Glomerulosclerosis in the diet-induced obesity model correlates with sensitivity to nitric oxide inhibition but not glomerular hyperfiltration or hypertrophy

Abstract

The diet-induced obesity (DIO) model is frequently used to examine the pathogenesis of obesity-related pathologies; however, only minimal glomerulosclerosis (GS) has been reported after 3 mo. We investigated if GS develops over longer periods of DIO and examined the potential role of hemodynamic mechanisms in its pathogenesis. Eight-week-old male obesity-prone (OP) and obesity-resistant (OR) rats (Charles River) were administered a moderately high-fat diet for 5 mo. Radiotelemetrically measured blood pressure, proteinuria, and GS were assessed. OP (n = 10) rats developed modest hypertension (142 ± 3 vs. 128 ± 2 mmHg, P < 0.05) and substantial levels of proteinuria (63 ± 12 vs. 12 ± 1 mg/day, P < 0.05) and GS (7.7 ± 1.4% vs. 0.4 ± 0.2%) compared with OR rats (n = 8). Potential hemodynamic mechanisms of renal injury were assessed in additional groups of OP and OR rats fed a moderately high-fat diet for 3 mo. Kidney weight (4.3 ± 0.2 vs. 4.3 ± 0.1 g), glomerular filtration rate (3.3 ± 0.3 vs. 3.1 ± 0.1 ml/min), and glomerular volume (1.9 ± 0.1 vs. 2.0 ± 0.1 μm³ × 10⁻⁶) were similar between OP (n = 6) and OR (n = 9) rats. Renal blood flow autoregulation was preserved in both OP (n = 7) and OR (n = 7) rats. In contrast, N^ω-nitro-l-arginine methyl ester (l-NAME) administration in conscious, chronically instrumented OP (n = 11) rats resulted in 15% and 39% increases in blood pressure and renal vascular resistance, respectively, and a 16% decrease in renal blood flow. Minimal effects of l-NAME were seen in OR (n = 9) rats. In summary, DIO-associated GS is preceded by an increased hemodynamic sensitivity to l-NAME but not renal hypertrophy or hyperfiltration.

the growing epidemic of obesity is widely recognized to contribute to major cardiovascular risk factors that include hypertension, diabetes, and metabolic syndrome. As extensively discussed in recent review articles (38, 44, 48, 75, 81), obesity has also been linked with the development and progression of chronic kidney disease. Moreover, while the presence of hypertension and/or diabetes that often accompany obesity likely strongly influence this association, recent clinical evidence suggests that obesity per se may promote the development of nephropathy even in their absence. The renal lesion typically associated with obesity nephropathy per se is that of focal segmental glomerulosclerosis (GS); however, its pathogenesis has remained poorly understood (38). It has been widely postulated that obesity-associated glomerular hyperfiltration and hemodynamic stress result in GS in a manner analogous to that postulated for chronic kidney disease states (38, 44, 48, 75, 81). However, unlike renal mass reduction models of chronic kidney disease, GS has been observed infrequently in animal models of obesity (23–26, 43, 45, 46) except in those due to leptin receptor mutations (21, 35, 49, 60), such as the obese Zucker rat and substrains derived from it. However, interpretations as to the potential role of the leptin pathway in the pathogenesis of obesity-associated GS are confounded by the observation that the leptin-deficient ob/ob mouse develops obesity but does not exhibit significant renal pathology (16, 76). Moreover, given the somewhat limited relevance of these monogenic obesity models to most human obesity, investigators have increasingly focused on polygenic diet-induced obesity (DIO) models.

The high-fat DIO model in the Sprague-Dawley rat closely mimics the associated pattern of metabolic and cardiovascular abnormalities, and, as in humans, its polygeneic pathogenesis is dependent on excess caloric intake. Sprague-Dawley rats fed a moderately high-fat (MHF) diet (∼30% kcal from fat) for a period of several months diverge into a bimodal distribution based on weight gain corresponding to obesity-prone (OP) and obesity-resistant (OR) groups (56, 58). Moreover, drugs that produce significant weight loss in obese humans have shown similar efficacy in the DIO model (63, 77). The DIO model has, therefore, been considered one of the best surrogates of human obesity to investigate the underlying mechanisms of obesity-related pathologies (63). However, although OP rats exhibit hypercholesterolemia, hypertriglyceridemia, activation of the renin-angiotensin and sympathetic nervous systems, oxidative stress, and modest levels of hypertension (18, 26–28, 57), very limited evidence of histological renal injury has been reported after 10 wk of MHF diet, which consists primarily of mesangial expansion (26). A similar paucity of quantitatively significant GS has been seen in the DIO model in the mouse (23, 24, 46). Therefore, the purpose of the present study was to examine if long-term (∼5 mo) administration of a MHF diet to OP and OR rats would result in the development of significant GS and allow an examination of its relationship to the commonly postulated hemodynamic initiators of obesity-associated GS.

