ORIGINAL
ARTICLE
E n d o c r i n e
C a r e
Brown Adipose Tissue and Its Relationship to Bone
Structure in Pediatric Patients
Skorn Ponrartana, Patricia C. Aggabao, Houchun H. Hu, Grace M. Aldrovandi,
Tishya A. L. Wren, and Vicente Gilsanz
Departments of Radiology (S.P., P.C.A., H.H.H., T.A.L.W., V.G.), Pediatrics (G.M.A., V.G.), and
Orthopedic Surgery (T.A.L.W., V.G.), Children’s Hospital Los Angeles, and Department of Electrical
Engineering (H.H.H.), Viterbi School of Engineering, University of Southern California, Los Angeles,
California 90027
Context: Emerging evidence suggests a possible link between brown adipose tissue (BAT) and bone
metabolism.
Objective: The objective of this study was to examine the relationships between BAT and bone
cross-sectional dimensions in children and adolescents.
Design: This was a cross-sectional study.
Setting: The study was conducted at a pediatric referral center.
Patients: Patients included 40 children and teenagers (21 males and 19 females) successfully treated
for pediatric malignancies.
Interventions: There were no interventions.
Main Outcome Measures: The volume of BAT was determined by fluorodeoxyglucose-positron
emission tomography/computed tomography. Measures of the cross-sectional area and cortical
bone area and measures of thigh musculature and sc fat were determined at the midshaft of the
femur.
Results: Regardless of sex, there were significant correlations seen between BAT volume and the
cross-sectional dimensions of the bone (r values between 0.68 and 0.77; all P ⱕ 0 .001). Multiple
regression analyses indicated that the volume of BAT predicted femoral cross-sectional area and
cortical bone area, even after accounting for height, weight, and gender. The addition of muscle
as an independent variable increased the predictive power of the model but significantly decreased
the contribution of BAT.
Conclusions: The volume of BAT is positively associated with the amount of bone and the crosssectional size of the femur in children and adolescents. This relation between BAT and bone
structure could, at least in part, be mediated by muscle. (J Clin Endocrinol Metab 97: 2693–2698,
2012)
rown adipose tissue (BAT) is specialized in dissipating
energy in the form of heat as a defense against the
cold through its control by the BAT-specific uncoupling
protein-1 (1). BAT was thought to disappear after infancy, but studies finding BAT in patients undergoing
B
positron emission tomography (PET)/computed tomography (CT) have renewed the interest in deciphering the
biology and function of this tissue past infancy. Although the function of BAT beyond thermoregulation
remains uncertain, emerging evidence suggests signifi-
ISSN Print 0021-972X ISSN Online 1945-7197
Printed in U.S.A.
Copyright © 2012 by The Endocrine Society
doi: 10.1210/jc.2012-1589 Received March 5, 2012. Accepted April 25, 2012.
First Published Online May 16, 2012
Abbreviations: BAT, Brown adipose tissue; BMI, body mass index; CBA, cortical bone area;
CSA, cross-sectional area; CT, computed tomography; PET, positron emission tomography;
SUV, standardized uptake value; WAT, white adipose tissue.
J Clin Endocrinol Metab, August 2012, 97(8):2693–2698
jcem.endojournals.org
2693
2694
Ponrartana et al.
Brown Adipose Tissue Volume and Bone
J Clin Endocrinol Metab, August 2012, 97(8):2693–2698
cant interactions between brown adipose tissue, other
soft tissues, and bone.
Brown adipocytes are derived from mesenchymal stem
cells, which can also differentiate into white adipocytes,
myocytes, and osteoblasts. We recently observed that pediatric patients undergoing PET/CT examinations with
visualized BAT had significantly greater muscle volume
than patients with no identifiable BAT, a relationship that
was independent of age, sex, body mass, and season of
scan (2). We also found that the volume of BAT, like that
of muscle, increases rapidly during puberty in both boys
and girls (3). These clinical observations are consistent
with evidence from in vitro studies suggesting that classical brown adipocytes derive from myoblastic-like myf-5positive precursors (4). The observations are also in agreement with data indicating that brown adipocytes and
skeletal muscle cells share many features: an abundance of
mitochondria, energy expenditure via oxidative phosphorylation, and sympathetically mediated adaptive thermogenesis (5).
