Combination of 5‐aminolevulinic acid and ferrous ion reduces plasma glucose and hemoglobin A1c levels in Zucker diabetic fatty rats

Mitochondrial dysfunction is associated with type 2 diabetes mellitus (T2DM). 5‐Aminolevulinic acid (ALA), a natural amino acid produced only in the mitochondria, is a precursor of heme. Cytochromes that contain heme play an important role in aerobic energy metabolism. Thus, ALA may help reduce T2DM‐associated hyperglycemia. In this study, we investigated the effect of ALA combined with sodium ferrous citrate (SFC) on hyperglycemia in Zucker diabetic fatty (ZDF) rats. We found that the gavage administration of ALA combined with SFC (ALA/SFC) for 6 weeks reduced plasma glucose and hemoglobin A1c (HbA1c) levels in rats without affecting plasma insulin levels. The glucose‐lowering effect depended on the amount of ALA/SFC administered per day. Furthermore, the glucose tolerance was also significantly improved by ALA/SFC administration. Although food intake was slightly reduced in the rats administered ALA/SFC, there was no effect on their body weight. Importantly, ALA/SFC administration induced heme oxygenase‐1 (HO‐1) expression in white adipose tissue and liver, and the induced expression levels of HO‐1 correlated with the glucose‐lowering effects of ALA/SFC. Taken together, these results suggest that ALA combined with ferrous ion is effective in reducing hyperglycemia of T2DM without affecting plasma insulin levels. HO‐1 induction may be involved in the mechanisms underlying the glucose‐lowering effect of ALA/SFC.

