Disruption of the putative splice acceptor site for SIVmac239 Vif reveals tight control of SIV splicing and impaired replication in Vif non-permissive cells

Virology 338 (2005) 281 – 291 www.elsevier.com/locate/yviro

Disruption of the putative splice acceptor site for SIVmac239 Vif reveals tight control of SIV splicing and impaired replication in Vif non-permissive cells Joseph G. Victoriaa, W. Edward Robinson Jr.a,b,c,* a

Department of Microbiology and Molecular Genetics, University of California, Irvine, CA 92697-4800, USA b Department of Pathology, University of California, Irvine, CA 92697-4800, USA c Department of Medicine, University of California, Irvine, CA 92697-4800, USA Received 23 March 2005; returned to author for revision 25 April 2005; accepted 6 May 2005 Available online 13 June 2005

Abstract Vif is dispensable for simian immunodeficiency virus (SIV) replication in some cells, termed permissive (i.e., CEM-SS), but not in others, termed non-permissive (i.e., H9, CEMx174, and peripheral blood lymphocytes). Non-permissive cells express the RNA editing enzyme, APOBEC3G. To determine whether vif mRNA could be alternatively spliced, a mutation altering the putative vif splice acceptor site (SA1) was introduced into SIVmac239 (SIVDvif-SA). Despite three consensus splice acceptor sites nearby SA1, SIVDvif-SA did not efficiently generate alternatively spliced vif mRNA. SIVDvif-SA was growth attenuated in CEMx174 and H9 cells but not in CEM-SS cells. Following SIVDvif-SA, but not SIVmac239, infection in either H9 or CEMx174 cells viral cDNA contained numerous G to A mutations; no such differences were observed in CEM-SS cells. This pattern is consistent with mutations generated by APOBEC3G in the absence of Vif. Therefore, efficient splicing of SIV vif mRNA is tightly controlled and requires the SA1 site. D 2005 Elsevier Inc. All rights reserved. Keywords: mRNA; DNA hypermutation; APOBEC3G; Real-time polymerase chain reaction; Cytosine deamination; AIDS

Introduction Lentiviruses, excluding equine infectious anemia virus, contain the accessory gene vif. Replication of human immunodeficiency virus (HIV) in the absence of functional Vif is unaffected in some cell types, termed permissive (such as SupT1 and CEM-SS), while required in others, termed non-permissive (such as peripheral blood mononuclear cells, macrophages, and H9 cells). Recent data indicate that Vif counters the antiviral effects of the cytosine deaminase APOBEC3G (Mangeat et al., 2003; Mariani et al., 2003; Sheehy et al., 2002), and possibly the closely related APOBEC3F (Wiegand et al., 2004; Zheng et al., 2004). APOBEC3G deaminates cytosine residues on * Corresponding author. Department of Pathology, University of California, Irvine, CA 92697-4800, USA. Fax: +1 949 824 2505. E-mail address: [email protected] (W.E. Robinson). 0042-6822/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.virol.2005.05.007

the nascent viral genome during reverse transcription, resulting in non-functional proviruses containing G to A mutations (Mangeat et al., 2003; Suspene et al., 2004). Additionally, viral replication may be inhibited directly through degradation of mutated cDNAs via the action of uracil DNA glycosylase (Harris et al., 2003; Mariani et al., 2003). Recent data also indicate that APOBEC3G may have additional antiviral effects separate from cytosine deamination. Several investigators have shown that APOBEC3G containing active site mutations retains antiviral activity when overexpressed (Newman et al., 2005; Shindo et al., 2003; Turelli et al., 2004). Vif counters APOBEC3G antiviral activity by targeting it for degradation (Conticello et al., 2003; Marin et al., 2003; Mehle et al., 2004; Sheehy et al., 2003; Stopak et al., 2003) as part of the Vif-Cul5-Scf complex (Yu et al., 2003), thereby preventing APOBEC3G incorporation into the virion (Yu et al., 2003).

282

J.G. Victoria, W.E. Robinson Jr. / Virology 338 (2005) 281 – 291

APOBEC3G is broadly active against multiple retroviruses such as murine leukemia virus and simian immunodeficiency virus (SIV) (Harris et al., 2003; Mangeat et al., 2003). Overexpression of APOBEC3G orthologues results in species-specific differences during HIV or SIV infection (Mangeat et al., 2003; Mariani et al., 2003; Yu et al., 2003). For example, both mouse and African green monkey (AGM) APOBEC3G disrupt HIV viral replication regardless of Vif expression (Mariani et al., 2003). Sensitivity to HIV Vif can be restored by a single amino acid substitution of aspartic acid for lysine at residue 128 of AGM APOBEC3G (Bogerd et al., 2004; Mangeat et al., 2004; Schrofelbauer et al., 2004). Vif from SIVagm is active against monkey APOBEC3G but not human APOBEC3G (Mariani et al., 2003; Takeuchi et al., 2005; Yu et al., 2004a). SIVmac239 Vif is more broadly reactive and is capable of targeting human, rhesus, and AGM APOBEC3G for degradation (Gaddis et al., 2004; Mariani et al., 2003). While a great deal of study has focused on the mechanism of APOBEC3G/Vif interactions and antiviral effects, regulation of Vif transcription and expression remains understudied. Productive SIV infection requires both a temporal regulation and balance between at least 20 different splicing events to generate mRNAs encoding accessory proteins, viral polyproteins, and genomic RNA (Amendt et al., 1994; Reinhart et al., 1996; Viglianti et al., 1990). Multiple-spliced 2.0-kb class mRNA messages encoding Rev, Tat, and Nef are first transcribed and expressed. Rev and Tat facilitate transcription and nuclear export of single-spliced 4.0-kb class mRNA encoding accessory proteins and envelope glycoproteins and the 9.2-kb mRNA encoding Gag and Pol polyproteins or unspliced genomic RNA (Furtado et al., 1991; Reinhart et al., 1996; Schwartz et al., 1991; Unger et al., 1991; Viglianti et al., 1990). Adding to this complexity, significant variability in the SIV consensus splice acceptor sequence, (U/C)N(C/U)AG/(G/A) (Amendt et al., 1994; Park et al., 1991; Reinhart et al., 1996; Unger et al., 1991; Viglianti et al., 1990), allows for potential alternative splicing of viral mRNA (Fig. 1). Direct sequencing of mRNA reveals more than three alternative splice sites for rev, tat, and nef within 100 base pairs of the authentic splice sites in both infectious molecular clones (Park et al., 1991; Unger et al., 1991) and uncloned isolates from sooty mangabey monkeys (SIVsmm) (Reinhart et al., 1996) (Fig. 1). In addition to multiple downstream splice acceptor sites, alternative splicing of rev, tat, and nef can occur from one of two splice donor sites within the viral LTR, SD1, and SD1V. Mechanisms of alternative splice site choice for SIV have not been characterized; however, HIV alternative splicing is tightly regulated by cis-acting purine-rich exonic splicing enhancers (ESEs) (Caputi and Zahler, 2002; Reinhart et al., 1996) and pyrimidine-rich exonic splicing silencers (ESSs) (Bilodeau et al., 2001; Domsic et al., 2003).

