Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Feb 5.
Published in final edited form as: Circ Res. 2009 Nov 25;106(2):337. doi: 10.1161/CIRCRESAHA.109.204529

COX-2 Dependent Prostacyclin Formation and Blood Pressure Homeostasis: Targeted Exchange of COX Isoforms in Mice

Ying Yu 1, Jane Stubbe 1, Salam Ibrahim 1, Wen-liang Song 1, Emer M Symth 1, Colin D Funk 2, Garret A FitzGerald 1
PMCID: PMC2818801  NIHMSID: NIHMS166933  PMID: 19940265

Abstract

Rationale

Cyclooxygenase (COX)-derived prostanoids (PGs) are involved in blood pressure (BP) homeostasis. Both traditional(t) nonsteroidal anti-inflammatory drugs (NSAIDs) that inhibit COX-1 and COX-2 and NSAIDs designed to be selective for inhibition of COX-2 cause sodium retention and elevate BP.

Objective

To elucidate the role of COX-2 in BP homeostasis using COX-1>COX-2 mice, in which the COX-1 expression is controlled by COX-2 regulatory elements.

Methods and Results

COX-1>COX-2 mice developed systolic hypertension relative to WTs on a high salt diet (HSD); this was attenuated by a PGI2 receptor (IP) agonist. HSD increased expression of COX-2 in WT mice and of COX-1 in COX-1>COX-2 mice in the inner renal medulla (IM). The HSD augmented in all strains urinary prostanoid metabolite excretion, with the exception of the major PGI2 metabolite that was suppressed on regular chow and unaltered by the HSD in both mutants. Furthermore, IM expression of the receptor for PGI2, but not for other prostanoids, was depressed by HSD in WT and even more so in both mutant strains. Increasing osmolarity augmented expression of COX-2 in WT renal medullary interstitial cells and again the increase in formation of PGI2 observed in WTs was suppressed in cells derived from both mutants. Intramedullary infusion of the IP agonist increased urine volume and sodium excretion in mice.

Conclusion

These studies suggest that dysregulated expression of the COX-2 dependent, PGI2 biosynthesis/response pathway in the IM undermines the homeostatic response to a HSD. Inhibition of this pathway may contribute directly to the hypertensive response to NSAIDs.

Keywords: Cyclooxygenase-2, Nonsteroidal anti-inflammatory drugs, Hypertension, Prostacyclin, IP receptor

Introduction

Prostaglandins contribute to BP homeostasis via their direct effects on vascular tone and on fluid and electrolyte transport in the kidney. Nonsteroidal anti-inflammatory drugs (NSAIDs) block PGs biosynthesis by inhibition of the activity of the cyclooxygenase (COX) isozymes, COX-1 and COX-2 1. Both traditional (t) NSAIDs and those designed to be selective for inhibition of COX-2 may increase systemic BP and/or undermine BP control with antihypertensive drugs2, 3. PG formation is generally reactive and elaboration of vasodilator PGs preserves renal blood flow in renoprival conditions4. Similarly, NSAIDs, in vulnerable populations, such as the elderly or in response to a hypertensive stimulus, such as a HSD, decrease total renal perfusion and cause redistribution of renal blood flow5. This may lead to medullary ischemia, and even acute renal failure 6. Even short-term studies of NSAIDs in apparently healthy, but susceptible populations may result in decreased glomerular filtration rate (GFR) and urinary sodium retention7.

In rodents, COX-1 deletion causes natriuresis, accentuates the effects of angiotensin converting enzyme (ACE) inhibitors and reduces BP despite activation of the renin-angiotensin system (RAS)8. Indeed, both pharmacological inhibition and genetic deletion of COX-1 abolish the hypertensive response to Angiotensin II (Ang II) in mice 9, 10. Deletion or inhibition of COX-2, by contrast, reduces renal medullary blood flow and sodium excretion and increases the vasoconstrictive response to Ang II 10, and elevates basal BP11. These observations have prompted the suggestion that hypertension on NSAIDs is a function of both inhibition of COX-2 and the selectivity with which it is attained12. While this is consistent with some evidence3, 13, both the relative importance of selectivity and the mechanism by which COX-2 preserves BP homeostasis remain to be rigorously addressed.

Although the COX isozymes are structurally similar, their method of regulation is quite distinct1. One mechanism by which COX-2 might play a role in BP homeostasis is by its ready induction in renal medullary interstitial cells (RMICs) where it is coexpressed with COX-114 and induced by high salt or fluid deprivation15, 16. Inhibition of medullary COX-2 in rats17, 18 or global deletion of the PGI2 receptor (IP)19, 20 or the EP2 PGE2 receptor 21, 22 in mice results in salt sensitive hypertension. We designed COX-1>COX-2 mice23 with which to address the hypothesis that dysregulated expression of COX-2, rather than structural distinctions of COX-224 from COX-1, might have relevance to BP homeostasis. Our results support the notion that dysregulated expression of the COX-2 dependent, PGI2 biosynthesis/response pathway in the renal internal medulla (IM) undermines the homeostatic response to a HSD. Inhibition of this pathway may contribute directly to the hypertensive response to NSAIDs.

Methods

An expanded methods section is available in the Online Supplement.

Mice

All COX-1>COX-2 and COX-2 null mice used for the experiments were initially produced on a mixed C57BL/6 × Sv129 genetic background (50%:50%). All procedures were approved by the Institutional Animal Care and Use Committee of the University of Pennsylvania.

BP measurement

Resting systolic blood was measured in conscious mice using a computerized noninvasive tail cuff system (Visitech Systems Inc.) as previously described11.

Analysis of renal medullar perfusion and urinary sodium excretion

Mice were anesthetized with a mixture of ketamine (100 mg/kg) and xylazine (10 mg/kg i.m.), and the right kidney was removed. After one week of recovery, catheters were implanted in the jugular vein, bladder and renal medullary interstitium as described previously25. Saline or Cicaprost tested compound was infused into the renal medulla at 20 μl/h. After one hour of equilibration, urine was collected every 30 min. Urinary sodium concentration was determined by Electrolyte Analyzer (PL 1000B).

