(Hypertension. 1997;29:815-821.)
© 1997 American Heart Association, Inc.
Articles |
Sodium Pump Isoform Specificity for the Digitalis-Like Factor Isolated From Human Peritoneal Dialysate
the Departments of Medicine (Endocrine-Hypertension Division) and Radiology, Harvard Medical School, Brigham and Women's Hospital, Boston, Mass.
Correspondence to Dr Steven W. Graves, Endocrine-Hypertension Division, Brigham and Women's Hospital, 221 Longwood Ave, Boston, MA 02115. E-mail swgraves@bics.bwh.harvard.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
We have isolated a labile, specific sodium pump inhibitor or digitalis-like factor from the peritoneal dialysate of volume-expanded renal failure patients whose levels correlated closely with volume status and blood pressure. This study characterizes the inhibitory profile of this agent compared with that of ouabain against the three
-isoforms of the sodium pump. We prepared microsomal Na,K-ATPase from rat tissues representing the highest proportion of one of the -isoforms. Both Northern and Western blot analyses confirmed that kidney had predominantly the 1-isoform, skeletal muscle the 2-isoform, and fetal brain the 3-isoform. Ouabain (5x10-6 mol/L) produced little inhibition of kidney Na,K-ATPase (3.4±2.0%) but significant inhibition of skeletal muscle (37.2±3.7%, P<.001) and fetal brain (38.8±3.5%, P<.001) activity. In contrast, the labile digitalis-like factor, causing comparable inhibition of fetal brain Na,K-ATPase activity (33.3±4.7%), produced markedly greater inhibition of kidney (42.5±5.6%, P<.001) and moderately greater inhibition of skeletal muscle pump activity (57.7±6.3%, P<.05). In addition, the labile digitalis-like factor produced a marked concentration-dependent inhibition of the 2- and 3-isoforms (r=.79, P=.00005). Experiments combining the labile digitalis-like factor and ouabain confirmed that digitalis-like factor, unlike ouabain, was an effective inhibitor of all three isoforms in rat, in particular 2. The different pattern of isoform sensitivity displayed by the labile digitalis-like factor and ouabain further differentiates the two agents and raises some interesting possibilities about the functional implications of the endogenous factor.
Key Words: mRNA • Western blot analysis • ouabain • Na+,K+-transporting ATPase • sodium pump • digitalis-like factor
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Maintenance of an intracellular milieu low in sodium ion and high in potassium ion is central to eukaryotic cell function and depends on a ubiquitous membrane enzyme, the Na,K-ATPase or sodium pump.1 During the past decade, it has become increasingly clear that there are both tissue and species differences in the distribution of isoforms of the catalytically active
-subunit of the sodium pump and that this distribution might have functional implications.2 3 4 5 In some tissues, such as the kidney, the dominant or sole form of the isoenzyme is 1.1 2 In a number of tissues, including the heart and skeletal muscle, both 1- and 2-isoforms are present.6 7 An additional isoform, termed 3, is prominent in brain.8 9 The three -isoforms vary in their affinity for sodium and their rate of activity. Moreover, they manifest a differential sensitivity to digitalis glycosides, such as ouabain; these differences were among the first lines of evidence for multiple forms of the -subunit.1 6
Several observations, including the tight conservation of the digitalis binding site over many phyla, have suggested that there may be an endogenous sodium pump inhibitor that can modulate catalytic activity, a premise now supported by substantial evidence.10 11 12 13 14 15 16 Any such agent is predicted to be digitalis-like in its mode of action and may participate in normal sodium homeostasis or in the pathogenesis of sodium-sensitive hypertension.10 11 14 15 16 Such a role remains controversial17 18 although there is considerable support from many lines of investigation.10 11 12 13 14 15 16 Intriguingly, evidence is emerging to suggest that the individual -subunit digitalis binding sites have had their distinct binding characteristics conserved to afford selective modulation of the activity of a given isoform in vivo as part of physiology and pathophysiology.2 6 This may reflect more than one endogenous inhibitor, each having differential effects on the three isoforms, and allow for regional or local control.19
This study examines the possibility that different sodium pump inhibitors may have a unique influence on the different isoforms. We compared the influence of ouabain and an endogenous sodium pump inhibitor or digitalis-like factor (DLF) that we have isolated from the peritoneal dialysate of volume-expanded renal failure patients20 21 on three different tissues, each chosen because it represented the tissue with the greatest abundance of a particular isoform. We chose rat tissue in part because of the marked difference in sodium pump isoform affinity for the cardiac glycosides, including ouabain, to increase the chances of finding differences.2 6 As indexes, we used both inhibition of ouabain binding to the sodium pump and inhibition of Na,K-ATPase hydrolysis. Our findings confirmed the relative resistance of the 1-isoform to the plant-derived digitalis glycoside ouabain. The endogenous ligand, on the other hand, was far more effective than ouabain in blocking the 1-isoform, while also producing marked inhibition of 2- and 3-isoform activities. The effect on 2 appears greater than the effect on 3. An action of the endogenous agent on the 1-isoform, which has a crucial influence on transport and metabolism in many organs, may have important physiological and pathological implications.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Microsome Preparation
