Biological and Pharmaceutical Bulletin
Online ISSN : 1347-5215
Print ISSN : 0918-6158
ISSN-L : 0918-6158
Regular Articles
Long-term High-Soybean Oil Feeding Alters Regulation of Body Temperature in Rats
Hiromi TsushimaKazuyo YamadaDaisuke MiyazawaMayumi MoriYoko HashimotoTakeshi OhkuboHidehiko HibinoHarumi Okuyama
ジャーナル フリー HTML

2014 年 37 巻 6 号 p. 1003-1013


We investigated whether body temperature (BT) regulatory mechanisms are influenced by dietary fatty acids (FA). Male Wistar rats were fed a high-fat diet containing fish oil (HFD), soybean oil (HSD) or lard (HLD). At the 20-week intervention, the BT of the HSD and HLD groups were lower than that of the normal diet (ND) group in the light and dark periods. The intracerebroventricular injections of interleukin-1β and bombesin in the HSD group induced greater hyperthermia and weaker hypothermia, respectively, than in the ND group. The HSD differentially affected BT under both physiological and pharmacological conditions. In the hypothalamus, the ratio of n-6/n-3 FAs was higher in the HSD group compared with the ND group. DNA microarrays revealed increased expression of thyroid-stimulating hormone β-subunit, and decreased expression of several genes in the hypothalamus of the HSD group compared with the ND group. The HSD feeding increased several adipokine concentrations in the plasma. However, there were no adipokines or gene expressions that changed in only the HSD and HLD groups showing significant hypothermia under the physiological condition. These findings suggested that long-term HSD intake produces abnormal BT regulation. It is less likely that adipokines or proteins/peptides are involved in abnormal BT regulation under the physiological conditions after HSD feeding.

We now obtain more energy from fat and glucose than ever before. These changes have induced hyperlipidemia and adipocyte hypertrophy, resulting in the risk of cardiovascular diseases, diabetes mellitus and obesity. From this point of view, there are many reports on lipid and/or glucose metabolisms during high-fat diet feeding in living animals.13) However, recent reports demonstrate that lipids not only influence metabolisms, but also neuronal function in the central nervous system (CNS).46) Psychiatric disorders such as Alzheimer’s disease and depression are suggested to be influenced by fat intake.79) These are probably brought through qualitative and quantitative changes in adipokines and/or lipid mediators after high-fat diet feeding. Most of these mediators have receptors/binding proteins and are integrated into intracellular signaling pathways in the CNS.7,1013) In addition, fatty acids (FA) and cholesterol are important components of membranes of cell, synaptic vesicles and mitochondria, and are responsible for the characteristics of membranes. The FA composition in the CNS can also be changed by fat intake.4,14)

Body temperature (BT) is maintained at the constant level on balance of heat production and expenditure in whole living body. The control center for BT is located in the preoptic area and/or the organum vasculosum of the lamina terminalis of the hypothalamus, and these neurons regulate metabolism resulting in heat production. Many biochemical studies suggested that high-fat feeding or obesity bring about hyperthermia induced by increase in thermogenesis, because these induced increases in uncoupling protein (UCP) mRNAs/proteins in brown adipose tissue and skeletal muscle, although they did not show data on BT.1518) However, when BT is actually measured during intervention of high-fat diet, various results on BT, hyperthermia, hypothermia or no changes, are reported.5,16,1921) As one of the reasons of the inconsistent results, the differences among experimental conditions, such as kinds of high-fat diets and feeding period, are raised. In this study, we used three kinds of high-fat diet, and investigated effects of dietary fat on BT regulation.


Animals and Diets

Four-week-old male Wistar rats were obtained from SLC Inc. (Shizuoka, Japan). The animals were housed individually at 23°C under a 12-h light/12-h dark cycle (light on at 08:00 a.m.) and were allowed free access to food and water. The experiments were carried out according to the Guidelines for Animal Care and Use of Kinjo Gakuin University College of Pharmacy, and the protocols were approved by the Institutional Animal Care and Use Committee of Kinjo Gakuin University College of Pharmacy.

We prepared three high-fat diets containing 20 w/w% fish oil (HFD), soybean oil (HSD) or lard (HLD) based on a normal powdered diet (ND; F-1 fish free: Funabashi Farm Co., Ltd.; Chiba, Japan). Total energy content and the energy percentages of fat were 467.5 kcal/100 g and 45.2% for the high-fat diets, and 360.0 kcal/100 g and 11.0% for the normal diet, respectively. The dietary interventions were carried out from 5 weeks of age to 30 weeks of age (tissue sampling). The diets were provided in a container with a dome-shaped cover (Roden Café: Oriental Yeast Co., Ltd., Tokyo, Japan) to prevent the diet from dropping out of the container and to prevent wood chips from entering the container. And these diets were exchanged twice per week, in order to avoid oxidation.

