Thromboxane prostanoid receptors enhance contractions, endothelin-1 and oxidative stress in microvessels from mice with CKD

Abstract

Cardiovascular disease (CVD) is frequent in chronic kidney disease (CKD) and has been related to angiotensin II (ANG II), endothelin-1 (ET-1), thromboxane A₂ (TxA₂) and reactive oxygen species (ROS). Since activation of thromboxane prostanoid receptors (TP-Rs) can generate ROS which can generate ET-1, we tested the hypothesis that CKD induces cyclooxygenase (COX)-2 whose products activate TP-Rs to enhance ET-1 and ROS generation and contractions. Mesenteric resistance arterioles were isolated from C57/BL6, or TP-R +/+ and TP-R −/− mice 3 months after SHAM-operation (SHAM) or surgical reduced renal mass (RRM, n=6/group). Microvascular contractions were studied on a wire myograph. Cellular (ethidium: dihydroethidium) and mitochondrial (mitoSOX) ROS were measured by fluorescence microscopy. Mice with RRM had increased excretion of markers of oxidative stress, thromboxane, and microalbumin, increased plasma ET-1 and increased microvascular expression of p22^phox, COX-2, TP-Rs, preproendothelin and endothelin-A receptors and increased arteriolar remodeling. They had increased contractions to U-46,619 (118±3 vs. 87±6, P<0.05) and ET-1 (108±5 vs. 89±4, P<0.05), which were dependent on cellular and mitochondrial ROS, COX-2, and TP-Rs. RRM doubled the ET-1-induced cellular and mitochondrial ROS generation (P<0.05). TP-R −/− mice with RRM lacked these abnormal structural and functional microvascular responses and lacked the increased systemic and the increased microvascular oxidative stress and circulating ET-1. In conclusion, RRM leads to microvascular remodeling and enhanced ET-1-induced cellular and mitochondrial ROS and contractions that are mediated by COX-2 products activating TP-Rs. Thus, TP-Rs can be upstream from enhanced ROS, ET-1, microvascular remodeling and contractility and may thereby coordinate vascular dysfunction in CKD.

Introduction

Although chronic kidney disease (CKD) is progressive, most patients die from cardiovascular disease (CVD) before reaching end stage renal disease (ESRD)¹. CVD in patients with CKD is unusual since it is not clearly associated with hypertension, hypercholesterolemia or obesity (unless extreme)^2–5. Uncertainty concerning the underlying vascular mechanisms has hampered the development of strategies to prevent the CVD.

Patients with CKD have vascular endothelial dysfunction and oxidative stress that can be detected in CKD stage 1^{6, 7}. These changes can be produced by angiotensin II (ANG II) acting on angiotensin type I receptors, endothelin-1 (ET-1) acting on type A receptors (ETA-Rs) or cyclooxygenase (COX) dependent prostaglandins (PGs), and thromboxane A₂ (TxA₂) acting on thromboxane-prostanoid receptors (TP-Rs)^8–11. TxA₂ has been implicated in the hypertension, renal vasoconstriction and glomerulosclerosis of rats with CKD¹² has been modeled in mice with surgically reduced renal mass (RRM)^{13, 14}. This C57/BL6 mouse model has severe oxidative stress but maintains a normal conscious mean blood pressure (MBP) measured telemetrically over 3 months and has only a modest reduction in glomerular filtration rate (GFR) over 3 months¹⁴. Therefore, it is a convenient model to study early CKD without the confounding effects of hypertension or uremia. Moreover, TP-R −/− mice maintain a normal MBP measured telemetrically and basal phenotype which simplifies interpretation of results with this strain^{11, 15}

