Dynamics of appetite-mediated gene expression in daidzein-fed female rats in the meal-feeding method

Abstract

We previously found that daidzein decreased food intake in female rats. The present study aimed to elucidate the relationship between dynamics of appetite-mediated neuropeptides and the anorectic effect of daidzein. We examined appetite-mediated gene expression in the hypothalamus and small intestine during the 3 meals per day feeding method. Daidzein had an anorectic effect specifically at the second feeding. Neuropeptide-Y (NPY) and galanin mRNA levels in the hypothalamus were significantly higher after feeding in the control but not in the daidzein group, suggesting that daidzein attenuated the postprandial increase in NPY and galanin expression. The daidzein group had higher corticotrophin-releasing hormone (CRH) mRNA levels in the hypothalamus after feeding, and increased cholelcystokinin (CCK) mRNA levels in the small intestine, suggesting that CCK is involved in the hypothalamic regulation of this　anorectic effect. Therefore, daidzein may induce anorexia by suppressing expression of NPY and galanin and increasing expression of CRH in the hypothalamus.

Graphical abstract

Daidzein feeding changes expression of NPY, galanin, and CRH in the hypothalamus and CCK in the small intestine, resulting in the anorectic effect.

Key words:

• daidzein
• appetite
• neuropeptide-Y(NPY)
• galanin
• cholelcystokinin

The soybean isoflavones genistein and daidzein are known to be analogous to 17β-estradiol. These isoflavones are ligands for estrogen receptors (ER) and result in the induction of estrogen-activated factors.¹⁾ The binding capacities of these isoflavones are >1000-fold lower than 17β-estradiol.^2–4) Soybean isoflavones function as estrogenic agents in estrogen-deficient conditions, and they conversely function as anti-estrogenic agents or have no effect in estrogen-sufficient conditions.⁵⁾ Increases in 17β-estradiol suppress food intake, and consequently induce decreases in body weight gain.^6,7) Administration of 17β-estradiol into female rats was found to lower food intake just after administration.⁸⁾ Continuous estradiol treatment in ovariectomized (OVX) rats has been shown to induce the estrus cycle, leading to reduced food intake and lower body weight.⁹⁾ In ERα knockout mice, administration of estradiol did not change food intake, suggesting that the effect of estrogen on food intake is mediated via ERα.¹⁰⁾

We¹¹⁾ and others¹²⁾ found that dietary daidzein reduced food intake in female rats, while genistein feeding did not change food intake. Guo Y et al. also reported that daidzein feeding induced an anorectic effect in obese mice fed high-fat diets.¹³⁾ However, some findings have shown different results of estradiol treatment and daidzein feeding. We showed that 17β-estradiol treatment reduced food intake in both male and female rats.¹¹⁾ Meanwhile, feeding with an isoflavone aglycone-rich fermented soybean extract or dietary daidzein decreased food intake in female rats but not in male rats.¹¹⁾ Further, estradiol treatment in OVX rats recovered uterine weight to levels similar to those seen in sham rats, but daidzein feeding did not.¹¹⁾ These observations suggest that anorectic effect of daidzein might be mediated by a different mechanism than estradiol. However, daidzein appears to function as selective estrogen receptor modulators¹⁴⁾ which are a group of estrogen receptor agonist/antagonist with tissue selective effects. Daidzein could suppress food intake in part through binding to the estrogen receptor and activation of the transcriptional activity of the receptor.

