Ndufa6 regulates adipogenic differentiation via Scd1

ABSTRACT Obesity and associated complications are becoming a pandemic. Inhibiting adipogenesis is an important intervention for the treatment of obesity. Despite intensive investigations, numerous mechanistic aspects of adipogenesis remain unclear, and many potential therapeutic targets have yet to be discovered. Transcriptomics and lipidomics approaches were used to explore the functional genes regulating adipogenic differentiation and the potential mechanism in OP9 cells and adipose-derived stem cells. In this study, we found that NADH:ubiquinone oxidoreductase subunit A6 (Ndufa6) participates in the regulation of adipogenic differentiation. Furthermore, we show that the effect of Ndufa6 is mediated through stearoyl-CoA desaturase 1 (Scd1) and demonstrate the inhibitory effect of a SCD1 inhibitor on adipogenesis. Our study broadens the understanding of adipogenic differentiation and offers NDUFA6-SCD1 as a potential therapeutic target for the treatment of obesity.


Introduction
Epidemiological investigations have shown that the prevalence of obesity has increased dramatically worldwide. Increased obesity contributes to four million deaths annually and 120 million disability-adjusted life-years [1]. Therefore, the prevention, reduction and treatment of obesity and its complications are urgent issues for public health [2,3].
Obesity is characterized by imbalanced food intake and energy expenditure [4]. Excessive energy is stored in the form of triglycerides, leading to an augmented adipocyte number and/or size [5]. In addition, aberrant adipogenic differentiation plays a critical role in the progression of obesity [6]. Adipogenic differentiation is the process that converts preadipocytes into adipocytes, and this process is accompanied by fat synthesis and follow-up lipid droplet formation and enlargement [7]. Differentiation of preadipocytes into adipocytes occurs through various mechanisms. In mammalian cells, there are three main adipogenesis regulators: peroxisome proliferator-activated receptor γ (PPARγ), CCAAT/enhancer binding protein α (C/EBPα) and sterol regulatory element-binding protein (SREBP-1) [8]. C/EBPs are expressed in the early differentiation phase and activate PPARγ expression [9]. PPARγ promotes lipogenic gene expression, and SREBP-1 controls fatty acid (FA) biosynthesis, such as acetyl-CoA carboxylase and fatty acid synthase [10]. Fatty acid binding protein 4 (FABP4), adiponectin, and fatty acid synthase (FASN) are responsible for the formation of mature adipocytes. FABP4 is mainly expressed in adipocyte tissue and is known to promote the storage of lipids [11]. Adiponectin is an adipokine that is secreted by adipocytes, and it is involved in the crosstalk between adipose tissue and other metabolic tissues [12]. FASN is an enzyme involved in endogenous fatty acid synthesis and is considered a central enzyme in this process [13]. In contrast, a transcription factor, preadipocyte factor-1 (PREF-1), is downregulated during adipogenic differentiation [14]. Therefore, adipogenic differentiation depends on the coordinated regulation of gene expression.
Inhibiting key adipogenic genes may be an effective antiobesity therapeutic strategy. Indeed, several inhibitors targeting fat synthesis enzymes (e.g., FASN and acetyl-CoA carboxylases or ACC) have been shown to exert therapeutic effects in obesity and related metabolic disorders [15][16][17][18][19]. However, the adipogenic differentiation process involves numerous pathways, and many potential targets for suppressing fat synthesis have yet to be discovered.
In the present study, we conducted transcriptomics analyses in two classical adipogenic differentiation models. We found that NADH:ubiquinone oxidoreductase subunit A6 (Ndufa6) participates in the regulation of adipogenic differentiation. Furthermore, we show that the effect of Ndufa6 is mediated through stearoyl-CoA desaturase 1 (Scd1). Thus, the newly discovered NDUFA6-SCD1 pathway may serve as an attractive therapeutic target for obesity.

