Mice with loxP sites flanking exons 11 and 13 (Pkd2^fl/fl) of the Pkd2 gene were crossed with tamoxifen-inducible, endothelial cell-specific Cre (Cdh5-creERT2) mice, producing a Pkd2^fl/fl:Cdh5-creERT2 line. Genomic PCR confirmed that tamoxifen stimulated Pkd2 recombination in mesenteric arteries of Pkd2^fl/fl:Cdh5-creERT2 mice, but not in arteries of Pkd2^fl/fl mice (Figure 1—figure supplement 1). Genomic PCR also amplified an identical product in tamoxifen-treated Pkd2^fl/fl and Pkd2^fl/fl:Cdh5-creERT2 mouse arteries due to Pkd2 in cells such as smooth muscle, where DNA would not undergo recombination (Figure 1—figure supplement 1; Bulley et al., 2018).

Western blotting was performed to quantify proteins in lysate collected from second- through fifth-order mesenteric artery branches. PKD2 protein in mesenteric arteries of tamoxifen-treated Pkd2^fl/fl:Cdh5-creERT2 mice was ~ 67.2% of that in tamoxifen-treated Pkd2^fl/fl controls (Figure 1A,B). This reduction in total arterial protein is expected given that smooth muscle cells, which also express PKD2, are far more abundant than endothelial cells in vessels of this size (Bulley et al., 2018). These data are also consistent with our previous observation that smooth muscle-specific PKD2 knockout reduced total mesenteric arterial wall PKD2 protein by ~ 75% (Bulley et al., 2018). In contrast, SK3, IK, TRPV4, Piezo1, GPR68 and PKD1 (polycystin-1, PC-1), which can form a complex with PKD2 (Qian et al., 1997; Tsiokas et al., 1997), were similar in arteries of both genotypes (Figure 1a and b). Immunofluorescence demonstrated that PKD2 protein was present in endothelial cells of intact arteries from tamoxifen-treated Pkd2^fl/fl mice, but absent in endothelial cells of tamoxifen-treated Pkd2^fl/fl:Cdh5-creERT2 mice (Figure 1C). These results indicate that PKD2 is expressed in endothelial cells and suggest that tamoxifen treatment of Pkd2^fl/fl:Cdh5-creERT2 mice abolishes PKD2 protein. Tamoxifen-treated Pkd2^fl/fl:Cdh5-creERT2 mice will thus be referred to as Pkd2 ecKO mice. Tamoxifen-treated Pkd2^fl/fl mice were used as controls in all experiments.

To investigate physiological functions of endothelial cell PKD2 channels, diameter responses to vasoactive stimuli were measured in pressurized (80 mmHg) third-order mesenteric arteries of Pkd2^fl/fl and Pkd2 ecKO mice. Vasodilation to ACh, a muscarinic receptor agonist, was similar in control and Pkd2 ecKO arteries, suggesting that endothelial cell PKD2 channels do not contribute to this response (Figure 2A and C). Repetitive intravascular flow (15 dyn/cm²) stimuli produced sustained, reproducible and fully reversible vasodilation in pressurized (80 mmHg) mesenteric arteries (Figure 2—figure supplement 2A–D). In pressurized Pkd2 ecKO arteries, mean vasodilation to single on-off flow stimuli were ~35.1% of those in Pkd2^fl/fl arteries (Figure 2B and C). Endothelial cell-denudation abolished vasodilation to both flow and ACh (Figure 2—figure supplement 1A and B). In contrast, endothelial denudation did not alter dilation to sodium nitroprusside, a NO donor, indicating that smooth muscle function was not altered by this procedure (Figure 2—figure supplement 1A and B). To determine the range over which endothelial cell PKD2 channels function, we measured vasoregulation to flow rates that produced shear stress between 3 and 35 dyn/cm². Cumulative increases in flow caused progressive dilation in Pkd2^fl/fl arteries, with a maximum at 27 dyn/cm² (Figure 2D,E). Further increasing flow partially reduced this maximal vasodilatory response (Figure 2D and E). Flow stimulated less vasodilation in Pkd2 ecKO arteries over the range studied (Figure 2D,E; Figure 2—figure supplement 3). Specifically, flow-mediated vasodilation was between ~ 45.5% and 60.1% of that in Pkd2^fl/fl arteries, regardless of rate (Figure 2D and E, Figure 2—figure supplement 3). These data indicate that endothelial cell PKD2 channels function over a broad flow range to stimulate vasodilation in pressurized arteries.

