Chronic potassium depletion induces renal injury, salt sensitivity, and hypertension in young rats

Kidney International, Vol. 59 (2001), pp. 1850–1858

Chronic potassium depletion induces renal injury, salt sensitivity, and hypertension in young rats PATRICIO E. RAY, SHIN-ICHI SUGA, XUE-HUI LIU, XIULING HUANG, and RICHARD J. JOHNSON1 Center for Molecular Physiology Research, Children’s Research Institute, Children’s National Medical Center and Department of Pediatrics, The George Washington University, Washington D.C., and Division of Nephrology, University of Washington, Seattle, Washington, USA

Chronic potassium depletion induces renal injury, salt sensitivity, and hypertension in young rats. Background. Chronic hypokalemia has been associated with renal hypertrophy, interstitial disease, and hypertension in both adult animals and humans. However, the effects of potassium (K⫹) depletion on the rapidly growing infant have not been well studied. The purpose of this study was to determine the effects of severe chronic dietary K⫹ depletion on blood pressure (BP) and renal structural changes in young rats. Methods. Sprague-Dawley rats (50 ⫾ 5 g) were fed either a control or a potassium-deficient diet (⬍0.05% K⫹) for 14 to 21 days. At the end of this period, the blood pressure (BP) was measured in all rats, and six rats in each group were sacrificed to determine changes in renal histology and renin-angiotensin system (RAS) activity. The remaining rats in each group were then switched to a high-salt (6% NaCl)–normal-K⫹ (0.5%) diet or were continued on their respective control or K⫹-deficient diet for an additional six days. Blood pressure measurements were done every three days until the end of the study. Results. K⫹-depleted animals had significant growth retardation and increased RAS activity, manifested by high plasma renin activity, recruitment of renin-producing cells along the afferent arterioles, and down-regulation of angiotensin II receptors in renal glomeruli and ascending vasa rectae. K⫹-depleted kidneys also showed tubulointerstitial injury with tubular cell proliferation, osteopontin expression, macrophage infiltration, and early fibrosis. At week 2, K⫹-depleted rats had higher systolic BP than control rats. Switching to a high-salt (6% NaCl)– normal-K⫹ diet resulted in further elevation of systolic BP in K⫹-depleted rats, which persisted even after the serum K⫹ was normalized. Conclusion. Dietary potassium deficiency per se increases the BP in young rats and induces salt sensitivity that may involve at least two different pathogenic pathways: increased RAS activity and induction of tubulointerstitial injury.

1 Present address is Division of Nephrology, Department of Medicine, Baylor University, Houston, Texas, USA.

Key words: chronic hypokalemia, renal hypertrophy, interstitial disease, blood pressure, kidney development, renin-angiotensin system. Received for publication June 23, 2000 and in revised form November 9, 2000 Accepted for publication November 16, 2000

 2001 by the International Society of Nephrology

Potassium is one of the most important electrolytes involved in cellular function, but potassium depletion is also one of the most common medical conditions and frequently complicates diuretic usage, vomiting, and conditions associated with hyperaldosteronism. Most frequently, hypokalemia is asymptomatic, but chronically persistent potassium deficiency can be associated with significant hemodynamic and renal structural changes. Epidemiologic studies, for example, have correlated diets low in potassium with an increased prevalence of hypertension [1–4]. However, while potassium supplementation can usually be shown to lower blood pressure (BP) in human hypertension and in experimental models of hypertension [5, 6], dietary potassium depletion in animals and humans has led to more conflicting results, with some studies showing potassium depletion to increase systemic BP [7–9] and other studies showing the opposite [10–12]. This suggests that the effects of potassium on the control of BP may be variable depending on the experimental or clinical condition. Potassium depletion may also cause renal structural changes. Several studies have shown that potassium depletion can induce marked renal hypertrophy [13, 14], and chronically, one can observe the development of interstitial fibrosis in both animals and humans [13]. Renal cysts have also been reported to develop in potassium-depleted subjects in which primary or secondary hyperaldosteronism is present [14]. The importance of these changes is debatable, as renal functional changes appear to be minor, and the structural changes are thought to be largely reversible on repletion of potassium. However, the effects of potassium depletion may be more severe in conditions in which there is rapid growth, such as in young infants. Potassium depletion is not uncommon in such infants and may be observed in infants and children treated with diuretics for hypertension, chronic lung disease, or congestive heart failure, and in children with Bartter’s-like syndromes. It is known that potassium depletion in young rats can activate the renin-