Available data from animal models also suggest a possible link between BAT and bone metabolism. The retinoblastoma protein was recently identified as a mesenchymal cell-fate regulator that controlled differentiation
into either the brown adipocyte or the osteoblast (6). Several reports have found this key regulator to be capable of
both inhibiting adipogenic differentiation and promoting
osteoblast maturation (6 – 8). Moreover, a clinical study
by Bredella et al. (9) found that subjects with functionally
active BAT had higher bone density values and lower levels
of preadipocyte factor-1, a negative regulator of adipocyte
differentiation, compared with those without BAT. Although the study was conducted in young women with
anorexia nervosa, women recovered from anorexia nervosa, and normal-weight women, the positive relation between BAT and bone density remained significant after
controlling for disease status and body mass. The authors
concluded that BAT may be involved in the regulation of
stem cell differentiation into the bone lineage at the expense of adipogenesis (10).
Because mechanical strains derived from muscle function are essential anabolic signals for bone, the possibility
exists that BAT could play an important role in the maintenance of skeletal integrity. In the current study, we examined whether there is an association between BAT volume and the development of the appendicular skeleton of
children and teenagers, and we explored the contribution
of muscle to this potential relationship.
Materials and Methods
The study subjects were patients successfully treated for pediatric
malignancy at Children’s Hospital Los Angeles (Los Angeles,
CA). This retrospective study was compliant with the Health
Insurance Portability and Accountability Act, and the investigational protocol was approved by the hospital’s institutional review board for clinical investigations. The requirement for informed consent was waived because all imaging was performed
for clinical purposes. For the purpose of this study, we reviewed
all 444 PET/CT studies performed from October 2008 to January 2012 and selected the last PET/CT examination of patients
who met the following criteria: 1) the patients were successfully
treated for cancer and were disease free, 2) they depicted functionally active BAT, and 3) they had imaging studies that included the entire femur. Most of the study subjects were included
in previous investigations (2, 3, 11).
All PET/CT scans were performed 1– 8 wk after their last form
of treatment on a Gemini GXL PET/CT system (Philips Healthcare, Cleveland, OH) after a 12-hour fast. Patients were injected
with fluorodeoxyglucose at a dose of 0.14
mCi/kg (12). All studies were performed after iv and oral contrast enhancement, and
all patients were indoors in room temperature (22 C) for at least 2 h before the examination. Age, height, weight, body mass index (BMI), and BMI percentile using the
Centers for Disease Control and Prevention’s calculator were assessed at the time of
the study.
A patient was considered to have BAT
when two radiologists independently confirmed its presence. The radiologists also assured that the underlying CT Hounsfield
units were negative and indicative of fat tissue density and that the regions of interest
FIG. 1. Coregistered CT and PET images from the same subject at two different levels.
excluded muscle and other tissue boundarArrows point to symmetrical uptake of radiotracers by BAT in foci with high SUV (⬎1.5),
ies. The distributions of Hounsfield units
which correspond to negative Hounsfield unit voxels (⫺20 to ⫺200) in the associated CT
from the CT and PET standardized uptake
images. The third row illustrates the corresponding segmented adipose tissue masks from
values (SUV) within these BAT-visualized
Matlab. White voxels denote mostly sc WAT, whereas interscapular BAT is gray.
J Clin Endocrinol Metab, August 2012, 97(8):2693–2698
jcem.endojournals.org
2695
TABLE 1. Age, anthropometric measures, and measures of muscle area, femoral CBA and CSA, sc fat area, and
BAT
Age (yr)
Weight (kg)
Height (cm)
BMI (%)
Thigh muscle (cm2)
Femoral CBA (cm2)
Femoral CSA (cm2)
Thigh sc fat (cm2)
Body BAT volume (cm3)
All (n ⴝ 40)
15.1 ⫾ 3.3
60.1 ⫾ 22.4
159.6 ⫾ 16.3
64.3 ⫾ 29.4
98.5 ⫾ 32.5
4.4 ⫾ 1.18
5.3 ⫾ 1.37
124.9 ⫾ 59.0
417.3 ⫾ 273.8
Males (n ⴝ 21)
14.7 ⫾ 3.6
61.2 ⫾ 23.2
162.7 ⫾ 19.3
63.9 ⫾ 32.9
105.1 ⫾ 35.5
4.5 ⫾ 1.28
5.5 ⫾ 1.56
102.8 ⫾ 43.3
456.6 ⫾ 323.9
areas were then measured to correspond to the volume of BAT
(cubic centimeters) from the base of the skull to the pubic bones.
