All procedures were approved and conducted in accordance to the St. Louis University Institutional Animal Care and Use Committee.

All mice were housed in a 12-h light/dark cycle. At the end of the experiment, mice were euthanized by CO₂ asphyxiation followed by cervical dislocation. Body composition was analyzed weekly by NMR using BioSpin LF50 Body Composition Analyzer (Bruker, Germany). Tissues were collected, snap frozen and stored at −80°C for qPCR or western blotting, or fixed in 4% formalin, incubated in 20% sucrose overnight before freezing for cryosection.

The floxed RORα mouse was created by the transgenics core at the Pennington Biomedical Research Center (Baton Rouge, LA, USA) by inserting LoxP sites by homologous recombination flanking the exon three of the rora gene. Female Rora^flox/flox mice were bred with male mice homozygous for the Cre recombinase gene under the control of the adenovirus EIIA promoter to generate the Rora⁺/ΔCre F1 generation. EIIa-Cre transgenic mice were obtained from The Jackson Laboratory (Bar Harbor, ME, USA; stock number: 003724). Genotypes were determined by PCR using the oligonucleotide primers: primer 1: 5′-GCCCCCTTTCCCGCCAGTAGCTG-3′; primer 2: 5′-GACCCAGATCCCCTAATAACG-3′; primer 3: 5′-GCCCTCAAGAAGAGGCTGCAATTT-3′ (Figure 1A).

Respiratory exchange ratio (RER), VO₂, VCO₂, heat production, and activity were measured using a monitoring system (Oxymax Comprehensive Lab Animal System; Columbus Instruments, Columbus, OH, USA). Eight aged-matched male mice (EIIaRORα WT and EIIaRORα KO) were housed individually in metabolic cages on a 12 h day–night cycle, fed with either a normal chow or high fat diet (HFD) (D12492, Research diet). After a 24-h acclimatization period, data were collected every 16 min over a 5-day period. The hourly or average values during light and dark periods were calculated. Two-way ANOVA followed by Bonferroni posttest was used to calculate the p value.

After 6-h fast, EIIaRORα WT and EIIaRORα KO (n = 8) were injected intraperitoneally with glucose (2 g/kg of fat free mass) or insulin (0.75 U/kg of fat free mass) (Sigma-Aldrich, St. Louis, MO, USA) to examine glucose tolerance test or ITT, respectively. Blood glucose was measured before the injection (t = 0 min), 15, 30, 60, and 120 min after injection using OneTouch Ultra^®2 glucometer. Blood was collected by tail snip; the first drop was wiped out, and the second drop was used to quantify glucose level.

Mice were euthanized, and blood was collected via cardiac puncture. Concentration of plasma total cholesterol, high-density lipoprotein (HDL) cholesterol, LDL cholesterol, triglyceride, and liver enzymes were assessed using Rx Daytona clinical chemistry analyzer (Randox). Plasma TNFα or IL-1β concentrations were detected by ELISA (Mouse TNF alpha ELISA KIT, EMD Millipore and Mouse IL-1 beta Quantikine SixPak, R&D Systems, respectively) according to the manufacturer’s recommendations.

Total RNA was isolated from mouse tissues using PureLink RNA mini kit (Ambion, Life Technologies). RNA was reverse transcribed to make cDNA using qScript™ cDNA Synthesis Kit (Quanta biosciences) according to the manufacturer’s instructions. Real-time PCR was performed using a SYBR-green PCR master mix kit (SYBR SELECT MASTER MIX, Life Technologies). Primers were purchased from Integrated DNA Technologies (see Table S1 in Supplementary Material for sequences).

Spleens and lymph nodes were harvested into a tissue culture dish and teased apart into a single cell suspension. Cell suspension was passed through a cell 0.22 µM filter (EMD Millipore) and centrifuged (300–400 × g) at 4°C. Cell pellet was resuspended in Flow Cytometry Staining Buffer (BioLegend, San Diego, CA, USA) at the final concentration of 2 × 10⁷ cells/mL. Cell surface and intracellular staining were performed according BioLegend protocol (BioLegend). Antibodies used for analysis are listed in Table S2 in Supplementary Material.

Data are expressed as mean ± SEM. Student’s t-test or two-way ANOVA was used to calculate statistical significance. p < 0.05 was considered significant.

