|Home | About | Journals | Submit | Contact Us | Français|
The role of SKCa and IKCa channels in myogenic tone and endothelium-derived hyperpolarizing factor (EDHF) responsiveness was investigated under control conditions and after ischemia and reperfusion (I/R) in parenchymal arterioles (PA) vs. middle cerebral arteries (MCA).
MCA and PA were dissected from male Wistar rats that were ischemic for 1 hour with 24 hours of reperfusion (n=12) or sham controls (n=12). Basal tone and reactivity to apamin (300 nM), TRAM-34 (1.0 µM) and nitro-L-arginine (L-NNA, 0.1 mM) were compared in PA and MCA pressurized to 40 mmHg and 75 mmHg, respectively. SKCa and IKCa channel mRNA expression was measured using real-time PCR.
PA developed greater basal tone than MCA (42±4% vs. 19±3% p<0.01). Addition of apamin and TRAM-34 increased tone of PA by 25±3% and 16±2%, respectively, whereas MCA had no response to either inhibitor. After I/R, the response to NOS inhibition was diminished in PA, whereas EDHF responsiveness was preserved. In addition, stimulated EDHF dilation was partially reversed by apamin and completely reversed by TRAM-34 in both control and ischemic PA. SKCa and IKCa channel mRNA expression was similar in PA and MCA and not altered by I/R. However, IKCa channel mRNA expression was 4–5 fold greater than SKCa channels.
It appears that SKCa and IKCa channel activity diminish basal tone of PA, but not MCA. The preservation of EDHF responsiveness of PA after I/R suggests an important role for this vasodilator under conditions when NOS is inhibited.
The vasoactive response of cerebral arteries and arterioles to changes in pressure, termed myogenic reactivity, is likely an important contributor to autoregulation of cerebral blood flow (CBF).1 In addition, myogenic tone is a major determinant of cerebrovascular resistance (CVR), providing a crucial protective mechanism by limiting transmission of damaging hydrostatic pressure to the microcirculation.2 In the middle cerebral artery (MCA) territory, both cerebral (pial) arteries and penetrating parenchymal arterioles (PA) contribute significantly to CVR.3 However, PA are unique in that they are long and largely unbranched vessels that connect the surface vessels to the microcirculation, where they appear to function as a bottleneck to blood flow.4 These vessels possess greater basal tone than the MCA and seem to be more resistant to ischemia and reperfusion (I/R) in that they do not lose myogenic tone and reactivity to pressure in the same manner as MCA.5–7
The cerebral endothelium controls vascular tone through production of vasoactive factors, most notably nitric oxide, prostacylcin (PGI2) and endothelium-derived hyperpolarizing factor (EDHF).8 The relative contribution of these vasodilators to vascular tone may depend on the size of the vessel.9,10 In particular, EDHF’s contribution to regulation of cerebral blood flow and vascular resistance may be greatest at the level of small PAs.10 In addition, whereas PGI2 has little influence in cerebral vessels, both cerebral arteries and arterioles have considerable basal NO production that mitigates tone under resting conditions.5,8 This is demonstrated by the fact that NOS inhibition produces constriction and increases tone in both MCA and PA.5,8,11–13
Activation of ion channels has an important role in regulating endothelial membrane potential and calcium homeostasis, and hence vasodilator production, in response to stimulation.14 In particular, small- and intermediate calcium-activated potassium channels (SKCa and IKCa) produce endothelial hyperpolarization when stimulated.14–16 In the absence of voltage-operated calcium channels, hyperpolarization of endothelial cells may increase the electrochemical driving force for calcium entry, and appears to be an important initial step in endothelial-mediated smooth muscle hyperpolarization necessary for EDHF dilation.17,18 SKCa and IKCa channels are expressed on endothelium, not vascular smooth muscle,14,19 and their inhibition by specific antagonists blocks EDHF-mediated relaxation and hyperpolarization.20–22 In the rat MCA, IKCa channel inhibition alone is sufficient to block EDHF dilation.20,21 However, when NOS is active, SKCa channels also contribute to endothelial hyperpolarization.22 Interestingly, the level of basal tone in the MCA is unaffected by apamin and TRAM-34, specific inhibitors of SKCa and IKCa channels, respectively.21,22 This result suggests that unlike NO, EDHF is not affecting basal tone under normal conditions. However, the functional importance of EDHF in PA, a vessel that has been shown to possess a greater EDHF component than MCA, is not known.10 The functional role for EDHF under basal conditions may represent an important yet unrecognized determinant of CVR.
