Frizzled: A brief summary of the characteristics and signalling versatility
Frizzled (FZD) receptor family is a small atypical class of G-protein-coupled receptors (GPCRs), comprising FZD subtypes and a structurally related Smoothened (SMO) found in animal species. FZDs initiate intracellular signalling cascades by interacting with a family of secreted lipoglycoproteins Wingless/Int-1 (Wnt), and with their specific co-receptors. SMO differs from FZDs in its mode of operation: its high tendency to constitutive activity is repressed by Patched (Ptc), a 12-spanning transmembrane receptor for hedgehog (Hh) family lipoglycoproteins. Hh binding to Ptc frees SMO, allowing the downstream signalling cascades to occur. The intracellular signals initiated by these receptors are prominent during embryogenesis and in tissue homeostasis, controlling cell proliferation, the motility, retaining stem cells, and directing cell fate. The de-regulation of the signalling events has clinical implications in the progress of cancer metastasis and degenerative diseases. This article focuses to summarise the functional characteristics of FZDs and SMO (in part 1), and the diverse intracellular signalling events initiated by these receptors (in part 2).
The contents may be changed at occasional updating: additions and corrections.ABBREVIATIONS: APC, adenomatous polyposis coli; βTRCP, β-transducin repeat-containing protein; CaMKII, calmodulin kinase II; CKI, casein kinase I; Ci, Cubitus interruptus; CRD, Cys-rich domain; Dkk1, dickkopf1; DVL, dishevelled; FZD, frizzled; GSK3, glycogen synthase kinase 3; Hh, hedgehog; JNK, c-jun NH2-terminal kinase; LRP, Low density lipoprotein-related protein; NFAT, nuclear factor of activated T cells; NLK, Nemo-like kinase; PCP; planar cell polarity; PDZ, PSD-95-discs-large-ZO-1; PKA, cAMP-dependent protein kinase; PKC, protein kinase C; Ptc, Patched; ROR, Receptor tyrosine kinase-like orphan receptor; RYK, receptor related to tyrosine kinase; Wg, wingless; WIF, Wnt inhibitory factor; Wnt, Wingless/Int-1.
1. Frizzled Family
1.1. Frizzled (FZD)
The major members of the frizzled family is a range of Frizzled (FZD): ten FZDs (FZD1-10) have been identified in mammals, chicken and clawed frogs of genus Xenopus; and four in fruit flies of genus Drosophila and nematode worm species Caenorhabditis elegans. The diverse signalling
FZDs can be activated by interacting with Wingless/Int-1 (Wnt) family proteins alongside the specific co-receptors low density lipoprotein-related receptor (LRP6 or 5). Its function could possibly be affected by receptor tyrosine kinase-like orphan receptor (ROR), or receptor related to tyrosine kinase (RYK), though how explicitly these receptors interact is mostly uncertain. Numerous genes encoding Wnt have been identified, at least 19 in mammals, seven in Drosophila, and five in C. elegans. By associating with specific Wnt and the co-receptor, FZDs are capable of initiating distinct patterns of intracellular signalling cascades appropriate to the conditions.Three streams of signalling cascades are known to be initiated by FZDs. Upon associating with LRP5/6, the FZD signalling causes β-catenin to accumulate in the cytosol and translocate to the nucleus for gene transcription: this signal event is known as the canonical pathway, for its capability in inducing an ectopic axis in Xenopus embryos and morphological transformation in mammalian cells. Another pathway initiated by FZDs involves small GTPases and c-jun N-terminal kinase (JNK); it causes a coordinated polarised orientation of cells, the phenomena known as planar cell polarity (PCP), and so the signalling is often called the PCP pathway. In addition, certain FZDs including FZD1 and FZD2 can couple to Gq (Malbon et al., 2001); although the major signalling via FZD1 is the canonical pathway, it is also capable of activating Gq/G11 in the presence of Wnt3A (Gao and Wang 2007) for an instance. By that means FZDs facilitate calcium signalling, activating protein kinase C (PKC) and calmodulin kinase II (CaMKII) (Sheldahl et al., 1999; Kuhl et al., 2000).
