Altered Endothelial Ca2+ Regulation After Ischemia/Reperfusion Produces Potentiated Endothelium-Derived Hyperpolarizing Factor–Mediated Dilations
Background and Purpose— Endothelium-derived hyperpolarizing factor (EDHF)–mediated dilations are potentiated after several pathologies, including ischemia/reperfusion (I/R). However, no study to date has addressed the mechanism by which this potentiation occurs. This study tested the hypothesis that potentiated EDHF-mediated dilations are due to altered endothelial Ca2+ handling after I/R.
Methods— Rat middle cerebral arteries (MCAs) were isolated after 2 hours of MCA occlusion and 24 hours of reperfusion (or sham surgery). This model has been previously demonstrated to produce potentiated EDHF-mediated dilations. MCAs were studied in a pressurized/perfused vessel chamber equipped for the simultaneous measurement of endothelial Ca2+ (with fura 2) and artery diameter. Measures were made after luminal administration of UTP (P2Y2 purinoceptor agonist), 2 MeS-ATP (P2Y1 purinoceptor agonist), and Br-A23187 (receptor-independent Ca2+ ionophore) for sham and I/R MCAs.
Results— I/R resulted in significantly potentiated UTP-mediated dilations (through a P2Y2 purinoceptor) and endothelial Ca2+ responses in the presence of NG-nitro-l-arginine methyl ester (L-NAME) and indomethacin. Endothelial Ca2+ and diameter responses were also significantly potentiated with 2 MeS-ATP (through a P2Y1 purinoceptor) when L-NAME and indomethacin were absent. Br-A23187, a receptor-independent Ca2+ ionophore, produced significantly potentiated endothelial Ca2+ responses after I/R in the presence of L-NAME/indomethacin. Evaluation of artery diameter as a function of endothelial Ca2+ demonstrated no differences between sham and I/R groups.
Conclusions— These findings demonstrate that I/R results in augmented endothelial Ca2+ responses that appear to be downstream of the receptor level. Moreover, these data suggest that this augmented Ca2+ response contributes to the potentiated EDHF-mediated dilations after I/R.
- cerebral arteries
- cerebral ischemia, transient
- endothelium-derived hyperpolarizing factor
- middle cerebral artery
- middle cerebral artery occlusion
Vascular tone may be modulated by several endothelium-derived compounds such as nitric oxide (NO)1,2⇓ and prostacyclin (prostaglandin I2 [PGI2]).3,4⇓ However, an additional vasorelaxation pathway has subsequently been revealed and termed endothelium-derived hyperpolarizing factor (EDHF) (for review, see Golding et al5 and Feletou and Vanhoutte6). Since the discovery of the EDHF-dependent mechanism, EDHF-dependent dilations have been described with increasing regularity in both the peripheral and cerebral circulations.
In addition, recent studies have described potentiated EDHF-mediated dilations after several pathological conditions such as ischemia/reperfusion (I/R),7–9⇓⇓ hypercholesterolemia,10,11⇓ and congestive heart failure.12 With regard to I/R in particular, studies have shown potentiated dilations in response to EDHF-dependent agonists (UTP and A23187) in rat cerebral arteries.7,8⇓ After 2 hours of ischemia and 24 hours of reperfusion, the EDHF-mediated component of the dilation was potentiated approximately 7-fold compared with control.7 Additionally, dog coronary arteries subjected to 1 hour of ischemia and 2 hours of reperfusion also showed evidence of a potentiated EDHF-mediated component.9 These combined findings from varied pathologies have led to the speculation that EDHF-mediated responses may play an additional role as a compensatory or backup mechanism in certain conditions. However, despite the increasing descriptions of potentiated EDHF-mediated dilations in multiple pathologies, no study to date has addressed the mechanism by which that potentiation occurs.
Although the exact mechanism by which EDHF-mediated dilations are initiated is not completely understood, it appears clear that an elevation of endothelial Ca2+ is necessary.13–16⇓⇓⇓ Furthermore, recent data have demonstrated that a specific endothelial Ca2+ threshold exists that, when met, initiates an EDHF-mediated dilation.17 Thus, agents that increase Ca2+ above the EDHF threshold result in the production of EDHF-mediated dilations, whereas those that fail to increase Ca2+ to the threshold do not. Along these lines, one might also expect the potency of EDHF-producing agonists to vary depending on conditions that affect endothelial Ca2+ regulation. For example, one might expect potentiated EDHF-mediated dilations under conditions that favor greater increases in endothelial Ca2+.
