The trigeminal ganglion has three main peripheral axonal branches, the ophthalmic, maxillary, and mandibular, which innervate the

corresponding regions of the face. Sensory information is then conveyed from the ganglia to the brainstem nuclei via a centrally projecting axonal bundle. The neurons that innervate each of these regions in the face are spatially segregated into specific domains within the ganglia and exhibit distinct gene expression profiles, reflecting the division of these otherwise similar trigeminal neurons into distinct subtypes (Hodge et al., 2007). Some of these differentially expressed genes affect axonal pathfinding programs that allow the central projections of these neurons to innervate the brainstem (Hodge AZD2281 molecular weight et al., 2007). Studies on the mechanism of the acquisition of these distinct identities have focused on BMP4, a TGF-β family member expressed in the distal epithelium of the maxillary and ophthalmic regions in the face (Hodge et al., 2007). As axons grow into these regions, they encounter BMP4, which results in a retrograde signal that leads to nuclear accumulation

of the phosphorylated and transcriptionally active forms of the SMAD1, 5, and 8 transcription factors (Nohe et al., 2004). Additionally, Tbx3, a predicted SMAD1 target ( Chen et al., 2008), is also selectively induced in selleck screening library the ophthalmic- and maxillary-innervating neurons in a BMP4-dependent manner ( Hodge et al., 2007). This retrograde signaling contributes to the gene expression differences between the ophthalmic- and maxillary-innervating neurons and mandibular-innervating trigeminal neurons ( Hodge et al., 2007). However, the nature of the retrograde BMP4 signal, and whether other factors are also involved in patterning the trigeminal ganglia remain unknown. We first sought to recapitulate retrograde BMP4 signaling in vitro by culturing dissociated

E13.5 rat trigeminal ganglia neurons in microfluidic chambers (Taylor et al., 2005). In these devices, axons grow through a 450 μm microgroove barrier and appear in the axonal compartment Bay 11-7085 by 2 days in vitro (DIV). Because the axons are fluidically isolated from the cell bodies, this approach allows experimental treatments to be applied selectively to axons (Taylor et al., 2005; Figure S1A, available online). The majority of the neurons that are adjacent to the microgrooves send an axon to the axonal compartment, as detected by retrograde labeling of cell bodies by axonal application of CM-DiI (Figure S1A). Selective application of BMP4 to the axonal compartment resulted in an increase in nuclear pSMAD1/5/8 (Figures 1A, 1B, S1B, and S1C). pSMAD1/5/8 levels nearly doubled within 15 min, with further increases over 1–2 hr (Figure S1D). Total SMAD1/5/8 localization and levels were unaffected (Figures S1E and S1F), indicating that BMP4 increases the fraction of SMAD1/5/8 that is phosphorylated.