In the present study, we investigated whether SKCa and IKCa inhibition affects basal tone in PA and the role of these channels in EDHF-mediated dilation. In addition, we investigated the effect of I/R on these channels and the role of EDHF in PA as this is a condition in which NO-mediated responses are largely abolished and it has been hypothesized that EDHF may act as a back-up vasodilator when NO production is compromised.5,9
All procedures were approved by the Institutional Animal Care and Use Committee and complies with the NIH guidelines for the use and care of animals. Temporary filament occlusion of the MCA was used to induce I/R in male Wistar rats (N=35;280–300g), as previously described.5,7 Ischemic animals were exposed to 1 hour of ischemia and 24 hours of reperfusion by suture removal. Sham control animals underwent anesthesia and a midline incision, but were not exposed to I/R. All animals received buprenorphine (0.025 mg/kg s.c.) for analgesia.
Animals were anesthetized with isoflurane (2% in oxygen), decapitated and the brain quickly removed and placed in cold, oxygenated physiologic saline solution (PSS). PA were identified as branches off the MCA that penetrate at right angles into the brain parenchyma. Once identified, surrounding brain tissue was carefully cleared and the vessel removed, placed in the arteriograph chamber (Living Systems Instrumentation, Burlington, Vermont) and mounted on glass cannulas. In some experiments, we compared the response of PA to that of MCA from the same animal. In these cases, both MCA and PA were carefully dissected and placed in arteriograph chambers. Isolated vessel experiments were conducted as previously described.5,7,11
Expression of SKCa and IKCa channel mRNA was determined using standard techniques for real time quantitative PCR (qPCR) and was performed by the DNA Facility at the University of Vermont. Briefly, MCA and PA from the ipsilateral and contralateral sides of the brain were dissected separately from ischemic (n=4) and sham (n=4) animals and rapidly frozen in RNAlater. The vessels were then removed and placed in a RNase free microcentrifuge tube, homogenized in the presence of RLT Buffer, and total RNA isolated using an RNeasy Micro Kit (Qiagen, Valencia, CA) animal tissue protocol. To increase yield, samples were not DNAase treated. The concentration of total RNA was obtained from a NanoDrop ND-1000 Spectrophotometer (Wilmington, DE) and its integrity analyzed by an Agilent 2100 Bioanalyzer (Santa Clara, CA). cDNA was made from total RNA using SuperScript III Kit (Invitrogen, Carlsbad, CA). The cDNA reaction was according to standard protocol, using random hexamers and 4.8 ng of total RNA. Real-time PCR was set up as follows: 10 µl Universal PCR Master Mix (Applied Biosystems, Foster City, CA), 1 µl Assay on Demand (Applied Biosystems, Foster City, CA), 8 µl water, 1µl cDNA. Target genes SK3 and SK4 were used to assess mRNA expression of SKCa2.3 and SKCa3.1 channels, which are known to be expressed in endothelium and contribute to the EDHF response.14,15,17,18 All target genes (SK3 for SKCa, SK4 for IKCa and B2m for an endogenous control) were assessed using Assays on Demand from Applied Biosystems that were validated for efficiency and did not amplify genomic DNA. All samples were run in technical duplicates using a 7900HT Sequence Detection System (Applied Biosystems, Foster City, CA). The PCR was cycled for 2 minutes at 50 C, 10 minutes at 95 C, then 40 cycles at 95 C for 15 s and 60 C for 1 minute.
