The regulation of HBP1, SIRT1, and SREBP-1c genes and the related microRNAs in non-alcoholic fatty liver rats: The association with the folic acid anti-steatosis.

Folic acid is one of the vital micronutrients that contribute to the genetic stability and other biological activities. In addition, microRNAs regulate gene expression through a multittude of pathways. Our current work aimd to explore the possible ameliorative potency of folic acid and its association with the hepatic miR-21, -34a, and -122 expression as well as their targeted genes, HBP1, SIRT1, and SREBP-1c in rats with non-alcoholic fatty liver disease (NAFL). A total of 50 Wistar rats were randomly divided into two groups, a control group (n = 10) and NAFL group (n = 40). Rats in NAFL group were fed a high-fat diet (HFD) containing 20% fats for 14 weeks. The NAFL group was further subdivided into four groups (n = 10/group), one untreated and three orally folic acid-treated groups (25, 50, and 75 μg/Kg b.wt). NAFL characteristics was evaluated in rats in addition to the miR-21, -34a, and -122 profile as well as the transcriptional levels of HBP1, SIRT1, and SREBP-1c genes. NAFL rats exhibited the classic traits of fatty liver disease profile and dysregulation in the pattern of miR-21, -34a, and -122 expression as well as their targeted genes (HBP1, SIRT1, and SREBP-1c, respectively) in the liver. Additionally, NAFL rats had altered levels of TNF-α and adiponectin. These alterations were significantly ameliorated in a dose-dependent pattern following the folic acid treatments. In conclusions, the anti-steatotic, insulin-sensitizing, glucose-lowering and lipotropic potencies of folic acid in NAFL rats may be linked to the epigenetic modulation of the hepatic microRNAs (miR-21, -34a, and -122) and the expression of their target genes (HBP1, SIRT1, and SREBP-1c).

Introduction

Non-alcoholic fatty liver (NAFL) disease is a common hepatic disease characterized by the ectopic deposition of lipids in hepatocytes in the absence of excessive consumption of alcohol. The excessive accumulation of hepatic fats occur due to the elevated triglycerides (TG) delivery to the liver or enhanced de novo lipogenesis. Although NAFL is a mild condition, it may progress to non-alcoholic steatohepatitis (NASH), and in severe cases may lead to liver failure, permanent cirrhosis and hepatocellular carcinoma [1]. Dyslipidemia, arterial hypertension, insulin resistance and central obesity are all features of metabolic syndrome. Data strongly suggest that NAFL is the hepatic manifestation of metabolic syndrome. Moreover, insulin resistance is considered the major cause of hepatic steatosis and the first hit in the development of NAFL [2].

The small non-coding RNAs, 20–22 nucleotides, are microRNAs (miRNAs or miRs) and act by inhibiting the translation or promote degradation of the messenger RNA (mRNA) of their target genes [3]. They are pleiotropic regulators of most cellular pathways, including the pathways of NAFL pathogenesis [4]. Many miRNAs are involved in the development of NAFL such as miR-122, the most frequent liver-specific miRNA. The expression of about 24 hepatocyte-specific genes are either directly or indirectly affected by miR-122 levels [5]. It can influence the metabolism of lipids via targeting sterol-regulatory element-binding proteins-1c (SREBP-1c), SREBP-2, hydroxymethylglutaryl-CoA reductase and fatty acid synthase (FAS). During the NAFL development, the hepatic expression of miR-122 is diminished [6]. MiR-34a is the other miRNA that is mostly linked to metabolic impairment in the liver. Its expression was enhanced in NASH and type 2 diabetes mellitus (T2DM) [7], and associated with the severity of NASH [8]. Sirtuin 1 (SIRT-1), an NAD+-dependent enzyme involved in gene silencing by de-acetylating both histone and non-histone targets, is the main target of miR-34a [9]. NAFL development is associated with enhanced hepatic expression of miR21, one of the most abundant miRNAs in the liver [10]. The proliferation of hepatocyte and lipid metabolism are regulated by MiR-21 as a result of targeting the pathway of a high mobility group (HMG) Box Transcription Factor 1 (HBP1)-p53-SREBP1c [11].

There are no drugs approved specifically for NAFL treatment. The available thrtapeutic approaches strive to enhance the metabolic status through weight loss and exercise, reducing insulin resistance and controlling other feaures of metabolic syndrome.

The bioactive metabolites of natural sources and the whole extracts of various natural origins were reported to possess vital processes such as chemoprotection [12-14], anti-genotoxicity [15], and anticancer potencies [16]. Folic acid, a B-class vitamin member, is essential for purines and pyrimidines synthesis, and for crucial methylation of biological substances, including DNA, phospholipids, and neurotransmitters [16]. The water-soluble synthetic folate is a donor of carbon atom in serine from glycine, purines/pyrimidine bases, and methylcobalamin synthesis [17]. Moreover, it plays a role in homocysteine re-methylation to methionine for the synthesis of thymine and, for DNA methylation via SAM [18]. The decline in folic acid causes mitochondrial and nuclear DNA damage in addition to inhibited regeneration of normal tissue [19].

