WBSCR16 is essential for mitochondrial 16S rRNA processing in mammals

Abstract

Mitochondrial rRNAs play important roles in regulating mtDNA-encoded gene expression and energy metabolism subsequently. However, the proteins that regulate mitochondrial 16S rRNA processing remain poorly understood. Herein, we generated adipose-specific Wbscr16^-/-mice and cells, both of which exhibited dramatic mitochondrial changes. Subsequently, WBSCR16 was identified as a 16S rRNA-binding protein essential for the cleavage of 16S rRNA-mt-tRNA^Leu, facilitating 16S rRNA processing and mitochondrial ribosome assembly. Additionally, WBSCR16 recruited RNase P subunit MRPP3 to nascent 16S rRNA and assisted in this specific cleavage. Furthermore, evidence showed that adipose-specific Wbscr16 ablation promotes energy wasting via lipid preference in brown adipose tissue, leading to excess energy expenditure and resistance to obesity. In contrast, overexpression of WBSCR16 upregulated 16S rRNA processing and induced a preference for glucose utilization in both transgenic mouse models and cultured cells. These findings suggest that WBSCR16 plays essential roles in mitochondrial 16S rRNA processing in mammals, and is the key mitochondrial protein to balance glucose and lipid metabolism.

Graphical Abstract

Graphical Abstract.

Introduction

Mitochondria are vital organelles producing the majority of ATP through oxidative phosphorylation (OXPHOS) via the electron transport chain (ETC) (1,2). The ETC consists of five protein complexes, with most of the subunits encoded by the nuclear genome and the remaining 13 proteins encoded by the mitochondrial genome (mtDNA) (3,4). Apart from being regulated by nuclear genome-encoded factors, mitochondrial OXPHOS is affected by mtDNA gene expression (5–7). As mitochondrial RNA is transcribed in the form of large polycistronic precursors, there are limitations in regulating mtDNA gene expression at the transcription level (8). After being transcribed, polycistronic mitochondrial RNA processing is thought to be an important way to regulate mitochondrial OXPHOS (9). To separate individual messenger RNA (mRNA), transfer RNA (tRNA), and ribosomal RNA (rRNA) from nascent mitochondrial RNA, it is necessary to cleave at specific sites with RNase P and RNase Z (ELAC2) (10–12). Mitochondria-targeted RNase P is a protein complex composed of three subunits, including mitochondrial RNase P protein 1 (MRPP1), MRPP2 and MRPP3 (11,13). MRPP3 and ELAC2 both contain RNA-binding domains (RBDs) and are directly involved in the 5′- and 3′-end of tRNA cleavage respectively (9,12,14). After that, rRNA undergoes certain post-transcriptional modifications (PTMs) such as methylation and pseudouridylation to generate mature RNA molecules that assemble with mitochondrial ribosomal proteins (15). MTG2 and PTCD1 have been identified as PTM regulators of mitochondrial 16S rRNA maturation (16,17), but the molecular mechanisms of RNA cleavage regulation remain poorly understood.

We previously identified WBSCR16 as a mitochondrial guanine nucleotide exchange factor (GEF), regulating mitochondrial dynamics and cell survival (18). WBSCR16 deficiency has been demonstrated to downregulate mtDNA gene expression due to impaired mitochondrial ribosome assembly (19–21). Besides, WBSCR16 was also shown to form a complex with NME6 and boost its nucleoside diphosphate kinase activity to maintain local mitochondrial pyrimidine triphosphate levels, which is necessary for sustaining mitochondrial RNA abundance (22). Despite these findings, the precise mechanism underlying how WBSCR16 contributes to mitochondrial ribosome assembly and its physiological significance remains elusive. In humans, alternative splicing gives rise to three different WBSCR16 isoforms, termed RCC1L^V1, RCC1L^V2 and RCC1L^V3, which share the same N-terminus (19). Among them, RCC1L^V1 is the predominant isoform and shares significant homology with the only isoform of mouse WBSCR16 (19). RCC1L^V1 and RCC1L^V3 were shown to play roles in mitochondrial ribosome biogenesis and protein translation, with RCC1L^V1 interacting with mt-LSU and RCC1L^V3 interacting with mt-SSU, respectively (19). To further elucidate the physiological roles of WBSCR16 in vivo, we adopted mouse models and mouse embryo fibroblasts (MEFs) systems in this study.

Adipose tissue maintains energy homeostasis through fat storage or lipolysis. Brown adipose tissue (BAT), characterized by its abundance of mitochondria, is capable of dissipating energy through mitochondrial OXPHOS and thermogenesis (23,24). Under cold exposure, BAT relies on circulating free fatty acids and glucose to maintain optimal thermogenesis (25,26). Both glucose and fatty acids fuel mitochondria as critical metabolic substrates, for which the preference is altered upon abnormal mitochondrial OXPHOS, a process termed metabolic flexibility (27,28). For instance, adipocytes were shown to passively take more advantage of fatty acids rather than glucose as fuels under metabolic dysfunctions such as insulin resistance (29,30). Currently, the modulation of mitochondrial activity in adipose tissues is recognized as an effective approach to improving insulin sensitivity, enhancing energy dissipation and restoring whole-body energy homeostasis (31–33). However, which mitochondrial factors govern metabolic flexibility and substrate preference in adipose tissues remains elusive.

In this study, we found that WBSCR16 bound to nascent 16S rRNA transcripts and recruited MRPP3 to facilitate the cleavage of 16S rRNA-mt-tRNA^Leu, thereby promoting 16S rRNA processing and mtDNA gene expression. WBSCR16 acts as a key determinant of metabolic flexibility in adipocytes, as Wbscr16 deficiency impaired mitochondrial translation and induced substrate utilization of more fatty acids to replace glucose. Ultimately, Wbscr16 deficiency increased energy wastage to contribute to hyperlipidemia prevention and obesity resistance. Targeting WBSCR16 offers a new strategy to combat obesity or other metabolic diseases.

Materials and methods

Animals

All animal experiments were performed according to procedures approved by the Institutional Animal Care and Use Committee for animal care and handling at the Shanghai Institute of Endocrine and Metabolic Diseases. Wbscr16 (Rcc1l)^flox/flox mice, Wbscr16-Del^tg/tg mice (with HA-Tag), Wbscr16 (Rcc1l)^tg/tg mice (with HA-Tag) and Adiponectin-Cre mice were obtained from Cyagen Biosciences. The Wbscr16-Del variant lacks exon 9 of the Wbscr16 gene. Ucp1-CreERT2 mice were obtained from the Shanghai Model Organisms. We hybridized Wbscr16^flox/flox (floxed) mice with Adiponectin-Cre mice to obtain Wbscr16^−/- (AWBSCR16 KO) mice and hybridized Wbscr16^tg/tg mice or Wbscr16-Del^tg/tg with Ucp1-CreERT2 mice to obtain Wbscr16^UCP1tg or Wbscr16-Del^UCP1tg mice. We also crossed Wbscr16^flox/flox with Adiponectin-Cre mice and Wbscr16^tg/+ mice to generate AWBSCR16 KO + KI mice. Mice were maintained on a standard rodent chow at ambient temperature (22°C) under 12-h light/12-h dark cycles. Littermates expressing no Cre (floxed mice) were used as controls within the above animal experiments. Samples were obtained from male mice at 8–10 weeks. Mice under a high-fat diet (HFD) were treated with 60% fat, 20% protein and 20% carbohydrate for 3 months.

Metabolic studies

Body fat composition was determined via MRI (Minispec TD-NMR Analyzers).

Glucose tolerance test (GTT) and insulin tolerance test (ITT)

For GTT, mice were fasted for 14 h and injected with D-glucose (2 g·kg⁻¹ body weight) (Sigma) intraperitoneally. For ITT, mice were fasted for 4 h and injected with recombinant human insulin (1 U kg⁻¹, Novo Nordisk A/S) intraperitoneally. Blood glucose was measured by a glucose meter (Abbott) using whole blood from the tail vein, and fasting glucose was measured after 14 h of starvation, with ad libitum access to water.

Serum analyses

After mice were anesthetized with chloral hydrate, their blood was collected and centrifuged for 15 min to obtain serum after 30 min of standing. Serum non-esterified fatty acids (NEFAs) (Labassay) and insulin (Mercodia) levels were measured by ELISA.

