Cholesterol induces renal vasoconstriction and anti-natriuresis by inhibiting nitric oxide production in anesthetized rats

Abstract

Although hypercholesterolemia is implicated in the pathophysiology of many renal disorders as well as hypertension, its direct actions in the kidney are not yet clearly understood. In the present study, we evaluated renal responses to administration of cholesterol (8 μg·min⁻¹·100 g body wt⁻¹; bound by polyethylene glycol) into the renal artery of anesthetized male Sprague-Dawley rats. Total renal blood flow (RBF) was measured by a Transonic flow probe, and glomerular filtration rate (GFR) was determined by Inulin clearance. In control rats (n = 8), cholesterol induced reductions of 10 ± 2% in RBF [baseline (b) 7.6 ± 0.3 μg·min⁻¹·100 g⁻¹], 17 ± 3% in urine flow (b, 10.6 ± 0.9 μg·min⁻¹·100 g⁻¹), 29 ± 3% in sodium excretion (b, 0.96 ± 0.05 μmol·min⁻¹·100 g⁻¹) and 24 ± 2% in nitrite/nitrate excretion (b, 0.22 ± 0.01 nmol·min⁻¹·100 g⁻¹) without an appreciable change in GFR (b, 0.87 ± 0.03 ml·min⁻¹·100 g⁻¹). These renal vasoconstrictor and anti-natriuretic responses to cholesterol were absent in rats pretreated with nitric oxide (NO) synthase inhibitor, nitro-l-arginine methylester (0.5 μg·min⁻¹·100 g⁻¹; n = 6). In rats pretreated with superoxide (O₂⁻) scavenger tempol (50 μg·min⁻¹·100 g⁻¹; n = 6), the cholesterol-induced renal responses remained mostly unchanged, although there was a slight attenuation in anti-natriuretic response. This anti-natriuretic response to cholesterol was abolished in furosemide-pretreated rats (0.3 μg·min⁻¹·100 g⁻¹; n = 6) but remained unchanged in amiloride-pretreated rats (0.2 μg·min⁻¹·100 g⁻¹; n = 5), indicating that Na⁺/K⁺/2Cl⁻ cotransport is the dominant mediator of this effect. These data demonstrate that cholesterol-induced acute renal vasoconstrictor and antinatriuretic responses are mediated by a decrease in NO production. These data also indicate that tubular effect of cholesterol on sodium reabsorption is mediated by the furosemide sensitive Na⁺/K⁺/2Cl⁻ cotransporter.

cholesterol serves many roles in the body, and it is present in all cell membranes. There it plays a variety of important functions such as membrane fluidity, function of lipid rafts in cell signaling, caveolae formation, regulation of several membrane transport mechanisms, etc. (3, 6, 11, 12, 40). Although cholesterol is an essential substance for a healthy body, a high level of cholesterol in the blood is a known risk factor for cardiovascular diseases including hypertension (10, 15, 41). Moreover, substantial evidence supports an important role for hypercholesterolemia and dyslipidemia in the development of chronic renal disease. These studies suggest a causal relationship between cholesterol levels, proteinuria, and renal injury (1, 2, 9, 17, 30). It leads to the notion that hypercholesterolemia also significantly influences kidney function and may exacerbates other well known effects of cholesterol. These effects could influence long-term blood pressure regulation controlled by the kidney, lead to inappropriate salt excretion, and thus contribute to the development of hypertension. Although an impairment of renal vascular function was reported in an in vitro study using isolated kidney from hypercholesterolemic rats (33) or renal artery rings isolated from hypercholesterolemic pigs (8), little is known about the prospective relationship between lipids, particularly cholesterol and acute changes in renal function in vivo.

