amount of virus production was detectable by a highly sensitive ELISA system, and that stimulation via CD40 on HIV-infected DCs increased the level of virus production approximately three times above the level of that without appropriate stimulation, probably through activation of NF-kB binding factor in DCs as described in crosslinking of CD40 on B cells (Lalmanach-Girard et al., 1993). In fact, we observed increased activity of DNA binding factor on the NF-kB site of HIV-1 LTR through CD40 cross-linking of DCs (Y. Tsunetsugu-Yokota, unpublished observation). This result not only supports our previous finding that HIV-1 does enter into DCs, but also indicates that appropriate activation signals to DC through interaction with T cells may enhance virus production in HIV-infected DCs. In our system, stimulation of CD40 during DC±T-cell coculture further increased the level of virus production. Because CD40L is expressed when T cells are activated (Armitage et al., 1992; Noelle
265
et al., 1992), it is possible that CD40/CD40L interaction plays some role in virus production during expansion of HIV in activated T cells as described by others (Pinchuk et al., 1994). However, further study is necessary for clarification of the precise role of CD40/CD40L interaction in virus production in our DC±T cell coculture system. It has been reported that exogenously added HIV-1 was expanded during conjugate formation of primary LCs and memory T cells, where virions were confined to heterologous syncytia induced by the virus (Pope et al., 1994). Their results suggested that cell fusion between HIV-infected LCs and T cells is an important factor for efficient virus production. In the early phase of HIV-1 infection in our culture system, at days 3 or 5 after cultivation, when numerous buddings of HIV-1 virions from activated T cells were already visible, we observed a tightly attached membrane structure of DCs and T cells, but no obvious cell fusion in the DC±T-cell cluster by thin-section electron microscopy (Tsunetsugu-Yokota, unpublished observation). Therefore, cell fusion is not likely to be essential for efficient virus transmission and production in the early phase of HIV-1 infection described here. A close membrane association of DCs and T cells may induce cytoskeleton-mediated focal virus budding, as observed in a model system of cell-to-cell infection (Pearce-Pratt et al., 1994), and such focal budding may help to preserve the stability of the virions (Dimitrov et al., 1993). Upon antigen recognition of T cells in the context of MHC molecules of APC, adhesion is strengthened further by ICAM-1/LFA-1 and LFA-3/CD2 interactions, whereas the interaction between B7 and CD28 is thought to be critical during a later stage of T-cell activation and for sustaining the proliferation of antigen-primed T cells (Damle et al., 1992). Furthermore, it is known that the expression of the IL-2 gene is not tightly associated with the LFA-3/CD2 pathway, but is regulated exclusively by a signal from B7/CD28 (Parra et al., 1994). Therefore, our results showing that the inhibition of ICAM-1/LFA-1 or LFA-3/CD2 interaction reduced virus production without affecting IL-2 production is consistent with their findings. Using PBMC derived from an HIV-infected donor, Diegel et al., reported that the inhibition of ICAM-1/LFA-1 or LFA-3/CD2 interaction resulted in a significant reduction of HIV production in a culture stimulated with superantigen or anti-CD3 mAb, without affecting T-cell proliferation (Diegel et al., 1993). Because B7/CD28 interaction between DCs and T cells is crucial for sustaining T-cell proliferation by producing IL-2, it is predictable that blocking of the B7/CD28 interaction may affect virus production in association with the inhibition of IL-2 production. In fact, Diegel et al., (1993) observed that exogenous addition of IL-2 abrogated the inhibitory effect on virus production by anti-B7 or anti-CD28 mAbs.
266
TSUNETSUGU-YOKOTA ET AL.
FIG. 5. Blocking effect of anti-LFA-1 and anti-LFA-3 mAbs on virus production in the coculture of HIV-infected HOS-CD4.Fusin and activated T cells. (A) HOS-CD4.Fusin cells were infected with HIV-1Lai, washed thoroughly, and untreated or treated with mAbs against LFA-1b, LFA-3, LFA-1b plus LFA-3, or CD64 at 20 mg/ml each. Purified CD41 T cells were activated by PMA and IM for 1 day, washed, and cocultured with HIV-infected HOS-CD4.Fusin cells. Two days after the coculture, T cells were separated from HOS-CD4.Fusin cells and cultured further in the presence of each mAb. The level of p24 antigen was determined at day 14 and expressed as percentage production as described in the legend of Fig. 4A. The representative result of two experiments were shown. (B) The effect of mAbs against these adhesion molecules on activated T-cell conjugates. Purified CD41 T cells were activated by PMA and IM in the presence of mAbs as in (A). The culture was observed at day 3 under the microscope, and photographed (320). The typical activated T-cell conjugates were observed in the presence of anti-LFA-3 or anti-CD64 mAb (left). In contrast the presence of anti-LFA-1b mAb markedly reduced the size of T-cell conjugates (middle and right).
