Dietary trans-10, cis-12 conjugated linoleic acid reduces early glomerular enlargement and elevated renal cyclooxygenase-2 levels in young obese fa/fa zucker rats.

Conjugated linoleic acid (CLA) slows the progression of disease in models of chronic kidney disease. Because obesity is associated with nephropathy and increased renal cyclooxygenase (COX) levels, the effects of dietary CLA on kidney function, morphology, and COX protein levels in the kidneys of young obese (fa/fa) Zucker rats, a model of metabolic syndrome, were examined. In study 1, 6-wk-old fa/fa and lean Zucker rats were given a mixture of CLA isomers (1.5% CLA, wt:wt) or the control diet (CTL) with no CLA for 8 wk. To examine specific isomer effects, study 2 used the same model with the following diets: 0.4% (g/g) cis-9, trans-11 (c9,t11) CLA; 0.4% trans-10, cis-12 (t10,c12) CLA; a combination of these 2 isomers (0.4% each); or CTL diets with no CLA. In study 1, fa/fa rats given the CLA mixture had 11% smaller kidney weights and 28% smaller glomeruli, and feed intake and body weight did not differ from the CTL rats. In study 2, diet also did not affect body weights, but fa/fa rats given a diet containing t10,c12 CLA had 7% lower kidney weights, 20% smaller glomeruli, and 39% lower COX-2 protein levels than CTL rats. In conclusion, dietary t10,c12 CLA reduces the enlargement of glomeruli in young obesity-associated nephropathy and is associated with lower protein levels of renal COX-2. Long-term studies with CLA supplementation are required to determine whether these changes would lead to reduction in development of renal disease associated with obesity.


Introduction
Conjugated linoleic acid (CLA) 8 describes a group of positional (carbon 7-12) and geometric (cis-cis, cis-trans, trans-cis, and trans-trans) isomers of octadecadienoic acid in which the double bonds are conjugated (1). CLA in the food supply is found in dairy products and ruminant meats, in which the predominant isomer is cis-9, trans-11 (c9,t11). The next most abundant isomer naturally occurring in food is trans-7, cis-9 (t7,c9), followed by c11,t13, t8,c10, and t10,c12. The formation of synthetic CLA also can occur under strong alkaline conditions during commercial hydrogenation of linoleic acid-rich vegetable oils at high temperatures, resulting in the formation of predominantly c9,t11 and t10,c12 isomers (2,3).
CLA has been associated with several health benefits, including antiadipogenic, antidiabetogenic, anticarcinogenic, and antiatherosclerotic effects. The mechanisms of these effects are hypothesized to involve modulation of prostanoid production, enhanced oxidation of lipids, cell apoptosis and/or differentiation, decreased cell proliferation, and alterations in gene expression (2,3). The majority of research investigating the effects of CLA has been with dietary supplementation of a CLA isomer mixture; however, recent studies have moved toward uncovering which CLA isomer(s) are responsible for the effects of CLA.
Although the concept of CLA supplementation in individuals with obesity and the metabolic syndrome currently is garnering much interest (4), its effects on the kidney in obesity are not known. Obesity and the metabolic syndrome, often described as a type 2 diabetes ''prediabetic'' state, are independently associated with increased risk of microalbuminuria and kidney disease (5,6). In chronic kidney disease models, CLA appears to reduce disease progression (7)(8)(9). Therefore, the hypothesis of the current study was that early CLA supplementation in young obese rats would alter one of the earliest signs of nephropathy associated with obesity, namely glomerular enlargement (10)(11)(12)(13).
Glomerular hypertrophy in obesity appears to result from chronically increased afferent arteriole flow caused by an increase in vasodilatory mediators such as prostanoids (12,14). Renal prostanoid production and rate-limiting enzymes involved in prostanoid production, such as cyclooxygenase (COX)-1 and COX-2, are altered in models of established kidney disease (7,(15)(16)(17)(18)(19)(20). CLA alters COX enzymes and their products in several other tissues (21)(22)(23)(24), possibly by reducing the arachidonic acid content of membrane phospholipids, by inhibiting COX-1 and/or COX-2 enzymes or by acting as an antagonist for COX-1 and/or COX-2. Because reducing the production of COX-derived prostanoids may be among the potential effects of CLA, we further hypothesized that renal COX-1 and COX-2 levels would be affected by CLA supplementation in young obese rats.
Thus, the objective of study 1 was to determine the effect of a CLA isomer mixture on glomeruli size and steady-state renal levels of COX-1 and COX-2 in young fa/fa rats. In study 2, the objective was to determine specific isomer effects by supplementing the diet with either c9,t11, t10,c12, or a combination of the 2 CLA isomers.

