The role of the sarcoplasmic reticulum in neonatal uterine smooth muscle: enhanced role compared to adult rat

Abstract

Little is known about contractile activity, response to agonists or excitation-contraction coupling in neonatal smooth muscle. We have therefore investigated 10-day rat uterus to better understand these processes, and compared it to adult uterus to elucidate how control of contractility develops. Spontaneous contractions are present in the 10-day neonatal uterus, although they are not as large or as regular as those present in adult tissues. External Ca²⁺ entry via L-type Ca²⁺ channels is the sole source of Ca²⁺ and is essential for the spontaneous activity. The neonatal uterus was responsive to carbachol or prostaglandin F_2α application; it showed a marked stimulation and a clear dissociation between the force and Ca²⁺ changes. Such sensitization was not apparent in adult rat myometrium. The sarcoplasmic reticulum (SR) had more releasable Ca²⁺ and contributed more to the response to agonists in neonatal compared to adult tissues. Thus, Ca²⁺ entry as opposed to SR Ca²⁺ release contributed much less to the uterine response to agonists in the neonatal, compared to adult tissues. Inhibition of the SR by cyclopiazonic acid also caused a more vigorous increase in Ca²⁺ and contractile activity, particularly frequency, in the neonatal compared to the adult uterus. Taken together these data suggest that: (1) spontaneous activity is already present by day 10, (2) receptor-coupling and excitation-contraction signalling pathways are functional, (3) the SR and Ca²⁺ sensitization mechanisms play a more prominent role in the neonate, and (4) there is a shift to a greater reliance on Ca²⁺ entry and excitability with development of the myometrium.

Our understanding of the processes controlling and producing contractions in smooth muscle is growing, but is still far from complete. One area of focus concerns the role of the intracellular Ca²⁺ store within the myocytes, the sarcoplasmic reticulum (SR) (Wray, 2002). It was initially anticipated, by extrapolation from studies on striated muscles, that the SR would release Ca²⁺, in response to Ca²⁺ itself or IP₃, and augment the contractile process. This role of the SR was, however, seriously questioned when work on first rat (Taggart & Wray, 1998) and then human (Tribe, 2001; Kupittayanant et al. 2002) uterine smooth muscle showed that both spontaneous force production and Ca²⁺ transients were increased when the SR was inhibited. This has led to the suggestion that the SR has a role in limiting contraction. The mechanism appears to be due in part to the SR releasing Ca²⁺ and activating K⁺ channels, causing hyperpolarization of the membrane and relaxation, as has been shown to be the case in vascular smooth muscle (Brenner et al. 2000). The uterine SR contains both IP₃ and ryanodine receptors (Martin et al. 1999) and agonists have been shown to be able to release Ca²⁺ from the SR and produce small increases in force, in the absence of external Ca²⁺ (Taggart & Wray, 1998; Luckas & Wray, 2000). Thus in the uterus, the role of the SR in physiological conditions is not fully understood. It may indeed change during pregnancy or labour, switching from being inhibitory to stimulatory. It has, for example, been reported that Ca²⁺-ATPase expression is increased in labouring women compared to non-labouring women (Tribe, 2001). To better understand the role and importance of the SR our approach here has been to study neonatal uterus, as this will represent a state where there is no pro-gestational influence and the SR will reflect the uterus at its least contractile. It is hypothesized that this is background activity, which will be altered with pregnancy and labour.

Relatively little is known for any smooth muscle about the contribution of the SR to contraction in neonatal animals, and what is known does not present a consistent pattern. Thus, in comparing the contribution of the SR or external Ca²⁺ entry to agonist-evoked contractions in neonatal and adult tissues, relatively more dependence on the SR was found in some (Hillemeier et al. 1991; Paul et al. 1994; Nakanishi et al. 1997), but not all (Hillemeier et al. 1991; Zderic et al. 1995; Akopov et al. 1998) tissues. Only one of the above studies measured intracellular [Ca²⁺] ([Ca²⁺]_i) (Akopov et al. 1998) and therefore it is unclear which mechanisms were being affected. We can find no data concerning this or excitation-contraction coupling in neonatal uterus for any species. Indeed there have been no studies of any aspect of excitation-contraction coupling in neonatal uterus. We have therefore also studied the relative contributions of SR and extracellular Ca²⁺ to contraction in this tissue, as well as the influence of agonists on these processes.

