Aging effects on the statistical and structural properties of the fornix of the brain

The brain consists of a complex network of axons, transmitting electrical impulses between interconnected neurons across distances that range from fractions of millimeters to meters. Myelinated axons, or nerve fibers, are axons that are wrapped by a myelin sheath, serving as an electrical insulation that increases the propagation speed of the signal along the nerve fiber while conserving the energy consumed and the space needed to maintain such propagation speed without myelin. Changes in the axon and surrounding myelin sheath during development and aging, or as a consequence of pathology, affect conduction and the proper functioning of the axon bundles. It is therefore important to be able to quantify the properties of these axons and their bundles and to discern which features best characterize the observed differences. We study the effects of aging on the myelinated axons in the fornix of the brain. The fornix is the principal subcortical output tract of the hippocampal formation, which plays a central role in memory. We obtain a collection of 328 high-quality electron micrographs from the fornix of 25 different rhesus monkey brains, ranging from young adults to the elderly, with both males and females. In this work, we develop a novel advanced recognition algorithm for automatically identifying myelinated axons and their surrounding myelin sheath. We extract multiple features of the nerve fibers and fully characterize their spatial structure. Using a feature selection algorithm, we discriminate between young and aged rhesus monvi keys with a high level of accuracy and pinpoint the differences in the aging process at the ultrastructural level across the life span. We observe a decline in the density of myelinated axons as well as in the fraction of occupied axon area with age, while the average axon area shows no dependence on the age of the subjects. We show an increase in the myelin thickness of axons for the female subjects, while no dependence is observed for the male subjects. This sex dichotomy is also present in the g-ratio of the myelinated axons, i.e., the ratio of the axon diameter to the fiber diameter. The method detailed here could be adapted to enable recognition in other areas as well as for changes caused by brain pathologies or by developmental disorders. Furthermore, the data collected will ultimately be useable in better modeling conduction properties in myelinated axons and better understanding how the aging process affects them.