The computation of BAT volume via threshold SUV values and
CT Hounsfield unit ranges was performed with scripts written in
Matlab software (Mathworks Inc., Natick, MA). For the procedure, any voxel with a CT Hounsfield unit value between ⫺20
and ⫺200 Hounsfield units was considered adipose tissue, regardless of being brown or white. Even though Hounsfield unit
values for BAT are significantly higher than those of white adipose tissue (WAT) and uptakes of these two adipose tissues do
not overlap at levels exceeding 1.5 SUV (11), we used thresholds
of 1 SD above the mean Hounsfield unit and a standardized uptake value of greater than 1.5 to calculate the volume of BAT (Fig.
1). This approach for estimating BAT volume is similar to those
previously reported in literature (13–15).
The femoral cross-sectional area (CSA) and cortical bone area
(CBA), the CSA of thigh musculature (quadricep muscles, hamstring muscles, and adductor muscles), and the CSA of sc WAT
were determined for both legs at the midshaft of the femur by
SliceOmatic (Tomovision, Montréal, Québec, Canada). The coefficients of variations for these measurements have been previously reported to range from 0.6 to 2.5% (16).
Descriptive and linear regression analyses were performed
with Statview software (version 5.0.1; SAS Institute, Cary, NC).
In most models, either the femoral CSA or the CBA was used as
the outcome variable, and anthropometric measures, gender,
muscle area, and the volume of BAT were used as possible independent variables. To exclude the possibility of multicollinearity, the goodness of fit of the regression models was evaluated
using the postestimation procedures of Stata (StataCorp, College
Station, TX). All values are expressed as mean ⫾ SD.
Females (n ⴝ 19)
15.5 ⫾ 3.0
58.9 ⫾ 21.9
156.1 ⫾ 11.7
64.8 ⫾ 25.8
91.2 ⫾ 28.0
4.4 ⫾ 1.08
5.1 ⫾ 1.11
149.3 ⫾ 65.2
373.8 ⫾ 205.2
P value
0.448
0.750
0.209
0.920
0.180
0.727
0.284
0.011
0.346
Results
The study cohort consisted of 40 patients (21 males and 19
females) between the ages of 6 and 19.4 yr: 35 had lymphoma, two had acute lymphoblastic leukemia, and one
each had Ewing sarcoma, melanoma, and neuroblastoma.
Table 1 describes the age, weight, height, BMI percentile, and the values for muscle, femoral CBA and CSA, sc
fat, and BAT in all patients and for males and females
separately. Although there were no significant gender differences in anthropometric and PET/CT measures for
muscle, bone, and BAT, male patients tended to be taller
and heavier and had greater values for muscle, bone,
and BAT despite their relatively younger age. In contrast, measures of sc fat at the midthigh were significantly greater in females than in males.
Twenty-one patients were examined in winter months,
six in spring, three in summer, and 10 in autumn. There
were no significant differences between the BAT volumes
of patients scanned in winter months compared with those
scanned in nonwinter months (483 ⫾ 313 vs. 344 ⫾ 208;
P ⫽ 0.11); the lack of statistical significance is likely a
reflection of the small number of examinations performed
in nonwinter months.
As expected, there were significant correlations between the cortical and cross-sectional areas of the femur
TABLE 2. Simple correlations of age, anthropometric measures, and measures of musculature, bone, and white and
brown adiposity
Age
(yr)
Age (yr)
Weight (kg)
Height (cm)
BMI (%)
Muscle (cm2)
CBA (cm2)
CSA (cm2)
sc fat (cm2)
BAT volume (cm3)
0.78
0.80
0.18
0.56
0.80
0.77
0.53
0.74
Weight
(kg)
0.72
0.55
0.65
0.74
0.76
0.77
0.65
0.84
Height
(cm)
0.85
0.81
BMI (%)
⫺0.07
0.52
0.06
0.01
0.50
0.68
0.67
0.34
0.58
There were 19 female patients (bottom) and 21 male patients (top).
0.30
0.23
0.22
0.32
0.60
Muscle
(cm2)
0.72
0.85
0.72
0.42
0.83
0.84
0.71
0.63
CBA
(cm2)
0.80
0.81
0.84
0.28
0.81
0.99
0.77
0.68
CSA
(cm2)
0.78
0.73
0.84
0.11
0.78
0.96
0.75
0.68
sc fat
(cm2)
0.62
0.80
0.62
0.38
0.62
0.71
0.69
0.56
BAT volume
(cm3)
0.70
0.71
0.60
0.25
0.70
0.77
0.73
0.60
2696
Ponrartana et al.
