Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

There is increasing evidence that erythromycin (EM), a 14-membered ring macrolide, is effective in the treatment of chronic inflammatory airway diseases such as diffuse panbronchiolitis (1). The mechanism of this action is thought to be anti-inflammatory and immunoregulatory rather than antibacterial, because EM (200-600 mg/d) below the minimum concentrations for killing of common superinfecting organisms is clinically effective in the treatment of these diseases (2). One major improvement of symptoms after EM therapy is a decrease in sputum volume (3). The change is supported by in vitro experimental evidence that EM inhibits mucin secretion (4) and may inhibit water secretion by inhibiting Cl secretion (5) in airway epithelial cells. However, the intracellular site of action of EM in airway epithelium remains unclear.

Intracellular Ca²⁺ plays an important role as a second messenger in Cl ion transport and mucin secretion stimulated by inflammatory mediators such as adenosine triphosphate (ATP) (6, 7). Recently, another immunosuppressive macrolide, FK506, has been shown to affect Ca²⁺ dynamics in cardiomyocytes (8) and airway epithelial cells (9) through the inhibition of FK binding protein (FK-BP), which is an anchor protein of ryanodine and inositol 1,4,5-triphosphate (IP₃) receptors (10). This led us to examine whether EM also modulates intracellular Ca²⁺ dynamics. In this study, we investigated the effect of EM on the ATP-induced increase in intracellular Ca²⁺ in bovine tracheal epithelium. Furthermore, the effect of EM on ATP-induced Ca²⁺ oscillations was studied in individual cells using Ca²⁺ image analysis.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Culture

Bovine tracheas were obtained from a slaughterhouse. Strips of epithelium were pulled off the submucosa, washed four times with phosphate-buffered saline (PBS) containing 5 mM dithiothreitol, and rinsed twice with PBS. Then epithelial tissues were digested with PBS containing 0.05% protease (type XIV; Sigma Chemical Co., St. Louis, MO) at 4°C overnight. After terminating the digestion by addition of fetal calf serum (FCS; final concentration ~ 2.5%), cells were pelleted (200 × g, 10 min) and suspended in 50% Dulbecco's modified Eagle's medium (DMEM) and 50% Ham's nutrient F12 containing 5% FCS, nonessential amino acids, penicillin (10⁵ U/liter), streptomycin (100 mg/ liter), and gentamicin (50 mg/liter). The isolated cells were plated at 2.5 × 10⁵ cells/cm². The medium was changed every 2 d. In our experimental system, it usually took 3 to 4 d to obtain the confluence.

Intracellular Calcium Concentration Measurement

The isolated cells were cultured on round coverslips (15 mm diameter; Matsunami Ltd., Tokyo, Japan), which were coated with human placental collagen (20 µg/cm²). After confluence was achieved, the coverslips were washed with Hanks' balanced salt solution (HBSS), which contained 10 mM N-2-hydroxyethylpiperazine-N'-ethane sulfonic acid (Hepes) (pH 7.4), and the cells were loaded with 2 µM acetoxymethyl ester of fura-2 (fura2-AM; Dojin, Kumamoto, Japan) for 20 min at 37°C. The coverslips were then washed again and held with a rigid holder in a continuously stirred cuvette containing Hepes-buffered HBSS maintained at 37°C, and the fluorescence intensity was measured with a spectrophotometer (CAF-110; Japan Spectroscopic Co., Tokyo, Japan) (11). For excitation of fura-2, ultraviolet (UV) lights of 340- and 380-nm wavelength were automatically exchanged at a rate of 50 Hz; the emitted light from cells (F340 and F380) was detected with a photomultiplier tube through a 510 ± 10 nm bandpass filter, and the fluorescence intensity ratio F340/F380 was automatically calculated. Maximal and minimal values for the ratio were determined in the presence of ionomycin (10 µM) and 5 mM ethyleneglycol-bis-(-aminoethylether)-N,N'-tetraacetic acid (EGTA), respectively. Intracellular calcium concentration ([Ca²⁺]_i) was calculated using the formula described by Grynkiewicz and associates (12).

