https://orcid.org/0000-0002-4917-033X
Abstract
Obesity is associated with perturbations in incretin production and whole-body glucose metabolism, but the precise underlying mechanism remains unclear. Here, we tested the hypothesis that nicotinamide phosphoribosyltransferase (NAMPT), which mediates the biosynthesis of nicotinamide adenine dinucleotide (NAD+), a key regulator of cellular energy metabolism, plays a critical role in obesity-associated intestinal pathophysiology and systemic metabolic complications. To this end, we generated a novel mouse model, namely intestinal epithelial cell-specific Nampt knockout (INKO) mice. INKO mice displayed diminished glucagon-like peptide-1 (GLP-1) production, at least partly contributing to reduced early-phase insulin secretion and postprandial hyperglycemia. Mechanistically, loss of NAMPT attenuated the Wnt signaling pathway, resulting in insufficient GLP-1 production. We also found that diet-induced obese mice had compromised intestinal NAMPT-mediated NAD+ biosynthesis and Wnt signaling pathway, associated with impaired GLP-1 production and whole-body glucose metabolism, resembling the INKO mice. Finally, administration of a key NAD+ intermediate, nicotinamide mononucleotide (NMN), restored intestinal NAD+ levels and obesity-associated metabolic derangements, manifested by a decrease in ileal Proglucagon expression and GLP-1 production as well as postprandial hyperglycemia in INKO and diet-induced obese mice. Collectively, our study provides mechanistic and therapeutic insights into intestinal NAD+ biology related to obesity-associated dysregulation of GLP-1 production and postprandial hyperglycemia.
Obesity is associated with postprandial hyperglycemia, an important predictor for the development of cardiovascular diseases (1, 2), type 2 diabetes, and nonalcoholic fatty liver disease (3, 4). The mechanisms by which obesity induces postprandial hyperglycemia have not been fully elucidated but likely involve dysfunction of intestinal hormones, such as glucose-dependent insulinotropic polypeptide (GIP) and glucagon-like peptide-1 (GLP-1) (5). GIP and GLP-1 are produced by intestinal endocrine K cells and L cells, respectively, both of which are released in response to oral nutrient ingestion that stimulates insulin release in a glucose-dependent manner (6, 7). Importantly, high-fat diet (HFD)-fed obese rats displayed diminished L-cell numbers, accompanied by a decrease in postprandial GLP-1 concentration and hyperglycemia (8). In addition, data obtained from studies conducted in humans suggest that alterations in nutrient-induced GLP-1 response are associated with obesity and its metabolic complications, such as postprandial hyperglycemia (9). For example, GLP-1 response is significantly decreased in subjects who are obese and glucose intolerant compared with healthy-lean subjects (10-12), although there are conflicting results showing that meal-induced GLP-1 secretion is increased or unchanged in subjects with obesity (9). In contrast, diet-induced weight reduction improved postprandial GLP-1 response and hyperglycemia in subjects with obesity (13). These studies underscore the importance of intestinal GLP-1 production as a mediator in obesity-associated postprandial hyperglycemia; however, the molecular mechanisms underlying the dysfunctional intestinal GLP-1 production associated with obesity remain unclear.
Nicotinamide adenine dinucleotide (NAD+) is a key coenzyme regulating cellular energy metabolism and redox status in various species (14-16). Regulation of cellular NAD+ levels involves multiple NAD+ biosynthetic enzymes, including nicotinamide phosphoribosyltransferase (NAMPT) and nicotinamide mononucleotide adenylyl transferases, as well as NAD+-consuming enzymes, such as CD38 and poly ADP-ribose polymerases (14-16). NAMPT is the rate-limiting enzyme in mammalian NAD+ biosynthesis, and impaired NAMPT-mediated NAD+ biosynthesis has been shown to cause obesity-associated metabolic abnormalities, such as glucose intolerance, beta-cell dysfunction, insulin resistance, and type 2 diabetes (14-17). For example, NAMPT deficiency induces adipose tissue dysfunction and insulin resistance, and altered whole-body energy and glucose metabolism (18-21). In liver, Nampt deletion impairs fatty acid oxidation and accelerates hepatic steatosis (22). In the intestines of mice, a conditional whole-body knockout of the Nampt gene significantly decreases intestinal NAD+ levels and triggers dysfunctional nutrient absorption associated with intestinal villi atrophy and the downregulation of ATP-dependent transporters (23). In addition, NAD+ levels in the colon decline with age and are accompanied by aging-associated colonic degeneration (24). Together, these findings suggest the importance of intestinal NAD+ biology in intestinal function.
Although it was recently reported that NAMPT inhibition decreased GLP-1 secretion in vitro studies (25), the present study was focused on investigating the roles of intestinal NAMPT-mediated NAD+ biosynthesis in maintaining intestinal integrity, including incretin production and whole-body glucose metabolism. We hypothesized that intestinal NAMPT-mediated NAD+ biosynthesis plays an essential role in regulating GLP-1 production and, therefore, impaired NAMPT-mediated NAD+ biosynthesis in intestines contributes to the development of obesity-associated postprandial hyperglycemia. To test this hypothesis, we generated intestinal epithelial cell-specific Nampt knockout (INKO) mice and examined their GLP-1 production and whole-body glucose metabolism. In addition, we evaluated changes in intestinal NAMPT-mediated NAD+ biosynthesis induced in HFD-fed obese mice and further explored the therapeutic potential of promoting intestinal NAD+ biology against diet-induced dysfunction in GLP-1 production and postprandial glucose metabolism.