METHODS

Animals

Male OP and OR Sprague-Dawley rats were purchased at 6–8 wk of age from Charles River, where they are maintained as separate outbred colonies and fed a standard chow diet. As previously described (58), these colonies of OP and OR rats were derived from an outbred line of Sprague-Dawley rats from Charles River through selective breeding based on weight gain phenotype during consumption of a high-energy diet. All rats were fed a standard chow diet (13% kcal from fat, 1% NaCl, LabDiet 5001) upon arrival at the Hines Veterans Affairs Hospital. After baseline measurements, the diet was switched to a MHF diet (32% kcal from fat, 0.5% NaCl, Research Diets D12266B) for the duration of the experiments. Water was provided ad libitum. All animals were cared for in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals, and all protocols were approved by the Hines Veterans Affairs Institutional Animal Care and Use Committee.

Experimental Procedures

Four experiments were performed in separate groups of OP and OR rats. Experiment 1 examined the effect of prolonged (5 mo) administration of a MHF diet on radiotelemetrically measured blood pressure (BP) and renal injury. Experiments 2–4 were performed after 3 mo of administration of a MHF diet, before the development of substantial renal injury (26), to characterize potential antecedent hemodynamic mechanisms of obesity-associated renal injury, which included 1) glomerular filtration rate (GFR) measurements in conscious rats and the associated morphometrically determined glomerular volume, 2) steady-state renal blood flow (RBF) autoregulatory capacity after step changes in BP in anesthetized rats, and 3) BP-RBF relationships and dynamic autoregulation of RBF before and during the administration of N^ω-nitro-l-arginine methyl ester (l-NAME) in conscious rats.

Experiment 1: BP and renal injury responses after 5 mo of MHF feeding.

OP and OR rats were chronically instrumented with a radiotelemetric BP transmitter (model TA11PA-C40, Data Sciences, St. Paul, MN) for the continuous assessment of BP (every 10 min for 10 s, 24 h/day), as previously described (9, 35, 37, 39, 40). In one group of OP (n = 4) and OR (n = 4) rats, BP transmitters were implanted 1 wk before the administration of the MHF diet so that BP could be assessed at baseline, before the administration of the MHF diet, and over the entire 5-mo duration of MHF diet feeding. In another group of OP (n = 6) and OR (n = 4) rats, BP transmitters were implanted 1 wk before the fifth month of administration of the MHF diet so that BP was continuously monitored during the final month of the protocol. In either case, the average systolic BP measured over the entire duration of implantation (i.e., 5 or 1 mo, respectively) was used to characterize the BP of an individual rat during MHF feeding. A 24-h urine collection was performed at baseline and at the end of the protocol in all rats. Proteinuria (in mg/day) was measured using the sulfosalicylic acid method, as previously described (9, 35–37, 39, 40). At the completion of the study, rats were euthanized, and kidneys were cleared via retrograde perfusion with 0.9% saline followed by modified Karnovsky's fixative [2% (wt/vol) paraformaldehyde and 2.5% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.4] for histological assessment.

Experiment 2: ambient GFR and glomerular volume after 3 mo of MHF feeding.

OP (n = 6) and OR (n = 9) rats fed the MHF diet for 3 mo were anesthetized [pentobarbital sodium (50 mg/kg ip)], and an osmotic minipump (2ML1, Durect, Cupertino, CA) containing 10 mg/ml FITC-inulin (Sigma-Aldrich, St. Louis, MO) dissolved in 10 mM PBS (pH 7.4) was surgically implanted (subcutaneously) in the subscapular region for the determination of GFR in the conscious state, similar to methods previously described by others (15, 59). Three days after surgery, rats were placed in metabolic cages, and a 24-h urine collection and tail vein blood sample were obtained. After the 24-h urine collection, rats were anesthetized, and the kidneys were perfusion fixed for morphometric analysis.

Experiment 3: steady-state step renal autoregulation after 3 mo of MHF feeding.

