HIV PATHOGENS IN HUMAN LYMPHOID TISSUE
Jean-Charles Grivel, PhD, Staff Scientist
Beda Brichacek, PhD, Senior Research Fellow
Angelique Biancotto, PhD, Visiting Fellow
Cristian Condack, MD, Visiting Fellow
Yana Kiselyeva, PhD, Visiting Fellow
Andrea Lisco, MD, PhD, Visiting Fellow
Christophe Vanpouille, PhD, Visiting Fellow
Wendy Fitzgerald, BS, Technician
Sarah Iglehart, BS, Predoctoral Fellow
![Photo of Leonid Margolis, P h. D.](graphics/margolis.jpg)
HIV infection is transmitted in lymphoid tissues by CCR5-using HIV-1 (R5), which often evolves later into variants able to use CXCR4 (X4 or R5X4); these variants, in turn, are associated with accelerated progression to AIDS. HIV-1 pathogenesis is critically dependent on, among other factors: the site of infection, with gut-associated lymphoid tissue playing an important role in R5/X4 selection; tissue immune activation; and coinfecting microbes that interact with HIV. To investigate these aspects of HIV infection in a complex tissue microenvironment, we studied viral pathogenesis in explants of various human tissues infected with HIV-1 alone or in combination with other viruses. In one project, we compared R5 and X4 HIV-1 infection of human rectosigmoid and tonsillar tissues. In another, we focused on lymph-node immune activation in HIV-1 pathogenesis. We also continued our investigations into the effect of non–HIV microbes on HIV infection. The results of these studies provide new opportunities for designing strategies to prevent and contain HIV infection by manipulating the response of human tissues to HIV or mimicking the negative effects of non–HIV microbes on HIV replication. The studies also permit us to expand our understanding of the basic mechanisms of HIV transmission, evolution, and interactions with other viruses in human tissues.
CCR5- and CXCR4-using HIV-1 infection of human rectosigmoid and tonsillar tissues
In vivo, HIV-1 uses CCR5 and/or CXCR4 coreceptors for cell entry. As stated above, infection is transmitted by CCR5-using HIV-1 variants (R5) that often later evolve into variants able to use CXCR4 in addition to CCR5 or exclusively CXCR4 (R5X4 or X4); this evolution is associated with an accelerated progression to AIDS. The viral determinants conferring cellular tropism are mainly associated with the variable region 3 (V3) of the viral envelope protein. Irrespective of route of transmission (sexual, vertical, or intravenous), acute infection is almost exclusively associated with CCR5-tropic HIV-1 (R5), although body fluids that transmit HIV-1 infection (semen, vaginal secretions, blood, and milk) contain both R5 and X4. Later in the course of infection, X4 HIV-1 variants emerge. In human lymphoid tissue, the presence of CCR5-binding chemokines accelerates the selection for X4 HIV-1 variants. Several mutations or a single pathway of V3 mutations may underlie the conversion of R5 into X4. The mechanism and cause of selective R5 transmission and its early dominance are not fully understood, but they probably involve several sequential and parallel barriers (gatekeepers). Recently, researchers have been focusing on the role of gut-associated lymphoid tissue (GALT) independently of the route of transmission. GALT is thought to be one of the barriers that selects R5 over X4 variants.
To study mechanisms of selection in GALT, we used a system of intestinal explants derived from both endoscopy and surgical resection specimens. In this system, which we and others developed some years ago, we tested whether GALT—under controlled laboratory conditions—is more vulnerable to R5 than to X4 HIV variants. We compared infection of rectosigmoid tissue with infection of tonsillar tissue by these two HIV variants. As previously described for human tonsillar and lymph node explants, colorectal explants support productive ex vivo infection by both R5 and X4 HIV-1 without exogenous stimulation and with similar replication kinetics in both tissues. We found that the level of R5 replication in rectosigmoid tissue is much greater than that in tonsillar tissue. The most evident explanation for the higher production of R5 HIV-1 in human colorectal tissue lies in the large difference in the presence of CCR5+ T cells. While such cells in tonsillar tissue constitute about 15 percent of the total CD4+ T cells at the time of infection, these cells in GALT accounted for up to 70 percent of CD4 T cells. The abundance of CCR5+ T cells in both tissues was numerically close to the relative replication levels of R5 HIV-1 in these tissues. Furthermore, tonsillar tissue responds to X4 HIV-1 infection by upregulating the secretion of CC-chemokines, providing a potential CCR5 blockade and further resistance to R5 infection, whereas rectosigmoid tissue failed to increase such innate immune responses. Our results show that rectosigmoid tissue is more prone than tonsillar lymphoid tissue to R5 HIV-1 infection primarily because of the high prevalence and availability of R5 cell targets and reduced chemokine blockade. Nevertheless, most CD4+ T cells express CXCR4, and X4 HIV-1 readily replicates in both tissues, suggesting that the differential expression of coreceptors is one of the gatekeepers of HIV-1 infection and is complemented by other barriers, accounting for selective R5 infection of the rectal mucosa in vivo.
