INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

SEROTONIN'S MANY cerebrovascular effects include responses of both physiological and pathophysiological importance. Perivascular serotonergic nerves may help couple cerebral metabolism and perfusion (5) and may also contribute to migraine (28). Serotonin participates in cerebrovascular responses to aggregating platelets and may also contribute to development of the vasospasm that typically follows intracranial hemorrhages (38). Although the exact role of serotonin in many cerebrovascular responses remains under investigation, its importance in the cerebral vasculature seems certain.

Perhaps as a consequence of the many diverse cerebral functions of serotonin, cerebrovascular reactivity to this agonist is not static and instead can be modulated by many influences. Acclimatization to chronic hypoxia alters cerebrovascular reactivity to serotonin and other amines (26), as do pregnancy (20) and maturation (45). Cerebrovascular reactivity to serotonin and other amines also varies in relation to artery size and type (10, 45). Although the extent and nature of this variability is well documented, the mechanisms responsible remain unclear.

Aside from factors such as contractile protein content and intracellular calcium pool size that govern overall contractile capacity, one of the most important determinants of arterial reactivity to any agonist is the receptor type to which it binds. Indeed, shifts in receptor type are often closely associated with shifts in vascular function. For example, the receptor mediating responses to norepinephrine shifts from the ₁-subtype in large arteries to the ₂-subtype in small arteries (23, 24). Given the importance of receptor type in determining reactivity, it is possible that changes in serotonergic reactivity related to differences in artery size, type, age, and hypoxic exposure could also be associated with shifts in receptor type. The present experiments were conducted to evaluate these possibilities.

Serotonergic receptors at present are categorized into seven main families, of which the 5-hydroxytryptamine types 1 and 2 (5-HT₁ and 5-HT₂, respectively) families most commonly mediate contractile responses to 5-HT in vascular smooth muscle (18, 19, 39). Within these two main receptor families, at least five different 5-HT₁ subtypes and three different 5-HT₂ subtypes have been pharmacologically characterized and cloned (39). To examine how these various 5-HT receptor types might vary with cerebral artery type (or origin) and diameter, all experiments were carried out in common carotid (Com), main branch middle cerebral arteries (Main), and second branch order segments (2BR) of middle cerebral arteries. Owing to its extracranial origin, the Com was expected to be more similar to peripheral, than to cerebral, arteries. To examine how 5-HT receptor type can vary with maturation, arteries from both term fetal lambs and nonpregnant adult sheep were studied. To examine how hypoxic acclimatization can influence 5-HT receptor type we also examined arteries from nonpregnant adult sheep acclimatized at an altitude of 3,820 m for 110 days. To identify the serotonergic receptors involved in each of these experimental groups, we determined both relative agonist potencies and antagonist affinities according to the criteria established by Hoyer et al. (18, 19). With the use of these approaches, we investigated the extent to which maturation, chronic hypoxia, artery size, and artery type affect 5-HT receptor type in the ovine cerebral circulation.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

General methods. For these studies, we examined Com, Main, and 2BR arteries from adult rams and nonpregnant ewes that had been maintained near sea level or at high altitude (3,820 m; Barcroft Laboratory, White Mountain Research Station, Bishop, CA) for 110 days (arterial PO₂  64 Torr) as well as near-term normoxic fetal sheep (139-141 days gestation) using methods previously described (10, 33). Both males and females were included in the normoxic fetus group, only nonpregnant fully mature ewes (24 mo or older) were included in the hypoxic adult group, and the normoxic adult group consisted mainly of nonpregnant fully mature ewes with inclusion of a few young males (18-24 mo). Statistical comparisons of basic characteristics, such as reactivity to 5-HT, maximum responses to potassium, and maximum responses to endothelium-dependent vasodilators, as previously described (33), were performed and verified that no significant differences existed between animals of different sex.

Briefly, after the initial dissection and removal of excess adipose and connective tissue, we cut the arteries into individual ring segments 1-3 mm long. To avoid endothelial influences, we removed the endothelium by mechanical abrasion achieved in Com and Main segments by carefully passing a small needle through the lumen and in 2BR segments by passing a human hair through the lumen of the vessels several times. We then mounted the segments on paired tungsten wires (OD 250 µm for Com, 125 µm for Main, and 24 µm for 2BR) between a low-compliance force transducer (Kulite BG-10) and a post attached to a micrometer used to vary resting tension. We equilibrated the arteries at 38.5°C (normal ovine core temperature) for at least 30 min in baths superfused with a bicarbonate Krebs solution containing (in mM) 120 NaCl, 5.56 dextrose, 25.6 NaHCO₃, 5.17 KCl, 2.49 MgSO₄, and 1.60 CaCl₂, in addition to 114 µM ascorbic acid and 27 µM disodium EDTA. The baths were continuously bubbled with 95% O₂-5% CO₂. We obtained micrometer readings for each segment under unstressed conditions (100 mg total tension in Com and Main; 10 mg in 2BR) and calculated unstressed diameter as twice the distance between the two wires divided by pi. For 2BR segments, optimum diameter was taken as 2.9 and 2.2 times unstressed diameter in adult and fetal segments, respectively, which in our previous studies has proven to be optimum in terms of both reproducibility and maximum force generation (10). This extent of stretch yielded resting tensions of 150-250 mg. Com and Main segments were stretched until resting tensions of 1.0 and 0.5 g were attained, respectively. During all contractility experiments the contractile tensions from all artery segments were continuously recorded on both an ink-writing oscillograph (model 8373-20 recorder, Cole-Parmer Instrument) and an online computer, which digitized and normalized the data.

