The evolving roles of Wnt signaling in stem cell proliferation and differentiation, the development of human diseases, and therapeutic opportunities

Abstract

The evolutionarily conserved Wnt signaling pathway plays a central role in development and adult tissue homeostasis across species. Wnt proteins are secreted, lipid-modified signaling molecules that activate the canonical (β-catenin dependent) and non-canonical (β-catenin independent) Wnt signaling pathways. Cellular behaviors such as proliferation, differentiation, maturation, and proper body-axis specification are carried out by the canonical pathway, which is the best characterized of the known Wnt signaling paths. Wnt signaling has emerged as an important factor in stem cell biology and is known to affect the self-renewal of stem cells in various tissues. This includes but is not limited to embryonic, hematopoietic, mesenchymal, gut, neural, and epidermal stem cells. Wnt signaling has also been implicated in tumor cells that exhibit stem cell-like properties. Wnt signaling is crucial for bone formation and presents a potential target for the development of therapeutics for bone disorders. Not surprisingly, aberrant Wnt signaling is also associated with a wide variety of diseases, including cancer. Mutations of Wnt pathway members in cancer can lead to unchecked cell proliferation, epithelial-mesenchymal transition, and metastasis. Altogether, advances in the understanding of dysregulated Wnt signaling in disease have paved the way for the development of novel therapeutics that target components of the Wnt pathway. Beginning with a brief overview of the mechanisms of canonical and non-canonical Wnt, this review aims to summarize the current knowledge of Wnt signaling in stem cells, aberrations to the Wnt pathway associated with diseases, and novel therapeutics targeting the Wnt pathway in preclinical and clinical studies.

Keywords: Cancer; Canonical Wnt; Disease; Non-canonical Wnt; Stem cells; Targeted therapy; Wnt signaling; β-Catenin.

Figure 1

The simplified canonical Wnt/β-catenin signaling pathway. (A) In the absence of Wnt ligands, a destruction complex of adenomatous polyposis coli (APC) protein, Axin, serine/threonine kinase glycogen synthase kinase 3β (GSK-3β), and casein kinase 1 (CK1) phosphorylates β-catenin. Following phosphorylation, the E3-ubiquitin ligase β-TrCP ubiquitinates β-catenin, leading to its proteasomal degradation. The TCF/LEF transcription factor complex remains bound to the transducing-like enhancer protein (TLE/Groucho), and gene transcription does not proceed. (B) In the presence of Wnt ligands, Fz receptors and LRP5/6 coreceptors recruit the disheveled protein (Dsh/Dvl), phosphorylating the cytoplasmic tails of LRP5/6. LRP5/6 binds Axin, causing the destruction complex to disassemble and freeing β-catenin. β-Catenin is stabilized and accumulates in the cytoplasm before translocating to the nucleus to stimulate gene expression. Wherein, β-catenin binds to TCF/LEF, displacing TLE/Groucho, and target gene transcription proceeds. A number of extracellular regulators also play a role in modulating canonical Wnt signals. R-spondin ligands potentiate Wnt signaling through interaction with LGR4/5 and inhibition of Rnf43/Znrf3 (a transmembrane E3 ubiquitin ligase) mediated degradation of Fzd. Norrin is capable of binding to Fz receptor subtype-4 (Fz4) in order to activate the canonical pathway in an LRP5-dependent manner. Dkk family proteins and SOST bind to the LRP5/6 co-receptor to prevent the binding to Wnt. Wif-1, Cerberus, and members of the sFRP family sequester Wnts in the extracellular space to prevent the triggering of signaling.

Figure 2

GSK3β and CK1 phosphorylation sites on β-catenin. Prior to ubiquitination by β-TrCP, β-catenin is phosphorylated at 4 key residues. Destruction complex member CK1 first phosphorylates β-catenin at the Ser45 residue, enabling GSK3β to phosphorylate the Ser33, Ser37, and Thr41 residues. The result is the creation of a binding site for β-TrCP. Thereafter, phosphorylated Ser37 and Ser33 on β-catenin are recognized by the F-box protein β-TrCP for ubiquitination, leading to the subsequent proteasomal destruction of β-catenin.

