Introduction

Aging is associated with hyperinsulinemia, insulin resistance, and glucose intolerance (1, 2, 3, 4). It has been suggested that the insulin resistance that develops with age is a consequence of increased adiposity (5, 6, 7, 8, 9). Supporting a primary role of adiposity are studies in rodents that have shown that, when normalized to lean body mass (LBM),¹ peripheral insulin sensitivity is unchanged in old vs. young (7, 10, 11, 12, 13). Moreover, while Barzilai and colleagues observed a hepatic defect in insulin action in aged rodents, they found that this defect was strongly correlated with adiposity and, in particular, intraabdominal adiposity (8, 13, 14, 15).

As first suggested by Vague in 1947 (16), in comparison with other fat depots, increasing intraabdominal adiposity has the most influence on insulin sensitivity (17, 18, 19, 20, 21). Intraabdominal fat has increased lipolytic activity (22, 23) as well as differential gene expression in key metabolic pathways (24) compared with subcutaneous fat. Thus, increased intraabdominal fat, such as that observed during normal aging, may lead to increased concentrations of circulating fat secretory factors, such as free fatty acids (FFAs), leptin, and adiponectin. It has been proposed that the concentrations of such factors would increase in the portal vein, thereby leading to hepatic insulin resistance (18). Researchers have shown that removal of intraabdominal fat restores hepatic insulin sensitivity in old rats (14, 15). Moreover, the adipocyte insulin receptor knockout, or FIRKO, mouse, which is unable to perform insulin-mediated fat storage, is protected against insulin resistance (25). Taken together, these data support the theory that certain metabolic defects that occur with age are likely caused by intraabdominal obesity.

It was known as early as 1935 (26) that a calorie-limited diet can prevent many of the maladies associated with aging, including obesity, and lengthen the lifespan of rodents. Data currently being gathered in nonhuman primates suggest a similar phenomenon (27, 28). While there have been multiple hypotheses as to how caloric restriction retards the aging process, including decreasing free radical formation (29), caloric restriction prevents the intraabdominal obesity that typically appears with age. Barzilai et al. (15) showed that rats fed a restricted diet had comparable insulin sensitivity to those animals in which they performed surgical removal of fat. When these rats were re-fed a normal ad libitum diet, intraabdominal fat returned concomitantly with insulin resistance. Therefore, by reducing the amount of fat deposited, one might prevent insulin resistance and the associated pathophysiologies, subsequently lengthening lifespan.

Short periods of fat feeding are also known to induce obesity and insulin resistance (30, 31). Storlien et al. (30) showed that male Wistar rats fed high fat diets (60% calories from fat) had greater fat mass and insulin resistance in both the periphery and the liver as measured by hyperinsulinemic euglycemic clamp. In addition, studies performed on dogs in this laboratory have shown that even a moderate increase in calories from fat (8% to 9%) without any increase in total caloric intake is sufficient to increase intraabdominal adiposity by 75% (21). Moreover, this latter increase in adiposity was associated with a complete inability of physiological hyperinsulinemia to suppress glucose production, further emphasizing an important role for intraabdominal fat in the development of insulin resistance, especially at the liver.

One approach to study the role of visceral adiposity is to compare the components of insulin sensitivity in the old ad libitum–fed vs. young fat-fed rat. We hypothesize that, if intraabdominal obesity causes insulin resistance, the manner in which that fat is accumulated (either by age or diet) should be of little importance to the overall pattern of insulin resistance. That is, old rats should exhibit similar impairments in insulin sensitivity as observed when feeding fat to young rats.

Research Methods and Procedures

Animals

Four groups of male Fischer Brown Norway rats (F344 X BN F1; National Institute on Aging, Bethesda, MD) were used in this study: 1) young rats fed ad libitum (YL: 6 months); 2) young rats fed a high-fat diet (YF: 6 months); 3) old rats fed ad libitum (OF: 24 months); and 4) old rats fed a calorically restricted diet (OL: 24 months). Animals were housed in separate cages under controlled temperature and lighting (standard 12:12 light:dark cycle). Ad libitum–fed rats were given a standard chow diet provided by NIH (Diet NIH-31; 4.02 kcal/g). The high-fat diet consisted of 5.0 kcal/g, 66.5% of which were from fat (lard), 21% from protein, and 12.5% from carbohydrate (Harlan Teklad, Madison, WI). Restricted animals were raised at the National Institute on Aging, where they were limited to 60% of ad libitum calorie intake per day using an NIH-31–fortified diet (3.95 kcal/g). All animals were given free access to water. Animals from each group were monitored for food intake (grams per day) and body weight for 2 to 4 weeks before use in the study. All procedures were approved by the Institutional Animal Care and Use Committee at the University of Southern California.