Experimental Protocol

To measure BT and locomotor activity, a transmitter of a telemetry system (Mini Mitter Co., Inc., Bend, Oregon, U.S.A.) was implanted intraperitoneally under pentobarbital anesthesia (50 mg/kg). The data were collected every 10 min for BT and every 60 min for locomotor activity using a computer running Vitalview Software (Mini Mitter Co., Inc.). The dietary intervention was started several days after the rats had recovered from the surgery.

At week 20 of the dietary intervention, the rats were anesthetized with pentobarbital, and a guide cannula was implanted into the right lateral ventricle (5.8 mm anterior to lambda, 1.8 mm lateral to the midline and 3.8 mm ventral to the skull surface) to administer interleukin-1β (IL-1β; 50 ng/5 µL) (R&D Systems, Minneapolis, MN, U.S.A.) and bombesin (Bom; 0.1 µg/5 µL) (Sigma Chemical, St. Louis, MO, U.S.A.) as described elsewhere.2224) The coordinate of the lateral cerebroventricle was determined using the atlas of Konig and Klippel.25) The first administration was performed 1 week after surgery, with a recovery period of at least 4 d before the second administration. On the experimental days, a cannula connected to a 10-µL syringe using 50-cm polyethylene tube was inserted into the guide cannula without anesthesia. The polyethylene tube contained the drug and saline to wash the cannula, with a small bubble between them. Then, the animal was placed in another cage (30×30×50 cm in size) for 30 min before drug administration. Five minutes after drug administration, a dummy cannula was inserted into the guide cannula and the rat was returned to its home cage. The drug administration took approximately 0.5 min and was performed at around 11:30. The rats were acclimatized to handling and the drug administration conditions during the recovery period after implanting the guide cannula.

After all experiments had been completed (1 week after the last drug administration; 24 weeks after starting the dietary intervention), the animals were decapitated under anesthesia with halothane, between 10:00–15:00 under normal eating condition. Trunk blood was collected ice-cold tubes without and with heparin, and centrifuged at 3000 rpm for 15 min at 4°C to separate plasma/serum. The plasma, serum and the brain were stored at −80°C until required for measurements.

Measurements of Plasma Parameters

Glucose, cholesterol, triglyceride, γ-glutamyl transpeptidase (GTP) and high density lipoprotein (HDL)-cholesterol were measured using a clinical autoanalyzer (TMS-1024: Tokyo Boeki Medical System Ltd., Tokyo, Japan).

Determination of Serum, Hypothalamic and Dietary FA Profiles

The hypothalamus was dissected from the whole brain using a brain matrix (ASI Instruments; Warren, MI, U.S.A.), and then divided into two blocks along the midline (52.8±4.4 mg/half block). One half was placed in a microtube containing 400 µL of buffer (50 mM Tris–HCl, pH 7.5; 150 mM NaCl; 1 mM ethylenediaminetetraacetic acid (EDTA); 1 mM dithiothreitol (DTT)) with protease inhibitors and phosphatase inhibitors. After sonication, the sample was used for FA analysis. Total lipids of hypothalamic homogenate, serum and diet sample were extracted by the modified method of Bligh and Dyer.26) FAs were converted to their methyl esters by treatment with 0.5% HCl in methanol and the contents were determined by capillary column (DB-225, J&W Scientific, CA, U.S.A.) gas–liquid chromatography (Shimadzu, Kyoto, Japan).27)

Hypothalamic mRNA Expression Analysis

The second half of the hypothalamus was sectioned and the sections were soaked in RNAlater®-ICE (Applied Biosystems Ambion; Foster, CA, U.S.A.) for RNA isolation, and stored at −80°C until analysis. The slices were dispersed into RNAwiz® (Applied Biosystems Ambion) and total RNA was extracted according to the manufacturer’s instructions. Total RNA was quantified by absorbance at 260 nm. One microgram of total RNA from each sample in the same group was combined and analyzed using a GeneSQUARE® (Kurabo Industrial Ltd., Osaka, Japan) multi-sample DNA microarray system to analyze the expression of 293 genes ( associated with obesity and lipid/glucose metabolism. Sample processing and data acquisition were carried out by Kurabo Industrial Ltd.

Some gene expressions, which were selected from the array results, were determined using a RT-qPCR system. Their primers (TaqMan® gene expression assays) and chemicals for PCR and cDNA syntheses (High Capacity RNA-to-cDNA kit) were obtained from Life Technologies (Carlsbad, CA, U.S.A.). PCR experiments were carried out according to the manufacturer’s instruction.

Detection of Various Adipokines in Plasma

The analysis was carried out using a commercial kit, Proteome Profiler Array® for rat adipokine (R&D Systems, Inc., MN, U.S.A.), which is able to detect thirty adipokines. One hundred microliters of plasma collected from five rats (20 µL/rat) in each diet group were diluted 15 times, and it was used as the sample. The procedures were performed according to the attached manual. Chemiluminescence was exposed on X-film for 5 min, and its density was measured using Image J 1.44p (NIH, U.S.A.).