Reactive oxygen species (ROS) upregulate COX-2 in the vessel wall⁹. Moreover, ROS react with nitric oxide (NO) to generate peroxynitrite that irreversibly inactivates prostacyclin synthase thereby redirecting COX-products to vasoconstrictor actions¹⁶. Indeed, COX-2 generates endoperoxides (PGH₂) and TxA₂ that activate TP-Rs in vascular smooth muscle cells (VSMCs) that, during increased ROS, mediate an endothelium-dependent contracting factor (EDCF) response^{9, 17}. Thus, TP-R signaling can be downstream from increased ROS. Surprisingly, however, genetic deletion of TP-Rs prevented an increased excretion of the oxidative stress marker 8-isoprostane F_2α in mice infused with ANGII¹¹. Thus, TP-R signaling also can be upstream of increased ROS. ROS¹⁸ and CKD¹⁹ both increase ET-1 generation. However, the mechanisms interlinking TP-R activation, ROS and ET-1 generation are unclear²⁰. Therefore, we tested the hypothesis that ROS and ET-1generated in microvessels of mice with RRM depend on COX-2 generation of PGs and/or TxA₂ that activate TP-Rs. These studies have translational impact because drugs that block TP-Rs are already in clinical trials with promising responses²¹. If they were found to reduce vascular contractility, remodeling, ROS and ET-1 generation in CKD, they could fill a therapeutic void to prevent CVD which is now the principal cause of death and disability in these patients.

Methods

Animal Preparation and Surgery

Male C57BL6 mice weighing 25 to 30 g (Charles River Laboratory, Germantown, MD), TP-R knockout (−/−), and TP-R wild-type (TP+/+) mice (C57/BL6J background, kind gift from Thomas Coffman, MD, Duke University, Chapel Hill, NC) were maintained on tap water and standard chow (Na⁺ content 0.4 g ·100g⁻¹, Harlan Teklad) and allowed free access to tap water. All of the procedures conformed to the Guide for Care and Use of Laboratory Animals by the NIH Institute for Laboratory Animal Research, and were approved by the Georgetown University Animal Care and Use Committee.

A 2-step surgical 5/6 nephrectomy procedure was used to create RRM under inhalational anesthesia with 2% isoflurane and oxygen mixed with room air in a vaporizer^{13, 14}. Approximately, two thirds of the mass of the left kidney was ablated by stitching off each pole using an absorbable hemostat (Ethicon, Inc.). At a second surgery 1 week later, the right kidney was removed. SHAM-operated control mice (SHAM) were subjected to a similar 2-stage procedure without the removal of renal mass. Mice were studied after 3 months, at which time they had considerable oxidative stress with a 7-fold increase in excretion of 8-isoprostane F_2α but an unchanged BP and only mild albuminuria, glomerulosclerosis, tubulointerstitial fibrosis, and a 33% reduction in measured overall glomerular filtration rate (GFR)^{13, 14}.

Measurement of plasma endothelin 1 concentration

Plasma concentrations of ET-1 were measured using a Quantikine ELISA kit (R&D Systems, Minneapolis, MN, USA²².

Measurement of urinary 8-isoprostane F_2α (8-Iso), thromboxane B₂ (TxB₂), malondialdehyde (MDA), microalbumin, and creatinine

Mice were housed in metabolic cages (Nalgene Nunc International, Rochester, NY). Urine was collected for 24 hours into tubes containing antibiotics as described^{23, 24}. 8-Iso and TxB₂ in urine were quantified by ELISA (Enzo Life Science Inc. Farmingdale, NY) after purification, extraction, and measurement of individual recovery by spiking with radiolabelled PGE₂, as described and validated previously against GC/MS²⁴. MDA was measured by an assay kit (Cayman Chemical Company, Ann Arbor, MI)¹⁷ and microalbumin by an ELISA kit (Exocell, Philadelphia, PA). Values were normalized with creatinine, which was measured by a urinary creatinine assay kit (Exocell, Philadelphia, PA) (see Online Data Supplement).

Protein expression of mesenteric resistance arterioles

The expression of ETA-R, p22^phox, p47^phox, COX-1 and -2, and TP-R of mesenteric resistance arterioles were quantified using specific antibodies as described^{13, 14, 23, 25, 26}. (details see Online Data Supplement)

RNA isolation and real-time quantitative RT-PCR

Total RNA isolation and real-time quantitative PCR was performed as previously described with some modifications²⁷. (details see Online Data Supplement)

Preparation and study of mesenteric resistance arterioles

Vessels (mean luminal diameter 125±3 µm and length 2 mm) were separated from the superior mesenteric bed, mounted in a wire myograph (M610, Danish myotechnology A/S; Aarhus, Denmark) and studied as described²⁸ (see Online Data Supplement).

