INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

THE EFFECTS OF HYPERTHERMIA on the cardiovascular system are little known. Hyperthermia produced by acute heating in rats induces an homeostatic response characterized by increased arterial blood pressure, heart rate, and regional vascular resistance (5, 8, 9). Failure of this adaptation may conduce the pathological condition known as heatstroke, which is characterized by marked hypotension and cardiovascular shock (13), but the pathogenesis of this disorder remains unclear. One possibility is that sympathetic activity or sympathetic vascular reactivity is altered. There is evidence that during heat stress an increased activity of perivascular sympathetic fibers (4, 6) and raised plasmatic catecholamine levels occur (4), which probably contribute to maintain arterial pressure homeostasis by producing vasoconstriction. However, it is unclear whether reactivity of blood vessels is altered, as there are conflicting reports that elevated temperature increases (14), does not modify (16), or decreases (8, 18) the vasoconstriction to norepinephrine. Other vascular regulatory mechanisms, involving the endothelium, may be also functioning during hyperthermia. This hypothesis is based on data indicating endothelial cell damage (17) and increased plasma levels of endothelin-1 (1) in heatstroke patients, as well as excessive production of nitric oxide during hyperthermia, possibly due to endotoxemia and induction of nitric oxide synthase (6). However, more studies are required to elucidate the role of the endothelium in the vascular effects of hyperthermia.

The purpose of this study was to assess the effect of hyperthermia on vascular reactivity, analyzing the role of the endothelium in this effect. We hypothesize that hyperthermia may increase vasoconstriction and/or reduce vasodilatation in some vascular beds by altering the release of endothelial vasoactive factors. This effect of hyperthermia would act as a homeostatic mechanism aiming to maintain systemic arterial pressure. To analyze this hypothesis, we used isolated femoral arteries from rabbits to examine the constriction to potassium, to norepinephrine, and to endothelin-1, as well as the relaxation to the nitric oxide donor sodium nitroprusside, the nitric oxide-dependent relaxant acetylcholine, and the nitric oxide-independent relaxant adenosine. These experiments were performed in the arteries exposed to normotemperature (37°C) and moderate (41°C) and severe (44°C) hyperthermia. The use of in vitro preparations may be useful in approaching this issue as they avoid complicating factors that may arise from in vivo experiments (e.g., desensitization of adrenoceptors because of prolonged sympathetic stimulation). Femoral artery was selected as it is a muscular artery, and it has been hypothesized that muscular vasculature is a potential target of vasoconstriction as part of the integrated pattern of blood flow redistribution during heat stress (15).

	METHODS

Top Abstract Introduction Methods Results Discussion References

Seventy-seven New Zealand White rabbits, weighing 2-2.5 kg, were killed by intravenous injection of pentobarbital sodium (100 mg/kg). Femoral arteries (OD after dissection = 1-1.3 mm) were dissected free and cut into cylindrical segments 2 mm in length. Each segment was prepared for isometric tension recording in a 4-ml organ bath containing modified Krebs-Henseleit solution with the following composition (in mM): 115 NaCl, 4.6 KCl, 1.2 KH₂PO₄, 1.2 MgSO₄, 2.5 CaCl₂, 25 NaHCO₃, and 11.1 glucose. The solution was equilibrated with 95% O₂-5% CO₂ to give a pH of 7.3-7.4, which was measured with a pH meter (micropH 2001, Crison Instruments). Briefly, the method consists of passing two fine, stainless steel pins, 150 µm in diameter, through the lumen of the vascular segment. One pin is fixed to the organ bath wall, whereas the other is connected to a strain gauge for isometric tension recording, thus permitting the application of passive tension in a plane perpendicular to the long axis of the vascular cylinder. The recording system included a universal transducing cell (UC3, Statham Instruments), a Statham microscale accessory (UL5, Statham Instruments), and a Beckman type RS recorder (model R-411, Beckman Instruments). The optimal passive tension was determined in preliminary experiments by recording the contraction to potassium chloride (10¹ M) after applying, at random sequence, different passive tensions (0.1, 0.25, 0.5, 1, 2, and 5 g) at 37 (20 segments), 41 (10 segments), and 44°C (10 segments). In these experiments it was found that the maximal response was obtained at a passive tension of 0.5 g at the three temperatures studied. Therefore, the experiments were performed in vascular segments stretched to this optimal passive tension of 0.5 and equilibrated for 60-90 min before any drug was added, renovating the solution in the baths every 30 min. In a group of experiments (n = 7) designed to study the effect of temperature on the basal tone of the arteries, the temperature was set from the beginning at 37, 41, or 44°C, and, after the equilibration period, the temperature was changed from 37 to 41 or 44°C, or from 41 or 44 to 37°C. In the experiments performed to study the response to vasoconstrictor or vasodilator stimuli, the temperature of the bath was adjusted from the beginning of the experiment at 37, 41, or at 44°C, and the arteries remained at the chosen temperature throughout the duration of the experiment.

