Polyadenylation of Friend murine leukemia virus env-mRNA is affected by its splicing

Authors

  • Akihito Machinaga,

    1. Department of Bioinformatics, Faculty of Engineering, Soka University, Hachioji-shi, Tokyo, Japan
    Search for more papers by this author
  • Sayaka Takase-Yoden

    Corresponding author
    1. Department of Bioinformatics, Faculty of Engineering, Soka University, Hachioji-shi, Tokyo, Japan
    • Correspondence

      Sayaka Takase-Yoden, Department of Bioinformatics, Faculty of Engineering, Soka University, 1-236, Tangi-machi, Hachioji-shi, Tokyo 192-8577, Japan. Tel: +81-42-691-2375; fax: +81-42-691-2375; email: takase@soka.ac.jp

    Search for more papers by this author

Abstract

As splicing was previously found to be important for increasing Friend murine leukemia virus env-mRNA stability and translation, we investigated whether splicing of env-mRNA affected the poly(A) tail length using env expression vectors that yielded unspliced or spliced env-mRNA. Incomplete polyadenylation was detected in a fraction of the unspliced env-mRNA products in an env gene-dependent manner, showing that splicing of Friend murine leukemia virus plays an important role in the efficiency of complete polyadenylation of env-mRNA. These results suggested that the promotion of complete polyadenylation of env-mRNA by splicing might partially explain up-regulation of Env protein expression as a result of splicing.

List of Abbreviations
CAE

cytoplasmic accumulation element

Fr MLV

friend murine leukemia virus

hnRNP

heteronuclear ribonucleoprotein

LTR

long terminal repeat

luc

luciferase

NRS

negative regulator of splicing

NXF1

nuclear RNA export factor 1

SA

splice acceptor site

SD

splice donor site

SR protein

serine/arginine rich protein

XMRV

xenotropic murine leukemia virus related virus

Friend murine leukemia virus (Fr-MLV) is a member of the simple retroviruses in the Retroviridae family, with a genome that contains (from 5′ to 3′) a 5′ LTR, a 5′ leader sequence, gag, pol, env, and a 3′ LTR. The gag gene encodes the structural proteins of the virion and the pol gene encodes a protease, reverse transcriptase and integrase. The env gene encodes the Env protein, which has a surface domain and a transmembrane domain. There is a SD in the 5′ leader sequence and a SA in the 3′ end of the pol gene. Both full-length unspliced and spliced viral mRNAs are produced in MLV-infected cells. Gag and Pol proteins are translated from unspliced mRNAs and the Env protein is translated from spliced mRNA [1]. The MLV Env protein plays important roles both in viral adsorption to host cells and in induction of neuropathogenic disease in MLV-infected mouse and rat hosts [2-8]. In previous studies, we showed that the expression level of neuropathogenic A8-MLV Env is correlated with neuropathogenicity [9, 10]. Therefore, definition of the regulatory mechanism of Env expression is important for understanding the functions of the Env protein. We previously found that splicing is important for increasing env-mRNA stability and translation [11]. However, the detailed mechanism for up-regulation of Env expression as a result of splicing is still not clear.

The mRNAs of eukaryotic cells have 3′ poly(A) tails that are produced by a two-step reaction involving endonucleolytic cleavage and subsequent poly(A) tail synthesis [12, 13]. These two reactions are well conserved from yeast to humans. The poly(A) tail of mRNA plays an important role in mRNA metabolism [14] and translational control. In general, a long poly(A) tail correlates with active translation whereas a short poly(A) tail is linked with translational repression [15]. In the ‘closed loop’ model [16], mRNA circularizes by protein–protein interactions, and the interaction of the 5′ cap and the 3′ poly(A) tail enhances translation [17, 18]. The circularization of mRNA also promotes translation by shunting of terminating ribosomes or, alternatively, it can influence initiation factor activity and thereby aid in ribosome recycling [19]. The objective of the present study was to examine whether the polyadenylation of Fr-MLV env-mRNA is affected by splicing. This information should lead to increased understanding of the detailed mechanism(s) by which the MLV Env expression is up-regulated as a result of splicing.

