INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

NEUROTRANSMITTERS INDUCE contraction of smooth muscle cells initially by mobilizing Ca²⁺ from the intracellular Ca²⁺ store through inositol 1,4,5-trisphosphate (InsP₃)-gated Ca²⁺ channels (11, 31). Recent data have suggested that intracellular Ca²⁺ signals are organized as a hierarchy (4, 6, 8, 18, 29). The opening of individual Ca²⁺ release channels gives rise to fundamental events referred to as blips in the case of InsP₃ receptors or quarks for the ryanodine receptors. The next level of organization is represented by small groups of InsP₃- or ryanodine-sensitive channels releasing Ca²⁺ as localized units to give puffs and sparks, respectively. Propagated Ca²⁺ waves may be obtained by recruitment of a variable number of these elementary events. Typical examples of homologous hierarchical Ca²⁺-signaling systems have been demonstrated in nonexcitable and excitable cells (4, 6, 18).

In smooth muscle, Ca²⁺ signals can be generated by InsP₃ and ryanodine receptors (31), and there are indications that these two Ca²⁺ release channel types are located on the same intracellular store in some smooth muscle cells (15). In vascular myocytes, Ca²⁺ sparks may appear spontaneously (2, 23, 28) or may be triggered by activation of L-type Ca²⁺ currents (1). Repetitive activation of Ca²⁺ sparks associated with the progressive recruitment of isolated Ca²⁺ release channels is needed to trigger propagated Ca²⁺ waves in rat portal vein myocytes (1). In these myocytes, elementary Ca²⁺ events, similar to Ca²⁺ puffs, have never been observed by increasing the InsP₃ concentration by flash photolysis of the caged compound (5). Moreover, when ryanodine receptors are inhibited, activation of InsP₃ receptors does not trigger propagated Ca²⁺ waves (5). Therefore, it remains to be established whether a cooperativity between InsP₃- and ryanodine-sensitive Ca²⁺ release channels to induce propagated Ca²⁺ waves in response to activation of membrane receptors is a physiological mechanism in smooth muscle cells.

The present study shows that, in vascular myocytes, norepinephrine activates InsP₃- and ryanodine-sensitive channels to induce propagated Ca²⁺ waves and maximal contraction. After an initial Ca²⁺ release from InsP₃ receptors, Ca²⁺ sparks are activated, leading to an increase in amplitude and upstroke velocity of the norepinephrine-induced Ca²⁺ responses.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Cell preparation, solutions, and membrane current recordings. Wistar rats (150-160 g) were stunned and killed by cervical dislocation. The portal vein and duodenum were removed quickly, cut into several pieces, and incubated for 10 min in low-Ca²⁺ (40 µM) physiological solution, then 0.8 mg/ml collagenase, 0.25 mg/ml pronase E, and 1 mg/ml BSA were added at 37°C for 20 min. The solution was removed, and the pieces of vein were incubated again in a fresh enzyme solution at 37°C for 20 min. Tissues were then placed in enzyme-free solution and triturated using a fire-polished Pasteur pipette to release cells. Cells were stored on glass coverslips and used on the same day or maintained in short-term primary culture in medium 199 containing 5% FCS, 2 mM glutamine, 1 mM pyruvate, 20 U/ml penicillin, and 20 µg/ml streptomycin; they were kept in an incubator gassed with 95% air-5% CO₂ at 37°C and used within 30 h.

Confocal microscopy and fluorescence measurements. A confocal scanning head (model MRC 1000, Bio-Rad, Paris, France) was coupled to an inverted microscope (Diaphot, Nikon, Tokyo, Japan). In all experiments a Nikon Plan Apo ×60, 1.4 NA objective lens was used. The iris aperture was set to 40-50% of maximum, providing axial (z) resolution of ~1.5 µm and x-y resolution of 0.4 µm. Illumination was provided by a 25-mW argon ion laser (Ion Laser Technology, Salt Lake City, UT). The excitation wavelength (488 nm) was selected using interference filters. For some experiments, two fluorescent dyes, fluo 3 and fura red (30 µM each) were dialyzed into the cells through the patch pipette, as previously reported (1, 2). The emitted fluorescence was collected at wavelengths >515 nm, and fluo 3 and fura red fluorescences were separated by a dichroic mirror. Each fluorescence beam was filtered and detected by photomultiplier tubes, allowing simultaneous measurements of the fluorescence emitted by fluo 3 and fura red for dual-emission imaging (19). Dividing the fluorescence records pixel by pixel resulted in ratio images. For calibrating the fluo 3/fura red fluorescence, cells were dialyzed with pipette solutions adjusted to six different free ionized Ca²⁺ concentrations with appropriate ratios of EGTA to CaCl₂ (final EGTA concentration was always 10 mM). Fluorescence ratios were measured when Ca²⁺ equilibration was reached (within 5-8 min). The calibration curve was fitted using a least-squares analysis program. The operational dissociation constant (K_d) for the fluo 3-fura red mixture was 260 nM on our setup. In flash photolysis experiments with caged InsP₃, fluo 3 (60 µM) was used alone. For other experiments, cells were loaded by incubation in physiological solution containing 1 µM fluo 3-AM for 1 h at room temperature. These cells were washed and allowed to cleave the dye to the active fluo 3 compound for 1 h. In the presence of fluo 3 alone, intracellular Ca²⁺ concentration ([Ca²⁺]_i) was estimated from the fluorescence ratio (R, calculated as F/F_rest, where F is fluorescence and F_rest is resting fluorescence) by use of the following equation: [Ca²⁺]_i = K_d · R/{(K_d/[Ca²⁺]_rest + 1)  R} (8), where K_d is the dissociation constant of the indicator (316 nM) and [Ca²⁺]_rest is resting [Ca²⁺]_i, estimated at 45 nM in control conditions and 55 nM in the presence of 10 µM caged InsP₃. Image acquisition and data analysis were performed by using COMOS, TCSM, and MPL-1000 software (Bio-Rad). Images were acquired in the line-scan mode of the confocal microscope; this mode repeatedly scanned a single line through the cell every 2 ms. In this line-scan image the spatial average fluorescence can be measured in a 2-µm region on the x-axis, illustrating temporal changes of [Ca²⁺]_i in cell volumes of ~1 µm³.

