A single nucleotide mutation in Nppc is associated with a long bone abnormality in lbab mice

Background The long bone abnormality (lbab) mouse is a new autosomal recessive mutant characterized by overall smaller body size with proportionate dwarfing of all organs and shorter long bones. Previous linkage analysis has located the lbab mutation on chromosome 1 between the markers D1Mit9 and D1Mit488. Results A genome-based positional approach was used to identify a mutation associated with lbab disease. A total of 122 genes and expressed sequence tags at the lbab region were screened for possible mutation by using genomic DNA from lbabl/lbab, lbab/+, and +/+ B6 mice and high throughput temperature gradient capillary electrophoresis. A sequence difference was identified in one of the amplicons of gene Nppc between lbab/lbab and +/+ mice. One-step reverse transcriptase polymerase chain reaction was performed to validate the difference of Nppc in different types of mice at the mRNA level. The mutation of Nppc was unique in lbab/lbab mice among multiple mouse inbred strains. The mutation of Nppc is co-segregated with lbab disease in 200 progenies produced from heterozygous lbab/+ parents. Conclusion A single nucleotide mutation of Nppc is associated with dwarfism in lbab/lbab mice. Current genome information and technology allow us to efficiently identify single nucleotide mutations from roughly mapped disease loci. The lbab mouse is a useful model for hereditary human achondroplasia.

(TJL) [4]. Homozygous mutants exhibit proportionate dwarfing of all organs and shorter long bones. The mutation has been transferred to the C57BL/6J strain to improve reproduction. The genetic locus responsible for the phenotype has previously been mapped to chromosome 1 (Chr 1) between markers D1Mit9 and D1Mit488 (53.5 cM) at TJL [4], but the responsible gene and the nature of mutation remained unclear.
Disease gene hunting has always been time-consuming and labor-intensive. For successful map-based cloning, a complicated fine mapping of a major locus is generally essential [5]. Recently, we developed an alternative, sequence-based, positional candidate cloning approach to bypass this bottleneck of cloning, and we have successfully identified several mutated genes in different mouse spontaneous mutants by applying this strategy [6,7]. Our strategy takes advantage of the availability of comprehensive murine sequence databases, polymerase chain reaction (PCR), high throughout PCR product analysis, and sequencing technologies to speed up the process of disease-related gene hunting. Interestingly, we identified a nucleotide mutation (C→G transversion) in gene Nppc in the lbab mice. Herein, we describe the detailed process of our cloning and validation.

Results
Body growth of lbab/lbab mice All lbab/lbab mice housed at the University of Tennessee Health Science Center died before 7 days of age with remarkable changes in organ weight and body size (Fig. 1, Table 1). No detectable differences in survival rate or body size were noted between the lbab/+ and +/+ mice.

Target region of the mutation in lbab locus
Previous genetic analysis showed that the lbab mutation is located on mouse Chr 1 and is flanked by molecular markers D1Mit9 and D1Mit488 [4]. According to the Ensembl database, D1Mit488 is located between 91902920-91903043 bps. However, there is no physical position for D1Mit9 in the database, although we know from TJL database that it is positioned at 53.5 cM. From TJL's mouse genome informatics database, we found 13 molecular markers at the 53.5 cM position. The positions of most of these [8] markers are near 84 Mb [9]. Accordingly, we decided to start our investigation in the region between 83 and 90 Mb ( Fig. 2A). Genomic sequences within this region are complete in the Ensembl database. There are a total of 122 transcripts, 70 of which represent genes and 52 represent expressed sequence tags (ESTs) ( Table 2).

Initial screening of the targeted region
Because lbab mice are bred on a C57BL/6J (B6) background, we assumed that the majority of the background genome carrying the lbab mutation was from the B6 strain. Therefore, we isolated genomic DNA from both B6 and lbab/lbab mice. We designed 528 pairs of primers flanking first and last exons of 122 candidate transcripts by using Primer3 software [10]. We obtained the primers commercially (Illumina, San Diego, CA) and conducted PCR amplification with genomic DNAs from lbab/lbab and B6 mice, which were regarded as normal controls. The PCR products were then analyzed for the presence of sequence differences between lbab/lbab and normal mice by using the RevealTM system (SpectruMedix LLC, PA). We found variations in 25 PCR products between normal B6 and lbab/lbab mice (Table 3).
We speculated that the identified sequence variations were likely to arise from the close linkage between the mutated gene and nearby sequences in the parental PL/J strain, even though the background of lbab mice was mainly on the B6 strain. To determine if this were the case, we isolated DNA from PL/J mice obtained from TJL and amplified the 25 variable fragments by using the same panel of primers. By comparing those DNA fragments with those from lbab/lbab mice, we did not find sequence differences between DNA products from lbab/lbab and PL/ J mice (data not shown), suggesting that those 25 variations represent polymorphisms between PL/J and B6.

