NADH dehydrogenase reverses dietary and clock metabolic syndrome

Abstract

Circadian clocks are internal timing systems that enable organisms to anticipate and adapt to daily environmental changes. These rhythms arise from a transcription-translation feedback loop in which CLOCK/BMAL1 regulate the expression of thousands of genes, including their repressors PER/CRY¹. Disruption of circadian rhythms contributes to obesity, metabolic disease, and cancer^2-4, yet how the clock maintains metabolic homeostasis remains limited. Here we report that the clock regulates oxidative metabolism through diurnal respiration of mitochondrial respiratory chain complex I. Genetic loss of the clock and high fat diet feeding in male mice led to reduced complex I respiration within adipocytes, leading to suppression of PPAR and insulin signaling pathways. In contrast, preserving complex I function maintained adipogenic and metabolic gene networks and protected against diet- and circadian-induced metabolic dysfunction independently of weight gain. These findings reveal that circadian disruption impairs metabolic health through mitochondrial complex I dysfunction, establishing clock control of complex I as a key regulator of transcriptional and metabolic homeostasis.

Keywords: BMAL1; Circadian clock; NDI1; adipocytes; complex I; metabolism; mitochondria.

Extended Data Figure 1.. The circadian clock regulates the abundance of mitochondrial complex I, related to Figure 1.

a, OCR in iWAT mitochondria of Control and Bmal1-KO mice at ZT2 and ZT14 in the presence of indicated substrates and inhibitors (n = 5). TMPD, Tetramethyl-p-phenylenediamine; Asc acid; ascorbic acid. b, Representative bioluminescence of PER2::LUC oscillations in synchronized primary white adipocytes from mPer2:Luc mice. c, Relative mRNA expression of Bmal1 in synchronized white adipocytes in vitro (n = 6). d, NADH-dependent respiration in synchronized white adipocytes in vitro (n=6). Data are represented as mean ± SEM. Statistical significance was calculated by two-way ANOVA followed by Dunnett’s multiple comparisons test with ZT14 Control set as the reference group in a.

Extended Data Figure 2.. The circadian clock regulates the abundance of mitochondrial complex I, related to Figure 1.

a, mRNA abundance of complex I subunits in adipocytes from the visceral WAT depot of Control (Bmal1^fx/fx) and Bmal1-KO (Adiponectin-Cre; Bmal1^fx/fx) mice (n=6). b, SDS-PAGE immunoblot of electron transport chain proteins in isolated WAT mitochondria from Control and Bmal1-KO mice harvested at the indicated timepoints (n=6). c, BN-PAGE immunoblot on isolated WAT mitochondria from Control and Bmal1-KO mice at ZT14 using an antibody mix against OXPHOS complex subunits (n=4). d, BN-PAGE immunoblot on isolated WAT mitochondria from Control mice at ZT2 and ZT14 using an antibody mix against OXPHOS complex subunits (n=5). Data are represented as mean ± SEM. Statistical significance was calculated by two-way ANOVA followed by multiple comparisons.

Extended Data Figure 3.. High fat diet feeding leads to reduced and loss of diurnal CI respiration in visceral WAT mitochondria, related to Figure 2.

a, OCR in WAT mitochondria of wildtype mice at ZT14 following 12 weeks of chow or HFD feeding. Mitochondria were treated with pyruvate/malate/FCCP followed sequentially with rotenone, succinate, antimycin A, and N,N,N′,N′-Tetramethyl-p-phenylenediamine (TMPD)/ascorbic acid (n = 5). b, OCR in WAT mitochondria of wildtype mice at ZT2 and ZT14 following 12 weeks of HFD feeding. Mitochondria were treated with pyruvate/malate/NADH followed sequentially with rotenone/antimycin A, TMPD/ascorbic acid, and azide (n = 6). Data are represented as mean ± SEM. Statistical significance was calculated by two-way ANOVA followed by multiple comparisons in a-b. **p < 0.01.