It has been shown that hypercholesterolemia causes a decrease in nitric oxide (NO) production and an increase in superoxide (O₂⁻) activity in the kidney (1, 2, 9, 11, 31), leading to endothelial dysfunction. In agreement, renal expression of NAD(P)H oxidase subunits is increased, and, moreover, NO synthase as well as SOD activity is reduced in hypercholesterolemic animals (2, 8); thus it may cause enhanced O₂⁻ activity and reduced NO level in the kidney. Since endothelium-produced NO plays a major role in modulating renal function (16, 22–24, 39), we tested the hypothesis that an increased level of cholesterol impairs renal hemodynamics and tubular function by a reduction of NO and/or by an enhancement of O₂⁻ activity.

We evaluated the hemodynamic and excretory responses to cholesterol bound by polyethylene glycol (PEG) infused directly into the left renal artery of anesthetized male Sprague-Dawley rats. To understand the mechanism involved in the renal responses to cholesterol, these experiments were also conducted in rats pretreated with nitro-l-arginine methyl ester (l-NAME), a NO synthase inhibitor, and also with tempol, a O₂⁻ scavenging agent. Additionally, experiments were also conducted in other groups of rats pretreated with amiloride and furosemide to evaluate the role of tubular epithelial Na⁺ channel (ENaC) and Na⁺/K⁺/2Cl⁻ co-transport (NKCC) activity, respectively, in mediating sodium excretory responses to cholesterol.

METHODS

The study was performed in male Sprague-Dawley rats (280–320 g body wt), obtained from Charles River Laboratories (Wilmington, MA), in accordance with the guidelines and practices established by the Tulane University Animal Care and Use Committee. After at least 3 days of acclimatization of the rats in our vivarial facility after receiving the shipment from the company, the acute clearance experiments were performed in these rats to determine renal responses to cholesterol. On the morning of the experimental day, rats were anesthetized with pentobarbital sodium (50 mg/kg ip), and the surgical procedures required for acute renal clearance experiments were conducted as described previously (18–21). Briefly, the right jugular vein was catheterized for intravenous administration of saline and drug solutions. The right femoral artery was cannulated with a catheter connected to a pressure transducer, which allowed continuous monitoring of arterial blood pressure (AcqKnowledge data acquisition system; Biopac). The right femoral artery catheter was also used for blood sampling. A tapered PE-10 catheter was inserted into the left renal artery via the left femoral artery to allow intra-arterial administration of drugs directly into the left kidney (18–21). This catheter was kept patent by a continuous infusion of heparinized isotonic saline, as a vehicle, at a rate of 5 μl/min throughout the experiment. The left kidney was exposed via a flank incision and placed in a Lucite cup, and the ureter was cannulated with a PE-10 catheter for urine collection. An ultrasonic flow probe (Transonic System, Ithaca, NY) was placed on the left renal artery to measure total renal blood flow (RBF).

Renal responses to intra-arterial administration of cholesterol bound by PEG (vehicle) were examined in these experiments. PEG allows cholesterol to dissolve in saline, and, moreover, PEG cholesterol can easily be incorporated into the cellular membrane (32); thus it is expected that a constant rate delivery of this compound in the kidney has occurred during the infusion period. As circulating cholesterol in the plasma is bound mostly by protein (very low-density lipoprotein, low-density lipoprotein, and high-density lipoprotein particules), PEG-bound cholesterol in the plasma resembles free cholesterol that exerts its effects mainly by its integration into the cellular membrane. The dose of cholesterol used in the present study was 8 μg·min⁻¹·100 g body wt⁻¹ that could cause an increase of ∼10% in free cholesterol in the kidney. This dose (8 μg·min⁻¹·100 g body wt⁻¹) was selected since it was found to cause 10–15% reductions in RBF in initial few pilot experiments in rats. Administration of the vehicle (PEG) alone in control rats revealed that this vehicle did not cause any significant effects in the kidney.

