C57BL/6J, staggerer Rora spontaneous mutant (JAX strain number 002651; Rora^sg/sg) and B6.SJL-Ptprc^aPepc^b/BoyJ (JAX strain number 002014; CD45.1⁺) mice were purchased from Jackson Laboratories (Bar Harbor, MD, USA). Adult 6–10 weeks old female mice were used in all experiments. Conditional Rora floxed (Rora^fl/fl; Lexicon Pharmaceuticals, USA) and Rora^sg/sg mice were crossed with Lyz2^tm1(cre)Ifo (JAX strain number: 004781; LysM^Cre) mice to generate LysM^CreRora^fl/sg animals with a conditional deletion of Rora in cells of a myeloid lineage. LysM^CreRora^fl/sg were used due to the variable levels of Lyz2 in different macrophage populations that may result in incomplete excision as seen in LysM^CreIL-4Rα^flfl mice (15). Animals were housed in a specific pathogen-free facility in individually ventilated and filtered cages under positive pressure. All animal experiments were performed in compliance with the Irish Medicines Board (Figures 1, 2) or Health Product's Regulatory Authority (Figures 3–5) and approved by the Trinity College Dublin's BioResources ethical review board.

It is relevant that due to the importance of RORα in regulating circadian rhythm (16), and the corresponding circadian rhythm known to govern circulating monocytes and thus potentially impacting upon the inflammatory response (17), all experiments on mice and tissue isolations were performed at 10 a.m. to ensure no variations occurred due to alterations in the circadian rhythm due to time differences in experiments.

Endotoxic shock was induced in age-matched male C57BL/6J, Rora^sg/sg and LysM^CreRora^fl/sg mice by intraperitoneal injection of ultra-pure lipopolysaccharide (UP-LPS) isolated from Escherichia coli (Invivogen, France) at a dose of 10 mg/kg, as previously described (18). At 3 h post-injection, blood was collected via submandibular bleed for serum cytokine quantification. Temperature was monitored throughout using subcutaneously implanted temperature transponder chips (Bio Medic Data Systems; IPTT 300). Mice were visually scored, using a scale from 1 to 10, at regular intervals, with criteria dependent upon a combination of body temperature, appearance, condition, and behavioral characteristics. The scoring system was: Score 1, <1°C reduction in body temperature, no other change in condition; Score 2, <1°C reduction in body temperature accompanied by reduced interest in surroundings; Score 3, 1–2°C reduction in body temperature accompanied by a reduced interest in surroundings; Score 4, 1–2°C reduction in body temperature accompanied by piloerection; Score 5, 1–2°C reduction in body temperature accompanied by piloerection and reduced interest in surroundings; Score 6, 1–2°C reduction in body temperature accompanied by piloerection and reduced mobility; Score 7, >2°C reduction in body temperature accompanied by piloerection and reduced mobility; Score 8, >2°C reduction in body temperature accompanied by piloerection and nasal and ocular discharge and reduced mobility; Score 9, >2°C reduction in body temperature accompanied by piloerection and nasal and ocular discharge and vocalization; Score, 10 animals that are moribund. All mice were culled at humane end-points or after 48 h (Figures 1, 2) or 24 h (Figures 4, 5).

Clinical scores were performed blinded with individual mice tracked by the transponder chip code. Rora^sg/sg mice are visually ataxic and stunted and therefore blinding was not practical. All further analysis of cells and serum was performed blinded.

In addition, a lower dose (0.2 mg/kg) of UP-LPS was used to induce cell trafficking into the peritoneal cavity. Groups of age-matched C57BL/6J and Rora^sg/sg mice were injected intraperitoneally with 0.2 mg/kg UP-LPS and peritoneal cells collected by lavage after 3 h. Cell expression was quantified by flow cytometry.

Peritoneal macrophages were cultured from naive C57BL/6J and Rora^sg/sg mice as previously described (18). Briefly, the peritoneal cavity was lavaged with 5 ml ice-cold PBS and the resulting cells plated at 2 × 10⁶ cells/ml in RPMI 1640 supplemented with 2 mM L-glutamine, 100 U/ml penicillin and 100 μg/ml streptomycin and incubated for 2 h at 37°C with 5% CO₂, and any non-adherent cells removed. Adherent cells were stimulated with media-alone or UP-LPS (200 ng/ml) for 4 h, after which culture supernatants were collected for cytokine analysis.

