SEARCH

SEARCH BY CITATION

Keywords:

  • caloric restriction;
  • cidea;
  • ghsr;
  • klf15;
  • ppara;
  • ucp1

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

Insulin sensitivity deteriorates with age, but mechanisms remain unclear. Age-related changes in the function of subcutaneous white adipose tissue (sWAT) are less characterized than those in visceral WAT. We hypothesized that metabolic alterations in sWAT, which in contrast to epididymal WAT, harbors a subpopulation of energy-dissipating UCP1+ brown adipocytes, promote age-dependent progression toward insulin resistance. Indeed, we show that a predominant consequence of aging in murine sWAT is loss of ‘browning’. sWAT from young mice is histologically similar to brown adipose tissue (multilocular, UCP1+), but becomes morphologically white by 12 months of age. Correspondingly, sWAT expression of ucp1 precipitously declines (~300-fold) between 3 and 12 months. Loss continues into old age (24 months) and is inversely correlated with the development of insulin resistance. Additional age-dependent changes in sWAT include lower expression of adbr3 and higher expression of maoa, suggesting reduced local adrenergic tone as a potential mechanism. Indeed, treatment with a β3-adrenergic agonist to compensate for reduced tone rescues the aged sWAT phenotype. Age-related changes in sWAT are not explained by the differences in body weight; mice subjected to 40% caloric restriction for 12 months are of body weight similar to 3-month-old ad lib fed mice, but display sWAT resembling that of age-matched ad lib fed mice (devoid of brown adipose-like morphology). Overall, findings identify the loss of ‘browning’ in sWAT as a new aging phenomenon and provide insight into the pathogenesis of age-associated metabolic disease by revealing novel molecular changes tied to systemic metabolic dysfunction.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

Insulin resistance increases during aging. Although obesity often leads to insulin resistance, many under-weight older individuals develop diabetes; hence, qualitative changes in adipose tissue, which has considerable function outside of energy storage, are potentially important in the development of age-associated diabetes.

Adipose tissue can be broadly categorized as white (WAT) or brown (BAT). WAT serves as the primary reservoir for lipid storage, but is also an active endocrine organ, with deposits under the skin (subcutaneous, sWAT) and around internal organs (visceral). In contrast, BAT primarily functions to produce heat and is found predominantly between the scapulae of mice and in the supraclavicular/thoracic regions of adult humans (Cypess et al., 2009). The adipocytes that comprise WAT and BAT display several distinguishing features. White adipocytes contain a single (‘unilocular’) large lipid droplet, few mitochondria, and little cytosol, whereas brown adipocytes have numerous smaller (‘multilocular’) lipid droplets, abundant mitochondria, and normal cytosolic volume. Brown adipocytes are also defined by high expression of uncoupling protein-1 (UCP1), a mitochondrial protein that provides these cells with the unique ability to dissipate large quantities of energy as heat (thermogenesis) (Cannon & Nedergaard, 2004).

Importantly, UCP1+ adipocytes are found interspersed within WAT. These brown-like adipocytes, often described using terms such as ‘brite’ (brown in white) (Petrovic et al., 2010) or ‘beige’ (Ishibashi & Seale, 2010), can be induced with adrenergic stimuli or exposure to the cold, reportedly via β-3 adrenergic receptor (β3AR)-mediated transdifferentiation of white adipocytes (Himms-Hagen et al., 2000; Barbatelli et al., 2010), although this remains a topic of debate (Petrovic et al., 2010). It has recently been reported that brown adipocytes residing within classic BAT depots share lineage with skeletal muscle, based on tracing studies that revealed a common precursor positive for the myogenic marker myogenic factor 5 (myf5). In contrast, brown adipocytes residing in WAT do not display similar evidence of myf5+ lineage (Timmons et al., 2007).

Accumulating evidence suggests that UCP1 is physiologically relevant outside of thermoregulation. Ucp1−/− mice display metabolic abnormalities that include impaired diet-induced thermogenesis (Feldmann et al., 2009) and increased susceptibility to diet-induced obesity with age (Kontani et al., 2005), whereas transgenic mice overexpressing ucp1 are leaner and more insulin sensitive than wild-type mice (Kozak & Anunciado-Koza, 2008). Similarly in humans, a polymorphism that reduces ucp1 expression is linked to obesity (Sramkova et al., 2007), and high expression of ucp1 specifically in sWAT is associated with improved glucose tolerance in obese individuals (Timmons & Pedersen, 2009). Ucp1 is thus an attractive therapeutic target, holding particular promise for the treatment of type 2 diabetes.

With advancing age, loss of BAT activity occurs in both rodents (Yamashita et al., 1999) and humans (Saito et al., 2009; Yoneshiro et al., 2011). Aging is also associated with increased production of pro-inflammatory signals in visceral WAT (Wu et al., 2007). However, little is known about consequences of aging specific to sWAT. The present studies demonstrate that a predominantly brown-like phenotype in sWAT from young adult mice is progressively lost during aging, and changes that occur later in life are linked to the development of age-related insulin resistance.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

Loss of ‘browning’ is a predominant effect of aging in murine sWAT

Sections of sWAT from six-week-old male C57Bl/6J mice stained with hematoxylin and eosin (H&E) reveal significant ‘brown-like’ morphology, with prominent pink cytosolic staining surrounding multilocular adipocytes. In contrast, sWAT from mice at 12 months of age displays morphology that is uniformly white (Fig. 1A). Consistent with the brown-like morphology, brown adipocyte–specific gene expression is abundant in young sWAT, with transcript levels of ucp1 and cell death-inducing DFFA-like factor-a (cidea) 133-fold and 3-fold lower, respectively, than that of interscapular brown adipose tissue (BAT) from the same animals (Fig. S1). UCP1 protein is also readily detected in sWAT from 6-week-old mice (Fig. 1B).

