SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Objectives: Obese cats show many similarities to obese people, including insulin resistance and an increased diabetes risk. However, atherosclerosis and cardiovascular disease are not seen in cats. In people, they are associated with the development of an inflammatory response, which, we hypothesized, does not occur in cats.

Design and Methods: Twenty neutered cats of equal gender distribution were allowed to gain weight by offering food ad libitum and were examined before and at 10, 30, 60, and 100% weight gain. All cats reached 60% of weight gain, 12 cats gained 100 % in 12 months.

Results: Fat was equally distributed between subcutaneous and visceral depots. Insulin-independent glucose uptake increased and insulin sensitivity decreased with increasing adiposity. However, baseline glucose concentrations were unchanged suggesting a decrease in EGP. Inflammatory cytokines (Il-1, IL-6, TNFa) and catalase, superoxide dismutase, glutathione peroxidase did not change. Insulin, proinsulin, and leptin were positively and adiponectin negatively correlated with adiposity. Heat production increased with obesity, but became less when body weight gain was > 60 %.

Conclusions: This indicates that metabolism adapts more appropriately to the higher intake of calories in the initial phase of obesity but slows at higher body fat content. This likely contributes to the difficulty to lose weight.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Obesity is a risk factor for diabetes in humans and cats. In humans, it is also an independent risk factor for cardiovascular diseases (CVD) [1]. In rodents and humans, lipid accumulation in the adipocyte increases reactive oxygen species (ROS) and cellular stress pathways are activated leading to the secretion of pro-inflammatory cytokines such as tumor necrosis factor αTNF, αIL-6, and IL-1, among others [2]. Immune cells including monocytes and macrophages are recruited to adipose tissue and produce many of the same cytokines, further increasing oxidative stress and a vicious cycle ensues, in which both adipocytes and recruited macrophages participate [2]. Circulating cytokine concentrations in obese people are positively correlated with impaired glucose tolerance and insulin resistance [1, 3] because insulin signaling is inhibited. Other adipokines have been shown to be intricately involved in changes in insulin action. Two of them, leptin and adiponectin, may play a role in the relationship between obesity and inflammation [3]. Leptin is positively and adiponectin is negatively correlated with insulin resistance in cats similar to other species [4-6]. Leptin acts in the hypothalamus to signal satiety and to increase basal metabolic rate. However, it is also pro-inflammatory [7, 8] and is a member of the IL-6 family of cytokines. Adiponectin is the most abundant gene product of adipose tissue [9, 10]. In both humans and animals, adiponectin mRNA expression and serum levels decrease in obesity and rise with weight loss [9]. Adiponectin has potent anti-inflammatory and anti-atherogenic effects. In macrophages, it inhibits phagocytosis and scavenger receptor expression, and decreases the ability of Toll-like receptors to elicit TNFα and IL-6 production [11, 12], apparently as a result of reduced NF-κB signaling [12]. In leukocytes, adiponectin induces the anti-inflammatory cytokine IL-10 and the endogenous IL-1 receptor antagonist (IL-1RA). The increase in pro-inflammatory leptin and the decrease in anti-inflammatory adiponectin maintain the obesity-inflammation-insulin resistance circle in rodents and humans. While insulin resistance and oxidative stress are associated with accelerated atherosclerosis in humans [13], atherosclerosis is not a feature of feline obesity or diabetes despite alterations in lipid particle size and number usually seen in obese people with cardiovascular problems [14]. The reason for this is unknown.

We hypothesized that the reason for the lack of atherosclerosis and cardiovascular problems in obese cats is their lack to respond to the increase in fat mass with an inflammatory response. Antioxidant diet, therefore, will have no effect on insulin sensitivity. The goal of this longitudinal study was to characterize markers of oxidative stress in cats during the development of obesity.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Animals and diets

Twenty lean and neutered adult domestic shorthair cats (10 males and 10 females) were used in this study. Cats were maintained at the University of Illinois College of Veterinary Medicine Animal Care Facility under standard colony conditions. They were housed individually and were given free access to water. Animal studies were approved by the University of Illinois Animal Care and Use Committee and conducted in accordance with guidelines established by the Animal Welfare Act and the National Institutes of Health Guide for the Care and Use of Laboratory Animals, which included a cycle of 12 hours of light and 12 hours of darkness. Room temperature was maintained at 21°C. Cats were determined healthy based on the results of physical examination and clinical laboratory data. All cats were socialized daily and were accustomed to daily handling.

Prior to the beginning of this study all cats had been fed a commercial dry diet (Purina Pro Plan, Nestle Purina, St. Louis, MO) for 8 months. The cats were equally and randomly assigned to one of two food groups, control diet CON (10 cats; 5 males and 5 females) or diet containing antioxidants AOX (10 cats; 5 males and 5 females). There was no difference in the age (years) of the cats assigned to the 2 groups (CON: 2.4 ± 0.3; AOX: 2.1 ± 0.2). The weight at the beginning of the study was also not different between the groups (3.4 ± 0. 1 kg in CON, and 3.4 ± 0.2 kg in AOX).

The composition of the diets (Nestle Purina Company, St. Louis, MO) is listed in Table 1. The diets differed in the amount of vitamin E. The AOX diet contained beta-carotene, pyridoxine and astaxanthin, which were not included in the control diet. The cats were fed once daily. Food intake was measured daily, and the cats were weighed once weekly. The total time of the study was 12 months.