After 30 min of equilibration at optimum stretch, we contracted all arteries with a potassium-Krebs solution similar to that described above, with the exception that NaCl was eliminated and exchanged for KCl on an equimolar basis that yielded a final potassium ion concentration of 120 mM. Exposure to potassium-Krebs was repeated until reproducible maximal contractions were obtained, which generally required two or three exposures. The arteries were then contracted with either 10 µM 5-HT (Com and Main) or UTP (2BR) segments. UTP was used in the 2BR segments because it produced a much more stable contractile response than that to 5-HT. Once agonist-induced tone had stabilized, the arteries were then exposed to either 1 µM ADP (Com and Main) or 1 µM acetylcholine (2BR). ADP was used in Com and Main segments because it is the most potent endothelium-dependent relaxant we have found in these artery types. Correspondingly, acetylcholine was used in the 2BR segments for similar reasons. Arteries of either type exhibiting relaxations >10% to ADP or acetylcholine were discarded.

Agonist dose-response measurements. After verification of endothelial denudation, the arteries were equilibrated for another 30 min at optimum stretch in normal Krebs to which we added 0.1 µM prazosin and 0.2 µM cocaine to block any possible effect of ₁-receptors and neuronal uptake, respectively. Thirty minutes later, cumulative concentration-response curves were generated for 5-HT; 5-carboxamidotryptamine (5-CT), a preferential agonist of 5-HT_1A and 5-HT_1E receptors; 8-hydroxy-2(di-n-propylamino)tetraline (8-OH-DPAT), a preferential agonist of 5-HT_1A receptors; and sumatriptan (Suma), a preferential agonist of the h1B and 1D serotonergic subtypes. Each agent was added in half-log increments from 0.1 nM to 100 µM. Each segment was used to determine only one concentration-response relation to only one agonist. On completion of the concentration-response determinations, the contractile data were normalized relative to the maximal response to potassium and then fitted to the logistic equation using nonlinear regression to obtain pD₂ values (log EC₅₀) and the maximal contractile responses (E_max) for each agonist.

Determination of antagonist pK_b values. In preliminary experiments, 5-HT caused significant desensitization in repeated concentration-effect determinations. To avoid this effect, which violates classical Schild conditions, we determined antagonist dissociation constants (pK_b) by comparing concentration-effect curves obtained from control and antagonist treatments in paired adjacent artery segments. Owing to the large number of antagonist-agonist combinations employed, tissue availability could not accommodate allocation of more than one or two artery segments from each animal to determination of each estimate of antagonist affinity. Therefore, pK_b values were determined instead of pA₂ values, which require analysis of the effects of multiple concentrations of anatogonist using a Schild plot.

To normalize for segment-to-segment differences in contractile capacity, all responses were normalized relative to the maximum response within each segment to 120 mM potassium-Krebs. Given the reasonable assumption that equal magnitudes of response corresponded to equal numbers of agonist-receptor interactions in adjacent segments from the same artery, we then calculated equieffective dose ratios between control and antagonist-treated segments at the pD₂ concentration of the antagonist-treated segments. For this calculation, we determined coefficients for the logistic equation for both control and antagonist-treated arteries and then solved for the concentration in the untreated control segments, which produced the tension observed at the pD₂ concentration in the antagonist-treated segments. Because this approach meets the general requirements necessary for a Schild analysis, we then used the Furchgott equation to determine pK_b: [A']/[A] = 1 + ([B]/K_B) where [A] is the concentration of agonist in the control tissues, [A'] is the concentration of agonist in the blocked tissues at the pD₂, [B] is the concentration of antagonist, and pK_b is the log K_B. In this manner, each estimate of pK_b was obtained from a matched pair of artery segments, one treated with the antagonist and the other served as the control. The antagonists used included the 5-HT_2A-selective antagonist ketanserin, the mixed 5-HT₁/5-HT₂ antagonist methiothepin, the 5-HT_h1B- and 5-HT_1D-selective antagonist GR-127935, and the selective 5-HT_2B/5-HT_2C antagonist SB-206553.

