INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE SIGNAL-TRANSDUCTION PATHWAYS that couple G protein-receptor agonists to contractile responses in vascular smooth muscle involve a multitude of sequential steps that begin with the binding of agonist to the cell surface receptor and culminate in increased force production by myosin-actin interactions. For many agonists, this sequence can be conveniently divided into the initial steps linking agonist-receptor interaction to second-messenger release, and the subsequent steps linking increased second-messenger availability to increased contractile activity (Fig. 1). Agonist-induced second-messenger production, in turn, is governed by three main factors: agonist affinity for the receptor, receptor density, and receptor gain, which determines the rate of second-messenger production for each activated agonist-bound receptor. Each of these factors appears to be under physiological control for many different receptor types. For example, both receptor density and agonist affinity for a given receptor type can vary dramatically with age (12), tissue (7), or environmental conditions such as chronic hypoxia (15). In addition, members of the recently discovered family of regulators of G protein-signaling (RGS) proteins also appear to function as modulators of the coupling between receptor binding and intracellular effector activation (20). Thus recent work has identified physiological mechanisms capable of modulating each of the main factors governing coupling agonist-concentration to second-messenger production. How these factors are coordinately influenced in response to well-defined physiological stresses remains unexplored.

View larger version (6K): [in this window] [in a new window]   Fig. 1.   Basic mechanisms of pharmacomechanical coupling. Pharmacomechanical coupling of receptor agonists to contractile responses involves multiple sequential events including receptor activation, second-messenger release, and action of this second messenger on downstream response elements. The coupling between agonist stimulation and second-messenger production can be simplified by viewing the receptor complex as a cofactor-regulated enzyme. From this perspective, the agonist serves as the cofactor, agonist affinity is conceptually related to enzyme-cofactor affinity, and receptor density is equivalent to enzyme concentration. Receptor gain is related to enzyme specific activity in that it defines the rate of second-messenger production for any given set of agonist concentration, affinity, and receptor-density conditions. Coupling between the second messenger and the response, in turn, is governed by both the concentration time course of the second messenger and the sensitivity of the response element to the messenger. In many systems, receptor affinity, density, and gain as well as second-messenger sensitivity all appear subject to physiological regulation. See text for details.

Acute hypoxia is a well-defined physiological stress that influences a wide variety of steps in many signal-transduction pathways. With the use of serotonin as an agonist, we recently demonstrated that acute hypoxia (PO₂ of 9-12 Torr) of 30-min duration can reduce agonist affinity both in adult and fetal ovine common carotid arteries and receptor density in adult arteries (4). Acute hypoxia also altered the relationship between numbers of receptors bound and contractile force in an age-specific manner. In adult ovine common carotid arteries, the size of the contractile response produced for each femtamole bound was quite similar under hypoxic and normoxic conditions, suggesting that hypoxia had relatively little effect downstream from agonist-receptor binding. In the fetus, however, any given number of receptors bound always produced less contractile tone under hypoxic than under normoxic conditions, indicating that events downstream from agonist-receptor binding were vulnerable to attenuation by hypoxia. Together, these observations suggest that in addition to modulating receptor density and agonist affinity, acute hypoxia may also modulate serotonergic receptor gain in an age-specific manner.

The present studies were conducted to test the hypothesis that both acute hypoxia and maturation selectively and differentially influence the gain of the coupling between agonist-concentration and second-messenger production. For these studies, we have examined serotonin-induced inositol 1,4,5-trisphosphate (IP₃) production in ovine common carotid arteries because this preparation offers several important advantages. First, in this preparation, serotonin acts on a single receptor type, the 5-hydroxytryptamine (5-HT) 2A, to stimulate increases in IP₃ synthesis (47). Second, sufficient tissue is available from a single animal to permit measurements of the time courses of IP₃ responses. Third, we have already quantitated the effects of acute hypoxia and maturation on 5-HT_2A-receptor density and agonist affinity in this tissue. The overall strategy of these studies was to correct for the effects of hypoxia on affinity and density by normalizing the numbers of receptors bound in each experimental group and thereby directly observe the effects of acute hypoxia and maturation on both contraction and the relationship between the numbers of receptors bound and the subsequent IP₃ response.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

