Dissecting the interaction between HSP70 and vascular contraction: role of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hbox{Ca}^{2+}$$\end{document}Ca2+ handling mechanisms

Heat-shock protein 70 (HSP70) is a ubiquitously expressed molecular chaperone with various biological functions. Recently, we demonstrated that HSP70 is key for adequate vascular reactivity. However, the specific mechanisms targeted by HSP70 to assist in this process remain elusive. Since there is a wealth of evidence connecting HSP70 to calcium (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hbox {Ca}^{2+}$$\end{document}Ca2+), a master regulator of contraction, we designed this study to investigate whether blockade of HSP70 disrupts vascular contraction via impairment of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\text{Ca}}^{2+}$$\end{document}Ca2+ handling mechanisms. We performed functional studies in aortas isolated from male Sprague Dawley rats in the presence or absence of exogenous \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hbox {Ca}^{2+}$$\end{document}Ca2+, and we determined the effects of VER155008, an inhibitor of HSP70, on \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hbox {Ca}^{2+}$$\end{document}Ca2+ handling as well as key mechanisms that regulate vascular contraction. Changes in the intracellular concentration of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hbox {Ca}^{2+}$$\end{document}Ca2+ were measured with a biochemical assay kit. We report that blockade of HSP70 leads to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hbox {Ca}^{2+}$$\end{document}Ca2+ mishandling in aorta stimulated with phenylephrine, decreasing both phasic and tonic contractions. Importantly, in \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hbox {Ca}^{2+}$$\end{document}Ca2+ free Krebs’ solution, inhibition of HSP70 only reduced the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hbox {E}_{\mathrm{max}}$$\end{document}Emax of the phasic contraction if the protein was blocked before IP3r-mediated \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hbox {Ca}^{2+}$$\end{document}Ca2+ release, suggesting that HSP70 has a positive effect towards this receptor. Corroborating this statement, VER155008 did not potentiate an IP3r inhibitor’s outcomes, even with partial blockade. In another set of experiments, the inhibition of HSP70 attenuated the amplitude of the tonic contraction independently of the moment VER155008 was added to the chamber (i.e., whether it was before or after IP3r-mediated phasic contraction). More compelling, following re-addition of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hbox {Ca}^{2+}$$\end{document}Ca2+, VER155008 amplified the inhibitory effects of a voltage-dependent \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hbox {Ca}^{2+}$$\end{document}Ca2+ channel blocker, but not of a voltage-independent \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hbox {Ca}^{2+}$$\end{document}Ca2+ channel inhibitor, indicating that HSP70 has a positive impact on the latter. Lastly, the mechanism by which HSP70 modulates vascular contraction does not involve the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hbox {Ca}^{2+}$$\end{document}Ca2+ sensitizer protein, Rho-kinase, nor the SERCA pump, as blockade of these proteins in the presence of VER155008 almost abolished contraction. In summary, our findings shed light on the processes targeted by HSP70 during vascular contraction and open research avenues for potential new mechanisms in vascular diseases.


Results
Impact of HSP70 blockade in the total concentration of Ca 2+ in PE-stimulated aorta. We have previously demonstrated that HSP70 is key to PE-induced vascular contraction 4 . In Fig. 1A, we confirmed that, upon inhibition of HSP70, aortic rings challenged with a single dose of PE ( 10 µmol/l ) display a reduction in the force generated. Since vascular responses rely on the intracellular concentration of free Ca 2+ , in this study, we measured the levels of this cation in control (CTL)-and VER155008-treated aortic samples stimulated with this α -1 adrenergic agonist. We found that, compared with CTL samples, rings incubated with VER155008 have a substantial decrease in the total concentration of free Ca 2+ (Fig. 1B), which indicates that the mechanism targeted by HSP70 to influence vascular contraction might involve Ca 2+ dynamics.
