J Nutr Sci Vitaminol, 58, 415–422, 2012

l-Tryptophan Suppresses Rise in Blood Glucose and Preserves Secretion in Type-2 Diabetes Mellitus Rats

Tomoko Inubushi1, Norio Kamemura2, Masataka Oda3, Jun Sakurai3, Yutaka Nakaya4, Nagakatsu Harada4, Midori Suenaga3, Yoichi Matsunaga3,*, Kazumi Ishidoh2 and Nobuhiko Katunuma2

1 Faculty of Life Science, 2 Institute for Health Sciences, and 3 Faculty of Pharmaceutical Sciences, Tokushima Bunri University, 180 Nishihamabouji, Yamashiro-cho, Tokushima, Tokushima 770–8514, Japan 4 Department of Nutrition and Metabolism, School of Medicine, The University of Tokushima, 3–18–15 Kuramoto-cho, Tokushima, Tokushima 770–8503, Japan (Received May 24, 2012)

Summary Ample evidence indicates that a high-/low-carbohydrate diet increases glucose energy expenditure and is beneficial in patients with type-2 diabetes mellitus (T2DM). The present study was designed to investigate the effects of l-tryptophan in T2DM. Blood glucose was measured by the glucose dehydrogenase assay and serum insulin was measured with ELISA in both normal and hereditary T2DM rats after oral glucose administration with or without l-d-tryptophan and tryptamine. The effect of tryptophan on glucose absorption was examined in the small intestine of rats using the everted-sac method. Glucose incor- poration in adipocytes was assayed with [3H]-2-deoxy-d-glucose using a liquid scintillation counter. Indirect computer-regulated respiratory gas-assay calorimetry was applied to assay energy expenditure in rats. l-Tryptophan suppressed both serum glucose and insulin levels after oral glucose administration and inhibited glucose absorption from the intestine. Trypt- amine, but not l-tryptophan, enhanced insulin-stimulated [3H]-glucose incorporation into differentiated adipocytes. l-Tryptophan increased glucose-associated energy expenditure in rats in vivo. l-Tryptophan-rich chow consumed from a young age preserved the secretion of insulin and delayed the progression of T2DM in hereditary diabetic rats. The results sug- gested that l-tryptophan suppresses the elevation of blood glucose and lessens the burden associated with insulin secretion from b-cells. Key Words l-tryptophan, OGTT, tryptamine, type-2 diabetes mellitus

The main mode of action of currently available medi- amine and (5-HT) modulate insulin secretion cations for type-2 diabetes mellitus (T2DM) is stimula- (9) and glucose uptake into skeletal muscle (10), as well tion of insulin secretion from pancreatic b-cells (1). as melatonin-stimulated insulin secretion (11). Recently, However, such glucose-lowering agents could burden it was reported that oral administration of l-tryptophan b-cells and worsen prognosis. Regular muscle exercise for 15 d did not modify glycermia or insulinemia, but enhances insulin sensitivity and consequently reduces raised melatonin in T2DM model rats (12). the burden on b-cells. Clinical evidence indicates that a In the present study, we report two novel effects for high-protein and low-carbohydrate diet is beneficial in l-tryptophan: l-tryptophan itself decreases glucose the treatment of T2DM (2–4). absorption from the small intestine, and its metabo- Although the high protein in the diet reduced blood lite, tryptamine, increases glucose uptake into adipo- glucose levels, resulting in the reduction of the bur- cytes. Finally, l-tryptophan-rich-chow consumed from den of pancreatic b-cells, its constituent amino acids a young age effectively prevents exhaustion of b-cells have various effects on blood glucose level (5); it was in old T2DM rats. We speculate that l-tryptophan and reported that amino acid combinations such as isoleu- its metabolites play an important role in “the specific cine and serine increased insulin sensitivity (6), whereas dynamic action” and that l-tryptophan is potentially l-glutamine, glycine and threonine decreased maximal useful therapeutically in T2DM. insulin responsiveness of the glucose transport system Experiments in isolated adipocytes (7). The effects of l-tryptophan on blood glucose level have been complex, because not only Experimental animals. Four-week-old male normal l-tryptophan itself upregulates gluconeogenesis in liver Sprague-Dawley (SD) rats from Japan SLC, Inc. and (8), but also its metabolites including 5-hydroxy-trypt- hereditary type-2 diabetes Sprague-Dawley Torii rats (SDT) from CLEA Japan, Inc., which develop spontane- * To whom correspondence should be addressed. ous T2DM, were used in this study. All rats were fasted E-mail: [email protected] at 17:00 for 16-h before the experiments. SDT rats were

415 416 Inubushi T et al.

