Trophic sympathetic influence weakens pro-contractile role of Cl− channels in rat arteries during postnatal maturation

Membrane transporters and their functional contribution in vasculature change during early postnatal development. Here we tested the hypothesis that the contribution of Cl− channels to arterial contraction declines during early postnatal development and this decline is associated with the trophic sympathetic influence. Endothelium‐denuded saphenous arteries from 1- to 2-week-old and 2- to 3-month-old male rats were used. Arterial contraction was assessed in the isometric myograph, in some experiments combined with measurements of membrane potential. mRNA and protein levels were determined by qPCR and Western blot. Sympathectomy was performed by treatment with guanethidine from the first postnatal day until 8–9-week age. Cl− substitution in the solution as well as Cl−-channel blockers (MONNA, DIDS) had larger suppressive effect on the methoxamine-induced arterial contraction and methoxamine-induced depolarization of smooth muscle cells in 1- to 2-week-old compared to 2- to 3-month-old rats. Vasculature of younger group demonstrated elevated expression levels of TMEM16A and bestrophin 3. Chronic sympathectomy increased Cl− contribution to arterial contraction in 2-month-old rats that was associated with an increased TMEM16A expression level. Our study demonstrates that contribution of Cl− channels to agonist-induced arterial contraction and depolarization decreases during postnatal development. This postnatal decline is associated with sympathetic nerves development.


Scientific Reports
| (2020) 10:20002 | https://doi.org/10.1038/s41598-020-77092-0 www.nature.com/scientificreports/ Cl − homeostasis have not been described for the vascular system, but they were reported for the central nervous system. In neurons, the intracellular Cl − concentration is increased in early postnatal period compared to an adult organism 13 due to high expression level of NKCC1 (Na + -K + -Cl − -cotransporter-1, mediates Cl − influx) and low level of KCC2 (a K + -Cl − -cotransporter, which extrudes Cl − from the cell) 14 . Therefore, activation of GABA A receptors in immature neurons causes Cl − efflux and depolarization, while in mature neurons GABA A receptors mediate Cl − influx and hyperpolarization 13 . We have prevoiusly demonstrated that the contribution of Cl − to arterial contraction decreases during postnatal maturation of vasculature 15 . However, the role of Cl − channels in these changes and the mechanisms behind it remain to be elucidated.
As such a mechanism, we propose here that postnatal reduction in the contribution of Cl − conductance to arterial contraction is associated with trophic influence of sympathetic nerves. This suggestion is based on several findings: (1) the development of sympathetic vasomotor innervation occurs during the first postnatal month in rats 8,16 ; (2) sympathetic nerves are known to have trophic influence on arterial structure and function 8,17,18 ; (3) sympathetic nerves modulate the expression and activity of several ion channels and pumps [19][20][21][22] ; (4) TMEM16A protein and sympathetic nerves exhibit opposite expression gradients along the vascular tree 23,24 . However, experimental evidence for the trophic influences of vascular sympathetic innervation on Cl − channel expression and/or function is not available yet.
Thus, in the present study, we tested the hypothesis that the contribution of Cl − channels to arterial contraction declines during early postnatal development and this decline is associated with the trophic influence of sympathetic nerves.
Cl − substitution depolarizes membrane potential but does not affect intracellular Ca 2+ . In normal PSS, smooth muscle cell membrane potential in endothelium-denuded arteries was less negative in 1to 2-week-old rats compared to 2-to 3-month-old rats (− 42.6 ± 8.6 mV (n = 17) and − 57.7 ± 6.6 mV (n = 13), respectively, p < 0.05). In Cl − -free PSS, membrane potential was less negative in comparison with normal PSS in both age-groups (Table 1). Surprisingly, this significant depolarization was followed only by moderate smooth muscle contraction in both 2-to 3-month-old and 1-to 2-week-old rats (Table 1).
