BoneKEy-Osteovision | Commentary
LRP5 at the crossroad of human genetics
DOI:10.1138/2001048
The human chromosome 11q12-13 locus has long been under scrutiny by human geneticists for one reason: both the osteoporosis pseudoglioma (OPPG) syndrome (OMIM 259770) and the high bone mass syndrome (HBM, OMIM 601884) were mapped to this region (). OPPG is an autosomal recessive childhood disorder involving both skeletal and eye abnormalities. OPPG patients present normal bone growth associated with severe osteopenia, although without abnormal collagen synthesis or hormonal defects. They also show congenital or juvenile-onset blindness due primarily to hyperplasia of the primary vitreous. Interestingly, obligate heterozygotes for the OPPG allele (i.e., parents of affected children) have an increased incidence of osteoporotic fractures, suggesting the existence of a skeleton-restricted haploinsufficient phenotype. In contrast, carriers of the autosomal dominant HBM trait show very high spinal bone mineral density (Z(BMD)=5.54 for affected individuals of the kindred (n=12), compared to 0.41 for unaffected individuals (n=16) (2)) without other clinical features.
This fall has seen the genetic origin of these two diseases identified. As Johnson et al. suggested in 1997, it turned out that they are allelic disorders, resulting from different mutations of the same gene. In the November 16th issue of Cell, the Osteoporosis-Pseudoglioma Syndrome Collaborative Group, led by Matt Warman (Case Western Reserve University School of Medicine) shows that OPPG patients harbor inactivating mutations in the LRP5 gene (), and that heterozygous carriers of the mutations indeed have reduced bone mass. In the meantime, a study by a collaborative effort between Genome Therapeutics Corporation and the Osteoporosis Research Center at Creighton University reports that an activating mutation (Gly171Val) in LRP5 is responsible for the high bone mass syndrome (). That the same gene was responsible for such opposite phenotypes did not come as a surprise. In fact this case can be classified as a textbook example where absence of a factor causes a loss of function (translating to a loss of bone) and constitutive activation of this same factor induces a gain of function (translating to a gain of bone). The real surprise came from the identity of the factor in question.
LRP5 stands for low-density lipoprotein receptor-related protein 5. It is a single pass membrane receptor whose extracellular domain contains four modules consisting of six YWTD repeats followed by an epidermal growth factor (EGF)-like motif and a LDLR-like ligand-binding domain (). A recent study has shown that LRP5 is involved in the Wnt canonical signaling pathway (). Wnt are secreted factors playing critical roles early during development, for instance controlling mesoderm induction, patterning, cell fate determination and morphogenesis (). A role for these proteins beyond development has never been demonstrated. The 1st and 2nd most N-terminal modules of LRP5 mediate its interaction with the Wnt-frizzled seven-transmembrane ligand-receptor complex. This association results in inhibition of β-catenin phosphorylation by GSK3-β as part of a multi-protein cytoplasmic complex that includes Disheveled, casein kinase I and the scaffolding protein Axin. GSK-3 phosphorylation of β-catenin facilitates its ubiquitination, targeting it for degradation by proteasomes. Therefore, inhibition of β-catenin phosphorylation prevents its degradation and by raising its cytoplasmic level increases its translocation to the nucleus. Nuclear β-catenin then interacts with TCF/LEF transcription factors to activate gene expression. Along with another receptor, termed LRP6, LRP5 defines a subclass of LDL-receptor related proteins homologous to Drosophila arrow, a co-receptor of the Wnt homologue wingless. In Drosophila, inactivation of arrow results in a wingless-like phenotype (). Likewise, inactivation of LRP6 in mice reproduces many developmental abnormalities observed as the result of deficiency in various Wnt proteins (). However, LRP6 but not LRP5 overexpression can induce dorsal axis duplication in Xenopus, suggesting that these receptors control distinct function (). The present findings that LRP5 loss of function does not recapitulate any known feature of a Wnt deficiency reinforce this hypothesis. They also suggest that LRP5 deficiency does not cause developmental defects, therefore extending the importance of the Wnt pathway to the regulation of perinatal and postnatal functions.
Both studies show that LRP5 is expressed in osteoblasts, although at a very low level. In addition, Matt Warman's group showed that LRP5 but also LRP6 expression is stimulated by BMP2 treatment of the ST2 marrow stromal cell line. In the same cells, Wnt 1, Wnt2 and Wnt3a but not Wnt4 and Wnt5a could induce the expression of alkaline phosphatase (ALP). However, LRP5 overexpression did not enhance Wnt3a induction of ALP, whereas a gain-of-function mutation in this gene does induce the HBM phenotype in humans. This suggests that other Wnts, or even other ligands, may be involved in signaling through LRP5 and that Wnt-induced ALP stimulation may reflect only one aspect of their activity. A more mechanistic understanding of the function of the Wnt/LRP5/β-catenin signaling pathway in osteoblast biology will have to await the analysis of mouse models reproducing for instance the LRP5 gain-of-function and loss-of-function mutations found in affected patients. Considering the preliminary data presented at this year's ASBMR this should be in a very short wait. Indeed, M. Johnson (Creighton University) mentioned during a satellite symposium that their group has overexpressed the HBM mutation in mice and that these mice exhibit increased bone mass and biomechanical strength. Likewise, the existence of LRP5-deficient mice was reported during an oral presentation by L. Chang and G. Karsenty's laboratories (Baylor College of Medicine).
This work is licensed under a
Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License.