1. Cell Biology
  2. Structural Biology and Molecular Biophysics
Download icon

A Gs-coupled purinergic receptor boosts Ca2+ influx and vascular contractility during diabetic hyperglycemia

  1. Maria Paz Prada
  2. Arsalan U Syed
  3. Olivia R Buonarati
  4. Gopireddy R Reddy
  5. Matthew A Nystoriak
  6. Debapriya Ghosh
  7. Sergi Simó
  8. Daisuke Sato
  9. Kent C Sasse
  10. Sean M Ward
  11. Luis F Santana
  12. Yang K Xiang
  13. Johannes W Hell
  14. Madeline Nieves-Cintrón  Is a corresponding author
  15. Manuel F Navedo  Is a corresponding author
  1. University of California, Davis, United States
  2. University of Louisville, United States
  3. Sasse Surgical Associates, United States
  4. University of Nevada, United States
  5. VA Northern California Healthcare System, United States
Research Article
  • Cited 4
  • Views 841
  • Annotations
Cite this article as: eLife 2019;8:e42214 doi: 10.7554/eLife.42214

Abstract

Elevated glucose increases vascular reactivity by promoting L-type CaV1.2 channel (LTCC) activity by protein kinase A (PKA). Yet, how glucose activates PKA is unknown. We hypothesized that a Gs-coupled P2Y receptor is an upstream activator of PKA mediating LTCC potentiation during diabetic hyperglycemia. Experiments in apyrase-treated cells suggested involvement of a P2Y receptor underlying the glucose effects on LTTCs. Using human tissue, expression for P2Y11, the only Gs-coupled P2Y receptor, was detected in nanometer proximity to CaV1.2 and PKA. FRET-based experiments revealed that the selective P2Y11 agonist NF546 and elevated glucose stimulate cAMP production resulting in enhanced PKA-dependent LTCC activity. These changes were blocked by the selective P2Y11 inhibitor NF340. Comparable results were observed in mouse tissue, suggesting that a P2Y11-like receptor is mediating the glucose response in these cells. These findings established a key role for P2Y11 in regulating PKA-dependent LTCC function and vascular reactivity during diabetic hyperglycemia.

https://doi.org/10.7554/eLife.42214.001

Introduction

Diabetes is a major risk factor contributing to cardiovascular complications, including hypertension, increased risk of stroke, coronary disease, and organ failure (Whelton et al., 2018). These complications are often linked to the effects of elevated extracellular glucose (e.g. hyperglycemia - the hallmark metabolic abnormality in diabetes) on vascular function (Brown et al., 2010; Cooper et al., 2001; Tousoulis et al., 2013). In addition to endothelium-dependent alterations (Lagaud et al., 2001; Sena et al., 2013), endothelium-independent modifications, including enhanced contractility of arterial myocytes, are emerging as critical contributing factors to these complications (Aronson, 2008; DeFronzo et al., 2013; Fleischhacker et al., 1999; Montero et al., 2013; Nieves-Cintrón et al., 2018). Consistent with this notion, we and others recently demonstrated augmented vasoconstriction in response to acute and chronic elevations in extracellular glucose that were independent of endothelial function (Fleischhacker et al., 1999; Jackson et al., 2016; Nystoriak et al., 2017). We linked these glucose-mediated effects on vascular reactivity to an increase in L-type CaV1.2 channel (LTCC) activity in human and murine arterial myocytes (Navedo et al., 2010b; Nystoriak et al., 2017). This is critical, as Ca2+ influx via LTCCs is the main Ca2+ entry pathway in these cells and determines muscle contractility and arterial diameter (Knot and Nelson, 1998). Unexpectedly, we found that glucose-mediated potentiation of LTCCs involves phosphorylation of CaV1.2 at serine 1928 (Ser1928) by protein kinase A (PKA) (Morotti et al., 2017; Navedo et al., 2010b; Nystoriak et al., 2017). Yet, how glucose leads to PKA activation is unknown.

Elevations in cellular glucose have been shown to trigger autocrine release of nucleotides (e.g. ATP, UTP) into the extracellular space of many cells, including arterial myocytes (Costa et al., 2009; Hazama et al., 1998; Nilsson et al., 2006; Parodi et al., 2002). These extracellular nucleotides can then act on G protein-coupled P2Y receptors to induce a plethora of cellular responses (von Kügelgen and Harden, 2011). Previous studies demonstrated that elevations in extracellular glucose led to increases in intracellular Ca2+ ([Ca2+]i) in isolated arterial myocytes (Nilsson et al., 2006). Indeed, this glucose-mediated response was associated with enhanced autocrine release of nucleotides engaging P2Y receptors, but the contributing mechanisms as well as the identity of the P2Y receptor involved are unclear.

Eight functionally diverse G protein-coupled P2Y receptors have been identified (von Kügelgen and Harden, 2011). Most of these P2Y receptors are coupled to Gq/i proteins. Conversely, P2Y11 is the only P2Y receptor that is coupled to Gs (Communi et al., 1997; Kennedy, 2017; Schuchardt et al., 2012; von Kügelgen and Harden, 2011). Activation of this receptor could then activate adenylyl cyclase (AC) and PKA to underlie the glucose-induced effects on LTCC activity, but its role in arterial myocytes is unclear. P2Y11 receptors were first cloned from human tissue and subsequently identified in other species (Kennedy, 2017; von Kügelgen and Harden, 2011). Although the P2Y11 gene has not been found at the expected position in the mouse genome, recent rodent annotations related to this gene (e.g. XM_008766009.2 and XM_0130655917.2), as well as studies defining a functional role based on pharmacological data obtained from rat and mouse tissue hinted at the presence of at least a P2Y11-like receptor in mice (Kennedy, 2017). We, therefore, undertook a comprehensive approach to investigate whether upstream engagement of this Gs-coupled P2Y receptor is implicated in PKA activation leading to increased Ser1928 phosphorylation of vascular CaV1.2 that results in potentiation of LTCC activity during diabetic hyperglycemia. Accordingly, Western blot analysis, immunofluorescence confocal microscopy, Ground State Depletion (GSD) super-resolution nanoscopy, FRET-based cAMP imaging and voltage-clamp electrophysiology confirmed functional expression of P2Y11 in human arterial myocytes. We found that elevated glucose stimulates cAMP synthesis with the same magnitude as that observed upon application of the selective P2Y11 agonist NF546 or when both stimuli were given simultaneously, thus suggesting that glucose and NF546 signaling proceed through the same pathway. Nanometer proximity was observed between P2Y11 and PKA and P2Y11 and CaV1.2 at the sarcolemma of arterial myocytes, which facilitates a structural arrangement that may be essential for glucose effects on LTCC activity. Indeed, the NF546/glucose-induced cAMP signal resulted in enhanced Ser1928 phosphorylation and LTCC activity in human arterial myocytes, and these changes were prevented by the selective P2Y11 inhibitor NF340. Intriguingly, comparable functional data were observed in mouse arterial myocytes, suggesting that a P2Y11-like receptor could be mediating the glucose response in these cells that results in vasoconstriction (Kennedy, 2017). Thus, our data in human and murine tissue indicate a key role for a Gs-coupled P2Y receptor, which fits the distribution, pharmacological, and signaling profile of a P2Y11 or P2Y11-like receptor, respectively, as an upstream player in the regulation of LTCCs and vascular reactivity during diabetic hyperglycemia. These findings may have important clinical and therapeutic implications as they may point to P2Y11 as a potential target for treating diabetic vascular complications.

Results

Glucose-induced extracellular nucleotide release mediates Ser1928 phosphorylation, increased LTCC activity and vasoconstriction

In our first series of experiments, we pressurized (60 mmHg) mouse cerebral arteries to develop stable and spontaneous tone (Supplementary file 1 and 2). When increasing extracellular glucose from 10 mM to 20 mM D-glucose, there was a significant constriction (Figure 1A and Supplementary file 1 and 2), which is consistent with previous reports by our group and others (Jackson et al., 2016; Nystoriak et al., 2017). These extracellular D-glucose concentrations are used because they are within the range of observed nonfasting glucose levels in nondiabetic and diabetic mouse models (Lagaud et al., 2001; Moien-Afshari et al., 2008; Navedo et al., 2010b; Nieves-Cintrón et al., 2015; Nystoriak et al., 2014; Nystoriak et al., 2017). Considering that elevating glucose in the external milieu may trigger autocrine release of nucleotides (Costa et al., 2009; Hazama et al., 1998; Nilsson et al., 2006; Parodi et al., 2002), we investigated whether hydrolysis of extracellular nucleotides would ablate glucose-mediated vasoconstriction. For these experiments, pressurized arteries (60 mmHg) were pre-treated with the ectonucleotidase apyrase (0.32 U/mL) in a 10 mM D-glucose solution for 10 min. Apyrase-treated arteries failed to constrict to 20 mM D-glucose (Figure 1A and Supplementary file 2). Vasoconstriction in response to 60 mM potassium (K+) in which the membrane potential closely follows the equilibrium potential for K+ (~−20 mV) (Knot and Nelson, 1998), was similar in control and apyrase-treated arteries (Figure 1—figure supplement 1A and Supplementary file 1 and 2). These results suggest that differences in glucose-induced vasoconstriction were not due to an inability of apyrase-treated arteries to respond to 20 mM D-glucose. Furthermore, the glucose-induced constriction was not attributed to changes in osmolarity as treatment with equimolar concentrations of the nonpermeable mannitol (20 mM = 10 mM D-glucose +10 mM mannitol) had no effect on vascular reactivity (Figure 1—figure supplement 1B and Supplementary file 1 and 2).

Figure 1 with 2 supplements see all
Extracellular nucleotides promote vasoconstriction, CaV1.2 activity and Ser1928 phosphorylation in response to 20 mM D-glucose in murine arterial myocytes.

(A) Representative diameter recordings and summary arterial tone data from pressurized (60 mmHg) wt mouse cerebral arteries before and after application of 20 mM D-glucose in the absence (n = 6 arteries from six mice) and presence (n = 6 arteries from six mice) of apyrase (apy; 0.32 U/ml; *p<0.05, Wilcoxon matched pairs test; Figure 1—source data 1). (B) Characteristic IBa recordings from the same cell and summary IBa data from wt mouse cerebral arterial myocytes evoked by step depolarizations from −70 to +10 mV before and after application of 20 mM D-glucose in the absence (n = 11 cells from five mice) and presence of apyrase (n = 9 cells from five mice) (*p<0.05, paired t test; Figure 1—source data 2). (C) Representative immunoblot detection of phosphorylated Ser1928 (pSer1928) and total CaV1.2 from wt mouse cerebral and mesenteric arteries after 10 min incubation with 10 mM or 20 mM D-glucose in the absence and presence of apyrase (n = 10 arterial lysates per condition), and quantification of pSer1928 (AU = arbitrary units) (*p<0.05, Kruskal-Wallis with Dunn’s multiple comparisons; Figure 1—source data 3).

https://doi.org/10.7554/eLife.42214.002

LTCC activity is critical for vascular reactivity (Knot and Nelson, 1998), and its function is enhanced in response to acute increases in glucose and in diabetes due to increased Ser1928 phosphorylation in the CaV1.2 carboxy terminal (Morotti et al., 2017; Navedo et al., 2010b; Nystoriak et al., 2017). We, thus, examined whether nucleotide degradation with apyrase prevents glucose-mediated increases in LTCC activity and Ser1928 phosphorylation. To assess this, we used patch-clamp electrophysiology in the whole-cell perforated configuration with barium (Ba2+) as the charge carrier, before and after application of nifedipine to determine the nifedipine-sensitive Ba2+ current (IBa) associated with LTCC activity in control and apyrase-treated cerebral arterial myocytes. We found that 20 mM D-glucose significantly enhanced IBa in wild type (wt) control arterial myocytes, but not in apyrase-treated cells (Figure 1B). The glucose effect on LTCC activity was not due to changes in osmolarity as 20 mM mannitol had no impact on IBa (Figure 1—figure supplement 1C).

We next performed Western blot analysis with a specific phospho-antibody against Ser1928 at CaV1.2 and a FP1 antibody for total CaV1.2 detection that have been extensively validated by our group (Buonarati et al., 2017; Davare et al., 1999; Hall et al., 2013; Nystoriak et al., 2017; Patriarchi et al., 2016). Indeed, we have previously shown that the immunoreactivity at 250 kDa of CaV1.2 with the phospho-specific antibody for Ser1928 (but not a phospho-specific antibody for Ser1700) and FP1 are completely eliminated in tissue from S1928A knockin mice (Patriarchi et al., 2016) and conditional CaV1.2 knockout mice (Buonarati et al., 2017), respectively. Data revealed increased phosphorylation of this residue in response to 20 mM D-glucose in control arterial lysates (Figure 1C and Figure 1—figure supplement 1D). However, this response was completely absent in arterial lysates exposed to 20 mM mannitol, thus ruling out any osmolarity effects (Figure 1—figure supplement 1E). Moreover, glucose-induced Ser1928 phosphorylation was hindered in lysates from apyrase-treated arteries (Figure 1C and Figure 1—figure supplement 1D). These results suggest that elevated glucose increases Ser1928 phosphorylation to potentiate LTCC activity through an autocrine mechanism involving secreted nucleotides.

To further test this possibility, LTCC activity was recorded in response to elevated glucose during continuous perfusion or static bath conditions (as above). The rationale for these experiments was to determine whether the wash out of nucleotides with flow can impact the glucose-induced potentiation of LTCCs. We found statistically similar IBa in cells exposed to 10 mM and 20 mM D-glucose under continuous bath perfusion (Figure 1—figure supplement 2). Conversely, IBa was significantly elevated in arterial myocytes exposed to 20 mM D-glucose when the bath perfusion was stopped (e.g. static bath conditions). These results suggest that secreted nucleotides could diffuse away under continuous flow, whereas under static bath conditions, they could accumulate in or around the area of release at the surface membrane of arterial myocytes to induce the optimal activation of the signaling pathway underlying increased LTCC activity in response to elevations in extracellular glucose.

P2Y11 receptors are in close proximity to CaV1.2 and PKAcat in arterial myocytes

Extracellular nucleotides are well-known activators of G-coupled P2Y receptors (von Kügelgen and Harden, 2011). P2Y11 is a Gs-coupled P2Y receptor that could activate PKA (Communi et al., 1997; Kennedy, 2017; Schuchardt et al., 2012; von Kügelgen and Harden, 2011), and this could lead to enhanced Ser1928 phosphorylation, LTCC activity, and vasoconstriction during diabetic hyperglycemia. Using Western blot analysis, we first validated the P2Y11 antibody in tsA-201 cells, which endogenously express P2Y11 (Dreisig and Kornum, 2016), by over-expressing human GFP-tagged P2Y11 and by knocking down the endogenous and over-expressed receptor with specific P2Y11 antisense oligodeoxynucleotides (ODNs) (Figure 2A and B). We subsequently detected an immunoreactive band of the expected molecular weight for P2Y11 (~40 kDa) in lysates from freshly dissected human adipose arteries (Figure 2C and Figure 2—figure supplement 1A). This band seems specific for P2Y11 since treatment of human arteries with specific human P2Y11 antisense ODNs knocked down ~50% of the basal P2Y11 protein abundance (Figure 2—figure supplement 1B). Confocal imaging and line profile analysis of P2Y11-associated fluorescence and the plasma membrane marker wheat germ agglutinin (WGA) suggest a prominent distribution of the receptor on the surface membrane of freshly dissociated human arterial myocytes (Figure 2D and E). This was not observed when the antibody for P2Y11 was omitted from the preparation (Figure 2—figure supplement 1C). These results suggest expression of P2Y11 in human arterial myocytes. We observed comparable results in mouse arterial lysates and arterial myocytes, including detection and knock down with P2Y11 ODNs of an immunoreactive band of the expected molecular weight for P2Y11 (Figure 2C and Figure 2—figure supplement 1A, B and D). A band of ~110–120 kDa was observed in mouse arterial lysates, perhaps reflecting the formation of heterodimers between the P2Y11 receptor itself or other purinergic/G protein-coupled receptors, as previously reported (Barragán-Iglesias et al., 2015; Barragán-Iglesias et al., 2014; Ecke et al., 2008; Nishimura et al., 2016). We also found expected distribution of P2Y11-like-associated fluorescence at the surface membrane of mouse arterial myocytes (Figure 2D and E), which was not observed when the primary antibody was omitted or boiled (Figure 2—figure supplement 1C and E).

Figure 2 with 1 supplement see all
P2Y11 protein and distribution in arterial myocytes.

