![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Department of Internal Medicine, Division of Endocrinology, Sahlgrenska University Hospital, SE-41345 Göteborg, Sweden;
Department of Woman and Child Health, Pediatric Endocrine Unit, Karolinska Hospital, SE-17176 Stockholm, Sweden; and
Department of Integrative Pharmacology, AstraZeneca R&D Mölndal, S-43183 Mölndal, Sweden
**
*
2To whom correspondence should be addressed. E-mail: claes.ohlsson{at}medic.gu.se
ABSTRACT
The introduction of transgenic mice as animal models in medical research has increased the need for methods to study the phenotype of mice. The aim of the present study was to develop and evaluate a method for in vivo prediction of fat content in living mice. We combined a modified dual-energy X-ray technique with an image analysis procedure. This combined procedure calculates the percentage of fat area, defined as the percentage of the total area of the mice consisting of >50% fat. A high correlation between the percentage of fat area and dissected adipose tissue was seen in both male and female mice (males, r = 0.92, P < 0.001; females, r = 0.88, P < 0.001). A high correlation was also seen between the percentage of fat area and serum levels of leptin (males, r = 0.95, P < 0.001; females, r = 0.86, P < 0.001). An additional experiment demonstrated a very strong correlation between the percentage of fat area and total body fat as determined by chemical extraction (r = 0.97, P < 0.001). In summary, the percentage of fat area, as measured with the dual-energy X-ray/image combined procedure, provides a good in vivo estimation of total body fat content in mice.
KEY WORDS: • fat • mice • leptin • body mass index • dual energy X-ray absorptiometry
Several researchers studying nutritional and obesity-related
questions are currently using transgenic mouse models. This has
increased the need for methods to study the phenotype of mice. Body
composition measurements are of obvious importance in the study of many
mechanisms of metabolic regulation. Traditionally, these measurements
are performed ex vivo on killed animals, using chemical determinations
of body fat or weighing of dissected tissues. In vivo determinations of
body fat would, in many experimental situations, simplify procedures,
save time and reduce costs. It is also possible to do repeated
measurements before and after treatment in living mice. Serum leptin
levels and body mass index
(BMI)3
are two indirect parameters, which are positively correlated with the
amount of fat in humans and rodents (1
).
Many methods are used for in vivo measurements of body composition in
humans and animals. Dual-energy X-ray absorptiometry (DXA) was
originally developed for the determination of bone mineral content and
areal density This method is now rapidly gaining acceptance for body
composition measurements (2
). The main advantage of the
DXA method is a high precision, generally 1–2% in human measurements,
although accuracy may vary with scanning equipment and software
solutions for scan analysis.
Commercially available DXA equipments have been modified for use in
small experimental animals and successfully applied for bone
determinations in several species, including pigs, rats and mice
(3
–5
). More recent studies have demonstrated that
currently available devices seem to perform well for soft tissue (lean
and fat) measurements in rats and larger animals but not in mice
(3
, 6
).
The Norland pDEXA Sabre (Fort Atkinson, WI) combined with the Sabre Research software (Version 3.9.2) are claimed by the manufacturer to quantify the amount of fat in mice. However, we were unable to obtain any correlation between the amount of fat given by the software and the amount of adipose tissue dissected from the mice or serum leptin levels. We observed that the image given on the setting the percentage of fat was correlated with the amount of adipose tissue dissected. Therefore, we developed a modified technique to estimate body fat in mice based on the DXA/image procedure of the Norland pDEXA Sabre, combined with an image analysis. In this article, we report the first results of body fat determinations in mice using this procedure.
MATERIALS AND METHODS
Mice
Experiment 1. All mice were siblings and of a mixed C57BL/6J/129 background (M&B, Ry, Denmark). The mice were analyzed at 4 mo of age. They had free access to fresh water and food pellets (B&K Universal AB, Sollentuna, Sweden) consisting of cereal products (76.9% barley, wheat feed, wheat and corn germ), vegetable proteins (14.0% hipro soya) and vegetable oil (0.8% soya oil).
Experiment 2. Twenty-four C57Bl/6 female mice (M&B, Ry, Denmark) with initial weight of 20.2–23.6 g were housed in groups of six to seven under controlled conditions 19–21°C, 50% ± 10% humidity and 12:12-h light-dark cycle. Mice were fed on standard, nonpurified diet (R3, Lactamin, Sweden) for 1 wk to allow them to adapt to the new environment. After wk 1, they received a mixed diet (cream cheese, chocolate, almond paste and chocolate cake) and standard, nonpurified diet so that they would gain weight. The food and water were consumed ad libitum throughout the experiment. Body weight was monitored three times a week and mice were killed by cervical dislocation at different time-points during a period of 22 d, so that the weight of the killed mice differed from 21.4 to 34.5 g (-1.8% - +55.1% body weight gain). All animal procedures were approved by the ethical committee at the University of Göteborg.
