Hepatic NADH reductive stress underlies common variation in metabolic traits


The cellular NADH/NAD+ ratio is fundamental to biochemistry, but the extent to which it reflects versus drives metabolic physiology in vivo is poorly understood. Here we report the in vivo application of Lactobacillus brevis (Lb)NOX1, a bacterial water-forming NADH oxidase, to assess the metabolic consequences of directly lowering the hepatic cytosolic NADH/NAD+ ratio in mice. By combining this genetic tool with metabolomics, we identify circulating α-hydroxybutyrate levels as a robust marker of an elevated hepatic cytosolic NADH/NAD+ ratio, also known as reductive stress. In humans, elevations in circulating α-hydroxybutyrate levels have previously been associated with impaired glucose tolerance2, insulin resistance3 and mitochondrial disease4, and are associated with a common genetic variant in GCKR5, which has previously been associated with many seemingly disparate metabolic traits. Using LbNOX, we demonstrate that NADH reductive stress mediates the effects of GCKR variation on many metabolic traits, including circulating triglyceride levels, glucose tolerance and FGF21 levels. Our work identifies an elevated hepatic NADH/NAD+ ratio as a latent metabolic parameter that is shaped by human genetic variation and contributes causally to key metabolic traits and diseases. Moreover, it underscores the utility of genetic tools such as LbNOX to empower studies of ‘causal metabolism’.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Fig. 1: LbNOX alters compartment-specific free NADH/NAD+ ratio in hepatocytes.
Fig. 2: In vivo manipulation of the hepatic cytosolic NADH/NAD+ ratio alters hepatic and circulating αHB levels.
Fig. 3: A common GCKR variant is associated with plasma αΗΒ levels in humans.
Fig. 4: Direct oxidation of free hepatic cytosolic NADH improves glucose tolerance and hepatic insulin sensitivity in vivo.
Fig. 5: Many GCKR-associated metabolic traits lie downstream of hepatic NADH reductive stress.

Data availability

All data generated and used in this study are either included in this article (and its Supplementary Information) or are available from the corresponding author on reasonable request.


  1. 1.

    Titov, D. V. et al. Complementation of mitochondrial electron transport chain by manipulation of the NAD+/NADH ratio. Science 352, 231–235 (2016).

    ADS  CAS  PubMed  PubMed Central  Google Scholar 

  2. 2.

    Cobb, J. et al. α-Hydroxybutyric acid is a selective metabolite biomarker of impaired glucose tolerance. Diabetes Care 39, 988–995 (2016).

    CAS  PubMed  Google Scholar 

  3. 3.

    Gall, W. E. et al. α-Hydroxybutyrate is an early biomarker of insulin resistance and glucose intolerance in a nondiabetic population. PLoS ONE 5, e10883 (2010).

    ADS  PubMed  PubMed Central  Google Scholar 

  4. 4.

    Thompson Legault, J. et al. A metabolic signature of mitochondrial dysfunction revealed through a monogenic form of Leigh syndrome. Cell Rep. 13, 981–989 (2015).

    CAS  PubMed  PubMed Central  Google Scholar 

  5. 5.

    Rhee, E. P. et al. A genome-wide association study of the human metabolome in a community-based cohort. Cell Metab. 18, 130–143 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

  6. 6.

    Goodman, R. P., Calvo, S. E. & Mootha, V. K. Spatiotemporal compartmentalization of hepatic NADH and NADPH metabolism. J. Biol. Chem. 293, 7508–7516 (2018).

    CAS  PubMed  PubMed Central  Google Scholar 

  7. 7.

    Trammell, S. A. J. et al. Nicotinamide riboside opposes type 2 diabetes and neuropathy in mice. Sci. Rep. 6, 26933 (2016).

    ADS  CAS  PubMed  PubMed Central  Google Scholar 

  8. 8.

    Cantó, C. et al. The NAD+ precursor nicotinamide riboside enhances oxidative metabolism and protects against high-fat diet-induced obesity. Cell Metab. 15, 838–847 (2012).

    PubMed  PubMed Central  Google Scholar 

  9. 9.

