Download Inherited variation at the epigenetic level: paramutation from the

Available online at www.sciencedirect.com
Inherited variation at the epigenetic level: paramutation from the
plant to the mouse
François Cuzin1,2, Valérie Grandjean1,2 and Minoo Rassoulzadegan1,2
In contrast with a wide definition of the ‘epigenetic variation’,
including all changes in gene expression that do not result from
the alteration of the gene structure, a more restricted class had
been defined, initially in plants, under the name ‘paramutation’.
It corresponds to epigenetic modifications distinct from the
regulatory interactions of the cell differentiation pathways,
mitotically stable and sexually transmitted with non-Mendelian
patterns. This class of epigenetic changes appeared for some
time restricted to the plant world, but examples progressively
accumulated of epigenetic inheritance in organisms ranging
from mice to humans. Occurrence of paramutation in the
mouse and possible mechanisms were then established in the
paradigmatic case of a mutant phenotype maintained and
hereditarily transmitted by wild-type homozygotes.
Together with the recent findings in plants indicative of a
necessary step of RNA amplification in the reference maize
paramutation, the mouse studies point to a new role of RNA, as
an inducer and hereditary determinant of epigenetic variation.
Given the known presence of a wide range of RNAs in human
spermatozoa, as well as a number of unexplained cases of
familial disease predisposition and transgenerational
maintenance, speculations can be extended to possible
roles of RNA-mediated inheritance in human biology and
pathology.
Addresses
1
Inserm U636, F-06108 Nice, France
2
Université de Nice-Sophia Antipolis, Laboratoire de Génétique du
Développement Normal et Pathologique, F-06108 Nice, France
Corresponding author: Cuzin, François ([email protected])
Current Opinion in Genetics & Development 2008, 18:193–196
This review comes from a themed issue on
Chromosomes and expression mechanisms
Edited by Sarah Elgin and Moshe Yaniv
Available online 15th February 2008
0959-437X/$ – see front matter
# 2008 Elsevier Ltd. All rights reserved.
DOI 10.1016/j.gde.2007.12.004
been, however, noted. As it is often the case, they were
first considered somewhat secondary aspects, better left
aside ‘for further studies.’ And at some point, they
reached the front. One such event was the discovery
by McClintock in the 1940s of genetic transposition in
maize, a notion that was to be generalized only much later
to a variety of organisms. Again in maize and at about the
same time, cases of non-Mendelian inheritance were
reported under the name paramutation [1]. Initially,
the notion was, at best, considered an interesting
peculiarity of the plant. With time, significant advances
were, however, registered, especially with the development of the present-day powerful molecular techniques.
And more recently, inheritance of epigenetic variations
was observed in the mouse, the premier experimental
model of mammalian genetics [2].
Further studies will tell us to what extent the mechanisms
in the plant and the mouse are alike. Real or not, some
differences may be noted. For instance, the well-documented instances of paramutation in plants [3] correspond
to either partial or total gene silencing, and this is also the
case of somatic epigenetic regulations, such as the transcriptional inactivation of the mammalian X chromosome
[4] and of heterochromatin regions of the genome [5]. By
contrast, the first few cases of mouse paramutation rather
appear to correspond to the upregulation of gene expression. However, the mouse and the plant share significant
common features. Paramutation occurs in both the cases as
a consequence of interallelic talk. It modifies the expression of a gene, the ‘paramutable’ one, when segregated
from a heterozygote parent carrying a distinct allelic form,
the ‘paramutagenic’ one. Unlike a typical mutation, paramutation results in quantitatively variable phenotypes,
which led to the concept of a ‘rheostat’ rather than an
‘on-off’ switch of gene expression [6]. But the most
remarkable property common to the plant and the animal
is to involve RNA molecules as determinants of the epigenetic variation, not only at the somatic level, as it is the
case of X chromosome inactivation and gene imprinting
[4,5,7], but also as carriers of a hereditary information.
A brief history of paramutation
Paramutation: a formal definition
Owing to its intrinsic beauty and unique operational
power, Mendelian genetics dominated 20th-century
biology and is still at the core of biomedical science.
