Identification and Purification of Cellular Proteins That Specifically Interact with the RNA Constitutive Transport Elements from Retrovirus D

VIROLOGY

228, 333–339 (1997) VY968407

ARTICLE NO.

Identification and Purification of Cellular Proteins That Specifically Interact with the RNA Constitutive Transport Elements from Retrovirus D HENGLI TANG,* YUAN XU,†,1 and FLOSSIE WONG-STAAL*,†,2 *Department of Biology and †Department of Medicine, University of California, San Diego, La Jolla, California 92093-0665 Received September 19, 1996; returned to author for revision October 29, 1996; accepted December 12, 1996 Human immunodeficiency virus (HIV) encodes a transacting protein, Rev, which interacts with an RNA element (RRE) to mediate nuclear export of unspliced viral mRNA. Recently, the RNA constitutive transport elements (CTE) from MasonPfizer monkey virus (MPMV) and simian retrovirus type I (SRV-1) were shown to render Rev-independent expression of gag, pol, or env genes in subgenomic constructs of HIV-1 and to support replication of HIV-1 mutants lacking RRE and Rev. Since CTEs act in cis, in the absence of any viral regulatory proteins, it is widely believed that they interact directly with the cellular export machinery by means of RNA–protein interaction. In this report, Electrophoretic mobility shift and UVcrosslinking assays were carried out to identify nuclear proteins that interact specifically with CTEs from both MPMV and SRV-1. Two of the four proteins (65 and 40 kDa, respectively) that bound CTE RNA did not interact in the same assays with RNA of a nonfunctional CTE mutant generated by site-directed mutagenesis. Both proteins have been partially purified. The nature of these proteins and their roles in RNA intracellular trafficking are currently under investigation. q 1997 Academic Press

INTRODUCTION

transport pathway (Ruhl et al., 1993; Bogerd et al., 1995; Fritz et al., 1995; Stutz et al., 1995; Bevec et al., 1996). The situation for the simple retroviruses is more puzzling because they do not seem to encode any Rev-like transacting proteins, and yet, like the complex retroviruses, they also utilize unspliced RNAs to produce their structural (gag and pol) proteins. One way to circumvent this problem is to encode cis-acting elements in their transcripts to interact directly with one of the cellular RNA export systems. The constitutive transport element (CTE) from Mason-Pfizer monkey virus (MPMV) and simian retrovirus type-1 (SRV-1) have recently been identified as such cis-acting elements (Bray et al., 1994; Zolotukhin et al., 1994). Since they render Rev-independent expression of gag, pol, or env genes in HIV subgenomic constructs and can support replication of HIV-1 Rev- and RRE-deficient mutants in the absence of viral regulatory proteins, it is likely that these RNA elements interact directly with components of a particular cellular machinery by means of RNA–protein interaction. In this report, we carry out studies to identify and partially purify nuclear proteins that interact specifically with functional CTEs from both MPMV and SRV-1.

Retroviruses utilize a single primary transcript that is processed to engender multiple mRNAs encoding all the viral structural and regulatory proteins (reviewed by Vaishnav and Wong-Staal, 1991; Cullen, 1992). For the human immunodeficiency viruses (HIV), the multiply spliced mRNAs are produced in the early phase of infection and encode regulatory proteins such as Tat, Rev, and Nef, while the singly spliced and unspliced transcripts involved in the production of structural and accessory proteins are made in the late phase. This temporally regulated expression of spliced and unspliced RNA is achieved by posttranscriptional regulation. Export of the RNAs from the nucleus to the cytoplasm is an important aspect of this regulation. While the multiply spliced mRNAs can be transported into the cytoplasm and get translated as normal cellular messenger RNAs, the singly spliced and unspliced RNAs apparently cannot be transported by the usual cellular mRNA export machinery and need a separate mechanism for export. In complex retroviruses, expression from unspliced and singly spliced RNAs is regulated by a viral protein and its cognate RNA element, namely, Rev and RRE for HIV and Rex and RxRE for human T-cell lymphotropic virus (HTLV), respectively. Several Rev cofactors have been cloned and implicated to be involved in a cellular

