Human Tumor-Associated Viruses and New Insights Into the Molecular Mechanisms of Cancer

Home , B virus

Oncogene (2009) 27, S31–S42 & 2009 Macmillan Publishers Limited All rights reserved 0950-9232/09 $32.00 www.nature.com/onc REVIEW Human tumor-associated viruses and new insights into the molecular mechanisms of cancer

D Martin and JS Gutkind

Cell Growth Regulation Section, Oral and Pharyngeal Cancer Branch, National Institute of Dental and Craniofacial Research, National Institutes of Health, Bethesda, MD, USA

The study of acute-transforming retroviruses and their It took 40 years to isolate and characterize the actual oncogenes and of the multiple mechanisms deployed by infectious agent, the Tobacco mosaic virus (Lecoq, DNA viruses to circumvent the growth-suppressive and 2001). However, Ivanovski’s operational definition of a proapoptotic function of tumor suppressor genes has virus as an agent capable of transmitting an infection provided the foundation of our current understanding of following filtration provided a simple strategy that cancer biology. Unlike acute-transforming animal viruses, became the workhorse of early virologists, which was however, human tumor-associated viruses lead to malig- used extensively to test whether filtrates from disease nancies with a prolonged latency and in conjunction with tissues or body fluids were able to transmit pathological other environmental and host-related cooperating events. conditions. This approach enabled the Italian physician The relevance of viral infection to human cancer develop- Giuseppe Ciuffo to identify the viral etiology of human ment has often been debated. We now know that at least warts in 1907 (Ciuffo, 1907), a discovery that was not six human viruses, Epstein–Barr virus (EBV), hepatitis B fully appreciated by his contemporaries and remained in virus (HBV), hepatitis C virus (HCV), human papilloma the dark until the human papillomaviruses were virus (HPV), human T-cell lymphotropic virus (HTLV-1) associated with human cancer in the 1970s. Using this and Kaposi’s associated sarcoma virus (KSHV) contribute procedure, Peyton Rous at the Rockefeller Institute to 10–15% of the cancers worldwide. Hence, the op- showed in 1911 that a solid tumor—a spontaneous portunity exists to fight cancer at the global scale by spindle cell sarcoma arising in chicken in a nearby preventing the spread of these viruses, by the development farm—could be readily transmitted to other healthy and distribution of effective and safe antiviral vaccines, chickens using tumor filtrates (Rous, 1911). This and by identifying their oncogenic mechanism. Here, we observation led to the identification of the Rous discuss the molecular events underlying the neoplastic sarcoma virus (RSV) as the first tumor-causing virus potential of the human tumor-associated viruses, with (Rous, 1910). RSV belongs to the same Retroviridae emphasis on the enigmatic KSHV and its numerous virally family than the avian leukemia virus (ALV) (Vogt, hijacked proangiogenic, immune-evasive and tumor-pro- 2009), which the Danish scientists Ellerman and Bang moting genes. The emerging information may facilitate had shown to transmit chicken leukemia (Ellermann and the development of new molecular-targeted approaches to Bang, 1908). Leukemia was not regarded as a malig- prevent and treat virally associated human malignancies. nancy at that time, and hence ALV was not considered a Oncogene (2009) 27, S31–S42; doi:10.1038/onc.2009.351 tumor-causing virus (Epstein, 1971). In spite of the simplicity, elegance and reproducibility of Rous’s Keywords: cancer; signal transduction; AIDS malignan- findings, his work was fiercely debated and argued cies; G proteins; viral oncogenes against by a scientific community that was likely not prepared to grasp the paramount importance of his observation. Ultimately, Rous was awarded the Nobel Prize 55 years later, in 1966, as his discovery set the Introduction foundation of our current understanding of experimental and human cancer biology. 0 The discovery of viruses as infectious agents can be Retroviruses share a common genomic structure 5 0 directly linked to the development of filtration devices LTR-gag-pol-env-LTR 3 . The viral gag gene encodes removing all cells and bacteria in the late 19th century. for structural proteins necessary for virion formation, Building on this technical advancement, the Russian and pol encodes several enzymatic activities, including biologist Dmitri Ivanovski identified in 1892 a new class the critical reverse transcriptase, whereas env encodes of infectious agents causing mottling in tobacco plants. envelope proteins involved in host cell recognition and infection (Coffin et al., 1997). The long terminal repeats display transcriptional promoting activity and are Correspondence: Dr JS Gutkind, Oral and Pharyngeal Cancer Branch, required for the random integration of the retrotran- National Institute of Dental and Craniofacial Research, National Institutes of Health, 30 Convent Drive, Building 30, Room 211, scribed viral sequences into the host DNA. Intense Bethesda, MD, 20892-4330, USA studies revealed that unlike ALV, RSV included a E-mail: [email protected] nonessential extra segment in its 30 end, encoding a gene Human viral-associated malignancies D Martin and JS Gutkind S32 termed src for its ability to induce sarcomas, which was including nasopharyngeal carcinoma, gastric carcinoma, thus considered the first bona fide oncogene (Duesberg almost half of the cases of Hodgkin’s lymphoma and B- and Vogt, 1970). This information was exploited by cell lymphoma in immunocompromised patients (Javier Bishop, Varmus and their colleagues, who, using a and Butel, 2008; Parkin, 2006) (Table 1). clever DNA-hybridization strategy, identified avian EBV infection immortalizes B lymphocytes causing cellular sequences with homology to the RSV-encoded polyclonal proliferation, which could result in the escape src. The avian gene related to v-src was termed c-src,and of some transformed clones, but the exact mechanism by in so doing, they described the first cellular counterpart of which EBV causes lymphoproliferation is far from being a cancer-causing gene or proto-oncogene (Stehelin et al., completely understood. A virally encoded six-span 1976). This finding also provided a novel conceptual transmembrane protein, latent membrane protein 1 framework, as it suggested that human cancers might arise (LMP1), appears to mediate some of the proliferation from the aberrant expression or alterations in normal signals driving the B-cell population expansion (Arva- human genes, thus setting the stage for contemporary nitakis et al., 1995). LMP1 acts as a constitutively active molecular oncology. Indeed, many other transforming receptor that mimics activated CD40, a member of the retroviruses were identified whose oncogenes encode tumor necrosis factor receptor family (Mosialos et al., altered versions of some of the most frequently mutated 1995). The cytoplasmic carboxyl terminus of LMP1 has human cancer genes (Weinberg, 1997). a critical role in EBV-induced B-cell transformation through its binding to a tumor necrosis factor receptor- associated factor and the tumor necrosis factor receptor- associated death domain protein (Brown et al., 2001). Human tumor-associated viruses This, in turn, induces the activation of several key signaling molecules such as PI3K, JNK and JAKs Animal acute-transforming retroviruses and their en- leading to the activation of transcription factors coded oncogenes were the center of attention in cancer including NF-kB, AP-1 and STATs (Kilger et al., biology in the 1970s and 1980s, as they enabled a wealth 1998), all of which have been extensively related to of discoveries on the molecular mechanisms involved in human malignancies. cellular transformation. In addition, DNA viruses such Despite the ability of LMP1 to induce B-cell as the Murine Polyoma Virus and Simian Vacuolating transformation, it is not expressed in Burkitt’s lympho- Virus 40 (SV40) were shown to cause tumors in ma cells. In that regard, the proposed role of EBV in experimental animals and to transform human cells in Burkitt’s lymphoma was initially met with great skepti- culture (reviewed in Howley and Livingston, 2009). cism (Epstein, 2001). Finding EVB linked to a particular However, detractors of the field of viral oncology at that African tumor, it was quite surprising when ensuing time argued that none of these viruses had been related epidemiological studies showed >90% seropositivity in to human cancer, as attempts to identify viruses causing adults worldwide, making this fact initially one of the human malignancies had been unsuccessful. That was major arguments for the detractors of EBV-induced going to change dramatically just a few years later when tumorigenesis (Epstein, 2001). Nowadays, however, it is the first human tumor viruses were identified. We now widely accepted that EBV has a central role in African know that at least six viruses contribute to human Burkitt’s lymphoma, Hodgkin’s lymphoma and naso- malignancies (Table 1). Indeed, 10–15% of the cancers pharyngeal carcinoma, the latter a highly aggressive worldwide, or over 1.3 million cases of cancer each year cancer type that is quite prominent in Southeast Asia resulting in almost a million deaths, can be linked to (Young and Rickinson, 2004). This disparity between viral infection (Parkin, 2006). In this study, we will the widespread EBV infection and the delayed and low discuss the emerging information on the molecular incidence of infected individuals developing cancer, such events underlying the neoplastic potential of these as Burkitt’s lymphoma, led to the current view that human tumor-associated viruses. genetic or environmental factors (for example, parasitic infections such as malaria, nutrition, maturity and status of the immune system and translocation of myc that is Epstein–Barr virus frequently found in Burkitt’s lymphoma cells) also have a crucial role in the development of this disease In 1958, Denis Burkitt first described a childhood B-cell (Lombardi et al., 1987). This example illustrates a malignancy in Uganda, now known as Burkitt’s general concept in tumor virology, which is that human lymphoma, which he suspected to have a viral etiology oncogenic viruses often lead to malignancies with a closely associated with malaria (Burkitt, 1962). In 1965, prolonged latency and in conjunction with other Tony Epstein and Yvonne Barr identified herpesvirus environmental or host-related cooperating events. particles by electron microscopy in Burkitt’s lymphoma cells. This virus was distinct from other previously described herpesviruses and was named the Epstein– Barr virus (EBV) or human herpesvirus 4 (HHV-4) Hepatitis B virus (Epstein et al., 1965). Currently, we know that EBV is associated with a human benign disease, infectious A viral etiology for certain forms of hepatitis had been mononucleosis and with multiple human malignancies, suspected for years because of the infectious patterns of

