Combined proteomic and RNA microarray analysis after TRIzol extraction of eutopic endometrium for endometriosis biomarker discovery

Abstract

OBJECTIVE: Endometriosis (endo) is an important gynaecological disease defined by the presence of ectopic endometrium (EM) outside the uterus and associated with pelvic pain and infertility. In this study we evaluated for the first time the feasibility of EM TRIzol extraction to obtain both a protein and mRNA fraction of sufficient quality from the same EM sample in order to perform combined proteomic analysis and mRNA microarray.DESIGN: Proteomic analysis of TRIzol-extracted eutopic endometrium.MATERIALS AND METHODS: Endometrial tissues were obtained during the late luteal phase (day 23-26) from 16 patients including controls with a normal pelvis (n=5) and patients with endometriosis (n=11) including minimal-mild (n=5) and moderate-severe (n=6) disease according to ASRM classification. We used Surface Enhanced Laser Desorption/Ionisation Time-of-Flight Mass Spectrometry (SELDI-TOF-MS): all samples were measured in duplicates on CM10 and IMAC30 protein chips in low molecular weight (CHCA) and high molecular weight (SPA). All mass peaks were used as input for a Least Squares Support Vector Machine (LS-SVM). Next, Leave one out cross validation (LOO-CV) was used to estimate independent test set performance.RESULTS: Profiling of endometrial samples by using different ProteinChip surfaces showed differential expression of proteins and peptides in the range between 7 - 17 kDa depending on chip and matrix type. The LOO-CV performance of the LS-SVM models indicated an Area under the ROC curve (AUC) between 85%-100% (depending on chip and matrix type) for the diagnosis of endometriosis. These results will be discussed in relationship to the mRNA microarray results obtained on the same EM samples (Sherwin et al, 2008).CONCLUSIONS: Protein profiling using ProteinChip array after TRIzol preparation is feasible and reproducible and allows combined proteomic analysis and mRNA microarray on the same EM sample. OBJECTIVE: Endometriosis (endo) is an important gynaecological disease defined by the presence of ectopic endometrium (EM) outside the uterus and associated with pelvic pain and infertility. In this study we evaluated for the first time the feasibility of EM TRIzol extraction to obtain both a protein and mRNA fraction of sufficient quality from the same EM sample in order to perform combined proteomic analysis and mRNA microarray. DESIGN: Proteomic analysis of TRIzol-extracted eutopic endometrium. MATERIALS AND METHODS: Endometrial tissues were obtained during the late luteal phase (day 23-26) from 16 patients including controls with a normal pelvis (n=5) and patients with endometriosis (n=11) including minimal-mild (n=5) and moderate-severe (n=6) disease according to ASRM classification. We used Surface Enhanced Laser Desorption/Ionisation Time-of-Flight Mass Spectrometry (SELDI-TOF-MS): all samples were measured in duplicates on CM10 and IMAC30 protein chips in low molecular weight (CHCA) and high molecular weight (SPA). All mass peaks were used as input for a Least Squares Support Vector Machine (LS-SVM). Next, Leave one out cross validation (LOO-CV) was used to estimate independent test set performance. RESULTS: Profiling of endometrial samples by using different ProteinChip surfaces showed differential expression of proteins and peptides in the range between 7 - 17 kDa depending on chip and matrix type. The LOO-CV performance of the LS-SVM models indicated an Area under the ROC curve (AUC) between 85%-100% (depending on chip and matrix type) for the diagnosis of endometriosis. These results will be discussed in relationship to the mRNA microarray results obtained on the same EM samples (Sherwin et al, 2008). CONCLUSIONS: Protein profiling using ProteinChip array after TRIzol preparation is feasible and reproducible and allows combined proteomic analysis and mRNA microarray on the same EM sample.