Mouse strains to study cold-inducible beige progenitors and beige adipocyte formation and function

Abstract

Cold temperatures induce formation of beige adipocytes, which convert glucose and fatty acids to heat, and may increase energy expenditure, reduce adiposity and lower blood glucose. This therapeutic potential is unrealized, hindered by a dearth of genetic tools to fate map, track and manipulate beige progenitors and 'beiging'. Here we examined 12 Cre/inducible Cre mouse strains that mark adipocyte, muscle and mural lineages, three proposed beige origins. Among these mouse strains, only those that marked perivascular mural cells tracked the cold-induced beige lineage. Two SMA-based strains, SMA-Cre(ERT2) and SMA-rtTA, fate mapped into the majority of cold-induced beige adipocytes and SMA-marked progenitors appeared essential for beiging. Disruption of the potential of the SMA-tracked progenitors to form beige adipocytes was accompanied by an inability to maintain body temperature and by hyperglycaemia. Thus, SMA-engineered mice may be useful to track and manipulate beige progenitors, beige adipocyte formation and function.

Figure 1. Cre drivers with high beige adipocyte labelling.

(a–m) Two-month-old male Cre mice (AdipoTrak-Cre and SM22-Cre); R26R^RFP mouse models (b,c,f,g) were maintained at RT (23 °C) or cold exposed (6 °C) for 7 days as shown in a. Subcutaneous inguinal adipose depot explants were imaged for direct RFP fluorescence or sectioned and immunostained for reporter (RFP, red) and UCP1 (green); co-expression (RFP+UCP1) is a yellow-gold hue. Cell nuclei were counterstained with DAPI (blue). Two-month-old male, inducible Cre mice (NG2-Cre^ERT2; SMA-Cre^ERT2 and SMA-rtTA-TRE-Cre); R26R^RFP mice (h–m) were administered one dose of TM on 2 consecutive days or administered Dox 3 days before experimentation. Mice were maintained at RT or cold exposed for 7 days as shown in a. Subcutaneous inguinal adipose depot explants were imaged or were sectioned and immunostained for reporter (RFP, red) and UCP1 (green). For AdipoTrak-GFP suppression and reactivation studies, AdipoTrak-GFP mice were maintained on Dox from conception until 2 months of age (a,d,e). Dox was removed for 2 weeks and subsequently mice were maintained at RT or cold exposed for 7 days off Dox as shown in a. Subcutaneous inguinal adipose depot explants were imaged or sectioned and immunostained for reporter (GFP, false coloured red) and UCP1 (green) (e). Quantification of total UCP1+ beige adipocytes that were RFP+ is indicated to the right of the images. Images are representative from n=4 mice per group replicated twice. Scale bar, 10 μm (b,d,f,h,j,l). Scale bar, 100 μm (c,e,g,i,k,m).

Figure 2. Cre models with low beige adipocyte labelling.

(a–h) Two-month-old male Cre mice (Myf5-Cre and Myogenin (MyoG)-Cre); R26R^RFP mouse models were maintained at RT (23 °C) or cold exposed (6 °C) for 7 days. Subcutaneous inguinal adipose depot explants were imaged for direct RFP fluorescence or sectioned and immunostained for reporter (RFP, red) and UCP1 (green) (a–d). Two-month-old male, inducible Cre mice (PDGFRα-Cre^ERT2 and Myh11-Cre^ERT2); R26R^RFP mice were administered one dose of TM on 2 consecutive days. Mice were maintained at RT or cold exposed for 7 days. Subcutaneous inguinal adipose depot explants were imaged or sectioned and immunostained for reporter (RFP, red) and UCP1 (green) (e–h). (i,j) Myh11-Cre^ERT2; R26R^RFP were also housed in the cold or RT for 14 days. Subcutaneous inguinal adipose depot explants were imaged or sectioned and immunostained for reporter (RFP, red) and UCP1 (green). Quantification of UCP1+ beige adipocytes that were RFP+ is denoted next to the IHC image. Images are representative from n=4 mice per group replicated twice. Scale bar, 10 μm (a,c,e,g,i). Scale bar, 100 μm (b,d,f,h,j).

Figure 3. Existing white adipocytes do not contribute to beiging.

(a) Two-month old C57BL/6J mice were maintained at RT or cold temperature for 7 days. Subcutaneous inguinal adipose depots were haematoxylin and eosin (H&E) stained and IHC stained for UCP1. Insets are magnified images. Scale bar, 100 μm (in both images). (b–d) Two-month-old Adiponectin-Cre^ERT2; RFP (b), aP2-Cre^ERT2: RFP (c) or UCP1-Cre^ERT2; RFP (d) male mice were administered one dose of TM for 2 consecutive days and examined (pulse) or mice were maintained at RT for 7 days (TM washout period). Subsequently, mice were cold exposed for 7 days. Subcutaneous inguinal adipose depots were sectioned and immunostained for RFP (red) and UCP1 (green). Cell nuclei were visualized by DAPI staining. Quantification of UCP1+ beige adipocytes that were RFP+ is denoted next to the IHC image. Images are representative from n=3 mice per group replicated twice. Scale bar, 100 μm.

