Abstract
Kaposi’s sarcoma (KS) and Primary effusion lymphoma (PEL) are two Kaposi’s sarcoma-associated herpesvirus (KSHV/HHV-8) –induced cancers that are clinically challenging to treat and often have poor prognoses. While these diseases have declined dramatically in the developed world, they are responsible for significant morbidity and mortality in the developing world. Further, current treatments are often associated with considerable toxicity. While KS and PEL are composed of cells harboring KSHV in the latent state, the paracrine hypothesis of tumor development proposes that the tumor is actually being driven by the minority of cells undergoing lytic replication. Therefore, it is believed that potential targets for therapy are the KSHV-infected cells undergoing lytic replication. Lytic replication ensures the production and spread of virions, and allows for the expression of potentially pathogenic genes which have proposed roles in the paracrine neoplasia thought to drive tumorigenesis. The lytic-inductive paradigm for therapy proposes to induce massive lytic reactivation, while concurrently administering a potent antiviral, thereby specifically causing the death of virally infected cells while abrogating the potential increase in viremia, which may lead to unwanted negative clinical sequelae. However, the studying induction of and inhibition of viral lytic replication are generally reliant upon in vitro chemical induction which may not recapitulate the temporally ordered cascade of events that is thought to occur in vivo. Facile animal models are urgently needed in which to test antineoplastic and antiviral strategies for the development of effective targeted therapies. Herein, we test the lytic-inductive paradigm in a direct xenotransplant murine model of PEL. We use two clinically approved drugs, one, Vorinostat (suberoylanilide hydroxamic acid, SAHA) is a histone deacetylase inhibitor and a known potent inducer of viral replication. The other, Velcade (Bortezomib, Btz) is an inhibitor of the 26S proteasome and has been shown in other viral models to inhibit viral replication. We found that in PEL inoculated mice, the combination of SAHA and Btz leads to significantly increased survival over mice treated with the single agents. Further, we show that there is massive apoptosis with the combination that correlates with strong lytic induction of KSHV. Importantly, we also show that a potential unwanted side effect of lytic replication, namely increased viremia in vivo, is abrogated by Btz. Indeed, Btz appears to have very potent antiviral effects in the setting of KSHV while the combination of the SAHA and Btz have very potent antitumoral effects. Our results suggest that the lytic-inductive paradigm for treating PEL, and potentially other herpesvirus-induced cancers, may be a viable option in the setting of an immunocompromised host. To test the lytic-inductive paradigm in the context of KS, a solid tumor, we needed to develop a suitable model. By employing murine bone marrow-derived endothelial-lineage cells (mEC) we have now developed two novel productively infected murine models of KS. We show that the mEC are non-tumorigenic when uninfected, but, upon infection with rKSHV.219, a recombinant replication competent virus, the KSHV-infected cells efficiently form tumors that pathologically, phenotypically and molecularly resemble Kaposi’s sarcoma. The tumors are composed of LANA positive spindle-cells and that express antigens associated with human KS spindle-cells such as podoplanin and CD31. Further, the virus is able to transcribe a wide array of viral genes representing all stages of the viral replicative cycle culminating with the in vivo production of a herpesvirus-like particle of the correct size and morphology to be rKSHV.219. Given the productive nature of the tumors, these models should provide excellent substrates in which to test the lytic-inductive paradigm for the treatment of KS along with other antineoplastic and antiviral strategies.