A suboptimal 5' splice site downstream of HIV-1 splice site A1 is required for unspliced viral mRNA accumulation and efficient virus replication

Abstract

Background: Inefficient alternative splicing of the human immunodeficiency virus type 1(HIV-1) primary RNA transcript results in greater than half of all viral mRNA remaining unspliced. Regulation of HIV-1 alternative splicing occurs through the presence of suboptimal viral 5' and 3' splice sites (5' and 3'ss), which are positively regulated by exonic splicing enhancers (ESE) and negatively regulated by exonic splicing silencers (ESS) and intronic splicing silencers (ISS). We previously showed that splicing at HIV-1 3'ss A2 is repressed by ESSV and enhanced by the downstream 5'ss D3 signal. Disruption of ESSV results in increased vpr mRNA accumulation and exon 3 inclusion, decreased accumulation of unspliced viral mRNA, and decreased virus production.

Results: Here we show that optimization of the 5'ss D2 signal results in increased splicing at the upstream 3'ss A1, increased inclusion of exon 2 into viral mRNA, decreased accumulation of unspliced viral mRNA, and decreased virus production. Virus production from the 5'ss D2 and ESSV mutants was rescued by transient expression of HIV-1 Gag and Pol. We further show that the increased inclusion of either exon 2 or 3 does not significantly affect the stability of viral mRNA but does result in an increase and decrease, respectively, in HIV-1 mRNA levels. The changes in viral mRNA levels directly correlate with changes in tat mRNA levels observed upon increased inclusion of exon 2 or 3.

Conclusion: These results demonstrate that splicing at HIV-1 3'ss A1 is regulated by the strength of the downstream 5'ss signal and that suboptimal splicing at 3'ss A1 is necessary for virus replication. Furthermore, the replication defective phenotype resulting from increased splicing at 3'ss A1 is similar to the phenotype observed upon increased splicing at 3'ss A2. Further examination of the role of 5'ss D2 and D3 in the alternative splicing of 3'ss A1 and A2, respectively, is necessary to delineate a role for non-coding exon inclusion in HIV-1 replication.

Figure 1

Inefficient inclusion of HIV-1 exon 2 is dependent upon a suboptimal signal at 5'ss D2. (A) Map of HIV-1 genome (NL4-3) showing the locations of 5' and 3' splice sites. The positions of Exon 2, Exon 3, and ESSV are indicated above the viral genome. Probes used to analyze HIV-1 splicing are shown above and below the viral genome and splice sites. Oligonucleotide primers used for RT-PCR analysis of viral splicing are shown above the viral genome. The BSS/SJ4.7A primer pair were used to detect the 1.8 kb, completely spliced viral mRNA species. The BSS/KPNA primer pair were used to detect the 4.0 kb incompletely spliced viral mRNA species. The probe complementary to the 3'-end of the viral mRNAs used for Northern analysis is indicated by NB. The probes used for the RNase protection assays (DPHV, A1D2, A2D3, and 601c) are represented by lines and are complementary to the splice sites to which they overlap. (B) 5'ss D2 within pNL4-3 was mutagenized as shown resulting in a consensus 5'ss signal in the infectious molecular clone NLD2UP. The previously described plasmid NEVM [15] was used as a control for increased splicing at 3'ss A2. Total RNA samples from Hela cells 48 hours post transfection with the indicated plasmids were analyzed by RT-PCR using primers specific for completely spliced viral mRNA (C) or incompletely spliced viral mRNA (D). HIV-1 RNA species are indicated on the right side of the gel by exon content, the mRNA to which they encode, and mRNA spliced at 3'ss A1 are indicated by plus signs and 3'ss A2 by asterisks. (E) Total cellular RNA from 293T cells 24 hours post transfection with the indicated plasmids was subjected to Northern blot analysis with a radiolabeled probe (NB) complementary to all HIV-1 mRNAs. (F) Northern blots were quantitated and the values shown were normalized to β-actin and β-galactosidase mRNA levels and represent the average of three independent experiments. RNA was also subjected to RPA analysis using the following riboprobes: DPHV (G), A1D2 (H), A2D3 (I), and 601c (J). Individual panels are representative of a single experiment. (K) Viral splice site utilization is represented relative to NL4-3 for each splice site. The values shown represent the average of three independents experiments and were normalized to β-actin and β-galactosidase mRNA levels.

Figure 2

Efficient HIV-1 replication is dependent upon the presence of a suboptimal signal at 5'ss D2. (A) Reverse transcriptase activity of cell-free supernatants from 293T cells transfected with either NLD2UP or NEVM mutants. Asterisks indicate a significant difference when compared to mock transfected cells from three independent experiments (p < 0.01 by Student's t-test). (B) HIV-1 p24 Gag production from transfected 293T cells was measured by subjecting ten-fold serial dilutions of cell-free supernatants to Western blot analysis using serum from an HIV-1 infected patient. (C and D) Protein from transfected 293T cells was subjected to Western blot analysis using serum from an HIV-1 infected patient or antibodies to the indicated cellular or viral gene product.

Figure 3

Restoration of HIV-1 virion production upon transient Gag, Gag-Pro, and Gag-Pro-Pol expression. (A) Open reading frames and signals contained within HPdBs. A representation of the HPdBs mRNA (grey box) is shown with the respective viral splice sites and splicing signals indicated. (B) Reverse transcriptase activity of cell-free supernatants from 293T cells transfected with the indicated plasmids, with or without the co-expression of the vector HPdBs. Black bars indicate the reverse transcriptase activity measured upon transient transfection of either HPNd, HPBs, or the indicated pNL4-3 derivative alone. Grey bars indicate the reverse transcriptase activity measured upon transient transfection of the indicated NL4-3 derivative along with HPBs. Reverse transcriptase activity represents the average of three independent experiments, normalized to the reverse transcriptase activity of supernatants from pNL4-3 transfected cells. Single asterisk indicates there is no significant difference when compared to NL4-3 transfected cells (p > 0.02 by Student's t-test) and double asterisk indicates there is no significant difference when compared to mock transfected cells (p > 0.2), from three independent experiments. (C) Protein from transfected 293T cells was subjected to Western blot analysis using serum from an HIV-1 infected patient. The HIV-1 Gag precursor (p55 Gag) and Gag proteolytic products (CA and MA) are indicated on the right.

Figure 4

Effects of non-coding exon inclusion on viral mRNA stability. (A) RNase-protection mapping of HIV-1 spliced mRNAs using the DPHV riboprobe after transient transfection of the indicated plasmids and treatment with actinomycin D for the indicated times. Quantitation of the changes in accumulation of the spliced (B) viral mRNA species after the addition of actinomycin D (normalized to cellular β-actin) is shown relative to the onset of the experiment. The co-transfected pCMVβgal110 control was used to measure LacZ turnover, and the data shown represents LacZ mRNA levels within NL4-3 co-transfected cells. The data shown represent the average of three independent experiments.