Figure 4. AdipoTrak and Myf5 model characterization.

(a) Merged (H2B-GFP and RFP) whole-mount images of denoted adipose depots and other tissues from 2-month-old AdipoTrak (PPARγ^τTA,TRE-Cre); TRE-H2B-GFP; R26R^RFP male mice and control (PPARγ^tTA; RFP) mice. BAT, classical brown adipose tissue; IGW, inguinal; MWAT, mesenteric white adipose tissue; PGW, perigonadal; PSCW, periscapular; RPW, retroperitoneal. Scale bar, 10 μm. (b) Image of control (R26R^RFP) and Myf5-Cre; RFP male mice. Please note the RFP fluorescence was easily observed in regular room lighting from the Myf5-Cre; RFP mouse. (c) Fluorescence microscopic whole-mount images of RFP fluorescence from denoted adipose depots from 2-month-old Myf5-Cre; RFP RT mice. Scale bar, 10 μm. (d) Two-month-old Myf5-Cre; RFP male mice were maintained at RT or cold exposed for 7 days. Histological sections from periscapular (top: PSCW) and retroperitoneal (bottom: RPW) adipose depots were examined for RFP fluorescence and UCP1 (green) expression. Cell nuclei were counterstained with DAPI (blue). Quantification of UCP1+ beige adipocytes that were RFP+ is denoted next to the IHC image. (e) Adipose depot cells from 2-month-old Myf5-Cre; RFP mice were analysed by flow cytometry for co-expression of RFP with the indicated genes. (f) Adipose depot cells from 2-month-old Myf5-Cre; RFP male mice were separated into RFP+ and RFP− fractions using FACS and examined for mRNA expression of denoted genes using qPCR. ND, not detected. Data are means±s.e.m. (n=4–5 mice per group replicated twice). Scale bar, 100 μm. Student's t-test, *P<0.05, RFP+ compared with RFP−.

Figure 5. SMA-Cre^ERT2 characterization.

(a) Subcutaneous inguinal adipose depots from RT 2-month-old wild-type male mice were immunostained for PECAM, SMA and perilipin, and nuclei detected with DAPI. (b) Subcutaneous inguinal adipose depots from RT 2-month-old wild-type male mice were IHC stained for endogenous SMA. (c) Subcutaneous inguinal adipose depots from 2-month-old uninduced (no TM) SMA-Cre^ERT2; RFP male mice were immunostained for RFP, SMA and perilipin, and nuclei stained with DAPI. (d) Subcutaneous inguinal adipose depot from 2-month-old pulsed SMA-Cre^ERT2; RFP male mice were immunostained for RFP, SMA and perilipin, and nuclei highlighted with DAPI. (e) SV cells and adipocytes were isolated from 2-month-old TM-pulsed SMA-Cre^ERT2; RFP male mice and mRNA expression of endogenous SMA and SMA-driven RFP were assessed by qPCR. Data are means±s.e.m. (n=4–5 mice per group replicated thrice). Student's t-test, #P<0.001, SVF compared with adipocyte compartment. (f) SV cells were isolated from 2-month-old TM-pulsed SMA-Cre^ERT2; RFP male mice and visualized for RFP and lipid (green; LipidTox green). (g) Adipocytes were floatation isolated from 2-month-old TM-pulsed induced SMA-Cre^ERT2; RFP male mice and visualized for RFP and lipid (green). (h) Adipocytes were isolated from 2-month-old uninduced SMA-Cre^ERT2; RFP male mice. TM was then administered ex vivo for 24 h and the floated adipocytes examined for RFP and lipid (green). (i) Two-month-old SMA-Cre^ERT2; RFP male mice were administered one dose of TM for 2 consecutive days. Mice were maintained at RT for 14 days (TM washout). Mice were subsequently cold exposed for 7 days. Subcutaneous inguinal adipose depots were immunostained for RFP (red) and UCP1 (green). Cell nuclei were counterstained with DAPI (blue). Quantification of UCP1+ beige adipocytes that were RFP+ is denoted next to the IHC image. Images are representative from n=4 mice per group replicated twice. Scale bar, 100 μm.

Figure 6. SMA-rtTA characterization.

