Peroxisome proliferator-activated receptor γ (PPARγ) agonists improve insulin sensitivity and lipemia partly through enhancing adipose tissue proliferation and capacity for lipid retention. The agonists also reduce local adipose glucocorticoid production, which may in turn contribute to their metabolic actions. This study assessed the effects of a PPARγ agonist in the absence of glucocorticoids (adrenalectomy, ADX). Intact, ADX, and intact pair-fed (PF) rats were treated with the PPARγ agonist rosiglitazone (RSG) for 2 wk. RSG increased inguinal (subcutaneous) white (50%) and brown adipose tissue (6-fold) weight but not that of retroperitoneal (visceral) white adipose tissue. ADX but not PF reduced fat accretion in both inguinal and retroperitoneal adipose depots but did not affect brown adipose mass. RSG no longer increased inguinal weight in ADX and PF rats but increased brown adipose mass, albeit less so than in intact rats. RSG increased cell proliferation in white (3-fold) and brown adipose tissue (6-fold), as assessed microscopically and by total DNA, an effect that was attenuated but not abrogated by ADX. RSG reduced the expression of the glucocorticoid-activating enzyme 11β-hydroxysteroid dehydrogenase 1 (11β-HSD1) in all adipose depots. RSG improved insulin sensitivity (reduction in fasting insulin and homeostasis model assessment of insulin resistance, both −50%) and triacylglycerolemia (−75%) regardless of the glucocorticoid status, these effects being fully additive to those of ADX and PF. In conclusion, RSG partially retained its ability to induce white and brown adipose cell proliferation and brown adipose fat accretion and further improved insulin sensitivity and lipemia in ADX rats, such effects being therefore independent from the PPARγ-mediated modulation of glucocorticoids.
- peroxisome proliferator-activated receptor-γ
- 11β-hydroxysteroid dehydrogenase
peroxisome proliferator-activated receptor-γ (PPARγ) is a ligand-activated nuclear receptor whose γ2 form is highly expressed in white adipose tissue (WAT) where it regulates the expression of a number of genes involved in adipogenesis and anabolic lipid metabolism (3, 41). The net result of PPARγ activation is a remodeling of WAT with a larger number of smaller adipocytes (2, 14, 48, 57). PPARγ activation also efficaciously ameliorates whole body and muscle insulin resistance (41, 43, 48) as well as hypertriacylglycerolemia (2, 29), albeit at the expense of some gain in fat mass. Synthetic thiazolidinedione PPARγ ligands are used clinically to treat insulin resistance and type 2 diabetes (41, 55).
In addition to beneficial alterations in the pattern of adipocytokine production, the insulin-sensitizing action of PPARγ agonists is believed to derive largely from an enhanced lipid retention by adipose tissue, which in turn reduces exposure of other insulin-sensitive tissues to fatty acids (3, 25). This is achieved through several pathways, which include a depot-specific induction of the expression of proteins involved in triacylglycerol-derived fatty acid uptake and binding, de novo fatty acid and triacylglycerol synthesis, and fatty acid reesterification, along with downregulation of genes involved in lipid release (3, 41). Another relevant gene that has been found to be downregulated in WAT by PPARγ agonists (4, 28) is that which codes for 11β-hydroxysteroid dehydrogenase 1 (11βHSD1), an enzyme expressed in the brain, liver, and adipose tissue that locally converts inactive glucocorticoids into bioactive forms such as cortisol in humans and corticosterone (Cort) in rodents (45). This enzyme is particularly relevant to the metabolic effects of PPARγ agonism because excess glucocorticoids promote visceral obesity, hyperlipidemia, and insulin resistance (52, 56), as seen for instance in human Cushing's syndrome (5, 40). Beyond systemic glucocorticoids, the removal of which prevents or reverses many forms of obesity (7, 42), the importance of 11βHSD1-mediated local amplification of glucocorticoid action in adipose tissue is being increasingly recognized (44). This is vividly illustrated by the development of visceral obesity and the metabolic syndrome in mice overexpressing adipose 11βHSD1 (32) and by the lack thereof in response to high-fat feeding in mice with 11βHSD1 gene invalidation (26, 36, 37). Of note is the fact that, in contrast with PPARγ agonism, glucocorticoids upregulate 11βHSD1 expression in WAT (22, 45, 49). Finally, in addition to glucocorticoid-PPARγ interactions at the level of 11βHSD1, glucocorticoids are well known to be required for adipocyte differentiation in vitro and may conceivably impact PPARγ-mediated modulation of adipose cellularity in vivo.
