Frizzled homolog proteins, microRNAs and Wnt Signaling in Cancer

Abstract

Wnt signaling pathways play important roles in tumorigenesis and are initiated by binding of Wnt to various receptors including frizzleds (FZDs). FZDs are one of several families of receptors comprised of FZD/LRP/ROR2/RYK in the Wnt signaling pathway. Expression of some FZD receptors are up-regulated, thereby activating the Wnt signaling pathway and is correlated with cancer malignancy and patient outcomes (recurrence and survival) in many cancers. The FZD family contains ten genes in humans and their function has not been completely examined including the regulatory mechanisms of FZD genes in cancer. Knockdown of FZDs may suppress the Wnt signaling pathway resulting in decreased cell growth, invasion, motility and metastasis of cancer cells.

Recently a number of microRNAs (miRNAs) have been identified and reported to be important in several cancers. MiRNAs regulate target gene expression at both the transcription and translation levels. The study of miRNA is a newly emerging field and promises to be helpful in understanding the pathogenesis of FZDs in cancer. Also miRNAs may be useful in regulating FZDs in cancer cells.

Therefore the aim of this review is to discuss current knowledge of the functional mechanisms of FZDs in cancer, including regulation by miRNAs and the potential for possible use of miRNAs and FZDs in future clinical applications.

Introduction

Frizzled homolog protein (FZD) is a seven-pass transmembrane type receptor and 10 members have been identified (FZD1–FZD10) in humans (1) (Figure 1). Wnt is a ligand of FZDs and consists of nineteen family genes in humans. The Wnt signaling pathway is initiated by the binding of Wnt ligands to the complex compromised of FZD and low-density lipoprotein receptor–related proteins 5/6 (LRP5/6)/ROR2/RYK, resulting in regulation of diverse cellular functions (2–5). The Wnt signaling pathway is comprised of canonical and noncanonical signals. In the canonical Wnt signaling pathway, Wnts bind to a complex of FZD and LRP5/6. The resultant signals prevent β-catenin phosphorylation by a multiprotein complex composed of adenomatous polyposis coli (APC), glycogen synthase kinase 3β, casin kinase 1, and axins, causing its proteosomal degradation. The beta-catenin associates with T cell factor (TCF)/lymphocyte enhancer transcription factors to activate target genes involved in cell survival, proliferation, or invasion (Figure 1B). The non-canonical Wnt signaling pathway consists of the Wnt/Ca²⁺ pathway and Wnt/c-Jun N-terminal kinase (JNK) (planar cell polarity) pathway (6). In the non-canonical Wnt signaling pathway, Wnts bind to a complex comprised of FZD and ROR2/RYK while the Wnt/Ca²⁺ pathway, Wnt activates intracellular Ca²⁺ signaling, as well as Ca²⁺-dependent protein kinases, such as protein kinase C (PKC) and calmodulin dependent protein kinase II (7) (Figure 1C). In the Wnt/JNK pathway, receptor stimulation activates Dishevelled (Dvl), which in turn activates the Rho family of GTPases such as RhoA and Rac. RhoA stimulates c-Jun expression through phosphorylation of c-Jun by Rho associated kinase (ROCK) (8–10) (Figure 1D). Therefore FZD plays a crucial role in both canonical and non-canonical pathways and the expression of FZD has been reported to be up-regulated in some cancer tissues.

Figure 1. Schematics of Wnt signaling pathway in cancer cells.

A. Regulation of Wnt signaling pathway in normal cells. Beta-catenin is usually phosphorylated and degradated by a multiprotein complex composed of APC, GSK3, CK1 and AXIN.

B. Wnt/beta-catenin pathway. Wnt ligands bind to the complex compromised of FZD/LRP/ROR/RYK, resulting in the constitutive stabilization of beta-catenin. The beta-catenin associates with TCF to activate target genes.

C. Wnt/Ca²⁺ pathway (non-canonical pathway). Wnt activates intracellular Ca²⁺ signaling as well as Ca²⁺-dependent protein kinases such as PKC and CaMK II. TAK1 and NLK can interfere with TCF/beta-catenin signaling.

D. Wnt/JNK pathway (non-canonical pathway). Receptor stimulation activates DVL, which in turn activates the Rho family of GTPases. Rho stimulates c-Jun expression through phosphorylation of c-Jun by ROCK.

