Previous Article | Next Article ![]()
Journal of Virology, March 2002, p. 3007-3014, Vol. 76, No. 6
0022-538X/02/$04.00+0 DOI: 10.1128/JVI.76.6.3007-3014.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.
Department of Infectious Diseases and Microbiology, Graduate School of Public Health,1 Departments of Pathology,2 Cell Biology and Physiology, School of Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania 15261 ,3 Immunex Corporation, Seattle, Washington 981014
Received 11 July 2001/ Accepted 29 November 2001
|
|
|---|
|
|
|---|
Dendritic cells (DC) are the most potent antigen-presenting cells for the induction of antiviral T-cell responses through their expression of high levels of major histocompatibility complex (MHC) class I and II molecules and costimulatory molecules, such as CD40, CD80, and CD86, and the production of immunomodulating cytokines such as interleukin-12 (IL-12) and IL-15 (6). Current evidence indicates that immature DC (iDC) are highly efficient at capturing and processing antigens (6). Subsequent maturation of the iDC by ligation of CD40 with CD40 ligand (CD40L or CD154) expressed on CD4+ T cells upregulates MHC class I and II molecules and costimulatory molecules, greatly enhancing the presentation of antigen to T cells by these mature DC (mDC) (47). In the classic endogenous pathway, proteins produced during viral replication in the antigen-presenting cells are proteolytically cleaved in the cytosol (34). The resulting peptides are transported to the endoplasmic reticulum, where they complex with MHC class I molecules and then travel through the Golgi to the cell surface. In the exogenous pathway, viral proteins are ingested from the extracytosolic space into endosomal vesicles. There, the proteins are digested and the viral peptides are complexed with MHC class II molecules before transport to the cell membrane.
Some viruses do not replicate efficiently in DC, suggesting that there are alternative mechanisms to the conventional, endogenous MHC class I pathway for the induction of CD8+ T-cell responses to these viral antigens (29, 50). This has been related to uptake by DC of exogenous antigen in the form of virus-infected, apoptotic, or necrotic cells, followed by processing through nonconventional pathways and cross-presentation of antigen in the context of MHC class I molecules to CD8+ T cells (2, 3, 20, 21, 29, 45, 46). In HIV-1 infection, iDC do not support efficient replication of X4 strains due to low expression of the CXCR4 coreceptor, whereas they express higher levels of CCR5 and more efficiently support R5 virus replication (11, 17, 29). Thus, induction of anti-HIV-1 CD8+ T-cell responses to X4 virus may at least in part be due to uptake of X4 antigens by iDC and cross-presentation by HLA class I molecules on mDC. These viral antigens could be derived from cells that have been productively infected by X4 strains and have undergone apoptosis (4). A similar process of uptake of exogenous, nonreplicating viral antigens by iDC, with processing through the HLA class II pathway, presumably leads to induction of anti-HIV-1 CD4+ T-cell responses.
We therefore studied activation of anti-HIV-1 T-cell responses in peripheral blood mononuclear cells (PBMC) of persons with chronic HIV-1 infection by autologous DC loaded in vitro with various forms of HIV-1 X4 antigen. The study subjects for the T-cell immunity experiments were eight HIV-1-seropositive homosexual men from the Pittsburgh, Pa., portion of the Multicenter AIDS Cohort Study who were chronically infected with HIV-1 (Table 1). One of these subjects (S5) was not receiving antiretroviral therapy, while the seven other subjects were being treated with combination antiretroviral drug therapy of a protease inhibitor and two reverse transcriptase inhibitors. An additional HIV-1 chronically infected person on combination drug therapy was used for the DC phenotyping studies. Each subject gave written, informed consent approved by the University of Pittsburgh Institutional Review Board. Viral loads in plasma ranged from undetectable (<50 copies/ml) to >30,000 copies per/ml at the time of this study, and CD4+ T-cell counts ranged from 339 to 1,215 per ml of blood. Seven healthy HIV-1-negative persons served as controls.