Fig. 1. Genomic organization and splicing of SIV RNA. (A) SIV encodes gag, pol, and envelope proteins plus additional regulatory proteins, Tat and Rev, and accessory proteins Vif, Vpr, Vpx, and Nef. The 5Vand 3VLTRs are critical for viral replication. Splicing of rev, nef, tat, and vif mRNAs occurs through splice donor site 1 (SD1) or an alternate splice donor site (SD1V) in the 5V LTR to several downstream splice acceptor sites (SA1 to SA4). Superscript letters denote alternative splice acceptor sites. Rev, Nef, and Tat can be translated at least three different, multiply spliced mRNAs. Inclusion of a small, non-coding exon (EX1) (light grey box) is present in some alternatively spliced forms of rev, tat, nef, and env. (B) Strong consensus sequences within 250 base pairs of the putative vif splice acceptor sequence, SA1, are aligned and denoted as SA1a, SA1b, and SA1c. Underlined sequences match the SIV consensus splice acceptor sequence (Reinhart et al., 1996; Viglianti et al., 1990) illustrated at the bottom. Asterisks indicate engineered mutations to generate SA1-deficient virus (SIVDvif-SA).

Published reports suggest vif mRNA is the product of a single splicing event (Garrett et al., 1991; Park et al., 1991; Schwartz et al., 1991; Viglianti et al., 1990). The putative SIV vif splice acceptor site (SA1), identified by northern hybridization, genomic position, and sequence, is located in the carboxyl terminus of integrase (IN) (Fig. 1). No alternatively spliced forms of vif have been reported; however, levels of spliced vif mRNAs are low throughout the course of infection, thus infrequent use of alternative splice sites may not be easily detected (Schwartz et al., 1991; Viglianti et al., 1990). While much study has focused on the role of HIV Vif and its interaction with APOBEC3G these interactions are understudied in SIV replication, arguably the best animal model for HIV. In this study, we mutated the putative vifsplice acceptor site, SA1, in SIVmac239 and measured viral

J.G. Victoria, W.E. Robinson Jr. / Virology 338 (2005) 281 – 291

283

replication, vif RNA splicing, and mutational frequency in SIV from both Vif permissive and non-permissive cell lines. These results demonstrate that vif splicing must occur at SA1 for Vif to be expressed at adequate levels to counteract the antilentiviral properties present in vif non-permissive cells, indicating that splicing of vif mRNA is tightly controlled.

Results Mutation of the SIV SA1 splice acceptor site leads to a growth defect in Vif non-permissive cells Although no reports have conclusively demonstrated that messages spliced at SIV SA1 encode for functional Vif protein, both its genomic position and Northern blot analysis have led to the conclusion that it is the authentic vif splice acceptor site (Park et al., 1991; Viglianti et al., 1990). However, three strong consensus sequences are also located within 200 base pairs of SA1 (Fig. 1B). Using site-directed mutagenesis, mutations in the consensus SIV SA1 sequence, UACAGA to UAUCGC (nucleotides 5463– 5468), were introduced to generate SIVDvif-SA (Fig. 1B). The mutations were silent within the overlapping IN coding sequence. SIVDvif-SA failed to replicate in H9 cells (Fig. 2A), known to express APOBEC3G which confers a Vif non-permissive phenotype for both HIVDvif and SIVDvif (Mangeat et al., 2003; Sheehy et al., 2003; Zhang et al., 2003). Replication of SIVDvif-SA in CEMx174 cells was severely attenuated (Fig. 2B). Vif-deficient SIV have been reported to be either attenuated (Park et al., 1994; Zou and Luciw, 1996) or replication incompetent (Desrosiers et al., 1998) in CEMx174 cells. Levels of APOBEC3G in CEMx174 cells are comparable to those in H9 cells (data not shown and Takeuchi et al., 2005). SIVDvif-SA replication in cells lacking APOBEC3G, CEM-SS, is superimposable on that of the reference strain (Fig. 2C). Taken together, these data indicate that despite multiple strong consensus sequences nearby SA1, SIVDvif-SA replicates with Vif-deficient kinetics. Splicing of SIV vif mRNA To ascertain whether alternative splice acceptor sites for vif could be used in the absence of the consensus SA1 site, SIV cDNA was generated by reverse transcriptase polymerase chain reaction (RT-PCR) from cellular RNA. Total cellular RNA was extracted from CEMx174 cells 4 days after inoculation with either SIVDvif-SA or SIVmac239. Viral cDNA was generated using a primer which binds downstream of the vif start codon. Spliced RNAs were enriched by amplifying across the splice junction (Fig. 3A). Southern hybridization of the cDNA revealed a product of the predicted size for SD1 to SA1 splicing during SIVmac239 infection that was absent in CEMx174 cells infected with SIVDvif-SA (Fig. 3B). Additionally, significant hybridization was observed to a larger, potentially alternatively spliced

Fig. 2. Replication of SIVDvif-SA in vif non-permissive and permissive cells. Equal amounts, as measured by RT assay, of SIVmac239 (circles) or SIVDvif-SA (triangles) were inoculated upon either (A) semi-permissive CEMx174, (B) non-permissive H9 cells, or (C) permissive CEM-SS cells at a multiplicity of infection of less than one. Closed and open symbols represent duplicate infections. Infections were monitored by IFA using a modification of the methods for HIV, described previously (King and Robinson, 1998; Robinson et al., 1989).

284

J.G. Victoria, W.E. Robinson Jr. / Virology 338 (2005) 281 – 291

Fig. 3. Splicing of vif mRNA. (A) Primer S-Vif-REV was used to generate cDNA from CEMx174 cells infected with either SIVmac239 or SIVDvif-SA. cDNA was enriched for either authentic spliced message or alternatively spliced products by PCR using the S-M667/S-IN-Rev or S-M667/S-SA1V-Rev primer sets, respectively. Radiolabeled S-AA55 was used to probe PCR products spliced at either the SD1 or SD1V5Vsplice sites. (B) RT-PCR products from S-M667/S-INREV amplification were transferred to zeta probe (BioRad) membrane and probed with 32P-labeled S-AA55. The asterisk indicates the expected size of products spliced from SD1 to SA1, while the open arrow indicates a potential, alternatively spliced product. (C) RT-PCR products, performed in duplicate, from RNA extracted at either 24, 48, or 72 h post-inoculation were amplified using the S-M667/S-SA1V-Rev primer set and probed with labeled S-AA55.

product in both SIVmac239- and SIVDvif-SA-infected cells. This product was not present when blots were analyzed using a probe that bound between SD1 and SD1V(data not shown) indicating a preferential usage of SD1V in the putative alternatively spliced product. To distinguish alternative versus cryptic splicing, cDNA products were enriched for splicing events that occurred upstream of SA1 using the S-SA1V-Rev primer (Fig. 3A, dotted line). These reactions, performed in duplicate, indicate an unmasking of a cryptic splice site when SA1 is mutated, rather than alternative splicing, as no product is observed in the SIVmac239 samples (Fig. 3C).

Sequence of SIVDvif-SA-spliced products To determine if this cryptic splicing occurred at one of the three nearby consensus sequences (Fig. 1B), cDNA products of SIVmac239 and SIVDvif-SA were cloned into the pCR-Script cloning vector and sequenced. Sequences from SIVmac239-infected cells demonstrated correct splicing from the 5Vsplice donor site, SD1, within the viral 5V-LTR to the SA1 sequence (Fig. 4B). All of the cDNA sequences analyzed from SIVDvif-SA confirmed the presence of the engineered mutations; surprisingly, none of the clones were spliced at any of the three consensus splice acceptor

Fig. 4. Inefficient splicing from SD1Vto a cryptic splice acceptor site. Products from RT-PCR were cloned into pCR-SCRIPT (Stratagene) cloning vectors and sequenced with both T7 and T3 primers using automated sequencing. (A) Expected splicing pattern for vif mRNA. (B) cDNA from SIVmac239 spliced at the SD1-SA1 junction was present in 4/4 clones tested. A single clone of SIVDvif-SA was identified that spliced from the alternative 5VSD1Vsite and a near consensus splice acceptor sequence (5V-AGAAG-3V) 308 bp upstream of SA1. In the remaining four clones SIVDvif-SA cDNA reads through the IN sequence, retaining the engineered SA1 splice site mutations. The nucleotide sequence of each group of clones is illustrated.

