Prenatal hypoxia inhibited propionate‐evoked BK channels of mesenteric artery smooth muscle cells in offspring

Abstract As a common complication of pregnancy, gestational hypoxia has been shown to predispose offspring to vascular dysfunction. Propionate, one of short‐chain fatty acids, exerts cardioprotective effects via reducing blood pressure. This study examined whether prenatal hypoxia impaired propionate‐stimulated large‐conductance Ca2+‐activated K+ (BK) channel activities in vascular smooth muscle cells (VSMCs) of offspring. Pregnant rats were exposed to hypoxia (10.5% oxygen) and normoxia (21% oxygen) from gestational day 7‐21. At 6 weeks of age, VSMCs in mesenteric arteries of offspring were analysed for BK channel functions and gene expressions. It was shown firstly that propionate could open significantly BK single channel in VSMCs in a concentration‐dependent manner. Antagonists of G protein βγ subunits and inositol trisphosphate receptor could completely suppress the activation of BK by propionate, respectively. Gαi/o and ryanodine receptor were found to participate in the stimulation on BK. Compared to the control, vasodilation and increments of BK NPo (the open probability) evoked by propionate were weakened in the offspring by prenatal hypoxia with down‐regulated Gβγ and PLCβ. It was indicated that prenatal hypoxia inhibited propionate‐stimulated BK activities in mesenteric VSMCs of offspring via reducing expressions of Gβγ and PLCβ, in which endoplasmic reticulum calcium release might be involved.

demonstrated in middle cerebral arteries, 2 mesenteric arteries, [3][4][5][6] renal arteries 7,8 and femoral arteries. 9,10 More studies described increased blood pressure responses and greater vasoconstrictions of resistance arteries in the offspring exposed to prenatal hypoxia, [11][12][13] which were associated with altered vascular constrictor and dilator mechanisms. Among underlying factors are calcium and potassium channels, which play pivotal keys in contraction and hyper-polarization of vascular smooth muscle cells (VSMCs), 7,11,12,14 respectively. Nevertheless, the detailed molecular mechanisms through which hypoxia in utero affects vascular functions of foetuses and offspring have yet to be elucidated.
Short-chain fatty acids (SCFAs) are gut microbial metabolites produced by fermentation of dietary fibres, mainly including acetate, propionate and butyrate. 15 Beyond the important physiological roles of shaping the gut environment and energy homeostasis, SCFAs are known to regulate several cellular processes in other tissues by activating membrane receptors (Gpr41, Olr59, Ffar2 and so on), inhibiting histone deacetylases, and stabilizing the hypoxia-inducible factor. [16][17][18] Moreover, the protection of SCFAs against hypertension has been demonstrated recently in different models.
On one hand, SCFAs reduce risk factors of cardiovascular diseases, including obesity by improving appetite regulation 19 and type 2 diabetes mellitus by improving insulin sensitivity. 20 On the other hand, SCFAs could regulate blood pressure by relaxing resistance arteries [21][22][23] and releasing renin. 24 In addition, the anti-hypertensive effect of propionate has been proposed to be linked with anti-inflammatory properties via regulating T helper cell homeostasis. 25 In spite of uncovering the beneficial effects of SCFAs in various ways, understanding of the intracellular mode of actions is limited.
Based on the characteristics of SCFAs-induced vasodilatation, effects of propionate on large-conductance Ca 2+ -activated K + (BK) channel activity in VSMCs were investigated in this study for the first time. Downstream signal pathway of propionate receptors was tested to explore the underlying mechanisms. Considering the increased vasoconstrictions in rat offspring suffered from hypoxia in utero, we hypothesized that prenatal hypoxia inhibited propionate-dependent BK activities in mesenteric arteries of the offspring.
To test the hypothesis, basal open probability (NPo) of BK channels, propionate-stimulated NPo of BK with and without specific antagonists, and expressions of relative proteins in mesenteric VSMCs were compared between prenatal hypoxia offspring and the control.
The information obtained extends the physiological functions of propionate and the influence of prenatal hypoxia on BK dysfunction in the offspring.

| Animals
Sprague-Dawley rats (female: 300-350 g, male: 400-450 g) were purchased from Animal Center of Soochow University and housed in a controlled environment (24 ± 0.2°C) under a 12 light-dark cycles (8:00 am-8:00 pm) with standard rat diets. One female rat was caged with two male rats overnight and pregnancy was confirmed by the presence of vaginal plug next day, which was regarded as the first day of gestation. Pregnant rats were randomly divided into two groups (N = 15 each): control (CON) and prenatal hypoxia group (HY).
From gestation day 7-21, HY rats were placed in hypoxia chambers (10.5% O 2 ) and the CON rats in normal circumstances (21% O 2 ), with free access to food and water. All rats were raised in normal environment after gestational day 21. After delivering, all rats were provided with tap water and standard rat food, and eight pups per litter were culled to receive breastfeeding and maternal care. A 6-weekold male offspring with different mothers were killed using sodium pentobarbital (100 mg/kg intraperitoneally) and used for following experiments. It has been reported that there exist sex differences in vascular function of offspring exposed to prenatal hypoxia. 26,27 In order to get rid of the effects of sex hormones, the focus of the present study was on male offspring. All experimental procedures were approved by the Ethical Committee of First Hospital of Soochow University and carried out conforming to the SSR's specific policies and procedures.