Statistical Analyses

Data were presented as means±standard error (S.E.). Statistical analysis was carried out using Student’s t-test for within-group comparison or one-way ANOVA followed by Tukey’s post hoc test for between-group comparison. Values of p<0.05 were considered statistically significant.


Effects on Growth

The body weight of rats and plasma biochemical parameters were determined for all four groups of rats and are summarized in Table 1. Only the HLD group showed significant weight gain compared with the ND group after week 9 of dietary intervention (data not shown), and at 20-week intervention, with a 13% increase in body weight compared with the ND group. The body weights of the HFD and HSD groups were similar to those of the ND group. In contrast, in HSD group, weight of white and brawn adipose tissues were significantly increased during the intervention.

Table 1. Growth and Biochemical Parameters in the Plasma after Dietary Intervention
Diet intake (g)2662±521821±43*1817±45*2113±43*
Energy intake (kcal)9439±1898511±199*8495±211*9877±200
Body weight (g)491±8.2498±10.9510±12.2556±18.4*
BAT weight (g)0.47±0.020.69±0.090.78±0.07*0.70±0.06
WAT weight (g)19.45±1.4023.77±2.1035.00±3.08*34.93±4.66*
Glucose (mg/dL)151.2±3.9172.7±4.4162.3±6.1183.6±9.7*
Cholesterol (mg/dL)79.6±8.156.3±2.7*79.0±3.692.4±3.8
HDL-C (mg/dL)57.3±6.248.5±2.764.2±3.960.7±4.6
Triglyceride (mg/dL)64.8±17.915.7±1.5*43.7±5.267.2±7.7
γ-GTP (IU/L)3.3±2.73.7±1.54.7±2.22.0±1.6

ND: normal diet group; HSD: 20% soybean oil diet group; HFD: 20% fish oil diet group; HLD: 20% lard diet group; BAT: interscapular brawn adipose tissue; WAT: visceral white adipose tissue. Values are expressed as mean±S.E. of 5–6 rats. * p<0.05 vs. ND (one-way ANOVA followed by Tukey’s post hoc test).

Plasma biochemical parameters were measured after completing all of the experiments. The glucose concentrations tended to be higher in all three high-fat diet groups compared with the ND group, although this was only statistically significant between the HLD and ND groups. Cholesterol and triglyceride concentrations were significantly lower in the HFD group compared with the ND group. The other parameters measured were not significantly different between the high-fat diet groups compared with the ND group.

Effects on Body Temperature

Under physiological conditions before starting the dietary intervention, there were no significant differences in BT among the four groups in either the light or dark cycle. However, at the 20-week intervention (Figs. 1A, B), BTs were significantly lower in both light and dark cycles of the HSD and HLD groups than the ND group. In particular, BTs measured between 7:00 and 10:00 in these groups were notably different from that of the ND group (Fig. 1A). However, the HFD group showed similar BTs to the ND group throughout one day. These phenomena were not found at week before 16 after dietary intervention.

Fig. 1. Body Temperatures under Physiological Conditions after 20 Weeks of Dietary Intervention

(A) Diurnal BT profiles measured after 20 weeks of dietary intervention. BT was measured at 10-min intervals. The thin and thick lines of X-axis show light and dark periods, respectively. (B) BT in the light and dark cycles of panel A. * p<0.05: ND vs. HLD; #p<0.05: ND vs. HSD. Values are mean±S.E. n=5 rats/group. BT values ≤35°C or >40°C were excluded from the analysis.

After 20 weeks of the dietary intervention, the effects of IL-1β (50 ng/5 µL) and Bom (0.1 µg/5 µL) on BT were investigated following administration into the lateral cerebroventricle (Fig. 2), IL-1β and Bom act as pyrogenic and hypothermic drugs, respectively.22,24) In all of the high-fat diet and ND groups, IL-1β increased BT with two peaks. Because the first coinciding with drug administration were also observed after the injections of vehicle, the second were caused by IL-1β, which showed statistically significant increases in BT, compared with BT after vehicle administration in each groups (Figs. 2A, C). In the ND group, the hyperthermia occurred approximately 60 min after injection and lasted for over 120 min, and BTs from 80 min to 310 min after injection increased significantly by approximately 1.2°C relative to the basal level. The IL-1β-induced hyperthermia in the HLD group was similar to that in the ND group. In contrast, this effect was delayed in the HFD group, as BTs at 80 and 90 min after injection were lower than that in the other groups. However, the subsequent increases in BT were not different from those of the ND group. On the other hand, administration of IL-1β in the HSD group increased BT by approximately 2°C relative to baseline, which lasted for over 360 min after injection. These increases during 220–360 min after injection were statistically higher than those in the ND group.