Several studies have been performed analyzing interactions between estradiol and neuropeptide-Y (NPY). Bonavera et al. reported that 17β-estradiol treatment decreased NPY release in the hypothalamic paraventricular nucleus (PVN) and induced anorexia.¹⁵⁾ Ainslie et al. also reported that OVX-induced estrogen deficiency increased hypothalamic NPY.¹⁶⁾ These findings suggest that estrogen downregulates hypothalamic NPY and subsequently reduces food intake. In addition, estradiol treatment increased gene expression of the anorectic corticotrophin-releasing hormone (CRH) in vivo and in vitro.^17,18) Accordingly, estradiol treatment induces the variation of the hypothalamic appetite-mediated neuropeptides. Asarian and Geary ¹⁹⁾ demonstrated that one mechanism by which estradiol decreases eating is increasing the satiating action of cholelcystokinin (CCK), at least in rats. Estradiol treatment increased CCK-induced c-Fos expression in the brain of OVX rats.^20,21) Estradiol treatment also induced CCK mRNA expression in the central portion of the medial preoptic nucleus and posterodorsal medial amygdale.²²⁾ In the periphery, estradiol treatment elevated plasma levels of estradiol and CCK in a dose-dependent manner. ²³⁾ Thus, we inferred that the anorexic action of estradiol may be mediated also by an increased gene expression of CCK in upper intestine, where CCK is primarily produced and secreted, although estradiol enhanced the satiating potency of exogenous CCK in OVX rats.¹⁹⁾ It would be indispensable to investigate the dynamics of appetite-mediated gene expression in the hypothalamus and small intestine of female rats fed a daidzein diet.

We used the 3 meals per day feeding (MF3) method to determine whether daidzein-induced anorexia is directly mediated by gene expressions of appetite-mediated neuropeptides. The meal feeding method is effective at comparing levels of gene expression just before and just after ingestion of the test diet. In the past, the meal-feeding method has been utilized to compare changes in gene expression before and after a meal.^24–26) In the present study, we examined the dynamics of hypothalamic and gastrointestinal appetite-mediated gene expression when daidzein feeding suppressed food intake.

Materials and methods

Animals

Female Sprague-Dawley rats (Japan SLC, Hamamatsu, Japan) at 6-week olds were raised in stainless wire mesh cages in a room controlled by a 12 h light–dark cycle (dark phase: 15:00‒3:00) and constant temperature (23 ± 1 °C). They were housed separately for a week to acclimate to the environment. Animals were fed regular tap water and regular chow (trade name: MF, Oriental Yeast, Tokyo, Japan) ad libitum. This study was conducted in accordance with the ethical guidelines of the Ehime University Animal Experimentation Committee and was in complete compliance with the National Institutes of Health: Guide for the Care and Use of Laboratory Animals. All efforts were made to minimize the number of animals used and limit experimentation to what was necessary to produce reliable scientific information.

Diets

The control (C) diet contained (in g/kg): casein (New Zealand Dairy Board, Wellington, New Zealand), 200; cellulose (Danisco Japan., Inc, Tokyo, Japan), 50; soybean oil (J-oil Mills., Inc, Tokyo, Japan), 70; AIN-93 mineral mixture, 35; AIN-93 vitamin mixture, 10; l-cystine (Nacalai Tesque, Kyoto, Japan), 3; sucrose (Nippon Beet Sugar Manufacturing, Tokyo, Japan), 100; and α-corn starch (Nihon Shokuhin Kako, Tokyo, Japan), 532. The major aglycone isoflavone daidzein was purchased from the LC laboratories (MA, USA). The daidzein (D) diet was included at a ratio of 0.15 g/kg diet replacing α-corn starch.

Training protocols for MF3

All rats were given the C diet 3 times per day for 1 h each session (15:00‒16:00, 18:00‒19:00, and 21:00‒22:00). The MF3 schedule is shown in Fig. 1. Training on the MF3 schedule was completed when their food intake corresponded with the intake of free feeding rats for 14 days.

Fig. 1. Diagram for MF3 schedule.

Notes: Rats were given control (C) or daidzein (D) diet 3 times per day for 1 h each (15:00‒16:00, 18:00‒19:00, and 21:00‒22:00). The black zone shows the dark phase without food availability; the gray zones show food availability during the dark phase; the white zone shows the light phase without food availability.

