Increased β-adrenergic stimulation augments vascular smooth muscle cell calcification via PKA/CREB signalling

In chronic kidney disease (CKD), hyperphosphatemia promotes medial vascular calcification, a process augmented by osteogenic transdifferentiation of vascular smooth muscle cells (VSMCs). VSMC function is regulated by sympathetic innervation, and these cells express α- and β-adrenergic receptors. The present study explored the effects of β2-adrenergic stimulation by isoproterenol on VSMC calcification. Experiments were performed in primary human aortic VSMCs treated with isoproterenol during control or high phosphate conditions. As a result, isoproterenol dose dependently up-regulated the expression of osteogenic markers core-binding factor α-1 (CBFA1) and tissue-nonspecific alkaline phosphatase (ALPL) in VSMCs. Furthermore, prolonged isoproterenol exposure augmented phosphate-induced calcification of VSMCs. Isoproterenol increased the activation of PKA and CREB, while knockdown of the PKA catalytic subunit α (PRKACA) or of CREB1 genes was able to suppress the pro-calcific effects of isoproterenol in VSMCs. β2-adrenergic receptor silencing or inhibition with the selective antagonist ICI 118,551 blocked isoproterenol-induced osteogenic signalling in VSMCs. The present observations imply a pro-calcific effect of β2-adrenergic overstimulation in VSMCs, which is mediated, at least partly, by PKA/CREB signalling. These observations may support a link between sympathetic overactivity in CKD and vascular calcification.

Therefore, the present study explores a putative modulating role of β2-adrenergic activation by isoproterenol during phosphate-induced osteogenic transdifferentiation and calcification of VSMCs in vitro.

RNA isolation and RT-PCR
Total RNA was isolated from HAoSMCs by using Trizol reagent (Fisher Scientific). cDNA was synthesized with SuperScript III Reverse Transcriptase and oligo(dT) [12][13][14][15][16][17][18] primers (Fisher Scientific). RT-PCR was performed in duplicate with iQ™ Sybr Green Supermix (Bio-Rad Laboratories) and CFX96 Real-Time PCR Detection System (Bio-Rad Laboratories). The specificity of the PCR products was confirmed by analysis of the melting curves. Relative mRNA expression was calculated by the 2 −ΔΔCt method using GAPDH as housekeeping gene, normalized to the control group. The following human primers were used (Fisher Scientific) [1,28,56,76]:

ALP activity assay
ALP activity in HAoSMCs was determined by using a colorimetric assay kit (Abcam). Results are shown normalized to total protein concentration measured by the Bradford assay (Bio-Rad Laboratories) [69] and to the control group.

Calcium content analysis
HAoSMCs were decalcified in 0.6 M HCl at 4 °C overnight, and calcium content in the supernatant was determined by using the QuantiChrom Calcium assay kit (BioAssay Systems). HAoSMCs were lysed with 0.1 M NaOH/0.1% SDS, and protein concentration was measured by the Bradford assay (Bio-Rad Laboratories) [14,44,46,64]. Results are shown normalized to total protein concentration [1,28,56] and to the control group.

Statistics
Data are shown as scatter dot plots and arithmetic means ± SEM. Normality was tested with Shapiro-Wilk test. Non-normal datasets were transformed (log, reciprocal, or sqrt) prior to statistical testing to provide normality. Statistical testing was performed by one-way ANOVA followed by Tukey (homoscedastic data) or Games-Howell (heteroscedastic data) post hoc test and by the Steel-Dwass method (non-normal data). Two groups were compared by unpaired two-tailed t-test or Mann-Whitney U-test. p < 0.05 was considered statistically significant.