Statistical Analysis

Data are presented as mean ±SEM. Analyses were performed by the analysis of variance and subsequent pairwise comparisons as appropriate. A p valueof <0.05 was considered significant. Prism 4.0 software (GraphPad InStat 3) was used for all the calculations.

Results

COX-1 rescues the BP and renin response to COX-2 deletion on a normal diet but not on a HSD

COX-2 deficient mice exhibited systolic hypertension (145.0±5.4 mmHg vs WT littermates, 122.0±7.8 mmHg; p<0.01 Figure 1A and Online Figure I), while BP was unaltered in COX-1>COX-2 mice BP (127.0±5.5 mmHg vs WT littermates, 123.3±4.4 mmHg, Figure 1A) on a normal chow diet. These results indicate COX-1 could substitute for COX-2 in maintaining BP homeostasis under physiological conditions. This is also consistent with the ability of COX-1 under the COX-2 promoter to rescue impaired renal development, characteristic of COX-2 null mice23, 26. The renal cortical hypoplasia and biochemical evidence of renal dysfunction in COX-2 KOs was absent in COX-1>COX-2 mice (Online Figure II). However, in response to the HSD, COX-1>COX-2 mice developed hypertension (149.7±7.3 mmHg vs WT, 128.1± 6.5 mmHg, p<0.05, n=9–12, Figure 1A), just like COX-2 KOs (151.0 ± 2.9 mmHg; p < 0.05). Thus, while COX-1 could rescue the renal developmental impact of COX-2 deficiency, including hypertension on a chow diet, it was unable to compensate for the absence of COX-2 in regulating BP homeostasis in the face of a high salt challenge.

Figure 1. Effects of normal or HSD on blood pressure and renin expression in COX-1>COX-2, COX-2 KO and WT mice.

Figure 1

Open in a new tab

A. Systolic blood pressure (BP) was measured in conscious COX-1>COX-2, WT and COX-2 KO mice by tail cuff (n = 9–12 for each group). Mice (6–7 weeks old) were fed either a normal chow diet (0.7% NaCl) or a HSD (8% NaCl) for 2 weeks. *, p<0.05 vs WT controls; #, p<0.05 vs normal diet. B. Quantitative real time RT-PCR was performed to analyze renin mRNA expression in the kidneys from COX-1>COX-2, WT and COX-2 KO mice. *, p<0.05 vs WT controls; #, p<0.05 vs normal diet, n=4–6.

Downstream products of COX-2 include renin secretagogues, like PGI227, 28. Renin expression was reduced on a chow diet by ~35% in COX-2 KO mice and this phenotype was rescued in COX-1>COX-2 mice (p<0.05, Figure 1B). However, analogous to the BP response to HSD, the homeostatic decline in renin observed in COX-2 KOs was not rescued in the COX-1>COX-2 mice.

Insertion of COX-1 does not substitute for COX-2 derived biosynthesis of PGI2

We have previously shown in mice and humans that COX-2 is the dominant source of PGI2 biosynthesis, as measured by its urinary metabolite, 2,3-dinor-6-keto PGF1α (PGIM) 11, 24. By contrast, the major urinary metabolite of PGE2, 9,15-dioxo-11α-hydroxy-2,3,4,5-tetranor-prostane-1,20-dioic acid (PGEM), derives from both COX-1 and COX-211. Deletion of COX-2 suppressed both urinary PGIM and PGEM in mice on a chow diet. However, while COX-1 insertion could rescue the decline in PGEM (1.72 ±0.38 ng/mg creatinine in COX-2 KO vs 3.15±0.47 ng/mg creatinine in COX-1>COX-2 vs 3.89±0.92 ng/mg creatinine in WT,) (Figure 2A), it had no impact on the suppression of urinary PGIM (Figure 2B). A similar disparity in the ability of COX-1 to substitute for COX-2 applied under HSD conditions. HSD increased excretion of all metabolites consistent with induction of COX-2 in renal medulla (see below) in WT mice. Deletion of COX-2 had no impact on excretion of major metabolites of either PGD2 (11,15-dioxo-9alpha-hydroxy-2,3,4,5-tetranorprostan-1,20-dioic acid) or Tx(2,3-dinor TxB2), both of which derive predominantly from COX-1 29, 30, on either chow or HSD (Figure 2C and 2D).

Figure 2. Effect of high salt intake on PG biosynthesis in COX-1>COX-2, COX-2 KO and WT mice.

Figure 2

Open in a new tab

Twenty four hour urine samples from COX-1>COX-2, COX-2 KO and WT male mice (8 weeks old) were collected, the urinary PGE2 metabolite-PGE-M (A), PGI2 metabolite PGI-M (B) PGD2 metabolite-PGD-M (C), Tx2 metabolite-Tx-M (D) were measured. Data are presented as mean ± SEMs, *, p<0.05 vs WT controls (n=7–20).

Medullary COX-1 expression in response to HSD in the absence of COX-2

HSD is recognized to have divergent effects on expression of COXs in kidney15. As expected, the HSD had no significant effect on WT COX-1 expression in both cortex and medulla (Figure 3A and 3C), whereas it downregulated COX-2 expression in cortex and upregulated COX-2 in medulla in WT mice (Figure 3B and 3D). Indeed, cortical expression of COX-1 tended to be higher in the mutants on regular chow and this change reached significance on a HSD. COX-2 expression was absent, as expected, in both cortex and medulla in the mutants (Figure 3B and 3D). By contrast, HSD upregulated expression of COX-1 in the cortex of COX-1>COX-2 (760 ± 30 ×106/18S rRNA, p<0.05) and COX-2 KO vs WT (717± 45 ×106/18S rRNA vs 503 ± 34 ×106/18S rRNA, p<0.05) mice (Figure 3A), and of COX-1 in the medulla of COX-1>COX-2 (3768± 242 ×106/18S rRNA vs 2302 ± 167 ×106/18S rRNA, p<0.05), while it did not regulate expression of medullary COX-1 when the isozyme was expressed under its own promoter (Figure 3C). Thus, COX-2 is usually the dominant isoform regulated by HSD in the medulla.

