INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

ACTIVATION OF ₁-adrenergic receptors by norepinephrine or its structural analogs typically results in the constriction of blood vessels. Such action in the venous vasculature represents a physiologically important mechanism in regulating venous return and ultimately the cardiac output of the heart. van Breemen's lab (19) has previously demonstrated that phenylephrine (PE)-induced constriction of the rabbit inferior vena cava (IVC) is signaled and modulated by repetitive intracellular Ca²⁺ waves or intracellular Ca²⁺ concentration ([Ca²⁺]_i) oscillations in the individual in situ vascular smooth muscle cells (VSMC) within the media of the intact vessel. Such repetitive Ca²⁺ waves or [Ca²⁺]_i oscillations appear to be a ubiquitous signaling mechanism for tonic contraction in various blood vessels because similar agonist-induced wavelike [Ca²⁺]_i oscillations have been observed in both the resistance vessels such as the rat tail artery (1, 8), the rat mesenteric artery (15, 18), and the conduit vessel like the rat aorta (1). Mechanistically, we have reported that Ca²⁺ used to generate these repetitive Ca²⁺ waves is immediately derived from the sarcoplasmic reticulum (SR). These [Ca²⁺]_i oscillations are the result of repetitive cycles of SR Ca²⁺ release followed by SR Ca²⁺ store refilling (9, 20). To maintain the asynchronous [Ca²⁺]_i oscillations that signal the tonic contraction, stimulated Ca²⁺ entry from the extracellular space is required for the repetitive refilling of the SR Ca²⁺ store. Ca²⁺ entry involved in the maintenance of [Ca²⁺]_i oscillations and venoconstriction is mediated by a putative nonselective cationic channel (NSCC) component, which is coupled to Na⁺/Ca²⁺ exchange, and by a L-type voltage-gated Ca²⁺ channel (VGCC) component (9). However, questions regarding both the identity of this putative NSCC as well as the mechanism(s) of activation of the NSCC and the L-type VGCC remain unanswered. We did observe that the putative inositol 1,4,5-trisphosphate (InsP₃)-sensitive Ca²⁺ release channel (IP₃R) channel blocker 2-aminoethoxydiphenyl borate (2-APB) can abolish the Ca²⁺ signal in response to PE, and this indicates that the opening of the IP₃R channel is the very first requisite event in the generation of these [Ca²⁺]_i oscillations. But the use of this compound has been recently criticized for its nonspecific effects on other ion transport mechanisms in cultured cell preparations (3, 4, 13, 16). Thus precise interpretation of the findings is difficult without proper controls. As for the identity of the putative nonselective cationic channels, the potential candidates are the receptor-operated channel (ROC) and the store-operated channels (SOC). Given the reports that certain transient receptor potential (Trp) molecules, which have been shown to be the molecular substrates for the SOC, are expressed in mammalian blood vessels (14, 26) and given that the [Ca²⁺]_i oscillations involve repetitive cycles of SR store emptying followed by store refilling (20), it is highly likely that the SOC may also be present in the rabbit IVC and may be involved in refilling the SR Ca²⁺ store to sustain the [Ca²⁺]_i oscillations and tonic contraction in the PE-stimulated rabbit IVC.

In this paper, we report that both the opening of the IP₃R channels and the SOC are required for PE-mediated smooth muscle [Ca²⁺]_i oscillations and tonic constriction of the rabbit IVC. In addition, correlative evidence also indicates that this SOC may be encoded at least in part by the Trp1 gene, which is highly transcribed in the smooth muscle of the rabbit IVC.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Tissue preparation and contraction study. The use of animals for this study complies with the regulations of the local animal ethics committee. Detailed methods have been previously described for this preparation (9, 19). Briefly, IVCs from male New Zealand White rabbits (Animal Care, University of British Columbia) were used. The vessels were then turned inside out, and the endothelium was completely removed by rubbing the luminal surface of the inside-out vessels with filter paper as shown previously by van Breemen's laboratory (19). The endothelium-denuded, inside-out vessels were then cut into small rings of ~4 mm in width. Vessel contractility was monitored with a force transducer in the small rings of IVC under isometric conditions in 37°C physiological salt solution (PSS) bubbled with 100% O₂. Data were acquired and analyzed using Chart v3.4.5 (ADI instruments). All the contraction traces shown here represent findings from a minimum of eight tissues. One sample nonparametric test (Wilcoxon Signed-Rank test) was used to assess statistical significance.

