Inactivation of Cys⁶⁷⁴ in SERCA2 increases BP by inducing endoplasmic reticulum stress and soluble epoxide hydrolase

2.1 Animals

All animal care and study protocols complied with the guidelines of the ethical use of animals and were approved by the Animal Care and Use Committee (Third Military Medical University, Chongqing, China). All efforts were made to minimize animal suffering and reduce the number of animals used. Animal studies are reported in compliance with the ARRIVE guidelines (Kilkenny et al., 2010; McGrath & Lilley, 2015) and with the recommendations made by the British Journal of Pharmacology. Experimental protocols and design adheres to BJP guidelines (Curtis et al., 2015).

All mice used in this study were of C57BL/6J background (RRID:IMSR_JAX:000664, https://www.jax.org/strain/000664) and were originally obtained from The Jackson Laboratory (Bar Harbor, USA). Mice were bred and housed in the laboratory animal room at Chongqing University under specific pathogen-free conditions. Mice were kept in open polypropylene cages with clean chip bedding. The animal room was maintained at a controlled temperature (22 ± 3°C) and a 12-hr cycle of light and dark. Five mice in each cage were free to drink water and were fed on a regular diet (0.3–0.8% sodium, 1.0–1.8% calcium, 0.6–1.2% phosphorus, Beijing Keao Xieli Feed Corporation, China) unless otherwise indicated. To obtain tissues, all mice were killed by i.p. injection with 2,2,2-tribromoethanol (250 mg·kg⁻¹, Sigma-Aldrich, Cat# T48402).

2.2 Angiotensin II-induced hypertension

Angiotensin II was used to induce hypertension and oxidative stress in LDL receptor-deficient mice (The Jackson Laboratory, RRID:IMSR_JAX:002207, https://www.jax.org/strain/002207) originally designed for modelling aortic aneurysms. Male mice at 8 weeks old (18–22 g) were fed on a Western diet (21% w/w milk fat, 0.15% w/w cholesterol, Beijing Keao Xieli Feed Corporation) for 1 week and then were randomized to receive subcutaneous implantation of ALZET osmotic pumps (Model 2004, DURECT Corporation, USA), delivering angiotensin II (1.44 mg·kg⁻¹·day⁻¹, dissolved in PBS) or PBS for 4 weeks (n = 5 per group). Before implantation, mice were given buprenorphine (0.08 mg·kg⁻¹, subcutaneous) as analgesic, and during pump implantation, they were anaesthetized with isoflurane (1.5%) inhalation. The mouse back was depilated with depilatory ointment. After disinfection with iodophor, the pump was implanted subcutaneously into the back of the mouse. The incision was sealed with sterile sutures and disinfected with iodophor. Buprenorphine (0.08 mg·kg⁻¹, s.c., every 8 hr) was given to the mouse the first day after surgery. The mice were monitored every day in the first week after operation. Mice were killed by i.p. injection with 2,2,2-tribromoethanol (250 mg·kg⁻¹) and perfused with ice-cold PBS, and then the kidneys were embedded in optimum cutting temperature formulation for detecting ROS and SERCA2 C674 irreversible oxidation.

2.3 BP measured by tail-cuff plethysmography

BP was measured in conscious mice using tail-cuff plethysmography (Softron Biotechnology, China) between 9 a.m. and 11 a.m. Each mouse was warmed at 37°C for 10 min to detect tail artery pulsations and to acquire pulse levels and then was put in a plastic restrainer with its tail wrapped in a cuff containing a pneumatic pulse sensor. To minimize stress-induced fluctuations on BP, all mice were acclimated to the procedures for 1 week, before actual data collection. The arterial BP was measured at least five times, and systolic BP, diastolic BP, and mean BP were recorded.

2.4 Measurement of ROS generation

Dihydroethidium (DHE, APEXBIO, Cat# C3807) was used to detect intracellular oxidants, mainly superoxide. In brief, fresh frozen sections of renal cortex from angiotensin II or PBS-infused mice were incubated with 15-μM DHE at 37°C for 40 min and then washed with PBS to remove free DHE. Fluorescence at λ605 nm was recorded by fluorescence microscopy (Leica DM6, Germany), and the average DHE fluorescence was calculated. To avoid different fluorescence background, the average value of the PBS group was used to normalize the data in each measurement.