(A) (B)

250 6000 Glucose D-Tryptophan L-Trptophan(-)

L-Tryptophan L-Tryptophan (ip) ) L-Trptophan (+) 200 mL

) 4000

dL * g/ 150 (m e os * Insulin (pg/ 2000 uc

gl 100 * ood Bl * * 0 50 15 30 45 60 Time after administration (min)

0 0306090 Time after administration (min)

(C)

2500 Glucose L-tryptophan 0.1g/dL L-tryptophan 0.2g/dL L-tryptophan 0.3g/dL D-tryptophan 0.3g/dL Fig. 1. l-Tryptophan alters the uptake of glucose. SD 2000 rats at 5 wk of age were subjected to OGTT [oral glucose (2 g/kg BW) alone (n58), plus l-tryptophan (62.5 mg/

(mg/dL ) kg BW) orally (n56) or intraperitoneally (n57), or 1500 plus d-tryptophan (62.5 mg/kg BW) orally (n55)]. * Blood sugar level (A) and serum insulin level (B). Data in (A) and (B) are mean6SD. * p,0.05, by Wilcoxon-

absorption 1000 Mann-Whitney’s test. Glucose uptake in everted-sac (C). Everted-small intestine isolated from 8-wk-old rat. * Glucose 500 Sac containing 1 mL of saline was incubated in 10 g/dL * ** * of glucose solution add l-tryptophan (0.1–0.3 g/dL) or d-tryptophan (0.3 g/dL) at 37˚C. Glucose concentration 0 in the saline was assayed at 30 min and 60 min. Data 30 60

Incubation period (min) are mean6SD. * p,0.05, and ** p,0.01 by ANOVA. fed normal chow (MF, Oriental Yeast Co., Ltd.), and Wiseman was used for measurement of glucose absorp- l-tryptophan-rich-chow (add 37.5 mg l-tryptophan tion (14). A 20-cm-long portion of the small intestine to 100 g chow, total 280 mg of l-tryptophan in 100 g was isolated from 8- or 9-wk-old SD rats and one end of normal-chow and 317.5 mg of that in 100 g of of the small intestine was bound to form a sac, and the l-tryptophan-rich-chow). The rats were allowed food ad sac was everted so that the serosal layer was inside and libitum. All animal experiments described in this study the mucosal layer was outside. The sac was filled with were approved by the Animal Review Committee of 1 mL of saline and incubated in glucose solution (10 g/ Tokushima Bunri University before the study. dL) at 37˚C under aeration. The saline samples (120 mL) Oral glucose tolerance test (OGTT) and insulin assay. were pumped out from the inside of the sac for 30 and The OGTT was performed after 16-h fasting in SD or 60 min, and glucose concentrations in saline were SDT rats with glucose (2 g/kg body weight (BW)), plus assayed using the Somogyi-Nelson method (15, 16). either l- or d-tryptophan (62.5 mg/kg BW) (12). Blood Preparation of differentiated adipocytes and glucose uptake samples were collected from the inferior vena cava at assay. Preadipocytes 3T3-L1 were obtained from the the indicated times after oral administration of glucose. Health Science Research Resources Bank, Japan (17). Blood-glucose levels in the samples were assayed using a They were inoculated at a concentration of 1.03104 Nipro-Free Meter (Nipro Co., Tokyo, Japan) with a hydro- cells/mL in 24 well gelatin coated plates (IWAKI, Japan) gen electrode and the glucose dehydrogenase method. and cultured for 5-d. The cells were allowed to differenti- Insulin concentration was determined after OGTT using ate in high-glucose Dulbecco’s modified Eagle medium an enzyme-linked immunosorbent assay (ELISA) kit (DMEM) containing 1 mm dexamethasone, 10 mg/mL (Shibayagi Laboratories, Gunma, Japan) according to insulin, 0.5 mm 3-isobutyl-1-methyl-xanthine and 10% the protocol provided by the manufacturer (13). fetal bovine serum (FBS) for another 2-d. The culture Assay of glucose absorption using an everted-sac of rat medium were replaced with high-glucose DMEM con- small intestine in vitro. The method of Wilson and taining 10 mg/mL insulin and 10% FBS, and cultured Beneficial Effects ofl -Tryptophan in Diabetes 417

(A) (B)

cpm(x103) cpm(x103) 10 10 L-Tryptophan Insulin - Tryptamine Insulin - * Insulin + Insulin + 8 8