Confocal [Ca²+]_i imaging. For Ca²⁺ studies, rings of IVC were loaded with 5 µM fluo-3 AM (with 5 µM pluronic F-127) for 90 min at 25°C and then left to equilibrate for 30 min in normal PSS. Details regarding image collection and analysis have been previously described by van Breemen and colleagues (9, 19). The tissue was isometrically mounted on a purpose-built microscope stage. [Ca²⁺]_i changes were recorded using a Noran Oz laser scanning confocal microscope. The luminal surface of the inside-out rings of IVC was illuminated using the 488-nm line of an argon-krypton laser, and a high-gain photomultiplier tube collected the emission after it had passed through a 525/52 BP filter. The emission fluorescence (F525) collected reflects [Ca²⁺]_i. The representative fluorescence traces shown in this report reflect the averaged fluorescence signals from a 3 × 3 pixels region (1.36 µm²) of the ribbon-shaped vascular smooth muscle cell. Changes in fluorescence intensity directly reflect changes in the [Ca²⁺]_i. Numerical data were analyzed in Excel and Sigma Plot. One-sample nonparametric tests (Wilcoxon signed-rank test) was used to assess statistical significance.

RNA extraction. Total cellular RNA from endothelium-denuded rings of rabbit IVC was extracted using a RNeasy Mini Kit according to manufacturer's instructions. It is important to note that all the dissection equipment was pretreated with RNAseZap before use. RNA was quantified by measuring absorbance spectrophotometrically at 260 nm, and its integrity was assessed after electrophoresis in nondenaturing 1% agarose gels stained with ethidium bromide.

RT-PCR. Reverse transcription of 5 µg total RNA was performed in 60-µl reaction volumes containing 200 units of Superscript II reverse transcriptase, 60 units RNase inhibitor, 3 mM MgCl₂, 1× buffer II (Sigma), 0.3 µg random primers, and 1 mM dNTP for 50 min at 42°C. Contaminating genomic DNA present in the RNA preparations was removed by digesting the reaction with 5 units of DNase I for 45 min at 37°C before the addition of reverse transcriptase. Five microliters of the RT product were used in each 100-µl PCR reaction. The PCR mixture contained 250 µM dNTP, 2 mM MgCl₂, 1× volume of buffer, and 2.5 unit Hotstar Taq polymerase, 1 µl of forward primers (100 ng/µl), and 1 µl of reverse primers (100 ng/µl). The PCR program for the amplification was 40 cycles of 1 min at 94°C, 1 min at 55°C, and 1 min at 72°C. The final extension was completed at 72°C for 7 min.

