Objective: The objective of this study was to characterize immune function in the fa/fa Zucker rat, and to determine the effects of feeding conjugated linoleic acid (CLA) isomers on immune function.
Methods and Procedures: Lean and fa/fa Zucker rats were fed for 8 weeks nutritionally complete diets with different CLA isomers (%wt/wt): control (0%), c9t11 (0.4%), t10c12 (0.4%), or MIX (0.4% c9t11 + 0.4% t10c12). Isolated splenocytes were used to determine phospholipid (PL) fatty acid composition and cell phenotypes, or stimulated with mitogen to determine their ability to produce cytokines, immunoglobulins (Ig), and nitric oxide (NO).
Results: Splenocyte PL of fa/fa rats had a higher proportion of total monounsaturated fatty acids and n −3 polyunsaturated fatty acids (PUFA), and lower n −6 PUFA and n −6-to-n −3 PUFA ratio (P < 0.05). Feeding CLA increased the content of CLA isomers into PL, but there were lower proportions of each CLA isomer in fa/fa rats. Splenocytes of fa/fa rats produced more amounts of IgA, IgG, and IgM, NO, and interleukin-1β (IL-1β), IL-6, and tumor necrosis factor-α (TNF-α) (P < 0.05). Obese rats fed the t10c12 diet produced less TNF-α and IL-1β (lippopolysaccharide (LPS), P < 0.05). Splenocytes of fa/fa rats produced less concanavalin A (ConA)-stimulated IL-2 (P < 0.0001) than lean rats, except fa/fa rats fed the c9t11 diet (P < 0.05).
Discussion: The c9t11 and t10c12 CLA isomers were incorporated into the membrane PL of the fa/fa Zucker rat, but to a lesser extent than lean rats. Splenocytes of obese rats responded in a proinflammatory manner and had reduced T-cell function and feeding the t10c12 and c9t11 CLA isomers may improve some of these abnormalities by distinct methods.
Obesity is associated with an increased risk of infection (as reviewed in ref. 1) and immune-related forms of cancer (2,3), poor antibody responses to vaccines (4,5), and increased levels of systemic (6) and tissue inflammatory mediators (7), indicating abnormalities in immune function. Although there is considerable evidence that chronic low-grade inflammation is associated with the obese state, the etiology of this inflammation remains unknown. Most studies have focused on the role of the inflammatory cells and the adipocytes in the etiology of inflammation; however, T cells have an important role in regulating inflammation (8) and their contribution to inflammation in the obese state remains unknown.
The Zucker fa/fa rat is a monogeneic model of obesity that expresses a dysfunctional leptin receptor that severely limits its ability to respond to leptin (9), a condition which has been identified in only a few individuals (10). However, many of the metabolic abnormalities present in the Zucker fa/fa rat, including leptin resistance (as reviewed in ref. 11), are observed in human obesity (12,13). A limited number of studies have been conducted on immune function in this animal model. Abnormalities in the innate immune system have been identified in the Zucker fa/fa rat including an impaired capacity to kill yeast cells, despite normal phagocytic function (14,15). Additionally, there are several reports of T-cell lymphopenia affecting both the CD4+ and the CD8+ T cells (16) and a decreased ability of lymphocytes to respond in vitro to mitogen stimulation (17,18). Little is known about the effect of obesity on mitogen-stimulated cytokine, immunoglobulin (Ig) production, or immune cell types (beyond the relative proportion of CD4+ and CD8+ cells). Although these studies are suggestive of immune dysfunction, currently there is no animal model of obesity with identified chronic inflammation or clearly characterized T-cell dysfunction that would explain the immune abnormalities observed in human obesity. Therefore, the first objective of this study was to characterize immune function in the Zucker fa/fa rat to determine its suitability as a model for the inflammatory immune dysfunction associated with obesity in humans.