Inheritance of the mutations – modifications of the
primary genetic text – is determined by the strict rules
of the chromosome ballet at meiosis. Exceptions had
www.sciencedirect.com
1. Discovery in plants. The thought provoking paper by
RA Brink [1] started with the statement ‘‘Departures
from the law [of Mendel] have been reported from
time to time, only to prove unfounded, in most
instances at least, on close analysis. The present study
is concerned with a seeming exception to the rule. . ..’’
Current Opinion in Genetics & Development 2008, 18:193–196
194 Chromosomes and expression mechanisms
The ‘seeming exception’ was that the plant color
determined by one particular allele was modified in
segregants from heterozygote parents with defined
allelic combinations. The altered form of the gene was
maintained through the subsequent generations but
eventually reversed to the original state. The author
noted that this phenomenon was reminding of
reversible and hereditary changes of phenotype that
had been described in ascomycetes [8,9] and considered a number of possible mechanisms, including,
‘in the light of McClintock’s recent work’, integration/
excision of extrachromosomal material. The hypothesis of transposon insertion was far from absurd, as it
was much later shown to be the cause of a hereditary
epigenetic variation at the Agouti locus of the mouse
[10]. It was not, however, to be confirmed in this case,
nor in the subsequent observations of unexplained
hereditary variants occurring in various plant species
[3]. The term ‘paramutation’ was coined to describe
this strange phenomenon [11]. It was assumed from
the beginning, and fully confirmed with the progress of
molecular techniques, that the affected genes were not
modified in their primary structure.
2. The plant models. While observations in a number of
species suggested that paramutation is not a rare event
in plants, the most detailed analysis was to be
performed on the maize, especially on the B0 *
paramutation of the B-I locus, characterized by an
extreme stability [3,6]. Epigenetic changes in B-I
expression lead to a lighter color of the plant. The
modified state, B0 *, is induced when the original,
paramutable B-I allele is associated in a heterozygote
with a paramutagenic allele. A tandemly repeated
region 100 kb upstream of B-I is sufficient to induce
the epigenetic modification, even in non-allelic
locations. Transcribed into the non-coding RNA, it
is differentially methylated and shows distinct chromatin structures in the paramutant and the wild-type
genomes. A recent important finding was the identification as an RNA-dependent RNA polymerase of a
gene product required for paramutation (Mop1 for
Mediator Of Paramutation) [12]. A complete picture of
the molecular mechanisms at work is still to be
generated, but it is clear that these convergent results
point to a role of RNA in the establishment and control
of the epigenetic state.
Inheritance of epigenetic states in the mouse
1. Interallelic cross-talk. Several indications of interallelic
cross-talk in the mouse (‘paramutation-like’ effects)
pointed to the possibility of non-Mendelian heredity
comparable to the plant systems. One of them was the
interallelic transfer of methylation patterns established
after heterologous recombination during meiosis [13].
Both at the Rosa26 locus modified by the insertion of a
Current Opinion in Genetics & Development 2008, 18:193–196
LoxP cassette and at the Rxra locus containing a single
LoxP site, Cre expression during meiosis led to
methylation in the LoxP insert. This modification
extended to the neighboring regions of the same
chromosome and, furthermore, was also transferred at
the next generation to the un-recombined allele on the
homologous chromosome. Both cis- and trans-methylation were found stably reproduced over several
generations. Remarkably, they involved both CpG
methylations and the methylation of other di-nucleotides in hereditary stable patterns [14]. Interallelic crosstalk was independently observed at the imprinted
Rasgrf1 locus [15]. A repeated DNA element upstream
of the transcription start is required to establish
methylation and then expression of the active paternal
allele. When this region had been replaced by one that
controls methylation and allele-specific expression of
the imprinted Igf2r gene, paternal inheritance of the
mutated Rasgrf1 locus allowed the expression of the
paternal locus, but it also caused the activation of the
wild type, normally silent maternal allele. Again, transactivation was heritable and reversible.
2. Agouti and Axin-fused: transposon directed epigenetic
states. A distinct instance of transgenerational maintenance of an epigenetic variation was observed at the
Agouti viable yellow (Avy) and the Axin-fused (AxinFu)
loci of the mouse [10]. Variation of the phenotype is
linked at the two loci to the DNA methylation state of
a retrotransposon responsible for their abnormal
expression. The parental phenotype is transmitted
either both maternally and paternally (Avy) or only
maternally (AxonFu). Unlike the typical cases of
paramutation, these variations do not appear, however,
to involve interallelic cross-talk.
