Na+-H+ Antiporter Phenotype, Abundance, and Phosphorylation of Immortalized Lymphoblasts From Humans With Hypertension
Abstract Previous studies have demonstrated an elevated Na+-H+ exchanger activity in various cell types from patients with essential hypertension. The phenotype of an increased maximal transport capacity is preserved in Epstein-Barr virus immortalized lymphoblasts from hypertensive patients. The mechanisms underlying this abnormality are unclear. In this study, we used lymphoblasts from hypertensive patients and normotensive control subjects with and without a family history of hypertension to determine (1) Na+-H+ exchanger activity using fluorometry with the pH indicator 2′,7′-bis(carboxyethyl)-5(6)-carboxyfluorescein, (2) Na+-H+ exchanger isoform 1 abundance with specific polyclonal antibodies, and (3) Na+-H+ exchanger phosphorylation by immunoprecipitation of the 32P-labeled transporter. Na+-H+ exchanger activity (in millimoles per liter per minute) measured when pHi was clamped at 6.0 was significantly higher in cells from hypertensive patients (18.8±0.6, P<.001) and those subjects with a family history of hypertension (16.4±0.6, P<.001) compared with normotensive control subjects (12.9±0.6). Exchanger abundance was identical in all three groups of subjects, indicating that increased activity in the hypertensive group was due to an elevated turnover number of the exchanger. Na+-H+ exchanger phosphorylation in quiescent cells was significantly elevated in cells from hypertensive patients (1.58±0.16, P<.001) compared with control subjects (1.00±0.07), and cells from normotensive subjects with a hypertensive family history showed intermediate values (1.23±0.14). Identical changes in Na+-H+ exchanger function and phosphorylation have been demonstrated in vascular smooth muscle cells from spontaneously hypertensive rats. Our findings suggest that the elevated Na+-H+ exchanger activity in cells from human hypertensive patients is not associated with an increased exchanger abundance but may be related to increased exchanger phosphorylation.
A variety of ion transport abnormalities have been reported in hypertension in both animals and humans. One of these is an elevation of Na+-H+ exchanger (NHE) activity (for reviews, see References 11 through 33 ). An increased NHE activity has been demonstrated in vascular smooth muscle cells from spontaneously hypertensive rats4 5 and in leukocytes,6 red blood cells,7 and platelets8 9 from humans with hypertension. Furthermore, direct evidence of increased NHE activity in vivo has been obtained by nuclear magnetic resonance studies.10 Kinetic studies have indicated an increased Vmax of NHE,5 6 7 although the cellular mechanisms underlying these changes (eg, altered transporter number, turnover, or posttranslational modification) have not been elucidated.
Since the cloning of the ubiquitous NHE, named isoform 1 or NHE-1,11 other members of this growing family have been described.12 13 14 These isoforms—NHE-2, NHE-3, and NHE-4—are less sensitive to inhibition by amiloride derivatives and have a more restricted distribution, which may be consistent with a role in transepithelial Na+ transport. However, the changes described in NHE activity in leukocytes from human hypertensive (HT) patients are sensitive to ethylisopropyl amiloride, and only NHE-1 mRNA transcripts are reported in lymphoid cells.12 13 Thus, the earlier reports on increased NHE activity in hypertension may reflect either altered NHE-1 expression or its posttranslational processing.
The changes of NHE activity may have been attributed to environmental factors. Evidence against this was obtained in Epstein-Barr virus immortalized lymphoblasts from HT patients,15 in which the phenotypic changes persisted despite transformation and culture in vitro. This increased Vmax of NHE thus may be genetically determined but was not associated with any difference in the mRNA transcripts of NHE-1,15 implying that there was no difference in NHE-1 expression. Cellular NHE-1 protein content was not directly determined in this in vitro cell culture model of hypertension.15
Another possible mechanism is an increased NHE activity related to an elevated NHE-1 turnover number. This may result from increased phosphorylation of NHE-1, because this posttranslational modification has been documented to activate Na+-H+ exchange.16 17 18 In the present study, we therefore used the pH-sensitive fluorophore 2′,7′-bis(carboxyethyl)-5(6)-carboxyfluorescein (BCECF) to measure NHE activity in lymphoblasts from HT patients and normotensive control subjects with (NTFH) and without (NT) a family history of hypertension. In addition, to define whether this increased NHE activity in HT patients was due to an elevated number of NHE-1 molecules or an increased turnover number per site, we used NHE-1–specific antibodies to assess the abundance of the NHE-1 protein in these cell lines. Finally, we estimated NHE-1 phosphorylation in quiescent cells after immunoprecipitation of 32P-labeled transporter. Our results suggest that NHE activity is elevated in HT patients and NTFH compared with NT subjects. As there were no associated changes in NHE-1 abundance, turnover number was increased in cells from HT patients. One possible mechanism underlying this phenotype is an increased NHE-1 phosphorylation in lymphoblasts from HT patients.