6 6

ucose uptake 4 4 ucose uptake Gl Gl 2 2

0 0 0110 100 0110 100 Concentration of L-Tryptophan (µM) Concentration of Tryptamine (µM)

Fig. 2. Tryptamine and l-tryptophan modulate glucose uptake. Differentiated adipocytes were cultured in [3H]-2-deoxy-d- glucose plus various concentrations of l-tryptophan (A) and tryptamine (B). Uptake of 2-deoxy-d-glucose was measured by a scintillation counter (n53). for a further 10-d. Cell differentiation was confirmed by RESULTS microscopic examination after Oil Red O staining (Sup- plementary Fig. 1) and these cells were used as the differ- Per-oral-administrated l-tryptophan suppresses hyperglyce- entiated adipocytes. The differentiated adipocytes were mia with glucose load incubated in insulin-free high-glucose DMEM without l-Tryptophan was co-administrated orally in com- FBS for 4-h and washed 3 times with 5 mm potassium bination with glucose, and blood glucose levels were dihydrogen phosphate, 20 mm 4-(2-hydroxyethyl)-1-pi- assayed using tail blood samples every 30 min. l-Tryp- perazineethanesulfonic acid (HEPES) (pH 7.4), 1 mm tophan administration both via oral and intraperito- MgSO4, 1 mm CaCl2, 136 mm NaCl, and 4.7 mm KCl, neal routes suppressed the increase in blood glucose designated as Krebe’s Ringer phosphate HEPES (KRPH) level to levels similar to those after glucose load at 30 buffer (18). The cells were incubated with 0–100 mm and 60 min (Fig. 1A). Interestingly, the suppression by l-tryptophan or tryptamine in the presence or absence d-tryptophan was lower than that by l-tryptophan at all of 10 mg/mL insulin in KRPH buffer at 37˚C for 20 min. time periods of the OGTT. After the addition of [3H]-2-deoxy-d-glucose (75 kBq/ l-Tryptophan suppresses secretion of insulin after glucose mL, Perkin-Elmer, Walton, MA) and 2-deoxy-d-glucose load (100 mm), the cells were incubated at room tempera- Insulin levels was measured before and at 30, 40, ture for 20 min. After washing 3 times with KRPH buf- and 60 min after oral administration of glucose plus fer, the culture medium was replaced with 0.1 n NaOH l-tryptophan. l-Tryptophan significantly suppressed the at 0.5 mL/well for 16-h and the lysates (400 mL) were insulin levels after glucose load at both 30 and 40 min neutralized with 1 m HCl. The samples were added to (Fig. 1B). Clearsol® scintillation cocktail (Nacalai Tesque, Inc., l-Tryptophan inhibits glucose-absorption from small intes- Kyoto, Japan) and incorporated [3H]-2-deoxy-d-glucose tine was measured by a liquid scintillation counter LSC- The amount of glucose absorbed from the small intes- 6100 (Hitachi-Aloka Medica, Tokyo, Japan). tine was quantified directly in the in vitro system of the Assay of energy expenditure in rats. An indirect com- everted small intestine. Glucose added to the mucosal puter-regulated respiration gas-assay calorimeter (Oxy- side was transported to the serosal side. l-Tryptophan, max®: Columbus Instruments, Columbus, OH) was used but not d-tryptophan, inhibited the transport of glucose to assay the energy expenditure in rats. The respiratory in a dose-dependent manner (Fig. 1C). gas was circulated at 2.0 L/min. Either glucose (2 g/kg Tryptamine enhances blood glucose incorporation into adipo- BW) or glucose plus l-tryptophan (62.5 mg/kg BW) was cytes administrated orally. l-Tryptophan, but not d-tryptophan, reduced blood Statistical analysis. Data are presented as mean6 glucose level in OGTT (Fig. 1A). To elucidate the mecha- standard deviation. Differences in blood glucose levels nisms of the effect of l-tryptophan, we analyzed insu- between animal groups were analyzed by the Wilcoxon- lin-stimulated glucose incorporation into differentiated Mann-Whitney test using Kaleidagraph (Ver. 3.6, Syn- adipocytes in the presence or absence of either l-trypto- ergy Co., PA). Group data were compared by one-way phan or its metabolite, tryptamine. Tryptamine, but not analysis of variance (ANOVA). A p value ,0.05 (aster- l-tryptophan, enhanced insulin-stimulated [3H]-glucose isk) or ,0.01 (double asterisks) denoted the presence of incorporation into the cells in a dose-dependent manner a statistically significant difference. All statistical analy- (Fig. 2A and B). Neither d-tryptophan nor 5-HT affected ses were performed using The Statistical Package for insulin-dependent 2-deoxy-d-glucose uptake (Supple- Social Sciences (Ver 13.0, SPSS, Tokyo, Japan). mentary Fig. 2A and B). Indirect respiratory calorimetry revealed that rats fed 418 Inubushi T et al.