Flash photolysis. Caged InsP₃ [D-myo-InsP₃, P⁴⁽⁵⁾-1-(2-nitrophenyl)ethyl ester] or caged Ca²⁺ (DM-nitrophen) was introduced into the cell via the patch pipette, with 3-4 min allowed for equilibration. Photolysis was produced by a 1-ms pulse from a xenon flash lamp (Hi-Tech Scientific, Salisbury, UK) focused to a ~2-mm-diameter spot around the cell. Light was band-pass filtered with a UG11 glass between 300 and 350 nm. Flash intensity could be adjusted by varying the capacitor-charging voltage between 0 and 385 V, which corresponded to a change in the energy input into the flash lamp from 0 to 240 J. On flash photolysis, Ca²⁺ or InsP₃ was released within 2 ms. A small percentage of conversion of caged compounds (~10%) was useful if repetitive pulses were applied to obtain similar responses. Flash intensities up to 37 J could be applied repetitively without altering the reserve of caged InsP₃ (10 µM) or DM-nitrophen (1 mM, in the presence of 0.25 mM CaCl₂) and, consequently, the amount of photoreleased compounds.

Immunocytochemistry. Myocytes were immunostained as previously described (21), except donkey serum was used instead of FCS. Myocytes were incubated in the presence of anti-ryanodine and anti-InsP₃ receptor antibodies (at 1:100 and 1:200 dilution, respectively) for 20 h at 4°C, and the secondary antibodies [donkey anti-rabbit IgG conjugated to FITC or donkey anti-mouse IgG conjugated to tetramethylrhodamine isothiocyanate (TRITC) diluted 1:200] were incubated for 3 h at 20°C. Thereafter, cells were mounted in Vectashield.

Contraction. Isometric contraction of longitudinal strips from rat portal vein were recorded in an experimental chamber, as described previously (24), by means of a highly sensitive isometric force transducer (model 801 AME, Akers). The circulating physiological solution was maintained at 30 ± 1°C.

Microsomal membrane preparation. Microsomal membranes of portal vein, heart, and cerebellum from Wistar rats were prepared by homogenization with a Kontes potter in a solution containing 20 mM Tris · HCl, 1 mM EGTA, and 1 mM phenylmethylsulfonyl fluoride, pH 7.4. The homogenate was centrifuged at 1,200 rpm for 10 min at 4°C. Microsomal membranes were obtained as a pellet by centrifugation of the supernatant at 40,000 rpm for 90 min at 4°C. Microsomal membranes were then resuspended in the buffer and stored at 80°C. Protein concentration was determined according to Bradford (7).

Immunoblotting. For Western blotting analysis, microsomal proteins were separated on 5% SDS-PAGE minigels and transferred to polyvinylidene difluoride membranes for 16 h at 30 V in a transfer buffer containing 192 mM glycine and 25 mM Tris · HCl (pH 8.3). Membranes were blocked for 1 h in blocking buffer containing 20 mM Tris · HCl and 3% BSA (pH 7.4) and then incubated overnight with the primary antibody at 1:200 (rabbit anti-InsP₃ receptor antibody) or 1:100 (mouse anti-ryanodine receptor antibody) dilution. After extensive washing, membranes were incubated for 2 h with the secondary antibody coupled to the peroxidase (anti-rabbit or anti-mouse, 1:5,000 dilution). Specific antigen detection was performed using H₂O₂ and diaminobenzidine to detect peroxidase activity on polyvinylidene difluoride membranes and the Kodak EDAS 120 (Rochester, NY).

