Translation of the poly(GR) frame in C9ORF72-ALS/FTD is regulated by cis-elements involved in alternative splicing

Abstract

GGGGCC (G₄C₂) repeat expansion in the first intron of C9ORF72 is the most common genetic cause of amyotrophic lateral sclerosis and frontotemporal dementia, two devastating age-dependent neurodegenerative disorders. Both sense and antisense repeat RNAs can be translated into 5 different dipeptide repeat proteins, such as poly(GR), which is toxic in various cellular and animal models. However, it remains unknown how poly(GR) is synthesized in patient neurons. Using a reporter construct containing 70 G₄C₂ repeats flanked by human intronic and exonic sequences, we show that translation of the poly(GR) frame does not depend on repeats or the CUG start codon in the poly(GA) frame, suggesting poly(GR) is not produced after ribosomal frameshifting in the poly(GA) frame. However, deletion analysis suggests that translation of the poly(GR) frame depends on the length of the intronic sequence 5' adjacent to G₄C₂ repeats. Moreover, several 5´ cis elements that are predicted to be involved in alternative splicing regulates poly(GR) synthesis. These results suggest that translation of repeat RNAs in the poly(GR) frame is regulated by multiple cis elements, likely through RNA secondary structures and/or associated RNA binding proteins.

Keywords: ALS; Cis elements; FTD; Poly(GR); Repeats; Translation.

Fig. 1.

Translation of the poly(GR) frame does not depend on G₄C₂ repeats or the CUG start codon in the poly(GA) frame. (A) Schematic of the bicistronic C9-Nluc reporter constructs used to analyze the GR or GA frame translation levels. The location of the CUG start codon or the UAG stop codon is shown. (B) Poly(GA) frame translation levels of different reporter constructs in which the nanoluciferase gene (Nluc) is in-frame with poly(GA). (C) Poly(GR) frame translation levels of different reporter constructs in which Nluc is in-frame with poly(GR). (D) Schematic of the intron 1 deletion constructs generated to identify the translation initiation site for the GR frame. (E) Poly(GR) frame translation levels of the constructs shown in (D). Values are mean ± S.E.M. of four to five independent transfections. **p < 0.01, ***p < 0.001 (one-way ANOVA followed by Dunnett’s multiple comparisons test). ns: not significant.

Fig. 2.

Mutations in AAGAAAA motifs increase poly(GR) translation. (A) Mutations in AAGAAAA motifs in the C9GR-Nluc construct are highlighted in red. AAGAAAA motifs are shown in blue. (B) Mutations in AAGAAAA motifs significantly increase poly(GR) production as measured by Nluc activity normalized to Fluc. (C) Motif mutations do not affect alternative splicing. Values are mean ± S.E.M. of five biological replicates. **p < 0.01, by one-way ANOVA followed by Dunnett’s multiple comparisons test. (D) Relative RNA levels of Nluc, Fluc, and pre-mRNA are not affected by motif mutations. Values are mean ± S.E.M. of three biological replicates. ns: not significant, by two-way ANOVA followed by Dunnett’s multiple comparisons test.

Fig. 3.

Alternative splicing donor sites regulate poly(GR) translation. (A) Schematic of a new splice variant caused by mutations at both V1 and V3 splice donor sites (V1V3 5´ SM). (B) The new splice variant in V1V3 5´ SM construct is detected by PCR. (C) Mutations at both V1 and V3 splice donor sites do not affect the RNA levels of Nluc, Fluc, and pre-mRNA. Values are mean ± S.E.M. of three biological replicates. ns: not significant, by two-way ANOVA followed by Dunnett’s multiple comparisons test. (D) Mutations at both V1 and V3 splice donor sites increases poly(GR) translation. Values are mean ± S.E.M. of four biological replicates. ***p < 0.001, by Welch’s t test. (E) Schematic of C9GR-Nluc with mutated V1 5´ splice donor site (V1 5´ SM) that produces variant 3 (top) and C9GR-Nluc with mutated V3 5´ splice donor site (V3 5´ SM) that produces variant 1 (bottom). (F) The effects of slice donor site mutations are revealed by PCR analysis. (G) Mutations in the V1 or V3 splice donor site do not affect the RNA levels of Nluc, Fluc, and pre-mRNA. Values are mean ± S.E.M. ns: not significant, by two-way ANOVA followed by Dunnett’s multiple comparison test. (H) Mutations in the V1 or V3 splice donor site increase poly(GR) production. Values are mean ± S.E.M. of five to 10 biological replicates. **** p < 0.0001, by one-way ANOVA followed by Dunnett’s multiple comparisons test.