Journal of Gerontology: BIOLOGICAL SCIENCES
Cite journal as: J Gerontol A Biol Sci Med Sci
2010 Vol. 65A, No. 1, 31–40
Two-Year Body Composition Analyses of Long-Lived
GHR Null Mice
© The Author 2009. Published by Oxford University Press on behalf of The Gerontological Society of America.
All rights reserved. For permissions, please e-mail: email@example.com.
Advance Access published on November 9, 2009
life span (as reviewed in ( 1 )). For example, several studies
using mouse models with disruptions to the GH/IGF-I axis,
including the Ames dwarf ( 1 ), Snell dwarf ( 2 ), and growth
hormone receptor/GH-binding protein gene – disrupted
(GHR − / − ) dwarf ( 3 ), have all reported extended longevity.
Unfortunately, in the Ames and Snell dwarf mice, pituitary
defi ciencies in prolactin- and thyroid-stimulating hormone
production exist in addition to GH; therefore, it is not
possible to determine which of the absent hormones are
responsible for the increased longevity ( 4 – 9 ). Because the
GHR − / − mice have a disruption specifi cally in the GHR
gene, these mice are useful for studying the impact of GH
on aging ( 3 ). Caloric restriction (CR) also has been shown
to enhance longevity in fl ies, mice, and most recently in
nonhuman primates ( 10 ). Interestingly, CR of GHR − / −
mice does not further extend life span ( 11 ), suggesting GH
resistance/insensitivity and CR work via similar mecha-
nisms for life-span extension even though gene expression
profi ling suggests that several genes are differentially ex-
pressed ( 2 , 12 – 14 ).
Attempts to determine mechanisms responsible for the
increased life span in GHR − / − mice are ongoing. These
mice remain dwarf throughout life and have elevated circu-
lating GH and markedly reduced IGF-I levels ( 15 ). GHR − / −
mice have a slight decrease in fasting glucose levels at
younger ages (up to 10 months) ( 15 – 21 ), although no sig-
nifi cant differences are found in older male mice ( 15 , 22 , 23 ).
Insulin levels are signifi cantly lower when compared with
ECREASED growth hormone (GH)/insulin-like
growth factor-I (IGF-I) signaling is known to extend
controls, independent of gender or age, and have a concom-
itant improvement in insulin sensitivity ( 11 , 15 , 16 , 18 – 25 ).
Reduced levels of thyroid hormone and lower core body
temperature in these mice have also been reported ( 21 ). Al-
though less consistent, lipid levels are typically improved
with total cholesterol and low-density lipoprotein choles-
terol decreased in male GHR − / − mice ( 17 , 20 , 25 ). Finally,
these mice are protected from fatal neoplasia ( 26 ). Collec-
tively, these characteristics are generally considered favor-
able for the health of the mice and likely contribute to
improving their longevity.
Based on the known lipolytic/antilipogenic actions of
GH, it is expected that an absence of GH action will result
in increased adipose tissue in GHR − / − mice. Indeed, sev-
eral reports in which whole-body composition analyses
were performed consistently show that male GHR − / − mice
are relatively obese in comparison to littermate controls
( 18 , 20 , 27 , 28 ), with the only exception in one report in
which young mice (6 – 7 weeks) did not show signifi cant dif-
ferences in percent whole-body fat ( 28 ). Furthermore, de-
spite the signifi cantly reduced body size and weight of the
male GHR − / − mice, the absolute weight of their total fat
mass is comparable to that of littermate controls in 6-month-
old male mice ( 18 ). Thus, whole adipose tissue mass ap-
pears to be one of the few tissues not reduced in adult male
mice. Importantly, the accumulation of fat mass is not uni-
form among different depots; in particular, the subcutane-
ous depot is disproportionately enlarged ( 16 , 18 , 27 , 29 ).
Increases in adiposity are often associated with decreases
in life span and impaired insulin sensitivity. Thus, these
Darlene E. Berryman , 1 , 2 Edward O. List , 2 Amanda J. Palmer , 1 Min-Yu Chung , 1 , 3 Jacob Wright-Piekarski , 2
Ellen Lubbers , 2 Patrick O’Connor , 4 Shigeru Okada , 2 , 5 and John J. Kopchick 2 , 4
1 School of Human and Consumer Sciences, College of Health and Human Services and 2 Edison Biotechnology Institute,
Ohio University, Athens .
3 Department of Nutritional Science, University of Connecticut, Storrs .
4 Department of Biomedical Sciences and 5 Department of Pediatrics, College of Osteopathic Medicine, Ohio University, Athens .
Growth hormone receptor gene – disrupted (GHR − / − ) mice exhibit increased life span and adipose tissue mass. Although
this obese phenotype has been reported extensively for young adult male GHR − / − mice, data for females and for other
ages in either gender are lacking. Thus, the purpose of this study was to evaluate body composition longitudinally in both
male and female GHR − / − mice. Results show that GHR − / − mice have a greater percent fat mass with no signifi cant dif-
ference in absolute fat mass throughout life. Lean mass shows an opposite trend with percent lean mass not signifi cantly
different between genotypes but absolute mass reduced in GHR − / − mice. Differences in body composition are more
pronounced in male than in female mice, and both genders of GHR − / − mice show specifi c enlargement of the subcutane-
ous adipose depot. Along with previously published data, these results suggest a consistent and intriguing protective
effect of excess fat mass in the subcutaneous region.
Key Words: Body composition — Growth hormone — Obesity — Adipose depots — Gender differences .
BERRYMAN ET AL.
long-lived mice offer a unique and counterintuitive situation
in which obesity is associated with improved life span and
many health parameters.
Previous studies that have assessed body composition in
GHR − / − mice have not used both genders and have assessed
body composition at a particular age. Although one study
did report body composition in both genders at several time
points ( 28 ), this study used a separate cohort of mice for
only three ages and had a wide range of ages to represent
young, adult, and aged mice. No studies have tracked body
composition in the same cohort of mice over their life span.