Although the therapeutic effects of folic acid against a rat model of high-fat diet-induced NAFL was previously investigated, to the best of our knowledge, this is the first report linking the probable molecular effect of folic acid with the the expression of mature miR-21, -34a, and -122 as well as their targeted genes HBP1, SIRT1, and SREBP-1c. This may introduce promising therapeutic approaches in managing and treating NAFL by targeting distinct pathways implicated in its onset and progression.

Materials and methods

Animals

Animal management procedures were undertaken in accordance with the requirements of the Institutional Animal Care and Use Committee (IACUC), Alexandria University, Egypt. The protocol of this study has been approved by the ethics review board of the IACUC of Medical Research Institute, Alexandria University (ID: AU0122172833). The experiments in this study were in compliance with the ARRIVE guidelines.

Healthy male Wistar rats (50 rats, ~ 150 g & 8 weeks old), were obtained from the Medical Technology Center, Medical Research Institute, Alexandria University, Egypt. The housing of rats (5 per cage) was done at an ambient temperature of 23 °C in a 12-h light/dark cycle. Free access to water and a commercial chow diet (El-Haramain® standard rodent diet, Egypt) for two weeks before the experiment was supplied.

Drugs

Folic acid (Nile Pharmaceutical and Chemical Industries Company, Alexandria, Egypt) was dissolved in a saline solution to prepare 25, 50, and 75 μg/ml stock solutions.

Experimental design

Rats were grouped into a control group (n = 10) and the non-alcoholic fatty liver (NAFL) group, (n = 40). The control rodent were given a standard commercial rats chow containg 7% simple sugars, 3% fat, 50% polysaccharide, 15% protein (w/w), and energy 3.5 Kcal/g. On the other hand, the NAFL rats were given a palatable high-fat diet (HFD) for 14 weeks [20], which was prepared by grounding the sweet biscuits, milk chocolate, and peanuts to the commercial rodent chow in the ratio of 1:2:2:3 [21]. The prepared HFD achieved 20% protein, 20% fat, 48% carbohydrate, and energy 5.15 Kcal/g (35% of calories as fat). After 14 weeks of NAFL induction, rats were divided into four groups (n = 10), one of them was the untreated group, and the other 3 groups received daily oral treatment of 25, 50, or 75 μg/Kg b.wt of folic acid once per day for a month.

Sample collection

At the end of the treatment period, overnight fasting rats were weighed and sacrificed by cervical dislocation. The obtaining of blood samples was done by cardiac puncture and centrifuged at 3000 ×g, 4°C for 20 minutes to separate sera for the assessment of glucose, insulin, bilirubin, lipid profile, alanine aminotransferase (ALT), and aspartate aminotransferase (AST). Livers of treated and control animals were washed using ice-cold saline quickly, dried, and weighed to calculate the relative liver weight (= 100 × Liver weight/Body weight)). Four parts of each liver were prepared, one part (0.5 g) was homogenized in phosphate-buffered saline (1:9) for subsequent evaluation of adiponectin, tumor necrosis factor-α (TNF-α), and total protein, the second part was used for the determination of hepatic triglycerides content, the third part was used for the molecular studies and the last part was used in the histological study.

Biochemical assays

The commercial kits (Randox, Antrim, UK) were used for blood glucose level, triglycerides, cholesterol, low-density lipoprotein-cholesterol (LDL-C), high-density lipoprotein-cholesterol (HDL-C), ALT, AST, and total bilirubin assessment. The Enzyme Immunolinked Assay (EIA) kit (Mercodia, Uppsala, Sweden) was used to determine insulin levels. The homeostasis model assessment (HOMA), a method of insulin resistance evaluation [22], was used to investigate the index of insulin resistance (IRI) as follows:

Hepatic triglycerides content

The lipids content was first extracted from the hepatic tissues following the method of Bligh and Dyer [23]. Lipids-containing chloroformic layer was used to assess triglycerides level using a routine kit (Randox, Antrim, UK).

ELISA assay of TNF-α and adiponectin

Enzyme-linked immunosorbent assay kits (eBioscience, California, USA) were used to evaluate serum levels of TNF-α and adiponectin according to the manufacturer’s instructions.

The expression of miRNAs and their target genes

For the extraction of total RNA, thirty milligrams of liver tissues were processed following the instructions of miRNeasy kit (Qiagen, Hilden, Germany) manufacturer. Further, the reverse transcription of extracted RNA was done using miRCURY LNA RT Kit (Qiagen, Hilden, Germany).

