Gene expression profiling in patients with polymyalgia rheumatica before and after symptom-abolishing glucocorticoid treatment

Comprehensive multi-cohort transcriptional meta-analysis of muscle diseases identifies a signature of disease severity

Muscle diseases share common pathological features suggesting common underlying mechanisms. We hypothesized there is a common set of genes dysregulated across muscle diseases compared to healthy muscle and that these genes correlate with severity of muscle disease. We performed meta-analysis of transcriptional profiles of muscle biopsies from human muscle diseases and healthy controls. Studies obtained from public microarray repositories fulfilling quality criteria were divided into six categories: (i) immobility, (ii) inflammatory myopathies, (iii) intensive care unit (ICU) acquired weakness (ICUAW), (iv) congenital muscle diseases, (v) chronic systemic diseases, (vi) motor neuron disease. Patient cohorts were separated in discovery and validation cohorts retaining roughly equal proportions of samples for the disease categories. To remove bias towards a specific muscle disease category we repeated the meta-analysis five times by removing data sets corresponding to one muscle disease class at a time in a "leave-one-disease-out" analysis. We used 636 muscle tissue samples from 30 independent cohorts to identify a 52 gene signature (36 up-regulated and 16 down-regulated genes). We validated the discriminatory power of this signature in 657 muscle biopsies from 12 additional patient cohorts encompassing five categories of muscle diseases with an area under the receiver operating characteristic curve of 0.91, 83% sensitivity, and 85.3% specificity. The expression score of the gene signature inversely correlated with quadriceps muscle mass (r = -0.50, p-value = 0.011) in ICUAW and shoulder abduction strength (r = -0.77, p-value = 0.014) in amyotrophic lateral sclerosis (ALS). The signature also positively correlated with histologic assessment of muscle atrophy in ALS (r = 0.88, p-value = 1.62 × 10^-3) and fibrosis in muscular dystrophy (Jonckheere trend test p-value = 4.45 × 10^-9). Our results identify a conserved transcriptional signature associated with clinical and histologic muscle disease severity. Several genes in this conserved signature have not been previously associated with muscle disease severity.

Methylomic and transcriptomic predictors of one-month exposure to cortisol in healthy individuals

Allostatic load (AL) refers to the cumulative "wear and tear" on an organism throughout its lifetime. One of the primary contributing factors to AL is prolonged exposure to stress or its primary catabolic agent cortisol. Chronic exposure to stress or cortisol is associated with numerous diseases, including cardiovascular disease, metabolic disorders, and psychiatric disorders. Therefore, a molecular marker capable of integrating a past history of cortisol exposure would be of great utility for assessing disease risk. To this end, we recruited 87 healthy males and females of European ancestry between 18 and 60 years old, extracted genomic DNA and RNA from leukocytes, and implemented a gene-centric DNA enrichment method coupled with bisulfite sequencing and RNA-Seq of total RNA for the determination of genome-wide methylation and gene transcription, respectively. Sequencing data were analyzed against awakening and bedtime cortisol data to identify differentially methylated regions (DMRs) and CpGs (DMCs) and differentially expressed genes (DEGs). Six candidate DMCs (p_unadjusted < 0.005) and nine DEGs (p_unadjusted < 0.0005) were used to construct a prediction model that could capture past 30+ days of both bedtime and awakening cortisol levels. Utilizing a cross-validation approach, we obtained a regression coefficient of R² = 0.308 for predicting continuous awakening cortisol and an area under the curve (AUC) = 0.753 for dichotomous (high vs. low tertile) awakening cortisol, and R² = 0.224 and AUC = 0.723 for continuous and dichotomous bedtime cortisol levels, respectively. To our knowledge, the current study represents the first attempt to identify genome-wide predictors of cortisol exposure that utilizes both methylation and transcription targets. The utility of our approach needs to be replicated in an independent cohort of samples for which similar cortisol metrics are available.