INTERACTION OF NEF WITH THE HOST T CELL
Nef Structure and Interaction with Cellular Proteins
Nef is a small viral regulatory protein of approximately 27 kDa that is expressed abundantly in the early stages of viral replication.
In fact, it is expressed even before the virus is integrated into the genome.6 Nef is posttranslationally modified by the irreversible attachment of myristic acid to its N-terminus, which targets Nef to the cellular membrane. So far, it seems that membrane targeting is absolutely essential for most, if not all, functions of Nef.Nef achieves its biochemical effect by requisite interactions with cellular components. More than 30 putative Nef targets have been reported, and the number is still growing.7 Structurally, HIV-1 Nef has three distinct features. The first 70 amino acids constitute a genetically diverse and structurally flexible N-terminal arm that is followed by a well-conserved and folded core domain of approximately 120 residues. Out of the core domain projects a 30-amino-acid loop, which is again highly flexible. Only HIV-2 and SIV have an additional C-terminal tail of 10 to 30 amino acids. The core domain is the only part of Nef that adopts a stable tertiary fold. Its high-sequence conservation in all Nef alleles implies that this region has the same structure in HIV-2 and SIV Nef. Each of the described domains has interaction sites for the binding and interaction of and with cellular proteins. For example, mutational analysis of the core domain identified several residues relevant for the binding of SH3 domains of src kinases and the Pak kinase. Also, the C-terminal flexible loop contains three defined protein motifs that interact with components of the endocytosis machinery, an example of which would be the regulatory subunit of the v-ATPase.8 In addition to these better-defined interactions, most interacting proteins associate with Nef in the context of one or more protein complexes.
In these instances, several domains of Nef seem to be involved in the association with those complexes. For example, the N-terminus of Nef interacts with a protein complex that contains Lck and PKC and the Polycomb protein Eed.9 Surprisingly, domains in the far C-terminus contribute to the binding of this complex. Some interactions cannot occur simultaneously, as their binding sites on Nef are overlapping. For example, the association of Nef with Pak, the human thioesterase, involve the same hydrophobic patch on the core domain, forming a collar around F121.Given the dynamic interaction of Nef with its cellular environment, it is likely that Nef is sequentially directed to distinct cellular locations and different sets of interacting effectors. Therefore, Nef may be viewed as an adapter protein, assembling and connecting several protein and signaling complexes in a timely, defined fashion. The fairly extensive unstructured regions of Nef provide a vast accessible surface for serving this function.7
Interaction of Nef with the Endocytosis Machinery
The accelerated endocytosis and lysosomal degradation of CD4 was one of the first functions attributed to Nef.10 Meanwhile, additional factors that are required were identified. These include cellular proteins generally involved in trafficking, such as adapter protein complexes (AP-complexes), B-COP, the vacuolar ATPase, thioesterase, and the PACS-1 protein.11 To recruit CD4 into the endocytotic pathway, Nef binds CD4 and acts as an adapter between the receptor and components of clathrin-coated pits. This involves the interaction of Nef with AP-complexes and the regulatory unit of the vacuolar proton pump of the v-ATPase.12 The precise meaning of the latter interaction is not obvious and awaits further clarification. Subsequently, Nef binds to the -subunit of COPI coatomers (β-COP) to direct CD4 to a degradation pathway.13
In subsequent studies, the downregulation of other surface receptors was reported, including CD2814 and CD3 (only SIV).15 However, there are more receptors that are affected (O.
Keppler, personal communication). Functionally, it was shown that downregulation of CD4 increases viral infectivity by preventing the formation of complexes between gp120 and CD4 on viral particles, as well as intracellularly.16,17 It was further argued that CD4 downregulation may also prevent reinfection (also called superinfection), in order not to overload the biosynthetic capability of the cell.18 On the other hand, it is known that HIV genomes recombine within an infected cell; therefore, the latter argument is on shaky grounds. Taken together, CD4 downregulation by Nef may have an additional role that has not yet been described. Unlike CD4, other receptors that are downregulated do not bind directly to Nef, perhaps with the exception of CD3 in SIV Nef. Therefore, the rather promiscuous downregulation of a seemingly unrelated set of surface receptors may point to a general interference with trafficking pathways that is not fully explained or understood until now.Surprisingly, Nef uses a different mechanism and also different protein domains to downregulate another cell surface protein: the major histocompatibility complex (MHC) class I, A and B.19 The mechanism was recently described in more detail. Nef binds sequentially a PACS-1/AP 1 complex and PI3 kinase before reaching the plasma membrane. There, the ARF6 endocytic pathway is activated, leading to increased downregulation of MHC class I. In addition, Nef seems to block the recycling of MHC molecules back to the cell surface, which ultimately leads to an accumulation of MHC complexes in the trans-Golgi network (TGN).20 Downregulation of MHC class I by Nef is not as efficient as CD4 downregulation and seems to require higher expression levels of Nef. The effect, however, has been shown to be sufficient to diminish recognition of HIV-infected cells by cytotoxic T lymphocyte (CTL).21 Therefore, it is a classical viral immune escape mechanism22 (see below).
