Interaction between epigenetics and genetics

From: Michael Ragland (ragland37_at_webtv.net)
Date: 08/22/04 

Date: Sun, 22 Aug 2004 00:32:33 +0000 (UTC)

There have been two opposing viewpoints here on S.B.E. recently. One
view is that genes and the environment interact with each other and
therefore it is erroneous to state either the environment or DNA is
dominant in producing and shaping phenotypes. The other view while
acknowledging the interaction between genes and the environment takes
the position DNA ultimately has more influence on phenotype. It would
seem these two positions are irreconciliable with each other. My own
view is supportive of both positions. Upon recently reading an article
on epigenetics [included below] I was struck by the complicated
interaction between the environment and DNA. It certainly seemed to put
a dent in the idea DNA ultimately has more influence on phenotype. The
consensus of the piece was that epigenetics didn't replace Mendelian
genetics but rather the two went hand in hand.

Rather than primarily focusing on epigenetics in humans, however, the
article focused on agouti mice and Drosophilia Melanogastor physical
phenotypes. In agouti mice DNA methylation pattern which produced
different color coats and in heat shock protein Hsp90 appendages
protruding from the eyes of flies.

In the case of humans the Dutch hunger winter during WWII was cited. The
article states, "Detailed birth records collected during that so-called
Dutch Hunger Winter have provided scientists with useful data for
analyzing the long-term health effects of prenatal exposure to famine.
Not only have researchers linked such exposure to a range of
developmental and adult disorders, including low birth weight, diabetes,
obesity, coronary heart disease, breast and other cancers, but at least
one group has also associated exposure with the birth of
smaller-than-normal grandchildren." Needless to say, I don't think it is
a surprise severe environmental trauma on a developing embryo/fetus will
result in genetic deficits and damage.

The article also mentions the importance of epigenetics of the
epidimeology of disease, particularly cancer and negative epigenetic
changes as a result of assisted reproductive technology and how
epigenetic changes present obstacles to human cloning.

According to Vercelli, the environmental susceptibility of epigenetics
probably explains why genetically identical organisms such as twins can
have dramatically different phenotypes in different environments. This
is based on variability in CpG methylation at the agouti locus causes
differences in coat color among genetically identical mice. Maternal
nutrition affects the phenotype of offspring by influencing the degree
of CpG methylation at the agouti locus.

Assuming this can be extrapolated to human twins I would agree. However,
since human maternal twins are almost without exception 100% genetically
identical and share a similar environment the appearance of dramatically
different phenotypes in *different environments* is going to be limited.
Despite methylation and other epigenetics involved in the development of
identical twins most are going to have similar phenotypes. More similar
than say fraternal twins.

At the end the article states, "Not that Mendelian genetics is wrong;
far from it. The increased understanding of epigenetic change and the
recent evidence indicating its role in inheritance and development
doesn't give epigenetics greater importance than DNA. Genetics and
epigenetics go "hand in hand," says Ohlsson. In the case of disease,
says Reik, "there are clearly genetic factors involved, but there are
also other factors involved. My suspicion is that it will be a
combination of genetic and epigenetic factors, as well as environmental
factors, that determine all these diseases."

I've certainly gained a new appreciation of the importance between the
interaction of genes and the environment. Particularly as it relates to
disease and assisted reproductive technology and human cloning. While
the relationship between Mendelian genetics and epigenetics is complex
and varied in its interactions I continue to believe some phenotypes of
our species are dominated by our DNA.

Michael Ragland

Epigenetics: Genome, Meet Your Environment
As the evidence accumulates for epigenetics, researchers reacquire a
taste for Lamarckism | By Leslie A. Pray
©Mehau Kulyk/Photo Researchers, Inc.

Toward the end of World War II, a German-imposed food embargo in western
Holland--a densely populated area already suffering from scarce food
supplies, ruined agricultural lands, and the onset of an unusually harsh
winter--led to the death by starvation of some 30,000 people. Detailed
birth records collected during that so-called Dutch Hunger Winter have
provided scientists with useful data for analyzing the long-term health
effects of prenatal exposure to famine. Not only have researchers linked
such exposure to a range of developmental and adult disorders, including
low birth weight, diabetes, obesity, coronary heart disease, breast and
other cancers, but at least one group has also associated exposure with
the birth of smaller-than-normal grandchildren.1 The finding is
remarkable because it suggests that a pregnant mother's diet can affect
her health in such a way that not only her children but her
grandchildren (and possibly great-grandchildren, etc.) inherit the same
health problems.

