EP1 Disruption Attenuates End-Organ Damage in a Mouse Model of HypertensionNovelty and Significance
Prostaglandin E2 is a major prostanoid found in the kidney and vasculature contributing to the regulation of blood pressure. The prostaglandin E2 receptor EP1 has been shown to contribute to hypertension by mediating angiotensin II-dependent vasoconstriction, although its precise role is incompletely characterized. Disruption of the EP1 receptor in C57BL/6J mice reduced the incidence of mortality during severe hypertension induced by uninephrectomy, deoxycorticosterone acetate, and angiotensin II. Mortality was dependent on all components of the model. Death was a result of aortic aneurysm rupture or occurred after development of anasarca, each of which was reduced in EP1−/− mice. Mean arterial pressure was increased in treated EP1+/+ and EP1−/− mice; however, this elevation was significantly lower in EP1−/− mice. Blood pressure reduction via administration of hydralazine phenocopied EP1−/− mice. Thus, reduction in blood pressure by disruption of EP1 reduced incidence of mortality and decreased organ damage, suggesting that EP1 receptor blockade may be a viable target for antihypertensive therapy.
Hypertension is a major risk factor for cardiovascular diseases, increasing the risk of stroke, myocardial infarction, arterial aneurysms, and heart failure. Approximately 30% of adults in the United States have hypertension, and the incidence of cardiovascular diseases remains greater in hypertensive patients than in normotensive patients, highlighting the need for novel therapeutic agents.1
Prostaglandin E2 (PGE2) is a major prostanoid found in the mouse kidney and vasculature contributing to the regulation of blood pressure, where it can exert either vasopressor or vasodepressor effects.2–4 Four PGE2 receptors (EP1 through EP4) mediate these effects, with the EP1 and EP3 receptors primarily mediating the pressor response of PGE2, whereas the EP2 and EP4 receptors mediate the depressor response.2,5–10
Each PGE2 receptor has distinct tissue localization and elicits characteristic signal transduction pathways.11 EP1 couples to Gq proteins, resulting in mobilization of intracellular calcium and stimulation of phosphoinositide turnover activating protein kinase C. The EP2 and EP4 receptors couple to Gsproteins, which increase intracellular cAMP. The EP3 receptor couples to Gi proteins, decreasing intracellular cAMP (for reviews, see Coleman et al and Sugimoto et al.11,12). Receptors couple to alternate signal transduction pathways as well, including arrestin-mediated signaling pathways.13,14
Systemic infusion of PGE2 results in a vasodepressor response,2,5,15 primarily through EP2 activation. In the absence of EP2, a PGE2 pressor response is unmasked,2 mediated by the EP3 receptor.8 Agonist-induced EP3 tachyphylaxis in the background of EP2−/− mice uncovers a depressor action of EP4.8,16 EP1 does not appear to play a significant role in the blood pressure effects of systemically administered PGE2; however, it does contribute to hypertension. Genetic disruption of the EP1 receptor in mice has been shown to decrease blood pressure, particularly when mice are fed a low-sodium diet.17 Furthermore, EP1−/− mice have blunted pressor responses to both acute and chronic angiotensin II (Ang II) administration.10 In isolated vessel preparations, pretreatment with the EP1 selective antagonist SC51322 reduced Ang II-mediated vasoconstriction.10 Treatment of spontaneously hypertensive rats with SC51322 significantly reduces blood pressure,10 indicating that blockade of the EP1 receptor may be a target for the treatment of hypertension.
EP1 blockade has been shown to affect renal function positively in stroke-prone spontaneously hypertensive rats,18 as well as cerebrovascular dysfunction induced by Ang II,19 implicating the EP1 receptor in hypertension and resultant end-organ damage. This has yet to be investigated in detail and in the context of other organ damage. Therefore, EP1+/+ and EP1−/− mice were studied in a model of severe hypertension.
Complete methods can be found in the expanded Methods section in the online-only Data Supplement.
