Background: Metabolomics is one of the main areas to understand cellular process at molecular
level by analyzing metabolites. In recent years metabolomics has emerged as a key tool to understand
molecular basis of diseases, to find diagnostic and prognostic biomarkers and develop new
treatment opportunities and drug molecules.
Objective: In this study, untargeted metabolite and lipid analysis were performed to identify potential
biomarkers on premature ovarian insufficiency plasma samples. 43 POI subject plasma samples were
compared with 32 healthy subject plasma samples.
Methods: Plasma samples were pooled and extracted using chloroform:methanol:water (3:3:1 v/v/v)
mixture. Agilent 6530 LC/MS Q-TOF instrument equipped with ESI source was used for analysis. A
C18 column (Agilent Zorbax 1.8 μM, 50 x 2.1 mm) was used for separation of the metabolites and lipids.
XCMS, an “R software” based freeware program, was used for peak picking, grouping and comparing
the findings. Isotopologue Parameter Optimization (IPO) software was used to optimize XCMS parameters.
The analytical methodology and data mining process were validated according to the literature.
Results: 83 metabolite peaks and 213 lipid peaks were found to be in semi-quantitatively and statistically
different (fold change >1.5, p <0.05) between the POI plasma samples and control subjects.
Conclusion: According to the results, two groups were successfully separated through principal component
analysis. Among the peaks, phenyl alanine, decanoyl-L-carnitine, 1-palmitoyl lysophosphatidylcholine
and PC(O-16:0/2:0) were identified through auto MS/MS and matched with human metabolome
database and proposed as plasma biomarker for POI and monitoring the patients in treatment period.