The expression of microRNAs in liver

The quantification of mature miRNAs (miR-21, miR-34a, and miR-122) was performed using the obtained cDNAs by miRCURY LNA miRNA PCR assay and SYBR Green PCR kit. As a house-keeping gene, U6 was used. Quantitative PCR reactions began at 95°C for 10 minutes as an initial denaturation and amplification via 45 PCR cycles as follows: 95°C for 5 seconds, 55°C for 15 seconds for annealing, and then 15 seconds at 60°C for the extension. Data collection was done using Rotor-Gene Q-Pure Detection version 2.1.0 (Qiagen, Maryland, USA). The relative miRNAs expression was computed relative to the expression of U6 of the same sample by standardizing the values of threshold cycles (Ct) of target miRNAs to that of U6 using the method of 2-ΔΔCt.

The expression of HBP1, SIRT1, and SREBP-1c genes in liver

The gene expression of HBP1, SIRT1, and SREBP-1c was also quantified in the cDNA by Rotor-Gene qRT-PCR (Qiagen, Maryland, USA) using QuantiTect SYBR Green PCR Master Mix (Qiagen, Hilden, Germany). The conditions of PCR amplification were adjusted as a denaturation for 10 minutes at 95°C and then 40 cycles of PCR for amplification as follows: at 95°C for 5 seconds as a denaturation, 15 seconds at 55°C for annealing, and 15 seconds for extension at 60°C. The house-keeping gene, GAPDH, was used for normalization. The used primers are presented in Table 1. The values of Ct were determined using Rotor-Gene Q-Pure Detection version 2.1.0 (Qiagen, Valencia, CA, USA). For each studied gene, the relative alteration in samples was estimated using the method of 2-ΔΔCt.

Table 1

PCR primers sequences.

Gene	Accession No.	Primer Sequence
HBP1	NM_013221.2	F: AGTTGCTGCAGTGTAATGAGAATTTG
		R: GGTGAGTATTTTCCGGTATATCTGAGG
SIRT1	NM_001372090.1	F: TGTTTCCTGTGGGATACCTGA
		R: TGAAGAATGGTCTTGGGTCTTT
SREBP-1c	NM_001276708.1	F: GACGACGGAGCCATGGATT
		R: GGAAGTCACTGTCTTGGTTGTT’
GAPDH	NM_017008.4	F: GGGTGTGAACCACGAGAAATA
		R: AGTTGTCATGGATGACCTTGG

Histological examination

Liver tissues obtained from rats were immediately fixed in 10% formalin before being treated with a conventional grade of alcohol and xylol, embedded in paraffin and sectioned at 4–6 μm thickness. The sections were stained with Haematoxylin and Eosin (H&E) stain in order to study the histopathological changes.

Statistical analysis

All data were stated as mean ± SD. The Kolmogorov-Smirnov test was done to validate the normal distribution of variables. A one-way analysis of variance (ANOVA) followed by the Tukey-Post Hoc test was applied for the comparison between the mean values of different groups. Differences at P < 0.05 reflected the presence of significance. SPSS statistical software version 18 (IBM, Chicago, IL, USA) was used software for all data.

Results

Effect of folic acid on body and liver weights

Results revealed that folic acid-treated NAFL rats exhibited a decrease in the weights of liver and body dose-dependently, as well as the relative liver weights. All NAFL rats were significantly heavier and had a higher index of liver weights (14.17±0.09 g) and relative liver weights of 3.65±0.03 g compared with control rats (5.95±0.11 g and 2.82±0.05 g, respectively). However, folic acid-treated rats recorded a marked dose-dependent decline in the body (~ 21.5%), liver (~ 35.8%) weights and relative liver weights by ~ 22.5% following the 75 μg/Kg dosing (Table 2).

Table 2

Body and liver weights, relative liver weight, and serum liver functions.

	Control	NAFL rats
		Untreated	Folic acid treatments (μg/Kg)
			25	50	75
Body weight (g)	198.5± 3.54	338.3±8.03a	332.4±6.74a	325.2±8.38a	300.2±4.89a,b,c,d
Liver weight (g)	5.95± 0.11	14.17±0.09a	11.97±0.22a,b	10.59±0.28a,b,c	9.10±0.14a,b,c,d
Relative liver (%) weight (g)	2.82±0.05	3.65±0.03a	3.25±0.04a,b	3.00±0.06a,b,c	2.83±0.03b,c,d
ALT (U/l)	31.4 ± 4.7	63.0a ± 12.8	55.8a ± 8.2	37.2bc ± 5.1	32.6bc ± 5.3
AST (U/l)	111.0 ± 13.7	175.5a ± 16.0	146.2ab ± 13.8	124.6bc ± 8.8	119.7bc ± 10.1
Bilirubin (mg/dl)	0.46 ± 0.14	1.28a ± 0.31	0.93ab ± 0.11	0.89ab ± 0.06	0.68abc ± 0.10

Data were shown as mean ± SD, (n = 10). NAFL, non-alcoholic fatty liver rats; ALT, alanine aminotransferase; AST, aspartate aminotransferase.

a, significant vs. control rats;

b, significant vs. untreated NAFL rats;

c. significant vs. NAFL rats treated with folic acid 25 μg/Kg;

d, significant vs. NAFL rats treated with folic acid 50 μg/Kg;

e, significant vs. NAFL rats treated with folic acid 75 μg/Kg; (p ≤ 0.05).

