In the cleidomastoid muscle in E18–P0 animals, each labeled terminal axonal branch covered, on average, 14.2% (±11.4%, n = 151) of the total AChR area per contacted junction. This small percentage of occupation probably overestimates the actual area of synaptic contact, because it includes nonsynaptic connector

branches (see electron microscopy section below). Even so, of the 151 junctions studied, only one was innervated by an axon that overlapped with more than 50% of the junctional area (Figure 1J). The typically small contact area of single axonal input to neuromuscular junctions suggests that each developing neuromuscular junction may be shared by many different axons. Indeed, when we looked at neonatal neuromuscular junctions in a transgenic

fluorescent protein-expressing mouse line that labels all motor axons (“YFP-16”; Feng et al., 2000), we saw that the cumulative synaptic PLX3397 drive to each neonatal neuromuscular junction was much greater than that shown by single axon labeling (compare Figures 1A–1D with 1K). With all axons labeled, each perinatal junction was nearly fully occupied (92.4% ± 5.0%, n = 33, of the receptor area covered; Figure 1K). The synaptic vesicle marker synaptophysin was also present throughout each junction (Figure 1K), arguing that the majority of these contacts are synaptic. However, the small size of perinatal neuromuscular junctions Resminostat compounded by the tight see more fasciculation of the incoming axons and their small caliber made it impossible to directly assess the number of converging axons at neonatal junctions by fluorescence microscopy given the limitations imposed by diffraction (see below). To learn when axonal arbors projected to the greatest number of muscle fibers, we also screened embryonic muscles from YFP-H and GFP-S mice for ones that contained a single fluorescent motor axon. Analysis of motor neuron

axon arbors from embryonic periods (E16–E18) showed that the size of motor units increased over prenatal life to reach a peak just before birth. We found that at E18 (1 day before birth), motor units are larger than the first day after birth. An example of this change is presented in Figure 2A, which shows a clavotrapezius motor unit at E18 whose arbor extends to 331/412 muscle fibers. This axon projects to 80.3% of the neuromuscular junctions, whereas the average axonal projection was 4.6% of the muscle fibers in P23 animals (Table 1). However, 3 days before birth (E16), motor unit sizes were, on average, ∼6-fold smaller than at E18 (n = 5; see Table 1). Figure 2B shows an axon reconstructed from an E16 cleidomastoid muscle in which the labeled axon innervates 52 of 161 (32.3%) of the total number of neuromuscular junction sites.