Resumen de GATA4 links differentiation and inflammation in pancreatic carcinogenesis
Mónica Victoria Pérez de Andrés
Gata4 and Gata6 are necessary for murine pancreas development and, recently, GATA6 has been described to be an important regulator of pancreatic acinar cell differentiation. In addition, a number of studies have shown that GATA6 has a tumor suppressor role in pancreatic carcinogenesis, both in mouse and human. Here, I have characterized Gata4 function in the murine pancreas by specifically deleting the gene during embryonic development and have found that, unlike GATA6, GATA4 is not required to maintain acinar homeostasis in the adult. Strikingly, Gata4 deletion in a mutant Kras-induced model of pancreatic carcinogenesis (Kras*;Gata4-/- mice) led to development of pancreatic cancer (PDAC) in the absence of Pancreatic Intraepithelial Neoplasias (PanINs), which are believed to be the most common PDAC precursors.
To investigate the origin of PDAC present in Kras*;Gata4-/-, a thorough histological analysis of Kras* and Kras*;Gata4-/- pancreata was performed revealing the presence of a new potential precursor lesion that had not been described before and was designated as PALe (Pale Acinar Lesion). I have performed a systematic characterization of PALes in Kras* and Kras*;Gata4-/- contexts, unraveling their similarities and differences when compared to acino-ductal metaplasia and PanINs. Quantification of PALes in Trp53-proficient and -null contexts revealed an increased prevalence upon Trp53 loss. This was linked to increased proliferation rates when compared to Trp53-proficient PALes, supporting their role as potential precursor lesions of PDAC.
Transcriptomic analyses on Kras* and Kras*;Gata4-/- showed that Gata4 is required for Kras*-induced inflammation. To understand the effects resulting from Gata4 inactivation, I used a combination of transcriptomic, chromatin immunoprecipitation and mass spectrometry analyses and identified Reg3b and Cxcl12 as two possible mediators of PanIN formation. Nevertheless, Reg3b administration and inhibition or the Cxcl12 receptor did not impact on PanIN development. To understand the shared and unique functions of Gata4 and Gata6, I analyzed transcriptomic and ChIP-Seq and found that they share most of their binding sites and share some transcriptomic effects but regulate inflammation in an opposite fashion.
Finally, I studied the role of Gata4 in the tumor phenotype. Survival analysis of Kras* and Kras*;Gata4-/- mice showed a significantly decreased survival upon Gata4 loss, uncovering its role as a tumor suppressor. A study in a large series of mouse PDAC samples revealed that GATA6 and GATA4 expression correlate and are strongly associated with PDAC differentiation. Expression of GATA4 and GATA6 in a large cohort of human PDAC samples revealed that reduced expression of both proteins in tumors was associated with the worst patient survival providing compelling clinical rationale for assessing their expression as biomarkers of poor prognosis.
Overall, this work highlights the relevant role of Gata4 as a mediator of PanIN formation and as a tumor suppressor in PDAC, while uncovering the presence of a new potential precursor lesion of PDAC