8 l Glucose Glucose+L-tryptophan with glucose plus -tryptophan increased their energy 7 * expenditures. When l-tryptophan was co-administered with glucose, the energy expenditure increased signifi- 6 cantly without any exercise compared to glucose only (kcal/kg/h) 5 after 30 min of administration (Fig. 3).

4 Therapeutic effect of l-tryptophan-rich chow in hereditary type-2 diabetes rats 3

expenditure Next, we used l-tryptophan-rich chow in SDT rats 2 to address the effect of l-tryptophan-rich chow on the onset and progression of T2DM. At first, we planned Energy 1 to add 18.75 mg of l-tryptophan to the normal chow 0 (total 298.75 mg/100 g chow, designated as trypto- 03060 phan-moderate-chow), and to feed SDT rats trypophan- Time after administration (min) moderate-chow from 5 wk to 18 wk of age. The results Fig. 3. l-Tryptophan modulates energy expenditure. of OGTT at various ages of SDT rats with tryptophan- Energy expenditure of 5-wk-old SD rats fed glucose moderate-chow did not showed statistical difference (2 g/kg BW) with or without l-tryptophan (62.5 mg/kg from those with the normal chow. Thus, we used tryp- BW, each n55) was measured by an indirect computer- tophan-rich (317.5 mg tryptophan/100 g chow) chow

regulated calorimeter. Data are mean6SD. * p,0.05 by for breeding SDT rats. ANOVA. The mean fasting glucose level in 5-wk-old SDT rats (52.265.2 mg/dL) was similar to that of the control

(A) (B)

300 600

) Tryptophan(-) Tryptophan(+) 500 dL

) 250 g/

dL 400 (m g/

200 e (m os 300 e * uc

os 150 gl 200 uc * gl

100 ood 100 Bl ood Tryptophan (-) Tryptophan (+) Bl 50 0 0306090 120 0 Time after administration (min) 0306090 120 Time after administration (min)

(C) (D)

700 Tryptophan (-) Tryptophan (+) 2000 *P=0.013 600 *P=0.008 )

dL 500 1500 ) g/ (m mL

e 400 os

uc 300 1000 gl

200 * Insulin (pg/ ood Bl 100 * 500

0 0306090120 0 Time after administration (min) Fasten 30 min Fasten 30 min Normal Chow L-Tryptophan-rich Chow

Fig. 4. Glucose tolerance test in SDT rats. SDT rats were bred on normal chow (tryptophan (2), 280 mg/100 g chow, n57) and tryptophan-rich chow (tryptophan (1): 375 mg of l-tryptophan was added to 1 kg chow, total 317.5 mg/100 g chow, n56). They were subjected to OGTT at 5 wk of age (A), 12 wk of age (B), and 18 wk of age (C and D). Blood glucose (A–C) and insulin levels (D) were measured. Data are mean6SD. * p,0.05 by Wilcoxon-Mann-Whitney’s test. Beneficial Effects ofl -Tryptophan in Diabetes 419

