1 Research Centre, Centre Hospitalier de lUniversité de Montreal, Hôtel Dieu, Montreal, Quebec, H2W 1T8, Canada
2 Department of Biomembranes, Faculty of Biology, Moscow State University, 119899 Moscow, Russia
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
quantitative trait loci; genetic hypertension; telemetry
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Atp1a1 encodes the 1-isoform of the Na-K-ATPase subunit ubiquitously expressed in all tissues studied so far (10, 18). It is implicated in the BP regulation, because the
1-Na-K-ATPase is the sole active Na+ transporter located on the basolateral membrane of renal epithelial cells (12), which plays a pivotal role in the extracellular fluid volume adjustment (9, 23). An elevated Na+:K+ stoichiometry of this enzyme could be sufficient to maintain an augmented transcellular salt and fluid transport via a membrane hyperpolarization and an activation of electrogenic ion carriers (16). In addition to its implication in the kidney function, it was also suggested that an inhibition of the
1-Na-K-ATPase by endogenous ouabain or ouabain-like factors could maintain the elevated vascular smooth muscle tone and peripheral resistance (1).
The role of the 1-Na-K-ATPase in genetic hypertension of the S rat remains controversial. Herrera and Ruiz-Opazo (13) sequenced cDNA clones from kidney libraries and found an adenine (A) to thymidine (T) transversion at the nucleotide 1079 in the Atp1a1 coding region of the S rat compared with the Dahl salt-resistant rat (R). This mutation led to the G276L substitution in the enzyme. However, Kurtz and coworkers (21) could not reproduce this result. Later on, Ruiz-Opazo and coworkers (20) argued that the negative result obtained by Kurtz and coworkers might be caused by a consistent Tag PCR error that selectively substituted A to T-1079. Subsequently, Herrera and coworkers (14) developed transgenic S rats bearing the
1-Na-K-ATPase cDNAs from the R rat. These rats exhibited smaller salt-sensitive hypertension. In the same study, they also found a linkage between the alleles of the Atp1a1 locus and BP variations in an F2(S x R) population. In contrary, Rapp and Dene (19) detected no such linkage while studying another independently derived F2(S x R) population. Zicha and coworkers (28) found no linkage between Atp1a1 polymorphisms and BP in an F2(S x R) population, although BP of these F2 hybrids correlated positively with ouabain-sensitive Na+ extrusion and Na+ content in their erythrocytes. Recently, Harris and coworkers (11) carried out a site-directed mutagenesis to artificially introduce a mutation of the A-to-T base change in the Atp1a1 coding region. They found that the S rat does not have the mutation at all compared with the positive control that does (11). Therefore, the mutation detected by Herrera and coworkers (13) remains to be duplicated by other investigators.
Conflicting results were also obtained in the studies of the 1-Na-K-ATPase activity in salt-sensitive hypertension. No systematic differences in the rate of ouabain-sensitive Na+ and K+ fluxes were observed by Zicha and Duhm (27) while comparing the erythrocytes from S and R rats fed on the low-salt diet. A high-salt diet led to a rise in the ouabain-sensitive Na+ and K+ fluxes in S rats only (26). In the Xenopus laevis oocyte expression system, cells carrying the total RNA from the S kidneys or cDNAs derived from the in vitro transcribed Atp1a1-mRNA from S exhibited a decreased ouabain-sensitive 86Rb influx and non-altered ouabain-sensitive 22Na efflux compared with R rats (13). The same differences in the ouabain-sensitive ion fluxes in erythrocytes were observed comparing the S and R rats (2). Based on these results, Herrera et al. (14) speculated that G276L substitution could lead to an elevated stoichiometry of the
1-Na-K-ATPase (3Na+:1K+ in S vs. 3Na+:2K+ in R).
Regardless of the controversy over the base difference between S and R, Atp1a1 was a candidate for a BP QTL potentially determining an allelic difference between S and MNS, because a substitution of the S chromosome region containing Atp1a1 by that of the MNS lowered BP (4). To further examine the candidacy of Atp1a1, we constructed two congenic strains, each by replacing a fragment of the S chromosome by that of MNS (8). The obvious difference between them is one congenic strain is homozygous SS, whereas the other is homozygous MM, at the Atp1a1 locus. We then compared the effect of the salt diet on BP and the activity of the erythrocyte Na/K pump in these congenic strains with those of the S strain. Since the 1-Na-K-ATPase is the only isoform expressed in mammalian erythrocytes (10, 18) we used these cells to examine the effect of the Atp1a1 substitution on the
1-Na-K-ATPase activity and its stoichiometry. Other cell types such as vascular muscle cells would have other isoforms of the Na-K-ATPase in addition to the
1-isoform (15).
