Inhibition of HIV replication by dominant negative mutants of ... - Nature

2 downloads 0 Views 445KB Size Report
domain also constitutes an RNA-binding domain that specifi- cally interacts with the cognate target sequence, Rev response - element (RRE). Substitutions of ...
© 1999 Nature America Inc. • http://medicine.nature.com

ARTICLES

Inhibition of HIV replication by dominant negative mutants of Sam68, a functional homolog of HIV-1 Rev THIPPARTHI R. REDDY1, WEIDONG XU1, JONATHAN K. L. MAU2, CHRISTOPHER D. GOODWIN1, MODEM SUHASINI1, HENGLI TANG2, KENNETH FRIMPONG1, DAVID W. ROSE3 & FLOSSIE WONG-STAAL1,2 Departments of 1Medicine and 2Biology, University of California San Diego, La Jolla, California 92093, USA Whittier Diabetes Program and Veteran’s Medical Research Foundation, University of California San Diego, La Jolla, California 92093, USA Correspondence should be addressed to F.W.-S.

© 1999 Nature America Inc. • http://medicine.nature.com

3

The HIV-1 Rev protein facilitates the nuclear export of mRNA containing the Rev response element (RRE) through binding to the export receptor CRM-1. Here we show that a cellular nuclear protein, Sam68 (Src-associated protein in mitosis), specifically interacts with RRE and can partially substitute for as well as synergize with Rev in RRE-mediated gene expression and virus replication. Differential sensitivity to leptomycin B, an inhibitor of CRM-1, indicates that the export pathways mediated by Rev and Sam68 are distinct. C-terminally deleted mutants of Sam68 inhibited the transactivation of RRE-mediated expression by both wild-type Sam68 and Rev. They were retained in the cytoplasm and impeded the nuclear localization of Rev in coexpressed cells. These mutants also inhibited wild-type HIV-1 replication to the same extent as the RevM10 mutant, and may be useful as anti-viral agents in the treatment of AIDS.

Complex retroviruses, typified by HIV-1, encode an essential regulatory protein Rev, which mediates the nuclear export of unspliced or incompletely spliced viral RNA (refs. 1,2). Rev is a shuttle protein3, composed of a basic, nuclear localization sequence4 and a leucine-rich nuclear export sequence5. The basic domain also constitutes an RNA-binding domain that specifically interacts with the cognate target sequence, Rev response element (RRE). Substitutions of leucine and glutamic acid residues within nuclear export sequence yield a mutant Rev protein (RevM10) with a dominant negative phenotype4. RevM10 confers human CD4 cells with antiviral resistance in cell culture6 and preferential survival in HIV-1-infected patients7,8. The Rev nuclear export sequence binds the nuclear export receptor CRM1, a member of the importin-β family9. This interaction is functionally relevant, as leptomycin B (LMB), a drug that disrupts the complex formation of Rev, CRM-1 and RanGTP, also inhibits the nuclear export of Rev–nuclear export sequence conjugates9. CRM-1 probably bridges the indirect interaction of Rev with members of the nucleoporin family10, including Rab/hRIP, reported to be Rev-binding proteins11,12. Additional cellular proteins that bind Rev (refs. 13–15) and/or RRE (refs. 16,17) have been identified, which modulate Rev activity either positively or negatively. However, no cellular counterpart of Rev has been reported. Here, we demonstrate that an RNA-binding protein, Sam68, binds to RRE in vitro and in vivo, and functionally replaces Rev in RRE-mediated gene expression and virus replication. Furthermore, Sam68 synergizes with Rev and Cterminally deleted mutants of Sam68 act as potent inhibitors of HIV replication. Sam68 is a target of the c-Src tyrosine kinase18,19. It is also phosphorylated at threonine residues by the cyclin-dependent protein kinase Cdc2 (ref. 20). Such phosphorylation events occur specifically during mitosis. Sam68 contains an hnRNP K homolNATURE MEDICINE • VOLUME 5 • NUMBER 6 • JUNE 1999