J.G. Victoria, W.E. Robinson Jr. / Virology 338 (2005) 281 – 291

285

sequences near SA1 (Fig. 4B). A single clone, of the five recovered, contained a spliced product approximately 750 bp in length that utilized a loose consensus splice acceptor sequence, GAAG/TAA located at position 5160, over 300 base pairs upstream of SA1. Consistent with the hybridization of spliced cDNA (Fig. 3A and data not shown), the SIVDvif-SA message utilized the alternative 5V splice donor site, SD1V. Several clones from both SIVmac239 and SIVDvif-SA contained non-viral sequences which had a high degree of homology with human chromosome 11 (data not shown), which may account for some of the non-specific bands seen in Fig. 3B. Decreased cDNA synthesis is observed during initial infection of CEMx174 cells with SIVDvif-SA Considering SIVDvif-SA growth kinetics (Fig. 2) mirrored that of Vif-deleted virus (Desrosiers et al., 1998; Gibbs et al., 1994; Park et al., 1994; Zou and Luciw, 1996), it is unlikely that the cryptically spliced vif mRNA is capable of encoding sufficient levels of Vif to counteract the antiviral effects of APOBEC3G. We therefore employed a more sensitive technique, quantitative real-time PCR, to measure growth kinetics during the first round of infection. Previously, we demonstrated that SYBR green-I quantitative real-time PCR using HIV-specific primers is a sensitive and reproducible method to distinguish defects associated with impaired HIV entry, reverse transcription, and proteolytic processing (Victoria et al., 2003). Using SIV-specific primer pairs, homologous to the previously characterized HIV primers, the replication of SIVDvif-SA in CEMx174 cells was studied. A minimal effect on viral entry was observed, as seen by similar levels of SIV minus strand strong stop DNA (-sssDNA) between 1 and 24 h postinoculation (Fig. 5A). However, a greater than 90% reduction in completely reverse transcribed cDNA was observed throughout infection (Fig. 5B). These results mirror SIV infection in the presence of the RT inhibitor PMPA (data not shown). Since neither a non-coding nucleotide change in IN nor mutation of SA1 would be expected to cause a reverse transcription defect, these data are consistent with the degradation of provirus mutated by APOBEC3G reported for infections with HIV lacking Vif in non-permissive cells (Goncalves et al., 1996; Mariani et al., 2003; Sova and Volsky, 1993; von Schwedler et al., 1993). Integration itself was not adversely affected, as no defect was observed using the recombinant IN from SIVDvif-SA in in vitro assays or in vivo by measuring the accumulation of two-LTR circle DNA, a by-product of failed integration (data not shown). Mutational frequency of SIVDvif-SA DNA was extracted 48 h post-inoculation from CEM-SS, CEMx174, or H9 cells infected with either SIVmac239 or SIVDvif-SA. The IN genes from each virus were amplified and

Fig. 5. Kinetics of reverse transcription for SIVDvif-SA in semi-permissive CEMx174 cells. Quantitative real-time PCR was performed on CEMx174 cellular lysates between 1 and 72 h post-inoculation with either SIVmac239 (closed circles) or SIVDvif-SA (open triangles). At each time point 1.0  106 cells were lysed. Lysates from the equivalent of 2.0  105 cells were amplified with primers design to detect either (A) -sssDNA (S-AA55/ S-M667) or (B) complete products of reverse transcription (S-M661/ S-M667). Infected cell equivalents were calculated from standard curves generated by serial dilution of SIVmac251 chronically infected H9 cell lysates into lysates from uninfected H9 cells. PCR reactions are representative of duplicate experiments.

cloned. Multiple clones from triplicate infections were subjected to sequence analysis. Overall mutational frequency in IN genes was statistically higher during SIVDvif-SA infection than in SIVmac239 in both H9 and CEMx174 cells. This increased frequency was driven by G to A mutation, indicative of APOBEC3G-mediated cytosine deamination. Indeed, during SIVDvif-SA infections 1.5% of all guanine residues sequenced were mutated to adenine in H9 (19 of 1233) and CEMx174 (26 of 1781). Conversely, infections with SIVmac239 displayed only 0.12% mutational frequency of guanines in CEMx174 cells (2 of 1707) and no such mutations in H9 cells. The G to A mutational frequency of SIVDvif-SA in permissive CEM-SS cells was not statistically different from SIVmac239 (Fig. 6A). Although the increase in G to A mutations in Vif non-permissive cells is substantial,

286

J.G. Victoria, W.E. Robinson Jr. / Virology 338 (2005) 281 – 291

frequencies in H9 and CEMx174 cells in the presence of Vif, suggest that A to G mutations were not Vif regulated. However, in H9 cells A to G mutations were increased in the absence of Vif. A to G mutations have been observed previously (Kobayashi et al., 2004; Mangeat et al., 2003; Wiegand et al., 2004) as the second most common mutation during HIV Vif-deficient viral replication in non-permissive cells and may be the result of another cellular editing enzyme.

Discussion

Fig. 6. G to A and A to G mutational frequencies of SIVmac239 and SIVDvif-SA. Whole cell lysates were obtained from H9, CEMx174, or CEM-SS cells infected with either SIVmac239 or SIVDvif-SA 120 h post-inoculation. Following PCR amplification of the IN gene, automated sequencing was performed. In H9 cells, six SIVmac239 clones (3386 bp) and ten SIVDvif-SA clones (7835 bp) were sequenced. In CEMx174 cells, fourteen clones of both SIVmac239 (9858 bp) and SIVDvif-SA (9717 bp) were sequenced. In CEM-SS cells, five SIVmac239 clones (4034 bp) and four SIVDvif-SA clones (3096 bp) were sequenced. Panel A: G to A mutational frequency; or panel B: A to G mutational frequency for each cell type inoculated with either SIVmac239 (open bars) or SIVDvif-SA (shaded bars). Asterisks indicate statistical significance ( P < 0.05) by one-tailed two-sample t test assuming unequal variance.

the absolute frequency is lower than that generally reported for vif-deleted HIV in non-permissive cells (1.1 – 14%) (Mangeat et al., 2003; Suspene et al., 2004; Yu et al., 2004b, 2003; Zhang et al., 2003). The relatively low frequencies observed here could be due to inherent differences between HIV and SIV susceptibility to human APOBEC3G. However, we cannot rule out low level expression of Vif from cryptically spliced vif mRNA during SIVDvif-SA infection. By far the second most common mutation observed was an A to G. Strikingly, A to G mutations (Fig. 6B) exceed G to A (Fig. 6A) mutations in both H9 and CEMx174 cells when Vif was present. These data, combined with the relatively similar A to G mutational