Accordingly, intracellular Ca 2+ in the arteries from 2-to 3-month-old rats was not significantly different under control and in Cl − -free conditions (Fura-2 emission ratio values were 0.24 ± 0.03 a.u. and 0.25 ± 0.03 a.u., The effect of Cl − channel blockers on arterial contraction is stronger in 1-to 2-week-old rats than in 2-to 3-month-old rats. In order to identify whether the elevated Cl − contribution to vasocontraction in 1-to 2-week-old rats is associated with greater impact of Cl − channels, we inhibited Cl − membrane transport with 1 mM DIDS 26 . DIDS did not affect arterial contraction to methoxamine in 2-to 3-month-old rats ( Fig. 3a) but almost completely abolished contraction in 1-to 2-week-old rats (Fig. 3b,c). This suggests larger contribution of Cl − transport to arterial responses of 1-to 2-week-old rats.
The abundance of Cl − channels is higher in the arteries from 1-to 2-week-old rats than in 2-to 3-month-old rats. To explore the molecular background for larger contribution of Cl − transport to arterial contraction in 1-to 2-week-old rats, we studied the expression of Cl − channels in endothelium-denuded arteries from 2-to 3-month-old and 1-to 2-week-old rats. The mRNA expression level of ClC3 was not different in the arteries from two age-groups (Fig. 5a). CFTR mRNA was not detected in saphenous artery of either 2-to 3-month-old or 1-to 2-week-old rats though primer's efficiency was confirmed using rat testis.
In accordance to our functional findings, mRNA content of the CaCC-associated proteins, TMEM16A and bestrophin 3, was significantly higher in endothelium-denuded arteries of 1-to 2-week-old rats compared to 2-to 3-month-old rats (Fig. 5b,c). This was further supported by Western blot showing an increased protein abundance of TMEM16A in endothelium-denuded arteries from 1-to 2-week-old rats compared to 2-to 3-month-old rats (Fig. 5d,e).
Chronic sympathetic denervation preserved larger Cl − contribution to the arterial contraction in 2-to 3-month-old rats. To test our hypothesis that postnatal decline in Cl − contribution is associated with trophic influences of sympathetic nerves, we compared the contribution of Cl − to arterial contraction in 2-month-old chronically sympathectomized and control rats. Rats treated with guanethidine were characterized by decreased body weight (179 ± 35 g (n = 9) and 263 ± 26 g (n = 8) for sympathectomized and control groups, respectively, p < 0.05). Adrenergic nerve plexus was not observed in saphenous artery of sympathectomized rats in contrast to control group suggesting successful sympathetic denervation (Supplementary Figure S3). Saphenous arteries of sympathectomized rats had smaller inner relaxed diameter than control group (454 ± 39 µm (n = 8) and 595 ± 37 µm (n = 8), respectively, p < 0.05) as well as maximum active wall tension (6.1 ± 0.8 N/m (n = 8) and 9.1 ± 1.1 N/m (n = 8), p < 0.05). The sensitivity to methoxamine was estimated by calculating pD 2 Table 1. Active force, membrane potential and pH i in endothelium-denuded arteries of 1-to 2-week-old and 2-to 3-month-old rats in normal and Cl − -free PSS. Data are shown as mean ± SD. Numbers in the parentheses indicate the number of rats. Active force values in each of the experimental conditions were calculated as percentage of the maximum obtained during the CRC for methoxamine. # p < 0.05 Cl − -free PSS versus PSS in the corresponding group (unpaired Student's t-test). www.nature.com/scientificreports/ values that were higher in sympathectomized group compared to the control group (5.83 ± 0.18 (n = 8) and 5.54 ± 0.24 (n = 8), respectively, p < 0.05). In Cl − -free PSS, the arterial contractile responses were reduced in both control (Fig. 6a) and sympathectomized ( Fig. 6b) groups compared to normal PSS. Importantly, Cl − substitution reduced area under CRC in sympathectomized group more than in control group (Fig. 6c) suggesting that Cl − contribution to the contraction is increased in chronically denervated arteries.