Experiments were performed to examine the hypothesis that endothelial cell PKD2 channel knockout modifies smooth muscle cell contractility, thereby indirectly altering responses to flow. An increase in extracellular potassium (60 mm K⁺) or intravascular pressure (80 mmHg) similarly constricted arteries of Pkd2^fl/fl and Pkd2 ecKO mice, indicating that endothelial cell PKD2 channels or their knockout does not influence depolarization-induced vasoconstriction or myogenic tone, respectively (Figure 2—figure supplement 4A–D). Similarly, arterial passive diameter, determined by removal of extracellular Ca²⁺ from the bath solution, was similar in Pkd2^fl/fl and Pkd2 ecKO arteries (Figure 2—figure supplement 4E). Thus, knockout of endothelial cell PKD2 channels does not modify smooth muscle cell function.

To investigate mechanisms by which endothelial cell PKD2 channels regulate contractility, membrane potential was measured in pressurized mesenteric arteries using glass microelectrodes. At 10 mmHg, the mean membrane potential of Pkd2^fl/fl and Pkd2 ecKO arteries were similar at ~ −62.1 and −61.1 mV, respectively (Figure 3A,B). Increasing intravascular pressure to 80 mmHg similarly depolarized Pkd2^fl/fl and Pkd2 ecKO arteries by ~ 19.3 and 17.7 mV, respectively (Figure 3A,B). Intravascular flow stimulated a mean hyperpolarization of ~ 11 mV in Pkd2^fl/fl arteries (Figure 3A,B). In contrast, flow only hyperpolarized Pkd2 ecKO arteries by ~ 3 mV, or ~ 25.5% of that in controls (Figure 3A,B). These data suggest that flow modulates PKD2 channels in endothelial cells, leading to arterial hyperpolarization and vasodilation.

The contribution of PKD2 channels to currents was investigated in mesenteric artery endothelial cells using patch-clamp electrophysiology. Temporal responses to flow were recorded using the whole-cell configuration with physiological ionic gradients and steady-state voltage of −60 mV. In a static bath, Pkd2^fl/fl endothelial cells generated a mean steady-state inward current of ~−80 pA (Figure 4A,C). Flow stimulated an initial, transient peak increase in mean inward current of ~−21 pA that was followed by a sustained reduction in inward current that plateaued at ~−11 pA in Pkd2^fl/fl cells (Figure 4-C). In the continuous presence of flow, the removal of bath Ca²⁺ increased mean inward current to ~−53 pA in Pkd2^fl/fl cells (Figure 4A,C). In a static bath, mean steady-state inward current was similar in Pkd2^fl/fl and Pkd2 ecKO cells (Figure 4A,C). In contrast, flow activated a transient peak inward current in Pkd2 ecKO cells that was only ~ 15% of that in Pkd2^fl/fl cells (Figure 4A,B). Similarly, the sustained flow-mediated reduction in inward current in Pkd2 ecKO cells was ~ 40% of that in Pkd2^fl/fl cells (Figure 4A,C). In the continuous presence of flow, removal of bath Ca²⁺ resulted alsoin a smaller increase in inward current in Pkd2 ecKO cells than in Pkd2^fl/fl cells (Figure 4A,C). Specifically, under flow Ca²⁺ removal increased inward current only ~ 25 pA in Pkd2 ecKO cells, which was ~ 48.8% of the response in Pkd2^fl/fl cells (Figure 4A,C). This differential response to Ca²⁺ removal in Pkd2^fl/fl and Pkd2 ecKO endothelial cells was due to flow, as inward current in a static condition was similar in cells of both genotypes regardless of whether the bath solution contained Ca²⁺ or was Ca²⁺-free (Figure 4C). These data demonstrate that flow stimulates a biphasic current response that is composed of an initial transient inward current followed by a sustained Ca²⁺-dependent reduction in inward current in endothelial cells. Data also indicate that PKD2 channels contribute to both of these flow-mediated phases.

Attenuation of the flow-mediated sustained reduction in steady-state inward current by both PKD2 knockout and extracellular Ca²⁺ removal suggests the involvement of IK and SK channels. Under flow, the co-application of apamin and Tram-34, SK and IK channel blockers respectively, increased mean inward current by ~ 23.2 pA in Pkd2^fl/fl cells (Figure 5A,B,D). In contrast, the apamin/tram-34-mediated increase in inward current under flow in Pkd2 ecKO cells was only ~ 11.6 pA or ~50% of that in Pkd2^fl/fl cells (Figure 5A,B,D). In a static bath, apamin/Tram-34 produced a far smaller and similar increase in inward current in Pkd2^fl/fl and Pkd2 ecKO cells (Figure 5C,D). These data suggest that flow stimulates PKD2-mediated Ca²⁺ influx that activates SK/IK channels in endothelial cells, leading to a steady-state reduction in inward current.