Table 1. Macronutrient composition of diets
Diet CompositionControlAntioxidant
Protein, %45.248.10
Carbohydrate, % (by Subtraction)21.1318.42
Fat, %18.318.5
Ash, %7.467.61
Moisture, %7.206.56
Crude Fiber, %0.710.81
Kcal/G, ME (by Calculation)3.53.5

Study design

Time 0 refers to the time of experiments when all cats were lean; time 10, 30, 60, and 100 refers to the time of experiments when cats had increased their body weight by approximately 10, 30, 60%, and 100% respectively. The body weights are shown in Table 3. The increase in weight was achieved by allowing increased food intake. Indirect calorimetry measurements were performed on all cats 1 week prior to other testing procedures. Blood sampling for baseline measurements of oxidative stress enzymes and hormones were followed by a low-dose intravenous glucose tolerance test (IVGTT). Food was withheld for 20-24 hours before testing. Two baseline samples were collected for measurement of insulin, and glucose. A glucose bolus (50% dextrose, w/v) of 0.5 g/kg body weight was then injected and blood samples were taken at 5, 10, 15, 30, 45, 60, 90, 120, 180 minutes for the measurements of glucose and insulin.

The cats had 1 day of rest and on day 3, heparin was injected intravenously at a dose of 100 units/kg in order to measure heparin-releasable superoxide dismutase. Blood samples were taken before and at 2, 10, and 15 minutes post injection.

Nuclear magnetic resonance imaging to quantify body fat mass was performed 1 week after completion of the IVGTT.

Blood sampling and processing

To allow for blood sampling, intravenous catheters were placed into the jugular vein 15-18 hours before each testing period after tranquilization of the cats with 1-2 mg/kg Telazol®(tiletamine/zolazepam; Fort Dodge Animal Health, Fort Dodge, IA) subcutaneously. Catheter patency was maintained by injection of 0.5 mL of 0.38% sterile citrate flush (citric acid, trisodium salt dehydrate, Sigma Co., MO) every 8-14 hours. All blood samples were collected through the jugular catheter into tubes containing EDTA. The blood was placed on ice immediately and centrifuged at 405 × g for 15 minutes. The plasma was stored at −80°C until assayed. Red blood cell lysate was obtained by washing red blood cells four times with 0.9% NaCl and then adding cold double distilled water as described by the manufacturer (Cayman Chemical Ann Arbor, MI).

Indirect calorimetry

Indirect calorimetry was performed as described previously [5]. The cats were fasted overnight and indirect calorimetry was performed for 2 hours to determine baseline RER and heat production in conscious cats. They were then fed their daily food ration and measurements were taken for an additional 21 hours.

The following calculations were used:

  • RER = liters CO2 produced/ liters O2 consumed
  • Heat production (kcal) = 3.82 × litersO2 consumed + 1.15 × litersCO2 produced.
  • Heat/metablic body weight (MBW) = heat production (kcal/kg)/(body weight)0.75

Liters O2 consumed was determined from the accumulated O2 at a given time minus O2 consumed at time 0. Liters CO2 consumed was determined from the accumulated CO2 at a given time minus CO2 consumed at time 0.

Enzyme assays

The following assays were performed on samples taken at baseline during the IVGTT: The activity of catalase and glutathione peroxidase in red blood cell lysate was measured using kits from Cayman Chemical (Ann Arbor, MI). The glutathione peroxidase samples were diluted 1:200 with sample buffer from the kit prior to assay. The catalase samples were diluted 1:2,000 with sample buffer from the kit prior to assay. All assays were performed following the manufacturer's directions. The activity of superoxide dismutase (SOD) in lysed RBC's was measured using a kit from RANDOX (Oceanside, CA). The samples were diluted 1:200 using RANSOD diluent prior to assay. Heparin-released superoxide dismutase was measured using a kit from Cayman Chemical (Ann Arbor, MI). All enzyme assays were assayed within 4 weeks of sample collection. Assays were validated for the use in cats by measuring recovery and parallelism. All samples were tested in duplicate. The standard curve for serial dilutions of plasma from cats was observed to be parallel to the standard curve provided by the manufacturer for all assays. Addition of four concentrations of the standards provided in the catalase kit to feline plasma resulted in mean ± SE recovery of 97 ± 1.1%. The assay had a working range of 2-34 nmol/min/mL. Addition of five concentrations of the standards provided in the SOD (RBC) kit to feline plasma resulted in mean ± SE recovery of 91 ± 0.5%. The assay had a working range of 0.168-4.68 units/mL. Addition of five concentrations of the standards provided in the SOD (plasma) kit to feline plasma resulted in the mean ± SE recovery of 94 ± 1.4%. The assay had a working range of 0.025-0.25 units/mL.

Inflammatory cytokines and adipokines

IL-6, IL-1, and TNF alpha were measured in baseline samples from the IVGTT using feline ELISA kits (R and D systems, Minneapolis, MN). Leptin and adiponectin were measured as described [5]. These assays had previously been validated in our laboratory [5].

Other assays

Glucose measurements were performed using a colorimetric glucose oxidase method (Genzyme Diagnostics, Charlottetown, PE Canada). Insulin [15] and feline proinsulin [16] measurements were performed as described previously. Non-esterified fatty acids were measured using a kit from Wako Diagnostics (Richmond, VA). Cholesterol and triglyceride concentrations were determined by enzymatic methods using the Beckman AU480 chemistry analyzer with reagents from Beckman Diagnostics (Fullerton, CA).

