Human immunodeficiency virus gp120 downregulates CD1d cell surface expression

Immunology Letters 98 (2005) 131–135

Human immunodeficiency virus gp120 downregulates CD1d cell surface expression Chadi A. Hagea,b,c , Lisa L. Kohlia,b , Sungyoo Choa,d , Randy R. Brutkiewicza,d , Homer L. Twigg IIIa,b , Kenneth S. Knoxa,b,∗ a Indiana University School of Medicine, Indianapolis, IN, USA Department of Medicine, Division of Pulmonary and Critical Care Medicine, Indiana University Medical Center, Roudebush VA Medical Center, 1481 West 10th Street, VA-111P, Indianapolis, IN 46202, USA c Department of Medicine, Division of Infectious Diseases, USA d Department of Microbiology and Immunology, The Walther Oncology Center, The Walther Cancer Institute, USA

b

Received 30 September 2004; received in revised form 27 October 2004; accepted 28 October 2004 Available online 24 November 2004

Abstract CD1d is an MHC class I-like surface molecule that presents endogenous glycoplipid antigens. The effect of HIV infection on CD1d surface expression has not yet been reported. FACS analysis revealed significantly lower levels of CD1d on CD14+ monocytes from HIV-infected subjects compared to HIV-infected subjects on HAART and healthy controls. CD1d expression correlated inversely with viral load in infected individuals. CD1d surface expression on human cell lines was downregulated after infection with M-tropic HIV, T-tropic HIV, or after exposure to HIV gp120 in vitro. These data suggest that CD1d-mediated responses are altered during HIV infection and may thus contribute to the global immunodeficiency seen in these patients. © 2004 Elsevier B.V. All rights reserved. Keywords: HIV; CD1d; MHC; Monocyte; CD14; U937; M-tropic; gp120

1. Introduction In addition to the rapid depletion of CD4+ T cells, infection with the human immunodeficiency virus (HIV) leads to several immunologic consequences including apoptosis of bystander cells [1], activation of primary macrophages [2], and downregulation of cell surface MHC-1 molecules [3]. It is hypothesized that the loss of MHC-I on the cell surface may allow the virus to evade immune surveillance, limiting the diversity of the cytotoxic T cell (CTL) response. CD1d is a non-polymorphic MHC class I-like surface molecule that presents endogenous glycoplipid antigens to a specific subset of T lymphocytes termed natural killer T

∗

Corresponding author. Tel.: +1 317 554 1739; fax: +1 317 554 1743. E-mail address: [email protected] (K.S. Knox).

0165-2478/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.imlet.2004.10.025

(NKT) cells. In humans, CD1d is highly expressed on peripheral blood monocytes, cortical thymocytes and at lower levels on B lymphocytes [4]. CD1d-restricted NKT cells are immunoregulatory cells that participate in tumor immunity as well as autoimmunity [5]. Increasing evidence suggests that CD1d-mediated antigen presentation also orchestrates an effective immune response to various infections. CD1drestricted NKT cells are necessary for optimum control and clearance of several pathogens, including Mycobacterium tuberculosis [6], respiratory syncytial virus [7], and herpes simplex virus [8]. Immune responses to viral infections may be uniquely dependent upon CD1d antigen presentation and NKT cell regulatory function. In a CDld knockout mouse model, it is clear that CD1d is needed to orchestrate an appropriate response to infection with lymphocytic choriomeningitis virus [9,10]. Recently, a patient with CD1d-restricted NKT cell

132

C.A. Hage et al. / Immunology Letters 98 (2005) 131–135

deficiency was described who suffered from disseminated varicella-zoster virus infection after receiving an attenuated vaccine [11]. In vitro, EBV-transformed B cells lose CD1d during the course of infection (unpublished observation). Thus, in the context of a chronic viral infection that is unable to be cleared effectively by the host immune system, CD1d-generated immune responses are likely aberrant. It is not known whether HIV infection alters CD1d cell surface expression. As CD1d is structurally similar to MHCI, which is known to be downregulated during the course of HIV infection, we hypothesized that CD1d surface expression would also be reduced by HIV. We studied surface expression of CD1d on human monocytes from infected patients and compared them to healthy controls. Utilizing a variety of CD1d-expressing human cell lines, the effect of in vitro HIV infection on CD1d surface expression was characterized. This brief report, to our knowledge, is the first to show that CD1d is downregulated in HIV-infected patients. We provide in vitro evidence to support this clinical observation.

