In mammals, center nerve axons are often wrapped by compact membranous sheaths called myelin (Nave & Werner, 2014). This electrical insulator-like structure enables rapid saltatory propagation of nerve impulses along a myelinated axon, by restricting action potentials (APs) across the axonal membrane only around the unmyelinated intervals (known as Ranvier’s nodes) (Nave and Werner, 2014, Tasaki, 1939). Therefore, myelin function is essential for the nervous systems to ensure a fast signal transmission over a long axonal path length (Monje, 2018). In our auditory system, for instance, axons are often extensively myelinated to conduct APs rapidly and reliably, providing sound sensation with extraordinary temporal resolution. However, it remains elusive how AP conduction velocity is coordinated among axons of varying path lengths to ensure a synchronized signal processing (Ford et al., 2015, Stange-Marten et al., 2017).

Superior olivary complex (SOC) of the brainstem is dedicated to the sound source localization by detecting interaural time and intensity differences (Grothe et al., 2010, Yu and Goodrich, 2014). Anatomically, the SOC is composed of medial nucleus of the trapezoid body (MNTB), medial superior olives (MSO), lateral superior olives (LSO), and multiple peri-olivary nuclei including lateral nucleus of the trapezoid body, ventral nucleus of the trapezoid body, superior paraolivary, dorsolateral periolivary region, dorsomedial periolivary region, anterolateral periolivary region, and mediodorsal periolivary region (Grothe and Park, 2000, Thompson and Schofield, 2000). The MNTB receives excitatory inputs exclusively from the cochlear nucleus (CN) on the contralateral side by giant nerve terminals (known as the calyx of Held, CH) between globular bushy cells (GBCs) in the CN and postsynaptic MNTB somata (Yu & Goodrich, 2014). In turn, the glycinergic MNTB neurons project to the ipsilateral MSO, where excitatory inputs from both contra- and ipsilateral CN spherical bushy cells (SBCs) are converged, allowing interaural time differences (ITD) determination (Grothe et al., 2010, Yu and Goodrich, 2014). In the LSO, interaural level differences (ILDs) are detected by comparing the ipsilateral CN inputs with the MNTB-relayed signals from the contralateral CN (Grothe et al., 2010, Kandler, 2004, Sanes and Rubel, 1988, Yu and Goodrich, 2014).

In gerbils, axons of GBC and SBC from the same CN project to contralateral MNTB and MSO, respectively. As the GBC axons have a greater inner diameter and internodal length compared with the SBC axons (Ford et al., 2015), faster AP conduction on the GBC axons was expected. In fact, the delay of the CN(GBC)-MNTB-MSO pathway can be even smaller than that of the CN(SBC)-MSO pathway, in spite of longer axonal pathlength and bisynaptic transmission, resulting in that the MNTB-derived inhibitory input driven by contralateral GBC can even precede the excitatory input driven by contralateral SBC from the same CN at MSO (Brand et al., 2002, Ford et al., 2015, Roberts et al., 2013). Similarly, MSO receives SBC inputs from CNs of both sides and the long-traveling SBC axons from the contralateral CN appear greater in internodal length and inner diameter than their ipsilateral counterparts (Seidl & Rubel, 2016). These observations argue for precise coordination in AP propagation velocities by fine-tuned axonal morphology and myelin profile throughout the circuit for biaural sound source localization.

Moreover, the tonotopic arrangement of axon projections is well characterized in auditory nuclei, including MNTB (Kandler et al., 2009). GBC axons that encode high-frequency sounds terminate preferentially in the medial MNTB region, whereas those carrying low-frequency information travel a longer distance and end more laterally (Kandler et al., 2009). One recent study in gerbils found that GBC axons responding to low-frequency sounds have shorter myelinated internodes and larger axonal diameters than those to high-frequency sounds, resulting in faster propagation of APs along low-frequency axons (Ford et al., 2015). This suggests the presence of an intra-nucleus structural tuning of the GBC axons to modulate AP conduction velocity (Ford et al., 2015, Stange-Marten et al., 2017). However, the same morphological heterogeneity in GBC axons was missing in the CBA mice (Stange-Marten et al., 2017), probably owing to the species- or mouse strain-specific difference in the temporal resolution of binaural hearing (Fischl et al., 2016, Grothe and Pecka, 2014, Stange-Marten et al., 2017).

The CBA mouse can preserve normal hearing function into adulthood without notable hearing loss till 18 months of age (Li and Borg, 1991, Li and Hultcrantz, 1994, Spongr et al., 1997). The C57BL/6 mouse (another widely used mouse strain), by contrast, (Park et al., 2010, Sha et al., 2008), is known for the early-onset loss of high-frequency hearing (Li and Borg, 1991, Li and Hultcrantz, 1994, Park et al., 2010, Spongr et al., 1997), giving the opportunity to study age-related changes in GBC axons. By means of serial block-face scanning electron microscopy, we performed comprehensive structural mapping of MNTB tissues from the C57BL/6 mice at different ages, and assessed the differences in GBC axon morphology across juvenile, adult, and old mice. Our quantifications revealed distinct fiber ultrastructure and myelin profile of GBC axons targeting high and low-frequency ranges of the MNTB, and their degenerative trajectories during aging.