2.5 Histology and immunohistochemistry

Kidneys were dissected and fixed in 10% formalin for 2 days, followed by 30% sucrose solution for 1 day before embedding in optimum cutting temperature formulation to prepare serial frozen sections. Masson's trichrome staining was used for morphological analysis. The antibody-based procedures used in this study comply with the recommendations made by the British Journal of Pharmacology (Alexander et al., 2018). For immunohistochemistry, sections were heated for antigen retrieval in 10-mM citrate acid buffer before being blocked in goat serum. Primary antibodies against SERCA2 C674-SO₃H (Ying et al., 2008; C674-SO₃H antibody was a custom polyclonal antibody from Bethyl Laboratories, Inc.), soluble epoxide hydrolase (sEH; Santa Cruz, Cat# sc-87099, RRID:AB_2293440), or dopamine D₁ receptors (Abcam, Cat# ab81296, RRID:AB_2814742) were incubated overnight at 4°C according to the staining procedure of HRP-DAB kit (CWBIO, Cat# CW0125M). IgG acted as a negative control. The positive areas (brown) of cortex tubules were scored by four individual observers blinded to the experimental groups using a scale of 0 to 4, where 0 equals negative intensity, 1 weak intensity, 2–3 medium intensity, and 4 strong intensity.

2.6 SERCA2 C674S knock-in mouse construct

The construct of the SERCA2 C674S knock-in (SKI) mouse (RRID:MGI:5780876) was generated as previously described (Thompson et al., 2014). Briefly, the genetic variation is TGT at C674 to TCC (Ser⁶⁷⁴ [S674]) in exon 14 of SERCA2. The presence of the C674S mutation was verified by sequencing of both genomic DNA and cDNA from heart. Homozygous SKI mice died before birth, so only heterozygous SKI mice that express 50% of C674 and 50% of S674 were used in this study, and their littermate wild-type (WT) mice without S674 were used as controls.

2.7 Mice treated with inhibitors for sEH or ER stress

Both male WT and SKI mice at 12–16 weeks old (25–30 g) were randomly divided into PEG400 solvent control group or treatment group (n = 5 per group). In treatment group, mice were given either a sEH inhibitor, 1-trifluoromethoxyphenyl-3-(1-propionylpiperidin-4-yl) urea (TPPU, 0.3 mg·kg⁻¹·day⁻¹; Zhou et al., 2017), or ER stress inhibitor, 4-phenylbutyrate (4-PBA, 320 mg·kg⁻¹·day⁻¹) individually for 1 week via oral gavage. Both TPPU and 4-PBA were dissolved in PEG400. The BP in conscious WT and SKI mice was measured by tail-cuff plethysmography. Urine was collected in a metabolic cage, and 24-hr urinary sodium excretion and other urinary compounds were measured by the AU5800 clinical chemistry analyser (Beckman Coulter, USA).

2.8 Primary renal proximal tubule cell culture

Primary renal proximal tubule (RPT) cells from 4- to 8-week-old male WT or SKI mice were isolated with a modified method (Terryn et al., 2007). Briefly, the sliced renal cortex was digested in collagenase II (Worthington Biochemical Corporation, Cat# LS004176) solution at 37°C for 30 min before being ground on the first nylon sieve (100 mesh) and the second nylon sieve (200 mesh). The solution collected by flushing the second sieve in the reverse direction was centrifuged for 10 min at 1,000 x g, and the pellet was suspended in DMEM/F12 medium (Hyclone, Cat# SH30023.01) and cultured at 37°C in a humidified atmosphere containing 5% CO₂, supplemented with 10% FBS (ExCell Bio, Cat# FSP500), 10 μg·L⁻¹ EGF (Novoprotein, Cat# CH28), 5 mg·L⁻¹ insulin (Solarbio, Cat# I8830), 5 mg·L⁻¹ transferrin (Solarbio, Cat# T8010), 4 mg·L⁻¹ dexamethasone (Solarbio, Cat# D8040), and 1% penicillin–streptomycin. The medium was replaced every 2 days, and confluent monolayer was formed 7 days later. Cells from passages 0 to 1 were used.