Impact of HSP70 blockade in PE-induced phasic contraction: role of IP3r-mediated intracellular Ca 2+ release. Inhibition of HSP70 decreases PE-induced phasic contraction in the aorta 4 . Here, to better understand this process, we performed functional studies in the absence of exogenous Ca 2+ . Under this condition, PE induced a transient contractile response (Fig. 2A). Then, we confirmed that the blockade of HSP70 (VER-Before) decreases the total and the maximum response elicited by this agonist (Fig. 2B,C, respectively). Additionally, inhibition of maximum response only occurred if HSP70 was targeted before IP3r-mediated intra- www.nature.com/scientificreports/ cellular Ca 2+ release (Fig. 2C), which suggests that blockade of HSP70 attenuates phasic contraction by blunting the response elicited by this receptor. Next, to build into this finding, we performed functional studies in the presence of 2-aminoethoxydiphenyl borate (2-APB), an inhibitor of the IP3r. Noteworthy, it is known that 2-APB also inhibits receptor-and storeoperated Ca 2+ entry 23,24 , and therefore, this inhibitor can be used to study phasic and tonic vascular responses to α -1 adrenergic stimulation 12 . In this set of experiments, aiming at blocking the IP3r, 2-APB was added at the moment we replaced the Krebs' solution with 0[Ca 2+ ] Krebs' solution. As expected, 2-APB produced a dosedependent inhibitory effect (Fig. 2D). The hyporesponsive pattern detected in samples incubated with 2-APB was similar to that evoked by samples exposed to VER155008 (Fig. 2E vs. D). In subsequent experiments, rings were stimulated with PE in the presence of 2-APB and VER155008. The HSP70 inhibitor did not potentiate the total nor the maximum inhibitory effect of 2-APB (Fig. 2F,G, respectively), which, again, indicates that HSP70 affects PE-induced phasic contraction by acting upon IP3r-mediated mechanisms. HSP70 contributes to phenylephrine-induced phasic contraction via IP3r-mediated intracellular Ca 2+ release. Aortic rings were challenged with phenylephrine ( 10 µmol/l ) in 0[Ca 2+ ] Krebs' solution in the presence of vehicle (CTL) or VER155008 ( 10 µmol/l , DMSO diluted), which was added before or after IP3rmediated Ca 2+ release (A). In another set of experiments, samples were stimulated in the presence of (D) 2-aminoethoxydiphenyl borate (2-APB) (50 and 100 µmol/l , DMSO diluted), or (E) the combination of 2-APB and VER155008. Inhibitors were added before IP3r-mediated Ca 2+ release. (B,F) Area under the curve and (C,G) E max . Panels (B,C) and (F,G) use the same color scheme as panels (A) and (D,E), respectively. Data are expressed as mean ± SEM. n = 14 for CTL and VER-before, n = 9 for VER-after, and n = 6 for all other groups, * p < 0.05 vs. CTL and # p < 0.05 vs. VER-before. www.nature.com/scientificreports/ Impact of HSP70 blockade in PE-induced tonic contraction: role of voltage-dependent and -independent Ca 2+ channels. Subsequently, we sought to identify the mechanism by which inhibition of HSP70 reduces PE-induced vascular tonic contraction. To accomplish this goal, we evaluated the force developed by aortic rings after restoring the initial concentration of Ca 2+ to the Krebs' solution. The re-addition of Ca 2+ triggered a prolonged contraction phenotype in aortic rings (Fig. 3A). Then, in a first set of experiments, we added VER155008 before and after PE-induced IP3r-mediated contraction. The presence of VER155008 significantly decreased the total and the maximum response elicited by PE (Fig. 3B,C, respectively), and this hyporesponsive pattern was independent of the moment VER155008 was added to the chamber (i.e., whether it was before or after IP3r-mediated phasic contraction). Since Ca 2+ influx following α -1 receptor stimulation is mediated by voltage-dependent and independent channels, we next investigated whether blockade of HSP70 impairs tonic contraction by interfering with these channels. Here, we show that the response exhibited by aortic rings, in the presence of VER155008 and following the re-addition of Ca 2+ , was similar to the one observed in samples incubated with verapamil ( In fact, no statistical difference was observed between the use of the HSP70 inhibitor and verapamil or 2-APB in the total nor the maximum response elicited by the agonist (Figs. 4C,D; 5C,D, respectively). More interestingly, we observed that the combination of VER155008 and verapamil increased verapamil's inhibitory effect, which reveals that the HSP70 inhibitor potentiates the effects of this L-type Ca 2+ channel (LTCC) blocker and points to these drugs impacting different mechanisms. Corroborating this statement, the combination of VER155008 and 2-APB, did not produce a synergism (Fig. 5B), which corroborates the notion that these drugs affect the same mechanism. Impact of HSP70 blockade in PE-induced phasic/tonic contraction: role of Rho-kinase and SERCA pump. Rho-kinase contributes to PE-induced contraction by promoting Ca 2+ sensitization [19][20][21] .