Experimental protocol

The vascular and luminal areas were measured as described^{28, 29}. Concentration-response curves to phenylephrine (PE: 10⁻⁸ to 10⁻⁵ mol·L⁻¹, α-adrenoceptor agonist), U-46,619 (10⁻⁹ to 10⁻⁶ mol·L⁻¹, TP-Rs agonist,) and endothelin −1 (ET-1, 10⁻¹⁰ to 10⁻⁷ mol·L) were compared to vehicle. Since PE-mediated contractions were not enhanced in RRM, ET-1 was selected for further studies.

To examine the functional effects of cellular and mitochondrial ROS, vessels were incubated with vehicle, the membrane permeable redox cycling nitroxide tempol (10⁻⁴ mol·L⁻¹, Sigma, St. Louis, MO), or the mitochondrial accumulated form MitoTEMPO (10⁻⁵ mol·L⁻¹, Sigma, St. Louis, MO)³⁰ for 30 minutes prior to testing with ET-1 (10⁻⁷ mol·L⁻¹).

To examine the role of PGs and TP-Rs, concentration-dependent responses to ET-1 (10⁻⁷ mol·L⁻¹) were obtained after incubation (30 min) with either vehicle, SC-560 (10⁻⁶ mol·L⁻¹; SC, an inhibitor of COX-1, Sigma, St. Louis), paracoxib (10⁻⁵ mol·L⁻¹; Para, an inhibitor of COX-2, Sigma, St. Louis, MO), SC+Para, OKY-046NA (10⁻⁵ mol·L⁻¹; an inhibitor of TxA₂ synthase (TxA₂-S), Sigma, St. Louis, MO), or SQ-29,548 (10⁻⁶ mol·L⁻¹; inhibitor of TP-Rs, Cayman Chemical Company, Ann Arbor, MI). These are fully effective concentrations⁷ (see online Data Supplement).

Endothelin-1-stimulated cellular and mitochondrial ROS in mesenteric resistance arterioles

ET-1 (10⁻⁷ mol·L⁻¹)-induced cellular and mitochondrial ROS production were determined after loading with dihydroethidium (DHE) or MitoSox™ Red as described^{9, 31} and fluorescence quantified by PTI RatioMaster™ (Photon Technology International, London, Ontario, Canada) (see Online Data Supplement).

Prolonged effect of TP-Rs in mice with RRM

RRM or SHAM models were created in TP-R +/+ and −/− mice and studied 3 months after the surgery. Plasma ET-1, 24 hour urinary 8-Iso, TxB₂, MDA, microalbumin, mesenteric arteriolar contractions to PE, ET-1, and U-46,619 and generation of cellular and mitochondrial ROS with ET-1 were obtained as described.

Chemicals and solutions

Agents^were purchased from Sigma (St. Louis, MI) and dissolved in physiologic salt solution (PSS).

Statistical analysis

Data are presented as mean ± SEM. Cumulative concentration-response experiments were analyzed by nonlinear regression (curve fit) for repeated measurement and differences assessed by two-way, repeated-measures ANOVA with interaction to assess the effects of RRM vs SHAM, intervention vs vehicle or TP-R −/− vs TP-R +/+ and the interaction (the effects of the intervention or genotype on the response to RRM). This was followed, if appropriate, with Bonferroni post hoc t-tests for multiple comparisons. A probability value <0.05 was considered statistically significant.

Results

Body, kidney and heart weights, mesenteric resistance arteriole remodeling, plasma endothelin-1and renal excretion of biomarkers (Table 1)

Table 1.

