KSHV Latency Proteins
Rationale: During latency, KSHV gene expression is limited and only a subset of viral proteins are expressed. These include latent nuclear antigen-1 (LANA-1) encoded by ORF73, viral Cyclin (v-Cyc) encoded by ORF72, viral FLICE inhibitory protein (v-FLIP) encoded by ORF71, Kaposin encoded by ORFK12, and viral interferon regulatory factor 3 (vIRF-3; also known as LANA-2) encoded by ORFK10.5. Because latency predominates in KSHV-associated tumors, elucidation of latent protein structure and function is essential for understanding KSHV persistence and pathogenesis. The KSHV latent genes may also provide ideal targets for anti-tumor and/or anti-viral drugs. Inhibition of latent protein expression and/or function could block essential tumorigenic pathways. For these reasons, resolving the structure of the KSHV latency proteins should have a profound impact on basic and clinical KSHV-related research. A brief outline of each latency protein follows.
LANA-1 (also ORF 73)
Background: LANA-1 is approximately 1,162-aa with some inter-isolate variation due to variation in internal repeat elements. The theoretical molecular mass of 145 kDa yields an apparent size of 220-230 kDa when analyzed by Western blot. There are 3 distinct domains: a C and an N terminal domain that are rich in basic amino acids, and a central hydrophilic domain consisting of multiple tandem repeat elements. LANA-1 provides essential functions in the KSHV life. It tethers the viral episome to the host chromosome to ensure segregation to daughter cells during mitosis, inhibits lytic replication via repression of the lytic switch protein RTA (ORF50), and confers a growth advantage on latently infected cells via modulation of the Notch signaling pathway. LANA-1 has also been reported to modulate the expression/function of other cellular proteins, including cell cycle proteins and those involved in chromatin remodeling and several important signaling pathways. The ability of LANA-1 to replicate and maintain the KSHV episome is essential for establishment of latency and the persistence of the KSHV genome. Determining the structure of the specific LANA domains that bind KSHV DNA and host chromatin will therefore aid our understanding of KSHV persistence and suggest strategies to eliminate latent virus. To date, only the structure of a 23-aa N-terminal peptide bound to nucleosomes has been determined.
v-Cyc (also v-Cyclin; ORF 72):
The structure of v-Cyc has been solved and it will not be considered in this application.
v-FLIP (also ORF 71):
Background: KSHV v-FLIP, a 188-aa protein, is a homolog of cellular FLICE (FADD like IL-1b-converting enzyme) inhibitory proteins (cFLIPs) that block death receptor–induced apoptosis by inhibiting caspase 8 recruitment to death-induced signaling complexes (DISCs) via death effector domains (DEDS). vFLIP contains two DEDs and blocks signaling through death receptors. KSHV encodes two additional apoptosis modulators, vIAP/ORF7 (see below) and vBcl-2/ORF16, but only vFLIP is expressed during latency. In addition to functioning as an anti-apoptotic protein, vFLIP associates with IKK and activates NFkB signaling, a function essential for PEL cell survival and one that no doubt influences several downstream pathways. Other viruses express v-FLIPS to inhibit host cell apoptosis, including several -herpesviruses and the molluscum contagiosum virus, a member of the human poxvirus family (74). Although all of these proteins contain tandem DEDs and hinder recruitment and activation of caspases, the essential mechanisms through which cellular and v-FLIPS inhibit death receptor signaling remain obscure. The crystal structure of the v-FLIP of molluscum contagiosum virus MC159 was recently solved, and revealed unexpected insights into its function. Determining the crystal structure of KSHV v-FLIP should yield insights into mechanisms through which KSHV inhibits apoptosis. Selective blocking of v-FLIP function could thus allow eradication of KSHV-infected cells.
