Stem Cell Reviews

, Volume 3, Issue 1, pp 30–38

Networking of WNT, FGF, Notch, BMP, and Hedgehog Signaling Pathways during Carcinogenesis

Authors

    • Genetics and Cell Biology SectionNational Cancer Center Research Institute
Article

DOI: 10.1007/s12015-007-0006-6

Cite this article as:
Katoh, M. Stem Cell Rev (2007) 3: 30. doi:10.1007/s12015-007-0006-6

Abstract

The biological functions of some orthologs within the human genome and model-animal genomes are evolutionarily conserved, but those of others are divergent due to protein evolution and promoter evolution. Because WNT signaling molecules play key roles during embryogenesis, tissue regeneration and carcinogenesis, the author’s group has carried out a human WNT-ome project for the comprehensive characterization of human genes encoding WNT signaling molecules. From 1996 to 2002, we cloned and characterized WNT2B/WNT13, WNT3, WNT3A, WNT5B, WNT6, WNT7B, WNT8A, WNT8B, WNT9A/WNT14, WNT9B/WNT14B, WNT10A, WNT10B, WNT11, FZD1, FZD2, FZD3, FZD4, FZD5, FZD6, FZD7, FZD8, FZD10, FRAT1, FRAT2, NKD1, NKD2, VANGL1, RHOU/ARHU, RHOV/ARHV, GIPC2, GIPC3, FBXW11/βTRCP2, SOX17, TCF7L1/TCF3, and established a cDNA-PCR system for snap-shot and dynamic analyses on the WNT-transcriptome. In 2003, we identified and characterized PRICKLE1, PRICKLE2, DACT1/DAPPER1, DACT2/DAPPER2, DAAM2, and BCL9L. After completion of the human WNT-ome project, we have been working on the stem cell signaling network. WNT signals are transduced to β-catenin, NLK, NFAT, PKC, JNK and RhoA signaling cascades. FGF20, JAG1 and DKK1 are target genes of the WNT-β-catenin signaling cascade. Cross-talk of WNT and FGF signaling pathways potentiates β-catenin and NFAT signaling cascades. BMP signals induce IHH upregulation in co-operation with RUNX. Hedgehog signals induce upregulation of SFRP1, JAG2 and FOXL1, and then FOXL1 induces BMP4 upregulation. The balance between WNT-FGF-Notch and BMP-Hedgehog signaling networks is important for the maintenance of homoestasis among stem and progenitor cells. Disruption of the stem cell signaling network results in pathological conditions, such as congenital diseases and cancer.

Keywords

WNTFGFNotchBMPHedgehogStem cell biologyComparative integromicsSystems medicine

Introduction

Biological functions of some orthologs within the human genome and model-animal genomes are evolutionarily conserved, but those of other orthologs are divergent mainly due to protein evolution and promoter evolution. Because WNT signaling molecules play key roles in embryogenesis and carcinogenesis [1, 2], comprehensive characterization of human genes encoding WNT signaling molecules is the starting platform to elucidate the mechanism of human diseases and also to develop new therapeutics for human diseases. The author’s group carried out the human WNT-ome project (http://esi-topics.com/fmf/2005/september05-MasaruKatoh.html).

Genome science, characterized by high-throughput technology, omics data, and bioinformatics, is the hub for various kinds of sciences, such as oncology, gastroenterology, developmental biology, regenerative medicine, and stem cell biology [3, 4]. Comparative genomics and network analyses are hot fields for the genome science in the post-genome era. After the completion of the human WNT-ome project, we have been working on the stem cell signaling network with the focus on WNT, FGF, Notch BMP, and Hedgehog signaling pathways.

Overview of WNT signaling cascades and accomplishments of the human WNT-ome project will be reviewed in the first part of this article, and then cross-talk of WNT, FGF, Notch BMP, and Hedgehog signaling pathways will be reviewed.

Overview of WNT Signaling Pathway

WNT family members, characterized by conserved Cys and other residues, are secreted-type glycoproteins implicated in organ development, tissue regeneration and tumor formation [1, 2]. Frizzled (FZD) family members are seven-transmembrane-type WNT receptors with extracellular Frizzled domain, cytoplasmic DVL-, and PZD-binding motifs. WNT signals are transduced through FZD family receptors to the canonical and non-canonical pathways in the context dependent manner [5, 6].