Solutions and chemicals. Normal physiological salt solution (PSS) containing (in mM) 140 NaCl, 5 KCl, 1.5 CaCl₂, 1 MgCl₂, 10 glucose, and 5 HEPES, (pH 7.4 at 37°C) was used for all the studies. High K⁺ (80 mM extracellular K⁺) PSS is identical in composition to normal PSS with the exception of (in mM) 65 NaCl and 80 KCl. Fluo-3 AM and pluronic F-127 were purchased from Molecular Probes and were dissolved in dimethyl sulfoxide (DMSO). PE (Sigma), caffeine (Sigma), phentolamine (Sigma), SKF-96365 (Calbiochem), NiCl₂ (Sigma), and LaCl₃ (Calbiochem) were prepared in normal PSS. Stocks of nifedipine (Sigma) and 2-APB (Sigma) were prepared in ethanol, and stocks of cyclopiazonic acid (CPA, Calbiochem) were prepared in DMSO. For the RT-PCR study, SuperscriptII reverse transcriptase, RNase inhibitor, and random primers were obtained from GIBCO-BRL. Buffer II (10×) was obtained from Sigma/Aldrich. MgCl₂, dNTP, 10× volume PCR buffer, Hotstar Taq polymerase, and RNeasy mini kit were purchased from Qiagen. Ribosomal RNA (18S) and RNAseZap were purchased from Ambion.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The concentration of 2-APB used in all our experiments is 75 µM, which is above its IC₅₀ for the inhibition of IP₃R-channels (16, 25). When the rings of rabbit IVC were pretreated with 2-APB, PE (5 µM) failed to induce any measurable Ca²⁺ signal (Fig. 1A) compared with the PE-induced [Ca²⁺]_i oscillations observed before the pretreatment. In terms of force generation, as shown in Fig. 1B, PE (5 µM) typically elicits a significant sustained increase in tension with an average amplitude of 1.00 ± 0.01 (means ± SE) g (n = 16 rings from 4 rabbits) in the IVC. Such effect was, as expected, mediated through activation of the ₁-adrenergic receptor, because it can be prevented by the application of the ₁-adrenergic receptor antagonist phentolamine (10 µM, data not shown). Interestingly, PE failed to elicit any significant increase in tension (0.03 ± 0.02 g, n = 16 rings from 4 rabbits) following 2-APB pretreatment in the same vessels (Fig. 1B). These results indicate that 2-APB prevents the development of PE-induced venoconstriction by blocking the initiation of PE-induced [Ca²⁺]_i oscillations. We then proceeded to examine whether 2-APB can disrupt ongoing PE-induced venoconstriction and [Ca²⁺]_i oscillations as well. As shown in Fig. 1, C and D, introduction of 2-APB immediately halted ongoing PE-mediated [Ca²⁺]_i oscillations and fully relaxed the PE-mediated contraction (1.03 ± 0.05 g) to baseline (0.03 ± 0.04 g, n = 16 rings from 4 rabbits). These findings clearly show that 2-APB at 75 µM can effectively prevent or abolish the tonic contraction induced by PE, and such inhibition is mediated by preventing or abolishing PE-induced [Ca²⁺]_i oscillations. It therefore appears that the opening of IP₃R channel is required for PE-mediated venoconstriction. However, before such a conclusion is reached, we have to examine the selectivity of 2-APB (75 µM) in the rabbit IVC, especially with regard to important Ca²⁺ translocators such as the ryanodine-sensitive SR Ca²⁺ release channels (RyrR channels), the sarco(endo)plasmic-reticulum Ca²⁺ ATPase (SERCA), the SOC, and the L-type VGCC.

View larger version (34K): [in this window] [in a new window]   Fig. 1.   Effects of 75 µM 2-aminoethoxydiphenyl borate (2-APB) on the initiation and the maintenance of phenylephrine (PE)-induced vascular smooth muscle cell (VSMC) intracellular Ca2+ concentration ([Ca2+]i) oscillations and venoconstriction in the rabbit inteferior vena cava (IVC). A: left still-frame image shows the ribbon-shaped fluo-3-loaded VSMC that resides in the vessel wall of the rabbit IVC visualized with confocal microscopy. Two subcellular regions from two different VSMC in the field of view were chosen, and the changes in fluorescence units over time in these regions are depicted by the representative traces to the right. Changes in fluorescence reflects changes in [Ca2+]i. Applications of PE initiated [Ca2+]i oscillations in the cells from the control group (dotted line), whereas it resulted in no measurable Ca2+ signal in the cells from the 2-APB-pretreated group (solid line). Experimental control and 2-APB-pretreated traces are representative of the results in 70 cells from 4 different rings of rabbit IVC. B: application of PE-stimulated tension development in the control IVC (dotted line) but elicited no response in the 2-APB-pretreated IVC (solid line). C: application of 2-APB immediately abolished ongoing [Ca2+]i oscillations mediated by PE. Experimental trace is representative of the results in 70 cells from 4 different rings of rabbit IVC. D: application of 2-APB abolished ongoing tonic contraction mediated by PE.

