INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE ARTERIOLAR WALL consists of a layer of endothelial cells (EC) surrounded by a layer of smooth muscle cells (SMC) (13, 16, 23). In cheek pouch arterioles, the cells within each of these layers are coupled to each other via gap junctions, which enable the rapid transmission of electrical signals from cell to cell (21, 23, 25). Thus focal activation of muscarinic and -adrenergic receptors evokes vasodilation (hyperpolarization) and vasoconstriction (depolarization), respectively, which then travel rapidly and bidirectionally along the arteriole from the site of stimulation (23). However, it has not been resolved whether EC, SMC, or both cell layers are critical to the initiation and conduction of vasodilation and vasoconstriction.

The endothelial layer has been favored as the pathway for conduction based on the longitudinal orientation of EC along the arteriolar axis, pronounced dye coupling between EC, and the extent and location of immunolabeling for gap junctions (10, 13, 20, 23). In contrast, intracellular recordings have indicated that PE-induced depolarization can travel along the smooth muscle layer, independent of EC (23). Although ACh evokes hyperpolarization along SMC as well as EC (23), the nature of communication between these cell layers has not been established. Evidence suggests direct coupling between EC and SMC through gap junctions (4, 13, 16), as well as the generation of endothelium-derived hyperpolarizing factor(s) (1, 3). Thus multiple pathways are apparent for cell-to-cell communication along the arteriolar wall.

A fundamental limitation of previous studies of conduction is that the cellular components of the arteriolar wall have remained intact. In the present study, we have refined the light-dye technique (11, 17) to selectively damage EC or SMC within a defined segment of the conduction pathway along cheek pouch arterioles in vivo. We tested the hypotheses that conduction of vasodilation can be disrupted by damage to EC and that conduction of vasoconstriction can be disrupted by damage to SMC.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Hamster Cheek Pouch Preparation

With the use of a stereo microscope, the cheek pouch was everted, opened, and pinned onto a Plexiglas board. The superficial connective tissue was removed to allow clear observation of microvessels. The preparation was superfused continuously with a bicarbonate-buffered physiological saline solution (PSS; pH 7.35-7.45, 37°C) of the following composition (in mM): 137.0 NaCl, 4.7 KCl, 1.2 MgSO₄, 2.0 CaCl₂, and 18.0 NaHCO₃; the PSS was equilibrated with 95% N₂-5% CO₂. These reagents were obtained from J. T. Baker (Phillipsburg, NJ) or Sigma (St. Louis, MO) and dissolved in deionized water (dH₂O). Other reagents (below) were from Sigma unless stated.

Upon completion of surgical procedures, the preparation was secured onto a fixed stage and equilibrated for at least 30 min. The intravital microscope (modified 20T, Zeiss; Thornwood, NY) was mounted onto a movable platform, which enabled defined regions along an arteriole to be observed without disturbing the preparation or the placement of micropipettes (described below). The cheek pouch was transilluminated through a heat filter (700 nm cutoff) with a 100-W halogen lamp [ACH/APL condenser, numerical aperture (NA) = 0.32; Zeiss]. The image was acquired using a Leitz UM 32 objective (NA = 0.20) coupled to a video camera (NC-70X, Dage-MTI; Michigan City, IN); total magnification on the monitor face (PVM 133; Sony; Japan) was ×700. Internal diameter was measured (resolution 2 µm) from the edges of the vessel lumen using a video caliper (modified 321; Colorado Video; Boulder, CO) calibrated using a stage micrometer (100 × 0.01 = 1 mm).

After equilibration, arteriolar reactivity (second- and third-order branches) was determined in each preparation by evaluating diameter responses to topical sodium nitroprusside (SNP, 10 µM) added to the superfusate and response to raising superfusate PO₂ from 0 to 10% (with 5% CO₂-85% N₂). In all vessels studied, topical SNP increased diameter by ~80% above rest (to ~70 µm; taken as maximal diameter), and 10% oxygen reduced resting diameter by ~25%.

