| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1Department of Physiology, New York Medical College, Valhalla, New York; and 2Xuzhou Medical College, Xuzhou, People's Republic of China
Submitted 25 October 2005 ; accepted in final form 5 April 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
51% and 58%, respectively. Administration of both inhibitors eliminated the dilation. In WT arterioles, inhibition of COX-2 did not significantly affect FID, whereas inhibition of COX-1 decreased the dilation by 57%. The residual portion of the response was abolished by additional administration of N
-nitro-L-arginine methyl ester. Western blot analysis indicated a comparable content of COX-1 protein in arterioles of WT and eNOS-KO mice. COX-2 protein, which was not detectable in arterioles of WT mice, was strongly expressed in arterioles of eNOS-KO mice, together with an upregulation of COX-2 gene expression. Immunohistochemical staining confirmed the presence of COX-2 in the endothelium of eNOS-KO arterioles. In conclusion, COX-2-derived PGs are the mediators responsible for maintenance of FID in arterioles of eNOS-deficient mice.
nitric oxide; endothelium
Shear stress stimulates the release of PGs, as well as NO, through multiple pathways, one of which involves upregulation of cyclooxygenase (COX) and eNOS gene expression, respectively (5, 7, 9, 25). The PG biosynthetic pathway is initiated by activation of phospholipase A2, which is responsible for shear stress-induced, as well as agonist-induced, arachidonic acid release from membrane phospholipids (21, 25). COX-1 and COX-2 act to convert arachidonic acid to PGH2, which is subsequently converted to a variety of eicosanoids, depending on the downstream enzymatic machinery present in a particular cell type.
Whether there is "cross talk" between the NOS and COX pathways remains open to question. The evidence supporting a negative relation between NO and PG synthesis is provided by studies that indicate an enhanced production of PGs as a consequence of inhibition of NO synthesis (3, 25, 26). In contrast, other studies demonstrated that NO stimulates PG synthesis in endothelial cells and macrophages via a mechanism involving a reaction with the heme component that binds to the active site of COX enzymes (20, 27). Our own data, from study of shear stress-mediated regulation of arteriolar tone, seem to support a compensatory upregulation of PG synthesis when NO-mediated regulation is impaired (31, 34). In the present study, we aimed to identify the specific isoform of COX that is responsible for the maintenance of flow- and shear stress-induced dilation in skeletal muscle arterioles of eNOS-KO mice. In addition, we attempted to clarify the molecular basis of the upregulation of COX-2, as well as the site of its formation.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Heterozygous eNOS (+/–) mice, originally developed by Shesely et al. (28), were interbred in the Department of Comparative Medicine, New York Medical College, to generate eNOS wild-type (WT, +/+) and homozygous mutant (eNOS-KO, –/–) mice. The mice were killed by inhalation of 100% CO2 at 22–24 wk of age. First-order gracilis muscle arterioles were isolated and cannulated in a vessel chamber (13, 31). All protocols were approved by the Institutional Animal Care and Use Committee of New York Medical College and conform to the National Institutes of Health and American Physiological Society guidelines for the use and care of laboratory animals.
Experimental Procedures
Flow-induced dilation. Changes in arteriolar diameter response to increases in perfusate flow (0–10 µl/min in 2 µl/min steps) were studied at 80 mmHg perfusion pressure.
In the first protocol, the role of specific isoforms of COX (COX-1 and COX-2) in the mediation of flow-induced dilation was studied in arterioles of both strains of mice. After flow-diameter curves were obtained in the control condition, vessels were subjected to valeryl salicylate (VS, 3 mM) or NS-398 (10 µM), specific inhibitors of COX-1 and COX-2 (1, 35), respectively, for 30 min, and then flow-induced dilation was reassessed.
In the second series of experiments, flow-induced dilation was obtained before and after exposure of vessels of both strains of mice to the simultaneous presence of COX-1 and COX-2 inhibitors. In arterioles of WT mice, flow-induced dilation was also assessed in the control condition and after incubation of the vessels with 100 µM N-nitro-L-arginine methyl ester (L-NAME) for 30 min.