Fletcher P, Elliott J, Grivel J-C, Margolis LB, Anton P, McGowan I, Shattock R. Ex vivo culture of human colorectal tissue for the evaluation of candidate microbicides. AIDS 2006;20:1237-45.
Grivel J-C, Elliott J, Lisco A, Biancotto A, Condack C, Shattock R, McGowan J, Margolis LB, Anton P. HIV-1 pathogenesis differs in rectosigmoid and tonsillar tissues infected ex vivo with CCR5- and CXCR4-tropic HIV-1. AIDS 2007;21:1263-72. (See comment on this paper in: Poli G. Busting a gut understanding HIV pathogenesis in lymphoid tissue. Future HIV Ther 2007;1:247-50.)
Kiselyeva Y, Nedellec R, Ramos A, Pastore C, Margolis LB, Mosier DE. Evolution of CXCR4-using human immunodeficiency virus type 1 SF162 is associated with two unique envelope mutations. J Virol 2007;81:3657-61.
Margolis L, Shattock R. Selective transmission of CCR5-utilizing HIV-1: the ‘gatekeeper’ problem resolved? Nat Microbiol 2006;4:312-7.
Activation of HIV-infected lymphoid tissue as a driving force for HIV infection
The depletion of CD4+ T cells associated with redistribution and sequestration of various lymphocyte subpopulations is a hallmark of HIV infection. Immune hyperactivation followed by cytokine secretion seems to be the centerpiece of such redistribution and sequestration, the actual mechanisms of which remain to be elucidated. To investigate the mechanisms, we studied cytokine expression spectra and distributions of T cell subsets in freshly excised and cultured lymph nodes obtained from either chronically HIV-1–infected patients or healthy controls. We found that a reduction in the fraction of CD4+ T cells in lymph nodes from HIV-1–infected patients was associated with an increase in effector T cell frequencies and a profound upregulation of CD38, an activation marker, in naive, central memory, and effector CD4+ and CD8+ T cells. Likewise, the death receptor Fas (CD95) was more frequently detectable on T cells from HIV+ nodes. Dendritic cell (DC) depletion was dramatic, with plasmacytoid DC 40-fold and myeloid DC 20-fold less frequent in HIV+ nodes than in control nodes. Cytokine dysregulation was evident, with IL-2 and IL-15 secretion as much as two or three orders of magnitude greater in infected than in control lymph nodes. Further, IL-10 and IL-1b secretion was dramatically upregulated in HIV-1+ nodes. Extrapolating our data to the situation in vivo, we propose that the deterioration in lymph node structure and function in the course of chronic HIV-1 infection is a complicated phenomenon that includes the death of infected cells and activation of other cells, both infected and uninfected. In this setting, activated effector cells are excessively attracted or retained in the infected nodes and serve as a source of cytokines, which in turn may activate new cells and trigger a massive activation of lymphoid tissue. In the case of chronic HIV-1 infection, this phenomenon leads to the lymphadenopathy and massive bystander activation that characterizes HIV-1 infection. Given that the activated cells are primed for apoptosis, this mechanism contributes to a massive death of T cells and dendritic cells. Such depletion may contribute significantly to the impairment of both innate and adaptive immune defenses in chronic HIV-1 infection.