Results
A first series of experiments investigated the effects of chronic β-adrenergic stimulation on markers of osteogenic transdifferentiation of VSMCs. To this end, HAoSMCs were treated with increasing concentrations of isoproterenol, a nonspecific β-adrenergic receptor agonist, and the expression of osteogenic transcription factor CBFA1 and osteogenic enzyme ALPL was determined. As shown in Fig. 1a and b, isoproterenol treatment up-regulated CBFA1 and ALPL mRNA expression in HAoSMCs in a concentrationdependent manner, effects reaching statistical significance at 1 μM isoproterenol concentration. Accordingly, prolonged isoproterenol treatment significantly increased ALPL protein abundance in HAoSMCs (Fig. 1c). In contrast, exposure of HAoSMCs to isoproterenol did not significantly modify macrophage markers CD68 and LGALS3 or smooth muscle markers ACTA2 and TAGLN mRNA expression ( Fig. 1d-g).
Next experiments explored the effects of isoproterenol on osteogenic markers CBFA1 and ALPL expression as well as calcification of HAoSMCs during elevated phosphate conditions. Isoproterenol treatment up-regulated osteogenic markers mRNA expression in HAoSMCs to similarly high levels as following exposure to the phosphate donor β-glycerophosphate ( Fig. 2a, b). Additional treatment with isoproterenol tended to enhance β-glycerophosphate-induced CBFA1 mRNA expression, a difference, however, not reaching statistical significance (p = 0.057; Fig. 2a), and significantly augmented β-glycerophosphate-induced ALPL mRNA expression in HAoSMCs (Fig. 2b). Similarly, isoproterenol significantly increased ALP activity in HAoSMCs during control as well as elevated phosphate conditions (Fig. 2c). As shown by quantification of calcium content, isoproterenol did not significantly modify calcification of HAoSMCs during control conditions, but significantly aggravated mineralization induced by a calcification medium (Fig. 2d). These effects were confirmed by Alizarin red staining (Fig. 2e) and calcium NIR fluorescent imaging (Fig. 2f) showing an enhancement of calcification induced by calcification medium in the presence of isoproterenol. Taken together, β-adrenergic stimulation with isoproterenol aggravated phosphate-induced HAoSMC osteogenic marker expression and calcification in vitro.
To elucidate the underlying mechanisms of the pro-calcific effects of isoproterenol, the possible involvement of protein kinase A (PKA)/cAMP-responsive element-binding protein (CREB) signalling was investigated. As shown by Western blotting, isoproterenol significantly up-regulated the phosphorylation at RRXS*/T* substrate motif for PKA (where R refers to arginine, X refers to any amino acid, and S and T represent serine and threonine residue, respectively) of total cellular proteins and, thus, PKA activity in HAoSMCs following 5 min of treatment, levels remaining significantly higher up to 1 h of treatment (Fig. 3a). Similarly, isoproterenol increased the activation of CREB in HAoSMCs, as shown by increased phosphorylation at Ser 133 , following 5 min and up to 30 min of treatment without significantly affecting total CREB protein abundance (Fig. 3b).
Furthermore, suppressing the endogenous expression in HAoSMCs of the PKA catalytic subunit α by silencing of the PRKACA gene using small interfering RNA (siRNA) (Fig. 4a) was able to significantly blunt isoproterenolinduced CBFA1 and ALPL mRNA expression (Fig. 4b, c). In addition, knockdown of the CREB1 gene by siRNA (Fig. 4d) significantly suppressed isoproterenol-induced osteogenic markers mRNA expression in HAoSMCs (Fig. 4e, f). PRKACA (Fig. 4a), or CREB1 (Fig. 4d) mRNA expression was not significantly affected by isoproterenol treatment. More importantly, silencing of the PRKACA or CREB1 genes significantly reduced calcification of HAoSMCs induced by isoproterenol together with calcification medium (Fig. 4g). Thus, PKA and CREB are activated by isoproterenol and participate in isoproterenol-induced osteogenic signalling and augmentation of calcification in HAoSMCs.
To identify the β-adrenergic receptor mediating the osteoinductive effects of isoproterenol in HAoSMCs, the mRNA levels of the β-adrenergic receptor subtypes were determined. In accordance with the previous reports [57], the β2-adrenergic receptor, encoded by the ADRB2 gene, had the highest relative expression in HAoSMCs (Fig. 5a). Next, the endogenous expression of the β2-adrenergic receptor in HAoSMCs was suppressed by silencing of the ADRB2 gene using siRNA followed by additional treatment without or with isoproterenol. As a result, ADRB2 mRNA expression was significantly reduced in ADRB2 siRNA-transfected HAoSMCs as compared to negative control siRNA-transfected HAoSMCs, levels not significantly affected by additional isoproterenol treatment (Fig. 5b). Moreover, isoproterenol significantly up-regulated osteogenic markers mRNA expression in control-transfected HAoSMCs, effects significantly suppressed by ADRB2 silencing (Fig. 5c, d). Furthermore, additional treatment with the highly selective β2-adrenergic receptor antagonist ICI 118,551 significantly blunted isoproterenol-induced osteogenic markers mRNA expression (Fig. 5e, f) and significantly reduced calcification of HAoSMCs induced by isoproterenol together with calcification medium (Fig. 5g). Thus, the β2-adrenergic receptor mediated, at least partly, the pro-calcific effects of isoproterenol in HAoSMCs.

