Small mitochondrial protein NERCLIN regulates cardiolipin homeostasis and mitochondrial ultrastructure

Significance Recent studies have shown that the human mitochondrial proteome is enriched for small proteins with vital cellular functions. We describe here a primate-specific small mitochondrial protein NERCLIN as a negative regulator of cardiolipin homeostasis and mitochondrial ultrastructure. NERCLIN has a protective role in heat stress, contributing to the stress-induced adaption of mitochondrial dynamics. Our findings add NERCLIN to the group of small mitochondrial proteins with important regulatory functions.


In silico analysis of GRPEL2 sequences in primates
The mRNA sequences were extracted from UCSC Genome Browser (4). The sequences corresponding to exon 3 were removed and the obtained sequences lacking exon 3 were translated to amino acid sequence. The amino acid sequences of GRPEL2 gene skipping exon 3 were aligned using ClustalW algorithm.

DNA and RNA analysis
For expression analysis human multiple tissue cDNA panel (MTC™ Clontech, 636742) and human normal brain tissue qPCR array (OriGene Technologies, HBRT101) were used. Prior to their use these samples were de-identified by manufacturers. 1-3 ng of cDNA was used for realtime PCR assay. After completion of PCR amplification of the samples from a human multiple tissue cDNA panel, the PCR products were loaded on 0.7% agarose gel for electrophoresis. Isoform expression of GRPEL2 was analyzed using the data in the Genotype-Tissue Expression (GTEx) Project (https://gtexportal.org/home/).

Real-time PCR assay
Total RNA was extracted from cultured cells using Mini spin kit (Macherey-Nagel). RNA extracted from canine fibroblasts was obtained from Prof. Hannes Lohi's group. 1000 ng of RNA was reverse transcribed using Maxima First Strand cDNA Synthesis Kit for RT-qPCR (Thermo Fisher Scientific, K1641). Real-time PCR analysis was done by DyNAmo Flash SYBR Green qPCR Kit (Thermo Fisher Scientific, F-415L) using CFX96 TM Real-Time PCR Detection System (Bio-Rad). The PCR program started with 95°C for 7 min followed by 40 cycles at 95°C for 10 s and 60°C for 30 s. Beta-2-Microglobulin (B2M) was used as a reference gene for normalization and the mRNA expression level was calculated using the comparative Ct (threshold cycle) method. All primer sequences are in Dataset S5.

Cell culture
Human embryonic kidney cells (HEK293), human osteosarcoma cells (143B), human fibroblasts, mouse embryonic fibroblasts (MEFs) or African green monkey kidney fibroblasts (COS7) were cultured (37°C, 5% CO2) in DMEM (Lonza, BE12-614F), supplemented with 10% fetal bovine serum (Life Technologies, 10270106), l-glutamine (Gibco, 25030024) and penicillin/streptomycin (Gibco, 15140122). Previously characterized(5) human induced pluripotent stem cells (iPS) were maintained on Matrigel-coated (Corning, 356231) plates with E8-medium supplemented with E8supplement (Gibco, A1517001) and 50 µg/mL uridine (Sigma, U3003). iPS cells were passaged with 0.5 mM EDTA (Invitrogen, 15575020) in PBS (Life Technologies, BE17-516F) when 70% confluent. iPS cells at passage number 33 were used in the experiments. To induce heat stress, HEK293 cells were incubated at 42°C or 45°C for indicated times in a cell incubator in normoxia and 5% CO2. Control cells were kept at 37°C. After the treatment, representative images were taken using a bright field microscope, and the samples were processed immediately. To induce oxidative stress HEK293 cells were treated with 250 μM or 500 μM H2O2 (Sigma, H1009) for the indicated times. After the treatment, the cells were scraped, pelleted at 1000 g for 10 min and washed once with cold PBS (Life Technologies, BE17-516F). To inhibit mitophagy we used chloroquine (Sigma, C6628). JetPRIME reagent (Polyplus, 114-15) was used for cell transfection. After transfection cells were cultured for 24 or 48 h. This study involves human fibroblasts and iPS cells derived from healthy donors after providing written consent for the use of patient material. Human patient stem cell research was approved by the Coordinating Ethics Committee of the Helsinki and Uusimaa Hospital District (Nr 95/13/03/00/15).

