Allergology International (1998) 47: 219-224

Original Article Effects of phosphodiesterase inhibitors on secretions of human monokines

Yuji Sudn,' Gen Tcrnurc;' lsco Ohno;' Kimito Maeda,l Yi l.iu;' Kohei Yamauchi 2 Fumihiko Kurimoto? and Kunio Shirato 1 '

1First Department of Internal Medicine, Tohoku University School of Medicine, Sendai, 2Third Department of Internal Medicine, Iwate Medical University School of Medicine, Morioka, 3Mitsubishi Kagaku Bio Clini­ cal Laboratory, Tokyo, Japan

ABSTRACT Key words: cilostazol, interleukin-JB, KF19514, mono­ cytes, phosphodiesterase inhibitor, theophylline, tumor The purpose of this study was to evaluate the effect of necrosis factor a. newly developed selective phosphodiesterase (PDE) inhibitors, KF19514 (type lilY) and cilostazol (type III), and theophylline on the secretions of tumor necrosis INTRODUCTION factor a (TNFa) and interleukin- 1~ (IL-1~) from human ~ (IL-1~) peripheral monocytes stimulated by lipopolysaccha­ Tumor necrosis factor a (TNFa) and interleukin-1 l- 3 ride (LPS). Human blood monocytes were incubated have been found in the broncho-alveolar lavage fluid with LPS in the absence or presence of KF19514, and in bronchial biopsy specimens" obtained from cilostazol or theophylline. TNFa and IL-1 ~ in the cell­ patients with bronchial asthma. These cytokines have a free supernatants were measured with enzyme-linked role in airway inflammation through leukocyte recruit­ immunosorbent assay. KF19514 showed significant ment into the airway by inducing leukocyte adhesion s,6 inhibition on the release of TNFa (% inhibition ± SEM molecules on endothelial cells and airway epithelial was 82.8 ± 7.4% at 1 urnol/L) and IL-1~ (34.4 ± 7.5% cells." Peripheral blood monocytes secrete these cyto­ at 10 umol/L). In addition, KF19514 inhibited the kines by both irnrnunoloqlcol'' and non-immunological stimulation, such as that by lipopolysaccharide expression of TNFa mRNA. Cilostazol inhibited the (LPS).9,lO release of TNFa significantly (60.2 ± 8.9% at 30 urnol/l.) Thus, an in vitro system that employs the secretions of ~ but not IL- 1~. Theophylline inhibited slightly but TNFa and IL-1 from LPS-stimulated blood monocytes significantly the release of TNFa at a therapeutic con­ could be used to evaluate the anti-inflammatory proper­ centration (17.4 ± 5.1 % at 100 urnol/L]. These resuIts ties of anti-asthmatic drugs. suggest that theophylline may not only have a broncho­ Phosphodiesterase (POE) inhibitors modulate mono­ 11 14 dilating action but also an anti-inflammatory prop­ kine secretion via the elevation of cyclic nucleotides. - erty in the treatment of bronchial asthma, and that KF 19514 isa newly synthesized type IIIV dual POE inhibitor. KF19514 may have an anti-inflammatory action on at It is a derivative of KF 17625 and has a more potent bron­ 1S least the transcriptional level. chodilator action than KF17625. ICso values for POE I, POE II, POE III, POE 1'1, and POE V in vitro have been shown to be 0.27, > 10, > 10, 0040 and > 10 umol/L, respec­ tively.16 Cilostazol, a derivative of cilostornlde.V:"? is a type Correspondence: Kunio Shirato, MD, First Department of III selective POE lnhibitor./? This drug is clinically used in Internal Medicine, Tohoku University School of Medicine, Sendai, 980-8574, Japan. Japan for the treatment of obstructive arteriosclerosis. Received 18 June 1997. Accepted for publication 30 April In this study, we examined the effect of these selective 1998. POE isoenzyme inhibitors on the secretion of TNFa and 220 Y SUDA ET AL.

