Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Xanthines such as theophylline and aminophylline are widely used in the treatment of asthma. Although xanthines operate through multiple mechanisms in the airways and lungs, the most likely mechanism of the bronchodilating effect has been thought to be the accumulation of cAMP in bronchial smooth muscle by inhibition of cyclic nucleotide PDE. Furthermore, because eosinophilic inflammation plays a crucial role in the pathogenesis of asthma, much attention has been paid to the anti-inflammatory and immunomodulatory effects of xanthines (Howell, 1990; Ward et al., 1993; Kidney et al., 1995). However, the complex multiple actions of xanthines include side effects on the cardiovascular system, GI system and/or CNS, and these side effects limit their use.

PDEs comprise at least seven isozyme families, PDE1 through PDE7. Individual PDE isozymes are encoded by a distinct cDNA sequence and differ with respect to their tissue distributions, physical and kinetic activities and substrate preferences (cAMP and/or cGMP) (Beavo, 1995). The inhibitions of PDE activities by xanthines are weak and are not isozyme-selective. Thus, when one isozyme of PDE is inhibited selectively and strongly, it may result in more useful effects and fewer undesirable effects in the treatment of asthma.

The inhibition of PDE3 or PDE4, both of which hydrolyze cAMP, is of special interest because either inhibition induces bronchodilation by the accumulation of cAMP (Nicholson et al., 1991; Hall, 1993). Although some investigators reported that PDE3 inhibition relaxed airway smooth muscle more than PDE4 inhibition did (Rabe et al., 1993; Torphy et al., 1993), PDE3 inhibition causes a positive inotropic action, and it may result in arrhythmogenic potency of xanthines (Nicholson et al., 1991; Hall, 1993). In contrast, PDE4 inhibition has only mild effects on the cardiovascular system. In addition, PDE4 inhibition induces anti-inflammatory effects such as the reduction of chemical mediator release from mast cells, basophils, eosinophils, neutrophils and macrophages; it also inhibits eosinophil infiltration and airway vascular leakage in the airways in guinea pigs (Dent et al., 1991; Souness et al., 1991; Ortiz et al., 1992; Raeburn and Karlsson, 1993; Lagente et al., 1994; Souness et al., 1994; Underwood et al., 1994). These facts suggest that the selective inhibition of PDE4 may be useful in treating asthma. There are, however, only a few studies in the human on the relaxing effect of the selective inhibition of PDE4 on precontracted bronchial smooth muscle (Cortijo et al., 1993; Rabe et al., 1993; Torphy et al., 1993), and there is no study on the preventive effect of the selective inhibition of PDE4 on bronchial smooth muscle contraction induced by histamine, ACh and allergen.

T-440, 6,7-diethoxy-2,3-bis(hydroxymethyl)-1-[1-(2-methoxyethyl)-2-oxo-pyrid-4-yl]naphthalene, is a novel PDE4 inhibitor (fig. 1; table 1 and Iwasaki et al., 1996). T-440 inhibits antigen- and chemical mediator-induced bronchoconstrictions in guinea pigs (Kaminuma et al., 1996). However, the effect on human bronchial tissues is not yet known. The goal of this study is to explore the reversal and preventive effects of the selective inhibition of PDE4 by T-440 on the human bronchial contraction induced by histamine, ACh and allergen. We also studied the effect of T-440 on cAMP accumulation in human bronchial tissue when the tissue was contracted by histamine, comparing it with the effect of aminophylline, one of the clinically used isozyme-nonselective PDE inhibitors.

View larger version (13K): [in this window] [in a new window]   Fig. 1.   Chemical structure of T-440.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Human bronchial tissues were obtained from 41 patients (32 males, 9 females) undergoing surgery for lung cancer. They had no history of atopic diseases and bronchial asthma, and their pulmonary function was within normal limits (FEV_1.0% > 70%, %VC > 80%). They were 31 to 82 years old (mean age 64.0 ± 1.8). Immediately after resection of a lung or a lobe, the tissues were placed in Krebs-Henseleit solution of the following millimolar composition: NaCl, 118.3; KCl, 4.7; MgSO₄, 1.2; KH₂PO₄, 1.2; CaCl₂, 2.5; NaHCO₃, 25.0; glucose, 11.1. Segmental and subsegmental bronchi were dissected free from apparently normal lung tissue and cut into ring preparations. The rings were cut open through the cartilage. All tissues were prepared within 2 h after the resection.