The aims of this paper were therefore to investigate the role of the SR in neonatal uterus and compare it with data obtained in adults. We have done this by simultaneously recording force and intracellular [Ca²⁺]_i in rat myometrium (1) in the presence or absence of a functioning SR, (2) with and without external Ca²⁺ present in the bathing solution, and (3) in the presence or absence of an agonist. We find significant differences between adult and neonatal tissue, which suggest a greater dependence on the SR in the neonate.

Methods

Small strips (5-10 mm length), of uterine horn (1 mm width, < 1 mm depth) were dissected from 2- and ‘10-day-old’ (range 8- to 12-day-old) neonatal Wistar rat pups that had been humanely killed by cervical dislocation. Some experiments were performed on similar sized strips of longitudinal myometria dissected from virgin adult Wistar rats, killed humanely by cervical dislocation following stunning. All procedures were carried out according to the UK Animals (Scientific Procedures) Act 1986.

Strips were incubated for 3-4 h at room temperature (22-23 °C) in oxygenated physiological solution (composition (mm): NaCl 154, KCl 5.4, MgSO₄ 1.2, glucose 12, CaCl₂ 2, Hepes 10; pH 7.4 with NaOH) with 12 μm of the membrane-permeant form of the Ca²⁺-sensitive dye Indo-1 (Molecular Probes) and then refrigerated overnight (4 °C) in Krebs solution. Preliminary control experiments showed that these procedures did not alter contractile activity, compared to that recorded in freshly isolated tissues.

Strips were warmed to room temperature for 1-2 h and then mounted in a small chamber (300 μl) on a Nikon Diaphot inverted microscope with a ×10 fluor objective. In all tissues, the fluor objective was focused on the longitudinal myometrial fibres. A metal hook was attached to either end of the strip with one hook fixed to a tension transducer (Grass FT 03). Neonatal strips were stretched to 0.2-0.5 mN force; adult strips to 2-5 mN force. The chamber was superfused with pre-heated, oxygenated physiological solution at a flow rate of 3 ml min⁻¹. Experiments examining the influence of SR function on spontaneous contractions were performed at 37 °C; all other experiments were performed at 30 °C to improve the retention of the Indo-1 signal. Tissue was excited via a xenon arc lamp (75 W) with light of wavelength 340 nm. The light emitted at 400 nm and 500 nm was recorded via photo multiplier tubes and digitally recorded (10 Hz). The ratio of the shifting in opposite directions of 400 nm:500 nm emissions was used as an indicator of [Ca²⁺]_i (Taggart et al. 1997).

At the start of each experiment, tissues were equilibrated in oxygenated physiological solution for at least 40 min. In order to compare experiments, agonist-induced changes in the amplitude of Ca²⁺ and force are described as percentage of control. The control is the maximum amplitude of Ca²⁺ and force induced by a 1 min application of a 40 mm K⁺-depolarizing solution (isosmotic replacement of NaCl) at the start of each experiment. The effects of agonists on SR releasable Ca²⁺ were investigated in Ca²⁺-free solutions with 1 mm EGTA to chelate any contaminating extracellular Ca²⁺, apart from when comparison with adult tissue was being made, to avoid any differential effect of EGTA on membrane permeability. The duration of the force and Ca²⁺ changes was measured by extrapolating the slope obtained during the control period, and calculating the time for the responses to return to this line. The effects of agonist in the neonate were dose dependent, but the myometrial samples were stimulated with supramaximal doses of carbachol (100 μm), prostaglandin F_2α (1 μm) or oxytocin (100 nm) for 1 min, as mentioned in the text, to ensure similar stimulation to both adult and neonatal tissues. The responses to carbachol were abolished by atropine. Nifedipine (used to inhibit L-type Ca²⁺ entry) was dissolved in ethanol and added from a 10 mm stock to make a final concentration of 0.1 μm. Cyclopiazonic acid (CPA) was dissolved in DMSO and diluted in external solution at volumes of ≤ 0.1 %.

All chemicals were from Sigma (Poole, Dorset, UK), apart from Indo-1 (Molecular Probes, Eugene, OR, USA).