In another study, unrelated to the Hunger Winter, researchers correlated
grandparents' prepubertal access to food with diabetes and heart
disease.2 In other words, you are what your grandmother ate. But, wait,
wouldn't that imply what every good biologist knows is practically
scientific heresy: the Lamarckian inheritance of acquired
characteristics?

If agouti mice are any indication, the answer could be yes. The
multicolored rodents make for a fascinating epigenetics story, which
Randy Jirtle and Robert Waterland of Duke University told last summer in
a Molecular and Cell Biology paper; many of the scientists interviewed
for this article still laud and refer to that paper as one of the most
exciting recent findings in the field. The Duke researchers showed that
diet can dramatically alter heritable phenotypic change in agouti mice,
not by changing DNA sequence but by changing the DNA methylation pattern
of the mouse genome.3 "This is going to be just massive," Jirtle says,
"because this is where environment interfaces with genomics."

EPIGENETICS EXPLAINED This type of inheritance, the transmission of
non-DNA sequence information through either meiosis or mitosis, is known
as epigenetic inheritance. From the Greek prefix epi, which means "on"
or "over", epigenetic information modulates gene expression without
modifying actual DNA sequence. DNA methylation patterns are the
longest-studied and best-understood epigenetic markers, although ethyl,
acetyl, phosphoryl, and other modifications of histones, the protein
spools around which DNA winds, are another important source of
epigenetic regulation. The latter presumably influence gene expression
by changing chromatin structure, making it either easier or more
difficult for genes to be activated.

Because a genome can pick up or shed a methyl group much more readily
than it can change its DNA sequence, Jirtle says epigenetic inheritance
provides a "rapid mechanism by which [an organism] can respond to the
environment without having to change its hardware." Epigenetic patterns
are so sensitive to environmental change that, in the case of the agouti
mice, they can dramatically and heritably alter a phenotype in a single
generation. If you liken the genome to the hardware of a computer,
Jirtle explains, then "epigenetics is the software. It's the grey area.
It's just so darn beautiful if you think about it."
Courtesy of Museum Online
 LAMARCK: Jean-Baptiste Lamarck (1744-1829) is best remembered for a
discredited theory of heredity, the "inheritance of acquired traits." He
proposed that environment changes caused changes in behavior which in
turn led to the increase or decrease of particular structures. Lamarck
had a colorful and distinguished career: in turns soldier, bank clerk,
Professor of "insects and worms" he died a poor man and was buried in a
rented grave.

The environmental lability of epigenetic inheritance may not necessarily
bring to mind Lamarckian images of giraffes stretching their necks to
reach the treetops (and then giving birth to progeny with similarly
stretched necks), but it does give researchers reason to reconsider
long-refuted notions about the inheritance of acquired characteristics.
Eighteenth-century French naturalist Jean Baptiste de Lamarck proposed
that environmental cues could cause phenotypic changes transmittable to
offspring. "He had a basically good idea but a bad example," says Rohl
Oflsson, Uppsala University, Sweden.
Although the field of epigenetics as it is known today (that is, the
study of heritable changes in gene expression and regulation that have
little to do with DNA sequence) has been around for only 20 years or so,
the term epigenetics has been in use since at least the early 1940s.
Developmental biologist Conrad Waddington used it back then to refer to
the study of processes by which genotypes give rise to phenotypes (in
contrast to genetics, the study of genotypes). Some reports indicate
that the term is even older than Waddington, dating back to the late
1800s. Either way, early use of the term was in reference to
developmental phenomena.

In 2001, Joshua Lederberg proposed the use of more semantically, or
historically, correct language.4 But it appears that today's use of the
term is here to stay, at least for now, as are its derivatives:
epiallele (genes with different degrees of methylation), epigenome (the
genome-wide pattern of methyl and other epigenetic markers), epigenetic
therapy (drugs that target epigenetic markers), and even epigender (the
sexual identity of a genome based on its imprinting pattern).

Terminology aside, biologists have long entertained the notion that
certain types of
cellular information can be transmitted from one generation to the next,
even as DNA sequences stay the same. Bruce Stillman, director of Cold
Spring Harbor Laboratory (CSHL), NY, traces much of today's research in
epigenetics back to Barbara McClintock's discovery of transposons in
maize. Methyl-rich transposable elements, which constitute over 35% of
the human genome, are considered a classical model for epigenetic
inheritance. Indeed, the epigenetic lability of Jirtle's agouti mice is
due to the presence of a transposon at the 5' end of the agouti gene.
But only over the past two decades has the evidence become strong enough
to convince and attract large numbers of epigenetics researchers.