Effect of T-440 on Bronchial Contraction

The tissues were placed in 10-ml organ baths containing Krebs-Henseleit solution. The solution was maintained at 37°C and continuously aerated with a mixture of 95% oxygen and 5% carbon dioxide. A glass hook at the bottom of the organ bath held one end of the tissue, and the other end was fixed by a silk thread to an isometric force displacement transducer (TB-611T; Nihonkohden Kogyo Co., Tokyo, Japan). The bronchial tissues were set up for the tension measurement under a 1-g load and equilibrated for more than 60 min in the solution, during which time the solution was replaced every 15 min. The responses were amplified (AP-621G; Nihonkohden Kogyo Co., Tokyo, Japan) and recorded (RJG-4128; Nihonkohden Kogyo Co.) (Honda et al., 1991).

For the studies of contraction induced by allergen, tissues were passively sensitized as previously described (Yamaguchi et al., 1992). Briefly, before being placed in 10-ml organ baths, the bronchial tissues were incubated for 1 h at 25°C with the serum from asthmatic patients with a highly positive specific IgE against house dust, Dermatophagoides pteronyssius (DP), and Dermatophagoides farinae (DF), which are major constituents of house dust antigen.

First, 10³ M ACh was added to the tissues to test the viability of the bronchial smooth muscles. Next, the tissues were washed with Krebs-Henseleit solution until the tension returned to the baseline, and then the tissues were set up under a 1-g load again.

Reversal effects of T-440 on bronchial contraction induced by histamine, ACh and allergen. To study the reversal effects of T-440 on bronchial contraction induced by histamine and ACh, we prepared 3 to 6 bronchial tissues from each of 11 patients and six tissues from each of four patients, forming two groups. We added 10⁵ M histamine to one group and 10⁴ M ACh to the other. Confirming the plateau of the contraction, we added T-440 (10⁷ to 10⁵ M), aminophylline (10⁵ or 3.3 × 10⁵ M) or their solvent to the tissue. We defined this point as 100% contraction and observed the response for 100 min.

6

5

5

5

Preventive effects of T-440 on bronchial contraction induced by histamine, ACh and allergen. To study the preventive effects of T-440 on bronchial contraction induced by histamine and ACh, we prepared 5 to 6 tissues from each of 8 patients and 6 tissues from each of 4 patients, forming two groups. We incubated the tissues with T-440 (10⁷ to 10⁵ M), aminophylline (10⁵ or 3.3 × 10⁵ M) or their solvent for 30 min. Then we added histamine (10¹⁰ to 10³ M) to one group and ACh (10⁹ to 3.3 × 10³ M) to the other cumulatively to construct a concentration-response curve.

6

5

5

5

Extraction, Separation and Characterization of PDE Isozymes from Human Bronchus

To confirm that T-440 inhibits the PDE4 activity of human bronchial tissues, we evaluated its effect on isolated PDE4 activity.

Human bronchial tissue (2.1 g) was obtained as described above and stored at 80°C until use. The tissue was homogenized (Polytron Model PT3000, Kinematica GmbH, Luzern, Switzerland) at 15,000 rpm for 3 × 20 sec in three volumes of ice-cold homogenization buffer (50 mM Tris-HCl [pH 7.5], 0.25 M sucrose, 5 mM benzamidine, 0.2 mM phenylmethylsulfonyl fluoride [PMSF], 2 µM leupeptin, 0.1 mM EGTA, 5 mM 2-mercaptoethanol). The extract was centrifuged to obtain the supernatant (cytosolic fraction) at 25,000 × g for 30 min at 4°C. The cytosolic fraction was filtered (0.45 µm) and applied on a Mono-Q HR5/5 column (Pharmacia, Uppsala, Sweden) pre-equilibrated in elution buffer (50 mM Tris-HCl [pH 7.5], 2 mM benzamidine, 0.2 mM PMSF, 0.1 mM EGTA, 5 mM 2-mercaptoethanol), attached to a fast protein liquid chromatography (FPLC) system (Pharmacia, Uppsala, Sweden). After the column was washed with elution buffer, bound proteins were eluted with a linear gradient of 0 to 1 M NaCl in 30 ml of elution buffer at a flow rate of 0.5 ml/min, and 0.5 ml fractions were collected. Each fraction was assayed for PDE activity, and PDEs in each fraction were characterized by specificity for a substrate and by influences of cGMP and rolipram (a specific inhibitor of PDE4). The PDE4 fractions were collected and used for inhibition assays.

