Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Nov 1.
Published in final edited form as: Diabetes Obes Metab. 2024 Aug 19;26(11):5272–5282. doi: 10.1111/dom.15879

Nicotinamide N-methyltransferase Inhibition Mitigates Obesity-Related Metabolic Dysfunctions

JoAnne J Babula 1, Dinh Bui 1, Heather L Stevenson 2, Stanley J Watowich 1,3, Harshini Neelakantan 1,*
PMCID: PMC11622326  NIHMSID: NIHMS2015151  PMID: 39161060

Abstract

Aims:

Nicotinamide N-methyltransferase (NNMT) is a critical regulator of cellular metabolism and energy homeostasis and a validated novel target for obesity-linked Type 2 diabetes. This study assessed the effects of a small-molecule NNMT inhibitor, 5A1MQ, on body composition, metabolic parameters, fatty liver pathologies, and circulating biomarkers in diet-induced obese (DIO) mice and characterized its plasma pharmacokinetics (PK) and tissue distribution in vivo.

Materials and Methods:

DIO mice were administered vehicle or 5A1MQ once-daily for 28 days. Longitudinal measures of body composition, blood glucose and plasma insulin levels, and terminal measures of liver histopathology and serum markers were evaluated. Plasma and tissue PK were established in age- and strain-matched mice after intravenous, oral, and subcutaneous dosing of 5A1MQ.

Results:

5A1MQ treatment dose-dependently limited body weight and fat mass gains, improved oral glucose tolerance and insulin sensitivity, and suppressed hyperinsulinemia in DIO mice. Liver histology from 5A1MQ-treated DIO mice showed attenuated hepatic steatosis and macrophage infiltration, and correspondingly reduced liver weight, size, and triglyceride levels. 5A1MQ treatment normalized circulating levels of alanine transaminase, aspartate transaminase, and ketone bodies, supporting an overall improvement in liver and metabolic functions. The pharmacodynamic effects of 5A1MQ were further corroborated by its high systemic exposures and effective distribution to metabolically active tissues, including adipose, muscle, and liver, following subcutaneous dosing to mice.

Conclusions:

This work validates NNMT inhibition as a viable pharmacological approach to ameliorate metabolic imbalances and improve liver pathologies that develop with obesity.

Introduction

Obesity, characterized by a body mass index (BMI) ≥ 30 kg/m2 and extensive body fat accumulation, is an emerging global crisis with continually increasing prevalence1,2. Among the hallmarks of obesity is the disruption of metabolic homeostasis, including glucose handling and metabolism leading to reduced insulin sensitivity3. Obese non-diabetic patients exhibit similar levels of insulin resistance as non-obese diabetic patients4, indicating a dominant effect of obesity in blunting systemic insulin response. Consequently, increased insulin resistance underlies many chronic diseases5, including comorbidities such as hypertension, dyslipidemia, metabolic dysfunction associated steatotic liver disease (MASLD), and cancers6.

Nicotinamide N-methyltransferase (NNMT) has been recognized and advanced as a therapeutic target for obesity and Type 2 diabetes (T2D)7–11. NNMT primarily methylates intracellular nicotinamide to 1-methylnicotinamide (1-MNA) using the cofactor S-Adenosyl methionine (SAM)12 and is highly expressed in the liver13. In high-fat diet (HFD)-fed mice, NNMT expression is upregulated in adipose and liver tissues7, with greater obesity propensity correlating with increased NNMT expression and activity in adipose tissue7,14. In humans with T2D, insulin sensitivity is inversely correlated with adipose tissue NNMT expression and circulating levels of 1-MNA10,11. Furthermore, in morbidly obese patients (BMI ≥ 40 kg/m2), altered methylation patterns of several genes related to metabolic processes, proliferation, inflammation, and extracellular matrix remodeling are strongly associated with NNMT expression15. Thus, there exists strong evidence for NNMT’s pivotal role in regulating energy metabolism through the glucose-insulin axis, lipid metabolism7,12,16, and epigenetic effectors17.

NNMT knockdown in adipose and liver tissues (but not brown adipose tissue and kidney) using antisense oligonucleotides protected HFD-fed mice from obesity and insulin resistance phenotypes7. Similar improvements in body weight, adiposity, and insulin sensitivity have been reported with whole-body NNMT knockout in mice fed HFD18. Our group previously demonstrated that treatment with the NNMT inhibitor 5-amino-1-methylquinolium (5A1MQ) limited body weight and white adipose tissue mass gains, decreased adipocyte size, and lowered total plasma cholesterol levels in diet-induced obese (DIO) mice maintained on a HFD8. Additionally, we recently demonstrated that treating obese mice with 5A1MQ and a lean diet accelerated and enhanced body weight and whole-body adiposity losses, dramatically improving associated liver pathologies (e.g., steatosis) in obese mice compared to lean diet alone19. This work extends these investigations by demonstrating 5A1MQ’s effects on prediabetic phenotypes (i.e., hyperinsulinemia, impaired glucose tolerance), metabolic markers, and impaired liver pathologies in DIO mice and establishing 5A1MQ’s pharmacokinetic (PK) profile.

Methods

Chemicals

The NNMT inhibitor, 5A1MQ, was synthesized as previously described20. 5A1MQ was prepared weekly by dissolving in 0.9% sterile saline and filtering through a sterile 0.2 μm membrane. Solutions were stored at 4⁰C. LC-MS/MS standards were purchased from commercial suppliers, including 1-MNA from Cayman Chemical Company (Ann Arbor, MI).

Animals

Subcutaneous (SC) PK and tissue distribution studies in mice were conducted by XenoBiotic Laboratories, Inc. (Cranbury, NJ). Intravenous (IV) and oral (PO) dosing PK and DIO efficacy studies were conducted by WuXi AppTec Co., Ltd. (Nantong, China). Studies were conducted in accordance with each sites’ Institutional Animal Care and Use Committee (IACUC) protocols and national and local regulations, including the National Research Council Guide for the Care and Use of Laboratory Animals21 and the Animal Welfare Act.

