Department of Pediatrics, Shinshu University School of Medicine, Matsumoto, Japan
Correspondence: Kozo Yasui, Department of Pediatrics, Shinshu University School of Medicine, Asahi 3-1-1, Matsumoto 390-8621, Japan. E-mail: koyasui{at}gipac.shinshu-u.ac.jp
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSREFERENCES |
|---|
Key Words: superoxide anion production • chemotaxis • adenosine • receptor antagonism • airway inflammation
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSREFERENCES |
|---|
There is increasing evidence that theophylline has anti-asthma properties other than bronchodilation, which include anti-inflammatory actions [4 , 5 ]. The inhibition of a late asthmatic response by theophylline has been demonstrated in conjunction with significant reduction of inflammatory cells in the airway [6 , 7 ], and these effects were seen at plasma concentrations below those traditionally regarded as therapeutically useful [6 ]. Bronchial hyper-responsiveness is an important feature in asthma, while inflammation associated with granulocyte (neutrophil as well as eosinophil) infiltration is involved in its development. Granulocytes are thought to damage the normal airway tissue through releasing toxic substances such as superoxide and harmful proteins.
Because neutrophil and eosinophil activation is associated with late-phase asthmatic reactions [3 , 6 , 7 ], to elucidate the effects of theophylline on these cells may be of importance.
Several studies have been published on the effects of theophylline on neutrophil functions such as generation of oxygen metabolites and chemotactic movement [8 9 10 11 12 ]. It has been established that the physiological mechanism for the effect of theophylline on the modulation of neutrophil functions may be dependent on adenosine receptor antagonism and/or on PDE inhibition. Adenosine, which is normally present in plasma at concentrations up to 3 x 10-7 M [13 , 14 ], is one of the pharmacological and hormonal agents that modulate neutrophil functions in vitro [10 , 14 15 16 17 18 ]. Neutrophils release a certain amount of adenosine [13 ], and adenosine A1 and A2A but not A2B receptors are known to be strongly expressed on neutrophils [19 , 20 ]. Theophylline (an A1, A2A, and A2B antagonist) is believed to modulate neutrophil functions at least in part through selective adenosine receptor antagonism. However, very limited information on the effects and mechanism of theophylline on eosinophil functions is available compared with those on neutrophils, and little confused [21 22 23 24 ].
This study was undertaken to determine the involvement of adenosine receptor antagonism and PDE inhibition in the effect of theophylline on eosinophil functions and to compare it with that on neutrophils to reach a consolidated view of the functions and mechanisms of theophylline.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSREFERENCES |
|---|
Preparation of cells
Heparinized venous blood was obtained from healthy volunteers.
Neutrophils were isolated by using dextran sedimentation and
centrifugation on a Histopaque gradient (without endotoxin) as
previously described [26
]. Contaminating red blood cells
were removed by means of cold hypotonic water lysis. Cells were then
subjected to another density cut using centrifugation over a Percoll
gradient in order to isolate neutrophils from contaminating
eosinophils. Eosinophils were purified from the peripheral blood of
slightly allergic donors (mild rhinitis) with the aid of gradient
centrifugation and negative selection with anti-CD16 mAb-coated
immunologic magnetic beads and a magnetic cell sorter. The purity of
the isolate was assessed by preparing cytocentrifuged smears and
staining with May-Grünwald-Giemsa (Merck, Darmstadt, Germany)
stain. The purity of the neutrophils was >97% and that of the
eosinophil preparations was >98%, whereas the granulocyte viability
was >99% as determined by trypan blue dye exclusion (Sigma). Purified
cells were suspended in Hanks’ balanced salt solution (HBSS) without
calcium chloride or magnesium sulfate.
Superoxide anion (O2-) production
Superoxide production was determined at 37°C, with the
modified method of Cohen and Chovaniec [26
,
27
]. The amount of superoxide released was determined
with a Hitachi spectrophotometer U 2000 (Tokyo, Japan) as the change in
absorbance at 550 nm resulting from superoxide dismutase
(SOD)-inhibitable cytochrome c reduction. The reaction was
carried out for 5 min with fMLP (10-6 M), C5a
(10-7 M), and PMA (5 ng/mL), after which the reaction was
stopped by the addition of 0.5 mM N-ethylmalemide. The
generation of superoxide was calculated by subtracting the change in
absorbance in the presence of SOD (1 mM) from that in its absence, and
then dividing this value by 21.1 x 103/M/cm for the
molar extinction coefficient.
