Previous Article | Next Article ![]()
Journal of Virology, November 2002, p. 11033-11041, Vol. 76, No. 21
0022-538X/02/$04.00+0 DOI: 10.1128/JVI.76.21.11033-11041.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.
Department of Pathology,1 Department of Genetics, Stanford University School of Medicine, Stanford, California 94304,2 Department of Pathology, University of California, San Francisco, California 941103
Received 18 March 2002/ Accepted 19 July 2002
|
|
|---|
/CCL3 when exposed to HIV in vitro. IPC also induce naïve CD4 T cells to proliferate and would therefore preferentially infect these cells. These results indicate that IPC may play an important role in the dissemination of HIV. |
|
|---|
) in response to viral infection. Mature IPC, also called PDC or DC2, function as antigen-presenting cells capable of inducing immunity in naïve allogeneic T cells similar to classical myeloid DC (MDC). Viral infection in particular triggers IPC to induce Th1 immunity (7, 18). In contrast to MDC, which are derived from a committed MDC precursor or granulocyte-monocyte precursor, IPC express high levels of the interleukin 3 (IL-3) receptor (CD123) and lack myeloid markers such as CD11c (8, 15, 25, 31). A recent study suggests that IPC may be of lymphoid origin (35). Originally termed "plasmacytoid T cells" because of their unique plasma-cell-like morphology and their localization in the T-cell zones of lymph nodes (LN) (38), these cells have been identified in blood, tonsil, and LN (15, 17, 25). IPC precursors require IL-3 for survival and an activation stimulus, such as CD40 ligand (CD40L), for maturation into PDC (15). While the susceptibility of DC to human immunodeficiency virus (HIV) has been extensively studied, most reports have focused on either MDC or bulk DC populations that would represent a mixture of MDC and IPC (4, 20, 26, 28). The susceptibility of IPC to HIV infection has been difficult to examine because of their scarcity, representing only 0.1% of peripheral blood mononuclear cells (PBMC). Nevertheless, IPC have been identified in sites, such as blood and tonsil (15), where HIV is transmitted and, therefore, could also be exposed to virus during infection. Moreover, a progressive loss of IPC in the blood has been observed in HIV-infected patients, suggesting an interaction between IPC and HIV (9, 34). A preliminary report suggests that IPC may be susceptible to HIV, although neither the molecular basis for infection nor the effect of IPC maturational status on infectivity was explored (27).
|
|
|---|
In vivo expansion of human DC. HIV-seronegative study patients participating in a cancer vaccine trial, who provided signed informed consent that fulfilled Institutional Review Board guidelines, received 10 daily subcutaneous injections of Flt3L (Immunex, Seattle, Wash.) at a dose of 20 µg/kg of body weight/day to a maximum daily dose of 1.5 mg/day. Patients then underwent leukapheresis, and PBMC were obtained with Ficoll separation of the apheresis product.
Chemotaxis assays.
Chemotaxis assays were performed as previously described (24). Briefly, 106 PBMC in 100 µl of complete medium (CM) comprised of RPMI 1640 (BioWhittaker, Walkersville, Md.) with 10% human AB+ serum supplemented with IL-3 at a saturating concentration of 10 ng/ml (Peprotech, Rocky Hill, N.J.) were added to each 24-well 5-µm-pore-size transwell insert (Corning, Acton, Mass.) with 600 µl in the lower chamber, with or without chemokine. SDF-1/CXCL12 and MIP-1
/CCL3 (R&D Systems, Minneapolis, Minn.) were added at concentrations ranging from 1 to 1,000 ng/ml. After 3 h of incubation at 37°C, the transwell inserts were removed. The cells in the lower chamber were collected, stained with antibody, and enumerated by collecting events for a fixed time (60 s) on a FACSCalibur. By counting a 1:5 dilution of input cells, the absolute number of cells that transmigrated could be obtained. To determine which subsets of DC were migrating, cells were stained and analyzed by four-color flow cytometry as described. Results are expressed as the percentage of input DC per transwell. Standard deviations are shown for three replicate transwells. Data are representative of three separate experiments.
Human DC isolation. PBMC were labeled with mouse monoclonal antibodies to human CD3, CD14, CD19, and CD56 and washed. Antibody-labeled PBMC were then depleted with anti-mouse immunomagnetic beads (Dynal, Lake Success, N.Y.). This negative selection was followed by positive selection with immunomagnetic beads for major histocompatibility complex class II (MHC-II) (Miltenyi, Auburn, Calif.) to yield DC with >85% purity. DC were then further purified with a fluorescence-activated cell sorter (FACS) (Becton Dickinson and Cytomation, Fort Collins, Colo.) to exclude residual CD3-, CD14-, CD19-, and CD56-staining cells and to separate IPC from MDC. IPC and MDC were subsequently cultured in CM.
