Premature termination codons in PRPF31 cause retinitis pigmentosa via haploinsufficiency due to nonsense-mediated mRNA decay

Abstract

Dominant mutations in the gene encoding the mRNA splicing factor PRPF31 cause retinitis pigmentosa, a hereditary form of retinal degeneration. Most of these mutations are characterized by DNA changes that lead to premature termination codons. We investigated 6 different PRPF31 mutations, represented by single-base substitutions or microdeletions, in cell lines derived from 9 patients with dominant retinitis pigmentosa. Five of these mutations lead to premature termination codons, and 1 leads to the skipping of exon 2. Allele-specific measurement of PRPF31 transcripts revealed a strong reduction in the expression of mutant alleles. As a consequence, total PRPF31 protein abundance was decreased, and no truncated proteins were detected. Subnuclear localization of the full-length PRPF31 that was present remained unaffected. Blocking nonsense-mediated mRNA decay significantly restored the amount of mutant PRPF31 mRNA but did not restore the synthesis of mutant proteins, even in conjunction with inhibitors of protein degradation pathways. Our results indicate that most PRPF31 mutations ultimately result in null alleles through the activation of surveillance mechanisms that inactivate mutant mRNA and, possibly, proteins. Furthermore, these data provide compelling evidence that the pathogenic effect of PRPF31 mutations is likely due to haploinsufficiency rather than to gain of function.

Figures

Figure 1. Effect of the PRPF31 c.1115_1125del mutation on its own transcripts.

Mutant PRPF31 alleles carrying this mutation (vertical dotted bar) result in either a long mRNA form, which bears the deletion and a PTC just before the natural termination codon (stop) in exon 14, or a short mRNA form, in which exon 11 is skipped during splicing and a PTC is created in exon 12. The gray areas indicate the portion of the transcripts that have an out-of-phase reading frame. The drawing is not to scale.

Figure 2. Capillary electrophoresis of semiquantitative RT-PCRs spanning 6 different PRPF31 mutations.

We analyzed unprocessed PCR products obtained with 6 different primer pairs, specific for each mutation, that could simultaneously amplify the wild-type and the mutant mRNA (cDNA) in the same reaction. The single curves shown are representative of 5 replicates from each of 3 independent cultures. The x axis indicates the approximate size (in bp) of the DNA fragments, whereas the y axis shows the amount of PCR product normalized to the peak height of the wild-type allele. PCR products originating from the mutant alleles are indicated by asterisks. For c.1115_1125del, the peak at ~197 bp is the short form produced by the skipping of exon 11, whereas the peak at ~260 bp is the NMD-insensitive long-form mRNA allele containing the 11-bp deletion. For the c.856-2A>G mutation, the red curve corresponds to cell line 12688, whereas the blue curve corresponds to cell line 13190. For the c.1115_1125del mutation, the PCR products from cell lines AG0293, AG0305, and AG0307 are indicated by the green, red, and blue curves, respectively.

Figure 3. Allelic expression of nuclear PRPF31 pre-mRNA in c.1115_1125del heterozygotes.

(A) The mutant and wild-type alleles in cell lines from 3 patients at steady state were expressed at a ratio of approximately 50:50. (B) When cell line AG0307 was treated with actinomycin D to block RNA transcription, the amount of PRPF31 pre-mRNA was progressively depleted because of splicing (0–60 minutes) followed by its intrinsic decay (after 60 minutes; data on specific effects of splicing and decay are not shown). (C) Nonetheless, the ratio between wild-type and mutant nuclear pre-mRNA was approximately the same and remained constant over time, which indicated that the half-lives of the 2 forms are the same.

Figure 4. Differential protein abundance and absence of truncated PRPF31 proteins in cell lines derived from patients.

(A) Quantitative SDS-PAGE (10%) of total protein extracts from various patient cell lines and controls. Simultaneous infrared detection of proteins was achieved with a highly specific N-terminal anti-PRPF31 antibody (green) and a β-actin control (red). The white dotted boxes indicate the expected position of mutant proteins derived from NMD-sensitive alleles, if they were present. Protein size cannot be predicted from the mutation in cell line 14523; however, it is likely to lack the N-terminal epitope for detection with this antibody. (B) Original magnification of another quantitative, high-resolution SDS-PAGE (15%) using protein extracts from a cell line carrying the c.1115_1125del mutation (AG0307) and from an unaffected control (AG0309). A high-percentage acrylamide gel, run over a longer time, was used to resolve the small size difference between PRPF31 proteins derived from the wild-type and the mutant long-form mRNA. However, only the full-length (wild-type) PRPF31 form was visible. (C) Quantification of PRPF31 protein abundance from cell lines derived from patients using a combination of 4 independent gels, including the one depicted in A, and normalized to the abundance of β-actin. Black bars, control cell lines; white bars, cell lines from patients with PRPF31 mutations.

