Single bolus injection of nicotinamide mono nucleotide increases systemic insulin sensitivity in association with activation of NAD salvage pathway in the liver and adipose tissue in mice

Abstract

Rising nicotinamide adenine dinucleotide (NAD⁺) levels mitigate the onset and progression of age-related diseases including metabolic disorders. Several studies have demonstrated that NAD⁺ levels can be efficiently replenished via nicotinamide mononucleotide (NMN) intake and thereby prevent metabolic disorders. However, the acute effects of NMN administration on metabolism remain unclear. We observed metabolic dynamics after a single bolus injection of NMN. Sirt1 and Nampt mRNA levels were increased in the liver suggesting that intracellular NAD⁺ increased after injection. During OGTT, glucose tolerance and insulin secretion did not change significantly in response to NMN administration, while during ITT, increased insulin sensitivity was observed in muscle. NMN administration decreased serum non-esterified free fatty acid (NEFA) concentrations, which would presumably be responsible for the increased muscle insulin sensitivity. Furthermore, NMN administration reduced the respiratory quotient, confirming that NMN promotes utilization of lipids as an energy source. Our data demonstrate acute effects of NMN on metabolism and raise the possibility of NMN as a treatment for metabolic disorders.

Graphical Abstract

Introduction

Nicotinamide adenine dinucleotide (NAD⁺) is an important coenzyme known to play a role in energy metabolism. In fact, many studies have demonstrated that NAD⁺ levels decrease in liver and muscle in the settings of metabolic disorders such as obesity [1-4]. Nicotinamide mononucleotide (NMN) effectively increases the NAD⁺ concentrations in various tissues in rodent models, which reportedly exerts beneficial effects under many pathological conditions such as obesity- and aging-associated complications [5-8]. Imai et al. made the critical discovery that Sirtuin (Sirt)1 has NAD⁺-dependent protein deacetylase activity, serving as a major catalyst for further research attention to NAD⁺ biology [9]. There is a growing body of evidence demonstrating that NMN exerts beneficial, potentially therapeutic, actions in various disease models, with the effects of NMN on metabolism being of particular note [6, 10]. Most studies have analyzed the effects of multiple or long-term administrations of NMN, and have involved genetically engineered mice, diabetic murine models, and aged mice, with none, to our knowledge, examining in detail the metabolic effects of a single dose on young wild-type (WT) male mice [1, 11-14].

A single bolus injection of NMN reportedly increases glucose-stimulated insulin secretion and improves glucose tolerance in diabetic mice [1, 15]. Long-term administration of NMN to mice was reported to improve mitochondrial function in organs involved in metabolism including the liver [14]. Clinical trials have focused mainly on the safety and pharmacokinetics of NMN [16, 17]. A meta-analysis performed to assess the effects of NAD⁺ suggested that supplementation with NAD⁺ precursors improved dyslipidemia in humans, but also resulted in hyperglycemia [18]. On the other hand, Yoshino et al. found that 10-week NMN supplementation enhanced muscle insulin sensitivity in prediabetic women, as shown employing the hyperinsulinemic-euglycemic clamp and skeletal muscle insulin signaling [19]. NMN apparently improves blood lipid levels, but the effects on glucose metabolism are difficult to interpret because it is influenced by both the duration and the method of administration.

Herein, we analyzed the effects of NMN on glucose and lipid metabolism in WT mice given a single bolus injection of this coenzyme.

Materials and Methods

Animals

Mouse experimental protocols were approved by the Ethics of Animal Experimentation Committee at Yamaguchi University School of Medicine (approval no. J22008). C57BL/6J mice were housed in a temperature-controlled (22°C ± 1°C) room under a 12-h light:12-h dark cycle. Zeitgeber time (ZT) 0 is usually designated as lights on and ZT12 as lights off. The high-fat diet (rodent diet with 60% energy from fat; D12492) was purchased from Research Diet and was freely accessible.

