Cilostazol, a cyclic AMP phosphodiesterase inhibitor, stimulates nitric oxide production and sodium potassium adenosine triphosphatase activity in SH-SY5Y human neuroblastoma cells

Life Sciences, Vol. 65, No. 13, pp. 1413-1422, 1999 Copyright 8 1999 Elswier Science Inc. Printed in the USA. All rights reserved 0024-3205/99/S-see front matter

ELSEVIER

CILOSTAZOL, A CYCLIC AMP PHOSPHODIESTERASE INHIBITOR, STIMULATES NITRIC OXIDE PRODUCTION AND SODIUM POTASSIUM ADENOSINE TRIPHOSPHATASE ACI-IVITY IN SH-SY5Y HUMAN NEUROBLASTOMA CELLS Hidetoshi Inada, Hideo Shindo, Masato Tawata and Toshimasa Onaya Third Department of Internal Medicine, Yamanashi Medical University, Tamaho, Y amanashi, 409-3898, Japan (Received in final form May 27,1999) Summary

Deficiencies in cellular cyclic AMP (CAMP) and nitric oxide (NO) production are We used a thought to be involved in the pathogenesis of diabetic neuropathy. human neuroblastoma cell line, SH-SYSY, to investigate the effect of cilostazol, a specific CAMP phosphodiesterase inhibitor, on NO production and Na+, K+SH-SYSY cells were cultured under 5 or 50 mM glucose for ATPase activity. the cells were then exposed to cilostazol or other chemicals and nitrite, 5-6 days, CAMP and Na+, K+-ATPase activity were measured. In cells grown in 50 mM glucose, cilostazol was observed to increase significantly both NO production and cellular CAMP accumulation in a time- and dose-dependent manner. Cilostazol also significantly recovered reduced levels of protein kinase A activity (PKA) in 50 mM glucose. Furthermore, a PKA inhibitor, H-89 significantly suppressed the increase in NO production stimulated by cilostazol, suggesting Cilostazol did not that cilostazol stimulates NO production by activating PKA. affect either sorbitol or myo-inositol concentrations. Dexamethazone, which is known to induce inducible NO synthase, had no effect on NO production stimulated by cilostazol, suggesting that cilostazol stimulates NO production catalyzed by neuronal constitutive NO synthase (ncNOS) in SH-SYN cells. L-arginine, which is an NO agonist enhanced Na+, K+-ATPase activity in cells grown in 50 mM glucose, NG-nitro-L-arginine methyl ester (L-NAME), which is an NOS inhibitor inhibited basal Na+, K+-ATPase activity in 5 mM glucose and suppressed the increased enzyme activity induced by cilostazol in 50 mM glucose. The above results confirmed our previous observation that NO regulates Nat, Kf-ATPase activity in SH-SYSY cells and suggest that cilostazol increases Na+, K+-ATPase activity, at least in part, by stimulating NO production. The present results also suggest that cilostazol has a beneficial effect on diabetic neuropathy by improving Na+, K+-ATPase activity via directly increasing CAMP and NO production in nerves. Key Words: cilostazol, nitric oxide, cyclic AMP, protein kinase A, sodium potassium ATF’ase activity. SH-SYSY

cell, diabetic neuropathy

The specific CAMP phosphodiesterase inhibitor cilostazol (6-[4-( 1-cyclohexyl-lH-tetrazol-5-yl)butoxyl-3,4-dihydro-2-( lH)-quinolinone), which has an anti-platelet and vasodilative action (l), has been used for the treatment of arteriosclerosis obliterance. Recently, cilostazol has also Corresponding author: Hideo Shindo, M.D. Third Department of Internal Medicme. Yamanashi Medsal University. 409-3898, Japan Tel: +81 552 73 9602. Fax: +8 I 552 73 9685

Tamaho, Yamanashi

1414

Cilostazol Increases NO in SH-SYSYCells

Vol. 65, No. 13, 1999

been reported to have a beneficial effect on symptoms and clinical tests of diabetic neuropathy preliminary clinical open studies (2,3).