OP (n = 7) and OR (n = 7) rats fed the MHF diet for 3 mo were anesthetized [inactin (100 mg/kg ip)] and surgically prepared for the assessment of renal autoregulatory capacity, as previously described (11, 14, 36, 37, 39, 40). Autoregulatory indexes were calculated as the fractional change in RBF/fractional change BP, as previously described (1, 11, 14, 36, 37, 39, 40).

Experiment 4: ambient renal hemodynamics and dynamic autoregulation after 3 mo of MHF feeding.

OP (n = 11) and OR (n = 9) rats fed the MHF diet for 3 mo were anesthetized [pentobarbital sodium (50 mg/kg ip)] and chronically instrumented with a radiotelemetric BP transmitter and a Transonic RBF probe on the left renal artery, as previously described (1, 11, 34, 35, 37). After 7 days of recovery from surgery, baseline recordings (2–4 h at 200 Hz) of BP and RBF were made in conscious rats. After one to three of such recordings at 24-h intervals, l-NAME was administered in drinking water at a dose of 100 mg/l (∼10 mg·kg⁻¹·day⁻¹). Starting at least 48 h after the initiation of l-NAME, BP and RBF recordings were again obtained. After 4–6 days and two to three of such recordings, the dose of l-NAME was increased to 250 mg/l (∼25 mg·kg⁻¹·day⁻¹), and BP and RBF recordings were obtained in a similar manner. Averaged results over the two or three separate 2- to 4-h recordings of BP and RBF in each rat before and during both doses of l-NAME administration were used to characterize the ambient baseline renal hemodynamics and effects of l-NAME on such measurements.

Histological and Morphometric Methods

Transverse sections (3 μm) of paraffin-embedded kidneys were stained with hematoxylin and eosin and periodic acid-Schiff for histological analyses. Glomerular injury was quantified in a blinded fashion as the total percentage of glomeruli (at least 100 glomeruli/section) exhibiting lesions of segmental or global GS, as previously described (9, 34–37, 39, 40). The glomerular cross-sectional area was measured using Image-Pro Premier image-analysis software (version 9.0, MediaCybernetics, Rockville, MD) using a ×20 objective lens on a DMCD Leica microscope (Wetzlar, Germany) equipped with a DFC295 Leica digital camera in periodic acid-Schiff-stained sections, as previously described (39–41). The glomerular cross-sectional area of 100 glomeruli/kidney section was evaluated by two individuals whose values were then averaged. The mean glomerular volume was calculated as follows: glomerular volume = β/κ(A_G^3/2), where A_G is glomerular cross-sectional area, β is size distribution coefficient (equal to 1.38), and κ is the shape coefficient for glomeruli idealized as spheres (equal to 1.1) (62, 79).

Determination of GFR

FITC-inulin standards were made from a 10 mg/ml FITC-inulin solution by serial dilution in 10 mM HEPES buffer (pH 7.4, Sigma-Aldrich) in concentrations ranging from 0.15 to 20 μg/ml. Urine samples were prediluted in HEPES buffer (1:100), and samples, plasma, and prediluted urine were then further diluted in HEPES buffer (1:39). Standards, samples, and blank (HEPES buffer) were run in triplicate (200 μl/well) on a black 96-well microplate (Costar). Plasma and urine from rats not infused with FITC-inulin were included to correct for differences in background fluorescence. Fluorescence was measured using a fluorescent microplate reader (Synergy HT, BioTek, Winooski, VT) with excitation and emission filters of 485 and 530 nm, respectively. GFR was calculated by dividing the excretion rate of FITC-inulin by its plasma concentration.

Dynamic RBF Autoregulation

Transfer function analysis between BP (input) and RBF (output) was performed at baseline and at both doses of l-NAME using previously published methods (1, 11, 34, 35, 37, 66). The natural frequencies of myogenic and tubuloglomerular feedback mechanisms were determined from their characteristic signature resonance peaks in fractional gain in admittance between 0.1 and 0.3 Hz and between 0.025 and 0.05 Hz, respectively, by inspection of individual records and averaged across each record. We used several other components of the transfer function analyses that have been considered potential indexes of the strength of the renal myogenic mechanism, including 1) the slope of admittance magnitude reduction immediately below the myogenic peak, 2) the peak phase associated with the frequency of the signature resonance peak of the myogenic mechanism, and 3) the average coherence between 0.5 and 1 Hz (34, 66).

