The Role Of Hippuric Acid In The Prognostic Evaluation Of Primary Hepatocellular Carcinoma And Its Mechanism

Study Background and Aims Hepatocellular carcinoma(HCC)is currently the sixth most common cancer,and the third leading cause of cancer-related deaths in the world.The newest data revealed that among all malignancies reported in China,the incidence and mortality of HCC rank fourth and third,respectively,which seriously endangering the life and health of people.Hepatectomy is the most important treatment for patients with HCC to obtain long-term survival,while recurrence after surgery restricts therapeutic effect.HCC shows a high tendency to vascular invasion.Tumor cells invade the main portal vein or its branches to form portal vein tumor thrombus(PVTT),which is an important sign of HCC progression.The key to further improving the therapeutic effect of liver resection depends on the exploration of the mechanism of occurrence and development of HCC.Metabolites are small molecules that undergo chemical transformation during metabolism and can directly reflect changes in the organism.Studying all small molecule metabolites in biological systems through metabolomics can provide overall information on the dynamic metabolic reactions of living systems to endogenous and exogenous factors.This study explored the application of metabonomics detection platform based on ultra performance liquid chromatography-mass spectrometry(UPLS-MS)technology in HCC.The study aimed to find and evaluate the predictive value of characteristic metabolic molecules for postoperative prognosis of HCC,and to further study the mechanism of hippuric acid and its key regulatory enzymes in the development of HCC.In the first part,we detected postoperative tissue samples of patients with HCC and PVTT.We found the characteristic metabolite of hippuric acid,and use targeted metabolomics to quantitatively verify the expression content of hippuric acid in different tissues;the second part of the study explored the predictive value of hippuric acid in the prognosis of HCC.We used targeted metabolomics to quantitatively detect the content of hippuric acid in HCC tissues,and analyzed correlation between hippuric acid level and tumor pathological characteristics.The level of hippuric acid was incorporated into the multivariate analysis to screen for risk factors related to overall survival and recurrence-free survival.Nomograms were constructed based on the multivariate analysis to accurately predict the long-term prognosis after liver resection.The third part of the study further explored the mechanism of differential metabolite hippuric acid in HCC.The expression of hippuric acid was detected in HCC cell lines,and themechanism of HCC development was explored by using public database combined with molecular biology research strategy.We intended to explore the mechanism of hippuric acid and its key enzymes in regulating the progression of hepatocellular carcinoma,so as to find potential prognostic markers and therapeutic targets.Study Methods In the first part,we established a tissue sample analysis method by ultra-high performance liquid chromatography mass spectrometry technology platform.Tissue samples of 50 HCC patients combined with PVTT were collected after surgery.Metabolomics analysis was used to analyze HCC tissue,proximal adjacent tumor tissues(<2cm),distal adjacent tumor tissues(? 2cm)and portal vein tumor thrombus tissues of each patient.An orthogonal partial least squares discriminant analysis(OPLS-DA)model was established and the model interpretation rate and prediction rate of the OPLS-DA model were calculated.The reliability and stability of the OPLS-DA model were verified by internal validation methods.The differential metabolites were screened according to the screening conditions.The primary and secondary mass spectra of the differential metabolites obtained from the above screening were compared with the spectra in the public metabolism database to identify the types of differential expressed metabolites.Finally,the targeted metabolomics was further adopted to quantitatively detect the level of the characteristic metabolite in four tissues to confirm the changes in the metabolites in HCC.In the second part,the tissue samples of 426 HCC patients who underwent hepatectomy were subjected to targeted metabolomics testing to quantitatively determine the level of hippuric acid in HCC tissues,by using the above-mentioned targeted metabonomics method.We used X-tile software to explore the optimal cutoff value of hippuric acid for prognostic prediction.The correlation between the expression of hippuric acid and the pathological characteristics of patients was analyzed.All patients are divided into training set and validation set according to the principle of complete randomization.Data of the training set is used for variables screening and model construction,and the prediction performance of the model is validated in the validation set.Cox regression analysis was used to determine the risk factors that affect OS and RFS in HCC patients.The OS and RFS-related risk factors were used as predictors to construct prediction models.C-index and calibration curve in the training set and the validation set were obtained to evaluate the discrimination and calibration of the prediction model.Decision