General methods. Adult carotid arteries were obtained from healthy adult nonpregnant sheep (18-24 mo old) of either sex killed with 100 mg/kg iv pentobarbital sodium. Fetal carotid arteries were obtained from near-term fetuses (139-141 days gestation) of either sex weighing 2.5-4.0 kg that were delivered by cesarean section and killed with 100 mg/kg iv pentobarbital sodium. All procedures were reviewed and approved by the Institutional Animal Use and Care Committee of Loma Linda University. After the arteries were dissected, they were all cleaned of extraneous connective and adipose tissue and cut into multiple segments (1 mm in length for adult sheep; 3 mm in length for the fetal lamb). Two sets of eight matched segments were obtained from each animal and studied in parallel. To avoid possible endothelium-mediated effects, the endothelium was removed from all segments by passing a roughened hypodermic needle through the lumen of the vessel several times and gently flushing it with cold isotonic Krebs solution. As previously shown (4), this method reliably and predictably removes the endothelium from this preparation. Physically denuded segments were thereafter incubated in the continuous presence of the nitric oxide synthase inhibitors nitro-L-arginine methyl ester (L-NAME, 100 µM) and nitro-L-arginine (L-NA, 100 µM). All segments were equilibrated at optimum resting tensions of ~1 g on paired hand-made tungsten wires placed between a low-compliance force transducer (0.6 g/µm displacement, Kulite BG-10) and a post attached to a micrometer used to vary resting tension. The artery segments were equilibrated at 38.5°C (normal ovine core temperature) for 30 min in a bicarbonate-Krebs solution containing (in mM) 122 NaCl, 25.6 NaHCO₃, 5.56 dextrose, 5.17 KCl, 2.49 MgSO₄, 1.60 CaCl₂, 0.114 ascorbic acid, and 0.027 disodium-EDTA with 100 µM L-NA and 100 µM L-NAME continuously bubbled with 95% O₂ and 5% CO₂. Contractility measurements were recorded, digitized, and normalized via an online computer, as previously described in detail (47).

Quantitation of IP₃ responses. After the arteries were equilibrated, they were first contracted by exposure to an isotonic potassium-Krebs solution. Once contractile tensions had plateaued, the segments were washed with sodium-Krebs solution and exposed for 30 min to either 95% 0₂-5% CO₂ (control vessels) or 95% nitrogen-5% CO₂ (hypoxic vessels) mixture. The sodium-Krebs solution contained 0.1 µM prazosin to inhibit -receptor activation and 0.2 µM cocaine to inhibit neuronal uptake of 5-HT. After 30 min of equilibration, IP₃ accumulation responses to specific concentrations of serotonin (see below) were determined at multiple durations of exposure to serotonin to bracket the rise and fall of inositol phosphate responses and thus enable determination of the IP₃ time courses. Specifically, the arteries were contracted with serotonin and then frozen in liquid nitrogen after 0, 15, 30, 45, 60, 90, 120, and 180 s of exposure to serotonin. When all samples were frozen, each segment was homogenized in trichloroacetic acid, extracted in ether, lyophilized overnight, resuspended in purified water, and analyzed for IP₃ content using the IP₃ kit from New England Nuclear Life Science Products (Boston, MA). To facilitate comparisons among individual animals and groups, all IP₃ values were normalized relative to milligrams of cellular protein. Proteins were quantified using the Bradford method, as previously described (33).

Calculation of serotonin concentrations yielding equivalent numbers of receptors bound. Because our previous studies using ovine carotids have indicated that acute hypoxia depresses both the density and agonist-binding affinity of the 5-HT_2A receptors in this preparation (4), a key goal of these studies was to standardize the numbers of ligand-receptor binding events in each preparation. By achieving equivalent numbers of bound receptors in all groups, independent of treatment with acute hypoxia, any subsequent differences among groups in the effects of acute hypoxia on contractile tensions would thus necessarily be attributable to effects downstream from IP₃ synthesis. To standardize the numbers of receptors bound to serotonin, we used the basic derivation of Hulme and Birdsall (16). Briefly, agonist dissociation constants determined in previous experiments (4) were used to calculate fractional occupancy at any given agonist concentration according to the following relationship as first demonstrated by Furchgott and Bursztyn (11)

(1)

where [A] is the agonist concentration, and K_A is the agonist dissociation constant. The numbers of receptors bound were then calculated as the product of fractional occupancy and receptor density

(2)

where B_max is the receptor density (in fmol/mg tissue wet wt). From this relationship, the concentration of agonist yielding any given level of binding could be calculated by simple rearrangement

(3)

where Bound is the number of receptors bound (in fmol/mg tissue wet wt) at [A]. The next step was to determine the standardized number of receptors to be bound in each of the experimental groups. Given that the agonist dissociation constant and receptor density were lowest in the hypoxic adult group, we chose 80% of the total receptor density in this group as our target for the standardized number of receptors bound. As indicated in Table 1, this number was calculated to be 8.54 fmol receptor/mg tissue wet wt. With the use of equation 3, we then calculated the serotonin concentrations that would yield this number of receptors bound in each preparation using previously determined values of receptor density and agonist dissociation constants, as indicated in Table 1.

