Generation and phenotyping of two different murine models for chondrodysplasias caused by defects in proteoglycan biosynthesis.

Among the recent classification of genetic skeletal disorders, there is a cluster of diseases such as, Desbuquois dysplasia, diastrophic dysplasia that display common features including short stature, dysplasia of skeletal elements, and congenital joint dislocations. At biochemical level all these disorders are characterized by defects in proteoglycans (PGs) that represent one of the most important components of the cartilage extracellular matrix. alterations of PG synthesis, due to mutations in genes encoding for enzymes or proteins involved in the process, lead to the onset of genetic diseases affecting cartilage.The molecular basis of many skeletal disorders are poorly understood, thus mechanistic studies using an in vivo approach is necessary to investigate the role of the disease gene in the disorder.Among this scenario, this work has been focused, using in vivo models, on the study of two chondrodysplasias in which proteoglycan synthesis is impaired: Chondrodysplasia with joint dislocation gPAPP type and Desbuquois dysplasia type 1. Chondrodysplasia with joint dislocation gPAPP type is a recessive osteochondrodysplasia caused by mutations in IMPAD1 gene that encodes for a Golgi resident adenosine 3‟,5‟-bisphosphate phosphatase crucial for proteoglycan sulfation. Desbuquois dysplasia type 1 is a rare chondrodysplasia caused by mutations in the CANT1 gene encoding for calcium-activated nucleotidase 1, a Golgi protein that preferentially hydrolyzes UDP.The molecular knowledge of the two disorders is different; thus, in this thesis the two models have been used to pursue different objectives: I) the generation and characterization of a conditional knock-in mouse as model for the study of the chondrodysplasia with joint dislocations gPAPP type, II) the validation of a Cant1 knock-out mouse (dbqd mouse) as an animal model of Desbuquois dysplasia type 1 and the study of Cant1 role in PG biosynthesis. In this work, we have tested an innovative strategy called “the Cre-mediated genetic switch” in vivo, with the aim to generate a Impad1 conditional knock-in mouse for a missense mutation.The Cre-mediated genetic switch combines the ability of Cre recombinase to stably invert or excise a DNA fragment depending upon the orientation of flanking mutant loxP sites. Targeting constructs were generated in which the Impad1 exon 2 and an inverted exon 2* containing the point mutations, were flanked by mutant loxP sites in a head-to-head orientation. When the Cre recombinase is present, the DNA flanked by the mutant loxP sites is expected to be stably inverted leading to the activation of the mutated exon.The targeting vectors were used to generate heterozygous floxed mice in which inversion of the wild-type with the mutant exon has not occurred yet. To generate Impad1 knock-in mice, floxed animals were mated to a global Cre-deleter mouse strain for stable inversion and activation of the mutation.Unexpectedly the phenotype of homozygous Impad1 knock-in animals overlaps with the lethal phenotype described previously in Impad1 knock-out mice. Expression studies demonstrated that mutant mRNA underwent abnormal splicing leading to the synthesis of non-functional proteins. the skeletal phenotypes were not caused by the missense mutations, but by aberrant splicing.The dbqd mouse, an animal model of human Desbuquois dysplasia type I, was previously generated by our research group. The clinical phenotype of Cant1 knock-out mice showed the same typical features of patients with Desbuquois dysplasia type 1 confirming the dbqd mouse as an animal model of the human disorder. Biochemical studies highlighted the contribution of CANT1 in PG synthesis. GAG synthesis was reduced and chains were shorter and oversulfated