|
View this table: [in a new window] |
TABLE 1. IFN- response of PBMC stimulated with autologous DC loaded with HIV-1 antigen
|
)-producing cells (data not shown). Thus, we used CD8- cell preparations in subsequent studies of anti-HIV-1 T-cell responses. The CD8- cells were stimulated with anti-CD3 monoclonal antibody (MAb) (200 ng of OKT-3/ml; Ortho, Raritan, N.J.) and human recombinant IL- 2 (rIL-2; 100 U/ml; Chiron, Emeryville, Calif.) for 4 to 5 days in RPMI 1640 medium supplemented with 20% heat-inactivated fetal calf serum and antibiotics (Gibco, Grand Island, N.Y.). The CD8- cells were then treated with Polybrene (5 µg per 106 cells; Sigma Biosciences, St. Louis, Mo.) for 1 h, pelleted by centrifugation at 250 x g for 10 min, and superinfected with HIV-1 IIIB (5) for 2 h at 37°C (50 to 100 ng of p24 per 106 cells). The superinfected cells were washed and adjusted to 106 cells/ml and cultured in complete RPMI 1640 medium supplemented with rIL-2 (100 U/ml) for another 4 to 5 days. Infection was monitored by production of HIV-1 p24 antigen in culture supernatant by a commercial p24 enzyme immunoassay (NEN Life Sciences, Boston, Mass.). Levels of p24 in the CD8- cells from the HIV-1-infected subjects ranged from 0.1 to 540 ng/ml for cells not superinfected with HIV-1 IIIB compared to 6 to >1,160 ng/ml for cells superinfected with HIV-1 IIIB (Table 1). To inactivate HIV-1 and induce apoptosis, the CD8- cells were treated with psoralen [10 µg/ml; 4'-(aminomethyl)-4,5',8-trimethylpsoralen hydrochloride; Calbiochem, La Jolla, Calif.] and UV light irradiation (312-nm wavelength calibrated to provide 0.86 mJ/cm2/s for 6 min; Derma Control, Fisher Scientific, Pittsburgh, Pa.) (referred to here as psoralen and UV irradiation [PUV] treatment). The cells were washed three times and cultured for an additional 6 h to allow time for apoptosis. Apoptosis was confirmed by flow cytometry (EPICS XL; Beckman Coulter, Fullerton, Calif.) by using the TUNEL (terminal deoxynucleotidyltransferase-mediated dUTP-biotin nick end labeling) assay (In Situ Cell Death Detection Kit; Boehringer Mannheim, Indianapolis, Ind.). Without PUV treatment, ca. 10% of the CD8- cell preparations not superinfected with HIV-1 IIIB were apoptotic, compared to 25 to 30% apoptotic cells in CD8-cell preparations that were superinfected with HIV-1 IIIB. Apoptosis in both of these cell preparations increased to 70% after PUV treatment (termed AP and AP-IIIB, respectively). Necrotic preparations of HIV-1 IIIB-superinfected (NE-IIIB) and nonsuperinfected (NE) CD8- cells were made by three or four rapid freeze-thaw cycles of the cells in dry ice-ethanol and 37°C water baths.
DC were derived from autologous CD14+ cells that were enriched by positive selection with anti-CD14 MAb-coated immunomagnetic beads (MACS Isolation Kit; Miltenyi Biotec, Auburn, Calif.). The purity of CD14+ cells was >95%, based on staining with MAb to CD14 (BD) and analysis by flow cytometry. The CD14+ cells were cultured in 1,000 U of granulocyte-monocyte colony-stimulating factor/ml and 1,000 U of IL-4 (Schering-Plough, Kenilworth, N.J.)/ml in complete RPMI 1640 medium for 4 to 5 days to obtain iDC as previously described (14). To induce maturation, the iDC were treated with 1.0 to 2.5 µg of trimeric CD40L (Immunex, Seattle, Wash.)/ml at day 4 or 5 for two additional days at 37°C in the cytokine-supplemented medium.