HIV strains. HIV-1BaL (from Suzanne Gartner, Mikulas Popovic, and Robert Gallo), HIV-1LAI/IIIB (from Gallo), and human recombinant IL-2 (from Maurice Gately, Hoffmann-La Roche, Inc.) were obtained through the AIDS Research and Reference Reagent Program, Division of AIDS, National Institute of Allergy and Infectious Diseases, National Institutes of Health (NIH). HIV-1BaL was prepared by infection of IL-2-stimulated PBMC isolated from buffy coats of HIV-seronegative blood donors. PBMC were infected and then stimulated with 50 U of recombinant IL-2/ml. Virus-containing culture supernatants were harvested 6 days later and stored at -80°C. The 50% tissue culture infective dose (TCID50) was determined in IL-2-stimulated PBMC. HIV-1LAI/IIIB was prepared by infection of T-cell blasts (PBMC stimulated for 3 days with 5 µg of phytohemagglutinin/ml, washed extensively, and resuspended in CM and 50 U of recombinant IL-2/ml). Virus-containing culture supernatants were harvested 6 days later and stored at -80°C. The TCID50 was determined in IL-2-stimulated T-cell blasts.
HIV infection.
DC were infected by a 2-h incubation with the macrophage-tropic R5 HIV-1BaL or the T-lymphotropic X4 HIV-1LAI/IIIB strain at 1,500 TCID50s per 106 cells and were then washed three times, resuspended in CM, and plated in 96-well plates at a concentration of 0.2 x 106 cells/ml (40,000 DC/well). When indicated, following the 2-h incubation, soluble CD40L trimer (CD40LT) (gift of Immunex) was added at a final concentration of 500 ng/ml. Cell-free supernatants were collected at 24 h and assessed for IFN-
, RANTES, and MIP-1
secretion by enzyme-linked immunosorbent assay (ELISA) (R&D Systems). Cell-free supernatants were collected every 7 days and stored at -80°C for determination of reverse transcriptase (RT) activity and p24 antigen concentration (NEN, Boston, Mass.). Where indicated, IPC were preincubated with RANTES or SDF-1
at a concentration of 1 µg/ml and were then infected with virus. Where indicated, neutralizing monoclonal anti-IFN-
antibody (R&D Systems) or a matched isotype control antibody was added to the cultures to a final concentration of 10 µg/ml following infection, and the cultures were supplemented with antibody every 2 days. For viability experiments, infected IPC cultured in parallel with and without CD40LT (40,000 DC/well) were harvested, stained with antibodies, annexin V and PI, and enumerated and analyzed by flow cytometry.
Immunohistochemistry. Following heat-induced epitope retrieval with 10 mM citrate buffer, pH 6.0, formalin-fixed tissue was incubated with anti-human CD123 antibody (Becton Dickinson) and detected with DAKO's (Carpinteria, Calif.) LSAB+ peroxidase system and diaminobenzidine (brown). Double staining was achieved by first staining with anti-CD123 as described but utilizing Vector AS (Vector, Burlingame, Calif.) and then denaturing any residual antibody-enzyme complexes with acid. Following a buffer wash, the tissues were incubated with anti-p24 antibody (DAKO 1:10, overnight at room temperature) and detected with DAKO's LSAB+ Alkaline Phosphatase system with Vector Red (Vector). Paraffin sections from rectal biopsies, cervical biopsies, and tonsil were obtained from the tissue bank at San Francisco General Hospital under an approved human subject protocol. Rectal tissues were derived from routine colonoscopy. Cervical tissues were derived from a patient with cervical dysplasia. The tonsillar tissue was derived from a 30-year-old homosexual male infected with HIV who was on no antiviral medications. He underwent tonsillectomy for adenotonsillar hypertrophy.
IPC/T-cell coculture experiments. Resting CD4+ T cells were purified from autologous PBMC by negative selection against CD8, -14, -19, and -56 and HLA DR with immunomagnetic beads (Miltenyi) and lacked surface expression of CD40L as assessed by flow cytometry. These unstimulated CD4+ T cells were cocultured with HIV-infected DC at a concentration of 0.5 x 106 CD4 T cells/ml (100,000 CD4 T cells/well). In separate cultures, unstimulated CD4+ T cells were also exposed to HIV for 2 h, washed, and cultured without IPC. When indicated, CD40LT was added to the coculture at a final concentration of 500 ng/ml. Cell-free supernatants were collected every 7 days and stored at -80°C for determination of RT activity.