Kaposin (ORF K12):
Background: Kaposin is encoded by the K12 locus, which was first described as encoding a 0.7-kb transcript (T0.7) for a 60-aa hydrophobic protein termed Kaposin. The K12 locus is now known to be more complex, encoding not only T0.7 but also two different upstream GC-rich direct repeat sequences, DR1 and DR2. The number of DR repeats varies between KSHV isolates, yielding transcripts of different sizes. Transcriptional regulation of the K12 transcript is complex and yields at least 3 Kaposin proteins - A, B and C - of which Kaposin B is the most abundant. Kaposin B is encoded only by the upstream sequences, Kaposin A is the product of the original T0.7 transcript, while Kaposin C is a chimera of T0.7 and the upstream repeats. Kaposin A has been implicated in transformation via activation of the ERK/MAPK signaling pathway; Kaposin B has been shown to stabilize ARE-containing cytokine transcripts via activation and association with MK-2. Kaposin B consists primarily of two proline-rich direct repeats termed DR1 and DR2 and binding to MK-2 has been mapped to the N-terminal element of DR2. The region of MK2 that binds to Kaposin B was mapped to residues 200-270 within the C-lobe of the MK2 catalytic domain. Interestingly, DR2 binding to MK2 is insufficient for MK2 activation, and DR1 is required for the RNA-stabilizing activity of Kaposin B. The conservation of DR1 across KSHV isolates also suggests a critical function. To date DR1 binding partners have not been identified and whether DR1 performs functions in addition to cooperating in RNA stabilization remains unknown. We are interested in solving the structure of full-length Kaposin B, as well as DR1 and DR2 multimers, since this should illuminate mechanisms of mRNA stabilization, suggest ways to block pro-inflammatory pathways in KSHV-infected cell, and uncover the function of DR1.
Viral Immune modulators
The ‘K’-proteins of KSHV modulate both the adaptive cellular immune response. Several of these proteins are homologues of host cell proteins that either inhibit or modulate their human homologues, or have adopted additional functions. The structure of several host modulators has been solved previously and these will not be considered here. However, the function Since KSHV does not infect animals, we will also study the structure of proteins of the closely related Rhesus Rhadino-virus (RRV) whenever possible. This will allow us to examine whether eliminating the respective immune modulators will affect disease as expected.
Viral CD200 homologues:
Background: The KSHV and RRV genomes encode homologues of cellular CD200, a membrane glycoprotein that is found on the surface of many cell types including epithelium, neurons, and lymphocytes. Structurally, CD200 contains two immunoglobulin domains, a transmembrane domain, and a very short cytoplasmic domain lacking any identifiable signaling motifs. To elicit an effect, CD200 binds to its receptor, CD200R, which is largely restricted to the surface of cells of myeloid lineage. Ligation of CD200R results in a decrease in the production of type 1 (inflammatory) cytokines such as tumor necrosis factor (TNF), and a shift toward a type 2 (anti-inflammatory) cytokine response from the cells displaying CD200R.The KSHV vCD200 is a 271-aa protein which shares 40% identity with cellular CD200, whereas the RRV vCD200 is a 253-aa glycoprotein sharing 30% and 41% aa identity with human CD200 and KSHV vCD200, respectively. Both the KSHV and RRV vCD200 molecules have been shown to be expressed on the surface of infected cells undergoing lytic replication and both are capable of downmodulating macrophage activation through binding to CD200R on activated macrophages and thereby reducing the level of TNF production. Defining the structure of a vCD200/CD200R complex might thus help to develop compounds that are able to counteract this viral evasion pathway, thus enabling the activation of macrophages that are an essential part of the anti-viral host response. To develop virus-specific compounds, it will be important to solve both the host and the viral CD200.