![]() |
METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
To study the isotope influx rate, after 3 h of preincubation, the cells were centrifuged, and the supernatant was aspirated. Then, 0.2 ml of medium B containing 4 µCi/ml 22Na or 0.5 µCi/ml of 86Rb with or without ouabain was added. It is well-documented that the 1-isoform of Na-K-ATPase from rodents possesses an extremely low affinity for ouabain compared with other species (24). Indeed, in preliminary experiments, we found that an elevation in the ouabain concentration from 2.5 to 10 mM increased the ouabain-sensitive 86Rb influx by 1015% (data not shown). However, because of possible side effects of ouabain in a high concentration on the membrane integrity and on ion transporters other than the Na/K pump (see below), 2.5 mM ouabain was used in our study. In the absence of ouabain, the kinetics of isotope influx and efflux is linear up to 2030 min (data not shown). Based on these results, the incubation time was limited to 15 min, and the isotope uptake was terminated by an addition of 1 ml ice-cold medium C containing 100 mM MgCl2 and 10 mM HEPES-Tris buffer (pH 7.4). The erythrocytes were centrifuged, washed twice with the ice-cold medium C, and lysed by a mixture of 0.5 ml of 0.5% Triton X-100 and 0.5 ml of 10% trichloroacetic acid (TCA). These samples were centrifuged (2,000 g, 10 min), and the protein-free lysate was transferred into 10 ml of scintillation cocktail.
To measure the isotope efflux, 50 µl of cells preincubated during 5 h in 0.2 ml of medium B containing 22Na or 86Rb were mixed with 1 ml of ice-cold medium B, centrifuged, and washed twice with the same ice-cold medium. Then, 1 ml of medium B with or without 2.5 mM ouabain was then added. After an incubation of 15 min at 37°C, these samples were centrifuged, and 0.8 ml of the supernatant was measured for radioactivity with a liquid scintillation analyzer (model TR-1600; Canberra-Packard Canada, Mississauga, Ontario, Canada).
To measure the total content of the intracellular Na+ and K+, after 3 h of incubation in medium B, erythrocytes were sedimented, washed three times with ice-cold medium C, and lysed by Triton-TCA as indicated above. Aliquots of the protein-free supernatant were diluted with water and subjected to flame photometry (Flapho-4, Carl Zeiss, Germany).
The stoichiometry of the Na/K pump (S) was calculated as S = (VOSNaE - VOSNaI) x (VOSKI - VOSKE)-1, where VOSNaE and VOSKI are values of ouabain-sensitive (subscript "OS") components of Na+ efflux (efflux indicated by subscript "E") and of K+ influx (influx indicated by subscript "I"), respectively. These parameters were also used as a measure of the Na/K pump activity. VOSNaI and VOSKE are values of ouabain-sensitive components of Na+ influx and K+ efflux, respectively; these values were used to evaluate the Na+/Na+ (VOSNaI) and K+/K+ (VOSKE) mode of operation of the Na/K pump (7). To determine VOSNaI and VOSKI, the rate of K+ (86Rb) and Na+ (22Na) influx in the absence and presence of ouabain was measured and calculated as V = A x (amt)-1, where A is the radioactivity of cell lysate (in cpm), a is the specific radioactivity of 86Rb or 22Na in the incubation medium (in cpm/nmol), m is the content of erythrocytes in the samples (in liters), and t is the time of incubation (0.25 h).
The calculation of absolute values for the rate of isotope efflux is complicated because of a lack of information on the content of intracellular exchangeable K+ and Na+, which makes it difficult to determine the values of the intracellular specific radioactivity of 86Rb and 22Na. To overcome this problem, we assumed that under steady-state conditions the total values of inward and outward ion fluxes are the same, i.e., VNaE = VNaI and VKE = VKI, where V is the rate of Na and K efflux and influx in the absence of ouabain. This assumption is applicable for our study because the same conditions were used to measure the rates of isotope influx and efflux (17). Based on the above assumption, VOSNaE and VOSKE were calculated as VOSNaE = VNaI x (ANat - ANao) x ANat-1, and VOSKE = VKI x (ARbt - ARbo) x ANat-1, where ANat, ARbt, ANao, and ARbo are the radioactivity of 22Na (subscript Na) or 86Rb (subscript "Rb") in the incubation medium after 15 min of incubation without (subscript "t") and with ouabain (subscript "o"), respectively. The values of the ouabain-resistant (subscript "OR") Na+ and K+ efflux were calculated as VORNaE = VNaI x ANao x ANat-1, and VORKE = VKI x ARbo x ARbt-1. All measurements of ion fluxes were done in triplicates. Isotopes and ouabain were purchased from Amersham (Mississauga, Ontario, Canada) and Sigma (St. Louis, MO), respectively. The remaining chemicals were obtained from Sigma, GIBCO-BRL (Life Technologies, Gaithersburg, MD) and Anachemia (Montreal, Quebec, Canada).
