INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

OVER THE PAST few years, a number of independent laboratories have demonstrated that Ca²⁺ modulates cell function by activating a membrane-spanning Ca²⁺ receptor (CaR) that signals changes in extracellular Ca²⁺ to the cell interior by means of G protein-coupled pathways (3, 11, 12, 22, 23). The CaR has been demonstrated in a variety of tissues including thyroid and parathyroid glands, kidney, brain, and nerves (3, 11, 12, 22, 23). The discovery of a Ca²⁺-sensing receptor, together with previous data suggesting that Ca²⁺ may mediate relaxation by activating a G protein-coupled pathway (2), led us to look for a homologous CaR in the blood vessel wall with the idea that a CaR could serve as a sensor that signals changes in extracellular Ca²⁺ to the vascular smooth muscle.

We recently reported the presence of CaR protein in the perivascular nerve network of mesenteric resistance arteries as well as in dorsal root ganglia which house cell bodies for perivascular sensory neurons (5). In addition, we reported that Ca²⁺-induced relaxation is not dependent on the presence of an intact endothelium and is not blocked by inhibitors of either nitric oxide synthase or cyclooxygenase. In contrast, treatment of arteries with phenol, which reportedly destroys perivascular nerves (1), completely abolished the relaxation, supporting the idea that Ca²⁺-induced relaxation is nerve mediated (5). One limit of the phenol treatment, however, is that it is not selective for any single type of nerve fiber. The present experiments took advantage of the fact that capsaicin, the pungent ingredient of most red peppers of the genus Capsicum, selectively destroys sensory nerve fibers (15, 17). We therefore used capsaicin-induced sensory denervation to test the hypothesis that Ca²⁺-induced relaxation is mediated by activation of the sensory nerve CaR. A previous report of this study has appeared in abstract form (19).

	METHODS

Top Abstract Introduction Methods Results Discussion References

All animal procedures were carried out in accordance with regulations of the Institutional Animal Care and Use Committee. Pregnant female Wistar rats were purchased from Harlan Sprague Dawley and maintained on Purina Rodent Chow and deionized water. On days 1 and 2 after delivery, a subgroup of the pups (n = 8) was anesthestized with halothane and given a subcutaneous injection of capsaicin (50 mg/kg) dissolved in a vehicle containing 5% ethanol, 5% Tween 80, and 90% saline. This dose of capsaicin has been demonstrated to cause permanent morphological ablation of thin unmyelinated sensory neurons (15, 17). Litter-matched controls received an equivalent volume of the same vehicle (n = 5) or saline (n = 6). The rats were weaned at 3 wk of age and placed on Purina Rodent Chow and deionized water. At age 5 wk, the rats were anesthetized with a mixture of ketamine and xylazine (100 mg/kg, 5 mg/kg) and killed by exsanguination. Mesenteric resistance arteries were isolated and maintained in ice-cold physiological salt solution (PSS) of the following composition (in mM): 150 NaCl, 4.7 KCl, 1.17 MgSO₄ · 7H₂O, 5 NaHCO₃, 1.15 KH₂PO₄, 1.10 Na₂HPO₄, 1.0 CaCl₂, 20 HEPES, and 5 glucose, pH 7.4.

Biophysical measurements. Cylindrical segments of mesenteric resistance arteries that were 2 mm long were mounted on wire myographs (Kent Scientific) by means of two tungsten wires inserted through the lumen. Each wire was attached to a plastic foot by stainless steel screws; one foot was connected to an isometric force transducer and the other to a microdrive. The mounted segments were placed in tissue baths filled with PSS that was maintained at 37°C and continuously gassed with 95% air-5% CO₂. After equilibration for 15 min, the vessels were set to an internal diameter of 200-225 µm (see below). After equilibration for a further 15 min, the segments were contracted with 5 µM norepinephrine until reproducible contractile responses were obtained (3 or 4 times). In preliminary experiments, we noted that contraction of the vessel segments with a solution containing a high concentration of K⁺ decreased subsequent Ca²⁺-induced relaxation. We therefore avoided the use of solutions containing a depolarizing concentration of K⁺. Ca²⁺-induced relaxation was assessed by measuring the effect of cumulative addition of Ca²⁺ from 1 to 5 mM to a segment precontracted with 5 µM norepinephrine. The magnitude of relaxation for each concentration of Ca²⁺ was calculated and expressed as percent initial tension. Relaxation response to 1 µM acetylcholine was also assessed. For acetylcholine response, the results are expressed as percent relaxation, where 100% represents baseline tension in response to 5 µM norepinephrine and 0% relaxation represents complete absence of active tension.

