Relevant conflicts of interest/financial disclosures: The authors have no conflicts of interest to declare.
Quantification of total and visceral adipose tissue in fructose-fed rats using water-fat separated single echo MRI
Version of Record online: 21 MAY 2013
Copyright © 2012 The Obesity Society
Volume 21, Issue 9, pages E388–E395, September 2013
How to Cite
Rönn, M., Lind, P. M., Karlsson, H., Cvek, K., Berglund, J., Malmberg, F., Örberg, J., Lind, L., Ortiz-Nieto, F. and Kullberg, J. (2013), Quantification of total and visceral adipose tissue in fructose-fed rats using water-fat separated single echo MRI. Obesity, 21: E388–E395. doi: 10.1002/oby.20229
- Issue online: 23 SEP 2013
- Version of Record online: 21 MAY 2013
- Accepted manuscript online: 12 DEC 2012 11:15AM EST
- Manuscript Accepted: 19 NOV 2012
- Manuscript Revised: 5 NOV 2012
- Manuscript Received: 20 JAN 2012
- The Swedish Research Council for Environment, Agricultural Sciences, and Spatial Planning (Formas)
The aim of this study was to setup a rodent model for modest weight gain and an MRI-based quantification of body composition on a clinical 1.5 T MRI system for studies of obesity and environmental factors and their possible association.
Design and Methods
Twenty-four 4-week-old female Fischer rats were divided into two groups: one exposed group (n = 12) and one control group (n = 12). The exposed group was given drinking water containing fructose (5% for 7 weeks, then 20% for 3 weeks). The control group was given tap water. Before sacrifice, whole body MRI was performed to determine volumes of total and visceral adipose tissue and lean tissue. MRI was performed using a clinical 1.5 T system and a chemical shift based technique for separation of water and fat signal from a rapid single echo acquisition. Fat signal fraction was used to separate adipose and lean tissue. Visceral adipose tissue volume was quantified using semiautomated segmentation. After sacrifice, a perirenal fat pad and the liver were dissected and weighed. Plasma proteins were analyzed by Western blot.
The weight gain was 5.2% greater in rats exposed to fructose than in controls (P = 0.042). Total and visceral adipose tissue volumes were 5.2 cm3 (P = 0.017) and 3.1 cm3 (P = 0.019) greater, respectively, while lean tissue volumes did not differ. The level of triglycerides and apolipoprotein A-I was higher (P = 0.034, P = 0.005, respectively) in fructose-exposed rats.
The setup induced and assessed a modest visceral obesity and hypertriglyceridemia, making it suitable for further studies of a possible association between environmental factors and obesity.
Obesity is a growing worldwide problem affecting both children and adults, not only in developed countries but also in developing areas. There are some proposed reasons for this growing problem, but no established answers, though the presumed key pathogenetic mechanisms behind obesity are “the Big Two”: increased consumption of energy-dense foods high in saturated fats and sugars, and reduced physical activity. However, it has also been suggested that environmental factors could interfere with the control of energy balance and homeostasis including, e.g., stress , sleep deprivation , adenovirus infection , and intestinal microbiota . Also the kind, not only the amount, of energy may matter , e.g., the types of fats and carbohydrates, and ratios thereof. Environmental pollutants, so-called obesogens, may also promote an increase in fat mass by interfering with the endocrine system and thus altering the finely tuned system, regulating the organism by mimicking, potentiating, or inhibiting the normal endocrine functions, thereby possibly also regulating lipid metabolism and adipogenesis inappropriately to promote obesity [6-10].
