Evidence for Vasculoprotective Effects of ETB Receptors in Resistance Artery Remodeling in Diabetes

  1. Kamakshi Sachidanandam1,
  2. Vera Portik-Dobos2,
  3. Alex K. Harris1,
  4. Jim R. Hutchinson2,
  5. Erin Muller2,
  6. Maribeth H. Johnson3 and
  7. Adviye Ergul1,2
  1. 1Program in Clinical and Experimental Therapeutics, University of Georgia College of Pharmacy, Augusta, Georgia
  2. 2Department of Physiology, Medical College of Georgia, Augusta, Georgia
  3. 3Department of Biostatistics, Medical College of Georgia, Augusta, Georgia
  1. Address correspondence and reprint requests to Adviye Ergul, MD, PhD, Medical College of Georgia, Dept. of Physiology CA-2094, Augusta, GA 30912. E-mail: aergul{at}


OBJECTIVE—Vascular remodeling, characterized by extracellular matrix deposition and increased media-to-lumen (M/l) ratio, contributes to the development of microvascular complications in diabetes. Matrix metalloproteinases (MMPs) play an important role in the regulation of extracellular matrix (ECM) turnover and vascular remodeling. Vasoactive factor endothelin (ET)-1 not only causes potent vasoconstriction but also exerts profibrotic and proliferative effects that change vessel architecture, which makes it a likely candidate for a key role in vascular complications of diabetes. Thus, this study investigated the regulation of MMP activity of resistance arteries under mild-to-moderate diabetes conditions, as seen in type 2 diabetes, and the relative role of ET receptors in this process.

RESEARCH DESIGN AND METHODS—Vessel structure, MMP activity, and ECM proteins were assessed in control Wistar and diabetic Goto-Kakizaki (GK) rats treated with vehicle, ETA receptor antagonist atrasentan (5 mg · kg−1 · day−1), or ETB receptor antagonist A-192621 (15 mg · kg−1 · day−1) for 4 weeks.

RESULTS—M/l ratio was increased in diabetes. Atrasentan prevented this increase, whereas A-192621 caused further thickening of the medial layer. Increased MMP-2 activity in diabetes was prevented by atrasentan treatment. Collagenase activity was significantly decreased in diabetes, and while ETA antagonism improved enzyme activity, ETB blockade further reduced collagenase levels. Accordingly, collagen deposition was augmented in GK rats, which was reversed by atrasentan but exacerbated with A-192621.

CONCLUSIONS—ET-1 contributes to the remodeling of mesenteric resistance arteries in diabetes via activation of ETA receptors, and ETB receptors provide vasculoprotective effects.

Cardiovascular complications contribute to the increased morbidity and mortality in type 2 diabetes. Pathological changes in vascular function and structure underlie these complications. Vasoactive factor endothelin (ET)-1 not only causes potent vasoconstriction but also exerts profibrotic and proliferative effects that change vessel architecture, which make it a likely candidate for a key role in vascular complications of diabetes. A significant correlation has been observed between plasma ET-1 levels and diabetes complications (1,2). ETA receptor antagonism prevents mesenteric vascular hypertrophy in type 1 diabetes (3). Fukuda et al. (4) reported that blockade of ET-1 action inhibits aortic extracellular matrix (ECM) deposition. We showed that ET-1 levels are elevated and that an ETA antagonist prevents ECM deposition and MMP activation in middle cerebral arteries but not in the kidney of Goto-Kakizaki (GK) rats—a nonobese type 2 diabetes model (5,6). ET-1 mediates its diverse effects via distinct G protein–coupled receptor subtypes ETBAB and ETBBB. While these past studies indicate the involvement of the ETA receptor subtype in mediating detrimental effects of ET-1, the role of ETB receptors in diabetes-induced changes in the vasculature remains unknown. ETBAB receptors, localized mainly on vascular smooth muscle cells (VSMCs), are responsible for the vasocontractile and proliferative response to ET-1 (7). Endothelial ETBB receptors mediate vasodilatation, whereas VSMC ETB receptors can also lead to vasoconstriction in certain vascular beds (7). This duality of function of ETBB receptors underscores the importance of binding and function of both ET receptor subtypes. Especially given that inhibition of the ETBBB receptor system in a knockout mouse model or pharmacological blockade by an ETBBB antagonist leads to enhanced intimal hyperplasia observed in carotid arteries after injury induced by ligation, suggesting a vasculoprotective effect (8), the question remains whether ETB receptors contribute to or balance detrimental effects of ETA receptor activation in diabetic vascular remodeling.