Nicotinamide Mononucleotide administration

NMN was kindly provided by MIRAI LAB. The NMN powder was dissolved in 0.9% saline and administered intraperitoneally at 500 mg/kg body weight. Control mice were intraperitoneally injected with the same volume of 0.9% saline.

Oral glucose tolerance test (OGTT)

Glucose tolerance tests were performed on 12-week-old male mice fasted for 16 h. Saline with or without NMN (500 mg/kg) was injected intraperitoneally at ZT 23 or 11, 6 h before initiating the glucose tolerance test. Glucose was given orally by gavage (2 g/kg) at ZT5 or 17. Tail blood samples were collected at 0 (just before oral administration), 15, 30, 60, 90, and 120 min after oral administration, and blood glucose concentrations were determined using ANTSENCE II (Horiba Industry). Plasma insulin concentrations were measured using an enzyme-linked immunosorbent assay (Morinaga).

Insulin tolerance tests (IP-ITT)

Insulin tolerance tests were performed on random-fed 13-week-old male mice. Saline with or without NMN (500 mg/kg) was injected intraperitoneally at ZT 23 or 11, 6 h before the start of the IP-ITT. Insulin (0.75 units/kg) was injected intraperitoneally, and tail blood was collected.

Serum non-esterified fatty acid (NEFA) and blood lactate assay

Saline with or without NMN (500 mg/kg) was injected intraperitoneally at ZT23. Tail blood was collected from 13-week-old random-fed male mice at ZT5, and then centrifuged for 20 min at 1,200 g. Serum NEFA and blood lactate levels were measured employing an enzyme assay kit (FUJIFILM Wako, Abcam).

Indirect calorimetry

The respiratory quotient (RQ) was measured by indirect calorimetry (MK-5000RQ, Muromachikikai), as previously reported [20].

Western blotting

Random-fed 13-week-old male mice were injected intraperitoneally with saline with or without NMN (500 mg/kg) at ZT23, and then intraperitoneally with bolus saline containing insulin (0.75 units/kg) at ZT1. Liver, gastrocnemius muscle (GM), or epidydimal fat were harvested before (0 min) or 15 min after insulin injection. Total cellular protein was extracted using Cell Lysis Buffer (CST) after homogenization employing a GentleMACS Dissociator (Miltenyi Biotec). 30 μg of the total cellular protein samples were separated by SDS-PAGE. Blotting results were visualized and quantified with an Amersham Imager 680 (Cytiva). Antibodies used included anti-Akt (CST #9272, 1:1,500), anti-Phospho-AKT (Ser473) (CST #9277, 1:1,500), and HRP-conjugated anti-Rabbit (Jackson ImmunoResearch Laboratories #711-035-152, 1:10,000).

RNA isolation and Real-Time RT-PCR

Total RNA extraction from the liver or epidydimal fat was performed with the RNeasy Mini Kit (Qiagen), and cDNA was synthesized with Superscript Ⅱ Reverse Transcriptase (Life Technologies) and subjected to real-time PCR using Power SYBR Green PCR Master Mix (Life Technologies) on an Applied Biosystems StepOnePlus Real Time PCR System (Life Technologies). Each cDNA value was calculated using the Δ Ct method and normalized to the value of the housekeeping gene, Gapdh. The sequences of the primers were as follows:

mSirt1 forward: 5'-TTGACCTCCTCATTGTTATTGG-3'

mSirt1 reverse: 5'-TGTAGATGAGGCAAAGGTTCC-3'

mSirt2 forward: 5'-ATCAGCAAGGCACCACTAGC-3'

mSirt2 reverse: 5'-CGTCCCTGTAAGCCTTCTTG-3'

mSirt3 forward: 5'-GGTGGACACAAGAACTGCTG-3'

mSirt3 reverse: 5'-CCCAGGTGAAGAAGCCATAG-3'

mNampt forward: 5'-AGCGGGGAACTTTGTTACAC-3'

mNampt reverse: 5'-TTTGTCACCTTCCCATTCTTG-3'

mGapdh forward: 5'-AGTATGACTCCACTCACGGCAA-3'

mGapdh reverse: 5'-TCTCGCTCCTGGAAGATGGT-3'

