Main Page Dynamic Development
The Foundations of Developmental Biology
From Sperm and Egg to Embryo
Genetic Regulation of Development
Organizing the Multicellular Embryo
Generating Cell Diversity
Dynamic Development at a Glance
The Developmental Biology Journal Club
Developmental Biology Tutorial
C. Cristofre Martin
3307 N. Broad Street
Philadelphia, PA 10140
Genomic imprinting can be loosely defined as the gamete-of-origin dependent
modification of phenotype. That is, the phenotype elicited from a locus is differentially
modified by the sex of the parent contributing that particular allele. This process
results in a reversible gamete-of-origin specific marking of the genome that ultimately
produces a functional difference between the genetic information contributed by each
parent. In mammals, the term genomic imprinting has been restricted to describing
mono-allelic gene expression or the inactivation of either the maternal or paternal allele
of a particular locus.
Similarly, dominance modification involves the production of phenotypic differences in the
degree of penetrance of a particular locus (Fisher, 1928). This epigenetic modification is
produced by the action of other, distinct modifier loci, most commonly acting in trans.
Genomic imprinting is thought to be a particular sub-type of dominance modification
(Sapienza, 1990), whereby imprinting would result when one or more of these modifiers is
sex-linked and would therefore produce a dosage difference.
Imprinting-like phenomena have been observed in a wide range of phyla from both the plant
and animal kingdoms. Some manifestations of this gamete-of-origin dependent modification
include: parental dominance in hybrid plants (Heslop-Harrison, 1990); paternal chromosome
elimination in Sciara (Metz, 1938; Crouse et al., 1971); inactivation of the
paternal genome by heterochromatization in the scale insects (coccids) (Nur, 1990);
parent-of-origin specific modification of position effect variegation in Drosophila
(Spofford, 1976); phenotypic differences in progeny produced from interspecific crosses in
fish (Whitt et al., 1977) and between donkey and horse; preferential inactivation of the
paternally derived X chromosome in marsupials and rodent extraembryonic tissues (VandeBerg
et al., 1987); parent-of-origin dependent switching of yeast mating types (Klar,
1987); and allelic exclusion of immunoglobulin genes (Holliday, 1990). Probably the most
extensively studied and best understood examples of genomic imprinting are the
gamete-of-origin dependent modifications of transgene methylation and expression in mice
(Swain et al., 1987; Reik et al., 1987; Sapienza et al., 1987) and
more recently in a line of transgenic zebrafish (Martin and McGowan, 1995a; 1995b). It is
not at all clear yet whether all of the above phenomena are occurring via a similar
molecular mechanism. That a parent-of-origin effect occurs in such a wide range of
organisms suggests that either there is strong evolutionary conservation of this
phenomenon or that it arose independently a number of times. The final result in all cases
is an individual that appears to be functionally hemizygous at an imprinted locus. Some of
these functionally hemizygous loci would be the result of imprinting by the male parent
and some would be the result of imprinting by the female parent (Figure 1).
Figure 1: Two examples of a hypothetical imprinted gene responsible
for body color. (LEFT) In this example the pigment gene is maternally imprinted (maternal
allele is inactivated). Matings between a male who possesses the allele for pigment and a
female who possesses the allele for no pigment produces offspring that show only the
pigmented phenotype. In this example, the mother's allele is imprinted and inactivated in
the offspring. Therefore, the only actively-expressing allele is the father's pigment
allele, which is not imprinted in the offspring. (RIGHT) In this example the pigment gene
is paternally imprinted (paternal allele is inactivated). Matings between a male who
possesses the allele for pigment and a female who possesses the allele for no pigment
produces offspring that show only the pigmented phenotype. In this example, the father's
allele is imprinted and inactivated in the offspring. Therefore, the only actively
expressing allele is the mother's no pigment allele, which is not imprinted in the
offspring. (Figure courtesy of Ross McGowan, Dept. Zoology, University of Manitoba).
Because there is monoallelic parent-of-origin specific expression from an imprinted
autosomal locus, genomic imprinting is counter to classical Mendelian genetic theory,
which states that there is equal inheritance of parental traits. An imprinted mutant
allele would appear to be recessive when inherited from one sex, because it would be
inactive (and consequently invisible), whereas it would be the only active allele when
inherited from the other sex and, therefore, appear to be dominant (Figure 2).