Whereas the metabolic consequences of excess glucocorticoids are well established, the contribution of alterations in the glucocorticoid status by PPARγ agonists to the beneficial metabolic effects of these compounds is unknown. To gain insight into this issue, the present study was conducted to evaluate the importance of glucocorticoid modulation in PPARγ action by assessing in rats the impact of the PPARγ agonist rosiglitazone (RSG) in the absence of glucocorticoids, achieved through adrenalectomy (ADX). The end points included indexes of adipose tissue accretion and cellularity, as well as insulin sensitivity and lipemia, which are major metabolic targets of PPARγ agonism. The hypothesis to be tested was that, because PPARγ agonism normally reduces the impact of glucocorticoids in adipose tissue, the absence of glucocorticoids diminishes the influence of PPARγ agonism on adipose tissue morphology/function and consequently on whole body metabolic variables. Brown adipose tissue (BAT) was also studied because in rodents it is a major target of PPARγ agonism, which transforms BAT into a major site of lipid accumulation (8, 24, 28, 46). Because of the reciprocal modulatory interactions between PPARγ and glucocorticoid signaling on 11βHSD1 and the key influence of the enzyme on local Cort action in adipose tissue, the above determinations were complemented by assessment of treatment effects on 11βHSD1 expression levels. Glucocorticoids greatly affect whole body energy balance through favoring food intake and reducing energy expenditure (12, 50). To dissociate the consequences of glucocorticoid removal per se from those due to its reducing action on food intake and energy deposition, additional groups of intact rats were pair fed (PF) to the ADX animals.
MATERIALS AND METHODS
Animals and treatments.
Thirty-six male Wistar rats initially weighing 150–175 g were purchased from Charles River Laboratories (St. Constant, Quebec, Canada) and housed individually in stainless steel cages in a room kept at 23 ± 1°C with a 10:14-h light-dark cycle (lights on at 0700). The animals were cared for and handled in conformance with the Canadian Guide for the Care and Use of Laboratory Animals, and the protocols were approved by our institutional animal care committee. Rats had free access to tap water and a stock diet (Charles River Rodent Diet 5075, Ralston Products, Woodstock, Canada; digestible energy content, 12.9 kJ/g). Four days after their arrival, one-third of the animals underwent ADX. The bilateral removal of the adrenals was achieved through two small lateral skin incisions performed under isoflurane anesthesia. The adrenals were pulled out through the incision by holding the periadrenal fat and were severed with scissors. After the excision surgery, incisions were appropriately sutured. Intact, sham-operated animals were handled in the same way as ADX animals except that the adrenals were not excised. All rats were given 0.15 M NaCl in lieu of water to drink throughout the experiment. Food intake of the ADX animals was monitored daily during the first 4 days after ADX. Food provided to one-half of the intact animals was then matched to the intake of the ADX rats (PF rats). To avoid intake of food in a single large meal, PF rats were given one-third of their ration at 0800 and the remaining two-thirds at 1700. One group in each of the intact, ADX, and PF cohorts was given the ground stock diet, whereas the other group was fed the stock diet containing the PPARγ agonist RSG (purchased as Avandia at a local pharmacy, 30 mg·kg−1·day−1) for the 2-wk treatment period. The amount of food provided to the PF groups was adjusted every other day to food intake of the corresponding ADX group. The amount of RSG was adjusted twice weekly to the average food consumption of each group so as to provide the same amount per unit body weight to all groups. On the day before last, both control and RSG-treated PF groups were fed at 2200 to minimize the duration of fasting before death. On the last day, food was removed at 0730, and rats were killed at the beginning of the afternoon. Because of the food intake adjustment period, food intake and weight gain were calculated from day 4 to day 14.
