Pyridoxine treatment alters embryonic motility in chicks: Implications for the role of proprioception.

The requirement for sensory feedback to coordinate motor activity in adults is unequivocal. However, the role that somatosensory information plays during early spontaneous embryonic motility is not clear. In order for somatosensation to play a role in motility, it stands to reason that functional sensory circuitry must first be established. In chicks, spontaneous movement begins 4 days after an egg is laid (embryonic day 4 - E4). It take roughly one day for a fertilized egg to be laid by the hen and roughly 21 more days for the embryo to hatch. Monosynaptic connections between proprioceptive neurons and motor neurons are first formed at the end of E7 (Davis, Frank, Johnson, & Scott, 1989; Lee, Koebbe, & O’Donovan, 1988). It is possible that proprioceptive information could be incorporated into motor patterns at this time. In fact, Oppenheim (1972) reported that reflex responses to flipping the limb are elicited as early as E7.5 and that responses to limb stroking appear by E8.5. However, the ability to evoke a reflexive behavior does not mean that sensory information is able to modulate spontaneous/intentional motility. In the chick, observations of changes in electromyographic (Bekoff, 1976) and kinematic (Bradley, 1999; Sharp, Ma, & Bekoff, 1999) recordings between E9 and E13 have been suggested to result from sensory modulation of spinal/motor circuitry during this time. Likewise, results from experiments which manipulate physical constraints on the embryo are most consistent with the idea that sensory feedback may alter motility at this time. Bradley (1997) showed that acute reduction of amniotic fluid altered motility in E9 chicks. However, it is not clear from this study whether alterations were responsive, due to changes in sensory feedback, or passive, due to movement restriction. On the other hand, Bradley and Sibelski (2000) showed that although both intra- and inter-limb joint coordination could be altered by near immobilization of the ankle on E12, this manipulation did not cause changes on E9. Likewise, it has been shown in rat fetuses that limb yoking on E19–21 alters inter-limb coordination (Robinson, 2005; Robinson, Kleven, & Brumley, 2008). This time frame in rats (Kucera, Walro, & Reichler, 1988; 1989; Milburn, 1973) is similar to the E9-E13 period in chick with respect to sensorimotor development. Taken as a whole, these data strongly indicate that embryonic motility can be modulated by somatosensation shortly after synaptic incorporation, but these experiments do not explicitly determine what sensory modalities are responsible for these changes. To more directly determine the role of somatosensation in modulating embryonic motility, it would be desirable to selectively remove a specific sensory modality and to then assay for changes in behavior. This is not an easy task, but pyridoxine toxicity is a reasonable approach. High doses of pyridoxine (vitamin B6) have been shown to mostly kill large proprioceptive neurons with somata in dorsal root ganglia (DRGs) and to cause severely ataxic locomotion in many adult mammals (human - Schaumburg, Kaplan, Windebank, Vick, Rasmus, Pleasure, & Brown, 1983; rat and dog - Antropol and Tarlov, 1942; rat and guinea pigs - Xu, Sladky, & Brown, 1989; rat - Helgren, Cliffer, Torrento, Cavnor, Curtis, DiStefano, Wiegand, & Lindsay, 1997; Krinke, Naylor, & Skorpil, 1985; cat - Pearson, Misiaszek, & Hulliger, 2003). Unfortunately, the mechanism of pyridoxine toxicity is unknown and damage to large diameter cutaneous neurons has been reported (e.g. Pearson et al., 2003). None the less, pyridoxine toxicity in cats has been used to gain an understanding of how proprioception is involved in balance (Stapley, Ting, Hulliger, & Macpherson, 2002) and recovery from peripheral nerve injury (Pearson et al., 2003). It is now known that pyridoxine administration in the chick embryo on E7 and E8 results in the loss of most large diameter, TrkC positive, sensory neurons when assayed on E13 (Sharp and Fedorovich, submitted). Additionally, this treatment spares motor neurons and small diameter sensory neurons expressing substance P and CGRP. As greater than 90% of TrkC positive neurons are proprioceptive in chick embryos (Oakley, Garner, Large, & Frank, 1995; Oakley, Lefcort, Plouffe, Ritter, & Frank, 2000), this suggests that pyridoxine is more selective in chicks than in mammals or that it is more selective during embryogenesis. Therefore, we used injections of pyridoxine to explore the ability of one somatosensory modality, proprioception, to modulate spontaneous motility of E9 chick embryos