Vascular ECM displays a very dynamic equilibrium where there is constant synthesis, degradation, and reorganization. Turnover of matrix proteins are regulated by matrix metalloproteinases (MMPs) (9). While decreased MMP activity is generally believed to contribute to ECM accumulation in diabetic kidney and in vascular tissue from patients with diabetes, we and others have recently reported that there is an early activation of MMPs in hypertension and diabetes and that ET-1 is involved in regulation of MMP activity (5,6,10,11). Recent reports indicated that MMPs are also important in generating myogenic tone of mesenteric arteries (12,13). However, regulation of the MMP activity of these resistance arteries under mild-to-moderate diabetes conditions as seen in type 2 diabetes and the relative role of ET receptors in this process remain unknown. Accordingly, this study investigated vessel structure, MMP activity, and ECM proteins in control Wistar and diabetic GK rats chronically treated with vehicle, ETA receptor antagonist atrasentan, or ETB receptor antagonist A-192621. The central hypothesis was that increased MMP activity would be associated with hypertrophic remodeling of the mesenteric resistance arteries in diabetes, and while ETA receptor blockade would prevent this effect, antagonism of vasculoprotective ETB receptor would display opposing effects augmenting pathological remodeling in type 2 diabetes.


Animal and tissue preparation.

All experiments were performed on male Wistar (Harlan, Indianapolis, IN) and GK (in-house bred, derived from the Tampa colony) rats (14,15). The animals were housed at the Medical College of Georgia animal care facility, which is approved by the American Association for Accreditation of Laboratory Animal Care. All protocols were approved by the institutional animal care and use committee. Animals were fed standard rat chow and tap water ad libitum until killed at 18 weeks of age. Weight and blood glucose measurements were monitored twice a week until they were killed. Blood glucose was measured from the tail vein using a commercially available glucometer (Freesytle, Alameda, CA). Blood pressure was monitored either by telemetric method (as previously reported) (6) or via the tail- cuff method (Kent Scientific, Torrington, CT), which we have previously validated on telemetry-implanted animals (5). After the spontaneous onset of diabetes, starting at 14 weeks of age, animals were divided into groups and treated for 4 weeks as follows: ETA receptor blockade, atrasentan (Abbott Labs) 5 mg · kg−1 · day−1 in drinking water; ETB receptor blockade, A-192621 (Abbott Labs) 15 mg · kg−1 · day−1 by oral gavage split into two daily doses; or vehicle as recommended by the manufacturer (6,16,17). Daily water consumption was measured for atrasentan treatment arm (6). Vehicle for the A-192621 consisted of 83% deionized water, 10% polyethylene glycol-400, 5% ethanol, and 2% cremaphor EL. Tap water was used as vehicle for atrasentan. Animals were anesthesized with sodium pentobarbital and exsanguinated via cardiac puncture. The mesenteric bed was then harvested, and third-order mesenteric arteries were isolated for morphometry and biochemical studies. For immunohistochemistry, mesenteric arteries were perfused with Histogel (Richard Allen Scientific, Kalamazoo, MI), then excised and embedded in the same matrix. Upon gelling of the matrix, the embedded vessel was placed in 10% formalin for storage. For protein studies, vessels were excised, snap frozen in liquid nitrogen, and stored at −80°C.

Plasma measurements.

Plasma ET-1 and insulin were measured by specific enzyme-linked immunosorbent assay kits from ALPCO Diagnostics (Windham, NH).