HIV TRANSMISSION AND PRODUCTION BY DC
We demonstrated in this study that direct contact between HIV-infected DCs and T cells upon antigen stimulation was crucial for efficient virus transmission and prominent virus production during T-cell activation in vitro. Interaction through costimulatory molecules such as ICAM-1/LFA-1 or LFA-3/CD2 could be responsible for efficient virus transmission by enforcing focal adhesion of DCs and T cells. Furthermore, the finding that activation of DCs by cross-linking of CD40 molecules was able to increase virus production of DCs also suggested that mutual activation of HIV-infected DCs and T cells through ICAM-1/LFA-1 or LFA-3/CD2 interaction might play an important role in this enhanced virus production. Transmission and expansion of HIV-1 mediated by DC±T cell interaction is unique in the following aspects: (1) virus activation is coupled to T cell activation, (2) high and efficient virus production compared to virus transmission from non-APC to activated T cells, (3) DC±T cell adhesion especially through LFA-3/CD2 interaction may be important for virus transmission. Because mAb against LFA-1 inhibited virus production in activated T cells coculture with HIV-1-infected HOS-CD4.Fusin or HeLa-CD4 cells, both of which do not express LFA-1, it is highly likely that LFA-1 plays a role mainly for expansion of the virus among T cells. Thus, both focal adhesion and subsequent mutual activation signal(s) delivered through LFA-1/ICAM and LFA-3/CD2 interactions could be crucial for efficient virus transmission and maximum virus production during direct interaction between HIV-infected DCs and T cells. The exact nature of activation signal(s) for HIV-1 through LFA-1/ICAM and LFA-3/CD2 interactions between DCs and T cells, which is not necessarily accompanied by a signal for T cells to induce IL-2 production, remains to be elucidated. ACKNOWLEDGMENTS We are grateful to Dr. Hiroshi Yoshikura for his helpful review of the manuscript. We also thank Miss Mari Takizawa and Mrs. Machiko Akahane for their excellent technical help. This work is supported in part by grants from the Ministry of Public Health and Welfare, the Ministry of Education and Science, and the Science and Technology Agency of the government of Japan.
REFERENCES Akagawa, K., Takasuka, N., Nozaki, Y., Komuro, I., Azuma, M., Ueda, M., Naito, M., and Takahashi, K. (1996). Generation of CD11RelB1 dendritic cells and TRAP-positive osteoclast-like multi-nucleated giant cells from human monocytes. Blood 88, 4029±4039. Armitage, R. J., Fanslow, W. C., Strockbine, L., Sato, T. A., Clifford, K. N., Macduff, B. M., Anderson, D. M., Gimpel, S. D., Davis-Smith, T., Maliszewski, C. R., Clark, E. A., Simth, C. A., Grabstein, K. H., Cosman, D., and Spriggs, M. K. (1992). Molecular and biological characterization of a murine ligand for CD40. Nature 357, 80±82. Bukrinsky, M. I., Stanwick, T. L., Dempsey, M. P., and Stevenson, M. (1991). Quiescent T lymphocytes as an inducible virus reservoir in HIV-1 infection. Science 254, 423±427.