Materials and Methods
Animal Experimentation
Three- to four-week-old C57BL/6J male mice were purchased from the CLEA Japan. Male and female INKO mice were generated by crossing Villin-Cre transgenic mice (#004586; Jackson Laboratory, Bar Harbor, ME, USA) (26) with floxed-Nampt (fl/fl) mice (27). Because of limited availability, a minimum sample size of 3 was used in physiological assessments. Experiments assessing metabolic phenotype were conducted in multiple cohorts. Statistical outliers were identified and removed from analysis using Grubbs’ test. Mice were housed in specific pathogen-free conditions with 12-hour light/dark cycles and were maintained on regular-chow diet (RCD) (CE-2; CLEA Japan) or HFD (60% kcal from fat) (Research Diets D12492; Research Diets, Inc., New Brunswick, NJ), with access to water ad libitum. Food intake of individually housed male mice was measured daily for 4 days. Tissue samples harvested from mice were immediately frozen in liquid nitrogen and stored at -80°C until further use. All animal studies were approved by the Keio University School of Medicine Institutional Animal Care and Use Committee.
Glucose and Energy Metabolism
For oral glucose tolerance tests (OGTTs), mice were fasted for approximately 10 hours before a 25% glucose solution (1 g/kg of body weight) was administered by oral gavage. For intraperitoneal glucose tolerance tests (IPGTTs), mice were fasted for approximately 16 hours before an intraperitoneal injection of 50% glucose solution (2 g/kg of body weight). Blood glucose levels were measured at 0, 15, 30, 60, and 120 minutes after glucose loading using the One Touch Ultra View (Johnson & Johnson KK, Tokyo, Japan). Blood samples (approximately 200 µL) were collected through retro-orbital sampling in tubes containing 5 µL of dipeptidyl peptidase IV inhibitor (Merck Millipore, Billerica, MA, USA), 0.15 to 0.40 trypsin inhibitory units of aprotinin (#A1153; Sigma-Aldrich, St. Louis, MO, USA), and 0.5 M EDTA (#E7889; Sigma-Aldrich) at 0 and 5 minutes for plasma GLP-1 measurements (28, 29). We established a cohort of mice used exclusively for collecting blood for GLP-1 measurements. For plasma insulin and glucagon measurements, blood samples (approximately 100 µL) were collected via tail vein at 0, 10, and 20 minutes. Plasma insulin (#M1104; Morinaga Institute of Biological Science, Yokohama, Japan; RRID: AB_2811268. https://antibodyregistry.org/search.php?q=AB_2811268) (30), glucagon (#10-1281-01; Mercodia, Uppsala, Sweden; RRID: AB_2783839 https://antibodyregistry.org/search.php?q=AB_2783839) (31), and GLP-1 (#27700; Immuno-Biological Laboratories, Gunma, Japan; RRID: AB_2892225. https://antibodyregistry.org/search.php?q=AB_2892225) (32) concentrations were determined using commercially available ELISA kits.
Nicotinamide Mononucleotide Rescue Experiments
Nicotinamide mononucleotide (NMN; #44501900, Oriental Yeast Co., Tokyo, Japan) at a dose of 500 mg/kg-body weight/day (33) was administered to male INKO mice fed with RCD and C57BL/6 male mice fed with HFD by oral gavage for up to 14 days (NMN-treated group). Based on our previous studies (18, 19), NMN was administered to male INKO mice soon after weaning, at 3 to 4 weeks of age. For OGTTs, mice were fasted for approximately 10 hours after oral gavage of NMN, and a 25% glucose solution (1 g/kg of body weight) was orally administered. The control group included age-matched fl/fl, INKO, and HFD-fed male mice receiving water by oral gavage (NMN-untreated group). Mice were euthanized 1 hour after gavage.
Ex Vivo Studies
Ileal extracts obtained from 8-week-HFD-fed C57BL/6J male mice that had received NMN or water by oral gavage were cut into small pieces and incubated in RPMI 1640 (#11875093; Thermo Fisher Scientific, Germany) containing Complete Protease Inhibitor Cocktail (#11697498001; Roche Diagnostics GmbH, Germany), 0.1% dipeptidyl peptidase IV inhibitor (Merck Millipore, Billerica, MA, USA), and 300 mM glucose. After 4 hours of incubation at 37°C, the supernatant was collected to measure GLP-1 (#27700; Immuno-Biological Laboratories, Gunma, Japan; RRID: AB_2892225. https://antibodyregistry.org/search.php?q=AB_2892225) concentrations (32, 34), which were normalized to tissue weights (35).
Indirect Calorimetry
Metabolic parameters were measured using a biogas analytical mass spectrometer (ARCO-2000; Arco System, Chiba, Japan). Briefly, male mice were individually housed and monitored with a 12-hour light/dark cycle and ad libitum access to water and food for 3 days, which included 2 days of acclimation. Whole-body oxygen consumption, CO2 production, and the respiratory exchange ratio were measured; energy expenditure was calculated and expressed per mouse for analysis and presentation.
Histology
Tissue samples were fixed in neutral formalin, embedded in paraffin, and sectioned at 4 μm at the Institute of Nutrition & Pathology Inc. Images were captured using an Axiocam camera and analyzed using AxioVision software (Carl Zeiss, Germany). Intestinal submucosal layer thickness in Masson trichrome-stained specimens was also measured using AxioVision (36).
