Non-identical Monozygotic Twins
Most monozygotic twins are remarkably similar. However, there are some equally remarkable exceptions that show that monozygotic twins do not have to be "identical."
It is usually assumed that monozygotic twins always share the same genome and that their hereditary endowments are the same. Indeed, twins have long been used in studies that attempt to separate "nature" from "nurture". Galton (1875) proposed that the comparison of monozygotic twins to dizygotic twins would show which characteristics are genetic ("nature") and which are conditioned by the environment ("nurture"). However, recent data have suggested that each of a pair of one-egg twins can develop very differently from the other, and these findings of "discordant monozygotic twins" have raised concerns about the assumptions underlying those studies that presume each of a pair of identical twins will develop the same way. In fact, Hall (1996) hypothesizes that the mechanism underlying the separation of blastomeres (which leads to identical twinning) may be genetic changes that distinguish the cells from each other. In this scenario, "identical" twinning is caused by the non-identity of the early blastomeres.
There are several cases in which identical twins can have discordant phenotypes. The major routes involve aneuploidy, X-chromosome inactivation differences, imprinting, and circulatory differences.
For example, take the situation of an egg fertilized by a Y-bearing sperm. It has been found (Opitz, 1993) that monozygotic twinning is associated with higher than normal amounts of aneuploidy; so it is possible that if twinning were to occur through the failure of the first two blastomeres to adhere to one another, aneuploidy might also occur. In that case, the twins would have different chromosome complements. If the aneuploidy were for the X chromosome, one twin might by male (XY or XYY) while the other would be female (XO). Such male/female "identical twins" have been found (Edwards et al., 1966; Machin, 1996). These twins would be assumed dizygotic, when they actually had originated monozygotically.
The randomization of X-chromosome inactivation can also give rise to monozygotic twins who are profoundly different. Dosage compensation for the human X chromosome is such that one of the two X chromosomes becomes inactive in each somatic female cell. However, there is no determining which X chromosome will be inactivated (see Chapter 11 of Gilbert, 1997). Take, for example, the case of identical twins who are heterozygous for an X-linked form of muscular dystrophy. Most heterozygous women do not express any symptoms because the cells expressing the wild-type allele can compensate for those cells expressing the mutant allele. However, if by chance the wild-type allele is on the inactivated X chromosome in a large proportion of her muscle cells, the woman will manifest the disease. There have been several instances where one girl shows the symptoms of the disease while her identical twin sister does not (Pena et al., 1987; Norman and Harper, 1989; Richards et al., 1990; Tremblay et al., 1993). Similarly, there are cases where monozygous female twins are discordant with respect to colorblindness or Hunter disease due to X-chromosome inactivation (Jorgensen et al., 1992; Winchester et al., 1992; Goodship et al., 1996). It is possible that X-inactivation is the cause of late monozygotic twinning, since recent studies have shown that a large percentage of monoamniotic-monochorionic twins are female and that they have oppositely skewed patterns of X-chromosome activation. For more information on this hypothesis, click here.
The Wiedemann-Beckwith syndrome (WBS) is a rare genetic condition consisting of giganistism, macroglossia (enlarged and protruding tongue), and exomphalos (protruding umbilicus). It usually occurs spontaneously and is associated with a duplication of paternal chromosome 11p. It is thought that the gene encoding insulin-like growth factor II (IGF2; which maps to 11p15) is imprinted such that the paternal allele is exclusively expressed during embryogenesis, while the maternally derived allele is inactive (Ohlsson et al., 1993; Rainer et al., 1993). Uniparental disomy (where a cell has two copies of a chromosome or chromosomal part derived from one parent and none from the other parent) can result when a cell contains three copies of a chromosome. It can exclude one of the chromosomes, but in so doing, it can then have both chromosomes derived from the same parent. In the case of imprinted genes, this means that the chromosomes would be homozygous for either the active or inactive allele. Uniparental disomy can happen spontaneously after the twinning event, and in most twins with WBS, one twin manifests the condition, while the other does not (Leonard et al., 1996).
Discordance for growth and major malformations
Boklage (1990) estimates that only one out of every six twin pregnancies come to term as twins. In the other cases, there is one or more "vanishing twin," and the pregnancy delivers one baby or none at all (Hall, 1996). While the mechanisms producing such vanishing twins is not known, research has focused on vascular problems that result from identical twins sharing the same placental circulation. Fetal growth discordance is likely when one twin draws blood from a larger arterial zone in the placenta than the other. In other cases, however, growth discordance can be the result of blood flowing from one twin into the other. This can occur through the placental venous return regions or from anastomeses between blood vessels. Such twin-twin transfusions can result in the death of one twin and developmental abnormalities in the survivor (Benirschke, 1961; Machin, 1996). Donnenfeld and colleagues (1989) have shown that selective termination of a malformed twin fetus can be dangerous to the normal twin, since it may accidentally result in the exsangination of blood from the normal fetus.