Magnetic resonance imaging

Fat mass was determined with nuclear magnetic resonance as previously described [5]. MRI measurements were performed in all cats after injection of tiletamine HCL and zolazepam HCL (2-4mg/kg subcutaneously, Fort Dodge Animal Health, Fort Dodge, IA). All cats were investigated on a Siemens Magnetom Trio 3T horizontal spectroscopy and imaging system with a 16 × 16-cm field of view. Magnetic resonance imaging was performed at 0, 30, 60, and 100% body weight increase. The slice thickness was 3mm with 0.3125 × 0.3125 points. Repetition time (Tr) was 0.4 seconds, and echo time (TE) was 9.4 milliseconds. Total fat mass was analyzed with the Amira 5.3 program from Visage Imaging (San Diego, CA). Slices where both kidneys were visible were used for analysis.

Statistical analysis

The statistical model-based data analysis was carried out using a non-linear mixed effects (NLME) population approach, which consists of fitting all experimental data collected in all cats simultaneously while considering the possible effect of covariates, such as time of study, gender, age, diet, BW, and measured factors associable to insulin resistance. The significance level required for including a covariate in the population model was, in general, p = 0.05. A repeated measures analysis of variance was used to compare parameters between groups and time-points to account for repeated measurements over time of the same individuals and an unstructured covariance structure was assumed (p < 0.05). The area under the curve for glucose and insulin (AUCG and AUCI, respectively), which were defined mathematically as the integral under the continuous concentration curve, were estimated as the sum of all the trapezoids and triangles bounded by the time-versus-concentration curve.

Minimal model analysis of intravenous glucose tolerance test was applied to characterize blood glucose kinetics in terms of insulin dependent (SI) and insulin independent (SG) elimination processes as described previously [17]. In particular, SG represents the component of fractional glucose clearance that is independent of deviations of insulin concentrations from basal, whereas SI represents the steady state increase in fractional glucose clearance per unit increase from basal of insulin concentration. The time lag between variations of insulin concentrations and variations of glucose disappearance is described by first order kinetics with rate constant P2. Both SI and SG quantify implicitly also the effects of insulin and glucose on hepatic glucose production.

The dynamics of glucose concentration and insulin action, following an intravenous glucose bolus of dose D, are described by the two differential equations

  • display math(1)
  • display math(2)

where G(t) is the model prediction of glucose concentration [mg/dL]; X(t) is insulin action [min-1]; I(t) represents the insulin concentration profile [pmol/L], which is normally obtained by linear interpolation of the insulin concentration measurements; VG is the glucose distribution volume, which is expressed in this study in [dL/kg] because the administered glucose dose D was fixed at 500 mg/kg. The initial glucose concentration after the glucose bolus, G(0), is the sum of basal glucose concentration Gb and the concentration increment caused by instantaneous dilution of the injected dose D in the distribution space VG. Because this volume consists mainly of vascular and interstitial space and mixing is not instantaneous, glucose measurements up to 5 minutes were ignored during model fitting. Basal glucose, Gb, and basal insulin, Ib, were determined from blood samples collected before the glucose bolus.

Estimates of the minimal model parameters SG, SI , P2 and glucose distribution volume (VG) were obtained within a population pharmacokinetic framework using the NLME parameter estimation approach [18]. Estimated model parameters were log-transformed to ensure positivity of the back-transformed parameters and to reduce the dependence on parameter scaling.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Body weight, BMI, girth, energy intake, and total fat

The body weight (BW), BMI, girth, energy intake, and total fat of the cats are shown in Table 2. There were no gender or diet differences and all parameters increased significantly throughout the study (all p < 0.001) and were highly correlated among each other (all p < 0.001). The highest correlation was seen between BMI and girth (r2 = 0.887) followed by BMI and BW (r2 = 0.870) and BW and girth (r2 = 0.866). Girth was also highly correlated with % body fat (r2 = 0.7975). Food intake was highly correlated with BMI (r2 = 0.629), girth (r2 = 0.715), BW (r2 = 0.708), and % body fat (r2 = 0.486). The distribution of fat (i.e., visceral and subcutaneous abdominal) did not change during the weight increase. Visceral fat constituted 64 ± 3% of total fat at 0% and 64 ± 3% at 100% weight increase. Twelve of the 20 cats (equal diet and gender distribution) achieved the 100% body weight increase within a 12 month time period.

image

Figure 1. Mean (±SE) values for area under the curve for glucose and insulin in cats (n = 20, 10 males and 10 females) at different % body weight increases. Diet had no effect and the results were combined.

Download figure to PowerPoint

Table 2. Body weight, body mass index (BMI), girth, energy intake, and total fat in 20 cats
 Weight (kg)BMI (kg/m2)Girth (cm)Energy intake (kcal/kg)Total fat (%)
  1. Values with the same superscript letter differ significantly (all P < 0.001). ND=not done

  2. The data (mean ± SEM) are from 10 males and 10 females at 0-60 % body weight increases and 12 cats (6 males and 6 females) at 100%. There was no difference between the two diet groups and the results were combined.