All tissues were obtained from pregnant Sprague-Dawley rats on day 18 of gestation. All experimental protocols using rats were reviewed and approved by the Committee on Animal Experimentation at Harvard Medical School in compliance with National Institutes of Health guidelines on the use of animals in research. Rat tissues (maternal kidney and hindlimb skeletal muscle as well as fetal brain) were removed and frozen immediately after death and bleeding of the rats and were stored at -70°C until use. For preparation of microsomal Na,K-ATPase, kidney and fetal brain were minced with a razor and then homogenized with six strokes of a glass-polytetrafluoroethylene homogenizer at medium speed in ice-cold medium A (mmol/L: sucrose 250, histidine 30, EDTA 1, and imidazole 5, pH 7.25). The homogenate was centrifuged at 10 000g for 20 minutes at 4°C. The supernatant was then collected and subjected to ultracentrifugation (100 000g, 1 hour at 4°C). The resultant pellets were suspended in medium A and adjusted to a final concentration of 2 mg protein/mL. The microsomal preparations were then stored at -20°C. Skeletal muscle microsomal Na,K-ATPase fraction was prepared by the method of Rogus et al22 with some modification. Briefly, the tissue was minced with scissors in medium B (mmol/L: sucrose 250, histidine 24, EDTA 4, pH 6.8, containing 0.08% [wt/vol] deoxycholate), homogenized with a tissue disrupter (Tissue Tearor, Biospec Products) at setting 3 (medium speed), and then homogenized (six strokes at medium speed) with a glass-polytetrafluoroethylene homogenizer. The homogenate was centrifuged at 8500g for 20 minutes. The supernatant was collected and stored at -20°C overnight and then thawed and ultracentrifuged at 100 000g for 35 minutes. The pellet was resuspended in medium C (mmol/L: sucrose 250, EDTA 1, Tris 125, pH 7.2) by brief homogenization and then centrifuged at 20 000g for 50 minutes. The pellet was resuspended in medium A and stored at -20°C until use. The microsomal preparations at the time of use had the following total and ouabain-sensitive activities, respectively: fetal brain, 4.8 and 3.5 µmol Pi/min per milligram protein (74%); skeletal muscle, 14.0 and 9.4 (67%); and kidney, 17.9 and 15.6 (88%).
Northern Blot Analysis
Total RNA was isolated by homogenization of 50 to 100 mg tissue in 1 mL of TRI-Reagent (Molecular Research Center). The homogenate was centrifuged at 16 000g for 20 minutes at 4°C, and RNA was isolated from the clear supernatant according to the manufacturer's protocol. RNA was dissolved in FORMAzol and stored at -70°C until analysis. For Northern analyses, 7 or 10 µg RNA was loaded in each lane of a multilane gel (1% agarose) and separated by electrophoresis and transferred to a nylon membrane (Schleicher & Schuell). The cDNA probes (gifts of Dr Jerry Lingrel, University of Cincinnati) were labeled with [32P]dCTP and added to the hybridization solution to achieve an activity of 2x106 cpm/mL. The membrane was prehybridized, hybridized, and washed sequentially. Hybridization was carried out in (final concentration) 5x SSC and 50% formamide solution at 42°C for approximately 18 hours. Northern blots were exposed to x-ray film for autoradiography and were analyzed with a laser scanning densitometer (Personal Densitometer, Molecular Dynamics). Values for the individual samples were corrected for differences in the total added RNA by dividing the optical density of the -subunit band by the optical density of the corresponding individual ribosomal 18s rRNA band.
Western Blot Analysis
A constant amount of rat microsomal Na,K-ATPase (20 µg protein) was subjected to 7% sodium dodecyl sulfate–polyacrylamide gel electrophoresis. After protein separation on the gel, proteins were transferred electrophoretically to nitrocellulose membranes. The protein bands on a finished gel were then probed for a single sodium pump -isoform using -subunit–specific monoclonal antibodies (McK1, a monoclonal antibody specific for the 1-isoform, and McB2, a monoclonal antibody specific for the 2-isoform were gifts of Dr Kathleen J. Sweadner (Harvard Medical School); the antibody specific for the 3-isoform was purchased from Affinity Bioreagents Inc). A second antibody-coupled chemiluminescent detection method (Enhanced Chemiluminescence, Amersham) was used to register the -isoform–first antibody complex. This procedure was performed according to the manufacturer's protocol. Blots were exposed to x-ray film (X-OMAT AR, Eastman Kodak), and the signal density was measured with a laser scanning densitometer (Personal Densitometer, Molecular Dynamics).
Isoform Distribution
To validate our assumptions about isoform distribution and confirm previous reports, we determined isoform expression in different tissues. The tissues used were selected because they represented the tissues having the greatest abundance of a particular isoform of the sodium pump. We determined the distribution of the sodium pump -isoforms by Northern analysis of the selected tissues (Table
). In kidney, 1 was the dominant, if not exclusive, form of the sodium pump. In skeletal muscle, there was approximately 95% 2 and approximately 5% 1, similar to previous reports.23 In brain, 3 mRNA represented approximately 40% of the total, with 1 and 2 each accounting for about half of the remaining 60%. Tissue-specific expression of Na,K-ATPase isoform protein was determined by Western analysis and was generally consistent with the mRNA results (Table).
|
Preparation of DLF from Peritoneal Dialysate
DLF was isolated from the peritoneal dialysate of several volume-expanded renal failure patients sampled on multiple occasions and was purified as described in detail elsewhere.20 Only one dialysate from a single patient was used for a given set of experiments. Briefly, the purification of the DLF included ultrafiltration as a first step to exclude molecular species greater than 1000 D. The factor was isolated from the ultrafiltrate by solid-phase extraction with a short preparative C18 reversed-phase high-performance liquid chromatographic (HPLC) column and was subsequently eluted with a small volume of absolute methanol. This was dried, and the redissolved residue was introduced onto a semipreparative C18 HPLC column followed by gradient elution. The location of the DLF, previously established,20 remained constant. The appropriate HPLC eluate was collected, the organics were removed under vacuum, and the active residue was submitted to analytical C18 HPLC purification using different elution parameters. The final product, the location of which had been previously determined, was collected, dried completely to remove all organics, and redissolved in a small volume of pure water or assay buffer. The agent was used immediately because of the inherent chemical lability of this DLF.24 The quantity or molar concentration of this DLF was not known.