Mitochondrial dysfunction is associated with type 2 diabetes mellitus (T2DM). 5-Aminolevulinic acid (ALA), a natural amino acid produced only in the mitochondria, is a precursor of heme. Cytochromes that contain heme play an important role in aerobic energy metabolism. Thus, ALA may help reduce T2DM-associated hyperglycemia. In this study, we investigated the effect of ALA combined with sodium ferrous citrate (SFC) on hyperglycemia in Zucker diabetic fatty (ZDF) rats. We found that the gavage administration of ALA combined with SFC (ALA/SFC) for 6 weeks reduced plasma glucose and hemoglobin A1c (HbA1c) levels in rats without affecting plasma insulin levels. The glucose-lowering effect depended on the amount of ALA/SFC administered per day. Furthermore, the glucose tolerance was also significantly improved by ALA/SFC administration. Although food intake was slightly reduced in the rats administered ALA/SFC, there was no effect on their body weight. Importantly, ALA/SFC administration induced heme oxygenase-1 (HO-1) expression in white adipose tissue and liver, and the induced expression levels of HO-1 correlated with the glucose-lowering effects of ALA/SFC. Taken together, these results suggest that ALA combined with ferrous ion is effective in reducing hyperglycemia of T2DM without affecting plasma insulin levels. HO-1 induction may be involved in the mechanisms underlying the glucose-lowering effect of ALA/SFC. Type 2 diabetes mellitus (T2DM) is a major metabolic disease characterized by both fasting and postprandial hyperglycemia [1]. T2DM involves reduced insulin secretion due to failure of pancreatic b-cell function and insulin resistance in the target organs, such as the adipose tissue, skeletal muscle, and liver [1]. In 2010, approximately 310 million people were diagnosed with T2DM worldwide [2], and this number is expected to increase in the future [1]. T2DM is known to cause various complications such as cardiovascular diseases and retinopathy [3], and the severity of these complications is closely linked to the risks of morbidity and mortality. Thus, the worldwide epidemic of T2DM represents a major health care problem in the 21st century.
Multiple factors, including genetic factors, physical inactivity, and changes in food intake, in combination with a sedentary lifestyle can lead to T2DM [4]. Chronic overnutrition induces intracellular lipid accumulation in the peripheral tissues, which is associated with mitochondrial dysfunction [5]. Mitochondrial dysfunction is linked to decreased ATP production and boxidation and to accumulation of lipid intermediates and reactive oxidative species (ROS) [1,6,7]. There is growing evidence that mitochondrial dysfunction is associated with T2DM [5,[8][9][10]; for example, skeletal muscle biopsy samples from T2DM patients show that mitochondrial complex I activity was reduced by approximately 40% [9]. Moreover, microarray analysis has indicated that the expression of the genes involved in mitochondrial oxidative metabolism is reduced in T2DM patients [8]. In addition, the rates of mitochondrial oxidative phosphorylation in offspring of T2DM patients are reduced [10]. Therefore, improvements in the mitochondrial activity, which can be induced by pharmacological drugs or supplements, may reduce hyperglycemia of T2DM.
5-Aminolevulinic acid (ALA) is a natural amino acid and exists in all kinds of animals and plants [11]. ALA is a common precursor of tetrapyrrole compounds including heme and chlorophylls and synthesized from glycine and succinyl CoA by mitochondrial ALA synthase in animal cells [11]. The polymerization of eight molecules of ALA produces a precursor of heme, protoporphyrin IX (PpIX), which is used as a photosensitizer for photodynamic diagnosis and therapy to identify and kill tumor cells [11][12][13]. Heme is generated by the insertion of ferrous ion into PpIX in the mitochondria [11]. A recent study has suggested that ALA enhances aerobic energy metabolism, especially cytochrome c oxidase activity and protein expression in the mitochondria [14]. In addition, abnormal heme biosynthesis can cause porphyria cutanea tarda and is often associated with T2DM [15]. Furthermore, ALA has been demonstrated to induce heme oxygenase-1 (HO-1) expression in the kidney as well as in cultured cells [16][17][18]. HO-1 is a rate-limiting enzyme in heme metabolism [11], and the upregulation of HO-1 generates cytoprotective products such as bilirubin and carbon monoxide [19]. Interestingly, increased intracellular heme levels lead to upregulation of HO-1 expression [20], and HO-1 has been shown to play a role in reducing hyperglycemia in several diabetes models [21][22][23].
There are two previous large-scale intervention studies in which ALA combined with sodium ferrous citrate (ALA/SFC) was administered to prediabetes volunteers [24,25]. Rodriguez et al. [24] reported that the oral administration of ALA/SFC for 12 weeks improved the glucose levels of 2-h postoral glucose tolerance test (OGTT) in prediabetes volunteers in Hawaii. These improved levels were particularly noticeable in prediabetes subjects with 2-h post-OGTT baseline levels of ≥ 140 mgÁdL À1 (7.8 mmolÁL À1 ), who displayed significantly better glucose-lowering effects compared to those with baseline levels of < 140 mgÁdL À1 . Higashikawa et al. [25] also studied subjects with mild hyperglycemia selected from the general population in Hiroshima, Japan, and reported that the oral administration of ALA/SFC for 12 weeks reduced the fasting and 2-h post-OGTT plasma glucose levels without adverse events. These reports suggest that ALA/SFC improves glucose tolerance in prediabetes subjects. However, how ALA combined with ferrous ion reduces hyperglycemia of T2DM remains to be clarified.
As ALA combined with ferrous ion effectively induces the production of heme [26], we examined the effects of ALA/SFC on the plasma blood glucose and hemoglobin A1c (HbA1c) levels using Zucker diabetic fatty (ZDF) rats, which is a well-known diabetes animal model. Furthermore, we examined the effects of ALA/SFC on the plasma insulin levels, pancreatic bcell mass, and HO-1 expression in various tissues in order to explore possible mechanisms of the reduction in plasma glucose level by ALA/SFC in diabetic conditions.
institutional guidelines on the management and welfare of laboratory animals.

ALA/SFC administration
All experiments involving ALA/SFC administration to the ZDF rats were performed at the same time. During the acclimation period, the ZDF rats were divided (Table 1) to match the average plasma glucose levels, HbA1c levels, body weight, and food intake. ALA, SFC, and pioglitazone were dissolved in distilled water for injection (Otsuka Pharmaceutical Factory, Inc., Tokushima, Japan). Ten-week-old (groups 1-7, set 1) or 11-week-old (group 8, set 2) ZDF rats were orally administered ALA/SFC or a vehicle by gavage for 6 weeks (Fig. 1). The dosing volume was 10 mLÁkg À1 , and the molar ratio of ALA to ferrous ion was 1:0.05. The rats were administered ALA/SFC once daily except for rats administered 150/23.6 mgÁkg À1 ALA/ SFC twice daily. The control pair-feeding group (group 8) was administered the vehicle 1 week after the start of administration of the first set, and were administered the same amount of food (once daily) as the rats administered 300/47.1 mgÁkg À1 ALA/SFC (group 7) in the previous week. Time course blood sampling, measurement of body weight, OGTT, and necropsy were also performed 1 week after the first set. ALA/SFC was administered between 09:00 and 12:00 in all rats. The second administration was performed > 6 h after the first dosing.