HeLa cells were grown in Dulbecco's Modified Eagle’s Medium–low glucose (Sigma–Aldrich, St Louis, MO, USA) supplemented with 10% FCS (MP Biomedicals, Santa Ana, CA, USA), 50 units penicillin/mL (Gibco, Gaithersburg, MD, USA) and 50 µg streptomycin/mL (Gibco) at 37 °C in 5% CO2. The cells (1 × 106 cells) were transfected with 8 µg env or luc expression vectors using Lipofectamine 2000 Reagent (Invitrogen, Carlsbad, CA, USA) diluted with Opti MEM (Invitrogen) according to the manufacturer’s instructions. Construction of the vectors used in the present study was described previously [11]. However, vectors m1 and m1gpL were previously designated proA8m1 and proA8gpL, respectively. G to T (nt 2608), G to T (nt 2614), and G to T (nt 2629) mutations were introduced into the pol region in m1 and m1gpL to suppress progeny virus production. m1 and m1gpL also had A to T (nt 2126) and T to A (nt 2777) point mutations.

HeLa cells were transfected with vectors and total RNA was extracted 24 hr post-transfection using an RNeasy Mini Kit (Qiagen, Gaithersburg, MD, USA) according to the manufacturer's instructions. After treatment with RNase-free DNase (Qiagen), 4 µg RNA was added to each ligation reaction. As shown in Figure 1, the anchor primer oligo1 (5′-GGGACAGCCTATTTTGCTAG-3′) was ligated to the 3′ end of the RNA and RT was then carried out using the poly(dT) + oligo2 primer (5′-CTAGCAAAATAGGCTGTCCCTTTTTTTTTT-3′), which has the oligo1 antisense sequence and 10 T nucleotides, or oligo2 primer (5′-CTAGCAAAATAGGCTGTCCC-3′), which is complementary to the oligo1 sequence. Nested PCR was carried out to amplify the viral mRNA poly(A) tail. In the first PCR, a forward oligo3 primer targeting the 3′ end of U3 in the 3′ LTR (5′-GCCCTATAAAAGAGCTCACAACC-3′) and a reverse oligo2 primer were used. After purification of the first PCR products using a MicroElute Clean-Up Column (Favorgen Biotech Corp., Pingtung County, Taiwan), a second PCR was carried out. In the second PCR, a forward oligo4 primer targeting the 5′ end of the R region in the 3′ LTR (5′-AGTCCTCCGACAGACTGAGT-3′) and a reverse oligo5 primer targeting the 3′ end of the oligo2 and poly(dT) sequence (5′-AAAATAGGCTGTCCCTTTTT-3′) were used. To determine the poly(A) tail length of gapdh-mRNA as a control, in the first PCR, a forward primer hGAPDH3end (5′-ACCACACTGAATCTCCCCT-3′) targeting the upstream region of the poly(A) signal and a reverse oligo2 primer were used. After purification of the first PCR products using a MicroElute Clean-Up Column (Favorgen Biotech Corp.), a second PCR was carried out. In the second PCR, a forward hGAPDH3end2 (5′-CATGTAGACCCCTTGAAGAG-3′) primer targeting the downstream region of hGAPDH3end and a reverse oligo5 primer were used. The resulting PCR products were separated by electrophoresis on a 5% polyacrylamide gel in TBE buffer (50 mM Tris-HCl [pH 8.0], 48.5 mM boric acid, 2 mM EDTA) and stained with ethidium bromide. The size of DNA contained in the band was determined by reference to a standard curve generated using the known sizes of the marker DNAs and their mobilities. To analyze the sequences of the bands separated by electrophoresis, the bands were extracted from the gels using a QIAquick Gel Extraction Kit (QIAGEN) and cloned into a pGEM-T-easy vector (pGEM-T-easy Vector System; Promega, Madison, WI, USA). The sequences of the cloned fragments in T-easy vectors were amplified using T7 (5′-GTAATACGACTCACTATAGGGC-3′) or sp6 (5′-ATTTAGGTGACACTATAGAA-3′) primers and a BigDye Terminator v 3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA, USA). The sequences were analyzed using an ABI PRISMOR 3100 Genetic Analyzer (Applied Biosystems).

