Aged Male Mice Remain Glucose Tolerant Despite Increased Energy Storage Efficiency Favoring Diet‐Induced Obesity
Liz Gray, Kaylynn Vidmar, Marita Rivir, Vishnupriya J. Borra, Diego Perez‐Tilve
TL;DR
Older male mice can store more fat without developing glucose intolerance, but this makes them more prone to obesity.
Contribution
The study reveals that aged mice have increased fat storage efficiency and reduced lipid turnover, which protects against glucose intolerance but increases obesity risk.
Findings
Old mice on a high-fat diet gained more fat mass but remained glucose tolerant.
Old mice showed reduced lipid turnover in visceral fat and increased plasticity in subcutaneous fat.
Increased fat storage efficiency in old mice did not protect against weight loss during calorie restriction.
Abstract
Obesity and aging are converging health challenges, contributing to morbidity in older populations. However, the specific contribution of age to susceptibility to obesity is unclear. This study examined the impact of age on susceptibility to diet‐induced obesity (DIO) and calorie restriction (CR) in male mice. Young (2–3 months) and old (17–24 months) lean C57BL/6J male mice were fed a standard chow diet (CD) or a high‐fat diet (HFD) for 28 days, then underwent 18 days of CR. We monitored body weight, body composition, energy intake and expenditure, glucose tolerance, and gene expression in metabolically relevant tissues. HFD‐fed old mice exhibited more fat mass gain but, surprisingly, protection from glucose intolerance. In comparison, young controls exhibited resistance to DIO due to reduced calorie storage efficiency. Gene expression analysis suggested reduced plasticity and lipid…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
FIGURE 1
FIGURE 2
FIGURE 3
FIGURE 4
FIGURE 5- —College of Medicine, University of Cincinnati10.13039/100014452
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAdipose Tissue and Metabolism · Regulation of Appetite and Obesity · Genetics, Aging, and Longevity in Model Organisms
Introduction
1
Over the past five decades, advances in health policies and living standards have led to a rise in life expectancy (World Health 2025) and a decline in birth rates which is expected to persist in the next decades (Fertility and Forecasting 2024). This demographic shift highlights the need for maintaining health among older individuals, as they may need to stay in the workforce or depend on social support systems for a longer period of time. Notably, the decades‐old overall increase in lifespan has been undermined by the rising mortality associated with obesity (Dowd et al. 2024; Preston et al. 2018). Hence, investigating the interaction between age and obesity may therefore reveal mechanistic insights critical for developing strategies to mitigate their combined impact on lifespan.
Obesity presents a major challenge to overall health and places a substantial burden on healthcare systems worldwide (Nagi et al. 2024). Rates of obesity are rising across age groups, trend that is projected to persist in the coming decades (G. B. D. Adult BMI Collaborators 2025a; G. B. D. Adolescent BMI Collaborators 2025b). Importantly, recent studies reveal that young adults with obesity show elevated markers of aging, implying that sustained excess body fat in early adulthood may accelerate biological aging (Correa‐Burrows et al. 2025; Nunan et al. 2022; Ruperez et al. 2025). Furthermore, aging itself is believed to contribute to increased fat accumulation (Kuk et al. 2009), and mid‐life obesity has been linked to a higher risk of developing sarcopenic obesity in later years [4], a condition associated with greater mortality (Atkins et al. 2014; Batsis et al. 2014). Although early life obesity tends to contribute to obesity in adulthood (Whitaker et al. 1997), a large segment of the population without a history of childhood obesity becomes obese later in life (Simmonds et al. 2016). Importantly, development of obesity in older adults significantly increases mortality in the elderly (Berry et al. 2022), highlighting the importance of maintaining a healthy body weight over time.
Although preexisting obesity contributes to the association between aging and increased fat mass (Kuk et al. 2009; Palmer and Jensen 2022), the extent to which age itself increases the susceptibility to obesity is not entirely understood. A decrease in physical activity and overall reductions in energy expenditure throughout the last decades of life in humans (Pontzer et al. 2021) is consistent with an increase in the susceptibility to weight gain. Interestingly, the contribution of this reduction in expenditure may be offset by the decrease in appetite exhibited with age (Dericioglu et al. 2024; Giezenaar et al. 2016). On the other hand, emerging evidence of age‐specific adipocyte proliferation is consistent with facilitation of adipose tissue expansion favoring susceptibility to obesity in older individuals (Wang et al. 2025).
We performed studies to determine the impact of age on susceptibility to DIO and glycemic control of old male C57/Bl6 mice following a 4‐week HFD feeding. We measured body composition, glucose tolerance and the relative contribution of energy intake and expenditure to energy balance. In addition, we compared the efficacy of CR for weight loss between aged and young mice. Our data revealed increased susceptibility to DIO in old mice due to increased calorie storage efficiency but a remarkably preservation of glucose tolerance. In contrast, they exhibit increased susceptibility to body weight loss following CR, suggesting an increased vulnerability to withstand periods of negative energy balance.
Materials and Methods
2
Mice and Diets
2.1
All studies were approved by the Institutional Animal Care and Use Committee of the University of Cincinnati in accordance with the US National Institutes of Health Guide for the Care and Use of Laboratory Animals. Mice were housed in an AAALAC‐approved room with a 12‐h light and 12‐h dark cycle held at 22°C, and with free access to food and water.
2–3‐month‐old (Young) and 17–24‐month‐old (Old) male C57BL/6J mice (Cat. Number #000664; Jackson Laboratory, Bar Harbor, ME, USA) were fed either pelleted chow diet (CD) (5053 Rodent Diet 20; PicoLab; 3.02 kcal/g of metabolizable energy; 24.4/13.1/62.4% from protein/fat/carbohydrates) or high‐fat diet (HFD) (D12331; Research Diets Inc., New Brunswick, NJ, USA; 5.56 kcal/g of metabolizable energy; 16.4/58/25.5% from protein/fat/carbohydrates) as indicated. Body weight (BW) and food intake (FI) measurements were taken daily by scale. Mice were group housed, mostly as 4 per cage, to prevent age‐dependent differences in food spillage seen with singly‐housed mice (Starr and Saito 2012). Daily FI was calculated by dividing the total number of grams eaten per cage by the number of mice in that cage to obtain an average FI per mouse.
Calorie Restriction
2.2
Lean Young and Old mice, always maintained on CD, received either CD or HFD ad libitum for 28 days. At the end of this period, the daily food intake of half of the mice from each group was limited to achieve 35%–38% calorie restriction (CR) after 18 days. CR mice received fresh food daily, after accounting for any remaining food from the previous day. CR mice received a single meal during the light phase, approximately 5 h prior to the onset of the dark cycle during the duration of the CR period.
Glucose Tolerance Test
2.3
Mice were fasted for 4–6 h and then injected intraperitoneally (IP) with glucose (2 g/kg, 20% wt/vol d‐glucose [Sigma‐Aldrich] in phosphate‐buffered solution). Blood glucose (BG) measurements were taken from tails directly prior to injection (0 min) and then at 15, 30, 60, 90, and 120 min post‐injection. A handheld glucometer and corresponding test strips were used to take BG measurements (FreeStyle, Abbott Diabetes Care Inc.).