Real-time PCR
Total RNA was isolated using an RNeasy Plus Mini Kit (#74134; Qiagen, Valencia, CA, USA) or TRIzol reagent (#15596018; Invitrogen, Carlsbad, CA, USA) (37), then reverse-transcribed into cDNA using PrimeScript RT Master Mix (#RR036B; Takara Bio, Shiga, Japan). Relative gene expression levels were determined using a KAPA SYBR FAST Universal Kit (#KK4602; Sigma-Aldrich). Expression levels were calculated using the 2−ΔΔCt method and normalized to levels of the housekeeping control gene, 18S ribosomal RNA. Primers are listed in Table S1 (38).
NAD+ Measurements
NAD+ was extracted from frozen tissue samples and STC-1 cells using ice-cold perchloric acid and neutralized with potassium carbonate. NAD+ concentrations were determined using an HPLC system (Prominence; Shimadzu Scientific Instruments, Kyoto, Japan) fitted with a Supelco LC-18-T column (#58970-U; Sigma-Aldrich) as previously described (18, 19, 39, 40). NAD+ concentrations were normalized to tissue or wet cell weights.
Western Blotting
Total protein was extracted from frozen tissues homogenized in radioimmunoprecipitation assay buffer (#182-02451; FUJIFILM Wako Pure Chemical, Osaka, Japan) containing protease inhibitor (#P8340; Sigma-Aldrich). Extracts were analyzed by western blotting as described previously (41, 42). Rabbit monoclonal anti-NAMPT (#86634; Cell Signaling Technology, Tokyo, Japan; RRID: AB_2800084. https://antibodyregistry.org/search?q=AB_2800084) (43) and mouse monoclonal anti-β actin (#A5316; Sigma-Aldrich; RRID: AB_476743; https://antibodyregistry.org/search?q=AB_476743) (44) were used as primary antibodies. Horseradish peroxidase-conjugated anti-rabbit IgG (#111-035-003; Jackson ImmunoResearch Laboratories, West Grove, PA, USA; RRID: AB_2313567; https://antibodyregistry.org/search?q=AB_2313567) (45) and anti-mouse IgG (#115-035-003; Jackson ImmunoResearch Laboratories; RRID: AB_10015289; https://antibodyregistry.org/search?q=AB_10015289) (46) were used as secondary antibodies. Blots used to detect NAMPT were stripped and reprobed with anti-β actin for normalization. Proteins were visualized using ECL Prime Western Blotting Detection Reagent (#RPN2232; GE Healthcare, Pittsburgh, PA, USA) and quantified using ImageJ software (National Institutes of Health).
Cell Culture
STC-1 cells (#CRL3254; ATCC; RRID: CVCL_J405; https://web.expasy.org/cellosaurus/CVCL_J405) (47) were cultured in Dulbecco’s modified Eagle’s medium (#11995-065; Gibco, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum and 1% antibiotic/anti-mycotic solution (#A5955; Sigma-Aldrich). To assess the expression of Proglucagon, molecules involved in the Wnt signaling pathway, and GLP-1 production in cell-culture supernatant, STC-1 cells were incubated in the medium in the presence or absence of 0.1% dimethyl sulfoxide (DMSO), 100 nM FK866 (Cayman Chemical, Ann Arbor, MI, USA), 100 µM NMN (Sigma-Aldrich), and 10 or 25 µM ICG-001 (#S2662; Selleckchem, Houston, TX, USA) for 24 hours. GLP-1 concentrations were normalized to cell numbers (1 × 106 cells).
Quantification and Statistical Analysis
Data are presented as mean ± standard error of the mean. Statistical analyses were performed using GraphPad Prism (version 8.0). Differences between groups were assessed using Student unpaired t test. Comparisons among multiple groups were performed using one-way analysis of variance (ANOVA) with Tukey's post-hoc test. Differences in continuous metabolic parameters were evaluated using repeated-measures ANOVA. Results with P < 0.05 were considered statistically significant.
Results
Generation of INKO Mice
To investigate the pathophysiological roles of intestinal NAMPT-mediated NAD+ biosynthesis in whole-body glucose metabolism, we generated INKO mice by crossing the mice harboring loxP sites flanking the Nampt (fl/fl) gene (27) with transgenic mice expressing Cre recombinase controlled by the Villin gene promoter (26). Although no differences in intestinal Nampt expression and NAD+ levels were detected between fl/fl and Villin-Cre (Cre) mice (Fig. S1a, b) (38), NAMPT gene and protein expression in INKO mice were markedly reduced in the jejunum, ileum, and colon, but not in other metabolic organs, such as the liver and kidneys (Fig. 1a and 1b and S2a, b) (38). Additionally, NAD+ levels were reduced by 43.4%, 48.4%, and 43.1% in the jejunum, ileum, and colon, respectively, but not in the liver and kidneys of INKO mice compared with fl/fl mice (Fig. 1c and S2c) (38), suggesting that NAMPT is an important regulator of NAD+ biosynthesis in the intestine.