To determine the changes in the cholesterol content in the kidney due to such intra-arterial infusion of PEG-bound cholesterol, left kidneys were isolated from rats infused with PEG cholesterol for 60 min (n = 7) as well as from control rats infused with vehicle (n = 6) for 60 min. Kidney tissues were homogenized on ice in isotonic buffer using a Polytron at high speed with a saw-tooth generator, followed by sonication at high power. Lipids were extracted from homogenates using methanol/chloroform as described by Bligh and Dryer (7). Extracted lipids were dried under nitrogen and resuspended in reaction buffer provided with the Amplex Red Cholesterol Assay (Invitrogen, Eugene, OR). Cholesterol content was measured using the above kit according to manufacturer instructions and calculated from a standard curve. Fluorescence was measured on a BioTek fluorescence microplate reader using excitation and emission at 530 and 590 nm, respectively. Data were normalized to protein content measured using a BCA protein assay (Thermo Scientific, Rockford, IL).

The acute responses to PEG cholesterol infusion were examined in control group and also in four other groups of rats pretreated with different drugs. These groups were as follows: 1) control group (n = 8), 2) l-NAME pretreated group (n = 6), 3) tempol pretreated group (n = 6), 4) amiloride pretreated group (n = 5), and 5) furosemide pretreated group (n = 6). These groups were different from the groups used for the measurement of cholesterol content as mentioned earlier. After a 50-min stabilization period after completion of surgical procedures, the experimental protocol was started with two 30-min control clearance periods to assess baseline values of renal hemodynamic and excretory parameters between groups. This was followed by the intra-arterial infusion of vehicle (PEG) in the control group and l-NAME (0.5 μg·min⁻¹·100 g⁻¹), tempol (50 μg·min⁻¹·100 g⁻¹), amiloride (0.2 μg·min⁻¹·100 g⁻¹), or furosemide (0.3 μg·min⁻¹·100 g⁻¹) in individual pretreated groups, which were administered continuously for the remainder of the experimental period. These doses of drugs were used in previous studies reported from our laboratory (18–20, 23, 25–27). After initiation of pretreatment with drugs, a 15-min period was allowed for stabilization before collection of two 30-min clearance periods to determine renal functional responses to pretreatment. PEG cholesterol was then administered intra-arterially at a dose of 8 μg·min⁻¹·100 g⁻¹ as mentioned earlier. After allowing 15 min for stabilization, another two 30-min clearance collections were made to assess renal functional responses to cholesterol in these rats.

Urine volume was measured gravimetrically. Sodium concentration in plasma and urine samples was determined by flame photometry. Renal vascular resistance (RVR), absolute (U_NaV), and fractional sodium excretion (FE_Na) were calculated according to standard formulas. Inulin concentration in plasma and urine samples was measured colorimetrically to assess inulin clearance that was used for the determination of glomerular filtration rate (GFR). Nitrite/nitrate (NOx) concentration in the urine was measured by colorimetric assay (Cayman Chemical; Ann Arbor, MI) to determine urinary excretion rate of NO metabolites (U_NOxV) (21). The responses to cholesterol infusion was calculated by deducting the baseline values obtained during the pretreatment of drugs (l-NAME/tempol/amiloride/furosemide) or vehicle (PEG) from the values obtained during the period of PEG cholesterol infusion. The average values obtained during two 30-min clearance collection periods before cholesterol infusion were taken as “pretreatment” values (depicted as 100% in figures), and the average values obtained in two 30-min clearance collection periods during cholesterol infusion were taken as experimental values for “cholesterol” treatment.

Data are expressed as means ± SE. Statistical comparisons between control and experimental values in the same group were conducted by paired Student's t-test. Statistical comparisons among the groups were conducted by two-way ANOVA for repeated measurements, followed by Newman-Keuls test. A P value of ≤0.05 was considered statistically significant.

RESULTS

Kidney cholesterol content during infusion of PEG cholesterol.

Figure 1 illustrates the results obtained from the left kidneys of rats infused with PEG cholesterol (n = 7) or vehicle (PEG alone; n = 6). Administration of PEG cholesterol into the left renal artery over 60-min period led to a 28% increase of kidney tissue renal cholesterol content compared with vehicle-treated rats. The mean values of cholesterol contents in the left kidney in PEG cholesterol-treated and vehicle-treated rats were 3.19 ± 0.13 and 2.48 ± 0.14 μg cholesterol/100 μg protein, respectively, which were significantly (P < 0.05) different from each other. This indicates that the PEG cholesterol can rapidly and acutely modify nephron membrane cholesterol levels.