Measurement of body weight and food intake
Body weight was measured at least once a week, as well as on each starting day of administration, just before dosing. Food intake was measured once or twice a week. If the food intake was measured twice a week, the food weight was measured at 3-4 day intervals, and the daily food intake was calculated from the difference.

Biochemical analyses
Blood sampling was performed in the morning, after removal of food for 1-2 h. Blood samples were transferred to microfuge tubes and centrifuged at 800 g at room temperature for 5 min. The hematocyte fractions were used for measurements of the HbA1c levels. The plasma was obtained and centrifuged again at 2000 g at room temperature for 10 min and used for measurements of the plasma glucose and insulin levels. The plasma glucose levels were determined by the glucose oxidase method using CicaLiquid GLU (Kanto Chemical, Tokyo, Japan). The HbA1c levels were estimated by an enzymatic method using Norudia N HbA1c (Sekisui Medical Co., Ltd., Tokyo, Japan). The plasma insulin concentration was determined using a rat insulin enzyme-linked immunosorbent assay kit (Morinaga Institute of Biological Science, Inc., Kanagawa, Japan).

Oral glucose tolerance test
An OGTT was conducted 2-3 days after the 6-week blood sampling. After the last ALA/SFC administration, the rats were fasted overnight. On the day of the test, the body Rats were administered ALA/SFC once daily except for rats administered 150/23.6 mgÁkg À1 ALA/SFC twice daily (group 5). Group 8 is the pair-feeding group, which received the vehicle 1 week after the start of administration of the first set (groups 1-7). Group 8 was given the same amount of food once daily as group 7 in the previous week. The number of rats in all groups is eight. The molar ratio of ALA to ferrous ion was 1:0.05. All ZDF rats survived for the duration of the ALA/SFC administration.  Fig. 1. Experimental schedule. The experimental schedules for groups 1-7 (A) and group 8 (pair-feeding group, B) are shown. The ZDF rats were orally administered the experimental treatment for 6 weeks. The plasma glucose and HbA1c levels were measured before the start of administration (0 weeks) and after 3 and 6 weeks. An OGTT was conducted 2-3 days after the 6-week blood sampling. After the OGTT, on the same day, the rats underwent necropsy. Group 8 underwent vehicle administration 1 week after the start of administration of the first set. They were given the same amount of food once daily as the rats administered 300/47.1 mgÁkg À1 ALA/SFC (group 7) in the previous week. The number of rats in all groups is eight. weight was measured and blood samples were taken from the tail vein using heparinized capillary tubes. Glucose (2 g glucose/10 mLÁkg À1 ) was subsequently administered orally, and blood samples were collected at 15, 30, 60, 90, and 120 min after glucose administration.

Measurement of pancreatic b-cell mass
After the OGTT, on the same day, the rats underwent necropsy. The pancreatic b-cell mass was measured as follows: the pancreas was fixed with paraformaldehyde solution and then embedded in paraffin. Five sections from the head to the tail of the pancreas were created and stained with anti-insulin antibody (Dako, Kyoto, Japan). Using a microscope equipped with a 3CCD digital camera (Olympus Corporation, Tokyo, Japan) and image analysis software (FLVFS-LS ver. 1.12; Flovel, Tokyo, Japan), the b-cell area per total pancreatic area was measured (at Gotemba Laboratory, BoZo Research Center Inc., Shizuoka, Japan). The mean areas of the five sections were calculated for each animal, and the b-cell mass per pancreas was calculated using the following formula: b-cell mass per pancreas (mg) = average b-cell area per total pancreatic area 9 pancreatic weight Total RNA extraction and real-time polymerase chain reaction Total RNA was isolated from a tissue sample taken at necropsy for gene expression analysis using an RNA extraction kit (RNeasy Plus Universal Mini Kit, Qiagen, Tokyo, Japan) according to the manufacturer's instructions. Quantitative and qualitative analyses were performed using an Experion automated electrophoresis system (Bio-Rad Laboratories, Hercules, CA, USA). cDNA was synthesized using the SuperScript VILO cDNA Synthesis Kit (Life Technologies, Carlsbad, CA, USA). The expression levels of HO-1 mRNA were measured with TaqMan Gene Expression Assays (Life Technologies, Rn01536933_m1) using the 7500FAST real-time polymerase chain reaction (PCR) system (Life Technologies). One of the cDNA samples was serially diluted and analyzed in parallel to obtain standard curves. The expression levels of all samples were normalized according to the expression levels of an endogenous gene control, b-actin.