Figure 2: Schematic representation of the phenotypic effects of
maternal imprinting of a mutant allele. Darkened body indicates individual that is mutant
for the hypothetical imprinted locus. A cross is used to indicated the imprinted/inactive
allele. (CENTER) Both parents are homozygous for the normal allele at the imprinted locus.
Although only one allele is active (the paternal copy) in the offspring produced from
these parents, it must be a normal allele and therefore all offspring will have a normal
phenotype. (LEFT) The mother is homozygous mutant at the imprinted locus, and the father
is normal. Since this hypothetical locus is maternally imprinted, the maternal mutant copy
will be inactivated in their offspring and the paternal normal copy will be the only
active allele. The offspring will be phenotypically normal, and the mutant allele will
appear to be a recessive mutation. (RIGHT) The mother is homozygous normal at the
imprinted locus, and the father is homozygous mutant. The maternal normal allele is
imprinted and inactivated in the offspring of these parents. The only allele that is
active is the mutant paternal copy. Therefore, all offspring produced from these parents
will display the mutant phenotype, and the mutant allele will appear to be a dominant
mutation. (Figure courtesy of Ross McGowan, Dept. Zoology, University of Manitoba).
As stated previously, the genome imprinting phenomenon has been most intensively studied
in mammals, particularly mice. One type of epigenetic modification in mammals involves the
addition of a methyl group to position 5 of cytosine and occurs most frequently at CpG
dinucleotides. This type of DNA modification is significant because, in general,
methylated DNA sequences are transcriptionally inactive. Furthermore, the methylation of
DNA residues can be stably preserved through the replication process by the action of
maintenance methylases, which use the replicated hemimethylated DNA as a template (Razin
and Riggs, 1980). It has, therefore, been suggested that DNA methylation may be the
epigenetic marking responsible for the imprinting phenomenon in mammals, or at least
reflects the imprinted state of a locus produced during gametogenesis (reviewed in Sasaki et
al., 1993). It is logical to assume that the epigenetic differences observed in the
paternal and maternal genetic contributions would be established during the only time when
the two genomes are segregated; i.e., during gametogenesis.
If methylase differences exist in the gametes, one would expect to find different DNA
methylation patterns in sperm and eggs. These differences should be perpetuated at least
through early development when the final somatic tissue methylation pattern is
established. Monk et al. (1987) and Sanford et al. (1987) demonstrated in
mice that sperm DNA is more methylated than oocyte DNA. In the mouse, a period of overall
demethylation has been observed from fertilization, lasting until approximately the
blastocyst stage, followed by extensive de novo methylation. This period begins at
gastrulation and ultimately produces methylation levels higher than those observed in
either of the gametes (reviewed in Monk, 1990). Similar changes in the methylation
phenotype of a variety of transgenes in mice have been observed during development
(Chaillet et al., 1991; Ueda et al., 1992).
A number of laboratories have demonstrated a gamete-of-origin influence on the level of
methylation of a variety of transgenes in mice (Hadchouel et al., 1987; Reik et
al., 1987; Sapienza et al., 1987; Swain et al., 1987). With one
exception, methylation tends to increase after passage of a transgene locus through the
female germline and decrease after passage through male gametogenesis. In a number of
cases an accompanying effect has been found with respect to expression of the transgene:
Expression is decreased after passage through a female and increased by passage through a
male (Swain et al., 1987; McGowan et al., 1989; Allen et al., 1990).
A mosaic pattern of transgene expression has also been found in several lines of
transgenic mice, which suggests that the imprinted state may not be retained in all cells
(McGowan et al., 1989) or that only a sub-set of early stem cells is affected by
Several endogenous genes, as well as regions of many mouse chromosomes (Cattanach and
Beechey, 1990), have been shown to be imprinted. To date only a small, but still growing,
list of mammalian genes has been demonstrated unequivocally to be imprinted (Table 1).
Estimates of the number of imprinted genes in the mammalian genome range from less than
100 to greater than 200. Like the imprinted transgenes, these endogenous genes show high
levels of DNA methylation at the imprinted/inactive allele. At present, no consensus
sequences for imprinted loci have been identified at/or surrounding these imprinted genes.