Body Composition
2.4
Nuclear magnetic resonance (echoMRI, Houston, TX) was used to determine body composition.
Determination of Energy Expenditure via Indirect Calorimetry
2.5
Mice were individually housed in sealed chambers (TSE systems, Chesterfield, MO, USA) with unrestricted access to food and water for 96 h to measure locomotor activity, respiratory exchange ratio (RER) and energy expenditure (EE). Data from the final 72 h of this period were utilized to calculate the average daily EE. To determine EE over the 28‐day period, we calculated the average daily EE for each group using indirect calorimetry measurements collected at the conclusion of the study. For CD‐fed groups, daily EE was assumed to remain constant from day 0. For HFD‐fed groups, we assumed that daily EE increased at the same rate for each group from day 0, when all subjects were initially fed the CD, through day 28. The cumulative EE for the 28‐day period was then determined by summing these daily estimates for each respective group.
Estimation of Energy Balance and Energy Stored
2.6
Energy balance (EB) was calculated by subtracting the estimated average 4‐week cumulative EE for each group from the 4‐week cumulative energy intake (EI) calculated as the average per mouse for each cage. Energy stored (ES) was calculated assuming a caloric content of 9.4 and 1.8 kcal/g of fat mass and lean mass, respectively (Guo and Hall 2011), gained or lost during the 28‐day period.
Tissue Collection and Gene Expression Analysis
2.7
Liver, epididymal white adipose tissue (eWAT), inguinal white adipose tissue (iWAT), brown adipose tissue (BAT), and quadriceps (Quad) were collected immediately after decapitation, weighed, flash‐frozen using liquid nitrogen, and then stored at −80°C. Tissues were homogenized in the lysis buffer provided by the kit using a tissue homogenizer (TissueLyser, Qiagen). RNA from the liver lysate was extracted using the RNeasy mini kit (Qiagen, Cat #74106). RNA was extracted using the RNeasy lipid tissue mini kit (Qiagen, Cat #74804) or RNeasy fibrous tissue mini kit (Qiagen, Cat #74704) in the case of muscle. cDNA was generated using the iScript cDNA synthesis kit (Bio‐Rad, Cat #1708891). Gene expression was performed with TaqMan probes and TaqMan gene expression master mix (Cat #4370074) from Applied Biosystems using the QuantStudio 5 realtime PCR system (Applied Biosystems).
Plasma Measurements
2.8
We used commercial assays for Insulin (Crystal Chem, #90080), Leptin (Crystal Chem, #90030), Adiponectin (Crystal Chem, #80569), Cholesterol (Infinity Cholesterol, #TR13421), Glycerol (Sigma Aldrich, #MAK117), and Triglycerides (Infinity Triglycerides, #TR222421).
Statistical Analysis
2.9
The statistical analysis was performed using GraphPad Prism (v10), except for the analysis of energy expenditure via two‐way ANCOVA, which was performed using SPSS (v31). The data are expressed as mean and standard error of the mean. Statistical differences between groups were determined using t‐test (paired, unpaired, or nested) for pairwise comparisons. Two‐, three‐way, or repeated measurements (RM) ANOVA were followed by pairwise post hoc comparison using Fischer's LSD or Tukey's multiple comparison test if statistically significant main effects were detected. p < 0.05 were considered statistically significant.
Results
3
Impact of Age and Diet on Body Composition and Glucose Tolerance
3.1
To determine the impact of aging on the susceptibility to diet‐induced obesity (DIO), we switched the diet of young and old C57Bl6/J male mice from standard chow (CD) to a high‐fat diet (HFD) and monitored body weight (BW) and body composition for 28 days. The old mice exhibited a significantly higher BW (Figure 1a), lean mass (Figure 1b), and fat mass (Figure 1c) at the beginning of the study compared to the young counterparts. HFD feeding significantly increased BW (Figure 1a) and fat mass (Figure 1c,d). The change in fat mass due to HFD feeding was significantly larger in the old mice compared to the young mice (Figure 1d). Lean mass was unaffected by HFD (Figure 1b,d), although the young mice exhibited an increase in lean mass during the 28‐day period (Figure 1d). When normalized as percentage change relative to CD‐fed controls, HFD‐fed old mice exhibited a larger percent BW gain (Figure S1a), while HFD‐fed young mice had a greater percent lean mass gain compared to HFD‐fed old ones (Figure S1b). The difference in fat mass gain relative to CD controls between HFD‐fed old and young mice was statistically significant only at 14 days post‐HFD (Figure S1c).
Effect of 4‐week HFD vs CD feeding on body composition, storage efficiency, and glucose tolerance in young and old CD WT C57BL/6J male mice either maintained on CD or fed HFD for 28 days. Mice were housed as 4 per cage. (a) Body weight measurements across the 4‐week CD/HFD feeding period. (b) 4‐week lean mass change. (c) 4‐week fat mass change. (d) Lean and fat mass change between day 0 and day 28. (e) Plasma leptin, plasma adiponectin (f) and hepatic triglyceride content per gram of wet tissue (g). Glucose tolerance test (GTT, 2 g/kg i.p.) (h) on a subset (n = 6) of mice from each group, maintained on their respective diets through the 36th day. (i) Area under curve of blood glucose (BG) levels during the GTT, relative to the baseline BG levels prior administration of the glucose bolus (j). (k) Baseline plasma insulin. Data presented as mean ± s.e.m. (a–c) 3‐way ANOVA, followed by Tukey's post hoc test. (d–f, h) 2‐way ANOVA followed by Fisher's LSD post hoc test. * = p < 0.05, ** = p < 0.01, *** = p < 0.001. (a–d, n = 19–20; e–j, n = 5–6).
Leptin levels increased with HFD feeding without a significant effect of age (Figure 1e), whereas both age and diet significantly (p < 0.023 and p < 0.01) influenced adiponectin levels, with old CD mice exhibiting significantly higher levels than the young counterparts (Figure 1f). Both age and HFD significantly increased hepatic triglyceride content (p < 0.032 and p < 0.0001, respectively), with HFD‐fed old mice exhibiting significantly higher hepatic lipid content than their HFD‐fed young counterparts (Figure 1g). HFD feeding impaired glucose tolerance in young mice (Figure 1h,i), impairment also seen in the baseline BG (Figure 1j). Interestingly, old HFD‐fed mice remained glucose tolerant (Figure 1h,i) and preserved baseline BG (Figure 1j) despite increased DIO (Figure 1a,d) and increased hepatic steatosis (Figure 1g). A two‐way ANOVA showed that both age (p < 0.026) and diet (p < 0.0001) significantly affected baseline insulin levels. Post hoc analysis found HFD feeding raised insulin (3.05‐fold) in young mice and to a lesser extent (1.90‐fold) in old mice (Figure 1k).