Catheters

Three to 4 days before the experiment, animals were placed in cages with wire floors. One end of the cage was equipped with a hole large enough to accommodate the rat's tail. The distal one-third of the tail was drawn through this hole and secured using a rubber stopper and tape. With this arrangement, the animal was allowed to move in the cage to gain free access to food and water. This restraint was required to allow the animals to acclimate to the restrained position necessary to protect the patency of the catheters placed later for the experiment. Two tail vein catheters for infusion were inserted the day before the experiment. On the morning of the experiment (6:00 AM), under local anesthesia, one tail artery catheter was placed for sampling (2% lidocaine; Phoenix Pharmaceuticals, St. Joseph, MO). Animals were returned to the cage without food in the original restrained arrangement and were allowed to move freely up to and during the time of the experiment. Tail artery catheter patency was maintained by an infusion of saline with 10 U/mL heparin (1.0 mL/h). Animals were allowed to sit quietly until the beginning of the experiment at 12:00 PM. Each animal underwent a euglycemic hyperinsulinemic clamp (EGC) as described below.

EGC

D-[3-³H]-glucose (0.2 Ci/min; "tracer") was infused beginning at t = -240 minutes. Four arterial samples to measure basal (B) glucose turnover were taken at t = -60, -45, -30, and –15 minutes. At time 0, an intravenous infusion of insulin (5 mU/min per kilogram) was started. Samples were taken at 10-minute intervals, and plasma glucose was monitored after the onset of infusion. Dextrose (20% unlabeled) was infused through the remaining venous catheter at a variable rate to maintain euglycemia. Euglycemia was defined as the average basal glucose for each individual rat. Steady state (SS) was defined as the last 30 minutes of the clamp (t = 120 to 150 minutes). Blood samples were assayed for glucose, insulin, tracer, FFAs, and glycerol.

Blood Sampling and Body Composition

Blood samples for glucose, insulin, and tracer determination were collected in tubes coated with heparin and lithium fluoride and centrifuged immediately for separation of plasma. Samples collected for FFAs and glycerol were collected in tubes containing diethyl p-nitrophenyl phosphate (Paraoxan, Sigma, St. Louis, MO) to inhibit lipoprotein lipase within samples (32) and centrifuged immediately, and plasma was stored at –20 °C until assay. To estimate abdominal adiposity (IA Fat), epididymal and perirenal fat pads were excised and weighed for each rat at the end of their respective experiment. LBM was estimated by subtracting total IA Fat tissue weight from whole body weight.

Assays

Plasma glucose was measured using the automated glucose analyzer YSI 2300 STAT Plus (YSI, Yellow Springs, OH). Insulin was measured using a rat insulin ELISA kit obtained from Linco (St. Charles, MO). Plasma nonesterified FFAs were measured using the FFA Assay Kit from Wako Chemicals (Neuss, Germany), and glycerol was measured using Triglyceride Reagent from Sigma Diagnostics (St. Louis, MO).

To determine D-[3-³H]-glucose, plasma samples were deproteinized with BaOH₂ and Zn₂SO₄ as described by Somogyi (33). Briefly, the supernatant was placed in scintillation vials and dried at 70 °C in a vacuum oven. Once dry, samples were reconstituted in double distilled water and counted in ReadySafe scintillation fluid (Beckman Instruments, Fullerton, CA) on a scintillation counter (Beckman).

Materials

High-performance liquid chromatography–purified D-[3-³H]-glucose was obtained from New England Nuclear (Boston, MA). Purified pork insulin was obtained from Novo Nordisk (Bagsvaerd, Denmark). Dextrose (20%) was purchased from Braun (Irvine, CA).

Data Analysis

Glucose turnover, the rates of glucose uptake (Rd), and endogenous glucose production (EGP) were calculated using classic tracer dilution methodology (34). Because Rd is dependent primarily on muscle mass, Rd was normalized to kilograms of LBM, whereas EGP was expressed per kilogram of body weight. Whole body insulin sensitivity (Si) was defined as the SS glucose infusion rate (G_{INF SS}) divided by the change in plasma insulin (SS - B) times the SS glucose (G_{INF SS}/(insulin glucose_SS). The ability of hyperinsulinemia to stimulate Rd (Si_Rd) was defined as the increase in Rd (SS - B) divided by the incremental insulin times the SS glucose [(Rd_SS - Rd_B)/( insulin glucose_SS)]. Finally, the effect of insulin to suppress EGP (S_{I EGP}) was defined as the decrease in EGP (B - SS) divided by the incremental insulin times the SS glucose [(EGP_B - EGP_SS)/(insulin glucose_SS)]. All units for insulin sensitivity measures were expressed in units of 10³ dl/min kg per pM.

Statistics

All data are represented as means SE. Group comparisons were made using ANOVA, and when significance was observed, post hoc Student's t tests or paired t tests, when appropriate, were performed. Both linear and nonlinear regression analysis were used to assess the relationship between sensitivity and adiposity. ANOVAs and regressions were performed using MINITAB Statistical Software (State College, PA) and Student's t tests using Excel 2000, with statistical significance set at p 0.05.