    Madiraju, A. K. et al. Metformin suppresses gluconeogenesis by inhibiting mitochondrial glycerophosphate dehydrogenase. Nature 510, 542–546 (2014).

    ADS  CAS  PubMed  PubMed Central  Google Scholar 

  10. 10.

    Krebs, H. A. The redox state of nicotinamide adenine dinucleotide in the cytoplasm and mitochondria of rat liver. Adv. Enzyme Regul. 5, 409–434 (1967).

    CAS  PubMed  Google Scholar 

  11. 11.

    Ferrannini, E. et al. Early metabolic markers of the development of dysglycemia and type 2 diabetes and their physiological significance. Diabetes 62, 1730–1737 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

  12. 12.

    Hung, Y. P., Albeck, J. G., Tantama, M. & Yellen, G. Imaging cytosolic NADH-NAD+ redox state with a genetically encoded fluorescent biosensor. Cell Metab. 14, 545–554 (2011).

    CAS  PubMed  PubMed Central  Google Scholar 

  13. 13.

    Masia, R. et al. Live cell imaging of cytosolic NADH/NAD+ ratio in hepatocytes and liver slices. Am. J. Physiol. Gastrointest. Liver Physiol. 314, G97–G108 (2018).

    PubMed  Google Scholar 

  14. 14.

    Trammell, S. A. J. et al. Nicotinamide riboside is uniquely and orally bioavailable in mice and humans. Nat. Commun. 7, 12948 (2016).

    ADS  CAS  PubMed  PubMed Central  Google Scholar 

  15. 15.

    King, M. P. & Attardi, G. Human cells lacking mtDNA: repopulation with exogenous mitochondria by complementation. Science 246, 500–503 (1989).

    ADS  CAS  PubMed  Google Scholar 

  16. 16.

    Kruse, S. E. et al. Mice with mitochondrial complex I deficiency develop a fatal encephalomyopathy. Cell Metab. 7, 312–320 (2008).

    CAS  PubMed  PubMed Central  Google Scholar 

  17. 17.

    Jain, I. H. et al. Hypoxia as a therapy for mitochondrial disease. Science 352, 54–61 (2016).

    ADS  CAS  PubMed  PubMed Central  Google Scholar 

  18. 18.

    Machiela, M. J. & Chanock, S. J. LDlink: a web-based application for exploring population-specific haplotype structure and linking correlated alleles of possible functional variants. Bioinformatics 31, 3555–3557 (2015).

    CAS  PubMed  PubMed Central  Google Scholar 

  19. 19.

    Teslovich, T. M. et al. Biological, clinical and population relevance of 95 loci for blood lipids. Nature 466, 707–713 (2010).

    ADS  CAS  PubMed  PubMed Central  Google Scholar 

  20. 20.

    Diabetes Genetics Initiative of Broad Institute of Harvard and MIT et al. Genome-wide association analysis identifies loci for type 2 diabetes and triglyceride levels. Science 316, 1331–1336 (2007).

    Google Scholar 

  21. 21.

    Cornelis, M. C. et al. Genome-wide meta-analysis identifies six novel loci associated with habitual coffee consumption. Mol. Psychiatry 20, 647–656 (2015).

    CAS  PubMed  Google Scholar 

  22. 22.

    Schumann, G. et al. KLB is associated with alcohol drinking, and its gene product β-Klotho is necessary for FGF21 regulation of alcohol preference. Proc. Natl Acad. Sci. USA 113, 14372–14377 (2016).

    CAS  PubMed  Google Scholar 

  23. 23.

    Kilpeläinen, T. O. et al. Genome-wide meta-analysis uncovers novel loci influencing circulating leptin levels. Nat. Commun. 7, 10494 (2016).

    ADS  PubMed  PubMed Central  Google Scholar 

  24. 24.

    Cheung, C. Y. Y. et al. An exome-chip association analysis in Chinese subjects reveals a functional missense variant of GCKR that regulates FGF21 levels. Diabetes 66, 1723–1728 (2017).

    CAS  PubMed  Google Scholar 

  25. 25.