Effect of folic acid on serum liver functions

Dose-dependent changes were observed in the ALT and AST serum activities and bilirubin levels similar to the effects on liver weight and index with an improvement following folic acid treatments. The NAFL rats showed a significant increase in serum ALT, AST, and bilirubin levels compared to controls. However, the treatment with folic acid revealed a significant dose-dependent decline in these levels which completely normalized to the highest dose (75 μg/Kg), and the changes were ~ 48, 31.7 & 47% for ALT, AST, and bilirubin respectively (Table 2).

Effect of folic acid on glucose homeostasis parameters

In order to assess glucose homeostasis, fasting blood glucose (FBG), insulin, and HOMA-insulin resistance (HOMA-IR) were investigated in control and treated rats (Table 3). The NAFL rats had significantly higher levels of FBG, insulin and HOMA-IR (164 mg/dl, 15.48 mIU/ml and 6.29) with respect to control rats (75.40 mg/dl, 7.33 mIU/ml and 1.36) respectively. The folic acid-treated NAFL rats at the lowest dose (25 μg/Kg) showed a non-significant effect compared with the untreated NAFL rats. However, the higher doses of folic acid (50 and 75 μg/Kg) reduced (p ≤ 0.05) the FBG, insulin and insulin resistance index respecting the untreated NAFL group. Furthermore, 75 μg/Kg treatment of folic acid had a significant effect on FBG which were changed from 164±8.77 to 143.1±7.78, and HOMA-IR from 6.29±1.07 to 4.80±0.44 compared with the treatment with the lower folic acid dosing (25 μg/Kg).

Table 3

Glucose homeostasis parameters, serum lipid profile and hepatic triglycerides content.

	Control	NAFL rats
		Untreated	Folic acid treatment (μg/Kg)
			25	50	75
FBS (mg/dl)	75.40±5.95	164.0a± 8.77	161.9a ±8.09	143.8abc ±4.78	143.1abc ±7.78
Insulin (mIU/ml)	7.33±1.81	15.48a±2.08	14.34a ±0.71	13.86a ±0.88	13.65ab ±1.07
HOMA-IR	1.36±0.30	6.29a±1.07	5.74a ±0.45	4.92abc ±0.41	4.80abc ±0.44
TG (mg/dl)	36.80±6.0	63.70a±9.76	53.90a ±3.84	46.40b ±9.47	40.40bc ±11.55
Cholesterol (mg/dl)	135.1±11.68	182.30a ±17.61	155.4ab ±8.22	140.30bc ±6.40	136.8bc ±9.37
HDL-C (mg/dl)	51.20 ±4.85	30.76a ±2.26	36.50ab ±4.53	46.60bc ±5.54	48.32bc ±3.73
LDL-C (mg/dl)	76.54 ±11.24	138.8a ±18.86	108.1ab ±8.63	84.42bc ±5.80	80.40bc ±7.51
Hepatic TG (mg/g tissue)	34.20 ±3.98	107.8a ±3.33	92.81ab ±9.01	70.42abc ±8.92	56.43abcd ±6.97

Data were shown as mean ± SD, (n = 10). NAFL, non-alcoholic fatty liver; FBS, fasting blood sugar; TG, triglycerides; HDL-C, high-density lipoprotein-cholesterol; LDL-C, low-density lipoprotein-cholesterol.

a, significant vs. control rats;

b, significant vs. untreated NAFL rats;

c, significant vs. NAFL rats treated with folic acid 25 μg/Kg;

d, significant vs. NAFL rats treated with folic acid 50 μg/Kg;

e, significant vs. NAFL rats treated with folic acid 75 μg/Kg, (p ≤ 0.05).

Effect of folic acid on serum lipid profile

Results of serum lipid profile indicated that folic acid normalized the lipid profile (TG, TC and LDL-C) of NAFL rats (Table 2). NAFL rats showed a significant elevation in serum levels of TG, TC, and LDL-C with respect to the control rats with an elevation from 36.80±6.0, 135.1±11.68 and 76.54±11.24 mg/dl to 63.70±9.76, 182.30±17.61 and 138.8±18.86 mg/dl respectively. However, the serum level of HDL-C was significantly decreased in NAFL rats from 51.20 mg/dl to 30.76 mg/dl. The treatment of NAFL rats with folic acid resulted in a decline in the serum levels of TG, TC, LDL-C, and elevation in the serum levels of HDL-C in a dose-dependent manner. Respecting the highest dose (75 μg/Kg), results indicated that folic acid treatment completely normalized the serum levels of TC, HDL-C, while the lower dose showed a non-significant effect on the TG levels compared with the untreated NAFL rats.

