Bovine Growth Hormone Transgenic Mice Are Resistant to Diet-Induced Obesity but Develop Hyperphagia, Dyslipidemia, and Diabetes on a High-Fa
http://www.100md.com
内分泌学杂志 2005年第2期
Department of Physiology (B.O., M.B.-Y., S.M.F., F.F., A.L., J.T., G.B., J.O.), G?teborg University, S-405 30 G?teborg, Sweden; Department of Internal Medicine (B.O., M.B.-Y., J.T.), Division of Endocrinology, Wallenberg Laboratory for Cardiovascular Research (A.L., J.O.), Sahlgrenska University Hospital, S-413 45 G?teborg, Sweden; Department of Medicine (B.A.), Lund University, S-22184 Lund, Sweden; and AstraZeneca R&D (M.B., F.F., J.T., J.O.), S-43183 M?lndal, Sweden
Address all correspondence and requests for reprints to: Bob Olsson, Department of Internal Medicine, Sahlgrenska University Hospital, G?teborg University, Vita Str?ket 12, S-413 45 G?teborg, Sweden. E-mail: bob.olsson@medic.gu.se.
Abstract
It is known that bovine GH (bGH) transgenic mice have increased body mass, insulin resistance, and altered lipoprotein metabolism when fed a normal diet (ND). In this study, the effects of 8 wk of high-fat diet (HFD) were investigated in 6-month-old male bGH mice. Although littermate controls had unchanged energy intake, energy intake was higher in the bGH mice on a HFD than on a low-fat diet. Nevertheless, the bGH mice were resistant to diet-induced weight gain, and only in the bGH mice did the HFD result in increased energy expenditure. Glucose oxidation was higher in the bGH mice compared with littermate controls on both a HFD and ND. In addition, the bGH mice had 0.5 C higher body temperature throughout the day and increased hepatic uncoupling protein 2 expression; changes that were unaffected by the HFD. On a HFD, the effect of bGH overexpression on serum triglycerides and apolipoprotein B was opposite to that on a ND, resulting in higher serum concentrations of triglycerides and apolipoprotein B compared with littermate controls. Increased serum triglycerides were explained by decreased triglyceride clearance. The HFD led to diabetes only in the bGH mice. In conclusion, bGH transgenic mice were resistant to diet-induced obesity despite hyperphagia, possibly due to increased energy expenditure. On a HFD, bGH mice became dyslipidemic and diabetic and thereby more accurately reflect the metabolic situation in acromegalic patients.
Introduction
GH TRANSGENIC MICE have been extensively used to study long-term effects of GH overproduction. GH transgenic mice have increased body size (1), organomegaly (2), and reduced body fat mass (3, 4, 5). The correlation between GH and body fat is illustrated in ovine GH gene transgenic mice controlled by the metallothionine promoter, where zinc is used to regulate GH dosage. In that model, epidydimal and sc fat mass were found to decrease linearly with increased GH levels (4). GH transgenic mice are normoglycemic but highly hyperinsulinemic and show slightly impaired or normal glucose tolerance when fed normal chow (6). The bovine GH (bGH) transgenic mice also show several alterations in lipid and lipoprotein metabolism. These alterations include lower very-low-density lipoprotein (VLDL) levels, triglyceride levels, and hepatic triglyceride secretion rate and increased lipoprotein lipase activity (3). Furthermore, bGH mice also have higher high-density lipoprotein (HDL) and total cholesterol levels (2, 3).
Patients with acromegaly and GH transgenic mice share similar traits resulting from the elevated serum levels of GH, e.g. organomegaly, decreased body fat content (7), insulin resistance (8), and disturbed lipoprotein metabolism (9, 10). However, in contrast to acromegalic patients, who have elevated serum triglyceride levels and, in some individuals, elevated triglyceride secretion (9, 10), bGH mice have lower triglyceride levels (3). Moreover, acromegalic patients develop hyperglycemia, which does not occur in GH transgenic mice. The reason for these discrepancies may be due to differences between the species or due to the differences in terms of changes in other hormonal axis, e.g. unchanged or substituted cortisol deficiency in acromegaly and increased corticosterone levels in GH transgenic mice (11). However, the differences may also be attributable to the larger intake of fat in humans than that contained in the ordinary mouse diet.
Other aspects of the regulation of metabolism that may be influenced by GH include food intake, energy expenditure (EE), and metabolic fuel preference, measured as respiratory exchange ratio (RER). The effect of GH on food intake has been studied both in man and experimental animals. GH has been shown to increase food intake in children (12) and in experimental animals (13) but not in all studies (14, 15, 16) including bGH transgenic mice given ordinary low-fat chow (5). It has also been shown that eating behavior changes after GH administration (16). However, no studies have been performed to investigate the effect of GH on intake of low- vs. high-fat diet (HFD). It is generally accepted that GH increases lipolysis and fatty acid oxidation (17). However, the effect of GH overproduction, as in the case of acromegaly, on RER is less clear (18, 19), as is the relation between estimates of GH release and RER in young and old adults (20).
It is a general finding in clinical GH studies that GH treatment (21), as well as acromegaly, results in increased EE (18, 19). However, few studies have addressed the influence of GH on EE (22) or body temperature in rodents (23). In a previous paper (11), we observed that bGH transgenic mice have an increased locomotor activity, indicating that they have an increased EE, but basal metabolic rate has not previously been investigated in these mice.
The consumption of a HFD fed ad libitum changes body composition and metabolic status, resulting in diet-induced obesity and altered levels of cholesterol, triglycerides, and insulin (24, 25, 26). However, important genetic differences exist, as illustrated by the observations that different mouse strains have different susceptibility to diet-induced obesity (26, 27). Moreover, when findings in animal models are extrapolated to the human situation, HFD feeding better mimics the human diet in the Western world, where 30–40% of the dietary energy comes from fat, compared with less than 10% in normal mouse chow.
The effect of GH on body composition, food intake, and EE, as well as lipid- and carbohydrate metabolism, during the influence of a HFD has not previously been studied. In this study, the effect of 8 wk of HFD was investigated in 6-month-old male transgenic bGH mice. We conclude that bGH transgenic mice are resistant to diet-induced obesity despite hyperphagia, possibly due to increased resting EE. The bGH transgenic mice became dyslipidemic and diabetic on a HFD and thereby more accurately reflect the metabolic situation in acromegaly compared with those fed a low-fat diet.
Materials and Methods
Animals
In this study, 5- to 6-month-old bGH transgenic mice and littermate controls were used. The bGH mice have previously been described by Sandstedt et al. (28). The mice were housed with a 12-h light, 12-h dark cycle (0700–1900 h, with a 1-h dawn/sunset function) and had free access to tap water and mouse standard chow (R-34, Lactamin, Vadstena, Sweden) or HFD (R-638, AnalyCen Nordic AB, Lidk?ping, Sweden). The standard chow contained (in energy percent) 9.4% fat, 20.2% protein, 0.8% fiber, and 69.6% nonfat energy; and the HFD contained 39.9% fat, 17% protein, 0.7% fiber, and 42.3% nonfat energy. However, no cholesterol was added to the HFD. In all of the HFD experiments, the mice received a HFD for 8 wk before the analysis, except in one experiment of body temperature measurements where the mice only received a HFD for 4 wk. Six bGH male mice and eight littermate controls on either normal chow or a HFD were housed three per cage in two cages (bGH mice) or four per cage (controls) in two cages during recording of food intake. Every morning, for 5 consecutive days, the food in each cage was weighed.
The mice were anesthetized using medetomidine (Domitor, Orion Espoo, Finland; 0.5 mg/kg, ip) and ketamine (Ketalar, Parke-Davis, Detroit, MI; 75 mg/kg, ip) and killed by heart puncture. The liver, white adipose tissue (WAT), brown adipose tissue (BAT), and soleus muscle were excised and immediately frozen in liquid nitrogen and stored at –135 C. The study was performed after prior approval from the local ethics committee for animal experimentation at the G?teborg University, Sweden.
Real-time PCR
Liver, skeletal muscle, BAT, and WAT from five bGH males and littermate controls on normal diet (ND) were homogenized. In addition, livers from five bGH mice and littermate controls on a HFD for 8 wk were homogenized. Total RNA was then extracted by Tri Reagent (Sigma Diagnostics, St. Louis, MO). First-strand cDNA was synthesized from total RNA using Superscript preamplification system (Life Technologies, Rockville, MD). Real-time PCR analysis was performed with an ABI Prism 7900 Sequence Detection System (Applied Biosystems, Foster City, CA) using FAM and TAMRA- or VIC and TAMRA-labeled fluorogenic probes. The expression of uncoupling protein (UCP) 1, UCP2, UCP3, peroxisome proliferator-activated receptor (PPAR), sterol regulatory element-binding protein (SREBP)-1, apolipoprotein (apo)CIII, and fatty acid synthase (FAS) were normalized against mouse acidic ribosomal phosphoprotein P0 (M36B4). The relative expression levels were calculated according to the formula: 2–CT, where CT is the difference in cycle threshold values between the target and the M36B4 internal control (User Bulletin no. 2, PerkinElmer, Foster City, CA).
Surgical implantation of telemetry transmitters and signal acquisition
Radiotelemetry transmitters, designed to detect core body temperature, were implanted in bGH mice and littermate controls. The mice were anesthetized using medetomidine (Domitor, 0.5 mg/kg, ip) and ketamine (75 mg/kg, ip). A telemetry transmitter (TA11ETA-F20, weight 3 g; Data Sciences International, Inc., St. Paul, MN) was then implanted into the abdominal cavity through a midline incision. The abdominal incision was closed with staples. Anesthesia was reversed by atipamezole (Antisedan, Orion Espoo, Finland, 4 mg/kg, ip). Mice were allowed 5–7 d to recover from surgery. The cage with the animal was placed on a receiver plate, and the signal was collected using the Dataquest LabPRO Acquisition System (version 3.01, Data Sciences International, Inc.). The temperature signal was sampled at a frequency of 500 Hz, in 6-sec bursts, every 2 min for 48 h. Body temperature was then recalculated as 2-h averages and displayed as a 24-h curve.
Indirect calorimetric analyses
To determine the thermoneutral zone of bGH mice and littermate controls, the resting metabolic rate was measured at 24 C, 28 C, 32 C, and finally 36 C, for 2 h at each temperature, by indirect calorimetric analysis. A 1-h period was allowed for acclimatization, after which the VO2 and EE were determined at the lowest level of O2 consumption (VO2) during the final hour of the measuring period and were expressed as milliliters of O2 x hours–1 x (kilograms of body weight)–0.75. EE (kcal/h·kg) was calculated using a rearrangement of the Weir equation as supplied by Columbus Instruments: (3.815 + 1.232 x RER) x VO2.
VO2 and CO2 production were measured using an indirect open circuit calorimeter with a flow rate of 500 ml/min and a chamber vol of 2.7-liters (Oxymax, Columbus Instruments, Columbus, OH; cage dimensions, 20 x 10 x 12.7 cm). The system was calibrated daily using a standard gas mixture (0.49% CO2-20.5% O2-79.0% N2). Measurements were done pair-wise, and expired air was analyzed for a 30-sec period every 6 min using an electrochemical O2 analyzer and a CO2 sensor (Oxymax). Mice were conscious and unrestrained during the measuring period but did not have access to food or water during the VO2 and EE measurements.
After 8 wk on the ND or HFD, VO2 and EE of individual animals were measured at thermoneutrality (30 C). One hour of acclimatization was allowed, after which VO2 and EE were determined at the lowest level of VO2 during the final 2 h of the measuring period and were expressed as milliliters of O2x hours–1 x (kilograms of body weight)–0.75.
To determine the RER, the mice had free access to water and either normal chow or HFD ad libitum. The RER was calculated as VCO2/VO2 [volume of CO2 produced per volume of O2 consumed (milliliters per kilogram per minute)].
IVGTT (iv glucose tolerance test)
IVGTT was performed in 6-month-old bGH mice and littermate controls fed a ND or HFD for 8 wk. The studies were performed in the late morning hours after 4 h withdrawal of food from the cages. The animals were anesthetized with midazolam (Dormicum, Hoffman-La-Roche, Basel, Switzerland; 0.4 mg/mouse, ip) and a combination of fluanison (0.9 mg/mouse) and fentanyl (0.02 mg/mouse; Hypnorm, Janssen, Beerse, Belgium). Thereafter, a blood sample was taken from the retrobulbar, intraorbital, capillary plexus, after which the animals were given an iv injection of D-glucose (1 g/kg; British Drug Houses, Poole, UK). The vol load was 10 μl/g body weight. Blood samples were then taken at 1, 5, 10, 20, 30, and 50 min (75 μl) after the glucose load. The samples for glucose and insulin were taken in heparinized tubes; and after immediate centrifugation, plasma was separated and stored at –80 C until analysis. From the IVGTT, the glucose elimination rate was calculated as the glucose elimination constant (KG), i.e. the elimination of glucose in percent per minute from min 1 to min 20 after iv glucose administration, as calculated after logarithmic transformation of data.
