Background
Non-alcoholic fatty liver disease (NAFLD) represents a liver disease spectrum characterized by excessive accumulation of fat in the liver, with no alcohol abuse [
1,
2]. NAFLD could be classified into the non-alcoholic fatty liver (NAFL; which is simple steatosis) and the non-alcoholic steatohepatitis (NASH) [
3]. Steatosis is a benign status with mild fat deposition, which could be reversed by the lifestyle modification (such as diet and exercise) [
4]. On the other hand, for NASH, in addition to the fat deposition, there would be intralobular inflammation and hepatocyte ballooning. Moreover, NASH can progress into advanced liver fibrosis, cirrhosis, and ultimate hepatocellular carcinoma (HCC) [
1,
5].
NAFLD is strongly associated with obesity, dyslipidemia, diabetes, and insulin resistance, which has been therefore regarded as the hepatic manifestation of metabolic syndromes [
6]. Despite massive advances in elucidating the genetic mechanism in NAFLD development, understanding of the disease pathogenesis remains incomplete [
1]. Recently, the
two-hit theory has been widely accepted to elucidate the pathogenesis of NAFL and NASH. The first
hit refers to the accumulation of triglyceride (TG) in hepatocytes, i.e., the simple steatosis. This process is closely associated with abnormal lipid metabolism involved in central obesity and insulin resistance. The second
hit includes mechanisms contributing to the development of inflammation and fibrosis, such as oxidative stress and mitochondrial dysfunction [
7,
8].
Patients with NAFLD are always asymptomatic in clinic. The disease is often diagnosed when there is evidence for liver steatosis on imaging modality, which is associated with the metabolic syndromes, including obesity (high body mass index, BMI, and waist circumference) and diabetes (high blood glucose with hypertriglyceridemia) [
5,
6]. Ultrasonography is a non-invasive method frequently used in the assessment of hepatic lipid accumulation [
9,
10], so as other imaging techniques like computed tomography (CT) and nuclear magnetic resonance (NMR) [
10,
11]. In addition, the blood biochemistry results could also give a hint on the diagnosis of NAFLD, such as the elevated transaminase level [
12].
Recently, there are advances in the non-invasive techniques intending to assess the NASH/fibrosis level, including the NAFLD fibrosis score (NFS) [
5], Fibro Meter [
13,
14], and Fibro Scan [
15], with, however, relatively low accuracy. Up to now, the liver biopsy is still considered to be the gold standard for the diagnosis of stages of NASH, as well as distinguishing NAFL, NASH, and liver fibrosis [
16]. However, no factors against NAFLD have been elucidated to date.
Yin Yang 1 (YY1), a ubiquitous, is a multifunctional zinc-finger transcription factor from the protein family, which can work as transcriptional repressor, activator, or initiator element binding protein [
17]. A myriad of potential YY1 target genes have already been identified, important for cell proliferation and differentiation process. YY1 has been shown to play an important role in regulating proliferation and apoptosis of tumor cells [
18]. Moreover, YY1 promotes the triglyceride accumulation in the adipocytes via repressing Chop10 transcription, implying its potential role in the development of obesity [
19]. Furthermore, YY1 has also been found to be able to repress the genes associated with the insulin/insulin-like growth factor (IGF) signaling pathway, such as IGF1–2, IRS1–2, and Akt1–3 in skeletal muscles [
20]. A recent study has also found that YY1 might be related to the body weight, glucose level, and cholesterol or free fatty acid level [
21]. In addition, compared with control subjects, the YY1 levels are significantly down-regulated in the liver tissues in NAFLD patients [
22]. However, the association between the YY1 expression and the NAFLD progression has not completely elucidated.
In this study, the obese patients undergoing bariatric surgery were divided into four groups according to the liver pathogenesis. The mRNA and protein expression levels of YY1 were determined, and the association between the YY1 expression and the NAFLD progression was investigated.
Methods
Study subjects
This study was approved by the Ethics Committee of the Affiliated Drum Tower Hospital of the Medical School of Nanjing University (Permit Number: 2017–030-02). This study was registered in International Clinical Trial Registry Platform (ICTRP), with the clinical trial number NCT03296605. Patients were selected from a cohort undergoing laparoscopic Roux-en-Y gastric bypass surgery at the Department of Hepatobiliary Surgery of the Affiliated Drum Tower Hospital of the Medical School of Nanjing University. Exclusion criteria were included the patients with evidence for viral hepatitis, hemochromatosis, or alcohol consumption (> 20 g/d for females and > 30 g/d for males) [
23]. The participants were recruited from April 2017 to February 2018. Written informed consent was obtained from all subjects.
Data collection
Liver tissue samples were obtained during surgery. One half was put into lipid nitrogen and stored at − 80 °C; and the other half was fixed by 10% formaldehyde, embedded in paraffin, and subjected to the hematoxylin-eosin (H&E) staining. Specimen was stored in the Nanjing Multicenter Biobank, the Biobank of Nanjing Drum Tower Hospital, and the Affiliated Hospital of Nanjing University Medical School. We conducted this study from February 2018. We had access to information that could identify individual participants during or after data collection. Histological characteristics were determined according to the Kleiner scoring system [
24]. Steatosis was assessed and scored in a scale of 0–3, inflammation grades of 0–3, and hepatocellular ballooning of 0–2. These histopathological features were used to estimate the NAFLD activity score (NAS). These subjects were classified into the control (without steatosis), hepatic steatosis (NAS of 1–2), non-defining NASH (NAS of 3–4), and NASH (NAS of ≥5) groups [
25]. Fibrosis was staged in based on the grades of 0–4. For biochemical measurement, blood samples were taken after an overnight (10-h) fast. Samples were analyzed and tested for the liver function, insulin level, C-reactive protein level, glucose level, and lipid panels (including total cholesterol, LDL, HDL, and triglycerides). Insulin activity was determined by the homeostatic model assessment for insulin resistance (HOMA-IR) index [
26,
27].