in

Many factors have been shown to contribute to the pathogenesis of diabetic neuropathy (4). Using a rat model of diabetes, we previously demonstrated that cyclic AMP (CAMP) concentration is decreased in the sciatic nerve, and that this decrease is related to nerve conduction velocity (NCV) and ouabain sensitive Na f, K+-ATPase (osspATPase) activity in the sciatic nerve These findings suggested that a reduced concentration of CAMP is an of diabetic rats (5-7). important factor in the pathogenesis of diabetic neuropathy. In the same studies, we also observed that cilostszol increased CAMP and improved NCV and osspATPase activity in the sciatic nerve of diabetic rats (5-7). On the other hand, nitric oxide (NO) deficiency was also reported to be an important factor in the pathogenesis of diabetic neuropathy using streptozotosin (STZ)-induced diabetic rats (8). In addition, in a study involving SH-SYSY human neuroblastoma cells as a model of peripheral nerve cells, we observed that a high-concentration of glucose in medium decreases NO production, alters myo-inositol metabolism, and decreases protein kinase C activity, suggesting that all of these processes are involved in the regulation of NO production (9). Recently, CAMP was reported to stimulate NO production via an induction of inducible nitric oxide synthase (iNOS) (10) and cilostazol was found to increase NO production through an induction of iNOS in vascular smooth muscle cells (11). However, the effect of cilostazol on Thus, we investigated the effect of cilostazol NO production catalyzed by ncNOS is unknown. on NO production in SH-SYN human neuroblastoma cells, which express only ncNOS (9), as well as the effect of cilostazol on osspATPase activity in these cells. Methods

1) Cell culture SH-SYSY cells (passage 75-85, kindly provided by Dr. Douglas A Greene, University of Michigan, Ann Arbor, MI) were grown in Dulbecco’s modified eagle medium (D-MEM) containing 10% bovine calf serum, 100 U/ml penicillin and 100 pg/ml streptomycin (ah from The GIBCO BRL, Grand island, NY), and 5 or 50 mM glucose in humidified 10% C@. Cells were plated in a 6-cm dish, and exposed one day medium was changed every 3 days. The cells were then exposed to cilostazol (Otsuka later to 5 or 50 mM glucose for 5-6 days. Pharmaceutical Co. Ltd., Tokushima, Japan), H-89, dexamethazone and L-NAME for the duration indicated in the text, Tables and Figures. 2) Assay procedures NO production in the cultured cells was estimated by measuring the nitrite Nitrite: Nitrite concentrations were measured using calorimetric concentrations in the culture medium. nitrite assay kits (Cayman Chemical Co., Ann Arbor, MI) following the manufacturer’s Briefly, nitrite concentrations were determined by mixing 100 recommended procedure (12). ~1 ahquots of phenol-red-free D-MEM culture media with 50 ~1 of Griess reagents 1 and 2. After 10 min, the difference in absorbance at 650 nm and 540 nm was measured using a microplate reader (Toyo Soda,Tokyo, Japan) and compared to a standard curve generated by O-40 PM NaN02. CAMP: Cells and media were both homogenixd with ethanol (final 50 %) and then centrifuged Supematants were evaporated under a N2 stream and the residues at9OOx gfor5 minat4OC. RIA assays for CAMP were then resuspended in an appropriate volume of imidazole buffer. were performed using commercially-available kits (Yamasa Co., Choshi, Japan) following the manufacturer’s recommended procedure ( 13). Protein kinme A (PKA): buffer A (20 mM Tris-HCl

Cells were collected and lysed using a sonicator while in ice-cold (pH7.2), 0.25 M sucrose, 5 mM EGTA, 1 mM phenylmethylsulfonyl