To better understand the underlying mechanisms behind such potential association between environmental factors and obesity and to explore epidemiological findings [8, 11, 12], there is a need to develop adequate experimental animal models. There is also a need for accurate techniques that allow detailed studies of body composition. Today the most common measure of obesity in humans is body mass index (BMI). This is a very simple and cheap way to estimate total body fatness, but it does not measure its distribution. Another possible anthropometric method is the waist-circumference model mostly used in humans but also evaluated in the rat by Gerbaix et al. . However, visceral adipose tissue (VAT) poses a larger risk compared to subcutaneous adipose tissue (SAT) [14-16] for outcomes as diabetes and cardiovascular diseases. Therefore, magnetic resonance imaging (MRI) is of great value, as a noninvasive technique that allows direct assessment of the adipose tissue and its distribution. MRI also allows repeated measures, and thereby longitudinal studies of body composition.
This study had two aims. The first aim was to develop a rodent model for studying adipose tissue in an environment of modest caloric excess. The second aim was to develop a method for assessing total, subcutaneous, visceral adipose tissue, and lean tissue volumes using a clinical 1.5 T MRI scanner.
Animal treatment was conducted in accordance with Swedish regulations. The local ethical committee for animal experimentation approved the study.
Twenty-four 4-weeks-old female Fischer rats, supplied by Charles River Laboratories, Salzfeld, Germany, were housed in special wire cages (length, width, height: 59 × 38 × 43 cm) to provide enough space for them to move around freely. The temperature was held at 23°C and relative humidity at 75%, with a 12-h light 12-h dark cycle (light on at 7:00 h). Animals were fed a standard laboratory chow (Nova-SCB RM1A) and water, both ad libitum. Glass bottles were used to reduce contamination from plastic materials (Ancare Corp., Bellamore, NY). After acclimatization for one week, the animals were divided into a control group (n = 12) and a fructose exposed group (n = 12) with six rats in each cage. The weight at the start was 60.9 g ± 4.3 g (mean ± SD) in both groups. The exposed group was given water containing fructose (5% for 7 weeks and then 20% for 3 weeks), and the control group was given water without fructose. Food and liquid intake were measured by weighing (grams of pellet/water/fructose solution per cage) during the entire experiment. Mean consumption per individual and day was calculated. Body weights were measured at the beginning of the study, weekly thereafter and before the rats were euthanized. The MRI scanning was performed over a period of four days at the end of the experiment. The last common weight was used to compare weights between the groups. The weight before euthanization was used in the comparison to the body weights estimated by MRI. Before the MRI exam, the rats were anesthetized with Ketalar 90 mg/kg bw (Pfizer, New York, NY) and Rompun 10 mg/kg bw (Bayer, Leverkusen, Germany). Immediately after the scanning, they were killed by exsanguinations from the abdominal aorta while still under anesthesia.
MR imaging protocol
The MR imaging was performed on a 1.5 T clinical MR system (Achieva; Philips Healthcare, Best, Netherlands) using a quadrature knee coil. The rats lay in prone position. MR compatible pads were used to position the animal in the coil center. The volume of interest, 100 × 100 × 150 mm (sagittal × coronal × axial), was positioned to cover the volume from neck to tail. A spoiled 3D single gradient-echo protocol with imaging parameters repetition time 8 ms, echo time 3.2 ms, and flip angle 12° was used. The acquired voxel size was 0.5 × 0.5 × 1.0 mm. The reconstructed voxel size was 0.45 × 0.45 × 1.0 mm. Slice-encoding direction was feet-head, phase-encoding direction was anterior–posterior, and frequency-encoding direction was left–right. Total imaging time, using one signal average was 4 min 17 s. A receiver bandwidth of 446.4 Hz/pixel was used (i.e., water fat shift was 0.486 pixels). No parallel imaging and no gating was used.
