Introduction

Top Abstract Introduction Methods Results Discussion References

Although our knowledge is still fragmentary and sometimes controversial, uterine mast cell and its mediators appear to play a role in the regulation of uterine contractility that may be important during parturition. This is supported by anatomical and functional considerations. For example, mast cells are present around blood vessels and between myometrial cells in sections of uterine mouse and human tissue (Padilla et al., 1990; Rudolph et al., 1993). The suggestion that they may play a role in the regulation of uterine contractility is supported by the fact that in the myometrium of animals undergoing pregnancy, a gradual increase in the number of mast cells and in the concentration of histamine, a marker of mast cell concentration in this tissue (Rudolph et al., 1997), has been demonstrated. These changes are observed during the second half and peak at the end of the gestational period (Padilla et al., 1990; Tabb, 1994). The increment in the concentration of histamine returns to pregestational values after labor, which suggests a massive activation of mast cells with the consequent release of histamine and possibly other contractile and proinflammatory mediators during the process of parturition (Padilla et al., 1990).

On stimulation, uterine mast cells may release histamine, serotonin and lipid mediators such as prostaglandins, leukotrienes and platelet-activating factor as well as other mediators (Chegini and Rao, 1988; Massey et al., 1991; Nishihira et al., 1984). Smooth muscle myometrial human tissue has been shown to be hyperresponsive to the action of these compounds at the end of pregnancy (Cruz et al., 1989; González et al., 1994; Tetta et al., 1986). Furthermore, histamine and serotonin not only contract myometrium but also potentiate each other (together with prostaglandin F₂) in evoking uterine contractions (Rudolph et al., 1992, 1993). In addition, these autacoids, as well as some cytokines secreted by mast cells like tumor necrosis factor-, may not only have an important role in the regulation of contractions but also participate in processes such as gap junction formation, glycogenolysis, collagenase synthesis or leukocyte recruitment, which are involved in uterine preparation for parturition (Czametski and Rosenbach, 1986; Gaboury et al., 1995; Garfield, 1994; Hunt et al., 1996; Wilcox et al., 1992; Zanglis et al., 1996).

Ethodin (Rivanol; 6,9-diamino-2-oxyethyl acridine lactate) has been described as an effective drug for inducing uterine contractions similar to a physiological labor in stillbirth with uterine inertia (Schubert and Cullberg, 1987). This is a pathological situation in which estrogen levels (which under normal conditions are driven by the C19 steroids synthesized by the living fetus) decrease in maternal plasma, but progesterone levels remain constant, supporting uterine quietness and suppressing labor. Ethodin overcomes this unfavorable situation (Schubert and Cullberg, 1987). The biochemical mechanisms involved in that effect are unknown. It has been found that when ethodin is infused, amniotic fluid levels of prostaglandin F₂, prostaglandin E₂, prostacyclin and thromboxane A₂ are increased (Ölund et al., 1980). The increase in the amniotic concentration of these phospholipid derivatives is gradual as labor progresses, similar to the pattern observed in a normal labor (Keirse et al., 1977).

Suzuki-Nishimura et al. (1995) recently described a number of polybasic amines that fit the pharmacological profile of allergic substances that evoke mast cell degranulation. A classic example is compound 48/80, largely known as a mast cell degranulator, which has been shown to contract rat myometrium (Tabb, 1994). Ethodin has also been defined as a polyamine that "irritates skin and mucous membranes and causes sneezing on inhalations (Windholz et al., 1976)." Therefore, it is important to analyze the possible role that ethodin may have in evoking uterine contractions through the activation of uterine mast cells.

This research was designed with the hope of deepening our understanding of the mechanisms involved in parturition through study of the actions of ethodin in evoking myometrium contractions. This led us to search for possible interactions between mast cells and smooth muscle in the uterus. We examined the contractile properties of ethodin in a primary culture of mouse myometrium smooth muscle cells free of mast cells, and we also evaluated the effect of the addition of mast cells to the culture medium in the presence of the drug.

	Methods

Top Abstract Introduction Methods Results Discussion References

Materials. Ethodin, serotonin, acrolein, collagenase I, compound 48/80, penicillin and percoll were obtained from Sigma Chemical (St. Louis, MO). Oxytocin and human chorionic gonadotropin were from Sandoz Laboratories (Santiago, Chile). Trypsin, albumin, FBS and RPMI 1640 (GIBCO, Grand Island, NY). DNase and anti-desmin were from Boehringer-Mannheim (Mannheim, Germany).