Statistical analysis

Results are expressed as means ± s.e.m. and n is the number of animals. Differences were taken as significant if P values were < 0.05 in the appropriate Student t test, or ANOVA.

Results

Spontaneous and high K⁺ depolarized uterine activity in 10-day-old neonatal rats

The neonatal uterus was spontaneously active, producing more or less regular Ca²⁺ transients followed by phasic contractions (Fig. 1A) at a frequency of 4.5 ± 0.7 contractions per 10 min at 37 °C (n = 6). These contractions were preceded by increases in [Ca²⁺]_i in all cases. The duration of the Ca²⁺ and force transients were 2.3 ± 0.4 and 2.9 ± 0.6 min, respectively, based on averaging three contractions in each of six animals.

Figure 1. Effects of 0 Ca²⁺ and nifedipine on spontaneous [Ca²⁺]_i and force transients in uterus from 10-day-old neonatal rats.

A, simultaneous force (top) and [Ca²⁺]_i (Indo-1 ratio, bottom) recordings from spontaneous activity in uterus from10-day neonatal rats. B, effect of 0 Ca²⁺ solution containing 1 mm EGTA on spontaneous force and [Ca²⁺]_i transients, and C, effect of nifedipine (0.1 μm) on spontaneous force and [Ca²⁺]_i transients.

One minute of 40 mm K⁺ depolarizing solution induced a rapid rise in [Ca²⁺]_i (0.4 ± 0.03 min to peak, n = 4), a maintained maximal [Ca²⁺]_i (plateau phase) for 0.5 ± 0.01 min, and a more gradual return to basal [Ca²⁺]_i over 1.5 ± 0.09 min. Changes in force followed those of [Ca²⁺]_i but differed in that there was little plateau phase. Force increased over 1.3 ± 0.06 min to its maximum and then gradually decreased to basal levels over 3.6 ± 0.2 min (see Fig. 5).

Figure 5. Agonist-induced SR release and force responsiveness in uterus from 10-day-old neonatal rats.

Normalized [Ca²⁺]_i (Indo-1 ratio) and force records in response to high K⁺ and then carbachol (100 μm; A) and PGF_2α (1 μm; B), applied 2 min after exposure to 0 Ca²⁺ solutions containing 1 mm EGTA. In A and B the dotted line represents basal [Ca²⁺]_i prior to the addition of 0 Ca²⁺ solution. C, the normalized [Ca²⁺]_i records from three experiments are superimposed, comparing the carbachol-induced [Ca²⁺]_i transient obtained in a control solution with Ca²⁺, in the presence of 0.1 μm nifedipine and in 0 Ca²⁺ solution containing 1 mm EGTA. Control (black) and nifedipine-treated animals (grey) were normalized to a high-K⁺ control (100 %) at the start of each experiment.

To investigate whether spontaneous contractions are dependent upon external Ca²⁺ entry, the solution was changed to one with 0 Ca²⁺ and 1 mm EGTA (Fig. 1B). The Ca²⁺ transients and force were rapidly abolished (typical of three others). Both were restored upon return of Ca²⁺ to the bathing solution (not shown). Nifedipine (0.1 μm), a blocker of L-type voltage-gated Ca²⁺ channels rapidly abolished Ca²⁺ and force transients in the neonatal uterus (n = 4 Fig. 1C). Taken together these data indicate that spontaneous uterine activity in these 10-day neonatal uteri is dependent upon external Ca²⁺ entry through voltage-gated, L-type Ca²⁺ channels.

The effects of cyclopiazonic acid on myometria of 10-day-old neonatal rats - evidence for an SR Ca²⁺ store

Application of CPA (20 μm) to spontaneously contracting uterus, to inhibit the SR Ca²⁺-ATPase (Kosterin et al. 1996), caused a marked stimulation and clear changes to Ca²⁺ and spontaneous phasic activity (Fig. 2). There was a significant increase in the baseline [Ca²⁺]_i and force, which slowly returned to control levels after CPA withdrawal; time to recovery was 16 ± 3 and 36 ± 4 min, respectively, the difference was significantly different.

Figure 2. The effect of CPA (20 μm) on spontaneous force and [Ca²⁺]_i transients (Indo-1 ratio) in the uterus in 10-day-old neonatal rats.