Test protocols.In a test session, female rats were divided into two groups given free access to either the C or D diet on the MF3 schedule for 13 days. Their food intake was measured after each meal. The diet was replenished after measuring. The body weight of each rat was measured every morning. Thirteen days after the beginning of the test, each group was further subdivided into 6 groups (n = 6). The rats were killed by decapitation, and a blood sample corresponding to a non-fasting state was collected from the neck within 30 min before and after each feeding (14:30‒15:00, 16:00‒16:30, 17:30‒18:00, 19:00‒19:30, 20:30‒21:00, and 22:00‒22:30). The blood was stored at room temperature for at least 30 min before separating the serum by centrifugation at 3000 × g at 4 °C for 15 min, and then stored at −50 °C until analysis. In order to measure mRNA expression level, the hypothalamus and the mucosa of the upper small intestine was extracted from rats of all groups before and after each meal feeding. We isolated total RNA from fresh samples of the upper small intestine mucosa as previously described.²⁷⁾ The hypothalamus was excised on ice from the basal area of the brain after decapitation, soaked immediately in RNase inhibitor, and kept at 4 °C for a day. These samples were then stored at −80 °C until the assay. The perirenal and ovary fat tissues were removed and weighed; the sum of the weights of the right and left tissues was regarded as the total tissue weight. The carcass was stored at −20 °C until the analysis of body composition.

Body composition measurement

The carcass was minced. Total body water amount was determined by the difference in the mince weight before and after drying at 105 °C. Total body lipid amount was determined gravimetrically after extraction using the method of Folch et al.²⁸⁾ Total body protein amount was determined using the Kjeldahl method with an N-to-protein conversion factor of 6.25.²⁹⁾

Real-time monitoring polymerase chain reaction (PCR)

Total RNA was isolated from the hypothalamus as previously described.³⁰⁾ mRNA was isolated from total RNA using　Oligotex-dT30 (Takara Bio, Inc., Shiga, Japan). cDNA was synthesized using reverse transcriptase (Reverse Transcriptase XL[AMV]）for RT-PCR, 5 U/μL, Takara Bio Inc., Shiga, Japan) using a thermal cycler (ABI GeneAmp2400; PerkinElmer, Inc., Waltham, USA). Quantitative PCR to assay mRNA expression was carried out using a StepOnePlus™ real-time PCR system (Applied Biosystems, Carlsbad, Calfornia, USA). The sequences of primers were as follows; rat NPY: forward primer 5′-gcccagagcagagcaccc-3′ and reverse primer 5′-caagtttcatttcccatcacca-3′ (359 bp); rat galanin: forward primer 5′-ttggccacctcctcagaaga-3′ and reverse primer 5′-aatgactttaaattatagcagaggacac-3′ (84 bp); rat orexin: forward primer 5′-gctccagacaccatgaacct-3′ and reverse primer 5′-gaacacgtcttctggcga-3′ (146 bp); rat agouti-related protein (AGRP): forward primer 5′-gttcccaggtctaagtctgaa-3′ and reverse primer 5′-tgaagaagcggcagtagcac-3′ (205 bp); rat melanin-concentrating hormone (MCH): forward primer 5′-gttcccaggtctaagtctgaa-3′ and reverse primer 5′-tgaagaagcggcagtagcac-3′ (205 bp); rat cocaine and amphetamine-regulated transcript (CART): forward primer 5′-taaagtttgcgttcccccc-3′ and reverse primer 5′-cgaaagtccctcttcttccc-3′ (180 bp); rat proopiomelanocortin (POMC): forward primer 5′-gacaccaaaaccctcatca-3′ and reverse primer 5′-caaactcagaatggggtgaa-3′ (226 bp); rat long form of the leptin receptor (Ob-Rb): forward primer 5′-gattccacaaggggttcta-3′ and reverse primer 5′-ctggaggattctgatgtc-3′ (226 bp); rat CCK: forward primer 5′-cgcactgctagcccgataca-3′ and reverse primer 5′-tttctcattccgcctcctcc-3′ (216 bp); rat CRH: forward primer 5′-cgcagccgttgaatttcttg-3′ and reverse primer 5′-agcagcgggacttctgttga-3′ (116 bp); rat β-actin: forward primer 5′-ctatgagctgcctgacggtc-3′ and reverse primer 5′-agtttcatggatgcacagg-3′ (115 bp). The basic amplification program was set to perform 50 cycles of 15 s denaturizing at 95 °C, 1 min annealing, and primer extension at 60 °C. Fluorescence was recorded at 530 nm during extension. Relative mRNA expression was calculated using the crossing point of each target gene, that of β-actin gene as the reference, and the corresponding real-time PCR efficiency of respective primer sets by the method previously reported.³⁰⁾

