Clone wars: From molecules to cell competition in intestinal stem cell homeostasis and disease

Abstract

The small intestine is among the fastest self-renewing tissues in adult mammals. This rapid turnover is fueled by the intestinal stem cells residing in the intestinal crypt. Wnt signaling plays a pivotal role in regulating intestinal stem cell renewal and differentiation, and the dysregulation of this pathway leads to cancer formation. Several studies demonstrate that intestinal stem cells follow neutral drift dynamics, as they divide symmetrically to generate other equipotent stem cells. Competition for niche space and extrinsic signals in the intestinal crypt is the governing mechanism that regulates stemness versus cell differentiation, but the underlying molecular mechanisms are still poorly understood, and it is not yet clear how this process changes during disease. In this review, we highlight the mechanisms that regulate stem cell homeostasis in the small intestine, focusing on Wnt signaling and its regulation by RNF43 and ZNRF3, key inhibitors of the Wnt pathway. Furthermore, we summarize the evidence supporting the current model of intestinal stem cell regulation, highlighting the principles of neutral drift at the basis of intestinal stem cell homeostasis. Finally, we discuss recent studies showing how cancer cells bypass this mechanism to gain a competitive advantage against neighboring normal cells.

Fig. 1. Organization of the intestinal villus and crypt.

Crypts of Lieberkühn surround intestinal villi (six or more crypts per villus) and replenish them with newly formed cells. Crypt base columnar (CBC) stem cells, located at the base of each crypt, represent the driving force that maintains intestine homeostatic renewal. CBC cells divide and originate transit-amplifying (TA) cells located in the upper zone of the crypt, which undergo multiple rounds of mitosis before differentiating into any of the cells of either secretory or absorptive lineages (right panel). CBC cells are intermingled with Paneth cells, specialized secretory cells that support CBC cells by providing growth factors. The balance between self-renewal and differentiation is regulated by morphogenetic signals, including BMP and Wnt. Differentiated cells move to the top of the villus, where they are shed into the intestinal lumen and die by anoikis at the end of their life cycle.

Fig. 2. Activation of the canonical Wnt signaling pathway.

Binding of Wnt ligands to cognate receptors Frizzled and Lrp5/6 leads to the formation of a multiprotein complex known as the signalosome. The signalosome recruits Dvl, Axin, Gsk3, and casein kinases to the plasma membrane, leading to the dissociation of the cytosolic destruction complex. In turn, this promotes β-catenin stabilization, which translocates into the nucleus. Here, together with TCF/LEF transcription factors, β-catenin induces Wnt target gene activation. Axin2, Rnf43 and Znrf3 are both target genes and inhibitors of the Wnt pathway, thus forming a negative feedback loop.

Fig. 3. Rnf43 and Znrf3 are transmembrane E3 ligases that promote Wnt receptor turnover.

Rnf43/Znrf3 (R/Z) promotes Frizzled and Lrp5/6 ubiquitination, assisted by the cytosolic adaptor Dvl. Once ubiquitinated, Wnt receptors are internalized by endocytosis and degraded via the lysosomal system. R/Z activity is regulated by the phosphorylation of the cytosolic tail by casein kinase 1 (CK1). The secreted Wnt agonist Rspo inhibits R/Z by forming a ternary complex with Lgr4/5/6 receptors, which in turn triggers the autoubiquitination of R/Z followed by the endocytosis of the complex. The deubiquitinase Usp42 stabilizes R/Z at the plasma membrane by removing ubiquitin from R/Z proteins, hence inhibiting Wnt signaling.

Fig. 4. Confetti reporter-based multicolor lineage tracing can be used to study neutral drift dynamics in stem cell populations.

Top: the R26R-confetti allele functions as a stochastic multicolor reporter, in which Cre-dependent recombination induces the mutually exclusive expression of one of four different fluorescent proteins (green, yellow, red, and cyan). The specific activation of confetti in intestinal stem cells allows us to follow clonal dynamics in the intestinal crypt, as individual ISCs and their direct progeny express the same unique fluorescent protein. Bottom: an intestinal crypt (visualized from a bottom view) containing ISCs labeled with different fluorescent proteins (red, yellow, and cyan) or unlabeled (light gray). Paneth cells (dark gray) are intercalated between the ISCs. Over time, some clones will expand (see the yellow clone in the example), while others will shrink or disappear completely (cyan and red clones, respectively) as a result of neutral competition.

Fig. 5. Red2Onco system allows us to study the effects of cancer cells on their surrounding environment.

Top: Red2Onco design is based on the original confetti allele. CRISPR/Cas9 genome editing was used to insert a 2 A peptide-Oncogene cassette between the red and cyan fluorescent proteins. This converts the original RFP into a bicistronic construct coexpressing the red fluorescent marker together with an oncogene of choice. Cre recombination occurs as usual, inducing the stochastic expression of one of the four fluorescent reporters. Cells expressing RFP will also express the selected oncogene, while the other colors label wild-type (WT) cells. Bottom: expression of an oncogene, such as Kras^G12D or PIK3CA^H1047R, confers a survival advantage to red clones, which will rapidly expand into their home crypt. Over time, this will lead to the fixation of the mutant clone into the crypt, which becomes monoclonal red. Due to the cancer cell influence, nearby crypts will also accelerate fixation and become monoclonal (visualized in cyan). Distal crypts are not affected by the red mutant clones. Mutant cells express secreted factors, including prodifferentiation signaling molecules (BMPs) and Wnt antagonists (Notum, Dkk, and Sfrp), which reduce the number of stem cells in proximal WT crypts, driving biased drift. Note that niche remodeling induced by cancer cells also affects nearby stromal cells.