"[Epigenetics] has very deep roots in biology," says Stillman," but the
last few years have been just an explosion in understanding."
METHYLATION AND MORE One of the prominent features of DNA methylation
is the faithful propagation of its genomic pattern from one cellular or
organismal generation to the next. When a methylated DNA sequence
replicates, only one strand of the next-generation double helix has all
its methyl markers intact; the other strand needs to be remethylated.
According to Massachusetts Institute of Technology biologist Rudy
Jaenisch, the field of epigenetics took its first major step forward
more than two decades ago when, upon discovering DNA methyltransferases
(DMTs, the enzymes that bind methyl groups to cytosine nucleotides),
researchers finally had a genetic handle on how epigenetic information
was passed along. Now, it is generally believed that DMTs bind methyl
groups to the naked cytosines based on the methylation template provided
by the other strand. This is known as the maintenance methylase theory.
© The American Society of Human Genetics

 THE EPIGENOME IS REPROGRAMMED DURING DEVELOPMENT: Erasure of
epigenetic marks, including DNA methylation and genomic imprinting,
occurs as primordial germ cells migrate along the genital ridge.
Reestablishment of mark takes place during gametogenesis, differentially
in sperm (blue) and egg (pink). After fertilization another round of
erasure occurs--apart from imprinted genes (dotted line), which are
protected--followed by tissue-specific patterning. (Reprinted with
permission, Am J Hum Genet, 74:599-609 2004)

But even a decade ago, says Wolf Reik of the Babraham Institute,
Cambridge, UK, "a lot of epigenetics was phenomenology, and so people
looked at it and said, well, this is all very interesting, but what's
the molecular mechanism?" Reik points to recent evidence suggesting a
critical link between the two main types of epigenetic regulation, DNA
methylation and histone modification, as one of the most interesting
recent developments in the field. Because of that link, researcher Eric
Selker and colleagues at the University of Oregon, Eugene, have proposed
that there may be more to methylation propagation than maintenance,
despite 25 years of evidence. In 2001, Selker and coauthor Hisashi
Tamaru showed that dim-5, a gene that encodes a histone H3 Lys-9
methyltransferase, is required for DNA methylation in the filamentous
fungus, Neurospora crassa.5 The histone enzyme is, in turn, influenced
by modifications of histone H3. So even though DNA methylation is guided
by a DNA methyltransferase encoded by dim-2, it still takes orders from
the chromatin.
In a study by CSHL researchers Robert Martienssen, Shiv Grewal, and
colleagues, evidence suggests that histone modifications are, in turn,
guided by RNA interference (RNAi).6 Using the fission yeast
Schizosaccharomyces pombe, the researchers deleted genes that encode
RNAi molecular machinery and observed a loss of histone H3 lys-9
methylation and impaired centromere function. "This new understanding
has created a lot of excitement," says Stillman.

EPIGENETICS AND DISEASE More than two decades ago, anyone who proposed
that epigenetic regulation played a role in carcinogenesis was a "lone
prophet in the desert," explains Jaenisch. Researchers didn't seriously
entertain the notion until Andy Feinberg and Bert Vogelstein, both at
Johns Hopkins University, reported a link between human cancer cells and
aberrant DNA methylation patterns.7 Even then, Feinberg says "the
initial reaction was disbelief. I think that people ignored it. Now,
everyone accepts that epigenetics is important in cancer." The
etiological link between epigenetic change and cancer has fueled both
academic and pharmaceutical interest in the field.

Methylation usually silences gene expression. Normally, about 70% of all
CpG dinucleotides in the mammalian genome are methylated. The remainder,
clusters near the 5' end of genes known as CpG islands, are protected
from it. Too little methylation across the genome or too much
methylation in the CpG islands can cause problems, the former by
activating nearby oncogenes, and the latter by silencing tumor
suppressor genes. When Feinberg and Vogelstein linked cancer to
epigenetics in the early 1980s, they linked it specifically to
genome-wide hypomethylation. A few years later, German and US research
teams discovered connections between cancer and tumor- suppressing
silencing caused by hypermethylation. Both hypo- and hypermethylation
can play significant regulatory roles even in the same tumor.
© 2003 American Society for Microbiology