Because the method using the CAF-110 system described above represents mean Ca²⁺ change in numerous cells on a coverslip, Ca²⁺ oscillations that might occur in individual cells could not be detected. Therefore, the single-cell Ca²⁺ image technique was used to detect Ca²⁺ oscillations (13). Cells cultured on glass-bottomed petri dishes (P35GC-101; MatTek Co., Ashland, MA) were washed with Hepes-buffered HBSS and loaded with 10 µM fura2-AM for 1 h. UV light of 340- or 380-nm wavelength was produced by a xenon lamp and narrow bandpass filters, and applied to the cells through a ×40 objective lens (Fluor 40; Nikon, Tokyo, Japan). Emission fluorescence was led to a silicon intensifier target camera through a 510 ± 10-nm bandpass filter. Ca²⁺ images of single epithelial cells were sampled at 3-4-s intervals. Data sets were stored on the hard disk of the computer as 8-bit (256 × 256 pixels) digital images, and processed to calculate the ratio later. All of these procedures were performed using an image processor (Argus 50; Hamamatu Photonics, Hamamatu, Japan). Ca²⁺ concentration of Hepes-buffered HBSS in the absence or presence of EM was measured by the OCPC method (14).

Drugs

EM, ATP, uridine triphosphate (UTP), ionomycin, EGTA, Hepes, LaCl₃, ampicillin, cephasolin, penicillin, streptomycin, gentamicin, and verapamil were purchased from Sigma. Clarithromycin was a gift from Taisho Pharmaceutical Co. (Tokyo, Japan). DMEM, Ham's F12, and nonessential amino acids were purchased from Gibco Co. (Tokyo, Japan).

Statistics

Data are shown as means ± SE. Statistical analysis was performed by two-tailed paired or unpaired Student's t test, and a P value of less than 0.05 was considered significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Basal [Ca²⁺]_i in the bovine tracheal epithelium was 120 ± 14 nM (n = 45). Exposure of cultured tracheal epithelial cells to ATP (10⁴ M) rapidly increased F340/F380 (Figure 1A). This [Ca²⁺]_i response was biphasic, consisting of an initial transient rise that peaked within 15 s followed by a sustained response. The latter gradually decreased but continued for more than 10 min. In this study, the value at 2 min after addition of ATP was selected as the sustained response. Transient and sustained increases in [Ca²⁺]_i in response to ATP (10⁴ M) were 640 ± 56 and 204 ± 5 nM, respectively (n = 20). EM per se caused no significant change in [Ca²⁺]_i. Pretreatment of the cells with EM (10⁷ M to 10⁴ M) for 10 min had little effect on the ATP-induced transient response. However, EM significantly inhibited the ATP-induced sustained increase in [Ca²⁺]_i in a dose-dependent manner (Figures 1B and 2). The decreases in ATP-induced sustained response in the presence of 10⁵ M and 10⁴ M EM were 47.8 ± 13.7% and 79.2 ± 14.1%, respectively (n = 6 for each concentration; P < 0.05 and P < 0.01, respectively). Another 14-membered ring macrolide, clarithromycin (10⁵ M) similarly inhibited the ATP-induced sustained response, but ampicillin (10⁵ M) and cephasolin (10⁵ M) did not (Figure 3).

View larger version (13K): [in this window] [in a new window]   Figure 1.   Representative recordings of fura-2 fluorescence ratio in bovine tracheal epithelial cells exposed to ATP (104 M) in the absence or presence of EM. (A) Control; (B) EM (104 M). Cells were pretreated with EM for 10 min. EM had little effect on ATP-induced transient response, whereas EM inhibited ATP- induced sustained response.

View larger version (46K): [in this window] [in a new window]   Figure 3.   Effects of various antibiotics on ATP-induced sustained responses. CAM, AB-PC, and CEZ represent clarithromycin, ampicillin, and cephazolin, respectively, all at 105 M. Cells were pretreated with the antibiotics for 10 min; n = 6 from six individuals. *P < 0.05, significantly different inhibitions from control.

UTP, a potent agonist for P_2u purinoceptors, induced a biphasic [Ca²⁺]_i increase (Figure 4A). Transient and sustained increases in [Ca²⁺]_i in response to UTP were 681 ± 22 and 247 ± 14 nM, respectively (n = 6). Furthermore, EM strongly inhibited the UTP-induced sustained response but not transient response (Figure 4B). The decreases in UTP-induced sustained response in the presence of 10⁴ M EM were 85.7 ± 11.8% (n = 6, P < 0.01).

View larger version (13K): [in this window] [in a new window]   Figure 4.   Representative recordings of fura-2 fluorescence ratio in bovine tracheal epithelial cells exposed to UTP (104 M) in the absence or presence of EM. (A) Control; (B) EM (104 M). Cells were pretreated with EM for 10 min. EM had little effect on UTP-induced transient response, whereas EM inhibited UTP- induced sustained response.