Histology and immunoblotting

Mouse tissues were fixed in 4% paraformaldehyde and embedded in paraffin. Sections were stained with hematoxylin and eosin or oil red O according to standard protocols and observed with an inverted microscope (Olympus). Quantification of cross-sectional areas of lipid droplets (LDs) in adipose tissues was performed using ImageJ software. The electron microscopic images of MEFs or BATs were obtained with an H-7650 (HITACHI) transmission electron microscope. Quantification of mitochondrial number and size was performed using ImageJ software.

Cell culture and genetic modification

MEFs were derived from wild-type C57BL/6 mice at embryonic days (E) 13.5–15.5, and subsequently immortalized with large T-antigen. The gRNA sequence CCCGACGCAGGATCCAGCCGGT was designed to delete Wbscr16 in MEFs by using CRISPR technology. pLV-Puro-EF1A containing Wbscr16-FLAG, pLV-Bsd-EF1A containing Mrpp3-HA, pLV-U6-Scramble, pLV-Puro-U6-shWbscr16 and pLV-Bsd-U6-shMrpp3 lentiviral vectors were obtained from VectorBuilder. MSCV-IRES-Puro containing Wbscr16-Del-FLAG was obtained from Genscript. Retroviruses and lentiviruses were produced according to the manufacturer’s instructions. After viral infection, cells were selected for 4 days with 1 mg/ml puromycin or 5 mg/ml blasticidin and left to recover for over 24 h before analysis.

Mitochondrial DNA content

Total DNA was extracted and purified from BAT with the Mammalian genomic DNA extraction kit (Tiangen). 10 ng was used to perform qPCR to quantify mitochondrial and nuclear DNA markers. Mitochondrial DNA content (16S) was expressed relative to the genomic Tert gene. Specific primers were listed in Supplementary Table S1.

Primary brown adipocyte preparation and substrates detection

Primary cultures of brown or white adipocytes were obtained by isolating preadipocytes from interscapular brown fat depots of newborn Wbscr16^flox/flox mice or inguinal white adipose depots of 2-month-old Wbscr16^flox/flox mice, followed by induced differentiation into mature adipocytes. Preadipocytes were induced in a growth medium with 5 μg/ml insulin (Sigma), 0.5 mM isobutylmethylxanthine (IBMX, Sigma), 1 μM dexamethasone (DEX, Sigma), 1 nM triiodothyronine (T3, Sigma) and 5 μM Rosiglitazone (Sigma) for 48 h, and then refreshed with growth medium supplemented with insulin, T3 and Rosiglitazone for the next 4 days. To obtain Wbscr16 knockout, the above primary adipocytes were transduced with adenovirus containing Cre (OBiO Technology) after differentiating for 6 days. Dulbecco's Modified Eagle Medium (DMEM) with glucose or palmitic acid (dissolved in bovine serum albumin without fatty acids) was refreshed to carry out the substrate utilization assays. Mitochondria and LDs of primary adipocytes were stained using 0.5 μM Mito-Tracker Red/Green (Invitrogen, 37°C, 5 min) or 0.5 μM BODIPY (493/503) (Invitrogen, 37°C, 5 min). 1 μg/ml BODIPY (558/568) C12 (Invitrogen) was incubated for 15 h to label the fatty acids recruited into the LDs of mature adipocytes. A high-content analysis system (PerkinElmer, Opera Phenix) was used to capture images and quantify fluorescence intensity and LD size.

Glucose utilization, palmitic acid utilization and ATP content measurement

MEFs and preadipocytes were seeded onto 96-well plates. After inducing differentiation to mature adipocytes, glucose utilization during the indicated time period was detected by the Glucose Oxidase Method (GOD) (Applygen) according to the manufacturer’s protocol. Palmitic acid utilization during the indicated time period was measured by the Free Fatty Acid Quantification Kit (BioVision) according to the manufacturer’s protocol. ATP production in MEFs and mature brown adipocytes was measured using the ATP Assay Kit (Beyotime) according to the manufacturer’s protocol.

Metabolomics analysis

The metabolites were extracted from the cell residue of mature adipocytes with 1 ml pre-chilled methanol/acetonitrile/water (v/v, 2:2:1) under sonication for 1 h in ice baths. The mixture was incubated at −20°C for 1 h followed by centrifugation at 14 000 g, 4°C for 20 min and then transferred to the sampling vial for liquid chromatography-mass spectrometry (LC-MS) analysis (Bioprofile). The LC-MS portion of the platform was based on a Shimadzu Nexera X2 LC-30AD system equipped with an ACQUITY UPLC BEH Amide column (1.7 μM, 2.1 mm × 100 mm, Waters) and a triple quadrupole mass spectrometer (5500 QTRAP, AB SCIEX). Raw MRM data files were processed by peak finding, alignment, extraction and filtering using MultiQuant software. SIMCAP 14 software (Umetrics, Umeå, Sweden) was used for all multivariate data analyses and modeling. Data were mean-centered using Pareto scaling. Models were built on principal component analysis (PCA). The discriminating metabolites were obtained using a statistically significant threshold of fold change (FC) and two-tailed Student’s t-test (P-value) on the normalized raw data. The P-value was calculated by one-way analysis of variance (ANOVA) for multiple group analyses.

RNA immunoprecipitation (RIP) assay, RNA isolation and qRT-PCR

MEFs expressing Wbscr16-FLAG, MEFs expressing Mrpp3-HA with Wbscr16 knockdown and MEFs expressing Mrpp3-HA along with Wbscr16-FLAG or Wbscr16-Del-FLAG were lysed with RNA immunoprecipitation (RIP) buffer containing 25 mM Tris–HCl (pH 7.4), 150 mM KCl, 5 mM EDTA, 0.5 mM dithiothreitol (DTT), 0.5% NP-40 in the presence of RNasin (Thermo Fisher Scientific) and protease inhibitors. After centrifugation at 13 000 rpm for 10 min, the supernatants were incubated with anti-FLAG magnetic agarose (Invitrogen) or anti-IgG magnetic beads for 3 h at 4°C. Beads were washed with RIP buffer three times and then eluted with TRIzol reagent (Invitrogen). After that, the eluates were extracted and analyzed by quantitative reverse transcription PCR (qRT-PCR) (Applied Biosystems QuantStudio 7) or RNA Immunoprecipitation Sequencing (RIP-Seq) analysis (Bioprofile). The total RNA of BATs was extracted using TRIzol reagent according to the manufacturer’s instructions. Relative RNA expression was detected by qRT-PCR. Primers used for RIP assay and qRT-PCR were provided in Supplementary Table S1. Data were normalized to Gapdh and analysed by using the ΔΔC_T method.

Northern blot

The total RNA of BAT was extracted using TRIzol reagent. 15 μg of RNA was loaded onto 1% formaldehyde gels alongside an RNA Ladder (Roche). After electrophoresis overnight at 25 V, RNA was transferred onto a positively charged nylon membrane (Amersham) overnight and crosslinked at 80°C for 2 h. Membranes were then pre-hybridized for 2 h at 50°C in digoxigenin (DIG) Easy Hyb buffer (Roche). Digoxigenin-labeled oligonucleotide probes were hybridized to the membranes overnight at 50°C. Afterward, membranes were washed with a stringency wash solution three times, incubated in anti-digoxigenin-AP (Roche) for 30 min at room temperature, and exposed on Lumifilm with CSPD chemiluminescence solution. α-Tubulin was used as a loading control. Oligonucleotide probes were synthesized with the PCR DIG Probe Synthesis Kit (Roche), and corresponding sequences were listed in Supplementary Table S1.

RNA sequencing and alignments

RNA sequencing (RNA-Seq) was performed on total RNA extracted from three floxed and three AWBSCR16 KO mice after removing cytoplasmic rRNA (Ribo-Zero^TM mouse rRNA removal kit). The RNA sequencing was carried out using the Illumina HiSeq platform, according to the Illumina Tru-Seq protocol for Library preparation, sequencing, and bioinformatics analyses (Gene Denovo).