Results

Body Composition

Body composition and food intake data are shown in Table 1. Although it was somewhat surprising that OL rats ate a similar number of calories per day compared with OF, it was predictable that intake for restricted rats was significantly reduced vs. YL rats of similar size and LBM (Table 1). Old rats were markedly heavier than YL rats (p < 0.0001), whereas caloric restriction prevented this weight gain, resulting in reduced body weight compared with YL rats (p = 0.001). In contrast, YF rats did not differ from young controls (p = 0.12). Interestingly, despite negligible weight gain, YF rats had markedly increased IA Fat/LBM vs. both YL (p < 0.0001) and OF rats (p = 0.0013).

Table 1. - Food intake and body composition for all animals used in study.

Full table

Basal Plasma Chemistry and Turnover

Basal glucose was similar among all groups (p 0.074; Table 2) except OL vs. YF rats; basal glucose was modestly increased in YF vs. OL rats (10%; p = 0.018). Both obese groups (OF and YF) exhibited marked basal hyperinsulinemia (>2-fold increase) compared with lean groups (YL and OL), suggestive of insulin resistance. Despite increased adiposity and likely insulin resistance, basal FFAs were not elevated in OF rats (p = 0.12 vs. YL). In contrast, FFAs were significantly increased in YF rats (p = 0.016 vs. YL). Basal glucose turnover was similar between most groups (p 0.07) except for OL rats vs. obese animals (OF and YF), in which case obese animals had a significantly smaller basal turnover (p 0.024; Table 2).

Table 2. - Basal plasma chemistry and basal glucose turnover for all animals used in study.

Full table

Insulin Sensitivity

During clamps, euglycemia was achieved by 90 minutes in all groups (B vs. SS: p 0.29; Figure 1A). Lactate increased significantly by hyperinsulinemia in all groups (p 0.012; Figure 1B). Both obese groups required 40% to 50% less G_inf to maintain euglycemia (OF: 21.27 1.7 and YF: 17.38 1.0 mg/min per kilogram) than their lean counterparts (YL: 36.65 1.1 and OL: 38.73 1.8 mg/min per kilogram; p 0.0001; Figure 1C). A decreased G_inf was required despite similar incremental insulin (B vs. SS: 10 to 11 pM; Figure 1D) among all groups (p 0.10), suggesting insulin resistance in both obese models.

Figure 1.

Time-course of (A) glucose, (B) lactate, and (C) glucose infusion rate during clamps of YL (n = 8; ), YF (n = 7; ), OF (n = 8; ), and OL (n = 8; ) rats. (D) Change in insulin between basal and steady state for all groups; no statistical differences were found in insulin by ANOVA.

Full figure and legend (94K)

Hyperinsulinemia augmented Rd in all groups (p < 0.001; Figure 2A). However, while YL and OL rats exhibited an 4-fold increase in Rd (SS Rd; YL: 39.35 2.18 and OL: 41.82 2.28 mg/min per kg LBM), we observed only a 3-fold increase in YF rats (SS Rd: 24.54 1.79 mg/min per kg LBM). Similar to YF rats (p = 0.54), OF rats also showed a tendency for impairment in insulin's ability to stimulate Rd (SS Rd: 24.80 1.79 mg/min per kg LBM), although this difference did not reach significance (vs. YL: p = 0.078; vs. OL: p = 0.054). Interestingly, hyperinsulinemia suppressed EGP substantially in all groups (YL: 67 12; OF: 66 10; OL: 66 16%; p 0.001) except YF rats (p = 0.10; Figure 2B). Furthermore, suppression was similar for YL, OF, and OL rats (p 0.92).

Figure 2.

Basal (solid black bars) and steady-state (hatched bars) (A) glucose uptake (Rd) and (B) endogenous glucose production (EGP) during clamps for YL (n = 8), YF (n = 7), OF (n = 8), and OL (n = 8) rats, *p < 0.05 vs. basal, paired Student's t test.

Full figure and legend (67K)

Insulin sensitivity was measured as whole body (Si_Clamp) and separated into peripheral (Si_Rd) and hepatic (Si_EGP) components (Figure 3). Old rats exhibited significant impairment of Si_Clamp compared with YL (17.69 1.89 vs. 28.99 1.54 10³ dL/min kg per pM, respectively; p < 0.001; Figure 3A), with a similar impairment in fat-fed rodents (13.59 1.18 10³ dL/min kg per pM; p = 0.092 vs. OF). Caloric restriction not only prevented resistance in old rats, but significantly improved Si_Clamp compared with YL (36.99 2.62 10³ dL/min kg per pM; p = 0.023). Si_Rd was affected similarly (Figure 3B). In stark contrast, Si_EGP was not impaired in OF rats compared with YL (OF: 4.37 0.86 vs. YL: 5.03 0.93 10³ dL/min kg per pM; p = 0.61) or OL (6.44 1.51 10³ dL/min kg per pM; p = 0.26; Figure 3C). YF rats, however, exhibited an 75% decrement in hepatic insulin sensitivity compared with YL rats (p = 0.013).