Statistical Analysis

Results are expressed as means ± SE. The nonparametric Mann-Whitney test was used to examine whether statistically significant differences in glomerular injury existed between OP and OR rats. Two-way repeated-measures ANOVA was used to compare baseline and final body weight and proteinuria between OP and OR rats as well as hemodynamic variables at baseline and during l-NAME administration between OP and OR rats. A Student-Newman-Kuels post hoc test was used for multiple comparisons. An unpaired t-test was used to examine whether statistically significant differences were present among all other variables in OP versus OR rats. Linear regression analysis was used to calculate the slope of relationship between GFR and glomerular volume. P values of <0.05 were considered statistically significant.

RESULTS

Effect of a 5-mo Administration of a MHF Diet on Body Weight, Kidney Weight, BP, and Renal Injury

At 6–8 wk of age before the administration of a MHF diet, baseline body weight tended to be higher in OP versus OR rats but did not reach statistical significance (P = 0.16; Table 1). After administration of the MHF diet, OP rats gained a significantly (P < 0.01) greater amount of weight compared with OR rats (net gain of 527 ± 20 vs. 444 ± 16 g, respectively), resulting in significant (P < 0.001) differences in final body weight. Similarly, whereas no differences in proteinuria were observed at baseline, OP rats developed significantly greater levels of proteinuria compared with baseline (P < 0.001) and compared with OR rats (P < 0.001) by the end of the protocol. In the four OP rats and four OR rats in which baseline BP was obtained, systolic BP was modestly but significantly greater (P < 0.05) in OP rats. Similarly, during MHF diet feeding, OP rats (n = 10) exhibited modest but significantly greater (P < 0.005) systolic BP compared with OR rats (n = 8). Yet, despite the modest level of hypertension observed in OP rats, they developed significant (P < 0.001) levels of GS compared with the minimal levels observed in OR rats. Vascular and tubulointerstitial injury were not observed in kidneys from either OP or OR rats. Finally, although absolute kidney weight (sum of both kidneys) was higher in OP versus OR rats, this difference did not reach statistical significance (P = 0.06) despite the significant differences in body weight. Conversely, when normalized to body weight, kidney weight tended to be higher in OR versus OP rats, but this also did not reach statistical significance (P = 0.09).

Table 1.

Anthropometric, hemodynamic, and renal injury data in OP and OR rats fed a MHF diet for 5 mo

	Body Weight, g		Proteinuria, mg/day		Systolic Blood Pressure, mmHg
	Baseline	Final	Baseline	Final	Baseline	Final	Kidney Weight, g	Kidney Weight/Body Weight	Glomerulosclerosis, %
OP rats	173 ± 5	699 ± 19*†	5 ± 1	63 ± 12*†	132 ± 3†	142 ± 3†	5.0 ± 0.1	7.2 ± 0.9	7.7 ± 1.4†
OR rats	144 ± 4	588 ± 18*	6 ± 1	12 ± 1	120 ± 2	128 ± 2	4.7 ± 0.1	8.0 ± 0.3	0.4 ± 0.2

Values are means ± SE; n = 10 obese-prone (OP) rats and 8 obese-resistant (OR) rats. MHF, moderately high-fat diet.

^*P < 0.05 vs. baseline;

^†P < 0.05 vs. OR rats.

GFR and Glomerular Volume in Rats Administered the MHF Diet for 3 mo

These experiments were performed to precisely assess GFR and glomerular volume in OP and OR rats administered a MHF diet for 3 mo before the development of significant renal injury. At 3 mo after the administration of the MHF diet, the body weight of OP rats (729 ± 15 g) was significantly (P < 0.001) greater compared with OR rats (455 ± 7 g; Fig. 1). Nevertheless, despite the substantial differences in body weight, kidney weight (4.3 ± 0.2 vs. 4.3 ± 0.1 g, respectively, P = 0.8), GFR (3.3 ± 0.3 vs. 3.1 ± 0.1 ml/min, respectively, P = 0.6), and glomerular volume (1.9 ± 0.1 vs. 2.0 ± 0.1 μm³ × 10⁻⁶, respectively, P = 0.4) were similar between OP and OR rats. Of note, GFR and glomerular volume were significantly correlated in the collective group of OP and OR rats (Fig. 1D). These data demonstrate that OP rats exhibit neither glomerular hyperfiltration nor hypertrophy compared with OR rats at 3 mo after the administration of the MHF diet.

Fig. 1.

A–C: there were no significant differences in kidney weight (A), glomerular filtration rate (GFR; B) or glomerular volume (C) observed between OP and OR rats after 3 mo of administration of a moderately high-fat (MHF) diet. D: a significant correlation was noted between GFR and glomerular volume in OP and OR rats. Values are expressed as means ± SE.