Recombinant protein purification

To obtain recombinant WBSCR16-GST fusion protein, full-length Wbscr16 was cloned into the pGEX-4T-1 vector. The construct was validated by sequencing and transformed into competent cells of Escherichia coli BL21 (DE3). The transformed bacteria were cultured in the Lysogeny Broth medium until the OD₆₀₀ reached ∼1.0. Induction was performed by adding 0.5 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) and incubating the cells for 20 h at 16°C. The cells were then harvested by centrifugation at 4000 g for 20 min, resuspended in lysis buffer and lysed on ice for 30 min with gentle vortexing. The resulting lysate was subjected to centrifugation at 15 000 g for 10 min. Afterward, the supernatant was incubated overnight with GST-tag purification resin (Beyotime, P2262), following which the resin was washed using lysis buffer. Proteins bound to the resin were subsequently eluted with elution buffer (50 mM Tris, 150 mM NaCl, 10 mM GSH, pH 8.0). Additionally, the recombinant MRPP3-His fusion protein was acquired from Proteintech (Ag15629, 230–583 amino acids encoded by MRPP3, not the full-length protein). The purity of recombinant proteins was analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) with Coomassie Brilliant Blue staining.

Protein preparation, co-immunoprecipitation and western blot analysis

Proteins from MEFs or tissues were extracted by RIPA buffer and quantified using BCA kits (Thermo Fisher Scientific). Protein samples were electrophoresed by SDS-PAGE and probed on polyvinylidene fluoride (PVDF) membranes using primary antibodies (Supplementary Table S2). Detailed information on the antibodies was provided in Supplementary Table S2. To generate a better polyclonal antibody to detect WBSCR16 in western blot analysis, purified recombinant WBSCR16 protein was coupled to BSA as an antigen to immunize rabbits. Anti-serum was collected after four doses of immunization (WBSCR16 polyclonal antibody in this study).

MEFs were transduced with lentivirus or retrovirus carrying the vector, Wbscr16-FLAG (WT-WBSCR16-FLAG), Wbscr16-Del-FLAG (WBSCR16-Del-FLAG) or Mrpp3-HA (MRPP3-HA), as indicated. After screening for 4 days with puromycin or blasticidin, proteins were extracted using NP-40 lysis buffer (Beyotime) added with protease inhibitor (Roche). Total proteins for co-immunoprecipitation were extracted from interscapular brown adipose tissue (iBAT) and liver tissues of Wbscr16^UCP1tg mice or Wbscr16-Del^UCP1tg mice by RIPA lysis buffer (Epizyme) supplemented with protease inhibitor (Roche). The co-immunoprecipitation was performed using anti-FLAG magnetic agarose (Invitrogen) or anti-HA magnetic beads (Thermo Fisher Scientific). Immunopurified protein was eluted with SDS containing 2x loading buffer (Bio-Rad) and subjected to western blots along with input samples. For pull-down assays, WBSCR16-GST and/or MRPP3-His (Proteintech) recombinant proteins were incubated with Glutathione agarose (Thermo Fisher Scientific) or anti-HA magnetic beads (Thermo Fisher Scientific). Bound proteins were eluted in SDS-containing 2x loading buffer and subjected to western blots.

RNA pull-down assay

MEFs were lysed in 500 μl of NP-40 lysis buffer (Beyotime) containing an EDTA-free protease inhibitor cocktail and RNasin (Thermo Fisher Scientific) on ice for 2 h. After centrifugation at 13 000 rpm for 10 min, the supernatant was hybridized with 100 nM biotinylated DNA oligo probes (Biosune, detailed information on the probes was provided in Supplementary Table S1) against endogenous RNA at room temperature for 30 min. Streptavidin magnetic beads (Invitrogen) were added to each binding reaction containing 20 mM Tris–HCl (pH 7.5), 50 mM NaCl, 2 mM MgCl₂, 15% glycerol and 0.1% Tween 20, and further incubated at 4°C for 3 h. The beads were washed with wash buffer containing 20 mM Tris–HCl (pH 7.5), 10 mM NaCl and 0.1% Tween 20 for five times, and eluted with SDS-containing 2x loading buffer. The retrieved proteins were detected by western blots.

Sucrose gradient analysis of mitochondrial ribosomes

Mitochondria isolated from BAT (34) were solubilized in lysis buffer containing 260 mM sucrose, 100 mM KCl, 20 mM MgCl₂, 10 mM Tris–HCl (pH 7.5), 1% Triton X-100 with EDTA-free protease inhibitor cocktail and RNasin. The lysate was incubated on ice for 20 min and then centrifugated at 9200 g for 45 min at 4°C. The supernatant was loaded onto a 12.5 ml of linear 10–30% sucrose gradient containing 10 mM Tris–HCl (pH 7.5), 100 mM KCl, 20 mM MgCl₂ with EDTA-free protease inhibitor cocktail, and centrifuged at 24 000 rpm for 15 h at 4°C by SW 41Ti (Beckman Coulter). The gradients were collected from the top into 15 fractions. Each fraction was diluted in SDS-containing 2x loading buffer or TRIzol reagent and analysed by western blots or qRT-PCR.

PET-CT imaging

Glucose uptake into BAT was quantified using positron emission tomography (PET) analysis of ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) uptake. Mice were fasted for 12 h and subsequently administered with ¹⁸F-FDG (200 μCi/mouse) via the tail vein under isoflurane anesthesia. After 50 min, PET/CT imaging (Siemens) was performed for 20 min, with imaging and anesthesia conditions remaining consistent. The PET image was then reconstructed using the OSEM3D method, and the final PET images were directly viewed on the Inveon Research Workstation (IRW). To quantify glucose uptake of BATs, a region of interest (ROI) was manually delineated on the merged PET/CT image specifically for the iBAT area, enabling the measurement of the maximum standardized uptake values (SUV max) of the BAT.

Surface plasmon resonance

The BIAcore T200 equipment (BIAcore, GE Healthcare) was used for the surface plasmon resonance (SPR) analyses at 25°C. Initially, the equipment was preconditioned in PBST by serially injecting two 20-s pulses of 50 mM HCl, 0.05% SDS and 50 mM NaOH at a flow rate of 30 μl/min. Subsequently, the CM5 research-grade sensor chip (GE Healthcare) was activated by EDC (1-(3-dimethylkeratin)-3-ylcarbodiimide hydrochloride) and N-hydroxyapatite at a 5 μl/min flow for 20 min. Following activation, MRPP3-His protein was diluted to a concentration of 50 μg/ml in 10 mM sodium acetate at pH 4.0. The sample was then injected at a flow rate of 5 μl/min for 4 min, resulting in an expected coupling amount of ∼4233.3 RU. Subsequently, serially diluted WBSCR16-GST protein samples (ranging from 0 to 100 μg/ml) were injected for 120 s over the oligonucleotide-derivatized surfaces during the association phase, with a buffer flow for 10 s in the dissociation phase. The chip surfaces were regenerated with a 60-s injection of 10 mM NaOH. Additionally, a buffer duplication procedure (0 μM protein concentration) was carried out alongside the series of injections. Before data processing, each set of injections was background-subtracted (0 μM injection). Finally, the data were analyzed by fitting it to a 1:1 binding model using Biacore Evaluation 2.0 to determine the equilibrium dissociation constant (K_D).

Statistical analysis

All of the experiments were performed at least in triplicate, or otherwise indicated. Data of western blots were digitalized and analyzed using either the histogram panel of Adobe Photoshop or the ImageJ software. Statistical analyses were performed using Microsoft Excel and the Prism 7 software. The data are presented as mean ± S.E.M. of absolute values or percentages of control. The values obtained in vivo or in vitro for the different parameters studied were compared using a Student’s two-tailed unpaired t-test or two-way analysis of variance for comparison of two groups.

Results

Mitochondrial alterations in Wbscr16 knockout adipose tissues and cells

Whole-body deficiency of Wbscr16 causes early embryonic lethality and mitochondrial fragmentation (35). BATs contain abundant and metabolically active mitochondria and are capable of regulating energy homeostasis and thermogenesis (24). To investigate the physiological roles of mitochondrial WBSCR16 in adipose tissues, adipocyte-specific Wbscr16 knockout (AWBSCR16 KO) mice were generated by crossing Wbscr16^flox/flox mice with Adiponectin-Cre mice. As expected, WBSCR16 protein levels were significantly decreased in iBAT and white adipose tissue (WAT) of AWBSCR16 KO mice, while remaining unchanged in the liver (Figure 1A). Interestingly, the relative tissue weight and LD size of iBAT were increased in AWBSCR16 KO mice compared to controls (Figure 1B–D). However, the relative tissue weights of inguinal white adipose tissue (iWAT) and epididymal white adipose tissue (eWAT) decreased, resulting in a lower fat/lean ratio in AWBSCR16 KO mice (Figure 1E–G). Additionally, there were no differences in whole body weight and plasma insulin levels between AWBSCR16 KO and control mice (Supplementary Figure S1A and B). The LD size and anatomical structure of iWAT, eWAT and liver remained normal (Figure 1E and Supplementary Figure S1C–E).