Fig. 1.

Infusion of PEG-cholesterol increase renal membrane cholesterol content. Content of cholesterol in the kidney in vehicle-infused group (n = 6) and cholesterol-infused group (n = 7). Data normalized to total protein content. *Significant difference vs. vehicle-infused group (P < 0.05).

Baseline values of renal parameters in different experimental groups used in acute studies.

Between the experimental groups of rats in the present study, there were no significant differences in basal values of mean arterial pressure (MAP) and renal parameters before administration of l-NAME, tempol, amiloride, furosemide, or cholesterol vehicle in rats (mean values in each group are given in Table 1). Table 2 summarizes the control values of MAP and renal hemodynamic and excretory functions obtained due to pretreatment of drugs (l-NAME, tempol, amiloride, and furosemide) and vehicle (PEG) given intra-arterially in the left kidney before administration of cholesterol. Administration of PEG alone in rats did not cause any significant changes in MAP or any renal parameters (Table 2) compared with values in control untreated groups (Table 1). Intra-arterial administration of l-NAME directly into the left kidney caused a minimal increase in MAP but caused significant decreases in RBF, urine flow (V), U_NaV, and FE_Na without altering GFR (Table 2). Tempol administration did not cause any significant change in MAP or in renal parameters (Table 2). As expected, amiloride or furosemide pretreatment markedly increased V, U_NaV, and FE_Na without altering RBF, GFR, or MAP (Table 2). Urinary excretion rate of NO metabolites nitrate/nitrite (U_NOxV) was significantly decreased during l-NAME infusion but not during tempol, amiloride, or furosemide administration compared with the values obtained in vehicle pretreated rats (Table 2). Renal and systemic responses to the administration of l-NAME, tempol, and amiloride observed in the present study were similar to those studies reported earlier from our laboratory (18–20, 23, 25–27).

Table 1.

Basal values of mean arterial pressure, renal hemodynamics and excretory function in experimental animals

Group of Rats	n	MAP, mmHg	RVR, mmHg·ml−1·min−1·g−1	RBF, ml·min−1·g−1	GFR, ml·min−1·g−1	V, μl·min−1·g−1	UNaV, μmol·min−1·g−1	FENa, %	UNOxV, nmol·min−1·g−1
Control untreated	8	115±2	15.3±0.6	7.6±0.2	0.88±0.05	10.6±0.9	0.96±0.04	0.72±0.03	0.22±0.01
l-NAME pretreated	6	116±2	15.8±0.8	7.3±0.2	0.91±0.03	9.8±1.1	0.94±0.04	0.75±0.05	0.23±0.02
Tempol pretreated	6	114±4	15.5±0.9	7.4±0.4	0.90±0.08	9.7±0.8	0.92±0.03	0.73±0.06	0.26±0.05
Amiloride pretreated	5	118±3	15.1±1.1	7.9±0.6	0.86±0.07	9.4±0.9	0.95±0.05	0.78±0.06	0.21±0.02
Furosemide pretreated	6	114±3	15.0±1.0	7.8±0.6	0.90±0.03	9.9±1.1	0.94±0.07	0.77±0.05	0.24±0.02

Values are means ± SE. MAP, mean arterial pressure; RBF, renal blood flow; GFR, glomerular filtration rate; V, urine flow; U_NaV, absolute sodium excretion; FE_Na, fractional sodium excretion; U_NOxV, urinary nitrite/nitrate excretion. There were no significant differences in basal parameters between experimental groups.

Table 2.