It is well established that dietary nutrients, particularly lipids, can influence both the inflammatory response and T-cell function (19). More recently, conjugated linoleic acid (CLA), which describes a group of the geometrical and positional isomers of the dietary essential linoleic acid, has been reported to have anti-inflammatory and immunoregulatory effects in healthy animals (reviewed in ref. 20). Contrary to the animal studies, clinical trials in healthy individuals have reported few effects of CLA on immune function (21,22,23,24,25,26,27), including one report of a minor elevation in serum C-reactive protein levels (28). Overall, these results suggest that feeding CLA to healthy individuals has minimal impact on immune function. However, studies in animals suggest that CLA isomers may have a greater impact when the immune system is challenged. In support of this hypothesis, feeding a 1.3% wt/wt CLA mixture improved mucosal inflammation and increased mRNA levels of interleukin-10 (IL-10) in colonic lymph nodes in a piglet model of bacterial-induced colitis (29). Furthermore, plasma levels of proinflammatory cytokines were lower and IL-10 levels were higher in animals with acute inflammation (immune-mediated catabolism) that were fed a CLA mixture (1–2% wt/wt) (30,31). The majority of studies that have been conducted in animals have fed a mixture of the two major isomers of CLA, the cis 9,trans 11 (c9t11) and trans 10,cis 12 (t10c12) CLA.
These findings suggest that feeding CLA isomers might be beneficial to treat the immune dysfunction associated with obesity. Contrary to what might be predicted from animal studies, two studies conducted in obese men reported minor elevations in blood C-reactive protein (t10c12 CLA only) and urinary 15-ketodihydroprostaglandin F2α with no effect on blood cytokine levels after 13 months of supplementation with either c9t11 or t10c12 CLA isomer (32,33). It is well established for other dietary fats, such as the long-chain n −3 fatty acids, that the level and type of fats in the diet influence their ability to modulate inflammation (19). Unfortunately, in the studies by Riserus et al. (32,33) the fat content and composition of the subject's diet were not determined.
To determine the potential benefits of CLA isomers in the treatment of obesity, systematic animals studies are required where the diet is controlled and the impact of the two main isomers is studied on the major immune abnormalities that occur with the obese state. The second objective of this study was to determine the effects of feeding c9t11 and t10c12 CLA, either singly or combined, on parameters of immune function and inflammation in the fa/fa Zucker rat.
Methods and Procedures
RPMI-1640 culture media, fetal calf serum, antimycotic-antibiotic solution (10,000 μg/ml penicillin G sodium, 10,000 μg/ml streptomycin sulfate, and 25 μg/ml amphotericin B), 2-mercaptoethanol, and HEPES were purchased from Invitrogen (Burlington, Ontario, Canada). Phorbol myristate acetate (PMA) and concanavalin A (ConA) were purchased from ICN (Montreal, Quebec, Canada) and lippopolysaccharide (LPS), ionomycin (I) and pokeweed mitogen (PWM) were obtained from Sigma-Aldrich (Oakville, Ontario, Canada). BD OptEIA anti-rat enzyme-linked immunosorbent assay (ELISA) sets were used to detect IL-4, IL-6, IL-10, interferon-γ (IFN-γ), and tumor necrosis factor-α (TNF-α) (BD Biosciences Pharmingen, Mississauga, Ontario, Canada). IL-2 CytoSet was purchased from Biosource (Medicorp, Montreal, Quebec, Canada). IL-1β ELISA kit was purchased from R&D Systems (Cedarlane Laboratories, Hornby, Ontario, Canada) and Ig (G, M, and A) ELISA quantitation sets were purchased from Bethyl Laboratories (Cederlane Laboratories, Hornby, Ontario, Canada). Fluorescent prelabeled monoclonal antibodies were purchased from BD Biosciences Pharmingen (Mississauga, Ontario, Canada) except OX62 and OX12, which were purchased from Serotec (Cedarlane Laboratories, Hornby, Ontario, Canada). Streptavidin-Quantum Red was purchased from Sigma-Aldrich (Oakville, Ontario, Canada). Sterile 4-ml tubes, 96-well “V”-bottom and flat-bottom plates, and 1.5-ml microcentrifuge tubes were purchased from Fisher Scientific (Ottawa, Ontario, Canada). High-performance thin layer chromatography plates were purchased from Fisher Scientific (Ottawa, Ontario, Canada) and glass methylation vials were purchased from Chromographic Specialties (Brockville, Ontario, Canada). All dietary components except cornstarch (Best Foods, Etobicoke, Ontario, Canada), CLA (Natural ASA, Hovdebygda, Norway), and tert-butylhydroquinone (Aldrich Chemical, Milwaukee, WI) were purchased from Harlan Teklad (Madison, WI).