(A) Representative blot of immunoreactive bands of expected molecular weight for endogenous P2Y11 (~40 kDa), overexpressed P2Y11-GFP (~70 kDa), and β-actin (~43 kDa) in untransfected, vehicle-treated (empty transfection) and P2Y11-GFP transfected tsA-201 cells (n = 3 lysates per condition). Note that tsA-201 cells endogenously express P2Y11 (Dreisig and Kornum, 2016). (B) Representative blot of immunoreactive bands of expected molecular weight for endogenous P2Y11 (~40 kDa), overexpressed P2Y11-GFP (~70 kDa), and β-actin (~43 kDa) in tsA-cells transfected with P2Y11-GFP as well as corresponding P2Y11 sense (SNS) or antisense (ANS) ODNs (64% reduction in endogenous P2Y11 expression in cells treated with ANS; 62% reduction in P2Y11-GFP expression in P2Y11-GFP-transfected cells treated with ANS; n = 3 lysates per condition; Figure 2—source data 1). (C) Representative immunoblot detection of P2Y11 (~40 kDa) in lysates from human and wt mouse arteries (n = 3 arterial lysates per sample). (D) Representative confocal images of P2Y11-associated fluorescence (green), wheat germ agglutinin (WGA, red) and merged channels in human (n = 11 cells from three humans) and wt mouse (n = 14 cells from three mice) arterial myocytes. (E) Line profile of the P2Y11- and WGA-associated fluorescence from the area highlighted by the dotted lines in the representative human and mouse arterial myocytes in D.

https://doi.org/10.7554/eLife.42214.013

If P2Y11 is involved in PKA-mediated activation of LTCCs in response to elevated glucose, we hypothesized that a subpopulation of these receptors should be localized in close proximity to CaV1.2 and PKAcat. To test this possibility, we examined the spatial relationship of P2Y11 with CaV1.2 and PKAcat in freshly dissociated human arterial myocytes using GSD super-resolution nanoscopy in the Total Internal Reflection Fluorescence (TIRF) configuration. This approach achieves a lateral resolution of ~20 nm with selective illumination of the (sub)sarcolemmal region. Note that the FP1 antibody against CaV1.2 has been extensively validated by our group (Buonarati et al., 2017; Hall et al., 2013; Nystoriak et al., 2017; Patriarchi et al., 2016). The PKAcat antibody was validated in arterial myocytes by pre-absorption with a blocking peptide (Figure 3—figure supplement 1). The GSD super-resolution localization maps (Figure 3A and B, bottom panels) for P2Y11, CaV1.2, and PKAcat obtained from conventional TIRF images (Figure 3A and B, top panels) showed that these proteins form clusters of various sizes (Figure 3C) and density (Figure 3D) at the sarcolemma of human arterial myocytes. Whereas clusters were broadly expressed throughout the sarcolemma, merged spatial maps of P2Y11 with CaV1.2 or PKAcat suggested a close association between a subset of these proteins (Figure 3E and F). Indeed, histograms of the P2Y11-to-nearest CaV1.2 or PKAcat distances revealed that the closest centroids for P2Y11-CaV1.2 and P2Y11- PKAcat were at 59 nm and 58 nm, respectively (Figure 3G and H). Similar results were observed using freshly dissociated mouse cerebral arterial myocytes (Figure 3—figure supplement 2). Clusters of P2Y11, CaV1.2, or PKAcat were never observed when primary antibodies were omitted from the human or mouse preparation (Figure 3—figure supplement 3A and B). To determine whether the interaction between a subpopulation of P2Y11 with a pool of CaV1.2 and PKA in arterial myocytes is the result of specific organization between two proteins, we compared the percentage of P2Y11 and CaV1.2/PKAcat overlap between experimental super-resolution localization maps and images containing randomized distribution of P2Y11, CaV1.2 and PKAcat. Randomized images for P2Y11-CaV1.2 and P2Y11-PKAcat were generated based on data derived from the experimental super-resolution localization maps for these pairs of proteins using the Coste’s randomization algorithm included in the JACoP plug-in in ImageJ (Bolte and Cordelières, 2006). Our analysis found that the percentage of overlap between P2Y11-CaV1.2 and P2Y11-PKAcat obtained from the experimental localization maps was significantly higher than that observed for a simulated random distribution between P2Y11-CaV1.2 and P2Y11-PKAcat (Figure 3I and Figure 3—figure supplement 3C). These results suggest an intimate association between subpopulations of P2Y11-CaV1.2 and P2Y11-PKAcat.

Figure 3 with 3 supplements see all
Nanometer organization of P2Y11 with CaV1.2 and PKAcat in human arterial myocytes.

Representative conventional TIRF images (top) and corresponding GSD reconstruction maps (bottom) from human arterial myocytes labeled for (A) P2Y11 and CaV1.2 and (B) P2Y11 and PKAcat. Lower panels display enhanced magnifications of areas shown in yellow boxes (scale bar, 400 nm). (C) Histograms of the area of P2Y11, CaV1.2 and PKAcat clusters in isolated human arterial myocytes (1621 ± 29, 1209 ± 16 and 1322 ± 20 nm2, respectively; Figure 3—source data 1). (D) Bar plot of cluster density for P2Y11, CaV1.2 and PKAcat (38 ± 2, 29 ± 4, and 25 ± 3 clusters/µm2, respectively; Figure 3—source data 2). Enlarged merged image (left) and associated x-y fluorescence intensity profile (right) from area highlighted by the dotted lines of sites of close proximity between (E) P2Y11 (red) and CaV1.2 (green) and (F) P2Y11 (red) and PKAcat (green) (scale bar, 200 nm). Histograms of the lowest intermolecular distance to P2Y11 centroids for (G) CaV1.2 (n = 19,611 particles from 6 cells; Figure 3—source data 3) and (H) PKAcat (n = 22,425 particles from 6 cells; Figure 3—source data 4) fluorescence particles. Data were fit with a sum of two Gaussian functions with depicted R2 and centroids. (I) Bar plot of % overlap of P2Y11 with CaV1.2 or PKAcat for experimental (CaV1.2: n = 36 segments from 12 cells; PKAcat: n = 22 segments from 11 cells) and randomized simulations images (CaV1.2: n = 6 segments from 6 cells; PKAcat: n = 6 segments from 6 cells) (*p<0.05, unpaired t test with Welch’s correction; Figure 3—source data 5).

https://doi.org/10.7554/eLife.42214.017

We used the proximity ligation assay (PLA) as an additional analytical tool to examine the association between P2Y11 and CaV1.2 or PKAcat. PLA fluorescence puncta are only observed if proteins of interest are 40 nm or less apart (Fredriksson et al., 2002). We have extensively validated this approach in previous studies (Nieves-Cintrón et al., 2017; Nystoriak et al., 2017). Whereas PLA signals were absent when at least one primary antibody was omitted (Figure 4—figure supplement 1), robust fluorescent puncta were detected in human (Figure 4A and B) and mouse (Figure 4C and D) arterial myocytes co-labeled for P2Y11 and CaV1.2 or P2Y11 and PKAcat. Altogether, these results suggest that human P2Y11 or mouse P2Y11-like receptors are located on the surface membrane of arterial myocytes within nanometer proximity (~≤40 nm) of a subpopulation of CaV1.2 and PKAcat.

Figure 4 with 1 supplement see all
P2Y11 associates with CaV1.2 and PKAcat in human and murine arterial myocytes.

(A) Exemplary fluorescence PLA (red)/DAPI (blue) and differential interference contrast (right) images of human arterial myocytes labeled for P2Y11 + CaV1.2 and P2Y11 + PKAcat. (B) Quantification of PLA fluorescent puncta per cell area (puncta/µm2) for human arterial myocytes labeled for P2Y11 (n = 26 cells from three human samples), CaV1.2 (n = 20 cells from three humans), PKAcat (n = 17 cells from three humans), P2Y11 + CaV1.2 (n = 23 cells from three humans), and P2Y11 + PKAcat (n = 20 cells from three humans) (*p<0.05, Kruskal-Wallis with Dunn’s multiple comparisons; Figure 4—source data 1). (C) Representative fluorescence PLA (red)/DAPI (blue) and differential interference contrast (right) images of mouse arterial myocytes labeled for P2Y11 + CaV1.2 and P2Y11 + PKAcat. (D) Quantification of PLA fluorescent puncta per µm2 cell area for mouse arterial myocytes labeled for P2Y11 (n = 44 cells from six mice), CaV1.2 (n = 15 cells from six mice), PKAcat (n = 19 cells from six mice), P2Y11 + CaV1.2 (n = 25 cells from six mice), and P2Y11 + PKAcat (n = 29 cells from six mice) (*p<0.05, Kruskal-Wallis with Dunn’s multiple comparisons; Figure 4—source data 2).

https://doi.org/10.7554/eLife.42214.030

Increased sarcolemmal cAMP synthesis in arterial myocytes in response to a P2Y11 agonist recapitulates glucose effects

P2Y11 can stimulate adenylyl cyclase (AC) activity to produce cAMP (Communi et al., 1997; Kennedy, 2017; Schuchardt et al., 2012; von Kügelgen and Harden, 2011). Thus, we first used a membrane-targeted Epac1-camps-based FRET sensor (ICUE3-PM) (Allen et al., 2006; Liu et al., 2011) expressed in tsA-201 cells to validate the P2Y11 pharmacology (Figure 5—figure supplement 1A) (Dreisig and Kornum, 2016). As expected, the highly selective P2Y11 agonist NF546 (500 nM) (Meis et al., 2010) increased cAMP synthesis in these cells. This NF546-induced cAMP response remained intact in cells pretreated with either selective P2Y1 (10 µM MRS2179) or P2Y6 (100 nM MRS2578) inhibitors but was prevented in cells exposed to the selective P2Y11 inhibitor NF340 (10 µM; 500-fold more selective for P2Y11 than other P2Y receptors) (Meis et al., 2010).

We recapitulated the NF546-induced cAMP response and its ablation by pretreatment with NF340 in primary, unpassaged human arterial myocytes expressing the ICUE3-PM sensor (Figure 5A and B). Increasing D-glucose concentration from 5 mM to 15 mM, which are glucose concentrations comparable to those observed in nondiabetic and diabetic patients respectively, induced a small yet significant increase in cAMP synthesis in human cells (Figure 5A and B). The glucose effects are not due to osmotic changes as equimolar concentrations of the non-permeable mannitol did not induce any cAMP synthesis (Figure 5—figure supplement 1B). As expected, the synthesis of cAMP was further amplified by the broad AC activator forskolin (1 μM) (Figure 5A) even in the presence of 20 mM mannitol (Figure 5—figure supplement 1B). Simultaneous application of 15 mM D-glucose +NF546 resulted in cAMP synthesis of a similar magnitude to that induced by independent treatments (Figure 5A and B). This suggests that the effects of elevated glucose and the NF546 compound may be acting through a similar pathway that requires activation of P2Y11. Consistent with this notion, application of NF546 by itself or 15 mM D-glucose +NF546 failed to induce cAMP synthesis in cells pretreated with NF340 (Figure 5A and B). Yet forskolin, which bypasses P2Y11, was still able to stimulate global cAMP production. Similar results were observed in murine arterial myocytes expressing the ICUE3-PM sensor (Figure 5C and D and Figure 5—figure supplement 1C). Intriguingly, cAMP synthesis in response to elevated glucose and the P2Y11 agonist NF546 was significantly larger in human myocytes compared to mouse cells (glucose: 0.077 ± 0.003 in human vs 0.045 ± 0.002 in mouse; NF546: 0.083 ± 0.005 in human vs 0.045 ± 0.004 in mouse; p<0.05). The reasons for this are currently unclear but may include intrinsic differences associated with species (human vs mouse), blood vessels (human adipose arteries vs mouse aortae) and expression patterns of the biosensor in human vs mouse cells (Reddy et al., 2018), as well as variations in protein expression and receptor/enzyme activity (e.g. P2Y11, AC and PDE expression/activity) in human vs mouse arterial myocytes. Whether either or all of these factors contribute to distinctive cAMP responses in human vs mouse arterial myocytes as well as their physiological implications remain to be determined. Nevertheless, our findings suggest a role for human P2Y11 and mouse P2Y11-like receptors in glucose-induced cAMP synthesis.

Figure 5 with 1 supplement see all
Glucose and the P2Y11 agonist NF546 increase sarcolemmal cAMP synthesis in arterial myocytes, and these effects are prevented by the P2Y11 antagonist NF340.

(A) Time course of YFP/CFP (donor/acceptor) FRET ratios (normalized to resting levels before treatment) in human arterial myocytes expressing the ICUE3-PM biosensor in response to 15 mM D-glucose (black; n = 19 cells from two humans), 500 nM NF546 (green; n = 18 cells from two humans) and 15 mM D-glucose +500 nM NF546 (red; n = 14 cells from two humans) before and after application of broad adenylyl cyclase agonist forskolin (1 µM). In a set of experiments, cells were first pre-treated with the P2Y11 antagonist NF340 (10 µM) for at least 15–20 min before treatment with 500 nM NF546 (orange; n = 15 cells from two humans) and 15 mM D-glucose +500 nM NF546 (blue; n = 13 cells from two humans). Horizontal bars indicate treatment. Increases in cAMP production are represented by decreases in YFP/CFP ratio due to binding of cAMP to the biosensor. (B) Bar plot of maximum FRET responses (YFP/CFP) for human arterial myocytes in response to the indicated treatment (*p<0.05, Kruskal-Wallis with Dunn’s multiple comparisons; Figure 5—source data 1. Statistical differences were compared between 15 mM D-glucose vs NF546, 15 mM D-glucose vs 15 mM D-glucose +NF546, 15 mM D-glucose vs NF546 +NF340, 15 mM D-glucose vs 15 mM D-glucose +NF546+NF340, NF546 vs NF546 +NF340, NF546 vs 15 mM D-glucose +NF546, NF546 vs 15 mM D-glucose +NF546+NF340, 15 mM D-glucose +NF546 vs 15 mM D-glucose +NF546+NF340). (C) Time course of YFP/CFP (donor/acceptor) FRET ratios (normalized to resting levels before treatment) in mouse arterial myocytes expressing the ICUE3-PM biosensor in response to 20 mM D-glucose (black; n = 83 cells from three mice), 500 nM NF546 (green; n = 13 cells from three mice) and 20 mM D-glucose +500 nM NF546 (red; n = 13 cells from three mice). Horizontal bars indicate treatment. In a set of experiments, cells were first pre-treated with the P2Y11 antagonist NF340 (10 µM) for at least 15–20 min before treatment with 500 nM NF546 (orange; n = 17 cells from three mice) and 20 mM D-glucose +500 nM NF546 (blue; n = 18 cells from three mice). Increases in cAMP production are represented by decreases in YFP/CFP ratio due to binding of cAMP to the biosensor. (D) Bar plot of maximum FRET responses (YFP/CFP) for mouse arterial myocytes in response to the indicated treatment (*p<0.05, Kruskal-Wallis with Dunn’s multiple comparisons; Figure 5—source data 2. Statistical differences were compared between all datasets).

https://doi.org/10.7554/eLife.42214.034

The P2Y11 inhibitor NF340 prevents glucose-induced Ser1928 phosphorylation, LTCC activity and vasoconstriction

In our next series of experiments, we confirmed the glucose-induced increase in IBa detected in mouse arterial myocytes (Figure 1B) using freshly dissociated human arterial myocytes in which external glucose was elevated from 5 mM to 15 mM (Figure 6A and B). These results indicate that elevations in extracellular glucose also stimulate LTCC activity in human arterial myocytes (Nystoriak et al., 2017). The increase in IBa evoked by 15 mM D-glucose was prevented in human cells pretreated with NF340 (Figure 6A and B). Glucose-induced potentiation of IBa was also prevented in mouse arterial myocytes pretreated with NF340 (Figure 6C and D). However, inhibition of P2Y6 receptors with MRS2578, which has been implicated in myogenic tone regulation and glucose-induced NFAT activation in arterial myocytes (Brayden et al., 2013; Nilsson et al., 2006), failed to stop the increase in IBa in response to 20 mM D-glucose (Figure 6C and D). Note that basal IBa were similar in cells exposed to 10 mM D-glucose and 10 mM D-glucose +MRS2578, suggesting that inhibition of P2Y6 did not alter basal LTCC activity (Figure 6C and D). Altogether, these results suggest that P2Y6 receptors are not involved in glucose-induced potentiation of LTCC activity, and that these glucose effects are prevented with a P2Y11 inhibitor.

Figure 6 with 1 supplement see all
The P2Y11 inhibitor NF340 prevents glucose-induced elevations in Ser1928 phosphorylation, LTCC activity, and vasoconstriction in human and murine arterial myocytes.