Steady-State Step Renal Autoregulation in Rats Administered the MHF Diet for 3 mo

These experiments were performed to examine the ability of the renal vasculature to buffer large steady-state changes in mean arterial pressure (MAP) under anesthesia, which provides the best index of susceptibility to hypertensive-induced renal damage (11). At 3 mo of administration of the MHF diet, the body weight of OP (n = 7) rats was significantly (P < 0.001) greater compared with OR (n = 7) rats (561 ± 23 vs. 427 ± 9 g, respectively). However, RBF autoregulation was well preserved in both OP and OR rats, as evidence by their similar autoregulatory indexes (0.1 ± 0.2 vs. 0.1 ± 0.1, respectively).

Ambient Renal Hemodynamics and Dynamic Autoregulation Before and After l-NAME in Conscious Rats Administered the MHF Diet for 3 mo

At 3 mo after the administration of the MHF diet, OP rats were significantly (P < 0.0001) heavier compared with OR rats (595 ± 16 vs. 411 ± 14 g, respectively) at the time of BP radiotelemetry and RBF probe implantation (Fig. 2). At 1 wk after surgery, no significant differences in baseline BP, renal vascular resistance (RVR) and RBF assessed over 2–3 separate days were noted between conscious OP and OR rats. In contrast, very robust hemodynamic differences were seen after the administration of increasing doses of l-NAME. For example, 100 and 250 mg/l l-NAME led to progressive increases in MAP in OP rats compared with baseline levels (11% and 15% increases, respectively). In contrast, whereas 100 mg/l l-NAME led to a significant 8% increase in MAP in OR rats, the response to 250 mg/l l-NAME was greatly blunted and not significantly different from baseline levels. MAP was significantly greater in OP versus OR rats during the administration of 250 mg/l l-NAME. A similar pattern was noted with respect to RVR responses after l-NAME administration in OP rats as RVR increased by 31% and 39% with 100 and 250 mg/l l-NAME, respectively, compared with baseline levels. In contrast, RVR only modestly increased (5%) in response to 100 mg/l l-NAME in OR rats and fell to baseline levels in response to 250 mg/l l-NAME; however, neither change was significantly different from baseline values. The results of two-way repeated-measures ANOVA indicated that RBF was significantly lower in OP versus OR rats across the entire protocol, which was mainly manifest during l-NAME administration. For example, baseline RBF was similar in OP versus OR rats (9.2 ± 0.5 vs. 10.0 ± 0.6 ml/min, respectively); however, the administration of 100 and 250 mg/l l-NAME led to 14% and 16% reductions in RBF in OP rats compared with baseline levels, whereas RBF was largely unaffected by l-NAME in OR rats. In summary, these data demonstrate that while baseline BP, RVR, and RBF were similar in conscious OP and OR rats fed a MHF diet for 3 mo, significant differences in the hemodynamic response to l-NAME were observed in the conscious state.

Fig. 2.

A–C: mean arterial pressure (MAP; A), renal vascular resistance (RVR; B), and renal blood flow (RBF; C) in conscious OP and OR rats after 3 mo of administration of the MHF diet. Hemodynamic data were obtained before (i.e., baseline) and after escalating doses of N^ω-nitro-l-arginine methyl ester (l-NAME) in the drinking water. In general, OP rats exhibited an increased sensitivity to l-NAME administration, as demonstrated by greater increases in MAP and RVR and a tendency for greater decreases in RBF, compared with OR rats. Values are expressed as means ± SE. *P < 0.05 vs. baseline; †P < 0.05 for the OR group at the respective time point.

The transfer function analysis between BP (input) and RBF (output) in OP and OR rats before and after l-NAME is shown in Table 2. Because no significant differences in any component of the transfer functions were observed between the two doses of l-NAME (i.e., 100 vs. 250 mg/l l-NAME) in either OP or OR rats, these data were averaged and presented as a single value for l-NAME administration for presentation clarity. Aside from a faster operating frequency of the myogenic mechanism in OP versus OR rats, no significant differences were seen in any of the transfer functions classically used to assess the strength of the myogenic mechanism (i.e., fractional gain, phase peak, coherence, and slope of gain reduction) during baseline (34, 66). As expected, l-NAME administration increased the operating frequency and strength of the myogenic mechanism, as indicated by a higher phase peak and slope of gain reduction in both groups compared with baseline, although such changes were larger in OR versus OP rats. This was especially true with respect to the fractional gain of the signature resonance peak, which was significantly increased only in OR rats in response to l-NAME. No significant differences in transfer function components associated with the tubuloglomerular feedback mechanism were observed between OP and OR rats at baseline or in response to l-NAME administration.