While many studies have focused on APOBEC3G interactions with Vif (Conticello et al., 2003; Marin et al., 2003; Mehle et al., 2004; Sheehy et al., 2003; Stopak et al., 2003; Yu et al., 2003), expression of APOBEC3G (Rose et al., 2004; Sheehy et al., 2002), and species-specific antiviral activities of APOBEC3G (Bogerd et al., 2004; Kobayashi et al., 2004; Lake et al., 2003; Mangeat et al., 2004; Mariani et al., 2003; Schrofelbauer et al., 2004; Xu et al., 2004), this is the first study to report a tightly controlled transcription of vif mRNA through splicing at SA1. Despite significant selective pressure on SIVDvif-SA, efficient splicing of vif mRNA at one of three strong consensus splice acceptor sequences nearby SA1 was not observed. In fact, only a low level of splicing, through a cryptic splice site that preferentially utilized the unconventional upstream SD1V site, was observed. This finding is in sharp contrast to mutational analyses of HIV splice acceptor sites 2 –4, in which many cryptic and alternative splice sites were identified (Purcell and Martin, 1993). In addition to consensus splice donor and splice acceptor sequences, splicing is governed by several cis-acting elements: intron branch point sequences, ESSs, ESEs, and intronic polypyrimidine tracts (reviewed in Garcia-Blanco, 2003; Nilsen, 2003). Because degenerate branch point sequences are located throughout the SIV genome, it is unlikely that alternative splicing at SA1a, SA1b, or SA1c is prohibited via this mechanism. ESEs control the alternative splicing of rev, tat, and nef through recruitment of ASF/ SF2, Srp40, SC35, and other serine/arginine-rich (SR) proteins in HIV (Bilodeau et al., 2001; Caputi and Zahler, 2002; Domsic et al., 2003; Reinhart et al., 1996; Ropers et al., 2004). At least five ESSs have been identified for HIV (ESSI-ESSV) as well as purine-rich ESEs for both HIV and SIVsmm (Caputi and Zahler, 2002; Reinhart et al., 1996). No putative strong SR protein binding sites were identified near SA1a, SA1b, or SA1c using the ESEfinder Web interface (Cartegni et al., 2002). However, ESEfinder also failed to identify any significant ESE binding sites for SIV nearby SA1, despite evidence of a weak ESE that binds ASF/SF2 to enhance splicing at SA1 for HIV (Ropers et al., 2004). Although several ESS sequences have been identified for HIV (Bilodeau et al., 2001; Caputi and Zahler, 2002; Domsic et al., 2003), no studies have been performed to

J.G. Victoria, W.E. Robinson Jr. / Virology 338 (2005) 281 – 291

analyze the role of ESS in SIV. High levels of spliced tat and rev during infection may be in part due to extensive 12 – 20 nucleotide polypyrimidine tracts immediately upstream of the SA2, SA3, and SA4 3V splice sites. No such polypyrimidine tracts are located upstream of either the SA1a, SA1b, or SA1c sites; however, the upstream sequence of SA1 contains only a single short, 10-base pair interrupted polypyrimidine tract. Data presented here suggest lack of a strong polypyrimidine tract and a potential unidentified ESS may be acting to repress splicing of vif mRNA at alternative splice sites nearby SA1. Due to this tightly controlled expression of vif mRNA, replication of SIVDvif-SA is consistent with the phenotype of viruses containing deleted or otherwise non-functional vif. In CEMx174 cells, SIV lacking vif has been reported to be completely replication incompetent (Desrosiers et al., 1998), severely attenuated (Park et al., 1994), or mildly attenuated (Zou and Luciw, 1996). SIV containing point mutations in the start codon or premature stop codons of vif display a growth delay of 35 days and reach only 33% of the maximal RT activity of SIVmac239 in CEMx174 cells (Park et al., 1994). Similarly, we demonstrate that CEMx174 cells, at least those that we have used, are best characterized as semipermissive. Growth of SIVDvif-SA in CEMx174 is delayed by greater than 14 days and only a maximum of 40% of the cells were positive for SIV antigens by indirect immunofluorescence assay (IFA) (Fig. 2A). We observed no viral replication in the truly non-permissive cell line, H9 (Fig. 2B). Conversely, SIVDvif-SA replication in permissive CEMSS cells mirrored that of SIVmac239 (Fig. 2C). This latter finding is important because it suggests that only vif is being affected by the mutation. SIVmac239 and SIVsmm produce a measurable fraction of mRNAs that utilize SA1 to generate a small non-coding exon, EX1, in alternatively spliced tat, rev, and nef (Fig. 1) (Reinhart et al., 1996; Unger et al., 1991; Viglianti et al., 1990). Our data suggest that inclusion of EX1 is not necessary for SIV replication as SIVDvif-SA replication in CEM-SS cells was unaffected. Previous studies on exons 2 and 3 of HIV demonstrated that inclusion of exon 3 can reduce protein expression (Krummheuer et al., 2001). For SIV, inclusion of exon 1 would appear to have little positive or negative effect. Defects associated with vif deletion in HIV include abnormal viral core assembly (Borman et al., 1995; Gabuzda et al., 1992; Mariani et al., 2003; Sakai et al., 1993; von Schwedler et al., 1993), decreased proviral cDNA synthesis (Borman et al., 1995; Gabuzda et al., 1992; Mariani et al., 2003; Sakai et al., 1993; von Schwedler et al., 1993), misincorporation of envelope with abnormal posttranslational modification of gp41 (Gaddis et al., 2003), G to A hypermutability in the viral genome (Harris et al., 2003; Kobayashi et al., 2004; Mangeat et al., 2003; Mariani et al., 2003; Suspene et al., 2004; Wiegand et al., 2004; Yu et al., 2004b), and a failure to exclude APOBEC3G from virions (Kao et al., 2003; Luo et al., 2004; Mariani et al., 2003; Schafer et al., 2004; Zennou et

287

al., 2004). The myriad of functions ascribed to Vif are often contradictory and some may be artifacts of in vitro studies. Decreases in reverse transcribed viral cDNA may be one of the most controversial effects associated with vif disruption. While most reports agree that the initiation of reverse transcription is unaffected in the absence of Vif for both SIV and HIV (Borman et al., 1995; Gaddis et al., 2003; Goncalves et al., 1996; Sova and Volsky, 1993; von Schwedler et al., 1993), effects of Vif deletion on late products of reverse transcription are less clear. Vif disruption in non-permissive cells has been shown to cause little or no defect in cDNA synthesis in some studies (Gaddis et al., 2003; Zou and Luciw, 1996) or greater than 25% reduction in cDNA synthesis during the initial round of infection in others (Borman et al., 1995; Sova and Volsky, 1993; von Schwedler et al., 1993). Using the most sensitive method developed to date, quantitative real-time PCR, we demonstrate a 90% reduction in late products of reverse transcription (Fig. 5B), with little to no affect on the initiation of reverse transcription (Fig. 5A). One possible explanation for the defects in reverse transcription observed for SIVDvif-SA are alterations in RNA secondary structure resulting from the engineered mutations. However, this seems unlikely as replication in both Vif permissive and non-permissive cells should have been affected equally. It is clear from Fig. 2 that replication was impaired only in Vif non-permissive cells. A more likely explanation is that APOBEC3G-induced mutations target the proviral cDNA, either directly or indirectly, for degradation. Thus, our real-time PCR data support degradation of SIV cDNA by uracil DNA glycosylase as a mechanism for the antiviral activities of APOBEC3G (Harris et al., 2003; Mariani et al., 2003). Similar to viral replication of vif-deleted HIV (Kobayashi et al., 2004; Mangeat et al., 2003; Mariani et al., 2003; Suspene et al., 2004; Yu et al., 2004b; Zhang et al., 2003), SIVagm (Mangeat et al., 2003; Mariani et al., 2003; Schrofelbauer et al., 2004), and SIVmac (Mangeat et al., 2003; Mariani et al., 2003), we observed hypermutability of the nascent -sssDNA, specifically G to A transitions. Reports of the mutation rate for vif-deficient HIV during APOBEC3G overexpression are highly varied, ranging from 1.1% to 14% (Mangeat et al., 2003; Mariani et al., 2003). Even endogenous levels of APOBEC3G induce up to 2.5% G to A mutations in HIV lacking Vif (Zhang et al., 2003). In concordance with previously reported studies, we observed a significantly higher mutational frequency for SIVDvif-SA relative to SIVmac239 in CEMx174 and H9 cells, but not in CEM-SS cells. The percentage of G to A mutations observed, ¨1.5%, is at the low end of those reported for HIV. Replication of some SIV strains appears to be less sensitive to inhibition by APOBEC3G from human cell lines (Gaddis et al., 2003), which may explain the lower mutational frequency we observed. Alternatively, low level of Vif expression from the cryptically spliced mRNA (Fig. 3) may partially abrogate APOBEC3G-mediated G to A