The distorted activation patterns of lymphocytes in lymph nodes and tonsils from HIV-1–infected individuals emphasize the important role of tissue activation in HIV-1 disease progression. To reveal mechanisms connecting cell activation and HIV-1 replication under controlled laboratory conditions, we inoculated blocks of human tonsillar tissue with the virus and analyzed T cell activation status. HIV-1 readily infects various CD4+ T cells in this system, and the viral load depends on the number of activated target cells, but only those of a particular pattern—CD25+/HLA-DR+. We observed no positive correlation between viral production and the number of cells expressing other activation markers, namely, CD69, CD38, or CD95. Moreover, HIV-1 infection of lymphoid tissue was associated with activation of both HIV-1–infected and –uninfected (bystander) T cells. Remarkably, HIV-1 infection seems to facilitate the same pattern of lymphoid tissue activation that is associated with enhanced HIV-1 replication. HIV infection of lymphoid tissues was followed by CD4+ T cell apoptosis that was selectively elevated in T cells expressing both CD25/HLA-DR and p24gag, but not in cells expressing either of these markers alone. The HIV-1–induced activation of bystander cells should promote viral spreading by creating a pool of new viral targets that, upon infection, produce virus at a higher rate.
In summary, in lymphoid tissues, HIV induces a vicious circle of infection of activated cells, driving them into apoptosis and triggering cytokine dysregulation that, in turn, leads to activation of uninfected cells and makes the cells susceptible to productive HIV infection and inappropriately attracts and/or retains the cells in lymphoid tissue. Targeting individual elements of this cycle may constitute a new anti–HIV-1 strategy. In general, a better understanding of the mechanisms by which HIV-1 infection induces immune activation resulting in CD4+ T cell depletion may provide new insight into HIV-1 pathogenesis and reveal new targets for immune-based interventions to slow HIV-1 disease progression.
Biancotto A, Grivel J-C, Iglehart SJ, Vanpouille C, Lisco A, Sieg SF, Debernardo R, Garate K, Rodriguez B, Margolis LB, Lederman MM. Abnormal activation and cytokine spectra in lymph nodes of people chronically infected with HIV-1. Blood 2007;109:4272-9. (See comments on this paper in: Grossman Z. Looking at HIV-infected lymph nodes. Blood 2007;109:4116-7.)
Biancotto A, Iglehart SJ, Vanpouille C, Condack CE, Lisco A, Ruecker E, Hirsch I, Margolis LB, Grivel J-C. HIV-1-induced activation of CD4+ T cells creates new targets for HIV-1 infection in human lymphoid tissue ex vivo. Blood 2007 [E-pub ahead of print].
Effect of non–HIV pathogens on HIV infection in human lymphoid tissue
We continue to study the effects of non–HIV pathogens on HIV infection in human lymphoid tissue. Study of interactions between vaccinia virus (VACV) and HIV occurring in coinfected individuals is of particular clinical relevance in view of possible antipoxvirus immunizations of HIV-infected subjects and the use of vaccinia as a vector to deliver an anti–HIV vaccine. To understand VACV pathogenesis in the context of human tissues, we used human tonsils ex vivo and found that the tissue supports productive infection by the New York City Board of Health’s VACV strain (used to make Dryvax vaccine). VACV readily infected both T and B lymphocytes and depleted cells of both subsets equally. Among T lymphocytes, CD8+ cells were preferentially depleted because they were preferentially infected; the probability that a CD8+ T cell would be productively infected, and hence depleted, was almost six times higher than that for a CD4+ T cell. T cells expressing CCR5 or the activation markers CD25, CD38, and HLA-DR were other major targets for VACV infection. With CCR5 one of the two HIV coreceptors, the effect of VACV on CCR5-expressing cells caused the two viruses to enter into complex interactions in coinfected individuals. To study VACV/HIV-1 interactions, we coinfected human lymphoid tissue ex vivo with VACV and either a CCR5-using HIV variant or a CXCR4-using HIV-1. Given that VACV induced CCR5 downregulation and depleted activated CCR5-expressing cells at the principal site of productive HIV-1 infection, R5 replication was dramatically inhibited, whereas replication of CXCR4-using HIV-1 was only mildly inhibited.