As well as elevating basal [Ca²⁺]_i and force, CPA increased the frequency of spontaneous Ca²⁺ transients and contractions in such a way that they were initially fused, contributing to the increase in baselines. Between 5 and 15 min following CPA removal, the frequency of spontaneous contractions remained significantly greater than during the preceding control period (7.1 ± 0.1 contractions vs. 3.9 ± 0.8 contractions (10 min)⁻¹ respectively, n = 4). It is also clear from Fig. 2 that the durations of Ca²⁺ transients and phasic contractions were significantly reduced from 1.9 ± 0.4 to 0.8 ± 0.06 min and from 2.4 ± 0.5 to 1.3 ± 0.08 min, respectively, and on occasion their amplitude was increased by CPA.

Agonist-induced Ca²⁺ and force measurements

Three agonists; carbachol, prostaglandin (PG) F_2α and oxytocin were applied to the neonatal uterus following a control stimulation with high K⁺. Figure 3A-C shows the comparative response of the neonatal uterus to the three agonists - each response is normalized to the high-K⁺ control (= 100 %). Compared with high-K⁺ controls, carbachol induced a greater maximal rise in [Ca²⁺]_i (140 ± 4 % n = 6; Fig. 3A) than PGF_2α (116 ± 6 %; n = 5, Fig. 3B). Carbachol and PGF_2α induced a similar maximal rise in force of 180 ± 19 % and 184 ± 6 %, respectively. Thus in both cases, maximal force was significantly greater than maximal [Ca²⁺]_i. Figure 3A and B clearly shows that force remained high when [Ca²⁺]_i had returned to basal levels. This was most pronounced with carbachol-induced contractions, when force took 9.7 ± 1.8 min longer than [Ca²⁺]_i to return to basal levels (cf. 2.8 ± 0.6 min with PGF_2α). Figure 3C clearly shows that oxytocin induced only a small contraction in neonatal rat uterus. Cytosolic Ca²⁺ and force increased to 21.6 ± 6 % and 54.5 % ± 6 %; n = 4, respectively, compared to control high-K⁺. The responses to carbachol were investigated in more detail, and compared to adult myometrium.

Figure 3. Agonist-induced changes in force (black) and [Ca²⁺]_i (grey) in neonatal and adult uterus.

Agonists were applied for 1 min and changes in force and Indo-1 ratio were normalized to those obtained during a 1 min application of 40 mm K⁺ solution at the start of each experiment (100 %). Neonate: 100 μm carbachol (A), 1 μm PGF_2α (B), 100 nm oxytocin (C). Adult: 100 μm carbachol (D).

Figure 3D shows the effect of carbachol in adult myometrium. Three aspects can be noted. Firstly, that the response to carbachol in the neonate was significantly prolonged compared to the adult. Thus in the neonate force remained at maximal values for 3-4 min and remained elevated for at least 15 min (n = 6), whereas in the adult, force was back to baseline values in around 1 min (n = 6). Secondly, it is clear that force and [Ca²⁺]_i mirrored each other well in the adult myometrium, but this was not the case in the neonates; the Ca²⁺ response was restored to control level within 7 min, when force was still near maximal values, suggesting that Ca²⁺ sensitization had occurred. Thirdly, when compared to a control high-K⁺ depolarization-evoked contraction in each tissue, carbachol induced significantly greater maximal [Ca²⁺]_i and force in neonates compared to adults. These values were 140 ± 4 % vs. 113 ± 10 %, and 182 ± 19 % vs. 120 ± 7 %, respectively (high-K⁺ control amplitude + 100 %).

Comparison of the contribution of Ca²⁺ entry and SR release to the carbachol response in adult and neonatal uterus

To assess the contribution of voltage-gated Ca²⁺ entry to the responses in adult and neonatal tissues, nifedipine (0.1 μm) was used. In both tissues nifedipine was applied for 10 min and then carbachol applied, in the continued presence of nifedipine. Paired control experiments were performed without nifedipine. As shown in Fig. 4A, in 10-day neonatal rats, nifedipine attenuated the [Ca²⁺]_i and force responses to carbachol, but did not alter their overall pattern. Nifedipine significantly reduced carbachol-induced maximal [Ca²⁺]_i and force by 23 ± 5 % and 28 ± 4 %, respectively (n = 5). The effects of nifedipine on myometria from adult rats were much more striking. Compared with neonatal rats, nifedipine attenuated carbachol-induced [Ca²⁺]_i and force in the myometrium of adult rats to a significantly greater degree, reducing carbachol-induced maximal [Ca²⁺]_i and force by 58 ± 4 % and 68 ± 5 %, respectively (n = 4; Fig. 4B).