[³H]ryanodine and [³H]InsP₃ binding assays. [³H]ryanodine binding to microsomal membranes of rat portal vein was measured, as reported previously (26), in a medium containing 1 M KCl, 25 mM HEPES (pH 7.8 at 37°C), 1 mM dithiothreitol, 0.1 mM CaCl₂, 1 mg/ml BSA, and 0.1 mM phenylmethylsulfonyl fluoride. [³H]ryanodine was used in the concentration range of 2-30 nM. After a 3-h incubation at 37°C, aliquots were filtered through Whatmann GF/C glass fiber filters and washed three times with 5 ml of ice-cold binding buffer. The filters were placed in scintillation vials filled with 4 ml of liquid scintillation cocktail, shaken for 1 h, and counted in a Packard 1500 Tri-Carb. Nonspecific binding was measured in the presence of 10 µM ruthenium red and subtracted before calculation. At 20 nM [³H]ryanodine, nonspecific binding was <50% of total binding.

Chemicals and drugs. Collagenase was obtained from Worthington (Freehold, NJ); pronase E, BSA, norepinephrine, ACh, prazosin, heparin, ruthenium red, D-myo-InsP₃, Triton X-100, and sodium azide from Sigma Chemical (St. Louis, MO); medium 199 from Flow Laboratories (Puteaux, France); FCS from Flobio (Courbevoie, France); ryanodine, indo 1-AM, caged InsP₃ [D-myo-InsP₃, P⁴⁽⁵⁾-1-(2-nitrophenyl)ethyl ester], and DM-nitrophen from Calbiochem (Meudon, France); caffeine from Merck (Nogent sur Marne, France); fluo 3, fluo 3-AM, and fura red from Molecular Probes Europe (Leiden, The Netherlands); and [³H]ryanodine (68 Ci/mmol) and [³H]InsP₃ (20 Ci/mmol) from Du Pont NEN (Boston, MA). Oxodipine was a gift from Dr. Galiano (Instituto de Investigación y Desarrollo Químico Biológico, Madrid, Spain).

Analysis of data. Values are means ± SE. Significance was tested by Student's t-test. P < 0.05 was considered significant. Concentration-response curves were analyzed by a nonlinear least-squares fitting program. Ca²⁺ responses were fitted by the Boltzmann equation. Binding data were analyzed with the program GraphPad Prism (version 2.0, GraphPad Software, San Diego, CA) for one- or two-site model.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Characterization of anti-InsP₃ and anti-ryanodine receptor antibodies on InsP₃- and Ca²⁺-induced Ca²⁺ releases. To study the role of InsP₃ and ryanodine receptors in triggering propagated Ca²⁺ waves in smooth muscle cells, we first analyzed the properties of antibodies directed against the COOH terminus of InsP₃ and ryanodine receptors. Because these antibodies recognize all the isoforms of each receptor, we chose to use rat heart and cerebellum as positive controls. Figure 1A shows a Western blot analysis on samples of rat portal vein, heart, and cerebellum. The anti-ryanodine receptor antibody recognized a high-molecular-weight band (~500,000) in cardiac and portal vein membranes, whereas the anti-InsP₃ receptor antibody recognized a lower-molecular-weight band (~250,000) in cerebellum and portal vein membranes. No cross-reactivity of the antibodies with the two proteins was detectable. These values are in good agreement with those previously reported for ryanodine and InsP₃ receptors (3, 9). Immunodetection of InsP₃ and ryanodine receptors in 0.5-µm cell confocal sections was performed with these antibodies, and the binding sites were detected with FITC and TRITC, respectively. As illustrated in Fig. 1B, the two types of receptors appeared to be distributed in the whole confocal sections, with spots of ryanodine receptors in areas corresponding to the cell periphery and to infoldings of the plasma membrane in close association with the sarcoplasmic reticulum.

View larger version (59K): [in this window] [in a new window]   Fig. 1.   Immunologic detection of inositol 1,4,5-trisphosphate (InsP3R) and ryanodine (RyR) receptors. A: microsomes of rat portal vein (200 µg), heart (100 µg), and cerebellum (50 µg) were separated on 5% SDS-PAGE and analyzed by Western blot with anti-InsP3 or anti-ryanodine receptor antibodies (anti-InsP3 Ab and anti-RyR Ab, respectively). Molecular mass of myosin is indicated. B, top: immunolocalization of InsP3 and ryanodine receptors in a confocal cell section of a rat portal vein myocyte prepared by double-staining protocol. In absence of primary antibodies or after inactivation of antibodies by their antigen peptides, only a faint background fluorescence was observed. B, bottom: part of cell section (white box) with a higher magnification illustrating spots of ryanodine receptors in vicinity of caveolae (C) and in cell periphery and homogeneous distribution of InsP3 receptors in same areas.