The C-terminal tails of FZDs
Among the FZD subtypes, the C-terminal region varies largely in length. The study employed chimeric Drosophila FZD1 (fz) and FZD2 (fz2) revealed that a determinant of the differential sub-cellular localisation and signalling of these receptors is within the C-terminal domain (Wu et al., 2004). The length of C-terminal tail in the fz is about one third of that in fz2, the former mainly comprising two functional motifs KTxxxW and PDZ binding site, both of which are also present in fz2. The conserved KTxxxW is present two amino acids after the seventh transmembrane domain; this motif is required for the activation of dishevelled (DVL), which initiates the canonical and PCP pathways by utilising its distinct domains. The shorter C-terminal domain of fz appear to favour its apical localisation where the PCP pathway predominates (Wu et al., 2004). The lengths of the putative C-terminal tails are as followed (n. residues): Drosophila fz(27), fz2(88), fz3(71), fz4(154); human FZD1(24), FZD2(24), FZD3(167), FZD4(38), FZD5(63), FZD6(211), FZD7(24), FZD8(88), FZD9(61), FZD10(57). The uncanny coincidence for FZD8 makes it memorable for sure.1.2. Smoothened (SMO)
An unique member of this family, smoothened (SMO), has a high level of constitutive activity which is impeded by its affiliate, a 12-pass transmembrane protein, Patched (Ptc). Ptc family proteins are structurally related to bacterial transporters of the larger resistance nodulation division family proteins, which utilise proton as an energy source. The activation of SMO is initiated by hedgehog (Hh) family proteins, which bind to the Ptc and releases SMO. Differences in the mode of SMO operation exist between Drosophila and mammals. In Drosophila, Ptc-bound SMO is retained in cytoplasmic vesicles. Hh binding to Ptc frees SMO, which then becomes phosphorylated by cyclic AMP-dependent protein kinase (PKA) or casein kinase I (CKI), and accumulate at the plasma membrane (Denef et al. 2000; Jia et al. 2004; Zhang et al. 2004).
Mammalian SMO, on the other hand, is exported to the cell-surface at inactive state bound to Ptc. Hh-binding to Ptc causes it to internalise, upon releasing SMO. The free SMOs translocate and accumulate at the primary cilium, which is a microtubule-based structure protruding from the surface of effectively all mammalian cells. Though only one gene encoding Hh is found in Drosophila, in mammals there are three paralogous genes encoding for Hh proteins: Desert Hedgehog (DHh), Indian Hedgehog (IHh), and Sonic Hedgehog (SHh).
Sterols influence SMO activities
All Hh proteins have a sterol-recognition region (SRR) at the C-terminal tail. The cholesterol modification appears to help Hh localising at the cell-membrane (Gallet et al., 2006; Zeng et al., 2001). The sterols have been shown as a necessity for SMO function (Cooper et al., 2003). The activity of SMO can be positively regulated by oxysterols, which may act as the maximum full agonist at SMO (Corcoran et al., 2006; Dwyer et al., 2007). Elevated plasma concentration of 24-S-hydroxycholesterol was seen in the sufferers of Alzheimer’s disease and vascular dementia (Lütjohann D et al., 2000); however, the effect of that on SMO is yet to be identified. The SMO activity is negatively regulated by pro-vitamin D3 (Bijlsma et al., 2006) and by steroidal alkaloid, cyclopamine (11-deoxojervine), isolated form the corn lily, Veratrum californicum. Ptc has a sterol-sensing domain (SSD) which participates in SMO inhibition (Martin et al., 2001). Ptc family contains Niemann-Pick C1 protein (NPC1) with sterol transporter activity, thereby taking a major role in intracellular trafficking of cholesterol; NCP1 and several other Ptc family are known to have Hh receptor activity, suggesting probable functional redundancies existing among the family members.
