Silencing Srsf6 does not modulate incomplete splicing of the huntingtin gene in Huntington's disease models

Abstract

We have previously shown that the incomplete splicing of exon 1 to exon 2 of the HTT gene results in the production of a small polyadenylated transcript (Httexon1) that encodes the highly pathogenic exon 1 HTT protein. There is evidence to suggest that the splicing factor SRSF6 is involved in the mechanism that underlies this aberrant splicing event. Therefore, we set out to test this hypothesis, by manipulating SRSF6 levels in Huntington's disease models in which an expanded CAG repeat had been knocked in to the endogenous Htt gene. We began by generating mice that were knocked out for Srsf6, and demonstrated that reduction of SRSF6 to 50% of wild type levels had no effect on incomplete splicing in zQ175 knockin mice. We found that nullizygosity for Srsf6 was embryonic lethal, and therefore, to decrease SRSF6 levels further, we established mouse embryonic fibroblasts (MEFs) from wild type, zQ175, and zQ175::Srsf6^+/- mice and transfected them with an Srsf6 siRNA. The incomplete splicing of Htt was recapitulated in the MEFs and we demonstrated that ablation of SRSF6 did not modulate the levels of the Httexon1 transcript. We conclude that SRSF6 is not required for the incomplete splicing of HTT in Huntington's disease.

Figure 1

Characterisation of the Srsf6 knockout mouse line. (a) Schematic of mouse Srsf6 with scissors denoting approximate Cas9 cleavage sites used to generate the knockout allele. Sanger sequencing was used to confirm the deletion breakpoint for the Δ990 bp line. UTR untranslated region. (b) qPCR analysis showed Srsf6 mRNA levels to be 50% of WT whereas, Srsf4 or Srsf5 levels were unchanged in cortex from 2 month old Srsf6^+/− Δ990 mice. n = 7 WT and 4 Srsf6^+/− mice. (c) Western blot analysis showed a 50% reduction in SRSF6 in cerebellum and cortex from 2 month old Srsf6^+/− mice compared to WT littermates. See Supplementary Fig. S7 for uncropped blots and total protein loading controls. n = 5/genotype. Statistical analyses were by unpaired Student’s t-tests. Test statistics can be found in Supplementary Table S4. WT wild type.

Figure 2

QuantiGene analysis of Htt transcripts in brain regions from the progeny of the zQ175 and Srsf6^+/− mouse cross. (a) Schematic of the location of the QuantiGene plex probe sets on the mouse Htt transcript. (b) Srsf6^+/− female mice were bred to zQ175 knockin male mice to generate progeny with four genotypes: WT, Srsf6 heterozygous knockout (Srsf6^+/−), zQ175 knockin and double mutants (zQ175::Srsf6^+/−). (c) Httexon1 was detected in the cortex, striatum, hippocampus and cerebellum of 2 month old zQ175 mice but was not altered by heterozygosity for Srsf6 knockout. (d) Full-length Htt was measured in the cortex, striatum, hippocampus and cerebellum using the Htt exon 50–53 assay. Cortical full-length Htt was significantly lower in zQ175 mice compared to WT and this was not changed by heterozygosity for Srsf6 knockout. n = 6/genotype. Statistical analysis was by one-way ANOVA with Bonferroni correction for multiple pairwise comparisons, ***p < 0.001, p < 0.2 values are indicated. Test statistics can be found in Supplementary Table S5. WT wild type.

Figure 3

Generation and characterisation of zQ175 mouse embryonic fibroblasts (MEFs). (a) Schematic shows workflow for derivation and characterisation of MEF cell cultures. Srsf6^+/− mice were bred to zQ175 mice. The female was sacrificed at approximately E14.5 and embryos were dissected. Mouse embryonic fibroblasts (MEFs) were isolated, cultured, passaged and expanded as required. MEFs were seeded for qPCR or QuantiGene assays as required. (b) QuantiGene analysis showed that Httexon1 was present in both the zQ175 and zQ175::Srsf6^+/− MEFs at comparable levels, and that full-length Htt was decreased in both the zQ175 and zQ175::Srsf6^+/− cells to a similar degree. n = 3 biological replicates/genotype. Statistical analysis was by one-way ANOVA with Bonferroni correction for multiple pairwise comparisons, ***p < 0.001. Test statistics can be found in Supplementary Table S6. (c,d) Htt allele discrimination qPCRs were used to measure the (c) WT and (d) zQ175 knockin Htt alleles in the MEF lines. n = 3 biological replicates/genotype. Statistical analysis was by unpaired Student’s t-test or one-way ANOVA with Bonferroni correction for multiple pairwise comparisons, ***p < 0.001, p < 0.2 values are indicated. Test statistics can be found in Supplementary Table S7.

Figure 4

RNA interference experimental plan, cytotoxicity of MEFs, Srsf6 transcript and SRSF6 protein quantification. (a) Schematic showing experimental workflow for RNA interference experiments in WT, zQ175 and zQ175::Srsf6^+/− MEFs. MEFs were seeded and transfected with either an Srsf6-targeting (siSRSF6) siRNA or a non-targeting negative control (siNC) using DharmaFECT 1 transfection reagents and harvested after 24, 48, 72 or 96 h. Cells were used in alamarBlue, QuantiGene or western blot experiments. (b) Cytotoxicity was measured using the alamarBlue assay. The viability was calculated as the signal from siSRSF6 or siNC transfected cells as a percentage of the signal from DharmaFect 1 vehicle-treated MEFs. n = 9 cell cultures/treatment group. Data were analysed by one-way ANOVAs, ***p < 0.001. Test statistics can be found in Supplementary Table S8. Data stratified by genotype can be found in Supplementary Fig. S4. (c) QuantiGene analysis showed that Srsf6 mRNA levels decreased to 80–90% of that in the siNC-treated cells by 72 h post transfection. n = 3 biological replicates/genotype. Statistical analysis was by two-way ANOVA, ***p < 0.001. Test statistics can be found in Supplementary Table S9. (d) Western blots of the SRSF6 protein in MEFs 72 h after siSRSF6 or siNC transfection. To quantify SRSF6 levels, the SRSF6 signal was normalised to a total protein loading control. SRSF6 protein levels are plotted relative to WT. n = 3 biological replicates/genotype. Uncropped blots and total protein loading controls in Supplementary Figs. S8, S9 and S10.

Figure 5

Measurement of Httexon1 and Htt transcripts in WT, zQ175 and zQ175::Srsf6^+/− MEFs after transfection with an siRNA targeting Srsf6 (siSRSF6). QuantiGene analysis was used to measure Httexon1 and Htt mRNA levels 72 h after siSRSF6 or siNC transfection. Neither Httexon1 nor full-length Htt levels were changed in MEFs treated with siSRSF6 compared to siNC. n = 3 biological replicates/genotype. Statistical analysis was by two-way ANOVA, ***p < 0.001. Test statistics can be found in Supplementary Table S10.