Mechanistic insights into non-coding Y RNA processing

Figure 1.. Intact RO60 binding site is essential for 3’ end ysRNA biogenesis. (A) Wild type human RNY5 secondary structure (predicted) including the three mutant pool regions that were selected for 3’ end mutagenesis. In each of the mutant pools, L1 (red), L2 (green) and L3 (blue) near the 3’ end cleavage site (indicated between positions 49 and 50) 5 nt substitution mutations were introduced resulting in 1,024 possible combinations for each library. (B) Workflow of the 3’ end RNY5 mutagenesis. The mutations for L1, L2 and L3 were introduced using primers containing random nucleotides. sRNA cDNA libraries were generated and sequenced. (C) Sequence logo analysis of most abundant L3 reads shows strong preference for G at the 5th position. (D) Sequence logo analysis of missing mutant pool 3 reads shows absence of G at the 5th position. (E) Predicted structure of a highly abundant mutant pool 3 sequence motif folds similar to wild type RNY5 retaining an intact cytidine bulge for RO60 binding. The mutated region of L3 is depicted in dark blue whereas RO60 is shown in light blue binding to the cytidine bulge. (F) Predicted structure of a missing mutant pool 3 sequence motif indicates a change in…

Figure 2.. 3’ end ysRNA production is Y RNA structure dependent. L1 mutants generate longer ysRNAs when compared to L2/L3 mutants. (A) Total read percentage for each cDNA library replicate of L1/L2/L3 mutant pools at each size ranging from 25–34 nts. L2/L3 mutant derived ysRNAs (green and blue) are mainly 32/33 nt. L1 mutant derived ysRNA reads (red) are longer with a length of 34/35 nt. (B) Predicted structures of the two most abundant L1 RNY5 mutants generate longer ysRNA reads compared to wild type RNY5. (C) Most abundant L1 RNY5 mutants generate longer ysRNAs compared to wild type RNY5. (D) Northern blot of the least abundant L1 RNY5 mutants shows no ysRNAs. (E) L1 RNY5 mutants with the same predicted structure as wild type generate wild type sized ysRNAs. Apart from the L1 RNY5 mutants that fold the same way as wild type RNY5 the structural features of wild type RNY5 including the stem S, internal loop L2a and stem S3 close to the 3’ end cleavage site are shown. (F) L1 RNY5 with the same structure than wild type RNY5 produce wild type sized ysRNAs.

Figure 3.. A bulge of 1 nt is required for 3’ end Y RNA cleavage. (A) All L2a loop mutants except for L2a Δ8 nt and L2a Δ9 nt were predicted to fold similarly to wild type RNY5. (B) Sequential deletion of internal loop L2a had a small effect on ysRNA production from either the 3’ or the 5’ end of RNY5. Except for L2a loop Δ3 nt to L2a loop Δ6 nt, mutants showed decreased ysRNA levels when compared to wild type RNY5 derived ysRNAs. L2a loop Δ9 nt did not produce ysRNAs from neither the 3’ nor the 5’ end of RNY5. (C) Most of the RNY5 substitution mutants except for M9 and M14 folded similarly to wild type RNY5. (D) Only M5 and M7 were less efficiently processed to ysRNAs when compared to wild type RNY5.

Figure 4.. 5’ end Y RNA cleavage is UGGGU sequence dependent. (A) Structure of wild type human RNY5 including the three mutant pool regions that were selected for 5’ end mutagenesis analysis. In each of the regions of mutant pool L4 (blue), L5 (Orange) and L3 (purple) near the 5’ end cleavage site (indicated with an arrow between positions 32 and 33) 5 nt substitution mutations were introduced resulting in 1,024 possible combinations for each library. (B) L4/L5 mutants produced mainly 30 nt ysRNAs whereas L6 mutants produced mainly 30/31 nt ysRNAs. (C) Sequence logo analysis showed a strong selection for the wild type RNY5 sequence motif UGGGU. (D) Sequence logo analysis showed that for all abundant L5 produced ysRNAs there was strong selection for the first three positions to be UAU. (E) Predicted L4 RNY5 mutants fold similar to wild type RNY5 and the two most abundant L4 RNY5 mutants are shown as representatives. (F) L4 RNY5 mutants generate wild type sized ysRNAs. (G) Predicted RNY5 mutant structures with mutations at position 22. U was replaced by A, C and G. (H) Y RNA cleavage was affected at the 5’ end of RNY5 if the U at position 22 of RNY5 was mutated.

Figure 5.. Conserved UUAU motif involved in 3’ and 5’ end cleavage. (A) Wild type RNY5 structure with the structural feature loop L2b. The structural feature loop L2b at positions 23–26 is shown in Orange. (B) Predicted structures of RNY5 L1 mutants with high or no ysRNA accumulation. RNY5 L1 mutants that produce ysRNAs in a good amount are shown in the upper panel whereas the RNY5 L1 mutants with no ysRNAs from the 3’ and 5’ end are shown in the lower panel. (C) RNY5 L1 mutants with an intact internal loop L2b generate a good amount of ysRNAs whereas RNY5 mutants with a shorter internal loop L2b produce less/no ysRNA from the 3’ nor the 5’ end of RNY5. (D) Predicted structures with mutated loop L2b with the same structure than wild type RNY5 (upper panel) or different structure than wild type RNY5. (E) Mutations in loop L2b resulted in a lower accumulation of ysRNAs from the 3’ and 5’ end of RNY5.

Figure 6.. RO60 is critical for Y RNA cleavage. (A) Mouse embryonic stem cells lack Ro60 protein in a Ro60−/− knock out cell line. (B) Ro60−/− mES cells did not generate ysRNAs from the 3’ or 5’ end of Rny1 (mY1) when cells were treated with poly (I:C) using the human RNY1 probe. (C) RO60 is critical for RNY5 cleavage at the 3’ end. The 5’ probe for RNY5 appears to hybridize with an unknown RNA that is slightly longer than RNY5. (D) RNASEL contributes to Y RNA cleavage in mouse cells using wild type and RNASEL−/− mouse fibroblast cells. (E) tRNA cleavage was abolished in RNASEL−/− cells human cells upon poly (I:C) treatment confirming RNASEL removal in this mutant. RNASEL knock out did not affect ysRNA generation in human cells as much as it did in mouse cells. (F) RNY5 substitution mutants in which the U of the RNASEL favoured UN^N motif was replaced by A, G or C. These mutants were transfected into 3T3 cells. The mutations introduced at position 49 did not have any effect on Y RNA expression or cleavage from the 3’ or 5’ end of RNY5 in mouse cells.