Figure 3. Effect of high salt intake on COX-1 and COX-2 expression in renal tissue obtained from COX-1>COX-2, COX-2 KO and WT mice.

Figure 3

Open in a new tab

COX-1 and COX-2 mRNA in renal cortex (A, B) and inner medulla (C, D) were determined by real time RT-PCR; HS, HSD. *, p<0.05 vs WT controls; #, p<0.05 vs normal diet, n=4–6.

Medullary COX-1 cannot restore the COX-2 dependent capacity to generate PGI2

Given the relevance of COX-2 in the renal medulla to BP homeostasis17, 18, the tissue capacity to generate PGs was assessed. This is quite distinct from estimation of systemic PG biosynthesis, as reflected by urinary metabolites; the capacity of tissues to generate PGs greatly exceeds actual synthetic rates in vivo31.

The most abundant product was PGE2 (Figure 4). Both PGE2 and PGI2 (detected as 6-keto-PGF1α) are vasodilators and PGI2 is also a potentrenin secretagogue. Consistent with hypertension and hyporeninema (Figures 1A and 1B), the capacity to generate these PGs is reduced in COX-2 KOs on a chow diet (Figures 4A and C). By contrast, the rather trivial medullary capacity to generate PGD2 and PGF2α, both of which derive predominantly from COX-1, is retained (Figures 4B and D).

Figure 4. Effect of high salt intake on the capacity of renal medulla to generate PGs.

Figure 4

Open in a new tab

Mice were fed a HSD or normal chow diet for 2 weeks, the renal medulla was dissected and the PG profile (A, PGE2; B, PGD2; C, 6-keto-PGF1α; D, PGF2α) was analyzed by mass spectrometry. *, p<0.05 vs WT controls, n=5–8; #, p<0.05 vs normal diet group, n=5–8.

HSD augments the capacity to generate renal medullary PGs in WT mice. However, this capacity is attenuated in the case of PGI2 in both mutant strains (Figures 4C). This indicates that COX-2 is the dominant source of this PG and it cannot be substituted for by COX-1 under these conditions. The HSD induced increase in PGE2 is restrained only in COX-2 KOs, indicating the ability of COX-1 to compensate for COX-2 in the generation of this PG. Given the failure of COX-1 insertion to compensate for the BP effect of deletion of COX-2 (Figure 1), these results implicate PGI2 as the dominant product of COX-2 restraining the hypertensive response to HSD.

Reduced expression of the medullary PGI2 receptor on a HSD

While the EP1 and EP3 receptors for PGE2 were by far the most abundantly expressed in renal medulla, their expression was unaltered in the mutants or by HSD. This was also true for the other receptors for PGE2 – EP2 and EP4 – and also for receptors for PGF2α-the FP – and for PGD2 – the DP2. Expression of DP1 was not detectable (Online Figure III). Expression of the receptor for TxA2, the TP, was suppressed compared to WT by COX-2 deletion on both chow and HSD (Online. Figure III) although this was not evident in the renal cortex (Online Figure IV) where it is expressed in the glomeruli32. Pertinent to its proposed role in the hypertensive response to HSD, medullary expression of the PGI2 receptor, the IP, is depressed by HSD, even in WT mice and expression of its transcript is detectably further depressed in both mutants (Figure 5A). Accordingly, we observed IP protein suppression by HSD was detectable in renal medulla although no significant difference detected among genotypes (Figure 5B and 5C). Thus, decreased expression of the IP could interact with depressed synthesis of PGI2 in the renal medulla to underlie the hypertensive response to COX-2 deletion and the failure of COX-1 to compensate for this deficiency. A similar pattern of decreased expression of the IP in both mutants on HSD was observed in the renal cortex (Online Figure IV), where it may be relevant to the attendant hyporeninema (Figure 1B).

Figure 5. HSD decreases the medullary IP receptor expression in COX-1>COX-2, COX-2 and WT mice.

Figure 5

Open in a new tab

A. IP mRNA levels in medulla from COX-1>COX-2, COX-2 KO and WT mice were quantitated by real time RT-PCR; HS, HSD treatment. #, p<0.05 vs normal diet; *, p<0.05 vs WT controls, n=4–6. B. Representative Western blot of medullary IP receptor in COX-1>COX-2, COX-2 KO and WT mice. N, Normal salt diet; H, HSD. C. Relative level of IP protein to β-actin expression. #, p<0.05 vs normal diet, n=4.

IP receptor activation restrains salt-induced hypertension in COX-1>COX-2 mice

Systemic and renal PGI2 biosynthesis were impaired under both basal and HSD conditions and renal IP receptor expression was further depressed by HSD in COX-1>COX-2 mice. To examine whether down-regulation of PGI2/IP signaling contributes to salt sensitive hypertension in COX-2 mutant mice, COX-1>COX-2 mice were subcutaneously infused with the IP specific agonist, cicaprost (50 μg/kg/day), using an osmotic mini-pump. BP was recorded before implantation and after exposure to a HSD by tail-cuff. Systolic BP in COX-1>COX-2 mice was elevated markedly by the HSD and this effect was significantly attenuated by cicaprost infusion (from 140.8 ± 4.1 to 119.4 ± 4.2 mmHg, p<0.01, n=9. Online Figure V).

