Apoptosis Induction by Iron Radiation via Inhibition of Autophagy in Trp53+/- Mouse Testes: Is Chronic Restraint-Induced Stress a Modifying Factor?

Abstract

We used chronic restraint-induced stress (CRIS) and iron ionizing radiation (IR) to mimic human exposure to psychological stress (PS) and IR in a mouse model, and to investigate the relationship among endoplasmic reticulum stress (ERS), apoptosis and autophagy in testicular toxicity. Male Trp53+/- C57BL/6N mice were restrained for 6 h/day for 28 consecutive days, and total body irradiation with 0.1 or 2 Gy iron ion beam was performed on the day 8. Histopathological observation showed severely damaged spermatogenic cells, increased apoptotic cells, caspase-3 activation and cytochrome c release, indicating that IR and CRIS+IR induced testicular cell apoptosis. Upregulation of GRP78 (78-kDa glucose-regulated protein) suggested that IR and CRIS+IR induced ERS in the testes, and further analysis showed that apoptosis was enhanced by ERS through activation of the PERK/eIF2α/ATF4/CHOP pathway. Decreased expression of LC3II, Atg5 (autophagy related 5) and Beclin 1, and increased expression of p62, combined with ultrastructural changes seen under transmission electron microscopy, suggest that IR and CRIS+IR inhibit autophagosome formation. This process was related to inhibition of autophagy via activation of the PI3K/AKT/mTOR pathway under ERS. We showed that apoptosis was strengthened and autophagy was inhibited by ERS in mouse testes induced by IR and CRIPS+IR. These results showed that CRIS+IR had no difference in apoptosis induction and autophagy inhibition compared with IR alone. CIRS alone could induce apoptosis only in Leydig cells and its induction of pathological and molecular changes in testicular tissues was only a small extent as compared to those induced by IR. Of note, we showed that 28 consecutive days of CRIS did not exacerbate IR effects (no additive effect with IR). These findings also suggest that studies on the concurrent exposure to PS and IR should be done using different endpoints in both short and long-term CRIS models.