Insulin-mediated glucose uptake measurement

6 h after the NMN administration at ZT23, random-fed 13-week-old male mice were injected intraperitoneally with 2-deoxy-D-glucose (2-DG) (200 mmol/kg) and insulin (0.75 units/kg) at ZT5. Liver GM tissues, and epididymal fat were harvested 90 min after the injection at ZT5 and subsequently homogenized with 10 mM of Tris-HCl (pH 8.0) employing the GentleMACS Dissociator. The supernatants were used for the 2-DG measurement, after centrifugation, employing a 2-DG Uptake Measurement Kit (COSMO BIO #OKPPMG-K01H).

Serum LDL + VLDL level determination

A fluorescence-based assay kit (Abcam) was used serum LDL + VLDL levels according to the manufacturer’s instructions.

Statistical analysis

Two groups were compared using the two-sided paired t-test. All data are presented as means ± standard error. Statistical significance was set at p < 0.05.

Results

Single bolus injection of NMN increases Sirt1 and Nampt mRNA levels

Nicotinamide is converted to NMN by nicotinamide phosphoribosyltransferase (Nampt) in the salvage pathway. It is difficult to measure intracellular NAD⁺ even when using metabolomics-based analytical methods. Therefore, to evaluate the ability of NMN to increase intracellular NAD⁺ levels, we determined Sirts and Nampt mRNA levels in the liver 3 h or 6 h after the single bolus injection of NMN. NMN increased Sirt1 and Nampt mRNA expressions but not those of Sirt2 and Sirt3 mRNA at 3 h after injection, suggesting that NMN increases intracellular NAD⁺ even with only a single bolus injection (Fig. 1). 6 h after injection, Nampt mRNA expression was still significantly elevated, but the increase in Sirt1 mRNA expression was no longer statistically significant (Fig. 1).

Fig. 1. Relative gene expression profile of Sirts and Nampt after single bolus injection of NMN. 0.9% Saline only (Control) or NMN dissolved in 0.9% saline (NMN) were administrated intraperitoneally to mice at ZT23. Liver was harvested 3 h (upper) and 6 h (lower) after administration. n = 4–6. Data are presented as means ± SD. NS means not significant. Significance was determined by paired t-test. *p < 0.05, **p < 0.01.

Single bolus injection of NMN attenuates systemic insulin sensitivity

In OGTT, single bolus injections of NMN 6 h before the tests did not significantly alter either glucose tolerance or insulin secretion (Fig. 2A, B). In the IP-ITT performed at ZT5, blood glucose levels were found to be significantly lower in the NMN group at 60 and 90 min after insulin injection than in the control group (Fig. 2C). In the IP-ITT performed at ZT17, blood glucose levels were significantly lower in the NMN group only at 30 minutes after injection (Fig. 2D). IP-ITT performed at ZT5 on high-fat diet-induced obese mice also showed significantly lower blood glucose levels in the NMN group at 30 and 60 minutes after injection (Fig. 2E). Body weight of obese mice did not differ significantly between control and NMN group (control vs. NMN = 37.2 vs. 37.5 g, n = 6, p = 0.95).

Fig. 2. The effects of bolus injection of NMN on glucose metabolism. (A) OGTT at ZT 5. Mice were fasted from ZT13. At ZT23, control and NMN mice were injected saline or NMN, respectively. Mice received an oral administration of 2 g/kg glucose at ZT5. n = 7. (B) OGTT at ZT17. Mice were fasted from ZT1. At ZT11, they were injected saline or NMN. They received glucose at ZT17. n = 6. (C) IP-ITT at ZT5. At ZT23, random-fed mice were injected with saline or NMN. They were injected with insulin at ZT5. n = 7. (D) IP-ITT at ZT17. At ZT11, random-fed mice were injected with saline or NMN. They were injected with insulin at ZT17. (E) IP-ITT at ZT5 on high-fat induced-obese mice. The protocol was the same as for non-obese mice. The value expressed as a percentage of the blood glucose level before insulin administration. n = 6 or 7. Data are presented as means ± SD. Significance was determined by paired t-test. *p < 0.05, **p < 0.01.

