One more factor joins the plot: Pbx1 regulates differentiation and survival of midbrain dopaminergic neurons

Abstract

Midbrain dopaminergic neurons have been the subject of intense research, because of their importance in pathologies such as Parkinson's disease. In this issue of The EMBO Journal, Villaescusa et al (2016) show the transcription factor Pbx1 is required for the differentiation and survival of midbrain dopaminergic neurons. Notably, Pbx1 protects from oxidative stress by inducing Nfe2l1, and the expression of both genes is reduced in dopaminergic neurons from Parkinson's patients. This is an important study, with possible implications in regenerative medicine and drug design.

How the myriad of cell types that make up the nervous system are specified during embryogenesis has been a major focus of developmental neuroscience. One of the best studied cases is that of dopaminergic neurons of the substantia nigra (SN), a nucleus of cells in the midbrain whose degeneration is the main cause of Parkinson's disease (PD) (Lees et al, 2009). Two main reasons have driven the efforts to unravel the molecular mechanisms controlling the differentiation and maintenance of midbrain dopaminergic (mDA) neurons. First, PD is one of the most promising cases for the use of cell replacement therapies, urging the establishment of in vitro culture protocols that could function as a source of cells for regenerative medicine. Second, our knowledge on the etiology of PD is still scarce, and understanding how mDA neurons are specified and maintained in the developing embryo may shed light on the cellular processes and pathways that are disrupted in pathological conditions.

Given their importance in differentiation and reprogramming, the role played by sequence‐specific transcription factors has been the subject of intense scrutiny, with mouse genetics being at center stage. More recently, the advent of genomic approaches has allowed for the extensive characterization of genetic programs driven by transcription factors, providing important mechanistic insights (Farnham, 2009). In some cases, the identification of target genes uncovers additional roles that may be masked by the complexity of phenotypes observed in mouse models. Recent work by Villaescusa et al (2016) is a fine example of a very effective combination of mouse genetic and genomic approaches, identifying the homeobox transcription factor Pbx1 as a novel regulator of development and survival of mDA neurons.

Multiple studies have focused on the role of Pbx1 in organogenesis and cell differentiation, but a regulatory function in mDA neurons had not been fully investigated (Sgadò et al, 2012). The authors were prompted to consider this possibility due to the specific expression of Pbx1 (its full‐length isoform Pbx1a) in the intermediate and marginal zones of the midbrain floorplate, the region where post‐mitotic neuroblasts and mDA neurons reside. A detailed analysis during neurogenesis placed the onset of Pbx1 expression after that of Nurr1, but before Pitx3, two other transcription factors that promote maturation and survival of mDA neurons (Zetterstro et al, 1997; Nunes et al, 2003) (see Fig 1). Of significance was also the persistence of Pbx1 expression in the SN of adult mice, altogether suggesting roles in both differentiating and adult mDA neurons.

Figure 1. Pbx1 is a new intrinsic regulator of mid‐brain dopaminergic (mDA) neurons.

(A) Schematic representation of the mDA neuron lineage, along which Pbx1 (its full‐length isoform Pbx1a) is expressed in post‐mitotic neuroblasts (co‐expressing Nurr1) and in mDA neurons that also co‐express Pitx3 and TH. (B) Characterization of the Pbx1 transcriptional program revealed that Pbx1 both activates and represses gene expression. Pbx1 activates Pitx3, thereby promoting the mDA neuron fate, while simultaneously repressing Onecut2 and alternative lateral fate identity. In addition, Pbx1 promotes the survival of mDA neurons by activating Nfe2l1. Reduced levels of Pbx1 and Nfe2l1 were found in mDA neurons from PD patients, establishing these transcription factors as potential targets for drug design.

To investigate the function of Pbx1 in this important lineage, Villaescusa et al (2016) analyzed the midbrain of Pbx1 knockout embryos. The phenotype was relatively mild, with ablation of the Pbx1 gene resulting only in a small decrease in the number of mDA neurons. Loss of Pbx1 occurred with the concomitant up‐regulation of the closely related factor Pbx3, suggesting a more severe phenotype may be compensated by genetic redundancy. Because Pbx1/3 double knockout embryos are not viable, a conditional strategy was followed to restrict the ablation of both Pbx genes to cells expressing tyrosine hydroxylase (TH), the rate‐limiting enzyme for the synthesis of DA. This resulted in progressive decrease of mDA neurons at mid‐embryonic stages, with increased cell death and a massive loss of this population before birth. Thus, Pbx1 is required for the proper differentiation of post‐mitotic precursors into mature mDA neurons, and their survival.