ogy domain (KH domain) that mediates RNA binding and protein–protein interaction, as well as proline-rich sequences that facilitate binding to other proteins21,19. A nuclear localization signal has been mapped to a C-terminal tyrosine-rich region22. A splice variant of Sam68 with a deletion in the KH domain is specifically expressed at growth arrest, and overexpression of this variant is able to inhibit DNA synthesis23, indicating that Sam68 isoforms are involved in the regulation of the G1/S transition of the cell cycle. The exact mechanism of action of Sam68 is not known. It is postulated to be involved in the post-transcriptional regulation of gene expression, but no definitive evidence has been gathered so far. Our results provide direct evidence that Sam68 is involved in post-transcriptional regulation of gene expression. The observation that Sam68 synergizes with HIV-1 Rev and the derivation of Sam68 mutants that inhibit HIV replication indicate new research avenues into HIV pathogenesis and therapeutic intervention. Sam68 enhances RRE-directed reporter gene expression To explore a potential role for Sam68 in RRE-mediated transactivation, we assessed the effect of exogenously expressed Sam68 under the control of the CMV promoter on an RRE-regulated reporter gene (RRE–CAT; chloramphenicol acetyltransferase) in transient co-transfection assays. To more easily distinguish Sam68 transgene expression from endogenous Sam68, we also constructed a mutant with a 96-amino-acid deletion from the N terminus (Sam68∆96). Deletion of 103 amino acids from the N terminus does not affect the RNA binding and multimerization of Sam68 (ref. 21). We studied both wild-type and a variety of mutant Sam68 constructs (Fig. 1a). Sam68 is expressed at low levels but ubiquitously in human cells (data not shown). Both wild-type Sam68 and Sam68∆96 were over-expressed in transfected cells (Fig. 1a), and both induced a 12-fold to15-fold increase in RRE635

© 1999 Nature America Inc. • http://medicine.nature.com

c

b

CAT activity (CPM) I

II

mediated CAT reporter gene expression over basal levels, whereas Rev expression yielded a 20-fold to 30-fold increase (Fig. 1b). The increase in CAT expression mediated by Sam68∆96 was dose dependent, and at times was as efficient as that with Rev (data not shown). Therefore, we used Sam68∆96 instead of the full-length gene in most of our subsequent studies. We also assessed the effect of a mutant in which the KH domain was deleted (Sam68∆96∆KH; ref. 23). The KH domain of Sam68 is important

a

Splice acceptor

CMV promoter

Splice acceptor pCM228 (RRE-LacZ)

LacZ Reporter alone

+Rev

+sam68∆96

8.5 ± 1.1

39.6 ± 6.7

LacZ expression

Injected cells % blue cells

44.0 ± 5.8

III

IV

V

VI

VII

VIII

for self-association as well as RNA binding21. Sam68∆96∆KH failed to enhance the RRE-mediated transactivation (Fig. 1b). Similarly, mutants deleted at the N terminus (∆42-329) and C terminus (∆330-443 and ∆410-443) did not transactivate RRE-directed gene expression (Fig. 1b). As negative controls, we examined the effect of hnRNP A1, another RNA binding protein, on RRE-directed transactivation (data not shown), as well as the effect of Sam68∆96 on other expression constructs, including CTE–CAT (constitutive transport element of type-D retroviruses, a cellular functional analogue of Rev-RRE; ref. 24) and HIV LTR–CAT (CAT reporter gene linked to the HIV long terminal repeat; Fig. 1c). The results indicate that transactivation of RRE by Sam68 is specific. We obtained independent confirmation of the specific effects of Sam68∆96 on RRE-dependent activation by microinjection assays25. Co-injection of pSam68∆96 expression vector into the nuclei of HS68 cells transactivated the expression of RRE-directed lacZ (pCM228) as efficiently as pRev expression vector at the same concentration (Fig. 2a). Of the pRev- and pSam68∆96injected cells, 39% and 44%, respectively, stained blue. In contrast, pSam68∆96 did not stimulate the expression of a control reporter plasmid (3XUASp36LacZ), which has low basal activity (Fig. 2b), again indicating that Sam68∆96 does not activate reporter gene expression through a general transcriptional activation mechanism.

LacZ

b UAS × 3

Reporter alone

+sam68∆96

LacZ expression

Injected cells % blue cells

636

Sam68∆96

Endogenous

a

Relative CAT activity (fold)

© 1999 Nature America Inc. • http://medicine.nature.com

Fig. 1 Effects of wild-type and mutant Sam68 on RRE-mediated CAT expression. a, Wild-type and mutant Sam68. P, Proline-rich motifs; KH, K homology domain; RGG, Arg-Gly-Gly; 1 and 443, amino acid numbers. Right, lane 1 (endogenous Sam68), extracts from untransfected cells; lanes 2 and 3, wild-type and ∆96 mutant; in lane 3, the two bands could represent different phosphorylated forms of the protein. b, Sam68 and Sam68∆96 enhance RRE directed reporter gene expression. CAT activity of 293T cells transfected with RRE–CAT alone (I) or RRE–CAT plus Sam68 (II), Sam68∆96 (III), Sam68∆96∆KH (IV), Sam68C’∆330-443 (V), Sam68C’∆410-443 (VI), Sam68∆42-329 (VII) and/or Rev (VIII). Data represent fold increase relative to basal levels (RRE–CAT alone) c, RREindependent expression constructs. CAT assay of cells transfected with CTE–CAT or HIV LTR–CAT Sam68∆96 and/or Tat (+, present; –, absent). DNA was equalized to 2.5 µg for each transfection.

Sam68 (wild-type)

ARTICLES