Sensational site: the sodium pump ouabain-binding site and its ligands

Abstract

Cardiotonic steroids (CTS), used by certain insects, toads, and rats for protection from predators, became, thanks to Withering's trailblazing 1785 monograph, the mainstay of heart failure (HF) therapy. In the 1950s and 1960s, we learned that the CTS receptor was part of the sodium pump (NKA) and that the Na⁺/Ca²⁺ exchanger was critical for the acute cardiotonic effect of digoxin- and ouabain-related CTS. This "settled" view was upended by seven revolutionary observations. First, subnanomolar ouabain sometimes stimulates NKA while higher concentrations are invariably inhibitory. Second, endogenous ouabain (EO) was discovered in the human circulation. Third, in the DIG clinical trial, digoxin only marginally improved outcomes in patients with HF. Fourth, cloning of NKA in 1985 revealed multiple NKA α and β subunit isoforms that, in the rodent, differ in their sensitivities to CTS. Fifth, the NKA is a cation pump and a hormone receptor/signal transducer. EO binding to NKA activates, in a ligand- and cell-specific manner, several protein kinase and Ca²⁺-dependent signaling cascades that have widespread physiological effects and can contribute to hypertension and HF pathogenesis. Sixth, all CTS are not equivalent, e.g., ouabain induces hypertension in rodents while digoxin is antihypertensinogenic ("biased signaling"). Seventh, most common rodent hypertension models require a highly ouabain-sensitive α2 NKA and the elevated blood pressure is alleviated by EO immunoneutralization. These numerous phenomena are enabled by NKA's intricate structure. We have just begun to understand the endocrine role of the endogenous ligands and the broad impact of the ouabain-binding site on physiology and pathophysiology.

Keywords: cardiotonic steroids; digoxin; marinobufagenin; protein kinase cascade; signalosome.

Graphical abstract

Figure 1.

Cardiotonic steroids, NKA topography, and the NKA ouabain-binding site. A and B: the structures of two cardenolides (A), ouabain and digoxin, and two bufadienolides (B), bufalin and marinobufagenin, pertinent to this review, are illustrated. All four have both A/B and C/D rings in a cis conformation so that A and D are bent toward one another. Ouabain, digoxin, and bufalin are hydroxylated at C14, whereas marinobufagenin has an epoxide at C14–C15. C: diagram of the NKA α subunit topography showing the 10 transmembrane domains and the N and C termini. The colored ovals indicate the activator domain (A), spanning parts of the N terminus and the cytoplasmic M2-M3 loop, and the nucleotide-binding domain (red N) and phosphorylation domain (P), both spanning parts of the large M4-M5 cytoplasmic loop. The positions of the amino acids, Gln111, Asn122, Asp 892, and Arg906 are also indicated. D: structure of the NKA protomer in the E2P conformational state with bound ouabain (OBN, orange). This figure also shows the A domain (yellow), the N domain (red), and the P domain (blue) of the α subunit. Also shown are the β subunit (maroon), consisting of an extracellular (glycosylated) domain and a single transmembrane domain, and an FXYD peptide within the membrane (light orange). Reproduced from Ref. with permission. E: detailed model structures of the pig kidney ouabain-sensitive (α1^S) NKA ouabain-binding site with bound ouabain (blue), and the rat kidney ouabain-resistant (α1^R) NKA with bound ouabain (red). Note that the two amino acid substitutions, Gln111Arg and Asn121Asp, enables the uncharged Arg and Asp to form a hydrogen bond and accounts for the ouabain resistance of rat α1. This bond prevents ouabain from penetrating as deeply into the pocket as it does in α1^S, as illustrated. NKA, Na⁺, K⁺-ATPase (Na⁺ pump). Reproduced from Ref. with permission.

Figure 2.

