Serum ANGPTL2 and ANGPTL3 as potential biomarkers for diagnosis of non-alcoholic fatty liver disease

Yan Ma; Yan Wang; Qiang Zhang; Jia-Nan Lang; Long-Yan Yang; Dong Zhao

doi:10.4103/ed.ed_33_19

ORIGINAL ARTICLE

Year : 2020 | Volume : 5 | Issue : 2 | Page : 29-37

Serum ANGPTL2 and ANGPTL3 as potential biomarkers for diagnosis of non-alcoholic fatty liver disease

Yan Ma, Yan Wang, Qiang Zhang, Jia-Nan Lang, Long-Yan Yang, Dong Zhao
Center for Endocrine Metabolism and Immune Disease, Beijing Key Laboratory of Diabetes Research and Care, Beijing Luhe Hospital, Capital Medical University, Beijing, China

Date of Submission	04-Nov-2019
Date of Decision	14-Apr-2020
Date of Acceptance	29-May-2020
Date of Web Publication	06-Jul-2020

Correspondence Address:
Dr. Dong Zhao
Center for Endocrine Metabolism and Immune Disease, Beijing Key Laboratory of Diabetes Research and Care, Beijing Luhe Hospital, Capital Medical University, Beijing 101149
China
Dr. Long-Yan Yang
Center for Endocrine Metabolism and Immune Disease, Beijing Key Laboratory of Diabetes Research and Care, Beijing Luhe Hospital, Capital Medical University, Beijing 101149
China

Source of Support: None, Conflict of Interest: None

Check

DOI: 10.4103/ed.ed_33_19

Abstract

Purpose: Nonalcoholic fatty liver disease (NAFLD) is a chronic liver disease that has a serious effect on worldwide public health. We sought relationships among ANGPTL2, ANGPTL3, and ANGPTL6 with NAFLD metabolic and biochemical parameters, to assess their potential as diagnostic tools for NAFLD.
Materials and Methods: Serum levels of ANGPTL2, ANGPTL3, and ANGPTL6 in NAFLD patients (n = 52) and non-NAFLD participants (n = 51) were quantified by ELISA. The utility of ANGPTLs as biomarkers for NAFLD prediction was assessed using receiver operating characteristic (ROC) analysis.
Results: Serum ANGPTL2 levels were increased (P = 0.0102), serum ANGPTL3 levels were decreased (P < 0.0001), and ANGPTL6 levels did not change (P = 0.3174) in 52 patients with NAFLD when compared to 51 participants without NAFLD. The areas under the ROC curves for ANGPTL2 and ANGPTL3 were 0.647 and 0.746, respectively. When two factors were jointly analyzed, the area under the ROC curve for ANGPTL2 and ANGPTL3 was 0.773.
Conclusion: ANGPTL2 and ANGPTL3 levels may be novel and important predictors of NAFLD severity.

Keywords: Angiopoietin-like proteins, nonalcoholic fatty liver disease, nuclear factor-kappa B/c-Jun N-terminal kinase signaling

How to cite this article:
Ma Y, Wang Y, Zhang Q, Lang JN, Yang LY, Zhao D. Serum ANGPTL2 and ANGPTL3 as potential biomarkers for diagnosis of non-alcoholic fatty liver disease. Environ Dis 2020;5:29-37

How to cite this URL:
Ma Y, Wang Y, Zhang Q, Lang JN, Yang LY, Zhao D. Serum ANGPTL2 and ANGPTL3 as potential biomarkers for diagnosis of non-alcoholic fatty liver disease. Environ Dis [serial online] 2020 [cited 2023 Jun 2];5:29-37. Available from: http://www.environmentmed.org/text.asp?2020/5/2/29/289030

Introduction

Nonalcoholic fatty liver disease (NAFLD) is a chronic and growing, worldwide disease that seriously affects public health.^[1],[2] NAFLD is a chronic metabolic disease characterized with the excessive accumulation of fat in hepatocytes and except resulting from alcohol intake or other secondary liver disease. NAFLD is accompanied by an increased risk for serious metabolic diseases such as diabetes and cardiovascular disease.^[3],[4],[5] as well as increased mortality.^[6] NAFLD may progress to a more severe disease status characterized by hepatocyte loss, steatosis, ballooning, inflammatory necrosis, and fibrosis known as nonalcoholic steatohepatitis (NASH).^[7],[8] NASH is one of the important factors of liver cirrhosis or liver cancer.^[9],[10] Precise histological diagnosis, including disease stage, is commonly based on liver biopsy.^[2] However, this approach has a number of limitations, including the need for a specialized performance site. Hence, considerable attention has been given to the development of noninvasive diagnostic approaches such as the identification of reliable biomarkers for disease prediction and prognosis.

