Mouse knockout solves endocrine puzzle and promotes new pituitary lineage model

Abstract

Spontaneous and genetically engineered mouse mutants have greatly enhanced our understanding of congenital and acquired endocrine disorders in humans. The dwarf mouse identified by George Snell in 1929 revealed the pituitary as the master gland, regulating function of the thyroid, gonads, and growth (Snell 1929; Carsner and Rennels 1960). Report of a missense mutation in the homeodomain of PIT1 in Snell’s dwarf quickly led to the realization that humans with cretinism and growth insufficiency had lesions in the same gene (Li et al. 1990; Tatsumi et al. 1992). This theme has been repeated. As spontaneous and genetically engineered mutants revealed the importance of other transcription factors in mouse pituitary development, corresponding mutations were found in human patients with endocrine deficiencies (Sheng et al. 1996, 1997; Sornson et al. 1996; Cogan et al. 1998; Dattani et al. 1998; Wu et al. 1998; Parks et al. 1999; Netchine et al. 2000; Machinis et al. 2001; Thomas et al. 2001). In general, the mutant mouse phenotypes correspond well with the symptoms reported in human patients. In this issue of Genes & Development, Drouin and colleagues add another gene to this collection, with their report of the features of humans and mice deficient in the T-box transcription factor, TPIT (Pulichino et al. 2003a,b). Pulichino et al. investigated the consequences of TPIT deficiency during mouse development and obtained surprising results that shift the paradigm for pituitary cell specification. In addition, Drouin and collaborators from France, the Netherlands, Germany, and Turkey found TPITmutations in pediatric patients with isolated ACTH deficiency. This demonstrates conservation of the critical role of TPIT in pituitary corticotrope development and function, and reveals an underlying cause of medical emergency in human newborns: neonatal hypoglycemia caused by adrenal insufficiency. Previous studies by Drouin and colleagues focused on pituitary corticotropes, the pituitary cells that coordinate the stress response (Drouin and Goodman 1980). Their strategy has been to identify the trans-acting factors that regulate expression of pro-opiomelanocortin, the protein precursor of adrenocorticotropin (ACTH; Tremblay et al. 1988; Therrien and Drouin 1991, 1993). In 1995 their group identified a gene now called Pitx1 that encodes a paired homeodomain transcription factor that binds bicoid-like sites in the POMC promoter (Lamonerie et al. 1996). Targeted disruption of Pitx1 revealed that it is dispensable for POMC transcription and corticotrope development, probably because of functional overlap with the related Pitx2 gene (Lanctot et al. 1999; Szeto et al. 1999; Suh et al. 2002). Together with Pitx2, Pitx1 acts early in the genetic hierarchy, activating expression of the LIM homeodomain gene Lhx3 and stimulating initial expansion of the pituitary primordium (Tremblay et al. 1998; Suh et al. 2002; S.A. Camper, unpubl.). Although Pitx1 is not essential for corticotrope specification or POMC transcription, it can act synergistically with several other transcription factors to stimulate hormone gene expression, suggesting that it may be important in normal pituitary gland function (Szeto et al. 1996; Amendt et al. 1999; Kurotani et al. 1999; Tremblay and Drouin 1999; Tremblay et al. 1999; Quirk et al. 2001; Quentien et al. 2002a,b; Suszko et al. 2002). The pursuit of genes critical for the corticotrope lineage and POMC transcription next led to the demonstration of synergistic interaction between the basic helix– loop–helix factor NEUROD1 (also known as BETA2), ubiquitous HLH proteins, and PITX1 (Poulin et al. 1997, 2000). Both humans and mice with NEUROD1 deficiency develop diabetes because of abnormalities in pancreas development, but POMC expression is not affected, revealing that NEUROD1 is not essential for POMC transcription (Naya et al. 1997; Malecki et al. 1999; Liu et al. 2001). Like Pitx1 and Neurod1, Nr4a1 (also known as NUR77 or NFGI-B) enhances POMC transcription in cell culture, but it is dispensable for POMC transcription mice (Crawford et al. 1995; Philips et al. 1997a,b). Collectively, these observations suggested that POMC transcription relies on a combination Corresponding authors. E-MAIL keeganc@med.umich.edu; FAX (734) 936-6684. E-MAIL scamper@umich.edu; FAX (734) 763-7672. Article and publication are at http://www.genesdev.org/cgi/doi/10.1101/ gad.1085903.