Zhang et al., 2020 - Mitochondrial peptide BRAWNIN is essential for vertebrate respiratory complex III assembly. Nature communications   11:1312 Full text @ Nat. Commun.

Fig. 1 Mito-SEP prediction pipeline identifies uncharacterized endogenous mitochondrial SEPs.

a Size distribution of curated human proteins annotated as mitochondria (1172 proteins), cytosolic (5806 proteins), or secretory (3351 proteins) on Uniprot. p-value by two-sided Mann–Whitney U-test. n.s. = not significant. b Size distribution of Uniprot proteins (All) (20417 proteins) annotated to reside in mitochondria (1172 proteins) or inner mitochondrial membrane (IMM) (292 proteins). p-value by two-sided Mann–Whitney U-test. c Percentage of proteins smaller than 100 a.a. in different cellular localizations. d Human SEP selection and mitochondrial prediction workflow. Blue numbers indicate number of SEPs and green numbers indicate number of genes encoding these SEPs. e HA immunofluorescence (green) of mito-SEP candidates showing colocalization with Mitotracker Red (red) in HeLa cells. Uncharacterized SEPs are labeled as mitochondrial SEP with unknown function (MSUF). Scale = 20 μM. f Circular plot of expression correlation between MSUF and MSigDB gene ontology (GO) gene sets that relate to oxidative phosphorylation. The width of each connector represents the normalized enrichment score (NES) value of the gene set enrichment analysis (GSEA) analysis performed on a ranked list of MSUF-coexpressed genes determined by weighted gene co-expression network analysis (WGCNA). The black bar indicates the score of the 90th percentile of the score of 1000 randomly selected genes. CDC42SE2 depicts the position of a non-mitochondrial negative control with NES = 0.

Fig. 2 BRAWNIN (BR) is a conserved SEP at the inner mitochondrial membrane.

a Ribosome protected fragment (RPF) reads from the BRAWNIN gene (C12orf73) in human cell lines. P1 and P2 refer to the 2 potential ORF isoforms generated by alternative splicing. b BR (P1) peptide is conserved throughout vertebrate evolution and possesses an N-terminal hydrophobic region that is predicted to be a transmembrane domain or signal peptide. Underlined regions are peptides detected by mass spec. Regions used as immunogens for raising the polyclonal antibodies used in this study are indicated. c Endogenous BR is detected in HeLa and HEK293T with an α-BR antibody. OE = overexpression. d Western blot of HEK293T lysates transfected with control or BR-targeting siRNAs. e Immunofluorescence of endogenous BR detected by α-BR in HeLa transfected with control or BR-targeting siRNAs. Scale = 5 μM. f Western blot of indicated proteins from subcellular fractions of HEK293T. W = whole cell, M = mitochondria enriched fraction, C = cytosol, P = nuclear fraction and debris, including un-extracted mitochondria; BR∆N25 = synthetic BR peptide lacking the N-terminal 25 a.a. g Endogenous BR detected by α-BR in HeLa costained with mitotracker. Scale = 5 μM. h Endogenous BR detected by α-BR in HEK293T costained with mitotracker. Scale = 5 μM. i α-BR and α-citrate synthase (CS) immunofluorescence in mouse skeletal muscle, longitudinal section. Scale = 20 μM. j α-BR immunofluorescence and GFP epifluorescence of Mito-Dendra2 in mouse skeletal muscle, longitudinal section. Scale = 20 μM. k Extraction of endogenous BR from HEK293T mitochondria with buffer alone, Na2CO3 solution or buffer containing 1% Triton X. S = supernatant; P = pellet. l Extraction of endogenous BR from HEK293T mitochondria using increasing concentrations of digitonin. TOMM20, TOMM70, and VDAC1 are outer mitochondrial membrane (OMM) proteins, are MTCO1, ATP5A, and TIM23 are IMM associated. m Proteinase K protection assay of HEK293T mitochondria. The α-BR epitope is C-terminal to the predicted transmembrane domain. TOMM20 and TOMM70 have cytosolic domains. TIM23 has IMS domains. Both CS and ATP5A are in the matrix.

Fig. 3 BR is an AMPK target that potentiates OXPHOS in human U87MG cells.

a Heatmap depicting GSEA NES scores of mitochondrial pathways (Supplementary Data 2) in the ranked list of pairwise correlations between genes in the human liver, heart, or skeletal muscle (SKM) with either BR gene, known electron transport gene ATP5H or CDC42SE2, a representative non-mitochondrial candidate SEP. b SDS-PAGE of HEK293T treated with AMPK activator AICAR (500 µM) for the indicated time periods. c SDS-PAGE of HEK293T treated with indicated starvation regimes for the indicated time periods. Etomoxir was used at 3 µM. d Mitostress test (Agilent Seahorse) analysis of control versus stable BR knockdown U87MG cells. Oxygen consumption rate (OCR) reads were normalized with citric synthase (CS) activity. Data are mean and SEM of six technical replicates. Experiment is representative of two biological replicates. Oligo = oligomycin, FCCP = carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone, Rot/AA = rotenone/antimycin A. e Analogous to d. except using short-term (72 h) siRNA-mediated depletion of BR. The experiment is representative of three biological replicates each with six technical replicates. Error bars represent SEM. f Spare respiratory capacity calculated from experiments in d and e. Data are mean and SEM of six technical replicates which are representative of two and three biological replicates for shRNA and siRNA-mediated depletion, respectively. p-values from two-sided unpaired t-test.