(A) Representative IBa recordings from the same cell and (B) summary current density data obtained from freshly dissociated human arterial myocytes before and after increasing extracellular D-glucose from 5 mM to 15 mM in the absence (n = 12 cells from six humans) and presence (n = 10 cells from five humans) of NF340 (10 µM) (*p<0.05, paired t test; Figure 6—source data 1). (C) Exemplary IBa traces from the same cell and (D) amalgamated current density data recorded from mouse cerebral arterial myocytes treated with 10 µM NF340 (n = 8 cells from three mice) or 100 nM MRS2578 (n = 7 cells from five mice) before and after increasing extracellular D-glucose from 10 mM to 20 mM (*p<0.05, Kruskal-Wallis with Dunn’s multiple comparisons; Figure 6—source data 2). Representative immunoreactive bands for phosphorylated Ser1928 (pSer1928) and total CaV1.2 and densitometry for pSer1928/CaV1.2 ratio in lysates from (E) human arteries exposed to 5 mM and 15 mM D-glucose in the absence and presence of 10 µM NF340 (n = 6 arterial lysates per condition; *p<0.05, Kruskal-Wallis with Dunn’s multiple comparisons; Figure 6—source data 3) and (F) mouse cerebral and mesenteric arteries exposed to 10 mM or 20 mM D-glucose in the presence of 10 µM NF340 or 100 nM MRS2578 (n = 4 arterial lysates per condition; *p<0.05 Kruskal-Wallis with Dunn’s multiple comparisons; Figure 6—source data 4). Representative diameter recordings and summary data from pressurized (60 mmHg) mouse cerebral arteries exposed to (G–H) 10 µM NF340 (n = 4 arteries from four mice; Friedman with Dunn’s multiple comparisons; Figure 6—source data 5) or (I–J) 100 nM MRS2578 (n = 7 arteries from seven mice; *p<0.05, Wilcoxon matched pair test; Figure 6—source data 6) before and after application of 20 mM D-glucose.

https://doi.org/10.7554/eLife.42214.041

Since the P2Y11 inhibitor hindered the glucose effects on LTCC activity, we also explored whether the NF340 compound interfered with the Ser1928 phosphorylation state of the channel and vascular reactivity in response to hyperglycemia. As observed in mouse arteries (Figure 1C), we found that elevating extracellular glucose significantly increased Ser1928 phosphorylation in human arterial lysates (Figure 6E; Figure 6—figure supplement 1A). Yet, this glucose-mediated increase in Ser1928 phosphorylation was prevented if arterial lysates were pretreated with NF340, whereas phosphorylation still occurred if arterial lysates were pretreated with the P2Y6 inhibitor MRS2578 (Figure 6E and F, and Figure 6—figure supplement 1A and B). Furthermore, vasoconstriction in response to elevated glucose was blocked in pressurized (60 mmHg) mouse cerebral arteries exposed to NF340 (Figure 6G and H and Supplementary file 2) but not to MRS2578 (Figure 6I and J and Supplementary file 2). Note that while arterial tone was similar in control and NF340-treated arteries, it was reduced in arteries exposed to MRS2578 (Figure 6I and J), which is in line with recent data showing that P2Y6 is a sensor of pressure-induced constriction (Brayden et al., 2013). Vasoconstriction in response to 60 mM K+ was similar in control and NF340 as well as MRS2578-treated arteries (Figure 6—figure supplement 1C and Supplementary file 1 and 2). We also evaluated a potential role for P2Y1 receptors in stimulating PKA-mediated vasoconstriction as these receptors have been suggested to crosstalk with P2Y11 to regulate their function in HEK cells and neurons (del Puerto et al., 2012; King and Townsend-Nicholson, 2008). Data revealed that the glucose-induced vasoconstriction was not altered by adding the selective P2Y1 inhibitor MRS2179 (10 μM) either before or after 20 mM D-glucose stimulation (Figure 6—figure supplement 1D and E and Supplementary file 1 and 2). These results suggest that the P2Y1 receptor is not involved in glucose-mediated vasoconstriction. Together, these data demonstrate the effectiveness of a P2Y11 inhibitor in blocking the remodeling of Ser1928 phosphorylation and LTCC activity that mediates vasoconstriction in response to elevated glucose.

The P2Y11 agonist NF546 stimulates Ser1928 phosphorylation, PKA-dependent LTCC activity and vasoconstriction

We next investigated whether a selective P2Y11 agonist could recapitulate the glucose-mediated Ser1928 phosphorylation and PKA-dependent LTCC activity. Using freshly dissociated human arterial myocytes, we found that IBa was significantly increased by application of NF546 (Figure 7A). This P2Y11 agonist also increased IBa at multiple membrane potentials in mouse arterial myocytes with no change in the current-voltage (I-V) relationship (Vmax = 14.9 ± 2.0 mV for 10 mM D-glucose and Vmax = 10.9 ± 1.1 mV for NF546; p=0.1085, extra sum-of-squares F test) (Figure 7B). The NF546-mediated increase in IBa was similar in magnitude to that induced by elevated glucose (Figure 7—figure supplement 1A). Similar IBa potentiation was also observed after application of the non-hydrolyzed ATP analog and potent P2Y11 agonist ATPγS (1 μM; Figure 7—figure supplement 1B) (von Kügelgen and Harden, 2011). Note that ATPγS stimulates P2Y11 activity with EC50 that ranges from 31 nM to 23 μM, depending on readout, cell type and species used, but it is still a more potent activator than ATP itself (Communi et al., 1999; Haas et al., 2014; Kennedy, 2017; Meis et al., 2010; Qi et al., 2001; Schuchardt et al., 2012; von Kügelgen and Harden, 2011). We further correlated the potentiation in IBa by NF546 to an elevation of Ser1928 phosphorylation in freshly dissected human and mouse arterial lysates (Figure 7C and Figure 7—figure supplement 1C). Since PKA is essential for glucose-mediated remodeling of IBa in arterial myocytes and can be activated downstream of P2Y11 (Navedo et al., 2010b; Nystoriak et al., 2017), we also examined its involvement in NF546-mediated potentiation of LTCC activity in arterial myocytes. Consistent with a key role for PKA, NF546 failed to upregulate IBa in arterial myocytes incubated with the PKA inhibitors PKI (100 nM) or rpcAMP (10 μM) (Figure 7D).

Figure 7 with 1 supplement see all
The P2Y11 agonist NF546 increases Ser1928 phosphorylation, LTCC activity, and induces vasoconstriction.

(A) Representative IBa recordings from the same cell (top) and summary IBa data (bottom) from freshly dissociated human arterial myocytes in response to step depolarizations from −70 to +10 mV before and after application of 500 nM NF546 (n = 11 cells from five humans; *p<0.05, paired t test; Figure 7—source data 1). (B) Representative IBa recordings from the same cell (left) triggered by step depolarization from −70 mV to voltages ranging from −60 to +60 mV before and after application of 500 nM NF546 in mouse cerebral arterial myocytes and corresponding IBa-voltage relationship (right) (n = 8 cells from five mice) (*p<0.05, paired t test; Figure 7—source data 2). (C) Exemplary immunoblot detection of phosphorylated Ser1928 (pSer1928) and total CaV1.2 from human (left) and mouse cerebral and mesenteric arteries (right) incubated with 500 nM NF546 and respective densitometry quantification of pSer1928/CaV1.2 ratio (n = 6 arterial lysates per condition for humans; n = 6 arterial lysates per condition for mice) (*p<0.05, Wilcoxon matched pairs test; Figure 7—source data 3). (D) Representative IBa recordings from the same cell (top) and summary IBa data (bottom) from mouse arterial myocytes evoked by step depolarizations from −70 to +10 mV before and after application of 500 nM NF546 in the absence (n = 9 cells, four mice) and presence of 100 nM PKI (n = 9 cells, five mice) or 10 µM rpcAMP (n = 9 cells, four mice) (*p<0.05, paired t test; Figure 7—source data 4). (E) Representative diameter recording and summary arterial tone data from pressurized (60 mmHg) mouse cerebral arteries exposed to 500 nM NF546 (n = 6 arteries, six mice) (*p<0.05, Wilcoxon matched pairs test; Figure 7—source data 5). (F) Representative diameter recordings and summary arterial tone data from pressurized (60 mmHg) wt mouse cerebral arteries exposed to 20 mM D-glucose before and after application of 500 nM NF546 (n = 6 arteries, six mice, left; *p<0.05, Friedman with Dunn’s multiple comparisons; Figure 7—source data 6) and S9128A mouse cerebral arteries after NF546 application (n = 4 from three mice, right; Figure 7—source data 6).

https://doi.org/10.7554/eLife.42214.052

Having established that application of the P2Y11 agonist NF546 influences Ser1928 phosphorylation state as well as LTCC activity, we investigated whether it could also modulate arterial tone. Accordingly, pressurized (60 mmHg) arteries that developed stable arterial tone and responded to 60 mM K+ (Figure 7—figure supplement 1D and Supplementary file 1 and 2) showed significant constriction upon NF546 application (Figure 7E). Interestingly, NF546 did not induce further constriction in wt arteries previously exposed to 20 mM D-glucose (Figure 7F). Indeed, arterial tone was similar in arteries treated with 20 mM D-glucose or 20 mM D-glucose +NF546 (Supplementary file 1 and 2). Confirming a critical role for Ser1928, arteries from a S1928A knockin mouse in which phosphorylation of this CaV1.2 amino acid is prevented (Lemke et al., 2008; Nystoriak et al., 2017; Qian et al., 2017), failed to constrict to NF546 (Figure 7F and Supplementary file 1 and 2). Response of S1928A arteries to 60 mM K+ was similar to those observed for wt arteries (Figure 7—figure supplement 1D and Supplementary file 1 and 2). These results suggest that a P2Y11 agonist can induce Ser1928 phosphorylation, PKA-dependent LTCC potentiation and vasoconstriction, thus recapitulating glucose effects (Navedo et al., 2010b; Nystoriak et al., 2017), and providing further evidence of the involvement of human P2Y11 and mouse P2Y11-like receptors in these alterations.

Increased Ser1928 phosphorylation and LTCC activity during chronic diabetic hyperglycemia are prevented by a P2Y11 inhibitor

To explore whether a P2Y11 inhibitor could prevent changes in Ser1928 phosphorylation and LTCC activity during chronic diabetic hyperglycemia, we exposed isolated arteries for 48 hr to 10 mM D-glucose, 20 mM mannitol, 20 mM D-glucose or 20 mM D-glucose +NF340. This organ culture method prevents arterial myocyte phenotypic changes associated with prolonged culturing conditions. Furthermore, the exposure time is sufficient to induce many of the vascular remodeling phenomena associated with elevated glucose in arterial myocytes, including increased LTCC activity, decreased K+ channel activity and downstream activation of Ca2+-dependent transcription factors (Navedo et al., 2010b; Nieves-Cintrón et al., 2015; Nilsson et al., 2006; Nystoriak et al., 2014; Nystoriak et al., 2017). LTCC function was assessed by measuring single-channel activity using the cell-attached configuration from dissociated arterial myocytes under each experimental condition described above. We found that wt cells isolated from the 20 mM D-glucose but not the 20 mM mannitol treated group had significantly enhanced LTCC open probability (nPo) compared to those in the 10 mM D-glucose group (Figure 8A and B). This increase in LTCC activity during diabetic hyperglycemia, however, was prevented in arterial myocytes from arteries treated in the presence of the P2Y11 inhibitor NF340. Ser1928 phosphorylation was significantly elevated in arteries chronically exposed to 20 mM D-glucose, but this effect was completely abolished in the presence of NF340 (Figure 8C and Figure 8—figure supplement 1A). Consistent with a key role for Ser1928 phosphorylation in potentiation of LTCC activity in response to chronic elevations in glucose, LTCC nPo was similar in arterial myocytes from S1928A arteries incubated for 48 hr in 10 mM or 20 mM D-glucose (Figure 8—figure supplement 1B). These results suggest that treatment with a P2Y11 inhibitor can avert PKA-mediated Ser1928 phosphorylation and stimulated LTCC activity during diabetic hyperglycemia.

Figure 8 with 1 supplement see all
Augmented LTCC activity and Ser1928 phosphorylation in response to chronic extracellular glucose elevations are prevented in the presence of the P2Y11 antagonist NF340.

(A) Representative single LTCC recordings obtained during a 2 s step depolarization from −80 to −30 mV and (B) bar plot of LTCC nPo in arterial myocytes isolated from mouse cerebral arteries incubated for 48 hr in 10 mM D-glucose (n = 10 cells from three mice), 20 mM mannitol (n = 13 cells from four mice), 20 mM D-glucose (n = 10 cells from four mice) and 20 mM D-glucose +10 µM NF340 (n = 10 cells from four mice). Channel openings (o) are represented by downward deflections from baseline (c) (*p<0.05, one-way ANOVA with Tukey post hoc test; Figure 8—source data 1). (C) Representative immunoblot detection of phosphorylated Ser1928 (pSer1928) and total CaV1.2 from mouse cerebral and mesenteric arteries incubated for 48 hr in 10 mM D-glucose, 20 mM D-glucose and 20 mM D-glucose +10 µM NF340 and densitometry quantification of pSer1928/CaV1.2 ratio (n = 7 arterial lysates per condition) (*p<0.05, Kruskal-Wallis with Dunn’s multiple comparisons; Figure 8—source data 2). (D) Proposed model for the role of P2Y11 in PKA-dependent stimulation of LTCC activity and vasoconstriction during diabetic hyperglycemia (NUC = nucleotides).

https://doi.org/10.7554/eLife.42214.063

Discussion

Increased arterial myocyte contractility and vasoconstriction in response to elevated extracellular glucose contribute to vascular complications in diabetes (Cooper et al., 2001; Fleischhacker et al., 1999; Jackson et al., 2016; Montero et al., 2013; Nieves-Cintrón et al., 2015; Nystoriak et al., 2017). This endothelium-independent component is attributed, at least in part, to glucose-induced decreases in K+ channel expression and activity (Jackson et al., 2016; Nieves-Cintrón et al., 2015; Nieves-Cintrón et al., 2017; Nieves-Cintrón et al., 2018; Nystoriak et al., 2014; Straub et al., 2009), transcriptional remodeling through NFATc3 signaling (Nieves-Cintrón et al., 2015; Nieves-Cintrón et al., 2018; Nilsson et al., 2006; Nystoriak et al., 2014), and alterations in PKC/Rho kinase signaling (Abd-Elrahman et al., 2015; Hien et al., 2016; Kizub et al., 2010). Our group has additionally established, in human and mouse arterial tissue, PKA-mediated potentiation of LTCC activity via increased phosphorylation of the pore-forming CaV1.2 subunit at Ser1928 as a key event that contributes to vascular complications during diabetic hyperglycemia (Morotti et al., 2017; Navedo et al., 2010b; Nystoriak et al., 2017). The classic molecular machinery stimulating PKA activity implicates activation of upstream effector enzymes (e.g. AC) and G protein-coupled receptors (GPCR). In the present study, using native human arteries and arterial myocytes, we identified a Gs-coupled P2Y receptor with the molecular, pharmacological, and signaling properties of P2Y11 as the GPCR underlying the glucose triggering cascade that results in PKA-dependent potentiation of LTCCs (Figure 8D). Indeed, P2Y11 is known to associate with ACs to stimulate cAMP/PKA signaling (Communi et al., 1997; Pucéat et al., 1998). Similar results were observed in mouse arteries and arterial myocytes, indicating that a P2Y11-like receptor could be mediating the glucose response in these cells. The implications of these observations are profound, as they may shed light on unique mechanisms of diabetic vascular complications.

Elevations in extracellular glucose have been shown to promote autocrine release of nucleotides in a number of cells, including arterial myocytes (Costa et al., 2009; Hazama et al., 1998; Nilsson et al., 2006; Parodi et al., 2002). This glucose-induced release of nucleotides has been associated with increased global [Ca2+]i in arterial myocytes, but the mechanisms and functional implications were not clearly established (Nilsson et al., 2006). This prompted us to examine whether extracellular nucleotides could be mediating glucose effects on LTCC activity and vascular reactivity. Our first major finding was that hydrolysis of extracellular nucleotides by the ectonucleotidase apyrase prevented the increases in Ser1928 phosphorylation, LTCC activity and vasoconstriction in response to elevations in extracellular glucose (Figure 1). Outcomes from LTCC recordings under continuous flow or static bath conditions (Figure 1—figure supplement 2) further support the involvement of extracellular nucleotide signaling in mediating the glucose effects. Given the well-established function of extracellular nucleotides in activating P2Y receptors (see (von Kügelgen and Harden, 2011)), our studies also pointed to the potential contribution of at least one of the eight known P2Y receptors in this process. In line with this concept, our multidisciplinary approach using human tissue suggested involvement of a P2Y receptor fitting the profile for the P2Y11 subtype. Intriguingly, similar results were observed using mouse tissue/cells, suggesting a role for a P2Y11-like receptor. To date, this is the only Gs-coupled P2Y, which could activate AC, leading to increased cAMP synthesis and thereby PKA activation. Considering the essential actions of PKA on LTCC activity and vasoconstriction in response to elevated glucose (Morotti et al., 2017; Navedo et al., 2010b; Nystoriak et al., 2017), we submit that such changes may proceed through activation of this purinergic signaling pathway.

