Insulin resistance is positively associated with plasma cathepsin D activity in NAFLD patients

Subjects characteristics

Fifty-five patients were enrolled in our study including 29 men and 26 women, with an age ranging from 20 to 65 years and BMI ≥25 kg/m². All subjects were proven NAFLD by liver biopsy (based on the criteria of NAS (NAFLD activity score)) or chemical shift MRI (magnetic resonance imaging MRI). Metabolic syndrome (MetS) was determined according to the definition of International Diabetes Federation (IDF) [7]. Subjects were selected based on exclusion criteria including consuming excessive ethanol (male ≥ 14 and female ≥ 7 standard beverages in one week), suffering from secondary hepatic fat accumulation (causes as previously described [8]), a history of bariatric surgery, liver cirrhosis and/or hepatocellular carcinoma as well as extrahepatic malignancy(s) within the past 5 years. Besides, subjects who are pregnant and breastfeeding, use insulin and are about to undergo or recovering from a surgical /medical procedure were also ruled out from our study. All participants were collected from Maastricht University Medical Centre (MUMC+) and CO-EUR (a second line eating disorder clinic) during the period September 2015-October 2018.

Informed consent: Informed consent has been obtained from all individuals included in this study.

Ethical approval: The research related to human use has been complied with all the relevant national regulations, institutional policies and in accordance the tenets of the Helsinki Declaration (revised in 2008), and approved by the Medical-Ethical Committee of Maastricht University with ClinicalTrials.gov Identifier: NCT02422238.

Biochemical analyses

Blood samples were drawn after overnight fasting at Maastricht University Medical Center (MUMC+). All blood samples were centrifuged to yield plasma (1000 × g; 10 min; 4°C) and then stored at −80°C freezer until analyses. Plasma samples were further analyzed at the clinical chemistry department of Maastricht UMC+ hospital to determine aspartate aminotransferase (AST), alanine aminotransferase (ALT), γ-glutamyl transpeptidase (GGT), bilirubin, alkaline Phosphatase, total cholesterol, triglyceride, low density lipoprotein cholesterol (LDL-cholesterol) and high density lipoprotein cholesterol (HDL-cholesterol) as previously described [8]. Besides, fasting plasma glucose, hemoglobin A1c (HbA1C) and plasma insulin levels were evaluated using routine laboratory methods. Insulin resistance was calculated by (fasting plasma insulin (mU/L) × fasting plasma glucose (mmol/L)) / 22.5 according to the homeostasis model of assessment-insulin resistance (HOMA-IR) [9]. Plasma CTSD activity was evaluated using a CTSD activity assay kit by following manufacturer's protocol (MBL International, Woburn, USA) as previously described [4]. To each assay, 5 μL 10 × diluted plasma sample was added in black 96-well plate and 52 μL mastermix (50 μL CTSD reaction buffer and 2 μL CTSD substrate) was added. After mixing, the 96-well plate was incubated at 37°C for 2 h (protection from light) and then the plate was measured via using a fluorescence plate reader with 328-nm excitation wavelength and 460-nm emission wavelength. The activity of plasma CTSD was expressed by the relative fluorescence units.

Magnetic Resonance Imaging

The hepatic fat fraction (steatosis) was determined by using chemical shift gradient echo MRI. Four circular region of interests (ROIs) of 5 cm² from three MRI sections were drawn in the liver, where artefact, vascular and biliary structures were avoided. Then, the ROI was copied from the in-phase (IP) image to the opposed-phase (OP) image and the average ROI signal intensity (SI) for the IP and OP image were calculated. At the end, the mean signal intensity loss of all 12 ROIs was determined by using the formula: (SI IP- SI OP)/ 2*SI IP*100% [10].

Data statistics

Statistical analyses were performed by using SPSS 25.0 (IBM, Armonk, NY, USA) for Microsoft Windows. All data were expressed as mean ± SD. A bivariate correlation was used to analyze the simple correlations between plasma CTSD activity and other parameters. Then multiple linear regression analyses were conducted for analyzing the association between plasma CTSD activity and insulin/HOMA-IR, where plasma CTSD activity was added as dependent variable and either HOMA-IR or insulin as independent variables generating model 1 (simple regression), model 2 (model 1 + correcting for age), model 3 (model 2 + correcting for sex), model 4 (model 3 + correcting for BMI) and model 5 (model 4 + correcting for waist circumference). p-value <0.05 was statistically significant. Interactions between co-variates (age, sex, BMI and waist circumference) in the multiple regression analyses were also tested but no significant interactions were found (data not shown).

General clinical characteristics of the study participants

The baseline characteristics of the population are listed in Table 1. Fifty-five NAFLD patients were included in our study, including 29 males and 26 females with an average age of 51.4 years old and mean BMI of 32.3 Kg/m². The hepatic fat fraction (%) ranged from 5.2 to 45.5%. Furthermore, plasma CTSD activity ranged from 93.9 RFU/μL to 324.3 RFU/μL, with a mean of 179.6 ± 58.9 RFU/μL.

HOMA-IR and plasma insulin levels are positively associated with plasma CTSD activity

To investigate whether insulin resistance-related parameters correlate with plasma CTSD activity in NAFLD patients, multiple linear regression analyses were performed. As shown in Table 2, HOMA-IR was positively associated with plasma CTSD activity (Model 1: standardized β=0.413, 95% Cl: 0.156~0.667, p=0.002), even after adjustment for age (Model 2: standardized β=0.417, 95%Cl: 0.162~0.669, p=0.002), sex (Model 3: standardized β=0.394, 95%Cl: 0.129~0.656, p=0.004), BMI (Model 4: standardized β=0.410, 95%Cl: 0.143~0.674, p=0.003) and waist circumference (Model 5: standardized β=0.412, 95%Cl: 0.142~0.679, p=0.004). Furthermore, we found that plasma insulin levels were also positively associated with plasma CTSD activity (Model 1: standardized β=0.472, 95%Cl: 0.225~0.723, p=0.000), independent of age (Model 2: standardized β=0.487, 95%Cl: 0.243~0.735, p=0.000), sex (Model 3: standardized β=0.468, 95%Cl:0.214~0.726, p=0.001), BMI (Model 4: standardized β=0.493, 95%Cl: 0.237~0.753, p=0.000) and waist circumference (Model 5: standardized β=0.495, 95%Cl: 0.236~0.758, p=0.000). Our data therefore demonstrate that insulin resistance is independently associated with plasma CTSD activity in the context of NAFLD.