1850

Ray et al: K⫹ depletion-induced hypertension

angiotensin system (RAS) [15–17] despite inhibiting aldosterone secretion by the adrenal gland [15, 18]. However, how these two opposing mechanisms affect BP and the renal structural changes has not been studied. The purpose of this study was to determine the effects of severe chronic dietary potassium depletion on BP and renal structural changes in young rats. Postweanling rats were subjected to either a control or K⫹-deficient (⬍0.05%) diet for a two- to three-week period. Subsequently, rats in both groups were switched to a high-salt (NaCl 6%)– normal-K⫹ (0.5%) diet or were continued on their respective control or K⫹-deficient diets for an additional six days. We report that potassium depletion rapidly induces activation of the RAS, with the induction of severe renal injury and the development of salt-sensitive hypertension, and the hypertension persists despite correction of the potassium deficit. These studies suggest that chronic potassium depletion can increase the BP of young rats and can induce significant injury to the kidney within days; also emphasized is the need to treat hypokalemia in the rapidly growing infant. METHODS Experimental design Male postweanling Sprague-Dawley rats, 50 ⫾ 5 g body weight, were purchased from Harlan SpragueDawley (Indianapolis, IN, USA). Animals were housed in a temperature-controlled room (24⬚C) with a 12-hour on/12-hour off lighting schedule. They were divided in two groups of 18 rats each and were fed either a control diet (Na⫹ 0.3%, K⫹ 0.5%) or a potassium-deficient diet (Na⫹ 0.3%, K⫹ ⬍ 0.05%) for 14 to 21 days. These diets were similar in the composition of other major nutrients and were obtained from ICN Nutritional Biochemical (Cleveland, OH, USA). All rats were given food and water ad libitum. Changes in body weight were measured weekly. The BP was measured every three to five days throughout the course of the study. After 14 to 21 days, six rats in each group were sacrificed. Plasma was collected for biochemical analysis, and kidneys were harvested for histologic studies. Subsequently, all remaining rats in the control group were divided in two groups of six rats each and fed either a normal diet (Na⫹ 0.3%, K⫹ 0.5%) or a high-salt–normal-K⫹ diet: (Na⫹ 6%, K⫹ 0.5%) for an additional six days. In a similar manner, all remaining rats in the potassium-deficient group were divided in two groups of six rats each and were continued on the K⫹-deficient diet (Na⫹ 0.3%, K⫹ 0.05%) or were placed on the high-salt–normal-K⫹ diet (NaCl 6%, K⫹ 0.5%) for an additional six days. At the end of this period, the experiment was terminated. Rats in the K⫹-deficient group were not switched to a high-salt–K⫹-deficient diet to test their salt-sensitive BP response under K⫹-deficient conditions, because salt loading induces significant

1851

cardiovascular and growth-related adverse effects in K⫹depleted young rats [19]. Thus, under the current experimental conditions, the salt-sensitivity BP response was tested only after the K⫹ deficit was corrected. Blood pressure measurements Systolic arterial BP measurements were taken in conscious restrained rats using an automated system with a photoelectric tail-cuff [20]. Rats were first adapted to the chamber by performing several measurements each day. After the rats were preconditioned to the chamber, the mean of three BP readings was collected. Animals were returned to their cages for approximately one hour, and a second set of BP pressure readings was done as described before to confirm the initial results. This procedure was repeated every three to seven days throughout the study period. Biochemical analysis Blood was collected to measure the hematocrit (Hct) as described before [15]. Plasma samples were collected in ice-cold heparinized tubes to measure plasma electrolytes, blood urea nitrogen, and creatinine using standard techniques as previously described [15, 18]. Sodium and potassium concentrations were measured using a KNA 2 sodium-potassium analyzer (Radiometer, Copenhagen, Denmark). Ethylenediaminetetraacetic acid (EDTA)containing tubes were used to collect samples for the measurement of plasma renin activity (PRA). PRA was determined by radioimmunoassay [21]. All samples were collected at the time of sacrifice. Renal histology and immunohistochemistry studies Methyl Carnoy’s fixed renal tissues were processed and embedded in paraffin. Four micrometer sections were cut and stained either with the periodic acid-Schiff reagent (PAS) or the Masson trichrome stain [22] to determine the degree of renal injury. Immunohistochemistry studies were done as previously described [23] in order to identify the following antigens: osteopontin (OPN), with OP 199 a goat anti-rat OPN antibody (gift of C. Giachelli, University of Washington, Seattle, WA, USA), macrophages with ED-1, a monoclonal IgG1 to rat macrophages (Harlan, Bioproducts, Indianapolis, IN, USA), and renin with an anti-rat renin polyclonal antibody [diluted 1:2500 in phosphate-buffered saline (PBS) containing 3% bovine serum albumin (BSA); gift of Dr. T. Inagami, Vanderbilt University, Nashville, TN, USA]. The specificity and characteristics of all these antibodies have been previously described [16, 23, 24]. As described in previous studies [23, 24], the OPN and macrophage stainings were selected as markers of renal injury. Cell proliferation was evaluated by using an antibody to the proliferating cell nuclear antigen (PCNA) with a PCNA kit (Zymed, South San Francisco, CA, USA)