Tissue homogenization and MMP activity.

Snap-frozen mesenteric arteries were homogenized in modified radioimmunoprecipitation assay buffer (50 mmol/l Tris-HCl; 1% nonidet P-40; 0.25% Na-deoxycholate; 150 mmol/l NaCl; 1 mmol/l phenylmethylsulfonyl fluoride; 1 μg/ml aprotinin, leupeptin, and pepstatin; 1 mmol/l sodium orthovanadate; and 1 mmol/l sodium fluoride) as we previously described (6). Gelatinolytic activity was assessed by densitometric analysis (Gel-Pro version 3.1; Media Cybernetics, Carlsbad, CA) (6). Recombinant MMP-2 and MMP-9 proteins (Calbiochem, San Diego, CA) were run in parallel with all samples, and the band intensity on zymogram gels was normalized to that of standard to prevent gel-to-gel variability. Collagenase activity of vascular MMP-13 was determined using a fluorescein-conjugated collagen assay kit as recommended by the manufacturer (Molecular Probes, Eugene, OR). Briefly, homogenates (20 μg total protein) were incubated with the substrate, and increased fluorescence that is directly proportional to the proteolytic activity of MMP-13 was measured at time 0, 2, 4, 8, and 24 h using a microplate fluorometer. Other serine proteases in the tissue extracts were blocked by using 50 mmol/l phenylmethylsulfonyl fluoride. Tissue inhibitor of MMP (TIMP)-2 levels were measured by enzyme-linked immunosorbent assay (Amersham Biosciences, Piscataway, NJ).

Vessel morphometry and immunohistochemistry.

Fixed vessel segments were embedded in paraffin, sectioned at 4 μ, and mounted on treated slides. Sections were stained with Masson trichrome stain. Slides were viewed using a Zeiss Axiovert microscope (Carl Zeiss, Thornwood, NY), and media-to-lumen (M/l) ratios were analyzed using Spot software (Diagnostic Instruments, Sterling Heights, MI). Four measurements were made per section; each animal had at least three sections. For immunohistochemistry, slides were then deparaffinized, blocked (Super Block; Biogenex Labs, San Ramon, CA), and placed in PBS for 5 min. Then, slides were incubated with MMP-2 primary antibody at room temperature, washed, and incubated in secondary antibody (LSAB2-HRP kit; Dako, Carpinteria, CA), followed by incubation with Streptavidin-HRP. Bound antibody was detected with a DAB substrate kit. Additional slides were incubated with only the secondary antibody to determine nonspecific staining.


Protein levels of MMP-2 were determined by immunoblotting as previously described (6,18), and antibodies were from Calbiochem (Cambridge, MA). Collagen type 1 and collagen type 4 levels were evaluated by slot-blot analysis, and antibodies were from BD Transduction Laboratories (San Jose, CA). All blots were restained with anti-actin antibody (Sigma, St. Louis, MO) for equal protein loading. Using a subset of samples, blots were hybridized with secondary antibody alone to evaluate nonspecific binding; there was no detectable signal.

Statistical analysis.

A rank transformation was applied to the data before analysis to address distributional issues for all outcome variables (19). A 2 × 3 ANOVA was used to investigate the main effects of disease (control vs. diabetic) and drug (saline vs. atrasentan vs. low-dose A-192621) and the interaction between disease and drug. Effects were considered statistically significant at P < 0.05. SAS version 9.1.3 was used for all analyses.


Metabolic parameters.

Metabolic parameters for all study groups are summarized in Table 1. Diabetic animals were significantly smaller than controls, and ET receptor antagonism did not affect animal weight. GK animals displayed elevated blood glucose in all treatment groups. There was a disease-drug interaction such that A-192621 treatment caused a further elevation than vehicle in GK rats but not in controls.


Metabolic parameters for animals in treatment groups

Vascular structure.