Animals. We used female albino mice (3-5 months) weighing 30 to 40 g were maintained under a 12-hr dark/light cycle in a temperature-controlled room (24-25°C) with free access to drinking water and laboratory chow.

Contraction studies in mouse uterine horns. Longitudinal pieces of uterine horns of mice in estrous were mounted vertically into an isolated organ bath containing oxygenated RBS at 35°C. After a stabilization period, ethodin (10 µM) or compound 48/80 (1 µg/ml) were added, and the evoked contractions were recorded on a Grass model 7 polygraph, as previously described (Rudolph et al., 1992). The RBS was composed of 153 mM. NaCl, 5.3 mM KCl, 1.13 mM MgCl₂, 24.9 mM NaHCO₃, 2.7 mM glucose and 1.2 mM ascorbic acid. It was continuously gassed with 95% O₂ and 5% CO₂ and adjusted to pH 7.4.

Preparation of dispersed cells. Myometrial smooth muscle cells were prepared for primary culture from uterine horns from virgin mice in estrous that were administered human chorionic gonadotropin (30 U/kg i.p.) 18 hr before the experiment. The method was similar to that of Boulet and Fortier (1987). After the uteri were removed and dissected free of fat, they were placed in ice-cold Hanks'-HEPES solution containing HEPES (0.47% w/v), penicillin (200 U/ml), streptomycin (200 µg/ml) and fungizone (10 µg/ml) and were slit longitudinally and transversally into small pieces (<10 mg). To remove endometrium, tissue was immersed in Hanks' solution containing trypsin (0.37% w/v), incubated for 90 min at 4°C and then incubated for a similar time at room temperature. After this procedure, the semidispersed tissue was washed with Hanks' solution and incubated at 37°C for 45 min in a Hanks' solution containing trypsin (0.03% w/v), collagenase I (0.03% w/v) and DNase (0.015% w/v) to remove the stroma. At the end of the second incubation and after the medium was washed, tissue was incubated at 37°C for 1 hr with a Hanks' solution containing trypsin (0.02% w/v), collagenase I (0.05% w/v), DNase (0.01% w/v) and EDTA (0.02% w/v). Digestion was stopped with 2 ml of 10% FBS and filtered out through 500-µm Nitex. Final suspension of the myometrial fraction was centrifuged at 300 × g for 10 min and washed out three times with Hanks' solution. The cells were counted in a hemocytometer, and viability (>85%) was determined by trypan blue exclusion. Each uterine horn yielded ~800,000 cells.

Cell culture. Dispersed cells were plated to a final concentration of 2 to 3 × 10⁶ cells/5 ml in RPMI 1640 containing glutamine (250 µg/ml), FBS (5% v/v), penicillin (100 U/ml), streptomycin (100 µg/ml) and fungizone (10 µg/ml). The cell suspension was seeded into 25-cm² flasks (Corning) and maintained for 16 hr at 37°C in a humidified air atmosphere containing 5% CO₂, which allowed fibroblasts to adhere to the culture flasks. Unattached muscle cells were transferred to 60-mm-diameter wells containing coverslips precoated with laminin (20 µg/ml) plus collagen IV (20 µg/ml) to maintain the cells in a contractile phenotype (Thyberg and Hultgardh-Nilsson, 1994). They were cultured for 70 hr under identical conditions and used for contractile activity before they became confluent. The presence of smooth muscle cells in the cell culture was evaluated by immunocytochemistry using a desmin monoclonal antibody and a peroxidase-conjugated goat anti-mouse immunoglobulin that showed a purity of >95%. The capacity of the cells to contract was determined by the addition of oxytocin (0.1 µM) and 5-hydroxytryptamine (10 µM).