Only Ct duplicate values of <35 were used (if replicates were >1.0 apart, they were rerun). Ct values were averaged and used for comparison using the ΔΔCt method.23 Three comparisons of mRNA expression were made between MCA and PA, and between ischemic and sham animals: 1) SKCa and IKCa channel expression of PA relative to MCA in sham and ischemic animals to determine if channel expression was different between these two segments of the vasculature; 2) SKCa and IKCa channel expression in PA after I/R relative to sham to determine if I/R affected channel expression; and, 3) IKCa channel expression relative to SKCa channel expression in PA to determine if one channel was expressed to a greater extent than the other.
MCA and PA were dissected from animals that were either sham control (n=6) or after I/R (n=6), mounted in the arteriograph chamber and equilibrated for 1 hour at either 40 mm Hg for PA or 75 mm Hg for MCA. These pressures were used because they approximate what these vessel experience in vivo.3 After the equilibration period in which spontaneous tone developed, apamin (300 nM), TRAM-43 (1.0 µM) and nitro-L-arginine (L-NNA, 0.1 mM) were sequentially added to the bath and lumen diameters and wall thicknesses measured. These concentrations of apamin and TRAM-34 have been shown to be selective for SKCa and IKCa channels, respectively.18 Because apamin and TRAM-34 can cause depolarization of perivascular sensory nerves that are present in MCA, but not PA11, capsaicin (1.0 µM) was added to the bath prior to addition of the inhibitors for the MCA only. At the conclusion of the experiment, papaverine (0.1 mM) in zero calcium PSS was added to obtain fully relaxed diameters.
Because the role of SKCa and IKCa in EDHF production has been previously studied in MCA,20–22 we investigated the role of SKCa and IKCa channels in EDHF responsiveness of PA only. PA were dissected from sham control (n=6) or after I/R (n=6) and mounted in the arteriograph chamber. Arterioles were equilibrated for 1 hour at 40 mm Hg during which time spontaneous tone developed. L-NNA (0.1 mM) and indomethacin (10 µM) were added to the bath to inhibit NOS and cyclooxygenase (COX), respectively. These concentrations have been shown to maximally inhibit NOS/COX in these vessels in the rat.10 In the presence of the inhibitors, A23817, a calcium ionophore and endothelium-dependent agonist, was cumulatively added to the bath (0.01 to 1.0 µM) and diameters recorded at each concentration once stable. This is a common approach to investigating EDHF.5,8,10,13,20 Next, apamin (300 nM) and TRAM-34 (1.0 µM) were sequentially added to the bath and diameter recorded. In some experiments (n=3), TRAM-34 was added prior to apamin to determine if inhibition of IKCa channels alone could reverse the EDHF dilation. Finally, papaverine (0.1 mM) in zero calcium PSS was added to obtain fully relaxed diameters.
All experiments were conducted using Krebs PSS (a bicarbonate-based buffer aerated with 5% CO2, 10% O2 and 85% N2 to maintain pH), the ionic composition was (mmol/L): NaCl 119.0, NaCHO3 24.0, KCl 4.7, KH2PO4 1.18, MgSO4A7H2O 1.17, CaCl2 1.6, EDTA 0.026, and glucose 5.5. PSS was made each week and stored without glucose at 4 C; glucose was added to the PSS prior to each experiment. Zero calcium PSS was made without adding calcium. Apamin and TRAM-34 were purchased from Tocris Biosciences (Ellisville, MO) and kept frozen until used. Capsaicin, L-NNA, indomethacin, calcium ionophore A23187 and papaverine were purchased from Sigma (St. Louis, MO).
Vascular tone was calculated as a percent decrease in diameter from the fully relaxed diameter in papaverine by the equation: (1-(tone/papav))*100%; where tone = diameter of vessel with tone and papav=diameter in papaverine. Percent constriction was calculated as a percent change in diameter from baseline by the equation: (1-(drug/baseline))*100%; where drug = diameter of vessel in either apamin, TRAM-34 or L-NNA and baseline = diameter prior to giving the drug.
RQ values were determined by first calculating the ΔCt (difference between target gene Ct and endogenous control gene Ct) and the ΔΔCt (difference between ΔCt’s to be compared). RQ value was then calculated as 2−ΔΔCt.
