Effects of 1-Methylnicotinamide (MNA) on Exercise Capacity and Endothelial Response in Diabetic Mice

1-Methylnicotinamide (MNA), which was initially considered to be a biologically inactive endogenous metabolite of nicotinamide, has emerged as an anti-thrombotic and anti-inflammatory agent with the capacity to release prostacyclin (PGI2). In the present study, we characterized the effects of MNA on exercise capacity and the endothelial response to exercise in diabetic mice. Eight-week-old db/db mice were untreated or treated with MNA for 4 weeks (100 mg·kg-1), and their exercise capacity as well as NO- and PGI2-dependent response to endurance running were subsequently assessed. MNA treatment of db/db mice resulted in four-fold and three-fold elevation of urine concentrations of MNA and its metabolites (Met-2PY + Met-4PY), respectively (P<0.01), but did not affect HbA1c concentration, fasting glucose concentration or lipid profile. However, insulin sensitivity was improved (P<0.01). In MNA-treated db/db mice, the time to fatigue for endurance exercise was significantly prolonged (P<0.05). Post-exercise Δ6-keto-PGF1α (difference between mean concentration in the sedentary and exercised groups) tended to increase, and post-exercise leukocytosis was substantially reduced in MNA-treated animals. In turn, the post-exercise fall in plasma concentration of nitrate was not affected by MNA. In conclusion, we demonstrated for the first time that MNA improves endurance exercise capacity in mice with diabetes, and may also decrease the cardiovascular risk of exercise.

It is well known that PGI 2 production is increased during exercise [7,8] and PGI 2 release from the vascular endothelium in response to exercise appears to be an important factor regulating exercise tolerance and exercise capacity [9]. Furthermore, Zoladz et al. [9] have suggested that impairment of the exercise-induced release of PGI 2 may be responsible for the increased cardiovascular risk of vigorous exercise. Since it has been reported that diabetic patients have decreased ability to release PGI 2 during exercise [10], and are characterized by higher cardiovascular risk during vigorous exercise [11] pharmacological stimulation of post-exercise PGI 2 production may prove beneficial.
NO is also involved in the regulation of exercise capacity, and NO generated by NO synthase is metabolized in the body to inorganic anions: nitrite (NO 2 -) and nitrate (NO 3 -) [12]. On the other hand, nitrite may be reduced back to NO by enzymatic and non-enzymatic pathways, particularly in acidic environments with low oxygen availability [12], which occurs during exercise [13]. It has been reported that single bout of strenuous physical exercise had no effect on plasma nitrate concentrations in humans [14]. However, others have demonstrated a small post-exercise increase in plasma nitrate concentrations [15] or increase in plasma nitrite concentrations [16]. Furthermore, exogenous nitrate and the subsequent increase in plasma nitrite concentrations was accompanied by enhanced exercise tolerance in humans [17]. Thus, enhanced NO bioavailability appears to enhance exercise capacity in humans.
We previously showed that endogenous MNA was involved in the regulation of exercise capacity, since the NNMT-MNA pathway was activated by a single bout of strenuous exercise, with an elevated post-exercise plasma concentration of MNA [18]. Considering the pharmacological profile of MNA, including PGI 2 release and improvement of NO-dependent function, one could speculate that MNA supplementation could improve exercise capacity in diabetics and therefore, could be considered as a protective agent against cardiovascular risk during physical activity.
Accordingly, the aim of this work was to characterize the effects of MNA supplementation on exercise capacity and endothelial-, PGI 2 -and NO-dependent response to exercise in diabetic db/db mice. For this purpose, db/db mice were treated with MNA in drinking water (100 mg. kg -1 ) for 4 weeks, their exercise capacity during an endurance running test and post-exercise MNA, nitrite, nitrate and 6-keto-PGF 1α concentrations were subsequently assessed.

Materials and Methods Animals
Male C57BL6/J db/db mice (henceforth referred to as db/db mice) purchased from Charles River Laboratories were housed with five mice per cage and a 12-hours light/dark cycle. Animals had free access to drinking water and standard rodent chow. All procedures involving animals were approved by the Local Bioethics Committee in Krakow, Poland (Permit Number: 914/2012; 127/2014) and conducted in accordance with the institutional guidelines.