Furthermore, to get an insight into the molecular background for augmented contribution of Cl − to the regulation of arterial contraction in sympathectomized group, we looked at mRNA expression levels of TMEM16A and bestrophin 3. The expression of TMEM16A was larger in sympathectomized group compared to control group (Fig. 7a), while the difference in bestrophin 3 mRNA levels did not achieve significance (Fig. 7b).

Discussion
Here we studied the developmental alterations in the depolarizing and pro-contractile role of Cl − in peripheral artery of systemic circulation. Arteries of 1-to 2-week-old rats demonstrated higher contribution of Cl − conductance to the regulation of arterial contraction as compared to 2-to 3-month-old rats. This was also suggested by larger effects of Cl − substitution, DIDS and MONNA in the arteries of younger age-group. Moreover, we showed www.nature.com/scientificreports/ that vascular smooth muscle cells of 1-to 2-week-old rats contained larger quantity of TMEM16A and bestrophin 3 in comparison to 2-to 3-month-old group. Finally, we revealed that chronic sympathetic denervation prevents postnatal decline of Cl − contribution to the contraction in rat arteries. Cl − substitution in PSS reduced arterial contraction to methoxamine in the arteries from 2-to 3-month-old rats. This is in accordance with previous findings showing reduced sensitivity to noradrenaline of mesenteric arteries from adult rats and mice in Cl − -free PSS 28 . Moreover, in the present study, we evaluated the arterial contraction of 1-to 2-week-old rats in Cl − -free PSS and found that it was markedly suppressed compared to normal PSS. This result is in accordance with our previous observation in another study 15 . Importantly, the reduction of contraction in Cl − -free PSS was larger in the arteries of 1-to 2-week-old rats compared to adult rats.
Notably, substitution of Cl − ions in PSS by aspartate caused considerable depolarization of arterial smooth muscle cells from both 1-to 2-week-old and adult rats. Accordantly, previous data reported that Cl − substitution by aspartate or sulfate salts depolarized rat mesenteric arteries 28 . Intriguingly, substitution of Cl − depolarized the arteries from both age-groups but did not cause severe increase in vascular tone (an elevation of tone was appr. 4% and 8% of maximum active force for 2-to 3-month-old and 1-to 2-week-old rats, respectively). Accordantly, the depolarization was not accompanied by an increase of intracellular Ca 2+ in vascular smooth muscle cells. This can be due to change of pH i that accompanies substitution of extracellular Cl − in arteries of both 1-to 2-week-old and 2-to 3-month-old rats. Our results on pH i alkalization due to substitution of extracellular Cl − are in accordance with previous reports 29 . These changes of pH i might in turn modulate enzymatic activity in smooth muscle  www.nature.com/scientificreports/ cells 30 . However, the exact mechanism responsible for inhibition of contraction despite strong depolarization of smooth muscle cells is unknown and beyond the scope of the present study. Importantly, an alkalization in response to Cl − substitution was observed in both age-groups and was similar between them. Thus, this cannot explain the difference in methoxamine-induced contractions. The importance of extracellular Cl − ions for depolarization of arterial smooth muscle cells is also significantly larger in 1-to 2-week-old than in 2-to 3-month-old rats. In arteries of 2-to 3-month-old rats, methoxamineinduced depolarization was decreased in Cl − -free PSS. Similarly, in rat mesenteric arteries noradrenaline-induced depolarization was reduced in Cl − -free PSS 28 . In contrast to arteries of adult rats, in the arteries of younger rats substitution of Cl − completely abolished depolarization. We did not measure the Cl − currents in smooth muscle cells of 1-to 2-week-old rats in order to compare the Cl − conductance between the different age-groups. However, our membrane potential measurements suggest that transmembrane Cl − ion gradient is essential for agonistinduced depolarization and contraction of arteries from 1-to 2-week-old rats. A higher mRNA expression of NKCC1 might point to the increased Cl − influx in smooth muscle cells of 1-to 2-week-old compared to 2-to 3-month-old rats (Supplementary figure S4) but functional significance of this difference needs to be validated in future studies.