The contraction to potassium chloride (5 × 10³ to 5 × 10² M), norepinephrine (10⁹ to 10⁴ M), and endothelin-1 (10¹¹ to 10⁷ M) was studied in the arteries under control conditions, after removal of endothelium, and after treatment with the inhibitor of nitric oxide synthase N-nitro-L-arginine (L-NNA; 10⁴ M), at 37, 41, and 44°C. The response to the specific ₁-adrenergic agonist phenylephrine (10⁹ to 10⁴ M), the specific ₂-adrenergic agonist BHT-920 (10⁹ to 10⁴ M), and the specific endothelin ET_B-agonist IRL-1620 (10¹¹ to 10⁷ M) was also analyzed in control conditions at these temperatures. In addition, the contraction to endothelin-1 was studied in the presence of the antagonist of endothelin ET_A receptors cyclo D--aspartyl-L-propyl-D-valyl-L-leucyl-D-tryptophyl (BQ-123; 10⁵ M) and the antagonist of ET_B receptors N-[N-[N-[(2,6-dimethyl-1-piperidinyl) carbonyl]-4-methyl-L-leucyl]-1-(methoxycarbonyl)-D-tryptophyl]-D-norleucine monosodium (BQ-788; 10⁵ M), the response to phenylephrine in the presence of the ₂-adrenergic antagonist yohimbine (10⁶ M), and the response to BHT-920 in the presence of the ₁-adrenergic antagonist prazosin (10⁶ M) at the three temperatures indicated.

The relaxation to acetylcholine (10⁸ to 10⁴ M), sodium nitroprusside (10⁸ to 10⁴ M), or adenosine (10⁸ to 10⁴ M) was was studied in segments precontracted with endothelin-1 (10⁸ to 3 × 10⁸ M) at 37, 41, and 44°C. The relaxation to acetylcholine and adenosine was also recorded in the presence of L-NNA (10⁴ M).

The agonists were added to the organ bath in a cumulative manner, and L-NNA, BQ-123, BQ-788, yohimbine, or prazosin was added to the bath 20 min before the concentration-response curves for the corresponding agonists was begun. Removal of the endothelium was accomplished by gently rubbing the vascular lumen with a steel rod, and the adequacy of the procedure was tested by abolition of the relaxing response to acetylcholine (10⁶ M) in the arteries precontracted with endothelin-1 (10⁸ to 3 × 10⁸ M).

EC₅₀ values were calculated as the concentration producing 50% of the maximal effect by arithmetic (for potassium curves) or geometric (for the rest of the agonists) interpolation. The results were expressed as the pD₂ (log EC₅₀).

The values of the contraction are shown as absolute values, the relaxation values are shown as percentage of the active tone achieved with endothelin-1, and both responses are expressed as means ± SE. Data were evaluated by a two-way analysis of variance applied to each group of data, followed by one-way analysis of variance at each agonist concentration and Dunnett's test to compare each experimental condition with its control. P < 0.05 was considered significant.

Drugs used were the following: acetylcholine chloride, L-NNA, norepinephrine ()arterenol (bitartrate salt), prazosin hydrochloride, and yohimbine hydrochloride (from Sigma Chemical); BQ-123 (free base), BQ-788, and endothelin-1 (8-21), N-Suc-[Glu⁸, Ala^11,15] free base (IRL-1620) from Research Biochemicals International; adenosine free base from Carlo Erba (Italy); endothelin-1 (human, porcine) from Peninsula Laboratories; sodium nitroprusside (Nitroprussiat Fides) from Fides-Rottapharm; and 5-allyl-2-amino-5,6,7,8-tetrahydro-⁴H-thiazolo-[4,5]-dazepin hydrochloride (BHT-920), a gift from Europharma.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Effect of temperature changes on basal tone. At the basal tension of 0.5 g, warming from 37 to 41 or 44°C, or cooling from 41 or 44 to 37°C, did not induce any change in the tension of the vascular segments.

