Editors: Norio Fujiki, M.D. & Darryl R.J. Macer, Ph.D.
Director, Dept. of Human Genetics, Nagasaki University School of Medicine, JAPAN
First I would like to define genomic imprinting. Genomic imprinting is a newly observed, genetically important mechanism of gene expression in mammals (1). Through the mechanism, one of the parental genes is marked to reduce its expression in offspring. In other words, each allele in a child is different in expression depending on its parental derivation. Until this was discovered it was believed that according to Mendelian law, alleles inherited from mother and father were identical. But, there are some genes which are not so. We do not know the cause yet, but we can see the effects.
Supportive evidence for the imprinting in the human genome has been provided by the observation of abnormal phenotypes (growth deficiency or over-growth) in persons with uniparental disomy (UPD) (2-4). UPD is defined as an event where both of the homologous chromosomes are derived from one parent (Figure 1) (5). The expression of an allele on UPD chromosomes of such persons has been thought to be suppressed (imprinted), and the cause of their abnormal growth might be due to the lack of an active allele from the counter parent. There are two types of UPD, as indicated.
Figure 2: Imprinting map of the mouse with probable positions of human imprinting disorders (Cattanach 1991).
Figure 3: Partial pedigree of Igf-2 mutant mice (DeChiara et al. 1991).
In mice, there is a model of a human disease, Igf-2 deficiency. The parental inheritance of this disease is represented in Figure 3. The pedigree demonstrated that the Igf2 gene is imprinted through the mother. This is a typical example of maternal imprinting.
Now I would like to discuss clinical subjects. Several human chromosomes or chromosomal regions are supposed to be involved in the imprinting, and candidate genetic diseases of which occurences are related to the underlying imprinting mechanism include Beckwith-Wiedemann syndrome (the presumed locus 11p15.5), Wilms tumour (11p15.5), Prader-Willi syndrome (15q11-q12), Angelman syndrome (15q11-q12), Huntington's chorea (4pter-p16.3), myotonic dystrophy (19q13.2-q13.3), neurofibromatosis type I (17q11.2), and fragile X syndrome (Xq27.3) (6). Some examples are shown in Table 1. I will discuss some of these.
Cytogenetic analysis revealed that 60-70% of patients have a 15q11-q12 deletion. The deletion can be confirmed at the molecular level by Southern blot analysis using several chromosome 15-linked marker DNAs as probes (7), as shown in Figures 4 and 5. The common deleted region among the patients studied is D15S9. Furthermore, maternal uniparental disomy (UPD) is sometimes observed in karyotypically normal PWS patients. A proposed model in Figure 6 is an explanation for these findings. The presumed PWS locus is maternally imprinted.
Figure 5: Parental origin of chromosome 15 in PWS patients.
Southern blots (upper) show the lack of a 17.5kb pIR10/ScaI allele that should have been derived from her father. Dinucleotide repeat polymorphisms (lower left) show that a patient (P) in a family is homozygous for a 72-base allele of the cardiac muscle actin gene (locus, 15q), the father (F) is homozygous for an 88-base allele, and the mother (M) homozygous for the 72-base allele, indicating uniparental disomy in the patient. A two-copy density of each of the other markers on chromosome 15 in the patient was confirmed by densitometry on Southern blots (data not shown). Likewise, chromosomes 15 in the patient (lower right) are of biparental origin.
Figure 6: Proposed mechanism for the occurence of PWS and AS
Figure 7: DNA deletion and its segregation in an AS family (Saitoh et al. 1992).
Southern blots of XbaI cut DNA (a) and MspI cut DNA (b), hybridized to p28§3-H3 (GABRB3), and those of HindIII cut DNA (c) hybridized to p28§3-H3 (2.5kb) and p21-4U (3.2kb) (located at 21q11.2 or 21q21.2).
This disease is associated with exomphalos, microglasia, gigantism and hypoglycemia in infancy, and susceptibility to Wilms tumour. There are four classes of the disease: (1) familial cases, (2) sporadic cases with partial trisomy for 11p15.5, (3) sporadic cases with a balanced translocation involving 11p15, and (4) sporadic cases with a normal karyotype. The disorder in Class 1 is of maternal origin, while the origin of an additional segment in Class 2 is paternal and that of the translocation in Class 3 is maternal. A part of Class 4 cases has paternal UPD of chromosome 11. These complex findings may be explained by a proposed mechanism (10) shown in Figure 8.
As shown in Table 1, in the human genome there are several candidate disorders of which occurrence is related to the imprinting. Furthermore, the human chromosomes, other than those mentioned above, involving UPD with parent-of-origin include chromosome 4 of maternal origin (4mat), 6pat, 7mat, 14pat, 14mat, 22mat, and X&Ypat (11). These UPD individuals often have abnormal clinical pictures, especially a developmental delay. It is thus expected that with the advancement of the human genome project, many other imprinted genes/regions or their related disorders will be identified.
1. Cattanach, B.M. "Chromosome imprinting and its significance for mammalian development", pp. 41-71 in K.E. Davies & S.M. Tilghman, eds., Gene Expression and Its Control (Cold Spring Harbor Laboratory Press 1991).
2. Nicholls, R.D. et al. (1989) "Genetic imprinting suggested by maternal heterodisomy in non-deletion Prader-Willi syndrome", Nature 342: 281-285.
3. Malcolm, S. et al. (1991) "Uniparental disomy in Angelman's syndrome", Lancet 337: 694-7.
4. Henry, I. et al. (1991) "Uniparental paternal disomy in a genetic cancer-predisposing syndrome", Nature 351: 665-7.
5. Engel, E. (1980) "A new genetic concept: Uniparental disomy and its potential effect, isodisomy", Amer. J. Medical Genetics 6: 137-43.
6. Hall, J. (1990) "Genomic imprinting: Review and relevance to human diseases", Amer. J. Human Genetics 46: 857-73.
7. Hamabe, J. et al. (1991a) "Molecular study of Prader-Willi syndrome: deletion, RFLP, and phenotype analyses on 50 patients", Amer. J. Medical Genetics 41: 54-63.
8. Hamabe, J. et al. (1991) "DNA deletion and its parental origin in Angelman syndrome patients", Amer. J. Medical Genetics 41: 64-8.
9. Saitoh, S. et al. (1992) "Famalial Angelman syndrome caused by imprinted submicroscopic deletion encompassing GABA receptor beta 3-subunit gene", Lancet339: 366-7.
10. Mannens, M. et al. (In press) "Parental imprinting of human chromosome region 11p15.4-pter involved in the Beckwith-Wiedemann syndrome and various human neoplasia", Cell.
11. FrŽzal, J. & Schinzel, A. (1991) "Report of the committee on clinical disorders, chromosome aberrations and uniparental disomy. Human Gene Mapping 11. London Conference (1991). Eleventh International Workshop on Human Gene Mapping", Cytogenet. Cell Genetics 58: 986-1052.