BCECF-AM, nonesterified fatty acid–free bovine serum albumin (BSA), glutamine, nigericin, monensin, isopropyl β-d-galactopyranoside, polyoxyethylene-8-lauryl ether, and tissue culture medium 199 (TC199) were from Sigma Chemicals Ltd. RPMI 1640 culture medium, penicillin, and streptomycin were purchased from Gibco BRL, Life Technologies Ltd. Fetal calf serum (FCS) was supplied by Globepharm Ltd. The RPMI 1640 growth medium contained 10% FCS, 4 mmol/L l-glutamine, and penicillin/streptomycin (105 U of each per liter). The TC199 contained 15 mmol/L HEPES and 1 g/L BSA, with pH adjusted to 7.4 with NaOH. Iscove’s modified Dulbecco’s medium containing 25 mmol/L HEPES (pH 7.2) was from Gibco and contained 15% FCS (Sigma), 4 mmol/L l-glutamine, and 105 U each of penicillin and streptomycin per liter. Cyclosporin A was from Sandoz Inc and stored as a stock solution of 20 μg/mL of medium. Protein A–Sepharose CL4B, glutathione–Sepharose 4B, and the pGEX-2T plasmid were from Pharmacia LKB Biotechnology. [3H]3-O-Methyl-d-glucose, Hybond C-extra nitrocellulose, and enhanced chemiluminescence Western blotting reagents were purchased from Amersham International. [32P]Orthophosphate and phosphate-free RPMI 1640 medium were supplied by ICN Biomedicals Ltd.
Patients and Lymphoblast Culture
The study groups consisted of 16 normotensive, healthy nondiabetic control subjects without a family history of hypertension (NT subjects), 12 normotensive control subjects with a family history of one or both parents suffering from hypertension (NTFH subjects), and 13 HT patients. Secondary hypertension was excluded by clinical and biochemical evaluations. All study subjects were selected from among unrelated individuals in Boston and were nondiabetic whites. In addition to interview and physical examination, all individuals had fasting blood samples drawn for the establishment of immortalized cell lines. All subjects gave informed consent. The protocol was approved by the Committee on Human Studies at the Joslin Diabetes Center.
Serum creatinine was measured with standard methods. Height and weight were recorded for determination of body mass index (kilograms per meter squared). Systolic and diastolic (fifth Korotkoff sound) blood pressures were obtained by averaging two blood pressure measurements taken with a standard cuff 5 minutes apart after the subject had been resting in a sitting position for at least 10 minutes. Hypertension was defined as a diastolic pressure greater than 90 mm Hg or systolic pressure greater than 140 mm Hg. A parental history of hypertension was recorded if one or both parents had hypertension diagnosed and treated with antihypertensive medication. This information was obtained directly from the parents or from the study subjects if the parents were not available.
The establishment of lymphoblast cultures has been described previously in detail.19 Briefly, peripheral blood lymphocytes suspended in Iscove’s modified Dulbecco’s growth medium were immortalized by transformation with Epstein-Barr virus with the use of cyclosporin A to improve the efficiency of establishing permanent cell lines. The immortalized cells were harvested by centrifugation (800 rpm, 7 minutes) and suspended in complete growth medium containing 10% dimethyl sulfoxide for cryostorage in liquid nitrogen.
The immortalized lymphoblasts were recovered from liquid nitrogen storage and cultured in RPMI 1640 growth medium; the same batch of serum was used throughout the study. Cell density was determined daily on a Coulter counter model ZM (Coulter Electronics) and was maintained between 0.25 and 0.75×106/mL. The rate of lymphoblast proliferation was determined in RPMI 1640 medium containing 10% FCS by resuspending cells at an initial density of 0.25×106/mL, measuring cell number every day over a period of 3 days and fitting the data with the following equation: N=No ekt, where N is the cell density at an elapsed time t, No is the initial cell density, and k is the time constant for cell proliferation.