For the DIO study, ~18-week-old male C57BL/6 mice maintained on a HFD (D12492i, Research Diets Inc.) for at least 12 weeks were purchased from Gempharmatech (Nanjing, China). Mice were singly housed upon arrival, continued HFD and water ad libitum, and acclimated for 2–4 weeks until body weights stabilized (~40–50 g). For the acute IV and PO PK studies, ~8-week-old male C57BL/6 mice were purchased from LingChang Biotech Co. Ltd. (Shanghai, China). Mice were fasted for ~12 h before dosing, and food was returned 4 h post-dosing. For the SC PK and biodistribution studies, 18–20-week-old male C57BL/6 mice were purchased from Hilltop Lab Animals, Inc. (Scottdale, PA, USA) and maintained on a certified diet (Diet #5002, LabDiet) and water ad libitum. Mice were group-housed, acclimated for at least 3 days before study started, and maintained in colony rooms at 21–23°C, 45–50% relative humidity, and 12 h light-dark cycles.

Efficacy of 5A1MQ in DIO mice

DIO mice were acclimated to handling for a week and received sham saline SC injections for three days before randomization into control (Group 1, n=8; sterile 0.9% saline), low-dose 5A1MQ (Group 2, n=8; 10 mg/kg/day), and high-dose 5A1MQ (Group 3, n=8; 32 mg/kg/day) groups. Randomization ensured similar mean (± standard deviation) baseline body composition (e.g., total weight, lean and fat masses) across groups. Treatments continued for 30 days as a once-daily SC injection, administered between ~4:00–5:00 PM before the dark cycle began.

Body weight, Whole-body composition, and food intake

Longitudinal body weight and food intake were measured twice weekly over the study period. Whole-body composition (body weight, fat mass, lean mass, and free and total water) was measured on days −1, 8, 15, 23, and 29 using an EchoMRI™-100 analyzer.

Non-fasted blood glucose and plasma insulin measurements

Non-fasted blood glucose and plasma insulin were measured on days −7, 0, 7, 14, 21, and 28 using tail blood collected ñoon-1:00 PM. Blood glucose and plasma insulin were measured using OneTouch® glucose test strips (LifeScan Inc., Shanghai, China) and EZRMI-13K ELISA kit (Merck-Millipore, Billerica, MA), respectively.

Oral Glucose Tolerance Test (OGTT)

The OGTT was performed on Day 21. After non-fasted blood collections, animals were fasted for 4 h with a single administration of saline, low-dose, or high-dose 5A1MQ at 3 h post-fasting. At 1 h post-dosing, 2 g/kg bolus glucose was orally administered, and blood glucose were measured at 0 min (pre-bolus glucose) and 15, 30, 60, and 120 min-post glucose. Plasma insulin was measured in 0- and 15-min blood samples.

Terminal blood and tissue collection

On Day 30, animals received an injection of vehicle or drug one hour before being euthanized. Animals were fasted for 4 h, sacrificed by CO2 inhalation, and tissues were harvested. Blood samples (~700 μL) were collected by cardiac puncture; 300 μL was processed for Complete Blood Count (CBC) and 150 μL for serum chemistry analysis. The whole liver from each animal was collected and weighed. The left lobe was divided into ~30 mg sections, which were flash-frozen for triglyceride (TG) analysis and bioanalytical evaluation of 5A1MQ and 1-MNA or fixed in formalin and paraffin-embedded for hematoxylin and eosin (H&E) staining and F4/80 immunohistochemistry (see Supplementary Information).

Bioanalysis of plasma and liver tissues

Terminal blood samples were processed to obtain ~70 μL of plasma for bioanalytical quantitation of 5A1MQ and 1-MNA using appropriate LC-MS/MS methods (see Supplementary Information).

Acute intravenous and oral dosing PK of 5A1MQ in mice

Mice were randomized into IV (Group 1, n=3; 5 mg/kg 5A1MQ) and PO ( Group 2, n=3; 30 mg/kg 5A1MQ) dosing groups. Blood samples were collected by cross sampling via the saphenous veins at 0.0833 (IV group), 0.25, 0.5, 1, 2, 4, 8, 12 (PO group), and 24 hours post-dosing (both groups) and processed for plasma using K2EDTA tubes. Samples were stored at −70°C until bioanalysis (see Supplementary Information).

Acute and repeat SC dosing PK and tissue distribution profiling of 5A1MQ in mice

Non-fasted mice were randomized into acute SC (Group 1, n=6; 25 mg/kg 5A1MQ) and repeated SC (Group 2, n=12; 25 mg/kg 5A1MQ, once-daily for 5 days) dosing groups. Using a cross-sampling design, 40 μL blood samples were collected via the saphenous veins at 0.25, 0.5, 1, 2, 4, 8, 12, 24, 32, and 48 h post-acute dosing in Group 1 and post-5 days of once-daily dosing in Group 2. Blood samples were collected via cardiac puncture at 1 h, 4 h, and 8 h (n=3 mice/timepoint) post-dosing on Day 5 from subsets of Group 2 mice, with the epididymal white adipose tissue (EWAT), liver, and quadriceps muscle terminally collected. Processed plasma and tissue samples were stored at −70°C for LC-MS/MS analysis (see Supplementary Information).

Statistical Analysis

Statistical analyses were performed using Graphpad Prism (v10.0.2) and RStudio (v4.1.0) with an experiment-wise error rate of α = 0.05. Datasets were assessed for normality using the Shapiro-Wilk test and homoscedasticity using the Brown-Forsythe and Barlett’s (GraphPad) or Levene’s (RStudio) tests, followed by appropriate parametric or non-parametric statistical tests. Longitudinal outcomes were analyzed using a repeated measure two-way analysis of variance (ANOVA) with Geisser-Greenhouse correction where sphericity was not met (confirmed by Mauchly’s test; RStudio) and Benjamini, Krieger, and Yuketieli multiple comparisons posthoc test. Terminal datasets were analyzed using an ordinary one-way ANOVA with Tukey’s multiple comparisons posthoc test, the Kruskal-Wallis test with Dunn’s multiple comparisons posthoc test, or the Brown-Forsythe ANOVA test with Dunnett’s T3 posthoc test, as appropriate to determine treatment effects. As needed, datasets were log-transformed. One-way or two-way analysis of covariance (ANCOVA; RStudio) was performed to analyze the effect of treatment on food intake and metabolic measures using body weight and fat mass as covariates.