Figure 3.

(A) Si_Clamp (10³ dL/min kg per pM), (B) Si_Rd (10³ dL/min kg LBM per pM), and (C) Si_EGP (10³ dL/min kg per pM) for YL (n = 8), YF (n = 7), OF (n = 8), and OL (n = 8) rats. Different letters indicate a significant difference between groups as assessed by ANOVA and post hoc Student's t tests, p < 0.05.

Full figure and legend (56K)

The effects of insulin in lean and obese animals on FFA and glycerol release were also examined (Figure 4). Hyperinsulinemia suppressed FFAs significantly in all groups (p 0.002; Figure 4A). However, while FFA suppression in lean rats was 85% to 90% (p = 0.10 for YL vs. OL), FFAs in obese rats were suppressed by only 55% (p = 0.63 OF vs. YF; p 0.0039 lean vs. obese; Figure 4B). Glycerol measurements tended to be more variable, but the trends were similar to those observed for FFAs (Figure 4 C and D).

Figure 4.

The effect of hyperinsulinemia on plasma FFA and glycerol. (A) Basal (solid black bars) and steady-state (hatched bars) FFA and (B) percent suppression of FFA during clamps for YL (n = 8), YF (n = 7), OF (n = 8), and OL (n = 8) rats. (C) Basal (solid black bars) and steady-state (hatched bars) glycerol and (D) percent suppression of glycerol. *p < 0.05 vs. basal, paired Student's t test. Different letters indicate a significant difference between groups as assessed by ANOVA and post hoc Student's t tests, p < 0.05.

Full figure and legend (98K)

Effect of Adiposity on S_I

Insulin sensitivity at both the muscle and the liver was negatively correlated by linear regression analysis to IA Fat (p < 0.001 and p = 0.001, respectively; Figure 5 A and B). However, the effect of IA Fat may not be the only factor limiting Si_Rd and Si_EGP; IA Fat/LBM accounted for 41% of the variation in Si_Rd (r = 0.64). Similarly, IA Fat explained only 32% of the variation in Si_EGP (r = 0.57). Of greater interest, when animals were grouped by age (young: YL and YF; old: OF and OL) and regression analysis was repeated, old animals exhibited a significantly steeper negative slope compared with young animals (Si_Rd: 5.19 vs. 2.85; Si_EGP: 1.76 vs. 0.90; p < 0.05) between sensitivity and IA Fat/LBM (Figure 5 C and D). These data indicate that, while increased adiposity may explain part of the resistance in old rats, old animals exhibit increased vulnerability to increased fat deposition compared with young animals.

Figure 5.

Linear regression analysis of Si_Rd vs. IA Fat/LBM (A and C) and Si_EGP vs. IA Fat/LBM (B and D) for YL (n = 8; ), YF (n = 7; ), OF (n = 8; ), and OL (n = 8; ) rats. A and B depict a single linear regression for all groups, whereas C and D depict separate regression lines for young (YL and YF; solid line) and old (OF and OL; dashed line) animals. Comparisons of regression lines in C and D were made using a Student's t test for analysis of slope equivalence.

Full figure and legend (61K)

Discussion

Studies suggest that the insulin resistance observed with age is a result of increased adiposity rather than aging per se (5, 7, 8, 9, 17, 35). In apparent concordance with this hypothesis, in this study, old rats exhibited comparable whole body and peripheral resistance compared with young rats fed a high-fat diet known to induce obesity. However, there were important deviations from the hypothesis that all changes can be accounted for by increased fat mass. YF rats had a significantly greater degree of adiposity compared with older rats when normalized to lean body mass. However, despite greater relative adiposity, YF rats did not manifest a corresponding decrease in peripheral insulin sensitivity compared to older animals. More profound was our demonstration that, despite marked intraabdominal obesity, old rats failed to express significant hepatic resistance, unlike YF rats, which exhibited a 75% decrement in liver insulin action. Furthermore, we observed a markedly steeper relationship between adiposity and insulin sensitivity for old vs. young animals: old rats, regardless of treatment, exhibited an increased propensity to develop insulin resistance with greater adiposity. Here we show that, while intraabdominal adiposity may contribute significantly to the insulin resistance observed with age, age per se imparts an increased susceptibility to obesity-induced insulin resistance.

Old rats were nearly 2-fold more obese than young, and caloric restriction not only prevented this fat gain, it reduced adiposity by 40% compared with young. These data are in agreement with other studies in rodents (7, 14, 36) and rhesus monkeys (28, 37, 38). For example, Bertrand et al. (36) reported that caloric restriction prevents the 2-fold increase in adipose mass observed in male Fischer 344 rats between 6 and 24 months of age. Moreover, it is well established that humans garner increased adipose tissue between the ages of 30 and 70 years (39), the period in which decrements in insulin sensitivity also become apparent. This further signifies the importance of studying insulin resistance in an aging model in the context of coincident obesity.