(a) Subcutaneous inguinal adipose depots from 2-month old uninduced (no Dox) SMA-rtTA; RFP (SMA-rtTA; TRE-Cre; R26R^RFP) male mice maintained at RT were immunostained for RFP, SMA and perilipin. Cell nuclei were counterstained with DAPI (blue). (b) Subcutaneous inguinal adipose depot sections from 2-month old Dox-pulse SMA-rtTA; RFP male mice maintained at RT were immunostained for RFP, SMA and perilipin; nuclei were highlighted with DAPI. (c) SV cells and adipocytes were isolated from 2-month old pulse SMA-rtTA; RFP male mice maintained at RT. mRNA expression of RFP were assessed by qPCR. Data are means±s.e.m. (n=4–5 mice per group). Students t-test, #P-value <0.001, SVF compared with adipocyte compartment. (d) Subcutaneous inguinal adipose depots from 2-month old pulsed SMA-rtTA; RFP male mice maintained at RT were sectioned and immunostained for RFP and PECAM (green), and nuclei stained with DAPI. (e) Adipocytes were isolated pulse SMA-rtTA; RFP mice described in d and visualized for RFP and lipid (green). (f) Adipose depot cells of pulse SMA-rtTA; RFP mice as described in d were analysed for co-expression of RFP and SMA. (g) RFP− and RFP+ cells were FACS isolated from pulse SMA-rtTA; RFP mice and SMA mRNA expression was assessed using qPCR. (h) Adipose cells of pulse SMA-rtTA; RFP mice were flow analysed for co-expression of RFP with mural or endothelial markers. (i) RFP− and RFP+ cells were FACS isolated from pulse SMA-rtTA; RFP mice and mRNA expression of various adipose progenitor, mural and endothelial markers were assessed by qPCR. Data are means±s.e.m. (n=5 mice per group replicated twice). Images are representative from n=4 mice per group replicated twice. Scale bar, 100 μm. Students t-test, *P-value <0.01, RFP+ compared with RFP−.

Figure 7. Endogenous SMA and SMA-driven reporter expression profile in beige adipocytes.

(a) Subcutaneous adipose depots from 2-month-old, cold-exposed wild-type mice were immunostained for endogenous SMA and UCP1 expression, and nuclei detected with DAPI. (b) Subcutaneous adipose depots from 2-month-old, cold-exposed wild-type male mice were IHC stained for endogenous SMA and UCP1. (c,d) Two-month-old uninduced (no TM) UCP1-Cre^ERT2 (c) or Adiponectin-Cre^ERT2 (d) male mice were cold exposed. After 7 days of cold exposure, mice were administered TM and maintained at RT for 24 h. Subcutaneous inguinal adipose depots were immunostained for RFP and UCP1 (green); nuclei were highlighted with DAPI. (e) Subcutaneous adipose depots from cold-exposed uninduced SMA-Cre^ERT2; RFP were immunostained for RFP and UCP1 (green); nuclei were highlighted with DAPI. (f) Uninduced SMA-Cre^ERT2; RFP mice were cold exposed for 7 days; mice were then pulsed with TM and housed at RT for 24 h. Sections from the subcutaneous inguinal adipose depots were immunostained for RFP and UCP1 (green). Cell nuclei were counterstained with DAPI (blue). (g) Uninduced SMA-rtTA; RFP mice were exposed to cold for 7 days. Subcutaneous inguinal adipose depot were sectioned and immunostained for RFP and UCP1 (green). Cell nuclei were counterstained with DAPI (blue). (h) Uninduced SMA-rtTA; RFP mice were cold exposed for 7 days and then administered Dox for 24 h and maintained at RT. Subcutaneous inguinal adipose depot sections were analysed for RFP (red), UCP1 (green) and DAPI (blue).

Figure 8. SMA+ mural cells are required for beige adipocyte formation.

(a–d) Two-month-old SMA-Cre^ERT2; PPARγ^fl/fl or SMA-Cre^ERT2; R26R^DTA were administered one dose of TM for 2 consecutive days. Mice were then randomized to RT or cold exposed. Seven days later, mice were analysed for beige adipocyte formation by the following: rectal temperature (a), blood glucose levels (b), haematoxylin and eosin (H&E) staining (c), UCP1 IHC (d) and mRNA expression of beige and thermogenic markers (e). (f) SV cells were isolated from TM-induced SMA-Cre^ERT2 or SMA-Cre^ERT2; PPARγ^fl/fl mice and incubated in beige adipogenic culture conditions. Triglyceride content and mRNA expression of beige and thermogenic genes were analysed to determine adipocyte differentiation. Data are means±s.e.m. (n=4 mice per group replicated thrice). Students t-test, *P-value <0.05, SMA-Cre^ERT2; PPARγ^fl/fl or SMA-Cre^ERT2; R26R^DTA compared with control. Students t-test, #P-value <0.05, cold control compared with RT control. Students t-test, **P-value <0.01, control compared with undifferentiated SV cells. (g–k) Two-month-old Myh11-PPARγ male mice were administered one dose of TM for 2 consecutive days. Mice were then randomized to RT or cold for 1 or 2 weeks. Mice were analysed for beige adipocyte formation by the following: rectal temperature (g), blood glucose levels (h), H&E staining (i), UCP1 IHC (j) and mRNA expression of beige markers (k). Data are means±s.e.m. (n=4 mice per group replicated thrice. Scale bar, 100 μm. (l) SV cells were isolated from TM-induced Myh11Cre^ERT2; PPARγ^fl/fl mice and incubated in adipogenic conditions. Triglyceride content and mRNA expression of beige and thermogenic genes were analysed to determine adipocyte differentiation. Data are means±s.e.m. (n=4 mice per group replicated twice).