Extended Data Figure 4.. NDI1 expression does not affect adipocyte physiology or glucose homeostasis in chow-fed mice, related to Figure 2.

a, Body weight of Control (NDI1^LSL) and NDI1 (Adiponectin-Cre; NDI1^LSL) mice at their indicated age on a chow diet (n = 5–6). b, Glucose tolerance test (GTT) and insulin during the GTT at 3 months of age (n = 5–6). c, Gene expression of indicated genes involved in thermogenesis in brown adipose tissue (BAT), inguinal white adipose tissue (iWAT), and gonadal white adipose tissue (gWAT) of Control and NDI1 mice at 3 months of age (n = 4). d, Representative images of hematoxylin and eosin staining of indicated adipose tissue depots in Control and NDI1 mice at 3 months of age (10x magnification, scale bars 200 μm). e, Mitochondria/nuclear DNA ratio in gWAT from control and NDI1 mice at 3 months of age (n = 5–6). Data are represented as mean ± SEM. Statistical significance was calculated by two-way ANOVA followed by multiple comparisons in a-c and unpaired t-test in e.

Extended Data Figure 5.. Expression of NDI1 does not affect activity, food intake, or energy expenditure during high fat diet feeding, related to Figure 2.

a, Daily activity rhythms of Control and NDI1 mice after 10 weeks of HFD feeding (n = 8). b, Total daily food intake in Control and NDI1 mice after 10 weeks of HFD feeding (n = 8). c, Total energy expenditure over 24 hours versus body weight after 10 weeks of HFD feeding (n = 8). d, Gene expression of indicated genes involved in thermogenesis in BAT, inguinal WAT, and gonadal WAT from Control and NDI1 mice after 12 weeks of HFD feeding (n = 8). e, SDS-PAGE immunoblot of complex I and II subunits in isolated gWAT mitochondria from Control and NDI1 mice after 10 weeks of HFD feeding (n=6). f, Mitochondria/nuclear DNA ratio in gWAT from control and NDI1 mice after 10 weeks of HFD feeding (n = 6). Data are represented as mean ± SEM. Statistical significance was calculated by two-way ANOVA followed by multiple comparisons in a, d, and f, unpaired t-test in b and ANCOVA in c. *p < 0.05, ***p < 0.001.

Extended Data Figure 6.. Loss of Ndufs2 in adipocytes leads to reduced complex I and III activity in gWAT, related to Figure 3.

a, SDS-PAGE immunoblot of Ndufs2 in isolated mitochondria from Control (Ndufs2^fx/fx) and Ndufs2-KO (Adiponectin-Cre; Ndufs2^fx/fx) mice at 3 months of age on a chow diet (n = 2). b, SDS-PAGE immunoblot of electron transport chain proteins in isolated gWAT mitochondria from Control and Ndufs2-KO mice at 3 months of age (n = 3–4). c, Quantification of protein abundance of samples in B. d, OCR in gWAT mitochondria of Control and Ndufs2-KO mice at ZT14 in the presence of indicated substrates and inhibitors (n = 7–7). TMPD, Tetramethyl-p-phenylenediamine; Asc acid; ascorbic acid. e, OCR of differentiated adipocytes from Control and Ndufs2-KO mice in the presence of permeabilizer (PERM), ADP, and complex I to IV substrates followed sequentially by oligomycin and antimycin A (n = 5). Data are represented as mean ± SEM. Statistical significance was calculated by two-way ANOVA followed by multiple comparisons in c, d, and e.*p < 0.05, **p < 0.01, ***p < 0.001.

Extended Data Figure 7.. Deletion of Bmal1 leads to impaired glucose tolerance on a chow diet, which is prevented by NDI1, related to Figure 4.

a, Body weight of Control, Bmal1-KO, and Bmal1-KO + NDI1 mice at 4 months of age on a chow diet (n = 5–7). (B) Glucose tolerance test (GTT) and insulin during the GTT of the mice in A at 4 months of age (n = 5–7). Data are represented as mean ± SEM. Statistical significance was calculated by one-way ANOVA in a and two-way ANOVA followed by Dunnett’s multiple comparisons test with Bmal1-KO mice set as the reference group in b. # denotes significance between Bmal1-KO vs other groups (p < 0.05).

