Nitric oxide in the contractile action of bradykinin, oxytocin, and prostaglandin F₂ α in the estrogenized rat uterus

Abstract

Experiments were performed on uteri from estrogen-primed female rats. Bradykinin (BK) (10⁻⁸ M) significantly augmented biosynthesis of prostaglandin F₂ α (PGF₂α) and prostaglandin E₂ (PGE₂), and this synthesis was completely blocked by N^G-monomethyl l-arginine (NMMA) (300 μM), a competitive inhibitor of nitric oxide synthase (NOS). Blockade of prostaglandin synthesis by indomethacin caused rapid dissipation of isometric developed tension (IDT) induced by BK. Blockade of NOS with NMMA had similar but less marked effects. Combining the two inhibitors produced an even more rapid decay in IDT, suggesting that BK-induced NO release maintains IDT by release of prostanoids. The decline of frequency of contraction (FC) was not significantly altered by either indomethacin or NMMA but was markedly accelerated by combination of the inhibitors, which suggests that PGs maintain FC and therefore FC decline is accelerated only when PG production is blocked completely by combination of the two inhibitors of PG synthesis. The increase in IDT induced by oxytocin was unaltered by indomethacin, NMMA or their combination indicating that neither NO nor PGs are involved in the contractions induced by oxytocin. However, the decline in FC with time was significantly reduced by the inhibitor of NOS, NMMA, suggesting that FC decay following oxytocin is caused by NO released by the contractile process. In the case of PGF₂α, NMMA resulted in increased initial IDT and FC. The decline in FC was rapid and dramatically inhibited by NMMA. Receptor-mediated contraction by BK, oxytocin, and PGF₂α is modulated by NO that maintains IDT by releasing PGs but reduces IDT and FC via cyclic GMP.

Nitric oxide (NO), a soluble gas, is synthesized by endothelial NO synthase (NOS), a constitutive enzyme that converts arginine to the free radical NO plus citrulline in the presence of NADPH and other cofactors. Endothelial (NOS) is a constitutive enzyme in vascular endothelial cells, it is activated by cholinergic stimulation. The NO produced diffuses to overlying vascular smooth muscle and activates soluble guanylate cyclase with the subsequent formation of cyclic GMP (cGMP) (1). NO interacts with the heme group in soluble guanylate cyclase altering its conformation, thereby activating the enzyme that converts guanine triphosphate into cGMP. cGMP then relaxes the overlying smooth muscle (2).

Another constitutive isoform of NOS, neural (n) NOS, has been found in various parts of the central and peripheral nervous systems where it acts as a gaseous neurotransmitter (3, 4). In the hypothalamus, it activates the release of luteinizing hormone-releasing hormone that induces sexual behavior on the one hand, and release of luteinizing hormone on the other, both of which are required for reproduction (5, 6).

In the periphery, terminals of NOergic neurons are also present in the corpora cavernosa penis and release of NO from these terminals activates guanylate cyclase in the smooth muscle causing relaxation, a requirement for penile erection (7). In uteri from rats primed with estrogen, NOergic terminals in the uterine smooth muscle activate cyclooxygenase (COX), producing various prostanoids (8, 9). These induce contraction of the uterus. On the other hand, the contractility of the estrogenized uterus spontaneously decreases during in vitro incubation. Blockade of NOS with N^G-monomethyl l-arginine (NMMA), a specific inhibitor of NOS, instead of speeding the relaxation of the uterus, as would be expected by the inhibition of prostaglandin (PG) release, decreased the rate of relaxation. We hypothesized that the release of NO in the uterus activated guanylate cyclase and led to production of cGMP, which as in vascular smooth muscle, also relaxed uterine smooth muscle. We concluded that NO has complex opposing actions on uterine contractility, relaxing uterine smooth muscle by inducing release of cGMP and contracting it via stimulation of synthesis and release of PGs by activation of COX. The present work was designed to clarify this paradoxical double action of NO by exploring the effects of three compounds which cause contraction of the uterine smooth muscle, bradykinin (BK), oxytocin (OX), and PGF₂α.

