Salt-Sensitive Hypertension in Transgenic Mice Overexpressing Na+-Proton Exchanger
Abstract Essential hypertension is one of the most common diseases that exacerbate the risk of cardiovascular or cerebrovascular attacks. Although the etiology of essential hypertension remains unclear, recent investigations have revealed that an enhancement of Na+-proton (Na+-H+) exchange activity is a frequently observed ion transport abnormality in hypertensive patients and animal models. To test the hypothesis that increased Na+-H+ exchange causes hypertension, we produced transgenic mice overexpressing Na+-H+ exchanger and analyzed their Na+ metabolism and blood pressure. Urinary excretion of water and Na+ was significantly decreased in transgenic mice, and systolic blood pressure was elevated after salt loading. The impaired urinary excretion of Na+ suggested that the Na+-H+ exchanger overexpressed in the renal tubules increased reabsorption of Na+, which caused a blood pressure elevation by Na+ retention after excessive salt intake. Our results demonstrate that overexpression of Na+-H+ exchanger can be a genetic factor that interacts with excessive salt intake and causes salt-sensitive blood pressure elevation.
More than one fourth of the total population and more than half of the population aged ≥65 years in the United States are estimated to suffer from hypertension.1 2 Because hypertension is a major risk factor for both coronary artery disease and cerebrovascular disease, it has become a significant human health problem. Although the etiology of essential hypertension is unknown, epidemiologic studies have revealed that interactions between genetic and environmental factors, especially excessive intake of dietary salt, play a key role in its pathogenesis.3 Among Na+ metabolism–related factors, an enhancement of Na+-H+ exchange activity has recently been reported in blood cells of hypertensive patients or their normotensive relatives4 and in various types of cells derived from hypertensive animal models5 (for a review, see Reference 66 ). On the basis of these findings, involvement of the Na+-H+ exchanger in the pathophysiology of hypertension has long been speculated. So far, however, there is no direct evidence to show its role in blood pressure control.
The Na+-proton exchanger is a plasma membrane protein with 10 or 12 putative transmembrane domains.7 Driven by the Na+ gradient produced by Na+,K+- ATPase, the Na+-H+ exchanger countertransports Na+ and H+, regulating the maintenance of intracellular pH or Na+ reabsorption in renal tubules. To determine whether increased Na+-H+ exchange activity could be a cause of hypertension, transgenic mice that overexpress the rabbit type-I Na+-H+ exchanger were produced, and their phenotypes were analyzed. In the present study, we demonstrate that the transgenic mice overexpressing the Na+-H+ exchanger in renal tubules showed Na+ retention and became transiently hypertensive during salt loading. These data suggested that overexpression of the Na+-H+ exchanger may be a genetic factor in salt-sensitive blood pressure elevation.
Materials and Methods
Construction of the Transgene
The rabbit Na+-H+ exchanger cDNA clone was isolated from a rabbit pup aortic cDNA library8 by using oligonucleotide probes synthesized according to the nucleotide sequence of the human amiloride-sensitive (type-I) Na+-H+ exchanger.7 It was confirmed that the nucleotide sequences of the cloned Na+-H+ exchanger were completely identical to those reported by Tse et al.9 The transgene was designed so that cDNA was expressed under the control of human elongation factor 1α promoter10 11 (Fig 1⇓). The promoter and Na+-H+ exchanger fusion gene was constructed as follows. The 1.6-kb EcoRI fragment of pEF321CAT11 containing the chloramphenicol acetyltransferase gene was replaced by the 4.2-kb rabbit full-length cDNA of the Na+-H+ exchanger containing a 5′-untranslated region (−350 nucleotides from the ATG codon) and the poly(A) signal. Subsequently, this plasmid was partially digested with EcoRI, filled-in with Klenow fragment, and relegated to disrupt the EcoRI site between the promoter and Na+-H+ exchanger cDNA. This plasmid was used for pronuclear microinjection after linearization by EcoRI digestion.
