A bone-specific adipogenesis pathway in fat-free mice defines key origins and adaptations of bone marrow adipocytes with age and disease

Figures

• 

  Figure 1.. Adiponectin is expressed by AdipoqCre+/mTmG+ bone marrow stromal cells in vitro. (A) Bone marrow stromal cells from 16 week old male AdipoqCre+/mTmG+ mice were cultured at low density for 14 days to promote formation of colony-forming units (CFUs). Endogenous fluorescence was then amplified by immunostaining for green fluorescent protein (GFP) and red fluorescent protein (RFP). In addition, spontaneous adipogenesis was assessed based on immunostaining for perilipin 1 (PLIN1, pink). Scale = 1 mm. (B) Quantification of GFP and RFP expression in fibroblastic stromal cells within each culture dish (n = 3 independent mice, 166 total colonies counted). Data presented as mean ± SD. One-way ANOVA. *p≤0.05. (C) Representative mixed colony with both GFP+ and RFP+ fibroblasts demonstrating GFP+ perinuclear granules in red cells (white arrowheads), indicating upregulation of adiponectin expression. (D) Day 14 adipogenic colony demonstrating PLIN1+ lipid droplets (pink, white arrows) in GFP+ adipocytes. Nearby RFP+ fibroblasts show early signs of conversion (GFP+ perinuclear granules, white arrowheads). (E) Day 14 adipogenic colony with uniformly GFP+ stromal cells and PLIN1+ adipocytes. (F)…

• 

  Figure 1—figure supplement 1.. Adiponectin is expressed by BMAT adipocytes in the AdipoqCre+/mTmG+ mouse in vivo. (A) Representative longitudinal cross-section of the proximal tibial metaphysis and epiphysis at the region of the growth plate (*). Bone marrow adipocytes (BMAds) were identified by expression of perilipin (PLIN1, pink, arrowheads). BMAds were positive for green fluorescent protein (GFP). GFP expression was also noted in cells lining the bone surface (arrows) and in cells of the stromal reticular network. Four month old, male AdipoqCre+/mTmG+. Scale = 100 µm. (B) Four month old, male AdipoqCre-/mTmG+ negative control. All cells, including BMAds, were positive for red fluorescent protein (RFP). Scale = 100 µm. (C) Representative proximal tibia, 4 month old, female AdipoqCre+/mTmG+. As in males, regulated bone marrow adipose tissue (rBMAT) adipocytes are GFP+. Prominent labeling is also observed in the stromal reticular network and on the bone surface. Scale = 50 µm. (D) Representative tail vertebrae, 3 week old male AdipoqCre+/mTmG+. Constitutive BMAT adipocytes (cBMAT) are GFP+ (arrowheads). GFP expression is absent in the stroma and on the bone surface. Scale = 50 µm. Images representative of the…

• 

  Figure 2.. AdipoqCre+/DTA+ fat free (FF) mice lack white and brown adipose tissues and circulating adiponectin. (A) Serum adiponectin (ACRP30) of AdipoqCre-/DTA+ control (Con), AdipoqCre+/DTA+ fat free (FF), and Adipoq knockout (KO) mice by western blot. Blood albumin levels by Ponceau S staining (loading control). (B) Representative pictures showing the absence of white and brown adipose tissues (white arrowheads) and the enlarged liver (black arrows) in 16 week old male FF mice relative to control. Identical gross phenotypes were observed in females (data not shown). (C) Representative hematoxylin and eosin (H&E) stained sections of inguinal white adipose tissue. Areas of adipocytes have been replaced by loose fibrous tissue in FF mice. Ad = adipocytes. LN = lymph node. Inset scale = 50 µm. (D) Representative (H&E) stained sections of liver. CV = central vein. Scale = 50 µm. (E) Random fed blood glucose, measured using a glucometer. (F) End point tibia lengths, measured using a caliper. (G) Body mass. Sample size for control and FF mice, respectively: 4 months Male n = 5, 6, Female n = 8, 5; and 8 months Male n = 4, 5; Female n = 3, 3. (E) Two-tailed t-test, (F, G) two-way ANOVA with…

• 

  Figure 3.. Trabecular bone is increased in AdipoqCre+/DTA+ fat free (FF) mice. (A) Representative 3D µCT-based reconstructions of tibiae from 4 month old AdipoqCre+/DTA+ fat free (FF) and AdipoqCre-/DTA+ controls (Con). Scale = 1 mm. (B–F) Quantification of trabecular parameters in the proximal tibial metaphysis. Region of interest as indicated in (A). (B) Trabecular bone volume fraction (Tb. BVF). (C) Trabecular number. (D) Trabecular thickness. (E) Trabecular spacing. (F) Trabecular bone mineral density (Tb. BMD). Sample size for control and FF mice, respectively: 4 months Male n = 5, 6, Female n = 8, 5; and 8 months Male n = 4, 5; Female n = 3, 3. Statistical significance was assessed by two-way ANOVA with Tukey’s multiple comparisons test. ANOVA results as indicated. *P ≤ 0.05. Data presented as mean ± SD. WT and FF mice were housed at 30°C on a 12 hr/12 hr light/dark cycle. Figure 3—source data 1.Trabecular bone parameters.