Bone marrow-derived macrophages (BMDM) were cultured from the tibia and fibula of naive C57BL/6J and Rora^sg/sg mice as previously described (18). Briefly, cells were flushed from the tibia and fibula to prepare a single cell suspension. After red blood cell lysis, the resultant cells were cultured at 3 × 10⁶ cells/ml in RPMI 1640 supplemented with 2 mM L-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin and 10% fetal calf serum, macrophage colony-stimulating factor (M-CSF) from L929 mouse fibroblast supernatants at 37°C with 5% CO₂ for 7 days. Cells were stimulated with media alone or UP-LPS (200 ng/ml) for 4 h, after which culture supernatants were collected for cytokine analysis.

CD45.1⁺ C57BL/6 recipient mice were irradiated using an X-Ray irradiator (XStrahl CIX3), receiving 9 Gy in two doses (5 Gy and 4 Gy) 3 h apart. Mice were reconstituted with 1 × 10⁷ BM cells from CD45.2⁺ C57BL/6 mice or Rora^sg/sg mice. To ensure efficient irradiation and reconstitution, expression of CD45.1 vs. CD45.2 was assessed by flow cytometry of the spleen after 8 weeks (Supplementary Figure 1).

Mice were treated with SR3335 (N-[4-[2,2,2-trifluoro-1-hydroxy-1-(trifluoromethyl)ethyl]phenyl]-2-thiophenesulfonamide; Cayman) (Cambridge Biosciences, UK) a synthetic inverse agonist of RORα (19). C57BL/6 mice were treated with vehicle control, or SR3335 at a dose of 15 mg/kg per day i.p. for 7 days (19). On day 8 all animals were injected with UP-LPS i.p. as described above.

Peritoneal macrophages and BMDM, prepared as above, were analyzed with an XF-24 Extracellular Flux Analyzer (Seahorse Biosciences, Agilent Technologies) to determine oxygen consumption rate (OCR) and extracellular acidification rate (ECAR). Cells cultured at a density of 5 × 10⁵ cells/ml were incubated with non-buffered RPMI supplemented with 25 mM glucose with no CO₂ for 1 h prior to analysis. Three consecutive measurements were taken at basal conditions and then after the sequential addition of 1 μM oligomycin, 1.5 μM FCCP (fluoro-carbonyl cyanide phenylhydrazone) and 1.25 μM rotenone plus 2.5 μM antimycin A. The OCR and ECAR were measured basally and after the addition of each inhibitor. The spare respiratory capacity (SRC) is determined as the difference between the basal OCR and maximal OCR, after the addition of FCCP (20).

Cytokine and chemokine levels were quantified in tissue culture supernatants and serum by sandwich ELISA. IL-6 was quantified using matched antibody pairs from BD Biosciences (Oxford, UK), IL-1β, TNFα, IL-10, GROα, and CCL2 were quantified using the DuoSet ELISA development system from R&D Systems (Abingdon, UK) following the manufacturer's protocol.

Surface marker expression was assessed by flow cytometry with data collection on a CyAn ADP (Beckman Coulter, High Wycombe, UK) and data analyzed using FlowJo software (Tree Star, OR USA). Cells were isolated from the peritoneal cavity by lavage with ice-cold PBS. Cells were stained with BD Biosciences (Oxford, UK) mAbs; F4/80-FITC (BM8), CD45.2-PE-CF594 (104), SiglecF-AlexaFluor647 (E50-2440), CD3-PECF594 (145-2C11); eBiosciences (Dublin, Ireland) mAbs; CD11b-PerCP (M1/70), F4/80-eFluor450 (BM8), CD5-APC (53-7.3), MHC class II-eFluor 450 (MS/114.15.2); and BioLegend (London, UK) mAbs; Ly6G-APCCy7 (1A8), CD19-PerCP (6D5). Prior to surface staining, cells were incubated with LIVE/DEAD Fixable Aqua stain (Molecular probes, Invitrogen, Dublin, Ireland) to isolate dead cells. Using appropriate controls, quadrants were drawn and data were plotted on logarithmic scale density-plots.

Statistical analysis was performed using GraphPad InStat®. Results are presented as mean +/- SEM. Differences, indicated as two-tailed P-value, were considered significant when P>0.05 as assessed by unpaired Student's t-test with Welch correction applied as necessary. Survival statistics were calculated using log-rank correlation. Area Under Curve (AUC) for the clinical scores of individual mice were determined and difference between group clinical score AUC were analyzed by Student's t-test.