Basal parameters, plasma endothelin-1 and 24-hour urinary excretion in C57BL/6 mice: comparison of Sham operated and reduced renal mass mice

Parameter	SHAM	RRM	P
Body weight (BWt) (g)	32.6±0.8	33.1±0.9	NS
Kidney weight/BWt (mg·g−1)	6.6±0.2	7.8±0.3	<0.05
Heart weight/BWt (mg·g−1)	4.5±0.2	4.4±0.2	NS
Aorta weight/BWt (mg·g−1)	0.13±0.007	0.15±0.008	NS
Vessel lumen CSA (×1000 µm2)	99.6±31.3	75.4±15.0	NS
Vessel media CSA (×1000 µm2)	12.9±3.7	17.7±2.7	NS
Vessel media:lumen (µm2·µm−2)	0.17±0.02	0.27±0.01	<0.01
Plasma endothelin-1 (pg/ml)	1.6±0.3	3.4±0.3	<0.05
Urinary 8- isoprostane F2α /creatinine (ng·mg−1)	2.8±0.38	5.1±0.6	<0.05
Urinary malondialdehyde /creatinine (nmol·mg−1)	48.0±3.4	69.2±6.8	<0.05
Urinary TxB2/creatinine (ng·mg−1)	2.01±0.15	2.52±0.13	<0.05
Urinary microalbumin/creatinine (ug·mg−1)	7.3±0.3	12.6±2.0	<0.05

Mean ±SEM values, (n=6 /group). CSA: cross-section area

Mice with RRM had normal body, heart and aorta weights but, despite removal of two-thirds of the left kidney to create RRM, its mass at 3 months exceeded that of SHAM mice, as reported previously¹⁴. Mice with RRM had an increased media:lumen ratio of mesenteric resistance arterioles and increased plasma ET-1 and renal excretions of 8-Iso, MDA, TxB₂, and microalbumin (P<0.05).

Protein expression in mesenteric resistance arterioles (Table 2 and supplementary Figures Suppl 1, Suppl 2 and Suppl 3)

Table 2.

Expression of proteins and mRNA in mesenteric resistance arterioles from C57BL/6 mice

A	Proteins	SHAM (n=5)	RRM (n=5)	P value
	Protein	N=5	N=5
	Endothelin A-R	0.36±0.06	0.52±0.04	<0.01
	p47phox	0.37±0.04	0.30±0.04	NS
	p22phox	0.08±0.02	0.69±0.15	<0.01
	TP-R	0.30±0.04	0.52±0.04	<0.01
	COX-1	0.28±0.03	0.33±0.03	NS
	COX-2	0.42±0.12	1.20±0.27	<0.05

B
	mRNA	SHAM (n=5)	RRM (n=5)	P value
	Endothelin A-R	1.02±0.05	1.89±0.17	<0.01
	Preproendothelin-1	1.01±0.08	1.46±0.13	<0.05
	p22phox	1.01±0.05	2.42±0.13	<0.005
	TP-R	1.03±0.05	0.96±0.01	<0.01
	COX-1	0.28±0.03	0.33±0.03	NS
	COX-2	1.01±0.09	2.35±0.28	<0.05

Mean ±SEM values as fold relative to beta-actin for protein (see figures in supplements S1 to S3 for blots) or relative to 18s rRNA for mRNA (see figure in supplement S4). Endothelin A-R, endothelin type A receptor; TP-R, thromboxane-prosanoid receptor; COX, cyclooxygenase.

Vessels from mice with RRM had increased protein expression of ETA-R, p22^phox, COX-2, and TP-R but no significant change in p47^phox or COX-1.

Gene expression in mesenteric resistance arterioles (Table 2 and supplementary figure S4)

Vessels from mice with RRM had increased mRNA expression of ETA-R, preproendothelin-1, p22^phox and COX-2 but no significant change in COX-1 or TP-R.

Contractions and ROS generation with ET-1 in mesenteric resistance arterioles (Figure 1 and Table 3)

Figure 1.

Concentration-response relationships for phenylephrine (Panel A), U-46,619 (Panel B), endothelin-1 (Panel C), and cellular and mitochondrial reactive oxygen species generation with 10⁻⁷ mol·L⁻¹ endothelin-1 (Panel D) in mesenteric resistance arterioles from SHAM (open circles and broken lines or open boxes) and reduced renal mass mice (closed circles and continuous lines or shaded boxes). Comparing groups: *, P<0.05; **, P<0.01, ***, P<0.005. E:Dht; ethidium: dihydroethidium fluorescence ratio.

Table 3.

