Vasodilation of intramuscular arterioles under shear stress in dystrophin-deficient skeletal muscle is impaired through decreased nNOS expression

Summary

Duchenne muscular dystrophy (DMD) is a lethal X-linked disorder of striated muscle caused by the absence of dystrophin. Recently, impairment of vascular dilation under shear stress has been found in DMD, but the underlying molecular mechanism is not fully understood. Moreover, dilation of intramuscular arterioles, which may be a key to the molecular pathogenesis, has not been addressed yet. We examined dilation of arterioles in the mouse cremaster muscle under shear stress due to ligation. The vasodilation was significantly impaired in dystrophin-deficient mdx mice as well as in neuronal nitric oxide synthase (nNOS)-deficient mice; however, neither endothelial NOS-deficient mice nor α1-syntrophin-deficient mice showed any difference in vasodilation from control mice. These results indicate that nNOS is the main supplier of nitric oxide in shear stress-induced vasodilation in skeletal muscle, but that the sarcolemmal localization of nNOS is not indispensable for the function. In contrast, the response to acetylcholine or sodium nitroprusside was not impaired in mdx or nNOS-deficient mice, suggesting that pharmacological treatment using a vasoactive agent may ameliorate skeletal and cardiac muscle symptoms of DMD.

Introduction

Nitric oxide (NO) is a vasoactive agent generated by nitric oxide synthase (NOS). Neuronal NOS (nNOS) is highly expressed in skeletal muscle compared with endothelial NOS (eNOS) and inducible NOS (iNOS). nNOS is anchored by α1-syntrophin, a member of the dystrophin-glycoprotein complex (DGC), at the sarcolemma in skeletal muscle (1–6). Dystrophin is a cytoskeletal protein, and its absence together with the secondary loss of DGC from the sarcolemma is responsible for Duchenne muscular dystrophy (DMD), a severe muscle disease characterized by progressive skeletal muscle degeneration complicated with cardiomyopathy (5). nNOS expression is greatly reduced at the mRNA level in dystrophin-deficient muscle (2). Moreover, the attenuation of α-adrenergic vasoconstriction is impaired in contracting dystrophin-deficient muscle, suggesting that nNOS has a specific role in protection from sympathetic vasoconstriction (7, 8). In addition, the localization of nNOS at the sarcolemma through α1-syntrophin is indispensable for the attenuation of α-adrenergic vasoconstriction during muscle contraction (9). Recently, Loufrani et al. showed that the carotid and mesenteric arteries of mdx mice, an animal model of DMD, do not dilate properly under shear stress, although they are dilated normally by treatment with either an NOS stimulator, such as acetylcholine (ACh), or an NO donor, such as sodium nitroprusside (SNP) (10). They concluded that the endothelial dystrophin plays an invaluable role in vasodilation under shear stress. In addition, the molecular background is not clearly understood, although flow-induced remodeling in arterial wall is deficient in mdx mice when stimulated by arterial ligation or hydralazine (11, 12). To clarify the role of nNOS in intramuscular arterioles in vivo, we studied vasodilation in the mouse cremaster muscle. We caused the modified parallel occlusion of arterioles by microsurgical nylon thread ligation (13–16). We enlisted the participation of DGC in shear-stress vasodilation by using mdx mice. We also determined the significance of the localization of nNOS at the sarcolemma by using α1-syntrophin knockout mice (α1syn^-/-). In addition, we used nNOS knockout (nNOS^-/-) and eNOS knockout (eNOS^-/-) mice to clarify which NOS is involved in vasodilation under shear stress.

Materials and methods

Animals

Mdx mice and their controls, C57Bl/10 mice (B10), α1syn^-/- mice generated in C57Bl/6 mice (B6), and their wild-type littermates (α1syn^+/+), aged 8-10 weeks were used (17). Eight- to 10-week-old nNOS^-/- and eNOS^-/- mice (B6 background) were supplied by the Jackson Laboratory. They were anesthetized by intraperitoneal injection of 1.2x10^-3 g carbamic acid ethyl ester per gram of body weight. At the end of the experiment, animals were sacrificed by an overdose of pentobarbital. All protocols were approved by the Institutional Animal Care and Use Committee of the National Institute of Neuroscience and were performed in compliance with the Guide for the Care and Use of Division of Laboratory Animal Resources.