1852

Ray et al: K⫹ depletion-induced hypertension

according to the manufacturer’s instructions but using aminoethylcarbazole as the chromogen. Briefly, tissue sections were heated twice for five minutes each in 0.01 mol/L sodium citrate (pH 6.0) in a microwave oven (2450 mHz, 850 W) to augment antigen retrieval. Endogenous peroxidase activity was blocked by treating with 3% H2O2 in 100% methanol for 10 minutes. Sections were treated according to the kit instructions and counterstained with hematoxylin. All sections were examined and scored independently by two investigators in a blinded manner. Angiotensin II receptor binding studies Renal sections were harvested and frozen immediately in isopentane at ⫺30⬚C and were stored at ⫺70⬚C until studied. Cross sections (20 ␮m) of the kidney were cut in a cryostat at ⫺18⬚C, thaw mounted on gelatin-coated glass slides, and dried overnight in a desiccator at 4⬚C. Angiotensin II (Ang II) receptor subtypes were localized by autoradiography as we have previously described [15, 18]. [125I-Sar1-Ile8]-Ang II (specific activity, 2200 Ci/ mmol; New England Nuclear Life Science Products, Boston, MA, USA) was used as the ligand. Tissue sections were preincubated for 15 minutes in 10 mmol/L phosphate buffer (pH 7.4) containing 120 mmol/L NaCl, 5 mmol/L Na2EDTA, 0.005% bacitracin (Sigma Chemical Co., St. Louis, MO, USA), and 0.2% proteinase-free BSA (Sigma), followed by incubation for 120 minutes at 22⬚C in fresh buffer containing the appropriate concentration of the ligand. After incubation, the sections were washed four times for one minute each in fresh ice-cold 50 mmol/L Tris-HCl buffer (pH 7.6), followed by 30 seconds in distilled water at 0⬚C. To characterize Ang II receptor subtypes, consecutive sections were incubated with [125I-Sar1-Ile8]-Ang II (0.5 nmol/L) and either 5 ␮mol/L Ang II (Research Biochemicals International, Natick, MA, USA), 10 ␮mol/L angiotensin II subtype 1 (AT1) receptor antagonist losartan (2-nbutyl-4-chloro-5-hydroxymethyl-1-[2⬘(1H-tetrazol-5-yl)biphenyl-4-yl-methyl]imidazole, potassium salt (Merck, Rahway, NJ, USA), or 0.1 ␮mol/L of the AT2 competitor CGP 42112 (nicotinic acid-Try-N-benzyloxycarbonylArg-Lys-His-Pro-Ile-OH; Research Biochemicals International). Sections were dried after washing and were exposed to Hyperfilm-[3H] (Amersham Corporation, Arlington Heights, IL, USA), along with 20 ␮m sections of [125I]labeled Micro-scale standards (Amersham). Film was developed in a Kodak D-19 developer (Eastman Kodak, Rochester, NY, USA) for four minutes at 4⬚C and was fixed in Kodak rapid fixer (with hardener) for four minutes at 22⬚C and was rinsed in water for 10 minutes. The films with autoradiographic images of binding and standards were used to measure mean optical densities from the different areas of interest on the images. Protein

content of the total renal sections was determined by the Bradford’s Coomassie blue method after scraping the sections from the slides. The protein concentration in the specific renal areas used for receptor quantitation was measured using the method of Miller, Curella, and Zahniser [25] as we have previously described [18]. Mean optical density measurements per unit area were first obtained from the images of both standards and samples. Mean optical density measurements from images of plastic standards and their corresponding disintegrations per minute (dpm) per mg of plastic standard were used to generate standard curves by nonlinear fitting using a computerized image processing and analysis program, Scion Image (Scion, Frederick, MD, USA). These standard curves were used to obtain dpm/mg of plastic standard for the samples. Based on the experimental relationship between plastic and protein standards [26], the dpm/mg of plastic standard measured from the autoradiograms were transformed to corresponding values of dpm/mg protein. After correcting for the specific activity of the ligand, values of fmol per mg of protein were obtained. The apparent number of AT1 receptors was determined as the specific binding displaced by 10 ␮mol/L losartan, while that of AT2 receptors was determined as the specific binding displaced by 0.1 ␮mol/L CGP 42112A. Renal injury/growth studies For each section, 50 glomeruli were examined. Tubulointerstitial injury was graded (0 to 5⫹) in a blinded manner based on the presence of tubular cellularity, basement membrane thickening, dilatation, atrophy, sloughing, or interstitial widening as follows: 0, no changes present, grade 1, ⬍10% tubulointerstitial changes present; grade 2, 10 to 25% tubulointerstitial change involvement; grade 3, 25 to 50%, grade 4, 50 to 75%; and grade 5, 75 to 100% tubulointerstitial changes, respectively [23]. For each renal section, the entire cortical and medullary regions were evaluated, and a mean score per renal section was calculated. A second tubulointerstitial injury score was based on observations that OPN expression by injured tubules is a sensitive marker of tubulointerstitial injury [23]. Utilizing computer-assisted image analysis software (Optimas, v6.2; Media Cybernetics, Silver Springs MD, USA) and digitized images, the percentage of area occupied by OPN-positive tubules (including the entire cortex and outer medulla, exclusive of glomeruli) was measured at ⫻50, and the mean percentage area was calculated for each renal section [24]. The number of macrophages (ED-1–positive cells/mm2) in the cortex and the medulla was also quantitated. Statistical analysis Results were expressed as mean ⫾ SEM, and significance was analyzed by unpaired Student t tests or oneway analysis of variance using the Newman–Keuls test. P values of less than 0.05 were considered statistically significant.

Fig. 1. Potassium depletion induces the recruitment of renin along the afferent arterioles. (A) A representative renal section from a control rat showing the typical normal distribution of renin in juxtaglomerular cells by immunohistochemistry. (B) A representative renal section from a rat subjected to the K⫹-depleted diet showing the recruitment of renin along the afferent arteriole. Original magnification of A and B ⫻400.