Few studies have examined the role of P2Y11 in the cardiovascular system. Current knowledge suggests that this receptor is associated with positive inotropic effects of ATP on cardiomyocytes and its activity seems to be impaired during cardiomyopathy (Balogh et al., 2005). Moreover, a polymorphism (Ala-87-Thr) in P2Y11 has been linked to increased risk of acute myocardial infarction and C-reactive protein blood levels (Amisten et al., 2007). These results indicate a role for P2Y11 in cardiac (dys)function. However, the functional significance of P2Y11 in the vasculature, and particularly in arterial myocytes, is virtually unknown. In the present study, Western blot analysis detected an immunoreactive band of the expected molecular weight for P2Y11 (e.g. ~40 ± 10 kDa) in human arterial lysates (Figure 2C). We confirmed robust distribution of the receptor at the expected plasma membrane location in isolated human arterial myocytes (Figure 2D). In line with the expected function for this GPCR, we corroborated localized subplasmalemmal cAMP synthesis upon receptor stimulation with a highly selective P2Y11 agonist (NF546 (Meis et al., 2010); Figure 5). Notably, 20 mM D-glucose triggered cAMP synthesis of the same magnitude as the NF546 compound, and simultaneous application of NF546 and 20 mM D-glucose did not have any additional effect when compared with individual treatment exposure, suggesting that both stimuli proceed through the same signaling pathway. In support of this possibility, the specific P2Y11 antagonist NF340 (Meis et al., 2010) blocked the NF546 and NF546 + 20 mM D-glucose effects on cAMP synthesis (Figure 5). The P2Y11 antagonist also prevented increased Ser1928 phosphorylation and LTCC activity in response to acute elevations in extracellular glucose (Figure 6), while the P2Y11 agonist NF546 recapitulated all the glucose-induced, PKA-dependent effects (Figure 7). Moreover, the NF340 compound hampered glucose-induced changes in Ser1928 phosphorylation levels and LTCC activity during chronic hyperglycemic conditions resembling those observed during diabetes (Figure 8). Altogether, these results revealed a role for human P2Y11 or mouse P2Y11-like receptor in arterial myocytes, particularly during glucose signaling, and suggest that this GPCR may be involved in the initiation and/or progression of arterial myocyte dysfunction leading to vascular complications during diabetic hyperglycemia. Our results may also have profound clinical implications as they add to a growing body of evidence implicating a role for altered P2Y11 function in a number of pathological conditions (Nishimura et al., 2017), which could make them a potential therapeutic target.

Our findings indicating that elevated extracellular glucose triggers PKA-dependent potentiation of LTCC activity to induce vasoconstriction were unexpected, as PKA activity has been traditionally associated with vasodilation due, in part, to increased K+ channel activity. A reasonable prediction was that glucose could influence LTCCs via a cAMP/PKA signaling pathway that is distinct from that governing K+ channels as occurs upon β adrenergic stimulation (Moore et al., 2014). Such separation in cAMP would require precise subsarcolemmal compartmentalization of key proteins and signaling molecules (including GPCR, signaling enzyme, second messenger, effector protein and substrate). In the case of PKA-dependent effects on LTCC function in response to elevated glucose, this separation would include seclusion of a Gs-coupled receptor, AC, PKA and CaV1.2. Indeed, Tajada et al recently showed that optimal ion channel regulation by effector proteins is constrained by their nanometer distance between each other (Tajada et al., 2017). In support of this possibility, our recent work revealed nanometer proximity of a subpopulation of CaV1.2 (~10% of CaV1.2 clusters) to pools of PKA (Nystoriak et al., 2017). Here, we found a close association (≤40–90 nm) between pools of P2Y11 and CaV1.2 as well as P2Y11 and PKA (Figure 3). Moreover, given the cAMP biosensor data demonstrating that both glucose and the P2Y11 agonist NF546 induce localized subsarcolemmal cAMP synthesis (Figure 5), it is also likely that P2Y11 may be in close proximity to a specific AC isoform in arterial myocytes - the identity of which remains to be established. Nevertheless, this implicates the formation of a distinctive nanomolecular domain likely containing P2Y11, AC, PKA and CaV1.2 in arterial myocytes (Figure 8D), although its confirmation is beyond the scope of this study. This nanodomain may ensure the necessary signaling compartmentalization required for PKA-mediated effects on LTCC activity and vascular reactivity in response to elevated glucose.

The spatial organization of the P2Y11/AC/PKA/CaV1.2 nanodomain could be facilitated by scaffold proteins, such as A kinase anchoring proteins (AKAPs). These AKAPs fine tune signal transduction by localizing GPCRs, signaling enzymes and effector proteins in close proximity to their substrates (Johnstone et al., 2018; Langeberg and Scott, 2015). The A kinase anchoring protein 150 (AKAP150; murine ortholog of human AKAP79) is of particular interest as it has been shown to interact with AC, PKA and CaV1.2 (Bauman et al., 2006; Efendiev et al., 2010; Gao et al., 1997; Murphy et al., 2014; Navedo et al., 2008; Nystoriak et al., 2017; Zhang et al., 2013). Intriguingly, although AKAP79/150 can associate with multiple GPCRs (Fraser et al., 2000; Zhang et al., 2011), no report to date has linked it to any purinergic receptor. Localization of P2Y11 within the same AKAP79/150-driven complex may be critical as this scaffold protein is necessary for increased L-type Ca2+ channel activity in response to elevated glucose (Nystoriak et al., 2017). Two possibilities that do not exclude a fundamental role for AKAP79/150 are 1) direct interaction of P2Y11 with the scaffold protein and/or 2) association of the receptor with a pool of CaV1.2 within the AKAP79/150 complex. Future studies should examine these exciting alternatives as they may provide further insight into mechanisms regulating LTCCs and vascular reactivity during diabetic hyperglycemia.

A lingering question is how nucleotides, such as ATP, are released in response to glucose to modulate arterial myocyte excitability. Although our data do not identify the specific nucleotide, ATP release is particularly interesting as this nucleotide is an effective vasoactive agent and the endogenous P2Y11 agonist (Communi et al., 1997; Kennedy, 2017; Schuchardt et al., 2012; von Kügelgen and Harden, 2011). Indeed, ATP can activate P2Y11 with an EC50 that ranges from 1.8 to 17 μM, depending on the readout and the cell type used (Communi et al., 1999; Haas et al., 2014; Meis et al., 2010; Qi et al., 2001). Moreover, P2Y11 can be activated by ATP/ATP-derived nucleotides with the following potency order: ATPγS > dATP > ATP>>>ADP, with other nucleotides such as UTP and UDP having minimal or no influence on receptor activation (Kennedy, 2017; Schuchardt et al., 2012; von Kügelgen and Harden, 2011). Our recent study suggested that glucose effects on LTCC activity and vascular reactivity require glucose transport into the cell and subsequent metabolism that can generate ATP (Nystoriak et al., 2017). ATP can then be transported out of the cell through ATP-binding cassettes, vesicular exocytosis, plasma membrane F1F0-ATPase, connexin hemichannels and pannexin channels to induce activation of P2Y11 (Lohman et al., 2012). We thus speculate that one of these ATP transporters could form part of the AKAP/P2Y11/AC/PKA/CaV1.2 nanocomplex to provide another layer of compartmentalization that facilitates PKA-dependent regulation of LTCC function and vascular reactivity in response to elevations in extracellular glucose.

In summary, our data in human and murine tissue provide evidence of a Gs-coupled purinergic receptor that fits the profile of P2Y11 or a P2Y11-like receptor, respectively, as a key upstream component in the signaling cascade regulating LTCC activity and vascular reactivity during diabetic hyperglycemia. Data start to define a unique nanomolecular complex likely formed by AKAP/P2Y11/AC/PKA/CaV1.2. This nanocomplex can organize glucose signaling and help explain the intriguing consequences of glucose-induced PKA activation on LTCC function and vascular reactivity, which may have significant pathological implications, and therapeutic potential to treat vascular complications during diabetic hyperglycemia.

Materials and methods

Key resources table
Reagent
type (species)
or resource
DesignationSource or referenceIdentifiersAdditional
information
Strain
(Mus musculus),
C57BL/6J
wild-typeJackson Laboratoriesstock # 000664
Strain
(Mus musculus)
S1928A(Lemke et al., 2008)
Cell line
(human
embryonal
kidney)
tsA-201Sigma-Aldrich96121229SV40 transformed
Oligodeoxynucleotide, senseP2Y11 SNS ODNIntegrated DNA Technologieshuman P2Y11 sequence
(NC_000019.10): 5’-CAACGTCTCGGGTAAGGAGAA-3’ and 5’-ATGAGGAAGGAAACGTGGGT-3’
last three bases on the 3’ end were phosphoro-thioated
Oligodeoxynucleotide, antisenseP2Y11 ANS ODNIntegrated DNA Technologieshuman P2Y11 sequence
(NC_000019.10): 5’-CAAGGCCACCCTAACCACTG-3’
and 5’-CTCTCCCTTCCCTGCGTTA-3’
last three bases on the 3’ end were phosphoro-thioated
DNA constructhuman P2Y11UMR cDNA
Resource Center
www.cDNA.org; clone ID
P2Y1100000
tagged with GFP at C-
terminus
Antibodyanti-FP1 (CaV1.2; custom rabbit)(Davare et al., 2000)dilutions: 1:100 for
immunoblot and PLA;
10 µg/mL for GSD
Antibodyanti-CH3P
(pSer1928,
custom rabbit)
(Davare et al., 2000)1:50-1:100 dilutions
Antibodyanti-β-actin (mouse monoclonal)Abcamab8226
RRID: AB_30637
1:1000 dilution
Antibodyanti-α-tubulin (mouse monoclonal)Active Motif39527; clone 5-B-1–21:500 dilution
Antibodyanti-P2Y11 (rabbit polyclonal)Abcamab1807391:100-1:200 dilutions
Antibodyanti-P2Y11 (goat polyclonal)Santa Cruz
Biotechnology
sc-69588; clone C-18
RRID: AB_21559
dilutions: 1:100 for
stainings (PLA,
classical) and 10 µg/mL
for GSD
Antibodyanti-PKAcatα, β, γ (mouse
monoclonal)
Santa Cruz
Biotechnology
sc-28315; clone A-21:200 dilution
AntibodyPKAcatα, β, γ blocking peptideSanta Cruz
Biotechnology
sc-28315 P; clone A-21:20 dilution for 1 µg of primary
antibody
Antibodyanti-PKAcat α, β, γ (rabbit polyclonal)Santa Cruz Biotechnologysc-28892; clone H-95dilutions: 1:200 for PLA and 10 µg/mL for GSD
AntibodyAlexa Fluor 488 conjugate of wheat germ agglutininLife TechnologiesW7024
AntibodyAlexa Fluor 568-conjugated donkey anti-goatMolecular ProbesA11057
RRID: AB_142581
5 mg/mL dilution
AntibodyAlexa Fluor 568-conjugated donkey anti-mouseMolecular ProbesA100375 mg/mL dilution
AntibodyAlexa Fluor 568-conjugated donkey anti-rabbitMolecular ProbesA11011
RRID: AB_143157
2 µg/mL dilution
AntibodyAlexa Fluor 647-conjugated donkey anti-goatMolecular ProbesA21447
RRID: AB_141844
2 µg/mL dilution
Antibodygoat anti-rabbit IgG (H + L)-horseradish peroxidase conjugateBio-Rad170–6515
RRID: AB_11125142
1:10000 dilution
Antibodygoat anti-mouse IgG (H + L)-horseradish peroxidase conjugateBio-Rad170–65161:10000 dilution
Chemical
compound,
drug
sodium pentobarbital (Fatal-Plus)Vortech Pharma-ceuticalsNDC 0298-9373-68
Chemical
compound,
drug
mannitolFisher ScientificBP686
Chemical
compound,
drug
NF340Santa Cruz Biotechnologysc-361274
Chemical
compound,
drug
NF546Tocris3892
Chemical
compound,
drug
apyraseNew England
Biolabs
M0398L
Chemical
compound,
drug
nifedipineSigma-AldrichN7634
Chemical
compound,
drug
Bay K-8644Sigma-Aldrich71145-03-4
Chemical
compound,
drug
forskolinSigma-AldrichF6886
Chemical
compound,
drug
amphotericin BSigma-AldrichA4888
Chemical
compound,
drug
MRS2578Santa Cruz Biotechnologysc-204103A
Chemical
compound,
drug
MRS2179Tocris1454889-37-2
Chemical
compound,
drug
protein kinase A inhbitor (PKI)Sigma-AldrichP9115fragment
14–22, myristoylated
trifluoroacetate salt
Chemical
compound,
drug
Rp-Adenosine 3’,5’-cyclic
monophos-phorothioate triethylam-monium salt (rpcAMP)
Sigma-AldrichA165
Chemical
compound,
drug
Adenosine 5’-[γ-thio] triphosphate tetralithium salt (ATPγS)Sigma-AldrichA1388
Software,
algorithm
GraphPad PrismGraphPad Prism, RRID: SCR_002798
Software,
algorithm
ImageJFiji,
RRID: SCR_002285
Software,
algorithm
pCLAMP10Molecular Deviceselectrophysiology
Software,
algorithm
LASAFLeicaGSD
Software,
algorithm
IonOptixIonOptixarterial diameter
recordings
Software,
algorithm
MetaflorMolecular DevicesFRET

Animals

Male wild type C57BL/6J (wt) or knockin mice in which Ser1928 of CaV1.2 was mutated to Ala (S1928A) (Lemke et al., 2008) of 5 to 8 weeks of age were euthanized with a lethal dose of sodium pentobarbital (250 mg/kg; intraperitoneally), as approved by the University of California, Davis Animal Care and Use Committee (protocol #: 20321).

Human tissue

Request a detailed protocol

Excised adipose arteries from human patients undergoing surgical sleeve gastrectomy were used for this study (Supplementary file 3). Samples were obtained after Institutional Review Board (IRB) approval from the University of Nevada Reno School of Medicine (IRB ID: 2013–019) and in accordance with the guidelines of the Declaration of Helsinki. The need for informed consent was waived by IRBs at the University of Nevada Reno School of Medicine (IRB ID: 2013–019) and the University of California Davis (IRB ID: 597267–1) because the tissue is considered ‘waste’, has no codification that could be used to identify patients and was determined not to be human subject research in accordance with United States of America federal regulations, as defined by 45 CFR 46.102(f). This precludes the acquisition of detailed clinical profiles other than sex, age and whether the patient is nondiabetic or diabetic. Only samples from nondiabetic patients were used. No exclusions were made due to medication history, sex or presence of comorbidities. Collected tissue was placed in cold phosphate-buffered saline (PBS) solution containing (in mM): 138 NaCl, 3 KCl, 10 Na2HPO4, 2 NaH2PO4, 5 D-glucose, 0.1 CaCl2, and 0.1 MgSO4, pH 7.4 with NaOH until used.

Arterial myocyte isolation

Request a detailed protocol

Mouse cerebral arteries were dissected in ice-cold dissection buffer containing (in mM): 140 NaCl, 5 KCl, 2 MgCl2, 10 D-glucose, 10 HEPES, pH 7.4, with NaOH. Arteries were digested in dissection buffer containing papain (1 mg/mL) and dithiothreitol (1 mg/mL) at 37° C for 7 min, followed by incubation in dissection buffer containing collagenase type F (0.7 mg/mL) and collagenase type H (0.3 mg/mL) at 37° C for 7 min. Arteries were washed in ice-cold dissection buffer and gently triturated with glass pipettes to disperse the cells, which were kept in ice-cold dissection buffer until use.

Single human arterial myocytes (Supplementary file 3) were isolated as previously described (Nieves-Cintrón et al., 2017; Nystoriak et al., 2017). Briefly, small diameter adipose arteries from human samples were enzymatically digested in dissection solution supplemented with papain (26 U/mL) and dithiothreitol (1 mg/mL) at 37° C for 15 min, followed by incubation in dissection solution containing collagenase type H (1.95 U/mL), elastase (0.5 mg/mL), and trypsin inhibitor (1 mg/mL) at 37° C for 10 min. Cells were then washed three times in ice-cold dissection buffer, triturated with glass pipettes to disperse the individual cells, and maintained in ice-cold dissection buffer until use.