Data analysis and statistics. The data are expressed throughout the paper as means ± SE. Significant differences in agonist potency (pD₂) and E_max among 5-CT, 5-HT, 8-OH-DPAT, and Suma were evaluated using ANOVA techniques followed by a Duncan's multiple-range analysis. Given that pooled variances for common carotid pD₂ and E_max values were significantly different than observed in the cerebral arteries (Main and 2BR), the common and cerebral arteries were analyzed separately. Comparisons between the different agonists were performed using the Duncan's multiple-range analysis at the P < 0.05 level. Significant differences in antagonist affinities (pK_b) were analyzed using two-way ANOVAs with experimental group as one factor and artery type as the other. Each antagonist was analyzed separately. Comparisons of pK_b values among experimental groups and among agonists were performed using a Duncan's post hoc analysis. Unless indicated otherwise, statistical significance was taken at the P < 0.05 level.

Drugs. The drugs used were 5-HT, 8-OH-DPAT, cocaine hydrochloride, prazosin, UTP, acetylcholine (all from Sigma); methiothepin, ketanserin, 5-CT, and SB-206553 (all from Research Biochemicals); and Suma and GR-127935 (supplied by Glaxco Pharmaceuticals). Stock solutions of all drugs were made using distilled water, were stored as aliquots at 20°C until time of use, and were never frozen more than once.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

General results. From a total of 41 normoxic adults, 34 normoxic fetuses, and 29 hypoxic adults, a total of 380 Com, 313 Main, and 346 2BR artery segments were taken for study. Com segment diameters averaged ~3.0 mm in adults and 2.6 mm in fetuses. Main diameters averaged ~640 µm in adults and 470 µm in fetuses. For 2BR segments, the diameters averaged 201 ± 3 µm in normoxic adults, 197 ± 5 µm in hypoxic adults, and 203 ± 3 µm in normoxic fetuses. The maximal contractile tensions produced by these arteries in response to 120 mM KCl averaged 6.2 ± 0.2 g in Com segments, 3.4 ± 0.1 g in Main segments, and 0.63 ± 0.01 g in 2BR segments. When precontracted with 1 µM 5-HT (Com and Main) or 20 µM UTP (2BR), neither 10 µM ADP (Com and Main) nor 1 µM acetylcholine (2BR) produced any significant vasodilatation, indicating that endothelium removal was effective and complete.

Agonist potencies. In Com arteries, the concentration-response curves for 5-CT and 8-OH-DPAT were right-shifted relative to those for 5-HT (Fig. 1, Table 1). Suma produced no significant contractile effect in Com segments from any experimental group. As shown in Table 1, the relative order of agonist potency was 5-HT > 5-CT and 5-HT > 8-OH-DPAT in Com arteries from all experimental groups, although the relation between 5-CT and 8-OH-DPAT potency in the Com segments varied with group. Com E_max values followed the same general pattern observed for pD₂: 5-HT > 5-CT and 5-HT > 8-OH-DPAT, with relation between 5-CT and 8-OH-DPAT varying among the different experimental groups (Table 2).

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Concentration-response relations for 5-hydroxytryptamine (5-HT), 5-carboxamidotryptamine (5-CT), 8-hydroxy-2(di-n-propylamino)tetraline (DPAT), and sumatriptan (Suma). Cumulative additions of 5-HT, 5-CT, 8-OH-DPAT, and sumatriptan to common carotid (A), main branch middle cerebral (B), and second branch middle cerebral (C) arteries of normoxic adult sheep yielded the concentration-response relations indicated. All contractile responses are expressed as %maximum responses to K+. Suma produced no measurable response in common carotid segments from any experimental group. All values are indicated as means ± SE, and vertical error bars indicate SEs for the numbers of animals indicated in Table 1.

In Main segments, the concentration-response curves for 5-CT, 8-OH-DPAT, and Suma were all right-shifted relative to those for 5-HT (Fig. 1, Table 1). The relative order of agonist potencies was 5-HT > 5-CT  Suma  8-OH-DPAT in middle cerebral arteries from both the normoxic and hypoxic adult groups but was different than this in Main segments from normoxic fetuses: 5-HT  5-CT > Suma > 8-OH-DPAT (as indicated in Table 1, > indicates significant differences at the P < 0.05 level and  indicates no significant difference and P > 0.05). In regard to E_max values in Main segments, again the normoxic and hypoxic adult groups exhibited identical patterns that were 5-HT > 5-CT > 8-OH-DPAT  Suma. In the normoxic fetus group, this order was similar: 5-HT > 5-CT  8-OH-DPAT  Suma (Table 2).

In contrast to the Com and Main segments, concentration-response curves for 5-HT, Suma, and 8-OH-DPAT were all right-shifted relative to those for 5-CT in 2BR artery segments (Fig. 1, Table 1). In this artery group, the relative order of agonist potencies was 5-CT > 5-HT > Suma > 8-OH-DPAT in normoxic adults, but was markedly different in both hypoxic adults and normoxic fetuses: 5-CT  5-HT  Suma > 8-OH-DPAT. In 2BR artery segments, the relative order of E_max values varied with experimental group and exhibited the order 5-HT > 5-CT  8-OH-DPAT  Suma in normoxic adults, but in hypoxic adults the order was 5-HT > Suma  5-CT  8-OH-DPAT, which was similar to that observed in normoxic fetuses: 5-HT > Suma  8-OH-DPAT  5-CT (Table 2).