0%3.4 ± 0.2a32.4 ± 1.1b30.2 ± 0.7c40.6 ± 1.1d25 ± 1e
10%3.9 ± 0.2a36.9 ± 1.3b35.3 ± 1c52.9 ± 1dND
30%4.4 ± 0.2a42.4 ± 1.5b38.6 ± 0.8c59.7 ± 1.1d45 ± 2 e
60%5.7 ± 0.3a54.7 ± 2.1b45.2 ± 0.8c64.4 ± 1.8d57 ± 2e
100%6.9 ± 0.5a64.6 ± 2.3b49.2 ± 1.0c81.5 ± 6.3d63 ± 1e

Glucose and insulin

The results for baseline glucose (mg/dL) and AUCG are shown in Table 3 and Figure 1, respectively. There was no significant difference between the baseline glucose concentrations in any of the groups and there was no diet effect. An increase in weight up to 30% did not lead to an increase in the AUCG but the AUCG was significantly higher at 60% and 100% (Figure 1). In contrast to glucose, baseline insulin concentrations (pmol/L) increased and correlated significantly with increasing body weight (Table 3; r2 = 0.544; p < 0.0001). The AUCI also significantly increased with increasing body weight (Figure 1); however, the average increase from basal remained more or less constant (not shown). It can be deduced that the weight-related increase in the AUC during the IVGTT parallels the increase in basal insulin concentrations.

Table 3. Concentrations of glucose (mg/dl), insulin (pmol/l), adiponectin (ug/mL), leptin (ng/mL) and proinsulin (pmol/L) in cats
 0%10%30%60%100% 
  1. Values with the same superscript differ significantly (P < 0.05)

  2. The data (mean + SEM) are from 20 cats (10 males and 10 females) at 0-60% body weight increases and 12 cats (6 males and 6 females) at 100%. There was no difference between the two diet groups and the results were combined.

Glucose89 ± 489 ± 288 ± 187 ± 290 ± 1 
Insulin32.4 ± 3.6a58.1 ± 5.4a85.7 ± 9.4a110.9 ± 12.8a193.0 ± 19.4a 
Adiponectin2.9 ± 0.3a-c2.8 ± 0.3d,e,f1.8 ± 0.2a,d,g,h1.5 ± 0.2b,e,g1.0 ± 0.2c,f,h 
Leptin3.4 + 0.2a5.2 + 0.5a7.8 + 0.7a14.9 + 1.4a25.9 + 3.4a
Proinsulin193 ± 18a- d238 ± 15a,e,f271 ± 22b,e269 ± 27c,g371 ± 34d,f,g 

Table 4 reports the estimated fixed effects of the final model after back-transformation (exponentiation) of the actually estimated parameters, together with the coefficient of variation of the parameter estimate (equal to 100 times the standard error of the log-transformed parameter) and the 95% confidence interval of the parameter estimate. The minimal model parameters P2, SI, and VG were characterized each with a single fixed effect, that is, no effect of covariates on these parameters was found. Only glucose effectiveness SG was found significantly related to changes in body weight. This relationship was expressed for the log-transformed parameter as a linear function centered at 3 kg body weight. Thus, SG in Table 4 represents the average glucose effectiveness at basal insulin for a cat having a BW of 3 kg, while αSG represents the multiplier of SG for a 1 kg increment of BW above the reference 3 kg. The formula, SG (BW) = SG * αSG BW-3 predicts a 15% (=1 − 0.8532 = 0.1468) reduction of glucose effectiveness for each kg weight gain, or a 38% (=1 − 0:8532[3] = 0.3789) reduction for a 3 kg weight gain. In an attempt to include the increase in basal insulin concentrations associated with weight gain in the estimation of insulin sensitivity in addition to the glucose disappearance associated with dynamic supra-basal insulin excursions, a different parameterization of the minimal model was adopted that expresses SG as sum of glucose effectiveness at zero insulin (GEZI) and insulin sensitivity SI multiplied by basal insulin Ib (SG = GEZI + SI IB). Using this new model, a reduction of insulin sensitivity SI of about 17% (=1 − 0.8284 = 0.1716) was predicted for each kg increase.

Table 4. Minimal model parameter estimates and precision*
ParameterValueCV%95% CI
  1. The estimated fixed effects of the final model are reported after back-transformation (exponentiation) of the actually estimated parameters, together with the coefficient of variation of the parameter estimate (equal to 100 times the standard error of the log-transformed parameter) and the 95% confidence interval of the parameter estimate.

  2. SI, SG, and VG represent the insulin sensitivity index, glucose effectiveness at basal insulin, and glucose distribution volume. P2 is a rate constant.

P20.0126413.5[0.0097, 0.0165]
SG0.025757.9[0.0220, 0.0301]
αSG0.85323.1[0.8027, 0.9068]
SI1.218*10−49.9[1.00*10−4, 1.48*10-4]
VG1.8662.32[1.783, 1.952]

Lipid analysis

Diet had no effect on plasma baseline cholesterol, triglyceride, and NEFA. Baseline cholesterol did not change significantly with increasing body weight (not shown). An increase in body weight significantly increased triglyceride concentrations (p < 0.05 for all). They were (mg/dL) 18 ± 2 at 0% and 65 ± 8 at 100% obesity. A significant difference was also seen in NEFA concentration when body weight increased to ≥30 % compared to 0%. The concentrations (mEq/L) were 0.42 ± 0.03 at 0% and between 0.5 ± 0.04 and 0.66 ± 0.05 at 30-100%.

Enzyme assays

The results for catalase, glutathione peroxidase, and superoxide dismutase activity in red blood cells (RBCs) are shown in Table 5. Diet had no effect on enzyme activity and there was no significant difference among the cats with increasing body weights. There was no effect of diet on heparin-releasable SOD and there was no significant difference in enzyme activity among the cats with increasing body weight (Table 5).

