Function of the adult nervous system relies on the appropriate establishment of neural circuits during embryogenesis. In vertebrates, the neurons that make up motor circuits form in distinct domains along the dorsoventral (DV) axis of the neural tube. Each domain is characterized by a unique combination of transcription factors (TFs) that promote a specific fate, while repressing the fates of adjacent domains. The prdm12 TF is required for the expression of eng1b and the generation of V1 interneurons in the p1 domain, but the details of its function remain unclear. We used CRISPR/Cas9 genome editing technology to generate the first germline mutants for the prdm12 gene and used this resource, together with classical luciferase reporter assays and co-immunoprecipitation experiments, to study prdm12b function in zebrafish. We also generated germline mutants for bhlhe22 and nkx6.1 to examine how these TFs act with prdm12b to control p1 formation. We find that prdm12b mutants lack eng1b expression in the p1 domain and also possess an abnormal Mauthner cell-dependent escape response. Using cell culture-based luciferase reporter assays, we demonstrate that Prdm12b acts as transcriptional repressor, most likely by recruiting EHMT2/G9a. We also show that the Bhlhe22 TF binds to the Prdm12b zinc finger domain to form a Bhlhe22:Prdm12b complex. However, bhlhe22 mutants display normal eng1b expression in the p1 domain. While prdm12 has been proposed to promote p1 fates by repressing expression of the nkx6.1 TF, we do not observe an expansion of the nkx6.1 domain upon loss of prdm12b function, nor is eng1b expression restored upon simultaneous loss of prdm12b and nkx6.1. We conclude that prdm12b germline mutations produce a phenotype that is indistinguishable from that of morpholino-mediated loss of prdm12 function. In terms of prdm12b function, our results indicate that Prdm12b acts as transcriptional repressor and interacts with both EHMT2/G9a and Bhlhe22. However, bhlhe22 function is not required for eng1b expression in vivo, perhaps indicating that other bhlh genes can compensate for its loss during embryogenesis. Lastly, we do not find evidence for nkx6.1 and prdm12b acting as a repressive pair in the formation of the p1 domain – suggesting that prdm12b is not solely required to repress non-p1 fates, but is also needed to promote p1 fates.