Vol. 65, No. 13, 1999

Cilostazol Increases NO in SH-SYSY Cells

1415

fluoride, 0.1 mM dithiothreitol (DTT), 0.1 % Triton X- 100, and 10 &ml leupeptin/aprotinin). The lysate was centrifuged at 14,000 x g for 15 min at 4°C and the supematant was kept for an The phosphotransferase activity of PKA was measured using a analysis of enzyme activity. CAMP-dependent protein kinase assay kit (Upstate Biotechnology, NY, USA) according to the Briefly, 500 pmol of a specific PKA substrate, kemptide manufacturer’s instructions. (LRRASLG), was diluted in either 20 yl of assay dilution buffer B (20 mM 3morpholinopropanesulphonic acid (pH7.2), 25 mM B-glycerophosphate, 5 mM EGTA, 1 mM sodium vanadate, and 1 mM MT) or 20 ~1 of assay dilution buffer B containing 20 PM of a PKC inhibitor peptide (PKC substrate, FRARKGALRQKNV) and 20 PM of a calmodulinThe reaction mixture dependent protein kinase inhibitor (calmidazolium chrolide, R24.571). was supplemented with 50 pg of sonicated cellular extract (5.0 pg protein/PI) and 10 ~1 of kinase reagents (75 mM MgCl2, 500 PM ATP, and 10 &i[y-32P]ATP at 3000 Ci/mmol). Kinase reactions were conducted for 10 min at 30°C and 25 ~1 ahquots were then spotted onto Whatman Clifton, NJ) and dried at 25°C. The P81 P8 1 phosphocellulose paper (Whatman, phosphocellulose paper was washed five times with 0.75% H3P04 and then once with acetone. The amount of radioactivity trapped on the P81 phosphocellulose paper was quantified using a liquid scintillation counter. After aspirating the media, the cells were rinsed with Dulbecco’s phosphate-buffered An equal volume of 6% saline (D-PBS), scraped into 1 ml of pure water, and sonicated. perchloric acid and a l/10 volume of 2.5 M KzCO3 were added to the homogenate, and the Sorbitol concentrations were mixture was then centrifuged at 900 x g at 4°C for 5 min. determined using a fluorometric enzyme assay as reported previously ( 14). SorbitoL

A myo-inositol assay was performed according to the spectroIluorometric Myo-inositol: Brielly, cells wcrc enzymatic procedure reported previously (15) with slight modifications. The homogenate was washed with D-PBS, scraped into pure water and sonicated. deproteinized with 0.15 M ZnS04 and Ba(OH)z, and then centrifuged at 1,000 x g. The supematant was mixed with glycine-NaOH buffer (final 60 mM, pH9.5) containing 0.5 mM NAD, and absorbance was measured using a specrophotometer (Hitachi, Tokyo, Japan) both before and one hour following the addition of 0.5 U of myo-inositol dehydrogenase (excitation wave length: 340 nm, emission wave length: 460 nm). Ouabain sensitive Na’, K’-ATPme (osspATPnre) activity : OsspATPase activity was measured Cells were rinsed with D-PBS and scraped into in freshly-prepared homogenates of cells. buffer containing 50 mM imidazole (pH 7.4), 100 mM KCl, 4 mM MgC12, and 1 mM EDTA and sonicated with Sonifier 250 (BRANSON, CT, USA). The crude homogenate was then centrifuged at 200 x g for 5 min at 4°C and the supematant was assayed for osspATPase. OsspATPase assay was performed following an inorganic phosphate assay procedure (16). Briefly, 50 ~1 of supematant was added to incubation buffer containing 50 mM imidazole (pH7.4) with or without 1 mM ouabain, and the reaction was initiated by adding ATP to a final concentration of 2 mM in a total volume of 500 ~1. After 10 min incubation at 37-C, the reaction was terminated by adding 250 ~1 of 5% SDS containing 90 mM EDfA. Then, 750 ~1 of a reducing agent solution containing 3.5% ferrous ammonium sulfate, 1.0% thiourea, 1.0% H2SO4, 150 ,ul of an ammonium molybdate and 9% H2SO4 were added to 750 ~1 of solution to be assayed. After 10 min incubation at room temperature, the absorbance at 750 nm was measured using a spectrophotometer (Hitachi, Tokyo, Japan). Protein: Results were normalized against cell protein concentrations. Cellular protein concentrations were determined by the BCA method using commercially available kits according to the manufacturer’s recommended procedure ( 17).

3) Statistical analysis Results are shown as the mean f SEM. We repeated each experiment twice and obtained almost same results. Statistical analyses were performed on samples (n=6 or 8) determined from individual wells in a single experiment. Differences among groups were evaluated by one-factor (Table II) and two-factor (Fig. l-4 and Table I) analysis of variance (ANOVA)

1416

Cilostazol Increases NO in SH-SYSY Cells

followed by post hoc multiple comparison (Abacus Concepts, Inc. CA,USA).

test (BonferroniDunn

Vol. 65, No. 13, 1999

test) with Statview software

Results

Efect oJ‘cilostazo1 on nitrite and CAMP concentrations In our preliminary experiments, we confirmed that cilostazol has no direct effect on nitrite assay. The concentration of nitrite was significantly higher (p
than in in cells 50 mM did not

