Hepatic fatty acid biosynthesis in KK‐Ay mice is modulated by administration of persimmon peel extract: A DNA microarray study

Abstract Scope Previously, we showed that the intake of a persimmon peel (PP) extract altered hepatic gene expression associated with the insulin signaling pathway and enhanced tyrosine phosphorylation of insulin receptors in nonobese type 2 diabetic Goto‐Kakizaki rats. Our objective was to evaluate the effect of fat‐soluble PP extract on obese type 2 diabetic KK‐Ay mice with insulin resistance. Methods and results KK‐Ay mice were fed a diet mixed with 0.1% of the extract for 8 weeks. The total ketone body levels in the plasma of PP extract‐fed mice were significantly lower than those in the normal diet‐fed mice. Hepatic nonesterified palmitic acid content was higher in the PP extract‐fed mice than in normal diet‐fed mice. The hepatic gene expression profiles of the treated mice indicated upregulation of fatty acid synthesis and downregulation of inflammation‐associated genes, predicting SREBP‐1c and PPARγ activation. Conclusion These results suggest that the PP extract enhances hepatic fatty acid synthesis via SREBP‐1c and PPARγ, as well as anti‐inflammatory activity in KK‐Ay mice.

. High amounts of carotenoids, polyphenols, and vitamins (Gorinstein et al., 2001) are present in the persimmon peel (PP); however, mostly, the peel is wasted because its benefits have not been thoroughly explored.
For the effective use of PP, we prepared a fat-soluble PP extract (Izuchi, Takahashi, & Inada, 2009) and administered it to Goto-Kakizaki (GK) rats, a nonobese type 2 diabetes model (Izuchi et al., 2011). Our previous study showed that GK rats fed a PP extract showed altered hepatic gene expression associated with the insulin signaling pathway. Moreover, hepatic tyrosine phosphorylation of the insulin receptor beta-subunit was enhanced by the intake of PP extract. These results suggest that PP extract has the potential to reduce insulin resistance in GK rats.
Abnormal insulin secretion, rather than insulin resistance, is a major cause for the development of diabetes in GK rats (Kimura et al., 1982). Improvement of insulin sensitivity in GK rats has little effect on alleviation of hyperglycemic state. However, the major causes of insulin resistance in KK-A y mice are excessive eating as well as a genetic background of the development in obesity-induced diabetes (Iwatsuka, Shino, & Suzuoki, 1970). Owing to these reasons, they are suitable experimental animals for determining the effects of PP extract on insulin resistance of obesity-induced type 2 diabetes mellitus. To investigate the effect of PP extract on obesity-induced diabetes, we used KK-A y mice fed a diet containing PP extract in the present study.

| Preparation of PP extract
The PP extract was prepared as described previously (Izuchi et al., 2009). In brief, dried PP powder was extracted with ethanol; then, a methyl tert-butyl ether-soluble fraction of the extract was collected and evaporated. Chemical components present in PP extract are shown in Supporting Information Table S1.

| Experimental animals and diets
Five-week-old male KK-A y /TaJcl mice were purchased from CREA Japan (Tokyo, Japan). They were housed individually in plastic cages and maintained at a temperature of 22 ± 1°C, under a 12-hr light/dark cycle (lights on from 08:00 to 20:00 daily). They were fed a commercial diet (AIN-93G; Oriental Yeast, Tokyo, Japan) for a week and then divided into two groups with similar average body weight: a control diet group (CD, n = 7) fed a commercial diet and a PP extract diet group (PD, n = 6) fed a commercial diet containing 1 g/kg of PP extract. The mice were allowed free access to food and drinking water.
After 8 weeks, each mouse was fasted for 3 hr and then anesthetized intraperitoneally with pentobarbital. Blood samples were collected from the carotid artery, treated with heparin, and centrifuged at 800×g. The supernatants were collected and stored at −20°C until use. Livers were excised and placed in RNAlater ® (Ambion ® , Austin, TX, USA) for subsequent RNA analysis. The protocol for the animal experiments was approved by the Animal Use Committee of the Faculty of Agriculture at The University of Tokyo (P10-403).

| Measurement of plasma biochemical parameters
Aspartate aminotransferase, alanine aminotransferase, total cholesterol, triacylglycerol, nonesterified fatty acid, LDL-and HDLcholesterol, glucose, total ketone bodies, and glycoalbumin levels in plasma were analyzed using 7180 clinical analyzer (Hitachi High-Technologies, Tokyo, Japan). Commercial ELISA kits were used to measure the concentrations of plasma adiponectin (Otsuka Pharmaceutical Co. Ltd, Tokyo, Japan), insulin, leptin (Morinaga Institute of Biological Science Inc., Yokohama, Japan).