2.9 Na⁺/K⁺-ATPase activity assay

Na⁺/K⁺-ATPase activity of primary RPT cell membranes was assessed by using Na⁺/K⁺-ATPase analysis kit (Nanjing Jiancheng Bioengineering Institute, Cat# A070-2) with a modified method (Chen et al., 2016) using malachite green dye to measure inorganic phosphate (Pi) released from ATP. Na⁺/K⁺-ATPase activity was corrected for protein concentration and expressed as μmol phosphate released per mg protein per min.

2.10 Real-time quantitative PCR

Total RNA was isolated from the renal cortex using TRIzol reagent (Invitrogen, Cat# 15596026) and retro-transcribed to cDNA using cDNA PCR kit (Takara Bio Inc., Cat# RR037Q). Real-time quantitative PCR was performed using SYBR-Green-based detection (Takara Bio Inc., Cat# RR420L) with primers commercially synthesized (Sangon Biotech, China) and listed in Table S1. The following cycling conditions were used: denaturation, annealing, and extension at 95°C, 57°C, and 72°C for 10, 30, and 10 s, respectively, for 40 cycles. β-actin was used as internal control. Relative expression of mRNA was analysed using the comparative Ct method (2^−ΔΔCt), and the relative copy number of SERCA was analysed using a 2^−ΔCt method.

2.11 Western blot

The renal cortex or RPT cells cultured in 0.2% FBS DMEM/F12 overnight were lysed in RIPA buffer (Enogene, Cat# E1WP106). Experimental details of Western blots are in accordance with BJP guidelines (Alexander et al., 2018). Proteins were separated by SDS-PAGE electrophoresis using standard methods, transferred to PVDF membrane, and immunoblotted with specific antibodies overnight at 4°C against SERCA2 (C498 antibody was a custom polyclonal antibody from Bethyl Laboratories, Inc.), p-PERK (Thr981, Santa Cruz, Cat# sc-32577, RRID:AB_2293243), BIP (EnoGene, Cat# E90019, RRID:AB_2814761), ATF6 (EnoGene, Cat# E90009A, RRID:AB_2814762), CHOP (EnoGene, Cat# E90221, RRID:AB_2814763), sEH (Abcam, Cat# ab155280, RRID:AB_2814743), D1 receptors (Abcam, Cat# ab81296, RRID:AB_2814742), β-actin (EnoGene, Cat# E12-051-3, RRID:AB_2814764), GAPDH (EnoGene, Cat# E1C604-1, RRID:AB_2814765), followed by incubation with HRP-conjugated goat-anti-rabbit secondary antibody (Sino Biological Inc., Cat# SSA003, RRID:AB_2814815) 1 hr at room temperature. Proteins were visualized with an X-ray film system (Fujifilm, Japan) or ChemiDoc™ Touch System (Bio-Rad, USA). Band density was quantified by NIH ImageJ software (RRID:SCR_003070, https://imagej.net/) and normalized to GAPDH or β-actin.

2.12 Intracellular calcium measurements

Primary RPT cells were subcultured onto cover slides overnight in a 24-well plate. According to the manufacturer's procedure, the cells were loaded with 4-μM Fluo-4 AM (Solarbio, Cat# F8500) at 37°C for 20 min, then were retained for 40 min after adding fivefold volume of HBSS containing 1% FBS, and were washed three times with HEPES buffer (10-mM HEPES, 1-mM Na₂HPO₄, 137-mM NaCl, 5-mM KCl, 1-mM CaCl₂, 0.5-mM MgCl₂, 5-mM glucose, 0.1% BSA; pH 7.4). The fluorescence signals with excitation λ494 nm and emission λ516 nm were recorded by fluorescence microscopy (Leica DM6), and the average fluorescence signals were analysed by Las X software (Leica, Germany).