Impact of Age and Diet on Gene Expression
3.2
To gain insights on why older mice are more susceptible to fat mass gain, we analyzed gene expression related to adipose tissue function in epididymal (eWAT) and inguinal (iWAT) white adipose tissue (Figure 2). Expression of adipogenesis regulators Cebpa, Cebpb, and Pparg increased significantly due to the HFD feeding in eWAT, but this effect was significantly reduced in aged mice, as determined by a main effect of age (Figure 2b) or following pairwise comparisons (Figure 2a,c). In iWAT, age alone diminished Cebpa and Cebpb expression in control diet (CD) fed mice. Unlike in eWAT, HFD did not raise Pparg or Lep (encoding leptin) expression in iWAT of young mice but did so in old mice (Figure 2c,d). Diet had a significant effect on Adipoq (encoding adiponectin) expression in both eWAT (p < 0.0001) and iWAT (p < 0.0249). Specifically, HFD feeding significantly increased and decreased Adipoq expression in eWAT and iWAT of the young mice, respectively (Figure 2e). In contrast, Adipoq expression remained unaffected in old mice.
Gene expression analysis of epididymal and inguinal adipose tissue depots of young and old C57BL/6J male mice, fed with CD or HFD for 28 days. Expression of genes involved in adipogenesis (a–e), lipogenesis (f–i), lipolysis (j–n) and inflammation (o–q). Data presented as mean ± s.e.m. n = 5–6. Outliers detected using the Grubb's test (alpha = 0.05) were removed from the analysis. Data for each gene are presented as relative expression compared to CD (Yng) for eWAT and analyzed using 2 way‐ANOVA, followed by Fisher's LSD post hoc test for each fat depot. The expression of adrenergic receptors and Gipr are shown as relative to the levels of Adrab1. * = p < 0.05, ** = p < 0.01, *** = p < 0.001.
In eWAT, HFD elevated the lipogenic genes Insig1, Fasn, and Scd1, with a diminished response in aged mice (Figure 2f,g,i). HFD reduced the expression of the lipogenic gene Elovl6 in eWAT regardless of age and in iWAT of young mice (Figure 2h). HFD did not impact Insig1 or Scd1 expression in iWAT of young mice, but both were increased in old mice (Figure 2f,i). HFD increased the expression of Pnpla2, encoding the key lipolytic enzyme adipose triglyceride lipase (ATGL) in eWAT of young mice and decreased it in iWAT (Figure 2j). In CD‐fed old mice, iWAT Pnpla2 expression was lower than in young mice. In contrast to eWAT, HFD increased Pnpla2 expression only in old iWAT (Figure 2j). The expression of Adrb1 and Adrb3 and of Gipr, all encoding g‐protein coupled receptors (GPCRs) involved in the control of lipolysis, was upregulated by HFD in eWAT of young mice but not in old mice, mirroring the pattern of Pnpla2 expression (Figure 2k,m,n). Adrb2 did not change with HFD but was higher in older eWAT (Figure 2l). HFD reduced Adrb1 and Adrb3 expression in iWAT of young mice, while in old mice it stayed mostly unchanged, except for an increase in Adrb1 (Figure 2k,m). Markers for leukocyte (Itgam) and macrophage (Emr1) infiltration rose in eWAT with HFD but stayed unchanged in iWAT (Figure 2o,p); the upregulation of Itgam in eWAT was less pronounced in old mice (Figure 2o). Tnf expression increased with HFD in both fat depots, regardless of age (Figure 2q). Taken together, these data indicate a reduction in the plasticity and lipid turnover of visceral fat in old mice following 4‐week HFD feeding, which could contribute to a net increase in lipid deposition. Neither age nor HFD influenced the expression of genes involved in adipogenesis, inflammation, or thermogenesis in interscapular brown adipose tissue (iBAT) (Figure S2a). However, lipid metabolism markers were differentially regulated: age increased Elovl6 and Fasn, while HFD upregulated Insig1 and Scd1. In the liver (Figure S2b), HFD significantly increased the expression of the key glucoregulatory genes Gck and G6pc in old mice but reduced Pck1 in young mice. Emr1 and Itgam, inflammatory markers, were unchanged. In young mice, lipid metabolism markers were unaffected by HFD, but aging increased Elovl6, Fasn, and Scd1, in line with age‐dependent hepatic lipid accumulation (Figure 1g). Hepatic Ppara expression was reduced in CD‐fed old mice but was restored by HFD (Figure S2b). Neither age nor diet had a significant effect on the expression of hypertrophy‐associated genes (Myh1, Myh2, Igf1, Fst, or Igfbp5) in quadriceps (Figure S2c). Interestingly, age did significantly decrease the expression of muscle atrophy‐related genes Mstn and Trim63 (Figure S2c). Energy metabolism gene expression in skeletal muscle was largely unchanged, except for the lactate transporter Slc16a3, which was significantly reduced by both age and HFD (Figure S2c).
Impact of Age and HFD Feeding on EE, RER, and Locomotor Activity
3.3
To determine whether changes in EE contributed to the increased susceptibility to fat mass gain in HFD‐fed old mice, we performed indirect calorimetry for 72 h after 28 days of the dietary intervention in a subset of mice from each group (n = 6). EE significantly increased due to HFD feeding (Figure 3a,d). Furthermore, old mice exhibited increased EE compared to young mice, irrespective of the diet (Figure 3a,d).
Energy expenditure, respiratory exchange ratio, and locomotor activity of young and old mice over the course of 72 h after 28 days of HFD feeding. (a) Hourly energy expenditure (EE), (b) respiratory exchange ratio (RER) and (c) locomotor activity. (d) Average EE per group over the 72‐h period. (e) Average EE per mouse, relative to lean mass. Average RER (f) and total locomotor activity (g) over the 72‐h period. Data presented as mean ± s.e.m. n = 6. (d–g) 2‐way ANOVA, followed by Fisher's LSD post hoc test. * = p < 0.05, ** = p < 0.01, *** = p < 0.001, **** = p < 0.0001. (e) Linear regression analysis failed to detect a slope significantly different from 0 when comparing Yng CD vs. Yng HFD (dotted lines) and Old CD vs. Old HFD (dashed lines). When the analysis considered only diet as a variable (bold solid lines), the slopes of both groups differed from 0 (CD: R 2 = 0.47, p = 0.014; HFD: R 2 = 0.63, p = 0.002). Although the slopes did not differ significantly, their intercept was statistically significant (p < 0.001).