Stimulation and inhibition of a dog kidney NKA membrane preparation by ouabain. A: graph shows the dose-response relationships for the NKA preparation after rapid washing in high Na⁺ or K⁺ solutions. The Na⁺-washed NKA (O) retained its ability to be stimulated by low concentrations of ouabain, whereas the K⁺-washed enzyme (•) did not. Each data point is the mean of 4 determinations ± SD and is presented as a % of control activity. B: data from the same experiments but presented in terms of absolute-specific NKA activity. The specific activity of the potassium-washed NKA increased but lost all ability to be stimulated by ouabain. C: the green area highlighted in B is enlarged in C to illustrate the physiological (blue) range for ouabain- and digoxin-like eCTS, as well as the therapeutic (orange) and toxic/lethal (red) values for ouabain and digoxin in the human circulation. The upper value for these eCTS is at least one order of magnitude lower than the concentration of their exogenous counterparts given therapeutically. CTS, cardiotonic steroid; NKA, Na⁺, K⁺-ATPase (Na⁺ pump). Illustrations reproduced and modified from Ref. 88) with permission of The Company of Biologists Ltd.

Figure 3.

Ouabain-induced hypertension and ouabain-digoxin antagonism. A: prolonged infusion of ouabain, but not digoxin, elevates blood pressure (BP) in rats; digoxin antagonizes the effect of ouabain. Sprague-Dawley rats were infused by subcutaneous osmotic minipumps with vehicle, ouabain (15 µg/kg/day) or digoxin (30 µg/kg/day) for 6 wk. After 5 wk, some ouabain-infused rats (n = 8/group) received a secondary infusion of vehicle, digoxin (30 µg/kg/day) or digitoxin (30 µg/kg/day). Mean BPs were measured by tail cuff. *P < 0.05 vs. ouabain; ***P < 0.001 vs. ouabain; #P < 0.005 vs. vehicle; **P < 0.001 vs. digoxin. In a second study, after infusion of vehicle, ouabain, or digoxin for 5 wk, plasma steroid immunoreactivity was measured with ouabain- or digoxin-specific immunoassays. The levels were: vehicle + endogenous ouabain (EO), 0.66 ± 0.12 nM; ouabain-infused, 5.2 ± 0.39 nM ouabain + EO; digoxin-infused, 3.3 ± 0.32 nM digoxin (n = 8 rats/group). Data are regraphed from Ref. and presented with permission. B: the abilities of various cardiotonic steroids (CTS) to alter myogenic tone in isolated rat mesenteric small arteries pressurized to 70 mmHg (MT₇₀). Summarized data show the relative abilities of various CTS to antagonize the 3 nM ouabain-induced increase in MT₇₀ (upper row of bars), and the ability of 10 nM digoxin to antagonize the increase in MT₇₀ induced by various CTS (lower row of bars). The data (means of 5–8 arteries ± SE per bar) are expressed as: b/a × 100% where “a” is the 3 nM ouabain-induced (top) or other CTS-induced (bottom) constriction and “b” is the constriction when 10 nM other CTS (top graph) or digoxin (bottom graph) was added. Positive values indicate vasodilation (or antagonism); negative values indicate vasoconstriction (or synergism) induced by the second CTS. The ability of the various CTS to induce hypertension when infused subcutaneously (96, 326) is indicated at the bottom: “+” agents induce hypertension; “–“ agents do not induce hypertension and are antihypertensinogenic (i.e., they antagonize the hypertensinogenic effect of ouabain); ND, not determined. The figure is regraphed from Ref. and reproduced with permission.

Figure 4.