Aging was associated with marked basal hyperinsulinemia despite normal glycemia (2-fold increase) comparable basal with that observed with high-fat feeding, and this age-related increase in basal insulin was prevented in food-restricted animals. Basal hyperinsulinemia is suggestive of an insulin-resistant state. Consistent with this, both old and fat-fed rats exhibited marked whole body insulin resistance as indicated by an 45% decrease in the glucose infusion rate necessary to maintain euglycemia. Caloric restriction not only prevented age-related resistance, but improved whole body sensitivity by 25%. It has been suggested that prevention of obesity in restricted animals is protective against age-related insulin resistance (14, 40, 41, 42). Our data support this hypothesis; restricted rats were both 40% leaner and more insulin sensitive than young controls.

Peripheral insulin sensitivity was also markedly impaired in age-related and diet-induced obesity (Figure 3), which is consistent with reports in humans (5, 43, 44) and nonhuman primates (45). In contrast, some have suggested that, in rodents, peripheral sensitivity is not altered by the aging process per se (7, 12), once indices are expressed per LBM. However, Fink et al. (43) reported that elderly humans manifest a marked impairment in glucose uptake even after taking into account differences in both body fat and LBM. The apparent disparity between these studies and those of other rodent studies (7, 10, 12) may be explained by methodological differences; this study was designed to give an accurate measure of insulin sensitivity such that the individual components of sensitivity (i.e., peripheral, hepatic) could be measured simultaneously at physiological hyperinsulinemia.

Surprisingly, despite marked obesity and significant peripheral resistance, old rats did not exhibit a detectable hepatic defect in insulin action. These data are in stark contrast to those obtained in the fat-fed model, in which we found a 75% deficit in hepatic insulin sensitivity. Although others have shown severe hepatic insulin resistance in aged rodents, which they attribute to intraabdominal obesity (8, 13), it is possible that the rats used in this study did not attain the same degree of intraabdominal obesity as that observed in these previous models of similar age and weight. Alternatively, our data suggest that while adiposity, and specifically intraabdominal adiposity, may play an important role in the development of insulin resistance with age and fat feeding, it may not be the sole cause. This theory is supported by analyses shown in Figure 5.

Although regression analyses revealed significant correlations between insulin sensitivity and adiposity, adiposity could explain only 30% to 40% of the variation in either peripheral or hepatic sensitivity. Two very important findings in this study reveal why the correlations were not stronger. First, despite 50% greater adiposity in fat-fed rats, peripheral insulin sensitivity was not further impaired compared with old rats. Evidence for this limiting effect of adiposity on peripheral sensitivity has been previously reported (7, 35). Second, as discussed earlier, despite marked obesity in old compared with young controls, hepatic insulin sensitivity was not changed. In addition, once groups were separated according to age, linear regression revealed a distinct difference between the responses of old and young rats to increased adiposity. These data suggest that age per se may, in fact, play a significant role in the propensity for animals to develop insulin resistance with ensuing obesity.

It is equally likely that the mechanism(s) of insulin resistance in the fat-fed and old rat models are independent of IA Fat accumulation. Another important consideration in hepatic insulin resistance is the delivery of certain factors directly to the liver through the portal vein. The so-called "Portal Hypothesis" proposes that visceral fat accumulation (the circulation of which drains directly into the portal vein) generates an increased flux of adipocyte secretory factors to the liver (18). In the rodent, mesenteric fat is the only fat depot that drains into the portal vein; therefore, it is possible that fat-fed, but not old, rats had enlarged mesenteric fat depots. Furthermore, because subcutaneous fat was not quantified in this study, we cannot exclude the possible influence of differences in IA Fat-to-subcutaneous fat ratio between fat-fed and aged rats on insulin sensitivity measures. Further study will be necessary to clarify this possibility.

We are unable to exclude possible influences of differences in diet composition in this study. Specific diet effects on insulin sensitivity, as well as effects caused by the age of diet exposure, have been well documented in rodents (46, 47). Moreover, it has been shown that triglyceride deposition in insulin-sensitive tissues is highly correlated with the incidence of resistance (48, 49). The A-ZIP/F mouse, an example of severe lipoatrophy, suffers from fatty liver and muscle, with severe insulin resistance (50, 51). In this study, triglyceride deposition was likely greater in fat-fed animals compared with old animals because 66% of the fat-fed animals' caloric intake was derived from lard. This may help explain the presence of hepatic resistance in fat-fed, but not aged, rats (46). It is therefore likely that IA Fat accumulation alone cannot fully explain insulin resistance in either fat-fed or aged animals.

Although both surgical removal of IA Fat and caloric restriction completely reverse/prevent hepatic resistance in aged rodent models, the mechanisms by which such results occur with these interventions may be mutually exclusive. Lane et al. (52) showed that, in short-term restricted rhesus monkeys, changes in insulin responsiveness become apparent before any significant changes in body composition. Additionally, Davidson (2) suggests that, because of the nature of caloric restriction paradigms, animals eat over a much shorter period of time; therefore, feeding regimens in these animals more resemble meal feeding rather than a mere decrease in calories. Still further, it might be argued that the link between epididymal and perirenal adipose tissue depot removal and reversal of hepatic resistance is indirect rather than direct, because their ablation can lead to changes in gene/protein expression in other fat depots (15). Hence, it is difficult to substantiate the connection between surgical removal of a tissue vs. reduction of this tissue mass by alteration of food intake and feeding patterns.

