The nature of intestinal stem cells' nurture

EMBO reports 12, 6, 558–564. doi:; DOI: 10.1038/embor.2011.52

The ‘nature versus nurture’ debate concerns the relative contributions to an individual's identity of its nature (that is, its genetic make-up) compared with its nurture, defined as the totality of external, environmental factors. A similar type of debate is ongoing among developmental and stem-cell biologists: is the intrinsic nature (that is, its (epi)genetic make-up) of a stem cell what makes it self-renew and differentiate according to the physiological needs of a given tissue, or is it the immediate environment (nurture) that regulates stemness? Irrespective of the relative weight of each contribution, there is little doubt that both cell-autonomous and environmental factors play crucial roles in the maintenance of homeostasis in self-renewing tissues such as the skin, mammary gland, blood and intestine. In an article published last month in EMBO reports (Mustata et al, 2011), the Lgr4 gene is shown to have a rate-limiting role in establishing the stem-cell niche of the proximal intestinal tract.

…the Lgr4 gene is shown to have a rate-limiting role in establishing the stem-cell niche of the proximal intestinal tract

The epithelial lining of the proximal intestine is characterized by a unique tissue architecture consisting of villi and crypts. The intestinal crypt of Lieberkühn is a highly dynamic niche with stem cells in its lower third, which give rise to a population of fast-cycling transit-amplifying cells. Transit-amplifying cells undergo a limited number of cell divisions and eventually differentiate into four specialized cell types of the small intestine: absorptive, enteroendocrine, goblet and Paneth cells. Notably, Paneth cells are the only terminally differentiated cell type of the proximal intestinal tract that (i) move downwards along the crypt–villus axis and (ii) retain canonical Wnt signalling activity upon differentiation (van Es et al, 2005).

On the basis of clonal analysis and knock-in experiments, it was shown that the crypt base columnar (CBC) cells—located in the lower third of the crypt and characterized by Lgr5 expression—represent actively cycling stem cells that are able to give rise to all differentiated cell types of the intestinal epithelium (Barker et al, 2007). More recently, it has also been shown that Paneth cells, apart from their well-known bactericidal function, are in close physical association with Lgr5⁺ stem cells, to which they provide essential niche signals such as EGF, Wnt3a and Dll4 (Sato et al, 2011). This is also important in the light of the observation that single Lgr5⁺ stem cells, when cultured ex vivo, can generate crypt–villus organoids without a (mesenchymal) niche (Sato et al, 2009). In fact, the latter is only partly true, as these organoids are cultured in matrigel and in the presence of specific growth factors that are probably released by the niche in vivo.

Lgr5, together with Lgr4 and Lgr6, belongs to the family of leucine-rich repeat-containing G-protein-coupled seven-transmembrane receptors. Recently, both Lgr5 and Lgr6 have received attention from the stem-cell community: Lgr5 is a downstream Wnt target gene and a marker of cycling stem cells in the intestinal tract and the hair follicle, whereas Lgr6 expression marks adult stem cells in the skin (Barker & Clevers, 2010). However, whether they merely represent stem-cell markers or also have a functional role in stemness is unknown.

Mustata et al (2011) report on the functional role of another member of the Lgr family, Lgr4, by studying the effects of a targeted loss-of-function mutation (Lgr4 KO) on the development and differentiation of the mouse small intestine both in vivo and ex vivo. Endogenous Lgr4 expression is detected in transit-amplifying cells above the Paneth-cell zone, in CBC cells, and in rare Paneth cells. Loss of Lgr4 function results in a reduction in crypt depth due to a 50% decrease in epithelial-cell proliferation and, surprisingly, in an 80% reduction in Paneth-cell differentiation. Strikingly, these phenotypic features are apparently antagonistic to those of Lgr5 KO mice, in which premature Paneth-cell development was observed (Garcia et al, 2009). Accordingly, loss of Lgr4 function partly rescues the perinatal lethality of Lgr5 KO mice indicating non-redundancy of their individual functions.