Immunoblotting
Liver or epididymal fat was lysed in RIPA buffer (182-02451; Wako Pure Chemicals, Tokyo, Japan) containing 1% Halt Protease Inhibitor Cocktail (1009) for 1 h on ice. After centrifugation (20 400 g, 20 min, 4°C) twice, the protein concentration in the supernatant was measured by Pierce TM BCA Protein Assay Kit (Thermo Fisher Scientific, Rockford, IL, USA) and then the supernatant was diluted with SDS sample buffer. Twenty microgram of liver protein or 16 lg of epididymal fat protein were loaded per lane, fractionated using 4-15% Mini-PROTEANÒ TGX TM Gel (456-1085, Bio-Rad Laboratories), and electrotransferred onto a Trans-BlotÒ Turbo TM Mini PVDF Transfer Pack (170-4156, Bio-Rad Laboratories) by Trans-Blot Turbo transfer system (Bio-Rad Laboratories). Then the membranes were treated with 5% nonfat milk (Wako Pure Chemicals) in TBST (50 mM Tris-HCl pH7.6, 150 mM NaCl and 0.1% Tween 20) for 1 h at room temperature, followed by incubation with antisera raised against HO-1 (kindly provided from Dr. Shigeru Taketani, Department of Biotechnology, Kyoto Institute of Technology) or anti-b-actin antibody (ab8227; Abcam, Cambridge, UK) at 4°C overnight. Furthermore, the membranes were washed three times for 10 min with TBST, incubated with HRP-linked anti-rabbit IgG antibody (NA934; GE Healthcare, Piscataway, NJ, USA) for 1 h at room temperature and washing three times with TBST for 10 min. Then specific protein was visualized with Immuno Star LD (292-69903; Wako Pure Chemicals) using ChemiDoc MP system (Bio-Rad Laboratories).

Statistical analysis
All results are expressed as the mean AE SE. Statistical differences were analyzed using one-way analysis of variance followed by Tukey's test (except for the data in Fig. 7C). For the Fig. 7C, statistical significance was determined using a two-tailed unpaired Student's t-test. P-values < 0.05 were considered statistically significant.

ALA/SFC reduces food intake, but does not affect body weight
We examined the effects of ALA/SFC on body weight and food intake in ZDF rats. A previous report showed that the administration of ALA/SFC to Otsuka Long-Evans Tokushima Fatty (OLETF) rats for 6 weeks reduced plasma glucose levels and improved impaired glucose tolerance [27]. Thus, we used the maximum doses of 300/47.1 mgÁkg À1 ALA/SFC. During administration of ALA/SFC for 6 weeks, the body weight of the ZDF rats slightly increased, but there was no significant difference between the vehicle-administered and ALA/SFC-administered rats ( Fig. 2A). The rats administered pioglitazone, as a positive control, displayed a marked increase in body weight.
During administration of ALA/SFC, the food intake in rats administered 200/31.4 and 300/47.1 mgÁkg À1 ALA/SFC decreased compared with that in rats administered the vehicle (Fig. 2B). At 6 weeks, the food intake in rats administered 300/47.1 mgÁkg À1 ALA/SFC significantly decreased. However, the food intake in rats administered 200/31.4 mgÁkg À1 ALA/ SFC did not significantly differ from that in vehicleadministered rats (P = 0.11). These data indicate that ALA/SFC reduces food intake, but does not affect body weight in ZDF rats.