Given the linear relationship between lean mass and whole‐body energy expenditure (EE) and considering that old mice exhibited increased lean mass and total EE compared to young mice, we evaluated the average EE relative to the lean mass for each group (see Figure 3e). Neither linear regression analysis nor Pearson correlation analysis detected a significant correlation between EE and lean mass within each group with the given number of replicates. When considering pairwise comparisons of the linear regression fit of groups on the same diet, the Extra‐sum‐of‐squares F test indicated that a single fit was the preferred model over separate fits (F = 0.077 (2, 8), p = 0.93 for CD; F = 1.64 (2, 8), p = 0.25 for HFD). EE and lean mass exhibited a significant positive correlation (r = 0.69, p = 0.01 for CD; r = 0.71, p = 0.01 for HFD) (Figure 3e), and the linear regression analysis indicated statistically significant positive slopes (p = 0.014 for CD; p = 0.01 for HFD). This suggests that lean mass contributes to differences in EE when comparing groups on the same diet, regardless of age. Furthermore, there was a statistically significant intercept (p < 0.0001) between the CD and HFD slopes (Figure 3e), indicating that in addition to the effect of lean mass, HFD significantly increased EE. These findings align with those from a two‐way ANCOVA, which demonstrated that, after adjusting for lean mass alone (data not presented) or for both lean and fat mass, only diet significantly influenced differences in energy expenditure (Table S1). Additionally, a 2‐way ANCOVA conducted for energy intake during the calorimetry experiment indicated a statistically significant effect of diet (p < 0.002) when controlling for lean mass (data not shown); however, this effect was not statistically significant when both lean and fat mass were considered (Table S2).
As expected, consumption of HFD reduced RER (Figure 3b,f), but there was also a significant decrease due to aging, particularly between both CD‐fed groups (Figure 3f). HFD significantly reduced home cage activity, whereas aging had no effect (Figure 3c,g), suggesting activity did not contribute to the diet and age‐dependent increase in EE (Figure 3a,d).
Determination of Energy Balance and Energy Storage Efficiency During HFD Feeding
3.4
To determine the contribution of energy intake (EI) to fat mass gain, we monitored daily food intake over 28 days. CD‐fed mice had similar daily (Figure 4a) and total EI (Figure 4b). Both young and old mice showed significantly higher EI immediately after starting HFD, with old mice exhibiting significantly increased EI compared to young mice (Figure 4a). The HFD groups exhibited significantly higher 28‐day EI (Figure 4b), with old HFD‐fed mice exhibiting significantly higher EI compared to young HFD‐fed mice (Figure 4b).
Determination of energy balance, and storage efficiency in young and old CD‐fed WT C57BL/6J male mice maintained on CD or fed HFD for 28 days. (a) Daily home cage energy intake (EI) per mouse. The daily EI was calculated by dividing the total caloric intake per cage by the number of mice within in cage (n = 5 cages/group, n = 3–4 mice/cage). (b) Cumulative 4‐week EI expressed as kcal/mouse from each of the 5 cages per group. Caloric intake accounts for spilled food, which was separated from bedding and feces. (c) Estimated cumulative energy expenditure (EE) per day and per mouse over the 4‐week period. The 72‐h average EE for each group (see Figure 2d) was used as estimated EE for day 28. Estimated daily EE assumes a linear increase from day 0 for the HFD groups and that the EE for the CD groups remained constant throughout the 28‐day period. (d) Estimated 4‐week cumulative EE for each group. (e) Energy balance (EB) during the 4‐week period. EB was calculated by subtracting the estimated daily average EE for each group (c) from each of the individual replicates used to calculate the daily EI as summarized in (a). (f) Cumulative 4‐week EB expressed as kcal/mouse. (g) Energy balance (CD: Black, HFD: Red) and energy stored as fat (CD: Blue, HFD: Purple) during the 4‐week period. Energy stored (ES) was calculated assuming a caloric content of 9.4 kcal and 1 kcal per gram of fat and lean mass, respectively, gained (or lost) between days 0 and 28 (see Figure 1d). (h) Storage efficiency, calculated as percentage of ES relative to EB. Grubb's test (alpha = 0.05) detected an outlier (628%) in the Old CD group which was removed from the analysis. (i) Correlation between the EB‐ES fat and the change in lean mass exhibited by young and old mice. Data presented as mean ± s.e.m. of the number of cages per group (n = 5). The EI was calculated considering the total food consumption per cage divided by the number of mice per cage. ES was calculated considering the total amount of fat and lean mass per cage, divided by the number of mice per cage. (a, e) 3‐way ANOVA detected statistically significant differences for all main effects and interactions, except for age in (e). Asterisks indicate statistically significant differences between Yng and Old HFD‐fed mice after pairwise RM‐ANOVA, followed by Sidak's post hoc test. (b–h) 2‐way ANOVA, followed by Fischer's LSD post hoc test. (g) Differences between EB and ES were analyzed using a paired t‐test. (i) Pearson r = 0.73, p = 0.016 for young mice. * = p < 0.05, ** = p < 0.01, **** = p < 0.0001. n = 5 (except c, d, which depict the average for each group).
To estimate EE over the 28‐day period, we first calculated the average daily EE for each group (Figure 4c, day 28; CD (Young): 9.36 kcal/d, CD (Old): 9.99 kcal/d, HFD (Young): 11.27 kcal/d, HFD (Old): 12.17 kcal/d) based on hourly EE measurements obtained at the study's conclusion (Figure 3d). We then assumed that daily EE remained constant in the CD‐fed groups since day 0, while EE increased daily at the same rate in the HFD groups (Figure 4c; HFD (Young): 0.078 kcal/d, HFD (Old): 0.068 kcal/d), beginning from day 0 when all mice were initially fed with CD. We then estimated total average EE over 28 days for each group (Figure 4d). Subtracting EE from EI, we calculated daily (Figure 4e) and cumulative 28‐day energy balance (EB) (Figure 4f). A pairwise comparison using 2‐way ANOVA showed a significant age effect (p = 0.045) on daily EB for CD‐fed groups, although their total 28‐day EB did not differ (Figure 4f). HFD significantly increased daily (Figure 4e) and total (Figure 4f) EB for both age groups compared to CD controls (p < 0.001). Old HFD‐fed mice had a much higher total EB than young HFD‐fed mice (Figure 4f), indicating more calories available for storage.
We then calculated energy stored (ES) by considering changes in fat and lean mass (Figure 1d), using caloric values of 9.4 kcal/g of fat and 1.8 kcal/g of lean mass, gained or lost. Pairwise comparisons showed significant differences between 28‐day EB and ES for both young groups (Figure 4g). Two‐way ANOVA detected significant effects of age (p = 0.004) and diet (p = 0.002) on storage efficiency. It is worth noting that CD‐fed old mice showed considerable variability in % storage efficiency (Figure 4g), which could influence the overall impact of age. This highlights the sensitivity limitations of our calculations for both EB and ES within this group, as both values are close to 0 kcal. Old HFD‐fed mice exhibited an average storage efficiency of 92%, which was not statistically different from the expected 100% (t‐test, p = 0.26). Conversely, the storage efficiency in young HFD‐fed mice was approximately 60%, significantly lower than 100% (t‐test, p = 0.003), and significantly lower than that of the old HFD group (Figure 4h). The change in lean mass for both groups of young (but not old) mice showed a significant positive correlation (Pearson r = 0.73, p = 0.016) with the EB remaining after only subtracting the energy stored as fat (ESfat) (Figure 4i). These results suggest that the reduced storage efficiency observed in the young mice (Figure 4h) may be attributed to an increased energy requirement for accruing lean mass (Figure 1d).