All data are presented as mean ± SEM. One-way analysis of variance was used to determine differences in tone and percent constriction, with a posthoc Student-Newman-Keuls test for multiple comparisons, where appropriate.
In the first experiments, we investigated the role of SKCa and IKCa in modulating basal tone in PA and MCA from the same animals. The percent tone of MCA and PA from sham control and ischemic animals is shown in Figure 1. Under control conditions, PA developed considerably greater tone compared to MCA (42±4% vs. 19±3%; p<0.01). Cumulative addition of apamin and TRAM-34 constricted the PA, suggesting that EDHF was affecting basal tone in these vessels (Figure 1A). Unlike the PA, and similar to previous studies20–22, apamin and TRAM-34 had no effect on MCA diameter (Figure 1B). However, NOS inhibition with L-NNA caused constriction in both PA and MCA (Figure 1).
Similar to previous studies5–7, I/R did not significantly affect myogenic tone in PA (42±4% vs. 33±2%; p>0.05), but caused a significant loss of tone in MCA (19±3% vs. 8±2%; p<0.05). Addition of apamin and TRAM-34 to ischemic PA caused constriction, but had no effect on MCA after I/R. Importantly, PA had little to no response to NOS inhibition with L-NNA after I/R (Figure 1).
Figure 2 shows the percent constriction of PA and MCA from sham controls and ischemic animals in response to cumulative addition of apamin, TRAM-34 and L-NNA. Comparing percent constriction allows for comparison of the reactivity of the vessels to each inhibitor. Figure 2A shows that in PA from control animals, apamin caused significantly greater constriction than TRAM-34, an effect that was reversed with I/R. Importantly, constriction to L-NNA was robust in control arterioles, but almost completely abolished after I/R. Figure 2B shows that percent constriction to apamin and TRAM-34 was minimal in MCA from both control and ischemic animals. Unlike SKCa and IKCa channel inhibition, NOS inhibition caused considerable constriction in MCA that was not diminished after I/R.
Although the role of SKCa and IKCa channels in EDHF production has been established in MCA, the role of these channels in EDHF-mediated responses in PA is unknown. Figure 3 shows the response to A23187 in the presence of NOS/COX inhibition from sham control and ischemic animals. Notice that the diameter of ischemic PA is greater than control vessels due to their lack of responsiveness to L-NNA. Addition of A23187 caused considerable dilation of both types of arterioles with NOS/COX inhibition that was maximal at 1.0 µM and similar to fully relaxed diameters, suggesting a considerable role for EDHF in these arterioles. There was little difference in the EDHF component of the response with I/R except that ischemic arterioles started with less tone, but still dilated fully.
It has been shown that EDHF-mediated dilation can be reversed by IKCa channel inhibition alone with little to no contribution of SKCa channels in MCA.20–22 However, it is not known what role these channels play in EDHF-mediated responses in PA. Figure 4A shows a diameter tracing of a PA in the presence of NOS/COX inhibition that dilated fully to A23187. Addition of apamin caused considerable vasomotion and partially reversed the EDHF-based dilation. However, addition of TRAM-34 completely reversed this dilation and restored diameter to starting levels. When TRAM-34 was given prior to apamin (n=3), it alone fully reversed the EDHF dilation (Figure 4B). These data suggest that similar to the MCA, IKCa channel inhibition alone is sufficient to reverse EDHF dilation in PA.
Figure 4C shows the percent tone in PA from control and ischemic animals under these conditions. Similar to the first set of experiments, PA developed less tone after I/R, but was not significantly different than controls. PA from control animals responded to L-NNA with increased tone, but had little reactivity to NOS inhibition after I/R. However, arterioles from both groups dilated fully to 1.0 µM A23187, again suggesting a considerable role for EDHF in these vessels. Interestingly, apamin caused constriction in both types of vessels, but only partially reversed the EDHF-based response, whereas TRAM-34 completely reversed the dilation. It is worth noting that while the percent tone in TRAM-34 was less in PA after I/R compared to sham controls, these vessels started out less constricted because of their lack of response to L-NNA. However, TRAM-34 reversed the dilation to the starting diameter of those vessels as well.