Detecting the mutation in lbab locus in a more focused region
Because of the recognition of sequence polymorphisms between PL/J and B6 mice in the targeted region, we made two changes in our follow-up screening. First, we switched our controls from B6 to PL/J mice. Using another panel of primers (n = 240 pairs) for PCR amplification of every exon of all candidate genes, we identified only one DNA fragment from lbab/lbab mice that was different from PL/J mice DNA. This fragment was from exon 2 of the gene ENSMUSG00000026241, representing the gene for natriuretic peptide precursor C (Nppc). The same pair of primers for this fragment was used for further genotyping as indicated in material and methods.
Because we used PL/J mice as controls for this cycle of screening, there was another concern that the sequence variation might be a polymorphism derived from the B6 strain. To address this issue, we used PCR amplification of genomic DNA from B6 mice and compared results to the PCR products from PL/J and lbab/lbab mice. The data showed that the amplified DNA fragment from lbab/lbab mice was different from both B6 and PL/J mice, while the fragments from PL/J and B6 were the same (Fig. 2B). To find out which nucleotide(s) was different between lbab/ lbab and PL/J controls, we sequenced genomic DNA fragments from PL/J and lbab/lbab mice. The data revealed a C→G change from PL/J to lbab/lbab mice (Fig. 3A, left panel).
To confirm the same difference at the cDNA level, we performed reverse transcriptase PCR (RT-PCR) on total RNAs from lbab/lbab, lbab/+, and +/+ mice by using primers that covered the mRNA sequence from exon 2 to exon 3 of the Nppc. The resultant RT-PCR products were sequenced using the SpectruMedix system, and the same C→G transversion from PL/J to lbab/lbab mice was found (Fig. 3A, right panel) To evaluate the potential consequence of this point mutation in Nppc, we examined the translated amino acid sequence of Nppc. We found that this transversion predicts the substitution of arginine (R) for glycine (G) in a conserved domain of Nppc protein (Fig. 3B).

Confirmation of mutation
We conducted further experiments to confirm the single nucleotide change in the Nppc gene. Based on our initial screening, we think that it is very likely that the C→G change in exon 2 of the Nppc gene is causally related to the phenotypes in lbab/lbab mice. Because the lbab mutation arose from PL/J, theoretically, there should be no difference in the Nppc sequence between lbab and PL/J mice except the mutation. In addition, this single nucleotide replacement is the only change between homozygous lbab/lbab and homozygous normal PL/J mice in 122 transcripts. However, we could not completely rule out the possibility that we may have missed other mutations/polymorphisms. Moreover, one could question whether this is a spontaneous polymorphism in mouse strains or a random mutation that arose after the lbab was separated from the PL/J. Therefore, we carried out two more experiments to further ensure the association between the mutation and the disease. First, we examined sequence polymorphism in exon 2 of Nppc from nine other inbred strains (Fig. 2C). As shown in Fig. 2D, each such mixture showed multiple bands, indicating the difference between lbab/ lbab and those strains. Second, we bred 200 mice from heterozygous lbab/+ parents to evaluate allele frequency in relation to the phenotype. We genotyped every offspring by using the original pair of primers that flank exon 2 of the Nppc gene. From those progeny, we found 14 lbab/lbab mice that exhibited a genotype of homozygous G/G nucleotide and a phenotype of lbab mice, while the remaining progeny had 74 homozygous C/C and 112 heterozygous C/G genotypes, all with a normal phenotype. Taken together, our data indicate that the C→G transversion in exon 2 of the Nppc gene is associated with the phenotypes observed in lbab mice.
Phenotypic assay of normal and lbab mice Figure 1 Phenotypic assay of normal and lbab mice. A wholebody microCT scanned image of a pair of lbab/lbab and +/+ B6 mice in the same litter at the age of 6 days. A: lbab/lbab mouse, B: +/+ mouse.

Discussion
For the first time, we have identified a single nucleotide mutation by high throughput screening of a large genome region without fine mapping. The initial mapping at TJL was conducted with only 27 F 2 animals. Linkage of lbab was first detected on Chr 1 with D1Mit231 and D1Mit9 by using the pooled sample. DNA samples were then typed for the individual 96 animals with these two markers and three additional Chr 1 markers [4]. By the standard strategy of classical positional cloning, the lbab locus could be further mapped. However, with the availability of mouse genome information and a tested protocol for high throughput screening of mutations [6,7], we directly searched genes based on the map from TJL. With the success of finding this mutation and others [6,7], we feel con-fident that we no longer need fine mapping for most mutations.
In this study, several lines of evidence indicate that a single nucleotide mutation of Nppc is associated with the lbab phenotype. First, Nppc is located within the genetic region of the lbab locus. Second, the Nppc mutation was the only defect detected among candidate genes and ESTs within the lbab locus from lbab mice. Because the lbab mutation was transferred from the PL/J strain to the B6 inbred strain, we evaluated the possibility of close linkages of nearby sequences from the PL/J mice by screening any sequence difference near the mutation area and later crossing with PL/J mice. There were no other differences between lbab/lbab mice and their two parental strains, so PCR products from each of those strains were mixed separately with that of lbab/lbab mice. Each mixture showed multiple bands of signal, indicating the difference in their DNA sequences. The X-axis represents relative size of the PCR products. The Y-axis represents relative strength of signal or the amount of the PCR products.

