Identity-by-descent refines mapping of candidate regions for preaxial polydactyly in a large Chinese pedigree

Preaxial polydactyly (PPD) is congenital hand malformation characterized by the duplication of digit. Herein, we scan the genome-wide SNPs for a large Chinese family with PPD-II/III. We employ the refined IBD algorithm to identify the identity-by-decent (IBD) segments and compare the frequency among the patients and normal relatives. A total of 72 markers of 0.01 percentile of the permutation are identified as the peak signals. Among of them, 57markers locate on chromosome 7q36 which is associated with PPD. Further analyses refine the mapping of candidate region in chromosome 7q36 into two 380 Kb fragments within LMBR1 and SHH respectively. IBD approach is a suitable method for mapping cause gene of human disease. Target-enrichment sequencing as well as functional experiments are required to illustrate the pathogenic mechanisms for PPD in the future.

Duplication of ZRS is unclear how this contributes to ectopic expression [9].
The common PPD only involves in hands/feet. In extreme and rare cases, PPD occur both in hands and feet. To investigate the genetic basis, Li et al. adopted a candidate gene approach to genotype nine microsatellite markers of 7q36 chromosomal region in a Chinese family with PPD both in hands and feet. By linkage analysis and haplotype construction, they located the linkaged region spanning 1.7 Mb between D7S2465 and D7D2423 [18]. It includes the 450Kb candidate region previously identified by Henus [7]. Nevertheless, the other part of genome is not investigated yet. Herein, we genotyped genome-wide SNPs and employed the identity-by-descent (IBD) to refine the mapping of potential candidate loci for PPD in the same family.

Patients
This study has been approved by the internal review board of Kunming Institute of Zoology, Chinese Academy of Sciences (SMKX 2012013). The six-generation pedigree (including 21 patients and 24 normal relatives) involved in this study has been described previously in Li et al [18]. All patients show hexadactyly of hands and feet. They have been diagnosed by physical examination & X-ray and assigned as isolated PPD-II on hand and isolated PPD-III on feet according to Temtamy and McKusick's classification [19]. PPD shows autosomal dominant inheritance in this pedigree.

SNP array
We genotyped 900015 markers in 45 individual with HumanOmniZhongHua-8 BeadChip v1.0 (Illumina) . We exported the chip data in accordance with the reference sequence GRCh37 into PLINK format via GenomeStudio (Illumina). The markers on mitochondrial DNA and sex chromosomes were disregarded. We adopted a series of quality control strategies [20] by using PLINK 1.9 [21]. Two individuals with call rate < 90% were removed.
The SNPs with call rate < 90%, minor allele frequency < 1%, and deviation of Hardy-Weinberg equilibrium (P<1e-6) were excluded. After filtering, a total of 595534 autosomal SNPs for 43 individuals were utilized in subsequent analyses. The data have been deposited into Dryad (XXXXXX).

IBD detection
We used BEAGLE 4.0 [22] to phase and impute the genotype data referring to the pedigree information and the genetic map of HapMapⅡ [23]. We detected the IBD segment with the refined IBD in BEAGLE 4.1 [24]. The IBD segment length shorter than 1cM and the logarithm of odds (LOD) score under 3 were excluded before permutation [25]. The threshold of the genome-wide significance was set to the 0.05 percentile of the distribution of the permutation p-value.

Results
The length distribution of detected IBD segments approximates a Pareto distribution (Fig. S1).
The permutation result shows the significant segments distributing widely across genomes ( Fig. 1). When considering the top 0.01% outliers of signals, we find the peak signals of 72 SNPs, of which 57 markers located at 7q36 chromosomal region (Table S1). We map the markers into the IBD fragments including LMBR1 and SHH ( Table 1). The minimal IBD segments within LMBR1 and SHH are around 380 Kb, respectively (Table S2). The IBD segments are more frequently in patient-patient (ratio; percentage) than normal-normal (ratio; percentage) ( Table 2). We make annotation for the significant SNPs (Table S1). All the SNPs haven't been reported to be associated with PPD before.

Discussion
Our IBD analyses refine the mapping of the candidate regions for PPD into two ~380 Kb segments in 7q36 referring to LMBR1 and SHH genes, respectively (Table S2). The segment for LMBR1 includes three genes (i.e. LMBR1, NOM1, and RNF32) and lies within the 450Kb candidate region identified before [7]. Given that the intron 5 of LMBR1 performed as an enhancer for SHH playing an important role in the pathogenesis of PPD (Table 3). Li et al.
sequenced the exons of the five candidate genes and the intron 5 of LMBR1. But no candidate mutations were found [18]. The candidate mutations for PPD may locate in the other introns of LMBR1.
In addition to the segment of LMBR1, we also identified a segment of SHH. The SHH gene encodes sonic hedgehog, a secreted protein, which plays a key role in the limb development [26]. The ectopic expression of SHH in the anterior limb margin can cause PPD in mouse [27].
This has been described as well in humans [27], Hemingway cats [8] and chicken [28]. The candidate causal mutations are located in the intron 5 of LMBR1 [9]. In summary, we refine the mapping of the candidate regions for PPD based on high-density genomic SNPs. The potential candidate mutations are most likely to locate in LMBR1 and/or SHH gene. It is much improved compared with previous results [7,18]. Our study suggests that the IBD approach is a suitable method for mapping the cause genes of human diseases.

Moreover, as disruptions of topological chromatin domains can result in limb malformations
[31], more attention should be paid when studying PPD in the future on this aspect.

Competing interests
The authors declare no conflict of interest.

Ethics approval and consent to participate
This study has been approved by the internal review board of Kunming Institute of Zoology, Chinese Academy of Sciences (SMKX 2012013). The patients consent to participate in this study by signing a Consent Form allowing the use of biological samples and clinical data.

Availability of data and supporting materials section
If the paper is accepted the data will be deposited into Dryad (XXXXXX).

Funding
The research protocol of the study entitled "Identity-by-descent Refines Mapping of