RESULTS Hypokalemia-induced growth retardation Postweanling rats were placed on a potassium-deficient diet for 14 to 21 days. This resulted in the development of marked hypokalemia (2.3 ⫾ 0.3 vs. 4.5 ⫾ 0.4 mEq/L, K⫹-depleted vs. controls, P ⬍ 0.001), whereas serum sodium was not affected (136 ⫾ 1.2 vs. 139 ⫾ 0.9 mEq/L, K⫹-depleted vs. controls, P ⬎ 0.5). Rats fed the potassium deficient diet for two to three weeks also showed significant growth retardation. There was an approximately 20% reduction in food intake in K⫹-depleted rats when compared with controls. This was associated with a decreased daily weight gain when compared with control animals, averaging 5 ⫾ 0.4 and 2.8 ⫾ 0.6 g for control and K⫹-depleted rats, respectively (P ⬍ 0.05).

䉴 Fig. 4. Immunohistochemistry staining for proliferating cell nuclear antigen (PCNA; brown color) in representative renal sections from control (A) and potassium-depleted young rats (B). Original magnification ⫻100.

Fig. 3. Potassium depletion induces tubulointerstitial injury in young rats. The renal medulla of representative renal sections from control (A) and potassium-depleted young rats (B) stained with trichrome Mason (light-blue color). Original magnification ⫻200.

Ray et al: K⫹ depletion-induced hypertension

1854

Table 1. Hematocrit (Hct), blood urea nitrogen (BUN), serum creatinine (SCr), and plasma renin activity (PRA)

Hct % BUN mg/dL SCr mg/dL PRA ng/mL/h

Control

K⫹ depletion

41 ⫾ 0.8 13 ⫾ 2.6 0.4 ⫾ 0.04 2.7 ⫾ 0.4

45 ⫾ 1.3a 16 ⫾ 2.1 0.5 ⫾ 0.07 17.3 ⫾ 2a

Data are mean ⫾ SD. a P ⬍ 0.05

At the end of this period, control animals weighed an average of 142 ⫾ 7 g, and K⫹-depleted rats weighed 101 ⫾ 17 g (P ⬍ 0.05). An increase in Hct was also observed in K⫹-depleted rats (Table 1). These findings, in the absence of changes in red cell mass and in the context of previous studies [15–18], suggest the presence of extracellular volume (ECV) contraction in K⫹-depleted rats. Renal function was also slightly depressed in K⫹-depleted animals, although these changes were not statistically significant. Activation of the renin angiotensin system in Kⴙ-depleted young rats Plasma renin activity. PRA was sixfold greater in K⫹depleted rats as compared with controls (Table 1). Renin immunohistochemistry. Consistent with the increase in PRA in hypokalemic rats, an increase in immunoreactive renin was also observed in renal vessels of K⫹-depleted rats. Whereas in control rats renin staining was limited to the juxtaglomerular cells (Fig. 1A), in K⫹-depleted rats, renin staining was also detected upstream in the afferent arterioles (Fig. 1B). Angiotensin II receptor binding. As shown in Figure 2, K⫹ depletion also induced a significant reduction in Ang II receptor binding in the renal glomeruli, vasa recta bundles, and the inner zone of the outer kidney medulla (glomeruli, 250 ⫾ 9 vs. 176 ⫾ 12 fmol/mg protein, for control and K⫹-depleted rats, respectively, P ⬍ 0.001; renal medulla, 395 ⫾ 25 vs. 245 ⫾ 16 fmol/mg protein, for control and K⫹-depleted rats, respectively, P ⬍ 0.001). These receptors were predominately type I Ang II receptors. These findings may be due to the activation of RAS in K⫹-depleted rats, since Ang II seems to be the primary factor modulating the number Ang II receptors expressed in the kidney [18]. No significant renal AT2 binding sites were detected in both control and K⫹-depleted rats. Renal structural changes induced by hypokalemia Renal hypertrophy. There was no difference in absolute kidney weight between K⫹-depleted (0.71 ⫾ 0.04 g) and control (0.68 ⫾ 0.03 g) rats. However, when the kidney weight was expressed as a percentage of the total body weight, this ratio was significantly increased in the K⫹-depleted group (0.70 vs. 0.47%, K⫹ depletion vs. controls, P ⬍ 0.05).

Fig. 2. Quantitative autoradiography Ang II binding studies. (A and B) Representative renal sections from control and potassium-depleted rats. Original magnification ⫻10.