2.13 Immunofluorescence staining

Primary RPT cells were incubated with specific antibodies for sEH, D₁ receptors, or cytokeratin 18 (Bioss Antibodies, Cat# BSM-33103M, RRID:AB_2814819) for 12 hr, followed by Alexa Fluor 488 goat anti-rabbit IgG (H + L; Jackson ImmunoResearch, Cat# 111-545-144, RRID:AB_2338052) or Cy3-conjugated AffiniPure goat anti-mouse IgG (H + L; Proteintech, Cat# SA00009-1, RRID:AB_2814746) for 2 hr. Nuclei were stained with DAPI (Solarbio, Cat# C0065) for 10 min. The fluorescence signals were monitored by microscopy (Leica DM6) and analysed by Las X software (Leica).

2.14 Primary RPT cells treated with TPPU or 4-PBA

Primary RPT cells isolated from 4- to 8-week-old male WT or SKI mice were cultured in DMEM/F12 medium and treated with 1-μM TPPU or 0.5-M 4-PBA for 48 hr, before being collected for measurement of Na⁺/K⁺-ATPase activity and protein expression. PEG400 served as a solvent control.

2.15 Adenoviral infection in rat RPT cells

The rat RPT cells were infected with adenovirus sEH or empty adenovirus at 50 MOI per cell in DMEM/F12 respectively without serum and antibiotics for 3 hr before adding DMEM/F12 containing 2% FBS for 2 days.

2.16 Data and statistical analysis

The data and statistical analysis comply with the recommendations of the British Journal of Pharmacology on experimental design and analysis in pharmacology (Curtis et al., 2018). All studies were designed to generate groups of equal size, using randomisation and blinded analysis. Group size is the number of independent values, and statistical analysis was carried out using these independent values. Sample sizes in each groups subjected to statistical analysis were determined based on our previous studies, preliminary results, and power analysis (Curtis et al., 2015). Statistical analysis was undertaken only using these independent values with n ≥ 5. Fold change over control was used in Western blot and fluorescent image analysis to avoid the larger variation among different experiments. The mean values of the control group were normalized to 1. In the figures, the Y axis shows the ratio of the experimental group to that of the corresponding matched control values and is labelled as “fold matched control values.” All results were expressed as mean ± SEM. Statistical analysis was performed using GraphPad Prism 7.00 (RRID:SCR_002798, http://www.graphpad.com/). Unpaired Student's t test was used to analyse data from two groups. When F achieved statistical significance and there was no significant variance heterogeneity, two-way ANOVA with Bonferroni post hoc analysis was used to analyse the differences among multiple groups. For determining whether groups differ, the level of probability was set at P < .05 to constitute the threshold for statistical significance. No data were excluded from any study.

2.17 Materials

The following compounds were supplied by Sigma-Aldrich - 4-PBA, Cat# SML0309; angiotensin II, Cat# A9525; buprenorphine, Cat#BP1062. The sHE inhibitor TPPU was provided by Dr. Lee and Dr. Hammock (University of California-Davis, CA)

2.18 Nomenclature of targets and ligands

Key protein targets and ligands in this article are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY (RRID:SCR_013077) (Harding et al., 2018), and are permanently archived in the Concise Guide to PHARMACOLOGY 2019/20 (Alexander, Christopoulos, et al., 2019; Alexander, Fabbro, et al., 2019).