These facts altogether clearly indicate that the SMO functions are largely influenced by sterols and similar lipids. The relative complexity of mammalian Hh/SMO systems, in comparison to fruit flies, is thought to have emerged since mammals utilise and metabolise sterols more extensively than insects.
1.3. Shared characteristics between FZD and SMO
Cysteine-rich domain (CRD)
A common feature between FZDs and SMO is a large N-terminal region containing highly conserved Cys residues forming a Cys-rich domain (CRD) also known as Frizzled domain, via intramolecular disulfide bonds. CRD engages in transient heterotrimerisation comprising the receptor, the lipoglycoprotein ligand, and the co-receptor. A high resolution crystal structure of CRD has been determined by Dann et al. in 2001 (Fig. 1). The overall sequence similarity within the domain among FZDs varies between 30 and 50% approximately.Using fusion proteins, IgG fused with CRD from murine FZD8 and the soluble alkalylphosphatase fused with Xenopus Wnt8 produced in Drosophila S2 cells, Hsieh et al. demonstrated that FZD can bind to the Wnt via this domain with a high affinity (Kd ≈ 9 nM by a solid phase binding assay) (Hsieh et al., 1999). The affinity of Wnt for CRD in specific ligand-receptor (i.e. Wnt-FZD) pairs has been said to correlate with the magnitude of the signalling (Rulifson et al., 2000; Wu & Nusse, 2002); however in Drosophila FZDs (fz and Dfz2), CRD was suggested to have fewer importance in Wingless (Wg)-activation of the canonical pathway involving β-catenin orthologous Armadillo, based on their observation upon the mutant constructs lacking CRD (Chen et al., 2004). Nonetheless, the deletion of CRD in the study was identified to be incomplete with 8 residues of CRD still intact; hence a complete CRD-deletion was subsequently performed by Povelones and Nusse. The study confirmed the necessity of CRD in Armadillo signalling, in particular prominently in the cultured S2 cell system. The study further showed the specificity of CRD by replacing fz CRD with that of SMO; the construct was non-functional, whilst the fz CRD substitution with Wnt inhibitory factor (WIF) domain resulted in a functional construct (Povelones and Nusse, 2005). In SMO, the necessity of CRD in achieving the maximum activation has been shown by the studies constructed SMO lacking CRD (Murone et al., 1999, Aanstad et al., 2009).
 | ||
| Fig. 1. The Crystal Structure of the Cysteine-rich Domains of murine Frizzled 8 (MFZD8): PDB id 1IJY; | The structure was obtained by X-ray diffraction at 1.35 angstrom resolution by Dann et al. in 2001. The conserved Cys residues forming disulfide bonds are shown in yellow. The PDB file was visualised using UCSF Chimera version 1.4.1. (University of California, San Francisco). |
A mutuality between Wnt/Int1 and hedgehogs
1.4. The distant family members
A cellular slime mould species Dictyostelium discoideum is known to have two types of FZD/SMO-like proteins, 16 with CRD and 9 without. So far, these GPCRs named above are all categorised under this family to this date; these are identifiable at UniprotKB (www.uniprot.org).2. Signalling by Frizzled and Smoothened
2.1. Frizzled signalling
FZDs are atypical GPCRs because they often employ small phosphoprotein dishevelled (DVL) as their primary transducer, though some (e.g. FZD1 and FZD2) can also signal through Gi/0 or Gq/11. Drosophila has only one type of DVL, while three subtypes are found in mammals and in Xenopus. DVLs function as adaptors, typically having three interacting domains: Dishevelled and Axin binding (DIX) domain at the N-terminal region, PDZ domain in the middle, and Dishevelled Egl-10 Pleckstrin (DEP) domain near the C-terminal region. By the DIX domain DVL can dimerise; hence three DVL types, each offering two other binding sites, presumably with distinct accessibility to potentially interacting partners, could offer intracellular signals in variation by different combinatory pairs depending on the cellular component.