IL-1~ from human peripheral blood monocytes (BMo) Blood monocytes stimulation stimulated by LPS. In addition, we evaluated the effect of A total of 400 000 BMo were incubated in the absence theophylline. and presence of 1 Ilg/mL LPS with or without KF19514, cilostazol or theophylline at the following concentrations: METHODS KF19514 at 1, 10 and 100 urnol/L: cilostazol at 1, 10 and 30 urnol/L: and theophylline at 10, 100 and 1000 Reagents and culture materials umol/L. After 18 h incubation (3JOC, 5% CO2), the super­ Complete medium (MEM) consisted of alpha-modified natant was harvested after centrifugation and frozen at Eagle's medium + 1% fetal calf serum + penicillin (100 -20°C until the assay was undertaken. units/mL) + streptomycin (100 1l9/mLi all from Gibco, In order to verify that KF19514, cilostazol or theophyl­ Paisley, UK). KF19514, 5-phenyl-3-(3-pyridyl)methyl-3H­ line did not influence the cell viability during the incuba­ imidazo[4,5-C][l,8] naphthyridin-4(5H)-one was donated tion, colorimetric ossov?' using tetrazolium compound, by Kyowa Hakko Kogyo, Shizuoka, Japan. Cilostazol, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bro­ 6-(4-[1-cyclohexyl-1 H-tetrazol-5-yl] butoxy)-3,4-dihydro­ mide (Sigma), was performed. There was no difference 2(1 H)-quinolinone was donated by Otsuka Pharmaceutical in cell viability among these conditions. company Ltd, Osaka, Japan. Theophylline was obtained from Sigma (St Louis, MO, USA). KF19514 and theo­ Measurement of TNFa and IL-l ~ phylline were dissolved in phosphate buffered saline (PBSi Gibco), whereas cilostazol was dissolved in N,N­ TNFa and IL-1~ were assayed with an enzyme linked Dimethylformamide (Gibco) and further diluted by immunosorbent assay using human TNFa and IL-1 ~ kits adding MEM to the stock solution. (Otsuka Pharmaceutical Company, Osaka, Japan). The assay procedure is briefly described as follows: the monoclonal antibody specific for each cytokine is Isolation of blood monocytes coated onto the microtiter plate provided in the kit. Blood monocytes were purified by cdherence.? In brief, Standards with known amounts of the cytokine and blood was drawn from normal volunteers and mono­ samples are pipetted into the wells and left at 4°C for nuclear cells (MNC) were obtained by Ficoll-Hypaque 24 h. After washing, an enzyme-linked polyclonal anti­ centrifugation (LSM, Organon Teknika Corp, NC, USA). body specific to the cytokine is added to the wells and The concentration of BMo was adjusted to 4 X 10s/mL left at room temperature for 2 h. Following a wash to using non-specific esterase staining (Sigma) for MNC. remove any unbound antibody-enzyme reagent, a sub­ Following this, 1 mL of cell suspension was poured into strate solution is added to the wells and the intensity of culture wells and incubated (3JOC, 5% CO2) for 1 h in the color is measured by densitometry. A curve is pre­ order for BMo to adhere to the bottom of the well. After pared as a standard and, by comparing the optical the incubation, non-adherent cells were removed by density of the samples to this curve, the concentration of washing three times with PBS to purify the BMo. Trypan the cytokine in the sample is determined. blue dye exclusion showed 100% viability of the cells before incubation. Northern analysis