curve analysis was used to evaluate the clinical benefit of the prediction models.Based on the predictive value of the nomograms,the risk stratification of all patients was carried out,and the prognosis difference of patients was compared.In the third part,targeted metabolomics was used to detect the expression of hippuric acid in HCC cell lines and normal liver cell line.The effect of hippuric acid on the biological phenotype of HCC cells were explored.Public databases and metabolic pathway databases were searched to identify key enzymes Glycine N-acyltransferase(GLYAT)that regulate expression of hippuric acid.We further explored the expression of key enzymes that regulate hippuric acid in HCC through the TCGA public database,and analyzed its expression level with tumor characteristics and patient prognosis.We explored the function of GLYAT in HCC cell lines,and constructed GLYAT overexpression and knockdown cell lines using lentivirus infection.We performed molecular functional experiments to explore the effect of GLYAT on the biological phenotype of HCC cells and downstream signaling pathways.On the basis of the mechanism,the role of GLYAT on the occurrence and development of HCC was further explored through in vivo animal model.Study Results In the first part,non-targeted metabolic profiling was performed on four kinds of tissue samples from 50 patients with HCC with portal vein tumor thrombus.The orthogonal partial least squares discriminant analysis(OPLS-DA)model was established with model interpretation rate R2 Y of 0.726 and prediction rate Q2 of 0.708.Internal validation revealed that the model fits well,indicating that the OPLS-DA model is stable and effective.Finally,a total of 17 metabolites with significant differences in expression in HCC,proximal adjacent cancer,distal adjacent cancer and PVTT tissues were identified,through comparison with the metabolic profile data from public databases.These metabolites belong to amino acids,purines,organic acids,bile acids,weak acids,ketones,saturated and unsaturated fatty acids.Metabolic pathway analysis showed that differential metabolites were mainly enriched in the following pathways: primary bile acid biosynthesis,taurine and hypotaurine metabolism,purine metabolism and phenylalanine metabolism.Analysis of metabolic pathways and networks shows that the differential expressed metabolites are mainly concentrated in the following pathways: primary bile acid biosynthesis,taurine and subtaurine metabolism,purine metabolism,and phenylalanine metabolism.We further compared the relative content of each differential metabolite in different groups and found that the content of hippuric acid gradually decreased from to distal adjacent tumor tissues(> 2cm),proximal adjacent tumor tissues(<2cm),HCC tissues,to portal vein tumor thrombus tissues.We used the targeted metabolomics method to quantitatively detect the content of hippuric acid in four tissues,and analyze the differences in its content in distal adjacent tumor tissues,proximal adjacent tumor tissues,HCC tissues and portal vein tumor thrombus tissues.The results showed that the changing trend of hippuric acid content was consistent with the foregoing.Therefore,hippuric acid was identified by non-targeted and targeted metabonomics,which is a characteristic differential metabolite in HCC with portal vein tumor thrombus.In the second part,a total of 426 HCC patients were enrolled into analysis.Targeted metabolomics was used to quantitatively detect the level of hippuric acid in the HCC tissues of the enrolled patients.X-tile software was used to determine the optimal cutoff value to distinguish high and low expression of hippuric acid.The optimal cutoff value was 0.215 ng/m L.All patients were divided into training set(N=286)and validation set(N=140)according to the principle of entirely randomization.In the training set,the Cox multivariate regression analysis method was used to study the factors affecting the OS of HCC patients.The results showed that the preoperative high AFP level,low hippuric acid expression,maximum tumor diameter,multiple tumors,macrovascular invasion,intraoperative blood transfusion were risk factors affecting OS in patients with HCC;The risk factors associated RFS in patients with HCC included Child-Pugh score,low hippuric acid expression,maximum tumor diameter,tumor number,large vessel invasion,and resection margin ? 1 cm.The OS and RFS-related risk factors were adopted as predictors included into the nomograms to obtain 1,3,and 5 years of OS or RFS survival probabilities.The C-index of the OS nomogram prediction model is 0.763 and 0.795 in the training set and validation set,respectively,and the C-index of the RFS nomogram prediction model is 0.750 and 0.765 in the training set and validation set,respectively.Calibration curves were plotted to evaluate the consistency of OS and RFS nomogram prediction models in the training set and validation set.The results showed that the predicted results fit well with the actual results,indicating that the consistency of the two prediction models is ideal.Decision curve analysis shows that OS and RFS nomogram prediction models have better clinical benefits than BCLC staging and TNM staging.In the third part of the study,we adopted targeted metabonomics for detection of hippuric acid content