Polar body twins
Between the concepts of monozygous twins and dizygous twins is the idea of polar body twins. In such cases, one sperm is thought to fertilize the egg, while another sperm is thought to fertilize a polar body. Bieber and colleagues (1981) analysed an acardiac fetus of a normal co-twin. Such embryos without hearts are only found to exist in situations of twinning, where circulation from the normal twin keeps the abnormal twin alive. In this case, the abnormal, acardiac, twin was an XXX triploid, having three complete chromosome sets. Analysis of chromosomes and major histocompatability antigens showed that the acardiac twin was produced from the fertilization of a sperm with the diploid first meiotic division polar body.
Concerning Galton's (and society's) questions of hereditary versus environmental influence, one must realize that identical zygotic genomes do not mandate identical development. Intrauterine differences in placental circulation, aneuploidy, X-inactivation, and nuclear changes involving imprinted chromosomes indicate that the newborn monozygotic twins may already be different. Moreover, as changes in experience alter the neuronal pathways and changes in microenvironments alter the lymphocyte population, it is difficult to separate out a pure "nature" and a pure "nurture".
Benirschke, K. 1961. Twin placenta in perinatal mortality. N.Y. State Med. J. 61: 1499-1508.
Bieber, F. R., Nance, M. M., Morton, C. C., Brown, J. A., Redwine, F. O., Jordan, R. L., and Mohanakumar, T. 1981. Genetic studies of an acardiac monster: evidence for polar body twinning in man. Science 213: 775-777.
Bocklage, C. E. 1990. Survival probability of human conceptions from fertilization to term. Internat. J. Fertil. 35: 75-94.
Donnenfeld, A. E. , Glazerman, L. R., Cutillo, D. M., Librizzi, R. J., and Weiner, S.1989. Fetal exsanguination following intrauterine angiographic assessment and selective termination of a hydrocephalic monozygous co-twin. Prenat. Diagn. 9: 301-308.
Edwards, J. H., Dent, K., and Kahn,J. 1966. Monozygotic twins of different sex. J. Med. Genet. 3: 117-123.
Galton, F. 1875. The history of twins, as a criterion of the relative powers of nature and nurture. J. Anthropol. Inst. 12: 566-576.
Goodship, J., Carter, J., and Burn, J. 1996. X-inactivation patterns in monozygotic and dizygotic female twins. Amer. J. Med. Genet. 61: 205-208.
Jorgensen, A. L., Philip, J., Raskind, W. H., Matsushita, M., Christensen, B., Dreyer, V., and Motulsky, A.G. 1992. Different pattern of X-inactivation in MZ twins discordant for red-green color-vision deficiency. Amer. J. Human Genet. 51: 291-298.
Hall, J. 1996. Twins and twinning. Amer. J. Med. Genet. 61: 202-204.
Leonard, N. J., and several others. 1996. Two pairs of male monozygotic twins discordant for Wiedemann-Beckwith syndrome. Amer. J. Med. Genet. 61: 253-257.
Machin, G. A. 1996. Some causes of genotypic and phenotypic discordance in monozygotic twin pairs. Amer. J. Med. Genet. 61: 216-228.
Norman, A. and Harper, P. 1989. Survey of manifesting carriers of Duchenne and Becker muscular dystrophy in Wales. Clin. Genet. 36: 31-37.
Ohlsson, R. and several others. 1993. IGF2 is parentally imprinted during human embryogenesis and in the Beckwith-Wiedemann syndrome. Nature Genet. 4: 94-97.
Opitz, J. 1993. Blastogenesis and the "primary field" in development. In Blastogenesis : Normal and Abnormal (J. M. Opitz, ed). Birth Defects 29: 3-37.
Pena, S.D. J., Karpati, G., Carpenter, S., and Fraser, F. C. 1987. The clinical consequences of X-chromosome inactivation: Duchenne muscular dystrophy in one of monozygotic twins. J. Neurol. Sci. 79: 337-344.
Rainer, S., Johnson, L. A., Dobry, C. J., Ping, A. J., Grundy, P. J., and Feinberg, A. P. 1993. Relaxation of imprinted genes in human cancer. Nature 362: 747-749.
Richards, C. S. and several others. 1990. Skewed X-inactivation in a female MZ twin results in Duchenne muscular dystrophy. Amer. J. Human Genet. 46: 672-681.
Tremblay, J. P. and several others. 1993. Myoblast transplantation between monozygous twin girl carriers of Duchenne muscular dystrophy. Neuromusc. Disord. 3: 583-592.
Winchester, B. and several others. 1992. Female twin with Hunter disease due to non-random inactivation of the X-chromosome: a consequence of twinning. Amer. J. Med. Genet. 44: 834-838.