To determine uptake of the apoptotic and necrotic cell preparations by iDC or mDC, HIV-1 IIIB-infected and uninfected, apoptotic and necrotic CD8- cells were stained with membrane fluorophore PKH 26 as per the manufacturer's instructions (Sigma). They were then cocultured for 12 h at 37°C at a ratio of 2:1 with iDC or mDC that were stained with HLA-DR-fluorescein isothiocyanate (FITC). Flow cytometric analysis was performed by gating on large cells with high side scatter, and double-positive cells were enumerated. As shown in Fig. 1A, the uptake of both the CD8- apoptotic and necrotic cell preparations by iDC was much greater than by mDC (38% versus 6% for apoptotic cells and 23% versus 0.2% for necrotic cells). Furthermore, 30 to 50% of the iDC contained intracellular CD8- cells and cell debris as visualized by confocal laser scanning microscopy (Leica Lasertechnik, Heidelberg, Germany) (Fig. 1B). Therefore, for the T-cell stimulation experiments, the apoptotic and necrotic CD8- cell preparations were added to iDC at a 2-to-1 ratio (cell equivalent portions were used for the necrotic cell preparations) on day 4 or 5, incubated for 12 h, and then treated with either medium (to maintain iDC) or CD40 ligand (CD40L) (to induce maturation) for 2 additional days before use in the T-cell studies. Additionally, iDC were infected on day 4 or 5 with 300 ng of p24 per 106 cells of replication-competent, cell-free HIV-1 IIIB or PUV-treated HIV-1 IIIB (prepared by PUV treatment as with the CD8- cells and three sequential washes in medium for 1 h each at 22,000 x g at 4°C) for 2 h at 37°C and washed twice with medium to remove unadsorbed virus. These HIV-1 IIIB-infected iDC were then cultured with medium alone or CD40L, as stated above, for the T-cell activation experiments.
![]() ![]() View larger version (158K): [in a new window] |
FIG. 1. (A) Uptake of apoptotic and necrotic, non-HIV-1 IIIB-infected, CD8- cells (subject N7) by iDC and mDC, as demonstrated by flow cytometry. Similar results were obtained for HIV-1 IIIB-infected, CD8-cells. (B1) iDC ingested the apoptotic CD8- cells, as demonstrated by confocal laser scanning microscopy. (B2 to B5) Section series were collected from the apical to the basal surface of cells at 0.5-µm section intervals at a resolution of 1,024 x 1,024 pixels.
|
The addition of CD40L to DC of three uninfected and three HIV-1-infected persons resulted in a similar pattern of surface marker changes, so the results were combined as shown in Fig. 2. The data show that there were significant increases in surface marker expression on iDC, i.e., HLA DR, CD80, and CD86, after loading with some of the antigen preparations (Fig. 2). Interestingly, all of these increases were noted with the iDC+AP-IIIB preparations. However, iDC+AP preparations only increased the expression of CD86, and iDC+NE only increased the expression of HLA-DR and CD86. These changes were found in DC from both HIV-1-infected and uninfected subjects (data not shown). Thus, the modifications in surface marker expression were not dependent on HIV-1 infection in the antigen preparations or the study subjects.
![]() View larger version (30K): [in a new window] |
FIG. 2. Effect of CD40L and various antigen preparations on the expression of costimulatory and HLA class I and II molecules on DC. Data shown are the means (+ the standard error) for the MFI of DC from three uninfected persons (N2, N3, and N4) and three HIV-1-infected subjects (S7, S8, and S9). *, Comparison of the effect of antigen loading on iDC expression of HLA-DR (iDC+AP-IIIB, P = 0.019; iDC+NE, P = 0.047), CD80 (iDC+AP-IIIB, P = 0.04), and CD86 (iDC+AP, P = 0.032; iDC+AP-IIIB, P = 0.001; iDC+NE, P = 0.029) (paired t test); **, comparison of the iDC with matched mDC (P 0.02, except for P = 0.038 and 0.044 for CD83 expression on NE-loaded and NE-IIIB-loaded DC, respectively). All other comparisons were P > 0.05.
|
We next investigated the capacity of DC loaded with the apoptotic, necrotic, and cell-free X4 virus preparations to induce IFN-
production in autologous T cells. An enzyme-linked immunospot assay was used to assess the production of IFN-
from single, antigen-specific T cells, as described elsewhere (14). A concentration of 104 antigen-loaded iDC or mDC was added to 105 autologous PBMC (i.e., a 1:10 ratio) and incubated overnight at 37°C. The results were expressed as the number of spot-forming cells (SFC) per 106 PBMC. This assay had an average, within-sample variation of 12% among replicates of >100 PBMC samples tested in our laboratory.