Whereas obesity is associated with muscle and liver insulin resistance, it has also been linked to resistance at the adipocyte (53). Although the intent of this study was not to measure lipolysis directly, the sensitivity of this process to insulin can be inferred. We showed that the ability of hyperinsulinemia to suppress systemic FFAs during clamp conditions was impaired by 30% in both obese models and completely recovered with dietary restriction. These data indicate that insulin action at the adipocyte is impaired in both aging and fat feeding. It has been suggested that FFAs are one likely candidate for the indirect effects on insulin action on the liver (54, 55). Also, FFAs have been implicated as a possible secretory factor responsible for hepatic insulin resistance in the portal theory (18, 19). Assuming the portal theory is correct, fat-fed, but not old, rats would have elevated portal FFAs. Future experiments will attempt to address this hypothesis.

In conclusion, age-associated insulin resistance may not be explained solely by concomitant intraabdominal obesity. Although significant, the variation in intraabdominal adiposity could account for only 30% to 40% of the changes in both peripheral and hepatic sensitivity in rodents studied here. Moreover, old animals showed a significantly higher propensity to develop peripheral insulin resistance with smaller fat deposition. Therefore, the linkage between IA Fat depot accretion and resistance may merely be correlative in nature. The lack of hepatic insulin resistance in old animals studied here further supports this hypothesis. We propose that mesenteric fat accumulation, rather than epididymal and perirenal fat, and subsequent increased delivery of adipose secretory factors such as FFAs to the liver are more likely a direct link to hepatic resistance in obese models. Future studies will be performed to address how the mechanisms responsible for insulin resistance in these two obese models differ.

Notes

¹ Nonstandard abbreviations: LBM, lean body mass; FFA, free fatty acid; YL, young ad libitum fed; YF, young fat fed; OF, old ad libitum fed; OL, old calorically restricted; EGC, euglycemic hyperinsulinemic clamp; IA Fat, intraabdominal fat; B, basal; SS, steady state; Rd, rate of glucose uptake; EGP, endogenous glucose production; Si, insulin sensitivity.