Contraction to potassium chloride. Potassium chloride (5 × 10³ to 5 × 10² M) produced concentration-dependent contraction of femoral arteries at every temperature studied. Figure 1 summarizes the results with potassium chloride, and the corresponding pD₂ values are shown in Table 1. At 41 and 44°C, both the sensitivity and maximal effect were significantly higher than at 37°C (Fig. 1A). At 37°C endothelium removal increased the sensitivity, and pretreatment with L-NNA increased both the sensitivity and maximal effect, compared with intact arteries (Fig. 1B). At 41 or 44°C, neither endothelium removal nor L-NNA pretreatment modified the contraction to potassium (Fig. 1, C and D).

View larger version (31K): [in this window] [in a new window]   Fig. 1.   Contraction of rabbit femoral arteries to potassium chloride at 37, 41, and 44°C (A) and after endothelium removal and pretreatment with N-nitro-L-arginine (L-NNA) at 37 (B), 41 (C), and 44°C (D). Values are means ± SE; n = 5 animals. Statistically significant compared with 37°C: * P < 0.05; ** P < 0.01. Statistically significant compared with control at same temperature:  P < 0.05; P < 0.01

Contraction to adrenergic stimulation. Norepinephrine produced a concentration-dependent contraction of femoral arteries, and the sensitivity and maximal contraction were significantly higher at 41 and 44°C than at 37°C (Fig. 2A). Compared with control arteries, endothelium removal or treatment with L-NNA did not modify significantly the contraction to norepinephrine at any of the temperatures studied (Table 1).

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Contraction of control rabbit femoral arteries to norepinephrine (A; n = 5-6 animals), phenylephrine (B; n = 7 animals), and BHT-920 (C; n = 7 animals) at 37, 41, and 44°C. Values are means ± SE. Statistically significant compared with 37°C: * P < 0.05; ** P < 0.01.

6

4

4

Contraction to endothelin-1. Endothelin-1 (10¹¹ to 10⁷ M) contracted femoral vascular segments in a concentration-dependent way, and the contraction for 10¹¹ to 3 × 10¹⁰ concentrations was significantly higher at 41 and 44°C compared with 37°C (Fig. 3A). Compared with control arteries, endothelium removal or treatment with L-NNA increased the response to endothelin-1 at 37°C (Fig. 3B), but not at 41 (Fig. 3C) or 44°C (Fig. 3D).

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Contraction of rabbit femoral arteries to endothelin-1 at 37, 41, and 44°C (A) and after endothelium removal and pretreatment with L-NNA at 37 (B), 41 (C), and 44°C (D). Values are means ± SE; n = 7 animals. * Statistically significant (P < 0.05) compared with 37°C.  Statistically significant (P < 0.05) compared with control at same temperature.

View larger version (28K): [in this window] [in a new window]   Fig. 4.   Contraction of rabbit femoral arteries to endothelin-1 at 37 (A), 41 (B), and 44°C (C) in arteries nontreated or treated with endothelin ETA-receptor antagonist BQ-123 or with ETB-receptor antagonist BQ-788 (n = 7 animals). D: contraction of rabbit femoral arteries to IRL-1620 at 37, 41, and 44°C (n = 7 animals). Values are means ± SE. Statistically significant compared with control: * P < 0.05; ** P < 0.01.

Relaxation to acetylcholine, sodium nitroprusside, and adenosine. The level of active tone induced with endothelin-1 10⁸ to 3 × 10⁸ M (3.4 ± 0.1 g) was not different in any of the experimental conditions studied. In these precontracted vascular segments, acetylcholine (10⁸ to 10⁴ M) (Fig. 5), sodium nitroprusside (10⁸ to 10⁴ M) (Fig. 6), or adenosine (10⁸ to 10⁴ M) (not shown) produced concentration-dependent relaxation. Table 2 shows the pD₂ values for these vasodilators at the three temperatures studied. The sensitivity to sodium nitroprusside was significantly higher at 41 or 44°C compared with 37°C, whereas the sensitivity to acetylcholine or adenosine was not significantly different at any of the temperatures studied. Maximal relaxation, relative to the tension attained with endothelin-1, at 37, 41, and 44°C was, respectively, 82 ± 4, 91 ± 7, and 90 ± 6% for acetylcholine; 93 ± 1.4, 98 ± 1.4, and 96 ± 1.4% for sodium nitroprusside; and 73 ± 10, 80 ± 17, and 88 ± 3% for adenosine, and it was not significantly different between the temperatures studied. L-NNA treatment partially reduced the relaxation to acetylcholine in a similar degree at all the temperatures studied (Fig. 5), and it did not modify the relaxation to adenosine (not shown).