Discussion
The present study identifies a promoting effect of chronic β-adrenergic stimulation by isoproterenol through the β2-adrenergic receptor on osteogenic marker expression and calcification of VSMCs. These observations seem surprising, since β2-adrenergic activation ameliorates calcification in valvular cells [53]. Furthermore, acute β2-adrenergic receptor activation induces vasodilation [11], an effect that may be impaired by aging or atherosclerosis [36,57]. In contrast, chronic isoproterenol exposure augments vasoconstrictor responses and induces vascular dysfunction [18,19,68]. Injections of high doses of adrenaline induce microcrystalline calcifications in aortic mitochondria of rabbits [7]. Also, repeated isoproterenol injections in rabbits induce oxidative stress and DNA damage in cerebral artery VSMCs [39]. Although further studies on the vascular effects of β-adrenergic activation are required, β-adrenergic overstimulation has been discussed as a putative factor in sympathetic overactivity [68]. Isoproterenol treatment of VSMCs promotes the expression of osteogenic markers CBFA1 and ALPL. These effects appear to be mediated through the β2-adrenergic receptor and are blocked by a selective antagonist. The pro-calcifying effects of isoproterenol can also be abrogated by silencing of the β2-adrenergic receptor, which did not abolish, but only reduce ADRB2 expression. Thus, it is tempting to speculate that not physiological activation but unphysiological overactivation of the downstream signalling pathways may be responsible for the pro-calcific effects of the β2-adrenergic receptor. These downstream effects of β2-adrenergic stimulation on osteogenic marker expression involve PKA and CREB. Although β2-adrenergic signalling events in VSMC are complex, sustained PKA activation has been identified after isoproterenol exposure [31]. Activation of PKA by TNF-α induces VSMC calcification, suggesting a detrimental role of chronic PKA activation in vascular calcification [32,33,65]. PKA has also been implied in the osteogenic effects of uremic serum on VSMC calcification [9]. Accordingly, PKA inhibition attenuates osteogenic transdifferentiation and calcification of VSMCs [37,65] or aortic rings [67]. However, the role of PKA may be more complex, since also anti-calcific effects of PKA due to inhibition of endoplasmic reticulum stress are described [8].
A downstream target of PKA, activated by Ser 133 phosphorylation, is the transcription factor CREB [48]. CREB has been shown to activate CBFA1/RUNX2 [79], and activation of the PKA/CREB pathway plays an important role in osteogenic differentiation of bone marrow stromal cells [59] and mesenchymal stem cells [80]. CREB has also been implied as a factor downstream of PKApromoting vascular calcification [6,65]. Furthermore, CREB mediates the pro-calcific effects of low potassium conditions in VSMCs [61]. CREB may also be involved in the pro-calcific effects of transforming growth factor β1 [27]. The present observations show a transient Knockdown or blockade of the β2-adrenergic receptor blunts the pro-calcific effects of isoproterenol in HAoSMCs. a Scatter dot plots and arithmetic means ± SEM (n = 4; arbitrary units, a.u.) of β-adrenergic receptors ADRB1, ADRB2, and ADRB3 relative mRNA expression in HAoSMCs. b Scatter dot plots and arithmetic means ± SEM (n = 6; a.u.) of ADRB2 relative mRNA expression in HAoSMCs following transfection with negative control (Negsi) or ADRB2 (ADRB2si) siRNA and treatment for 24 h with control or 1 μM isoproterenol (ISO). *(p < 0.05) statistically significant vs Negsi-transfected HAoSMCs. c, d Scatter dot plots and arithmetic means ± SEM (n = 6; a.u.) of CBFA1 (c) and ALPL (d) relative mRNA expression in HAoSMCs following transfection with negative control (Negsi) or ADRB2 (ADRB2si) siRNA and treatment for 24 h with