On the day before the studies described below, approximately 30×106 lymphoblasts were recovered from the culture media by centrifugation and washed twice with RPMI 1640 medium containing 1 g/L BSA and no serum. The cells were subsequently resuspended gently in RPMI 1640 medium containing 1 g/L BSA, 4 mmol/L glutamine, and penicillin/streptomycin (105 U of each per liter). Preliminary studies had indicated that cells were rendered quiescent after 24 to 48 hours of serum withdrawal, with no increase in cell numbers. At this time, trypan blue exclusion exceeded 98%. These initial studies also indicated that a 24-hour period of serum withdrawal did not significantly alter NHE activity or NHE-1 protein abundance. Therefore, all subsequent experiments were performed on cells after 24 hours of serum withdrawal. Measurements of pHi, NHE activity, and NHE-1 abundance and its phosphorylation in the lymphoblasts were performed blind so that the origin of the cells was not revealed until the study was completed.
Measurement of pHi and Na+-H+ Antiport Activity
Lymphoblasts (5×106 cells) were incubated with BCECF-AM (3 μmol/L in TC199) at 37°C for 0.5 hour. After extensive washes, cells were left in this medium for 0.5 hour at room temperature to ensure complete deesterification of the BCECF-AM. Measurements of pHi have been described in detail previously.19 Briefly, resting pHi was measured in HEPES-buffered saline (mmol/L: NaCl 140, KCl 5, CaCl2 1.8, MgSO4 0.8, glucose 5, and HEPES 15 as well as 1 g/L BSA, pH 7.4 at 37°C). Excitation wavelengths were set at 500 and 439 nm and emission at 530 nm in a fluorometer (Deltascan, Photon Technology International). Calibration was achieved with the use of isotonic KCl buffers and monensin and nigericin as described.19
All measurements of buffering and H+ efflux were performed at 37°C, with pHi clamped to 6.0, since H+ efflux mediated by Na+-H+ exchange was near the Vmax of the transporter in lymphoblasts.15 19 Intrinsic buffering was determined at pHi 6.0 with the use of 50 mmol/L NH4Cl.19 H+ efflux into HEPES-buffered saline and into Na+-free medium (substituting N-methyl-d-glucamine chloride for Na+) enabled the measurement of Na+-dependent H+ efflux. Because of the sensitivity of H+ efflux to 10 μmol/L ethylisopropyl amiloride, this NHE activity was attributed to the ubiquitous isoform NHE-1. Furthermore, this conclusion was substantiated by the use of probes specific for the NHE isoforms 1 through 414 (kindly provided by Dr J. Orlowski, McGill University, Montreal, Canada), which demonstrated that these lymphoblasts contained only mRNA transcripts for NHE-1 and none of the other isoforms. NHE activity was determined in triplicate on two separate occasions, and the reported values are means of these measurements.
Estimation of NHE-1 Abundance in Lymphoblasts
The NHE-1–specific polyclonal antibody G252 was described previously19 20 and is a protein A–Sepharose partially purified immunoglobulin fraction from antiserum raised against a β-galactosidase–NHE-1 carboxy terminal fusion protein containing the final 157 amino acids of NHE-1 (obtained from Prof J. Pouyssegur and Dr C. Sardet, University of Nice, France).16 Another antibody, G253, was raised concurrently and used for immunoprecipitation experiments (see below). A glutathione S–transferase NHE-1 carboxy terminal fusion protein (GST fusion protein) was also constructed so that the same amino acids of NHE-1 from the original β-galactosidase fusion protein were present.19 20 Titration with different amounts of this GST fusion protein enabled us to estimate the NHE-1 abundance in cell extracts.19 20
To estimate NHE-1 abundance per cell, we resuspended known numbers of cells (determined on a Coulter counter) in 50 mmol/L Tris, pH 7.4, containing (mmol/L) NaCl 150, EDTA 5, phenylmethylsulfonyl fluoride 1, o-phenanthroline 1, and iodoacetamide 1, and an equal volume of buffer (composed of 0.125 mmol/L Tris, pH 6.8, 5% sodium dodecyl sulfate [SDS], 20% glycerol, and 0.004% bromophenol blue) was added. After extracts were boiled for 3 minutes, samples were resolved on 7.5% SDS-polyacrylamide gels and electroblotted to supported nitrocellulose as described.19 After an overnight incubation in “blocking buffer” (10% low-fat milk powder [Marvel] in TBS-Tween, which contained 20 mmol/L Tris, pH 7.4, 137 mmol/L NaCl, and 0.1% Tween 20), membranes were incubated with 1 μg/mL of G252 