Prolonged incubation with ouabain, but not digoxin, augments Ca²⁺ transporter expression in primary cultured rat arterial and cardiac myocytes; Src kinase inhibition prevents the upregulation in the arterial myocytes (cardiomyocytes were not tested). Primary cultured rat superior mesenteric artery myocytes and left ventricular cardiomyocytes were incubated for 72 h with vehicle, ouabain (Ouab, 50 or 100 nM), digoxin (Dig, 100 nM), or ouabain + digoxin (Ouab + Dig, 100 nM, each). In some instances, the arterial myocytes were incubated with the Src kinase inhibitor, PP2 (4-amino-5-(4-chlorophenyl)-7-(t-butyl)pyrazolo[3,4-d]pyrimidine, 1 µM), or its inactive analog, PP3 (4-amino-7-phenylpyrazol[3,4-d]pyrimidine, 1 µM), both without and with 100 nM ouabain. A and B: after washout of the cardiotonic steroids, the cytosolic Ca²⁺ concentration ([Ca²⁺]_CYT) in the arterial myocytes was measured with Fura-2, both at rest and during stimulation, with 1 µM phenylephrine (PE). A: representative cell recordings. B: resting: n = 298–316 cells/bar ± SE; peak: n = 39–49 cells/bar ± SE. C–E: the effects of ouabain (100 nM), digoxin (100 nM) and ouabain + digoxin, and of PP2 (Src kinase inhibitor) and PP3 (inactive PP2 analog), both without and with ouabain, were tested on arterial myocyte Na⁺/Ca²⁺ exchanger-1 (NCX1) and transient receptor potential C6 (TRPC6) protein expression. Each bar shows the mean of 4–8 immunoblots ± SE. F: Ouabain (50 and 100 nM), but not digoxin (100 nM), also upregulates NCX1 expression in cardiomyocytes. *P < 0.05; **P < 0.01; ***P < 0.001 vs. vehicle and digoxin. Data from Ref. (A–E) and Ref. (F) regraphed and reproduced with permission.

Figure 5.

Models for the interaction of ouabain and digoxin with the Na⁺ pump. In all models, four α2 Na⁺ pump subunits are shown for consistency. White and red ovals denote active and inactive α2 subunits, respectively. For simplicity, in Models 1–4, each reaction initially involves addition of one CTS (usually ouabain, black triangle) at its EC50, so that, at steady state, exactly one half of the subunits are inactivated (residual pump activity = 50%). Addition of a second, structurally different, CTS (e.g. digoxin, gray triangle), at a concentration approximately twice the EC50 for ouabain, is then shown. Model 1 uses monoprotomers, while Models 2–4 invoke diprotomers, either without (Model 2), or with (Models 3 and 4), half-of-the-sites reactivity. As illustrated, none of those models can display ouabain antagonism. Models 5 and 6 invoke tetraprotomers each with quarter-site reactivity. In the latter models, binding of one CTS molecule to a single tetramer is sufficient to stop all ion pumping. Models 4 and 6 invoke disaggregration of complexed subunits upon addition of a second CTS. In each case, the relative ion-pumping activity after addition of the first and second CTS (as a percentage of baseline) is shown below each model. In Model 6A, the binding of a single ouabain molecule (inverted black triangle) to an active protomer (white subunit, 100% activity; red subunits are inactive) fully inhibits the NKA (red subunit with inverted black triangle, 0% activity). Subsequent binding of one or more digoxin molecules (inverted gray triangles) to an unbound protomer(s) then causes the tetraprotomer to disaggregate and one of the protomers (white subunit) recovers 100% activity, i.e., ouabain-digoxin antagonism. Conversely, in Model 6B, binding of a digoxin molecule to one of the four protomers also inhibits the NKA; subsequent binding of one or more ouabain molecules to an inactive protomer(s) causes the tetraprotomer to disaggregate and one of the protomers (white subunit) to recover 100% activity. CTS, cardiotonic steroid. Reproduced from Ref. with permission.

Figure 6.

Signaling pathways involved in cardiotonic steroid (CTS)-mediated positive cardiac inotropy and ouabain-activated cardiac hypertrophy. A: normal operation of the NKA (Ion Pumping). B: acute NKA inhibition by any cardiotonic steroid (“Pump Inhibition”) raises cytosolic Na⁺ ([Na⁺]_CYT), thereby inhibiting Na⁺/Ca²⁺ exchanger-1 (NCX1)-mediated Ca²⁺ extrusion (and promoting NCX1-mediated Ca²⁺ entry - not shown). The result is a net rise in cytosolic Ca²⁺ ([Ca²⁺]_CYT), enhanced Ca²⁺ signaling and, in the heart, positive inotropy. C: sustained NKA block by (endogenous) ouabain (EO) activates protein kinase cascades, which lead to reactive oxygen species (ROS) generation, altered protein expression, altered Ca²⁺ signaling (not mediated by NCX1) and, in the heart, cardiac hypertrophy. . NKA, Na⁺, K⁺-ATPase (Na⁺ pump). The model is modified from Fig. 10 in Ref. and Fig. 11 in Ref.

