Leonid Margolis, PhD, Head, Section on Intercellular Interactions
Jean-Charles Grivel, PhD, Staff Scientist
Angelique Biancotto, PhD, Visiting Fellow
Cristian Condack, MD, Visiting Fellow
Andrea Lisco, MD, PhD, Visiting Fellow
Christophe Vanpouille, PhD, Visiting Fellow
Silvia Chen, PhD, Special Volunteer
Sarah Igelhart, BS, Predoctoral Fellow

Most of our experimental knowledge of microbial pathogenesis comes from experiments with a microbe infecting otherwise sterile, cultured, isolated cells. In vivo, however, critical events of microbial pathogenesis occur in tissues where diverse cells are immersed in a heterogeneous microenvironment populated by pathogenic and commensal microbes affecting each other. HIV interactions with other microbes, such as human herpesviruses (HHV), may determine the clinical course of HIV disease. In particular, the interactions may differentially affect CCR5-using HIV-1 (R5), which selectively transmits infection and dominates HIV's early stages, and CXCR4-using HIV-1 (X4), which often evolves at the disease's late stages. The "gatekeeping" mechanisms selecting R5 over X4 are not understood and are thought to be associated with the properties of gut-associated lymphoid tissue (GALT), which is an early HIV target. The study of human microbial pathogenesis requires an adequate experimental tissue model. We proposed a new hypothesis on the "gatekeeping" mechanism. In particular, we study the role of GALT in selective transmission of R5 HIV-1, identifying the mechanisms of the differential interactions of HHV-7 with R5 and X4 HIV-1 in co-infected human lymphoid tissue ex vivo, and we develop new tissue-like models based on experimental primate embryonic stem cell differentiation.

Differential infection of human tissue ex vivo by CCR5- and CXCR4-using HIV-1

Grivel, Biancotto, Lisco, Condack, Margolis; in collaboration with Anton, McGowan, Shattock

As one of the main goals of HIV research, an understanding of the mechanisms of HIV-1 transmission is critical for the development of effective topical microbicides and of other preventive strategies. HIV-1 entry into target cells is dependent on the sequential interactions of the viral gp120 glycoprotein with the cellular receptors CD4+ and either CCR5 (R5 HIV-1) or CXCR4 (X4 HIV-1). Although both R5 and X4 phenotypes have been described for many HIV-1 subtypes (or clades), R5 virus is almost exclusively associated with acute infection irrespective of route of transmission (sexual, vertical, or intravenous). However, body fluids that transmit HIV-1 infection (semen, vaginal secretions, blood, and milk) contain both R5 and X4. Despite much effort, researchers have been unable to identify the "gatekeeping" mechanism that restricts the more efficient transmission of CXCR4-utilizing HIV-1 (X4) versus the transmission of CCR5-utilizing HIV-1 (R5). We hypothesize that there is no single "gatekeeper" mechanism but that, instead, a range of mechanisms (innate and acquired) at various stages of HIV transmission restrict the establishment of infection by X4, although no one stage on its own seems to provide a perfect barrier. Nevertheless, the superimposition of several of these leaky "gatekeepers" provides almost perfect protection against X4 transmission in a more reliable manner than a single, efficient firewall. Thus, the collective efforts of several laboratories may have unexpectedly solved the problem of the selective transmission of R5 HIV-1 variants, thereby suggesting that the search for one critical "gatekeeper" mechanism should be abandoned.

