Inhibiting HDACs in a preclinical model of HTLV-1-induced adult T-cell lymphoma

The human T-lymphotropic retrovirus type 1 (HTLV-1) infects and immortalizes CD4⁺ T-cells and causes adult T-cell leukemia/lymphoma (ATL), an aggressive hematological malignancy that is resistant to most anticancer treatment modalities. With few exceptions, the clinical prognoses associated with advanced acute- and lymphoma-stage ATL are dismal – with typical patient life expectancies of 6 months to 2 years. There is an urgent need to develop improved chemotherapeutic strategies for the clinical management of HTLV-1-induced cancers. Herein, Zimmerman et al. report that the histone deacetylase inhibitors (HDAC-Is), valproic acid (VPA) and AR-42, induce histone hyperacetylation and apoptosis in HTLV-1-transformed T-cell-lines [1]. As the in vivo efficacy of AR-42 has been previously demonstrated in a xenograft model of prostate cancer [2,3], the authors further show that AR-42, administered orally, decreases the serum level of the human interleukin-2-receptor alpha chain (hIL-2Rα, an indicator of tumor burden) and prolonged the survival of treated animals in a NOD/SCID model of ATL [1]. These findings suggest that the global activation of cellular and/or proviral genes, as a result of epigenetic histone modifications and chromatin remodeling, has pleiotropic consequences in rapidly dividing HTLV-1-infected tumor cells which lead to their elimination through apoptosis. Moreover, these results allude to an intriguing alternative approach to reduce the leukemic tumor burden in ATL patients using orally bioavailable HDAC-Is.

The HTLV-1 encodes seven non-structural/regulatory proteins (Tax, Rex, Hbz, p8^I, p12^I, p13^II, and p30^II) within a highly conserved 3′ nucleotide sequence, known as pX, which modulate retroviral gene expression and replication, maintain proviral silencing, and deregulate cellular signaling pathways associated with aberrant lymphoproliferation [4,5]. These factors exert antagonistic control over HTLV-1 gene expression through a complex cross-regulatory network, dependent upon an intricate balance of retroviral proteins. This is of primary importance, since HTLV-1-infected T-lymphocytes must evade immunological detection in order to establish long-term persistent infections as a prelude to neoplastic disease. The retroviral transactivator, Tax, interacts with cellular transcription factors and transcriptional coactivators (p300/CREB-binding protein, PCAF, TORC-1/2, CARM-1, P-TEFb) on three Tax-responsive elements (TREs) within the HTLV-1 promoter to drive proviral transcription and replication [5]. Tax-dependent transactivation is negatively influenced by other pX proteins: p30^II, p13^II, and Hbz [6–10]. p30^II interacts with p300 and competes against Tax for coactivator-binding, and thereby inhibits the formation of Tax-CREB-p300-TRE quaternary complexes on the HTLV-1 promoter [6]. Ectopic overexpression of p30^II suppresses nuclear export of the doubly spliced HTLV-1 tax/rex mRNA and negatively regulates proviral gene expression at the post-transcriptional level [7]. Interestingly, Younis et al. have shown that p30^II is recruited to the HTLV-1 promoter in a cotranscriptional manner, through interactions with Tax, and travels with the RNA Pol II elongation complex to reach its nascent mRNA target [8]. The antisense-encoded, HTLV-1 basic domain/leucine zipper (bZIP) factor, HBZ, interacts with CREB and the transcriptional coactivators p300/CBP and inhibits Tax-dependent transactivation from the retroviral promoter [9]. HBZ also counters the negative regulation of Tax expression by p30^II by reducing the level of p30II mRNA [11]. Also, Andresen et al. recently demonstrated that the HTLV-1 p13^II protein binds to Tax and interferes with the recruitment of p300/CBP to Tax-containing promoter complexes [10]. The retroviral pX ORF-I and ORF-II products, p12^I and p30^II, are important for infectivity and the maintenance of HTLV-1-immortalized cells in in vivo models of pathogenesis [5]. In addition to facilitating immune-escape by HTLV-1-infected cells, it is possible that the repression of Tax-dependent transactivation and expression of retroviral antigens prevents cellular apoptosis induced by viral proteins. Tax has been shown to induce caspase-dependent apoptosis through the titration of limiting amounts of nuclear p300 [12]. The p30^II protein interacts with the transcriptional cofactors, p300/CBP and TIP60, and deregulates cellular transcription pathways, associated with genomic instability/multinucleation, increased lymphoproliferation, and oncogenic transformation [6,13]. p30^II has also been shown to modulate ATM kinase activity and inhibit the cellular DNA damage-response which could promote the accumulation of genetic lesions [14,15]. The p13^II protein targets mitochondria and sensitizes T-cells to oncogene-induced cell death. Thus, it cannot be excluded that the reduced ATL tumor burden and prolonged survival of AR-42-treated NOD/SCID animals, engrafted with HTLV-1-infected MET-1 cells, in the present study could be attributable, at least in part, to increased proviral gene expression and apoptosis induced by retroviral proteins [1]. Although the authors did not observe significant changes in Tax expression in the HDAC-I-treated HTLV-1-transformed T-cell-lines, MT-2 and C8166, it remains to be determined whether HDAC-Is influence proviral gene expression in the tumorigenic ATL-derived MET-1 cells.

