Phospholamban Is Present in Endothelial Cells and Modulates Endothelium-Dependent Relaxation
Evidence From Phospholamban Gene-Ablated Mice
Abstract—Vascular endothelial cells regulate vascular smooth muscle tone through Ca2+-dependent production and release of vasoactive molecules. Phospholamban (PLB) is a 24- to 27-kDa phosphoprotein that modulates activity of the sarco(endo)plasmic reticulum Ca2+ ATPase (SERCA). Expression of PLB is reportedly limited to cardiac, slow-twitch skeletal and smooth muscle in which PLB is an important regulator of [Ca2+]i and contractility in these muscles. In the present study, we report the existence of PLB in the vascular endothelium, a nonmuscle tissue, and provide functional data on PLB regulation of vascular contractility through its actions in the endothelium. Endothelium-dependent relaxation to acetylcholine was attenuated in aorta of PLB-deficient (PLB-KO) mice compared with wild-type (WT) controls. This effect was not due to actions of nitric oxide on the smooth muscle, because sodium nitroprusside–mediated relaxation in either denuded or endothelium-intact aortas was unaffected by PLB ablation. Relative to denuded vessels, relaxation to forskolin was enhanced in WT endothelium-intact aortas. The endothelium-dependent component of this relaxation was attenuated in PLB-KO aortas. To investigate whether these changes were due to PLB, WT mouse aorta endothelial cells were isolated. Both reverse transcriptase–polymerase chain reaction and Western blot analyses revealed the presence of PLB in endothelial cells, which were shown to be >98% pure by diI-acetylated LDL uptake and nuclear counterstaining. These data indicate that PLB is present and modulates vascular function as a result of its actions in endothelial cells. The presence of PLB in endothelial cells opens new fields for investigation of Ca2+ regulatory pathways in nonmuscle cells and for modulation of endothelial-vascular interactions.
Vascular endothelial cells regulate tone by releasing vasorelaxing factors such as endothelium-derived nitric oxide (NO) or prostacyclin and endothelium-derived constricting factors such as endothelin. [Ca2+]i levels modulate the production and release of these vasoactive factors.1 In general, agonist-mediated increases in [Ca2+]i occur in a biphasic manner. An initial transient increase in [Ca2+]i reflects inositol 1,4,5-trisphosphate–mediated release from intracellular stores.2 3 This fractional Ca2+ release from the endoplasmic reticulum activates an influx of [Ca2+]o.4 Restoration of [Ca2+]i occurs via sequestration into intracellular stores and extrusion through the plasma membrane.
Phospholamban (PLB) is a 24- to 27-kDa phosphoprotein that modulates the activity of the sarco(endo)plasmic reticulum Ca2+ ATPase (SERCA). PLB has 3 phosphorylation sites that are critical for the regulation of Ca2+ uptake into intracellular stores. In its unphosphorylated form, PLB inhibits SERCA uptake of Ca2+. Phosphorylation of PLB relieves this inhibition by increasing the affinity of SERCA for Ca2+. Using a mouse model in which the PLB gene has been ablated (PLB-KO), PLB was shown to be an important modulator of cardiac muscle contractility (for review, see Reference 55 ) and the inotropic response of the heart to β-adrenergic stimulation.6 In addition, contractility in slow-twitch skeletal7 and vascular smooth muscle8 was markedly affected by PLB gene ablation.
In the present study, we establish for the first time that PLB is also expressed in the endothelium, a nonmuscle tissue. Furthermore, the lack of expression of this protein in the endothelium of the PLB-KO aorta is associated with altered endothelium-dependent relaxation. Moreover, we show that PLB is an important modulator of the endothelium-dependent component of protein kinase A–mediated relaxation in mouse aorta. This opens new fields for investigation of Ca2+ regulatory pathways in nonmuscle cells and for modulation of endothelial-vascular interactions.
Materials and Methods
Generation of PLB-KO Mice
Mice deficient in the PLB gene were generated by gene-targeting methodology in murine embryonic stem cells as previously described.6 Homozygous breeding was used for the generation of both the wild-type (WT) and PLB-KO mice. Both male and female mice were used for the present study. Care was taken to ensure that gender-matched mice were used for each experiment. All experiments were completed in accordance with institutional animal care guidelines.