Experimental protocol
The scheme of the protocol is presented in Fig 1. 8-week-old db/db mice were randomly assigned into the following experimental groups: sedentary or exercised mice not treated with MNA (sedentary or exercised control) and sedentary or exercised mice treated with MNA (sedentary or exercised MNA). MNA was given in drinking water for 4 weeks at a dose of 100 mg·kg -1 . Mice were weighed once a week in order to adjust the MNA dosage. After 4 weeks of MNA supplementation, the animals assigned into the exercised groups were subjected to endurance running tests as described below.
For the assessment of their running performance capacity, a closed two-line treadmill equipped with an electrode was used (Columbus Instruments, Columbus, OH, USA). Three days before the exercise experiment, the mice were acclimatized. On the first and second day of acclimatization, the mice were placed on the immobile treadmill for 5 min; on the third day, they spent 5 min on the immobile treadmill, followed by 10 min of walking at a velocity of 5 m·min -1 . The exercise capacity of the db/db mice supplemented and non-supplemented with MNA was evaluated by measuring their endurance running time on the treadmill at 5°incline. The treadmill was started at 5 m Á min -1 and the speed was incrementally increased by 1 m Á min -1 every 2.5 minutes to a final velocity of 8 m Á min -1 . The animals were run on the treadmill until they reached fatigue, which was defined as when they being unable to keep running for at least 10 s despite electrical stimulation (current 0.34 mA, voltage 25 V, electrical stimulation frequency 3 Hz). The time from start to finish was recorded. Simultaneously, sedentary mice were placed on the immobile treadmill. The endurance exercise protocol described above was established in the preliminary study.
Immediately after completion of endurance running to fatigue, the mice were anaesthetised with pentobarbital (50 mg·kg -1 , i.p.), and blood samples were then taken from the right heart ventricle on EDTA-anticoagulant and centrifuged (1000 x g, 10 min, 4°C) to obtain plasma samples. After blood sampling, the mice were euthanized with an excessive dose of pentobarbital (100 mg·kg -1 , i.p.). Plasma samples were deep frozen (-80°C) and stored until further analysis. Furthermore, one day before the endurance running test, mice from the sedentary control and MNA groups were placed in individual metabolic cages for 24 h urine sample collection. Urine samples were deep frozen (-80°C) and stored until further analysis.

Blood cell count and biochemical parameters
Blood cell count was assessed by an Animal Blood Counter (ABC Vet, Horiba, Germany). Biochemical parameters including glucose, HbA 1c , creatinine concentrations and lipid profile were determined using automatic biochemistry analyser (ABX Pentra 400, Horiba, Germany).

Measurement of MNA, Met-2PY and Met-4PY
Concentrations of MNA, Met-2PY and Met-4PY in plasma and urine samples were determined using the LC/MS/MS method. Prior to analysis, plasma samples were deproteinized with acidified acetonitrile. Chromatographic analysis was performed on an UltiMate 3000 HPLC system (Thermo Scientific Dionex, Sunnyvale, CA, USA). Chromatographic separation was carried out on an Aquasil C18 analytical column (4.6 mm x 150 mm, 5 μm, Thermo Scientific, Waltham, MA, USA). The mobile phase consisted of acetonitrile (A) and water (B), with the addition of 0.1% formic acid. The flow rate was set at 0.8 ml·min -1 with isocratic elution (A: B, 20/80).
Urine samples were diluted 1:10. HPLC analysis was performed on the Transcend TLX-2 system with an HTS PAL System autosampler (Thermo Scientific). Compounds were separated from the matrix using a TurboFlow Cyclone-P polymer column (0.5 x 50 mm, Thermo Scientific). From the TurboFlow column, the analytes were eluted with acidified acetonitrile onto an Aquasil Detection was performed on a TSQ Quantum Ultra triple quadrupole mass spectrometer (Thermo Scientific) equipped with a heated electrospray ionization interface (HESI II Probe) operating in the positive ion mode. Data acquisition and processing were accomplished using Xcalibur 2.1 software.

Measurement of 6-keto-PGF 1α , nitrite and nitrate in plasma
For the measurement of PGI 2 , the plasma concentration of its stable metabolite 6-keto-PGF 1α was determined using a commercially available ELISA kit according to the manufacturer's instructions.
The concentration of nitrite and nitrate were measured using ENO-20 -NOx Analyzer (Eicom Corp., Kyoto, Japan). The ENO-20 uses a liquid chromatography method with postcolumn derivatisation using Griess reagent. Nitrite and nitrate were separated from other substances in matrices on an NO-PAK column, 4.6x50mm (Eicom Corp.). Nitrate was reduced to nitrite using a cadmium-copper column (NO-RED, Eicom Corp.). Nitrite was mixed with Griess reagent to form a purple azo dye in a reaction coil placed in a column oven at 35°C, and the absorbance of the dye product was measured at 540 nm. The flow of the mobile phase (Carrier Solution) was 0.33 ml·min -1 . The Griess reagent (Reactor A and B Solution) was delivered by the pump at a rate of 0.11 ml·min -1 . The plasma sample was precipitated with methanol at a ratio of 1:1 (v/v), and centrifuged at 10 000 x g for 10 min, and the supernatant was used for analysis.

