INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

OUR LABORATORY HAS RECENTLY demonstrated that the perivascular sensory nerve network expresses a receptor for extracellular Ca²⁺ (CaR) (5). This receptor is homologous to that initially cloned from the bovine parathyroid gland (2, 3) and subsequently shown to be present in thyroid gland (11), kidney (22), and brain (23). We also showed that low, physiological concentrations of extracellular Ca²⁺ are capable of inducing relaxation of isolated segments of mesenteric resistance arteries of the rat. The relaxation is endothelium and nitric oxide independent, is inhibited by chemical denervation with phenol and by selective sensory denervation, and is associated with the release of a hyperpolarizing vasodilator that we have called nerve-derived hyperpolarizing factor (5, 19). On the basis of these findings, we have proposed that the perivascular sensory nerve CaR serves as a molecular link between changes in whole animal Ca²⁺ homeostasis and vascular reactivity. In this capacity, activation of the sensory nerve CaR by extracellular Ca²⁺ would be associated with the release of a vasodilator transmitter, which would then cause relaxation of adjacent vascular smooth muscle cells.

For this dilator system to operate under physiological conditions, it is necessary for the concentration of Ca²⁺ in the interstitial fluid that is in contact with the adventitial surface of an artery to achieve levels that activate perivascular sensory nerve-mediated relaxation. To our knowledge, however, the concentration of ionized Ca²⁺ has not been directly measured in the interstitium of any tissue. We have therefore adapted standard in situ microdialysis methods for the measurement of the concentration of free ionized Ca²⁺ in the submucosal interstitium of the duodenum. We then used this method to test the hypothesis that, under physiological conditions, the concentration of free Ca²⁺ in the duodenal submucosa undergoes dynamic changes over a range that is sufficient to activate the perivascular sensory nerve-linked dilator system.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Animal preparation for microdialysis. All animal procedures were carried out in accordance with regulations of the Institutional Animal Care and Use Committee. Male Wistar rats (8-10 wk old) were purchased from Harlan Sprague Dawley and maintained on Purina rodent chow and deionized water. On the day of study the rats were anesthetized with a mixture of ketamine and xylazine (100 and 5 mg/kg ip, respectively). Additional anesthesia was given at one-half the initial dose every 50 min or as needed throughout the experiment. A venous line was inserted into the left jugular vein, and the small intestine was exposed through a midline incision. A linear microdialysis probe with a 5-mm window and a molecular weight cutoff of 30,000 (Bioanalytical Systems) was tunneled through the interstitial space of the duodenal submucosa, ~2 cm distal to the pyloric sphincter, by means of a 23-gauge needle, with care taken not to enter the lumen. The probe was then fixed in place with veterinary bonding glue (3M Animal Care Products, St. Paul, MN) and perfused at 1 µl/min with use of a microperfusion pump (Bioanalytical Systems) with perfusion buffer containing 120 mmol/l NaCl, 20 mmol/l HEPES (pH 7.4), and various concentrations of Ca²⁺.

Zero-net flux protocol. To determine the concentration of Ca²⁺ in the interstitial space of the duodenal submucosa, we used an equilibrium dialysis method similar to that previously used by Siragy and Linden (24) to estimate interstitial adenosine concentration. After the equilibration period the microdialysis probe was perfused at 1 µl/min with buffer containing increasing concentrations of Ca²⁺ (0.5, 1, 2, and 3 mmol/l). In preliminary experiments we determined the time required for equilibration of buffer in the dialysis probe with fluid in the interstitial compartment. This was done by perfusing the probe with known amounts of Ca²⁺ and then taking samples of effluent (dialysate) for measurement of Ca²⁺ every 15 min for 1 h. On the basis of the results, subsequent experiments were performed by allowing a 35-min equilibration period for each new concentration of Ca²⁺, and then dialysate was collected for 15 min. The concentration of free ionized Ca²⁺ in dialysate and perfusate was determined using a microfluorometric method (see below). The difference between Ca²⁺ in the dialysate and Ca²⁺ in the perfusate was then plotted as the dependent variable against the concentration of Ca²⁺ in the perfusate. Through the use of regression analysis, the point where the difference between Ca²⁺ in the dialysate and Ca²⁺ in the perfusate was zero (zero-net flux point) was identified and used as an estimate of the concentration of Ca²⁺ in the interstitium. Ca²⁺ in the interstitium was estimated when the correlation coefficient was 0.9 (95% of the time). At the end of each experiment the segment was examined under ×40 magnification to confirm placement of the probe in the submucosa. Only data from animals in which the probe was correctly placed in the submucosa were included for analysis. Some segments were processed for histology to confirm placement of the probe within the submucosa.