PDE activity was measured by a modification of the two-step radioisotope method (Thompson et al., 1979). The reaction mixture contained 50 mM Tris-HCl (pH 8.0), 5 mM MgCl₂, and 4 mM 2-mercaptoethanol. All assays were performed at 30°C for 30 min, and the concentration of substrate was 1 µM for [³H]-cAMP or [³H]-cGMP. Inhibition assays for T-440, rolipram and theophylline were performed in triplicate. For each drug, 3 to 5 different concentrations were tested. IC₅₀ values were obtained by fitting concentration-response curves to the four-parameter logistic equation.

Measurement of cAMP Content

We prepared 40 mg of human bronchial tissues (diameter 1 to 2 mm), denuding the epithelium by gentle rubbing. To examine whether T-440 causes the accumulation of cAMP in the tissue, we measured cAMP content in the presence and absence of histamine. In this study, we used smaller bronchi because they are easily homogenized. The tissues were incubated free-floating in Krebs-Henseleit solution aerated with a mixture of 95% oxygen and 5% carbon dioxide at 37°C. The tissues were equilibrated for more than 60 min in the solution, during which time the solution was replaced every 15 min. At 20 min after the addition of histamine (10⁵ M) or its solvent, when the histamine-induced bronchial contraction reached a plateau in the study described above, we added T-440 (10⁶ or 10⁵ M), aminophylline (3.3 × 10⁵ M) or their solvent to the tissues. They were then incubated for 60 min, because the reversal effect of the drugs reached plateau at around 60 min. Next they were rapidly removed, blotted on dry absorbent paper and snap-frozen by submersion in liquid nitrogen. The frozen tissues were stored at 80°C until assay for cAMP content.

We added 1 ml of cold 10% trichloroacetic acid (0°C) to a cold glass homogenizing tube containing approximately 40 mg of frozen tissue. The tissue was homogenized for 6 × 10-sec bursts at setting 6 (Polytron PCU-2, Kinematica GmbH, Luzern, Switzerland) and transferred to a 1.5-ml microtube. The homogenate was centrifuged at 600 × g for 15 min at 4°C (MR-150, Tomy, Tokyo, Japan) to form soluble extract and particulate fractions. Trichloroacetic acid in soluble fraction was removed by four successive extractions with water-saturated ether. After being heated at 70°C for 5 min to remove ether, the soluble fraction was stored at 20°C until assay for cAMP content. The residual precipitation was used for the measurement of protein content. cAMP was acetylated, and its amount was subsequently estimated by enzyme immunoassay kits (Cayman Chemical Co., Ann Arbor, MI). cAMP content was expressed as picomoles of cAMP per milligram of protein. Protein content was estimated by the method of Lowry et al. (Lowry et al., 1951).

Data Analysis and Statistics

Data were expressed as the mean ± S.E. In the study of reversal effect, contraction was expressed as the percent of contraction induced by each spasmogen.

In the study of preventive effect, the contractile response was expressed as follows: The EC₅₀ value of the histamine- and ACh-induced concentration-response curve was calculated by linear regression analysis and expressed as a negative logarithm.

Unless otherwise mentioned, data were analyzed by one-factor analysis of variance (ANOVA), and the P values were corrected by Fisher's protected least significant difference (Fisher's PLSD) for multiple comparisons. A P value smaller than .05 was considered significant.