View larger version (37K): [in this window] [in a new window]   Fig. 2.   Effects of anti-ryanodine receptor antibody on increases in intracellular Ca2+ concentration ([Ca2+]i) evoked by flash-photolytic Ca2+ jumps in voltage-clamped portal vein myocytes. A: Ca2+ response evoked by flash pulse of 37 J shown as a line-scan image. UV, ultraviolet light. B: spatial averaged fluorescence from a 2-µm region of line-scan image in presence of intracellular Ca2+/DM-nitrophen (trace a), in presence of intracellular Ca2+/DM-nitrophen and 10 µg/ml anti-ryanodine receptor antibody for 7-8 min (trace b), and in absence of both compounds (trace c). Trace a-b, time course of Ca2+-induced Ca2+ release (CICR) component. C: effects of increasing concentrations of anti-ryanodine receptor antibody, 10 µg/ml boiled anti-ryanodine receptor antibody, and 10 µg/ml anti-InsP3 receptor antibody on peak Ca2+ responses evoked by 37-J flash pulses. Values are means ± SE, with number of experiments in parentheses.  Significantly different from control (P < 0.05). Myocytes were loaded with fluo 3-fura red mixture and caged Ca2+ (DM-nitrophen) and held at 50 mV. External solution contained 1 µM oxodipine.

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Increase in [Ca2+]i evoked by flash photolysis of caged D-myo-InsP3 in voltage-clamped portal vein myocytes. A: Ca2+ response evoked by a 37-J flash pulse shown as a line-scan image. B: spatial averaged fluorescence from a 2-µm region of line-scan image in presence of intracellular caged InsP3 (trace a), in presence of intracellular caged InsP3 and 10 µg/ml anti-ryanodine receptor antibody (trace b), and in absence of both compounds (trace c). Trace a-b, time course of CICR component. C and D: effects of increasing concentrations of anti-InsP3 receptor antibody and 10 µg/ml boiled anti-InsP3 receptor antibody (C) and anti-ryanodine receptor antibody and 10 µg/ml boiled anti-ryanodine receptor antibody (D) on peak InsP3-induced Ca2+ responses. Antibodies were applied intracellularly for 7-8 min. Values are means ± SE, with number of experiments in parentheses.  Significantly different from control (P < 0.05). Myocytes were loaded with fluo 3 and caged InsP3 and held at 50 mV. External solution contained 1 µM oxodipine.

A common intracellular Ca²⁺ store activated by caffeine and neurotransmitters. The experiments described below were carried out under Ca²⁺-free conditions to eliminate the contribution of extracellular Ca²⁺, so that the release of Ca²⁺ from intracellular stores could be resolved. Repeated stimulations by 10 µM ACh or norepinephrine, which induced InsP₃ accumulations (16) on indo 1-loaded myocytes, did not produce a second rise in [Ca²⁺]_i after recovery in Ca²⁺-free solution (n = 5; Fig. 4A). Similarly, successive applications of 10 mM caffeine, which acted at the ryanodine receptors, produced a first Ca²⁺ response, whereas a second application was ineffective (n = 5; Fig. 4A). To determine the extent of overlap between ACh- and caffeine-sensitive Ca²⁺ stores, cells were exposed to combinations of agents to determine whether prior exposure to one agent would diminish a subsequent response to the other agent. As illustrated in Fig. 4B, prior exposure to ACh completely eliminated subsequent responses to caffeine (n = 4). Similarly, prior applications of caffeine suppressed subsequent responses to ACh (n = 4).

View larger version (19K): [in this window] [in a new window]   Fig. 4.   Effects of caffeine and ACh on intracellular Ca2+ store in portal vein myocytes. A: in Ca2+-free, 0.5 mM EGTA-containing solution, Ca2+ responses evoked by 2 successive applications of 10 µM ACh or 10 mM caffeine (Caf). B: after an ACh- or caffeine-induced Ca2+ response in Ca2+-free solution, further applications of 10 mM caffeine or 10 µM ACh were ineffective. C and D: after pretreatment with 10 µM ryanodine for 30 min, first 10 mM caffeine application induced a Ca2+ response, but second application was ineffective (C). Similarly, no response was evoked with a second application of 10 µM ACh (D).