References

1. Andres, R., Robin, J. D. (1975) Aging and the disposition of glucose. Adv Exp Med Biol. 61: 239–249.
2. Davidson, M. (1979) The effect of aging on carbohydrate metabolism: a review of the English literature and a practical approach to the diagnosis of diabetes mellitus in the elderly. Metabolism. 28: 688–705.
3. Reaven, G. M., Reaven, E. P. (1985) Age, glucose intolerance, and non-insulin-dependent diabetes mellitus. J Am Geriatr Soc. 33: 286–290.
4. Fraze, E., Chiou, M., Chen, I., Reaven, G. M. (1987) Age-related changes in postprandial plasma glucose, insulin, and free fatty acid concentrations in nondiabetic individuals. J Am Geriatr Soc. 35: 224–228.
5. Coon, P., Rogus, E., Drinkwater, D., Muller, D., Goldberg, A. (1992) Role of body fat distribution in the decline of insulin sensitivity and glucose tolerance with age. J Clin Endocrinol Metab. 75: 1125–1132.
6. Ferrannini, E., Vichi, S., Beck-Nielsen, H., Laakso, M., Paolisso, G., Smith, U. (1996) Insulin action and age. Diabetes. 45: 947–953. | PubMed | ISI | ChemPort |
7. Barzilai, N., Rossetti, L. (1995) Relationship between changes in body composition and insulin responsiveness in models of the aging rat. Am J Physiol. 269: E591–E597.
8. Barzilai, N., Rossetti, L. (1996) Age-related changes in body composition are associated with hepatic insulin resistance in conscious rats. Am J Physiol. 270: E930–E936.
9. Barzilai, N., Gupta, G. (1999) Interaction between aging and syndrome X: new insights on the pathophysiology of fat distribution. Ann NY Acad Sci. 892: 58–72. | PubMed | ChemPort |
10. Goodman, M., Dluz, S., McElaney, M., Belur, E., Ruderman, N. (1983) Glucose uptake and insulin sensitivity in rat muscle: changes during 3–96 weeks. Am J Physiol. 244: E93–E100.
11. Narimiya, M., Azhar, S., Dolkas, C., Mondon, C., Sims, C., Wright, D. (1984) Insulin resistance in older rats. Am J Physiol. 246: E397–E404. | PubMed | ChemPort |
12. Nishimura, H., Kuzuya, H., Okamoto, M., et al. (1988) Change of insulin action with aging in conscious rats determined by euglycemic clamp. Am J Physiol. 254: E92–E98.
13. Gupta, G., Cases, J. A., She, L., et al. (2000) Ability of insulin to modulate hepatic glucose production in aging rats is impaired by fat accumulation. Am J Physiol. 278: E985–E991. | ChemPort |
14. Barzilai, N., Banerjee, S., Hawkins, M., Chen, W., Rossetti, L. (1998) Caloric restriction reverses hepatic insulin resistance in aging rats by decreasing visceral fat. J Clin Invest. 101: 1353–1361. | PubMed | ISI | ChemPort |
15. Barzilai, N., She, L., Liu, B., et al. (1999) Surgical removal of visceral fat reverses hepatic insulin resistance. Diabetes. 48: 94–98. | Article | PubMed | ISI | ChemPort |
16. Vague, J. (1947) La differenciation sexuelle, facteur determinant des formes de l'obesity. Presse Med. 30: 339–340.
17. Kohrt, W. M., Kirwan, J. P., Staten, M. A., et al. (1993) Insulin resistance in aging is related to abdominal obesity. Diabetes. 42: 273–281. | PubMed | ISI | ChemPort |
18. Björntorp, P. (1990) "Portal" adipose tissue as a generator of risk factors for cardiovascular disease and diabetes. Arteriosclerosis. 10: 493–496. | PubMed | ISI | ChemPort |
19. Bergman, R. N., Van Citters, G. W., Mittelman, S. D., et al. (2001) Central role of the adipocyte in the metabolic syndrome. J Invest Med. 49: 119–126.
20. Peiris, A., Mueller, R., Smith, G., Struve, M., Kissebah, A. (1986) Splanchnic insulin metabolism in obesity: influence of body fat distribution. J Clin Invest. 78: 1648–1657. | PubMed | ISI | ChemPort |
21. Kim, S., Ellmerer, M., VanCitters, G., Bergman, R. N. (2003) Primacy of hepatic insulin resistance in the development of the metabolic syndrome induced by an isocaloric moderate fat diet in the dog. Diabetes. 52: 2453–2460. | PubMed | ChemPort |
22. Morimoto, C., Kameda, K., Tsujita, T., Okuda, H. (2001) Relationships between lipolysis induced by various lipolytic agents and hormone-sensitive lipase in rat fat cells. J Lipid Res. 42: 120–127.
23. Montague, C., O'Rahilly, S. (2000) Perspectives in diabetes: the perils of portliness: causes and consequences of visceral adiposity. Diabetes. 49: 883–888. | Article | PubMed | ChemPort |
24. Atzmon, G., Yang, X., Muzumdar, R., Ma, X., Gabriely, I., Barzilai, N. (2002) Differential gene expression between visceral and subcutaneous fat depots. Hormone Metab Res. 34: 622–628.
25. Bluher, M., Kahn, B. B., Kahn, C. R. (2003) Extended longevity in mice lacking the insulin receptor in adipose tissue. Science. 299: 572–574. | Article | PubMed | ISI | ChemPort |
26. McCay, C., Crowell, M., Maynard, L. (1935) The effect of retarded growth upon the length of life span and upon ultimate body size. J Nutr. 10: 63–79. | ISI | ChemPort |
27. Lane, M. A., Baer, D. J., Tilmont, E., et al. (1995) Energy balance in rhesus monkeys subjected to long-term dietary restriction. J Gerontol. 50A: B295–B302.
28. Roth, G., Ingram, D., Lane, M. (1999) Caloric restriction in primates: will it work and how will we know? J Am Geriatr Soc. 47: 896–903. | PubMed | ISI | ChemPort |
29. Masoro, E. (1996) Possible mechanisms underlying the antiaging actions of caloric restriction. Toxicol Pathol. 24: 738–741.