.Fffect of dexamethazone on nitrite levels stimulated by cilostazol Two and a half PM of dexamethazone, which is known to inhibit the expression of iNOS mRNA, Dexamethazone had no effect on the concentration was incubated with cilostazol for 24 hours. of nitrite stimulated by cilostazol in cells grown in either 5 or 50 mM glucose (Fig. 3). Ejfect of cilostazol and /or NOS inhibitor on osspATPase activiry OsspATPase activity in cells grown in 50 mM glucose was significantly lower than in cells grown Following a 6 hour incubation, the NO agonist, L-arginine (2 mM) in 5 mM glucose. significantly recovered impaired osspATPase activity in cells grown in 50 mM glucose (data not shown). In contrast, the NOS inhibitor, L-NAME (10 @I) significantly inhibited osspATPase Cilostazol significantly increased impaired activity in 5 mM glucose (Fig.4)@0.05). osspATPase activity in cells grown in 50 mM glucose, which was significantly inhibited by LNAME(1x0.05) (Fig.4). Discussion

Two main hypotheses have been proposed for the pathogenesis of diabetic neuropathy. One hypothesis states that metabolic derangement such as acceleration of the polyol pathway and The other states that reduced decreased concentration of CAMP lead to diabetic neuropathy. peripheral nerve blood flow and consequent endoneurial hypoxia play a major role in the Many investigators have reported a beneficial effect of pathogenesis of diabetic neuropathy. Cilostazol has been reported to improve cilostazol on diabetic neuropathy in animal studies. nerve function via metabolic improvements, such as an increase in CAMP concentration (5,7,18).

Cilostazol

Vol. 65, No. 13. 1999

Increases

1417

NO in SH-SYSY Cells

(a) 50Z '5

40 -

5 L % 020.

E 6 9) .E

IO-

.z n

“A

o-l 0

1

6

12

24

I

I

1

t

I

0

1

6

12

24

Time(h)

Time (h)

FIG. 1 Effect of cilostazol

on nitrite (a) or CAMP(~)

contents in SH-SYSY cells. 10 FM of cilostazol

Cells were cultured at 5 or SO mM glucose for 6 days.

or vehicle were

added_ and nitrite were determined at the time indicated. + 50 mM glucose for both (a) and (b). U 5 mhl glucose S=h for each point * t

: significant diffcrcnce YS 0 hour in 5 mhl glucose (p < 0.05). : significant difference vs 0 hour in SO mM glucose (p < 0.05).

(b)

(a)

10-7

0

10-s

10-s

0

10-4

10-6

10-5

10-4

cilostazol (M)

cilostazol (M)

FIG. Effect of cilostazol

10.7

2

and nitrite (a) or CAMP (b) contents in SH-SYSY

cells.

Cells were cultured at 5 or 50 mM glucose for 6 days and treated with the indicated concentration of cilostazol for 12 hours. + 5 mM glucose

+

50 mM glucose for both (a) and (b).

N=6 for each point * t

: significant : significant

vs without

cilostazol

in 5 mM glucose (p < 0.05).

difference vs without

differmce

cilostazol

in 50 mM glucose (p < 0.05).

Cilostazol Increases NO in SH-SY5Y Cells

1418

Vol. 65, No. 13, 1999

TABLE I Effect of Cilostazol or Protein Kinase A Inhibitor, H-89 on Nitrite Concentration and Protein Kinase A Activitv nitrite (% of 5 mM glucose)

5 mM glucose

Protein kinase A (% of 5 mM glucose)

N

H-89 (-)

10 PM H-89

H-89 (-)

6

100

67.6 + 3.9”

100 47.0 f 3.1 *

# 35.3 + 1.3*

80.3 f 4.8 #

39.2 f 1.2*t

50 mM glucose

6

52.8k3.1

*

# 34.2 f 1.7*

50 mM glucose + 10 PM cilostazol

6

7x.5 + 4.4 #

41.9 + 2.3*t

Cells were cultured

in 5 or SO mM glucose for 6 days and then incubated

for 6 hours.

Values arc expressed

mill glucose

without

glucose

11-89.

+ 10 PM cilostazol

Significance

was determined

#

as the mean -c SEM.