Table 5. Activity of anti-oxidant enzymesx
 0%10%30%60%100%
  1. Values with the same superscript differ significantly (all P < 0.05).

  2. xMean (±SEM) catalase (Units/mg Hb), glutathione peroxidase (Gpx) (Units/mg Hb), red blood cell superoxide dismutase (SOD RBC) activity (Units/g Hb), as well as baseline (SOD) and AUC (Units/mL; SOD AUC) concentrations of heparin-releasable plasma superoxide dismutase activity in 20 cats (n = 20, 10 males and 10 females) at 0-60% body weight increases and 12 cats at 100%. The enzyme activities were not significantly different between the diet groups and the results were combined.

Catalase205 ± 27214 ± 26200 ± 25191 ± 28230 ± 33
Gpx97 ± 5102 ± 599 ± 397 ± 4111 + 5
SOD RBC3271 ± 264a4183 ± 184a,b2351 ± 132b-d2834 ± 250c,e3829 ± 241b-e
SOD5.8 ± 0.95.7 ± 0.65.8 ± 1.15.9 ± 1.14.9 ± 0.8
SOD AUC3779 ± 7712719 ± 3192670 ± 4653753 ± 7294277 ± 1191

Cytokine and adipokine analysis

Adiponectin decreased with increasing body weight (Table 3) and was negatively correlated with insulin concentrations (r2 = 0.2808; p < 0.0001), whereas leptin increased with increasing body weight (Table 3) and was positively correlated with insulin concentrations (r2 = 0.4202; p < 0.0001).

Cytokine concentrations were measured in 16 cats at 0-60% weight increase and in 12 cats at 100% weight increase. Diet had no effect on cytokine concentrations. There was no significant difference in IL-6 and Il-1 concentrations (ug/mL) among the cats with increasing body weight. The IL-6 concentrations ranged from ranged from 2.8 ± 0.5 (30% weight increase) to 3.7 ± 0.5 (0% weight increase) and the IL-1 concentrations from 1.9 ± 0.5 (100% weight increase) to 2.7 ± 0.5 (0% weight increase) for IL-1. There was also no significant difference among the TNF-alpha concentrations (ug/mL), ranged from 1.9 ± 0.5 at 0% weight increase to 0.9 ± 0.2 at 60% weight increase.

Indirect calorimetry

Diet did not have an effect on baseline (resting) heat production in cats and the results were combined. Baseline heat production in cats increased significantly with increasing body weight when expressed on a per cat basis but tapered off between 60% and 100% weight gain (Figure 2, Table 6). This was also seen in the postprandial period (Table 6). When resting heat production was expressed on a metabolic body size basis (kg0.75), heat production was highest at 60% weight increase and then decreased at 100% and was similar to the value seen at 0%.

image

Figure 2. Resting heat production (mean ± SE; A. Kcal/hr; B. Kcal/kg0.75/hr) in 20 cats (n = 20, 10 males and 10 females) at different % body weight increases. Diet had no effect and the results were combined.

Download figure to PowerPoint

Table 6. Heat production (Kcal/hr) in 20 cats at different % body weight increasesx
 0%10%30%60%100%
  1. Values with the same superscript differ significantly (all P < 0.05).

  2. xMean (±SEM) values are shown for 20 cats (10 males and 10 females) at 0-60 % body weight increases and 12 cats (6 males and 6 females) at 100%. Diet had no effect and the results were combined. Cats were monitored for 24 hours (2 hours baseline and 22 hours post feeding).

baseline5.6 ± 0.4a-c6.8 ± 0.4d,e,f8.4 ± 0.4a,d,g,h10.4 ± 0.4b,e,g11.2 ± 0.4c,f,h
3-125.8 ± 0.4a-d7.7 ± 0.4a,e,f8.9 ± 0.4b,g,h11.2 ± 0.4c,e,g12.0 ± 0.4d,f,h
12-245.6 ± 0.5a-d7.5 ± 0.5a,e,f8.9 ± 0.5b,g,h11.2 ± 0.5c,e,g12.0 ± 0.6d,f,h

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

To our knowledge, this is the first longitudinal study that examines anthrophometric, hormonal, and biochemical changes during the development of obesity in cats. The obese cat has many similarities to obese people; however, there are also some striking differences. Atherosclerosis, clinical hypertension and cardiovascular disease, which are associated with obesity in people, have not been reported in obese cats, despite the fact that we have shown that obese cats have most of the prerequisites for their development. Obese cats have dyslipidemia characterized by abnormalities in lipoprotein particle number and size [14]. They have more and larger VLDL particles than lean cats, and have a higher number of small LDL and HDL particles [14], all changes associated with an increased risk for cardiovascular disease in people [19, 20]. However, atherosclerosis and clinical hypertension have not been described in obese cats. In a recent study of long-term obese cats in our laboratory, atherosclerosis was not found in any vessels examined during postmortem examination (M. Hoenig and A. Gal, unpublished).