30. Storlien, L. H., James, D. E., Burleigh, K. M., Chisholm, D., Kraegen, E. W. (1986) Fat feeding causes widespread in vivo insulin resistance, decreased energy expenditure, and obesity in rats. Am J Physiol. 251: E576–E583. | PubMed | ChemPort |
31. Storlien, L. H., Jenkins, A. B., Chisholm, D. J., Pascoe, W. S., Khouri, S., Kraegen, E. W. (1991) Influence of dietary fat composition on development of insulin resistance in rats. Diabetes. 40: 280–289. | Article | PubMed | ISI | ChemPort |
32. Zambon, A., Hashimoto, S., Brunzell, J. (1993) Analysis of techniques to obtain plasma for measurement of levels of free fatty acids. J Lipid Res. 34: 1021–1028.
33. Somogyi, M. (1945) Determination of blood sugar. J Biol Chem. 160: 69–73. | ISI | ChemPort |
34. Steele, R., Wall, J. S., De Bodo, R. C., Altszuler, N. (1956) Measurement of size and turnover rate of body glucose pool by isotope dilution method. J Biol Chem. 187: 15–24.
35. Bogardus, C., Lillioja, S., Mott, D., Hollenbeck, C., Reaven, G. M. (1985) Relationship between degree of obesity and in vivo insulin action in man. Am J Physiol. 248: E286–E291.
36. Bertrand, H., Lynd, F., Masoro, E., Yu, B. (1980) Changes in adipose mass and cellularity through the adult life of rats fed ad libitum or a life-prolonging restricted diet. J Gerontol. 35: 827–835. | PubMed | ISI | ChemPort |
37. Lane, M., Ingram, D., Roth, G. (1999) Calorie restriction in nonhuman primates: effects on diabetes and cardiovascular risk. Toxicol Sci. 54: 41–48.
38. Roth, G., Ingram, D., Lane, M. (2001) Caloric restriction in primates and relevance to humans. Ann NY Acad Sci. 928: 305–315. | PubMed | ChemPort |
39. Norris, A. H., Lundy, T., Shock, N. W. (1963) Trends in indices of body composition in men between ages 30–70 years. Ann NY Acad Sci. 110: 623–639. | PubMed | ChemPort |
40. Kemnitz, J., Roecker, E. B., Weindruch, R., Elson, D. F., Baum, S. T., Bergman, R. N. (1994) Dietary restriction increases insulin sensitivity and lowers blood glucose in rhesus monkeys. Am J Physiol. 266: E540–E547. | PubMed | ISI | ChemPort |
41. Barzilai, N., Gupta, G. (1999) Revisiting the role of fat mass in the life extension induced by caloric restriction. J Gerontol. 54A: B89–B96.
42. Barzilai, N., Gabriely, I. (2001) The role of fat depletion in the biological benefits of caloric restriction. J Nutr. 131: 903S–906S. | PubMed | ChemPort |
43. Fink, R. I., Kolterman, O. G., Griffin, J., Olefsky, J. M. (1983) Mechanisms of insulin resistance in aging. J Clin Invest. 271: 1523–1535.
44. Rowe, J. W., Minaker, K. L., Pallotta, J. A. (1983) Characterization of the insulin resistance of aging. J Clin Invest. 71: 1581–1587. | PubMed | ISI | ChemPort |
45. Bodkin, N., Ortmeyer, H., Hansen, B. (1995) Long-term dietary restriction in older-aged rhesus monkeys: effects on insulin resistance. J Gerontol. 50: B142–B147.
46. Podolin, D. A., Wei, Y., Pagliassotti, M. J. (1999) Effects of a high-fat diet and voluntary wheel running on gluconeogenesis and lipolysis in rats. J Appl Physiol. 86: 1374–1380.
47. Pagliassotti, M. J., Gayles, E. C., Podolin, D. A., Wei, Y., Morin, C. L. (2000) Developmental stage modifies diet-induced peripheral insulin resistance in rats. Am J Physiol Regul Integr Comp Physiol. 278: R66–R73.
48. Virkamaki, A., Korsheninnikova, E., Seppala-Lindroos, A., et al. (2001) Intramyocellular lipid is associated with resistance to in vivo insulin actions on glucose uptake, antipolysis, and early insulin signalling pathways in human skeletal muscle. Diabetes. 50: 2337–2343. | PubMed |
49. Seppala-Lindroos, A., Vehkavaara, S., Hakkinen, A., et al. (2002) Fat accumulation in the liver is associated with defects in insulin suppression of glucose production and serum FFA independent of obesity in normal men. J Clin Endocrinol Metab. 87: 3023–3028. | Article | PubMed | ISI | ChemPort |
50. Moitra, J., Mason, M. M., Olive, M., et al. (1998) Life without white fat: a transgenic mouse. Genes Develop. 12: 3168–3181.
51. Reitman, M., Mason, M. M., Moitra, J., et al. (1999) Transgenic mice lacking white fat: models for understanding human lipoatrophic diabetes. Ann NY Acad Sci. 892: 289–296.
52. Lane, M., Tilmont, E., Angelis, H., et al. (1999) Short-term calorie restriction improves disease-related markers in older male rhesus monkeys (Macaca mulatta). Mech Ageing Dev. 112: 185–196.
53. Mittelman, S. D., Van Citters, G. W., Kirkman, E. L., Bergman, R. N. (2002) Extreme insulin resistance of the central adipose depot in vivo. Diabetes. 51: 755–761. | Article | PubMed | ChemPort |
54. Rebrin, K., Steil, G. M., Getty, L., Bergman, R. N. (1995) Free fatty acid as a link in the regulation of hepatic glucose output by peripheral insulin. Diabetes. 44: 1038–1045. | PubMed | ISI | ChemPort |
55. Rebrin, K., Steil, G. M., Mittleman, S. D., Bergman, R. N. (1996) Causal linkage between insulin suppression of lipolysis and suppression of liver glucose output in dogs. J Clin Invest. 98: 741–749. | PubMed | ISI | ChemPort |

Acknowledgments

This work was supported by NIH Grants to R.N.B. (DK27619 and DK29867) and M.A. (AG15111). During these studies, K.J.C. was supported by NIH Training Grant AG00093.