To validate the lower‐than‐expected storage efficiency observed in young HFD‐fed mice, we submitted a separate cohort of young mice to a 28‐day exposure to HFD feeding and repeated the measurements of BW, change in body composition, determination of ES, EE, EI, and EB (Figure S3a–h). Determination of EI took into consideration food spillage, which was collected weekly. In that replication study, the storage efficiency of EB into ES for HFD‐fed mice was 49% (Figure S3i), which differed statistically from the expected 100% (t‐test, p < 0.001). This confirmatory study further suggests that the near 100% storage efficiency exhibited by old HFD‐fed mice contributes to their higher susceptibility to DIO compared to younger mice.
Susceptibility to Body Weight Loss During Calorie Restriction
3.5
We exposed a second cohort of young and old mice, always fed with CD, to ad libitum HFD feeding for 28 days (Figure S4) and then subjected them to CR for 18 days (Figure 5). Older mice had significantly higher body weight (Figure S4a), fat mass (Figure S4c), and lean mass (Figure S4d) before HFD feeding.
Effect of 18‐day calorie restriction (CR) on body weight, glucose tolerance and baseline insulin levels young and old C57BL/6J male mice, previously fed with CD or HFD for 28 days. (a) Cumulative energy intake (EI) during the 18‐day CR period. (b) Percentage CR compared relative to the EI consumed by the ad libitum counterparts during the 18‐day period. (c) Difference in EI between the CR groups and the corresponding ad libitum control at the end of the 18‐day period. (d) BW during the 18‐day CR period. (e) Body weight change of the CR groups at day 18 compared to their initial BW. (f) Body weight change of the CR groups at day 18 compared to the average BW of their ad libitum fed counterparts on day 18. (g) 6‐h fasted glucose tolerance test (GTT), performed on day 13. Mice were injected with 2 g/kg (20% wt/vol) of dextrose at time = 0 min. (h) Area under curve (AUC) of the blood glucose (BG) levels during the GTT, relative to the baseline BG levels prior administration of the glucose bolus (i). (j) Baseline insulin levels in plasma collected at time of euthanasia on day 18. Data presented as mean ± s.e.m. (e) 2way‐ANOVA, followed by Fisher's LSD post hoc test. (h, i) 3way‐ANOVA, followed by Šídák's Multiple Comparisons test. * = p < 0.05, ** = p < 0.01, *** = p < 0.001. (a–c) Mice were housed as 2–4 per cage. Food intake was collected from each cage and normalized to the number of mice per cage. (d–i) The study begun with n = 15 (CD old), n = 16 (CD/HFD Yng, HFD old) and n = 8 (CR) mice per group.
HFD feeding for 28 days increased BW and EI in both young and old mice (Figure S4a,b). Consistent with our earlier findings (Figure 1d), old HFD‐fed mice experienced a significantly larger change in fat mass gain when compared to young HFD‐fed mice (Figure S4c). HFD modestly but significantly increased lean mass regardless of age, with young mice exhibiting a significantly larger gain compared to old mice (Figure S4d).
After 28 days on HFD, eight mice from each group underwent an 18‐day CR based on the daily EI of the remaining mice kept feeding ad libitum (Figure 5a). At the end of the 18‐day period, the percentage of CR for CD‐fed mice ranged between 38.5% (young) and 37.4% (old), and for HFD‐fed mice between 35.7% (young) and 35.9% (old) (Figure 5b). Considering the total EI for each group (Figure 5a and Figure S4f,g), CR young mice fed either with CD or HFD received 11% or 9% fewer calories, respectively, compared to their old counterparts (Figure 5c). In contrast to the initial 4 weeks of HFD feeding (Figure S4a–e), old ad libitum HFD‐fed mice did not gain BW during the subsequent 18‐day period, unlike their young counterparts (Figure 5d and Figure S4l). This could be due in part to a relatively lower caloric intake, considering that, despite a higher BW and lean mass (Figure S4d), old HFD‐fed mice exhibited similar caloric intake compared to their lighter young counterparts during the same period (Figure 5a and Fifure S4f,g). All CR groups lost BW regardless of their initial BW (Figure 5d and Figure S4c,d,l). Both age (p < 0.0002) and diet (p < 0.025) had a significant impact on the BW change since the onset of the CR (Figure 5e). Specifically, old mice lost significantly more BW compared to their young counterparts, regardless of the diet (Figure 5e), although this difference remained statistically significant only for the HF‐CR groups when considering the BW change calculated on a percentage basis (Figure S4l). Pertaining to the effect of diet, pairwise comparisons were statistically significant only between CD‐CR and HF‐CR young mice (Figure 5e). On the other hand, there were no differences in BW loss when considering the BW of the ad libitum controls on day 18 (Figure 5f).
The weights of visceral (eWAT) and subcutaneous (iWAT) white adipose tissue depots, and the liver were significantly higher in old mice compared to the young and increased significantly with HFD feeding (Figure S5a,b,d). There was also a significant effect of age on the eWAT/iWAT ratio (3‐way ANOVA, p = 0.03). HFD increased interscapular brown adipose tissue (iBAT) depot weight, but age had no effect (Figure S5c). CR significantly reduced the weight of the four tissues (Figure S5a–d), although pairwise post hoc comparisons only detected a statistically significant reduction in the liver of old HFD‐fed mice (Figure S5d). HFD significantly increased plasma leptin levels in both young and old ad libitum fed mice, while CR had an overall statistically significant effect lowering them (Figure S5e). Adiponectin levels were significantly influenced by diet and by an interaction among age, diet and CR (Figure S5f). Pairwise post hoc comparisons only detected a statistically significant reduction in old HFD‐fed mice compared with their CD fed counterparts (Figure S5f). Age, HFD, and calorie restriction significantly influenced plasma triglycerides, although pairwise post hoc comparisons only detected significant differences between young HFD‐fed mice and their CD counterparts (Figure S5g). HFD feeding significantly increased cholesterol levels regardless of age, whereas neither age nor CR had a significant impact (Figure S5h).
To determine the impact of CR on whole‐body glucose homeostasis, we performed a GTT on day 13. Ad libitum‐fed young, but not old, HFD‐fed mice exhibited glucose intolerance, which was significantly reversed by CR (Figure 5g,h). CR improved glucose tolerance in CD‐fed groups (Figure 5g,h and Figure S4j,k) but reduced baseline BG levels in all groups (Figure 5i). Baseline insulin levels at the end of the 18‐day period revealed significant differences due to HFD and CR, but not age (Figure 5j). Specifically, CR effectively reduced hyperinsulinemia in HFD‐fed groups (Figure 5i,j). HFD feeding significantly increased circulating glycerol levels, a product of lipolysis serving as a gluconeogenic substrate, with levels significantly higher in young HFD‐fed mice compared to their old counterparts (Figure S5i). CR reduced circulating glycerol levels in HFD‐fed mice regardless of age (Figure S5i). This data suggests that the increased efficiency in calorie storage does not reduce the effectiveness of calorie restriction in promoting body weight loss in old mice.