After addition of 1 mM EGTA, which lowered the calculated Ca²⁺ concentration of the medium to 70 nM, [Ca²⁺]_i decreased slightly (Figure 5A) from 123 ± 10 to 111 ± 4 nM (n = 6). Under this condition, the ATP- induced transient response remained, whereas the sustained response was abolished (Figure 5A). ATP (10⁴ M)-induced transient and sustained levels of [Ca²⁺]_i in the presence of 1 mM EGTA were 687 ± 60 and 95 ± 8 nM, respectively (n = 6). This suggests that the sustained response reflects Ca²⁺ influx from the extracellular solution. The ATP- induced transient response of EM-treated cells remained in the presence of 1 mM EGTA, and the recording of EM-treated cells closely resembled that of control cells in the presence of EGTA (Figures 5A and 5B). To elucidate what type of Ca²⁺ channel is involved in ATP-induced Ca²⁺ influx, the effects of Ca²⁺ channel blockers were examined. As a result, the ATP-induced sustained response was abolished in the presence of 1.5 mM La³⁺ (Figure 6A), whereas verapamil (10⁶ M) had no effect on the ATP- induced sustained response (Figure 6B).

View larger version (16K): [in this window] [in a new window]   Figure 5.   Effect of Ca2+-deficient medium on ATP-induced responses. (A) In the presence of EGTA (1 mM), ATP (104 M)- induced sustained response was abolished. (B) After pretreating cells with EM (104 M) for 10 min, then adding EGTA (1 mM), ATP-induced transient response remained and the sustained response disappeared as in A.

View larger version (14K): [in this window] [in a new window]   Figure 6.   The effect of Ca2+ channel blockers on ATP-induced responses. (A) In the presence of La3+ (LaCl3, 1.5 × 103 M), ATP-induced sustained response was abolished. (B) Pretreating the cells with verapamil (106 M) for 5 min had no effect on ATP-induced transient and sustained response.

In Ca²⁺ image analysis of single cells, spontaneous Ca²⁺ oscillations were not observed in the unstimulated state. However, more than 30% of subconfluent cells showed Ca²⁺ oscillations in response to low concentration of ATP (10⁶ M). The Ca²⁺ oscillations occurred at a period of 27.0 ± 2.2 s (n = 30), having nearly constant peak values of 315 ± 10 nM (n = 36) (Figure 7A), and continued at least for 30 min. In contrast, high concentration of ATP (10⁴ M) induced biphasic responses, which consisted of a transient response followed by a sustained response (Figure 7B) similar to responses of confluent cell sheets in Figure 1A. The transient response was larger than the peak of Ca²⁺ oscillations induced by ATP (10⁶ M). After addition of EM (10⁴ M), Ca²⁺ oscillations were strongly attenuated in both frequency and amplitude, and eventually abolished within ~ 3 min (Figure 7C). EM (10⁵ M) also decreased the amplitude and the frequency of ATP (10⁶ M)-induced Ca²⁺ oscillations, but the oscillations were not abolished (Figure 7D). After addition of 1 mM EGTA, Ca²⁺ oscillations were attenuated with longer intervals and eventually abolished (Figure 7E).

View larger version (19K): [in this window] [in a new window]   View larger version (8K): [in this window] [in a new window]   View larger version (17K): [in this window] [in a new window]   View larger version (15K): [in this window] [in a new window]   View larger version (15K): [in this window] [in a new window]   Figure 7.   Representative recordings of ATP-induced Ca2+ responses in single cells. (A) Low concentration of ATP (106 M) induced Ca2+ oscillations. (B) High concentration of ATP (104 M) induced transient response followed by sustained response. (C) ATP (106 M)-induced Ca2+ oscillations were strongly inhibited by EM (104 M). (D) EM (105 M) reduced the amplitude and the frequency of Ca2+ oscillations, but oscillations were not abolished. (E) EGTA (1 mM) abolished Ca2+ oscillations.

To rule out the possibility of a chelating effect of EM, the Ca²⁺ concentration of the extracellular solution was measured by the OCPC method (14). Ca²⁺ concentration of HBSS in the presence of EM (10⁴ M) was 1 mM, which was similar to that of HBSS alone.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The present study demonstrated that EM inhibited ATP-induced sustained Ca²⁺ response, which mainly reflected Ca²⁺ influx from extracellular solution. This inhibitory effect of EM is unlikely to be due to chelation of Ca²⁺, because the Ca²⁺ concentration of HBSS was not changed in the presence of EM. These data suggest that EM specifically inhibits Ca²⁺ influx.