Stimulus delivery. Vasomotor stimuli were applied through borosilicate glass micropipettes (GC120F-10, Warner; Hamden, CT) pulled (P-87, Sutter; Novato, CA) to a tip internal diameter (ID) of ~1 µm for microiontophoresis or 2-3 µm for pressure microejection. Solutions were filtered (0.2 µm Acrodisc, Gelman Sciences; Ann Arbor, MI) and backfilled into micropipettes. ACh (1.0 M) and PE (0.5 M) were used as the criterion stimuli (23). These were dissolved in dH₂O and delivered via a microiontophoresis programmer (260, World Precision Instruments, Sarasota; FL) as a current pulse (500 nA, 500 ms); retain current was 200 nA. Bradykinin (an EC-dependent vasodilator; 1 mM in dH₂O) and sodium nitroprusside (SNP; an EC-independent vasodilator; 10 µM in PSS) served as control stimuli (see RESULTS) and were delivered using pressure microejection (PLI 100; Medical Systems; Greenvale, NY): 8-10 psi, 300-500 ms for bradykinin, and 500-1,000 ms for SNP. The micropipette was positioned with its tip perpendicular and adjacent to the arteriolar wall using a micromanipulator. To control for the passage of current, NaCl (1 M) was microiontophoresed as above and was without effect. Pressure microejection of PSS was also without effect, whereas that of dH₂O produced a small (3-5 µm) transient vasodilation. Vasomotor responses were abolished with micropipette tips positioned >100 µm from the vessel wall, demonstrating that the direct effects of stimuli were highly localized.

Light-dye treatment. Treatment with light-dye relies on production of oxygen free radicals in physiological solution when a fluorochrome is illuminated at the appropriate wavelength (14). Previous studies that have used the light-dye approach to damage EC in microvessels have injected a bolus of the salt (sodium) form of fluorescein into the peripheral circulation of rodents and illuminated arterioles for varying time periods (11, 17). Whereas this approach damages the endothelial layer, the hydrophilic nature of the dye salt enables it to readily escape the vascular compartment and simultaneously affect SMC (I. S. Bartlett and S. S. Segal, data not shown). To minimize such nonselective actions in the present experiments, FITC was conjugated to BSA, resulting in a single protein band of ~60 kDa, as determined with gel electrophoresis. We reasoned that with FITC-BSA retained within (or excluded from) the vascular compartment, the effects of illumination on EC (or SMC) should be more specific than when illuminating with the dye salt. We determined the molar substitution ratio of FITC to albumin to be ~3:1, consistent with previous determinations (8).

Experimental Protocols

Baseline responses. Second- and third-order arterioles suitable for study were located away from the edge of the preparation and had an unbranched segment ~1 mm in length and without an adjacent venule. A micropipette was positioned perpendicular to the vessel axis at the "distal" end (with respect to blood flow) of the segment with its tip adjacent to the vessel wall (Fig. 1A). A stimulus was delivered, and changes in diameter were recorded locally at the site of stimulation. The stimulus was then repeated at the distal site while observing at the "central" site located 500 µm upstream and again while observing at the "proximal" site located 1,000 µm upstream from the distal site. At least 2 min were allowed between stimuli to ensure full recovery to resting diameter. The micropipette was then moved to the central and proximal sites, and the respective local responses to the stimulus were recorded. Once control data for ACh and PE had been obtained at each site, the central segment was subjected to light-dye treatment to disrupt cells along the conduction pathway. At the central site, the illuminated segment was ~300 µm long (Fig. 1A). Light-dye treatment consisted of one of the following protocols.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   A: diagram illustrating sites along arteriole defined relative to direction of blood flow (Flow). Microiontophoresis micropipette containing ACh or phenylephrine (PE) positioned at distal site. Central site is centered 500 µm upstream from distal site and is contained within illumination area (shaded circle). Proximal site is 1,000 µm upstream from distal site. B: original recordings showing vasomotor responses observed at each site along an arteriole in response to microiontophoresis (500 nA for 500 ms; arrows) of ACh (top trace) or PE (bottom trace) at distal site.

Protocol 1: Endothelial damage with luminal light-dye treatment. In cheek pouch arterioles, the endothelial layer forms a barrier that restricts the passage of hydrophilic molecules from the blood (12) and thereby prevents the contact of FITC-BSA with SMC. Thus the damage produced by illumination of luminal FITC-BSA should initially be confined to the endothelial layer and be reflected by the selective loss of responsiveness to an endothelium-dependent vasodilator [e.g., ACh (7, 11, 17)].