In vivo dimension analysis. The inner diameters of branch I and II mesenteric resistance arteries were measured in vivo using intravital video microscopy to provide an estimate of what diameter they should be set to when studied in vitro. Rats were anesthetized with ketamine and xylazine, and the femoral artery was cannulated with PE-50 tubing for direct measurement of intra-arterial pressure using a pressure transducer (Statham) and chart recorder (Gould). Body temperature was maintained at 37°C and continuously monitored using a rectal temperature probe. A section of jejunum was exteriorized through a midline incision, and the mesenteric arcade perfusing it was exposed. Fat and connective tissue were gently removed by blunt dissection. After a 20-min recovery period, mesenteric arteries were viewed by means of a videocamera, and the image was stored on videotape. The tape was then played back, and the vessel diameter was measured using a calibration scale derived from a stage micrometer. The mean inner diameter of 20 small mesenteric arteries from 4 different rats with a mean pressure of 81 ± 2 mmHg was 213 ± 9.2 µm. The vessels were subsequently studied while stretched to internal diameters of 200-225 µm.

Immunocytochemistry. Immunocytochemical techniques were used to quantify relative density of calcitonin gene-related peptide (CGRP)-containing nerves and to verify denervation. Relative density of nerves staining positively for the CaR was also assessed. Mesenteric resistance vessels were carefully dissected of excess connective tissue and fat. Blood in the lumen was removed by flushing the mesenteric tree with cold PSS using a syringe and needle inserted into the lumen of the mesenteric artery trunk. Whole mount vessel segments were prepared as follows. Cylindrical segments of mesenteric resistance vessels ~2 mm in length were fixed in ice-cold methanol for 10 min. After the segments were rinsed with Tris-buffered saline (TBS), intrinsic peroxidase activity was blocked with H₂O₂/NaN₃ (DAKO peroxidase blocker), and nonspecific protein binding was blocked using a serum-free blocker (DAKO, Carpinteria, CA). The segments were then incubated overnight at 4°C with a polyclonal rabbit anti-CaR antibody (KK3 1:2,000, 6 µg/ml raised against a fusion protein containing amino acids 340-620 and generously provided by Dr. Kim Rogers of NPS Pharmaceuticals) or an anti-CGRP antibody (1:250, Sigma Chemical, St. Louis, MO). After the segments were washed with TBS, intrinsic biotin was blocked using a biotin-blocking kit (DAKO). The segments were then incubated with biotin-conjugated secondary antibody (1:250). The resulting antigen-antibody complex was stained using horseradish peroxidase-conjugated avidin with Vector SG substrate as chromogen (Vector Labs, Burlingame, CA). After dehydration with ethanol and xylene, the segments were permanently mounted.

Statistical analysis. Statistical analysis was performed using the SYSTAT software package. All data are presented as means ± SE. Comparisons among groups were performed using ANOVA with a repeated measures design where appropriate. A P value of <0.05 was taken to indicate significant difference among groups.

Drugs and chemicals. Norepinephrine (arterenol bitartrate salt), acetylcholine chloride, and 8-methyl-N-vanillyl-6-nonemide (capsaicin) were obtained from Sigma Chemical.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Immunocytochemistry. Visual examination of photomicrographs indicated that neonatal treatment of rats with capsaicin significantly reduced the density of neurons containing CGRP. Morphometric analysis performed as described in METHODS confirmed this impression (Fig. 1; control CGRP density, 51.1 ± 3.9 µm²/mm²; capsaicin-treated CGRP density, 31.4 ± 2.8 µm²/mm², P = 0.01, n = 3). This result confirmed effectiveness of the capsaicin treatment as a means of destroying sensory nerves.

View larger version (78K): [in this window] [in a new window]   Fig. 1.   Representative photomicrographs demonstrating that neonatal capsaicin treatment significantly reduces density of perivascular nerves that stain positively for calcitonin gene-related peptide (CGRP). Animals were treated with capsaicin on days 1 and 2 of life and killed at 5 wk of age. Mesenteric resistance arteries were isolated and processed for immunocytochemistry using anti-CGRP antibody. Density of CGRP-containing nerves was estimated using a method used by Wang and Prewitt (26). A: segment from a sham-treated rat. B: segment from a capsaicin-pretreated rat. C: photomontage of CGRP-positive fibers in A. D: photomontage of CGRP-positive fibers in B.

View larger version (86K): [in this window] [in a new window]   Fig. 2.   Representative photomicrographs demonstrating that neonatal capsaicin treatment significantly reduces density of perivascular nerves that stain positively for Ca2+ receptor (CaR). Animals were treated with capsaicin as described above. Mesenteric resistance arteries were isolated and processed for immunocytochemistry using polyclonal rabbit anti-CaR antibody (KK3). A: segment from a sham-treated rat. B: segment from a capsaicin-pretreated rat. C: photomontage of CaR-positive fibers in A. D: photomontage of CaR-positive fibers in B.