The effect of folic acid on triglycerides content was evaluated in hepatic tissues (Table 3). The NAFL rats had a marked elevation in the hepatic TG contents by about 215% relative to the control. However, the folic acid treatments diminished (p ≤ 0.05) the hepatic TG contents in a dose-dependent manner in NAFL rats (92.8±9.01, 70.4±8.92 & 56.4±6.97 for 25, 50 and 75 μg/Kg respectively) when compared with untreated NAFL rats (107.8±3.33).

Effect of folic acid on the circulating TNF-α and adiponectin

The NAFL rats had marked higher circulating levels of TNF-α which were elevated by ~ 19-Folds, however, adiponectin levels were lowered (~ 51%) compared with control rats (Fig 1). Dose-dependently, the treatment of NAFL rats with folic acid significantly ameliorated the TNF-α and adiponectin levels compared with the untreated NAFL rats. The adiponectin levels were completely normalized at the doses of 50 and 75 μg/Kg, while TNF-α levels were still significantly higher than the control values (28.83 pg/ml), (Fig 1).

Fig 1

The effect of folic acid (25, 50 and 75 μg/Kg) treatment on TNF-α and the adiponectin levels in NAFL rats.

Data were shown violin plots (n = 10) and the level of significance was ≤ 0.05. a, significant vs. control rats; b, significant vs. untreated NAFL rats; c, significant vs. NAFL rats treated with folic acid 25 μg/Kg; d, significant vs. NAFL rats treated with folic acid 50 μg/Kg.

Effect of folic acid on the miR-21, miR-34a and miR-122 expression

Considering the effect of folic acid on NAFL rats, the expressions of both miR-21, -34a and -122 were evaluated in the liver (Fig 2). The up-regulation of both miR-21 and miR-34a was noticed (p ≤ 0.05) in the untreated NAFL rats to be about 4 and 3.8 folds, respectively compared to the control values. The treatment of NAFL rats with folic acid resulted in a significant decline of these microRNAs dose-dependently. However, the highest dose of folic acid (75μg/Kg) exerted the most down-regulation in the expression of hepatic miR-21 and miR-34a by about 57 and 63%, respectively compared to the NAFL rats. In the case of the hepatic expression of miR-122, a significant down-regulation by about 64% in NAFL rats respecting the control values was observed. Folic acid treatments dose-dependently up-regulated (p ≤ 0.05) the expression of miR-122 compared with the untreated NAFL rats, while the doses of 50 and 75μg/Kg that normalized (p ≤ 0.05) the expression of miR-122 by about 2.7 and 3 folds, respectively relative to the untreated NAFL rats.

Fig 2

The effect of folic acid (25, 50 and 75 μg/Kg) treatments on the hepatic levels of miR-21, -34a and -122 genes in NAFL rats.

Data were shown as violin plots (n = 10) and the level of significance was ≤ 0.05. a, significant vs. control rats; b, significant vs. untreated NAFL rats; c, significant vs. NAFL rats treated with folic acid 25 μg/Kg; d, significant vs. NAFL rats treated with folic acid 50 μg/Kg.

Effect of folic acid on the expression of hepatic HBP1, SIRT1, and SREBP-1c genes

The expression of HBP1, SIRT1, and SREBP-1c transcripts was investigated and results were shown in Fig 3. Untreated NAFL rats had significant suppression of the transcription of HBP1 and SIRT1 by ~ 44 and 58%, respectively compared with the normal rats. Folic acid treatments of NAFL rats induced a significant dose-dependent up-regulation of the transcript levels of these genes compared with the untreated NAFL rats. In comparison with the normal rats, the highest folic acid dose (75 μg/Kg) exerted approximately complete normalization of HBP1 and SIRT1 mRNA expression. In contrast, the hepatic expression of SREBP-1c showed a significant dose-dependent up-regulation in the untreated NAFL rats (3.66 folds) compared with the normal group. Folic acid treatments significantly down-regulated the SREBP-1c expression in NAFL rats compared with the untreated rats. The highest dose of folic acid (75 μg/Kg) exerted the most ameliorative effect by a down-regulation (~ 70%) of the SREBP-1c mRNA expression in NAFL rats.

Fig 3

The effect of folic acid treatments (25, 50 and 75 μg/Kg) on the HBP1, SIRT1, and SREBP-1c gene expression in NAFL rats’ liver.

Data were shown as violin plots (n = 10) and the level of significance was ≤ 0.05. a, significant vs. control rats; b, significant vs. untreated NAFL rats; c, significant vs. NAFL rats treated with folic acid 25 μg/Kg; d, significant vs. NAFL rats treated with folic acid 50 μg/Kg.