Results

5A1MQ limited body weight and fat mass gain without altering food intake in DIO mice

Repeated 5A1MQ treatment dose-dependently suppressed body weight gain in DIO mice [Treatment effect: F(2,21) = 3.734, p<0.05]. By study termination, mice treated with high-dose 5A1MQ displayed a weight gain of 0.9 g, relative to weight gains of 5.4 g and 5.2 g in control and low-dose 5A1MQ groups, respectively (Fig. 1A; p<0.01, high-dose 5A1MQ vs. controls and low-dose 5A1MQ groups). There was no main effect of 5A1MQ treatment on food intake in DIO mice (Fig. 1B; Treatment effect: F(2,21) = 2.186, p = n.s.), and the terminal cumulative food intake analyzed using terminal body weight as a covariate (one-way ANCOVA) also verified no main effect of 5A1MQ treatment on food intake after adjusting for body weight measures [F(1, 21) = 1.31, p = 0.265]. For longitudinal fat mass measures, there was a significant main effect of treatment × time interaction [F(8,84) = 6.56, p<0.0001], suggesting significant longitudinal treatment effects. Consequently by study termination, while absolute fat mass increased by 4.7 g in control and 5A1MQ low-dose groups, respectively, a limited 1.3 g fat mass gain was noted in the 5A1MQ high-dose group (Fig. 1C; p<0.01, high-dose 5A1MQ vs. control and low-dose 5A1MQ groups). There were no significant main treatment effect on longitudinal lean mass measures (Fig. 1D). For detailed statistical results, see Supplementary Information (Table S7).

Figure 1. 5A1MQ dose-dependently limits body weight and fat mass gain without altering food intake and lean mass in DIO mice.

Figure 1.

Open in a new tab

Longitudinal body weight (A), cumulative food intake (B), weekly fat mass (C) and lean mass (D) measured in control, low-dose, and high-dose 5A1MQ treated DIO mice. All data represent mean ± SEM (n=8 per group) and analyzed by repeated measures two-way ANOVA followed by Benjamini, Krieger and Yuketieli’s multiple comparison test. *p<0.05, **p<0.01 vs. control group and ^p<0.05, ^^p<0.01, vs. 10 mg/kg mg/kg 5A1MQ group; significant p values reported are after false discovery rate corrections.

5A1MQ improved fed-state plasma insulin levels and glucose tolerance in DIO mice

5A1MQ treatment did not alter weekly measures of fed-state blood glucose (Fig. 2A) but significantly suppressed rising plasma insulin levels representative of hyperinsulinemia in DIO mice [Treatment effect: F(2,21) = 5.436, p<0.05] (Fig. 2B). Plasma insulin levels increased by 112% and 46% from baseline (Day 0) to Day 28 in control and 5A1MQ low-dose groups, respectively, while decreasing by 9% in the 5A1MQ high-dose group (Fig. 2B). After adjusting for terminal body weight [F(2,20) = 3.664, p = 0.044] and fat mass [F(2,20) = 4.954, p = 0.018], there remained a main effect of 5A1MQ treatment on fed-state insulin measures suggesting body weight and fat mass-independent effects of the drug on circulating insulin. Similarly, after adjusting for body weight [F(2,104) = 7.55, p = 0.0008]and fat mass [F(2,104) = 8.58, p = 0.0003] as covariates, a two-way ANCOVA on the oral glucose tolerance data indicated significant main effect of 5A1MQ treatment (Fig. 2C). High-dose 5A1MQ treatment significantly decreased circulating glucose levels at 15- and 30-min post-bolus glucose compared to the other groups (Fig. 2C; 15-min post-bolus glucose: p<0.0001 and 30-min post-bolus glucose: p<0.05, high-dose 5A1MQ vs. control and low-dose 5A1MQ groups). Correspondingly, 5A1MQ high-dose treatment suppressed plasma insulin at 15 min post-glucose loading compared to the insulin levels noted in the other groups (p<0.0001, high-dose 5A1MQ vs. control and low-dose 5A1MQ; Fig. 2D). For detailed statistical results, see Supplementary Information (Table S7).

Figure 2. 5A1MQ dose-dependently reduced fed-state plasma insulin and improved oral glucose tolerance and insulin sensitivity in DIO mice.

Figure 2.

Open in a new tab

Weekly fed-state blood glucose (A) and plasma insulin (B) and blood glucose (C) and plasma insulin (D) measured during oral glucose tolerance test (OGTT performed on Day 21) in control, low-dose, and high-dose 5A1MQ treated DIO mice. All data represent mean ± SEM (n=8 per group, A-D) and analyzed by repeated measures two-way ANOVA (A-D) and ANCOVA (C); Benjamini, Krieger and Yuketieli’s multiple comparison tests followed by two-way ANOVA analysis. *p<0.05, **p<0.01, ****p<0.0001 vs. control group and ^p<0.05, ^^p<0.01, ^^^^p<0.0001 vs. 10 mg/kg/day 5A1MQ group; significant p values reported are after false discovery rate corrections.

5A1MQ protected against liver insults and steatosis in DIO mice

At study completion, control DIO mice exhibited significantly enlarged livers (average liver weight 2.1 g; Fig. 3A–B), whereas 5A1MQ treatment produced significant dose-dependent decreases in terminal liver weights, even after adjusting for terminal body weights [F(2,20) = 4.953, p=0.018]. Liver weights in the high-dose group averaged 1.4 g, which was significantly lower than the liver weights in the other groups (p<0.001, 5A1MQ high-dose group vs. control and low-dose groups; Fig. 3A). Consistent with the effect of 5A1MQ on liver weight, there were significant main effects of treatment on whole-liver cross-sectional area [Treatment effect: F(2,21) = 9.031, p=0.0015] and hepatocyte cell count [Treatment effect: F(2,21) = 15.71, p<0.0001] (Fig. 3B). 5A1MQ treatment also significantly reduced liver TG levels [Treatment effect: F(2,8.746) = 12.6, p=0.0169] (Fig. 3C).

Figure 3. 5A1MQ dose-dependently reduced liver weight and size, attenuated hepatic steatosis, and improved inflammation, ballooning, and NAFLD activity score (NAS) in DIO mice.

Figure 3.