We further tested the contribution of Cl − conductance to the arterial contraction pharmacologically. We used two different blockers, DIDS and MONNA. DIDS is a conventional blocker widely used to inhibit Cl − conductance 31,32 . Structurally different from DIDS, MONNA was suggested to be a putative TMEM16A channel blocker 27 . Importantly, both blockers showed stronger effects in the arteries of 1-to 2-week-old rats compared to 2-to 3-month-old animals. www.nature.com/scientificreports/ The pharmacology of Cl − membrane transport is relatively poor. DIDS was shown to inhibit other membrane transporters including anion exchanger AE2 33 , K ATP channels 34 and Ca 2+ -ATPase 35 . Similarly, MONNA was reported to have Cl − -independent vasorelaxing effects in the arterial wall 28 . However, in this study, the effects of both blockers were qualitatively similar to the effects of Cl − -free PSS. Based on these independent approaches, we concluded that the contribution of Cl − channels to arterial contraction is higher in the arteries of 1-to 2-weekold rats compared to 2-to 3-month-old rats.
The molecular background of augmented contribution of Cl − conductance in 1-to 2-week-old rats was addressed by comparing the expression of Cl − channels in these two age-groups. In accordance with our functional data, smooth muscle cell expression of TMEM16A and bestrophin 3 was considerably higher in 1-to 2-week-old rats than in 2-to 3-month-old rats.
Two antibody-stained bands seen in TMEM16A Western blot may be a result of either different splice variants or post-transcriptional modification of the protein. In fact, TMEM16A was shown to have many splice variants in different tissues 36 , including vasculature 37 . TMEM16A is known to be essentially important for Ca 2+ -activated Cl − conductance in vascular smooth muscle cells 37-40 although its specific contribution to Cl − channel formation remains uncertain 4 . In accordance with previous observations the expression of TMEM16A and bestrophin 3  www.nature.com/scientificreports/ is interdependent i.e. the expression of one protein can affect another protein's expression 38 . We suggest similar relation in the early postnatal development where an augmented expression of TMEM16A in the arteries from 1-to 2-week-old rats leads to an upregulation of bestrophin 3. However, the mechanism and importance of this interdependence remains to be elucidated.
To the best of our knowledge, the relative expression of CaCCs has never been explored in developing arteries. However, higher TMEM16A expression during early stages of development was previously described in olfactory epithelium of mice 41 . Moreover, larger abundance of bestrophin family member, bestrophin 1, was shown in murine eyes in early postnatal development 42 . Thus, high level of CaCCs in early postnatal period discovered in the present study is not the unique feature of immature smooth muscle cells and may accompany the development of different tissues.
Sympathetic nerves are well-known to have trophic influence on arterial structure and function 17,18 . Arterial sympathetic innervation develops during the first month of postnatal development in rats 8,16 . In order to identify the trophic influence of sympathetic nerves, we prevented their development by treating rats with guanethidine from the first postnatal day until the age of 8-9 weeks. Despite delayed growth rate, sympathectomized rats did not demonstrate any alterations in behavior activity or health status, which is in line with previously reported data 43 . Chronic sympathetic denervation increased smooth muscle sensitivity of agonist-induced contraction. This is in accordance with previous findings showing the increase in arterial sensitivity to contractile stimuli 21,44 .