Figure 1.

Adipose tissue and mitochondrial characterizations in Wbscr16 knockout mice and cells. (A) Representative western blots of WBSCR16 in iBAT, WAT and liver from floxed and AWBSCR16 KO mice were shown as indicated. Asterisks denote non-specific signals, and arrows indicate the band corresponding to WBSCR16. (B–D) Representative images and HE staining of iBAT (B), relative weight normalized to body weight (C) (n = 9–11), and cross-sectional area of LDs in iBAT (D) (n = 100) were shown. (E–G) Representative images and HE staining of WAT (E), relative weight normalized to body weight (F) and fat mass-to-lean mass ratio (G) (n = 10–15) were shown. (H–J) Electron microscopic images of iBAT (H), along with quantification of mitochondrial number (I) (n = 8) and size (J) (n = 52–68) in iBAT from floxed and AWBSCR16 KO mice were shown. (K) Representative western blots of WBSCR16 in overexpressed (OE) or knockout (KO) MEFs were shown. (L and M) Electron microscopic images of MEFs (L) and quantification of mitochondria size (M) (n = 31–42) in control, Wbscr16-OE, and KO cells were shown. Data were presented as mean ± SEM; t-test: *P < 0.05, **P < 0.01, ***P < 0.001.

In comparison to controls, dramatic alterations in mitochondrial morphology were observed in iBAT from AWBSCR16 KO mice, including increased mitochondrial numbers and sizes, as well as a substantial reduction in tubular cristae (Figure 1H–J). However, the copy number of mtDNA remained normal in iBAT from AWBSCR16 KO mice (Supplementary Figure S1F). Mitochondria contain a considerable amount of iron, which imparts a brown color to BAT. Thus, the color of BAT partly reflects its mitochondrial content (24). Notably, the color of crude mitochondrial extracts from iBAT was lighter in AWBSCR16 KO mice (Supplementary Figure S1G). These data support the conclusion that WBSCR16 plays an important role in mitochondrial maintenance.

To explore whether mitochondrial alterations in adipocytes were directly caused by the loss of Wbscr16, rather than by secondary effects in vivo, we established stable cell lines with Wbscr16 knockout (KO) or overexpression (OE) in mouse embryonic fibroblasts (MEFs) (Figure 1K). As expected, the mitochondrial morphology of MEFs was dramatically affected, with variations in mitochondrial size and the number of tubular cristae corresponding to WBSCR16 protein levels (Figure 1L–M). These findings indicate that WBSCR16 is crucial for maintaining mitochondrial integrity in cultured cells and adipose tissue.

Mitoribosomal assembly defects in Wbscr16^−/− cells

The respiratory chain resides in cristae formed by the mitochondrial inner membrane. Given the dramatic loss of cristae caused by Wbscr16 deficiency in adipose tissues and MEFs, we examined the levels of specific proteins in respiratory chain complexes and found significant decreases in most mtDNA-encoded proteins (Figure 2A). Simultaneously, proteins encoded by the nuclear genome, such as NDUFA4 and NDUFB8 in complex I, and COX4 in complex IV, were also downregulated (Figure 2B). Additionally, mitochondrial 16S rRNA was significantly decreased in iBAT of AWBSCR16 KO mice. 12S rRNA, the other mitochondrial rRNA, was slightly downregulated (Figure 2C). However, mRNA levels of most mtDNA genes in iBAT from AWBSCR16 KO mice remained unchanged except for upregulated mt-Nd1, mt-Nd2 and mt-Nd3 (Figure 2D). Increased mtDNA gene expression (mt-Nd1, mt-Nd2 and mt-Nd3) and ribosomal proteins (MRPS35 and MRPS16) were considered to be compensatory effects due to respiratory chain defects caused by Wbscr16 deficiency (Figure 2D and E). Similar consequences of mRNA levels were observed upon defective ribosome assembly due to the loss of Tfb1m, Mterf3 or Ptcd1 (36–38). To further investigate whether Wbscr16 deficiency affects mitochondrial ribosome assembly, sucrose gradient sedimentation analysis of mitochondrial extracts was performed to separate different mitochondrial ribosome particles. The results revealed a significant reduction of the 55S monosome after Wbscr16 deficiency. Furthermore, MRPL12 protein (39S large subunit) decreased in 39S large subunit (mtLSU) fractions, while it notably increased in the less dense fractions (Figure 2F and G). Additionally, MRPS35 and MRPS16 proteins were elevated in the fractions of the 28S small subunit (mtSSU) (Figure 2H and I). Increased 28S mtSSU fractions were considered to be compensatory consequences of impaired assembly of mtLSU. Similar compensatory responses have been observed in cells with unsuccessfully assembled mtLSU (38,39). These findings imply that Wbscr16 deficiency hinders 39S mtLSU assembly, thereby leading to impaired mitochondrial ribosomal biogenesis and compromised protein translation of mtDNA genes.

Figure 2.

Disruptions of Mitochondrial Ribosome Assembly in Wbscr16 Knockout Adipose Tissues. (A and B) Representative western blots of mtDNA-encoded proteins (MT-ND4, MT-ND6, MT-CO1, MT-CO2 and MT-ATP6) (A), as well as nuclear genome-encoded mitochondrial proteins (NDUFA4, NDUFB8, UQCRC2, SDHA, SDHB, COX4, ATP5A and VDAC1) (B) in iBAT were shown (n = 4). (C) qRT-PCR analysis of Wbscr16 and two mtDNA-encoded rRNAs in iBAT from controls and AWBSCR16 KO mice was shown (n = 4). (D) qRT-PCR analysis of Wbscr16 and mtDNA encoded mRNAs in iBAT from floxed and AWBSCR16 KO mice was shown (n = 4). (E) Representative western blots of WBSCR16, the mtLSU protein MRPL37 and MRPL12, and the mtSSU protein MRPS35 and MRPS16 in iBAT from controls and AWBSCR16 KO mice were shown (n = 4). (F) The distribution of WBSCR16, the mtLSU protein MRPL12, and the mtLSSU protein MRPS35 and MRPS16 in mitochondrial ribosome fractions of iBAT from controls and AWBSCR16 KO mice were shown (n = 4). (G–I) The distribution of the mtLSU protein MRPL12 (G), the mtLSSU protein MRPS35 (H) and MRPS16 (I) in mitochondrial ribosome fractions of iBAT from controls and AWBSCR16 KO mice was quantified using ImageJ software. Data were presented as mean ± SEM; t-test: *P < 0.05, **P < 0.01, ***P < 0.001; NS, not significant.

Requirement of WBSCR16 for 16S rRNA processing

To investigate the mechanism underlying the disruption of mitochondrial ribosome assembly in this study, RNA processing was taken into consideration since 16S rRNA was significantly downregulated in Wbscr16 knockout adipose tissues. The reduction of mature 16S rRNA in iBAT from AWBSCR16 KO mice was further confirmed by a northern blot assay, while mature 12S rRNA levels remained unchanged (Figure 3A). In the initiation region of the mitochondrial large polycistronic precursors (40), 12S rRNA and mt-tRNA^Val are located at the upstream of 16S rRNA, while mt-tRNA^Leu and Nd1 are located at the downstream of 16S rRNA (Figure 3B). Unprocessed transcripts containing regions from 12S rRNA to mt-tRNA^Leu were significantly upregulated in iBAT from AWBSCR16 KO mice, especially the mt-tRNA^Leu-Nd1 transcripts (Figure 3C). Transcriptome-wide RNA sequencing (RNA-seq) confirmed the decrease in mature 16S rRNA and revealed an accumulation of unprocessed RNAs spanning tRNA regions adjacent to 16S rRNA, as well as regions near mt-Nd1, mt-Co3, mt-Nd3 and mt-Nd4l (Figure 3D). The upregulated 16S rRNA-mt-tRNA^Leu transcripts in iBAT from AWBSCR16 KO mice indicated that WBSCR16 is involved in RNA cleavage to regulate 16S rRNA processing at the 3′-end of 16S rRNA, which corresponded to the 5′-end of mt-tRNA^Leu (Figure 3E). We performed RNA sequencing of transcripts binding to FLAG-tagged WBSCR16. The results revealed that rRNA, particularly 16S rRNA, showed a high affinity to WBSCR16 rather than other regions in the large polycistronic precursor (Figure 3F). Additionally, the specificity of the interaction between 16S rRNA and WBSCR16 was validated through RNA pull-down assays (Figure 3G). Thus, mitochondrial alterations in Wbscr16 knockout cells or adipose tissues may be attributed to aberrant 16S rRNA processing.