Measurement of Na,K-ATPase Activity
The Na,K-ATPase activity of the individual microsomal preparations was measured as [
-32P]ATP hydrolysis as described previously.21 25 In brief, microsomal Na,K-ATPase (20 µg protein per tube) was incubated in 150 µL buffer (containing in mmol/L: sodium 100, potassium 5, magnesium 3, EDTA 1, Tris 80 [pH 7.5], and ATP 4, spiked with [-32P]ATP; final specific activity, approximately 70 mCi/mol) for 30 minutes at 37°C. The reaction was terminated by addition of 850 µL of a solution of 40 g/L charcoal and 0.1N HCl. Ouabain-sensitive activity was defined as the activity inhibitable by 10 mmol/L ouabain. DLF or submaximal concentrations of ouabain were assessed by addition to the enzyme of the inhibitor in 10 µL buffer, followed by a 30-minute incubation at 37°C before the initiation of hydrolysis with the addition of ATP.
Ouabain Binding
Ouabain binding to microsomal Na,K-ATPase was measured as the counts of radioactive [3H]ouabain bound in the presence of magnesium and PO43- and absence of potassium. Microsomal Na,K-ATPase was incubated with [3H]ouabain and unlabeled ouabain or DLF in 200 µL of a solution containing (mmol/L) magnesium 5, Pi 5, and Tris 50 (pH 7.2) for 3 hours at 37°C. Free ouabain was separated from bound ouabain by centrifugation (200 000g for 25 minutes at 4°C). In each experiment, the nonspecific binding, determined in the presence of 1 mmol/L cold ouabain, was less than 5% of total radioactivity bound to the enzyme and was subtracted.
Comparison of Endogenous DLF With Ouabain
For both ouabain and DLF, assays were carried out simultaneously with microsomal preparations of fetal brain, muscle, and kidney. As opposed to ouabain, for which a wide range of known concentrations could be studied, we used a different approach for the endogenous DLF: Each HPLC-purified preparation of DLF, obtained from a single patient sample, was split into three equal aliquots and assayed against the three tissues. As part of each study, a single concentration of ouabain was assessed simultaneously. Because brain contains all three isoforms, we used it as our reference tissue and used a ouabain concentration of 5x10-6 mol/L because it produced approximately 30% to 50% inhibition of brain Na,K-ATPase in preliminary studies.
Initially, we attempted to isolate the activity of one isoform by using a monoclonal antibody specific to a second isoform present to block inhibitor access. We chose skeletal muscle because it has both 1- and 2-isoforms. We incubated monoclonal antibody against 1 (McK1; final titer, 1:500) or independently monoclonal antibody against 2 (McB2; final titer, 1:80) with 20 µg microsomal enzyme for 60 minutes and assayed Na,K-ATPase activity as described above. Thereafter, ouabain (5x10-6 mol/L) or DLF—enough to produce approximately 30% to 50% inhibition—was added in the presence or absence of antibody and activity was measured. The effects of the antibody on enzyme activity in the absence of inhibitor were also assessed.
Because this approach proved unsuccessful, we developed two different approaches to assess the relative influence of an agent on each of the three isoforms.
Isolation of 1-Isoform Activity
Even though kidney contains almost exclusively the 1-isoform, the absence of tissues containing exclusively the 2- or 3-isoform required that we be able to "isolate" the contribution of 1 to the overall activity and identify the effects of each inhibitor on 1 in tissues containing other isoforms. The previously reported finding that the 1-isoform is markedly less sensitive to ouabain than are the 2- and 3-isoforms allowed for alternative approaches that "isolated" the effect of DLF on either 1 or 2 and 3. The first of these approaches used ouabain at 5x10-6 mol/L, a concentration that inhibits almost all of the 2- and 3-isoforms of the pump but very little (<10%) of the 1. This in effect leaves only the 1 available. The effects of DLF on the 1-isoform could then be assessed in the three tissues. Ouabain alone (5x10-6 mol/L) or equal amounts of a single DLF preparation alone or equal portions of DLF in combination with this ouabain concentration were tested against the three tissues. Inhibition due to the addition of DLF to ouabain thus reflects the effects of DLF on 1.
2- and 3-Isoform Isolation
The second of these approaches looked at the effect of DLF on the 2- and 3-isoforms by isolating their response. This was accomplished by means of the ouabain binding assay but in this case using a very low concentration of labeled ouabain (5x10-9 mol/L). At this concentration, ouabain is bound only to the 2- and 3-isoforms, and the displacement of labeled ouabain by the addition of another sodium pump inhibitor, which binds to the digitalis binding site, would represent an interaction with only the 2- or 3-isoform. Brain microsomal Na,K-ATPase was used because of the presence of 2 and 3. The addition of increasing amounts of cold ouabain or DLF was assayed to test the ability of each inhibitor to displace the [3H]ouabain tracer.