Drugs

T-440 and rolipram were synthesized by Tanabe Seiyaku Co. Ltd., (Osaka, Japan). Aminophylline, ACh and histamine were purchased from Sigma Chemical Co. (St. Louis, MO), theophylline was from Nacalai Tesque (Kyoto, Japan) and house dust antigen was from Torii Co. (Tokyo, Japan). T-440, rolipram, aminophylline and theophylline were dissolved in dimethyl sulfoxide (DMSO) and subsequently diluted in the vehicle solution or reaction mixture. The final concentration of DMSO was 0.01%. ACh, histamine, and house dust antigen were dissolved in Krebs-Henseleit solution just before each experiment. The concentration given is the final bath concentration.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Effect of T-440 on Bronchial Contraction

Reversal effects of T-440 on bronchial contraction induced by histamine, ACh and allergen. The magnitude of the bronchial contraction induced by 10⁵ M histamine was similar between groups (control, 0.87 ± 0.14 g; 10⁷ M T-440, 0.59 ± 0.14 g; 10⁶ M T-440, 0.51 ± 0.14 g; 10⁵ M T-440, 0.82 ± 0.11 g; 10⁵ M aminophylline, 0.65 ± 0.18 g; 3.3 × 10⁵ M aminophylline, 0.91 ± 0.11 g; P > .05). T-440 and aminophylline reversed the contraction in a concentration-dependent fashion (fig. 2A). 10⁶ and 10⁵ M T-440 and 3.3 × 10⁵ M aminophylline significantly reversed the contraction compared with the control, but 10⁵ M aminophylline did not (ANOVA with repeated measures). The reversal effect of 10⁵ M T-440 was significantly greater than that of 10⁵ and 3.3 × 10⁵ M aminophylline.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Reversal effect of T-440 on 105 M histamine-, 104 M ACh- and allergen-induced bronchial contraction. We defined the magnitude of each spasmogen-induced contraction and the time at which each drug or solvent was added as 100% contraction and 0 min, respectively. Each value is mean ± S.E. A) Histamine (n = 6-11). B) ACh (n = 4). C) Allergen (n = 6  7).  control,  107 M T-440,  106 M T-440,  105 M T-440,  105 M aminophylline,  3.3 × 105 M aminophylline. * P < .05, ** P < .01 and *** P < .0001 compared with control.  P < .05 compared with 105 M aminophylline.  P < .0001 and P < .01 compared with 105 and 3.3 × 105 M aminophylline.

4

7

6

5

5

5

5

5

6

5

5

5

5

5

5

5

5

Preventive effects of T-440 on bronchial contraction induced by histamine, ACh and allergen. T-440 and aminophylline relaxed the basal tension in a concentration-dependent fashion (fig. 3). When we expressed the relaxation as the percentage of contraction induced by 10³ M ACh, 10⁶ and 10⁵ M T-440 and 3.3 × 10⁵ M aminophylline caused significant relaxation compared with the control. Moreover, 10⁵ M T-440 caused significant relaxation compared with 10⁵ M aminophylline. After adjusting the tension to 1 g again, we cumulatively added histamine, ACh or allergen to the tissues.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Relaxing effect of T-440 on basal tension. Here % relaxation was expressed as the percentage of contraction induced by 103 M ACh. Each value is mean ± S.E.; n = 12 to 20. * P < .05 and ** P < .01 compared with control.  P < .05 compared with 105 M aminophylline.

3

7

6

5

5

5

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Preventive effect of T-440 on histamine-, ACh- and allergen-induced bronchial contraction. Here % contraction was defined as described in "Materials and Methods." Each value is mean in histamine-induced contraction (n = 8, panel A) and ACh-induced contraction (n = 4, panel B) and mean ± S.E. in allergen-induced contraction (n = 8, panel C). The inset shows the preventive effect of T-440 on allergen-induced maximal contraction.  control, 107 M T-440,  106 M T-440,  105 M T-440,  105 M aminophylline,  3.3 × 105 M aminophylline. * P < .05, ** P < .01 compared with control.

