Purinergic and adrenergic Ca²⁺ transients during neurogenic contractions of rat mesenteric small arteries

Preparation of arteries and experimental solutions

All experiments were carried out according to the guidelines of the Institutional Animal Care and Use Committee of the University of Maryland School of Medicine. Male Sprague-Dawley rats, weighing 140–210 g, were anaesthetized with intramuscular ketamine (50–100 mg kg⁻¹), and killed by cervical dislocation. Segments, 1.3–2.0 mm in length, of 3rd or 4th order mesenteric small arteries were dissected by methods described in detail previously (Duling et al. 1981). The dissection was carried out in cold dissection solution with the following composition (mm): 3 Mops, 145 NaCl, 5 KCl, 2.5 CaCl₂, 1 MgSO₄, 1 KH₂PO₄, 0.02 EDTA, 2 pyruvate, 5 glucose, and 1.0 % albumin (pH 7.4, 5 °C). In the confocal myograph (Danish Myo Technlogy A/S, Aarhus, Denmark) segments of artery were mounted between two 40 μm wires, one of which was fixed, the other of which was connected to a force transducer. After mounting, the artery was left slack between the wires during the loading period. The vessels were loaded with calcium indicator in dissection solution without albumin to which had been added 15 μm fluo-4 AM, 1.5 % DMSO (v/v), 0.03 % Cremaphor EL (v/v). Loading was allowed to proceed for 3 h at room temperature. Perfusion of the vessel with physiological saline solution (i.e. gassed Krebs solution (mm): 112 NaCl, 25.7 NaHCO₃, 4.9 KCl, 2 CaCl₂, 1.2 MgSO₄, 1.2 KHPO₄, 11.5 glucose, and 10 Hepes (pH 7.4, 23 °C)) was then started and the arteries were left undisturbed for a further 15 min. All of the physiological saline solutions had 1 μm capsaicin and 1 μm scopolamine added to inhibit the sensory and cholinergic nerves, respectively. The internal lumen diameter of vessels was then set using a procedure developed previously (Mulvany & Halpern, 1977). Briefly, the ring was progressively stretched by ∼1 mN increments and the internal circumference was determined at each level of stretch. From these measurements, we calculated the effective transmural pressure that would have been required to maintain the (relaxed) artery at the measured internal circumference, according to the law of Laplace. The stepwise stretch was stopped when the effective transmural pressure exceeded 70 mmHg. The relation between effective transmural pressure and internal circumference was then fitted to an exponential curve and the internal circumference of the vessel corresponding to a transmural pressure of 70 mmHg (L₇₀) was calculated from the fitted curve. The arteries were then set to the calculated internal circumference, L₁ = 0.9 L₇₀. The normalized internal lumen diameter (l₁) of the arteries was calculated as l₁ = L₁/π. The l₁ values for the vessels used were 272 ± 12 μm (mean ± s.e.m., n = 15). After this procedure, arteries were allowed to equilibrate for 1 h. Prazosin was prepared in a stock solution and diluted in the superfusate reservoir. Cremophor EL, capsaicin (N-vanillylnonamide) and scopolamine were obtained from Sigma Chemical Co. (St Louis, MO, USA); fluo-4 AM, was purchased from Molecular Probes (Eugene, OR, USA). EFS was applied via two platinum wires placed in the myograph bath parallel to the long axis of the artery and connected to a stimulator (Astro-Med, Inc., West Warwick, RI, USA, Model S48). Because arteries tended to lose fluo-4 at mammalian temperature, all experiments were carried out at room temperature (23 °C). The neurogenic origin of the force and calcium responses was confirmed by their abolition by application of 30 μm guanethidine for 1 h.

