Prenatal exposure of diabetes and progestin-mediated autistic biomarker in peripheral blood mononuclear cells.

ABSTRACT

Despite the importance of early diagnosis and intervention, the diagnosis of autism spectrum disorders (ASDs) remains delayed as it is mostly based on clinical symptoms and abnormal behaviours appearing after 2 years of age. Identification of autistic markers remains a top priority in achieving an early and effective ASD diagnosis. We have previously reported that prenatal exposure of hormones or diabetes triggers epigenetic changes and oxidative stress, resulting in gene suppression with autism-like behaviours in offspring. Here, a potential biomarker for ASD diagnosis was established through gene analysis in peripheral blood mononuclear cells (PBMCs). The study from in vivo mouse showed that prenatal hormone exposure or maternal diabetes suppresses mRNA expression of estrogen-related receptor α (ERRα), superoxide dismutase 2 (SOD2), G protein-coupled estrogen receptor (GPER) and retinoic acid-related orphan receptor α (RORA) in the brain as well as oxidative stress and mitochondrial dysfunction, subsequently triggering autism-like behaviour in mouse offspring. Also, similar gene suppression was found in hematopoietic stem cells (HSCs) and PBMC, with inherited epigenetic changes being identified on the related promoters. The human case-control study found that mRNA levels of ERRα, SOD2, GPER and RORA were significantly reduced in PBMC from ASD subjects (n = 132) compared with typically developing (n = 135) group. The receiver operating characteristic curve showed a .869 ± .021 of area under the curve for ASD subjects with 95% confidence interval of .829-.909, together with 1.000 of sensitivity and .856 of specificity. In conclusion, the combined mRNA expression in PBMC based on prenatal factor exposure-mediated gene suppression could be a potential biomarker for ASD diagnosis.