GK animals exhibited increased media thickness of mesenteric arteries with a significantly increased M/l ratio. ETA antagonism prevented the increase in M/l, but ETB antagonism caused a further increase in the M/l ratio in diabetic GK rats but not in controls. A representative image for each group is shown in Fig. 1A and morphometry data from all animals studied summarized in Fig. 1B. Total collagen type 1 levels were quantified by slot-blot analysis, in addition to the qualitative assessment of collagen staining in Masson-trichrome–stained sections. Densitometric analysis (Fig. 2A) demonstrated increased collagen type 1 in diabetes. Furthermore, there was a disease and treatment interaction such that atrasentan treatment attenuated collagen deposition, whereas A-192621 caused augmented accumulation in GK animals (P = 0.047). ETB blockade increased collagen type 1 in control rats, but it did not reach statistical significance. There was no difference in collagen type 4 levels in control versus GK animals.

FIG. 1.

Vessel segments were analyzed for morphologic changes and collagen deposition by Masson staining. Diabetes induced a twofold increase in M/l ratio; ETA receptor antagonism by atrasentan prevented this increase, whereas ETB receptor blockade with A-192621 caused a further medial thickening. Representative sections are shown in A and combined analysis given in B. n = 5–8/group. *P < 0.001 vs. control; #P < 0.001 vs. diabetes vehicle or A-192621; ζP < 0.01 vs. diabetes vehicle. (Please see for a high-quality digital representation of this figure.)

FIG. 2.

Effects of ET receptor antagonism on collagen deposition in diabetes. A: Mesenteric collagen type 1 levels, assessed by slot-blot analysis, demonstrated increased deposition in diabetes that was attenuated by atrasentan but worsened by A-192621. n = 5–9/group. *P < 0.01 vs. control vehicle; #P < 0.05 vs. diabetes vehicle; ζP < 0.01 vs. diabetes vehicle or atrasentan. B: MMP-13 collagenase activity was measured by incubating tissue homogenates with a fluorogenic MMP-13 substrate, which was decreased in diabetes. Similar to collagen results, atrasentan treatment restored collagenase activity in diabetes, but A-192621 reduced enzyme activity even in control animals. n = 5/group. *P < 0.001 vs. diabetes vehicle; #P < 0.001 diabetes vehicle vs. diabetes atrasentan; ζP < 0.001 control A-192621 vs. control vehicle. Diabetic groups are indicated by broken lines.

Mesenteric MMP expression and activity.

Since there is increased collagen accumulation in diabetes, first mesenteric collagenase (MMP-13) activity was assessed using a fluorogenic assay. In diabetic animals, this activity was significantly decreased (Fig. 2B). Similar to morphometry results, ETA blockade restored collagenase activity to levels seen in control animals, and ETB antagonism caused a further but not significant decrease in GK rats. Surprisingly A-192621 treatment significantly reduced collagenase activity, whereas ETA blockade had no effect on enzyme activity in control animals. Gelatinolytic activity was evaluated using gelatin zymography, which detects MMP-2–and MMP-9–based lytic activity. A representative zymogram is shown in Fig. 3A. Lytic activity was detected mainly at 62 kDa and to a much lesser extent at 72 kDa, corresponding to the active and latent forms of MMP-2, respectively. Densitometric analysis of the bands corresponding to active form demonstrated that MMP-2 activity was increased in diabetes and that ETA receptor antagonism prevented this increase in activation (Fig. 3B). ETB receptor antagonist–treated group also displayed lower activity levels than vehicle-treated diabetic rats.

FIG. 3.

MMP-2 activity is increased in diabetes. A: Representative zymogram showing changes in vascular MMP-2 activity. B: Densitometric analysis of lytic bands indicates an increase in MMP-2 activity, which is ameliorated by both ETA and ETB receptor blockade. n = 5–10/group. *P < 0.01 vs. control vehicle; #P < 0.001 vs. diabetes vehicle; ζP < 0.05 vs. diabetes vehicle.