Therefore, we next investigated the impact of a Rho-kinase inhibitor in the phasic and tonic contractions elicited by this agonist. The presence of Y27632, a Rho-kinase inhibitor, did not affect the phasic E max of PE in 0[Ca 2+ ] Kreb's solution (Fig. 6A). However, it consistently reduced the total response evoked by the drug (Fig. 6A). Similarly, the combination of Y27632 and VER155008 did not exacerbate the the drugs' effect on maximum response, but it decreased the total response elicited by PE (Fig. 6B). Next, we concentrated our efforts to evaluate the effects of Y27632 and VER155008 in samples following the re-addition of Ca 2+ to the Kreb's solution. The combination of Y27632 and VER155008 significantly reduced the drugs' response in comparison with effects of these inhibitors independently (Fig. 6B vs. A), which suggests that these drugs are acting upon different mechanisms. Details about the total and maximum response elicited by phenylephrine can be found in Fig. 6C,D, respectively.
Up to this point, our results corroborate the notion that blockade of HSP70 weakens vascular contraction by decreasing the intracellular concentration of free Ca 2+ . Thus, we next investigated if a SERCA pump inhibitor, which prevents Ca 2+ transfer from the cytoplasm to the SR, would affect the outcome measured in the presence of the HSP70 inhibitor. From our observations, it is clear that blocking the SERCA pump did not affect PE-induced phasic contraction, but it completely disrupted the vessels' ability to develop tonic contraction (Fig. 7A). The www.nature.com/scientificreports/   www.nature.com/scientificreports/ latter was maximized in the presence of VER155008 (Fig. 7B), and points toward a complete impairment of Ca 2+ handling mechanisms in aortic rings. The results of the total and maximum response elicited by the agonist are presented in Fig. 7C,D, respectively.

Discussion
In the present study, we expanded the concept that HSP70 is a role player in vascular physiology, as we show, for the first time, that proper vascular Ca 2+ handling stimulated by PE requires this protein. Our findings are of utmost importance because Ca 2+ is the chief mediator of vascular contraction and fluctuations in its intracellular levels modulate the contractile phenotype of blood vessels 1 . Therefore, notwithstanding the limitations of our study, it fills in a knowledge gap in vascular biology and it will, in turn, guide future research in this field, particularly due to the emergent link between HSP70 and cardiovascular/renal diseases [25][26][27][28][29] . Throughout this study, we applied a well-established, yet indirect, protocol to investigate Ca 2+ changes in the presence of VER155008, which functions as an ATP-competitive inhibitor 30 . To overcome this limitation and strengthen our claims, a biochemical assay kit was also used to evaluate the free levels of this cation. In Fig. 1B, we specifically show that inhibition of HSP70 decreases the total levels of free Ca 2+ , which ultimately, reduces the force of contraction. Here, it is important to recognize some pitfalls of this method. In recent years, researchers have evaluated Ca 2+ fluctuations with a fluorescent indicator, such as fura-2, and since the concentration of Ca 2+ rapidly change in vascular structures, it is considered a more accurate measurement for this cation. Additionally, in order to perform the biochemical assay, samples need to be homogenized and it might include mitochondrial Ca 2+ , which does not affect vascular smooth muscle contraction. Still, we argue that the claim that blockade of HSP70 impairs vascular contraction by affecting Ca 2+ handling mechanisms are based on our collective findings, which include not only the indirect measurement of the free levels of Ca 2+ , but also an extensive and well-detailed set of functional studies as well as evidence from previous studies. The literature shows a complex interaction between HSP70 and Ca 2+ . In fact, the ATPase domain of HSP70 binds two Ca 2+ ions 9 and changes in the intracellular concentration of this cation modulates the expression of HSP70 10,11 . Additionally, the genetic deletion of this protein impairs Ca 2+ homeostasis in cardiac and skeletal muscle 7,8 . Thus, our data elegantly builds upon previous knowledge as it uncovers a new biological process where HSP70 interacts with Ca 2+ .