SD rats (63.267.1 mg/dL) (Figs. 1A and 4A). A glu- SDT rats on normal chow. To confirm this conclusion, cose tolerance test in SDT rats on normal chow showed we measured insulin contents in the blood of rats with increased glucose levels to 177633.2 mg/dL at 30 min, or without l-tryptophan-rich chow (Fig. 4D). The basal reaching a peak level of 220626.7 mg/dL at 60 min, insulin level in SDT rats fed normal-chow was compa- and 128624.0 mg/dL later at 120 min. These were rable to that in rats fed l-tryptophan-rich chow. Blood comparable to the value at 30 min but significantly insulin level at 30 min after glucose administration in higher than those at 60 and 90 min after administra- the former was lower than that in the latter, though tion in the normal SD rats. These results suggested blood glucose level in the former was higher than that that the potential insulin secretion level in SDT rats in the latter. Thus, l-tryptophan-rich chow consumed was lower than in SD rats at 5 wk of age. No differ- from a young age effectively prevented exhaustion of ences were observed between the levels in SDT rats on b-cells in old-SDT rats, as well as the uptake of glucose l-tryptophan-rich chow and those on normal chow. At from the small intestine in young and aged SDT rats. 12 wk of age, the fasting glucose levels in SDT rats with or without l-tryptophan-rich chow were 8866.89 and 9169.1 mg/dL, respectively (Fig. 4B). In the glucose- tolerance test, blood glucose levels in SDT rats on nor- mal-chow increased to 346635.1 mg/dL at 30 min, reached a peak level of 425658.3 mg/dL at 60 min and decreased to 269692.8 mg/dL at 120 min, whereas the values in SDT rats on l-tryptophan-rich chow were Glucose 282638.3, 356636.2, and 187666.6 mg/dL at 30, Small-Intestine L-Tryptophan 60, and 120 min, respectively. The differences between Inhibition blood glucose levels at 30 and 60 min were significant. Absorption Decarboxylase At 18 wk of age, the fasting blood glucose levels in SDT rats with or without l-tryptophan-rich chow were CO increased (9469.1 and 117621.1 mg/mL, respec- Glucose 2 tively), although the difference between the two groups was not significant (Fig. 4C). In the glucose-tolerance Blood Insulin Glucose Tryptamine test, the blood glucose levels in SDT rats on normal chow Tryptophan increased to 3476130.0 mg/dL at 30 min, reached a Activation peak level of 5176118.0 mg/dL at 60 min, and then Uptake Muscle decreased to 4326114.6 mg/dL at 120 min, while in rats on l-tryptophan-rich chow these levels were 217673.4 mg/dL at 30 min, 3546100.7 mg/dL at 60 min, and 253690.2 mg/dL at 120 min. The differ- Fig. 5. Schematic illustration of the effects of l-trypto- ences between 60 and 120 min were significant. These phan and tryptamine on blood glucose level. See text for details. results suggest possible diminished insulin secretion in After differentiation Before differentiation (Oil Red O staining)

Supplementary Fig. 1. Preadipocytes 3T3-L1 (A) and differentiated adipocytes (B) stained by Oil Red O. The differentiated adipocytes were used in the glucose uptake study. 420 Inubushi T et al.

(A) (B) cpm(x103)cpm (x103) 6 D-Trp 12 5-HT Insulin - Insulin + 5 10

4 ke 8 upta upta ke 3 6

Glucose 4 Glucose 2 Insulin - 2 1 Insulin +

0 0 0204060 0204060 Concentration of 5-Trp (µM) Concentration of 5-HT (µM) Supplementary Fig. 2. Insulin-dependent glucose uptake into the differentiated adipocytes in the presence of various con- centrations of d-tryptophan (A, n53) and 5-hydroxytryptamine (B, n53).

4,500 38mM L-tryptophan 4,000

3,500 19mM L-tryptophan ) 3,000 U 38 mM Glucose

R

(

s 2,500

t i 38mMD-tryptophan n 19 mM Glucose U 2,000

e

s

n

o 1,500

p s 19mM D-tryptophan

e

R 1,000

500

0

-500 200 400 600800 1,000 1,2001,400 1,600 Scan times(s)

Supplementary Fig. 3. Interaction of SGLT-1 and tryptophan. Surface plasmin resonance assay of SGLT-1 conjugated sen- sor chip and glucose, d- and l-tryptophan as analytes were carried out using the BIACore 3000 system.