MATERIALS AND METHODS Plasmid construction Molecular cloning was carried out according to standard procedures (Sambrook et al., 1989). Plasmids pSRV1 and p72DXB9DBgl2 No. 18 were kind gifts from Dr. Paul Luciw, pSRV-1 contains a 8.2-kb permuted form of

1

Present Address: Genentech, Inc., South San Francisco, CA 94080. To whom correspondence and reprint requests should be addressed. Fax: (619) 534-7743. E-mail: [email protected]. 2

0042-6822/97 $25.00

333

AID

VY 8407

/

6a29$$$101

01-22-97 06:15:20

Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.

vira

AP: Virology

334

TANG, XU, AND WONG-STAAL

SRV-1 genome, and p72DXB9DBgl2 No. 18 carries MPMV sequence 7886–8558. They serve as sources for CTEs of SRV-1 and MPMV, respectively. MPMV sequence 8005–8240 and SRV-1 sequence 7620–7859 were amplified by PCR and cloned into pDM138 (gift from Dr. Tom Hope, Donello et al., 1996) and pSP72 (Promega) using ClaI sites that were introduced by the PCR primers, giving rise to plasmids pDM M240, pDM S240, p72 M240, and p72 S240. All plasmids have their antisense version in which the transcripts would represent the antisense RNA of the corresponding parts of viral transcripts. Plasmid pDM M21 is generated by PCR amplifying MPMV sequence 8031–8240 and cloning into the ClaI site of pDM138. pDM M19 and pDM M14 were made in a similar way with MPMV sequences 8005–8190 and 8005–8140 cloned into pDM 138, respectively. Mutagenesis were carried out by overlapping PCR on pDM M240 (Ho et al., 1989), and the appropriate mutant form of CTE was cloned into pSP72, generating p72-D240. Transfection and CAT assays HeLa cells were transfected with the standard calcium phosphate precipitation method. Transfection efficiencies were normalized by cotransfecting a b-gal reporter plasmid and measuring b-gal activity before the CAT assays. CAT assays were done by the 14C method and after autoradiography, the radioactive spots were cut out and radioactivity was measured by scintillation counter to determine the percentage of conversion of chloramphenicol to acetylchloramphenicol.

other 15 min at room temperature. The RNA–protein complexes were then resolved on a 5% polyacrylamide gel (acrylamide:bis Å 60:1) with 5% glycerol. For competition experiments, various cold RNAs were included in the preincubation step and the volume of reaction was kept constant by nuclease-free water. For UV-crosslinking, approximately 20–30 mg nuclear extract or a column fraction was incubated at room temperature for 10 min in the binding buffer containing 12 mM K/HEPES (pH 7.9), 250 mM NaCl, 40 mM MgCl2 , 100 mM DTT, 2 mg of bovine liver rRNA (Sigma), 5 mg of yeast tRNA (Clontech), 20 units RNasin, and 10% glycerol. Reaction volumes were 30 ml for nuclear extract and 40 ml for column fractions. For competition experiments, cold RNAs were included at this incubation step. 32P-labeled RNA probes (1 1 106 cpm) were then added and incubation continued for another 10 min. Crosslinking was done by exposing the reaction mixture to a UV source of 254 nm, 7000 mW/cm2 at a distance of 2 cm, on ice. RNase was added to digest the unbound probe and the samples were then electrophoresed on a 12% SDS–polyacrylamide gel, which was subsequently dried and autoradiographed. Protein purification

Nuclear extracts from HeLa and H9 cells were prepared according to Dignam (Dignam et al., 1983). A gel shift assay was carried out as follows: 10 mg of nuclear extract was preincubated in a binding buffer (12 mM HEPES, pH 7.9, 50 mM NaCl, 40 mM MgCl2 , 100 mM DTT, 10 units RNasin, and 5% glycerol) at room temperature for 15 min, then 5 1 104 cpm of radioactive probe was added to the mixture, and the reaction was carried out for an-