A bolus of FITC-BSA (1 ml/kg body wt of a 3% wt:vol solution) was injected into the hamster via the femoral vein and equilibrated for 10 min. The central segment of the arteriole was viewed through a ×40 water immersion objective (NA = 0.75; Zeiss) and epi-illuminated using a 75-W xenon lamp (excitation: 450-490 nm; emission: >520 nm; Zeiss filter set 487709). Each period of illumination lasted 3 min and was followed by 5 min of recovery; the response to ACh release at the illuminated (central) site was then reevaluated. Periods of illumination (each for 3 min plus 5 min recovery) were repeated until vasodilation in response to ACh was abolished (Fig. 2). The response to PE at the central site was then reevaluated to assess the integrity of surrounding SMC.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Effect of illuminating central segment of arterioles containing FITC-BSA on local vasomotor responses (change in diameter, µm) to microiontophoresis of ACh (open bars, n = 9) and PE (hatched bars, n = 5) at central site (see Fig. 1A). Initial control responses to PE and ACh shown at left. Successive periods of illumination (3 min each + 5 min recovery) caused progressive loss of response to ACh until it was abolished (four periods; i.e., 12 min total illumination). Responses to PE were not different from control at this time. Continued illumination caused progressive loss of response to PE until it too was abolished (8 periods; i.e., 24 min total illumination).

Protocol 2: Smooth muscle damage with luminal light-dye treatment. As the integrity of EC is disrupted, light-dye damage initiated from within the vessel lumen should spread into the surrounding SMC. To evaluate this effect, the loss of response to PE was taken as an index of smooth muscle damage. Thus, once the response to application of ACh at the illuminated site had been abolished (as in protocol 1), treatment periods were continued until the response to application of PE at the illuminated site was also abolished (Fig. 2).

Protocol 3: Smooth muscle damage with abluminal light-dye treatment. The FITC-BSA solution was perifused over the central segment using a micropipette (10 µm tip ID, 2-5 psi) while it was illuminated (as described in protocol 2). On the basis of preliminary studies, each period of illumination with abluminal FITC-BSA lasted 1 min and was followed by 5 min of recovery. The shorter period of illumination (relative to luminal treatment) is attributable to the greater dye concentration at the site of illumination, because it was not first diluted in the intravascular compartment. The local response to PE was then reassessed at the illuminated site. Illumination periods continued until responses to PE had been nearly abolished. In two experiments, the arteriole lost tone following light-dye treatment. To counteract this dilation, the superfusate oxygen concentration was increased to 10% to restore tone (21). Results were not different from arterioles that did not lose tone and were therefore pooled. We have previously shown that elevating superfusate oxygen to increase tone is without effect on conduction (19, 21).

Illumination in absence of dye. Control experiments were performed to evaluate the effect of epi-illumination in the absence of dye (with filter set in place) on local and conducted vasomotor responses. Arterioles were illuminated for periods that corresponded with the abolition of direct responses to ACh and PE in protocols 1 and 2, respectively.

Localization and effects of light-dye damage on conduction and cellular integrity. After each of the above protocols, the localization of functional damage was determined by the local responses to ACh and PE delivered at the sites proximal and distal to the illuminated (central) site. The ability of responses initiated by ACh and PE at the distal site to conduct into and through the illuminated segment was then reevaluated, as was the ability of a stimulus delivered at the illuminated site to evoke a conducted response at the proximal site.