For both 5-HT and 5-CT, agonist potencies increased progressively with the transition from Com to Main to 2BR in all experimental groups, although the magnitudes of these increases varied somewhat among the different groups. Similarly, Suma potencies also increased with the transition from Main to 2BR segments in all experimental groups. For 8-OH-DPAT, potencies did not vary consistently among artery types in the different experimental groups but tended to be least potent in Com segments and most potent in Main segments.

Antagonist affinities. As indicated in Fig. 2 and Table 3, ketanserin potently antagonized 5-HT-induced contractions, particularly in the Com segments where it exhibited subnanomolar affinities that did not vary significantly across the three experimental groups. Ketanserin pK_b values against 5-HT in the Com segments were also slightly but not significantly greater than those in the Main segments in all experimental groups. Correspondingly, ketanserin pK_b values within the Main segments did not vary with experimental group. In contrast, ketanserin pK_b values within the 2BR segments were 2.5- to 50-fold less than those in the Main segments in all three groups, and this difference was significant in normoxic and hypoxic adults but not in normoxic fetuses. Within the 2BR segments, ketanserin pK_b values were significantly greater in normoxic fetuses than in either adult group although the values within the adult groups did not differ significantly from one another.

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Effects of ketanserin, methiothepin, and GR-127935 on serotonin concentration-response relations. Shown are the effects of ketanserin, methiothepin, and GR-127935 on serotonin concentration-response relations for common carotid (A), main branch middle cerebral (B), and second branch middle cerebral (C) arteries taken from normoxic adult sheep. Concentrations of the antagonists used were adjusted for each artery type. Concentrations of ketanserin, methiothepin, and GR-127935 used in the common carotid segments were 1.0, 0.5, and 100 nM, respectively. For the main branch middle cerebral segments shown, corresponding concentrations were 2.8, 1.0, and 100 nM, respectively. For the second branch middle cerebral segments shown, concentrations were 200, 17, and 85 nM, respectively. All values are indicated as means ± SE, and vertical error bars indicate SEs for the numbers of animals indicated in Tables 1 and 3.

Against 5-CT, ketanserin produced the same general pattern of results as that produced against 5-HT (Fig. 3 and Table 3). Ketanserin pK_b values against 5-CT were in the subnanomolar range in all Com and Main segments, did not vary with experimental group in either artery type, and did not differ significantly between Com and Main segments within any experimental group. The ketanserin pK_b values against 5-CT in the 2BR segments were all significantly lower than observed in Com and Main segments by at least 100-fold in all experimental groups. Within the 2BR segments, ketanserin pK_b values against 5-CT did not vary with experimental group. It is important to note, however, that of eight experiments carried out to determine the pK_b for ketanserin against 5-CT in fetal second branch segments, only three of these exhibited any effect. Against Suma-induced contractions, ketanserin had no measurable effect in any artery type from any experimental group.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Effects of ketanserin and methiothepin on 5-CT concentration-response relations. Indicated are the effects of ketanserin and methiothepin on 5-CT concentration-response relations for common carotid (A), main branch middle cerebral (B), and second branch middle cerebral (C) arteries taken from normoxic adult sheep. Concentrations of the antagonists used were adjusted for each artery type. Concentrations of ketanserin and methiothepin used in the common carotid segments were 0.5 and 0.5 nM, respectively. For the main branch middle cerebral segments shown, corresponding concentrations were 2.7 and 2.7 nM, respectively. For the second branch middle cerebral segments shown, concentrations were 230 and 46 nM, respectively. All values are indicated as means ± SE, and vertical error bars indicate SEs for the numbers of animals indicated in Tables 1 and 3.

Methiothepin affinities against 5-HT also exhibited a pattern similar to that observed for ketanserin (Fig. 2 and Table 3). Methiothepin pK_b values against 5-HT were in the subnanomolar range in all Com and Main segments and did not differ significantly between Com and Main segments within any experimental group, although the values in the Main segments tended to be greater than those in the Com segments in all groups. Methiothepin pK_b values did not differ significantly among the experimental groups in Com segments, but in Main segments the values obtained in normoxic fetuses were slightly but significantly less than observed in the normoxic adults. Methiothepin pK_b values against 5-HT in the 2BR segments were all significantly less than in corresponding Main segments, and these differences averaged 25-fold in the normoxic adults, 8-fold in the hypoxic adults, but just 3-fold in the normoxic fetuses.

