© 2001 by European Society of Cardiology
Copyright © 2000, European Society of Cardiology
The importance of calcium in interpretation of NaK-ATPase isoform function in the mouse heart
Institute of Molecular Pharmacology and Biophysics, University of Cincinnati College of Medicine, 231 Bethesda Avenue, Cardiovascular Research Center, Cincinnati, OH 45267-0828, USA
* Corresponding author. Tel.: +1-513-5582-400; fax: +1-513-5581-778
Received 2 April 2001; accepted 23 April 2001
KEYWORDS Calcium (cellular); Na/K-pump
We have recently emphasized an important role for intracellular calcium in changes that occur in the adrenergic system in cardiac hypertrophy and failure in the mouse heart [1,2]. For example, the blunting or loss of function of the β-adrenergic receptor signaling pathway that has been described for a number of mouse transgenic models can be restored to normal by "unloading" the heart of calcium [2]. It is well known that the isolated rat heart responds in a perfectly normal manner to a variety of pharmacological agents including digitalis and to physiological stimuli, but only when the perfusion media contains "low" calcium, ranging from 0.25 to 1.0 mM [3–7].
Less is known about the isolated mouse heart although most agree that the calcium requirements for pharmacological and physiological studies are the same as the rat [5].
In the case of the NaK-ATPase (NKA) three 
isoforms have been described. The adult heart contains all three, the adult rat heart expresses 1 and 2and the adult guinea pig heart expresses only the 1. The subcellular distribution for these isoforms has been described [8]. In terms of characteristics, all three isoforms are quite similar in kinetics, affinities for the monovalent ions, Mg2+ and ATP. There are some differences described in isolated heart studies for the relative affinities of cardiac glycosides, but nothing, in our opinion, of such magnitude to have pharmacological or toxicological significance. For some years, there has been much discussion about different functions for these isoforms and conjecture about co-localization or not of specific isoforms with the Na/Ca exchange protein. The latter has bearing on a concept of a putative endogenous ouabain-like factor as well as assigning a specific role for one isoform to digitalis-induced positive inotropy and for another isoform to the toxic actions of cardiac glycosides [9,10]. The Blaustein group has adumbrated a "plasmerosome" concept for smooth muscle in which the above aspects are emphasized [9].
In this article we wish to show that while attractive, the evidence that a particular isoform of the (NKA) is NOT connected to contractile effects in the heart is flawed.
In a recent paper by James et al. [10], a conclusion was reached that the two isoforms of the (NKA) in the mouse heart are functionally and geographically distinct. The conclusion was based on physiological and pharmacological assessment of the hearts of phenotypes with genetically reduced levels of the 1- and 2-isoforms, respectively. Further, the authors concluded that it is only the 2-isoform that physically interacts with the cellular machinery responsible for calcium regulation and that only when this isoform is inhibited by ouabain, a cardiac glycoside, will one observe the usual positive inotropic effect. Fig. 5 of this paper [10] is a cartoon that clearly illustrates the separation of these two isoforms with the 1-isoform assigned a task of "overall electrolyte balance of Na and K", but not that of calcium. Further, the authors state that reduction of the 1-isoform
...results in a cardiac phenotype resembling aspects ofcardiac glycoside toxicity (decreased cardiac contractility). Therefore, cardiac glycoside toxicity may be due, at least in part, to inhibition of theThe authors also state that
...concentrations of ouabain that positively effect contractility in the
The conclusions of this paper are questionable. The animals were engineered correctly and the molecular analyses of the isoform mRNA and protein levels are correct. A difficulty lies in the physiological and pharmacological analyses of the phenotypes and of the wild type leading to, in our opinion, an incorrect compartmentation model for isoform-specific functions [10].
In the paper [10], a concentration–response to ouabain on contraction of the wild type and the 1 (±) using an isolated working (perfused) heart preparation is described. The concentration of calcium used in the perfusates generally was 2.0 mM, although in one experiment 1.5 mM was used. It is well known that both rat and mouse hearts are unique compared to all other species in that intracellular Na is higher [3,5]; at a concentration of calcium in the range used by the authors, both species exhibit a negative force-frequency, a very short action potential and a marked insensitivity to cardiac glycosides [3,6,7]. All of the latter are due to a major difference in the distribution of intracellular calcium, with most of the calcium derived from the SR in rat and mouse heart as opposed to all other mammalian species. Further, upon stimulation or twitches, the rat and mouse hearts accrue calcium while the rabbit, guinea pig, etc., extrude calcium [5]. Therefore, the mouse and rat hearts are already "loaded" with calcium during stimulation. In order to measure normal physiological and pharmacological responses, it is necessary to utilize much lower concentrations of calcium in perfusing media [4,6]. It is not surprising therefore that a monophasic curve was obtained [10] because the so-called "low dose" effect is masked by the high calcium.
