Structure, Function, and Endothelium-Derived Hyperpolarizing Factor in the Caudal Artery of the SHR and WKY Rat
Objective— To quantify structural and functional characteristics of the caudal artery from spontaneously hypertensive (SHR) and normotensive Wistar Kyoto (WKY) rats with particular reference to endothelium-derived hyperpolarizing factor (EDHF).
Methods and Results— Ultrastructural studies showed that the number of myoendothelial gap junctions, smooth muscle cell (SMC) layers, and medial cross-sectional area were significantly greater in SHR than WKY. Intracellular dye labeling demonstrated hyperplasia of SMCs in SHR. Analysis of nerve-mediated excitatory junction potentials recorded in SMCs at the adventitial and luminal borders demonstrated decreased radial coupling of SMCs in SHR. In both SHR and WKY, in the presence of NG-nitro-l-arginine methyl ester and indomethacin, acetylcholine-elicited EDHF was abolished by charybdotoxin and apamin, while iberiotoxin had no effect, implicating the involvement of small and intermediate, but not large, calcium-activated potassium channels. EDHF was abolished by Gap-mimetic peptides, 18β-glycyrrhetinic acid, and endothelial removal but not affected by the NO scavengers hydroxocobalamin and carboxy-PTIO.
Conclusions— Significant differences in SMC morphology and homocellular and heterocellular coupling exist between the caudal artery of SHR and WKY rats. In the caudal artery of SHR, significantly greater heterocellular coupling compensates for other structural changes in the media to maintain a functional role for EDHF.
Heterocellular and homocellular coupling of arterial smooth muscle cells (SMCs) and endothelial cells (ECs) within the walls of blood vessels plays an integral role in the coordination of vasomotor activity.1,2⇓ Furthermore, the variability in vasomotor responses reported in and between different vascular beds1 depends in part on differences in the incidence and characteristics of vascular gap junctions and their constituent connexin proteins.2,3⇓ Thus, structural variation is an important determinant of functional variation within the vasculature. For example, myoendothelial gap junctions (MEGJs), which permit the electrical and chemical coupling of EC and SMC layers, are increasingly considered integral to the activity of endothelium-derived hyperpolarizing factor (EDHF), one of three important endothelium-derived vasodilatory factors.2–5⇓⇓⇓ The importance of EDHF varies between different vessels and generally increases as vessel size decreases.2 Indeed it has been shown that the prevalence of MEGJs in different segments of the mesenteric vascular bed of the rat corresponds to the participation of EDHF in vasodilatory responses previously reported in these vessels.6,7⇓ Furthermore, in the femoral artery of the Wistar rat, which lacks EDHF, there are no MEGJs.8
Interestingly, while the majority of studies have emphasized the greater importance of EDHF in smaller-sized resistance vessels,2,3⇓ recent studies have demonstrated a significant role for EDHF in larger resistance arteries and in conduit vessels.5,9,10⇓⇓ While MEGJs have been implicated in EDHF activity in smaller vessels, it is not clear whether they may be important in larger vessels which have more layers of smooth muscle cells.1 Our preliminary studies showed that MEGJs are present in large arteries, such as the caudal artery. However, no studies have examined the role of MEGJs in EDHF function in such larger vessels, nor have they correlated the incidence of MEGJs with changes in EDHF, which might occur during hypertension.
In hypertension, remodeling of the vascular wall accompanies changes in blood pressure. This remodeling typically comprises a decrease in luminal diameter and, frequently, an increase in medial thickness.11 The underlying basis for the change in the media is heterogeneous, being variously attributed to hypertrophy, hyperplasia, or rearrangement of SMCs.12,13⇓ Nothing is known, however, about cellular coupling in the media of hypertensive vessels following vascular remodeling, nor whether other anatomical changes may occur during hypertension to modify the functional coupling in and between the two cell layers.
A number of vascular diseases, including hypertension, are characterized by endothelial dysfunction whereby alterations in the production and action of endothelium-derived vasodilatory substances, such as NO, prostaglandins, and EDHF occur.3,5⇓ To date, the only studies to examine the hyperpolarization associated with EDHF have been in the mesenteric arteries of spontaneously hypertensive rats (SHR)14 and stroke-prone SHR,15 where EDHF is reduced. Clearly, increases in the number of SMCs in the media of hypertensive vessels could have an impact on the ability of an endothelial-derived factor like EDHF to hyperpolarize the smooth muscle. Furthermore, while the incidence of cytoplasmic projections of ECs toward SMCs has been shown to increase in steroid and ligation-induced hypertension,16,17⇓ no studies have determined the incidence of pentalaminar MEGJs and correlated these with the activity of EDHF in hypertension.