image

Figure 1. Brown-to-white transitioning in murine subcutaneous white adipose tissue (sWAT) with age. (A) Representative hematoxylin and eosin staining of sWAT from young (6-week-old) and old (1-year-old) mice. (B) Representative staining of sWAT sections from young mice with UCP1-specific antibodies. UCP1 antigen is stained green on the left, nuclei are counterstained blue with DAPI in the middle, and a merged image is on the right. Relative mRNA expression of (C) uncoupling proteins (left), cidea and cidec (middle), and oxidative markers (right) in sWAT from young and old mice (qPCR; endogenous control = tbp). *< 0.05, **< 0.01 ***< 0.001 vs. young, with fold-change values indicated. (D) Scatterplot illustrating the changes in adipogenic gene expression, as determined using targeted PCR arrays, in sWAT from young and old mice. Genes altered >2-fold (old vs. young) are above (increased) or below (decreased) the three diagonal lines; genes altered >4-fold are identified by name, with fold-change values in parentheses. (E) Corresponding expression of beta adrenergic receptors (top), maoa and th (middle), and dio2, thra, and thrb (bottom) in sWAT (qPCR; endogenous control = tbp). *< 0.05, **< 0.01, ***< 0.001 vs. young; n = 3–4.

Download figure to PowerPoint

However, by 12 mo of age, a marked reduction in ucp1 (276-fold) and cidea (42-fold) expression is observed in sWAT (Fig. 1C), but not in BAT (Fig. S2). In contrast, sWAT levels of nonbrown-specific gene family members are either unchanged (ucp2/3) or modestly induced (1.5-fold, cidec) with age (Fig. 1C). Consistent with a reduction in oxidative brown adipocytes, sWAT from older mice displays lower expression of brown-specific oxidative enzymes cytochrome c oxidase subunit VIIa polypeptide 1 (cox7a1, 17-fold, < 0.01) and cytochrome c oxidase subunit VIIIb (cox8b, 8-fold, < 0.001) (Fig. 1C).

Using PCR arrays to simultaneously quantify the expression of 84 genes related to white and/or brown adipogenesis in sWAT from old vs. young mice, we find that the most pronounced effect of aging is reduced expression of ucp1 (246-fold; Fig. 1D, Table 1). Additional genes significantly down-regulated with age include fatty acid synthase (fasn, 10-fold), peroxisome proliferator-activated receptor alpha (ppara, 8-fold), krüppel-like factor 15 (klf15, 8-fold), angiotensinogen (agt, 7-fold), adipsin (cfd, 5-fold), and deiodinase iodothyronine, type II (dio2, 4-fold), as well as adrenergic signaling modulators beta-2-adrenergic receptor (adrb2, 8-fold) and cAMP responsive element binding protein (creb1, 1.8-fold) (Fig. 1D, Table 1). Further, the levels of CCAT/enhancer binding protein, beta (cebpb), vitamin D receptor (vdr), early growth response 2 (egr2), tsc22 domain family, member 3 (tsc22d3), and sirtuin 2 (sirt2) are modestly reduced (2-3-fold, < 0.05) in sWAT from older animals (Fig. 1D, Table 1).

Table 1. Adipogenic gene expression in subcutaneous white adipose tissue is altered during aging
Up-regulatedDown-regulated
Gene symbolFold-change (old vs. young)P-valueGene symbolFold-change (old vs. young)P-value
  1. PCR arrays (Qiagen, #PAMM-049) were used to determine the expression of 84 ‘adipogenic’ genes in subcutaneous white adipose tissue isolated from mice at 6 weeks (young) or 1 year (old) of age. The genes significantly up-regulated (left) or down-regulated (right) in old vs. young mice (< 0.05) mice are listed; n = 3.

Lep27.600.036Ucp1−246.200.053
Sfrp543.110.001Fasn−9.460.019
Fgf16.600.037Ppara−8.350.051
Src1.600.035Adrb2−8.070.015
   Klf15−7.550.043
   Agt−7.450.049
   Cfd−4.980.028
   Dio2−4.210.019
   Retn−3.370.020
   Cebpb−2.490.017
   Vdr−2.290.001
   Egr2−1.990.041
   Tsc22d3−1.940.037
   Creb1−1.760.022
   Sirt2−1.620.001

Consistent with depot expansion, as well as brown-to-white transitioning, genes induced with age include leptin (lep, 28-fold), secreted frizzled-related protein 5 (sfrp5, 43-fold), and fibroblast growth factor 1 (fgf1, 7-fold) (Fig. 1D, Table 1). A comprehensive list of PCR array findings can be found in Table S1 (Supporting information), and qPCR confirmation of select genes produced very similar results (Fig. S3). Taken together, PCR array findings indicate the most pronounced adipogenic change during aging is a decline in the expression of genes that favor brown adipogenesis (Fig. 1D, Table 1).

As lower adrb2 expression in sWAT from older mice suggests that loss of browning with age could be due to decreased adrenergic tone, we quantified the expression of other beta adrenergic receptors in sWAT from young and old mice. Levels of adrb3, but not adrb1, similarly decline with age (4-fold; Fig. 1E). Additionally, we find that sWAT expression of monoamine oxidase a (maoa) is ~3-fold higher in the older animals (Fig. 1E). In contrast, tyrosine hydroxylase (th) levels are similar in sWAT from young and old mice (Fig. 1E), as are circulating concentrations of adrenaline and noradrenaline (Fig. S4).

Because sWAT expression of dio2 declines with age, we investigated how aging impacts other components of the thyroid axis. Interestingly, we observe 42% higher triiodothyronine (< 0.01) and 66% higher thyroxine (= 0.03) levels in plasma from the older mice (Fig. S4). Likewise, sWAT expression of thyroid hormone receptor b (thrb), but not thyroid hormone receptor a (thra), is significantly increased (2-fold) with age (Fig. 1E).