3.1 Inactivation of SERCA2 C674 increases BP and decreases urine output and sodium excretion

The irreversible oxidation of C674 in SERCA2 was increased in the aorta of hypertensive mice induced by high fat, high sucrose diet (Qin et al., 2014; Weisbrod et al., 2013). As shown in Figure S1A,B, in the renal cortex of angiotensin II-induced hypertensive mice (SBP in mmHg, PBS 99.6 ± 2.8 vs. angiotensin II 133.6 ± 3.3, n = 5, P < .05), we observed increased superoxide production and increased irreversible oxidation of C674, as detected by the specific antibody targeting SERCA2 C674-SO₃H that was previously fully validated (Qin et al., 2013; Tang et al., 2010; Tong et al., 2010; Ying et al., 2008). To address if the C674 of SERCA2 is key to control BP and whether its irreversible oxidation under hypertension-prone conditions drives the direction of hypertension, we used heterozygous SKI mice, where half of the C674 was substituted by S674 to represent the partial irreversible oxidation of C674. In the heterozygous SKI mice, body weight was not different from that in the WT mice (WT 27.2 ± 0.67 g vs. SKI 27.5 ± 0.53 g, n = 20), but BP was increased, as shown in Figure 1a. SKI mice also had decreased daily urine volume and daily urine sodium excretion, compared with WT mice (Figure 1b). The concentrations of urinary sodium and other components were not different between WT mice and SKI mice (Table S2), indicating that the decreased urinary sodium excretion was due to the decrease of urine volume. Overall, there was no difference in renal histological structure between WT mice and SKI mice (Figure S2). The renal proximal tubule in the renal cortex accounts for 60–80% reabsorption of filtered sodium (Burnier, Bochud, & Maillard, 2006), in which the Na⁺/K⁺-ATPase is the most important pump to transport sodium from proximal tubule to extracellular space to reduce urinary sodium excretion (Harris & Zhang, 2012) and is responsible for sodium homeostasis. Activation of Na⁺/K⁺-ATPase in the proximal tubule induces the retention of water and sodium (Te Riet, van Esch, Roks, van den Meiracker, & Danser, 2015), so we measured the activity of Na⁺/K⁺-ATPase in the cultures of primary RPT cells. As shown in Figure 1c, the activity of Na⁺/K⁺-ATPase was higher in SKI RPT cells than in WT RPT cells. These results suggest that an enhancement of water and sodium retention may account for the elevated BP in SKI mice.

3.2 Inactivation of SERCA2 C674 increases intracellular calcium level in RPT cells

C674 inactivation could occur in both SERCA2a and SERCA2b, two major isoforms of SERCA2. There was no difference in the mRNA abundance of both SERCA2a and SERCA2b in renal cortex between WT mice and SKI mice, although the relative copy number of SERCA2b was about 100 times higher than that of SERCA2a (Figure 2a), consistent with previous report that SERCA2b as a housekeeping gene is the predominant one among all SERCA isoforms in the kidney, while other isoforms are negligible (Guo et al., 2016). The substitution of S674 for C674 did not affect the protein expression of SERCA2 in the renal cortex and primary RPT cells (Figure 2b), suggesting that the elevation of BP in SKI mice was not due to the different expression levels of SERCA2 in the renal cortex. The major function of SERCA2 is to take up intracellular calcium into sarcoplasmic reticulum and ER to maintain calcium homeostasis. Next, we used Fluo-4 to detect intracellular calcium. As shown in Figure 2c, the intracellular calcium level of SKI RPT cells was higher than that of WT RPT cells, indicating that inactivation of C674 inhibited the normal function of SERCA2 in RPT cells without affecting the expression of SERCA2.

3.3 The sustained induction of ER stress by inactivation of SERCA2 C674 accounts for the elevated BP

We further explored the causes of water and sodium retention and elevated BP in SKI mice. SERCA, including SERCA2, is the only enzyme in the ER to take up intracellular calcium into ER to maintain ER calcium stores, which is essential for ER normal function, and sustained ER calcium depletion causes ER stress (Mei, Thompson, Cohen, & Tong, 2013). ER stress is an important contributor to hypertension (Mohammed-Ali et al., 2017; Wang et al., 2017; Yum et al., 2017). We previously reported that endothelial cells from SKI mice had depleted ER calcium stores (Thompson et al., 2014), implying that inactivation of C674 could induce ER stress. In the renal cortex (Figure 3a) and primary RPT cells (Figure 3b), compared with the samples from WT mice, SKI mice had higher protein expression levels of ER stress markers, p-PERK, BIP, ATF6, and CHOP. To determine if the sustained induction of ER stress in SKI mice accounts for the elevated BP, we applied the ER stress inhibitor, 4-PBA. The increased activity of Na⁺/K⁺-ATPase in primary SKI RPT cells was reversed by 4-PBA (Figure 3c). Furthermore, treatment in vivo with 4-PBA for 1 week reversed the elevated BP in SKI mice to levels similar to those in WT mice treated with solvent control, but it did not affect BP in WT mice (Figure 3d). 4-PBA therapy also reversed the decreased daily urine volume and urinary sodium excretion in SKI mice (Figure 3e). These results indicate that the elevation of BP in SKI mice is caused by persistent ER stress, which enhances Na⁺/K⁺-ATPase activity in primary RPT cells to cause the retention of water and sodium.