One potential biomarker is angiopoietin-like protein 8 (ANGPTL8). The levels of ANGPTL8 are increased in moderate-to-severe NAFLD patients when compared to levels in patients with no NAFLD or mild NAFLD.^[11] ANGPTL8 belongs to the “angiopoietin-like protein” (ANGPTL) family, which exhibits a similar structure to angiogenic-regulating factors.^[12] ANGPTLs have been suggested to play a role in controlling lipid, carbohydrate, and energy metabolism.^[13] ANGPTL1–7 share a common structure but exhibit distinct functions.^[14] Several studies have found that ANGPTL2 levels were raised in obesity,^[15] metabolic syndrome, type 2 diabetes, certain tumor^[16] and atherosclerosis,^[17] and excessive ANGPTL2 led to chronic inflammation and systemic insulin resistance. It was reported that ANGPTL2 could activate nuclear factor-kappa B (NF-κB) signaling leading to the release of inflammatory factor, such as tumor necrosis factor-α (TNF-α) and CCL2, and the aggregation of inflammatory cells which promoted the development of chronic inflammation of the liver.^[18] Lu et al. also found that berberine could inhibit inflammatory activity of NAFLD through ANGPTL2 pathway.^[18] ANGPTL3 is predominantly expressed in the liver^[19] and has been shown to modulate plasma lipoprotein metabolism.^[20] ANGPTL6, also known as angiopoietin-like growth factor, counteracts obesity through increasing systemic energy expenditure and preventing subsequent metabolic disease.^[21] Previous studies suggested that serum levels of ANGPTL6 increased significantly in various chronic metabolic diseases and showed associations between ANGPTL6 levels and prognosis of the diseases. Increased serum ANGPTL6 levels could be an independent predictive value for metabolic syndrome.^[22] Serum ANGPTL6 levels were significantly higher in type 2 diabetic patients.^[23] To date, there is no evidence that serum ANGPTL2 or ANGPTL6 levels can serve as reliable biomarkers for NAFLD severity prediction or disease prognosis. In this study, nondiabetic NAFLD patients and non-NAFLD participants were assessed for circulating levels of ANGPTL2, ANGPTL3, and ANGPTL6. The purpose of this study was to determine whether these molecules were related to subject metabolic and/or biochemical parameters. In this manner, the efficacy of ANGPTLs for the assessment of NAFLD was examined.

Materials and Methods

Study population

The study comprised 103 individuals, including 51 healthy people enrolled in a medical examination center (control group) and 52 NAFLD patients admitted to the Center for Endocrine Metabolism and Immune Diseases, at Beijing Luhe Hospital, Capital Medical University in Beijing. The inclusion criteria were as follows: age >30 and <50 years and no excessive alcohol intake (defined as >30 g/day for males and >20 g/day for females). All individuals underwent an abdominal ultrasound, implemented by the same operator. Ultrasonography was considered an accepted tool for steatosis screening, although it lacks sensitivity. Mild hepatic steatosis was defined as mildly increased echogenicity, normal diaphragm, and vascular margin definition. Moderate hepatic steatosis was defined as moderately increased echogenicity, mild loss of diaphragm, and vascular margin definition. Severe hepatic steatosis was defined as severely increased echogenicity, loss of diaphragm, and vascular margin definition. We excluded patients with coronary heart disease, cerebrovascular disease, hypothyroidism, pituitary related diseases, chronic renal insufficiency, chronic liver disease, pregnancy, women within 1-year postpartum, those who had taken lipid-lowering drugs, those who were hypoglycemic, on corticosteroids, or other drugs within 3 months before examination.

Clinical measurements

Before examination, individuals fasted from 22:00 and venous blood samples were drawn between 7:00 and 8:00 the next morning. Serum was separated within 1 h. Blood analysis included fasting plasma glucose (mmol/L), glutamic-pyruvic transaminase (ALT, U/L), glutamic-oxaloacetic transaminase (AST, U/L), gamma-glutamyl transpeptidase (GGT, U/L), serum total bilirubin (μmol/L), direct bilirubin (μmol/L), triglyceride (TG, mmol/L), total cholesterol (TC, mmol/L), low-density lipoprotein cholesterol (LDL-C, mmol/L), and high-density lipoprotein cholesterol (HDL-C, mmol/L). Serum was stored at −80°C until further analysis.