MR image reconstruction and evaluation
Water and fat images were reconstructed from the complex single echo image data using a previously presented model-based method . In brief, the water and fat content in each voxel was determined by use of three assumptions. First, the majority of voxels was assumed to have one of two different water:fat signal ratios. The assumed ratios were 100:0 for muscles and organs, and 0:100 for adipose tissue. Second, the static magnetic field distribution was assumed to be smooth. Third, voxels with an equal amount of fat and water were located on interfaces between water-dominant regions and adipose tissue. The first assumption left two possible alternatives for the static magnetic field in each voxel. Using the second assumption, the right alternative could be selected using optimization. In this study, a multiscale belief propagation approach was used . To allow a continuous spectrum of water:fat ratios, the phase map was filtered using an averaging filter. This step relaxes the initial assumption of only two possible water:fat ratios. However, the final result is biased by the initial assumption, and fat fractions greatly different from the initial assumptions cannot be trusted. This limits measurements of liver fat content, for example. The determination of the static magnetic field distribution allowed direct calculation of the water and fat components. Method feasibility has previously been demonstrated in whole-body scans of a human subject at both 1.5 and 3.0 T .
The quality of the image reconstruction was evaluated by visual identification of locations where the water and fat signal were erroneously separated, commonly denoted swap artifacts, and by manual determination of their volumes. Two operators with experience from water and fat reconstruction techniques inspected all images. Both operators noted what they determined were swap artifacts. Afterwards, the operators jointly reviewed all located artifacts and reached a consensus. The volumes of the swap artifacts from the consensus were determined by manual segmentation by one of the operators.
Volumes of total, visceral, subcutaneous adipose tissue, and lean tissue (TAT, VAT, SAT, LT, respectively) were quantified using a semiautomated approach. Fat fraction images, defined by fat/(fat + water), were calculated, and adipose tissue and lean tissue were separated by thresholding at 50% fat fraction.
To reduce the effect of fat fractions originating from background and low signal regions in the analysis, the tissue of the rats was separated from background by clustering. The water and fat images were clustered into three classes (adipose tissue, lean tissue, and background) using a version of Fuzzy C-means that incorporates spatial continuity . Fat fractions originating from noise in low signal regions were suppressed by multiplication of the background cluster inverse.
The VAT volume was segmented from the fat fraction image using a previously described semiautomated method . The operator manually placed foreground seeds in the VAT depot and background seeds in the SAT, muscles, organs, and in the background. The algorithm then determined the boundary between VAT and other tissues. The operator interactively added/removed seeds in a three-plane view until the segmentation was satisfactory. One operator segmented the VAT depot from all animals. The segmentation was performed twice, on two consecutive days, to measure the reproducibility. The average of the repeated segmentations was used as the VAT volume. The subcutaneous adipose tissue volume was calculated as the difference between the TAT and VAT volumes. An illustration of the different image processing steps is shown in Figure 1.
The weights of the rat volumes imaged were estimated by summing the weights of adipose and lean tissue, assuming densities of 0.923 kg/L and 1.05 kg/L, respectively .
At termination, blood was sampled from the abdominal aorta and centrifuged directly for 15 min to separate plasma. Aliquots were stored at −70°C pending biochemical analyses of the following circulating markers: triglycerides, cholesterol, high-density lipoproteins (HDL), and apolipoprotein A-1 (apo A-1). Blood sampling from one rat in the control group failed because of death during anesthesia.
The liver and the left perirenal fat pad (Figure 2) were dissected and weighed. The liver weight was used to calculate the liver somatic index: (liver weight/bodyweight)*100 (LSI).
Lipids: The analysis of cholesterol and triglycerides was a standard laboratory technique and was performed on an Architect C 8000 analyzer (Abbott Laboratories, Abbott Park, IL, USA) and reported using SI units.
Analysis of protein apo A-I: Prior to Western blot, 1 μL of plasma proteins from fructose-exposed rats (n = 12) and controls (n = 11) was separated on SDS-polyacrylamide gradient gels (T = 5-20%, C = 1.5%) with stacking gels (T = 5%, C = 1.5%) for 1 h (180 V, 60 mA) in electrode buffer (0.15% w/v Tris, 0.72% w/v glycine, 0.05% w/v SDS) using a Mini Protean II electrophoresis cell (Bio Rad). Samples were diluted in sample cocktail (4% w/v SDS, 200 mM DTT, 20% w/v sucrose) and boiled for 3 min.