Figure 3.

Impaired 16S rRNA Processing Due to Lack of WBSCR16 Binding. (A) Northern blot analysis of 16S and 12S rRNA in iBAT from controls and AWBSCR16 KO mice was shown. Tubulin mRNA was detected as a loading control. (B) Schematic illustration of the mtDNA regions containing rRNA (12S and 16S) and mRNA ORFs punctuated by tRNAs was shown. (C) Analysis of mitochondrial unprocessed RNA transcripts containing 16S rRNA in iBAT from controls and AWBSCR16 KO mice using qRT-PCR was shown (n = 8). (D) Changes in RNA-seq coverage (log2 fold change [KO_mean/floxed_mean]) from three control (floxed) and three knockout (KO) mice were shown. Increases are shown in the centrifugal direction, and decreases are shown in the centripetal direction. The mitochondrial genome is displayed, with rRNAs, mRNAs, tRNAs, and the non-coding region (NCR) marked in different patterns (n = 3). (E) The genome browser view displayed the mean RNA-seq coverage, showing the raw sequencing reads of three control (floxed) and three knockout (KO) mice at the top, and the log2 fold change [KO_mean/floxed_mean] at the bottom, highlighting the 5′ cleavage sites of mt-tRNA^Leu in the absence of WBSCR16 (n= 3). (F) Mapping of FLAG-tagged WBSCR16 binding sequences in RIP-Seq was shown (n = 3). (G) Representative western blots of the pull-down products from FLAG-tagged WBSCR16 in MEFs with biotin-labeled DNA probes targeting 12S rRNA, mt-tRNA^Val and 16S rRNA were shown. Data were presented as mean ± SEM; t-test: *P < 0.05, ***P < 0.001.

The essentials of RNA-binding domain (RBD) for WBSCR16 binding to 16S rRNA

To further investigate how WBSCR16 is involved in 16S rRNA processing, we attempted to analyze the RNA-protein interaction from the perspective of WBSCR16 protein structure. WBSCR16 binding to 16S rRNA was further validated by performing RIP of FLAG-tagged WBSCR16 in MEFs, followed by qRT-PCR (Figure 4A). The mouse Rcc1l (Wbscr16) gene consists of 11 exons. It encodes the WBSCR16 protein with a circular seven-bladed β-propeller structure, and each blade is composed of four antiparallel β-strands with loops between each strand (Figure 4B). We previously found that homozygosity for Wbscr16 mutant alleles led to lethality in mouse models (18). We stably overexpressed the ninth exon-deleted Wbscr16 mutant (Wbscr16-Del) in MEFs (Figure 4C and D). Of note, the deleted exon encoded one of the seven bladed β-propeller structures, which is a potential RBD homologous to human WBSCR16 (PDB: 5XGS) (Figure 4B) (41). To determine whether the potential RBD of WBSCR16 is involved in specific binding to 16S rRNA, we performed RIP-qPCR. The mutant WBSCR16 failed to interact with 16S rRNA but retained its 12S RNA-binding activity (Figure 4E and F). This suggests that the structural integrity of WBSCR16, especially the RBD domain, is crucial for 16S rRNA binding, rather than for 12S rRNA.

Figure 4.

Disrupted Binding of WBSCR16 to 16S rRNA in the absence of RBD. (A) RIP of FLAG-tagged WBSCR16 in MEFs was performed using FLAG magnetic beads or IgG magnetic beads, followed by qRT-PCR analysis of the eluted rRNA and mRNA (n = 6). (B) Protein structure of human WBSCR16 (PDB ID: 5XGS) illustrated regions encoded by different exons (pink; Exon 1–8 and 10–11, red; Exon 9). (C) Schematic illustrations of the exons in wild-type Wbscr16 (WT) and its ninth exon-deleted mutant (Del) were shown. (D) Representative western blots of FLAG tag and WBSCR16 in wild-type WBSCR16 (WT) or its ninth exon-deleted mutant (Del) after overexpressing them into MEFs were shown. (E and F) RIP of FLAG-tagged wild-type WBSCR16 (WT) and its ninth exon-deleted mutant (Del) from MEFs by FLAG magnetic beads or IgG magnetic beads. qRT-PCR analysis of eluted 16S rRNA (E) and 12S rRNA (F) was shown. Data were normalized to Gapdh (n = 3). Data were presented as mean ± SEM; t-test: *P < 0.05, ***P < 0.001; NS, not significant.

WBSCR16 interaction with MRPP3 in vitro and in vivo

RNase P and RNase Z are responsible for general RNA processing in mitochondria (11,12). We further investigated whether WBSCR16 facilitates the RNase activity to regulate 16S rRNA processing. Interestingly, the protein level of MRPP3, the catalytic subunit of RNase P, was upregulated in iBAT from AWBSCR16 KO mice (Supplementary Figure S2A–C) but downregulated in Wbscr16-OE cell lines (Supplementary Figure S2D–G). However, ELAC2 (RNase Z) and other non-enzymatic proteins that have been reported to be involved in mitochondrial RNA processing or other genes reported to regulate mtDNA-encoded gene expression associated with WBSCR16 remained nearly unchanged (Supplementary Figure S2B and C, Supplementary Figure S2E and F, and Supplementary Figure S2H and I). We generated transgenic mice that allow inducible expression of wild-type WBSCR16 or its ninth exon-deleted mutant in BAT (Wbscr16^UCP1tg or Wbscr16-Del^UCP1tg) using the Cre-LoxP system. We observed that WBSCR16 interacted with MRPP3 but not ELAC2 or other non-enzymatic proteins using cultured cells or adipose tissues expressing tagged WBSCR16 (Figure 5A and B and Supplementary Figure S3A). Similar results were observed in liver extracts (Supplementary Figure S3B). Additionally, the ninth exon-deleted mutant of WBSCR16 failed to interact with MRPP3 (Figure 5C and D). Together, the structural integrity of WBSCR16 was critical for its interaction with MRPP3 and 16S rRNA to facilitate 16S RNA processing. To further investigate the relationship between WBSCR16 and MRPP3, we purified the full-length of recombinant WBSCR16-GST from E. coli and obtained commercial MRPP3-His proteins (230–583 amino acids of MRPP3) (Supplementary Figure S3C–E). Pull-down assays confirmed a direct association between WBSCR16 and MRPP3 in vitro (Figure 5E and F). SPR analysis was performed to further validate the binding between WBSCR16 and MRPP3 in vitro. WBSCR16-GST fusion protein interacted strongly with recombinant MRPP3-His protein with a dissociation constant of 36.97 nM (Figure 5G). Since RNase P possesses the ability to perform all RNA processing at the 5′-end of tRNA (11,14), these data suggest that WBSCR16 recruits more MRPP3 to nascent 16S rRNA to facilitate the cleavage at the 5′-end of mt-tRNA^Leu and accelerate 16S rRNA processing. To validate our speculation about MRPP3 distributions regulated by WBSCR16, we performed RIP to isolate nascent mitochondrial RNA which binds to MRPP3 under different WBSCR16 expression levels. Overexpression of wild-type Wbscr16 increased the binding of MRPP3 to 16S rRNA but not 12S rRNA (Figure 5H). Under Wbscr16 knockdown or overexpression of its ninth exon-deleted mutant, the binding of MRPP3 to 16S rRNA and 12S rRNA was significantly downregulated (Figure 5H). These findings suggest that WBSCR16 efficiently recruits more MRPP3 to nascent 16S rRNA and specifically enhances 16S rRNA processing.

Figure 5.

