Tyrosine hydroxylase gene regulation in human neuronal progenitor cells does not depend on Nurr1 as in the murine and rat systems.

A previous study on the human tyrosine hydroxylase (TH) promoter revealed remarkable differences in the mechanism of TH gene regulation between the human and murine models. Indeed, a low degree of homology was observed in the sequence of TH promoters among human, mouse, and rat systems. Only five short conserved regions (CRs) could be identified among the three species. A human TH minimal promoter was engineered and assembled into a self-inactivating lentiviral vector system. This human TH minimal promoter contained the five CRs plus the first -194 bp from the transcription start of the human TH promoter and the first 35 bp of the untranslated messenger RNA leader of the human TH gene. A significant degree of specificity for this human TH minimal promoter was observed only for human neuronal progenitor cells (hNPCs), but not for TH-positive differentiated mouse primary striatal and substantia nigra cells, indicating a significant difference in TH gene regulation between the human and mouse systems. Not only is the degree of homology between the human and mouse promoters in the range of only 46%, but also those few elements that share a high degree of homology display totally different functions in human and mouse brain-derived cells. In the rodent system, NR4A2 (Nurr1) is required for the transactivation of TH minimal promoters. Intriguingly, neither the dimeric nor the heterodimeric binding sites for Nurr1 are present in the 13 kb DNA sequence that contains the human TH promoter. Instead, the CRs termed one and four of the human TH promoter encode only for a half palindromic binding site sequence for Nurr1, which failed to bind Nurr1 in an in vitro electrophoretic mobility shift assay (EMSA). Additionally, of the three monomeric NGFI-B response element (NBRE) core sites (AGGTCA) and two NBRE-related sites present in the human TH promoter, only one core and two NBRE-related sites formed protein binding complexes. Interestingly, there was no increase of protein binding complex formation upon TH induction and in no case could antibodies supershift Nurr1 from the complex. These findings, taken together, demonstrate that NBRE-related binding sites for Nurr1 do not play a direct role in mediating an interaction between Nurr1 and the human TH promoter. Likewise, immunohistochemical and Western blot analysis have also confirmed that both endogenous and exogenous Nurr1 expression does not positively correlate with TH gene expression in hNPCs, in contrast to the mouse model. In addition, real-time PCR analysis revealed that the downregulation of human Nurr1 gene expression mediated by silencing RNA molecules did not affect human TH gene expression in differentiated hNPCs. A better understanding of human TH gene regulation may have important implications both for the development of novel therapeutic approaches and the study of the pathogenesis of a variety of neurological illnesses, including Parkinson's disease, bipolar disorder, and schizophrenia.