Differential substitution of COX-1 for hypertonic induction of COX-2 dependent PG formation in RMICs

In rodents, COX-2 expression is restricted in RMICs in renal medulla14. RIMCs were isolated to address the impact of COX-1 substitution for COX-2 in an in vitro model of regulated expression of COX-2. The RMICs derived from COX-2 KOs were hypersensitive to cell dissociation, more than 45% of 3rd passage COX-2 KO RMICs died when treated with 0.025% trypsin/0.5mM EDTA compared to less than 3% in cells from either COX-1>COX-2 or WT groups. Unlike the cells from COX-1>COX-2 and WT mice (which passaged for more than 10 generations), even passage 4 RMICs were not derived successfully from COX-2 KO mice. The results are consistent with previous reports that COX-2 plays important role in RMIC survival 16, 33

Consistent with our analysis of transcripts in the renal medulla of animals on a HSD (Figure 3), exposure of RMICs to hypertonic conditions (using addition of NaCl and mannitol to the culture medium) induced expression of COX-2 in WT cells and of COX-1 in those derived from COX-1>COX-2 mice (Figure 6A). The capacity of these cells to make PGs was augmented by 630 mOsm/kg H2O (Figure 6B); again PGE2 is the most abundant product. While the capacity to form PGI2 is modest in these cells, induction of its formation is evident. Like PGE2, deletion of COX-2 depresses significantly RMIC production of PGI2. Substitution of COX-1 for COX-2 in the COX-1>COX-2 mice differentially rescued PG formation under hyperosmolar conditions. Thus, while 81%, 75% and 60% capacity to form PGE2, PGD2 and PGF2α was restored, COX-1 compensated only one third of the capacity to form PGI2 (measured as 6-keto-PGF1α). Expression of the IP receptor in the renal medulla is largely restricted to the vasa recta 34, 35. Primary ECs and RMICs from same mice were grown in a transwell co-culture system to address the possibility of paracrine signaling from RMICs via endothelial IPs. RMICs in the transwell insert were pre-incubated in media of high osmolality (630 mOsm/kg H2O) to induce COX-2 expression, and intracellular cAMP (as a surrogate measure of IP activation) was quantified in ECs grown in the bottom chamber. Again, all PGs released from RMICs from WT in the co-culture medium were augmented significantly by hypertonic stress. COX-1 replacement failed to recapitulate PGI2 induction (measured as 6-keto-PGF1α) in RMICs from COX-1>COX-2 mice (Online Figure VI A). Correspondingly, cAMP production of WT ECs was increased by 2.8-fold in response to stressed RIMCs (from 3.6±0.27 to 9.8±1.48 pmol/105 cells, n=6, Online Figure VI B), while only minimal cAMP in ECs from COX-1>COX-2 mice was detected (0.22±0.01 to 0.25±0.03 pmol/105 cells, Online Figure VI B) and that was not increased by high osmolality. The cAMP response of ECs to cicaprost, did not differ between WT and mutant mice, demonstrating the capacity of the IP to generate the measured signal. These observations are consistent with aberrant IP receptor signaling in the vasa recta in COX-1>COX-2 mice.

Figure 6. COX protein expression and PG production by cultured RMICs from COX-1>COX-2 and WT mice.

Figure 6

Open in a new tab

RMICs grown to sub-confluence were subjected to gradual changes in media osmolality from 330 to 630 mOsm/kg H2O. After 24 h, the cells were incubated with arachidonic acid (20 μM) for 15 min. The cell lysates were prepared for Western blot of COX-1 and COX-2 (A) and the supernatants underwent PG analysis (B) by mass spectrometry. *, p<0.05 vs WT; #, p<0.01 vs 330 mOsm/kg H2O group; n=6, repeated 3 times.

Natriuresis evoked by activation of the renal medullary IP receptor

The IP agonist cicaprost (10 ng/20 μl/h) significantly and progressively increased urine volume (UV) from 0.69 ± 0.09 μl/min (basal level-before infusion) to 1.67 ± 0.10μl/min (p<0.01, Figure 7A) and urinary sodium excretion (UNaV) from 0.095± 0.002μEq/min (basal level) to 0.1688± 0.005μEq/min (p<0.01, Figure 7B) over two hours of direct intramedullary infusion. In contrast, no significant effects on urine volume were observed over the same period during infusion of vehicle control (0.67± 0.12μl/min vs 0.74± 0.05μl/min, p=n.s., n=4). Thus, activation of the renal medullary IP is capable of evoking natriuresis.

Figure 7. Activation of the IP by cicaprost evokes natriuresis in anesthetized mice.

Figure 7

Open in a new tab

The right kidney was removed 1 wk before the experiment. The mice were anesthetized and catheterized as described in Methods. After 1 hr equilibration, cicaprost (10 ng/h) was infused into the renal medulla over 120 min. Urine was collected every 30 min before and during cicaprost infusion. Urine volume (UV, A) and urinary sodium excretion (UNaV, B) were determined. Values presented as means ± SEM. *, p<0.05 vs basal level, n=6.

Discussion

Both tNSAIDs and NSAIDs purposefully developed to inhibit selectively COX-2 (pdNSAIDs) can elevate BP3. Vasodilator PGs – such as PGI2 and PGE2- are critical to the preservation of renal blood flow in renoprival conditions4, 25 and COX-2 is the dominant source of biosynthesis of PGI2, as assessed by excretion of its major urinary metabolite in humans and in mice 11, 36. Deletion of the IP results in salt sensitive hypertension19, 20 and an increased sensitivity to thrombogenic stimuli11, while mutations of the IP and PGI2 synthase have been associated with hypertension and cardiovascular events in humans37, 38. Both COX-2 and COX-1 contribute to PGE2 formation11 and deletion of at least one of the EPs – the EP2 - coupled, like the IP, to adenylate cyclase activation34, also results in salt sensitive hypertension 21, 22. By contrast, activation of other PGE2 receptors - EP1 and EP3 - elevates BP39, 40 and activates platelets41 respectively in rodents. Thus, PGI2 and PGE2 differ with respect to their relative derivation from COX-2 and their receptor dependent impact on cardiovascular function. Deletion and pharmacological inhibition of the COX isozymes suggests that their products have opposite effects on BP10, just as in hemostasis3; thus, the likelihood of a hypertensive response to any NSAID may relate both to inhibition of COX-2 and the selectivity with which it is attained12. Large scale clinical trials to address this hypothesis have not been performed. An overview analysis of 19 trials suggested that hypertension was more likely with pdNSAIDs than tNSAIDs. However, these trials were small and heterogeneous with respect to both groups of NSAIDs. Also, many tNSAIDs, such as diclofenac and meloxicam, are similarly selective to the pdNSAID, celecoxib, for inhibition of COX-242.