Protein purification procedures were detailed in a previous study (Xu Y. et al., 1996). Briefly, a nuclear extract of 10 liters of H9 cells was applied to a mono-S column, the flow-through proteins were loaded onto a mono-Q column, the bound proteins were eluted with 500 mM NaCl, and (NH4)2SO4 was added to a final concentration of 1.0 M, the protein solution was then applied to a butyl– Sepharose 4 fast-flow column, and the bound proteins were eluted with 0.5 M (NH4)2SO4 . All fractions were tested for CTE binding activity by crosslinking and fractions containing the appropriate crosslinking band were subjected to biotinylated RNA selection. Selection of proteins on biotinylated RNA was carried out according to Yeakley et al. (1996). Briefly, 50 ml of streptavidin–agarose beads (Life Technologies) was washed with 5 ml of WB350 (20 mM Tris, pH 7.8, 350 mM KCl, 0.01% Nonidet P-40) and resuspended in 50 ml WB350 with 500 mg/ml yeast tRNA. The beads were blocked by tRNA at room temperature for 10 min while the RNA–protein interaction was carried out: about 60– 100 mg of biotinylated RNA was incubated with 1.4 mg nuclear extract in the presence of 500 mg yeast tRNA in 11 FSP (20 mM Tris, pH 7.8, 60 mM KCl, 2.5 mM EDTA, 0.1% Triton X-100) on ice for 10 min. The sample volume was brought up to 1 ml with 11 FSP, the salt concentration adjusted by adding 150 ml 2 M KCl, and then the agarose beads were added and the sample was rocked overnight in a cold room. The beads were then washed three times with WB350 and then once in 11 FSP. Proteins were eluted by heating in SDS sample buffer for 5

AID

vira

In vitro transcription Large amounts of cold RNAs were made by a T7 Megashortscript in vitro transcription kit (Ambion). Biotinylated RNAs were made with the same kit, with the inclusion of biotin-21–UTP (Clontech) in the transcription reaction; the amount of biotin-21–UTP is controlled to ensure incorporation of three to four biotins per RNA molecule. High specificity radioactive RNAs were made by an in vitro transcription kit from Promega, using [32P]UTP as label. After the reactions, RNAs were either purified by polyacrylamide gel or directly precipitated by ethanol depending on the application of the RNA. RNA–protein interaction assays

VY 8407

/

6a29$$$102

01-22-97 06:15:20

AP: Virology

CTE BINDING PROTEINS

335

FIG. 1. CTE interacts with nuclear proteins in vitro. (A) RNA gel shift assay of radioactive MPMV CTE RNA incubated with HeLa nuclear extract in the presence or the absence of proteinase K. (B) Competition of the protein binding with yeast tRNA and cold CTE RNA. The competitors are of 0, 20, 80, and 3201 excess of the radioactively labeled CTE RNA.

min at 657 and spun down to pellet the beads and the supernatant was run on a 10% PAGE gel. RESULTS CTE interacts with nuclear proteins in vitro Gel shift assays were first carried out to establish that CTE RNA does interact with nuclear factors specifically. Nuclear extract was incubated with radioactively labeled CTE RNA in the presence or the absence of proteinase K. A shift was observed only when nuclear extract was included in the reaction and proteinase K was not (Fig. 1A), indicating that CTE RNA binds to a certain protein factor(s) in the nuclear extract. To show the specificity of the interaction, yeast tRNA (Fig. 1B, lanes 2–5), bovine liver rRNA (data not shown) and cold CTE RNA (Fig. 1B, lanes 6–9) were included in the preincubation step as competitor RNA. Neither tRNA nor rRNA effectively competed for the interaction even at a concentration of 3201 excess, while cold CTE RNA readily competed the binding away at low concentrations. To further identify individual nuclear factors that bind to CTE RNA, UV-crosslinking assays were carried out with both MPMV CTE and SRV-1 CTE. These RNAs were generated by in vitro transcription with T7 polymerase from EcoRI-linearized plasmids p72 M240 and p72 S240 in the presence of [32P]UTP. Probes were incubated with nuclear extracts and the RNA–protein mixture was irradiated by short-wavelength UV light and digested by RNase before being run on an SDS–polyacrylamide gel. Only proteins that interact with the labeled RNA in the mixture will become detectable after radioautography. As shown in Fig. 2, both MPMV CTE and SRV-1 CTE crosslinked to a set of four bands with molecular weights ranging from 40 to 85 kDa (Fig. 2A). When increasing amounts

