QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online March 30, 2004
doi: 10.1242/10.1242/dev.01117

This Article Summary Full Text Full Text (PDF) Supplemental Figures Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by He, X. Articles by Zeng, X. Search for Related Content PubMed PubMed Citation Articles by He, X. Articles by Zeng, X.

LDL receptor-related proteins 5 and 6 in Wnt/ß-catenin signaling: Arrows point the way

Division of Neuroscience, Children's Hospital, Harvard Medical School, Boston, MA 02115, USA

View larger version (28K): [in a new window]   Fig. 1. A simplified prevailing view of Wnt/ß-catenin signaling (see Box 1 for some alternative views). (A) Without Wnt, the scaffolding protein Axin assembles a protein complex that contains Apc, Gsk3, Ck1 and ß-catenin. In this complex, ß-catenin is sequentially phosphorylated by Ck1 and Gsk3. Phosphorylated ß-catenin is recognized by ß-Trcp, which is a component of an ubiquitin-ligase complex that conjugates ß-catenin with ubiquitin. Poly-ubiquitinated ß-catenin is degraded by the proteosome. TCF/LEF-associated co-repressors, such as Groucho (Cavallo et al., 1998), and Axin-associated Diversin (Schwarz-Romond et al., 2002), PP2A (Hsu et al., 1999) and other proteins (Kikuchi, 1999) are omitted for simplicity. (B) In the presence of Wnt, ß-catenin phosphorylation and degradation is inhibited. Accumulated ß-catenin forms a nuclear complex with the DNA-bound TCF/LEF transcription factor, and together they activate Wnt-responsive genes. This signaling cascade is perhaps initiated by a Wnt-induced Fz-Lrp5/Lrp6 co-receptor complex, which recruits Axin to the plasma membrane through Lrp5/Lrp6-Axin association. Fz-associated Dishevelled (Dvl) protein may bind Axin and inhibit Axin-Gsk3 phosphorylation of ß-catenin, either directly or indirectly via Dvl-associated proteins. Lrp5/Lrp6-Axin binding may also promote Axin degradation. Either or both of these events can lead to ß-catenin accumulation. This description represents one of several possibilities. The composition of the Axin complex upon Wnt stimulation is not well defined. Gsk3-binding protein (GBP/Frat) (Farr et al., 2000; Salic et al., 2000), and nuclear ß-catenin-associated Legless/Bcl9 and Pygopus (Belenkaya et al., 2002; Kramps et al., 2002; Parker et al., 2002; Thompson et al., 2002) are omitted for simplicity. Modified with permission from He (He, 2003).

View larger version (24K): [in a new window]   Fig. 2. Schematic diagram of Arrow/LRP5/LRP6 and other members of the LDLR family. All proteins depicted are of human origin except for Arrow (Drosophila), and are conserved in vertebrates; some have invertebrate homologs (Herz and Bock, 2002). YWTD (Tyr, Trp, Thr and Asp)-type ß-propeller domains, LDLR type A (LA) domains and EGF (epidermal growth factor)-like domains are shared by all LDLR family proteins, although the domain arrangements vary. Arrow/LRP5/LRP6 are highly homologous and have the same extracellular domain arrangements. Intracellularly they do not have the NPxY [Asp, Pro, X (any amino acid) and Tyr] motif, but instead each have five copies of PPP(S/T)P (P, Pro; S/T, Ser or Thr) motifs. Besides Arrow/LRP5/LRP6, VLDLR (very low-density lipoprotein receptor) and APOER2 (apolipoprotein E receptor 2, also called LRP8) have established signaling roles by acting as receptors for the secreted signaling molecule Reelin. Other members of the LDLR family participate in lipoprotein/cholesterol uptake, steroid hormone uptake, regulation of cell surface protease activity and Ca2+ homeostasis, or are less characterized (reviewed by Herz and Bock, 2002). Many members have multiple names. The commonly used names are listed in bold.