In order to evaluate whether KF19514 inhibits the Time course of secretion of TNFa and IL-1~ expression of TNFa mRNA, Northern analysis 22 was by lipopolysaccharide stimulation performed. Ten million BMo were cultured in large For the time course study, 400 000 BMo, purified as wells under the following conditions: MEM only, MEM + described above, were incubated in the absence and LPS (1 uq/rnl), and MEM + LPS (1 uq/rnl) + KF19514 presence of 1 uq/rnl, LPS from Escherichia coli (serotype (100 urnol/L). After 2 h incubation, the cells in each well 026:B6, Sigma) for 0, 2, 6, 12 and 18 h in two subjects. were denatured with guanidine isothiocyanate and RNA The supernatant was harvested after centrifugation was purified by phenol-chloroforrn.P After electropho­ (6009, 10 min) at each incubation time and frozen at resis, the RNA was transferred to nitrocellulose filters. -20°C until the assay was undertaken. The filters were hybridized with human TNF eDNA probe SUPPRESSION OF MONOKINE BY POE INHIBITOR 221

(820 bp EcoRI fragment of TNFa cloned in a pUC 2355 pg/mL for IL-1 ~ in subject 1, and 189 pg/mL and vector}" labeled with [32P]dCTP (Amersham, UK) by nick 598.5 pg/mL in subject 2. translation and the hybridized signals were detected by autoradiography. Effects of KF19514, cilostazol and theophylline on the secretion of TNFa Statistical analysis from LPS-stimulated BMo

The paired two-tailed Student's t-test was performed for As shown in Fig. 2, KF19514, cilostazol and theophyl­ statistical analysis. Differences were considered significant line dose-dependently inhibited the secretion of TNFa for P < 0.05. Values are indicated as mean ± SEM. from LPS-stimulated BMo. The mean of the actual amount of TNFa secreted after 18 h-stimulation by LPS was 693 ± 84 pg/mL. ICso values of each agent were as RESULTS follows: KF19514, < 1 umol/L: cilostazol, 28/-lmol/L; Time course of secretion of TNFa and IL-l ~ theophylline, > 100 umol/L. Thus, KF19514 had the from LPS-stimulated BMo most potent inhibitory action. Although theophylline showed the least inhibitory effect on a molar basis, a The secretions of TNFa and IL-1~ from LPS-stimulated significant inhibition was observed at 100 umol/L of BMo were evaluated after 0, 2, 6, 12 and 18 h of incu­ theophylline, which is in the upper range of the thera­ bation. Neither cytokine was detected at any period peutic concentration (18 /-lg/mL). without LPS-stimulation. As shown in Fig. 1, the patterns ~ of secretions for TNFa and IL-1 were different from Effects of KF19514, cilostazol and each other. Thus, 40% of the maximal secretion ob­ theophylline on the secretion of IL-1~ tained at 18 h culture was noted at 2 h for TNFa, from LPS-stimulated BMo whereas the secretion of IL-1~ was not detected at 2 h incubation. However, the maximum level of secretion for As shown in Fig. 3, KF19514 also inhibited the secretion both cytokines was observed by 12 h culture and of IL-1~ from BMo dose-dependently. The mean of the remained at a plateau for up to 18 h. The means of the actual amount of the cytokines secreted after 18 h 120 stimulation by LPS were 524.5 pg/mL for TNFa and ~ 100 .\~~ \ ". * 120 E] " . ~ 8 80 . E ~ 100 Z Q) " c- ~~ o 0 60 " ·2 ~ 80 c ...c: ~~ o u " u .... :.:: g " ~ o 60 ~ >- 40 " ~2 Q)- V') 8- " ~ ~ 40 o "" ** c.. 20 >..~ 1 u .... 20 ...... ~ ...... -...... ** ...... ~ -._j-.....