Control values of mean arterial pressure and renal parameters obtained during pretreatment of drugs (l-NAME, tempol, amiloride and furosemide) and cholesterol-vehicle prior to administration of cholesterol

Group of Rats	n	MAP, mmHg	RVR, mmHg·ml−1·min−1·g−1	RBF, ml·min−1·g−1	GFR, ml·min−1·g−1	V, μl·min−1·g−1	UNaV, μmol·min−1·g−1	FENa, %	UNOxV, nmol·min−1·g−1
Vehicle pretreated	8	113±3	15.6±0.7	7.2±0.3	0.85±0.04	10.4±0.8	0.97±0.05	0.74±0.04	0.23±0.02
l-NAME pretreated	6	120±4	22.7±0.9*	5.2±0.3*	0.85±0.02	8.6±1.0*	0.72±0.05*	0.60±0.03*	0.16±0.02*
Tempol pretreated	6	115±4	15.3±0.8	7.5±0.5	0.91±0.06	9.9±0.7	0.95±0.04	0.75±0.03	0.28±0.04
Amiloride pretreated	5	114±2	15.1±1.0	7.6±0.5	0.84±0.06	10.9±1.2	2.16±0.24*	1.83±0.19*	0.24±0.03
Furosemide pretreated	6	115±3	14.7±1.0	7.9±0.6	0.94±0.04	16.5±2.2*	2.27±0.27*	1.92±0.22*	0.26±0.02

Values are means ± SE.

^*Significant difference vs. values in vehicle pretreated group (P < 0.05).

Responses to cholesterol infusion on renal hemodynamics.

Intra-arterial infusion of cholesterol into the left kidney caused increases in RVR and reductions in RBF and GFR without affecting MAP. Figures 2–4 illustrate the percent changes in renal hemodynamic parameters in response to cholesterol as normalized to the pretreatment values. As shown in Fig. 2A, cholesterol infusion into the kidney caused a significant increase in RVR [absolute change (Δ), 2.3 ± 0.4 mmHg·ml⁻¹·min⁻¹·g⁻¹]. Similar increases in RVR were observed in tempol (Fig. 2A), amiloride, and furosemide pretreated rats (Fig. 2B). Although l-NAME pretreatment markedly increased RVR (Table 2), cholesterol did not cause further significant change In RVR (Fig. 2A). There was a significant reduction in RBF (Δ, −0.9 ± 0.2 ml·min⁻¹·g⁻¹) in response to cholesterol administration in vehicle pretreated rats (Fig. 3A). Cholesterol infusion did not show a significant change in RBF in l-NAME pretreated rats (Fig. 3A), although basal RBF was decreased due to l-NAME infusion (Table 2). Tempol, amiloride, or furosemide pretreatment did not alter basal RBF significantly as well as the responses in RBF to cholesterol that were similar to responses in vehicle pretreated group (Fig. 3, A and B). Cholesterol infusion did not cause any appreciable changes in GFR in vehicle pretreated rats as well as in rats pretreated with l-NAME and tempol (Fig. 4A) as well as amiloride and furosemide (Fig. 4B).

Fig. 2.

A: renal vascular resistance (RVR) responses in percent changes to intra-arterial infusion of cholesterol in vehicle pretreated group (●; n = 8), l-NAME pretreated group (△; n = 6), and tempol pretreated group (□; n = 6). B: RVR responses in percent changes to intra-arterial infusion of cholesterol in amiloride pretreated group (◊; n = 5) and furosemide pretreated group (▿; n = 6). *Significant difference vs. pretreatment values (P < 0.05).

Fig. 3.

A: renal blood flow (RBF) responses in percent changes to intra-arterial infusion of cholesterol in vehicle pretreated group (●; n = 8), l-NAME pretreated group (△; n = 6), and tempol pretreated group (□; n = 6). B: RBF responses in percent changes to intra-arterial infusion of cholesterol in amiloride pretreated group (◊; n = 5) and furosemide pretreated group (▿; n = 6). *Significant difference vs. pretreatment values (P < 0.05).

Fig. 4.

A: glomerular filtration rate (GFR) responses in percent changes to intra-arterial infusion of cholesterol in vehicle pretreated group (●; n = 8), l-NAME pretreated group (△; n = 6), and tempol pretreated group (□; n = 6). B: GFR responses in percent changes to intra-arterial infusion of cholesterol in amiloride pretreated group (◊; n = 5) and furosemide pretreated group (▿; n = 6). There were no appreciable changes in all groups.