Maximum vascular contractions and ROS generation of mesenteric arterioles in C57BL/6 mice: Comparison of sham operated and reduced renal mass mice

Stimulus	SHAM	RRM	P
PE (10−5M) contraction (%)	67.8±3.8	48.1±3.4	<0.05
U-46,619 (10−6M) contraction (%)	87.1±5.9	118.0±2.6	<0.01
ET-1 (10−7M) contraction (%)	89.1±4.0	108.1±5.1	<0.05
ET-1 (10−7M) cellular ROS generation (Eth/DHE,Δf/f0)	0.13±0.03	0.33±0.09	<0.05
ET-1 (10−7M) mitochondrial ROS generation (mitoSOX, Δf/f0)	0.08±0.02	0.17±0.02	<0.05

Mean ± SEM values (n = 6 per group). PE, phenylephrine; ET-1, endothelin-1, ROS, reactive oxygen species

Vessels from mice with RRM had reduced contractions to PE, but increased contractions to U-46,619 and ET-1 and generated more cellular and mitochondrial ROS with 10⁻⁷ mol·L⁻¹ of ET-1 than SHAM mice.

Effect of metabolism of cellular or mitochondrial ROS or blockade of COXs, TxA₂-S, or TP-Rs on contractions to ET-1 in mesenteric resistance arterioles (Table 4 and Figure 2)

Table 4.

Effects of metabolism of cellular or mitochondrial ROS or antagonist of cyclooxygenase-1 or -2, thromboxane A₂ synthase or thromboxane-prostanoid receptors on endothelin-1 (10⁻⁷ mol·L⁻¹) contractions of mesenteric resistance arteries from C57BL/6 mice: comparison of sham operated and reduced renal mass mice

Treatment	Contraction (%)		By ANOVA, effects of :
	SHAM	RRM	RRM	Treatment	Interaction
Vehicle	89.2±4.1	108.8±5.1	P<0.01	P<0.05
Tempol	79.0±3.4	73.6±2.5	P<0.05	P<0.01	P<0.01
MitoTEMPO	92.2±6.7	85.1±1.5	P<0.05	P<0.05	P<0.05
SC-560	84.1±5.4	96.1.±2.3	P<0.05	NS	NS
Paracoxib	79.1.±3.3	88.2±3.5	P<0.05	P<0.05	NS
SC-560+Paracoxib	83.1±5.1	55.6±7.8	P<0.05	P<0.01	P<0.05
OKY-046NA	85.1±4.3	94.4.±3.4	P<0.05	NS	NS
SQ-29,548	87.8±3.2	85.3±4.6	P<0.05	P<0.05	P<0.05

Mean ± SEM values (n = 6 per group)

Figure 2.

Contractile concentration responses to endothelin-1 in mesenteric resistance arterioles from mice with reduced renal mass after bath addition of: Panel A. Reactive oxygen species inhibitors: vehicle, tempol (10⁻⁴ mol·l⁻¹) or mitoTEMPO (10⁻⁴ mol·l⁻¹) or Panel B, cyclooxygenase (COX) inhibitors: SC-560 (10⁻⁶ mol·l⁻¹; COX-1 inhibitor), paracoxib (10⁻⁵ mol·l⁻¹; COX-2 inhibitor) or SC-560 plus paracoxib; or Panel C, thromboxane inhibitors: OKY-046NA (10⁻⁵ mol·l⁻¹; thromboxane A₂ synthase inhibitor), or SQ-29,548 (10⁻⁶ mol·l⁻¹; thromboxane-prostanoid inhibitor). Compared with vehicle: *, P<0.05; **, P<0.01; ***, P<0.005

The enhanced contractions to ET-1 in vessels from mice with RRM were reduced by incubation with tempol, MitoTEMPO or with paracoxib (a selective COX-2 inhibitor) alone or plus SC-560 (a selective COX-1 inhibitor) or SQ-29,548 (a TP-R inhibitor) but not with SC-560 alone or OKY-046NA (a TxA₂-S inhibitor).

Effect of TP-Rs in mice with RRM (Table 5 and Figures 3 and 4)

Table 5.

