LC-MS Based Serum And Tissue Metabolomic Studies On Hepatocellular Carcinoma

Hepatocellular carcinoma(HCC)is one of the most common malignancies.Because of its concealed pathogenesis,HCC is always diagnosed in the advanced stage that brings a poor prognosis after treatment.Thus,it is of great significance to explore new methods for the early diagnosis of HCC.Meanwhile,as the boomingly developed branch in system biology,metabolomics could sensitively,systematically and dynamically characterize the global metabolic alterations of small-molecular in diversely stimulated organisms based on the high-throughput analytical approaches.Therefore,this study acquired the metabolic profiles from sera and tissues by using liquid-chromatography coupled to mass spectrometry(LC-MS).Uni-variate and multi-variate analysis methods were then utilized to select the serum discriminative metabolites with robust diagnostic performances.Furthermore,metabolic pathway and network analysis were performed to investigate the latent relationship between the differential metabolites.On this basis,cellular and molecular biological approaches were employed to preliminarily explore the mechanisms responsible for the perturbations of related pathways.High-throughput data was obtained using LC-MS from 587 sera including 267 HCC cases and 320 control subjects.Principal components analysis(PCA)and partial least square discriminant analysis(PLS-DA)models were then constructed and demonstrated the differences between metabolic profiles of HCC and controls.Furthermore,87 differential metabolites were selected via random forest classification and further evaluated by logistic regression analysis,Pearson correlation analysis and receiver operating characteristic(ROC)analysis based on the validation dataset.Consequently,D-galactose,undecanoyl-L-carnitine and PE(P-18:0/0:0)were selected as the discriminative metabolites which had sensitivity,specificity and uni-variate AUC of 0.96,0.92 and 0.98 via combination with alphafetoprotein(AFP).Besides,the combination was able to discriminate the cases of AFP-negative by the accuracy of 78%,which demonstrated good performances of the discriminative metabolites in the assistance of AFP.Moreover,metabolic pathway and network analysis were conducted based on the differential metabolites.It was found that the acylcarnitines scattered according to both the length and the degree of unsaturation of the acyl-chain,which could provide valuable information to the further study upon related mechanisms.On the basis of the dramatic differences between metabolic profiles of HCC and control sera,tissue metabolomics studies were carried out as well.First of all,binary-phase extraction based tissue preparation and LC-MS based analytical method were both established to obtain the polar metabolic profiles of 80 pairs of HCC and adjacent noncancerous tissues(ANTs).By using orthogonal projection to latent structure discriminant analysis(OPLS-DA)model,29 differential metabolites were selected and further evaluated in the logistic regression model,in which the affections of clinical confounding factors such as gender,age and pathological stage to the differential metabolites were characterized.Besides,to investigate the origin s of the metabolic alterations in serum metabolic profiles,Spearman correlation analysis was performed to explore the underlying relationship between the serum and tissue differential metabolites.According to the findings above,the global decreases of short-and mediumchain acylcarnitines(S/MCACs)and increases of long-chain acylcarnitines(LCACs)were observed in both serum and tissue polar metabolic profiles.To investigate related regulation and mechanism,we performed metabolic target analysis upon the metabolic profiles of 26 acylcarnitines in polar metabolomes of 80 pairs of tissues.The expressions of carnitine palmitoyltransferases(CPTs)which regulate the metabolism of acylcarnitines were analyzed with molecular biological techniques including quantitative polymerase chain reaction(q PCR)and western blot analysis.As a result,CPT1 A revealed no significant difference between HCCs and ANTs whereas CPT2 was down-regulated in HCCs compared to ANTs,which indicated that the decreased expression of CPT2 might be the cause of the opposite trends between S/MCACs and LCACs.To validate this deduction,we performed metabolic target analysis upon acylcarnitines in BEL-7402/CPT2-knockdown and negative control cells,and found that S/MCACs and LCACs respectively revealed decrea sed and increased levels in BEL-7402/CPT2-knockdown cells,which was consistent with the results in tissue acylcarntine profiles.Therefore,the down-regulation of CPT2 might be one of the inducements responsible for the opposite trends of acylcarnitines with different chain-lengths.In addition to the polar metabolic profiling analysis,we optimized the method for lipidomic analysis and further systematically characterized the alterations of lipidomic profiles between HCC and ANT tissues.The high-throughput lipidomic data was then used to construct the PLS-DA models for the discrimination of HCC and ANT tissues and 113 differential lipids were finally identified.In the following metabolic topological analysis,glycerophospholipid metabolism,sphingolipid metabolism and glycerolipid metabolism showed significant changes according to the criteria including pathway impact larger than 0.1 and the p-value of coverage significance less than 0.05.These results suggested that the enzymes in these pathways were greatly influenced.More specifically,the metabolic pathway analysis showed that phosphatidylethanolamine(PE)decreased and phosphatidylserine(PS)increased in HCC tissues compared to ANTs,which were agreed to the higher expression of phosphatidylserine synthase 2(PTDSS2)that catalyzes the conversion of PE into PS.Similarly,the decreases in triacylglycerols(TGs)and increases in diacylglycerols(DGs)were also coincidence with the up-regulated expression of lipoprteinlipase(LPL)in HCC tissues suggesting that the altered expressions of corresponding enzymes might be responsible for the disturbances of metabolic profiles.Particularly,as the second messenger in the phosphatidylinosital biphosphate(PIP2)cellular signaling pathway,DG is always playing a key role in the cellular signaling in tumor cells.To sum up,we obtained the discriminative metabolites with satisfactory diagnostic performances with the employment of the metabolomics strategy adopted in this study.Meanwhile,in combination with the cellular and molecular biological approaches,the mechanism responsible for the alterations of ac ylcarnitines based on the tissue metabolomics platform was investigated.At last,to explore the systematic changes of lipids in HCC tissues,metabolic topological and pathway analysis were conducted on the tissue lipidomics data.Therefore,the present study was capable of providing valuable information for clinical diagnosis and mechanism studies on HCC.