Animals and diet
Experimental procedures were reviewed by the University of Manitoba, Fort Garry Protocol and Management committee and approved in accordance with the Canadian Council on Animal Care guidelines. Lean and obese male fa/fa Zucker rats (n = 80; Harlan, Indianapolis, IN) were obtained at 5 weeks of age and acclimatized for 5–7 days. Rats were fed a semi-purified diet formula, based on the AIN-93G diet (34), differing only in the amounts of CLA isomers (Table 1). Lean and obese animals (n = 10/treatment) were randomly assigned to one of the following dietary treatments for 8 weeks: (i) 0.4% wt/wt c9, t11 CLA (c9t11), (ii) 0.4% wt/wt t10, c12 CLA (t10c12), and (iii) 0.4% wt/wt c9, t11 plus 0.4% wt/wt t10, c12 CLA (MIX), or (iv) 0% CLA (control diet, Ctl). The total amount of fat provided in the diet was 8.5% wt/wt. Our rational for providing the same concentration of the individual isomers in the mixture diet was to interpret the effects of the individual isomers and then their effect at the same concentration when provided together. The commercially prepared CLA isomers were in free fatty acid form. All dry ingredients were premixed and fresh batches of diet containing oil were prepared weekly and stored at −20 °C until fed.
Table 1. Diet composition
The rats were individually housed in a temperature- (21–23 °C) and humidity- (55%) controlled environment with 14:10 light-to-dark cycles. All rats had free access to water and were fed ad libitum. Feed cups were filled three times a week and feed intake, adjusted for feed spillage, was recorded at that time. Animal body weights were recorded weekly. After consuming the experimental diets for 8 weeks, rats were killed by CO2 asphyxiation and cervical dislocation, and the spleens were removed.
Isolation of splenocytes and primary culture conditions
The spleens were weighed and placed in sterile 0.5% wt/vol bovine serum albumin in Krebs-Ringer HEPES buffer (pH 7.4) and isolated as we have described previously (35). Isolated splenocytes were resuspended in complete culture media (RPMI-1640 supplemented with 5% (vol/vol) heat-inactivated fetal calf serum, 1% (vol/vol) antimycotic-antibiotic solution, HEPES (25 mmol/l), and 2-mercaptoethanol (2.5 mmol/l)) and counted on a hemacytometer (Fisher Scientific, Edmonton, Alberta, Canada). The fetal calf serum contains fatty acids and therefore the cell culture media contained 0.2 μmol/l c9, t11 CLA isomer. Splenocytes were resuspended in the culture media described earlier (1.0 × 106 cells/ml) and incubated in 4-ml sterile polystyrene tubes in a humidified atmosphere at 37 °C in the presence of 5% vol/vol CO2. The final cell culture medium either contained no mitogen (unstimulated cells) or was supplemented with mitogens at concentrations determined previously in our laboratory and/or published in the literature [ConA (2.5 mg/l), LPS (0.1 g/l) (36), PWM (55 mg/l), or PMA + I (20 μg/l + 0.5 nmol/l)] (37) in healthy normal body weight rats. After 48 h of culture, the supernatant was removed and stored at −80 °C until cytokine, Ig, and nitric oxide (NO) assays were performed. The cell pellets were washed with phosphate-buffered saline and frozen at −80 °C for fatty acid analysis.
In the control-fed rats, immune cell subsets in splenocytes were identified by one or two color direct immunofluorescence assay as we have described previously (38). The prelabeled monoclonal antibodies used were CD3, RT1B (Class II monomorphic), CD28 (fluorescein isothiocyanate labeled); CD4, CD8, CD86, CD3, CD11b/c, and OX12 (phycoerythrin labeled); and CD25, OX62, CD86, and CD80 (biotin labeled). Streptavidin-Quantum Red (R-PE-Cy5 flurochrome) was added to wells containing biotin-labeled Ab. After final wash, plates were aspirated and 200 μl of cell fix (1% wt/vol paraformaldehyde) was added to each well. The proportion of cells that were positive for each marker was determined by flow cytometry (FACScan; Becton Dickinson, Sunnyvale, CA) according to the relative fluorescence intensity using CellQuest software (Becton Dickinson, Sunnyvale, CA).