Table 2.

Transfer functions related to myogenic and tubuloglomerular feedback autoregulatory mechanisms before and after l-NAME administration in conscious OP and OR rats fed the MHF diet for 3 mo

	Myogenic					Tubuloglomerular Feedback
	Frequency, Hz	Fractional gain	Phase peak, °	Coherence at 0.05–0.1 Hz	Slope of the gain reduction, db/decade	Frequency, Hz	Fractional gain
OP rats
Baseline	0.26 ± 0.01	2.0 ± 0.2	113 ± 11	0.5 ± 0.05	26 ± 3	0.04 ± 0.002	0.4 ± 0.1
l-NAME	0.24 ± 0.01	2.4 ± 0.2	126 ± 6	0.5 ± 0.06	30 ± 3	0.04 ± 0.003	0.5 ± 0.1
OR rats
Baseline	0.23 ± 0.01	2.3 ± 0.2	92 ± 6	0.5 ± 0.03	29 ± 2	0.04 ± 0.003	0.5 ± 0.1
l-NAME	0.21 ± 0.01	3.6 ± 0.6*†	116 ± 13	0.4 ± 0.04	37 ± 5	0.03 ± 0.002	0.5 ± 0.1
ANOVA effects
Strain	P < 0.01	P = 0.06	NS	NS	NS	NS	NS
l-NAME	P < 0.01	P < 0.01	P < 0.05	NS	P < 0.01	NS	NS
Interaction	NS	P < 0.05	NS	NS	NS	NS	NS

Values are means ± SE. N^ω-nitro-l-arginine methyl ester (l-NAME) values represent the average of transfer function values at 100 and 250 mg/l dosages of l-NAME, which were not significantly different from each other within OP and OR rats. NS, not significant.

^*P < 0.05 vs. baseline;

^†P < 0.05 vs. OP rats administered l-NAME.

DISCUSSION

Attempts to elucidate the pathogenesis of obesity-associated GS have had limited success, in part due to a lack of relevant obesity models that exhibit significant GS (38). In contrast to a previous study (26) in rodent DIO models, the present study clearly demonstrates that unambiguous and substantial GS develops in the DIO model after more prolonged feeding of the MHF diet (5 mo). As only OP rats but not OR rats developed such GS, this allowed an investigation of the potential antecedent mechanisms that may contribute to the development of obesity-associated GS in the DIO rat. In this context, while a large number of potential mechanisms have been proposed for the pathogenesis of obesity-associated GS, it is the hyperfiltration/hyperperfusion pathway that has received the greatest emphasis (38, 44, 48, 75, 81). It has been postulated that the increased metabolic needs associated with the increased body mass in obesity by necessity lead to an increase in absolute GFR. Therefore, the lack of hyperfiltration in OP rats in the present study seems unexpected. However, it is of note that increases in GFR are only indirectly related to body mass and are more a function of the increases in body surface area, metabolic rate, and/or lean body mass (17, 20, 29, 72, 80). Given these considerations and the many variables that can impact GFR measurements, such as age, dietary protein intake, etc., it is perhaps not surprising that increases in GFR and/or renal perfusion have been found in some but not all human obese populations (38, 44, 48, 75, 81). Data obtained in animal models have, if anything, displayed even greater variability as well as species differences. While the DIO model in the dog is consistently associated with significant glomerular hyperfiltration (42, 45), this is observed much less consistently in obese rodents, with reports of increased (60, 65), unchanged (25, 49, 64, 70), or even decreased (68, 73) levels of GFR. Of greater relevance, the relationship between hyperfiltration and GS is also fairly inconsistent in these animal models. Despite significant increases in GFR, the dog obesity model fails to exhibit quantitatively significant GS (42, 45). Similarly, the melanocortin-4 receptor-deficient mouse, a monogenic model of obesity with elevated circulating leptin levels, does not exhibit significant GS even at 52–55 wk of age despite a long duration of obesity and the presence of several metabolic abnormalities consistent with metabolic syndrome (25). It should be acknowledged that even longer durations of high-fat feeding or obesity may be required to elicit significant GS in these models and that the development of such GS may interact, in part, with the pathogenesis of aging GS per se, particularly in rodents. Nevertheless, these data suggest that hyperfiltration, per se, in obesity may not be sufficient to result in GS. These interpretations are in accordance with data obtained in several nonobesity models. For instance, experiments from our laboratory have consistently demonstrated that hyperfiltration and hyperperfusion, per se, are not injurious, as evidenced by the lack of GS in normotensive models of renal mass reduction that exhibited comparable significant elevations in single nephron GFR (39). Similarly, additional hyperfiltration superimposed by repetitive pregnancies in the uninephrectomized female Munich Wistar rat fails to result in GS (5). Strong clinical parallels indicating the generally benign consequences of hyperfiltration per se are provided by the lack of significant GS in the vast majority of women with multiple pregnancies or in kidney donors (7, 55, 67).