However, the imprint imposed at these loci is a very specific and localized process. The
imprinted genes H19 (maternal) and Igf2 (paternal) are tightly linked to
each other and localized to the same region of mouse chromosome 7 but are imprinted in the
The pronuclear transplantation experiments of McGrath and Solter (1984) and Surani et
al. (1984; 1986) using mice have unequivocally demonstrated that there is an absolute
requirement for a genetic contribution from both sexes in order for development to proceed
normally; i.e., maternal and paternal contributions are not equivalent. Embryos
containing only maternal contributions develop minimal extraembryonic tissues
(trophectoderm), whereas a poorly developed embryo proper is characteristic of embryos
containing only paternal genomes. The importance to the mammalian embryo of both maternal
and paternal genomes is most apparent in the very low viability of mammalian
parthenogenotes. The timing of developmental arrest and the phenotype of parthenogenetic
mouse embryos is highly variable and is most dependent on the genetic background of the
embryos. Parthenogenetic embryos produced from one inbred strain of mice may only develop
to blastula or gastrula stages, whereas parthenogenetic embryos produced from another
strain of mice might develop to the 40 somite stage. Understanding of the role of genomic
imprinting in development has been hampered by the lack of a distinct/consistent embryonic
phenotype in parthenogenetic embryos and the lack of a precise role for the presently
known imprinted genes in development.
Considerable interest has been generated by the realization that the penetrance and
severity of many complex human diseases can be affected by the sex of the parent
contributing the allele responsible for the disease. The parent of origin effects
associated with a disease suggest that imprinting may be involved. Some of these putative
imprinted diseases include Huntington's disease, cystic fibrosis, Prader-Willi and
Angelman syndromes (reviewed in Hall, 1990; Sapienza and Hall, 1995). Parent of origin
effects are also seen for birth defects such as spina bifida (Chatkupt et al.,
1992) and certain cancers. The role of imprinting in the development of cancers such as
Wilm's tumour, rhabdomyosarcomas and osteosarcomas has been particularly well studied. By
examining DNAs collected from tumors and normal tissues, it has been revealed that the
development and proliferation of cancerous tumors may be associated with a loss of
imprinting at a number of different loci. Two examples of these imprinted loci are: Igf2
(insulin-like growth factor), which is involved in beta-cell tumorigenesis,
hepatoblastoma, Wilm's tumor and rhabdomyosarcomas (Christofori et al., 1995; Li et
al., 1995; Pedone et al., 1994) and the gene H19, which has been shown
to be involved in hepatoblastoma, lung cancer, Wilm's tumor and bladder carcinoma (Li et
al., 1995; Kondo et al., 1996; Ariel, 1995). The development of these cancers
is thought to be due to the relaxation of the normal monoallelic expression or 'imprinted'
status of these genes resulting from the loss of DNA methylation at the imprinted allele.
Numerous investigators have presented theories regarding the evolutionary
origins/advantages of genomic imprinting. Few of these theories have been able to satisfy
all the disparate information from our present understanding of the phenomenon, such as:
1) broad phylogenetic distribution, 2) the different types of imprinted genes, and 3) both
paternal and maternal imprinting. Chandra and Nanjundiah (1990) put forward several
postulates describing in a general way the kinds of influences that could produce
imprinting and suggest several roles for imprinting: 1) imprinted and non-imprinted
alleles of a locus may confer different phenotypes; 2) modifiers of imprinting
(so-called"imprintor genes") may have pleiotropic effects such that they are
selected for reasons other than their action on the imprinted locus; and 3) imprinting
could have co-evolved with other traits. Holliday (1990) suggests that the function of
genomic imprinting is to produce functionally haploid gene sets in order for the cell's
regulatory machinery to be able to 'fine tune' itself on just a single gene copy. It was
further implied that imprinting would also serve to prevent possible detrimental
cross-talk between maternal and paternal transcripts at developmentally important
regulatory loci. Barlow (1993) has suggested that genome imprinting as a mechanism of gene