The integrity of EC and of SMC was evaluated qualitatively with Hoechst 33342 or propidium iodide (PI, Molecular Probes, Eugene, OR); these fluorescent dyes bind to DNA and reveal the cell nucleus. After some experiments, the cheek pouch was superfused for 20 min with either dye (1 µM in PSS). Whereas PI is excluded from healthy cells and becomes permeable following membrane damage (9, 22), Hoechst 33342 is membrane permeable and labels all cells when added to the superfusate (I. S. Bartlett and S. S. Segal, data not shown). Therefore, after the loss of the ACh response (Fig. 2), an overdose of pentobarbital sodium was given, and the left carotid artery was cannulated (PE-50) in the direction of blood flow. The artery was perfused for ~10 min with control PSS to flush the vasculature of the pouch (n = 4), then for ~10 min with Hoechst 33342 (1 µM in PSS), and for a final ~10 min with control PSS; effluent exited from a hole in the jugular vein. Dye labeling was evaluated using the ×40 immersion objective and appropriate filters (Zeiss). Photomicrographs were obtained using a 35-mm camera (Contax 167MT containing Kodak T160 slide film) coupled to the intravital microscope; slides were scanned (35T Plus, Microtek; Redondo Beach, CA) to digitize these images. Alternatively, dye labeling was observed using a Nikon E800 microscope (magnification, NA: ×20, 0.5; ×40, 0.75; Plan Fluor objectives) with appropriate filters (Chroma; Brattleboro, VT); epifluorescent illumination was provided with a 100-W mercury lamp. These images were digitized using a cooled charge-coupled device camera (SPOT, Diagnostic Instruments; Sterling Heights, MI) coupled to a frame grabber (Matrox Millenium II; Quebec, Canada) housed in a Pentium-based personal computer. Photomicrographs were arranged for presentation using Adobe Photoshop (version 4.01, Adobe Systems; San Jose, CA).

Data Acquisition and Analysis

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Microiontophoresis of ACh or PE onto cheek pouch arterioles evoked vasodilation or vasoconstriction, respectively, that conducted along the arteriolar wall (Fig. 1B).

Protocol 1: Endothelial Damage With Luminal Light-Dye Treatment

Strikingly, the selective loss of response to ACh at the central site had no effect on conducted responses to ACh that were initiated at the distal site and observed either at the central site or the proximal site (Fig. 3). Furthermore, when initiated by PE at the central site, conducted vasoconstriction at the proximal site was not different from control (diameter change: 11 ± 2 µm). To distinguish whether the effects of light-dye damage were specific to the ACh signaling pathway or caused a more general dysfunction of EC, the local response to bradykinin at the central site was also evaluated. The vasodilation observed before light-dye treatment was converted to vasoconstriction following loss of the ACh response (Fig. 4). At the same time, vasodilation to SNP was not significantly affected (Fig. 4).

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Endothelial cell (EC) damage with luminal light-dye treatment. A: responses of arterioles to ACh (n = 5) and PE (n = 5) after loss of local response to ACh in central (illuminated) segment (refer to Figs. 1 and 2). ACh and PE were applied at the distal site; diameter changes observed at distal, central, and proximal sites. B: original recordings showing vasomotor responses to ACh and PE observed at distal site (local) and at central and proximal sites (conducted) after treatment. Note lack of effect of EC damage on conducted responses to ACh or to PE at both central and proximal sites. Arrows indicate delivery of stimuli.

View larger version (17K): [in this window] [in a new window]   Fig. 4.   Local responses of arterioles to PE, bradykinin (BK), and sodium nitroprusside (SNP) at central site before and after loss of response to ACh induced by luminal light-dye treatment (n = 5 each). ACh, PE, BK, and SNP were applied using micropipettes as described in METHODS. * P < 0.05; Student's paired t-tests.

In control arteriole segments that contained FITC-BSA, but were not illuminated, perfusion with Hoechst-33342 dye intensely stained EC but not SMC (Fig. 5). After loss of response to ACh in the illuminated segment, the dye labeled SMC as intensely as EC (Fig. 5).

View larger version (73K): [in this window] [in a new window]   Fig. 5.   Fluorescence micrographs of labeling with Hoechst 33342 perfused through lumen of an arteriole following loss of response to ACh at site of luminal light-dye treatment. A: control (right) and illuminated (left) regions along arteriole. B: higher magnification of illuminated region. Note labeling of smooth muscle cells (SMC) (nuclei oriented perpendicular to vessel axis) as well as endothelial cells (nuclei parallel to vessel axis). Scale bars = 50 µm.