3.4 Inactivation of SERCA2 C674 up-regulates the expression of sEH and down-regulates the expression of D₁ receptors

In the renal cortex, a number of physiological systems control sodium excretion and BP, including the renin-angiotensin system, intrarenal dopaminergic neurons, a range of renal sodium transporters (Wang, Luo, et al., 2010; Harris & Zhang, 2012; te Riet et al., 2015), and some other important factors, such as sEH, endothelin-1, and caveolin-1. We screened genes to determine which factors contribute to the reduction of water and sodium excretion and elevated BP in SKI mice and found that both sEH and D₁ receptors were affected. Compared with WT mice, the sEH of SKI mice increased and D₁ receptors decreased at the level of mRNA (Figure S3), which were further supported at the level of protein in both renal cortex and primary RPT cells (Figure 4). sEH was mainly located in the RPT of kidney (Figure S4; Liu, 2018) and this enzyme metabolizes epoxyeicosatrienoic acids to less active dihydroxyeicosatrienoic acids (He, Wang, Zhu, & Ai, 2016), and thus inhibits sodium excretion by enhancing the activity of Na⁺/K⁺-ATPase in RPT (Dos Santos, Dahly-Vernon, Hoagland, & Roman, 2004; Imig, 2000; Imig, 2004). On the contrary, activation of D₁ receptors promotes sodium excretion by inhibiting Na⁺/K⁺-ATPase in RPT (Zeng & Jose, 2011; Harris & Zhang, 2012). Therefore, the increase of sEH and the decrease of D₁ receptors in SKI RPT cells could reduce sodium excretion by enhancing the activity of Na⁺/K⁺-ATPase, thus causing the retention of water and sodium, and ultimately increasing BP.

3.5 Inhibition of ER stress reverses the expression of sEH and D₁ receptors in SKI RPT cells

Next, we examined whether persistent induction of ER stress explained the increase in sEH expression and the decrease in D₁ receptor expression in SKI RPT cells. In WT RPT cells, both p-PERK and BIP were down-regulated by 4-PBA, and sEH showed a trend towards decrease, while ATF6, CHOP, and D1R were not affected by 4-PBA (Figure S5). However, 4-PBA reversed the elevation of BP in SKI mice, down-regulated ER stress markers and sEH expression, and up-regulated the reduced D₁ receptor expression in SKI RPT cells (Figure 5). These results suggest that the persistent ER stress in the renal cortex of SKI mice controls sodium secretion and BP by regulating sEH and D₁ receptors.

3.6 sEH inhibitor suppresses ER stress and reverses the elevated BP in SKI mice

Several studies indicate that sEH could regulate ER stress (Bettaieb et al., 2017; Harris et al., 2015; Wagner, McReynolds, Schmidt, & Hammock, 2017), so we applied the sEH inhibitor TPPU to RPT cells from both WT mice and SKI mice. In WT RPT cells, TPPU itself did not affect the expression of sEH compared with solvent control, nor did it affect the expression of ER stress markers and D₁ receptors (Figure S6). Although TPPU did not affect sEH expression, it did reverse the increased ER stress markers and decreased D₁ receptor expression in SKI RPT cells (Figure 6a,b). The increased Na⁺/K⁺-ATPase activity in SKI RPT cells was also reversed by TPPU (Figure 6c). As observed with 4-PBA, treatment in vivo with TPPU for 1 week reversed the elevated BP and the decreased daily urine volume and urinary sodium excretion in SKI mice (Figure 6d,e). These results suggest that up-regulation of sEH is a major cause of elevated BP in SKI mice.

3.7 sEH up-regulates ER stress markers and down-regulates expression of D₁ receptors

To investigate the roles of the up-regulated sEH in SKI mice, we overexpressed sEH in rat RPT cells. As shown in Figure 7, compared with empty adenovirus control, overexpression of sEH increased the expression of ER stress markers, while decreasing the expression of D₁ receptors. This indicates that while ER stress and sEH might regulate each other, but both are upstream of D₁ receptors.