CD4 stimulation by IPC. Naïve CD4 T cells were purified from PBMC of HIV-seronegative study subjects by negative selection for CD8, CD14, CD19, HLA DR, and CD45RO with immunomagnetic beads (Miltenyi). Monocytes and B cells were purified by positive selection with immunomagnetic beads specific for CD14 and CD19, respectively. IPC were purified as previously described. These different antigen-presenting cells were irradiated to 30 Gy and coincubated at 10,000 cells/well with 100,000 autologous CD4+ CD45RA+ T cells/well in CM in 96-well plates. T-cell proliferation in triplicate cultures was assessed after 6 days for [3H]thymidine incorporation over 18 h and measured with a scintillation counter (Wallac, Toruc, Finland).
|
|
|---|
, and MIP-1
(Fig. 1d). IPC express high levels of CD62L (Fig. 1c), which is required for migration across high endothelial venules (HEV) (3), and lack expression of DC-SIGN, an HIV binding C-type lectin (12).
![]() View larger version (40K): [in a new window] |
FIG. 1. IPC express HIV coreceptors. (a) Blood DC from healthy subjects were phenotypically defined by flow cytometry as expressing MHC-II and lacking lineage markers CD3, -14, -19, and -56 (box). (b) These gated cells were further divided into two distinct subsets either expressing high levels of CD123 and lacking CD11c (IPC) (solid box) or expressing CD11c (MDC) (dashed box). (c) IPC were then further characterized for their surface expression of CD4, CXCR4, CCR5, CD62L, and DC-SIGN. Open histograms represent cells stained with isotype-matched control antibodies. (d) The capacity of IPC (black bar) and MDC (white bar) to migrate in response to SDF-1 and MIP-1 was assessed in a transwell chemotaxis assay. Error bars represent the standard deviation for three replicate transwells. (e) Following 10 daily doses of Flt3L to expand DC in vivo, analysis of blood reveals that approximately 16% of PBMC are DC. (f) The proportion of IPC relative to MDC is essentially unchanged by Flt3L treatment. Data shown are representative of three experiments.
|
IPC and MDC were purified from Flt3L-mobilized blood with immunomagnetic beads and separated by FACS on the basis of their expression of CD123 and CD11c to >99% purity. These purified cells represent either IPC (pre-DC2) or immature MDC that lack surface expression of CD80 and 86. We then exposed IPC and MDC to either the macrophage-tropic R5 HIV-1BaL or the T-cell-tropic X4 HIV-1LAI/IIIB strains that use CCR5 or CXCR4, respectively, as a coreceptor for viral entry (1, 11, 30). Macrophage-tropic strains of HIV are responsible for primary HIV infection, while the T-cell-tropic strains usually arise late in infected individuals and are responsible for syncytium formation. After 2 h of incubation, cells were washed free of unbound virus and cultured. HIV replication was assessed by measuring Mg2+-dependent RT activity (19). IPC did not efficiently replicate either HIV-1BaL or HIV-1LAI/IIIB (Fig. 2a). In contrast, MDC that were simultaneously purified replicate HIV-1BaL but not HIV-1LAI/IIIB, consistent with prior reports (23).
![]() View larger version (39K): [in a new window] |
FIG.2. HIV replication in IPC requires CD40L activation. IPC and MDC were enriched with immunomagnetic beads and were then separated by FACS based on their differing expression of CD123 and CD11c. IPC and MDC were incubated with HIV for 2 h, washed, and then cultured in parallel in 96 wells at a concentration of 0.2 x 106 cells/ml (40,000 DC/well). (a) Following incubation of IPC and MDC with either HIVLAI/IIIB (triangles) or HIVBaL (circles), triplicate cultures were assessed weekly for viral replication on the basis of RT activity. (b) IPC and MDC were also stimulated with CD40LT following exposure to virus and were assessed weekly for RT activity. (c) IPC exposed or unexposed to HIVBaL and then stimulated with CD40LT were also assessed by flow cytometry for intracellular HIV p24 following 1 week of culture. The histogram on the right represents infected cells stained with an irrelevant isotype-matched control antibody. (d) IPC cultures were also assessed for T-cell contamination and for viability by annexin V and PI staining at day 14. Numbers represent the percentage of cells present in the indicated gates. (e) IPC were incubated with either RANTES or SDF-1 at a concentration of 1 µg/ml for 1 h prior to and during exposure to medium alone (black bars), HIVBaL (gray bars), or HIVLAI/IIIB (black bars). Cultures were then assessed for p24 production at 1 week by ELISA. Error bars represent standard deviations from triplicate wells. Data shown are representative of three experiments.