3. The Kit* paramutation [2]. Null mutants of the Kit
gene are lethal in the homozygous state, while
heterozygotes with a wild-type allele exhibit characteristic white spots, white tail tips and white feet in the
case of a null mutant created by the insertion of a LacZ
cassette [16]. In the progeny of intercrosses between
heterozygotes or crosses with a wild-type partner, the
fraction that inherited two wild-type (Kit+) alleles,
estimated from the distribution of tail colors, appeared
strangely reduced. Upon analysis, it became apparent
that a Kit+/+ offspring was indeed generated but that
most of it maintained the mutant phenotype.
Furthermore, these genotypically wild-type mice
efficiently transmitted their white spots to their
progeny. Molecular analysis of Kit expression indicated
a complex situation. Levels of transcripts were reduced
owing to post-transcriptional cleavage. At the same
time, however, the overall rate of transcription was
increased, in all somatic cells and most strikingly, in the
testicular post-meiotic germ cells in which the gene is
normally silent. In crosses with normal wild-type
partners, the modified (Kit*) state was transmitted,
both maternally and paternally. Given the apparently
www.sciencedirect.com
Inherited variation at the epigenetic level: paramutation from the plant to the mouse Cuzin, Grandjean and Rassoulzadegan 195
simple structure of the sperm genome, paternal
transmission raised the question of the signal that
induces paramutation in the embryo. Suggested by the
unusual amounts of RNA in the sperm of heterozygotes
and Kit* males, its crucial role was established by
microinjection in wild-type one-cell embryos. Injecting
RNA from somatic and germ cells of heterozygotes
resulted in a high frequency of establishment of the
epigenetic change, while RNA of a normal mouse
exhibited only a low efficiency. Another observation of
interest was that two microRNAs with sequences
partially complementary to that of Kit RNA (miR-221
and -222) induced the paramutated state. A variety of
other microRNAs tested in the same way were fully
negative as far as Kit expression was concerned (our
unpublished results). On the contrary, microRNA levels
were not modified in the paramutants, thus making
unlikely that the modified expression of the gene was
due to one of their known activities in post-transcriptional regulation, a point that will be discussed in a
subsequent section.
Other instances of paramutation in the mouse
In order to extend the panel of phenotypes possibly
altered by paramutation, a series of tissue- and organspecific microRNAs were injected in normal 1-cell
embryos. Heritable epigenetic changes of distinct phenotypes were so far observed in two cases, miR-1 and
miR-124. The interest of the ‘miR-1* paramutation’ is
that it recapitulates several of the characteristics of a
known, serious human pathology—moreover one that is
known for its familial occurrence with complex patterns
not amenable to a simple Mendelian scheme (K Wagner,
N Wagner, H Ghanbarian, P Gounon, VG, FC & MR,
manuscript in preparation). The microRNA is normally
expressed in myocytes, both in the heart and the skeletal
muscles. Mice born after the microinjection of miR-1 in
one-cell embryos showed much enlarged hearts (hypertrophic cardiomyopathy), a condition that in humans may
lead to premature death. In the ‘miR-1* paramutants’, it
resulted from increased expression of the RNA encoding
the Cdk9 kinase, the main effector of normal and pathological development of the heart [17]. Heart hypertrophy
was efficiently inherited for several generations, both
paternally and maternally, with a variable quantitative
expression. RNA was carried in the spermatozoon head,
including minute amounts of miR-1 sequences. Among
other hereditary epigenetic variations generated by the
same strategy, families derived after the microinjection of
the brain-specific microRNA miR-124 and showing an
accelerated growth starting at a very early developmental
stage are currently under study.
A role of paramutation in human heredity?
Among many tantalizing questions for which there is no
answer yet is the possibility that epigenetic heredity
might play a significant role in humans. A familial mode
www.sciencedirect.com
of hereditary transmission of a disease-prone condition is
conceivable on the model of the mouse paramutation, the
closest one so far being the miR-1-induced cardiac hypertrophy. There is also little doubt that it will be extremely
difficult to demonstrate such a proposition. Beside the
obvious limitations to experimental approaches, the
extreme heterogeneity of our fully outbred genomes will
constitute a considerable limitation. A handful of observations are nevertheless pointing to that possibility.