Oncogene Human viral-associated malignancies D Martin and JS Gutkind S33 Table 1 Human tumor-associated viruses Virus Family Cancer type Cases per Mechanismb Oncogenes Oncogene function References yeara

Hepatitis B Hepadnaviridae Hepatocellular 340 000 Chronic X protein Molecular-signaling (Ganem and Prince, virus (HBV) carcinoma inflammation dysregulation 2004; Guidotti and inhibition of p53 Chisari, 2006; Kao and Chen, 2002) Hepatitis C Flaviviridae Hepatocellular 195 000 Chronic None — (Colombo et al., virus (HCV) carcinoma inflammation 1989; Thomas et al., 2000) Epstein–Barr Herpesviridae Burkitt’s lymphoma, 113 400 Oncogenic LMP-1 Molecular signaling (Arvanitakis et al., virus (EBV) Hodgkin’s lymphoma, dysregulation 1995; Brown et al., (HHV-4) Post transplantation NF-kB activation 2001; Mosialos et al., lymphoma, lymphoproliferation 1995) Nasopharyngeal carcinoma Human Papilloma Cervical cancer, 561 180 Oncogenic E6/E7 Inhibition of p53 (Beaudenon et al., papillomavirus viridae Anal cancer, and Rb Cell 1986; Dyson et al., (HPV) Penis cancer, adhesion 1989b; Scheffner Head and neck dysregulation et al., 1990) carcinoma Human T Retroviridae Adult T-cell 3300 Oncogenic Tax Molecular-signaling (Matsuoka and lymphotropic leukemia dysregulation Jeang, 2007; Poiesz virus type 1 NF-kB activation et al., 1980) (HTLV-1) immortalization Kaposi’s Herpesviridae Kaposi’s sarcoma, 102 300 Oncogenic vGPCR, vIL-6, Multiple-signaling (Arvanitakis et al., sarcoma- Pleural effusion vBcl2, vMIPs, events Cell cycle 1997; Bais et al., associated lymphoma, vFlip, vCyclin dysregulation 1998; Cesarman herpesvirus Multicentric LANA, Inhibition of et al., 1995; Chang (KSHV) Castleman’s Kaposin B apoptosis Immune et al., 1994; Ganem, (HHV-8) disease evasion Autocrine 2006; Montaner and paracrine et al., 2003; Yang functions et al., 2000) aSource: (Parkin, 2006) bHuman cancer-associated viruses often lead to malignancies with a prolonged latency and in conjunction with other environmental or host-related cooperating events, including nutritional status, environmental factors, parasitic or viral infection, chronic inflammation, and immunosuppression.

transmission of this disease. After an unfortunate yielding several overlapping and frame-shifted proteins accident, Blumberg identified in 1965 a blood antigen because of the alternative use of start codons. Its that reacted with antibodies from patients with hepatitis pathogenic mechanism is far from being completely (Blumberg, 1977). This antigen was a surface protein of understood (Guidotti and Chisari, 2006). None of the a hepatotropic DNA virus, which was called hepatitis B virally encoded proteins has been directly linked to an virus (HBV) (Hirschman et al., 1969). In most indivi- acutely oncogenic activity, as the interval between the duals, infection with this virus is asymptomatic or leads initial viral infection and the development of HCC to an acute hepatitis after which the infection is cleared normally spans several decades. Rather, persistent by the immune response (Guidotti et al., 1999). infection with HVB is associated with varying degrees However, HBV infection fails to clear, leading to of chronic liver disease, progressing to cirrhosis and chronic infection in a large fraction of individuals lately HCC (Guidotti and Chisari, 2006). During this infected as neonates or young children and in a small progression, sustained immune-mediated tissue injury, percentage of adults. We now know that HBV has a hepatocellular regeneration and continuous inflamma- causative role in the majority of the hepatocellular tion are thought to result in random genetic and carcinomas (HCC), leading to more than 300 000 deaths chromosomal damage leading to HCC (Nakamoto every year from a total 350 million people infected et al., 1998). Chronic liver cell injury could be worldwide (Kao and Chen, 2002) (Table 1). A vaccine considered as a premalignant state in which increased against HBV was developed in the late 1970s (Buynak DNA synthesis, proinflammatory byproducts such as et al., 1976). This provided an effective means to prevent reactive oxygen species, and decreased cellular detox- the spread of HVB infection and the subsequent ification and repair functions could result, over an development of HBV-associated HCC, which can there- extended period of time, in multiple genetic and fore be considered the first anticancer vaccine. chromosomal alterations and ultimately in cancer HBV possesses an unusual partially double-stranded (Ganem and Prince, 2004). In a few cases, HBV circular DNA encoding four open-reading frames, each DNA integration could also have a positional effect

Oncogene Human viral-associated malignancies D Martin and JS Gutkind S34 reminiscent of that of some retroviruses when integra- cancers are associated with the high-risk HPV types 16 tion occurs in the proximity of oncogenic genes (Wang and 18 (Beaudenon et al., 1986; zur Hausen, 2009). In et al., 1990). most people, infection with HPV is asymptomatic and is Interestingly, one of the HVB-encoded proteins, X cleared within 1 or 2 years in 90% of the cases. protein, has been proposed to activate several molecular However, genital HPV is considered the most common mechanisms involved in proliferation and inflammation, sexually transmitted disease that is responsible for more such as the AP-1 and NF-kB pathways, activation of than 500 000 new annual cases of human cancer the JAK1, protein kinase C, PI3K and MAPK (Tang worldwide, including those of the cervix, anus, vulva, et al., 2006). The X protein has also been described to penis and of the head and neck area and oropharynx directly interact and interfere with several transcription (Parkin, 2006; zur Hausen, 2009) (Table 1). factors including CREB, ATF, NF-IL6, EGR1, OCT1 The molecular pathogenesis of cervical cancer and RXR and the p53 tumor suppressor, all of which other HPV-associated cancers is much better under- may participate in HCC progression (Guidotti and stood than for other viral-induced malignancies. Pre- Chisari, 2006). ventive programs set in place worldwide after the establishment of HPV as a causative agent of cervical cancer through the routine Papanicolau test or PAP smear have provided extensive information about the Hepatitis C virus natural history of HPV infection and the development of cervical cancer. It is a lengthy process, spanning A particular subset of transfusion-associated serum normally about 12–15 years, involving chronic inflam- hepatitis patients was found to be not infected with mation and a multistage complex interaction of the host HBV (Dienstag et al., 1977). Interestingly, these patients immune system, in combination with the expression of developed a rather mild hepatitis that gave rise to two viral genes, E6 and E7, followed by integration persistent infection in an elevated percentage of the cases of the viral DNA into the host genome and a series of (Thomas et al., 2000). This disease was associated with a epigenetic changes occurring in dysplastic lesions small enveloped virus that Houghton and colleagues (Kadaja et al., 2009). The molecular function of E6 identified as an RNA virus of the flaviviridae family in and E7 provided one of the first evidences that human 1989, which was termed the Hepatitis C virus (HCV). A tumor viruses have evolved alternative strategies to year later, the same group established the link between induce cell proliferation and transformation. This HCV and HCC and chronic liver disease (Colombo mechanism was first observed during the analysis of et al., 1989). There are more than 270 million people another virus with oncogenic properties in human cells infected by HCV worldwide and 20% of them are in vitro, the SV40 polyomavirus. Unlike retroviruses, expected to eventually develop liver complications and instead of repurposing a hijacked human gene and cirrhosis, and those will ultimately have a 4–7% annual transforming it into an oncogene, DNA tumor viruses risk to progress to HCC (Thomas et al., 2000) (Table 1). sport genes of exclusive viral origin that are essential for The mechanism by which HCV causes liver damage and viral replication (Javier and Butel, 2008). Although no HCC is not much unlike HBV. It involves sustained liver human neoplasms have been associated with SV40 damage and inflammation by inducing a strong im- infection, its ability to induce tumors in animal models munological reaction in the liver that ultimately leads to enabled the identification of a virally encoded protein, cirrhosis and cancer after several decades of chronic large T antigen, harboring potent oncogenic activity. In infection. Unfortunately, the development of an 1979, molecular studies by Lane and Levine led to the effective HCV vaccine has been challenging because of identification of a cellular protein of 53 KDa that the weak immune response elicited by HCV. interacted strongly with large T and was thus named TP53 (Levine, 1997). It took several years before the impact of this discovery was fully realized. Vogelstein High-risk human papilloma viruses and Kinzler (2004) reported in 1989 that the TP53 gene is commonly deleted in colorectal cancers, thus raising In 1842, the Italian physician Rigoni-Stern noted that the possibility that TP53 represented a tumor suppres- cases of cervical cancer in women were associated with sor gene. This observation was rapidly extended to sexual contacts (Gasparini and Panatto, 2009). Genital numerous other cancer types. Indeed, p53 is the most warts had also been associated with sexual promiscuity, commonly mutated gene in human cancer, being altered and so a contagious etiology was suspected, but a in approximately 50% of all human malignancies demonstration of the contagious nature of genital warts (Vogelstein and Kinzler, 2004). did not arrive until 1907 when Ciuffo observed the It was a surprising finding that, similar to SV40, HPV transmission of warts using cell-free extracts (Ciuffo, viruses have evolved to deploy a similar strategy to 1907). To date, more than 130 HPV types have been induce transformation. The E6 oncogene product identified and classified into low- or high-risk groups associates with a cellular protein of 100 KDa called depending on their likelihood of inducing cervical E6-associated protein (E6-AP) with E3 ubiquitin ligase cancer on persistent infection (zur Hausen, 2009). activity (Scheffner et al., 1993). Although large T HPV is associated with nearly all the carcinomas of impairs the function of p53 based on physical interac- the uterine cervix, and in particular, 70% of the cervical tion and inhibition, E6 and E6-AP induce the rapid