Materials and Methods
Animals and diets. Male fa/fa and lean Zucker rats (n ¼ 10 per group) were used in both studies; however, rats were obtained from different sources (Charles River for study 1 and Harlan for study 2). Rats were received at 5 wk of age and acclimatized for 5-7 d before the start of each study. The protocols for the animal care procedures were approved by the University of Manitoba Protocol Management and Review Committee.
Diet formulations were based on the AIN-93G diet (25) and contained 8.5% total lipid (wt:wt) in both studies ( Table 1). Dry ingredients were premixed and fresh batches of feed containing oil were prepared weekly and stored at 220°C. Both studies originally included a weight-matched group that was to be fed restricted amounts of control (0% CLA) diet such that their body weight would not differ from the treatment group with the lowest body weight, within each genotype. This approach was used because pair-feeding would not result in similar body weights if feed efficiency differed among the groups. During both studies, however, feed intake of the rats designated to be the weightmatched group did not have to be restricted. Statistical analyses also determined that final body weights did not differ within genotypes in both studies. Thus, the weight-matched group was removed from further analyses. Both studies lasted for 8 wk, during which all rats consumed the diet ad libitum with fresh feed 3 times/wk.
In study 1, fa/fa and lean rats were each randomly assigned to 1 of 2 treatment groups, where they were offered diet containing 1.5% (wt:wt) CLA (faCLA and lnCLA) or control (CTL) containing no CLA (faCTL and lnCTL) for 8 wk (n ¼ 10 in each group). The CLA mixture (86% purity, FFA form) contained 4 major isomers: t10,c12, c9,t11, c11,t13, and t8,c10, found at final amounts in the diet at 3.91, 3.83, 2.38, and 2.04 g/kg, respectively. The dose of CLA was chosen to parallel research in Zucker Diabetic Fatty rats demonstrating improved glucose tolerance and attenuated hyperinsulinemia (26). Details of the CLA composition, diet formulations, diet fatty acid composition, and other indices for these rats related to hepatic steatosis, pancreatic function, glucose disposal, inflammatory markers, adiposity, and adipokine status have been reported elsewhere (27)(28)(29).
Rats were housed individually in hanging stainless steel wire-bottom cages, and kept in a controlled environment with temperature maintained at 21-23°C, 55% relative humidity, and a 14-h-light:10-h-dark cycle. During this time, feed consumption (corrected for spillage) and weekly body weights were recorded. After 7 wk, rats were food deprived overnight (12 h) in metabolic cages for urine collection.  1 The dry ingredients included: cornstarch (363 g/kg), maltodextrin (132 g/kg), sucrose (100 g/kg), egg white (213 g/kg), cellulose (50 g/kg), AIN-93M mineral mix for study 1 (35 g/kg), AIN-93G mineral mix for study 2 (35 g/kg), AIN-93-VX vitamin mix (10 g/kg), biotin premix (10 g/kg) prepared with 200 g biotin/kg cornstarch, choline (2.5 g/kg), and tert-butylhydroquinone (0.014 g/kg). All dry ingredients, were purchased from Harlan Teklad, except cornstarch, which was supplied Best Foods (for study 1), and tertbutylhydroquinone, supplied by Fisher Scientific (study 1) or Aldrich Chemical (for study 2). 2 NuChek Prep. 3 Natural ASA. 4 Vita Health for study 1 and Harlan Teklad (study 2). Histology and image analysis. Formalin-fixed kidney cross sections were embedded in paraffin, sectioned at 5 microns, and processed using previously described methods for histologic analysis (32). After staining with hematoxylin and eosin, mean glomerular volume was determined as described (33). By random stage movement through the sections, the diameter of at least 30 randomly chosen glomeruli per kidney was measured by light microscopy with an Olympus BX60 microscope (Olympus Optical) using 203 magnification. After being captured using a Spot Junior CCD high-resolution camera (Diagnostic Instruments), images were analyzed using Image Pro Plus software (32). We converted the measured diameters to an estimate of glomerular volume using the formula developed by Hirose et al. (34). Observers for study 1 (L.W.W.) and study 2 (D.J.H.) were unaware of the treatments when performing the analysis.
Renal biochemistry. Serum creatinine and urinary creatinine were determined (35) and results were used to calculate creatinine clearance. We measured urinary protein using the method developed by Smith et al. (36).
