Mountain States Genetics Foundation

From the Editor and Authors: The mendelian model of inheritance, in which dominant and recessive traits are passed on according to the segregation of chromosomes, provides the framework for most of our thinking about human genetic disorders; however, most clinicians encounter a few families whose stories do not fit quite so easily into this scheme.

Now, molecular genetics is revealing new facts about human inheritance that are beginning to explain what Dr. Judith Hall, clinical geneticist from Vancouver, British Columbia has called "nontraditional inheritance". Nontraditional inheritance is both prevalent and clinically relevant. This issue of Genetic Drift will describe how recent discoveries about mosaicism, mitochondiral inheritance, genetic imprinting, uniparental disomy and the instability of triplet nucleotide repeats are reshaping medicine's view of inheritance.

This issue was written by David K. Manchester, M.D.(CO) and Suzanne Cassidy, M.D. (AZ+OH), with helpful input from Annette Taylor, Ph.D., Carole Green, M.D., William Seltzer, Ph.D.(CO), and Lori Ballinger, M.S.(NM)

Adults have roughly 10 trillion (i.e. 10 million, million) cells. Mutations occur at any given locus as frequently as once in every million cell divisions, on average, and at some loci as frequently as once in every 50 thousand cell divisions. Hence, it is safe to say that all people are mosaic, with the effects of this mosaicism dependent upon the number of cells involved, the specific genes affected, and locations of the mutant cells.

Many somatic mutations are deleterious and kill the cells in which they occur. Others may be selected against in rapidly dividing or differentiating populations of cells. When either of these effects of somatic mutation occurs early enough during development to disrupt embryonic structures, birth defects may be produced. Studies in experimental animals exposed to mutagenic agents confirm these mechanisms, but the extent to which they are involved in human birth defects in not known. Chromosomal mosaicism, however, has been well documented in some children with birth defects. Approximately one in every fifty cases of down syndrome is found to be mosaic by routine cytogenetic analyses. Other trisomies, such as trisomy 9, are found only as mosaics, presumably because complete trisomy would be lethal. Some geneticists have speculated that all viable aneuplodies may be mosaic.

Deleterious mutations may occur at any time during life, and it is very likely that the same processes of cell death and negative selection also contribute to aging. Again, the extent to which this occurs has not yet been determined. The current technical barrier to more definitive investigations of the contributions of mosaicism to human disorders is the lack of sensitive, quantitative assays that can detect low frequency mutations against normal backgrounds.

Some somatic mutations have opposite effects on cells and provide them with selective advantages for growth. These are mutations that initiate or contrubute to the development of cancer. With great effort, many of the important genes that are mutated during carcinogenesis are now being identified (see Genetic Drift, Vol 9, 1993). Clinicians can anticipate that the techniques applied in this setting will also be useful in delineating the contributions of somatic mutations to birth defects, aging and other disorders.

The phenotypic effects of somatic mutations depend on when the mutations occur; the earlier in development, the more prevalent will be their contributions. There is mounting evidence that mitotic missegregation of chromosomes in very early gestation frequently leads to mosaicism that is confined to the placenta.

At implantation, the majority of cells that make up the blastocyst differentiate into trophoblast, tissue that eventually makes up the placenta. Less than 10% of blastocyst cells contribute to the inner cell mass that is destined to form the embryo, amnion and extraembryonic mesoderm. Hence, chromosomal missegregation or other mutations occurring during this early stage of development are most likely to affect cells contributing to the placenta.

Such "abnormal" pregnancies may remain viable because fusion of normal and abnormal nuclei to form the functional syncytium of the placenta complements genetic imbalances. Compensation within syncytiotrophoblast may not always be complete, however, and confined placental chromosomal mosaicism is associated with intrauterine growth retardation.

Most importantly for genetic counseling, early embryonic somatic mutations and chromosomal missegregation can introduce genetic disorders into the germ-line by producing gonadal mosaicism. Several rounds of cell division occur between fertilization and differentiation of germ cells in the developing embryo.

Mutations that antedate formation of germ stem cells can lead to mosaicism from which two or more populations of gametes, each possessing a different genotype, can be derived. Classical teaching of mendelian segregation of traits has not recognized this possibility, and so it came as a surprise to many in medicine when geneticists began describing recurrences of dominant mutations in families where parents were unaffected (see Figure 1).

There are now well documented cases of gonadal mosaicism for dominant mutations in what would otherwise have been considered as recessively inherited cases of skeletal dysplasias, connective tissue disorders, and a few chromosomal aberrations. Dr. Peter Byers' group at U. Washington in Seattle has taken a particular interest in gonadal mosaicism and has directly demonstrated mosaicism in sperm for mutations leading to osteogenesis imperfecta and Ehlers-Danlos syndrome Type IV. In their experience with osteogenesis imperfecta, severe phenotypes formerly considered as recessively inherited are in fact caused by mutations expressed in heterozygotes (i.e. dominant traits). These recur as a result of gonadal mosaicism at an empiric rate of 6% (Figure 2).

Gonadal mosaicism is even more prevalent in families with Duchenne/Becker muscular dystrophy. Researchers analyzing changes in the dystrophin gene that cause this disorder report gonadal mosaicism in 15 to 25% of families in which they have tracked mutations to their origins. These numbers underscore the need to now include the possibility of gonadal mosaicism in analyzing essentially all pedigrees and to consider it routinely in genetic counseling for recurrence risks.

Finally, when somatic mutations lead to gonadal mosaicism, mildly affected mosaic individuals may have offspring with severe symptoms. In some cases, parent and offspring may appear to have different disorders. This has recently been demonstrated in a family where a parent with apparent Stickler syndrome had a child with Kneist dysplasia. These are both dominantly inherited disorders of connective tissue in which there are joint and spine abnormalities, near sightedness, hearing loss, and, often, cleft palate. Kneist dysplasia is the more severe of the two. Each disorder involves a mutation in a gene for type II collagen. Ordinarily, different mutations account for the differing phenotypes, but in this family, the variation was found to be due to mosaic expression of the severe mutation in the parent and full expression in the child.

A few mechanisms of somatic mutation are now beginning to be understood. An example is the normal rearrangement of genes that produces multiple clones of B and T-cell precursors during the development of lymphocytes. This recombination is catalyzed by enzymes that target specific DNA sequences.

Since target sequences also occur in other genes, illegitimate recombination occurs with low frequency at loci other than those coding for variable regions of antibodies and T-cell receptors. This mechanism likely contributes to development of childhood leukemia and lymphomas.

As other mechanisms of mutagenesis become understood, it is likely that their footprints will be found in other cancers, in birth defects, and in an unknown number of germline disorders entering into families through gonadal mosaicism.

The Genetic Drift Newsletter is not copyrighted. Readers are free to duplicate all or parts of its contents. The Genetic Drift Newsletter is published semiannually by the Mountain States Regional Genetic Services Network for associates & those interested in Human Genetics. In accordance with accepted publication standards, we request acknowledgement in print of any article reproduced in another publication. The views expressed in the newsletter do not necessarily reflect local, state, or federal policy. For additional information, contact Carol Clericuzio, M.D., Editor, Department of Pediatrics, The University of New Mexico, Albuquerque, NM, 87131

Vol. 10: Winter, 1994

Nontraditional Inheritance

Mosaicism