The COX isozymes differ quite dramatically with respect to their transcriptional regulation1. These COX-1>COX-2 mice23 permit assessment of whether harmonization of regulatory control allows rescue of COX-2 deficiency by COX-1. Our results suggest that COX-2 plays a unique role in maintaining BP homeostasis and that this relates to its preferential linkage to biosynthesis of renal medullary PGI2.

The hypertensive response to NSAIDs is quite heterogeneous in humans43 and is markedly influenced by genetic strain in mice44. The precise genetic modifiers of this response remain to be identified. Here we studied mice on a mixed C57BL/6 × Sv129 background, that exhibit hypertension after pharmacological inhibition or genetic deletion of COX-2 when maintained on regular chow11. COX-2 plays important roles in development26 and mice deficient in COX-2 typically exhibit renal hypoplasia and biochemical evidence of renal compromise. Here COX-1 can rescue the developmental effects, hypertension and hyporeninemia of COX-2 KOs under such physiological conditions. By contrast, COX-1 was unable to compensate for the impact of COX-2 deletion when the mice were exposed to the hypertensive stimulus of a HSD.

This response to a HSD reflects a selective impact on the medullary COX-2/PGI2 biosynthetic/response pathway that mediates natriuresis. Evidence in the present study suggesting that COX-2 is uniquely coupled to synthesis of PGI2 derives from metabolite measurement in vivo and estimates of the capacity to generate PGs in both kidney medulla and RIMCs. Previous observations that PGI2 was preferentially produced through COX-2 in macrophages45, ECs46 and COX/PGI synthase co-transfected HEK293 cells47 are also consistent with this notion. COX-1 cannot compensate for the failure to augment PGI2 formation in response to a HSD. This effect is likely amplified by the unexpected depression by the HSD of IP transcript and protein, uniquely amongst the PG receptors. Indeed, COX-2 deficiency in the mutant mice may have further augmented this effect, attaining significance at the level of IP mRNA.

Renal medullary COX-2 has previously been implicated in buffering the hypertensive response to dietary salt intake 17, 18. Both COX isoforms are abundant in medulla, but they are distributed differentially. COX-1 is expressed in both the medullary collecting duct (MCD) and in RMICs10, 14, 15. COX-2 is most prominent in RMICs, which are involved in the maintenance of the medullary microcirculation and urinary salt excretion10. Although the renal papilla receives less than 1% of total renal blood flow, primarily through the descending vasa recta48, medullary blood flow promotes renal salt excretion, so called pressure-natriuresis48.

Deletion of either the IP19, 20 or EP2 receptors21, 22 results in salt sensitive hypertension, while deletion of the EP140 and PGF2α receptor (FP)49 reduces BP. IP receptors are expressed abundantly in the renal vasculature. IPs in the vasa recta34, 35, can indirectly modulate sodium reabsorption by altering blood flow to the adjacent medullary thick ascending limbs and collecting ducts50, 51. Expression of the EP2 is also evident in the vasa recta52, where it too plays an important role in maintaining fluid and electrolyte balance and BP21, 22, 25. Indeed, PGE2 is the most abundant PG product in the measurements of biosynthetic capacity in renal medulla and RIMCs. However, unlike PGI2, both systemic and renal medullary synthesis of PGE2 can be rescued by COX-1 and rescue of PGE2 is much more efficient than of PGI2 in RIMCs. Thus, PGI2 is the COX-2 product more directly implicated in the preservation of BP homeostasis in response to dietary salt.

Placebo controlled trials have revealed a cardiovascular hazard attributable to pdNSAIDs3. Although a predisposition to thrombosis is dominant, hypertension and cardiac failure are also evident in the spectrum of this cardiovascular risk53. Multiple lines of evidence suggest that the risk of thrombosis is attributable to suppression of COX-2 dependent PGI224, 54. The current results integrate this mechanism with a predisposition to hypertension and provide in vivo evidence to support the notion that COX-2 is preferentially coupled to biosynthesis of PGI2.

In summary, these results suggest that dysregulated expression of the COX-2 dependent, PGI2 biosynthesis/response pathway in the renal internal medulla (IM) undermines the homeostatic response of BP to a HSD. Inhibition of this pathway may contribute directly to the hypertensive response to both pdNSAIDs and tNSAIDs in humans.

Supplementary Material

Supp1

Acknowledgments

Sources of funding.

This study was supported by grants from the American Heart Association--Jon Holden DeHaan Scientist Development Grant 0730314N, the National Heart Lung and Blood Institute (HL62250 and HL066233) and the Canadian Institutes of Health Research (MOP-79459). G.A.F. is the McNeill Professor in Translational Medicine and Therapeutics. C.D.F. holds a Tier I Canada Research Chair in Molecular, Cellular and Physiological Medicine and is recipient of a Career Investigator Award from the Heart and Stroke Foundation of Ontario.

Non-standard Abbreviation and Acronyms

PG

Prostanoid

NSAIDs

Nonsteroidal anti-inflammatory drugs

BP

Blood pressure

HSD

High salt diet

IM

Renal inner renal medulla

RMIC

Renal medullary interstital cell

EC

Endothelial cell

Footnotes

Disclosures

G.A.F. is a consultant to Logical Therapeutics, receives support for an investigator initiated study from Crystal Genomics and serves of the scientific advisory board of Nicox, all of which companies have an interest in NSAIDs.