2.1.1. Canonical pathway
A few consequences
1. Beta-catenin activation
The major event established in the canonical pathway is the activation of a transcription factor, β-catenin. In the absence of the Wnt stimulation, low cytosolic free β-catenin concentration is maintained by proteolysis initiated by glycogen synthase kinase 3β (GSK3β) which phosphorylates β-catenin along with casein kinase 1α (CKIα) for its ubiquitylation. GSK3β has a tendency associating with its substrates, adenomatous polyposis coli (APC) and Axin. APC binds to the ends of microtubules; some phosphorylated β-catenin associates with APC to localise at protruding pseudopodia (Faux et al., 2010), whilst some contribute to the cell to cell adhesions by linking E-cadherin and actin cytoskeletons (Drees et al., 2005; Yamada et al., 2005). A rapid increase in cytosolic β-catenin concentration in the canonical pathway is thought to be achieved by de-phosphorylating and releasing these from the cytoskeletal captivity.
Specific hetero-oligomerisation of Wnt, FZD and LRP6/5 activates DVL as CKIε phosphorylates it, and DVL binds to Axin; β-arrestin2 also binds to both DVL and Axin (Vryja et al., 2007). The multi-protein complexes stay by the plasma membrane as Axin associates with the C-terminal domain of LRP6/5, which is then sequentially phosphorylated by CKIγ (Davidson et al., 2005) and by GSK3β (Zeng et al., 2008). The affinity of the region of LRP6 to GSK3β appears sufficiently high to confine GSK3β function at the vicinity (Piao et al., 2008). The reduced GSK3β activity increases free un-phosphorylated β-catenin concentration in the cytosol, allowing its translocation to the nucleus where it initiates gene transcriptions in collaboration with DNA-bound T cell-specific transcription factor/lymphoid enhancer-binding factor 1 (TCF/LEF) family members.
1.2a Stabilisation of axonal microtubules
In neurones GSK3β phosphorylates microtubule-associated proteins such as MAP-1B (Lucas et al., 1998) and Tau (Hanger et al., 1992) to make them stabilise axonal microtubules in assisting neurite outgrowth (Cuchillo-Ibanez et al., 2008). Wnt-7a has been shown to enlarge growth cone, induce branching, and shorten axon (Lucas and Salinas, 1997; Hall et al., 2000); such effects can be initiated via DVL1 and LRP6 which prevents GSK3β from phosphorylating those (Krylova et al., 2000; Cianni et al., 2003). APC also binds to the microtubules and stabilise the assembly in GSK3β-dependent manner (Zumbrunn et al., 2001).
1.2b The re-localisation of APC in neurones
Wnt-7a has shown to induce localisation of APC at the presynaptic region of hippocampal neurones upon the canonical pathway activation; the study revealed a co-localising tendency of APC with a highly Ca2+-permeable transporter, α7-nicotinic acetylcholine receptor, possibly implicating its participation in promoting neurotransmitter release (Farias et al., 2007).
The effect of G-proteins from other signals
Axin has a RGS (regulators of G-protein signalling) domain. The RGS domain binds to Gα to accelerate GTP hydrolysis. The domain has been shown to bind specifically to the G12 at an active state, but without affecting the rate of catalysis, thereby reducing the activation of G12 effectors such as small G-protein Rho by competing G12 binding (Stemmle et al., 2006). The RGS domain of Axin has also been found to bind to Gs activated by prostanoid EP2 receptor, and that in turn reduce GSK3β phosphorylation of β-catenin (Castellone et al., 2005).
The receptor internalisation
The co-receptor LRP6 is subjected to caveolae-induced endocytosis (Li. et al., 2010). One might wonder whether β-arrestin1/2, either of which functions as a scaffolding component in activating the canonical pathway (Chen et al., 2001; Vryja et al., 2007), to effectively induce subsequent receptor internalisation; that appears to occur at least for FZD4 which has been reported to internalise via β-arrestin2 recruited by DVL2 (Chen et al., 2003).