It is known that phasic contraction in response to PE in the aorta involves IP3r-mediated Ca 2+ release from the SR 12,15 . We previously demonstrated that, in vessels stimulated with this α − 1 agonist when HSP70 is blocked, there is a reduction in the amplitude of the fast component 4 . However, it was yet-to-be-determined if a direct relationship exists with the IP3r. Here, we confirmed that blockade of HSP70 also weakens PE-induced phasic contraction in aorta under 0[Ca 2+ ] Krebs' solution ( Fig. 2A). Similar results were also detected in samples incubated with 2-APB, an inhibitor of the IP3r (Fig. 2D). Interestingly, we found that the combination of VER155008 and 2-APB does not augment the latter's inhibitory effect (Fig. 2E vs. D). Such findings indicate both inhibitors acting upon similar mechanisms. Corroborating this statement, we also detected that, if we block HSP70 after PE-mediated IP3r-induced phasic E max , the impact of VER155008 is abolished ( Fig. 2A,C), which strongly suggests that HSP70 contributes to PE-induced phasic contraction via IP3r-mediated Ca 2+ release. In a counterintuitive manner, it has been previously demonstrated that upregulation of HSP70 reduces IP3r protein levels following ischemia/reoxygenation in PC12 cells 31 . Here, it is important to consider that Ca 2+ overload can occur during ischemia/reoxygenation 32 , and as discussed by the authors, the ultimate outcome observed was that HSP70 contributes to maintaining Ca 2+ homeostasis in these cells 31 . In this sense, our results align with the previous literature as we also show that the precise control of Ca 2+ handling requires HSP70.
Next, we turned our attention to try at understanding the mechanism(s) by which HSP70 affects vascular tonic contraction. A previous study demonstrated that PE-induced tonic contraction includes Ca 2+ influx via voltage-dependent and independent channels 12 . While there is limited information about an interaction between HSP70 and LTCC, it has been suggested that HSP70 might act by inhibiting voltage-gated Ca 2+ channel to prevent Ca 2+ overload, and consequently, apoptosis 33 . However, as highlighted by the authors experimental evidence was lacking. Here, we found that the HSP70 inhibitor potentiates the inhibitory effect of an LTCC blocker (Fig. 4B  vs. A), but not of a non-selective inhibitor of voltage-independent Ca 2+ channels (Fig. 5B vs. A). Therefore, our data corroborate the idea that HSP70 contributes to tonic contraction by acting upon voltage-independent Ca 2+ channel-facilitated Ca 2+ influx. In support of this statement, we also confirmed that VER155008 reduces tonic contraction independently of the moment it is added to the chamber (i.e., whether it was before or after IP3rmediated phasic contraction) (Fig. 2). Noteworthy, we used 2-APB to target voltage-independent Ca 2+ channels, which has an inhibitory effect towards NSCC and CRAC channels 12,34 . Consequently, we are unable to pinpoint the exact channel targeted by the HSP70 inhibitor. Another possibility one should consider in this context is the fact that a previous study has demonstrated that the constitutive HSP70 interacts with lipid membranes leading to the generation of a functional ATP-dependent cationic pathway 35 . Therefore, further studies are required to uncover the precise mechanism by which HSP70 targets Ca 2+ influx in PE-stimulated aorta.