The present study extends these interpretations by demonstrating that hyperfiltration is not even necessary for the development of GS in this rodent obesity model. Of interest, significant GS without antecedent hyperfiltration has also been reported in another model of obesity induced by bilateral hypothalamic electrolytic lesions in the rat (6). Collectively, these data strongly suggest that even when hyperfiltration and GS are observed together in obesity, the relationship is likely not causal. Indeed, it is worth emphasizing that even in the seminal studies that first linked hyperfiltration to GS in renal ablation models, it was the increased glomerular pressures rather than hyperfiltration per se that was demonstrated to be pathogenic (2). Although the increased glomerular pressure was initially postulated to be integral to the hyperfiltration response, subsequent studies (12, 41) in normotensive models of renal mass reduction have shown that comparable glomerular hyperfiltration can be achieved by coordinated increases in glomerular filtration surface area (hypertrophy) and plasma flow without pathogenic increases in glomerular pressure (12, 41). The consistent presence of glomerular hypertrophy with increased glomerular capillary surface area in chronic hyperfiltration states is consistent with such interpretations. Accordingly, we suggested that the elevated glomerular pressure observed in hypertensive hyperfiltration states is a consequence of the glomerular transmission of coexistent systemic hypertension rather than being intrinsic to the hyperfiltration phenomena, per se (vide infra) (8, 13, 38). A recent study (55) in long-term kidney donors has confirmed both the generally benign course of hyperfiltration and its mediation by increased renal plasma flow and glomerular capillary surface area. The absence of glomerular hypertrophy in the present study is thus in accordance with the observed lack of hyperfiltration and/or hyperperfusion in OP rats. Of relevance, the lack of differences in glomerular hypertrophy in OP versus OR rats also reduces the likelihood that reduced podocyte density, a well-recognized consequence of such hypertrophy (53), is playing a major role in the development of obesity-associated GS in OP rats.

Similar to a recent study (22) in which BP was measured in OP and OR rats via radiotelemetry methods, our study showed that BP was modestly but significantly higher in OP versus OR rats at baseline and during administration of the MHF diet. Given the clearly demonstrated importance of elevated systemic and glomerular pressures in the pathogenesis of GS in nonobesity models (8, 10, 13, 38), the increased average systolic BP observed in OP versus OR rats on a MHF diet is likely relevant. However, such modest levels of hypertension, as observed in OP rats, usually do not lead to substantial proteinuria or GS in the absence of mechanisms that facilitate its transmission to glomerular capillaries, such as preglomerular vasodilation and an impairment of the normally protective renal autoregulatory responses that are present in renal mass reduction models. Although there is some controversy as to the relative merits of the methods used to assess autoregulatory efficiency/capacity, the susceptibility to hypertension-induced renal damage seems to correlate best with the ability of the renal vasculature to respond to large steady-state step changes in BP (11). This was found to be intact in both OP versus OR rats at 3 mo after the initiation of the MHF diet and before the development of substantial GS. However, given that BP does not change from one steady-state level to the next in the conscious state, such steady-state assessments of renal autoregulation may not address the ability of the vasculature to respond to more modest but rapid BP fluctuations (8, 11, 13). Therefore, we also examined the kinetics of the two major components of the renal autoregulatory response, myogenic and tubuloglomerular feedback mechanisms, using transfer function methodology to assess dynamic autoregulation in a separate group of conscious chronically instrumented OP versus OR rats also fed the MHF diet for 3 mo. Similar to the steady-state step autoregulation results, no significant differences were observed between OP and OR rats in any of the transfer functions regarding the strength of either the myogenic or tubuloglomerular feedback mechanism (34, 66). Moreover, the operating frequency of the myogenic mechanism was actually higher in OP versus OR rats, which would suggest a faster renal autoregulatory buffering of BP transmission. Thus, these autoregulatory data combined with the lack of evidence of renal hyperperfusion strongly rule out a role for preglomerular vasodilation and/or impaired autoregulation in contributing to the greater levels of GS in OP versus OR rats observed in the present study.

