Experiments by nature: lessons on type 1 diabetes.

The etiology and pathogenesis of type 1 diabetes (T1D) - one of the most frequent chronic, life-debilitating diseases in humans - have long fascinated endocrinologists, pathologists and biologists alike. Currently conventional wisdom portrays T1D as a chronic T cell-mediated autoimmune disease that leads to the specific destruction of pancreatic insulin-producing β cells. The process of β cell destruction is accompanied (or preceded) by the production of autoantibodies (autoAb) to β cell antigens (i.e. insulin, GAD65, IA-2 and ZnT8). These autoAb have proved to be instrumental in identifying subjects at risk of developing the disease prior to overt hyperglycemia, and they help to distinguish T1D from T2D patients (who have no autoAb), but are not deemed to be pathogenic. This review will examine to which extent this well-established disease-dogmas are sustained by experiments by nature, which should not suffer from the common biases and errors of experiments by humans.