Statistical analysis.
Repeated measures ANOVA followed by Dunnett in the SYSTAT 9 program (SPSS Sci., Chicago, IL) was used to compare the significance level for a difference or a lack of it in BP between a congenic strain and the S strain. To analyze ion flux data, means and standard errors were calculated and statistical significance was assessed by a Students t-test for unpaired data.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
BP measurements and comparisons.
For the simplicity of presentation, only the averaged values of 24 h for the day when the pump activity was analyzed were included. For a more detailed presentation of BPs for the congenic strains, please see Ref. 8. The mean arterial pressures (MAPs) for the strains are given to show the BP effect in a congenic strain. The variations in both systolic and diastolic pressures were the same (data not shown). MAPs were significantly lowered (P < 0.0001) in the S.M5 (n = 5) and S.M6 (n = 7) congenic strains compared with that of the S (n = 9) strain.
Effect of salt diet on ion transport in erythrocytes from S.
Neither the intracellular Na+ nor K+ content measured in erythrocytes incubated for 3 h in medium B was different in rats fed on high- and low-salt diet (data not shown). Table 1 shows that none of the components of ion fluxes in erythrocytes from the S rats was affected by the salt diet. These negative results are consistent with previous data of other investigators (27). Also in line with the work of Canessa and coworkers (2), a low-salt diet was used in the further studies on ion transports in erythrocytes from congenic rats.
|
|
Ouabain-resistant ion fluxes in erythrocytes.
While measuring ion fluxes in the presence of the Na-K-ATPase inhibitor ouabain, we did not notice any differences between S.M5 and S rats (Fig. 3). However, the rates of ouabain-resistant Na+ influx and efflux were both increased by 5080% in S.M6 rats compared with S and S.M5 rats.
|
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Although MNS rats per se were not directly analyzed for either BP or the pump activity, S.M6 served as an effective control because it contained a small segment from the MNS rat not including Atp1a1. If any gene from the MNS genome adjacent to, but excluding Atp1a1, affected the pump activity, then S.M6 should have provided the adequate information. S.M6 as a control is important in another respect. As S.M6 was derived from the same progenitor congenic strain S.M (8) as S.M5, any BP effects of the residual MNS background as well as potential position effects resulting from making the congenic strains had to be ruled out (3). The MNS rat would not be a good control, because it differs greatly from S in the genome on the rest of the chromosomes and because it gives no information on the effects of genome changes during the congenic construction. If one could observe a difference in the 1-Na-K-ATPase activity between MNS and S, then this difference could not be a test for the Atp1a1 gene alone. It could be due to many other genes from MNS modifying it such as the adducin gene (22). Of course, if there were no differences, then the MNS genome other than Atp1a1 might be influencing the outcome.
The rates of ouabain-resistant Na+ influx and efflux mediated by ion transporters distinct from the Na-K-ATPase were both increased in S.M6 compared with S and S.M5 (Fig. 3). As an explanation, an elevated ouabain-resistant Na+ flux could be considered as a mechanism for the increase in an intracellular Na+ content in S.M6, which in turn could lead to an augmented ouabain-sensitive Na+ efflux (Fig. 2) by a feedback activation of the Na-K-ATPase. Because the values of the ouabain-resistant K+ influx and efflux in congenic and S rats were the same, it appeared unlikely that the Na-K-Cl cotransporter and a nonselective monovalent ion leakage across the erythrocyte membrane contribute to the elevated ouabain-resistant Na+ fluxes in S.M6 rats. Because our experiments were conducted in the bicarbonate-free medium, the impact of NaCO3-/Cl- exchanger seemed also unlikely. One aspect is certain: this increase in the ouabain-resistant Na+ fluxes in S.M6 is not due directly to the Atp1a1 gene, because S.M6 has the same Atp1a1 alleles as S.