Open in a new tab

Whole-liver weight (A), cell count and sectional area (B), liver triglyceride [TG] levels (C), representative H&E-stained images of liver sections (D), macrovesicular and microvesicular liver steatosis scores (%) (E), inflammation, ballooning, and NAS scores (F), F4/80-stained areas quantified by Image J software (G), and representative images of liver sections stained for F4/80 (macrophage marker) (H) in control, low-dose, and high-dose 5A1MQ treated DIO mice. For panels D and H, scale bar = 50μm. All data represent mean ± SEM (n = 8 per group, A-C and E-G). Data were analyzed by one-way ANOVA followed by Tukey’s multiple comparisons test or Kruskal-Wallis test with Dunn’s multiple comparisons posthoc test and using ANCOVA where applicable; data were log-transformed where datasets were non-normally distributed. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001 vs. control group and ^p<0.05, ^^p<0.01 vs. 10 mg/kg/day 5A1MQ group.

Histological examination of liver sections revealed extensive steatosis in the control group compared to the 5A1MQ treatment groups (Fig. 3D). 5A1MQ treatment resulted in dose-related reductions in hepatic steatosis relative to controls [Kruskal-Wallis test = 12.31, p = 0.0021], with a 73% reduction in microvesicular steatosis compared to controls (Fig. 3E; p<0.01, high-dose 5A1MQ vs. control). Control DIO mice also showed mild lobular inflammation and hepatocellular ballooning, not exceeding grade 1, as observed in similar DIO models22 (Fig. 3F). 5A1MQ treatment significantly reduced liver lobular inflammation (Kruskal-Wallis test = 10.91, p = 0.0043; p<0.01, high dose 5A1MQ vs. control) and ballooning (Kruskal-Wallis test = 7.355, p = 0.0253; p<0.05, high dose 5A1MQ vs. control) (Fig. 3F), with an overall dose-dependent improvement in the non-alcoholic fatty liver disease (NAFLD) activity score (NAS) score [Treatment effect: F(2,21) = 15.48, p<0.0001; p<0.01 and p<0.0001, high dose 5A1MQ vs. low dose group and control, respectively] (Fig. 3F). Furthermore, immunohistochemical detection of F4/80 inflammatory marker in liver sections revealed significantly lowered F4/80 signal in the 5A1MQ high-dose group compared to controls [Kruskal-Wallis test = 7.246, p = 0.0267; p<0.05, high dose 5A1MQ vs. control] (Fig. 3G–H).

5A1MQ improved serum chemistry and hematological parameters in DIO mice

Consistent with 5A1MQ’s effect on liver TG levels, significantly lower serum TG levels were noted with 5A1MQ treatment [Treatment effect: F(2,21) = 8.278, p = 0.0022; p<0.01, high-dose 5A1MQ vs. control; Fig. 4A] in DIO mice. Also observed were trends in 5A1MQ-mediated reductions in serum lipid markers, including free fatty acid, total cholesterol, and high- and low-density lipoproteins (Table S1). DIO mice treated with 5A1MQ exhibited dose-related decreases in serum ALT [Kruskal-Wallis test = 13.40, p = 0.0012; p<0.001, high-dose 5A1MQ vs. control] and AST [Kruskal-Wallis test = 10.70, p = 0.0047; p<0.01, low-dose 5A1MQ vs. control] levels (Fig. 4B). Control DIO mice had high serum ketone body levels (Fig. 4C), as seen typically after several weeks on HFD and the initiation of hepatic insulin resistance in DIO mice23. This dose-dependently decreased with 5A1MQ treatment [F(2,20) = 14.29, p=0.000; p<0.0001, high dose 5A1MQ vs. control; p<0.05, high dose 5A1MQ vs. low dose 5A1MQ] (Fig. 4C). Additionally, 5A1MQ treatment significantly lowered the levels of two circulating enzymes sensitive to metabolic disruptions, CK [F(2,21) = 5.854, p=0.0095] and LDH [F(2,21) = 8.489, p=0.0020]; effects were only statistically significant at the low but not the high dose of 5A1MQ (Fig. 4D; p<0.05, low-dose 5A1MQ vs. control). Lastly, 5A1MQ treatment significantly reduced total white blood cell count [F(2,21) = 5.237, p = 0.0142; p<0.05, high-dose 5A1MQ vs. control] that was driven by large reductions in lymphocytes (p<0.05, 31% decrease in high-dose 5A1MQ vs. control) and monocytes (p<0.001, 50% decrease in high-dose 5A1MQ vs. control; p<0.01, low-dose 5A1MQ vs. control) (Table S2).

Figure 4. 5A1MQ corrects serum chemistry markers and ketone body levels in DIO mice.

Figure 4.

Open in a new tab

Key metabolic markers and enzymes were measured in the serum of control, low-dose, and high-dose 5A1MQ treated DIO mice, including TG (A), ALT and AST (B), ketone bodies (C), and CK and LDH (D). All data represent mean ± SEM (n=8 per group) and are analyzed by either Kruskal-Wallis with Dunn’s multiple comparisons posthoc test or one-way ANOVA followed by Tukey’s multiple comparisons test as appropriate. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001 vs. control group and ^p<0.05 vs. 10 mg/kg/day 5A1MQ group. TG: Triglyceride; ALT: Alanine transaminase; AST: Aspartate transaminase; CK: Creatine kinase; LDH: Lactate dehydrogenase.

5A1MQ exhibited favorable PK via non-parenteral routes of administration to mice

Following IV dosing of mice at 5 mg/kg 5A1MQ, the Cmax and AUC0-∞ were 2,009 ng/mL and 825 ng*h/mL, respectively, with high plasma clearance (101 mL/min/kg), large estimated steady-state volume of distribution of (39 L/kg), and long terminal half-life (6.3 h) (Fig. 5A). With 25 mg/kg 5A1MQ single and repeated once-daily SC dosing for 5 days, Cmax reached 7,010 ng/mL and 5,130 ng/mL, respectively, by ~15 minutes (Tmax). Plasma AUC0-∞ were 7,031 ng*h/mL and 4,687 ng*h/mL, and the estimated terminal half-lives (T1/2) were 13.3 h and 12.8 h for acute and repeated SC 5A1MQ dosing, respectively (Fig. 5B). 5A1MQ concentrations in the plasma remained above IC50 concentration (5A1MQ mouse NNMT IC50 = 78 ng/mL) for ~2–4 h post-parenteral dosing (Fig. 5A–B).

Figure 5. 5A1MQ elicits substantial systemic exposure when administered non-parenterally but has poor oral bioavailability in mice.

Figure 5.