|
To determine what proportion of IPC became productively infected following exposure to virus, we performed intracellular staining for HIV p24 2 weeks after exposure to HIV-1BaL. p24 staining could be detected in over 17% of IPC cultured in the presence of CD40LT (Fig. 2c). At this time point, IPC cultures had minimal contamination by CD3+ T cells (Fig. 2d, upper panel), arguing against contaminating lymphocytes as the source of viral replication. The viability of IPC was not significantly altered by CD40LT or HIV infection (Fig. 2d, lower panel), nor was the number of cells recovered (data not shown). These results argue against CD40LT augmenting viral replication by acting as a survival factor for IPC. HIV-1BaL entry is dependent on CCR5 binding, since preincubation of IPC prior to viral exposure with RANTES/CCL5, a ligand for CCR5, blocked infection while preincubation with SDF-1
, a ligand for CXCR4, had no effect (Fig. 2e). Conversely, preincubation of IPC with SDF-1
but not with RANTES blocked infection with HIV-1LAI/IIIB.
To assess the capacity of IPC to produce IFN-
in response to HIV, we measured IFN-
in IPC culture supernatants 1 day following addition of HIV-1BaL. As shown in Fig. 3a, the virus-exposed IPC produced significant amounts of IFN-
that were augmented with CD40L activation. The addition of CD40L without virus also induced IFN-
secretion by IPC. IPC exposed to supernatants derived from IL-2-stimulated PBMC in the absence of virus did not secrete IFN-
(data not shown). To determine whether the produced interferon affected viral replication, we cultured HIV-1BaL-infected IPC in the presence or absence of neutralizing anti-IFN-
antibody. While there was no effect on viral production without CD40LT-induced maturation, addition of anti-IFN-
antibody significantly accentuated viral replication in IPC incubated with CD40LT (Fig. 3b). IPC were also assessed for their capacity to produce RANTES (Fig. 3c) and MIP-1
/CCL3 (Fig. 3d) in response to viral infection and/or maturation with CD40LT. While MIP-1
production was induced only by exposure to HIV, RANTES production could also be induced by CD40LT without virus in a pattern similar to IFN-
production. Thus, HIV infection leads to activation of IPC that may be distinct from maturation signals delivered through CD40L.
![]() View larger version (31K): [in a new window] |
FIG. 3. IFN- production by IPC exposed to HIV. (a) IPC production of IFN- was measured by ELISA 1 day following exposure to HIVBaL with or without the addition of CD40LT. (b) IPC were also incubated with HIVBaL in the presence or absence of neutralizing anti-IFN- antibody, and p24 production was assessed after 1 week. Error bars represent standard deviations from triplicate wells. Data shown are representative of three experiments. IPC production of RANTES (c) and MIP-1 (d) was measured by ELISA 1 day following exposure to HIVBaL with or without the addition of CD40LT.
|
![]() View larger version (126K): [in a new window] |
FIG. 4. Tissue distribution of IPC. (a) IPC can be identified within the extrafollicular T-cell-rich zones of LN by staining with anti-CD123 antibody (brown) (magnification, x100). IPC can be visualized within the HEV. (b) IPC can also be identified in the parafollicular T-cell zone in tonsil (magnification, x100). (c) Large numbers of IPC are seen in the cervical submucosa (magnification, x100). (d) Rare IPC can also be found in the rectal submucosa (arrow) (magnification, x100).
|
![]() View larger version (109K): [in a new window] |
FIG. 5. Evidence for HIV infection of IPC in vivo. (a) In double-stain experiments without counterstaining of tonsillar tissues, uninfected CD123+ (black) IPC that lack p24 staining (red) can be identified (magnification, x400). (b) p24+ (red) cells can be identified in the tonsils of an HIV-infected individual (magnification, x400). (c) HIV+ IPC can also be seen that are CD123+ (black) and p24+ (red) in tonsil from an HIV-infected individual. (d) Multinucleated giant cells are also identified with CD123 (black) and p24 (red) staining.
|
antibody accentuated viral production in IPC/T-cell cocultures, whether or not CD40LT was added (Fig. 6f).
