Methods

Animal models

Experiments were performed on mesenteric arteries from Male Wistar rats (Charles River, Margate, UK) ages 11-14 weeks (200-350 g) from the Biological Research Facility, St George’s, London, UK; and from the Animal Resources Center, Perth, Australia. Animals were housed in cages with free access to water and food (RM1; Dietex Inter-national, UK)ad libitum , with a 12-hour light/dark cycle and constant temperature and humidity (21 ± 1°C; 50% ± 10% humidity) in accordance with the Animal (Scientific Procedures) Act 1986, the guidelines of the National Health and Medical Research Council of Australia and the UNSW Animal Ethics and Experimentation Committee (AEEC #18/86B). Animals were kept in a bedding of LSB Aspen woodchip. Animals were culled by either cervical dislocation with secondary confirmation via cessation of the circulation by femoral artery severance or were anaesthetized with sodium pentathol (intraperitoneal, 100 mg/kg) in accordance with Schedule 1 of the ASPA 1986.
Either whole mesenteric plexus or 2nd/3rd/4th order MA were used with vessel order identified from the second bifurcation of the superior mesenteric artery. Arteries were dissected, cleaned of fat and adherent tissue and stored on ice within physiological salt solution (PSS) of the following composition (mmol-L-1); 119 NaCl, 4.5 KCl, 1.17 MgSO4.7H20, 1.18 NaH2PO4, 25 NaHCO3, 5 glucose, 1.25 CaCl2.

Reverse transcription quantitative polymerase chain reaction

To investigate gene expression in fresh ECs we used a subtractive approach rather than generating pure native EC samples. Relative fold-changes in expression levels of VSMC/EC markers and Kcnqtranscript was determined in denuded MA samples (an EC(-) population) compared to whole MA samples (an EC(+) population) via Reverse transcription quantitative polymerase chain reaction (RT-qPCR).
EC(-) MA samples were prepared as described previously (Askew Page et al., 2019); Briefly, vessels were cut open longitudinally and pinned on a Sylgard dish, the lumen of the vessel was then rubbed with human hair, and vessels washed in 0.1% (phosphate buffered saline) PBS-Triton X for 1 x 1 min, then 3 times in PBS (1 minute each) on ice. EC(+) MA samples were whole MA plexus that had not undergone EC removal as above.
mRNA from both whole MA EC(+) and MA EC(-) was extracted using Monarch Total RNA Miniprep Kit (New England BioLabs, Ipswich, Massachusetts, USA) with a LunaScript RT SuperMix Kit (New England BioLabs, Ipswich, Massachusetts, USA). Quantitative analysis of relative gene expression was assessed via CFX-96 Real-Time PCR Detection System (BioRad, Hertfordshire, UK). Samples were run in duplicate to account for variation. Samples were run in BrightWhite qPCR plate (Primer Design, Camberley, UK), with each well containing 20 µL of reaction solution containing: 10 µL of PrecisionPLUS qPCR Master Mix (Primer Design, Camberley, UK), 300 nmol-L-1 of gene specific target primer (Thermofisher scientific, Waltham, Massachusetts, USA) and 10 ng of cDNA sample made up to 20 µL total volume with nuclease free water. Run protocol: 1. activation step (15 min:95°C), 2. denaturation step (15 sec: 94°C), 3. annealing step (30 sec: 55°C) and 4. extension step (30 sec: 70°C). Steps 2- 4 were repeated x 40. Quantification cycle (Cq) was determined via Bio-Rad CFX96 Manager 3.0. Values are Cq values normalised to housekeeper genes expressed as a 2-ΔΔCqof MA EC(-) compared to MA EC(+) samples. Appropriate reference genes including ubiquitin C (UBC), cytochrome C1 (CYC1) and Beta-2-microglobulin (B2M; Primer design, UK) were determined using qbase+ PCR analysis system (Biogazelle, Ghent, Belgium) geNorm programme. See Table 1.1 for a list of the primers used in the following (Askew Page et al., 2019; Thomas A. Jepps et al., 2011); ThermoFisher Scientific).