2. Materials and methods We recruited eight HIV-infected individuals with no recent infection or illness. Four HIV subjects were clinically stable and candidates for HAART, but were na¨ıve to antiretroviral therapy and four were compliant with highly active antiretroviral therapy (HAART). Patients who were categorized by their treating physician as a long-term nonprogressor or those who had failed HAART and discontinued therapy were not enrolled. Patients on HAART had undetectable viral loads within 3 months prior to enrollment. Four healthy volunteers were enrolled as controls. After informed consent was obtained, blood samples were collected by venipuncture into heparinized syringes. Peripheral blood mononuclear cells (PBMC) were isolated using Ficoll (Sigma) density gradient centrifugation. Cells were washed and immediately used for fluorescent antibody labeling and flow cytometric analysis. In addition to PBMCs, several cell lines were tested for CD1d expression. CD1d expressing cell lines were tested for downregulation of surface CD1d after in vitro HIV infection or after exposure to gp120 recombinant protein. Uninfected cells and cells pulsed with BSA were cultured in parallel as controls. Cell lines (Jurkat, MOLT-4, U937, THP-1, H9) were infected as previously described [12]. Briefly, 2 × 106 cells were exposed to either Ada-M or HTLVIIIb (NIH AIDS Research and Reference Reagent Program, Rockville, MD) at a TCID50 of 103 for 2 h with constant gentle shaking. Cells were then washed and cultured in 10 ml cell culture flasks in RPMI 1640 with 10% FBS, penicillin (50 U/ml), streptomycin (50 ␮g/ml) and incubated 6 days at 37 ◦ C with 5% CO2 . In some experiments, 3 ␮g/ml PHA (Sigma) was added to the final 3 days of culture to augment infectivity. In parallel experiments, BSA or 110 ng/ml HIV gp-120 protein (Mtropic HIV-1 SF162 Cat# 7363, NIH AIDS Research and

Reference Reagent Program, Rockville, MD) was added in lieu of viral infection. Cells were harvested on the first and the sixth day of culture and CD1d expression was analyzed by flow cytometry. p24 levels in the supernatant of the cell culture was measured using a commercially available ELISA kit (Coulter, Miami, FL; sensitivity = 7.5 pg/ml). Monoclonal antibodies used included PE-conjugated anti-CD1d, PerCP-conjugated anti-CD14, FITC-conjugated CCR5, APC-conjugated anti-HLA-A, B, C (Becton, Dickinson) and APC-conjugated CXCR4 (R&D). FITC, APC and PerCP isotype (Becton, Dickinson) and PE isotype (Caltag) were used as controls in all analysis. Samples were acquired on a FACSCalibur flow cytometer with CellQuest software (Becton, Dickinson) and analyzed using FCS Express (De Novo Software). Gating on the CD14+ cell population, the CD1d surface expression was quantified by geometric mean (Gm) of the fluorescent intensity. Comparisons between group means was performed using the Mann–Whitney Rank Sum test for nonparametric data. Correlation between CD1d surface levels and HIV viral load after log transformation was analyzed using Spearman’s Rank Order correlation coefficient. P values <0.05 were considered significant. All analyses were done using SigmaStat Version 3.0.1a (Systat Software, Inc.).

3. Results Eight HIV-infected subjects were enrolled. CD4 T cell counts averaged 347 cells/␮l (range: 108–482) for HIVinfected patients not receiving therapy. CD4 cell counts were comparable in patients receiving HAART (mean: 271 cells/␮l, range: 240–336). HIV RNA viral loads were below the level of detection (<40 copies RNA/ml) in all HIVinfected individuals who were receiving HAART, and ranged between 22,000–200,000 copies RNA/ml (mean: 87,000 copies RNA/ml) in patients naive to therapy. Compared to healthy controls, CD1d surface expression was significantly lower on CD14+ monocytes from HIV-infected individuals (P = 0.010). Patients receiving HAART exhibited levels of CD1d comparable (P = 0.615) to those of healthy controls (Fig. 1). As such, in HIV-infected individuals, CD1d expression showed an inverse correlation (r = −0.79; P = 0.03) with viral load that was not seen with the absolute CD4 count (r = −0.19; P = 0.61). We attempted to use primary human cells for our infection assays as recently reported [13], but experiments could not be performed in monocyte-derived macrophages (cultured with or without M-CSF) or monocyte-derived dendritic cells (using GM-CSF and IL-4) as CD1d was not reproducibly or reliably expressed (only approximately 10% of experiments) on the cell surface in either of these primary cell types. Ficoll separation does not affect CD1d surface expression on monocytes when compared to whole blood from the same donor. However, Ficoll separation does enhance the rate of downregulation of CD1d on monocytes when placed in cul-

