HTLV-1 Rex: the courier of viral messages making use of the host vehicle

Abstract

The human T-cell leukemia virus type 1 (HTLV-1) is a retrovirus causing an aggressive T-cell malignancy, adult T-cell leukemia (ATL). Although HTLV-1 has a compact RNA genome, it has evolved elaborate mechanisms to maximize its coding potential. The structural proteins Gag, Pro, and Pol are encoded in the unspliced form of viral mRNA, whereas the Env protein is encoded in singly spliced viral mRNA. Regulatory and accessory proteins, such as Tax, Rex, p30II, p12, and p13, are translated only from fully spliced mRNA. For effective viral replication, translation from all forms of HTLV-1 transcripts has to be achieved in concert, although unspliced mRNA are extremely unstable in mammalian cells. It has been well recognized that HTLV-1 Rex enhances the stability of unspliced and singly spliced HTLV-1 mRNA by promoting nuclear export and thereby removing them from the splicing site. Rex specifically binds to the highly structured Rex responsive element (RxRE) located at the 3' end of all HTLV-1 mRNA. Rex then binds to the cellular nuclear exporter, CRM1, via its nuclear export signal domain and the Rex-viral transcript complex is selectively exported from the nucleus to the cytoplasm for effective translation of the viral proteins. Yet, the mechanisms by which Rex inhibits the cellular splicing machinery and utilizes the cellular pathways beneficial to viral survival in the host cell have not been fully explored. Furthermore, physiological impacts of Rex against homeostasis of the host cell via interactions with numerous cellular proteins have been largely left uninvestigated. In this review, we focus on the biological importance of HTLV-1 Rex in the HTLV-1 life cycle by following the historical path in the literature concerning this viral post-transcriptional regulator from its discovery to this day. In addition, for future studies, we discuss recently discovered aspects of HTLV-1 Rex as a post-transcriptional regulator and its use in host cellular pathways.

Keywords: B-23; CRM1; HIV-1 Rev; HTLV-1 Rex; HTLV-2; importinβ; post-translational regulator; retroviruses.

FIGURE 1

Concerted functions of viral proteins for HTLV-1 expression. Postinfection, the HTLV-1 provirus expresses viral proteins at appropriate times to control the early productive phase and the late shut-down phase leading to latency in the HTLV-1 life cycle. At the very beginning, without Rex, the viral transcripts are fully spliced and thus, Tax and Rex are selectively translated (stage 1). Tax transactivates HTLV-1 LTR promoter activity, whereas Rex inhibits splicing and actively exports the unspliced and singly spliced viral mRNA from the nucleus resulting in the expression of structural proteins and production of viral particles (stage 2). In the late phase, p30II from a minor, doubly spliced transcript binds to tax/rex mRNA and confines it to the nucleoli (stage 3) resulting in decreased Tax/Rex protein levels leading to latency. N, nucleus; C, cytoplasm; Nu, nucleolus.

FIGURE 2

Molecular mechanism of HTLV-1 Rex function. HTLV-1 Rex specifically binds to the RxRE motif of HTLV-1 transcripts. Rex also interacts with the cellular nucleocytoplasmic shuttling protein, CRM1, through its NES. Consequently, the Rex–viral mRNA complex is exported from the nucleus by CRM1. In the cytoplasm, Rex subjects viral transcripts to the cellular translational machinery to enhance viral production. Released Rex binds to importinβ via its NLS and returns to the nucleus by the importin complex shuttling activity. P30II binds to Rex through its NLS and retains Rex in the nucleolus for suppression. Rex not only transports viral transcripts, but also inhibits splicing of viral mRNA that encode structural proteins. hnRNP A1, which governs the processing/splicing of pre-mRNA and transport of mature mRNA, was found to bind to RxRE in a competing manner against Rex. Another major splicing factor, SF2/ASF, was found to influence the processing of HTLV-1 mRNA (i.e., overexpression of SF2/ASF resulting in differential pX splice site utilization), although the direct physiological interaction to the viral proteins has not been examined. Recently, Rex was shown to directly interact with Dicer and inhibit its processing of shRNA to siRNA (Abe et al., 2010). Overall, interactions between Rex and other cellular mRNA processing proteins may lead to an unknown molecular mechanism of Rex in the inhibition of the splicing machinery. N, nucleus; C, cytoplasm; Nu, nucleolus.