Brown Adipose Tissue Volume and Bone
J Clin Endocrinol Metab, August 2012, 97(8):2693–2698
Discussion
We previously found that BAT is positively associated with muscle mass (2).
The current study shows that BAT also
influences bone structure. Both the
cross-sectional area and the amount of
cortical bone at the midshaft of the femur were positively associated with the
volume of BAT from the base of the
skull to the pubic bone in children and
adolescents. Moreover, these relationships were independent of known major determinants of bone acquisition,
such as height, weight, and gender.
The reason(s) for the link between
FIG. 2. Coronal PET scans of the neck and upper chest of girls with high (top panel) and low
BAT and skeletal growth in children is
(bottom panel) BAT volumes. Corresponding CT images of the midshaft of the femurs of the
unknown. However, finding that the
same subjects show larger cross-sectional femoral dimensions (top panel) in the subject with
effect of BAT became nonsignificant
greater BAT volume.
with the inclusion of muscle in the
model supports the notion that the
and the age, anthropometric measures, and muscle area, BAT-bone link could, at least in part, be mediated by musregardless of gender (Table 2). The cross-sectional di- cle. Support for this notion comes from the knowledge that
mensions of the femur were also strongly associated maintaining optimal bone mass and structure depends on
with BAT volume (Figs. 2 and 3) and with the amount sensing and transducing mechanical loading information
of muscle at the midthigh in both males and females. derived from muscle contractions (17, 18). This adaptaThe relationship between BAT and CT measures of tion of the skeleton to stress leads not only to an increase
muscle and bone tended to be stronger than those be- in bone mass but also to an increase in the geometric ditween BAT and WAT (Table 2).
mensions of the bone leading to increased bone strength
Multiple regression analyses indicated that the vol- (19). The theory of a functionally unified muscle-bone
ume of BAT was independently associated with femoral system is further supported by a common embryogenesis
CSA and CBA, even after accounting for height, weight, of both tissues signaled via the Wnt pathway and evidence
and gender (Table 3). The addition of muscle as an showing that they are regulated by the same hormones and
independent variable increased the predictive power of genes (20). Alternatively, the possibility exists for a direct
the model but significantly decreased the contribution link between BAT and skeletal development. Animal studof BAT (Table 4). Similar results were obtained when ies support a causal relationship between BAT and osteousing BMI percentile rather than height and weight in blastogenesis. Mice lacking functional BAT were found to
have very low bone mass, reduced osteoblast activity, and
the model.
FIG. 3. Correlation between femoral CBA and BAT volume (A) and femoral CSA and BAT volume (B).
J Clin Endocrinol Metab, August 2012, 97(8):2693–2698
jcem.endojournals.org
TABLE 3. Multiple regression models for the femoral
CBA and CSA
␤
Cortical bone area
Height (cm)
Weight (kg)
Gender
BAT volume (cm3)
Cross-sectional area
Height (cm)
Weight (kg)
Gender
BAT volume (cm3)
SE
P value
R2
0.423
0.292
0.086
0.276
0.009
0.007
0.201
0.001
0.001
0.05
0.32
0.04
0.758
0.510
0.157
⫺0.017
0.294
0.011
0.009
0.243
0.001
0.0003
0.30
0.85
0.03
0.739
increased bone resorption (21–23). Moreover, heterotrophic ossification modeled by the bone morphogenic
protein-2 is known to induce the accumulation of brown
adipocytes and subsequently trigger chondrocyte development and bone formation (24).
Mechanical strains are essential anabolic signals for
bone. A decrease in skeletal loading is known to diminish
osteoblastic bone formation and increase osteoclastic
bone resorption, resulting in a significant decrease in bone
accretion and an overall loss of bone. There are physiological situations, however, in which no significant signs
of disuse are seen despite minimal mechanical loading.
Large hibernating mammals, for example, remarkably
maintain their muscle and bone mass despite losing a third
of their weight and remaining immobile over a period of
nearly 7 months. Interestingly, the relative mass of BAT
has been shown to markedly increase during hibernation
in rodents (25). Similarly, infancy is a developmental stage
associated with large amounts of BAT and rapid increases
in muscle and bone mass despite the lack of significant
skeletal loading. Studies are needed to determine the degree to which BAT contributes to the physiological processes necessary for the maintenance of muscle function in
hibernating animals and for musculoskeletal development
in infants.