Cytokine, Ig, and NO production
The following assays were performed according to the manufacturer's instructions. The cultured cell supernatants of PMAI, ConA, LPS, and PWM-stimulated splenocytes were used to determine IL-1β and TNF-α (31.2–2,000 pg/ml), IL-2 (23.4–1,500 pg/ml), IL-4 (1.6–100 pg/ml), IL-6 (78–5,000 pg/ml), IL-10 (15.6–1,000 pg/ml), and IFN-γ (31.25–2,000 pg/ml) levels with commercial ELISA kits (detection limits are indicated in brackets). Ig levels were quantified in LPS-stimulated and unstimulated (UNS) supernatant using ELISA quantitation kits. The range of detection for IgG, IgA, and IgM were 7.8–500, 15–1,000, and 31.2–2,000 ng/ml, respectively. NO production was determined by analyzing nitrite (NO2−, a product of the l-arginine-dependent NO pathway) concentration in splenocyte culture supernatants using a colorimetric assay based on the Griess reaction (39). All samples were measured in duplicate and the absorbance was measured at 540 nm for NO2− or 450 nm for cytokines and Ig on a microtiter plate reader (SpectraMax 190; Molecular Devices, Sunnyvale, CA). The average of the duplicate data was used for statistical analysis if the coefficient of variance was ≤10%.
Splenocyte phospholipid fatty acid composition
A modified Folch method was used to extract lipids from freshly isolated splenocytes before mitogen stimulation as described previously (40). Total phospholipids (PLs) were separated on silica G plates as described previously (41) and visualized with 8-anilino-1-naphthalenesulfonic acid under ultraviolet light against the appropriate standards. PL fatty acid methyl esters were prepared from the scraped silica band. Methyl esters were prepared by the base-catalyzed method using sodium methoxide as described elsewhere (42). Prepared PL fatty acid methyl esters were flushed with nitrogen and stored at −35 °C until analysis by gas chromatography. Fatty acids were separated by automated gas-liquid chromatography (Varian 3800; Varian Instruments, Mississauga, Ontario, Canada) using a 100-m CP-Sil 88 fused capillary column (Varian, Mississauga, Ontario, Canada) as described elsewhere (42).
Statistical analysis was conducted using the SAS software statistical package (Version 9.1; SAS Institute, Cary, NC). All data were reported as mean ± s.e.m. The effects of diet and phenotype were determined by two-way ANOVA and significant differences between groups were identified by least-square means at P < 0.05. Blocking was imposed to account for any unexplained error associated with days the animals were killed for cytokine, Ig, and NO data. Phenotype analysis was conducted only on rats fed the control diet and a two-tailed t -test was used to compare differences between the phenotypes. Statistical significance was reported at P ≤ 0.05.
Body weight, feed intake, and spleen measurements
Obese Zucker (fa/fa) rats consumed more feed, had significantly higher body and spleen weights and a lower spleen weight per gram (g) body weight, a lower number of total splenocytes, and a lower number of splenocytes per gram spleen weight than lean rats (Table 2). There was no effect of diet on any of the parameters listed in Table 2 in the lean animals. Obese rats fed the MIX or t10c12 CLA diets consumed significantly less feed than obese rats fed the Ctl diet, and obese rats fed the t10c12 CLA diet consumed significantly less than the c9t11 and MIX CLA diets (Table 2). Obese rats fed the t10c12 and MIX diet had lower spleen weights than obese rats fed the Ctl or c9t11 diet (Table 2). Obese rats fed t10c12 and MIX diets had similar absolute spleen weights and number of splenocytes (t10c12 only) compared with lean rats fed the same diet.