Figure 1.

(a) Experimental design to determine the poly(A) tail length of mRNA by nested PCR. (b) Putative structure of the nested PCR product. *It is generally known that cleavage of the 3′ end of mRNA for poly(A) tail addition usually occurs 15–30 nt downstream of the poly(A) addition signal [20].

To detect the cDNAs from env- and luc-mRNA in the RT reaction products, after RT synthesis using the poly(dT) + oligo2 primer, PCR with forward primer s1 (5′-GAGACCCTTGCCCAGGGA-3′) and reverse primer s2 (5′-TGCCGCCAACGGTCTCC-3′) was done using Go Taq (Promega). The PCR products were separated by electrophoresis on 2% agarose gels in TBE buffer and stained with ethidium bromide. Control samples without the cDNA synthesis step did not yield these specific bands. To evaluate the copy numbers of cDNAs from env- and luc-mRNA in the RT reaction products after RT synthesis using the poly(dT) + oligo2 primer, quantitative real-time PCR was carried out as described previously [11]. Briefly, the primers and probe used to quantitate cDNA of env- and luc-mRNA were forward s1-primer, reverse s2-primer, and TaqMan ss-probe (5′-CACCACCGGGAGCTCATTTACAGGCAC-3′). The copy numbers of cDNAs corresponding to env- and luc-mRNA were determined by standard curves that were created using serially diluted splA8 plasmid and splA8L plasmid, respectively. The primers and probe used to quantitate cDNA corresponding to total RNA from the m1 and splA8 vectors were forward e1-primer (5′-AGGACCTCGGGTCCCAATAG-3′), reverse e2-primer (5′-TTAGGTAGCGGGAACGAAAGTT-3′), and TaqMan e-probe (5′-CCGAACCCCGTCCTGGCAGAC-3′). The copy numbers of cDNAs corresponding to total RNAs from the m1 and splA8 vectors were determined using standard curves that were created using serially diluted splA8 plasmid. The primers and probe used to quantitate total RNA from the m1gpL and splA8L vectors were forward L1-primer (5′-CGGCTTCGGCATGTTCA-3′), reverse L2-primer (5′-TACATGAGCACGACCCGAAA-3′), and TaqMan L-probe (5′-CACGCTGGGCTACTTGATCTGCGG-3′). The copy numbers of cDNAs corresponding to total RNA from the m1gpL and splA8L vectors were determined using standard curves that were created using serially diluted splA8L plasmid. To quantitate cDNA from gapdh-mRNA, TaqMan Human GAPDH Control Reagents containing primer sets and a probe (Applied Biosystems) were used. The copy number of cDNA from gapdh-mRNA was determined by standard curves that were created by using serially diluted gapdh T-easy vector containing a fragment of the human gapdh gene. The copy number of cDNA from mRNA of vectors was normalized relative to the copy number of cDNA from cellular gapdh-mRNA. Negative control samples without the cDNA synthesis step did not show specific amplification. To quantitate the vector DNA in transfected cells, cellular genomic DNA was extracted using a DNeasy Blood and Tissue Kit (QIAGEN) according to the manufacturer’s instructions. A sample of the resulting DNA was analyzed by real-time PCR using specific primers and the TaqMan probe with a 7500 Real-Time PCR System (Applied Biosystems). The primers and probe used to quantitate the DNA introduced by m1 and splA8 were forward e1-primer, reverse e2-primer, and TaqMan e-probe. The copy numbers of m1 and splA8 DNAs were determined using standard curves that were created using serially diluted m1 plasmid and splA8 plasmid, respectively. The primers and probe used to quantitate the DNA introduced by m1gpL and splA8L were forward L1-primer, reverse L2-primer, and TaqMan L-probe. The copy numbers of DNAs of m1gpL and splA8L were determined using standard curves that were created using serially diluted m1gpL plasmid and splA8L plasmid, respectively. The amount of gapdh DNA was measured as an internal control using TaqMan Human GAPDH Control Reagents containing primer sets and a probe (Applied Biosystems). The copy number of gapdh DNA was determined using a standard curve that was created using serially diluted gapdh T-easy vector plasmids. The copy number of DNA of introduced plasmid was normalized relative to the copy number of gapdh DNA.