Verification of integrity of duodenal submucosa. The supravital colloidal dye trypan blue, which is taken up by dead cells, was used in some experiments to assess viability of the duodenal mucosa (18, 21, 25). In these cases the duodenal lumen was perfused with buffer containing 0.06% trypan blue, the effluent from the dialysis probe was collected, and the optical density (OD) at 562 nm was determined to test for the presence of dye. After collection of the samples, the perfused bowel segment was slit open longitudinally, washed with buffer, and visually examined (×50 magnification) to determine whether dye had permeated into the submucosa.

In vitro characterization of the microdialysis probe. An in vitro recovery assay was carried out to establish whether there was a linear relationship between the concentration of Ca²⁺ measured in the dialysate and the concentration of Ca²⁺ in fluid bathing the external surface of the probe. The dialysis probe was placed in 5 ml of buffer containing 120 mmol/l NaCl, 20 mmol/l HEPES, and 0-6 mmol/l Ca²⁺. The probe was perfused with Ca²⁺-free buffer at 1 µl/min, and samples of dialysate were collected over 15-min periods. The concentration of Ca²⁺ in the dialysate was then plotted as a function of Ca²⁺ concentration in the bathing medium.

Measurement of Ca²⁺. Ca²⁺ in the experimental samples was measured using a microfluorometric assay. Fifty microliters of Ca²⁺-free perfusion buffer containing 20 µmol/l mag-fura 5 were placed in a stainless steel chamber with a glass coverslip attached to its bottom. The chamber was then placed on the stage of a Nikon Diaphot microscope that was interfaced with a dual-excitation wavelength fluorometer (model AR/CM, Spex, Edison, NJ). The fluorescence of this buffer at 510 nm was recorded during sequential excitation at 340 and 380 nm (1,000 Hz), and then 50 µl of solution containing a known amount of Ca²⁺ (12.5-125 µmol/l) were added and the fluorescence was recorded. The ratio of fluorescence at 340 nm to that at 380 nm of solutions containing known amounts of Ca²⁺ was used to construct a standard curve, which in turn was used to estimate the concentration of ionized Ca²⁺ in diluted aliquots of dialysate.

Isolated vessel studies. Ca²⁺-induced relaxation was determined using previously described methods (5). After induction of anesthesia, branch II mesenteric resistance arteries were isolated from the mesentery, with care taken not to disturb the periadventitial surface, and mounted on a wire myograph. The segment was allowed to equilibrate for 30 min at 37°C in physiological salt solution of the following composition (in mmol/l): 150 NaCl, 4.7 KCl, 1.17 MgSO₄ · 7H₂O, 5 NaHCO₃, 1.10 Na₂HPO₄, 1.0 CaCl₂, 20 HEPES, and 5 glucose; the pH of this solution was 7.4 when it was gassed with 95% air-5% CO₂. After equilibration the segments were set to an internal diameter of 200-225 µm, as previously described (19), and allowed to equilibrate for another 30 min. The segments were then contracted with 10 µmol/l methoxamine or 5 µmol/l serotonin, and the relaxation response was recorded during cumulative addition of Ca²⁺ from 1 to 5 mmol/l. Relaxation induced at each concentration of Ca²⁺ was calculated and expressed as percent initial tension, where 100% represents baseline tension after contraction with 10 µmol/l methoxamine or 5 µmol/l serotonin and 0% represents complete absence of active tension. Plots of percent relaxation vs. extracellular Ca²⁺ were constructed for each experiment to estimate the concentration of Ca²⁺ that induced 50% relaxation. This value was then taken as the ED₅₀ for Ca²⁺.

