Pathogenesis Of Oral Lichen Planus Based On Meta Bolomic And Transcriptomic Association Analysis

AimsMetabolic biomarkers related to erosive and reticulated oral lichen planus(OLP)were discovered by non-targeted metabolomic methods.The diagnostic models of erosive and reticulated OLP were constructed respectively,which can be used to predict clinical outcomes in the future.This study will provide a scientific theoretical basis for the classification changes and early diagnosis of OLP.The metabolites and gene expression were analyzed by the gene expression database to reveal the pathological network pathway of OLP from the level of genes and metabolism.It provides new ideas for the study of the pathogenesis of OLP.MethodsA total of 153 subjects were included in this research,including 50 patients with erosive oral lichen planus(EOLP),51 patients with reticulated oral lichen planus(ROLP),and 52 healthy controls(HC).A total of 10 EOLPs,11 ROLPs and 12 HCs were selected for the validation group.Ultra-high performance liquid chromatography-quadrupole/electrostatic field orbitrap tandem high-resolution mass spectrometer(UHPLC-Q-Orbitrap HRMS)was used to analyze the metabolite components of 40 EOLPs,40 ROLPs and 40 HCs.Differential metabolic markers among the three groups were screened and identified,then diagnostic models among the three groups was constructed.The diagnostic efficiency of biomarkers was verified in the validation group.The related genes were further screened by the shared metabolites between EOLP and ROLP.The key differential genes were found by cross-correlation shared metabolite-regulated genes with OLP-related differential genes in the network database.Finally,the OLP "gene-metabolite" co-expression network was constructed.Results1.The analysis of population baseline characteristics showed that there was no significant difference of EOLP,ROLP and HC(P>0.05).The number of females in the diseased group was more than that of the healthy group with insufficient sleep,and they preferred hot food(P>0.05).The ROLP group accounted for the largest proportion of smokers(P>0.05).There was no significant difference in the number of people with drinking history,lack of exercise,preference for spicy food and bad chewing habits.2.This study was based on UHPLC-Q-Orbitrap HRMS analysis of serum samples to explore new methods for early diagnosis of different types of OLP and HC..A total of 19 differential metabolites were identified between EOLP and HC,and 25 differential metabolites were identified between ROLP and HC.14 differential metabolites were identified between EOLP and ROLP.Metabolites include Amino acids:Glutamine,Glutamic acid,Tryptophan,etc.;Carnitines:Acetyl-L-carnitine,Decanoylcarnitine,etc.;Cholines:LysoPC(22:5),LysoPC(16:1),etc.;Sphingolipids:Sphingosine,etc.;Fatty acids:Eicosapentaenoic acid,Arachidonic acid,Eicosahexaenoic acid,etc.;Organic acids:Citric acid,Lactic acid,etc.;and Benzamide,Uric acid,Indole acrylic acid and other compounds.3.According to the area under the curve,Eicosapentaenoic acid,Citric acid,Serine as diagnostic models between EOLP and HC;Eicosapentaenoic acid,Taurine and Serine as diagnostic models between ROLP and HC;Sphingosine,Deoxycholic acid,LysoPE(18:2)as diagnostic model between EOLP and ROLP.Binary logistic regression was used to evaluate the diagnostic effect of the model in the validation group.According to the sensitivity and specificity of model prediction,the test variable value with the highest Youden index(EOLP/HC=0.500,ROLP/HC=0.500,EOLP/ROLP=0.512)was selected as the best diagnostic critical value between groups.The diagnostic accuracy between the EOLP,ROLP and HC groups all reached 100%,and the diagnostic accuracy between EOLP and ROLP groups reached 92.68%.4.A total of 14 shared differential metabolites were found in the EOLP group and the ROLP group,including Glutamic acid,Tryptophan,Indole Acrylic acid,Benzamide and other compounds.Among them,Eicosapentaenoic acid,Citric acid and Benzamide showed a significant positive correlation in EOLP and ROLP.Spearman’s correlation coefficient was performed on these 14 metabolites using,and it was found that each metabolite had a strong correlation.5.Metabolic pathway analysis was carried out on the shared differential metabolites of EOLP and ROLP screened in the previous stage,and 6 important metabolic pathways were found,including:D-glutamine and D-glutamic acid metabolism;Alanine,Aspartic acid and Glutamic acid metabolism;Sphingolipid metabolism;Arginine and Proline metabolism;Glyoxylic acid and Dicarboxylic acid metabolism;Nitrogen metabolism.Therefore,metabolic changes in OLP patients are closely related to the abnormal amino acid metabolism.6.The GSE52130 dataset was found by the Gene Expression Omnibus,and 2013 significant differential genes(adjusted P value<0.05)between the OLP and HC groups were screened,of which 782 were up-regulated and 1231 were down-regulated.According to the condition of |LogFC|>0.6,1227 differential expression genes were screened out,and a protein-protein interaction network was constructed using the STRING database.7.Using HMDB and KEGG database,223 genes closely related to 14 shared differential metabolites were found,and no related gene information was retrieved for Benzamide and Indole acrylic acid.A "metabolite-gene" network was constructed using Cytoscape software.19 genes including IDO1,OPLAH,CD1B were highlighted by crossing the GSE52130 dataset and metabolite-regulated genes.Using KEGG,it was found that 19 shared genes related to OLP were mainly involved in 3 pathways,and they were involved in Alanine,Aspartate and Glutamate metabolism and Sphingolipid metabolism together with metabolic pathways.ConclusionsIn this research,UHPLC-Q-Orbitrap HRMS technology was used to deeply elucidate the metabolic changes of erosive and reticulated OLP,and three high-precision diagnostic models of EOLP and ROLP were successfully constructed with good diagnostic efficiency.The association analysis of metabolomic and transcriptomic data provided preliminary information on the potential molecular perturbations of OLP.The disordered gene-metabolic pathways may be related to the occurrence and development of OLP lesions.This research revealed the potential targets of OLP from the level of genes and metabolism for the first time,and provided a strong support for further exploration of the pathogenic mechanism of OLP in the future.