N: number

78.8*3.1”

1~1-89. n.s.: not significant

7: p
vs SO mM glucose

by ANOVA and post hoc multiple comparison

I ,uM H-89 : p < 0.05YS5

with or without

of samples*

: p < 0.05 vs SOmM glucose without 11-89.

without

10 ,uM H-89

without

II-X9.

trst (Bonfcroti;Dumr

test).

TABLE II Effect of Cilostazol on Sorbitol and Myo-inositol

Concentrations

N

sorbitol (nmollmg protein)

myo-inositol (nmohmg protein)

5 mM glucose

6

1.08 + 0.07 *

44.28 + 2.35 *

50 mM glucose

6

3.27 + 0.24

21.04 + 0.78

50 mM glucose + 10 LIM cilostluol

6

3.09 f 0.25n.s.

23.30 + 1.59n.s.

Cells wcrc cultured

in 5 or 50 mM glucose for 6 days and treated with or without 10 ,uM cilostazol : p < 0.05 vs for 12 hours. Values are expressed as the mean + SEM. N: number of samples.* SO mM glucose without cilostazol. n.s. : not significant vs SO mM glulcose without cilostazol. Significance was determined by ANOVA and post hoc multiple comparison test (BonferomriiDunn test).

1419

Cilostazol Increases NO in SH-SYSY Cells

Vol. 65, No. 13, 1999

r

5 mM glucose

n.s.,

r

50 mM glucose

n.s.,

50 mM glucose + 10 pM cilostazol

FIG. 3 Effect of dexamethazone

on NO production

in SH-SYSY

cells.

Cells were cultured at 50 mM glucose for 6 days and treated with 2.5 yM dexamethazone N=6 without dexamethezone n.s.: not significant with 2.5 ,uM dexamethezone

for 24 hours

m

F= ? ‘L +y 5 200 “n + ;z g% gf 3E 100 E 4

a

0 .

5 mM glucose

50 mM glucose 50 mM glucose + 10 pM cilostazol

FIG. 4 Effect of cilostazol

or NOS inhibitor,

L-NAME on Na*, K+-ATPase activity.

5 or 50 mM glucose and treated with 10 PM cilostazol N=6 m without L-NAME l : p < 0.05 m with 1OpM L-NAME

or 10 ,uM of L-NAME

Cells were cultured at for 6 hours.

1420

Cilostazol

Increases

NO in SH-SYSY

Cells

Vol. 65, No. 13, 1999

On the other hand, other investigators reported that cilostazol improved impaired nerve function, such as osspATPase activity, and nerve structure by increasing nerve blood flow (19-21). However, the effect of cilostazol in these studies on nerve blood flow could not be kept separate from metabolic effects. In the present study, we investigated the effects of cilostazol on NO production and osspATPase activity in SH-SYN cells by focusing on the metabolic effects of cilostazol .