Tubulointerstitial injury. K⫹-depleted rats developed marked tubulointerstitial injury that predominantly affected the epithelium of the collecting ducts of the outer medulla (Table 2). The injury was characterized by areas of tubular cell hypercellularity interspersed with tubular dilation, atrophy, and cast formation. There were substantial increases in Masson trichrome staining in interstitial areas consistent with early fibrosis (Fig. 3). Tubular proliferation (PCNA staining) was prominent (Fig. 4), as well as expression of OPN by tubules in the outer medulla and cortex (Fig. 5A, B). Osteopontin staining was predominantly detected in collecting ducts and medullary thick ascending limb (mTAL) tubular epithelial cells. The greatest expression of OPN was detected in the outer medulla and less so in the inner medulla. The tubular hyperplasia in K⫹-depleted rats also preferentially involved epithelial cells from the collecting ducts and less so from mTAL. A marked macrophage (ED-1– positive cells) infiltration was observed, especially in the outer medulla (Fig. 5C, D), in association with renal tubules expressing OPN. In contrast to the tubular changes, glomeruli appeared normal (data not shown). Control rats had no renal abnormalities. Hypokalemic rats develop salt-sensitive hypertension At the end of two to three weeks, systolic BP levels were significantly elevated in K⫹-depleted rats (114 ⫾ 7.4* mm Hg) when compared with control rats (91 ⫾ 5.3 mm Hg) mean ⫾ SEM, respectively (*P ⫽ 0.01; Fig. 6). In order to determine whether the increase in

Ray et al: K⫹ depletion-induced hypertension

1855

Table 2. Renal histologic findings in potassium depleted baby rats

Tubulointerstitial score in the cortex (0–5⫹) Tubulointerstitial score in the medulla (0.5⫹) Cortical osteopontin % Medullary osteopontin % Cortical macrophages ED-1 positive cells/mm2 Medullary macrophages ED-1 positive cells/mm2 PCNA positive cells in medulla hpf

Control

K⫹ depletion

0.05 ⫾ 0.10 0.16 ⫾ 0.33 0.82 ⫾ 0.05 1.4 ⫾ 0.3 21 ⫾ 3 15 ⫾ 2 1.5 ⫾ 0.4

0.08 ⫾ 0.10 2.76 ⫾ 0.75a 1.4 ⫾ 0.3 3.4 ⫾ 0.5a 25 ⫾ 4 231 ⫾ 28a 16 ⫾ 1a

Data are mean ⫾ SD. a P ⬍ 0.05

Fig. 5. Tubular expression of osteopontin (OPN), which is a sensitive marker of injured tubules, was barely detected in controls (A, ⫻25 and ⫻100), but was markedly up-regulated in tubules in the medulla of potassium-depleted rats at two weeks (B, ⫻25 and ⫻100). Macrophage infiltration, as noted by ED-1 staining, was minimized in the medulla of control animals (C, ⫻200) but was marked in rats with hypokalemic nephropathy (D, ⫻200).

BP was associated with salt sensitivity, rats were switched to a normal or high-salt (6% NaCl)–normal-K⫹ (0.5%) diet. Interestingly, K⫹-depleted rats showed a more marked increase in BP on switching to a high-salt– normal-K⫹ diet (155 ⫾ 14* mm Hg) compared with the control rats subjected to the same diet (123 ⫾ 8.9 mm Hg) mean ⫾ SEM, respectively (*P ⫽ 0.01; Fig. 6). BP values at the end of the six-day high-salt period were markedly increased in the rats that were K⫹ depleted as compared with controls, despite the fact that the K⫹ levels in these rats had corrected back to the normal range (4.2 ⫾ 0.5 vs. 4.5 ⫾ 0.6 in previously K⫹-depleted vs. control rats, respectively, P ⫽ NS). Histologic studies obtained at the end of the study showed persistent tubulointerstitial injury, although the expression of renin in juxtaglomerular cells and renal arterioles (data not shown) and PRA levels were now equally suppressed in both control (2.7 (0.4 ng/mL/hour) and K⫹-depleted animals switched to the high-salt–normal-K⫹ diet (2.3 (0.5 ng/mL/hour, P ⫽ NS). In a similar manner, no differences were detected in Hct values between controls (41 ⫾ 0.8%) and K⫹depleted rats switched to the high-salt–normal-K⫹ diet (42 ⫾ 0.6%, P ⫽ NS). Finally, when rats subjected to the

Fig. 6. Blood pressure values in the four groups of rats at the end of two to three weeks. Groups were defined as: Control, rats subjected to the control diet for a period of three weeks as described in the Methods section; LK, rats subjected to the potassium-deficient diet for a period of three weeks; Control ⫹ HS, rats subjected to the control diet for two weeks and then switched to a control high-salt (6% Na⫹)–normalK⫹ (0.5%) diet for six additional days; LK ⫹ HS, rats subjected to the potassium-depleted diet for two weeks and then switched to a control high-salt (6% Na⫹)–normal-K⫹ (0.5%) diet for six additional days. BP values were taken at the end of each period. *P ⬍ 0.05 when compared with control group. **P ⬍ 0.05 when compared with all other groups.