AID

VY 8407

/

6a29$$$102

01-22-97 06:15:20

of cold CTE RNA were included in the incubation mixture prior to adding the probes, all four bands faded away (Fig. 2B), confirming the detection of RNA–protein interaction. To determine whether these are nonspecific RNA binding proteins, we carried out the same assay with a CTE antisense RNA probe, which is of the same length and has the same GC composition as the CTE RNA. We found that two of the four bands crosslinked to this antisense RNA (Fig. 2C, lane 2). The other two proteins that did not bind to the antisense RNA have molecular weights of approximately 40 and 65 kDa, respectively (Fig. 2C, lane 1). These proteins apparently are specific for binding CTE. Mutagenesis study of MPMV CTE Antisense RNA provides a control for sequence specificity, but not for functional relevance. We reasoned that a mutant form of CTE with a limited number of base changes will be a more appropriate control for identifying CTE binding factors that are functionally important. MPMV CTE was originally mapped to start at 8022– 8039 and end at 8240–8140 (Bray et al., 1994); our mutagenesis summarized in Fig. 3 showed that deletion of 8005–8030 completely abrogated CTE function, while deletion at the 3* end from 8240–8190 had little effect on CTE’s ability to promote mRNA export, and trimming further down to 8140 abolished function as expected. Since CTE RNA is predicted to form an extensive secondary structure and there is evidence that these secondary structures are important for its function (Tabernero et al., 1996), we went on to make minimal internal mutations (one, two, or three substitutions in the 5* half) that can potentially disrupt the predicted structure. One mutant with a short stretch of three As changed to three Ts was found to lose 85–90% of its ability to promote CAT

vira

AP: Virology

336

TANG, XU, AND WONG-STAAL

FIG. 2. Two cellular factors bind specifically to functional CTE. (A) Both MPMV CTE and SRV-1 CTE RNA were labeled and used in crosslinking experiments with HeLa nuclear extract. They crosslinked to the same set of four bands. (B) The same crosslinking experiments carried out in the presence of excess amounts of cold CTE RNA. Left to right: 0, 20, 80, and 3201 excess of cold CTE RNA was included in the preincubation step (see Materials and Methods). (C) CTE antisense RNA crosslinked to two of the four bands. Arrows show the 65- and 40-kDa protein that did not crosslink to CTE antisense RNA. (D) CTE and DCTE have different affinity for the 65- and 40-kDa protein in crosslinking experiments. Arrows show that the two bands only crosslink to CTE but not to DCTE. (E) Cold DCTE RNA does not compete with CTE RNA for binding to 65- and 40-kDa protein. Increasing amounts (0, 20, 80, and 1601) of cold DCTE RNA were included in the preincubation step of the crosslinking experiments.

expression from the pDM 138 system (Donello et al., 1996). This region is well conserved in SRV-1, SRV-2, and MPMV CTEs; it is predicted to form a double-stranded stem with another stretch of the RNA (Table 1). The fact that different CTEs have complimentary changes flanking the three bases (at least one base pairing is kept by complimentary variation on each side) indicates that the base pairing in this region is important for function. The substitutions we made were also predicted by computer analysis to change the secondary structure of CTE dramatically (data not shown). This mutant, designated DCTE, was used in subsequent experiments as the nonfunctional form of MPMV CTE.

kDa proteins, further suggesting that they are probably general RNA binding proteins (Fig. 2D, lane 2). However, the 40- and 65-kDa bands have a much lower affinity for DCTE RNA than for wild-type CTE RNA (Fig. 2D, lane 1). In the competition assay with cold DCTE RNA, we observed that of the four bands that crosslink to CTE, the 50- and 85-kDa bands were easily competed away by DCTE, while the 40- and 65-kDa proteins were relatively resistant to competition with increasing amounts of cold DCTE RNA (Fig. 2E). Note that the relative intensity of the 50- and 40-kDa bands changed with increasing amounts of competitors. Thus, these two protein species appear to preferentially bind the functional wild-type CTE.

