To test the hypothesis that collateral sprouting of motoneurons can occur without the intervention of metabolic changes in the cell body, we examined the levels of growth-associated protein 43 (GAP-43) mRNA in mouse motoneurons induced to sprout by muscle inactivity (following marcaine or botulinum toxin treatment) or by partial denervation. GAP-43 mRNA was selected as an appropriate marker for cell body metabolic changes because it is expressed at low levels in mature motoneurons, but is strongly expressed during developmental or regenerative axonal growth in motoneurons. Sprouting motoneurons were identified by retrograde labelling with fluorescent tracers applied to the muscle in which sprouting occurred. Both a full-length cDNA probe and an oligonucleotide probe were used for in situ hybridization. We were unable to detect any significant increases in GAP-43 mRNA levels in fluorescent motoneurons after any treatment, except 4 days after partial denervation (but not at 2 or 8 days). This amounted to a 1.6-fold increase in signal level compared to control motoneurons, while presumed axotomized motoneurons in the same spinal cords displayed on average an 8.7-fold increase. We conclude that collateral sprouting can occur in motoneurons without a detectable increase in cell body levels of GAP-43 mRNA. The modest increase observed in the 4-day partial denervation situation may be a response to the more vigorous and extensive nodal axonal sprouting occurring in these motoneurons. Our results do not deny a role for pre-existing GAP-43 in collateral sprouting, but support the hypothesis that sprouting can occur in motoneurons without necessarily requiring increased GAP-43 mRNA levels in the cell body.