Open in a new tab

5A1MQ plasma concentration-time profiles after single intravenous (IV; 5 mg/kg) and oral (PO; 30 mg/kg) dosing (A) and single and multiple subcutaneous (SC; once-daily [QD]x1 or QDx5, 25 mg/kg) dosing of adult male, C57Bl/6J mice (B). All data in A-B represent mean ± SD (n=3–6 at each timepoint). IC50 values represent 5A1MQ IC50 measured against mouse NNMT enzyme in an in vitro biochemical assay (A-B).

In contrast to the high systemic exposures observed with parenteral dosing, low systemic exposure occurred with PO dosing to mice (Fig. 5A). A low Cmax of 14.5 ng/mL and a delayed Tmax of 4 h were observed following 30 mg/kg 5A1MQ PO dosing, suggesting slow and limited absorption in the intestine, with an estimated terminal half-life of 14.8 h. The AUC0-∞ after PO dosing was 224 ng*h/mL, resulting in a low oral bioavailability of 3.5%. Since 5A1MQ is highly soluble in aqueous (e.g., >100 mg/mL in phosphate buffered saline) and simulated gastric fluids (e.g., >10 mg/mL) and displays low binding to mouse plasma protein (Fb = 7%), the low oral bioavailability of 5A1MQ is likely attributed to limited enteric absorption and high first-pass metabolism. The latter factor is supported by 5A1MQ’s short half-life (T1/2 < 7 min) and high clearance (57.9 mL/min/kg) observed in cryopreserved mouse hepatocytes.

Compartmental modeling analysis concluded that three-compartment models with first-order absorption kinetics best fit the 5A1MQ PK datasets (Fig. S1). The IV model suggested 5A1MQ to be quickly distributed to peripheral compartments with a higher rate of distribution over elimination (i.e., K12 = 2.2 > K21 = 1.3; K13 = 2.4 > K31 = 0.02, Table S3). The PO model predicted lower 5A1MQ absorption (Ka = 0.07, Table S3) compared to SC dosing model (Ka = 6.0, Table S3). The SC dosing model also suggested rapid distribution to, but slow elimination kinetics from, the peripheral compartments (i.e., K12 = 1.8 > K21 = 1.67E-05; K13 = 0.7 > K31 = 0.2, Table S3). A similar profile was only noted in one peripheral compartment in the 5A1MQ PO model (i.e., K12 = 1.4 > K21 = 1.32E-05, Table S3), with rapid efflux predicted from the second peripheral compartment (i.e., K13 = 0.0002 < K31 = 0.2, Table S3). The latter in the PO model supports high hepatic first-pass metabolism and rapid elimination of 5A1MQ from the liver.

5A1MQ exhibited favorable tissue distribution profile following SC dosing to mice

Following once-daily repeated SC dosing to steady-state, significant total 5A1MQ (bound and unbound combined) exposure was noted at 1 h post-dose in EWAT and quadriceps muscle tissues that rapidly declined between 4 and 8 h post-dose (Table S4). This corresponded with inverse time-dependent measures of the target engagement biomarker 1-MNA in these tissues (Fig. S2), suggesting nearly 70% and 40% inhibition of 1-MNA at 1 h post-dose in EWAT and muscle, respectively, relative to the 1-MNA levels observed at 8 h post-dose. Increasing tissue-to-plasma 5A1MQ concentration ratios were, however, observed over time in EWAT and muscle tissues, suggesting slow efflux and sustained exposures; EWAT-to-plasma ratio increased from 1.4 at 1 h to 31.1 at 8 h post-dose, and muscle-to-plasma ratio increased from 2.2 at 1 h to 17.0 at 8 h post-dose (Table S4). Relatively, much lower 5A1MQ exposure was noted in mouse liver (mean total exposure level at 1 h post-dose = 83 ng/ng; liver-to-plasma ratio = 0.09) corresponding with a weaker PK/PD target engagement effect (Fig. S2), but with a similar slow efflux profile as noted in the other tissues with delayed plasma equilibrium at ~8 h post-dose (liver-to-plasma ratio at 8 h = 1.4).

Given the notable pharmacodynamic effects of 5A1MQ treatment in the livers of DIO mice, concentrations of 5A1MQ and the target engagement biomarker 1-MNA were evaluated at study termination at ~1–2 h post-5A1MQ dosing on Day 30 to assess the magnitude of 5A1MQ exposure-response. The liver concentrations of 5A1MQ in control and low-dose groups were below the lower limit of quantitation (i.e., 1 ng/mL) and averaged 28.5 ng/g in the high-dose group (Table S5). The latter supported a mean 39% reduction in 1-MNA level in the liver of the high-dose 5A1MQ group, relative to controls (Table S5). These results indicated effective target engagement by 5A1MQ at the pharmacological dose of 32 mg/kg/day in the liver of DIO mice, consistent with its observed liver-specific improvements.

5A1MQ displayed acceptable off-target profile

5A1MQ was tested at 10 μM concentration against the Cerep panel of major receptors, enzymes, and uptake transporter proteins. Results indicated no significant activity against all targets tested except a 67.4% inhibition of monoamine oxidase-A (MAO-A).

Discussion

In this study, 5A1MQ treatment effectively prevented body weight and fat mass gain in obese mice without altering food intake, extending prior research on NNMT inhibition in DIO mouse models7,8,9. NNMT upregulation with body weight gain24 shunts nicotinamide out of the NAD salvage pathway, thereby disrupting NAD biosynthesis, cellular energy metabolism7, and Sirtuin 1 (SIRT1) stability, the latter a crucial modulator of genes associated with metabolic processes25. We and others have shown that NNMT inhibition increases NAD and SAM levels in adipose systems7,8, enhancing energy expenditure and regulating polyamine flux7. Additionally, combining 5A1MQ with a lean diet produces a unique EWAT metabolic signature in HFD-fed mice, associating with strong inhibition of lipid and fatty acid synthesis pathways19. These previous studies, in conjunction with the present study support that NNMT inhibition can reduce adiposity and normalize adipose tissue metabolism without altering food intake in the context of obesity.