Laboratory measurements

Serum ANGPTL2 (1F-716, Immuno-Biological Laboratories Co., Ltd., Japan), ANGPTL3 (1G-823, Immuno-Biological Laboratories Co., Ltd., Japan), and ANGPTL6 (L180719405, Cloud-Clone Corp., USA) levels were measured by ELISA kit according to manufacturer's instructions.

Gene set enrichment analysis

The GEO data set GSE48452 was used for differential gene expression screening. The relationship between biological pathway and ANGPTL expression levels was analyzed using gene set enrichment analysis (GSEA v2.2, http://www.broad.mit.edu/gsea/). Default settings were used. Thresholds for significance were determined by permutation analysis (1000 permutations). A false discovery rate (FDR) was calculated. FDR <0.05 was considered significant.

Statistical analysis

Statistical analyses were performed using SPSS 18.0 (SPSS Inc., Chicago, IL, USA). Two-tailed unpaired Student's t-tests were used to determine statistical significance. Data are presented as means ± standard deviation or N (%). We conducted Spearman correlation analysis to examine the relationships between serum ANGPTL levels and clinical and biochemical variables, followed by multivariate regression analysis with adjustment for various covariates. P < 0.05 was considered for statistical significance. Receiver operating characteristic (ROC) curve and area under the ROC curve (AUC) analyses were used to detect the diagnosis and recognition ability of ANGPTLs to NAFLD.

Results

Baseline characteristics of the study population

The baseline characteristics of the study population, according to NAFLD status, are shown in [Table 1]. Among 103 individuals, 52 patients were found, by imaging, to have NAFLD. Patients with NAFLD had significantly higher body mass index (BMI), systolic pressure, diastolic pressure, serum levels of fasting glucose, ALT, AST, GGT, TC, and HDL-C when compared to those without NAFLD. There was no significant difference in age, blood pressure, TG, or LDL-C between the two groups.

Table 1: Clinical data from patients

Click here to view

ANGPTL levels in human subjects with or without nonalcoholic fatty liver disease

Serum levels of ANGPTLs were analyzed by ELISA. Patients with NAFLD had elevated ANGPTL2 levels when compared to participants without NAFLD [3.357 ± 0.147 vs. 3.979 ± 0.162 μg/L, P = 0.0102, [Figure 1]a, whereas serum ANGPTL3 levels were decreased in patients with NAFLD when compared to participants without NAFLD [410.4 ± 21.17 vs. 582.9 ± 28.07 μg/L, P < 0.0001, [Figure 1]b. Serum ANGPTL6 levels for patients with NAFLD when compared to participants without NAFLD [38.08 ± 2.987 vs. 32.95 ± 2.060 μg/L, P = 0.3174, [Figure 1]c were not significantly different between the two groups. These data suggest that serum ANGPTL2 and ANGPTL3 levels were associated with NAFLD.

Figure 1: Comparison of serum ANGPTL2, ANGPTL3, and ANGPTL6 levels in healthy controls and nonalcoholic fatty liver disease patients. (a) Serum ANGPTL2 levels were elevated in nonalcoholic fatty liver disease patients. (b) Serum ANGPTL3 levels were decreased in nonalcoholic fatty liver disease patients. (c) Serum ANGPTL6 levels were unchanged in nonalcoholic fatty liver disease patients. Data are expressed as means ± standard deviation. Statistical significance was accepted at P < 0.05

Click here to view

To elucidate the role of ANGPTLs in NAFLD progression, serum levels of ANGPTLs in NAFLD patients at different levels of severity were analyzed. The results revealed that with an increase in NAFLD severity, serum ANGPTL2 levels [3.640 ± 0.180 vs. 4.367 ± 0.270 μg/L, P = 0.0345, [Figure 2]a increased. Serum ANGPTL3 levels [424.3 ± 28.61 vs. 390.5 ± 32.66 μg/L, P = 0.3037, [Figure 2]b and serum ANGPTL6 levels [34.19 ± 3.055 vs. 31.76 ± 2.913 μg/L, P = 0.4269, [Figure 2]c were unchanged. These data suggest that serum levels of ANGPTL2 increased with NAFLD progression.

