SLIRP Regulates the Rate of Mitochondrial Protein Synthesis and Protects LRPPRC from Degradation

Abstract

We have studied the in vivo role of SLIRP in regulation of mitochondrial DNA (mtDNA) gene expression and show here that it stabilizes its interacting partner protein LRPPRC by protecting it from degradation. Although SLIRP is completely dependent on LRPPRC for its stability, reduced levels of LRPPRC persist in the absence of SLIRP in vivo. Surprisingly, Slirp knockout mice are apparently healthy and only display a minor weight loss, despite a 50-70% reduction in the steady-state levels of mtDNA-encoded mRNAs. In contrast to LRPPRC, SLIRP is dispensable for polyadenylation of mtDNA-encoded mRNAs. Instead, deep RNA sequencing (RNAseq) of mitochondrial ribosomal fractions and additional molecular analyses show that SLIRP is required for proper association of mRNAs to the mitochondrial ribosome and efficient translation. Our findings thus establish distinct functions for SLIRP and LRPPRC within the LRPPRC-SLIRP complex, with a novel role for SLIRP in mitochondrial translation. Very surprisingly, our results also demonstrate that mammalian mitochondria have a great excess of transcripts under basal physiological conditions in vivo.

Fig 1. Loss of SLIRP compromises the stability of mitochondrial mRNAs and LRPPRC.

(A) Immunoblot of SLIRP and LRPPRC protein levels in heart, liver and kidney mitochondria from 12-week old wild-type (Slirp ^+/+), Slirp homozygous knockout (KO, Slirp ^-/-) and Lrpprc heterozygous KO (Lrpprc ^+/-) mice. VDAC was used as a loading control. (B) Representative confocal microscopy images of Slirp ^+/+ (left) and Slirp ^-/- (right) MEFs, stained for LRPPRC and TOM20 as a mitochondrial marker. Magnifications of the dashed boxed areas show merged channels of LRPPRC and TOM20 (bottom, left) and single channels of LRPPRC (bottom, right). Scale bars presented, 10 μm. (C) In organello de novo transcription assays performed on heart and liver mitochondria isolated from 12-week old Slirp ^+/+ and Slirp ^-/- mice. VDAC was used as a loading control. (D) Mitochondrial tRNA steady-state levels assessed by northern blotting in hearts of 12-week old Slirp ^+/+ and Slirp ^-/- mice. (E) Mitochondrial transcript steady-state levels assessed by qRT-PCR in hearts from 12-week old Slirp ^+/+ (white bars) and Slirp ^-/- (black bars) mice, as well as in hearts from 12-week old Lrpprc control (Lrpprc p/p, light grey bars) and conditional KO (Lrpprc p/p, cre, dark grey bars), n = 5. Error bars represent SEM. * p value < 0.05. ** p value < 0.01. *** p value < 0.001. (F) Measurement of mitochondrial transcript poly(A) tail length from heart mitochondria of Slirp ^+/+ (white bars), Slirp ^-/- (black bars), Lrpprc ^+/- (yellow bars) and Lrpprc p/p, cre (grey bars) mice. Error bars represent SEM. n.s. means not significant. * p value < 0.05.

Fig 2. SLIRP loss affects the engagement of mitochondrial mRNAs with the mitochondrial ribosome.