For unpassaged, cultured human and mouse arterial myocytes, human adipose arteries and mouse aortas were dissected out and placed in ice-cold Dulbecco’s Modified Eagle Medium (DMEM; Gibco – Life Technologies, Grand Island, NY) containing 1X glutamate, 1X pyruvate, 1X penicillin/streptavidin and fungizone (0.25 g/mL). Artery segments were subsequently transferred and incubated in a DMEM solution containing 2.2 mg/mL of collagenase Type 2 (Worthington) at 37° C for 15 min to remove the adventitia. To disperse and culture unpassaged arterial myocytes, the tissue was cut into 2–5 mm segments and incubated at 37° C with constant shaking in a buffer containing (in mM): 134 NaCl, 6 KCl, CaCl2, 10 HEPES, and 7 D-glucose supplemented with 2.2 mg/mL collagenase Type 2 (Worthington). The digestion was stopped by adding an equal volume of DMEM containing 5% fetal bovine serum. Digested tissue was then centrifuged for 5 min at 14,000 rpm. The supernatant was removed and the pellet containing the digested tissue was resuspended in DMEM containing 1X glutamate, 1X pyruvate, 5% serum and 5 mM D-glucose, resulting in dispersion of individual arterial myocytes. Cells were then seeded on glass coverslips coated with laminin and kept in an incubator at 37° C with 5% CO2 for 2–3 days before adenoviral transduction or lysis for immunoblotting (see section below).

Arterial diameter measurements

Request a detailed protocol

Freshly isolated posterior mouse cerebral arteries were cannulated on glass micropipettes mounted in a 5 mL myograph chamber (University of Vermont Instrumentation and Model Facility), as described previously (Nieves-Cintrón et al., 2017; Nystoriak et al., 2014; Nystoriak et al., 2017). Arteries were allowed to equilibrate at an intravascular pressure of 20 mmHg, while being continuously superfused (37° C, 30 min, 3–5 mL/min) with physiological saline solution composed of (in mM): 119 NaCl, 4.7 KCl, 2 CaCl2, 24 NaHCO3, 1.2 KH2PO4, 1.2 MgSO4, 0.023 EDTA, and 10 D-glucose aerated with 5% CO2/95% O2. Bath pH was closely monitored and maintained at 7.35 to 7.4. After equilibration, intravascular pressure was increased to 60 mmHg and arteries allowed to develop stable myogenic tone. Arteries not exhibiting stable tone after ~1 hr were discarded. To assess the response of arterial diameter to hyperglycemia, D-glucose was increased from 10 mM to 20 mM in the perfusion solution. Arterial tone data are presented as a percent decrease in diameter relative to the maximum passive diameter at 60 mmHg obtained using Ca2+-free saline solution containing nifedipine (1 µM) at the end of the experiment.

Electrophysiology

Request a detailed protocol

All experiments were performed at room temperature (22–25° C). Both whole-cell and single-channel data were acquired using an Axopatch 200B amplifier and Digidata 1440 digitizer (Molecular Devices). Recording electrodes were pulled from borosilicate capillary glass using a micropipette puller (model P-97, Sutter Instruments) and subsequently polished to achieve resistances that ranged from 3.5 to 6.5 MΩ. L-type Ca2+ channel currents were assessed in freshly dissociated arterial myocytes using the perforated whole-cell mode of the patch-clamp technique with Ba2+ as a charge carrier (IBa) after myocytes were allowed to attach (10 min) to a glass coverslip in a recording chamber. Borosilicate glass pipettes were filled with a solution containing (in mM): 120 CsCl, 20 tetraethylammonium chloride (TEA-Cl), 1 EGTA, and 20 HEPES with amphotericin B (250 µg/mL; pH adjusted to 7.2 with CsOH). The bath solution consisted of (in mM): 115 NaCl, 10 TEA-Cl, 0.5 MgCl2, 10 D-glucose, 5 CsCl, 20 BaCl2, and 20 HEPES, pH adjusted to 7.4 with CsOH. IBa were elicited by 200 ms depolarizing pulses from a holding potential of −70 mV to +10 mV. IBa were first recorded in the presence of 10 mM D-glucose under static flow conditions. This solution was then exchanged for a solution containing either 20 mM D-glucose, 20 mM mannitol or NF546 at a rate of 2.1 mL/min. Flow was stopped after 3 min, and IBa were recorded again under static flow conditions at least 5 min after the indicated treatment was initiated. In some experiments, arterial myocytes were pretreated for 10–15 min with indicated inhibitors prior to IBa recordings. For experiments comparing IBa before and after elevated glucose under continuous flow and static bath conditions, cells were patched in a bath solution containing 10 mM D-glucose at a perfusion rate of 2.1 mL/min. After establishment of a stable gigaseal for at least 5 min, IBa were recorded in the presence of 10 mM D-glucose under continuous flow. Those cells that showed robust IBa in 10 mM D-glucose were then perfused with a bath solution containing 20 mM D-glucose under continuous flow for at least 5 min before recording of IBa again. Subsequently, IBa were recorded one more time on the same cell, 5 min after stopping the bath perfusion (e.g. static bath conditions). Currents were sampled at 10 kHz and low pass-filtered at 5 kHz. The CaV1.2 blocker nifedipine (1 μM) was applied at the end of each experiment to determine the nifedipine-sensitive component. The I-V relationship of nifedipine-sensitive IBa was determined by 200 ms depolarizing steps from −70 mV to voltages ranging from −60 to +60 mV in increments of +10 mV. The I-V relationship for averaged data sets was fit with a peak Gaussian function: I(V)=Imax x exp(−0.5((V – Vmax)/b)2), where Imax is peak I, Vmax is V at Imax, and b is the slope of the distribution, as described previously (Nystoriak et al., 2017). A voltage error of 10 mV attributable to the liquid junction potential of the recording solutions was corrected offline.

The cell-attached configuration of the patch-clamp technique was used to examine single-channel Ca2+ currents in arterial myocytes using Ca2+ as the charge carrier, as previously described (Cheng et al., 2011; Dixon et al., 2015; Dixon et al., 2012; Navedo et al., 2010a; Nystoriak et al., 2017). Resting membrane potential was fixed to ~0 mV using a high K+ bath solution (Hess et al., 1986; Zampini et al., 2006) composed of (in mM): 145 KCl, 10 NaCl, and 10 HEPES, pH adjusted to 7.4 with NaOH. Pipette solution was composed of (in mM): 120 TEA-Cl, 110 CaCl2, and 10 HEPES with BayK-8644 (500 nM), which promotes longer open times of the channel thereby increasing the probability of sweeps with channel activity. Single L-type Ca2+ currents were elicited by a 2 s depolarizing step from a holding of −70 mV to −30 mV. Single-channel currents were sampled at 10 kHz and low pass filtered at 2 kHz, followed by a Gaussian filter (500 Hz) during analysis. Capacitive currents were compensated by subtraction of blank sweeps. When nifedipine was included in the patch pipette, CaV1.2 activity was negligible. Single-channel openings and nPo were analyzed using the single-channel event half-amplitude detection algorithm from pCLAMP 10.

Immunoprecipitation and immunoblotting

Request a detailed protocol

Human adipose arteries, mouse cerebral and mesenteric arteries, and mouse arterial myoyctes were homogenized in a RIPA lysis buffer solution composed of 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 5 mM EGTA, 10 mM EDTA, 1% nonyl phenoxypolyethoxylethanol-40 (NP-40), 10% glycerol, 0.05% sodium dodecyl sulfate (SDS), 0.4% deoxycholic acid (DOC) with protease inhibitors (1 µg/ml pepstatin A, 10 µg/ml leupeptin, 20 µg/mL aprotinin, and 200 nM phenylmethylsulfonyl fluoride) and phosphatase inhibitors (2 µM microcystin LR, 1 mM p-nitrophenyl phosphate, 50 mM Na-pyrophosphate, and 50 mM NaF), then sonicated for 5 min at 4° C and cleared by centrifugation (15,000 X g) for 20 min at 4° C. For Ser1928 phosphorylation experiments, the soluble fraction was incubated on a head-over-head tilter with protein A-Sepharose beads and a purified custom antibody against CaV1.2 (5 µg of FP1; see Davare et al., 2000) or nonspecific rabbit immunoglobulin G (IgG) control for 4 hr at 4° C. Beads were washed three times with washing buffer containing (in mM): 150 NaCl, 10 EDTA, 10 EGTA, 10 Tris-HCl (pH 7.4), 0.1% Triton X-100, pH 7.4. Samples were extracted in Laemmli Sample Buffer (Bio-Rad) for 5 min at 95° C for immunoprecipitation (Ser1928 phosphorylation) experiments and 15 min at 80° C for other immunoblot experiments. Proteins were separated by SDS-polyacrylamide gel electrophoresis at 75–100 V for 1.5 hr in a stacking gel polymerized from 3% acrylamide and a resolving phase polymerized from 7.5% (Ser1928 phosphorylation) or 10% acrylamide. Proteins were then transferred to a polyvinylidene difluoride membrane at 50 V for 600 min at 4° C. All membranes, except for P2Y11 blots (10% Odyssey blocking buffer; LI-COR Biosciences), were blocked in 5% nonfat dried milk in tris-buffered saline with 0.05% Tween 20 (TBS-T) for 1 hr at room temperature before primary antibody incubation for 2–3 hr at room temperature. Antibodies and dilutions were as follows: rabbit anti-CH3P (pSer1928; 1:50-1:100),(Davare et al., 2000) rabbit anti-FP1 (CaV1.2; 1:100) (Davare et al., 2000), rabbit anti-P2Y11 (Abcam, 1:100-1:200); mouse anti-β-actin (Abcam, 1:1000), and anti α-tubulin (Active Motif, 1:500). Except for the P2Y11 antibody, which was diluted in 1% bovine serum albumin, all antibodies were diluted in 5% nonfat dried milk in TBS-T. After washing, membranes were then incubated 1 hr at room temperature with horseradish peroxidase-labeled goat anti-rabbit (Bio-Rad; 1:10,000) or goat-anti-mouse antibodies (Bio-Rad; 1:10,000) diluted in 5% nonfat dried milk in TBS-T or 5% Odyssey in TBS-T (P2Y11 only) and developed on autoradiography film using chemiluminescence (Classico [Millipore] and Femto [Thermo Fisher Scientific]). For Ser1928 phosphorylation and P2Y11 knock down experiments, total CaV1.2 and β-actin, respectively, were used for normalization (density expressed as percentage of total CaV1.2/β-actin and these ratios were normalized to the ratio of the control band to get the relative density of pSer1928/CaV1.2 or P2Y11/β-actin for all treatments). For Ser1928 phosphorylation quantification in Figure 1—figure supplement 1E, triplicates were run in parallel and all values were normalized to the ratio of the loading control band (e.g. α-tubulin obtained from the same samples) to get the pSer1928/α-tubulin ratio for all treatments. The pSer1928/α-tubulin ratios were then normalized to the first control band ratio to obtain the relative density of pSer1928/α-tubulin. Densitometry analysis for bands was performed with ImageJ software (National Institutes of Health) as follow. Developed films were scanned to tiff files, uploaded to ImageJ and color inverted. Immunoreactive bands were outlined and light intensity per area was measured. Doublet bands were outlined and measured as one signal. Intensity of equal area above and below each immunoreactive band was measured to average and then subtract the background signal. Importantly, films were exposed for increasing time periods to ensure signals were in the linear range as described in Davare and Hell (2003) and Hall et al. (2006). β-actin and α-tubulin were used as a loading control in some experiments.

Organ culture

Request a detailed protocol

Freshly isolated arteries for acute Ser1928 phosphorylation experiments were incubated for 5 min at room temperature in a physiological saline solution composed of (in mM): 119 NaCl, 4.7 KCl, 2 CaCl2, 24 NaHCO3, 1.2 KH2PO4, 1.2 MgSO4, 0.023 EDTA, and 10 (5 for humans) D-glucose aerated with 5% CO2/95% O2 to reach a pH of 7.35–7.4. Arteries were then incubated for 10 min at room temperature in a similar physiological solution containing the specified treatment. Following incubation, arteries were immediately placed in RIPA lysis buffer with protease and phosphatase inhibitors (as described in immunoblotting section). For chronic exposure experiments, arteries were placed in DMEM/F-12 (Gibco-Life Technologies, Grand Island, NY) supplemented with L-glutamine (2 mM) containing 10 mM, 20 mM D-glucose or 20 mM mannitol in the absence or presence of 10 µM NF340 and incubated for 48 hr at 37° C. After incubation, arteries were quickly washed in PBS and then placed in RIPA lysis buffer with protease and phosphatase inhibitors for immunoblotting.

Immunofluorescence

Request a detailed protocol

Immunofluorescent labeling of freshly dissociated arterial myocytes was performed as described previously (Navedo et al., 2008), using a goat-anti P2Y11 (Santa Cruz Biotechnology, clone C-18; 1:100) or a mouse-anti PKAcat α, β, γ (Santa Cruz Biotechnology, clone A-2, 1:200) antibodies in PBS supplemented with 0.1% BSA. Alexa Fluor 488 conjugated wheat germ agglutinin (WGA; Life Technologies) was used to stain for the plasma membrane. The secondary antibodies were Alexa Fluor 568-conjugated donkey anti-goat and donkey anti-mouse (5 mg/mL; Molecular Probes). In control experiments, the PKAcat antibody was preabsorbed with a PKAcat blocking peptide (Santa Cruz Biotechnology, A-2; 1:20 for 1 μg of primary antibody). Cells were imaged (512 × 512 pixel images) using an Olympus FV1000 confocal microscope paired with an Olympus 60x oil immersion lens (NA = 1.4) and a zoom of 3.0 (pixel size = 0.1 µm). P2Y11-associated fluorescence was not detected in negative control experiments in which the primary antibody was substituted for PBS or boiled to test for antibody specificity. Cells for each group were imaged with the same laser power, gain settings, and pinhole.

Proximity ligation assay

Request a detailed protocol

A Duolink In Situ PLA kit (Sigma) (Fredriksson et al., 2002) was used to detect complexes consisting of P2Y11 and CaV1.2 and P2Y11 and PKAcat in freshly isolated arterial myocytes as previously described (Nieves-Cintrón et al., 2017; Nystoriak et al., 2017). Briefly, cells were plated on glass coverslips and allowed to adhere (1 hr, room temperature) prior to fixing with 4% paraformaldehyde (20 min), quenching in 100 mM glycine (15 min), and washing in PBS (2 × 3 min). Cells were permeabilized with 0.1% Triton X-100 (20 min) and then blocked (1 hr, 37° C) in 50% Odyssey blocking solution (LI-COR Bioscience). Cells were incubated overnight at 4° C with a specific combination of two primary antibodies in 0.1% Odyssey +0.05% Triton X-100 PBS solution: goat anti-P2Y11 (Santa Cruz Biotechnology, clone C-18; 1:100), custom rabbit anti-FP1 (CaV1.2, 1:100) (Davare et al., 2000) and rabbit anti-PKAcat α, β, γ (Santa Cruz Biotechnology, clone H-95, 1:200). Cells were incubated with only one primary antibody as negative controls. After primary antibody incubation, cells were washed with Duolink buffer A (2 × 5 min). Oligonucleotide-conjugated secondary antibodies (PLA probes: anti-goat MINUS and anti-rabbit PLUS) were used to detect P2Y11 and CaV1.2 and P2Y11 and PKAcat (1 hr, 37° C). Following incubation with probes, cells were washed with Duolink buffer A (2 × 5 min) and a ligation solution composed of two distinct oligonucleotides, complementary to the probes, and ligase was added (30 min, 37° C) to allow hybridization with the probes and formation of a closed DNA circle at sites of dual labeling, which serves as a template for a rolling circle amplification reaction (100 min, 37° C). After the amplification step, cells were washed with Duolink buffer B (2 × 10 min) and 1% buffer B (1 × 1 min). Coverslips were allowed to dry and subsequently mounted on a microscope slide with Duolink mounting medium. The fluorescence signal was detected using an Olympus FV1000 confocal microscope coupled with a 60x oil immersion lens (NA, 1.4). Images were collected at different optical planes (z-axis step = 0.5 µm). The stack of images for each sample was combined into a single-intensity projection image that was subsequently used for analysis of number of puncta/µm2 per cell.

Immunolabeling and Ground State Depletion (GSD) microscopy

Request a detailed protocol

Isolated arterial myocytes were allowed to adhere to a coverslip (1 hr) before fixing with 3% paraformaldehyde +0.1% glutaraldehyde solution in PBS (10 min) followed by washes with PBS (3 × 15 min). Cells were then incubated for 5 min with 0.1% sodium borohydride in H2O, followed by 3 × 5 min washes with PBS and incubated in permeabilization/blocking solution consistent of 0.05% Triton X-100% and 20% SEA BLOCK (Thermo Scientific) for 1 hr at room temperature. Cells were exposed overnight to primary antibodies [goat anti-P2Y11 (Santa Cruz Biotechnology, clone C-18), custom rabbit anti-CaV1.2 (FP1 (Davare et al., 2000)) or rabbit anti-PKAcat α, β, γ (Santa Cruz Biotechnology, clone H-95)] diluted in blocking buffer to a concentration of 10 µg/mL. Cells were briefly washed 3x with PBS, followed by additional longer washes (3x for 15 min). For secondary antibodies, donkey Alexa Fluor 647-conjugated antibody recognizing goat IgG (2 µg/mL; Molecular Probes) and donkey Alexa Fluor 568-conjugated antibody recognizing rabbit IgG (2 µg/mL; Molecular Probes) diluted in blocking buffer were added for 1 hr at room temperature. Secondary Ab was washed 3x with PBS, followed by longer washes (3x for 15 min). Specificity of secondary antibodies was tested in control experiments in which primary antibodies were omitted from the preparation (no 1° antibody controls). For imaging, coverslips were mounted on microscope slides with a round cavity containing MEA-GLOX imaging buffer (NeoLab Migge Laborbedarf-Vertriebs GmbH, Germany) and sealed with Twinsil (Picodent, Germany). The imaging buffer was composed of 10 mM MEA (cysteamine), 0.56 mg/mL glucose oxidase, 34 µg/mL catalase, and buffer containing 10% w/v glucose, 10 mM NaCl, and 50 mM Tris-HCl, pH 8.