In the present study, we have quantified the structural and functional differences between the caudal artery of the hypertensive SHR and normotensive Wistar Kyoto (WKY) rats with particular reference to the role of EDHF. We hypothesized that the incidence of MEGJs would reflect the importance of EDHF as a vasodilatory factor in normotensive rats and that the incidence of MEGJs would be altered in the caudal artery of the SHR in line with changes in the relative contribution of EDHF as a vasodilatory factor in hypertension.
For detailed methods, please see http://atvb.ahajournals.org.
Blood Pressure Recordings
Systolic blood pressure of SHR was significantly greater than that of WKY animals (P<0.05; 188±6.4 and 121±2.6 mm Hg, n=10, for SHR and WKY, respectively).
Arterial and Smooth Muscle Cell Morphology
The number of layers of SMCs and the medial cross sectional area was significantly larger in SHR than in WKY, while the circumference as measured at the level of the internal elastic lamina (IEL) was significantly smaller (P<0.05; Table 1).
Following morphometric analysis of intracellular dye labeling with fluorescein-conjugated dextran, the surface areas of SMCs in SHR and WKY were found not to be significantly different (P>0.05; 782±56 μm2, 18 cells, from 5 rats, and 757±41 μm2, 23 cells, from 6 rats, for SHR and WKY, respectively), although the cells were significantly longer and thinner in the SHR than in the WKY (P<0.05; length, 171±6.7 μm and 138±4.4 μm; width, 8.1±0.4 μm and 9.8±0.5 μm; Figure 1; compare 1B with 1D; for SHR and WKY, respectively).
With lucifer yellow, dye was commonly seen to spread among adjacent SMCs in arteries from both SHR and WKY (Figure 1, compare 1A with 1C, 1E, and 1F; for SHR and WKY, respectively). This spread of dye was in the radial direction between cells, rather than between cells along the length of the vessel (Figure 1A, 1C, 1E, and 1F) and occurred irrespective of whether electrodes were backfilled with KCl or LiCl. In all cases, the impaled cell lay superficial to the other labeled cells. That is, closer to the adventitial border when impaled from the outside or closer to the lumen, when impaled from the luminal surface. Endothelial cells were never labeled after impalements of SMCs at either the adventitial or luminal border. No dye was observed to spread when electrodes were filled with fluorescein-conjugated dextran.
At the ultrastructural level, very small gap junctions (≤100 nm in profile), which usually appeared in only one serial section (Figure IA and IB, available at http://atvb.ahajournals.org), were found between SMCs. Such junctions were found between adjacent processes from two SMCs (Figure IA,B, online data supplement). On the other hand, large gap junctions (≥300 nm in profile), which appeared in several or more consecutive sections, were readily found between adjacent ECs (Figure IC, online data supplement).
Electrical Coupling of Smooth Muscle Cells
Resting membrane potential of SMCs did not differ significantly between SHR and WKY (P>0.05; −59±1 mV, n=19, from 7 different animals for SHR and −59±1 mV, n=23 from 7 different animals for WKY). Stimulation of nerves located at the adventitial border of the media evoked excitatory junction potentials (EJPs), which were significantly larger in the SMCs at the adventitial border of the vessels from SHR, than those recorded in SMCs at the adventitial border in the WKY (P<0.05; Figure 2; Table 2), although there was no significant difference in the time constants of these potentials in the two rat strains (P>0.05; 239±15 ms for SHR and 278±18 ms for WKY). Repetitive stimulation (2 to 10 impulses at 10 Hz) produced summation of EJPs and contraction. In some cases, muscle action potentials were recorded; their incidence was not different between the SHR and WKY.