Serum and plasma analyses
Triglyceride and cholesterol concentrations were determined by enzymatic colorimetric assays (MPR2: TG/GPO-PAP and Chl/CHOD-PAP, Roche Diagnostics, Mannheim, Germany). Serum apoB concentrations were determined by an electroimmunoassay as previously described (3, 29).
Plasma insulin was determined using a RIA with the use of a guinea pig antirat insulin antibody, 125I-labeled human insulin as tracer, and rat insulin as standard (Linco Research, St. Charles, MO). Plasma glucose was determined with the glucose oxidase method (30).
Size distribution of serum lipoproteins
Determination of size distribution of lipoproteins was performed by gel filtration using fast protein liquid chromatography (FPLC) equipment (Pharmacia Upjohn, Uppsala, Sweden) as described previously (3). Briefly, serum from six mice was pooled to a total vol of 1.5 ml and the density adjusted to 1.215 g/ml with KBr in 0.9% NaCl. After ultracentrifugation (35,000 x g, 4 C, 24 h), the total lipoprotein fraction was recovered by aspiration and the final vol adjusted to 2 ml with FPLC-buffer (0.15 M NaCl, 0.01% EDTA, 0.02% NaAz, pH 7.3). After filtration through a 0.45-μm low-protein filter, the sample was loaded on a 25-ml Superose 6B column (Pharmacia Upjohn) using a constant flow rate of 0.35 ml/min. Eluted samples were collected in 0.5-ml fractions. The fractions were stored at –20 C until assay. Triglycerides and cholesterol concentrations were determined with enzymatic colorimetric assays as described above. Western blots for determination of apoB in the fractions were performed as described previously (3).
In vivo hepatic triglyceride secretion rate
Triglyceride secretion rate in vivo was measured after iv administration of Triton WR1339 as described before (3, 31). After a 5-h period without access to mouse standard chow (0700–1200 h), anesthetized bGH mice and littermate controls were injected iv with Triton WR1339 (Sigma Diagnostics) diluted in saline (200 mg/ml) via the jugular vein (500 mg/kg body weight). Blood samples (70 μl) were taken before and 30, 60, and 90 min after the Triton WR1339 injection. The triglyceride accumulation was linear during this time period. The triglyceride concentration was analyzed as described above. Triglyceride clearance (milliliters per minute) was calculated from the baseline fasting triglyceride concentration (micromoles per milliliter) and the hepatic triglyceride secretion rate (micromoles per minute).
Hepatic lipid content
Hepatic triglyceride content was determined as described before (3). In brief, the livers were homogenized, and lipids were extracted. After evaporation of the final chloroform phase, lipids were dissolved in isopropanol, and triglyceride concentrations were determined as described above.
Statistics
Values are given as means ± SEM. Comparison between groups was performed with unpaired Student’s t test or two-way or one-way ANOVA followed by Bonferroni’s test. P values less than 0.05 were considered significant. When appropriate, values were normalized by logarithmic transformation.
Results
Food intake and body composition
Male bGH mice and littermate controls of 6 months of age were given a HFD for 8 wk and compared with corresponding mice receiving a ND. bGH mice fed a HFD had a similar final weight and weight gain as those receiving normal chow (Table 1). However, littermate controls receiving the HFD had an increased body weight gain and final body weight compared with those receiving normal chow, indicating that the overexpression of GH protected mice from gaining more adipose tissue during a HFD. Retroperitoneal WAT weight was increased by the fat feeding in control mice but was unaffected in the bGH mice. Furthermore, the epidydimal fat depot weight tended to decrease in the bGH mice and to increase in the littermate controls on a HFD. Similarly, the BAT weight increased in the littermate controls and was unchanged in the bGH mice on the HFD. Liver weight was unaffected in either genotype by the HFD (Table 1).
TABLE 1. Body weight, body weight gain, and tissue weights in bGH mice and littermate controls given a ND or a HFD for 8 wk
The food intake was recorded during 5 consecutive days in two groups of animals (three to four mice in each cage) per treatment group (the coefficient of variance, with respect to food intake between the two cages receiving the same treatment, was 15%). There was no difference in food intake or energy intake per gram body weight between the bGH mice and the littermate controls when fed normal chow (Fig. 1B). However, 8 wk of HFD resulted in decreased food intake and unchanged energy intake in the littermate controls. In contrast, the bGH mice did not change their food intake and therefore increased their energy intake on a HFD (Fig. 1, A and B). We calculated food efficiency for the bGH mice and littermate controls according to the formula: body weight gain (grams)/food intake (grams). The bGH mice had a higher food efficiency on a normal chow (0.045) compared with littermate controls (0.016). However, on a HFD, the littermate controls (0.082) and the bGH mice (0.061) had a similar food efficiency.
FIG. 1. Analysis of energy intake in bGH transgenic mice and littermate controls (C) on a ND or a HFD (A and B). Values are presented as mean ± SEM (n = 4). Values with different superscripts denote statistically significant differences between groups (P < 0.05, one-way ANOVA followed by Bonferroni post hoc test).
Energy expenditure, VO2, and RER
Together these results indicate that the bGH transgenic mice prefer a HFD, alternatively that they increase their energy intake as a result of increased EE when receiving a HFD as indicated by the resistance to diet induced obesity. We therefore determined EE, VO2, and RER by indirect calorimetry.
First, the thermoneutral zone was determined, i.e. the temperature range where no energy is used for maintaining regular body temperature, in the bGH mice and littermate controls receiving normal chow. We found no difference in the thermoneutral zone between the bGH mice and littermate controls (Fig. 2A). However, EE was higher in the bGH mice compared with littermate controls (P < 0.05, 2-way ANOVA). After an 8-wk period of HFD, the bGH mice and littermate controls were compared with their respective control groups that had received ND, regarding VO2, EE, and RER (Fig. 2, B–E). At thermoneutrality (30 C), EE was significantly higher in the bGH mice compared with respective controls both on the ND (+14%) and on the HFD (+51%). The HFD increased EE only in the bGH mice (+27% bGH HFD vs. bGH ND; Fig. 2B). VO2 was higher in the bGH mice on a ND (+29%) compared with littermate controls. The difference in VO2 between the bGH mice and littermate controls was further enhanced by the HFD (+68%) (Fig. 2C). RER measurements revealed that both genotypes had the expected diurnal variation of RER on a ND, with a decreased RER during the light phase and an increased RER during the dark phase (Fig. 2D). Interestingly, RER increased in the early afternoon in the bGH mice but at about 1900 h in the littermate controls, indicating a different feeding behavior in the bGH mice. Around noon, the genotypes had a similar RER, but the mean RER was higher in the bGH mice during the light phase (control, 0.81 ± 0.02; bGH, 0.88 ± 0.01, P = 0.005) and during the dark phase (control, 0.89 ± 0.02; bGH, 0.96 ± 0.01, P = 0.02), indicating higher glucose oxidation (Fig. 2D). On a HFD, the diurnal variation in RER was less marked in both genotypes (Fig. 2E). The bGH transgenic mice had a higher RER during the dark phase compared with littermate controls (control, 0.77 ± 0.02; bGH, 0.85 ± 0.02, P = 0.02) but not during the light phase (control, 0.80 ± 0.02; bGH, 0.81 ± 0.02, P = 0.85).
FIG. 2. Determination of the thermoneutral zone (panel A), EE (panel B), VO2 (panel C), and RER (panels D and E) by indirect calorimetry, in bGH transgenic mice and littermate controls (C). A, The thermoneutral zone, defined as the outside temperature range, where no energy is needed for maintaining body temperature, was investigated by the determination of EE at 24, 28, 32, and 36 C in bGH mice and littermate controls receiving a ND. Values are presented as mean ± SEM (n = 4) (P < 0.0005 between groups and P < 0.0001 between temperatures, two-way ANOVA). B, EE analysis of bGH transgenic mice and littermates on a normal diet and on a HFD, by indirect calorimetry, at thermoneutrality (30 C). Values are presented as mean ± SEM (n = 5–8). Panel C, Analysis of VO2 at thermoneutrality (30 C) of bGH transgenic mice and littermates on a normal diet and a HFD by indirect calorimetry. Values are presented as mean ± SEM (n = 5–8). In panels B and C, values with different superscripts denote statistically significant differences between groups (P < 0.05, one-way ANOVA followed by Bonferroni post hoc test). D and E, Analysis of energy substrate preference during 24 h in bGH transgenic mice and littermate controls by analysis of RER using indirect calorimetry. The gray bar indicates the dark hours from 1900–0700 h. Values are presented as mean ± SEM (n = 5–8; P < 0.0001 between groups, both D and E, two-way ANOVA).
Body temperature measurement and uncoupling protein expression
Because the bGH mice showed increased EE, we analyzed body temperature by telemetry transmitters that accurately measure body temperature in freely moving conscious mice. The telemetry analyses of body temperature showed that the bGH mice follow the normal diurnal body temperature variation but have an approximately 0.5 C elevation of body temperature throughout the day when fed normal chow (Fig. 3A). In another experiment, the body temperature was monitored in the same way at three different occasions during a period of 4 wk on a HFD (Fig. 3B). We hypothesized that the HFD would result in an increased body temperature in the bGH mice but not in the control mice. However, we found no significant effect of the HFD on body temperature in the two groups of animals.
FIG. 3. Twenty-four-hour body temperature measurements of unrestrained and freely moving bGH transgenic mice and littermates on normal chow (A) and a HFD (B) by telemetry. Values are presented as mean ± SEM (n = 5–7) (P < 0.0001 between groups, both A and B, two-way ANOVA).
To determine the possible contribution of the uncoupling proteins to the increased EE and body temperature in the bGH transgenic mice, we performed real-time PCR of UCP 1–3 mRNAs in liver, skeletal muscle, WAT, and BAT. We observed no difference in the gene expression of either UCP1, 2, or 3 in BAT or an effect of GH overexpression on muscle UCP3 expression (data not shown). However, we observed a 5-fold higher hepatic UCP2 expression (Table 2) and a 2.8-fold lower expression of UCP2 in WAT (data not shown) in the bGH mice compared with littermate controls on a ND. Hepatic UCP2 mRNA expression was unchanged by the HFD in the bGH mice but increased in littermate controls (Table 2).
TABLE 2. Hepatic gene expression of male bGH mice and littermate controls given a ND or a HFD for 8 wk
Serum lipids and lipoproteins
We have previously investigated the lipoprotein metabolism in bGH transgenic mice on a ND (3). In that study, we found that the bGH mice had decreased serum triglycerides, decreased VLDL cholesterol and VLDL-apoB levels, and decreased hepatic triglyceride secretion compared with littermate controls. Hence, we decided to investigate the effect of a HFD on lipoprotein metabolism more thoroughly in the bGH mice (Figs. 4–6).
FIG. 4. Total serum cholesterol (panel A), triglyceride (panel B), and apoB levels (panel C) in bGH transgenic mice and littermate controls (C) fed either a normal diet or a HFD. Values are presented as mean ± SEM (n = 5–8). Values with different superscripts denote statistically significant differences between groups (P < 0.05, one-way ANOVA followed by Bonferroni post hoc test).
FIG. 5. Lipoprotein size distribution in bGH transgenic mice and littermates fed either normal chow or a HFD. Pooled total lipoprotein fraction [density (d) < 1.215 g/ml], from four to eight mice in each group, was subjected to FPLC as described in Materials and Methods. A, Serum cholesterol; B, triglyceride distribution; C, ApoB distribution as determined by Western blot of the FPLC fractions. SeeBlue prestained standard was used as molecular size marker (NOVEX, San Diego, CA). LDL, Low-density lipoprotein.
FIG. 6. Triglyceride (TG) secretion rate (panel A) and corresponding serum triglyceride levels (panel B) and triglyceride clearance rate (panel C) in bGH transgenic mice and littermate controls (C) on a HFD. A, Hepatic triglyceride secretion rate in vivo was measured by iv injection of WR1339 (500 mg/kg body weight). Serum triglycerides were measured at baseline and 30, 60, and 90 min after the injection of WR1339. The triglyceride secretion rate was calculated from the slope of the curve and expressed as micromoles of triglycerides per hour per kilogram body weight. B, Serum triglyceride levels in the mice before injection of WR1339. Panel C, Triglyceride clearance (milliliters per minute) was calculated from the baseline fasting triglyceride concentration (micromoles per milliliter) and the hepatic triglyceride secretion rate (micromoles per minute). Values are presented as mean ± SEM (n = 5–8). Values with different superscripts denote statistically significant differences between groups (P < 0.05, Student’s t test).
The high-fat feeding resulted in increased serum levels of cholesterol in the controls, but no effect of the diet was observed in the bGH mice (Fig. 4A). Furthermore, fat feeding resulted in opposite effects on serum triglycerides in the bGH mice and littermate control (Fig. 4B). A similar trend was observed regarding serum apoB levels (Fig. 4C). Thus, the fat feeding experiments show that overexpression of GH results in dyslipidemia, including increased serum concentrations of cholesterol, triglycerides, and apoB (Fig. 4).