Isoflavone levels measurement

Serum isoflavone levels were carried out as previously described.³¹⁾ Briefly, frozen plasma samples from rats were thawed and 100 μL aliquots were mixed with 100 μL of hydrolysis buffer (0.1 mol/L sodium acetate pH 5 with 0.1% (wt/vol) ascorbic acid and 0.01% (wt/vol) EDTA), 8 μL of glucuronidase, and 4 μL of sulfatase. The reaction mixture was allowed to hydrolyze to glucuronide and/or sulfate metabolites at 37 °C for at least 15 h. Subsequently, 10 μL of an internal standard (formononetin, 5 μg/mL in dimethyl sulfoxide), 120 μL of water, 75 μL of ammonium acetate buffer (1 mmol/L, pH 7), and 83 μL of triethylammonium sulfate buffer (3 mol/L, pH 7) were added, the samples were then heated to 60 °C for 10 min to facilitate the dissociation of isoflavones from plasma proteins; the mixture was then centrifuged. The deproteinized samples were passed over 0.5-g Sep-Pak C-18 cartridges (Nihon Waters, Tokyo, Japan) that had been previously washed with 5 mL of ammonium acetate buffer (10 mmol/L, pH 5) and 5 mL of water, at room temperature. The absorbed isoflavones were eluted with 1.5 mL of methanol. The methanol effluent was evaporated to dryness under a gentle stream of nitrogen at 45 °C, dissolved in 100 μL of (40:60 vol/vol) methanol/aqueous acetic acid (1%), and stored at −20 °C until HPLC analysis. A 30 μL aliquot of the sample was applied to a reverse phase HPLC column (4.6 × 150 mm, CAPCELL PAK C18, particle size 5 μm, Shiseido, Tokyo, Japan). The mobile phase was potassium phosphate buffer containing 40% of a mixture of methanol and acetonitrile (3:2, vol/vol) at 40 °C, and the flow rate was 1.0 mL/min. Daidzein, genistein, and equol were detected using an electrochemical detector (electrochemical detector 3005, Shiseido, Tokyo, Japan) under the following conditions: working electrode, glassy carbon; applied voltage, 800 mV. Formononetin was detected simultaneously using a UV-Visible detector (SPD-10AV, Shimadzu Corporation, Kyoto, Japan) at 254 nm.

Leptin, insulin, and glucose levels measurement

Leptin, insulin, and glucose levels in the serum samples were measured using commercial kits: rat leptin measurement kit (Morinaga Institute of Biological Science, Inc, Tokyo, Japan), rat insulin measurement kit (Morinaga Institute of Biological Science, Inc, Tokyo, Japan), and Glucose Assay Kit (Wako Pure Chemical Industries, Ltd).

Statistical analysis

Data are expressed as the mean ± standard error (SEM). For weekly food intake, the comparison was separately analyzed by one-way repeated measure analyses of variance (ANOVA) with Bonferroni post hoc significance testing. For data in Table 1, Figs. 3, 4, and 5, comparison was separately analyzed by three-way repeated measure ANOVA with Bonferroni post hoc significance testing. The comparisons between each group were made using an unpaired Student’s t-test with Bonferroni corrections. All statistical tests were done with IBM SPSS Statistics software (SPSS Japan Inc., an IBM company). Statistical significance was defined as p < 0.05.

Table 1. Effect of daidzein on serum leptin, insulin, glucose and isoflavones levels in rats.