Total protein in each sample was measured using Bio Rad protein assay. Protein levels in 1 μL of each sample were 65 μg+/−5 μg and did not correlate with apo A-I intensities.
Rat plasma proteins separated by SDS PAGE were transferred to a PVDF membrane. After blocking 1 h (5% milk in TBS) and incubation overnight with primary antibodies 1:1000 (2% milk in TTBS) against apo A-I (rabbit, Ab 20453) or apo E (rabbit, Epitomics#1831-1), the membranes were incubated for 1 h with goat antirabbit HRP-conjugated secondary antibodies 1:40,000 (2% milk in TTBS). Proteins were visualized using an ECL Plus Western Blotting Detection System exposed to x-ray film. Gel images were evaluated using Quantity one gel analysis software, v. 7.1.0 (Bio Rad, Hercules, CA) and protein intensities were determined as optical density/mm2 in % of the total gel density.
Unpaired two tailed t tests were used to test differences between the groups. Pearson correlation coefficients (R2) were used to measure linear correlations. The reproducibility of the VAT segmentation was measured by coefficient of variation defined by standard deviation of the two measurements divided by their mean. The statistical analyses were performed using Prism 5 by GraphPad Software Inc. (San Diego, CA, USA).
|Body weight week 10 (g) (mean ± SD)||159.8 ± 5.97||12||166.7 ± 6.20||12||0.010|
|Weight gain week 1–7 (g) (mean ± SD)||88.7 ± 5.59||12||94.3 ± 6.21||12||0.028|
|Weight gain week 8–10 (g) (mean ± SD)||12.0 ± 2.92||12||11.5 ± 2.02||12||0.66|
|Weight gain week 1–10 (g) (mean ± SD)||100.6 ± 5.46||12||105.8 ± 6.35||12||0.042|
|Food week 1–7 (g/rat/day)||9.2; 9.5||2a||9.1; 9.4||2a|
|Food week 8–10 (g/rat/day)||8.5; 8.5||2a||5.3; 5.8||2a|
|Food week 1–10 (g/rat/day)||9.0; 9.2||2a||8.1; 8.4||2a|
|Liquid week 1–7 (g/rat/day)||12.7; 9.7||2a||17.7b; 18.7b||2a|
|Liquid week 8–10 (g/rat/day)||10.5; 9.4||2a||22.4c; 18.2c||2a|
|Liquid week 1–10 (g/rat/day)||12.1; 9.6||2a||19.0; 18.5||2a|
|Food energy week 1–7 (kcal/rat/day)||26.1; 26.9||2a||25.7; 26.6||2a|
|Food energy week 8–10 (kcal/rat/day)||24.0; 24.1||2a||15.0; 16.5||2a|
|Food energy week 1–10 (kcal/rat/day)||25.4; 25.9||2a||22.1; 23.2||2a|
|Fructose energy week 1–7 (kcal/rat/day)||0; 0||2a||3.53b; 3.74b||2a|
|Fructose energy week 8–10 (kcal/rat/day)||0; 0||2a||17.5c; 14.5c||2a|
|Fructose energy week 1–10 (kcal/rat/day)||0; 0||2a||8.0; 7.3||2a|
|Total energy week 1–7 (kcal/rat/day)||26.1; 26.9||2a||29.2; 30.3||2a|
|Total energy week 8–10 (kcal/rat/day)||24.0; 24.1||2a||32.5; 31.0||2a|
|Total energy week 1–10 (kcal/rat/day)||25.4; 25.9||2a||30.1; 30.5||2a|
|Controls (mean ± SD)||N||Fructose-exposed (mean ± SD)||N||P|
|Body weight (g)||160.9 ± 5.76||12||167.8 ± 6.25||12||0.010|
|Estimated body weight, MRI (g)||119 ± 4.9||9||122 ± 4.9||8||0.20|
|Total adipose tissue (cm3)||23.6 ± 2.1||9||28.8 ± 5.3||8||0.017|
|Visceral adipose tissue (cm3)||12.6 ± 1.64||9||15.7 ± 3.1||8||0.019|
|Subcutaneous adipose tissue (cm3)||11.0 ± 1.1||9||13.2 ± 2.4||8||0.031|
|Lean tissue (cm3)||92.5 ± 4.03||9||91.0 ± 1.98||8||0.35|
