An imprinted gene(s) for diabetes?

• Published in: Nature genetics Vol. 9; no. 2; pp. 110 - 112
• Main Authors: Temple, I. Karen, James, Rowena S, Crolla, John A, Sitch, Fiona L, Jacobs, Patricia A, Howell, W. Martin, Betts, Peter, Baum, J. David, Shield, Julian P.H
• Format: Journal Article
• Language: English
• Published: United States 01.02.1995
• Subjects: Adolescent; chromosome 6; Chromosome Aberrations; Chromosomes, Human, Pair 6; diabetes mellitus; Diabetes Mellitus - genetics; Female; Genomic Imprinting; Humans; Infant; Infant, Newborn; INS gene; insulin; Male; man; ring
• ISSN: 1061-4036; 1546-1718
• DOI: 10.1038/ng0295-110

Transient neonatal diabetes mellitus (TNDM) classically occurs in growth retarded infants in the first few weeks of life. The condition is usually selflimiting and most patients recover by one year of age. Subsequent insulin-dependent diabetes may occur in later childhood. A failure of beta cell maturation has been proposed as the underlying defect although the real cause remains unknown. Individuals with TNDM do not appear to carry the common HLA susceptibility haplotypes associated with childhood diabetes nor evidence of autoimmune mediated disease, hence other processes must be at play. Most TNDM cases are sporadic but the disorder has been described in three male siblings born to unrelated parents, three of four offspring of an unaffected father by three different mothers, and in male twins, presumed to be monozygous and first cousins. These familial cases do not conform to a simple mendelian mode of inheritance. In the May 1994 issue of Nature Genetics, Haig presented growing evidence that insulin genes are imprinted. First, Giddings et al. have shown that only the paternal copies of both mouse insulin genes (ins1 and ins2) are expressed in the yolk sac of mouse embryos. Second, in some populations susceptibility to diabetes appears to be associated with paternal and not maternal transmission of a 4.1 kb DNA fragment of 11p15 which includes the human insulin (INS) gene. Most imprinted genes so far discovered have a developmental role and Haig proposed that TNDM, which is developmental disease of insulin production that improves in postnatal life, could potentially be due to an imprinted gene. We have evidence that this may well be the case in humans and infer the imprinted gene to be on chromosome 6. During a systematic search for uniparental disomy in a series of 32 individuals with supernumerary marker chromosomes, we identified patient A with TNDM. Chromosome analysis showed her to have a female karyotype with two cell lines; 13 out of 50 cells had a chromosome complement of 46,XX and the remaining 37 cells had a chromosome complement of 47,XX,+r(6). Paternal uniparental isodisomy (UPD) of chromosome 6 was shown by demonstrating that the proband had inherited only paternal alleles at loci outside the region of chromosome 6 represented by the ring. At all 16 informative loci, heterozygosity in the father was reduced to homozygosity in the proband. Maternal alleles were demonstrable only for chromosome 6p11-6p21 confirming that the small ring chromosome was of maternal origin.