Against 5-CT, methiothepin affinities were more variable than observed against 5-HT (Fig. 3 and Table 3). In the Com segments, the pK_b values were again in the subnanomolar range and did not vary significantly among the different experimental groups. In the Main segments, the pK_b values were 5-fold less in normoxic adults, 3-fold greater in hypoxic adults, but 20-fold less in normoxic fetuses than observed in corresponding Com segments. Only the latter difference was statistically significant. Interestingly, methiothepin pK_b values against 5-CT did not differ significantly between Main and 2BR segments in any experimental group and did not vary with experimental group in either artery type.

Consistent with observations made in other preparations (6, 25), both ketanserin and methiothepin exhibited some noncompetitive character against 5-HT and 5-CT (Figs. 2 and 3). These effects were most evident for methiothepin against 5-HT and 5-CT in the main branch middle cerebral but were also apparent for ketanserin against 5-CT in the main branch middle cerebral. In light of these effects, all calculations of pK_b values were not based on pD₂ shifts, which assume no effect on maximum response, but instead were based on dose-ratio analyses, which make no assumptions about possible effects on maximum response. These noncompetitive effects were not apparent in all arteries and were completely absent for the antagonists GR-127935 and SB-206553.

For the antagonist GR-127935, pK_b values against 5-HT were low and ranged from 7.1 to 7.7 across all experimental groups (Figs. 2 and 4). These values did not differ significantly between artery types in any experimental group and also did not vary significantly with experimental group within any artery type. Against Suma, GR-127935 pK_b values were somewhat greater and ranged from 7.5 to 8.5. However, again these values did not differ significantly between artery types in any experimental group and also did not vary significantly with experimental group within any artery type.

View larger version (19K): [in this window] [in a new window]   Fig. 4.   Effects of GR-127935 on Suma concentration-response relations. Shown are the effects of GR-127935 on sumatriptan concentration-response relations for main branch middle cerebral (A) and second branch middle cerebral (B) arteries taken from normoxic adult sheep. Concentrations of GR-127935 used were adjusted for each artery type and were 92 and 49 nM for the main branch middle cerebral and second branch middle cerebral segments, respectively. All values are indicated as means ± SE, and vertical error bars indicate SEs for the numbers of animals indicated in Tables 1 and 3.

In the Com segments, the antagonist SB-206553 only weakly antagonized 5-HT-induced contractions, with pK_b values averaging from 7.2 to 7.6 (Table 3). These values did not differ significantly among the three experimental groups.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

The present experiments address the general hypothesis that changes in reactivity to 5-HT associated with maturation, acclimatization to chronic hypoxia, and differences in artery type involve differences in the receptor subtype mediating vasoconstrictor responses to 5-HT. As indicated by measurements of both relative agonist potency and antagonist affinity, the data suggest that 1) the receptors mediating responses to 5-HT changed from predominantly the 5-HT_2a subtype in large Com arteries to a mixed 5-HT₁/5-HT₂ population in Main arteries to a predominantly 5-HT₁ population in second branch segments of the middle cerebral arteries of adult sheep; 2) in small cerebral arteries the receptor mediating responses to 5-HT changed from a mixed 5-HT₂/5-HT₁-like population in term fetuses to a predominantly 5-HT₁ population in adults; and 3) chronic hypoxia has little if any effect on 5-HT receptor type in adult carotid and cerebral arteries.

Serotonergic receptor type in ovine Com. In Com segments from all experimental groups, 5-HT was a more potent contractile agonist than 5-CT, 8-OH-DPAT, or Suma (Table 1). In addition, maximum contractile responses in the Com were far greater for 5-HT than any other agonist tested in all experimental groups (Table 2). Together, these data are consistent with the presence of the 5-HT₂ receptor type according to the criteria of Hoyer et al. (18, 19). The absence of any contractile response to Suma, an agonist at the 1B, 1D, and 1F subtypes (2, 18, 19), suggests that these subtypes may not be present in the ovine Com. The low potency of 8-OH-DPAT further argues against the presence of the 1A subtype in Com segments, particularly in normoxic and hypoxic adults.

The subnanomolar affinities observed for ketanserin against 5-HT in Com segments (Table 3) also suggest the presence of the 5-HT₂ receptor type (18, 19). Given that ketanserin generally exhibits near micromolar affinities against most other serotonin receptors, the high affinity of ketanserin against 5-CT observed in Com arteries suggests that these responses to 5-CT were mediated by action on 5-HT₂ receptors. Consistent with this interpretation, 5-CT exhibited low potency in the present experiments (pD₂  5.3) and in other experiments has been shown to activate 5-HT₂ receptors with low affinity (pK_i  4.7) (18).

Whereas ketanserin exhibits high affinity toward the 5-HT₂ receptor, the antagonist GR-127935 exhibits much lower affinities (pK_i  7) toward this receptor but much higher affinities toward the 5-HT_1B and 5-HT_1D subtypes (pK_i  9-10) (36). In the present study, the affinities (pK_b) observed for GR-127935 against 5-HT ranged from 7.1 to 7.5 across all experimental groups, thus again arguing in support of the presence of the 5-HT₂ and against the 5-HT_1B and 5-HT_1D receptor types in ovine Coms.