Responses to cholesterol infusion on renal excretory function.

Figures 5–8 illustrate the percent changes in renal excretory parameters in response to cholesterol infusion as normalized to the pretreatment values. Cholesterol significantly reduced V (Δ, −1.8 ± 0.3 μl·min⁻¹·g⁻¹) in vehicle pretreated rats, as shown in Fig. 5A. Inhibiting NO production reduced baseline V (Table 2) in l-NAME pretreated rats; however, cholesterol infusion in these l-NAME pretreated rats did not alter urine flow. In the tempol pretreated group, cholesterol also induced a decrease in V (Δ, −1.2 ± 0.4 μl·min⁻¹·g⁻¹). Although, amiloride treatment in rats increased V (Table 2), responses to cholesterol (Δ, −1.8 ± 0.4 μl·min⁻¹·g⁻¹) were the same as observed in the vehicle pretreated group (Fig. 5B). However, antidiuretic response to cholesterol infusion was abolished in furosemide pretreated rats (Fig. 5B), although furosemide treatment increased V in these rats (Table 2).

Fig. 5.

A: urine flow (V) responses in percent changes to intra-arterial infusion of cholesterol in vehicle pretreated group (●; n = 8), l-NAME pretreated group (△; n = 6), tempol pretreated group (□; n = 6). B: V responses in percent changes to intra-arterial infusion of cholesterol in amiloride pretreated group (◊; n = 5) and furosemide pretreated group (▿; n = 6). *Significant difference vs. pretreatment values (P < 0.05).

Fig. 6.

A: absolute sodium excretion (U_NaV) responses in percent changes to intra-arterial infusion of cholesterol in vehicle pretreated group (●; n = 8), l-NAME pretreated group (△; n = 6), and tempol pretreated group (□; n = 6). B: U_NaV responses in percent changes to intra-arterial infusion of cholesterol in amiloride pretreated group (◊; n = 5) and furosemide pretreated group (▿; n = 6). *Significant difference vs. pretreatment values (P < 0.05). ^#Significant difference vs. response in vehicle pretreated group (P < 0.05).

Fig. 7.

A: fractional sodium excretion (FE_Na) responses in percent changes to intra-arterial infusion of cholesterol in vehicle pretreated group (●; n = 8), l-NAME pretreated group (△; n = 6), and tempol pretreated group (□; n = 6). B: FE_Na responses in percent changes to intra-arterial infusion of cholesterol in amiloride pretreated group (◊; n = 5) and furosemide pretreated group (▿; n = 6). *Significant difference vs. pretreatment values (P < 0.05). #Significant difference vs. response in vehicle pretreated group (P < 0.05).

Fig. 8.

A: urinary nitrite/nitrate excretion (U_NOxV) responses in percent changes to intra-arterial infusion of cholesterol in vehicle pretreated group (●; n = 8), l-NAME pretreated group (△; n = 6), and tempol pretreated group (□; n = 6). B: U_NOxV responses in percent changes to intra-arterial infusion of cholesterol in amiloride pretreated group (◊; n = 5) and furosemide pretreated group (▿; n = 6). *Significant difference vs. pretreatment values (P < 0.05).