As shown in Fig. 2A, pretreatment of IVC with 75 µM 2-APB did not significantly affect the peak amplitude of the caffeine-induced Ca²⁺ transient (105.9 ± 13.1% of the control, n = 11 rings from 4 rabbits, P = 0.66) and therefore appears to be inactive against ryanodine-sensitive SR Ca²⁺ release channels (RyrR channels). Furthermore, the fact that 2-APB pretreatment did not affect the amplitude nor the profile of the caffeine-induced Ca²⁺ transient implies that it had no significant effect on the plasma membrane Ca²⁺ extrusion system responsible for removing the excess cytoplasmic Ca²⁺ released from the SR. The ability of SERCA to replenish the SR Ca²⁺ store was assessed by examining the extent of refilling of the caffeine-sensitive store in the presence of 75 µM 2-APB in tissues that have been depleted of their SR Ca²⁺ stores (with 25 mM caffeine). Figure 2A shows that 2-APB only marginally affects the refilling of the caffeine-sensitive SR Ca²⁺ store because the presence of 2-APB reduced the peak amplitude of the third caffeine-induced Ca²⁺ transient slightly but nonsignificantly by 14.3 ± 9.7 g/100 ml (n = 11 rings from 4 rabbits, P = 0.12). Our finding indicates that 75 µM 2-APB may partially inhibit the SERCA, as has been shown by Missiaen et al. (16). Complete inhibition of SERCA by high concentration of CPA or thapsigargin will, as we have reported earlier, abolish the [Ca²⁺]_i oscillations (9). However, it will also lead to a sustained elevation in [Ca²⁺]_i, presumably due to the opening of SOC as a result of store depletion. Such elevation in [Ca²⁺]_i was not observed following IP₃R-channel blockade because the application of 75 µM 2-APB during PE stimulation promptly abolished ongoing [Ca²⁺]_i oscillations and returned the [Ca²⁺]_i to the baseline. Our result reveals that the SR was not depleted at this point as caffeine stimulated a large Ca²⁺ transient (Fig. 2B), even though Ca²⁺ release by PE was completely blocked. Therefore, such disruption of [Ca²⁺]_i oscillations must be the consequence of the potent inhibition of IP₃R-channel opening, rather than the weak inhibition of SERCA.

View larger version (26K): [in this window] [in a new window]   Fig. 2.   Effects of 75 µM 2-APB on the ryanodine-sensitive sarcoplasmic reticulum (SR) Ca2+ release (RyrR) channels, the sarco(endo)plasmic reticulum Ca2+-ATPase (SERCA), the L-type VGCC, and the store-operated channel (SOC) in the rabbit IVC. A: three pulses of caffeine are applied with a 5-min interval between each pulse. Maximum amplitude of the caffeine-induced Ca2+ transient from the first pulse reflects control SR Ca2+ level as Ca2+ from the SR is released through the opened RyrR channel. Right; bar graph comparing the average maximum amplitude of the second and third pulses to the first pulse (n = 11 rings from 4 rabbits). Error bars represent SE. After the addition of 2-APB, the second pulse of caffeine resulted in a single Ca2+ transient whose maximum amplitude is similar to the first pulse, indicating that 2-APB did not interfere with the opening of RyrR channels. The third pulse of caffeine resulted in Ca2+ transient whose maximum amplitude was slightly but not significantly diminished compared with the first pulse, indicating that the 2-APB may weakly inhibit the SERCA, which mediates refilling of the SR Ca2+ store. B: after complete inhibition of PE-mediated [Ca2+]i oscillations by 2-APB, application of caffeine resulted in a Ca2+ transient with a comparable maximum amplitude to that of the control condition before PE stimulation (dotted line), indicating that the SR Ca2+ store was replenished. Experimental traces shown are representative of the results obtained in 45 cells from 3 different rings of rabbit IVC. C: pretreatment of rabbit IVC with 2-APB (solid line) significantly reduced high K+ (80 mM extracellular K+)-mediated tonic contraction compared with the control (dotted line). High K+ contractions in both the control and the 2-APB-treated vessels were performed in the presence of 10 µM phentolamine, which is used to block of effects of neurotransmitters released by the nerve endings. Experimental trace shown is representative of findings in 16 rings from 4 rabbits. D: application of 5 µM PE followed by 10 µM CPA resulted in a maintained elevation in [Ca2+]i of VSMC (black solid line). Application of 75 µM 2-APB did not affect this plateau response, whereas the addition of 50 µM SKF-96365 abolished the maintained [Ca2+]i elevation and returned the [Ca2+]i to prestimulation baseline level (indicated by gray solid line). Representative trace shown is typical of the responses obtained in 30 cells from 3 rings of IVC.

