Serum metabolomics analysis in patients with alcohol dependence

Yanjie Zhang; Yajun Sun; Qin Miao; Shilong Guo; Qi Wang; Tianyuan Shi; Xinsheng Guo; Shuai Liu; Guiding Cheng; Chuansheng Wang; Ruiling Zhang

doi:10.3389/fpsyt.2023.1151200

Serum metabolomics analysis in patients with alcohol dependence

Front Psychiatry. 2023 Apr 17:14:1151200. doi: 10.3389/fpsyt.2023.1151200. eCollection 2023.

Authors

Yanjie Zhang^{1

2}, Yajun Sun^{1

3}, Qin Miao^{1

4}, Shilong Guo⁵, Qi Wang^{1

2}, Tianyuan Shi^{1

2}, Xinsheng Guo^{1

2}, Shuai Liu^{1

2}, Guiding Cheng^{1

2}, Chuansheng Wang^{1

2}, Ruiling Zhang^{1

2}

Affiliations

¹ Department of Psychiatry, Henan Mental Hospital, The Second Affiliated Hospital of Xinxiang Medical University, Xinxiang, China.
² Henan Key Lab of Biological Psychiatry, Xinxiang Medical University, Xinxiang, China.
³ Department of Scientific Research, The Second Affiliated Hospital of Xinxiang Medical University, Xinxiang, China.
⁴ Department of Addiction, The Second Affiliated Hospital of Xinxiang Medical University, Xinxiang, China.
⁵ Department of Oncology, The Third Affiliated Hospital of Xinxiang Medical University, Xinxiang, China.

Abstract

Objective: Alcohol dependence (AD) is a chronic recurrent mental disease caused by long-term drinking. It is one of the most prevalent public health problems. However, AD diagnosis lacks objective biomarkers. This study was aimed to shed some light on potential biomarkers of AD patients by investigating the serum metabolomics profiles of AD patients and the controls.

Methods: Liquid chromatography-mass spectrometry (LC-MS) was used to detect the serum metabolites of 29 AD patients (AD) and 28 controls. Six samples were set aside as the validation set (Control: n = 3; AD group: n = 3), and the remaining were used as the training set (Control: n = 26; AD group: n = 25). Principal component analysis (PCA) and partial least squares discriminant analysis (PCA-DA) were performed to analyze the training set samples. The metabolic pathways were analyzed using the MetPA database. The signal pathways with pathway impact >0.2, value of p <0.05, and FDR < 0.05 were selected. From the screened pathways, the metabolites whose levels changed by at least 3-fold were screened. The metabolites with no numerical overlap in their concentrations in the AD and the control groups were screened out and verified with the validation set.

Results: The serum metabolomic profiles of the control and the AD groups were significantly different. We identified six significantly altered metabolic signal pathways, including protein digestion and absorption; alanine, aspartate, and glutamate metabolism; arginine biosynthesis; linoleic acid metabolism; butanoate metabolism; and GABAergic synapse. In these six signal pathways, the levels of 28 metabolites were found to be significantly altered. Of these, the alterations of 11 metabolites changed by at least 3-fold compared to the control group. Of these 11 metabolites, those with no numerical overlap in their concentrations between the AD and the control groups were GABA, 4-hydroxybutanoic acid, L-glutamic acid, citric acid and L-glutamine.

Conclusion: The metabolite profile of the AD group was significantly different from that of the control group. GABA, 4-hydroxybutanoic acid, L-glutamic acid, citric acid, and L-glutamine could be used as potential diagnostic markers for AD.

Keywords: alcohol dependence; liquid chromatography-mass spectrometry (LC–MS); metabolic signal pathways; multivariate statistical analysis; potential biomarkers.