Non-coding RNAs are very numerous and potentially important in gene and protein regulation. Currently microRNAs are well known as examples of non-coding RNAs (11, 12) and over 1500 human miRNAs have been identified based on miRBase [http://www.mirbase.org/]. miRNAs bind to the 3’UTR of target gene mRNA and repress translation or induce mRNA cleavage, thereby inhibiting translation from mRNA to protein (13). Thus miRNAs regulate target genes including FZDs, resulting in inhibition of diverse Wnt signaling pathways. Aberrant expression of miRNAs has been reported in many types of cancers and can function as tumor suppressor genes or oncogenes (14–16). Decreased expression of tumor suppressor miRNAs result in increased expression of target oncogenes. In contrast, increased expression of oncogenic microRNAs leads to loss or decreased expression of target tumor suppressor genes. So far, there have been few reports regarding the mechanism of FZD gene expression in cancer. Therefore identifying miRNAs regulating FZD expression in cancer tissues will be helpful to understand the mechanisms of FZD gene expression in cancer. Based on previous literature, five miRNAs (miR-204, miR-31, miR-493, miR-194, miR-23b) have been reported to regulate five FZD genes.

In this review, we focus on the function of FZDs (FZD1–FZD10) in cancer and discuss current regulatory mechanisms including miRNAs in the context of understanding their potential roles in tumorigenesis (Table 1).

Table 1.

FZDs and Wnt signaling pathway in Cancer

FZD	chromosome	Wnt signaling pathway		Over-expressed in Cancer/reference		microRNA/function in cancer/reference
FZD1	7q21	Canonical		colon, ovarian, breast, neuroblastoma	19, 20, 21, 22	miR-204/suppressor	29, 30, 31
FZD2	17q21.1	Canonical	Non-canonical (Ca2+)	Wilms’ tumour, melanoma, lung	33, 34, 35, 36
FZD3	8p21	Canonical	Non-canonical (PKA)	lung, leukemia, myeloma, lymphoma, sarcoma	36, 41, 42	miR-31/suppressor	48
FZD4	11q14-q21	Canonical		cervical uterus, leukemia colon, melanoma, pancreatic	49	miR-493/suppressor	55
FZD5	2q33.3-q34	Canonical		kidney, prostate	57, 58
FZD6	8q22.3-q23.1		Non-canonical (Ca2+)	squamous cell carcinomas	64	miR-194/unknown	70
FZD7	2q.33	Canonical	Non-canonical (JNK)	esophageal, gastric, nasopharyngeal, adenoid cystic, hepatocellular, colon, Wilms’ tumour	71, 72, 73, 74, 75, 76, 77, 78	miR-23b/suppressor	88, 89, 90
FZD8	10p11.2	Canonical		cervical uterus, kidney, leukaemia, lung	41, 57, 91, 92
FZD9	7q11.23		Non-canonical (ERK)	glioblastoma, astrocytoma	96
FZD10	12q24.33	Canonical	Non-canonical (JNK)	colon, lung, sarcoma	104, 105, 106, 107

FZD1 (Frizzled-1) and miRNA

Human FZD1 was first cloned and mapped to chromosome 7q21 by Sagara et al (17). Human FZD1 interacts with Wnts1-3 and Wnt3a increasing Wnt/beta-catenin signaling and resulting in stimulation of diverse tumorigenic processes. (18). FZD1 expression has been reported to be up-regulated in several cancers including colon, ovarian, breast neuroblastoma (19–22). In one study, Wang et al used thiazolidinediones, a novel cancer drug for breast cancer, and found that the drug decreased mRNA expression of both FZD1 and LRP6, inhibited beta-catenin mediated transactivation, and resulted in the inhibition of cell growth in breast cancer cell lines (23).

Two doxorubicin resistant neuroblastoma cancer cell lines had amplification of the chromosome 7q21 region (24) and overexpression of FZD1 and MDR1 (multidrug resistant gene) was found (22) in these doxorubicin resistant neuroblastoma cell lines. In this study, FZD1 silencing dramatically reduced MDR1 expression (22). Since this report showed the association of FZD1 and MDR1, these results may be helpful in understanding the function of FZD1 in chemo-resistant cancer cells.