To test for direct effects on the Ca²⁺ entry pathways, we examined the effects of 2-APB on the L-type VGCC and the SOC, two plasmalemmal channels important in PE-mediated [Ca²⁺]_i oscillations. The L-type VGCC provides a portion of the Ca²⁺ used to refill the SR Ca²⁺ store and to sustain the tonic contraction. Given its voltage dependence, we stimulated it with 80 mM [K⁺] PSS, which resulted in tonic contraction that can be completely abolished with 10 µM nifedipine (blocker of the L-type VGCC) (9). Figure 2C shows that 75 µM 2-APB pretreatment inhibited high K-induced tonic contraction by 12.9 ± 5.0% (n = 16 rings from 4 rabbits, P = 0.039). However, this slight inhibition of L-type VGCC cannot account for the complete inhibition of the force generation by 2-APB, because we know that only 27% of the IVC constriction induced by PE is contributed by Ca²⁺ influx through the L-type VGCC (9). Most notably, this result indicates that 75 µM 2-APB exerts only marginal direct inhibition on the L-type VGCC. Therefore, during ₁-adrenergic stimulation, 2-APB must be inhibiting Ca²⁺ influx through the L-type VGCC indirectly by inhibiting upstream activation mechanism(s).

In addition to the L-type VGCC, a putative NSCC is important for sustaining the [Ca²⁺]_i oscillations and mediates nearly 73% of tonic contraction induced by PE. Its sensitivity to 2-APB observed here implies that it is most probably a SOC-type channel. To test for the presence of SOC in the rabbit IVC, we used 5 µM PE to discharge Ca²⁺ from the SR and then applied 10 µM CPA to inhibit SERCA. As shown by the representative trace (n = 30 cells from 3 rings of IVC) in Fig. 2D, this resulted in a maintained elevation of [Ca²⁺]_i above baseline [Ca²⁺]_i. This time the elevated [Ca²⁺]_i could not be returned to baseline by 2-APB, because closing of the IP₃R did not lead to refilling because SERCA was blocked. However, the [Ca²⁺]_i returned to baseline upon subsequent addition of 50 µM of the ROC-SOC blocker SKF-96365. This suggests that this maintained elevation of [Ca²⁺]_i is due to increased Ca²⁺ influx through the SOC, which is opened as a consequence of depletion of the SR Ca²⁺ store with PE and CPA. Theoretically, 2-APB can prevent activation of SOC either by preventing SR Ca²⁺ depletion or by blocking the SOC channel directly. However, the representative trace depicted in Fig. 2D shows that 2-APB did not affect the maintained elevation in [Ca²⁺]_i following SR store depletion with PE and CPA, even though SKF-96365 completely abolished it. This finding indicates that in the VSMC of the rabbit IVC 2-APB does not inhibit the SOC directly.

From these new findings, we can rule out direct inhibition on the RyrR channels, the SERCA, the L-type VGCC, and the SOC as the primary mechanism of inhibition by 2-APB of PE-mediated [Ca²⁺]_i oscillations and tonic contraction. The prevention of the generation of any Ca²⁺ signal or force with 2-APB pretreatment indicates that IP₃R channel-mediated SR Ca²⁺ release is crucial. In addition to inhibiting Ca²⁺ release, 2-APB also prevented stimulated Ca²⁺ entry through both the NSCC component and the L-type VGCC component, because no force can be generated or maintained in the presence of 2-APB. This indicates that the putative NSCC is most probably a SOC-type channel, which, as we demonstrated earlier, does appear to exist in the rabbit IVC. This speculated involvement of SOC in PE-mediated [Ca²⁺]_i oscillations and constriction of the rabbit IVC is also consistent with the findings that the nifedipine-resistant, SKF-96365-sensitive component of [Ca²⁺]_i oscillations, and tonic contraction mediated by the putative NSCC is also sensitive to Ni²⁺ and La³⁺, agents commonly used to block the SOC (11, 14, 20). As shown in Fig. 3A, application of 2 mM NiCl₂ or 300 µM LaCl₃ in IVC pretreated with nifedipine (10 µM) completely abolished the [Ca²⁺]_i oscillations stimulated with 5 µM PE. In vessels pretreated with nifedipine, application of 2 mM NiCl₂ reduced PE-mediated tonic contraction (0.71 ± 0.03 g) to a baseline level of 0.02 ± 0.01 g (n = 9 rings from 4 rabbits). Similarly, application of 300 µM LaCl₃ decreased PE-mediated tonic contraction (0.70 ± 0.04 g) to 0.01 ± 0.01 g (n = 8 rings from 4 rabbits) (Fig. 3B). It should be noted that even though La³⁺ and Ni²⁺ are commonly used to block the SOC, they are not selective for the SOC. In this context, these results do help to characterize this putative NSCC as La³⁺ and Ni²⁺ sensitive. These characteristics, together with the sensitivity of the PE response to 2-APB, indicate that this putative NSCC is a SOC-type channel.