Quantitative real-time PCR
Total RNA was extracted from the liver tissue using Trizol (Invitrogen, Carlsbad, CA, USA). RNA (500 ng) was used for cDNA synthesis using random primers and Primescriptreverse transcriptase (Takara, Dalian, Liaoning, China). Quantitative real-time PCR was carried out using the SYBR Green qPCR kit (Takara), on a fluorescent temperature cycler. Primer sequences were as follows: YY1, forward 5′-ACGGCTTCGAGGATCAGATTC-3′ and reverse 5′-TGACCAGCGTTTGTTCAATGT-3′; and GAPDH, forward 5′-TGACTTCAACAGCGACACCCA-3′ and reverse5′-CACCCTGTTGCTGTAGCCAAA-3′. Reaction conditions were set as: 95 °C for 30 s, followed by 40 cycles of 95 °C for 5 s and 60 °C for 34 s. Target gene expression was calculated with semi-quantitative method. GAPDH was used as internal reference.
Western blot analysis
Tissues were lysed with the RIPA buffer containing phosphatase inhibitors. The protein concentration was determined using the BCA method (Pierce, Rockford, IL, USA). Totally 24 mg protein was separated on 10% SDS-PAGE, and then electronically transferred onto a PVDF membrane. After blocking with 3% BSA in 10 mM Tris-HCl (pH 7.4) containing 0.05% Tween-20, the membrane was incubated with the mouse anti-YY1 (Abcam, Cambridge, MA, USA) and anti-β-actin primary antibody (1:1000 dilution; Key GEN Bio TECH, Nanjing, Jiangsu, China) at 4 °C overnight. After washing, the membrane was incubated with the peroxidase-conjugated secondary antibody (Santa Cruz, Santa Cruz, CA, USA), and developed in the Super Signal West Pico Chemiluminescent Substrate (Pierce). The protein was visualized and quantified with the Imagine J software.
Immunohistochemistry analysis
Formalin-fixed liver tissue samples were subjected to the immunohistochemistry analysis. Briefly, liver sections were deparaffinized and treated by citrate, and then blocked with Immuno Detector Peroxidase Blocker (Bios SB, Santa Barbara, CA, USA). Sections were incubated with the rabbit anti-YY1 primary antibody (1:250dilution; Abcam, Cambridge, MA, USA) at 4 °C overnight. Liver sections were then treated with peroxidase-conjugated secondary antibody (Santa Cruz) and DAB chromogen. Then the samples were counterstained with hematoxylin, and observed under light microscope.
Statistical analysis
Data were expressed as mean ± SD. Statistical analysis was performed using the SPSS 19.0 software (SPSS Inc., Chicago, IL, USA). Group comparison of numeric variables was performed using the ANOVA or Kruskal-Wallis test, depending on the variables’ distribution. The χ2test was used for comparison of nominal categorical variables. Correlation analysis was conducted with the Spearman’s test. Multivariate logistic regression model was used to identify the significant clinical and metabolic factors that predicted the NAFLD absence, after adjusting for other factors such as BMI.
Discussion
In the present study, the factors associated with normal liver histology in patients with obesity were identified and investigated. Our findings identifying the protective factors could help guide the NAFLD screening among the patients with high risk, as well as further understand the disease pathogenesis. Patients undergoing weight-loss surgery offered insights into the unique patient subset. This cohort allowed for the identification of the protective factors against the development of NAFLD confirmed by histology in the high risk group. Our results showed thatYY1 was associated with the NAFLD progression. Furthermore, YY1 had strong association with glucose, insulin, HOMA-IR, ALT, and AST. These findings suggest that besides NAFLD, YY1 is also associated with the hepatic metabolism.
Recently, it has been shown that the hepatic YY1 expression level is increased in the diabetic rats [
28]. Moreover, YY1 promotes the hepatosteatosis and insulin resistance, mainly via FXR, in the animal model [
22]. FXR is a metabolic nuclear receptor, abundantly expressed in the liver, intestine, and kidney, which has been first identified as a key regulator in the cholesterol and bile acid homeostasis [
29]. Moreover, FXR is also a major transcriptional factor participating in the regulation of the glucose and lipid metabolism in liver. In line with this, our results showed that YY1 might influence the liver metabolism. Moreover, YY1 and ALT were most important factors to predict the NAFLD activity, further supporting the important interaction between the YY1 and NAFLD progression. The more severe NAFLD was, the higher the YY1 expression level would be. Taken together, these results suggest that the hepatic YY1 expression is an important factor involved in the progression of NAFLD. This study has important clinical significance for diagnose and treatment of NAFLD. And combination of YY1 and NAS scores can serve as a more accurate diagnostic indicator for NAFLD.
There are also limitations about this study. The data was derived from the cohort of patient undergoing bariatric surgery. However, there is need for confirmation in an additional cohort which including patients selected at daily routine in a hepatological setting for NASH, outside the setting for bariatric surgery, and it should be evaluated more broadly in healthy people. Actually, the data is difficult to collect because healthy people and patients without NASH usually reject invasive testing especially in China, so it’s not feasible to confirm our conclusion in another cohort in this study. Of course, further in-depth studies are still needed to investigate the correlation between YY1 and NAFLD progression in broader populations in the future.
Acknowledgements
We thank Dr. Biyun Xu for the kind assistant in statistical analysis.