The poly(A) tail length of mRNA from MLV was analyzed as described in Figure 1a. The putative structure of the nested PCR product is presented in Figure 1b. The m1 vector carried the full-length A8-MLV provirus genome and generated spliced env-mRNA (Fig. 2a). Electrophoretic analysis of the products of nested PCR using oligo4 and oligo5 primers showed a smear in the 100–200-bp range in the mRNA from samples obtained from m1-transfected cells (lane 1 of Fig. 2b). Sequence analysis showed that these smears contained multiple A nucleotides 11–21 nt downstream of the poly(A) addition signal (AAUAAA). As a control, the poly(A) tail length of gapdh-mRNA was determined in m1-transfected cells. There were no differences in the poly(A) tail smear patterns of the gapdh-mRNAs from m1- and mock-transfected cells (bottom of Fig. 2b).

Figure 2.

Analysis of poly(A) tail length of mRNA from m1- and splA8-transfected cells. (a) Structure of the MLV provirus genome, the vectors used in this study, and env-mRNA. (b) Determination of poly(A) tail length of mRNA from m1- and splA8-transfected cells by using the poly(dT) + oligo2 primer for the RT reaction. The poly(A) tail length of the mRNA from each sample was analyzed by nested RT-PCR as described in the present study and in Figure 1. The PCR products were separated by electrophoresis. As a control, the gapdh-mRNA poly(A) tail length was analyzed in m1- and splA8-transfected cells. This figure is representative of five independent experiments that yielded similar results. Sequence analysis of the indicated bands in (b) was carried out by TA-cloning as described in the present study. (c) Detection of cDNA synthesized from env-mRNA obtained from m1- and splA8-transfected cells with a poly(dT) + oligo2 primer. (d) Measurement of the copy number of cDNA synthesized from env-mRNA and total viral RNA with the poly(dT) + oligo2 primer, and the copy number of vector DNA in (□) m1- and (mim12170-gra-0001) splA8-transfected cells by real-time PCR. Amount of cDNA from env-mRNA was measured by quantitative real-time PCR using s1 and s2 primers and an ss-probe. Amount of cDNA from total viral RNA and vector DNA was measured by quantitative real-time PCR using e1 and e2 primers and an e-probe. Mean values from three independent experiments and SEM are shown. Statistical comparison was done using the t -test. *P < 0.01. (e) Determination of the poly(A) tail length of mRNA from m1- and splA8-transfected cells by using the oligo2 primer for the RT reaction. The poly(A) tail length of the mRNA from each sample was analyzed by nested RT-PCR as described in the present study and in Figure 1. As a control, the gapdh-mRNA poly(A) tail length was analyzed in m1- and splA8-transfected cells. This figure is representative of three independent experiments that yielded similar results.