288

J.G. Victoria, W.E. Robinson Jr. / Virology 338 (2005) 281 – 291

hypermutation. Unfortunately, no SIV-specific antibodies are available to determine if Vif was expressed at low levels. Consistent with the hypothesis that low levels of Vif expression occurred from the cryptically spliced product was the nature of the G to A mutations. Some clones had many G to A mutations while others had one or none (data not shown). Nevertheless, as shown in Fig. 2, the growth defects of SIVDvif-SA were severe in Vif non-permissive cells, indicating that even low levels of Vif expression were insufficient to support SIV replication. Our laboratory has found that identical mutation of the HIV SA1 site results in more severely impaired HIV replication with a higher rate of G to A mutations than that observed for SIV. Moreover, no evidence for alternatively spliced HIV vif mRNA or for Vif protein synthesis was observed (B. Herrera and W.E. Robinson, Jr., unpublished results). Thus, low level of expression of SIV Vif from the cryptically spliced product seems a viable interpretation of our findings that G to A mutations are lower in SIVDvif-SA than observed for Vif deleted HIV. Another important finding in these studies concerns the species specificity of Vif. SIVagm Vif fails to down regulate human APOBEC3G (Mariani et al., 2003; Takeuchi et al., 2005; Yu et al., 2004a). Indeed, the block of SIVagm in some human cell lines has been attributed to this failure. Clearly, SIVmac239 replicates in Vif non-permissive H9 and CEMx174 cells, indicating SIVmac239 Vif interacts with human APOBEC3G. Moreover, the G to A mutational frequency of SIVmac239 was quite low in H9 and CEMx174 cells when Vif was expressed (Fig. 6). Thus, while some SIV Vif proteins may differ sufficiently in sequence and cannot interact with human APOBEC3G, not all do. Indeed, work from several laboratories clearly indicates HIV Vif, SIVmac239 Vif, and SIVAGM Vif show differential abilities to interact with human, AGM, and rhesus APOBEC3G (Bogerd et al., 2004; Mariani et al., 2003; Yu et al., 2004a). These studies on SIVDvif-SA identify another interesting phenotype that may be Vif dependent. Previous reports on HIV replication have identified that the second most common mutation observed in the absence of Vif is A to G (Kobayashi et al., 2004; Mangeat et al., 2003; Wiegand et al., 2004). As indicated in Fig. 6B, A to G mutations were

common in our study as well. It is possible that A to G conversions result from base pair misincorporation biases inherent in RT or PCR. However, biochemical assays of SIV RT indicate T:G mismatches are the most prevalent RT substitutions. A:C mismatches during the first round of synthesis would result in A to G mutations we observed; however, these mutations were much less frequent than T:G mismatches (Manns et al., 1991). Alternatively, A to G mutations may represent a second class of Vif-responsive innate antiviral editing enzymes, distinct from APOBEC3. The RNA editing enzyme ADAR1, for example, is expressed within T-lymphocytes and has been shown to induce A to G mutations in hepatitis delta virus (Jayan and Casey, 2002; Wong and Lazinski, 2002). Increased A to G mutational frequency during SIVDvif-SA infection in H9 and CEM-SS cells suggests that A to G mutational frequency is suppressed by Vif, consistent with this hypothesis. However, relatively high rates of A to G mutations were also seen during SIVmac239 replication in H9 and CEMx174 cells with equivalent mutational frequencies between SIVmac239 and SIVDvif-SA during replication in CEMx174. Nevertheless, further studies on the possibility that Vif affects other RNA editing enzymes should be explored. Despite more than 20 years of research on HIV and SIV, the role of differential RNA splicing in pathogenesis of lentiviruses remains largely unknown. In HIV, there appears to be a great deal of plasticity in the splicing of mRNAs around SA2, SA3, and SA4 (Caputi and Zahler, 2002; Domsic et al., 2003; Schwartz et al., 1991). It is not clear whether such plasticity exists for SIV. Nevertheless, these data suggest that splicing of SIV vif is tightly controlled at SA1. Although a cryptic splice site was uncovered, Vif expression from this mRNA was inadequate to counteract the anti-SIV properties present in CEMx174 and H9 cells.

Materials and methods Oligonucleotides All oligonucleotides were synthesized and desalted by Sigma-Genosys.

Name

Sequence (5Vto 3V)

Location*

S-AA55 S-M667 S-M661 S-MH536 S-MH535 SIVDvif-SA5V SIVDvif-SA3V S-Vif-REV S-SA1V-Rev S-IN-Rev

CAAAGCAGAAAGGGTCCTAACAGAC TCCAGCACTAGCAGGTAGAGCC CACTCTCCTTCAAGTCCCTGTTC CTCAATAATAAGAAGACCCTGGTCTG GTATATGTCTAAGATTCTATGTCTTC GTCTATTATCGCGAAGGCAGAGATC GATCTCTGCCTTCGCGATAATAGAC GGCTATGCCACCTCTCTAGC GTAGTGATCATGTTAATTAATCTTTCTG TTAGCCTTTCTTCTGGGTACTACCTTA

1022 – 1046 (anti-sense) 821 – 842 (sense) 1086 – 1108 (anti-sense) 10462 – 10488 (sense) 281 – 306 (sense) 5457 – 5482 (sense) 5457 – 5482 (anti-sense) 5649 – 5668 (anti-sense) 5376 – 5403 (anti-sense) 5556 – 5582 (anti-sense)

Underlined nucleotides represent introduced mutations. * Nucleotide positions based upon SIVmac239 (accession no. M33262).

J.G. Victoria, W.E. Robinson Jr. / Virology 338 (2005) 281 – 291

289

Generation of SIVDvif-SA

Cell lysis and sample preparation for real-time PCR

A subclone of SIVmac239 containing a BamHI (nt 2105) and SphI (nt 6706) fragment was constructed and used for site-directed mutagenesis. PCR-based mutagenesis was performed using the QuikChange kit (Stratagene). Complementary primer pairs, SIVDvif-SA5Vand SIVDvif-SA3V(1 AM), were used in a PCR containing 250 AM dNTPs, 1 reaction buffer, and 2.5 U of Pfu Turbo polymerase. PCR conditions were as follows: initial denaturation at 95 -C for 30 s followed by 18 rounds of cycling at 95 -C for 30 s, 55 -C for 60 s, then 68 -C for 12 min. Following DpnI treatment to digest the parental clone, Escherichia coli (DH5-a) cells were transformed with 2 Al of the PCR product. Plasmids were isolated from independent colonies. Sequencing of plasmids confirmed the introduction of only the designed mutations. Full-length SIVDvif-SA was constructed in SIVmac239 using the BamHI and SphI restriction sites.

Cell lysis was performed as previously described (Victoria et al., 2003). Briefly, 1.0  106 cells were lysed in 100 Al of lysis solution (Kellogg and Kwok, 1990) containing 100 mM KCl, 10 mM Tris –HCl (pH 8.3), 2.5 mM MgCl2, 0.5% Tween-20, 0.5% Nonidet NP40, and 20 Ag of proteinase K. Cells were lysed for 1 h at 60 -C, followed by heat inactivation of proteinase K for 15 min at 95 -C.