control blood glucose levels (Fig. 1A). This is based on Discussion the physiological functions of l-tryptophan, which Based on the results of the present study, we propose inhibits glucose absorption from the small intestine. We two physiological functions for l-tryptophan: inhibition tested the interaction of the purified porcine sodium of glucose absorption from the small intestine and stim- glucose transporter (SGLT)-1 (21) and either d-glucose, ulation of blood glucose incorporation into cells. Figure l-tryptophan or d-tryptophan using the BIA Core 3000 5 shows a schematic model of the effects of l-trypto- system. Our preliminary results showed that both d-glu- phan on the absorption of orally administrated glucose, cose and l-tryptophan interacted with SGLT-1 stron- including its metabolite, tryptamine, which stimulates ger than d-tryptophan did (see Supplementary Fig. 3). blood glucose incorporation into the cell. Unfortunately, we failed to obtain the results of compe- Several physiological effects have been reported for tition analysis between d-glucose and l-tryptophan on l-tryptophan, mainly in the veterinary medicine field SGLT-1 because of the small amount of SGLT-1 conju- (e.g., horses, oxen or sheep) with regard to its nutritional gated on the sensor chip. Further analysis of the compe- value, such as its effects on physical abilities (19). The tition assay between d-glucose and l-tryptophan should present study demonstrated that co-administration of be done using a larger amount of SGLT-1 conjugated on l-tryptophan with glucose increased the energy expen- the sensor chip. diture without any exercise (Fig. 3). The results suggest In addition, the results showed that tryptamine, but that l-tryptophan is an amino acid responsible, at least not 5-HT, stimulated the rate of glucose uptake into in part, for the “specific dynamic action” associated with adipocytes (Fig. 2B and Supplementary Fig. 2). Trypt- a high-protein diet (20). amine is rapidly synthesized from l-tryptophan with Oral administration of l-tryptophan was more effec- the activity of aromatic amino-acid decarboxylase (22). tive than intraperitoneal injection with regard to the Accordingly, the effects of l-tryptophan are thought to Beneficial Effects ofl -Tryptophan in Diabetes 421 be mediated by locally-produced tryptamine. Biochemi- support to the inclusion of l-tryptophan in the treat- cally, 1% of l-tryptophan converts to tryptamine (23). ment of T2DM. The results of glucose tolerance tests As shown in Fig. 2B, 10 mm tryptamine significantly in SDT rats of various ages showed l-tryptophan-rich stimulated insulin-stimulated glucose uptake into differ- chow prevented the rise and kept the decrease of blood entiated adipocytes. Under physiological conditions, the glucose level after administration (Fig. 4). Furthermore, blood l-tryptophan level is controlled at approximately the possible serum insulin level after administration was 55 mm (24). Thus, an l-tryptophan-rich diet is neces- higher in SDT rats fed l-tryptophan-rich chow than in sary to achieve effective concentrations of tryptamine. rats fed normal chow. The results suggest that the addi- Tryptamine also modulated insulin secretion, result- tion of l-tryptophan to food is a potentially effective ing in hypoglycemia in normal mice but not in T2DM treatment in these experimental diabetic rats. Hundley mice (25). The effect of tryptamine was inhibited by the et al. (38) reported that administration of tryptophan 5-HT antagonist (26). The ratio of intracellu- increased blood glucose levels in fasting diabetic rats, lar and extracellar 5-HT concentrations of b-cells con- which opposite to the effect seen in our study. The differ- trolled insulin secretion (10). Thus, the effect was pos- ence between the two studies could be due to differences sibly accomplished by intracellular 5-HT converted from in the study design; in our study, l-tryptophan-rich food tryptamine. was provided before the development of T2DM. Consid- Experimental evidence suggests that l-tryptophan ered together, the above results and our findings suggest deficiency modulates glucose tolerance (8). Tryptophan that the l-tryptophan is useful