LRP5 and LRP6 with extracellular WNT-binding region and cytoplasmic AXIN-binding motif are WNT co-receptors activating the canonical pathway (Fig. 1a). In the absence of canonical WNT signals, β-catenin associated with AXIN and APC is phosphorylated by GSK3β and casein kinase Iα (CKIα) [6, 7]. Phosphorylated β-catenin is recognized by βTRCP1 and βTRCP2 for ubiquitination, and poly-ubiquitinated β-catenin is degraded in the proteasome complex [810]. In the presence of canonical WNT signals, Frizzled-Dishevelled complex is assembled with LRP5/6-AXIN-FRAT complex to release β-catenin from GSK3β [6, 7]. Unphosphorylated β-catenin is resistant to the ubiquitination-mediated degradation, and is accumulated in the nucleus. Nuclear β-catenin is associated with TCF/LEF family transcription factors (TCF1, LEF1, TCF3 and TCF4), Legless family docking proteins (BCL9 and BCL9L), and PYGO family co-activators (PYGO1 and PYGO2) for the transcriptional activation of target genes of the WNT-β-catenin signaling cascade, such as FGF20, JAG1, DKK1, MYC, CCND1, and AXIN2.
https://static-content.springer.com/image/art%3A10.1007%2Fs12015-007-0006-6/MediaObjects/12015_2007_6_Fig1_HTML.gif
Fig. 1

WNT signaling pathway and human WNT-ome project. a WNT signaling pathway. WNT signals are transduced through Frizzled (FZD) family receptors to the canonical and non-canonical pathways in the context dependent manner. Canonical WNT signals are transduced through FZD receptor and LRP5/LRP6 co-receptor to the β-catenin-TCF signaling cascade for the cell fate determination. Non-canonical WNT signals are transduced through FZD receptor and ROR2/RYK co-receptor to NLK, NFAT, PKC, JNK, and RhoA signaling cascades for the control of tissue polarity and cell movement. b Human WNT-ome project. Human WNT2B, WNT3, WNT3A, WNT5B, WNT6, WNT7B, WNT8A, WNT8B, WNT9A (WNT14), WNT9B (WNT14B), WNT10A, WNT10B, WNT11, FZD1, FZD2, FZD3, FZD4, FZD5, FZD6, FZD7, FZD8, FZD10, FRAT1, FRAT2, NKD1, NKD2, VANGL1, RHOU (ARHU), RHOV (ARHV), GIPC2, GIPC3, FBXW11 (βTRCP2), SOX17, and TCF7L1 (TCF3) were cloned characterized in our laboratory during 1996–2002

ROR2 and RYK are WNT co-receptors activating the non-canonical pathways (Fig. 1a). Non-canonical signals are transduced to Dishevelled for the activation of JNK and RhoA signaling cascades, and also to PLC for the catalysis of PIP2 to DAG and IP3 [6]. DAG activates PKC, while IP3 activates Ca2+-Calmodulin signaling for the following activation of CAMK-MAK3K7-NLK and Calcineurin-NFAT signaling cascades [11].

Human WNT-ome Project

The author entered the WNT field in 1995 through the isolation of a novel member of WNT gene family. The novel gene was initially designated WNT-13, because it was the 13th member of the WNT gene family [12]. WNT-13 was then renamed WNT2B due to the fact that WNT-13 was the paralog of WNT2 proto-oncogene [13]. WNT2B-ST7L-CAPZA1 locus at human chromosome 1p13 and WNT2-ST7-CAPZA2 locus at human chromosome 7q31 are paralogous regions within the human genome [14, 15]. WNT2B splicing variant 1 (WNT2B1) consists of exons 1, 2 and 4∼7 of WNT2B gene, while WNT2B splicing variant 2 consists of exons 3∼7 [13]. Function of WNT signaling molecules were previously investigated in the model-animal experimental system, such as Xenopus axis duplication assay and knock-out mice, to report phenotypic change associated with a single WNT signaling molecule. Using the Xenopus axis duplication assay, We demonstrated that WNT2B2 rather than WNT2B1 is the canonical WNT activating the β-catenin signaling cascade [16].

From 1996 to 2002, we engaged in wet biology to clone and characterize the following human genes encoding WNT signaling molecules (Fig. 1b): WNT2B [12, 13], WNT3 [17, 18], WNT3A [18, 19], WNT5B [20, 21], WNT6 [22], WNT7B [23, 24], WNT8A [25, 26], WNT8B [26, 27], WNT9A (WNT14) [19, 28], WNT9B (WNT14B) [28, 29], WNT10A [22, 30, 31], WNT10B [32, 33], WNT11 [34], FZD1 [35], FZD2 [35], FZD3 [36], FZD4 [37, 38], FZD5 [39], FZD6 [40], FZD7 [35, 41] FZD8 [42], FZD10 [4346], FRAT1 [47, 48], FRAT2 [48, 49], NKD1 [50], NKD2 [50], VANGL1 [51], RHOU (ARHU) [52], RHOV (ARHV) [53], GIPC2 [54, 55], GIPC3 [56], FBXW11 (βTRCP2) [810, 57], SOX17 [58], and TCF7L1 (TCF3) [59]. In 2003, we engaged in dry biology of WNT signaling molecules to identify and characterize PRICKLE1 [60], PRICKLE2 [60], DACT1 (DAPPER1) [61], DACT2 (DAPPER2) [61], DAAM2 [62], and BCL9L [63].