267
Cameron, P. U., Forsum, U., Teppler, H., Granelli-Piperno, A., and Steinman, R. M. (1992a). During HIV-1 infection most blood dendritic cells are not productively infected and can induce allogeneic CD41 T cells clonal expansion. Clin. Exp. Immunol. 88, 226±236. Cameron, P. U., Freudenthal, P. S., Barker, J. M., Gezelter, S., Inaba, K., and Steinman, R. M. (1992b). Dendritic cells exposed to human immunodeficiency virus type-1 transmit a vigorous cytopathic infection to CD41 T cells. Science 257, 383±387. Caux, C., Massacrier, C., Vanbervliet, B., Dubois, B., Van Kooten, C., Durand, I., and Banchereau, J. (1994). Activation of human dendritic cells through CD40 cross-linking. J. Exp. Med. 180, 1263±1272. Cella, M., Scheidegger, D., Palmer-Lehmann, K., Lane, P., Lanzavecchia, A., and Alber, G. (1996). Ligation of CD40 on dendritic cells triggers production of high levels of interleukin-12 and enhances T cell stimulatory capacity: T±T help via APC activation. J. Exp. Med. 184, 747±752. Damle, N. K., Klussman, K., Linsley, P. S., and Aruffo, A. (1992). Differential costimulatory effects of adhesion molecules B7, ICAM-1, LFA-3, and VCAM-1 on resting and antigen-primed CD41 T lymphocytes. J. Immunol. 148, 1985±1992. Diegel, M. L., Moran, P. A., Gilliland, L. K., Damle, N. K., Hayden, M. S., Zarling, J. M., and Ledbetter, J. A. (1993). Regulation of HIV production by blood mononuclear cells from HIV-infected donors: II. HIV-1 production depends on T cell-monocyte interaction. AIDS Res. Hum. Retroviruses 9, 465±473. Dimitrov, D. S., Willey, R. L., Sato, H., Chang, L.-J., Blumenthal, R., and Martin, M. A. (1993). Quantitation of human immunodeficiency virus type 1 infection kinetics. J. Virol. 67, 2182±2190. Embretson, J., Zupancic, M., Ribas, J. L., Burke, A., Racz, P., TennerRacz, K., and Haase, A. T. (1993). Massive covert infection of helper T lymphocytes and macrophages by HIV during the incubation period of AIDS. Nature 362, 359±362. Fagnoni, F. F., Takamizawa, M., Godfrey, W. R., Rivas, A., Azuma, M., Okumura, K., and Engleman, E. G. (1995). Role of B70/B7-2 in CD41 T-cell immune responses induced by dendritic cells. Immunology 85, 467±474. Hauss, P., Selz, F., Cavazzana-Calvo, M., and Fischer, A. (1995). Characteristics of antigen-independent and antigen-dependent interaction of dendritic cells with CD41 T cells. Adv. Exp. Med. Biol. 378, 409±411. Lalmanach-Girard, A.-C., Chiles, T. C., Parker, D. C., and Rothstein, T. L. (1993). T cell-dependent induction of NF-kB in B cells. J. Exp. Med. 177, 1215±1219. Lehner, T., Hussain, L., Wilson, J., and Chapman, M. (1991). Mucosal transmission of HIV. Nature 353, 709. Ludewig, B., Holzmeister, J., Gentile, M., Gelderblom, H. R., Rokos, K., Becker, Y., and Pauli, G. (1995). Replication pattern of human immunodeficiency virus type 1 in mature Langerhans cells. J. Gen. Virol. 76, 1317±1325. McIlroy, D., Autran, B., Cheynier, R., Wain-Hobson, S., Clauvel, J. P., Oksenhendler, E., Debre, P., and Hosmalin, A. (1995). Infection frequency of dendritic cells and CD41 T lymphocytes in spleens of human immunodeficiency virus-positive patients. J. Virol. 69, 4737± 4745. McLellan, A. D., Sorg, R. V., Williams, L. A., and Hart, D. N. (1996). Human dendritic cells activate T lymphocytes via a CD40:CD40 ligand-dependent pathway. Eur. J. Immunol. 26, 1204±1210. Mueller, D. L., Jenkins, M. K., and Schwartz, R. H. (1989). Clonal expansion versus functional clonal inactivation: A costimulaotry signaling pathway determines the outcome of T cell antigen receptor occupancy. Annu. Rev. Immunol. 7, 445±480. Noelle, R. J., Roy, M., Shepherd, D. M., Stamenkovic, I., Ledbetter, J. A., and Aruffo, A. (1992). A 39-KDa protein on activated helper T cells binds CD40 and transduces the signal for cognate activation of B cells. Proc. Natl. Acad. Sci. USA 89, 6550±6554.
268
TSUNETSUGU-YOKOTA ET AL.