Fig. 2. Effects of High-Fat Dietary Intervention on the Change in Body Temperature from Basal Levels Following the Administration of IL-1β or Bom

(A) IL-1β (50 ng/5 µL) was injected at time 0 into the cerebroventricle and elicited significant hyperthermia compared with vehicle in each group (C) (* p<0.05; 80 to 310 min after injection in ND and HLD groups; 80 to 360 min after injection in HSD; 120 to 300 min after injection in HFD group). However, HFD group showed lower BT at 80 and 90 min after injection of IL-1β than the other groups (#p<0.05). On the other hand, in HSD group, IL-1β produced the potent effect between 220 and 360 min after injection, compared with the other groups (#p<0.05). (B) Bom (0.1 µg/5 µL) was injected at time 0 into the cerebroventricle. Bom decreased BT in the ND, HFD and HLD groups, but not in the HSD group, compared with vehicle (* p<0.05). (C) Saline (5 µL) was injected at time 0 into the cerebroventricle. Body temperature was measured at 10 min-intervals and values are means±S.E. ND: n=3–4; HFD: n=5–6; HSD: n=4–5; HLD: n=4–5. BT at baseline was 38.0±0.46°C, 37.7±0.15°C, 36.6±0.20°C and 36.9±0.19°C in the ND, HFD, HSD and HLD groups, respectively. Basal BT values ≤35°C or >40°C were excluded from the analysis.

Administration of Bom quickly decreased BT in the ND group with a maximum change in BT of –2.1±0.47°C from baseline at 80 min after injection (Figs. 2B, C). BT returned to baseline levels by 190 min after injection in the ND group. The artificial increase in BT, which was observed after vehicle injection, also disappeared. Changes in BT following Bom administration tended to be smaller in all three high-fat diet groups compared with the ND group, with the smallest change in the HSD group. When the Bom-induced changes in BT were compared with those after vehicle administration in each group, there were significant differences in the ND, HFD and HLD groups, but not in the HSD group.

Effects on Locomotor Activity

Locomotor activity in all four groups is shown in Fig. 3. Twenty weeks after dietary intervention, the HFD group was relatively active, but their changes of the 1-h intervals or the periods were not statistically significant compared with ND group. No significant effects were detected in the other groups.

Fig. 3. Locomotor Activity under Physiological Conditions after 20 Weeks of Dietary Intervention

(A) Diurnal motor activity profiles measured after 20 weeks of dietary intervention. Locomotor activity was measured at 1-h intervals. The thin and thick lines of X-axis show light and dark periods, respectively. (B) Locomotor activity in the light and dark cycles of panel A. ND: normal diet group; HSD: 20% soybean oil diet group; HFD: 20% fish oil diet group; HLD: 20% lard diet group. Values are expressed as mean±S.E. of 5–6 rats.

Effects on Hypothalamic and Serum Fatty Acid Profiles

Shown in Table 2A, the ratio of n-6/n-3 FAs was lower in the HFD compared with the normal diet. In the HLD, there were a large proportion of saturated and monounsaturated FAs, and the high ratio of n-6/n-3 FAs. These characteristics were reflected in FA profiles of the serum (Table 2B).