Aorta Force Measurements
Isometric forces were measured as described previously.8 9 Briefly, aortic rings were threaded with 2 triangular 100-μm stainless steel wires; each completed mounting formed a double triangle. The aorta and holder were then mounted on a hook that was attached to a Harvard Apparatus differential capacitor force transducer. Resting tension on each aorta was set to 30 mN, to approximate an in vivo aortic pressure of 100 mm Hg, and this passive tension was maintained throughout the experiment. For experiments using denuded aortas, the endothelium was removed by rubbing the aorta between the thumb and index finger. Removing the endothelium in this manner does not affect force development in response to either KCl or phenylephrine. Relaxation responses to acetylcholine (ACh, 1 nmol/L to 10 μmol/L), forskolin (1 nmol/L to 1 μmol/L), and sodium nitroprusside (SNP) (0.1 to 300 nmol/L) were determined in aortas contracted with phenylephrine (3 μmol/L). If the relaxation response to ACh (10 μmol/L) of the denuded vessels exceeded 10%, these data were excluded from analysis. Data were obtained using MP100W hardware and analyzed using AcqKnowledge software (Biopac, Goleta, Calif).
Endothelial Cell Isolation
Primary mouse aortic endothelial cells were isolated as described previously.9 Briefly, endothelial cells were collected from a freshly dissected thoracic aorta. The vessel was ligated at each end with a 5-0 silk suture, and a PBS solution (in mmol/L: KCl 2.68, KH2PO4 1.47, NaCl 136.9, and Na2HPO4 8.1) containing 8 mg/mL collagenase B and 8 mg/mL BSA was injected into the vessel. After a 40-minute incubation at 37°C, the vessel was opened longitudinally and loosely adhered cells were dislodged by repeated flushing with a pipette. These cells were collected, centrifuged, and resuspended in DMEM containing 5.5 mmol/L glucose and 0.1% BSA and plated onto glass coverslips coated with fibronectin. These cells were allowed to grow for 4 passages. To confirm that these cells were endothelial cells, separate coverslips were incubated with 10 mg/mL diI-acetylated LDL for 4 hours at 37°C and then counterstained with 5 mg/mL bis-benzamide to identify individual cells. More than 98% of the isolated cells, as identified by diI-acetylated LDL uptake and nuclear counterstaining were determined to be endothelial cells.
Reverse Transcriptase–Polymerase Chain Reaction (RT-PCR) of Isolated Cells
Primary endothelial cells from 2 aortas were isolated as described above. Pooled endothelial cells and the remaining vessel medial layers were each homogenized in 0.8 mL of RNAzol B (Tel-Test Inc, Friendswood, Tex) and the RNA extracted, per the manufacturer’s instructions, using glycogen as a carrier during RNA precipitation. The entire RNA sample was transcribed into first-strand cDNA using Superscript Reverse Transcriptase (Gibco BRL, Grand Island, NY) at 42°C. Serial dilutions (1:1, 1:2, 1:4, and 1:8) of the cDNA were amplified by hot-start PCR for PLB (Tm=56°C, 35 cycles) or α-actin (Tm=57°C, 27 cycles) transcripts. The primers used for detection were as follows: α-actin, upper, AGACAGCTATGTGGGGGATG; lower, GAAGGAA TAGCCACG CTCAG); PLB, upper, TGTGGGTTGCAAAGTTAGGC; lower, TGTG GGTTGCAA AGTTAGGC.
Western Blot Analysis
Cell homogenates were solubilized in SDS sample buffer and loaded onto a 10% to 20% gradient SDS polyacrylamide gel. Samples were transferred electrophoretically onto a 0.22-μm nitrocellulose membrane. After blocking the nitrocellulose with 5% dry milk, the membrane was incubated for 1 hour at room temperature with a monoclonal antibody to PLB (1:1000 dilution). The antibody-antigen complex was detected after incubation of the blot with horseradish peroxidase–conjugated secondary antibody and visualized using enhanced chemiluminescence Western blotting reagents (Amersham). Blots were quantitated by densitometry (Zenith soft laser scanning densitometer).
Data from concentration-relaxation curves were compared using 2-way ANOVA for repeated measures. EC50 values were determined using a logistic fit with Origin software (Northampton, Mass) and compared with t test analysis. Significance was defined as P<0.05 for all tests.
Effects of PLB Gene Ablation on Endothelium-Dependent Relaxation
To assess the effects of PLB gene ablation on endothelium-dependent relaxation, mouse aortas were stimulated with 3 μmol/L phenylephrine (≈ED80) and subsequently treated with ACh. Figure 1A⇓ depicts a typical tracing obtained from WT and PLB-KO mouse aorta in response to cumulative additions of ACh. Concentration-relaxation relationships indicated that the PLB-KO aortas were less responsive, particularly at lower concentrations of ACh, than the WT controls (Figure 1B⇓). Additionally, the EC50 values were significantly greater in PLB-KO than WT aortas (471±140 versus 130±37 nmol/L, n=7; P<0.05).