    Orho-Melander, M. et al. Common missense variant in the glucokinase regulatory protein gene is associated with increased plasma triglyceride and C-reactive protein but lower fasting glucose concentrations. Diabetes 57, 3112–3121 (2008).

    CAS  PubMed  PubMed Central  Google Scholar 

  26. 26.

    Speliotes, E. K. et al. Genome-wide association analysis identifies variants associated with nonalcoholic fatty liver disease that have distinct effects on metabolic traits. PLoS Genet. 7, e1001324 (2011).

    CAS  PubMed  PubMed Central  Google Scholar 

  27. 27.

    Raimondo, A., Rees, M. G. & Gloyn, A. L. Glucokinase regulatory protein: complexity at the crossroads of triglyceride and glucose metabolism. Curr. Opin. Lipidol. 26, 88–95 (2015).

    CAS  PubMed  PubMed Central  Google Scholar 

  28. 28.

    Beer, N. L. et al. The P446L variant in GCKR associated with fasting plasma glucose and triglyceride levels exerts its effect through increased glucokinase activity in liver. Hum. Mol. Genet. 18, 4081–4088 (2009).

    CAS  PubMed  PubMed Central  Google Scholar 

  29. 29.

    Dupuis, J. et al. New genetic loci implicated in fasting glucose homeostasis and their impact on type 2 diabetes risk. Nat. Genet. 42, 105–116 (2010).

    CAS  PubMed  PubMed Central  Google Scholar 

  30. 30.

    Holland, W. L. & Summers, S. A. Sphingolipids, insulin resistance, and metabolic disease: new insights from in vivo manipulation of sphingolipid metabolism. Endocr. Rev. 29, 381–402 (2008).

    CAS  PubMed  PubMed Central  Google Scholar 

  31. 31.

    Perry, R. J., Samuel, V. T., Petersen, K. F. & Shulman, G. I. The role of hepatic lipids in hepatic insulin resistance and type 2 diabetes. Nature 510, 84–91 (2014).

    ADS  CAS  PubMed  PubMed Central  Google Scholar 

  32. 32.

    Nakatani, Y. et al. Involvement of endoplasmic reticulum stress in insulin resistance and diabetes. J. Biol. Chem. 280, 847–851 (2005).

    CAS  PubMed  Google Scholar 

  33. 33.

    Shoelson, S. E., Lee, J. & Goldfine, A. B. Inflammation and insulin resistance. J. Clin. Invest. 116, 1793–1801 (2006).

    CAS  PubMed  PubMed Central  Google Scholar 

  34. 34.

    Jang, C. et al. Metabolite exchange between mammalian organs quantified in pigs. Cell Metab. 30, 594–606.e3 (2019).

    CAS  PubMed  PubMed Central  Google Scholar 

  35. 35.

    Titchenell, P. M., Chu, Q., Monks, B. R. & Birnbaum, M. J. Hepatic insulin signalling is dispensable for suppression of glucose output by insulin in vivo. Nat. Commun. 6, 7078 (2015).

    ADS  CAS  PubMed  PubMed Central  Google Scholar 

  36. 36.

    Ader, M. & Bergman, R. N. Peripheral effects of insulin dominate suppression of fasting hepatic glucose production. Am. J. Physiol. 258, E1020–E1032 (1990).

    CAS  PubMed  Google Scholar 

  37. 37.

    Saxena, R. et al. Genetic variation in GIPR influences the glucose and insulin responses to an oral glucose challenge. Nat. Genet. 42, 142–148 (2010).

    CAS  PubMed  PubMed Central  Google Scholar 

  38. 38.

    Kim, W. et al. Polyunsaturated fatty acid desaturation is a mechanism for glycolytic NAD+ recycling. Cell Metab. 29, 856–870.e7 (2019).

    CAS  PubMed  PubMed Central  Google Scholar 

  39. 39.

    Baraona, E. & Lieber, C. S. Effects of ethanol on lipid metabolism. J. Lipid Res. 20, 289–315 (1979).

    CAS  PubMed  Google Scholar 

  40. 40.