View larger version (30K): [in this window] [in a new window]   Fig. 5.   Relaxation of precontracted rabbit femoral arteries to acetylcholine in control arteries at 37, 41, and 44°C (A) and after L-NNA treatment at 37, (B), 41 (C), and 44°C (D). Values are means ± SE (n = 7 animals). Statistically significant compared with control at same temperature: * P < 0.05; ** P < 0.01.

View larger version (25K): [in this window] [in a new window]   Fig. 6.   Relaxation of precontracted rabbit femoral arteries to sodium nitroprusside at 37, 41, and 44°C. Values are means ± SE; n = 5 animals. Statistically significant compared with 37°C: * P < 0.05; ** P < 0.01.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The purpose of this study was to analyze the effects of hyperthermia on the constrictor and dilator responses of a muscular artery. Our results with sodium nitroprusside suggest that the sensitivity of the vascular smooth muscle to exogenous nitric oxide is increased during hyperthermia, and this effect may be specific for nitric oxide because the relaxation to adenosine was not modified by warming. The observations with sodium nitroprusside apparently contrast with those found with a substance that produces relaxation by releasing nitric oxide, such as acetylcholine, for which no increase in sensitivity was observed during warming. These results, however, might be reconciled if we suppose that the release of nitric oxide during cholinergic stimulation is reduced during warming, and this is compensated for by the increased sensitivity of the smooth muscle to nitric oxide. The reduction in nitric oxide production might be due to functional impairment of endothelial cells and/or of nitric oxide synthase, produced by elevated temperature. We have also found that warming from 37 to 41 or 44°C, or cooling to 37°C in segments previously warmed at 41 or 44°C, produced no changes in the basal tension of the arteries. This may be related to the lack of active tone in our vascular preparations because inhibition of the release or effects of nitric oxide may fail to induce vasoconstriction in the absence of vasoconstrictor tone (19).

On the other hand, the results with the vasoconstrictors studied suggest that during hyperthermia the arterial contraction is increased. Because cooling may inhibit and warming may enhance Ca²⁺ influx in rat vascular smooth muscle (20), a facilitation of Ca²⁺ entry may underlie, in part, the increased arterial contraction to potassium found during hyperthermia in the present study. This suggestion is in line with the finding that hyperthermia in the anesthetized rat increases renal vasoconstriction to membrane Ca²⁺ channels opening with BaCl₂ (8).

Our results with norepinephrine and phenylephrine suggest that warming may increase the contraction to activation of ₁-adrenoceptors, which may be predominant in the adrenergic contraction of rabbit femoral arteries (3). Although BHT-920 produced some contraction in this preparation, this response may be due to unspecific activation of ₁-adrenoceptors by this agonist because this response was reduced by prazosin. The small contraction to BHT-920 in the presence of prazosin was not modified by warming, further supporting the hypothesis that hyperthermia facilitates the response to activation of ₁-, but not ₂-adrenoceptors.

Regarding the effects of endothelin-1, we found that the endothelin ET_A-antagonist BQ-123 shifted to the right the concentration-response curve to this peptide, suggesting that the response to endothelin-1 is mediated mainly by ET_A receptors in our preparation. A smaller participation of endothelin ET_B receptors may also be present because the endothelin ET_B-antagonist BQ-788 inhibited slightly the contraction to endothelin-1, and the endothelin ET_B agonist IRL-1620 produced a small contraction. During hyperthermia, the contraction induced by low concentrations of endothelin-1, but not that by IRL-1620, was increased, and the blockade produced by BQ-123, but not that by BQ-788, was also greater during hyperthermia than at 37°C. These observations suggest that hyperthermia may increase the sensitivity and/or effects of endothelin ET_A receptors but not those of endothelin ET_B receptors. This phenomenon may be due to an increased sensitivity and/or concentration of endothelin ET_A receptors or to a facilitation of postreceptor mechanisms by warming.