Discussion
4
Here we provide evidence that age increases the susceptibility to obesity in male mice following acute HFD feeding. This increased susceptibility to obesity is contributed to a large extend by maximizing energy storage efficiency. Thus, whereas HFD‐feeding promotes both fat and lean mass gain in young mice, it only increased fat mass gain in old mice. This is consistent with the effect of aging in humans, which favors fat mass gain despite loss in muscle mass, contributing to the age‐dependent increase incidence of sarcopenic obesity (JafariNasabian et al. 2017; Prado et al. 2024). Clinical data indicate that persistence of obesity after mid‐life increases the likelihood of sarcopenic obesity later in life (Lutski et al. 2020). As expected, HFD feeding deteriorated glycemic control in young mice by promoting baseline hyperglycemia and impaired glucose tolerance. Interestingly, old male mice displayed normal baseline blood glucose and glucose tolerance despite increased baseline fat mass and, unexpectedly, both parameters were unaffected following HFD feeding despite increased susceptibility to fat mass gain. Of note, we did not monitor survival on our aging mouse cohorts. This could become a significant limitation should that be unusually reduced, as that could have led to an overrepresentation of the most metabolically fit mice, potentially contributing to the enhanced glycemic control observed. That aside, although these findings are at odds with evidence that glucose tolerance diminishes with age in humans (Basu et al. 2006; DeFronzo 1981; Elahi et al. 1993; van den Beld et al. 2018), they are also consistent with evidence that such deterioration may be reversed in very old individuals (Paolisso et al. 1996), who also benefit from being overweight for overall longevity (Duan et al. 2023; Toss et al. 2012). Although age had a significant impact increasing baseline insulin level, old did not develop greater hyperinsulinemia than young counterparts when challenged with HFD, despite increased fat mass gain. Notably, adiponectin levels were elevated in old CD‐fed mice, potentially contributing to enhanced glycemic regulation through improved insulin sensitivity. However, this benefit would be diminished by HFD feeding, which significantly decreased adiponectin concentrations in old HFD‐fed mice. It is also possible that alternative mechanisms improving glucose disposal may contribute to preserve glucose tolerance in old mice, despite any age‐related decline in insulin sensitivity. Hence, the increased hepatic glucokinase and lipogenic enzyme gene expression exhibited by HFD‐fed old mice is consistent with increased efficacy by the liver storing and converting glucose as fat. This may explain their greater liver weight and elevated hepatic triglyceride levels. Although aging is associated with a higher risk of steatohepatitis (Kim et al. 2015), our results indicate that older mice accumulated more hepatic triglycerides without a corresponding rise in proinflammatory markers. This pattern is consistent with clinical evidence showing that age‐related liver fat accumulation is driven more by increased adiposity than by age itself (Kuk et al. 2009). Nonetheless, given the short duration of our study, we cannot exclude the possibility that a shift toward de‐novo lipogenesis, while potentially beneficial for short‐term glycemic control, may become maladaptive and contribute to steatohepatitis over longer periods. Interestingly, plasma glycerol, a product of lipolysis and marker of insulin resistance, was significantly lower in HFD fed old mice compared to their younger counterparts. This reduction correlated with a reduced induction of the expression of genes involved in regulating lipolysis (i.e., Pnpla2, Adrb3 or Gipr) within the visceral WAT of HFD‐fed old mice. These findings are consistent with previous reports demonstrating increased Adrb3 expression in visceral as opposed to subcutaneous fat from obese subjects (Hoffstedt et al. 1997); and with the effect of age reducing adipose tissue catecholamine‐induced lipolysis in humans (Lonnqvist et al. 1990) and mice (Camell et al. 2017). Notably, the reduced induction of Gipr expression, known to promote lipolysis in adipose tissue (Regmi et al. 2024), suggests an attenuated lipolytic activity in visceral WAT resulting from broader regulation of the adipocyte transcriptional program by aging. Reduced lipolysis is also consistent with lesser increase in Itgam gene expression, a marker of myeloid cell recruitment linked to increased lipolysis (Kosteli et al. 2010) and to the onset of insulin resistance. Given this, such reduction may contribute to the preservation of glucose tolerance, as lower levels of circulating free fatty acids and glycerol can mitigate insulin resistance and reduce gluconeogenic substrate supply. This is consistent with observations in tissue‐specific hormone sensitive lipase knockout mice, which exhibit protection from glucose intolerance during HFD feeding associated with decreased plasma glycerol levels (Pajed et al. 2021).
The increase in eWAT depot weight in HFD‐fed old mice is consistent with previous findings (Chinnapaka et al. 2024), and the higher eWAT/iWAT ratio exhibited by old mice is consistent with an age‐related increase in visceral fat (Hunter et al. 2010). Interestingly, our data suggest that aging regulates lipid metabolism differently across fat depots. Thus, although old mice fed a HFD exhibited greater acute fat mass gain, they demonstrated lower expression of genes related to adipogenesis and lipid synthesis in visceral adipose tissue (such as Cebpa, Pparg, or Adipoq) when compared to young mice. This suggest that fat mass increase in young mice may be primarily driven by expansion of adipose tissue, whereas in older mice it results mainly from lipid accumulation within pre‐existing adipocytes. Increased adipogenesis may account for the sustained BW gain seen in young mice compared to the reduced BW gain observed in old mice after the initial four weeks of HFD feeding. In contrast certain genes—such as Pparg, Lep, Insig1, and Scd1—are upregulated in the subcutaneous fat (iWAT) of HFD‐fed aged mice. Subcutaneous fat is particularly responsive to fat accumulation (Okuno et al. 1998) and the insulin sensitizing effect of PPARγ activators (Ciaraldi et al. 2002); and loss of PPARγ expression in subcutaneous fat impairs glucose homeostasis (Xu et al. 2018), emphasizing its importance for glycemic control. In keeping with this, mice with reduced (le Lay et al. 2022) or increased (Kim et al. 2007; Kusminski et al. 2012) subcutaneous adipose tissue expansion capacity exhibit impaired or improved glucose tolerance, respectively. Hence, our data suggest that enhanced sensitivity of subcutaneous fat to insulin‐driven lipid deposition in response to HFD feeding in older mice could contribute in part to improved glycemic control in the old HFD‐fed mice.
Studies using transcriptomic analyses across multiple tissues have demonstrated that both aging and DIO are associated with reduced “gene elasticity,” defined as a diminished capacity of gene expression to adapt to acute challenges in energy availability (i.e., ad libitum versus fasting versus refeeding) (Zhou et al. 2023). Although these studies did not examine the effects of DIO in aged mice, their findings suggest that aging and DIO together should exert a compounding, detrimental effect on gene regulation. Our gene expression analysis—despite being limited by the small number of genes examined and the absence of groups reflecting acute changes in energy status—reveals alterations in specific tissues, particularly in iBAT and liver consistent with this notion. In contrast, gene expression patterns in subcutaneous and visceral WAT indicate that this compounding effect is tissue‐specific, with distinct genes being differentially affected by the interaction between age and DIO. Further in‐depth studies will be required to identify the factors underlying these tissue‐dependent differences.