Serum and tissue sampling.
Rats were killed by decapitation, and blood was collected from the neck wound and centrifuged (1,500 g, 15 min at 4°C), and the separated serum was stored at −80°C until later biochemical measurements. Inguinal (IWAT) and retroperitoneal white adipose tissues (RWAT), taken as representative of subcutaneous (SF) and visceral fat (VF), respectively, as well as interscapular brown adipose tissue (BAT), were excised, weighed, and prepared for further analysis as described below. A sample of liver was immediately frozen and stored at −80°C for later quantification of triacylglycerol content in lipid extracts (19).
Serum glucose concentrations were measured by the glucose oxidase method with the Beckman glucose analyzer. Insulin and Cort were determined by RIA using reagent kits from Linco Research (St. Charles, MO) with rat insulin and Cort as standards, respectively. An index of fasting insulin resistance, consisting of the product of fasting serum glucose and insulin, was calculated according to the homeostasis model assessment of insulin resistance (HOMA-IR) described by Matthews et al. (33). Triacylglycerol concentrations in serum and liver lipid extracts were measured by an enzymatic method using a reagent kit from Roche Diagnostics (Montreal, Quebec, Canada) that allows correction for free glycerol.
Adipocyte morphology by light microscopy and DNA content.
IWAT, RWAT, and BAT samples were fixed in 0.1 mmol/l PBS (pH 7.3) containing 4% paraformaldehyde and embedded in paraffin. Thin sections were mounted on glass slides and dyed with hematoxylin/eosin. Digital images of tissue slices were captured using an Olympus BX60 microscope equipped with a Sony RT Slider Spot Camera (Camsen Group, Markham, ON, Canada) at a magnification of 10× (IWAT and RWAT) or 40× (BAT). Total DNA content of RWAT and BAT samples was determined with the DNeasy Tissue Purification kit from Qiagen (Mississauga, ON, Canada) according to manufacturer's instructions.
Adipose tissue RNA isolation and analysis of 11βHSD1 mRNA.
Total RNA was prepared from IWAT, RWAT, and BAT using the Trizol RNA extraction method. RNA concentration was estimated from absorbance at 260 nm, and RNA was reverse transcribed using the Expand reverse transcriptase (Roche Diagnostics, Laval, QC, Canada). The expression level of mRNA was quantitated using quantitative fluorescent real-time PCR (Corbett Research, New South Wales, Australia). Amplification and detection of target mRNA were performed with Platinum Taq polymerase and the intercalating dye SybrGreen I. The primers, designed using the Vector NTI program and synthesized by Invitrogen (Burlington, ON, Canada), were the following: for 11βHSD1, 5′-primer (5′-3′) AATGGGAGCCCATGTGGTATTG; 3′-primer (5′-3′) GCACAGAGTGGATATCATCGTGG; for L273431 (L27), 5′-primer (5′-3′) CTGCTCGCTGTCGAAATG; 3′-primer (5′-3′) CCTTGCGTTTCAGTGCTG. The levels of 11βHSD1 mRNA were normalized to the amount of L27 mRNA (a gene not affected by treatments) detected in each sample, and results are expressed as 11βHSD1/L27 mRNA.
Data are presented as means ± SE and were analyzed by a three × two factorial ANOVA. For convenience, the ADX and PF interventions and their control (intact, ad libitum fed) were grouped into a factor termed Cort status, and the main and interactive effects of the Cort status, with three levels (intact, ADX, intact PF), and drug treatment, with two levels (control, RSG), were analyzed. Some variables were log transformed before analysis to ensure homogeneity of variance. Pairwise differences between individual group means were analyzed by Fisher's protected least significant difference test. Differences were considered statistically significant at P < 0.05.