Imaging and data analysis

For 2-D confocal imaging, we used a ‘real-time’ confocal imaging system (Solamere Technology Group, SLC, UT, USA) comprised of a YOKOGAWA confocal scanner (Model CSU10) and an intensified CCD camera (Model XR/Mega-10). This produced 30 images s⁻¹. We also used a Bio-Rad MRC 600 confocal microscope (Hercules, CA, USA) for some 2-D imaging and it produced 2 images s⁻¹. Both confocal imaging systems utilized Nikon inverted microscopes equipped with ‘water’ objective lenses (× 60; NA, 1.2). Images were analysed using custom software written with IDL (Interactive Data Language, Research Systems, Inc). This software was used to obtain average fluorescence from areas-of-interest (AOIs) within the images. Fluorescence ‘pseudo-ratios’ (F/F₀) were then constructed by dividing the average fluorescence (F) within the AOI, at each frame, by the average fluorescence during the first 2 s of the recording (F₀), before EFS began. Sites where jCaTs arose within smooth muscle cells were identified with the use of a computer program that allowed us to place AOIs in the region of a local change in fluorescence, and then interactively position the AOI such that the change was largest. The pseudo-ratio within the AOI was then calculated. The time of occurrence was defined as the first point at which the rate of change of the ratio exceeded 4.8F/F₀ s⁻¹ (with additional criteria for rejection of noise). The peak value was then obtained by fitting the next 10 points (330 ms) with a 4th order polynomial, and taking the maximum value. Ca²⁺ waves were characterized by a similar procedure except that the rate of rise that had to be achieved was reduced to 0.22F/F₀ s⁻¹. Significant motion of the artery would preclude accurate confocal imaging. Although the ‘confocal’ myograph is designed to maintain arteries isometrically a small external shortening (typically ∼3 μm) does occur (see videos available in Supplementary material) and internal shortening inevitably occurs as series elastic elements are stretched. As will be shown, jCaTs are typically observed under conditions of very low force however (usually less than 10 % of peak force, Fig. 2) and motion is correspondingly reduced. Ca²⁺ waves were measured during strong adrenergic contractions (Fig. 4). If cells moved out of the plane of focus in these experiments, the experiment was rejected. If an AOI could be maintained in the same cell, the measurement was accepted.

Purinergic component

In order to study selectively the Ca²⁺ signals and contractions generated by neurally released ATP, arteries were exposed to prazosin (1.0–10.0 μm) to block α₁-adrenergic receptors. After prazosin treatment, 73.7 ± 14.0 % (n = 7) of the initial transient contraction remained and 5.00 ± 0.98 % (n = 7) of the maintained contraction remained. Others (Gitterman & Evans, 2001) have shown previously that purinergic receptor antagonists, such as suramin, abolish these small contractions that remain in prazosin (Fig. 2A, lower trace), and that L-type Ca²⁺ channel blockers are without effect. Similarly, we found that in the presence of prazosin, suramin (300 μm) abolished jCaTs (as reported previously: Lamont & Wier, 2002) as well as contraction. In five arteries also exposed to prazosin, the blocker of L-type Ca²⁺ channels, nifedipine (0.3 μm), had no effect on the amplitude of jCaTs or the contraction they produced. As is the case with sparks, the peak amplitude (F/F₀) of jCaTs was not normally distributed. We therefore express the data as mean, median and range. In prazosin, these were: mean, 2.77; median, 2.56; range, 1.50–6.27 (n = 299). Nifedipine produced no significant change: mean, 2.90; median, 2.71; range 1.47–6.98 (n = 344). The amplitude of the initial contraction in prazosin was not significantly changed by nifedipine (prazosin, 0.19 ± 0.04 mN; prazosin and nifedipine, 0.17 ± 0.04 mN; mean ± s.e.m., n = 5). The results with prazosin, suramin and nifedipine all suggest that jCaTs and accompanying contractions are mediated by purinergic receptors, in agreement with previous studies (Gitterman & Evans, 2001; Lamont & Wier, 2002).