Basal parameters, plasma endothelin-1, renal excretions of makers of reactive oxygen species, thromboxane and microalbumin, vessel structure, maximum vascular contractions and reactive oxygen species generation in mesenteric resistance arterioles: effects of reduced renal mice and thromboxane-prostanoid receptor knockout

Parameter	TP-R+/+		TP-R−/−		By ANOVA, effect of
	SHAM	RRM	SHAM	RRM	RRM	Genotype	Interaction
Body weight (BWt) (g)	29.4±0.5	30.5±1.6	29.7±1.4	30.3±1.6	NS	NS	NS
Kidney weight /BWt (mg·g−1)	6.1±0.3	9.4±0.5‡	7.1±0.2	8.9±0.4†	<0.005	NS	NS
Heart weight/BWt (mg·g−1)	4.7±0.1	5.2±0.2	4.3±0.2	4.5±0.4	NS	NS	NS
Aorta weight/ BWt (mg·g−1)	0.14±0.005	0.15±0.008	0.15±0.006	0.16±0.007	NS	NS	NS
Plasma endothelin −1 (pg/ml)	2.2±0.3	4.6±0.2‡	3.0±0.2	3.1±0.5	<0.01	NS	<0.01
Urinary 8-isoprostane F2α /creatinine (ng·mg−1)	1.43±0.18	5.36±0.62‡	1.79±0.11	3.32±0.04*¶	P<0.005	P<0.05	P<0.01
Urinary malondialdehyde /creatinine (nmol·mg−1)	47.7±9.1	72.5±7.5*	52.1±5.3	53.1±3.4‖	P<0.05	NS	<0.05
Urinary TxB2/creatinine (ng·mg−1)	1.2±0.1	2.7±0.2*	0.9±0.1	2.1±0.1*	P<0.01	P<0.05	P<0.05
Urinary microalbumin/creatinine (µg·mg−1)	56±4	124±36*	34±8	44±6¶	P<0.05	<0.05	<0.05
Vessel lumen CSA (×1000 µm2)	89.6±4.2	71.4±4.1	80.6±6.2	75.4±3.0	NS	NS	NS
Vessel media CSA (×1000 µm2)	12.0±1.2	16.4±2.5	11.5±2.0	11.4±2.0	NS	NS	NS
Vessel media:lumen (µm2: µm2)	0.13±0.01	0.23±0.02†	0.14±0.02	0.16±0.02‖	P<0.05	NS	<0.05
PE (10−5M) contraction (%)	76.1±5.9	64.4±7.3*	71.7±3.3	62.1±4.2*	NS	NS	NS
U-46,619 (10−6M) contraction (%)	94.3±4.5	110.1±3.2*	1.7±0.7#	2.8±3.3#	P<0.05	P<0.001	P<0.01
ET-1 (10−7M) contraction (%)	81.3±4.5	106.7±2.3*	84.7±4.7	95.5±3.5¶	P<0.05	P<0.01	P<0.05
ET-1 (10−7M) cellular ROS generation (Eth/DHE, Δf/f0)	0.13±0.03	0.35±0.05†	0.20±0.03	0.25±0.05	P<0.01	NS	P<0.05
ET-1 (10−7M) mitochondrial ROS generation (mitoSOX, Δf/f0)	0.06±0.02	0.19±0.02*	0.06±0.04	0.12±0.06	P<0.05	NS	P<0.05

Mean ± SEM values (n = 6 per group). CSA: cross-sectional area. Compared to SHAM:

^*, P<0.05,

^†, P<0.01,

^‡P<0.001,

Compared to equivalent TP+/+:

^‖, P<0.05;

^¶, P<0.01,

^#, P<0.001

Figure 3.

Effect of thromboxane-prostanoid receptors and reduced renal mass on contractile responses in mesenteric resistance arterioles. Data from SHAM (open symbols) or RRM (shaded symbols) in TP-R +/+ (circles) or TP-R −/− (square) mice. Mean± SEM values (n=6 per group). Compared with TP-R +/+: *P<0.05; ***, P<0.005.

Figure 4.