Additional loss of ucp1 expression in sWAT between 1 and 2 years of age is inversely related to the development of age-associated hyperglycemia

A further 15-fold reduction in sWAT ucp1 expression is evident from 1 to 2 years of age (< 0.05; Fig. 2A). Consistent with the development of age-associated insulin resistance, 2-year-old mice also display significantly increased blood glucose levels (Fig. 2B). Importantly, correlation analyses highlight a significant inverse association (Spearman's r = −0.83; = 0.015) between the loss of ucp1 in sWAT and increased blood glucose levels (Fig. 2C).

image

Figure 2. Loss of ucp1 expression in sWAT with age is inversely correlated with the development of age-associated hyperglycemia. (A) sWAT expression of ucp1 and (B) blood glucose levels in 1- and 2-year-old mice. *< 0.05; n = 4–8. (C) sWAT expression of ucp1 (log fold-change, y-axis) vs. circulating glucose concentrations (x-axis), in one- (circles) and 2-year-old animals (triangles).

Download figure to PowerPoint

Age-related loss of ucp1 expression is first evident at 4 months of age and specific to sWAT

To identify more precisely when brown-to-white transitioning begins, we studied mice at 6, 9, 12, 16, 20, or 26 weeks of age. As expected, body weight increases steadily from 6 to 26 weeks (Fig. 3A). sWAT expression of ucp1 is similar at 6, 9, and 12 weeks, but lower at 16 (8-fold) and 26 weeks (87-fold), compared to 6 weeks. In addition, cidea (17-fold), cox7a1 (10-fold), and cox8b (7-fold) levels are significantly lower at 26 weeks (Fig. 3B). Accordingly, H&E-stained sWAT sections reveal predominant brown adipose-like characteristics at 6 and 16 weeks, features consistent with white adipocytes becoming more frequent at 20 weeks, and white morphology that is largely exclusive by 26 weeks (Fig. 3C).

image

Figure 3. Age-related loss of browning in sWAT is first evident at four months of age. (A) Body weight and (B) sWAT expression of brown adipocyte-specific genes (qPCR; endogenous control = 36B4) in mice at 6, 9, 12, 16, 20, and 26 weeks of age. *< 0.05, **< 0.01 vs. 6 weeks, (one-way ANOVA); n = 4–5. (C) Representative hematoxylin and eosin staining of sWAT at 6, 16, 20, and 26 weeks.

Download figure to PowerPoint

From 6 to 20 weeks of age, sWAT accumulation is indistinguishable from epididymal WAT (eWAT), with eWAT outgrowing sWAT by 26 weeks (Fig. 4A). Consistent with this, production of white mediators (lep, srfp5) increases similarly with age in sWAT and eWAT (Fig. 4B). However, in contrast to sWAT, ucp1 levels in eWAT are very low and do not change as a function of age (Fig. 4C). As a result, the > 1000:1 ratio of ucp1 expression in sWAT/eWAT at 12 weeks decreases to 6:1 by 26 weeks (age × depot interaction ‘Pi’ = 0.006). Likewise, the depot differential for cidea is > 100:1 at 6 weeks, but diminishes to 6:1 by 26 weeks (Pi = 0.028), with similar expression patterns evident for other brown-specific genes cox7a1 and cox8b (Fig. 4C).

image

Figure 4. Subcutaneous (sWAT) and visceral (epididymal, eWAT) white adipose tissue depots grow similarly with age, but the loss of browning is specific to sWAT. sWAT (white bars) and eWAT (black bars) (A) depot weight (milligrams) and (B) expression of white-indicator genes at 6, 9, 12, 16, 20, and 26 weeks (qPCR; endogenous control = 36B4); corresponding expression of (C) brown-indicator or (D) brown regulatory genes, with age x depot interaction P-values (Pi) indicated. *< 0.05, **< 0.01, ***< 0.001 vs. 6 weeks; #< 0.05, ##< 0.01, ###< 0.001 vs. sWAT; n = 4–5.

Download figure to PowerPoint

As both sWAT and eWAT display similar overall growth, but only the former is undergoing brown-to-white transitioning, we compared age-dependent variation in adipogenic gene expression between these depots. Pro-brown factor klf15 displays an expression pattern most consistent with that of the brown-specific genes, with a significant age x depot interaction (Pi = 0.03). Levels are significantly higher in sWAT than in eWAT at 6, 9, and 12 weeks, and age-dependent loss is exclusive to sWAT at (26 vs. 6 weeks < 0.05) (Fig. 4D). Ppara expression also displays a significant age x depot interaction (Pi = 0.05), with lower concentrations in sWAT at 9 and 26 weeks (vs. 6 weeks), and significantly higher expression in sWAT than in eWAT at 6 and 12 weeks (Fig. 4D). In contrast, pparg2 expression is similar in sWAT and eWAT, with age-related induction evident at 20 and 26 weeks (vs. respective 6 weeks; Fig. 4D). Ppargc1a, notably more abundant in sWAT at 9 weeks only, does not vary with age in either depot. Finally, dio2 expression transiently increases in both depots at 9 weeks (vs. 6 weeks), although this is statistically significant only in sWAT (Fig. 4D).

Loss of ucp1 expression in sWAT with age is independent of adiposity

To address a potential relationship between age-related loss of sWAT browning and associated increases in adiposity, we next studied mice displaying varying degrees of age-associated weight gain. Mice lacking the growth hormone secretagogue receptor (ghsr −/−) are indistinguishable from ghsr +/+ mice at 6 weeks, but gain significantly less weight by one year of age (Fig. 5A). Consistent with less adiposity, lep (4-fold) and sfrp5 (2-fold) levels are significantly lower in sWAT from the older ghsr −/− mice, compared to age-matched ghsr +/+ littermates (Fig. 5B). Despite these differences in weight gain, age-induced loss of ucp1 in ghsr −/− mice is similar to that of ghsr +/+ mice (316- vs. 276-fold; age × genotype interaction Pi = 0.34), with analogous findings for cidea, cox7a1, and cox8b (Fig. 5C), suggesting weight gain is not a critical factor for age-dependent loss of brown adipocytes in sWAT.