Cells and viruses Cells were cultured at 37 -C in RPMI-1640 containing 2 mM l-glutamine supplemented with 11.5% fetal bovine serum (Omega Scientific Inc.). All cell lines were reduced by half and resuspended in fresh media every other day. Human CD4+ T-lymphoblastoid cells CEM-SS and H9, and a human B/T lymphocyte hybrid cell line, CEMx174, were obtained from the NIH AIDS Research and Reference Reagent Program. Plasmids encoding SIVmac239 or SIVDvif-SA were transformed into CEM-SS cells via single pulse electroporation at 200 mV for 50 ms. Infection was monitored by a modification of IFA previously described for HIV (Robinson et al., 1989). Fixed cells were incubated with either polyclonal monkey sera from SIV-infected animals or cross-reactive polyclonal human anti-HIV immune globulin, followed by fluorescein-conjugated goat anti-human IgG. The percentage of SIV antigen-positive cells was determined on a Nikon epifluorescent microscope. Cells were maintained for 6 – 10 days post-electroporation until cultures reached >50% SIV antigen positive by IFA, the supernatant was then clarified of cells by low-speed centrifugation followed by filtration through 0.45 Am cellulose acetate filters. Replication kinetics of SIVDvif-SA and SIVmac239 Early replication kinetics of SIVmac239 and SIVDvif-SA were measured in CEMx174, CEM-SS, and H9 cells using real-time PCR. Equal amounts of each virus by RT assay were inoculated onto 1  107 cells. Next, 1  106 cells were harvested for real-time PCR at 1, 2, 4, 8, 12, 24, 36, 48, and 72 h post-inoculation. To measure the time necessary for each virus to spread through the culture, IFA was performed and the percentage of cells positive for SIV antigens was calculated over a 2- to 6-week period in each cell line.

Quantitative real-time PCR Real-time PCR was performed using the Smart Cycler (Cepheid) and identical conditions to those previously reported for HIV-specific primers AA55, M661, and M667 (Victoria et al., 2003). Lysate from the equivalent of 2.0  104 cells was used in PCR under conditions: 95 -C for 150 s followed by 40 cycles of 95 -C for 15 s and 61 -C for 30 s. Melt curve analysis was utilized to distinguish bona fide products of amplification from primer dimers (Victoria et al., 2003). PCR reactions contained 0.2 mM Tris –HCl (pH 8.4), 2.0 mM MgCl2, 50 mM KCl, 0.625 U Taq DNA polymerase (Invitrogen), 0.2 AM of each primer, 200 AM dNTPs, and 0.25 SYBR Green I (Sigma-Aldrich or Molecular Probes) in a 25-Al reaction volume. Primer pairs were used to detect: -sssDNA, S-AA55/S-M667; completely synthesized SIV cDNA, S-M661/S-M667; or two LTR circle DNA, SMH535/S-MH536. PCR amplification of S-MH535/SMH536 was carried out in an identical manner as above except 5 Ag of non-acetylated bovine serum albumin, 0.15 M trehalose, and 0.2% Tween-20 were added to the reaction tubes. The extension time temperature was 59 -C and the number of cycles was increased to 45. Identifying the vif splice acceptor site Total cellular RNAs were isolated according to manufacturer’s instructions (Gentra) from 1.0  107 CEM-SS cells infected with either SIVmac239 or SIVDvif-SA. Next, cDNA was generated first by preincubation of 1 Ag of cellular RNA with S-Vif-REV primer for 5 min at 70 -C. Then, 200 U of Superscript II RT (Invitrogen), 1 manufacturer’s buffer (Invitrogen), 500 AM RNase-free dNTPs, and 10 AM DTT were added and the reactions were heated at 42 -C for 50 min followed by 72 -C for 15 min. PCR amplification of the cDNA was performed using Triple Master enzyme mix (Eppendorf). Amplification of 7 Al of cDNA from either SIVmac239 or SIVDvif-SA was performed using 1 manufacturer’s buffer, 0.4 AM of S-M667, and 0.4 AM of either S-INRev or S-SA1V-Rev primers, 2.5 units of enzyme, and 200 AM dNTPs in a 25-Al total reaction volume. Products from PCR were separated by electrophoresis through a 1.5% agarose gel and DNA was transferred to zeta probe membrane (BioRad Laboratories) in 0.4 M NaOH. S-AA55 probe was end-labeled using T4 polynucleotide kinase (New England Biolabs) and g-[32]P-ATP. Probe was

290

J.G. Victoria, W.E. Robinson Jr. / Virology 338 (2005) 281 – 291

hybridized to membrane for 2 h at room temperature in 2.5% non-fat milk containing 100 Ag/ml sonicated salmon sperm DNA. Unbound probe was removed by sequential 15 min washes in 2 SSC 0.5% SDS, 0.1 SSC 0.5% SDS, and 0.1 SSC 0.1% SDS at 60 -C. Images were scanned using a Molecular Dynamics Storm phosphorimager and quantified using IQ tool software (Molecular Dynamics). PCR amplified products were cloned into pGEM T-easy vector (Promega) and sequenced. Sequence analysis of SIVDvif-SA and SIVmac239 SIV-infected cells were lysed 48 h post-inoculation as above. Total cell lysate was used in high fidelity PCR to amplify the IN gene (between nucleotides 4776 and 5668) from the proviral DNA. PCR products were cloned into the pGEM cloning vector (Stratagene) and used for automated sequencing (Laguna Scientific) using the T7 primer.

Acknowledgments This work was supported in part by grants from the Burroughs-Wellcome Fund (App #99-2609 to W.E.R.) and a training grant from the University of California Universitywide Biotechnology Training Program (J.G.V). W.E. Robinson Jr. is a Burroughs-Wellcome Fund Clinical Scientist in Translational Research. The authors would like to thank B. McDougall and A. Harris for technical assistance and D. Lee, B. Herrera, and D. Gerth for critical reading of the manuscript and discussion.

References Amendt, B.A., Hesslein, D., Chang, L.J., Stoltzfus, C.M., 1994. Presence of negative and positive cis-acting RNA splicing elements within and flanking the first tat coding exon of human immunodeficiency virus type 1. Mol. Cell. Biol. 14, 3960 – 3970. Bilodeau, P.S., Domsic, J.K., Mayeda, A., Krainer, A.R., Stoltzfus, C.M., 2001. RNA splicing at human immunodeficiency virus type 1 3’ splice site A2 is regulated by binding of hnRNP A/B proteins to an exonic splicing silencer element. J. Virol. 75, 8487 – 8497. Bogerd, H.P., Doehle, B.P., Wiegand, H.L., Cullen, B.R., 2004. A single amino acid difference in the host APOBEC3G protein controls the primate species specificity of HIV type 1 virion infectivity factor. Proc. Natl. Acad. Sci. U. S. A. 101, 3770 – 3774. Borman, A.M., Quillent, C., Charneau, P., Dauguet, C., Clavel, F., 1995. Human immunodeficiency virus type 1 Vif-mutant particles from restrictive cells: role of Vif in correct particle assembly and infectivity. J. Virol. 69, 2058 – 2067. Caputi, M., Zahler, A.M., 2002. SR proteins and hnRNP H regulate the splicing of the HIV-1 tev-specific exon 6D. EMBO J. 21, 845 – 855. Cartegni, L., Chew, S.L., Krainer, A.R., 2002. Listening to silence and understanding nonsense: exonic mutations that affect splicing. Nat. Rev. Genet. 3, 285 – 298. Conticello, S.G., Harris, R.S., Neuberger, M.S., 2003. The Vif protein of HIV triggers degradation of the human antiretroviral DNA deaminase APOBEC3G. Curr. Biol. 13, 2009 – 2013. Desrosiers, R.C., Lifson, J.D., Gibbs, J.S., Czajak, S.C., Howe, A.Y.,