|Fat pad (g)||0.67 ± 0.069||12||0.85 ± 0.10||12||<0.001|
|Fat pad/bodyweight ratio (%)||0.42 ± 0.043||12||0.51 ± 0.057||12||<0.001|
|Liver (g)||4.70 ± 0.10||12||5.24 ± 0.37||12||0.001|
|Liver/bodyweight ratio (%)||2.9 ± 0.18||12||3.1 ± 0.19||12||0.012|
|Cholesterol (mmol/L)||2.6 ± 0.18||11||2.6 ± 0.18||12||0.92|
|Triglycerides (mmol/L)||1.15 ± 0.36||11||1.50 ± 0.39||12||0.034|
|HDL (mmol/L)||0.74 ± 0.050||11||0.81 ± 0.051||12||0.003|
|Apo A-1 (optical density, %)||8.66 ± 1.07||11||10.46 ± 1.61||12||0.005|
Rats in the fructose group consumed substantially more liquid than rats in the control group during the whole study. During the last three weeks, when the fructose concentration was changed from 5% to 20%, this consumption increased, but at the same time, the chow intake decreased in the fructose group. Energy intake was fairly constant, but the amount of energy from fructose rose from about 12% to 50% (Table 1).
As shown in Table 1, the weight gain was significantly greater in rats exposed to fructose than in controls (P < 0.05), and the main difference in weight gain took place during the first seven weeks (Figure 3). The mean fat pad weight was 27% greater in the fructose group (P < 0.001), and the ratio fat pad/body weight was 21% greater, compared with the control group (P < 0.001). Liver weights and LSI were also greater (P = 0.001 and P = 0.012) in the fructose-exposed group.
Regarding the circulating markers, the most pronounced effects were seen for triglycerides (Table 2) and HDL associated protein apo A-I (Table 2, Figure 4). The levels of triglycerides and apo A-1 were 30% (P = 0.034) and 21% (P = 0.005) higher, respectively, in the fructose-exposed individuals than in the controls.
Technical problems with cooling of the MRI equipment resulted in four missing measurements in the exposed group and three missing measurements in the control group.
From the evaluation of the image quality, the average volume of swap artifacts was 0.020 cm3, corresponding to 0.017% of total body volume (range 0-0.109 cm3 or 0-0.09%). No swap artifacts were noted in six of the image volumes. Swap artifacts were typically located in interfaces between air and tissue. Metal artifacts were seen in the abdomen of one of the animals in the control group, affecting a volume of 1.22 cm3 (1.09%).
The time required for the semiautomated segmentation was 3 min 17 s ± 51 s per animal, including image loading and saving. The coefficient of variation from the repeated segmentations of VAT was 0.71% ± 0.71%.
MRI measurements showed that rats in the fructose group had greater TAT, VAT, and SAT volumes compared with rats in the control group (P = 0.017, 0.019, and 0.031, respectively), while LT volumes did not differ significantly between the two groups (Table 2). The fat pad weights were best correlated with VAT volume (R2 = 0.59, P = 0.0003) (Figure 5) and not with LT volume (R2 = 0.024, P = 0.55). The body weights estimated from the MRI data were, as expected, strongly correlated to body weight, and underestimated compared to the real weights (Table 2, Figure 6).