				Before feeding				After feeding
				C		D		C		D
First feeding
Leptin			ng/mL	3.52 ± 0.63		3.26 ± 0.29		4.58 ± 0.34		8.69 ± 2.67
Insulin			ng/mL	1.55 ± 0.33		1.71 ± 0.33		1.47 ± 0.16		1.70 ± 0.36
Glucose			mM	15.22 ± 0.59		16.74 ± 0.91		14.32 ± 0.34		16.19 ± 0.80
Daidzein			nM	137.4 ± 26.3		187.0 ± 43.1		181.3 ± 26.1		198.0 ± 27.4
Equol			nM	ND		1303.3 ± 190.4*		3.4 ± 2.6		1057.4 ± 114.6*
Second feeding
Leptin			ng/mL	7.68 ± 1.02		8.49 ± 0.32		9.20 ± 1.45		9.76 ± 0.92
Insulin			ng/mL	1.36 ± 0.17		1.04 ± 0.13		1.76 ± 0.37		1.35 ± 0.18
Glucose			mM	15.67 ± 0.47		15.94 ± 0.49		15.86 ± 0.41		15.82 ± 0.60
Daidzein			nM	137.4 ± 16.3		213.6 ± 28.3*		150.6 ± 22.2		238.8 ± 14.7*
Equol			nM	ND		863.5 ± 66.6*		ND		828.8 ± 105.9*
Third feeding
Leptin			ng/mL	8.53 ± 1.18		7.33 ± 0.36		9.42 ± 2.78		7.16 ± 1.18
Insulin			ng/mL	1.44 ± 0.20		1.11 ± 0.15		1.39 ± 0.19		1.45 ± 0.18
Glucose			mM	14.87 ± 0.24		16.45 ± 0.56		14.22 ± 0.47		13.47 ± 0.66
Daidzein			nM	170.1 ± 25.1		221.7 ± 12.9		170.7 ± 18.2		305.4 ± 40.0*
Equol			nM	ND		917.3 ± 67.3*		ND		969.7 ± 75.3*

Each value represents the mean ± SEM, n=6. C, control group; D, daidzein group. Asterisks show a significant difference compared with the control group, determined by three-way repeated ANOVA with Bonferroni's test.

Results

Food intake, body weight gain, and adipose tissue weight after the MF3

Fig. 2 shows changes in food intake during the 13-day test session in rats. Rats fed the daidzein diet significantly decreased food intake on days 6, 9, 11, and 13 relative to rats fed the control diet (Fig. 2(A)). The daidzein-induced reduction of food intake began 6 days after initiation of the experiment. The anorectic effect was most consistently observed in the second feeding, but was also observed in the first and third feedings (Fig. 2(B)–(D). On day 6, food intake in group D was significantly lower than that observed in group C during the first feeding (Fig. 2(B)). From day 7 to 13, food intake in group D was consistently significantly or suggestively lower than that observed in group C during the second feeding (Fig. 2(C)). On day 13, food intake in group D was significantly lower than that observed in group C during the third feeding (Fig. 2(D)). The total food intake and total body weight gain over all 13 days was significantly lower in group D than in group C (p < 0.05, data not shown).

Fig. 2. Daily changes in the amount of food intake in female rats on the MF3 schedule.

Notes: Total food intake (A) and intake during the first (B), second (C), and third (D) feeding sessions are shown. Rats were provided with the control or 0.015% daidzein diets during the 13day experimental period on the MF3 schedule. Each value represents the mean ± standard error for 6 rats. The asterisks show a significant difference relative to the control group (p < 0.05), and the number signs show a trend relative to the control group (0.05 < p < 0.1), determined by two way repeated ANOVA with Bonferroni post hoc analysis. p values of main effects and interactions of diet (D) and feeding session (S) are obtained by two-way repeated ANOVA (D: p < 0.05; S: p < 0.001; D × S: NS).