Data analysis. All values were calculated as means ± SE. Reported values of n always refer to the number of animals used in any given experimental group. The numbers of animals used were chosen to yield statistical powers of at least 0.95 for each experimental group. Before statistical analysis, the distributions of all data sets were analyzed for normalcy; all values in all groups were normally distributed, and no log transformations were necessary. Age-related differences in the effects of acute hypoxia on contractile tensions were determined by performing an ANOVA on the paired differences between the normoxic and hypoxic responses to serotonin. To calculate the IP₃ time courses, baseline values were first subtracted, after which the increases above baseline were normalized relative to the maximum response for each individual set of arteries. This approach ensured that each artery set had the same statistical weight and contributed equally to the calculation of the overall time course in each group. Once average values of percent maximum had been calculated at each time of exposure to serotonin, the means were multiplied by the average amplitude of change observed in each group to obtain the values (see Fig. 4). Areas beneath the IP₃ time course curves were calculated using Simpson's approximation. Time courses of IP₃ responses to hypoxia were compared using a repeated-measures ANOVA. All other values were compared using ANOVA followed by Fisher's least-significant difference test to assess between-group differences.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

From a total of 11 adult sheep and 9 fetal lambs, we obtained a total of 320 common carotid segments (16 segments/animal, 8 each for the hypoxic and normoxic groups). The maximum normoxic contractile tensions produced by these arteries in response to 120 mM KCl averaged 4.12 ± 0.43 and 6.50 ± 0.88 g in the fetal and adult arteries, respectively.

Effects of acute hypoxia on initial contractile responses to 5-HT. After 3 min of normoxic exposure to concentrations of 5-HT that yielded equivalent numbers of bound receptors in all experimental groups, contractile tensions averaged 50.2 ± 13.6 and 40.0 ± 8.5% of corresponding maximum responses to 120 mM potassium in the fetal and adult segments, respectively (Fig. 2). Corresponding values observed after 30-min exposure to hypoxia averaged 17.9 ± 5.2 and 21.4 ± 4.1%, respectively. ANOVA revealed no significant effects of age on the magnitudes of tension under either normoxic or hypoxic conditions. To more closely analyze the effects of hypoxia on contractile tensions, we calculated the paired differences between corresponding hypoxic and normoxic values of tension at each time point during exposure to hypoxia in each experimental group. This calculation yielded two difference curves indicating the time courses of response to hypoxia in the fetal and adult artery groups (not shown). Comparison of these difference curves revealed that the fetal contractile responses during hypoxia followed a significantly different time course and were of significantly greater magnitude (endpoint: 30 ± 10%) than those of the adult (endpoint: 17 ± 11%), particularly during the first minute of exposure to hypoxia.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Effects of acute hypoxia on initial contractile responses to 5-hydroxytryptamine (5-HT). Shown here are the early time courses of contractile responses to 5-HT, expressed relative to the maximum normoxic responses to 120 mM isotonic potassium. As determined by analysis of variance, acute hypoxia significantly reduced maximum contractile responses in both fetuses (A) and adults (B). All values are expressed as means and standard errors for 9 fetuses and 11 adults.

Effects of acute hypoxia on IP₃-response time courses. Basal values of IP₃ content (in pmol/mg protein) were significantly greater in fetal (94 ± 16) than adult (44 ± 6) arteries (Fig. 3) for both the normoxic and hypoxic groups. Exposure to hypoxia for 30 min had no significant effect on basal values, which after hypoxia averaged 84 ± 16 and 32 ± 7 pmol/mg protein in fetal and adult arteries, respectively. Treatment with serotonin produced a complex IP₃ response that, in normoxic fetal arteries, attained a peak value of 83 ± 24 pmol/mg protein above baseline at 45 s. Hypoxia appeared to alter this response and produced dual peaks of 57 ± 13 and 57 ± 16 pmol/mg protein above baseline at 30 and 60 s, respectively (Fig. 4). The peak magnitudes of the fetal normoxic and hypoxic IP₃ responses were significantly higher than baseline in all time points.

View larger version (39K): [in this window] [in a new window]   Fig. 3.   Effects of maturation and acute hypoxia on baseline inositol 1,4,5-trisphosphate (IP3). Baseline IP3 contents were significantly greater in fetal than adult arteries, but they were not affected by acute hypoxia. All values are expressed as means and standard errors for 9 fetuses and 11 adults.