In separate experiments, flow-induced dilation (10 µl/min) was assessed in the control condition and after inhibition of the PGI2 receptor with 10–6 M CAY-10441. The efficacy of the inhibition was assessed by vasodilator responses to iloprost (10–7 M), a stable analog of PGI2, before and after administration of CAY-10441. Dilation to 10–6 M adenosine was used to assess smooth muscle-dependent responses before and after administration of the inhibitors.
Passive diameter. At the conclusion of each experiment, the suffusion solution was changed to a Ca2+-free solution containing 1 mM EGTA. The vessels were incubated for 10 min to reach maximal diameter (passive diameter) at 80 mmHg perfusion pressure.
Western blotting.
Gracilis arteries (3 mm long) were solubilized in Laemmli buffer, sonicated twice for 1 min each, and boiled for 1 min to denature the proteins. Samples were separated on 10% SDS-polyacrylamide gels. Membranes were probed with polyclonal antibodies of COX-1 and COX-2 (Cayman Chemical) overnight at 4°C. The secondary antibody was conjugated to horseradish peroxidase according to the Amersham ECL-Plus protocol. 
-Actin was used to normalize for loading variations.
Real-time RT-PCR.
Gracilis muscle arterioles were pulverized in liquid nitrogen, and total RNA was extracted using TRI Reagent (Sigma). RNA (1 µg) was reverse transcribed with random primers (Promega) and SuperScript III (Invitrogen). cDNA was utilized for quantitative real-time RT-PCR using the LightCycler with the FastStart DNA Master SYBR Green I kit (Roche Diagnostics). Primers for COX-1 and COX-2 (Sigma Genosys) were designed as follows: 5'-TGGCCAAGGTCTACCCCG-3' (sense) and 5'-CTCTGTACCCAAAGACTGCC-3' (antisense) for COX-1 and 5'-AGTCTCTCAATGAGTACCGGAAA-3' (sense) and 5'-AAGTTCTTCAAATGATGTGTACGG-3' (antisense) for COX-2. A relative quantification method (
Ct, where Ct is cycle threshold) (22) was used to evaluate the expression of each gene in vessels of KO and WT mice. RT-PCR of GAPDH was used as an internal control and for normalization of all data.
Immunohistochemistry. Gracilis muscle was perfused (80 mmHg) and fixed with freshly prepared 4% paraformaldehyde. Paraffin sections (5 µm) were probed with anti-COX-1 and anti-COX-2 antibodies (Cayman Chemical) and further processed with the Vectastain Elite ABC staining method using 3,3'-diaminobenzidine as a peroxidase substrate, according to the manufacturer's protocol (Vector), and counterstained with hematoxylin. Control staining was performed by incubation of COX-1 or COX-2 blocking peptides with corresponding antibodies for 1 h before application to the slides or by elimination of the primary antibodies from the staining process.
Chemicals
VS, NS-398, iloprost, and CAY-10441 were obtained from Cayman Chemical. VS was dissolved in physiological salt solution directly with a brief sonication. NS-398 was dissolved in DMSO at 50 mM and further diluted with physiological salt solution. All other chemicals were obtained from Sigma (St. Louis, MO).