Intracellular recordings were made from cells at the luminal border of vessels from both SHR and WKY and these cells were identified as SMCs by dye labeling (Figure 1E). Stimulation of nerves at the adventitial border evoked EJPs in these luminal SMCs due to electrical coupling of SMCs in the media. There were no differences in the properties of these EJPs when recorded from the SHR or WKY (P>0.05; Table 2). In arteries of both SHR and WKY, there was a significant increase in rise time of EJPs recorded from cells on the luminal compared with the adventitial border (P<0.05; Table 2). In SHR, there was a significant decrease in the amplitude of the EJPs recorded from SMCs located at the adventitial and luminal borders. On the other hand, in WKY, there was no significant difference in the amplitude of EJPs recorded from SMCs located at the adventitial and luminal borders. In both SHR and WKY arteries, EJPs were followed by small, slow membrane depolarizations, as reported previously.18
Incidence of MEGJs
There were significantly more MEGJs in the SHR compared with WKY (P<0.05; Table 1). Two types of MEGJs were found: those between projections of ECs and SMCs (Figure 3A–C) and those between the projections of ECs and the surface of SMCs (Figure 3D–F). Of the total of 32 MEGJs found in both SHR and WKY, 59% were found on projections of both ECs and SMCs, while the remainder were found on projections originating from ECs only. No differences in the proportion of these two types of MEGJ were found between SHR and WKY.
The proximity of MEGJs and gap junctions between ECs was examined in SHR and WKY. In both cases, MEGJs were within 2 μm of endothelial gap junctions (Figure 3C), and no significant difference was found for the distance between the MEGJs and endothelial gap junctions in the two groups (P>0.05; 1.62±0.18 μm, n=19 for SHR and 1.68±0.35 μm, n=12, for WKY).
The majority of MEGJs had single plaques or regions of pentalaminar membrane structure (Figure 3A through 3C, for example). Five of the 20 MEGJs from SHR and 1 of the 12 MEGJs from WKY, however, had multiple plaques (Figure 3D through 3F, for example). The longitudinal distance over which MEGJs with single plaques were seen in vessels from SHR and WKY was determined by comparing the number of transverse sections in which each plaque appeared. The number of sections in which individual plaques appeared was not significantly different between SHR and WKY (P>0.05; 3.4±0.52 sections, n=15 for SHR and 3.4±0.48 sections, n=11, for WKY). The width of single plaques was measured from high-magnification micrographs and found not to differ between SHR and WKY (P>0.05; 135±16 nm, n=15 for SHR and 125±18 nm, n=11 for WKY). The MEGJs with multiple plaques were found in more sections than MEGJs with single plaques (6.3±0.6, n=4 for SHR, and 4 for the single MEGJ in WKY). The remaining MEGJ in the SHR series was found at the end of a series and therefore was not included in the present analysis, as the pentalaminar region may have continued outside the 50 sections examined.
EDHF Activity in Caudal Arteries
In the presence of L-NAME and indomethacin, ACh produced a concentration-dependent relaxation in caudal arteries from both SHR and WKY, which was not significantly different between the two strains (n=5 and 6, respectively; Figure 4A). The combined addition of charybdotoxin and apamin abolished the EDHF relaxation in both SHR and WKY (n=4, for each) and uncovered a contraction that was not significantly different between the two strains. 18β-Glycyrrhetinic acid significantly reduced the EDHF relaxation in SHR by 63% (remaining vasodilation was 10±4.6% of phenylephrine-induced constriction, n=4) and by 100% in the WKY (n=3). Recirculation of Krebs’ solution containing the Gap-mimetic peptide combination (43Gap26, 40Gap27, and37,43Gap27), L-NAME, and indomethacin, however, significantly reduced the EDHF relaxation by 95.8±4.3% in the SHR and by 79.3±9.4% in the WKY (remaining vasodilation was 1.8±1.8% and 7±3% of phenylephrine-induced constriction, n=4 for each; Figure II, available at http://atvb. ahajournals.org). Control data showed that the EDHF relaxation (31.8±3.6%, n=7) was not altered after 60 minutes of recirculation of the Krebs’ solution (109.4±7.2%, n=7, of the initial response to ACh). In the presence of indomethacin and the Gap-mimetic peptide combination, but in the absence of L-NAME, ACh did elicit a relaxation (82.4±4.4%; n=5, WKY data only; Figure II, available at http://atvb.ahajournals.org), that was abolished by subsequent incubation in L-NAME (n=4; Figure II). In the presence of L-NAME and indomethacin and in the additional presence of hydroxocobalamin (n=5, for each) or carboxy-PTIO (n=3, for each), the EDHF relaxation was unchanged (SHR, 51.7±7.9% and 26.8±4.9%; WKY, 38.2±9.6% and 34.5±5.2% for hydroxocobalamin and carboxy-PTIO, respectively, for SHR and WKY, respectively). The subsequent addition of charybdotoxin and apamin abolished the EDHF relaxation in both SHR and WKY (n=3, for each) and uncovered a contraction that was not significantly different between the two strains.