(A) Sedimentation profile in sucrose density gradient of transcripts and ribosomal proteins from liver mitochondria from 12-week old wild-type (Slirp ^+/+, left panel) and Slirp homozygous knockout (Slirp ^-/-, right panel) animals. Individual mitochondrial transcripts were detected by qRT-PCR. The abundance of a given RNA in each fraction is shown as percentage of the total level in the control. The migration of the small (28S) and large (39S) mitochondrial ribosomal subunits, and of the assembled mitochondrial ribosome (55S) is assessed by the profiles of the 12S (blue line) and 16S (red line) rRNAs as well as by the migration of subunit-specific ribosomal proteins (MRPL37 and MRPS35) detected by immunoblotting. (B) Individual mRNA sedimentation profiles from the gradient described in (A). Slirp ^+/+ profiles are depicted in purple and Slirp ^-/- profiles are depicted in orange. Individual mitochondrial mRNAs were detected by qRT-PCR and the mRNA distribution profile is shown after normalization to controls, i.e. the quantity of a given mt-mRNA, named RNAx, in each fraction of the Slirp ^-/- gradient was normalized to the total RNAx_Slirp+/+/total RNAx_Slirp-/- ratio, where total RNAx is the sum of the RNAx quantity detected across all the fractions. (C) Hierarchical clustered expression levels of mitochondrial transcripts (log10FPKM) across all fractions from both Slirp ^+/+ and Slirp ^-/- mitochondria. (D) Hierarchical clustered log2 fold changes in transcript expression for each fraction, showing the overall decrease in mRNA levels of Slirp ^-/- compared to Slirp ^+/+ mitochondria.

Fig 3. SLIRP regulates the rate of translation but is not essential for respiratory chain activity.

(A) Mitochondrial translation rate assessed by in cellulo ³⁵S-methionine/cysteine pulse labelling for 10, 30 and 60 minutes in wild-type (Slirp ^+/+) and Slirp homozygous knockout (Slirp ^-/-) primary MEFs. (B) Mitochondrial translation rate assessed by in organello ³⁵S-methionine pulse labelling for 10, 30 and 60 minutes in isolated heart and liver mitochondria from 12-week old Slirp ^+/+ and Slirp ^-/- mice. (C) Steady-state levels of the mitochondria- and nucleus-encoded subunits of the respiratory chain complexes as assessed by western blot analysis of protein extracts from heart and liver mitochondria from 12-week old Slirp ^+/+ and Slirp ^-/- mice. (D) Oxygen consumption rate of isolated liver mitochondria from 12-week old Slirp ^+/+ (white bars) and Slirp ^-/- (black bars) mice. Isolated mitochondria were incubated with complex I or complex II substrates. Each set of substrates was successively combined with ADP (to assess the phosphorylating respiration), oligomycin (to assess the non-phosphorylating respiration) and CCCP (to assess uncoupled respiration). n = 3. Error bars represent SEM. (E) Activity of the respiratory chain complexes I (CI), II (CII) and IV (CIV) of liver mitochondria from 12-week old Slirp ^+/+ and Slirp ^-/- mice. Citrate synthase activity (CS) was used as a control. n = 3. Error bars represent SEM.

Fig 4. LRPPRC is degraded by LONP1 in the absence of SLIRP and LRPPRC alone cannot preserve mitochondrial transcript stability.

(A) Immunoblotting of SLIRP and LRPPRC protein levels in liver and kidney mitochondria from 12-week old wild-type (Slirp ^+/+, Lrpprc ^+/+) and Slirp homozygous knockout (Slirp ^-/-, Lrpprc ^+/+) mice and mice overexpressing Lrpprc on a Slirp homozygous knockout background (Slirp^-/-, Lrpprc^+/T). VDAC was used as a loading control. (B) Immunoblotting of LRPPRC protein levels in Slirp ^+/+ and Slirp ^-/- primary MEFs after transfection with a siRNA directed against the expression of the LONP1 protease (siLonp1) or with a scrambled siRNA (siCtrl). LONP1 was detected to assess the efficiency of the knockdown (KD), SUV3L1 and PNPT1 were detected to assess the steady-state level of the mitochondrial RNA degradosome and SDHA was used as a loading control (left panel). The right panel represents the quantification of three independent experiments. Error bars represent SEM. * p value < 0.05. *** p value < 0.001. (C) QRT-PCR analysis of the mitochondrial transcript steady-state levels after KD of Lonp1 expression in Slirp ^+/+ and Slirp ^-/- MEFs as described in (B). n = 6. Error bars represent SEM. * p value < 0.05. ** p value < 0.01. *** p value < 0.001. n.s. means not significant.