Recruitment of MRPP3 by WBSCR16 to Facilitate 16S rRNA Processing. (A) Representative western blots of MRPP3 and ELAC2 in tissue lysate from iBAT with HA-tagged WBSCR16 overexpression immunoprecipitated with anti-HA magnetic beads were shown. (B) Immunoblots showed WBSCR16, FLAG and MRPP3 in cultured cells with FLAG-tagged WBSCR16 and HA-tagged MRPP3 immunoprecipitated with anti-HA magnetic beads. (C) Representative western blots of WBSCR16, FLAG, and MRPP3 in cultured cells with FLAG-tagged wild-type WBSCR16 or its ninth exon-deleted mutant (WBSCR16-Del) overexpression immunoprecipitated with anti-FLAG magnetic beads were shown. (D) Immunoblot analysis of MRPP3, WBSCR16, and HA in tissue lysates from iBAT, with HA-tagged wild-type WBSCR16 or its 9th exon-deleted mutant (WBSCR16-Del) overexpression, immunoprecipitated with anti-HA magnetic beads. (E and F) Interactions between WBSCR16 and MRPP3 were analyzed in vitro with GST- (E) or His- (F) pull-down assays, followed by immunoblotting with the indicated antibodies. (G) SPR analysis of the binding between recombinant WBSCR16-GST and MRPP3-His proteins was shown. The MRPP3-His was immobilized on a CM5 BIAcore chip and interacted with WBSCR16-GST at the indicated concentrations; the K_d value (equilibrium dissociation constant) obtained was shown in the graph. (H) RIP of MRPP3 was performed with anti-HA magnetic beads or anti-IgG magnetic beads in MEFs with MRPP3-HA overexpressing under Wbscr16 knockdown, FLAG-tagged WBSCR16 or its ninth exon-deleted mutant (WBSCR16-Del) overexpression. qRT-PCR analysis of eluted 12S rRNA (left) and 16S rRNA (right) was shown. Data were normalized to Gapdh control (n = 3). Data were presented as mean ± SEM; t-test: ***P < 0.001; NS, not significant.

It’s worth noting that, besides 16S rRNA, 12S rRNA was also slightly downregulated in iBAT from AWBSCR16 KO mice (Figure 2C). A weaker decrease in binding to 12S rRNA by MRPP3 was observed under Wbscr16 knockdown or overexpression of its ninth exon-deleted mutant (Figure 5H). Besides the rRNAs, we observed that the binding of MRPP3 to other mitochondrial RNAs also changed in accordance with WBSCR16 levels, such as mt-Atp8 and mt-Atp6(Supplementary Figure S3F and G). However, their mRNA levels were not significantly altered in iBAT from AWBSCR16 KO mice. Therefore, there may be unknown functions beyond 16S rRNA processing for which WBSCR16 recruits MRPP3 here.

In addition to MRPP3, we investigated whether the other two protein subunits of RNase P and protein factors reported to be associated with WBSCR16 were involved in our mechanisms. We observed no interactions of WBSCR16 with MRPP1 and MRPP2 (Supplementary Figure S3A), which corresponded to the weak affinity among protein subunits of RNase P (11). The human homolog RCC1L^V1 was previously shown to interact with mt-LSU and GTPBP10 to promote mt-LSU subunit assembly in human cells (19). However, GTPBP10, not GTPBP5, interacted with WBSCR16 among the associated proteins we detected (Supplementary Figure S3A) (16,42–44). Moreover, the protein levels of GTPBP10 were not affected in cultured cells with Wbscr16 knockout or overexpression (Supplementary Figure S3H–I). Based on our data and the general mitochondrial central dogma, we proposed that WBSCR16 first facilitated 16S rRNA processing and then modulated mitochondrial ribosome subunit assembly through additional protein interactions with mtLSU and GTPase proteins.

Fatty acid utilization preference in Wbscr16^−/− iBAT

Cells consume fuels including sugars (such as glucose), amino acids and fatty acids to generate energy in the form of ATP, GTP or others (45–48). The nutrients mentioned above are usually metabolized or shuttled into mitochondria to generate energy at certain percentages. However, how mitochondria integrate different fuel metabolism remains largely unknown. Given significant alterations in Wbscr16 KO mitochondria from cultured cells and adipose tissues, we were curious about the physiological functions of WBSCR16 in regulating 16S rRNA processing and mtDNA gene expressions. The ability to efficiently adjust metabolic situations by fuel sensing, trafficking, storage, and utilization, depending on availability and energy requirements, is known as metabolic flexibility. Higher plasma glucose (Figure 6A) but lower NEFA levels (Figure 6B) were observed in AWBSCR16 KO mice. Thus, we suspected that WBSCR16 might affect the metabolic flexibility of adipose tissues through mitochondrial alterations. We profiled the transcriptome of iBAT from AWBSCR16 KO mice and revealed significant changes in lipid metabolism activity (Supplementary Figure S4A). Notably, genes associated with fatty acid oxidation, including Acot2, Scd3, Gpat3 and Mogat, were found to be the most differentially expressed metabolic genes (DEGs) in iBAT from AWBSCR16 KO mice (Figure 6C). These data, validated by subsequent qRT-PCR analysis (Supplementary Figure S4B), indicated that the ablation of Wbscr16 enhanced fatty acid oxidation in BAT. Metabolome data of iBAT-stromal vascular fraction (SVF) differentiated mature adipocytes indicated that glycolysis and the citrate cycle associated with glucose metabolism were severely impaired (Supplementary Figure S4C). Collectively, these findings suggest that mitochondrial defects in the iBAT of AWBSCR16 KO mice lead to adaptations of metabolic flexibility that reduce glucose utilization while favoring more fatty acids as mitochondrial fuel.

Figure 6.

Increased fatty acid utilization in Wbscr16 knockout adipose tissues to prevent obesity. (A and B) Levels of fed and fasted glucose (A) and plasma NEFA (B) in floxed and AWBSCR16 KO mice were shown (n = 7–10). (C) A Heatmap displaying mRNA levels of upregulated or downregulated genes in iBAT from floxed and AWBSCR16 KO mice was shown (n = 3). (D) Glucose utilization during 24 and 48 h in mature adipocytes derived from floxed iBAT-SVF transduced with or without Cre lentivirus was shown. Values were normalized by total protein (n = 3–4). (E and F) Representative staining (E) and quantitative analysis (F) of the BODIPY C12 accumulated in LDs of mature adipocytes derived from iBAT-SVF transduced with or without Cre lentivirus were shown (n = 4–5). (G) Quantitative analysis of palmitic acid utilization during 24 h in mature adipocytes derived from iBAT-SVF infected with or without Cre lentivirus was shown (n = 3). (H and I) Representative PET-CT images (H) in coronal, sagittal and transaxial views and the ratio of SUV Max (I) (n = 6) of iBAT (arrows) from floxed and AWBSCR16 KO mice were shown. (J) Representative western blots of CD36, p-HSL^Ser660, total-HSL and CPT1α in iBAT from floxed and AWBSCR16 KO mice were shown. (K and L) Growth curves (K) (n = 6, two-way analysis of variance) and representative images (L) of HFD-treated floxed and AWBSCR16 KO mice were shown. (M–O) Representative images and HE staining of iBAT (M), iWAT (N) and eWAT (O) from mice after HFD treatment were shown. (P) Representative images, HE staining, and oil red O staining of the liver from HFD-treated mice were shown. (Q) Relative tissue weights normalized to body weight in chow diet (CD) or HFD fed Floxed and AWBSCR16 KO mice were shown (n = 4–11). (R and S) Levels of fed and fasted glucose (R), and fasted plasma NEFA (S) in HFD-treated floxed and AWBSCR16 KO mice were shown (n = 4–5). (T) Representative western blots of WBSCR16, MT-CO2 and PPARγ in iBAT from CD or HFD mice were shown. (U) Analysis of human WBSCR16 mRNA levels in adipose tissues from obese and non-obese subjects (BMI 16.7–50.2) with normal or impaired glucose tolerance (GSE27951) was shown (n = 11–13). Data were presented as mean ± SEM; t-test: *P < 0.05; **P < 0.01; ***P < 0.001.