Calculations and Statistics
Flow-induced dilation was normalized by the passive diameter and expressed as percentage of passive diameter or as change in diameter as percentage of passive diameter [(diameter obtained after flow-induced dilation – basal diameter)/passive diameter·100%] to eliminate the influence of the difference in vessel size on data comparison between the two groups. The difference in flow-diameter curves was compared by repeated-measures ANOVA, followed by the Tukey-Kramer multiple-comparison test. The ratio of band densities of Western blots of COX-1 or COX-2 to -actin was compared by Student's t-test. P < 0.05 was considered significant. Values are means ± SE; n indicates the number of mice.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Arteriolar active and passive diameters of eNOS-KO mice (69 ± 3 and 114 ± 4 µm, respectively) were significantly smaller than those of WT mice (79 ± 4 and 127 ± 6 µm, respectively). As a result, the basal tone of arterioles, expressed as percentage of passive diameter, was similar in both strains of mice (60 ± 2 and 62 ± 1%). Arteriolar dilations to increases in perfusate flow were not significantly different in eNOS-KO and WT (at maximal flow rate) mice (77 ± 2 and 78 ± 1% of passive diameter, respectively). Also, calculated shear stress at maximal flow rate was similar (13–16 dyn/cm2) in vessels of the two strains of mice. The specific role of COX-1 and COX-2 in the mediation of PG-dependent, flow-induced dilations of arterioles of eNOS-KO mice was tested with VS and NS-398, respectively. Neither VS nor NS-398 affected significantly the basal tone and adenosine-induced dilations (not shown) in arterioles of either strain of mice. Each inhibitor significantly inhibited the dilation by 50% (Fig. 1, A and B); combination of both inhibitors essentially eliminated the responses (Fig. 1C).
|
50% (Fig. 2, A and B). Flow-induced dilation was abolished by additional administration of L-NAME (Fig. 2C).
|
40% and 70%, respectively (Fig. 3).
|
Evidence provided by Western blotting (Fig. 4) indicates that COX-1 protein content in gracilis muscle arterioles of WT and eNOS-KO mice was comparable. However, COX-2 protein, which was basically undetectable in arterioles of WT mice, was strongly expressed in eNOS-KO vessels. The densitometric ratio of COX-1 to -actin (Fig. 4B) was not different in vessels of the two groups, but the densitometric ratio of COX-2 to -actin (Fig. 4C) increased significantly, by 250-fold, in vessels of eNOS-KO mice compared with WT mice. Normalized gene expression of COX-2 increased by 16-fold in arterioles of eNOS-KO mice compared with WT mice (Fig. 5). COX-1 mRNA was not different in the vessels of WT and eNOS-KO mice.
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
As demonstrated previously, active and passive diameters of arterioles isolated from eNOS-KO mice were smaller than those from WT mice (31). We presumed that this was caused by a vascular remodeling attributable to the elevated blood pressure in eNOS-null mice, but a causal relation has not been established. On the other hand, the basal tone of arterioles in eNOS-KO mice was not significantly different from that in WT mice; moreover, the capacity of the arteriolar endothelium to sense shear stress, followed by vasodilation, was not significantly altered in eNOS-KO arterioles. This indicates that endothelial cells adapt to the chronic lack of NO synthesis and maintain a normal or close-to-normal regulation of vascular tone via upregulation of an NO-independent mechanism.
Similar to our previous findings (31), corelease of NO and PGs mediates flow-induced dilation of skeletal muscle arterioles of WT mice, as shown by our findings that VS attenuated the dilation by 50% and abolished the response when administered in combination with L-NAME (Fig. 2). Also, the PG-mediated responses were not affected by NS-398 (Fig. 2B), and COX-2 protein was not expressed in these vessels (Fig. 4), suggesting that, under normal conditions, COX-2-derived PGs do not contribute to the mediation of flow-induced dilation.
