摘要

The CYP17A1 gene encodes 17OHase and 17,20-lyase activities. We have previously reported an E305G mutation in CYP17A1 that was found by expression studies to cause only isolated 17,20-lyase deficiency. Further clinical investigations and urinary steroid examination have revealed that this mutation has in fact an extensive effect on 17OHase activity. The aim of the present study was to analyze the metabolic patterns of our patients using a novel application of a mathematical approach. Implementing an unbiased systems biology metabolomics approach, together with methods adapted from gene expression data analysis, we investigated the steroid metabolomic network relevant for 17,20-lyase deficiency of patients with the CYP17A1 gene mutation E305G. Six homozygote patients with a CYP17A1 gene E305G mutation and eight unaffected controls. We used gas chromatography-mass spectrometry (GC-MS) to quantitatively assess 31 urinary steroid metabolites. We applied statistical methods adopted from transcriptomics to identify differential metabolites and differential metabolite ratios, and a heatmap to visualize the results. We assigned Threshold Number of Misclassifications (TNoM) scores to metabolite ratios. Implementation of the unbiased approach revealed alteration in the metabolic balance of unsuspected enzyme systems in our patients. Of the 166 metabolite ratios that reached the discriminatory TNoM score of 0 or 1, 138 discriminated between the groups on the basis of CYP17A1 activity, three indicated inhibition of 3 beta HSD2 activity, ten indicated inhibition of 3 beta HSD2/21OHase/11OHase activity, and four enhanced 11 beta HSD2 activity. The unbiased all-inclusive metabolomic approach powerfully discriminated between the patients and the control groups on the basis of CYP17A1 activity. Other adrenal and extra-adrenal enzymes were affected by this mutation. We present an informatics tool that suggests disease-specific profiles and opens interesting research avenues.

  • 出版日期2010-9