Effects of anti-InsP₃ and anti-ryanodine receptor antibodies on norepinephrine-induced Ca²⁺ release. To verify whether InsP₃ and ryanodine receptors participate in the Ca²⁺ responses evoked by neurotransmitters, we tested the effects of these antibodies on the Ca²⁺ release evoked by various concentrations of norepinephrine in the continuous presence of 1 µM oxodipine. Under these conditions, the norepinephrine-induced Ca²⁺ release was inhibited by prazosin in a concentration-dependent manner (data not shown), indicating that this response was activated by ₁-adrenoceptors, as previously demonstrated (17). Figure 5 illustrates line-scan images of voltage-clamped myocytes and the time courses of the Ca²⁺ responses evoked by norepinephrine when increases in [Ca²⁺]_i were analyzed in 2-µm regions. The delay between application of norepinephrine and the onset of the Ca²⁺ response was estimated to be 1.1 ± 0.1 s (n = 10). At 1 µM, norepinephrine induced localized and transient Ca²⁺ responses (n = 13; Fig. 5A, see Fig. 7) or propagated Ca²⁺ waves (n = 5; see Fig. 7). In the presence of 1 µM norepinephrine, spatially localized [Ca²⁺]_i transients are obtained in >70% of the cells (n = 18). It can be postulated that norepinephrine stimulates these Ca²⁺ signals rather than increases the number of spontaneous Ca²⁺ releases, which are detected in only 30% of the cells tested in control conditions (23). When norepinephrine was applied at 10 µM, propagated Ca²⁺ waves were recorded in all the cells tested (n = 19; Fig. 5B). Intracellular application of an anti-ryanodine receptor antibody (10 µg/ml) for 7-8 min removed the localized and transient Ca²⁺ responses activated by 1 µM norepinephrine; however, a small and slow increase in [Ca²⁺]_i persisted (Fig. 5C). This antibody also transformed the norepinephrine-induced propagated Ca²⁺ waves in reduced Ca²⁺ responses showing a uniform increase in [Ca²⁺]_i, whatever the region of the line-scan image (Fig. 5D). These results indicate that the localized and transient Ca²⁺ responses activated by 1 µM norepinephrine and removed by the anti-ryanodine receptor antibody correspond to Ca²⁺ sparks, as previously identified in these vascular myocytes (1, 2). They also show that propagated Ca²⁺ waves evoked by norepinephrine are dependent on activation of ryanodine-sensitive Ca²⁺ release channels, whereas activation of InsP₃-gated channels alone gives rise to slow and small Ca²⁺ responses showing a similar time course in all regions of the cell.

View larger version (55K): [in this window] [in a new window]   Fig. 5.   Increase in [Ca2+]i evoked by norepinephrine (NE) in voltage-clamped portal vein myocytes. A: transient and localized Ca2+ responses evoked by 1 µM norepinephrine shown as line-scan image and spatial averaged fluorescence from different 2-µm regions (vertical bars). B: Ca2+ wave evoked by 10 µM norepinephrine shown as a line-scan image and spatial averaged fluorescence from 2-µm regions within or outside initiation site of Ca2+ wave (vertical bars). C and D: effects of 10 µg/ml anti-ryanodine receptor antibody (applied intracellularly for 7-8 min) on Ca2+ responses evoked by 1 µM norepinephrine (C) or 10 µM norepinephrine (D) shown as line-scan image and spatial averaged fluorescence from 2-µm regions (vertical bars). Myocytes were loaded with fluo 3-fura red mixture and held at 50 mV. External solution contained 1 µM oxodipine.

View larger version (24K): [in this window] [in a new window]   Fig. 6.   Upstroke velocity of norepinephrine-evoked Ca2+ responses. A: fitting with Boltzmann equation of rising phase of norepinephrine-evoked Ca2+ responses (measured in 2-µm region of line-scan image corresponding to an initiation site) in control conditions (left) or after internal application of 10 µg/ml anti-ryanodine receptor antibody for 7-8 min (right). Arrow, critical [Ca2+]i threshold. B: upstroke velocity obtained in control conditions (left) and in presence of 10 µg/ml anti-ryanodine receptor antibody or 10 µM ryanodine (applied externally and internally) for 7-8 min (right). Values are means ± SE, with number of experiments in parentheses. Myocytes were loaded with fluo 3-fura red mixture and held at 50 mV. External solution contained 1 µM oxodipine.

View larger version (19K): [in this window] [in a new window]   Fig. 7.   Peak sizes of Ca2+ responses plotted against norepinephrine concentration. A: [Ca2+]i was measured in 2-µm region of line-scan image in control conditions (), in presence of 10 µg/ml anti-ryanodine receptor antibody or 10 M ryanodine (), and in presence of 10 µg/ml anti-InsP3 receptor antibody () for 7-8 min. After ryanodine receptor inhibition, Hill coefficient and norepinephrine concentration producing half-maximal response were 1.1 and 3 µM, respectively. B: [Ca2+]i was measured in entire line-scan image in control conditions (), in presence of 10 µg/ml anti-ryanodine receptor antibody or 10 µM ryanodine (), and in presence of 10 µg/ml anti-InsP3 receptor antibody () for 7-8 min. After ryanodine receptor inhibition, Hill coefficient and norepinephrine concentration producing half-maximal response were 1.1 and 3.5 µM, respectively. Values are means ± SE, with number of experiments in parentheses. Myocytes were loaded with fluo 3-fura red mixture and held at 50 mV. External solution contained 1 µM oxodipine.

[³H]ryanodine and [³H]InsP₃ binding on rat portal vein membranes. InsP₃ binding to rat portal vein microsomal preparations was measured over a broad range of InsP₃ concentrations (2-200 nM), but as shown in Fig. 8, Scatchard plots of these data were nonlinear. The InsP₃-binding data could be fitted by assuming the presence of two InsP₃-binding sites. The calculated K_d values were 1.4 and 132.5 nM, with 90% of the binding sites being low affinity. The maximal binding capacity (B_max) corresponding to the total number of InsP₃-binding sites (high-affinity + low-affinity sites) was 1.59 ± 0.21 pmol/mg protein (n = 3). [³H]ryanodine binding was performed on the same membranes used for [³H]InsP₃ binding. The K_d and B_max values of [³H]ryanodine binding were 8.7 ± 0.1 nM and 5.70 ± 0.65 pmol/mg protein (n = 5), respectively. These results show that the B_max of InsP₃ in rat portal vein myocytes is rather low: three to four times less than that of ryanodine.