Ndufa6 is upregulated in adipogenic differentiationx
To uncover potential genes that are involved in the regulation of adipogenic differentiation, transcriptomics was performed comparing differentiated to undifferentiated OP9 or differentiated to undifferentiated ASC. There were 2005 and 2381 genes upregulated after adipogenic differentiation in OP9 and ASC, respectively (Figure 1a). Among the two upregulated gene sets, 632 genes were the same ( Figure 1b). Subsequently, these genes were subjected to Gene Ontology (GO) analysis, and the top two terms were 'oxidation-reduction process' and 'TCA cycle and respiratory electron transport' (Figure 1c). This result indicated that adipogenic differentiation was accomplished by mitochondrial oxidative metabolism. Electron transport is an important step in mitochondrial oxidative metabolism. Thus, the GO term 'TCA cycle and respiratory electron transport' was subdivided, and the results presented a wealth of terms related to complex I ( Figure 1c). Complex I is the largest mitochondrial respiratory chain complex, and its deficiency accounts for almost one-third of respiratory chain disorders [20]. To validate this analysis and identify novel candidate genes related to adipogenic differentiation, the top 10% of the 632 upregulated genes were selected. Most genes were already known to participate in adipogenic differentiation, such as Adipoq (4800-fold in OP9 and 780-fold in ASC) and Fabp4 (3400-fold in OP9 and 74-fold in ASC). Among the top-induced genes that had not been previously studied in adipogenic differentiation, we focused on Ndufa6, a subunit of complex I that is essential for correct complex assembly [21]. Its expression was upregulated >20-fold in ASCs and >5-fold in OP9 cells after adipogenic differentiation, as measured by qPCR (Figure 1d).

Ndufa6 silencing prevents adipogenic differentiation
OP9 cells have the advantages of rapid adipogenic differentiation and high transfection efficiency [22]. To uncover its role in adipogenic differentiation, Ndufa6 was silenced during adipogenic differentiation. A siRNA targeting Ndufa6 substantially downregulated Ndufa6 gene expression in OP9 cells (Figure 2a). Cells transfected with Ndufa6 siRNA had markedly smaller lipid droplets and lower TG content than cells transfected with control siRNA (Figure 2b). Consistently, 4 adipogenic differentiation marker genes, Pparg, Fasn, Fabp4 and C/ebpα, were significantly downregulated in Ndufa6-silenced cells (Figure 2c). These data indicate that Ndufa6 silencing prevents adipogenic differentiation in OP9 cells.

Transcriptomics analysis implicates Scd1 as a target of Ndufa6
To explore the potential mechanism of Ndufa6 in adipogenic differentiation, gene expression profiles were determined by RNA-seq in OP9 cells with or without Ndufa6 knockdown. A total of 934 differentially expressed genes  The cellular TG content was tested in OP9 cells after adipogenic differentiation (n = 3), and the test results were normalized by the protein content. (c) The relative mRNA levels of adipogenic differentiation marker genes were tested with RT-qPCR (n = 3). Values are presented as the mean ± standard deviation. Statistical analysis was performed by the LSD t-test. * P < 0.05; ** P < 0.01; *** P < 0.005. (≥1.5-fold change, p value<0.05) were identified, of which 238 were downregulated and 696 were upregulated ( Figure  4a). Significantly enriched functional categories, such as respiratory electron transport and lipid and fatty acid metabolic processes, were observed among the downregulated genes ( Figure 4b). However, no specifically enriched biological processes were found among the upregulated genes. A heatmap of the top 20 downregulated genes was plotted (Figure 4c), and Ndufa6 and Scd1 had similar expression patterns. Decreased Scd1 expression in Ndufa6 knockdown cells was confirmed by RT-qPCR (Figure 4d).
The expression of Scd2, Scd3 and Scd4 was not detectable in OP9 cells. Transcriptomics data corroborate the lipidomics results, implicating Scd1 as a target of Ndufa6.

Ndufa6 knockdown inhibits adipogenic differentiation through Scd1
To investigate whether Scd1 participates in Ndufa6mediated adipogenic differentiation, Scd1 was silenced or overexpressed in OP9 cells. When Scd1 was silenced individually, markedly decreased lipid droplets, TG content  These results indicate that Ndufa6 knockdown inhibits adipogenic differentiation through Scd1.