A reduction in lipid turnover has been associated with the increased susceptibility to weight gain in humans (Valentine et al. 2022). The extent to which gene expression differences between visceral and subcutaneous adipose tissues contribute to preferential fat deposition in older mice remains unclear. Nonetheless, the simultaneous reduction in genes linked to adipogenesis, lipid synthesis, and lipolysis within visceral fat in old HFD‐fed mice is consistent with a lower rate of local lipid turnover, which could favor fat mass gain. Regardless, overall fat mass gain ultimately depends on a net positive energy balance, either from decreased energy expenditure or increased energy intake.
There is evidence that increased caloric intake accounts for most weight gain in humans (Swinburn et al. 2009). In our studies, old mice exhibited a transient but significant increase in caloric intake when first exposed to HFD, compared to young controls, which contributed to a higher net energy balance in old mice. This increase is consistent with previous reports in rats (Judge et al. 2008), but at odds with evidence in humans indicating a decrease in appetite with age (Dericioglu et al. 2024). The cause for this transient increase in caloric intake following acute exposure to HFD remains to be elucidated. A study exposing C57/Bl6 mice to multiple flavors found no substantial age‐related differences (Tordoff 2007). However, that study did not address fatty food specifically. Age may raise the hedonic value of fat apart from taste, potentially causing a temporary rise in calorie intake. However, this hypothesis has yet to be tested.
However, when considering the energy stored as either fat or lean mass gained during the first 4 weeks of HFD feeding, our data suggest that the age‐dependent susceptibility to acute fat mass gain is not solely due to increased calorie intake in the old mice. Hence, our data suggest that old mice are more susceptible to obesity due to increased efficiency in energy storage compared to the young mice. Although the lower‐than‐expected efficiency in young mice is consistent with findings in human studies (Bouchard et al. 1990), the explanation for the age‐related discrepancy remains elusive. Our determination of net energy balance is limited by several factors. Pertaining to energy intake, our calculations assume that neither diet nor age affects the efficiency of metabolized energy absorption. This is supported by data showing that wildtype mice have consistently low fecal lipid levels, regardless of dietary fat content (Petit et al. 2007). Reduced storage efficiency could be due to overall malabsorption in the young compared to the old mice. There is evidence suggesting lipid malabsorption in old compared to young mice (Yamamoto et al. 2014), but this should result in decreased storage efficiency in the old mice, suggesting a negligible impact of age‐related malabsorption in our cohort.
Regarding energy expenditure, some studies have suggested that energy expenditure declines with age in humans (Palmer and Jensen 2022). However, recent data indicate that energy expenditure remains relatively stable in adults aged 20–60 (Pontzer et al. 2021), even as obesity rates increase during these decades (Emmerich et al. 2024; Kuk et al. 2009). In keeping with this, we observed energy expenditure proportional to lean mass irrespective of age, with a comparable increase for both HFD‐fed groups. However, several limitations in determining energy expenditure could contribute to the observed differences in energy storage between young and old HFD‐fed mice. For example, our calculations assume that energy expenditure increases in a linear manner for both young and old HFD‐fed mice. This assumption accounts for the efficient energy storage seen in older HFD‐fed mice. In contrast, a non‐linear (e.g., logarithmic) increase in energy expenditure in younger HFD‐fed mice could explain their lower observed storage efficiency. Alternatively, there is evidence that indirect calorimetry might underestimate basal metabolic rate in CD‐fed mice when compared to direct calorimetry. Although both methods yield similar results for HFD‐fed mice (Burnett and Grobe 2014), there could be age‐dependent underestimation of energy expenditure in the young HFD‐fed mice using indirect calorimetry. Finally, our calculations assume similar caloric cost in fat and lean mass deposition for young and old mice (Guo and Hall 2011). Given that young, but not old, mice exhibit lean mass gain following HFD feeding, an underestimation of the energy cost of anabolic processes, including lean mass gain, could have contributed to the reduced calculated storage efficiency in the young mice. This is, however, a hypothesis we cannot test with our current data.
Reduced lipolysis in adipose tissue with age may play a role in promoting fat mass gain and conferring protection against impaired glucose metabolism. However, this reduced lipolytic capacity does not appear to hinder the mobilization of lipids during sustained caloric restriction. In fact, aged mice subjected to chronic CR exhibited greater BW loss than their young counterparts. Notably, although the benefits of CR on lifespan diminish significantly with age in mice (Hahn et al. 2019), increased fat mass correlates with increased lifespan in old mice during CR (Liao et al. 2011). Some factors limit the interpretation of the differences in body weight loss due to calorie restriction. Thus, both young and old HFD‐CR groups were slightly less calorie restricted than their CD‐CR counterparts, which may have resulted in less BW loss among CR‐HFD mice. Another limitation is that we did not precisely measure changes in overall fat and lean mass or their impacts on BW loss. Keeping these caveats in mind, our results support the idea that factors underlying the age‐related propensity for fat accumulation might serve to buffer against pronounced BW loss and diminished lifespan during extended periods of negative energy balance.
In summary, our findings demonstrate lower than expected storage of calories as fat that correlates with increased lean mass gain in young mice. In contrast, increased caloric intake and greater energy storage efficiency contribute to the increased susceptibility to fat mass gain in old male mice. Notably, the maintenance of normal body weight throughout life by consuming a low‐fat diet helps preserve glycemic control even when faced with an obesogenic challenge. Despite this resilient glycemic control, the long‐term health risks associated with persistent obesity underscore the importance of a healthy dietary lifestyle throughout life.
Author Contributions
L.G. and K.V. designed and executed experiments, analyzed data, and wrote the manuscript. V.J.B. and M.R. executed experiments, analyzed data, and wrote the manuscript. D.P.‐T. conceptualized the studies, designed research, analyzed data, and wrote the manuscript.
Funding
This work was supported by the College of Medicine, University of Cincinnati, D101228.
Conflicts of Interest
Diego Perez‐Tilve maintains research collaborations and receives funding from MBX Biosciences, BlueWater Biosciences, and Penguin Bio.