View larger version (17K): [in this window] [in a new window]   Fig. 8.   Specific binding of [3H]InsP3 to rat portal vein membranes. Inset: saturation binding experiments (83 µg/ml of protein) incubated with increasing concentrations of [3H]InsP3 at 4°C for 10 min. Each point is mean of 3 experiments, each carried out in duplicate. , Total binding; , nonspecific binding; , specific binding. Scatchard analysis of specific binding. B/F, bound/free. Each point is mean of 3 experiments, each carried out in duplicate.

Effects of ryanodine receptor inhibition on Ca²⁺ release evoked by various neurotransmitters in non-voltage-clamped vascular and visceral myocytes. To ensure that the inhibition of norepinephrine-induced Ca²⁺ responses by ryanodine receptor inhibitors was not due to a dilution of the InsP₃ concentration when the cells were dialyzed with the pipette solution, portal vein and duodenal myocytes were loaded with 1 µM fluo 3-AM or indo 1-AM, respectively, and not voltage clamped. In the continuous presence of 1 µM oxodipine, norepinephrine (10 µM) or ACh (10 µM) evoked propagated Ca²⁺ waves, the amplitude and upstroke velocity of which are listed in Table 1. In the presence of 10 µM ryanodine for 10-12 min in the extracellular medium, the amplitude and maximal upstroke velocity of the neurotransmitter-induced Ca²⁺ responses (measured in a 2-µm region of the line-scan image) were significantly reduced (Table 1). In duodenal myocytes the ACh-induced global rise in [Ca²⁺]_i was also reduced (~35%) in the presence of 10 µM ryanodine. Taken together, these results indicate that, in non-voltage-clamped vascular and visceral myocytes, inhibition of ryanodine-sensitive Ca²⁺ release channels results in neurotransmitter-induced Ca²⁺ responses of reduced amplitude.

Effects of ryanodine receptor inhibition on neurotransmitter-induced contraction. Involvement of ryanodine-sensitive Ca²⁺ release channels in neurotransmitter-induced contraction was first investigated by measuring the variation of the scanned line length through the myocyte after application of 10 µM norepinephrine (Fig. 9A). Within 1 s after norepinephrine ejection, the increase in [Ca²⁺]_i was observed without any noticeable variation in the scanned line length (L₀). Then the scanned line length decreased as the myocyte contracted (L₁) before it returned to initial length within 5-10 s (not shown). The variation in cell length was normalized to the resting value and expressed as (L₀  L₁)/L₀. The ratio was equal to zero when L₁ = L₀ and became positive when L₁ < L₀, then reflecting a contraction. After application of 10 µM ryanodine (in non-voltage-clamped myocytes) or 10 µg/ml anti-ryanodine receptor antibody (in voltage-clamped myocytes), the ratio (L₀  L₁)/L₀ was significantly reduced (by ~50%) compared with control conditions (Fig. 9B), suggesting that the norepinephrine-induced contraction was reduced. Second, we recorded isometric contractions from thin isolated strips of portal vein smooth muscle (24). After pretreatment with 10 µM ryanodine for 30 min, the norepinephrine-induced contractions were reduced by ~45% compared with control conditions, as shown in Fig. 9C. These results support the role of a functional interaction between InsP₃- and ryanodine-sensitive Ca²⁺ release channels in the amplitude of the norepinephrine-induced contraction in vascular myocytes.