Scd1 inhibitor suppresses adipogenic differentiation
The above results demonstrate that the Ndufa6-Scd1 pathway may be a potential therapeutic target in obesity intervention. Ndufa6 is a subunit of complex I, and its inhibition may induce unintended side effects, as a previous study demonstrated that complex I inhibition The cellular TG content was tested in OP9 cells after adipogenic differentiation (n = 3), and the test results were normalized by the protein content. (d) RT-qPCR was used to test the relative mRNA level of Scd1 and other adipogenic differentiation marker genes (n = 3). (e) After adipogenic differentiation, the cellular fatty acids were tested, and the Scd1 product and substrate (C16:0, C16:1, C18:0 and C18:1) were analysed (n = 3). Statistical analysis was performed by the LSD t-test and oneway ANOVA. * P < 0.05; ** P < 0.01; *** P < 0.005. Different letters represent significant differences (P < 0.05, ANOVA). results in glycolysis enhancement and lactic acidosis [25]. Thus, we investigated whether a Scd1 inhibitor alone could suppress adipogenic differentiation. The Scd1 inhibitor A939572 markedly decreased lipid droplets, the cellular TG content (Figure 6a), the expression of adipogenic differentiation marker genes Pparg, Fasn, Fabp4 and C/ebpα (Figure 6b) and the levels of C16:0, C16:1, C18:0 and C18:1 (Figure 6c).

Discussion
In the present study, we found that Ndufa6 is a key regulator in adipogenic differentiation. Ndufa6 is a subunit of complex I, the largest and most complex enzyme in the oxidative phosphorylation system [21,26]. Oxidative phosphorylation is essential for normal adipogenic differentiation [27,28]. Complex I assembly errors or conformational changes impair the cellular oxidative phosphorylation system [20]. Notably, Ndufa6 knockdown markedly inhibited adipogenic differentiation, whereas Ndufa6 overexpression did not affect adipogenic differentiation (data not shown). Hence, Ndufa6 likely exerts its role in adipogenic differentiation through assembled complex I but not itself. During adipogenic differentiation, the cells retained a high level of metabolic activity and increased substrate consumption along with a marked increase in the mitochondrial content and higher β-oxidation [29]. Lipid metabolism, including lipid transport, synthesis, and catabolism, requires high levels of energy and fully functional mitochondria, and the normal function of complex I is critical for these processes [30]. Therefore, it is anticipated that loss of other subunits of complex I will also inhibit adipogenic differentiation.
Our results suggested that Ndufa6 may be a therapeutic target for obesity. However, it should be noted that complex I dysfunction induced by inhibiting Ndufa6 may lead to impairment of the ability of the respiratory chain to oxidize NADH to NAD + and the overproduction of ROS [21,31]. Excessive levels of ROS provoke lipid peroxidation and damage cellular proteins and DNA [32]. Therefore, on the basis The cellular TG content was tested in OP9 cells after adipogenic differentiation (n = 3), and the test results were normalized by the protein content. (b) RT-qPCR was used to test the relative mRNA level of Scd1 and other adipogenic differentiation marker genes in OP9 cells with or without Scd1 inhibitor after adipogenic differentiation (n = 3). (c) After adipogenic differentiation, the cellular fatty acids were tested, and the Scd1 product and substrate (C16:0, C16:1, C18:0 and C18:1) were analysed (n = 3). Statistical analysis was performed by the LSD t-test. * P < 0.05; ** P < 0.01; *** P < 0.005. of our study, the development and proper selection of Ndufa6 inhibitors may contribute to the treatment of obesity.
We identified Scd1 as a target of Ndufa6 in regulating adipogenic differentiation. Ndufa6 knockdown prominently decreased MUFA levels in cells. This phenotype is consistent with the phenotype of Scd1 knockdown in previous reports [33][34][35]. SCD1, a stearoyl-CoA desaturase converting stearic acid (C18:0) and palmitic acid (C16:0) into oleic acid (C18:1n9) and palmitoleic acid (16:1n7), respectively, requires electrons for the desaturation reaction [36]. Thus, it is likely that complex I dysfunction induced by Ndufa6 knockdown leads to inhibition of SCD1 activity in addition to the downregulation of transcription. Numerous studies have revealed that blocking SCD1 leads to significant inhibition of adipogenesis [33,[37][38][39]. Some metabolic diseases, such as obesity and nonalcoholic fatty liver disease, involve adipogenesis [24]. Thus, targeting SCD1 can be an effective treatment for these metabolic diseases. The results suggested that targeting Ndufa6, the upstream gene of SCD1, can also cause inhibitory effects of SCD1. This provides a new idea and a new way to develop SCD1 inhibitors in the future.
Other fatty acid desaturases also require electrons for the desaturation reaction. It is unclear why polyunsaturated fatty acid levels were not affected in Ndufa6 knockdown cells (Fig. S1A).
Numerous studies have reported that key adipogenic genes, such as Pparg, Fasn, Fabp4 and C/ebpα, directly participate in adipogenic differentiation [8,40,41]. Pparg and C/ebpα together promote differentiation by activating adipose-specific gene expression, such as Fasn and Fabp4, and by maintaining each other's expression at high levels [40]. Knockdown of both Ndufa6 and Scd1 markedly decreased the expression of adipogenic marker genes, indicating that the Ndufa6-Scd1 pathway may be an upstream regulator of adipogenic differentiation. Therefore, the Ndufa6-Scd1 pathway can be a key therapeutic target for the management of obesity.