Effect of folic acid on the liver histology

The liver sections of control stained with H&E showed anastomosing plates of hepatocytes radiating from the central vein. Most hepatocytes are mononucleated and some hepatocytes are binucleated. The hepatocytes connected with blood sinusoids and Kupffer cells were associated with sinusoidal lining cells as shown in control (Fig 4). The hepatic architecture in NAFL group showed zonal degeneration; congested central and portal veins impacted with hemolyzed blood, increased inflammatory cellular infiltration in portal area, dilated sinusoids, necrosis, macrosteatosis and deeply eosinophilic cytoplasm. Also, fatty degeneration (steatosis) was present, where hepatocytes contained ballooned clear cytoplasm in contact with wide sinusoids. Many apoptotic figures were also seen in addition to necrosis and increased eosinophilia in some hepatocytes. The treatment of NAFL rats with folic acid in a dose of 25 μg/Kg per day showed patent sinusoids attached with central vein, increased apoptosis in many hepatocytes away from central vein with increased steatosis as shown in (Fig 4). Folic acid treatment in a dose of 50 μg/Kg/day showed normal portal area; portal vein, hepatic artery and bile duct in addition to little hepatocytes with apoptotic figures as compared with their counterparts. Treatment of NAFL rats with the highest dose of folic acid (75 μg/Kg/day) revealed evidence of histological features more or less similar to the normal control with minimal number of apoptotic figures (Fig 4).

Fig 4

Haematoxylin and Eosin (H&E)-stained liver sections showing the effect of folic acid (25, 50 and 75 μg/Kg) treatment on liver histology of NAFL rats.

The sections of control show normal hepatocytes with central vein (CV), sinusoids (S), Kupffer cells (K), and binucleated hepatocytes (thin arrow). The sections of NAFL rats (untreated) show fatty degeneration (steatosis); where hepatocytes have areas of clear cytoplasm, wide sinusoids (S) apoptotic figures (⇧), necrosis (N) and increased eosinophilia (E) in some hepatocytes. The liver section of NAFL rat treated with folic acid (25 μg/Kg) shows patent sinusoids attached with central vein), apoptosis in many hepatocytes with increased steatosis. The liver section of NAFL rat treated with folic acid (50 μg/Kg) shows normal portal area; portal vein (PV), hepatic artery (HA) and bile duct (B). Note little hepatocytes with apoptotic figures and sinusoids lined with kupffer cells. Finally, liver sections of NAFL rat treated with folic acid (75 μg/Kg) show plates of almost normal hepatocytes in contact with sinusoids, central vein (CV) and minimal number of apoptotic figures (H & E; Bar = 50 μm).

Discussion

Folic acid has therapeutic effects in rats with HFD-induced NAFL, as well as a function in modulating hepatic miR-21, -34a, and -122 and their target genes (HBP1, SIRT1, and SREBP-1c, respectively). The development and progression of NAFL are multi-factorial because many different pathways are involved but the order and extent of each pathway remain unclear [24]. One of the compromised pathways in NAFL is the one-carbon metabolism that involves interrelated methionine and folate pathways affecting many biological processes such as DNA synthesis and methylation, oxidative stress defense and detoxification [25]. In cellular metabolism, S-adenosylmethionine (SAM) seems to be the most essential methyl-group donor [26]. Folic acid helps remethylate homocysteine into methionine, which is a SAM precursor. The reduced SAM concentration promotes the reduction of DNA methylation capacity that affects the epigenetic regulation of gene expression [27], and inhibits the synthesis of carnitine and phosphatidylcholine and alters hepatic microRNAs (miRNAs) levels and DNA methylation patterns [28]. A folate-deficient diet reduces the liver’s methylation capacity and causes liver steatosis through three potential mechanisms: (1) a decrease in de novo phophatidylcholine (PC) synthesis, which results in an increase in liver TG synthesis and a diminish in very low-density lipoprotein (VLDL) secretion; (2) suppression of carnitine synthesis, which influences the transport of long-chain fatty acids into thethe mitochondria and (3) an increase in the expression of genes associated with hepatic lipid production [29], moreover, the regulation of hepatic genes related to oxidative stress and AMP-activated protein kinase (AMPK) activation can be involved in the mechanism [29]. All these changes can encourage the initiation and progression of NAFL through the modulation of lipid metabolism, a finding that renders the one-carbon metabolism pathway a lead target for the pharmacological industry [30]. Respecting the controls, the NAFL rats had a significant body weight gain and increased relative liver weights indicating the behavior of obesogenicity induced by HFD. Moreover, NAFL-rats demonstrated a significant hepatic triglycerides accumulation that may induce histological manifestations of fatty liver with signs of steatohepatitis [31]. At the histological level, feeding the male rats a HFD for 14 weeks resulted in histological manifestations of fatty liver with signs of steatohepatitis; macrosteatosis, portal inflammation, inflammatory cell infiltration, apoptosis, and cirrhotic changes. These changes in hepatic tissues are commonly associated with significant alterations in the pathways involved in the pathogenesis of the NAFL [31]. The serum AST and ALT activities and serum bilirubin levels together with the significant elevation in the circulatory levels of TNF-α suggested the occurrence of hepatic inflammation and necrosis in NAFL rats [28].