Loss of Lgr4 function results in […] an 80% reduction in Paneth-cell differentiation

To further investigate the role of Lgr4 in crypt development, the ex vivo ‘minigut’ culture system (Sato et al, 2009) was used; in contrast to crypts from wild-type mice that give rise to self-renewing structures encompassing all the differentiated cell lineages of the adult gut, organoids derived from age-matched Lgr4 KO animals are initially present as hollow spheres, mainly composed of stem and transit-amplifying cells, which disaggregate within 2–3 days and die within a week in culture. In agreement with their apparently opposite and non-redundant functions, crypt cultures from Lgr5 KO mice survive long-term culture and develop into differentiated organoids comparable with those of normal mice. Whereas loss of Lgr4 function partly rescues the lethality of Lgr5 KO mice in vivo, this is not true ex vivo; compound homozygous Lgr4/5 KO crypts give rise to hollow spheres that collapse and die as observed in Lgr4 KO organoids. Hence, under these experimental conditions—that is, in the absence of a mesenchymal niche—the Lgr4 defect is dominant over the Lgr5 one.

Analysis of Paneth-cell differentiation markers and of Wnt targets, including Lgr5, confirmed their downregulation in Lgr4 KO organoids, thus suggesting a role for Lgr4 in Wnt signalling. Notably, lithium chloride treatment partly rescues the ex vivo phenotype of Lgr4 KO crypts, although this is not the case for other Wnt-signalling agonists, such as Wnt3a and Gsk3β inhibitors. On the basis of these observations, the authors conclude that Lgr4 probably has a permissive, rather than a direct and active role in Wnt signalling.

In view of this and other studies, a revisitation of the cell-autonomous and niche-independent features of the Lgr5⁺ cycling stem cell (CBC cells) in the intestinal crypt seems to be necessary (Fig 1). First, the capacity of CBC cells to recapitulate ex vivo the complexity of the crypt–villus unit is mostly dependent on Paneth cells (Sato et al, 2011). When they are sorted as single cells, CBC cells perform poorly in organoid formation, whereas doublets of CBC and Paneth cells show high clonogenicity (Sato et al, 2009, 2011). However, rather than occurring exclusively through the secretion of niche signals in the form of Wnt ligands, the nature of the interdependency between Paneth cells and CBC cells seems to involve additional mechanisms. As shown by Mustata et al, loss of Lgr4 function causes a Paneth-cell differentiation blockade in the presence of wild-type levels of Wnt3a and Wnt11, a defect that can be rescued by lithium chloride, but not by the Wnt3a ligand or Gsk3β inhibitors. This indicates that additional factors secreted by epithelial and possibly mesenchymal cells—for example, stromal myofibroblasts (Vermeulen et al, 2010)—and the physical association of Paneth with Lgr5⁺ cells underlies their ‘partnership’ in preserving homeostasis within such a highly dynamic tissue. Hence, Paneth cells apparently constitute an essential component of the stem-cell niche in the upper intestinal tract.

…rather than occurring exclusively through the secretion of niche signals […] the nature of the interdependency between Paneth cells and CBC cells seems to involve additional mechanisms

Figure 1.

Schematic illustration of the intestinal stem-cell compartment in the upper intestinal tract: Lgr4 (expressed in CBC and TA cells) positively stimulates Paneth-cell differentiation and, indirectly, stem-cell homeostasis, while Lgr5 (expressed in CBC cells) has been reported to inhibit Paneth-cell differentiation (Garcia et al, 2009). CBC, crypt base columnar; Dll4, delta-like 4; EGF, epidermal growth factor; TA, transit amplifying.

As it is always the case, good science leads to new questions. Which cell type provides this niche function in the colon where Paneth cells are not present? Of note, it has been shown that in the colon Lgr5⁺ cells are intermingled with yet uncharacterized CD24⁺ cells (Sato et al, 2011), a cell-surface antigen known to enrich for Paneth cells in the upper intestinal tract. As CD24 expression does not mark CBC cells, but rather their flanking cells, these observations could again reflect the supportive, niche role of Paneth cells and CD24⁺ cells in the upper and distal intestinal tract, respectively. This might also be true for colon cancer, where Paneth cells are often present, possibly to provide niche support for cancer stem cells. Alternatively, premature (in the colon) and/or fully differentiated (in the upper intestine) Paneth cells might have a dual function by providing physical and paracrine support for cycling stem cells in homeostasis, as well as representing the hitherto elusive quiescent stem cells that underlie tissue regeneration after tissue insults. Whatever the truth, the intestinal scene is now set to further dissect the complexity of the nature–nurture interaction between intestinal (cancer) stem cells and their niche.