Fig. 4 Knockout of <italic>Br</italic> in zebrafish causes lethal mitochondrial deficiency.

a In-situ hybridization (ISH) of brawnin (br) in zebrafish larvae. dpf = days post fertilization. Left panel: scale = 100 µm, right panel: scale 200 µm. b Western blot analysis of purified mitochondria from adult WT and KO (homozygous br/br) skeletal muscle using zebrafish-specific α-Br. c Mendelian percentages of offspring from heterozygous (Het) intercrosses genotyped at the indicated dpf. Data represent the mean of at least 2 clutches per time point (except 28 and 124 dpf) with at least 15 animals per clutch. Error bars indicate SEM. d Standard length (mm) measurements of WT, Het, and zygotic KO (zKO) larvae generated from an F2 intercross genotyped and segregated at 3 dpf. Maternal zygotic KOs (mzKOs) were generated from a homozygous incross and measured until 42 dpf when all animals succumbed to disease. Data are representative of four independent experiments from two independent Cas9/gRNA injected br KO allele founders. Number of animals at the beginning of the experiment: WT = 14; HET = 37; zKO = 8; mzKO = 22. Error bars indicate SEM. e Representative images of WT, Het, and zKO clutch-mates at 65 dpf. Scale = 5 mM. f Basal motility monitoring continuous swim tracking of WT and mzKO larvae at 11 dpf. 23 animals per genotype monitored. Error bars indicate SEM. g Coupled respiratory analysis of isolated SKM mitochondria measured on Agilent Seahorse in the presence of succinate + rotenone, read out as CS-normalized OCR. Basal, ADP stimulated, maximal respiration and ATP production were plotted. Data represent mean and SEM of four biological replicates. p-values from two-sided paired t-test. h Basal respiratory rate of free moving 1 dpf WT and mzKO larvae measured on Agilent Seahorse platform. Number of animals: WT1 = 30; WT2 = 15; KO1 = 21; KO2 = 23. Data represent mean and SEM, p-values from two-sided unpaired t-test.

Fig. 5 Br functionally and physically interacts with electron transport chain complex III.

a Lactate concentrations in 5 dpf WT and mzKO larvae. Data represent mean and SEM of four independent clutches each with 50 larvae per genotype. p-values from two-sided paired t-test. b Citric acid cycle organic acid concentrations in 5 dpf WT and mzKO larvae. Data represent mean and SEM of four independent clutches each with 50 larvae per genotype. p-values from two-sided paired t-test with Bonferroni correction. *α-KG concentrations are a 10-fold multiple to enable visualization on the same scale. c Distributions of the export rate of lactate and succinate (5000 flux samples) comparing the unconstrained MitoCore model versus the CIII-blocked model derived by MitoCore flux balance analysis (FBA). d Venn diagram depicting intersections of proteins identified in 4 co-IP/mass spec experiments of BR in HEK293T mitochondria. Endogenous BR and overexpressed untagged BR were immunoprecipitated with α-BR, overexpressed BR-FLAG with α-FLAG. e Immunoprecipitation of BR-FLAG followed by immunoblotting for indicated ETC proteins from mouse mitochondria with AAV9-mediated expression of BR-FLAG. MTCYB, UQCRFS1, UQCRC1, and UQCRH are components of CIII. NDUFA9 and COX4 are components of CI and CIV, respectively. f. Reciprocal IP of BR with CIII components. UQCRQ-FLAG and BR-HA were co-transfected into HEK293T. UQCRQ-FLAG immunoprecipitation was followed by immunoblotting for BR-HA and indicated proteins. CS is an unrelated matrix protein.

Fig. 6 Br is required for complex III assembly and activity.

a Quantitative proteomic analysis of WT and br zKO isolated adult skeletal muscle (SKM) mitochondria. Mean values of log2 fold change (zKO/WT) for each detected protein from three biological replicates are plotted in a volcano plot. Subunits of each of the ETC complexes are labeled in the indicated colors. CI–CV = Complex I–Complex IV. p-values from two-sided unpaired t-test. b The median log2 fold change (zKO/WT) of all detected ETC proteins are further depicted in a dendrogram heatmap and separated by complex membership. AF = assembly factors. c The median log2 fold change (zKO/WT) of CI–CIV proteins were superimposed onto the structures of human CI–CIV to visually depict the observed changes. The same scalebar from panel 6b applies. Undetected proteins are shown in gray. d SDS-PAGE of purified WT and zKO SKM mitochondria followed by immunoblotting for the indicated ETC proteins. Image is representative of two experiments. e Quantitation of SDS-PAGE in 6d. Data represent mean and SEM. n = 5, except for Uqcrc2 where n = 3. p-values from two-sided unpaired t-test. f BN-PAGE and western blot analysis of WT and zKO SKM mitochondria. Position of free III2 (red) and III2 + IVn (black) supercomplexes are indicated by arrowheads. g Spectrophotometric respiratory chain activity assay (RCA) of WT and zKO SKM mitochondria. Data represent mean and SEM. Each dot represents one animal, n = 12 for WT and zKO animals. p-values from two-sided unpaired t-test.

Fig. S4 ZFIN is incorporating published figure images and captions as part of an ongoing project. Figures from some publications have not yet been curated, or are not available for display because of copyright restrictions.

Acknowledgments:
ZFIN wishes to thank the journal Nature communications for permission to reproduce figures from this article. Please note that this material may be protected by copyright. Full text @ Nat. Commun.