Given the critical role of Ca2+ in normal EDHF-mediated dilations, it is reasonable to suspect that alterations in endothelial Ca2+ regulation might account for the potentiated EDHF-mediated dilations after I/R. Therefore, the following studies were designed to evaluate the hypothesis that endothelial Ca2+ regulation is altered after I/R, thus resulting in amplified Ca2+ responses and potentiated dilations for EDHF-dependent agonists. It was predicted that endothelial Ca2+ would reach the EDHF-producing threshold at lower agonist concentrations in I/R arteries compared with shams. To test this hypothesis, endothelial Ca2+ was increased via 2 separate receptor systems (P2Y1 and P2Y2 purinoceptors) and through a receptor-independent mechanism (Br-A23187) in pressurized rat middle cerebral arteries (MCAs). Selective measurement of endothelial Ca2+ was performed in pressurized MCAs from sham and I/R rats with the use of recently developed fluorescence techniques.17,18⇓ Simultaneous measurement of artery diameter permitted the evaluation of endothelial Ca2+ in the vasodilatory responses.
Materials and Methods
All experiments were approved by the Animal Protocol Review Committee at Baylor College of Medicine. A total of 29 male Long-Evans rats were used for these studies (weight, 250 to 350 g). Rats were anesthetized with 3% isoflurane before any procedures.
MCA Occlusion/Reperfusion Surgery
Rats were maintained with 1.5% to 2.5% isoflurane through a nose cone and allowed to breathe freely during surgery. Rats were also treated with 2% lidocaine at the site of surgery before the surgical procedure. Core body temperature was measured rectally and maintained at 37°C with a temperature controller coupled to a heating pad.
The right MCA was occluded for 2 hours, as previously described.7,19⇓ In brief, the right carotid artery was exposed in the area of the carotid bifurcation. The external carotid artery was tied off, severed, and retracted for use as a route to introduce the occluding device. The occluding device consisted of a nylon monofilament (≈242 μm in diameter) with the tip enlarged and rounded with a small bulb of epoxy. The occluder was introduced into a small hole in the external carotid artery and advanced in a retrograde fashion back to the bifurcation and into the internal carotid artery. The occluder was further advanced toward the circle of Willis until the tip met resistance at the opening of the anterior cerebral artery. This placement of the occluder resulted in occlusion of the right MCA at the point of origin off of the circle of Willis. Note that the occluder did not enter the MCA lumen; rather, it simply blocked the opening. Furthermore, transmission electron microscopy studies have shown that the MCA endothelium remains intact after occlusion/reperfusion in this model.8 The occluder was left in place for 2 hours before complete removal and restoration of blood flow. Rats were ambulatory before they returned to the animal care facilities. Sham surgeries were the same as above except that the occluder was removed immediately after placement.
Mounting of Pressurized/Perfused Cerebral Arteries
After 24 hours of reperfusion (after occluder removal), rats were anesthetized with isoflurane, and the brain was removed and placed in ice-cold physiological salt solution (PSS). A 2- to 3-mm section of the MCA ipsilateral to the injury was removed and mounted on 2 glass micropipettes within an isolated vessel chamber designed specifically for fluorescence microscopy (Chuel Tech). The chamber was filled with warmed (37°C) Krebs’ buffer at a pH of 7.4. The vessel was tested for the absence of leaks by clamping the luminal inflow and outflow tubings and monitoring the pressure from in-line pressure transducers. A constant pressure (≈50 mm Hg) over a 30-second time period indicated a leak-free vessel. Each vessel was then pressurized to a mean pressure of 85 mm Hg, and flow was established through the lumen at a rate of 100 μL/min.18 The artery was allowed to equilibrate for 40 to 60 minutes before the fura 2-AM loading procedure. Sham and I/R arteries developed spontaneous tone during the course of the equilibration period.