View larger version (29K): [in this window] [in a new window]   Fig. 3.   Effects of Ni2+ and La3+ on the nifedipine (Nif)-resistant component of PE-mediated [Ca2+]i oscillations and tonic contraction of the rabbit IVC. A: representative trace on left shows that applications of 2 mM NiCl2 to the bath abolished the Nif (10 µM)-resistant component of the [Ca2+]i oscillations stimulated with PE. Similarly, in a different cell from a different set of experiments, the representative trace on the right shows that application of 300 µM LaCl3 abolished Nif-resistant component of PE-mediated [Ca2+]i oscillations as well. These experimental traces are representative of the results in 60 cells from 4 different rings of rabbit IVC. B: representative contraction traces shows that the application of either 2 mM NiCl2 (n = 9 rings from 4 rabbits) or 300 µM LaCl3 (n = 8 rings from 4 rabbits) to the bath completely inhibited the Nif-resistant component of the tonic contraction elicited by PE.

Given that the functional evidence points to the nonselective cationic SOC as a crucial component for PE-mediated [Ca²⁺]_i oscillations and tonic contraction in the rabbit IVC, we then proceeded to determine whether SOC mRNA is expressed in the smooth muscle of the rabbit IVC by RT-PCR study. The most well-characterized genes that encode the SOC belong to the family of Trp genes with a large subfamily of Trp1~7 (6, 14). We therefore examined the mRNA expression of Trp1~7 genes in the smooth muscle of the rabbit IVC. It should be noted here that due to the fact that the rabbit equivalent of Trp1~7 have not been sequenced, we used primers based on mouse or rat sequences. However, because there is high interspecies sequence homology when comparing mouse and rat sequences for the same type of Trp channels, it is highly plausible that the primers that we use can identify rabbit Trp channels as well. As depicted in Fig. 4A, only a single band of the predicted size (372 bp) of Trp1 was detected in the smooth muscle of the rabbit IVC (n = 5 rabbits). In parallel, both rabbit and rat brains were used as positive controls for Trp1~7 expression. With the same primers, only Trp1, 3, and 4 mRNA were detected in the rabbit brain (n = 3 rabbits), whereas all Trp1~7 mRNA were detected in the rat brain (n = 3 rats). Furthermore, because Ca²⁺ influx via the L-type VGCC has been shown to be important for PE-mediated [Ca²⁺]_i oscillations in the rabbit IVC, we also tested the expression of _1C-subunit (the pore-forming unit) of the L-type VGCC in the rabbit IVC (5). RT-PCR analysis of the IVC (n = 5 rabbits) showed mRNA expression for the _1C-subunit of the L-type VGCC. Sequencing of the Trp1~7 and _1C-subunit amplification products revealed 100% homology with the respective sequences obtained from GenBank.

View larger version (68K): [in this window] [in a new window]   Fig. 4.   Transient receptor potential (Trp)1 and 1C-mRNA expression in the smooth muscle of the rabbit IVC. Exemplary agarose gel electrophoresis from RT-PCR analysis of Trp channel family members and 1C-mRNAs in the rabbit IVC smooth muscle (A), rabbit brain (B), and rat brain (C). PCR products were generated through the use of specific primers for Trp1-Trp7 and 1C-subunit of the L-type voltage-gated Ca2+ channel (VGCC). Only genes for Trp1 (372 bp) and 1C (371 bp) were found to be expressed in the smooth muscle of the rabbit IVC. In the rabbit brain-positive control, only mRNA for Trp1 (372 bp), Trp3 (331 bp), and Trp4 (265 bp) were detected. In contrast, Trp1~7 and 1C-mRNA expression were detected in the rat brain as positive controls. Expression of 18S ribosomal RNA was used as an internal control. RT-PCR reactions run in the absence of reverse transcriptase (RT) or cDNA (cDNA) were used as negative controls. Gels shown are representative of findings in a minimum of 3 animals.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The above experimental observations have provided us with the following insights into PE-mediated constriction of the rabbit IVC. First, as suggested before (9), the generation of individual Ca²⁺ waves requires the opening of IP₃R channels. This is supported by the finding that the generation of PE-mediated venoconstriction can be prevented with IP₃R-channel blockade by 2-APB. These observations indicate that during ₁-adrenergic stimulation, Ca²⁺ is not delivered via plasmalemmal channels directly to activate the myofilaments in the rabbit IVC. Instead, the SR network in the ₁-adrenergic-stimulated IVC delivers Ca²⁺ directly to the myofilaments. One may speculate that this may represent a more efficient and effective way of activating the myofilaments, because the SR network penetrates deep into the myoplasm. The IP₃R channels on the SR play a crucial role as Ca²⁺ is released by the IP₃R channels to activate the myosin light chain kinase units tethered to the myofilaments (10). The opening of IP₃R channels is thus required for PE-mediated [Ca²⁺]_i oscillations and constriction of the rabbit IVC. However, this does not rule out that Ca²⁺ release via the RyrR channel may serve to amplify the Ca²⁺ signal (initiated by IP₃R channel-mediated Ca²⁺ release) through a Ca²⁺-induced Ca²⁺ release mechanism.

