Although oligodendrocytes are widely distributed throughout the adult central nervous system, they originate in restricted domains in the embryonic neural tube. One such domain resides in the ventral neuroepithelium of the developing spinal cord, and it has been suggested that this domain might supply the full complement of oligodendrocytes across the dorsoventral axis. However, two recent reports in Neuron identify a more dorsal domain that could be responsible for a second phase of oligodendrogenesis.

Ventrally, oligodendrocytes are produced in the pMN domain, which initially generates motor neurons, but later switches to producing oligodendrocytes under the influence of sonic hedgehog (SHH) signalling and the transcription factors NKX6.1 and NKX6.2. To find out whether oligodendrocytes could be generated in the absence of these factors, Cai et al. and Vallstedt et al. examined mouse embryos from an Nkx6.1−/−;Nkx6.2−/− double mutant line, and Cai et al. also investigated a Shh−/− line. In both cases, the pMN domain failed to produce oligodendrocytes. Nevertheless, some oligodendrocyte precursor cells (OLPs) — identified by the expression of Olig2 and PDGFRα (platelet-derived growth factor receptor α) — were still generated, albeit some time after the usual time of onset of oligodendrogenesis. These OLPs originated near the midline in the dorsal spinal cord.

On their own, these findings do not necessarily indicate that the dorsal spinal cord normally produces oligodendrocytes — the OLPs could have migrated from the ventral neural tube, or the dorsal neuroepithelium could have been respecified in the mutant embryos. However, closer inspection of wild-type embryos revealed that an Olig2-expressing domain is normally present in the dorsal spinal cord, and in both mutant and wild-type embryos, the cells in this domain also expressed the dorsal neuroepithelial markers Pax7 and Mash1 . In addition, Cai et al. showed that if the dorsal spinal cord was explanted at embryonic day 11.5 (before the onset of oligodendrogenesis) and grown in culture, the tissue still generated OLPs. Therefore, OLP production in the dorsal spinal cord does not require signals or cellular contribution from the ventral neural tube.

The findings of Cai et al. and Vallstedt et al. support a model in which the dorsal domain is responsible for a phase of oligodendrogenesis that begins around embryonic day 14.5 — 2 days later than in the ventral spinal cord. It was not possible to determine whether the dorsal OLPs develop into mature myelinating oligodendrocytes, because the mutants that were investigated in these studies died at birth. Therefore, further studies will be required to establish what contribution, if any, the dorsal domain makes to the oligodendrocyte population of the mature spinal cord.