LETTERS NATURE MEDICINE VOLUME 10 | NUMBER 2 | FEBRUARY 2004 197 Human T-cell leukemia/lymphoma virus type 1 (HTLV-1) persists despite a vigorous virus-specific host immune response, and causes adult T-cell leukemia and lymphoma in approximately 2% of infected individuals. Here we report that HTLV-1 has evolved a genetic function to restrict its own replication by a novel post-transcriptional mechanism. The HTLV-1-encoded p30 II is a nuclear-resident protein that binds to, and retains in the nucleus, the doubly spliced mRNA encoding the Tax and Rex proteins. Because Tex and Rex are positive regulators of viral gene expression 1,2 , their inhibition by p30 II reduces virion production. p30 II inhibits virus expression by reducing Tax and Rex protein expression. The stoichiometry and catalytic activity of Tax, the viral transactiva- tor 3–5 , determine T-cell progression through the G1 phase of the cell cycle (reviewed in ref. 6). The Rex protein promotes viral production by regulating the transport of genomic and envelope viral mRNA to the cytoplasm 7 , and influences the expression of other cellular genes (reviewed in ref. 6). Because Tax is highly immunogenic 8,9 , we hypothesized that HTLV-1 may have evolved a dedicated genetic function to reduce the expression of viral proteins (including Tax) and become transiently dormant in order to evade host immune sur- veillance—a strategy commonly used by DNA viruses 10 . To this end, we chose to investigate the effect on viral replication of the HTLV-1 p30 II protein, encoded by the doubly spliced mRNA from open read- ing frame II, because the nuclear and nucleolar localization of this protein suggested that it might have regulatory function(s). We coexpressed the HTLV-1 molecular clone p-BST (ref. 11) with the cDNA encoding p30 II , and measured viral p19 Gag in cell super- natants (Fig. 1a). p30 II decreased viral production in a dose- dependent manner, even when additional Tax was provided (Fig. 1a). HTLV-1 molecular clones p-CSH (ref. 12) and p-ACH (ref. 13) were also inhibited by p30 II (Fig. 1b,c, respectively). The decrease in p24 Gag and p19 Gag did not result from inhibition of the viral protease, as the p55 Gag precursor was cleaved normally in the presence of p30 II (Fig. 1d). As Tax alone was unable to counteract the negative effect of p30 II (Fig. 1a), we assessed whether Rex, the other positive regulator of viral gene expression, could restore viral production. Even at a high concentration, Rex did not substantially restore viral production in the presence of p30 II (Fig. 1e). Together, these data suggest that nei- ther Tax nor Rex alone is sufficient to reverse the negative effect of p30 II on viral replication. We next investigated whether the negative effect of p30 II may be due to inhibition of Tax-stimulated transcription. We cotransfected p30 II with a cDNA construct expressing Tax from a cytomegalovirus (CMV) promoter, and tested Tax transcriptional activity on both a viral long-terminal repeat (LTR) and a NF-κB–luciferase (Luc) reporter gene construct, as Tax also affects the NF-κB transcription pathway at several levels 14 . p30 II did not affect basal (Fig. 2a) or Tax- mediated (Fig. 2b) transcription from the viral LTR. Similarly, there was no effect on transcription of the NF-κB–Luc reporter construct (Fig. 2c). We also studied the expression and cellular localization of Tax in the presence of p30 II . p30 II did not affect the expression (Fig. 2b) or cellular localization (Fig. 2d–i) of Tax protein. Having excluded a direct effect of p30 II on Tax-mediated tran- scription, we assessed whether Tax is affected by p30 II at a post- transcriptional level. We coexpressed an HTLV-1 LTR–Luc reporter gene with the p-BST molecular clone, in the presence of increasing amounts of p30 II . Under this condition, Tax was produced from the doubly spliced mRNA transcript originating from the HTLV-1 molecular clone. p30 II decreased Tax activity on the viral LTR in a dose-dependent manner (Fig. 3a), suggesting a post-transcriptional inhibition of Tax by p30 II . To further investigate the mechanism of p30 II suppression of viral gene expression, we performed semiquantitative or real-time RT- PCR on total, nuclear and cytoplasmic RNA from cells cotransfected with the p-BST provirus and a p30 II expression construct. In part of this experiment, we provided additional Tax to increase the level of viral mRNA. p30 II did not substantially alter the levels of Gag-Pol, Env and p21 Rex spliced mRNA 15 in the total cellular RNA of transfected cells, both in the absence and presence of Tax (Fig. 3b,c). We observed only a modest decrease in the bicistronic Tax/Rex mRNA at the highest concentration of p30 II (Fig. 3c); this effect was negligible when Tax 1 Animal Models and Retroviral Vaccines Section and 2 Laboratory of Receptor Biology and Gene Expression, National Cancer Institute, 41/D804, Bethesda, Maryland 20892, USA. 3 Basic Research Laboratory, Center for Cancer Research, National Cancer Institute at Frederick, Frederick, Maryland 21702-1201, USA. 4 Present address: Department of Microbiology, University of Kansas Medical Center, 3025 Wahl Hall West, Kansas City, Kansas 66160, USA. Correspondence should be addressed to G.F. (franchig@mail.nih.gov). Published online 18 January 2004; doi:10.1038/nm984 HTLV-1-encoded p30 II is a post-transcriptional negative regulator of viral replication Christophe Nicot 1,4 , Miroslav Dundr 2 , Julie M Johnson 1 , Jake R Fullen 1 , Norma Alonzo 1 , Risaku Fukumoto 1 , Gerald L Princler 3 , David Derse 3 , Tom Misteli 2 & Genoveffa Franchini 1 © 2004 Nature Publishing Group http://www.nature.com/naturemedicine