Search

Most diseases are defined by a symptom, not a mechanism. Consequently, therapies remain symptomatic. In reverse, many potential disease mechanisms remain in arbitrary search for clinical relevance. Reactive oxygen species (ROS) are such an example. It is an attractive hypothesis that dysregulation of ROS can become a disease trigger. Indeed, elevated ROS levels of various biomarkers have been correlated with almost every disease, yet after decades of research without any therapeutic application. We here present a first systematic, non-hypothesis-based approach to transform this field as a proof of concept for biomedical research in general. We selected as seed proteins 9 families with 42 members of clinically researched ROS-generating enzymes, ROS-metabolizing enzymes or ROS targets. Applying an unbiased network medicine approach, their first neighbours were connected, and, based on a stringent subnet participation degree (SPD) of 0.4, hub nodes excluded. This resulted in 12 distinct human interactome-based ROS signalling modules, while 8 proteins remaining unconnected. This ROSome is in sharp contrast to commonly used highly curated and integrated KEGG, HMDB or WikiPathways. These latter serve more as mind maps of possible ROS signalling events but may lack important interactions and often do not take different cellular and subcellular localization into account. Moreover, novel non-ROS-related proteins were part of these forming functional hybrids, such as the NOX5/sGC, NOX1,2/NOS2, NRF2/ENC-1 and MPO/SP-A modules. Thus, ROS sources are not interchangeable but associated with distinct disease processes or not at all. Module members represent leads for precision diagnostics to stratify patients with specific ROSopathies for precision intervention. The upper panel shows the classical approach to generate hypotheses for a role of ROS in a given disease by focusing on ROS levels and to some degree the ROS type or metabolite. Low levels are considered physiological; higher amounts are thought to cause a redox imbalance, oxidative stress and eventually disease. The source of ROS is less relevant; there is also ROS-induced ROS formation, i.e. by secondary sources (see upwards arrow). The non-hypothesis-based network medicine approach uses genetically or otherwise validated risk genes to construct disease-relevant signalling modules, which will contain also ROS targets. Not all ROS sources will be relevant for a given disease; some may not be disease relevant at all. The three examples show (from left to right) the disease-relevant appearance of an unphysiological ROS modifier/toxifier protein, ROS target or ROS source.

Additional file 1: Additional figures and tables.

Additional file 1: Fig. S1. Datasets and experimental design. Fig. S2. Detailed comparison of the WID general clock and other epigenetic clocks, and assessments of cell composition in response to menopause, hormone replacement and cancer. Fig. S3. Detailed epidemiological assessment of the WID general clock age acceleration. Fig. S4. Association of the WID general, epithelial and immune clocks with menopause, replicative age, and solo-WCGW methylation. Fig. S5. Association of WID-relative-epithelial- and -immune-age with chronological age and immune cell proportion and breast cancer subtype. Fig. S6. Epidemiological associations of the WID-relative-epithelial-age. Fig. S7. Epidemiological associations of the WID-relative-immune-age. Table S1. Epidemiological and clinical characteristics of the training set. Table S2. Epidemiological and clinical characteristics of the validation set 1. Table S3. WID-REA quantiles and pre-menopausal breast cancer risk. Table S4. Ages of individuals in the breast tissue sets. Table S5. Epidemiological and clinical characteristics of Validation set 2. Table S6. List of samples used to assess the performance of age prediction.

Additional file 2. Is the protocol for Clinical Trial 1, ‘Mifepristone treatment prior to insertion of a levonorgestrel releasing intrauterine system for improved bleeding control – a randomized controlled trial’ (EudraCT registration number 2009-009014-40).