image

Figure 5. Age-dependent loss of browning is evident when mice gain less weight while aging. (A–C) Young (6-week-old) and old (1-year-old) growth hormone secretagogue receptor (ghsr) −/− mice (black bars) are compared to ghsr +/+ (white bars). (A) Body weight and sWAT expression of (B) white-specific or (C) brown-specific genes (qPCR; endogenous control = tbp). *< 0.05, **< 0.01, ***< 0.001 vs. young; ###< 0.001 vs. ghsr +/+ (two-way ANOVA); n = 3–8. (D–G) Mice calorically restricted for 1 year (CR) are compared to age-matched ad libitum fed mice (AM) and weight-matched ad libitum fed 12-week-old mice (WM). (D) Body weight (left) and sWAT accumulation (right), and sWAT expression of (E) white-specific or (F) brown-specific genes (qPCR; endogenous control = 36B4). *< 0.05, **< 0.01, ***< 0.001 vs. WM; #< 0.05, ###< 0.001 vs. AM (two-way ANOVA); n = 4. Percent values indicate the degree to which the effect of aging (AM vs. WM) is attenuated with CR. (G) Representative hematoxylin and eosin staining of sWAT from WM, AM, and CR (arrows point to multilocular cells) mice.

Download figure to PowerPoint

We next used a restrictive feeding paradigm to more fully eliminate weight gain as a causative factor in age-related changes in sWAT. Mice fed a 40% calorically restricted (‘CR’) diet for one year (initiated at 10 weeks of age) were compared to ad libitum fed age-matched (‘AM’) littermates, as well as 12-wk-old nonrestricted mice with a mean body weight (BW) matching that of the CR group (weight-matched, ‘WM’). CR (22.7 g) and WM mice (22.7 g) are substantially smaller than AM mice (40.5 g), with ~75% less sWAT after normalizing for BW (Fig. 5D). Consistent with little change in adiposity, lep and sfrp5 induction is > 90% precluded in the CR mice (Fig. 5E). Conversely, we observe only a very slight attenuation (< 1%) of age-associated changes in ucp1 expression with CR (Fig. 5F), although the loss of cidea, cox7a1, and cox8b expression is partially (~50%) attenuated (Fig. 5F). Again utilizing adipogenesis-targeted PCR arrays, we find that CR also induces other genes we previously show decrease with age, including ppara, fasn, acacb, and cfd (< 0.05 vs. AM; Fig. S5).

Consistent with the ucp1 expression data, H&E staining indicates that sWAT from CR mice lacks the significant BAT-like areas we find characteristic of younger animals. Instead, CR tissue has little cytosolic staining and appears more similar to AM, but with smaller cells (Fig. 5G). However, in sWAT from CR mice only, we find occasional cells that appear multilocular (indicated by arrows in Fig. 5G). Similar multilocular cells are also observed in eWAT from CR mice (Fig. S6).

Older mice maintain the ability to induce brown adipocytes in response to adrenergic stimuli

Exogenous stimulation of the β3-adrenoreceptor (β3AR) is an appreciated means of inducing brown adipocytes within WAT (Himms-Hagen et al., 2000; Barbatelli et al., 2010). Because older mice exhibit substantially reduced brown-like characteristics, we wondered whether the ability to induce ucp1+ brown adipocytes in sWAT remains intact with age. To test whether older mice maintain the ability to produce ucp1+ brown adipocytes in response to adrenergic stimuli, we injected 13-mo-old mice with the highly selective β3AR agonist CL 316,243 (CL) once daily (0.1 mg kg−1) for seven days. Indeed, CL treatment elicits the appearance of multilocular adipocytes and UCP1 immunoreactivity in sWAT from older mice (Fig. 6A,B). In addition, compared to those receiving the saline vehicle, mice receiving CL display significantly increased ucp1, cidea, cox7a1, cox8b, dio2, and ppara mRNA expression in sWAT (Fig. 6C), with similar changes evident in eWAT (Fig. S7).

image

Figure 6. Exogenous β-3-adrenergic receptor activation induces brown adipocytes in sWAT of older mice. Old mice (13 months) were injected with vehicle (saline) or the β-3-adrenoreceptor agonist CL 316,243 (CL, 0.1 mg kg−1) once daily for 7 days. (A) Representative hematoxylin and eosin staining of sWAT from control (left) and CL-treated (right) mice. (B) Representative staining of sWAT from CL-treated mice with UCP1-specific antibodies; UCP1 antigen is stained green on the left, nuclei are counterstained blue with DAPI in the middle, and a merged image is on the right (with translucent light to illustrate morphology). (C) Corresponding sWAT expression of brown adipocyte–related genes in control (white bars) and CL-treated (black bars) mice (qPCR; endogenous control = 36B4). *< 0.05, **< 0.01 vs. Veh; n = 3–6.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

In the present study, we describe an inverse association between age and brown-like characteristics in WAT. Although it has been previously suggested that UCP1 expression is detectable only in WAT from very young mice (Rossmeisl et al., 2002), to our knowledge, we present the first detailed characterization of how aging affects the sWAT adipocyte phenotype. We demonstrate that with increasing age, there is a progressive loss of ucp1+ brown adipocytes in murine sWAT, but not eWAT or interscapular BAT.

C57Bl/6J mice have completed maturational growth and are ‘mature adults’ from 3 to 6 months of age, with the human equivalent of 20–30 years, and middle-aged from 10–14 months, corresponding to 38- to 47-year-old humans (Flurkey et al., 2007). We find that sWAT expression of ucp1 decreases nearly 300-fold between 3 and 12 months of age, followed by another 15-fold by 2 years of age. The pronounced changes we observe during early aging are important, as cellular events that link aging to associated metabolic disease are likely to begin prior to the development of systemic consequences. But additionally, and in further support of such a link, we observe a significant inverse relationship between the loss of ucp1 that occurs later in life and associated increases in circulating glucose levels. Although it remains to be determined whether the phenotypic changes we have identified in sWAT are causal to the development of age-associated insulin resistance, these correlative findings highlight the relevance of sWAT to age-related metabolic disease and provide an important foundation for subsequent mechanistic studies.

