Grouppe Kurosawa: HIV - The Biology of TAT (print)

The Biology of TAT

TAT is a transactivator protein that is generally considered essential for the massive initial output of virus that is thought to enable HIV mutational variants to outpace and overwhelm the immune system. This leads to chronic infection, the eventual destruction of the immune system and progression to AIDS. During acute HIV infections, TAT is released from infected cells. In its extracellular form, TAT stimulates HIV gene expression, the growth of cells derived from Kaposi sarcomas, angiogenesis and the promotion of tumor metastasis, the development of lymphoid hyperplasia, the secretion of TGFb, TNF alpha and beta, IL-6, the malignant transformation of keratinocytes, the inhibition of IL-2 and IL-2 receptor gene expression, and the inhibition of the anti-viral alpha/beta interferon system. The cell growth-promoting activity and the virus-transactivating effects of extracellular TAT are mediated by different pathways. The accompanying essay, The Folly of Modeling HIV Vaccine Efficacy in Chimpanzees, discusses additional effects of TAT on immune responsiveness.

The TAT protein has two functional domains that are of biological significance to the pathogenesis of AIDS. The amino terminal domain contains a cluster of cysteines and basic amino acids that allow TAT to accumulate in the nucleus. The carboxyl terminal domain contains an RGD amino acid sequence like fibronectin, which allows TAT to bind various membrane integrin proteins. When secreted TAT is taken back up into cells, the cell attachment site is the basic sequence in the amino terminal domain. There are two major receptor sites on membranes that bind the amino terminal domain of TAT. The first is a low affinity sulfated proteoglycan site(s) that is involved in the endocytosis of TAT and stimulation of HIV replication. The sulfated proteoglycan that regulates endocytosis is a rapidly turning over molecule under the synthetic control of TGFb. In the presence of TGFb, TAT may have an enhanced ability to enter infected and uninfected cells. The second site is 100 to 1000 times more sensitive to TAT and is involved in the stimulation of cell growth.

The high affinity membrane binding site for TAT has recently been identified as the FLK-1/KDR receptor for vascular endothelial growth factor (VEGF), a tyrosine kinase that stimulates both the activation and proliferation of endothelial cells and vascular permeability. TAT is known to activate endothelial cells and to be a powerful angiogenic growth facior. Expression of TAT in transgenic mice induces Kaposi sarcoma-like lesions, squamous cell papillomas and carcinomas, adenocarcinomas of skin adnexa glands, and B-cell lymphomas. Very small amounts of extracellular TAT (0.1 ng/ml) may mimic VEGF by activating its receptor. This would explain a number of AIDS-related dysfunctions associated with endothelial cells, such as Kaposi's sarcoma, arteriopathy, and intravascular coagulopathy or hypercoaguloability of the blood. Increased blood levels of von Willebrand factor, a protein released from activated or damaged endothelial cells, negatively correlates with the CD4 T cell count and positively correlates with disease progression towards AIDS. In addition to endothelial cells, VEGF is a chemotatic factor for monocytes, and an inhibitor of the functional maturation of dendritic cells. Dendritic cells, as previously discussed, are antigen processing cells that are critically important for activating naïve T cells during the primary immune response. Less differentiated, therefore less functional, dendritic cells are commonly found in HIV associated follicular hyperplasia, and are associated with changes in the B cell population resulting in an inverted centroblast to centrocyte ratio. Centroblasts are the sites of somatic mutation, the mutational process that initiates the development of high affinity antibodies. In the absence of proper centrocyle development, however, B cells harboring high affinity antibodies will not be selected for survival. Both NEF and TAT stimulate hypergammaglobulinemia, but this does not imply the secreted antibodies are high affinity or biologically useful in neutralizing virus or viral proteins. Both VEGF and TAT inhibit apoptotic cell death in a variety of different cell types, and VEGF is an autocrine growth factor for Kaposi's sarcoma cells. It is secreted from KS cell lines and is probably responsible along with TAT for the enhanced vascular lesions and enhanced vascular permeability associated with Kaposi sarcoma. The viral TAT protein truly functions as a protein hormone, which enables it to be biologically active in very low serum concentrations.