Previous studies have shown that homozygous staggerer mice are leaner and display improved insulin sensitivity but also have a severe cerebellar ataxia (9). To help normalize feeding, these ataxic mice are typically provided food on the bottom of the cage so as to provide more ready access given their locomotor deficits (9). We generated mice heterozygous for a transgene encoding Cre recombinase expressed under the direction of the EIIa promoter (EIIa-Cre), which directs expression in all tissues, and homozygous for loxP-flanked exon 3 of Rora gene [Figure 1A, Rora^flox/floxCre^+/WT (EIIaRORα KO)]. Rora^flox/flox littermates without the EIIa-Cre transgene (EIIaRORα WT) served as controls. When we assessed the level of Rora expression in different tissues, we found that the deletion was not complete. Rora expression ranged from 45 to 30% of normal levels depending on the tissue examined. We observed a significant decrease of expression in several tissues: 65% in white adipose tissue (WAT), 65% in liver, 70% in brain, 55% in BAT, and 60% in the skeletal muscle (Figure 1B). The EllaRORα KO mice displayed normal locomotion (Figure S1 in Supplementary Material, activity data at 3-month old) and exhibited no signs of ataxia, from weaning time (day 21) to 15 months of age (data not shown). The residual RORα expression in the brain may explain why the EIIaRORα KO mice do not display ataxia. Overall, no differences in the size or the weight of several organs (liver, kidney, spleen, quadriceps, brown or WATs, heart) were observed (data not shown). There were no gross alterations in brain morphology, but fine characterization of brain morphology and function will be examined in later studies.

Three-month-old male mice (EIIaRORα KO or EIIaRORα WT) were fed with a chow diet and used to decipher the role of RORα in energy and glucose metabolism. Whole body composition was determined weekly using NMR (BioSpin LF50, Bruker Germany), and no significant differences were observed in body weight, lean mass, or fat mass (Figure 2A). Using indirect calorimetry, we were able to determine that deletion of RORα leads to an increase in fat oxidation during both day- and nighttime (Figures 2C,D). No changes in carbohydrate consumption were observed, but we observed an increase in the total quantity of fat oxidized. This increase of fat utilization led to an increase in energy expenditure with neither an increase in food intake nor activity (Figure S1 in Supplementary Material). Typically, an RER of 0.7 indicates that mice are fasted and a difference of 0.05 in the RER reflects a change of 20% of fuel utilization. Fat oxidation was increased by 20% in EIIaRORα KO compared to EIIaRORα WT mice when fed a normal chow diet. No differences were observed in levels of blood triglycerides, total cholesterol, HDL, or blood glucose in fasted or fed mice (Figure 2B). RORα deletion did not alter glucose or insulin tolerance (Figure 2E). Expression of several lipogenic and cholesterologenic enzyme genes in liver was also unaffected by RORα deletion (Figure 2F). Interestingly, Akt signaling was affected by RORα deletion. All Akt isoforms (Akt1, 2, and 3) were upregulated in EIIaRORα KO compared to EIIaRORα WT mice (Figure 2G). These results demonstrate that RORα deletion leads to an increase in fat utilization when mice are fed a normal chow diet, but loss of RORα has no effect on glucose utilization or hepatic cholesterol synthesis.

Three-month-old male mice were challenged with an HFD (60% calories from fat, Harland TD06414) over 14 weeks and monitored for body composition weekly, fuel utilization (at week 1 of HFD) and glucose metabolism. EIIaRORα KO mice exhibited resistance to DIO displaying significantly reduced body weight gain and reduced fat mass relative to control mice (Figure 3A). After 7 weeks on an HFD, a significant difference in body weight was observed (Figure 3A, left panel) and, surprisingly, after only 2 weeks on an HFD, fat mass was lower in EIIaRORα KO mice compared to EIIaRORα WT animals (Figure 3A, right panel). Three-month-old mice were maintained in the metabolic chambers for 2 days on a regular chow diet and then challenged with the same HFD. Metabolic parameters such as RER, oxygen consumption, activity, and energy expenditure were monitored for 5 days. After transition to the HFD, the RER displayed a less pronounced circadian pattern of oscillations (Figure 3B). During both day- and nighttime, the RER of EIIaRORα KO animals was lower than the wild-type counterparts (Figure 3B) consistent with an increase in fatty acid oxidation in the KO mice. We calculated the amount of fatty acid oxidized per hour per kilogram of mouse using the Frayn’s equation (11). We observed 13.5% increase in the quantity of fatty acid oxidized in the EIIaRORα KO animals compared to control mice (Figure 3C, right panel). No change in carbohydrate oxidation was observed (Figure 3C, left panel). These data suggest that EIIARORα KO mice were resistant to DIO due to an increase of fatty acid oxidation even with food intake remaining constant (Figure S1 in Supplementary Material).