View larger version (23K): [in this window] [in a new window]   Fig. 4.   Effects of acute hypoxia on time courses of IP3 responses. Activation of the same numbers of ligand-receptor complexes produced quite different effects in the different experimental groups, as indicated here by the time courses of the IP3 responses, expressed as values above baseline in pmol IP3 per mg base-soluble protein. Acute hypoxia appeared to alter the timing of the peak IP3 responses to 5-HT in both fetal and adult arteries. In both age groups, the IP3 responses were multiphasic and were significantly higher than baseline at all measured time points after administration of 5-HT. All values are expressed as means and standard errors for 9 fetuses and 11 adults.

View larger version (43K): [in this window] [in a new window]   Fig. 5.   Effects of acute hypoxia on areas beneath the IP3-time curves. Total IP3 released in response to activation of a standardized number of receptors bound to 5-HT (8.54 fmol bound per mg wet wt) was estimated by integration of the area beneath the IP3 time courses of response shown in Fig. 2. For this standardized number of ligand-receptor interactions, fetal arteries produced significantly more total IP3 than did adult arteries. Acute hypoxia had no significant effect on the mass of IP3 produced per ligand-receptor interaction. For details of the calculations of the numbers of receptors bound, please see Table 1. All values are expressed as means and standard errors for 9 fetuses and 11 adults.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Contemporary views of pharmacomechanical coupling have been continuously revised since its basic principles were first articulated by A. J. Clark in the 1930's (44). Clearly, receptor density and agonist affinity play key roles in determining the magnitude of the second-messenger signal (Fig. 1), and correspondingly numerous studies have demonstrated that physiological modulation of these variables enables adaptation to influences such as hypoxia (4, 37), pregnancy (22), diet (14), exercise (10), maturation (4, 36), hypertension (46), and diabetes (8). On the other hand, appreciation of the regulatory importance of changes in coupling efficiency between receptor activation and second-messenger production is much more modest, due in large part to a relative absence of direct evidence documenting physiological changes in this coupling. In his initial description of pharmacomechanical coupling, Stephenson (41) defined intrinsic efficacy as the receptor characteristic that determined the strength with which any receptor was coupled to its effector. Interestingly, intrinsic efficacy was viewed as an inherent receptor property that should not vary from tissue to tissue. However, several recent studies have suggested that the efficiency of the coupling between receptor activation and second-messenger production may be quite variable (29, 48). To distinguish Stephenson's concept of intrinsic efficacy from processes coupling receptor activation to second-messenger production, we have coined the term "receptor gain." From a physiological perspective, "gain" is a composite variable that certainly includes intrinsic efficacy, but also includes many other mechanisms. For IP₃-dependent responses, these other mechanisms include the phospholipase C (PLC) isoform, activity, and concentration, as well as the phosphatidyl inositol bisphosphate pool size (24). In addition, receptor gain should also be influenced by any of the mechanisms that modulate IP₃ turnover and metabolism.

Evidence of physiological regulation of receptor gain comes from a variety of sources. For example, several different isoforms of inositol-specific PLC have been cloned and sequenced (17), and the activities of some of these have been shown to vary with age (18), phosphorylation by protein kinase A (28), and binding to G protein beta-gamma subunits (27). Whether or not such changes are typical of PLC-, which is the main form in vascular smooth muscle (6, 23), is at present unknown but appears possible. Similarly, metabolism of membrane lipids related to phosphatidyl inositol bisphosphate, the substrate for PLC, may be influenced by age and/or diet (3). The activity of IP₃ kinase, a major enzyme inactivating IP₃, has also been shown to vary with ischemia (26, 43), maturation (35), and phosphorylation mediated by either protein kinase A or C (25, 40). Another key enzyme in IP₃ metabolism is IP₃ 5-phosphatase, whose activity also varies with age (35). Because the present experiments were purposefully completed in the absence of inhibitors of IP₃ metabolism, all these latter influences could potentially contribute to the observed variations in receptor gain. Equally important, the main action of RGS proteins is to modulate receptor-mediated activation and coupling to heterotrimeric G proteins (5). Together, this evidence strongly suggests that receptor gain is not only a complex composite variable, but is also subject to physiological regulation.