Treatment effects on food intake, body weight, food efficiency, and corticosteronemia are summarized in Table 1. As expected, ADX reduced food intake (−15%). The Cort status and RSG interacted with each other on this variable, as the RSG-induced increase (+10%) in food intake seen in intact rats was abolished in ADX and PF rats. In fact, RSG unexpectedly tended to reduce food intake in ADX rats and did so significantly in PF animals, as RSG-treated PF animals did not eat all of the restricted amount of food provided. Treatment effects on body weight gain were in accordance with and proportional to those on food intake, as indicated by their strong correlation (r = +0.87, P < 0.0001). Feeding efficiency was not significantly affected by RSG and was slightly reduced by ADX, particularly those treated with RSG, but was nevertheless strongly correlated with weight gain (r = +0.94, P < 0.0001). Serum Cort concentrations confirmed the adequacy of the ADX procedure. In non-ADX rats, serum Cort levels were quite variable within experimental groups. As expected (11, 31, 38), PF increased serum Cort, which was augmented slightly further by RSG but not significantly so when groups were compared pairwise.
ADX decreased IWAT weight 30% in untreated animals, whereas inguinal mass in PF rats remained identical to that of their counterparts fed ad libitum (Fig. 1A). The effect of RSG on inguinal mass strongly depended on food intake: in intact animals, RSG treatment resulted in a 54% increase in IWAT weight, which was totally abolished in ADX and PF animals. The Cort status affected RWAT (Fig. 1B) in a manner similar to IWAT, as the visceral depot was smaller in ADX than in intact and PF rats. RSG did not increase RWAT weight in intact animals, whereas it tended to decrease depot weight in ADX rats and did so significantly in intact PF rats. The combined weight of the two WAT depots (not shown), taken as a gross index of global WAT mass, was decreased 20% by ADX relative to intact, untreated animals, whereas PF had no effect. RSG increased the combined weight of the two WAT depots 42% in intact rats but remained without effect in ADX and PF rats. As depicted in Fig. 1C, BAT weight was unaffected by ADX or PF in untreated animals. RSG increased BAT weight sixfold in intact rats, whereas the increase was dampened in ADX (2.8-fold) and in PF rats (4-fold). Body weight gain was closely correlated with the sum of the two WAT depots (r = +0.84, P < 0.0001) but less so with BAT weight (r = +0.44, P < 0.008).
Treatment effects on the microscopic appearance of adipocytes are depicted in Fig. 2A. In untreated control animals, RWAT cell size appeared smaller in both ADX and PF groups compared with the intact group, but much less so in the PF group (Fig. 2A, left). In RSG-treated (Fig. 2A, right) intact animals, there appeared to be less of the largest adipocytes and relatively more smaller cells per unit area compared with untreated rats. Cell size was markedly decreased by RSG in ADX and PF groups, but less so in the latter. Essentially similar microscopic observations were made in IWAT (not shown) except that the difference in cell size in the intact, RSG-treated group relative to its untreated control was on average slightly less than that of the RWAT depot, reflecting its larger fat content. Total DNA content was quantified in RWAT and BAT (IWAT not available) as an estimate of changes in total cell number. Although total DNA includes that of nonadipocyte cells, adipocytes are the major target of the proliferative action of PPARγ agonists. Figure 2B shows that ADX, but not PF, decreased total DNA in RWAT compared with the intact group in both control and RSG-treated cohorts. RSG increased RWAT DNA in all groups, but the magnitude of the increase was smaller in ADX than in intact and PF groups. In absolute terms, the RSG-induced increase in RWAT DNA was approximately threefold larger in intact than in ADX animals. PF rats were statistically indistinguishable from intact rats and different from the ADX groups. Regarding BAT, its microscopic appearance was not notably altered by ADX or PF in untreated rats (Fig. 2C, left), confirming the weight data. RSG treatment (Fig. 2C, right) was associated with the appearance of larger lipid droplets, an effect that was somewhat dampened in ADX and PF rats. Observation at higher magnitudes (not shown) indicated that in intact rats some lipid droplets had begun to coalesce, leading to the appearance of monolocular fat cells. BAT DNA (Fig. 2D) displayed responses to treatment conditions that resembled those in RWAT, except that the RSG-induced DNA increase in PF rats was less pronounced than in the latter.