In adapting to the lack of endothelium-derived NO, flow-induced dilation of arterioles of male eNOS-KO mice is mediated solely by PGs that are derived from COX-1 and COX-2. This is evidenced by the fact that each inhibitor of the COX isoforms blocks flow-induced dilation by 50% (Fig. 1, A and B) and a combination of both inhibitors eliminates the dilations (Fig. 1C). Moreover, as shown in Fig. 3, PGI2 is the major mediator responsible for dilation in these vessels. The cascade of events, involving conversion of the physical stimulus of shear stress to chemical signals, followed by the synthesis/release of vasoactive autacoids, takes place in the vascular endothelium. Signaling cascades responsible for shear stress-stimulated PG release have been demonstrated to be linked to the activation of phospholipase C at the cell membrane, as well as increased intracellular Ca2+, followed by stimulation of downstream COX enzymes to initiate conversion of arachidonic acid to PGs (5, 7). Our results provide evidence of a predominant expression of COX-2 protein (Figs. 4 and 6) in endothelial cells of eNOS-KO, but not WT, arterioles, suggesting that a COX-2-dependent signaling cascade is responsible for the shear stress-initiated release of PGs. Interestingly, COX-2 was originally believed to be induced only in response to cytokines to act continuously to release increased amounts of PGs, as observed during the process of inflammation (11). In the present study, however, COX-2 seems to be functional only in response to shear stress, not in basal conditions, because although NS-389 inhibited flow-induced dilation by 50%, it had no effect on the basal tone of arterioles. An increased expression of endothelial COX-2 in response to shear stress has previously been reported (25, 33). On the other hand, we, as well as others, have noted that, in eNOS-KO and WT mice, treatment of arteries/arterioles with L-NAME or indomethacin or removal of the endothelium barely affected basal vascular diameters or smooth muscle resting membrane potential, indicating that basal release of endothelial NO and PGs may not contribute significantly to the maintenance of vascular tone (6, 28, 31). Together, these data suggest that, in response to NO deficiency, COX-2 is upregulated and accounts for the compensatory preservation of flow-induced dilation.
The literature regarding the cross talk between the NOS and COX pathways is divided with respect to whether NO activates or inhibits PG synthesis. Although our findings are likely to be consistent with the studies that support the hypothesis of a negative correlation between the two mediators (2, 3, 11, 26), still the question arises as to why this compensatory activity depends specifically on COX-2. This issue could perhaps be best explained by the evidence showing that NO exerts divergent effects on the COX isoforms, with potentiating COX-1 but inhibiting COX-2 (8). With the use of cultured fibroblasts of COX-1-KO and COX-2-KO mice, it was demonstrated that exposure of COX-2-KO cells to NO significantly increases the release of PGE2 without changing COX-1 mRNA or protein expression, suggesting a nongenomic regulation of COX-1 enzymatic activity. In contrast, NO inhibited COX-2-derived PGE2 production in COX-1-KO cells and LPS-stimulated macrophages, associated with a decreased expression of COX-2 mRNA and protein, as well as nitration of COX-2, leading to an impaired catalytic activity of the enzyme (10). In addition, a recent study reported a role for L-NAME in an initiation of damage to the intestine of rats via upregulation of COX-2 mRNA, further suggesting a negative relation between NOS and COX-2 (24). Thus these studies support our findings that an upregulation of endothelial COX-2 expression (gene and protein) in response to shear stress could be attributable to NO deficiency (25, 33). Consistent with the present findings, upregulation of COX-2 in the course of adaptation of an attenuated dilation to shear stress in mesenteric arterioles of L-NAME-treated rats has also been reported (12). Interestingly, we also found that the VS-sensitive portion of flow-induced dilation, as well as COX-1 protein expression, was comparable in arterioles of WT and eNOS-KO mice, indicating that the function of COX-1 in the mediation of flow-induced dilation of microvessels seems to be relatively stable and less affected by NO than by COX-2, a phenomenon that has also been observed in previous studies (2, 11, 12, 14–16, 29, 30, 32). It has also been suggested that COX-1 is more resistant than COX-2 to nitration by NO, because micromolar concentrations of NO elicit COX-2 nitration without effects on COX-1 (8). Results of the present study lead to the conclusion that COX-2-derived PGs, together with those derived from COX-1, compensate for the deficiency of eNOS to preserve shear stress-dependent mechanisms in the microvascular endothelium.