In Fig. 1A and B illustrated below, we present a concentration–response for ouabain on contraction of a wild-type mouse perfused with a solution containing 0.75 mM [Ca2+] and in separate experiments 1.5 mM [Ca2+]. Although not shown, this heart at 0.75 mM [Ca2+] in the perfusate displays positive staircase, is quite stable and reacts to a variety of pharmacological interventions in a robust manner. In particular, we observed a clear two-phase effect, one at extremely low concentrations of ouabain clearly in the range of the binding constant referred to in the James et al. paper [10], and the second phase produced by higher concentrations of ouabain. Thus, at the appropriate ionic conditions, the drug affects both isoforms in an identical manner, obviously inhibiting the NaK-ATPase and producing the characteristic time-dependent increase in contraction.
|
It is highly unlikely that cardiac glycosides affect any system other than the NaK-ATPase. It is possible that at very low calcium, ouabain causes ingress of calcium via Na channels as described by Santana et al. [11], according to a "slip-mode conductance". Even if correct the explanation for the latter still invokes an interaction with the NKA which in turn somehow affects a Na channel so that an increase in [Ca2+]i occurs [11]. Another possibility is that very low concentrations of ouabain (EC50 of 0.4 nM) activates the ryanodine receptor, RyR2, and may serve as an amplification factor for glycoside actions (J.A. Wasserstrom, personal communication).
It is hard to imagine, although it is always possible, that simply lowering the cellular calcium would produce a sudden re-distribution of putative "compartmented" isoforms and unmask a particular species. In any case, the conclusions reached by James et al. [10] and, in particular the cartoon in Fig. 5 [10] are misleading. Further, cardiac glycoside toxicity cannot be assigned to a particular isoform based on these experiments since no toxicity was observed. The authors are incorrect in suggesting that "depressed contractility" is a sign of toxicity due to digitalis. Toxicity is generally regarded as due to arrhythmicity and elevation of serum K+ levels and neither of these were observed by these authors. In point of fact, the so-called "hypocontractile" animals were only 80% of the wild-type and the "hypercontractile" animals were only 12% greater than wild-type. In the case of the latter, there was a remarkable 50% increase in amplitude of the calcium transient, and in the former, no change in the calcium transient. In fact the baseline calcium of both isoform animals was the same. This means that diastolic calcium was unchanged. The contractile changes, compared to the wild-type, as described for these animals, in our opinion, seem to be minimal at best and do not match up with the calcium data [10].
It is of interest that the authors cite a paper by themselves (Askew et al., 1993 in Ref. [10]) in which the well known biphasic contractile effect is illustrated and assigned to both isoforms, a conclusion opposite to the one they present in James et al. [10]. Another citation in [10] [Grupp et al., 1985, the present author (A.S.) for some reason excluded] clearly shows the relationship between inhibition of the pump by ouabain in rat corresponding to an increase in intracellular Na which leads to an increase in [Ca2+]i. In a paper by Adams et al. [12] the well-known biphasic effect of ouabain is illustrated along with a high affinity-binding site (KD, 124 nM for rat heart). Noteworthy is that in the latter study [12], the concentration of calcium employed in the bathing medium was 1.8 mM and the lowest concentration of ouabain that produced a statistically significant effect on contractility was 100 nM. Again, the "two contractile effects" of ouabain was attributed to two isozymes of Na, K-ATPase. Hickerson et al. [13] described an isolated rat heart perfused with a solution containing [Ca2+] of 0.5 mM, in which ouabain produced a clear biphasic inotropic response with the lowest concentration of drug at 10 nM and again both responses were attributed to the two isoforms of the NaK-ATPase. Taken together, the data provide good evidence that lower concentrations of calcium in the perfusing medium allows for observation of the "two-inotropic" effects of ouabain in the mouse heart but at even much lower concentrations of drug than previously reported well within the KD cited by James et al. [10].