Super-resolution images of arterial myocytes were obtained using a super resolution ground state depletion system (SR-GSD, Leica) dependent on stochastic single-molecule localization, equipped with high-power lasers (532 nm, 2.1 kW/ cm2; 642 nm, 2.1 kW/ cm2) and an additional 30 mW, 405 nm laser. A 160x HCX Plan-Apochromat (NA 1.47) oil immersion lens and an electron-multiplying charge-coupled device (EMCCD) camera (iXon3 897; Andor Technology) were used to acquire images (Dixon et al., 2015). The camera was running in frame-transfer mode at a frame rate of 100 Hz (10 ms exposure time). Fluorescence was detected through Leica high-power TIRF filter cubes (532 HP-T, 642 HP-T) with emission band-pass filters of 550–650 nm and 660–760 nm. Reconstruction of P2Y11, CaV1.2, and PKA distribution from 30,000 images used the coordinates of centroids obtained by fitting single-molecule fluorescence signals with a 2D Gaussian function using a LASAF software (Leica). The localization accuracy of the system is limited by the statistical noise of photon counting; the precision of localization is proportional to DLR/√N, where DLR is the diffraction-limited resolution of a fluorophore and N is the average number of detected photons per switching event, assuming the point-spread functions are Gaussian (Dempsey et al., 2011; Fölling et al., 2008). The full width at half maximum for single-molecule signals was ~20 nm as recently calculated by our group (Tajada et al., 2017). Localizations produced from less than 800 photons were filtered out of the reconstruction. All pixels with intensity above a user-defined threshold were binarized, evaluated and segmented into individual objects and included as clusters in our analysis. Cluster size and density were determined using the Analyze Particle option in the ImageJ software (National Institute of Health). The JACoP plug-in in the ImageJ software was used to unbiasedly and automatically determine the shortest intermolecular distances for P2Y11 and CaV1.2 as well as P2Y11 and PKAcat following the protocol described by Bolte and Cordelières (2006). Intermolecular distance histograms were generated from the JACoP plug-in output data and fitted with a sum of two Gaussian functions of the following equation: Y =Y0 + (A1 / (w1 x √(π/2))) x exp(−2 x ((X - XC1)/w1)2) + (A2 / (w2 x √(π/2))) x exp(−2 x ((X - XC2)/w2)2), where Y0 is the Y offset, A1 and A2 are the areas of the distribution of distances, XC1 and XC2 are the x values of distance at the center of the distribution, and w1 and w2 are the widths of each distribution in nanometers. To calculate the percentage of complete P2Y11 overlap, we first quantified the number of clusters/objects in the thresholded, binarized localization maps for P2Y11 and CaV1.2 as well as P2Y11 and PKAcat using the Analyze Particle tool in ImageJ. The binarized localization maps of P2Y11 were then multiplied by those corresponding to the paired CaV1.2 or PKAcat localization maps. The number of objects obtained from this operation were then divided by the number of clusters/objects present in the original binarized localization maps for P2Y11. This method only detects objects that are 100% overlapping with each other in two independent images (e.g localization maps for P2Y11 and CaV1.2). Random simulation of image pairs for P2Y11 and CaV1.2 or P2Y11 and PKAcat were generated based on the original super-resolution localization maps for each pair of proteins using the Coste’s randomization algorithm included in the JACoP plug-in in ImageJ (Bolte and Cordelières, 2006) from six different cells per condition. Parameters for the Coste’s randomization algorithm were selected to generate simulated images with relative similar cluster size and density to those observed in the original super-resolution localization maps for each protein. Each randomized image was generated after undergoing 1000 randomization rounds. Randomized images were binarized prior to analysis and the percentage of complete P2Y11 overlap from the randomized image pairs was calculated as described above.

Cell culture, transfection of tsA-201 cells, and reverse permeabilization

Request a detailed protocol

tsA-201 were obtained from Sigma-Aldrich (cat#: ECACC 96121229). These cells are included in the European Collection of Authenticated Cell Cultures. tsA-201 cells were cultured in DMEM supplemented with 1X pyruvate, 1X glutamax, 8% fetal bovine serum (FBS) and 5 mM glucose (without phenol red) at 37° C in a 5% CO2 incubator. Cells were transfected at 60–70% confluence with human P2Y11 tagged with GFP at C-terminus (UMR cDNA Resource Center - www.cDNA.org; clone ID P2Y1100000) with JetPRIME transfection reagent (Polyplus transfection SA, NY) for approximately 36 hr. P2Y11 sense and antisense oligodeoxynucleotides (ODNs; 2 µM; Integrated DNA Technologies) were transfected in tsA-201 cells 24 hr after initial transfection with P2Y11-GFP. The cells were gently washed with 1X PBS and then removed with a cell scrapper using a RIPA lysis buffer solution composed of 50 mM Tris base, 150 mM NaCl, 5 mM EGTA, 10 mM EDTA, 1% nonyl phenoxypolyethoxylethanol-40 (NP-40), 10% glycerol, 0.05% sodium dodecyl sulfate (SDS), 0.4% deoxycholic acid (DOC) with protease inhibitors (1 µg/mL pepstatin A, 10 µg/mL leupeptin, and 20 µg/mL aprotinin). Antisense and sense ODNs for P2Y11 were designed and checked for specificity using Prime BLAST against the human P2Y11 sequence (NC_000019.10). The sequences used for the antisense ODNs were as follows: 5’-CAAGGCCACCCTAACCACTG-3’ and 5’-CTCTCCCTTCCCTGCGTTA-3’. The sequences for sense ODNs were 5’-CAACGTCTCGGGTAAGGAGAA-3’ and 5’-ATGAGGAAGGAAACGTGGGT-3’. For all ODNs, the last three bases on the 3’ end were phosphorothioated to reduce degradation by cellular nucleases. ODNs were dissolved in nuclease-free water to a 1 mM concentration.

Human adipose arteries were permeabilized in a solution containing (in mM): 120 KCl, 2 MgCl2, 10 EGTA, 5 Na2ATP, 20 TES (2-[(2-Hydroxy-1, 1-bis(hydroxymethyl)ethyl)amino] ethanesulfonic acid, N-[Tris(hydroxymethyl)methyl]−2-aminoethanesufonic acid), pH adjusted to 6.8 with NaOH. Arteries were first incubated in permeabilization solution for 20 min at 4°C, followed by a 4 hr incubation at 4°C in a solution supplemented with ODNs (2 µM). Arteries were placed in an ODN-containing solution with elevated MgCl2 (10 mM) for 1 hr at room temperature. Permeabilization was reversed by incubating arteries for 30 min at room temperature in MOPS physiological solution composed of (in mM): 140 NaCl, 5 KCl, 10 MgCl2, 5 D-glucose, and 2 MOPS, pH adjusted to 7.1 with NaOH. Arteries were incubated in a MOPS solution containing 0.01, 0.1, and 1.8 mM CaCl2 in 15 min intervals, increasing Ca2+ gradually. After completing reverse permeabilization, arteries were cultured in D-MEM/F-12 culture media supplemented with L-glutamine (2 mM) for 2.5 days at 37°C. Arteries were then lysed and used for Western blot (see Immunoblotting section).

Adenovirus infection of unpassaged arterial myocytes and Fluorescence Resonance Energy Transfer (FRET)

Request a detailed protocol

Laminin (Life Technologies, Grand Island, NY) diluted 100x in sterile-filtered PBS (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4, pH = 7.4) was used to coat #0 glass coverslips (Karl Hecht, Sondheim, Germany). After adding diluted laminin (100 µL per coverslip), coverslips were placed in a 37° C incubator with 5% CO2 for a minimum of 2 hr, then moved to 24-well plate wells (Falcon, Tewksbury, MA), and washed 3x with sterile-filtered PBS. Freshly dissociated human adipose arterial myocytes and mouse aortic cells were plated on the laminin-coated coverslips with 500 μL of serum-containing media for 48 hr in a 37° C incubator with 5% CO2. Media was then replaced with 500 μL of serum-free media-containing virus coding for the membrane-targeted Epac1-camps-based FRET sensor (ICUE3-PM) (Allen et al., 2006; Liu et al., 2011) and placed at 37° C with 5% CO2 for another 36 hr. Viruses were produced using the AdEasy system (Qbiogene, Carlsbad, CA) (Luo et al., 2007). After infection, media was changed to serum-free media without virus. Glass coverslips were transferred to glass bottom culture dishes (MatTek, Ashland, MA) containing 3 mL PBS at room temperature.

A Zeiss AXIO Observer A1 inverted fluorescence microscope (San Diego, CA) equipped with a Hamamatsu Orca-Flash 4.0 digital camera (Bridgewater, NJ) and controlled by Metaflor software (Molecular Devices, Sunnyvale, CA) acquired phase contrast, CFP480, and FRET images. Phase contrast and CFP480 images were collected with 20x and 40x oil immersion objective lenses, while FRET images were collected using only the 40x oil immersion objective lens. Images for FRET analysis were recorded by exciting the donor fluorophore at 430–455 nm and measuring emission fluorescence with two filters (475DF40 for cyan and 535DF25 for yellow). Images were subjected to background subtraction and acquired every 30 s with exposure time of 200 ms for each channel. The donor/acceptor FRET ratio was calculated and normalized to the ratio value of baseline. CFP480 images were acquired by exciting the donor fluorophore at 430–455 nm and measuring emission fluorescence with the 475DF40 filter for 25 ms. Averages of normalized curves and maximal response to stimulation were graphed based on FRET ratio changes. Binding of cAMP to ICUE3-PM led to decreases in the YFP/CFP ratio, indicating increases in cAMP levels.

Chemicals and statistics

Request a detailed protocol

All chemical reagents were from Sigma-Aldrich (St. Louis, MO) unless otherwise stated. Data were analyzed using GraphPad Prism software and expressed as mean ±SEM. Data were assessed for potential outliers using the GraphPad Prism Outlier Test and for normality of distribution using appropriate tests. Statistical significance was then determined using suitable paired or unpaired Student’s t-test, nonparametric tests or One-way analysis of variance (ANOVA) for multiple comparisons with proper post hoc test. p<0.05 was considered statistically significant (denoted by * in figures).

References

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10
  11. 11
  12. 12
  13. 13
  14. 14
  15. 15
  16. 16
  17. 17
  18. 18
  19. 19
  20. 20
  21. 21
  22. 22
  23. 23
  24. 24
  25. 25
  26. 26
  27. 27
  28. 28
  29. 29
  30. 30
  31. 31
  32. 32
  33. 33
  34. 34
  35. 35
  36. 36
  37. 37
  38. 38
  39. 39
  40. 40
  41. 41
  42. 42
  43. 43
  44. 44
  45. 45
  46. 46
  47. 47
  48. 48
  49. 49
  50. 50
  51. 51
  52. 52
  53. 53
  54. 54
  55. 55
  56. 56
  57. 57
  58. 58
  59. 59
  60. 60
  61. 61
  62. 62
  63. 63
  64. 64
  65. 65
  66. 66
  67. 67
  68. 68
  69. 69
  70. 70
  71. 71
  72. 72
  73. 73
  74. 74
  75. 75
  76. 76
  77. 77
  78. 78
  79. 79
  80. 80
  81. 81
  82. 82
  83. 83
  84. 84

Decision letter

  1. Richard S Lewis
    Reviewing Editor; Stanford University School of Medicine, United States
  2. Richard Aldrich
    Senior Editor; The University of Texas at Austin, United States
  3. Richard S Lewis
    Reviewer; Stanford University School of Medicine, United States
  4. Mark T Nelson
    Reviewer; University of Vermont, United States

In the interests of transparency, eLife includes the editorial decision letter and accompanying author responses. A lightly edited version of the letter sent to the authors after peer review is shown, indicating the most substantive concerns; minor comments are not usually included.

Thank you for submitting your article "A Gs-coupled purinergic receptor boosts Ca2+ influx and vascular excitability during diabetic hyperglycemia" for consideration by eLife. Your article has been reviewed by three peer reviewers, including Richard S Lewis as the Reviewing Editor and Reviewer #1, and the evaluation has been overseen by Richard Aldrich as the Senior Editor. The following individual involved in review of your submission has agreed to reveal their identity: Mark T Nelson (Reviewer #2). A further reviewer remains anonymous.

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

Summary:

Cardiovascular complications of diabetes arise from augmented vasoconstriction of arterial myocytes in response to elevated glucose. In this manuscript, Navedo and colleagues demonstrate a pivotal role for the Gs- coupled metabotropic purinergic receptor, P2Y11, in mediating the vasoconstrictor response to glucose in mouse and human vascular smooth muscle myocytes. Using molecular, pharmacological/functional and FRET-based approaches, the authors provide evidence that P2Y11 is expressed in mouse and human vascular smooth muscle cells, and that glucose causes vasoconstriction through a cascade involving ATP release, P2Y11 activation, cAMP synthesis, and PKA-dependent phosphorylation of CaV1.2. In support of the PKA-dependent activation of CaV1.2 and subsequent vasoconstriction, they show using proximity analyses that a subpopulation of P2Y11 is in close proximity to PKA and CaV1.2. The signaling specificity afforded by such microdomains may explain the paradoxical vasoconstriction-promoting effect of PKA, which is generally associated with vasodilatory responses through effects on K+ channels. The role of P2Y11 is convincing and is a major new insight. Also significant is the identification of smooth muscle P2Y11 as a potential target in the management of diabetic vasculopathy.

Essential revisions:

1) There were concerns about whether the glucose responses are affected by possible changes in osmolarity. From the Materials and methods, it does not appear that changes in osmolarity were controlled for, and the effect of glucose may be different in hypertonic and isotonic conditions. A series of controls for all experiments should be included in the manuscript rather than referencing an abstract (Syed et al., 2018).

2) Antibodies need to be validated and quantification described in more detail.a) For assessing Ser1928 phosphorylation, showing full-length blots with only a single molecular weight does not demonstrate specificity. The method of quantifying band densities should be described in more detail (more than just saying ImageJ was used). Were band densities integrated, and how were doublets quantified (e.g., Figures 1, 6, and especially Figure 7C)? In Figure 7C, MW markers are needed; is the upper Cav band of the doublet phosphorylated, and could this be used to estimate the proportion of channels that are phosphorylated? It is difficult to assess the linearity of the antibody without showing on the same blot dephosphorylated channels and fully phosphorylated channels. Does a 30% increase mean 30% of the channels are phosphorylated?

b) Primary antibodies to P2Y11 and PKAcat need to be validated. It appears that P2Y11 ab is not specific for P2Y11 in mouse (extra bands in Figure 2—figure supplement 1A). This is important because the only evidence that P2Y11 is expressed in arterial myocytes comes from cell staining (Figure 2); the Westerns are from lysates also containing endothelial cells which are known to express P2Y11.

3) While ATP might be expected to accumulate in restricted spaces in tissue to activate P2Y11, there is some doubt about whether ATP release could be responsible for the increased Ba2+ currents in patch-clamp experiments where ATP would be free to diffuse. More evidence should be provided to support the role of ATP. For example, does continuous bath perfusion reduce or eliminate the increase in Ba2+ current in response to glucose? Can ATP release be detected by luciferase in the bath (e.g., see Ospichuk and Cahalan, Nature 359:241, 1992)? Finally, please discuss possible mechanisms of ATP secretion and what is known about the affinity of P2Y11 for ATP.

4) To show that phosphorylation of Ser1928 is essential for the regulation of contractility by P2Y11 agonists (and not merely correlated), vasoconstriction measurements (Figure 7) should be repeated with a S1928A knockin mouse. Similarly, the single-channel Po (Figure 8) should also be measured in the knockin animals to counter the possibility that prolonged incubation in glucose may have increased Po by affecting other signaling pathways and gene expression.