Oncogene Human viral-associated malignancies D Martin and JS Gutkind S35 degradation of p53 through the proteasome pathway on The molecular mechanism by which HTLV-1 causes ubiquitination (Scheffner et al., 1993). Another onco- cancer is still under active investigation, but it is clear genic driver in HPV-infected cells is the viral oncogene that it involves the function of a virally encoded E7. In 1971, Knudson identified a gene that was mutated essential protein called Tax rather than of insertional in a rare form of a sporadic pediatric cancer, retino- mutagenesis that is typical for leukemia-causing animal blastoma, which could also be inherited (Knudson, retroviruses (Yoshida, 2001). The mechanism by which 1971). These studies led to the hypothesis that cancer Tax induces transformation seems to involve the development required two sporadic mutations in a single perturbation of several cell growth regulatory pathways, recessive trait, and that tumor progression could be epigenetic mechanisms and interference with the cellular greatly accelerated if one of the mutations was inherited. DNA repair apparatus (Yoshida, 2001), although only a In the mid 1980s the Rb tumor suppressor gene, small proportion of patients would develop ATL after responsible for this disease, was identified (Friend several decades from the initial infection. Similar to et al., 1986). Led by previous observations on p53,it EBV’s LMP-1, Tax strongly activates the NF-kB, AP-1 was rapidly found that its protein product, Rb, is the and CRE pathways that have a particularly potent target for some tumor virus proteins, including the SV40 pro-proliferative effect on lymphocytes, whereas Tax large T antigen and the E1A protein form of human also interacts with several factors regulating chromatin adenovirus (Dyson et al., 1989a). HPV-encoded E7 remodeling, such as HDAC-1 and SWI/SNF, and causes competes with the E2F transcription factor for binding the inhibition of DNA repair mechanisms, thus leading to Rb, releasing E2F and enabling DNA synthesis by to genomic instability and ultimate transformation entering into the S phase of the cell cycle (Dyson et al., (Matsuoka and Jeang, 2007). There is no vaccine 1989b). Unscheduled inactivation of the Rb protein available for this virus and treatment is restricted to triggers a robust antiproliferative response mediated by management of the opportunistic infections resulting the p53 protein as a mechanism to abort oncogenic from the immunosuppression caused by ATL transformation (Hickman et al., 2002). It is therefore (Tsukasaki et al., 2009). not surprising that viruses such as SV40, Adenovirus and HPV have co-evolved to inactivate both p53 and Rb to induce transformation. Recent studies suggest that E6 can also affect multiple cell–cell adhesion molecules and Kaposi’s sarcoma-associated herpesvirus (KSHV) signaling events (Mantovani and Banks, 2001). Ultimately, the recent development of HPV vaccines Kaposi’sarcoma (KS) is an angioproliferative tumor against some of the high-risk strains, including HPV16 that was first described by the Hungarian pathologist and HPV18, may help reduce the estimated more than Moritz Kaposi in 1872. KS lesions develop in connective 450 000 annual cases of cervical cancer worldwide tissues typically under the skin, giving rise to very (zur Hausen, 2009), as well as limiting the emerging characteristic purplish raised blotches because of the spread of HPV-associated oral cancers (Chaturvedi high vascularity of these lesions (Geilen et al., 2006). et al., 2008). Although most lesions remain asymptomatic, sometimes they can be disfiguring or painfully swollen, and KS lesions can be fatal if they develop in internal organs such as the lungs, liver or gastrointestinal tract (Levi, Human T-cell lymphotropic virus (HTLV-1) 2005). This disease was traditionally considered rare, affecting elderly men of Mediterranean or Jewish origin The stage for the discovery of the first human oncogenic and Northern African men (classic or sporadic KS) retrovirus was set in 1977 by Takatsuki and colleagues (Safai, 1987). The incidence of this disease changed who studied a particular leukemia termed adult T-cell radically during the last few decades because of the leukemia (ATL) in Japan (Uchiyama et al., 1977). widespread appearance of lymphoadenopathic KS cases Owing to its distinct geographic distribution, an in the sub-Saharan region in Africa (endemic KS), in infectious etiology was suspected. In 1980, after many transplanted patients undergoing immunosuppressive years of failed attempts to isolate human tumor retro- therapy (iatrogenic KS), and because of its tragic viruses by the scientific community, Robert Gallo association with AIDS (epidemic KS) (Ganem, 2006) identified a novel retrovirus in cultured human T-cell (Table 1). In fact, the description of a new, very lymphoma cells, which was named the human T-cell aggressive variant of KS heralded the beginning of the lymphotropic virus (HTLV-1) (Poiesz et al., 1980). worldwide AIDS epidemic (Wood and Harrington, Between 10 and 20 million people worldwide are 2005). AIDS–KS is a life-threatening condition char- infected with HTLV-1 and the prognosis remains poor acterized by extensive local edema and disfiguring for ATL patients (Table 1). In less than 2% of infected dermal lesions with involvement of the internal organs, individuals, HTLV-1 is also associated with a progres- oral cavity, bleeding and respiratory failure (Ganem, sive demyelinating upper motor neuron disease known 2006). Although the incidence of KS has been greatly as HAM/TSP for HTLV-1-associated myelopathy/ reduced since the introduction of the triple antiretroviral Tropical Spastic Paraparesis characterized by sensory therapy (HAART) for the treatment of HIV-infected and motor deficits, particularly of the lower extremities, individuals (Bower et al., 2006), it still remains the most incontinence and impotence (Araujo and Silva, 2006). prevalent cancer in children in Africa, and is still more