antibody in 5% Marvel in TBS-Tween for 2.5 hours. We have previously validated the specificity of this technique with concurrent incubations of G252 with an excess of GST fusion protein, which abolished NHE-1 immunoreactivity on the blots.19 The second antibody was a 1:1500 dilution of horseradish peroxidase–linked donkey anti-rabbit immunoglobulin (1 hour of incubation) followed by enhanced chemiluminescence developing reagent (1 minute). Bands on the preflashed x-ray film corresponding to NHE-1 (approximately 92 to 114 kD) were measured with a densitometer (Bio-Rad Laboratories Ltd). A serial dilution of GST fusion protein treated as above enabled an estimate of NHE-1 in cell extracts to be made.19 20 Experiments were performed in duplicate. Since most NHE-1 in lymphoblasts was associated with plasma membrane19 20 and mRNA from NHE-1 and no other isoform was detectable in these cells, a turnover number for NHE-1 in the different groups could be calculated.19 20
Immunoprecipitation of 32P-Labeled NHE-1 From Lymphoblasts
Cells (107) were washed three times in phosphate-free HEPES-buffered saline. Identical numbers of cells from each of the three groups (4 to 8×106) were then resuspended in 1 mL phosphate-free RPMI 1640 medium containing 30 mmol/L HEPES, 4 mmol/L glutamine, and penicillin/streptomycin (105 U of each per liter), with pH adjusted to 7.4 with NaOH. Preliminary studies had indicated that the pH of this medium remained between 7.35 and 7.4 after 3 hours of 37°C incubation at this cell density. Carrier-free [32P]orthophosphate (2 MBq) was then added to the 1 mL of cell suspension, and these samples were incubated for 3 hours at 37°C. Similar numbers of cells were used for each of the three groups because Western blots of the same cells had revealed no differences in NHE-1 abundance among the different groups (see “Results”). This enabled the phosphorylation of similar amounts of NHE-1 to be estimated among the three groups. The method for immunoprecipitation of NHE-1 was adapted from that of Sardet et al.16 The cells were recovered by centrifugation and washed extensively in HEPES-buffered saline. One milliliter of extraction buffer composed of 10 g/L polyoxyethylene-8-lauryl ether and (mmol/L) Tris 30, NaCl 130, EDTA 5, phenylmethylsulfonyl fluoride 1, o-phenanthroline 1, iodoacetamide 1, sodium fluoride 100, sodium orthovanadate 5, ATP 10, and sodium pyrophosphate 10 as well as 1 mg/L pepstatin A and 2 mg/L leupeptin was then added to the cells. The extracts were sonicated for 15 minutes, and cell debris was removed by centrifugation at 14 000g. The supernatant was preabsorbed with protein A–Sepharose CL4B beads. The antibody G253 was then added to the supernatant at a final concentration of 100 μg/mL, and the samples were rotated end on end for 16 hours at 4°C. Immunoprecipitates of NHE-1 were recovered after a 1-hour incubation with protein A–Sepharose CL4B beads that had been pretreated with unlabeled lymphoblast extracts to reduce nonspecific binding. These beads were washed between six and eight times in extraction buffer containing 1 g/L ovalbumin. The remaining pellet was solubilized in Laemmli sample buffer for SDS–polyacrylamide gel electrophoresis on 7.5% gels. Gels were dried and subjected to autoradiography on preflashed x-ray films. The densities of the 32P-labeled NHE-1 were determined with a Bio-Rad densitometer, and values were normalized to an arbitrary value of 1 for the NT control cell extracts. These determinations were performed in duplicate on different days 4 to 6 weeks apart, and the mean data are reported. The inorganic phosphate content was also determined in trichloroacetic acid extracts from 107 cells with the use of acid ammonium molybdate for measurement of phosphate in a colorimetric assay.
Results are expressed as mean±SEM. Comparisons by ANOVA and Student’s t test were performed with an oxstat statistics package (Microsoft Corp). Spearman correlation coefficients (rs) were also calculated. Two-tailed probability values less than .05 were considered significant.
Table 1⇓ shows the clinical characteristics of the three study groups. The HT group had significantly higher systolic and diastolic pressures than both of the other groups (P<.001). HT patients were also more overweight, and nine had a family history of hypertension. Plasma creatinine in the HT compared with NT or NTFH groups did not differ significantly.