Statistics
Data are expressed as mean±SE. Student's t test was used for comparison of the effects of DLF with those of ouabain for normally distributed data. ANOVA was used for three-way comparisons among the three tissues. Paired data were statistically analyzed with the Pearson product-moment correlation test. A value of P=.05 was considered significant. For complete dose-response curves, data were plotted as the line of best fit.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Ouabain-Induced Inhibition of Na,K-ATPase Activity
Microsomal preparations of the three tissues were exposed to graded concentrations of ouabain to establish its concentration dependence (Fig 1
). Increasing ouabain concentrations produced a monophasic inhibition of Na,K-ATPase activity in kidney microsomes, with an IC50 of 5x10-5 mol/L, consistent with the inhibition of a single isoform, 1, which was relatively ouabain resistant.
|
), skeletal muscle (
), and fetal brain (
) were preincubated with graded concentrations of ouabain in buffer containing (mmol/L) sodium 100, potassium 5, and magnesium 3 for 3 hours. The reaction was started by addition of [32P]ATP and ended by addition of a 4% charcoal/0.1N HCl solution after 30 minutes. Data represent means of six sets of experiments for each tissue preparation.
In contrast, the effect of ouabain on the microsomal Na,K-ATPase activity from skeletal muscle was biphasic, representing at least two different binding sites of differing affinities. Of these two responses, the inhibition occurring at relatively high ouabain concentrations represented approximately 30% of total Na,K-ATPase activity and had an IC50 identical to that of rat kidney. The second occurred at lower ouabain concentrations and represented a more ouabain-sensitive isoform (2, IC50 approximately 2x10-8 mol/L), comprising approximately 70% of the total activity.
Fetal brain likewise exhibited a biphasic response to increasing ouabain concentration. Since rat fetal brain contains all three isoforms, our results confirmed that the ouabain affinities of the 2- and 3-isoforms are close and jointly represented approximately 60% of the total Na,K-ATPase activity in this experiment.8 9
Effect of DLF on the Different Isoforms
The effect of DLF on tissue Na,K-ATPase was compared with a single concentration of ouabain (5x10-6 mol/L), producing approximately 30% to 50% inhibition of rat brain Na,K-ATPase. Although responses to this ouabain concentration differed greatly among the three tissues, being least for kidney, responses for a single concentration of DLF were similar for brain and kidney (Fig 2). In rat fetal brain, Na,K-ATPase inhibition was not significantly different for this dose of ouabain and the average DLF present (DLF, 33.3±4.7% versus ouabain, 38.8±3.5%; P=NS). However, DLF in matched aliquots and ouabain (5x10-6 mol/L) produced significantly different inhibition of rat kidney (DLF, 42.5±5.6% versus ouabain, 3.4±2.0%; P<.001) and skeletal muscle (DLF, 57.7±6.3% versus ouabain, 37.2±3.7%; P<.05) Na,K-ATPase activity (Fig 2). DLF produced significantly more inhibition against skeletal muscle than it did against brain (muscle, 57.7±6.3% versus brain, 33.3±4.7%; P=.02), whereas the effect of ouabain against these two tissues was not different.
|
1-Isoform Isolation
The significantly reduced affinity of rat 1-isoform for cardiac glycosides provided a means by which 1 activity could be assessed independently of ("isolated from") 2 and 3 activities. Ouabain (5x10-6 mol/L) inhibited more than 90% of both 2- and 3-isoforms, leaving only the 1-isoform available for inhibition by a second specific inhibitor (see Fig 3, top), in this case DLF.
|
The inhibitory effects of a single concentration of DLF in the absence and presence of 5x10-6 mol/L ouabain on 1 activity are shown in Fig 3
, bottom, and represent the mean of six experiments. In kidney, as anticipated, ouabain alone had very little effect, whereas DLF alone demonstrated significantly greater inhibition (DLF, 32.1±2.7% and ouabain, 7.9±1.9%; P<.0001). When the two agents were applied to the kidney together, inhibition was additive but not significantly different from that with DLF alone (DLF plus ouabain, 37.8±0.9%). In the other tissues, ouabain by itself produced a marked inhibition, and the combination of ouabain and DLF produced an additive effect (fetal brain: DLF, 23.8±3.0%; ouabain, 39.5±6.1%; DLF plus ouabain, 74.9±4.1%; skeletal muscle: DLF, 63.3±10.1%; ouabain, 45.8±3.4%, DLF plus ouabain, 98.0±2.0%), providing additional evidence for the 1-isoform having a higher affinity for the endogenous inhibitor DLF than for ouabain.
Furthermore, these experiments confirm those summarized in Fig 2 in that DLF alone produced significantly more inhibition of skeletal muscle than of fetal brain (skeletal muscle, 63.3±10.1% versus brain, 23.8±3.1%; P=.02).