Figure 7.

High dietary salt increases plasma endogenous ouabain (EO) in normal human subjects and in Dahl salt-sensitive (DS) rats. A: on day “0,” 10 g (171 meq) NaCl was added to the normal salt diet (NS) of normal human males for 5 days (= high salt, HS). Endogenous ouabain (EO) was measured on C-18-extracted plasma samples with an ouabain-specific antibody immunoassay (229) for 5 days before the dietary salt increase and during the 5 days on HS. HS led to a 13-fold increase, on average, in plasma EO on day 3 (5.8 ± 2.2 nM; n = 13, P <0.05), after which the EO level fell to an elevated plateau relative to the level during NS. Data from two representative subjects are shown (○ and ●, respectively). Reproduced from Ref. with permission. B–D: an HS diet increases plasma endogenous ouabain (EO), arterial myocyte Ca²⁺ transporter expression, and systolic blood pressure (SBP) in DS rats. Dahl salt-resistant (DR; n = 5) and DS (n = 7) rats were initially fed a 0.4% NaCl (NS) diet and SBP was measured by tail cuff for 5 days before switching to a 6% NaCl (HS) diet. B: graph shows the mean SBP in the rats for the 5 days on the NS diet and for days 7–11 on the HS diet, when the SBP had reached a plateau. ***P < 0.001 for the indicated pairs. C: representative immunoblots indicate that the HS diet, elevated plasma EO, and the high BP in DS rats on the HS diet are all associated with enhanced expression of the Ca²⁺ transporters, Na⁺/Ca²⁺ exchanger-1 (NCX1), and transient receptor potential channel-6 (TRPC6). This is also seen after prolonged ouabain treatment of arterial myocytes (Fig. 4, C–E) and in arterial myocytes from rats with ouabain-induced hypertension (Supplemental Fig. S6E). D: plasma EO is higher in DS than DR rats on a HS diet. EO was measured in pooled plasma samples following solid-phase extraction by direct spray-tandem multistage mass spectroscopy (SPE-MS-MS-MS) using exogenous dihydro-ouabain (DHO) as an internal reference standard. Samples were monitored for lithiated molecular adducts (M+Li⁺) at 379.2 and 381.2 m/z (mass/charge ratio), respectively. D, a and b: standards: absolute ion intensities are unsuppressed. D, c and d: highly concentrated C18 extracts of pooled plasma samples were spiked with 50 nM DHO; the absolute ion intensities are suppressed by ∼75%. The relevant numbers are the EO/DHO peak ratios: 0.65 for DR-HS vs. 8.50 for DS-HS rats, a 13-fold increase. Data in B and C reproduced from Ref. with permission; D from the online supplement to Ref. .

Figure 8.

Subcutaneous (sc) low-dose angiotensin II (Ang II) + high dietary salt (HS) increases plasma endogenous ouabain (EO) and expression of arterial smooth muscle Ca²⁺ transporters, and raises mean blood pressure (MBP). Intracerebroventricular (ICV) very low-dose Ang II also raises MBP and plasma EO. A–C: Sprague-Dawley rats were fed a normal salt diet (0.4% NaCl, NS) for 1 wk, and then NS or an HS diet (2% NaCl) and infused sc with vehicle or low-dose Ang II (150 ng/kg/min via minipump) for 2 wk. MBP was measured by telemetry (A). Plasma EO was measured by radioimmunoassay (RIA) (B). Relative expression of aortic Na⁺/Ca²⁺ exchanger-1 (NCX1) and sarco-/endoplasmic reticulum ATPase-2 (probably mostly SERCA2b, the predominant isoform in vascular smooth muscle) was measured by immunoblot (C). *P < 0.05, ***P < 0.001 vs. NS; #P < 0.05 vs. HS and NS+ Ang II (ANOVA). Data from Ref. regraphed and published here with permission. D: n = 4 (NS) or 8 rats (all others); intracerebroventricular (ICV) infusion of Ang II (2.5 ng/min) in male Wistar rats for 14 day elevates plasma EO and MBP. Concurrent ICV infusion of eplerenone (Epl, 5 µg/day), a mineralocorticoid receptor blocker, or FAD-286 [FAD, 4-(5,6,7,8-tetrahydroimidazo[5,1-f]pyridin-5-yl)benzonitrile hydrochloride; 25 µg/day], an aldosterone synthase inhibitor, prevents EO and MBP elevation. MBP was measured via a right femoral artery catheter. EO was measured by RIA on solid phase-extracted plasma samples; EO was verified off-line by liquid chromatography-multistage mass spectroscopy (LC-MS-MS-MS) on plasma samples pooled from each of the groups. *P < 0.05 vs. all other groups (two-way ANOVA); n = 7 or 8 rats/group. Data from Ref. regraphed and published with permission.