In contrast, significant differences in BP and RBF responses to l-NAME were observed between OP and OR rats, which may be relevant to the pathogenesis of GS in OP rats. We have previously suggested that carefully quantitated BP responses to nitric oxide (NO) inhibition in conscious rats may serve as a surrogate marker for the susceptibility to hypertensive glomerular injury (34). Rodent strains/colonies that exhibit greater hypertensive and/or renal vasoconstrictor responses to l-NAME also tend to exhibit greater susceptibility to hypertensive GS (Wistar, Sprague-Dawley colony from Harlan, Dahl) (4, 19, 31, 34). Conversely, rodent strains/colonies that exhibit relatively attenuated BP and renal vasoconstrictor responses to l-NAME tend to be resistant (Wistar Furth, Sprague-Dawley colony from Charles River, Brown-Norway rats, and C57BL6 mice) (31, 34, 52, 54, 61). The mechanisms responsible for these associations remain to be definitively established. However, based on observations by us and others, we have suggested that the pathophysiological basis of these relationships likely stems from the underemphasized vasodilatory role of NO at the efferent arteriole (34). Topographic studies (3, 30) of the distribution of NO synthase (NOS) isoforms have shown significant expression of both NOS1 and NOS3 in efferent arteriolar endothelial cells. A micropuncture study (4) has similarly shown efferent arteriolar constriction to play a significant role in the development of glomerular hypertension and injury in the l-NAME-treated model. More significantly, quantitative differences have been observed between genetic strains in the efferent arteriolar and glomerular pressure responses to NOS blockade that correlate with proteinuria and susceptibility to GS (31). In this context, it is of note that NO is a vasodilator of the afferent arteriole and inhibits preglomerular autoregulatory responses (47, 71, 78) and the blockade/loss of such NO effects per se would be expected to reduce BP transmission and GS, contrary to what is observed. The expected consequences of loss of NO at the efferent arteriole to increase glomerular pressure, proteinuria, and GS provide a mechanism to resolve this apparent contradiction. The concordance between the hypertensive response to NO inhibition and GS susceptibility in various genetic strains/colonies additionally suggests that the efferent arteriolar sensitivity to NO loss may parallel BP responses. In this context, it is of note that although endothelial dysfunction has been widely documented in obesity and postulated to play a role in renal injury (22, 32, 33, 50, 51, 69, 74), the responsible mechanisms have not been defined. The proposed pathogenic construct based on the loss of NO at the efferent arteriole provides a potential mechanism to more directly link endothelial dysfunction to GS. It should also be acknowledged that decreased production of other renal vasodilators, such as prostaglandins, EDHF, etc., or increased production of vasoconstrictors, such as angiotensin II, sympathetic nerve activation, oxidative stress, etc., may contribute to the increased sensitivity of the renal vasculature to NO inhibition in OP versus OR rats.

In summary, the novel aspects of this study include 1) the demonstration that substantial GS develops in the OP strain of the DIO model, 2) the development of such obesity-related GS occurs in the absence of glomerular hypertrophy and hyperfiltration, and 3) the increased hemodynamic sensitivity to l-NAME administration precedes the development of obesity-related GS in OP rats. These data have very important implications regarding the potential role of decreased NO in the pathogenesis of GS in this DIO model.

GRANTS

This work was supported by Office of Research and Development of the Department of Veterans Affairs Career Development Award 1IK2BX001285 (to A. J. Polichnowski) and a Merit Review Award (to K. A. Griffin) and by National Institute of Diabetes and Digestive and Kidney Diseases Grants DK-40426 (to A. K. Bidani) and DK-61653 (to K. A. Griffin).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: A.J.P., A.K.B., and K.A.G. conception and design of research; A.J.P. and H.L.-V. performed experiments; A.J.P., M.M.P., J.L., R.B., G.A.W., A.K.B., and K.A.G. analyzed data; A.J.P., M.M.P., G.A.W., A.K.B., and K.A.G. interpreted results of experiments; A.J.P. prepared figures; A.J.P., A.K.B., and K.A.G. drafted manuscript; A.J.P., G.A.W., A.K.B., and K.A.G. edited and revised manuscript; A.J.P., H.L.-V., M.M.P., J.L., R.B., G.A.W., A.K.B., and K.A.G. approved final version of manuscript.