Dual X-ray absorptiometry (DXA)
DXA measurements were performed with the Norland pDEXA Sabre and
the Sabre Research software (Version 3.9.2). Three mice were analyzed
in each scan. The mice were anesthetized (Figs. 2
and 3)
or killed
(Fig. 4)
before scanning and placed with their stomach down in the DXA.
A dead mouse was included in all the scans as an internal standard to
avoid interscan variations. After scanning, the software percent fat
procedure was used together with a setting that made areas with > 50% fat appear white in the image. The accuracy of this setting was
checked daily with a standard consisting of a container of Plexiglas
with 100% water in one end and 100% olive oil in the other end and a
continuous gradient in between. The image was then printed, scanned and
imported to the software Scion Image (Scion Corporation, Frederick,
MD). The imported image was set to a threshold of 50 arbitrary units,
making lean mass and bone black, whereas the fat area appeared as white
holes in the mice. Thereafter, the analyze particle procedure was
performed, first with the white areas of the mice included (A1 = total mouse area) and then without the white area included (A2 = lean area + bone area). The percentage of fat area was then calculated
as [(A1 - A2)/A1]·100.
|
|
|
Serum leptin levels were measured by an ELISA (90930; Chrystal Chem, Chicago, IL) with intra- and interassay CV of 5.4% and 6.9%, respectively.
BMI and dissection of adipose tissue
The mice were weighed before killing. The crown-rump length was measured on the same scan as the DXA fat was measured and it was defined as the distance between the crown of the skull and a point located in the middle of a line between the two caput femoris. The BMI was then calculated as the body weight (g)/[crown-rump length (mm)]2. The retroperitoneal and gonadal adipose tissues were dissected and weighed (wet weight).
Chemical extraction of fat
After the DXA analysis, mice were homogenized in a blender (IKA T25 basic; Tamro MedLab, Mölndal, Sweden). The samples were hydrolyzed (8 M HCl) and the fat was extracted by dietyl-ether and petroleum-ether (repeated three times). After evaporation, the amount of residual fat was measured by a gravimetric method.
Statistical methods
Linear regression analysis was performed with DXA fat as the independent variable and adipose tissue dissected, leptin levels or total body fat extracted as the dependant variable. Pearson’s correlation coefficient (r) was calculated. Student’s t tests were used to analyze data and P < 0.05 was considered significant.
RESULTS
Development of a DXA/image procedure for determination of the percentage of fat area.
According to the manufacturer, the Norland pDEXA Sabre together with the Sabre Research software (Version 3.9.2) could be used to quantify the amount of fat in mice. However, we could not obtain any correlation between the amount of fat given by the software and the amount of adipose tissue dissected from the mice or the serum leptin levels (data not shown). We noticed that the image given on the setting the percentage of fat seemed to correlate with the amount of dissected adipose tissue. Repeated measurements and evaluations made us choose a setting of > 50% fat as being made white on the image and regarded as fat area. The percentage of fat area was then calculated as described in the Materials and Methods section by a combination of the DXA technique and an image-processing technique. The interassay CV for the measurements of percent fat area was 1.0% for mice with 30% DXA fat area, 2.3% for mice with 3% DXA fat area and 17.6% for mice with 1.4% DXA fat area (the mouse was repositioned before each scan when the interassay CV was calculated). The large CV for mice with 1.4% DXA fat area indicated that the sensitivity of the assay is decreased in very lean mice.
The DXA/image procedure as a predictor of dissected adipose tissue.
We first used the DXA/image procedure to see whether the percent fat
area was correlated with the amount of adipose tissue dissected from
the mice. A high correlation between the percentage of fat area and
dissected adipose tissue was seen in both male and female mice
(Table 1
; Figs. 1
and
2). The serum leptin levels displayed a good correlation with the
dissected adipose tissue (Table 1)
and also with the percentage of fat
area (Fig. 3
), as measured with our DXA/image procedure.
|
|
Calculations of BMI require highly reproducible length measurements. We
have standardized the length measurements using X-ray technique as
described in the Materials and Methods section, resulting in interassay
CV of < 3%. BMI displayed a high correlation with the dissected
adipose tissue in both male and female mice (Table 1)
. However, this
correlation was weaker when males and females were pooled (Table 1)
.