2 6 12 18 1 10 30 100 1000 Incubation time (h) Drug concentration (fJmol/L)

Fig. 1 Time course of secretion of TNFa (0) and IL-l ~ (.) Fig. 2 The effects of theophylline (0; n = 11), cilostazol from human blood monocytes stimulated by lipopolysaccharide (0; n = 6) and KF19514 (~; n = 6) on the secretion of TNFa (LPS). The cytokines in the supernatant obtained after 0,2,6, 12 from human blood monocytes stimulated by lipopolysaccharide and 18 h culture were assayed. Each plot is the mean of %secre­ (LPS). Results are expressed as mean ± SEM of % sec­ tion of the amount secreted after 18 h culture. Bars show the retion of TNFa to positive control stimulation. *P < 0.01 and ranges of two subjects. **P < 0.001 denote significant difference from the control level. 222 Y SUDA ET Al. actua l amount of Il - l psecreted after 18 h stimulation by A lPSwas 13 72 ± 191 pg/ml . The ICso value ofKF1 951 4 was 55 umol/ L. However, neither cilostazol nor theo ­ phylline showed statistically sign ificant inhibition on the secretion of Il - l Pfrom SMa.

Effect of KF19514 on the expression of the TNFa gene by LPS-stimuloted BMo

In order to eva luate whet her KF195 14 inh ib its the expression of TNFa mRNA, we perfo rmed No rthern ana lysis in two sub jects. RNA was obtained fro m 1 2 3 4 10 million SMa in each of the fo llowing cu ltu re con ­ dition s: MEM a lone, M EM con ta ining lP S, and M EM con ta ining lPS and KF195 14 (1 00 u rnol / L}. Ten micro ­ B grams of RNA we re obta ined in each condition. The aut oradiogra m fo r TNF a mRNA in the la ne obtained fro m cells cultured with lPS an d KF195 14 285- was wea ker than that ob ta ined fro m cel ls cultu red with LPS. No ge ne express io n was no ted in t he lane without l PS-stimulation. This result indicated that KF19514 part ial ly inhibited the gene transcription for TNFu. in the SMa stimu lated by lPS (Fig. 4) .

21 3 4

120 Fig. 4 No rthern ana lysis for TNFa RNA extracted fro m mo no­ cyte culture with medium a lone (lone 1), with LP5 (lone 2), ~ 1 0 0 with l PS and KF1951 4 (lane 3) and rRNA from calf liver (lane 4), were hybr idized with hTN FcDNA probe . An equal a mount of o , " RNA in each lo ne was con fir med with the assessment of 285 ~ ~ 80 and 185 rRNA stained with ethid ium bro mide after electro­ =I •U ...... -1 * -, pho resis (Pa nel B). Solid ar rows indicate the pos itions of 185 -0 "";: 60 " o 0 an d 285 rRNA. Cl osed ar row head indicates the position of " .2 -5 " TNFamRNA. - U ~ 0 " ii:l1-- 40 '1 "' 0Q. o Q. 20 DISCUSSION

In the present study, we demonstrated that KF195 14 signi­ ficantly inhibited the secretions of both TNFa and Il-l P 10 30 100 1000 from SMa stimulated by l PS, and that cilostazol and theo­ Drug concentra tion {pmo l/ L] phylline significantly inhibited the secretion of TNFu., but not Il - l p, in the same system. In add ition, we showed Fig. 3 The effects of theophylline (.; n ~ 11), cilostozol Ie; that 100 umol/ L of KF 195 14 decreased the TNFu. mRNA n ~ 6) and KF19514 I. ; n ~ 6)) on the secretion of Il - I Pfrom signal. huma n blood mo nocytes stimula ted by lipopolysaccharide (LPS). Results ar e expressed as mea n ± SEM of % sec retion of The rank order of the inhibito ny effect on the secretion IL- l ~ to positive control stimulation. *P < 0.0 1 and **P < 0 .001 of TNFu. was KF19514 > cilostazol > theophylline. denote sign ifica nt difference from the co ntro l leveL KF195 14 showed an inhibito ny effect on the release of SUPPRESSION OF MONOKINE BY POE INHIBITOR 223