It has been thought that the expanding adipose tissue mass secretes inflammatory cytokines, and that adiposity is associated with low grade inflammation [2]. IL-6 and IL-1 were shown to decrease insulin sensitivity by interfering with insulin signaling through SOCS-3 up-regulation and IRS-1 down-regulation [21, 22]. They are both positively correlated with obesity, insulin resistance and glucose intolerance [1]. Plasma concentrations of IL-6 have been shown to predict the development of T2DM in people [23]. Another inflammatory cytokine, TNF-α, is produced by adipocytes as well as by monocytes and macrophages. Obesity has been shown to lead to increased plasma concentrations in some [24] but not all studies [25, 26]. We have previously shown a higher expression of TNFα in adipose tissue of obese cats which was associated with decreased expression and activity of lipoprotein lipase and speculated that TNFα might serve as a regulator of lipoprotein lipase [27] and be responsible for a partitioning of fatty acids to muscle tissue. However, in that study, plasma concentrations of TNFα were not measured. In the current study, we found that development of obesity in the cat was not associated with higher plasma TNFα concentrations. Tissue expression was not examined because of animal welfare concerns. Because circulating levels are quite low in obesity despite high levels of expression in adipose tissue, it has been suggested by others that TNFα action in obesity is local [28] and that insulin resistance of obesity is more strongly related to local, not circulating TNFα concentrations. It also has been suggested that circulating levels of TNFα receptors may be better indicators of the status of activation of the TNFα system [29]. Higher local production of TNFα has been described in patients with inflammation of the airways, which was not reflected in higher systemic levels [30]. The fact that adipocyte TNFα expression was higher in obese compared to lean cats in our previous study may support the notion that TNFα is produced and acts locally.

ROS have been implicated in the development of insulin resistance and other obesity related metabolic diseases [31]. The increased ROS leads to a decrease in activities of anti-oxidant enzymes, such as glutathione peroxidase, Cu,Zn-superoxide dismutase (SOD), and catalase. In cats, none of the inflammatory mediators were different at 100% obesity than at baseline. The activity of SOD in red blood cells was even higher at 10% obesity than at any other time. The reason for this increase is unknown. Urinary isoprostane concentrations were also not found to be different between long-term obese and aged-matched lean cats (K. Viviano, unpublished). Our findings that none of the antioxidant enzymes or the pro-inflammatory cytokines showed changes indicative of an inflammatory process and that inflammatory cells were also not seen when white adipose tissue was examined histologically (Gal and Hoenig, unpublished) support the notion that inflammation does not play a role in the development of insulin resistance with increasing obesity in cats. This is also supported by results from a study of immune responses in lean and obese cats [32]. In that study, we examined complete blood counts and lymphocyte distribution by flow cytometric analysis with specific fluorescein-conjugated subset markers. Immune function was assessed by measuring the proliferative activity of different cellular fractions in response to several polyclonal mitogens. Phagocytosis and natural killer cell cytotoxicity was examined to test innate immune functions. A similar immune innate and adaptive immune response was elicited regardless of body condition and it was concluded that obesity does not alter immune responses in cats. It is unclear why insulin resistance of obesity in cats is not associated with an inflammatory response and a change in immune function. Future studies should be directed to examine tissue specific up-regulation of protective mechanisms.

Similar to what we described previously [17] using the euglycemic hyperinsulinemic clamp and the frequently sampled high dose (1 g glucose/kg body weight) intravenous glucose tolerance test (FSIVGTT), which included an injection of exogenous insulin, we saw a marked decrease in the insulin-independent uptake of glucose SG. Baseline insulin concentrations increased significantly with increasing body weight, even with a weight gain of just 10%, whereas baseline glucose concentrations remained unchanged. The unchanged basal glucose in the obese versus lean state can be explained only with a simultaneous reduction in endogenous glucose production (EGP). We have previously shown that EGP is indeed decreased in obese cats in both the fasting and postprandial state compared to lean cats [5, 6]. The minimal model data analysis did not evidence significant variations of SI with body weight and all changes in glucose kinetics with body weight were attributed to SG. However, the total fractional glucose clearance during a euglycemic hyperinsulinemic clamp would be SG + SI*delta_insulin, that is, not totally independent of SG unless SG would be negligible. One can therefore assume that insulin sensitivity measured during a clamp and predicted by the current minimal model parameters decreases with increasing body weight because SG decreases with body weight. This was the case, that is, insulin sensitivity was found to decrease with increasing weight gain, when we adopted a different parameterization of the minimal model, which expressed SG as the sum of glucose effectiveness at zero insulin and the product of SI and Ib. A direct comparison between the current results and previous results from the FSIVGTT [17] might not be possible because the FSIVGTT had an additional insulin injection, and the between-animal variability in glucose kinetics was attributed (according to statistical criteria) to SI rather than to SG. This suggests that exogenous insulin administration changes the outcome of the minimal model, especially due to the different route of insulin administration.

As expected, increasing food intake is highly correlated with an increase in body weight, girth, BMI, and % fat. This suggests that weight, girth or BMI are excellent measures of body fat if applied in one and the same animal longitudinally and can be used in a clinical setting where more sophisticated techniques to measure fat mass are not available. The distribution of fat is different than what we reported before [5]. Our previous measurements were made more caudally where fat was equally distributed between the abdominal subcutaneous and visceral space. In this study, we examined a more cranial area between the kidneys. There we found a larger percentage of visceral fat; however, similar to our previous findings, the distribution did not change with changing body condition, that is, visceral and subcutaneous fat increased in parallel, suggesting that obesity does not preferentially increase visceral fat depots in cats. In people, visceral depots are strongly associated with insulin resistance and obesity-related diseases such as dyslipidemia and atherosclerosis [33]. Even in the absence of any change in body weight or total body fat, the amount of visceral fat predicted the changes in glucose tolerance and insulin sensitivity in a prospective study in people [34]. Visceral fat also was related to fasting insulin levels in children, independent of total body fat [35]. In a recent study, the degree of large artery stiffening during weight gain was related to the amount of visceral fat gain and was independent of total fat mass [36]. In obese people, weight loss reduces visceral fat and improves insulin sensitivity [37]. Similar benefits are seen when visceral fat is removed surgically, whereas no beneficial effect is seen with removal of subcutaneous fat [38]. Few investigators have found that subcutaneous abdominal fat is more strongly associated with insulin resistance [39].