RORα has been widely implicated in the control of inflammatory signaling, with many studies demonstrating an anti-inflammatory role for RORα through suppression of IκB, a negative regulator of the NFκB pathway (3, 12). RORα is also integral in the generation of a functional type 2 immune response, through expression in ILC2 (21). To further explore roles for RORα in inflammation Rora^sg/sg mutant mice, which produce a truncated form of the RORα protein, were challenged with LPS to induce endotoxic shock. Mutant mice had significantly improved survival (P < 0.01) after LPS challenge compared to WT C57BL/6 mice (Figure 1A), which was associated with a significant (P < 0.01) reduction in clinical signs of endotoxemia (Figure 1B).

Assessment of serum cytokines and chemokines in response to LPS treatment shows significantly (P < 0.05–0.0001) decreased levels of the pro-inflammatory IL-1β and IL-6 and the monocyte chemoattractant CCL2 in Rora^sg/sg mice relative to WT animals (Figure 1C). Interestingly, there was no significant reduction in the serum expression of IL-10, TNFα or the neutrophil chemoattractant KC/GRO in Rora^sg/sg mice compared to WT animals (Figure 1C). These data show RORα deficient mice have reduced susceptibility to LPS-induced endotoxic shock, with selective alteration in production of pro-inflammatory cytokines and chemokines.

To further assess the in vivo cellular response to LPS, mice were administered a lower dose (0.2 mg/kg) to elicit local activation and cellular recruitment to the peritoneal cavity. In WT animals, the cellular profile in response to LPS can be characterized by an infiltration of neutrophils (CD11b⁺Ly6G⁺; Supplementary Figure 2B), B1a cells (CD19⁺CD5⁺; Supplementary Figure 2D) and small peritoneal macrophages (SPM; CD11b⁺F4/80⁺MHCII^hi; Supplementary Figure 2C) which temporarily replace the resident large peritoneal macrophage population (LPM; CD11b⁺F4/80^hiMHCII^lo; Supplementary Figure 2C). There were no differences in neutrophils and B1a cells in the peritoneum of naive WT and Rora^sg/sg mice, with comparable influx of both cells following LPS treatment (Figure 2A). However, while there was no differences in the number of resident LPM in naïve WT and Rora^sg/sg mice, there were significantly (P < 0.01) fewer SPM in the peritoneal cavity of naïve Rora^sg/sg mice (Figure 2B), in accordance with previously published studies (8). In response to LPS the influx of SPM in the peritoneal cavity was comparable between WT and Rora^sg/sg animals (Figure 2B). Whereas, there was a significant decrease in the number of LPM in the peritoneal cavity of Rora^sg/sg mice compared to WT animals after intraperitoneal LPS treatment (Figure 2B).

Alterations in recruited monocytes has also been observed in the adipose tissue, with myeloid-derived macrophages pro-inflammatory macrophages significantly decreased in the absence of RORα (8). In both the obesity model and in response to LPS, there is reduced response to the inflammatory stimuli in the absence of RORα. The response to inflammatory stimuli can be influenced by the metabolic state of the macrophage populations themselves, with inflammatory macrophages relying on aerobic glycolysis while anti-inflammatory macrophages, like utilizing fatty acid oxidation to fulfill their energy requirements, reflecting M1-like and M2-like cells, respectively (20, 22). To further examine the differences in the macrophage populations in Rora^sg/sg mice, the metabolic status of peritoneal macrophages isolated from naive WT and Rora^sg/sg mice were assessed by extracellular flux analysis. We compared the oxygen consumption by unstimulated peritoneal macrophages from WT and Rora^sg/sg mice and show enhanced OCR and increased SRC in macrophages isolated from Rora^sg/sg mice compared to WT macrophages (Figure 2C), which is indicative of increased oxidative phosphorylation, as associated with M2-like macrophages (20). Conversely, macrophages isolated from WT mice showed significantly (P < 0.01) increased ECAR, which is indicative of increased reliance upon anaerobic glycolysis and a switch toward an M1-like phenotype (Figure 2C). Additionally, comparison between the ratio of OCR to ECAR shows a significantly (P < 0.05) altered commitment toward oxidative phosphorylation and anaerobic glycolysis in cells from Rora^sg/sg mice (Figure 2C). The characterization of the metabolic profile of peritoneal macrophages from Rora^sg/sg mice suggests a reduction in an M1 phenotype in the absence of RORα.