Generation of intestinal epithelial cell-specific Nampt knockout (INKO) mice. INKO mice were generated by crossing floxed-Nampt (fl/fl) mice with transgenic mice expressing the Cre recombinase under the Villin gene promoter. (a) Nampt expression in whole extracts of jejunum, ileum, kidneys, and liver measured in 2- to 4-month-old fl/fl and INKO male mice (n = 4-7 per group). (b) Western blotting and quantification of NAMPT and β-actin in whole extracts of jejunum, ileum, kidneys, and liver harvested from 3- to 5-month-old fl/fl and INKO male mice (n = 4 per group). (c) Tissue NAD+ levels in whole extracts of jejunum, ileum, kidneys, and liver obtained from 4- to 7-month-old fl/fl and INKO male mice (n = 4-6 per group). Data were analyzed using Student unpaired t test and are presented as mean ± SEM. *P < 0.05; **P < 0.01. NAD+, nicotinamide adenine dinucleotide; NAMPT, nicotinamide phosphoribosyltransferase; SEM, standard error of the mean.
Intestinal Epithelial NAMPT Deletion Attenuates GLP-1 and Early-phase Insulin Secretion After Oral Glucose Loading
Intestinal epithelial NAMPT deletion did not affect body weight, food intake, or whole-body energy expenditure, suggesting that intestinal epithelial NAMPT deficiency had no obvious effects on overall energy intake or expenditure (Fig. S3a-c and S4a) (38). However, INKO mice, under ad libitum feeding conditions, had significantly higher blood glucose levels than did fl/fl mice (fl/fl, 170 ± 15.15 mg/dL vs INKO, 205 ± 6.45 mg/dL; n = 4-9 per group, P < 0.05), suggesting that INKO mice had impaired glucose metabolism. Therefore, we evaluated glucose tolerance using IPGTTs and OGTTs. Compared with the fl/fl mice, the INKO mice showed increased glucose levels and decreased insulin concentrations during OGTTs but not IPGTTs (Fig. 2a-d and S4b, c) (38). Furthermore, intestinal expression of glucose transporter genes, including Sglt1, Glut5, and Glut2, was comparable between INKO and fl/fl mice (Fig. S5) (38), suggesting that the loss of intestinal NAMPT contributed to reduced incretin activity. Supporting this idea, the ileal expression of Proglucagon, a GLP-1 precursor (35), and GLP-1 levels during OGTTs were significantly lower in INKO mice than in fl/fl mice (Fig. 2e, f). Importantly, ileal gene expression of Proglucagon was not different between fl/fl and Cre mice (Fig. S1c) (38). In contrast, plasma concentrations of glucagon, which also is processed from Proglucagon in the ileum as well as the pancreas (48, 49) during OGTTs, were similar between fl/fl and INKO mice (Fig. 2g). These results indicated that intestinal epithelial cell-specific deletion of Nampt causes impaired GLP-1 production, early-phase insulin secretion, and postprandial glucose metabolism. We next determined whether derangements of GLP-1 secretion and postprandial glucose metabolism induced by intestinal Nampt deletion were dependent on intracellular NAD+ biosynthesis in the small intestine. Therefore, we administered NMN, a product of NAMPT reaction, by oral gavage to INKO mice soon after weaning for 7 days (500 mg/kg-body weight/day) (33). NMN administration significantly increased NAD+ concentrations (Fig. 2h). NMN administration improved glucose levels during OGTTs in NMN-treated INKO mice compared with the age-matched NMN-untreated INKO mice (Fig. 2i). The glucose levels during OGTTs in NMN-treated INKO mice were similar to those in the age-matched fl/fl mice, suggesting that NMN administration normalized the postprandial glucose metabolism. NMN administration also restored the ileal expression of Proglucagon mRNA in INKO mice (Fig. 2j). In addition, the jejunal Gip expression of INKO mice was lower than that of fl/fl mice (Fig. S6a) (38). However, NMN administration did not increase Gip expression in the jejunum (Fig. S6b) (38). These findings further support our notion that NAMPT-mediated NAD+ biosynthesis in intestines is an important regulator of ileal Proglucagon expression, GLP-1 production, and postprandial glucose metabolism.
Effects of Nampt deletion in intestinal epithelial cells on glucose metabolism. (a) Intraperitoneal glucose tolerance tests (IPGTTs) (n = 8-11 per group) and (c) oral glucose tolerance tests (OGTTs) conducted in 2- to 5-month-old male mice (n = 8-10 per group). Area under the curve (AUC) for glucose is shown next to each curve. Plasma insulin concentrations during (b) IPGTTs (n = 5 per group) and (d) OGTTs (n = 8 per group) in 2- to 5-month-old male mice. (e) Expression of Proglucagon mRNA in whole extracts of ileum obtained from 2- to 5-month-old male INKO and fl/fl mice (n = 6 per group). (f) Active GLP-1 concentrations in plasma during OGTTs in 2- to 5-month-old male mice (n = 5–12 per group). (g) Plasma glucagon concentrations during OGTTs in 2- to 5-month-old male mice (n = 4-5 per group). (h-j) NMN (500 mg/kg-body weight/day: NMN-treated group) or water (NMN-untreated group) was administered to male fl/fl and INKO mice by oral gavage for 7 consecutive days soon after weaning at 3 to 4 weeks of age. (h) Ileal NAD+ concentrations in 4- to 5-week-old NMN-treated (gray bar) and NMN-untreated (black bar) INKO male mice (n = 3 per group). (i) OGTTs after 6-day treatment with NMN. OGTTs from 4- to 5-week-old NMN-treated INKO (gray circles, n = 4) and age-matched untreated INKO (black circles, n = 4) or fl/fl (white circles, n = 5) male mice. The AUC for glucose is shown next to the glucose tolerance curves. (j) Gene expression of Proglucagon in whole extracts of ileum from 4- to 5-week-old NMN-treated INKO (gray bar) and NMN-untreated INKO (black bar) or fl/fl (white bar) male mice (n = 3 per group). Data were analyzed using Student unpaired t test or one-way ANOVA with Tukey's post-hoc test. Data are presented as mean ± SEM. *P < 0.05; **P < 0.01. GLP-1, glucagon-like peptide-1; INKO, intestinal epithelial cell-specific Nampt knockout; NAD+, nicotinamide adenine dinucleotide; NMN, nicotinamide mononucleotide; SEM, standard error of the mean.