To further validate this finding, glucose utilization in primary brown adipocytes and MEFs was detected. As expected, glucose utilization decreased significantly in Wbscr16 KO cells (Figure 6D and Supplementary Figure S4D). However, the ATP levels in Wbscr16 KO cells remained well-maintained, possibly due to compensatory fatty acid oxidation (Supplementary Figure S4E and F). To further detect the differences in fatty acid uptake after Wbscr16 deficiency, mature primary brown adipocytes were incubated with BODIPY C12 (Red C12) labeled fatty acids. Quantitative analysis showed that Wbscr16 KO cells took up more fatty acids than control cells (Figure 6E and F). Moreover, when cultured with exogenous palmitic acid in the medium, Wbscr16 KO cells were found to consume more palmitic acid than controls (Figure 6G). LDs were larger in Wbscr16 KO mature primary brown adipocytes, corresponding to enhanced fatty acid uptake (Supplementary Figure S4G and H). In contrast, LDs in mature primary white adipocytes showed no significant changes (Supplementary Figure S4I and L). To further investigate the effect of adipose-specific Wbscr16 ablation on glucose uptake in vivo, a 2-deoxy-2-[fluorine-18] fluoro-D-glucose (¹⁸F-FDG) PET/CT scan was performed on both AWBSCR16 KO and control mice. Reduced ¹⁸F-FDG uptake was observed in the iBAT of AWBSCR16 KO mice, indicating downregulated glucose uptake here (Figure 6H and I). These data are consistent with previous studies showing that fatty acid oxidation increases when glucose metabolism is compromised by mitochondrial dysfunctions (49,50).

To provide enough fatty acids for BAT in AWBSCR16 KO mice, triglycerides stored in LDs and food intake from external sources were the main energy sources. Since we observed unchanged food intake (Supplementary Figure S4M) but reduced fat mass in AWBSCR16 KO mice (Figure 1G), we suspected that the increased fatty acids were possibly derived from triglycerides in LDs from iBAT, WAT or other tissues. Significantly increased CD36, p-HSL^Ser660 and CPT1α protein levels in iBAT from AWBSCR16 KO mice supported our hypothesis (Figure 6J). The adapted metabolic flexibility associated with fuel metabolism in AWBSCR16 KO mice potentially contributed to higher plasma glucose levels and lower NEFA levels (Figure 6A and B). Under these physiological conditions, AWBSCR16 KO mice exhibited improved insulin sensitivity (Supplementary Figure S4N) despite having slightly higher plasma glucose levels and impaired glucose tolerance compared to floxed controls (Figure 6A and Supplementary Figure S4O).

Prevention of diet-induced obesity in AWBSCR16 KO mice

To further determine the metabolic effects of Wbscr16 ablation in BAT with adapted metabolic flexibility in vivo, control and AWBSCR16 KO mice were subjected to a HFD for 3 months. After HFD treatment, AWBSCR16 KO mice gained significantly less weight than the controls on HFD (Figure 6K). At the end of HFD treatment, AWBSCR16 KO mice had ∼40% less body weight and significantly reduced WAT weight than the controls (Figure 6L and Q). The body weight of AWBSCR16 KO mice treated with HFD remained similar to that of CD-treated control mice (Supplementary Figure S5A and B). Although we observed slightly increased tissue weights and sizes of LDs in iBAT from AWBSCR16 KO mice after HFD treatment, the tissue structure was comparable to the iBAT from CD-treated AWBSCR16 KO mice (Figure 6M). Importantly, HFD-treated AWBSCR16 KO mice had dramatically smaller iWAT and eWAT, which were only 20–25% of those tissues from the control mice (Figure 6N, O and Q). Since LDs in the WAT of AWBSCR16 KO mice remained normal, there was not enough evidence to define adipocyte atrophy in this case (Figure 6N and O and Supplementary Figure S1B and C). Lower triglyceride storage in WAT corresponded to higher fatty acid utilization in AWBSCR16 KO mice. Although there were no differences in plasma insulin levels and glucose sensitivity under HFD (Supplementary Figure S5C and D), fewer LDs were found in the livers of AWBSCR16 KO mice (Figure 6P). Notably, AWBSCR16 KO mice still exhibited slightly higher plasma glucose levels but lower plasma NEFA levels under HFD treatment (Figure 6R and S), which was similar to the CD-treated conditions. The PPARγ protein levels in iBAT reflected that fatty acid metabolism was also enhanced in HFD-treated AWBSCR16 KO mice (Figure 6T). These results indicated that mtDNA gene expression mediated by WBSCR16 was important for providing different mitochondrial fuels to adapt to energy requirements. To investigate the potential correlation between WBSCR16 levels in adipose tissues and obesity development in humans, we analysed expression profiles of human subcutaneous adipose tissue (sWAT) obtained from 20 individuals with a range of body mass indices (51). Interestingly, the expression of WBSCR16 in sWAT of obese individuals was higher than that of non-obese controls (Figure 6U), highlighting the physiological roles of WBSCR16 in modulating energy metabolism.

Promotion of 16S rRNA processing by WBSCR16 overexpression

To elucidate the effects of WBSCR16 on 16S rRNA processing, we generated transgenic mice with inducible WBSCR16 expression in BAT (Wbscr16^UCP1tg) using the Cre-LoxP system (Wbscr16^tg/+; UCP1^ERT-Cre+/–). After tamoxifen induction, the protein level of WBSCR16 was increased significantly in iBAT (Figure 7A). Both 12S rRNA and 16S rRNA levels increased, along with markedly decreased unprocessed precursors in Wbscr16^UCP1tg iBAT (Figure 7B–D). As expected, the mitochondrial morphology from Wbscr16^UCP1tg iBAT was also noticeably different from that of controls, including significantly increased tubular cristae, smaller mitochondria and slightly more mitochondrial numbers (Figure 7E–G). These alterations of mitochondrial morphology and RNA processing were contrary to those observed in Wbscr16 knockout iBAT, further indicating that WBSCR16 played a significant role in mitochondria via 16S rRNA processing. In Wbscr16^UCP1tg iBAT, the tissue weight and size of LDs were both decreased (Figure 7H–J), while the overall body weight, WAT weight, and fat content remained unchanged (Supplementary Figure S6A–F). Increased glucose utilization was detected in WBSCR16-overexpressing MEFs (Figure 7K). In contrast, the Wbscr16^UCP1tg mice showed lower plasma glucose levels, improved glucose sensitivity (Figure 7L–N), and impaired insulin sensitivity (Supplementary Figure S6G and H). Moreover, Wbscr16^UCP1tg mice gained more body weight and fat mass after HFD treatment without hepatic lipid accumulation (Supplementary Figure S6I and L). Most phenotypes of mitochondria and energy metabolism contrasted with AWBSCR16 KO mice. These results also suggested that WBSCR16 regulated 16S rRNA processing to provide different mitochondrial fuels which benefited metabolic flexibility.

Figure 7.

Restoration of impaired 16S rRNA processing with WBSCR16. (A) Representative western blots of WBSCR16 in iBAT from control and Wbscr16^UCP1tg mice were shown. (B) qRT-PCR analysis of 12S and 16S rRNAs in iBAT from control and Wbscr16^UCP1tg mice was shown (n = 4). (C) Northern blot analysis of 16S and 12S rRNAs in iBAT was shown. The tubulin mRNA was detected as a loading control. (D) Mitochondrial unprocessed RNA transcripts involved in 16S rRNA were analysed by qRT-PCR (n = 6–18). (E–G) Electron microscopic images of iBAT (E), along with quantification of mitochondrial number (F) (n = 5–7) and size (G) (n = 82) of iBAT from control and Wbscr16^UCP1tg mice were shown. (H–J) Representative images and HE staining of iBAT from control and Wbscr16^UCP1tg mice were shown (H). Relative weight normalized to body weight (n = 6–8) (I) and cross-sectional area of LDs of iBAT (n = 256) (J) were shown. (K) Glucose utilization in Wbscr16-overexpressing MEFs was shown (n = 4–5). (L–N) Daily resting glucose levels (L), glucose tolerance test (M) and AUC analysis (N) of control and Wbscr16^UCP1tg mice were shown (n = 8). (O) Representative western blots of WBSCR16 and HSP90 in iBAT were shown. (P) qRT-PCR analysis of 12S and 16S rRNAs in iBAT from floxed, AWBSCR16 KO and AWBSCR16 KO + KI mice was shown (n = 6). (Q) Mitochondrial unprocessed RNA transcripts related to 16S rRNA were analysed by qRT-PCR in iBAT from floxed, AWBSCR16 KO and AWBSCR16 KO + KI mice (n = 6). (R and S) Representative images and HE staining of iBAT (R) and cross-sectional area of LDs (S) of iBAT, iWAT and eWAT from floxed, AWBSCR16 KO and AWBSCR16 KO + KI mice were shown (n = 60–95). (T) A schematic diagram of the working model for mitochondrial metabolic substrate preference regulated by WBSCR16 was shown. Data were presented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001; NS, not significant.