Our functional data with Cl − -free PSS for the first time demonstrate that sympathectomized rats had larger Cl − contribution to the agonist-induced contraction than matched control group. We did not perform experiments with MONNA or DIDS on arteries from sympathectomized rats. However, our suggestion is supported by the fact that larger Cl − contribution to the agonist-induced contraction is associated with the increased level of TMEM16A mRNA in smooth muscle cells of sympathectomized rats compared to control animals. Our results show that development of sympathetic nerves and a change in the Cl − dependence of contraction are related processes. Therefore, we conclude that in accordance with our hypothesis the postnatal declines in Cl − contribution to the regulation of arterial contraction and CaCCs expression are associated with trophic influences of sympathetic nerves but the exact mechanisms are still unknown.
To conclude, our comprehensive study for the first time demonstrates that contribution of Cl − channels to the arterial contraction and agonist-induced depolarization of smooth muscle cells is larger at early postnatal development. This is enabled by higher expression of TMEM16A and bestrophin 3, the molecules responsible for the Ca 2+ -activated Cl − conductance, in 1-to 2-week-old than in 2-to 3-month-old arteries. We suggest that higher contribution of the Ca 2+ -activated Cl − conductance in vasculature of immature organism facilitates their vasoconstriction by counteracting the high activity of many types of K + channels in 1-to 2-week-old rats 12 . Our study uncovered a novel target of trophic influences of sympathetic nerves. We showed for the first time that trophic influences of sympathetic nerves govern postnatal decline in the functional contribution of Cl − channels (Fig. 8). To study alteration in the contribution of Cl − channels to arterial contraction with age, we used 10-to 14-dayold (referred as 1-to 2-week-old rats) and 2.5-to 3.5-month-old (i.e. 2-to 3-month-old rats) male Wistar rats.

Animals. Experiments were conducted in accordance with
To study the trophic influence of sympathetic nerves, we performed chronic pharmacological sympathectomy 45 . The animals were chronically treated with guanethidine monosulfate (Santa Cruz, dissolved in 0.9% NaCl) starting from the first postnatal day until the age of 8-9 weeks. Rat litters were randomly divided into two groups immediately after birth; sympathectomized group that received injections of guanethidine (n = 6 litters) and control rats (n = 5 litters). Guanethidine was injected subcutaneously 6 days per week in the dose of 25 mg/kg for 2 weeks (1-15 postnatal days) followed by 50 mg/kg (from the 16th postnatal day to 8-9-week age). Control animals received similar volume of 0.9% NaCl (i.e. 1.7-2.5 ml/kg). The injections of guanethidine were performed by skilled researcher using syringes with 30G needles. All rat pups survived and did not show any sign of discomfort during injections. Two days after the last injection, the rats were sacrificed and saphenous arteries were isolated for further wire myograph and qPCR studies. Adrenergic nerve plexus was visualized using glyoxylic acid staining 46 using Axiovert 200 microscope and S25 filter set.
Wire myograph experiments. Several 2-mm-long segments of saphenous artery were dissected and mounted in the isometric myograph (Danish Myo Technology (DMT) A/S, Denmark). The endothelium was removed by rubbing inner lumen of the artery with a rat whisker. Force transducer signals were digitalized at 10 Hz and recorded on a PC hard drive using an analogue-to-digital converter (E14-140 M, L-CARD, Russia) and the PowerGraph 3.3 software (DISoft, Russia, https ://www.power graph .ru/). PSS in myograph chamber was warmed to 37° and arterial segments were gradually stretched to 0.9d 100 ; d 100 is an inner diameter of relaxed artery subjected to the transmural pressure of 100 mmHg 47 . A standard start-up procedure consisted of (1) noradrenaline (10 µM) application; (2) an application of acetylcholine (10 µM) to methoxamine (1-3 µM) preconstricted artery (a test for successful endothelium removal) and (3)  www.nature.com/scientificreports/ The experimental protocol included (unless other specified) two CRC for the α 1 -adrenoceptor agonist methoxamine (in the range from 0.01 to 100 µM). The first CRC was performed in normal PSS for all studied segments. Subsequently, some arterial segments were subjected to experimental challenges while other served as a timecontrol. In part of experiments, the arterial preparations were further incubated for 30 min in Cl − -free PSS (for composition see below) or in normal PSS (i.e. time-control). Five to 10 min after beginning of the incubation period the preparations were stimulated with 10 µM of methoxamine for 5 min followed by washout. This was done in order to stimulate Cl − efflux from smooth muscle cells 28 . For pharmacological challenges in another part of experiments, one arterial segment was exposed for 30 min to Cl − channel blocker, either 1 mM DIDS 26,48 or 3 µM MONNA, while another arterial segment was incubated with a vehicle. The concentration of MONNA 3 µM was selected based on our previous studies, where it caused submaximal relaxation 28 .