Perspectives
PGs accompanied by NO participate in the mediation of shear stress-sensitive regulation of microvascular tone. Whereas many aspects of the action of shear stress on endothelial NO production remain open to question, even less is known about the equally potent ability of shear stress to stimulate endothelial PG synthesis, an issue that becomes especially important when NO synthesis is impaired or absent. The interactions between NO and PGs are complex and depend on the local environment in which these endothelial mediators are synthesized. Also, changes in the expression of endothelial mediators in response to NO deficiency are species, tissue, and cell specific, as well as sex and stimulus dependent (13, 17). For example, in skeletal muscle arterioles of eNOS-KO mice, dilation to ACh is counterbalanced by endothelium-derived hyperpolarizing factor-mediated responses (4, 16) and by neuronal NOS-derived NO in cerebral arteries (23) and PGs in mesenteric arteries (6), whereas flow-induced dilation in coronary arteries is counterbalanced by a neuronal NOS-dependent mechanism (15). Thus it is tempting to speculate that perhaps one or more redundant compensatory mechanisms become activated after selective gene deletion or enzyme inactivation. Moreover, it is important to point out that, in contrast to other NO deficiency models, such as hypertension or heart failure, in which NO deficiency is a consequence of the pathological process, in animals that underwent deletion of the eNOS gene or were treated with an NOS inhibitor, NO deficiency acts as an initial cause to activate the compensatory process. Therefore, it could be expected that in the former, the compensation, if any, is usually not complete and is eventually followed by decompensation. In this context, we found previously that, in coronary arterioles of dogs, maintenance/compensation of flow-induced dilation was only observed in the early stage of heart failure (3 wk) and was followed quickly by permanent impairment of the responses (32). In adult spontaneously hypertensive rats, shear stress-induced dilation of skeletal muscle arterioles was attenuated by 50%, attributable simply to the loss of the NO-mediated portion of the response (18). However, in rats of the same age treated chronically with L-NAME, high blood pressure was associated with a preserved shear stress-induced dilation that was mediated by PGs (34), a response similar to that observed in the present study. Although we do not have information about the time course of PG-dependent compensation of shear stress-induced responses in eNOS-KO mice and L-NAME-treated rats, our results suggest a relatively complete compensation in these NO deficiency models. Thus an understanding of different compensatory mechanisms in response to NO deficiency may be important in the investigation of adaptive mechanisms, as well as interactions among the various endothelial mediators, to maintain normal vascular function.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
-Estradiol restores endothelial nitric oxide release to shear stress in arterioles of male hypertensive rats. Circulation 101: 94–100, 2000.CT) method. Methods 25: 402–408, 2001.[CrossRef][Web of Science][Medline]This article has been cited by other articles:
|
|
 |
|
  J. L. Park, L. Shu, and J. A. Shayman Differential involvement of COX1 and COX2 in the vasculopathy associated with the {alpha}-galactosidase A-knockout mouse Am J Physiol Heart Circ Physiol, April 1, 2009; 296(4): H1133 - H1140. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Dumont, F. Pinaud, A.-L. Guihot, C. Baufreton, L. Loufrani, and D. Henrion Alteration in flow (shear stress)-induced remodelling in rat resistance arteries with aging: improvement by a treatment with hydralazine Cardiovasc Res, February 1, 2008; 77(3): 600 - 608. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. L. Haas, J. L. Doyle, M. R. Distasi, L. E. Norton, K. M. Sheridan, and J. L. Unthank Involvement of MMPs in the outward remodeling of collateral mesenteric arteries Am J Physiol Heart Circ Physiol, October 1, 2007; 293(4): H2429 - H2437. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Focardi, G. M. Dick, A. Picchi, C. Zhang, and W. M. Chilian Restoration of coronary endothelial function in obese Zucker rats by a low-carbohydrate diet Am J Physiol Heart Circ Physiol, May 1, 2007; 292(5): H2093 - H2099. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. B. Anning, B. Coles, J. Morton, H. Wang, J. Uddin, J. D. Morrow, S. K. Dey, L. J. Marnett, and V. B. O'Donnell Nitric oxide deficiency promotes vascular side effects of cyclooxygenase inhibitors Blood, December 15, 2006; 108(13): 4059 - 4062. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