Subsequently, we focused on determining the contribution of Rho-kinase, which promotes Ca 2+ sensitization, and therefore, affects the contractile phenotype of vascular structures 19,20 . From our data, it is clear that the combination of VER155008 with Y27632 amplifies the hyporesponsive pattern observed in the aorta in comparison with blocking these proteins independently (Fig. 6). Given our findings regarding the role of HSP70 in Ca 2+ influx, one can argue that these results were to be expected, especially because Rho-kinase impacts vascular contraction by inhibiting the myosin light chain phosphatase, which, in turn, prevents relaxation 36,37 . Therefore, it appears that two different mechanisms were targeted in this set of experiments. Corroborating this statement, a previous study found that heat shock-mediated vascular hypercontractility does not directly involve Rho-kinase 38 .
Finally, we investigated a potential interaction between HSP70 and the SERCA pump, which mediates Ca 2+ re-uptake by the SR 22 . In this study, we used the SERCA pump inhibitor, cyclopiazonic acid (CPA), which does www.nature.com/scientificreports/ not affect phasic contraction 15 . In fact, it has been demonstrated that it increases the intracellular concentration of Ca 2+ without changing the force of contraction 12 . Interestingly, in the presence of CPA, there is a shift in the contractile phenotype of vascular smooth muscle to rely mainly on Ca 2+ influx via voltage-independent Ca 2+ channels 12 . Here, as showed by others, we confirmed that CPA does not impact PE-induced phasic contraction (Fig. 7). More interestingly, we found that, under the conditions examined in this study, it disrupts the ability of the vessel to elicit tonic contraction following re-addition of Ca 2+ (Fig. 7). It is important to consider the importance of Ca 2+ re-uptake by the SR, as the influx of this ion is gated by its depletion from the SR 16 . When we combined the HSP70 inhibitor with CPA, we observed a complete impairment of the vessels' ability to elicit contraction, which could be due to (a) the effects of CPA upon Ca 2+ re-uptake and/or (b) the effects of VER155008 on voltage-independent Ca 2+ channels, which was, under this condition, the main source of Ca 2+ influx. While, to the best of our knowledge, data regarding an interplay between HSP70 and the SERCA pump in vascular structures are nonexistent, the literature shows that this chaperone has a protective role towards this protein 33 . For example, in cardiomyocytes, the deletion of HSP70 associates with a decrease in the expression of SERCA2a 7 whereas, in PC12 cells, overexpression of HSP70 increases the levels of SERCA2a and SERCA2b 31 . Importantly, it has been demonstrated that, in HEK-293 cells, HSP70 prevents thermal inactivation of SERCA2a, potentially by decreasing its oxidation and nitrosylation 39 . Likewise, HSP70 prevents the thermal inactivation of SERCA1a in fast-twitch skeletal muscle 40 . Nevertheless, these studies differ in many aspects from our work, especially the fact that we aimed at investigating this interaction in the absence of a pathological condition.