Experimental Design

We mounted and fixed mouse on experimental stage under anesthesia and scrotum of each mouse was placed on a clear silicone dish as shown in Figure 1a. The cremaster muscle was exposed as described with minor modification (18), and was observed under an intravital microscope at 450 magnifications. The exposed cremaster muscle was deoxygenated by continuous superfusion (5 ml/min) of buffered Tyrode solution (34 ± 0.5 ˚C, pH 7.35-7.45) bubbled with 95% N₂ and 5% CO₂ gas. Captured microcirculatory images were converted to digital images by the computer and recorded by VTR (Fig. 1a). To calculate the shear stress, we used CapiFlow^® (IM-Capiflow, Kista, Sweden), a fully computerized system for the measurement of red blood cell velocity, as previously described (19, 20).

Figure 1.

Observation and measurement of dilation of intramuscular arterioles induced by drug treatment or by shear stress in mouse cremaster muscle. (a) Optical system consisting of a cool light unit, mirror box, 450X intravital microscope (MZFL3, Leica Microsystems, Heidelberg, Germany), cooled, color 3 charge-coupled device (CCD) camera, image capture unit (C5810, Hamamatsu Photonics K.K., Hamamatsu, Japan), computer (Apple Macintosh G4, Apple, Cupertino, California), and video cassette recorder (HR-STG300, Victor JVC, Yokohama, JAPAN). (b) Arterioles in the mouse cremaster muscle are classified as indicated. A1; first-order arterioles, A2; second-order arterioles, A3; third-order arterioles. Observation area was indicated by circle. (c) Measurements of arteriole diameter were performed 120-1000 µm from the point of divergence, and the observation point was decided in reference to the border of muscle fibers. (d). Points for measurements of tissue pO₂ during parallel occlusion in A3 area: A, around the main arteriole; B, around the ligation site; and C, the original point for measurements of dilation of arterioles.

Drug treatment

We first examined the vasodilatory response of third-order arterioles (A3; about 20 μm) in mouse cremaster muscles (Fig. 1b) (22). ACh or SNP was added to the buffer solution and applied directly to the muscle, based on previous reports with modification (23, 24). The vessel diameter was measured before and just after drug administration and the dilatory ratio was calculated as: diameter of arteriole after drug treatment/ before drug treatment. To determine adequate dose, ACh or SNP was exposed from its lower concentration to higher concentration. Before increasing dose, we waited for maximum ten minutes until no more dilatory effect was observed by previous dose. Papaverin was added at the final part of experiments to know the extent of maximum dilation of vessels. We also examined the effect of NO synthesis inhibition by adding N-omega-nitro-L-arginine methyl ester (L-NAME, 0.1 mmol/L) to the buffer from 10 minutes before ACh or SNP administration.

Shear stress

We used parallel occlusion method to increase the blood flow velocity in nonoccluded parallel arteriolar branches in vivo, based on the previous studies (13–16). The arteriole was ligated using 10-0 nylon thread with needle to produce shear stress (Fig. 1c) (13). The ligated portion and the measured point (A3) were remote enough from the branching point to avoid artifactual effects. The dilatory ratio for shear stress experiments was calculated as: diameter of arteriole after ligation/ before ligation. The dilatory ratio was also examined under indomethacin (1.0x10^-3, 5.0x10^-2, 1.0x10^-2 or 0.5 mmol/L) administration, when Prostaglandin I₂ (PGI₂) (1.0x10^-4 mmol/L) was added to the buffer solution or we induced vasodilation by parallel occlusion. L-NAME and indomethacin were supplied from 10 minutes before ligation. Without L-NAME treatment, shear stress-induced vasodilation was observed for a longer period as long as 20 minutes in 4 B10 and 4 mdx mice.