The intriguing fact is that S.M6 has a shorter MNS chromosome segment than S.M5. When the nonoverlapping region was removed, which contained the Atp1a1 gene and was unique to S.M5, the rates of ouabain-resistant Na+ influx and efflux went up. It seemed that some gene(s) in this unique segment from the MNS rat actually modified the permeability of erythrocyte membranes to Na+. This phenomenon might involve gene-gene interactions/modifications between the Atp1a1 alleles from the S rat and another unknown modifying gene(s) of the MNS rat. Alternatively, the unidentified residual MNS genetic background might be different between S.M5 and S.M6 and consequently might have influenced the Na-K-ATPase activity differently. In our genomic scan, we found 88 markers to be SS homozygous for both S.M5 and S.M6 strains (data not shown). Thus a new congenic strain by replacing the MNS Atp1a1 alleles with those of the S rat might have to be constructed to examine this Atp1a1-independent phenomenon.
It is noted that possible altered properties of the 1-isoform of the Na/K pump of the S rat were based on two observations from the same research team, which showed decreased values of the ouabain-sensitive K+ influx in the Xenopus laevis oocytes transfected with RNAs from the S kidneys (13) and in the erythrocytes from the S rat compared with the R rat (2). Because the values of the ouabain-sensitive Na+ influx were not altered, it was speculated that the stoichiometry of the
1-isoform of the Na/K pump in the S rat could be 3Na+:1K+ compared with a "normal" 3Na+:2K+ ratio of this transporter in the R rat (14). Nevertheless, the data obtained in our study indicated that the stoichiometry of the Na-K-ATPase was not affected by the substitution of the Atp1a1 alleles from the S rats by these of the MNS rats (Fig. 4), despite a decrease in BP as the result of it. In addition, the values of stoichiometry in both congenic strains as well as in S rats were in the range from 1.4 to 2.0, which is closer to the 1.5 value of the "normal" ratio rather than to the 3.0 value predicted from the 3Na+:1K+ stoichiometry. These observations are not consistent with the notion of a functional mutation in the
1-Na-K-ATPase of the S rat.
Because both intracellular Na+ and extracellular phosphate affect the stoichiometry of the erythrocyte Na-K-ATPase (7), different conditions used in separate experiments could be a factor for the differing outcomes. Canessa and coworkers (2) measured a unidirectional 22Na and 86Rb influx and efflux. Afterwards, the total content of intracellular monovalent cations was used to calculate the specific intracellular activity of isotopes (2), which might not allow a precise measurement of the Na+:K+ stoichiometry (see METHODS). Moreover, they used the phosphate-free medium and Na+-loaded K+-depleted cells to study the isotope efflux. In our studies, we used cells with physiologically low intracellular Na+/K+ ratio, and the medium contained inorganic phosphate at the concentration corresponding to its level in rat plasma (2.5 mM).
It is noteworthy that an addition of 2.5 mM ouabain led to an elevation of the Na+ influx by 25%, which resulted in the negative values of ouabain-sensitive component of Na+ influx used to evaluate the Na+/Na+ mode of operation of the Na-K-ATPase (Table 1, Fig. 2). Presumably, the addition of ouabain could activate other Na+ transport pathways via an interaction with the inactivated Na-K-ATPase. This hypothesis is supported by the recent data on the production of oxygen reactive species in ouabain-treated cardiac myocytes independently of the intracellular Na+/K+ ratio (25). However, the side effect of a high concentration of ouabain on Na+ transporters other than the Na-K-ATPase cannot be excluded. An inhibition of the ouabain-sensitive Na+ efflux extruding a small portion of the intracellular 22Na could also lead to this phenomenon.
In conclusion, our data showed that neither the 1-Na-K-ATPase activity nor its stoichiometry was affected by the substitution of the Atp1a1 alleles of S by those of MNS. Thus the BP-lowering effects observed in S.M5 and S.M6 could not be attributed to the
Na-K-ATPase activity or its stoichiometry. Atp1a1 is not supported as a candidate to be a BP QTL.
![]() |
ACKNOWLEDGMENTS |
---|
This work was supported by grants from American Heart Association (AHA) National Center, from the Heart and Stroke Foundation of Canada (Quebec), and from Canadian Institutes for Health Research (CIHR) (to A. Y. Deng), from the Kidney Foundation of Canada (to S. N. Orlov), and by a CIHR grant (to P. Hamet). A. Y. Deng is an Established Investigator of the AHA. J. Dutil is a holder of the CIHR studentship.
![]() |
FOOTNOTES |
---|
Address for reprint requests and other correspondence: A. Y. Deng, Research Centre, Centre Hospitalier de lUniversité de Montreal, 7-132 Pavillon Jeanne Mance, 3840, rue St. Urbain, Montreal, Quebec, Canada H2W 1T8 (E-mail: alan.deng{at}umontreal.ca).
10.1152/physiolgenomics.00059.2001.
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|