Figure 2: Correlations between expression levels of ANGPTL2, ANGPTL3, and ANGPTL6 with nonalcoholic fatty liver disease progression. (a) ANGPTL2 was increased with an increase in the clinical stage of nonalcoholic fatty liver disease. (b) ANGPTL3 was decreased with an increase in the clinical stage of nonalcoholic fatty liver disease. (c) ANGPTL6 was unchanged with an increase in the clinical stage of nonalcoholic fatty liver disease. Data are expressed as means ± standard deviation. Statistical significance was accepted at P < 0.05

Click here to view

To further assess a role for serum ANGPTL2, ANGPTL3, and ANGPTL6 in NAFLD occurrence and progression, constituent ratio analysis was performed. Consistent with previous results, in the healthy controls, only 20/51 (39%) cases exhibited high levels of ANGPTL2 (>median). However, for mild and moderate/severe stages of NAFLD, 13/27 (48%) cases and 18/24 (75%) cases showed high levels of ANGPTL2, respectively [Figure 3]a. Similarly, 36/51 (71%) healthy controls exhibited high levels of ANGPTL3 (>median), whereas only 10/27 (37%) and 6/24 (25%) NAFLD cases for mild and moderate/severe stages of NAFLD, respectively, showed high levels of ANGPTL3 [Figure 3]b. No differences in ANGPTL6 levels were observed for any groups [Figure 3]c. These data suggest that increased levels of serum ANGPTL2, as well as decreased serum levels of ANGPTL3, may contribute to NAFLD development.

Figure 3: Constituent ratios, with different levels of ANGPTLs, were significantly different between healthy samples and mild or moderate/severe stage nonalcoholic fatty liver disease. (a) ANGPTL2 levels gradually increased with the development of nonalcoholic fatty liver disease. (b) ANGPTL3 levels gradually decreased with the development of nonalcoholic fatty liver disease. (c) ANGPTL6 was unchanged with the development of nonalcoholic fatty liver disease

Click here to view

Correlation of ANGPTLs with clinical parameters

Relationships were sought for serum ANGPTL2 and ANGPTL3 levels and various clinical and biochemical profiles [Table 2]. By Spearman correlation analysis, ANGPTL2 levels had strong positive correlations with diastolic pressure (r = 0.3297, P = 0.0007), ALT (r = 0.3158, P = 0.0012), and GGT (r = 0.4031, P < 0.0001). NAFLD status (r = 0.2547, P = 0.0094), BMI (r = 0.2948, P = 0.0028), systolic pressure (r = 0.2886, P = 0.0033), AST (r = 0.2106, P = 0.0327), TC levels (r = 0.2616, P = 0.0079), and HDL-C levels (r = −0.2366, P = 0.0167).

Table 2: Association of angiopoietin-like 8 with clinical parameters

Click here to view

ANGPTL3 levels had strong positive correlations with NAFLD status (r = −0.4262, P < 0.0001), TC levels (r = −0.3892, P < 0.0001), GGT (r = 0.3433, P = 0.0004), BMI (r = −0.2671, P = 0.0069), fasting blood glucose (r = −0.2717, P = 0.0057), and HDL-C (r = −0.3016, P = 0.0021).

Serum ANGPTL levels and cytokeratin-18 as predictors of an nonalcoholic fatty liver disease diagnosis

Because the serum levels of ANGPTL2 and ANGPTL3 changed significantly with the occurrence and development of NAFLD, we evaluated the utility of ANGPTL2 and ANGPTL3 as predictive biomarkers for NAFLD by performing ROC analysis [Figure 4]a and [Figure 4]b. The area under the ROC curve (AUC) for ANGPTL2 was 0.647. The area under the ROC curve for ANGPTL3 was 0.746. Cytokeratin-18 (CK18) is a strong candidate as an NAFLD biomarker.^[24] We, therefore, evaluated CK18 levels in patients, and the AUC for CK18 was 0.741 [Figure 4]c. These results demonstrated that ANGPTL3 was a stronger predictor for a NAFLD diagnosis.

Figure 4: Receiver operating characteristic curves for serum ANGPTL2, ANGPTL3, and cytokeratin-18 levels for the prediction of nonalcoholic fatty liver disease. (a) Receiver operating characteristic curve for ANGPTL2 serum levels for the diagnosis of nonalcoholic fatty liver disease. The AUC for the receiver operating characteristic curve was 0.647 (95% confidence interval, 0.547–0.739). (b) Receiver operating characteristic curve for ANGPTL3 protein levels for the diagnosis of nonalcoholic fatty liver disease. The AUC was 0.746 (95% confidence interval, 0.651–0.827). (c) Receiver operating characteristic curve for cytokeratin-18 protein levels for the diagnosis of nonalcoholic fatty liver disease. The AUC was 0.741 (95% confidence interval, 0.637–0.829)

Click here to view

Next, we established a logistic regression model and carried out multivariate analysis by ROC curve. When two factors were jointly analyzed, the area under the ROC curve for ANGPTL2 and ANGPTL3 was 0.773 [Figure 5].