| Separation and quantitative determination of nonesterified fatty acids
The lipid extracts from livers were mixed with an appropriate amount of undecanoic acid as an internal standard and then TA B L E 1 Plasma biochemical parameters and hepatic nonesterified fatty acid levels in KK-A y mice administered with PP extract for 8 weeks separated by silica gel TLC. A part of the nonesterified fatty acids were methyl-esterified with boron trifluoride-methanol complex (14% in methanol). The reaction solutions were mixed with hexane and saturated saline. The hexane layers were filtered (pore size 0.45 μm, ADVANTEC, Tokyo, Japan) and analyzed by GC-TOF MS.
Deal conditions are shown in Supporting Information.
The concentration of each fatty acid in the lipid extract was determined by comparing the values of peak areas in total ion chromatogram with those in methyl-esterified fatty acid standards: palmitic, stearic, oleic, and linoleic acids.

| DNA microarray experiments and data analysis
The livers of four mice from each group were subjected to DNA microarray analysis based on the levels of plasma total ketone bodies close to the average value. Total RNA was isolated from the liver using TRIzol ® (Life Technologies, Carlsbad, CA,

USA) reagent and purified using an RNeasy mini kit (QIAGEN,
Hilden, Germany). The quality and quantity of the purified total RNA were checked by agarose gel electrophoresis and spectrophotometry, respectively. DNA microarray was performed using Affymetrix GeneChip ® system. Details are described in Supporting Information.
Gene-annotation enrichment analysis of the differentially expressed genes (DEGs, a false discovery rate <0.

| Statistical analysis
Each value was expressed as mean ± SEM. Differences between the groups were calculated using an unpaired Student's t test and a p < 0.05 was considered statistically significant.
F I G U R E 1 Venn and Euler diagrams showing the relationship between significantly enriched Gene Ontology (GO) terms in differentially expressed genes (DEGs) following administration of persimmon peel extract in KK-A y mice. The DEGs included in the GO terms are indicated as gene symbols. Left side of the dashed line shows downregulated genes and the right side shows upregulated genes. "Metabolic process" (dot-dashed line), "response to stimulus" (dot-dot-dashed line), and "development process" (dotted line) show the common ancestors of the enriched GO terms surrounded by the respective lines

| Effects of PP extract administration on plasma biochemical parameters and hepatic fatty acids
We evaluated the effects of PP extract administration on plasma biochemical parameters and nonesterified fatty acids of KK-A y mice (Table 1). Plasma total ketone bodies were significantly lower in PD than in CD mice (p = 0.014). The detected fatty acids mostly belonged to four types: palmitic acid (C16:0), stearic acid (C18:0), oleic acid (C18:1 [n-9]), and linoleic acid (C18:2 [n-6]). Liver palmitic acid content was significantly higher in PD than in CD mice (p = 0.027).
The other fatty acid levels tended to be higher in PD than in CD mice.

| Alterations of hepatic gene expression profile by PP extract and identification of enriched GO terms in DEGs
It has been suggested that a decrease in total ketone body levels and increase in hepatic palmitic acid of PD mice are caused by an alteration of hepatic fatty acid metabolism. Accordingly, we evaluated the effect of PP extract administration on hepatic lipid metabolism using a DNA microarray. DEGs (Supporting Information  Information Tables S5 and S6) and DEGs contained in each GO term are shown in Figure 1.

| Effects of PP extract administration on energy metabolism
We used KK-A y mice to evaluate the effects of PP extract administration on obesity-induced diabetes and observed that the total ketone body levels in the plasma of KK-A y mice were decreased by the administration of PP extract. In diabetes mellitus, reduction of glucose uptake into the cells owing to deficiency of insulin signal enhances β-oxidation of fatty acids and results in further generation of ketone bodies as a byproduct. In contrast, hepatic nonesterified palmitic acid levels in KK-A y mice were significantly increased by the administration of PP extract. It is suggested that fatty acid synthesis is promoted by PP extract because most of the de novo synthesized fatty acid is palmitic acid (Kuhajda et al., 1994).
In addition to the increase in nonesterified palmitic acid in the liver, detailed pathway analysis suggested the enhancement of fatty acid biosynthesis in PD mice, such as upregulation of Fasn and Acacb ( Figure 2), genes encoding the key enzymes for fatty acid synthesis (Wakil, Stoops, & Joshi, 1983). Moreover, a decrease in plasma total ketone bodies can be explained by the upregulation of Aacs.
Aacs, encoding the ketone body-utilizing enzyme AACS (Endemann, Goetz, Edmond, & Brunengraber, 1982), which converts acetoacetate derived from ketone bodies to acetoacetyl-CoA and then promotes a reaction in the direction of acetyl-CoA production.
The activation of hepatic PPARγ and SREBP-1 and 2 by PDadministered mice was predicted by IPA. SREBP-1c is an important factor in controlling blood glucose and regulating the expression of glycolysis-associated genes and fatty acid biosynthesis (Shimano et al., 1997). We infer that SREBP-1c activated in the liver of PD mice upregulated the expression of Fasn and Acc and fatty acid synthesis.
Peroxisome proliferator-activated receptor gamma is a nuclear receptor associated with adipocyte differentiation and fat accumulation. Therefore, activation of PPARγ is a possible way to improve insulin resistance and hyperglycemia. Activation of PPARγ induces the upregulation of genes associated with lipogenesis (Way et al., 2001),