C.A. Hage et al. / Immunology Letters 98 (2005) 131–135

Fig. 1. CD1d surface expression on CD14+ monocytes. PBMC from HIVinfected individuals and healthy controls were labeled with PE-conjugated anti-CD1d and PerCP-conjugated anti-CD14 monoclonal antibodies, and analyzed for CD1d expression by gating on the CD14+ monocyte population. Untreated, HIV-infected subjects had significantly less CD1d expression on the cell surface compared to healthy controls (P = 0.010) and patients on HAART (P = 0.008, Mann–Whitney Rank Sum test). There was a strong correlation between viral load and CD1d expression in the eight HIV-infected subjects (r = −0.79; P = 0.03, Spearman’s Rank Order correlation coefficient, not shown). Data represented as mean (dark line), median (thin line), and range. Intensity index represents the fold increase over isotype control for each study subject.

ture. This was true regardless of culturing conditions (human serum, FBS, plastic adherence, Teflon bags). Thus, we examined CD1d expression on various cell lines. MOLT-4, U937, THP-1 and Jurkat cell lines all express CD1d on their cell surface. In contrast, H9 cells do not express surface CD1d. Surface expression is not altered by culturing with PHA, but is downregulated if PMA is used. Length of time in culture does not affect CD1d expression. Because our clinical observation was with monocytes from HIV-infected subjects, we selected the monocyte/macrophage U937 cell line for the in vitro experiments shown, despite the fact that Jurkat T cells expressed the highest levels of CD1d on the cell surface and are more permissive to infection in vitro. Infection of U937 cells with the M-tropic virus Ada-M showed very little change in CD1d surface expression after 24 h of culture. However, Fig. 2 shows representative data from four experiments in which HIV M-tropic virus Ada-M reduced CD1d surface levels on U937 cells after 6 days of culture (black histogram). Interestingly, addition of gp120 from an M-tropic strain, SF162, produced similar levels of CD1d downregulation (light gray histogram), suggesting active replication may not be necessary for this effect. Similar results were obtained for MOLT-4 and Jurkat T cell lines after infection with T-tropic HTLVIIIb (not shown). In parallel, MHC class I surface expression was measured as a positive control and was downregulated to varying degrees on all cell lines at 6 days, confirming previous reports [3]. CCR5 and CXCR4 surface expression was also measured to delineate the possibility that HIV tropism was responsible for the effects seen. Importantly, U937 cells express very little CCR5 and are not overtly permissive to infection with

133

Fig. 2. CD1d is downregulated by HIV infection in vitro. The open histogram represents uninfected cells and the dark gray filled histogram represents isotype control. U937 cells were exposed to Ada-M for 2 h, washed and cultured. Alternatively, HIV gp120 protein was added and cultured in parallel experiments. A significant, and surprisingly similar, decrease in CD1d surface expression is noted after exposure to either live virus (black histogram) or gp120 protein alone (light gray histogram). Live virus produced low grade infection as p24 levels in supernatant were 100–300 pg/ml.

M-tropic virus. Thus, cells remained viable during culture and the impact of tropism on CD1d downregulation is speculated to be minimal. CXCR4 is abundantly expressed on all cell lines, including U937 cells, and when exposed to Mtropic HIV or M-tropic gp120, CXCR4 remained expressed at high levels, suggesting a specific effect on CD1d and MHC molecules.

4. Discussion Our small, descriptive study is the first to show that surface CD1d expression is downregulated on CD14+ monocytes of HIV-infected patients. In contrast, patients receiving HAART have levels of CD1d surface expression comparable to healthy controls. CD1d-expressing cell lines lose CD1d after in vitro HIV infection and this loss may be independent of productive infection as the gp120 protein is similarly capable of downregulating CD1d. These findings may provide insight as to the role of CD1d in chronic human infections and the pathogenesis of HIV viral escape. The role of CD1d in human disease is yet to be defined. Studies examining the role of CD1d antigen presentation typically use ␣-galactosylceramide (␣-GalCer), the only known exogenous ligand for CD1d, to stimulate responses [5]. These ␣-GalCer responses are much more robust than responses generated against endogenous glycolipids and can elicit immunological effects with very low levels of CD1d [14]. From these studies, it is known that CD1d-restricted NKT cells are regulatory, skewing the immune response to a TH2 phenotype. However, upon stimulation, CD1d-restricted cells release a burst of both TH1 and TH2 cytokines critical in ini-