Because earlier studies in bovine GH transgenic mice and
growth hormone releasing hormone receptor – defi cient lit/lit
mice show clear age- and gender-dependent effects on body
composition ( 30 , 31 ), there is a need to evaluate the patterns
of body composition changes throughout the life span in
these GHR − / − mice. This might offer insight as to how rela-
tive obesity can be accompanied by improvements in life
span. Thus, the purpose of this study was to assess system-
atically the changes in body composition more than 2 years
in male and female GHR − / − mice in order to provide a bet-
ter understanding of the metabolic dysfunction surrounding
adiposity and its contribution to longevity.
M aterials and M ethods
Male and female GHR − / − in a C57BL/6J background
and wild-type (WT) littermate controls were used in this
study. The generation of this gene-disrupted mouse and
subsequent backcrossing into the C57Bl/6J strain have been
described previously ( 3 ). Although the body composition
measurements were initiated with more mice, ultimately
only six male GHR − / − , six male WT, eight female GHR − / − ,
and eight female WT mice were used for the fi nal analyses
due to death of several mice during the 2-year measurement
period. For bone mineral density and liver triglyceride
(TAG) measurements, tissues from these same mice were
assayed; however, an additional cohort of age-matched mice
was used to confi rm the results. In this additional cohort of
mice, there were 10 male GHR − / − , 10 male WT, 9 female
GHR − / − , and 6 female WT mice. Mice were bred and
housed within the animal facility at Ohio University with a
14-hour light/10-hour dark cycle in a temperature-controlled
environment (21°C – 23°C). Mice were housed up to four
per cage with ad libitum access to food and water. Mice
were weaned at 28 days of age onto a standard rodent diet
(ProLab RMH 3000; PMI Nutrition International, Inc., St.
Louis, MO; 14% of kilocalories from fat, 16% from protein,
and 60% from carbohydrates). Mice were maintained on the
standard rodent diet throughout the study. All procedures
were approved by the Ohio University Institutional Care
and Use Committee and fully complied with federal, state,
and local policies.
Weight and Body Composition Measurements
Body weights were measured for each animal in dupli-
cate just prior to body composition measurements using a
standard scale, and the mean body weight was used for
analysis ( 32 ). The Bruker Minispec (Bruker Optics, The
Woodlands, TX), which employs nuclear magnetic reso-
nance (NMR) technology to estimate the fat, lean, and fl uid
mass of the animals, was used to assess body composition.
Weight and body composition measurements of all mice
were taken every 2 weeks from 6 weeks of age until 16
weeks of age (measured at 6, 8, 10, 12, 14, and 16 weeks).
Starting at 16 weeks, measurements were taken every 4
weeks until 104 weeks of age (or 2 years). The percent fat
mass, percent lean mass, and percent fl uid mass of each
animal were calculated at each time point using fat, lean,
and fl uid mass divided by body weight, respectively. Addi-
tionally, a single body composition measurement was taken
at 104 weeks for the second cohort of mice.
Tissues were harvested at 105 weeks of age after sacrifi cing
the mice by cervical dislocation. Several tissues were weighed
and collected, including liver, heart, spleen, kidney, and four
adipose depots (subcutaneous, epididymal/parametrial, ret-
roperitoneal, and mesenteric). Tissues were frozen in liquid
nitrogen and stored at − 80°C for future analysis.
Liver TAG Measurements
Liver samples were digested in a 3 M KOH in 65% ethanol
solution overnight for the extraction and measurement of TAG
levels, as described previously ( 33 ). A Triglycerides-GPO
kit (Pointe Scientifi c, Canton, MI) was used to measure the
glycerol content of the samples. TAG levels were estab-
lished assuming that the average molecular weight of TAG
is 885 g/mol.
Bone Length and Mineral Density
In order to obtain estimates of bone mineral density and
femur length, the right hind limbs were fi rst dissected at the
hip following which most soft tissues were removed. Fem-
ora were then mounted in foam blocks and scanned using
the GE eXplore Locus Small Animal MicroCT Scanner (GE
Healthcare, London, Ontario, Canada) using a 20- m m voxel
protocol with the following scan parameters: 80 kV, 450
m A, and 2000-milliseconds exposure time. Bone density was
normalized using an acrylic calibration phantom that in-
cluded densities equivalent to air, water, and bone. A mid-
diaphyseal region of the femur was selected using a threshold
of 800 HU in order to separate bone tissue from the back-
ground image. Femur length was measured between the
proximal end of the greater trochanter and the distal edge of
the intercondylar fossa. Bone mineral density of the selected
tissue was then calculated using the bone analysis package
TWO-YEAR BODY COMPOSITION ANALYSES
in GE Microview 2.2.1. For the fi rst cohort of mice that were
used for body composition analyses, only four male GHR − / −
femora were available for bone length and mineral density
analyses. Because of the resultant low sample size for male
GHR − / − mice in this original group, data from the second
cohort of animals are provided in Figure 4; yet, results from
statistical analyses for both groups are reported.
Data are presented as mean ± SEM . Statistics were per-
formed using the SPSS version 16.0 software (Chicago, IL).
Comparisons of longitudinal data were done with two-way
repeated measures analysis of variance (ANOVA). When
the treatment by time interactions were signifi cant, pairwise
contrasts were conducted for specifi c weeks using a Bonfer-
roni procedure. Data for other measurements were analyzed
using the appropriate two-way ANOVA or Student ’ s t test.
All data generated with the second cohort of animals were
analyzed separately. Differences were considered signifi -
cant at p < .05.
The mean body weights of both male and female GHR − / −
mice were signifi cantly less than their littermate controls at
all time points measured ( Figure 1 ). Even at the start of the
study at 6 weeks of age, the weight of both genders of
GHR − / − mice was 53% of corresponding littermate con-
trols. By the last measurement at 2 years or 104 weeks of
age, weights of male and female GHR − / − mice were 56%
and 44%, respectively, of their littermate controls. Thus,
GHR − / − mice were consistently about one half the weight
of control mice. Both genders and genotypes exhibited
accelerated pubertal growth rates up until ~ 16 weeks of age
followed by a slower progressive increase in weight in sub-
sequent weeks. Body weights for all mice except female
WT peaked by or before 96 weeks of age and began to
decline by the fi nal measurement at 104 weeks of age.