• 

  Figure 3—figure supplement 1.. Cortical parameters are increased in young AdipoqCre+/DTA+ fat free (FF) mice, but tends to decline with age. (A) Representative ex vivo micro-CT cross sections of 4 month old AdipoqCre+/DTA+ fat free (FF) and AdipoqCre-/DTA+ control (Con) cortical bone at 6 mm, 8 mm, and 12 mm distal to the proximal end of the tibia. (B–H) Quantification of cortical bone parameters. (B) Cortical thickness. (C) Cortical bone volume fraction (Ct. BVF). (D) Cortical medullary area. (E) Cortical total area (medullary+ bone area). (F) Cortical bone area. (G) Cortical bone mineral content (Ct. BMC). (H) Polar moment of inertia (pMOI). Sample size for control and FF mice, respectively: 4 months Male n = 5, 6, Female n = 8, 5; and 8 months: Male n = 4, 5; Female n = 3, 3. Statistical significance was assessed by two-way ANOVA with Tukey’s multiple comparisons test. ANOVA results as indicated. *p≤0.05. Data presented as mean ± SD. WT and FF mice were housed at 30°C on a 12 hr/12 hr light/dark cycle. Figure 3—figure supplement 1—source data 1.Cortical bone parameters.

• 

  Figure 4.. BMAT is present in AdipoqCre+/DTA+ fat free (FF) mice and expands with age. (A, B) Representative μCT images of osmium-stained tibiae for both male and female AdipoqCre+/DTA+ fat free (FF) and AdipoqCre-/DTA+ control (Con) mice at (A) 4 months and (B) 8 months of age. BMAT is in dark grey and bone is in light gray. (C) Quantification of total tibial BMAT volume as a percentage of total bone marrow volume. (D) Regional analysis of BMAT within the proximal end of the same tibiae as in (C), expressed as the total volume of osmium-stained lipid from the proximal end of the tibia to the tibia/fibula junction. (E) Regional analysis of BMAT within the distal end of the same tibiae as in (C), expressed as the total volume of osmium-stained lipid from tibia/fibula junction to the distal end of the bone. (F) Representative hematoxylin and eosin (H&E) stained sections of BMAT within tail vertebrae. Ad = BMAT adipocytes. B = bone. Scale = 50 µm. Sample size for control and FF mice, respectively: 4 months Male n = 5, 6, Female n = 8, 5; and 8 months Male n = 4, 5; Female n = 3, 3. Statistical significance was assessed by two-way ANOVA with Tukey’s multiple comparisons test. ANOVA results as…

• 

  Figure 5.. Fat free (FF) bone marrow adipocytes express perilipin, but not CD68, and have increased lipid storage relative to controls. Representative images from both male and female AdipoqCre+/DTA+ fat free (FF) and AdipoqCre-/DTA+ control (Con) mice at 8 months of age. (A) Representative longitudinal hematoxylin and eosin (H&E) stained sections of the femur and tibia, including the knee and surrounding soft tissue. Scale = 1 mm. (B) Representative serial sections stained with H&E, perilipin 1 (PLIN1, red; DAPI, blue), and CD68 (amplified with DAB, CD68+ cells are brown). Sections from the insets depicted in (A). ipWAT = infrapatellar white adipose tissue, located within the knee joint. BMAT = bone marrow adipose tissue. Scale = 50 µm. (C) Representative serial sections of the ectopic adipocytes within the tibial diaphysis in the AdipoqCre+/DTA+ (DTA) mice. Stained with H&E (left) and perilipin 1 (PLIN1, red; DAPI, blue). (D) Bone marrow adipocyte size distribution in the proximal tibia of the control and FF mice at 8 months of age. Scale = 500 µm. Sample size for control and FF mice, respectively: 8 months Male n = 4, 5; Female n = 3, 3. Statistical significance was assessed by two-way…

• 

  Figure 5—figure supplement 1.. Extra-skeletal adipocytes are depleted in the foot and tail of FF mice. Representative images from AdipoqCre+/DTA+ fat free (FF) and AdipoqCre-/DTA+ control (Control) mice at 8 months of age. Representative longitudinal hematoxylin and eosin (H&E) stained sections of the (A) foot and (B) tail vertebrae, including the bone (B) and surrounding soft tissues. Ad = adipocytes; BMA = bone marrow adipocyte. Scale = 0.5 mm. T and FF mice were housed at 30°C on a 12 hr/r12 hr light/dark cycle.

