Amnionless function is required for cubilin brush-border expression and intrinsic factor-cobalamin (vitamin B12) absorption in vivo.

Amnionless (AMN) and cubilin gene products appear to be essential functional subunits of an endocytic receptor called cubam. Mutation of either gene causes autosomal recessive Imerslund-Gräsbeck syndrome (I-GS, OMIM no. 261100) in humans, a disorder characterized by selective intestinal malabsorption of cobalamin (vitamin B12) and urinary loss of several specific low-molecular-weight proteins. Vital insight into the molecular pathology of I-GS has been obtained from studies of dogs with a similar syndrome. In this work, we show that I-GS segregates in a large canine kindred due to an in-frame deletion of 33 nucleotides in exon 10 of AMN. In a second, unrelated I-GS kindred, affected dogs exhibit a homozygous substitution in the AMN translation initiation codon. Studies in vivo demonstrated that both mutations abrogate AMN expression and block cubilin processing and targeting to the apical membrane. The essential features of AMN dysfunction observed in vivo are recapitulated in a heterologous cell-transfection system, thus validating the system for analysis of AMN-cubilin interactions. Characterization of canine AMN mutations that cause I-GS establishes the canine model as an ortholog of the human disorder well suited to studies of AMN function and coevolution with cubilin.

Canine Imerslund-Gräsbeck syndrome maps to a region orthologous to HSA14q.

Selective malabsorption of cobalamin (vitamin B(12)) accompanied by proteinuria, known as Imerslund-Gräsbeck syndrome or megaloblastic anemia 1 (I-GS, MGA1; OMIM 261100), is a rare autosomal recessive disorder. In Finnish kindreds, I-GS is caused by mutations in the cubilin gene ( CUBN), located on human Chromosome (Chr) 10. However, not all patients have CUBN mutations, and three distinct mutations in the amnionless gene, AMN, were very recently identified in patients from Norwegian and Israeli families. The present study demonstrates that in a large canine I-GS pedigree, the disease is genetically linked (peak multipoint LOD score 11.74) to a region on dog Chr 8 that exhibits conserved synteny with human Chr 14q. Multipoint analysis indicates that the canine disease gene lies in an interval between the echinoderm microtubule-associated, protein-like 1 ( EML1) gene and the telomere. A single critical recombinant further suggests that the disease gene is between markers in EML1 and the G protein-coupled receptor ( G2A) gene, defining an I-GS interval in the human genome that contains the AMN gene. Thus, these comparative-mapping data provide evidence that canine I-GS is a homologue of one form of the human disease and will provide a useful system for understanding the molecular mechanisms underlying the disease in humans.