It is noteworthy that the HDAC-Is, VPA and AR-42, induce Poly (ADP-ribose) polymerase-cleavage and apoptosis in HTLV-1-transformed T-cells, which have impaired p53 functions due to Tax-dependent inactivation. Others have demonstrated that the wildtype p53 tumor suppressor present in HTLV-1-infected lymphocytes can be reactivated by certain small molecules that inhibit NF-kappa-B [16]. Indeed, Mori et al. have previously shown that another HDAC-I, FR901228, inhibits NF-kappa-B activation and induces apoptosis in HTLV-1-transformed T-cells and primary ATL leukemic lymphocytes in vitro [17]. As the inactivation of p53 by Tax is dependent upon the ability of Tax to activate NF-kappa-B-signaling pathways [18], it will be interesting to determine whether AR-42 interferes with NF-kappa-B transactivation in the current NOD/SCID model of HTLV-1-induced T-cell lymphomagenesis [1].

Histone deacetylases are recruited to repressor complexes on the promoters of transcriptionally inactive genes, and negatively regulate transcription by removing the acetyl chemical moiety from the ε-NH₃⁺ group of acetylated lysines in histones H3 and H4. This causes the phosphate-sugar backbone of DNA to interact tightly with positively charged nucleosomes and condenses the chromatin architecture into a suppressive state. Pharmacological inhibitors of HDACs are being widely investigated in numerous preclinical and clinical studies as anti-inflammatory modulators, treatments for neuropsychiatric and cognitive disorders, and as chemotherapies against solid tumors and hematological malignancies, including small cell lung cancers (SCLCs), non-Hodgkin’s lymphoma (NHL), mantle cell lymphoma, chronic lymphocytic leukemia (CLL), cutaneous T-cell lymphoma (CTCL), as well as myelodysplastic syndrome. Others have previously demonstrated that HDAC-Is, such as FR901228, LBH589, MS-275, and suberoylanilide hydroxamic acid (SAHA), inhibit cellular proliferation and induce apoptosis in cultured HTLV-1-transformed T-cell-lines and ATL leukemic lymphocytes [17,19,20]. The present study extends these findings by demonstrating the efficacy of a novel HDAC-I, AR-42, in an in vivo preclinical model of ATL [1].

The general success of various HDAC-Is in Phase I and II clinical trials of anticancer chemotherapies, suggests that these pharmacological agents could hold great promise for the eventual treatment of ATL patients. Human T-lymphotropic retrovirus type-1-infected ATL cells are refractive to most therapeutic strategies currently in use, which include CHOP (cyclophosphamide, hydoxydaunorubicin, oncovin, prednisolone)-regimen chemotherapy, arsenic trioxide, interferon-alpha in combination with zidovudine, proteasome inhibitors, a monoclonal antibody against the CCR4 chemokine receptor, and allogeneic stem cell transplantation. The discovery by Zimmerman et al. that an orally bioavailable HDAC-I reduces tumor burden and increases the survival of engrafted animals in a NOD/SCID model of ATL [1], represents a hopeful advance for the future clinical management of HTLV-1-induced lymphomalignancies.