Statistical analysis
Statistical analysis was performed using GraphPad Prism 5. The area under the curve (AUC) was calculated using the trapezoidal rule in Microsoft Excel. P values < 0.05 were considered as statistically significant.

Discussion
In the present work, we demonstrated for the first time that long-term supplementation with MNA improved endurance exercise capacity in diabetic mice. We suggest that this MNA-induced effect could be linked to PGI 2 and improvement of insulin sensitivity, but not to direct anti-diabetic effects of MNA. It has been previously reported that anti-thrombotic, anti-inflammatory and gastroprotective effects of MNA are mediated by PGI 2 [3][4][5]. In the present study, we demonstrated that there were no statistically significant differences in post-exercise 6-keto-PGF 1α plasma concentrations between untreated and MNA-treated mice (Fig 7A). However, increases in plasma 6-keto-PGF 1α concentrations induced by exercise were remarkably greater in the MNA-treated group (the magnitude of the Δ increase was augmented by approximately 45%). On the other hand, MNA did not appear to have any effects on fasting glucose, HbA 1c concentrations or lipid profile, although an increase in insulin sensitivity was observed (Fig 2A-2C). At the start of current experiment 8-week-old db/db mice were diabetic, as evidenced by significant insulin resistance in comparison with wild-type mice. The lack of the effect of MNA on HbA 1c concentrations but the apparent effect on insulin resistance may be due to the short period of MNA supplementation. It may have been long enough to improve insulin sensitivity but not HbA 1c . On the other hand, changes in HbA 1c concentrations caused by anti-diabetic treatment, e.g. pioglitazone, has been shown to occur after 4-week-long treatment in a rat model of diabetes [19], suggesting that MNA may affect insulin resistance rather than directly causing hypoglycemia, in line with the previous work by Watala et al. [20]. Altogether, our results suggest that the MNA-induced effect on exercise capacity could perhaps be partially linked to the  improvement in insulin sensitivity, although it was most likely associated with PGI 2 -mediated mechanisms.
It is commonly accepted that exercise alone leads to an increase in PGI 2 release as assessed by measuring stable metabolites in human plasma [7,21] and urine [8]. Moreover, the magnitude of the increase in PGI 2 concentration in the interstitial muscle fluid in response to exercise  depends on exercise intensity [22]. Recent data suggest that PGI 2 plays a role in the regulation of exercise capacity, as PGI 2 release in response to exercise was positively correlated with V'O 2max in healthy men [9]. Moreover, the training-induced increase in V'O 2max was accompanied by increased PGI 2 release during exercise in the responders group. Interestingly, in the group of subjects in whom no increase was found in V'O 2max after training (non-responders), no changes were observed in the exercise-induced release of PGI 2 after training [23]. These findings strongly suggest that PGI 2 plays a role in the training-induced regulation of V'O 2max in humans. It is also well known that PGI 2 alone or its stable analogue iloprost are able to increase exercise capacity in patients with pulmonary hypertension [24,25] and stable angina pectoris [26]. It is important to add that, such individuals unaccustomed to habitual physical activities who undertake vigorous exercise have a 50-fold increase in the risk of sudden death and a 100-fold increase in the risk of acute myocardial infarction [11]. For example, patients suffering from diabetes have impaired ability to release PGI 2 during exercise [10] and are characterized by high cardiovascular risk during vigorous exercise [11]. Accordingly, the magnitude of exercise-induced PGI 2 release is an important factor that determines exercise tolerance, as well as the cardiovascular risk of vigorous exercise [9]. PGI 2 -mediated safeguarding effects of MNA on exercise capacity may rely on the protection of coronary, pulmonary and peripheral microcirculation through the inhibition of platelets from forming aggregates during vigorous exercise or/and the improvement of cardiac output [9].
In the present study, we did not find any differences in post-exercise 6-keto-PGF 1α plasma concentrations between untreated and MNA-treated mice (Fig 7A). However, the relative values of Δ increase in plasma 6-keto-PGF 1α concentration induced by exercise, were remarkably greater in the MNA-treated group. It might be that the post-exercise peak of the plasma 6-keto-PGF 1α concentration reflecting PGI 2 production occurs immediately after the end of exercise, and quickly declines. This could explain why we did not see any evidence for the release of PGI 2 by MNA after exercise in the blood taken within 3-5 minutes, the period of time needed for anaesthesia (pentobarbital) and blood sampling. Catheter placement and instant post-exercise sampling would be required to confirm the effect of MNA on PGI 2 release during exercise. Additionally, in contrast to humans, measurement of pre-exercise (baseline) and post-exercise plasma concentrations of 6-keto-PGF 1α in the same mice is technically challenging. It is also important to note that plasma 6-keto-PGF 1α might be quickly metabolized during exercise into 2,3-dinor-6-keto-PGF 1α , and excreted into the urine. It is not possible to collect urine from mice during and immediately after endurance running in order to compare the urine concentrations of 2,3-dinor-6-keto-PGF 1α between MNA-treated and untreated mice after endurance running.