Effect of thromboxane-prostanoid receptors and reduced renal mass on cellular and mitochondrial reactive oxygen species generation with 10⁻⁷ mol·l⁻¹ endothelin-1 in mesenteric resistance arterioles. Mean± SEM values (n=6 per group) for cellular ROS generation (Panel A) and mitochondrial ROS generation (Panel B) from TP-R +/+ or TP-R −/− mice from SHAM (open boxes) or RRM (shaded boxes) groups. Compared to TP-R +/+: *P<0.05.

Similar to C57/BL6 mice of the prior series, TP+/+ mice with RRM had increased plasma ET-1 and increased excretion of 8-Iso, MDA, TxB₂ and microalbumin (Table 5). Their mesenteric resistance arterioles had increased vascular remodeling, decreased contractions to PE, but increased contractions to U-46,619 and ET-1(Figure 3) and increased cellular ROS and mitochondrial ROS with ET-1 (Figure 4). TP-R −/− abolished all of these effects of RRM, except for the reduced PE contractions.

Discussion

We confirm that 3 months of RRM in C57Bl/6 mice increases oxidative stress, albuminuria and growth of the remaining renal mass^{13, 14}. The main new findings are that these mice had increased plasma ET-1 and generated more TxA₂, their mesenteric resistance arterioles were considerably remodeled and had reduced contractions to PE, but enhanced contractions to U-46,619 and ET-1 and enhanced cellular and mitochondrial ROS with ET-1. The enhanced ET-1 contractions were dependent on cellular and mitochondrial ROS and on TP-Rs and products of COX-2>-1. Unlike TP-R +/+, TP-R−/− mice with RRM did not have significantly enhanced plasma levels of ET-1 or enhanced contractions or cellular or mitochondrial ROS in response to ET-1 and did not have vascular remodeling. Thus, TP-Rs mediate enhanced generation of ET-1 and microvascular ROS and enhanced contractility and remodeling in this mouse model of CKD. Figure 5 presents a schema for the proposed central role for TP-Rs in progressive CKD.

Figure 5.

A flow diagram to demonstrate the potential central role for thromboxane-prostanoid receptors in mediating the increased endothelin-1, reactive oxygen species and microvascular remodeling that may contribute to progression of chronic kidney disease.

TP-Rs are activated by PGH₂ (a primary product of COX-1 and-2), TxA₂ and by the stable mimetic, U-46,619. The vascular protein expression of both COX-2 and TP-Rs was upregulated in mice with RRM and the enhanced responsiveness to ET-1 was dependent on COX-2 and TP-Rs (although blockade of COX-1 with COX-2 was more effective than COX-2 alone, suggesting some adaptive interaction). Since blockade of TxA₂-S did not prevent enhanced ET-1 contractions, it is likely that PGH₂, generated by COX-2>-1 was the principal PG activating the TP-Rs. Activation of the TP-Rs enhances the generation of PGs and TxA₂ which further activate the TP-R in a feed-forward manner^{32, 33}. This may account for the reduced excretion of TxB₂ in TP-R −/− mice with RRM. Although mice with RRM had increased TP-R protein expression and increased TP-R responses to U-46,619, TP-R mRNA was unchanged. This may relate to oxidative stress that post-transcriptionally prevents TP-R protein degradation and enhances its membrane expression³⁴.

The enhanced contractions to ET-1 and U-46,619 in vessels from mice with RRM were probably not a consequence of vascular remodeling since contractions to PE were actually diminished. Adrenergic agonists usually do not provoke vascular oxidative stress³⁵. Moreover, the structural and functional changes in mice with RRM were likely independent of BP and uremia since prior telemetric measurement of mean arterial pressure (MAP) were unchanged after 3 months of RRM and the global GFR was reduced by only 33%^{13, 14}