Production of the Transgenic Mice
Pronuclear microinjection was performed by the standard method. Briefly, female mice (C3H×C57BL/6 F1) were superovulated and mated with C3H males. The fertilized eggs were recovered, and male pronuclei were microinjected with the transgene. The microinjected eggs were then transferred into the oviducts of pseudopregnant ICR females, which mated with vasectomized ICR males.
RNase Protection Assay
To differentiate between the expression of endogenous murine Na+-H+ exchanger and introduced rabbit Na+-H+ exchanger, an RNase protection assay was performed by using a cDNA fragment of the rabbit Na+-H+ exchanger as a probe. Poly(A+) RNA was extracted from various tissues with a QuickPrep mRNA Purification Kit (Pharmacia). The 181-bp Pst I–Sma I fragment of the Na+-H+ exchanger cDNA was subcloned into pBluescript IISK(−) (Stratagene), and an antisense riboprobe was synthesized by using T3RNA polymerase according to the manufacturer’s protocol. The 239-bp antisense riboprobe was hybridized with poly(A+) RNA (1 μg) in hybridization buffer (80% formamide, 40 mmol/L PIPES, 400 mmol/L NaCl, and 1 mmol/L EDTA, pH 6.7) overnight at 45°C and then digested with RNase A. The mRNA for rabbit Na+-H+ exchanger was detected as a 181-nt radioactive band on polyacrylamide gel.
Preparation of the Monoclonal Antibody
The peptide encoding the carboxyl terminal end of the rabbit type-I Na+-H+ exchanger (98 amino acids) was produced as a fusion protein with glutathione S-transferase (GST) by using the Escherichia coli expression vector pGEX-2T (Pharmacia). The fusion protein was affinity-purified by using glutathione-agarose beads and was used in the immunization of rats. Rat monoclonal antibodies that reacted with the fusion protein but not with GST were screened by enzyme-linked immunosorbent assays. The specificity of monoclonal antibodies was further determined by immunoblot analysis, and the cross-reactivity with the murine Na+-H+ exchanger was also confirmed (data not shown).
The kidneys of control and transgenic mice were fixed in situ by perfusing the fixative (95% ethanol, 4% H2O, and 1% acetic acid) via the renal artery before excision. The kidneys were further fixed in the same fixative for 12 hours, embedded in paraffin, and sectioned in 1-μm slices. After deparaffinization and rehydration, immunoenzymatic staining was performed by using biotinylated anti-rat goat immunoglobulin and a Dako LSAB kit (Dako Corp) including peroxidase-labeled streptavidin.
From 235 offspring, 28 carried the transgene and, of these, 20 successfully transmitted the transgene to their progeny. Of 20 transgenic strains, at least 3 (strains 48, 64, and 92) were found to express the rabbit Na+-H+ exchanger at the mRNA level. Since human elongation factor 1α promoter shows strong activity in various murine tissues throughout all developmental stages,12 it was expected that the transgenic mice would universally overexpress the rabbit Na+-H+ exchanger. In fact, expression of the transgene was detected in various tissues, such as the heart, lung, liver, kidney, spleen, testis, skeletal muscle, and aorta, although expression of the transgene could not be detected in other tissues (Fig 2⇓). The partially protected band (120 nt) was considered to detect the endogenous murine type-I Na+-H+ exchanger, because the longest band of partial protection was predicted to be 120 nt in length when sequences of the cRNA probe were compared with the corresponding sequence of the murine type-I Na+-H+ exchanger cDNA (authors’ unpublished data). Actually, only the partially protected band with a length of 120 nt could be detected in the control brain. Therefore, expression of the rabbit Na+-H+ exchanger was assumed to be almost equal to or several times that of the endogenous exchanger at the mRNA level.