Figure 1.. The circadian clock regulates respiration at mitochondrial respiratory chain complex I.

a, OCR in gWAT mitochondria isolated from Control (Bmal1^fx/fx) and Bmal1-KO (Adiponectin-Cre; Bmal1^fx/fx) mice at ZT2 and ZT14 in the presence of indicated substrates and inhibitors (n = 6–7). Pyr, pyruvate; FA-carn, fatty acid carnitine; Succ, succinate; Rot, rotenone; Antim A, antimycin A; Oligo, oligomycin; FCCP, Carbonyl cyanide-p-trifluoromethoxyphenylhydrazone. b, State III respiration in gWAT mitochondria isolated from Control and Bmal1-KO mice at ZT14 in response to CI- and II-linked substrates (n = 5). c, OCR in gWAT mitochondria of Control and Bmal1-KO mice at ZT2 and ZT14 in the presence of indicated substrates and inhibitors (n = 5). TMPD, Tetramethyl-p-phenylenediamine; Asc acid; ascorbic acid. d, OCR in gWAT mitochondria of wildtype mice at ZT2 and ZT14 following 12 hours of fasting in response to indicated substances as in C (n = 5). e, State III respiration in adipocytes differentiated from Control and Bmal1-KO mice in the presence of CI-IV substrates (n = 5). f, NADH-dependent OCR in adipocytes differentiated from Control and Bmal1-KO mice (n=6). g, NADH-dependent OCR in Control, Bmal1^−/− and Cry1/2^−/− MEFs (n=6). Data are represented as mean ± SEM. Statistical significance was calculated by two-way ANOVA followed by Dunnett’s multiple comparisons test with ZT14 Control set as the reference group in a and c, multiple unpaired t-tests in b and e, two-way ANOVA followed by multiple comparisons in d, unpaired t-test in f, and one-way ANOVA followed by multiple comparisons in g. The # symbol denotes significance between ZT14 Control and all other groups. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 2.. Expression of the yeast NDI1, an alternative NADH dehydrogenase, in adipocytes improves metabolic health during high fat diet feeding.

a, Schematic of the mitochondrial electron transport chain with ectopic yeast NDI1. NDI1 transfers electrons to ubiquinone and regenerates NAD+. b, OCR of adipocytes differentiated from Control (NDI1^LSL) and NDI1 (Adiponectin-Cre; NDI1^LSL) mice in the presence of indicated substrates and inhibitors (n = 5). c, Body weight during high fat diet (HFD) feeding for 12 weeks in Control and NDI1 mice (n = 8). d, RER over 24 hours after 10 weeks of HFD feeding (n = 8). e, Glucose tolerance test (GTT) and insulin during the GTT at 12 weeks of HFD feeding (n = 8). f, Representative images of hematoxylin and eosin staining of gWAT and liver after 12 weeks of HFD feeding (20x and 10x magnification, respectively, scale bars 200 μm). g, Distribution in visceral WAT adipocyte size after 12 weeks of HFD feeding (n = 5). h, Liver weights after 12 weeks of HFD feeding (n = 8). i, Mitochondria/nuclear DNA ratio in gWAT after 12 weeks of HFD feeding (n = 4–5). j, State III respiration in gWAT mitochondria in response to CI- and II-linked substrates after 12 weeks of HFD feeding (n = 6). k, Heatmap of differentially abundant metabolites (p < 0.05) in visceral WAT from Control and NDI1 mice after 12 weeks of HFD feeding at ZT14 (n = 8). l, The NAD+/NADH and ATP/ADP ratios in WAT from Control and NDI1 mice after 12 weeks of HFD feeding at ZT14 (n = 8). Data are represented as mean ± SEM. Statistical significance was calculated by two-way ANOVA followed by multiple comparisons in b-f, g, and j and unpaired t-test in h, i and l. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 3.. Adipocyte complex I deficiency limits adipose expansion and impairs systemic glucose homeostasis.