in delaying the progres- enhanced glucose-mediated insulinotropic polypeptide sion of T2DM. (GIP) secretion in tryptophan-deficient or tryptophan- In conclusion, the present study demonstrated that adequate animals (27). However, these reports did not l-tryptophan suppresses the rapid rise in blood glucose focus on functions of monoamine, which is closely after a glucose-rich meal and lessens the burden of related to glucose metabolism (22). Orally administered insulin secretion from b-cells. These results suggest that tryptophan raised melatonin levels in T2DM model l-tryptophan is potentially beneficial in patients with rats (12). Melatonin improved insulin action and b-cell T2DM. Further studies are needed to confirm the benefi- function (11). 5-HT increased uptake of glucose into cial effects of l-tryptophan or its derivatives for dietary the skeletal muscle and liver (9, 28), and then glycogen supplementation in patients with T2DM. synthesis through cyclin-dependent-kinase-5 activation (29). Brain 5-HT content peaked at 1-h after refeeding Conflict of interest: The authors declare no conflict of whereas brain l-tryptophan content increased until 2-h interest. and decreased after 4-h (30). This fact suggested brain REFERENCES monoamine profiles were possibly different from those of l-tryptophan in brain and monoamine in periph- 1) Nagayama D, Saiki A, Endo K, Yamaguchi T, Ban N, eral organs. Further analyses of the roles of brain and Kawana H, Ohira M, Oyama T, Miyashita Y, Shirai K. peripheral monoamines in glucose metabolism need to 2010. Improvement of cardio-ankle vascular index by glimepiride in type 2 diabetic patients. Int J Clin Pract 64: be carried out. 1 1796–1801. l-Tryptophan is also the precursor of NAD in the 2) Katsilambros NL. 2001. Nutrition in diabetes mellitus. 1 liver (22). Although the ratio of NADH/NAD is impor- Exp Clin Endocrinol Diabetes 109: 250–258. tant for glucose metabolism (31), the amount of NADH 3) Stephenson TJ, Setchell KD, Kendall CW, Jenkins DJ, itself also regulated them, because GAPDH activity was Anderson JW, Fanti P. 2005. Effect of soy protein-rich inhibited by NADH attachment (32). Thus, de novo diet on renal function in young adults with insulin- NAD1 synthesis from l-tryptophan possibly participated dependent diabetes mellitus. Clin Nephrol 64: 1–11. in upregulation of glycogen synthesis (29). 4) Andersén E, Hellström P, Kindstedt K, Hellström K. The blood glucose level was controlled by not only 1987. Effects of a high-protein and low-fat diet versus glucose intake but also gluconeogenesis, which set the a low-protein and high-fat diet on blood glucose, serum fasting blood glucose level (2). It was reported that lipoproteins, and cholesterol metabolism in noninsulin- dependent diabetics. Am J Clin Nutr 45: 406–413. l-tryptophan inhibited hepatic gluconeogenesis at the 5) Manders RJF, Wagenmakers AJM, Koopman R, Zorenc level of phosphoenolpyruvate formation (33). An in AHG, Menheere PPCA, Schaper NC, Saris WHM, van vivo study showed a paradoxical effect of l-tryptophan Loon LJC. 2005 Co-ingestion of a protein hydrolysate on the activity of phosphoenolpyruvate carboxykinase, and amino acid mixture with carbohydrate improves a key enzyme for phosphoenolpyruvate synthesis (34). plasma glucose disposal in patients with type 2 diabetes. This effect on the gluconeogenesis of l-tryptophan is Am J Clin Nutr 82: 76–83. active in the normal rats but not in T2DM rats (35). The 6) Marshall S, Monzon R. 1989. Amino acid regulation of effects of l-tryptophan and its metabolites on in vivo insulin action in isolated adipocytes. J Biol Chem 264: gluconeogenesis should be further analyzed. 2037–2042. The current therapies for T2DM are mainly designed 7) Traxinger RR, Marshall S. 1989. Role of amino to stimulate insulin secretion from pancreatic b-cells acids in modulating glucose-induced desensitization of the glucose transporter system. J Biol Chem 264: (36). However, resting pancreas b-cells is a new para- 20910–20916. digm in the treatment of T2DM (37) based on lessen- 8) Wittman JS 3rd. 1976. Alteration of glucose tolerance ing the burden on b-cell function. The above studies add 422 Inubushi T et al.