FIGURE 3

Similarities and differences between Rex-1 and Rex-2. The phylogenetic tree of HTLV, which is drawn based on the report by Vandamme et al. (1998), shows that the major branches of PTLV-1 and PTLV-2 separated at an early stage. Sub-branches of STVL-1 and HTLV-1 or STLV-2 and HTLV-2 were separated within each major branch thereafter. Yet, the genomic structures of HTLV-1 and HTLV-2 are very similar and both viruses encode Tax and Rex, the major transcriptional and post-transcriptional regulators, respectively. Both Tax-1 and Tax-2 have NLS, NES, and ATF/CREB binding domains, whereas only Tax-1 has a distinct NF-κB activating domain and p300 binding domain, as well as a number of PTM sites, for phosphorylation, ubiquitination, and SUMOylation, resulting in stronger transactivation and transforming activities than those of Tax-2. Both Rex-1 and Rex-2 are phosphoproteins sharing 60% similarity with overlapped major functional domains, such as NLS, NES, and SD at the 3′-terminus. The RxRE motif of HTLV-1 mRNA (RxRE-1) is located in the U3/R region and all HTLV-1 mRNA have intact RxRE-1s in the 3′UTRs. On the other hand, the RxRE of HTLV-2 (RxRE-2) is located at the R/U5 region and only unspliced mRNA maintains the intact RxRE-2. Thus, Rex-1 is capable of transporting all viral mRNA including tax/rex mRNA and enhancing the expression of Tax for further transactivation of LTR, whereas Rex-2 is not. Overall, Rex-1 may have a stronger impact on viral replication through the enhancement of Tax-1 expression compared with Rex-2. Different roles of Tax and Rex may be related to the differences in pathophysiologies of HTLV-1 and HTLV-2.

FIGURE 4

Similarities and differences between HTLV-1 Rex and HIV-1 Rev. HTLV-1 and HIV-1 are evolutionally distinct, but both belong to a family of complex retroviruses. Both viruses encode transactivators, HTLV-1 Tax and HIV-1 Tat, and post-transcriptional regulators, HTLV-1 Rex and HIV-1 Rev. Although Tax and Tat transactivate their respective LTRs, they act though different mechanisms and cannot be replaced with each other. On the other hand, even though the homology in the primary sequence of HTLV-1 Rex and HIV-1 Rev is low, they carry out similar functions through common domains, such as NES, NLS, and multimerization domains. They are RNA binding proteins and specifically bind to respective viral mRNA with high affinity through RxRE for Rex and RRE for Rev. They stabilize unspliced or partially spliced viral mRNA and actively transport them to the cytoplasm via CRM1 binding for selective translation of viral structural proteins and return to the nucleus by interactions with Importinβ. Rex can function through RRE, while Rev cannot bind to RxRE. HTLV-1 Tax is translated only from fully spliced viral mRNA; thus, stabilization and active nuclear transport of unspliced HTLV-1 mRNA by Rex reduces the relative expression rate of Tax. On the contrary, HIV-1 Rev does not suppress Tat activity, since it enhances truncated, yet active, Tat proteins. Thus, it is expected that HTLV-1 Rex favors reduction of viral production, whereas HIV-1 Rev may support active viral production. N, nucleus; C, cytoplasm; Nu, nucleolus.

FIGURE 5

Interactions of Rex with host pathways, uncovered and covered. Besides the known interactions between Rex and CRM-1, importinβ, and B-23, a number of potential interactions between Rex and cellular proteins, based on the high-throughput yeast two-hybrid system, were reported by Simonis et al. (2012). Rex is suspected of interacting with a series of proteins that play crucial roles in mRNA surveillance, nucleocytoplasmic shuttling, tumor growth regulation, and SUMOylation. Air1 (ZCCHC7) is a component of the TRAMP complex, which is involved in nuclear mRNA surveillance. NUP62 is one of three nucleoporins (NUP54, 58, and 62) composing the nuclear pore complex and is essential for nuclear transport. LZTS2 is a tumor suppressor and its expression is transcriptionally regulated by NF-κB. LZTS2 expression levels affect cell proliferation and tumor growth through the Wnt/β-catenin pathway in various cancer cell lines. An E2 SUMO ligase, UBC9, and an E3 SUMO ligase, PIAS2, are also expected to interact with Rex. Rex is also expected to interact with SP100, a major component of the nuclear body, which has a transactivating function and is induced in stimulated and malignant cells. Since Rex impacts the host cell through an unknown mechanism, such as increasing il-2rα mRNA level and fyn-b mRNA expression level by enhancing unusual exon 7 usage, unknown interactions between Rex and cellular proteins may be related to Rex functions. Solid lines indicate reported interactions. Dashed lines indicate potential interactions. N, nucleus; C, cytoplasm; NB, nuclear body.