3

7

6

5

5

5

3

6

5

5

5

5

5

Extraction, Separation and Characterization of PDE Isozymes from Human Bronchus

Cytosolic cAMP-PDE activities of human bronchial tissue were separated by Mono-Q column chromatography into three peaks (fig. 5A). In accordance with the nomenclature proposed by Beavo and Reifsnyder (Beavo and Reifsnyder, 1990), the first peak (fraction 14) and the second peak (fraction 22) of cAMP-PDE activity were characterized as PDE1 and PDE2, respectively. But because the first peak overlapped another peak of cGMP-PDE activity (fig. 5A), it was suggested to be a mixture of PDE1 and PDE5. The third peak (fraction 26) was the major peak of cAMP hydrolysis (45% of total cAMP-PDE activity). It did not hydrolyze cGMP, and its cAMP-PDE activity was unaffected by the addition of cold cGMP. Furthermore, it was potently inhibited by rolipram, an inhibitor of PDE4 (fig. 5B). Therefore, it was characterized as PDE4. The PDE4 fractions (Kleine et al., 1992; Columbo et al., 1993; Griswold et al., 1993) were mixed, and the preventive effects of T-440, rolipram and theophylline were evaluated. The IC₅₀ values of these inhibitors were 0.08 µM, 2 µM and >100 µM, respectively (fig. 6). PDE3 activity could not be detected.

View larger version (22K): [in this window] [in a new window]   Fig. 5.   Elution profile of PDE activities from human bronchus after separation on a Mono-Q anion exchange column. A) PDE activity when 1 µM cAMP () or 1 µM cGMP () was used as a substrate. B) PDE activity when cAMP was used as a substrate in the absence () and presence of 1 µM cold cGMP () or in the presence of 10 µM rolipram ().

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Inhibition of PDE4 activity by T-440, rolipram and theophylline in human bronchial tissue. IC50 values for the effect of T-440 (), rolipram () and theophylline () on PDE4 activity were 0.08 mM, 2 mM and >100 mM, respectively.

Measurement of cAMP Content

T-440 (10⁵ M) significantly increased cAMP content in the absence and presence of histamine (fig. 7). Interestingly, 10⁵ M T-440 evoked a greater accumulation of cAMP in the presence of histamine than in its absence. On the other hand, 3.3 × 10⁵ M aminophylline evoked a slight accumulation of cAMP, but this effect was not significant. The accumulation induced by 10⁵ M T-440 was significantly greater than that induced by 3.3 × 10⁵ M aminophylline.

View larger version (17K): [in this window] [in a new window]   Fig. 7.   Effect of T-440 on accumulation of cAMP in human bronchial tissue. Each value is mean ± S.E.; n = 5 to 11. * P < .05 compared with histamine-added and nonadded control (open column). ** P < .0001 compared with histamine-added and nonadded control.  P < .01.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

We examined the effects of the selective inhibition of PDE4 by the novel PDE4 inhibitor T-440 on human bronchial contraction induced by histamine, ACh and allergen. T-440 reversed in a concentration-dependent fashion the contraction induced by histamine, ACh and allergen. Pretreatment with T-440 also inhibited the maximal contraction induced by allergen, but not by histamine or ACh. The potency of T-440 was greater than that of isozyme-nonselective PDE inhibitor aminophylline. We also confirmed that T-440 inhibits human bronchial PDE4 activity and that inhibition of PDE4 by T-440 causes the accumulation of cAMP in human bronchial tissues at the concentration that reverses histamine-induced contraction.

In the human bronchial tissues, we could not detect PDE3 activity in the cytosolic fraction, but PDE3 is detected in the human airway tissues, including smooth muscle, and PDE3 inhibitors have reversal effects on precontracted human bronchial tissues (Torphy et al., 1992; Rabe et al., 1993). We speculate that the cytosolic fraction in the present study did not contain enough PDE3 isozyme. Therefore, we examined the potency of T-440 against PDE isozymes in guinea pigs and dogs (table 1 and Iwasaki et al., 1996). T-440 is a PDE4 inhibitor with preferential potency against PDE4 isozyme (Iwasaki et al., 1996). As table 1 shows, T-440 potently inhibits PDE4 activity in the guinea pig lung (IC₅₀ = 0.071 µM) and is approximately 10-fold more potent than rolipram (IC₅₀ = 0.71 µM). and it displayed a 690-fold selectivity for PDE4 compared with PDE3 in the guinea pig heart (IC₅₀ against PDE3, 49 µM) (Yamagata et al., 1997). In the present study, the preventive potency of T-440 against PDE4 in the human bronchus (IC₅₀ = 0.08 µM) was similar to that against PDE4 in the guinea pig lung, and T-440 was approximately 25-fold more potent than rolipram (IC₅₀ = 2 µM). We used T-440 at a concentration lower than 10 µM because the inhibition of PDE4 is selective at that concentration.