We and our colleagues reported measles virus (MV) as another pathogen that interacts with HIV in coinfected tissues. The mechanisms of MV’s immunosuppression, as well as its pathogenesis in human tissues, remain poorly understood. Therefore, to assess the extent of MV replication in human lymphoid tissues and to identify its target cells, we infected ex vivo human tonsillar tissue with either wild-type or vaccine strains of MV. We characterized the spread of MV following low-multiplicity infection. We isolated B lymphocytes (CD19+), memory and naive T lymphocytes (CD3+), macrophages (CD14+), natural killer cells (CD16+/CD56+), and other cells from tissue blocks nine days after infection and analyzed them with flow cytometry. MV infected up to 50 percent of B lymphocytes, whereas macrophages and natural killer and other (including dendritic) cells were infected at lower levels (10 to 20 percent). Cells that express the MV receptor–signaling lymphocyte activation molecule (SLAM, CD150) were preferentially infected by all wild-type MV strains. Memory (CD45RO+) T lymphocytes expressing high levels of SLAM were infected by wild-type MVs about five times more efficiently than were naive (CD45RA+/CD62L+) T lymphocytes that do not express SLAM. The vaccine strain, however, did not discriminate between memory and naive T lymphocytes. Depletion of SLAM-expressing T lymphocytes was more marked than depletion of other SLAM-expressing lymphoid cells. The results provide new insights into the mechanisms of MV-induced immune suppression and vaccine attenuation. Furthermore, the preferential infection and depletion of activated T cells by wild-type MV provides new insight into the mechanism by which MV infection suppresses HIV replication in regions where both viruses are endemic. Indeed, as we have shown (see above), the number of activated T cells is paramount in determining the viral load in HIV-infected tissues. Therefore, a reduction in the number of memory T cells activated by MV infection will, in addition to previuosly reported mechanisms, diminish HIV viral load in coinfected patients.
Thus, the study of VAC and MV, together with earlier reported studies of human herpesviruses 6 and 7, revealed diverse mechanisms of HIV interactions with non–HIV viruses. The mechanisms involve competition for activated T cells—particularly for CCR5-expressing CD4 T cells—downregulation of HIV receptors and coreceptors, upregulation of chemokines, and depletion of target cells. Interactions between HIV and non–HIV pathogens may generate both positive and negative signals for HIV replication. We attempted to mimic one of the negative signals generated by non–HIV pathogens (HHV-7) by specifically downmodulating CD4 with cyclotriazadisulfonamide (CADA). Application of CADA resulted in significant suppression of HIV-1 infection in human lymphoid tissue ex vivo. Simulation of such negative signals with (inactivated) pathogens, their components, or other compounds may be an important strategy in suppressing HIV infection. The model of ex vivo–infected human tissues may be used to evaluate how various microbes or microbe-based vectors affect HIV-1 in coinfected tissues.
Condack C, Grivel J-C, Devaux P, Margolis L, Cattaneo R. Measles virus vaccine attenuation: suboptimal infection of lymphatic tissue and tropism alteration. J Infect Dis 2007;196:541-9.
Vanpouille C, Biancotto A, Lisco A, Brichacek B. Interactions between HIV-1 and vaccinia virus in human lymphoid tissue ex vivo. J Virol 2007;81:12458-64.
COLLABORATORS
Peter Anton, MD, David Geffen School of Medicine at UCLA, Los Angeles, CA
Roberto Cattaneo, PhD, MayoClinicCollegeof Medicine, Rochester, MN
Ivan Hirsch, PhD, INSERM, Centre de Recherche en Cancérologie de Marseille, Institut Paoli-Calmettes, Université Méditerranée, Marseille, France
Michael Lederman, MD, Case Western Reserve University/University Hospitals of Cleveland, Cleveland, OH
Paolo Lusso, MD, PhD,Laboratory of Immunoregulation, NIAID, Bethesda, MD
Donald E. Mosier, PhD, MD, The Scripps Research Institute, La Jolla, CA
Robin Shattock, MD, St. George’s Hospital Medical School, London, UK
Dominique Schols, PhD, Rega Institute for Medical Research, Katholieke Universiteit, Leuven, Belgium
Kurt Vermeire, PhD, Rega Institute for Medical Research, Katholieke Universiteit, Leuven, Belgium
For further information, contact margolil@mail.nih.gov.