Cell motility (chemotaxis)
Cell motility was measured with the agarose method
[26
, 28
]. Briefly, 5 mL of 1.2% agarose
dissolved in HBSS containing Ca2+ and Mg2+ and
supplemented with 10% fetal calf serum (FCS) was placed in 60 x
15-mm Petri dishes. A 10-µL aliquot of cell suspension, containing
5 x 105 cells, was placed in the center well of the
plate, and equal volumes of the chemoattractant (2 x
10-7 M fMLP) and HBSS were placed in the outer and inner
wells, respectively. The wells were 3 mm in diameter, and the outer and
inner wells were located at a distance of 7 mm from the center one.
After incubation at 37°C in 5% CO2 for 2 h, the
plates were fixed with ethanol and formalin and the cells stained with
Wright’s stain (Sigma). Cell motility was defined as the linear
distance the cells moved from the center well in the direction of the
well containing the chemoattractant (chemotaxis), and the migration
distance was compared to that of controls. All results represent the
averages of duplicate measurements.
Statistical analysis
All data are presented as means ± SD.
Comparisons between paired conditions were made with the aid of paired
t tests. One-way factorial analysis of variance and multiple
comparison tests (Fisher’s PLSD) were used for comparative analyses of
data for several groups. P < 0.05 was considered
significant in all cases.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSREFERENCES |
|---|
 View larger version (13K): [in a new window] |
Figure 1. Effects of theophylline on superoxide anion
(O2-) production from neutrophils
(107 cells/mL) and eosinophils (2 x 106
cells/mL) stimulated with several agents. Cells were preincubated with
buffer or the indicated concentrations of theophylline for 15 min at
37°C before addition of stimulants, after which
O2- production was determined. The reaction
was conducted for 5 min. Results from five donors were normalized to
obtain percentages of control values and are shown as the mean ±
SD. *Significant differences from control
values: P < 0.05.
|
Effects of theophylline on chemotaxis
The maximum linear distance that cells had migrated from the
margin of the center well was measured and compared with that in
control experiments (Fig. 2
). The control migration distance for four individual donors was
3120 ± 160 µm in neutrophils, and 2250 ± 150 µm in
eosinophils. As shown, theophylline caused a concentration-dependent
reduction in the chemotactic response of both cells. The addition of
theophylline to neutrophils caused a significant decline in chemotaxis
at lower concentrations than those for eosinophils.
 View larger version (17K): [in a new window] |
Figure 2. Effects of theophylline on fMLP (2 x 10-7 M)-induced
chemotaxis of neutrophils and eosinophils. Chemotaxis was defined as
the maximum distance cells migrated under agarose (2 h) from the edge
of the center well. Results are shown as means ± SD
from three separate experiments and normalized as control percentage.
*Significant differences from control values:
P < 0.05.
|
 View larger version (18K): [in a new window] |
Figure 3. Effects of 8-sulfophenyltheophylline (8-SPT) or enprofylline on
O2- production from neutrophils (A) and
eosinophils (B) stimulated with fMLP (10-6 M). Cells were
preincubated with buffer or the indicated concentrations of drugs for
15 min at 37°C before addition of the stimulant, after which
O2- production was determined. The reaction
was conducted for 5 min. The cells were also incubated with 0.1 U/mL
ADA (filled triangles) or 1 µM KF-17837 (open triangles). Results are
shown as means ± SD from three separate experiments
and normalized as control percentage. *Significant
differences from control values: P < 0.05.
|
10-4 M).
Preincubation with 0.1 U/mL of ADA also significantly inhibited
neutrophil chemotaxis, whereas treatment of eosinophils with 8-SPT or
ADA (Fig. 4B)
had no effect on eosinophil chemotaxis. Inhibitory effect
was observed in eosinophils treated with enprofylline at higher
concentrations.
 View larger version (17K): [in a new window] |
Figure 4. Effects of 8-SPT or enprofylline on fMLP (2 x
10-7 M)-induced chemotaxis of neutrophils (A) and
eosinophils (B). Chemotaxis was defined as the maximum distance cells
migrated under agarose (2 h) from the edge of the center well. The
cells were also treated with 0.1 U/mL ADA (filled triangles) or 1 µM
KF-17837 (open triangles). Results are shown as means ±
SD from three separate experiments and normalized as
percentages of control values. *Significant differences
from control values: P < 0.05.
|
All of the adenosine analogs inhibited O2- production from neutrophils stimulated with fMLP (Fig. 5A ), with a rank order for potency obtained with selective adenosine receptor agonists of CGS-21680 (A2A) > NECA (A1 and A2A, A2B) > CHA (A1) > IB-MECA (A3). CGS-21680 was most potent in adenosine analogs, suggesting that an A2A-receptor was involved in neutrophil events. Exposure to 5 x 10-7 M CGS-21680 for 15 min resulted in a 50% inhibition of O2- production from neutrophils.