Arthur, L.O., Johnson, R.P., 1998. Identification of highly attenuated mutants of simian immunodeficiency virus. J. Virol. 72, 1431 – 1437. Domsic, J.K., Wang, Y., Mayeda, A., Krainer, A.R., Stoltzfus, C.M., 2003. Human immunodeficiency virus type 1 hnRNP A/B-dependent exonic splicing silencer ESSV antagonizes binding of U2AF65 to viral polypyrimidine tracts. Mol. Cell. Biol. 23, 8672 – 8762. Furtado, M.R., Balachandran, R., Phalguni, G., Wolinksy, S.M., 1991. Analysis of alternatively spliced human immunodeficiency virus type-1 mRNA species, one of which encodes a novel tat-env fusion protein. Virology 185, 258 – 270. Gabuzda, D.H., Lawerence, K., Langhoff, E., Terwilliger, E., Dorfman, T., Haseltine, W.A., Sodroski, J., 1992. Role of vif in replication of human immunodeficiency virus type 1 in CD4+ T lymphocytes. J. Virol. 66, 6489 – 6495. Gaddis, N.C., Chertova, E., Sheehy, A.M., Henderson, L.E., Malim, M.H., 2003. Comprehensive investigation of the molecular defect in vifdeficient human immunodeficiency virus type 1 virions. J. Virol. 77, 5810 – 5820. Gaddis, N.C., Sheehy, A.M., Ahmad, K.M., Swanson, C.M., Bishop, K.N., Beer, B.E., Marx, P.A., Gao, F., Bibollet-Ruche, F., Hahn, B.H., Malim, M.H., 2004. Further investigation of simian immunodeficiency virus Vif function in human cells. J. Virol. 78, 12041 – 12046. Garcia-Blanco, M.A., 2003. Messenger RNA reprogramming by spliceosome-mediated RNA trans-splicing. J. Clin. Invest. 112, 474 – 480. Garrett, E.D., Tiley, L.S., Cullen, B.R., 1991. Rev activates expression of the human immunodeficiency virus type 1 vif and vpr gene products. J. Virol. 65, 1653 – 1657. Gibbs, J.S., Regier, D.A., Desrosiers, R.C., 1994. Construction and in vitro properties of SIVmac mutants with deletions in ‘‘nonessential’’ genes. AIDS Res. Hum. Retroviruses 10, 333 – 342. Goncalves, J., Korin, Y., Zack, J., Gabuzda, D., 1996. Role of Vif in human immunodeficiency virus type 1 reverse transcription. J. Virol. 70, 8701 – 8709. Harris, R.S., Bishop, K.N., Sheehy, A.M., Craig, H.M., Petersen-Mahrt, S.K., Watt, I.N., Neuberger, M.S., Malim, M.H., 2003. DNA deamination mediates innate immunity to retroviral infection. Cell 133, 803 – 809. Jayan, G.C., Casey, J.L., 2002. Increased RNA editing and inhibition of hepatitis delta virus replication by high-level expression of ADAR1 and ADAR2. J. Virol. 76, 3819 – 3827. Kao, S., Khan, M.A., Miyagi, E., Plishka, R., Buckler-White, A., Strebel, K., 2003. The human immunodeficiency virus type 1 Vif protein reduces intracellular expression and inhibits packaging of APOBEC3G (CEM15), a cellular inhibitor of virus infectivity. J. Virol. 77, 11398 – 11407. Kellogg, D.E., Kwok, S., 1990. Detection of human immunodeficiency virus. In: Innis, M.A., Gelfand, D.H., Sninsky, J.J., et al., (Eds.), PCR Protocols: A Guide to Methods and Applications. Academic Press, Inc., San Diego, pp. 337 – 347. King, P.J., Robinson Jr., W.E., 1998. Resistance to the anti-human immunodeficiency virus type 1 compound l-chicoric acid results from a single mutation at amino acid 140 of integrase. J. Virol. 72, 8420 – 8424. Kobayashi, M., Takaori-Kondo, A., Shindo, K., Abudu, A., Fukunaga, K., Uchiyama, T., 2004. APOBEC3G targets specific virus species. J. Virol. 78, 8238 – 8244. Krummheuer, J., Lenz, C., Kammler, S., Scheid, A., Schaal, H., 2001. Influence of the small leader exons 2 and 3 on human immunodeficiency virus type 1 gene expression. Virology 286, 276 – 289. Lake, J.A., Carr, J., Feng, F., Mundy, L., Burrell, C., Li, P., 2003. The role of Vif during HIV-1 infection: interaction with novel host cellular factors. J. Clin. Virol. 26, 143 – 152. Luo, K., Liu, B., Xiao, Z., Yu, Y., Yu, X., Gorelick, R., Yu, X.F., 2004. Amino-terminal region of the human immunodeficiency virus type 1 nucleocapsid is required for human APOBEC3G packaging. J. Virol. 78, 11841 – 11852. Mangeat, B., Turelli, P., Caron, G., Friedli, M., Perrin, L., Trono, D., 2003. Broad antiretroviral defence by human APOBEC3G through lethal editing of nascent reverse transcripts. Nature 424, 99 – 103.