Pantaleo, G., Graziosi, C., Demarest, J. F., Butini, L., Montroni, M., Fox, C. H., Orenstein, J. M., Kotler, D. P., and Fauci, A. S. (1993). HIV infection is active and progressive in lymphoid tissue during the clinically latent stage of disease. Nature 362, 355±358. Parra, E., Wingren, A. G., Hedlund, G., Bjorklund, M., Sjogren, H. O., Kalland, T., Sansom, D., and Dohlsten, M. (1994). Costimulation of human CD41 T lymphocytes with B7 and lymphocyte function-associated antigen-3 results in distinct cell activation profiles. J. Immunol. 153, 2479±2487. Patterson, S., Gross, J., English, N., Stackpoole, A., Bedford, P., and Knight, S. C. (1995). CD4 expression on dendritic cells and their infection by human immunodeficiency virus. J. Gen. Virol. 76, 1155± 1163. Pearce-Pratt, R., Malamud, D., and Phillips, D. M. (1994). Role of the cytoskeleton in cell-to-cell transmission of human immunodeficiency virus. J. Virol. 68, 2898±2905. Pinchuk, L. M., Polacino, P. S., Agy, M. B., Klaus, S. J., and Clark, E. A. (1994). The role of CD40 and CD80 accessory cell molecules in dendritic cell-dependent HIV-1 infection. Immunity 1, 317±325. Pope, M., Betjes, M. G. H., Romani, N., Hirmand, H., Cameron, P. U., Hoffman, L., Gezelter, S., Schuler, G., and Steinman, R. M. (1994). Conjugates of dendritic cells and memory T lymphocytes from skin facilitate productive infection with HIV-1. Cell 89, 389±398. Sallusto, F., and Lanzavecchia, A. (1994). Efficient presentation of soluble antigen by cultured human dendritic cells is maintained by granulocyte/macrophage-colony-stimulating factor plus interleukin 4 and downregulated by tumor necrosis factor a. J. Exp. Med. 179, 1109±1118. Schnittman, S. M., Psallidopoulos, M. C., Lane, H. C., Thompson, L., Baseler, M., Massari, F., Fox, C. H., Salzman, N. P., and Fauci, A. S. (1989). The reservoir for HIV-1 in human peripheral blood is a T cell that maintains expression of CD4. Science 245, 305±308.
Spira, A. I., Marx, P. A., Patterson, B. K., Mahoney, J., Koup, R. A., Wolinsky, S. M., and Ho, D. D. (1996). Cellular targets of infection and route of viral dissemination after an intravaginal inoculation of simian immunodeficiency virus into rhesus macaques. J. Exp. Med. 183, 215±225. Steinman, R. M. (1991). The dendritic cell system and its role in immunogenicity. Annu. Rev. Immunol. 9, 271±296. Teunissen, M. B. M., Rongen, H. A. H., and Bos, J. D. (1994). Function of adhesion molecules lymphocyte function-associated antigen 3 and intercellular adhesion molecule-1 on human epidermal Langerhans cells in antigen-specific T cell activation. J. Immunol. 152, 3400±3409. Thomas, R., Davis, L. S., and Lipsky, P. E. (1993). Comparative accesory cell function of human peripheral blood dendritic cells and monocytes. J. Immunol. 151, 6840±6852. Tsunetsugu-Yokota, Y., Akagawa, K., Kimoto, H., Suzuki, K., Iwasaki, M., Yasuda, S., HaÈusser, G., Hultgren, C., Meyerhans, A., and Takemori, T. (1995). Monocyte-derived cultured dendritic cells are susceptible to human immunodeficiency virus infection and transmit virus to resting T cells in the process of nominal antigen presentation. J. Virol. 69, 4544±4547. Weissman, D., Barker, T. D., and Fauci, A. S. (1996). The efficiency of acute infection of CD41 T cells is markedly enhanced in the setting of antigen-specific immune activation. J. Exp. Med. 183, 687±692. Weissman, D., Li, Y., Ananworanich, J., Zhou, L. J., Adelsberger, J., Tedder, T. F., Baseler, M., and Fauci, A. S. (1995). Three populations of cells with dendritic morphology exist in peripheral blood, only one of which is infectable with human immunodeficiency virus type 1. Proc. Natl. Acad. Sci. USA 92, 826±830. Zack, J. A., Arrigo, S. J., Weitsman, S. R., Go, A. S., Haislip, A., and Chen, I. S. Y. (1990). HIV-1 entry into quiescent primary lymphocytes: Molecular analysis reveals a labile, latent viral structure. Cell 61, 213± 222.