 View larger version (19K): [in a new window] |
Figure 5. Effects of adenosine analogs on
O2- production from neutrophils (A) or
eosinophils (B) stimulated with fMLP (10-6 M). Neutrophils
(107 cells/mL) and eosinophils (2 x 106
cells/mL) were preincubated with adenosine analogs (final DMSO
concentration < 0.1%) for 15 min at 37°C. Representative data
are shown. Essentially the same results were obtained in three
similarly designed experiments.
|
Comparative study between theophylline and 8-SPT in
O2- production from neutrophils
A likely explanation of the inhibitory effects of
O2- production at high concentrations of
theophylline (>10-3 M) can be found in the intracellular
cAMP accumulation as a result of PDE inhibition. On the basis of the
finding that 8-SPT does not inhibit PDE activity [10
,
29
], we used 8-SPT and an excess of dbcAMP to reproduce
the biphasic effect of theophylline on O2-
production from neutrophils stimulated with fMLP (10-6 M).
dbcAMP was used as a cell-permeant cAMP analog that can modulate
intracellular signaling and cellular function. The results are shown in
Figure 6
. The potentiation of O2- production from
neutrophils by 8-SPT was reversed by the addition of dbcAMP, and an
excess of dbcAMP (5 x 10-4 M) significantly inhibited
O2- production from neutrophils in comparison
to that seen in controls (ANOVA: F3,9 = 60.642,
P < 0.0001). In other words, the biphasic effect of
theophylline could be reproduced with these chemicals.
 View larger version (21K): [in a new window] |
Figure 6. Effects of 8-SPT and dbcAMP on O2- production
from neutrophils. Neutrophils (107 cells/mL) were
preincubated for 15 min at 37°C with buffer and the indicated
concentrations of 8-SPT alone or in the presence of dbcAMP.
O2- production was conducted for 5 min under
stimulation with fMLP (10-6 M). Results are shown as
means ± SD from three separate experiments and
normalized as percentages of control.
|
It should be noted that a different effect of theophylline was observed in eosinophils, where neither theophylline nor A2A-specific adenosine analog had any potentiating effect on O2- production. This appears to be consistent with the results of receptor binding studies that few adenosine A2A receptors occur in eosinophils [20 , 34 ]. The inhibitory effect of theophylline on O2- production from eosinophils was previously reported by several investigators [21 , 23 , 24 ]. With regard to the effect of theophylline, contradictory findings to ours have been reported by different parties [21 , 24 ]. They reported that lower concentrations of theophylline may potentiate O2- production from eosinophils. Yukawa and colleagues [25 ] reported that eosinophil activation is caused by competing with circulating adenosine for eosinophil A2 receptors. One explanation for this clear discrepancy could be the insufficient purity of eosinophils. Although high-purity eosinophils should have been obtained for their functional study, the centrifugation over Percoll density gradients was used, resulting in low-purity eosinophils (80%). In contrast, the magnetic bead method is recommended for the isolation of eosinophils with a high purity (>98%) without functional priming [35 ]. It can therefore be assumed that the eosinophils used in the study contained 20% neutrophils; the potentiation of superoxide production by theophylline could therefore be the result of neutrophil contamination. In a recent study, however, Ezeamuzie and Al-Hage [24 ] again reported a biphasic effect of theophylline on the activation of highly purified eosinophils. Furthermore, they observed significant inhibition of O2- production from eosinophils at much lower concentrations (10-6 and 10-5 M). These findings are contrary to other studies [21 , 23 ], but so far the causes of these discrepancies are not fully understood.
Higher concentrations (
10-3 M) of theophylline caused
significant inhibition of O2- production from
both neutrophils and eosinophils. Our findings showed that theophylline
analog 8-SPT, more potent A1 and A2A receptor
antagonists, which lacks PDE inhibitory activity [10
,
29
], had only a potentiating effect on
O2- production in neutrophils. On the other
hand, enprofylline, which is known as an adenosine A1 and
A2B antagonist [11
, 30
] with
PDE inhibitory activity, inhibited O2-
production from both cells. This suggests that the difference in
effects between theophylline and enprofylline might be due to the
presence or absence of A2A antagonism and that the
inhibitory effects of these xanthine derivatives on
O2- production may be due to PDE inhibition.
The inhibitory effects of theophylline and enprofylline on cAMP-PDE are
already well known [11
, 36
]. In addition,
we were able to reproduce the biphasic effect of theophylline on
O2- production from neutrophils by using 8-SPT
and an excess of dbcAMP. The addition of dbcAMP reversed the
potentiation of O2- production from
neutrophils by 8-SPT. Selective PDE IV inhibitors have been found to be
very effective for the inhibition of superoxide production from
activated eosinophils [14
, 23
,
37
]. These findings support our contention that the
inhibitory effects of O2- production at a high
concentration of theophylline can be explained by intracellular cAMP
accumulation as a result of PDE inhibition.