ET-1 is generated in vascular endothelial cells. Its production is increased by ROS^{18, 36} and CKD¹⁹. Whereas ANG II increases ROS in VSMCs largely by activation of NADPH oxidase³⁵, ET-1 also activates other source of ROS including the mitochondria^{37, 38}. The present study demostrates that ET-1 activated ethidium: dihydroethiduim fluorescence ratio and mitoSOX™ Red fluorescence³⁹ and both tempol (distributed throughout the cell)^{40, 41} and mitoTEMPO (partitioned into mitochondria)^39–41 prevented the enhanced contractility to ET-1 in vessels from mice with RRM. Thus, both sources of ROS were activated in mice with RRM and apparently both contributed to enhanced contractions to ET-1. p22^phox is an essential chaperone protein for neutrophial oxidases (NOXs)^{26, 42}. It was strongly upregulated in vessels from mice with RRM and could account for the observed increase in cellular ROS since in vivo silencing of p22^phox prevents the progressive increase in excretion of 8-Iso and the hypertension of rats infused with ANG II²⁶. NOX-2 is expressed in resistance arterioles and the kidney⁴² and was upregulated > 3-fold in the kidneys of mice with RRM¹⁴. An uncoupled endothelial NOS may have contributed also to the increased ROS⁴³.

In contrast to TP-R +/+, TP-R −/− mice with RRM did not have significantly increased plasma levels of ET-1 or excretions of MDA, TxB₂ or microalbumin or significantly enhanced cellular or mitochondrial ROS or contractions to ET-1 (although there were trends suggesting some residual effects). However, the reduced contractions to PE persisted in TP-R-1 −/− mice with RRM which might represent downregulation of vascular α-adrenoreceptors during enhanced sympathetic nervous activity in RRM⁴⁴. The absence of vascular remodeling in arterioles from TP-R −/− mice with RRM may represent less vascular ROS^{45, 46}. Vascular remodeling in mice with ANGII is dependent on TP-Rs in VSMCs⁴⁷. Our findings extend microvascular studies⁹ that have reported that ROS enhance TP-R activity and responsiveness to ANG II and ET-1 by demonstrating that this pattern occurs in a model of CKD and that vascular TP-Rs are required to generate cellular and mitochondrial ROS with ET-1. Thus, TP-Rs are both upstream and downstream of ROS and thereby may play essential mediating and reinforcing roles in the generation of ROS from cellular and mitochondrial sources. They could thereby enhance remodeling and contractility of microvessels in CKD.

Perspective

Future CVD events are predicated by endothelial dysfunction and vascular remodeling^{48, 49} which are frequently accompanied by oxidative stress⁴⁹, as in CKD⁵⁰. ROS^{7, 29, 50}, ET-1¹⁹ and TxA₂^{14, 51, 52} are all increased in patients with CKD. The vascular remodeling, enhanced ROS and responsiveness to ET-1 and thromboxane in mice with RRM were prevented by genetic deletion of TP-Rs. Thus TP-R antagonists, which have already been used in clinical trials²¹, could be novel drugs to prevent vascular oxidative stress and CVD in patients with CKD.

ETA-R blockade improves renal function¹⁹ and reduces glomerulosclerosis in a rat model of RRM⁵³ and markedly reduces albuminuria in patients with diabetic⁵⁴ and non-diabetic⁵⁵ CKD. Since we now show the importance of TP-Rs in activating the ET-1 system in RRM, TP-R antagonists may reduce renal disease progression in addition to vascular injury.

Novelty and Significance.

What is New?

1. The plasma levels of ET-1, the microvascular protein expression of p22^phox, cyclooxygenase-2 (COX-2), TP-Rs and endothelin-A receptors, the remodeling and the contractions to both ET-1 and thromboxane are increased in a mouse model of CKD.

2. The increases in microvascular cellular and mitochondrial ROS of mice with RRM depend on TP-Rs

3. TP-R gene deletion prevents ET-1 generation, microvascular remodeling, enhanced contractile response and ROS generation in mice with RRM.

What is Relevant?

The results from clinical trials in patients treated with TP-R antagonists are promising. Vascular TP-Rs are novel targets to correct the remodeling, dysfunction and oxidative stress in microvessels that precedes cardiovascular disease in patients with CKD.

Summary

Mice with RRM have increased ET-1 and microvascular remodeling and enhanced ET-1-induced cellular and mitochondrial ROS and contractions that are mediated by COX-2 products activating TP-Rs. Thus, TP-Rs can be upstream from ROS and contribute to vascular dysfunction in CKD.