**Background:** NAD+ levels decline in metabolic tissues during obesity, contributing to insulin resistance, NAFLD, and metabolic dysfunction. While NAD+ boosters are promising therapies, the circadian nature of NAD+ bioavailability has been largely ignored. This study investigates whether timed restoration of NAD+ oscillations can treat diet-induced metabolic disease in mice.
**Methods:** C57Bl/6J mice were fed a high-fat diet (HFD) for 11 weeks. At week 8, a subgroup received daily IP injections of NAD+ (50 mg/kg) at ZT11 (one hour before lights off, HFN group) for 22 days. Controls received saline (HF) or were chow-fed (CD). A separate group received NAD+ at ZT23 (HFN23). Outcomes included body weight, hepatic NAD+ by HPLC, glucose tolerance tests (GTT), insulin tolerance tests (ITT), serum insulin and triglycerides, hepatic histology (Oil-Red-O), protein carbonyls, mtDNA, transcriptomics (Clariom D arrays), western blotting for clock and metabolic proteins, BMAL1 ChIP-qPCR, fatty acid oxidation (14C-palmitate), and mitochondrial respiration (Seahorse).
**Key Results:** NAD+ at ZT11 restored hepatic NAD+ oscillations with a peak at ZT12 (vs. disrupted rhythms in HF). HFN mice lost ~5% body weight by week 11, while HFN23 gained ~2%. GTT and ITT improved significantly only in HFN mice (AUC HF vs HFN at day 10 GTT ZT16: P<0.001; ITT day 10: P<0.005). Serum insulin was markedly lower in HFN vs HF at ZT12-18 (P<0.001). Hepatic triglycerides were reduced and rhythmic in HFN (P<0.05). Oil-Red-O staining showed reduced steatosis in both HFN and HFN23. Protein carbonyls were normalized in HFN. Transcriptomics revealed 1327 day/night DE genes exclusively in HF (enriched for immune processes), while HFN had only 306 exclusive DE genes, with no enrichment for inflammation. At ZT6, 182 shared DE genes between CD-HF and HF-HFN were enriched for immune response; at ZT18, shared genes were enriched for lipid metabolism. GSEA showed IL6-JAK-STAT3 and TGFβ signaling were top hallmarks in HF vs HFN at ZT6; cholesterol homeostasis and MTORC1 signaling at ZT18. AKT phosphorylation (S473) was restored at ZT12 in HFN (P<0.0001 vs HF). AMPK phosphorylation (T172) was rhythmic with a peak at ZT12 in HFN. mTORC1 signaling (p-S6K T389, p-4EBP1) was reduced in HFN. BMAL1 ChIP showed increased recruitment at ZT18 in HFN23 for clock and metabolic genes, corresponding to inverted clock gene expression (phase shift of 10-12 h for Bmal1, Cry1, Per1, Per2, Rev-Erbα). Clock-controlled genes Dbp, Tef, Noct, Nfil3 were also phase-inverted in HFN23. SCN clock gene expression remained intact. Locomotor activity and food intake remained aligned to light-dark cycles in all groups. Fatty acid oxidation genes (Cpt1a, Cpt2, Acox1) were significantly upregulated at ZT18 only in HFN (P<0.05-0.001). Ex vivo 14C-palmitate oxidation and CPT-1-dependent mitochondrial respiration were increased in HFN (P<0.05).
**Clinical Implications:** This study provides the first evidence that the efficacy of NAD+ therapy for metabolic disease is critically time-of-day dependent. NAD+ administered at the onset of the active phase (ZT11) reverses obesity, insulin resistance, and NAFLD through restoration of hepatic NAD+ rhythms, AMPK/AKT/mTOR signaling, and fatty acid oxidation. In contrast, administration at the rest phase (ZT23) inverts the hepatic clock, fails to improve glucose homeostasis, and only partially corrects metabolic parameters. These findings strongly advocate for chronotherapy-based NAD+ supplementation in clinical trials and practice, suggesting that timing of NAD+ booster intake (e.g., morning vs. evening) may determine therapeutic outcomes in metabolic disease.