To determine which lipoprotein fractions are responsible for these effects, we fractionated serum total lipoproteins using size exclusion chromatography and determined the content of cholesterol, triglycerides, and apoB in the fractions (Fig. 5). In contrast to the situation when the mice received normal chow, the bGH mice had increased levels of cholesterol, triglycerides, and apoB in the VLDL/intermediate density lipoprotein fractions compared with littermate controls on a HFD (Fig. 5). Because we previously observed a lower hepatic secretion of triglycerides in the bGH mice compared with littermate controls on a ND (3), we hypothesized that the bGH mice on a HFD would have a higher hepatic triglyceride secretion than the littermate controls. We therefore investigated the hepatic triglyceride secretion rate in another group of bGH mice and littermate controls after 8 wk on a HFD. However, the fat-fed animals of both genotypes had similar hepatic triglyceride secretion rates (Fig. 6A), although a markedly higher serum concentration of triglycerides was observed in the bGH mice (Fig. 6B). The higher serum triglyceride levels in the bGH mice on a HFD were explained by a decreased triglyceride clearance (Fig. 7C). In contrast, when the mice were fed a ND, there was no difference in triglyceride clearance between the bGH mice and controls (controls, 43.3 ± 13.3 ml/min; bGH, 59.0 ± 12.3.ml/min, P = 0.43) (calculations are based on data in Ref.3). Thus, fat feeding resulted in a more pronounced decrease in triglyceride clearance in the bGH mice compared with littermate controls.
FIG. 7. Analysis of glucose tolerance in bGH transgenic mice and littermate controls. Plasma insulin (A and B) and glucose levels (C) were measured immediately before and 1, 5, 10, 20, 30, 50, and 75 min after an iv injection of glucose (1 g/kg) in anesthetized bGH mice and littermate controls on either a normal diet (n = 12 and 16, respectively) or a HFD (n = 12 and 10, respectively). Values are presented as mean ± SEM.
We have previously shown that overexpression of GH results in increased lipoprotein lipase activity in both adipose tissue and heart (3). However, apoCIII mRNA expression was higher in the bGH mice on a HFD [8.68 ± 0.64 (2–CT)] compared with littermate controls [4.75 ± 0.50 (2–CT), P < 0.05], indicating that increased expression of apoCIII could contribute to the decreased clearance of triglycerides in the bGH mice (32).
We found decreased mRNA levels of PPAR, SREBP-1a, SREBP-1c, and FAS in the bGH transgenic mice on a HFD compared with littermate controls (Table 2). These results indicate a further decrease in lipogenesis, as well as ?-oxidation, in the bGH mice in contrast to control mice as a result of a HFD. We therefore investigated the liver content of triglycerides in bGH mice and littermate controls after 8 wk of high-fat feeding. The liver triglyceride content was lower in the fat-fed bGH mice compared with the fat-fed controls (4.0 ± 0.2 μmol/g liver vs. 4.9 ± 0.2 μmol/g liver, P < 0.05).
IVGTT
One of the features of insulin resistance is a decreased turnover of triglyceride-rich lipoproteins and hypertriglyceridemia (33, 34). Elevated serum levels of GH have been shown to result in insulin resistance both in humans and animal models. Also a HFD fed ad libitum leads to obesity-induced insulin resistance. We therefore performed an IVGTT and compared the plasma levels of glucose and insulin in bGH mice and littermate controls receiving either normal chow or a HFD, to separate the contributing components GH and HFD in insulin resistance.
The bGH mice were hyperinsulinemic, but normoglycemic, on a ND (Fig. 7A and Table 3). The hyperinsulinemia was aggravated by the intake of the HFD in both genotypes (Fig. 7B and Table 3). Despite the increased serum insulin levels, the bGH mice became diabetic and had an impaired glucose elimination rate (KG) when fed a HFD, in contrast to the controls (Fig. 7C and Table 3).
TABLE 3. Basal insulin levels, basal glucose levels, and glucose disposal in bGH mice and littermate controls given a ND or a HFD for 8 wk
Discussion
The interaction between GH and HFD on weight gain, EE, and lipid and carbohydrate metabolism has not previously been investigated. We showed that bGH transgenic mice are resistant to diet-induced weight gain and obesity despite hyperphagia on a HFD. Moreover, we showed that the bGH transgenic mice increase their energy intake and EE on a HFD in contrast to control mice, indicating that the dietary fat induces an increased EE that results in increased food intake. Furthermore, we extended our previous findings of altered lipoprotein metabolism in bGH transgenic mice on a ND (3) by showing an opposite effect of GH overexpression on serum levels of triglycerides and apoB when the mice were given a HFD. Thus, the HFD resulted in dyslipidemia only in the bGH mice. Moreover, only the bGH mice became diabetic on a HFD. Together, our results show that the bGH mice on a HFD more accurately reflect the metabolic situation in acromegaly, for which they have long since been a model.
In line with a recent study (5), we observed no effect of GH overexpression on food intake of a ND. However, the bGH mice were clearly hyperphagic when fed a HFD. An appetite-promoting effect of GH has also been seen in a replacement study of children with Turner syndrome and Silver Russell syndrome who suffer from growth retardation (12). However, many food intake studies in rodents have failed to show an effect of GH on food intake (5, 14, 15, 16). It can therefore be speculated that the lack of, or small, effect on food intake observed in different GH models depends on the low percentage of fat in the food. The appetite-promoting effect of GH found in this study further strengthens the already approved treatment of AIDS wasting with GH where only the anabolic effects of GH have previously been acknowledged (36). Based on these experimental studies, the effect of the dietary-fat content on EE should be evaluated when treating AIDS patients with GH.
We observed that the RER increased in the early afternoon in the bGH mice but several hours later in the littermate controls. This finding indicates that the transgenic mice eat more during the light phase, which is in line with the finding of another study (16) showing that GH treatment of Zucker rats made them eat more during the day period without changing the total food intake. Thus, it can be concluded that the effect of GH on food intake is dependent on the content of dietary fat and that GH probably influences the diurnal rhythm of food intake.
In search for possible explanations for the complete resistance to diet-induced obesity of the bGH mice, we investigated EE. We found that the bGH mice had higher EE on a ND, that clearly increased by a HFD. This mimics the human situation where GH substitution to GH-deficient patients (17, 21), as well as acromegaly, increases EE (18, 19). However, it is not known whether the effect of GH on EE in humans is maintained on a low-fat diet. O'Sullivan et al. (18) observed that oral glucose increased EE in controls only, and not in acromegalic patients, indicating that low-fat diet might mitigate the calorigenic effect of GH overexpression.
Several alterations in other hormone levels have been described in these bGH transgenic mice, including increased serum levels of IGF-I (3), corticosterone, and T3 but lower levels of T4 (11) compared with littermate controls. Increased T3 levels may explain the increase in EE. However, in an elegant clinical study comparing the effect of GH and T3 on EE, it was concluded that increased T3 levels as a result of GH treatment cannot solely explain the calorigenic effect of GH (37). Increased corticosterone levels have several effects on intermediary metabolism, including decreased insulin sensitivity. However, it is unlikely that the effects on EE in the bGH mice are mediated by corticosterone, because acromegalic patients and GH treatment result in increased EE without concomitant changes in glucocorticoid levels. Moreover, dexamethasone treatment of lean mice decreases EE (38). Infusion of glucose and insulin to normal adults has been shown to increase EE (39), indicating that the hyperinsulinemia and hyperglycemia observed after fat feeding of the bGH mice may contribute to increased EE. However, the marked insulin resistance of the bGH mice might also counteract the calorigenic effect of hyperinsulinemia and hyperglycemia.
Few studies have addressed the influence of GH on EE or body temperature in rodents (22, 23). In another study of bGH transgenic mice, VO2 tended to be lower in the transgenic mice compared with controls. It was concluded that shivering thermogenesis is reduced but body temperature is unchanged in these animals (22). The reason for the different results is most likely due to differences in the experimental setup. They used restrained mice and measured the body temperature for an hour, whereas we measured the body temperature for 24 h in freely moving mice using telemetry. In line with our results, Hauck et al. (23) observed that GH receptor knock-out mice have reduced body temperature, indicating that marked changes in GH action for a prolonged period influence body temperature in both directions.
The liver and kidney size and, to a lesser extent, the heart and intestine size have been shown to account for more than 50% of the strain variation of EE in laboratory mice (40). The livers of bGH mice account for approximately 9% of the body weight, which is twice that of littermate controls, indicating that the change in liver size might contribute to the increased EE in the bGH mice. We also noted that the brown-fat pad (BAT) size was increased in the bGH mice, which is in contrast to a previous observation (41). In that study (41), they also observed lower expression of UCP-1 in the BAT depot, in contrast to our results. The reason for the discrepant results is unclear but may be due to the difference in age of the mice in the two studies. UCP1 has been implicated in thermogenesis by uncoupling the ATP production in the mitochondrial respiratory chain (42). The closely related UCP2 and UCP3 also function as uncouplers (43), but their metabolic role is less clear. Because UCP1 is exclusively expressed in BAT, its role in adult humans is thought to be minor, because they have minimal amounts of BAT. Therefore, the attention has been focused on UCP2, which is expressed in most tissues, and UCP3, which is mainly expressed in skeletal muscle (44, 45). Different polymorphisms in UCP2 have been linked to EE and substrate specificity for energy production (46, 47). Taken together, these findings indicate that higher expression of UCP2 in the enlarged livers of bGH mice might contribute to the increased EE observed on a ND. However, the contribution of UCP2 to the increase in EE observed in the bGH mice on a HFD, compared with those on normal chow, remains unclear. We have previously shown that hepatic gene expression of enzymes in ?-oxidation is lower in the bGH mice (48). This, together with the present finding of increased RER, indicates that fatty acids in the liver, to a lesser extent, are used for energy production. However, they can activate UCPs (43). We also observed that UCP2 expression increased by the dietary fat in littermate controls, in contrast to the bGH mice, indicating that the bGH mice are resistant to the stimulatory effect of fatty acids on UCP2 expression. UCP2 mRNA levels have previously been shown to be up-regulated by fatty acids and PPAR agonists (49, 50). The down-regulation of PPAR in the liver in the bGH mice, especially in the fat-fed state, might contribute to the unresponsiveness of the bGH mice.
We have previously studied hepatic gene expression in bGH mice on a ND and showed that they have reduced hepatic gene expression of SREBP-1c and lipogenic enzymes (48). Now, we extend these studies by showing a decrease in SREBP-1a mRNA levels in the bGH mice, compared with littermate controls, when fed a HFD. This finding shows a novel regulation of SREBP-1a and that the combined action of chronic overexpression of GH and HFD might reduce the hepatic de novo lipogenesis. Moreover, FAS and SREBP-1c decrease by fat feeding in bGH mice, whereas fat feeding had no effect or even tended to increase the expression of FAS in the control mice. These results shows that overexpression of GH represses the expression of genes involved in de novo lipogenesis.
Fat feeding altered the effect of GH overexpression on lipoprotein metabolism. Instead of decreased triglyceride secretion, unchanged triglyceride clearance, and decreased serum triglyceride levels on a ND (3), the fat feeding resulted in similar triglyceride secretion, markedly decreased triglyceride clearance, and hence increased serum triglyceride levels in the bGH mice compared with littermate controls. The increased serum triglyceride levels were associated with increased apoB levels in the VLDL/intermediate density lipoprotein fraction, showing that an increase in the number of lipoprotein particles in these fractions was responsible for the change. Because GH has been shown to markedly increase lipoprotein lipase activity in both mice and rats (3, 51), it is possible that our findings of increased hepatic apoCIII expression take part in the decreased triglyceride clearance. However, it must be pointed out that other mechanisms may be involved in the changed triglyceride clearance, including VLDL size, apoCI or CII levels, or the activity of lipases other than lipoprotein lipase. We are not aware of any experimental studies on the effect of GH on apoCIII expression or plasma levels, but it has been shown that GH treatment of children results in increased plasma apoCIII levels (35). Together, these findings show that fat feeding results in dyslipidemia of bGH mice that is not observed when these mice are given normal chow.
The RER was increased in the bGH mice, compared with littermate controls, on both diets. Increased glucose oxidation has also been reported in acromegalic patients (18). The reason for this can be speculated upon but is further supported by the decreased hepatic expression of PPAR in bGH mice on a HFD, indicating decreased hepatic ?-oxidation. In line with the finding of increased RER, we have also observed that the hepatic expression of genes involved in fatty acid activation, ?-oxidation, and ketone body formation is decreased in the bGH transgenic mice (48). This was also reflected in the decreased capacity of these mice to produce ketone bodies, both in the fed and fasted states (48). Together, these results may give a clue to the interesting result that young adults, but not old adults, show a positive correlation between RER and measures of GH secretion (20).