The pHi of quiescent lymphoblasts measured in HEPES-buffered saline was very similar in all three groups of subjects (Table 2⇓). The intrinsic buffering capacities in either quiescent cells or those clamped to pHi 6.0 did not differ among groups (Table 2⇓). However, H+ efflux rates due to NHE at pHi 6.0 among groups differed when compared by ANOVA (P<.005). NHE activity is very close to the Vmax of the lymphoblast exchanger at this pHi.15 19 Thus, NHE activity (in millimoles per liter per minute, Fig 1⇓) was significantly higher in lymphoblasts from HT patients (18.8±0.6) compared with NT (12.9±0.6, P<.001) or NTFH (16.4±0.6, P<.008) subjects. NHE activity of cells from the NTFH group was significantly elevated compared with the NT group (P<.001). Lymphoblast proliferation rate constants were also significantly higher in the HT group compared with both NT (P<.001) and NTFH (P<.002, Table 2⇓) groups.
Fig 2⇓ shows a representative Western blot of lymphoblast cell extracts obtained from 106 cells from the different groups of subjects. The NHE-1–specific polyclonal antibody G252 exhibited strong specific immunoreactivity toward protein bands in the molecular weight range of 92 to 114 kD. This concurs with the experiments of Sardet et al,16 who reported the molecular weight of N-linked glycosylated NHE-1 in Chinese hamster lung fibroblasts to be approximately 105 to 110 kD. We had previously demonstrated the specificity of the G252 antibody by abolishing immunoreactivity when antibody was coincubated with the GST NHE-1 fusion protein.19 20 The molecular weight of NHE-1 in the three different groups was very similar (Fig 2⇓), with no gross changes that could be attributed to altered N-linked glycosylation. The G253 antibody that was used in immunoprecipitation experiments (see below) reacted with protein bands identical to those of G252 antibody in this molecular weight range, and its reactivity was also abolished specifically with the GST fusion protein (data not shown). By loading known amounts of GST fusion protein and assuming that transfer of cellular NHE-1 and the fusion protein from gel onto the nitrocellulose was comparable, we estimated NHE-1 cellular abundance in the three groups of cell lines, as previously described.19 20 Table 3⇓ illustrates the similarity in cellular abundance of NHE-1 among the three groups. Thus, the elevated NHE activity at pHi 6.0 in HT patients compared with NT subjects was not due to an increase in the number of NHE-1 sites per cell. As we had previously demonstrated that virtually all lymphoblast NHE-1 is concentrated on the plasma membrane fraction with no intracellular compartmentalization,19 20 a turnover number could be estimated from the NHE activity at pHi 6.0 and the number of NHE-1 sites per cell. The turnover number of NHE-1 at pHi 6.0 was significantly higher in HT cells compared with NT cells (Table 3⇓), with intermediate values in NTFH cells. From these data, the increased NHE activity in HT cells cannot be due to increased NHE-1 expression but might be due to posttranslational processes such as phosphorylation.
Since NHE activity is modulated by phosphorylation,16 17 18 identical numbers of quiescent cells from the three groups (and by inference, similar numbers of NHE-1 protein molecules) were labeled with [32P]orthophosphate in phosphate-free RPMI 1640 medium. Antibody G253 immunoprecipitated a phosphoprotein of approximately 100 kD (Fig 3⇓), which is in the molecular weight range corresponding to NHE-1. Immunoprecipitation with nonspecific antibodies showed no phosphoprotein in this region of the gel (data not shown). To compare phosphorylation of NHE-1 among the three groups, we normalized the data so that the NT group had a mean value of 1.0. Fig 4⇓ demonstrates that the phosphorylation of NHE-1 in the HT group was significantly higher than in the NT group (P<.001). Phosphorylation of NHE-1 in the NTFH group had a broad range, spanning values obtained in the other two groups of subjects. This increased phosphorylation of NHE-1 in HT cells was not explained by an increased specific activity of [32P]orthophosphate in these cells, because the uptake of the label was similar among groups, as was the inorganic phosphate content (Table 3⇑).
Spearman regression analysis revealed weak correlations between NHE activity and NHE-1 proliferation (rs=.30, P<.06), NHE activity and cell proliferation rate (rs=.32, P<.04), and NHE-1 phosphorylation and cell proliferation rate (rs=.31, P<.05). However, these relationships do not indicate which of these changes was the primary abnormality.