The DXA/image procedure as a predictor of total body fat as determined using chemical extraction.
An additional experiment was performed to evaluate the DXA/image
procedure as a predictor of total body fat as determined using chemical
extraction. Total body fat displayed a very strong correlation with the
percentage of fat area as measured with the DXA/image procedure
(Fig. 4
).
DISCUSSION
We have presented a modified DXA-based method for the in vivo
estimation of body fat in mice. The amount of fat as determined with
the DXA/image procedure was well-correlated with the amount of
adipose tissue dissected in both female and male mice. Furthermore, a
very strong correlation was found between the percentage of fat area
and total body fat as determined by chemical extraction. The DXA/image
procedure makes it possible to follow mice longitudinally in vivo
during development and during pharmaceutical treatments. In the present
study, as previously reported, the serum levels of leptin correlated
very well with the amount of dissected adipose tissue (1
).
Interestingly, the amount of fat, as determined with the DXA/image
procedure, was also correlated with serum levels of leptin, supporting
the notion that the DXA/image procedure can predict the fat content in
mice.
Body composition measurements by DXA make the model assumption that the
subject or animal measured consists of three components that are
distinguishable by their X-ray attenuation properties: fat, bone
mineral and residual lean soft tissue. DXA systems measure the ratio of
photon attenuation at two energy levels. This ratio or R
value can be calculated theoretically from the measured mixture’s mass
fractions and mass attenuation at each energy level. The fat, lean and
bone mineral content of all pixels in the DXA scan can be quantified
from the measured X-ray attenuation, the component R
values, image-processing methods and soft tissue distribution
models (7
). The complexity of the calculations requires
considerable efforts in the development of analysis software, and a
software solution developed for one application, e.g., human total body
or a small animal, cannot always be expected to perform correctly
outside the original application. Currently available DXA systems seem
to perform well for soft tissue measurements in larger animals and also
in small animals such as rats but not in mice (3
, 6
). The
DXA system we used (Norland pDEXA Sabre) together with the Sabre
Research software (Version 3.9.2) are claimed by the manufacturer to
quantify the amount of fat in mice. However, we were unable to obtain
any correlation, between the amount of fat given by the software and
the amount of adipose tissue dissected from the mice or the serum
leptin levels. However, we observed that the image given on the setting
percent fat correlated with the amount of dissected adipose tissue.
Therefore, we developed a simplified technique to estimate body fat in
mice based on the Norland pDEXA Sabre, combined with an image analysis.
Our simplified approach, using only the amount/area of soft tissue
pixels containing 50% fat or more, circumvents some problems of point
typing and assumptions of soft tissue distribution, but it does not
obtain absolute values of body fat in mice. Another disadvantage of the
DXA/image procedure was that the sensitivity of the method for changes
in fat content was decreased in very lean mice. The interassay CV for
the measurements of the percentage of fat area was 1.0% for mice with
30% DXA fat area, 2.3% for mice with 3% DXA fat area and 17.6% for
mice with 1.4% DXA fat area, indicating that the lower sensitivity
limit for the DXA/image procedure is between 1.4% and 3% DXA fat
area.
BMI was well correlated with the amount of dissected adipose tissue in both male and female mice. However, a gender difference was seen with higher BMI in the males than in the females with the same amount of dissected adipose tissue (data not shown). This gender difference probably is because the male mice have more muscle tissue than the females. In summary, the percentage of fat area, as measured with the DXA/image procedure, is a reliable in vivo method to estimate the fat content in mice.
FOOTNOTES
1 Supported by the Swedish Medical Research
Council, the Swedish Foundation for Strategic Research, the Swedish
Cancer Society, the Bergvall Foundation, the Lundberg Foundation, the
Swedish Medical Society, Torsten and Ragnar Söderbergs Foundation
and the Swedish Association against Rheumatic Disease. 
3 Abbreviations used: BMI, body mass index; DXA,
dual energy X-ray.
Manuscript received 9 April 2001. Revision accepted 22 August 2001.
LITERATURE CITED
1. Maffei, M., Halaas, J., Ravussin, E., Pratley, R. E., Lee, G. H., Zhang, Y., Fei, H., Kim, S., Lallone, R., Ranganathan, S., Maffei, M., Halaas, J., Ravussin, E., Pratley, R. E., Lee, G. H., Zhang, Y., Fei, H., Kim, S., Lallone, R., Ranganathan, S., Kern, P. A. & Friedman, J. M. (1995) Leptin levels in human and rodent: measurement of plasma leptin and ob RNA in obese and weight-reduced subjects. Nat. Med. 1:1155-1161.[Medline]
2. Jebb, S. A. (1997) From chemical analysis of the body to metabolic insights provided by the new methodology. Br. J. Nutr. 78(suppl. 2):S101-S112.