The elevation of intracellular cAMP inhibits activation of several functions of human granulocytes including O2- production induced by receptor-linked agonists [38 ]. In contrast, when PMA was used to activate cells, theophylline had no effect on O2- production from either neutrophils or eosinophils. cAMP PDE inhibitor in combination with prostaglandin E1 markedly inhibits superoxide production from neutrophils stimulated with fMLP, but not from those stimulated with phorbol ester [39 ], whereas inactive C5b67 inhibits C5a- and fMLP-, but not PMA-induced superoxide production [40 ]. These results indicate that such inhibitions occur independent of protein kinase C activation, and that cAMP levels represent the second messenger interfering with intracellular signaling stimulated with receptor-linked agonists and not with PMA in granulocytes.
The addition of theophylline to neutrophils caused a significant decline in chemotaxis at lower concentrations than those used for eosinophils. The effect thus seems to depend on the existence of a specific receptor for adenosine. A1 receptor activation in neutrophils is known to be associated with augmentation of chemotaxis [15 ]. Theophylline’s antagonism does not discriminate between A1 and A2A receptor subtypes [20 ], and neutrophils express both A1 and A2A receptors [15 , 19 ]. The difference in effects of 8-SPT and enprofylline on neutrophil chemotaxis was according to their affinities for adenosine A1 receptor [20 ]. On the other hand, eosinophils possess fewer binding sites for adenosine A1 and A2A agonists [34 ] and cAMP is known as an inhibitory modulator of leukocyte chemotaxis [41 ]; the inhibitory effects on eosinophil chemotaxis at a high concentration of enprofylline and theophylline can be explained by intracellular cAMP accumulation as a result of PDE inhibition.
Recently, it was found that human eosinophils express higher levels of adenosine A3 receptors [34 ], and that A3 receptor stimulation modulates the functions of eosinophils [30 , 42 , 43 ]. We also confirmed that a selective A3 receptor agonist (IB-MECA) reduces O2- production in eosinophils most effectively [43 ]. Furthermore, theophylline did not influence the effect of IB-MECA (data not shown), which may be due to the lack of A3 receptor antagonism.
The primary aim of this study was to organize the currently available information on the effects of theophylline on granulocyte functions, and to understand the mechanism that mediates cellular functions. We can conclude that theophylline modulates granulocyte functions in conjunction with adenosine receptor antagonism (A1 and A2A) and cAMP-PDE inhibition. Furthermore, the difference in effect on neutrophils and eosinophils appears to depend on the existence of specific adenosine receptors. In connection with the anti-inflammatory actions of theophylline, the results presented here may raise fears that neutrophils subjected to theophylline therapy may exhibit enhanced superoxide anion production and cause normal tissue damage. However, the priming effect of theophylline on O2- production is very small compared to that of inflammatory cytokines and is normalized by the inhibitory action of PDE at therapeutic concentrations. In addition, accelerated neutrophil apoptosis by theophylline [44 ] may well prevent any prolonged airway inflammation. In conclusion, the anti-inflammatory effects of theophylline on granulocyte functions are partly the result of inhibition of the release of toxic oxygen metabolites from eosinophils and/or suppression of chemotactic migration by both neutrophils and eosinophils.
Received November 22, 1999; revised February 29, 2000; accepted March 1, 2000.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  W.-J. Luo, J.-F. Qian, and H.-H. Jiang Pretreatment with aminophylline reduces release of Troponin I and neutrophil activation in the myocardium of patients undergoing cardioplegic arrest Eur. J. Cardiothorac. Surg., March 1, 2007; 31(3): 360 - 365. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G Caramori and A Papi Oxidants and asthma Thorax, February 1, 2004; 59(2): 170 - 173. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Yasui, H. Kanda, T. Iwanami, and A. Komiyama Increased serum concentration of urinary trypsin inhibitor with asthma exacerbation Eur. Respir. J., November 1, 2003; 22(5): 739 - 742. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. V. Culpitt, C. de Matos, R. E. Russell, L. E. Donnelly, D. F. Rogers, and P. J. Barnes Effect of Theophylline on Induced Sputum Inflammatory Indices and Neutrophil Chemotaxis in Chronic Obstructive Pulmonary Disease Am. J. Respir. Crit. Care Med., May 15, 2002; 165(10): 1371 - 1376. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. D. Loike, L. Cao, S. Budhu, S. Hoffman, and S. C. Silverstein Blockade of {{alpha}}5{{beta}}1 Integrins Reverses the Inhibitory Effect of Tenascin on Chemotaxis of Human Monocytes and Polymorphonuclear Leukocytes Through Three-Dimensional Gels of Extracellular Matrix Proteins J. Immunol., June 15, 2001; 166(12): 7534 - 7542. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