Longitudinal measurements of fat mass and lean mass as
well as fat and lean mass normalized to body weight showed
markedly different trends by genotype and gender. For fat
mass, there was very little difference between GHR − / − male
mice and littermate controls, with the exception that fat
mass gain in GHR − / − mice was more rapid and prominent
between 10 and 28 weeks of age ( Figure 2A ). In contrast,
female mice had no signifi cant difference in fat mass until
92 weeks of age, at which point WT mice had higher fat
mass through the remainder of the study ( Figure 2B ). Two-
way repeated measures ANOVA revealed no signifi cant
difference based on genotype, F (1,24) = 0.00, p = .994, or
gender, F (1,24) = 4.2, p = .052, although there was a signifi cant
interaction between genotype and gender, F (1,24) = 5.38,
p = 0.3. Because of the extraordinary differences in body
weight, it is also necessary to consider fat mass changes
relative to body weight. Percent fat mass ( Figure 2C ) was
markedly elevated at all time points in male GHR − / − rela-
tive to WT male mice. A similar trend was observed for fe-
male mice except that the difference between genotypes
was not as large and the difference dissipated by 88 weeks
of age ( Figure 2D ). Two-way repeated measures ANOVA
revealed a signifi cant difference based on genotype, F (1,24) =
69.8, p = 1.5 × 10 − 8 , and gender, F (1,24) = 5.4, p = .03, as
well as a signifi cant interaction between genotype and
gender, F (1,24) = 12.2, p = .002. All groups of mice experi-
enced fat mass and percent fat mass loss as they approached
2 years of age. In males, fat mass peaked by 80 and 72
weeks of age for GHR − / − and control mice, respectively.
The fat mass loss in females occurred a bit later with control
mice reaching peak fat mass at 96 weeks of age and GHR − / −
mice by 84 weeks of age.
Whereas absolute fat mass was similar between geno-
types but signifi cantly increased in GHR − / − mice when
normalized to body weight, lean mass showed the opposite
trend. That is, absolute lean mass was signifi cantly reduced
in GHR − / − mice ( Figure 3A and B ), whereas percent lean
mass was relatively proportional to body size ( Figures 3C
and D ). Specifi cally, two-way repeated measures ANOVA
Figure 1. Body weight over time for male and female growth hormone receptor gene – disrupted (GHR − / − ) mice and littermate controls. Weight measurements
were taken every other week from 6 weeks of age until 16 weeks and taken monthly from 16 weeks until 104 weeks or 2 years of age. Data are expressed as mean ±
SEM . Two-way repeated measures analysis of variance revealed a signifi cant impact of genotype, F (1,24) = 163.2, p = 2 × 10 − 12 , and gender, F (1,24) = 14.2, p = .001.
WT, wild type.
BERRYMAN ET AL.
revealed a signifi cant difference in absolute lean mass
based on genotype, F (1,24) = 434, p = 7.5 × 10 − 17 , and
gender, F (1,24) = 18.3, p = 0.00025, as well as a signifi cant
interaction between genotype and gender, F (1,24) = 8.9,
p = .007. For percent lean mass, the data revealed no sig-
nifi cant difference based on genotype, F (1,24) = 0.06, p =
.82, or gender F (1,24) = 1.4, p = .24 or interaction between
genotype and gender, F (1,24) = 1.9, p = .18. All mice
showed a more drastic increase in lean mass in early life (6
weeks until ~ 20 weeks) that remained relatively stable
throughout the remainder of the study. An exception was
found in control female mice in which lean mass gains
were seen even during the fi nal months of measurement.
Much like lean mass, fl uid mass, as predicted with this
nuclear magnetic resonance technology, was proportional
to body size (data not shown).
Figure 2. Absolute fat mass ( A and B ) and percent fat mass ( C and D ) for male ( A and C ) and female ( B and D ) growth hormone receptor gene – disrupted
(GHR − / − ) and littermate control (wild-type [WT]) mice. Data are expressed as mean ± SEM .
Figure 3. Absolute lean mass ( A and B ) and percent lean mass ( C and D ) for male ( A and C ) and female ( B and D ) growth hormone receptor gene – disrupted
(GHR − / − ) and littermate control (wild-type [WT]) mice. Data are expressed as mean ± SEM .
TWO-YEAR BODY COMPOSITION ANALYSES
Mean values for femoral length were 10.9 ± 0.1 for male
GHR − / − mice, 11.4 ± 0.1 for female GHR − / − mice, 15.6 ±
0.1 for male WT mice, and 16.1 ± 0.07 for female WT mice.
Femoral length exhibited signifi cant differences between
both genotype, F (1,30) = 1531.8, p = 2.6 × 10 − 27 , and
gender, F (1,30) = 15.77, p = 0.004 ( Figure 4 ) with females
having longer femur lengths at 2 years. However, there was
no signifi cant interaction between genotype and gender,
F (1,30) = 0.034, p = 0.8. Although signifi cant differences
were noted in bone mineral density estimates between gen-
otypes, F (1,30) = 27.89, p = 0.000011, no signifi cant differ-
ence was identifi ed between genders, F (1,30) = 0.141, p =
.71 ( Figure 4 ). Similar to femoral length, there was no
signifi cant interaction effects between genotype and gender,
F (1,30) = 0.005, p = 0.94. Data are shown for the second
cohort of mice. The data generated with the fi rst cohort of
mice, used for the body composition analyses, showed a
similar trend as the data shown for the second cohort of
mice; however, femoral lengths for this set of mice did not
reveal a statistically signifi cant difference between genders,
F (1,22) = 1.6, p = .16, possibly due to the smaller sample size.
Because of the signifi cant difference in body weights for
GHR − / − and WT mice, it is important to consider both abso-
lute weight of tissues as well as tissue weights normalized to
body weight ( Figure 5 ). For absolute weight, most nonadipose
Figure 4. Femoral length ( B ) and bone mineral density estimates ( C ) of mid-diaphyseal cortex in male and female growth hormone receptor gene – disrupted
(GHR − / − ) and wild-type (WT) mice. Bone mineral density estimates were calculated subsequent to scanning on a GE eXplore Locus Small Animal MicroCT Scan-
ner. Data are expressed as mean ± SEM , n = 10 (male GHR − / − and WT), n = 6 (female WT), n = 9 (GHR − / − female). Right femora ( A – D ) of 2-year-old mice were
extracted, mounted in foam, and scanned using the following protocol: 20- m m voxel, 80 kV, 450 m A, and 2000-milliseconds exposure time.