by dietary L-tryptophan in rats. J Nutr 106: 631–635. analysis of tryptophan metabolites in plasma. Biochem- 9) Hajduch E, Renurel F, Balendaran A, Batty IH, Downes istry 118: 142–146. CP, Hundal HS. 1999. Serotinin (5-hydroxytryptamine), 24) Japan Biochemistry Society (eds). 1979. Data Book of a novel regulator of glucose transport in rat skeletal Biochemistry, p 1548. Tokyo Kagaku Dojin, Tokyo. muscle. J Biol Chem 274: 13563–13568. 25) Sugimoto Y, Kimura I, Yamada J, Watanabe Y, Takeuchi 10) Paulman N, Grohmann M, Voigt J-P, Bert B, Vowinckel J, N, Horisaka K. 1991. The involvement of insulin in Bader M, Skelin M, Fink H, Rupnik M, Walther DJ. 2009. tryptamine-induced hypoglycemia in mice. Life Sci 48: Intracellular serotonin modulates insulin secretion from 1679–1683. pancreatic b-cells by protein serotonylation. PLOS Biol 26) Yamada J, Sugimoto Y, Kimura I, Takeuchi N, Hori- 10: e1000229. saka K. 1988. The hypoglycemia effects of tryptamine 11) Agil A, Rosado I, Ruiz R, Figuueroa A, Zen N, Fernández- in mice: mediation by 5-HT receptors. Eur J Pharmacol Vázquez G. 2012. Melatonin improves glucose homeo- 156: 299–301. stasis in young Zuker diabetic fatty rats. J Pineal Res 52; 27) Ponter AA, Sève B, Morgan LM. 1993. Intragastric 203–210. tryptophan reduces glycemia after glucose, possibly via 12) Tormo MA, Romero de Tejada A, Morales I, Paredes glucose-mediated insulinotropic polypeptide, in early- S, Sánchez S, Barriga C, Hernández R. 2004. Orally weaned pignets. J Nutr 124: 259–267. administered tryptophan and experimental type 2 dia- 28) Moore MC, Kimura K, Shibata H, Honjoh T, Saito M, betes. Mol Cell Biochem 261: 57–61. Everett CA, Smith MS, Cherrington AD. 2005. Portal 13) Kishino C, Kojima M, Kubota T, Hibi N. 1997. Develop- 5-hydroxytryptophan infusion enhances glucose dis- ment of high sensitive quick determination ELISA kit posal in conscious dogs. Am J Physiol Endorinol Metab to insulin determination. The 44th Annual Meeting of 289: E225–231. Japanese Association for Laboratory Animal Science, p 29) Tudhope SJ, Wang CC, Petrie JL, Potts L, Malcomson F, 175 (Abstract). Kieswich J, Yaqoob MM, Arden C, Hampson LJ, Agius L. 14) Wilson TH, Wiseman G. 1954. The use of sacs of everted 2012. A novel mechanism for regulating hepatic glyco- small intestine for the study of the transference of sub- gen synthesis involving serotonin and cyclin-dependent stances from the mucosal to the serosal surface. J Physiol kinase-5. Diabetes 61: 49–60. 123: 116–125. 30) Murtman RJ, Fernstrom JD. 1975. Control of brain 15) Somogyi M. 1952. Notes on sugar determination. J Biol monoamine synthesis by diet and plasma amino acid. Chem 195: 19–23. Am J Clin Nutr 28: 638–647. 16) Nelson N. 1944. A photometric adaptation of the Somo- 31) Ainscow EK, Brand MD. 1999. Internal regulation of gyi method for the determination of glucose. J Biol Chem ATP turnover, glycolysis and oxidative phosphorylation 153: 375–381. in rat hepatocytes. Eur J Biochem 266: 737–749. 17) Wu Z, Xie Y, Morrison RF, Bucher NLR, Farmer SR. 32) Mohr S, Stamler JS, Brüne B. 1996. Posttranslational 1998. PPARg induces the insulin-dependent glucose modification of glyceraldehyde-3-phosphate dehydroge- transporter GLUT4 in the absence of C/EBPa during the nase by S-nitrosylation and subsequent NADH attach- conversion of 3T3 fibroblasts into adipocyes. J Clin Invest ment. J Biol Chem 271: 4209–4214. 101: 22–32. 33) Ray PD, Foster DO, Lardy HA. 1966. Paths of carbon 18) Yang M, Zhang Y, Pan J, Sun J, Liu J, Libby P, Shuhova in gluconeogenesis and lipogenesis. IV. Inhibition by GK, Doria A, Katunuma N, Peroni OD, Gureer-Millo M, L-tryptophan of hepatic gluconeogenesis at the level Kahn BB, Clement K, Shi GP. 2007. Cathepsin L activ- of phosphoenolpyruvate formation. J Biol Chem 241: ity controls adipogenesis and glucose tolerance. Nat Cell 3904–3908. Biol 9: 970–977. 34) Foster DO, Ray PD, Lardy HA. 1966. A paradoxical in 19) Farris JW, Hinchcliff KW, McKeever KH, Lamb DR, vivo effect of L-tryptophan on the phosphoenolpyruvate Thompson DL. 1998. Effect of tryptophan and of glu- carboxykinase of rat liver. Biochemistry 5: 563–569. cose on exercise capacity of horses. J Appl Physiol 85: 35) Alvares FL, Ray PD. 1974. Lack of inhibition by L-tryp- 807–816. tophan or quinolinate of gluconeogenesis in diabetic 20) Stephen M. 2009. Specific dynamic action: a review of rats. J Biol Chem 249: 2058–2062. the postprandial metabolic response. J Comp Physiol B 36) Serrano-Gil M, Jacob S. 2010. Engaging and empower- 179: 1–56. ing patients to manage their type 2 diabetes, Part I: a 21) Halaihel N, Gerbaud D, Vasseur M, Alvarada F. 1999. knowledge, attitude, and practice gap? Adv Ther 27: Heterogeneity of pig intestinal D-glucose transport sys- 321–333. tems. Am J Physiol 277: C1130–1141. 37) Brown RJ, Rother KI. 2008. Effects of beta-cell rest on 22) Keszthelyi D, Troost FJ, Masclee AA. 2009. Understand- beta-cell function: a review of clinical and preclinical ing the role of tryptophan and serotonin metabolism in data. Pediatr Diabetes 9: 14–22. gastrointestinal function. Neurogastroenterol Motil 21: 38) Hundley JM, McDaniel EG, Sebrell WH. 1956. Trypto- 1239–1249. phan-niacin metabolism in alloxan diabetic rats. J Nutr 23) Morita I, Masujima T, Yoshida H, Imai H. 1981. Enrich- 59: 407–423. ment and high-performance liquid chromatography