RNA function. Ribosome stalling induced by mutation of a CNS-specific tRNA causes neurodegeneration

Abstract

In higher eukaryotes, transfer RNAs (tRNAs) with the same anticodon are encoded by multiple nuclear genes, and little is known about how mutations in these genes affect translation and cellular homeostasis. Similarly, the surveillance systems that respond to such defects in higher eukaryotes are not clear. Here, we discover that loss of GTPBP2, a novel binding partner of the ribosome recycling protein Pelota, in mice with a mutation in a tRNA gene that is specifically expressed in the central nervous system causes ribosome stalling and widespread neurodegeneration. Our results not only define GTPBP2 as a ribosome rescue factor but also unmask the disease potential of mutations in nuclear-encoded tRNA genes.

Fig. 1. Progressive neurodegeneration in the nmf205^−/− mice

(A to F) Hematoxylin and eosin (H and E) stained sagittal sections of nmf205^−/− and wild-type (WT; +/+) cerebella. (E and F) Higher-magnification images of cerebellar lobule IX from (C and D). (G and H) Cerebellar sections were immunostained with antibodies to cleaved caspase-3 (c-Casp3; green) and counterstained with Hoechst 33342. (I to L) H and E-stained sagittal sections of the dentate gyrus (I and J) and retina (K and L). Scale bars, 1 mm (D), 50 µm (F and J), and 100 µm (H and L). ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer.

Fig. 2. The nmf205 mutation disrupts the Pelota-interacting protein, GTPBP2

(A) Western blot analysis of GTPBP2 in wild-type (WT, +/+) and nmf205^−/− (−/−) cerebellar extracts using antibodies to the N-terminal of GTPBP2. GAPDH was used as a loading control. (B) Co-immunoprecipitation (IP) from HEK293T cells co-transfected with FLAG-fusion proteins as indicated, and Pelota-HA. Input levels were determined by immunoblotting (IB). (C) GST-pull down of brain lysate (BL). The pulldown eluate and GST-fused baits were immunoblotted as indicated. (D) Bacterially expressed MBP-GTPBP2-His or MBP-His were purified and were incubated with GST or GST-Pelota. GST-pull down products (top panel) and input (middle and bottom panels) were immunoblotted with anti-His antibody (top two panels) or visualized by Coomassie blue (bottom panel).

Fig. 3. Mutation of the CNS-specific tRNA^Arg gene n-Tr20 underlies nmf205-mediated neurodegeneration

(A) Acylated (pH 5) and deacylated (pH 9) n-Tr20 detected by northern blot analysis of acid PAGE gels. (B) Northern blot analysis of n-Tr20 in brain RNA from P30 mice. (C) H and E-stained sagittal sections of cerebella from 2-month-old B6J-nmf205^−/− (nmf205^−/−), 6-month-old Tg(n-Tr20^B6N)609; B6J-nmf205^−/−and 6-month-old wild-type (WT; B6J) mice. Higher magnification images of lobule IX from are shown in the lower panel. Scale bars: 1 mm (C, top panel) and 50 µm (bottom panel). (D) Northern blot analysis of n-Tr20, its isodecoders, and all tRNA^Arg_UCU genes. 5S rRNA was used as loading control. (E) Northern blot analysis of n-Tr20 in P0, P10, and P30 B6N and B6J brain RNA. Mature and immature n-Tr20 transcripts were quantified relative to the P0 B6N brain. (F) Northern blot analysis of n-Tr20 in the P30 B6N and B6J cortex (Cx), cerebellum (Cer), and hippocampus (Hip). Mature and immature n-Tr20 was quantified relative to the B6N cortex. (G) Northern blot analysis of B6N and B6J P30 brain RNA using pooled probes to n-Tr20/21/22/23/25 tRNAs. Bands were quantitated relative to the P0 B6N brain. Error bars indicate SEMs. All data are representative of independent experiments with ≥ 3 mice. *p < 0.05, **p < 0.005 and ***p < 0.0005 (two-tailed Student's t test).

Fig. 4. The n-Tr20 mutation induces ribosome stalling which is resolved by GTPBP2

(A) Cumulative distribution of pauses at all codons averaged between replicates. (B) The mean number of pauses ≥10 standard deviations above the background translation levels of their genes (P≥10). (C) Cumulative distribution of pause scores at AGA codons averaged between replicates. (D) The mean number of pauses (P≥10) at AGA codons. (E) Read profile for Zfp238 CDS. * indicates an AGA pause with P=45. Arrows indicate AGA codons. (See fig. S16). (F) Average pausing magnitude at AGA codons, calculated by dividing genome-wide observed reads at AGA codons by expected reads. Expectations are based on read densities in genes containing an AGA. Error bars indicate standard deviations across biological replicates. *p < 0.05 (two-tailed Student's t test). (G) Difference in the codon frequency observed in the A site at pauses (P≥10) versus the genome-wide codon frequency.