Identifying the steps involved in striatal development is important both for understanding the striatum in health and disease, and for generating protocols to differentiate striatal neurons for regenerative medicine. The most prominent neuronal subtype in the adult striatum is the medium spiny projection neuron (MSN), which constitutes more than 85% of all striatal neurons and classically expresses DARPP-32. Through a microarray study of genes expressed in the whole ganglionic eminence (WGE: the developing striatum) in the mouse, we identified the gene encoding the transcription factor Forkhead box protein P1 (FoxP1) as the most highly up-regulated gene, thus providing unbiased evidence for the association of FoxP1 with MSN development. We also describe the expression of FoxP1 in the human fetal brain over equivalent gestational stages. FoxP1 expression persisted through into adulthood in the mouse brain, where it co-localised with all striatal DARPP-32 positive projection neurons and a small population of DARPP-32 negative cells. There was no co-localisation of FoxP1 with any interneuron markers. FoxP1 was detectable in primary fetal striatal cells following dissection, culture, and transplantation into the adult lesioned striatum, demonstrating its utility as an MSN marker for transplantation studies. Furthermore, DARPP-32 expression was absent from FoxP1 knock-out mouse WGE differentiated in vitro, suggesting that FoxP1 is important for the development of DARPP-32-positive MSNs. In summary, we show that FoxP1 labels MSN precursors prior to the expression of DARPP-32 during normal development, and in addition suggest that FoxP1 labels a sub-population of MSNs that are not co-labelled by DARPP-32. We demonstrate the utility of FoxP1 to label MSNs in vitro and following neural transplantation, and show that FoxP1 is required for DARPP-32 positive MSN differentiation in vitro.

Huntington’s disease (HD) is an autosomal dominant inherited disorder leading to the loss inter alia of DARPP-32 positive medium spiny projection neurons (“MSNs”) in the striatum. There is no known cure for HD but the relative specificity of cell loss early in the disease has made cell replacement by neural transplantation an attractive therapeutic possibility. Transplantation of human fetal striatal precursor cells has shown “proof-of-principle” in clinical trials; however, the practical and ethical difficulties associated with sourcing fetal tissues have stimulated the need to identify alternative source(s) of donor cells that are more readily available and more suitable for standardization. We now have available the first generation of protocols to generate DARPP-32 positive MSN-like neurons from pluripotent stem cells and these have been successfully grafted into animal models of HD. However, whether these grafts can provide stable functional recovery to the level that can regularly be achieved with primary fetal striatal grafts remains to be demonstrated. Of particular concern, primary fetal striatal grafts are not homogenous; they contain not only the MSN subpopulation of striatal projection neurons but also include all the different cell types that make up the mature striatum, such as the multiple populations of striatal interneurons and striatal glia, and which certainly contribute to normal striatal function. By contrast, present protocols for pluripotent stem cell differentiation are almost entirely targeted at specifying just neurons of an MSN lineage. So far, evidence for the functionality and integration of stem-cell derived grafts is correspondingly limited. Indeed, consideration of the features of full striatal reconstruction that is achieved with primary fetal striatal grafts suggests that optimal success of the next generations of stem cell-derived replacement therapy in HD will require that graft protocols be developed to allow inclusion of multiple striatal cell types, such as interneurons and/or glia. Almost certainly, therefore, more sophisticated differentiation protocols will be necessary, over and above replacement of a specific population of MSNs. A rational solution to this technical challenge requires that we re-address the underlying question—what constitutes a functional striatal graft?