MATERIALS AND METHODS

Female rats of the Wistar strain (200–230 g body weight) were used. They were housed in group cages under controlled conditions of light (lights on 500-1700 h) and temperature (23–25°C). Rat chow and water were freely available. The rats were treated with estrogen by injecting 1 μg of 17β-estradiol in 0.2 ml of ethanol (30%) subcutaneously. Control animals were injected with an equal volume of the solvent solution only. Twenty-four hours later, the animals were killed and their uterine horns were removed.

Metabolism of [¹⁴C]Arachidonic Acid Once the uterine horns were obtained, each horn was opened, trimmed of visible fat, and placed in a Petri dish containing a modified Krebs–Ringer bicarbonate (KRB) solution with glucose (11 mM).

The metabolism of exogenous arachidonic acid by rat uterine tissue was determined by incubating the tissue for 60 min in KRB medium containing 0.25 μCi of [¹⁴C]arachidonic acid (52.9 Ci/mol; l Ci = 37 GBq) (New England Nuclear) in an atmosphere of 95% O₂/5% CO₂ with constant shaking at 60 cycles per min at 37°C. For each determination, ≈200 mg of uterine tissue was used. The uterine strips were randomly treated with BK (10⁻⁸ M) or NMMA (300 μM). These reagents were added to the incubation medium alone or in combination as described in Results. The above-mentioned drugs were obtained from Sigma. The controls were incubated in medium alone. At the end of the incubation period, the incubation medium was acidified to pH 3.0 with 1.0 M HCl in 1 volume of ethyl acetate and extracted twice for PGs. Pooled ethyl acetate extracts were dried under nitrogen. The residues were suspended in chloroform/methanol and applied to silica gel TLC plates. The plates were developed in a solvent system of benzene/dioxane/glacial acetic acid (60:30:3; vol/vol.). The position of the authentic eicosanoids was visualized by spraying the dried plates with 10% phosphomolybdic acid in ethanol followed by heating at 110°C for 10 min. Average R_f values were 0.30 for 6-keto-PGF₁α, 0.35 for PGF₂α, 0.47 for PGE₂, 0.75 for thromboxane B₂, and 0.8 for arachidonic acid. Radioactivity from TLC plates for different prostanoids was measured by liquid scintillation counting. The area of each of the radioactive peaks corresponding to authentic prostanoids was calculated and expressed as a percentage of the total radioactivity of the plates.

Motility Studies.

Uterine horns were obtained as described above. The tissue was removed and each horn was divided by a transverse cut into two segments of an equal length. The exposed segments were placed in Petri dishes containing KRB at room temperature and constantly gassed with 95% O₂/5% CO₂. Each segment was immediately opened by a cut along the mesosalphinx insertion; one end was attached to a glass holder and immersed in a tissue chamber filled with 20 ml of KRB (pH 7.4, 37°C) and continuously gassed. The other end of the tissue was attached to a strain gauge coupled to an amplifier driving a direct writing oscillograph. After a resting tension of 1 g was applied to each strip by means of a micrometric device, isometric developed tension (IDT) and frequency of contractions (FC) were measured. IDT values (expressed in mg) were obtained by measuring the mean amplitude of all the contractions recorded over a 10-min period. FC values were obtained as the mean number of the contractile cycles observed during the same period.

All uterine strips were preincubated in the chambers without tension in the presence of NMMA (300 μM), indomethacin (1 μM) alone, or in combination, or an equal volume of medium alone for 30 min (Control).

After preincubation, either BK (10⁻⁸ M), Ox (0.25 milliunits⋅ml⁻¹), or PGF₂α (10⁻⁷ M) were added. Uterine stimulation induced by these agents was recorded for 60 min.