Measurement of Endothelial Ca2+ in Pressurized/Perfused Arteries
After 40 to 60 minutes of equilibration, an autofluorescence measurement was made of the pressurized artery. The autofluorescence value was later subtracted from the total fluorescence to yield fluorescence resulting purely from fura 2. The endothelium was then selectively loaded with a cell-permeant fluorescent Ca2+-indicating dye, fura 2-AM, by delivering the dye luminally for 5 minutes.18 Fura 2-AM (0.67 μmol/L) was mixed with 0.02% pluronic F-127 in dimethyl sulfoxide. At the end of the loading period, the fura 2 solution was washed out by perfusing with standard PSS. A period of 15 minutes was allowed after the fura 2-AM washout to allow for complete deesterification of fura 2-AM to fura 2 by endogenous endothelial esterases.
Measurement of endothelial Ca2+ in fura 2–loaded arteries was performed with a system combining a fluorometer and a vessel edge-detection device. The fluorometer (C&L Instruments) used an excitation filter wheel to cycle rapidly between 340, 380, and 360 nm with the use of bandpass filters. A dichroic mirror was used to deflect the varying excitation wavelengths to the artery and allow the resulting fluorescence emission to pass through to a beam splitter. An infrared light source was used to transilluminate the artery for simultaneous measurement of artery diameter. The infrared light also passed through the dichroic mirror to the beam splitter. Approximately 20% of the fluorescence/infrared signal was diverted to a charge-coupled device (CCD) camera with the remaining diverted to a photomultiplier detector. The photomultiplier detector was equipped with a 540/40-nm bandpass filter to selectively measure fura 2 fluorescence without interference from the infrared wavelengths. The image from the CCD was directed to the vessel edge-detector device (Living Systems), where the diameter was measured and output in real time into the fluorescence/Ca2+ measurement. The ratio of 340 to 380 fluorescence (R340/380) was converted to intracellular free Ca2+ ([Ca2+]i) on the basis of the following equation:
where β is the ratio of the 380 fluorescence unbound to bound Ca2+, R is the R340/380, Rmax is the R340/380 in saturating Ca2+ conditions, and Rmin is the R340/380 in Ca2+-free conditions. Values for these constants were obtained from a series of in situ calibration curves yielding values as follows: β=5.218, Rmax=2.017, and Rmin=0.1365. The dissociation constant for fura 2 to Ca2+ (Kd) in intact cerebral arteries was 282 nmol/L.20
Concentration-Response Curves to Luminal Agents
Concentration-response curves were performed with the use of luminally administered UTP, 2-methylthioadenosine 5′-triphosphate (2 MeS-ATP), and Br-A23187. A manifold system was used to substantially reduce the dead volume of the luminal tubing and thus greatly speed the rate of getting the drug from the reservoir to the endothelium. Concentration-response curves to UTP and Br-A23187 were performed in the presence of NG-nitro-l-arginine (L-NAME; 50 μmol/L) and indomethacin (10 μmol/L) to rule out the effects of NO and PGI2 formation, respectively. Concentration-response curves to 2 MeS-ATP did not include L-NAME/indomethacin.
After 24 hours of reperfusion, the brain was evaluated for the presence of ischemic injury with 2,3,5-triphenyltetrazolium chloride (TTC). After removal of the MCA segment, the brain was placed in a rat brain matrix (Braintree Scientific) and sliced into 2-mm coronal sections. The sections were incubated in 2.5% TTC for at least 30 minutes. Areas with profound ischemic injury remained white, whereas viable tissue stained red.7,21⇓ Brains were scored on a scale from 0 to 4 as follows: 0, no lesion; 1, small lesion confined to the striatum; 2, lesion largely consuming the striatum; 3, lesion consuming the striatum and involving some cortex; and 4, lesion consuming the striatum and cortex. “I/R” arteries were used only from rats with an injury score of 3 or 4. Note that this reflects stricter inclusion criteria for I/R arteries compared with previous studies, which included animals with scores of 1 through 4.7,22⇓
Chemicals and Buffer Compositions
All drugs and chemicals were obtained from Sigma with the exceptions of fura 2-AM and pluronic F-127 (TefLabs) and Br-A23187 (Molecular Probes). The Krebs’ buffer (PSS) consisted of the following (in mmol/L): 119 NaCl, 4.7 KCl, 21 NaHCO3, 1.18 KH2PO4, 1.17 MgSO4, 0.026 EDTA, 1.6 CaCl2, and 5.5 glucose.18 The buffer was bubbled continuously with 20% O2/5% CO2/balance N2 to yield a pH of 7.4 at 37°C.