Figure 4. Comparison of the effect of nifedipine (0.1 μm) on carbachol (100 μm)-induced force and [Ca²⁺]_i in neonatal (A) and adult rat uterus (B).

Simultaneous measurements of force and Indo-1 ratio were made and data from control (black) and nifedipine-treated animals (grey) were normalized to a high-K⁺ control (100 %) at the start of each experiment.

In order to determine the contribution of IP₃-dependent release of Ca²⁺ from the SR to the carbachol response, experiments were undertaken in Ca²⁺-free solutions (1 mm EGTA) so that any rise in [Ca²⁺]_i would be from the SR.

Two minutes of 0 Ca²⁺ EGTA superfusion caused a small drop in basal [Ca²⁺]_i and force of 24.3 ± 8 % and 8.4 ± 3 %, respectively (Fig. 5A and B, n = 12). Carbachol induced an immediate rise in [Ca²⁺]_i and force, as shown previously. However, the maximal carbachol-induced increase in [Ca²⁺]_i was reduced significantly to 75.7 ± 6 % compared to either the response in the presence of Ca²⁺ (140 ± 4 %) or the high-K⁺ control (100 %). The Ca²⁺ rise was also now extremely transient; not even lasting as long as the carbachol application and significantly shorter than the 7 min duration seen in the presence of external Ca²⁺. The carbachol-induced rise in force closely followed the rise in [Ca²⁺]_i and reached a maximum amplitude of 61.3 ± 5 % of the high-K⁺ control, significantly less than that seen in the presence of [Ca²⁺]_i (Fig. 3B). The rate of decrease in force was much less than that of [Ca²⁺]_i, and force remained at 29.7 ± 2 %, 2 min following carbachol washout, whereas Ca²⁺ had returned to basal levels at this point (Fig. 5A). As shown in Fig. 5B, PGF_2α also produced an immediate, transient rise in [Ca²⁺]_i which was similar in shape to that induced by carbachol. However, the maximal amplitude of the rise was significantly less than that observed with carbachol, i.e. 32.3 ± 1 % compared to the high-K⁺ control (n = 4). The rise in force was similar with both agonists (62.9 ± 6 % and 61.3 ± 5 %). Thus, release from the SR can be evoked by different agonists, but the quantitative response may differ between agonists. In all cases prior incubation with CPA (for 30 min) abolished the carbachol-induced increases in [Ca²⁺]_i and force (n = 3, data not shown). If the prior incubation was with nifedipine (5 min) both parameters were unaffected (n = 2, data not shown). Figure 5C summarizes the contribution of Ca²⁺ from different sources to the carbachol response in the neonatal myometrium.

Characteristics of the neonatal SR and comparison with adult

As can be seen in Fig. 5A, in 0 Ca²⁺ solution the release of Ca²⁺ from the SR in response to carbachol or PGF_2α was transient. It is unclear whether this was because there was no more releasable Ca²⁺ in the SR, or whether the agonists were unable to release further Ca²⁺. To distinguish between these two mechanisms, we performed the experiments illustrated in Fig. 6. In the continued presence of 0 Ca²⁺ solution, carbachol was applied for 1 min, three times with 2 min washout periods between successive applications (n = 4). Only the first application of carbachol produced a rise in [Ca²⁺]_i, i.e. in the myometrium of 10-day neonatal rats one application of 100 μm carbachol mobilized all the releasable Ca²⁺, as subsequent applications of carbachol did not alter [Ca²⁺]_i. It may be argued that the reason that the second and third applications of carbachol were unsuccessful in releasing Ca²⁺ from the SR, was because the prolonged period in 0 Ca²⁺ solution had led to an emptying of the SR, rather than agonist application releasing all the Ca²⁺. However, this was not the case, because even after 10 min in 0 Ca²⁺ EGTA solution, significant changes were still evident (n = 4, data not shown) that were 75 % of the response after 2 min. Figure 6 also shows that the second and third applications of carbachol resulted in small increases in force without any Ca²⁺ transient (a similar but smaller force transient can also be seen in Fig. 5A). This was a consistent response and did not occur when changes of the control solution were made. This carbachol-induced, Ca²⁺-insensitive elevation of force was also noted in the presence of Ca²⁺, as shown in Fig. 3, and again suggests carbachol-induced sensitization of the contractile apparatus.