Although IP-ITT revealed that single bolus injections of NMN consistently raised insulin sensitivity significantly, the effect was most evident at ZT5. Therefore, we performed our subsequent analyses at the ZT5 time point.

Single bolus injection of NMN increases insulin sensitivity in muscle but not in liver

The level of phosphorylated AKT (p-AKT) following an insulin bolus injection has been employed as a marker of liver and muscle insulin sensitivity. NMN or saline was injected at ZT23, followed by insulin at ZT5, and the liver and GM were harvested before or 15 min after the ZT5 injection (Fig. 3A). NMN administration did not change the phosphorylation of AKT prior to insulin administration. However, NMN administration decreased AKT phosphorylation in the liver and increased that in GM after insulin injection (Fig. 3A).

Fig. 3. The effects of NMN on insulin signaling and glucose uptake in the liver and gastrocnemius muscle (GM). (A) At ZT23, mice were injected with saline or NMN. Western blot of p-AKT and AKT in the liver and GM isolated 15 min after (bottom) insulin (0.75 U/kg) administration at ZT5. The time course of the experiment is shown on the upper left, and representative Western blot image is shown on the upper right. Relative densitometric bar graphs are shown on the lower. n = 5. (B) Glucose uptake in the liver, GM, and epididymal fat (epi-fat) isolated 90 min after insulin and 2-DG (200 mmol/kg) administration at ZT5. Data are presented as means ± SD. Significance was determined by paired t-test. NS means not significant, *p < 0.05.

To determine insulin-mediated glucose uptake in liver, muscle and adipose tissue, 2-DG containing insulin was injected intraperitoneally at ZT5. NMN significantly increased 2-DG accumulation in the GM, but not in the liver and epididymal fat (Fig. 3B).

Single bolus injection of NMN increases lipids utilization for energy

We then examined the effect of a single bolus injection of NMN on the energy substrate proportion. Administration of NMN at ZT23 did not affect blood glucose levels (Fig. 4A). Blood lactate levels tended to decrease from ZT23 to ZT5, and NMN administration further increased the change, with a significant difference in the amount of change compared to the control mice (Fig. 4B).

Fig. 4. The effects of NMN on the energy substrate utilization. (A, B) At ZT23, mice were injected intraperitoneally with saline or NMN. Blood glucose levels at ZT5 (A) (n = 6), blood lactate levels before (ZT23) and after (ZT5) administration (left), and the change (Δ) of blood lactate levels (right) (B) (n = 7). (C) Serum NEFA concentrations before (ZT23) and after (ZT5) administration (left), and the change (Δ) of serum NEFA concentrations (right). n = 7. Data are presented as means ± SD. Significance was determined by paired t-test. NS means not significant, *p < 0.05, **p < 0.01.

Serum NEFA concentrations tended to increase from ZT23 to ZT5 in the controls, though NMN administration decreased serum NEFA concentrations, and the amount of change differed markedly from that seen in the control mice (Fig. 4C).

RQ is usually higher during the dark phase and lower during the light phase, and would be expected to decrease from ZT0 to ZT6. NMN administration at ZT23 reduced RQ without affecting locomotion (control vs. NMN = 62.9 vs. 52.0/min, n = 4, p = 0.74) and food intake (control vs. NMN = 0.49 vs. 0.62 g/7 h, n = 4, p = 0.71), as compared to control mice, at least from ZT0 to ZT6, confirming that NMN promotes utilization of lipids as an energy source even in response to a single administration (Fig. 5A). Interestingly, in the same time frame (from ZT0 to ZT6) on the day following NMN administration (day2), the RQ of the NMN group was significantly higher than that of the control group (Fig. 5B).