ALA/SFC reduces blood glucose and HbA1c levels
Next, we investigated the effect of ALA/SFC on the plasma glucose and HbA1c levels in ZDF rats. As shown in Fig. 3A, the administration of 300/ 47.1 mgÁkg À1 ALA/SFC once daily for 6 weeks significantly reduced the plasma glucose levels (P = 0.0012 vs the vehicle). The administration of 200/ 31.4 mgÁkg À1 ALA/SFC seemingly reduced the plasma glucose levels, but the difference was not significant (P = 0.056 vs the vehicle). At 3 weeks, the administration of 300/47.1 mgÁkg À1 ALA/SFC reduced the plasma glucose levels compared to that of vehicle (P = 0.012). The effects of 30 mgÁkg À1 pioglitazone and 300/47.1 mgÁkg À1 ALA/SFC were almost the same at 6 weeks (P > 0.9). In addition, as shown in Fig. 3B, the HbA1c levels in the rats administered 300/ 47.1 mgÁkg À1 ALA/SFC were reduced at 6 weeks compared with those in rats administered the vehicle; however, the difference was not significant (P = 0.065, vs the vehicle, Tukey's test). When Dunnett's test was used, the HbA1c levels in the rats administered 300/ 47.1 mgÁkg À1 ALA/SFC were significantly reduced (P = 0.013).
The glucose-lowering effect of ALA/SFC is not a secondary effect of reduction in food intake Because ALA/SFC administration appeared to the affect food intake in ZDF rats, as shown in Fig. 2B, pair-feeding was conducted to examine the extent to which food intake reduction by ALA/SFC contributed to the plasma glucose-lowering effect. Food intake in the pair-feeding rats without ALA/SFC administration was matched to that in rats administered 300/ 47.1 mgÁkg À1 ALA/SFC. The plasma glucose levels in the pair-feeding rats were slightly, but not significantly, lower than those in the ad-lib feeding vehicle-administered rats (Fig. 4A). In addition, the plasma glucose levels in rats administered 300/47.1 mgÁkg À1 ALA/ SFC were significantly reduced compared with those in the pair-feeding rats. The HbA1c levels in the pairfeeding rats were not significantly different from those in the vehicle-administered rats (Fig. 4B), while the HbA1c levels in rats administered 300/47.1 mgÁkg À1 ALA/SFC were decreased significantly compared with that in the pair-feeding rats (P = 0.010). These results suggest that the glucose-lowering effect of ALA/SFC in ZDF rats is not a secondary effect of the reduction in food intake. rats administered 150/23.6 mgÁkg À1 ALA/SFC once daily, but were not significantly different from those in rats administered the vehicle (P > 0.7) (Fig. 4). However, the plasma glucose and HbA1c levels in the rats administered 150/23.6 mgÁkg À1 ALA/SFC twice daily were markedly reduced (P < 0.001 vs the vehicle) at 6 weeks, and were almost the same as those observed with 300/47.1 mgÁkg À1 ALA/SFC administration once daily (plasma glucose and HbA1c, P > 0.9) (Fig. 4). In addition, the plasma glucose and HbA1c levels in the rats administered 150/23.6 mgÁkg À1 ALA/SFC twice daily were significantly decreased compared with those in the rats administered 150/23.6 mgÁkg À1 ALA/SFC once daily (P < 0.001 and P < 0.01, respectively).
These data suggest that the potency of the glucoselowering effect of ALA/SFC depends on the amount of ALA/SFC administered per day, and it might need more than 6 weeks to obtain a significant reduction in HbA1c level in ZDF rats administered 300/ 47.1 mgÁkg À1 ALA/SFC total per day.

ALA/SFC decreases fasting plasma glucose levels and improves glucose tolerance
Oral glucose tolerance tests were performed to determine whether administration of ALA/SFC for 6 weeks improves glucose tolerance in ZDF rats. The fasting plasma glucose levels were found to be significantly reduced by administration of both 200/31.4 and 300/ 47.1 mgÁkg À1 ALA/SFC (Fig. 5A). Pioglitazone also reduced the levels of fasting plasma glucose. In the OGTT, the glucose tolerance was significantly improved by administration of 200/31.4 and 300/ 47.1 mgÁkg À1 ALA/SFC (Fig. 5B). Consistently, the area under the curve (AUC) of the glucose levels significantly was decreased by administration of 200/31.4 and 300/47.1 mgÁkg À1 ALA/SFC compared to rats administered the vehicle (Fig. 5C). The AUC of the glucose levels was also significantly decreased by pioglitazone. These data indicate that ALA/SFC decreases the fasting plasma glucose and improves glucose tolerance.

Administration of ALA/SFC does not affect plasma insulin levels
We moreover examined the plasma insulin levels before administering ALA/SFC and after 3 and 6 weeks. The plasma insulin levels in rats administered 200/31.4 and 300/47.1 mgÁkg À1 ALA/SFC were not significantly different from those in rats administered the vehicle (Fig. 6A). The plasma insulin levels were also measured during the OGTT; the fasting plasma insulin levels of rats administered 200/31.4 and 300/ 47.1 mgÁkg À1 ALA/SFC were not significantly different from those of the rats administered the vehicle. At 15 min after glucose administration, there were no significant differences in the insulin levels among all groups (Fig. 6B). Furthermore, the average b-cell mass of rats administered 200/31.4 or 300/47.1 mgÁkg À1 ALA/SFC was not significantly different from that of the rats administered the vehicle (Fig. 6C). Thus, these data suggest that ALA/SFC does not affect the plasma insulin levels and b-cell mass in ZDF rats.