EP1−/− Mice Were Protected Against Nphx/DOCA-NaCl/Ang II Mortality
We used a model of severe hypertension to investigate the contribution of the EP1 receptor in hypertensive end-organ damage.20 Unexpected mortality was observed after implantation of the Ang II osmotic pump (Figure 1A). Of 58 EP1+/+ mice, 60% died within 14 days. EP1−/− mice were significantly protected against mortality; of 35 EP1−/− mice only 24% died (P=0.0044). Modified protocols omitting 1 of the 3 components (Nphx, DOCA-NaCl, or Ang II) demonstrated that all components of the model played an essential role in causing mortality (Figure 1B, Nphx/DOCA-NaCl/Ang II versus Nphx/DOCA-NaCl, Nphx/Ang II, or DOCA-NaCl/Ang II; N=10 per group; P=0.011, <0.005, or <0.005, respectively).
EP1−/− Mice Had Reduced Aortic Aneurysm Rupture but Comparable Aortic Histopathology
Postmortem analysis of EP1+/+ and EP1−/− mice in the full model indicated that a significant portion of the mice died as a result of rupture of the aorta. Aneurysms and dissections were observed in both the thoracic and abdominal aortas. A total of 37% of EP1+/+ and 13% of EP1−/− mice died because of aortic rupture. Aortic aneurysms were present in 67% of EP1+/+ mice and 40% of EP1−/− mice. Aneurysm severity was scored at death on a scale of 0 to 5 (Figure 2A). A reduction in aneurysm severity was observed in EP1−/− mice compared with EP1+/+ mice (Figure 2B; P=0.049). Hematoxylin and eosin staining revealed aneurysms, and dissections in the wall of the thoracic and abdominal aortas were accompanied by inflammation in both EP1+/+ and EP1−/− mice (Figure 3A through 3E; P=0.3975). Analysis of aortic sections from both thoracic and abdominal lesions of each genotype demonstrated macrophage, and neutrophils were most abundant, with no differences observed between the genotypes in any immune cell component (Figure 3F). Cyclooxygenase 2 (COX-2) has been shown to play a significant role in aortic aneurysm formation and macrophage infiltration.21,22 COX-2 mRNA was elevated in abdominal aortas 2 to 5 days after Ang II administration, although differences between genotypes were not observed (Figure 3G; P>0.774). Aortic sections stained with Masson trichrome showed that, regardless of genotype, there was less fibrillar collagen present in vessels that ruptured, and the amount and organization of collagen surrounding the lesion were not significantly different in EP1+/+ and EP1−/− intact aneurysms (Figure S1 in the online-only Data Supplement; P=0.1925).
Anasarca Was Observed in EP1+/+ Mice
A subset of EP1+/+ mice appeared to have substantial edema and displayed an increase in body weight, peaking ≈5 days after Ang II administration (Figure 4A). Average body weight in the EP1+/+ cohort subsequently decreased because of mortality in the animals with the largest weight gain. At baseline, EP1+/+ mice weighed more than EP1−/− mice, although this difference was modest (EP1+/+, 26.6±0.39 g; EP1−/−, 24.9±0.68 g; P=0.024). Body weight of EP1−/− mice was unchanged over the course of the study. EP1+/+ mice had a significantly greater fraction water weight compared with EP1−/− mice (Figure 4B; P=0.0138), and aneurysm incidence was lower in mice which developed anasarca compared with mice without anasarca (33% versus 76%). Anasarca is commonly a result of liver failure, nephrotic syndrome, or heart failure.23,24 Plasma alanine transaminase activity was analyzed as a marker of liver function in Nphx/DOCA-NaCl/Ang II-treated EP1+/+ mice. Alanine transaminase was not elevated above baseline values nor did it correlate with body weight (data not shown).
Modest Renal Injury Was Induced in EP1+/+ and EP1−/− Mice
The Nphx/DOCA-NaCl/Ang II model was initially developed to induce hypertensive renal damage on the C57BL/6 background.20 To quantify renal damage in the EP1+/+ and EP1−/− mice, we monitored urinary albumin excretion, serum urea nitrogen, renal histopathology, and biomarkers of acute kidney injury, neutrophil gelatinase-associated lipocalin, and kidney injury molecule 1 mRNA expression (Figure 5). Albumin creatinine ratio and serum urea nitrogen were elevated, but no significant differences were observed between genotypes (Figure 5A and 5B). Renal histopathology showed modest hypertensive renal damage compared with the contralateral kidney removed at time of uninephrectomy (Figure 5C). Dilated tubules with moderate glomerularsclerosis and tubulointerstitial fibrosis were observed. Significant increases in neutrophil gelatinase-associated lipocalin and kidney injury molecule-1mRNA expression in the kidney of EP1+/+ and EP1−/− mice were observed, although no differences were detected between genotypes (Figure 5D and 5E).