We found that bGH mice on a ND were severely insulin resistant, because they had markedly elevated insulin levels in the presence of normal glucose levels. Furthermore, because they had normal glucose elimination after an iv glucose challenge, it is clear that their hyperinsulinemia is adequate, i.e. the islet function seems normal. In contrast, after a HFD, when insulin resistance worsens, the hyperinsulinemia is inadequate, in view of the increased glucose levels and the glucose intolerance. Hence, the limit for islet compensation is reached in these mice. In fact, the insulin levels actually declined after glucose administration. This finding is hard to explain, but it is a clear indication of severe islet dysfunction. The elevated glucose levels were in the range of diabetic animals, and therefore the combination of bGH transgene and high-fat feeding is a challenge, inducing diabetes. The diabetic effect is due to a combined effect on insulin action, resulting in insulin resistance, and on islet function, resulting in inadequate hyperinsulinemia.
In conclusion, bGH transgenic mice were completely resistant to diet-induced obesity despite hyperphagia, possibly due to increased EE. Elevated body temperature and increased hepatic UCP2 expression may, together with increased locomotor activity (11), explain the increase in EE. Furthermore, the bGH transgenic mice became dyslipidemic and diabetic on a HFD. Thus, the bGH transgenic mice more accurately reflect acromegalic patients, with respect to metabolic parameters, when fed a HFD. Based on these findings, it would be interesting to investigate the effect of dietary manipulations on EE, RER, and lipid metabolism in acromegalic patients, anticipating that a change toward a carbohydrate-rich diet will reduce EE and ameliorate the dyslipidemia, as well as the carbohydrate metabolism.
Acknowledgments
We thank Lena Kvist and Mia Umearus for excellent technical assistance.
References
Palmiter RD, Brinster RL, Hammer RE, Trumbauer ME, Rosenfeld MG, Birnberg NC, Evans RM 1982 Dramatic growth of mice that develop from eggs microinjected with metallothionein-growth hormone fusion genes. Nature 300:611–615
Quaife CJ, Mathews LS, Pinkert CA, Hammer RE, Brinster RL, Palmiter RD 1989 Histopathology associated with elevated levels of growth hormone and insulin-like growth factor I in transgenic mice. Endocrinology 124:40–48
Frick F, Bohlooly YM, Lindén D, Olsson B, T?rnell J, Edén S, Oscarsson J 2001 Long-term growth hormone excess induces marked alterations in lipoprotein metabolism in mice. Am J Physiol Endocrinol Metab 281:E1230–E1239
Eisen EJ, Peterson CB, Parker IJ, Murray JD 1998 Effects of zinc ion concentration on growth, fat content and reproduction in oMT1a-oGH transgenic mice. Growth Dev Aging 62:173–186
Berryman DE, List EO, Coschigano KT, Behar K, Kim JK, Kopchick JJ 2004 Comparing adiposity profiles in three mouse models with altered GH signaling. Growth Horm IGF Res 14:309–318
Kopchick JJ, Bellush LL, Coschigano KT 1999 Transgenic models of growth hormone action. Annu Rev Nutr 19:437–461
Bengtsson BA, Brummer RJ, Edén S, Bosaeus I, Lindstedt G 1989 Body composition in acromegaly: the effect of treatment. Clin Endocrinol (Oxf) 31:481–490
Foss MC, Saad MJ, Paccola GM, Paula FJ, Piccinato CE, Moreira AC 1991 Peripheral glucose metabolism in acromegaly. J Clin Endocrinol Metab 72:1048–1053
Nikkil? EA, Pelkonen R 1975 Serum lipids in acromegaly. Metabolism 24:829–838
Oscarsson J, Wiklund O, Jakobsson KE, Petruson B, Bengtsson BA 1994 Serum lipoproteins in acromegaly before and 6–15 months after transsphenoidal adenomectomy. Clin Endocrinol (Oxf) 41:603–608
Bohlooly YM, Olsson B, Gritli-Linde A, Brusehed O, Isaksson OG, Ohlsson C, S?derpalm B, T?rnell J 2001 Enhanced spontaneous locomotor activity in bovine GH transgenic mice involves peripheral mechanisms. Endocrinology 142:4560–4567
Blissett J, Harris G, Kirk J 2000 Effect of growth hormone therapy on feeding problems and food intake in children with growth disorders. Acta Paediatr 89:644–649
Azain MJ, Roberts TJ, Martin RJ, Kasser TR 1995 Comparison of daily versus continuous administration of somatotropin on growth rate, feed intake, and body composition in intact female rats. J Anim Sci 73:1019–1029
Clark RG, Jansson JO, Isaksson O, Robinson IC 1985 Intravenous growth hormone: growth responses to patterned infusions in hypophysectomized rats. J Endocrinol 104:53–61
Roberts TJ, Azain MJ, Hausman GJ, Martin RJ 1994 Interaction of insulin and somatotropin on body weight gain, feed intake, and body composition in rats. Am J Physiol 267:E293–E299
Schulz C, Wieczorek I, Reschke K, Lehnert H 2002 Effects of intracerebroventricularly and intraperitoneally administered growth hormone on body weight and food intake in fa/fa Zucker rats. Neuropsychobiology 45:36–40
Carroll PV, Christ ER, Bengtsson BA, Carlsson L, Christiansen JS, Clemmons D, Hintz R, Ho K, Laron Z, Sizonenko P, Sonksen PH, Tanaka T, Thorne M 1998 Growth hormone deficiency in adulthood and the effects of growth hormone replacement: a review. Growth Hormone Research Society Scientific Committee. J Clin Endocrinol Metab 83:382–395
O’Sullivan AJ, Kelly JJ, Hoffman DM, Baxter RC, Ho KK 1995 Energy metabolism and substrate oxidation in acromegaly. J Clin Endocrinol Metab 80:486–491
M?ller N, Schmitz O, J?rgensen JO, Astrup J, Bak JF, Christensen SE, Alberti KG, Weeke J 1992 Basal- and insulin-stimulated substrate metabolism in patients with active acromegaly before and after adenomectomy. J Clin Endocrinol Metab 74:1012–1019
Christiansen JJ, Vahl N, Fisker S, M?ller N, Christiansen JS, J?rgensen JO 2002 Age dimorphism in the association between growth-hormone status and the respiratory quotient. Obes Res 10:284–290
Gerver WJ, De Bruin R, Delemarre v d Waal HA, Aldewereld B, Theunissen P, Westerterp KR 2000 Effects of discontinuation of growth hormone treatment on body composition and metabolism. Horm Res 53:215–220
Moura AS, Spiers DE, Lamberson WR 1998 Thermogenic activity of growth hormone transgenic mice. Growth Dev Aging 62:149–159
Hauck SJ, Hunter WS, Danilovich N, Kopchick JJ, Bartke A 2001 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 226:552–558
Yokode M, Hammer RE, Ishibashi S, Brown MS, Goldstein JL 1990 Diet-induced hypercholesterolemia in mice: prevention by overexpression of LDL receptors. Science 250:1273–1275
Masucci-Magoulas L, Goldberg IJ, Bisgaier CL, Serajuddin H, Francone OL, Breslow JL, Tall AR 1997 A mouse model with features of familial combined hyperlipidemia. Science 275:391–394
Surwit RS, Kuhn CM, Cochrane C, McCubbin JA, Feinglos MN 1988 Diet-induced type II diabetes in C57BL/6J mice. Diabetes 37:1163–1167
Ahrén B, Scheurink AJ 1998 Marked hyperleptinemia after high-fat diet associated with severe glucose intolerance in mice. Eur J Endocrinol 139:461–467
Sandstedt J, Ohlsson C, Norjavaara E, Nilsson J, T?rnell J 1994 Disproportional bone growth and reduced weight gain in gonadectomized male bovine growth hormone transgenic and normal mice. Endocrinology 135:2574–2580
Oscarsson J, Olofsson SO, Bondjers G, Edén S 1989 Differential effects of continuous versus intermittent administration of growth hormone to hypophysectomized female rats on serum lipoproteins and their apoproteins. Endocrinology 125:1638–1649
Bruss ML, Black AL 1978 Enzymatic microdetermination of glycogen. Anal Biochem 84:309–312
Li X, Catalina F, Grundy SM, Patel S 1996 Method to measure apolipoprotein B-48 and B-100 secretion rates in an individual mouse: evidence for a very rapid turnover of VLDL and preferential removal of B-48-relative to B-100-containing lipoproteins. J Lipid Res 37:210–220
Ebara T, Ramakrishnan R, Steiner G, Shachter NS 1997 Chylomicronemia due to apolipoprotein CIII overexpression in apolipoprotein E-null mice. Apolipoprotein CIII-induced hypertriglyceridemia is not mediated by effects on apolipoprotein E. J Clin Invest 99:2672–2681
Cummings MH, Watts GF, Umpleby AM, Hennessy TR, Naoumova R, Slavin BM, Thompson GR, S?nksen PH 1995 Increased hepatic secretion of very-low-density lipoprotein apolipoprotein B-100 in NIDDM. Diabetologia 38:959–967
Duvillard L, Pont F, Florentin E, Galland-Jos C, Gambert P, Verges B 2000 Metabolic abnormalities of apolipoprotein B-containing lipoproteins in non-insulin-dependent diabetes: a stable isotope kinetic study. Eur J Clin Invest 30:685–694
Schaefer GB, Greger NG, Fesmire JD, Blackett PR, Wilson DP, Frindik JP 1994 Lipids and apolipoproteins in growth hormone-deficient children during treatment. Metabolism 43:1457–1461
Grinspoon S, Mulligan K 2003 Weight loss and wasting in patients infected with human immunodeficiency virus. Clin Infect Dis 36:S69–S78
Wolthers T, Groftne T, M?ller N, Christiansen JS, ?rskov H, Weeke J, J?rgensen JO 1996 Calorigenic effects of growth hormone: the role of thyroid hormones. J Clin Endocrinol Metab 81:1416–1419
Chen HL, Romsos DR 1995 A single intracerebroventricular injection of dexamethasone elevates food intake and plasma insulin and depresses metabolic rates in adrenalectomized obese (ob/ob) mice. J Nutr 125:540–545
Ravussin E, Bogardus C, Schwartz RS, Robbins DC, Wolfe RR, Horton ES, Danforth Jr E, Sims EA 1983 Thermic effect of infused glucose and insulin in man. Decreased response with increased insulin resistance in obesity and noninsulin-dependent diabetes mellitus. J Clin Invest 72:893–902
Konarzewski M, Diamond J 1995 Evolution of basal metabolic rate and organ masses in laboratory mice. Evolution 49:1239–1248
Li Y, Knapp JR, Kopchick JJ 2003 Enlargement of interscapular brown adipose tissue in growth hormone antagonist transgenic and in growth hormone receptor gene-disrupted dwarf mice. Exp Biol Med Maywood 228:207–215
Enerb?ck S, Jacobsson A, Simpson EM, Guerra C, Yamashita H, Harper ME, Kozak LP 1997 Mice lacking mitochondrial uncoupling protein are cold-sensitive but not obese. Nature 387:90–94
Jaburek M, Varecha M, Gimeno RE, Dembski M, Jezek P, Zhang M, Burn P, Tartaglia LA, Garlid KD 1999 Transport function and regulation of mitochondrial uncoupling proteins 2 and 3. J Biol Chem 274:26003–26007
Vidal-Puig A, Solanes G, Grujic D, Flier JS, Lowell BB 1997 UCP3: an uncoupling protein homologue expressed preferentially and abundantly in skeletal muscle and brown adipose tissue. Biochem Biophys Res Commun 235:79–82
Fleury C, Neverova M, Collins S, Raimbault S, Champigny O, Levi-Meyrueis C, Bouillaud F, Seldin MF, Surwit RS, Ricquier D, Warden CH 1997 Uncoupling protein-2: a novel gene linked to obesity and hyperinsulinemia. Nat Genet 15:269–272
Esterbauer H, Schneitler C, Oberkofler H, Ebenbichler C, Paulweber B, Sandhofer F, Ladurner G, Hell E, Strosberg AD, Patsch JR, Krempler F, Patsch W 2001 A common polymorphism in the promoter of UCP2 is associated with decreased risk of obesity in middle-aged humans. Nat Genet 28:178–183
Le Fur S, Le Stunff C, Dos Santos C, Bougneres P 2004 The common –866 G/A polymorphism in the promoter of uncoupling protein 2 is associated with increased carbohydrate and decreased lipid oxidation in juvenile obesity. Diabetes 53:235–239
Olsson B, Bohlooly YM, Brusehed O, Isaksson OG, Ahrén B, Olofsson SO, Oscarsson J, T?rnell J 2003 Bovine growth hormone-transgenic mice have major alterations in hepatic expression of metabolic genes. Am J Physiol Endocrinol Metab 285:E504–E511
Cortez-Pinto H, Zhi Lin H, Qi Yang S, Odwin Da Costa S, Diehl AM 1999 Lipids up-regulate uncoupling protein 2 expression in rat hepatocytes. Gastroenterology 116:1184–1193
Tsuboyama-Kasaoka N, Takahashi M, Kim H, Ezaki O 1999 Up-regulation of liver uncoupling protein-2 mRNA by either fish oil feeding or fibrate administration in mice. Biochem Biophys Res Commun 257:879–885
Oscarsson J, Ottosson M, Vikman-Adolfsson K, Frick F, Enerb?ck S, Lithell H, Edén S 1999 GH but not IGF-I or insulin increases lipoprotein lipase activity in muscle tissues of hypophysectomised rats. J Endocrinol 160:247–255(Bob Olsson, Mohammad Bohl)
Address all correspondence and requests for reprints to: Bob Olsson, Department of Internal Medicine, Sahlgrenska University Hospital, G?teborg University, Vita Str?ket 12, S-413 45 G?teborg, Sweden. E-mail: bob.olsson@medic.gu.se.