It has been shown that a single echo-based reconstruction of water and fat images, from data collected on a clinical 1.5 T MRI scanner, in combination with a highly reproducible semiautomated postprocessing can be used for detailed assessment of rat body composition. The results also show that feeding with an addition of only 5% fructose in the drinking water could increase fat mass in rats. These results support the model set up as suitable for studies of obesity, and how different environmental factors affecting weight regulation could act in an environment of a modest caloric excess, mimicking a contemporary lifestyle.
The fructose model
Our main purpose in adding fructose was to give the rats a moderate excess of calories. Based on others research about different sources of sugar, fructose was the best choice. Results regarding the association between different forms of sugar and obesity/metabolic syndrome point to fructose, which is almost entirely cleared by the liver, contributing to adverse effects, e.g., altering the triglycerides in the circulation .
Other studies show that rats are good at spontaneously balancing their diet regarding energy content , but our knowledge of how different energy sources affect the metabolism and interact with other stressors and contaminants is limited. In this study, the fructose-exposed rats ate a lower amount of chow than the controls provided with tap water. After seven weeks, when the fructose concentration increased from 5% to 20%, the difference in liquid intake also increased, and the fructose group consumed about twice amount of liquid, but decreased their food intake by one third compared with the control group. The energy intake was even higher in the fructose-exposed group. Notably, the weight gain did not differ between the groups during this period. However, since the measurement of liquid and energy intake was not performed at the individual level, no firm conclusions could be drawn about the relationships between the impact of the quantitative intake of liquid and different nutrients on weight gain or fat tissue development.
Significantly higher levels of the HDL associated protein apo A-I, responsible for reverse cholesterol transport, were detected in fructose-exposed rodents than in controls (Figure 4). It has to be taken into account though, that apo A-I levels has been measured by Western blot with plasma volume and total protein content as loading controls. Optimal procedure would have been a strip of the PVDF membrane after the apo A-I blot followed by a second blot toward a “stable” protein such as actin in tissue preparations as loading control, but to our knowledge, there are no such proteins in plasma. 2-DE quantification based on staining was not an option because of low plasma volumes and ELISA would include the risk of unspecific binding contributing to the total signal.
The higher apo A-I levels is in line with the higher HDL levels found in the exposed group in this study. In contrast to humans, rodents carry most of their cholesterol in HDL . Therefore, we suggest that the higher apo A-I and HDL observed reflect metabolic alterations in the fructose exposed rats. Interestingly, increased HDL has previously been described in diabetic rats exposed to sucrose, and according to Eisenberg et al., a possible explanation might be reduced clearance rates .
In addition, higher liver weight was found in fructose-exposed rats. Although not histologically examined, it is likely that this represents fat infiltration in the liver, since we also found higher levels of triglycerides in the circulation. In this respect, the results are similar to what is seen in a positive caloric balance in humans. The higher triglyceride levels seen in the fructose exposed rats are also consistent with the results from a study by Dai and McNeill where they examined the effect of different concentration of fructose in male Wistar rats . They found the greatest increase of triglycerides at a concentration of 10%, supplied via the drinking water, a lesser increase for a 20% solution, and the lowest increase for a 5% solution, which was about the same magnitude as that seen in the present study. Regarding bodyweight gain, there was no significant difference between any of the groups in their study, though the group exposed to 10% fructose solution had the highest mean bodyweight gain during the exposure.
The MRI method
Imaging-based studies of body composition benefit from high image resolution and good adipose tissue to lean tissue contrast. The chemical shift based water–fat reconstruction technique used in this study allowed collection of relatively high-resolution image data in one single echo scan on a clinical 1.5 T scanner. The reconstructed images allow creation of fat fraction images that are in absolute scale, free from intensity nonuniformities, and based on fat and water-specific chemical shifts. This entails multiple theoretical advantages over, for example, T1-weighted images, which are commonly used in studies including adipose tissue quantification. Water–fat image reconstruction has also previously been performed on single echo acquisitions, however, with different solutions to the separation problem [26-28]. To the best of our knowledge, this is the first time these images are used for adipose tissue quantification.