In summary, at least the downmodulation of MHC class I by Nef clearly is a function that increases viral particle production.Interaction of Nef with Signaling Proteins of the T Cell
Receptor (TCR) Environment
The ability of Nef to activate T cells was originally described in 1994 using Jurkat cells stably transfected with a CD8-Nef chimeric protein.23 Cells that had higher concentrations of CD8-Nef at the membrane induced tyrosine phosphorylation and upregulation of activation markers such as CD69. In addition, these cells underwent activation-induced apoptosis, indirectly confirming that Nef activated these cells. In 2001, these results were confirmed by gene expression profiling of inducible T cell lines, showing that Nef and anti-CD3-mediated T cell activation overlap by 97%.24 As much as these results imply that Nef activates T cells, there are recognizable differences between T cell receptor (TCR)-mediated activation by antigen or antibodies and Nef expression with respect to the phenotype of activated T cells (see below). The use of tumor cell lines, rather than the primary cells, and chimerical proteins may mask subtle differences of classical T cell activation and Nef- induced signaling. Hence, we still do not know which T cell signaling pathways are, and which are not, activated by Nef.
On the other hand, the ability of Nef to induce T cell activation pathways was confirmed by a number of publications, showing the direct interaction of Nef with both the T cell receptor and its immediate downstream effectors.25 Functional as well as binding studies analyzed the interaction of Nef with the T cell receptor -chain26 and proteins of the T cell receptor environment, including adaptor proteins Vav27 and LAT,5 the tyrosine kinase Lck,9 the serine kinase Pak,28 PKC,29 the DOCK2-ELMO1 complex,30 the map kinases ERK1 and ERK2,31 and membrane micro domains (rafts).32 Although the picture is still not clear, these reports suggest that Nef is part of and probably acts through a T cell receptor-associated multiprotein complex.25 In line with these findings, development of an AIDS-like disease in an HIV transgenic mouse model correlated with Nef- mediated activation of mouse T cells.5 How Nef activates T cells is not known.
Because it has no intrinsic enzymatic activity, it likely functions as an adapter protein. By binding to signaling molecules of different compartments/lipid-rich micro domains (rafts) and possibly by forming oligomers, Nef may function as an intracellular cross-linker.From the sheer number of interacting molecules in the Nef signaling complex, it may be assumed that several distinct signaling pathways are activated, similar as seen, for example, when the T cell receptor complex is activated through a cross-linking antibody. Although detailed individual pathways have not yet been sufficiently documented, at least three different functional consequences of Nef signaling have been found:
1. Nef signaling and viral replication: Nef signaling or Nef-mediated T cell activation seems to perfectly fill the needs of HIV, as one of the very early findings was that T cells had to be activated for HIV replication to start.33 T cell activation leads to the nuclear translocation of transcription factors such as NFAT and NFkB, probably priming the viral promoter or establishing a basal viral transcription that would lead to the expression of more Tat protein.34 Upon infection of resting primary T cells, HIV does, in fact, express Nef, even before the virus is integrated,6 but viral replication remains very low or not detectable. Conversely, stimulation of the T cell receptor by antibodies potently activates replication. Therefore, Nef- and antibody-mediated T cell receptor activation clearly differs, at least with respect to viral replication. In other words, Nef- mediated T cell activation does not start viral replication. On the other hand, there is little doubt that Nef supports HIV replication. This was shown in a number of studies using primary cells. Particularly revealing were experiments using co-cultures of resting T cells and immature dendritic cells.35-37 Immature dendritic cells replicate HIV at a very low level. However, upon co-culture with resting T cells, a significant increase of viral replication is observed in the T cell compartment.
This, however, requires the presence of a functional Nef gene. These results point to a role for dendritic cell surface receptors as co-stimulators of viral replication. Therefore, besides Nef, a second, thus far not characterized, signal (receptor, cytokine, or chemokine) is required to start HIV replication.2. Nef signaling and cytoskeleton rearrangement/chemotaxis: More recently, at least two publications have demonstrated in detail that Nef signaling leads to cytoskeleton rearrangement in the infected cell. Both reports analyzed the Nef-induced pathways leading to this effect, but came up with different results. Fackler and colleagues27 found that the PxxP domain of Nef binds and activates the Vav protein, leading further downstream to the activation of Racl and the Pak kinase. The authors speculated that this Nef effect could play a major role in viral morphogenesis and budding, similar to that seen with other viruses and bacteria. This mechanism would basically lead to an increased release of viral particles. Janardhan and colleagues30 confirmed that Nef activates Racl but suggested a different mechanism. They found that Nef associates with the DOCK2- ELMOl proteins and showed that this complex is required to activate downstream Racl and the Pak kinase. They further found that the chronic activation of DOCK2-ELMO1 leads to impaired chemotaxis of Nef-expressing cells. They speculated and suggested that this would affect the ordered migration patterns of infected T cells and, therefore, the generation and maturation of the immune response to viral antigens. Although both studies come to different conclusions, it seems clear that Nef activates a pathway generally associated with cytoskeleton rearrangement and chemotaxis. How these molecular mechanisms support the viral life cycle remains to be elucidated.
3. Nef signaling and immune evasion/antiapoptosis: The third consequence of Nef signaling in T cells is the stimulation of antiapoptotic and, possibly later in the viral life cycle, proapoptotic signaling pathways. To date, only the antiapoptotic mechanisms are well described. A logical consequence of T cell signaling from the T cell receptor is in the understandable interest of the virus. Much less defined are the proapoptotic effects of Nef, as will be discussed below. In addition, Nef signaling from the T cell receptor leads to the upregulation of Fas ligand, which may be the only real but indirect proapoptotic signaling function of Nef. It is most likely a mechanism of viral immune evasion (see below).