Two nuclear factors bind specifically to functional CTE

Partial purification of the 40- and 65-kDa proteins

DCTE was labeled and used in crosslinking assays. Like the antisense CTE, it crosslinked to the 50- and 85-

Protein purification was conducted to enrich CTE binding proteins. CTE binding activity was monitored at each

AID

vira

VY 8407

/

6a29$$$102

01-22-97 06:15:20

AP: Virology

CTE BINDING PROTEINS

337

FIG. 3. Mutagenesis of MPMV CTE. Schematic map of pDM138 and mutations made on the MPMV CTE. A representative CAT assay result is included to show the ability of various fragments of CTE to promote the expression of a CAT gene placed in an intron (Donello, et al., 1996).

step by UV crosslinking. First, a procedure previous used for purification of RRE binding proteins was used (Xu et al., 1996). This procedures entails mono-S, mono-Q, and butyl–Sepharose columns in series. Several fractions (7, 8, and 9) from the butyl–Sepharose column contained the 40-kDa CTE binding protein. Fraction 7 also contained the 50-kDa nonspecific RNA binding protein, but fractions 8 and 9 contained only the 40-kDa CTE binding protein (Fig. 4A). The 40-kDa protein clearly did not crosslink to DCTE or CTE antisense RNA and is distinct from the RRE binding protein, RREBP49, even though the two copurified during this procedure (Fig. 4B, LT-D1 is an RNA probe that represents the RNA of the first 90 nucleotide of the RRE sequence). Since the 65-kDa protein was not significantly enriched by this procedure, fractions containing this protein were pooled and subjected to further purification by biotinylated RNA selection (see Materials and Methods). A prominent 65-kDa band could be seen in CTE selected proteins but not DCTE selected proteins (Fig. 4C). A band of lower gel mobility is seen in the DCTE selected proteins, it could be a general RNA TABLE 1 Potential Base Pairing Conserved among SRV CTEs by Complementary Variation of Nucleotides

AID

MPMV

8055 CAUUUC 8060 8132 GUAAAG 8127

SRV-1

7675 CAUUUU 7680 7751 GUAAAA 7746

SRV-2

7616 GUUUC 7620 7689 CAAAG 7685

VY 8407

/

6a29$$$102

01-22-97 06:15:20

binding protein, the absence of this band in the CTE lane may be due to the effective competition the 65-kDa protein, which has a higher affinity for the CTE RNA. DISCUSSION Regulation of RNA export from the nucleus is an important aspect of posttranscriptional control of retrovirus gene expression. Since removal of introns from transcripts is a prerequisite for normal messenger RNA export out of the nucleus, how retroviruses achieve export of their unspliced transcripts (e.g., gag/pol mRNA) in a normal cell environment has always been a puzzle. Studies of the Rev/RRE system of HIV suggest that the virus has tapped into a transport machinery that is distinct from the usual cellular mRNA export pathway. Supposed sequestration of cellular factors by excess BSA bearing the activation domain of Rev microinjected into Xenopus oocyte nuclei inhibited the nuclear export of 5 S rRNA and U snRNA, but had little effect on the mRNA or tRNA export from the nuclei (Fischer et al., 1995). cis-acting RNA elements such as CTE have been shown to replace, in part, Rev/RRE functionally and probably share some aspects of the same pathway with Rev/RRE. They could either interact with a cellular counterpart of Rev or components along the Rev transactivation cascade. Rev has been proposed to function at multiple posttranscriptional levels. Rev can inhibit splicing in vitro (Kjems et al., 1993) and interacts with splicing components in vivo (Luo et al., 1994), directly promote RNA export (Fischer et al., 1994, 1995; Meyer et al., 1994), stabilize RNA (Felber et al., 1989; Malim et al., 1993), and facilitate polysome assembly (Arrigo et al., 1991). These multiple effects are