References

  • 1.Smith WL, DeWitt DL, Garavito RM. Cyclooxygenases: structural, cellular, and molecular biology. Annu Rev Biochem. 2000;69:145–182. doi: 10.1146/annurev.biochem.69.1.145. [DOI] [PubMed] [Google Scholar]
  • 2.Armstrong EP, Malone DC. The impact of nonsteroidal anti-inflammatory drugs on blood pressure, with an emphasis on newer agents. Clin Ther. 2003;25:1–18. doi: 10.1016/s0149-2918(03)90003-8. [DOI] [PubMed] [Google Scholar]
  • 3.FitzGerald GA. COX-2 and beyond: Approaches to prostaglandin inhibition in human disease. Nat Rev Drug Discov. 2003;2:879–890. doi: 10.1038/nrd1225. [DOI] [PubMed] [Google Scholar]
  • 4.Fujino T, Nakagawa N, Yuhki K, Hara A, Yamada T, Takayama K, Kuriyama S, Hosoki Y, Takahata O, Taniguchi T, Fukuzawa J, Hasebe N, Kikuchi K, Narumiya S, Ushikubi F. Decreased susceptibility to renovascular hypertension in mice lacking the prostaglandin I2 receptor IP. J Clin Invest. 2004;114:805–812. doi: 10.1172/JCI21382. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Whelton A, White WB, Bello AE, Puma JA, Fort JG. Effects of celecoxib and rofecoxib on blood pressure and edema in patients > or =65 years of age with systemic hypertension and osteoarthritis. Am J Cardiol. 2002;90:959–963. doi: 10.1016/s0002-9149(02)02661-9. [DOI] [PubMed] [Google Scholar]
  • 6.Whelton A. Nephrotoxicity of nonsteroidal anti-inflammatory drugs: physiologic foundations and clinical implications. Am J Med. 1999;106:13S–24S. doi: 10.1016/s0002-9343(99)00113-8. [DOI] [PubMed] [Google Scholar]
  • 7.Catella-Lawson F, McAdam B, Morrison BW, Kapoor S, Kujubu D, Antes L, Lasseter KC, Quan H, Gertz BJ, FitzGerald GA. Effects of specific inhibition of cyclooxygenase-2 on sodium balance, hemodynamics, and vasoactive eicosanoids. J Pharmacol Exp Ther. 1999;289:735–741. [PubMed] [Google Scholar]
  • 8.Athirakul K, Kim HS, Audoly LP, Smithies O, Coffman TM. Deficiency of COX-1 causes natriuresis and enhanced sensitivity to ACE inhibition. Kidney Int. 2001;60:2324–2329. doi: 10.1046/j.1523-1755.2001.00072.x. [DOI] [PubMed] [Google Scholar]
  • 9.Francois H, Athirakul K, Mao L, Rockman H, Coffman TM. Role for thromboxane receptors in angiotensin-II-induced hypertension. Hypertension. 2004;43:364–369. doi: 10.1161/01.HYP.0000112225.27560.24. [DOI] [PubMed] [Google Scholar]
  • 10.Qi Z, Hao CM, Langenbach RI, Breyer RM, Redha R, Morrow JD, Breyer MD. Opposite effects of cyclooxygenase-1 and -2 activity on the pressor response to angiotensin II. J Clin Invest. 2002;110:61–69. doi: 10.1172/JCI14752. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Cheng Y, Wang M, Yu Y, Lawson J, Funk CD, Fitzgerald GA. Cyclooxygenases, microsomal prostaglandin E synthase-1, and cardiovascular function. J Clin Invest. 2006;116:1391–1399. doi: 10.1172/JCI27540. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.FitzGerald GA. The choreography of cyclooxygenases in the kidney. J Clin Invest. 2002;110:33–34. doi: 10.1172/JCI16044. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Aw TJ, Haas SJ, Liew D, Krum H. Meta-analysis of cyclooxygenase-2 inhibitors and their effects on blood pressure. Arch Intern Med. 2005;165:490–496. doi: 10.1001/archinte.165.5.IOI50013. [DOI] [PubMed] [Google Scholar]
  • 14.Campean V, Theilig F, Paliege A, Breyer M, Bachmann S. Key enzymes for renal prostaglandin synthesis: site-specific expression in rodent kidney (rat, mouse) Am J Physiol Renal Physiol. 2003;285:F19–32. doi: 10.1152/ajprenal.00443.2002. [DOI] [PubMed] [Google Scholar]
  • 15.Yang T, Singh I, Pham H, Sun D, Smart A, Schnermann JB, Briggs JP. Regulation of cyclooxygenase expression in the kidney by dietary salt intake. Am J Physiol. 1998;274:F481–489. doi: 10.1152/ajprenal.1998.274.3.F481. [DOI] [PubMed] [Google Scholar]
  • 16.Hao CM, Yull F, Blackwell T, Komhoff M, Davis LS, Breyer MD. Dehydration activates an NF-kappaB-driven, COX2-dependent survival mechanism in renal medullary interstitial cells. J Clin Invest. 2000;106:973–982. doi: 10.1172/JCI9956. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Ye W, Zhang H, Hillas E, Kohan DE, Miller RL, Nelson RD, Honeggar M, Yang T. Expression and function of COX isoforms in renal medulla: evidence for regulation of salt sensitivity and blood pressure. Am J Physiol Renal Physiol. 2006;290:F542–549. doi: 10.1152/ajprenal.00232.2005. [DOI] [PubMed] [Google Scholar]
  • 18.Zewde T, Mattson DL. Inhibition of cyclooxygenase-2 in the rat renal medulla leads to sodium-sensitive hypertension. Hypertension. 2004;44:424–428. doi: 10.1161/01.HYP.0000140924.91479.03. [DOI] [PubMed] [Google Scholar]
  • 19.Watanabe H, Katoh T, Eiro M, Iwamoto M, Ushikubi F, Narumiya S, Watanabe T. Effects of salt loading on blood pressure in mice lacking the prostanoid receptor gene. Circ J. 2005;69:124–126. doi: 10.1253/circj.69.124. [DOI] [PubMed] [Google Scholar]