Given that glucocorticoids and PPARγ agonism exert divergent actions on 11βHSD1 expression, we next determined whether the Cort status and RSG interacted on its modulation. Expression levels of the enzyme were quite variable in animals not treated with RSG, but there was a tendency for ADX to reduce 11βHSD1 expression in both WAT depots (Table 2), which nearly reached significance in RWAT (P = 0.07). Such reduction was absent in PF rats, suggesting an effect of Cort independent of that on food intake. RSG robustly reduced the expression of 11βHSD1 independently of the Cort status in both WAT depots. In BAT, 11βHSD1 mRNA was not affected by the Cort status, but RSG exerted a significant global decreasing effect that was statistically independent of the Cort status.
Finally, treatment effects were determined on indexes of insulin sensitivity and lipemia, which are major metabolic targets of PPARγ agonism (Fig. 3). The Cort status exerted a slight but significant (P < 0.03) global effect on fasting serum glucose (not shown), which was entirely explained by a reduction in glycemia in PF rats treated with RSG (PF RSG 7.3 ± 0.2 vs. PF control 8.0 ± 0.2 mM, P < 0.05). RSG did not affect fasting glycemia in intact and ADX groups. Both ADX and PF decreased fasting insulin by ∼30% (Fig. 3A). RSG decreased insulinemia in all groups to approximately one-half of the levels found in untreated rats. The hypoinsulinemic effect of ADX or PF and that of RSG were fully additive. These treatment effects were reflected in the HOMA-IR index (Fig. 3B), which is inversely proportional to insulin sensitivity. Both ADX and PF decreased serum triacylglycerol concentrations (Fig. 3C). RSG treatment also decreased triacylglycerolemia in all groups, although less so in PF animals. As in the case of insulinemia, the effects of ADX or PF and those of RSG tended to be additive. The same pattern of response was found for liver triacylglycerol concentration (data not shown), the main source of fasting circulating triacylglycerols, for which treatment additivity was robust.
This study investigated the effects of a PPARγ agonist on adipose tissue accretion and cellularity, lipemia, and indexes of insulin sensitivity in rats deprived of Cort, with control for food intake. The findings indicate that effects of PPARγ agonism on adipose tissue such as cell proliferation and weight accretion are dampened but partially preserved in the absence of Cort and that its beneficial global metabolic actions are maintained, such effects being therefore independent of PPARγ-mediated modulation of glucocorticoids.
As expected (1, 30, 50), ADX reduced food intake by ∼15%. In contrast, RSG brought about a 10% increase in food intake of intact rats, consistent with earlier reports (8, 28). This hyperphagic effect of RSG was completely lost in ADX and PF animals. In fact, RSG tended to paradoxically decrease food intake in ADX rats and did so significantly in PF animals, which did not ingest all of the restricted amount of food provided. Although the mechanisms by which PPARγ agonism impacts food intake have not been systematically addressed to date, its orexigenic effect has been tentatively attributed to the downregulation of adipose tissue leptin expression (53, 58). However, ADX, PF, and PPARγ agonism have all been reported to reduce circulating leptin levels (1, 6, 15, 16, 21, 27, 47), which is not congruent with the divergent effect of RSG on food intake of intact vs. ADX and PF rats. The present findings instead suggest the possibility that RSG may exert direct or indirect central effects on ingestive behavior that appear to be dependent on food availability.
The ADX- and PF-induced changes in energy balance had the expected consequences on fat accretion in rats not treated with RSG. Whereas ADX decreased fat deposition in both IWAT and RWAT, particularly in the latter, PF did not influence final depot weights. This is because of the divergent impact of ADX and PF on sympathetically activated thermogenesis, which results in fat loss in ADX and fat sparing in PF animals (17).