vira

AP: Virology

338

TANG, XU, AND WONG-STAAL

FIG. 4. Partial purification of the 40-kDa CTE binding protein. (A) Separation of the 50- and 40-kDa proteins. Fractions from the butyl–Sepharose column were tested for CTE binding activity using crosslinking assay. Lanes 1 and 2, column 7 contained both 50- and 40-kDa proteins; lanes 3 and 4, column 8 contained only the 40-kDa band of the four bands that crosslink to CTE RNA. (B) The 40-kDa protein does not crosslink to CTE antisense or RRE RNA. Interestingly, the 40-kDa protein copurified with the RRE binding protein (Xu et al., 1996). (C) Specific selection of the 65kDa protein by CTE RNA. Protein fractions containing the 65-kDa CTE binding protein were pooled and subjected to RNA selection. Protein was eluted from the beads by boiling in SDS–PAGE buffer and run on a 10% polyacrylamide gel; the proteins were visualized by silver staining.

not mutually exclusive; the full effect of Rev may be a summation of these individual activities. Increasing evidence shows that direct RNA export may be the most important aspect of all these (Fridell et al., 1996a,b). It is interesting to note that CTE cannot restore Rev activity to the wild-type level; Rev0 HIV mutants supplied with CTE are somewhat attenuated compared to wild-type virus or complementation by Rev (Bray et al., 1994; Zolotukhin et al., 1994). Similarly, insertion of CTE only restored 30–40% of the Rev activity in the pDM 138/ pDM128 system (data not shown). This partial effect of CTE could be due to the ability of CTE to complement only select aspects of Rev activities (e.g., export of unspliced RNA). Several cis-acting RNA elements similar to CTE have been identified from other viruses. The posttranscriptional regulatory element (PRE) from hepatitis B virus enhanced export of an intronless b-globulin transcript and promoted expression of a CAT gene placed in an intron (Huang et al., 1994, 1995; Donello et al., 1996). This element was shown to interact with several cellular factors in vitro (Huang et al., 1996). Another element within the herpes simplex virus thymidine kinase gene can also promote cytoplamic accumulation of unspliced b-globulin transcript; this element, named PPE for premRNA processing enhancer, was shown to bind heterogeneous nuclear ribonucleoprotein particle L (hnRNP L) (Liu et al., 1995). More recently, a direct repeat sequence in the 3* untranslated region of Rous sarcoma virus genome was found to facilitate Rev-independent expression of HIV Gag protein. RSV that lacks this element did not replicate well and seemed to have a significant reduction of unspliced viral RNA (Ogert et al., 1996). Like MPMV and SRV-1, the RSV genome does not encode Rev-like proteins, so it is likely that this direct repeat

element from RSV also directly interacts with cellular proteins to achieve function. Such proteins are not yet identified. Given the functional similarly of all these different elements, it will be interesting to find out whether they share common pathways or work by distinct mechanisms. The RSV element was found to be functional in chicken embryo fibroblast but not in COS cells, while MPMV CTE does not seem to function in CEF while active in COS and human cell lines (Ogert et al., 1996). This cell line restriction of one element or another indicates that their mechanisms do not entirely overlap. So far, no common protein has been identified to interact with CTE, PRE, and PPE. Different regulatory steps can contribute to the overall effect achieved by Rev, i.e., the cytoplasmic enrichment of unspliced transcripts. It is formally possible that these different elements emulate different aspects of Rev function, e.g., one type of element works directly on RNA export while another stabilizes the unspliced RNA either in the nucleus or cytoplasm or both. To resolve this question, molecular cloning and characterization of these RNA binding proteins will be of great interest.