Transfer of RNA by the male gamete has been demonstrated (reviewed in reference [18]). Human sperm contains in fact much greater amounts of RNA than that of a
‘normal’ laboratory mouse (P. Gounon, VG, MR & FC,
unpublished observations). It is tempting to speculate
about these relatively high amounts of RNA being related
to the extensive degree of heterozygosis of our genomes
compared to those of the inbred laboratory strains. On the
contrary, expression of a pathological trait dependent on
the heterozygote structure in the parental generation was
observed for the insulin VLTR gene in type 1 diabetes
[19]. Several recently reported instances of heritable
epimutations, reviewed and discussed by Whitelaw and
her colleagues [20,21], may or may not belong to the same
class as the typical mouse and plant paramutations.
Indications are also provided by epidemiological studies,
with for instance the familial occurrence of several malignancies for which no clear Mendelian heredity has been
established [22] and the intriguing case of the paternal
transmission of mortality risk ratios over several generations in the ‘Överkalix cohort’ [23]. There is clearly a
long way to go before clear cases of paramutation can be
established beyond any doubt in humans. The hope is
that a more advanced knowledge of the mouse models
may provide in the future useful clues and criteria to
apply to clinical situations.
Toward molecular mechanisms?
Coming back to more immediate issues, the mechanism
leading to the establishment of the paramutated state in
mice is still difficult to understand within the frame of our
current concepts. In spite of the fact that the microinjection
of miR-1 or miR-221 RNAs was a potent inducer, several
characteristics of the epigenetic modification make it difficult to relate them to one of the known activities of
microRNAs. Levels of the microRNAs were not modified
in the somatic and germ cells of Kit* animals, and the
upregulation of gene transcription of the Kit locus in germ
cells [2] cannot be simply explained by a microRNA
effect. Moreover, the epigenetic modification could be
induced by the microinjection of oligoribonucleotides
mimicking segments of the coding sequence. The conditions that lead to paramutation – RNA transfer by sperm,
by microinjection, a parental heterozygous structure with a
structurally modified allele – result in the presence of
abnormal transcripts in the newly formed embryo. It would
then be logical to consider that they are detected by a
highly sensitive surveillance mechanism that initiates a
Current Opinion in Genetics & Development 2008, 18:193–196
196 Chromosomes and expression mechanisms
hereditary program for protecting the embryos by an
augmented expression of the normal allele. The modified
expression thus established is likely to correspond to local
modifications of the chromatin structure in regulatory
elements, possibly at distant locations, as it is the case in
the B0 * paramutation of maize. The identification of these
primary targets would be an important step in our understanding of the mechanisms at work.
Acknowledgements
We are indebted to P Gounon, K Wagner and N Wagner for communicating
the unpublished data and fruitful discussions. Work performed in our
laboratory was made possible by grants from Agence Nationale de la
Recherche and as ‘‘Equipe Labellisée’’ of Ligue Nationale Française
contre le Cancer (France).
References and recommended readings
Papers of particular interest, published within the annual period of
review, have been highlighted as:
of special interest
of outstanding interest
1.
Brink RA: A genetic change associated with the R locus in
maize which is directed and potentially reversible. Genetics
1956, 41:872-879.
Of historical interest as the first clear-cut demonstration that, contrary to
the established ways of thinking of the time, heredity was not to be
completely explained by the Mendelian rules.
2.
Rassoulzadegan M, Grandjean V, Gounon P, Vincent S, Gillot I,
Cuzin F: RNA-mediated non-Mendelian inheritance of
an epigenetic change in the mouse. Nature 2006, 431:
469-474.
The first complete description of a mode of heredity in a mammalian
organism similar in many aspects to the paramutation of the plant geneticists.
3.
Chandler VL, Stam M: Chromatin conversations: mechanisms
and implications of paramutation. Nat Rev Genet 2004,
5:532-544.
4.
Masui O, Heard E: RNA and protein actors in X-chromosome
inactivation. Cold Spring Harb Symp Quant Biol 2006, 71:419-428.
5.