Western immunoblotting. Western immunoblotting of steady-state protein levels of COX-1 and COX-2 was performed as described (18,19). Homogenates were centrifuged at 100,000 3 g; 35 min at 4°C and the supernatant containing the cytosolic fraction was removed. The remaining pellet was resuspended in 15-20 volumes of homogenization buffer containing 1% Triton X-100, incubated on ice for 10 min, and centrifuged at 100,000 3 g; 35 min at 4°C. The supernatant was stored at 280°C for later analysis of COX proteins. After SDS-PAGE, detection of COX-1 (1:250, Santa Cruz Biotechnology) and COX-2 (1:250, Cayman Chemical) was carried out by incubating blots overnight at 4°C with primary antibodies, followed by incubation for 1 h at room temperature with a peroxidase-conjugated secondary antibody (1:20,000, Sigma). Immunoblots were incubated with ChemiGlow (Fisher Scientific) and image analysis and quantification of immunoreactive bands were performed using the Fluorchem FC digital imaging system (Alpha Innotech). A reference kidney homogenate was loaded on each gel in duplicate so that results could be compared across gels. Dose-response curves were used to determine the linear range of response (14-20 mg of protein was used).
Statistical analysis. Two-way (genotype; 6 CLA) and 3-way (genotype; c9,t11 CLA level; t10,c12 CLA level) ANOVA were used to determine significant main and all interaction effects (SAS Institute) for studies 1 and 2, respectively. Data were analyzed for normality using the Kolmogorov-Smirnov D statistic and for homogeneity of variance using Levene's test. In study 2, because of the larger number of groups and interactions observed in study 1, preplanned contrasts also were used to detect interactions. Preplanned contrasts compared lean vs. fa/fa for each of the following: 6 9,11; 6 10,12; 6 TOG; 9,11 vs. 10,12; 10,12 vs. TOG; 9,11 vs. TOG. If    Dietary conjugated linoleic acid in obesity-associated nephropathy 287 interactions were detected in either study, differences between means were tested by Duncan's multiple range test. All data are reported as means 6 SEM. The significance level was P , 0.05 for main effects and P , 0.10 for interactions (to reduce the risk of missing interactions) as determined by the ANOVA table or the preplanned contrasts.

Results
Feed intake, body, and kidney weight. In study 1, fa/fa compared with lean rats had 43% higher feed intakes, 58% higher body weights, 21% greater kidney weights, and 23% lower kidney weights relative to body weight ( Table 2). Dietary CLA supplementation did not alter feed intake or body weight within each genotype. Whereas CLA did not alter kidney size in lean rats, kidneys were 11% smaller by weight in faCLA rats than in faCTL rats.
In study 2, fa/fa compared with lean rats had 54% higher feed intakes, 67% higher body weights, 22% larger kidney weights, and 27% lower kidney weights relative to body weight ( Table 3). With respect to CLA effects, dietary t10,c12 CLA supplementation in fa/fa rats resulted in 9-16% lower feed intakes than fa/ fa rats on the other diets. Also, faTOG rats ate 8% less feed than faCTL rats. Analysis of weekly feed intake data revealed that all of the differences in feed intake occurred within the first 5 wk of the study and that weekly feed intakes within each genotype did not differ during the last part of the study. This accounts for the similar body weights at the end of the study. As in study 1, CLA supplementation did not alter kidney size in lean rats, but in fa/fa rats, the fa10,12 group had lower kidney weights (by 7%) than faCTL rats. Glomerular volume. In study 1, kidney glomeruli were 46% larger in faCTL vs. lnCTL ( Figs. 1 and 2). Whereas CLA supplementation did not alter glomerular size in lean rats, it did result in 28% smaller glomeruli volumes in the faCLA vs. faCTL rats. However, the glomeruli of faCLA rats remained larger than the lean rats.
In study 2, the glomeruli were 37% higher in faCTL vs. lnCTL rats (Figs. 3 and 4). Again, CLA isomer supplementation did not alter the glomerular size in lean rats. In fa/fa rats, only the fa10,12 rats had significantly smaller glomerular volumes than the faCTL rats, whereas the glomeruli in the other 2 groups did not significantly differ from either the fa10,12 or faCTL groups.
Urinary variables. In study 1, fa/fa rats had 32% higher urine volume, 31% lower urine creatinine, 55% higher serum creatinine, 66% lower creatinine clearance, and 109% higher urine protein excretion than lean rats. However, dietary CLA supplementation did not alter any of these variables (Supplemental Table 1).
In study 2, there also was a genotype effect observed in all urinary variables, where fa/fa rats had 47% higher urine volume, 31% lower urine creatinine, 25% higher serum creatinine, 49% lower creatinine clearance, and 76% higher urine protein excretion than lean rats. Again, dietary CLA supplementation did not alter any of these variables (Supplemental Table 2).