Cell cultures and adipogenic differentiation
OP9 cells were maintained in αMEM supplemented with 5% FBS. For OP9 adipogenic differentiation, cells were grown to 90% confluence and then changed to adipogenic differentiation medium (DMEM supplemented with 5% FBS, 1 mM rosiglitazone) every three days for a total of 8 days. Adipose-derived stem cells (ASCs) were isolated according to a previous report [43] with slight modification. In brief, 8-to 10-week-old male C57/BL6J mice were sacrificed, and subcutaneous adipose tissues were taken from the inguinal fat pads. Next, tissues were washed extensively with phosphate-buffered saline (PBS) containing 2% penicillin/ streptomycin. Upon removal of debris, the tissues were minced with scissors and digested with 0.1% type I collagenase in D-Hanks balanced salt solution for 1 h at 37°C at 200 rpm. Then, the collagenase activity was neutralized by adding an equal volume of DMEM/F12 supplemented with 5% FBS, and the sample was pipetted up and down several times. The floating adipocytes were separated by centrifugation at 400 g for 5 min. The pellet was collected and filtered through a 70 μm cell strainer. The cells were plated in a 10 cm dish with growth medium (DMEM/F12 containing 5 ng/ml EGF, 5 ng/ml bFGF and GlutaMAX). Growth medium was replenished every three days until the cells reached 90% confluence, and the cells were then passaged with trypsin/EDTA solution. Adipogenic and osteogenic differentiation assays were performed to evaluate the potential for pluripotent differentiation. For adipogenic differentiation, cells were grown to 90% confluence and then changed to adipogenic differentiation medium 1 (DMEM supplemented with 5% FBS, 0.5 M IBMX, 1 μM dexamethasone, 1 μM rosiglitazone and 1.7 μM insulin) for three days. Next, the cells were incubated with adipogenic differentiation medium 2 (DMEM supplemented with 5% FBS and 1 μM rosiglitazone) for 8 days, and the medium was changed every three days. For osteogenic differentiation, cells were grown to 90% confluence and then changed to osteogenic differentiation medium (DMEM supplemented with 5% FBS, 0.01 μM 1,25dihydroxyvitamin D3, 50 μM vitamin C and 10 mM β-sodium glycerophosphate). The osteogenic differentiation medium was changed every three days for a period of 27 days. ASCs from passage 4 to passage 10 were used in this study. Both types of cells were cultured in a 37°C incubator with 5% CO 2 .

siRNA or plasmid transfection
Transfection was performed using jetPRIME transfection reagent according to the manufacturer's protocol. OP9 cells were 50% or 80% confluent at the time of siRNA or plasmid transfection, respectively. The medium was replaced with adipogenic differentiation medium 24 h post-transfection. The siRNA sequence for mouse Ndufa6 was 5ʹ-CGAGAAAUGUUCAUGAAGAAUTT-3ʹ, that of Scd1 was 5ʹ-AGUUUCUAAGGCUACUGUCUUTT-3ʹ, and universal negative control siRNA (GenePharma, A06001) was used as a control. pcDNA3.1-Scd1 was used for SCD1 expression in OP9 cells, and the pcDNA3.1 empty vector was used as a control.

Oil red O staining
Following 8 days of adipogenic differentiation, cells were rinsed with PBS, fixed with 4% neutral buffered formalin at room temperature for 30 min, washed twice with PBS, stained with Oil Red O reagent for 15 min at room temperature, and then counterstained with haematoxylin solution. Cell images were captured using a Nikon Eclipse Ti2-U inverted microscope.

Alizarin red staining
Alizarin red staining was performed based on the instructions of the Alizarin Red S staining kit. Cell images were captured using a Nikon Eclipse Ti2-U inverted microscope.