Measurements of partial pressure of oxygen (pO₂)

Observations of the microcirculation and in vivo partial pressure of oxygen (pO₂) measurements were made with a microscope and the oxygen-dependent quenching of phosphorescence decay technique, as previously described (21). We measured tissue pO₂ of B10 (n = 3) and mdx mice (n = 3) at three distinct points of cremaster muscles before and after ligation by the phosphorescence quenching method (Fig. 1d).

Histological analysis and immunohistochemistry

Ten-micrometer cryosections of cremaster muscles were prepared, air-dried, and stained with hematoxylin and eosin (H&E). Six-micrometer acetone-fixed cryosections were prepared, blocked with goat serum, and then incubated with primary antibodies, rabbit against nNOS (Zymed Laboratories) and rat against CD31/PECAM-1 (Southern Biotechnology Associates) at room air temperature. Alexa 488-labeled goat anti-rabbit IgG (H + L) (Molecular Probes) and Alexa 594-labeled goat anti-rat IgG (H + L) were used as the secondary antibody. The sections were viewed and photographed by a laser microscope, TCSSP^TM (Leica Microsystems).

Statistical analysis

Results were expressed as means ± standard error of the mean (SEM). Results were compared between mdx mice and B10, α1syn^-/- and α1syn^+/+ mice, and eNOS^-/- or nNOS^-/- mice and B6. The effect of L-NAME pretreatment was also evaluated. The significance of the differences between groups was determined by Mann-Whitney U test or ANOVA. Values of p < 0.05 were considered to be significant.

Results

Drug induced vasodilation

Maximum arteriolar dilation was determined for administration of ACh or SNP, and then compared with the dilation by the treatment with 1.0 mmol/L of Papaverine. The optimal dose of both ACh and SNP was 1.0 mmol/L for maximum dilatory ratio (Fig. 2) and the dose was used for subsequent examinations (Fig. 3).

Figure 2.

Responses of arterioles for vasodilatory agents, ACh and SNP in three B10. Graphs are showing dilatory ratio against various doses of ACh (a) or SNP (b), in reference to maximum dilation by treatment of 10^-3 M of Papaverin (Pap).

Figure 3.

Effects of vasodilative agents on dilation of mouse cremaster arterioles of B10 (black bar), mdx (white bar), B10 pretreated with L-NAME (black bar), mdx pretreated with L-NAME (white bar), α1syn^+/+ (black bar), α1syn^-/- (white bar), B6 (black bar), eNOS^-/- (white bar), and nNOS^-/- (black bar) mice. (a) After pretreatment with L-NAME, ACh-induced vasodilation was reduced both in B10 and in mdx mice. Values are indicated as mean ± SEM. Asterisk (*) shows statistical significance (p < 0.05). (b) Vasodilation induced by SNP was not statistically significant between the mice we examined.

The administration of ACh or SNP gave almost the same dilatory ratio between B10 (n = 7) and mdx mice (n = 7), α1syn^+/+ (n = 5) and α1syn^-/- (n = 5) and, eNOS^-/- (n = 4) or nNOS^-/- mice (n = 4) and B6 (n = 4) (Figs. 3a and 3b). This result does not conflict with the conclusion of a previous study using nNOS- and eNOS-deficient mice that expression of either nNOS or eNOS is sufficient for ACh-induced dilation (25). Pretreatment of L-NAME gave the same degree of inhibition in ACh-induced vasodilation in B10 (n = 5) and in mdx mice (n = 5), but did not significantly alter the dilatory ratio in SNP-induced vasodilation in these mice.