Figure 5. Absolute and normalized organ weights of male and female growth hormone receptor gene – disrupted (GHR − / − ) and wild-type (WT) mice at 2 years of
age. Weights of inguinal (SubQ), perigonadal (PG) for the epididymal fat pad in males and the parametrial fat pad in females, retroperitoneal (Retro), mesenteric
(Mes), spleen, liver, kidney, and heart were in 105-week-old mice at the time of dissection. Data are expressed as mean ± SEM . Asterisk indicates GHR − / − samples
were signifi cantly different than corresponding WT mice.
BERRYMAN ET AL.
tissues in GHR − / − mice were signifi cantly smaller in size in
comparison with WT mice, corresponding to the dwarf size of
the former. That is, absolute weights of liver, heart, and kidney
were decreased in size in comparison with control mice. Using
two-way ANOVA to compare the tissue weights for each
genotype and gender, there was an expected main effect of
genotype for the absolute weights of spleen, liver, kidney, and
heart (with F values ranging from 12.8 to 410, all with a p < .001)
but no signifi cant effect of gender or interaction between
gender and genotype. Many of the signifi cant differences ob-
served in absolute mass of tissues were not maintained when
normalized to body weight, suggesting that the lower tissue
mass in GHR − / − mice is merely proportional to the smaller
body size. One interesting exception for both male and
female mice was the kidneys, which were decreased in
GHR − / − mice for both normalized and absolute values.
Using two-way ANOVA for normalized tissue weights,
there was a main effect of genotype for only relative kidney
weights, F (1,25) = 27.0, p = .00003, with no effect of gen-
der or interaction between gender and genotype.
Adipose tissue showed a different trend. In contrast to
other tissues, the absolute weights of all adipose depots in
male GHR − / − mice were not signifi cantly different from
WT mice. Although not always signifi cant, the trend for
the absolute weights of most adipose depots was to be
reduced in female GHR − / − compared with littermate con-
trols, with the exception of the inguinal subcutaneous fat
pad. Two-way ANOVA for absolute adipose tissue weight
revealed only a signifi cant main effect for the mesenteric
fat, F (1,25) = 5.5, p = .03. When normalized to body
weight, the only signifi cant difference by genotype as
determined by two-way ANOVA was in the subcutaneous
fat pad, F (1,25) = 27.6, p = 0.00003, which had a value
nearly double the WT mice for both male and female mice.
Thus, the subcutaneous fat pad in both male and female
mice was preferentially enlarged relative to body size in
older mice as has been reported previously in younger
male mice ( 18 , 28 ).
The TAG concentration in livers of GHR − / − mice did
not differ signifi cantly from their WT littermates, and there
was no signifi cant difference between males and females
of the same genotype. Specifi c values for male GHR − / − ,
male WT, female GHR − / − , and female WT were 24.7 ± 2.2,
23.5 ± 5.1, 37.5 ± 16.1, and 36.2 ± 7.5 mg/g tissue, respec-
tively. Due to large standard errors and relatively small
group sizes in the cohort from the body composition study,
we repeated liver TAG measurements in the second cohort
of 24-month-old mice. The mean concentrations were sim-
ilar to the fi rst group of mice analyzed (e.g., male WT was
25.6 ± 4.3 mg/g tissue in the second cohort vs 23.5 in the
fi rst group of mice). Again, no signifi cant difference was
found in any group.
Because of their increased longevity, GHR − / − mice have
been the focus of intensive examination. Interestingly,
GHR − / − mice remain long lived despite reports of increased
adiposity ( 18 , 20 , 27 , 28 ), a feature more commonly associ-
ated with reduced longevity. The present study is the fi rst to
monitor body composition longitudinally in the same co-
hort of GHR − / − animals to determine the lifelong changes
in lean and fat mass of this long-lived mouse model. Over-
all, the data for GHR − / − mice show marked increases in
percent body fat mass and no signifi cant difference in per-
cent lean mass throughout the majority of their life, as well
as a decrease in bone mineral density at the end of the study.
There are distinct declines in body weight and adiposity in
later life for all groups of mice studied, which may serve as
a unique indicator of specifi c aging processes or deteriora-
tion in health. Furthermore, the previously reported increase
in the inguinal subcutaneous fat pad in younger male mice
( 16 , 18 , 27 , 29 ) is maintained in later life for male mice, and
this is the fi rst report of a similar depot-specifi c trend in
female mice. Despite these consistencies with previous
reports, there are some important gender- and age-specifi c
differences between GHR − / − and WT groups.
Several previous studies have assessed body composition
in GHR − / − mice. These studies reported consistently that
GHR − / − male mice have higher percent fat mass at 3.5
months ( 20 ), 5 months ( 27 ), 6 months ( 18 ), 7 – 10 months,
and 28 – 32 months ( 28 ) in comparison to controls. The only
exception in the literature is with very young mice (6 – 7
weeks) in which it has been reported that there is no signifi -
cant difference in percent fat mass ( 28 ). Importantly, these
studies were conducted on mice with differing genetic back-
grounds, suggesting that this trend is not strain specifi c in
male mice. The present data for male mice are consistent
with these previous reports except that increased percent fat
mass was seen at all ages, even in male mice as young as 6
weeks old. Female GHR − / − mice have been less rigorously
studied with only one previous study report, indicating that
female mice are also relatively obese but only in older mice
from 2 – 3 years of age; no increase in percent fat mass is
noted in younger ages (6 – 7 weeks and 7 – 10 months) ( 28 ).
Our data show signifi cant increases in percent fat mass for
female GHR − / − mice at all ages except in older (2-year-
old) mice. The apparent discrepancy could be due to the
differing background strain, to the cross-sectional versus
longitudinal experimental design, or the difference in the
methodology used to assess fat mass (Dual-Energy X-Ray
Absorptiometry [DXA] vs NMR).