Figure 6. Effects on [Ca²⁺]_i and force from uterus of 10-day-old neonatal rats of repeated carbachol exposure in 0 Ca²⁺ solution.

Following a 1 min exposure to high K⁺ which increased Ca²⁺ and force, the uterus was placed in 0 Ca²⁺ solution and exposed to carbachol (100 μm) for 1 min three times, with an interval of 2 min between each application.

The different effects of carbachol on adult and neonatal uterus shown in Fig. 3A and D suggest a difference in their ability to release Ca²⁺ from the SR. To directly compare the sizes of the agonist-induced Ca²⁺ releases in 2-day, 10-day and adult uteri we adopted the following protocol: the tissues were depolarized in high-K⁺ throughout, to maintain a similar membrane potential and SR loading and then switched to 0 Ca²⁺ solution. In all experiments high-K⁺ induced a rise in [Ca²⁺]_i, followed by a decline when Ca²⁺ was removed from the bathing solution. The rate of decrease of [Ca²⁺]_i, measured as the decrease in the 2 min period compared to high-K⁺ peak (100 %), was not different between 2-day, 10-day and adult uteri (61 ± 5 %, 88 ± 19 % and 60 ± 9 %, respectively, n = 5 in each group). After 2 min in 0 Ca²⁺ solution, carbachol was applied. In all tissues this induced a transient rise in [Ca²⁺]_i. There was a marked and significant decrease in the amount of Ca²⁺ released with age: 122 ± 5 %; 77 ± 3 %; 44 ± 5 %; high-K⁺ + 100 %, 2-day, 10-day and adult respectively (Fig. 7). There was no difference in the duration of the response.

Figure 7. Comparison of carbachol-induced SR Ca²⁺ release in 2-day neonatal, 10-day neonatal and adult rats.

[Ca²⁺]_i records in response to exposure to high K⁺ and then carbachol (100 μm) applied 2 min after exposure to 0 Ca²⁺ solution.

Discussion

The data in this paper have uncovered large differences in the contribution of the SR to contraction in neonatal uterine smooth muscle compared to adult tissue. In particular, we have shown that the role and capacity of the SR in the neonatal tissue is larger than that in adult myometrium. In addition there is clear evidence of sensitization of the contractile machinery to Ca²⁺ by agonists in the neonatal uterus, unlike adult tissue. The physiological significance of these data is discussed.

Patterns of force production and uterine development

The neonatal uterus was found to be capable of contracting spontaneously in response to high-K⁺ depolarization and to agonists. Thus even by day 10 all aspects necessary for excitation-contraction coupling and signal transduction are apparent and functional. There were, however, qualitative differences between the neonatal uterus and uterus from adult animals. As expected, developed force was less, but also the pattern of spontaneous activity was not quite as regular in neonates as in adults. There is little information on the postnatal development of the uterus. In a study on rat uterus (Brody & Cunha, 1989), it was reported that at 5 days there was the beginning of the circular myometrial layer, but that myometrial development was not complete until 20 days. Preliminary data show that at day 10 a separation into longitudinal and circular muscle layers is apparent, but incomplete (K. Noble, C. Vaillant & S. Wray, unpublished observations). Thus force can be developed by the smooth muscle layers, but communication and coordination appear to be functionally and anatomically incomplete at day 10, perhaps explaining the irregular nature of the spontaneous activity.

The small but surprising responses of both force and [Ca²⁺]_i to oxytocin in the neonatal uterus, may be due to the relatively high levels of oestradiol found between 7 and 14 days in the rat uterus (Dohler & Wuttke, 1975), stimulating production of some oxytocin receptors as well as uterine development. The clear responses to carbachol and PGF_2α, and demonstration of an SR-releasable Ca²⁺ store, show that the relatively small effects of oxytocin were not due to failure of the intracellular signalling pathways in the neonatal uterus.