Table 2. Effect of phenotype and CLA isomers on feed intake, body and spleen weight, and splenocyte numbers in lean and fa/fa Zucker rat
Proinflammatory cytokines—TNF-α, IL-1β, IL-6, and IFN-γ
Splenocytes obtained from obese Zucker rats produced more TNF-α, IL-1β, and IL-6 than lean rats after mitogen stimulation (Table 3). There was no main effect of diet on the production of these cytokines by splenocytes from either lean or obese rats, irrespective of the type of mitogen used. However, LPS-stimulated splenocytes of obese rats fed the t10c12 CLA diet produced less TNF-α and IL-1β than obese rats that were fed the Ctl diet (P < 0.05, Table 3). Feeding CLA did not affect the production of TNF-α, IL-1β, or IL-6 in lean rats. IFN-γ production in ConA-stimulated splenocytes was higher in obese rats (P < 0.002), but lower with PMAI (P < 0.0001) and LPS (P < 0.0001) stimulation. Lean animals fed the c9t11 CLA isomer produced less IFN-γ than lean animals fed the Ctl diet (P < 0.05, PMAI-stimulated splenocytes). Diet did not affect IFN-γ production in ConA-, LPS- or PWM-stimulated immune cells of either lean or obese rats.
Table 3. Effect of phenotype and CLA isomers on mitogen-stimulated cytokine production of splenocytes from lean and fa/fa Zucker rats
Splenocytes from obese Zucker rats produced less ConA-stimulated IL-2 (P < 0.0001) than lean rats (Table 3), but there was no main effect of phenotype with PMAI or PWM stimulation (Table 3). Splenocytes of obese rats fed the c9t11 produced similar levels of IL-2 compared with lean rats fed the same diet (ConA, P < 0.05), whereas splenocytes of obese rats fed the other diets had lower production compared with lean rats fed the same diet (ConA, P < 0.05) (Table 3). Obese rats fed the MIX diet had decreased PMAI-stimulated production of IL-2 compared with obese rats fed the Ctl diet (P < 0.05, Table 3). Feeding CLA did not affect PWM-stimulated IL-2 production in obese rats. Feeding either the MIX or the c9t11 diet to lean rats decreased IL-2 production in PWM-stimulated splenocytes (P < 0.05). Lean rats fed any of three CLA diets produced less IL-2 compared with lean rats fed the Ctl diet (ConA, P < 0.05) (Table 3).
IL-4 and IL-10
Immune cells from obese rats fed the c9t11 diet produced more IL-10 after ConA stimulation than lean rats fed the same diet and obese rats fed the MIX diet (Table 3). Obese rats fed the c9t11 CLA produced more IL-10 after stimulation with PMAI than cells from those fed the t10c12 and MIX CLA diets (P < 0.05, Table 3). Obese rats fed the c9t11 diet produced more IL-10 after PWM stimulation compared with obese rats in the Ctl group (P < 0.05, Table 3). Splenocytes of lean rats fed the c9t11 diet and stimulated with PWM produced more IL-10 than lean rats fed the MIX diet. Diet or phenotype did not affect production of IL-10 in immune cells stimulated with LPS in lean and obese rats. Cells from obese rats fed any of the CLA-containing diets produced less IL-4 than cells from Ctl-fed rats after stimulation with ConA or PMAI (Table 3). IL-4 production was not significantly different among the lean groups.
Ig and NO production
Splenocytes from obese animals produced more IgA, IgG, and NO in the unstimulated state and more IgM, IgA, IgG, and NO after stimulation with LPS (Table 4). Feeding any of the CLA-containing diets resulted in a lower (P < 0.05) production of IgA both with and without LPS but only the amounts produced by the cells from the c9t11- and t10c12-fed animals reached levels not significantly different from lean animals fed the same diets (Table 4). Feeding the t10c12 or MIX diet to obese animals resulted in a lower production of NO in the unstimulated condition to levels not significantly different from the lean animals fed the same diets (Table 4). For lean rats, diet did not alter NO levels in LPS-stimulated or unstimulated cells.
