Author information
1The Health & Nutrition Innovative International Research Center, Department of Epidemiology, Biostatistics and Community Health Sciences, Faculty of Health Sciences, School of Public Health, Ben-Gurion University of the Negev, Beer-Sheva, Israel.
2Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA, USA.
3Department of Medicine, University of Leipzig, Leipzig, Germany.
4Institute of Laboratory Medicine, Clinical Chemistry, and Molecular Diagnostics, University of Leipzig Medical Center, Leipzig, Germany.
5Department of Diagnostic Imaging, Soroka University Medical Center, Ben-Gurion University of the Negev, Beer-Sheva, Israel.
6Helmholtz Institute for Metabolic, Obesity and Vascular Research (HI-MAG) of the Helmholtz Zentrum München at the University of Leipzig and University Hospital Leipzig, Leipzig, Germany.
7Department of Gastroenterology and Liver Diseases, Soroka University Medical Center, Beersheba, Israel.
8Department of Nutrition, Harvard T.H. Chan School of Public Health, Boston, MA, USA.
9Department of Medicine, Channing Division of Network Medicine, Abstract
Abstract
Background and aims: We demonstrated in the randomized 18-month DIRECT PLUS trial (n = 294) that a Mediterranean (MED) diet, supplemented with polyphenol-rich Mankai duckweed, green tea, and walnuts and restricted in red/processed meat, caused substantial intrahepatic fat (IHF%) loss compared with 2 other healthy diets, reducing NAFLD by half, regardless of similar weight loss. Here, we investigated the baseline proteomic profile associated with IHF% and the changes in proteomics associated with IHF% changes induced by lifestyle intervention.
Approach and results: We calculated IHF% by proton magnetic resonance spectroscopy (normal IHF% <5% and abnormal IHF% ≥5%). We assayed baseline and 18-month samples for 95 proteomic biomarkers.Participants (age = 51.3 ± 10.8 y; 89% men; and body mass index = 31.3 ± 3.9 kg/m 2 ) had an 89.8% 18-month retention rate; 83% had eligible follow-up proteomics measurements, and 78% had follow-up proton magnetic resonance spectroscopy. At baseline, 39 candidate proteins were significantly associated with IHF% (false discovery rate <0.05), mostly related to immune function pathways (eg, hydroxyacid oxidase 1). An IHF% prediction based on the DIRECT PLUS by combined model ( R2 = 0.47, root mean square error = 1.05) successfully predicted IHF% ( R2 = 0.53) during testing and was stronger than separately inputting proteins/traditional markers ( R2 = 0.43/0.44). The 18-month lifestyle intervention induced changes in 18 of the 39 candidate proteins, which were significantly associated with IHF% change, with proteins related to metabolism, extracellular matrix remodeling, and immune function pathways. Thrombospondin-2 protein change was higher in the green-MED compared to the MED group, beyond weight and IHF% loss ( p = 0.01). Protein principal component analysis revealed differences in the third principal component time distinct interactions across abnormal/normal IHF% trajectory combinations; p < 0.05 for all).
Conclusions: Our findings suggest novel proteomic signatures that may indicate MRI-assessed IHF state and changes during lifestyle intervention. Specifically, carbonic anhydrase 5A, hydroxyacid oxidase 1, and thrombospondin-2 protein changes are independently associated with IHF% change, and thrombospondin-2 protein change is greater in the green-MED/high polyphenols diet.