ALA/SFC increases HO-1 expression in white adipose tissue and the liver
Increased intracellular heme levels have been shown to lead to upregulation of HO-1 expression [20], and HO-1 is known to play an important role in reducing hyperglycemia in several diabetes animal models [21][22][23]. Thus, we examined the expression levels of HO-1 in various tissues, including the epididymal white adipose tissue (WAT), brown adipose tissue (BAT), quadriceps muscle, and liver. In the liver, the HO-1 mRNA levels were significantly higher in rats administered 300/47.1 mgÁkg À1 ALA/SFC compared with those administered the vehicle (Fig. 7A). In the epididymal WAT, the HO-1 mRNA levels were significantly higher in rats administered 200/31.4 and 300/ 47.1 mgÁkg À1 ALA/SFC compared with those administered the vehicle (Fig. 7A). However, the HO-1 mRNA levels in rats administered 150/23.6 mgÁkg À1 ALA/SFC were not significantly different from those in rats administered the vehicle. The induced expression of HO-1 correlated with the glucose-lowering effect of ALA/SFC. There were no significant differences in the HO-1 expression levels in the BAT and quadriceps muscle between the vehicle-administered and ALA/SFC-administered rats. In addition, the expression levels of HO-1 proteins in the liver and WAT of the rats administered 300/47.1 mgÁkg À1 were higher than those in the vehicle (Fig. 7B and 7C). Accordingly, these data suggest that ALA/SFC upregulates HO-1 expression in the epididymal WAT and liver of ZDF rats.