Protein levels of COX enzymes. In study 1, COX-1 levels did not differ (Supplemental Fig. 1A) between fa/fa and lean rat kidneys, but fa/fa rats had 187% higher COX-2 protein levels (Supplemental Fig. 1B) than lean rats. CLA supplementation did not alter COX-1 or COX-2 protein levels. In study 2, COX-1 levels also did not differ (Supplemental Fig. 2) between fa/fa and lean rats, but faCTL rats had 67% higher COX-2 protein levels ( Fig. 5) than lnCTL rats. As with the glomerular size values, when CLA was given to fa/fa rats, COX-2 protein levels were lower but only significantly different from CTL when t10,c12 CLA alone was given (reduced by 39%).

Discussion
The current study demonstrates that CLA supplementation, particularly the t10,c12 isomer, significantly reduces the early glomerular enlargement and the higher levels of COX-2 in the FIGURE 3 Glomerular size in kidneys from lean and fa/fa rats fed CLA isomers alone or together for 8 wk (study 2). Data are expressed as means 6 SEM, n ¼ 9-10. Also significant preplanned contrast for lean vs. fa/fa for 10,12 vs. CTL, P ¼ 0.0624. Labeled bars without a common letter differ, P , 0.05. In both studies, larger glomeruli were observed in fa/fa compared with lean rats, consistent with previous reports in this rat model (10,11,37). Glomerular hypertrophy in obesity and its associated metabolic abnormalities appear to be due to afferent but not efferent arteriole dilation, which leads to glomerular hyperfiltration, glomerular enlargement, thickening of the basement membrane, mesangial expansion, proteinuria, and renal disease (12,14). This also has been observed in obesity-related glomerulopathy, a common occurrence in obese patients that is characterized by glomerulomegaly often accompanied by focal segmental glomerulosclerosis (38).
Interestingly, dietary CLA affected fa/fa but not lean rats, suggesting that CLA may not affect the kidney unless there is a perturbation in glomerular filtration. It may do this by altering the production of COX metabolites. In this regard, fa/fa rats have significantly higher renal COX-2 protein levels than lean rats, as demonstrated in the current studies. Other studies have demonstrated similarly elevated renal COX-2 levels in fa/fa and Zucker Diabetic Fatty rats (15,16,20,37). Prostanoids produced by COX enzymes are important in the regulation of renal hemodynamics, the renin-angiotensin system, and renal pathogenesis (39). Elevated prostanoids have been implicated in renal injury in both animal models of obesity-associated nephropathy (15)(16)(17)20,38) and in patients with type 2 diabetes (41). With COX-2 inhibition, prostanoid levels and renal injury are reduced in the fa/fa rat (15,20) and in patients with Type 2 diabetes (40). Supplementation of this isomer may ameliorate renal disease through the alteration of COX metabolites as in the Han:SPRD-cy rat model of kidney disease, which displays elevated COX-2 activity and in which dietary CLA competes with linoleic acid and inhibits the production of prostaglandin E 2 (7,19). The ability of CLA to alter prostanoid production has been demonstrated in other tissues (21)(22)(23)(24). In animal models of obesity and diabetic nephropathy, increased COX-2 may be due to the effects of high glucose on protein kinase C activation of prostanoid production (21)(22)(23)(24) and/or be due to activation of the renin-angiotensin system (20). Alternatively, it may be associated with the increased glomerular size observed in these kidneys, because COX-2 in the kidney is primarily associated with the macula densa. In the same way, the possibility that COX-2 protein levels in fa/fa rats given t10,c12 CLA are reduced because of changes in glomerular size also cannot be ruled out in the current study.
We have reported other effects of CLA mixture supplementation in the fa/fa rat, including preservation of pancreatic islets, improvement in peripheral glucose utilization, and reduction in inflammatory markers (28). As well, CLA reduces hepatic steatosis, improves liver function, and favorably modifies lipid metabolism (27). It also has recently been shown to decrease adipocyte size and favorably modify adipokine status and insulin resistance (29). In the JCR:LA-cp rat model of metabolic syndrome, insulin resistance and vascular disease, cosupplementation of chromium picolinate and CLA improves lipid profile, hyperinsulinemia, and circulating adipokines. As well, supplementation of CLA alone improves renal function by reducing urinary albumin levels and glomerulosclerosis (9). However, long-term studies on the efficacy, safety, and mechanism of action of the t10,c12 isomer are warranted to determine whether the renal effects are beneficial in the long term, especially because dietary supplementation with the t10,c12 isomer exhibits prodiabetic, pro-oxidative, and proinflammatory properties in humans (41,42).