Immunocytochemistry

2nd and 3rd order MA segments were enzymatically digested to obtain freshly isolated ECs as previous (Greenberg et al., 2016). Briefly, vessels were washed in Hanks’ Balanced Salt Solution (HBSS; ThermoFisher Scientific, GIBCO, 14170-088) containing 50 μmol-L-1 CaCl2 for 5 min at 37 °C and then placed in 1mg/mL collagenase IA (Sigma Aldrich, C9891, UK) in the same solution for 15 min at 37°C. Vessel were washed in HBSS containing 50 μmol-L-1 CaCl2 for 10 min at 37°C. The supernatant was removed and the vessels cells suspended in fresh HBSS containing 0.75 mmol-L-1CaCl2. ECs were dissociated using a wide-bore smooth-tipped pipette. The cell-containing solution was plated onto coverslips and left at RT for 1 h before use.

Freshly dispersed ECs, together with residual VSMC, were fixed in 4% paraformaldehyde (Sigma-Aldrich, UK ) in PBS for 20 min at RT as previously described (Barrese, Stott, Figueiredo, et al., 2018). Cells were treated with 0.1 mol-L-1 glycine for 5 min and incubated for 1 h with blocking solution (PBS-0.1% Triton X-100-10% bovine serum albumin) at RT. Following the incubation overnight at 4°C with primary antibodies (Table 1.2) diluted in blocking solution (anti-PECAM-1 for ECs, anti-α-actin for VSMCs and anti-KV7.4 or KV7.5 channel for ECs/VSMCs), cells were then washed for 20 min with PBS, incubated for 1 hr at RT with the secondary conjugated antibodies diluted in blocking solution. Excess secondary antibody was removed by washing with PBS and mounted using media containing DAPI for nuclei counterstaining. Using triple staining, ECs and VSMC were differentiated via the following: ECs were positive for anti-CD31 (endothelial cell-specific marker) and negative for anti-α-actin (data not showed); while VSMC was positive for anti-α-actin and negative for anti-PECAM-1 (data not showed). Cells were analysed using a Zeiss LSM 510 Meta argon/krypton laser scanning confocal microscope (Image Resource Unit St George’s University of London).

Immunohistochemistry

Animals were anaesthetized with sodium pentathol (intraperitoneal, 100 mg/kg) and perfusion fixed (Sandow, Goto, Rummery, & Hill, 2004) in 2% paraformaldehyde in 0.1 mol-L-1 PBS. Third to 4th order mesenteric artery segments were dissected, opened laterally and pinned as a sheet to a Sylgard dish. Segments were washed in PBS (3 x 5 min), incubated in blocking buffer (PBS with 1% BSA and 0.2% Triton) at room temperature (RT) for 2 h and then overnight with primary antibody (Table 1.2) in blocking buffer at 4oC, washed again (3 x 5 min with gentle agitation), and incubated in secondary antibody (Table 1.2; matched to the respective primary) in PBS with 0.1% Triton in PBS for 2 h at RT. Tissue was mounted on slides in anti-fade media containing propidium iodide (PI) or 4′,6-diamidino-2-phenylindole (DAPI; Table 1.2) and imaged with uniform confocal settings. Incubation of tissue with secondary only was used as a ‘zero’ setting for confocal imaging. Controls involved substitution of primary with isotype control, with concentration (where provided by manufacturer) matched, or 10-fold higher than the respective antibody of interest (Table 1.2). Working Ab dilutions were prepared in accordance with previous work (Chadha et al., 2012; Thomas A. Jepps, Greenwood, Moffatt, Sanders, & Ohya, 2009). Confocal image stacks were collected at 0.2 µm intervals. The optimal rinsing protocol was determined by incubating in secondary only; and rinsing after successive 5 min incubations until fluorescence was reduced to background. Note that if this was not done secondary alone was specifically highly localized to IEL hole sites; as potential false positives at such sites; suggesting that such sites have an affinity for IgG-secondary label alone.