![]() View larger version (27K): [in a new window] |
FIG. 6. IPC transmit HIV to autologous CD4 T cells. (a) Following a 2-h exposure to either HIVLAI/IIIB (triangles) or HIVBaL (circles), 4 x 104 IPC were washed and then coincubated with 105 autologous CD4 T cells. Viral replication was assessed weekly for RT activity in triplicate cultures. (b) Infected IPC/T-cell cocultures were also incubated in the presence of CD40LT following exposure to HIV. CD4 T cells that were directly exposed to either HIVLAI/IIIB (squares) or HIVBaL (diamonds), washed, and cultured alone served as controls. Data shown correspond to those for cultures performed in parallel with experiments for Fig. 2 and are representative of three experiments. By comparison, MDC/CD4 T-cell cultures were assessed in parallel either without (c) or with (d) the addition of CD40LT. (e) IPC/T-cell cocultures incubated in the presence or absence of HIVBaL and/or CD40LT were assessed by flow cytometry for intracellular HIV p24 following 1 week of culture. (f) IPC/T-cell cocultures incubated in the presence or absence of neutralizing anti-IFN- antibody were assessed for p24 production 1 week following exposure to HIVBaL.
|
![]() View larger version (13K): [in a new window] |
FIG. 7. IPC stimulate autologous T cells to proliferate. B cells, IPC, MDC, and monocytes were compared as antigen-presenting cells in an autologous mixed leukocyte reaction. Ten thousand antigen-presenting cells/well were cocultured for 6 days with 100,000 autologous CD45RA+ CD4+ T cells/well. T-cell proliferation was assessed by [3H]thymidine incorporation. Error bars represent standard deviations from triplicate wells. Data shown are representative of three experiments.
|
|
|
|---|
B activation. Nevertheless, the requirement of maturation on viral replication differs from the case for MDC, in which viral replication occurs without CD40L-induced maturation and can actually be inhibited by this signal (23). Thus, infected IPC may evade HIV-specific immunity, including cytotoxic T lymphocytes, until they interact with activated CD4 T cells within the extrafollicular T-cell zones of lymphoid tissues. Once this interaction occurs and CD40 is ligated, IPC then replicate virus. IPC may, therefore, serve as viral reservoirs for both macrophage and T-cell-tropic strains.
IPC begin secreting IFN-
within 24 h of HIV exposure, regardless of whether they replicate virus. Thus, viral infection triggers activation signals distinct from CD40L as has been suggested previously (18). IFN-
appears to partially suppress viral replication in both IPC and IPC/T-cell cocultures, consistent with prior reports on the effects of IFN-
/ß on HIV replication (32, 39). While CD40L has been described as a survival factor for MDC, we saw no survival advantage in CD40LT-activated IPC. Cellular viability may have been maintained, in part, by virally induced IFN-
, which has been shown to be an IPC survival factor (16). Nevertheless, the difference in HIV replication between IPC exposed to HIV in the presence of CD40LT and those exposed in the absence of CD40LT is not due to altered viability of the IPC mediated by CD40 ligation. IPC also produce ß-chemokines in response to viral exposure and therefore would attract CD4 T cells.
Importantly, IPC can infect CD4 T-lymphocytes in trans, even without significant viral replication. Such infection may stem from either internalized or surface-bound virus. CD40L-mediated activation of IPC profoundly increases viral production in the IPC/T-cell cocultures. Since HIV infects proliferating CD4 T cells more efficiently than resting CD4 T cells (13, 37), the ability of IPC to induce naïve T cells to proliferate may also lead to more efficient infection of CD4 T cells. Nevertheless, MDC are approximately 50-fold more efficient than IPC at infecting autologous CD4 T cells. As mentioned previously, the blood-derived IPC used in this study do not express DC-SIGN, which might explain differences in viral replication between IPC and MDC. Further work is necessary to confirm this possibility.
IPC may represent an important HIV reservoir in infected individuals. The capacity of HIV to exploit IPC, a key cell in both innate and adaptive antiviral immunity, could contribute to the chronic and evasive nature of HIV infection. Understanding how CD40 signaling in IPC leads to viral replication should further our understanding of the HIV life cycle and perhaps bring about additional targeted therapies for this devastating viral infection.
L.F. is supported by a physician-scientist award from the National Cancer Institute (K23 CA82584-01). M.M. is a fellow of Istituto Superiore di Sanita', Rome, Italy. This investigation was also supported in part by NIH grant P01-HL56443 (to E.G.E.), by Human Health Service grant M01-RR00070 (General Clinical Research Centers, National Center for Research Resources, NIH), by NIH grants MH-59037 and P30 MH59037 UCSF-GIVI CFAR (to B.G.H.), and by NIH grants CA-42509-14 and CA-81543-02 (to L.A. Herzenberg, for support of M.M.).
|
|
|---|
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»