Values are reported as mean±SE. Single measurements between groups were compared with Student’s t test. When multiple measurements were performed (such as for a concentration-response curve), a 2-way repeated-measures ANOVA was used to determine whether a statistical difference existed between groups. When group differences did exist, further evaluation of individual differences was performed with a Tukey test. Significance was defined as P<0.05.
Changes in artery diameter are reflected by the following equation:
where Ddrug is the diameter of the artery after administration of an agonist, Dbase is the resting diameter, and Dmax is the maximal diameter of the artery (equivalent to Ca2+-free diameter).
On pressurization (85 mm Hg) and establishment of luminal flow (100 μL/min), sham and I/R MCAs developed spontaneous tone over the course of 40 to 60 minutes. The percent tone at the end of the equilibration period was similar between the 2 groups (Table). The addition of L-NAME/indomethacin resulted in a significant increase in tone in the sham group but not in the I/R group. Tone in the presence of L-NAME/indomethacin was significantly greater in the sham group (38±3% versus 25±2% tone; P<0.01). I/R MCAs were not further constricted with a constricting agent because of the likely alterations that would result in endothelial Ca2+.23
To determine whether altered endothelial Ca2+ responses might account for the potentiated EDHF-mediated dilations after I/R, concentration-response curves to luminal UTP (10−7 to 10−5 mol/L) were performed in the presence of L-NAME and indomethacin. Dilations to UTP in the presence of L-NAME/indomethacin have been demonstrated to be EDHF dependent and potentiated after I/R in this preparation.7,24⇓ MCA diameter changes and endothelial [Ca2+]i were recorded simultaneously in fura 2–loaded arteries for both sham and I/R groups. Figure 1 shows representative experiments for both sham and I/R MCAs in response to UTP. Note that endothelial Ca2+ increased more for a given concentration of UTP in I/R MCAs compared with shams (Figure 1, bottom). In addition, note the potentiated dilation in the I/R MCA at 10−6 mol/L UTP (Figure 1, top). Interestingly, the dilations at 1 μmol/L UTP appeared more transient than the corresponding Ca2+ response in the I/R group. These data are summarized in Figure 2. From Figure 2, it is evident that while resting Ca2+ was similar between sham and I/R groups (n=5 each), UTP produced a significantly greater increase in endothelial Ca2+ in the I/R MCAs for a given concentration of UTP. The horizontal dashed line indicates the Ca2+ threshold (≈340 nmol/L) for initiating an EDHF-mediated dilation from a previous study17 and the present study (Figure 4, bottom). Endothelial Ca2+ in response to 1 μmol/L UTP in the I/R group exceeds the Ca2+ threshold for EDHF-mediated dilations, whereas Ca2+ from the sham group does not. The dilatory response to UTP was also significantly potentiated in the I/R group. In particular, dilations at 1 μmol/L UTP were significantly greater in the I/R group. Although endothelial Ca2+ was considerably higher in the I/R group at 10 μmol/L UTP, differences in percent diameter change did not exist between the groups because of maximal dilations that occurred in each group.
To determine whether the potentiated Ca2+ and diameter responses were specific to the receptor system used above, an additional agonist, 2 MeS-ATP, was evaluated. UTP and 2 MeS-ATP operate through 2 separate receptor systems and dilatory mechanisms. UTP stimulates a P2Y2 purinoceptor that results in an increase in endothelial Ca2+ through an inositol 1,4,5-triphosphate (IP3)–dependent mechanism with the subsequent production of NO and EDHF.24,25⇓ 2 MeS-ATP stimulates a P2Y1 purinoceptor that produces an increase in endothelial Ca2+ in an IP3–independent manner with subsequent release of NO alone under normal conditions.25–27⇓⇓ Figure 3 represents summary data for concentration-response curves to 2 MeS-ATP (10−8 to 10−5 mol/L) for sham and I/R MCAs (n=4 each). L-NAME and indomethacin were omitted from these experiments to obtain corresponding NO-dependent diameter measurements. Similar to the UTP-mediated responses, 2 MeS-ATP produced potentiated Ca2+ responses after I/R. Furthermore, the greater Ca2+ response after I/R was accompanied by a potentiated diameter response as well.