In contrast, one report showed LOH in the same region (chromosome 7q21) in follicular thyroid carcinoma and FZD1 expression was downregulated in these tumors. Cell growth and invasion ability were also decreased in follicular thyroid carcinoma cell lines (25). Therefore depending on FZD1 expression levels in cancer tissues or condition, the function of FZD1 may vary.

To date, one paper has documented the association of FZD1 with miRNA-204. MiR-204 is one of several down-regulated miRNAs in senescent human trabecular meshwork cells (26, 27). Over-expression of miR-204 decreased FZD1 mRNA expression in two primary human trabecular meshwork cell lines and there was lower luciferase activity using plasmid containg the FZD1 3’-UTR sequence in HEK 293 cells (28), suggesting that miR-204 regulates FZD1 expression directly.

Regarding miR-204 function, several reports have shown it to be a tumor suppressor gene in head and neck, endometrial and renal cancers (29–31). Additional research will be required to elucidate the regulation of FZD1 by miR-204.

FZD2

Human FZD2 has been mapped to chromosome 17q21.1 (32). Since FZD2 mRNA is expressed in most human adult and fetal tissues, FZD2 expression is up-regulated in several cancers including primary Wilms’ tumour, melanoma and lung squamous cell carcinoma. (33–36). Wnt5a binds to FZD2 and activates the WNT/Ca2⁺ signaling pathway in melanoma cell lines (35). In the presence of Wnt3a, FZD2 also activates Wnt/beta-catenin signaling in pulmonary carcinoma (37). These reports suggest that FZD2 in the presence of Wnts may activate both canonical and non-canonical Wnt signaling pathways in cancer. So far no studies have reported the association of miRNAs with FZD2 regulation.

FZD3 and miRNA

Human FZD3 was mapped to chromosome 8p21 by Kirikoshi et al (38) and Sala et al (39). Kirikoshi et al reported that FZD3 mRNA is expressed in normal tissues (skeletal muscle, kidney, pancreas, cerebellum and cerebral cortex) and cancer cell lines and Sala et al also reported that FZD3 mRNA was expressed in most normal tissues. Although FZD3 mRNA was down-regulated during progression of ovarian carcinoma (40), FZD3 expression was up-regulated in several cancers (lung squamous cell carcinoma tissues, primary acute and lymphoblastic leukemia, myeloma, lymphoma, Ewing sarcoma) (36, 41, 42).

In one report, expression of sFRP1 (secreted Frizzled-Related Protein 1), a Wnt antagonist, was significantly decreased by DNA methylation in acute leukemia and interestingly over-expression of FZD3 mRNA correlated with hypermethylation of the sFRP1 (secreted Frizzled-Related Protein 1) (43). This result suggested that activation of aberrant Wnt signalling may be caused by the cooperation of SFRP1 down-regulation and FZD3 over-expression (43).

High FZD3 expression levels correlated with Wnt target gene, c-Myc and Cyclin D1 in sporadic adenoma, familial adenomatous polyposis and chronic lymphocytic leukemia (44, 45). FZD3 activated several Wnt/beta-catenin signaling pathways in the presence of Wnt3 and LRP6 compared to Wnt3 only in chronic lymphocytic leukemia (45). FZD3 regulated Wnt-3a-dependent neurite outgrowth in Ewing sarcoma (46), activated Galpha(s)/cAMP/PKA signaling pathway in the presence of Wnt5a and inhibited cell migration in breast cancer cell lines (47). As these reports show that FZD3 activates or inhibits cancer cells in the presence Wnt ligands and LRP, if FZD3 were to be used in cancer as a biomarker, both FZD3 and Wnt ligand expression should be analyzed. Thus FZD3 function as oncogene and has potential as a therapeutic target gene. With regards to miRNA-FZD3, one report has been published about breast cancer. In this study, Valastyan et al focused on miR-31 in breast cancers since it was expressed in primary normal human mammary epithelial cells and non-metastatic breast tumor cells, but not expressed in metastatic breast cancer cell lines. Cell invasion/migration and lung metastasis were inhibited in miR-31 transfected MDA-MB-231 cells compared to controls. FZD3 protein level and luciferase activity using a plasmid vector with the FZD3 3’-UTR sequence in miR-31 transfected MDA-MB-231 cells was decreased compared to control transfectants. Knockdown of FZD3 decreased MDA-MB-231 cell invasion (48). So far this study is the only one showing miR-31 as a novel microRNA targeting FZD3.