Next, we tested the hypothesis that flow-stimulates vasodilation through PKD2-mediated SK/IK channel activation. Apamin/Tram-34 reduced both flow- and ACh -induced vasodilation in Pkd2^fl/fl arteries to ~ 77% and 57% of those that occurred in the control condition (Figure 6A,C). Apamin/Tram-34 reduced mean ACh-induced vasodilation to ~ 54% of that in control in Pkd2 ecKO arteries, which was a similar reduction to that in Pkd2^fl/fl arteries (Figure 6A,C). In contrast, apamin/Tram-34 did not alter flow-mediated vasodilation in Pkd2 ecKO arteries (Figure 6A,C). With static intravascular solution, bath application of apamin/Tram-34 did not alter the diameter of pressurized, myogenic Pkd2^fl/fl or Pkd2 ecKO mesenteric arteries, indicating that SK and IK channels are not active in the absence of flow or ACh (Figure 6B,C). These data indicate that flow stimulates PKD2-mediated SK/IK channel activation in endothelial cells to induce vasodilation. In contrast, ACh stimulates vasodilation via a PKD2-independent SK/IK channel-mediated mechanism.

Flow stimulates nitric oxide synthase (NOS) in endothelial cells, but the significance of PKD2 channels to this activation mechanism is unclear (Fleming, 2010; Balligand et al., 2009; Garcia and Sessa, 2019). Phosphorylation of bovine eNOS at serine 1179 and human eNOS at serine 1177 leads to activation (Fulton et al., 1999; Dimmeler et al., 1999). Western blotting was performed to measure both eNOS protein phosphorylated at serine 1176 (p-eNOS (S1176)) and total eNOS protein in mouse mesenteric arteries. Intravascular flow (15 dyn/cm², 5 min, 37°C) increased mean p-eNOS (S1176) protein ~ 1.4 fold in Pkd2^fl/fl arteries, but only ~ 1.08 fold in Pkd2 ecKO arteries (Figure 7A,B). In contrast, flow did not alter total eNOS in either genotype (Figure 7A,B). L-NNA, a NOS inhibitor, reduced flow-mediated vasodilation to ~ 64% of control in pressurized Pkd2^fl/fl arteries and to ~ 83% of control in Pkd2 ecKO arteries (Figure 7C,D). Thus, the L-NNA-induced reduction in flow-mediated vasodilation in Pkd2 ecKO arteries was ~ 47% of that in Pkd2^fl/fl arteries (Figure 7C,D). These data indicate that PKD2 channels are key for flow to activate eNOS in endothelial cells and to elicit vasodilation through this mechanism in mesenteric arteries.

In vitro evidence that endothelial cell PKD2 channels contribute to flow-mediated vasodilation suggests that these proteins may regulate blood pressure. Telemetry measurements were performed using implanted probes to measure systemic blood pressure in Pkd2^fl/fl and Pkd2 ecKO mice. Diastolic and systolic blood pressures were ~ 9 and 14 mmHg higher, respectively, in Pkd2 ecKO than Pkd2^fl/fl mice, which translated to a mean arterial pressure (MAP) that was raised by ~ 11% (Figure 8A,B). Locomotion was similar between genotypes, indicating that the higher blood pressure in Pkd2 ecKO mice was not due to higher activity (Figure 8—figure supplement 1). Echocardiography measurements indicated that cardiac output, fractional shortening, ejection fraction and heart rate were all similar in Pkd2^fl/fl and Pkd2 ecKO mice (Figure 8C–F). Proximal tubule diameter and glomerular area were also similar in kidneys of Pkd2^fl/fl and Pkd2 ecKO mice, indicating no renal dysfunction (Figure 8G–I). These results demonstrate that flow stimulates PKD2 channels in endothelial cells to induce vasodilation and reduce systemic blood pressure.

All procedures were approved by the Animal Care and Use Committee of the University of Tennessee (protocol 17–068.0). Pkd2^fl/fl mice were obtained from the Baltimore PKD Core Center. Cdh5(PAC)-creERT2 mice were a kind gift from Cancer Research UK (Wang et al., 2010). Pkd2^fl/fl mice with loxP sites flanking exons 11–13 of the Pkd2 gene were obtained from the John Hopkins PKD Core. Pkd2^fl/fl mice were crossed with tamoxifen-inducible endothelial cell-specific Cre mice (Cdh5(PAC)-CreERT2, Cancer Research UK) to generate PKD2^fl/fl:Cdh5(PAC)-CreERT2 mice. Male Pkd2^fl/fl:Cdh5(PAC)-CreERT2 or Pkd2^fl/fl mice (8–14 weeks of age) were injected with tamoxifen (1 mg/ml, i.p.) once per day for 5 days and studied 7–14 days after the last injection.