Table 2. Fatty Acid Profiles in the Diets (A), and the Serum (B) and the Hypothalamus (C) after 24 Week-Dietary Intervention
A) Diets
Fatty acidsDiet
14 : 00.37±0.052.34±0.29*0.11±0.01*1.09±0.10*
14 : 1Nd0.07±0.010.01±0.010.06±0.01
15 : 00.08±0.010.64±0.040.03±0.000.07±0.00
15 : 1Nd0.05±0.03Nd0.01±0.01
16 : 0 DMANdNdNdNd
16 : 015.76±0.1320.15±0.54*11.37±0.09*23.56±0.17*
16 : 10.49±0.013.56±0.15*0.09±0.00*1.77±0.07*
17 : 10.15±0.000.72±0.01*0.05±0.00*0.27±0.00*
18 : 0 DMANd0.04±0.01NdNd
18 : 05.83±0.065.51±0.104.04±0.05*10.32±2.23*
18 : 1 DMANdNdNd2.20±2.20
18 : 1 n-918.31±0.2219.76±0.1426.10±0.16*40.24±0.07*
18 : 1 n-71.11±0.012.29±0.01*1.11±0.041.98±0.08
18 : 2 n-650.63±0.0410.43±0.13*50.42±0.0915.51±0.21*
18 : 3 n-60.14±0.000.18±0.00*0.29±0.00*0.03±0.01*
18 : 3 n-34.12±0.021.24±0.01*5.22±0.01*0.97±0.03*
20 : 00.27±0.000.50±0.11*0.32±0.01*0.20±0.00
20 : 10.51±0.001.67±0.06*0.25±0.00*0.68±0.01
20 : 2 n-6Nd0.24±0.010.03±0.000.31±0.00
20 : 3 n-60.08±0.080.10±0.000.01±0.010.05±0.00
20 : 4 n-6Nd1.57±0.030.02±0.000.17±0.08
20 : 3 n-3Nd0.21±0.10Nd0.02±0.02
20 : 5 n-30.82±0.005.99±0.03*0.03±0.00*0.05±0.02*
22 : 00.31±0.120.16±0.01*0.30±0.020.28±0.24
22 : 10.04±0.040.90±0.05*Nd0.02±0.02*
22 : 2NdNdNdNd
22 : 4 n-6Nd0.20±0.010.04±0.010.10±0.02
22 : 5 n-6Nd0.98±0.030.03±0.030.01±0.01
22 : 5 n-30.73±0.010.68±0.19Nd0.01±0.01*
22 : 6 n-3Nd19.22±0.47Nd0.02±0.02
24 : 00.13±0.000.13±0.010.12±0.01Nd
24 : 10.12±0.000.49±0.04NdNd
PUFA n-650.85±0.1113.70±0.12*50.84±0.0916.18±0.13*
PUFA n-36.66±0.0127.33±0.78*5.25±0.01*1.07±0.02*
n-6/n-3 ratio8.98±0.030.50±0.02*9.68±0.00*15.18±0.20*
Table 2. Continued
B) Serum
Fatty acidsSerum
14 : 00.24±0.010.40±0.050.09±0.020.19±0.03
14 : 1NdNd0.03±0.02Nd
15 : 00.31±0.020.32±0.020.13±0.03*Nd
15 : 1Nd0.12±0.050.04±0.02Nd
16 : 0 DMANdNdNdNd
16 : 019.19±0.4316.12±0.52*13.26±0.84*15.97±0.78*
16 : 10.82±0.140.81±0.080.12±0.02*0.42±0.03*
17 : 1Nd0.29±0.020.02±0.010.05±0.02
18 : 0 DMANdNdNdNd
18 : 09.64±0.3413.50±0.37*13.75±0.87*15.48±0.32*
18 : 1 DMANdNdNdNd
18 : 1 n-99.18±0.488.91±0.4310.30±0.9616.40±0.82*
18 : 1 n-72.10±0.071.36±0.071.17±0.051.14±0.04
18 : 2 n-629.47±0.909.53±0.11*27.50±1.4516.61±0.33*
18 : 3 n-60.26±0.030.28±0.060.37±0.030.22±0.01
18 : 3 n-30.85±0.070.41±0.09*1.04±0.190.14±0.04*
20 : 0Nd0.20±0.030.21±0.020.10±0.03
20 : 10.16±0.010.67±0.24*0.15±0.030.24±0.02
20 : 2 n-60.27±0.010.15±0.010.28±0.060.15±0.03
20 : 3 n-60.45±0.050.33±0.010.37±0.020.42±0.01
20 : 4 n-622.27±1.3521.80±0.7626.40±2.2527.61±1.85
20 : 3 n-3Nd0.11±0.05NdNd
20 : 5 n-3Nd6.74±0.250.23±0.01Nd
22 : 0Nd0.19±0.030.34±0.04Nd
22 : 11.80±0.85NdNd3.17±0.88
22 : 2Nd0.14±0.14NdNd
22 : 4 n-6Nd0.12±0.050.38±0.03Nd
22 : 5 n-6Nd1.02±0.140.40±0.04Nd
22 : 5 n-3Nd1.08±0.430.57±0.16Nd
22 : 6 n-32.62±0.2213.55±0.34*1.99±0..531.26±0.13*
24 : 0Nd0.67±0.160.63±0.08Nd
24 : 10.37±0.021.18±0.03*0.23±0.030.44±0.03
PUFA n-652.72±0.6033.23±0.74*55.70±0.9545.00±1.68*
PUFA n-33.47±0.2421.88±0.57*3.83±0.471.40±0.14*
n-6/n-3 ratio15.51±1.131.53±0.06*15.35±1.3433.80±4.29*
Table 2. Continued
C) Hypothalamus
Fatty acidsHypothalamus
14 : 00.11±0.060.18±0.070.17±0.020.13±0.06
14 : 1Nd0.20±0.150.03±0.03Nd
15 : 0Nd0.21±0.180.05±0.030.06±0.03
15 : 1Nd0.37±0.270.02±0.020.19±0.12
16 : 0 DMA1.58±0.122.31±0.191.36±0.161.81±0.15
16 : 017.71±1.1516.68±0.7817.05±0.3316.66±0.26
16 : 10.20±0.080.39±0.180.29±0.080.23±0.06
17 : 1Nd0.14±0.110.03±0.030.53±0.24
18 : 0 DMA3.15±0.204.18±0.272.96±0.373.45±0.27
18 : 021.08±1.0920.91±0.6121.50±0.3521.08±0.60
18 : 1 DMA1.54±0.072.45±0.591.13±0.251.68±0.14
18 : 1 n-919.02±1.2021.72±0.4821.20±0.4121.77±1.04
18 : 1 n-73.20±0.253.17±0.093.29±0.103.57±0.20
18 : 2 n-60.64±0.050.52±0.021.47±0.22*0.42±0.05
18 : 3 n-60.25±0.180.25±0.250.08±0.050.09±0.02
18 : 3 n-30.03±0.030.26±0.170.06±0.040.02±0.02
20 : 02.04±1.360.73±0.020.71±0.031.35±0.69
20 : 11.83±0.212.05±0.352.39±0.152.29±0.45
20 : 2 n-60.52±0.230.08±0.08*0.30±0.180.17±0.06
20 : 3 n-60.60±0.170.23±0.03*0.28±0.01*0.24±0.02
20 : 4 n-68.27±0.375.16±0.40*7.54±0.188.09±0.11
20 : 3 n-3Nd0.09±0.07Nd0.46±0.44
20 : 5 n-3Nd0.18±0.04Nd0.17±0.17
22 : 00.56±0.100.71±0.050.56±0.320.60±0.06
22 : 13.28±3.220.24±0.030.17±0.051.14±0.81
22 : 2 n-6Nd0.12±0.120.10±0.07Nd
22 : 4 n-6Nd1.87±0.133.30±0.19Nd
22 : 5 n-60.10±0.060.45±0.01*0.55±0.06*0.71±0.04*
22 : 5 n-3Nd0.82±0.400.02±0.02Nd
22 : 6 n-39.95±0.508.69±0.899.38±0.278.99±0.34
24 : 0Nd1.54±0.061.27±0.35Nd
24 : 14.34±1.463.09±0.162.73±0.134.12±1.29
PUFA n-610.39±0.608.57±0.2513.52±0.299.72±0.64
PUFA n-39.98±0.4810.03±0.929.46±0.269.64±0.64
n-6/n-3 ratio1.05±0.070.87±0.051.43±0.04*1.03±0.07