Protocol 2: Smooth Muscle Damage With Luminal Light-Dye Treatment

View larger version (23K): [in this window] [in a new window]   Fig. 6.   Smooth muscle damage after endothelial damage with luminal light-dye treatment. A: responses of arterioles to ACh (n = 5) and PE (n = 5) following loss of response to ACh and PE in central segment (see Fig. 2). ACh and PE were applied at distal site; diameter changes observed at distal, central, and proximal sites (see Fig. 1A). B: original recordings showing vasomotor responses to ACh and PE observed at distal site (local) and at central and proximal sites (conducted) after treatment. Note loss of conducted responses at central and proximal sites after EC and SMC damage with irreversible constriction of the central segment. Arrows indicate delivery of stimuli. * P < 0.05; Student's paired t-tests.

View larger version (46K): [in this window] [in a new window]   Fig. 7.   Propidium iodide (PI) labeling of arteriolar EC and SMC after light-dye damage. Photomicrographs show illuminated segment of arteriole following loss of response to PE. A and B correspond to condition in Fig. 6 with luminal light-dye treatment. A: brightfield image showing irreversible constriction of illuminated segment. Arrows indicate leukocytes adhering to damaged EC. B: fluorescent image of A. Endothelial cell nuclei at the sites indicated in A are brightly labeled with PI. C: fluorescent image following abluminal light-dye treatment corresponding to condition in Fig. 8. Note strong labeling of SMC nuclei (arrowheads) and absence of EC labeling. Other labeled nuclei include mast cells along arteriolar wall and parenchymal cells in illuminated area (see Fig. 1A) during FITC-BSA perifusion. Scale bars = 50 µm.

Protocol 3: Smooth Muscle Damage With Abluminal Light-Dye Treatment

View larger version (20K): [in this window] [in a new window]   Fig. 8.   Smooth muscle damage with abluminal light-dye treatment. A: responses of arterioles to ACh (n = 5) and PE (n = 5) following loss of response to PE in central segment. ACh and PE were applied at distal site; diameter changes observed at distal, central, and proximal sites (refer to Fig. 1A). B: original recordings showing vasomotor responses to ACh and PE observed at distal site (local) and at central and proximal sites (conducted) after treatment. Note loss of conducted responses to PE but not to ACh following SMC damage and constriction of the central segment. Arrows indicate delivery of stimuli. At distal site the increase in response to PE and diminition in response to ACh are attributable to an increase in resting diameter (from 31 ± 4 to 36 ± 5 µm; not significant). * P < 0.05; Student's paired t-tests.

Effect of Illumination in Absence of Dye

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We have refined the light-dye technique to selectively damage endothelium or smooth muscle to resolve the role of each cell layer in the initiation and conduction of vasodilation and vasoconstriction along arterioles engaged in blood flow control. When a fluorochrome is excited in the presence of oxygen, free radicals are generated, which attack membrane components of nearby cells (2, 5). We reasoned that as the conduction of vasomotor responses in cheek pouch arterioles reflects electrical signaling along smooth muscle and endothelial layers (23-25), membrane damage should disrupt conduction through the damaged cells. Restricting a fluorochrome to the luminal or abluminal compartment enabled selective and site-specific disruption of EC or SMC, respectively, in a controlled, exposure-dependent manner. The specificity of light-dye damage was determined by the localized and selective loss of functional responsiveness to ACh and bradykinin (endothelial damage) and to PE and SNP (smooth muscle damage); labeling with nuclear dyes provided an independent, qualitative index of damaged cells. Our findings indicate that the smooth muscle layer alone conducts vasoconstriction in response to PE, whereas both endothelial and smooth muscle layers can conduct the signal for vasodilation in response to ACh.

The endothelial layer has been implicated as the critical pathway for conduction (10, 20). This led to our hypothesis that damage to EC would disrupt conducted vasodilation in response to ACh. However, neither the conduction of vasodilation nor of vasoconstriction was affected by endothelial damage, despite losing the ability to initiate vasodilation from damaged EC. This finding uniquely demonstrates that the smooth muscle layer can conduct vasodilation when initiated by ACh from a region in which the integrity of EC remains intact. In contrast (and consistent with our original hypothesis), when SMC were damaged, vasoconstriction could no longer be conducted through the damaged region. These findings support the conclusion that the smooth muscle layer alone provides the signaling pathway for conducted vasoconstriction in response to PE (23). Nevertheless, the initiation and conduction of vasodilation were unaffected along the region in which SMC were damaged. With hyperpolarization to ACh evoked in SMC as well as EC (23), the present data support the hypothesis that each cell layer can serve as a conduction pathway along arterioles of the hamster cheek pouch.