Oncogene Human viral-associated malignancies D Martin and JS Gutkind S36 prevalent among minority groups and in developing interplay with HIV and its encoded proteins, particu- countries. larly the HIV-Tat, combined with severe immunosu- KS lesions are histologically complex and are highly pression in the epidemic KS, can ultimately promote vascularized, containing proliferating spindle-shaped latently KSHV-infected cells to reactivate the viral cells that are believed to be of endothelial origin, replicative program of KSHV. In every case, KS together with infiltrating mononuclear cells and abun- progression requires the evasion of the immune surveil- dant slit-like neovascular spaces (Ganem, 2006). lance system, either by the deployment of a repertoire of Although it was initially thought that HIV could be KSHV viral strategies or by the inability of the host, to the causative agent of KS, intense research on AIDS and mount an appropriate immune response. This is HIV-related malignancies during the early 1990s, how- supported by the ability to cause KS regression once ever, led to the identification of a new human the patient’s immune system can fight back after the herpesvirus, HHV-8, as the etiological agent causing removal of immunosuppressive regimes, and in AIDS KS and two infrequent B-cell type lymphoproliferative patients on successful treatment with HAART (Ganem, diseases, primary effusion lymphoma (PEL) and multi- 2006). Although most spindle cells within the KS lesions centric Castleman’s disease (Cesarman et al., 1995). express exclusively latent viral genes, the presence of HHV-8, also know as KSHV, is a member of the g- 1–5% of cells undergoing the lytic cycle appears to be a herpesviridae family and as such, the shedding of KSHV prerequisite for KS development, suggesting that an in saliva appears to be the most frequent route of interesting interplay between latent infection gene transmission, as for other herpesviruses, including EBV programs and KSHV genes involved in lytic reactivation (Ganem, 2006; Pauk et al., 2000). The sequence of the may underlie KS development (Chiou et al., 2002). KSHV genome revealed the presence of at least 89 open- reading frames (ORFs), many of which are unique among the herpesviridae family (Russo et al., 1996). Surprisingly, many of these KSHV genes represent Latently expressed KSHV genes pirated versions of cellular genes, including those encoding molecules involved in cell cycle control, Owing to the destructive nature of the viral lytic cycle, in apoptosis prevention, immune system regulation and which infected cells are ultimately destined to die, genes inter- and intracellular communication, thus harboring expressed during latency were expected to hold the key proangiogenic and oncogenic potential (Russo et al., to understanding KSHV sarcomagenesis. During the 1996). As viral genes have historically represented latent cycle, only a small portion of the KSHV-encoded powerful tools for disclosing critical cellular processes, genes is expressed (Zhong et al., 1996). These include the the study of this angioproliferative tumor and its viral genes LANA, vCyclin, vFlip, the ORF K12/Kaposin enigmatic causative virus and numerous candidate and a miRNA cluster, which are expressed in nearly tumor-promoting genes may ultimately help to identify 70% of the spindle cells in advanced KS lesions (Dittmer novel mechanisms involved in cell growth control and et al., 1998). normal and tumor-induced angiogenesis. vCyclin is a virally encoded homolog of human cyclin D2 that forms functional complexes with cyclin-dependent kinase, which has broader substrate specificity than the endogenous complex and displays resistance to KSHV and novel viral oncogenes inhibition by the p16, p21 and p27 Cdk inhibitors (Sharp and Boshoff, 2000). vCyclin/cyclin-dependent KSHV could be arguably considered the virus harboring kinase complexes can even cause the degradation of p27 the highest number of potential oncogenes. As for other (Ellis et al., 1999). The oncogenic potential of vCyclin herpesviruses, on infection and after multiple rounds of can only be exposed, however, when p53 function is viral replication and spread, KSHV can enter into a defective (Verschuren et al., 2002) because of the strong latent viral program that in most cases lasts for the life engagement of p53-dependent cell cycle arrest and of the individual or undergoes further lytic replication apoptosis induced by vCyclin. In this regard, LANA- (Ganem, 2006). In latency, the most frequent viral mediated inhibition of p53 transcriptional activity (in choice, only a few of the viral genes are expressed; also, this section) could provide an appropriate environment the viral DNA is retained as a circular episome in the for vCyclin to exert its oncogenic function and nucleus, and no or few virions are produced, thus often contribute to KS progression in vivo. evading immune detection (Vider-Shalit et al., 2007). It vFlip is a viral homolog of the human FLICE is not known what causes KSHV reactivation and inhibitory protein. By analogy with other FLICE reentry into a lytic cycle. Although treatment with inhibitory protein proteins encoded by other herpes- phorbol esters can trigger the rapid replication of virus, it has been suggested that vFlip acts as an latently infected KSHV cells in vitro (Xie et al., 2008), inhibitor for procaspase 8 activation during FasL- accumulating evidence suggests that cytokines and triggered apoptosis (Irmler et al., 1997). This could chemokines produced during chronic inflammatory result in immune privilege by endowing the infected cells processes in the case of classical KS, additional with resistance to FasL and perhaps other death- environmental factors, parasitic infection and compro- receptor ligands, such as those used by cytotoxic T- mised immune system in the case of endemic KS, and an cells. vFlip can also stimulate the IKK complex leading

Oncogene Human viral-associated malignancies D Martin and JS Gutkind S37 to NF-kB activation (Field et al., 2003), a process that responses, and they could contribute to tumorigenesis has a key role in the survival of PEL cells (Keller et al., by altering the expression of a number of host 2000). vFlip is expressed in latently infected KS spindle genes involved in immunomodulation, apoptosis and cells and, as such, represents an ideal candidate for the angiogenesis. growth-promoting role of KSHV (Ganem, 2006). However, there is as yet no evidence that vFlip can cause endothelial cell transformation, except for the acquisition of a spindle phenotype in human umbilical KSHV lytic genes endothelial cells expressing this protein (Grossmann et al., 2006). Histological evidence supports the invariable existence LANA is a large nuclear protein with no known of a small portion (1–5%) of cells expressing KSHV lytic mammalian homolog (Rainbow et al., 1997). It is markers in KS lesions (Chiou et al., 2002). However, it is functionally similar to the EBV EBNA-1 protein, and unclear whether this cell population emerges from a is primarily implicated in the KSHV episomal persis- basal level of cells undergoing the lytic viral cycle or tence by tethering the KSHV episome to the host mitotic results from the dysregulated expression of lytic genes, chromosomes through interactions with cellular for example because of an aborted lytic program (Sodhi chromosome-associated proteins (Cotter and Robert- et al., 2004a). Owing to the nature and the potential son, 1999). Besides this crucial viral role, LANA oncogenic properties of several of these genes, intense dysregulates various cellular pathways involving the research has been conducted to address their specific tumor suppressors pRb and p53 (Hu et al., 2002). As roles in KS development. such, LANA can cooperate with H-Ras to transform vBcl-2 is distantly related to Bcl-2, but shares a high fibroblasts (Radkov et al., 2000). LANA interacts with degree of homology to the conserved Bcl-2 homology GSK3b, leading to the accumulation of cytoplasmatic domains BH1 and BH2 (Sarid et al., 1997). vBcl-2 and nuclear b-catenin (Fujimuro et al., 2003) and blocks apoptosis in diverse cellular systems as effectively aberrant c-Myc stabilization (Bubman et al., 2007). as do Bcl-2 and Bcl-xL (Sarid et al., 1997). Unlike some However, despite the oncogenic potential of these members of the Bcl-2 family of proteins, vBcl-2 is latently expressed genes, it is unclear whether the resistant to caspase-mediated conversion to a proapop- expression of these genes is sufficient to cause cell totic factor (Bellows et al., 2000). Although over- transformation. An interesting possibility is that the expression of endogenous Bcl-2 is transforming in B combined expression of these genes provides a signifi- lymphocytes (Chao and Korsmeyer, 1998), a role for cant growth advantage, causing the latently infected vBcl-2 in endothelial transformation has yet to be KSHV cells to hyperproliferate in response to paracrine established. signals produced by other lytically infected or inflam- vIRFs. Four IRFs have been identified in the KSHV matory cells. In this regard, it is possible to unmask their genome. These molecules share limited but significant intrinsic oncogenic potential when endothelial cells homology to the human IRF family of regulators of expressing vCyclin and vFlip (and LANA, DM and interferon signaling (Offermann, 2007). vIRFs inhibit JSGutkind, unpublished) are co-injected with a small interferon-b-induced signaling leading to activation of number of cells expressing the potent KSHV-encoded the p21 tumor suppressor and the ATM/p53 pathway, oncogene vGPCR (Montaner et al., 2003), thus likely and are able to induce fibroblast transformation recapitulating the interplay between latent and lytic (Offermann, 2007). These molecules are highly ex- KSHV genes often observed in KS lesions. pressed in B cells infected with KSHV, where they Kaposin B, one of the open-reading frames encoded may help circumvent the antiviral immune response by the poorly investigated KSHV latent gene K12, has mediated by interferons (Seo et al., 2002). vIRFs are not the potential to cooperate with the major latent usually detected in spindle cells and KS lesions, and gene transcripts to induce sarcomagenesis, as its gene their relevance in Kaposi’s sarcomagenesis is likely to be product binds and activates MK2 directly, and thus more limited (Gao et al., 1997). could stabilize mRNAs containing ARE sequences in Virally encoded cytokines and chemokines (vIL-6, their 30 UTRs, many of which are suspected to have a vCCL1, vCCL2, vCCL3). KSHV encodes a homolog role in KS, such as IL-8 or TNF-a (McCormick and of the human IL-6 cytokine, vIL-6. vIL-6 is able to Ganem, 2005). support the growth of a rat embryo fibroblast line that is KSHV miRNAs KSHV and other human herpes- dependent on IL-6 (Nicholas et al., 1997). Unlike its viruses also bear multiple small noncoding RNAs cellular counterpart, vIL-6 does not require binding to (miRNAs) that have recently been shown to modulate the IL-6 receptor subunit, but instead is able to signal gene expression by mediating posttranscriptional target- through the gp130-coreceptor subunit alone (Molden specific RNA interference (Cai et al., 2005). In the case et al., 1997). vIL-6 induces a potent transforming of KSHV, 17 distinct miRNAS encoded by 12 miRNA phenotype when expressed in murine fibroblast (Aoki genes have been identified within the latency-associated et al., 1999) and leads to the expression of vascular transcript, but are not required for viral replication as endothelial growth factor (VEGF) (Aoki et al., 1999). judged by experiments with genetically modified KSHV vIL-6 is expressed at low levels during latency, increas- viruses (Xu et al., 2006). It has been hypothesized that ing during lytic replication both in PEL and endothelial the role of these miRNAs is to regulate antiviral cellular cells (Staskus et al., 1999). Besides its transforming