    Desai, B. N. et al. Fibroblast growth factor 21 (FGF21) is robustly induced by ethanol and has a protective role in ethanol associated liver injury. Mol. Metab. 6, 1395–1406 (2017).

    CAS  PubMed  PubMed Central  Google Scholar 

  41. 41.

    Hondares, E. et al. Hepatic FGF21 expression is induced at birth via PPARα in response to milk intake and contributes to thermogenic activation of neonatal brown fat. Cell Metab. 11, 206–212 (2010).

    CAS  PubMed  PubMed Central  Google Scholar 

  42. 42.

    Xu, J. et al. Fibroblast growth factor 21 reverses hepatic steatosis, increases energy expenditure, and improves insulin sensitivity in diet-induced obese mice. Diabetes 58, 250–259 (2009).

    CAS  PubMed  PubMed Central  Google Scholar 

  43. 43.

    Schlein, C. et al. FGF21 lowers plasma triglycerides by accelerating lipoprotein catabolism in white and brown adipose tissues. Cell Metab. 23, 441–453 (2016).

    CAS  PubMed  Google Scholar 

  44. 44.

    Lonsdale, J. et al. The Genotype-Tissue Expression (GTEx) project. Nat. Genet. 45, 580–585 (2013).

    CAS  Google Scholar 

  45. 45.

    Kim, J. K. Hyperinsulinemic-euglycemic clamp to assess insulin sensitivity in vivo. Methods Mol. Biol. 560, 221–238 (2009).

    CAS  PubMed  Google Scholar 

  46. 46.

    Lu, W., Wang, L., Chen, L., Hui, S. & Rabinowitz, J. D. Extraction and quantitation of nicotinamide adenine dinucleotide redox cofactors. Antioxid. Redox Signal. 28, 167–179 (2018).

    CAS  PubMed  PubMed Central  Google Scholar 

  47. 47.

    Fiehn, O. Metabolomics by gas chromatography-mass spectrometry: the combination of targeted and untargeted profiling. Curr. Protoc. Mol. Biol. 114, 30.4.1–30.4.32 (2016).

    Google Scholar 

  48. 48.

    International HapMap Consortium et al. The International HapMap Project. Nature 426, 789–796 (2003).

    Google Scholar 

  49. 49.

    Chang, C. C. et al. Second-generation PLINK: rising to the challenge of larger and richer datasets. Gigascience 4, 7 (2015).

    PubMed  PubMed Central  Google Scholar 

  50. 50.

    1000 Genomes Project Consortium et al. A global reference for human genetic variation. Nature 526, 68–74 (2015).

    Google Scholar 

Download references


We thank J. Avruch and members of the Mootha laboratory for valuable discussions and feedback. R.P.G. was supported by the US National Institutes of Health (NIH) grant F32DK111132 and NIH K08DK1158811. R.P.G. and R.M. were both supported by a Pinnacle Research Award from the American Association for the Study of Liver Diseases (AASLD). O.S.S. was supported by NIH F32GM133047.  Y.-H.H.H. was supported by NIH T32DK110919. Part of this study was performed at the National Mouse Metabolic Phenotyping Center at University of Massachusetts supported by NIH 5U2C-DK093000 (to J.K.K.) This work was supported in part by a gift from the Marriott Mitochondrial Disorders Collaborative Research Network and grants from the NIH (R35GM122455 and TR01GM099683). V.K.M. is an investigator of the Howard Hughes Medical Institute.

Author information




R.P.G. and V.K.M. designed the study, which was supervised by V.K.M. R.P.G. and R.M. performed the Peredox experiments, which were supervised by G.Y. R.P.G., A.L.M., A.P. and O.G. performed the in vivo mouse experiments, tissue analyses and hepatocyte experiments. R.P.G., R.S., O.S.S., H.S., C.B.C. and A.D. processed and analysed in vitro and in vivo mouse metabolomics data. H.L.N., S.S. and J.K.K. processed and analysed the hyperinsulinaemic–euglycaemic clamp experiments. Y.-H.H.H. and J.N.H. contributed to the human genetic data analyses. R.P.G. wrote the initial manuscript, which was further edited by V.K.M.