Loss of NAMPT Impairs GLP-1 Synthesis via Attenuation of Canonical Wnt Signaling
To identify the underlying mechanism for reduced GLP-1 production in INKO mice, we evaluated whether intestinal epithelial NAMPT deletion alters intestinal function. The small intestine and colon were shorter in INKO mice than in fl/fl mice (Fig. 3a and S2d) (38). Masson trichrome staining revealed that all parts of the intestine in INKO mice exhibited a thickening of the submucosal layer with edematous changes, which is a hallmark for intestinal fibrotic change (36) (Fig. 3b and 3c and S2e, f) (38). Consistently, the expression of fibrotic marker genes, such as aSMA and Fibronectin, was higher in INKO mice than in fl/fl mice (Fig. 3d and S2g) (38). Moreover, intestinal epithelial cell-specific Nampt deletion decreased the expression of Proglucagon, encoding the enteroendocrine hormone, GLP-1 (35) (Fig. 2e). Because the canonical Wnt signaling pathway is involved in regulating Proglucagon gene expression and GLP-1 production, we analyzed the intestinal expression of Wnt target genes, including Lgr5, Axin2, β-catenin, and Wnt3 (50-52). Our real-time PCR analysis of the small intestine revealed significantly lower Lgr5 and Axin2 expression levels in INKO mice than in fl/fl mice and showed decreased Wnt3 expression (P = 0.05) (Fig. 4a). These results prompted us to hypothesize that the loss of NAMPT impairs Wnt signaling in the ileum, which could result in insufficient GLP-1 production. To test this hypothesis, we used the murine enteroendocrine L-cell line STC-1 (34). FK866, a specific NAMPT inhibitor, markedly reduced NAD+ concentration compared with that of the control treatment (Fig. 4b). Consistent with our in vivo observations, pharmacological inhibition of NAMPT decreased the expression of the intestinal Wnt target genes, such as Lgr5, Axin2, and β-catenin, but not of Wnt3 (Fig. 4c). Importantly, NAD+ deficiency also reduced Proglucagon gene expression and GLP-1 production (Fig. 4e and 4f). We further examined whether exogenous supplementation of NMN could rescue the detrimental effects of NAMPT inhibition in STC-1 cells. NAD+ concentration was fully restored in the presence of NMN (Fig. 4b). Consistent with the normalization of NAD+ levels, expression of intestinal Wnt target genes and the Proglucagon gene, along with GLP-1 production, were substantially restored (Fig. 4c, 4e, and 4f), demonstrating the importance of NAMPT-mediated NAD+ biosynthesis in GLP-1 production. Remarkably, adding the potent Wnt signaling pathway inhibitor ICG-001 (53) to STC-1 cells decreased Proglucagon gene expression in a dose-dependent manner (Fig. 4d) and further abrogated the effects of NMN on Proglucagon gene expression and GLP-1 production in FK866-treated STC-1 cells (Fig. 4e and 4f). Consistently, NMN administration to INKO mice restored Proglucagon expression in ileum, although the small-intestine length and fibrotic markers were not affected (Fig. 2j; Fig. 3e and 3f). We cannot exclude the possibility that the morphological phenotypes of INKO mice are partially mediated by the alterations in GLP-2, which is also processed from proglucagon in enteroendocrine L cells and regulates intestinal epithelial proliferation and intestinal growth (54-56). However, these findings suggested that NAMPT-mediated NAD+ biosynthesis in the intestine regulates GLP-1 production, at least in part, by Wnt signaling but not by fibrotic alterations.