TNFa similar to that of type IV POE isoenzyme inhibitors, present study, they evaluated the inhibitory effect on the rolipram,2s-28 nltroquozone." Ro-20-1724,27,28 and CP­ secretion after 2 h stimulation with LPS. Given that, as 77059.27 When compared with zordoverine.F? a type shown in the time-course study, the amount of TNFa after III/IV dual POE inhibitor, the inhibitory effect of KF19514 2 h stimulation with LPS is only 40% of the maximal secre­ was quite similar. By contrast, cilostazol showed a weaker tion which was observed after 12 h stimulation and lasted inhibitory effect than did KF19514, which was equal to up to 18 h, the greater rate of inhibition in their study may that of other type III POE isoenzyme inhibitors, namely have resulted from the shorter period of stimulation by LPS. cilostornide.i" milrinori." CI-930,26 rnotcpizone" and In conclusion, we showed that 100 umol/L of theo­ SK&F94836.28 Thus, it is suggested that in BMo a type IV phylline had a significant inhibition on the secretion of POE isoenzyme may have a major role in the regulation TNFa from purified BMo stimulated by LPS. Therefore, we of the secretion of TNFa, and that the type IV POE suggest that theophylline may have not only a broncho­ inhibitory action of KF19514 may be responsible for the dilating action but also an anti-inflammatory property inhibitory effect on the secretion of TNFa. In addition, when used in the treatment of bronchial asthma. In addi­ strong inhibition on the secretion of TNFa by KF19514 is tion, we demonstrated that KF19514 showed a potent at least partially due to the inhibition on the expression of inhibition on the secretion of TNFa and a moderate inhi­ TNFa mRNA, which is similar to the effect of rcliprorri." bition on the secretion of IL-1~ from purified BMo stimu­ KF19514 inhibits both POE I and POE I\/, and 70% lated by LPS, and that 100 urnol/L of KF19514 clearly inhibition on the secretion of IL-1~ was observed at 100 decreased the TNFa mRNA signal. Therefore, KF19514 urnol/L, which is 350 times the ICso for POE I. It has been may have an anti-inflammatory property, although its reported that the secretion of IL-1~ from LPS-stimulated clinical efficacy remains to be studied. MNC is downregulated by cGMp'14 Therefore, one pos­ sible mechanism is that complete inhibition of POE I by 100 umol/l, of KF19514 may downregulate the secretion ACKNOWLEDGEMENTS of IL-1~ via a possible increase in cGMp, although BMo We thank Kyowa Hakko Kogyo Co. Ltd, Shizuoka, Japan 30 contains POE I at a ratio of 1: 10 of POE IV However, we and Otsuka Pharmaceutical Company Ltd, Osaka, can not exclude the possibility that complete inhibition of Japan for their generous supply of KF19514 and cilo­ POE IV by KF19514 may inhibit the secretion of IL-1~ stazol, respectively, and Brent Bell for reviewing our because rolipram, a cAMP-elevating but not a cGMP­ manuscript. elevating agent, has been reported to inhibit the secretion of IL-1~.27 Cilostazol has been generally prescribed for the treat­ REFERENCES ment of obstructive arteriosclerosis in Japan. A recent Mattoli S, Mattoso VL, Soloperto M, Allegra L, Fasoli A. Cel­ study has shown that cilostazol has bronchodilator and lular and biochemical characteristics of bronchoalveolar bronchoprotective effects in normal subjects at a serum lavage fluid in symptomatic nonallergic asthma. J. Allergy C1in. Immunol. 1991; 87: 794-802. concentration of 6.7 urnol/L'" However, all subjects com­ 2 Broide DH, Lotz M, Cuomo AJ, Coburn DA, Federman EC, plained of mild to severe headaches at that concentration. Wasserman SI. Cytokines in symptomatic asthma airways. Although, as shown in this study's results, 10 urnol/L of J. Allergy C1in. Immunol. 1992; 89: 958-67. cilostazol demonstrated a significant inhibition on the 3 Marini M, Avoni E, Hollemborg J, Mattoli S. Cytokine mRNA secretion of TNFa from LPS-stimulated BMo, TNFa inhi­ profile and cell activation in bronchoalveolar lavage fluid bition by cilostazol is not clinically feasible. from nonatopic patients with symptomatic asthma. Chest 1992; 102: 661-9. Theophylline inhibited the secretion of TNFa dose­ 4 Ackerman V, Marini M, Vittori E, Bellini A, Vassali G, Mattoli dependently and a significant inhibition was observed at S. Detection of cytokines and their cell sources in bronchial 100 umol/Lwhich can be converted into a serum concen­ biopsy specimens from asthmatic patients. Relationship to tration of 18 ~g/mL. Thus, our findings suggested that we atopic status, symptoms, and level of airway hyperrespon­ may be clinically using the inhibitory action of theophylline siveness. Chest 1994; 105: 687-96. 5 Ebisawa M, Bochner BS, Georas SN, Schleimer RP. in the treatment of bronchial asthma. Recently, Spatafora Eosinophil transendothelial migration induced by cytokines. et 0/. reported that 100 umol/Loftheophylline inhibited the I. Role of endothelial and eosinophil adhesion molecules in secretion of TNFa from BMo stimulated by LPS.32 Although IL-1 beta-induced transendothelial migration. J. Immunol. their inhibitory rate was much greater than that in the 1992; 149: 4021-8. 224 Y SUDA ET AL.