Corresponding author

Correspondence to Vamsi K. Mootha.

Ethics declarations

Competing interests

V.K.M. is an inventor on the patent PCT/US2016/045015 filed by Massachusetts General Hospital on the use of the LbNOX technology as protein prosthesis for mitochondrial diseases or conditions. V.K.M. and R.P.G. are inventors on a patent provisionally filed by Massachusetts General Hospital on modulating hepatic reductive stress with chemicals.

Additional information

Peer review information Nature thanks Joshua D. Rabinowitz, Charles M. Brenner and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data figures and tables

Extended Data Fig. 1 The effects of LbNOX expression in primary hepatocytes.

a, Dose-dependent adenovirus-mediated expression of LbNOX in primary hepatocytes at 24 h. A representative western blot from two independent experiments is shown. b, Effect on basal or antimycin + rotenone-insensitive respiration with LbNOX or mitoLbNOX. n = 3. c, Effect of LbNOX on free cytosolic NADH/NAD+ as measured by Peredox with increasing alcohol concentrations. dg, Whole cell NAD+ (d), NADH (e), NADH/NAD+ (f) and NADH + NAD+ (g). n = 6 independent hepatocyte isolations. Nominal P values were determined using paired, two-sided Student’s t-tests between hepatocyte isolations (b, dg) or unpaired, two-sided Student’s t-test for Peredox experiments (c). Data are mean  ± s.e.m. Source Data

Extended Data Fig. 2 NR and LbNOX have distinct effects on pyridine dinucleotide pool sizes and redox ratios.

af, Relative total cellular NAD+ (a), NADH (b), NADH/NAD+ (c), secreted lactate/pyruvate ratio (d), secreted βHB/AcAc ratio (e) and αHB levels (f) in primary hepatocytes with or without NR supplementation and LbNOX. g, h, Effect of pyruvate (Pyr), LbNOX expression or NR on the inhibition of HeLa cell proliferation (g) and total NAD+ levels (h) by piericidin 4 days after seeding. Data are mean ± s.e.m. from n = 7 (ac) or 10 (df) independent hepatocyte isolation, or 3 independent HeLa cell experiments (g and h). Nominal P values were determined using paired (between hepatocyte isolations; af) or unpaired (HeLa cells; g and h) two-sided Student’s t-tests. Source Data

Extended Data Fig. 3 αHB levels and the hepatic NADH/NAD+ ratio are elevated in the Ndufs4-KO mouse model of mitochondrial disease.

a, Plasma αHB levels. b, Hepatic NADH/NAD+ ratio. n = 3 mice in each group for NADH/NAD+ measurements and n = 7 mice for αHB measurements. Data are mean ± s.e.m. P values were determined using one-sided Student’s t-test. WT, wild type. Source Data

Extended Data Fig. 4 A common GCKR variant increases hepatic cytosolic NADH reductive stress.

a, Gene structure, variant location, variant linkage disequilibrium blocks and haplotype frequency for the GCKR risk haplotype from the 1000 Genomes Project50. AFR, African; EAS, East Asian; EUR, European; SAS, South Asian. b, Enrichment of redox-sensitive metabolites in reported metabolite associations by loci from ref. 5. P values were determined from one-tailed Fisher’s exact test as described in the Methods. c, Effects of overexpression of Gckr or Gckr p.P446L in mouse primary hepatocytes. Data are mean ± s.e.m. Nominal P values were determined using paired, two-sided Student’s t-tests between n = 5 independent hepatocyte isolations. Source Data

Extended Data Fig. 5 Effects of LbNOX expression on metabolic parameters during hyperinsulinaemic–euglycaemic clamp.

ak, Effects of LbNOX expression in HFD-fed mice using a 2.5 mU min−1 kg−1 insulin infusion during hyperinsulinaemic–euglycaemic clamp on body weight (a), basal glucose levels (b), clamp glucose levels (c), glucose infusion rate (d), whole-body glucose turnover (e), whole-body glycolysis (f), whole-body glycogen synthesis (g), lean mass (h), fat mass (i), WAT glucose uptake (j) and skeletal muscle glucose uptake (k). P values were determined using two-sided Student’s t-test. Data are reported as mean ± s.e.m. from n = 8 (luciferase) or n = 9 (LbNOX) mice. Source Data

