Renal Medullary Interstitial COX-2 (Cyclooxygenase-2) Is Essential in Preventing Salt-Sensitive Hypertension and Maintaining Renal Inner Medulla/Papilla Structural Integrity

Abstract

COX (cyclooxygenase)-derived prostaglandins regulate renal hemodynamics and salt and water homeostasis. Inhibition of COX activity causes blood pressure elevation. In addition, chronic analgesic abuse can induce renal injury, including papillary necrosis. COX-2 is highly expressed in the kidney papilla in renal medullary interstitial cells (RMICs). However, its role in blood pressure and papillary integrity in vivo has not been definitively studied. In mice with selective, inducible RMIC COX-2 deletion, a high-salt diet led to an increase in blood pressure that peaked at 4 to 5 weeks and was associated with increased papillary expression of AQP2 (aquaporin 2) and ENac (epithelial sodium channel) and decreased expression of cystic fibrosis transmembrane conductance regulator. With continued high-salt feeding, the mice with RMIC COX-2 deletion had progressive decreases in blood pressure from its peak. After return to a normal-salt diet for 3 weeks, blood pressure remained low and was associated with a persistent urinary concentrating defect. Within 2 weeks of institution of a high-salt diet, increased apoptotic RMICs and collecting duct cells could be detected in papillae with RMIC deletion of COX-2, and by 9 weeks of high salt, there was a striking loss of the papillae. Therefore, RMIC COX-2 expression plays a crucial role in renal handling water and sodium homeostasis, preventing salt-sensitive hypertension and maintaining structural integrity of papilla.

Keywords: blood pressure; homeostasis; kidney; mice; prostaglandin.

Figure 1. Renal medullary interstitial cell COX-2 was effectively deleted in Tenascin-C-CreER2: COX-2^f/f (RMIC COX-2^f/f) mice.

A: After tamoxifen administration for 2 weeks, renal inner medullae/papillae were dissected and COX-2 mRNA was quantitated by qPCR. COX-2 mRNA levels in renal medulla/papilla were significantly reduced in RMIC COX-2^−/− mice. ***P < 0.001, n = 6 in COX-2^f/f mice and n = 5 in RMIC COX-2^−/− mice. All values are means ± SEM. P value was calculated by Student’s t test. B: After administration of a high salt diet for 2 weeks, COX-2 positive cells were found in many inner medullary interstitial cells in COX-2^f/f mice (arrows), but only a few inner medullary interstitial cells were COX-2 positive in RMIC COX-2^−/− mice. Scale bar: 50 μM.

Figure 2. RMIC COX-2 deficiency led to impaired diuresis and natriuresis in response to acute volume expansion.

Acute volume expansion was carried out in mice with normal chow or mice on a high salt diet for 2 weeks. A: RMIC COX-2 deletion led to decreases in urine volume and sodium excretion at baseline. *P < 0.05, n = 4 in both groups. B: RMIC COX-2 deficiency decreased urine volume and sodium excretion 2 weeks after high salt diet. *P < 0.05, n = 5 in COX-2^f/f mice and n = 3 in RMIC COX-2^−/− mice. All values are means ± SEM. P values were calculated by Student’s t test.

Figure 3. RMIC COX-2 deficiency led to increases in expression of sodium transporters and aquaporin 2 but decreases in CFTR.

Mice were fed with high salt diet for 2 weeks. A: RMIC COX-2 deficiency led to increases in aquaporin 2 (AQP2) mRNA and protein expression in the inner medullae/papillae. In addition, the immunostaining density in the apical plasma membrane was higher in RMIC COX-2^−/− mice than in COX-2^f/f mice (arrows), an indication of increased AQP2 activity. Original magnification: x 400. ***P < 0.001, n = 6 in the COX-2^f/f group and n = 8 in the RMIC COX-2^−/− group. Scale bar: 50 μM. B: The mRNA levels of CFTR decreased in RMIC COX-2^−/− mice. **P < 0.01, n = 6 in the COX-2^f/f group and n = 8 in the RMIC COX-2^−/− group. C: The mRNA levels of ENaCα, EnaCγ, and Atp1b1 were increased in RMIC COX-2^−/− mice. *P < 0.05, **P < 0.01, n = 6 in the COX-2^f/f group and n = 8 in the RMIC COX-2^−/− group. All values are means ± SEM. P values were calculated by Student’s t test.

Figure 4. RMIC COX-2 deficiency caused salt-sensitive hypertension and papillary damage in response to chronic salt loading Both COX-2^f/f and RMIC COX-2−/− mice were fed with a high salt diet (8% NaCl) after tamoxifen administration for 2 weeks.

A: Systolic blood pressure (SBP) was monitored with tail cuff microphonic manometer following initiation of high salt treatment (8% NaCl). High salt administration had no apparent effect on blood pressure in COX-2^f/f mice throughout experiment period, but increased blood pressure in RMIC COX-2 deficient mice for the first 5 weeks, and then led to progressively decreased blood pressure. N= 7 in COX-2^f/f mice and n = 6 RMIC COX-2^−/− mice. All values are means ± SEM. B: Diastolic blood perssure (SBP), systolic blood pressure (SBP) and mean arterial pressure (MAP) measured by carotid catheterization after a high salt diet for 4–5 weeks was markedly higher in RMIC COX-2^−/− mice than in COX-2^f/f mice. **P < 0.01; n = 8 in COX-2^f/f mice and n = 6 in RMIC COX-2^−/− mice. C and D: Both COX-2^f/f and RMIC COX-2^−/− mice were fed with a high salt diet (8% NaCl) for 9 weeks, and then fed with normal chow for 3 weeks. C: Blood pressure was markedly lower in RMIC COX-2^−/− mice than in COX-2^f/f mice. ***P < 0.001; n= 7 in COX-2^f/f mice and n = 6 RMIC COX-2^−/− mice. D: Urine osmolality was markedly lower in RMIC COX-2^−/− mice than in COX-2^f/f mice after 16 hours of water deprivation. **P < 0.01; n= 6 in COX-2^f/f mice and n = 5 RMIC COX-2^−/− mice. All values are shown as mean ± SEM. P values were calculated by Student’s t test.

Figure 5. Chronic high salt intake caused renal papillary loss in RMIC COX-2 deficient mice.

Direct morphologic examination (upper panels) or magnetic resonance imaging (MRI, lower panels) showed striking papillary loss after long-term high salt exposure (9–10 weeks) in RMIC COX-2^−/− mice, but not in COX-2^f/f mice.

Figure 6. RMIC COX-2 deficiency led to apoptosis in the inner medullae/papillae in response to high salt diet.

A: RMIC COX-2^−/− mice had increased papillary apoptotic cells as early as 2 weeks after a high salt diet, and more apoptotic cells were evident in both interstitial and epithelial cells 4 weeks after a high salt diet. ***P < 0.001 vs. corresponding COX-2^f/f mice, ^†††P < 0.001 vs. 2 weeks of RMIC COX-2^−/− mice; n = 6 in each group. Scale bar: 25 μM. B: Four weeks after high salt intake, there was increased papillary oxidative stress in RMIC COX-2^−/− mice, as indicated by 4-HNE-staining. Original magnification: x 160. All values are shown as mean ± SEM. P values were calculated by Student’s t test. Scale bar: 160 μM.