Prior findings led us to hypothesize that West Nile virus (WNV) preferentially transports along motor axons instead of sensory axons. WNV is known to undergo axonal transport in cell culture and in infected hamsters to infect motor neurons in the spinal cord. To investigate this hypothesis, WNV was injected directly into the left sciatic nerve of hamsters. WNV envelope-staining in these hamsters was only observed in motor neurons of the ipsilateral ventral horn of the spinal cord, but not in the dorsal root ganglion (DRG). To evaluate the consequence of motor neuron infection by WNV, the authors inoculated wheat germ agglutinin-horseradish peroxidase (WGA-HRP) 9 days after WNV sciatic nerve injection, and stained the spinal cord and the DRG for HRP activity 3 days later. The degree of HRP-staining in DRG was the same in WNV- and sham-infected animals, but the HRP-staining in the motor neuron in the ventral horn was considerably less for WNV-infected hamsters. To investigate the mechanism of WNV transport, hamsters were treated with colchicine, an inhibitor of membranous microtubule-mediated transport. The intensity of the WNV-stained area in the spinal cord of colchicine-treated hamsters at 6 days after WNV infection were significantly reduced (P<or=.05) compared to the placebo-treated hamsters. These data suggest that WNV is preferentially transported through the motor axons, but not the sensory axons, to subsequently infect motor neurons and cause motor weakness and paralysis.