Expression of the transgene was further examined immunohistochemically by using a monoclonal antibody against the carboxyl terminal region of rabbit Na+-H+ exchanger. This monoclonal antibody cross-reacted with murine type-I Na+-H+ exchanger because of the highly conserved nucleotide and amino acid sequences between rabbit and murine Na+-H+ exchanger. Although expression of the transgene was detected in various tissues at the mRNA level, the immunohistochemical study failed to detect expression in cell membranes except in the renal tubular cells and epithelial cells of the stomach, intestine, and glomerulus in all three transgenic strains (Nos. 48, 64, and 92). Endogenous expression of amiloride-sensitive Na+-H+ exchanger in the kidney was reported to be localized in the basolateral membrane of renal tubules13 and was confirmed to occur in the proximal tubules of control mice (Fig 3a⇓). However, expression was not detected in other areas of renal tubules (Fig 3b⇓) or in the glomeruli (Fig 3d⇓). Since the antibody could not distinguish between murine (endogenous) and rabbit (transgenic) Na+-H+ exchanger, the transgene expression must be identified as an ectopic signal. In transgenic mice, ectopic expression of Na+-H+ exchanger was detected only in the apical membrane (luminal side) of the medullary collecting tubules (Fig 3c⇓) and the epithelial cells of the glomeruli (Fig 3e⇓). In the present immunohistochemical study, the expression of type-I Na+-H+ exchanger was not detected in the cardiovascular system, such as in cardiac muscle cells, vascular smooth muscle cells, or vascular endothelial cells.
Electrolytes in the urine and serum were examined in age- and sex-matched control and transgenic mice (Table 1⇓). Transgenic mice showed a significant decrease in urinary volume, pH, and urinary excretion of Na+, K+, and Cl−, whereas water and food intakes and creatinine clearance were not different between the two groups. Fractional excretion was also decreased in Na+, K+, and Cl−, indicating that the decrease in urinary excretion of Na+ and other electrolytes was caused by an increase in tubular reabsorption. Increased reabsorption of Na+ in renal tubules accounts for the decrease in plasma renin activity and aldosterone levels, which also explains the decreased fractional excretion of K+ and the increased serum K+ concentration. Despite impaired Na+ excretion, systolic blood pressure was not increased in transgenic mice. Systolic blood pressure and pulse rate did not change between control and transgenic mice.
To test the effect of an environmental factor on blood pressure, chronic salt loading was performed by feeding an 8% NaCl–containing diet or 2% NaCl–containing drinking water. During salt loading, no significant difference was observed in body weight and food and water consumption between control and transgenic mice. Systolic blood pressure was not significantly changed in control mice even after 8 weeks of salt loading. In contrast, transgenic mice showed an elevation in systolic blood pressure after 4 (8% NaCl) or 5 (2% NaCl) weeks (Table 2⇓). Strains 64 and 92 also showed essentially the same blood pressure profile as strain 48 (systolic blood pressure was significantly elevated during the period of 4 to 7 weeks up to 124 and 117 mm Hg, respectively, by feeding an 8% NaCl diet). Along with the blood pressure increase during salt loading, urinary volume, urinary pH, and urinary excretion of Na+, K+, and Cl− increased and reached the same level as in the control mice after 3 to 4 or 5 to 6 weeks (Table 3⇓). After 8 weeks of salt loading, blood pressure in transgenic mice normalized, whereas the level of transgene expression was not significantly changed (Fig 4⇓).
These results lead to the conclusion that overexpression of Na+-H+ exchanger in renal tubules causes Na+ retention and salt-sensitive blood pressure elevation. Since identical phenotypes were observed in at least three independent transgenic strains, we concluded that the salt-sensitive hypertension was caused by overexpression of the Na+-H+ exchanger and not by insertional mutagenesis of the transgene.
The rise in blood pressure seen in the transgenic mice was ≈20 to 25 mm Hg, which seemed to be modest when compared with classical hypertensive rat models such as salt-sensitive Dahl rats, spontaneously hypertensive rats, or rats with deoxycorticosterone acetate (DOCA)-salt hypertension. For the range of blood pressure increase, there seems to be a species difference between rats and mice. Double transgenic mice overexpressing both renin and angiotensinogen14 15 showed hypertension, with a systolic pressure of ≈125 to 130 mm Hg, whereas the transgenic rats overexpressing renin alone showed marked hypertension (≈210 mm Hg). Furthermore, it should be noted that salt sensitivity in classic animal models is not determined by a single gene abnormality. In fact, salt-sensitive Dahl rats are assumed to carry several genes responsible for hypertension.16 In contrast, salt-sensitive blood pressure elevation in our transgenic mice was brought about by overexpression of a single gene product. Thus, it is not unusual that the rise in blood pressure was more modest than in salt-sensitive Dahl rats. DOCA-salt rats share similar pathophysiological features with our transgenic model in that they both show Na+ and water retention, although it is not caused by a hereditary abnormality but by the administration of mineralocorticoid.