Shear stress-induced vasodilation

In contrast, shear stress-induced vasodilation was significantly impaired in mdx mice (n = 10) compared with that of B10 (n = 10) (Fig. 4a), and in addition, the calculated shear stresses were different (Table 1). Interestingly, although nNOS^-/- mice (n = 5) showed impaired vasodilation, eNOS^-/- mice (n = 5) did not show significant differences in the dilatory ratio when compared with that of B6 (n = 5), indicating that nNOS is the main supplier of NO in the shear stress-induced vasodilation of arterioles in skeletal muscle. On the other hand, α1syn^-/- mice (n = 5) did not show significant differences in the dilation compared with the control α1syn^+/+ mice (n = 5), suggesting that the intramuscular localization of nNOS at the sarcolemma is not critical for shear stress-induced vasodilation. After pre-treatment with L-NAME, shear stress-induced vasodilation was significantly decreased in B10 (n = 5).

Figure 4.

Effects of shear stress-induced dilation of mouse cremaster arterioles of B10 (black bar), mdx (white bar), B10 pretreated with L-NAME (black bar), mdx pretreated with L-NAME (white bar), α1syn^+/+ (black bar), α1syn^-/- (white bar), B6 (black bar), eNOS^-/- (white bar), and nNOS^-/- (black bar) mice. (a) mdx mice, B10 pretreated with L-NAME, mdx mice pretreated with L-NAME and nNOS^-/- mice showed impaired vasodilation under shear stress. (b) Extended observation of shear stress-induced vasodilation. The vessel diameter in B10 rapidly increased after vessel ligation and reached a stable level within 10 minutes (n = 4). The dilation of arterioles was severely impaired in mdx mice (n = 4). The difference between mdx mice and B10 was observed as long as 20 minutes after the ligation.

Table 1. Relationship of vasodilation and shear stress in mouse cremaster arterioles.

	Blood cell velocity (cm/s)		Diameter (µm)		Shear stress rate
	before ligation	after ligation	before ligation	after ligation
B10 (n = 3)	0.48 ± 0.05	0.67 ± 0.20	18.7 ± 0.9	24.2 ± 0.2	1.02 ± 0.16
mdx (n = 3)	0.41 ± 0.05	0.91 ± 0.23*	18.8 ± 0.7	19.8 ± 0.1*	2.02 ± 0.23*

Shear stress rates were calculated as (shear stress before ligation) / (shear stress after ligation). Values are expressed as mean ± S.E.M. * = p < 0.05

As shown in Figure 4b, under a longer observation of shear stress-induced vasodilation in the absence of L-NAME, the difference between mdx mice and B10 was still observed at least 20 minutes after the ligation.

PGI₂ induced vasodilation

There were significant differences between shear stress-induced and PGI₂-induced vasodilation in dilatory ratios against high concentrations of indomethacin in B10 (Fig. 5). These data indicated that high concentration of indomethacin treatment could completely antagonize PGI₂-induced vasodilation, but the treatment cannot completely inhibit shear stress-induced vasodilation.

Figure 5.

Under various dose of indomethacin, vasodilation was induced either by treatment of Prostaglandin I₂ (PGI₂) or by parallel occlusion (ligation) in B10 cremaster muscle arterioles (n = 5).

Alternation of pO₂ before and after ligation

There were no significant differences in tissue pO₂ levels between before and after ligation not only in B10 but also in mdx mice (Fig. 6).

Figure 6.

Histogram showing pO₂ at the observation points. There are no significant differences in alterations of tissue pO₂ during parallel occlusion.

Immunohistochemical observation of NOS expression

In H&E stained tissues, centrally nucleated fibers, which represent muscle regeneration, were observed in only mdx mice (Figs. 7a-e). The immunohistochemical analysis showed that nNOS was observed mainly at the sarcolemma rather than in the endothelium and vascular smooth muscle in B10 and eNOS^-/- mice (Figs. 7b and 7h). In α1syn^-/- mice, nNOS was not localized at the sarcolemma but remained in the cytoplasm (Fig. 7f), as previously reported (14, 26). Less nNOS was found in mdx mice, and it was not detected in nNOS^-/- mice (Figs. 7d and 7j).