Male mice were euthanized with isoflurane (1.5%), followed by decapitation. Mesenteric artery branches from first- to fifth-order were removed, cleaned of adventitial tissue and placed into ice-cold physiological saline solution (PSS) that contained (in mM): 112 NaCl, 6 KCl, 24 NaHCO₃, 1.8 CaCl₂, 1.2 MgSO₄, 1.2 KH₂PO₄ and 10 glucose, gassed with 21% O₂, 5% CO₂ and 74% N₂ to pH 7.4. Endothelial cells were dissociated by introducing endothelial cell basal media (Endothelial cell GM MV2, Promocell) containing 2 mg/ml collagenase type 1 (Worthington Biochemical) into the arterial lumen and left to incubate for 30–40 min at 37°C. Cells isolated from mesenteric arteries contain multiple different types that exhibit similar visual phenotypes upon enzymatic isolation. To obtain a population of endothelial cells, cell isolate was placed into endothelial cell basal media containing growth supplements (Promocell) that support only endothelial cell survival. Endothelial cells were then studied < 5 days later.

Genomic DNA was isolated from mesenteric arteries using a Purelink Genomic DNA kit (Thermo Fisher Scientific). Reaction conditions used are outlined in the Baltimore PKD Center genotyping protocol (http://baltimorepkdcenter.org/mouse/PCR%20Protocol%20for%20Genotyping%20PKD2KO%20and%20PKD2%5Eneo.pdf). Genotyping was performed using a 3-primer strategy, with primers a (5’-CCTTTCCTCTGGTTCTGGGGAG), b (5’-GTTGATGCTTAGCAGATGATGGC) and c (5’-CTGACAGGCACCTACAGAACAGTG) used to identify floxed and deleted alleles.

Mesenteric artery segments comprising second- to fifth-order vessels were used for Western blotting. For experiments examining flow-mediated regulation of NOS and p-eNOS proteins, a glass cannula was inserted into the first-order branch of a mesenteric artery segment and flow introduced through to fifth-order arteries. Proteins were separated on 7.5% SDS-polyacrylamide gels and blotted onto nitrocellulose membranes. Membranes were blocked with 5% milk and incubated with one of the following primary antibodies: Piezo1 (Proteintech), PKD1 (Santa Cruz), PKD2 (Santa Cruz), SK3 (Abcam), eNOS (Abcam), IK (Alomone), p-eNOS (Cell Signaling), GPR68 (NOVUS), TRPV4 (MilliporeSigma) or actin (MilliporeSigma) overnight at 4°C. Membranes were washed and incubated with horseradish peroxidase-conjugated secondary antibodies at room temperature. Protein bands were imaged using a ChemiDoc Touch Imaging System (Bio-Rad), quantified using ImageJ software and normalized to actin.

Arteries were cut longitudinally and fixed with 4% paraformaldehyde in PBS for 1 hr. Following a wash in PBS, arteries were permeabilized with 0.2% Triton X-100, blocked with 5% goat serum and incubated overnight with PKD2 primary antibody (Rabbit mAB 3374 CT-14/4: Baltimore PKD Center) at 4°C. Arteries were then incubated with Alexa Fluor 555 rabbit anti-mouse secondary antibody (1:400; Molecular Probes) and 4’,6-diamidino-2-phenylindole, dihydrochloride (DAPI) (1:1000; Thermo Scientific) for 1 hr at room temperature. Arteries were washed with PBS and mounted in 80% glycerol solution. DAPI and Alexa 555 were excited at 350 nm and 555 nm with emission collected at ≤ 437 nm and ≥ 555 nm, respectively, using a Zeiss LSM 710 laser-scanning confocal microscope.

Experiments were performed using isolated third- and fourth-order mesenteric arteries using PSS gassed with 21% O₂/5% CO₂/74% N₂ (pH 7.4). Arterial segments 1–2 mm in length were cannulated at each end in a perfusion chamber (Living Systems Instrumentation) continuously perfused with PSS and maintained at 37°C. Intravascular pressure was altered using a Servo pump model PS-200-P (Living systems) and monitored using pressure transducers. Following development of stable myogenic tone, luminal flow was introduced during experiments using a P720 peristaltic pump (Instech). Arterial diameter was measured at 1 Hz using a CCD camera attached to a Nikon TS100-F microscope and the automatic edge-detection function of IonWizard software (Ionoptix). Myogenic tone was calculated as: 100 x (1-D_active/D_passive) where D_active is active arterial diameter and D_passive is the diameter determined in the presence of Ca²⁺-free PSS supplemented with 5 mM EGTA.

