Recently, it has been shown that different gene expression patterns delineate five progenitor domains in the developing ventral neural tube. Four of these domains (p0-p3) give rise to interneurons (V0-V3), while the other (pMn) gives rise to motoneurons. Although the cell migration patterns for these domains have been characterised in the embryo, the final somatic location and axonal connections of cells derived from each domain have yet to be determined in the adult. We have investigated the transmitter phenotype and connectivity of the V0 and V2 interneurons, by using expression of GFP driven by markers that are transiently expressed in the p0 and p2 domains (Dbx1 and Lim3 respectively)The relationships between progenitor-domain derived interneurons and their targets were examined with immunohistochemistry and confocal microscopy. To determine the transmitter phenotype, tissue sections were cut from mid-lumbar segments and reacted with antibodies against glutamic acid decarboxylase (GAD65, GAD67) and the neuronal glycine transporter GLYT2 (to identify GABAergic and glycinergic axon terminals), and against the vesicular glutamate transporters VGLUT1 and VGLUT2 (to identify glutamatergic axon terminals). Animals were anaesthetized with sodium pentobarbitone (1ml I.P.), killed humanely and fixed by perfusion.Dbx1-GFP cells made up approximately 40% of ventral horn neurons (excluding motoneurons) and included populations of inhibitory and excitatory interneurons. The Dbx1GFP cells in the ventral horn occupied a region extending laterally from the central canal, and also passing down the medial and lateral aspects of the ventral horn. Fewer Dbx1-GFP cells were found in the central part of lamina VII, and they were rare in the motor nuclei. Dbx1-GFP axon-terminals were distributed throughout the ventral horn and innervated the motor nuclei. Immunofluorescent labelling of GFP and ChAT (choline acetyl transferase) suggests that the large cholinergic terminals on motoneurons, C-boutons (Boidan 1966, Conradi 1969, Nagy et al 1993), are derived from the Dbx1-expressing cell population and probably originate from a group of cells lateral to the central canal in medial lamina VII and lamina X. Cholinergic cells in this area were previously shown to be immunoreactive for c-Fos following induction of fictive locomotion (Huang et al 2000). We also used GFP linked to a gene expressed in motoneurons (ISL-1) as an exclusion-marker, to show that C-boutons are not motoneuron collaterals. This was previously suspected because of the different vesicle constituents compared to motoneuron axon terminals. In contrast to the Dbx1 interneurons, the Lim3GFP cells were less numerous and most of their axon terminals were VGLUT2-immunoreactive. Some were immunoreactive for GLYT2 but they were very rarely positive for GAD. Lim3 axon-terminals were mainly distributed in the ventral horn, including the motor nuclei and around the central canal, although they were less evident in the central part of lamina VII. In the dorsal horn, axon terminals were distributed evenly across lamina VI with some axonal arbors in medial and central lamina V.In addition to examining cell populations derived from progenitor domains, we also assessed the transmitter phenotype and somatic location of cells labelled by an HB9-GFP construct. Previous studies of cells expressing HB9-GFP during chick embryo development indicated that HB9 expression was restricted to motoneurons. (Arber et al 1999). However, in our HB9-GFP construct (Wichterle et al 2002), GFP expression was also observed in interneurons throughout lamina VII, Lamina VIII and Lamina X in the adult mouse spinal cord. In particular, the transgene highlighted a population of interneurons in lamina VIII that receive GABAergic contacts on their somata and proximal dendrites. There also appeared to be glutamatergic interneuron contacts on proximal dendrites of motoneurons but not on their somata. In summary: These data suggest that progenitor domains in the developing neural tube give rise to functionally distinct populations of cells in the adult mouse. They also show that reporter genes can assist in the targeting of functional clusters for anatomical analysis. Using these methods, we have identified a population of cells lateral to the central canal that probably give rise to the large cholinergic terminals on motoneurons. We have also shown that HB9-GFP expression can be used to identify a novel cluster of interneurons in lamina VIII.