6 Howarth PH, Bradding P, Quint D, Redington AE, Holgate 19 Shintani S, Toba Y, Suzuki S, Ninomiya S, Umezato M, ST. Cytokines and airway inflammation. Ann. N. Y. Acad. Hiyama T. General pharmacological properties of cilostazol, Sci. 1994; 725: 69-82. a new antithrombotic drug. Part I. Effects on the central ner­ 7 Tosi MF, Stark JM, Smith CW, Hamedani A, Gruenert DC, vous system. Arzneimittelforschung. 1985; 35: 1157-62. Infeld MD. Induction of ICAM-1 expression on human air­ 20 Okuda Y, Kimura Y, Yamashita K. Cilostazol. Cardiovas. way epithelial cells by inflammatory cytokines: effects on Drug Rev. 1993; 11: 451-65. neutrophil-epithelial cell adhesion. Am. 1. Respir. Cel/ Mol. 21 Mosmann T. Rapid colorimetric assay for cellular growth BioI. 1992; 7: 214-21. and survival: application to proliferation and cytotoxicity 8 Borish L, Mascali JJ, Rosenwasser U. IgE-dependent cytokine assays.1. Immunol. Meth. 1983; 65: 55-63. production by human peripheral blood mononuclear phago­ 22 Ohno I, Tanno Y, Yamauchi K, Takishima T. Gene expres­ cytes.1. Immunol. 1991; 146: 63-7. sion and production of tumour necrosis factor by a rat 9 Martinet Y, Yamauchi K, Crystal RG. Differential expression basophilic leukaemia cell line (RBL-2H3): with IgE receptor of the tumor necrosis factor/cachectin gene by blood and triggering. Immunology 1990; 70: 88-93. lung mononuclear phagocytes. Am. Rev. Respir. Dis. 1988; 23 Chomczynski P, Sacchi N. Single-step method of RNA iso­ 138: 659-65. lation by acid guanidinium thiocyanate-phenol-chloroform 10 Bernaudin JF, Yamauchi K, Wewers MD, Tocci MJ, Ferrans extraction. Anal. Biochem. 1987; 162: 156-9. VJ, Crystal RG. Demonstration by in situ hybridization of 24 Pennica D, Nedwin GE, Hayflick JS et 01. Human tumour dissimilar IL-1 beta gene expression in human alveolar necrosis factor: precursor structure, expression and homol­ macrophages and blood monocytes in response to ogyto Iymphotoxin. Nature 1984; 312: 724-9. lipopolysaccharide. 1. Immunol. 1988; 140: 3822-9. 25 Molnar-Kimber K, Yonno L, Heaslip R, Weichman B. 11 Strieter RM, Remick DG, Ward PAet 0/. Cellular and mole­ Modulation of TNF alpha and IL-1 beta from endotoxin­ cular regulation of tumor necrosis factor-alpha production stimulated monocytes by selective PDE isozyme inhibitors. by pentoxifylline. Biochem. Biophys. Res. Commun. 1988; Agents Actions 1993; 39, Spec No: C77-9. 155: 1230-6. 