Figure 3

The non-canonical Wnt/PCP pathway. Wnt signaling is transduced through Fzd receptors and co-receptors such as protein tyrosine kinase 7 (PTK7), tyrosine kinase-like orphan receptor (ROR1/ROR2), and tyrosine kinase related receptor (RYK). As a result, Dvl is phosphorylated, leading to Inversin (Invs) recruitment. Smad ubiquitination regulatory factor (Smurf) is then recruited by the phosphorylated Dvl to Par6. Smurf is responsible for the ubiquitination of Wnt/PCP antagonizing Prickle, targeting Prickle for proteasomal destruction. Dvl also associates with DAAM (Disheveled-associated activator of morphogenesis), which mediates actin polymerization through the activation of profilin protein. Dvl and DAAM, respectively, stimulate the activation of the GTPases RHOA and Rac1. RHOA, in turn, activates ROCK (Rho kinase). ROCK mediates actin polymerization through its activation of mitogen-activated protein kinase (MRLC). Moreover, Rac1 activates JNK (c-Jun N-terminal kinase) which phosphorylates c-Jun, resulting in c-Jun activation. Furthermore, c-Jun, a well-characterized protein of the activator protein-1 (AP-1) complex, moves to the nucleus in order to initiate target gene expression. JNK also phosphorylates and activates CapZ-interacting protein (CapZIP), which serves to remodel actin filament assembly. Altogether, Wnt/PCP effectors influence the cytoskeletal rearrangements needed for cell polarity and cell motility.

Figure 4

The non-canonical Wnt/Ca²⁺ signaling pathway. Wnt/Ca²⁺ signaling is primarily initiated by the binding of the Wnt5a ligand and Fzd2 receptor. Wnt/Fzd interaction, along with the Ror1/Ror2 co-receptor, leads to the co-stimulation of Dvl and heterotrimeric G-protein to activate phospholipase C (PLC). PLC cleaves phosphatidylinositol-4,5-bisphosphate (PtdInsP2) into diacylglycerol (DAG) and inositol-1,4,5-trisphosphate (InsP3/IP3). DAG activates protein kinase C (PKC) to stimulate CDC42, which serves to mediate cellular polarity during development. InsP3 binds to InsP3R receptors on the ER surface, initiating the release of Ca²⁺ ions, thereby increasing cytoplasmic levels of calcium. The calcium sensor stromal interaction molecule 1/2 (STIM1/2) detects decreases in lumen ER Ca²⁺ concentration before activation of the Orai family proteins to induce store-operated Ca²⁺ entry. Sarcoplasmic/ER Ca²⁺ ATPases (SERCAs) also pump Ca²⁺ back from the cytosol into the ER. Increased cytoplasmic Ca²⁺ concentration activates the phosphatase calcineurin as well as calcium calmodulin-mediated kinase II (CAMKII). Calcineurin activates the nuclear factor of activated T cells (NFAT), leading to target gene transcription. CAMKII activates TGFβ-activated protein kinase 1 (TAK1), which then activates Nemo-like kinase (NLK). NLK is responsible for the phosphorylation of TCF, thereby inhibiting the formation of the β-catenin/TCF complex and preventing canonical Wnt target gene transcription.

Figure 5

Wnt signaling in multipotent hematopoietic stem cells. Canonical Wnt signaling via Wnt3a has been shown to play a role in HSC self-renewal capability. Wnt3a deficiency led to a reduction in both the number of HSCs in fetal mouse liver and long-term repopulation capacity. Moreover, Wnt5a has been shown to induce HSC quiescence through the activation of the non-canonical pathway. Furthermore, in another study, Fmi and Fz8 knockout resulted in the decline of quiescent LT-HSC populations, indicating non-canonical Wnt signaling plays an important role in the maintenance of LT-HSCs. Studies have shown that Wnt/β-catenin signaling and β-catenin/TCF interaction are required for normal T-cell development. Hossain et al demonstrated the disruption of β-catenin and TCF interaction hinders T-cell survival due to increased susceptibility of thymocytes and activated T-cells to apoptosis. It has been reported that canonical and non-canonical Wnt ligands have differing effects on B-cell development. Canonical Wnt signaling, through Wnt3a, has been shown to inhibit B-cell lymphopoiesis and promote the retention of HSC markers. On the other hand, Wnt5a, which may oppose the canonical path or act via the non-canonical cascade, increased B-cell lymphopoiesis.