1856

Ray et al: K⫹ depletion-induced hypertension

K⫹-deficient diet for two weeks were switched to a normal diet (Na⫹ 0.3%, K⫹ 0.5%) for one week, their BP (96 ⫾ 7 mm Hg) was not significantly different from that of the control rats fed with the same normal diet (91 ⫾ 5.3 mm Hg) mean ⫾ SEM, respectively (P ⫽ NS). DISCUSSION Several studies in human and rodents have shown that changes in dietary potassium intake have significant effects on BP [1–10]. However, this BP response of adult humans and rodents to changes in potassium balance varies greatly according to the clinical or experimental conditions [6–12]. Consequently, there is controversy in relation to the specific effects of potassium depletion on BP control. Even less is known about the effect of potassium depletion on BP in infants. The requirements of K⫹ and other electrolytes are increased during this period of rapid growth [27–29], and the infants’ growth rate is sensitive to subtle changes in electrolyte intake and extracellular fluid volume status. Therefore, we examined the effects of chronic dietary potassium depletion on BP using young rats as a model system to mimic the clinical situation frequently encountered in rapidly growing infants. We found that a two- to three-week period of dietary potassium restriction induces an increase in the BP of young rats that appears to involve at least two different pathogenic mechanisms: activation of the RAS and induction of renal tubulointerstitial injury. Furthermore, we found that the hypertension induced by potassium-depletion is salt sensitive. Our first observation was that potassium depletion resulted in an impaired growth, elevation of Hct, and stimulation of the RAS. In young rats, the increased Hct in the absence of changes in red cell mass suggests the presence of extracellular fluid (ECF) volume contraction. Although in the absence of ECF volume measurements this should be considered speculative, several findings support the possibility that chronic dietary K⫹ deficiency may induce volume contraction in young rats. First, in the presence of severe K⫹ restriction, there is movement of some extracellular sodium into cells to maintain electroneutrality [19]. Second, the presence of malnutrition and decreased food intake may lead to a negative sodium balance. Third, hypokalemia-induced diarrhea may also contribute to the presence of ECF volume contraction. However, in our study, none of the K⫹-depleted rats developed diarrhea during the initial two- to three-week period of K⫹ restriction. Fourth, K⫹ depletion impairs sodium chloride reabsorption in the thick ascending limb [30] and may cause renal chloride wasting by decreasing the number of chloride transporters in renal tubules [31]. In addition, young rats are more sensitive than adult rats to changes in dietary K⫹ and salt intake. Growing rats have a greater absolute need

for salt and K⫹ since these electrolytes are needed for tissue growth. In contrast, plasma volume may be actually increased in adult K⫹-depleted rats [30]. Hypokalemic rats also showed evidence for stimulation of the RAS, with increased PRA, increased renal renin content, and decreased Ang II receptor binding in glomeruli and in vasa rectae. Hypokalemia has been previously reported to stimulate renin secretion despite suppressing aldosterone synthesis [15, 32, 33], independent of the ECF volume status [30, 34]. This mechanism involves a direct stimulation of renin-secreting cells in the juxtaglomerular apparatus and is mediated by the renal vascular receptor [34]. In our study, ECF volume contraction may also be contributing to the increased PRA. We and others have reported that PRA activity is markedly influenced by subtle changes in sodium intake and ECF volume in young rats [35, 36]. We do not think that malnutrition or growth retardation per se could explain the changes observed in PRA, and renal Ang II receptors in K⫹-depleted rats. In previous studies [15, 18, 27], we have shown that young rats subjected to approximately a similar reduction in food intake, but given adequate quantities of K⫹, Na⫹, and Cl⫺, do not develop significant changes in PRA and/or Ang II receptors. Moreover, growth retardation can ameliorate the development of hypertension in spontaneously hypertensive rats (SHR) [37]. Hypokalemia had profound effects on renal structure in young rats. As has been previously reported [13, 14, 38], hypokalemic rats developed renal hypertrophy and hyperplasia, which are thought to be mediated by various growth factors, including insulin-like growth factor-1 [14, 39–41], and which were reflected in our study by the increase in kidney weight/body weight ratio and in the number of PCNA-positive tubules. Consistent with previous reports that hypokalemia can induce renal fibrosis [13], we observed macrophage infiltration and interstitial collagen deposition (as shown by the Masson trichrome stain). In addition, there was a marked up-regulation of a macrophage adhesive protein, OPN in the medulla. Moreover, OPN was specifically up-regulated in renal tubules undergoing hyperplastic changes and in renal areas showing a significant recruitment of macrophages. The mechanism for the development of the fibrosis is unclear, but may relate to the activation of the RAS with intrarenal vasoconstriction, leading to ischemia and/or direct stimulation of transforming growth factor-␤ synthesis by local Ang II [16, 42]. Tolins, Hostetter, and Hostsetter showed similar histologic findings in an adult rat model of hypokalemic nephropathy, and suggested that intrarenal complement activation secondary to amidation of C3 may be an important pathogenic factor in this process [43]. However, more studies are needed to determine whether similar pathogenic mechanisms are involved in young K⫹-deficient rats.