Viral cytokine and cc-chemokine homologues:
Background: The KSHV and RRV genomes encode homologues of cellular cytokine and cc-chemokines that are widely considered to be viral determinants of pathogenesis, participating in the direct development of lymphoproliferative lesions, in the recruitment of specific cell types to the site of infection or in the development of angiogenesis. First, KSHV encodes a functional homologue of cellular interleukin-6 (IL-6), referred to as viral IL-6 (vIL-6) that is capable of interacting with the cellular IL-6 receptor as well as the gp130 signaling subunit directly in the absence of the IL-6 receptor. Engagement of vIL-6 with gp130 in the presence or absence of the IL-6 receptor leads to induction of JAK/STAT signaling pathway. The RRV encoded vIL-6 is also capable of functioning in the absence of the IL-6 receptor, but shares limited amino acid sequence identity with cellular IL-6 (17.8%) and KSHV vIL-6 (12.7%), respectively. The structure of the KSHV vIL-6-human gp130 complex has been reported and indicates that KSHV vIL-6 utilizes specific hydrophobic interactions to initiate signaling through the gp130 subunit.
The viral macrophage inflammatory protein homologs (vMIPs) of KSHV and RRV possess the classical chemokine “CC” motif and are ~ 25-40% homologous to cellular MIP-1a. While KSHV encodes 3 vMIPs (vMIP-I, -II, and -III), RRV encodes a single MIP homolog (RRV vMIP). Several studies have been conducted on the KSHV vMIPs showing that these proteins bind to an array of chemokine receptors as well as induce a variety of biological functions.
Crystallographic studies have shown that KSHV vMIP-I and -II share the same three dimensional (3-D) structure and possess the typical chemokine tertiary fold. Furthermore, except for some variation in the N-terminus, the 3-D structures of vMIP-I, vMIP-II, RANTES, MCP-3, and Eotaxin are superimposable.
Since the structures of the KSHV vIL-6 and vMIPs have been deduced, we will focus our efforts solely on expression of the RRV homologues of vIL-6 and vMIP, as deduction of these divergent, yet related molecules could aid in the development of novel small compounds to inhibit the unique mechanisms utilized by the viruses and not the host proteins.
Viral B cell receptor homologues:
Background: The KSHV and RRV genomes both encode a unique ORF at the far left end of their genomes that are structurally similar to the B cell receptor (BCR). The cytoplasmic tail of both molecules contain an immunoreceptor tyrosine-based activation motif (ITAM) which has been shown to be capable of activating signaling pathways similar to those activated by the BCR complex in B lymphocytes. However, unlike the BCR, the KSHV ORF K1 and RRV ORF R1 proteins are thought to be constitutively active through an oligomerization process that is leads to phosphorylation of the ITAM and recruitment of the major B cell kinase, Syk, which is consistent with B cell survival. The structures of the KSHV K1 and RRV R1 proteins have not been deduced and are important molecules to for this contract, as they both possess oncogene-like properties.
Viral interferon regulatory factors:
Background: Interferon-regulatory factors represent the first line of the innate host response to viral infection since they are activated by signal transduction cascades triggered by pattern-recognition receptors. Structural analysis of IRF3 revealed how virus-induced phosphorylation induces a conformational change in IRF3 which allows the association with host transcription factors. As a result IRF3, and the similar protein IRF7, induce the transcription of interferons which induce an anti-viral response in neighboring cells by transcriptionally upregulating dozens is not hundreds of host cell genes. The importance if this innate host response is also stressed by the fact that many viruses have developed countermeasures against activating IRF3, as well as other IRFs. However, while the structure for several host IRFs has been solved, the structure of viral IRFs are unknown. KSHV and RRV encode four or eight, respectively, homologues of IRF. These homologues have been shown to both inhibit as well as promote IRF function. The structural analysis of these viral homologues might thus help to explain their various functions. Moreover, this approach might allow the development of inhibitors of IRF function that are specific for viral IRFs.