Generation of stably expressing cell lines
To generate stably overexpressing cell lines retroviral constructs were transiently transfected into phoenix package cell line with jetPRIME transfection reagent (Polyplus, 114-15) and HEK293 cells were infected by being exposed to virus-containing medium in the presence of polybrene (Millipore, TR-1003). Infected cells were treated with 1.5 µg/ml puromycin (Sigma, P8833) for five days until resistant clones formed. The expression of the constructs was confirmed by immunoblotting or/and by qPCR analysis.

Mitochondria isolation
Cultured HEK293 cells were scraped and harvested at 1000 g for 10 min at 4 °C. Cell pellets were washed with PBS (Life Technologies, BE17-516F), resuspended in ice-cold mitochondria isolation buffer (200 mM mannitol, 70 mM sucrose, 10 mM HEPES, 1mM EGTA, 0.2 % delipidated BSA (Biowest, P6154), protease inhibitor cocktail (Thermo Scientific, 78444), pH 7.5) and homogenized with 60 strokes using Dounce glass-glass homogenizer. Nuclei and unbroken cells were removed by centrifugation at 600 g for 20 min. The supernatant was centrifuged at 10 000 g for 10 min to pellet mitochondria. Mitochondrial pellets were resuspended in mitochondria isolation buffer without BSA and pelleted at 10 000 g for 10 min.

Plasmid construction
To clone the NERCLIN cDNA we extracted total RNA from 143B cells using Mini spin kit (Macherey-Nagel, 740955.50). Then RNA was reverse transcribed using Maxima First Strand cDNA Synthesis Kit for RT-qPCR (Thermo Scientific, K1641). The cDNA was PCR amplified using specific primers for TOPO cloning. Following gel electrophoresis, the correct band was cut, and the DNA was extracted from the gel for TOPO cloning. The TOPO construct containing NERCLIN was used as a template for cloning NERCLIN into EGFP or pBabe vectors. Briefly, NERCLIN without stop codon was amplified from TOPO vector containing NERCLIN and cloned into pEGFP-N1 vector using EcoRI/AgeI restriction sites. To generate NERCLIN-pBabe, fulllength NERCLIN was amplified from TOPO vector containing NERCLIN and cloned into pBabe using EcoRI/SalI restriction sites. To fuse BirA* to C-terminus of NERCLIN we used PCR overlap extension. pHA-BirA* and pBabe-NERCLIN constructs were used as templates. The resulting fused construct was inserted into pBabe vector using EcoRI and SalI restriction sites. GFP-BirA* and AIF-BirA* plasmids were as described in (6). The constructs for co-immunoprecipitation experiments (PTPMT1-V5, NERCLIN-Strep-HA, PTPMT1-Strep-HA, and NERCLIN-V5) were generated by the Genome Biology Unit supported by HiLIFE and the Faculty of Medicine, University of Helsinki, and Biocenter Finland. All plasmids were confirmed by Sanger sequencing. Primers used for the cloning are in Dataset S5. AcGFP1-Mito (MitoGFP) and pDsRed2-Mito Vector (MitoVector Red) were purchased from Clontech (Mountain View, CA).