The methiothepin affinities observed were also consistent with the presence of the 5-HT₂ receptor type in the ovine Com. Similar to ketanserin, methiothepin generally exhibits near nanomolar affinities for the 5-HT₂ receptor, but unlike ketanserin, methiothepin also exhibits submicromolar affinities (7.7-8.1) against most 5-HT₁ receptor types (18). The high pK_b values observed for methiothepin against 5-HT (9.2-9.5) and also against 5-CT (9.1-9.9) are consistent with the presence of the 5-HT₂ receptor type.

Given that the 5-HT₂ family includes three different subtypes (2A, 2B, and 2C), we used the antagonist SB-206553 to help determine which of these subtypes were present. As recently shown by Kennett et al. (21), this antagonist has 50- to 100-fold greater affinity toward the 2B and 2C subtypes (pK_i  9) than for the 2A subtype (pK_i  7.3). Thus the relatively low pK_b values (7.2-7.6) obtained for SB-206553 against 5-HT in the Coms argue against the presence of the 2B and 2C subtypes. Additional support for this interpretation is provided by the finding that mRNA for the 2C subtype is generally not found in extracranial tissues (1, 39). Furthermore, the affinity for ketanserin against the 2B subtype is also much lower (pK_i  5-6) (22) than observed in the present studies (pK_b  9). Together, the combined agonist and antagonist data strongly suggest that contractile responses to 5-HT in ovine Com arteries are mediated primarily by the 5-HT_2A receptor subtype.

Serotonergic receptor type in ovine Main. In the Main segments, the relative order of agonist potencies was similar to that observed in the Com segments, with the main exception that Suma produced significant contractile responses, suggesting the presence of an additional receptor type (Table 1), although the order of agonist potency was consistent with the predominance of the 5-HT₂ receptor type in Main segments (18). Another indication that a second receptor type might be involved was provided by the observation that 5-CT potency was at least eightfold greater in Main segments than in Com segments, across all experimental groups (Table 1). The potencies observed for 5-CT (pD₂  6.2-6.9), however, were still much lower than typically observed for 5-CT at most 5-HT₁ type receptors (7.9-8.6) (18), a result that might be expected if 5-CT were acting simultaneously with high affinity on 5-HT₁ receptors and with low affinity on 5-HT₂ receptors.

Given that the 5-HT_1A receptor is rarely expressed in arterial smooth muscle (18, 39) and that the potency observed for 8-OH-DPAT (pD₂  5.6-5.7) was markedly less than typical of interaction at 1A receptors (pD₂  8.2), but was consistent with interaction at other 5-HT_1B, 5-HT_1D, or 5-HT₂ receptors (pD₂  5-7) (18), the data were more consistent with the presence of the 1B or 1D subtype than with the presence of the 1A subtype. To evaluate the possible presence of the 1B/1D subtypes, pK_b values were measured for the antagonist GR-127935 against both 5-HT- and Suma-induced contractions. Against 5-HT, these pK_b values ranged from 7.4 to 7.7, values much lower than typical of interaction of GR-127935 with 1B or 1D receptors (8.5-10) but were typical of interaction with the 5-HT_2A subtype (pK_i  7.4) (36). Against Suma, however, the affinities for GR-127935 were significantly greater (8.1-8.5) than against 5-HT but were still too low to suggest the presence of the human 1B (pK_i  9.9) or 1D (pK_i  8.9) receptor subtypes (36). Alternatively, the range of values observed was consistent with the presence of the rat 1B receptor type (pK_i  8.5) (36). In further support of this interpretation, the potencies observed for Suma (pD₂  5.9-6.4) were also not typical of interaction with the 1D or h1B subtypes (pD₂  8.1-8.4) but were compatible with the presence of the r1B subtype (pD₂  6.3) (30). Previous studies have also concluded that the 5-HT_1B subtype is present in both bovine and human middle cerebral arteries (14). Thus the combined agonist and antagonist data are consistent with the interpretation that contractile responses to 5-HT in ovine Main arteries are mediated by a mixed population of 5-HT₁ (possibly the r1B) and 5-HT₂ receptors as previously suggested by Gaw et al. (13), and by Hamel et al. (15) for cat cerebral arteries and by Parsons and Whalley (31) for the rabbit basilar arteries.