We were concerned that the effects of light-dye treatment could be explained by the loss of a specific signaling pathway (18) rather than by a more generalized membrane disruption and loss of cellular integrity. To investigate this possibility, bradykinin was tested at the site of illumination. Before light-dye treatment, bradykinin caused a large dilation (Fig. 4), which was preceded by a small (~4 µm), transient vasoconstriction. After luminal light-dye treatment until the ACh response was abolished (Fig. 2), the vasomotor response to bradykinin was converted to a substantial vasoconstriction (Fig. 4). This result is consistent with the presence of bradykinin receptors on SMC as well as EC, such that activation of EC leads to vasodilation and activation of SMC to vasoconstriction (15). Whereas vasodilation (to ACh and bradykinin) predominated in the intact vessel, vasoconstriction was unimpaired after endothelial damage. Therefore, our finding that at least two distinct endothelium-dependent signaling pathways were disrupted supports the conclusion that EC were indeed damaged by luminal light-dye treatment.

In previous studies attempting to disrupt EC within illuminated arterioles, rodents were given an intravascular bolus of sodium fluorescein (11, 17). This salt form of the fluorochrome is hydrophilic and rapidly escapes from the vascular compartment. In our control experiments documenting this behavior (data not shown), evidence for a lack of specificity included the simultaneous impairment of conducted vasoconstriction and conducted vasodilation. Through delivering the fluorochrome conjugated to a large intravascular protein, illumination first caused exposure-dependent loss of endothelium-dependent responses, which was followed by the abolition of smooth muscle responses (Fig. 2). The irreversible vasoconstriction at this time (Fig. 7A) may be explained by uncontrolled Ca²⁺ entry into damaged SMC. With perifusion of FITC-BSA for abluminal light-dye treatment, the irreversible constriction occurred first, and the integrity of conducted responses to ACh indicated that the endothelial layer remained intact. If abluminal treatment was continued, however, conducted vasodilation could no longer be initiated from at the illuminated site. Collectively, these observations indicate that progressive luminal light-dye treatment damaged EC and then SMC, whereas progressive abluminal treatment damaged SMC and (if continued) then damaged EC. Moreover, these events occurred in a highly reproducible manner.

The illumination times employed when using FITC-BSA were longer (e.g., up to 24 min in protocol 2) than have been used in previous studies when using the unconjugated dye [2-10 min (11, 17)]. Such lengthy exposure to illumination might be expected to have effects on arterioles (6). Controls performed to account for such actions indicated that illumination in the absence of dye had no effect on resting diameter until after the fifth period (i.e., 15th min) of illumination, when arterioles began to lose tone in the illuminated segment. Whereas this local vasodilation was pronounced following the eighth period (24th min) of illumination, both dilation (to ACh) and constriction (to PE) could still conduct through (and be initiated from) the dilated segment. These observations demonstrate that vasomotor tone can be lost without interruption of either the conduction pathway or the ability to initiate conducted responses. Thus the signaling pathways for both ACh and PE remained intact when arterioles were illuminated in the absence of fluorochrome.

In summary, we have refined the light-dye technique to selectively damage endothelial or smooth muscle layers that initiate and conduct vasomotor signals along arterioles in vivo. By conjugating FITC to BSA, light-dye damage was constrained to the luminal (endothelial) or abluminal (smooth muscle) layer within the field of illumination. At the site of endothelial damage, vasodilation in response to ACh and bradykinin were eliminated, along with the ability to initiate conducted vasodilation. Nevertheless, neither the conduction of vasodilation nor of vasoconstriction was impaired through the segment with damaged EC. At the site of smooth muscle damage, the initiation and conduction of vasoconstriction in response to PE were abolished with no effect on the initiation or conduction of vasodilation. We therefore suggest that, for hamster cheek pouch arterioles in vivo, vasoconstriction can be initiated and conducted within the smooth layer, independent of endothelial integrity. In turn, whereas intact EC are necessary for ACh to initiate vasodilation, the smooth muscle layer can conduct vasodilation independent of endothelial integrity.