Serum leptin, insulin, glucose, and isoflavone levels

Table 1 shows the effects of daidzein on serum leptin, insulin, glucose, and isoflavone levels. Groups C and D showed no significant differences in serum leptin, insulin, and glucose levels (Table 1). Serum daidzein and equol levels in group D were significantly higher than in group C (Table 1).

Appetite-mediated gene expression in the hypothalamus and small intestine

Figs. 3, 4, and 5 show dynamic shifts in appetite-mediated gene expression in the hypothalamus and upper small intestine before and after feeding for the first, second, and third feeding sessions. For the first feeding session, there was no significant difference in gene expression between C group and D group (three-way ANOVA with Bonferroni post hoc significance testing; Figs. 3, 4, and 5 and other data not shown). For the second feeding session, expression of NPY and galanin mRNA in the hypothalamus was significantly higher after feeding in group C, but no change was observed after feeding in group D, suggesting that daidzein attenuated the postprandial increase of NPY and galanin expressions (Fig. 3). Expression of CRH mRNA in the hypothalamus was not changed by feeding in group C rats during the second feeding session, while group D showed significant increases in CRH mRNA after feeding　(Fig. 4). The expression of CCK mRNA in the upper small intestine of group C rats was not changed by feeding, while group D showed significant increases in CCK mRNA after feeding. Further, CCK mRNA was significantly increased in group D rats relative to group C rats both before and after feeding (Fig. 5). These results suggest that daidzein promoted the expression of CRH in the hypothalamus and CCK in the upper small intestine. There were no significant differences in AGRP, MCH, POMC, CART, and Ob-Rb gene expression in the hypothalamus after the second feeding session (data not shown). For the third feeding, a feeding-induced increase in NPY mRNA was observed in both groups. Daidzein feeding attenuated the feeding-induced increase of galanin mRNA after the third feeding in a manner to that observed after the second feeding. There were no significant differences in expression of AGRP, MCH, POMC, CART, CRH, and Ob-Rb in the hypothalamus and CCK in the upper small intestine between diet groups for the third feeding (three-way ANOVA with Bonferroni post hoc significance testing; Figs. 4 and 5 and other data not shown).

Fig. 3. Changes in expression of NPY and galanin mRNA before and after each meal in the rat hypothalamus.

Notes: Rats were provided with the control or daidzein (150 mg/kg) diets during the 13-day experimental period. Each value represents the mean ± standard error for 6 rats. Asterisks show a significant difference relative to the control group, determined by three-way ANOVA with Bonferroni post hoc analysis. p values of main effects and interactions of diet (D), feeding session (S) and timing (before or after feeding; T) are obtained by three-way ANOVA (D: NS; S: p < 0.001; T: p < 0.001; D × S: NS; D × T: NS; S × T: p < 0.01; D × S × T: NS in expression of NPY mRNA, and D: p < 0.05; S: p < 0.001; T: p < 0.05; D × S: NS; D × T: p < 0.05; S × T: NS; D × S × T: NS in expression of galanin mRNA).

Fig. 4. Change in expression of corticotropin-releasing hormone (CRH) mRNA before and after each meal in the rat hypothalamus.

Notes: Rats were provided with the control or daidzein (150 mg/kg) diets during the 13-day experimental period. Each value represents the mean ± standard error for 6 rats. Asterisks show a significant difference relative to the control group, determined by three-way ANOVA with Bonferroni post hoc analysis. P values of main effects and interactions of diet (D), feeding session (S) and timing (before or after feeding; T) are obtained by three-way ANOVA (D: NS; S: NS; T: p < 0.01; D × S: NS; D × T: NS; S × T: NS; D × S × T: NS).

Fig. 5. Changes in expression of cholecystokinin (CCK) mRNA before and after each meal in the rat upper small intestine.