  • 20.Francois H, Athirakul K, Howell D, Dash R, Mao L, Kim HS, Rockman HA, Fitzgerald GA, Koller BH, Coffman TM. Prostacyclin protects against elevated blood pressure and cardiac fibrosis. Cell Metab. 2005;2:201–207. doi: 10.1016/j.cmet.2005.08.005. [DOI] [PubMed] [Google Scholar]
  • 21.Kennedy CR, Zhang Y, Brandon S, Guan Y, Coffee K, Funk CD, Magnuson MA, Oates JA, Breyer MD, Breyer RM. Salt-sensitive hypertension and reduced fertility in mice lacking the prostaglandin EP2 receptor. Nat Med. 1999;5:217–220. doi: 10.1038/5583. [DOI] [PubMed] [Google Scholar]
  • 22.Tilley SL, Audoly LP, Hicks EH, Kim HS, Flannery PJ, Coffman TM, Koller BH. Reproductive failure and reduced blood pressure in mice lacking the EP2 prostaglandin E2 receptor. J Clin Invest. 1999;103:1539–1545. doi: 10.1172/JCI6579. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Yu Y, Fan J, Hui Y, Rouzer CA, Marnett LJ, Klein-Szanto AJ, FitzGerald GA, Funk CD. Targeted cyclooxygenase gene (ptgs) exchange reveals discriminant isoform functionality. J Biol Chem. 2007;282:1498–1506. doi: 10.1074/jbc.M609930200. [DOI] [PubMed] [Google Scholar]
  • 24.Grosser T, Fries S, FitzGerald GA. Biological basis for the cardiovascular consequences of COX-2 inhibition: therapeutic challenges and opportunities. J Clin Invest. 2006;116:4–15. doi: 10.1172/JCI27291. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Chen J, Zhao M, He W, Milne GL, Howard JR, Morrow J, Hebert RL, Breyer RM, Hao CM. Increased dietary NaCl induces renal medullary PGE2 production and natriuresis via the EP2 receptor. Am J Physiol Renal Physiol. 2008;295:F818–825. doi: 10.1152/ajprenal.90253.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Yu Y, Fan J, Chen XS, Wang D, Klein-Szanto AJ, Campbell RL, FitzGerald GA, Funk CD. Genetic model of selective COX2 inhibition reveals novel heterodimer signaling. Nat Med. 2006;12:699–704. doi: 10.1038/nm1412. [DOI] [PubMed] [Google Scholar]
  • 27.Yang T, Endo Y, Huang YG, Smart A, Briggs JP, Schnermann J. Renin expression in COX-2-knockout mice on normal or low-salt diets. Am J Physiol Renal Physiol. 2000;279:F819–825. doi: 10.1152/ajprenal.2000.279.5.F819. [DOI] [PubMed] [Google Scholar]
  • 28.Cheng HF, Wang JL, Zhang MZ, Wang SW, McKanna JA, Harris RC. Genetic deletion of COX-2 prevents increased renin expression in response to ACE inhibition. Am J Physiol Renal Physiol. 2001;280:F449–456. doi: 10.1152/ajprenal.2001.280.3.F449. [DOI] [PubMed] [Google Scholar]
  • 29.Yu Y, Cheng Y, Fan J, Chen XS, Klein-Szanto A, Fitzgerald GA, Funk CD. Differential impact of prostaglandin H synthase 1 knockdown on platelets and parturition. J Clin Invest. 2005;115:986–995. doi: 10.1172/JCI23683. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Song WL, Wang M, Ricciotti E, Fries S, Yu Y, Grosser T, Reilly M, Lawson JA, FitzGerald GA. Tetranor PGDM, an abundant urinary metabolite reflects biosynthesis of prostaglandin D2 in mice and humans. J Biol Chem. 2008;283:1179–1188. doi: 10.1074/jbc.M706839200. [DOI] [PubMed] [Google Scholar]
  • 31.FitzGerald GA, Pedersen AK, Patrono C. Analysis of prostacyclin and thromboxane biosynthesis in cardiovascular disease. Circulation. 1983;67:1174–1177. doi: 10.1161/01.cir.67.6.1174. [DOI] [PubMed] [Google Scholar]
  • 32.Mannon RB, Coffman TM, Mannon PJ. Distribution of binding sites for thromboxane A2 in the mouse kidney. Am J Physiol. 1996;271:F1131–1138. doi: 10.1152/ajprenal.1996.271.6.F1131. [DOI] [PubMed] [Google Scholar]
  • 33.Hao CM, Komhoff M, Guan Y, Redha R, Breyer MD. Selective targeting of cyclooxygenase-2 reveals its role in renal medullary interstitial cell survival. Am J Physiol. 1999;277:F352–359. doi: 10.1152/ajprenal.1999.277.3.F352. [DOI] [PubMed] [Google Scholar]
  • 34.Hao CM, Breyer MD. Physiological regulation of prostaglandins in the kidney. Annu Rev Physiol. 2008;70:357–377. doi: 10.1146/annurev.physiol.70.113006.100614. [DOI] [PubMed] [Google Scholar]
  • 35.Oida H, Namba T, Sugimoto Y, Ushikubi F, Ohishi H, Ichikawa A, Narumiya S. In situ hybridization studies of prostacyclin receptor mRNA expression in various mouse organs. Br J Pharmacol. 1995;116:2828–2837. doi: 10.1111/j.1476-5381.1995.tb15933.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.McAdam BF, Catella-Lawson F, Mardini IA, Kapoor S, Lawson JA, FitzGerald GA. Systemic biosynthesis of prostacyclin by cyclooxygenase (COX)-2: the human pharmacology of a selective inhibitor of COX-2. Proc Natl Acad Sci U S A. 1999;96:272–277. doi: 10.1073/pnas.96.1.272. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Arehart E, Stitham J, Asselbergs FW, Douville K, MacKenzie T, Fetalvero KM, Gleim S, Kasza Z, Rao Y, Martel L, Segel S, Robb J, Kaplan A, Simons M, Powell RJ, Moore JH, Rimm EB, Martin KA, Hwa J. Acceleration of cardiovascular disease by a dysfunctional prostacyclin receptor mutation: potential implications for cyclooxygenase-2 inhibition. Circ Res. 2008;102:986–993. doi: 10.1161/CIRCRESAHA.107.165936. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Nakayama T. Prostacyclin synthase gene: genetic polymorphisms and prevention of some cardiovascular diseases. Curr Med Chem Cardiovasc Hematol Agents. 2005;3:157–164. doi: 10.2174/1568016053544327. [DOI] [PubMed] [Google Scholar]