Proteinase K accessibility assay
Mitochondria were isolated as described in (7) and the proteinase K accessibility assay was performed as previously described (8). Briefly, after 24 h transient overexpression of NERCLIN cells from at least two 15-cm dishes were harvested. Cells were washed once with cold PBS (Life Technologies, BE17-516F) and spun at 500 g for 3 min at 4ºC. The cell pellet was resuspended in 10 mL of ice-cold extraction buffer (10 mM Tris-Mops, pH 7.4, 1 mM EGTA-Tris, pH 7.4, 0.2 M sucrose (Sigma, S0389), pH adjusted to 7.4) with 1x protease inhibitor cocktail (Thermo Scientific, 78444), and cells were disrupted using a glass-glass homogenizer. Unbroken cells were spun by centrifugation at 600 g for 10 min at 4ºC, and the pellet was subjected to the second round of homogenization. Mitochondria were recovered from the supernatant of both rounds of homogenization by centrifugation at 7 000 g for 10 min at 4ºC, and the pellet was washed once with a cold extraction buffer and centrifuged again. The final pellet was resuspended in 200 μL of extraction buffer and the mitochondrial protein concentration was determined with bicinchoninic acid (BCA) Protein Assay Kit (Thermo Scientific, 23227). For the Proteinase K accessibility assay, 50 μg of mitochondria were resuspended in 100 μL of extraction buffer (untreated), extraction buffer and Proteinase K (protease-treated mitochondria), 2 mM HEPES (Life Technologies, 15630-049), pH 7.4 and Proteinase K (protease-treated mitoplasts), or 2 mM HEPES, pH 7.4, 0.1% Triton X-100 (Thermo Scientific, 11473613) and Proteinase K (protease-and detergent-treated mitoplasts). 100 μg/mL proteinase K (Thermo Scientific, EO0491) was used in all cases. Samples were incubated on ice for 30 min and the reaction was inactivated with a final concentration of 1mM PMSF (Sigma, 93482). Samples were precipitated with 10% trichloroacetic acid, washed once with cold acetone and resuspended in 50 μL of 4 × Laemmli sample buffer (BioRad, 1610747) containing 4% (vol/vol) β-mercaptoethanol (BioRad, 1610710) and protease inhibitor cocktail (Thermo Scientific, 78444). Subsequently, 10 μL of each sample were boiled at 95ºC for 5 min and separated by SDS-PAGE, followed by western blotting analysis (described in 'Western blotting').

Immunocytochemistry
143B cells were plated to the coverslips in 6-well plates. To label mitochondria, the cells were transiently transfected with pDsRed2-Mito Vector (Clontech, 632421). In the experiments for studying intracellular localization of NERCLIN, the cells were co-transfected with the NERCLIN-EGFP construct. In the experiments with BirA* fused proteins, the cells were co-transfected with NERCLIN-BirA*. After 24 h of transfection, the cells were fixed with 4% paraformaldehyde for 10 min at RT and washed with PBS (Life Technologies, BE17-516F). Then the cells were permeabilized with Triton X-100 (Thermo Scientific, 11473613) for 15 min at RT, washed, and blocked with 5 % BSA (Jackson ImmunoResearch, 001-000-162) for 2 h at RT. Cells were then incubated with corresponding primary antibody against GRPEL2 (Novus Biological, NBP1-85099, 1:50) in a blocking buffer overnight at +4 ⁰C. After washing cells were incubated with secondary antibody for 1 h at RT (Alexa Fluor 488 goat anti-rabbit, Invitrogen, R37116, 1:400). Finally, the cells were washed, mounted using antifade mounting medium containing DAPI (Vector Laboratories, H-1000) and imaged with Axio Observer Z1 (Zeiss).

Generation of gene knockouts
CRISPR/Cas9 was used to generate knockout HEK293 cells. Cells were co-transfected with two gRNA transcriptional cassettes prepared by PCR and CAG-Cas9-T2A-EGFP plasmid (Addgene, 7831) as described elsewhere (9). To generate a Full KO cell line lacking GRPEL2 and NERCLIN, one of the guide RNAs was targeted to the 5' UTR region and the other to the beginning of exon 1. To generate GRPEL2-specific knockout cell line one of the guide RNAs was targeted to the intron 2 region and the other to intron 3 of the GRPEL2 gene to remove exon 3. After 24 h of transfection GFP-positive cells were sorted by FACS and single-cell clones were generated. The sequences of gRNAs are in Dataset S5.