Figures 6 and 7 illustrate U_NaV and FE_Na responses to cholesterol infusion in these rats. In the vehicle pretreated group, cholesterol markedly reduced both U_NaV (Δ, −0.29 ± 0.02 μmol·min⁻¹·g⁻¹) and FE_Na (Δ, −0.18 ± 0.02%). However, in l-NAME pretreated rats, cholesterol infusion did not cause any significant reduction in U_NaV and FE_Na (Figs. 6A and 7A), although l-NAME alone significantly decreased both U_NaV and FE_Na (Table 2). Tempol pretreatment did not affect U_NaV and FE_Na; however, compared with vehicle pretreated group, U_NaV and FE_Na responses to cholesterol were partially attenuated (Δ, −0.20 ± 0.02 μmol·min⁻¹·g⁻¹ and Δ, −0.13 ± 0.01%, respectively) in tempol pretreated rats. Although amiloride pretreatment markedly increased both U_NaV and FE_Na (Table 2), reductions in U_NaV and FE_Na (Δ, −0.60 ± 0.13 μmol·min⁻¹·g⁻¹ and Δ, −0.43 ± 0.09%, respectively) induced by cholesterol were similar as in vehicle pretreated rats (Figs. 6B and 7B). Furosemide pretreatment also greatly increased both U_NaV and FE_Na (Table 2); however, sodium excretion remains unaffected by cholesterol in furosemide pretreated rats (Figs. 6B and 7B).

Responses to cholesterol infusion on endogenous NO generation.

U_NOxV (marker of endogenous NO generation) was significantly decreased by intra-arterial infusion of cholesterol (Δ, −0.05 ± 0.01 nmol·min⁻¹·g⁻¹) in vehicle pretreated rats, as illustrated in Fig. 8A, indicating that cholesterol infusion induces a reduction in endogenous NO production. Similar reductions in U_NOxV in response to cholesterol were also observed in tempol pretreated (Δ, −0.04 ± 0.02 μmol·min⁻¹·g⁻¹), amiloride pretreated (Δ, −0.04 ± 0.01 μmol·min⁻¹·g⁻¹), as well as in furosemide pretreated rats (Δ, −0.05 ± 0.01 μmol·min⁻¹·g⁻¹). However, U_NOXV did not alter by cholesterol infusion in l-NAME pretreated rats (Fig. 8A) since NO production was already inhibited in this group of rats (Table 2).

DISCUSSION

The present study demonstrates that infusion of cholesterol directly into the renal artery in rats induces renal vasoconstrictor, antidiuretic, and antinatriuretic effects. Administration of the dose of cholesterol used in this present study caused an appreciable increase in cholesterol content in the renal tissue. Cholesterol bound by PEG that allows its integration into the cellular membrane belongs to the unique group of non-ionic amphipathic cholesterol derivatives (32) that exhibits low toxicity in in vitro as well as in vivo experiments (32, 34). Previous studies suggest that PEG cholesterol may interact with specific membrane components at very low dose (32, 34). In the present study, we have also used a relatively low dose of PEG cholesterol that causes an increase of ∼28% in cholesterol content in the kidney.

To our knowledge, this is the first study assessing direct effects of cholesterol on renal hemodynamics and excretory function in vivo. Interestingly, these renal responses to cholesterol were absent in rats pretreated with NO synthase inhibitor l-NAME. Cholesterol infusion also markedly reduced urinary excretion rate of NO metabolites nitrite/nitrate in the present study. These data indicate that cholesterol-induced changes in renal hemodynamics and excretory function are predominantly mediated by diminishing endogenous NO production. Cholesterol did not alter GFR, although there was a substantial decrease in RBF, indicating that it exerted proportionate influence on both preglomerular and postglomerular vascular resistance segments (24). The present finding is consistent with many of the previous reports of insignificant changes in GFR during administration of NO synthase inhibitors (19, 25, 26). Taken together, it is conceivable that cholesterol-induced renal vascular and glomerular function is mainly mediated by its inhibitory actions on endogenous NO production.

It can be argued that a marked reduction in RBF in l-NAME pretreated rats might have limited the further vasoconstriction induced by cholesterol. However, such possibility seems unlikely unless such vasoconstriction is caused by factors other than NO inhibition. It has been reported in many previous studies that the renal vasoconstriction can further be enhanced with administration of angiotensin II (4, 27, 29) or SOD inhibitor (25) in l-NAME pretreated animals. Thus a lack of cholesterol-induced vasoconstriction in l-NAME treated animals can only be explained by a mechanism that involved a decrease in NO production during cholesterol administration.