In single-cell Ca²⁺ image analysis, EM inhibited ATP-induced Ca²⁺ oscillations (Figures 7C and 7D). Although Ca²⁺ oscillations arise from repetitive Ca²⁺ release from intracellular Ca²⁺ stores in nonexcitable cells (15, 16), maintenance of these oscillations requires the refilling of intracellular Ca²⁺ stores (16). Our data showed that ATP-induced Ca²⁺ oscillations were abolished when extracellular Ca²⁺ was removed (Figure 7E). Therefore, EM action on Ca²⁺ oscillations may also reflect the inhibition of refilling Ca²⁺ stores.

In the present study, UTP, a potent agonist for P_2u purinoceptors, induced a biphasic [Ca²⁺]_i increase (Figure 4A). As previously shown by others (6), the potency of Ca²⁺-elevating effect was UTP  ATP. Furthermore, EM inhibited the UTP-induced sustained response as well as the ATP-induced sustained response (Figure 4B). These data suggest that EM inhibits Ca²⁺ influx induced by P_2u purinoceptor stimulation in bovine tracheal epithelial cells. ATP is reported to stimulate P_2u purinoceptors in human nasal (6) and hamster tracheal epithelium (7), activate phospholipase C, and produce IP₃, which, in turn, mobilizes intracellular Ca²⁺ from endoplasmic reticulum via IP₃ receptors (7). However, the mechanism for agonist-induced Ca²⁺ influx is not completely understood. One possible mechanism for Ca²⁺ influx in nonexcitable cells is store-dependent Ca²⁺ entry (17, 19, 20). The major pathway may be capacitative Ca²⁺ entry activated by emptying Ca²⁺ stores (19, 20). In the present study, ATP-induced Ca²⁺ influx was blocked by La³⁺ but not by an L-type, voltage-dependent Ca²⁺ channel blocker, verapamil (Figures 6A and 6B). Berridge (20) and Hoth and Penner (21) reported that bivalent and trivalent cations such as La³⁺ inhibit calcium release-activated calcium (CRAC) channels that refer to capacitative Ca²⁺ entry. CRAC channels may be regulated by several messengers, such as Ca²⁺ influx factor and 1,3,4,5-tetrakisphosphate (20, 22). Therefore, EM may affect these messengers and, as a result, inhibit store-dependent Ca²⁺ entry. Further studies are needed to prove this hypothesis.

Recent evidence suggests that FK506, an immunosuppressive macrolide, modulates Ca²⁺ dynamics via binding FK-BP, which is an anchor protein of IP₃ and ryanodine receptors (8). It is unknown whether EM affects IP₃ and ryanodine receptors. However, this possibility is unlikely because EM had little effect on the ATP-induced transient response (Figures 1, 2, and 5B), which mainly reflects Ca²⁺ mobilization from intracellular stores. Antilymphatic activity of EM is also reported to be distinct from that of FK506 (23).

View larger version (41K): [in this window] [in a new window]   Figure 2.   Concentration-dependent effect of EM on the ATP- induced transient and sustained responses; n = 6 from six individuals. Data are expressed as means ± SE. Significant inhibition from percent control is shown as *P < 0.05, P < 0.01. Hatched columns and open columns indicate transient and sustained responses, respectively.

In our study, another 14-membered ring macrolide, clarithromycin, also inhibited the ATP-induced sustained response (Figure 4). This is consistent with the finding that both clarithromycin and EM remarkably improve the prognosis of diffuse panbronchiolitis by decreasing in sputum volume (3, 24). In vitro, EM inhibits mucus secretion (4), active Cl secretion, and presumably water (5, 25). Although the possibility cannot be ruled out that the effects of EM on calcium influx in bovine tracheal epithelium could be unrelated to its effects on other cell parameters in human or canine airway epithelium, our data suggest that these effects of macrolides are due to inhibition of Ca²⁺-dependent signals.

In conclusion, EM inhibits the sustained increases in [Ca²⁺]_i and Ca²⁺ oscillations induced by ATP in airway epithelial cells. This inhibitory action is probably due to a block of Ca²⁺ entry. Thus, the therapeutic actions of this compound may be due to a suppression of Ca²⁺-dependent epithelial secretion of mucus and water.