To examine whether env-mRNA splicing affected polyadenylation of the mRNA, the poly(A) tail length of spliced env-mRNA and unspliced env-mRNA were compared using the m1 and splA8 vectors (Fig. 2a). The splA8 vector was designed to generate unspliced env-mRNA by deletion of the intron region in m1. Electrophoretic analysis of the nested PCR products showed a smear in the 100–200-bp range for mRNA obtained from splA8-transfected cells (lane 3 of Fig. 2b). Sequence analysis showed that these smears contained multiple A nucleotides 11–21 nt downstream of the poly(A) addition signal (AAUAAA). Interestingly, an approximately 70-bp band was also detected in RNA from splA8-transfected cells. Sequence analysis showed that this band came from mRNA in which five to eight A nucleotides were attached just downstream of the poly(A) addition signal and the oligo1 sequence (Fig. 2b). This band was not observed in m1-transfected cells, as shown in lane 1 of Figure 2b. In this experimental system, the length of the mRNA poly(A) tail could be determined in both unspliced full-length MLV mRNA and in spliced env-mRNA from m1-transfected cells. Therefore, to confirm that the first strand of cDNA synthesized by RT using the poly(dT) + oligo2 primer contained env-mRNA, PCR was carried out using s1 and s2 primers. These primers were designed to amplify the 94-bp fragment containing the splicing junction region in the cDNA from env-mRNA. As shown in Figure 2c, 94-bp bands were detected in cDNA of env-mRNA from m1- and splA8-transfected cells. All 94-bp bands were confirmed to come from env-mRNA by sequence analysis (data not shown). In addition, the copy number of cDNA synthesized from env-mRNA was evaluated by quantitative real-time PCR using s1 and s2 primers and the ss-probe. The copy number of cDNA from splA8 env-mRNA was 5.3-fold more than that synthesized from m1 env-mRNA (Fig. 2d). The copy number of cDNA synthesized from total viral RNA was also evaluated by quantitative real-time PCR using e1 and e2 primers and an e-probe, and it was not significantly different from results obtained for RNA from m1- and splA8-transfected cells. The transfection efficiency, which was measured by the copy number of plasmid DNA in transfected cells, was not significantly different in m1- and splA8-transfected cells. To confirm the finding that the approximately 70-bp band was detected only in splA8-transfected cells that contained unspliced env-mRNA but not in m1-transfected cells that contained spliced env-mRNA, the amount of template cDNA for the first PCR was normalized by the amount of cDNA synthesized from env-mRNA; the volume of the RT reaction mixture containing m1 cDNA used as a template for the first PCR was 5.3-fold that of the RT reaction mixture containing splA8 cDNA. The nested PCR reactions produced a smear in the 100–200-bp range for mRNA from m1-transfected cells, but the approximately 70-bp band was not detected (lane 2 of Fig. 2b).

As the poly(dT) + oligo2 primer used in the RT reaction contains 10 T nucleotides, mRNA with less than 10 A nucleotides cannot be reverse transcribed. To examine whether other kinds of abnormal mRNAs with less than 10 A nucleotides existed in the transcripts of m1 and splA8, the oligo2 primer, which has no T nucleotide, was used for the RT reaction and then nested PCR was carried out. Electrophoretic analysis of the nested PCR products showed a smear in the 100–200-bp range in the mRNA of the samples from m1-transfected cells (lane 1 of Fig. 2e). In RNA from splA8-transfected cells, bands in the 100–180-bp range were detected (lane 2 of Fig. 2e). Sequence analysis showed that these smears contained multiple A nucleotides, similar to the smear that was detected when the poly(dT) + oligo2 primer was used for the RT reaction (data not shown). The approximately 70-bp band was also detected in splA8-transfected cells (lane 2 of Fig. 2e). Sequence analysis showed that this band came from mRNA in which two to eight A nucleotides were attached just downstream of the poly(A) addition signal and the oligo1 sequence (Fig. 2e). There was no difference in the poly(A) tail smear patterns of control gapdh-mRNA from m1-, splA8-, and mock-transfected cells in the RT reactions using the oligo2 primer (the bottom panel of Fig. 2e).