Zaratiegui M, Irvine DV, Martienssen RA: Noncoding RNAs and
gene silencing. Cell 2007, 128:763-776.
6. Chandler VL: Paramutation: RNA-mediated instructions
passed across generations. Cell 2007, 23:641-645.
The most recent review of the experimental evidences for RNA-mediated
inheritance comparing the plant and the mouse systems.
7.
8.
Pauler F, Koerner M, Barlow D: Silencing by imprinted
noncoding RNAs: is transcription the answer? Trends Genet
2007, 23:284-292.
Mitchell MB: Further evidence of aberrant recombination in
Neurospora. Proc Natl Acad Sci U S A 1955, 41:935-937.
Current Opinion in Genetics & Development 2008, 18:193–196
9.
Rizet G: Les phénomènes de barrage chez Podospora
anserina. Rev Cytol Biol Veg 1952, 13:51-92.
10. Blewitt ME, Vickaryous NK, Paldi A, Koseki H, Whitelaw E:
Dynamic reprogramming of DNA methylation at an
epigenetically sensitive allele in mice. PLoS Genet 2006, 2:e49.
11. Brink RA: Paramutation. Annu Rev Genet 1973, 7:129-152.
12. Alleman M, Sidorenko L, McGinnis K, Seshadri V, Dorweiler JE,
White J, Sikkink K, Chandler VL: An RNA-dependent RNA
polymerase is required for paramutation in maize. Nature 2006,
442:295-298.
Establishes the notion that RNA amplification plays a role in a nonMendelian hereditary change in maize.
13. Rassoulzadegan M, Magliano M, Cuzin F: Transvection effects
involving DNA methylation during meiosis in the mouse. EMBO
J 2002, 21:440-450.
14. Grandjean V, Yaman R, Cuzin F, Rassoulzadegan M: Inheritance
of an epigenetic mark: the CpG DNA methyltransferase 1 is
required for de novo establishment of a complex pattern of
non-CpG methylation. Plos One 2007, 2:e1136.
15. Herman H, Lu M, Anggraini M, Sikora A, Chang Y, Yoon BJ,
Soloway PD: Trans allele methylation and paramutation-like
effects in mice. Nat Genet 2003, 34:199-202.
16. Bernex F, De Sepulveda P, Kress C, Elbaz C, Delouis C,
Panthier JJ: Spatial and temporal patterns of c-kit-expressing
cells in WlacZ/+ and WlacZ/WlacZ mouse embryos. Development
1996, 122:3023-3033.
17. Sano M, Abdellatif M, Oh H, Xie M, Bagella L, Giordano A,
Michael LH, DeMayo FJ, Schneider MD: Activation and function
of cyclin T-Cdk9 (positive transcription elongation factor-b) in
cardiac muscle-cell hypertrophy. Nat Med 2002, 8:1310-1317.
18. Miller D, Ostermeier GC, Krawetz SA: The controversy, potential
and roles of spermatozoal RNA. Trends Mol Med 2005,
11:156-163.
An up-to-date summary of the observation of the human sperm RNA and
speculations on its possible functions.
19. Bennett ST, Wilson AJ, Esposito L, Bouzekri N, Undlien DE,
Cucca F, Nistico L, Buzzetti R, Bosi E, Pociot F et al.: Insulin VNTR
allele-specific effect in type 1 diabetes depends on identity of
untransmitted paternal allele. The IMDIAB Group. Nat Genet
1997, 17:350-352.
20. Chong S, Whitelaw E: Epigenetic germline inheritance.
Curr Opin Genet Dev 2004, 14:692-696.
21. Chong S, Yougson NA, Whitelaw E: Heritable germline
epimutation is not the same as transgenerational epigenetic
inheritance. Nat Genet 2007, 39:574-575.
22. Czene K, Lichtenstein P, Hemminki K: Environmental and
heritable causes of cancer among 9.6 million individuals in the
Swedish Family—cancer database. Int J Cancer 2002,
99:260-266.
23. Pembrey M, Bygren L, Kaati G, Edvinsson S, Northstone K,
Sjöström M, Golding J: ALSPAC Study Team: sex-specific,
male-line transgenerational responses in humans. Eur J Hum
Genet 2006, 14:159-166.
www.sciencedirect.com