Table 4. Effect of phenotype and CLA isomers on IgM, IgA, IgG, and NO production in unstimulated or LPS-stimulated splenocytes from lean and fa/fa Zucker rats
Immune cell phenotypes in spleen
Immune cell phenotypes were measured only for rats fed the control diet. Obese animals had a lower proportion of CD3+ (42 ± 1.2 vs. 49 ± 1.6, P < 0.004), CD3+CD4+ (23 ± 1.2 vs. 27 ± 1.8, P < 0.04), and CD8+CD25+ (3 ± 0.5 vs. 4 ± 0.5, P < 0.02) cells but a higher proportion of CD11b/c+ (21 ± 1.5 vs. 18 ± 1.4, P < 0.05) and OX6+CD86+ (6 ± 0.8 vs. 4 ± 0.8, P < 0.04) cells in the spleen. There was no difference in the proportion of CD3+CD8+, CD4+CD25+, CD4+CD28+, CD8+CD28+, OX12+, OX6+CD11b/c+, OX6+OX62+, and OX6+CD80+ cells between lean and obese rats (P < 0.05).
PL fatty acid composition of splenocytes
The relative proportions of fatty acids from 14:0 to 24:1 (n −9) in total PLs were measured but only major fatty acids are reported (Table 5). Obese animals had a significantly higher proportion of C14:0, C16:0, C18:1(n −9), C18:1(n −7), C20:3(n −6), C20:5(n −3), C22:5(n −6), and C22:6(n −3) and lower proportion of C18:2(n −6), C18:3(n −3), and C20:2(n −6) compared with lean rats. The PLs from obese rats had a higher proportion of total monounsaturated fatty acids and n −3 polyunsaturated fatty acids (PUFA) and a lower proportion of total n −6 PUFA and n −-to-n −3 PUFA ratio. Compared with the Ctl diet-fed rats, feeding any of the CLA diets resulted in incorporation of the respective CLA isomer(s) into splenocyte PLs (Table 5). However, there were significantly lower proportions of the individual CLA isomers in the PLs of obese rats compared with lean rats (Table 5). Obese rats fed the c9t11 CLA diet incorporated more C18:1(n −9), C18:1(n −7), and total monounsaturated fatty acids and less C20:2(n −6) compared with obese rats fed the t10c12 or MIX diet (P < 0.05). Obese rats fed the c9t11 diet also had a higher proportion of C16:0 and a lower proportion of C20:3(n −6), C22:5(n −6), and n −3 PUFA compared with obese rats fed the MIX diet (P < 0.05).
Table 5. Effect of phenotype and CLA isomers on fatty acid composition of splenocyte phospholipids in lean and fa/fa Zucker rats
Immune dysfunction in the obese fa/fa Zucker rat
The results of this study demonstrate that the fa/fa Zucker rat, compared with its lean control, has altered immune function (refer to Figure 1). Other groups have also reported lower IL-2 production after mitogen stimulation in diet-induced obese rodents (36,43) and a lower T-cell proliferative response (estimated by the rate of 3H-thymidine incorporation) in both the fa/fa Zucker rat and diet-induced obese rats (16,18,36,44). Consistent with findings in diet-induced obesity (16,36,44), fa/fa Zucker rats had a lower proportion and concentration of total T cells, affecting only the T helper (CD3+CD4+) subset, in the spleen. Unlike Tanaka et al. (16), we did not notice a lower proportion of cytotoxic T cells (CD3+CD8+) in the fa/fa Zucker rats. The lower number of total T cells would have contributed to the lower IL-2 production after stimulation with a polyclonal T-cell mitogen, such as ConA (45,46). Although the proportion of CD8+ cells did not differ between lean and obese, obese rats also had a lower percentage of CD8+ splenocytes that expressed the IL-2 receptor, suggesting that cytotoxic T-lymphocytes of obese rats may have a reduced capacity to proliferate.