Surprisingly, there was no signifi cant difference in abso-
lute fat mass despite the extreme dwarf size of the male and
female GHR − / − mice compared with littermate controls.
Although many studies have provided data regarding per-
cent fat mass, only one previous article had reported data on
total absolute fat mass ( 27 ). This manuscript also reported
that there was no signifi cant difference in absolute fat mass.
TWO-YEAR BODY COMPOSITION ANALYSES
Because absolute fat mass is unchanged, it is tempting to
assume that adipose tissue is unscathed by the lack of GH
action and develops to its normal size in these dwarf mice.
However, a compelling argument against this theory is that
the accumulation of adipose tissue in GHR − / − mice is not
uniform, with different depots being disproportionately en-
larged and others being proportional to their dwarf size.
Specifi cally, multiple reports show a profound increase
preferentially in the mass of subcutaneous white adipose
depots in younger male and female GHR − / − mice from 4 to
6 months of age ( 16 , 18 , 27 , 34 ), in interscapular brown adi-
pose tissue in male mice 3 – 4 months of age ( 20 , 29 ), and
occasionally in retroperitoneal fat pad in 3- and 6-month-
old male mice ( 20 , 27 ). Other models of GH defi ciency, such
as the Sma1 mice that have a missense mutation in the GH
gene, show a similar preference in accumulation of fat in
the subcutaneous region ( 35 ). In the present study using
older mice, only the subcutaneous fat pad in GHR − / − mice
is enlarged, with the absolute weight being similar to WT
and with normalized weights being signifi cantly elevated in
both genders. It is interesting to note that the epididymal fat
pad, a male depot commonly studied because of its ease of
dissection, is reduced in terms of absolute mass and not
signifi cantly different when normalized to body weight,
indicating that there is no specifi c accumulation of fat in
this region. As such, studies reporting the impact of GH on
adipose tissue that utilized solely this fat pad should be
repeated using additional fat pads for comparison. Overall,
there is little doubt that an adipose depot – dependent effect
occurs in GHR − / − mice, which now appears to be main-
tained in later life in male and female mice, as demonstrated
in this study. It should be noted that marked increases
in subcutaneous adipose tissue have been reported for
GH-defi cient humans ( 36 , 37 ), although other depots may
also be enlarged ( 38 ).
Lean mass shows a trend opposite of fat mass. That is,
absolute levels of lean mass were signifi cantly decreased in
GHR − / − mice in both genders compared with controls.
Lean mass was remarkably stable for male WT and GHR − / −
mice as well as for female GHR − / − mice after a pubertal
growth spurt in early life. In contrast, female WT mice con-
tinued to have a modest increase in lean mass even at the
last time point measured. Completely opposite of fat mass,
lean mass when normalized to body weight was not signifi -
cantly different between genotypes or genders. Thus, lean
mass was proportional to body weight. Taken together, these
body composition data suggest that the increases in body
weight throughout adulthood and the reduction in body
weight seen in older mice are not due primarily to changes
in lean mass but rather refl ect alterations in body fat. Simi-
lar to these results, aging in humans is generally associated
with increases in total adiposity over the adult life span (re-
viewed in ( 39 )), until extreme old age when fat mass de-
creases ( 40 , 41 ). However, distinct from studies in human
populations, a loss of lean mass also tends to accompany
aging, with lean tissue reductions noted as early as 45 years
of age that continue with advanced aging ( 42 ).
Other studies have tracked body composition changes
over life span in mice. Two studies have utilized mice that
have altered GH function ( 30 , 31 ). In 1993, Donahue and
Beamer ( 31 ) used a chemical method to evaluate body com-
position in GH-defi cient lit/lit male and female mice along
with littermate controls up to 1 year of age. In another arti-
cle, Palmer and colleagues ( 30 ) reported body composition
for male and female bovine GH transgenic mice in the
C57JBl/6J background compared with littermate controls
also up to 1 year of age. Both articles reveal the importance
of assessing body composition in both genders and over
time as differing trends were noted in younger mice than in
older mice as well as in males versus females. For example,
bovine GH transgenic (bGH) male and female mice were
shown to be relatively lean but only at older ages. At younger
ages, the male bGH mice had more fat mass and the female
bGH mice had similar fat mass as littermate controls.
Another observation from this study was that WT male
C57Bl/6J mice have a notable increase in fat mass through-
out Year 1, a phenomenon that was also seen in female mice
yet the increase in fat mass was delayed throughout the pu-
bescent and early adult period. Finally, this study reported
rapid increases in lean mass prior to 24 weeks that stabi-
lized throughout the remainder of the study. Similar age-
and gender-dependent trends were reported for the lit/lit
study. Both previous studies only followed body composi-
tion out to 12 months of age for WT mice. The results of the
present study have some similarities to these previous re-
ports in that WT mice have a steady increase in fat mass
throughout the fi rst year of life, females had a delayed onset
of fat mass gains, and both male and female WT mice had a
rapid pubertal increase in lean mass. As the present study
continued to track WT mice for another year, the current
data add additional information as to the normal changes in
body composition in this commonly studied strain of mice.
WT mice continue to experience gains in body weight and
fat mass, but by 2 years, both male and female WT mice
experienced some loss in body weight, a weight loss that
can almost exclusively be attributed to fat mass loss. Al-
though it is possible that fat mass loss in later life could be
a signal that the health is deteriorating, this same trend is
noted for the long-lived GHR − / − mice.
Several studies have examined the skeletal phenotype as-
sociated with GHR − / − mice. The majority of these reports
have been conducted in young male mice where profound
effects have been noted. For example, GHR − / − mice exhibit
a reduction in epiphyseal plate width of the proximal tibia
by 20% (20 days old) to 30% (10 weeks old) compared with
WT littermates ( 43 , 44 ). Tibial linear growth rate also is re-
duced by ~ 65% in GHR − / − mice between postnatal Days
20 and 40, likely a result of reduced proliferation and hyper-
trophy of chondrocytes ( 43 ). Bone mineral density and
femoral length in 3.5-month-old male GHR − / − mice are
BERRYMAN ET AL.
also decreased by 32% and 74%, respectively ( 45 ). Only one
previous study has examined bone in older GHR − / − mice.