Oncogene Human viral-associated malignancies D Martin and JS Gutkind S38 ability, vIL-6 is induced by IFN-a in PEL cells and hCMV-1, express one or multiple GPCRs from their effectively blocks cell cycle arrest and apoptosis viral genome, suggesting that these viruses might have provoked by IFN-a, thereby protecting latently infected evolved to harness the potent signaling capacity of cells from this antiviral cytokine (Chatterjee et al., 2002). GPCRs for their proliferative and replicative advantage The genes for vCCL1, vCCL2 and vCCL3 (vMIPs) (Dorsam and Gutkind, 2007; Rosenkilde et al., 2008). virally encoded chemokines are expressed during viral In particular, vGPCR initiates the activity of a complex replication and appear to be involved in Th2 lympho- signaling network whose contribution to KS is cyte chemoattraction (Choi and Nicholas, 2008). It has beginning to be unraveled. been postulated that vMIPs could produce a polarized vGPCR contributes to KS development through its immune response shielding from mounting effective potent transforming and proangiogenic functions. Th1-driven antiviral responses (Choi and Nicholas, Among the multiple intracellular pathways stimulated 2008). vCCL2 acts as a non-signaling ligand with broad by vGPCR, the Akt–mTOR signaling axis has recently affinity for multiple CC and CXC receptors expressed in taken center stage, as its activation contributes to KS Th1 cells. By disrupting effective communication among development in genetically defined animal models innate and adaptive immune cells, the release of vMIPs (Sodhi et al., 2004b). The pharmacological inhibition could provide an effective immune evasion strategy for of mTOR by the use of rapamycin prevents KS KSHV-infected cells. Interestingly, these virally encoded progression in experimental model systems (Sodhi chemokines display proangiogenic properties through et al., 2006) and represents a remarkably effective the induction of VEGF and possibly other factors treatment option for renal transplanted patients who (Dittmer and Kedes, 1998), whereas their activation of the develop KS on immunosuppression (Stallone et al., CCR8 receptor can be antiapoptotic and might therefore 2005). KSHV GPCR activates the Akt–mTOR by contribute to the counteraction of the tumor cell stimulating PI3 K through bg-subunits released from proapoptotic defenses of the host (Louahed et al., 2003). both pertussis-toxin-sensitive and -insensitive G proteins, as well as in an autocrine manner by upregulating the expression of the VEGF receptor VEGFR2/KDR and the concomitant release of VEGF (Bais et al., 2003). The KSHV vGPCR oncogene Although the stimulation of the Akt–mTOR bio- chemical route contributes to endothelial cell transfor- Although the KSHV genome encodes multiple candi- mation by vGPCR, perhaps the most striking biological date oncogenes, extensive experimental information activity of this viral oncogene is to promote paracrine revealed that its virally encoded G protein-coupled neoplasia. Indeed, only a very few of the spindle-like receptor (GPCR) (KSHV GPCR) is unique in that it is cells within the tumor mass express vGPCR even when both transforming and proangiogenic and can promote KS-like lesions are initiated by the expression of vGPCR the tumoral growth of cells expressing KSHV latent in endothelial cells on retroviral infection in vivo, genes in a paracrine manner (Arvanitakis et al., 1997; (Montaner et al., 2003). Although a hit-and-run process Bais et al., 1998; Montaner et al., 2003; Yang et al., could explain the tumor-initiating activity of vGPCR 2000). The KSHV GPCR is encoded by the ORF74, and and its subsequent loss as the tumor progresses, its its closest human homologs are CXCR1 and CXCR2, decreased expression in inducible animal model systems the receptors for IL-8/CXCL-8 and Gro-a/CXCL-1 (Grisotto et al., 2006) or the pharmacological ablation chemokines (Arvanitakis et al., 1997). KSHV GPCR, of vGPCR-expressing cells cause rapid tumor regres- also often referred to as vGPCR, is constitutively active sion, even if the majority of the cells within the tumor do because of the presence of several structural changes, not express vGPCR and continue expressing KSHV- including a mutation (Asp142Val) within its DRY motif latent genes (Montaner et al., 2006). At the molecular in the second intracytoplasmatic loop that is highly level, the angiogenic and paracrine transforming effect conserved among GPCRs (Rosenkilde et al., 2000). of vGPCR involves the activation of multiple mitogen- Moreover, this receptor shows a higher degree of activated protein kinases and small GTPases of the Rho promiscuity than does its cellular homologs and is able family whose activities converge in the nucleus to to bind both CXC and CC chemokines, further control transcription factors such as hypoxia-inducible enhancing its constitutive activity (Gershengorn et al., factor 1a, AP-1 and NF-kB, thereby promoting the 1998). Expression of this receptor is transforming in expression and secretion of proinflammatory cytokines several cell types including fibroblasts and endothelial such as VEGF, IL-6, IL-8/CXCL8 and MIP-1/CCL3 cells (Arvanitakis et al., 1997; Bais et al., 1998; (Martin et al., 2008; Montaner et al., 2001; Montaner Montaner et al., 2003). Expression of vGPCR as a et al., 2004; Sodhi et al., 2000; Sodhi et al., 2001). These, transgene or specifically in endothelial cells by a tissue- in turn, may recruit and promote the aberrant growth of specific retroviral transduction system leads to the KSHV-latent infected cells, as well as of multiple progressive development of KS-like lesions in multiple stromal and endothelial-derived cells and their progeni- animal model systems (Jensen et al., 2005; Montaner tors, which constitute the majority of the cell popula- et al., 2003; Yang et al., 2000). Although the presence of tions in KS tumors (Montaner et al., 2003) (Figure 1). mutant forms of GPCRs or heterotrimeric G proteins Strikingly, these tumoral cells remain nonetheless is not a frequent event in human cancer, several dependent on the persistent secretion of growth and herpesviruses, including EVB, HHV-6, HHV-7 and survival factors from vGPCR-expressing cells.

Oncogene Human viral-associated malignancies D Martin and JS Gutkind S39

Figure 1 Contribution of multiple oncogenic, proinflammatory and immune-evasive strategies for the development of KS in KSHV- infected cells. KS lesions are heterogeneous, composed of endothelial-derived spindle cells as well as multiple inflammatory and mesenchymal cells. Viral production occurs in a limited number (1–5%) of the KS cells that undergo lytic replication. Expression of the vGPCR oncogene in these cells and in cells aberrantly expressing lytic gene programs may contribute to sarcomagenesis because of the activation of a complex signaling network, including the NF-kB and Akt–mTOR pathways and multiple Rho GTPases and MAPKs, concomitant with the expression of potent proangiogenic, proinflammatory and chemoattractant factors such as the virally encoded vIL-6 and the vGPCR-induced secretion of IL-8, VEGF and Gro-a. These cytokines and chemokines promote the recruitment and proliferation of other latently infected endothelial-derived cells in a paracrine manner, in which the viral proteins vFlip, vCyclin and LANA provide a proliferative advantage. The host antiviral responses are inhibited by the antiapoptotic vIRFs and vBcl2 molecules, whereas the adaptive immune responses are evaded by the expression of multiple immunomodulatory molecules such as vMIPs.

Conclusions initiate pro-proliferative and immunomodulating signals through the activation of NF-kB and its transcriptional We now know that human tumor-associated viruses regulatory program. KSHV, the most recently identified contribute to 10–15% of the cancers worldwide, or human tumor-associated virus, harbors a surprising approximately 1.3 million new cancer cases each year. number of virally pirated genes, which contribute to Hence, the opportunity exists to make remarkable immune evasion and promote the growth and survival of progress in our fight against cancer at the global scale by latently infected cells. This enigmatic angioproliferative preventing the spread of these viruses, by the development disease also revealed the existence of a virally initiated and distribution of effective and safe vaccines against them process of paracrine transformation, which provides an and by identifying the oncogenic mechanisms deployed by excellent example of how cell populations expressing these viruses, thereby enabling the development of new distinct gene programs can cooperate in tumor initiation, treatment options for these virally associated human maintenance and dissemination. Indeed, intercellular and malignancies. Clearly, the study of how viruses hijack cell even interclonal cooperativity may not be limited to viral- cycle regulatory and intracellular and even intercellular associated tumors but are also likely to occur in some of signaling mechanisms to achieve their replicative success the most prevalent and aggressive human neoplasias and continues to provide novel insights into multiple human their tumor-initiating stem cells. Ultimately, this cell–cell diseases, including cancer. In this regard, although HBV interdependence and their communication through cyto- andHCVinfectioncontributetoHCCbypromoting kine and chemokine-initiated signaling networks can chronic inflammation, EBV and HTLV-1 appear to expose multiple molecular targets that can soon be

Oncogene Human viral-associated malignancies D Martin and JS Gutkind S40 exploited for the development of novel mechanism-based Acknowledgements chemopreventive strategies and anticancer treatments. We apologize to our colleagues whose work could not be cited because of space constraints. This research was supported by Conﬂict of interest the National Institutes of Health Intramural AIDS Targeted Antiviral Program and the National Institute of Dental and The authors declare no conﬂict of interest. Craniofacial Research.