FZD4 and miRNA

Human FZD4 was mapped to chromosome 11q14-q21 (49) and was reported to be expressed in most normal human tissues. Sagara et al also reported that FZD4S, a splicing variant of the FZD4 gene (50), was expressed in adult heart, lung, fetal kidney and lung using RNA dot blot analysis. FZD4S inhibited and activated Wnt/beta-catenin signaling in the presence of Wnts in Xenopus (51). FZD4 activated the Wnt/beta-catenin signaling pathway and is related to epithelial to mesenchymal transition marker, E-cadherin and Snail1 expression in VaP cells (prostate cancer cell line) with TMPRSS2-ERG gene fusion and U87R4 cell lines being highly invasive (52, 53). Primers and antibodies able to differentiate between FZD4 and FZD4S should be used in the analysis of FZD4 expression because FZD4 and FZD4S may function differently in cancer. High methylation at the FZD4 loci was associated with progression-free survival in epithelial ovarian cancer (54). Though no studies have reported about FZD4 expression in cancer and normal tissues, methylation at the FZD4 locus may be a good cancer marker. One report has shown possible regulation of FZD4 by miR-493. MiR-493 was down regulated in bladder cancer tissues and bladder cancer cell lines (J82, T24, TCC-SUP) compared to normal tissues and cell lines (SV-HUC-1). Over-expression of miR-493 decreased T24 and J82 cell migration and motility and FZD4 protein levels and luciferase activity in miR-493 transfected T24 cells was decreased compared to controls, indicating that FZD4 is a target gene of miR-493. Also knockdown of FZD4 decreased cell migration and motility in T24 and J82 cells (55). As it is possible that FZD4S shares the same 3’UTR sequence with FZD4, both may be targets of miR-493, but there have been no reports regarding FZD4S in mammalian cells including cancer (51).

FZD5

Human FZD5 was mapped to chromosome 2q33.3-q34 (56) and its expression was reported to be up-regulated in renal cell carcinoma and advanced prostate cancer tissues compared to normal kidney and benign prostatic hyperplasia (BPH) tissues, respectively (57, 58). Additionally FZD5 protein levels correlated with the Wnt target gene, cyclin D1 protein expression levels in renal cell carcinoma (57). In another report, Wnt7a, a ligand of FZD5, activated the Wnt/beta-catenin signaling pathway and cell motility/invasion in metastatic melanoma, endometrial and ovarian cancer cell lines (59, 60, 61, 62). These reports suggest that FZD5 may be a biomarker and potential therapeutic target gene in cancer.

FZD6

Human FZD6 is on chromosome 8q22.3-q23.1 (63) and expressed in most adult normal tissues, fetal brain, liver, lung and kidney and cancer cell lines. FZD6 expression has been reported to be higher in cancers such as squamous cell sarcoma and some adenomas (64). According to array comparative genomic hybridization (aCGH) data, chromosome 8q22.3 in 61% of prostate cancer cases was amplified and FZD6 expression correlated with amplification of 8q22.3 in prostate cancer (65). Also in another report, high FZD6 expression was significantly correlated with poor survival in human neuroblastoma (66).

In contrast, FZD6 is a negative regulator of the Wnt/beta-catenin signaling pathway through the CaMKII-mediated TAK1-NLK pathway (Wnt/Ca2⁺ signaling pathway) (7). However mouse FZD6 interacted with Wnt4 through mouse FZD6 CRD (conserved cysteine-rich domain) (67) and activated the Wnt/JNK signaling pathway (68). In another report, FZD6 and Wnt4 protein were widely expressed in several adenoma tissues however localization of beta-catenin in the nuclei was not observed. ERK1/2, which is a non-canonical related gene, was highly activated in GHomas and TSHomas (69). Considering these reports, FZD6 may be involved in the non-canonical Wnt pathway in cancer but not in the Wnt/beta-catenin pathway.

Transcription factor 1 [Tcf1; hepatocyte nuclear factor 1a (HNF1a)] plays an important role in human hepatocytes. Down-regulation of the miR-192/-194 cluster was found in the livers of Tcf1^−/− mice. Over-expression of miR-194 decreased FZD6 mRNA and luciferase activity with plasmids containing FZD6 3’-UTR sequence in miR-194 transfected HEK 293 cells compared to controls (70).