Figure 5: Multivariate receiver operating characteristic analysis for the prediction of nonalcoholic fatty liver disease. Multivariate receiver operating characteristic curve analysis for ANGPTL2 and ANGPTL3 serum levels for the diagnosis of nonalcoholic fatty liver disease. The AUC was 0.773 (95% confidence interval, 0.680–0.850)

Click here to view

Measurement of ANGPTL mRNA levels in liver tissue of nonalcoholic fatty liver disease patients

The underlying mechanism of ANGPTL regulation in NAFLD is unknown. To examine ANGPTL expression in NAFLD patient livers, mRNA expression profiles were obtained from GSE48452 (http://www.ncbi.nlm.nih.gov/geo/). A total of 32 NAFLD patients and 41 healthy controls were analyzed. The results showed that ANGPTL2 mRNA expression levels were increased in NAFLD livers when compared to healthy controls [Figure 6]a. ANGPTL3 mRNA was downregulated in NAFLD livers [Figure 6]b. These results suggested that changes in serum ANGPTL2 and ANGPTL3 in NAFLD patients may be due to changes in the expression of ANGPTL2 and ANGPTL3 within patient livers.

Figure 6: ANGPTL levels were associated with nuclear factor-kappa B or c-Jun N-terminal kinase signaling in nonalcoholic fatty liver disease. (a) ANGPTL2 mRNA levels were elevated in nonalcoholic fatty liver disease patient livers. (b) ANGPTL3 mRNA levels were downregulated in nonalcoholic fatty liver disease patient livers. (c and e) Gene signatures for nuclear factor-kappa B and c-Jun N-terminal kinase signaling were enriched in nonalcoholic fatty liver disease patients in comparison to normal controls by gene set enrichment analysis plots. (d) Gene signatures for nuclear factor-kappa B signaling were enriched for patients in the ANGPTL2-high expression group. (f) Gene signatures for c-Jun N-terminal kinase activation were enriched in the ANGPTL3-low expression

Click here to view

The association of ANGPTL expression with nuclear factor κB or c-Jun N-terminal kinase signaling in nonalcoholic fatty liver disease patients

Hepatic fat accumulation leads to subacute inflammation through the activation of NF-κB signaling, by releasing proinflammatory cytokines such as TNF-α. Hepatic steatosis induces endoplasmic reticulum stress, which activates the c-Jun N-terminal kinase (JNK). As such, NF-κB and JNK signaling may play important roles in the development of NAFLD. We first assessed the activation of NF-κB (HALLMARK_TNFA_SIGNALING_VIA_NFKB) and JNK (HAN_JNK_SINGALING_UP) signaling in NAFLD in the GEO dataset by the GSEA method. Our results showed that gene signatures for NF-κB and JNK signaling were enriched in NAFLD patients, in comparison to normal healthy controls [Figure 6]c and [Figure 6]d, which is consistent with previous results. Therefore, we assessed whether ANGPTL expression was associated with NF-κB or JNK signaling in NAFLD patients, within the GEO dataset by separating the groups into high and low ANGPTL expression groups. Correlations for NF-κB/JNK signaling and ANGPTL expression were analyzed by the GSEA method. The gene set for NF-κB was highly enriched in the ANGPTL2 high expression group (FDR <0.05), but JNK was not (FDR >0.05) [Figure 6]e and [Figure 6]f. For ANGPTL3, the JNK gene set was highly enriched in the low expression group (FDR <0.05), but the NF-κB gene set was not (FDR >0.05) [Figure 6]e and [Figure 6]f. These results demonstrated a positive correlation between ANGPTL2 levels and the activation of NF-κB signaling in NAFLD, with a negative correlation between ANGPTL3 levels and the activation of JNK signaling in NAFLD. High expression levels of ANGPTL2 and low expression levels of ANGPTL3 in the liver may contribute to the development of NAFLD.

Discussion

The key findings of this cross-sectional study are: (1) serum levels of ANGPTL2 increased, whereas ANGPTL3 levels decreased with a transition from no NAFLD to NAFLD and from mild NAFLD to moderate/severe NAFLD. ANGPTL6 levels did not change; (2) ANGPTL2 levels were positively related to ALT, AST, and GGT, whereas ANGPTL3 levels were negatively related to ALT and GGT; (3) the area under the ROC curves for ANGPTL2 and ANGPTL3 were 0.647 and 0.746, respectively. (4) The changed serum ANGPTL2 and ANGPTL3 levels in NAFLD patients may be due to liver expression of mRNA in NAFLD patients; (5) gene enrichment analysis by GSEA demonstrated a positive relationship between ANGPTL2 levels and the activation of NF-κB signaling in NAFLD samples, with a negative relationship between ANGPTL3 levels and the activation of JNK signaling in NAFLD samples.