Discussion
In this study, we examined the effects of ALA/SFC on plasma blood glucose and HbA1c levels using ZDF rats. We found that ALA/SFC administration reduced the plasma glucose and fasting blood glucose levels. These effects were in an apparent dose-dependent manner. Furthermore, ALA/SFC administration reduced the HbA1c levels. ALA/SFC also reduced the food intake, but did not affect the body weight of the rats. In addition, ALA/SFC improved glucose tolerance in an apparent dose-dependent manner, which is consistent with the results of the previous study that found that administration of 300/47.1 mgÁkg À1 ALA/ SFC for 6 weeks in OLETF rats improved glucose tolerance [27]. Finally, ALA/SFC increased HO-1 expression in the WAT and liver. Taken together, these findings indicate that ALA, which is an endogenous d-amino acid, may be effective in reducing hyperglycemia of T2DM.
We considered a possible mechanism underlying the glucose-lowering effect of ALA/SFC. HO-1 has antioxidant activity through bilirubin production, which leads to decreased ROS and consequently protects against oxidative stress-induced cell injury in various tissues including the liver [19,28,29]. A decrease in the accumulated intracellular ROS is associated with enhanced mitochondrial function and insulin signaling [7,[30][31][32]. In fact, upregulation of HO-1 has been demonstrated to improve glucose metabolism and enhance insulin signaling in several diabetes animal models, which in turn is linked to antioxidant mechanisms [21][22][23]. Furthermore, the HO-1 mRNA expression was significantly reduced in T2DM patients, which is associated with dysfunction of the antioxidant defense mechanisms [33]. Our findings show that ALA/SFC upregulates HO-1 expression in the WAT and liver, which might be associated with decreased ROS, and induction of HO-1 could be partially involved in the mechanism underlying the glucose-lowering effect of ALA/SFC. ALA/SFC might induce HO-1 expression by increasing the intracellular content of heme [34]. Increased expression of HO-1 protein is correlated with the HO activity [21,35,36]. ALA combined with ferrous ion induces carbon monoxide, produced from heme through activation of HO-1 in mouse kidney [37]. This report suggests that ALA combined with ferrous ion induces HO-1 activity. Thus, in the next step, the measurement of HO activity in the ZDF rats administered ALA/SFC would in principle be investigated. In addition, the role of increases in HO activity could be investigated by use of HO-1 inhibitors such as zinc protoporphyrin. Furthermore, additional studies using HO-1 knockout mice would be useful to clarify whether upregulation of HO-1 is involved in the glucose-lowering effect of ALA/SFC. The exact molecular mechanisms underlying the upregulation of HO-1 by ALA/SFC remain to be elucidated.
Other mechanisms may be responsible for the glucose-lowering effect of ALA/SFC. Rats fed a normal diet and administered a low-dose ALA for 14 days display increased oxygen consumption and decreased WAT weight [38], suggesting that ALA may promote energy metabolism. Furthermore, ALA increases cytochrome c oxidase activity and enhances aerobic energy metabolism [14]. In addition, ALA induces the expression of cytochrome c oxidase, which is one of the heme-containing proteins [14]. Thus, ALA/SFC may enhance mitochondrial bioactivity, which consequently increases production of ATP and NAD+, decreases ROS production, and upregulates oxidative phosphorylation and the tricarboxylic acid cycle. These metabolic changes can enhance the intracellular glucose consumption, which would result in increased glucose uptake in various tissues such as the adipose tissues, muscles, and liver. However, further studies are required to elucidate the mechanisms underlying the glucose-lowering effect of ALA/SFC.
The glucose-lowering effect of ALA/SFC might be due to the reduction of food intake. However, as shown in Fig. 4, the pair-feeding rats did not show such a remarkable reduction in plasma glucose and HbA1c levels, suggesting that the glucose-lowering effect of ALA/SFC is not a secondary effect of food intake. It has been reported that eating speed is important for glucose metabolism [39]. Therefore, there is a possibility that ALA/SFC might reduce eating speed and thus lower the plasma glucose levels. In this study, we cannot rule out this possibility. The measurement of the precise eating speed of ZDF rats in each group would be required to answer this question.
In this study, there were two reasons why we used the doses of 300/47.1 mgÁkg À1 ALA/SFC. First, the administration of 300/47.1 mgÁkg À1 ALA/SFC to OLETF rats for 6 weeks reduced plasma glucose levels and improved impaired glucose tolerance [27]. Second, in preliminary experiments, low dose of ALA such as 3 or 10 mgÁkg À1 did not reduced plasma glucose levels in ZDF rats (unpublished data). However, the typical intake of ALA from food is approximately 1-2 mgÁday À1 [40]. Thus, ALA 300 mgÁkg À1 would be pharmacologic dose, but not physiologic dose.
In this study, the plasma insulin levels of ALA/ SFC-administered ZDF rats were not different from those of vehicle-administered rats. This finding is consistent with the results of the previous study in which prediabetes patients were administered ALA/SFC for 12 weeks, and the fasting insulin levels were not affected [25]. Decreased pancreatic b-cell mass is associated with development of T2DM [41,42]. Furthermore, pancreatic b-cell mass would be associated with insulin secretion [41][42][43]. In this study, there was no significant difference in the pancreatic b-cell mass between ALA/SFC-administered and vehicle-administered rats. Thus, the glucose-lowering effect of ALA/SFC might not be explained by regulation of plasma insulin level.
