Dissecting the mechanism of NOP56 GGCCUG repeat-associated non-AUG translation using cell-free translation systems

Abstract

The repeat expansion in the human genome contributes to neurodegenerative disorders such as spinocerebellar ataxia (SCA) and amyotrophic lateral sclerosis. Transcripts with repeat expansions undergo noncanonical translation called repeat-associated non-AUG (RAN) translation. The NOP56 gene, implicated in SCA36, contains a GGCCTG repeat in its first intron. In tissues of patients with SCA36, poly (Gly-Pro) and poly (Pro-Arg) peptides, likely produced through NOP56 RAN translation in (NOP56-RAN), have been detected. However, the detailed mechanism underlying NOP56-RAN remains unclear. To address this, we used cell-free translation systems to investigate the mechanism of NOP56-RAN and identified the following features. (i) Translation occurs in all reading frames of the sense strand of NOP56 intron 1. (ii) Translation is initiated in a 5' cap-dependent manner from near-cognate start codons upstream of the GGCCUG repeat in each frame. (iii) Longer GGCCUG repeats enhance NOP56-RAN. (iv) A frameshift occurs within the GGCCUG repeat. These findings provide insights into the similarities between NOP56-RAN and other types of RAN translation.

Keywords: NOP56; RAN translation; SCA; cell-free translation; non-AUG translation; spinocerebellar ataxia.

Figure 1

NOP56 gene structure and the GGCCUG-derived putative dipeptide repeats. A, NOP56 gene structure. The GGCCUG repeat in intron 1 of the NOP56 gene and its flanking regions are shown in detail. The red box highlights the region used in this study. B, schematic of putative dipeptide repeats derived from the NOP56 GGCCUG repeat.

Figure 2

NOP56-RAN translation in cell-free translation systems using nano-luciferase reporter. A, schematic of nano-luciferase (Nluc)-3 × HA reporters. Nluc was used as a reporter enzyme to quantify translation efficiency. B and C, anti-HA Western blot of translation products in each cell-free translation system using the Nluc reporters. B, Rabbit reticulocyte lysate (RRL), (C) HeLa lysate. Asterisks (∗) indicate non-specific bands. Predicted molecular sizes: AUG-Nluc: 25 kDa, GL (0): 41 kDa, AW (+1): 46 kDa, and GP (+2): 39 kDa. Anti-β-actin (bAct) was used as a loading control. D and E, expression of NOP56-Nluc reporters normalized to GGG-Nluc expressed in cell-free translation systems. D, RRL and (E) HeLa lysate. Error bars represent standard deviations (±SD) from three independent experiments.

Figure 3

Identification of initiation sites in each reading frame. A, mutations to identify NOP56-RAN start codons. The GGCCUG repeat length is 16 repeats. For example, “mut0-1” refers to the first candidate start codon in the 0 frame. (B-D) Expression of the reporters with mutation relative to the wild-type (WT) reporter expressed in HeLa lysate. B, GL (0), (C) AW (+1) and (D) GP (+2). Error bars represent ±SD from three independent experiments. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001, two tailed Student’s t test. E, LC-MS/MS analysis of the N-terminal region of the HA-immunoprecipitated and trypsin-digested poly AW-71 repeat protein expressed in RRL. LC-MS/MS analysis identified two peptides, 1-VGVSACVR, 2-MVGVSACVR. The MS spectrum for peptide 1 is shown, with related MS spectra provided in Fig. S5.

Figure 4

Repeat length dependency in NOP56-RAN. A, schematic of the NOP56-RAN reporter with varying repeat (rp) numbers in each frame. B-D, relative expression levels of NOP56-RAN reporters normalized to the 0 repeat in HeLa lysate. B, GL (0), (C) AW (+1), and (D) GP (+2). Error bars represent ±SD from three independent experiments. ∗∗∗p < 0.001, one-way ANOVA with Dunnett’s multiple comparison test.

Figure 5

Frameshift analysis in NOP56-RAN. A, Schematic of reporters with an AUG driving expression through an N-terminal FLAG (red) tag in AW (+1), followed by a C-terminal 3 × HA (orange) tag fused to the NOP56 intron 1 with 3 or 71 repeats (rp). B, Anti-FLAG and HA Western blot of the FLAG (AW)-NOP56-Nluc, HA reporter mRNAs expressed in each frame in RRL. Asterisks (∗) indicate non-specific bands. Predicted molecular sizes for anti-FLAG: AW-GL and AW-GP (+2) are smaller than 46 kDa (71 rp); AW-AW in frame: 46 kDa or high molecular weight (71 rp). Predicted molecular sizes for anti-HA: AW-GL not detected; AW-AW in frame: 46 kDa; AW-GP: 40 to 47 kDa. C, Luciferase activity of frameshifted products (AW-to-GL, AW-to-GP) expressed in RRL. Relative values are shown when the luciferase activity for in-frame (AW-to-AW) is set to 1. Error bars represent ±SD from three independent experiments. ∗p < 0.05; ∗∗p < 0.01, two tailed Student’s t test. D, Schematic of reporters with mutations at putative frameshift sites. E, Frameshift efficiency of the AW-to-GP mutations at putative frameshift sites. Relative expression of AW-AW expressed in RRL. Error bars represent ±SD from three independent experiments. n.s. = not significant; ∗p < 0.05; ∗∗p < 0.01, two tailed Student’s t test.

Figure 6

NOP56-RAN in human cultured cells. A, schematic of Nluc-3 × FLAG reporters. Nluc was used as a reporter enzyme to quantify translation efficiency. ATG-Fluc was included in all assays for normalization. B, anti-FLAG Western blot of the NOP56-Nluc reporter expressed in HeLa cells. Asterisks (∗) indicate non-specific bands. Predicted molecular sizes of the products: AUG + Nluc: 21 kDa, GL (0): 38 kDa, AW (+1): 43 kDa, and GP (+2): 36 kDa. C, relative expression of the NOP56-Nluc reporters normalized to ATG-Fluc in HeLa cells. Error bars represent standard deviations (±SD) from six independent experiments. D, top, schematic of stress induction by thapsigargin (TG). Bottom, relative Nluc/Fluc expression of the NOP56-Nluc reporters normalized to the HeLa cells without TG. Error bars represent standard deviations (±SD) from six independent experiments. n.s. = not significant; ∗∗p < 0.01; ∗∗∗p < 0.001, two tailed Student’s t test.