Despite a lower production of IL-2, splenocytes isolated from fa/fa Zucker rats produced higher levels of inflammatory cytokines and NO (Figure 1). These results are novel and suggest a vigorous proinflammatory response by T cells and macrophages. It is possible that the slightly higher proportion of macrophages (CD11b/c+) and activated antigen-presenting cells (RTB1+ or positive for major histocompatibility complex II) that express a co-stimulatory molecule (CD86+)) contributed to the higher production of these inflammatory mediators by cells from obese rodents. The small increase in the proportion of these innate immune cells is unlikely the sole contributor to the 2.5- to 2.8-fold increase in TNF-α production after LPS stimulation. In agreement with previous reports, mitogen-stimulated production of the regulatory/anti-inflammatory cytokine, IL-10, did not differ between lean and obese rats (36). This proinflammatory response may be unique to the Zucker rat as there was no difference in TNF-α production by splenocytes after mitogen stimulation in diet-induced obese rodents (C57BL/6J mice or Wistar rats) (36,43). Furthermore Lamas et al. (47) reported that mRNA levels of TNF-α and IL-6 in spleen were actually lower in diet-induced obese rats. Alternatively, it is possible that the highly saturated diets consumed by the rodents in these studies might have dampened the inflammatory reaction (48,49). It is interesting that the fa/fa Zucker rat in this study favored a proinflammatory response to a T-cell mitogen, whereas another study utilizing ob/ob mice reported lower inflammatory responses to allogeneic peripheral blood mononuclear cells or splenocytes (50). There is a growing body of evidence indicating that proinflammatory mediators can predict the onset of type 2 diabetes (51) and development of cardiovascular disease (52,53) in humans. Thus, the heightened inflammatory responses we observed in obese animals may contribute to the disease pathology of obesity-associated comorbidities.
To our knowledge, higher production of Igs in both the absence (increased IgA or IgG) and presence of LPS (increased IgG, IgA, and IgM) has not been previously reported in obese animals. The preliminary analysis used to determine the optimal incubation time for cytokine and Ig production was conducted in lean rats and we acknowledge that the 48 h time point selected may not have been optimal as it was determined based on the maximum response in lean rats. Despite this, the heightened proinflammatory response observed in the obese animals, in the absence of a difference in the proportion of B lymphocytes (OX12+ cells), likely contributed to Ig production as elevated circulating levels of Igs have been reported in inflammatory conditions such as rheumatoid arthritis (54,55).
It is well established that the type and amount of dietary fatty acids consumed influence the fatty acid composition of PLs in immune cells and this can modify membrane protein expression and function, membrane-mediated signaling, and gene transcription (as reviewed in refs. 56,57). Similar to previous studies that examined nonlymphatic tissues in the fa/fa Zucker rat, we observed abnormalities in the essential fatty acid concentration of splenocyte PLs (58,59,60,61). Consistent with the reported fatty acid composition of liver PLs in the fa/fa Zucker rat (58,59), we observed a lower n −6-to-n −3 PUFA ratio in immune cell PLs. This was the result of both a higher proportion of total n −3 fatty acids, including eicosapentaenoic acid (20:5n −3), and a lower proportion of linoleic acid (18:2n −6) in the PLs of obese rats. It has been demonstrated in both human and animal feeding studies that lowering the n −6-to-n −3 PUFA ratio lowers the proliferative response of T lymphocytes (62). Although this may have contributed to the lower IL-2 response to ConA stimulation, it is inconsistent with the higher production of proinflammatory cytokines. Lowering the n −6-to-n −3 PUFA ratio is reported to reduce the inflammatory response of immune cells in both healthy and inflammatory states (as reviewed in ref. 63).
The underlying mechanisms responsible for the immune abnormalities reported in the fa/fa Zucker rat are unknown, though a few hypotheses exist. Although this study was not designed to explore the underlying biological mechanisms, we propose that abnormalities in T-cell function may be related to the severe leptin resistance in this animal model. The Zucker fa/fa rat expresses a dysfunctional long form of the leptin receptor, which is present on B and T lymphocytes and monocytes/macrophages (64). When leptin was administered to ob/ob mice (leptin-deficient mouse model) it improved T-lymphocyte responses to mitogens (increased IL-2 production and 3H-thymidine incorporation) (50). The altered T-cell function we observed in the fa/fa Zucker rat might also be due to impairments in the protein kinase C pathway due to leptin resistance. In this study, splenocytes of obese rats stimulated with ConA, which directly binds to the T-cell receptor (65), produced more IFN-γ than lean rats. However, when splenocytes were stimulated with PMAI, which bypasses the plasma membrane receptors and activates protein kinase C (66), less IFN-γ was produced by obese rats compared to lean rats. Leptin has been reported to stimulate the protein kinase C pathway in peripheral blood mononuclear cells (67) and impairments in PMA-stimulated protein kinase C activity have been reported in fa/fa Zucker hepatocytes (68). Although leptin is reported to stimulate the production of inflammatory cytokines from macrophages (69), a leptin deficiency was protective against inflammatory experimental arthritis (70). Although we cannot completely rule out an effect of leptin resistance, our results suggest that additional mechanisms contribute to the heightened inflammatory response we observed in splenocytes of fa/fa Zucker rats. Clearly, further investigation is warranted to determine the underlying biological mechanisms involved in the inflammatory immune dysfunction present in the obese state.