Body composition of rodents has also been studied previously using MR-based methods [29-35]. Magnetic field strengths used span a wide range from 0.005 to 9.4 T. A low field system has previously been reported to yield reproducible results in living rats . However, this method does not allow analysis of adipose tissue distribution. Methods that combine two images, where the contrast is varied by spectral saturation of water and/or fat [30, 31] or flip angle  have also previously been described. These methods rely on acquisition of two image volumes and are therefore more time consuming and sensitive to motion. A recent study presented by Johnson et al.  uses a method similar to that presented in this paper. Their method also separated water and fat signal in postprocessing based on chemical shift. However, it uses four images with different echo times and therefore requires longer imaging time. An advantage of their method, and the use of multiple echo times, is that it allows more accurate spectral modeling and analysis of tissue fat infiltration, e.g., in the liver.
The quality of the image reconstruction was deemed sufficient for this application. Only minor swap artifact volumes were found (on average less than 0.1% of the total adipose tissue volume). The identification of swap artifacts was based on deviations from the correct anatomical features. The water image was mainly used for this, but the fat and the sum images were also used when needed. In complex regions, such as peri-intestinal, it is difficult to determine deviations from the correct anatomical features. The evaluation was therefore limited to regions were these deviations could be reliably detected.
The semiautomated segmentation was found to be relatively fast and to give highly reproducible segmentation of the VAT volume from the images used in this study. The use of a semiautomated segmentation also allows exclusion of fat signal from intestines, which is very difficult to achieve using fully automated methods.
The body weights estimated from the MRI data were underestimated, as expected, since the whole rats were not imaged and since bone and lung tissue was not included in the estimation. Nevertheless, the strong correlation between estimated and real rat weights (Figure 6), in combination with the good adipose to lean tissue contrast indicate that the MR method can reliably detect changes and differences in TAT and LT.
The perirenal fat pad is a quite large distinct fat depot, which is easy to distinguish and dissect and thus offers the possibility weighing a clearly defined piece of fat. Other fat depots have structures (or lack structure) that make them unsuitable to handle in a systematic way. We wanted to evaluate how well the fat pad correlated to the MRI measurements to validate the fat pad as an alternative fat measure to the more precise MRI measurement. MRI is an expensive way to analyze rats, so we speculated that if a good correlation was achieved the fat pad might be very useful as a cheaper and easier way to measure fat storage. The result was a fairly strong relationship between the VAT volume measured by MRI and the fat pad weight (Figure 5). Both depots were further found significantly larger in the fructose group.
There is a balance between animal welfare and precision in the measurements. On one hand social housing promotes motion and reduces psychological stress in the animals, but on the other hand, it makes it difficult to monitor the individual intake of food and liquid. Therefore, no formal statistic correlation analysis was performed between the intake of liquid and different nutrients on weight gain or fat tissue development.
We have presented a model with fructose-exposed rats evaluated by a noninvasive MRI method that allows assessment of adipose tissue volumes and distribution in living rats. Similar MRI methods have been used in studies in humans, which facilitates translational research. The model provides a useful starting-point to study the action of environmental factors, affecting fat mass under conditions of modest caloric excess.
- 12Relationship between serum concentrations of persistent organic pollutants and the prevalence of metabolic syndrome among non-diabetic adults: Results from the National Health and Nutrition Examination Survey 1999–2002. Diabetologia 2007;50( 9):1841–1851., , , , .
- 17Single-image water-fat separation in International Society for Magnetic Resonance in Medicine 2010, Stockholm, No 2907., , , .
- 26Improved single point water-fat imaging with virtual shimming. Proc Intl Soc Magn Reson Med 2001;9:789..
- 27Three-dimensional T1-weighted MR imaging using a one-point Dixon technique with arbitrary echo times. Proc Intl Soc Magn Reson Med 2005;13:893., , .