regulation originated from the use of DNA methylation as a host defense mechanism. The
same processes that inactivate incoming viruses in oocytes might also result in the
inactivation of some of the organism's own (maternal) loci. Recently, Thomas (1995)
proposed that imprinting acts as a mechanism of surveillance for chromosome loss and thus
protects against cancers and mono- and trisomic embryos. If imprinted loci are scattered
throughout the genome, loss of a chromosome or portion of a chromosome would result in the
absence of an active copy of an allele producing probable detrimental effects. Other
theories are based solely on genomic imprinting as it occurs is mammals. McGrath and
Solter (1984) noted that there is a phenotypic difference in the developing embryos of
mouse andro- and gynogenotes. More specifically, androgenetic embryos show poorly
developed embryos and excessive trophectoderm (extraembryonic supportive tissue), whereas
gynogenetic embryos have well developed embryos and poorly developed trophectoderm. These
"imprinting" phenotypes of andro- and gynogenetic embryos seem to suggest the
possibility of the male using imprinting to maximize maternal input to the embryo (Haig
and Westoby,1989). Finally, Varmuza and Mann (1994) suggested that imprinting in mammals
may be a means of protecting the female from ovarian germ cell tumors. Imprinting would
prevent the development of malignant trophoblast disease because of the inviability of
unfertilized eggs and inability of parthenogentically activated oocytes to implant and
It has been the intention of this review to provide a basic understanding of the process
of genomic imprinting (further review of the listed references will provide a much greater
appreciation of this very complex process). Although genomic imprinting has been
intensively studied in the past few years, it is still a very new field. Our present
knowledge of imprinting is limited to the identification of imprinted genes and to only
the very fundamental factors that contribute to the process. Researchers are now focusing
their efforts on identifying the trans- acting modifiers of imprinting -(i.e.,
"imprintor" genes), understanding the role of DNA conformation and chromatin
structure surrounding imprinted loci (ie., cis elements) and on determining the
role of imprinting in development (establishment of the imprint) and human disease.
- Define genomic imprinting.
- What is meant by the expression "parent of origin expression"?
- Diagram a cross showing the phenotypes and genotypes of a gene that is only active if it
is derived from the maternal parent. Do the same for a gene that is only active if it is
derived from the male parent.
- What modification of DNA is thought to be responsible for genomic imprinting?
- What is the evidence favoring this kind of modification?
- Discuss the pronuclear transplantation experiments of McGrath and Solter.
- What may be some consequences of the loss of imprinting? Give examples.
Allen, N.D., Norris, M.L., and Surani, M.A.H. (1990). Epigenetic control of
transgene expression and imprinting by genotype-specific modifiers. Cell 61:
Ariel, I. et al. (1995). The imprinted H19 gene as a tumor marker in bladder
carcinoma. Urology 45: 335-338.
Barlow, D.P., Stroger, R., Herrmann, B.G., Saito, K., and N. Schwiefer (1991).
The mouse insulin-like growth factor type-2 receptor is imprinted and closely
linked to the Tme locus. Nature 349: 84-87.
Barlow, D.P. (1993). Methylation and imprinting: From host defense to gene
regulation? Science 260: 309-310.
Bartolomei, M.S., Zemel, S., and S.M. Tilghman (1991). Parental imprinting of
the mouse H19 gene. Nature 351: 153-155.
Cattanach, B.M. and C.V. Beechey. (1990). Autosomal and X-chromosome
imprinting. Development Supplement, 63-72.
Christofori, G. et al. (1995). Deregulation of both imprinted and expressed
alleles of the insulin-like growth factor 2 gene during beta-cell tumorigenesis.
Nature Genetics 10: 196-201.
Crouse, H.V. (1960). The controlling element in sex chromosomes behaviour in
Sciara. Genetics 45: 1429-1443.
Dechiara, T.M., E.J. Robertson, and A. Efstratriadis (1991). Parental
imprinting of the mouse insulin-like growth factor II gene. Cell 64: 849-859.
Fisher, R.A. (1928). The possible modification of the response of the recurrent
mutations. American Naturalists 62: 115-126.
Giddings, F., King, C.D., Harman, K.W., Flood, J.F., and L.R. Carnaghi (1994).
Allele specific inactivation of insulin 1 and 2, in the mouse yolk sac,
indicating imprinting. Nat. Genet. 6: 310-313.