Biophysical measurements. Cumulative addition of extracellular Ca²⁺ (1-5 mM) to norepinephrine-preconstricted segments of mesenteric resistance artery caused a dose-dependent relaxation. Ca²⁺-induced relaxation of vessels from vehicle-pretreated rats did not differ significantly from saline-treated rats, indicating that the vehicle did not alter relaxation (vehicle treated vs. saline treated, overall P = 0.225, not significant). In contrast, Ca²⁺-induced relaxation was significantly depressed in vessels from capsaicin-pretreated rats, suggesting a perivascular nerve dependency of Ca²⁺-induced relaxation (Fig. 3, overall P < 0.001, n = 6 or 7). To verify selectivity of the capsaicin effect, we tested relaxation response to 1 µM acetylcholine. Unlike Ca²⁺, acetylcholine-induced relaxation was not significantly affected by neonatal capsaicin pretreatment (Fig. 4, saline treated vs. capsaicin treated, P = 0.35, n = 6 or 7). Similarly, the magnitude of contraction induced by 5 µM norepinephrine was not different between saline-treated rats and vessels from capsaicin-pretreated rats (Fig. 5, saline treated vs. capsaicin treated, P = 0.5, n = 6 or 7).

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Effect of capsaicin treatment on relaxation. Ca2+-induced relaxation was significantly depressed in vessels from capsaicin-pretreated rats (saline vs. capsaicin treated, overall P < 0.001, n = 6 or 7). * Significant difference at P < 0.05 saline vs. capsaicin treated, or vehicle vs. capsaicin. Treatment with vehicle alone (5% ethanol, 5% Tween 80, 90% saline) did not significantly alter Ca2+-induced relaxation (vehicle vs. saline treated, n = 5 or 6, P = 0.225, not significant).

View larger version (9K): [in this window] [in a new window]   Fig. 4.   Effect of capsaicin treatment on 1 µM acetylcholine relaxation. Neonatal treatment with capsaicin did not significantly alter acetylcholine relaxation of isolated mesenteric resistance arteries (saline vs. capsaicin treated, n = 6 or 7, P = 0.35, not significant).

View larger version (11K): [in this window] [in a new window]   Fig. 5.   Effect of capsaicin treatment on active tension development. Neonatal treatment with capsaicin did not significantly alter active tension developed by isolated mesenteric resistance artery segments in response to addition of 5 µM norepinephrine (saline treated vs. capsaicin treated, n = 6 or 7, P = 0.5, not significant).

View larger version (10K): [in this window] [in a new window]   Fig. 6.   Scatter plot showing relaxation induced by 3 mM Ca2+ in presence of 5 µM norepinephrine as a function of density of CaR in perivascular nerves. We found a significant positive correlation between Ca2+-induced relaxation and CaR content of mesenteric perivascular nerve network (n = 12, r2 = 0.559, r = 0.748, P = 0.005).

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

As noted above, we have recently found that perivascular nerves express a CaR and have performed experiments that indicate that this receptor mediates Ca²⁺ relaxation through coupling to the release of a nonpeptide hyperpolarizing factor (5). The important new finding of the present study is that chronic perivascular sensory denervation of rats with capsaicin significantly decreases the magnitude of Ca²⁺-induced relaxation. These data support the hypothesis that Ca²⁺ relaxes isolated arteries by activating a CaR that is present in perivascular sensory neurons.

We chose capsaicin as a means of inducing chronic sensory denervation because it is a neurotoxin that is highly selective for a population of primary afferent neurons that have thin unmyelinated (C) fibers (15, 17). The neurotoxic effect of capsaicin is dose dependent. Moreover, the dose of 50 mg/kg used in this study has been widely used to induce sensory denervation (7, 10, 13, 15). Our finding of decreased CaR content in capsaicin-treated rats supports the hypothesis that the CaR is present in perivascular sensory neurons.

Our strategy for verifying that capsaicin induces sensory denervation was based on the fact that sensory nerves express a high level of CGRP (7, 13). As predicted, capsaicin reduced the relative density of CGRP-containing fibers, and this is in agreement with previous reports showing that neonatal capsaicin treatment decreases CGRP content (7, 13). Because CGRP is a potent vasodilator, a possible interpretation of our data is that Ca²⁺ causes relaxation by inducing CGRP release. However, it is unlikely that a peptide vasodilator mediates Ca²⁺-induced relaxation, since we have shown that blockade of CGRP receptors with CGRP-837, neurokinin A receptors with SR-48968, and substance P receptors with spantide II and SR-140333 is without effect on Ca²⁺-induced relaxation (5, 8, 9).