In initial experiments with autologous PBMC from three HIV-1-infected subjects, we found that either CD8-cells, PBMC, iDC, or mDC did not produce appreciable IFN-
(0 to 5 SFC/105 cells; data not shown). The iDC, or iDC loaded with the various antigens, induced low levels of IFN-
in PBMC (Fig. 3). However, PBMC produced high levels of IFN-
when stimulated by mDC loaded with HIV-1 IIIB-superinfected, autologous, apoptotic CD8- cell preparations (mDC+AP-IIIB) compared to stimulation with mDC+AP (Fig. 3). This effect was dependent on the presence of the mDC, since only 0 to 1 SFC/105 cells was detected when the CD8- cell preparations were added to autologous PBMC without mDC. The IFN-
response was also dependent on the concentration of mDC+AP-IIIB added to the PBMC cultures, since ratios of 1:5 and 1:20 of DC to responder PBMC resulted in proportionally higher and lower numbers of IFN-
-producing cells, respectively, than the standard 1:10 ratio (data not shown). In contrast to the effects of the apoptotic cell preparations, mDC loaded with either of the necrotic cell preparations (mDC+NE or mDC+NE-IIIB), or infected with cell-free, viable HIV-1 IIIB (Fig. 3) or PUV-treated HIV-1 IIIB (data not shown) did not induce appreciable IFN-
production in the PBMC. Thus, mDC were superior to iDC at inducing IFN-
production in autologous PBMC. Moreover, HIV-1 IIIB-superinfected apoptotic cells were more potent immunogens in these mDC than were HIV-1 IIIB-superinfected necrotic cell preparations or cell-free HIV-1 IIIB.
![]() View larger version (17K): [in a new window] |
FIG. 3. IFN- production by PBMC from three subjects chronically infected with HIV-1 (S3, S7, and S8) induced by iDC or mDC loaded with the various antigen preparations.
|
in response to the mDC+AP-IIIB (Fig. 4A). In further experiments, the addition of either anti-HLA class I MAb or anti-HLA class II MAb to the DC-antigen preparations (14) inhibited induction of IFN-
production by autologous mDC+AP-IIIB; the addition of both of these MAb completely prevented induction of IFN-
in the PBMC (Fig. 4B). Therefore, mDC+AP-IIIB induced IFN-
production in autologous CD8+ and CD4+ T cells that was HLA class I and class II restricted, respectively.
![]() View larger version (15K): [in a new window] |
FIG. 4. (A) Both CD8+ and CD4+ T-cell subpopulations produced IFN- after treatment with mDC+AP-IIIB (subject S6). (B) Both HLA class I- and class II-associated, anti-HIV-1 IFN- responses were induced in PBMC by mDC+AP-IIIB (subject S5). These data are representative of five experiments.
|
in PBMC cultures from eight HIV-1-infected persons. The individual and composite data confirmed our findings that the greatest number of IFN-
-producing cells was induced by mDC+AP-IIIB (Table 1). The highest levels of IFN-
were stimulated by mDC+AP-IIIB in PBMC cultures from seven of the eight HIV-1-infected persons compared to mDC+AP. HIV-1-specific T-cell responses were not induced by HIV-1 antigen-loaded iDC or mDC preparations in PBMC from three HIV-1-negative subjects.
The one HIV-1-infected subject (S2) who did not have greater numbers of IFN-
-producing cells induced by the mDC+AP-IIIB had the highest response to mDC+AP. This same pattern of T-cell response to the apoptotic cell preparations was confirmed in PBMC obtained 6 months later from this subject (data not shown).