Inhibitory modules of the canonical pathway
1. Dickkopfs (Dkks)
The Wnt binding to its specific FZD-LRP6/5 pair can be competed by Dickkopf (Dkk) family secreted glycoproteins, which bind to LRP6/5 and single-pass transmembrane Kremen1/2 (Krm1/2) (Davidson et al., 2002; Mao et al., 2001). In the absence of Dkk1, Krm2 has been shown to facilitate Wnt signalling in Xenopus oocyte, presumably by retaining cell-surface population of LRP5/6 (Hassler et al., 2007). Dkk2 has been reported to be either agonist or antagonist to LRP6 depending on the cellular environment (Mao and Niehrs, 2003; Wu et al., 2000). A research article on cnidarian Dkks by Guder et al. offers further information on Dkks from evolutional perspectives (Guder et al., 2005).
2. SOST gene products
A similar mode of inhibition has been documented involving the SOST gene products, which associate with LRP5/6 to disrupt Wnt/FZD/LRP complexes (Li et al., 2005; Semenov et al., 2005). Sclerosteosis and Van Buchem disease are autosomal recessive diseases with bone overgrowth linked to defective SOST gene function. For those interested, the involvement of Dkk and SOST in bone diseases is neatly reviewed by Mason and Williams in 2010.
3. Proteins with PDZ domain and membrane-localised
Since DVL binds to the C-terminal domain of FZD via its PDZ domain (Wong et al., 2003), other proteins with PDZ domain at the vicinity could interfere the functional interaction. This has been demonstrated for Na(+)/H(+) exchange regulatory factor1 (NHERF), of which expression negatively correlated with the β-catenin level quantified in human breast carcinomas (Wheeler et al., 2011).
4. Gq activation
The canonical pathway could also be affected negatively by heterologous activation of Gq. The reason has been reported due to the calcium mobilisation activating a calcium-dependent protease calpain which degrades β-catenin. The negative effect on the β-catenin activation has been demonstrated by the co-activation of M3 muscarinic acetylcholine receptor (M3R) in human colon cancer cell-line SW480 and human embryonic kidney cell-line HEK293 (Li and Iyengar, 2002). The over-expression of murine FZD1, which couples to Gq, has been shown also to decrease Wnt1- or Wnt3a-induced β-catenin accumulation in mesenchymal cells or Xenopus oocytes; the effect was reversed by co-expression of the C-terminal peptide of Gq (305–359), which favours Gq-uncoupling of the receptors (Roman-Roman et al., 2004).
The calcium activated Nemo-like kinase (NLK), which inhibits TCF/LEF transcription, certainly contribute to the negative effect observed above (Ishitani et al., 1999). The regulation at the receptor level may also account for this, by the activities of major PKC isoforms induced by Gq: PKC may phosphorylate FZDs for β-arrestin induced internalisation, or for reducing the receptor association with DVL.
Some of the PKC isoforms can also inactivate GSK3β (Fang et al., 2002); though the reduction on GSK3β activity can favour increasing β-catenin concentration, the effect might be small if the sites are easily accessed by phosphatases. Alternatively, GSK3β complexed with Axin and APC might be protected from the inhibition if the associations make the sites inaccessible to PKC.
Activators of the canonical pathway
in vertebrates, the secreted R-spondins (R-spo1-4), each containing two furin-like Cys-rich domains at the N-terminus, a thrombospondin domain, and positive charge at the C-terminal domain, have been shown to enhance the activation of canonical pathway by increasing the phosphorylation of LRP6. In addition, the furin domains of R-spo1 prevent Dkk1-mediated LRP6 binding to Krm (Binnerts et al., 2007; Kim et al., 2008).
LiCl has also been shown to increase β-catenin accumulation by inhibiting GSK3β (Klein and Melton 1996).