Contraction studies in the smooth muscle cell primary culture. This was evaluated by measuring the decrease in the average length of a population of muscle cells exposed to a test compound as described previously (Bitar and Makhlouf, 1982). Briefly, the procedure was as follows. Cells were rinsed with PBS and incubated with either PBS (control) or PBS plus a test solution at 37°C for 30 sec. Then, they were fixed with acrolein (1% v/v), coverslips were rapidly extracted from the wells and photographs were taken randomly in successive microscopic fields with a goal of reaching a minimum of 100 cells per experiment by using a phase-contrast Carl Zeiss model G41-204-5 microscope. The procedure was repeated separately for each compound and/or experimental condition. The photomicrographs were developed, and lengths of cells were accurately measured considering their respective enlargements. Frequency histogram analysis of cell length distribution showed that this parameter was normally distributed (see figs. 3 and 4). Results of cellular contractions were presented as mean ± S.D. Data were subjected to one-way repeated-measures analysis of variance by using a statistical procedure with the software of program ORIGIN (MicoCal Software). Statistical difference was assessed with the Student's t test. Values of P < .05 were considered significantly different.

View larger version (K): [in this window] [in a new window]   Fig. 3.   Example of Gaussian curves developed from one of three different experiments in which contractile phenotype of the primary culture of myometrium smooth muscle cells was evaluated with the addition of 0.1 µM oxytocin (94 cells) () and 10 µM serotonin (117 cells) (X). , Control (138 cells). In the same experiment, 10 µM ethodin was ineffective for evoking contractions (+). The number in parentheses indicates the number of cells photographed and analyzed in each case. Inset, mean length ± S.D. in each case. * P < .01 compared with control ().

View larger version (K): [in this window] [in a new window]   Fig. 4.   Gaussian curves obtained from one of four different experiments showing the mast cell-mediated effect of ethodin on myometrium smooth muscle primary culture cells. X, 120,000 mast cells/ml (126 cells); , 10 µM ethodin (180 cells); , 120,000 mast cells/ml preincubated with 10 µM ethodin (155 cells) and *, 120,000 mast cells/ml previously degranulated by electrical stimulus (148 cells). The number in parentheses indicates the number of cells photographed and analyzed in each case. Inset, mean length ± S.D. in each case. * P < .01 compared with controls ( and X).

Isolation of mast cells. Uterine and peritoneal mast cells are known as connective tissue mast cells, sharing the characteristic of being sensitive to the action of estrogen (Cocchiara et al., 1992; Padilla et al., 1990). They were collected by washing the peritoneal cavity from female mice in diestrous with 2 ml of a solution containing 25 mM Tris · HCl, 120 mM NaCl, 5 mM KCl, 10 mM EDTA and 0.05% w/v BSA adjusted to pH 7.6 as described previously (Mousli et al., 1989). Cells were then subjected to two sequential gradient centrifugation steps through discontinuous gradients of percoll as described previously (MacGlashan and Guo, 1991). Cells were stained metachromatically with toluidine blue (0.1% w/v, pH 1.0) and quantified by using a Neubauer hemocytometer. Crude peritoneal cell suspensions contained 2.7% mast cells, and after gradient centrifugation they ranged between 60% to 75%. Purified mast cells were washed and maintained for a maximum of 30 min at 4°C (Takayama et al., 1994). The viability of the mast cells was determined by their ability to exclude trypan blue and by the measurement of histamine in the supernatant (Yamatodani et al., 1985). Nonspecific, spontaneous histamine release was always <6%.

Degranulation of mast cells. An aliquot of 2 ml of isolated and partially purified mast cells was incubated for 5 min at 37°C. Ethodin was added at a final concentration of 10 µM. The resulting suspension was poured into the culture flasks and processed as previously stated. To degranulate mast cells with an electrical stimulus, two platinum ring electrodes (distance between electrodes, 2 cm) were introduced into an aliquot of the mast cells suspension. An external field of supramaximal voltage was applied for 30 sec with biphasic pulses of 2-msec duration at a frequency of 30 Hz with a Grass model S48 stimulator (Cruz and Rudolph, 1986).