View larger version (28K): [in this window] [in a new window]   Fig. 9.   Effects of ryanodine and anti-ryanodine receptor antibody on norepinephrine-induced contraction. A: line-scan image recorded in presence of 10 µM norepinephrine showing successively maximal increase in [Ca2+]i and shortening of scanned line. B: plots of (L0  L1)/L0 (where L0 and L1 are lengths of scanned line before and during norepinephrine application, respectively) in control conditions (C) and in presence of 10 µg/ml anti-ryanodine receptor antibody or 10 µM ryanodine for 7-8 min. Myocytes were loaded with fluo 3-fura red mixture and held at 50 mV. C: contractile responses of isolated strips from rat portal vein smooth muscle to 10 µM norepinephrine in control conditions (C) and during application of 10 µM ryanodine for 30 min. Values are means ± SE, with number of experiments in parentheses.  Significantly different from control (P < 0.05). External solution contained 1 µM oxodipine.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The main conclusion from these results is that Ca²⁺ responses evoked by neurotransmitters in rat portal vein and duodenal myocytes depend on activation of InsP₃- and ryanodine-sensitive Ca²⁺ release channels. Evidence supporting this conclusion is as follows. 1) Propagated Ca²⁺ waves evoked by norepinephrine in portal vein myocytes were blocked by intracellular applications of heparin and an anti-InsP₃ receptor antibody. 2) The anti-ryanodine receptor antibody and ryanodine that completely inhibited Ca²⁺ signals evoked by flash-photolytic Ca²⁺ jumps decreased by ~50% the InsP₃- and norepinephrine-induced Ca²⁺ waves without affecting the Ca²⁺ content of the store. 3) When the upstroke velocity of norepinephrine-evoked Ca²⁺ waves was analyzed in 2-µm regions of the line-scan image, the maximal value was obtained in the region of the line-scan image corresponding to the initiation site of the Ca²⁺ wave. Whatever the regions analyzed, inhibition of ryanodine receptors strongly reduced the upstroke velocity to a similar value, indicating that this homogeneous remaining component corresponded to activation of InsP₃-sensitive Ca²⁺ channels. 4) A similar inhibition of the Ca²⁺ responses evoked by ACh was obtained with ryanodine treatment in portal vein and duodenal myocytes. These results show, for the first time, that activation of ryanodine-sensitive Ca²⁺ release channels is necessary for triggering neurotransmitter-evoked propagated Ca²⁺ waves of high amplitude and fast velocity that are, however, initiated by activation of InsP₃-sensitive Ca²⁺ channels.

On the basis of functional experiments and immunodetection of InsP₃ and ryanodine receptors, our results are consistent with the idea that a single intracellular store is mobilized by neurotransmitters and caffeine. First, after depletion of the intracellular Ca²⁺ store in Ca²⁺-free solution by maximal concentrations of caffeine or ACh, subsequent applications of both agents were ineffective. Similarly, when Ca²⁺-ATPases were blocked with thapsigargin, which induced Ca²⁺ leak from the intracellular Ca²⁺ store, ACh- and caffeine-induced Ca²⁺ responses were abolished. Second, after pretreatment of cells in Ca²⁺-containing solution with ryanodine and caffeine to deplete and prevent refilling of the intracellular Ca²⁺ store, the responses to ACh and caffeine were abolished. In addition, the basal [Ca²⁺]_i was increased in the presence of ryanodine and caffeine, suggesting that depletion of the Ca²⁺ store leads to increased Ca²⁺ entry into the cell (25). Third, immunodetection of InsP₃ and ryanodine receptors in 0.5-µm cell confocal sections showed that the two types of receptors were distributed in the whole sections. However, spots of ryanodine receptors are detected in several areas, whereas InsP₃ receptor immunostaining did not reveal any fluorescence spots. Therefore, it can be postulated that the density of ryanodine receptors may be higher than that of InsP₃ receptors. This hypothesis is supported by binding experiments to rat portal vein microsomal preparations which indicate that the B_max is three to four times higher for [³H]ryanodine than for [³H]InsP₃. [³H]InsP₃ and [³H]ryanodine binding sites have been identified in guinea pig ileal smooth muscle, but with InsP₃ receptors more abundant than ryanodine receptors (33). However, most of the [³H]InsP₃ binding was detected in circular muscle cells, whereas the binding in longitudinal muscle cells was weak (27). Therefore, it can be proposed that the functional implications of ryanodine receptors in neurotransmitter-induced Ca²⁺ responses depend on their relative density compared with InsP₃ receptor density. In smooth muscles displaying a higher density of InsP₃ receptors than of ryanodine receptors, the Ca²⁺ responses to neurotransmitters could mainly depend on activation of InsP₃ receptors. Comparative experiments in a greater number of tissues are, however, needed to expand this conclusion to all smooth muscles.