Statistical analysis. Statistical analysis was carried out using the SYSTAT software package. Values are means ± SE. Linear regression analysis was performed using a routine that provided the best fit to the following equation: y = mx + b, where m and b are regression coefficients. Comparisons among groups were made using ANOVA or unpaired Student's t-test. P < 0.05 was taken to indicate statistically significant difference.

Drugs and chemicals. Methoxamine and serotonin were obtained from Sigma Chemical (St. Louis, MO), mag-fura 5 from Molecular Probes (Junction City, OR), and ketamine and xylazine from the University of Texas Medical Branch Pharmacy; nonfat dry milk (Carnation) was purchased from Konsumer's Food Market (Galveston, TX).

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Position of the probe and mucosal viability. Examination of cross sections of segments of duodenum, with the probe in place, at ×40 magnification revealed placement of the microdialysis probe within the submucosal space. Only data from animals in which the probe was correctly placed in the submucosa (27 of 29 or 93% of animals tested) were included for analysis. Histological examination of cross sections of the duodenal segments also confirmed the position of the probe within the submucosal space. When trypan blue was perfused through the duodenal lumen, subsequent examination of the bowel wall revealed no evidence of penetration of the dye across the mucosal surface. Moreover, when the OD at 562 nm of the dialysate was determined before and after perfusion of the gut with trypan blue, there was no evidence that the dye was present in the lumen of the probe (OD = 0.00 ± 0.00 for both groups, n = 3, P = 0.48). These data indicate that there was no leakage of the dye from the mucosa to the probe in the submucosal space and support the conclusion that placement of the probe did not affect mucosal integrity.

Microfluorometric determination of Ca²⁺. Figure 1 illustrates the linear relationship between total Ca²⁺ in an aqueous solution and the ratio of fluorescence of mag-fura 5 during sequential excitation at 340 and 380 nm. The mean Ca²⁺ concentration of one sample assayed 10 times in the same assay was 0.59 ± 0.03 (SD) mmol/l (n = 10), giving an intra-assay coefficient of variation (SD/mean) of 5.1%. When the same sample was assayed on 13 separate occasions, the mean Ca²⁺ concentration was 0.50 ± 0.04 mmol/l (n = 13). Thus the interassay coefficient of variation was 8%. These results indicate good reproducibility of determination of Ca²⁺ by microfluorometry. When this method was used to estimate the concentration of free Ca²⁺ that is present in reconstituted nonfat dry milk, 7.78 ± 0.28 mmol/l (n = 4) was obtained.

View larger version (8K): [in this window] [in a new window]   Fig. 1.   Standard curve for determination of free ionized Ca2+ in microsamples. Ordinate, ratio of fluorescence of mag-fura 5 at 510 nm during sequential excitation at 340 and 380 nm. Intra-assay variation was 5.1%; interassay variation was 8%. Values are means ± SE; n = 10-13 (y = 0.956 + 0.004x, r = 0.998). [Ca2+], Ca2+ concentration.

In vitro Ca²⁺ recovery assay. The equilibrium dialysis method that we used depends on a constant percent recovery of Ca²⁺. Preliminary experiments were therefore carried out to establish the relationship between the concentration of Ca²⁺ in the bathing medium and the concentration of Ca²⁺ in the dialysate (17). When Ca²⁺ in the bath was increased from 0 to 6 mmol/l, a proportional increase in Ca²⁺ in the dialysate was observed such that there was a positive linear relationship with a correlation coefficient of 0.998 and a relative recovery rate of 30% (Fig. 2).