The DIO mouse is a translationally relevant model of human prediabetes characterized by hyperinsulinemia and oral glucose intolerance26,27. In the present study, control DIO mice demonstrated continually rising non-fasted plasma insulin levels27, which were dramatically suppressed by 5A1MQ treatment. Fasted insulin measures in the OGTT were also normalized by 5A1MQ treatment, suggesting efficient insulin response to glucose loading3. Since adipose, liver, and muscle tissues uptake and effectively clear glucose from systemic circulation28, and 5A1MQ effectively distributed to these tissues, it is hypothesized that 5A1MQ treatment in DIO mice may be enhancing insulin sensitivity in all these key metabolically active tissues. Overall, this data agrees well with previous studies where NNMT −/− mice on HFD displayed superior insulin sensitivity compared to their wild-type counterparts18. Similarly, NNMT knockdown by an antisense oligonucleotide or NNMT inhibitor treatment in DIO mice7,9, or in ob/ob and db/db mice9, significantly improved OGT, occasionally even in the absence of weight-loss effects. Our results on the strong 5A1MQ treatment effect observed on OGT after adjusting for body weight and fat mass readouts seem consistent with the previous reported findings. In addition to NNMT inhibition, 5A1MQ was observed to inhibit the human MAO-A enzyme in screening against a broad panel of target proteins. Hence, it is possible that the improved OGT by 5A1MQ may, in part, be mediated by its inhibitory activity against MAO-A, as previously reported for other MAO-A inhibitors29,30 and confirmed in NNMT −/− mice29.

Obesity-driven dyslipidemia27,31,32 and insulin resistance32,33 promote the development of MASLD. A therapeutic target for these diseases is NAD, a critical regulator of lipogenesis and fatty acid oxidation12,16,25. Decreased NAD levels lead to metabolic disruptions and non-alcoholic fatty liver disease development34. Supplements (e.g., nicotinic acid, a precursor of NNMT’s substrate nicotinamide) that promote NAD biosynthesis have shown to decrease plasma lipid levels, while the NNMT reaction product 1-MNA promotes hepatotoxicity and abnormal TG storage in the liver12. In this study, NNMT inhibition and subsequent 1-MNA reductions led to improved liver weight, steatosis, inflammation, which recapitulated the normalization of liver histopathology previously observed in obese mice treated with 5A1MQ and a reduced-calorie diet19. These effects occurred at significantly lower pharmacological exposure of 5A1MQ in the liver (AUC well under 1 μM), where 5A1MQ is expected to selectively engage NNMT drug target. This suggests that selective hepatic NNMT inhibition might restore liver health7,8,12 and reestablish metabolic equilibrium12,19.

Obesity-linked low-grade systemic inflammation and insulin resistance stimulate macrophage infiltration in the liver35,36. In this study, we observed elevated F4/80-positive macrophage levels in the control DIO livers, consistent with previous reports37,38. 5A1MQ treatment reduced liver F4/80-positive levels, similar to the effects reported for SGLT-2 and DPP-4 inhibitor drugs that regulate liver M1/M2 macrophage polarization and prevent the development of insulin resistance in HFD-fed mice37,38. In the case of NNMT inhibition, decreased liver inflammation may arise from stabilized SIRT1 activity resulting from NNMT inhibition-mediated increased NAD turnover, which downstream then impacts both inflammatory gene expression and insulin sensitivity39. Obese individuals typically also have elevated ketone body levels in plasma, liver, and cardiac tissues40,41. 5A1MQ treatment decreased serum ketone levels, potentially also by a SIRT1-mediated mechanism of regulating downstream transcriptional factors (e.g., peroxisome proliferator-activated receptor-γ [PPAR-γ]) that are known to modulate ketogenic enzymes41. Further, it is not uncommon in obesity for white blood cell counts to be elevated as they correlate with impaired glucose tolerance and systemic and adipose tissue inflammation42–46. 5A1MQ treatment greatly reduced white blood cell levels through large decreases in lymphocytes and monocytes, which aligned with improved insulin sensitivity. Also, serum chemistry and enzyme (ALT, AST) levels that are impacted with insulin resistance and obesity40,47,48 and routinely used to monitor liver function and fatty liver diseases were improved with 5A1MQ treatment, suggesting overall better liver function.

This study demonstrated that systemic NNMT inhibition using a pharmacological inhibitor such as 5A1MQ in obese animals can restore non-fasted plasma insulin levels, promote glucose clearance and insulin sensitivity during OGTT, and improve overall liver health by mitigating steatosis and inflammation. Multiple pathways support NNMT inhibition-mediated modulation of energy metabolism, steatosis, and inflammation, including i) the fundamental regulation of the NAD salvage pathway, resulting in increased capacity for NAD turnover7,8 that improves cellular energy expenditure25,49, and stabilized SIRT1 activity7 that regulates transcription of inflammatory genes (e.g., TNF-α, IL-1β, F4/80)39,50 and lipogenesis51 and ii) alterations to the methionine pathway by increasing intracellular SAM levels, allowing for increased polyamine flux-mediated increases in energy metabolism7,8 and downstream epigenetic changes, including fibrotic gene regulation52. Ongoing efforts are addressing further optimization of 5A1MQ, including improving species (e.g., mouse)-specific high metabolic clearance and off-target activity against MAO-A. Future studies will apply multi-omics analyses of adipose, liver, and muscle tissues to elucidate the pathways responsible for whole-body insulin sensitivity improvements that occur with NNMT inhibitor treatment.

Supplementary Material

Supinfo1
Supinfo2
Supinfo3
Supinfo4
Supinfo5
Supinfo6
Supinfo7
Supinfo8
Supinfo9
Supinfo10

Acknowledgments

We acknowledge our collaborators (Drs. Stanton McHardy and Hua-Yu Leo Wang) at the Center for Innovative Drug Discovery, University of Texas San Antonio, for their support in synthesizing gram-scale batches of 5A1MQ. We also acknowledge Wuxi Apptec Ltd. for conducting in vivo studies and biochemical, histological, and serum chemistry assays and HistoBridge LLC for immunohistochemical analysis of F4/80 in liver tissues. Grants: HLS (NCATS CTSA KL2, KL2TR001441-06); SJW (NIDDK SBIR, 1R41DK119052-01); HN (DoD, PR180216).

Footnotes

Competing Interests

SJW is the founder of Ridgeline Therapeutics, HN is a paid employee of Ridgeline Therapeutics, and JJB and DB are former employees of Ridgeline Therapeutics. HLS declares no competing interests.