Consistent with our previous report, endothelium-dependent relaxation to ACh was abolished by pretreatment with Nω-nitro-l-arginine (0.2 mmol/L, n=3; data not shown), suggesting that NO is the primary mediator of ACh relaxation.9 To determine if the differences in ACh-mediated relaxation were attributable to differential responses of the vascular smooth muscle to NO, concentration-relaxation relations to the NO donor SNP were generated (Figure 2⇓). SNP-mediated relaxation was unaffected by PLB gene ablation, indicating that the differences obtained with ACh were due to the endothelium. Interestingly, the endothelium-intact preparations were less sensitive to SNP than denuded aortas, although no differences between WT and PLB-KO aortas were observed.
The A-kinase pathway has been implicated in modulation of endothelium-dependent relaxation.10 11 12 Because PLB is a major effector of the A-kinase pathway in the heart,6 13 we assessed the effects of forskolin, an adenylate cyclase activator, on relaxation in endothelium-intact and denuded aortas. Concentration-dependent relaxation to forskolin was observed in both endothelium-intact and denuded aortas from WT and PLB-KO mice. Evaluation of the forskolin concentration-response curves (Figure 3⇓) revealed that WT aortas with an intact endothelium were significantly more sensitive than their denuded counterparts (ED50=31.0±3.3 versus 89.9±7.7 nmol/L, respectively, n=7; P<0.01). Importantly, endothelium-intact aortas from PLB-KO showed a much smaller increase in sensitivity (ED50=49.0±4.8 and 74.9±11.5 nmol/L in endothelium intact and denuded aortas, respectively, n=7; P>0.05). Thus, the differences in forskolin relaxation are mediated by the endothelium and depend on the presence or absence of PLB.
Evidence for the Presence of PLB in the Endothelium
Taken together, these observations provide functional evidence for PLB in the endothelium. Therefore, we directly tested for the presence of PLB in endothelial cells using both fresh isolates and primary cell cultures. Figure 4⇓ demonstrates the presence of PLB by both RT-PCR and Western blot analyses. RT-PCR analyses of freshly isolated aortic endothelial cells and cells from the remaining aortic media demonstrated the presence of PLB mRNA in both the endothelium and vascular smooth muscle. In parallel control experiments, α-actin transcripts were detected only in vascular smooth muscle and not in endothelial cells. This indicates that there was no significant smooth muscle contamination of the endothelial cell preparations (Figure 4A⇓). Furthermore, Western blot analysis of cultured mouse aortic endothelial cells confirmed the presence of PLB in the endothelium (Figure 4B⇓).
The present study establishes for the first time that PLB is present in the endothelium and modulates endothelium-dependent relaxation of mouse aorta. The importance of PLB in cardiovascular function in vivo has been further elucidated through the generation of PLB gene-manipulated mouse models.6 14 15 16 17 PLB, by virtue of its regulatory actions on SERCA, modulates Ca2+ in the heart and underlies inotropic responses to β-adrenoceptor stimulation.6 16 PLB in the vascular smooth muscle is also an important determinant of steady-state Ca2+ and isometric force.8 18 However, there is little known about the effects of PLB in nonmuscle tissue, because it has only been reported to be present in cardiac, slow skeletal, and smooth muscles. Our present data clearly show that not only is PLB present in the endothelium, but it can also modulate cardiovascular function through endothelium-mediated relaxation. Thus, PLB is important not only in muscle tissue, as previously held, but also can regulate the endothelium, a nonmuscle tissue.
Phosphorylation of PLB via the protein kinase A pathway is an important regulator of [Ca2+]i in cardiac muscle.6 13 The A-kinase pathway has also been implicated in an endothelium-dependent component of β-adrenoceptor relaxation of vascular tissues.10 11 12 Consistent with this observation, we also measured a significant increase in sensitivity to forskolin in endothelium-intact aortas from WT mice. Importantly, this difference in forskolin-mediated relaxation between endothelium-intact and denuded vessels was blunted in aortas from PLB-KO mice. This suggests that PLB gene ablation results in a loss of endothelium-dependent A-kinase vasorelaxation.