Abstract
It is known that bovine GH (bGH) transgenic mice have increased body mass, insulin resistance, and altered lipoprotein metabolism when fed a normal diet (ND). In this study, the effects of 8 wk of high-fat diet (HFD) were investigated in 6-month-old male bGH mice. Although littermate controls had unchanged energy intake, energy intake was higher in the bGH mice on a HFD than on a low-fat diet. Nevertheless, the bGH mice were resistant to diet-induced weight gain, and only in the bGH mice did the HFD result in increased energy expenditure. Glucose oxidation was higher in the bGH mice compared with littermate controls on both a HFD and ND. In addition, the bGH mice had 0.5 C higher body temperature throughout the day and increased hepatic uncoupling protein 2 expression; changes that were unaffected by the HFD. On a HFD, the effect of bGH overexpression on serum triglycerides and apolipoprotein B was opposite to that on a ND, resulting in higher serum concentrations of triglycerides and apolipoprotein B compared with littermate controls. Increased serum triglycerides were explained by decreased triglyceride clearance. The HFD led to diabetes only in the bGH mice. In conclusion, bGH transgenic mice were resistant to diet-induced obesity despite hyperphagia, possibly due to increased energy expenditure. On a HFD, bGH mice became dyslipidemic and diabetic and thereby more accurately reflect the metabolic situation in acromegalic patients.
Introduction
GH TRANSGENIC MICE have been extensively used to study long-term effects of GH overproduction. GH transgenic mice have increased body size (1), organomegaly (2), and reduced body fat mass (3, 4, 5). The correlation between GH and body fat is illustrated in ovine GH gene transgenic mice controlled by the metallothionine promoter, where zinc is used to regulate GH dosage. In that model, epidydimal and sc fat mass were found to decrease linearly with increased GH levels (4). GH transgenic mice are normoglycemic but highly hyperinsulinemic and show slightly impaired or normal glucose tolerance when fed normal chow (6). The bovine GH (bGH) transgenic mice also show several alterations in lipid and lipoprotein metabolism. These alterations include lower very-low-density lipoprotein (VLDL) levels, triglyceride levels, and hepatic triglyceride secretion rate and increased lipoprotein lipase activity (3). Furthermore, bGH mice also have higher high-density lipoprotein (HDL) and total cholesterol levels (2, 3).
Patients with acromegaly and GH transgenic mice share similar traits resulting from the elevated serum levels of GH, e.g. organomegaly, decreased body fat content (7), insulin resistance (8), and disturbed lipoprotein metabolism (9, 10). However, in contrast to acromegalic patients, who have elevated serum triglyceride levels and, in some individuals, elevated triglyceride secretion (9, 10), bGH mice have lower triglyceride levels (3). Moreover, acromegalic patients develop hyperglycemia, which does not occur in GH transgenic mice. The reason for these discrepancies may be due to differences between the species or due to the differences in terms of changes in other hormonal axis, e.g. unchanged or substituted cortisol deficiency in acromegaly and increased corticosterone levels in GH transgenic mice (11). However, the differences may also be attributable to the larger intake of fat in humans than that contained in the ordinary mouse diet.
Other aspects of the regulation of metabolism that may be influenced by GH include food intake, energy expenditure (EE), and metabolic fuel preference, measured as respiratory exchange ratio (RER). The effect of GH on food intake has been studied both in man and experimental animals. GH has been shown to increase food intake in children (12) and in experimental animals (13) but not in all studies (14, 15, 16) including bGH transgenic mice given ordinary low-fat chow (5). It has also been shown that eating behavior changes after GH administration (16). However, no studies have been performed to investigate the effect of GH on intake of low- vs. high-fat diet (HFD). It is generally accepted that GH increases lipolysis and fatty acid oxidation (17). However, the effect of GH overproduction, as in the case of acromegaly, on RER is less clear (18, 19), as is the relation between estimates of GH release and RER in young and old adults (20).
It is a general finding in clinical GH studies that GH treatment (21), as well as acromegaly, results in increased EE (18, 19). However, few studies have addressed the influence of GH on EE (22) or body temperature in rodents (23). In a previous paper (11), we observed that bGH transgenic mice have an increased locomotor activity, indicating that they have an increased EE, but basal metabolic rate has not previously been investigated in these mice.
The consumption of a HFD fed ad libitum changes body composition and metabolic status, resulting in diet-induced obesity and altered levels of cholesterol, triglycerides, and insulin (24, 25, 26). However, important genetic differences exist, as illustrated by the observations that different mouse strains have different susceptibility to diet-induced obesity (26, 27). Moreover, when findings in animal models are extrapolated to the human situation, HFD feeding better mimics the human diet in the Western world, where 30–40% of the dietary energy comes from fat, compared with less than 10% in normal mouse chow.
The effect of GH on body composition, food intake, and EE, as well as lipid- and carbohydrate metabolism, during the influence of a HFD has not previously been studied. In this study, the effect of 8 wk of HFD was investigated in 6-month-old male transgenic bGH mice. We conclude that bGH transgenic mice are resistant to diet-induced obesity despite hyperphagia, possibly due to increased resting EE. The bGH transgenic mice became dyslipidemic and diabetic on a HFD and thereby more accurately reflect the metabolic situation in acromegaly compared with those fed a low-fat diet.
Materials and Methods
Animals
In this study, 5- to 6-month-old bGH transgenic mice and littermate controls were used. The bGH mice have previously been described by Sandstedt et al. (28). The mice were housed with a 12-h light, 12-h dark cycle (0700–1900 h, with a 1-h dawn/sunset function) and had free access to tap water and mouse standard chow (R-34, Lactamin, Vadstena, Sweden) or HFD (R-638, AnalyCen Nordic AB, Lidk?ping, Sweden). The standard chow contained (in energy percent) 9.4% fat, 20.2% protein, 0.8% fiber, and 69.6% nonfat energy; and the HFD contained 39.9% fat, 17% protein, 0.7% fiber, and 42.3% nonfat energy. However, no cholesterol was added to the HFD. In all of the HFD experiments, the mice received a HFD for 8 wk before the analysis, except in one experiment of body temperature measurements where the mice only received a HFD for 4 wk. Six bGH male mice and eight littermate controls on either normal chow or a HFD were housed three per cage in two cages (bGH mice) or four per cage (controls) in two cages during recording of food intake. Every morning, for 5 consecutive days, the food in each cage was weighed.
The mice were anesthetized using medetomidine (Domitor, Orion Espoo, Finland; 0.5 mg/kg, ip) and ketamine (Ketalar, Parke-Davis, Detroit, MI; 75 mg/kg, ip) and killed by heart puncture. The liver, white adipose tissue (WAT), brown adipose tissue (BAT), and soleus muscle were excised and immediately frozen in liquid nitrogen and stored at –135 C. The study was performed after prior approval from the local ethics committee for animal experimentation at the G?teborg University, Sweden.
Real-time PCR
Liver, skeletal muscle, BAT, and WAT from five bGH males and littermate controls on normal diet (ND) were homogenized. In addition, livers from five bGH mice and littermate controls on a HFD for 8 wk were homogenized. Total RNA was then extracted by Tri Reagent (Sigma Diagnostics, St. Louis, MO). First-strand cDNA was synthesized from total RNA using Superscript preamplification system (Life Technologies, Rockville, MD). Real-time PCR analysis was performed with an ABI Prism 7900 Sequence Detection System (Applied Biosystems, Foster City, CA) using FAM and TAMRA- or VIC and TAMRA-labeled fluorogenic probes. The expression of uncoupling protein (UCP) 1, UCP2, UCP3, peroxisome proliferator-activated receptor (PPAR), sterol regulatory element-binding protein (SREBP)-1, apolipoprotein (apo)CIII, and fatty acid synthase (FAS) were normalized against mouse acidic ribosomal phosphoprotein P0 (M36B4). The relative expression levels were calculated according to the formula: 2–CT, where CT is the difference in cycle threshold values between the target and the M36B4 internal control (User Bulletin no. 2, PerkinElmer, Foster City, CA).
Surgical implantation of telemetry transmitters and signal acquisition
Radiotelemetry transmitters, designed to detect core body temperature, were implanted in bGH mice and littermate controls. The mice were anesthetized using medetomidine (Domitor, 0.5 mg/kg, ip) and ketamine (75 mg/kg, ip). A telemetry transmitter (TA11ETA-F20, weight 3 g; Data Sciences International, Inc., St. Paul, MN) was then implanted into the abdominal cavity through a midline incision. The abdominal incision was closed with staples. Anesthesia was reversed by atipamezole (Antisedan, Orion Espoo, Finland, 4 mg/kg, ip). Mice were allowed 5–7 d to recover from surgery. The cage with the animal was placed on a receiver plate, and the signal was collected using the Dataquest LabPRO Acquisition System (version 3.01, Data Sciences International, Inc.). The temperature signal was sampled at a frequency of 500 Hz, in 6-sec bursts, every 2 min for 48 h. Body temperature was then recalculated as 2-h averages and displayed as a 24-h curve.
Indirect calorimetric analyses
To determine the thermoneutral zone of bGH mice and littermate controls, the resting metabolic rate was measured at 24 C, 28 C, 32 C, and finally 36 C, for 2 h at each temperature, by indirect calorimetric analysis. A 1-h period was allowed for acclimatization, after which the VO2 and EE were determined at the lowest level of O2 consumption (VO2) during the final hour of the measuring period and were expressed as milliliters of O2 x hours–1 x (kilograms of body weight)–0.75. EE (kcal/h·kg) was calculated using a rearrangement of the Weir equation as supplied by Columbus Instruments: (3.815 + 1.232 x RER) x VO2.
VO2 and CO2 production were measured using an indirect open circuit calorimeter with a flow rate of 500 ml/min and a chamber vol of 2.7-liters (Oxymax, Columbus Instruments, Columbus, OH; cage dimensions, 20 x 10 x 12.7 cm). The system was calibrated daily using a standard gas mixture (0.49% CO2-20.5% O2-79.0% N2). Measurements were done pair-wise, and expired air was analyzed for a 30-sec period every 6 min using an electrochemical O2 analyzer and a CO2 sensor (Oxymax). Mice were conscious and unrestrained during the measuring period but did not have access to food or water during the VO2 and EE measurements.
After 8 wk on the ND or HFD, VO2 and EE of individual animals were measured at thermoneutrality (30 C). One hour of acclimatization was allowed, after which VO2 and EE were determined at the lowest level of VO2 during the final 2 h of the measuring period and were expressed as milliliters of O2x hours–1 x (kilograms of body weight)–0.75.
To determine the RER, the mice had free access to water and either normal chow or HFD ad libitum. The RER was calculated as VCO2/VO2 [volume of CO2 produced per volume of O2 consumed (milliliters per kilogram per minute)].
IVGTT (iv glucose tolerance test)
IVGTT was performed in 6-month-old bGH mice and littermate controls fed a ND or HFD for 8 wk. The studies were performed in the late morning hours after 4 h withdrawal of food from the cages. The animals were anesthetized with midazolam (Dormicum, Hoffman-La-Roche, Basel, Switzerland; 0.4 mg/mouse, ip) and a combination of fluanison (0.9 mg/mouse) and fentanyl (0.02 mg/mouse; Hypnorm, Janssen, Beerse, Belgium). Thereafter, a blood sample was taken from the retrobulbar, intraorbital, capillary plexus, after which the animals were given an iv injection of D-glucose (1 g/kg; British Drug Houses, Poole, UK). The vol load was 10 μl/g body weight. Blood samples were then taken at 1, 5, 10, 20, 30, and 50 min (75 μl) after the glucose load. The samples for glucose and insulin were taken in heparinized tubes; and after immediate centrifugation, plasma was separated and stored at –80 C until analysis. From the IVGTT, the glucose elimination rate was calculated as the glucose elimination constant (KG), i.e. the elimination of glucose in percent per minute from min 1 to min 20 after iv glucose administration, as calculated after logarithmic transformation of data.
Serum and plasma analyses
Triglyceride and cholesterol concentrations were determined by enzymatic colorimetric assays (MPR2: TG/GPO-PAP and Chl/CHOD-PAP, Roche Diagnostics, Mannheim, Germany). Serum apoB concentrations were determined by an electroimmunoassay as previously described (3, 29).