Previous studies that used m1gpL and splA8L vectors in which the env gene in m1 and splA8 was replaced by the luc gene (Fig. 3a) demonstrated that splicing of luc-mRNA did not influence its stability or translation efficiency [11]. The effect of luc-mRNA splicing on its polyadenylation in cells transfected with m1gpL and splA8L was investigated. As shown in Figure 3b, 94-bp bands were produced from cDNA synthesized from luc-mRNA obtained from m1gpL- and splA8L-transfected cells. The copy number of cDNA from luc-mRNA was also measured by quantitative real-time PCR using s1 and s2 primers and an ss-probe. The copy number of cDNA synthesized from luc-mRNA from splA8L-transfected cells was 1.9-fold more than that from m1gpL-transfected cells (Fig. 3c). The copy number of cDNA synthesized from total vector RNA was measured by real-time PCR using L1 and L2 primers and an L-probe, and it did not differ significantly between m1gpL- and splA8L-transfected cells. The transfection efficiency also did not differ significantly between m1gpL- and splA8L-transfected cells. As shown in Figure 3d (lanes 1–3 of the upper left panel), when the poly(dT) + oligo2 primer was used for the RT reaction, the nested PCR products showed a smear in the 100–200-bp range for mRNA from all samples obtained from m1gpL- and splA8L-transfected cells, but the approximately 70-bp band was not detected. There was also no difference in the poly(A) tail smear patterns of control gapdh-mRNA from m1gpL-, splA8L-, and mock-transfected cells (lower left panel of Fig. 3d). When the oligo2 primer was used for the RT reaction, nested PCR produced a smear in the 100–200-bp range in the mRNA of the samples from m1gpL- and splA8L-transfected cells, but the approximately 70-bp band was not detected (upper right panel of Fig. 3d). There was also no difference in the poly(A) tail smear patterns of control gapdh-mRNA from m1gpL-, splA8L-, and mock-transfected cells (lower right panel of Fig. 3d).

Figure 3.

Analysis of poly(A) tail length of mRNA from m1gpL- and splA8L-transfected cells. (a) Structure of the MLV provirus genome, the vectors used in this study, and luc-mRNA. (b) Detection of cDNA synthesized from luc-mRNA obtained from m1gpL- and splA8L-transfected cells with a poly(dT) + oligo2 primer. (c) Measurement of the copy number of cDNA synthesized from luc-mRNA and total RNA from the luc-vector with a poly(dT) + oligo2 primer, and the copy number of vector DNA in (□) m1gpL- and (mim12170-gra-0001) sp1A8L-transfected cells by real-time PCR. Amount of cDNA from luc-mRNA was measured by quantitative real-time PCR using s1 and s2 primers and an ss-probe. Amount of cDNA from vectors and vector DNA was measured by quantitative real-time PCR using L1 and L2 primers and an L-probe. Mean values from three independent experiments and SEM are shown. Statistical comparison was carried out using the t-test. *P < 0.01. (d) Determination of the poly(A) tail length of mRNA from m1gpL- and splA8L-transfected cells by using the poly(dT) + oligo2 primer and the oligo2 primer for the RT reaction. The poly(A) tail length of the mRNA from each sample was analyzed by nested RT-PCR as described in the present study and in Figure 1. As a control, the gapdh-mRNA poly(A) tail length was analyzed in m1gpL- and splA8L-transfected cells. This figure is representative of five (RT by poly(dT) + oligo2 primer) or three (RT by oligo2 primer) independent experiments that yielded similar results.