T-440 reversed in a concentration-dependent fashion the contraction induced by histamine and allergen. The effects of 10⁶ M T-440 on the contraction induced by histamine and allergen were similar to those of  3.3 × 10⁵ M aminophylline. The concentration 3.3 × 10⁵ M aminophylline is equivalent to 13.9 µg/ml, which is the adequate serum concentration clinically used. These data suggest that PDE4 inhibition is potent enough to relax human bronchial contraction, a result that supports previous reports (Cortijo et al., 1993). Reversal effects of T-440 and aminophylline on the contraction induced by ACh were remarkably smaller (fig. 2, A and B). This result, which is consistent with reports of previous studies in animals (Andersson et al., 1978; Souness et al., 1994; Kaminuma et al., 1996), may be due to the cholinergic inhibition of adenylate cyclase in the airway smooth muscle (Andersson et al., 1978; Jones et al., 1987). Harris and colleagues reported that the bronchorelaxant activity of a series of PDE4 inhibitors in guinea pigs in vivo and in vitro correlates with the ability of these agents to interact with high-affinity rolipram-binding site, not with their ability to inhibit PDE4 catalytic activity (Harris et al., 1989). We studied the potency of T-440 and rolipram against [³H]-rolipram binding site in the guinea pig brain, and ability of T-440 and rolipram to inhibit PDE4 in the guinea pig lung. We found that T-440 has lower potency against rolipram binding site and higher PDE4-inhibiting ability than rolipram (unpublished data). Although we did not study the reversal effect of rolipram, we speculate that T-440 may have lower reversal effect than rolipram.

T-440 and aminophylline relaxed bronchial basal tone in a concentration-dependent manner. We do not know which mediator regulates the basal tone, but cAMP may be one of the important regulators of airway tone. The effect of 10⁶ M T-440 on basal tension was similar to that of 3.3 × 10⁵ M aminophylline. These data suggest that T-440 is about 33-fold more potent than aminophylline in relaxing human bronchial basal tone.

Pretreatment with T-440 and aminophylline significantly prevented the contraction induced by allergen. They slightly prevented the histamine-induced contraction, but this effect was not significant. These results are also obtained in guinea pigs (Gustafsson and Persson, 1991). Thus there are difference between the reversal and the preventive effects of T-440 and aminophylline on the histamine- and ACh-induced contractions. Because T-440 and aminophylline relaxed the basal tension, we adjusted it before adding histamine and ACh. Therefore, when the preventive effects of T-440 and aminophylline were examined, the bronchial tissues did not contain the factors that contributed to the basal tone associated with T-440 and aminophylline. By contrast, when the reversal effects of T-440 and aminophylline were examined, the bronchial tissues did contain the factors that contributed to the basal tone associated with T-440 and aminophylline. Therefore, we speculate that the reversal effects were greater than the preventive ones. In spite of these facts, T-440 and aminophylline prevented the human bronchial contraction induced by allergen, which suggests that inhibition of PDE4 suppresses the release of chemical mediator from mast cells, thus inhibiting the contraction induced by allergen. Another possible mechanism in the inhibition of allergen-induced bronchial contraction is that T-440 inhibits the effects of other contractile mediators, especially leukotrienes, which are important mediators in asthma. But this possibility is remote, because T-440 did not have leukotriene B₄ and D₄ receptor antagonist activity in the guinea pig lung membrane (data not shown) and because PDE4 inhibition by rolipram did not inhibit LTD₄-induced bronchial contraction and constriction in guinea pigs (Howell et al., 1993; Underwood et al., 1994). In animal studies, rolipram inhibited histamine-induced bronchoconstriction significantly less than antigen-induced bronchoconstriction in the sensitized guinea pig in vivo (Howell et al., 1993; Underwood et al., 1993; Underwood et al., 1994). It also reduced the antigen-induced release of prostaglandin D₂ in the guinea pig trachea and that of histamine and leukotriene C₄ in murine mast cells (Griswold et al., 1993; Underwood et al., 1993). Thus our human bronchial ex vivo study is consistent with the animal studies. We conclude that 1) there are difference between the reversal and the preventive effects of T-440 and aminophylline on histamine- and ACh-induced bronchial smooth muscle contraction, and 2) T-440 and aminophylline prevent allergen-induced bronchial smooth muscle contraction. These findings are new insights in the pharmacology of PDE inhibitors in the human bronchus.