FZD7 and miRNA

Human FZD7 resides on chromosome 2q.33 (17) and is expressed in adult normal skeletal muscle, heart, brain, placenta, kidney, fetal kidney and lung. FZD7 mRNA is up-regulated in several cancers including esophageal, gastric, nasopharyngeal, adenoid cystic, hepatocellular, colon, Wilms’ tumour (71) (FzE3 primers shown in this paper were FZD7 based on BLAST [http://blast.ncbi.nlm.nih.gov/Blast.cgi].) (72–78). Additionally high FZD7 expression correlated with a significantly shorter survival time in colon and gastric cancer (10, 33, 79). FZD7 activates Wnt/beta-catenin signaling in several cancers including hepatocellular carcinoma, colon cancer and TNBC (triple negative breast cancer) (75–78, 80). It was reported that Wnt3 is a ligand of FZD7 (81). FZD7 also regulates Wnt/JNK signaling in colon cancer (10) and therefore may be involved in both canonical and non-canonical signaling pathways in cancer. Knockdown of FZD7 decreased cell growth in colon cancer cell lines with APC or CTNNB1 gene mutations (75), caused an increase in SNAI2 mRNA, a decrease in E-cadherin mRNA and inhibited MET (mesenchymal–epithelial transition) in non-adherent colorectal cancer cell carcinoid cell lines LIM1863-Mph. (82, 83). Expression levels of FZD7 mRNA were higher compared to other FZD family genes in colon cancer cell lines. It was report that the activity of the FZD7 gene promoter was regulated by beta-catenin in colorectal cancers (84). FZD7 mRNA levels were significantly higher in stage II, III or IV colon cancer tissues (10). Expression levels of Wnt11 mRNA were significantly higher in stage I-IV tumor tissues than in non-tumor tissues and correlated with those of FZD7 mRNA. Patient groups with high FZD7 and high Wnt11 were significantly associated with shorter disease free survival compared to low FZD7 and low Wnt11 groups (85).

Thus FZD7 may be a novel oncogene and several therapeutic options have been described. For example, compounds such as FJ9 and small interfering peptides (RHPDs) suppressed cell growth and activity of the Wnt/beta-catenin signaling pathway by inhibiting interaction between FZD7 and Dishevelled (DVL) in cancer cell lines (86, 87).

Currently miRNA has emerged as another FZD7 regulator. A previous study indicates that miR-23b targets FZD7 in colon cancer (88). MiR-23b down-regulated cell migration, invasion and induced apoptosis based on genome-wide functional screening (88). Primary tumor growth and lung metastasis were inhibited in miR-23b transfected colon cancer cells (HCT–116) compared to controls. FZD7 protein levels and luciferase activity with the FZD7 3’-UTR in miR-23b transfected HCT-116 cells were decreased compared to mock transfectants, indicating that miR-23 directly targets FZD7. Also knockdown of FZD7 decreased cell invasion in HCT-116 cells (88). The expression of miR-23b was significantly downregulated in several cancer tissues (89) and studies indicate a miR-23b tumor suppressor function targeting cMET in hepatocellular carcinoma (90).

FZD8

Human FZD8 is located on chromosome 10p11.2 (91) and expressed in fetal kidney and brain and in adult kidney, heart, pancreas, and skeletal muscle. In some cancers (renal cell carcinoma, acute lymphoblastic leukemia and lung cancer), FZD8 expression was found to be up-regulated (41, 57, 92) and related to lung cancer cell growth though activity of the Wnt/beta-catenin signaling pathway (92). FZD8 activate Wnt/beta-catenin signaling in the presence of LRP6 and Wnt-3a in cancer (93), though there are few reports about its role and mechanism.

FZD9

Deletion of a part of chromosome band 7q11.23 was reported in Williams syndrome and human FZD9, previously reported as FZD3, in the deletion region was cloned by Wang et al. (94, 95). FZD9 was reported to be expressed in normal brain, testis, eye, skeletal muscle and kidney and FZD9 expression was up-regulated in glioblastoma and astrocytoma (96). Intensity of FZD9 immunostaining was strongly associated with tumor grade in glioblastoma and astrocytoma (96). The FZD9 promoter is methylated in glioblastoma multiforme, myelodysplastic syndromes and acute myeloid leukemia (97, 98) and hypermethylation was associated with poor survival in these diseases (98).