Bonkowski and colleagues ( 28 ) compared several features of
bone in three groups of mice: young (6 – 7 weeks), adult (7 – 10
months), and aged (28 – 32 months) male and female mice
using DXA. They documented a reduction in total body bone
mineral density, bone mineral content, and bone area in
GHR − / − mice, regardless of age and that total body bone min-
eral density either increased or did not change as a function of
age for all genotypes. Relevant for the study conducted herein,
although Bonkowski and colleagues ( 28 ) reported that bone
mineral density was reduced in GHR − / − mice, it was only
female GHR − / − that showed further decreases in femoral
bone mineral density (BMD) with advancing age, suggesting
the presence of both age- and gender-dependent effects. Due
to the two-dimensional nature of DXA scans (grams per
square centimeter), compared with the three-dimensional
capabilities of microCT (milligrams per cubic centimeter),
microCT appears to be a more accurate method for measuring
BMD; thus, the use of different scanning methods could
account for the differences noted between the reports ( 46 ).
Femur length is also decreased in GHR − / − mice in both gen-
ders in this and the study of Bonkowski and colleagues ( 28 ).
Of note, there was a signifi cant effect of gender on femur length
with females exhibiting signifi cantly larger femora than males
of both genotypes. Bonkowski and colleagues ( 28 ) reported a
similar trend that did not reach statistical signifi cance.
High levels of TAG in nonadipose tissues are usually
linked to insulin resistance. Therefore, it is somewhat sur-
prising that the liver TAG content has been shown to be el-
evated in male 18-month-old Ames dwarf mouse, another
model with increased longevity ( 47 ). Furthermore, CR ap-
pears to reduce the TAG content in the Ames dwarf to the
levels seen in WT mice. In contrast, liver TAG content is
reduced in male and female bovine GH transgenic mice, a
model with excess GH action and physiologically opposite
to the model studied in the present study ( 30 , 48 ). Based on
these counterintuitive results from previous studies, we
predicted that liver TAG content might also be elevated in the
insulin-sensitive older GHR − / − mice. This prediction was
not supported by the data in that no signifi cant differences
were identifi ed in liver TAG content of older male or female
GHR − / − mice compared with littermate controls. This fi nd-
ing does not rule out that differences may be present at
younger ages as we have noted numerous other phenotypic
effects related to lipid distribution that are dependent on
age. However, by 2 years, this difference is no longer appar-
ent in two separate groups of mice, which may also be a
unique feature of the GHR − / − model as opposed to the
Ames dwarf mice. Other notable differences between these
models include the additional hormone defi ciencies in the
Ames dwarf mice and their differing responses to CR ( 1 , 11 ).
Of note, patients with Laron syndrome, who have GH gene
deletion comparable to that of GHR − / − mice, have also
been reported to have nonalcoholic fatty livers ( 49 ).
The results generated using the GHR − / − mice demon-
strate that mice can have a disproportionate amount of adi-
pose tissue and remain relatively obese throughout much of
their lives, yet still exhibit improved insulin sensitivity and
longevity. This is an intriguing situation that defi es much of
what we consider dogma. Potentially, the unique and consis-
tent pattern of fat deposition in the subcutaneous region
along with the apparent normal levels of TAG accumulation
in nonadipose tissue may offer insight as to why these mice
live longer. Based on recent transplant studies that demon-
strate inherent benefi cial effects of subcutaneous fat even
when transplanted in a visceral region ( 50 ), it is reasonable
to assume that the unique depot distribution in GHR − / − mice
would alter numerous physiological parameters, such as adi-
pokine expression and glucose homeostasis, among other
possibilities. Regardless, in this long-lived mouse model, it
appears that the excess adipose tissue is “ healthier ” in the
absence of the GHR. Further studies into the mechanisms of
fat metabolism in this mouse model are needed to explore
this phenomenon. Additionally, a better understanding of the
distinctive features of the subcutaneous depot in GHR − / −
mice in comparison to other depots also warrants further
study and will undoubtedly provide clues as to how an obese
state can be accompanied by improvements in many health
parameters. Analyzing other long-lived models for depot-
specifi c differences in fat mass may offer additional evidence
to the unique protective nature of this particular depot.
D.E.B. was supported in part by funds from NIDDK ( DK064905 ) and
National Institute on Aging (NIA; AG031736 ). E.O.L. is supported by funds
from NIA ( AG031736 ). J.J.K. is supported by funds from NIA ( AG19899
and AG031736 ), National Institute of Diabetes and Digestive and Kidney
Diseases ( DK075436 ), the State of Ohio ’ s Eminent Scholar Program that
includes a gift from Milton and Lawrence Goll, and a grant from DiAthegen.
All work was further supported by the Diabetes Research Initiative at Ohio
University and by a grant from American Veterans (AMVETS) .
Address correspondence to Darlene E. Berryman, PhD, RD, LD, W324
Grover Center, School of Human and Consumer Sciences, Ohio University,
Athens, OH 45701. Email: firstname.lastname@example.org .
1. Bartke A , Wright JC , Mattison JA , Ingram DK , Miller RA , Roth GS .
Extending the lifespan of long-lived mice . Nature . 2001 ; 414 : 412 .
2. Miller RA , Chang Y , Galecki AT , Al-Regaiey K , Kopchick JJ ,
Bartke A . Gene expression patterns in calorically restricted mice:
partial overlap with long-lived mutant mice . Mol Endocrinol .
2002 ; 16 : 2657 – 2666 .
3. Zhou Y , Xu BC , Maheshwari HG , et al . A mammalian model for
Laron syndrome produced by targeted disruption of the mouse growth
hormone receptor/binding protein gene (the Laron mouse) . Proc Natl
Acad Sci U S A . 1997 ; 94 : 13215 – 13220 .
4. Snell GD . Dwarf, A New Mendelian Recessive Character Of The
House Mouse . Proc Natl Acad Sci U S A . 1929 ; 15 : 733 – 734 .