RSG exerted robust effects on fat deposition that were highly depot specific and strongly dependent on the Cort status. In intact animals, the largest relative weight increase induced by RSG occurred in BAT (6-fold), followed by IWAT (∼50%), whereas RWAT weight remained unaffected. These findings and our previous study (28) suggest that PPARγ agonists favor fat accumulation mostly in BAT and subcutaneous WAT depots in the rat, rather than in visceral fat. In humans, PPARγ agonism also favors subcutaneous fat accretion and has been reported to decrease visceral fat mass (34, 35). Interestingly, fat accumulation in rodent BAT occurs in the face of a nonfunctional induction of uncoupling protein 1 expression, and the organ becomes a lipid-storing, rather than a thermogenic, tissue (46). Except for a slight decrease in ADX rats, food efficiency was not notably affected by RSG, indirectly suggesting a lack of effect on thermogenesis in the present conditions. Instead, lipid uptake and retention possibly constitute major means of BAT lipid accretion in view of the fact that fat accumulation is clearly present as early as 24 h after exposure to a PPARγ agonist (M. Laplante and Y. Deshaies, unpublished observations). The depot specificity of action of RSG was even more marked in ADX and PF animals. Indeed, whereas RSG no longer favored IWAT accretion in the latter groups, the drug actually reduced RWAT accretion, but partially maintained its weight-increasing action in BAT. This may explain the RSG-induced reduction in RWAT of these rats by virtue of a preferential substrate uptake and deposition by BAT under conditions of low energy availability. Therefore, an adequate supply of dietary energy is needed for PPARγ agonism to increase fat deposition in WAT, but such ability is retained in BAT under energy restriction, independently of the presence or absence of Cort. Finally, it should be noted that the effects of RSG on fat accretion in the various depots are fully restored in ADX rats receiving exogenous Cort (Berthiaume and Deshaies, unpublished observations).
Regarding adipose tissue cellularity, ADX is thought to bring about a decrease mostly in cell size rather than number (18), at least over the short time periods over which ADX is usually studied, whereas food restriction, obviously depending on severity and duration, has been reported to reduce mainly adipocyte size but also fat cell proliferation to some extent, at least in the WAT depots examined here (20). In rats not treated with RSG of the present study, the total DNA content of RWAT was only slightly reduced in ADX, and not at all in PF animals; no effect of the Cort status was observed in BAT. As expected (4, 41), RSG favored adipose cell proliferation in intact animals, as witnessed by the higher cell density per unit area in treated vs. untreated rats observed microscopically and confirmed by the higher DNA content of RWAT (3-fold) and BAT (6-fold). Although IWAT DNA was not determined here, PPARγ agonists are well known to induce cell proliferation also in subcutaneous fat (14, 39). Importantly, RSG retained its ability to induce cell proliferation in the absence of Cort. However, the DNA data indicate that the magnitude of the stimulation of cell proliferation elicited by RSG was clearly dampened in ADX rats. RSG increased DNA content in PF rats somewhat more than in ADX animals, particularly in RWAT, suggesting a facilitative role for Cort in PPARγ-mediated cell proliferation. The fact that in vitro Cort limits adipocyte proliferation, favoring differentiation instead (51), suggests the involvement of an alternate, Cort-controlled factor, which remains to be identified.
Because glucocorticoids and PPARγ agonism exert divergent actions on the expression of the 11βHSD1 gene, and given the importance of the enzyme in modulating glucocorticoid abundance in WAT, it was of particular interest to determine whether these pathways interacted on 11βHSD1 expression. The direct, food intake-independent upregulation of 11βHSD1 expression by Cort (22, 45, 49) tended to be confirmed in both WAT depots, as was its downregulation by RSG (4), which proved to be efficient independently of the Cort status. This provides novel evidence that under normal conditions the stimulatory action of Cort on 11βHSD1 expression does not counteract the strong downregulatory effect of PPARγ agonism. In BAT, the Cort status did not affect 11βHSD1 expression and the inhibitory effect of RSG was relatively less potent than in WAT, indicating that the gene may be under tissue-specific modulation. The precise impact of 11βHSD1 inhibition by PPARγ agonists on adipose lipid metabolism remains to be further characterized.