One side effect of obesity is impaired glucose metabolism, which is associated with insulin resistance and type-2 diabetes. We assessed the effect of RORα deletion in glucose metabolism in HFD-fed mice for 8 weeks. At this time point, a significant difference in body weight and adiposity was observed between EIIARORα KO mice and their control littermates (Figure 3A). As no differences were observed in lean mass and to avoid an effect due to the body weight, we decided to normalize the amount of glucose or insulin injected to lean mass. No difference in response to glucose (Figure 4A) was observed between the genotypes, but EIIARORα KO mice were more sensitive to insulin compared to their EIIARORα WT littermates (Figure 4B). No differences were observed in fasted blood glucose levels between EIIARORα KO animals and their control littermates (Figure S2 in Supplementary Material). As obesity is also considered an inflammatory disease, we also assessed the blood levels of pro-inflammatory cytokines IL-1β and TNFα. No differences were observed between the two genotypes with respect to circulating IL-1β level, but we found a significant decrease in circulating TNFα levels in EIIARORα KO animals compared to EIIARORα WT mice (Figure 4B; Figure S2 in Supplementary Material). As the WAT can be a source of inflammation, we quantified inflammatory gene expression in this tissue. Most of the pro-inflammatory markers (TNFα, IFNγ, CD36, CD11c) displayed a lower level of expression in EIIARORα KO animals compared to EIIARORα WT mice (Figures 4C,D; Figure S2 in Supplementary Material). T lymphocytes also play a critical role in WAT-associated inflammation and gene expression for FoxP3, a marker for regulatory T cells (T_reg), was increased in EIIARORα KO animals compared to EIIARORα WT suggesting more infiltration of T_reg cells (Figure S2 in Supplementary Material). Previously, we demonstrated that EIIARORα KO mice display an anti-inflammatory profile with a decrease in CD4⁺ and CD8⁺ cell populations in the spleen when fed a chow diet (5). These results show that EIIARORα KO mice display a lower level of inflammation and improved glucose metabolism relative to EIIARORα WT mice when they are challenged with an HFD. The expression of several genes encoding lipogenic enzymes was analyzed by qPCR, but no differences were observed between EIIARORα WT and EIIARORα KO animals (Figure 4E). Since EIIARORα KO mice are more sensitive to insulin, we assessed the role of RORα deletion in skeletal muscle. The expression of several genes involved in glucose and insulin metabolism was assessed by qPCR, and no major differences were observed between EIIARORα WT and EIIARORα KO except for Akt expression (Figure 4F).

Risk for development of metabolic disorders increases with age, thus we decided to study the effect of RORα deletion in aging. We used 15-month-old male EIIARORα KO and EIIARORα WT mice maintained on a normal chow diet. No differences in body composition were observed between the two groups (Figure 5A). To determine whether old-EIIARORα KO mice exhibited increased energy expenditure, we measured activity, heat production, as well as the RER over a 5-day period. No changes were observed in activity, but we observed a decrease in RER suggesting an increase in fat utilization during the resting period (day time) as well as an increase in heat production (Figure 5A). These data suggest that EIIARORα KO mice have a higher rate of energy expenditure and fatty acid oxidation. We also observed that old-EIIARORα KO mice were more glucose tolerant than old-EIIARORα WT mice (Figure 5B). When we challenged these mice with insulin, old-EIIARORα KO mice were more sensitive than the WT mice (Figure 5B). No differences were observed in fasted blood glucose levels (Figure S3 in Supplementary Material). Aging is also associated with increased pro-inflammatory cells and cytokines, thus we studied the proportion of different lymphocytes cell populations in the spleens of these mice by flow cytometry. An increase in anti-inflammatory T_H2 lymphocytes (CD4⁺CD3⁺Gata3⁺) was observed in old-EIIARORα KO mice compared to old-EIIARORα WT mice as well as a decrease of pro-inflammatory lymphocytes T_H17 (CD4⁺CD3⁺IL-17⁺; Figure 5C). No changes were observed in number of T_H1 lymphocytes or T_reg lymphocytes between the genotypes (Figure S3 in Supplementary Material). In WAT, a decrease in TNFα expression and an increase the expression of anti-inflammatory genes such as TGFβ and macrophage infiltration marker CD36 were observed in old-EIIARORα KO mice compared to old-EIIARORα WT mice (Figure 5D). These results suggest that deletion of RORα can be protective in age-induced metabolic disorders such as diabetes and obesity.