TA B L E 2 Predicted upstream regulators of DEGs by PP extract administration
Upstream transcription factors z-Score

Regulated genes by PP extract Predicted upregulation genes Predicted downregulation genes
Predicted activation factor SREBP-2 2.552 Aacs a , Acly, Cdkn1a b , Dhcr7, Fabp5, Fads2, Fasn, Fdps and decreases the formation of ketone bodies (Fujiwara, Yoshioka, Yoshioka, Ushiyama, & Horikoshi, 1988). PPARγ activated by PP extract can also promote fatty acid synthesis. However, it was presumed that the PPARγ activity of PP extract was low, and did not contribute to the improvement of insulin resistance and hyperglycemia.

| Effects of PP extract administration on inflammatory responses
The present study revealed that the expression of genes associated with inflammation such as Il1r1, Il6ra, and Saa 1, 2, and 4 was downregulated in the livers of PD mice (Supporting Information Table S4).
Inflammatory cytokines, including IL-1 and IL-6, are released from macrophages and adipocytes and lead to the pathogenesis of insulin resistance and development of type 2 diabetes mellitus (Hotamisligil, 2006). These cytokines induce the expression of acute-phase inflammatory markers, C-reactive protein (CRP), and SAA in the liver (Steel & Whitehead, 1994).
Inhibition of STAT3 and NF-κB signals, which are upstream activators of Il6 and Il1 (Libermann & Baltimore, 1990), was predicted by IPA. Obese diabetic KK-A y mice display obesity-induced chronic inflammation (Kato et al., 2001). We assumed that the reduction in reactive oxygen species (ROS) caused by the administration of PP

| Effect of PP extract on animal models of diabetes mellitus
In our previous study, the GK rats administered with PP extract indicated enhanced tyrosine phosphorylation of insulin receptor β-subunit and upregulation of Srebp-1c gene expression in the liver at mRNA level (Izuchi et al., 2011). These results suggested that the upregulation of Srebp-1c gene expression in the liver of PP extract-administered GK rats was induced by the enhancement of insulin receptor signaling, and consequently the expression of the genes involved in fatty acid synthesis, Fasn, Acaca, and Acacb, was upregulated. These results were consistent with the hepatic gene expression profile of the extract-administered KK-Ay mice. The functional characteristics of hepatic DEGs in GK rats administered PP extract showed upregulation of the genes related to glycolysis and fatty acid synthesis, and downregulation of the genes related to β-oxidation and gluconeogenesis. We expect that the effects of the extract on hepatic gene expression are induced by PPARγ activation, because the alterations were similar to the hepatic gene expression profile displayed in the presence of a PPARγ agonist (Way et al., 2001). This hepatic gene expression is probably caused by quercetin and ursolic acid present in PP extract, which are potential PPARγ agonists (Wang et al., 2014;Way et al., 2001).
Moreover, our previous study showed that GK rats administered with the extract reduced plasma ALT levels and downregulated the expression of hepatic Il1r and Crp genes with inflammation. In the present study, KK-Ay mice administered PP extract also showed a gene expression pattern similar to that of the GK rats. Fat-soluble antioxidants such as β-cryptoxanthin, quercetin, and vitamin E in the extract could act as anti-inflammatory agents and contribute to the improvement of insulin resistance.
In conclusion, we propose that PP extract promotes insulin receptor activity through multiple mechanisms, including the activation of PPARγ and anti-inflammatory effects, through which it alters fatty acid synthesis through SREBP-1c and PPARγ. PP extract administration altered hepatic gene expression in both obese and nonobese diabetic animals, but its effect was limited due to its very low activity, which may not overcome abnormal insulin secretion and excessive eating. Therefore, the extract could contribute to support improvement in insulin resistance alongside diet restriction and medication use.

CO N FLI C T O F I NTE R E S T
No potential conflict of interest was reported by the authors.

E TH I C A L S TATEM ENTS
These animal experiments were reviewed and approved by the Animal Use Committee of the Faculty of Agriculture at The University of Tokyo (Japan).