Cardiac Function Is Reduced in EP1+/+ and EP1−/− Mice
No structural differences in the heart were observed by echocardiography between EP1+/+ and EP1−/− mice at baseline. A modest increase in ejection fraction and fractional shortening was observed in EP1−/− mice at baseline. At 5 days after Ang II administration, EP1+/+ had increased left ventricular posterior wall and interventricular septum diastolic diameters. EP1−/− mice had increased interventricular septum diastolic diameters, although no significant change in left ventricular posterior wall was observed. EP1+/+ and EP1−/− mice displayed increased left ventricular interior diameter, although no differences were observed between genotypes. Cardiac function was significantly reduced in both genotypes on treatment with Nphx/DOCA-NaCl/Ang II, as demonstrated by decreased fractional shortening and ejection fraction (Table). Additionally, heart weights of EP1+/+ and EP1−/− mice after treatment with Nphx/DOCA-NaCl/Ang II showed no significant difference between the genotypes (EP1+/+: 198.9±9.6; N=21, EP1−/−: 190.0±4.3; N=17; P=0.461).
Hypertension Was Less Severe in EP1−/− Mice Than in EP1+/+ Mice
To determine the effect of disruption of the EP1 receptor on blood pressure in Nphx/DOCA-NaCl/Ang II-treated animals, intracarotid blood pressure was determined in EP1+/+ and EP1−/− mice 2 days after Ang II administration (Figure 6). Mean arterial pressure (MAP) was significantly increased compared with untreated animals in both EP1+/+ (76.79±5.33 mm Hg baseline versus 128.8±5.08 mm Hg; P=0.0004) and EP1−/− mice (74.36±5.61 mm Hg baseline versus 102.4±7.77 mm Hg; P=0.0423). However, the rise in MAP was significantly lower in EP1−/− compared with EP1+/+ mice (P=0.0295).
Reduction of Blood Pressure Protected Against Mortality
Fifteen EP1+/+ and EP1−/− mice treated with Nphx/DOCA-NaCl/Ang II were treated with the antihypertensive agent hydralazine. Hydralazine treatment significantly reduced MAP in EP1+/+ mice (106.1±5.76 mm Hg versus 128.8±5.08 mm Hg; P=0.024), but had no significant effect on EP1−/− mice (Figure 7A, F799.47±4.82 mm Hg versus 102.4±7.77 mm Hg; P=0.762). A significant decrease in the incidence of mortality was observed in EP1+/+, but not in EP1−/− mice (Figure 7B, P=0.007 and P=0.642, respectively). Hydralazine treatment reduced aneurysm incidence and severity in EP1+/+, although this did not achieve statistical significance (Figure 7C, P>0.05). Anasarca was not observed in hydralazine-treated EP1+/+ mice (body weight at 5 days post-Ang II: 21.40±0.65 g) as compared with untreated EP1+/+ mice (body weight at 5 days post-Ang II: 32.28±1.59 g; P<0.0001).
Disruption of the EP1 receptor affords substantial protection in the Nphx/DOCA-NaCl/Ang II-evoked hypertension. The incidence of mortality was significantly decreased and appeared to result from reduction in MAP. Mortality was a result of ruptured aortic aneurysm or occurred after developing anasarca. The results presented here are consistent with previous studies demonstrating the role of EP1 in modulating the rise in MAP in response to Ang II,10 and furthermore reveal the protective effect disruption that the EP1 receptor has on end-organ damage.