Respirometry analysis
Mitochondrial OCR in HEK293 cells was determined using Seahorse XF96 Extracellular Flux Analyzer (Agilent). Cells were seeded at density of 20 000 cells/well on 0.1mg / mL poly-D-lysine (Merck Millipore) coated XF96 Seahorse plate the day before the experiment. Next day, the medium was replaced 1 h before the measurement with Seahorse XF Base Medium (Agilent #103335-100) supplemented with 1 mM pyruvate, 2 mM glutamine, 25 mM glucose and 5 mM HEPES and pH adjusted to 7.4. The OCR was measured three times in baseline conditions and three times after each injection: 1.5 μM oligomycin (Sigma), 0.25 μM uncoupler carbonilcyanide p-triflouromethoxyphenylhydrazone (FCCP) (Sigma) and 1 μM rotenone (Sigma) and antimycin-A (Sigma). OCR data were normalized to DNA concentration using CyQUANT kit (Invitrogen). Only wells with a homogeneous monolayer of cells were included in the analysis.

BioID analysis
143B cells were grown on 15 cm dishes. The cells were transfected with corresponding BirA* fusion constructs using jetPRIME (Polyplus, 114-15) according to the manufacturer's manual. Following 24 h of transfection, 50 μM biotin (Sigma, B4639-1G) was added to each plate and allowed to biotinylate the proximal proteins for the next 24 h. The plates were washed with PBS, and cells scraped, pelleted, and frozen at -80°C. For each sample, four biological replicates were analyzed. Biotinylated proteins were extracted using streptavidin beads (Strep-Tactin sepharose resin, IBA, 2-1201-002) and analyzed by mass spectrometry as previously described (10). The mass spec analysis was done in technical duplicates. The mass spectrometry data were analyzed as previously described (10,11).

Co-immunoprecipitation
One day prior to the transfection, HEK293 CTRL and Full-KO cells were plated on 15 cm dishes. The cells were co-transfected with PTPMT1-V5 and NERCLIN-Strep-HA or with PTPMT1-Strep-HA and NERCLIN-V5 using jetPRIME (Polyplus, 114-15) by 15 μg of each plasmid (30 μg in total). After 24 h mitochondrial isolation was performed followed by immunoprecipitation of the HA-tagged proteins. Briefly, the mitochondria were resuspended in 200 μL of HENN buffer (50 mM HEPES pH 8. 5 mM EDTA, 150 mM NaCl, 50 mM NaF, 0.5% Np40, 1 mM DTT, 1 mM PMSF (Sigma, 93482), 1.5 mM Na3VO4 and protease inhibitor cocktail (Thermo Scientific, 78444)) and incubated for 20 min on ice. Meanwhile, monoclonal anti-HA−agarose beads produced in mouse (Sigma, A2095) were washed with HENN buffer and centrifuged for 1 min at 4 000 g at 4°C. Lysed mitochondria were centrifuged for 20 min at 14 000 g at 4°C, and 175 μL of the supernatant were added to the washed beads. The remaining 25 µL were used later as input. Samples were incubated on a rotating wheel for 6 h at 4°C and then were centrifuged at 4 000 g for 2 min 4°C. The resulting supernatant (flow-through) was collected into a new tube and the beads were washed three times with 500 µL of HENN buffer and centrifuged at 4 000g for 2 min 4°C. After the final wash, 20 µL of 4 x loading buffer and 20 µL of HENN buffer were added, and samples were boiled for 10 min at 95°C. Equal volumes of each sample were separated by SDS-PAGE followed by western blotting (described in 'Western blotting').