The increase in food intake with increasing obesity was accompanied by an increase in heat production. Although the highest total heat production was seen in cats at 100% weight increase, the increase became significantly smaller when their body weight was higher than 60%. This pattern was seen regardless whether the heat production was expressed on a per cat per time basis or on a metabolic body size basis. This indicates that metabolism adapts better to the higher intake of calories in the initial phase of obesity, however, slows at higher body fat content, which would accelerate fat deposition with high levels of obesity and contribute to the difficulty of obese cats losing weight. We have also shown in a previous study that long-term obese cats had higher total heat production than lean cats, although on a metabolic size basis, heat production was less [6]. The change in heat production was more marked than what was seen in this study. The caloric intake was also less compared to the cats at 100% weight gain despite similar body weights, likely because the long-term obese cats were maintaining their weight.

The increase in leptin and decrease in adiponectin with increasing body weight was expected. Leptin regulates food intake and energy expenditure in the hypothalamus and has been shown in previous studies to be higher in obese cats, indicating leptin resistance [5, 6, 17]. We also saw lower adiponectin levels in those obese cats. Adiponectin is a marker of PPAR-gamma activity. We found previously that adipose tissue expression of PPARs as well as the regulatory protein PGC-1 was lower in obese cats [40]. Adiponectin and leptin concentrations normalize with weight loss [5].

In summary, we have shown that development of obesity does not lead to the same changes in circulating inflammatory markers and activity of antioxidant enzymes that is seen in other species. This may explain the fact that obese cats do not develop cardiovascular problems, which are found in people. Diet containing antioxidants did not influence any of the parameters we studied. We have also shown that metabolism increases initially with increasing caloric intake but slows when cats become severely obese (>60% body fat). This may contribute to the difficulty, which obese cats have losing weight.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