control or 1 μM isoproterenol (ISO). *(p < 0.05), ***(p < 0.001) statistically significant vs Negsi-transfected HAoSMCs; †(p < 0.05), † † †(p < 0.001) statistically significant vs Negsi-transfected ISOtreated HAoSMCs. e, f Scatter dot plots and arithmetic means ± SEM (n = 10; a.u.) of CBFA1 (e) and ALPL (f) relative mRNA expression in HAoSMCs following treatment for 24 h with control (CTR) or 1 μM isoproterenol (ISO) without and with 1 μM β2-adrenergic receptor antagonist ICI 118,551 (ICI). g Scatter dot plots and arithmetic means ± SEM (n = 8; a.u.) of normalized calcium content in HAoSMCs following treatment for 11 days with control (CTR) or 1 μM isoproterenol (ISO) and calcification medium (Calc.; 10 mM β-glycerophosphate + 1.5 mM CaCl 2 ) without and with 1 μM β2-adrenergic receptor antagonist ICI 118,551 (ICI). *(p < 0.05), **(p < 0.01), ***(p < 0.001) statistically significant vs control HAoSMCs; † †(p < 0.01), statistically significant vs ISO-/Calc. + ISOtreated HAoSMCs 1 3 activation of PKA/CREB after isoproterenol exposure. The subsequent downstream effects of this CREB activation to induce a pro-calcific phenotype after isoproterenol exposure are currently unclear. CREB may modulate autophagy, which inherits a complex and important role in VSMC calcification [61]. CREB is further involved in pro-inflammatory signalling pathways in VSMCs [43]. After isoproterenol exposure, no alterations of the macrophage markers LGALS3 and CD68 or smooth muscle markers ACTA2 and TAGLN are detectable. Nonetheless, the PKA/CREB pathway seems to be required for the full pro-calcific effects of isoproterenol, since silencing of CREB or PKA abrogates these effects. These observations cannot rule out other putative mechanisms of isoproterenol exposure on VSMC calcification. β-adrenergic stimulation may induce intracellular signalling leading to RANKL release [35], ERK1/2 activation [58], PI3K pathway activation [10], oxidative stress [54], inflammation [18,30,58], or apoptosis [30,54], all known as regulators of VSMC osteogenic transdifferentiation and vascular calcification [51,72,73]. The effects of β2-adrenergic stimulation on VSMC calcification may be a relevant aspect in conditions of sympathetic hyperactivity, such as CKD [25,40,55], but more complex mechanisms may play a significant role. The current study is clearly limited by artificial VSMC culture conditions with isoproterenol administration, a condition different from the adrenergic system in the vasculature in vivo [47]. Isoproterenol treatment in cell culture would not mimic a rhythmicity of the sympathetic nervous system and its co-transmitters [47]. β2-adrenergic receptors are involved in the circadian oscillations of vascular adhesion molecules after TNF-α stimulation [20]. In endothelial cells, isoproterenol promotes nitric oxide (NO) release, an important inhibitor of vascular calcification [24,73]. Mice deficient for the β2-adrenergic receptor exhibit reduced aortic NO production [17]. Overexpression of the β2-adrenergic receptor in endothelial progenitor cells improves vascular repair after vascular injury [38]. However, overstimulation with isoproterenol induces endothelial synthase uncoupling [68] and up-regulates expression of inflammatory cytokines in endothelial cells [50]. A complex immunomodulatory role of the β2-adrenergic receptor has been discussed [77]. Clearly, caution is warranted when interpreting the current results, especially for mechanisms of vascular calcification in human patients.
In conclusion, β-adrenergic overstimulation by isoproterenol aggravates phosphate-induced VSMC calcification in vitro, effects mediated, at least in part, by the β2-adrenergic receptor and involving PKA/CREB signalling activation. Further studies are required to determine a possible association of adrenergic stimulation and vascular calcification.