References

Aoki Y, Jaffe ES, Chang Y, Jones K, Teruya-Feldstein J, Moore PS Chao DT, Korsmeyer SJ. (1998). BCL-2 family: regulators of cell et al. (1999). Angiogenesis and hematopoiesis induced by Kaposi’s death. Annu Rev Immunol 16: 395–419. sarcoma-associated herpesvirus-encoded interleukin-6. Blood 93: Chatterjee M, Osborne J, Bestetti G, Chang Y, Moore PS. (2002). 4034–4043. Viral IL-6-induced cell proliferation and immune evasion of Araujo AQ, Silva MT. (2006). The HTLV-1 neurological complex. interferon activity. Science 298: 1432–1435. Lancet Neurol 5: 1068–1076. Chaturvedi AK, Engels EA, Anderson WF, Gillison ML. (2008). Arvanitakis L, Yaseen N, Sharma S. (1995). Latent membrane protein- Incidence trends for human papillomavirus-related and -unrelated 1 induces cyclin D2 expression, pRb hyperphosphorylation, and loss oral squamous cell carcinomas in the United States. J Clin Oncol 26: of TGF-beta 1-mediated growth inhibition in EBV-positive B cells. 612–619. J Immunol 155: 1047–1056. Chiou CJ, Poole LJ, Kim PS, Ciufo DM, Cannon JS, ap Rhys CM Arvanitakis L, Geras-Raaka E, Varma A, Gershengorn MC, Cesar- et al. (2002). Patterns of gene expression and a transactivation man E. (1997). Human herpesvirus KSHV encodes a constitutively function exhibited by the vGCR (ORF74) chemokine receptor active G-protein-coupled receptor linked to cell proliferation. protein of Kaposi’s sarcoma-associated herpesvirus. J Virol 76: Nature 385: 347–350. 3421–3439. Bais C, Santomasso B, Coso O, Arvanitakis L, Raaka EG, Gutkind JS Choi YB, Nicholas J. (2008). Autocrine and paracrine promotion of et al. (1998). G-protein-coupled receptor of Kaposi’s sarcoma- cell survival and virus replication by human herpesvirus 8 chemo- associated herpesvirus is a viral oncogene and angiogenesis kines. J Virol 82: 6501–6513. activator. Nature 391: 86–89. Ciuffo G. (1907). Innesto positivo con filtrato di verruca volgare. Bais C, Van Geelen A, Eroles P, Mutlu A, Chiozzini C, Dias S et al. G Ital d mal ven ed pelle 48: 12. (2003). Kaposi’s sarcoma associated herpesvirus G protein-coupled Coffin JM, Hughes SH, Varmus H. (1997). Retroviruses. Cold Spring receptor immortalizes human endothelial cells by activation of the Harbor Laboratory Press: Plainview, NY. VEGF receptor-2/KDR. Cancer Cell 3: 131–143. Colombo M, Kuo G, Choo QL, Donato MF, Del Ninno E, Beaudenon S, Kremsdorf D, Croissant O, Jablonska S, Wain-Hobson Tommasini MA et al. (1989). Prevalence of antibodies to hepatitis S, Orth G. (1986). A novel type of human papillomavirus associated C virus in Italian patients with hepatocellular carcinoma. Lancet 2: with genital neoplasias. Nature 321: 246–249. 1006–1008. Bellows DS, Chau BN, Lee P, Lazebnik Y, Burns WH, Hardwick JM. Cotter II MA, Robertson ES. (1999). The latency-associated nuclear (2000). Antiapoptotic herpesvirus Bcl-2 homologs escape caspase- antigen tethers the Kaposi’s sarcoma-associated herpesvirus genome mediated conversion to proapoptotic proteins. JVirol74: 5024–5031. to host chromosomes in body cavity-based lymphoma cells. Blumberg BS. (1977). Australia antigen and the biology of hepatitis B. Virology 264: 254–264. Science 197: 17–25. Dienstag JL, Purcell HR, Alter HJ, Feinstone SM, Wong DC, Bower M, Palmieri C, Dhillon T. (2006). AIDS-related malignancies: Holland PV. (1977). Non-A, non-B post-transfusion hepatitis. changing epidemiology and the impact of highly active antiretrovir- Lancet 1: 560–562. al therapy. Curr Opin Infect Dis 19: 14–19. Dittmer D, Kedes DH. (1998). Do viral chemokines modulate Brown KD, Hostager BS, Bishop GA. (2001). Differential signaling Kaposi’s sarcoma? Bioessays 20: 367–370. and tumor necrosis factor receptor-associated factor (TRAF) Dittmer D, Lagunoff M, Renne R, Staskus K, Haase A, Ganem D. degradation mediated by CD40 and the Epstein-Barr virus (1998). A cluster of latently expressed genes in Kaposi’s sarcoma- oncoprotein latent membrane protein 1 (LMP1). J Exp Med 193: associated herpesvirus. J Virol 72: 8309–8315. 943–954. Dorsam RT, Gutkind JS. (2007). G-protein-coupled receptors and Bubman D, Guasparri I, Cesarman E. (2007). Deregulation of c-Myc cancer. Nat Rev Cancer 7: 79–94. in primary effusion lymphoma by Kaposi’s sarcoma herpesvirus Duesberg PH, Vogt PK. (1970). Differences between the ribonucleic latency-associated nuclear antigen. Oncogene 26: 4979–4986. acids of transforming and non-transforming avian tumor viruses. Burkitt D. (1962). A children’s cancer dependent on climatic factors. Proc Natl Acad Sci USA 67: 1673–1680. Nature 194: 232–234. Dyson N, Buchkovich K, Whyte P, Harlow E. (1989a). The Buynak EB, Roehm RR, Tytell AA, Bertland II AU, Lampson GP, cellular 107 K protein that binds to adenovirus E1A also Hilleman MR. (1976). Vaccine against human hepatitis B. JAMA associates with the large T antigens of SV40 and JC virus. Cell 58: 235: 2832–2834. 249–255. Cai X, Lu S, Zhang Z, Gonzalez CM, Damania B, Cullen BR. (2005). Dyson N, Howley PM, Munger K, Harlow E. (1989b). The human Kaposi’s sarcoma-associated herpesvirus expresses an array of viral papilloma virus-16 E7 oncoprotein is able to bind to the microRNAs in latently infected cells. Proc Natl Acad Sci USA 102: retinoblastoma gene product. Science 243: 934–937. 5570–5575. Ellermann V, Bang O. (1908). Experimentelle leukamie bei huhnern. Cesarman E, Chang Y, Moore PS, Said JW, Knowles DM. (1995). Zentralbl Bakteriol Parasitenkd Infektionskr Hyg 46: 595–597. Kaposi’s sarcoma-associated herpesvirus-like DNA sequences in Ellis M, Chew YP, Fallis L, Freddersdorf S, Boshoff C, Weiss RA AIDS-related body-cavity-based lymphomas. N Engl J Med 332: et al. (1999). Degradation of p27(Kip) cdk inhibitor triggered by 1186–1191. Kaposi’s sarcoma virus cyclin-cdk6 complex. EMBO J 18: 644–653. Chang Y, Cesarman E, Pessin MS, Lee F, Culpepper J, Knowles DM Epstein MA, Henle G, Achong BG, Barr YM. (1965). Morphological et al. (1994). Identification of herpesvirus-like DNA sequences in and biological studies on a virus in cultured lymphoblasts from AIDS-associated Kaposi’s sarcoma. Science 266: 1865–1869. Burkitt’s lymphoma. J Exp Med 121: 761–770.