Figure 9.

Impact of adrenocorticotropic hormone (ACTH) on endogenous ouabain (EO) and blood pressure (BP) in normal and genetically modified (ouabain-resistant) mice. A–C: ACTH injections elevated plasma (endogenous) ouabain-like compound(s) (eOLC) in wild-type (WT; α1^R/Rα2^S/S) mice and mice with ouabain-resistant α2 NKA (α1^R/Rα2^R/R) (B) but induced hypertension only in the WT mice (A). WT and α1^R/Rα2^R/R mice were injected subcutaneously with saline (n =15 per genotype) or 500 µg/kg ACTH fragment 1–24 (n =25 per genotype) for 5 days; systolic blood pressure (SBP) was measured by tail cuff 3 h after each injection. Bars represent means ± SE; *P < 0.05 vs. saline treatment. C: when α1 was made ouabain-sensitive in the α2^R/R mice (i.e., α1^S/Sα2^R/R or “SWAP” mice), ACTH elevated SBP even more than in WT mice. Protocol similar to that for A; n = 12 mice of each genotype injected with saline and n = 30 of each genotype injected with ACTH. #P < 0.05 vs. α1^R/Rα2^S/S ACTH. Data from Ref. regraphed and published with permission. D: mice (C57Bl/6) with cardiac-specific knockout of α2^S/S NKA (C-α2^S/S-KO), like wild-type (WT) mice, develop ACTH-induced hypertension. SBP was measured by tail cuff. Synthetic ACTH (Cortrosyn, 375 µg/kg) or saline was injected subcutaneously (sc) every 12 h; n = 5 WT and 5 C-α2^S/S-KO mice injected with saline and n = 6 WT and 4 C-α2^S/S-KO mice injected with ACTH. Bars represent means ± SE; *P < 0.05 vs. saline treatment (two-way ANOVA). Data from Ref. regraphed and published with permission. E: in contrast to C-α2^S/S-KO mice, mice with cardiovascular-specific KO of α2^S/S NKA (CV-α2^S/S-KO) do not develop ACTH-induced hypertension. Protocol similar to that in D; n = 6 WT and 9 CV-α2^S/S-KO mice injected with saline and n = 6 WT and 8 CV-α2^S/S-KO mice injected with ACTH. *#P < 0.05 vs. the corresponding ACTH or saline basal SBP (two-way ANOVA). Data from Ref. regraphed and published with permission.

Figure 10.