Serotonergic receptor type in ovine 2BR. In contrast to the Com and Main arteries, 2BR arteries from normoxic adults contracted with greatest sensitivity to 5-CT, and their relative order of agonist potencies was 5-CT > 5-HT > Suma > 8-OH-DPAT (Table 1). Although the results of statistical comparisons among the different agonists varied somewhat with experimental group, the same basic order of agonist potencies was observed in all groups. This order suggests that the 5-HT₁ type was predominant in 2BR arteries. Consistent with this interpretation, ketanserin affinities (Table 3) against 5-HT (pK_b = 7.4) and 5-CT (pK_b  7.2) in the normoxic adult second branch arteries were markedly less than typical of interaction at a 5-HT₂ receptor (pK_b  9) and more consistent with interaction at either a 5-HT_1D (pK_i  7.4) (44) or a mixed population of 5-HT_1B (pK_d  5.7) or 5-HT_1A (pK_d  5.9) and 5-HT₂ subtypes (17). In addition, the mixed 5-HT₁/5-HT₂ antagonist methiothepin exhibited 16-fold greater affinity than ketanserin against 5-HT and 8-fold greater affinity against 5-CT. The methiothepin affinities observed against 5-HT (pK_b = 8.6) and 5-CT (8.1), however, were much greater than their values at the 1A, 1B, or 1D subtypes (pK_d  7.1-7.3) and slightly less than their affinity against the 5-HT₂ subtype (pK_d  8.8). Together, these data suggest the possible presence of a mixed 5-HT₁/5-HT₂ receptor population with predominance of the 5-HT₁ subtype.

To help evaluate which 5-HT₁ subtype was involved in 5-HT-induced contractions of the 2BR arteries, we determined pK_b values for the antagonist GR-127935. The pK_b values obtained in second branch preparations from normoxic adults against 5-HT (7.7) were much lower than typical of interactions with the r1B (pK_i  8.5), h1B (pK_i  9.9), or 1D (pK_i  8.9) receptors and were much more typical of interaction with the 5-HT_2A (pK_i  7.4) (36), again suggesting the possible presence of 5-HT₂ receptors. Interestingly, the pK_b values obtained for GR-127935 against Suma (7.7) were identical to those obtained against 5-HT. Given that Suma has extremely low affinity toward the 5-HT₂ receptor (pK_a  3.7-4.1) (35), it is unlikely that the contractions induced by Suma (pD₂  6.6) involved 5-HT₂ receptors, and thus it seems unlikely that the low pK_b value obtained for GR-127935 against Suma (7.7) was due to involvement of 5-HT₂ receptors. Given that the apparent affinity of GR-127935 was the same against 5-HT and Suma, it remains possible that 5-HT and Suma activated the same receptor, as previously proposed by Parsons and Whalley (31) based on similar studies in the rabbit basilar artery. Although the nature of the 5-HT₁ receptor involved in the present studies remains uncertain, it is unlikely to be the 1A given the low potency (pD₂  5.6) exhibited in 2BRs. In turn, the potency exhibited by Suma in the second branch preparations (pD₂  6.6) was more consistent with interaction at the r1B subtype (pD₂  6.3) than with the 1D or h1B subtypes (pD₂  8.1-8.4) (30). In either event, the overall data support the view that contractile responses to 5-HT in 2BR arteries from normoxic adults are mediated predominantly by 5-HT₁ receptors (possibly the r1B), with a possible contribution from a less dominant population of 5-HT₂ receptors.

Effects of hypoxic acclimatization and maturation on 5-HT receptor type. A recurrent feature of many studies of the effects of high-altitude acclimatization include significant alterations in the reactivity of many vascular beds, a response we have demonstrated in the ovine cerebral circulation (26). Because receptor type is a key determinant of overall vasoreactivity, it seems reasonable to suspect that the receptor type mediating responses to 5-HT might change during high-altitude acclimatization. The present data, however, argue against this possibility. Aside from the typical (26) significant generalized increases in E_max that we observed in the Coms and Mains in hypoxic compared with normoxic adults (Table 2), there were no major differences in either agonist potencies (Table 1) or antagonist affinities (Table 3) between normoxic and hypoxic adults. Together these findings suggest that the changes in vascular reactivity associated with hypoxic acclimatization probably do not involve changes in receptor type, but rather changes in the coupling of the receptors to the contractile apparatus.

Compared with the general lack of effect of hypoxic acclimatization on 5-HT receptor type, maturation had a somewhat greater effect, which was most pronounced in the 2BR artery segments. In the normoxic fetus, the relative order of agonist potencies was essentially the same in the second branch segments as observed in the normoxic adult (Table 1), although maximum responses were somewhat depressed, particularly for 5-CT and 8-OH-DPAT. More importantly, ketanserin affinities against 5-HT were 25-fold greater in the fetus (pK_b  8.8) than in the adult (pK_b  7.4), suggesting much greater 5-HT₂ character in the fetal second branch arteries (Table 3). Against 5-CT, ketanserin affinities were also much less in the fetus (pK_b  6.7) than in the adult (pK_b  7.2), although this difference was not significant. It is important to note, however, that the majority (5 of 8) of attempts to determine the ketanserin pK_b against 5-CT yielded no measurable effect, suggesting that the receptor type at which 5-CT and ketanserin were interacting in the adult segments was either absent or present only in reduced abundance in the fetus. Despite this difference, no significant age-related differences in pK_b values were observed for any other antagonist in the second branch segments. Overall, the combined agonist and antagonist data suggest that in the normoxic fetus, as in the adult, contractile responses to 5-HT in the 2BR are mediated by a mixed population of 5-HT₁ (possibly the r1B) and 5-HT₂ receptors and the relative proportion of 5-HT₂ receptors is greater in the fetus than in the adult.