Active force values were calculated by subtraction the passive force values recorded in the preparation with fully relaxed smooth muscle (in PSS for vessel isolation, for composition see section "Solutions", supplemented with NO-donor DEA/NO, 1 µM) from baseline and each methoxamine concentration. All active force values in the second CRC were expressed as the percentage of the maximum active force achieved during the first CRC. In order to compare the effects of experimental challenges in two age-groups, the area under individual second CRCs to methoxamine were calculated in GraphPad Prism 7.0 (La Jolla, CA, USA, https ://www.graph pad.com/ scien tific -softw are/prism /) and then, expressed as a percentage of the mean value for matched time-control group. The arterial sensitivity to agonist was assessed as pD2 (i.e. the negative logarithm base 10 of EC 50 -the concentration of agonist, which causes a half-maximum effect) that was calculated by fitting the second individual CRCs to sigmoidal function (variable slope) using GraphPad Prism 7.0. pD2 values were obtained for each individual curve and then, their mean and SD values and were calculated for each group. . This was further supported by control experiments where 3 M KCl salt-agar bridge was placed between the Ag/AgCl pellet and the bath solution without any significant implications for the measurements. We therefore did not correct the data for the junction potential.
A segment of saphenous artery was mounted in a wire myograph (DMT A/S, Denmark). To exclude distortion of the electrode by gas bubbles, warm (37 °C) and equilibrated with 5% CO 2 in O 2 PSS was supplied from a separate reservoir to the myograph chamber (10 ml volume) by a peristaltic pump at the rate of 4.5 ml/min. The standard start-up procedure and methoxamine CRC were performed as described above. Then, the solution was changed to Cl − -free PSS for at least 30 min (with Cl − -depleting activation with 10 µM methoxamine-see above) and membrane potential was measured first in the absence and then in the presence of methoxamine (10 µM). In time-control group similar measurements were done in normal PSS.
Measurement of mRNA expression levels in arterial tissue by qPCR. Measurement of mRNA expression level was performed as described previously 12 . Saphenous arteries were isolated and cut in approximately 8-mm long segments, which were quickly mounted in an ice-cooled analogue of a wire myograph chamber and endothelium was removed using a rat whisker. Arteries were stored in RNA-later solution (Qiagen) at − 20 °C. RNA was extracted using the kit from Evrogen according to the manufacturer's instructions with RLT-buffer (Qiagen). All RNA samples were processed with DNase I (Fermentas, 1000 U/mL). RNA concentration was measured by a NanoDrop 1000 (Thermo Scientific, USA) and then all sample concentrations were adjusted to 70 ng/µL by dilution. Reverse transcription was performed using the MMLV RT kit (Evrogen, Russia) according to the manufacturer's manual. qPCR was run in the RotorGene6000 (Corbett Research, Australia) using qPCRmix-HS SYBR (Evrogen).