J.G. Victoria, W.E. Robinson Jr. / Virology 338 (2005) 281 – 291 Mangeat, B., Turelli, P., Liao, S., Trono, D., 2004. A single amino acid determinant governs the species-specific sensitivity of APOBEC3G to Vif action. J. Biol. Chem. 279, 14481 – 14483. Manns, A., Konig, H., Baier, M., Kurth, R., Grosse, F., 1991. Fidelity of reverse transcriptase of the simian immunodeficiency virus from African green monkey. Nucleic Acids Res. 19, 533 – 537. Mariani, R., Chen, D., Schrofelbauer, B., Navarro, F., Konig, R., Bollman, B., Munk, C., Nymark-McMahon, H., Landau, N.R., 2003. Speciesspecific exclusion of APOBEC3G from HIV-1 virions by Vif. Cell 114, 21 – 31. Marin, M., Rose, K.M., Kozak, S.L., Kabat, D., 2003. HIV-1 Vif protein binds the editing enzyme APOBEC3G and induces its degradation. Nat. Med. 9, 1398 – 1403. Mehle, A., Strack, B., Ancuta, P., Zhang, C., McPike, M., Gabuzda, D., 2004. Vif overcomes the innate antiviral activity of APOBEC3G by promoting its degradation in the ubiquitin-proteasome pathway. J. Biol. Chem. 279, 7792 – 7798. Newman, E.N., Holmes, R.K., Craig, H.M., Klein, K.C., Lingappa, J.R., Malim, M.H., Sheehy, A.M., 2005. Antiviral function of APOBEC3G can be dissociated from cytidine deaminase activity. Curr. Biol. 15, 166 – 170. Nilsen, T.W., 2003. The spliceosome: the most complex macromolecular machine in the cell? BioEssays 25, 1147 – 1149. Park, I.W., Steen, R., Li, Y., 1991. Characterization of multiple mRNA species of simian immunodeficiency virus from macaques in a CD4+ lymphoid cell line. J. Virol. 65, 2987 – 2992. Park, I.W., Myrick, K., Sodroski, J., 1994. Effects of vif mutations on cell-free infectivity and replication of simian immunodeficiency virus. J. Acquired Immune Defic. Syndr. 7, 1228 – 1236. Purcell, D.F., Martin, M.A., 1993. Alternative splicing of human immunodeficiency virus type 1 mRNA modulates viral protein expression, replication, and infectivity. J. Virol. 67, 6365 – 6378. Reinhart, T.A., Rogan, M.J., Haase, A.T., 1996. RNA splice site utilization by simian immunodeficiency viruses derived from sooty mangabey monkeys. Virology 224, 338 – 344. Robinson Jr., W.E., Montefiori, D.C., Gillespie, D.H., Mitchell, W.M., 1989. Complement-mediated, antibody-dependent enhancement of HIV-1 infection in vitro is characterized by increased protein and RNA syntheses and infectious virus release. J. Acquired Immune Defic. Syndr. (2), 33 – 42. Ropers, D., Ayadi, L., Gattoni, R., Jacquenet, S., Damier, L., Branlant, C., Stevenin, J., 2004. Differential effects of the SR proteins 9G8, SC35, ASF/SF2, and SRp40 on the utilization of the A1 to A5 splicing sites of HIV-1 RNA. J. Biol. Chem. 279, 29963 – 29973. Rose, K.M., Marin, M., Kozak, S.L., Kabat, D., 2004. Transcriptional regulation of APOBEC3G, a cytidine deaminase that hypermutates human immunodeficiency virus. J. Biol. Chem. 279, 41744 – 41749. Sakai, H., Shibata, R., Sakuragi, J., Sakuragi, S., Kawamura, M., Adachi, A., 1993. Cell-dependent requirement of human immunodeficiency virus type 1 Vif protein for maturation of virus particles. J. Virol. 67, 1663 – 1666. Schafer, A., Bogerd, H.P., Cullen, B.R., 2004. Specific packaging of APOBEC3G into HIV-1 virions is mediated by the nucleocapsid domain of the gag polyprotein precursor. Virology 328, 163 – 168. Schrofelbauer, B., Chen, D., Landau, N.R., 2004. A single amino acid of APOBEC3G controls its species-specific interaction with virion infectivity factor (Vif). Proc. Natl. Acad. Sci. U. S. A. 101, 3927 – 3932. Schwartz, S., Barbar, K.F., Pavlakis, G.N., 1991. Expression of human immunodeficiency virus type 1 vif and vpr mRNAs is Rev-dependent and regulated by splicing. Virology 183, 677 – 686. Sheehy, A.M., Gaddis, N.C., Choi, J.D., Malim, M.H., 2002. Isolation of a human gene that inhibits HIV-1 infection and is suppressed by the viral Vif protein. Nature 418, 646 – 650. Sheehy, A.M., Gaddis, N.C., Malim, M.H., 2003. The antiretroviral enzyme APOBEC3G is degraded by the proteasome in response to HIV-1 Vif. Nat. Med. 9, 1404 – 1407. Shindo, K., Takaori-Kondo, A., Kobayashi, M., Abudu, A., Fukunaga,

291

K., Uchiyama, T., 2003. The enzymatic activity of CEM15/Apobec3G is essential for the regulation of the infectivity of HIV-1 virion but not a sole determinant of its antiviral activity. J. Biol. Chem. 278, 44412 – 44416. Sova, P., Volsky, D.J., 1993. Efficiency of viral DNA synthesis during infection of permissive and nonpermissive cells with vif-negative human immunodeficiency virus type 1. J. Virol. 67, 6322 – 6326. Stopak, K., de Noronha, C., Yonemoto, W., Greene, W.C., 2003. HIV-1 Vif blocks the antiviral activity of APOBEC3G by impairing both its translation and intracellular stability. Mol. Cell 12, 591 – 601. Suspene, R., Sommer, P., Henry, M., Ferris, S., Guetard, D., Pochet, S., Chester, A., Navaratnam, N., Wain-Hobson, S., Vartanian, J.P., 2004. APOBEC3G is a single-stranded DNA cytidine deaminase and functions independently of HIV reverse transcriptase. Nucleic Acids Res. 32, 2421 – 2429. Takeuchi, H., Kao, S., Miyagi, E., Khan, M.A., Buckler-White, A., Plishka, R., Strebel, K., 2005. Production of infectious SIVagm from human cells requires functional inactivation but not viral exclusion of human APOBEC3G. J. Biol. Chem. 280, 375 – 382. Turelli, P., Mangeat, B., Jost, S., Vianin, S., Trono, D., 2004. Inhibition of hepatitis B virus replication by APOBEC3G. Science 303, 1829. Unger, R.E., Stout, M.W., Luciw, P.A., 1991. Simian immunodeficiency virus (SIVmac) exhibits complex splicing for tat, rev, and env mRNA. Virology 182, 177 – 185. Victoria, J.G., Lee, D.J., McDougall, B.R., Robinson Jr., W.E., 2003. Replication kinetics for divergent type 1 human immunodeficiency viruses using quantitative SYBR green I real-time polymerase chain reaction. AIDS Res. Hum. Retroviruses 19, 865 – 874. Viglianti, G.A., Sharma, P.L., Mullins, J.L., 1990. Simian immunodeficiency virus displays complex patterns of RNA splicing. J. Virol. 64, 4207 – 4216. von Schwedler, U., Song, J., Aiken, C., Trono, D., 1993. Vif is crucial for human immunodeficiency virus type 1 proviral DNA synthesis in infected cells. J. Virol. 67, 4945 – 4955. Wiegand, H.L., Doehle, B.P., Bogerd, H.P., Cullen, B.R., 2004. A second human antiretroviral factor, APOBEC3F, is suppressed by the HIV-1 and HIV-2 Vif proteins. EMBO J. 23, 2451 – 2458. Wong, S.K., Lazinski, D.W., 2002. Replicating hepatitis delta virus RNA is edited in the nucleus by the small form of ADAR1. Proc. Natl. Acad. Sci. U. S. A. 99, 15118 – 15123. Xu, H., Svarovskaia, E.S., Barr, R., Zhang, Y., Khan, M.A., Strebel, K., Pathak, V.K., 2004. A single amino acid substitution in human APOBEC3G antiretroviral enzyme confers resistance to HIV-1 virion infectivity factor-induced depletion. Proc. Natl. Acad. Sci. U. S. A. 101, 5652 – 5657. Yu, X., Yu, Y., Liu, B., Luo, K., Kong, W., Mao, P., Yu, X.F., 2003. Induction of APOBEC3G ubiquitination and degradation by an HIV-1 Vif-Cul5-SCF complex. Science 302, 1056 – 1060. Yu, Q., Chen, D., Konig, R., Mariani, R., Unutmaz, D., Landau, N.R., 2004a. APOBEC3B and APOBEC3C are potent inhibitors of simian immunodeficiency virus replication. J. Biol. Chem. 279, 53379 – 53386. Yu, Q., Konig, R., Pillai, S., Chiles, K., Kearney, M., Palmer, S., Richman, D., Coffin, J.M., Landau, N.R., 2004b. Single-strand specificity of APOBEC3G accounts for minus-strand deamination of the HIV genome. Nat. Struct. Mol. Biol. 11, 435 – 442. Zennou, V., Perez-Caballero, D., Gottlinger, H., Bieniasz, P.D., 2004. APOBEC3G incorporation into human immunodeficiency virus type 1 particles. J. Virol. 78, 12058 – 12061. Zhang, H., Yang, B., Pomerantz, R.J., Zhang, C., Arunachalam, S.C., Gao, L., 2003. The cytidine deaminase CEM15 induces hypermutation in newly synthesized HIV-1 DNA. Nature 424, 94 – 98. Zheng, Y.H., Irwin, D., Kurosu, T., Tokunaga, K., Sata, T., Peterlin, B.M., 2004. Human APOBEC3F is another host factor that blocks human immunodeficiency virus type 1 replication. J. Virol. 78, 6073 – 6076. Zou, J.X., Luciw, P.A., 1996. The requirement for Vif of SIVmac is celltype dependent. J. Gen. Virol. 77, 427 – 434.