Evidence for Ca²⁺ sensitization by agonists

While agonists have been shown to clearly alter the relationship between Ca²⁺ and force in several tonic smooth muscles, evidence for Ca²⁺ sensitization is not as apparent in the uterus (Taggart et al. 1997; McKillen et al. 1999). In many cases, force and Ca²⁺ mirror each other very well, as is indeed seen in our data, and it is only during non-physiological manoeuvres, such as prolonged exposure to agonist and tetanic conditions, or 0 Ca²⁺ solutions, that sensitization is seen. Thus it was surprising to see the extremely clear dissociation between Ca²⁺ and force in the neonatal preparations in response to carbachol and PGF_2α. We found extremely vigorous and prolonged responses to these agonists following brief exposure; the force response continued well after [Ca²⁺]_i had returned to near baseline levels and led to marked differences in the pattern of responses between the adult and neonatal tissues. Such Ca²⁺ sensitization is considered to arise by agonists modulating the activity of either myosin light chain kinase (MLCK) or phosphatase (MLCP) or both (Horowitz et al. 1996; Somlyo & Somlyo, 2000). In particular a decrease in the activity of the MLCP by rho-associated kinase (rak), and thereby force promotion, appears to be a physiologically important mechanism in many smooth muscles. In the (adult) uterus, however, where Ca²⁺ sensitization is difficult to demonstrate, this pathway, i.e. rak phosphorylation of MLCP, is not very active (Kupittayanant et al. 2001). From the data obtained in our study we would speculate that the rho-rak pathway is active in neonatal uterus and preliminary data using an inhibitor of the rho-kinase, Y-27632, supports this (K. Noble & S. Wray, unpublished data). Whatever the mechanism, it is apparent that Ca²⁺ sensitization plays a large role in the response to agonists in neonatal uterine smooth muscle, and that this mechanism is downregulated with development and pregnancy. Thus the adult pregnant and non-pregnant uterus is more controlled by Ca²⁺ entry pathways and hence in turn, by mechanisms controlling excitability, than that of the neonate (Wray et al. 2001). This in turn explains the simpler relation between force and Ca²⁺ seen in the adult tissues. The next question to arise is whether the role of the SR may be increased in the neonatal, as compared to the adult uterus. The data we have obtained, discussed next, show that this is indeed the case.

SR and neonatal uterus

The first issue we addressed was, could a functioning SR be demonstrated in the neonatal uterus? Our data revealed that the SR is a large and functionally important Ca²⁺ store in the neonate. Thus, when all external Ca²⁺ is removed, agonists could still elevate Ca²⁺ and produce force, albeit at reduced levels compared to those in the presence of external Ca²⁺. There were no spontaneous contractions in the absence of Ca²⁺ entry, either by using 0 Ca²⁺ solution or blocking L-type Ca²⁺ channels. Thus neonatal uterus, as in the adult (Wray, 1993; Wray et al. 2001), is entirely dependent on external Ca²⁺ entry for the production of spontaneous activity. We postulate from our data that both extracellular Ca²⁺ entry and SR Ca²⁺ release will contribute to the response of the neonatal uterus to agonists. Thus both nifedipine and CPA can reduce the effects of agonists. It was clearly demonstrated, however, that the reliance on Ca²⁺ entry was less in the neonate compared to adult tissues. We calculated that inhibition of Ca²⁺ entry reduces force and Ca²⁺ by around 20-30 % in the neonates but by at least double this amount in adults. This is consistent with a more prominent role for the SR in neonatal tissues, as discussed later. However, there was still some Ca²⁺ that could not be accounted for by L-type entry or SR release (Fig. 5C). It is tempting to speculate that this may be occurring via a capacitative entry pathway, stimulated by the agonist binding and emptying of the SR. Such a process may occur in uterine cells (Tribe, 2001), and complement L-type entry and SR release.