On the surface, it would seem paradoxical that diameter at 10−8 mol/L UTP was significantly different between groups while endothelial Ca2+ was not. However, there are 2 reasons for this apparent discrepancy. The first is that the measure for Ca2+ has a lower level of precision than that for diameter. Since the measure of endothelial Ca2+ is subject to more variability in measurement, it would require more animals to reach significance. The second part that contributes to this apparent discrepancy is due to the fact that the endothelial [Ca2+]i at 10−8 mol/L 2 MeS-ATP falls right around the Ca2+ threshold for NO-mediated responses in these arteries (220 nmol/L Ca2+).17 Therefore, increases in endothelial Ca2+ that fail to reach the threshold of 220 nmol/L Ca2+ do not produce an increase in diameter, whereas increases above the threshold do. Importantly, only relatively small increases from baseline in endothelial Ca2+ are necessary to reach that threshold. Furthermore, because a threshold exists, a linear relationship between Ca2+ and diameter would not be expected. Rather, significant increases in diameter would be expected once Ca2+ entered the threshold range. In this way, relatively small changes in endothelial Ca2+ (that do not reach statistical difference) could result in a more amplified difference in diameter responses (that does reach statistical difference).
To determine whether the potentiated Ca2+ and diameter responses were dependent on receptor-mediated events, concentration-response curves were conducted with Br-A23187 in the presence of L-NAME and indomethacin. Br-A23187 is a nonfluorescent Ca2+ ionophore used to increase endothelial [Ca2+]i through a non–receptor-dependent process. Concentration-response curves to luminally delivered Br-A23187 (1 to 6 μmol/L) resulted in a significantly greater increase in endothelial Ca2+ for a given concentration of Br-A23187 (Figure 4, top; n=4 each).
Data from individual experiments with Br-A23187 and UTP were combined and plotted as percent diameter versus endothelial Ca2+ (Figure 4, bottom). The data derived from UTP and Br-A23187 did not differ from each other and were therefore combined into 1 plot. Two things are apparent from these data. First, there is close overlap of the sham and I/R data. Fitting of the data to a Gompertz equation shows that a similar relationship between endothelial [Ca2+]i and diameter exists between sham and I/R groups. Note that the relationship is flat at the beginning (where the Ca2+ threshold has not yet been reached) and at the end (where the artery reaches maximal dilation). This similar relationship between endothelial [Ca2+]i and diameter between groups suggests that I/R does not affect the endothelial Ca2+ threshold for initiating the EDHF-mediated dilations. Second, there is a sharp transition between no dilation and maximal dilation for a given steady state endothelial [Ca2+]i (Figure 4, bottom). This steep response reflects the rather narrow endothelial [Ca2+]i range for no activation to maximal activation of the EDHF-dependent mechanism. A previous study has demonstrated that this Ca2+ range starts at approximately 340 nmol/L,17 indicated by a vertical dashed line.
A growing number of recent studies have reported upregulated EDHF-mediated dilations after a variety of pathological conditions, including I/R,7–9⇓⇓ hypercholesterolemia,10,11⇓ and congestive heart failure.12 However, none of these studies has addressed the mechanism by which EDHF-mediated responses are potentiated. Presumably, this lack of data regarding the potentiated EDHF mechanism has been due in part to the incomplete understanding of the mechanism by which EDHF-mediated responses occur.