References

  • 1.Mitchell NS, Catenacci VA, Wyatt HR, Hill JO. Obesity: overview of an epidemic. Psychiatr Clin North Am. 2011;34(4):717–732. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Uranga RM, Keller JN. The Complex Interactions Between Obesity, Metabolism and the Brain. Front Neurosci. 2019;13:513. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Cersosimo E, Triplitt C, Solis-Herrera C, Mandarino LJ, DeFronzo RA. Pathogenesis of Type 2 Diabetes Mellitus. In: Feingold KR, Anawalt B, Boyce A, et al. , eds. Endotext. South Dartmouth (MA)2000. [Google Scholar]
  • 4.DeFronzo RA. Lilly lecture 1987. The triumvirate: beta-cell, muscle, liver. A collusion responsible for NIDDM. Diabetes. 1988;37(6):667–687. [DOI] [PubMed] [Google Scholar]
  • 5.Castro AV, Kolka CM, Kim SP, Bergman RN. Obesity, insulin resistance and comorbidities? Mechanisms of association. Arq Bras Endocrinol Metabol. 2014;58(6):600–609. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Kahn BB, Flier JS. Obesity and insulin resistance. J Clin Invest. 2000;106(4):473–481. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Kraus D, Yang Q, Kong D, et al. Nicotinamide N-methyltransferase knockdown protects against diet-induced obesity. Nature. 2014;508(7495):258–262. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Neelakantan H, Vance V, Wetzel MD, et al. Selective and membrane-permeable small molecule inhibitors of nicotinamide N-methyltransferase reverse high fat diet-induced obesity in mice. Biochem Pharmacol. 2018;147:141–152. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Kannt A, Rajagopal S, Kadnur SV, et al. A small molecule inhibitor of Nicotinamide N-methyltransferase for the treatment of metabolic disorders. Sci Rep. 2018;8(1):3660. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Kannt A, Pfenninger A, Teichert L, et al. Association of nicotinamide-N-methyltransferase mRNA expression in human adipose tissue and the plasma concentration of its product, 1-methylnicotinamide, with insulin resistance. Diabetologia. 2015;58(4):799–808. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Liu M, Li L, Chu J, et al. Serum N(1)-Methylnicotinamide Is Associated With Obesity and Diabetes in Chinese. J Clin Endocrinol Metab. 2015;100(8):3112–3117. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Komatsu M, Kanda T, Urai H, et al. NNMT activation can contribute to the development of fatty liver disease by modulating the NAD (+) metabolism. Sci Rep. 2018;8(1):8637. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Pissios P Nicotinamide N-Methyltransferase: More Than a Vitamin B3 Clearance Enzyme. Trends Endocrinol Metab. 2017;28(5):340–353. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Riederer M, Erwa W, Zimmermann R, Frank S, Zechner R. Adipose tissue as a source of nicotinamide N-methyltransferase and homocysteine. Atherosclerosis. 2009;204(2):412–417. [DOI] [PubMed] [Google Scholar]
  • 15.Crujeiras AB, Pissios P, Moreno-Navarrete JM, et al. An Epigenetic Signature in Adipose Tissue Is Linked to Nicotinamide N-Methyltransferase Gene Expression. Mol Nutr Food Res. 2018:e1700933. [DOI] [PubMed] [Google Scholar]
  • 16.Canto C, Menzies KJ, Auwerx J. NAD(+) Metabolism and the control of energy homeostasis: A balancing act between mitochondria and the nucleus. Cell Metab. 2015;22(1):31–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Janke R, Dodson AE, Rine J. Metabolism and epigenetics. Annu Rev Cell Dev Biol. 2015;31:473–496. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Brachs S, Polack J, Brachs M, et al. Genetic Nicotinamide N-Methyltransferase (Nnmt) Deficiency in Male Mice Improves Insulin Sensitivity in Diet-Induced Obesity but Does Not Affect Glucose Tolerance. Diabetes. 2019;68(3):527–542. [DOI] [PubMed] [Google Scholar]
  • 19.Sampson CM, Dimet AL, Neelakantan H, et al. Combined nicotinamide N-methyltransferase inhibition and reduced-calorie diet normalizes body composition and enhances metabolic benefits in obese mice. Sci Rep. 2021;11(1):5637. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Neelakantan H, Wang HY, Vance V, Hommel JD, McHardy SF, Watowich SJ. Structure-activity relationship for small molecule inhibitors of nicotinamide N-methyltransferase. J Med Chem. 2017;60(12):5015–5028. [DOI] [PubMed] [Google Scholar]
  • 21.National Research Council (U.S.). Committee for the Update of the Guide for the Care and Use of Laboratory Animals., Institute for Laboratory Animal Research (U.S.), National Academies Press (U.S.). Guide for the care and use of laboratory animals. In: 8th ed. Washington, D.C.: National Academies Press,; 2011: http://grants.nih.gov/grants/olaw/Guide-for-the-Care-and-use-of-laboratory-animals.pdf. [Google Scholar]
  • 22.Hansen HH, HM AE, Oro D, et al. Human translatability of the GAN diet-induced obese mouse model of non-alcoholic steatohepatitis. BMC Gastroenterol. 2020;20(1):210. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Sunny NE, Satapati S, Fu X, et al. Progressive adaptation of hepatic ketogenesis in mice fed a high-fat diet. Am J Physiol Endocrinol Metab. 2010;298(6):E1226–1235. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Rudolphi B, Zapp B, Kraus NA, Ehebauer F, Kraus BJ, Kraus D. Body weight predicts Nicotinamide N-Methyltransferase activity in mouse fat. Endocr Res. 2017:1–9. [DOI] [PubMed] [Google Scholar]
  • 25.Trammell SA, Brenner C. NNMT: A Bad Actor in Fat Makes Good in Liver. Cell Metab. 2015;22(2):200–201. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.O’Brien PD, Hinder LM, Rumora AE, et al. Juvenile murine models of prediabetes and type 2 diabetes develop neuropathy. Dis Model Mech. 2018;11(12). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Thomas DD, Corkey BE, Istfan NW, Apovian CM. Hyperinsulinemia: An Early Indicator of Metabolic Dysfunction. J Endocr Soc. 2019;3(9):1727–1747. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Lopes M, Brejchova K, Riecan M, et al. Metabolomics atlas of oral 13C-glucose tolerance test in mice. Cell Rep. 2021;37(2):109833. [DOI] [PubMed] [Google Scholar]