To explore the hypothesis that receptor gain is physiologically regulated, the present experiments examined differences in gain associated with maturation and responses to acute hypoxia, both previously shown to alter receptor-IP₃ coupling (7, 9, 49). In studies from our own laboratory, we have demonstrated that both age and acute hypoxia can modulate receptor density and agonist affinity, at least for serotonin in ovine carotid arteries (4). Thus the experimental approach we adopted was designed to emphasize differences in receptor gain independent of possible changes in receptor density or agonist affinity. Accordingly, we standardized the numbers of receptors bound for each of our experimental groups and thus corrected for age- and hypoxia-related variations in receptor density and agonist affinity. The fraction of receptors bound varied from 37 to 78% across all groups and thus resided in the linear portion of the concentration-occupancy relationship and avoided potential complications associated with either subthreshold or near-maximal binding. To avoid possible problems associated with variable dynamics of the IP₃ responses in the different experimental groups, we quantified IP₃ responses as the integrated areas beneath the IP₃ time curves of response to serotonin. Although these measurements reflected average arterial IP₃ levels independent of possible subcellular compartmentation, the area results, and thus the gain estimates, were highly reproducible (Fig. 5).

With constant numbers of receptors bound in all experimental groups, acute hypoxia did not influence IP₃ area, and thus gain, in either fetal or adult arteries (Fig. 5). This observation suggests that previously reported hypoxic attenuation of inositol phosphate responses (9) was due more to significant depression of agonist affinity and/or receptor density, as we have demonstrated (4), than to factors downstream from receptor-ligand coupling, such as receptor gain. Interestingly, although acute hypoxia did not affect total IP₃ area, it did significantly affect the IP₃ time course over 3 min (Fig. 4). Acute hypoxia appeared to alter both the numbers and magnitudes of the IP₃ peaks produced by serotonin both in the fetus and the adult. The mechanisms responsible for these alterations cannot be discerned from the present data but could possibly involve dynamic differences in any of the determinants of gain, including either PLC activity or IP₃ turnover, as mentioned above.

In contrast to acute hypoxia, IP₃ areas, and thus estimates of receptor gain, were significantly greater in fetal than adult arteries (Fig. 5). In addition, the time courses of IP₃ responses were markedly different in fetal and adult arteries (Fig. 4), and baseline levels of IP₃ content were also elevated in fetal compared with adult arteries (Fig. 3), suggesting that some common feature of fetal arteries involving either IP₃ synthesis or degradation was in some way responsible for these observed age-related differences. Consistent with the possibility that age-related differences in IP₃ degradation may be involved, IP₃ 3'-kinase exhibited significantly lower activity in immature than in mature samples of either rabbit tracheal smooth muscle (35) or rat cardiac muscle (19). In addition, published K_m values for the 3'-kinase are 14- to 19-fold less than for IP₃ 5'-phosphatase, indicating the 3'-kinase is the primary pathway for IP₃ degradation, at least in rabbit tracheal smooth muscle (35). Although studies of the effects of age-related changes on PLC expression in vascular smooth muscle have yet to be reported, recent studies in developing neurons suggest that PLC- exhibits an age-dependent expression that is greater in immature than in mature, fully differentiated cells (13, 39). Together, these observations provide several possible mechanistic bases for the age-related differences observed in relation to the magnitudes and amplitudes of IP₃ responses and suggest potentially fruitful directions for future work in this area.

Although acute hypoxia did not significantly affect receptor gain in either age group, it did depress contractile tone in both fetal and adult arteries (Fig. 2). Because receptor gain, and therefore IP₃ production, was not affected by hypoxia, hypoxia's effect on contractility can be explained only by inhibition of the ability of IP₃ to induce contraction (Fig. 1). Furthermore, the observation that the magnitude of hypoxic relaxation was greater in fetal than adult arteries suggests that the ability of IP₃ to induce contraction is more sensitive to hypoxia in fetal than in adult arteries. This age-related difference may involve many possible mechanisms, including age-related differences in the density (4, 52) and isoform (45) of IP₃ receptors, hypoxia-induced changes in intracellular pH (30) or IP₃-receptor phosphorylation (21), or coupling between IP₃-receptor activation and calcium release (42). Clearly, many additional experiments will be required to distinguish among these possibilities.

Interpretation of the possible effects of acute hypoxia on IP₃-contractility coupling is complicated by possible parallel effects of hypoxia on other IP₃-independent pathways coupling receptor activation to contraction. In many preparations, receptor activation is associated with increases in calcium influx, and this enhanced influx can be inhibited by acute hypoxia (32). Receptor activation is also associated with activation of Rho-kinase, increased phosphorylation of myosin light chain phosphatase, and increased calcium sensitivity (2). Furthermore, stimulation of G protein-coupled receptors can lead to activation of other IP₃-independent transduction pathways including those mediated by increased activities of tyrosine kinase (34) and mitogen-activated protein kinase (31), pathways that may be particularly important for 5-HT_2a-mediated responses (50, 51). Together, these observations suggest caution when interpreting the possible effects of acute hypoxia on IP₃-contraction coupling. On the other hand, these findings also suggest additional mechanisms whereby acute hypoxia may mediate vasodilatation.

Perspectives