Although there is still considerable debate regarding the identity of EDHF (or whether EDHF is a factor per se), it is generally accepted that EDHF-mediated responses are initiated by an increase in endothelial Ca2+.13–16⇓⇓⇓ Furthermore, recent data from this laboratory have demonstrated that an endothelial Ca2+ threshold exists that, when met, initiates an EDHF-mediated dilation.17 The threshold for EDHF-mediated dilations is higher than that for NO-mediated dilations (340 versus 220 nmol/L), explaining why NO-mediated responses are typically found to precede EDHF-mediated responses. Additionally, it was shown that a normally non-EDHF–producing agonist (2 MeS-ATP) could elicit an EDHF-mediated response in conditions in which 2 MeS-ATP–mediated endothelial Ca2+ responses were augmented in order to reach the critical Ca2+ threshold.17 Thus, it appears that endothelial [Ca2+]i critically regulates the EDHF-mediated response. In the present study it was hypothesized that an augmented endothelial Ca2+ response accounts for the potentiated EDHF-mediated response after I/R. This hypothesis was addressed by simultaneously measuring endothelial Ca2+ and artery diameter in pressurized arteries.
Two separate and distinct endothelial purinoceptors (P2Y2 and P2Y1) were used in addition to a Ca2+ ionophore (Br-A23187) in order to increase endothelial Ca2+. UTP, which is selective for P2Y2 purinoceptors, has been demonstrated to produce both NO- and EDHF-mediated responses in the rat MCA.7,24⇓ 2 MeS-ATP, which is selective for P2Y1 purinoceptors, produces dilations through an exclusively NO-dependent mechanism.7,27⇓ Release of Ca2+ through the P2Y2 purinoceptor is believed to depend on the production of IP3, whereas that of the P2Y1 purinoceptor does not.25,26⇓ Br-A23187 elevates endothelial Ca2+ in a receptor-independent fashion, thus bypassing any effect of receptor number or coupling.
Potentiated endothelial Ca2+ responses were found with the use of both receptor-dependent and -independent mechanisms after I/R. The augmented Ca2+ release found in the present study with both UTP and 2 MeS-ATP might suggest either (1) upregulation of both P2Y1 and P2Y2 purinoceptors, (2) potentiation of a common mechanism between the 2 receptor systems, or (3) potentiation of the Ca2+ response downstream of receptor stimulation. However, the additional finding of potentiated Ca2+ responses with the Ca2+ ionophore (Br-A23187) suggests that the potentiation of the Ca2+ response occurs downstream of the receptors. Since increased Ca2+ responses were found with receptor-dependent mechanisms as well as with receptor-independent mechanisms, it would appear that the potentiated Ca2+ is due to more fundamental changes in Ca2+ handling. For instance, there could be a greater increase of [Ca2+]i (from internal stores or from external sources), or there could be a reduced removal of [Ca2+]i (to internal stores or extrusion across the plasma membrane) after I/R. Either one or both of the above possibilities would result in augmented Ca2+ responses.
On the basis of the aforementioned findings, it would be tempting to conclude that the effects of I/R appear to exclude the possibility of greater Ca2+ release from internal stores. The reasoning behind such a conclusion would be based on the assumptions that Br-A23187 acts purely as a direct mediator of Ca2+ influx28 and that the 2 MeS-ATP response does not involve an IP3-mediated Ca2+ increase.26 However, such a conclusion might be incorrect for the following 2 reasons. First, in addition to its ionophoretic properties, Br-A23187 has been suggested to stimulate Ca2+ entry through store-operated Ca2+ channels secondary to depletion of internal store Ca2+.29 This release of Ca2+ from internal stores may be due to the ability of Br-A23187 to promote the liberation of inositol phosphates such as IP3.30 Second, while 2 MeS-ATP does not appear to involve a significant IP3-dependent mechanism under normal circumstances, one cannot assume that IP3 is not involved in pathological circumstances. Thus, while the present data demonstrate significant alterations in Ca2+ handling after I/R, further studies will be needed to identify where that alteration occurs.
One additional possibility that could contribute to the potentiated dilations after I/R is that of Ca2+ sensitivity. Although the aforementioned data clearly demonstrate that potentiated Ca2+ responses occur, it is possible that an increased Ca2+ sensitivity could also contribute to the potentiated dilations. To evaluate the possibility of altered Ca2+ sensitivity, percent diameter change was plotted against endothelial [Ca2+]i (produced by various concentrations of UTP and Br-A23187). This plot demonstrates the sharp transition between no dilation and maximal dilation characteristic of these EDHF-mediated responses. Furthermore, it is clear that there is no significant leftward shift of the curve (which would indicate increased Ca2+ sensitivity) after I/R. Thus, the data are suggestive that increased Ca2+ sensitivity does not play a measurable role in the potentiated EDHF-mediated dilations. However, given the inherent limitations of the bioassay system, one cannot completely rule out the possibility of multiple offsetting factors (such as in the vascular reactivity to EDHF) downstream of Ca2+.
