Fig 1.
Phenotypic features of KYLB mice.
(A) Radiography of adult (8 week old) male KYLB and unaffected male littermate. The KYLB mouse has kyphosis of the thoraco-lumbar spine. (B-D) Alcian blue (cartilage) and Alizarin red (bone) staining of young (4 week old) male KYLB and male unaffected mice. KYLB mice had longer bodies (B), vertebrae (C), and lengthened long bones (tibia shown) (D), when compared to unaffected littermates.
Fig 2.
Mapping of Kylb locus and identification of Npr3 mutation.
(A) The Kylb locus, which originated in a BALB/c ENU-mutagenised male mouse and was hence inherited with the BALB/c alleles, was mapped to a 5.5Mb region flanked by the SNPs rs13459145 and rs13482436 on chromosome 15A1 containing 51 genes, including the most likely candidate, Npr3. (B) DNA sequence analysis of the Npr3 gene revealed a T to A transversion at codon 209 in exon 1, altering the wild-type (WT) sequence TAT, encoding tyrosine (Tyr), to the mutant (m) sequence AAT, encoding asparagine (Asn). The mutation introduced a Tsp509I restriction endonuclease site (/AATT) and this was used to confirm the mutation. (C) PCR primers specific for exon 1 were designed to amplify a 325bp product. Tsp509I did not cleave the WT PCR product but did cleave the mutant PCR product to yield 262bp and 63bp products, as shown by agarose gel electrophoresis of the digested Tsp509I PCR products. KYLB mice were homozygous for the mutant 262bp and 63bp (not shown) products (Kylb/Kylb), whilst unaffected mice were either homozygous for the 325bp product (WT, +/+) or heterozygous for the wild-type and mutant products (Kylb/+). (D) Protein sequence alignment (CLUSTALW) of NPR3 from 4 species revealed that the Tyr (Y) residue is evolutionarily conserved in the NPR3 homologs in mouse, human, monkey and zebrafish. Furthermore, the Asn209 (N) mutant (asterisk) generates a consensus N-linked glycosylation site (N-F-T) (indicated by horizontal line).
Fig 3.
Expression and localization of wild-type and His168Asn, Tyr209Asn and Ile384Phe NPR3 proteins in COS-7 cells.
(A) Western blot analysis of cell lysates from COS-7 cells transfected with either the wild-type (WT) or mutant (Asn168, Asn209, or Phe384) EGFP-tagged constructs. The WT and mutant Phe384 NPR3 proteins had molecular weights of 87kDa, whilst the Asn168 and Asn209 NPR3 mutants had a higher molecular weight of 89kDa, consistent with glycosylation at the mutant Asn168 and Asn209 residues. (B) Treatment with Endo H revealed some undigested WT NPR3 protein products whereas all the mutant NPR3 proteins were Endo H-sensitive, suggesting retention in the endoplasmic reticulum (ER). Tubulin was used as a loading control. (C-D) COS-7 cells transiently expressing EGFP-tagged WT or mutant (Asn168, Asn209 and Phe384) NPR3 proteins (green) were counterstained with anti-GM130 antibody (red) which is specific for the Golgi apparatus (C), or anti-PDI antibody (red) which is specific for the ER (D), and DAPI (blue) which is specific for nuclei. WT NPR3 localised to the plasma membrane with some expression in the Golgi apparatus, whereas all the mutant NPR3 proteins co-localised with the ER marker (orange). Scale bars = 20μm.
Table 1.
Body weight, length and DXA analysis of 12 week old mice.
Fig 4.
Histomorphometric analysis of vertebral and tibial growth plates.
Alcian blue and van Gieson staining of lumbar vertebrae (A, B) and tibiae (C, D) from 1 day old male mice showing reduced bone matrix (BM) formation (red stain) between intervertebral discs (IVD) and distal tibia in Kylb/Kylb mice compared to wild-type (WT) littermates. White boxes delineate detailed views shown in (B) and (D) of growth plates, showing expansion of the hypertrophic zone (HZ) in Kylb/Kylb mice compared to WT littermates. The resting (RZ) and proliferative (PZ) zones in the Kylb/Kylb and WT mice appeared to be similar and not affected. Scale bars = 100μm (A); 50μm (B, D); 200μm (C).
Fig 5.
p38 MAPK signaling in vertebral growth plates.
Vertebral sections from 1 day old wild-type (WT) and Kylb/Kylb male mice were stained with antibodies against phosphorylated p38 MAPK and MKK3/MKK6. Vertebrae from Kylb/Kylb mice had increased staining in the resting (RZ), proliferative (PZ) and hypertrophic (HZ) zones, but not in the bone matrix (BM), when compared to vertebrae from WT littermates. Scale bars = 50μm.
Table 2.
Comparison of mouse models with kyphosis due to Npr3 mutations.
Fig 6.
Schematic model illustrating the role of NPR3 as a decoy receptor.
In wild-type (WT) chondrocytes, NPR3 binds CNP thereby promoting its internalization and degradation, and reducing its availability to bind to NPR2 to stimulate the MAPK signaling pathway, whose activation inhibits termination of hypertrophic chondrocytes (HC) and bone formation. In KYLB mice who have a homozygous Tyr209Asn NPR3 mutation, NPR3 is abnormally glycosylated and retained in the ER. The absence of NPR3, and its loss as a decoy receptor for CNP, results in a greater availability of CNP to bind to NPR2, leading to increased MAPK signaling and enhanced survival of hypertrophic chondrocytes, resulting in the lengthening of bones.