The degree of promoter methylation and expression of FZD9 may be a tumor marker in cancer. Although FZD9 or Wnt7a individually did not inhibit growth of non-small cell lung cancer, co-expression of FZD9 and Wnt7a decreased cell growth and promoted cell differentiation through ERK-5-dependent activation of PPARγ and Sprouty-4 (99, 100, 101). Knockdown of FZD9 decreased cyclin D1 protein expression and suppressed cell growth/motility in hepatoma cell lines (102). Thus FZD9 may be a tumor suppressor or oncogene in the presence of Wnt ligands in different kinds of cancer.

FZD10

Human FZD10 has been mapped to chromosome 12q24.33 (103) and found to be expressed in adult normal placenta, brain, heart, lung, skeletal muscle, pancreas, spleen, and prostate and fetal kidney, lung and brain. Similar to others in the FZD family, FZD10 expression was reported to be higher in some cancer tissues (colon, lung squamous cell carcinoma, synovial sarcoma) (104–107). Concerning the regulatory mechanism of FZD10, HIG2 (hypoxia-inducible protein-2) activated Wnt/beta-catenin signaling by binding to FZD10 in renal cell carcinoma (108). FZD10 increased the phosphorylation level of c-jun in endometrial cancer in the presence of Wnt7A (61) and activated the Wnt/Dvl-Rac-JNK signaling pathway. These reports show that FZD10 regulates both the canonical and non-canonical signaling pathways in the presence of different ligands in cancer. Based on the recent literature, some groups have documented the effectiveness of anti-FZD10 antibody therapy to synovial sarcomas (SS) since FZD10 increases cell growth in synovial sarcoma (107, 109).

Although there are no reports about miRNAs targeting FZD10, the success of antibody based therapeutics will be helpful for miRNA replacement therapy in cancer treatment.

Relationship of FZD with co-receptors and other Wnt related genes in cancer

LRP5/6, RYK and ROR2 have been known as co-receptors of FZDs (110, 111). When exposed to Wnts, LRP5/6 forms a complex with Wnt and Fzs (Wnt/FZD/LRP complex), resulting in activation of the Wnt/beta-catenin pathway (112–118) (Figure 1B). The Wnt/FZD/LRP complex recruits axin to the plasma membrane and inhibiting destruction of the complex. Based on previous literature, LRP5 or LRP6 have been regarded as biomarkers in several cancers such as osteosarcoma (119) and breast cancer (120–122) and anti-LRP6 antibody blocks the Wnt/beta-catenin pathway inhibiting cell proliferation in cancer cells (123, 124). Two additional receptors such as receptor-like tyrosine kinase (RYK) and receptor tyrosine kinase-like orphan receptor 2 (ROR2) can bind to Wnts (125). RYK is a PTKs (protein tyrosine kinase) family protein (126–128) and is required to activate the Wnt/beta-catenin and Wnt/JNK signaling pathway (147–149) (Figure1B and 1D). It has been reported that RYK is a tumor marker in ovarian cancer (129, 130) and the RYK gene is truncated in leukemia (131). ROR2 has a tyrosine kinase-like domain (132) and activates the Wnt/beta-catenin pathway in the presence of FZD2/Wnt3a in lung cancer cells (133). ROR2 knockdown blocks the Wnt/JNK pathway (134) and inhibits cell metastasis in several cancer cells (osteosarcoma, melanoma, kidney and prostate cancer) (134–137) (Figure1B and 1D). ROR2 is a prognostic tumor marker and therapeutic target in leiomyosarcoma and gastrointestinal stromal cancer (138). Although it is highly possible that these co-receptors play an important role in Wnt signaling in cooperation with FZDs, there have been few reports regarding the relationship and interaction between FZDs and LRP/RYK/ROR2 in cancer.