The intracellular domains of FZD: key residues in the canonical pathway
Mutagenesis studies have identified a few residues involved in the activation of the canonical pathway. Umbhauer et al. identified the importance of highly conserved KTxxxW(Y) motif by the functional analysis of a series of mutant constructs in Xenopus oocytes (Umbhauer et al. 2000). Cong et al. substituted residues in the intracellular domains of rat FZD1 and human FZD5 one by one, and the LEF-dependent transcription activity of these mutant constructs were assessed in S2 cells: the study identified Arg in ICL1 (rFZD1-R340; hFZD5-R263), Leu in ICL3 (rFZD1-L524; hFZD5-L443), and Lys (rFZD1-K619; hFZD5-K525) and Trp (rFZD1-W624; hFZD5-W530) of the KTxxxW motif are all necessary components in initiating the canonical pathway; Arg in ICL, Leu in ICL3 and Lys in the motif have also been shown to be required for the receptors’ association with DVL (Cong et al., 2004).
Additional Information
Some aspects of more detailed descriptions for the canonical pathway and its components can be found in a comprehensive review article by MacDonald in 2009.
2.1.2. Non-Canonical pathway
Involvement of receptor tyrosine kinases
Some Wnts bind to specific receptor tyrosine kinase (RTK): first, ROR family of RTK, initially identified as RTK-like orphan receptor, ROR1/2 containing a CRD domain which binds to Wnt; second, RYK is an atypical RTK with an extracellular domain resembling the secreted WNT inhibitor factor (WIF) which binds to WNTs with a high affinity (Hsieh et al., 1999b; Liu 2005). The extracellular domains of ROR1/2 and C. elegans orthologue CAM-1 mutually contain an immunoglobulin (Ig)-like domain, CRD domain, and kringle domain, whereas the Ig-like domain is absent in Drosophila orthologues, Dror and Dnrk. Relatively well-studied RYK orthologous are Derailed (Drl) in Drosophila and LIN-18 in C.elegans.
These receptors may respond individually to Wnt, or could function as co-receptors to FZDs. The receptors may simply function accordingly to the receptor populations in the local environment, and the documented functions of these receptors might reflect degrees of heterogeneity of membrane protein species in the cell membrane. Along ROR1/2, muscle specific RTK (MuSK) and neuronal specific MuSK (NSK2) were identified to have CRD domain (Saldanha, Singh and Mahadevan, 1998); however their functions in relation with FZDs are unknown.
1. RTK activity of ROR1/2 and RYK
ROR1/2 were shown to exhibit RTK activity independent of FZD (Liu et al., 2008; Minami et al., 2009). ROR2 has a Pro-rich domain (PRD) involved in its auto-phosphorylation: by the domain it associates with casein kinase 1ε which phosphorylates Ser/Thr residues nearby; this induces its auto-phosphorylation of Tyr residues present within the PRD by its Tyr kinase domain (Kani et al., 2004). It has been reported that the kinase activity of RYK might be impaired relative to ordinary RTKs (Katso, Russell and Ganesan 1999), possibly implying its other role, for instance as being a co-receptor to FZDs utilising its PDZ domain, may be an integral function of RYK.
2. A unique property of RYK
An unusual characteristic of RYK was seen in Wnt3 stimulation induced a C-terminal cleavage of RYK and its translocation to the nucleus during neuronal development (Lu et al., 2004).
3. Possible (?) ternary complexes concerning a ligand & receptors
A) FZD-ROR-Wnt
The CRD domain of ROR2 could interact with Wnt5a and the CRD domain of rat FZD2 (Oishi et al., 2003).
B) FZD-RYK-Wnt
The WIF domain of RYK could associate with CRD of FZD8, and the complex could be bound by Wnt1 to mediate the canonical pathway (Lu et al., 2004). Though not a direct interaction confirmed, RYK-Wnt11 can mediate FZD7 internalisation via β-arrestin2 in X. laevis (Kim, Her and Han, 2008).
4. Downstream recruitment
Wnt-binding to FZD, also possibly in association with ROR1/2 or RYK, could recruit specific DVL to initiate non-canonical pathways wherein DVLs recruit Rho and Rho-associated kinase (Rho-kinase) (Strutt et al., 1997; Habas et al., 2001; Winter et al., 2001; Fanto and McNeill, 2004); the possible effector molecules RhoA, RhoU, RAC, CDC42, and also JNK (Boutros et al., 1998; Tao et al., 2001). In X. laevis, PKCδ was shown to be required for DVL recruitment by FZD7 and subsequent activation of JNK pathway (Kinoshita et al., 2003). Wnt 5a binding could recruit β-arrestin in activating RAC-1 signalling, influenced by CK1 and CK2 (Bryja et al., 2008).