2- and 3-Isoform Isolation
We then assessed the ability of DLF to bind to the 2- and 3-isoforms. This time we studied inhibition of [3H]ouabain binding by either cold ouabain or DLF. By use of a low concentration of [3H]ouabain (5x10-9 or 10x10-9 mol/L), ouabain will bind to the 2- and 3-isoforms; thus, any inhibition of labeled ouabain binding by DLF represents its competition for the 2- and/or 3-subunit. DLF produced a significant concentration-dependent inhibition of labeled ouabain binding from fetal brain microsomes (Fig 4). To determine whether the 1-isoform contributed to the observed ouabain binding, we repeated the experiments with rat kidney microsomes. Binding to kidney 1-isoform could not be demonstrated at this [3H]ouabain concentration, confirming that DLF was bound to 2 and 3.
|
When matched aliquots of DLF and ouabain were assayed by inhibition of both activity and ouabain binding, the relationship between binding and functional inhibition could be assessed. As shown in Fig 5, the effect of paired results of DLF specimens on ouabain binding and on inhibition of Na,K-ATPase activity (rat fetal brain) demonstrated a positive linear relationship whose slope differed significantly from that of ouabain (DLF: y=1.95x+3.6, r=.99, P=5x10-5; ouabain: y=0.61x+6.1, r=.99, P=2x10-4).
|
) were assayed in both the binding (with 5x10-9 mol/L [3H]ouabain) and Na,K-ATPase hydrolysis assay as described in "Methods." Regression lines of best fit are shown for both DLF and ouabain. Data points for DLF represent multiple DLF preparations. |
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Our goal was to ascertain whether an endogenous sodium pump inhibitor, obtained from volume-expanded patients, and ouabain would display identical or different patterns of sodium pump isoform inhibition. The patterns were markedly different: The endogenous agent was not only an efficient inhibitor of the
2- and/or 3-isoforms but also the 1-isoform of rat. Given the importance of the 1-isoform as a determinant of sodium transport in the kidney and elsewhere, this observation has potentially important functional implications. The endogenous sodium pump inhibitor used here has been previously shown to be the only inhibitor present in human peritoneal dialysate whose activity paralleled serum DLF levels, increased with volume expansion, and correlated with blood pressure.20 Several detailed characterizations of this DLF have demonstrated that it acts exclusively on the digitalis binding site of the sodium pump and on no other ATPase, producing inhibition by reducing the dephosphorylation rate of the catalytic cycle of the pump analogous to digitalis.20 21 24 25 DLF is a nonpeptide of intermediate polarity and is easily distinguished from lipids, fatty acids, common steroids, and cardiac glycosides by its HPLC mobility.26 Significantly, it is chemically labile, unlike cardiac glycosides.24 The HPLC fraction used here contained no other chemical species active against the sodium pump, even after two additional purification steps.27
The rat was selected because its sodium pump isoforms are well characterized, and isoform-specific mRNA riboprobes and monoclonal antibodies were available.7 8 An added reason was the marked difference in ouabain sensitivity among the sodium pump isoforms, optimizing the chances for detection of isoform-specific differences between ouabain and this DLF.2 6 On the other hand, our findings may not extend to other species, including the human, the source of our DLF.
We considered isolating individual isoforms6 for direct testing of DLF specificity but were concerned that purification might alter binding characteristics. Other researchers have modified insect cells to overexpress a single isoform but obtained isoforms with specific activity very much lower than that in their normal environment.28 Although there are no isoform-specific antagonists, we considered a comparable approach: We attempted to use monoclonal antibodies specific to a single isoform to block its respective digitalis binding site; however, neither the 1- nor 2-antibody interfered with the inhibitory action of ouabain or DLF on muscle Na,K-ATPase activity. For these reasons, we adopted the less direct approaches used in the present study to allow the sodium pump to be studied in a minimally perturbed membrane environment.
The tissue distribution of the sodium pump -isoforms, measured as either mRNA abundance or expressed protein levels, was comparable. The few differences seen for 3 may be due to the inherent quantitative imprecision of either Northern or Western analysis related to variability in probe affinity from one isoform to the next or to other factors. Some differences, especially those in brain, may reflect the inhomogeneity of isoform distribution and variable sampling of particular regions. Differences may be real; ie, individual isoform expression may not reflect the mRNA abundance, as has been found in other studies.29 30 Nevertheless, the findings confirm that one tissue was predominantly 1 (kidney), 2 (skeletal muscle), and 3 (fetal brain), although in brain there was also 1 and 2.
We confirmed the relative resistance of the 1-isoform2 and relative sensitivity of 2- and 3-isoforms to ouabain.1 4 5 8 30 The methods used here were not sufficiently sensitive to detect differences in ouabain affinity for the 2- and 3-isoforms, although 3 has been reported to have up to a 30-fold higher affinity.31 DLF clearly did not show 1 resistance (Fig 2 and Fig 3, bottom). Because we have no definitive measure of DLF concentration, the quantities available being too small to weigh or produce a UV spectrum, it was possible that all three -isoforms were actually resistant to DLF. For ouabain, the amount needed to produce 50% inhibition of rat fetal brain in six replicate assay tubes, as done for some experiments here, was 47 µg. DLF from a single purification produced comparable or greater inhibition. When the amount of DLF from single purifications was measured by supercritical fluid chromatography coupled with flame ionization detection, we consistently found less than 10 ng, often substantially less.27 Therefore, these findings are consistent with DLF having a high affinity for all isoforms. The results in Fig 3 confirm those of Fig 2 but further demonstrate that the effects of DLF and ouabain are additive when ouabain is present in submaximal concentrations. Previous studies demonstrated that DLF produced no added inhibition when a maximal concentration of ouabain was used, even in tissues with substantial non–Na,K-ATPase activity.21
In addition to the profound effect of DLF on the 1-isoform, it was shown here for the first time to be an effective inhibitor of the 2- and/or 3-isoform (Fig 4). Unfortunately, the limited amounts of DLF precluded full dose-response curves. However, the data in Fig 2 suggest that DLF does have a differential affinity for the different isoforms: If one focuses on the brain and skeletal muscle data, one finds that ouabain (5x10-6 mol/L) produced equivalent inhibition. This ouabain concentration blocks all the 2- and 3-isoforms in brain and all the 2-isoform in skeletal muscle. The balance of the activity—that not inhibited—is due to 1, which was equivalent for these two tissues. In these same tissues, however, DLF produced significantly more inhibition of skeletal muscle than brain. This suggests a preferential effect of DLF on 2 over 3. This is also suggested by the experiments reported in Fig 3, bottom, for the DLF alone.