One possible explanation for the discrepancy between expression at the mRNA and immunohistochemical levels is that the sensitivity of the present immunohistochemical study was too low to detect transgene expression. This is unlikely, however, because no signal was detected in the sections of aorta or testis of the transgenic mice, where the expression of mRNA was more abundant than in the kidney. Another possibility is that the translation of mRNA or the transport of the Na+-H+ exchanger to the cell membrane was inhibited by an unknown mechanism. Arguing for and against these mechanisms is beyond the objective of this article. An important issue is that the transgene is expressed as a membrane protein at least in epithelial cells of the glomeruli and the apical membrane of medullary collecting tubules and that the prominent phenotype observed in transgenic mice is the impairment of Na+ and water excretion by the kidney.
The cause of the decrease in urinary Na+ excretion was an enhancement of tubular reabsorption, because fractional excretion of Na+, which is a marker of tubular Na+ reabsorption, decreased while creatinine clearance remained normal (Table 1⇑). The mechanism for increased Na+ reabsorption is assumed to be overexpressed Na+-H+ exchanger in the apical membrane, which, along with endogenous Na+,K+-ATPase in the basolateral membrane, would constitute a net reabsorption pathway for Na+ through renal tubular cells (Fig 5⇓). Given that overexpression of the Na+-H+ exchanger primarily caused an increase in Na+ and water reabsorption, it is reasonable to assume that the renin-angiotensin-aldosterone system, which augments Na+ retention and K+ excretion, would be suppressed in compensation for volume expansion. In fact, an increase in serum K+ concentration and a decrease in urinary K+ excretion were observed together with a decrease in plasma renin activity and aldosterone level (Table 1⇑). These compensatory mechanisms may account for the normal blood pressure observed in transgenic mice, despite a tendency for Na+ and water to be retained.
Although the present immunohistochemical study failed to detect the expression of type-I Na+-H+ exchanger in the heart and vessels of transgenic mice, it cannot be ruled out that an increase in Na+-H+ exchange activity in such tissues may contribute to an increase of blood pressure. In smooth muscle or cardiac muscle cells overexpressing the Na+-H+ exchanger, it is possible that an excessive influx of Na+ via the Na+-H+ exchanger would activate Na+-Ca2+ exchange to increase the efflux of Na+ and the influx of Ca2+, which would enhance the contractility of vascular smooth muscle cells or cardiac muscle cells. The production and analysis of transgenic mice overexpressing the Na+-H+ exchanger in the cardiovascular system will be essential in supporting this hypothesis.
According to the hypothesis proposed by Guyton et al,17 which explains the mechanism of salt-sensitive hypertension, the increase in blood pressure observed in transgenic mice could be interpreted as a compensatory mechanism to excrete retained Na+ and water by increasing glomerular perfusion pressure. In fact, urinary volume and urinary excretion of Na+ began to increase along with the blood pressure elevation at 3 to 4 weeks. However, the systolic blood pressure in transgenic mice normalized after 8 weeks of salt loading (Table 2⇑), which suggested that another compensatory mechanism might have worked. Although the mechanisms of such normalization of blood pressure are unclear, possible mechanisms for this “escape” phenomenon may be (1) a decrease in the expression of the transgene (this was not the case as shown in Fig 4⇑), (2) a decrease in the expression of the endogenous Na+-H+ exchanger (type III) in the apical membrane,18 and (3) a decrease in the expression of endogenous Na+,K+-ATPase in the basolateral membrane. All these mechanisms will reduce the reabsorption of Na+ through renal tubular cells and explain the natriuresis without blood pressure elevation. We are currently examining the expression of the type-III Na+-H+ exchanger and the α1 subunit of Na+,K+-ATPase. Our preliminary data so far suggest that the expression of the type-III Na+-H+ exchanger might be slightly downregulated in some cases. Further investigations including the analysis of transgenic mice overexpressing the type-III Na+-H+ exchanger are necessary to determine whether the escape phenomenon can be fully explained by downregulation of this isoform.