In each experiment, one strip from one half horn was used as control, selected randomly, and the other half was submitted to experimental treatment.

Determination of NOS Activity.

The content of NOS in the tissue at the end of the experiments was estimated by the modified citrulline method (10). Because arginine is converted by NOS in the presence of O₂ and cofactors into equimolar concentrations of citrulline that remains in the tissue and NO that rapidly degrades, measurement of [¹⁴C]citrulline produced from [¹⁴C]arginine added to the homogenized muscle provides a convenient measure of the activity of NOS.

RESULTS

Role of NO in Prostanoid Synthesis Induced by BK.

Our previous work had shown the role of NO in uterine prostanoid synthesis in vitro (8). Therefore, we wished to determine if BK also acts via NO to stimulate prostanoid synthesis. In the present study, BK (10⁻⁸ M) significantly augmented the biosynthesis of PGF₂α and PGE₂ from estrogenized uteri incubated in vitro. Although there were significant increases in 6-keto-PGF₁α and thromboxane B₂ synthesis, as determined by t test versus values in the controls, these changes were not significant statistically by analysis of variance (Fig. 1). To determine the role of NO in the stimulation of prostanoid synthesis by BK, the tissues were preincubated with NMMA (300 μM), a competitive inhibitor of NOS, at a concentration which we found effective in previous experiments. NMMA blocked the stimulation of PGF₂α and PGE₂ induced by BK and had no significant effect on basal release of any of the prostanoids.

Figure 1.

Effect of NMMA (300 μM) and BK (10⁻⁸ M) on metabolism of [¹⁴C]arachidonic acid to various COX metabolites. Each column represents the mean of five preparations from different animals. ∗, P < 0.01 versus control.

Role of NO in BK-Induced Contractions of Uterine Muscle.

BK induced a highly significant IDT that decayed slowly over time, such that it was decreased by ≈15% from the initial maximal contraction by 50 min (Figs. 2 and 3A). To determine the role of PGs in the initiation and maintenance of the contraction induced by BK, the tissue was preincubated with indomethacin to block COX before BK was added. Under these conditions, the initial IDT was unchanged (Table 1) but dissipated significantly faster and reached resting values by 50 min (Fig. 3A). When NO formation was blocked by the inhibitor of NO synthesis, NMMA, results were similar but slightly more pronounced than those obtained with indomethacin, with a more rapid extinction of the IDT than with indomethacin that had decayed around 80% by 50 min. When the uteri were preincubated with both NMMA plus indomethacin, the initial IDT was unaltered but a much more pronounced inhibition of the IDT activity induced by BK was found, such that a 65% decay was reached by 20 min.

Figure 2.

Effect of 30-min preincubation with NMMA (300 μM) and indomethacin (INDO) (10⁻⁶ M) compared with control on typical recording of BK (10⁻⁸ M) induced contractions of uterine strips from an estrogen-treated rat.

Figure 3.

Effect of 30 min preincubation of NMMA (300 μM) and indomethacin (INDO) (10⁻⁶ M) compared with control on the time-course of BK (10⁻⁸ M) induced IDT (A) and FC (B). Points are means of six uterine strips. ∗, P < 0.05; ∗∗∗, P < 0.001 versus control values.

Table 1.

Initial IDT and FC induced by BK, OX, and PGF₂α

	IDT, mg	FC, no. of contractions/10 min
BK
BK	80  ±  9	13  ±  1
BK + NMMA	80  ±  8	13  ±  1
BK + INDO	86  ±  10	13  ±  2
BK + NMMA + INDO	79  ±  8	10  ±  1
OX
OX	55  ±  5	5  ±  0.5
OX + NMMA	64  ±  4	5  ±  1.0
OX + INDO	60  ±  3	6  ±  0.4
OX + NMMA + INDO	62  ±  5	5  ±  0.4
PGF2α
PGF2α	67  ±  7	5  ±  0.5
PGF2α + NMMA	90  ±  4*	8  ±  0.3**

Data show the effect of 30 min preincubation with NMMA (300 μm) or indomethacin (INDO; 10⁻⁶ m) or both together on spontaneous initial IDT and FC induced by Bk, PGF₂α, or OX. Results are the mean ±SEM of six uterine strips. ∗, P < 0.05; ∗∗, P < 0.01 vs. control. 