Plasma insulin was determined using a RIA with the use of a guinea pig antirat insulin antibody, 125I-labeled human insulin as tracer, and rat insulin as standard (Linco Research, St. Charles, MO). Plasma glucose was determined with the glucose oxidase method (30).
Size distribution of serum lipoproteins
Determination of size distribution of lipoproteins was performed by gel filtration using fast protein liquid chromatography (FPLC) equipment (Pharmacia Upjohn, Uppsala, Sweden) as described previously (3). Briefly, serum from six mice was pooled to a total vol of 1.5 ml and the density adjusted to 1.215 g/ml with KBr in 0.9% NaCl. After ultracentrifugation (35,000 x g, 4 C, 24 h), the total lipoprotein fraction was recovered by aspiration and the final vol adjusted to 2 ml with FPLC-buffer (0.15 M NaCl, 0.01% EDTA, 0.02% NaAz, pH 7.3). After filtration through a 0.45-μm low-protein filter, the sample was loaded on a 25-ml Superose 6B column (Pharmacia Upjohn) using a constant flow rate of 0.35 ml/min. Eluted samples were collected in 0.5-ml fractions. The fractions were stored at –20 C until assay. Triglycerides and cholesterol concentrations were determined with enzymatic colorimetric assays as described above. Western blots for determination of apoB in the fractions were performed as described previously (3).
In vivo hepatic triglyceride secretion rate
Triglyceride secretion rate in vivo was measured after iv administration of Triton WR1339 as described before (3, 31). After a 5-h period without access to mouse standard chow (0700–1200 h), anesthetized bGH mice and littermate controls were injected iv with Triton WR1339 (Sigma Diagnostics) diluted in saline (200 mg/ml) via the jugular vein (500 mg/kg body weight). Blood samples (70 μl) were taken before and 30, 60, and 90 min after the Triton WR1339 injection. The triglyceride accumulation was linear during this time period. The triglyceride concentration was analyzed as described above. Triglyceride clearance (milliliters per minute) was calculated from the baseline fasting triglyceride concentration (micromoles per milliliter) and the hepatic triglyceride secretion rate (micromoles per minute).
Hepatic lipid content
Hepatic triglyceride content was determined as described before (3). In brief, the livers were homogenized, and lipids were extracted. After evaporation of the final chloroform phase, lipids were dissolved in isopropanol, and triglyceride concentrations were determined as described above.
Statistics
Values are given as means ± SEM. Comparison between groups was performed with unpaired Student’s t test or two-way or one-way ANOVA followed by Bonferroni’s test. P values less than 0.05 were considered significant. When appropriate, values were normalized by logarithmic transformation.
Results
Food intake and body composition
Male bGH mice and littermate controls of 6 months of age were given a HFD for 8 wk and compared with corresponding mice receiving a ND. bGH mice fed a HFD had a similar final weight and weight gain as those receiving normal chow (Table 1). However, littermate controls receiving the HFD had an increased body weight gain and final body weight compared with those receiving normal chow, indicating that the overexpression of GH protected mice from gaining more adipose tissue during a HFD. Retroperitoneal WAT weight was increased by the fat feeding in control mice but was unaffected in the bGH mice. Furthermore, the epidydimal fat depot weight tended to decrease in the bGH mice and to increase in the littermate controls on a HFD. Similarly, the BAT weight increased in the littermate controls and was unchanged in the bGH mice on the HFD. Liver weight was unaffected in either genotype by the HFD (Table 1).
TABLE 1. Body weight, body weight gain, and tissue weights in bGH mice and littermate controls given a ND or a HFD for 8 wk
The food intake was recorded during 5 consecutive days in two groups of animals (three to four mice in each cage) per treatment group (the coefficient of variance, with respect to food intake between the two cages receiving the same treatment, was 15%). There was no difference in food intake or energy intake per gram body weight between the bGH mice and the littermate controls when fed normal chow (Fig. 1B). However, 8 wk of HFD resulted in decreased food intake and unchanged energy intake in the littermate controls. In contrast, the bGH mice did not change their food intake and therefore increased their energy intake on a HFD (Fig. 1, A and B). We calculated food efficiency for the bGH mice and littermate controls according to the formula: body weight gain (grams)/food intake (grams). The bGH mice had a higher food efficiency on a normal chow (0.045) compared with littermate controls (0.016). However, on a HFD, the littermate controls (0.082) and the bGH mice (0.061) had a similar food efficiency.
FIG. 1. Analysis of energy intake in bGH transgenic mice and littermate controls (C) on a ND or a HFD (A and B). Values are presented as mean ± SEM (n = 4). Values with different superscripts denote statistically significant differences between groups (P < 0.05, one-way ANOVA followed by Bonferroni post hoc test).
Energy expenditure, VO2, and RER
Together these results indicate that the bGH transgenic mice prefer a HFD, alternatively that they increase their energy intake as a result of increased EE when receiving a HFD as indicated by the resistance to diet induced obesity. We therefore determined EE, VO2, and RER by indirect calorimetry.
First, the thermoneutral zone was determined, i.e. the temperature range where no energy is used for maintaining regular body temperature, in the bGH mice and littermate controls receiving normal chow. We found no difference in the thermoneutral zone between the bGH mice and littermate controls (Fig. 2A). However, EE was higher in the bGH mice compared with littermate controls (P < 0.05, 2-way ANOVA). After an 8-wk period of HFD, the bGH mice and littermate controls were compared with their respective control groups that had received ND, regarding VO2, EE, and RER (Fig. 2, B–E). At thermoneutrality (30 C), EE was significantly higher in the bGH mice compared with respective controls both on the ND (+14%) and on the HFD (+51%). The HFD increased EE only in the bGH mice (+27% bGH HFD vs. bGH ND; Fig. 2B). VO2 was higher in the bGH mice on a ND (+29%) compared with littermate controls. The difference in VO2 between the bGH mice and littermate controls was further enhanced by the HFD (+68%) (Fig. 2C). RER measurements revealed that both genotypes had the expected diurnal variation of RER on a ND, with a decreased RER during the light phase and an increased RER during the dark phase (Fig. 2D). Interestingly, RER increased in the early afternoon in the bGH mice but at about 1900 h in the littermate controls, indicating a different feeding behavior in the bGH mice. Around noon, the genotypes had a similar RER, but the mean RER was higher in the bGH mice during the light phase (control, 0.81 ± 0.02; bGH, 0.88 ± 0.01, P = 0.005) and during the dark phase (control, 0.89 ± 0.02; bGH, 0.96 ± 0.01, P = 0.02), indicating higher glucose oxidation (Fig. 2D). On a HFD, the diurnal variation in RER was less marked in both genotypes (Fig. 2E). The bGH transgenic mice had a higher RER during the dark phase compared with littermate controls (control, 0.77 ± 0.02; bGH, 0.85 ± 0.02, P = 0.02) but not during the light phase (control, 0.80 ± 0.02; bGH, 0.81 ± 0.02, P = 0.85).
FIG. 2. Determination of the thermoneutral zone (panel A), EE (panel B), VO2 (panel C), and RER (panels D and E) by indirect calorimetry, in bGH transgenic mice and littermate controls (C). A, The thermoneutral zone, defined as the outside temperature range, where no energy is needed for maintaining body temperature, was investigated by the determination of EE at 24, 28, 32, and 36 C in bGH mice and littermate controls receiving a ND. Values are presented as mean ± SEM (n = 4) (P < 0.0005 between groups and P < 0.0001 between temperatures, two-way ANOVA). B, EE analysis of bGH transgenic mice and littermates on a normal diet and on a HFD, by indirect calorimetry, at thermoneutrality (30 C). Values are presented as mean ± SEM (n = 5–8). Panel C, Analysis of VO2 at thermoneutrality (30 C) of bGH transgenic mice and littermates on a normal diet and a HFD by indirect calorimetry. Values are presented as mean ± SEM (n = 5–8). In panels B and C, values with different superscripts denote statistically significant differences between groups (P < 0.05, one-way ANOVA followed by Bonferroni post hoc test). D and E, Analysis of energy substrate preference during 24 h in bGH transgenic mice and littermate controls by analysis of RER using indirect calorimetry. The gray bar indicates the dark hours from 1900–0700 h. Values are presented as mean ± SEM (n = 5–8; P < 0.0001 between groups, both D and E, two-way ANOVA).
Body temperature measurement and uncoupling protein expression
Because the bGH mice showed increased EE, we analyzed body temperature by telemetry transmitters that accurately measure body temperature in freely moving conscious mice. The telemetry analyses of body temperature showed that the bGH mice follow the normal diurnal body temperature variation but have an approximately 0.5 C elevation of body temperature throughout the day when fed normal chow (Fig. 3A). In another experiment, the body temperature was monitored in the same way at three different occasions during a period of 4 wk on a HFD (Fig. 3B). We hypothesized that the HFD would result in an increased body temperature in the bGH mice but not in the control mice. However, we found no significant effect of the HFD on body temperature in the two groups of animals.
FIG. 3. Twenty-four-hour body temperature measurements of unrestrained and freely moving bGH transgenic mice and littermates on normal chow (A) and a HFD (B) by telemetry. Values are presented as mean ± SEM (n = 5–7) (P < 0.0001 between groups, both A and B, two-way ANOVA).
To determine the possible contribution of the uncoupling proteins to the increased EE and body temperature in the bGH transgenic mice, we performed real-time PCR of UCP 1–3 mRNAs in liver, skeletal muscle, WAT, and BAT. We observed no difference in the gene expression of either UCP1, 2, or 3 in BAT or an effect of GH overexpression on muscle UCP3 expression (data not shown). However, we observed a 5-fold higher hepatic UCP2 expression (Table 2) and a 2.8-fold lower expression of UCP2 in WAT (data not shown) in the bGH mice compared with littermate controls on a ND. Hepatic UCP2 mRNA expression was unchanged by the HFD in the bGH mice but increased in littermate controls (Table 2).
TABLE 2. Hepatic gene expression of male bGH mice and littermate controls given a ND or a HFD for 8 wk
Serum lipids and lipoproteins
We have previously investigated the lipoprotein metabolism in bGH transgenic mice on a ND (3). In that study, we found that the bGH mice had decreased serum triglycerides, decreased VLDL cholesterol and VLDL-apoB levels, and decreased hepatic triglyceride secretion compared with littermate controls. Hence, we decided to investigate the effect of a HFD on lipoprotein metabolism more thoroughly in the bGH mice (Figs. 4–6).
FIG. 4. Total serum cholesterol (panel A), triglyceride (panel B), and apoB levels (panel C) in bGH transgenic mice and littermate controls (C) fed either a normal diet or a HFD. Values are presented as mean ± SEM (n = 5–8). Values with different superscripts denote statistically significant differences between groups (P < 0.05, one-way ANOVA followed by Bonferroni post hoc test).
FIG. 5. Lipoprotein size distribution in bGH transgenic mice and littermates fed either normal chow or a HFD. Pooled total lipoprotein fraction [density (d) < 1.215 g/ml], from four to eight mice in each group, was subjected to FPLC as described in Materials and Methods. A, Serum cholesterol; B, triglyceride distribution; C, ApoB distribution as determined by Western blot of the FPLC fractions. SeeBlue prestained standard was used as molecular size marker (NOVEX, San Diego, CA). LDL, Low-density lipoprotein.
FIG. 6. Triglyceride (TG) secretion rate (panel A) and corresponding serum triglyceride levels (panel B) and triglyceride clearance rate (panel C) in bGH transgenic mice and littermate controls (C) on a HFD. A, Hepatic triglyceride secretion rate in vivo was measured by iv injection of WR1339 (500 mg/kg body weight). Serum triglycerides were measured at baseline and 30, 60, and 90 min after the injection of WR1339. The triglyceride secretion rate was calculated from the slope of the curve and expressed as micromoles of triglycerides per hour per kilogram body weight. B, Serum triglyceride levels in the mice before injection of WR1339. Panel C, Triglyceride clearance (milliliters per minute) was calculated from the baseline fasting triglyceride concentration (micromoles per milliliter) and the hepatic triglyceride secretion rate (micromoles per minute). Values are presented as mean ± SEM (n = 5–8). Values with different superscripts denote statistically significant differences between groups (P < 0.05, Student’s t test).
The high-fat feeding resulted in increased serum levels of cholesterol in the controls, but no effect of the diet was observed in the bGH mice (Fig. 4A). Furthermore, fat feeding resulted in opposite effects on serum triglycerides in the bGH mice and littermate control (Fig. 4B). A similar trend was observed regarding serum apoB levels (Fig. 4C). Thus, the fat feeding experiments show that overexpression of GH results in dyslipidemia, including increased serum concentrations of cholesterol, triglycerides, and apoB (Fig. 4).