The complexities of brown adipogenesis are not well understood, but WAT contains a progenitor-rich stroma (Hauner, 2005), including those that can be induced in vitro to differentiate into fully functional brown adipocytes (Elabd et al., 2009). Although an age-induced loss of appropriate precursors would preclude brown adipogenesis, older mice maintain the ability to produce brown-like cells in response to β-3 agonism, arguing against this scenario. Unfortunately, whether brown adipocytes in sWAT arise from a unique progenitor or transdifferentiation of white adipocytes remains unclear, although the latter is supported for brown adipocytes induced in response to cold (Barbatelli et al., 2010). However, the source of basal brown adipocytes might differ from the ones that can be induced. In fact, an interesting question underscored by the present study is whether basal brown adipocytes present in young animals reflect a unique type of brown cell found within WAT or, instead, are inherently similar to the induced cells. A more detailed comparison of these two types of brown adipocytes within WAT is an important future study.

Promoter regions of ucp1 (Barbera et al., 2001) and cidea (Viswakarma et al., 2007) contain PPAR-responsive elements, and both PPARα and PPARγ (Nedergaard et al., 2005) can promote brown adipogenesis (Villarroya et al., 2007). However, we find that the expression of ppara, rather than pparg, declines in sWAT during aging, supporting a role of PPARα in age-dependent shifts in sWAT. In addition to ppara and ucp1, genes reduced in sWAT from older mice include cox7a1, cox8b, and dio2, which intriguingly are all regulated by adrenergic tone.

The sympathetic nervous system (SNS) acts through cell surface β-adrenergic receptors to modulate cAMP production, which then bind cAMP responsive elements within the ucp1 promoter [reviewed in (Richard & Picard, 2011)]. Although circulating catecholamine levels are unchanged, we observe attenuated sWAT expression of adrb2/3, as well as creb1, a downstream modulator of adrenergic signaling. Additionally, the expression of maoa, an enzyme that degrades endogenous catecholamines, is higher in sWAT from older mice. Collectively, these changes could promote relevant loss of adrenergic tone during aging.

Indeed, the sWAT aging phenotype is rescued by treating older mice with a β-3AR agonist. Thus, by administering an exogenous agonist that is unlikely to be degraded by maoa, reduced adrenergic tone can be overcome. It is interesting that pharmacological activation of β-3ARs induces browning in both sWAT and eWAT; however, this does not preclude endogenous depot differences that mediate selective WAT browning physiologically.

Acting synergistically with the SNS, the thyroid axis also modulates thermogenesis (Silva, 1995), with thyroid hormone responsive elements likewise present in the ucp1 promoter (Rabelo et al., 1995). Decreased expression of dio2 in older animals suggests an age-dependent decline in local concentrations of the active thyroid hormone triiodothyronine. Counterintuitively, we find that circulating levels of both triiodothyronine and thyroxine are increased in older animals, as are sWAT levels of thrb. Perhaps this reflects age-dependent development of peripheral thyroid hormone resistance, which has been hypothesized to occur in humans (Mooradian, 2008). While intriguing, determining whether such resistance plays a causal role in age-dependent shifts in the phenotype of sWAT presently awaits future study.

A number of genes significantly reduced in sWAT with age are PPARγ-target genes, including fasn, klf15, agt, and cfd. However, transcript levels of peroxisome proliferator-activated receptor gamma 2 (pparg2) do not vary with age. Reduced expression of these genes could be explained by altered interactions of PPARγ with transactivating partners. Indeed, over 300 cofactors have been identified that can associate with PPARγ and modulate gene transactivation, and many have been implicated in brown adipogenesis. In eWAT, total PPARγ levels are maintained upon aging, but the expression of an important cofactor, ncoa1 (src-1), decreases with age, resulting in impaired cofactor recruitment to PPARγ and dissociation of adipogenic and insulin-sensitizing gene regulation (Miard et al., 2009). However, in sWAT, we find an increase, rather than decrease, in src expression with age. PPARγ-induced ucp1 expression is also modulated by interactions with PGC1α (Puigserver et al., 1998), and exogenously introducing PGC1a causes human white adipocytes to develop brown-like characteristics (Tiraby et al., 2003). However, we do not observe a significant decrease in pgc1a expression in sWAT during aging. Therefore, while we cannot preclude a role of PPARγ in the loss of sWAT browning with age, our data are more supportive of a role for PPARα.

Recently, members of the krüppel-like zinc-finger (KLF) family of transcription factors have been highlighted as important regulators of adipogenesis. Of particular relevance, klf15 can activate the transcription of ucp1 by directly interacting with the promoter (Yamamoto et al., 2010). Consistent with this, we find substantially down-regulated expression of klf15, but not klf3 or klf4, in sWAT from older mice. In fact, klf15 expression patterning is collectively quite analogous to that of ucp1, supporting a regulatory role for klf15, with expression significantly higher in sWAT than in eWAT at 6, 9, and 12 weeks and not significantly modulated by CR. However, others have found that overexpression of klf15 in white adipose tissue does not increase the levels of ucp1, although increased oxygen consumption in adipocytes was observed (Nagare et al., 2011). It is possible that klf15 can modulate an oxidative program in adipocytes, but additional factors are ultimately required for the full brown phenotype. Indeed, it has been suggested that small changes in the levels of different transcriptional components of the ucp1 enhanceosome interact synergistically to achieve large differences in ucp1 expression (Xue et al., 2005). Thus, the various changes in gene expression we observe in sWAT during aging might collectively be a strong enough impetus to cause the dramatic decline in ucp1 expression. Further studies will be necessary to fully elucidate molecular mechanisms that promote age-dependent shifts in the phenotype of sWAT.

Interestingly, we find that age-related loss of ucp1 expression in sWAT does not require any associated weight gain. It is peculiar that chronic CR, which is known to slow aging, does not more substantially attenuate the loss of ucp1. However, expression is slightly increased in CR vs. AM mice, which could reflect a slower rate of aging, and if so, perhaps analysis performed after a shorter duration of restrictive feeding would reveal a more pronounced effect. It is also possible that other mechanisms inhibit the expression of ucp1 in the CR mice, as this protein promotes an energetically inefficient metabolism, which might be unfavorable when survival requires an exquisitely efficient use of fuel.