Figure 6

The influence of Wnt signaling on adipogenesis and osteogenesis in MSCs. Non-canonical Wnt signaling via Wnt5a was shown to be important to the chondrogenic differentiation of MSCs. Wnt5a enhanced cartilage formation, collagen fiber rearrangement, and glycosaminoglycan and collagen deposition in vivo. Wnt3a plays a dual role in modulating chondrogenesis. When acting via the canonical Wnt pathway, Wnt3a promotes MSC proliferation. By contrast, Wnt3a also inhibits MSC chondrogenesis via the CaMKII-mediated non-canonical Wnt pathway. In no specific order, the following Wnt pathway ligands have been shown to affect the osteoblastogenic differentiation of MSCs. Wnt3a and Wnt1 are capable of stimulating osteoblastogenesis through β-catenin activation. Wnt7a enhances the differentiation of MSCs into osteoblasts via enhancing TCF-1 binding to the promoter region of Runx2. Wnt11 increases the expression of Rspo2 as well as osteoblast-associated genes Dmp1 (dentin matrix protein 1), Phex (phosphate-regulating endopeptidase homolog), and Bsp (bone sialoprotein). BMP2 stimulates LRP5 expression and inhibits β-TrCP expression, leading to an increase in β-catenin levels in osteoblasts and the promotion of osteogenic differentiation. BMP2 also increases expression of the canonical Wnt ligands Wnt1, Wnt3a, and Wnt4 which function to increase transcription of osteogenic genes Id1, Dlx5, Msx2, Osx, and Runx2. Wnt10b was found to shift cell fate toward the osteoblast lineage by induction of the osteoblastogenic transcription factors Runx2, Dlx5, and osterix. Moreover, Wnt10b suppresses the adipogenic transcription factors peroxisome proliferator-activated receptor (PPARγ) and CCAAT/enhancer-binding protein a (C/EBPa). The non-canonical Wnt5a ligand inhibits PPARγ activation, thereby suppressing adipogenesis and promoting the osteogenic differentiation of MSCs. By contrast, Wnt inhibition has been shown to play a role in adipogenesis. The Wnt/β-catenin inhibitor sFRP1 is endogenously expressed by mature adipocytes in human adipose tissue and several sFRPs are associated with adipocyte dysfunction in obesity. Expressed by mature osteocytes but not by early osteocytes or osteoblasts, sclerostin (SOST) binds to the Wnt co-receptors LRP5/6, thereby antagonizing Wnt signaling with the effect of inhibiting bone formation and indirectly promoting bone resorption. BMP signaling may also restrict Wnt signaling through induction of SOST, as He et al found that BMP receptor type 1a (BMPr1a) induces the expression of SOST to limit cancellous bone accrual. Furthermore, a novel role for Axin2 in adipogenesis has also been described, as it was discovered that the adipogenic transcription factor PPARγ transcriptionally activates the destruction complex member Axin2, thereby impairing Wnt signaling via β-catenin degradation.

Figure 7

Wnt signaling in crypt intestinal stem cells. The Wnt target gene and transmembrane receptor Lgr5, is a stem cell marker for cycling crypt base columnar cells (CBCs). CBCs may either self-renew or give rise to specialized absorptive and secretory cell types. CBCs generate rapidly proliferating transit amplifying (TA) cells that move upwards from the crypt base before differentiating into specialized cell lineages. Usually located at the +4 position from the crypt base, label-retaining cells (LRCs) are slow cycling, quiescent stem cells. Active Wnt signaling stimulates crypt stem cell proliferation. Crypt stem cells receive Wnt3 signals from neighboring daughter Paneth cells. These signals play a role in Paneth cell induction and maturation, as well as Lgr5 stem cell growth efficiency. Interestingly, Farin et al demonstrated Wnt3 was dispersible for the maintenance of ISCs in vivo, suggesting the redundancy of Wnt signals from other sources in ISC homeostasis. As such, Foxl1-expressing mesenchymal cells have been identified as an essential source of Wnt2b, Wnt4, Wnt5a, and Wnt-activating Rspo3 within the intestinal stem cell niche. Aoki and colleagues demonstrated that diphtheria toxin-mediated loss of Foxl1 cells resulted in the shortening of the intestine as well as decreased epithelial cell proliferation, villi length, and crypt depth.

Figure 8

Aberrant activation of Wnt signaling in cancer. Wnt signaling plays numerous roles in normal development. The canonical pathway is vital for inducing cell proliferation, differentiation, and maturation, as well as proper body-axis specification. The non-canonical pathway is involved in cell polarization and migration. As such, the dysregulations of Wnt can lead to the development of diseases. Above are examples of aberrations to the WNT pathway that have been associated with cancer.