Immunoprecipitation
Three replicates were used to prepare samples for IP-MS analysis. 143B cells were plated on 10 cm dishes one day before the transfection. The cells were transfected with NERCLIN-Strep-HA or GFP-Strep-HA using Transit-X2 transfection reagent (Mirus, MIR6000). After 24 h the cells were lysed using HENN buffer (50 mM HEPES pH 8. 5 mM EDTA, 150 mM NaCl, 50 mM NaF, 0.5% Np40, 1 mM DTT, 1 mM PMSF (Sigma, 93482), 1.5 mM Na3VO4 and protease inhibitor cocktail (Thermo Scientific, 78444)) and incubated for 20 min on ice. The protein lysates were centrifuged at for 20 min at 14 000 g at 4°C. Monoclonal anti-HA−agarose beads produced in mouse (Sigma, A2095) or Strep-Tactin sepharose resin (IBA, 2-1201-002) were washed with HENN buffer and centrifuged for 1 min at 4 000 g at 4°C. The clear protein lysates were added to the prewashed beads and incubated on a rotating wheel for 6 h at 4°C. Then the beads were pelleted at 4 000 g for 2 min 4°C and washed three times with HENN buffer. Following immunoprecipitation, the samples were analyzed by mass spectrometry or by immunoblotting. For mass spectrometry analysis the protein complex were eluted from Strep-Tactin beads with the HENN buffer containing biotin (200 mM). The purified protein samples were reduced with 5 mM Tris(2carboxyethyl)phosphine (TCEP; Sigma-Aldrich) for 20 min at 37°C, and then alkylated with 10 mM iodoacetamide (IAA; Sigma-Aldrich) for 20 min at room temperature in the dark. Sequencing Grade Modified Trypsin (Promega) was then used to get a 1:100 enzyme to substrate ratio and the samples were incubated overnight at 37°C. The samples were desalted with C18 macrospin columns (Nest Group). For immunoblotting analysis Laemmli sample buffer (Bio-Rad, 1610737) was added onto the washed beads, and the bound complexes were eluted by boiling for 10 min at 95°C, followed by immunoblotting analysis (described in 'Western blotting').

Mass spectrometry analysis of proteins
The desalted samples were examined with an Evosep One liquid chromatography system connected to a Bruker timsTOF Pro hybrid trapped ion mobility quadrupole TOF mass spectrometer via a CaptiveSpray nano-electrospray ion source. For peptide separation using the 60 samples per day approaches, an 8 cm 150 m column with 1.5 m C18 beads (EV1109, Evosep) was employed (21 min gradient time). The concentrations of formic acid in mobile phases A and B were 0.1% in water and 0.1% in acetonitrile, respectively. The MS analysis was carried out in the positive-ion mode utilizing data-dependent acquisition (DDA) in PASEF mode, with a total of 10 PASEF scans each topN acquisition cycle. MSFragger (12) was used to analyze raw data (.d) obtained in PASEF (13) mode against the human Uniprot database. Only mitochondrially localized proteins were included for further analysis. The mass spectrometry data were filtered based on the FDR (set to 0,05) and the fold change (set to 2).

Transmission electron microscopy
HEK293 cells were cultured on glass coverslips in 6-well plates for 24 h, then transfected with the corresponding construct. After 48 h of transfection, the cells were fixed with a solution of 2% glutaraldehyde (Sigma, G7651) in 0.1 M sodium cacodylate buffer at pH 7.4 for 25 min at room temperature. Then the cells were washed two times for 3 min with 0.1 M sodium cacodylate buffer (pH 7.4). Fixed cells were then processed according to the standard protocol at the Electron Microscopy Unit of the Institute of Biotechnology, University of Helsinki. Images were acquired with the Jeol JEM-1400 transmission electron microscope. ImageJ software was used to quantify mitochondrial major axis length (mitochondrial length) or cristae width.