Despite low ucp1 expression, we observe a few multilocular cells in sWAT from CR mice. Others have reported the appearance of similar infrequent multilocular cells in eWAT of mice subjected to long-term CR and suggested that these cells could reflect a brown/white intermediate (Higami et al., 2004). Indeed, in contrast to ucp1, changes in the genes less exclusive to brown adipocytes (cidea, cox7a1, cox8b) are considerably attenuated with CR. Moreover, the frequency of these ‘intermediate’ adipocytes is similar in sWAT and eWAT, further supporting a flexible phenotype of white adipocytes. Finally, sWAT from CR mice displays increased expression of the regulatory factor ppara, as well as target genes fasn, acacb and srebf1. Although preferentially expressed in brown adipocytes, ppara is also found in white adipocytes and is required for oxidative gene induction in murine sWAT during white-to-brown transitioning in response to the cold (Li et al., 2005).

In summary, we demonstrate brown-to-white transitioning in murine sWAT during aging. Collectively, our data support the loss of adrenergic tone as a potential mechanism, as evidenced by age-dependently reduced expression of adrenergic receptors and increased local production of maoa in sWAT, and reversal of the phenotype by administering exogenous β3AR agonist. This agonist is also an appreciated insulin sensitizer (Bloom et al., 1992; de Souza et al., 1997), but attributing a theoretical benefit in an aging setting to sWAT is complicated by the fact that β3ARs are present in other metabolic tissues, including BAT (Arch, 1989) and skeletal muscle (Chamberlain et al., 1999). Alternatively, our data suggest that therapeutic strategies specifically targeting the browning of sWAT might be useful in treating metabolic diseases of aging, as we observe a negative correlation between the loss of ucp1 specifically in sWAT and the development of age-associated insulin resistance.

Supporting the relevance of our findings to humans, ucp1 expression in sWAT from obese individuals is higher in those without diabetes, which intriguingly suggests that ucp1 expression in human sWAT may promote metabolic homeostasis (Timmons & Pedersen, 2009). In future studies, it would be interesting to determine whether ucp1 expression in sWAT from elderly individuals is similarly correlated with insulin sensitivity. Pharmaceutically increasing UCP1 activity presently holds promise for treating obesity-associated metabolic disease and, based on the current findings, perhaps also metabolic diseases of aging.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

Animals

All experiments were approved by the Scripps Florida Institutional Animal Care and Use Committee. Male C57Bl/6J mice were housed in AAALAC-approved facilities maintained at 23 °C with 12-h light/dark cycles and, unless otherwise noted, given free access to water and standard chow. Blood glucose was measured using a glucometer (OneTouchUltra, LifeScan Inc., Milpitas, CA, USA). After sacrifice with CO2, blood was taken via cardiac puncture and bilateral (lower quadrant) inguinal subcutaneous white adipose tissue (lymph nodes removed), epididymal white adipose tissue, and interscapular brown adipose tissue (visible white fat trimmed away) were carefully collected. Ghsr −/− mice were generated as previously described (Sun et al., 2008).

Real-time RT–PCR

Real-time quantitative PCR was performed (7900HT, Applied Biosystems, Foster City, CA, USA) using commercially available Taqman assays (Applied Biosystems). Fold-changes were calculated as 2ΔΔCT with 36B4 and/or TATA-box binding protein (tbp) used as endogenous controls.

PCR arrays

Four hundred nanograms of total RNA was used to perform ‘adipogenesis’ (#PAMM-049) PCR arrays (Qiagen, Valencia, CA, USA) according to manufacturer's instructions without modification.

Histological analyses

Excised sWAT was fixed overnight with Z-fix (Anatech LTD, Battle Creek, MI, USA). The Scripps Research Institute histology core embedded samples in paraffin, sectioned the (3-µm) tissue, and performed H&E staining. For immunostaining, sections were blocked with 5% donkey serum in 0.2% Triton-X–PBS (1 h) and incubated with rabbit polyclonal UCP1 antibody (Abcam, Cambridge, MA, USA) overnight at 4 °C (1:100), followed by DyLight 649-conjugated donkey anti-rabbit IgG (Jackson Immunoresearch, West Grove, PA, USA) for 1 h (1:250). Hard-set medium with DAPI (Vector Laboratories Inc., Burlingame, CA, USA) was used to apply coverslips before imaging slides with laser scanning confocal microscopy (Olympus Fluoview FV1000, Center Valley, PA, USA).

Statistical analyses

All data are presented as means ± standard error of the mean. T-tests or anova (Newman–Keuls procedure for post hoc testing) was used as appropriate with  0.05 considered significant (GraphPad Prism v5, La Jolla, CA, USA).

Acknowledgments

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

We thank the Scripps ARC for excellent animal care and Donny Strosberg, PhD, for kindly providing CL 316,243. We gratefully acknowledge funding support by the NIA: R01AG29740 and R01AG19230.