Although PPARγ actions on insulin sensitivity that are not mediated by those exerted on adipose tissue clearly cannot be excluded (25, 54), ample evidence supports the notion that PPARγ agonists improve insulin sensitivity mainly through adipose tissue remodeling, increased capacity for lipid uptake/retention, and altered adipocytokine secretion pattern (2, 3, 14, 41, 48, 57). Hence, in the A-ZIP/F-1 mouse model of lipoatrophy, PPARγ agonism remains able to improve skeletal muscle insulin sensitivity, albeit at the expense of a severe deterioration of that of the liver (25), such that whole body insulin action is no longer ameliorated (10). In line with this concept, it is not unreasonable to interpret the additive RSG-induced improvement in indexes of insulin sensitivity over that of ADX being due to the partial maintenance of the effects of RSG on adipose tissue metabolism. Regarding triacylglycerolemia, the mechanisms underlying its reduction by PPARγ agonists in rodent models (13, 23, 28) remain unclear but may involve actions at the level of both liver very low density lipoprotein-triacylglycerol secretion (9) and triacylglycerol hydrolysis by adipose tissue lipoprotein lipase (28). The present findings clearly demonstrate that PPARγ agonism has the capacity to bring about a reduction in triacylglycerolemia beyond that elicited by glucocorticoid removal.
In summary, the RSG-induced increase in IWAT weight seen in intact animals was lost in ADX and food-restricted rats, but that in BAT was partially retained. RSG also retained its ability to induce WAT and BAT proliferation in the absence of glucocorticoids, albeit with lesser absolute potency than in intact rats, indicating that in vivo, Cort per se or an alternate factor controlled by Cort facilitates PPARγ-induced adipocyte proliferation. Inasmuch as the beneficial metabolic effects of RSG are due to its modulation of adipose tissue metabolism, the partial preservation of RSG-induced WAT and BAT proliferation and BAT lipid-retaining capability in ADX and PF animals was associated with a robust expression of the insulin-sensitizing and hypolipidemic actions of RSG, which were fully additive to those of ADX, and therefore independent from PPARγ modulation of glucocorticoids.
PPARγ and glucocorticoids are transcriptional factors that play important roles in adipose tissue metabolism. Both PPARγ agonism and glucocorticoids (in the low physiological range) favor adipogenesis and anabolic lipid metabolism. Remarkably, however, despite these similarities, the stimulation of the PPARγ pathway leads to a vast improvement in the metabolic profile at the whole body level, whereas the opposite occurs when glucocorticoids are in excess. Such extreme differences undoubtedly derive from many causative factors. The depot specificity of action of PPARγ agonism (mostly subcutaneous WAT, and BAT in rodents) and glucocorticoids (mostly visceral WAT), specific actions of PPARγ agonism on pathways of lipid retention that glucocorticoids do not share or even antagonize (e.g., adipose phosphoenolpyruvate carboxykinase, which favors fatty acid reesterification), as well as the interactions between the two pathways (e.g., PPARγ-mediated inhibition of 11βHSD1), may all be part of such divergent actions. That the metabolic effects of RSG persist in ADX animals should not be interpreted as meaning that a reduction in local glucocorticoids through 11βHSD1 downregulation does not partake in the mechanisms of action of PPARγ agonism on adipose tissue metabolism, such involvement perhaps being more apparent in the presence of normal energy intake. The present study, however, clearly demonstrates that the potency of PPARγ agonism to improve the metabolic profile extends over and beyond, and is therefore partly independent from, the beneficial metabolic effects of a reduction in glucocorticoids.
This work was supported by a grant from the Canadian Institutes of Health Research (CIHR) to Y. Deshaies. M. Berthiaume was the recipient of a PhD Studentship award from CIHR-Laval University. A. Tchernof was the recipient of a Scholarship award from the CIHR.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society