In summary, we showed that blockade of HSP70 affects Ca 2+ handling mechanisms in aorta stimulated with PE via crosstalk with (a) IP3r-mediated Ca 2+ release from the SR (phasic) and (b) voltage-independent Ca 2+ channels-facilitated Ca 2+ influx (tonic) (Fig. 8). There are, however, many points to enlighten, including the molecular aspects guiding this process. Here, we took an indirect approach to evaluate the role of HSP70 in vascular contraction, and therefore, further studies employing in situ enhancement of HSP70 in vascular smooth muscle are still required, since they could shed much light on the exact mechanisms involved in HSP70-mediated arterial contraction. A striking question arising at this point is whether the interaction between HSP70 and Ca 2+ remains in vascular diseases, such as diabetes and hypertension. Nevertheless, such a development in our understanding shifts the way one might approach disease-associated vascular complications, especially because we provided evidence that, under the conditions evaluated in this study, HSP70 contributes to vascular Ca 2+ dynamics, and Ca 2+ is a key player in healthy and diseased states.

Methods
Ethics statement. All animal procedures followed the Guide for the Care and Use of Laboratory Animals from the National Institutes of Health and were reviewed and approved by the Institutional Animal Care and Use Committees of Augusta University and Florida Institute of Technology. www.nature.com/scientificreports/ Animals. Sprague Dawley male rats were acquired from Charles River Laboratory and Taconic Biosciences.
Animals were housed at room temperature with light exposure cycles of 12 h and free access to food and water. Rats were sacrificed with 10-12 weeks under isoflurane anesthesia (5% in 100% O 2 ). We carefully excised the aorta of each animal, and placed it in ice-cold Krebs' solution (mmol/l: 130 NaCl, 4. • IP3r-vehicle or 2-APB (50 and 100 µmol/l ; IP3r inhibitor) or VER155008 ( 10 µmol/l ) or the combination of 2-APB and VER155008 was added at time zero, which was the moment that the 0[Ca 2+ ] Krebs' solution was placed into the chamber. After 3 min, samples were challenged with PE ( 10 µmol/l ). Then, the time-force curves generated were evaluated for 10 min. • Voltage-dependent and -independent Ca 2+ channels-0[Ca 2+ ] Krebs' solution was placed into the myograph chamber. After 3 min, samples were challenged with PE ( 10 µmol/l ). In this set of experiments, vehicle or VER155008 ( 10 µmol/l ) or verapamil (1 and 3 µmol/l ; a selective LTCC blocker) or the combination of verapamil and VER155008 or 2-APB (50 and 100 µmol/l ; a non-selective inhibitor of voltage-independent Ca 2+ permeable channels) or the combination of 2-APB and VER155008 was added after the addition of PE. We added the inhibitors at this point because (a) VER155008 affects the amplitude of the phasic contraction elicited by PE 4 and (b) 2-APB also affects the response of IP3r 23 . The time-force curves generated were evaluated for 10 min. Then, the extracellular Ca 2+ concentration was restored, and the force generated was evaluated for 15 min. • Rho-kinase and SERCA pump-vehicle or Y27632 (1 and 3 µmol/l ; Rho-kinase inhibitor) or VER155008 ( 10 µmol/l ) or the combination of Y27632 and VER155008 or CPA (5 and 10 µmol/l ; SERCA pump inhibitor) or the combination of CPA and VER155008 was added at time zero, which was the moment that the 0[Ca 2+ ] Krebs' solution was placed into the chamber. Following 3 min, samples were challenged with PE ( 10 µmol/l ). Then, the time-force curves generated were evaluated for 10 min. Subsequently, the extracellular Ca 2+ concentration was restored, and the tissue response was recorded for 15 min.
Data and statistical analysis. The relationship between time (T) and force (F) expressed in the curve plots was calculated in the following way. We considered the moment we added 0[Ca 2+ ] Krebs' solution into the chamber as t 0 . At any point in time after t 0 , F was calculated by subtracting the basal force, which is the observed measurement F at t 0 . Then, we computed the area under the curve (AUC) and the E max of the agonist. Results are expressed as mean ± SEM. Student t-test and one-way ANOVA followed by the Newman-Keuls method was used to determine statistical differences between two and three or more groups, respectively. n represents the number of animals and p < 0.05 was considered statistically significant using the software GraphPad Prism, version 5.0.

Data availability
All animal data generated or analyzed during this study are included in this published article.