Values of fatty acids (FA) represent mean±S.E. of % of total FA (A: µg/mg of tissue, B: µg/µL of serum, C: µg/mg of diet) of A: n=4–5, B: n=5–6 and C: n=3. * p<0.05 vs. ND (one-way ANOVA followed by Tukey’s post hoc test). Nd: not detected; DMA: dimethylacetal derivates; SFA: saturated FA; MUFA: monounsaturated FA; PUFA: polyunsaturated FA.

In the hypothalamus, saturated, monounsaturated and polyunsaturated FAs comprised 45–47%, 31–35% and 19–23%, respectively, of the total FA content in all four groups (Table 2C). Accordingly, the ratios of saturated/monounsaturated/polyunsaturated FAs in the hypothalamus were fairly similar among the four groups despite long-term feeding with four diets containing different FA contents. However, the ratios of n-6/n-3 FAs in the HSD group was highest among those in the four groups (p<0.05). In terms of individual FAs in the hypothalamus, the levels of 18 : 2, 22 : 4 and 22 : 5 n-6 FAs were higher and that of 20 : 3 n-6 FA was lower in the HSD group than in the ND group. Meanwhile, the levels of 22 : 4 n-6, 22 : 5 n-6 and 22 : 5 n-3 FAs were higher and those of 20 : 3 n-6 and 20 : 4 n-6 FAs were lower in the HFD group than in the ND group. The hypothalamic FA profile in HLD group was similar to that of the ND group.

Effects on Hypothalamic Gene Expression

The HSD group showed marked changes in BT compared with the ND group under both physiological and pharmacological conditions. Therefore, of 293 genes examined using the microarray experiment, genes expressed at >2-fold and <0.5-fold in this group compared with ND group were summarized in Table 3A.

Table 3. Changes in Hypothalamic Gene Expression after Dietary Intervention of High-Fat Diets
Official symbolOfficial dull nameHFDHSDHLD
TshbThyroid stimulating hormone, beta subunit2.118.8Nd
Acsl4Acyl-CoA synthetase long-chain family member
Bhlhb2Basic helix-loop-helix domain containing, class B,
CrkCG1587 gene product from transcript CG1587-RC0.20.40.8
LplLipoprotein lipase0.10.40.4
Nr3c1Nuclear receptor subfamily 3, group C, member
Nr3c2Nuclear receptor subfamily 3, group C, member
Pdha1Pyruvate dehydrogenase E1 alpha
Pdk4Pyruvate dehydrogenase kinase, isoenzyme
Pik3cbPhosphatidylinositol 3-kinase, catalytic, beta polypeptide0.10.50.4
Prkaa2Protein kinase, AMP-activated, alpha 2 catalytic subunit0.10.50.4
PrkcqProtein kinase C, theta0.10.50.3
Official symbolOfficial full nameHFDHSDHLD
TshbThyroid stimulating hormone, beta subunit5.98±2.059.48±2.67*1.88±1.14
Pik3cbPhosphatidylinositol 3-kinase, catalytic, beta polypeptide0.59±0.170.86±0.121.01±0.15
Prkaa2Protein kinase, AMP-activated, alpha 2 catalytic subunit0.42±0.15*0.70±0.090.83±0.08

Nd: not determined. The numbers in the tables show fold changes to gene expression of ND group, which were investigated with DNA microarray system (A) and RT-qPCR experiments (B). A: These shows gene expression changes of more than 2-fold or less than 0.5-fold after HSD intervention compared with ND. B: The values show mean±S.E. of 5 rats. * p<0.05 vs. ND (one-way ANOVA followed by Tukey’s post hoc test)

The mRNA expression of the thyroid-stimulating hormone (THS) bet subunit was remarkably up-regulated in the hypothalamus of HSD group. Eleven genes were down-regulated in the HSD group compared with the ND group; however, these genes were also decreased in the HFD group.