We wish to test this hypothesis by studying GALT, which has recently been shown to play a central role in the early stage of HIV infection by providing a strong barrier against X4 HIV-1 infection. To this end, we developed a system of intestinal explants derived from both endoscopy and surgical resection specimens. The expression of CCR5 and HLA-DR in these explants was higher on T cells than on blood-derived T cells, whereas the expression of CXCR4 and CD38 appeared to be more comparable to that observed in blood T cells. We also determined that the density of co-receptor expression varied between T lymphocytes isolated from blood and those isolated from both tissue sites. Not only did more CD4+ T cells express CCR5 in GALT than in blood and tonsils, but the cells also appeared to express a greater density of CCR5 per cell. The observed higher frequency of R5 target cells in rectal tissue suggests a mechanism for preferential R5 HIV-1 transmission. We demonstrated that it is possible to infect intestinal explants productively with HIV-1, thereby allowing us to compare GALT infectivity by X4 and R5 HIV-1 variants. In addition, the intestinal explant system can be used for screening candidate microbicides, and it is a useful tool for the preclinical evaluation of potential microbicides.

Effect of herpesviruses 7 on HIV-1 in co-infected human lymphoid tissue

Lisco, Grivel, Biancotto, Margolis; in collaboration with Lusso, Moss, Schols

Herpesviruses 7 (HHV-7) are highly prevalent in the general population and are a common viral co-pathogen in HIV-infected individuals. To understand the complex interactions between HHV-7 and HIV-1 in co-infected individuals, we investigated such interactions in the context of ex vivo human lymphoid tissue, a tissue in which critical events of viral pathogenesis occur in vivo. We first investigated whether human lymphoid tissue is capable of supporting productive HHV-7 infection ex vivo. We found that blocks of human lymphoid tissue inoculated with HHV-7 become productively infected without exogenous activation. As with HHV-6, which we studied earlier, HHV-7 suppresses the replication of CCR5-tropic (R5) HIV-1 but only mildly inhibits the replication of CXCR4-tropic (X4) HIV-1. Even though HHV-7 and HHV-6 are two closely related human herpesviruses with a high degree of genetic homology and, as noted, share striking similarities in the way they modulate HIV-1 infection, we found that the molecular mechanisms of HIV-1 suppression caused by these two closely related herpesviruses differ dramatically. In contrast to HHV-6, which suppresses CCR5-tropic HIV-1 by upregulating RANTES, HHV-7 does not enhance chemokine production. Instead, HHV-7 infection abrogates CD4+ expression in T lymphocytes in both productively infected and uninfected (bystander) cells. We demonstrated the abrogation of CD4+ expression on these T lymphocytes directly by measuring of CD4+ fluorescence intensity on these cells. In addition to the downregulation of CD4+, which leads to a generalized decline in the number of potential HIV targets and reduces the susceptibility of these tissues to HIV-l infection, mild depletion of CD4+ T cells may also contribute to the suppression of HIV-1. Interestingly, HHV-7 exerts its inhibitory effect predominantly on CCR5-utilizing HIV-1, just as replication of HHV-7 is severely inhibited in tissues co-infected with CXCR4-utilizing HIV-1. In summary, ex vivo co-infection of human lymphoid tissues with HHV-7 and either CCR5-utilizing or CXCR4-utilizing HIV-1 results in different outcomes. The former results in an almost complete suppression of R5 HIV-1 while HHV-7 replication is not affected; the latter results in a severe suppression of HHV-7 while replication of X4 HIV-1 is suppressed slightly.

The model system described here is the first to permit the study of HHV-7-induced pathogenesis in the context of human lymphoid tissue under controlled experimental conditions and the comparison of these phenomena with tissue pathogenesis of other lymphotropic viruses. In particular, this model system emphasizes the striking difference in pathogenesis between HHV-7 and the closely related HHV-6. The molecular mechanism of HIV-1 suppression by HHV-7 described here differs from the previously reported mechanisms of interactions of other microbes with HIV, such as measles, which is mediated by upregulation of chemokines. The model for studying interactions of these viruses in the context of human lymphoid tissue provides an opportunity to investigate the mechanisms of this interference under physiologically relevant conditions. Our ex vivo results suggest that HHV-7 may interfere with HIV-1 in lymphoid tissue in vivo, thus affecting disease progression, in particular facilitating a switch of dominance from CCR5-using to CXCR4-using HIV-1, which in turn may affect HHV-7 replication. Knowledge of the mechanisms of interactions of HIV-1 with HHV-7, as well as with other pathogens that modulate HIV-1 replication, may provide new insights into HIV pathogenesis and lead to the development of new anti-HIV therapeutic strategies.