  • 29.Ruf S, Rajagopal S, Kadnur SV, et al. Novel tricyclic small molecule inhibitors of Nicotinamide N-methyltransferase for the treatment of metabolic disorders. Sci Rep. 2022;12(1):15440. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Wickstrom L, Pettersson K. Treatment of Diabetics with Monoamine-Oxidase Inhibitors. Lancet. 1964;2(7367):995–997. [DOI] [PubMed] [Google Scholar]
  • 31.Marchesini G, Moscatiello S, Di Domizio S, Forlani G. Obesity-associated liver disease. J Clin Endocrinol Metab. 2008;93(11 Suppl 1):S74–80. [DOI] [PubMed] [Google Scholar]
  • 32.Fabbrini E, Sullivan S, Klein S. Obesity and nonalcoholic fatty liver disease: biochemical, metabolic, and clinical implications. Hepatology. 2010;51(2):679–689. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Chen Z, Yu R, Xiong Y, Du F, Zhu S. A vicious circle between insulin resistance and inflammation in nonalcoholic fatty liver disease. Lipids Health Dis. 2017;16(1):203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Guarino M, Dufour JF. Nicotinamide and NAFLD: Is There Nothing New Under the Sun? Metabolites. 2019;9(9). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Hotamisligil GS. Inflammation and metabolic disorders. Nature. 2006;444(7121):860–867. [DOI] [PubMed] [Google Scholar]
  • 36.Weisberg SP, McCann D, Desai M, Rosenbaum M, Leibel RL, Ferrante AW Jr. Obesity is associated with macrophage accumulation in adipose tissue. J Clin Invest. 2003;112(12):1796–1808. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Xu L, Nagata N, Nagashimada M, et al. SGLT2 Inhibition by Empagliflozin Promotes Fat Utilization and Browning and Attenuates Inflammation and Insulin Resistance by Polarizing M2 Macrophages in Diet-induced Obese Mice. EBioMedicine. 2017;20:137–149. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Zhuge F, Ni Y, Nagashimada M, et al. DPP-4 Inhibition by Linagliptin Attenuates Obesity-Related Inflammation and Insulin Resistance by Regulating M1/M2 Macrophage Polarization. Diabetes. 2016;65(10):2966–2979. [DOI] [PubMed] [Google Scholar]
  • 39.Yoshizaki T, Schenk S, Imamura T, et al. SIRT1 inhibits inflammatory pathways in macrophages and modulates insulin sensitivity. Am J Physiol Endocrinol Metab. 2010;298(3):E419–428. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Dhatariya K Blood Ketones: Measurement, Interpretation, Limitations, and Utility in the Management of Diabetic Ketoacidosis. Rev Diabet Stud. 2016;13(4):217–225. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Sikder K, Shukla SK, Patel N, Singh H, Rafiq K. High Fat Diet Upregulates Fatty Acid Oxidation and Ketogenesis via Intervention of PPAR-gamma. Cell Physiol Biochem. 2018;48(3):1317–1331. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Ziaee A, Ghorbani A, Kalbasi S, Hejrati A, Moradi S. Association of hematological indices with prediabetes: A cross-sectional study. Electron Physician. 2017;9(9):5206–5211. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Veronelli A, Laneri M, Ranieri R, et al. White blood cells in obesity and diabetes: effects of weight loss and normalization of glucose metabolism. Diabetes Care. 2004;27(10):2501–2502. [DOI] [PubMed] [Google Scholar]
  • 44.Ip BC, Hogan AE, Nikolajczyk BS. Lymphocyte roles in metabolic dysfunction: of men and mice. Trends Endocrinol Metab. 2015;26(2):91–100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Nemeth E, Ganz T. Anemia of inflammation. Hematol Oncol Clin North Am. 2014;28(4):671–681, vi. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Pouchot J, Sampalis JS, Beaudet F, et al. Adult Still’s disease: manifestations, disease course, and outcome in 62 patients. Medicine (Baltimore). 1991;70(2):118–136. [PubMed] [Google Scholar]
  • 47.Haan YC, Oudman I, Diemer FS, et al. Creatine kinase as a marker of obesity in a multi-ethnic population. Mol Cell Endocrinol. 2017;442:24–31. [DOI] [PubMed] [Google Scholar]
  • 48.Bonnet F, Ducluzeau PH, Gastaldelli A, et al. Liver enzymes are associated with hepatic insulin resistance, insulin secretion, and glucagon concentration in healthy men and women. Diabetes. 2011;60(6):1660–1667. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Akie TE, Liu L, Nam M, Lei S, Cooper MP. OXPHOS-Mediated Induction of NAD+ Promotes Complete Oxidation of Fatty Acids and Interdicts Non-Alcoholic Fatty Liver Disease. PLoS One. 2015;10(5):e0125617. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Zhu X, Liu Q, Wang M, et al. Activation of Sirt1 by resveratrol inhibits TNF-alpha induced inflammation in fibroblasts. PLoS One. 2011;6(11):e27081. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Ding RB, Bao J, Deng CX. Emerging roles of SIRT1 in fatty liver diseases. Int J Biol Sci. 2017;13(7):852–867. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Yoshino J, Patterson BW, Klein S. Adipose Tissue CTGF Expression is Associated with Adiposity and Insulin Resistance in Humans. Obesity (Silver Spring). 2019;27(6):957–962. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supinfo1
NIHMS2015151-supplement-Supinfo1.docx (19.1KB, docx)
Supinfo2
NIHMS2015151-supplement-Supinfo2.docx (63.8KB, docx)
Supinfo3
NIHMS2015151-supplement-Supinfo3.docx (166.2KB, docx)
Supinfo4
NIHMS2015151-supplement-Supinfo4.docx (33.2KB, docx)
Supinfo5
NIHMS2015151-supplement-Supinfo5.docx (65.1KB, docx)
Supinfo6
NIHMS2015151-supplement-Supinfo6.docx (40.2KB, docx)
Supinfo7
NIHMS2015151-supplement-Supinfo7.docx (29.3KB, docx)
Supinfo8
NIHMS2015151-supplement-Supinfo8.docx (31.9KB, docx)
Supinfo9
NIHMS2015151-supplement-Supinfo9.docx (48.2KB, docx)
Supinfo10
NIHMS2015151-supplement-Supinfo10.docx (135.8KB, docx)

RESOURCES