in HCC cell lines showed that its expression in HCC cell lines was lower than that in normal liver cell lines,and hippuric acid directly treated HCC cells can inhibit the malignant phenotype of HCC cell lines.The search results in the metabolic network database and the KEGG database showed that hippuric acid was the end product of the phenylalanine metabolic pathway,and its metabolic pathway was directly regulated by the key enzyme Glycine N-acyltransferase(GLYAT).Therefore,the level of hippuric acid is closely related to GLYAT.Based on the above results,we speculate that GLYAT may play an important role in HCC.The public database TCGA data was searched to analyze the expression of GLYAT in HCC.The results showed that the expression of GLYAT in HCC tissues was significantly lower than that in normal liver tissues.The HCC tissues and adjacent tissues of 67 patients with HCC were further collected,and the PVTT tissues of 11 HCC patients were also collected.The results showed that the expression of GLYAT gradually decreased in adjacent liver tissues,HCCtissues and portal vein tumor thrombus tissues.In addition,the expression of GLYAT is related to the presence of satellite lesions and microvascular invasion,that is,a lower expression level indicates a higher degree of malignancy of the tumor.Multivariate Cox regression analysis showed that low GLYAT expression was an independent risk factor for poorer OS and RFS.Similar results were also observed in HCC cell lines,and GLYAT expression was reduced in HCC cell lines compared with normal liver cells(LO2).The results suggest that up-regulation of GLYAT can significantly inhibit the proliferation and migration of HCC cells,while down-regulation of GLYAT can significantly increase the proliferation and migration of HCC cells.In vivo studies also confirmed that down-regulation of GLYAT expression promoted the growth of subcutaneous tumors in nude mice.Epithelial-mesenchymal transition(EMT)is a core part of tumor invasion and metastasis,which is often activated during tumor invasion and metastasis.The study further examined the changes of EMT-related proteins in HCC cell lines after GLYAT overexpression and knockdown.It was found that both epithelial marker proteins(E-cadherin)and interstitial marker proteins(N-cadherin and Vimentin)were changed.Overexpression of GLYAT increased expression of N-cadherin and Vimentin,while expression of E-cadherin was decreased.In GLYAT knockdown HCC cells,the expression of N-cadherin and Vimentin protein decreased,while the expression of Ecadherin increased.In mechanism research part,we examined the expression of various classical signaling pathway-related proteins after GLYAT overexpression and knockdown.The results showed that overexpression of GLYAT down-regulated the phosphorylation levels of PI3 K,Akt and m TOR,thereby inhibiting the activation of the PI3K/Akt/m TOR signaling pathway.Down-regulation of GLYAT activated the PI3K/Akt/m TOR pathway.Treatment with m TOR kinase inhibitor PP242 could reverse the malignant phenotype of GLYAT knockdown HCC cells,inhibiting cell invasion and migration caused by down- regulation of GLYAT.This result suggestes that down-regulation of GLYAT expression promotes the EMT process in HCC cells by activating the PI3K/Akt/m TOR signaling pathway,thereby promoting the invasion and migration of HCC cells.The above results suggest that GLYAT may be a potential prognostic biomarker for HCC and a key target for regulating HCC progression.Study Conclusion To sum up,this study established a non-targeted metabolomics analysis method for HCC tissues and detected postoperative samples of patients with HCC and PVTT.Differentially expressed metabolites during the occurrence and progression of HCC were identified.Quantitative detection using targeted metabonomics revealed that the content of hippuric acid gradually reduced from distal adjacent tumor tissues to proximal adjacent tumor tissues,to HCC tissues,to portal vein tumor thrombus tissues.In the second part,based on the targeted metabonomics analysis method established in the first part,quantitative detection of hippuric acid in HCC samples was carried out,and the optimal cut-off value of hippuric acid for prognosis prediction was determined.Multivariate Cox regression analysis confirmed that hippuric acid content was an independent risk factor for overall survival and relapse free survival,and nomograms integrating hippuric acid content was established to accurately predict the prognosis of HCCs.Finally,we explored the mechanism of hippuric acid in HCC progression.Through searching the public database data,it was confirmed that GLYAT was the key enzyme to catalyze formation of hippuric acid.Furthermore,we explored the biological function of GLYAT in HCC and its molecular mechanism in the progression of HCC,and explored the changes of related signal pathways.Clinical sample analysis confirmed that GLYAT is a potential prognostic marker for HCC.GLYAT can inhibit the proliferation and migration of HCC.Down regulation of GLYAT could promote the EMT process in HCC by activating PI3K/Akt/m TOR signaling pathway and improve the invasion and migration ability of HCC cells.Therefore,hippuric acid and its key enzymes may serve as potential biomarkers of HCC progression,and may be a new target of treatment intervention.