Second, as reported earlier by van Breemen's lab (9), stimulated Ca²⁺ entry dependent on a putative nifedipine-resistant, SKF-96365-sensitive, NSCC is crucial for sustaining PE-induced [Ca²⁺]_i oscillations and is responsible for nearly 73% of the force development. The complete inhibition of Ca²⁺ signal and force generation by 2-APB indicates that 2-APB, in addition to preventing Ca²⁺ release via the IP₃R channels, also prevents the activation of this putative NSCC. However, when SR depletion is maintained by SERCA blockade with CPA, 2-APB fails to block the NSCC, which subsequently could be blocked by SKF-96365, Ni²⁺, or La³⁺. These findings strongly suggest that the putative NSCC is a SOC-type channel. Even though the activation of this nonselective cationic SOC is dependent on IP₃R-mediated SR Ca²⁺ release, its precise mechanism of activation remains undefined at this time. The observation made in Fig. 2D, however, does exclude the conformational coupling model (12) as the mechanism of activation, because the SOC in store-depleted VSMCs remains activated in the presence of 2-APB. This channel thus may be similar to the calcium-influx, factor-activated, SOC-like nonselective cationic channel that has been described in VSMC (22, 23). Alternatively, it may be a Ca²⁺-release activated nonselective cationic channel that can be activated by Ca²⁺ released via the IP₃R channels or by the build up of Ca²⁺ in the plasma membrane-SR junctional space following SERCA blockade with CPA. Moreover, if a SOC-type channel is involved here, the intermittent Ca²⁺ release that produces the [Ca²⁺]_i oscillations should activate this SOC-type NSCC intermittently. On this note, it is interesting to point out that the oscillatory inward nonselective cationic current has been described in endothelin-stimulated rat aorta, a large vessel that exhibits asynchronous wavelike [Ca²⁺]_i oscillations as well. In addition, our finding of Trp1 mRNA expression in vascular smooth muscle of the rabbit IVC provides supporting evidence for the existence of a store-operated NSCC in this tissue (6). This observation of Trp1 mRNA expression in rabbit VSMC from the IVC is consistent with a recent report by Xu and Beech (26) that Trp1 protein is ubiquitously expressed in various human, rabbit, and mouse vessels. Trp1 is known to encode a component of the SOC (6, 11, 21, 27), and Trp1 protein is the pore-forming component, which has been localized to the plasma membrane of rabbit VSMC (26). Intriguingly, the SOC formed by the product of the Trp1 gene has been shown to be a NSCC (21, 27). This is consistent with our earlier investigations, which suggested that Na⁺ influx through this NSCC can raise the local [Na⁺] in the restricted subplasmalemmal space, which then drives the Na⁺/Ca²⁺ exchanger into its reverse mode of operation, bringing Ca²⁺ into the cell to refill the SR (9). It is important to note that even though only one particular Trp1 mRNA was detected here, more Trp-type subunits may also be expressed. Given that primers used to detect Trp1~7 mRNAs were not of rabbit origin and that only Trp1, 3, and 4 mRNAs were positively identified in the rabbit brain, one cannot dismiss the possibility that rabbit Trp2, and 5-7 mRNA with distinct rabbit sequences were not detected in this study.