As would be expected, given the evidence for Ca²⁺ sensitization presented above, there was also a disassociation between force and Ca²⁺ in response to agonist application, even when the SR was the only source of Ca²⁺. Thus the Ca²⁺-sensitization mechanism in the neonatal uterus is not dependent upon Ca²⁺ entry. That the SR was the source of the Ca²⁺ rise in response to agonist application was confirmed by the prior application of cyclopiazonic acid; when it was used to empty the SR no increase in Ca²⁺ was apparent. There was also no detectable rise in force under these conditions.

As both adult (Taggart & Wray, 1998) and now neonatal uterus have been shown to have an agonist-releasable Ca²⁺ store, it was of interest to compare their size. We found that the relative elevation of Ca²⁺ was 2-day > 10-day > adult. Thus, already between weeks one and two of postnatal life there is a reduction in the relative size of the Ca²⁺ release from the SR, which then continues into adult life. The SR thus appears to be more prominent in neonatal uteri compared to the adult. This conclusion is further borne out by the response of the two tissues to inhibition of the SR. Studies on adult myometrium of either rats (Taggart & Wray, 1998), humans (Kupittayanant et al. 2000; Tribe, 2001) or mice (Matthew & Wray, 2002) had shown that inhibition of the SR, via CPA, potentiated both force and [Ca²⁺]_i. When we examined the neonatal uterus, a large effect on Ca²⁺ and force was produced. Thus, as was the case for adult myometrium, the SR appears to be acting in an inhibitory way on spontaneous force and Ca²⁺ transients. This may be due to some of the Ca²⁺ entering the cell with action potential depolarization, being taken up by the SR and subsequently vectorially released to surface membrane ion channels (Shmigol et al. 1998, 1999). Ca²⁺-activated K⁺ channels (K_Ca) have been identified in the uterus (Khan et al. 2001) and would appear to be a likely target for SR Ca²⁺ release. In turn activation of the K_Ca channels would be expected to repolarize the membrane and curtail further Ca²⁺ influx and contraction. Such a mechanism has been established in arterial smooth muscle (Brenner et al. 2000), but remains to be clearly shown in the myometrium. Whether the enhanced stimulation of force and Ca²⁺ seen with CPA in neonatal compared to adult uterus is related to the increased amount of releasable Ca²⁺, as discussed above, to capacitative entry or to some other mechanism, such as the arrangement of the SR and submembrane space and ion channels, remains to be established.

Physiological significance

The neonatal uterus is not a dormant, unresponsive tissue. The demonstration that the neonatal uterus, like the non-pregnant uterus, is spontaneously active and capable of responding to agonists, suggests that this phasic activity is an inherent activity of uterine smooth muscle and probably essential for its health. It is this activity which will be built upon and augmented during pregnancy and parturition. The control of the uterine activity in the neonatal uterus shows developmental changes. Our data suggest that the SR and Ca²⁺ sensitization play more prominent roles in the neonatal smooth muscle and that there is a shift to a greater reliance on excitability and Ca²⁺ entry with development. We suggest that this process continues during pregnancy. Thus in the neonatal period, the uterus is closer in character to tonic vascular smooth muscles, which show clear Ca²⁺ sensitization and a dependence upon the SR as a Ca²⁺ source. It is worth noting that these smooth muscles are often much less excitable (i.e. action potentials may be absent) than phasic visceral muscles such as the uterus. It is suggested that as activity in the uterus increases with age and gestation, so either the expression or conductance of Ca²⁺ or K⁺ channels changes, and contraction depends more closely on Ca²⁺ entry than SR Ca²⁺ release. Indeed it is difficult to record inward Ca²⁺ current in neonatal uterine cells but several K⁺ channels can be readily recorded (Li et al. 2002). This fits well with a model of reduced excitability in the neonate compared to adult tissue. In both adult and neonatal uteri the SR appears to have a negative influence on contraction - its inhibition increases the frequency and strength of contraction. Thus the SR should not be regarded as simply a source of calcium for agonist-induced contraction, but rather as a moderator of contraction, probably due, at least in part, to its role in reducing membrane excitability.

By studying neonatal uterus we have uncovered differences pertaining to excitation-contraction mechanisms when compared to adult tissues. Although it is always able to contract, these differences show us the mechanisms in play when the uterus is in its least contractile state. In turn these differences highlight mechanisms that must be up- or downregulated, whether for successful parturition or for tocolytic therapy.