	Results

Top Abstract Introduction Methods Results Discussion References

Effect of ethodin in mouse uterine horns in vitro. Figure 1 depicts an example of the contractile effect evoked by ethodin (10 µM) or compound 48/80 (1 µg/ml) in an isolated organ bath preparation. As is shown, both drugs elicited uterine responses that consisted in a transient increase in contractile tension. The maximum uterine response was reached within 30 sec and declined to basal values at ~2 min. Tachyphylaxis to the response to ethodin and compound 48/80 was observed after three to five successive administrations of either drug to the isolated organ bath with an interval of 10 min, as shown in figure 1. Posterior addition of serotonin (10 µM) to the preparation evaluated the contractile capacity of smooth muscle cells (fig. 1). In addition, because the contractile response to ethodin was limited even when the drug was present in the bathing solution, we evaluated the capacity of this solution to evoke further contractions by changing it to a parallel organ bath preparation. As shown in figure 2, the preparation responded with a similar pharmacological contractile potency. Histochemistry of uterine horns incubated with ethodin (10 µM) showed a high number of degranulated mast cells in close proximity to the smooth muscle cells (data not shown).

View larger version (K): [in this window] [in a new window]   Fig. 1.   A. A representative recording of contractions elicited by 10 µM ethodin and 1 µg/ml compound 48/80 in isolated organ bath preparations of mouse uterine horns. B, Recording of one preparation showing the effect of successive additions of ethodin to the organ bath. Each one was applied after a washing period of 10 min. As shown, tachyphylaxis developed gradually after four or five successive stimuli. This was not due to muscle desensitization because tissue responded adequately to the addition of 10 µM serotonin (S).

View larger version (K): [in this window] [in a new window]   Fig. 2.   This was an experiment designed to evaluate the pharmacological contractile potency of 10 µM ethodin in two parallel isolated organ bath preparation of mouse uterine horns. A, The addition of ethodin evoked a contraction, but tissues returned to previous conditions even when the drug was present in the bathing solution. B, Evidence that the drug is active; this is a parallel isolated organ bath preparation in which the incubation solution of preparation A was added as indicated.

Contraction studies in the smooth muscle cells primary culture. The experimental protocol designed for these experiments provided us with a reproducible and reliable smooth muscle cell culture that allowed us to define how this cell type responds to ethodin. Morphometric measurements of the smooth muscle cells from the mouse myometrium preparations provided cells with lengths of 5 to 160 µm (figs. 3 and 4). The presence of the contractile phenotype as well as the presence of functional receptors for endogenous compounds was demonstrated by the administration of oxytocin (0.1 µM) or serotonin (10 µM). As shown from the curves in figure 3, both compounds elicited contractions of the smooth muscle cells. Under those experimental conditions, ethodin did not evoke contractions (fig. 3).

Effect of a purified fraction of mast cells on contractions evoked in a primary smooth muscle cell culture. More than 80% of the purified fraction of mast cells degranulated when incubated with ethodin 10 µM at 37°C or when stimulated with an electrical stimulus, as assessed through microscopic observation of toluidine blue-stained aliquots of mast cells. The addition of the degranulated mast cells to the primary culture of smooth myocytes did not evoke a significant shortening of the cells unless a minimum of 64,000 mast cells/ml was added (table 1).

	Discussion

Top Abstract Introduction Methods Results Discussion References

These studies were performed to determine the nature of ethodin interaction with smooth muscle myometrium cells to obtain a better understanding of the mechanisms regulating labor. As shown in figures 1 and 2, and as it has been previously reported, ethodin (10 µM) evoked a characteristic pattern of contractions and the release of histamine in isometrically suspended mice uterine horns. Both effects were inhibited by the previous addition of rithodrine, indomethacin or the mast cell stabilizers ketotifen and sodium chromolyn (Rudolph et al., 1997). The same pattern of contractions was observed with the addition of compound 48/80 to the isolated organ bath (fig. 1) or of ovalbumin to an isolated preparation of uterine horns of ovalbumin sensitized mice.³ Therefore, we hypothesized that the contractile effect of ethodin on smooth muscle cells may require a mechanism in which uterine mast cell activation should be involved.

The method followed to obtain the primary culture of smooth muscle myometrium cells allowed us to purify cells that maintained their contractile phenotype. As already described, 2- to 3-day-old cultures grown over laminin and collagen IV were appropriate to study the cells (Hedin et al., 1988; Kühl et al., 1986). At that stage, >95% of the cells demonstrated positive immunohistochemistry staining for the presence of smooth muscle specific desmin. Contractions evoked by the addition of oxytocin and serotonin showed that the smooth muscle cells expressed a contractile phenotype. Nevertheless, the cells did not contract by the action of ethodin, suggesting that this compound may not have a direct contractile effect on smooth muscle cells (fig. 3). The evidence that mast cells may be mediating the contractile action of ethodin is shown in figure 4, in which an aliquot of ethodin-degranulated mast cells (minimum, 64,000/ml; table 1) was added to the culture medium. Furthermore, the sole requirement of degranulation for evoking smooth muscle cell contraction was demonstrated when mast cells being degranulated by the electrical stimulus evoked the shortening of muscle cells (fig. 4).