As shown in Fig. 2, ALA/SFC decreased food intake in ZDF rats; however, the body weight was not affected. We have considered several possibilities to explain these phenomena. First, gavage administration of ALA/SFC by using a sonde might have caused irritation of stomach, because the pH of ALA/SFC solution in water is low. The irritation of stomach could decrease food intake and also suppress motions, resulting in reduction in energy consumption. Second, ALA/SFC might enhance caloric efficiency, because ALA enhances aerobic energy metabolism [14]. In this case, decreased food intake shall not affect body weight. Third, ALA/SFC might enhance intestinal absorption of nutrients, which results in decrease in food intake but no change in body weight. Although in this study we did not focus on the mechanisms by which ALA/SFC decreased food intake without a significant change in body weight, it would be interesting to examine these possibilities in the future studies. In addition, it is reported that GLP-1 plays important roles in not only insulin secretion but also appetite [44]. Therefore GLP-1 might be involved in decreased food intake by ALA/SFC. Although we did not measured the plasma levels of GLP-1, we found that the plasma levels of insulin in ZDF rats administrated ALA/SFC were not different from those in ZDF rats administered vehicle, implying that ALA/SFC does not upregulate GLP-1 secretion. Thus, GLP-1 may not account for decreased food intake in ZDF rats administered ALA-SFC in this study.
Some studies have reported that ALA induces DNA damage in cultured tumor cells, which may induce cytotoxicity [45,46]. A secondary effect of this cytotoxicity in normal tissues may be responsible for the plasma glucose-lowering effect of 300/47.1 mgÁkg À1 ALA/SFC in ZDF rats. However, it is well known that PpIX preferentially accumulates in tumor cells but not in normal cells [11][12][13]. In fact, because ALA administration leads to accumulation of PpIX in tumor cells [11][12][13], ALA is used for not only photodynamic diagnosis to identify tumor cells, but also used as a photodynamic therapy agent to induce cell death in tumors [11]. This cell death in tumors is probably due to ROS generation by PpIX after exposure to specific wavelength light [47]. Furthermore, PpIX can only induce DNA damage without light excitation in cultured tumor cells, through unknown mechanisms [48][49][50][51]. As heme is generated by insertion of ferrous ion into PpIX [11], the production of ferrous ion from ferric ion is important to reduce PpIX in normal cells; the presence of iron chelators alone in mouse skin promotes the accumulation of PpIX [52]. In contrast, ferrous ion promotes a decrease in intracellular PpIX and increases the production of heme [26,53]. In a previous study, the administration of 300/47.1 mgÁkg À1 ALA/ SFC for 6 weeks in OLETF rats did not show any changes in the mucosal sucrose activity or protein content in the small intestine [27]. In addition, there was no difference in the body weight between ZDF rats administered 300/47.1 mgÁkg À1 ALA/SFC and rats administered the vehicle in this study. Furthermore, to exclude a possibility of the cytotoxicity of ALA/SFC, we measured the plasma levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST). The ALT and AST in ZDF rats administered 300/ 47.1 mgÁkg À1 ALA/SFC were 405 AE 68 (IUÁL À1 ) and 485 AE 42 (IUÁL À1 ) (mean AE SE, n = 8), respectively, while those in ZDF rats administered the vehicle were 400 AE 153 and 371 AE 137, respectively. There was no significant difference between these two groups. Therefore, a dose of 300/47.1 mgÁkg À1 ALA/SFC in rats is not cytotoxic to normal tissues. Collectively, we can hence conclude that the glucose-lowering effect of ALA/SFC in ZDF rats is not a secondary effect of cytotoxicity.
On the basis of our findings, we conclude that administration of ALA/SFC for 6 weeks reduced the plasma glucose and HbA1c levels in ZDF rats. When the ZDF rat dose is converted to the human-equivalent dose based on body surface area [54], the human-equivalent dose appears high. However, administration of even low doses of ALA/SFC to prediabetic populations in USA and Japan for 12 weeks improved glucose tolerance effectively in the previous studies [24,25], implying that much lower doses of ALA/SFC may be effective in humans compared to those in ZDF rats. Of note, there are species-specific differences in iron metabolism. For example, the iron uptake, storage, and flux in humans are different from those in mice [55]. Thus, the difference of the effective dose between rats and humans might be explained by the different effective concentrations of ALA combined with ferrous ion in the target issues in each species.
The circadian rhythm controls numerous physiological processes. These processes include the metabolic pathways such as those related to glucose and lipid metabolism [56][57][58][59]. Dysregulation of the circadian rhythm is a risk factor of metabolic syndromes such as obesity and T2DM, as well as of cardiovascular diseases [58,60]. There is increasing evidence that heme plays an important role in the circadian rhythm [59,61,62]. As ALA is the sole precursor of heme, ALA has been speculated to affect the circadian rhythm through activation of the REV-ERBa nuclear receptor [63]. Thus, the difference of the effective dose in humans and rats might also be explained by differences in the heme biosynthesis and degradation, and in iron metabolism, which may all affect the circadian rhythm. Accordingly, it is important to determine the appropriate dose of ALA/SFC for diabetic patients in clinical trials.

Conclusions
Our results demonstrate that ALA/SFC can effectively reduce hyperglycemia of T2DM. However, further studies are required to elucidate the mechanisms of the glucose-lowering effect of ALA/SFC. Our findings indicate that ALA/SFC may be an effective antidiabetic drug, providing a glucose-lowering effect. Because ALA/SFC reduces the fasting and postprandial glucose levels in prediabetes patients, oral administration of ALA/SFC may represent a new therapeutic approach for not only new-onset T2DM, but also advanced T2DM in humans.
Research Center Inc.) for helpful discussion and conducting the animal experiments.