These data indicate that in the absence of functional RORα mice develop alterations in the activation or recruitment of macrophages within the peritoneum. Resident LPM and monocytes recruited to the peritoneal cavity and SPM have different origins, with resident macrophages derived from an embryonic precursor and maintained by self-proliferation and renewal, while circulating monocytes are myeloid-derived (23–25). The observation that solely the myeloid-derived macrophage population are altered in naive Rora^sg/sg mice suggests that during the embryonic stage where the resident cells are initially seeded, there is no apparent role for RORα, however, RORα does appear to impact on the recruitment and polarization of infiltrating monocytes into the peritoneal cavity. Indeed, circulating levels of CCL2, a chemokine well-defined as a recruitment signal for monocytes (26), are significantly decreased in Rora^sg/sg mice (Figure 1C).

To further address the diminished inflammatory response in Rora^sg/sg mice, in vitro assessment of macrophage populations was undertaken. There was a selective reduction in cytokine release from peritoneal macrophages isolated from naive Rora^sg/sg mice (Figure 3A), which mirrored the in vivo cytokine responses. As the altered macrophage profile was shown in myeloid-derived macrophage populations in ex vivo conditions, the metabolic profile of BMDM from WT and Rora^sg/sg mice was assessed. While macrophages from both strains display an OCR profile associated with oxidative phosphorylation as indicative of an alternatively activated phenotype—which would be expected as these macrophages were expanded in vitro with M-CSF—the BMDM from Rora^sg/sg mice showed an increased SRC and decreased ECAR relative to cells from WT mice which is in agreement with a bias toward an anti-inflammatory phenotype in the absence of RORα (Figure 3B). This metabolic analysis suggesting an inherent bias toward an anti-inflammatory phenotype in the absence of Rora, is interesting supporting previous studies have focused on the anti-inflammatory role of RORα (3, 6, 10, 12, 27). While the signaling pathways elicited by RORα are anti-inflammatory, these data suggest that in the absence of functional RORα there is an alteration in the myeloid compartment that results in a reduction in pro-inflammatory macrophages, which results in the observed diminished pro-inflammatory response.

In mutant mice, the ubiquitous deletion of functional RORα results in an amelioration of inflammation and a reduction and delay in LPS-associated morbidity (Figure 1). However, due to the wide-ranging roles of RORα in cerebellar development and lipid and glucose metabolism, the mutant Rora^sg/sg mice are stunted in growth, ataxic, and show increased mortality (28). To circumvent the impact of ubiquitous deletion of RORα, BMC mice were generated using bone marrow isolated from Rora^sg/sg mice, with a control BMC reconstituted with bone marrow from WT mice. It should be noted that Rora^sg/sg BMC and WT BMC were phenotypically comparable. Rora^sg/sg BMC had significantly (P < 0.05) decreased mortality (Figure 4A) and improved clinical score (Figure 4B) in response to LPS-induced endotoxic shock relative to WT BMC. Furthermore, the reduced sensitivity to LPS was associated with a concomitant decrease in pro-inflammatory cytokines and chemokines in serum of Rora^sg/sg BMC mice (Figure 4C).

To further explore roles for RORα endotoxic shock, mice were treated with SR3335, a synthetic RORα selective inverse agonist (19), for 7 days prior to the induction of LPS-induced endotoxic shock. Treatment of mice with SR3335 significantly improves mortality (Figure 4D) and clinical score (Figure 4E) from LPS-induced shock confirming that in this model blocking RORα reduces the severity of disease.

To validate the role of Rora expressing macrophages in the inflammatory response, we generated LysM^CreRora^fl/sg mice, where the Rora gene is excised in cells expressing the Lyz2 gene, which includes macrophages. When Rora^fl/sg or LysM^CreRora^fl/sg mice were injected with LPS, LysM^CreRora^fl/sg mice had significantly (P < 0.01) increased survival (Figure 5A), associated with improved clinical score (Figure 5B) when compared to Rora^fl/sg animals. In addition, mice with myeloid deletion of Rora had significantly (P < 0.05) reduced elevations in serum levels of IL-1β, IL-6, and CCL2, but not IL-10, TNF-a or KC/GRO after LPS treatment (Figure 5C). These results are all comparable to the disease progression and severity observed following LPS treatment of Rora^sg/sg and Rora^sg/sg BMC mice, suggesting that it is macrophage expression of Rora that is responsible for the impact of RORα on the progression of inflammation in this acute model.