Metilacijski označevalci za karakterizacijo in določanje izvora adenokarcinomov jeter

Govejšek, Tina (author), Hauptman, Nina (mentor), Zidar, Nina (comentor)

Metilacijski označevalci veljajo za eno izmed obetavnih epigenetskih značilnosti raka za diagnosticiranje in razlikovanje adenokarcinomov jeter, ki sodijo med najpogostejša maligna obolenja na svetu. Namen doktorske naloge je bil prepoznati in eksperimentalno potrditi nove diagnostične metilacijske označevalce, ki razlikujejo med dvema najpogostejšima primarnima karcinomoma jeter, hepatocelularnim karcinomom (HCC) in holangiokarcinom (CCA) ter adenokarcinomoma, ki pogosto zasevata v jetra, adenokarcinomom debelega črevesa in danke (CRC) in duktalnim adenokarcinomom trebušne slinavke (PDAC). Ker je metilacija DNA eden ključnih mehanizmov uravnavanja izražanja genov, smo želeli preučiti vpliv spremenjene metilacije DNA izbranih metilacijskih mest na izražanje pripadajočih genov. Prvi del naloge je zajemal obsežno dvostopenjsko bioinformacijsko analizo, ki je vključevala učno množico podatkov, pridobljenih iz javno dostopne baze TCGA z 2853 vzorci ter neodvisno testno množico podatkov, pridobljenih iz javno dostopne baze GEO s 782 vzorci. Z njo smo uspešno prepoznali nove diagnostične označevalce metilacije DNA, ki so hipermetilirani pri raku ter oblikovali diagnostične panele, ki razlikujejo med izbranimi adenokarcinomi. Paneli so dosegli občutljivost 77,8-95,9 %, specifičnost 91,5-97,7 % ter diagnostično natančnost 85,3-96,4 % za razlikovanje med izbranimi adenokacinomi in pripadajočimi zdravimi tkivi. Metilacijo DNA izbranih metilacijskih mest smo eksperimentalno potrdili z metodama metilacijsko občutljive talilne krivulje visoke ločljivosti (MS-HRM) in digitalnim PCR (dPCR) na 149 vzorcih tkiv, fiksiranih v formalinu in vklopljenih v parafin. Vključili smo vzorce HCC, CCA, CRC, PDAC, vzorce pripadajočih zdravih tkiv, zasevke CRC v jetrih in zasevke PDAC v jetrih. Paneli metilacijskih označevalcev, ki smo jih eksperimentalno potrdili, se nahajajo v promotorjih naslednjih genov in dolgih nekodirajočih RNA (lncRNA): RNF135 (angl. ring finger protein 135), EFNB2 (angl. ephrin-B2), RNF125 (angl. ring finger protein 125), ABLIM1 (angl. actin-binding LIM protein 1), OSMR (angl. oncostatin M receptor), LIFR-AS1 (angl. leukemia inhibitory factor receptor antisense RNA 1) in PROX1-AS1 (angl. prospero homeobox 1 antisense RNA 1). Paneli so razlikovali med HCC, CCA, zasevki CRC v jetrih, zasevki PDAC v jetrih in zdravim tkivom jeter z občutljivostjo 66,7-100 %, s specifičnostjo 94,9-100 % in z diagnostično natančnostjo 93-100 %. Rezultati dPCR so bili dodatno potrjeni z bioinformacijsko analizo z uporabo neodvisnega nabora podatkov 487 vzorcev iz podatkovnih zbirk TCGA in GEO. Rezultati MS-HRM, dPCR in bioinformacijske analize so pokazali visoko skladnost. Prav tako smo dokazali, da se hipermetilacija DNA preiskovanih promotorjev ohranja od primarnih CRC in PDAC do njihovih zasevkov v jetrih. Za izbrane gene smo izvedli metodo kvantitativne verižne reakcije s polimerazo v realnem času (qPCR) in analizo izražanja genov ter izračunali korelacijo med metilacijo DNA in izražanjem genov. Rezultati so pokazali, da lahko spremenjena metilacija preiskovanih promotorjev vpliva na izražanje genov pri preučevanih adenokarcinomih.