As the HT group was more overweight than the NT group, we compared 10 subjects from each of the two groups, so that the body mass indexes were more closely matched and not significantly different (NT, 30±2 versus HT, 32±2 kg/m2). The above-reported findings of differences in NHE activity, NHE-1 turnover number, cell proliferation rate, and NHE-1 phosphorylation between the HT and NT cell lines were essentially unaltered (NHE activity: NT, 13.6±0.6 versus HT, 18.8±0.8 mmol/L per minute, P<.001; NHE-1 turnover number: NT, 7085±397 versus HT, 8808±495 per second, P<.02; cell proliferation rate: NT, 17.1±1.0 versus HT, 22.6±1.9 ×10−3/h, P<.03; NHE-1 phosphorylation: NT, 1.00±0.10 versus HT, 1.67±0.20 arbitrary units, P<.01). Furthermore, regression analysis of data obtained in the two normotensive groups (NT and NTFH), which had similar body mass index values, demonstrated no significant correlations between body mass index and NHE activity, NHE-1 phosphorylation, or cell proliferation rate. Thus, these phenotypic differences are unlikely to be attributable to differences in body mass index alone.
In essential hypertension, studies have reported alterations in NHE activity in a variety of cells.2 4 5 6 7 8 9 10 15 Data on leukocytes or red blood cells ex vivo have consistently showed an increased Vmax for NHE. Canessa et al7 have also described a lower Hill coefficient for internal H+ binding in red blood cells from HT patients. The phenotype of a raised NHE Vmax, lowered Hill coefficient, and lowered pH0.5 for internal H+ has been shown to be conserved in Epstein-Barr virus immortalized lymphoblasts from HT patients.15 These cell lines are derived from B lymphocytes, in contrast to previous work on mixed leukocytes.6 In this cell culture model of human hypertension,15 it is likely that environmental or hormonal influences present in vivo may not be as important as the cellular genotype in determining these kinetic changes. However, although Rosskopf et al15 demonstrated no differences in NHE-1 mRNA transcripts in their HT cell lines, they performed no direct determinations of cellular NHE-1 protein content.
In the present study on lymphoblasts, we investigated the resting pHi and NHE activity when pHi was clamped at 6.0, near the Vmax of the exchanger. Measurements of pHi and buffering characteristics in quiescent lymphoblasts confirmed that no differences exist between HT and NT groups.15 In addition, the NTFH group had values similar to those of the other groups. However, NHE activity was elevated in the HT compared with the NT group, with intermediate values obtained in the NTFH group. These findings are in agreement with those of Rosskopf et al,15 although their absolute values of NHE H+ efflux at pHi 6.0 are higher, perhaps related to the use of cells stimulated by the addition of a phorbol ester to activate transport.15 The present study also confirms the finding of an enhanced lymphoblast proliferation rate in lymphoblasts derived from HT patients.15
In a previous study, we had used an NHE-1–specific antibody to characterize NHE-1 protein in cells from patients with diabetes.19 Such an approach has now been applied to lymphoblasts from HT patients as well as NTFH subjects. Our data suggest that the increased NHE activity in both HT patients and NTFH subjects was not due to altered NHE-1 protein abundance in the cell lines. Since NHE-1 is the major isoform present in lymphoblasts,12 13 it is unlikely that altered NHE activity in HT cell lines was due to the presence of the other isoforms, which may be mainly localized to epithelia. Moreover, calculated turnover numbers for NHE-1 suggest that altered activity in HT cells was due to an increased turnover per NHE-1 site. Several posttranslational modifications of NHE may alter its transport activity and hence its turnover. The molecular weight of NHE-1 is higher than that predicted from its amino acid sequence,11 16 and endoglycosidases have been shown to enhance the mobility of the protein in SDS gels.16 There are two putative N-linked glycosylation sites on NHE-1.16 Furthermore, previous experiments with enzymatic inhibitors of the pathways involved in N-linked glycosylation of proteins have demonstrated that NHE activity could be reduced by such inhibitors.21 It is thus possible that altered glycosylation of NHE-1 may account for some of the differences in transport activity between HT and NT cell lines. However, the Western blots of lymphoblast extracts in the present study indicate similarities in molecular weight between HT, NTFH, and NT lines. This provides some evidence against the existence of gross changes in glycosylation leading to differences in transport activity, although subtle changes in the structure of these complex glycosylation side chains cannot be excluded.