All primers used in the study were synthesized by Evrogen, their sequences are listed in Table 2. Gene expression levels were calculated using the RotorGene6000 software. Primer efficiency was identified using the LinRegPCR 2018.0 Software 49 (www.medis chebi ologi e.nl). The primer efficiencies for all studied genes were in the range 1.8-2.0. The expression level of mRNA was calculated as E −Ct , where E is primer efficiency and Ct is cycle threshold. These values were normalized to the geometric mean of the two housekeeping RNAs (Gapdh and Rn18s), detected in the same sample. Data are expressed as the percentage of mean value of the 2-to 3-monthold group and shown as median and interquartile range. Endothelium removal was confirmed by more than 15-fold reduction in eNOS mRNA content in endothelium-denuded versus endothelium-intact preparations (data not shown).

Western blotting.
Western blotting experiments were performed as described previously 23 . Endotheliumdenuded arterial segments were snap frozen in liquid nitrogen and lysed in ice cold Pierce lysis buffer using a pellet pestle (Sigma Aldrich, Denmark). The homogenate was sonicated for 45 s and centrifuged at 13,000 rpm for 10 min to collect supernatant. Protein quantification in the supernatant was carried out using bicinchoninic acid (BCA) Protein Assay kit (Pierce, Thermo Scientific, USA).
Ten µg protein were diluted with 4 × Laemmli sample buffer (Bio-Rad, USA) and 50 mM dithiothreitol (DTT) and denatured at 70 °C for 10 min. Gels (4-12% Bis-Tris, Bio-Rad, USA) were loaded and separation was performed by electrophoresis followed by transfer onto polyvinylidene fluoride (PVDF) membranes (Merck Millipore, Ireland). The membranes were blocked for 2 h with 5% non-fat dry milk (Blotting-Grade, Bio-Rad, www.nature.com/scientificreports/ USA) in TBS-T. Membranes were cut into two at approximately 75 kDa marker and incubated overnight at 4 °C in blocking buffer with primary eNOS (upper part of the membrane, control for endothelium removal; 1:500; Abcam) or TMEM16A (upper part of the membrane; 1:500; Abcam) or GAPDH (lower part of the membrane; 1:5000; Cell Signalling Technology, USA) antibody. The next day, the membranes were washed in TBS-T and incubated with horse-radish-peroxidase conjugated secondary goat-anti-rabbit antibody (1:2000, Cell Signalling Technology) for 2 h at room temperature. After washings in TBS-T, protein was detected using a chemiluminescent imaging system, ImageQuant LAS 4000 (GE Healthcare Life Sciences, USA, https ://www.cytiv alife scien ces.com/), quantified using ImageJ (v1.48, NIH, https ://image j.nih.gov/) and normalized to GAPDH from the same sample. Data are expressed as the percentage of mean value of the 2-to 3-month-old group and shown as median and interquartile range.
Ca 2+ -fluorimetry. Ratiometric measurements of intracellular Ca 2+ were made simultaneously with the isometric force measurements similarly to previously described 50 . Arterial segments mounted in myograph were loaded with Fura 2-acetoxymethyl ester (2.5 µM; Fura 2/AM) in DMSO with 0.1% (wt/vol) cremophor and 0.02% (wt/vol) pluronic F127 for 2 h. Arteries were excited by a 75 W xenon light source alternately at 340 and 380 nm, and emitted light was measured at 515 nm. Background fluorescence was determined before loading and subtracted from the measurements. Fluorescence was collected and stored digitally using Felix32 software (version 1.2, Photon Technology, USA, https ://pti-felix 32.softw are.infor mer.com/). Intracellular Ca 2+ was expressed as the ratio of fluorescence during excitation at 340 nm and 380 nm. pH-fluorimetry. Ratiometric measurements of pH i were made simultaneously with isometric force of the artery mounted in myograph as described previously 30 . Arterial preparation was loaded with 5 µmol/L AM-form of pH-sensitive fluorophore 2′,7′-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein (BCECF) in 0.02% DMSO for approximately 30 min at 37 °C. The excitation fluorescence ratio was recorded in BCECF-loaded arteries as the 510-nm emission during alternating 440-and 495-nm light excitation. The collected signal was converted to estimates of pH i using a high-K + -nigericin technique 51 . Solutions.