AID

vira

VY 8407

/

6a29$$$102

01-22-97 06:15:20

ACKNOWLEDGMENT We thank Drs. J. Yeakley and X. D. Fu for their help in RNA selection and Dr. Tom Hope for plasmid pDM138. We also thank Drs. T. R. Reddy and X. Li for advice on making nuclear extract. Dr. J. Li is thanked for his help with in vitro transcription of radioactive RNA. This work is supported by NIH Grant R 01 to F.W-S. and by grant P30-AI3621403 from the CFAR of the University of California, San Diego.

REFERENCES Arrigo, S. J., and Chen, I. S. (1991). Rev is necessary for translation but not cytoplasmic accumulation of HIV-1 vif, vpr, and env/vpu 2 RNAs. Genes Dev. 5, 808–819.

AP: Virology

CTE BINDING PROTEINS

339

Bevec, D., Jaksche, H., Oft, M., Wohl, T., Himmelspach, M., Pacher, A., Schebesta, M., Koettnitz, K., Dobrovnik, M., Csonga, R. et al. (1996). Inhibition of HIV-1 replication in lymphocytes by mutants of the Rev cofactor eIF-5A. Science 271, 1858–1860. Bogerd, H. P., Fridell, R. A., Madore, S., and Cullen, B. R. (1995). Identification of a novel cellular cofactor for the Rev/Rex class of retroviral regulatory proteins. Cell 82, 485–494. Bray, M., Prasad, S., Dubay, J. W., Hunter, E., Jeang, K. T., Rekosh, D., and Hammarskjold, M. L. (1994). A small element from the MasonPfizer monkey virus genome makes human immunodeficiency virus type 1 expression and replication Rev-independent. Proc. Natl. Acad. Sci. USA 91, 1256–1260. Cullen, B. R. (1992). Mechanism of action of regulatory proteins encoded by complex retroviruses. Microbiol. Rev. 56, 375–394. Dignam, J. D., Lebovitz, R. M., and Roeder, R. G. (1983). Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei. Nucleic Acids Res. 11, 1475–1489. Donello, J. E., Beeche, A. A., Smith, G. J., Lucero, G. R. et al. (1996). The hepatitis B virus posttranscriptional regulatory element is composed of two subelements. J. Virol. 7, 4345–4351. Felber, B. K., Hadzopoulou-Cladaras, M., Cladaras, C., Copeland, T., and Pavlakis, G. N. (1989). Rev protein of human immunodeficiency virus type 1 affects the stability and transport of the viral mRNA. Proc. Natl. Acad. Sci. USA 86, 1495–1499. Fischer, U., Meyer, S., Teufel, M., Heckel, C., Luhrmann, R., and Rautmann, G. (1994). Evidence that HIV-1 Rev directly promotes the nuclear export of unspliced RNA. EMBO J. 13, 4105–4112. Fischer, U., Huber, J., Boelens, W. C., Mattaj, I. W., and Luhrmann, R. (1995). The HIV-1 Rev activation domain is a nuclear export signal that accesses an export pathway used by specific cellular RNAs. Cell 82, 475–483. Fridell, R. A., Fischer, U., Luhrmann, R., Meyer, B. E., Meinkoth, J. L., Malim, M. H., and Cullen, B. R. (1996). Amphibian transcription factor IIIA proteins contain a sequence element functionally equivalent to the nuclear export signal of human immunodeficiency virus type 1 Rev. Proc. Natl. Acad. Sci. USA 93, 2936–2940. Fridell, R. A., Bogerd, H. P., and Cullen, B. R. (1996). Nuclear export of late HIV-1 mRNAs occurs via a cellular protein export pathway. Proc. Natl. Acad. Sci. USA 93, 4421–4424. Fritz, C. C., Zapp, M. L., and Green, M. R. (1995). Human nucleoporinlike protein that specifically interacts with HIV Rev. Nature 376, 530– 533. Ho, S. N., Hunt, H. D., Horton, H. M., Pullen, J. K., and Pease, L. R. (1989). Site directed mutagenesis by overlap extension using the polymerase chain reaction. Gene 77, 51–59. Huang, Z. M., and Yen, T. S. (1994). Hepatitis B virus RNA element that facilitates accumulation of surface gene transcripts in the cytoplasm. J. Virol. 68, 3193–3199. Huang, Z. M., and Yen, T. S. (1995). Role of the hepatitis B virus posttran-