In order to determine if the differential response to apamin and TRAM-34 between MCA and PA was due to differences in mRNA expression of SKCa and IKCa, we used qPCR to compare channel mRNA expression. Figure 5A shows SKCa and IKCa expression of PA and MCA from sham and ischemic animals. Notice that despite the lack of response of MCA to apamin and TRAM-34, both vessels expressed SKCa and IKCa. However, there was no significant difference in expression levels between PA and MCA, suggesting that differences in expression cannot account for the observed functional changes.
To determine if the effect of I/R on the response to apamin and TRAM-34 in PA related to channel mRNA expression, we compared the mRNA expression of SKCa and IKCa after I/R to sham controls (Figure 5B). There was no difference in channel expression with I/R, again suggesting that the differential response to apamin and TRAM-34 in PA was not related to differences in mRNA expression.
Lastly, because of the different roles of SKCa and IKCa channels in mediating EDHF in MCA and PA, we compared the mRNA expression of IKCa relative to SKCa. Figure 5C shows that there was a 4–5 fold increase in IKCa mRNA expression compared to SKCa in PA that was not altered by I/R.
In the present study, we show for the first time that SKCa and IKCa channels are involved in modulating basal tone in PA, but not MCA. This is demonstrated by the finding that apamin and TRAM-34, selective SKCa and IKCa inhibitors, both caused a significant increase in tone in PA, whereas the MCA was unaffected. The lack of response in MCA to these inhibitors confirms previous findings.20–22 The increase in basal tone in PA with channel inhibition importantly suggests that similar to NO, EDHF is affecting basal tone in the PA. A second major finding was that while I/R significantly diminished NO responsiveness, the reactivity to apamin and TRAM-34 was preserved. The decreased constriction to L-NNA in PA confirms our previous finding5 and suggests that basal NO production is decreased after I/R. This is not likely due to diminished responsiveness of vascular smooth muscle to NO (or the guanylate cyclase - cyclic GMP pathway) because sensitivity to the NO donor sodium nitroprusside was increased after I/R.5 It has been hypothesized that EDHF acts as a backup vasodilator under conditions when NO production is compromised.9,24 The finding that NO responsiveness was significantly diminished after I/R, but the role for EDHF was preserved seems consistent with this hypothesis. However, it appears that the EDHF-mediated response may be resistant to I/R.
Similar to previous studies, I/R caused a significant loss of myogenic tone in MCA, whereas tone in PA was more preserved.5–7 Clearly, diminished tone in MCA after I/R was not due to SKCa and IKCa channels because inhibition of these channels did not restore tone. This result was somewhat expected, however, because even under nonischemic conditions, MCA did not respond to apamin and TRAM-34. The explanation for the lack of responsiveness of the MCA to apamin and TRAM-34 under basal conditions is not clear, but does not appear to be related to lack of channel mRNA expression since qPCR in isolated MCA revealed that both channels were expressed at similar levels as PA. It is therefore possible that another component of the EDHF response is differentially affected in MCA and PA. The basic mechanism of EDHF-mediated response is that increased intracellular calcium in endothelium activates SKCa and IKCa channels causing K+efflux and membrane hyperpolarization.15 Hyperpolarization of endothelium is then transferred to vascular smooth muscle via synthesis or generation of signals capable of diffusing through membranes or myoendothelial gap junctions.25 In vascular smooth muscle, EDHF activates K+-channels, causing hyperpolarization and relaxation.14,21 There are several steps in this pathway that may be different between MCA and PA. First, basal endothelial cell calcium may be higher in PA compared to MCA such that SKCa and IKCa channels are more active basally in PA. Second, it may be that the generation of the signal responsible for transferring hyperpolarization from endothelium to vascular smooth muscle is different under basal conditions and/or in response to agonists. Third, there may be anatomic differences between MCA and PA such that there is a difference in how the signals are transferred from endothelium to vascular smooth muscle. Due to its smaller size and fewer smooth muscle cell layers, the endothelium of PA may have a greater influence on smooth muscle. It is therefore possible that this structural arrangement allows for transfer of hyperpolarization that is not possible in MCA under basal conditions.