During the clinical progression of NAFLD, the assessment of NAFLD is an important predictor of disease progression and is useful for therapeutic management decisions.^[25] Liver biopsy remains the gold standard for the assessment of progression, but it has significant limitations. Recently, liver fibrosis has been independently associated with long-term outcomes in patients with NAFLD.^[26] As such, there is an urgent need for a reliable and noninvasive test that can accurately assess the degree of NAFLD. Our study found that serum levels of ANGPTL2 were increased during the transition from no NAFLD to NAFLD, and from mild NAFLD to moderate/severe NAFLD, in non-T2DM patients. In contrast, Erkan et al. reported that serum ANGPTL2 levels were not significantly different between NAFLD patients and healthy controls.^[27] High levels of ANGPTL2 are expressed in adipose tissue. In obese mice and humans, circulating ANGPTL2 levels are increased and closely associated with inflammatory cytokines.^[15] Furthermore, our study revealed that serum ANGPTL2 levels were positively related to ALT, AST, and GGT. ALT, AST, and GGT are important biochemical indices of liver function and are closely related to NAFLD;^[28] hence, ANGPTL2 may be involved in the progression of NAFLD, but large-scale studies are required to validate these findings.

Liver-derived ANGPTL3 has been shown to be involved in a variety of metabolic processes.^[13],[29] Moreover, ANGPTL3 has been suggested as a potential surrogate biochemical marker for metabolic syndrome.^[30],[31] Yilmaz et al. reported levels of ANGPTL3 to be higher in patients with definitive NASH (n = 40) and borderline NASH (n = 8) when compared to controls (n = 14). No significant differences were found in patients with simple fatty liver (n = 9) when compared to controls.^[32] We found serum levels of ANGPTL3 decreased in NAFLD patients when compared to those without NAFLD. ANGPTL3 levels showed a downward trend from mild NAFLD to moderate/severe NAFLD. These results may be due to differences in sample size and nationality. The involvement of ANGPTL3 in metabolic processes is complex, and further study is required. No significant differences were found between NAFLD patients and non-NAFLD participants with respect to serum ANGPTL6, which is consistent with a previous study.^[27]

Since liver biopsy has many limitations, noninvasive measures for NAFLD diagnosis and disease severity measurement are urgently needed. Recently, research efforts have been directed toward the identification of reliable biomarkers for the prediction of NAFLD disease severity and prognosis. Novel noninvasive approaches have been proposed, including transient elastography with various scoring systems.^{[33],[34],[35]} However, this approach has less than optimal interobserver consistency and does not adequately distinguish early liver fibrosis from a nonfibrotic liver. Various biomarkers such as ALT: AST ratio, CK18, C-reactive protein, interleukin-1 receptor antagonist, CXCL10, PAI-1, adipokines, and FGF21 have been proposed as noninvasive measures of NAFLD severity but have not proven to be clinically valid.^[36] We found ANGPTL3 to be a better predictor of NAFLD than ANGPTL2. However, when these two factors were jointly analyzed, their predictive capacity was better than CK18.

Our results suggest that changes in serum ANGPTL2 and ANGPTL3 in NAFLD patients may be due to liver expression changes in ANGPTL2 and ANGPTL3 in patients. To further understand these observations, ANGPTL2 and ANGPTL3 relationships between cellular signaling molecules were carried out using gene enrichment analysis by GSEA plots. We found a positive relationship between ANGPTL2 levels and the activation of NF-κB signaling in NAFLD samples, with a negative relationship between ANGPTL3 levels and the activation of JNK signaling in NAFLD samples. Hence, high expression levels of ANGPTL2 and low expression levels of ANGPTL3 in the liver may contribute to the development of NAFLD.

The current study must be interpreted within the context of the following limitations. First, this is a cross-sectional study and thus longitudinal studies are required to determine whether ANGPTL2 and ANGPTL3 levels predict the NAFLD severity. Second, the relatively small sample size limits the generalizability of our conclusions. Due to the strict criteria for inclusion, the study comprised 103 individuals, including 51 healthy people and 52 NAFLD patients. The sample size of this study is relatively small, which is the limitation of the study, and we will further expand the sample size to verify the results of the study in the future. Third, ANGPTL levels were not measured in liver biopsy tissues.