The FC was less stable, decaying around 65% by 50 min. The initial FC induced by BK was similar in all treatment groups (Table 1), and the decrement of the FC, was only slightly increased by either NMMA or indomethacin alone, but was much greater in the presence of indomethacin plus NMMA, such that FC was essentially zero within 20 min (Fig. 3B).

Role of NO in OX-Induced Contraction of Uterine Muscles.

OX (0.25 milliunits⋅ml⁻¹) induced a large and sustained contractile activity in estrogenized uteri, characterized by an IDT that did not decrease at 50 min (Fig. 4A). Although the initial IDT induced by OX was slightly less than that induced by BK, the initial FC was significantly lower than that induced by BK (Table 1).

Figure 4.

Effect of 30 min preincubation of NMMA (300 μM) and indomethacin (INDO) (10⁻⁶ M) compared with control on the time course of OX (0.25 milliunits⋅ml⁻¹) induced IDT (A) and FC (B). Points are means of six uterine strips. ∗, P < 0.05; ∗∗, P < 0.01 versus control values.

To determine the importance of PGs in the maintenance of the contractions induced by OX, the tissue was preincubated with indomethacin (10⁻⁶ M) before incubation with OX. Contrary to the results obtained with BK, the inhibitor of COX did not modify either the initial IDT (Table 1) or the stability of IDT at 50 min, meaning that OX in this preparation is not acting through the synthesis of PGs. To determine if NO contributes to OX-induced contractions, the tissue was preincubated with NMMA, a specific NOS inhibitor, for 30 min before the addition of OX. Neither NMMA nor indomethacin modified the IDT induced by OX during the 50 min observation period (Table 1 and Fig. 4A).

The FC decay was 65% at 50 min with OX-treated uteri, similar to the decay in BK-treated uteri (Fig. 4B). On the other hand, the initial FC (Table 1) and its decay with time were also not modified by indomethacin, but significantly reduced by NMMA. The addition of NMMA in combination with indomethacin did not alter the action of NMMA on FC.

Role of NO in PGF₂α-Induced Contraction of Uterine Muscle.

PGF₂α evoked a marked contraction of the estrogenized uterus characterized by an IDT that only decayed by 15% at 50 min. (Fig. 5A). The initial IDT induced by PGF₂α was similar to that induced by BK and OX (Table 1), but the FC was significantly lower (P < 0.001) than that induced by BK and similar to that induced by OX. The decrement in IDT was similar to that obtained with BK.

Figure 5.

Effect of 30 min preincubation of NMMA (300 μM) on time course of PGF₂α (10⁻⁷ M) induced IDT (A) and FC (B). Points are means of six uterine strips. ∗∗∗, P < 0.001.

To determine the influence of NO in the contractions induced by PGF₂α, the tissues were preincubated with the NOS inhibitor, NMMA, for 30 min before the addition of PGF₂α. NMMA increased the initial IDT produced by PGF₂α (Table 1). Because NO induces PGF₂α release, and we added exogenous PGF₂α, we expected that NMMA would cause a more rapid decline in PGF₂α-induced contractions. In contrast to the results with BK and similarly to OX, NMMA did not change IDT stability with time, which at 50 min was not significantly different from that of the controls.

With PGF₂α-induced activity, the FC declined 90% at 50 min, whereas with BK it was around 65% at the same time (Fig. 5B). The FC in the presence of NMMA only decreased 20% by the termination of the experiment instead of 90% in the uteri treated with PGF₂α. This result means that NO decreases FC in PGF₂α-induced activity, as well as in OX-induced activity.