All-or-none Ca²⁺ responses evoked by neurotransmitters and flash-photolytic InsP₃ jumps have been previously reported in permeabilized basophilic leukemia cells (22) and guinea pig portal vein smooth muscle cells (12). These observations have been explained by a Ca²⁺-dependent positive-feedback control of InsP₃-induced Ca²⁺ release (12). This possibility seems unlikely in rat portal vein myocytes, since concentration-response curves to norepinephrine with Hill coefficients close to unity are obtained after blockade of ryanodine receptors with the anti-ryanodine receptor antibody or with high concentrations of ryanodine. In addition, the amplitude and the upstroke velocity of the norepinephrine-evoked Ca²⁺ responses were strongly reduced when ryanodine receptors were blocked. Activation of ryanodine channels by Ca²⁺ (CICR component) leads to Ca²⁺ responses with a high upstroke velocity (~7 µM/s) but limited amplitude (~100 nM). The fast upstroke velocity may be linked to a quasi-instantaneous release of Ca²⁺ in the whole scanned line that synchronously activates all the ryanodine-sensitive channels, whereas the limited amplitude of the Ca²⁺ response may depend on a very brief release of Ca²⁺ by flash photolysis of caged Ca²⁺ and on the Ca²⁺ buffering power of the cytoplasm. Accordingly, the amplitude of the maximal CICR component is reduced by ~30-50% within 0.7 s after the photolytic Ca²⁺ jumps (Figs. 2B and 3B). Interestingly, the CICR component obtained by subtracting the InsP₃-induced Ca²⁺ response in the presence of anti-ryanodine receptor antibody from that in the absence of the antibody shows time course and amplitude parameters similar to those of the CICR component obtained with flash photolysis of caged Ca²⁺. In contrast, the specific InsP₃-induced Ca²⁺ response (obtained in the presence of the anti-ryanodine receptor antibody) is slower (~0.3 µM/s) and more maintained than the CICR component. These observations are in agreement with the fact that, when InsP₃ is released by flash photolysis or by activation of phospholipase C in response to receptor stimulation, the maximal amplitude of the InsP₃-induced Ca²⁺ response seems to correspond to the addition of InsP₃ and CICR components. However, the maximal upstroke velocity of the CICR component activated by flash photolysis of caged InsP₃ or norepinephrine is slower (~3.5 µM/s) than that activated by flash photolysis of caged Ca²⁺ (~7 µM/s). A possible explanation is that a diffuse activation of discrete InsP₃-sensitive Ca²⁺ channels does not provide a sufficient Ca²⁺ trigger to activate all the ryanodine-sensitive channels across the scanned line. In addition, when a critical [Ca²⁺]_i threshold has been reached to open the ryanodine-sensitive channels (Fig. 3B), a part of the Ca²⁺ store has been released as a result of the previous InsP₃-sensitive channel opening. Inasmuch as the open probability of the ryanodine-sensitive channels has been shown to be modulated by the luminal Ca²⁺ concentration (20), a decrease in the Ca²⁺ content might account for the decrease in the upstroke velocity of the Ca²⁺ responses evoked by flash photolysis of caged InsP₃ or norepinephrine. In addition, the observation that the InsP₃-induced Ca²⁺ release reduces the upstroke velocity of the CICR component supports the existence of a single functional Ca²⁺ store. Our results also suggest that the InsP₃ receptors are less sensitive to Ca²⁺ than the ryanodine receptors. This possibility is supported by the observation that the [Ca²⁺]_i threshold for triggering Ca²⁺ sparks in response to activation of L-type Ca²⁺ current or flash-photolytic Ca²⁺ jumps has been estimated to be 75-95 nM in rat portal vein myocytes (1). A critical [Ca²⁺]_i threshold for inducing Ca²⁺-dependent positive feedback of InsP₃-induced Ca²⁺ has been estimated at ~150-180 nM in various cell types (6, 13). This value is larger than the [Ca²⁺]_i level corresponding to the transition between the two upstroke velocity components of the norepinephrine-induced Ca²⁺ waves obtained in some cells (~95 nM). Therefore, our results support the idea of a positive cooperativity between InsP₃- and ryanodine-sensitive Ca²⁺ channels in rat portal vein and duodenal myocytes; i.e., Ca²⁺ release through ryanodine-sensitive channels is critical for generation of Ca²⁺ waves evoked by neurotransmitters that are known to induce an increase in InsP₃ concentration (16). The cooperativity between InsP₃- and ryanodine-sensitive Ca²⁺ channels may not only depend on their relative proportion but also on their topical organization in the sarcoplasmic reticulum. Immunodetection of Ca²⁺ release channels in confocal sections shows that InsP₃ receptors are distributed homogeneously on the sarcoplasmic reticulum, even within the specialized areas showing clusters of ryanodine receptors. Elementary Ca²⁺ events, such as Ca²⁺ sparks, have been identified in smooth muscle cells and attributed to the opening of clusters of ryanodine-sensitive channels (1, 28). In contrast, elementary Ca²⁺ events corresponding to activation of clusters of InsP₃-gated Ca²⁺ channels have never been observed in portal vein myocytes (5), suggesting that these cells may lack clustered InsP₃ receptor units. Therefore, it is likely that the small increase in [Ca²⁺]_i due to opening of InsP₃-gated channels is able to activate the ryanodine receptors located in the vicinity of these InsP₃ receptors, then producing Ca²⁺ sparks and the subsequent Ca²⁺ waves. Such a microarchitecture of ryanodine and InsP₃ receptors has been proposed in the sarcoplasmic reticulum of rat portal vein myocytes (5) but remains, however, to be established in other smooth muscle cells. The physiological role of the Ca²⁺-amplifying mechanism described in this study is also supported by contraction experiments, in isolated cells and intact strips, showing that the norepinephrine-induced contractions are reduced by ~40-50% after inhibition of the ryanodine-sensitive Ca²⁺ release channels.

In conclusion, we have shown that, in portal vein and duodenal myocytes, ryanodine-sensitive Ca²⁺ release channels are involved in Ca²⁺ responses initiated by an increase in cytosolic InsP₃. This amplifying mechanism may be involved in a variety of cells expressing both types of Ca²⁺ release channels located on the same intracellular store in response to various constrictors that stimulate InsP₃ generation.