| Measurement of vessel tone
Third-order mesenteric arteries in 6-week-old male offspring were separated from connective tissue, cut into 3 mm in length and recorded with multi-myograph system (Radnoti) in 5 mL modified Krebs' solution (mmol/L: NaCl 115, NaHCO 3 25, KCl 4.6, NaH 2 PO 4 1.2, MgCl 2 1.2, CaCl 2 2.5, and glucose 10; pH 7.4 with NaOH), continuously gassed with 95% O 2 -5% CO 2 . A total of 120 mmol/L KCl was used to achieve maximal tension. After equilibration, cumulative concentration of phenylephrine (PE) was added into chambers and the contractions were normalized to maximal tension. Increasing dose of sodium propionate (SP) was added after application of 60 mmol/L KCl. Changes in tension by SP were expressed as per cent of the contraction induced by 60 mmol/L KCl.

| Cell culture
Vascular smooth muscle cells were isolated from mesenteric arteries in 6-week-old male rats, as previously mentioned. 28 Briefly, mesenteric arteries were stripped off adipose and connective tissues in pre-cooling phosphate-buffered saline (PBS). First order of mesenteric arteries were opened longitudinally and cut into pieces in a culture dish. Pieces attached to the dish were maintained in Dulbecco's Modified Eagle's Medium with high glucose (Thermo), including 100 U/mL penicillin, 100 mg/mL streptomycin (Thermo) and 10% foetal bovine serum (Bovogen) in the humidified incubator at 37°C with 95% air and 5% CO 2 . The cells would grow out successively from the original mesenteric arteries pieces after approximately three days.
All procedures were carried out under sterile conditions in laminar air flow bench. Confirmed as VSMCs using immunofluorescence of α-smooth muscle actin (α-SMA), the cells at third-fifth passages were used for the experiments.

| Immunohistochemistry and immunofluorescence
Third-order mesenteric arteries were isolated and cleared of adipose and connective tissues, followed by immediately fixed in 10% neutralbuffered formalin solution. Embedded in paraffin, vessel rings were serially sectioned at 6 μm. According to the standard immunohistochemical staining procedures, 29 Gpr41 and Olr59 were measured in the vessel

| Electro-physiological measurements on cells
Smooth muscle cells from mesenteric arteries were enzymatically dissociated as previously described. 30 To be specific, the third-order mesenteric arteries were cut into pieces in ice-cold Ca 2+ -free physiological saline solution (PSS), containing (in mmol/L) the following: KCl 5.6, NaCl 137, MgCl 2 1, HEPES 10, Na 2 HPO 4 0.42, NaH 2 PO 4 0.44, NaHCO 3 4.2 and glucose 10 (pH 7.4 with NaOH), and then incubated with 4 mg/mL papain (Solarbio), 2 mg/mL bovine serum albumin (Biosharp) and 1 mg/mL dithiothreitol (Biosharp) for 40 minutes at 37°C. Then, tissues were washed three times with ice-cold Ca 2+ -free PSS and dispersed to gain single smooth muscle cells. A 200-μm nylon mesh was used to percolate tissue debris. The cell suspension was stored at 4°C for study within 6 hours.

Gene name
Forward primer sequence Reverse primer sequence

| Western blot
Mesenteric arteries of offspring were homogenized in liquid ni- Blots were detected using enhanced chemiluminescence detection reagents (Advansta), and specific bands were analysed using a Bioimaging System (Tanon). Protein expressions were presented as relative richness normalized to β-actin.

| Statistical analysis
Results were exhibited as mean ± standard error of mean (SEM).

