A Beautiful Cell (or Two or Three?)

Abstract

The intestinal tract is the largest endocrine organ of the body, containing at least 15 different cells types that release more than 100 biologically active peptides and hormones [Fig. 1 (1–3)]. The vast majority of these enteroendocrine cells have been identified using immunohistochemical techniques, an approach that is inherently limited by the specificity of the antisera/antibodies, particularly because they are used in high concentrations with this technique (1). The current study by Habib et al. (4) takes advantage of two newly developed fluorescent enteroendocrine cell models (5, 6) to purify specific cells of interest, thereby permitting deeper interrogation of the gene profiles of these cells. The findings both confirm and extend previous observations that different enteroendocrine cell types have more features in common than originally believed. These findings have important implications for current investigations into the possible use of enteroendocrine cell stimulators for the treatment of diseases such as type 2 diabetes. The enteroendocrine cell of focus in the present study by Habib et al. (4) is the L cell, originally described on the basis of enteroglucagon immunoreactivity, using the K cell that produces glucose-dependent insulinotropic peptide (GIP) as a comparator. Enteroglucagon is a generic term for the intestinal proglucagon-derived peptides, including glucagon-like peptide (GLP)-1, an incretin hormone that stimulates glucose-dependent insulin secretion; both GLP-1 receptor agonists and degradation inhibitors are currently used in the clinic to treat patients with type 2 diabetes (7, 8). The other main L cell proglucagon-derived peptides are the gut growth factor GLP-2 and the anorexigenic hormone oxyntomodulin, whereas GIP also serves a role as an incretin hormone (7). Based on the expression of GIP vs. GLP-1, the highest numbers of K cells are found in the duodenum with decreasing numbers in the aboral direction, whereas L cells demonstrate the opposite gradient, with the greatest numbers in the ileum/colon (9, 10). However, the presence of GLP-1 has been demonstrated in subsets of duodenal K cells, whereas immunoreactive GIP has been detected in subpopulations of proximal L cells (5, 9, 11, 12). To further complicate the issue, numerous studies have also demonstrated the expression of peptide YY (PYY) in some ileal and colonic L cells (5, 10, 13, 14) as well as coexpression of GLP-1 with cholecystokinin (CCK) and with neurotensin (14, 15). Thus, it has long been suspected that different types of K and L cells may exist in the intestine. Studies on enteroendocrine cell development have elucidated the transcription factors required for lineage progression from early stem cells to differentiated enteroendocrine cells (Fig. 1). Thus, the development of related enteroendocrine cell populations requires sequential expression of Math1 (mammalian atonal homolog-1), Ngn3 (neurogenin 3), NeuroD (neurogenic differentiation), followed by the expression of paired box transcription factor (Pax)-4 and/or ARX (aristaless related homeobox) (16– 21). The specific transcription factors required for final differentiation of these progenitor cells into specific enteroendocrine cells are not well characterized but include both Pax6 and pancreatic duodenal homeobox-1 (Pdx1) for K cells as well as Pax6 for L cells: in this regard, expression of Pdx1 in L cells appears to be permissive for their coexpression of GIP (22). Recent findings of Dtx1 (deltex1),Egr2 (earlygrowth responseprotein2),Tbx3 (T box 3) in duodenal but not colonic L cells and Etv1 (ETS translocation variant 1) and Prox1 (prospero-related homeobox 1) in proximal L cells but not in K cells remain of uncertain significance (4). In the current study by Habib et al. (4), intestinal L cells from transgenic mice expressing a fluorescent marker (Ve-