Association between plasma endogenous ouabain (EO), Ca²⁺ transporter expression, and heart failure (HF) in humans and mice. Aa: plasma EO in normal human subjects and patients with idiopathic dilated cardiomyopathy. Severity of HF is denoted by the New York Heart Association (NYHA) classification: I. No limitation of physical activity; II. slight limitation of physical activity; III. marked limitation of physical activity (modest activity causes fatigue, palpitations or dyspnea). Plasma EO measured with an ouabain-specific radioimmunoassay (RIA); results were confirmed with liquid chromatography-mass spectroscopy (LCMS) in four random patients. “n” = numbers of patients in each group. Data are shown as means ± SE; ***P < 0.001 (ANOVA). Data from Table 3 in Ref. . b: expression of Na⁺/Ca²⁺ exchanger = 1 (NCX1) and sarco-/endoplasmic reticulum Ca²⁺ ATPase-2 (SERCA2; likely SERCA2a, the predominant cardiac isoform) in the left ventricle (LV) of patients with non-failing (NF) hearts and those in HF. Patients with HF due to either dilated cardiomyopathy (DCM) or coronary artery disease (CAD) increased expression of NCX1 and reduced expression of SERCA2a in the LV. *P < 0.05; **P < 0.01 vs. NF (ANOVA). Data regraphed from Ref. and used with permission. Ba: a left ventricular (LV) myocardial infarction (MI ≈ 45% of LV) was induced in C57Bl/6 mice (n = 5) by ligation of the left anterior descending coronary artery; controls (n = 5) were subjected to sham surgery. Four weeks after surgery, the LV ejection fraction, determined by echocardiography, had declined by >33% in the operated mice, indicative of HF, and plasma EO was significantly increased. b: NCX1 expression was increased and SERCA2 (likely SERCA 2a) was decreased in the LV of mice in HF, as in humans (Ab). Expression of SERCA2 (probably the 2 b isoform, which predominates in vascular smooth muscle) as well as NCX1 was increased in the aortae of the mice in HF (also see Fig. 6). *P < 0.05; **P < 0.01; ***P < 0.001 vs. sham (ANOVA). Data in Ba and Bb from Ref. used with permission.

Figure 11.

A and B: endogenous ouabain is elevated in pregnancy and plays a role in human fetal development. A: serum endogenous ouabain-like compound (eOLC) was measured in nonpregnant women and in women 1 wk before delivery, at the time of delivery, and 1 day after delivery. eOLC, which was measured by ELISA with selective anti-ouabain antibodies, was elevated during pregnancy and declined at the time of delivery and in the early postpartum period. *P < 0.05; **P < 0.01 vs. eOLC in nonpregnant women; n = 11–19 per group. B: neonatal birth weight correlates with maternal serum eOLC collected just before delivery: eOLC is lower in mothers of small-for-gestational-age neonates (2.0–2.5 kg) than in mothers of normal-for-gestational-age neonates (3.0–3.5 and 3.5–4.0 kg). *P < 0.05, n = 10–30/group. Data in A and B from Ref. regraphed and used with permission. C–F: decreasing endogenous ouabain (EO) in rats lowers neonatal birth weight and impairs organ development. C: in rats, treatment of dams with anti-ouabain antibodies (Ab, 10 mg/kg/day) on days 9–18 of pregnancy leads to reduced neonatal birth weight and slower gain of weight by the pups for at least the first 21 postnatal days. *P < 0.05 vs. offspring of dams injected with IgG (10 mg/kg/day); n = 22–35. D: offspring from anti-ouabain Ab-treated and IgG-treated dams were euthanized at age 17 days. The relative weights (organ wt/body wt) of the kidneys and livers of pups from anti-ouabain Ab-treated dams were less, and heart weights were greater than those from the offspring of control (IgG-treated) dams; *P < 0.05; n = 9–12. Data in C and D from Ref. were regraphed and used with permission. E: development of rat kidneys is impaired when they are cultured in fetal bovine serum-deprived media (0.2% FBS); adding ouabain to the media rescues the kidneys. [Note: FBS contains EO (224).] Embryonic day 14 (E14) explanted kidneys were cultured for 72 h in media containing 10% FBS; in some cultures, during the last 24 h, the medium was replaced with one containing only 0.2% FBS, either without or with 10 nM ouabain. Serum deprivation reduced the number of glomeruli in the kidneys by about 40%, but not if that medium was supplemented with 10 nM ouabain. **P < 0.01; n = 9–15. F: rats fed a low-protein diet (9% vs. normal 18% protein) throughout pregnancy gave birth to pups with abnormally low numbers of renal glomeruli; this loss was prevented by infusing ouabain into the dams (15 µg/kg/day via subcutaneous osmotic minipump) during the pregnancy. **P < 0.01; n = 8 per group. Data in E and F from Ref. regraphed and used with permission.