Oncogene Human viral-associated malignancies D Martin and JS Gutkind S41 Epstein MA. (1971). The possible role of viruses in human cancer. Keller SA, Schattner EJ, Cesarman E. (2000). Inhibition of NF- Lancet 1: 1344–1347. kappaB induces apoptosis of KSHV-infected primary effusion Epstein MA. (2001). Historical background. Philos Trans R Soc Lond lymphoma cells. Blood 96: 2537–2542. B Biol Sci 356: 413–420. Kilger E, Kieser A, Baumann M, Hammerschmidt W. (1998). Epstein- Field N, Low W, Daniels M, Howell S, Daviet L, Boshoff C et al. Barr virus-mediated B-cell proliferation is dependent upon latent (2003). KSHV vFLIP binds to IKK-gamma to activate IKK. J Cell membrane protein 1, which simulates an activated CD40 receptor. Sci 116: 3721–3728. EMBO J 17: 1700–1709. Friend SH, Bernards R, Rogelj S, Weinberg RA, Rapaport JM, Albert Knudson Jr AG. (1971). Mutation and cancer: statistical study of DM et al. (1986). A human DNA segment with properties of the retinoblastoma. Proc Natl Acad Sci USA 68: 820–823. gene that predisposes to retinoblastoma and osteosarcoma. Nature Lecoq H. (2001). Discovery of the first virus, the tobacco mosaic virus: 323: 643–646. 1892 or 1898? C R Acad Sci III 324: 929–933. Fujimuro M, Wu FY, ApRhys C, Kajumbula H, Young DB, Hayward Levi MJ. (2005). Classic Kaposi’s sarcoma. J Am Podiatr Med Assoc GS et al. (2003). A novel viral mechanism for dysregulation of beta- 95: 586–588. catenin in Kaposi’s sarcoma-associated herpesvirus latency. Nat Levine AJ. (1997). p53, the cellular gatekeeper for growth and division. Med 9: 300–306. Cell 88: 323–331. Ganem D, Prince AM. (2004). Hepatitis B virus infection–natural Lombardi L, Newcomb EW, Dalla-Favera R. (1987). Pathogenesis of history and clinical consequences. N Engl J Med 350: 1118–1129. Burkitt lymphoma: expression of an activated c-myc oncogene Ganem D. (2006). KSHV infection and the pathogenesis of Kaposi’s causes the tumorigenic conversion of EBV-infected human B sarcoma. Annu Rev Pathol 1: 273–296. lymphoblasts. Cell 49: 161–170. Gao SJ, Boshoff C, Jayachandra S, Weiss RA, Chang Y, Moore PS. Louahed J, Struyf S, Demoulin JB, Parmentier M, Van Snick J, Van (1997). KSHV ORF K9 (vIRF) is an oncogene, which inhibits the Damme J et al. (2003). CCR8-dependent activation of the RAS/ interferon signaling pathway. Oncogene 15: 1979–1985. MAPK pathway mediates anti-apoptotic activity of I-309/CCL1 Gasparini R, Panatto D. (2009). Cervical cancer: From Hippocrates and vMIP-I. Eur J Immunol 33: 494–501. through Rigoni-Stern to zur Hausen. Vaccine 27: A4–A5. Mantovani F, Banks L. (2001). The human papillomavirus E6 Geilen CC, Husak R, Steinhoff M. (2006). Pathogenesis and clinical protein and its contribution to malignant progression. Oncogene manifestation of Kaposi’s sarcoma. Front Radiat Ther Oncol 39: 20: 7874–7887. 59–67. Martin D, Galisteo R, Ji Y, Montaner S, Gutkind JS. (2008). An Gershengorn MC, Geras-Raaka E, Varma A, Clark-Lewis I. (1998). NF-kappaB gene expression signature contributes to Kaposi’s Chemokines activate Kaposi’s sarcoma-associated herpesvirus G sarcoma virus vGPCR-induced direct and paracrine neoplasia. protein-coupled receptor in mammalian cells in culture. J Clin Invest Oncogene 27: 1844–1852. 102: 1469–1472. Matsuoka M, Jeang KT. (2007). Human T-cell leukaemia virus type 1 Grisotto MG, Garin A, Martin AP, Jensen KK, Chan P, Sealfon SC (HTLV-1) infectivity and cellular transformation. Nat Rev Cancer 7: et al. (2006). The human herpesvirus 8 chemokine receptor vGPCR 270–280. triggers autonomous proliferation of endothelial cells. J Clin Invest McCormick C, Ganem D. (2005). The kaposin B protein of KSHV 116: 1264–1273. activates the p38/MK2 pathway and stabilizes cytokine mRNAs. Grossmann C, Podgrabinska S, Skobe M, Ganem D. (2006). Science 307: 739–741. Activation of NF-kappaB by the latent vFLIP gene of Kaposi’s Molden J, Chang Y, You Y, Moore PS, Goldsmith MA. (1997). A sarcoma-associated herpesvirus is required for the spindle shape of Kaposi’s sarcoma-associated herpesvirus-encoded cytokine homo- virus-infected endothelial cells and contributes to their proinflam- log (vIL-6) activates signaling through the shared gp130 receptor matory phenotype. J Virol 80: 7179–7185. subunit. J Biol Chem 272: 19625–19631. Guidotti LG, Rochford R, Chung J, Shapiro M, Purcell R, Chisari Montaner S, Sodhi A, Pece S, Mesri EA, Gutkind JS. (2001). The FV. (1999). Viral clearance without destruction of infected cells Kaposi’s sarcoma-associated herpesvirus G protein-coupled recep- during acute HBV infection. Science 284: 825–829. tor promotes endothelial cell survival through the activation of Akt/ Guidotti LG, Chisari FV. (2006). Immunobiology and pathogenesis of protein kinase B. Cancer Res 61: 2641–2648. viral hepatitis. Annu Rev Pathol 1: 23–61. Montaner S, Sodhi A, Molinolo A, Bugge TH, Sawai ET, He Y et al. Hickman ES, Moroni MC, Helin K. (2002). The role of p53 and pRB (2003). Endothelial infection with KSHV genes in vivo reveals that in apoptosis and cancer. Curr Opin Genet Dev 12: 60–66. vGPCR initiates Kaposi’s sarcomagenesis and can promote the Hirschman RJ, Shulman NR, Barker LF, Smith KO. (1969). Virus-like tumorigenic potential of viral latent genes. Cancer Cell 3: 23–36. particles in sera of patients with infectious and serum hepatitis. Montaner S, Sodhi A, Servitja JM, Ramsdell AK, Barac A, Sawai ET JAMA 208: 1667–1670. et al. (2004). The small GTPase Rac1 links the Kaposi sarcoma- Howley PM, Livingston DM. (2009). Small DNA tumor viruses: associated herpesvirus vGPCR to cytokine secretion and paracrine large contributors to biomedical sciences. Virology 384: neoplasia. Blood 104: 2903–2911. 256–259. Montaner S, Sodhi A, Ramsdell AK, Martin D, Hu J, Sawai ET et al. Hu J, Garber AC, Renne R. (2002). The latency-associated nuclear (2006). The Kaposi’s sarcoma-associated herpesvirus G protein- antigen of Kaposi’s sarcoma-associated herpesvirus supports latent coupled receptor as a therapeutic target for the treatment of DNA replication in dividing cells. J Virol 76: 11677–11687. Kaposi’s sarcoma. Cancer Res 66: 168–174. Irmler M, Thome M, Hahne M, Schneider P, Hofmann K, Steiner V Mosialos G, Birkenbach M, Yalamanchili R, VanArsdale T, Ware C, et al. (1997). Inhibition of death receptor signals by cellular FLIP. Kieff E. (1995). The Epstein-Barr virus transforming protein LMP1 Nature 388: 190–195. engages signaling proteins for the tumor necrosis factor receptor Javier RT, Butel JS. (2008). The history of tumor virology. Cancer Res family. Cell 80: 389–399. 68: 7693–7706. Nakamoto Y, Guidotti LG, Kuhlen CV, Fowler P, Chisari FV. (1998). Jensen KK, Manfra DJ, Grisotto MG, Martin AP, Vassileva G, Kelley Immune pathogenesis of hepatocellular carcinoma. J Exp Med 188: K et al. (2005). The human herpes virus 8-encoded chemokine 341–350. receptor is required for angioproliferation in a murine model of Nicholas J, Ruvolo VR, Burns WH, Sandford G, Wan X, Ciufo D Kaposi’s sarcoma. J Immunol 174: 3686–3694. et al. (1997). Kaposi’s sarcoma-associated human herpesvirus-8 Kadaja M, Isok-Paas H, Laos T, Ustav E, Ustav M. (2009). encodes homologues of macrophage inflammatory protein-1 and Mechanism of genomic instability in cells infected with the high- interleukin-6. Nat Med 3: 287–292. risk human papillomaviruses. PLoS Pathog 5: e1000397. Offermann MK. (2007). Kaposi sarcoma herpesvirus-encoded Kao JH, Chen DS. (2002). Global control of hepatitis B virus interferon regulator factors. Curr Top Microbiol Immunol 312: infection. Lancet Infect Dis 2: 395–403. 185–209.