26 Semmler J, Wachtel H, Endres S. The specific type IV 12 TaffetSM, Singhel KJ, Overholtzer JF, Shurtleff SA. Regulation phosphodiesterase inhibitor rolipram suppresses tumor of tumor necrosis factor expression in a macrophage-like necrosis factor-alpha production by human mononuclear cell line by lipopolysaccharide and cyclic AMP. Cel/lmmunol. cells. Int. 1. Immunopharmacol. 1993; 15: 409-13. 1989; 120:291-300. 27 Prabhakar U, LipshutzD, BartusJO et 01. Characterization of 13 Hurme M. Modulation of interleukin- 1 beta production cAMP-dependent inhibition of LPS-induced TNF alpha pro­ by cyclic AMP in human monocytes. Febs Lett. 1990; duction by rolipram, a specific phosphodiesterase IV (PDE 263: 35-7. IV) inhibitor. Int. 1. Immunopharmacol. 1994; 16: 805-16. 14 EndresS, Fulle HJ, Sinha Bet 0/. Cyclic nucleotides differen­ 28 Verghese MW, McConnell RT, Strickland AB et 01. tially regulate the synthesis of tumour necrosis factor­ Differential regulation of human monocyte-derived TNF alpha and interleukin- 1 beta by human mononuclear cells. alpha and IL-1 beta by type IV cAMP-phosphodiesterase Immunology 1991; 72: 56-60. (cAMP-PDE) inhibitors. 1. Pharmacol. Exp. Ther. 1995; 15 Suzuki F, Kuroda T, Kawakita T et 01. New bronchodilators. 272: 1313-20. 3. Imidazo[4, 5-c][l, 8]naphthyridin-4 (5H) -ones. 1. Med. 29 Schade FU, Schudt C. The specific type III and IV phospho­ Chem. 1992; 35: 4866-74. diesterase inhibitor zardaverine suppresses formation of 16 Manabe H, Akuta K, Sejimo H et 01. Anti-inflammatory tumor necrosis factor by macrophages. Eur. 1. Pharmacol. and bronchodilator properties of KF19514, a phospho­ 1993; 230: 9-14. diesterase 4 and 1 inhibitor. Eur J. Pharmacol. 1997; 30 Schudt C, Tenor H, Hatzelmann A. PDE isoenzymes as targets 332: 97-107. for anti-asthma drugs. Eur. Respir. 1. 1995; 8: 1179-83. 17 Hidaka H, Hayashi H, Kohri H et 01. Selective inhibitor 31 Fujimura M, Kamio Y, Saito M, Hashimoto T, Matsuda T. of platelet cyclic adenosine monophosphate phospho­ Bronchodilator and bronchoprotective effects of cilostazol diesterase, cilostamide, inhibits platelet aggregation. in humans in vivo. Am. 1. Respir. Crit. Care Med. 1995; J. Pharmacol. Exp. Ther. 1979; 211: 26-30. 151: 222-5. 18 Umekawa H, Tanaka T, Kimura Y, Hidaka H. Purification of 32 Spatafora M, Chiappara G, Merendino AM, D'Amico D, cyclic adenosine monophosphate phosphodiesterase from Bellia V, Bonsignore G. Theophylline suppressesthe release human platelets using new-inhibitor Sepharose chromato­ of tumour necrosis factor-alpha by blood monocytes and graphy. Biochem. Pharmacol. 1984; 33: 3339-44. alveolar macrophages. Eur. Respir. 1. 1994; 7: 223-8.