a, Body weight of Control (Ndufs2^fx/fx) and Ndufs2-KO (Adiponectin-Cre; Ndufs2^fx/fx) mice at 4 months of age on a chow diet (n = 6–7). b, Tissue weights of Control and Ndufs2-KO mice at 4 months of age (n = 6–7). c, Representative images of hematoxylin and eosin staining of gWAT and liver 4 months of age (10x magnification, scale bars 200 μm). d, Oral glucose tolerance test (GTT) and insulin during the GTT at 4 months of age (n = 6–7). e, Serum non-esterified fatty acids (NEFA) in Control and Ndufs2-KO mice at 4 months of age (n = 6–7). f, KEGG pathway analysis of downregulated (upper panel) and upregulated (lower panel) genes in Ndufs2-KO vs. Control adipocytes (n = 4). g, RNA-sequencing was performed on fractionated adipocytes from the visceral WAT of Control (Ndufs2^fx/fx), Ndufs2-KO (Adiponectin-Cre;Ndufs2^fx/fx), and Bmal1-KO (Adiponectin-Cre; Bmal1^fx/fx) mice. Pathway analysis of Gene Ontology biological process terms with highlighted genes among differentially expressed genes in Ndufs2-KO vs Control and Bmal1-KO vs Control fractionated adipocytes. h, Motif analysis of downregulated genes in Ndufs2-KO vs Control and Bmal1-KO vs Control fractionated adipocytes. Data are represented as mean ± SEM. Statistical significance was calculated by unpaired t-test in a and e, and two-way ANOVA followed by multiple comparisons in b and d. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 4.. Expression of NDI1 prevents diet-induced metabolic dysfunction in mice lacking the circadian clock in adipocytes.

a, Intracellular NAD+/NADH ratio and OCR in in the presence of indicated substrates and inhibitors in primary white adipocytes differentiated from Control, NDI1, Bmal1-KO, and Bmal1-KO + NDI1 mice (n = 5). b, Body weight during high fat diet (HFD) feeding for 12 weeks (Control n = 12, Bmal1-KO n=12, Bmal1-KO + NDI1 n=7). c, RER over 24 hours after 10 weeks of HFD feeding (Control n = 12, Bmal1-KO n=12, Bmal1-KO + NDI1 n=7). d, Oral glucose tolerance test (GTT) and insulin during the GTT at 6 weeks of HFD feeding (Control n = 12, Bmal1-KO n=12, Bmal1-KO + NDI1 n=7). e, Oral triglyceride clearance test at 6 weeks of HFD feeding (Control n = 12, Bmal1-KO n=12, Bmal1-KO + NDI1 n=7). f, Representative images of hematoxylin and eosin staining of gWAT and liver after 12 weeks of HFD feeding (10x magnification, scale bars 200 μm). g, Distribution in visceral WAT adipocyte size after 12 weeks of HFD feeding (n = 5). h, Liver weights after 12 weeks of HFD feeding (Control n = 12, Bmal1-KO n=12, Bmal1-KO + NDI1 n=7). i, The NAD+/NADH and ATP/ADP ratios in WAT after 12 weeks of HFD feeding at ZT14 (Control n = 12, Bmal1-KO n=12, Bmal1-KO + NDI1 n=7). Data are represented as mean ± SEM. Statistical significance was calculated by one-way ANOVA followed by multiple comparisons in a (left panel), h, and i, two-way ANOVA followed by Dunnett’s multiple comparisons test with Bmal1-KO set as the reference group in a (right panel), and b-e and g. The # symbol denotes significance between Bmal1-KO and indicated groups. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 5.. Mitochondrial complex I function regulates the transcriptional landscape in adipocytes.

a, Single-nuclei RNA-sequencing was performed on visceral WAT from Control (Bmal1^fx/fx; NDI1^LSL), Bmal1-KO (Adiponectin-Cre; Bmal1^fx/fx), and Bmal1-KO + NDI1 (Adiponectin-Cre; Bmal1^fx/fx; NDI1^LSL) mice. Uniform manifold approximation and projection (UMAP) plot showing WAT adipocyte nuclei from Control, Bmal1-KO, and Bmal1-KO + NDI1 mice (WAT from 4 mice was pooled for each sample). b, KEGG pathway analysis of downregulated genes in Bmal1-KO vs. Control adipocytes (upper panel) and upregulated genes in Bmal1-KO + NDI1 vs. Bmal1-KO adipocytes (lower panel). c, Heatmap of differentially abundant genes in adipocyte clusters between Control, Bmal1-KO, and Bmal1-KO + NDI1 mice. d, qPCR analysis of indicated genes in visceral WAT from Control, Bmal1-KO, and Bmal1-KO + NDI1 mice (Control n = 12, Bmal1-KO n=12, Bmal1-KO + NDI1 n=7). Data are represented as mean ± SEM. Statistical significance was calculated by one-way ANOVA followed by multiple comparisons in d. *p < 0.05, **p < 0.01, ***p < 0.001.