If VEGF and TAT function in a parallel manner by stimulating the VEGF receptor, TAT may mimic VEGF as a vascular permeability factor and inhibitor of dendritic cell development. VEGF is 50,000 times more potent as a vascular permeability factor than histamine. This is an important point because clinical AIDS does not technically develop until HIV infected cells cross the endothelial barrier, including the blood-brain barrier, and enter the interstitial spaces. Glucosteroids and TGFb both antagonize the cellular transmigration of immune cells across endothelial cells into tissues. This is one of the pathways by which these hormones control inflammation. The degree of TAT or VEGF induced vascular permeability may be a significant factor in determining the length of the asymptomatic phase of AIDS. TAT mimics VEGF and VEGF is released from TAT induced tumors, and activated endothelial cells and monocytes. The massive vascular permeability known to be induced by VEGF and possibly TAT has prompted us to reevaluate the conclusions of Ho, et.al. that the dramatic upturn in CD4 T cell counts after only a few weeks of protease inhibitor therapy is due to a rapid proliferation of normal CD4 cells. Their conclusions are based on a statistical analysis of the changes in CD4 T cell counts after protease inhibitor therapy. However, telomere length studies on the CD4 cells of HIV and uninfected individuals demonstrate no differences in CD4 T cell proliferation rates. The origin of the rebounding CD4 cell population after inhibitor therapy has tremendous clinical significance. Ho and others believe the increased CD4 cell count is due to the proliferation of fresh, uninfected cells. We believe the rebound effect is primarily due to a block in CD4 cell egress out of the blood, and to an enhanced expression of CD4 on the membranes of infected and non-infected cells. The cells already in the interstitial spaces will return to the blood via the thoracic duct, and this will RAPIDLY increase the CD4 T cell count. In the presence of HIV protease inhibitors, compounds that are also known to be powerful inhibitors of the proteasome complex, many virally infected cells will die from apoptosis while others will stop producing virus. The effects of protease inhibitors on productively infected cells is controversial and deserves further scientific scrutiny. Powerful, selective proteasome inhibitors such as epoxomicin inhibit the secretion of HIV virions from infected cells, and no doubt also induce apoptosis of viral reservoirs. In the absence of viral synthesis, TAT will no longer be synthesized. VEGF synthesis will decrease as will vascular permeability. To our knowledge, vascular permeability has never been previously proposed as a contributing factor to the decreased CD4 T cell counts (or increasing CD4 T cell counts associated with protease inhibitor therapy) associated with AIDS.

There are 1000 times more endothelial cells than circulating mononuclear cells in the body. Activated, proliferating endothelial cells can be infected with HIV, but they do not secrete virus unless further stimulated with the pro-inflammatory hormones IL-lb and TNFa. Activated endothelial cells express MHC class II molecules and act as antigen presenting cells in supporting T cell proliferation and IL-2 production. As a reservoir for virus, endothelial cells are capable of readily passing the HIV virus to the T cells they have just activated. Endothelial cells can also activate HIV gene transcription in latently or chronically infected monocytes. Interestingly, TNFa, which we feel is relatively absent during the asymptomatic phase of AIDS, is a potent inhibitor of VEGF receptor gene expression. In the absence of TNFa, endothelial and dendritic cells may be particularly sensitive to the presence of low concentrations of TAT and VEGF.

Although the carboxyl terminal portion of TAT appears to play no role in the endocytosis of extracellular TAT, the RGD sequence of TAT is not without biological significance. TAT modulates fibronectin expression in thymic epithelial cells and impairs in vitro thymocyte development. CD4-8- thymocyte precursors have a decreased ability to differentiate into CD4+8+ cells when the epithelial stromal cells are transfected with TAT-expressing plasmids. These cells are known to be productively infected with the HIV virus in vivo. The effects of TAT on thymocyte development may be TGFb-dependent.

TAT-mediated amplification of HIV transcription is absolutely dependent on the transcription factor NF-kB. TAT is not a DNA-binding transcription factor. TAT binds a stem loop structure in the viral RNA and increases the efficiency of transcriptional elongation. TAT binds a similar structure in the RNA of TNFb, which accounts for its ability to induce the transcription of this gene in transfected or infected lymphoid cells. TAT synergizes with TNFa in stimulating the transcription of HIV by binding a receptor site on the exterior membrane of infected cells. If this binding is blocked by anti-TAT antibodies, TNFa has a substantially reduced ability to activate NF-kB and subsequent HIV transcription. TAT may promote the activation of NF-kB by directly or indirectly inducing the depolymerization of microtubules, cytoskeletal proteins that indirectly control the degradation of IkB, the NF-kB inhibitor.

TNFa/b has substantial anti-viral activity, and synergizes with interferons in the induction of resistance to both RNA and DNA viral infections. TAT antagonizes the interferon response to viral infections in two ways. First, TAT binds the HIV TAR double stranded RNA stem loop preventing it from activating interferon inducible enzymes. Second, TAT directly binds the interferon-induced, double-stranded RNA-dependent kinase, PKR. This interaction inhibits the activity of PKR and prevents a PKR-induced inhibition of protein synthesis in infected and uninfected cells. As secreted TAT is readily taken up into uninfected cells, TAT may prevent interferon from maximally inducing the activity of PKR and establishing an anti-viral state in uninfected cells.

Anti-TAT antibodies naturally exist in people uninfected with the HIV virus. These antibodies bind the basic amino acid sequence in the amino terminal portion of the protein. When an individual becomes infected with HIV, these natural antibodies gradually disappear from the blood. There is a very strong inverse relationship between the presence of natural anti-TAT antibodies and p24 antigen in the blood. Monoclonal antibodies against TAT inhibit the replication of virus in culture. Natural anti-TAT antibodies may therefore play an important role in protecting against the spread of virus in the early stages of infection. The decrease in natural anti-TAT antibodies could be an indirect consequence of the generalized breakdown in the germinal center selection process for high affinity antibodies. Interestingly, passive immunization with sera from asymptomatic HIV+ persons reduces HIV-1 viremia levels in AIDS and ARCS patients to levels undetectable by PCR. If these sera contained high titers of anti-TAT antibodies, whether natural or developed as a consequence of infection, it would provide further support for our conclusion that TAT is an absolutely necessary component of an HIV vaccine.