Phosphorylation is another posttranslational process that could rapidly alter NHE activity without a change in the number of NHE-1 sites. Indeed, phosphorylation has been demonstrated to increase NHE-1 activity, whether stimulated by agonists16 17 or by the inhibition of phosphatases.17 18 The hypothesis of whether elevated NHE activity in HT and NTFH cell lines could be related to increased phosphorylation of NHE-1 was therefore tested by immunoprecipitating similar numbers of NHE-1 protein molecules from each of the three groups of cells. NHE-1 phosphorylation was significantly increased in the HT group only, with a broad range in the NTFH group that overlapped both HT and NT groups. This is the first direct demonstration that NHE-1 phosphorylation is increased in cells derived from humans with hypertension and may be associated with the increased NHE activity and turnover number. The NTFH group had values of NHE activity and phosphorylation between the extremes seen in the HT and NT groups, a finding to be expected in view of the genetic nature of hypertension. We5 22 and others4 23 have previously demonstrated increased NHE activity in vascular myocytes from the spontaneously hypertensive rat compared with the normotensive Wistar-Kyoto rat. This was associated with no change in the NHE-1 protein content, and thus by implication, the turnover number of NHE-1 was elevated.22 Furthermore, phosphorylation of NHE-1 was elevated approximately twofold in spontaneously hypertensive compared with Wistar-Kyoto rat cells.24
The precise mechanism underlying the increased NHE-1 phosphorylation is not known at present and may reflect increased kinase activity or a reduced phosphatase activity in HT cell lines. Growth factors that stimulate NHE activity and its phosphorylation may lead to an alkaline shift in the set point for transporter activation,16 17 and our current determinations of NHE Vmax would not have detected any shift in the activation curve for NHE-1. However, Rosskopf et al15 have reported a lowered pH0.5 for internal H+, which implied a reduced affinity for H+. Furthermore, in the present study, cells were serum deprived for 24 hours, and the persistence of higher NHE-1 phosphorylation in HT cell lines rendered quiescent by serum withdrawal may indicate a higher intrinsic kinase activity or lowered phosphatase activity independent of growth factors. The identity of the kinase(s) or phosphatase(s) involved would entail the identification of the consensus sequence in the C-terminal of NHE-1, which shows differentially enhanced phosphorylation in the HT cell lines.
Recent evidence has indicated the possible involvement of factors other than phosphorylation in the response of NHE-1 to growth factors.25 26 Deletion of the cytoplasmic C-terminal domain of NHE-1 containing the putative growth factor–sensitive phosphorylation sites led to a reduction but not complete abolition of the response of the transporter to growth factors. Thus, the activity of the exchanger is regulated by both phosphorylation and other undefined nonphosphorylation mechanisms. Until the site or sites involved in the difference in NHE-1 phosphorylation of HT cell lines are identified, it is currently not possible to assess the effect of the enhanced NHE-1 phosphorylation of these cell lines on exchanger function.
In conclusion, we have confirmed that cultured lymphoblasts from HT patients exhibit elevated NHE activity compared with those from NT control subjects. In addition, NTFH subjects have intermediate values of NHE activity. Elevated NHE activity in cell lines from both HT patients and NTFH subjects is purely associated with an increased turnover number of NHE-1 rather than an increase in the number of NHE-1 transporter molecules. Furthermore, phosphorylation of NHE-1 is increased in HT cell lines, with intermediate values present in cell lines from NTFH subjects. The persistence of these graded changes in phenotype, NHE-1 turnover number, and NHE-1 phosphorylation despite culture in vitro indicates that genetic factors may play a major role in the expression of this phenotypic cluster. The molecular mechanism underlying this HT phenotype remains to be explored in this cell culture model, and phosphopeptide mapping of 32P-labeled NHE-1 in these cell lines may reveal the nature of this biochemical defect in HT cell lines.
This research was supported by the Wellcome Trust, the British Heart Foundation, and grants DK 41526 and DK 36836 from the National Institutes of Health, Bethesda, Md.
- Received November 15, 1994.
- Revision received January 11, 1995.
- Accepted January 11, 1995.
Huot SJ, Aronson PS. Na+-H+ exchanger and its role in essential hypertension and diabetes mellitus. Diabetes Care. 1991;14:521-535.
Rosskopf D, Dusing R, Siffert W. Membrane sodium-proton exchange and primary hypertension. Hypertension. 1993;21:607-617.