Extended Data Fig. 6 Hepatic diacyl-glycerols and ceramides in DIO mice with LbNOX expression.

a, b, Effect of LbNOX expression of hepatic diacyl-glycerol (a) and ceramide (b) content in DIO mice. Data are reported as mean ± s.e.m. from n = 9 mice. P values were determined using two-sided Student’s t-test. Source Data

Extended Data Fig. 7 LbNOX improves hepatic insulin resistance in vivo independent of hepatic insulin signalling.

ac, Western blots of liver lysate from DIO mice 15 min after an intraperitoneal injection of saline or 2 U/kg insulin (a) with relative pS474 AKT/total AKT (b) and relative pT308 AKT/total AKT (c). n = 3 from representative western blots from 2 independent experiments. d, Transcriptional FOXO1 targets G6pc, Pepck1 and Pc in DIO mice with LbNOX or luciferase. n = 6. e, Western blots of liver lysates at the end of hyperinsulinaemic–eugylcaemic clamps. n = 3 representative of n = 8 (luciferase) and 9 (LbNOX). f, g, Relative pS474 AKT/total AKT and pT308 AKT/total AKT (f) and transcriptional FOXO1 targets G6pc, Pepck1 and Pc (g). n = 8 (luciferase) and 9 (LbNOX). h, Crossover analysis of relative abundance of gluconeogenic intermediates at the end of hyperinsulinaemic–euglycaemic clamps. Top, LbNOX versus Luc mice are compared. Bottom, samples are divided by high or low liver lactate/pyruvate (L/P) ratios and compared. *P < 0.05, **P < 0.01, using two-sided Student’s t-test. BPG, 1,3-bisphosphoglycerate; DAP, dihydroxyacetone phosphate; FBP, fructose 1,6-bisphosphate; F1P, fructose 1-phosphate; F6P, fructose 6-phosphate; G6P, glucose 6-phosphate; MAL, malate; PEP, phosphoenolpyruvate; 2PG, 2-phosphoglycerate; 3PG, 3-phosphoglycerate; PYR, pyruvate. i, Western blots and relative protein levels of GAPDH and triosephosphate isomerase (TPI) at the end of the insulin clamp. n = 8 (luciferase) and 9 (LbNOX). Data are reported as mean ± s.e.m. Source Data

Extended Data Fig. 8 NAD(P)(H) levels in LbNOX versus luciferase livers at the end of hyperinsulinaemic–euglycaemic clamps.

af, The relative abundance of total NAD+ (a), total NADH (b), total NADH/NAD+ (c), total NADP+ (d), total NADPH (e) and total NADPH/NADP+ (f). n = 4 mice for each group. P values were determined by one-sided Student’s t-test. Data are reported as mean ± s.e.m. Source Data

Extended Data Fig. 9 Metabolic origins and fate of αHB.

α-AB, α-aminobutyrate; BCKDH, branched-chain α-keto acid dehydrogenase complex; CGL, cystathionine γ-lyase; PDH, pyruvate dehydrogenase; S/TDH, serine/threonine dehydratase; TCA, tricarboxylic acid.

Supplementary information

Supplementary Figure

Supplementary Figure 1: Original westerns blots.

Reporting Summary

Supplementary Table

Supplementary Table 1: Literature summary of traits associated with GCKR genetic variants.

Supplementary Table

Supplementary Table 2: Hepatocyte metabolite screen related to Figure 1f.

Supplementary Table

Supplementary Table 3: Hyperinsulinemic-euglycemic clamp of LbNOX mice on high-fat-diet.

Source data

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Goodman, R.P., Markhard, A.L., Shah, H. et al. Hepatic NADH reductive stress underlies common variation in metabolic traits. Nature 583, 122–126 (2020). https://doi.org/10.1038/s41586-020-2337-2

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.