Western blot
Cells were washed with phosphate-buffered saline (PBS) and lysed in RIPA buffer with 1X protease inhibitor cocktail. Protein samples were separated by SDS-PAGE and electrically transferred to PVDF membranes. TBST containing 5% skim was used to block the membranes for 1 h, and the membranes were washed 3 times with TBST. The membranes were incubated overnight at 4°C with primary antibodies. The next day, membranes were washed 3 times with TBST and incubated with HRP-conjugated secondary IgG antibody for 1 h at room temperature. Before imaging, the membranes were washed with TBST 3 times, and an ECL reagent kit was used to detect expressed proteins.

Measurement of total triacylglycerol (TG)
A TG assay kit was used to measure the TG content, and total cellular protein was determined by a total protein assay kit according to the manufacturer's protocol. The TG content was normalized to the total cellular protein.

Fatty acid methyl ester (FAME) analysis
Cells were collected after adipogenic differentiation using RIPA buffer, lysed by ultrasound using a SONICS® (Sonics & Materials, Inc.) sonicator at a 20% amplitude setting (work 2 s and rest 3 s, 5 cycles). Total protein was measured using a total protein assay kit. Total lipid extraction was performed according to the method of Bligh and Dyer [44] using 14% boron trifluoride methanol (v/v) as the methylating agent [45]. Samples were quantified on a Q Exactive™ GC Orbitrap™ GC-MS/MS (Thermo Scientific) with a Rtx-Wax column (30.0 m × 0.25 mm, Restek, 12423). The temperatures of both the injection port and the detector were kept constant at 280°C. The column temperature was initially held at 40°C for 5 min, followed by an increase at a rate of 40°C/min to 120°C, then to 190°C at 10°C/min for 5 min, and then to 230° C at 5°C/min for 7 min; the total time was approximately 34 min for all fatty acid peaks. Peaks were identified by comparing retention times with known standards (Sigma Chemical). Individual fatty acids were quantified by reference to the internal standard (C17:0). Subsequently, the sample was normalized to total cellular protein.

Lipidomic analysis
Cell samples were prepared as described above. Total lipid extraction was performed according to the method of Matyash et al. [46]. Lipidomic analysis was performed on a Q Exactive Plus mass spectrometer (Thermo Scientific) equipped with a Vanquis UHPLC (Thermo Scientific). Mass spectrometry (MS) and tandem mass spectrometry (MS/MS) data were acquired in positive and negative ion modes. Before sample infusion into the MS, 1 µl of lipid mixture was separated at 40°C on an ACQUITY UPLC BEH C18 column (1.7 μm, 2.1 × 100 mm, Waters). For both the positive and negative modes, mobile phase A was 10 mM ammonium acetate in acetonitrile:H 2 O (60: 40), and mobile phase B was 10 mM ammonium acetate in isopropyl alcohol:acetonitrile (90:10). Liquid chromatography gradients were as follows: 0-3 min, 40% B-40% B; 3-20 min, 40% B-95% B; 20-22.5 min, 95% B-95% B; and 22.5-23 min, 95%-40% B. The flow rate was 0.25 ml/min. The quality control (QC) samples were prepared by mixing different samples, and the order of QC samples was evenly distributed among samples. Lipids were identified using LipidMap (www. lipidmaps.org), and statistical analysis of lipid profiling of samples was performed using SIMCA (14.1) software.

cDNA library construction and RNA sequencing (RNA-seq)
For cDNA library construction, first-strand cDNA was reverse-transcribed from mRNA and further converted into double-strand cDNA. Then, the double-stranded cDNA was resuspended in Tn5 transposase reaction mix, followed by digestion and tagmentation. Adapters and primers were synthesized according to published Illumina sequences. Enrichment PCR was performed using HIFI PCR Mix for NGS. The PCR amplification procedure was: 72 ℃ for 5 min, 98 ℃ for 30 s, and 25 cycles of 98 ℃ for 10 s, 65 ℃ for 30 s, 72 ℃ for 1 min, and 72 ℃ for 10 min. Libraries were quantified using Agilent 2100 BioAnalyzer, and sequenced using Illumina NovaSeq instrument (Sequencing was performed by GENEWIZ Biotech). Sequencing data were analysed using STAR (http:// code.google.com/p/rna-star/) and R studio (R studio Inc.). The differentially expressed genes were defined as genes with a P value < 0.05 and a fold change ≥ 1.5. Gene ontology (GO) analysis was performed using Metascape (http://metascape.org). Heatmap was generated using Tbtools software (https://github.com/CJ-Chen/TBtools).