Grapin-Botton Anne, Associate scientist e-mail pdf group members publications
Anne Grapin-Botton studied biology in Paris (France) with Nicole Le Douarin and received her PhD in 1995. She carried out postdoctoral work with Doug Melton in Harvard University (Cambridge, USA) and joined ISREC as an associate scientist in 2001. In 2005, she was appointed Tenure Track Assistant Professor in the School of Life Sciences at EPFL.
Pancreas development and cancer
The definitive endoderm gives rise to the epithelium of the digestive tract, the respiratory system, and several endocrine and exocrine glands. Our experiments in mouse and chick embryos are designed to:
(I) identify the signals that induce different organs at different positions in the digestive tract
(II) determine the pancreas-specific transcription factors induced and how their network controls organogenesis and
(III) determine how these signals lead to cancer or diabetes when inaccurately controlled.

Organizing endoderm organs along the body axis

Appropriate and reproducible organ layout is generally established at an early stage of development by the expression of patterning genes. These genes are expressed in restricted areas of the embryo and control differentiation and morphogenesis. In vertebrates, their activation occurs in the three germ layers after gastrulation. Although the ectoderm and mesoderm have been the focus of intensive work aimed at understanding how these layers are patterned, and how the patterning genes control differentiation, the endoderm has received less attention. Because signaling must occur between germ layers to achieve a properly organized body, our understanding of the coordinated development of all organs requires a more thorough consideration of the endoderm and its derivatives.

Our experiments demonstrated that the mesoderm at different locations sends signals that instruct the endoderm of its identity along the antero-posterior (A‑P, mouth to anus) axis (Kumar et al., 2003) (Figure 1). We recently demonstrated that the most anterior endoderm forms in the absence of FGF4 and that more posterior endoderm is progressively induced as the amount of FGF4 received increases (Dessimoz et al., 2006). FGF signaling from endoderm formation to gut tube closure is absolutely essential to form mid- and hindgut and accurately maintain the boundaries between different digestive tract organs. Our ongoing experiments suggest that accurate levels of retinoic acid (Figure 1) and Wnt signaling are also required in addition to FGFs. Taken together these observations suggest that the mesoderm sends similar signals to define regions in the nervous system and digestive tract.

Figure 1: Graded activity of Fgf4 (orange) and retinoic acid (blue) integrated over time induce specific transcription factors at specific positions along the antero-posterior axis in endoderm in the early embryo. These patterning factors then induce the formation of specific endoderm-derived organs and segments of the digestive tract.

Regulation of pancreas organogenesis: role of Neurogenin 3 (Ngn3) in endocrine cell differentiation and migration

Signaling activity from the mesoderm induces pancreatic genes at a given position along the A‑P axis. As a result exocrine and endocrine cells will differentiate in the pancreas. The transcription factor Ngn3 is absolutely necessary to generate endocrine cells. All pancreatic endocrine cells, producing glucagon, insulin, somatostatin or PP, differentiate from Pdx1+ progenitors that transiently express Neurogenin3 (Figure 2). To understand whether the competence of pancreatic progenitors changes over time, we generated transgenic mice expressing a tamoxifen-inducible Ngn3 fusion protein under the control of the Pdx1 promoter and backcrossed the transgene into the Ngn3-/- background, devoid of endogenous endocrine cells (Johansson et al., 2007). Early activation of Ngn3ERTM almost exclusively induced glucagon+ cells, while depleting the pool of pancreas progenitors. As from E11.5, Pdx1+ progenitors became competent to differentiate into insulin+ and PP+ cells. Somatostatin+ cells were generated from E14.5 while the competence to make glucagon+ cells was dramatically decreased. Hence pancreas progenitors, similar to retinal or cortical progenitors, go through competence states that each allow the generation of a subset of cell types. We further showed that the progenitors acquire the competence to generate late-born cells in a mechanism that is intrinsic to the epithelium. Ongoing experiments aim at understanding whether the change in competence is cell autonomous and what its molecular basis is.

This transgenic line was also used to identify targets of Ngn3 that may unravel the molecular mechanisms by which Ngn3 promotes endocrine cell migration and aggregation, a process relevant to the formation of islets of Langerhans.

Figure 2: Expression of Ngn3 (red) in a subset of pancreas progenitors at 14.5 days of development initiates their endocrine differentiation. Exocrine cells are in green.

Regulation of pancreas organogenesis: role of the Wnt pathway

Cell to cell signaling is important not only to initially induce the pancreas but also to generate specific cell types. We recently found that the Wnt pathway is active in endocrine cells during development (Dessimoz et al., 2005) (Figure 3). Inactivation of β‑catenin in the pancreas epithelium invalidates this pathway and results in reduced numbers of endocrine cells. Our ongoing efforts are aimed at understanding the mechanisms leading to this effect. In addition to its role in the Wnt pathway, β‑catenin is important for cell-cell adhesion. This may explain why exocrine cells in which Wnt pathway activity is not detected are also affected by β‑catenin inactivation. Activating β‑catenin mutations and Adenomatous Polyposis Coli (APC) loss of function lead to acinar cell carcinoma and pancreatoblastoma. Ongoing experiments are aimed at understanding how β‑catenin functions in adhesion and Wnt pathway activity lead to pancreatic cancer.

Figure 3: Wnt Pathway activity (black dots) in a pancreas section of a BAT‑gal mouse at 14.5 dpc. Endocrine cells are in red. Blue cells are exocrine cells and pancreas progenitors.

Significance of this work for diabetes and pancreatic cancer

Although the function of the pancreas in controlling glucose homeostasis is compensated by insulin injection in diabetic patients, the physiological effects are inexact and too variable. Among approaches that are currently being explored to find a cure for diabetes are the isolation and propagation of embryonic or adult stem cells that can be engineered to produce endocrine hormones and then transplanted to patients. Our experiments are aimed at identifying the critical cellular transcription factors and signaling molecules that are sufficient to transform cells into pancreas and β cells. Once found, the factors may be introduced in stem cells in vivo or in vitro to force their differentiation into β cells.

In addition, pre-cancerous and cancerous cells often reactivate the expression of developmental genes. In pancreatic carcinoma Pdx1 is up-regulated in pre-malignant metaplastic ductal epithelium. Our results show that Pdx1 expression is associated with enhanced invasive capacities of cells in the embryo. Further work on developmental genes may give a better understanding of pancreas cancer development and may point to new therapeutic targets.

Keywords

Development, embryo, gut, pancreas, diabetes, endoderm, Wnt, patterning, beta-cell, chick, mouse