  • 39.Guan Y, Zhang Y, Wu J, Qi Z, Yang G, Dou D, Gao Y, Chen L, Zhang X, Davis LS, Wei M, Fan X, Carmosino M, Hao C, Imig JD, Breyer RM, Breyer MD. Antihypertensive effects of selective prostaglandin E2 receptor subtype 1 targeting. J Clin Invest. 2007;117:2496–2505. doi: 10.1172/JCI29838. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Stock JL, Shinjo K, Burkhardt J, Roach M, Taniguchi K, Ishikawa T, Kim HS, Flannery PJ, Coffman TM, McNeish JD, Audoly LP. The prostaglandin E2 EP1 receptor mediates pain perception and regulates blood pressure. J Clin Invest. 2001;107:325–331. doi: 10.1172/JCI6749. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Ma H, Hara A, Xiao CY, Okada Y, Takahata O, Nakaya K, Sugimoto Y, Ichikawa A, Narumiya S, Ushikubi F. Increased bleeding tendency and decreased susceptibility to thromboembolism in mice lacking the prostaglandin E receptor subtype EP(3) Circulation. 2001;104:1176–1180. doi: 10.1161/hc3601.094003. [DOI] [PubMed] [Google Scholar]
  • 42.Capone ML, Tacconelli S, Di Francesco L, Sacchetti A, Sciulli MG, Patrignani P. Pharmacodynamic of cyclooxygenase inhibitors in humans. Prostaglandins Other Lipid Mediat. 2007;82:85–94. doi: 10.1016/j.prostaglandins.2006.05.019. [DOI] [PubMed] [Google Scholar]
  • 43.Johnson AG, Nguyen TV, Day RO. Do nonsteroidal anti-inflammatory drugs affect blood pressure? A meta-analysis. Ann Intern Med. 1994;121:289–300. doi: 10.7326/0003-4819-121-4-199408150-00011. [DOI] [PubMed] [Google Scholar]
  • 44.Yang T, Huang YG, Ye W, Hansen P, Schnermann JB, Briggs JP. Influence of genetic background and gender on hypertension and renal failure in COX-2-deficient mice. Am J Physiol Renal Physiol. 2005;288:F1125–1132. doi: 10.1152/ajprenal.00219.2004. [DOI] [PubMed] [Google Scholar]
  • 45.Brock TG, McNish RW, Peters-Golden M. Arachidonic acid is preferentially metabolized by cyclooxygenase-2 to prostacyclin and prostaglandin E2. J Biol Chem. 1999;274:11660–11666. doi: 10.1074/jbc.274.17.11660. [DOI] [PubMed] [Google Scholar]
  • 46.Caughey GE, Cleland LG, Penglis PS, Gamble JR, James MJ. Roles of cyclooxygenase (COX)-1 and COX-2 in prostanoid production by human endothelial cells: selective up-regulation of prostacyclin synthesis by COX-2. J Immunol. 2001;167:2831–2838. doi: 10.4049/jimmunol.167.5.2831. [DOI] [PubMed] [Google Scholar]
  • 47.Ueno N, Murakami M, Tanioka T, Fujimori K, Tanabe T, Urade Y, Kudo I. Coupling between cyclooxygenase, terminal prostanoid synthase, and phospholipase A2. J Biol Chem. 2001;276:34918–34927. doi: 10.1074/jbc.M100429200. [DOI] [PubMed] [Google Scholar]
  • 48.Cowley AW, Jr, Mattson DL, Lu S, Roman RJ. The renal medulla and hypertension. Hypertension. 1995;25:663–673. doi: 10.1161/01.hyp.25.4.663. [DOI] [PubMed] [Google Scholar]
  • 49.Yu Y, Lucitt MB, Stubbe J, Cheng Y, Friis UG, Hansen PB, Jensen BL, Smyth EM, FitzGerald GA. Prostaglandin F2alpha elevates blood pressure and promotes atherosclerosis. Proc Natl Acad Sci U S A. 2009;106:7985–7990. doi: 10.1073/pnas.0811834106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Romero JC, Bentley MD, Vanhoutte PM, Knox FG. Intrarenal mechanisms that regulate sodium excretion in relationship to changes in blood pressure. Mayo Clin Proc. 1989;64:1406–1424. doi: 10.1016/s0025-6196(12)65383-x. [DOI] [PubMed] [Google Scholar]
  • 51.Pallone TL, Silldorff EP, Turner MR. Intrarenal blood flow: microvascular anatomy and the regulation of medullary perfusion. Clin Exp Pharmacol Physiol. 1998;25:383–392. doi: 10.1111/j.1440-1681.1998.tb02220.x. [DOI] [PubMed] [Google Scholar]
  • 52.Therland KL, Stubbe J, Thiesson HC, Ottosen PD, Walter S, Sorensen GL, Skott O, Jensen BL. Cycloxygenase-2 is expressed in vasculature of normal and ischemic adult human kidney and is colocalized with vascular prostaglandin E2 EP4 receptors. J Am Soc Nephrol. 2004;15:1189–1198. doi: 10.1097/01.asn.0000124673.79934.24. [DOI] [PubMed] [Google Scholar]
  • 53.FitzGerald GA. COX-2 in play at the AHA and the FDA. Trends Pharmacol Sci. 2007;28:303–307. doi: 10.1016/j.tips.2007.05.007. [DOI] [PubMed] [Google Scholar]
  • 54.Seta F, Chung AD, Turner PV, Mewburn JD, Yu Y, Funk CD. Renal and cardiovascular characterization of COX-2 knockdown mice. Am J Physiol Regul Integr Comp Physiol. 2009;296:R1751–1760. doi: 10.1152/ajpregu.90985.2008. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp1
NIHMS166933-supplement-Supp1.pdf (448.3KB, pdf)

RESOURCES