• Language: Slovenian
• Keywords: Adenokacinomi jeter, hepatocelularni karcinom, holangiokarcinom, adenokarcinom debelega črevesa in danke, duktalni adenokarcinom trebušne slinavke, zasevki v jetrih, bioinformacijska analiza, MS-HRM, dPCR, metilacija DNA, izražanje genov.
• Work type: Doctoral dissertation
• Organization: MF - Faculty of Medicine
• Year: 2025
• PID: 20.500.12556/RUL-167566
• Publication date in RUL: 28.02.2025

Secondary language

• Language: English
• Title: Methylation Markers for Characterization and Determination of the Cancer Origin in Liver Adenocarcinomas

Abstract

DNA methylation biomarkers are one of the most promising epigenetic features for the diagnosis and differentiation of liver adenocarcinomas, which are among the most common malignancies worldwide. The aim of this doctoral thesis was to identify and experimentally validate novel diagnostic DNA methylation markers that differentiate between the two most common primary liver cancers, hepatocellular carcinoma (HCC) and cholangiocarcinoma (CCA), and adenocarcinomas that frequently metastasize to the liver, i.e., colorectal adenocarcinoma (CRC) and pancreatic ductal adenocarcinoma (PDAC). Since DNA methylation plays a key role in the regulation of gene expression, we aimed to investigate how altered DNA methylation at selected sites affects the expression of the corresponding genes. The first part of the study consisted of a comprehensive two-stage bioinformatics analysis comprising a TCGA identification dataset with 2853 samples and an independent GEO verification dataset with 782 samples. The analysis successfully identified novel diagnostic DNA methylation markers that are hypermethylated in cancer. Diagnostic panels that differentiate between selected adenocarcinomas were created. The panels differentiate between selected adenocarcinomas and healthy tissues with a sensitivity of 77.8-95.9%, a specificity of 91.5-97.7% and a diagnostic accuracy of 85.3-96.4%. To substantiate our bioinformatic findings, DNA methylation of selected methylation sites was experimentally validated by methylation-sensitive high-resolution melting (MS-HRM) and digital PCR (dPCR) on 149 formalin-fixed, paraffin-embedded tissue samples from HCC, CCA, CRC, PDAC, paired normal tissues, CRC liver metastases and PDAC liver metastases. The panels included the following genes and long non-coding RNAs (lncRNA): ring finger protein 135 (RNF135), ephrin-B2 (EFNB2), ring finger protein 125 (RNF125), homeobox-C 4 (HOXC4), actin-binding LIM protein 1 (ABLIM1), oncostatin M receptor (OSMR), lncRNA leukemia inhibitory factor receptor antisense RNA 1 (LIFR-AS1) and lncRNA prospero homeobox 1 antisense RNA 1 (PROX1-AS1). The panels differentiated HCC, CCA, CRC liver metastases, PDAC liver metastases and healthy liver tissue with a sensitivity of 66.7-100%, a specificity of 94.9-100% and a diagnostic accuracy of 93-100% in the dPCR experiment. The dPCR results were further validated by bioinformatic analysis using an independent dataset of 487 samples from the TCGA and GEO databases. The results of MS-HRM, dPCR and bioinformatic analysis showed a high concordance. Furthermore, we were able to show that DNA hypermethylation of the investigated promoter regions is maintained from the primary CRC and PDAC to their liver metastases. For selected genes and lncRNAs, we performed quantitative real-time polymerase chain reaction (qPCR) and gene expression analyses and calculated the correlation between DNA methylation and gene expression. The results showed that altered methylation can influence gene expression in certain groups of the investigated adenocarcinomas.

• Keywords: Liver adenocarcinomas, hepatocellular carcinoma, cholangiocarcinoma, colorectal adenocarcinoma, pancreatic ductal adenocarcinoma, liver metastases, bioinformatics analysis, MS-HRM, dPCR, DNA methylation, gene expression.