There are several limitations to these studies that deserve mention. First, high mortality observed in EP1+/+ mice confounds the analysis of measurements taken after implantation of Ang II, such as cardiac and renal function, because the analyses are only performed on surviving mice. Second, there was a modest (<10%), although statistically significant, difference in body weight observed between EP1+/+ and EP1−/− mice. Although the dosage of Ang II was adjusted by weight, the dosage of the DOCA pellet was not. However, one would predict the genotype receiving the greater dose/weight (EP1−/−) would have the worse phenotype, and this is opposite of what we observed. Lastly, EP1−/− mice were observed to have lower blood pressure than EP1+/+ mice after treatment with the Nphx/DOCA-NaCl/Ang II model. This is consistent with our data⇓ published previously,10 suggesting that EP1 mediates part of Ang II-induced hypertension. In this model we measured MAP at baseline or after treatment with all 3 model components; it is possible that EP1 also contributes to hypertension induced by Nphx or DOCA-NaCl as well.
Current models of aortic aneurysm include a combination of hyperlipidemic mice or high-fat diet with modulation of the renin-angiotensin-aldosterone axis or aberrant production of extracellular matrix components.25 Ang II-induced aortic aneurysms are characterized by accumulation of macrophages in the adventia and media, disruption of elastin fibers, expansion of the lumen, thrombus formation, and disordered extracellular matrix deposition.26 These characteristics were also observed in the Nphx/DOCA-NaCl/Ang II model, although no significant differences in macrophage accumulation, matrix deposition, or COX-2 mRNA expression were detected between the 2 genotypes. It should be noted that the aneurysms and dissections observed in this model occur after acute severe hypertension, and although the pathology appears similar to that observed in human disease, the disease genesis may not be. In humans, development of a true aneurysm is a slowly progressing disease initiating with local inflammation and disruption of the connective tissue matrix and is often associated with atherosclerosis. In contrast, development of false aneurysm or dissection as a result of a tear in the intima can occur more acutely by a sudden large rise in blood pressure or direct injury and may be more representative of the damage induced by the Nphx/DOCA-NaCl/Ang II model. Our data demonstrate that protection observed when EP1 is disrupted is likely because of the prevention of a large rise in blood pressure, because treatment with hydralazine phenocopied EP1−/− mice. This does not eliminate the possibility that EP1 receptors might also provide protection directly at the target tissue.
Data exist suggesting a role for prostaglandins, in particular PGE2, in aortic aneurysm formation. COX-2 initiates the production of prostaglandins, and its expression is induced by infusion of Ang II in the smooth muscle of the aorta surrounding aneurysms.21 Furthermore, either selective inhibition of COX-2 or genetic deletion of COX-2 significantly reduced aortic aneurysm formation and macrophage infiltration.21,22 Deletion of microsomal PGE synthase 1, which transforms the product of COX-2 metabolism into PGE2, has also been demonstrated to reduce aortic aneurysm formation and oxidative stress in low density lipoprotein receptor null mice with an Ang II infusion,27 suggesting that PGE2 plays an important role in development of aneurysms and the EP receptors may be viable targets for treatment of aneurysm progression.
Previous reports of the role of EP1 in renal injury are contradictory. In spontaneously hypertensive rats, treatment with an EP1 antagonist reduced proteinuria and tubulointerstitial damage,18 whereas in anti-glomerular basement membrane nephrotoxic serum nephritis genetic deletion of EP1−/− in mice resulted in enhanced mesangial expansion and tubular dilation and increased serum urea nitrogen and serum creatinine.28 In our studies, modest hypertensive renal damage was observed, although no significant differences in renal function were detected between genotypes. However, our interpretation was confounded by the differential mortalities in EP1+/+ and EP1−/− mice, potentially biasing our results. Examination of renal histopathology at time points before significant mortality failed to detect any severe renal damage or differences between the genotypes. This suggests that the role of EP1 in renal damage is highly context dependent.
Anasarca, or extreme generalized edema, can occur in many disease settings. It is commonly a result of liver failure, nephrotic syndrome, or heart failure.23,24 In our Nphx/DOCA-NaCl/Ang II model, a subset of EP1+/+ mice developed severe anasarca before mortality, whereas EP1−/− mice were protected. The EP1 receptor has been shown previously to be natriuretic.10 With this paradigm, one might predict that EP1−/− mice would retain more salt and water; however, in our results we demonstrate that EP1+/+ mice gain excessive fluid volume that is not observed in EP1−/− mice. This contradiction leads us to conclude that alterations in kidney function by disruption of EP1 do not play a dominant role in the development of the observed edema. Additionally, cardiac function was reduced to similar degrees in EP1+/+ and EP1−/− mice. Edema was prevented by treatment with hydralazine, suggesting elevation in blood pressure was responsible for development of edema. We hypothesize that hypertension induced by DOCA-NaCl and Ang II results in volume loading and enhanced vasoconstriction, which places excessive stress on the vascular wall leading to enhanced permeability, resulting in edema and susceptibility to dissections and rupture. Future experiments will be required to identify whether vascular permeability differences are observed between EP1+/+ and EP1−/−mice.
