The young and the Wnt-less: transplantable fetal intestinal spheroids without Wnts

Abstract

The differences between fetal and adult intestinal stem cells are unclear, and understanding this relationship could present novel therapeutic opportunities. Fordham et al. (2013) and Mustata et al. (2013) report a potential source of transplantable epithelial cells from fetal gut which can convert into adult intestinal stem and differentiated cells.

In the developing mouse small intestine, epithelial stem cells that line the newly formed intestine are distinct from those during late prenatal and postnatal stages. In the ‘early’ prenatal stage (~E10–E15.5) the epithelium is pseudo-stratified and shows little evidence of differentiation (Kaufmann, 1992). In contrast, during the ‘late’ prenatal stage (E15.5–E18) epithelial stem cells form a single cell layer and constitutively produce differentiated cells. The major role of the early prenatal stem cells is to ensure that rapid expansion of the epithelium occurs in concert with the elongation of the underlying muscle wall. Intestinal elongation during this phase is critical for initiating patterned movements of the intestine which position the gastrointestinal organs with respect to the abdominal body wall. In contrast, late stage prenatal stem cells function to pattern surfaces that create an absorptive barrier that is required after birth.

An important question is whether the functional differences between these stem cell populations could be recapitulated in vitro, as this would allow detailed comparisons between them. Two recent publications, one in this issue of Cell Stem Cell (Fordham et al., 2013) and the other in a recent issue of Cell Reports (Mustata et al., 2013), have tackled this question and examined the in vitro growth of intestinal epithelial cells isolated from various stages in pre- and post-natal mouse development. In addition,Fordham et al. (2013) extended their findings to fetal human intestine and human induced pluripotent embryonic stem cells (hiPS cells).

Both studies report findings in mice, which are concordant and quite striking. Both groups adopted methods developed to grow adult intestinal epithelial cells, which form complex structures termed organoids that consist of both stem cells and their differentiated progeny (Sato et al. 2009). These organoid cultures require canonical Wnt ligands, R-spondin, and Noggin to propagate.Mustata et al. (2013) found that cells isolated from the early prenatal stage grew in vitro almost exclusively as spheroids and not organoids. This morphologic distinction is quite important. Unlike organoids, spheroids are composed primarily of stem cells and lack budding and branching structures enriched for differentiated cells. The early fetal spheroids required Lgr4 for their growth and did not express the adult stem cell marker Lgr5.Mustata et al. (2013) also identified Cnx43 as an additional stem cell marker for early fetal spheroids. In contrast to late prenatal- and postnatal-derived organoids, early fetal spheroids did not require canonical Wnt ligands for maintenance of stem cells.

However, spheroids display complex responses to Wnt stimulation. In general, canonical Wnts are known primarily as pro-differentiation factors during development and as growth factors in adult stem cells. Here, Fordham et al. (2013) showed that Wnt signaling promoted the conversion of fetal spheroids into organoids. On the other hand,Mustata et al. (2013) showed that this conversion was induced by a γ-secretase inhibitor, which blocks Notch signaling. Notch inhibition induces differentiation of stem cells to secretary epithelial cell lineages which produce Wnt ligands (Sato et al. 2011a, Fordham et al. 2013). These two findings suggest that maturation of fetal intestinal stem cells proceeds in concert with establishment of the intestinal stem cell niche, which includes Wnt-producing mesenchymal and epithelial components. Importantly, Fordham and colleagues demonstrate that human fetal (10 week-old) intestinal cells can also be propagated as spheroids. They also demonstrate that a similar cell population highly resembling fetal intestinal progenitors can be obtained from human induced pluripotent stem cells (hiPSCs). They used similar growth conditions as for adult mouse organoids, greatly simplifying ex vivo culturing of human intestinal progenitors and suggesting a conserved developmental mechanism for intestinal growth.

Fordham et al. then assessed whether early fetal spheroids could be a good source of cells for transplantation and use in colonic regeneration. To test this possibility, they performed a proof-of-principle experiment by transplanting fetal mouse-derived spheroids into immunodeficient animals with injured colons. They observed incorporation of the transplanted cells into the regenerating colonic mucosa, reminiscent of previous reports using adult mouse colonic organoids (Yui et al., 2012) and suggesting that fetal stem cells could be utilized for regenerative therapy. One potential advantage of fetal spheroids is that they are enriched for stem cells, so there is no need to purify cell populations. Additionally, use of fetal cells could enable more timely administration of therapies, given the difficulties of isolating and propagating cells from human intestinal biopsies in a reasonable timeframe. The downsides to the use of fetal cells are potentially significant ethical issues as well as the need for immunosuppression in patients that receive these transplants. These hurdles could potentially be surmounted through the use of human induced pluripotent stem cells (hiPSCs), which would permit autologous transplants of mucosal epithelia. In vitro differentiation of hiPSCs into intestinal epithelium was first shown bySpence et al. (2011). In their study, Fordham et al. successfully derived intestinal spheroids from hiPSCs. Although these hiPSC-derived intestinal cells need to be further investigated to confirm their differentiation potential into adult colonic crypts, there is no doubt that these studies surmount a major translational roadblock and that this approach has potential clinical utility.

Together these papers raise several points for future investigation, including the relation between stem cell-based programs of development and injury repair. In contrast to the results obtained with early fetal spheroids highlighted here, accentuation of Wnt signaling in cultures of adult intestinal epithelium can also create spheroids that are enriched in stem cells (Sato et al., 2011a; Miyoshi et al., 2012). Adult spheroids evoke similarities to epithelial stem cells undergoing robust expansion in vivo during intestinal wound healing and, though this must be confirmed, it appears that these stem cell-based responses to injury do not recapitulate programs of very early development. The studies of Mustata et al. and Fordham et al. will greatly assist in understanding these distinct processes, and could provide insight into cancer biology. Over 80% of human colon tumors lack functional adenomatous polyposis coli (APC) protein, which results in robust activation of Wnt signaling. However, a subpopulation of human colon tumors harbors R-spondin gene fusions instead of APC mutations (Seshagiri et al., 2012), and the growth of these tumors may depend on R-spondin rather than Wnts. It would be interesting to see if cells within these tumors have transitioned to the early developmental stage described in Mustata et al. and Fordham et al., and thus have reduced requirements for canonical Wnt signaling. Further deciphering why early prenatal epithelial cells do not require Wnt will be an important area of investigation and may provide further insight into cancer biology and intestinal repair.