The data presented in Fig 5 also suggest a selective difference of DLF on 2 or 3 but do not indicate which. When graded concentrations of ouabain were assayed simultaneously for their effect on [3H]ouabain binding and on Na,K-ATPase hydrolysis, a tight correlation was found. This denotes the anticipated close coupling of ouabain binding to the pump, with inhibition of pump activity. However, because all three isoforms were present and the maximal concentration of ouabain used was limited (5x10-6 mol/L), the slope of the line of best fit was not 1.0 but approximately 0.6. This was anticipated: Cold ouabain at this concentration would displace all the labeled ouabain (100%, since none was bound to 1) but produce only 50% inhibition (because it would block only the 2- and 3-isoform activities). The effect of DLF differed. Although the correlation between displacement of ouabain and inhibition of activity was strong, suggesting that inhibition results from DLF binding to the digitalis receptor, the slope was 1.9, indicating that a greater proportion of activity was lost than ouabain binding inhibited. This is also the anticipated outcome because DLF inhibited 1, which represents proportionately more catalytic activity per pump unit than the other isoforms. On the basis of the data in Fig 2, 1 accounts for approximately 60% to 65% of the total activity, and DLF would inhibit this without affecting [3H]ouabain binding (since none is bound to 1). In addition, DLF was able to displace 50% of the labeled ouabain. Hence, this DLF concentration was adequate to inhibit at least an additional approximately 20% of the total Na,K-ATPase activity if 2 and 3 have equal catalytic rates. The final 20% inhibition may result from DLF preferentially inhibiting one isoform, which had greater catalytic activity. It may represent assay conditions: Ouabain bound to 3 has a dissociation rate 33 times slower than ouabain bound to 231 ; hence, it is possible that DLF competition with ouabain for 3 was not at equilibrium at the end of the 30-minute assay period, ie, that the DLF concentration may ultimately have been sufficient to replace more of the labeled ouabain. Alternatively, the inhibition assay requires no competition for the binding site, and the inhibition by DLF of the same 3 Na,K-ATPase hydrolysis may have occurred very rapidly compared with its dissociation rate. Hence, these data probably also reflect differences in isoform interaction with DLF.
Can inhibition of the sodium pump produce pathology? The answer is unresolved. Studies have demonstrated that long-term ouabain administration causes hypertension in rats.32 33 This may be due to a peripheral effect on vascular smooth muscle, a central nervous system effect, or both.34 In vitro, vascular smooth muscle responds to ouabain with sustained contraction.35 The DLF obtained from peritoneal dialysate produces the same effect,20 36 and preliminary data (not shown) suggest that this effect occurs at concentrations substantially lower than the concentration of ouabain needed for a comparable effect. Digoxin therapy in general does not require progressively increasing doses to maintain a constant therapeutic effect,37 and yet there is evidence that cultured cells exposed to ouabain increase sodium pump number,38 perhaps because of increased mRNA levels in response to increased cell ionized sodium.39 It is also possible that inhibition of one isoform of the catalytically active -subunit may have more pronounced effects than inhibition of another. Thus, it is possible that DLF may have a different effect than ouabain on sodium pump mRNA abundance.
In summary, these experiments demonstrate that DLF isolated from the peritoneal dialysate of volume-expanded patients with chronic renal failure, unlike ouabain, acts as a potent inhibitor of the 1-isoform of the sodium pump and suggest a preferential effect on 2 over 3. The varying effects of DLF and ouabain on the three isoforms suggest the potential for regional and local modulation of the sodium pump by circulating factors.
|
Acknowledgments |
|---|
We would like to thank Marion Merrell Dow for partial support of this research. We also express our gratitude to Dr Kathleen Sweadner (Boston) and Dr Jerry Lingrel (Cincinnati) for their generous gifts of molecular biological probes.
Received February 9, 1996;
first decision April 2, 1996;
first decision September 16, 1996;
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Lingrel JB, Orlowski J, Shull MM, Price EM. Molecular genetics of Na,K-ATPase. Prog Nucleic Acid Res. 1990;38:37-89.[Medline] [Order article via Infotrieve]
2.
Sweadner KJ. Two molecular forms of (Na+ + K+)-stimulated ATPase in brain: separation and difference in affinity for strophanthidin. J Biol Chem. 1979;254:6060-6067.
3.
Shull MM, Lingrel JB. Multiple genes encode the human Na+,K+-ATPase catalytic subunit. Proc Natl Acad Sci U S A. 1987;84:4039-4043.
4.