3. Elowsson, P., Forslund, A. H., Mallmin, H., Feuk, U., Hansson, I. & Carlsten, J. (1998) An evaluation of dual-energy X-ray absorptiometry and underwater weighing to estimate body composition by means of carcass analysis in piglets. J. Nutr. 128:1543-1549.
4. Andersson, D. K.-P., Arvidsson, A., Ohlsson, C., Sjögren, S., Andersson, K. & Ryberg, B. (1988) Human parathyroid hormone, hPTH(1–84) effects in distal femur following intermittent or continous administration in rats. Bone 23:S631.
5. Windahl, S. H., Vidal, O., Andersson, G., Gustafsson, J. A. & Ohlsson, C. (1999) Increased cortical bone mineral content but unchanged trabecular bone mineral density in female ERß(-/-) mice. J. Clin. Invest. 104:895-901.[Medline]
6. Bertin, E., Ruiz, J. C., Mourot, J., Peiniau, P. & Portha, B. (1998) Evaluation of dual-energy X-ray absorptiometry for body composition assessment in rats. J. Nutr. 128:1550-1554.
7. Pietrobelli, A., Formica, C., Wang, Z. & Heymsfield, S. B. (1996) Dual-energy X-ray absorptiometry body composition model: review of physical concepts. Am. J. Physiol. 271:E941-E951.
This article has been cited by other articles:
|
|
 |
|
  I. Berniakovich, M. Trinei, M. Stendardo, E. Migliaccio, S. Minucci, P. Bernardi, P. G. Pelicci, and M. Giorgio p66Shc-generated Oxidative Signal Promotes Fat Accumulation J. Biol. Chem., December 5, 2008; 283(49): 34283 - 34293. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Melgar, M. Bjursell, A.-K. Gerdin, L. Svensson, E. Michaelsson, and M. Bohlooly-Y Mice with experimental colitis show an altered metabolism with decreased metabolic rate Am J Physiol Gastrointest Liver Physiol, January 1, 2007; 292(1): G165 - G172. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Garcia, I. Wernstedt, A. Berndtsson, M. Enge, M. Bell, O. Hultgren, M. Horn, B. Ahren, S. Enerback, C. Ohlsson, et al. Mature-onset obesity in interleukin-1 receptor I knockout mice. Diabetes, May 1, 2006; 55(5): 1205 - 1213. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-O. Jansson, S. Moverare-Skrtic, A. Berndtsson, I. Wernstedt, H. Carlsten, and C. Ohlsson Leukemia inhibitory factor reduces body fat mass in ovariectomized mice Eur. J. Endocrinol., February 1, 2006; 154(2): 349 - 354. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Astrand, M. Bohlooly-Y, S. Larsdotter, M. Mahlapuu, H. Andersen, J. Tornell, C. Ohlsson, M. Snaith, and D. G. A. Morgan Mice lacking melanin-concentrating hormone receptor 1 demonstrate increased heart rate associated with altered autonomic activity Am J Physiol Regulatory Integrative Comp Physiol, October 1, 2004; 287(4): R749 - R758. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Faldt, I. Wernstedt, S. M. Fitzgerald, K. Wallenius, G. Bergstrom, and J.-O. Jansson Reduced Exercise Endurance in Interleukin-6-Deficient Mice Endocrinology, June 1, 2004; 145(6): 2680 - 2686. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Brommage Validation and calibration of DEXA body composition in mice Am J Physiol Endocrinol Metab, September 1, 2003; 285(3): E454 - E459. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Maeda, K. T. Uysal, L. Makowski, C. Z. Gorgun, G. Atsumi, R. A. Parker, J. Bruning, A. V. Hertzel, D. A. Bernlohr, and G. S. Hotamisligil Role of the Fatty Acid Binding Protein mal1 in Obesity and Insulin Resistance Diabetes, February 1, 2003; 52(2): 300 - 307. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Couplan, C. Gelly, M. Goubern, C. Fleury, B. Quesson, M. Silberberg, E. Thiaudiere, P. Mateo, M. Lonchampt, N. Levens, et al. High Level of Uncoupling Protein 1 Expression in Muscle of Transgenic Mice Selectively Affects Muscles at Rest and Decreases Their IIb Fiber Content J. Biol. Chem., November 1, 2002; 277(45): 43079 - 43088. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