ACh produced concentration-dependent hyperpolarizations in adventitial SMCs of arteries from both SHR and WKY in the presence of L-NAME and indomethacin, although the maximal response was 28% less in the SHR than in the WKY (P<0.05, SHR: −13 mV, WKY: −18 mV; Figure 4B). L-NAME had no effect on resting membrane potential in either strain, although membrane potentials were depolarized by 5 to 10 mV in the presence of indomethacin. Differences in the maximum response were not due to desensitization following the cumulative addition of ACh, because application of single concentrations of ACh produced similar data.
In both SHR and WKY, the addition of charybdotoxin (60 nmol/L) and apamin (0.5 μmol/L; n=8, for each) prevented the hyperpolarization following application of ACh. However, in all 8 SHR animals and in 4 of 8 WKY animals, ACh now produced a depolarization (7±1 mV hyperpolarization and 3±1 mV depolarization, 10±2 mV hyperpolarization and 2±1 mV depolarization, before and after addition of drug, respectively, for SHR and WKY, Figure 4Cb). Removal of the endothelium abolished both the hyperpolarization and the depolarization in SHR and WKY (n=3, for each). In both SHR and WKY, iberiotoxin (100 nmol/L) had no effect on the hyperpolarization induced by ACh (4±2 mV and 4±1 mV; 13±3 mV and 13±4 mV; before and after addition of drug, respectively, for SHR and WKY, respectively; n=3, for each, Figure 4D), while the addition 18β-glycyrrhetinic acid (20 μmol/L) abolished the hyperpolarization to ACh (5±1 mV and 0±0 mV; 12±2 mV and 0±0 mV; before and after addition of drug, respectively, for SHR and WKY, respectively, in both SHR and WKY; n=3, for each). In the presence of L-NAME and indomethacin, the addition of hydroxocobalamin (100 μmol/L), had no effect on the hyperpolarization elicited by ACh (8±1 mV and 8±2 mV and 9±2 mV and 8±3 mV, before and after addition of hydroxocobalamin, for SHR and WKY, respectively).
Significant differences have been found in the present study between the media of the caudal artery of hypertensive SHR and normotensive WKY rats. We found a significantly larger medial cross-sectional area in SHR due to an additional two layers of SMCs. In contrast to the concept that eutrophic remodeling predominates in arteries of SHR, the present study demonstrates that hypertrophic remodeling occurs in the caudal artery of the SHR, consistent with the concept that the type of remodeling may vary according to the model examined and the vascular bed under investigation.11 Furthermore, our intracellular dye labeling studies have shed light on the basis for the hypertrophic remodeling seen in the caudal artery of the SHR. We have found that SMCs in SHR were not significantly different in size, although they varied in shape from those in WKY. Given the increased number of SMC layers and medial area in the SHR, these results suggest that hypertrophy in the caudal artery results from hyperplasia of SMCs. Since previous studies in other vessels of SHR have variously described both SMC hypertrophy12 and hyperplasia,13 we conclude that considerable heterogeneity also exists in the basis for the hypertrophic remodeling in different vascular beds, even within the same model of hypertension.
