Nanomolar ouabain increases NCX1 expression and enhances Ca2+ signaling in human arterial myocytes: a mechanism that links salt to increased vascular resistance?

Abstract

The mechanisms by which NaCl raises blood pressure (BP) in hypertension are unresolved, but much evidence indicates that endogenous ouabain is involved. In rodents, arterial smooth muscle cell (ASMC) Na(+) pumps with an α(2)-catalytic subunit (ouabain EC(50) ≤1.0 nM) are crucial for some hypertension models, even though ≈80% of ASMC Na(+) pumps have an α(1)-subunit (ouabain EC(50) ≈ 5 μM). Human α(1)-Na(+) pumps, however, have high ouabain affinity (EC(50) ≈ 10-20 nM). We used immunoblotting, immunocytochemistry, and Ca(2+) imaging (fura-2) to examine the expression, distribution, and function of Na(+) pump α-subunit isoforms in human arteries and primary cultured human ASMCs (hASMCs). hASMCs express α(1)- and α(2)-Na(+) pumps. Further, α(2)-, but not α(1)-, pumps are confined to plasma membrane microdomains adjacent to sarcoplasmic reticulum (SR), where they colocalize with Na/Ca exchanger-1 (NCX1) and C-type transient receptor potential-6 (receptor-operated channels, ROCs). Prolonged inhibition (72 h) with 100 nM ouabain (blocks nearly all α(1)- and α(2)-pumps) was toxic to most cultured hASMCs. Treatment with 10 nM ouabain (72 h), however, increased NCX1 and sarco(endo)plasmic reticulum Ca(2+)-ATPase expression and augmented ATP (10 μM)-induced SR Ca(2+) release in 0 Ca(2+), ouabain-free media, and Ca(2+) influx after external Ca(2+) restoration. The latter was likely mediated primarily by ROCs and store-operated Ca(2+) channels. These hASMC protein expression and Ca(2+) signaling changes are comparable with previous observations on myocytes isolated from arteries of many rat hypertension models. We conclude that the same structurally and functionally coupled mechanisms (α(2)-Na(+) pumps, NCX1, ROCs, and the SR) regulate Ca(2+) homeostasis and signaling in hASMCs and rodent ASMCs. These ouabain/endogenous ouabain-modulated mechanisms underlie the whole body autoregulation associated with increased vascular resistance and elevation of BP in human, salt-sensitive hypertension.

Fig. 1.

Identification of Na⁺ pump α-subunit isoforms and Na/Ca exchanger-1 (NCX1) in human arterial smooth muscle cells (hASMCs) and mouse (m) tissues. A: both mouse aorta smooth muscle and primary cultured internal thoracic hASMCs cross-react with anti-HERED (α₂) antibodies. hASMCs also cross-react with anti-TSEP (human α₁) but not anti-NASE (rodent α₁) antibodies; mouse aorta smooth muscle cross-reacts with anti-NASE, but not anti-TSEP. B: mouse brain, but neither hASMCs nor mouse aorta smooth muscle, cross-reacts with anti-TED (α₃) antibodies. C: both hASMCs and mouse aorta smooth muscle cross-react with R3F1 (anti-NCX1) antibodies. Anti-β-actin served as a loading control. The expected molecular weights are as follows: α₁, 110 kDa; α₂, 110 kDa; α₃, 110 kDa; NCX1, 116 kDa; and β-actin, 42 kDa. Each panel is representative of replicates from at least 3 different patients or mouse preparations.

Fig. 2.

Expression of α₁- and α₂-Na⁺ pumps, NCX1, and C-type transient receptor potential (TRPC6) in smooth muscle from freshly harvested human mesenteric small arteries. Blots were probed with (top to bottom, as indicated) anti-HERED [polyclonal antibodies (pAb)], anti-TSEP (human α₁ pAb), and R3F1 [anti-NCX1 monoclonal antibodies (mAb)]; and anti-TRPC6 (pAb). Numbered lanes refer to mesenteric arteries (3^rd or 4^th order branches) from 4 different patients. Anti-β-actin served as the loading controls for all the blots. These data are representative examples from a total of 10 different patients (7 male, 3 female; 9 Caucasian, 1 Hispanic; age range = 21–63 years; 6 of the 10 patients died from trauma or suicide; no patients were hypertensive).

Fig. 3.

Expression of some Ca²⁺ transporters in hASMC. The data are representative blots of proteins from freshly harvested human mesenteric artery distal branches (Fresh; left lane) and primary cultured hASMCs (Cult; right lane). A: NCX1. B: TRPC6. C: sarco(endo)plasmic reticulum Ca²⁺-ATPase (SERCA2). D: inositol triphosphate receptor-1 (IP₃R1). Anti-β-actin served as the loading control. Data are representative of duplicate blots of proteins from each of 2 or (NCX1) 3 patients.

Fig. 4.

Distribution of α₁-Na⁺ pumps, PMCA, and the sarcoplasmic reticulum (SR) in primary cultured mesenteric hASMCs. A: images of a hASMC that was cross-reacted with anti-TSEP human α₁ pAb (a) and anti-PMCA mAb (b). The SR was then stained with ER tracker to label the SR (c). d: image (control) of a cell cross-reacted only with anti-rabbit secondary polyclonal antibody labeled with FITC; similar results were obtained with cells labeled with anti-mouse secondary monoclonal antibody labeled with Cy3. Scale bars = 25 μm. N, nucleus. B and C: higher magnification images from the boxed areas in the images in A. B includes the α₁ (a; red) and SR (c; green) images and the image overlay (d); b, black-and-white ER tracker (SR) image. C includes the PMCA (a; green) and SR (b; red) images and the image overlay (c). Note that the α₁ (TSEP) and PMCA staining patterns both differ markedly from that of ER tracker (an SR stain); this is reflected by the paucity of yellow-orange staining in the overlays.