Emanuel JR, Garetz S, Stone L, Levenson R. Differential expression of Na+,K+-ATPase - and ß-subunit mRNAs in rat tissues and cell lines. Proc Natl Acad Sci U S A. 1987;84:9030-9034.
5.
Herrera VLM, Emanuel JR, Ruiz-Opazo N, Levenson R, Nadal-Ginard B. Three differentially expressed Na,K-ATPase subunit isoforms: structural and functional implications. J Cell Biol. 1987;105:1855-1865.
6.
Sweadner KJ, Farshi SK. Rat cardiac ventricle has two Na+,K+-ATPases with different affinities for ouabain: developmental changes in immunologically different catalytic subunits. Proc Natl Acad Sci U S A. 1987;84:8404-8407.
7.
Sweadner KJ, Herrera VLM, Amato S, Moellmann A, Gibbons DK, Repke KRH. Immunologic identification of Na+,K+-ATPase isoforms in myocardium. Circ Res. 1994;74:669-678.
8.
Shull GE, Greeb J, Lingrel JB. Molecular cloning of three distinct forms of the Na+,K+-ATPase -subunit from rat brain. Biochemistry. 1986;25:8125-8132.[Medline]
[Order article via Infotrieve]
9. Schneider JW, Mercer RW, Gilmore-Hebert M, Utset MF, Lai C, Greene A, Benz EJ Jr. Tissue specificity, localization in brain, and cell-free translation of mRNA encoding the A3 isoform of Na+,K+-ATPase. Proc Natl Acad Sci U S A. 1985;85:284-288.
10. Buckalew VM Jr, Haddy FJ. Circulating natriuretic factors in hypertension. In: Brenner BM, Laragh JH, eds. Hypertension: Pathophysiology, Diagnosis, and Management. New York, NY: Raven Press Publishers; 1990:939-954.
11. Hamlyn JM, Manunta P. Ouabain, digitalis-like factors and hypertension. J Hypertens. 1992;10(suppl 7):99-111.
12. Goto A, Yamada K, Yagi N, Yoshioka M, Sugimoto T. Physiology and pharmacology of endogenous digitalis-like factors. Pharmacol Rev. 1992;44:377-399.[Medline] [Order article via Infotrieve]
13. MacGregor SE, Walker JM. Mini review. Inhibitors of the Na+,K+-ATPase. Comp Biochem Physiol. 1993;105S:1-9.
14. Schoner W. Endogenous digitalis-like factors. Prog Drug Res. 1993;41:249-281.[Medline] [Order article via Infotrieve]
15. Woolfson RG, Poston L, de Wardener HE. Digoxin-like inhibitors of active sodium transport and blood pressure: the current status. Kidney Int. 1994;46:297-309.[Medline] [Order article via Infotrieve]
16. Graves SW. The sodium pump in hypertension. Curr Opin Nephrol Hypertens. 1994;3:107-111.[Medline] [Order article via Infotrieve]
17.
Kelly RA, Smith TW. The search for the endogenous digitalis: an alternative hypothesis. Am J Physiol. 1989;256:C937-C950.
18. Wechter WJ, Benaksas EJ. Natriuretic hormones. Prog Drug Res. 1990;34:232-260.
19. Hollenberg NK, Graves SW. Endogenous sodium pump inhibition: current status and therapeutic opportunities. Prog Drug Res. 1996;46:9-42.[Medline] [Order article via Infotrieve]
20. Glatter KA, Graves SW, Hollenberg NK, Soszynski PA, Tao QF, Frem GJ, Williams GH, Lazarus JM. Sustained volume expansion and [Na,K]ATPase inhibition in chronic renal failure. Am J Hypertens. 1994;7:1016-1025.[Medline] [Order article via Infotrieve]
21. Tao Q, Soszynski PA, Hollenberg NK, Graves SW. Specificity of the volume-sensitive sodium pump inhibitor isolated from human peritoneal dialysate in chronic renal failure. Kidney Int. 1996;49:420-429.[Medline] [Order article via Infotrieve]
22.
Rogus E, Price T, Zierler KL. Sodium plus potassium-activated, ouabain-inhibited adenosine triphosphatase from a fraction of rat skeletal muscle, and lack of insulin effect on it. J Gen Physiol. 1969;54:188-202.
23.
Herrera VLM, Chobanian AV, Ruiz-Opazo N. Isoform-specific modulation of Na+,K+-ATPase -subunit gene expression in hypertension. Science. 1988;241:221-223.
24. Graves SW, Soszynski PA, Tao QF, Williams GH, Hollenberg NK. A labile endogenous Na-pump inhibitor. In: Bamberg E, Schoner W, eds. The Sodium Pump: Structure, Mechanism, Hormonal Control, and Its Role in Disease. New York, NY: Springer; 1994:771-774.
25. Tao QF, Soszynski PA, Hollenberg NK, Graves SW. A sensitive [Na,K]ATPase assay, specific for inhibitors acting via the digitalis-binding site. J Cardiovasc Pharmacol. 1995;25:859-863.[Medline] [Order article via Infotrieve]
26. Graves SW, Glatter KA, Lazarus JM, Williams GH, Hollenberg NK. Volume expansion in renal failure patients: a paradigm for a clinically relevant [Na,K]ATPase inhibitor. J Cardiovasc Pharmacol. 1993;22(suppl 2):S54-S57.