In these respects, transgenesis will be useful in the study of the pathophysiology of hypertension because the etiologic role of candidate genes can be evaluated separately by producing transgenic mice carrying each candidate gene or knockout mice that do not carry specific genes. It may then become possible to reconstruct the multiple genetic factors by mating these mice to reproduce similar pathophysiological features as in clinical hypertension. Interactions with environmental factors such as excessive salt intake can also be evaluated for each genetic factor. Using transgenic technology, the present study may be the first step in analyzing the genetic abnormalities in ion transport systems, which have long been assumed to play important roles in the etiology of essential hypertension.
This study was supported by the Uehara Memorial Foundation, Japan Health Sciences Foundation, and a grant from the Tokyo Hypertension Conference and in part by a grant-in-aid from the Japanese Ministry of Education. We wish to thank Dr T. Matsuzaki of the National Institute of Neuroscience for maintaining the experimental animals and Dr M. Aikawa of the Third Department of Internal Medicine, University of Tokyo, and Dr K. Kimura of the Second Department of Internal Medicine, University of Tokyo, for a helpful discussion.
- Received July 5, 1994.
- Accepted September 27, 1994.
- © 1995 American Heart Association, Inc.
Schoenberger JA. Epidemiology of systolic and diastolic systemic blood pressure elevation in the elderly. Am J Cardiol. 1986;57:45c-51c.
Intersalt Cooperative Research Group. Intersalt: an international study of electrolyte excretion and blood pressure: results for 24 hour urinary sodium and potassium excretion. Br Med J. 1988;297:319-328.
Feig PU, D’Occhio MA, Boylan JW. Lymphocyte membrane sodium-proton exchange in spontaneously hypertensive rats. Hypertension. 1987;9:282-288.
Rosskopf D, Dusing R, Siffert W. Membrane sodium-proton exchange and primary hypertension. Hypertension. 1993;21:607-617.
Kuro-o M, Nagai R, Nakahara K, Katoh H, Tsai R, Tsuchimochi H, Yazaki Y, Ohkubo A, Takaku F. cDNA cloning of a myosin heavy chain isoform in embryonic smooth muscle and its expression during vascular development and in arteriosclerosis. J Biol Chem. 1991;266:3768-3773.
Uetsuki T, Nakao A, Nagata S, Kaziro Y. Isolation and characterization of the human chromosomal gene for polypeptide chain elongation factor-1α. J Biol Chem. 1989;264:5791-5798.
Reilly RF, Hildebrandt F, Biemesderfer D, Sardet C, Pouyssegur J, Aronson PS, Slayman CW, Igarashi P. cDNA cloning and immunolocalization of a Na+-H+ exchanger in LCC-PK1 renal epithelial cells. Am J Physiol. 1993;261:F1088-F1094.
Ohkubo H, Kawakami H, Kakehi Y, Takumi T, Arai H, Yokota Y, Iwai M, Tanabe Y, Masu M, Hata J, Iwao H, Okamoto H, Yokoyama M, Nomura T, Katsuki M, Nakanishi S. Generation of transgenic mice with elevated blood pressure by introduction of the rat renin and angiotensinogen genes. Proc Natl Acad Sci U S A. 1990;87:5153-5157.
Fukamizu A, Sugimura K, Takimoto E, Sugiyama F, Seo MS, Takahashi S, Hatae T, Kajiwara N, Yagami K, Murakami K. Chimeric renin-angiotensinogen system demonstrates sustained increase in blood pressure of transgenic mice carrying both human renin and human angiotensinogen genes. J Biol Chem. 1993;268:11617-11621.
Biemesderfer DJ, Abu AA, Exner M, Reilly R, Igarashi P. NHE3: a Na+/H+ exchanger isoform of renal brush border. Am J Physiol. 1993;265:F736-F742.