Effects of BK, OX, and PGF₂α on NOS Activity.

All three of the compounds that induced uterine concentrations increased NOS activity measured at the end of the experiment by the citrulline method (Fig. 6). The increase induced by BK was significantly greater than that induced by OX and PGF₂α.

Figure 6.

Effect of BK, OX, and PGF₂α on NOS activity determined at the end of the experiment by formation of [¹⁴C] citrulline from [¹⁴C]arginine added to the homogenized tissue. Height of the columns equals mean ± SEM. ∗∗∗, P < 0.001 versus control; ∗∗, P < 0.01 versus control. NOS activity induced by OX and PGF₂α were similar and significantly less (P < 0.05) than that induced by BK.

DISCUSSION

Because inhibition of NOS by NMMA did not alter the initial FC and IDT induced by BK, OX, and PGF₂α, the most important point in these studies is that BK, OX, and PGF₂α act on uterine muscle to initiate IDT and FC, independent of NO that plays an important but differing role in maintaining IDT and FC, depending on the excitatory compound, as described below.

It is well known that BK stimulates PLA₂ in several tissues (11, 12) and consequently increases the production of arachidonate by hydrolysis of membrane phospholipids. This stimulation, in the presence of activated COX, leads to prostanoid biosynthesis (13, 14). Some studies also indicate that BK activates COX independently of phospholipase activity (15). We previously observed that in isolated uterine strips from ovariectomized rats, BK stimulates PGE₂ synthesis and release from endogenous arachidonic acid (16). The results obtained here with BK indicate that it stimulates the synthesis of both PGF₂α and PGE₂ from incubated estrogenized uterine muscle when labeled arachidonic acid is added to the incubation medium, indicating a stimulation of COX. This stimulation of COX is brought about by NO, because the inhibitor of NOS, NMMA, completely blocked the stimulation of PG synthesis and because BK increased the content of NOS in the tissue measured by the citrulline method.

To determine the role of PGs in initiation and maintenance of the contraction induced by BK the tissue was preincubated with indomethacin and, under this condition, the initial IDT was unchanged but the IDT dissipated significantly faster. This finding indicates that maintenance of IDT is due largely to production of PGs induced by BK. When NO formation was blocked by NMMA, the results were similar to those obtained with the inhibitor of COX because NMMA blocked NO release and consequent activation of COX and PG release. Moreover, when the uteri were preincubated with indomethacin plus NMMA, a much more pronounced reduction of the IDT induced by BK was found, probably because COX activity was partially inhibited by indomethacin and when activation by NO was decreased by NMMA, the inhibition was more complete. These results support the hypothesis that maintenance of contractions induced by BK results from NO-induced PG synthesis and release. The action of BK in the uteri is analogous to its action in the vascular system where BK acts on endothelial cells to increase intracellular Ca²⁺ concentration [Ca²⁺]_i which combines with calmodulin to activate NOS (17). The FC in BK-treated uteri was less stable than the IDT, but was little affected by either indomethacin or NMMA; however, when both were present, FC decayed rapidly. This result would suggest that FC is less sensitive to decreased COX formation than IDT, but is maintained by PGs. Bradykinin may induce contraction by activation of its receptors on the NOergic neuronal terminals and in the uterine smooth muscle itself increasing their [Ca²⁺]_i, which combines with calmodulin to activate NOS. The NO released activates COX in the muscle and generates prostanoids that maintain contraction as described (8).

In the case of OX, the inhibition of NOS by NMMA did not modify the initial IDT or its spontaneous decay over time. OX-induced contractions were also not affected by blockade of COX with indomethacin or incubation with both drugs. Thus, OX must combine with its receptors and stimulate contraction by an independent mechanism, perhaps by increasing [Ca²⁺]_i. Indeed, OX inhibits Ca²⁺–Mg²⁺ ATPase activity in the plasma membrane fraction of myometrium (18), thereby inhibiting Ca²⁺ influx and causing increase in [Ca²⁺]_i. Furthermore OX also stimulates the production of inositol triphosphate (19, 20) increasing release of Ca²⁺ from the sarcoplasmic reticulum. This maintained increase in [Ca²⁺]_i maintains the contractile state.