Third, our findings with 2-APB indicate that the activation of the nifedipine-sensitive, L-type VGCC component of Ca²⁺ entry, which is responsible for 27% of PE-mediated tonic contraction, is dependent on IP₃R-channel opening as well. It is plausible that the L-type VGCC may be activated by the depolarizing inward cationic (Na⁺ and Ca²⁺) current through the store-operated NSCC, which is activated by IP₃R channel-mediated SR Ca²⁺ release. In accordance with the functional data, the mRNA study showed that the _1C-subunit of the L-type VGCC is expressed in the smooth muscle of rabbit IVC.

The findings presented in this report have improved our understanding of the mechanism of asynchronous wavelike [Ca²⁺]_i oscillations, which is emerging as a ubiquitous and physiologically relevant form of Ca²⁺ signaling in VSMC from different vasculatures (1, 8, 15, 18, 19). Figure 5 depicts our current working model based on the evidence presented in this report and previous published results. It describes the sequence of events occurring during one cycle of SR store emptying and SR store refilling that, when repeated, gives rise to the observed [Ca²⁺]_i oscillations. Briefly, upon ₁-adrenergic receptor stimulation, one of the earliest events is the opening of IP₃R channels (Fig. 5A). The SR empties its Ca²⁺ through the IP₃R channels, and this gives rise to a Ca²⁺ wave (a single Ca²⁺ spike in a series of [Ca²⁺]_i oscillations). The Ca²⁺ released by the IP₃R channels not only elevates [Ca²⁺] in the myoplasm to activate the myofilaments, it also raises [Ca²⁺] near the IP₃R channels to activate neighboring IP₃R channels (2, 7). In the meantime, the release or emptying of the SR through IP₃R-channels mediated Ca²⁺ release leads to opening of the putative store-operated NSCC (Fig. 5B). Some Ca²⁺ and a large amount of Na⁺ then enter the plasma membrane-SR junctional space. This inward cationic current causes depolarization of the membrane potential, which activates the L-type VGCC. Meanwhile, [Na⁺] in the restricted subplasmalemmal space elevates. This drives the Na⁺/Ca²⁺ exchanger into its reverse mode of operation, bringing Ca²⁺ into the cell. With the IP₃R channels now inactivated by the greatly elevated [Ca²⁺] at the pore's outer surface and Ca²⁺ being supplied to SERCA by the Na⁺/Ca²⁺ exchanger, L-type VGCC and SOC, the SR starts to refill (Fig. 5C). Once the IP₃R channels are closed and the SR Ca²⁺ store is refilled, the signal to activate the SOC terminates and Ca²⁺ exchange or Ca²⁺ influx from the extracellular space ceases (Fig. 5D). As the SR is filled further, both the elevated SR luminal Ca²⁺ level and [Ca²⁺] at the cytoplasmic side of the IP₃R, fed by Ca²⁺ leakage from the SR reach the threshold for activation of IP₃R channels and regenerative SR Ca²⁺ release (17).

View larger version (22K): [in this window] [in a new window]   Fig. 5.   Hypothetical sequence of events during PE-mediated smooth muscle [Ca2+]i oscillations in rabbit IVC. Hypothetical sequence of events underlying smooth muscle [Ca2+]i oscillations is shown here in the order from A to D as described in details in DISCUSSION. NCX, Na+/Ca2+ exchanger; L-VGCC: L-type voltage-gated Ca2+ channel that contains the 1C-sub-unit; IP3R: inosital 1,4,5-trisphosphate-sensitive SR Ca2+ release channel; Em: membrane potential.

In summary, the work presented herein establishes that Ca²⁺ release through the IP₃R channels is required for ₁-adrenergic receptor-mediated tonic contraction in venous smooth muscle, whereas Ca²⁺ entry mediated by the putative store-operated NSCC activated by IP₃R channel-mediated Ca²⁺ release or store emptying is required to maintain such tonic contraction. In addition, our data demonstrate that the activation of the _1C-subunit containing L-type VGCC occurs secondarily to the opening of IP₃R channels and store-operated NSCC. Moreover, our results also indicate that the gene of a potential molecular candidate for the putative store-operated NSCC-Trp1 is actively transcribed in this venous smooth muscle. This appreciation for the central roles that IP₃R channels and SOCs play in excitation-contraction coupling of large capacitance vessels may be valuable in the management of patients when the reduction in venous return and ventricular preload are a priority.