Apart from co-receptors, several Wnt antagonists have been identified and reported as inhibitors of Wnt signaling pathways. Conventional Wnt antagonists such as sFRP (secreted Frizzled-Related Protein), Dkk (Dickkopf) and Wif1 (Wnt inhibitory factor 1) bind to Wnt and LRP and inhibit the Wnt signaling pathway (Figure 1A) (139). Based on previous reports, expression of sFRP, Dkk and Wif1 is down-regulated in many cancers because of promoter hypermethylation (140–145). In the Wnt/beta-catenin signaling pathways in cancer cells, the function of APC gene, a crucial tumor suppressor, is decreased or lost because of gene mutation. In contrast, the beta-catenin gene is mutated at phosphorylation sites, resulting in inhibition of beta-catenin degradation (146). Down regulation of the Wnt antagonists and mutation of APC/beta-catenin genes is often paralleled with up-regulation of FZDs (Table 1), but their direct relationships are not currently understood.

As mentioned above, miRNAs may be involved in the regulation of FZD protein expression. Interestingly some of hose miRNAs are mapped to deleted chromosomal regions based on previous reports as follows: 1. chromosome 9p21 (miR-31 targeting FZD3), 2. 9q21 (miR-204 targeting FZD1), 3. 9q22 (miR-23b targeting FZD7) and 4. 14q32 (miR-493 targeting FZD4) (147–153). These specific chromosome deletions may cause over-expression of these FZDs thorough loss of a particular miRNA (Figure 2).

Figure 2.

Association of FZD with other Wnt related genes and proteins. Chromosome deletions may cause over-expression of some FZDs thorough loss of a particular miRNA. Expression of sFRP, Dkk and Wif1 is down-regulated because of promoter hypermethylation. Mutation of APC/beta-catenin genes is detected in many cancers.

As shown in Figure 2, FZD plays an important role with other co-receptors such as LRP5/LRP6, RYK and ROR2 in Wnt signaling in cancer cells. Of course the expression pattern and function of FZD and other Wnt related genes are different depending on the cancer types. However many diverse factors are involved in Wnt signaling, thus it is important to focus on the various genes and proteins to identify the regulatory mechanisms of Wnt signaling in cancer cells.

Conclusions and future perspectives

As described above, currently ten FZD receptors have been identified. Based on previous literature, FZDs expression is higher in cancer tissues suggesting that FZDs will be potentially valuable therapeutic targets. As a therapeutic approach, some groups have used anti-human FZD10 antibody for cancer treatment. For instance, mouse anti-human FZD10 monoclonal antibody labeled with Yttrium-90 (⁹⁰Y) decreased in vivo tumor growth in mice with FZD10-positive synovial sarcoma cell lines SYO-1 and FZD10-transfected colon cancer cell line DLD-1 (109, 154).

Additionally new compounds such as FJ9 and small interfering peptides (RHPDs) have emerged as potential therapeutic tools by inhibiting the interaction between FZD7 and Dishevelled (DVL) in cancer cell lines (86, 87). Another study showed that rat anti-human FZD7 monoclonal antibody decreased the number of spheres, colonies and in vivo proliferation of the chick chorio-allantoic-membrane in primary FZD7-positive Wilms’ tumor (155). Antibody treatment against other FZDs may also affect cancer cell growth and metastasis. In a basic research setting, targeting therapy with small RNAs has been shown to be effective. For example, siRNA (small interfering RNA) and shRNA (short hairpin RNA) are loaded onto RISC (RNA inducible silencing complex) to induce mRNA cleavage of target genes (156) and clinical trials of siRNA and double-stranded RNA treatments have been performed in cancer therapy (156).

Recently microRNA has emerged as a new treatment option. MicroRNAs trigger mRNA cleavage or inhibition of target gene translation after being loaded onto RISC (156). So far, numerous labs have focused on the study of miRNAs and found that they dose-dependently inhibit cell proliferation and metastasis in an in vivo models (157–159). Thus miRNAs have been used as therapeutic agents in cancer therapy. In spite of much progress in miRNAs replacement therapy or siRNA therapy both in vitro and in vivo, there are still several problems to be addressed for clinical application. The most important issue is the safety of miRNA or siRNA delivery systems when these replacement therapies are performed in a clinical setting. Whether miRNA replacement therapy or siRNA therapy will be effective as novel anti-cancer therapy will to depend on the success of the delivery system.

Over-expression of tumor suppression miRNAs targeting FZDs, results in suppression of cell growth and metastasis through the Wnt signaling pathway. Since there are few reports describing miRNAs targeting FZDs for use in therapeutics, the antibody therapeutic successes will provide helpful information in this regard. Additional FZDs research will also contribute to uncovering and utilizing microRNAs as therapeutic options for cancer treatment.