Interestingly, HIV-1 p24 was detected in all of the AP preparations, although the levels were lower than in the AP-IIIB preparations (Table 1). This represents endogenous HIV-1 from naturally infected cells of the study subjects and could have resulted in stimulation of a portion of the IFN-
by the mDC+AP preparations. However, induction of IFN-
in the PBMC cultures was not significantly different when mDC+AP were used for antigen presentation compared to mDC alone.
We next determined whether the T-cell response to the apoptotic cell preparation of X4 virus was consistent over time. We found that there were comparable levels of IFN-
induced by autologous mDC+AP-IIIB over a >2-year period in one subject (S5) (Fig. 5). Very low levels of IFN-
were induced by mDC alone or mDC+AP. Persistent production of IFN-
by PBMC stimulated with mDC+AP-IIIB was confirmed at two separate time points 6 months apart in subject S3 (data not shown).
![]() View larger version (22K): [in a new window] |
FIG. 5. T-cell responses to mDC+AP-IIIB are persistent over time in the same individual (subject S5). Cryopreserved PBMC were used as responder cells for this experiment.
|
production was not related to the amount of HIV-1 p24 in the AP or AP-IIIB antigen preparations loaded in the DC. Further studies indicated that very low levels of p24 were maintained from the baseline level through 60 h of culture of iDC or mDC that had been loaded with HIV-1 IIIB-infected, PUV-treated, apoptotic CD8- cell preparations (Table 2). These low levels of p24 were similar to those in concurrent cultures of the HIV-1 IIIB-infected, PUV-treated, apoptotic CD8- cell preparations without the DC. These results show that PUV treatment inactivates HIV-1 in the apoptotic CD8- cell preparations, with a resultant lack of replication of HIV-1 in DC loaded with them. However, both iDC and mDC loaded with non-PUV-treated, HIV-1 IIIB-infected CD8- cells supported very high levels of replication of HIV-1 (Table 2). In contrast, iDC or mDC alone did not support efficient HIV-1 replication after infection with cell-free X4 virus (data not shown). |
View this table: [in a new window] |
TABLE 2. HIV-1 p24 levels in iDC and mDC loaded with untreated and PUV-treated, HIV-1 IIIB-infected CD8- autologous cells
|
We found that HIV-1-infected necrotic cells were not a potent T-cell immunogen compared to HIV-1-infected apoptotic cells. This is in agreement with the findings of Albert et al. (2), who demonstrated that activation of influenza A virus-specific CD8+ T cells by mDC loaded with influenza A virus-infected apoptotic monocytes was superior to that induced by mDC containing virus-infected necrotic cell preparations. Moreover, the influenza virus antigen from the apoptotic cell preparations was processed through a nonclassical, alternative HLA class I pathway by the DC. The efficiency of presentation of antigens of HIV-1 and other viruses may reflect differences in the mechanisms of processing and presentation of apoptotic and necrotic cell preparations by DC. These are likely based on differences in viral proteins, such as the predominance of glycine-alanine repeats that can act on proteosomes (8), and differential activity of cellular components such as chaperone molecules that regulate antigen trafficking (29).
There was little or no HIV-1 replication demonstrable in either the iDC or the mDC that had been loaded with the X4 virus-infected, PUV-treated apoptotic CD8- cells. The mDC loaded with this form of inactivated HIV-1 were, however, highly immunogenic for anti-HIV-1 CD8+ T cells. This suggests that the major processing of HIV-1 proteins in these DC for presentation to CD8+ T cells was not through a classic HLA class I endogenous pathway. We hypothesize that, after the ingestion of HIV-1-infected apoptotic cells by the iDC, HIV-1 proteins in the apoptotic cells are processed through an alternative HLA class I pathway for cross-presentation by mDC to HIV-1-specific CD8+ T cells. HIV-1 peptides processed by the infected CD8- cells may also serve as a source of antigen for the mDC. Additional studies are required to address directly these hypotheses, including inhibitors of various steps in HIV-1 replication and processing of viral proteins by DC.