Fig. 5.

Distribution of α₂-Na⁺ pumps, NCX1, and SR in primary cultured mesenteric hASMCs. A: hASMCs were cross-reacted with anti-HERED α₂ pAb and R3F1 anti-NCX1 mAb; the SR was then stained with ER tracker, as indicated by the labels. Insets: enlargements of the boxed areas. Pseudocolor images of the enlarged α₂ (red) and NCX1 (green) regions, and the overlay, are shown at right. B: hASMCs were cross-reacted with R3F1 anti-NCX1 mAb and anti-TRPC6 pAb; the SR was then stained with ER tracker, as indicated. N, nuclei. Insets: enlargements of the boxed areas. Pseudocolor images of the enlarged NCX1 (green) and TRPC6 (red) regions, and the overlay, are shown at right. In A and B, the patterns of staining by both antibodies were very similar to the pattern of ER tracker (i.e., SR) distribution. Scale bars = 30 μm. Note that the α₂, NCX1, and TRPC6 staining patterns are all very similar to that of ER tracker. This is reflected by the yellow-orange staining in the A and B overlay panels and indicates that α₂-Na⁺ pumps and NCX1 colocalize and overlie elements of SR.

Fig. 6.

Distribution of α₂-Na⁺ pumps and SERCA2 in a dissociated myocyte from a freshly harvested human internal thoracic artery. Myocytes were cross-reacted with anti-SERCA2 (SERCA2 mAb; A) and with anti-HERED (α₂ pAb; B). The confocal images of a freshly dissociated hASMC are shown at left, and enlarged images of the boxed areas are shown at right. C: image overlays at low (left) and high magnification. Scale bars = 5 μm. Note that the α₂ and SERCA2 staining patterns are very similar near the surface of the cell where the SR comes close to the plasma membrane (at right of low-power images), whereas only the SERCA2 mAb stains the interior of cell (at left of low-power images). This colocalization is reflected by the yellow-orange staining in the overlay panels; it indicates that the plasma membrane microdomains containing the α₂-Na⁺ pumps overlie junctional elements of SR.

Fig. 7.

Effects of prolonged incubation with 10 or 100 nM ouabain on the morphology of primary cultured human and rat mesenteric small artery myocytes. Morphology of hASMCs were cultured for 10 days in growth medium (A), or in growth medium containing 10 nM (B) or 100 nM (C) ouabain for the final 72 h of culture. In all cultures, the FBS-supplemented medium was replaced by FBS-free medium 24 h before the ouabain was added. After a 72-h incubation without or with ouabain, the cells were loaded with fura-2; the fura-2 images (360 nm excitation) are shown. Arrowheads in B and C point to some of the rounded-up cells; none are seen in A. For comparison, rat mesenteric artery myocytes cultured for 72 h without (D) or with (E) 100 nM ouabain are also shown; for details, see Ref. . Scale bars = 50 μm.

Fig. 8.

Effects of 72-h incubation with nanomolar ouabain on the expression of NCX1, TRPC6, SERCA2, and IP₃R1 in primary cultured hASMCs. Ouabain, 5 nM (A only) or 10 nM (A-D), was added to the culture medium of some cells, and the cells were maintained in culture for 72 h. The data are representative immunoblots (top) and averaged data from 3 to 5 patients (bottom) probed with anti-NCX1 (A), anti-TRPC6 (B), anti-SERCA2 (C), and anti-IP₃R1 (D) antibodies. Protein expression was normalized to GAPDH (A, C, D) or β-actin (B) expression (the 2 loading control proteins were used interchangeably because they gave similar results; also see Figs. 1–3). The blots from each patient were run in duplicate or triplicate. NCX1 and SERCA2 were significantly upregulated by 10 nM ouabain; *P < 0.05 vs. control (no ouabain).

Fig. 9.

Effects of 72-h incubation with 10 nM ouabain on Ca²⁺ signaling in primary cultured hASMCs. A: changes in cytosolic Ca²⁺ concentration ([Ca²⁺]_CYT) in a representative cell from each of 2 coverslips of hASMCs, 1 treated with 10 nM ouabain (red line) and 1 untreated control (blue line), both from the same patient. As indicated by the bars below the data records, normal physiological salt solution (PSS) was replaced by Ca²⁺-free PSS 2.5 min before 10 μM ATP was introduced. Then, after an initial, transient rise in [Ca²⁺]_CYT and recovery to the initial basal level, external Ca²⁺ was restored with ATP still present; this induced a second, more prolonged [Ca²⁺]_CYT transient. Finally, the ATP was removed, usually before the [Ca²⁺]_CYT had returned to the original basal level. None of the solutions used for Ca²⁺ imaging contained ouabain. B-D: averaged data for hASMCs from 1 mesenteric small artery from each of 3 patients; data in B are from the same patient as the data in A. The bars indicate the mean peak amplitudes ± SE of the first (Ca²⁺ release) and second (Ca²⁺ influx) Ca²⁺ transients, measured from the basal (resting) [Ca²⁺]_CYT (Δ[Ca²⁺]_CYT in nM). Blue bars, control cells; red bars, ouabain-treated cells. *P < 0.05 and **P < 0.01 vs. untreated controls. The basal [Ca²⁺]_CYT levels and numbers of cells (n) in which basal Ca²⁺ and Ca²⁺ release were measured are presented in Table 1 (patients 1–3). The expression of α₁- and α₂-Na⁺ pumps, NCX1 and TRPC6, in the smooth muscle from the freshly harvested mesenteric arteries of patients 2 and 3 of Fig. 2 correspond to patients 1 and 2, respectively, in this figure and in Table 1.