Ray et al: K⫹ depletion-induced hypertension

An important finding in the present study was that the K⫹-depleted rats developed an elevation of systemic BP within two weeks. The reason that K⫹-depleted rats showed BP elevation despite an ECV contraction is not clear. However, it is possible that the activation of the RAS initiates the hypertensive response, whereas the structural and functional changes of the kidney may serve to sustain the hypertension. In support of this concept, a previous study has shown that severe salt deprivation increases the BP of adult Sprague-Dawley rats, probably by stimulating the release of renin and increasing the sympathetic nervous system activity [44]. Indeed, the renal injury can theoretically affect BP through a variety of ways, including via a reduction in glomerular filtration rate, activation of the sympathetic nervous system, or intrarenal alterations in vasoactive mediators [45]. The renal structural and functional changes also are likely to explain the persistence of hypertension despite correction of the hypokalemia and later suppression of the RAS. Another important finding is the development of salt sensitivity in potassium-depleted rats. Recently, we reported that a salt-sensitive hypertension can be induced in rats by transient administration of Ang II, and we postulated that the structural and functional change in the kidney could be responsible for the development of salt sensitivity [24]. A similar mechanism may be operative in this model, in which endogenous activity of the RAS in conjunction with direct effects of hypokalemia leads to renal damage and the development of persistent hypertension despite correction of the hypokalemia and reversal of the RAS activation. In conclusion, we have shown that chronic dietary K⫹ restriction increases the BP in young rats and induces salt sensitivity that may involve at least two different pathogenic pathways: increased RAS activity and induction of tubulointerstitial injury. These findings suggest that severe chronic dietary K⫹ depletion in the rapidly growing infant may lead to renal structural and functional changes and the development of salt-sensitive hypertension. These results may have important implications for the therapy of pediatric diseases associated with severe and chronic hypokalemia. ACKNOWLEDGMENTS This work was supported by National Institutes of Health grants DK49419, DK-52121, DK 43422, HL 55605 and grants from F.A.D.I. Reprint requests to Patricio E. Ray, M.D., Room R-211, Children’s Research Institute, Children’s National Medical Center, 111 Michigan Avenue N.W., Washington D.C. 20010, USA. E-mail: [email protected]

APPENDIX Abbreviations used in this article are: Ang II, angiotensin II; AT1, angiotensin II subtype 1; BP, blood pressure; BSA, bovine serum albumin; BUN, blood urea nitrogen; CGP 42112A, nicotinic acid-Tyr-

1857

N-benzyloxycarbonyl-Arg-Lys-His-Pro-Ile-OH; dpm, disintegrations per minute; ECF, extracellular fluid; ECV, extracellular volume; EDTA, ethylenediaminetetraacetic acid; mTAL, medullary thick ascending limb; OPN, osteopontin; PAS, periodic acid/Schiff reagent; PBS, phosphate-buffered saline; PCNA, proliferating cell nuclear antigen; PRA, plasma renin activity; RAS, renin-angiotensin system; TGF-␤, transforming growth factor-␤.

REFERENCES 1. Langford HG: Dietary potassium and hypertension: Epidemiologic data. Ann Intern Med 98:770–772, 1983 2. Tannen RL: The influence of potassium on blood pressure. Kidney Int Suppl 22:S242–S248, 1987 3. Khaw KT, Barrett-Connor E: Dietary potassium and blood pressure in a population. Am J Clin Nutr 39:963–968, 1984 4. Krishna GG, Miller E, Kapoor S: Increased blood pressure during potassium depletion in normotensive men. N Engl J Med 320: 1177–1182, 1989 5. Tobian L, MacNeill D, Johnson MA, et al: Potassium protection against lesions of the renal tubules, arteries, and glomeruli and nephron loss in salt-loaded hypertensive Dahl S rats. Hypertension 6:I170–I176, 1984 6. MacGregor GA, Smith SJ, Markandu ND, et al: Moderate potassium supplementation in essential hypertension. Lancet 2:567–570, 1982 7. Krishna GG, Miller E, Kapoor S: Increased blood pressure during potassium depletion in normotensive man. N Engl J Med 320: 1177–1182, 1989 8. Lawton WG, Fitz AE, Anderson EA, et al: Effect of dietary potassium on blood pressure, renal functions, muscle sympathetic activity, and forearm vascular resistance and flow in normotensive and borderline hypertensive humans. Circulation 81:173–184, 1990 9. Gallen IW, Rosa KM, Esparaz DK, et al: On the mechanisms of the effects of potassium restriction on blood pressure and renal sodium retention. Am J Kidney Dis 31:19–27, 1998 10. Perera GA: Depressor effects of potassium-deficient diets in hypertensive man. J Clin Invest 32:633–638, 1953 11. Freed SC, Friedman M: Depressor effect of potassium restriction on the blood pressure of the rat. Proc Soc Exp Biol Med 78:74–78, 1951 12. Linas SL, Marzec-Calvert R: Potassium depletion ameliorates hypertension in spontaneously hypertensive rats. Hypertension 8: 990–996, 1986 13. Muehrcke RC, Rosen S: Hypokalemic nephropathy in rat and man. Lab Invest 13:1359–1373, 1964 14. Torres VE, Young WF Jr, Offord KP, Hattery RR: Association of hypokalemia, aldosteronism, and renal cysts. N Engl J Med 322: 345–351, 1990 15. Ray PE, Ruley EJ, Saavedra JM: Different effects of chronic K⫹ depletion on forebrain and peripheral angiotensin II receptors in young rats. Brain Res 556:240–246, 1991 16. Ray PE, McCune BK, Gomez RA, et al: Renal vascular induction of TGF-␤2 and renin by potassium depletion. Kidney Int 44:1006– 1013, 1993 17. Ray PE, Castre´n E, Ruley EJ, Saavedra JM: Different effects of chronic Na⫹, Cl⫺ and K⫹ depletion on brain vasopressin mRNA and plasma vasopressin in young rats. Cell Mol Neurobiol 11:277– 287, 1991 18. Ray PE, Castre´n E, Ruley EJ, Saavedra JM: Different effects of sodium or chloride depletion on angiotensin II receptors in rats. Am J Physiol 258:R1008–R1015, 1990 19. Meyer JH, Grunet RR, Zepplin MT, et al: Effects of dietary levels of sodium and potassium on growth and on concentrations in blood plasma and tissues of white rat. Am J Physiol 162:182–188, 1950 20. Vari RC, Coles JM: Atrial natriuretic peptide during the elevation in blood pressure induced by sodium restriction. J Hypertens 9:449– 455, 1991 21. Menard J, Catt KJ: Measurement of renin activity, concentration and substrate in rat plasma by radioimmunoassay of angiotensin I. Endocrinology 90:422–430, 1972 22. Masson PJ: Trichrome stainings and other preliminary techniques. J Tech Met 12:75, 1929

