H19, Insulin-like Growth Factors, and the Regulation of Embryonic Growth

Although growth hormone from the pituitary gland is critical for mammalian growth after birth, the proportional growth of the mammalian fetus is directed by a set of proteins called the insulin-like growth factors I and II (IGFI, IGFII). Both these molecules are capable of binding to the insulin receptor (IR). In addition, IGFI binds to its own receptor, IGF1R, and IGFII binds specifically to the mannose-6-phosphate receptor/IGF2R protein, which targets it for destruction in the lysosome.

Knockouts of IGFI from mice caused a severely growth retarded phenotype. Moreover, their retarded growth rate continues after birth. By six weeks after birth, they weighed only 30% of their heterozygous or wild-type sibs. Such mutants also have hypoplasia of their reproductive systems. Knockouts of the IGF-1 receptor achieve a body weight only half that of wild-type, and die shortly after birth from a combination of skin defects, muscle hypoplasia, and delayed skeletal ossification (Liu et al., 1991). The IGFI gene and its receptor appear to have normal Mendelian inheritance patterns.

Knockouts of the IGFII showed that homozygotic deficient mice were severely growth retarded, reaching only 60% of the normal birthweight. Moreover, the heterozygotes showed an unexpected pattern. If the mutant allele came from male mice, the heterozygous mice produced were as growth retarded as the homozygotes. If the mutant allele came from females, the growth of the heterozygotes resembled that of the wild-type fetuses. It was shown that only the paternal IGFII gene is active in most of those tissues expressing that gene (DeChiara et al., 1991a,b).

The IGFII gene of mice and humans are found in association with other imprinted genes. In the mice, the distal region of chromosome 7 contains five imprinted genes: (1) Mash-2, a helix-loop-helix transcription factor expressed from the maternal chromosome and active in the mouse placenta, (2) Insulin-2, a gene expressed from the paternal chromosome in the yolk sac but from both chromosomes in the pancreas, (3) Igf2, encoding insulin-like growth factor 2, expressed only from the paternal chromosome in most tissues, p57KIP2, a cdc2 kinase inhibitor that is maternally expressed, and H19, a gene encoding a nontranslatable RNA that is only produced by the maternal chromosome. A similar cluster is found on the small arm of human chromosome 11. The clustering of these genes on human and mouse chromosomes has raised the possibility that imprinting (like X-inactivation) is regulated by a signal that can act on several linked genes.

Recent hypotheses have focused on the linking of the Igf2 and H19 genes, partly because they are only 90 kb from each other, and partly because their expression patterns are strikingly similar, but from the "opposite" chromosome. One model (Leighton et al., 1995; Viville and Surani, 1995) contends that there is competition between the H19 and Igf-2 promoters for the same enhancers, and that a methylation imprint on the H19 gene from the germ cells determines which one will "win" the competition and become active (Figure 1).

Figure 1 Model for the regulation of imprinting at the wild type Igf2 locus. This model posits competition for the H19 enhancer (E, far right) by the H19 promoter and the Igf2 and insulin-2 promoters. On the paternal chromosome the promoters of both groups are methylated, and the enhancer interacts with the Igf2 and Ins-2 promoters. On the maternal chromosome, the H19 promoter is not methylated and the enhancer interacts with it. In the H19 deletion, the enhancer binds to the Ins-2 and Igf2 promoters, and in the methyltransferase deletion, the H19 promoters are not methylated, so the enhancer binds preferentially with them. (After Eden and Cedar, 1995).

Evidence for this model comes from several sources. First, the methylation pattern of the H19 gene showed a very clear difference between its active and inactive states. The promoter region of the repressed H19 gene was hypermethylated, while the active copy was hypomethylated and DNase I hypersensitive. Moreover, there are sites 3- 5 kb upstream of H19 which are methylated in the sperm and unmethylated in the oocyte. These could be the imprinted sites of the H19 gene (Tremblay et al., 1995).

Second, in the DNA methyltransferase gene knockout, both parental copies of H19 are expressed. This causes a reciprocal loss of Igf2 expression, and it suggests that the normally active Igf2 locus on the paternal chromosome has now become repressed. The loss of methylation on the paternal H19 gene may allow it to become expressed where it normally would not be (Li et al., 1993). When the H19 promoter is unmethylated (as it is on the maternal chromosome), the enhancers interact with the H19 enhancer and the H19 gene is turned on. When the H19 promoter is methylated (as it is on the paternal chromosome), it does not bind well to factors on the H19 enhancer, and the enhancer interacts instead with the Igf2 promoter, turning on the Igf2 gene.

Third, in a recent study (Leighton et al., 1995), the H19 gene (and its promoter) was deleted by gene knockout, but the H19 enhancer region was left intact. The paternal transmission of this deletion had no effect on the phenotype. However, if the mutation were transmitted by the mother, there was a 28% increase in the body size of the mouse at birth. This increase was attributable to an overproduction of IGF-II, since both the parental alleles of Igf2 were expressed. The H19 knockout mice had the phenotype opposite those of the DNA methyltransferase-deficient mice. Leighton and colleagues interpreted these results to show that when the H19 promoter on the maternal chromosome is removed, the enhancers interact with the maternal Igf2 gene, causing it to become expressed.

Literature Cited

DeChiara, T. M., Efstradiatis, A., and Robertson, E. J., 1991a. A growth deficiency phenotype in heterozygous mice carrying an insulin-like growth factor II gene disrupted by targeting. Nature 345: 78-80.

DeChiara, T. M., Robertson, E. J. and Efstratiadis, A. 1991b. Parental imprinting of the mouse insulin-like growth factor II gene. Cell 64: 849-859.

Eden, S. and Cedar, H. 1995. Action at a distance. Nature 375: 16-17.

Leighton, P. A., Ingram, R. S., Eggenschwiler, J., Efstratiadis, A., and Tilghman, S. M. 1995. Disruption of imprinting caused by deletion of the H19 gene region in mice. Nature 375: 34-39.

Li, E., Beard, C., and Jaenisch, R. 1993. Role of DNA methylation in genomic imprinting. Nature 366: 362-365.

Liu, J-P., Baker, J., Perkins, A. S., Robertson, E. J., and Efstratiadis, A. 1993. Mice carrying null mutations of the genes encoding insulin-like growth factor I (Igf-1) and type1 receptor (Igf1r ). Cell 75: 59-72.

Tremblay, K. D., Saam, J. R., Ingram, R. S., and Tilghman, S. M. 1995. A paternal-specific methylation imprint marks the alleles in the mouse H19 gene. Nature Genet. 9: 407-413.

Viville, S. and Surani, M. A. 1995. Towards unraveling the Igf2/H19 imprinted domain. BioEssays 17: 835-838.

© All the material on this website is protected by copyright. It may not be reproduced in any form without permission from the copyright holder.

HOME :: CHAPTER 5 :: IMPRINTING IN HUMANS AND MICE :: H19, INSULIN-LIKE GROWTH FACTORS, AND THE REGULATION OF EMBRYONIC GROWTH

PREVIOUS :: NEXT