The iDC or mDC infected with replication-competent or PUV-treated cell-free HIV-1 IIIB did not activate an HIV-1-specific T-cell response in our study. Buseyne et al. (10) recently reported that iDC that had been infected with X4 virus could stimulate HIV-1-specific CD8+ T cells in vitro. This required fusion and cytosolic processing of the virus by the iDC and was not related to endogenous virus replication in these cells. However, antigen presentation in that study was determined by stimulation of HIV-1 peptide-specific CD8+ T-cell lines and clones by the virus-fused DC. Such cell lines and clones should have a much greater proportion of HIV-1-specific CD8+ T cells than the PBMC used in our work. Their T-cell model may therefore detect lower amounts of HIV-1 peptide expressed in HLA class I molecules on the DC. Collectively, our studies suggest that DC have nonclassical, alternative mechanisms for cross-presentation of X4 viral antigens to CD8+ T cells.
Of importance is that maturation of DC induced by treatment with CD40L after loading of the iDC with the HIV-1 antigen preparations was essential for the efficient induction of anti-HIV-1 T-cell responses. This is likely related to enhanced expression of cell surface molecules CD80 and CD86 that are required for activation of T cells (6). CD40L also induces cytokines such as IL-12 (32) and IL-15 (28) that upregulate and prolong survival of antigen-specific T cells. In this regard, Ostrowski et al. (33) reported that CD40L treatment of HIV-1 peptide-treated DC enhances their activation of HIV-1-specific cytotoxic T lymphocytes in vitro, which is associated with production of IL-15 by the DC. However, in terms of the natural HIV-1 infection, these immunity-enhancing effects of CD40L on DC should be viewed in the context of potential pathogenic effects due to upregulation of HIV-1 X4 replication in cocultures of CD40L-treated mDC and CD4+ T cells (30). Indeed, we found high levels of virus replication in both iDC and mDC cultures loaded with non-PUV-treated, HIV-1 IIIB-infected CD8- cells. Thus, there may be competing events of uptake and presentation of HIV-1 X4-infected cells by mDC to T cells, with replication and cytopathic effects of this virus.
We (37) and others (26, 35, 36) have previously shown that CD8+ T-cell reactivity to HIV-1 is only temporarily increased in patients with long-term, chronic HIV-1 infection receiving combination antiretroviral drug therapy, even though the levels of circulating HIV-1 are greatly diminished. Additionally, there is usually poor recovery of anti-HIV-1 CD4+ T-cell activity in these persons (9, 12, 36, 38, 49). There do appear, however, to be residual T cells that are HIV-1 specific. This has been shown by reactivity of CD8+ T cells to HIV-1 during treatment interruptions (16, 39, 41) and by in vitro stimulation with DC loaded with HIV-1 peptides (33) or proteins complexed with liposome (14). The present study indicates that HIV-1-specific CD8+ T cells, as well as CD4+ T cells, can be readily activated to produce IFN-
in vitro by a single round of stimulation with DC that have been loaded with autologous, HIV-1-infected apoptotic cells and matured with CD40L. This approach may be useful as a therapeutic vaccine to enhance T-cell immunity to a person's own endogenous HIV-1 by autologous DC that have been loaded with apoptotic cells that contain strains of HIV-1 that are unique to the individual.
We thank K. Picha of Immunex for assistance in providing the CD40L; S. Narula of Schering-Plough (Kenilworth, N.J.) for providing hIL-4 and hGM-CSF; W. Buchanan and B. Calhoun for clinical assistance; R. Day, S. Barratt-Boyes, and L. Harshyne for technical advice; S. Alber, M. White, D. Damph, H. Li, C. Perfetti, C. Kalinyak, W. Jiang, M. Mather, and P. Zhang for technical assistance; and J. Malenka for administrative assistance. We give special thanks to the volunteers of the Pitt Men's Study-MACS, whose dedication made this study possible.
|
|
|---|
vß5 and CD36, and cross-present antigens to cytotoxic T lymphocytes. J. Exp. Med. 188:1359-1368.
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»