The effect of feeding diets containing CLA isomers on immune function
Studies investigating the impact of CLA isomers on immune function in human obesity are extremely limited and are mostly restricted to nonspecific markers of inflammation (32,33,71). This is the first study to examine the effects of CLA on immune function in a rodent model of obesity and although we observed few changes, these findings are important as CLA is marketed to the obese population for its weight reducing effects. Our results demonstrate that the individual CLA isomers modify some of the immune abnormalities in the obese fa/fa Zucker rat (Figure 1). Under the experimental conditions of this study, we observed that feeding the c9t11 CLA isomer to obese rats may have a beneficial influence on the proliferative and immunoregulatory response of T cells, whereas feeding the t10c12 CLA isomer reduced the inflammatory response after LPS stimulation, while feeding both of these isomers together appeared to negate the immunological effects of the single isomers. Immune changes in the obese rodents cannot be easily explained by the diet effects on feed intake (c9t11 only), body weight gain, or the distribution of T or B cells in spleen. It is unknown what effect the slight but significant decrease in feed intake (without a change in body weight) would have on immune parameters in obese rats fed the t10c12 or MIX CLA diet.
Dietary CLA has been reported to be incorporated into the PL fraction of peripheral blood mononuclear cells in healthy humans (72). As expected, the CLA isomers were incorporated into splenocytes membranes in CLA-fed rodents but to a lower relative extent than in the obese animals (incorporation was 67–70% of lean rats). This is in agreement with a previous report from our group, which determined that CLA isomers are incorporated less into liver PLs of fa/fa rats compared to lean Zucker rats (73). Interestingly, despite greater incorporation into the PL membrane of lean rats, CLA had little effect on the parameters of immune function measured in this study and does not explain the differences in immune responses between diets as splenocytes from rats fed MIX diet had similar levels of the two isomers as splenocytes from the groups fed the single isomer diets.
In conclusion, our results demonstrate that the fa/fa Zucker rat has T-cell lymphopenia (mainly affecting the T helper subset) in the spleen and that this affects both T- and B-cell functions. In addition to the lower incorporation of c9t11 or t10c12 CLA into the splenocyte PLs of obese rats, there was also a higher proportion of total monounsaturated fatty acids, n −3 PUFA, a lower n −6 PUFA, and n −6-to-n −3 PUFA ratio, and a lower proportion of linoleic acid. The proinflammatory response after stimulation is consistent with the inflammatory state of human obesity. A reduced ability to produce IL-2 after stimulation suggests a potential defect in T-cell function and is consistent with some of the immune abnormalities reported in obese humans. Feeding either the c9t11 (higher production of IL-10) or t10c12 (lower production of TNF-α and IL-1β) isomers singly but not together modulated the inflammatory response and proliferative response of splenocytes when stimulated. Further research is needed in obese humans to determine the physiological importance of these changes.
We thank Susan Goruk and Vanessa DeClercq for their excellent technical assistance and Dr. Laki Goonewardene for his help with statistical analyses. This study was supported by Dairy Farmers of Canada, the Natural Sciences and Engineering Research Council of Canada, and the CLA Network Canada. M.R.R. held the Muttart/Collip Diabetes Studentship from the Muttart Diabetes and Research Training Centre, University of Alberta, Edmonton, Alberta, Canada.