At the metabolic level, NAFL-rats showed a significant elevation in blood sugar, insulin, and HOMA-insulin resistance index indicating the insulin resistance status leading to the first hit of hepatic steatosis [32]. The NAFL rats of the current study have a significant decline in the serum adiponectin level, a marker of insulin sensitivity and inversely correlated with type 2 diabetes risk [33]. The increased blood glucose level may be due to the diminished capacity of insulin in limiting hepatic glucose synthesis [34]. These derangements in glucose homeostasis are associated with significant changes in serum lipid profile; TG, TC, and LDL-C which were significantly higher compared to control rats in this study, while the HDL-C serum level was significantly lower.

Impaired hepatic lipid deposition are the consequence of disturbances in the balance between the hepatic supply and synthesis of triglycerides and their consumption and oxidation. The main sources of lipids contributing to the fatty liver are the plasma non-esterified fatty acid (NEFA) pool from adipose tissue and the diet [35]. The increased influx of a NEFA to the liver from adipose tissues and diet through portal vein impairs insulin actions leading to insulin resistance, stimulating gluconeogenesis, and triglycerides synthesis [31]. These metabolic changes correlates with molecular changes in the hepatic tissues. miR-21, -34a, and -122 crucially affect the hepatic physiology and pathophysiology of NAFL. As we assessed their levels in the current work, we observed that untreated NAFL-rats exhibited a significant up-regulation in hepatic miR-21 and miR-34a consequently causing a significant down-regulation in their corresponding genes; HBP1 and SIRT1 respectively. On the other hand, levels of miR-122 in the livers of NAFL rats were significantly inhibited, whereas SREBP-1c gene expression as the targeted gene was markedly elevated. This pattern of changes may confirm the regulatory role of miR-122 on the expression of SREBP-1c through targeting it 3’UTR in its mRNA. Thus, the highly expressed SREBP-1c gene could be explained by the low level of miR-122 [36]. The validation of the resulted reverse expression patterns of hepatic miR-122 was linked to the various hepatic diseases [37].

Lipid metabolism in liver and adipose tissues is regulated by miR-21, miR-34a, and miR-122 [38, 39]. Similarly, many studies confirmed miR-21 up-regulation in hepatic tissues of HFD-induced NAFL in mice [40], and in NFLD and HCC patients [11]. MicroRNA-21 can promote hepatic lipid accumulation through the suppression of the HBP1-P53-SREBP1c pathway resulting in the induction of hepatic lipogenesis [41]. The down-regulation of p53 pathway can contributes to the development of NAFL through the involvement of miR-21 expression. The p53 transcriptional activation, cell cycle suppression, and inhibition of lipogenesis can be mediated by HBP1 via suppressing the transcription of SREBP-1c, a lipogenic transcription factor. The increased hepatic SREBP-1c in NAFL could be attributed to the insulin signaling pathway. The presence of insulin receptor substrate-1 (IRS-1) is required for SREBP-1c transcription, which is not down-regulated in insulin-resistant animal, however, the IRS-2 is down-regulated and affect the glucose production by insulin levels. Thus, the elevated SREBP-1c protein inhances FA synthesis and accelerates TGs accumulation. In addition, elevated insulin levels in the IR increase lipogenesis by activating SREBP-1c, which blocks IRS-2-mediated insulin signalling [42, 43]. The increased expression of SREBP-1c may be resulted directly from the suppressed miR-122 and indirectly from the elevated miR-21 expression through the down-regulation of the p53 pathway [41]. Furthermore, elevated levels of SREBP-1c were reported in obese, insulin-resistant, and hyperinsulinemic ob/ob mice suffering from fatty livers which may be due to the transcriptional dependency on the occurrence of insulin receptor substrate-1 (IRS-1) [44]. Moreover, lipogenesis could be stimulated by high insulin levels during the incidence of insulin resistance via the SREBP-1c protein activation [45] leading to the inhibition of insulin receptor substrate-2 that allows the action of insulin on glucose metabolism [46].

One of the main targets of miR-34a is the SIRT-1 gene which was suppressed in the present model of HFD-induced NAFL. The induction of miR-34a inhibits the hepatocyte nuclear factor 4 alpha (HNF4α) and acts as a pro-inflammatory agent via the activation of nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) [47]. P53 is inhibited by miR-34a expression stimulation in primary hepatocytes through the miR34a/SIRT-1/p53 pathway. The pathway of miR34a/SIRT-1/p53 inhibited p53, in primary hepatocytes, as a result of miR-34a stimulation [11]. SIRT-1, a crucial regulator of the metabolism, is involved in certain genes responsible for lipid and cholesterol synthesis, such as peroxisome proliferator-activated receptors (PPARs), PPAR- gamma co-activator 1-α (PGC-1α), forkhead box-protein 1 (FOXO1), p53, and SREBPs [11]. Parallel to the present results, the enhanced expression of miR-34a was documented in patients with NAFL and T2DM, and its expression was correlated with the severity of NAFL [47].