Incomplete polyadenylation was detected in a fraction of the unspliced env-mRNA products (Fig. 2b,e). Sequence analysis of the approximately 70-bp band showed that a fraction of the unspliced env-mRNA was cleaved just downstream of the poly(A) addition signal and two to eight A nucleotides were added (Fig. 2b,e). As the oligo5 primer, which was used for the second PCR, contains five T nucleotides, the first PCR product with less than five A nucleotides cannot be amplified. Thus, it is possible that other kinds of abnormal env-mRNA, such as env-mRNA having one or no A nucleotide just downstream of the poly(A) addition signal or env-mRNA that was cleaved upstream of the poly(A) addition signal, existed in a fraction of the unspliced env-mRNA. As cleavage of the 3′ end of mRNA for poly(A) tail addition usually occurs 15–30 nt downstream of the poly(A) addition signal [20], the 3′ ends of the approximately 70-bp band fraction of unspliced env-mRNA were probably abnormally processed unspliced env-mRNA. The mechanism(s) by which splicing affects processing of the 3′ end of env-mRNA is not well understood, but there are known to be functional interactions between spliceosomes and the 3′ end-processing machinery of eukaryotic mRNA [21]. Splicing factors that associate with the 3′ terminal intron also interact with downstream polyadenylation factors to promote both 3′ end cleavage/polyadenylation and terminal intron splicing [21-26]. In Rous sarcoma virus and avian sarcoma virus, splicing factors promote 3′ end processing through the cis-acting NRS elements [27, 28]. Wilusz and colleagues proposed a model in which hnRNP H and SR proteins compete to bind to the NRS. In this model, bound SR proteins bridge between the NRS and 3′ LTR, promoting recruitment of the 3′ end processing machinery. Therefore, splicing factors might assist in processing of the 3′ end of env-mRNA through interactions with factors involved in 3′ end processing. In contrast, electron microscopic analyses of nascent transcripts from Drosophila embryos have shown that splicing occurred co-transcriptionally [29]. In the current era, it is generally thought that RNA processing, including splicing, is coupled to transcription [30]. Thus, one cannot exclude the possibility that the abnormal env-mRNA found in a fraction of the unspliced env-mRNA was produced by pre-termination of transcription. When the oligo2 primer was used in the RT reaction to analyze the poly(A) tail length, the size of the smear band (100–180 bp) that was detected in splA8-transfected cells was smaller than that (100–200 bp) in m1-transfected cells (Fig. 2e). It is possible that the amount of mRNAs with a longer poly(A) tail may be larger in spliced env-mRNA than in unspliced env-mRNA. However, because in this experimental system, the length of the mRNA poly(A) tail could be determined for either the unspliced full-length MLV mRNA or the spliced env-mRNA from m1-transfected cells, this requires further analyses. The data presented herein, together with previous reports, suggest that splicing of MLV plays an important role in complete polyadenylation of env-mRNA, although the mechanisms are still unknown. Interestingly, when the env gene in m1 and splA8 was replaced by the luc gene, splicing did not affect the 3′ end structure of luc-mRNA (Fig. 3). These data suggested that there were positive cis-elements within the env region to complete the polyadenylation of env-mRNA by splicing.

Our previous study showed that following the introduction of splA8 to cells, the Env expression level was 0.2-fold lower than that for cells with m1. Furthermore, the half-life of env-mRNA from splA8 was significantly lower than that for m1 [11]. Recently, it was reported that the nuclear export receptor NXF1 was involved in nuclear export of RNA transcripts, especially unspliced mRNA, of gamma retroviruses including XMRV and MLV [31]. A conserved cis-acting element was identified in the pol gene of gamma retroviruses, named the CAE. The SA of XMRV (5638 nt) existed within the CAE (5607–5752 nt), but, in Fr-MLV A8, the SA at 5489 nt was actually used and it existed upstream of the CAE (5620–5766 nt) [32]. As unspliced env-mRNA, which was produced in splA8-transfected cells, had a complete CAE region, nuclear export of the unspliced env-mRNA was unaffected. In fact, our previous study revealed that there was no difference in the distribution of nuclear and cytoplasmic env-mRNA from m1- and splA8-transfected NIH3T3 cells, which produced spliced env-mRNA and unspliced env-mRNA, respectively [11]. In the present study, incomplete polyadenylation was detected in a fraction of the unspliced env-mRNA obtained from splA8-transfected cells (Fig. 2b,e). In retroviruses, polyadenylation of the 3′ end of mRNA plays an important role in translation initiation and mRNA stability [33, 34]. It is possible that the appearance of incomplete polyadenylation in unspliced env-mRNA is partially correlated with a reduction of its stability and translation efficiency. However, as the 3′ end of most unspliced env-mRNA was processed correctly, this suggests that other factors also contribute to promote env-mRNA stability and translation efficiency by splicing.

In conclusion, the present study shows that polyadenylation of env-mRNA in Fr-MLV was affected by its splicing in an env gene-dependent manner. The results suggest that promotion of complete env-mRNA polyadenylation by splicing might partially explain up-regulation of Env protein expression as a result of splicing.

ACKNOWLEDGMENT

This work was supported in part by funding from MEXT (Ministry of Education, Culture, Sports, Science and Technology): the Matching Fund for Private Universities, S0901015, 2009–2014.

DISCLOSURE

The authors have no conflicts of interest associated with this study.

Ancillary