Statistical differences were decided by unpaired Student's t test with
Welch's correction or two-way ANOVA with Bonferroni post test when appropriate. Data were analysed and curve fitted using GraphPad Prism 5.0 software. A value of P (two-tailed) <.05 was considered significance.

| Prenatal hypoxia increased PE-induced vasoconstriction and decreased propionate-induced relaxation in offspring
Mesenteric artery was used as a representative of peripheral resistance vessel. Vessel tone responding to PE was tested to evaluate the contraction function of mesenteric artery. There was no significant difference in KCl-mediated contraction ( Figure 1A). According to the concentration-response curves of PE-induced vasoconstriction shown in Figure 1B, mesenteric arterial contraction in HY responding to 10 -5 and 10 -4 mol/L PE was greater than that in CON.
Although the maximal relaxation induced by SP was similar between HY and CON, 25 mmol/L SP-induced vasodilation was weaker in HY than that in CON ( Figure 1C), with decreased pD 2 values (CON: 1.569 ± 0.088, HY: 1.424 ± 0.057, P < .01, shown in Figure 1D).

| Distribution of SCFAs receptors in mesenteric artery
Short-chain fatty acids and SCFAs receptors, Gpr41 and Olr59, were identified as novel regulators for blood pressure. In this context, presences of Gpr41 and Olr59 in mesenteric artery of rats were investigated by immunohistochemistry. As shown in Figure 2A

| Prenatal hypoxia inhibited propionatetriggered BK activities in mesenteric VSMCs
It is generally accepted that SCFAs are capable of inducing vasorelaxation in resistance arteries, such as caudal arteries and mesenteric arteries. 22,23 As a primary potassium channel, BK

is involved in hyper-polarization and subsequent relaxation of
VSMCs. Therefore, effects of SCFAs on activities of BK were detected via inside-out patch-clamp. At the testing potential of + 50 mV and 10 μmol/L [Ca 2+ ] free , it was found that SP activated BK single channel in a concentration-dependent manner ( Figure 3A). Twenty-five mmol/L SP, approximate EC 50 , was used in the subsequent experiments. Figure 3B

| Propionate-dependent mechanism of BK activation
Propionate receptors, Gpr41 and Olr59, are known to be Gα i/o and Gα s coupled receptors, respectively. Specific blockers were applied to address the mechanisms which underlie propionate-dependent BK opening. Pre-incubation of Gα i/o specific inhibitor, PTX, blocked SP-sensitive BK currents partially ( Figure 4A). Without PTX, SPevoked BK was 531.4 (NPo) times that at baseline, while SP-evoked BK with PTX was 21.52 times (P < .001; Table 2). G protein β subunit (Gnb) and IP3R-specific antagonists, gallein and 2-APB, respectively, inhibited completely SP-sensitive BK currents ( Figure 3B, 3), indicating Gnb and IP3R were indispensable for the stimulation of SP. NPo of BK was still increased by SP after blocking ryanodine receptor (RyR) by ryanodine ( Figure 3D), but the amplification was reduced (14.99 with ryanodine, 531.4 without ryanodine, P < .001).
As for NPo of BK in mesenteric VSMCs from HY in the presence of the four antagonists, there were no significant differences before and after stimulation of SP (Table 2), indicating complete inhibitions. These results suggest that G protein-IP 3 R/RyR might be involved in cellular mechanisms of SP-sensitive BK opening.

Firstly, expressions of SCFAs receptors existed in mesenteric
VSMCs were analysed. The mRNA and protein levels of Gpr41 and Olr59 in mesenteric arteries from HY were elevated compared with that from CON ( Figure 5A, 5). As shown in Figure 5C Gnb, was reduced in HY ( Figure 6B). It is known that phospholipase C (PLC), especially PLCβ, acts as a canonical effector of G protein βγ dimer to catalyse the release of diacyl glycerol and inositol 1,4,5-trisphosphate (IP 3 ). As shown in Figure 6C, 6, expressions of PLC family positive in mesenteric arteries were decreased in HY compared to CON. Furthermore, expression levels of IP 3 R and RyR in HY were examined. mRNA expressions of IP 3 R2 and IP 3 R3 rather than IP 3 R1 were elevated significantly in HY ( Figure 6E). Although RyR2 in HY mesenteric arteries was decreased, RyR3 (more abundant than RyR2) was raised ( Figure 6F). Taken together, reduction of Gnb5 and Plcb3 might be associated with insensitivity of BK to propionate in HY.

| D ISCUSS I ON
Epidemiological studies and investigations on animal models have suggested that the gut and gut microbiome could influence blood pressure. As the unique metabolites of gut microbiome, SCFAs were absorbed from the colon into the bloodstream to modulate host physiology in various tissues. 31,32 As early as 1990, it was manifested that three SCFAs alone or in combination could cause a concentration-dependent dilatation in human colonic resistance arteries. 23 Recently, Pluznick et al 33

ACK N OWLED G EM ENTS
We would like to thank Animal Center of Soochow University for experimental support. This work was supported by Suzhou Natural

CO N FLI C T O F I NTE R E S T
No conflicts of interest, financial or otherwise, are declared by the authors.

AUTH O R S ' CO NTR I B UTI O N S
XZ designed the project and wrote the manuscript; ZX revised the manuscript; WZ, XF and YZ performed experiments and analysed data; MS performed statistical analysis; LL, QG and JT assisted with critical paper assessment; PZ and JL assisted with data analysis.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding authors upon reasonable request.