5. Simultaneous but separate activations, or co-activated in association?
In C. elegans, Wnt (CWN-2), ROR (CAM-1) and FZD (CFZ-2) appears to be involved in locating developing nerve ring to be centred around the isthmus of the pharynx (Kennerdell, Fetter and Bergmann, 2009). The same combination with DVL working together in neurite outgrowth was subsequently confirmed (Song et al., 2010).
Aside: ROR could also mediate the canonical pathway
In H441 lung carcinoma cell-line, ROR2 could mediate the canonical pathway in response to Wnt3a along with FZD2 (Li et al., 2008).
.. to be continued.
References
(Articles accessible openly are shown in Bold)
Bijlsma M.F. et al., 2006. PLoS Biol 4:e 232.
Binnerts M.E. et al., 2007. PNAS 104(37): 14700–14705.
Bürglin T.R., 2008. Genome Biol 9: e 241.
Bryja V et al., 2007. PNAS 104(16): 6690–6695.
Bryja V. et al., 2008. EMBO reports 9(12): 1244-1250.
Castellone M.D. et al., 2005. Science 310: 1504–1510.
Chen C. et al., 2004. PNAS 101(45):15961–15966.
Chen M.H. et al., 2004. Genes & development 18(6):641-659.
Chen W. et al., 2001. PNAS 98(26): 14889–14894.
Chen W. et al., 2003. Science 301(5638): 1391-1394.
Cianni L. et al., 2003. J. Cell Biol 164(2): 243–253.
Cong F., Schweizer L. and Varmus H., 2004. Development 131(20): 5103-5115.
Cooper, M.K. et al., 2003. Nat Genet 33, 508–513.
Corcoran R.B. and Scott M.P., 2006. PNAS 103:8408-8413.
Cuchillo-Ibanez I. et al., 2008. FASEB 22(9): 3186-3195.
Dann C.E. et al., 2001. Nature 412:86-90.
Davidson G. et al., 2002. Development 129(24):5587-96.
Davidson G et al., 2005. Nature 438(7069) 867 872.
Dawber R.J., et al., 2005. BMC Developmental Biology, 5:e 21.
Drees F. et al., 2005. Cell 123: 903–915.
Dwyer J.R. et al., 2007. J Biol Chem 282:8959-8968.
Fang X. et al., 2002. Mol Cell Biol 22(7):2099-110.
Fanto M. and McNeill H., 2004. J. Cell Sci. 117(4): 527-533.
Farias G.G. et al., 2007. J Neurosci 27(20): 5313–5325.
Gallet, A. et al., 2006. Development 133 (3):407-418.
Gao Y. and Wang H.Y., 2007. JBC 282 (36):26490-26502.
Guder C. et al., 2005.Development 133(5): 901-911.
Hall A.C., Lucas F.R. and Salinas P.C., 2000. Cell. 100: 525–535.
Hanger D.P. et al., 1992. Neurosci. Lett. 147: 58–62.
Hassler C. et al., 2007. Development 134(23): 4255-4263.
Hsieh J.C. et al., 1999. PNAS 96(7):3546-51.
Ishitani et al., 1999. Nature 399(6738):798-802.
Kani S. et al., 2004. J Biol Chem 279:50102-9.
Katso RM, Russell RB and Ganesan TS., 1999. Mol Cell Biol. 19: 6427-6440.
Kennerdell JR, Fetter RD and Bergmann CI., 2009. Development 136: 3801-3810.
Kim K. et al., 2008. Mol. Cell. Biol. 19(6): 2588–2596.
Kinoshita N. et al., 2003. Genes &Development. 17: 1663-1676.
Klein and Melton 1996. PNAS 93: 8455-8459.