View larger version (10K): [in this window] [in a new window]   Fig. 2.   Relationship between Ca2+ bathing microdialysis probe and Ca2+ in dialysate. Probe was perfused with Ca2+-free buffer at 1 µl/min and bathed in buffer containing increasingly higher levels of Ca2+. At a constant flow rate, there was a linear relationship between Ca2+ concentration in dialysate and Ca2+ concentration in bathing medium (n = 5, r = 0.999).

Equilibrium microdialysis. The zero-net flux method that was employed also requires that measurements be made during steady-state conditions. Experiments were therefore performed to determine the time that is required to achieve equilibrium for each concentration of Ca²⁺ that was perfused through the microdialysis probe. The dialysis probe was perfused at 1 µl/min for 1 h with buffer containing 0.5, 1, 1.5, 2, or 3 mmol/l, and samples of dialysate were taken at 15-min intervals for determination of Ca²⁺. We found that equilibration at each level of Ca²⁺ occurs very rapidly, with a steady-state level being achieved within 30 min for all concentrations (Fig. 3; n = 3). To allow for variation between preparations, subsequent experiments were performed allowing for a 35-min equilibration period followed by a 15-min collection period for each concentration of Ca²⁺ that was perfused through the duodenal lumen.

View larger version (13K): [in this window] [in a new window]   Fig. 3.   Determination of in vivo time course of equilibration of Ca2+ in dialysate during perfusion of dialysate probe with 0.5-3 mmol/l Ca2+. Ca2+ concentration in dialysate leveled off within 30 min of equilibration; no statistically significant difference was detected between 30 and 45 min for 0.5 (), 1 (), 1.5 (), 2 (), and 3 mmol/l () Ca2+. P > 0.05, n = 3.

Effect of luminal Ca²⁺ on interstitial Ca²⁺. The zero-net flux method was used to estimate the concentration of interstitial Ca²⁺ that was present in the duodenal submucosa during perfusion of the lumen of the bowel with buffer that was nominally Ca²⁺ free or contained 3, 6, or 10 mmol/l Ca²⁺. When the duodenal lumen was perfused with nominally Ca²⁺-free buffer, interstitial Ca²⁺ was 1.0 ± 0.13 mmol/l (n = 4). When the concentration of Ca²⁺ perfusing the duodenal lumen was 3, 6, or 10 mmol/l, the concentration of interstitial Ca²⁺ was 1.52 ± 0.04, 1.78 ± 0.10, and 1.89 ± 0.10 mmol/l, respectively. These values are significantly greater than that observed during perfusion with Ca²⁺-free buffer (n = 4-5, P < 0.05; Fig. 4). When the concentration of interstitial Ca²⁺ was plotted against the concentration of Ca²⁺ in the duodenal lumen, there was a concentration-dependent relationship that showed saturation of interstitial Ca²⁺ content at ~2 mmol/l (Fig. 5).

View larger version (11K): [in this window] [in a new window]   View larger version (12K): [in this window] [in a new window]   View larger version (11K): [in this window] [in a new window]   View larger version (12K): [in this window] [in a new window]   Fig. 4.   Regression analyses of zero-net flux data obtained during perfusion of duodenal lumen with Ca2+-free buffer (A) or buffer containing 3 mmol/l (B), 6 mmol/l (C), or 10 mmol/l (D) Ca2+. Increasing Ca2+ in lumen of duodenum from 0 to 3, 6, and 10 mmol/l significantly increased interstitial Ca2+ (Ca2+isf): 1.0 ± 0.13 mmol/l (n = 4), 1.52 ± 0.04 (n = 4), 1.78 ± 0.1 (n = 5), and 1.89 ± 0.1 mmol/l (n = 4), respectively (P < 0.05). There was no statistically significant increase in interstitial Ca2+ concentration when lumen Ca2+ was raised from 6 to 10 mmol/l (P = 0.53).