Of them, gene expressions of TSH beta subunit, phosphatidylinositol 3-kinase catalytic beta polypeptide and AMP-activated protein kinase alpha 2 catalytic subunit were investigated with RT-qPCR (Table 3B). This showed a similar up-regulation of TSH beta subunit mRNA in the HSD group.

Effects on Adipokine Concentrations in Plasma

Of thirty adipokines analyzed, the fourteen showed at higher concentration more than 2 fold in the three high-fat diet groups compared with the ND group, although HGF, IL-1β, IL-6, IL-11, LIF, RANTES, serpin E1, TNF-α or VEGF were not detected in the all groups (Fig. 4). The remarkable increases were found of DPPIV, endocan, FGF-21, ICAM-1, IGFBP-1, −3, −5, leptin, MCP-1, M-CSF, Pref-1, RAGE, resistin, and TIMP-1 in the HFD and HSD groups, compared with the ND group. In the HLD group, the concentrations of some adipokines, DPPIV, IGFBP-3, -5, resistin and TIMP-1, increased in plasma; however, the other adipokines showed similar level of the plasma concentrations, compared with the ND group. However, there were no adipokines that changed in only the HSD and HLD groups showing significant hypothermia under the physiological condition.

Fig. 4. Detection of Various Adipokines in the Plasma after High-Fat Dietary Intervention

The vertical axis shows the chemiluminescent intensity of each spot. DPPIV: Dipeptidyl peptidase 4; FGF: Fibroblast growth factor, HGF: Hepatocyte growth factor; ICAM: Intracellular adhesion molecule, IGF: Insulin-like growth factor, IGFBP: IGF binding protein; LIF: Leukemia inhibitory factor; MCP: Monocyte chemotactic protein; M-CSF: Macrophage colony-stimulating factor; Pref-1: Preadipocyte factor-1; RAGE: Receptor for advanced glycation endproducts; RANTES: Regulated on activation, normal T cell expressed and secreted; TIMP: Tissue inhibitor of metalloproteinase; VEGF: Vascular endothelial growth factor.


After long-term dietary intervention, we detected marked hypothermia in the HSD and HLD groups relative to the ND group under physiological conditions, while the HFD group showed similar BT to the ND group. These results indicate that the different effects on BT were caused by the type of fat in the diets. The greatest hypothermic effects were found to occur between 7:00 and 10:00 a.m. in the HSD and HLD groups. This phenomenon may be connected with changes in the expression of certain clock genes, because the change from the dark cycle to the light cycle was set for 8:00 a.m., and it has been reported that a high-fat diet and hyperlipidemia affect the expression of genes involved in circadian rhythm and energy homeostasis.5,2830) Mendoza et al.5) showed that BT increased and decreased in the light and dark cycles, respectively, in high-fat diet-fed animals compared with low-fat diet-fed animals, and that the circadian BT rhythm is less apparent following high-fat diet feeding, findings that differ from our own. They used mice fed a saturated FA-rich diet.

Some studies have reported decreased UCP protein/mRNA in adipose tissue or skeletal muscle during high-fat feeding that may lead to hypothermia.3133) From experiments using tumor necrosis factor (TNF) receptor 1 knock out (KO) mice, it was concluded that TNF-α is increased by a high-fat diet, resulting in decreased UCP mRNA, and this reduces thermogenesis in diet-induced obesity, although BT was not measured.31,32) Their observation, reduced thermogenesis after the high-fat dietary intervention is not inconsistent with the hypothermia observed in this study. Under this experimental conditions, plasma concentrations of TNF-α of both ND and HSD groups are less than the limit of detection for this assay. In contrast, increased UCP mRNA/protein and increased inflammatory adipokine concentrations have also been observed in the hypothalamus/plasma in response to high-fat dietary intervention.18,20,34,35) In studies when BT was measured after high-fat dietary intervention, either an increase or no change in BT have both been observed.16,1921,36,37) Our results show that HSD and HLD produced hypothermia, whereas HFD did not influence BT. These inconsistent results are likely to be brought from the different kinds and content of fatty acids in the diets that influence biosynthesis and secretion of adipokines or lipid mediators, and characteristics of the cell membrane. Both HSD and HLD groups showed increased adipose tissue weight, and increased concentrations of certain adipokines in plasma. However, these groups did not show any changes in the plasma concentration of inflammatory adipokines such as IL-1β, TNF-α or IL-6, which are known to produce hyperthermia after administration. Furthermore, adipokine array experiment did not show any changes observed only in the HSD and HLD groups, in which hypothermia was induced, compared with HFD and ND groups. Recently, some of the other adipokines (visfatin, leptin, and IGF) were shown to produce hyperthermia after administration into the CNS.12,3841) In addition, because high-fat feeding alters the FA content in tissues and body fluids, the biosynthesis and secretion of lipid mediators may be altered. These have bioactive effects mediated through receptors or binding proteins, and some lipids are known to activate enzymes.7,9) Anandamide and sphingosine-1-phosphate led to increased BT following central injection.13,42) To date, no adipokines or lipid mediators have been shown to produce hypothermia. In contrast, there are several reports that FAs can influence membrane fluidity or characteristics of receptors/ion channels in the cell membrane. The FA composition in the CNS can also be changed by high-fat diets4,14); thus, a high-fat diet probably influences membrane functions of neurons and glia. Li et al. reported that cells cultured in eicosapentaenoic acid-rich medium show decreased caveolin-1 and eNOS protein expression in caveolae, along with changes in the FA profiles of the caveolae.43) They concluded that the microenvironment of the caveola modifies the location and function of proteins in the membrane. Furthermore, in cells cultured in medium containing certain FAs, the properties of ion channels/receptors and signaling pathways are modified by structural changes in the lipid raft/caveolae.4345) This may play a role in hypothermia.