Oncogene Human viral-associated malignancies D Martin and JS Gutkind S42 Parkin DM. (2006). The global health burden of infection-associated transformation induced by the Kaposi’s sarcoma-associated her- cancers in the year 2002. Int J Cancer 118: 3030–3044. pesvirus G protein-coupled receptor. Cancer Cell 10: 133–143. Pauk J, Huang ML, Brodie SJ, Wald A, Koelle DM, Schacker T et al. Stallone G, Schena A, Infante B, Di Paolo S, Loverre A, Maggio G (2000). Mucosal shedding of human herpesvirus 8 in men. N Engl J et al. (2005). Sirolimus for Kaposi’s sarcoma in renal-transplant Med 343: 1369–1377. recipients. N Engl J Med 352: 1317–1323. Poiesz BJ, Ruscetti FW, Gazdar AF, Bunn PA, Minna JD, Gallo RC. Staskus KA, Sun R, Miller G, Racz P, Jaslowski A, Metroka C et al. (1980). Detection and isolation of type C retrovirus particles from (1999). Cellular tropism and viral interleukin-6 expression distin- fresh and cultured lymphocytes of a patient with cutaneous T-cell guish human herpesvirus 8 involvement in Kaposi’s sarcoma, lymphoma. Proc Natl Acad Sci USA 77: 7415–7419. primary effusion lymphoma, and multicentric Castleman’s disease. Radkov SA, Kellam P, Boshoff C. (2000). The latent nuclear antigen J Virol 73: 4181–4187. of Kaposi sarcoma-associated herpesvirus targets the retinoblasto- Stehelin D, Varmus HE, Bishop JM, Vogt PK. (1976). DNA related to ma-E2F pathway and with the oncogene Hras transforms primary the transforming gene(s) of avian sarcoma viruses is present in rat cells. Nat Med 6: 1121–1127. normal avian DNA. Nature 260: 170–173. Rainbow L, Platt GM, Simpson GR, Sarid R, Gao SJ, Stoiber H et al. Tang H, Oishi N, Kaneko S, Murakami S. (2006). Molecular functions (1997). The 222- to 234-kilodalton latent nuclear protein (LNA) of and biological roles of hepatitis B virus x protein. Cancer Sci 97: Kaposi’s sarcoma-associated herpesvirus (human herpesvirus 8) is 977–983. encoded by orf73 and is a component of the latency-associated Thomas DL, Astemborski J, Rai RM, Anania FA, Schaeffer M, nuclear antigen. J Virol 71: 5915–5921. Galai N et al. (2000). The natural history of hepatitis C virus Rosenkilde MM, Kledal TN, Holst PJ, Schwartz TW. (2000). Selective infection: host, viral, and environmental factors. JAMA 284: elimination of high constitutive activity or chemokine binding in the 450–456. human herpesvirus 8 encoded seven transmembrane oncogene Tsukasaki K, Hermine O, Bazarbachi A, Ratner L, Ramos JC, ORF74. J Biol Chem 275: 26309–26315. Harrington Jr W et al. (2009). Deﬁnition, prognostic factors, Rosenkilde MM, Smit MJ, Waldhoer M. (2008). Structure, function treatment, and response criteria of adult T-cell leukemia-lymphoma: and physiological consequences of virally encoded chemokine seven a proposal from an international consensus meeting. J Clin Oncol transmembrane receptors. Br J Pharmacol 153Suppl 1: S154–S166. 27: 453–459. Rous P. (1910). A transmissible avian neoplasm (sarcoma of the Uchiyama T, Yodoi J, Sagawa K, Takatsuki K, Uchino H. (1977). common fowl). pp 696–705. Adult T-cell leukemia: clinical and hematologic features of 16 cases. Rous P. (1911). A sarcoma of the fowl transmissible by an agent Blood 50: 481–492. separable from the tumor cells. pp 397–411. Verschuren EW, Klefstrom J, Evan GI, Jones N. (2002). The Russo JJ, Bohenzky RA, Chien MC, Chen J, Yan M, Maddalena D oncogenic potential of Kaposi’s sarcoma-associated herpesvirus et al. (1996). Nucleotide sequence of the Kaposi sarcoma-associated cyclin is exposed by p53 loss in vitro and in vivo. Cancer Cell 2: herpesvirus (HHV8). Proc Natl Acad Sci USA 93: 14862–14867. 229–241. Safai B. (1987). Pathophysiology and epidemiology of epidemic Vider-Shalit T, Fishbain V, Raffaeli S, Louzoun Y. (2007). Phase- Kaposi’s sarcoma. Semin Oncol 14: 7–12. dependent immune evasion of herpesviruses. J Virol 81: 9536–9545. Sarid R, Sato T, Bohenzky RA, Russo JJ, Chang Y. (1997). Kaposi’s Vogelstein B, Kinzler KW. (2004). Cancer genes and the pathways they sarcoma-associated herpesvirus encodes a functional bcl-2 homo- control. Nat Med 10: 789–799. logue. Nat Med 3: 293–298. Vogt PK. (2009). A humble chicken virus that changed biology and Scheffner M, Werness BA, Huibregtse JM, Levine AJ, Howley PM. medicine. Lancet Oncol 10: 96. (1990). The E6 oncoprotein encoded by human papillomavirus types Wang J, Chenivesse X, Henglein B, Brechot C. (1990). Hepatitis B 16 and 18 promotes the degradation of p53. Cell 63: 1129–1136. virus integration in a cyclin A gene in a hepatocellular carcinoma. Scheffner M, Huibregtse JM, Vierstra RD, Howley PM. (1993). The Nature 343: 555–557. HPV-16 E6 and E6-AP complex functions as a ubiquitin-protein Weinberg RA. (1997). The cat and mouse games that genes, viruses, ligase in the ubiquitination of p53. Cell 75: 495–505. and cells play. Cell 88: 573–575. Seo T, Lee D, Shim YS, Angell JE, Chidambaram NV, Kalvakolanu Wood C, Harrington Jr W. (2005). AIDS and associated malignancies. DV et al. (2002). Viral interferon regulatory factor 1 of Kaposi’s Cell Res 15: 947–952. sarcoma-associated herpesvirus interacts with a cell death regulator, Xie J, Ajibade AO, Ye F, Kuhne K, Gao SJ. (2008). Reactivation of GRIM19, and inhibits interferon/retinoic acid-induced cell death. Kaposi’s sarcoma-associated herpesvirus from latency requires J Virol 76: 8797–8807. MEK/ERK, JNK and p38 multiple mitogen-activated protein Sharp TV, Boshoff C. (2000). Kaposi’s sarcoma-associated herpes- kinase pathways. Virology 371: 139–154. virus: from cell biology to pathogenesis. IUBMB Life 49: 97–104. Xu Y, Rodriguez-Huete A, Pari GS. (2006). Evaluation of the lytic Sodhi A, Montaner S, Patel V, Zohar M, Bais C, Mesri EA et al. origins of replication of Kaposi’s sarcoma-associated virus/human (2000). The Kaposi’s sarcoma-associated herpes virus G protein- herpesvirus 8 in the context of the viral genome. J Virol 80: coupled receptor up-regulates vascular endothelial growth factor 9905–9909. expression and secretion through mitogen-activated protein kinase Yang TY, Chen SC, Leach MW, Manfra D, Homey B, Wiekowski M and p38 pathways acting on hypoxia-inducible factor 1alpha. et al. (2000). Transgenic expression of the chemokine Cancer Res 60: 4873–4880. receptor encoded by human herpesvirus 8 induces an angioproli- Sodhi A, Montaner S, Miyazaki H, Gutkind JS. (2001). MAPK and ferative disease resembling Kaposi’s sarcoma. J Exp Med 191: Akt act cooperatively but independently on hypoxia inducible 445–454. factor-1alpha in rasV12 upregulation of VEGF. Biochem Biophys Yoshida M. (2001). Multiple viral strategies of HTLV-1 for Res Commun 287: 292–300. dysregulation of cell growth control. Annu Rev Immunol 19: Sodhi A, Montaner S, Gutkind JS. (2004a). Does dysregulated 475–496. expression of a deregulated viral GPCR trigger Kaposi’s sarcoma- Young LS, Rickinson AB. (2004). Epstein-Barr virus: 40 years on. Nat genesis? FASEB J 18: 422–427. Rev Cancer 4: 757–768. Sodhi A, Montaner S, Patel V, Gomez-Roman JJ, Li Y, Sausville EA Zhong W, Wang H, Herndier B, Ganem D. (1996). Restricted et al. (2004b). Akt plays a central role in sarcomagenesis induced by expression of Kaposi sarcoma-associated herpesvirus (human Kaposi’s sarcoma herpesvirus-encoded G protein-coupled receptor. herpesvirus 8) genes in Kaposi sarcoma. Proc Natl Acad Sci USA Proc Natl Acad Sci USA 101: 4821–4826. 93: 6641–6646. Sodhi A, Chaisuparat R, Hu J, Ramsdell AK, Manning BD, Sausville zur Hausen H. (2009). Papillomaviruses in the causation of human EA et al. (2006). The TSC2/mTOR pathway drives endothelial cell cancers — a brief historical account. Virology 384: 260–265.

Oncogene