NAMPT loss leads to a disrupted intestinal architecture. (a) Small-intestine lengths in 2- to 5-month-old male mice (n = 6-7 per group). (b) Representative images of Masson trichrome-stained longitudinal sections of jejunum and ileum from 2- to 6-month-old fl/fl and INKO male mice (scale bar, 50 µm). (c) Quantification of submucosal fibrous layer thickness in jejunum and ileum of 2- to 6-month-old fl/fl and INKO male mice (n = 3 per group). (d) Intestinal expression of fibrosis marker genes determined by real-time PCR in 2- to 5-month-old male INKO and fl/fl mice (n = 4-8 per group). αSMA,αsmooth muscle actin. (e, f) NMN (500 mg/kg-body weight/day: NMN-treated group) or water (NMN-untreated group) was administered to male INKO or fl/fl mice by oral gavage for 7 consecutive days soon after weaning at 3 to 4 weeks of age. Expression of fibrosis marker genes (e) in whole extracts of ileum from 4- to 5-week-old NMN-treated (gray bars) and NMN-untreated INKO (black bars) male mice (n = 3 per group). (f) Small-intestine lengths from 4- to 5-week-old NMN-treated INKO (gray bars), untreated INKO (black bars), and fl/fl (white bars) male mice (n = 3-4). Data were analyzed using Student unpaired t test or one-way ANOVA with Tukey's post-hoc test and presented as mean ± SEM. *P < 0.05; **P < 0.01. INKO, intestinal epithelial cell-specific Nampt knockout; NAMPT, nicotinamide phosphoribosyltransferase; NMN, nicotinamide mononucleotide; SEM, standard error of the mean
Loss of NAMPT impairs GLP-1 production by attenuating Wnt signaling. (a) Expression of genes encoding Wnt signaling targets in the small intestine of 2- to 4-month-old fl/fl and INKO male mice (n = 5-7 per group). (b-e) The mouse enteroendocrine L-cell line STC-1 was cultured in the presence or absence of 0.1% DMSO; 100 nM FK866, a potent NAMPT inhibitor; 100 µM nicotinamide mononucleotide (NMN); and ICG-001, a Wnt pathway inhibitor, for 24 hours. (b) NAD+ concentrations measured in STC-1 cells in the presence or absence of FK866 and NMN (n = 5 per group). (c) mRNA expression levels of genes involved in the Wnt signaling pathway in the presence or absence of FK866 and NMN (n = 4-8 per group). (d) Expression of Proglucagon mRNA in STC-1 cells after 24 hours of treatment with 0.1% DMSO, 10 μM or 25 μM of ICG-001 (n = 3–10 per group). (e, f) Expression of Proglucagon mRNA (n = 4-11 per group) and active GLP-1 concentrations (n = 5-6 per group) in the presence or absence of FK866, NMN, and 10 µM ICG-001 in FK866-treated STC-1 cells. GLP-1 concentrations were normalized to 1 × 106 cells. Data were analyzed using Student unpaired t test or one-way ANOVA with Tukey's post-hoc test. Lgr5; leucine-rich repeat containing G protein-coupled receptor 5. Data are presented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001. DMSO, dimethyl sulfoxide; GLP-1, glucagon-like peptide-1; INKO, intestinal epithelial cell-specific Nampt knockout; NAD+, nicotinamide adenine dinucleotide; NAMPT, nicotinamide phosphoribosyltransferase.
NMN Enhances GLP-1 Production in HFD-induced Obese Mice
We next asked whether intestinal NAMPT-mediated NAD+ biosynthesis is involved in the pathogenesis of obesity-associated dysregulation of whole-body glucose metabolism. We fed 5-week-old C57BL/6 male mice with a 60% HFD for 8 weeks. NAMPT gene and protein expression was significantly lower in the small intestine of HFD-fed obese mice than in that of RCD-fed mice (Fig. 5a and 5b). Consistent with these observations, NAD+ levels in the ileum were also significantly decreased by 44.3% in HFD-fed mice compared with those levels in RCD-fed mice (Fig. 5c). Furthermore, the ileal expression of Wnt target genes, including Axin2 and β-catenin was significantly lower in HFD-fed mice than in RCD-fed mice (Fig. 5d). These changes were accompanied by decreased Proglucagon expression in the ileum of HFD-fed mice (Fig. 5e). Accordingly, HFD-fed mice showed significantly increased glucose levels not only at 10 hours of fasting, but also 5 minutes after oral glucose loading (fasting glucose: RCD-fed mice, 167.75 ± 1.875 mg/dL vs HFD-fed mice, 202.6 ± 10.89 mg/dL; n = 4-5 per group, P < 0.05; glucose levels at 5 minutes after oral glucose loading at the dose of 1 g/kg-body weight: RCD-fed mice, 314.5 ± 29.67 mg/dL vs HFD-fed mice, 412.4 ± 22.29 mg/dL, n = 4–5 per group, P < 0.05). These results indicated that HFD impaired intestinal NAMPT-mediated NAD+ biosynthesis, the Wnt signaling pathway, GLP-1 production, and whole-body glucose metabolism. To investigate whether the abnormalities in GLP-1 production and whole-body glucose metabolism induced by HFD feeding depended on intracellular NAD+ biosynthesis in the intestine, we gave NMN (500 mg/kg-body weight/day) (33) by oral gavage to HFD-fed obese mice for up to 14 days. NMN increased NAD+ concentrations and Proglucagon expression in the ileum without changing body weight gain (Fig. 5f-5h). In addition, ileum explants (34) from NMN-treated HFD-fed obese mice displayed substantially higher GLP-1 secretion than did those from NMN-untreated HFD-fed obese mice (Fig. 5i). NMN-treated HFD-fed obese mice showed a trend toward improvements in glucose level and GLP-1 secretion during OGTTs (Fig. 5j, k). These findings demonstrated that impaired intestinal NAD+ biosynthesis is involved in obesity-associated dysregulation of GLP-1 production and postprandial glucose metabolism.