Supporting information
Figure S1: Effect of 4‐week HFD feeding body weight and body composition of young and old C57BL/6J male mice, expressed as percentage change compared to the CD‐control. Body Weight (a), Lean Mass (b) and Fat Mass (c) change, expressed as percentage change compared to the age‐matched CD‐controls. Data presented as mean ± s.e.m. (a–c) 2‐way repeated measurements ANOVA, followed by Sidak's post hoc test (n = 19–20). Figure S2: Gene expression analysis of interscapular brown adipose tissue, liver and quadriceps muscle of young and old C57BL/6J male mice, previously fed with CD or HFD for 28 days. Gene expression in interscapular brown adipose tissue (iBAT, a), liver (b) and quadriceps muscle (c). Data presented as mean ± s.e.m. n = 5–6. Grubb's test (alpha = 0.05) detected 10/312 (a), 5/240 (b) and 8/336 (c) and outliers that were removed from the analysis. Data for each gene were analyzed using 2 way‐ANOVA, followed by Fisher's LSD post hoc test * = p < 0.05, ** = p < 0.01, *** = p < 0.001. Figure S3: Replication study of the effect of 4‐week HFD feeding on energy balance and storage efficiency in young C57BL/6J male mice. (a) BW during the HFD feeding period. The mice were housed as 4 per cage. Change in fat (b) and lean mass (c) during the 4‐week period. Energy Stored (ES, d) and energy intake (EI, e) during the 4‐week period. The data were calculated per cage and divided by the number of mice per cage. (f) 72‐h energy expenditure (EE) determined after the 4‐week feeding period, in CD or HFD fed mice, singly housed in sealed chambers. (f, insert) Average daily EE. (g) Estimated daily EE the 4‐week period, calculated using the 72‐h average EE for each group (f insert), assuming that EE remained unchanged for the CD group and that it increased at the same rate (i.e., 2.49 kcal/wk) from day 0 for the HFD group. (h) Energy balance (EB) was calculated by subtracting the estimated total EE for each group (see g insert) from the total EI energy intake of each replicate over the 4‐week period (e). (i) Comparison of the calculated 4‐week EB and ES. Data presented as mean ± s.e.m. (a, d, f) RM‐ANOVA asterisks indicate main effect of diet (a, d) or difference detected by Fischer's LSD post hoc test (f). (b) Nested t‐test. (d, e, insert in f, h) Unpaired t‐test. (i) Paired t‐test. * = p < 0.05, ** = p < 0.01, *** = p < 0.001. (a–c) n = 24; (d, e, h, i) n = 6; (f) n = 12. Figure S4: Effect of a 4‐week HFD feeding of young and old C57BL/6J male mice, prior to undergoing an 18‐day calorie restriction. (a) body weight, (b) energy intake (EI), (c) fat and (d) lean mass. (e) Change in lean and fat mass. Mice were housed as 2–4 per cage. (f–l) Data presented in Figure 5 as shown in separate graphs as CD and HF fed groups from better visualization. (f, g) Energy intake, (h, i) BW during the 18‐day CR period. (j, k) Blood Glucose during a glucose tolerance test performed on day 13. (l) Percentage body weight change at the end of the 18‐day CR period. Data presented as mean ± s.e.m. (a–d) 3‐way ANOVA, followed by Tukey's post hoc test. (e) 2‐way ANOVA, followed by Fischer's LSD post hoc test. * = p < 0.05, *** = p < 0.001. (a, c, d, e) n = 16 (Yng) or 25 (old). (b) n = 4 (Yng) or 7 (Old). (l) 3w‐ANOVA, followed by Šídák's Multiple Comparisons post hoc test. Brackets indicate statistically significant differences. Main effects are summarized in the inserted tables. Figure S5: Tissue weights and metabolite concentration in plasma of ad libitum and calorie restricted CD and HFD WT C57BL/6J male mice. Weight of epididymal white adipose tissue (eWAT, a), inguinal white adipose tissue (iWAT, b), interscapular brown adipose tissue (iBAT, c) and the liver (d). Plasma leptin (e), adiponectin (f), triglycerides (g), cholesterol (h) and glycerol (i) levels at the end of the study. Data presented as mean ± s.e.m. Data were analyzed via 3 way‐ANOVA. Main effects are summarized in the inserted tables. Brackets indicate statistically significant differences detected by Šídák's Multiple Comparisons post hoc test. (a–f) n = 16 (ad lib), n = 8 (CR).
Table S1: Results of 2‐way ANCOVA on the energy expenditure data collected during the indirect calorimetry. Top Panel: Descriptive statistics. Means and standard deviation, as well as adjusted means and standard errors of energy expenditure for the four groups. Bottom Panel: Between‐subjects effects calculated via 2‐way ANCOVA. (n = 6).
Table S2: Results of a 2‐way ANCOVA on the energy intake data collected during the indirect calorimetry. Top Panel: Descriptive statistics. Means and standard deviation, as well as adjusted means and standard errors of energy intake for the four groups. Bottom Panel: Between‐subjects effects calculated via 2‐way ANCOVA. (n = 6, except n = 5 due to the exclusion of a HF‐Old replicate due to excessive shredding).
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Atkins, J. L. , P. H. Whincup , R. W. Morris , L. T. Lennon , O. Papacosta , and S. G. Wannamethee . 2014. “Sarcopenic Obesity and Risk of Cardiovascular Disease and Mortality: A Population‐Based Cohort Study of Older Men.” Journal of the American Geriatrics Society 62, no. 2: 253–260. 10.1111/jgs.12652.24428349 PMC 4234002 · doi ↗ · pubmed ↗
- 2Basu, R. , C. Dalla Man , M. Campioni , et al. 2006. “Effects of Age and Sex on Postprandial Glucose Metabolism: Differences in Glucose Turnover, Insulin Secretion, Insulin Action, and Hepatic Insulin Extraction.” Diabetes 55, no. 7: 2001–2014. 10.2337/db 05-1692.16804069 · doi ↗ · pubmed ↗
- 3Batsis, J. A. , T. A. Mackenzie , L. K. Barre , F. Lopez‐Jimenez , and S. J. Bartels . 2014. “Sarcopenia, Sarcopenic Obesity and Mortality in Older Adults: Results From the National Health and Nutrition Examination Survey III.” European Journal of Clinical Nutrition 68, no. 9: 1001–1007. 10.1038/ejcn.2014.117.24961545 · doi ↗ · pubmed ↗
- 4Berry, K. M. , S. Garcia , J. R. Warren , and A. C. Stokes . 2022. “Association of Weight at Different Ages and All‐Cause Mortality Among Older Adults in the US.” Journal of Aging and Health 34, no. 4–5: 705–719. 10.1177/08982643211059717.35220792 PMC 9411264 · doi ↗ · pubmed ↗
- 5Bouchard, C. , A. Tremblay , J. P. Despres , et al. 1990. “The Response to Long‐Term Overfeeding in Identical Twins.” New England Journal of Medicine 322, no. 21: 1477–1482. 10.1056/NEJM 199005243222101.2336074 · doi ↗ · pubmed ↗
- 6Burnett, C. M. , and J. L. Grobe . 2014. “Dietary Effects on Resting Metabolic Rate in C 57BL/6 Mice Are Differentially Detected by Indirect (O 2/CO 2 Respirometry) and Direct Calorimetry.” Molecular Metabolism 3, no. 4: 460–464. 10.1016/j.molmet.2014.03.003.24944905 PMC 4060218 · doi ↗ · pubmed ↗
- 7Camell, C. D. , J. Sander , O. Spadaro , et al. 2017. “Inflammasome‐Driven Catecholamine Catabolism in Macrophages Blunts Lipolysis During Ageing.” Nature 550, no. 7674: 119–123. 10.1038/nature 24022.28953873 PMC 5718149 · doi ↗ · pubmed ↗
- 8Chinnapaka, S. , H. Malekzadeh , Z. Tirmizi , and A. Ejaz . 2024. “Caloric Restriction Mitigates Age‐Associated Senescence Characteristics in Subcutaneous Adipose Tissue‐Derived Stem Cells.” Aging (Albany NY) 16, no. 9: 7535–7552. 10.18632/aging.205812.38728252 PMC 11131987 · doi ↗ · pubmed ↗