Our hypothesis is that unphosphorylated PLB is associated with inhibition of Ca2+ uptake into the endoplasmic reticulum, resulting in higher [Ca2+]i and increased activation of endothelial NO synthase. Phosphorylation or ablation of PLB increases endoplasmic reticulum Ca2+ uptake, leading to a lower [Ca2+]i and reduced endothelium-dependent relaxation. Therefore, PLB influences endothelium-dependent relaxation by modulating Ca2+ availability for activation of endothelial NO synthase.
These results are particularly interesting when considered along with recent results demonstrating reduced endothelium-dependent relaxation in mice deficient in sarco(endo)-plasmic reticulum Ca2+-ATPase isoform 3 (SERCA3).9 Previously, we concluded that the SERCA3 isoform is an important determinant of ACh-releasable Ca2+ stores. Our results showed that deletion of PLB, a protein associated with inhibition of Ca2+ sequestration, can also reduce endothelium-dependent relaxation. In this case, we hypothesized that this reduction was attributable to enhanced endoplasmic reticulum Ca2+ uptake leading to lower [Ca2+]i. Therefore, the interplay between Ca2+ release and uptake systems is an important determinant of endothelial steady-state Ca2+ levels that elicits endothelium-dependent relaxation.
In summary, our present results clearly demonstrate that PLB is present in the endothelium and can modulate vascular contractility. Importantly, endothelial PLB plays a dominant role in the endothelium-dependent component of A-kinase–mediated relaxation. Therefore, PLB may play more important roles in cardiovascular regulation as well as in Ca2+ homeostasis in nonmuscle tissues than previously thought.
This study was supported by HL-09781 (to R.L.S.); HL-54829 (to R.J.P.); and HL-26057 and P40RR12358 (to E.G.K.). The authors are grateful to Craig Weber for excellent technical assistance.
- Received November 13, 1998.
- Accepted January 4, 1999.
- © 1999 American Heart Association, Inc.
Schilling WP, Cabello OA, Rajan L. Depletion of the inositol 1,4,5-trisphosphate-sensitive intracellular Ca2+ store in vascular endothelial cells activates the agonist-/sensitive Ca(2+)-influx pathway. Biochem J. 1992;284:521–530.
Colden-Stanfield M, Schilling WP, Ritchie AK, Eskin SG, Navarro LT, Kunze DL. Bradykinin-induced increases in cytosolic calcium and ionic currents in cultured bovine aortic endothelial cells. Circ Res. 1987;61:632–640.
Luo W, Grupp IL, Harrer J, Ponniah S, Grupp G, Duffy JJ, Doetschman T, Kranias EG. Targeted ablation of the phospholamban gene is associated with markedly enhanced myocardial contractility and loss of beta-agonist stimulation. Circ Res. 1994;75:401–409.
Slack JP, Grupp IL, Luo W, Kranias EG. Phospholamban ablation enhances relaxation in the murine soleus. Am J Physiol. 1997;273:C1–C6.
Lalli J, Harrer JM, Luo W, Kranias EG, Paul RJ. Targeted ablation of the phospholamban gene is associated with a marked decrease in sensitivity in aortic smooth muscle. Circ Res. 1997;80:506–513.
Liu LH, Paul RJ, Sutliff RL, Miller ML, Lorenz JN, Pun RY, Duffy JJ, Doetschman T, Kimura Y, MacLennan DH, Hoying JB, Shull GE. Defective endothelium-dependent relaxation of vascular smooth muscle and endothelial cell Ca2+ signaling in mice lacking sarco(endo)plasmic reticulum Ca2+-ATPase isoform 3. J Biol Chem. 1997;272:30538–30545.
Kamata K, Miyata N, Kasuya Y. Involvement of endothelial cells in relaxation and contraction responses of the aorta to isoproterenol in naive and streptozotocin-induced diabetic rats. J Pharmacol Exp Ther. 1989;249:890–894.
Tada M, Kirchberger MA, Katz AM. Phosphorylation of a 22,000-dalton component of the cardiac sarcoplasmic reticulum by adenosine 3′:5′-monophosphate-dependent protein kinase. J Biol Chem. 1975;250:2640–2647.
Lorenz JN, Kranias EG. Regulatory effects of phospholamban on cardiac function in intact mice. Am J Physiol. 1997;273:H2826–H2831.
Luo W, Wolska BM, Grupp IL, Harrer JM, Haghighi K, Ferguson DG, Slack JP, Grupp G, Doetschman T, Solaro RJ, Kranias EG. Phospholamban gene dosage effects in the mammalian heart. Circ Res. 1996;78:839–847.