Immunoelectron microscopy

Animals were anaesthetized as above and perfusion fixed in 0.2% glutaraldehyde and 2% paraformaldehyde in 0.1 mol-L-1PBS (pH 7.4). Mesenteric artery segments (~2 mm in length) were washed (3 x 5 min) and processed in a Leica EMPACT 2 high-pressure freezer using 0.7% low melting agarose as a cryoprotectant. Samples were then freeze-substituted in a Leica AFS2 into 0.2% uranyl acetate in 95% acetone (from -85 to -50oC) and infiltrated with Lowicryl (at -50oC), before UV polymerization (2 d each at -50 and 20oC; (Zechariah et al., 2020).
Individual serial transverse sections (~100 nm) were mounted on Formvar-coated slot grids and processed for antigen localization as for confocal immunohistochemistry (per above and Table 1.2). The secondary used was 5 or 10 nmol-L-1colloidal gold-conjugated antibody (1:40; 2 h) in 0.01% Tween-20. Sections were imaged at x10-40,000 on a JEOL transmission electron microscope at 16 MP (Emsis, Morada G3). Background gold label density was determined from randomly selected (4 x) 1 x 1 µm regions per sample of lumen and IEL, compared to the same sized regions of interest in the endothelium.

Wire Myography

2nd order MA segments (~2 mm in length) were mounted on 40 µm tungsten wire in a tension myograph chamber (Danish Myo Technology, Arhus, Denmark) containing 5 mL of PSS (composition, as above) oxygenated with 95% O2 and 5% CO2 at 37°C. Vessels then underwent a passive force normalization process to achieve an internal luminal circumference at a transmural pressure of 100 mmHg (13.3 kPa) to standardize pre-experimental conditions (Mulvany, 1977). Force generated was first amplified by a PowerLab (ADInstruments, Oxford, UK), and recorded by LabChart software (ADInstruments, Oxford, UK). Vessels were then challenged with 60 mM [K+] to determine viability. Vessels were then constricted with 10 µmol-L-1methoxamine, an α-1 adrenoreceptor agonist, and EC integrity determined via addition of 10 µmol-L-1 CCh. Vessels displaying ≥90% vasorelaxation in response to CCh (10 µmol-L-1) were considered EC positive (EC+). Vessels were denuded of ECs by gently passing a human hair through the lumen. Vessels expressing ≤10% vasorelaxation in response to CCh (10 µmol-L-1) were considered EC negative (EC-). During functional investigations, all vessels were pre-constricted with the thromboxane A2 receptor agonist U46619 (300 nmol-L-1) to elicit an EC80 contraction. Concentration-dependent relaxant responses to S-1 (0.1-10 µmol-L-1), ML213 (0.1-10 µmol-L-1), ML277 (0.03-1 µmol-L-1), CCh (0.3-10 µmol-L-1) and S-nitroprusside (SNP; 0.01-3 µmol-L-1) were determined in the presence and absence of ECs, linopirdine (10 µmol-L-1), HMR-1556 (10 µmol-L-1), ML133 (20 µmol-L-1), barium chloride (BaCl; 100 µmol-L-1), L-nitroarginine methyl ester (L-NAME; 100 µmol-L-1), TRAM34 (1 µmol-L-1), Apamin (10 nmol-L-1), 4-aminopyridine (4-AP; 1 mmol-L-1) and tetraethylamonium (TEA; 1 mmol-L-1).

Dara and statistical analysis

All functional figures express mean data from at least 5 animals ±standard error of the mean (SEM). For functional experiments involving cumulative concentrations, a transformed data set was generated using; X=Log(X), to reduce representative skew. A four parametric linear regression analysis was then performed using the following equation; (Log(Agonist) vs. response – variable slope (four parameters bottom/hillslope/top/EC50)) using GraphPad Prism (Version 8.2.0) to fit a CEC to the figure. For data comparing multiple groups, a two way-ANOVA followed by a post hoc Bonferonni test, to account for type 1 errors in multiple comparisons was performed for comparison of mean values. Significance values are represented as follows; P <0.05 (*). The data and statistical analysis comply with the recommendations on experimental design and analysis in pharmacology (Curtis et al., 2018).