Fig. 5. RQ between ZT22 and ZT6. (A) RQ and average of the RQ for the indicated time period on day1 (injection day) (B) RQ and average of the RQ for the indicated time period on day2 (day after the injection date). n = 5. Significance was determined by paired t-test. NS means not significant, **p < 0.01, ***p < 0.001.

Single bolus injection of NMN decrease NEFA release from adipose tissue in association with increasing insulin sensitivity in adipose tissue

As shown Fig. 4C, NMN administration reduced serum NEFA concentrations, therefore, we analyzed the effect of NMN on NEFA release from adipose tissue and LDL and VLDL release from the liver.

To evaluate the ability of NMN to increase intracellular NAD⁺ levels in adipose tissue, we determined Sirts and Nampt mRNA levels in epididymal fat 3 h after the single bolus injection of NMN. NMN increased Sirt1 and Nampt mRNA expressions but not those of Sirt2 and Sirt3 mRNA at 3 h after injection, suggesting that NMN increases intracellular NAD⁺ even in adipose tissue (Fig. 6A).

Fig. 6. The effects of NMN on the lipid metabolism in adipose tissue and liver. (A) Relative gene expression profile of Sirts and Nampt after single bolus injection of NMN. 0.9% Saline only (control) or NMN dissolved in 0.9% saline (NMN) were administrated intraperitoneally to mice at ZT23. Epididymal fat was harvested 3 h after administration. n = 6. (B) Western blot of p-AKT and AKT in the epididymal fat isolated at ZT5 from mice injected saline (control) or NMN at ZT23. (C) Serum LDL + VLDL concentrations before (ZT23) and after (ZT5) administration (left), and the change (Δ) of serum LDL + VLDL concentrations (right). n = 4. Data are presented as means ± SD. Significance was determined by paired t-test. NS means not significant, *p < 0.05, **p < 0.01.

The level of p-AKT in the absence of insulin bolus injections was used as a marker of insulin sensitivity at basal levels in adipose tissue. NMN administration significantly increased AKT phosphorylation in adipose tissue, possibly followed by reduced NEFA release from adipose tissue (Fig. 6B).

In addition, to assess the contribution of NEFA sources to LDL and VLDL released from the liver, serum levels of LDL and VLDL at ZT23 and ZT5 and their changes were also analyzed. Serum LDL and VLDL levels slightly increased from ZT23 to ZT5 in the controls, whereas NMN administration decreased serum LDL and VLDL levels, and the amount of change was significantly different from that seen in the control mice (Fig. 6C). These findings suggest that a single bolus injection reduces both NEFA release from adipose tissue and lipoprotein (LDL and VLDL) release from the liver.

Discussion

We have herein demonstrated that a single bolus injection of NMN enhances insulin sensitivity but not insulin secretion in young male WT mice. A single bolus injection of NMN has been reported to improve glucose tolerance in Nampt heterozygous knockout mice and β cell-specific Sirt1 overexpressing mice, essentially by promoting insulin secretion [11, 12]. Yoshino et al. showed that NMN administered for longer than one-week enhanced insulin sensitivity in mice that were obese or diabetic [1]. Therefore, we considered it to be highly meaningful to administer a single bolus dose of NMN to young WT mice and analyze the resulting metabolic changes in detail.

This study showed that a single bolus injection of NMN reduced insulin resistance in WT mice, while not significantly changing either systemic glucose tolerance or insulin secretion. Most studies in mice have shown that NMN reduces the pathology of impaired insulin secretion, with only a few investigations demonstrating improved insulin resistance [6, 21]. On the other hand, human studies have revealed long-term administration of NMN to increase muscle sensitivity in healthy middle-aged men and alleviate postprandial hyperinsulinemia in prediabetic women [17, 19]. Our results indicate that the NMN effects observed in humans also occur in mice.