NMN administration improves impaired GLP-1 production induced by defects in NAMPT-mediated NAD+ biosynthesis in HFD-induced obese mice. (a-e) Five-week-old male C57BL/6 mice were fed a regular chow (RCD, white bars) or 60% fat (HFD, black bars) diets for 8 weeks. (a) Nampt expression in whole ileum extracts (n = 5 per group). (b) Western blotting and quantification of NAMPT and β-actin in small intestine (n = 3-4 per group). (c) Tissue NAD+ levels in whole ileum extracts (n = 4 per group). (d) mRNA expression of Wnt signaling pathway genes (n = 4-5 per group) and (e) Proglucagon (n = 5 per group) in whole ileum extracts. (f-k) Water (NMN-untreated group, black bars) or NMN (500 mg/kg-body weight/day, NMN-treated group, gray bars) was administered to HFD-fed male mice for up to 14 consecutive days by oral gavage. (f) Ileum NAD+ concentrations (n = 4 per group). (g) Body weight gain (n = 5 per group). (h) Ileal Proglucagon expression (n = 4-5 per group). (i) GLP-1 secretion from ileum extracts. Active GLP-1 in supernatants was normalized to tissue weights and is presented as a ratio relative to the NMN-untreated group (n = 4-5 per group). (j) OGTTs after 10-day treatment with NMN-treated (gray circles, n = 5) and age-matched untreated (black circles, n = 5) HFD-fed mice. (k) Active GLP-1 concentrations in plasma during OGTTs in NMN-treated (gray bars) and age-matched untreated (black bars) HFD-fed mice (n = 4-5 per group). Data were analyzed using Student unpaired t test and presented as mean ± SEM. *P < 0.05; **P < 0.01. GLP-1, glucagon-like peptide-1; HFD, high-fat diet; NAD+, nicotinamide adenine dinucleotide; NAMPT, nicotinamide phosphoribosyltransferase; NMN, nicotinamide mononucleotide; OGTT, oral glucose tolerance test; SEM, standard error of the mean.
Discussion
This study characterizes NAD+ as a novel pivotal regulator of GLP-1 production and whole-body postprandial glucose metabolism by analyzing an intestinal epithelial cell-specific Nampt knockout (INKO) and diet-induced obese mice. We found that intestinal epithelial cell-specific Nampt deletion (1) diminished GLP-1 production, accompanied by decreased early-phase insulin secretion and postprandial hyperglycemia without concomitant changes in body weight; and (2) attenuated the Wnt signaling that is likely responsible for decreased GLP-1 production. Furthermore, the diet-induced obese mice had a significant reduction in intestinal NAMPT expression and NAD+ levels, accompanied by impaired Wnt signaling, GLP-1 production, and postprandial glucose metabolism, compared with the lean mice. Finally, administration of NMN, a key NAD+ intermediate, restored these obesity-associated metabolic derangements. Collectively, these data reveal the crucial role of intestinal NAMPT-mediated NAD+ biosynthesis in the maintenance of GLP-1 production and normal postprandial glucoregulatory response, along with the therapeutic potential of NMN for these obesity-associated metabolic abnormalities.
Our results underscore the importance of intestinal NAMPT-mediated NAD+ biosynthesis in postprandial glucose metabolism. We found that an intestinal epithelial NAD+ deficiency induced defective GLP-1 production, which is consistent with previous in vitro findings (25). This diminished GLP-1 production contributed, at least in part, to decreased early-phase insulin secretion and resulted in postprandial hyperglycemia. Early-phase insulin secretion associated with GLP-1 production is disrupted in people with obesity accompanied by glucose intolerance and is critically involved in type 2 diabetes development (57, 58). GLP-1, secreted by L cells, plays a key role in the regulation of glucose homeostasis by stimulating glucose-dependent insulin secretion in pancreatic beta cells and by reducing food intake and body weight (5-7). Interestingly, although INKO mice produced less GLP-1, their food intake and body weight were comparable to those of control mice, consistent with a report on intestinal epithelial Proglucagon knockout mice (59). These findings demonstrate that intestinal NAMPT-mediated NAD+ biosynthesis is important for maintaining normal glucoregulatory response without concomitant changes in food intake or body weight.
The molecular mechanisms by which NAMPT-mediated NAD+ biosynthesis regulates GLP-1 production in the intestine remain unclear but may involve Wnt signaling pathway inactivation. Importantly, several studies have demonstrated that Wnt signaling is essential for regulating enteroendocrine cell functions such as incretin production (50, 60). For example, activation of the Wnt signaling pathway increases GLP-1 by modulating Proglucagon gene expression in intestinal enteroendocrine GLUTag and STC-1 L cells (50, 60). Indeed, our results, primarily from in vitro studies, demonstrate that NAD+ may be a pivotal endogenous regulator of Wnt signaling in the intestine. Although the precise mechanisms by which NAMPT-mediated NAD+ biosynthesis regulates Wnt signaling remain unclear, our results suggest that it is a downstream mediator of NAMPT-mediated NAD+ biosynthesis, required for GLP-1 secretion from L cells. Another possible explanation for the impaired GLP-1 production induced by intestinal NAD+ deficiency in INKO mice may be attributable to changes in intestinal morphophysiology, manifested by the shorter small intestine associated with fibrotic changes. Although NMN administration increased Proglucagon expression and improved postprandial hyperglycemia (Fig. 2i and 2j), neither small-intestine length nor fibrotic alterations were restored in INKO mice (Fig. 3e and 3f). These findings suggest that it is unlikely that alterations in intestinal morphophysiology, including fibrosis, play critical roles in the development of impaired GLP-1 production and postprandial glucose metabolism. The molecular mechanism linking intestinal NAD+ metabolism and morphophysiology is yet to be determined, but possibly involves the sirtuins, NAD+-dependent protein deacetylases. Indeed, 24-month-old intestinal epithelial cell-specific Sirt1 knockout mice displayed the shorter small intestine and colon associated with changes in inflammation and gut microbial composition (61). In addition, intestinal epithelial cell-specific Sirt1 knockout mice exhibited glucose intolerance under caloric restriction in OGTTs, but not in IPGTTs (62). Hence, although we cannot exclude the possibility that sirtuins mediate the metabolic effects induced by defects in NAD+ metabolism in INKO mice, our data indicate that the NAMPT-NAD+-Wnt signaling axes play key roles in maintaining early-phase insulin secretion via their regulation of GLP-1 production.