Although the identity of the vasodilator that mediates Ca²⁺ relaxation remains unknown, we previously reported that Ca²⁺-induced relaxation is not mediated by nitric oxide and that relaxation persists in endothelium-denuded segments of mesenteric resistance artery (5), thus ruling out an endothelium-derived relaxing vasodilator as a possible mediator. In the present study, vessels from capsaicin-treated rats that showed a depressed response to Ca²⁺ had a functionally intact endothelium as shown by acetylcholine-induced relaxation that was not different from control. This is in agreement with the findings of Ralevic et al. (21), who showed that endothelium-dependent relaxation to acetylcholine and ATP was not affected by neonatal capsaicin treatment (21). Moreover, demonstration of a normal response to acetylcholine implies that the decreased response to Ca²⁺ cannot be explained by a global reduction in vascular relaxation caused by capsaicin treatment. Finally, the magnitude of norepinephrine-induced contraction did not differ significantly from control, suggesting that vascular smooth muscle function was preserved in vessels from capsaicin-treated rats.

Our finding of a significant positive correlation between Ca²⁺-induced relaxation and the density of nerves staining positively for the CaR also warrants discussion. The apparent lack of effect of capsaicin at 5 mM Ca²⁺ is likely due to the fact that capsaicin treatment did not induce complete denervation. Capsaicin treatment significantly depressed relaxation induced by 1.5, 2, and 3 mM Ca²⁺. In addition, when we performed a correlation analysis at 3 mM Ca²⁺, we found a significant positive correlation between Ca²⁺-induced relaxation and the density of CaR-positive nerves. This finding has important implications for the local regulation of vascular reactivity. It is known that primary sensory neurons that have cell bodies in dorsal root ganglia provide a perivascular network of fibers around the arterial system. The role of perivascular sensory neurons, once considered as that of simply sensing stimuli and transmitting information centrally, is now known to include an efferent motor function (6, 18). This efferent motor function of sensory nerves is achieved by the release of many neurotransmitters including peptides, purines, and nitric oxide in response to local stimuli (14). Thus, for example, in response to painful stimuli in the tooth, intradental sensory fibers release peptides that cause vasodilation and inhibit sympathetic vasoconstriction (20), and increased acid secretion in the gastric mucosa induces release of CGRP that causes local vasodilation (16). The concept of local motor efferent function fits in with our hypothesis that local changes in extracellular Ca²⁺ activate a CaR located on perivascular sensory nerves causing release of vasoactive substances that alter vascular tone.

The fact that the Ca²⁺ receptor is the molecular sensor linking extracellular Ca²⁺ with vascular reactivity is consistent with its role in other tissues. For example, in parathyroid cells and thyroid C cells, activation of a Ca²⁺-sensing receptor by increased plasma Ca²⁺ results in decreased parathyroid hormone secretion and increased calcitonin secretion, respectively (11, 12). In this example, the CaR participates in whole animal Ca²⁺ homeostasis by modulating calciotropic hormone secretion. A somewhat different role for the CaR is seen in the terminal inner medullary collecting ducts of the kidneys where activation of apical CaR in response to elevated luminal Ca²⁺ reduces vasopressin-elicited osmotic water permeability (24). In the latter example, the CaR not only plays a modulatory role whereby it antagonizes the activity of another hormone, but it may also provide a signaling mechanism linking Ca²⁺ and water metabolism (4, 24). The CaR is also found in tissues such as cerebellum, hippocampus, hypothalamus, and sensory neurons that are not normally involved in whole animal Ca²⁺ homeostasis, and Ruat et al. (23) have suggested a possible role for the CaR in sensing and regulating nerve terminal responses to changes in Ca²⁺ resulting from local synaptic and neuronal activities.

In summary, we have shown that perivascular nerves express a Ca²⁺-sensing receptor and that raising extracellular Ca²⁺ from 1 mM to 5 mM induces nerve-dependent relaxation of isolated mesenteric arteries. The location of the perivascular Ca²⁺ receptor on the adventitial surface of blood vessel walls is ideal for detecting changes in interstitial Ca²⁺ that may be expected to fluctuate in tissues such as bone, gut, and kidney where considerable Ca²⁺ exchange occurs between blood and the interstitial fluid. We propose that the perivascular CaR serves as a molecular signal that regulates local vascular resistance and blood flow in response to changes in interstitial Ca²⁺. Furthermore, our data may provide insight into the physiological mechanisms by which whole animal Ca²⁺ balance may be linked to blood pressure regulation.