The EP1 receptor plays an important role in the development of hypertensive damage. In the Nphx/DOCA-NaCl/Ang II model, disruption of EP1 results in increased survival, lessened aneurysm severity, and the absence of anasarca. This effect is a result of a reduced rise in blood pressure observed in EP1−/− mice and suggests the EP1 receptor may be a viable pharmaceutical target for the treatment of hypertension and subsequent organ damage. Furthermore, the Nphx/DOCA-NaCl/Ang II model may prove to be a useful tool for studying the pathology of aortic aneurysm and dissection formation in a setting of acute severe hypertension.
We thank Jason Downey for careful critique of the article and Dr Matthew Breyer for helpful discussion.
Sources of funding
This work was supported by National Institutes of Health grants DK46205 (R.M.B.), DK37097 (R.M.B.), P50GM015431 (R.M.B.), 2P01DK065123 (R.Z.), DK075594 (R.Z.), DK9921 (R.Z.), DK62794 (R.C.H.), and O’Brien P30DK79341-01 (R.Z., and R.C.H.). The Mouse Metabolic Phenotyping Center is supported in part by grant DK059637. R.Z., R.C.H., and R.M.B. have merit awards from the Department of Veterans Affairs, and R.Z. has an American Heart Association established investigator award.
The online-only Data Supplement is available with this article at http://hyper.ahajournals.org/lookup/suppl/doi:10.1161/HYPERTENSIONAHA.112.199026/-/DC1.
- Received May 18, 2012.
- Revision received June 8, 2012.
- Accepted August 20, 2012.
- © 2012 American Heart Association, Inc.
- Breyer MD,
- Breyer RM
- Zhang Y,
- Guan Y,
- Schneider A,
- Brandon S,
- Breyer RM,
- Breyer MD
- Audoly LP,
- Ruan X,
- Wagner VA,
- Goulet JL,
- Tilley SL,
- Koller BH,
- Coffman TM,
- Arendshorst WJ
- Sugimoto Y,
- Narumiya S
- Kim JI,
- Lakshmikanthan V,
- Frilot N,
- Daaka Y
- Chun KS,
- Lao HC,
- Langenbach R
- Zhang Y,
- Guan Y,
- Schneider A,
- Brandon S,
- Breyer RM,
- Breyer MD
- Suganami T,
- Mori K,
- Tanaka I,
- Mukoyama M,
- Sugawara A,
- Makino H,
- Muro S,
- Yahata K,
- Ohuchida S,
- Maruyama T,
- Narumiya S,
- Nakao K
- Capone C,
- Faraco G,
- Anrather J,
- Zhou P,
- Iadecola C
- Gitlin JM,
- Trivedi DB,
- Langenbach R,
- Loftin CD
- King VL,
- Trivedi DB,
- Gitlin JM,
- Loftin CD
- Kemp CD,
- Conte JV
- Daugherty A,
- Cassis LA
- Wang M,
- Lee E,
- Song W,
- Ricciotti E,
- Rader DJ,
- Lawson JA,
- Puré E,
- FitzGerald GA
Novelty and Significance
What is New?
Describes new model of aortic aneurysm
Anasarca is observed
Disruption of the EP1 receptor improves the outcome of each pathology and overall survival
What Is Relevant?
Studies shed light on the role of EP1 in the pathophysiology of hypertension
Identify a therapeutic target for hypertension and its sequelae
The Nphx/DOCA-NaCl/Ang II model induces hypertension, aortic aneurysm formation, and anasarca. Disruption of the EP1 receptor reduced blood pressure in this model, leading to reduced incidence of mortality and decreased organ damage.