INTRODUCTION

The Wnt signalling pathway is involved in a wide range of processes, including bone development and homeostasis¹. In accordance with this, mutations have been identified in various components of the Wnt pathway that cause different musculoskeletal diseases². The canonical Wnt pathway begins with the formation of a heterotrimeric complex between a co-receptor, LRP5/6, a ligand, WNT, and a receptor, FZD, which produces an accumulation of β-catenin that, once in the nucleus will activate the transcription of numerous important target genes for bone¹. This activation is finely regulated by a series of extracellular inhibitors such as DKK1 and sclerostin that bind to LRP5/6, preventing the formation of the heterotrimeric complex. For DKK1 and sclerostin to exert their inhibitory activity, they must form another heterotrimeric complex with LRP5 and KREMEN1/2 or LRP4, respectively. Although in the case of DKK1 the presence of KREMEN does not seem to be necessary to carry out a correct inhibition, the presence of LRP4 is essential for the inhibitory function of sclerostin³,⁴.

LRP4 mutations have been described in humans which cause different diseases that affect not only bone mass, but also the regulation of the extremities and kidneys among other, depending on the position of where the mutation occurs². Specifically, mutations in the central cavity of the third β-propeller cause sclerosteosis, characterized by variable syndactyly and progressive bone overgrowth, particularly severe in the facial skeleton and skull. These patients present an increase in the Wnt pathway in osteoblasts, generating an increase in bone formation⁴-⁶. In addition to these mutations, in 2017, Merello et al.⁷ described a mutation in the second β-propeller domain (p.Thr851Arg) that cosegregated with the phenotype in a family with type I Chiari malformation (CMI). This is a malformation of the central nervous system characterized by a caudal displacement of the cerebellar tonsils, which generates varied symptoms both in onset and in severity (Orphanet, https://www.orpha.net/consor/cgi-bin/ index.php). Although some patients with CMI may be asymptomatic, others can present with a suboccipital headache and neck pain, among others. It is interesting to note that this malformation was described in a patient who presented a phenotype of high bone mass (HBM) due to gain-offunction mutations in LRP5, thus suggesting a possible relationship between CMI, the HBM phenotype and Wnt pathway⁸. Taking into account the role of LRP4 on the Wnt pathway and the important role that this pathway has in determining bone mineral density (BMD) and in the development of the skull, we hypothesized that mutations in LRP4 could be the cause of the HBM in women with this phenotype or of causing disease in patients with CMI. In this study, we have carried out a re-sequencing of the LRP4 gene in a cohort of 10 women with the HBM phenotype and in a cohort of 12 patients with CMI.

MATERIAL AND METHODS

LRP4 re-sequencing

Genomic DNA from CMI cases and their relatives and from HBM women was isolated. women was isolated from peripheral blood leukocytes using the Wizard^® Genomic DNA Purification Kit (Promega, Madison, Wisconsin, USA), according to with the manufacturer's instructions. Regarding the re-sequencing, specifically, we have amplified the exons that code for the mutations that cause Cenani-Lenz syndrome (p.D137N, p.C160Y, p. D449N, p. T461P, p.L473F, p. D529N, p.L953P, p.C1017R, p.R1277H, p.E1233K) and the third β-propeller domain where the mutations causing myasthenic syndrome and sclerosteosis are located. Amplification of each of these fragments was done by polymerase chain reaction (PCR) using GoTaq Flexi DNA polymerase (Promega). The PCR fragments were analyzed by agarose gel electrophoresis method and their purification was done on MultiScreen™ Vacuum Manifold 96-well plates (Merck Millipore). The purified PCR products were sequenced using the Sanger method in the genomics service of the CCiTUB (Genómica, Parc Científic, Barcelona, Spain). The labeling kit used was BigDye™ Terminator v3.1 Cycle Sequencing Kit (ThermoFisher), detection and electrophoresis were carried out on the 3730 Genetic Analyzer and 3730xl Genetic Analyzer (Thermo-Fisher) automatic capillary sequencers models. The design of the primers was based on the consensus sequence ENSG00000134569 (LRP4; GRCh37.p13). Their sequence is presented in table 1.

Table 1.  Primers used in LRP4 re-sequencing 

RESULTS

Re-sequencing of LRP4 in women with HBM phenotype and in patients with CMI

In the LRP4 re-sequencing have identified 12 variants, of which one was not previously described (c.3364+16A>C; table 2). This variant has been found in heterozygosity in the HBM2 woman who presents a Zscore sum (lumbar spine and femoral neck) of 4.6. Six of the 12 variants identified are found both in the cohort of women with the HBM phenotype and in the cohort of patients with CMI, with a similar frequency in both cohorts to that of the European population. Furthermore, we have found 4 variants only present in the HBM cohort and 2 variants only present in the CMI cohort. All the variants described, except rs558515201 and the variant c.3364+16A>C, appear as eQTL of different genes and tissues in the GTEx database. Specifically, the rs17790156, rs61898529, rs2306028, rs964551 and rs2306032 variants are LRP4 eQTL.

Table 2.  Variants found in LRP4 re-sequencing in women with HBM phenotype and in patients with CMI. The genotypes in the first column are indicated by the nucleotides of the coding strand of the LRP4 gene, which is the reverse of that which is used in a standard way to indicate genomic SNVs. For this reason, for example, the A allele of c.431-12G> A is equivalent to the T allele of rs139371503, and the same for the other variants 

MAF: minority allele frequency; EUR: non-Finish European population of gnomAD v2.1.1.; HBM: high bone mass phenotype; CMI: Chiari type I mal-formation; dup: dup: duplication; M: missense variant; I: intronic variant; S: synonymous variant; eQTL: described as eQTL in Gtex; T: tolerated by SIFT; B: benign by Polyphen-2.

DISCUSSION

Different studies have highlighted the role of the LRP4 gene in determining BMD, as mutations in it generate a phenotype of bone overgrowth. Furthermore, it has also been associated with CMI in a report on whole exome sequencing⁷. In the study presented here, we have re-sequenced the regions of the gene that contain mutations associated with different conditions. In it, we have only found a missense mutation in LRP4 in a patient with CMI that does not co-aggregate with the phenotype in the family, thus ruling out its ability to cause the disease (Figure 1). It is interesting to note that we have found a variant not previously described (c.3364+16A>C) and the variant rs558515201, which has a very low frequency in the European population (MAF=0.00006480), in heterozygosity in two women with HBM. Furthermore, the rs3751097 and rs540384558 variants, present in both HBM and CMI patients, are found in a cis regulatory element categorized as a distal enhancer-like signature by ENCODE. On the other hand, contrary to expectations, we found in both cohorts a slightly lesser frequency for the minor allele of the rs2306032 variant which has been defined as a protector in whole genome association study with total BMD¹⁰.

Figure 1.  Pedigree of the family of the CMI CH10 patient. NA: no data on the parents. Genotypes are indicated with the nucleotides of the coding strand of the LRP4 gene, which is the reverse of that used in a standard way to indicate genomic SNVs. Therefore, allele C is equivalent to allele G of rs72897663 

The low number of patients must be taken into account as a limitation of our study. This could explain both the absence of causal variants in the LRP4 gene in these two cohorts and the results contrary to those expected in allele frequency in the SNP rs2306032. Thus, increasing the size of the cohorts would provide us with a deeper understanding of the role of LRP4 on these phenotypes.

In conclusion, although LRP4 is undoubtedly an important gene for bone biology, we have not found any variant that could explain the HBM or CMI phenotype. Despite these negative results, it is interesting to consider LRP4 in the selection of candidate genes that may explain HBM or CMI-causing phenotypes.