Notes: Rats were provided with the control or daidzein (150 mg/kg) diets during the 13-day experimental period. Each value represents the mean ± standard error for 6 rats. Asterisks show a significant difference compared with the control group, determined by three-way ANOVA with Bonferroni post hoc analysis. P values of main effects and interactions of diet (D), feeding session (S) and timing (before or after feeding; T) are obtained by three-way ANOVA (D: p < 0.01; S: p < 0.001; T: NS; D × S: NS; D × T: NS; S × T: NS; D × S × T: NS).

Discussion

In the present study, we investigated whether changes in the expression of genes involved in appetite were related to daidzein-induced anorexia using the MF3 method. Feeding in meals is effective at demonstrating shifts in gene expression before and after ingestion of test diet.^24–26) Using the MF3 method, daidzein-induced suppression of feeding was observed primarily during the second feeding, while the anorectic effect was reduced during the first and third feeding (Fig. 2(B)–(D). We also observed significant differences in appetite-mediated gene expression due to daidzein feeding during the second feeding session (Figs. 3–5). Here, we assessed the relation between dynamics of appetite-mediated gene expression and the daidzein-induced anorectic effect during the second feeding.

Interestingly, our findings suggest that daidzein attenuated a feeding-induced increase of NPY expression (Fig. 3) concomitantly with the suppression of food intake (Fig. 2). NPY induces potent hyperphagia, and subsequently causes obesity.^32,33) It was previously reported that 17β-estradiol treatment decreased NPY expression in the hypothalamus.^15,34) In addition, both ovariectomy and estrogen deficiency significantly increased NPY mRNA expression in the hypothalamus.^16,35) Titolo et al. reported that NPY gene expression was regulated by estrogen in hypothalamic neurons in a manner dependent upon the ratio of ERβ to ERα; and further that ERα signaling downregulated NPY expression.³⁶⁾ NPY is expressed mainly in the arcuate nucleus (ARC) and dorsomedial hypothalamus (DMH), but also in other hypothalamic areas such as the PVN and ventromedial nucleus (VMH). Ainslie et al. reported that ovariectomy increases NPY expression in the PVN.¹⁶⁾ Some reviews suggest that estrogen induces anorexia by regulating leptin and CCK signaling in the hypothalamus.¹⁰⁾ ARC NPY is under the control of circulating leptin, whereas DMH NPY is regulated by brain CCK and not by leptin.³⁷⁾ These findings underscore the necessity of an investigation of NPY expression in specific hypothalamic areas such as ARC, DMH, PVN, and VMH. In addition, further study is required to clarify whether estradiol treatment and daidzein feeding induce downregulation of NPY expression by the same mechanism.

In the present study, daidzein feeding significantly increased the expression of CCK mRNA in the upper small intestine before and after feeding during the second feeding session (Fig. 5). The effect of estradiol on an increase in intraduodenal intralipid-induced satiation disappeared after injection of a CCK1 receptor (CCK1R) antagonist.³⁸⁾ Satiation by CCK was abolished in ER-α knockout mice.³⁹⁾ Estradiol treatment also increased CCK-induced c-Fos expression in the brain of OVX rats.^20,21) In addition, estradiol treatment increased plasma levels of CCK in a dose-dependent manner, and inhibited gastric emptying and gastrointestinal transit in OVX rats by CCK stimulation and CCK1R activation.²³⁾ Further, intraperitoneal injection of CCK reduces NPY expression in the rat hypothalamus.⁴⁰⁾ As such, the daidzein-induced downregulation of NPY gene expression might be partially attributed to increases in CCK expression induced by the estrogenic property of daidzein. However, we have also observed that daidzein feeding and estradiol treatment significantly decreased food intake in both female Long-Evans Tokushima Otsuka and Otsuka Long-Evans Tokushima Fatty (CCK1R-deficient) rats, suggesting that CCK signaling is not essential for the anorectic effects of daidzein and estradiol.⁴¹⁾ In rats given food on the MF3 schedule as well as rats fed ad libitum, ¹¹⁾ daidzein feeding significantly decreased food intake 6 days after initiation of the experiment (Fig. 2(A)), leading us to speculate that dietary daidzein as it is not able to suppress food intake by directly inducing the expression of CCK mRNA　in intestinal endocrine cells. On the other hand, gene expression of CCK was not constitutively enhanced in upper small intestine by daidzein feeding (Fig. 5). There is the possibility that daidzein feeding may enhance eating relative expression of CCK mRNA through change in the expression of other genes. Although CCK gene expression is not necessarily linked to secretion,⁴²⁾ suppressive effect on food intake of CCK is associated with the amount of secreted it. Analyses of protein levels will be needed to determine whether the anorectic effect of daidzein is partially mediated by increases in CCK expression.