27. Xie LQ, Markides KE, Lee ML, Hollenberg NK, Williams GH, Graves SW. Bioanalytical applications of the multidimensional open tubular column supercritical fluid chromatography. Chromatographia. 1993;35:363-371.
28.
Blanco G, Xie ZJ, Mercer RW. Functional expression of the 2 and 3 isoforms of the Na,K-ATPase in baculovirus-infected insect cells. Proc Natl Acad Sci U S A. 1993;90:1824-1828.
29.
Ewart HS, Klip A. Hormonal regulation of the Na+,K+-ATPase: mechanisms underlying rapid and sustained changes in pump activity. Am J Physiol.. 1995;269:C295-C311.
30. Sweadner KJ. Multiple digitalis receptors: a molecular perspective. Trends Cardiovasc Med. 1993;3:2-6.
31. O'Brien WJ, Lingrel JB, Wallick ET. Ouabain binding kinetics of the rat alpha two and alpha three isoforms of the sodium-potassium adenosine triphosphate. Arch Biochem Biophys. 1994;310:32-39.[Medline] [Order article via Infotrieve]
32.
Sekihara H, Yazaki Y, Kojima T. Ouabain as an amplifier of mineralocorticoid-induced hypertension. Endocrinology. 1992;131:3077-3082.
33.
Yuan CM, Manunta P, Hamlyn JM, Chen S, Bohen E, Yeun J, Haddy FJ, Pamnani MB. Long-term ouabain administration produces hypertension in rats. Hypertension. 1993;22:178-187.
34.
Huang BS, Leenen FHH. Brain ouabain and central effects of dietary sodium in spontaneously hypertensive rats. Circ Res. 1992;70:430-437.
35.
Stewart L, Hamilton C, Ingwall J, Naomi S, Graves S, Canessa M, Williams G, Hollenberg N. Vascular smooth muscle response to ouabain: relation of tissue sodium to the contractile response. Circ Res. 1993;73:1113-1120.
36. Krep HH, Graves SW, Price DA, Lazarus M, Ensign A, Soszynski PA, Hollenberg NK. Reversal of sodium pump inhibitor induced vascular smooth muscle contraction with digibind: stoichiometry and its implications. Am J Hypertens. 1995;9:39-46.
37. Schmidt TA, Allen PD, Colucci WS, Marsh JD, Kjeldsen K. No adaptation to digitalization as evaluated by digitalis receptor (Na,K-ATPase) quantification in explanted hearts from donors without heart disease and from digitalized recipients with end-stage heart failure. Am J Cardiol. 1993;7:110-114.
38. Yamamoto K, Ikeda U, Seino Y, Tsuruya Y, Oguchi A, Okada K, Ishikawa S, Saito T, Kawakami K, Hara Y, Shimada K. Regulation of Na,K-adenosine triphosphatase gene expression by sodium ions in cultured neonatal rat cardiocytes. J Clin Invest. 1993;92:1889-1895.
39.
Aperia A, Ibarra LB, Svensson CK, Greengard P. Calcineurin mediates -adrenergic stimulation of Na,K-ATPase activity in renal tubule cells. Proc Natl Acad Sci U S A. 1992;89:7394-7397.
This article has been cited by other articles:
 |
|
 |
|
  J. A. Wasserstrom and G. L. Aistrup Digitalis: new actions for an old drug Am J Physiol Heart Circ Physiol, November 1, 2005; 289(5): H1781 - H1793. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. M.A.M. Qazzaz, Z. Cao, D. D. Bolanowski, B. J. Clark, and R. Valdes Jr De Novo Biosynthesis and Radiolabeling of Mammalian Digitalis-Like Factors Clin. Chem., March 1, 2004; 50(3): 612 - 620. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Shimoda, J. Jasleen, J. D. Rounds, S. W. Ashley, and D. O. Jacobs Sepsis Increases the Plasma Membrane Content of {alpha}1 and {alpha}2 Isoforms of Na+-K+ Adenosine Triphosphatase in Rat Skeletal Muscle Arch Surg, January 1, 2001; 136(1): 95 - 100. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. W. Graves, K. E. Markides, and N. K. Hollenberg Application of Supercritical Fluid Chromatography to Characterize a Labile Digitalis-Like Factor Hypertension, December 1, 2000; 36(6): 1059 - 1064. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. M. A. M. Qazzaz, M. A. El-Masri, and R. Valdes Jr. Secretion of a Lactone-Hydrogenated Ouabain-Like Effector of Sodium, Potassium-Adenosine Triphosphatase Activity by Adrenal Cells Endocrinology, September 1, 2000; 141(9): 3200 - 3209. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q.-F. Tao, N. K. Hollenberg, and S. W. Graves Sodium Pump Inhibition and Regional Expression of Sodium Pump {alpha}-Isoforms in Lens Hypertension, November 1, 1999; 34(5): 1168 - 1174. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. N. Orlov, N. C. Adragna, V. A. Adarichev, and P. Hamet Genetic and biochemical determinants of abnormal monovalent ion transport in primary hypertension Am J Physiol Cell Physiol, March 1, 1999; 276(3): C511 - C536. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Y. Bagrov, O. V. Fedorova, R. I. Dmitrieva, W. N. Howald, A. P. Hunter, E. A. Kuznetsova, and V. M. Shpen Characterization of a Urinary Bufodienolide Na+,K+-ATPase Inhibitor in Patients After Acute Myocardial Infarction Hypertension, May 1, 1998; 31(5): 1097 - 1103. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||