In summary, potentiated EDHF-mediated dilations after I/R appear to result from augmented endothelial Ca2+ responses. These augmented Ca2+ responses are not receptor dependent but rather appear to result from a more fundamental alteration in endothelial Ca2+ regulation. Additional studies will be required to determine the specific mechanism by which endothelial Ca2+ regulation is altered.
This study was supported by American Heart Association grant 0230353N (to Dr Marrelli), National Institute of Neurological Disorders and Stroke grant NS-27616, and Bugher Foundation Award 0270110N.
- Received February 6, 2002.
- Revision received April 16, 2002.
- Accepted May 7, 2002.
- ↵Golding EM, Marrelli SP, You J, Bryan RM Jr. Endothelium-derived hyperpolarizing factor in the brain: a new regulator of cerebral blood flow? Stroke. 2002; 33: 661–663.
- ↵Marrelli SP, Childres WF, Goddard-Finegold J, Bryan RM Jr. Potentiated EDHF-mediated dilations in the rat middle cerebral artery following ischemia/reperfusion. In: Vanhoutte PM, ed. EDHF 2000. London, England: Taylor & Francis; 2001.
- ↵Tomioka H, Hattori Y, Fukao M, Watanabe H, Akaishi Y, Sato A, Kim TQ, Sakuma I, Kitabatake A, Kanno M. Role of endothelial Ni(2+)-sensitive Ca(2+) entry pathway in regulation of EDHF in porcine coronary artery. Am J Physiol. 2001; 280: H730–H737.
- ↵Marrelli SP. Mechanisms of endothelial P2Y(1)- and P2Y(2)-mediated vasodilatation involve differential [Ca2+]i responses. Am J Physiol. 2001; 281: H1759–H1766.
- ↵Longa EZ, Weinstein PR, Carlson S, Cummins R. Reversible middle cerebral artery occlusion without craniectomy in rats. Stroke. 1989; 20: 84–91.
- ↵Bederson JB, Pitts LH, Germano SM, Nishimura MC, Davis RL, Bartkowski HM. Evaluation of 2,3,5-triphenyltetrazolium chloride as a stain for detection and quantification of experimental cerebral infarction in rats. Stroke. 1986; 17: 1304–1308.
- ↵Marrelli SP, Johnson TD, Khorovets A, Childres WF, Bryan RM Jr. Altered function of inward rectifier potassium channels in cerebrovascular smooth muscle after ischemia/reperfusion. Stroke. 1998; 29: 1469–1474.
- ↵Dora KA, Doyle MP, Duling BR. Elevation of intracellular calcium in smooth muscle causes endothelial cell generation of NO in arterioles. Proc Natl Acad Sci U S A. 1997; 94: 6529–6534.
- ↵You J, Johnson TD, Marrelli SP, Mombouli JV, Bryan RM Jr. P2u receptor–mediated release of endothelium-derived relaxing factor/nitric oxide and endothelium-derived hyperpolarizing factor from cerebrovascular endothelium in rats. Stroke. 1999; 30: 1125–1133.
- ↵Frelin C, Breittmayer JP, Vigne P. ADP induces inositol phosphate-independent intracellular Ca2+ mobilization in brain capillary endothelial cells. J Biol Chem. 1993; 268: 8787–8792.
- ↵Taniguchi H, Tanaka Y, Hirano H, Tanaka H, Shigenobu K. Evidence for a contribution of store-operated Ca2+ channels to NO-mediated endothelium-dependent relaxation of guinea-pig aorta in response to a Ca2+ ionophore, A23187. Naunyn Schmiedebergs Arch Pharmacol. 1999; 360: 69–79.
- ↵Moscat J, Moreno F, Herrero C, Lopez C, Garcia-Barreno P. Endothelial cell growth factor and ionophore A23187 stimulation of production of inositol phosphates in porcine aorta endothelial cells. Proc Natl Acad Sci U S A. 1988; 85: 659–663.