Imprinted and Genotype-specific Expression of Genes at theIDDM2Locus in Pancreas and Leucocytes

P Vafiadis, ST Bennett, E Colle, R Grabs… - Journal of …, 1996 - Elsevier
P Vafiadis, ST Bennett, E Colle, R Grabs, CG Goodyer, C Polychronakos
Journal of autoimmunity, 1996Elsevier
One of the loci encoding susceptibility to insulin-dependent diabetes mellitus (IDDM)
isIDDM2, mapped to a variable number of tandem repeats (VNTR) polymorphism situated
596bp upstream of the insulin gene (INS). The shorter alleles (class I) predispose to IDDM,
while the longer class III alleles are protective. BesidesINS, it is possible that transcription
levels ofIGF2, the nearby gene encoding the insulin-like growth factor II, may be modulated
by allelic forms of the VNTR. In an effort to define the pathophysiologic mechanism of …
One of the loci encoding susceptibility to insulin-dependent diabetes mellitus (IDDM) isIDDM2, mapped to a variable number of tandem repeats (VNTR) polymorphism situated 596bp upstream of the insulin gene (INS). The shorter alleles (class I) predispose to IDDM, while the longer class III alleles are protective. BesidesINS, it is possible that transcription levels ofIGF2, the nearby gene encoding the insulin-like growth factor II, may be modulated by allelic forms of the VNTR. In an effort to define the pathophysiologic mechanism of theIDDM2effect, we examined the effect, incis, of VNTR genotype on steady-state mRNA levels ofINSin samples of human fetal pancreas, and ofIGF2in leucocytes of diabetic children. Relative levels of mRNA transcripts derived from each chromosome carrying a defined VNTR allele were measured by RT-PCR, taking advantage of transcribed polymorphisms at the 3′ untranslated region of each gene. In 10 samples of human fetal pancreas,INStranscripts from chromosomes carrying a class III VNTR were slightly but significantly (P=0.015) lower than those from class I (13% lower, 95% confidence limits 3–21%). In 10 leucocyte samples, mRNA from bothIGF2alleles was seen, indicating relaxation of the parental imprinting ofIGF2in these cells. However, this relaxation was incomplete as maternal allele mRNA was systematically at a lower level than paternal. The paternal/maternal ratio varied widely among individual subjects. Two of the most extreme cases, demonstrating almost complete repression of the maternal allele, were identical twins, suggesting that this variable relaxation of imprinting is genotype-dependent. However, this genotype-dependence cannot be accounted for by the maternal VNTR, as the mean ratios of paternal/maternalIGF2mRNA levels were not statistically different in individuals with a maternal VNTR of class I vs. class III (3.2±1.5 vs. 3.89±0.94). Thus, we present evidence that: (a) class III VNTR alleles are associated with lowerINSmRNA in fetal pancreas than class I alleles. The biologic importance of this difference remains to be determined; and (b) the variable relaxation ofIGF2imprinting seen in human leucocytes is not dependent on the presence of a class I vs. a class III VNTR.
Elsevier