It has also been suggested that cholesterol may alter distal nephron sodium reabsorption via changes in membrane lipid composition (3). This in turn could lead to the enhancement of sodium transporter activity (3, 5, 14, 35–38). As shown in several studies, membrane cholesterol extraction decreases and cholesterol enrichment increases sodium transport in A6 renal epithelial cells derived from Xenopus laevis (5, 14, 36–38). Moreover, our recent data also indicate an effect of membrane order or fluidity on ENaC activity (3). However, in our present in vivo study, we observed no appreciable change in the extent of anti-natriuretic responses induced by cholesterol in rats pretreated with an effective dose of amiloride to inhibit distal tubular ENaC activity that caused two- to threefold increases in U_NaV without altering its filtered load (23). On the other hand, inhibition of NKCC by furosemide prevented anti-diuretic and anti-natriuretic response induced by cholesterol but not its vascular effect in the kidney. Thus our data indicate that the acute cholesterol-induced tubular response involves NKCC predominantly. It needs to be mentioned here that NO-induced inhibition of tubular sodium reabsorption also involves NKCC present in thick ascending limb (TAL) of the loop of Henle (13, 35). Thus it is likely that cholesterol infusion influences NKCC activity due to its inhibiting effect on NO production. Alternatively, cholesterol may be involved directly in the regulation of sodium transport by NKCC.

The mechanism how cholesterol inhibits NO production has been postulated by many earlier studies (11, 12, 40). It has been demonstrated that high cholesterol levels decrease the production of NO in endothelial cells by upregulating the abundance of the structural protein caveolin and promoting its inhibitory interaction with eNOS enzyme. Another study (28) suggests that increased membrane cholesterol content can directly influence NO diffusion through the membrane and thus affect NO intracellular signaling. It has been shown that cholesterol increases caveolin-1 expression, which binds NO synthase, thus inhibiting its activity and, consequently, NO release (1, 10, 11, 40). Hypercholesterolemia has been shown to decrease renal NO synthase activity (1, 2, 11, 30). Endothelial dysfunction observed in hypercholesterolemic animals and patients is closely associated with reduced NO production (2, 10, 15, 17, 41). An impairment of renal vascular function was also observed in isolated kidney from hypercholesterolemic rats (30). In the kidney, NO controls not only vascular tone but also regulates tubular epithelial sodium transport (13, 20, 23, 24). Collectively, these findings support the notion that an increase in cholesterol level modulates renal hemodynamics and excretory function in a way that compromises the kidney's ability to excrete salt and water appropriately, and thus this mechanism contributes to the development of hypertension in hypercholesterolemic conditions. Inhibiting NO production by cholesterol could be a major factor in the pathophysiology of the renal disorders in hypercholesterolemic conditions. Future experiments are needed to determine the exact link between hypercholesterolemia, reduced NO production, and hypertension.

Hypercholesterolemia also leads to increased O2− activity in the rat kidney (2, 11, 16, 31), possibly due to increased renal expression of NAD(P)H oxidase subunits and reduced SOD activity (8). However, cholesterol-induced renal responses were not appreciably altered in tempol pretreated animals except that there was a slight attenuation in its anti-natriuretic effect. Thus these present data do not support a major contribution of O2− activity in mediating these renal effects of cholesterol.

In conclusion, the findings in the present study demonstrate that cholesterol-induced renal vasoconstrictor and anti-natriuretic responses are mediated by a decrease in NO production. These data also indicate that Na⁺/K⁺/2Cl⁻ cotransport in the thick ascending limb of the loop of Henle is involved in the acute antinatriuretic response to cholesterol.

GRANTS

This study was supported by National Heart, Lung, and Blood Institute Grant HL-66432, NIDDK 55626, and Tulane Enhancement Fund. L. Kopkan is currently supported by grants IGA NS/9699-4/2008, GAAV KJB 502030801, and partly supported from the Center for Cardiovascular Research (1M6798582302) and the institutional financial support of the Institute for Clinical and Experimental Medicine (MZO 00023001).

DISCLOSURES

No conflicts of interest are declared by the authors.