Consumption of the daidzein diet attenuated the postprandial increase in galanin mRNA expression (Fig. 3). It has been reported that galanin interacts with NPY in the hypothalamus. Immunostaining has shown that NPY and galanin contact at synapses in the hypothalamus and further that NPY-containing axons are observed in close proximity to galanin-containing cell bodies and dendrites.^43,44) Galanin is released in response to NPY administration,⁴⁴⁾ suggesting that NPY release provides excitatory input to galanin-containing neurons. It was reported that luteinizing hormone was secreted by an interaction between NPY and galanin.⁴⁵⁾ Our results suggest that daidzein feeding-induced increases in NPY expression might directly alter galanin expression.

We showed that daidzein feeding increased the expression of CRH mRNA in the hypothalamus (Fig. 4). Restricted feeding also increased CRH expression during the dark phase relative to the light phase in rats.⁴⁶⁾ The CRH receptor also colocalizes with NPY in the ARC.⁴⁷⁾ Estradiol treatment increases anorectic CRH gene expression in vivo and in vitro.^17,18) Although several potential mechanisms have been proposed, the mechanism of daidzein diet-induced increases in CRH gene expression has not been elucidated. As such, further studies will be needed.

Our results indicated that daidzein feeding did not significantly change the expression of AGRP, MCH, POMC, CART, and Ob-Rb in the hypothalamus (data not shown). Daidzein feeding also did not change serum leptin, insulin, and glucose levels (Table 1). This suggests that these peptides and hormones were not involved in the anorectic effect of daidzein feeding. It has previously been shown that estradiol treatment increases POMC gene expression.¹⁷⁾ However, daidzein feeding did not show any effect on POMC gene expression. Although both daidzein and estradiol reduce food intake, daidzein-induced anorexia may have a different mechanism.

It has been reported that equol, a metabolite of the daidzein, also induces anorexia. Rachon et al. reported that a 400 mg/kg diet of dietary equol decreased food intake and body fat in OVX rats.⁴⁸⁾ Wu et al. reported that the preventive effects of isoflavones on fat accumulation in early postmenopausal women depend on an individual’s capacity to produce equol.⁴⁹⁾ The present (Table 1) and previous¹¹⁾ studies showed that the serum level of equol was much higher than that of daidzein in rats fed daidzein diet. These results indicate that equol may be bioactive substance of anorectic action. We speculate that the duration of daidzein-induced anorexia (Fig. 2(A)) may also be associated with the production and internal dynamics of equol. Unexpectedly, in the present study, equol concentrations both before and after the second feeding were lower relative to the first and third feedings (Table 1). In this experiment, it is not clear why consumption of the daidzein diet reduced equol during the second feeding only, or why the daidzein-induced anorectic effect was weaker during the first and third feeding.

Our results suggests that daidzein-induced alterations in the dynamics of NPY, galanin, and CRH expression in the hypothalamus, and CCK expression in the upper small intestine, may play a key role in the anorectic effect of daidzein, although we have previously demonstrated that CCK signaling via CCK1R is not essential for the anorectic effects of daidzein.

Acknowledgments

mRNA expression measurements were performed at the Integrated Center for Science, Tarumi, Ehime University.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This research was supported by the Uehara Memorial Foundation, the Fuji Foundation for Protein Research, and a Grant-in-Aid for Scientific research from the Ministry of Education, Science, Sports, and Culture of Japan.