Additional file 1: Supplementary Methods. Fig. S1. Elevated levels of estradiol and progesterone in BRCA1 and BRCA2 mutation carriers compared to controls. Fig. S2. Epithelial cell composition in normal tissues adjacent to TNBC compared to cancer-free breasts. Fig. S3. Enrolment and randomisation in Clinical Trial 1. Fig. S4. Enrolment and randomisation in Clinical Trial 2. Fig. S5. Enrolment in Clinical Trial 3. Fig. S6. Effect of mifepristone treatment on breast epithelium cell subtype composition. Fig. S7. Effect of mifepristone treatment on breast epithelium cell subtype composition – RNA-seq data. Fig. S8. The WID-Breast5 index. Fig. S9. The Ki67 index. Fig. S10. Comparing the Ki67 index with the WID-Breast29 index. Fig. S11. Comparing the change (before and after ulipristal acetate) in percentage of immunohistochemistry Ki67-positive cell with change in Ki67-index in DNAme Dataset 4 (Clinical trial 3, ‘A pilot prevention study of the effects of the anti- progestin Ulipristal Acetate (UA) on surrogate markers of breast cancer risk’). Table S1a. Summary statistics about the 20 healthy women (12 BRCA1 mutation carriers and 8 BRCA1 wild type) enrolled in the trial to measure estradiol and progesterone levels throughout the menstrual cycle (see also Figs. 1 and 2). Table S1b. Summary statistics about the 21 healthy BRCA mutation carriers enrolled in Clincal Trial 2 who provided samples for DNAme analysis as part of DNAme Set 3. Table S2. The 37 WID-Breast29 CpGs. Table S3. Summary statistics about the donors of the breast cancer biopsies and matched surrounding normal tissue samples. Table S4. The reference CpGs identified for luminal progenitor, mature luminal, and basal breast epithelial cells. Table S5. Summary statistics about the 15 healthy women who provided samples for RT-PCR and RNA-seq before and after mifepristone treatment.

Additional file 4. Is the protocol for Clinical Trial 3, ‘A pilot prevention study of the effects of the anti- progestin Ulipristal Acetate (UA) on surrogate markers of breast cancer risk’ (EudraCT registration number: 2015-001587-19).

Additional file 3. Is the protocol for Clinical Trial 2, ‘The effect of a progesterone receptor modulator on breast tissue in women with BRCA1 and 2 mutations’ (EudraCT registration number: 2012-003703-35).

Additional file 5. Review history.

Additional file 1. Supplementary figures and tables: Figure S1. Cell-type composition in the combined Discovery Set and Diagnostic Validation Set as determined by the EpiDISH algorithm (A). Distribution of p-values after comparing hrHPV-positive controls to CIN3+ cases in the Discovery Set (based on a linear regression model with adjustment for age and immune cell proportion) (B). Distribution of the estimated epithelial and immune delta-betas (C). Performance of ridge and lasso classifiers based on out-of-bag estimates from 10-fold cross validation on the Discovery Set (D). Odds ratios when comparing the genomic annotation of the 5,000 CpGs comprising the WID-CIN-index to the 777,005 CpGs that were used in the analysis (E). Figure S2. Dependence of the WID-CIN-index on immune cell proportion in the hrHPV-positive controls and CIN3+ cases from the Diagnostic Validation Set (A). Dependence of the WID-CIN-index on age in the hrHPV-positive controls and CIN3+ cases from the Diagnostic Validation Set (B). The WID-CIN-index in different subgroups and HPV genotypes from the Diagnostic Validation Set (C). The other high risk genotype category consists of genotypes 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66 and 68. Only samples that tested positive for one of the three genotype categories were included in the plot. Figure S3. The cell-type composition of samples from the predictive validation set based on the EpiDISH algorithm (A). Dependence of the WID-CIN-index on age in the hrHPV-positive controls and CIN3+ cases from the Predictive Validation Set (B). Kaplan-Meier curves from the Predictive Validation Set in which samples have been split into those below and above the 75% specificity cutoff (C). Table S1. Summary of hrHPV genotypes in the Discovery Set (A), Diagnostic Validation Set (B) and Predictive Validation Set (C). The other high risk genotype category consists of genotypes 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66 and 68. Note that rows may not sum to the total sample number as some samples tested postive for multipe genotypes.