Regarding the initial IDT and FC induced by PGF₂α, it is interesting to comment that both parameters increased significantly after preincubation with NMMA, probably because it decreased cGMP formation which was relaxing the muscle (Fig. 7). In this case, the negative influence of NO in the initial IDT and FC was shown. In previous work, measuring the spontaneous activity of uterine strips from estrogenized rats (8), we demonstrated that IDT stability with time increased when the tissue was preincubated with NMMA. We speculated then, that NMMA had blocked the production of NO thereby blocking its effects on cGMP production, an agent that relaxes smooth muscle (21). The same interpretation can be applied here. The continued IDT induced by PGF₂α was unrelated to NO and further PGF₂α release because it was unaffected by either NMMA or indomethacin.

Figure 7.

Diagram of the mechanism by which BK, PGs, and oxytocin (OT) maintain frequency and force of contraction in estrogen-treated uterine muscle. For details, see text.

On the contrary, the decline in FC induced by these two agents was significantly inhibited by NMMA. These results indicate that NO is inhibiting FC in OX and PGF₂α-induced activity. With PGF₂α- and OX-stimulated uterine contraction, NO appears to decrease mainly the FC stability over time without affecting the stability of IDT. It is known that slow depolarization’s lead into spike discharges and the frequency of discharge of this pacemaker sets the frequency of contractions of the uteri (22). The ionic mechanisms underlying the generation of the slow depolarization’s in the myometrium are not well understood, but it was suggested that they result from an increase in membrane permeability involving sodium (23). There is Na⁺–Ca²⁺ exchange in the membrane and calcium influx favors Na⁺ extrusion (24). Thus, decreased Ca²⁺ influx caused by modifications in cGMP concentrations induced by activation of soluble guanylate cyclase by NO could induce membrane hyperpolarization that would cause the decrease in FC observed. Recent electrophysiological studies in smooth muscle suggest that Ca²⁺-activated K⁺ channels mediate hyperpolarizations by NO and cGMP (25–27). Release of smaller amounts of NO secondary to muscle contraction induced by OX or PGF₂α would release cGMP and this nucleotide by altering [Ca²⁺]_i, concentration in the myometrium pace maker cells would cause hyperpolarization and decrease the FC. Thus, in the case of OX and PGF₂α, the smaller quantities of NO released would still play a role in decreasing FC with time. The present findings are also consistent with earlier observations (21) indicating that the frequency but not the force of contractions of OX-stimulated rat uteri could be depressed by Br-cGMP. In fact, activation of NOS as determined by the citrulline method was greatest by BK and smaller in the case of OX- and PGF₂α-induced uterine contraction. We cannot discard the possibility that NO may act by another unknown mechanism to cause decay of FC.

Summarizing, the endogenous production of NO affects differently the IDT and FC decay over time—i.e., if the contractile activity was induced by BK that stimulates NOS directly, the NO stimulates the synthesis and release of PGE₂ and PGF₂α and these agents maintain IDT. On the other hand, if the contractile activity is induced by PGF₂α itself, or by OX, which do not require NO to maintain contraction, the smaller increase in NO production induced by the contraction process itself has no effect on IDT but decreases FC via release of cGMP.

ABBREVIATIONS

NOS

NO synthase

cGMP

cyclic GMP

NMMA

N^G-monomethyl l-arginine

PG

prostaglandin

BK

bradykinin

OX

oxytocin

KRB

Krebs–Ringer bicarbonate

IDT

isometric developed tension

FC

frequency of contractions

COX

cyclooxygenase

[Ca²⁺]_i, intracellular Ca²⁺ concentration.