Conclusion

These data demonstrated that serum levels of ANGPTL2 were increased, and ANGPTL3 levels were decreased in NAFLD patients, compared to participants without NAFLD, and in patients with mild NAFLD compared to moderate/severe NAFLD. Serum levels of ANGPTL2 and ANGPTL3 were the strongest diagnostic predictors of NAFLD. ANGPTL2 and ANGPTL3 levels may be novel and important predictors of NAFLD severity.

Financial support and sponsorship

This work was supported by the Beijing Natural Science Foundation (Project Number: 7184222) and National Science Funding of China (Project Number: 81800768).

Conflicts of interest

There are no conflicts of interest.

References

1.	Brunt EM, Wong VW, Nobili V, Day CP, Sookoian S, Maher JJ, et al. Nonalcoholic fatty liver disease. Nat Rev Dis Primers 2015;1:15080.
2.	Chalasani N, Younossi Z, Lavine JE, Charlton M, Cusi K, Rinella M, et al. The diagnosis and management of nonalcoholic fatty liver disease: Practice guidance from the American Association for the Study of Liver Diseases. Hepatology 2018;67:328-57.
3.	Targher G, Day CP, Bonora E. Risk of cardiovascular disease in patients with nonalcoholic fatty liver disease. N Engl J Med 2010;363:1341-50.
4.	Stefan N, Kantartzis K, Häring HU. Causes and metabolic consequences of Fatty liver. Endocr Rev 2008;29:939-60.
5.	Anstee QM, Targher G, Day CP. Progression of NAFLD to diabetes mellitus, cardiovascular disease or cirrhosis. Nat Rev Gastroenterol Hepatol 2013;10:330-44.
6.	Blachier M, Leleu H, Peck-Radosavljevic M, Valla DC, Roudot-Thoraval F. The burden of liver disease in Europe: A review of available epidemiological data. J Hepatol 2013;58:593-608.
7.	Brunt EM. Histopathology of non-alcoholic fatty liver disease. Clin Liver Dis 2009;13:533-44.
8.	Brunt EM. Pathology of nonalcoholic fatty liver disease. Nat Rev Gastroenterol Hepatol 2010;7:195-203.
9.	Ekstedt M, Franzén LE, Mathiesen UL, Thorelius L, Holmqvist M, Bodemar G, et al. Long-term follow-up of patients with NAFLD and elevated liver enzymes. Hepatology 2006;44:865-73.
10.	Söderberg C, Stål P, Askling J, Glaumann H, Lindberg G, Marmur J, et al. Decreased survival of subjects with elevated liver function tests during a 28-year follow-up. Hepatology 2010;51:595-602.
11.	Hong BS, Liu J, Zheng J, Ke W, Huang Z, Wan X, et al. Angiopoietin-like protein 8/betatrophin correlates with hepatocellular lipid content independent of insulin resistance in non-alcoholic fatty liver disease patients. J Diabetes Investig 2018;9:952-8.
12.	Kadomatsu T, Tabata M, Oike Y. Angiopoietin-like proteins: Emerging targets for treatment of obesity and related metabolic diseases. FEBS J 2011;278:559-64.
13.	Hato T, Tabata M, Oike Y. The role of angiopoietin-like proteins in angiogenesis and metabolism. Trends Cardiovasc Med 2008;18:6-14.
14.	Dijk W, Kersten S. Regulation of lipid metabolism by angiopoietin-like proteins. Curr Opin Lipidol 2016;27:249-56.
15.	Tabata M, Kadomatsu T, Fukuhara S, Miyata K, Ito Y, Endo M, et al. Angiopoietin-like protein 2 promotes chronic adipose tissue inflammation and obesity-related systemic insulin resistance. Cell Metab 2009;10:178-88.
16.	Odagiri H, Kadomatsu T, Endo M, Masuda T, Morioka MS, Fukuhara S, et al. The secreted protein angptl2 promotes metastasis of osteosarcoma cells through integrin alpha5beta1, p38 mapk, and matrix metalloproteinases. Sci Signal 2014;7:Ra7.
17.	Horio E, Kadomatsu T, Miyata K, Arai Y, Hosokawa K, Doi Y, et al. Role of endothelial cell-derived angptl2 in vascular inflammation leading to endothelial dysfunction and atherosclerosis progression. Arterioscler Thromb Vasc Biol 2014;34:790-800.
18.	Lu Z, He B, Chen Z, Yan M, Wu L. Anti-inflammatory activity of berberine in non-alcoholic fatty liver disease via the angptl2 pathway. BMC Immunol 2020;21:28.