TABLE 4. Multiple regression models for the femoral
CBA and CSA with the inclusion of muscle
␤
Cortical bone area
Height (cm)
Weight (kg)
Gender
Muscle (cm2)
BAT volume (cm3)
Cross-sectional area
Height (cm)
Weight (kg)
Gender
Muscle (cm2)
BAT volume (cm3)
SE
P value
R2
0.367
0.068
0.139
0.395
0.213
0.008
0.008
0.187
0.005
0.001
0.002
0.65
0.09
0.01
0.08
0.807
0.449
⫺0.091
0.041
0.435
0.224
0.010
0.009
0.222
0.006
0.001
0.0004
0.56
0.62
0.003
0.07
0.799
2697
Several technical issues in regard to PET/CT must be
considered for the appropriate interpretation of the current results. There is currently no consensus regarding a
standard SUV threshold value for objective BAT detection
and quantification by PET/CT. Unfortunately, a standard
threshold would be difficult to establish because SUV can
be greatly influenced by patient population, age, and vendor-specific imaging equipment and software. However,
based on our previous experience (11), a SUV threshold of
1.0 –1.5 for our pediatric and adolescent patient population appears adequate because there was no overlap in CT
values between subjects above and below 1.5.
This study has several notable limitations. It is based on
a retrospective, cross-sectional analysis of PET/CT examinations of children and adolescents with cancer. Moreover, possible confounding variables known to be major
regulators of bone, such as the degree of physical activity
and nutrition, were not included in our analysis. Although
we selected disease-free patients to minimize the bias associated with not recruiting healthy subjects, the findings
of this study need to be corroborated in healthy populations and adult cohorts. More importantly, prospective
longitudinal analyses are needed to clearly establish the
degree to which this association could be mediated by
muscle and how BAT could influence muscle contraction,
which is an essential anabolic signal for bone.
In conclusion, the current study provides evidence for
a positive association between the volume of BAT and the
amount and cross-sectional dimensions of the femur.
These findings in the appendicular skeleton complement
existing evidence showing that young patients with functionally active BAT have greater bone mineral density
measures in the axial skeleton than those without visualized BAT (10). Our results also raise the question of
whether the association between BAT and bone structure
is mediated by muscle. Improving our understanding of
the interactions between BAT, muscle, and bone could
lead to effective countermeasures to prevent the dramatic
bone loss associated with skeletal unloading.
Acknowledgments
Address all correspondence and requests for reprints to: Vicente
Gilsanz, M.D., Ph.D., Children’s Hospital Los Angeles, Department of Radiology, MS #81, 4650 Sunset Boulevard, Los Angeles, California 90027. E-mail: [email protected].
This work was supported by National Institutes of Health
Grants R21DK090778 and K25DK087931 (from the National
Institute of Diabetes and Digestive and Kidney Diseases).
Disclosure Summary: The authors have nothing to disclose.
2698
Ponrartana et al.