In general, cellular concentrations of nitrite and cyclic GMP (cGMP) are measured as indicators of NO production. However, since we found that cilostazol has a weak inhibitory effect on cGMP phosphodiesterase activity in SH-SYN cells in a preliminary experiment, we believed that cGMP content is not a reliable indicator of NO production in the present study. Thus, we assayed only the concentration of nitrite as an indicator of NO production. In our previous report (9), we found that glucose levels in the medium lowered NO production in a dosedependent manner from 5 to 100 mM glucose in SH-SYSY cells. Thus, present experiments were carried out using concentrations of 5 and 50 mM glucose in media as in the recent paper (22). The present findings, wherein cilostazol led to increased NO production, an accumulation of CAMP and increased PKA activity and further that H-89 inhibited cilostazol-stimulated NO production taken as a whole suggest that CAMP increased by cilostazol stimulates NO production. The concentration of CAMP is known to be regulated by the relative activities of adenylate cyclase and phosphodiesterase In another series of experiments, we found that whereas adenylate cyclase activity is decreased, phosphodiesterase activity is not altered in cells grown in high glucose medium (H Shindo et al. paper in preparation). We also confirmed that adenylate cyclase activity is impaired by high glucose in the present study (data not shown). In the present,experiments, cilostazol was seen to increase CAMP concentration in cells both at normal and high concentration of glucose by inhibiting phosphodiesterase activity. Consequently, cilostazol increased CAMP concentration two fold in cells both at normal and high concentration of glucose. Thus, the effect of cilostazol and glucose on CAMP accumulation are different and independent. Although many investigators reported that CAMP stimulates the expression of iNO! mRNA and increases NO production (10) and further that cilostazol increases NO production via expression of iNOS mRNA in vascular smooth muscle cells (1 l), no previous report has demonstrated that cilostazol or CAMP increases NO production in cells that express only ncNOS, a finding which we confirmed in a previous study (9). The observation that dexamethazone did not inhibit NO production stimulated by cilostazol, also supports the conclusion that cilostazol increased NO Furthermore, in a recent study we found production catalyzed by ncNOS in SH-SYSY cells. that CAMP analogs or agonists such as dibutyryl CAMP and beraprost sodium (prostacyclin analog) stimulate NO production catalyzed by ncNOS in SH-SYN cells (22). Although the precise mechanism by which cilostazol increases NO production in SH-SYSY cells remains unclear, the results of the present study suggest that cilostazol stimulates NO production PKA was reported to stimulate myo-inositol transport (23). by regulating PKA activity. Although the increase in CAMP observed in the present study in SH-SYN cells was accompanied by an increase in NO production and PKA activity, cilostazol was not found to affect sorbitol or Thus, we believe that PKA may directly affect NO regulating factors via a myo-inositol levels. pathway different from the myo-inositol-PI synthesis pathway, a conclusion which we reported In the present study, the accumulation of CAMP was not always accompanied previously (9). by the same magnitude of increase in other parameters such as PKA, NO or osspATPase activity. As we previously observed (22) that the relative increase in PKA activity paralleled osspATPase activity and NO production, we believe that this discrepancy can be located between CAMP and The relative increase in PKA to CAMP at 5 mM was lower than at 50 mM PKA activity. Although we can not sufficiently explain this discrepancy, it is possible that CAMP glucose. The different response to accumulations exceeded the maximum capacity of PKA activity. cilostazol between the increase in CAMP and nitrite at 5 mM in Fig. 2 can also be explained by the same mechanism. We previously reported that cilostazol treatment increased CAMP content and recovered impaired Similar to the results found in animal osspATPase activity in sciatic nerve of diabetic rats (7). models (7,21), CAMP accumulation and PKA activation by cilostazol in SH-SYN cells increased

Vol. 65, No. 13, 1999

Cilostazel Increases NO in SH-SYSYCells

1421

PKA has been observed to regulate osspATPase activity in many osspATPase activity. In contrast, NO has been found to stimulate osspATPase activity different cell lines (24). Additionally in the present study, L-NAME significantly, albeit directly in vascular wall (25). We observed a similar not completely, inhibited increased osspATPase activity by cilostazol. Although the precise mechanisms by which CAMP or NO result in a previous experiment (22). modulates osspATPase activity are not clearly known, the results so far indicate that cilostazol stimulates osspATPase activity via NO and partially via the direct effect of PKA activation. Furthermore, as cilostazol exerts a weak inhibitory effect on cGMP phosphodiesterase activity and further that cGMP- cyclic GMP dependent protein kinase (PKG) reportedly stimulates osspATPase activity (26), cilostazol may also stimulate osspATPase activity partially via a cGMP-PKG effect which is not mediated by NO. In the present study, we confirmed the metabolic effects of cilostazol via NO production on Since the concentration of cilostazol in human plasma in clinical use is 2-3 nerve cell function. PM (27), our study indicates that cilostazol may have a beneficial effect on diabetic neuropathy via improving impaired nerve function not only by increasing nerve blood flow but also by directly increasing CAMP and NO in such peripheral nerve tissues as SH-SY5Y cells. Acknowledgement

We thank Miss M. Orii and Mrs. Y. Satoh for their expert secretarial assistance. References