Our data demonstrate clearly that electrical and chemical coupling occurs between SMCs in the media of the caudal artery in both SHR and WKY rats. Dye transfer is seen within groups of SMCs and pentalaminar gap junctions are present between cells in adjacent layers of the media. However, our electrophysiological studies suggest that electrical coupling is reduced between SMCs in SHR compared with WKY. In WKY rats, differences in the time course of neurally evoked EJPs were observed between SMCs, near the sites of neurotransmitter release at the adventitial border, and at the luminal border, several cells away from these sites. However the amplitude of EJPs was not significantly attenuated, consistent with the properties of an electrically short cable.19 Previous studies in the caudal artery of Wistar rats did show a significant decrease in amplitude of EJPs in luminal versus adventitial cells,18 although the reason for this discrepancy is not known. In contrast to the data from WKY rats, we found a significant attenuation of the amplitude, in addition to slowing of the time course, of neurally evoked EJPs in the SMCs on the luminal border compared with the adventitial border in SHR. The observed 25% increase in the number of SMC layers in the caudal artery of SHR would result in an increase in the conduction distance for EJPs from the adventitial to the luminal border. However, the changes seen in the properties of EJPs between the adventitia and lumen were larger than would be expected solely from the increase in the number of SMC layers in the wall of the SHR, suggesting that coupling between SMCs in SHR has been reduced. Despite the attenuation in the amplitude of EJPs in SHR, there was no difference in the properties of EJPs recorded in the luminal SMCs in SHR compared with WKY. This occurred because the EJPs in adventitial SMCs of SHR were significantly larger and faster than those in WKY, in accord with previous studies in mesenteric arteries from SHR and WKY rats.20 These differences presumably relate to the approximate doubling in the density of the sympathetic innervation found in the caudal artery of SHR compared with WKY.21
In general, the role of EDHF is recognized to be more important in smaller resistance vessels, while NO is more important in larger vessels.2,3,7⇓⇓ Interestingly, however, in the present study we have described a significant EDHF-mediated hyperpolarization and relaxation in a large resistance artery22 of both hypertensive and normotensive rats. We found that the maximum hyperpolarizing response to ACh in SHR was only 28% less than that in WKY. This appeared to be due to the presence of an ACh-induced depolarization and contraction in this strain. On the other hand, EDHF induced similar-sized relaxations in the two strains. After blockade of EDHF in WKY, an ACh-induced contraction was also found, although this did not appear to result from depolarization like that in SHR. We confirmed that the hyperpolarizations and relaxations in both SHR and WKY were due to EDHF by demonstrating their abolition after application of charybdotoxin and apamin, universally accepted blockers of EDHF-mediated responses2,4,23⇓⇓ or by removal of the endothelium. The lack of effect of iberiotoxin on the hyperpolarization indicated that it did not result from the opening of large conductance calcium-activated potassium channels which have been shown to be activated by epoxyeicosatrienoic acid, a putative EDHF candidate.2–5⇓⇓⇓ The involvement of intercellular gap junctions in the EDHF response was confirmed by the abolition of the hyperpolarization in both strains and the marked attenuation and abolition of the relaxation in the SHR and WKY, respectively, by 18β-glycyrrhetinic acid. The involvement of intercellular gap junctions in the EDHF response was further confirmed by inhibition of the relaxation in both strains by Gap-mimetic peptides.8–10⇓⇓ These latter peptides had no effect on ACh-induced relaxation due to NO. Recent studies24 have demonstrated that endothelium-dependent relaxation attributed to EDHF, may in part be due to L-NAME-insensitive (non-NO synthase) basal NO. However, the relaxation due to this latter component, as with EDHF, varies between vascular beds,25 being absent in the femoral artery of the mature Wistar rat,8 for example. We found no evidence for such an L-NAME insensitive NO component following the use of the two different NO scavengers.
In the caudal artery of both SHR and WKY, MEGJs were found connecting the endothelium and smooth muscle. Serial section electron microscopy showed that the incidence of MEGJs was significantly greater in SHR than in WKY, without an apparent difference in the morphology or size of the plaques. This greater degree of heterocellular coupling in the SHR may serve to offset the observed changes in the number of SMCs and their coupling in the media of the caudal artery of these rats, thus maintaining EDHF-mediated relaxation.
Implicit in discussions of cell coupling and blood vessel function is that SMCs are equally well coupled throughout the media, in both the radial and longitudinal directions.1 Contrary to this view, our results suggest that the coupling of SMCs in the radial direction is different from the longitudinal coupling of SMCs in both SHR and WKY. Intracellular injection of lucifer yellow into SMCs at the adventitial or luminal borders, spread into adjacent cells in a radial, but not longitudinal direction. These data are consistent with previous studies in which lucifer yellow did not spread in a longitudinal direction between SMCs in arterioles containing only a single layer of medial SMCs.26,27⇓ While dye coupling indicates that electrical coupling can occur, a lack of electrical coupling cannot be inferred from a lack of dye coupling.1,2,28⇓⇓ Thus, in the present study, the selective movement of lucifer yellow in the radial axis of the caudal artery in both SHR and WKY suggests that there may be interesting differences in the connexin makeup of the gap junctions involved in radial compared with longitudinal cell coupling and hence an asymmetry in the communication pathways within the media.
In conclusion, we have shown that significant anatomical and functional differences are found in the caudal artery of SHR and WKY rats. In the SHR, these differences are characterized by hyperplasia of SMCs and a reduction in SMC coupling, but an increased incidence of MEGJs which provide the potential for increased heterocellular coupling between the two cellular layers. Physiologically, these structural changes result in the maintenance of EDHF activity, which is shown to be dependent on intercellular coupling via gap junctions.