Our results also illustrate the therapeutic potential of promoting intestinal NAD+ biosynthesis for obesity-associated metabolic derangements. A previous study (39) demonstrated that NAMPT-mediated NAD+ biosynthesis in metabolic organs was compromised by HFD and aging, 2 major risk factors for obesity. It was reported that the downregulation of the major NAD+-generating enzymes, including NAMPT, was accompanied by a decline in NAD+ levels in the colons of aged mice (24). Similarly, we found that HFD impaired intestinal NAMPT-mediated NAD+ biosynthesis associated with reduced GLP-1 production. Although the precise molecular mechanisms by which HFD compromises intestinal NAMPT-mediated NAD+ biosynthesis remain elusive, augmenting intestinal NAD+ biosynthesis by oral administration of NMN increased GLP-1 production in ileal extracts obtained from HFD-fed obese mice. These findings suggest that oral intake of NMN improves obesity-associated dysregulation of whole-body glucose metabolism in an incretin-dependent manner. Interestingly, it has also been reported that NMN elicits insulin secretion in an incretin-independent, beta cell-specific manner, improving glucose metabolism. For example, intraperitoneal injection of NMN increased insulin secretion in response to an intraperitoneal glucose challenge in HFD- and aged-induced diabetic mice (39). In addition, NMN stimulated insulin secretion from primary islet cells (63). Taken together, these findings highlight the promise of NMN in improving postprandial glucose metabolism through both incretin-dependent and incretin-independent pathways. Moreover, in contrast to the effect of oral NMN intake on GLP-1 production in mice, oral administration of nicotinamide riboside (NR), another NAD+ intermediate, did not affect GLP-1 secretion in nondiabetic people with obesity (64). Mechanisms explaining the difference in GLP-1 production between NMN and NR, two key NAD+ intermediates, remain unclear, but may involve the presence of the NMN transporter encoded by Slc12a8, which is expressed at high levels in the small intestine (33). Therefore, we speculate that oral intake of NMN, but not of NR, enhances intestinal NAD+ biosynthesis via NMN transporter in the small intestine, resulting in GLP-1 production. Human studies to investigate the effect of NMN on GLP-1 production are warranted.
In conclusion, the present study provides evidence showing that NAD+ is an important regulator of GLP-1 production and highlights the capability of the intestine to regulate postprandial glucose metabolism. Although further studies are required to determine the precise molecular mechanisms that link intestinal NAD+ biology, GLP-1 production, and whole-body glucose metabolism, it would be of great clinical importance to examine whether boosting intestinal NAD+ biosynthesis by oral intake of NMN is a potential therapeutic approach to the treatment of obesity-induced metabolic derangements.
Abbreviations
- DMSO
dimethyl sulfoxide
- GIP
glucose-dependent insulinotropic polypeptide
- GLP-1
glucagon-like peptide-1
- HFD
high-fat diet
- INKO
intestinal epithelial cell-specific Nampt knockout
- IPGTT
intraperitoneal glucose tolerance test
- NAD+
nicotinamide adenine dinucleotide
- NAMPT
nicotinamide phosphoribosyltransferase
- NMN
nicotinamide mononucleotide
- NR
nicotinamide riboside
- OGTT
oral glucose tolerance test
- RCD
regular chow diet
Acknowledgments
The authors thank Sayuri Suzuki (Department of Emergency and Critical Care Medicine, Keio University School of Medicine) and Asuka Uto (Department of Internal Medicine, Division of Endocrinology, Metabolism and Nephrology, Keio University School of Medicine) for technical assistance.
Financial Support
This study was supported by the Scientific Research Fund of the Ministry of Education, Culture, Sports, Science, and Technology of Japan (18K15399, 21K07377) and research grants from Keio University Medical Science Fund, Research Grants for Life Sciences and Medicine, Japan Diabetes Society Junior Scientist Development Grant supported by Novo Nordisk Pharma Ltd., Boehringer/Lilly Diabetes Research Grant in Japan Diabetes Foundation, MSD Life Science Foundation, Terumo Foundation for Life Sciences and Arts, Kowa Life Science Foundation, Mishima Kaiun Memorial Foundation, Japan Foundation for Applied Enzymology, YOKOYAMA Foundation for Clinical Pharmacology (YRY-2010), Takeda Science Foundation, The 2021 Inamori Research Grant Program (S.Y.), and the Japan Arteriosclerosis Prevention Fund (T.N.).
Author Contributions
S.Y. conceptualized the project. T.N., S.Y., and S.K. designed and performed experiments. T.N. and S.Y. wrote the manuscript. K.H., K.M., and J.I. helped to edit the main manuscript. J.Y. provided scientific suggestions and helped to edit the main manuscript. H.I. provided scientific suggestions, supervision, and detailed comments on the manuscript. All authors reviewed and edited the manuscript.
Disclosure Statement
The authors have nothing to disclose.
Data Availability
Some or all data generated or analyzed during this study are included in this published article or in the data repositories listed in the References. Supplemental material is available at https://doi.org/10.5281/zenodo.6085320.