5. Li S , Crenshaw EB 3rd , Rawson EJ , Simmons DM , Swanson LW ,
Rosenfeld MG . Dwarf locus mutants lacking three pituitary cell types
result from mutations in the POU-domain gene pit-1 . Nature .
1990 ; 347 : 528 – 533 .
TWO-YEAR BODY COMPOSITION ANALYSES
6. Brown-Borg HM , Borg KE , Meliska CJ , Bartke A . Dwarf mice and
the ageing process . Nature . 1996 ; 384 : 33 .
7. Andersen B , Pearse RV 2nd , Jenne K , et al . The Ames dwarf gene is
required for Pit-1 gene activation . Dev Biol . 1995 ; 172 : 495 – 503 .
8. Flurkey K , Papaconstantinou J , Harrison DE . The Snell dwarf muta-
tion Pit1(dw) can increase life span in mice . Mech Ageing Dev . 2002 ;
123 : 121 – 130 .
9. Flurkey K , Papaconstantinou J , Miller RA , Harrison DE . Lifespan
extension and delayed immune and collagen aging in mutant mice
with defects in growth hormone production . Proc Natl Acad Sci U S A .
2001 ; 98 : 6736 – 6741 .
10. Colman RJ , Anderson RM , Johnson SC , et al . Caloric restriction
delays disease onset and mortality in rhesus monkeys . Science .
2009 ; 325 : 201 – 204 .
11. Bonkowski MS , Rocha JS , Masternak MM , Al Regaiey KA , Bartke A .
Targeted disruption of growth hormone receptor interferes with the
benefi cial actions of calorie restriction . Proc Natl Acad Sci U S A .
2006 ; 103 : 7901 – 7905 .
12. Masternak MM , Al-Regaiey KA , Del Rosario Lim MM , et al . Effects
of caloric restriction on insulin pathway gene expression in the skeletal
muscle and liver of normal and long-lived GHR-KO mice . Exp Gerontol .
2005 ; 40 : 679 – 684 .
13. Masternak MM , Al-Regaiey KA , Del Rosario Lim MM , et al . Effects
of caloric restriction and growth hormone resistance on the expression
level of peroxisome proliferator-activated receptors superfamily in
liver of normal and long-lived growth hormone receptor/binding
protein knockout mice . J Gerontol A Biol Sci Med Sci . 2005 ; 60 :
1394 – 1398 .
14. Masternak MM , Al-Regaiey KA , Del Rosario Lim MM , et al . Caloric
restriction and growth hormone receptor knockout: effects on expres-
sion of genes involved in insulin action in the heart . Exp Gerontol .
2006 ; 41 : 417 – 429 .
15. Coschigano KT , Holland AN , Riders ME , List EO , Flyvbjerg A ,
Kopchick JJ . Deletion, but not antagonism, of the mouse growth
hormone receptor results in severely decreased body weights, insulin,
and insulin-like growth factor I levels and increased life span . Endo-
crinology . 2003 ; 144 : 3799 – 3810 .
16. Liu JL , Coschigano KT , Robertson K , et al . Disruption of growth
hormone receptor gene causes diminished pancreatic islet size and
increased insulin sensitivity in mice . Am J Physiol Endocrinol Metab .
2004 ; 287 : E405 – E413 .
17. Bartke A , Peluso MR , Moretz N , et al . Effects of Soy-derived diets on
plasma and liver lipids, glucose tolerance, and longevity in normal,
long-lived and short-lived mice . Horm Metab Res . 2004 ; 36 : 550 – 558 .
18. Berryman DE , List EO , Kohn DT , Coschigano KT , Seeley RJ ,
Kopchick JJ . Effect of growth hormone on susceptibility to diet-induced
obesity . Endocrinology . 2006 ; 147 : 2801 – 2808 .
19. Dominici FP , Arostegui Diaz G , Bartke A , Kopchick JJ , Turyn D .
Compensatory alterations of insulin signal transduction in liver
of growth hormone receptor knockout mice . J Endocrinol . 2000 ; 166 :
579 – 590 .
20. Egecioglu E , Bjursell M , Ljungberg A , et al . Growth hormone recep-
tor defi ciency results in blunted ghrelin feeding response, obesity, and
hypolipidemia in mice . Am J Physiol Endocrinol Metab . 2006 ; 290 :
E317 – E325 .
21. Hauck SJ , Hunter WS , Danilovich N , Kopchick JJ , Bartke A . Reduced
levels of thyroid hormones, insulin, and glucose, and lower body core
temperature in the growth hormone receptor/binding protein knockout
mouse . Exp Biol Med (Maywood) . 2001 ; 226 : 552 – 558 .
22. Bonkowski MS , Dominici FP , Arum O , et al . Disruption of growth
hormone receptor prevents calorie restriction from improving insulin
action and longevity . PLoS ONE . 2009 ; 4 : e4567 .
23. Al-Regaiey KA , Masternak MM , Bonkowski M , Sun L , Bartke A .
Long-lived growth hormone receptor knockout mice: interaction
of reduced insulin-like growth factor i/insulin signaling and caloric
restriction . Endocrinology . 2005 ; 146 : 851 – 860 .
24. Guo Y , Lu Y , Houle D , et al . Pancreatic islet-specifi c expression of an
insulin-like growth factor-I transgene compensates islet cell growth in
growth hormone receptor gene-defi cient mice . Endocrinology . 2005 ;
146 : 2602 – 2609 .
25. Masternak MM , Al-Regaiey KA , Del Rosario Lim MM , et al . Caloric
restriction results in decreased expression of peroxisome proliferator-
activated receptor superfamily in muscle of normal and long-lived
growth hormone receptor/binding protein knockout mice . J Gerontol
A Biol Sci Med Sci . 2005 ; 60 : 1238 – 1245 .
26. Ikeno Y , Hubbard GB , Lee S , et al . Reduced incidence and delayed
occurrence of fatal neoplastic diseases in growth hormone receptor/
binding protein knockout mice . J Gerontol A Biol Sci Med Sci . 2009 ; 64 :
522 – 529 .