Brown Adipose Tissue Volume and Bone
J Clin Endocrinol Metab, August 2012, 97(8):2693–2698
1. Cannon B, Nedergaard J 2004 Brown adipose tissue: function and
physiological significance. Physiol Rev 84:277–359
2. Gilsanz V, Chung SA, Jackson H, Dorey FJ, Hu HH 2011 Functional
brown adipose tissue is related to muscle volume in children and
adolescents. J Pediatr 158:722–726
3. Gilsanz V, Smith ML, Goodarzian F, Kim M, Wren TA, Hu HH
2012 Changes in brown adipose tissue in boys and girls during childhood and puberty. J Pediatr 160:604 – 609.e601
4. Kajimura S, Seale P, Spiegelman BM 2010 Transcriptional control
of brown fat development. Cell Metab 11:257–262
5. Farmer SR 2008 Brown fat and skeletal muscle: unlikely cousins?
Cell 134:726 –727
6. Calo E, Quintero-Estades JA, Danielian PS, Nedelcu S, Berman SD,
Lees JA 2010 Rb regulates fate choice and lineage commitment in
vivo. Nature 466:1110 –1114
7. Thomas DM, Carty SA, Piscopo DM, Lee JS, Wang WF, Forrester
WC, Hinds PW 2001 The retinoblastoma protein acts as a transcriptional coactivator required for osteogenic differentiation. Mol
Cell 8:303–316
8. Fajas L, Egler V, Reiter R, Hansen J, Kristiansen K, Debril MB,
Miard S, Auwerx J 2002 The retinoblastoma-histone deacetylase 3
complex inhibits PPAR␥ and adipocyte differentiation. Dev Cell
3:903–910
9. Abdallah BM, Ding M, Jensen CH, Ditzel N, Flyvbjerg A, Jensen
TG, Dagnaes-Hansen F, Gasser JA, Kassem M 2007 Dlk1/FA1 is a
novel endocrine regulator of bone and fat mass and its serum level
is modulated by growth hormone. Endocrinology 148:3111–3121
10. Bredella MA, Fazeli PK, Freedman LM, Calder G, Lee H, Rosen CJ,
Klibanski A 2012 Young women with cold-activated brown adipose
tissue have higher bone mineral density and lower Pref-1 than
women without brown adipose tissue: a study in women with anorexia nervosa, women recovered from anorexia nervosa, and normal-weight women. J Clin Endocrinol Metab 97:E584 –E590
11. Hu HH, Chung SA, Nayak KS, Jackson HA, Gilsanz V 2011 Differential computed tomographic attenuation of metabolically active
and inactive adipose tissues: preliminary findings. J Comput Assist
Tomogr 35:65–71
12. Gelfand MJ, O’hara S M, Curtwright LA, Maclean JR 2005 Premedication to block [(18)F]FDG uptake in the brown adipose tissue
of pediatric and adolescent patients. Pediatr Radiol 35:984 –990
13. Pfannenberg C, Werner MK, Ripkens S, Stef I, Deckert A, Schmadl
M, Reimold M, Häring HU, Claussen CD, Stefan N 2010 Impact of
age on the relationships of brown adipose tissue with sex and adiposity in humans. Diabetes 59:1789 –1793
Saito M, Okamatsu-Ogura Y, Matsushita M, Watanabe K,
Yoneshiro T, Nio-Kobayashi J, Iwanaga T, Miyagawa M, Kameya
T, Nakada K, Kawai Y, Tsujisaki M 2009 High incidence of metabolically active brown adipose tissue in healthy adult humans: effects of cold exposure and adiposity. Diabetes 58:1526 –1531
Lee P, Greenfield JR, Ho KK, Fulham MJ 2010 A critical appraisal
of the prevalence and metabolic significance of brown adipose tissue
in adult humans. Am J Physiol Endocrinol Metab 299:E601–E606
Kovanlikaya A, Loro ML, Hangartner TN, Reynolds RA, Roe TF,
Gilsanz V 1996 Osteopenia in children: CT assessment. Radiology
198:781–784
Seger RL, Cross RA, Rosen CJ, Causey RC, Gundberg CM, Carpenter TO, Chen TC, Halteman WA, Holick MF, Jakubas WJ,
Keisler DH, Seger RM, Servello FA 2011 Investigating the mechanism for maintaining eucalcemia despite immobility and anuria in
the hibernating American black bear (Ursus americanus). Bone 49:
1205–1212
Frost HM 2003 Bone’s mechanostat: a 2003 update. Anat Rec A
Discov Mol Cell Evol Biol 275:1081–1101
Fricke O, Schoenau E 2007 The ‘Functional Muscle-Bone Unit’:
probing the relevance of mechanical signals for bone development in
children and adolescents. Growth Horm IGF Res 17:1–9
Matsuoka T, Ahlberg PE, Kessaris N, Iannarelli P, Dennehy U, Richardson WD, McMahon AP, Koentges G 2005 Neural crest origins
of the neck and shoulder. Nature 436:347–355
Carmona MC, Hondares E, Rodríguez de la Concepción ML, Rodríguez-Sureda V, Peinado-Onsurbe J, Poli V, Iglesias R, Villarroya
F, Giralt M 2005 Defective thermoregulation, impaired lipid metabolism, but preserved adrenergic induction of gene expression in
brown fat of mice lacking C/EBP␤. Biochem J 389:47–56
Zanotti S, Stadmeyer L, Smerdel-Ramoya A, Durant D, Canalis E
2009 Misexpression of CCAAT/enhancer binding protein ␤ causes
osteopenia. J Endocrinol 201:263–274
Motyl KJ, Rosen CJ 2011 Temperatures rising: brown fat and bone.
Discov Med 11:179 –185
Olmsted-Davis E, Gannon FH, Ozen M, Ittmann MM, Gugala Z,
Hipp JA, Moran KM, Fouletier-Dilling CM, Schumara-Martin S,
Lindsey RW, Heggeness MH, Brenner MK, Davis AR 2007 Hypoxic
adipocytes pattern early heterotopic bone formation. Am J Pathol
170:620 – 632
Egginton S, Fairney J, Bratcher J 2001 Differential effects of cold
exposure on muscle fibre composition and capillary supply in hibernator and non-hibernator rodents. Exp Physiol 86:629 – 639
References
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.