View larger version (24K): [in a new window]   Fig. 6. Schematic representation of possible LRP6 domains involved in Wnt, Dkk and Wise binding, and of LRP5 mutations in human diseases. (Top) LRP6. Domains involved in Wnt, Wise and Dkk1 binding have only been mapped for LRP6 and are marked. Whether SOST binds LRP5/LRP6 is unknown. (Bottom) LRP5 mutations associated with osteoporosis-pseudoglioma (OPPG) syndrome, autosomal-dominant familial exudative vitreoretinopathy (FEVR), and various high bone density diseases are shown in red, purple and green, respectively. Arrows indicate mutation locations. *, nonsense mutation; fs, frame-shift mutation. OPPG is autosomal recessive, and the nine mutations indicated are from homozygous offspring of consanguineous families (Gong et al., 2001). FEVR discussed here is an autosomal-dominant form, possibly due to haploinsufficiency. Whether the three OPPG and three FEVR missense mutations (italicized) are loss-of-function mutations remains to be tested. T1449fs# in FEVR should be treated as hypothetical because the mutation occurs at a splice donor site in an intron. Note that the mutations associated with high bone density diseases, which are autosomal dominant because of probable `gain of function', are all missense mutations and are clustered in the first YWTD ß-propeller domain. SP, signal peptide; TM, transmembrane domain.

View larger version (26K): [in a new window]   Fig. 3. Schematic diagram of various Arrow/Lrp5/Lrp6 mutant proteins and their signaling properties. The wild-type Lrp5/Lrp6 is depicted on the left. Black bars in the intracellular domain represent PPPSP motifs. Lrp6C (without most of the intracellular domain), Lrp5N (secreted extracellular domain) and Lrp6m5 (alanine substitution of the S/T residue in all five PPP[S/T]P motifs) are dominant-negative reagents that can block canonical Wnt signaling. The following receptor mutants are constitutively active, i.e. they can activate ß-catenin signaling in the absence of Wnt: (1) Arrow/Lrp5/Lrp6N (without the extracellular domain but anchored on the membrane); (2) myristoylated Lrp5C (intracellular domain anchored to the plasma membrane via a form of lipid modification); (3) a single PPPSP motif linked to a truncated LDLR; and (4) Dfz2-Arr[intra], which is a fusion of the Arrow intracellular domain with the Dfz2 carboxyl-terminal tail. The Lrp5/Lrp6 intracellular domain that is not anchored to the plasma membrane is inactive. For reasons unclear at the moment, the Arrow intracellular domain designed to anchor to the plasma membrane via myristoylation is inactive in fly embryos, although its protein expression has not been examined (Tolwinski et al., 2003).

View larger version (6K): [in a new window]   Fig. 4. Schematic diagram of Axin, showing its binding sites for various interacting proteins, including Arrow/Lrp5. How Axin binds to Arrow/Lrp5 was mapped only via yeast two-hybrid assays and is not well defined. Axin-Lrp6 binding has not been mapped. Known binding sites for various Axin-binding proteins are depicted with square brackets on top. For Axin-Arrow/Lrp5 binding, the DIX domain and possibly the domains preceding it appear to be required (marked by the green line), whereas the RGS domain may be inhibitory (marked by the red line) (Mao et al., 2001b; Tolwinski et al., 2003). The broken green or red line indicates that the domain boundary has not been defined by mapping. The question mark indicates some ambiguity/inconsistency of the two mapping studies.

View larger version (28K): [in a new window]   Fig. 5. Three models of how Fz and Dsh/Dvl may function in relation to Arrow/Lrp5/Lrp6-Axin interaction. In all three models, Wnt-induced Arrow/Lrp5/Lrp6 phopshorylation on the PPP(S/T)P motifs provides docking sites for Axin. (A) The co-recruitment model. Wnt-induced Fz-Lrp5/Lrp6 complex co-recruits Dvl and Axin into the co-receptor complex via Fz-Dvl and Lrp5/Lrp6-Axin interactions. This proximity of Dvl and Axin, which can interact with each other, causes functional inhibition or degradation of Axin by Dvl, either directly or via Dvl-associated proteins. (B) The vesicle transport model. Axin is shuttled to the plasma membrane to its docking sites in Arrow/Lrp5/Lrp6 via Dvl-mediated `vesicular-type' transport, which relies on the ability of Dvl to bind Axin and phospholipids/vesicles. In this and the above co-recruitment model, Dsh/Dvl is downstream of and required for Arrow/Lrp5/Lrp6 function. (C) The parallel signaling model. Fz-Dvl-Axin and Arrow/Lrp5/Lrp6-Axin are two parallel and independent branches. Activation of ß-catenin signaling requires both branches under physiological conditions, but can occur when either branch is overactivated, for example, following overexpression of Dvl or Arrow/Lrp5/Lrp6N. This model accounts for the possibility that Dsh/Dvl may not be required for overexpressed Arrow/Lrp5/Lrp6 to activate ß-catenin signaling.