In the Main segments from the fetus, the relative order of agonist potencies was identical to that in the adult (Table 1), and the only significant difference in antagonist affinities was that methiothepin affinity against 5-HT was depressed in the fetus (pK_b  9.3) relative to the adult (pK_b  10.0). In the Com arteries, there were no age-related differences in antagonist affinities (Table 3), but agonist potencies (Table 1) and maximum responses (Table 2) tended to be generally greater in the fetus than the adult, reflecting the well-established influence of maturation on vasoreactivity (16, 33). These minor age-related differences not withstanding, the overall distribution of serotonergic receptor subtypes was therefore basically the same in the fetus and adult, with the exception that the second branch arteries exhibited greater 5-HT₂ character in the fetus.

Physiological significance of shifts in serotonergic receptor type. The present suggestion of a transition from the 5-HT_2a subtype in the larger and more proximal Com, to a mixed 5-HT1/5-HT2 population with 5-HT2 predominance in the Main segments, to a mixed 5-HT1/5-HT2 population with 5-HT1 predominance in the 2BR is similar to patterns suggested in other vascular beds. In rabbit aorta (9) and carotid artery (43), 5-HT-induced contractions are mediated by the 5-HT₂ receptor, whereas in the smaller mesenteric (42) and basilar arteries (31), a mixed receptor population of 5-HT₂ and 5-HT₁ receptors mediates responses to 5-HT. At the level of the rabbit middle cerebral artery, the responses to 5-HT are mediated predominantly by the 5-HT₁-like and/or 5-HT_1d receptor subtypes (8). In addition, the distribution of -adrenergic receptor subtypes has also been shown to vary with artery diameter and tends to shift from the ₁-subtype in larger arteries to a mixed ₁- and ₂-population in intermediately sized arteries, to predominantly the ₂-subtype in terminal arterioles (23, 24). The present findings are thus consistent with a broad variety of evidence suggesting an artery size-related shift in predominant receptor subtype.

A key difference among the characteristics of the receptors mediating contractile responses to 5-HT in small and large arteries is the signal transduction pathway involved. The 5-HT₂ receptor involved in large arteries is coupled to the synthesis and release of inositol trisphosphate, which in turn stimulates the release of intracellular calcium (34). The same is true of the ₁-receptor type also predominant in large arteries. In contrast, the 5-HT₁ receptors present in small arteries are coupled to decreases in cAMP, membrane depolarization, and subsequent entry of calcium from the extracellular space (12, 37). Again, the ₂-receptor predominant in small arteries also uses a similar signal transduction pathway. Thus the receptor type in large arteries most typically stimulates the release of intracellular calcium, whereas the receptor type in small arteries typically stimulates the entry of extracellular calcium. Correspondingly, intracellular stores of calcium are generally much greater and well developed in large than in small arteries (11). Certainly, other factors related to artery size and type, such as differences in electrophysiological characteristics (29), calcium sensitivity (4), and cyclic nucleotide metabolism (32) may also in some way influence and/or reflect different receptor distribution among arteries of differing type and size. Nonetheless, the 5-HT receptor type predominant in ovine cranial arteries appears closely related to the relative dependence on intracellular release, compared with extracelluar influx, as a source of activator calcium.

In parallel with the effects of changing artery size and type on receptor type, maturation further involves a generalized decreased reliance on extracellular calcium entry and an increased reliance on intracellular calcium release in many tissues (45). This general finding would predict that maturation would thus also involve a shift from the 5-HT₁-like receptor type, which depends more on extracellular calcium entry, to the 5-HT₂ type, which depends more on intracellular release. The fact that this shift was not observed, however, suggests at least three possibilities. First, the ability of Com and Main arteries to store and release intracellular calcium may be sufficient to mediate contractile responses to 5-HT in the term fetus. A second possibility is that the generally greater receptor densities typical of many receptor types in immature cerebral arteries (27) may in some way compensate for a decreased ability to release intracellular calcium. Third, the diacylglycerol released in tandem with inositol trisphosphate in response to 5-HT₂ stimulation may play a more significant role in increasing extracellular entry (40, 41) and or increased calcium sensitivity (3) in the fetal compared with adult arteries. Consistent with this possibility, protein kinase C activity, which is stimulated by diacylglycerol, is generally greater in fetal than adult vascular tissues (7). At present, the relative importance of each of these possibilities remains uncertain, particularly in the 2BR artery segments where maturation appeared to decrease the presence of 5-HT₂ receptors. Clearly, additional experiments will be required to differentiate among these possibilities.

Perspectives