The finding that both apamin and TRAM-34 caused constriction in PA suggests that both SKCa and IKCa channels are involved in mediating this response. Under nonischemic conditions, contraction to apamin was greater than that of TRAM-34, a result that was reversed after I/R. It is unclear what caused the differential response, or its significance, but does not appear to be related to differences in channel mRNA expression based on our PCR results. Because PCR was done on isolated PA that contained both endothelium and vascular smooth muscle, we cannot rule out the possibility that these channels are expressed on vascular smooth muscle. In a study by McNeish et al.,22 immunohistochemical staining of MCA found positive staining of IKCa in vascular smooth muscle, although a functional role for these channels in smooth muscle has yet to be demonstrated. In regards to the present study, we cannot rule out the possibility that these channels are expressed in vascular smooth muscle of PA and/or are altered by I/R that might contribute to the differential response.
Agonist stimulation of endothelial cell calcium in the presence of NOS/COX inhibition is commonly used to investigate EDHF responses.5,8,10,13,20 Previous studies in mesenteric vessels found that both SKCa and IKCa inhibition are necessary to block endothelial hyperpolarization and hence EDHF, making activation of these channels critical to agonist-induced EDHF production.9,15–17 Cerebral arteries are thought to be different than peripheral vessels in that IKCa channel inhibition alone is sufficient to block endothelial hyperpolarization and EDHF-mediated relaxation.20–22 In one study, endothelial and vascular smooth muscle membrane potential was measured in response to endothelial stimulation, and it was found that SKCa channels can contribute to endothelium-dependent hyperpolarization in MCA but only when the arteries could produce NO (eg., not in the presence of NOS inhibition).22 In the present study, under conditions of NOS/COX inhibition, A23187 caused dilation in PA that was similar in magnitude to that elicited by papaverine and zero calcium PSS, suggesting a major influence of EDHF on tone in both control and ischemic vessels. Addition of the specific SKCa inhibitor apamin partially reversed the dilation in both vessel types. This is in contrast to previous studies in the MCA in which apamin had no effect on EDHF-mediated dilation.20–22 However, similar to previous studies in the MCA, IKCa channel inhibition with TRAM-34 completely reversed the EDHF response in both types of PA. These data demonstrate another fundamental difference between MCA and PA in that there appears to be some involvement of SKCa channels in mediating agonist-induced EDHF responsiveness in PA, albeit to a lesser extent than IKCa channels. The finding that SKCa inhibition with apamin only partially reversed, whereas IKCa inhibition with TRAM-34 completely reversed EDHF-mediated vasodilation may be related to differences in channel expression in PA. Quantitative PCR revealed that IKCa expression was 4–5-fold higher than SKCa in these vessels.
In summary, there appear to be fundamental differences in mechanisms of endothelium-dependent relaxation between MCA and penetrating PA, including the influence of SKCa and IKCa on basal tone and the involvement of both channel types in agonist-induced EDHF-mediated dilations in PA, but not in MCA. During I/R, basal tone and reactivity to pressure is relatively intact in PA whereas it is significantly diminished in MCA, as demonstrated by this and other studies.5–7 There is also a selective loss of NO responsiveness in PA after I/R, yet the influence of EDHF, both under basal and stimulated conditions, appears to be preserved. The functional importance of this is not entirely clear, but may serve to preserve CBF under conditions in which NOS is impaired by disease.
We would like to thank Timothy Hunter, M.S., and the technical expertise of the DNA facility at UVM for their help with the PCR experiments. We also would like to thank Mark T. Nelson, Ph.D. for the helpful discussions.
We gratefully acknowledge the continued support of the National Institute of Neurologic Disorders and Stroke (grant NS043316 to MJC) and the Totman Medical Research Trust.