Membrane potential was measured by inserting sharp glass microelectrodes (50–90 MΩ) filled with 3 M KCl into the adventitial side of pressurized third- and fourth-order mesenteric arteries. Membrane potential was recorded using a WPI FD223a amplifier and digitized using a MiniDigi 1A USB interface, pClamp 9.2 software (Axon Instruments) and a personal computer. Criteria for successful intracellular impalements were: (Vane, 1994) a sharp negative deflection in potential on insertion; (Edwards et al., 1998) stable voltage for at least 1 min after entry; (Leffler et al., 2006) a sharp positive voltage deflection on exit from the recorded cell and (Garland et al., 2011) a < 10% change in tip resistance after the impalement.

The conventional whole-cell configuration was used to measure steady-state currents in isolated endothelial cells at a holding potential of −60 mV. The bath solution contained (in mM): NaCl 134, KCl 6, HEPES 10, MgCl₂ 1, CaCl₂ 2 and glucose 10 (pH 7.4, NaOH). Ca²⁺-free bath solution was the same composition as bath solution except Ca²⁺ was omitted and 1 mM EGTA added. The pipette solution contained (in mM): K aspartate 110, KCl 30, HEPES 10, glucose 10 and EGTA 1, with total MgCl₂ and CaCl₂ adjusted to give free concentrations of 1 mM and 200 nM, respectively. Free Mg²⁺ and Ca²⁺ were calculated using WebmaxC Standard (http://www.stanford.edu/~cpatton/webmaxcS.htm). The osmolarity of solutions was measured using a Wescor 5500 Vapor Pressure Osmometer (Logan, UT, USA). Currents were filtered at 1 kHz and digitized at 5 kHz using an Axopatch 200B amplifier and Clampex 10.4 (Molecular Devices. Offline analysis was performed using Clampfit 10.4. Flow-activated transient inward current was measured at its peak in each cell. Steady-state inward currents were calculated as the average of at least 45 s of continuous data.

Telemetric blood pressure recordings were performed by the University of Cincinnati Mouse Metabolic Phenotyping Center. Briefly, transmitters (PA-C10, Data Sciences International) were implanted subcutaneously into anesthetized mice, with the sensing electrode placed in the aorta via the left carotid artery. Blood pressures were measured prior, during and following tamoxifen injections (1 mg/ml, i.p) using a PhysioTel Digital telemetry platform (Data Sciences International). Dataquest A.R.T. software was used to acquire and analyze data.

Age- and sex-matched mice were anesthetized with isoflurane and placed on a warm pad on a recording stage of a Vevo 2100 ultrasound machine. The anterior chest was shaved and ultrasound coupling gel applied. Electrodes were connected to each limb and an electrocardiogram was recorded. Two-dimensional (short axis-guided) M-mode measurements were taken at the level of the papillary muscles using an 18–32 MHz MS400 transducer, as previously described (Parks et al., 2016). Images were also recorded in the parasternal long-axis. For analysis purposes, three or more beats were averaged using measurements within the same HR interval (450 ± 50 bpm) for analysis.

Kidney sections were stained with H and E and examined by Probetex, Inc (San Antonio, Texas). Briefly, image analysis was performed to measure glomerular size and tubular cross-sectional diameter. Glomerular size was measured by tracing the circumference of each of 25 random glomeruli and surface area calculated using the polygonal area tool of Image-Pro 4.5 image analysis software calibrated to a stage micrometer. Tubular size was measured using the linear length tool of Image-Pro 4.5 imaging software. The tracing tool was applied at the diameter of cross-sectional profiles of 5 proximal tubules/image (total of 25/section). Glomerular and tubular images were calibrated to a stage micrometer and data was transferred to an Excel spreadsheet and statistical analysis performed by Excel analysis pack.

OriginLab and GraphPad InStat software were used for statistical analyses. Values are expressed as mean ± SEM. Student t-test was used for comparing paired and unpaired data from two populations and ANOVA with Holm-Sidak post hoc test used for multiple group comparisons. p<0.05 was considered significant. Power analysis was performed to verify that the sample size gave a value of > 0.8 if P was > 0.05. Kidney histology, blood pressure and cardiac function experiments were all done single blind, wherein the person performing both the experiments and analysis of the results was not aware of the mouse genotype.