We investigated the effects of dietary intervention on changes in BT induced by representative pyrogenic (IL-1β) and hypothermic (Bom) drugs injected into the cerebroventricle to act directly on the CNS.22,24) Interestingly, the HSD group showed enhanced and prolonged hyperthermia in response to IL-1β, compared with the ND group. In contrast, the HFD delayed the onset of IL-1β-induced hyperthermia. The effects of IL-1β have been shown to involve the production of prostaglandin E2 (PGE2) following cyclooxygenase activation. In the HFD group, the concentration of arachidonic acid (20 : 4 n-6) may partly contribute to the different responses to IL-1β, because analysis of FA profiles revealed its low levels in the hypothalamus. However, the pool sizes may not be particularly small, although the levels of the precursors of PGE2 may be insufficient for the onset of hyperthermia. Therefore, we suggest that PGE2 is rapidly synthesized and the late stage of the hyperthermic effect induced by IL-1β is similar to that in the ND group. It was previously demonstrated that animals fed an n-3 FA-rich oil showed a very weak response to IL-1β.4648) In contrast, in the HSD group, the concentrations of arachidonic acid are similar in the serum and hypothalamus to those of the ND group. Pohl et al. show that high-fat-fed rats have prolonged fever after LPS injection, and concluded that this effect is mediated through an increase in circulating cytokines, especially leptin.37) Another report demonstrated that leptin causes fever mediated through increased release of IL-1β and prostaglandins.38) It has also been shown that FA modulates IL-1β-induced increases in monoamine turnover and PGE2 level.49) These observations suggest that leptin and monoamines are involved in the different responses to IL-1β-induced hyperthermia among groups fed different fats. However, because PGE2 plays a central role in hyperthermia, activities of enzymes involved in PG biosynthesis may be influenced by HSD feeding. In addition, the HSD group showed markedly weaker hypothermic responses to Bom. The effects of Bom on BT are mediated by gastrin-releasing peptide receptor, one of the three Bom receptor subtypes, and activation of PKC in the CNS, although the detailed mechanisms are still unclear.24) The HSD was found to influence BT regulation under both physiological and pharmacological conditions. These findings suggest that long-term HSD feeding influences the hypothalamic BT regulatory system.

The results of microarray analysis showed that the expression of 12 genes was affected by the HSD intervention. In the HSD group, only the TSH β-subunit gene, which enhances thermogenesis in BAT mediated by thyroid hormone secretion, was highly up-regulated. We propose that this enhancement was a result of hypothermia under physiological conditions. There are some reports showing increased plasma T3/T4 levels during high-fat dietary interventions, although BT was not measured in these studies.16,20) The other 11 genes were down-regulated in the HSD group compared with the ND group; these genes were also down-regulated in the HFD and HLD groups. Therefore, there were no genes showing altered expression in response to HSD alone. This suggests that proteins/peptides are unlikely to be involved in the changes in BT regulation following the HSD intervention.

In summary, this study showed that long-term intervention with the HSD produced hypothermia under physiological conditions, with different effects on BT after injections of IL-1β and Bom, compared with the ND group. It is less likely that adipokines, lipid mediators or proteins/peptides are involved in abnormal BT regulation under the physiological conditions after HSD feeding.


The authors would like to thank Yui Imai, Eri Yamada and Eri Takagi: Kinjo University College of Pharmacy, for experimental assistant and animal care. This work was supported by the Grants from the Ministry of Education, Culture, Sports, Science and Technology of Japan (070025, 24580204) and from Kinjo Gakuin University.

© 2014 The Pharmaceutical Society of Japan