Recent years have seen great advances in the refinement of cell culture protocols, whereby neural stem/progenitor cells are induced to differentiate into mDA‐like neurons upon exposure to a combination of extrinsic factors (Arenas et al, 2015). Having shown the requirement of Pbx1 in mDA neuron differentiation, the authors proceeded to test whether Pbx1 expression would enhance the efficacy of one of such protocols. Indeed, lentiviral mediated expression of Pbx1 significantly improved the yield (2.3‐fold) of TH⁺ cells generated from a human neuroepithelial stem cell line of iPS origin. These are promising results that will prompt further testing of Pbx1 expression in other cellular models, including more challenging attempts to obtain mDA neurons from differentiated somatic cells by direct reprogramming.

Having found the importance of Pbx1 in mDA neurons, Villaescusa et al (2016) proceeded to identify its transcriptional targets, using chromatin immunoprecipitation followed by deep‐sequencing (ChIP‐seq), starting with chromatin extracted from dissected ventral midbrain of mouse embryos. Gene Set Enrichment Analysis (GSEA) was used to compare the genomic location of Pbx1 with the transcriptome of mDA neurons (Subramanian et al, 2005). Pbx1 binds to genes that are either enriched or depleted in mDA neurons, indicating that it functions both as an activator and repressor of gene expression. The molecular basis for this dual activity remains to be understood. The promoter context is likely to play a role, as consensus binding sequences for distinct families of transcription factors are found near Pbx1 sites in the vicinity of activated and repressed target genes. In addition, Pbx proteins form heterodimeric complexes with distinct transcription factors, which could differ in their transcriptional activities (Cerdá‐Esteban & Spagnoli, 2014). In either case scenario, further work will be required to identify partners of Pbx1 in mDA neurons.

Striking insights were provided by the identity of Pbx1 target genes, many of which encode proteins involved in signal transduction, RNA metabolism, transcription and response to stress. The two most prominent targets were Pitx3 and Onecut2, which were activated and repressed by Pbx1, respectively. While Pitx3 is essential for mDA neuron maturation, expression of Onecut2 in midbrain occurs laterally and excluded from the Pbx1 domain. Thus, regulation of these two target genes demonstrates how induction of a cell type and repression of alternative fates can be coordinated by a single transcription factor. Most striking, however, was the finding of Nfe2l1 (aka Nrf1) amongst genes activated by Pbx1. Nfe2l1 is a member of a family of leucine zipper transcription factors that activate the expression of cytoprotective genes, both in homeostatic conditions and in response to toxicological insults, particularly by oxidative stress (Jennings et al, 2013).

From these studies, Pbx1 emerges as an important player in the differentiation and survival of mDA neurons. But how significant are these observations in the context of the human brain, and in particular in PD? Quite relevantly, Pbx1 is expressed in all TH⁺ cells in human embryonic midbrain during the peak of mDA neurogenesis, indicating its function may be evolutionarily conserved. The most striking observation, however, was made in the context of the human pathology. Analysis of brain samples from PD patients showed a significant reduction of Pbx1 expression in mDA neurons when compared to aged‐matched controls. In addition, while Nfe2l1 was detected in 20–50% mDA neurons of control donors, its expression was abolished or at least strongly reduced (< 10%) in PD patients. Oxidative stress is a well‐known pathogenetic mechanism in PD, and the expression of the related factor Nfe2l2 (aka Nrf2) was shown to be strongly induced in the surviving mDA neurons from PD patients (Ramsey et al, 2007). However, Nfe2l2 is not present in the nucleus of mDA neurons from healthy brains, which together with the observations in the present study suggests Nfe2l1 has a prominent cytoprotective role in homeostatic conditions (Ramsey et al, 2007). The reduced levels of this gene and of its direct activator Pbx1 in PD patients suggest these could be targets for new drugs to be developed. Modeling this pathway would aim at reestablishing neuroprotection in mDA neurons, in particular against oxidative stress, with the ambitious prospect of altering disease progression. This would be an important paradigm shift, since current therapies either by cell replacement or pharmacological approaches aim at restoring neurotransmitter balance, without having, however, the capability to alter the course of the disease.

See also: JC Villaescusa et al (September 2016)