134

C.A. Hage et al. / Immunology Letters 98 (2005) 131–135

tiating the adaptive immune response [5]. Shortly thereafter, CD1d-restricted NKT cells decline. This decline in NKT regulatory cells is thought to allow propagation of the specific TH1 response [9]. Because CD1d is downregulated during chronic infections and CD1d deficiency may predispose to viral infections [8,11], it is interesting to speculate that CD1drestricted NKT cell responses are necessary to contain viral escape mutants, and in the absence of CD1d viral escape is facilitated. Differences in CD1d-mediated immunity may exist for the acute response to chronic infection or an infection that is typically cleared by the host. CD1d-mediated immune responses also play a role in tumor surveillance and downregulation of CD1d may explain the increased incidence of malignancy in patients with HIV. Although CD1d expression has not been previously studied in HIV infection, NKT cells are among the cells that are depleted during the course of HIV infection [15]. The significance of NKT cell depletion and the mechanism that leads to selective NKT cell depletion during the course of HIV infection are not entirely clear. In a cross sectional study of HIV-infected individuals, van der Vliet et al. clearly showed a selective depletion of NKT cells. The decline in NKT cells did not correlate with the viral load or any other indices of disease severity [16]. A second study by Motsinger et al. suggested that the CD4+ subset of NKT cells are directly infected with HIV [17]. It is unclear whether CD1d downregulation contributes to the NKT cell loss seen in HIV-infected patients. However, data from animal models suggest that CD1d may provide a survival signal for NKT cells as CD1d knockout mice completely lack NKT cells [18,19]. In contrast to this observation, recent data using IL-7 knockout mice suggests that NKT cell survival and expansion might be independent of CD1d expression [20]. Thus, determining whether CD1d is necessary for NKT cell survival in the periphery requires further study. Our study has limitations that stem from the small size of the cohort studied, the cross sectional design, and the absence of genetic information in the eight HIV-infected subjects. In one patient studied, CD1d surface levels were unchanged after 1 month of HAART, suggesting that the kinetics of CD1d downregulation and recovery may take several months, allowing us to speculate that downregulation may be at the RNA level. Because patients infected with HIV are a heterogeneous population with a variable course of disease, we recently studied four other patients with HIV infection (three failed HAART with progressive disease and possible concomitant opportunistic infections and one was deemed a long-term nonprogressor) who represent a different population than shown in Fig. 1. These patients showed CD1d surface expression levels comparable to controls. Thus, careful characterization, ideally at the genetic level, of these subjects is vitally important. Despite these limitations, this initial observation is strongly supported by the in vitro cell line data. Longitudinal studies are ongoing in our laboratory to delineate whether the loss of CD1d has functional immunologic consequences

or correlates with disease severity. A recent study by Shinya showed that HIV nef is important in CD1a downregulation [21]. Thus, alternative mechanisms by which CD1d is downregulated requires further study as well. Acknowledgements HIV strain HTLVIIIb and HIV-1 SF162 gp120 were obtained through the NIH AIDS Research and Reference Reagent Program, Division of AIDS, NIH.We would like to thank Ron Collman, MD for his technical advice.Supported by grants K08 HL04545 to Kenneth S. Knox and RO1 AI46455 to Randy R. Brutkiewicz.