Author contributions

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

NR performed the experiments, with assistance from AL and SP. NR and RS designed the experiments and wrote the manuscript.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information
  • Arch JR (1989) The brown adipocyte beta-adrenoceptor. Proc. Nutr. Soc. 48, 215223.
  • Barbatelli G, Murano I, Madsen L, Hao Q, Jimenez M, Kristiansen K, Giacobino JP, De Matteis R, Cinti S (2010) The emergence of cold-induced brown adipocytes in mouse white fat depots is determined predominantly by white to brown adipocyte transdifferentiation. Am. J. Physiol. Endocrinol. Metab. 298, E1244E1253.
  • Barbera MJ, Schluter A, Pedraza N, Iglesias R, Villarroya F, Giralt M (2001) Peroxisome proliferator-activated receptor alpha activates transcription of the brown fat uncoupling protein-1 gene. A link between regulation of the thermogenic and lipid oxidation pathways in the brown fat cell. J. Biol. Chem. 276, 14861493.
  • Bloom JD, Dutia MD, Johnson BD, Wissner A, Burns MG, Largis EE, Dolan JA, Claus TH (1992) Disodium (R, R)-5-[2-[[2-(3-chlorophenyl)-2-hydroxyethyl]-amino] propyl]-1,3-benzodioxole-2,2-dicarboxylate (CL 316,243). A potent beta-adrenergic agonist virtually specific for beta 3 receptors. A promising antidiabetic and antiobesity agent. J. Med. Chem. 35, 30813084.
  • Cannon B, Nedergaard J (2004) Brown adipose tissue: function and physiological significance. Physiol. Rev. 84, 277359.
  • Chamberlain PD, Jennings KH, Paul F, Cordell J, Berry A, Holmes SD, Park J, Chambers J, Sennitt MV, Stock MJ, et al. (1999) The tissue distribution of the human beta3-adrenoceptor studied using a monoclonal antibody: direct evidence of the beta3-adrenoceptor in human adipose tissue, atrium and skeletal muscle. Int. J. Obes. Relat. Metab. Disord. 23, 10571065.
  • Cypess AM, Lehman S, Williams G, Tal I, Rodman D, Goldfine AB, Kuo FC, Palmer EL, Tseng YH, Doria A, et al. (2009) Identification and importance of brown adipose tissue in adult humans. N. Engl. J. Med. 360, 15091517.
  • Elabd C, Chiellini C, Carmona M, Galitzky J, Cochet O, Petersen R, Penicaud L, Kristiansen K, Bouloumie A, Casteilla L, et al. (2009) Human multipotent adipose-derived stem cells differentiate into functional brown adipocytes. Stem Cells 27, 27532760.
  • Feldmann HM, Golozoubova V, Cannon B, Nedergaard J (2009) UCP1 ablation induces obesity and abolishes diet-induced thermogenesis in mice exempt from thermal stress by living at thermoneutrality. Cell Metab. 9, 203209.
  • Flurkey KCJ, Harrison DE (2007) The mouse in aging research. In The Mouse in Biomedical Research, B.S. (Fox JG, Davisson MT, Newcomer CE, Quimby FW, Smith AL, eds). Burlington, MA: Elsevier Academic Press, pp. 637672.
  • Hauner H (2005) Secretory factors from human adipose tissue and their functional role. Proc. Nutr. Soc. 64, 163169.
  • Higami Y, Pugh TD, Page GP, Allison DB, Prolla TA, Weindruch R (2004) Adipose tissue energy metabolism: altered gene expression profile of mice subjected to long-term caloric restriction. FASEB J. 18, 415417.
  • Himms-Hagen J, Melnyk A, Zingaretti MC, Ceresi E, Barbatelli G, Cinti S (2000) Multilocular fat cells in WAT of CL-316243-treated rats derive directly from white adipocytes. Am. J. Physiol. Cell Physiol. 279, C670C681.
  • Ishibashi J, Seale P (2010) Medicine. Beige can be slimming. Science 328, 11131114.
  • Kontani Y, Wang Y, Kimura K, Inokuma KI, Saito M, Suzuki-Miura T, Wang Z, Sato Y, Mori N, Yamashita H (2005) UCP1 deficiency increases susceptibility to diet-induced obesity with age. Aging Cell 4, 147155.
  • Kozak LP, Anunciado-Koza R (2008) UCP1: its involvement and utility in obesity. Int. J. Obes. (Lond) 32(Suppl. 7), S32S38.
  • Li P, Zhu Z, Lu Y, Granneman JG (2005) Metabolic and cellular plasticity in white adipose tissue II: role of peroxisome proliferator-activated receptor-alpha. Am. J. Physiol. Endocrinol. Metab. 289, E617E626.
  • Miard S, Dombrowski L, Carter S, Boivin L, Picard F (2009) Aging alters PPARgamma in rodent and human adipose tissue by modulating the balance in steroid receptor coactivator-1. Aging Cell 8, 449459.
  • Mooradian AD (2008) Asymptomatic hyperthyroidism in older adults: is it a distinct clinical and laboratory entity? Drugs Aging 25, 371380.
  • Nagare T, Sakaue H, Matsumoto M, Cao Y, Inagaki K, Sakai M, Takashima Y, Nakamura K, Mori T, Okada Y, et al. (2011) Overexpression of KLF15 transcription factor in adipocytes of mice results in down-regulation of SCD1 protein expression in adipocytes and consequent enhancement of glucose-induced insulin secretion. J. Biol. Chem. 286, 3745837469.
  • Nedergaard J, Petrovic N, Lindgren EM, Jacobsson A, Cannon B (2005) PPARgamma in the control of brown adipocyte differentiation. Biochim. Biophys. Acta 1740, 293304.
  • Petrovic N, Walden TB, Shabalina IG, Timmons JA, Cannon B, Nedergaard J (2010) Chronic peroxisome proliferator-activated receptor gamma (PPARgamma) activation of epididymally derived white adipocyte cultures reveals a population of thermogenically competent, UCP1-containing adipocytes molecularly distinct from classic brown adipocytes. J. Biol. Chem. 285, 71537164.
  • Puigserver P, Wu Z, Park CW, Graves R, Wright M, Spiegelman BM (1998) A cold-inducible coactivator of nuclear receptors linked to adaptive thermogenesis. Cell 92, 829839.
  • Rabelo R, Schifman A, Rubio A, Sheng X, Silva JE (1995) Delineation of thyroid hormone-responsive sequences within a critical enhancer in the rat uncoupling protein gene. Endocrinology 136, 10031013.
  • Richard D, Picard F (2011) Brown fat biology and thermogenesis. Front. Biosci. 16, 12331260.
  • Rossmeisl M, Barbatelli G, Flachs P, Brauner P, Zingaretti MC, Marelli M, Janovska P, Horakova M, Syrovy I, Cinti S, et al. (2002) Expression of the uncoupling protein 1 from the aP2 gene promoter stimulates mitochondrial biogenesis in unilocular adipocytes in vivo. Eur. J. Biochem. 269, 1928.
  • Saito M, Okamatsu-Ogura Y, Matsushita M, Watanabe K, Yoneshiro T, Nio-Kobayashi J, Iwanaga T, Miyagawa M, Kameya T, Nakada K, et al. (2009) High incidence of metabolically active brown adipose tissue in healthy adult humans: effects of cold exposure and adiposity. Diabetes 58, 15261531.
  • Silva JE (1995) Thyroid hormone control of thermogenesis and energy balance. Thyroid 5, 481492.
  • de Souza CJ, Hirshman MF, Horton ES (1997) CL-316,243, a beta3-specific adrenoceptor agonist, enhances insulin-stimulated glucose disposal in nonobese rats. Diabetes 46, 12571263.
  • Sramkova D, Krejbichova S, Vcelak J, Vankova M, Samalikova P, Hill M, Kvasnickova H, Dvorakova K, Vondra K, Hainer V, et al. (2007) The UCP1 gene polymorphism A-3826G in relation to DM2 and body composition in Czech population. Exp. Clin. Endocrinol. Diabetes 115, 303307.
  • Sun Y, Butte NF, Garcia JM, Smith RG (2008) Characterization of adult ghrelin and ghrelin receptor knockout mice under positive and negative energy balance. Endocrinology 149, 843850.
  • Timmons JA, Pedersen BK (2009) The importance of brown adipose tissue. N. Engl. J. Med. 361, 415416; author reply 418–421.
  • Timmons JA, Wennmalm K, Larsson O, Walden TB, Lassmann T, Petrovic N, Hamilton DL, Gimeno RE, Wahlestedt C, Baar K, et al. (2007) Myogenic gene expression signature establishes that brown and white adipocytes originate from distinct cell lineages. Proc. Natl. Acad. Sci. U.S.A. 104, 44014406.
  • Tiraby C, Tavernier G, Lefort C, Larrouy D, Bouillaud F, Ricquier D, Langin D (2003) Acquirement of brown fat cell features by human white adipocytes. J. Biol. Chem. 278, 3337033376.
  • Villarroya F, Iglesias R, Giralt M (2007) PPARs in the control of uncoupling proteins gene expression. PPAR Res. 2007, 74364.
  • Viswakarma N, YuS NaikS, Kashireddy P, Matsumoto K, Sarkar J, Surapureddi S, Jia Y, Rao MS, Reddy JK (2007) Transcriptional regulation of Cidea, mitochondrial cell death-inducing DNA fragmentation factor alpha-like effector A, in mouse liver by peroxisome proliferator-activated receptor alpha and gamma. J. Biol. Chem. 282, 1861318624.
  • Wu D, Ren Z, Pae M, Guo W, Cui X, Merrill AH, Meydani SN (2007) Aging up-regulates expression of inflammatory mediators in mouse adipose tissue. J. Immunol. 179, 48294839.
  • Xue B, Coulter A, Rim JS, Koza RA, Kozak LP (2005) Transcriptional synergy and the regulation of Ucp1 during brown adipocyte induction in white fat depots. Mol. Cell Biol. 25, 83118322.
  • Yamamoto K, Sakaguchi M, Medina RJ, Niida A, Sakaguchi Y, Miyazaki M, Kataoka K, Huh NH (2010) Transcriptional regulation of a brown adipocyte-specific gene, UCP1, by KLF11 and KLF15. Biochem. Biophys. Res. Commun. 400, 175180.
  • Yamashita H, Sato Y, Mori N (1999) Difference in induction of uncoupling protein genes in adipose tissues between young and old rats during cold exposure. FEBS Lett. 458, 157161.
  • Yoneshiro T, Aita S, Matsushita M, Okamatsu-Ogura Y, Kameya T, Kawai Y, Miyagawa M, Tsujisaki M, Saito M (2011) Age-related decrease in cold-activated brown adipose tissue and accumulation of body fat in healthy humans. Obesity (Silver Spring) 19, 17551760.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
acel12010-sup-0001-DataS1.docxWord document128KData S1 Experimental procedures.
acel12010-sup-0002-TableS1-FigS1-S7.pdfapplication/PDF4148K