Kuhl, M. et al., 2000. J. Biol. Chem. 275 (17): 12701-12711.
Krylova O., Messenger M.J. and Salinas P.C., 2000. J Cell Biol 151(1): 83–93.
Li C. et al., 2008. BMC Molecular Biology 9: 11.
Li G. and Iyengar R., 2002. PNAS 99 (20): 13254–13259.
Li X. et al., 2005. J Biol Chem 280 (20): 19883–19887.
Liu Y. et al., 2005. Nat. Neurosci. 8: 1151-1159.
Liu Y et al., 2008. J Cell Biochem. 105: 497–502.
Li Y. et al., 2010. PLoS One. 5(6): e11014.
Lucas F.R. and Salinas P.C., 1997. Dev. Biol. 193:31–44.
Lütjohann D. et al., 2000. J Lipid Res. 41(2):195-8.
MacDonald B.T., Tamai K. and He X., 2009. Dev Cell. 17(1): 9–26.
Malbon C.C. et al., 2001.Biochem Biophys Res Commun. 287(3):589-93.
Mao B. et al., 2001. Nature 411: 321-325.
Mao B. and Niehrs C. 2003. Gene 302: 179-183.
Martin V. et al., 2001. Curr Biol 11(8):601-7.
Mason J.J. and Williams O., 2010. J. Osteoporosis. Article ID 460120, 9 pages.
Minami Y. et al., 2009. Dev Dyn 239: 1–15.
Murone M. et al., 1999. Curr Biol 9(2):76–84.
Nusse R., 2003. Development 130(22):5297-5305.
Oishi I. et al., 2003. Genes Cells 8(7):645-54.
Piao S. et al., 2008. PLoS ONE 3(12): e4046.
Povelones M. and Nusse R., 2005. EMBO 24: 3493–3503.
Roman-Roman S. et al., 2004. J Bio Chem 279(7): 5725–5733.
Rulifson, E.J., Wu, C. H. & Nusse, R., 2000. Mol. Cell 6(1): 117–126.
Semënov M., Tamai K. and He X., 2005. J. Biol. Chem. 280 (29): 26770-26775.
Sheldahl, L.C. et al., 1999. Curr. Biol. 9, (13): 695-698.
Song S. et al., 2010. PLos Genetics. 6(8): e1001056.
Stemmle L.N., Fields T.A. and Casey P.J. 2006. Mol Pharmacol 70(4):1461–1468.
Strutt D.I.,Weber, U. and Mlodzik, M., 1997. Nature 387: 292-295.
Strutt D. 2003. Development 130(19): 4501-4513.
Tao W. et al., 2001. Genes Dev. 15: 1796-1807.
Umbhauer M. et al., 2000. EMBO 10(18): 4944-4954.
van Amerongen R. and Nusse R., 2009. Development 136(19):3205-3214.
Varjosalo M. and Taipale J., 2008. Genes & Development 22:2454–2472.
Wheeler D.S. et al., 2011. Oncogene 30(1): 32-42.
Willert K. et al., 2003. Nature 423:448-52.
Winter C.G. et al., 2001. Cell 105(1): 81-91.
Wong et al., 2003. Mol Cell 12(5): 1251-1260.
Wu, C. H. & Nusse, R., 2002. J. Biol. Chem. 277(24):41762–41769.
Wu J., Klein T.J. and Mlodzik M., 2004. PLoS Biol 2(7):1004-1014.
Wu W. et al., 2000. Curr. Biol. 10: 1611-1614.
Yamada S. et al., 2005. Cell 123: 889–901.
Zeng, X. et al., 2001. Nature 411(6838): 716-720.
Zhang C. et al., 2004. PNAS 101(52):17900-17907.
Zeng X. et al., 2008. Development 135(2): 367-375.
Zumbrunn J. et al., 2001. Current Biology 11:44–49.
Further Information:
A fully comprehensive and detailed review on Frizzled class receptors:
Schulte G. 2010. Pharmacol Rev. 62(4):632-667.
No comments:
Post a Comment