Interestingly, although hyperthermia increased the vasoconstrictor response, the characteristics of this increase varied depending on the vasoconstrictor used. In the case of phenylephrine, hyperthermia increased sensitivity (pD₂ values) but not maximal contraction, whereas with potassium and norepinephrine both the sensitivity and the maximal effect were significantly increased by hyperthermia. The reason for these differences is not apparent, but it might be related to the existence of different levels of receptor reserve for the different vasoconstrictors. It may be hypothesized that a lower receptor efficacy at 37°C compared with hyperthermia would result in lower maximal effects for those agonists with no receptor reserve and a parallel shift for those with high receptor reserve.

At 37°C, inhibition of nitric oxide synthesis or endothelium removal increased the response of the rabbit femoral artery to potassium and endothelin-1 but did not modify significantly the response to norepinephrine. Thus, at normal temperature, the vascular contraction to potassium and endothelin-1 may be modulated by endothelial nitric oxide, and this phenomenon may not occur in the contraction to norepinephrine. Although the response to norepinephrine has been shown to be modulated by nitric oxide (10), this phenomenon was not observed in our results, which might be due to absence of ₂-adrenoceptors in our experimental preparation. During warming at 41 or 44°C, inhibition of nitric oxide synthesis did not modify the increased contraction to potassium, endothelin-1, and norepinephrine, thus suggesting that during hyperthermia the response to these vasoconstrictors is not modulated by nitric oxide. This suggests that hyperthermia by itself may inhibit the release of nitric oxide in response to potassium and endothelin-1 and might explain, at least in part, the elevated vasoconstriction to these two substances during warming. However, the increased response to norepinephrine may be related to mechanisms distinct to changes in release of nitric oxide.

There is little information in the literature concerning the effect of temperature changes in vascular production of nitric oxide. Hall et al. (6) have found in rats that hyperthermia increases blood levels of nitric oxide, probably due to endotoxemia resulting from damage of intestinal mucosa. Ryan and Gisolfi (16) observed in rat mesenteric arteries that the contraction to norepinephrine and the relaxation to acetylcholine were not significantly modified during exposure of these arteries to 42 and 43°C. On the other hand, in rat cremaster muscle arterioles, the relaxation to acetylcholine and sodium nitroprusside is reduced during local heat treatment, suggesting that warming can change production and efficacy of nitric oxide in this vascular bed (11). Our results may suggest a relatively reduced release of nitric oxide during warming, which may be compensated for in part by an increased sensitivity of smooth muscle to nitric oxide. Changes in temperature might affect the production of nitric oxide in a different way depending on vascular beds. In this sense, we have previously reported that cooling decreases the production of nitric oxide in rabbit femoral arteries, whereas it increases its production in rabbit ear arteries (2, 12).

Heatstroke is characterized by profound hypotension and cardiovascular failure (13), and it has been found that during heatstroke in rats there is a marked vasodilatation in the splanchnic circulation when central temperature increases above 41-42°C, a vasodilatation that may be responsible for the fall in arterial pressure (9). This hypotension might be due, at least in part, to increased sensitivity in some vascular beds to nitric oxide, as suggested by the present results in the rabbit femoral artery, where warming increased the vasodilatation to exogenous nitric oxide. However, during heatstroke, vasoconstriction may be present in other vascular beds, as evidenced by flow reductions in rat tail (7, 9), iliac (9), and renal (7) arteries. This vasoconstriction present in some vascular beds (7, 9) may be an attempt at compensating splanchnic vasodilatation and preventing arterial hypotension. Our results with potassium, norepinephrine, and endothelin-1 in the rabbit femoral artery are in line with those found in other vasculatures (7, 9), and they may be consistent with this compensatory mechanism. The increase in the response to vasoconstrictors, such as catecholamines and endothelin-1, in some vascular beds (e.g., muscular) during hyperthermia may be of relevance to counteract the increased vasodilatation in other vasculatures (e.g., splanchnic). However, when hyperthermia is severe and/or prolonged, this increased vasoconstriction may be insufficient to maintain arterial pressure, thus resulting in severe hypotension and cardiovascular failure.

In summary, the present study suggests that in muscular arteries hyperthermia 1) decreases the production of, and increases the dilation to, nitric oxide and 2) increases the responses to potassium chloride, norepinephrine, and endothelin-1. The potentiation of vasoconstriction to potassium and endothelin-1 may be due, at least in part, to a reduced release of nitric oxide during this condition, whereas the potentiation to norepinephrine may be independent of nitric oxide.