• 

  Figure 6.. BMAT in AdipoqCre+/DTA+ fat free (FF) mice is not responsive to cold temperature challenge (22°C vs 30°C). Male AdipoqCre+/DTA+ FF mice and controls were maintained in thermoneutral housing (30°C) or moved to room temperature (22°C) at 3–5 weeks of age. Bones were analyzed after 3–4 months, at 15–17 weeks of age. (A) Trabecular bone volume fraction (Tb. BVF) of tibia and femur. (B) Cortical thickness of tibia and femur analyzed by µCT. (C) Representative μCT images of osmium-stained tibiae at endpoint, respectively. Bone marrow fat is in dark grey and bone is in light gray. (D) Quantification of osmium-stained BMAT in the region proximal to the tibia/fibula junction (proximal tibia) or distal to this point (distal tibia). (E) Representative μCT images of osmium-stained femur at end point, respectively. (F) Quantification of osmium-stained BMAT in the 2 mm region below the growth plate (femur metaphysis, bracket) or from this point to the end of the femur flange (indicated by the arrow, diaphyseal BMAT). Sample size of control and FF, respectively: 30°C n = 4, 5; 22°C n = 5, 5. Statistical significance was assessed by two-way ANOVA with Tukey’s multiple comparisons test. ANOVA…

• 

  Figure 7.. BMAT in AdipoqCre+/DTA+ fat free (FF) mice is not regulated by β3-adrenergic stimulation. Male AdipoqCre+/DTA+ FF and AdipoqCre-/DTA+ controls (Con) were treated with CL316,243, a β3-adrenergic receptor (β3-AR) agonist, using a new chronic treatment regimen. Eight daily subcutaneous injections of 0.03 mg/kg CL316,243 were administered to 7.5 month old control and FF mice over the course of 10 days (weekdays only, M to F, M to W) prior to sacrifice on Day 11. (A) Serum glycerol at days 1 and 7 of the treatment regimen. (B) Average adipocyte cell size in the proximal tibia as assessed in ImageJ using H&E stained slides. (C) Representative H&E stained sections. Sample size of control and FF, respectively: 8 month old non-treatment control n = 4, 5 (same mice as in Figures 3—5), 8 month old CL316,243 treated n = 6, 5. Statistical significance was assessed by two-way ANOVA with Tukey’s multiple comparisons test. ANOVA results as indicated. (D) Gene expression of the indicated targets normalized to the geometric mean of housekeeping genes Ppia and Tbp in floated cell preparations enriched for bone marrow adipocytes (BMAe), each gene expressed relative to its respective control. Control…

• 

  Figure 8.. Subcutaneous fat transplant prevents BMAT expansion in AdipoqCre+/DTA+ fat free (FF) mice. Male and female AdipoqCre+/DTA+ FF and AdipoqCre-/DTA+ controls (Con) underwent sham surgery or were transplanted subcutaneously with WT inguinal white adipose tissue (iWAT) at 3–5 weeks of age. After surgery, mice were monitored for 12 weeks prior to sacrifice. (A) Body mass. (B) Random fed blood glucose, measured using a glucometer. (C) iWAT transplant mass at endpoint (week 12 after transplantation). (D) Liver mass at endpoint. (E) Serum triglyceride concentration at 4 weeks after the transplant surgery. (F, G) Representative μCT images of osmium-stained tibiae of (F) male and (G) female mice at endpoint. Bone marrow fat is in dark grey and bone is in light gray. (H) Quantification of total tibial BMAT volume. Sample size for control and FF mice, respectively: sham Male n = 4, 6; transplant Male n = 4, 5; sham Female n = 5, 7; transplant Female n = 5, 4. Statistical significance was assessed by (A, B) three-way ANOVA and (C–H) two-way ANOVA with Tukey’s multiple comparisons test. ANOVA results as indicated. *p≤0.05. Data presented as mean ± SD. All mice were housed at 30°C on a 12 hr/12…

• 

  Figure 8—figure supplement 1.. Engrafted fat transplant fragments resemble subcutaneous white adipose tissue at the terminal end point. Male and female AdipoqCre+/DTA+ fat free and AdipoqCre-/DTA+ controls underwent sham surgery or were transplanted subcutaneously with control adipose tissue at 3–5 weeks of age. After surgery, mice were monitored for 12 weeks prior to sacrifice. At the end point, the transplanted fat engrafted, became vascularized (arrowhead), and grossly resembled subcutaneous white adipose tissue in the FF mice (representative example pictured here).