View larger version (14K): [in this window] [in a new window]   Fig. 5.   Relationship between concentration of Ca2+ in lumen of duodenum and Ca2+ determined in submucosal interstitium. Increasing Ca2+ caused a significant rise in interstitial Ca2+ between 0 and 6 mmol/l (P < 0.05) and did not increase further between 6 and 10 mmol/l, achieving a maximal concentration of interstitial Ca2+ of ~2 mmol/l. Values are means ± SE; n = 4-5.

Comparison of interstitial Ca²⁺ in fasting and free-feeding rats. Because the preceding data indicate that changes in luminal Ca²⁺ induce changes in interstitial Ca²⁺, we tested the hypothesis that the concentration of Ca²⁺ in the duodenal submucosa changes with the feeding state of the animal. This was done by using the zero-net flux method to estimate the concentration of interstitial Ca²⁺ in fasting and free-feeding animals. In animals that were fasted for 24 h, interstitial Ca²⁺ was 1.1 ± 0.06 mmol/l, a level that was not different from that observed when the lumen was perfused with Ca²⁺-free buffer (P = 0.4; Fig. 6A). In contrast, in animals studied under free-feeding conditions, interstitial Ca²⁺ was 1.4 ± 0.06 mmol/l, which was significantly greater than that observed in the fasting animal (P < 0.05) and midway between the values observed when the duodenal lumen was perfused with Ca²⁺-free buffer and with 6 or 10 mmol/l Ca²⁺ (Fig. 6B).

View larger version (11K): [in this window] [in a new window]   View larger version (12K): [in this window] [in a new window]   Fig. 6.   Regression analyses of zero-net flux data obtained from animals that were studied after a 24-h fast (A) or under free-feeding conditions (B). Interstitial Ca2+ was significantly decreased in fasted animals compared with free-feeding rats: 1.1 ± 0.06 and 1.4 ± 0.06 mmol/l in fasted and free-feeding rats, respectively (P < 0.05, n = 5 each).

Comparison of interstitial Ca²⁺ and Ca²⁺ sensitivity of Ca²⁺-induced relaxation. To provide insight into the question of whether the concentration of interstitial Ca²⁺ observed in the present experiments is sufficient to activate the sensory nerve-mediated, Ca²⁺-activated dilator system that we recently described (5, 19), we performed experiments to assess Ca²⁺ sensitivity of Ca²⁺-induced relaxation during precontraction with two different agonists. Ca²⁺ induced relaxation of vessels precontracted with the ₁-agonist methoxamine (10 µmol/l) with an ED₅₀ of 1.67 ± 0.08 mmol/l and segments precontracted with 5 µmol/l serotonin with an ED₅₀ of 1.54 ± 0.05 mmol/l (Fig. 7; n = 5 each). These ED₅₀ values are midway between the minimal (1 mmol/l) and maximal (1.9 mmol/l) interstitial concentrations of Ca²⁺ that were observed in our equilibrium dialysis experiments and support the hypothesis that, under physiological conditions, interstitial Ca²⁺ achieves concentrations that can stimulate Ca²⁺-induced relaxation.

View larger version (12K): [in this window] [in a new window]   Fig. 7.   Concentration-response relationship for Ca2+-induced relaxation determined in vessels that were precontracted with 10 µmol/l methoxamine () or 5 µmol/l serotonin (). Values are means ± SE; not statistically significant, overall P = 0.25. ED50 values were 1.67 ± 0.08 and 1.54 ± 0.05 mmol/l for methoxamine- and serotonin-pretreated vessels, respectively. There was no statistically significant difference between ED50 values (P = 0.21, n = 5). Arrows, degree of relaxation that would be achieved as interstitial Ca2+ varies during fasting and feeding states or when 6 mmol/l Ca2+ is present in lumen of duodenum. Initial vessel tensions were not different between serotonin- and methoxamine-precontracted segments: 0.71 ± 0.157 and 0.96 ± 0.108 mN/mm, respectively (n = 5 each, not significantly different, P = 0.22).