1. H. UMEKAWA, T. TANAKA, Y. KIMURA and H. HIDAKA, Biochem. Pharmacol. 33 3339-3344 (1984). 2. M. SEKIGUCHI, A. MORIKAWA, K. NAKAJAMA, H. ITO, M. TOSHIMA and I. MAKINO, Jpn. Pharmacol. Ther. 19 3237-3277 (1991). 3. Y. OKUDA, M. MIZUTANI, I. ISAKA, T. IKEGAMI, M. SUMI, K. SATO, C, BANNAI, E. UENO and K. YAMASHITA, Anzneim-ForschlDrug Res. 42 540-542 (1992). 4. D.A. GREENE,A.A.F. SIMA, M.J. STEVENS,P.D. KILLEN, D.N. HENRY,T.P. THOMAS, J. DANANBERG and S.A. LATTIMER, Diabetes Metab Rev 9 189-2 17 (1993). 5. H. SHINDO, M. TAWATA, K. AIDA and T. ONAYA, J. Clin. Endocrinol. Metab. 74 393398 (1992). 6. H. SHINDO, M. TAWATA and T. ONAYA, J. Endocrinol. 136 431-438 (1993). 7. H. SHINDO, M. TAWATA and T. ONAYA, Endocrinology 132 510-516 (1993). 8. M.J. STEVENS, J. DANANBERG, E.L. FELDMAN, S.A. LATI’IMER, M. KAMIJO, T.P. THOMAS, H. SHINDO, A.A.F. SIMA and D.A. GRRENE, J. Clin. Invest. 94 853859 (1994). 9. H. SHINDO, T.P. THOMAS, D.D. LARKIN, A.K. KALIHALOO, H. INADA, T. ONAYA, M.J. STEVENS and D.A. GREENE, J.Clin. Invest. 97 736-745 (1996). 10. T. IMAI, Y. HIRATA, K. KANNO and F. MARUMO, J. Clin. Invest. 93 543-549 (1994). 11. U. IKEDA, M. IKEDA, S. KANO, T. KANBE and K. SHIMADA, Eur. J. Pharmacol. 3 14 197-202 (1996). 12. L.C. GREEN, D.A. WAGNER, J. GROGOWSKI, P.L. SKIPPER, J. S. WISHNOK and S.R. TANNENBAUM, Anal. B&hem. 126 131-138 (1982). 13. M. HOMMA, K SATOH, L. TAKEZAWA and M. UI, B&hem. Med. 18 257-273 (1977). 14. J.I. MALONE, G. KNOX, S. BENFORD and T.A. TEDESCO, Diabetes 29 861-864 (1980). 15. A. WEISSBACH, Myoinositol, H.U. Bergmeyer (ed.), 1333-1336, Verlag Chemie, Weinheim (1974). 16. C. HEGYVARY, K. KANG and Z. BANDI, Anal. Biochem. 94 397-401(1979). 17. P.K. SMITH, R.I. KROHN and A.K. HERMANSON, Anal. Biochem. 150 76-85 (1985). 18. K. NAKA, H. SASAKI, Y. KISHI, M. FURUTA, T. SANKE, K. NANJO and M. MUKOYAMA, Diabetes.Res. Clin. Pratt. 30 153-162 (1995). 19. M. KIHARA, J.D. SCHMELZER and P.A. LOW, Diabetologia 38 914-918 (1995). 20. N. HOTTA, K. KOH, F. SAKAKIBARA, J. NAKAMURA, Y. HAMADA,T. HARA, K.

1422

21. 22. 23. 24. 25. 26. 27.

Cilostazol Increases NO in SH-SYSY Cells

Vol. 65, No. 13, 1999

MORI, K. NARUSE, H. FUKASAWA, H. KAKUTA, and N. SAKAMOTO, Eur. J. Pharmacol. 313 201-209 (1996). K. UEHARA, K. SUGIMOTO, R. WADA,T. YOSHIKAWA,K. MARUKAWA,Y. YASUDA, Y. KIMURA and S. YAGIHASHI, J. Diabetes Complicat. 11 194-202 (1997). H. INADA, H. SHINDO, M. TAWATA and T. ONAYA, Diabetologia41 1451-1458 (1998). A. KARIHALOO, K. KATO, D.A. GREENEand T.P. THOMAS, Am. J. Physiol. 273 C671-C678 (1997). N.A. DELAMERE, R.R. SOCCI, and K.L. KING, Invest. Ophthalmol. Vis. Sci. 31 21642170 (1990). S. GUPTA, I. SUSSMAN, C. MCARTHUR, K. TORNHEIM, R.A. KOHEN and N.B. RUDERMENN, J. Clin. Invest. 90 727-732 (1992). C. SCAVONE, C. SCANLON, M. MCKEEand J.A. NATHANSON, J.Pharmacol. Exp. Ther. 272 1036-1043 (1995). H. AKIYAMA, S. KUDO, M. ODOMI and T. SHIMIZU, J. Chromgr. Biomed. Appl. 338 456-459 ( 1985).