To determine whether increased MMP-2 activity results from an increase in protein levels, total MMP-2 protein was assessed by immunoblotting (Fig. 4A). MMP-2 levels were higher in the GK rats, and treatment with either antagonist had no effect on protein levels in diabetic animals. On the other hand, ETB antagonism increased MMP-2 levels in the control group. In only control and GK rats, immunohistochemistry was performed to determine localization of MMP-2 protein in the vessel wall. There was intense staining in the entire wall, including both medial and adventitial layers (Fig. 4B).

FIG. 4.

MMP-2 protein is increased regulated in diabetes. A: Densitometric analysis of immunoreactive bands indicates that MMP-2 protein is increased in diabetes, and while ET receptor antagonism has no effect on MMP-2 levels in the diabetic group, ETB blockade increases MMP-2 protein in control rats. n = 5–10/group. *P < 0.01 vs. control vehicle; #P < 0.05 vs. control vehicle. B: To localize increased MMP-2 protein in diabetes, frozen mesenteric artery cross-sections were immunostained with an MMP-2 antibody (n = 3/group), which demonstrated diffuse staining along the vessel wall compared with controls. Nonspecific staining was determined in the absence of primary antibody. C: TIMP-2 protein is decreased in diabetes, and ETA receptor antagonism restores it to control levels, whereas ETB blockade had no effect. n = 5–10/group. *P = 0.05 vs. control vehicle; #P < 0.05 vs. diabetes vehicle; ζP < 0.05 vs. control vehicle.

Since TIMP-2 is the endogenous inhibitor of MMP-2, tissue levels were measured to determine whether the increase in MMP-2 activity arises from a decrease in its inhibitor (Fig. 4C). TIMP-2 was significantly decreased in diabetes. ETA antagonism restored TIMP-2 levels in GK rats without any effect on control animals. ETB blockade had no effect in the diabetic group and decreased TIMP-2 levels in controls.


This study tested the hypothesis that ETA receptor activation contributes to the remodeling of mesenteric microvessels via modulation of ECM dynamics by MMPs in type 2 diabetes, whereas ETB receptors exert a vasculoprotective effect by balancing detrimental effects of ETA receptors. This hypothesis was based on previous studies showing that ET-1 stimulates collagen accumulation in aortic and mesenteric vessels in type 1 diabetes and that pharmacological or genetic manipulation of ETB receptors results in augmented intimal remodeling in a carotid injury model (3,4,8,20). This study was designed to examine structural changes in resistance arteries of control and diabetic animals chronically treated with an ETA or ETB receptor antagonist and also to evaluate potential mechanisms of altered matrix dynamics in the microvasculature in diabetes. Our findings demonstrate for the first time that mild hyperglycemia causes medial thickening in the mesenteric arteries, which is associated with increased gelatinase MMP-2 but decreased collagenase MMP-13 activity. Furthermore, MMP activation and increased M/l ratio can be prevented by the administration of an ETA receptor antagonist. On the other hand, ETB receptor blockade worsens the remodeling process, suggesting a vasculoprotective effect of this receptor subtype.

The chemically induced streptozocin model of type 1 diabetes is most commonly used to study complications of diabetes. However, this model presents with very high glucose levels. Gilbert et al. (3) reported that ETA receptor antagonism prevents mesenteric vascular hypertrophy in this model by inhibiting macrophage infiltration and epidermal growth factor signaling. There is also evidence from a transgenic mouse model that overexpresses ET-1 only in endothelial cells that ET-1 causes remodeling and dysfunction of mesenteric arteries but not conduit vessels like aorta (21). We have shown that treatment with an ETA antagonist prevented diabetes-induced changes in expression of MMPs and procollagen type 1 in mesenteric arteries but not in aorta (11). However, the effect of ET-1 on resistance artery structure in models of type 2 diabetes in which blood glucose levels are comparable with those seen in patients remained unknown. More importantly, the relative role(s) of ET receptor subtypes in mediating vascular remodeling in diabetes was unclear. Murakoshi et al. (8) reported that inhibition of the ETB receptor system in a knockout mouse model or pharmacological blockade by an ETB antagonist led to enhanced intimal hyperplasia observed in carotid arteries after injury induced by ligation, suggesting that blockade of the receptor subtype may be detrimental. Thus, this study compared the effects of selective ETA versus ETB antagonism on vascular remodeling in diabetes. Our intriguing findings provide evidence that blockade of ETA receptors prevent diabetes-induced changes in vessel structure, whereas ETB antagonism causes an opposing effect and causes further medial thickening of the resistance arteries. These results support a vasculoprotective role for the ETB receptor in the regulation of vessel architecture.