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The goal of these experiments was to establish a method that could be used to determine the concentration of Ca²⁺ in the interstitial compartment of the duodenal submucosa and to use the method to test the hypothesis that under physiological conditions the concentration of free Ca²⁺ in this compartment undergoes dynamic changes over a range that is sufficient to activate the perivascular sensory nerve-linked, Ca²⁺-activated dilator system. The results of this study support this hypothesis by demonstrating that the concentration of Ca²⁺ in the duodenal submucosa changes significantly under normal feeding/fasting conditions over a range that can activate the perivascular sensory nerve-mediated Ca²⁺-induced relaxation.

We recently reported that low, near-physiological concentrations of extracellular Ca²⁺ cause nerve-dependent relaxation of isolated arteries, possibly by activation of a perivascular sensory nerve Ca²⁺ receptor (5, 19). These findings led to our hypothesis that this Ca²⁺-mediated dilator system serves to couple changes in interstitial Ca²⁺, as would occur during transcellular movement of Ca²⁺ in tissues such as the small intestine, kidney, and bony matrix, with local vascular tone (4). As a test of this hypothesis, we sought to directly measure the concentration of free Ca²⁺ in the interstitium of the duodenal submucosa. One reason that the small bowel was chosen over the renal and bony tissue is that it seemed that it would be relatively simple to alter interstitial Ca²⁺ by perfusing the lumen of the bowel with known amounts of Ca²⁺. The second reason is that we previously demonstrated high sensitivity of Ca²⁺-induced relaxation in branch II and III mesenteric resistance arteries, which feed this bed and might be expected to respond in a manner that is similar to the arteries in the submucosa. No studies have been done to document the existence of the perivascular sensory nerve CaR in submucosal vessels. However, there is evidence that supports the possibility that the pattern of perivascular innervation is similar in extra- and intramural arteries. One such line of evidence comes from studies showing that extrinsic denervation of the gut causes a significant loss of nonadrenergic, noncholinergic fibers in submucosal arteries (10). Other evidence supporting the existence of sensory nerve-mediated vasodilation in intramural vessels includes studies showing that extrinsic sensory afferent nerves project to the intestinal submucosal arterioles and mediate neurogenic vasodilation (29, 30) similar to that observed in extramural mesenteric resistance vessels (13).

In view of the fact that no direct measurements of Ca²⁺ concentration in the duodenal submucosa have been reported, it was necessary to develop a method for making these measurements. Several indirect methods have been used to collect interstitial fluid, including the use of chronically implanted capsules or microdialysis fibers, the liquid paraffin technique, or capillary ultrafiltration (6, 12, 15). Each of these methods, however, has distinct limitations for application to the intestinal wall. For example, because of the time required to achieve equilibration (several days to weeks), a chronic implantation method would not have been suitable for our studies, where we needed to acutely vary the intestinal Ca²⁺ content (6, 12). Similarly, the liquid paraffin technique would not have been ideal, inasmuch as it would have required us to make an incision on the intestinal wall to expose the submucosa, thus running the risk of contaminating the samples with blood cells and plasma proteins. We therefore adapted an in situ microdialysis method to measure interstitial Ca²⁺.