Guillemot, F., Caspary, T., Tilghman, S.M., Copeland, N.G., Gilbert, D.J.,
Jenkins, N.A., Anderson, D.J., Joyner, A.L., Rossant, J., and A. Nagy (1995).
Genomic imprinting of Mash-2, a mouse gene required for trophoblast development.
Nat. Genet. 9: 235-241.
Hadchouel, M., Farza, H., Simon, D., Tiollais, P., and C. Pourcel (1987).
Maternal inhibition of hepatitis B surface antigen gene expression in trangenic
mice correlates with de novo methylation. Nature 329: 454-456.
Haig, D.M., and M. Westoby (1989). Parent-specific gene expression and the
triploid endosperm. Am. Nat. 134: 147-155.
Hall, J.G. (1990). Genomic imprinting: review and relevance to human diseases.
Am. J. Hum. Genet. 46: 857-873.
Hatada, I., and T. Mukai (1995). Genomic imprinting of p57(KIP2),
cyclin-dependent kinase inhibitor, in mouse. Nat. Genet. 11: 204-206.
Hayashizaki, Y., Shibata, H., Hirotsune, S., Sugino, H., Okazaki, Y., Sasaki,
N., Hirose, K., Imoto, H., Okuizumi, H., Muramatsu, M., Komatsubara, H.,
Shiroishi, T., Moriwaki, K., Katsuki, M., Hatano, N., Saski, H., Ueda, T., Mise,
N., Takagi, N., Plass, C., and V.M. Chapman (1994). Identification of an
imprinted U2as binding protein related sequence on mouse chromosome 11 using
RLGS method. Nat. Genet. 6: 33-39.
Heslop-Harrison, J.S. (1990). Gene expression and parental dominance in hybrid
plants. Development (suppl.), 21-28.
Holliday, R. (1990). Genomic imprinting and allelic exclusion. Development
Jinno, Y., Yun, K., Nishiwaki, K., Kubota, T., Ogawa, O., Reeve, A., and N.
Niikawa (1994). Mosaic and polymorphic imprinting of the WT1 gene in humans.
Nat. Genet. 6: 305-309.
Kaneko-Ishino, T. (1995). Peg1/Mest imprinted gene on chromosome 6 identified
by cDNA subtraction hybridization. Nat. Genet. 11: 52.
Kay, G.F., Barton, S.C., Surani, M.A., and S. Rastan (1994). Imprinting and X
chromosome couting mechanisms determine Xist expression in early mouse
development. Cell 77: 639-650.
Klar, A.J.S. (1987). Differentiated parental DNA strands confer developmental
asymmetry on daughter cells in fission yeast. Nature 326: 466-470.
Kondo, M. et al. (1995). Frequent loss of imprinting of the H19 gene is
associated with its overexpression in human lung cancers. Oncogene 10:
Leff, S.E., Brannan, C.I., Reed, M.L., Ozcelick, T., Franke, U., Copeland, N.G.
and N. Jenkins (1992). Maternal imprinting of the mouse Snrpn gene and
conserved linkage homology with the human Prader-Willi syndrome region. Nat.
Genet. 2: 259-264.
Li, X. et al. (1995). Expression , promoter usage and imprinting status of
insulin-like growth factor II (IGF2) in human hepatoblastoma: uncoupling of
IGF2 and H19 imprinting. Oncogene 11: 221-229.
Mann, M, Latham, K., and S. Varmuza (1995). Identification of genes showing
altered expression in preimplantation and early postimplantation parthenogenetic
embryos. Dev. Genet. 17: 223-232.
Martin, C.C., and R. McGowan (1995a). Genotype-specific modifiers of transgene
methylation and expression in the zebrafish, Danio rerio. Genet Res Camb 65:
Martin, C.C., and R. McGowan (1995b). Parent-of-origin specific effects on the
methylation of a transgene in the zebrafish, Danio rerio. Developmental
Genetics 17: 233-239.
McGowan, R., Campbell, R., Peterson, A., and C. Sapienza (1989). Cellular
mosaicism in the methylation and expression of hemizygous loci in the mouse.
Genes and Development 3: 1669-1676.