Mass spectrometric analysis of lipids
To prepare mitochondrial samples four 15 cm dishes of HEK293 cells were used for each sample. The cells were washed on the plate two times with PBS (Life Technologies, BE17-516F) and once with ice-cold 0.25 M sucrose (Sigma, S0389). Next, the cells were scraped in 0.25 M sucrose and pelleted at 3 000 g for 10 min. The supernatant was removed, and the cell pellets were stored at 80°C. Lipids were extracted from the cell pellets or isolated mitochondria according to the Folch method (14). The samples were analyzed with electrospray ionization- . The MS source temperature was 250°C and instrument nitrogen was used as the nebulizing (40 psi) and the drying gas (3 L/min). Precursor ion scans were employed to produce lipid class specific scans as follows: m/z 184 for PC and SM; m/z 241 for PI; m/z 264 for ceramide (Cer) and hexosylceramide (HexCer); and m/z 369 for cholesterol ester (CE) as ammonium adducts (15)(16)(17). Neutral loss scans were employed for PE (loss of 141 amu) and PS (loss of 87 amu) (15). PE plasmalogens (PEp) were identified with fragment-specific scans for the vinyl ether chains at the sn-1 position (e.g., m/z 364, 390, and 392 for 16:0p, 18:1p, and 18:0p, respectively) and analyzed from the MS-scan (18). Triacylglycerol (TAG) species were analyzed as ammonium adducts from the MS+ scan (19). For CL and PG species analysis, the lipid extract was spiked with CL 14:0/14:0/14:0/14:0 and PG 20:1/20:1 (Avanti Polar Lipids) internal standards, methylated (20) and the lipids were extracted as above (14). The samples were then run on ACQUITY Ultra Performance LC system coupled to ESI source of Quattro Micro triple quadrupole MS (Waters, Manchester, UK) as described previously (21) and samples of the heat stress experiment were run with 6410 Triple Quadrupole LC/MS (Agilent Technologies). The CL were detected by MS+ scanning, and the PG species were selectively detected by scanning for the neutral loss of 203 amu (22). The spectra were extracted from the chromatogram and all lipid species were identified and quantified using Lipid Mass Spectrum Analysis software (23) and internal and external standards. The lipid species are marked as follows: [lipid class] [sum of acyl chain carbons]:[sum of acyl chain double bonds] (e.g., PC 34:1). The data are described as molar percentages (mol%) for each individual species relative to its lipid class. The lipidomics data were analyzed with R (https://www.R-project.org/) using limma (24). The raw intensities were normalized with the voom-function. The analysis was done using linear regression (lmFit), with empirical Bayesian statistics (eBayes).

Fluorometric measurement of total CL
To measure total CL in cells and isolated mitochondria fluorometric CL assay kit (BioVision, K944) was used. Cells or mitochondrial pellets were resuspended in the CL assay buffer and lysed by sonication. The debris was removed by centrifugation at 10 000 rpm for 10 min at +4°C. Then, the CL measurement was performed according to the manufacturer's instructions.

Cell viability assay
ATPlite viability assay kit (PerkinElmer, 6016943) was used to analyze cell viability. HEK293 cells were seeded on 96-well plates two days before the experiment to reach ~90% confluence the day of the experiment. The cells were washed twice with HBSS solution (Gibco, 14025092) and 100 µL of HBSS was added to each well. Then, the cells were exposed for 1 h or 2 h at 45°C in a cell incubator in normoxia and 5% CO2. Control cells were kept at 37°C. All samples were lysed at the same time prior to the assay, which was performed according to the manufacturer's recommendations. Luminescence was monitored with an EnSpire (PerkinElmer) microplate reader with 0.1 s measurement time and normalized to protein levels determined in parallel wells by bicinchoninic acid (BCA) assay (Thermo Fisher, 23227).

Statistical analysis
All data are presented as mean ± standard deviation (SD). For the statistical analyses with two samples, unpaired two-tailed t-tests or one-way ANOVA analysis were performed using Graph Pad Prism Software. All the graphs were prepared with Graph Pad Prism Software.      Total OPA1 level was taken as 100%. L, long OPA1 isoform, S, short OPA1 isoform (n = 3). (D, E) Quantification of western blot images presented in (A). Protein expression levels are normalized to β-tubulin level (n = 3). In all graphs data are presented as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, ns, not significant as compared to the cells transfected with empty vector (unpaired t-tests). in HEK293 cells exposed to H2O2 as determined by qPCR assay (n = 4). (B) Live microscopy images of HEK293 control cells (CTRL) or cells lacking GRPEL2 and NERCLIN (Full-KO) exposed to heat stress (42°C or 45°C). Scale bar: 200 μm. (C) Viability of CTRL, Full-KO or GRPEL2 KO HEK293 cells exposed to heat stress (45°C). ATPlite viability assay kit was used to analyze cell viability. ATP content was normalized to protein levels. (D) Viability of HEK293 control cells transiently transfected for 24 h with NERCLIN plasmid (NERCLIN OE) or with an empty vector (pBabe) exposed to heat stress (42°C). ATPlite viability assay kit was used to analyze cell viability. ATP content was normalized to protein levels. The viability relative to the . In all graphs data are presented as mean ± SD. *P < 0.05, **P < 0.01, ****P < 0.0001, ns, not significant (unpaired t-tests).