The in situ microdialysis technique initially described for neurobiology studies has been adapted for use in almost any tissue and can be used to measure virtually any substance of interest within the molecular weight cutoff limit (7-9, 24, 26, 29). The in vivo equilibrium dialysis technique developed by Lonroth et al. (17) allows in situ calibration of the microdialysis probe and permits high-precision estimates of intercellular substances of interest. This technique has been used to measure adenosine concentrations in the renal interstitium of rats and in subcutaneous tissue in humans (16, 24). In the present study we adapted the equilibrium microdialysis method to measure duodenal interstitial Ca²⁺ concentration. Although the data that were obtained in this study were highly reproducible in terms of estimated Ca²⁺ concentrations and the data obtained with trypan blue dye indicated that mucosal integrity was not grossly compromised, placement of the microdialysis fiber may have altered mucosal or submucosal blood flow, microvessel permeability, or interstitial volume. Any of these factors could then alter the absolute Ca²⁺ concentration in the interstitial compartment. It is unlikely, however, that any of these factors could be responsible for the dynamic changes in Ca²⁺ that were observed, because the microdialysis fiber was in place under all test conditions. Moreover, the interstitial Ca²⁺ concentrations that were observed under feeding/fasting conditions closely approximate the normal range of serum Ca²⁺ concentration. This finding, together with the observation that the Ca²⁺ concentration achieved in the submucosal compartment was saturable, makes it unlikely that the estimates of Ca²⁺ are grossly inaccurate. Therefore, in situ microdialysis appears to be a suitable method for estimating Ca²⁺ concentration in the interstitial space of the duodenal submucosa.

Using this method, we showed that Ca²⁺ in the interstitial compartment of the duodenal submucosa changes in response to perfusion of the intestinal lumen with buffers containing various amounts of Ca²⁺: from 1.0 mmol/l with Ca²⁺-free buffer to 1.8-1.9 mmol/l with 6 or 10 mmol/l Ca²⁺ in the lumen. The finding that the highest level that we observed seems to plateau near 2 mmol/l supports the idea that the Ca²⁺ concentration that can be achieved in the interstitial compartment is physiologically limited. This finding is consistent with previous reports suggesting that there is an upper limit to the amount of Ca²⁺ that may be absorbed by the intestine, even in the face of very high dietary intake (1, 14, 20).

In an attempt to address the question of the relevance of the Ca²⁺ concentrations that we used to perfuse the lumen of the bowel, we used the microfluorometric assay to measure the free Ca²⁺ concentration in reconstituted nonfat dry milk and found that it is ~7 mmol/l. This concentration is within the predicted range on the basis of the analysis of ingredients that was supplied with the product and implies that the Ca²⁺ concentrations that we used to perfuse the lumen of the bowel can be achieved during consumption of commercially available food products. Nevertheless, we were concerned that the process of perfusing the lumen of the bowel with artificial buffer may have caused changes that could alter the kinetics of transcellular ion transport. We therefore designed experiments to measure interstitial Ca²⁺ under conditions in which Ca²⁺ content might be expected to vary naturally, i.e., the fasting and free-feeding states. When Ca²⁺ was determined under these conditions, we found that it varied from 1.1 to ~1.5 mmol/l.

The results of the microdialysis portion of this study indicate that free Ca²⁺ concentration in the interstitial compartment of the duodenal submucosa is a function of Ca²⁺ concentration in the lumen of the bowel and varies between 1 and 2 mmol/l. This concentration range nearly perfectly matches the dose-response relationship of Ca²⁺-induced relaxation (Fig. 7), which, as reported previously, may be linked to the perivascular nerve CaR (5, 19). Although these data indicate that Ca²⁺ in the interstitium can vary over the range that could activate Ca²⁺-induced relaxation of isolated branch I and II mesenteric arteries and support the hypothesis that the Ca²⁺-activated dilator system may be functional in the gut wall, direct verification that CaR protein is present in the perivascular nerve network of intestinal submucosal vessels is yet to be shown.

In view of these findings, we propose that the perivascular sensory nerve-linked, Ca²⁺-activated dilator system that we have described (5, 19) is functional under physiological conditions and may serve as a link between changes in local Ca²⁺ transport activity and alterations in vascular reactivity, peripheral resistance, and regional blood flow.