5) The specificity of LTCC modulation by glucose is an important issue, and the GSD data as they stand are rather descriptive and do not establish whether P2Y11 exists in complexes with AC, PKA, and LTCC. While it is beyond the scope of this paper to establish the composition of a P2Y11 signaling complex, the existence of such a complex would make the distribution of P2Y11, PKA, and CaV1.2 non-random. To address this point, experimentally observed results should be compared with those of a simulated random distribution of the proteins at the same overall density, and the results shown in a supplementary figure.

6) Figures 4 and 5: The FRET response (cAMP production) is about twice as large in human myocytes than in mouse myocytes (normalized FRET ratio change, ~0.1 vs. ~0.05), despite the fact that the density of P2Y11-CaV1.2 clusters in human myocytes is about 20% that in mouse myocytes (~0.03/µm2 vs. >0.15/µm2) and P2Y11-PKA cluster density is about the same in both humans and mice (~0.06/µm2). Why is cAMP production noticeably higher in humans than in mice, and is this difference significant? Please provide comment in the text.

7) A microdomain containing P2Y11, CaV1.2, PKA and AC5 is discussed in the Discussion, but no data are shown for a role of AC5, only a citation of a FASEB meeting abstract (Discussion, first paragraph; Syed et al., 2018). Ideally, these data should be included in the current manuscript.

8) In the last paragraph of the subsection “The P2Y11 inhibitor NF340 prevents glucose-induced Ser1928 phosphorylation, LTCC activity and vasoconstriction”, the NF546 experiment does not rule out involvement of P2Y1 receptor, it merely shows it is not necessary for the vasoconstriction. To show it does not contribute, the authors should add the inhibitor after glucose to see if it inhibits (Figure 6—figure supplement 1D).

[Editors' note: further revisions were requested prior to acceptance, as described below.]

Thank you for resubmitting your work entitled "A Gs-coupled purinergic receptor boosts Ca2+ influx and vascular excitability during diabetic hyperglycemia" for further consideration at eLife. Your revised article has been favorably evaluated by Richard Aldrich (Senior Editor) and a Reviewing Editor.

The manuscript has been improved and the reviewers' comments have been addressed well with new experimental data and additions to the text. However, several points need to be explained more clearly before the manuscript can be accepted, as outlined below:

Point 1. Please show a bar graph quantifying the pSer Western in Figure 1—figure supplement 1E (like that shown in Figure 1C) to make the point.

Point 2b. Please show two MW markers (55, 35) for the P2Y11 blot in Figure 2—figure supplement 1D since the band is well below 55 kD. This is needed to show expression at the expected MW of 40 kD.

Point 3. Please describe the perfusion experiment in more detail – in particular, the speed of perfusion (ml/min), and the time interval between changes in perfusion rate and collection of data. Please state the EC50 for stimulating P2Y11 with ATP in the text, with a reference.

Point 5. The simulations and analysis need to be explained in the Materials and methods section. How is "overlap" quantified in Figure 3I? It is surprising that almost no overlap is visible in the experimental data in Figure 3—figure supplement 3C ("overlap" panel). Briefly describe how was randomization done and how many times it was repeated. The explanation in the second paragraph of the subsection “P2Y11 receptors are in close proximity to CaV1.2 and PKAcat in arterial myocytes” does not adequately explain how the simulation was done, and what we are looking at in the images in Figure 3—figure supplement 3C.

https://doi.org/10.7554/eLife.42214.073

Author response

Essential revisions:

1) There were concerns about whether the glucose responses are affected by possible changes in osmolarity. From the Materials and methods, it does not appear that changes in osmolarity were controlled for, and the effect of glucose may be different in hypertonic and isotonic conditions. A series of controls for all experiments should be included in the manuscript rather than referencing an abstract (Syed et al., 2018).

To address the concerns regarding possible osmolarity effects, we performed experiments using equimolar concentrations of the nonpermeable mannitol (20 mM mannitol = 10 mM D-glu + 10 mM mannitol). Our new data show that elevated mannitol failed to induce vasoconstriction as well as to potentiate L-type CaV1.2 channel (LTCC) activity, increase phosphorylation of CaV1.2 at serine 1928 (Ser1928) or stimulate cAMP production in arterial myocytes. These results suggest that effects observed in cAMP synthesis, Ser1928 phosphorylation, LTCC activity and vascular reactivity in response to an increase in extracellular glucose (e.g. 20 mM D-glu) are not due to osmolarity effects but are rather a direct consequence of elevated glucose. These data are consistent with previous studies from our group and others (Jackson et al., 2016; Navedo et al., 2010; Nystoriak et al., 2014; Nystoriak et al., 2017; Rainbow et al., 2006). Moreover, the new data are in line with our recent observations indicating that the elevated glucose effects on LTCC activity and vascular reactivity require glucose internalization and metabolic utilization (Nystoriak et al., 2017). These new data and discussion are now included as Figure 1—figure supplement 1B, 1C and 1E (subsection “Glucose-induced extracellular nucleotide release mediates Ser1928 phosphorylation, increased LTCC activity and vasoconstriction”, first, second and third paragraphs), Figure 5—figure supplement 1B and 1C (subsection “Increased sarcolemmal cAMP synthesis in arterial myocytes in response to a P2Y11 agonist recapitulates glucose effects”, last paragraph) and Figure 8A and 8B (subsection “Increased Ser1928 phosphorylation and LTCC activity during chronic diabetic hyperglycemia are prevented by a P2Y11 inhibitor”).

2) Antibodies need to be validated and quantification described in more detail.a) For assessing Ser1928 phosphorylation, showing full-length blots with only a single molecular weight does not demonstrate specificity.

We appreciate this comment and apologize for not addressing the specificity of the antibodies clearly in our original manuscript. The specificity of the phospho-specific antibody for Ser1928 as well as the FP1 antibody for total CaV1.2 measurement has been extensively and thoroughly validated in multiple published studies by our group, including for use in arterial myocytes (Buonarati et al., 2017; Hall et al., 2013; Nystoriak et al., 2017; Patriarchi et al., 2016). Importantly, immunoreactivity at 250 kDa of CaV1.2 with the phospho-specific antibody for Ser1928 (but not a phospho-specific antibody for Ser1700) and FP1 are completely eliminated in tissue from S1928A KI mice (Figure 3 in Patriarchi et al., 2016) and conditional CaV1.2 KO mice (Figure 2 in Buonarati et al., 2017), respectively. Full blots are shown in Buonarati et al., 2017, although the 100 kDa marker had been run to the bottom of the gel for better resolution of the high molecular mass CaV1.2. Blotted membranes were trimmed to save on the use of these antibodies, which are generated in-house and only available in limited quantity. We have modified the original text to clarify this issue (subsection “Glucose-induced extracellular nucleotide release mediates Ser1928 phosphorylation, increased LTCC activity and vasoconstriction”, third paragraph; subsection “P2Y11 receptors are in close proximity to CaV1.2 and PKAcat in arterial myocytes”, second paragraph).

The method of quantifying band densities should be described in more detail (more than just saying ImageJ was used). Were band densities integrated, and how were doublets quantified (e.g., Figures 1, 6, and especially Figure 7C)?

We apologize for this oversight. We have revised the Materials and methods section of our manuscript to include an expanded description on quantification of Western blot data (see subsection “Immunoprecipitation and immunoblotting”).

In Figure 7C, MW markers are needed.

Thank you for pointing this out. We have added MW markers to all our blots in the main and supplementary figures.

Is the upper Cav band of the doublet phosphorylated, and could this be used to estimate the proportion of channels that are phosphorylated?

The upper band represents the full length form of CaV1.2, which harbors Ser1928 (Hell et al., 1993). The lower band represents a shorter CaV1.2 form that is truncated around residue 1800 and thus does not contain Ser1928. Accordingly, only the upper band can be phosphorylated on Ser1928. The shorter channel form still has the proteolytically cleaved C-terminal domain attached to it, but this C-terminal fragment is difficult to detect with the small amounts of material from our arterial samples.

To define basal Ser1928 phosphorylation levels, we previously used CaV1.2 isolated from rat forebrain where we obtain much more material than from our arterial samples. There, we estimated that ~16.7% of the CaV1.2 long form is phosphorylated on Ser1928 under basal conditions in healthy adult rats (Davare and Hell, 2003). However, this is only an upper limit. With our arterial preparation (freshly isolated small resistance arteries from cerebral and mesenteric vessels) and the small amount of working material obtained from it, it could be less than that because the only way we can obtain such an estimate is by comparing the immunoblot signal from CaV1.2 from untreated samples with those where CaV1.2 had been incubated after the immunoprecipitation with purified PKA and Mg-ATP. The estimate assumes that all full-length CaV1.2 is phosphorylated on Ser1928 under these stimulatory conditions, but unfortunately, we cannot be completely certain about this. Hence, we will not be able to determine with high confidence the percentage of total CaV1.2 that is phosphorylated at Ser1928 under basal conditions in our preparation. Such a number would also not necessarily be too helpful because a substantial fraction of CaV1.2 resides inside the cells (Conrad et al., 2018; Ghosh et al., 2018) when only channels at the plasma membrane are relevant for potentiation by Ser1928 phosphorylation. For instance, if we would find that ~16% of CaV1.2 is phosphorylated in arterial lysates under basal conditions, it could quite well reflect that 50% of CaV1.2 at the surface membrane are Ser1928 phosphorylated if the intracellular pool constitutes 70% of total CaV1.2 and is largely not phosphorylated. We will be working in future studies to address this interesting point in a rigorous manner but would like to respectfully note that these data will not change the conclusions of the present study.

It is difficult to assess the linearity of the antibody without showing on the same blot dephosphorylated channels and fully phosphorylated channels.

We show dephosphorylated and phosphorylated channels on the same blot – e.g. total CaV1.2 reprobed after Ser1928 phosphorylation probe. Moreover, our blots are analyzed after films are exposed for increasing time periods (1 min, 2 min, 4 min) to ensure that the signals are in the linear range, as detailed in our previous work (Davare and Hell, 2003; Hall et al., 2006). This information is clarified and expanded in our revised manuscript (subsection “Immunoprecipitation and immunoblotting”).

Does a 30% increase mean 30% of the channels are phosphorylated?

No. It means the percentage of channels that were phosphorylated has increased by 30%. As an example, if 10% of the channels are phosphorylated basally and treatment increases the phosphorylation state by 30%, then ~13% of the channels would be phosphorylated upon treatment.

b) Primary antibodies to P2Y11 and PKAcat need to be validated. It appears that P2Y11 ab is not specific for P2Y11 in mouse (extra bands in Figure 2—figure supplement 1A). This is important because the only evidence that P2Y11 is expressed in arterial myocytes comes from cell staining (Figure 2); the Westerns are from lysates also containing endothelial cells which are known to express P2Y11.

To address this question, we performed additional experiments. The PKAcat Ab was validated using a blocking peptide provided by the vendor (Santa Cruz Biotechnology). Briefly, immunofluorescent labeling of PKAcat in arterial myocytes was performed using a mouse anti-PKAcat Ab (Santa Cruz Biotechnology) incubated with or without a PKAcat blocking peptide (A-2; Santa Cruz Biotechnology) before applying to cells on coverslips. Consistent with a previous study from our group (Nystoriak et al., 2017), arterial myocytes showed broad intracellular PKAcat-associated fluorescence, which was completely lost when the PKAcat Ab was preabsorbed with the PKAcat blocking peptide. These data are now presented in Figure 3—figure supplement 1 (subsection “P2Y11 receptors are in close proximity to CaV1.2 and PKAcat in arterial myocytes”, second paragraph) of the revised manuscript.

As for the P2Y11 Ab validation, we performed multiple tests. First, we detected single protein bands of expected molecular weight in lysates from untransfected and vehicle treated tsA-201 cells, which are known to endogenously express P2Y11 (Dreisig and Kornum, 2016). When tsA-201cells were transfected with a vector encoding a P2Y11 tagged with GFP (P2Y11-GFP), two bands were detected. The small molecular weight band corresponds to the endogenous P2Y11 and the other band is consistent with the addition of GFP to P2Y11 (see Figure 2A). Treatment of tsA-201 cells over-expressing P2Y11-GFP with specific P2Y11 antisense oligodeoxynucleotides (ODNs) resulted in a reduction in the intensity of both bands (Figure 2B). In native arterial tissue, we detected an immunoreactive band of the expected molecular weight for P2Y11 (~40 kDa) with the same P2Y11 Ab (Figure 2C and Figure 2—figure supplement 1A). Importantly, the abundance of this ~40 kDa band was reduced to about ~50% of the basal P2Y11 protein abundance in arteries treated with specific P2Y11 antisense ODNs (Figure 2—figure supplement 1B). A band of ~110-120 kDa was observed in mouse arterial lysates, perhaps reflecting the formation of heterodimers between the P2Y11 receptor itself or other purinergic/ G protein-coupled receptors, as previously reported (Barragan-Iglesias et al., 2015; Barragan-Iglesias et al., 2014; Ecke et al., 2008; Nishimura et al., 2016). This discussion is included in the first paragraph of the subsection “P2Y11 receptors are in close proximity to CaV1.2 and PKAcat in arterial myocytes” We also observed a band of expected molecular weight for P2Y11 in lysates from isolated mouse arterial myocytes (Figure 2—figure supplement 1D), and P2Y11-associated fluorescence was not detected in isolated mouse arterial myocytes exposed to a boiled P2Y11 Ab (Figure 2—figure supplement 1E). In addition, note that the functional expression of human P2Y11 and mouse P2Y11-like receptors in arterial myocytes is further supported by multiple well-controlled functional studies described throughout the manuscript (please see Figures 5-8). Finally, we reached out to several commercial companies that sell P2Y11 Ab to obtain antigenic peptides for P2Y11. They do not have available antigenic peptides for P2Y11 Ab nor were able to provide us with the amino acid sequence used to raise the antibody, thus prohibiting that additional approach to antibody validation. Regardless, we consider that our multiple biochemical and functional approaches provide strong support for validation of the P2Y11 Ab and the presence of human P2Y11 and mouse P2Y11-like receptor in arterial myocytes.

3) While ATP might be expected to accumulate in restricted spaces in tissue to activate P2Y11, there is some doubt about whether ATP release could be responsible for the increased Ba2+ currents in patch-clamp experiments where ATP would be free to diffuse. More evidence should be provided to support the role of ATP. For example, does continuous bath perfusion reduce or eliminate the increase in Ba2+ current in response to glucose? Can ATP release be detected by luciferase in the bath (e.g., see Ospichuk and Cahalan, Nature 359:241, 1992)? Finally, please discuss possible mechanisms of ATP secretion and what is known about the affinity of P2Y11 for ATP.

We recognize that our original cartoon identifying extracellular ATP as a primary mediator of glucose-induced activation of P2Y11 was premature. Although we are now including additional data showing that the well-known non-hydrolyzed ATP analog and P2Y11 agonist ATPγS (von Kugelgen and Harden, 2011) stimulates Ba2+ currents in arterial myocytes (Figure 7—figure supplement 1B; subsection “The P2Y11 agonist NF546 stimulates Ser1928 phosphorylation, PKA-dependent LTCC activity and vasoconstriction”, first paragraph), we have modified the illustration to avoid data overinterpretation.

Note, however, that increased glucose (20 mM D-glucose) has been shown to trigger the release of NUC into the extracellular space of many cells, including arterial myocytes (Costa et al., 2009; Hazama et al., 1998; Nilsson et al., 2006; Parodi et al., 2002). Our results with apyrase support this conclusion (Figure 1). In addition, we now provide additional data to support the concept that glucose-induced extracellular nucleotide release mediates the changes in Ba2+ current in isolated voltage clamped arterial myocytes by performing the proposed experiments with continuous bath perfusion. Accordingly, we were able to record robust Ba2+ current in isolated voltage clamped arterial myocytes under static and flow conditions. However, as predicted by the reviewers, whereas IBa recorded in a static bath was significantly elevated upon increasing external D-glucose from 10 mM to 20 mM, this response was not observed under continuous flow conditions. These results suggest that nucleotides (NUC) could indeed diffuse away under continuous flow conditions, whereas they could accumulate in or around the area of release at the surface membrane of arterial myocytes to induce the optimal activation of the signaling pathway underlying the increased Ba2+ current density in response to 20 mM D-glucose. We did attempt to detect glucose-induced NUC release (e.g. ATP) using a commercially available luciferin-luciferase kit. Unfortunately, we could not detect any signal. After consulting with the manufacturer, this was attributed to a need for optimizing the preparation to detect NUC release from native arterial myocytes. This is not a trivial task and may take more time than that allotted for resubmission of the manuscript. We are currently working to optimize our preparation/assay and rigorously address this point in a comprehensive follow up study. However, new data provide further support to our hypothesis that elevated glucose evokes LTCC activity and vasoconstriction through an autocrine mechanism involving secreted nucleotides, which are well-known activators of G-coupled P2Y receptors (von Kugelgen and Harden, 2011). These new data and discussion are now included as Figure 1—figure supplement 3 (subsection “Glucose-induced extracellular nucleotide release mediates Ser1928 phosphorylation, increased LTCC activity and vasoconstriction”).