We established the WNT-transcriptome cDNA-PCR system in 2002 [2]. The expression profile of WNT signaling molecules was measured in a variety of tumors, such as gastric cancer, pancreatic cancer, esophageal cancer, and breast cancer. Upregulation of WNT2 was found in all of ten cases of primary gastric cancer using cDNA-PCR, and in four of other eight cases using matched tumor/normal expression array analysis [64]. We also reported WNT5A upregulation in seven of ten cases of primary gastric cancer using cDNA-PCR, and in five of other eight cases using matched tumor/normal expression array analysis [65]. WNT2 and WNT5A are frequently upregulated in primary gastric cancer (Fig. 2a).
https://static-content.springer.com/image/art%3A10.1007%2Fs12015-007-0006-6/MediaObjects/12015_2007_6_Fig2_HTML.gif
Fig. 2

Carcinogenesis in the stomach. a Multi-stage carcinogenesis in the stomach. Genetic predisposition, chronic Helicobacter pylori infection, salt over-uptake, decreased vegetable/fruit consumption, smoking, metabolic syndrome, and aging are risk factors of human gastric cancer. WNT2 and WNT5A are frequently upregulated in primary gastric cancer. Results of cDNA-PCR are reproduction of figures in [64, 65]. bHelicobacter pylori infection and WNT. Helicobacter pylori activates NF-κB and AP-1 signaling cascades through lipopolysaccharide (LPS) recognition by Toll-like receptor (TLR), and also NF-κB and ERK signaling cascades through injection of peptidoglycan (PGN) and CagA, respectively. IL-1β, IL-6, IL-8, TNF-α, and IFN-γ are elevated in the stomach with Helicobacterpylori infection. IL-6 and TNF-α induce upregulation of WNT5A and WNT10B, respectively

Helicobacter pylori infection is one of risk factors of human gastric cancer [6669]. H. pylori is a Gram-negative spiral-shaped bacterium colonized in the human stomach. Lipopolysaccharide (LPS) derived from the cell wall of H. pylori is recognized by the cell-surface pathogen-recognition receptor TLR2 and TLR4 to activate TRAF6-MAP3K7-NF-kB and TRAF6-MAP3K7-AP-1 signaling cascades (Fig. 2b). In addition, H. pylori injects peptidoglycan (PGN) and CagA protein into human gastric epithelial cells using the type IV secretion system to activate PGN-NOD1-mediated NF-κB signaling cascade and CagA-SHP2-GRB2-mediated RAS-ERK signaling cascade (Fig. 2b). H.pylori induces the elevation of IL-6 and TNF-α in gastric mucosa. IL-6 and TNF-α then induce upregulation of WNT5A and WNT10B, respectively, to activate the WNT signaling pathway during chronic H.pylori infection (Fig. 2b).

Human NT2 cells, derived from an embryonal tumor, differentiate into neuron-like cells after retinoic acid treatment. Analyses on the WNT transcriptome during the early phase of retinoic acid induction in NT2 cells revealed upregulation of WNT2, WNT9B, and downregulation of WNT3A, WNT8A, WNT8B, WNT10B, and WNT11 (Fig. 3). Because the dynamic change of WNT transcriptome occurs during the phenotypical change in NT2 cells, we proposed the threshold model of WNT signaling in which the sum of the dynamic change in the WNT transcriptome rather than a change of a single WNT family member is important for the cell fate determination [2].
https://static-content.springer.com/image/art%3A10.1007%2Fs12015-007-0006-6/MediaObjects/12015_2007_6_Fig3_HTML.gif
Fig. 3

Dynamic change of the WNT transcriptome during the early phase of neuronal differentiation of NT2 cells induced by retinoic acid. Human NT2 cells are derived from embryonal tumor. NT2 cells differentiate into neuron-like cells after retinoic acid treatment. WNT-transcriptome analyses during the early phase of retinoic acid induction in NT2 cells revealed upregulation of WNT2, WNT9B, and downregulation of WNT3A, WNT8A, WNT8B, WNT10B, and WNT11. Results of cDNA-PCR are reproduction of figures in [2]

The WNT-transcriptome cDNA-PCR system has been applied to the snap-shot analyses on tumors as well as to the dynamic analyses after the treatments with TNF-α, IFN-γ, estrogen, and retinoic acid to investigate the roles of human WNT signaling molecules.