Table S1 Adipogenesis PCR-array results.

Fig. S1 Ucp1 and cidea expression (relative to the endogenous control 36B4) in interscapular brown adipose tissue (BAT) or subcutaneous white adipose tissue (sWAT) isolated from 5 week old male mice. N = 3.

Fig. S2 Ucp1 and cidea expression (relative to endogenous control gene 36B4) in interscapular brown adipose tissue isolated from young (5 week) or old (11 month) male mice. N = 6–7.

Fig. S3 QPCR-confirmation of PCR array results for ppara, pparg2, and sfrp5 expression in subcutaneous white adipose tissue isolated from young (6 weeks) and old (1 year) male mice (relative to the endogenous control tbp; n = 3–4).

Fig. S4 (A) Plasma concentrations of adrenaline (white bars) and noradrenaline (black bars) are similar in young (11 weeks, gray bars) and old (11–12 months), black bars) mice. N = 4–6. (B–C) Plasma concentrations of triiodothyronine (B) and thyroxine (C) are elevated in old (12 months) vs. young (6 weeks) mice. N = 3–11; *< 0.05, **< 0.01.

Fig. S5 Volcano plot illustrating adipogenic gene expression in subcutaneous white adipose tissue (sWAT) in mice calorically restricted (CR) vs. age-matched (AM) ad libitum fed mice. Fold changes (CR vs. AM) are plotted on the x-axis, with genes altered >2-fold either to the left (decreased) or right (increased) of the gray vertical lines, and P-values are indicated on the y-axis, with significantly altered genes above the horizontal line indicating = 0.05.

Fig. S6 Representative hematoxylin and eosin staining of epididymal white adipose tissue from mice calorically restricted for 1 year, with arrows pointing to multilocular cells.

Fig. S7 Brown-adipocyte related gene expression (relative to the endogenous control 36B4) in epididymal white adipose tissue isolated from 12–13 month old mice treated with vehicle (white bars) or β-3-adrenoreceptor agonist CL 316,243 (CL, black bars) for 7 days. *< 0.05, ***< 0.001 vs. veh.; n = 2–5.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.