So that the overall message should not be lost, we suggest that calcium plays a vital role in physiological and pharmacological action in the rat and murine heart. It is possible that there are subtle differences in the distribution of the two isoforms of the NKA in the mouse but the functions assigned by these authors and the models depicted are questionable. Clearly, both isoforms can link with the cellular machinery required to produce a positive inotropic effect when challenged by a cardiac glycoside. Thus, there is no reason at this point to ascribe a function for any isoform of the NKA in which a cardiac glycoside when applied does not lead to a positive inotropic effect [14].
N.B.: One of the authors (A.S.) brought this paper to the attention of the editorial staff of the new journal, Molecular Cell in which the James et al. [10] paper was published and after much discussion elected not to accept this "opinion" paper.
|
 |
Acknowledgements |
|---|
The studies reported herein were supported by NIH P01 HL22619 (A.S.) and by an NIH training grant K-32 HL07382 (A.S., N.N.P.).
|
References |
|---|
 Top References |
|---|
- Muth J.N, Bodi I, Lewis W, Varadi G, Schwartz A. A Ca2+-dependent transgenic model of cardiac hypertrophy: a role for protein kinase C. Circulation (2001) 103:140–147.
[Abstract/Free Full Text] - Serikov V.B, Petrashevskaya N.N, Canning A.M, Schwartz A. Reduction of [Ca2+] restores uncoupled β-adrenergic signaling in isolated perfused transgenic mouse hearts. Circ. Res. (2001) 88:9–11.
[Abstract/Free Full Text] - Shattock A.J, Bers D.M. Rat vs. rabbit ventricle: Ca flux and intracellular Na assessed by ion-selective microelectrodes. Am. J. Physiol. (1989) 256:C813–C822.[Web of Science][Medline]
- Forrester G.V, Mainwood G.W. Interval dependent inotropic effects in the rat myocardium and the effect of calcium. Plugers Arch. (1974) 352:189–196.[CrossRef][Web of Science][Medline]
- Bers D.M. Calcium fluxes involved in control of cardiac myocyte contraction. Circ. Res. (2000) 87:275–281.
[Free Full Text] - Chevalier B, Mouas C, Mansier C, Aumont M.C, Swynghdauw B. Screening of inotropic drugs on isolated rat and guinea pig hearts. J. Pharmacol. Methods (1987) 17:313–326.[CrossRef][Web of Science][Medline]
- Capogrossi M.C, Kort A.A, Spurgeon H.A, Lakotta E.G. Single adult rabbit and rat cardiac myocytes retain the Ca- and species-dependent systolic and diastolic contractile properties of intact muscle. J. Gen. Physiol. (1986) 88:589–613.
[Abstract/Free Full Text] - McDonough A.A, Zhang Y, Shin V, Frank J.S. Subcellular distribution of sodium pump isoform subunits in mammalian cardiac myocytes. Am. J. Physiol. (1996) 270:C1221–C1227.[Web of Science][Medline]
- Blaustein M.P, Juhaszova M, Golovina V.A. The cellular mechanism of action of cardiotonic steroids: a new hypothesis. Clin. Exp. Hypertens. (1989) 20:691–703.
- James P.F, Grupp I.L, Grupp G, Woo A.L, Askew G.R, Croyle M.L, et al. Identification of a specific role for the Na,K-ATPase 2-isoform as a regulator of calcium in the heart. Mol. Cell (1999) 3:555–563.[CrossRef][Web of Science][Medline]
- Santana L.F, Gomez A.M, Lederer W.J. Ca flux through promiscuous cardiac Na channels: slip-mode conductance. Science (1998) 279:1027–1033.
[Abstract/Free Full Text] - Adams R.J, Schwartz A, Grupp G, et al. High-affinity ouabain binding site and low-dose positive inotropic effect in rat myocardium. Nature (1982) 296:167–169.[CrossRef][Medline]
- Hickerson T.W, Grupp I.L, Schwartz A, Grupp G. The calcium dependence of the biphasic inotropic response to ouabain in the isolated rat heart. Prog. Clin. Biol. Res. (1988) 258:223–234.[Medline]
- Schwartz A, Allen JC, Lindenmayer GE. The sodium–potassium adenosine triphosphatase: pharmacological, physiological and biochemical aspects. Pharmacol Rev 1975:3–134.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||