Stem Cell Signaling Network

Canonical WNT signals activate the β-catenin signaling cascade for FGF20 and JAG1 upregulation (Fig. 4a). FGF binding to FGF receptors (FGFRs) induces autophosphorylation of FGFRs to recruit FRS2 and PLCγ. FGF signals are transduced through FRS2 to the RAS-ERK and PI3K-AKT signaling cascades, and through PLCγ to Ca2+-mediated PKC and NFAT signaling cascades [7072]. AKT phosphorylates GSK3β to inhibit its negative regulation on Snail, which is a repressor for CDH1 gene encoding E-cadherin. FGF-mediated E-cadherin downregulation leads to epithelial-to-mesenchymal transition (EMT), and potentiation of the β-catenin signaling cascade. In addition, WNT and FGF signals directly cross-talk through PLC to activate the Ca2+-mediated PKC and NFAT signaling cascades. On the other hand, JAG1-bindig to Notch family receptors activates CSL-NICD-MAML and NF-kB signaling cascades [73, 74]. Feedback loops from FGF and Notch signaling pathways to WNT signaling pathway remain unclear; however, WNT-FGF-Notch signaling network is the core component of the stem cell signaling network.
https://static-content.springer.com/image/art%3A10.1007%2Fs12015-007-0006-6/MediaObjects/12015_2007_6_Fig4_HTML.gif
Fig. 4

Stem cell signaling network. a Cross-talk of WNT and FGF signaing pathways. b Cross-talk of BMP and Hedgehog signaling pathways. c Simplified view of the stem cell signaling network, consisting of WNT, FGF, Notch, BMP and Hedgehog signaling pathways. JAG1 and JAG2 are ligands for Notch family receptors

BMP signals are transduced through type I receptor BMPR1A/BMPR1B/ACVR1 and type II receptor BMPR2 to phosphorylate R-SMAD proteins, such as SMAD1, SMAD5 and SMAD8, which is then complexed with SMAD4 [75]. The SMAD complex and RUNX transcription factor induce upregulation of Indian Hedgehog (IHH) [76]. Hedgehog signals are transduced through Patched (PTCH) family receptors to release Smoothened (SMO) signal transducer from functional inhibition. SMO then activates STK36 serine/threonine kinase for the stabilization and nuclear translocation of GLI family members (Fig. 4b). Hedgehog signals induce upregulation of SFRP1, JAG2 and FOXL1 [7678]. SFRP1 inhibits WNT signaling, JAG2 activates Notch signaling, and FOXL1 induces BMP4 upregulation [79]. BMP-Hedgehog signaling network counteracts with the WNT signaling pathway.

The balance between WNT-FGF-Notch and BMP-Hedgehog signaling networks is important for the maintenance of homoestasis among stem and progenitor cells (Fig. 4c). Disruption of the stem cell signaling network results in pathological conditions, such as congenital diseases and cancer [80].

Perspectives

Personalized medicine is the goal of medical sciences in the post-genome era. Because whole genome SNP screening and whole transcriptome analyses are expensive, SNP assay kits or transcriptome analyses kits detecting genes or mRNAs of WNT, FGF, Notch, BMP and Hedgehog signaling molecules could reduce the cost. Biomarker analyses on the stem cell signaling network will be utilized for genetic screening, prognosis prediction and therapeutic optimization of human cancer, such as gastric cancer and pancreatic cancer.

Acknowledgements

The author thanks to Drs. Norihiko Sagara, Jun Koike, Hiroyuki Kirikoshi and Tetsuroh Saitoh for their hard work in the Genetics and Cell Biology Section, to Ms. Yuriko Katoh and Masuko Kotoh for their bioinformatics works, to Drs. Momoki Hirai, Koichiro Shiokawa, Herbert Steinbeisser, Serge Fuchs, Scott Heller, and Jesús Chimal-Monroy for their collaboration on WNT signaling molecules, and to Dr. Takashi Sugimura for his warm encouragement. This study was supported in part by a Grant-in-Aid for Scientific Research on Priority Areas from the Ministry of Education, Culture, Sports, Science and Technology of Japan, and by a Grant-in-Aid for Cancer Research from the Foundation for Promotion of Cancer Research.

Copyright information

© Humana Press Inc. 2007