As suggested, we provide a discussion of possible mechanisms of NUC (including ATP) secretion and P2Y11 pharmacological profile (see Discussion, sixth paragraph).

4) To show that phosphorylation of Ser1928 is essential for the regulation of contractility by P2Y11 agonists (and not merely correlated), vasoconstriction measurements (Figure 7) should be repeated with a S1928A knockin mouse. Similarly, the single-channel Po (Figure 8) should also be measured in the knockin animals to counter the possibility that prolonged incubation in glucose may have increased Po by affecting other signaling pathways and gene expression.

To address this question, we performed additional experiments. First, we found that arteries from S1928A mice failed to constrict in response to the P2Y11 agonist NF546. These results are in stark contrast to those observed for wild type arteries (see Figure 7E and 7F), and further support the conclusion that phosphorylation of Ser1928 is essential for the regulation of vascular reactivity in response to a P2Y11 agonist. These new data and discussion are now included as Figure 7F in the revised manuscript (subsection “The P2Y11 agonist NF546 stimulates Ser1928 phosphorylation, PKA-dependent LTCC activity and vasoconstriction”, last paragraph).

Second, we performed the suggested single-channel experiments in arterial myocytes from S1928A mice under chronically elevated glucose stimulation (48 hours). Results indicate that S1928A cells isolated from either 10 mM and 20 mM D-glucose groups had similar LTCC open probability (nPo), which are comparable to that observed for LTCCs in wild type cells isolated from the group treated with 10 mM D-glucose. These results counter the possibility that prolonged incubation in elevated glucose may have increased LTCC nPo by affecting other signaling pathways and gene expression, and further support a key role for Ser1928 phosphorylation in regulation of LTCC activity during diabetic hyperglycemia. These new data and discussion are now included as Figure 8—figure supplement 1B (subsection “Increased Ser1928 phosphorylation and LTCC activity during chronic diabetic hyperglycemia are prevented by a P2Y11 inhibitor”) in the revised manuscript.

5) The specificity of LTCC modulation by glucose is an important issue, and the GSD data as they stand are rather descriptive and do not establish whether P2Y11 exists in complexes with AC, PKA, and LTCC. While it is beyond the scope of this paper to establish the composition of a P2Y11 signaling complex, the existence of such a complex would make the distribution of P2Y11, PKA, and CaV1.2 non-random. To address this point, experimentally observed results should be compared with those of a simulated random distribution of the proteins at the same overall density, and the results shown in a supplementary figure.

To address this point, we performed the suggested comparison between experimental and simulated random protein distribution. Accordingly, we compared the frequency of P2Y11-CaV1.2 and P2Y11-PKA overlap that was obtained experimentally with that from simulated images that were randomized for proteins of interest at the same overall density as those observed in the experimental super-resolution maps (Bolte and Cordelieres, 2006). The analysis revealed a higher level of overlap between P2Y11-CaV1.2 and P2Y11-PKA in the experimental data than that observed for a simulated random distribution. These results suggest a selective and distinctive interaction between a subpopulation of P2Y11 and a pool of CaV1.2 and PKA, which complements our super-resolution data in the first submission. These new data and discussion are now included as Figure 3I and Figure 3—figure supplement 3C (subsection “P2Y11 receptors are in close proximity to CaV1.2 and PKAcat in arterial myocytes”, second paragraph) in the revised manuscript.

6) Figures 4 and 5: The FRET response (cAMP production) is about twice as large in human myocytes than in mouse myocytes (normalized FRET ratio change, ~0.1 vs. ~0.05), despite the fact that the density of P2Y11-CaV1.2 clusters in human myocytes is about 20% that in mouse myocytes (~0.03/µm2 vs. >0.15/µm2) and P2Y11-PKA cluster density is about the same in both humans and mice (~0.06/µm2). Why is cAMP production noticeably higher in humans than in mice, and is this difference significant? Please provide comment in the text.

To answer this question, we first respectfully submit that it would not be appropriate at this time to correlate PLA data with the FRET responses as many factors may independently influence each measurement. The PLA data indicate relative abundance of the protein complex whereas the FRET data indicate relative signaling intensity after stimulation. There is not a necessary correlation between these two results. Accordingly, the goal of our PLA experiments is to define close association between P2Y11-CaV1.2 and P2Y11-PKA, whereas the goal of the FRET experiments is to determine whether glucose could modulate cAMP signaling as well as define potential mechanisms such as the involvement of the P2Y11. We strongly believe that our data have accomplished these goals.

The cAMP production in response to glucose and the P2Y11 agonist NF546 is indeed significantly larger in human arterial myocytes compared to mouse cells (P < 0.05). The reasons for this are currently unclear but may include intrinsic differences associated with species (human vs. mouse), blood vessels (human adipose arteries vs mouse aortae) and expression patterns of the biosensor in human vs. mouse cells, as well as variations in protein expression and receptor/enzyme activity (e.g. P2Y11, AC and PDE expression/activity) in human vs. mouse arterial myocytes. Whether either or all of the factors described above contribute to distinctive cAMP responses in human vs mouse arterial myocytes as well as their physiological implications, remain to be explored. Nevertheless, as suggested, we now provide a description of these issues in the revised manuscript (subsection “Increased sarcolemmal cAMP synthesis in arterial myocytes in response to a P2Y11 agonist recapitulates glucose effects”, last paragraph).

7) A microdomain containing P2Y11, CaV1.2, PKA and AC5 is discussed in the Discussion, but no data are shown for a role of AC5, only a citation of a FASEB meeting abstract (Discussion, first paragraph; Syed et al., 2018). Ideally, these data should be included in the current manuscript.

We recognize that our discussion of AC5 in complex withP2Y11, CaV1.2 and PKA is premature. We have therefore modified the text as reference to AC5 does not directly relate to the focus of this manuscript. We are working to rigorously and comprehensively examine the role of AC5 in a follow up manuscript.

8) In the last paragraph of the subsection “The P2Y11 inhibitor NF340 prevents glucose-induced Ser1928 phosphorylation, LTCC activity and vasoconstriction”, the NF546 experiment does not rule out involvement of P2Y1 receptor, it merely shows it is not necessary for the vasoconstriction. To show it does not contribute, the authors should add the inhibitor after glucose to see if it inhibits (Figure 6—figure supplement 1D).

To address this question, we performed additional experiments. We now show that glucose-mediated vasoconstriction is not affected by adding the P2Y1 inhibitor MRS2179 either before or after 20 mM D-glucose. These new data and discussion are now included as Figure 6—figure supplement 1E (subsection “The P2Y11 inhibitor NF340 prevents glucose-induced Ser1928 phosphorylation, LTCC activity and vasoconstriction”, last paragraph) in the revised manuscript.

[Editors' note: further revisions were requested prior to acceptance, as described below.]

The manuscript has been improved and the reviewers' comments have been addressed well with new experimental data and additions to the text. However, several points need to be explained more clearly before the manuscript can be accepted, as outlined below:

Point 1. Please show a bar graph quantifying the pSer Western in Figure 1—figure supplement 1E (like that shown in Figure 1C) to make the point.

A bar plot quantifying pSer1928 in the presence of 10 mM D-glucose and 20 mM mannitol is now included in the revised version of the manuscript (Figure 1—figure supplement 1E).

Point 2b. Please show two MW markers (55, 35) for the P2Y11 blot in Figure 2—figure supplement 1D since the band is well below 55 kD. This is needed to show expression at the expected MW of 40 kD.

As requested, we now show two MW marker for the WB blot in Figure 2—figure supplement 2D.

Point 3. Please describe the perfusion experiment in more detail – in particular, the speed of perfusion (ml/min), and the time interval between changes in perfusion rate and collection of data. Please state the EC50 for stimulating P2Y11 with ATP in the text, with a reference.

We have expanded our description of the perfusion experiments and have provided additional information about time intervals between perfusion rate and collection of data in the Materials and methods section and in the figure legend of the revised manuscript (subsection “Electrophysiology”, first paragraph and Figure 1—figure supplement 2 legend). We also stated the EC50 for stimulating P2Y11 with ATP and have provided citations for this in the revised manuscript (subsection “The P2Y11 agonist NF546 stimulates Ser1928 phosphorylation, PKA-dependent LTCC activity and vasoconstriction”, first paragraph, and Discussion, sixth paragraph).

Point 5. The simulations and analysis need to be explained in the Materials and methods section. How is "overlap" quantified in Figure 3I? It is surprising that almost no overlap is visible in the experimental data in Figure 3—figure supplement 3C ("overlap" panel). Briefly describe how was randomization done and how many times it was repeated. The explanation in the second paragraph of the subsection “P2Y11 receptors are in close proximity to CaV1.2 and PKAcat in arterial myocytes” does not adequately explain how the simulation was done, and what we are looking at in the images in Figure 3—figure supplement 3C.

To address these points, we have expanded the description of the generation of simulation/randomization data as well as the overall analysis in the revised manuscript. Randomization of P2Y11-CaV1.2 and P2Y11-PKAcat was repeated six times and then compared against experimental data obtained from the super-resolution localization maps. The percentage of P2Y11 overlap was determined by multiplying thresholded, binary localization maps (experimental and simulated data) of P2Y11by those from CaV1.2 or PKAcat. We apologize and acknowledge that the quality of original images in Figure 3—figure supplement 3C was not adequate, thus preventing proper evaluation of the data. We have modified the presentation of these images by changing the LUT to an inverted grayscale LUT that makes the overlapping signals more apparent and therefore, better reflects the results from the amalgamated data. Changes are found in subsection “P2Y11 receptors are in close proximity to CaV1.2 and PKAcat in arterial myocytes”, second paragraph, subsection “Immunolabeling and Ground State Depletion (GSD) microscopy”, last paragraph, and in the Figure 3—figure supplement 3 legend.

https://doi.org/10.7554/eLife.42214.074

Article and author information

Author details

  1. Maria Paz Prada

    Department of Pharmacology, University of California, Davis, Davis, United States
    Contribution
    Conceptualization, Data curation, Formal analysis, Validation, Investigation, Visualization, Methodology, Writing—orignal draft, Writing—review and editing
    Contributed equally with
    Arsalan U Syed and Olivia R Buonarati
    Competing interests
    No competing interests declared
  2. Arsalan U Syed

    Department of Pharmacology, University of California, Davis, Davis, United States
    Contribution
    Formal analysis, Validation, Investigation, Methodology, Writing—review and editing
    Contributed equally with
    Maria Paz Prada and Olivia R Buonarati
    Competing interests
    No competing interests declared
  3. Olivia R Buonarati

    Department of Pharmacology, University of California, Davis, Davis, United States
    Contribution
    Formal analysis, Validation, Investigation, Methodology, Writing—review and editing
    Contributed equally with
    Maria Paz Prada and Arsalan U Syed
    Competing interests
    No competing interests declared
  4. Gopireddy R Reddy

    Department of Pharmacology, University of California, Davis, Davis, United States
    Contribution
    Formal analysis, Validation, Investigation, Writing—review and editing
    Competing interests
    No competing interests declared
  5. Matthew A Nystoriak

    Diabetes & Obesity Center, Department of Medicine, University of Louisville, Kentucky, United States
    Contribution
    Formal analysis, Investigation, Writing—review and editing
    Competing interests
    No competing interests declared
  6. Debapriya Ghosh

    Department of Pharmacology, University of California, Davis, Davis, United States
    Contribution
    Formal analysis, Validation, Investigation, Methodology, Writing—review and editing
    Competing interests
    No competing interests declared
  7. Sergi Simó

    Department of Cell Biology & Human Anatomy, University of California, Davis, Davis, United States
    Contribution
    Formal analysis, Investigation, Methodology, Writing—review and editing
    Competing interests
    No competing interests declared
  8. Daisuke Sato

    Department of Pharmacology, University of California, Davis, Davis, United States
    Contribution
    Formal analysis, Investigation, Methodology, Writing—review and editing
    Competing interests
    No competing interests declared
  9. Kent C Sasse

    Sasse Surgical Associates, Reno, United States
    Contribution
    Resources, Writing—review and editing, Revising the article for important clinical and intellectual content
  10. Sean M Ward

    Department of Physiology & Cell Biology, University of Nevada, Reno, United States
    Contribution
    Resources, Writing—review and editing
    Competing interests
    No competing interests declared
  11. Luis F Santana

    Department of Physiology & Membrane Biology, University of California, Davis, Davis, United States
    Contribution
    Resources, Writing—review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-4297-8029
  12. Yang K Xiang

    1. Department of Pharmacology, University of California, Davis, Davis, United States
    2. VA Northern California Healthcare System, Mather, United States
    Contribution
    Resources, Formal analysis, Writing—review and editing
    Competing interests
    No competing interests declared
  13. Johannes W Hell

    Department of Pharmacology, University of California, Davis, Davis, United States
    Contribution
    Resources, Formal analysis, Supervision, Writing—review and editing
    Competing interests
    No competing interests declared
  14. Madeline Nieves-Cintrón

    Department of Pharmacology, University of California, Davis, Davis, United States
    Contribution
    Conceptualization, Resources, Data curation, Formal analysis, Funding acquisition, Supervision, Investigation, Visualization, Methodology, Writing—original draft, Writing—review and editing
    For correspondence
    mcnieves@ucdavis.edu
    Competing interests
    No competing interests declared
  15. Manuel F Navedo

    Department of Pharmacology, University of California, Davis, Davis, United States
    Contribution
    Conceptualization, Resources, Data curation, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing—original draft, Project administration, Writing—review and editing
    For correspondence
    mfnavedo@ucdavis.edu
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0001-6864-6594

Funding

National Heart, Lung, and Blood Institute (R01HL098200)

  • Manuel F Navedo

National Heart, Lung, and Blood Institute (R01HL121059)

  • Manuel F Navedo

National Institute of Neurological Disorders and Stroke (R01NS078792)

  • Johannes W Hell

National Institute on Aging (R01AG055357)

  • Johannes W Hell

National Institute of General Medical Sciences (T32GM099608)

  • Maria Paz Prada

National Heart, Lung, and Blood Institute (T32HL086350)

  • Arsalan U Syed
  • Matthew A Nystoriak

National Institute of Diabetes and Digestive and Kidney Diseases (R01DK57236)

  • Sean M Ward

National Institute of Neurological Disorders and Stroke (F31NS086226)

  • Olivia R Buonarati

American Heart Association (16SDG27260070)

  • Matthew A Nystoriak

American Heart Association (18POST34060234)

  • Debapriya Ghosh

National Heart, Lung, and Blood Institute (R01HL127764)

  • Yang K Xiang

National Heart, Lung, and Blood Institute (R01HL112413)

  • Yang K Xiang

University of California, Davis (Academic Federation Innovative Development Award)

  • Madeline Nieves-Cintrón

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

We thank members of the Navedo Lab for technical support. We also thank Dr. Charles Kennedy from the University of Strathclyde for helpful discussions and advice on purinergic signaling in the vasculature.

Ethics

Human subjects: This statement is already in our Materials and Methods section - Samples were obtained after Institutional Review Board (IRB) approval from the University of Nevada Reno School of Medicine (IRB ID: 2013-019) and in accordance with the guidelines of the Declaration of Helsinki. The need for informed consent was waived by IRBs at the University of Nevada Reno School of Medicine (IRB ID: 2013-019) and the University of California Davis (IRB ID: 597267-1) because the tissue is considered "waste", has no codification that could be used to identify patients and was determined not to be human subject research in accordance with United States of America federal regulations, as defined by 45 CFR 46.102(f). This precludes the acquisition of detailed clinical profiles other than sex, age and whether the patient nondiabetic or diabetic. Only samples from nondiabetic patients were used.

Animal experimentation: This statement is already in our Materials and Methods section -Male wild-type C57BL/6J (wt) mice of 5 to 8 weeks of age were euthanized with a lethal dose of sodium pentobarbital (250 mg/kg; intraperitoneally), as approved by the University of California, Davis Animal Care and Use Committee (protocol #: 20321).

Senior Editor

  1. Richard Aldrich, The University of Texas at Austin, United States

Reviewing Editor

  1. Richard S Lewis, Stanford University School of Medicine, United States

Reviewers

  1. Richard S Lewis, Stanford University School of Medicine, United States
  2. Mark T Nelson, University of Vermont, United States

Publication history

  1. Received: September 21, 2018
  2. Accepted: February 16, 2019
  3. Version of Record published: March 1, 2019 (version 1)

Copyright

© 2019, Prada et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 841
    Page views
  • 96
    Downloads
  • 4
    Citations

Article citation count generated by polling the highest count across the following sources: Crossref, Scopus, PubMed Central.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Download citations (links to download the citations from this article in formats compatible with various reference manager tools)

Open citations (links to open the citations from this article in various online reference manager services)