This study was, in part, supported by a grant from the Morris Animal Foundation.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  • 1
    Bastard JP, Maachi M, Van Nhieu JT, et al. Adipose tissue IL-6 content correlates with resistance to insulin activation of glucose uptake both in vivo and in vitro. J Clin Endocrinol Metab 2002;87:20842089.
  • 2
    Bastard JP, Maachi M, Lagathu C, et al. Recent advances in the relationship between obesity, inflammation, and insulin resistance. Eur Cytokine Netw 2006;17:412.
  • 3
    Shoelson SE, Lee J, Goldfine AB. Inflammation and insulin resistance. J Clin Invest 2006;116:17931801.
  • 4
    Harle P, Straub RH. Leptin is a link between adipose tissue and inflammation. Ann N Y Acad Sci 2006;1069:454462.
  • 5
    Hoenig M, Thomaseth K, Waldron M, Ferguson DC. Insulin sensitivity, fat distribution, and adipocytokine response to different diets in lean and obese cats before and after weight loss. Am J Physiol Regul Integr Comp Physiol 2007;292:R227R234.
  • 6
    Hoenig M, Jordan, ET, Glushka, J, et al. Effect of macronutrients, age, and obesity on 6 and 24-hour post-prandial glucose metabolism in cats. Am J Physiol Regul Integr Comp Physiol 2011;301:R1798R1807.
  • 7
    Brennan AM, Mantzoros CS. Leptin and adiponectin: their role in diabetes. Curr Diab Rep 2007;7:12.
  • 8
    Lord GM, Matarese G, Howard JK, Baker RJ, Bloom SR, Lechler RI. Leptin modulates the T-cell immune response and reverses starvation-induced immunosuppression. Nature 1998;394:897901.
  • 9
    Fruebis J, Tsao TS, Javorschi S, et al. Proteolytic cleavage product of 30-kDa adipocyte complement-related protein increases fatty acid oxidation in muscle and causes weight loss in mice. Proc Natl Acad Sci U S A 2001;98:20052010.
  • 10
    Bruun JM, Lihn AS, Verdich C, et al. Regulation of adiponectin by adipose tissue-derived cytokines: in vivo and in vitro investigations in humans. Am J Physiol Endocrinol Metab 2003;285:E527E533.
  • 11
    Ouchi N, Kihara S, Arita Y, et al. Adipocyte-derived plasma protein, adiponectin, suppresses lipid accumulation and class A scavenger receptor expression in human monocyte-derived macrophages. Circulation 2001;103:10571063.
  • 12
    Yamaguchi N, Argueta JG, Masuhiro Y, et al. Adiponectin inhibits Toll-like receptor family-induced signaling. FEBS Lett 2005;579:68216826.
  • 13
    Bansilal S, Farkouh ME, Fuster V. Role of insulin resistance and hyperglycemia in the development of atherosclerosis. Am J Cardiol 2007;99:6B14B.
  • 14
    Jordan E, Kley S, Le NA, Waldron M, Hoenig M. Dyslipidemia in obese cats. Domest Anim Endocrinol 2008;35:290299.
  • 15
    Hoenig M, Ferguson DC. Impairment of glucose tolerance in hyperthyroid cats. J Endocrinol 1989;121:249251.
  • 16
    Kley S, Caffall Z, Tittle E, Ferguson DC, Hoenig M. Development of a feline proinsulin immunoradiometric assay and a feline proinsulin enzyme-linked immunosorbent assay (ELISA): a novel application to examine beta cell function in cats. Domest Anim Endocrinol 2008;34:311318.
  • 17
    Hoenig M, Thomaseth K, Brandao J, Waldron M, Ferguson D. Assessment and mathematical modeling of glucose turnover and insulin sensitivity in lean and obese cats. Dom Anim Endo 2006;31:373389.
  • 18
    Pinheiro J, Bates D, DebRoy S, Sarkar D. nlme: Linear and nonlinear mixed effects models. R package version 2009;3:196.
  • 19
    Freedman DS, Otvos JD, Jeyarajah EJ, Barboriak JJ, Anderson AJ, Walker JA. Relation of lipoprotein subclasses as measured by proton nuclear magnetic resonance spectroscopy to coronary artery disease. Arterioscler Thromb Vasc Biol 1998;18:10461053.
  • 20
    Goff DC, Jr, D'Agostino RB, Jr, Haffner SM, Otvos JD. Insulin resistance and adiposity influence lipoprotein size and subclass concentrations. Results from the Insulin Resistance Atherosclerosis Study. Metabolism 2005;54:264270.
  • 21
    Rieusset J, Bouzakri K, Chevillotte E, et al. Suppressor of cytokine signaling 3 expression and insulin resistance in skeletal muscle of obese and type 2 diabetic patients. Diabetes 2004;53:22322241.
  • 22
    Jager J, Gremeaux T, Cormont M, Le Marchand-Brustel Y, Tanti JF. Interleukin-1beta-induced insulin resistance in adipocytes through down-regulation of insulin receptor substrate-1 expression. Endocrinology 2007;148:241251.
  • 23
    Vozarova B, Weyer C, Hanson K, Tataranni PA, Bogardus C, Pratley RE. Circulating interleukin-6 in relation to adiposity, insulin action, and insulin secretion. Obes Res 2001;9:414417.
  • 24
    Hotamisligil GS, Arner P, Caro JF, Atkinson RL, Spiegelman BM. Increased adipose tissue expression of tumor necrosis factor-alpha in human obesity and insulin resistance. J Clin Invest 1995;95:24092415.
  • 25
    Kellerer M, Rett K, Renn W, Groop L, Haring HU. Circulating TNF-alpha and leptin levels in offspring of NIDDM patients do not correlate to individual insulin sensitivity. Horm Metab Res 1996;28:737743.
  • 26
    Hauner H, Bender M, Haastert B, Hube F. Plasma concentrations of soluble TNF-alpha receptors in obese subjects. Int J Obes Relat Metab Disord 1998;22:12391243.
  • 27
    Hoenig M, McGoldrick JB, deBeer M, Demacker PN, Ferguson DC. Activity and tissue-specific expression of lipases and tumor-necrosis factor alpha in lean and obese cats. Domest Anim Endocrinol 2006;30:333344.
  • 28
    Inadera H. The usefulness of circulating adipokine levels for the assessment of obesity-related health problems. Int J Med Sci 2008;5:248262.
  • 29
    Hauner H. Secretory factors from human adipose tissue and their functional role. Proc Nutr Soc 2005;64:163169.
  • 30
    Dentener MA, Louis R, Cloots RH, Henket M, Wouters EF. Differences in local versus systemic TNFalpha production in COPD: inhibitory effect of hyaluronan on LPS induced blood cell TNFalpha release. Thorax 2006;61:478484.
  • 31
    Houstis N, Rosen ED, Lander ES. Reactive oxygen species have a causal role in multiple forms of insulin resistance. Nature 2006;440:944948.
  • 32
    Jaso-Friedmann L, Leary III JH, Praveen K, Waldron M, Hoenig, M. The effects of obesity and fatty acids on the feline immune system. J Vet Immunol Immunopathol 2008;122:146152.
  • 33
    Hamdy O, Porramatikul S, Al-Ozairi E. Metabolic obesity: the paradox between visceral and subcutaneous fat. Curr Diabetes Rev 2006;2:367373.
  • 34
    Thamer C, Machann J, Stefan N, et al. High visceral fat mass and high liver fat are associated with resistance to lifestyle intervention. Obesity (Silver Spring) 2007;15:531538.
  • 35
    Gower BA, Nagy TR, Goran MI. Visceral fat, insulin sensitivity, and lipids in prepubertal children. Diabetes 1999;48:15151521.
  • 36
    Orr JS, Gentile CL, Davy BM, Davy KP. Large artery stiffening with weight gain in humans: role of visceral fat accumulation. Hypertension 2008;51:15191524.
  • 37
    Shadid S, Jensen MD. Effects of pioglitazone versus diet and exercise on metabolic health and fat distribution in upper body obesity. Diabetes Care 2003;26:31483152.
  • 38
    Klein S, Fontana L, Young VL, et al. Absence of an effect of liposuction on insulin action and risk factors for coronary heart disease. N Engl J Med 2004;350:25492557.
  • 39
    Gohda T, Gotoh H, Tanimoto M, et al. Relationship between abdominal fat accumulation and insulin resistance in hemodialysis patients. Hypertens Res 2008;31:8388.
  • 40
    Hoenig M, Caffall Z. Ferguson DC. Triiodothyronine differentially regulates key metabolic factors in lean and obese cats. Dom Anim Endocrinol 2008;34:229237.