The most significant change was a decrease in serum NEFA, which might be one of the mechanisms enhancing insulin sensitivity and increasing glucose uptake in muscle [22-25]. Mice treated with NMN showed greater decreases in blood lactate levels and remarkable decreases in RQ after NMN administration. These results suggest that a single bolus injection of NMN would enhance lipid catabolism and shift the energy substrate from carbohydrates to lipids, possibly by promoting mitochondrial activity. In fact, long term administration of NMN to WT C57BL/N6 mice decreased RQ and increased mitochondrial oxidative metabolism, with serum NEFA levels tending to be lower in the mice given NMN for 9 or 12 months [21].

It is possible that the reduction of serum NEFA doesn’t contribute to only the activation of lipid catabolism but also the reduction of NEFA release from adipose tissue and LDL and VLDL release from liver. NMN administration increased insulin sensitivity of adipose tissue, which could reduce NEFA release from adipose tissue [26, 27]. In addition, NMN administration decreased serum LDL and VLDL level before insulin bolus injection, suggesting that single bolus injection NMN could decrease serum NEFA through both adipose tissue and liver.

Ramsey et al. found that the levels of Nampt RNA in the livers of WT mice showed a robust diurnal pattern with a peak at the beginning of the dark period, whereas the hepatic levels in Clock^Δ19 mutant mice were lower across the entire light-dark cycle [28]. 6 h after NMN injection, insulin sensitivity in the liver was reduced, contrary to muscle, suggesting that NMN increases glucose flow from liver to muscle. It is very similar to the early active phase alteration commonly seen in humans and mice [24].

Given that NMN increased Nampt mRNA expression in the present study, it might well be more effective to administer it around the onset of the active time period with the aim of increasing Nampt mRNA expression based on physiological conditions. On the other hand, the most prominent effect of a single bolus injection of NMN in this study was the promotion of lipid utilization. Normally, lipid utilization increases during periods of inactivity, followed by an increase in glucose influx from the liver to muscle during the early active phase [23, 24]. Therefore, it is inferred that NMN is more effective when administered before or after the onset of the active phase, even if lipid utilization and glucose flow are taken into account.

In conclusion, a single bolus injection of NMN in mice activates the NAD⁺ salvage pathway in the liver and adipose tissue, thereby increasing cellular NAD⁺ levels. This may result in increased lipid utilization as an energy source, suppression of serum NEFA derived from adipose tissue and liver, and ultimately increase insulin sensitivity in muscle (Graphical Abstract). Although the pharmacokinetics of NMN and NAD⁺ merit further investigation, we hope that our present findings will lead to the further use of NMN as a therapeutic agent for human metabolic diseases.

Graphical Abstract. The working model diagram regarding the acute effect of single bolus administration of NMN on liver, adipose tissue, and muscle metabolism. Briefly, single bolus injection of NMN activates NAD⁺ salvage pathway and then reduces serum NEFA derived adipose tissue and liver, followed by enhanced insulin sensitivity in skeletal muscle about for at least 8 h.

Disclosure

The authors have no known competing financial interests or personal relationships that could have influenced the work reported herein to declare.

Funding

This research was supported by grants, from the Japan Society for the Promotion of Science, 22K08626 (to YO) and 23K06401 (to AT). This work was also supported by research grants from the Japan Association for Diabetes Education and Care (to YO).

Author Contribution

Shunsuke Hiroshige; investigation and writing, Yasuko Kajimura; formal analysis, investigation and editing, Yuko Nagao; investigation, Akihiko Taguchi; conceptualization, funding acquisition and writing, Ryoko Hatanaka; investigation. Chika Yodokawa: investigation, Fujioka Yuka: investigation, Masaru Akiyama: supervision and investigation, Yukio Tanizawa; supervision, Yasuharu Ohta: conceptualization, funding acquisition, formal analysis, investigation and writing. All authors have read and agreed to the published version of the manuscript.