McGrath, J., and D. Solter (1984). Completion of mouse embryogenesis requires
both the maternal and paternal genomes. Cell 37: 179-183.
Metz, C.W. (1938). Chromosome behaviour, inheritance and sex determination in
Sciara. Amer. Nat. 72, 485-520.
Nicholls, R.D. (1994). New insights reveal complex mechanisms involved in
genomic imprinting. Am. J. Hum. Genet. 54: 733-740.
Nur, U. (1990). Heterochromatization and euchromatization of whole genomes in
scale insects (Coccoidea: Homoptera). Development (suppl.), 29-34.
Pedone, P.V. et al. (1994). Mono- and bi-allelic expression of insulin-like
growth factor II gene in human muscle tumors. Human Molecular Genetics 3:
Razin, A. and A.D. Riggs (1980). DNA methylation and gene function. Science
Reik, W., Collick, A., Norris, M.L., Barton, S.C., and Surani, M.A.H. (1987).
Genomic imprinting determines methylation of parental alleles in transgenic
mice. Nature 328: 248-251.
Sanford, J.P., Clark, H.J., Chapman, V.M. and J. Rossant (1987). Differences in
DNA methylation during oogenesis and spermatogenesis and their persistence
during early embryogenesis in the mouse. Genes & Development 1: 1039-1046.
Sapienza, C. (1990). Sex-linked dosage-sensitive modifiers as imprinting genes.
Development Supplement, 107-113.
Sapienza, C., Peterson, A., Rossant, J. and Balling, R. (1987). Degree of
methylation of transgenes is dependent on gamete of origin. Nature 328, 251-254.
Sapienza, C., and J.G. Hall (1995). Genome imprinting in human disease. In:
Scriver C.R., Beaudet, A.L., Sly, W.S., and D. Valle (eds): "The Metabolic and
Molecular Basis of Inherited Disease," 7th ed., McGraw-Hill, New York, New
York. pp 437-458.
Sasaki, H., Allen, N.A., and M.A. Surani (1993). DNA methylation and genomic
imprinting in mammals. In: DNA Methylation: Molecular Biology and Biological
Significance (J.P. Jost and H.P. Saluz, eds.). Birkhauser Verlag, Basel,
Switzerland. pp. 469-486.
Spofford, J.B. (1976). Position effect variagation in Drosophila. In: The
Genetics and Biology of Drosophila (Ashburner, M., Novitski, E., eds). Academic
Press, New York, New York. pp. 955-1018.
Surani, M.A.H., Barton, S.C., and M.L. Norris (1984). Development of
reconstituted mouse eggs suggests imprinting of the genome during gametogenesis.
Nature 308: 548-550.
Surani, M.A.H., Barton, S.C., and M.L. Norris (1986). Nuclear transplantation in
the mouse: Heritable differences between parental genomes after activation of
the embryonic genome. Cell 45: 127-136.
Swain, J.L., Stewart, T.A., and P. Leder (1987). Parental legacy determines
methylation and expression of an autosomal transgene: A molecular mechanism for
parental imprinting. Cell 50: 719-727.
Thomas, J.H. (1995). Genomic imprinting proposed as a surveillance mechanism
for chromosome loss. Proc. Natl. Acad. Sci. U.S.A. 92: 480-482.
VandeBerg, T.L., Robinson, E.S., Samallow, P.B., and P.G. Johnston (1987).
X-linked gene expression and X-chromosome inactivation: Marsupials, mouse and
man compared. In: Isozymes: Current Topics in Biological and Medical
Research (ed. C.L. Markert), Alan R. Liss, New York, New York.
Varmuza, S., and M. Mann (1994). Genomic imprinting - defusing the ovarian time
bomb. Trends Genet. 10: 118-123.
Villar, A.J., and R.A. Pederson (1994). Parental imprinting of the MAS
protocogene in mouse. Nat. Genet. 8: 373-379.
Wevrick, R., Kerns, J.A., and U. Francke (1994). Identification of a novel
paternally expressed gene in the Prader-Willi syndrome region. Hum. Mol. Genet.
Whitt, G.S., Phillip, D.P., and W.F. Childers (1977). Aberrant gene expression
during the development of hybrid sunfishes (Perciformes, Teleosti).
Differentiation 9: 97-109.