Differentiation of rhesus embryonic stem cells in tissue-like three-dimensional structures

Chen, Margolis; in collaboration with Kleinman, Revoltella, Zimmerberg

Deciphering the mechanism of human microbial pathogenesis in the context of various tissues requires the development of new tissue models. We wished to design various tissue-like structures by differentiating totipotent cells derived from the inner cell mass of pre-implantation rhesus embryos (embryonic stem cells [ESCs]) into particular lineages in three-dimensional structures so that the structures would become targets for various pathogens. To date, we have used several extracellular matrices to differentiate cells in multicellular structures. During embryonic development, the extracellular matrix (ECM) serves as a reservoir for growth factors, provides a substrate for cell adhesion and spreading, provides contact guidance for cell migration, and serves as a scaffold for building tissues.

We focused on the contribution of the physical properties of ECMs, in particular their malleability, to the adhesion and differentiation of rhesus monkey ESCs. We compared embryonic stem cell differentiation, adhesion, and spreading on two types of ECM: three-dimensional bioactive collagen I and Matrigel™ with a biologically inert agarose. We monitored cell differentiation from the expression of specific genes and by the detection of lineage-specific proteins. ESCs readily attach to and spread on type I collagen gels; on Matrigel™, ESCs formed aggregates with round cells while, in the periphery of these aggregates, cells started to spread. ESCs did not attach to agarose and instead remained in suspension-like cultures that facilitated cell-cell adhesion and the formation of large, often hollow aggregates ("cysts"). Under these conditions, ESCs progressed differently. Differentiation was low on collagen gels while Matrigel™ promoted strong expression of endoderm genes, and ESCs grown on agarose expressed liver-specific genes and proteins and differentiated into cardiomiocytes. This lineage differentiation was evident not only from the expression of specific markers and the activation of genes but also from ESCs' functional activity when the ESC aggregates started beating. We also observed differentiation into this same lineage in cell aggregates suspended in a bioreactor. The loss of the round cell shape from increasing cell-matrix spreading is associated with a decline of cardiomyocyte gene expression while recovery of the round cell shape is associated with the upregulation of cardiomiocyte genes. In general, the extent of cell-cell interaction versus cell-substrate adhesion is closely related to the pattern of ESC differentiation. Production of endogenous collagen is essential for ESC survival on both bioactive ECM and inert agarose. The in vitro model developed in the current study may be used as a model to recapitulate and study cardiac and liver developmental processes as well as heart and liver repair and malfunction. Understanding the effects of ECM on the differentiation of ESCs in various types of culture may lead to the design of tissue-like structures by stimulation of ESC aggregates for differentiation into particular lineages, especially endoderm and mesoderm lineages, thereby permitting the study of microbial pathogenesis.

COLLABORATORS

Peter Anton, MD, David Geffen School of Medicine at UCLA, Los Angeles, CA
Hynda Kleinman, PhD, Craniofacial Developmental Biology and Regeneration Branch, NIDCR, Bethesda, MD
Paolo Lusso, MD, PhD, Laboratory of Immunoregulation, NIAID, Bethesda, MD
Ian McGowan, MD, PhD, David Geffen School of Medicine at UCLA, Los Angeles, CA
William Moss, MD, The Johns Hopkins Bloomberg School of Public Health, Baltimore, MD
Robert Revoltella, MD, PhD, Institute of Biomedical Technologies, CNR, Pisa, Italy
Dominique Schols, PhD, Rega Institute for Medical Research, Katholieke Universiteit, Leuven, Belgium
Robin Shattock, MD, St. George's Hospital Medical School, University of London, London, UK
Joshua Zimmerberg, PhD, MD, Laboratory of Cellular and Molecular Biophysics, NICHD, Bethesda, MD

For further information, contact margolis@helix.nih.gov.