The lower expression levels of miR-122 in NAFL rats were confirmed by previous studies that indicated the liver-specific miR-122 knock-out which rapidly develops NASH as a result of enhanced lipogenesis, alterations in lipid secretion, and increased TNF-α secretion [48]. The lower level of miR-122 in the liver is a consequence of the high rate of the circulating levels instead of miRNA down-regulation [37]. Furthermore, higher levels of circulating miR-122 with a reverse expression pattern were reported [43, 49]. Collectively, these data suggest the opposite hepatic expression pattern of the studied miRNAs and their corresponding genes in the NAFL phenotype which has been induced by HFD in rats.

To the best of our knowledge, no drugs are currently approved for NASH therapy and as yet there have been no clinical studies that address the utility of one-carbon donors such as folic acid in this condition. Thus, the study was designed to explore the folic acid potency for the treatment of NAFL through targeting insulin sensitivity, cytokines, and microRNAs and their target genes. we found that folic acid treatment caused a significant dose-dependent amelioration in body and liver weights, hepatic TG contents, liver function parameters, ALT, AST, and bilirubin levels. Moreover, the dose-dependent decrease in FBG, insulin, and insulin HOMA-IR, in addition to the improvement in the serum lipid profile were reported following the folic acid treatments in this study. These findings are supported by a previous study that showed that a one-carbon donor, folic acid, could improve liver injury in HFD-induced NAFL rats by modulating DNA methylation and liver redox status [22]. Folic acid treatments of NAFL rats significantly increased the circulating adiponectin levels. Adiponectin has anti-inflammatory and insulin-sensitizing properties affecting the metabolism of glucose and lipids and may contribute, at least partially, to the glucose-lowering effect of folic acid [50]. Further, folic acid treatment significantly and dose-dependently decreased the elevated serum levels of TNF-α which indicated the anti-inflammatory properties of folic acid on NAFL rats. In agreement with our findings, folic acid and the methyl donor mixture reduced pro-inflammatory gene expression in monocytes and decreased the secretion of TNF-a and IL-1b cytokines from differentiated macrophages by inhibiting the NFkB pathway [49].

On the basis of molecular targets, the possible contribution of folic acid treatment was investigated among miR-21, miR-34a, and miR-122 to explore the mechanistic molecular involvement of folic acid in insulin sensitivity and lipid metabolism. The glucose-lowering, insulin-sensitizing, lipotropic, and anti-inflammatory effects of folic acid treatment in NAFL rats are linked to dose-dependent changes in the hepatic miR-22, -34a, and -122 and their target genes. However, a dose-dependent down-regulation resulted after folic acid treatments in the expression of miR-21 and -34a, while a dose-dependent up-regulation in the expression of their target’s genes; HPB1 and SIRT1, respectively was noticed. Folic acid treatments significantly enhanced the suppressed expression of miR-122 and significantly down-regulated its target gene; SREBP-1c. The mechanism of folic acid effects could be attributed to the modulation of the miR-21, -34a, and -122 expressions which in turn, induce alterations in the expression of their corresponding genes; HBP1, SIRT1, and SREBP-1c, respectively. These genes are involved in the NAFL development by inducing lipogenesis, insulin resistance, and imbalanced cytokine production [51]. Folic acid treatment may affect microRNAs expression in NAFL rats’ liver possibly through changing the methylation status of promoter regions of the gene. This suggestion is reinforced to a study on mice receiving a folic acid-deficient diet [28]. Future research should aim to identify the mechanism(s) by which the folic acid supplements can alter the levels of miRNAs and their association with the target genes expression and ultimately implicate such findings in NAFL treatment. Additionally, the futrue work should consider the application of a commercial standard diet for the NAFL induction instead of the home-made one. Furthermore, the perfusion of rats with a physiological saline is recommended for blood removal form the tested organs.

Conclusions

Our work illustrated the dysregulation in of hepatic miR-21, -34a, and -122 expression as well as their targeted genes, HBP1, SIRT1, and SREBP-1c in addition to altered TNF-α and adiponectin levels in NAFL rats. The potential anti-steatotic, insulin-sensitizing, glucose-lowering, and lipotropic effects of folic acid in HFD-induced NAFL in rats were detected. These effects may be mediated through epigenetic modulation of the hepatic microRNAs and the expression of their target genes.

The data supporting Figs 1–4 are available in the supporting information file.

(DOCX)

Click here for additional data file.