To determine which lipoprotein fractions are responsible for these effects, we fractionated serum total lipoproteins using size exclusion chromatography and determined the content of cholesterol, triglycerides, and apoB in the fractions (Fig. 5). In contrast to the situation when the mice received normal chow, the bGH mice had increased levels of cholesterol, triglycerides, and apoB in the VLDL/intermediate density lipoprotein fractions compared with littermate controls on a HFD (Fig. 5). Because we previously observed a lower hepatic secretion of triglycerides in the bGH mice compared with littermate controls on a ND (3), we hypothesized that the bGH mice on a HFD would have a higher hepatic triglyceride secretion than the littermate controls. We therefore investigated the hepatic triglyceride secretion rate in another group of bGH mice and littermate controls after 8 wk on a HFD. However, the fat-fed animals of both genotypes had similar hepatic triglyceride secretion rates (Fig. 6A), although a markedly higher serum concentration of triglycerides was observed in the bGH mice (Fig. 6B). The higher serum triglyceride levels in the bGH mice on a HFD were explained by a decreased triglyceride clearance (Fig. 7C). In contrast, when the mice were fed a ND, there was no difference in triglyceride clearance between the bGH mice and controls (controls, 43.3 ± 13.3 ml/min; bGH, 59.0 ± 12.3.ml/min, P = 0.43) (calculations are based on data in Ref.3). Thus, fat feeding resulted in a more pronounced decrease in triglyceride clearance in the bGH mice compared with littermate controls.
FIG. 7. Analysis of glucose tolerance in bGH transgenic mice and littermate controls. Plasma insulin (A and B) and glucose levels (C) were measured immediately before and 1, 5, 10, 20, 30, 50, and 75 min after an iv injection of glucose (1 g/kg) in anesthetized bGH mice and littermate controls on either a normal diet (n = 12 and 16, respectively) or a HFD (n = 12 and 10, respectively). Values are presented as mean ± SEM.
We have previously shown that overexpression of GH results in increased lipoprotein lipase activity in both adipose tissue and heart (3). However, apoCIII mRNA expression was higher in the bGH mice on a HFD [8.68 ± 0.64 (2–CT)] compared with littermate controls [4.75 ± 0.50 (2–CT), P < 0.05], indicating that increased expression of apoCIII could contribute to the decreased clearance of triglycerides in the bGH mice (32).
We found decreased mRNA levels of PPAR, SREBP-1a, SREBP-1c, and FAS in the bGH transgenic mice on a HFD compared with littermate controls (Table 2). These results indicate a further decrease in lipogenesis, as well as ?-oxidation, in the bGH mice in contrast to control mice as a result of a HFD. We therefore investigated the liver content of triglycerides in bGH mice and littermate controls after 8 wk of high-fat feeding. The liver triglyceride content was lower in the fat-fed bGH mice compared with the fat-fed controls (4.0 ± 0.2 μmol/g liver vs. 4.9 ± 0.2 μmol/g liver, P < 0.05).
IVGTT
One of the features of insulin resistance is a decreased turnover of triglyceride-rich lipoproteins and hypertriglyceridemia (33, 34). Elevated serum levels of GH have been shown to result in insulin resistance both in humans and animal models. Also a HFD fed ad libitum leads to obesity-induced insulin resistance. We therefore performed an IVGTT and compared the plasma levels of glucose and insulin in bGH mice and littermate controls receiving either normal chow or a HFD, to separate the contributing components GH and HFD in insulin resistance.
The bGH mice were hyperinsulinemic, but normoglycemic, on a ND (Fig. 7A and Table 3). The hyperinsulinemia was aggravated by the intake of the HFD in both genotypes (Fig. 7B and Table 3). Despite the increased serum insulin levels, the bGH mice became diabetic and had an impaired glucose elimination rate (KG) when fed a HFD, in contrast to the controls (Fig. 7C and Table 3).
TABLE 3. Basal insulin levels, basal glucose levels, and glucose disposal in bGH mice and littermate controls given a ND or a HFD for 8 wk
Discussion
The interaction between GH and HFD on weight gain, EE, and lipid and carbohydrate metabolism has not previously been investigated. We showed that bGH transgenic mice are resistant to diet-induced weight gain and obesity despite hyperphagia on a HFD. Moreover, we showed that the bGH transgenic mice increase their energy intake and EE on a HFD in contrast to control mice, indicating that the dietary fat induces an increased EE that results in increased food intake. Furthermore, we extended our previous findings of altered lipoprotein metabolism in bGH transgenic mice on a ND (3) by showing an opposite effect of GH overexpression on serum levels of triglycerides and apoB when the mice were given a HFD. Thus, the HFD resulted in dyslipidemia only in the bGH mice. Moreover, only the bGH mice became diabetic on a HFD. Together, our results show that the bGH mice on a HFD more accurately reflect the metabolic situation in acromegaly, for which they have long since been a model.
In line with a recent study (5), we observed no effect of GH overexpression on food intake of a ND. However, the bGH mice were clearly hyperphagic when fed a HFD. An appetite-promoting effect of GH has also been seen in a replacement study of children with Turner syndrome and Silver Russell syndrome who suffer from growth retardation (12). However, many food intake studies in rodents have failed to show an effect of GH on food intake (5, 14, 15, 16). It can therefore be speculated that the lack of, or small, effect on food intake observed in different GH models depends on the low percentage of fat in the food. The appetite-promoting effect of GH found in this study further strengthens the already approved treatment of AIDS wasting with GH where only the anabolic effects of GH have previously been acknowledged (36). Based on these experimental studies, the effect of the dietary-fat content on EE should be evaluated when treating AIDS patients with GH.
We observed that the RER increased in the early afternoon in the bGH mice but several hours later in the littermate controls. This finding indicates that the transgenic mice eat more during the light phase, which is in line with the finding of another study (16) showing that GH treatment of Zucker rats made them eat more during the day period without changing the total food intake. Thus, it can be concluded that the effect of GH on food intake is dependent on the content of dietary fat and that GH probably influences the diurnal rhythm of food intake.
In search for possible explanations for the complete resistance to diet-induced obesity of the bGH mice, we investigated EE. We found that the bGH mice had higher EE on a ND, that clearly increased by a HFD. This mimics the human situation where GH substitution to GH-deficient patients (17, 21), as well as acromegaly, increases EE (18, 19). However, it is not known whether the effect of GH on EE in humans is maintained on a low-fat diet. O'Sullivan et al. (18) observed that oral glucose increased EE in controls only, and not in acromegalic patients, indicating that low-fat diet might mitigate the calorigenic effect of GH overexpression.
Several alterations in other hormone levels have been described in these bGH transgenic mice, including increased serum levels of IGF-I (3), corticosterone, and T3 but lower levels of T4 (11) compared with littermate controls. Increased T3 levels may explain the increase in EE. However, in an elegant clinical study comparing the effect of GH and T3 on EE, it was concluded that increased T3 levels as a result of GH treatment cannot solely explain the calorigenic effect of GH (37). Increased corticosterone levels have several effects on intermediary metabolism, including decreased insulin sensitivity. However, it is unlikely that the effects on EE in the bGH mice are mediated by corticosterone, because acromegalic patients and GH treatment result in increased EE without concomitant changes in glucocorticoid levels. Moreover, dexamethasone treatment of lean mice decreases EE (38). Infusion of glucose and insulin to normal adults has been shown to increase EE (39), indicating that the hyperinsulinemia and hyperglycemia observed after fat feeding of the bGH mice may contribute to increased EE. However, the marked insulin resistance of the bGH mice might also counteract the calorigenic effect of hyperinsulinemia and hyperglycemia.
Few studies have addressed the influence of GH on EE or body temperature in rodents (22, 23). In another study of bGH transgenic mice, VO2 tended to be lower in the transgenic mice compared with controls. It was concluded that shivering thermogenesis is reduced but body temperature is unchanged in these animals (22). The reason for the different results is most likely due to differences in the experimental setup. They used restrained mice and measured the body temperature for an hour, whereas we measured the body temperature for 24 h in freely moving mice using telemetry. In line with our results, Hauck et al. (23) observed that GH receptor knock-out mice have reduced body temperature, indicating that marked changes in GH action for a prolonged period influence body temperature in both directions.
The liver and kidney size and, to a lesser extent, the heart and intestine size have been shown to account for more than 50% of the strain variation of EE in laboratory mice (40). The livers of bGH mice account for approximately 9% of the body weight, which is twice that of littermate controls, indicating that the change in liver size might contribute to the increased EE in the bGH mice. We also noted that the brown-fat pad (BAT) size was increased in the bGH mice, which is in contrast to a previous observation (41). In that study (41), they also observed lower expression of UCP-1 in the BAT depot, in contrast to our results. The reason for the discrepant results is unclear but may be due to the difference in age of the mice in the two studies. UCP1 has been implicated in thermogenesis by uncoupling the ATP production in the mitochondrial respiratory chain (42). The closely related UCP2 and UCP3 also function as uncouplers (43), but their metabolic role is less clear. Because UCP1 is exclusively expressed in BAT, its role in adult humans is thought to be minor, because they have minimal amounts of BAT. Therefore, the attention has been focused on UCP2, which is expressed in most tissues, and UCP3, which is mainly expressed in skeletal muscle (44, 45). Different polymorphisms in UCP2 have been linked to EE and substrate specificity for energy production (46, 47). Taken together, these findings indicate that higher expression of UCP2 in the enlarged livers of bGH mice might contribute to the increased EE observed on a ND. However, the contribution of UCP2 to the increase in EE observed in the bGH mice on a HFD, compared with those on normal chow, remains unclear. We have previously shown that hepatic gene expression of enzymes in ?-oxidation is lower in the bGH mice (48). This, together with the present finding of increased RER, indicates that fatty acids in the liver, to a lesser extent, are used for energy production. However, they can activate UCPs (43). We also observed that UCP2 expression increased by the dietary fat in littermate controls, in contrast to the bGH mice, indicating that the bGH mice are resistant to the stimulatory effect of fatty acids on UCP2 expression. UCP2 mRNA levels have previously been shown to be up-regulated by fatty acids and PPAR agonists (49, 50). The down-regulation of PPAR in the liver in the bGH mice, especially in the fat-fed state, might contribute to the unresponsiveness of the bGH mice.
We have previously studied hepatic gene expression in bGH mice on a ND and showed that they have reduced hepatic gene expression of SREBP-1c and lipogenic enzymes (48). Now, we extend these studies by showing a decrease in SREBP-1a mRNA levels in the bGH mice, compared with littermate controls, when fed a HFD. This finding shows a novel regulation of SREBP-1a and that the combined action of chronic overexpression of GH and HFD might reduce the hepatic de novo lipogenesis. Moreover, FAS and SREBP-1c decrease by fat feeding in bGH mice, whereas fat feeding had no effect or even tended to increase the expression of FAS in the control mice. These results shows that overexpression of GH represses the expression of genes involved in de novo lipogenesis.
Fat feeding altered the effect of GH overexpression on lipoprotein metabolism. Instead of decreased triglyceride secretion, unchanged triglyceride clearance, and decreased serum triglyceride levels on a ND (3), the fat feeding resulted in similar triglyceride secretion, markedly decreased triglyceride clearance, and hence increased serum triglyceride levels in the bGH mice compared with littermate controls. The increased serum triglyceride levels were associated with increased apoB levels in the VLDL/intermediate density lipoprotein fraction, showing that an increase in the number of lipoprotein particles in these fractions was responsible for the change. Because GH has been shown to markedly increase lipoprotein lipase activity in both mice and rats (3, 51), it is possible that our findings of increased hepatic apoCIII expression take part in the decreased triglyceride clearance. However, it must be pointed out that other mechanisms may be involved in the changed triglyceride clearance, including VLDL size, apoCI or CII levels, or the activity of lipases other than lipoprotein lipase. We are not aware of any experimental studies on the effect of GH on apoCIII expression or plasma levels, but it has been shown that GH treatment of children results in increased plasma apoCIII levels (35). Together, these findings show that fat feeding results in dyslipidemia of bGH mice that is not observed when these mice are given normal chow.
The RER was increased in the bGH mice, compared with littermate controls, on both diets. Increased glucose oxidation has also been reported in acromegalic patients (18). The reason for this can be speculated upon but is further supported by the decreased hepatic expression of PPAR in bGH mice on a HFD, indicating decreased hepatic ?-oxidation. In line with the finding of increased RER, we have also observed that the hepatic expression of genes involved in fatty acid activation, ?-oxidation, and ketone body formation is decreased in the bGH transgenic mice (48). This was also reflected in the decreased capacity of these mice to produce ketone bodies, both in the fed and fasted states (48). Together, these results may give a clue to the interesting result that young adults, but not old adults, show a positive correlation between RER and measures of GH secretion (20).
We found that bGH mice on a ND were severely insulin resistant, because they had markedly elevated insulin levels in the presence of normal glucose levels. Furthermore, because they had normal glucose elimination after an iv glucose challenge, it is clear that their hyperinsulinemia is adequate, i.e. the islet function seems normal. In contrast, after a HFD, when insulin resistance worsens, the hyperinsulinemia is inadequate, in view of the increased glucose levels and the glucose intolerance. Hence, the limit for islet compensation is reached in these mice. In fact, the insulin levels actually declined after glucose administration. This finding is hard to explain, but it is a clear indication of severe islet dysfunction. The elevated glucose levels were in the range of diabetic animals, and therefore the combination of bGH transgene and high-fat feeding is a challenge, inducing diabetes. The diabetic effect is due to a combined effect on insulin action, resulting in insulin resistance, and on islet function, resulting in inadequate hyperinsulinemia.