1858

Ray et al: K⫹ depletion-induced hypertension

23. Giachelli CM, Pichler R, Lombardi D, et al: Osteopontin expression in angiotensin II-induced tubulointerstitial nephritis. Kidney Int 45:515–524, 1994 24. Lombardi D, Gordon KL, Polinsky P, et al: Salt-sensitive hypertension develops after short-term exposure to angiotensin II. Hypertension 33:1013–1019, 1999 25. Miller JA, Curella P, Zahniser NR: A new densitometric procedure to measure protein levels in tissue slices used in quantitative autoradiography. Brain Res 447:60–66, 1988 26. Nazarali AJ, Gutkind JS, Saavedra JM: Calibration of 125Ipolymer standards with 125I-brain paste standards for use in quantitative receptor autoradiography. J Neurosci Methods 30:247–253, 1989 27. Ray PE, Schambelan M, Hintz R, et al: Plasma renin activity as a marker for growth failure due to sodium deficiency in young rats. Pediatr Nephrol 6:523–526, 1992 28. Bower TR, Pringle KC, Soper RT: Sodium deficit causing decreased weight gain and metabolic acidosis in infants with ileostomy. J Pediatr Surg 23:567–572, 1988 29. Ray PE, Holliday MA: Growth rate in infants with impaired renal function. J Pediatr 113:594–600, 1988 30. Mujais SK, Katz AI: Potassium deficiency, in The Kidney: Physiology and Pathophysiology (vol 2), edited by Seldin DW, Giebisch G, New York, Raven Press, 1992, pp 2249–2277 31. Amlal H, Wang Z, Soleimani M: Potassium depletion downregulates chloride-absorbing transporters in rat kidney. J Clin Invest 101:1045–1054, 1998 32. Luke RG, Lyerly RH, Anderson J, et al: Effect of potassium depletion on renin release. Kidney Int 21:14–19, 1982 33. Brunner HR, Baer L, Sealy JE, et al: The influence of potassium administration and of potassium deprivation on plasma renin in normal and hypertensive subjects. J Clin Invest 49:2128–2138, 1970

34. Linas SL: Mechanism of hyperreninemia in the potassiumdepleted rat. J Clin Invest 68:347–355, 1981 35. Ray PE, Lyon RC, Ruley EJ, Holliday MA: Sodium or chloride deficiency lowers muscle intracellular pH in growing rats. Pediatr Nephrol 10:33–37, 1996 36. Fine BP, Ty A, Lestrange N, Levine OR: Sodium deprivation growth failure in the rat: Alterations in tissue composition and fluid spaces. J Nutr 117:1623–1628, 1987 37. Toal CB, Leenen FH: Dietary sodium restriction and development of hypertension in spontaneously hypertensive rats. Am J Phyisol 245:H1081–H1084, 1983 38. Elger M, Bankir L, Kriz W: Morphometric analysis of kidney hypertrophy in rats after chronic potassium depletion. Am J Physiol 262:F656–F667, 1992 39. Rohan RM, Unterman TG, Liu L, Hise MK: Expression of the insulin-like growth factor system in the hypokalemic rat kidney. Am J Physiol 272:F661–F667, 1997 40. Flyvbjerg A, Frystyk J, Osterby R, Orskov H: Kidney IGF-I and renal hypertrophy in GH-deficient diabetic dwarf rats. Am J Physiol 262:E956–E962, 1992 41. Flyvbjerg A, Marshall SM, Frystyk J, et al: Insulin-like growth factor I in initial renal hypertrophy in potassium-depleted rats. Am J Physiol 262:F1023–F1031, 1992 42. Border WA, Noble NA: Interactions of transforming growth factor-beta and angiotensin II in renal fibrosis. Hypertension 31(1 Pt 2):181–188, 1998 43. Tolins JP, Hostetter MK, Hostsetter TK: Hypokalemic nephropathy in the rat. J Clin Invest 79:1447–1458, 1987 44. Ott CE, Welch WJ, Loren JN, et al: Effect of salt deprivation on blood pressure in rats. Am J Physiol 256:H1426–H1431, 1989 45. Johnson RJ, Schreiner GF: Hypothesis: The role of acquired tubulointerstitial disease in the pathogenesis of salt-dependent hypertension. Kidney Int 52:1169–1179, 1997