scriptional regulatory element in export of intronless transcripts. Mol. Cell. Biol. 15, 3864–3869. Huang, Z. M., Zang, W. Q., and Yen, T. S. (1996). Cellular proteins that bind to the hepatitis B virus posttranscriptional regulatory element. Virology 217, 573–581. Kjems, J., and Sharp, P. A. (1993). The basic domain of Rev from human immunodeficiency virus type 1 specifically blocks the entry of U4/ U6.U5 small nuclear ribonucleoprotein in spliceosome assembly. J. Virol. 67, 4769–4776. Liu, X., and Mertz, J. E. (1995). HnRNP L binds a cis-acting RNA sequence element that enables intron-dependent gene expression. Genes Dev. 9, 1766–1780. Luo, Y., Yu, H., and Peterlin, B. M. (1994). Cellular protein modulates effects of human immunodeficiency virus type 1 Rev. J. Virol. 68, 3850–3856. Malim, M. H., and Cullen, B. R. (1993). Rev and the fate of pre-mRNA in the nucleus: implications for the regulation of RNA processing in eukaryotes. Mol. Cell. Biol. 13, 6180–6189. Meyer, B. E., and Malim, M. H. (1994). The HIV-1 Rev trans-activator shuttles between the nucleus and the cytoplasm. Genes Dev. 8, 1538–1547. Ogert, R. A., Lee, L. H., and Beemon, K. L. (1996). Avian retroviral RNA element promotes unspliced RNA accumulation in the cytoplasm. J. Virol. 70, 3834–3843. Ruhl, M., Himmelspach, M., Bahr, G. M., Hammerschmid, F., Jaksche, H., Wolff, B., Aschauer, H., Farrington, G. K., Probst, H., Bevec, D. et al. (1993). Eukaryotic initiation factor 5A is a cellular target of the human immunodeficiency virus type 1 Rev activation domain mediating trans-activation. J. Cell Biol. 123, 1309–1320. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989). ‘‘Molecular Cloning: A Laboratory Manual,’’ 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Stutz, F., Neville, M., and Rosbach, M. (1995). Identification of a novel nuclear pore-associated protein as functional target of the HIV-1 Rev protein in yeast. Cell 82, 495–506. Tabernero, C., Zolotukhin, A. S., Valentin, A., Pavlakis, G. N., and Felber, B. K. (1996). The posttranscriptional control element of Simian Retrovirus type 1 forms an extensive RNA secondary structure necessary for its function. J. Virol. 70, 5998–6011. Vaishnav, Y. N., and Wong-Staal, F. (1991). The biochemistry of AIDS. Annu. Rev. Biochem. 60, 577–630. Xu, Y., Reddy, T. R., Fischer, W. H., and Wong-Staal, F. (1996). A novel HnRNP specifically interacts with HIV-1 RRE RNA. J. Biomed. Sci. 3, 82–91. Yeakley, J. M., Morfin, J. P., Rosenfeld, M. G., and Fu, X. D. (1996). A complex of nuclear proteins mediates SR protein binding to a purinerich splicing enhancer. Proc. Natl. Acad. Sci. USA 15, 7582–7587. Zolotukhin, A. S., Valentin, A., Pavlakis, G. N., and Felber, B. K. (1994). Continuous propagation of RRE(0) and Rev(0)RRE(0) human immunodeficiency virus type 1 molecular clones containing a cis-acting element of simian retrovirus type 1 in human peripheral blood lymphocytes. J. Virol. 68, 7944–7952.

AID

vira

VY 8407

/

6a29$$$103

01-22-97 06:15:20

AP: Virology