Mahnensmith RL, Aronson PS. The plasma membrane sodium-hydrogen exchanger and its role in physiological and pathophysiological processes. Circ Res. 1985;56:773-778.
Berk BC, Vallega G, Muslin AJ, Gordon HM, Canessa M, Alexander RW. Spontaneously hypertensive rat vascular smooth muscle cells in culture exhibit increased growth and Na-H exchange. J Clin Invest. 1989;83:822-829.
Canessa M, Morgan K, Goldszer R, Moore TJ, Spalvins A. Kinetic abnormalities of the red blood cell sodium-proton exchange in hypertensive patients. Hypertension. 1991;17:340-348.
Tokudome G, Tomonari H, Gardner JP, Aladjem M, Fine BP, Lasker N, Gutkin M, Byrd LH, Aviv A. Variations in the apparent set point for activation of platelet Na+/H+ antiport. Hypertension. 1990;16:180-189.
Tse C-M, Brant SR, Walker MS, Pouyssegur J, Donowitz M. Cloning and sequencing of a rabbit cDNA encoding an intestinal and kidney-specific Na+/H+ exchanger isoform (NHE-3). J Biol Chem. 1992;267:9340-9346.
Tse C-M, Levine SA, Yun CHC, Montrose MH, Little PJ, Pouyssegur J, Donowitz M. Cloning and expression of a rabbit cDNA encoding a serum-activated ethylisopropylamiloride-resistant epithelial Na+/H+ exchanger isoform (NHE-2). J Biol Chem. 1993;268:11917-11924.
Orlowski J, Kandasamy RA, Shull GE. Molecular cloning of putative members of the Na/H exchanger gene family. J Biol Chem. 1992;267:9331-9339.
Rosskopf D, Fromter E, Siffert W. Hypertensive sodium-proton exchanger phenotype persists in immortalized lymphoblasts from essential hypertensive patients: a cell culture model for human hypertension. J Clin Invest. 1993;92:2553-2559.
Sardet C, Counillon L, Franchi A, Pouyssegur J. Growth factors induce phosphorylation of the Na+/H+ antiporter, a glycoprotein of 110 kD. Science. 1990;247:723-726.
Sardet C, Fafournoux P, Pouyssegur J. α-thrombin, epidermal growth factor, and okadaic acid activate the Na+/H+ exchanger, NHE-1, by phosphorylating a set of common sites. J Biol Chem. 1991;266:19166-19171.
Bianchini L, Woodside M, Sardet C, Pouyssegur J, Takai A, Grinstein S. Okadaic acid, a phosphatase inhibitor, induces activation and phosphorylation of the Na+/H+ antiport. J Biol Chem. 1991;266:15406-15413.
Ng LL, Davies JE, Siczkowski M, Sweeney FP, Quinn PA, Krolewski B, Krolewski AS. Abnormal Na+/H+ antiporter phenotype and turnover of immortalized lymphoblasts from type 1 diabetic patients with nephropathy. J Clin Invest. 1994;93:2750-2757.
Siczkowski M, Davies JE, Ng LL. Activity and density of the Na+/H+ antiporter in normal and transformed human lymphocytes and fibroblasts. Am J Physiol. 1994;267:C745-C752.
Yusufi ANK, Szczepanska-Konkel M, Dousa TP. Role of N-linked oligosaccharides in the transport activity of the Na+/H+ antiporter in rat renal brush-border membrane. J Biol Chem. 1988;263:13683-13691.
Foster CD, Honeyman TW, Scheid CR. Alterations in Na+/H+ exchange in mesenteric arteries from spontaneously hypertensive rats. Am J Physiol. 1992;262:H1657-H1662.
Siczkowski M, Davies JE, Ng LL. Increased phosphorylation and turnover of the sodium proton exchanger (NHE-1) in cultured vascular smooth muscle from insulin resistant spontaneously hypertensive rats. Diabet Med. 1994;11(suppl 1):P82. Abstract.
Wakabayashi S, Bertrand B, Shigekawa M, Fafournoux P, Pouyssegur J. Growth factor activation and ‘H+-sensing’ of the Na+/H+ exchanger isoform 1 (NHE1). J Biol Chem. 1994;269:5583-5588.
Bertrand B, Wakabayashi S, Ikeda T, Pouyssegur J, Shigekawa M. The Na+/H+ exchanger isoform 1 (NHE1) is a novel member of the calmodulin-binding proteins: identification and characterization of calmodulin-binding sites. J Biol Chem. 1994;269:13703-13709.