While diabetes has been reported to promote medial hypertrophy of mesenteric vessels characterized by collagen deposition and not VSMC hypertrophy (22), regulation of ECM dynamics in diabetes is not fully understood. MMPs are very important for ECM degradation, and these enzymes are regulated at various levels (9). Increased ECM protein synthesis diminished MMP activity, and/or increased TIMP activity could contribute to matrix accumulation. To the best of our knowledge, the current study is the first to report regulation of proteins involved in ECM synthesis and degradation with respect to resistance vessel structure in type 2 diabetes. In diabetic animals, resistance arteries display increased gelatinase MMP-2 and decreased collagenase MMP-13 activity. Since MMP-13 is responsible for degrading fibrillar collagen, attenuated degradation may be one mechanism of increased collagen deposition. Moreover, it is becoming clear that MMPs have complex roles in the regulation of ECM. Degradation of basement membrane and internal elastic lamina by MMPs disrupts the boundaries between vascular layers and facilitating VSMC migration (23,24). Breakdown of fibrillar collagen reveals cryptic integrin signals buried in the ECM, which serve as chemotactic stimuli for VSMC migration (24,25). Equally important, MMPs activate membrane-bound proteins with growth-promoting properties via proteolytic cleavage (2628). Thus, it is likely that augmented MMP-2 activity may indeed be stimulating vessel restructuring as seen in this study. An alternative explanation is that enhanced MMP activity at this stage of the disease may represent a compensatory response to prevent ECM deposition. However, whether MMPs directly contribute to this process or upregulate as a consequence of hypertrophy in the mesenteric bed remains to be determined.

Flamant et al. (29) demonstrated that ET-1 enhances epidermal growth factor transactivation, a process that requires MMP activation (28). In the current study, increased MMP-2 activity was attenuated and blunted MMP-13 activity restored by ETA receptor antagonism, suggesting a dual mechanism for ET-1–mediated medial thickening in GK rats. It is also interesting to note that elevated MMP-2 activity paralleled increased MMP-2 protein levels in diabetes. However, ETA antagonism restored MMP-2 activity by not decreasing MMP-2 protein levels but rather stimulating the inhibitory TIMP-2 protein. As discussed above, ETB antagonism resulted in opposing effects compared with ETA blockade, causing a further increase in M/l ratio in diabetes. In addition, A-192621 treatment increased collagen levels and reduced collagenase activity not only in diabetic animals but also in control animals, indicating a role for this receptor subtype in the regulation of collagen turnover.

Taken together, we conclude that mild hyperglycemia without confounding effects of hyperlipidemia and hypertension in the GK model of type 2 diabetes appears to regulate MMP proteins and activity by both transcriptional and posttranslational mechanisms leading to hypertrophic remodeling of resistance arteries, and ET-1 mediates these effects via activation of mainly ETA receptors. Furthermore, ETB receptors seem to balance the detrimental growth-promoting effects of ETA receptor activation.


This work was supported by grants from Philip Morris USA, Inc., Philip Morris International, and the National Institutes of Health (grants R15HL76236 and RO1 DK074385).

We thank Abbott Laboratories for atrasentan and A-192621 compound.


  • Published ahead of print at on 1 August 2007. DOI: 10.2337/db07-0426.

  • The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

    • Received March 28, 2007.
    • Accepted July 27, 2007.


| Table of Contents

This Article

  1. Diabetes vol. 56 no. 11 2753-2758
  1. All Versions of this Article:
    1. db07-0426v1
    2. 56/11/2753 most recent