Our results represent pharmacological evidence that the contractile action of ethodin is mediated by the activation of uterine mast cells. However, uterine contractions evoked by the administration of ethodin have different evolution in vivo or in vitro conditions. In the first situation, contractions evolve toward labor and expulsion of the conceptus (Ölund et al., 1980, Schubert and Cullberg, 1987). However, in the in vitro preparation, uterine contractions evoked by ethodin were limited, lasting 2 to 3 min (fig. 1); then the myometrium returned to the preestimulation condition despite the presence of ethodin (which was neither metabolized nor reduced in its pharmacological contractile activity, as shown in fig. 2), demonstrating the presence of an autoregulatory process in the uterine mast cell. The above is also experimental evidence that mast cells are necessary but not sufficient for the uterotonic and labor inducer effect of ethodin in vivo. Labor may be initiated by mast cell activation, but this process should be followed by the sequence of events derived from leukocyte recruitment to convert the uterus into an active and reactive organ.

The proposal that mast cells may be mediating estrogen-regulated uterine changes is derived from biochemical and endocrine evidence reported >20 years ago (Spaziani, 1975). Additional evidence has been incorporated during the past years that support that hypothesis. It has been confirmed that uterine mast cells share, together with peritoneal and other genitourinary tissue mast cells, the special characteristic of being sensitive to the action of estrogen (Cocchiara et al., 1992; Padilla et al., 1990; Pang et al., 1995). Furthermore, as it has been previously analyzed, uterine mast cells may represent a significant source of histamine, serotonin, cytokines, phospholipid derivatives and enzymes that may be important as contractile and proinflammatory mediators for parturition. However, it would be misleading to analyze the production and possible actions of these compounds without emphasizing that the uterine mast cell represents just one among many potential sources of these biologically active compounds. It is our concept that as occurs in certain allergic and inflammatory responses, the mast cell may represent the most significant initial source of these compounds. Then, as these reactions evolve, additional sources of autacoids, lipid derivatives, cytokines and enzymes may be activated and/or recruited. For example, among the sources for additional mast cell mediators are the eosinophils and neutrophils. They may be recruited into the myometrium as a response to mast cell activation, which may release histamine, cytokines (like interleukin-4 and tumor necrosis factor-) or lipid derivatives (like leukotriene C₄ or platelet-activating factor) (Costa et al., 1994; Czametski and Rosenbach, 1986; Gaboury et al., 1995; Lukacs et al., 1995). Alternatively, leukocytes recruited to sites of mast cell activation may represent sources of other compounds not produced by the mast cell itself that may be important for parturition or postpartum uterine involution (Perez et al., 1996; Wilcox et al., 1992; Wolpe et al., 1988). It is also important to note that mast cells could also provide a significant source of arachidonic acid by secreting a stored phospholipase A₂ (Reddy and Herschman, 1996). In addition, histamine and mast cell constitutive cytokines: tumor necrosis factor- and interleukins-6 and -4 are also known to stimulate arachidonic acid metabolism in chorionic membranes (Adamson et al., 1994; Romero et al., 1992; Santhanam et al., 1991), which may represent the source of the prostaglandins, thromboxanes and leukotrienes detected in amniotic fluid when labor was evoked by ethodin in vivo (Ölund et al., 1980).

It is known that allergic reactions (Klein et al., 1984), intra-amniotic infections (Minkoff, 1983) and mastocytosis (Donahue et al., 1995) are clinical situations in which mast cell degranulation evoke myometrial contractions that could complicate pregnancy. Our hypothesis is that mast cell activation may represent at least one of the avenues taken by estrogens for remodeling myometrium leading to parturition. Many questions still need to be answered before this hypothesis can be established, rejected or complemented. Nevertheless, our research should serve to promote further investigation in this field.