Figure 7.

nNOS expression and localization in vascular endothelium and cremaster muscles of mice. H&E (a, c, e, g, and i) and double staining with nNOS (green) and PECAM-1 (red) antibodies (b, d, f, h, and i) of B10 (a, b), mdx (c, d), α1syn^-/- (e, f), eNOS^-/- (g, h), and nNOS^-/- (i, j) mice. Centrally located nuclei, a typical feature of regenerated muscle, are found only in mdx mice (c). In B10 and eNOS^-/- cremaster muscles, nNOS expression was observed at the sarcolemma. In contrast, the expression was greatly reduced or not detected in mdx or nNOS^-/- mice, respectively. Bar, 40 µm.

Discussion

Nitric oxide is one of the most important factors in shear stress-induced vasodilation especially by parallel occlusion method (10, 14, 27). Other factors, such as prostaglandins, were reported to contribute to shear stress-induced dilation in various models (15, 16, 28), but we showed that indomethacin, an inhibitor of prostaglandins, did not prevent the increase in diameter in shear stress condition. In addition, we concluded that the parallel occlusion method did not cause tissue hypoxia or acute ischemia. Thus, we demonstrated that dilation of arterioles in the mouse cremaster muscle under shear stress by the parallel occlusion method depends mainly on NO, especially that produced by nNOS. In particular, mdx and nNOS^-/- mice showed impaired vasodilation in parallel occlusion, whereas responses to ACh and SNP were unaltered. Decreased expression of nNOS in mdx skeletal muscle may be important as a cause of this finding.

It is intriguing to know the relationship between shear stress-induced vasodilation and the localization of nNOS. Koller et al. showed that shear stress-induced vasodilation of 80- to 156-μm arterioles was inhibited by removal of the endothelium or by addition of indomethacin in rat cremaster muscle, but they did not identify the responsible molecules of vascular dilation (29). In our study, nNOS expression was mainly found in the sarcolemma and less frequently in the endothelium or vascular smooth muscle, implying that skeletal muscle nNOS is possibly involved in dilation of intramuscular arterioles at the very end of the skeletal muscle circulation under shear stress. nNOS is anchored to the sarcolemma through α1-syntrophin. α1syn^-/- mice showed altered distribution of nNOS expression in cytoplasm, but showed no significant differences in shear stress-induced vasodilation between α1syn^-/- mice and α1syn^+/+ mice. Thus, the sarcolemmal localization of nNOS through expression of α1-syntrophin is not indispensable for vasodilation. However, how dystrophin or other molecules transduce mechanostress to soluble nNOS is unresolved (6). The defective vasodilation under shear stress due to nNOS deficiency in mdx mice might be related to its muscle degradation (14).

It is very interesting to note the amelioration of dystrophic phenotypes in nNOS transgenic mdx mice, although the localization of nNOS cannot have been improved (30). Decreased vasodilation just after muscle contraction has also been demonstrated in mdx skeletal muscle (31). Leinonen et al. found that capillary circulation in skeletal muscle was impaired in DMD (32), and deteriorated attenuation of α-adrenergic vasoconstriction during exercise may participate in this pathophysiology (7). Moreover, blood flow must be increased to accommodate the augmented metabolic demands of the muscle, not only in exercise. Intramuscular arterioles in mdx mice cannot afford to respond to the increased demands, and their failure may result in relative ischemia in the skeletal muscle and cardiac phenotypes of dystrophin deficiency. Asai et al. very recently showed that the functional ischemia in contraction-induced myofibers in mdx mice is due to nNOS deficiency and indicated that vasoactive drugs may ameliorate muscle damage (33). Even in dystrophin-deficient skeletal muscle, cholinergic vascular modulation was well preserved. Therefore, our study indicates that pharmacological treatment using a vasoactive agent is applicable to at least skeletal muscle symptoms in patients suffering from DMD.

In conclusion, we demonstrated that vasodilation of intramuscular arterioles under shear stress was impaired in dystrophin-deficient mdx mice. This impairment may be related to phenotypes of DMD, not only in skeletal muscle but also in cardiac muscle.