In contrast to the increase in the post-exercise concentration of 6-keto-PGF 1α , the concentration of nitrate decreased. Interestingly, this exercise-induced response was not modified by MNA treatment. Our data seem to be discordant with data from other studies showing an increase or preservation of post-exercise plasma nitrite and nitrate concentrations in healthy humans [14][15][16].
It is well established that under hypoxic conditions, nitrite and nitrate can be reduced back to NO in vivo, thereby being an alternative source of NO for the NOS-dependent pathway [12]. These conditions, with lower oxygen tension, occur in skeletal muscle during exhaustive exercise [13]. In particular, NO 2 --derived NO may be important in the setting of impaired endothelial NO production, as was the case for db/db mice at the age of 12 weeks [27], that were used in the present experiments. Our data showing a pronounced fall in the post-exercise plasma concentration of nitrate may suggest that exercise in db/db mice with endothelial dysfunction may indeed activate the reductive pathway of NO generation, i.e. NO 3 --NO 2 --NO. Accordingly, it appears as though exercise-induced NO formation in diabetic mice was mainly sustained by this reductive pathway, not by endothelial NO production, which was obviously impaired in diabetic mice. If so, it seems obvious that MNA did not modify the post-exercise fall in the plasma concentration of nitrate.
It is well known that the total number of white blood cells is increasing after exercise [28][29][30]. This phenomenon most likely occurs in response to exercise-induced skeletal muscle damage. The post-exercise increase in neutrophil count is correlated with increases in markers of skeletal muscle damage, such as plasma myoglobin concentration and plasma creatine kinase activity [30]. This notion is also supported by reports showing, leukocyte accumulation in exercised muscles, which was associated with a local inflammatory response resulting from exercise-induced muscle damage [31]. Interestingly, the function of the immune system is suppressed by acute bouts of endurance exercise, increasing the susceptibility to upper respiratory illness [32]. On the other hand, leukocytosis may be caused by sympathetic system-mediated mechanisms [33]. In the present study, we demonstrated that MNA decreased post-exercise leukocytosis, suggesting anti-inflammatory or/and anti-sympathetic profile of MNA activity.
Schmeisser et al. [34] has suggested that MNA increased the speed of crawling in nematodes C. elegans by reactive oxygen species (ROS)-dependent mechanism. This group also discovered that MNA, generated through the sirtuin-dependent pathway, extended the lifespan of nematodes by the induction of ROS and subsequent hydrogen peroxide generation by an aldehyde oxidase, GAD-3 [34]. It still remains to be established whether ROS-dependent mechanisms are involved in the MNA-induced effects on exercise capacity in db/db mice.
In conclusion, in the present work, we demonstrated for the first time that long-term supplementation with MNA results in an improvement of exercise capacity in diabetic mice, most likely by PGI 2 -dependent pathways. However, the underlying mechanisms need to be further investigated. As the release of PGI 2 in response to exercise appears to play a role in the regulation of exercise capacity [9], the impairment of exercise-induced PGI 2 release may lead to an increase in cardiovascular risk during high-intensity exercise. We assume that MNA-dependent stimulation of PGI 2 release not only improves exercise capacity in pathological states with impaired endothelial function and compromised exercise tolerance but also protects the coronary, pulmonary and peripheral microcirculation against the formation of platelet microaggregates, thereby preserving adequate tissue perfusion in skeletal muscle, as well as sustaining optimal cardiac output. Safeguarding the pro-aggregatory platelet response seems to be crucial for the safety of exercise in patients at high cardiovascular risk. In summary, we suggest that MNA affords protection against cardiovascular risk caused by long moderate-intensity exercise sessions as implemented in the current study, but may also protect diabetic or cardiovascular patients with impaired endothelial function during exercise of higher intensity and shorter duration. Further studies in humans are warranted to translate our findings to humans.