References [1] Gougeon ML. Apoptosis as an HIV strategy to escape immune attack. Nat Rev Immunol 2003;3:392–404. [2] Twigg III HL, Lipscomb MF, Yoffe B, Barbaro DJ, Weissler JC. Enhanced accessory cell function by alveolar macrophages from patients infected with the human immunodeficiency virus: potential role for depletion of CD4+ cells in the lung. Am J Respir Cell Mol Biol 1989;1:391–400. [3] Collins KL. How HIV evades CTL recognition. Curr HIV Res 2003;1:31–40. [4] Exley M, Garcia J, Wilson SB, Spada F, Gerdes D, Tahir SM, Patton KT, Blumberg RS, Porcelli S, Chott A, Balk SP. CD1d structure and regulation on human thymocytes, peripheral blood T cells, B cells and monocytes. Immunology 2000;100:37–47. [5] Vincent MS, Gumperz JE, Brenner MB. Understanding the function of CD1-restricted T cells. Nat Immunol 2003;4:517–23. [6] Chackerian A, Alt J, Perera V, Behar SM. Activation of NKT cells protects mice from tuberculosis. Infect Immun 2002;70:6302–9. [7] Johnson TR, Hong S, Van Kaer L, Koezuka Y, Graham BS. NK T cells contribute to expansion of CD8+ T cells and amplification of antiviral immune responses to respiratory syncytial virus. J Virol 2002;76:4294–303. [8] Grubor-Bauk B, Simmons A, Mayrhofer G, Speck PG. Impaired clearance of herpes simplex virus type 1 from mice lacking CD1d or NKT cells expressing the semivariant V␣14-J␣281 TCR. J Immunol 2003;170:1430–4. [9] Hobbs JA, Cho S, Roberts TJ, Sriram V, Zhang J, Xu M, Brutkiewicz RR. Selective loss of natural killer T cells by apoptosis following infection with lymphocytic choriomeningitis virus. J Virol 2001;75:10746–54. [10] Roberts TJ, Lin Y, Spence PM, Van Kaer L, Brutkiewicz RR. CD1d1-dependent control of the magnitude of an acute antiviral immune response. J Immunol 2004;172:3454–61. [11] Levy O, Orange JS, Hibberd P, Steinberg S, LaRussa P, Weinberg A, Wilson SB, Shaulov A, Fleisher G, Geha RS, Bonilla FA, Exley M. Disseminated varicella infection due to the vaccine strain of varicella-zoster virus, in a patient with a novel deficiency in natural killer T cells. J Infect Dis 2003;188:948–53. [12] Day RB, Wang Y, Knox KS, Pasula R, Martin II WJ, Twigg III HL. Alveolar macrophages from HIV-infected subjects are resistant to Mycobacterium tuberculosis in vitro. Am J Respir Cell Mol Biol 2004;30:403–10. [13] Knox KS, Day RB, Wood KL, Kohli LL, Hage CA, Foresman BH, Schnizlein-Bick CT, Twigg III HL. Macrophages exposed to lymphotropic and monocytotropic HIV induce similar CTL responses despite differences in productive infection. Cell Immunol 2004;229:130–8.

C.A. Hage et al. / Immunology Letters 98 (2005) 131–135 [14] Spada FM, Borriello F, Sugita M, Watts GF, Koezuka Y, Porcelli SA. Low expression level but potent antigen presenting function of CD1d on monocyte lineage cells. Eur J Immunol 2000;30:3468–77. [15] Unutmaz D. NKT cells and HIV infection. Microbes Infect 2003;5:1041–7. [16] van der Vliet HJ, von Blomberg BM, Hazenberg MD, Nishi N, Otto SA, van Benthem BH, Prins M, Claessen FA, van den Eertwegh AJ, Giaccone G, Miedema F, Scheper RJ, Pinedo HM. Selective decrease in circulating V␣24+ V␤11+ NKT cells during HIV type 1 infection. J Immunol 2002;168:1490–5. [17] Motsinger A, Haas DW, Stanic AK, Van Kaer L, Joyce S, Unutmaz D. CD1d-restricted human natural killer T cells are highly susceptible to human immunodeficiency virus 1 infection. J Exp Med 2002;195:869–79.

135

[18] Mendiratta SK, Martin WD, Hong S, Boesteanu A, Joyce S, Van Kaer L. CD1d1 mutant mice are deficient in natural T cells that promptly produce IL-4. Immunity 1997;6:469–77. [19] Chen YH, Chiu NM, Mandal M, Wang N, Wang CR. Impaired NK1+ T cell development and early IL-4 production in CD1-deficient mice. Immunity 1997;6:459–67. [20] Matsuda JL, Gapin L, Sidobre S, et al. Homeostasis of V␣ 14i NKT cells. Nat Immunol 2002;3:966–74. [21] Shinya E, Owaki A, Shimizu M, Takeuchi J, Kawashima T, Hidaka C, Satomi M, Watari E, Sugita M, Takahashi H. Endogenously expressed HIV-1 nef down-regulates antigen-presenting molecules, not only class I MHC but also CD1a, in immature dendritic cells. Virology 2004;326:79–89.