27. Berryman DE , List EO , Coschigano KT , Behar K , Kim JK ,
Kopchick JJ . Comparing adiposity profi les in three mouse models
with altered GH signaling . Growth Horm IGF Res . 2004 ; 14 : 309 – 318 .
28. Bonkowski MS , Pamenter RW , Rocha JS , Masternak MM , Panici JA ,
Bartke A . Long-lived growth hormone receptor knockout mice show a
delay in age-related changes of body composition and bone character-
istics . J Gerontol A Biol Sci Med Sci . 2006 ; 61 : 562 – 567 .
29. Li Y , Knapp JR , Kopchick JJ . Enlargement of interscapular brown
adipose tissue in growth hormone antagonist transgenic and in growth
hormone receptor gene-disrupted dwarf mice . Exp Biol Med (Maywood) .
2003 ; 228 : 207 – 215 .
30. Palmer AJ , Chung MY , List EO , et al . Age-related changes in body
composition of bovine growth hormone transgenic mice . Endocrinology .
2009 ; 150 : 1353 – 1360 .
31. Donahue LR , Beamer WG . Growth hormone defi ciency in ‘ little ’
mice results in aberrant body composition, reduced insulin-like
growth factor-I and insulin-like growth factor-binding protein-3
(IGFBP-3), but does not affect IGFBP-2, -1 or -4 . J Endocrinol .
1993 ; 136 : 91 – 104 .
32. Mystkowski P , Shankland E , Schreyer SA , et al . Validation of whole-
body magnetic resonance spectroscopy as a tool to assess murine body
composition . Int J Obes Relat Metab Disord . 2000 ; 24 : 719 – 724 .
33. Salmon DM , Flatt JP . Effect of dietary fat content on the incidence of
obesity among ad libitum fed mice . Int J Obes . 1985 ; 9 : 443 – 449 .
34. Flint DJ , Binart N , Boumard S , Kopchick JJ , Kelly P . Developmental
aspects of adipose tissue in GH receptor and prolactin receptor gene
disrupted mice: site-specifi c effects upon proliferation, differentiation
and hormone sensitivity . J Endocrinol . 2006 ; 191 : 101 – 111 .
35. Meyer CW , Korthaus D , Jagla W , et al . A novel missense mutation
in the mouse growth hormone gene causes semidominant dwarf-
ism, hyperghrelinemia, and obesity . Endocrinology . 2004 ; 145 :
2531 – 2541 .
36. Tanner JM , Whitehouse RH . The effect of human growth hormone on
subcutaneous fat thickness in hyposomatotrophic and panhypopitu-
itary dwarfs . J Endocrinol . 1967 ; 39 : 263 – 275 .
37. Toogood AA , Adams JE , O’Neill PA , Shalet SM . Body composition
in growth hormone defi cient adults over the age of 60 years . Clin
Endocrinol (Oxf) . 1996 ; 45 : 399 – 405 .
38. Laron Z , Ginsberg S , Lilos P , Arbiv M , Vaisman N . Body composition in
untreated adult patients with Laron syndrome (primary GH insensitivity) .
Clin Endocrinol (Oxf) . 2006 ; 65 : 114 – 117 .
39. Kuk JL , Saunders TJ , Davidson LE , Ross R . Age-related changes in
total and regional fat distribution . Ageing Res Rev . 2009 ; 8 : 339 – 348 .
40. Raguso CA , Kyle U , Kossovsky MP , et al . 3-year longitudinal study
on body composition changes in the elderly: role of physical exercise .
Clin Nutr . 2006 ; 25 : 573 – 580 .
41. Visser M , Pahor M , Tylavsky F , et al . One- and two-year change in
body composition as measured by DXA in a population-based
cohort of older men and women . J Appl Physiol . 2003 ; 94 :
2368 – 2374 .
42. Janssen I , Heymsfi eld SB , Wang ZM , Ross R . Skeletal muscle mass
and distribution in 468 men and women aged 18-88 yr . J Appl Physiol .
2000 ; 89 : 81 – 88 .
BERRYMAN ET AL.
43. Wang J , Zhou J , Cheng CM , Kopchick JJ , Bondy CA . Evidence
supporting dual, IGF-I-independent and IGF-I-dependent, roles for
GH in promoting longitudinal bone growth . J Endocrinol . 2004 ; 180 :
247 – 255 .
44. Davies JS , Gevers EF , Stevenson AE , et al . Adiposity profi le in
the dwarf rat: an unusually lean model of profound growth hormone
defi ciency . Am J Physiol Endocrinol Metab . 2007 ; 292 : E1483 – E1494 .
45. Sjogren K , Bohlooly YM , Olsson B , et al . Disproportional skeletal
growth and markedly decreased bone mineral content in growth
hormone receptor -/- mice . Biochem Biophys Res Commun . 2000 ; 267 :
603 – 608 .
46. Nazarian A , Cory E , Muller R , Snyder BD . Shortcomings of DXA
to assess changes in bone tissue density and microstructure induced
by metabolic bone diseases in rat models . Osteoporos Int . 2009 ; 20 :
123 – 132 .
47. Wang Z , Al-Regaiey KA , Masternak MM , Bartke A . Adipocytokines
and lipid levels in Ames dwarf and calorie-restricted mice . J Gerontol
A Biol Sci Med Sci . 2006 ; 61 : 323 – 331 .
48. Wang Z , Masternak MM , Al-Regaiey KA , Bartke A . Adipocytokines
and the regulation of lipid metabolism in growth hormone transgenic
and calorie-restricted mice . Endocrinology . 2007 ; 148 : 2845 – 2853 .
49. Laron Z , Ginsberg S , Webb M . Nonalcoholic fatty liver in patients
with Laron syndrome and GH gene deletion — preliminary report .
Growth Horm IGF Res . 2008 ; 18 : 434 – 438 .
50. Tran TT , Yamamoto Y , Gesta S , Kahn CR . Benefi cial effects of subcuta-
neous fat transplantation on metabolism . Cell Metab . 2008 ; 7 : 410 – 420 .
Received August 24 , 2009
Accepted October 14 , 2009
Decision Editor: Huber R. Warner, PhD