Schematic of the mutation identification in lbab mice
the possibility of other mutation involvement was ruled out. Third, cDNA sequence results agreed with the genomic DNA data. Last, we showed that the Nppc genotype is unique in lbab/lbab mice compared with nine other inbred strains, and the G/G Nppc genotype was closely associated with the phenotype in lbab mice.
Recently, several transgenic and knockout mouse studies have demonstrated that Nppc is critical in the prevention and rescue of achondroplasia [11,12]. A recent gene knockout study done by Chusho et al. [11] indicated that Nppc null mice of 129/Sv background showed severe dwarfism and early death. The lbab mice have a phenotype similar to Nppc knockout mice with two exceptions. First of all, the lbab mice develop an overall smaller body size. Second, the mutants exhibit proportionate dwarfing of all organs with the possible exception of the male reproductive tract, which appears extremely small [4]. However, much precise information may be obtained by a direct comparison between Nppc null mice and lbab mice. There may still be some difference between them because of the difference not only in the nature of the mutations but also in the genome backgrounds of those two models. Accordingly, we speculate that the identified point mutation of Nppc in lbab mice belongs to a loss-of-function mutation. As a key positive regulator of endochondral bone formation, Nppc seems to express its activity mainly through natriuretic peptide receptor 2 (Npr2) [11,[13][14][15]. On the other hand, a recent study indicated that Nppc counteracts the activities of fibroblast growth factor signalling, which is a major negative regulatory pathway for long bone development, in both direct and indirect ways [16].
Murine Nppc is structurally similar to that of other species. The affected Arg at codon 117 on the Nppc domain is highly conserved among all members of the natriuretic peptide system and different species (Fig. 3B). The mechanisms for regulating Nppc expression are currently  unknown. Importantly, the mutated nucleotide is also located in the common biologically active COOH-terminal 22 amino acid residue area, suggesting the critical significance of this amino acid residue in the functioning of these ligands during skeletal development.
Nppc was not in the list of candidate genes for the allelism test. According to the information on TJL webpage [4], allelisms tested were brachymorphic (bm) [17] with a ratio of disease/total of 0/42 progeny born, achondroplasia (cn) [18] with a ratio of 0/61 progeny born, osteochondrodystrophy (ocd) [19] with a ratio of 0/19 (4 unclassifiable) progeny born, and small (sml) [20] with a ratio of 0/59 progeny born. The first three loci are known to be located on Chr 19 [17], 4 [18], and 19 [19], respectively. The last one, sml, is either on Chr 6 or unknown [20]. Had Nppc been considered as a candidate gene, our initial screening would have been simpler, although we feel that some work is needed to exclude mutations in the nearby genes. In addition to the known function of Nppc, the fact that there is no other mutation in nearby genes in the lbab region is supporting evidence for the potential cause of the lbab phenotype by the single nucleotide mutation in Nppc.
Identifying the Nppc mutation in lbab mice provides useful information for human achondroplasia studies. It also demonstrates that, while candidate genes should be care-Schematic of the sequence changes of Nppc gene in lbab mice fully examined based on gene function, it is feasible to identify mutated genes that are roughly mapped by linkage analysis by sequence-based positional candidate cloning strategy. We speculate that this strategy will be particularly useful for familial human diseases with small numbers of patients; in those cases, researchers usually have either a rough map or name/number of the chromosome of the disease locus. Furthermore, by using functional genomics and rodent models with spontaneous mutations yielding measurable phenotypes, we can rapidly identify mutational events in a cost-effective manner.

Conclusion
In the present study, a sequence-based positional candidate cloning approach was applied to identify a gene mutation in lbab mice with abnormal endochondral ossification. Our results suggest that a single nucleotide mutation in gene Nppc is likely to be the causative factor and that the lbab mouse may be a useful model for human achondroplasia studies. some 19 encompassing the muscle-deficient osteochondrodystrophy (mdf-ocd) region. Mamm Genome 1998, 9:390-391. 20. Lundberg P, Welander P, Openshaw H, Nalbandian C, Edwards C, Moldawer L, Cantin E: A locus on mouse chromosome 6 that determines resistance to herpes simplex virus also influences reactivation, while an unlinked locus augments resistance of female mice. J Virol 2003, 77:11661-11673.