In conclusion, bGH transgenic mice were completely resistant to diet-induced obesity despite hyperphagia, possibly due to increased EE. Elevated body temperature and increased hepatic UCP2 expression may, together with increased locomotor activity (11), explain the increase in EE. Furthermore, the bGH transgenic mice became dyslipidemic and diabetic on a HFD. Thus, the bGH transgenic mice more accurately reflect acromegalic patients, with respect to metabolic parameters, when fed a HFD. Based on these findings, it would be interesting to investigate the effect of dietary manipulations on EE, RER, and lipid metabolism in acromegalic patients, anticipating that a change toward a carbohydrate-rich diet will reduce EE and ameliorate the dyslipidemia, as well as the carbohydrate metabolism.
Acknowledgments
We thank Lena Kvist and Mia Umearus for excellent technical assistance.
References
Palmiter RD, Brinster RL, Hammer RE, Trumbauer ME, Rosenfeld MG, Birnberg NC, Evans RM 1982 Dramatic growth of mice that develop from eggs microinjected with metallothionein-growth hormone fusion genes. Nature 300:611–615
Quaife CJ, Mathews LS, Pinkert CA, Hammer RE, Brinster RL, Palmiter RD 1989 Histopathology associated with elevated levels of growth hormone and insulin-like growth factor I in transgenic mice. Endocrinology 124:40–48
Frick F, Bohlooly YM, Lindén D, Olsson B, T?rnell J, Edén S, Oscarsson J 2001 Long-term growth hormone excess induces marked alterations in lipoprotein metabolism in mice. Am J Physiol Endocrinol Metab 281:E1230–E1239
Eisen EJ, Peterson CB, Parker IJ, Murray JD 1998 Effects of zinc ion concentration on growth, fat content and reproduction in oMT1a-oGH transgenic mice. Growth Dev Aging 62:173–186
Berryman DE, List EO, Coschigano KT, Behar K, Kim JK, Kopchick JJ 2004 Comparing adiposity profiles in three mouse models with altered GH signaling. Growth Horm IGF Res 14:309–318
Kopchick JJ, Bellush LL, Coschigano KT 1999 Transgenic models of growth hormone action. Annu Rev Nutr 19:437–461
Bengtsson BA, Brummer RJ, Edén S, Bosaeus I, Lindstedt G 1989 Body composition in acromegaly: the effect of treatment. Clin Endocrinol (Oxf) 31:481–490
Foss MC, Saad MJ, Paccola GM, Paula FJ, Piccinato CE, Moreira AC 1991 Peripheral glucose metabolism in acromegaly. J Clin Endocrinol Metab 72:1048–1053
Nikkil? EA, Pelkonen R 1975 Serum lipids in acromegaly. Metabolism 24:829–838
Oscarsson J, Wiklund O, Jakobsson KE, Petruson B, Bengtsson BA 1994 Serum lipoproteins in acromegaly before and 6–15 months after transsphenoidal adenomectomy. Clin Endocrinol (Oxf) 41:603–608
Bohlooly YM, Olsson B, Gritli-Linde A, Brusehed O, Isaksson OG, Ohlsson C, S?derpalm B, T?rnell J 2001 Enhanced spontaneous locomotor activity in bovine GH transgenic mice involves peripheral mechanisms. Endocrinology 142:4560–4567
Blissett J, Harris G, Kirk J 2000 Effect of growth hormone therapy on feeding problems and food intake in children with growth disorders. Acta Paediatr 89:644–649
Azain MJ, Roberts TJ, Martin RJ, Kasser TR 1995 Comparison of daily versus continuous administration of somatotropin on growth rate, feed intake, and body composition in intact female rats. J Anim Sci 73:1019–1029
Clark RG, Jansson JO, Isaksson O, Robinson IC 1985 Intravenous growth hormone: growth responses to patterned infusions in hypophysectomized rats. J Endocrinol 104:53–61
Roberts TJ, Azain MJ, Hausman GJ, Martin RJ 1994 Interaction of insulin and somatotropin on body weight gain, feed intake, and body composition in rats. Am J Physiol 267:E293–E299
Schulz C, Wieczorek I, Reschke K, Lehnert H 2002 Effects of intracerebroventricularly and intraperitoneally administered growth hormone on body weight and food intake in fa/fa Zucker rats. Neuropsychobiology 45:36–40
Carroll PV, Christ ER, Bengtsson BA, Carlsson L, Christiansen JS, Clemmons D, Hintz R, Ho K, Laron Z, Sizonenko P, Sonksen PH, Tanaka T, Thorne M 1998 Growth hormone deficiency in adulthood and the effects of growth hormone replacement: a review. Growth Hormone Research Society Scientific Committee. J Clin Endocrinol Metab 83:382–395
O’Sullivan AJ, Kelly JJ, Hoffman DM, Baxter RC, Ho KK 1995 Energy metabolism and substrate oxidation in acromegaly. J Clin Endocrinol Metab 80:486–491
M?ller N, Schmitz O, J?rgensen JO, Astrup J, Bak JF, Christensen SE, Alberti KG, Weeke J 1992 Basal- and insulin-stimulated substrate metabolism in patients with active acromegaly before and after adenomectomy. J Clin Endocrinol Metab 74:1012–1019
Christiansen JJ, Vahl N, Fisker S, M?ller N, Christiansen JS, J?rgensen JO 2002 Age dimorphism in the association between growth-hormone status and the respiratory quotient. Obes Res 10:284–290
Gerver WJ, De Bruin R, Delemarre v d Waal HA, Aldewereld B, Theunissen P, Westerterp KR 2000 Effects of discontinuation of growth hormone treatment on body composition and metabolism. Horm Res 53:215–220
Moura AS, Spiers DE, Lamberson WR 1998 Thermogenic activity of growth hormone transgenic mice. Growth Dev Aging 62:149–159
Hauck SJ, Hunter WS, Danilovich N, Kopchick JJ, Bartke A 2001 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 226:552–558
Yokode M, Hammer RE, Ishibashi S, Brown MS, Goldstein JL 1990 Diet-induced hypercholesterolemia in mice: prevention by overexpression of LDL receptors. Science 250:1273–1275
Masucci-Magoulas L, Goldberg IJ, Bisgaier CL, Serajuddin H, Francone OL, Breslow JL, Tall AR 1997 A mouse model with features of familial combined hyperlipidemia. Science 275:391–394
Surwit RS, Kuhn CM, Cochrane C, McCubbin JA, Feinglos MN 1988 Diet-induced type II diabetes in C57BL/6J mice. Diabetes 37:1163–1167
Ahrén B, Scheurink AJ 1998 Marked hyperleptinemia after high-fat diet associated with severe glucose intolerance in mice. Eur J Endocrinol 139:461–467
Sandstedt J, Ohlsson C, Norjavaara E, Nilsson J, T?rnell J 1994 Disproportional bone growth and reduced weight gain in gonadectomized male bovine growth hormone transgenic and normal mice. Endocrinology 135:2574–2580
Oscarsson J, Olofsson SO, Bondjers G, Edén S 1989 Differential effects of continuous versus intermittent administration of growth hormone to hypophysectomized female rats on serum lipoproteins and their apoproteins. Endocrinology 125:1638–1649
Bruss ML, Black AL 1978 Enzymatic microdetermination of glycogen. Anal Biochem 84:309–312
Li X, Catalina F, Grundy SM, Patel S 1996 Method to measure apolipoprotein B-48 and B-100 secretion rates in an individual mouse: evidence for a very rapid turnover of VLDL and preferential removal of B-48-relative to B-100-containing lipoproteins. J Lipid Res 37:210–220
Ebara T, Ramakrishnan R, Steiner G, Shachter NS 1997 Chylomicronemia due to apolipoprotein CIII overexpression in apolipoprotein E-null mice. Apolipoprotein CIII-induced hypertriglyceridemia is not mediated by effects on apolipoprotein E. J Clin Invest 99:2672–2681
Cummings MH, Watts GF, Umpleby AM, Hennessy TR, Naoumova R, Slavin BM, Thompson GR, S?nksen PH 1995 Increased hepatic secretion of very-low-density lipoprotein apolipoprotein B-100 in NIDDM. Diabetologia 38:959–967
Duvillard L, Pont F, Florentin E, Galland-Jos C, Gambert P, Verges B 2000 Metabolic abnormalities of apolipoprotein B-containing lipoproteins in non-insulin-dependent diabetes: a stable isotope kinetic study. Eur J Clin Invest 30:685–694
Schaefer GB, Greger NG, Fesmire JD, Blackett PR, Wilson DP, Frindik JP 1994 Lipids and apolipoproteins in growth hormone-deficient children during treatment. Metabolism 43:1457–1461
Grinspoon S, Mulligan K 2003 Weight loss and wasting in patients infected with human immunodeficiency virus. Clin Infect Dis 36:S69–S78
Wolthers T, Groftne T, M?ller N, Christiansen JS, ?rskov H, Weeke J, J?rgensen JO 1996 Calorigenic effects of growth hormone: the role of thyroid hormones. J Clin Endocrinol Metab 81:1416–1419
Chen HL, Romsos DR 1995 A single intracerebroventricular injection of dexamethasone elevates food intake and plasma insulin and depresses metabolic rates in adrenalectomized obese (ob/ob) mice. J Nutr 125:540–545
Ravussin E, Bogardus C, Schwartz RS, Robbins DC, Wolfe RR, Horton ES, Danforth Jr E, Sims EA 1983 Thermic effect of infused glucose and insulin in man. Decreased response with increased insulin resistance in obesity and noninsulin-dependent diabetes mellitus. J Clin Invest 72:893–902
Konarzewski M, Diamond J 1995 Evolution of basal metabolic rate and organ masses in laboratory mice. Evolution 49:1239–1248
Li Y, Knapp JR, Kopchick JJ 2003 Enlargement of interscapular brown adipose tissue in growth hormone antagonist transgenic and in growth hormone receptor gene-disrupted dwarf mice. Exp Biol Med Maywood 228:207–215
Enerb?ck S, Jacobsson A, Simpson EM, Guerra C, Yamashita H, Harper ME, Kozak LP 1997 Mice lacking mitochondrial uncoupling protein are cold-sensitive but not obese. Nature 387:90–94
Jaburek M, Varecha M, Gimeno RE, Dembski M, Jezek P, Zhang M, Burn P, Tartaglia LA, Garlid KD 1999 Transport function and regulation of mitochondrial uncoupling proteins 2 and 3. J Biol Chem 274:26003–26007
Vidal-Puig A, Solanes G, Grujic D, Flier JS, Lowell BB 1997 UCP3: an uncoupling protein homologue expressed preferentially and abundantly in skeletal muscle and brown adipose tissue. Biochem Biophys Res Commun 235:79–82
Fleury C, Neverova M, Collins S, Raimbault S, Champigny O, Levi-Meyrueis C, Bouillaud F, Seldin MF, Surwit RS, Ricquier D, Warden CH 1997 Uncoupling protein-2: a novel gene linked to obesity and hyperinsulinemia. Nat Genet 15:269–272
Esterbauer H, Schneitler C, Oberkofler H, Ebenbichler C, Paulweber B, Sandhofer F, Ladurner G, Hell E, Strosberg AD, Patsch JR, Krempler F, Patsch W 2001 A common polymorphism in the promoter of UCP2 is associated with decreased risk of obesity in middle-aged humans. Nat Genet 28:178–183
Le Fur S, Le Stunff C, Dos Santos C, Bougneres P 2004 The common –866 G/A polymorphism in the promoter of uncoupling protein 2 is associated with increased carbohydrate and decreased lipid oxidation in juvenile obesity. Diabetes 53:235–239
Olsson B, Bohlooly YM, Brusehed O, Isaksson OG, Ahrén B, Olofsson SO, Oscarsson J, T?rnell J 2003 Bovine growth hormone-transgenic mice have major alterations in hepatic expression of metabolic genes. Am J Physiol Endocrinol Metab 285:E504–E511
Cortez-Pinto H, Zhi Lin H, Qi Yang S, Odwin Da Costa S, Diehl AM 1999 Lipids up-regulate uncoupling protein 2 expression in rat hepatocytes. Gastroenterology 116:1184–1193
Tsuboyama-Kasaoka N, Takahashi M, Kim H, Ezaki O 1999 Up-regulation of liver uncoupling protein-2 mRNA by either fish oil feeding or fibrate administration in mice. Biochem Biophys Res Commun 257:879–885
Oscarsson J, Ottosson M, Vikman-Adolfsson K, Frick F, Enerb?ck S, Lithell H, Edén S 1999 GH but not IGF-I or insulin increases lipoprotein lipase activity in muscle tissues of hypophysectomised rats. J Endocrinol 160:247–255(Bob Olsson, Mohammad Bohl)