This work was carried out during the tenure of an award from the National Heart Foundation of Australia. S.L.S. was supported by a Peter Doherty Fellowship from the National Health and Medical Research Council of Australia. We thank Professor David Hirst for critical comments on the manuscript and Dr Marianne Tare for invaluable advice regarding the use of the wire myograph.
- Received February 10, 2003.
- Accepted March 5, 2003.
- ↵Beny J-L. Information networks in the arterial wall. News Physiol Sci. 1999; 14: 68–73.
- ↵Campbell WB, Gauthier KM. What is new in endothelium-derived hyperpolarizing factors? Curr Opin Nephrol Hyperten Res. 2002; 11: 177–183.
- ↵Sandow SL, Hill CE. The incidence of myoendothelial gap junctions in the proximal and distal mesenteric arteries of the rat is suggestive of a role in EDHF-mediated responses. Circ Res. 2000; 86: 341–346.
- ↵Shimokawa H, Yasutake H, Fujii K, Owada K, Nakaike R, Fukumoto Y, Takayanagi T, Nagao T, Egashira K, Fujishima M, Takeshita A. The importance of the hyperpolarizing mechanism increases as the vessel size decreases in endothelium-dependent relaxations in rat mesenteric circulation. J Cardiovasc Pharmacol. 1996; 28: 703–711.
- ↵Sandow SL, Tare M, Coleman HA, Hill CE, Parkington HC. Involvement of myoendothelial gap junctions in the actions of endothelium-derived hyperpolarizing factor. Circ Res. 2002; 90: 1108–1113.
- ↵Chaytor AT, Martin PE, Edwards DH, Griffith TM. Gap junctional communication underpins EDHF-type relaxations evoked by ACh in the rat hepatic artery. Am J Physiol. 2001; 280: H2441–H2450.
- ↵Griffith TM, Chaytor AT, Taylor HJ, Giddings BD, Edwards DH. cAMP facilitates EDHF-type relaxations in conduit arteries by enhancing electrotonic conduction via gap junctions. Proc Natl Acad Sci U S A. 2002; 99: 6392–6397.
- ↵Mulvany MJ, Baumbach GL, Aalkjaer C, Heagerty AM, Korsgaard N, Schiffrin EL, Heistad DD. Vascular remodeling. Hypertension. 1996; 28: 505–506.
- ↵Dickhout JG, Lee RM. Increased medial smooth muscle cell length is responsible for vascular hypertrophy in young hypertensive rats. Am J Physiol. 2000; 279: H2085–H2094.
- ↵Mulvany MJ, Baandrup U, Gundersen HJ. Evidence for hyperplasia in mesenteric resistance vessels of spontaneously hypertensive rats using a three-dimensional dissector. Circ Res. 1985; 57: 794–800.
- ↵Fujii K, Tominaga M, Ohmori S, Kobayashi K, Takata Y, Fujishima M. Decreased endothelium-dependent hyperpolarization to acetylcholine in smooth muscle of the mesenteric artery of spontaneously hypertensive rats. Circ Res. 1992; 70: 660–669.
- ↵Jack JJB, Noble D, Tsien RW. Electric Current Flow in Excitable Cells. Oxford, England: Oxford University Press; 1988.
- ↵Cassis LA, Stitzel RE, Head RJ. Hypernoradrenergic innervation of the caudal artery of the spontaneously hypertensive rat: an influence upon neuroeffector mechanisms. J Pharmacol Exp Ther. 1985; 234: 792–803.
- ↵Mulvany MJ. Small artery remodeling and significance in the development of hypertension. News Physiol Sci. 2002; 17: 105–109.
- ↵Chauhan S, Rahman A, Nilsson H, Clapp L, MacAllister R, Ahluwalia A. NO contributes to EDHF-like responses in rat small arteries: a role for NO stores. Cardiovasc Res. 2003; 57: 207–216.
- ↵Little TL, Xia J, Duling BR. Dye tracers define differential endothelial and smooth muscle coupling patterns within the arteriolar wall. Circ Res. 1995; 76: 498–504.
- ↵Qu Y, Dahl G. Function of the voltage gate of gap junction channels: selective exclusion of molecules. Proc Natl Acad Sci U S A. 2002; 99: 697–702.