Restoration of impaired 16S rRNA processing by exogenous WBSCR16

To delineate the contribution of WBSCR16 to 16S rRNA processing, transgenic mice expressing exogenous WBSCR16 were crossed to restore the expression of WBSCR16 in AWBSCR16 KO mice (Wbscr16^flox/flox: Wbscr16^tg/+: AdipoCre^+/−) (Figure 7O). Re-introduction of WBSCR16 successfully restored 16S rRNA levels and decreased the aberrantly accumulated precursors caused by Wbscr16 ablation (Figure 7P and Q). The mRNA levels of other mtDNA genes were minimally altered (Supplementary Figure S7A). Remarkably, re-expression of WBSCR16 also restored the sizes of LDs in iBAT and iWAT (Figure 7R and S, Supplementary Figure S7B and C). These results indicated that WBSCR16 regulated 16S rRNA processing to modulate mitochondrial fuels, which benefits systemic metabolic flexibility.

In summary, WBSCR16 is a 16S rRNA-binding protein that maintains mtDNA gene expression. WBSCR16 recruits MRPP3, the catalytic subunit of RNase P, to promote the 5′-end cleavage of the 16S rRNA precursor. WBSCR16 facilitates 16S rRNA processing to assist mitochondrial ribosome assembly. The mitochondrial alterations induced by Wbscr16 deficiency resulted in increased fatty acid utilization. These alterations provide opportunities to supply different mitochondrial fuels when glucose metabolism is impaired. The adaptive mitochondrial fuels in Wbscr16-deficient mice lead to more energy expenditure and even protect the mice from diet-induced obesity. In contrast, exogenous WBSCR16 promotes 16S rRNA processing and improves glucose metabolism (Figure 7T). Thus, WBSCR16 may be employed as a potent pharmacological target to combat metabolic diseases such as obesity and diabetes.

Discussion

In this study, we report that WBSCR16 binds to 16S rRNA and recruits the MRPP3 subunit of RNase P to cleave the 16S rRNA-mt-tRNA^Leu. Impaired 16S rRNA processing results in a deficiency in 39S mtLSU assembly and mitochondrial protein synthesis. Consequently, the mitochondrial alterations caused by Wbscr16 ablation in BATs lead to decreased glucose uptake and catabolism but increased fatty acid utilization as mitochondrial fuels. In contrast, overexpression of WBSCR16 enhances 16S rRNA processing and promotes glucose utilization in both cultured cells and transgenic mice. Moreover, overexpression of exogenous FLAG-tagged wild-type WBSCR16 in AWBSCR16 KO mice recovers the 16S rRNA levels and downregulates the unprocessed transcripts. These findings highlight the crucial role of WBSCR16 in promoting mitochondrial 16S rRNA processing, which benefits metabolic flexibility associated with mitochondrial fuels and whole-body energy homeostasis.

The ablation of Wbscr16 leads to dramatically decreased 16S rRNA, subsequently inhibiting the assembly of mitochondrial ribosomal large subunits. Similar changes have been observed in cells deficient in the RBP PTCD1 or FASTKD2 (38,52); however, the precise mechanisms of 16S rRNA processing remain unclear. Our findings demonstrate that WBSCR16 binds to 16S rRNA, directly recruits MRPP3 to the 16S rRNA-mt-tRNA^Leu site and enhances the cleavage at the 3′-end of 16S rRNA. Human RCC1L isoforms, the homolog of mouse WBSCR16, interact with various GTPases to promote mitochondrial ribosome assembly (19). RCC1L^V1 has been shown to specifically interact with mt-LSU and GTPBP10 to aid mt-LSU assembly (19). The expression of GTPBP10 has also been reported to change in association with WBSCR16 (19). Our results in mice, however, show no interaction between mtLSU proteins and WBSCR16 (Supplementary Figure S3A). Although we observed that GTPBP10 interacted with WBSCR16, Wbscr16 knockout or overexpression did not significantly alter GTPBP10 expression (Supplementary Figure S3G and H). Given that WBSCR16 bound to 16S rRNA and interacted with MRPP3, we concluded that WBSCR16 affected mitochondrial assembly through 16S rRNA processing rather than vice versa, at least in mice.

Recent studies demonstrated that RCC1L formed a complex with NME6 and potentiated its nucleoside diphosphate kinase activity. This process was crucial for maintaining local pyrimidine triphosphate levels necessary for mitochondrial RNA abundance (22). Deletion of NME6 disrupted mitochondrial pyrimidine homeostasis, resulting in reduced levels of most mtRNAs but not mt-Co1, 12S rRNA and 16S rRNA (22). A comparison of the phenotypes between Wbscr16 and NME6 knockouts revealed significant differences. Wbscr16 deficiency decreased 12S rRNA, 16S rRNA and MT-CO1 protein, while the expressions of these genes were elevated under NME6 knockout (22). Disruptions in the mitochondrial ribosomal proteins were not consistent in previous studies (22,53). These distinct phenotypes following Wbscr16 or NME6 knockout suggested that WBSCR16 may regulate mitochondrial ribosome assembly independent of NME6’s NPDK function.

Previous literature indicated that 5S rRNA was lacking in mammalian mitochondrial ribosomes, and its important role was replaced by mt-tRNA (54,55). A further study found that different mammals had different mt-tRNAs to replace 5S rRNA. mt-tRNA^Val predominated in the human mt-LSU while mt-tRNA^Phe was chosen for integration into the porcine mt-LSU (56). However, the reduction in the steady-state level of mt-tRNA^Val did not lead to abnormal assembly of human mitochondrial ribosomes as mt-tRNA^Phe was incorporated instead of mt-tRNA^Val (56). In our study, the RNA pull-down experiments showed that mt-tRNA^Val could not be precipitated by WBSCR16 and the level of mt-tRNA^Val increased in iBAT from AWBSCR16 KO mice compared to floxed controls. These results suggested that abnormal ribosome assembly caused by Wbscr16 deficiency could not be simply attributed to mt-tRNA^Val.

BAT was enriched with mitochondria and played a key role in energy homeostasis. In this study, we established Wbscr16 adipose tissue-specific knockout/transgenic mouse strains and discovered that Wbscr16 regulated 16S rRNA processing in vivo. This regulation leads to alterations in the assembly of ribosomal large subunits and then mtDNA gene expressions. The alterations consequently affected mitochondria to integrate different fuels to satisfy energy requirements. The increased fatty acid utilization but impaired glucose metabolism in Wbscr16 knockout BATs promoted energy expenditure in mice. The ablation of Wbscr16 in adipose tissues not only reduced lipid storage but also enhanced insulin sensitivity, potentially ameliorating diabetes caused by insulin resistance, even under HFD conditions. Conversely, increased expression of WBSCR16 led to pronounced glucose utilization, improving glucose sensitivity and reducing blood glucose levels. These findings were consistent with previous studies indicating that decreased glucose uptake and increased fatty acid oxidation were early events after mitochondrial damage when metabolic syndromes occurred such as obesity and Type 2 diabetes (49,50).

In conclusion, our results suggested that WBSCR16 enhanced cleavage at the 3′-end of 16S rRNA through simultaneous 16S rRNA binding and MRPP3 recruitment, and then promoted 16S rRNA processing. As a result, mitochondrial ribosome assembly and mtDNA gene expression were accelerated by WBSCR16. The alterations forced mitochondria to use different fuels to satisfy energy requirements. The whole-body energy homeostasis was affected by these fuel changes under normal or HFD situations. Further investigations of mechanisms involved in WBSCR16-mediated mitochondrial functions and metabolic reprogramming may uncover potential strategies for the treatment of metabolic syndromes.

Data availability

The data underlying this article are available in GEO under accession number GSE229693 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE229693).

Supplementary data

Supplementary Data are available at NAR Online.

Funding

National Natural Science Foundation of China [81770845, 91854124 to G.H.; 82088102 to G.N.; 32100923 to S.Z.; 82070880 to Z.Z.]; Shanghai Science and Technology Foundation [18140902800 to G.H.]. Funding for open access charge: The fund from our institute.

Conflict of interest statement. The authors declare no competing interests.