In human basophils, the PAF- or IgE-mediated release of histamine and leukotriene C₄ is reduced by the inhibition of PDE4, which appears to be the major PDE isozyme in mast cells and basophils (Kleine et al., 1992; Peachell et al., 1992; Torphy et al., 1992; Columbo et al., 1993). Inhibition of PDE4 reduces the functions of other inflammatory cells, including eosinophils (Dent et al., 1994; Souness et al., 1994), and of monocytes (Griswold et al., 1993; Molnar et al., 1993). Furthermore, T-440 inhibits interleukin-5 production in the peripheral mononuclear cells of asthmatic patients (Kaminuma et al., 1995). Inhibition of PDE4 also potentiates nonadrenergic, noncholinergic relaxation (Fernandes et al., 1994). In the present study, because the bronchial tissues are normal, these types of inflammatory cells may be not contributed.

T-440 accumulated cAMP content in bronchial smooth muscle in the presence or absence of histamine. This suggests that the inhibition of PDE4 causes the accumulation of cAMP, resulting in the relaxation of bronchial smooth muscle (Cortijo et al., 1993; Howell et al., 1993; Underwood et al., 1993). On the other hand, 3.3 × 10⁵ M aminophylline did not induce a significant accumulation of cAMP, although this concentration significantly reversed histamine-induced contraction as much as 10⁶ M T-440 did. These results suggest that aminophylline relaxed the bronchial contraction through mechanisms other than the inhibition of PDEs that hydrolyze cAMP.

The accumulation of cAMP induced by T-440 was significantly greater in the presence of histamine than in its absence. Histamine stimulates cAMP synthesis via H₂ receptors that are coupled to adenylate cyclase in the airway tissue, including human airway smooth muscle, though it remains in dispute whether this activation of H₂ receptors results in physiological changes (Dunlop et al., 1977; Chand and Eyre, 1978; Duncan et al., 1980; Florio et al., 1992). Therefore, our result may be due to the effect of T-440 on the accumulation of cAMP synthesized by histamine and on the accumulation of basal cAMP. In the present study, we could not show that histamine itself significantly induced cAMP synthesis. This may be due to the long incubation with histamine (60 min), because histamine-induced cAMP synthesis was rapid and took less than 6 min (Murad and Kimura, 1974; Duncan et al., 1980).

The bronchodilatory and anti-inflammatory effects of PDE4 inhibition may be useful in treating asthma. On the other hand, the inhibition of PDE3, which may relax the bronchus more than the inhibition of PDE4, has little or no effect on anti-inflammatory action (Rabe et al., 1993; Torphy et al., 1993). Additionally, the inhibition of PDE3 causes positive inotropic and arrhythmogenic actions (Nicholson et al., 1991; Hall, 1993), and long-term inhibition of PDE3 is associated with increased morbidity and mortality for patients with severe chronic heart failure (Packer et al., 1991). To treat asthmatic patients subject to chronic heart failure and arrhythmia, it is reasonable to try to use a selective PDE4 inhibitor.

The drawback of PDE4 inhibitors is emesis. Because the dose of T-440 that induced emesis is about 10-fold higher than that of rolipram in suncus (unpublished data), and because the PDE4-inhibiting activity of T-440 is 10-fold more potent than that of rolipram, the potency for emesis of T-440 may be 100-fold less than that of rolipram.

In summary, the inhibitor of PDE4, T-440, reversed histamine-, ACh- and allergen-induced bronchial contraction in the human bronchus. Pretreatment with T-440 also prevented allergen-induced contraction, but not histamine- and ACh-induced contractions, which suggests that T-440 inhibits the release of chemical mediators, probably from mast cells. The effects of T-440 were more potent than those of aminophylline. In addition, T-440 caused the accumulation of cAMP at the concentration that relaxed histamine-induced contraction. Thus the present study suggests that the selective inhibition of PDE4 is a candidate for the treatment of asthma.