To characterize jCaTs, confocal imaging of fluo-4 fluorescence at 30 images s⁻¹ was performed for three periods of 20 s at the beginning (0–20 s), middle (80–100 s) and end (160–180 s) of 3 min EFS (periods indicated by bars in Fig. 2A). Within the visible area of the arterial wall, typically 110 μm × 88 μm, we located all the sites at which jCaTs occurred (three of which are shown in Fig. 2Ba), and obtained the average fluorescence within an AOI, 0.85 μm × 0.85 μm, located at the site. Fluorescence was converted to a pseudo-ratio (viz. F/F₀) (see Methods) and representative traces from three AOIs are shown in Fig. 2Bb. When AOIs were close together, it was clear that some jCaTs appeared in two or more AOIs, since jCaTs spread over 5 μm (FWHM (full width at half-maximum), 4.68 μm) (Lamont & Wier, 2002). In these cases, it was essential to determine how many jCaTs actually occurred and avoid counting individual jCaTs twice or more. For example, we concluded that the site of origin of the jCaT marked ‘i’ is the AOI marked ‘1’, because it was larger in AOI ‘1’ and smaller in AOI ‘2’, while the opposite was true of jCaT marked ‘ii’. The jCaT marked ‘iii’ occurred in AOI ‘3’, distant from the others, and was clearly distinct from them. In five arteries exposed to prazosin and stimulated with EFS, 549 distinct jCaTs were identified in this fashion and the peak fluorescence ratio and times of occurrence were obtained. The frequency of jCaTs declined markedly during the 3 min of EFS (Fig. 2Cd-f). At sites where at least one jCaT occurred, the average probability of a jCaT was 0.008 ± 0.002 pulse⁻¹ in the first 20 s, 0.0014 ± 0.0002 pulse⁻¹ in the middle 20 s and 0.0007 ± 0.0002 pulse⁻¹ in the last 20 s. Since, however, some sympathetic varicosities may not have produced any jCaTs, this estimate is an upper limit. In contrast to the frequency, the peak amplitude of the jCaTs changed little during 3 min EFS (Fig. 2Ca-c). No propagating Ca²⁺ waves were ever observed during EFS in the presence of prazosin. In one case, jCaTs were recorded continuously at 30 images s⁻¹ throughout 80 s EFS. Throughout this time, the jCaTs occurred in sufficient numbers to produce a detectable elevation of average [Ca²⁺] (fluorescence, Fig. 2D, lower trace), and this paralleled closely the small contraction recorded simultaneously (Fig. 2D). Thus, it seems reasonable to attribute the contractile activation to jCaTs, despite their limited spatial extent and frequency.

In order to determine if the jCaTs originate at the perivascular nerves we measured the distances between the sites of origin of jCaTs (indicated by the white boxes in Fig. 3B) and identified nerve bundles (white lines Fig. 3B). Nerve bundles were identified by focusing 1.2 μm above the level (Fig. 3A) at which jCaTs were identified (Fig. 3B). Fifty per cent of the jCaTs originated at distances less than 0.85 μm from an identified nerve bundle, with the average distance being 1.29 μm (Fig. 3C). On the other hand, in the image shown, the average distance of a pixel from a nerve bundle was 5.57 μm, with 50 % occurring within 3.83 μm of a nerve bundle. Thus, jCaTs arise in close proximity to perivascular nerves.

Adrenergic component

Later during neurogenic contractions, in the absence of prazosin, asynchronous Ca²⁺ waves propagated within individual smooth muscle cells of the arterial wall (Fig. 1B and Fig. 4A). A representative Ca²⁺ signal, obtained as the average fluorescence with an AOI (1.35 μm square) within a single smooth muscle cell is shown in Fig. 4A (lower trace; right-hand scale). For these data, images were obtained at 2 s⁻¹, a rate that is too slow to resolve the jCaTs (Lamont & Wier, 2002) generated during the initial purinergic component. For the analysis of the adrenergic component of the experiment illustrated in Fig. 4, 74 individual smooth muscle cells were identified and an AOI was placed on each. In these 74 cells, 809 Ca²⁺ waves were detected (see Methods) and the time of onset and peak amplitude of each was determined. The histogram of times of occurrence (Fig. 4B) of these 809 Ca²⁺ waves reveals that very few occurred during the initial 20 s of EFS. The histogram of times of wave occurrence shows that wave frequency increased, reaching a peak by about 120 s, while force was still rising and then declined slightly. Ca²⁺ waves that occurred either early in the EFS, or after it's cessation tended to be smaller than those that occurred at other times (Fig. 4C). The line fit to the data (during EFS only) in Fig. 4C is a 2nd order polynomial, which provided a significantly better fit (χ² goodness of fit) than did a straight line. The Ca²⁺ signal exemplified in Fig. 4A was representative: Ca²⁺ waves begin only after a substantial delay and are initially relatively small. The peak amplitude and frequency increases, then both decline slightly, ceasing after EFS ceases. In five arteries subjected to the same EFS, the average number of waves per cell in the initial 20 s was 0.14 ± 0.06, in the period 80–100 s it was 1.55 ± 0.25 and in the period 160–180 s it was 1.14 ± 0.19 (n = 5, 254 cells). The peak amplitudes (F/F₀) in these same periods were 1.69 ± 0.07, 3.34 ± 0.06 and 2.68 ± 0.07, respectively.