Figure 5. Full-length PRPF31 protein is not mislocalized in patient-derived cell lines.

Full-length PRPF31 (red) was detected by indirect immunofluorescence using a C-terminal antibody. A speckled staining pattern that was exclusively nuclear was observed, similar to PRPF31 localization in other cell types. Merged images demonstrated extensive, but not complete, colocalization with the marker for splicing speckles SC35 (green). No differences in staining patterns could be observed between affected individuals and unaffected controls, which indirectly suggests that PRPF31 function may not be impaired in patients. Blue indicates DAPI nuclear staining, and the scale bar applies to all images.

Figure 6. Quantitative rescue of NMD-sensitive mutant mRNA alleles with cycloheximide and emetine treatment.

Representative images of agarose gel and capillary electrophoresis of semiquantitative RT-PCR products from mutant PRPF31 transcripts are shown (for emetine only). For each cell line, PCR products were always run on the same agarose gel, but they were not contiguous. Data shown summarize measurements of 5 PCR replicates from 3 (emetine) or 4 (cycloheximide) treatments on independent cultures for each cell line. Cell line identification codes are shown on the left. The term ctrl refers to semiquantitative RT-PCR from a control cell line (GM104848). Throughout, the symbols + and – indicate the presence or absence of NMD inhibitors in the cell culture medium. The terms mut and wt correspond to PCR products from plasmids containing the mutant and wild-type alleles, respectively. Electrophoretograms were normalized to the height of the wild-type peak and aligned automatically. Black and red curves indicate the absence or the presence of emetine in the culture medium, respectively, and asterisks highlight PCR products derived from the mutant mRNA form. Treatment with either cycloheximide (blue bars) or emetine (red bars) partly restored the expression of NMD-sensitive mutant alleles and had no effect on NMD-insensitive mutant alleles. For cell lines AG0293, AG0305, and AG0307, bars show only the change in the NMD-sensitive allele (short form). Data presented as bar graphs are the combination of both densitometry values of agarose gel bands (ImageJ software) and the integration of peak from capillary electrophoresis (Biocalculator software).

Figure 7. The mRNA containing the c.1115_1125del mutation (long form) is insensitive to NMD.

Treatment with emetine, puromycin, or cycloheximide partly rescued the relative abundance of mutant mRNA in the cell line 14266 (positive control, blue bars), but not the mRNA containing the 11-bp mutation in exon 11 (long form) extracted from cell lines AG0293, AG0305, AG0307 (respectively from left to right, bars in shades of orange), as measured by allele-specific real-time PCR. Nuclear (Nuc) and cytoplasmic (Cyt) mutant mRNA fold changes for each cell line (long form only) were calculated by dividing the mutant/wild-type allelic ratio following each NMD inhibitor treatment by the corresponding mutant/wild-type allelic ratio with no drug treatment. The rescue of the 14266 mutant mRNA was more pronounced in the presence of emetine and was more effective on Cyt RNA. Below: Allele-specific real-time PCR measurement of the long mRNA form in untreated cells showed that NMD did not affect the mRNA level of the long form of the c1115_1125del mutant allele, but was active on the 14266 mutant mRNA. Data represent the average of 3 real-time PCR replicates for each cell line and each treatment.

Figure 8. Inhibition of NMD and the lysosomal and proteasomal protein degradation pathways does not rescue the synthesis of mutant PRPF31 proteins.

(A) The addition of wortmannin (WRT) markedly enhanced the stability of most mutant forms of mRNA as measured by semiquantitative RT-PCR, except for mRNA derived from the c.177+1delG mutation that skips the initiation codon. A similar and sometimes more pronounced rescue was observed in the presence of WRT along with the lysosome inhibitor chloroquine (CHQ) and the proteasome inhibitor MG132. Data summarize measurements of 5 PCR replicates from 3 independent cultures treated as shown for each of the 6 cell lines. (B) Quantitative SDS-PAGE of the corresponding protein extracts from the same cell preparations. WRT treatment alone did not rescue the expression of any mutant proteins, nor did the addition of lysosomal and proteasomal inhibitors. Sample 14523 can be used as a negative control because it lacks the N-terminal epitope needed for recognition with this antibody.