Viral Inhibitor of Apoptosis vIAP (ORF K7):
Background: K7 is a viral homologue of the cellular protein survivin-deltaX3, a member of the inhibitor of apoptosis (IAP) family. K7 binds to cellular Bcl-2 via its BH-2 domain and to caspase 3 via its BIR-domain. K7 is a relatively small protein that occurs in two forms: a 19 kd N-glycosylated form and a 16kd non-glycosylated protein. K7 contains a single-transmembrane domain and a highly hydrophobic N-terminus that contains a mitochondrial-targeting signal. K7 locates both to the ER and to mitochondria. It forms both homo-dimers and heterodimers with cellular Bcl-2 via its BH2 domain. In addition, K7 binds to caspase 3 via its BIR-domain. Thus, K7 is an adaptor protein enabling anti-apopotic proteins (Bcl-2) to block effector caspases. Additionally, K7 binds to calcium-modulating cyclophilin ligand (CAML), a protein that regulates the intracellular Ca2+ concentration. K7 is unique to KSHV and does not have a homologue in RRV. The structure of K7 has not been solved. While the structure of K7 is not known, the structure of its human homologue survivin has been solved by X-ray (reviewed in and NMR. Therefore, we propose to solve the structure of K7 alone as well as in complex with the cellular proteins Bcl-2, caspase 3 and CAML. K7 was also found to interact with both K3 and K5 in a yeast-two hybrid screen. Since K3 and K5 are RING-CH-ubiquitin ligases (see below), this could indicate that K7 functions in part by interacting with either viral or cellular members of this ubiquitin ligase family. Importantly, K7 and its human homologue survivin are the only members of the IAP-family that do not contain a RING-domain. Finally, the host cell protein PLIC1 (protein-linking integrin-associated protein and cytoskeleton 1), also known as ubiquilin was shown to interact with K7. PLIC1 forms a dimer and binds to poly-ubiquitinated proteins via its carboxyterminal ubiquitin-associated domain (UBA). Taken together with the finding that K7 interacts with RING-CH type ubiquitin ligases, it is possible, that a trimeric complex consiting of K5 or K3, PLIC1 and K7 is formed. A co-crystal structure of K7 with ubiquilin, and potentially also with K3 or K5, is therefore expected to reveal how K7 might regulate the function of these cellular and viral proteins.
Viral Modulators of Immune Responses vMIR (ORFs K3 and K5):
Background: K3 and K5 are type III transmembrane proteins that are homologous to each other and related to ORFs in some of the other gamma-2 herpesviruses (but not Rhesus Rhadinovirus) as well as in some poxviral genomes. The NMR-structure of the amino-terminal fragment of K3 demonstrated that it represents a RING-domain consistent with the demonstration that this domain catalyzes the formation of poly-ubiquitin in vitro similar to RING-E3s. Thus, the K3-family represents a type III transmembrane ubiquitin ligase family. Interestingly, related protein are found in the human genome, a protein family termed membrane-associated RING-CH (MARCH) proteins. Members of the viral and cellular protein families share the ability to target ubiquitin to the cytoplasmic tail of subsets of transmembrane proteins which, depending on the intracellular site of ubiquitination, results in their destruction either by proteasomes or ubiquitin-mediated endocytosis and targeting to lysosomes. KSHV, as well as other viruses, “use” this protein family to evade detection by T cell, NK cells and NKT cells by downregulating their respective receptors on virally infected cells, namely MHC class I, B7.2, ICAM-1 and CD1. Thus, the virus becomes “invisible” to the cellular immune system. Animal models for poxviruses (Myxomavirus) and gamma-2 herpesviruses (murine herpesvirus 68) clearly showed that K3-family proteins are important for viral virulence or for the establishment and maintenance of viral latency. While the function of the human homologues is currently unknown, recent observation render it highly likely that MARCH proteins regulate MHC class II expression during dendritic cell maturation.
The development of specific inhibitors for proteins of this ubiquitin ligase family therefore has potential implications not only for the development of a novel type of antivirals that would expose the hidden virus to the immune system, but might also have applications as immune regulator of dendritic cell function. Ubiquitin ligases are a recognized novel class of drug targets. Solving the structures of members of the K3-protein family will be an essential first step in designing molecules that inhibit their function or interaction with cellular partners.