19.	Kim I, Kim HG, Kim H, Kim HH, Park SK, Uhm CS, et al. Hepatic expression, synthesis and secretion of a novel fibrinogen/angiopoietin-related protein that prevents endothelial-cell apoptosis. Biochem J 2000;346(Pt 3):603-10.
20.	Inaba T, Matsuda M, Shimamura M, Takei N, Terasaka N, Ando Y, et al. Angiopoietin-like protein 3 mediates hypertriglyceridemia induced by the liver X receptor. J Biol Chem 2003;278:21344-51.
21.	Oike Y, Akao M, Yasunaga K, Yamauchi T, Morisada T, Ito Y, et al. Angiopoietin-related growth factor antagonizes obesity and insulin resistance. Nat Med 2005;11:400-8.
22.	Namkung J, Sohn JH, Chang JS, Park SW, Kim JY, Koh SB, et al. Increased serum angiopoietin-like 6 ahead of metabolic syndrome in a prospective cohort study. Diabetes Metab J 2019;43:521-9.
23.	Sharma V, Pangtey GS, Gupta R, Rehan HS, Gupta LK. Correlation of long-term glycemic control as measured by glycated hemoglobin with serum angiopoietin-like 6 protein levels in type 2 diabetes mellitus patients. Indian J Pharmacol 2017;49:250-3. [PUBMED] [Full text]
24.	Mandelia C, Collyer E, Mansoor S, Lopez R, Lappe S, Nobili V, et al. Plasma cytokeratin-18 level as a novel biomarker for liver fibrosis in children with nonalcoholic fatty liver disease. J Pediatr Gastroenterol Nutr 2016;63:181-7.
25.	Shirabe K, Takeishi K, Taketomi A, Uchiyama H, Kayashima H, Maehara Y. Improvement of long-term outcomes in hepatitis C virus antibody-positive patients with hepatocellular carcinoma after hepatectomy in the modern era. World J Surg 2011;35:1072-84.
26.	Angulo P, Kleiner DE, Dam-Larsen S, Adams LA, Bjornsson ES, Charatcharoenwitthaya P, et al. Liver fibrosis, but no other histologic features, is associated with long-term outcomes of patients with nonalcoholic fatty liver disease. Gastroenterology 2015;149:389-97.e310.
27.	Erkan G, Muratoglu S, Ercin U, Bilgihan A. Angiopoietin-like protein 2 and angiopoietin-like protein 6 levels in patients with nonalcoholic fatty liver disease. Arch Med Sci 2018;14:781-7.
28.	Sharma DL, Lakhani HV, Klug RL, Snoad B, El-Hamdani R, Shapiro JI, et al. Investigating molecular connections of non-alcoholic fatty liver disease with associated pathological conditions in west virginia for biomarker analysis. J Clin Cell Immunol 2017;8:523.
29.	Oike Y, Akao M, Kubota Y, Suda T. Angiopoietin-like proteins: Potential new targets for metabolic syndrome therapy. Trends Mol Med 2005;11:473-9.
30.	Hassan M. ANGPLT3: A novel modulator of lipid metabolism. Glob Cardiol Sci Pract 2017;2017:e201706.
31.	Li Y, Teng C. Angiopoietin-like proteins 3, 4 and 8: Regulating lipid metabolism and providing new hope for metabolic syndrome. J Drug Target 2014;22:679-87.
32.	Yilmaz Y, Ulukaya E, Atug O, Dolar E. Serum concentrations of human angiopoietin-like protein 3 in patients with nonalcoholic fatty liver disease: Association with insulin resistance. Eur J Gastroenterol Hepatol 2009;21:1247-51.
33.	Yoneda M, Yoneda M, Fujita K, Inamori M, Tamano M, Hiriishi H, et al. Transient elastography in patients with non-alcoholic fatty liver disease (NAFLD). Gut 2007;56:1330-1.
34.	Harrison SA, Oliver D, Arnold HL, Gogia S, Neuschwander-Tetri BA. Development and validation of a simple NAFLD clinical scoring system for identifying patients without advanced disease. Gut 2008;57:1441-7.
35.	Sumida Y, Yoneda M, Hyogo H, Yamaguchi K, Ono M, Fujii H, et al. A simple clinical scoring system using ferritin, fasting insulin, and type IV collagen 7S for predicting steatohepatitis in nonalcoholic fatty liver disease. J Gastroenterol 2011;46:257-68.
36.	Wong VW, Adams LA, de Lédinghen V, Wong GL, Sookoian S. Noninvasive biomarkers in NAFLD and NASH-current progress and future promise. Nat Rev Gastroenterol Hepatol 2018;15:461-78.