 SAME GENOME, DIFFERENT EPIGENOME: Variability in CpG methylation at
the agouti locus causes differences in coat color among genetically
identical mice. Maternal nutrition affects the phenotype of offspring by
influencing the degree of CpG methylation at the agouti locus.
(Reprinted with permission, Molec Cell Biol, Aug 2003)

It has taken more than correlations between methylation and cancer,
however, to convince researchers that epigenetics is the cause, not
consequence, of malignancy. Feinberg points to two pieces of evidence
that have pushed epigenetics to the fore. First, several independent
observations of epigenetic aberrations (specifically, impaired
methylation patterns) in normal cells surrounding tumorous tissue
suggest that epigenetic abnormalities are not simply an epiphenomenon of
the cancer phenotype, as has been argued. But the real "smoking gun for
epigenetics," says Feinberg, has been the detection of a clear causal
link between Beckwith-Wiedemann syndrome (BWS) and a particular cluster
of imprinted genes, which include the insulin-like growth factor II
gene, Igf2.

Imprinting is the differential methyl tagging and expression of genes
depending on whether they came from the mother or father. Igf2 is one of
the best characterized imprinted genes: It is turned off on the maternal
chromosome (i.e., it is silenced by methylation) so that only its
paternal copy is expressed. But in the case of BWS, a rare birth defect,
Igf2 is biallelically expressed. It has been suggested that the double
dosage of Igf2 does its damage by inhibiting apoptosis. Babies born
with BWS are more likely to develop macroglossia (enlarged tongue),
abdominal wall defects, and various types of malignant tumors.

The etiological role of epigenetics in tumor formation has prompted
efforts to create antitumor drugs that correct disrupted epigenetic
inheritance. So far, says Jaenisch, no one has succeeded, although the
Food and Drug Administration approved the epigenetic inhibitor
azacitidine on May 19, for the treatment of the bone disorder
myelodysplastic syndrome. The drug reportedly turns on genes that have
been silenced by epigenetic methylation.

Epigenetic inheritance has been associated with a number of other human
health conditions, including some whose incidence is higher among babies
born with the aid of assisted reproduction technology (ART). As Reik
explains, embryos normally develop in a protective environment, the
womb. When they are put into the suboptimal environment of a culture
dish, many things can go wrong. Methylation sites initially established
in the oocyte may not be maintained properly, and imprinting patterns
may be lost during development. Individuals conceived by ART techniques
have a higher risk of being born with BWS, Angelman syndrome (AS), and
retinoblastoma (a tumor of the retina). Like BWS, AS has been linked to
imprinting errors. Typical features of babies born with AS include
developmental delay, absent speech development, and seizures.

Epigenetic inheritance also may be the reason that human cloning is all
but impossible. Indeed, Jaenisch considers cloning "the ultimate
bioassay for epigenetic changes." When a differentiated somatic cell is
put into an oocyte, its genome-wide epigenetic pattern must be
reprogrammed in order to restore totipotency. The difficulties
associated with reprogramming all the chromatin, histones, and
methylation patterns along the entire length of the DNA sequence may
explain why so many cloned embryos have so many developmental failures.

INHERITANCE OF METHYLATION STATES: Replication of methylated DNA…
Click to view enlarged diagram (PDF, 39K)

LAMARCKISM REVISITED Normally, the fur of agouti mice is yellow, brown,
or a calico-like mixture of the two, depending on the number of attached
methyl groups. But when Duke University researchers Jirtle and Waterland
fed folic acid and other methyl-rich supplements to pregnant mothers,
despite the fact that all offspring inherited exactly the same agouti
gene (i.e., with no nucleotide differences), mice who received
supplements had offspring with mostly brown fur, whereas mice without
supplements gave birth to mostly yellow pups with a higher
susceptibility to obesity, diabetes, and cancer. The methyl groups bound
to a transposon at the 5' end of the agouti locus, thereby shutting off
expression of the agouti gene, not just in the murine recipient but in
its offspring as well.

Although the study demonstrates that, at least in mice, folic acid
supplementation in pregnant mothers reduces the risk of their babies
having certain health problems, Jirtle warns that the results cannot be
extrapolated to humans. "Mice are not men," he emphasizes. But he
doesn't downplay the proof of principle. The take-home message is not
that folic acid supplements are a good thing. Rather, environmental
factors such as nutritional supplementation can have a dramatic impact
on inheritance, not by changing the DNA sequence of a gene or via
single-nucleotide polymorphism, but by changing the methylation pattern
of that gene. "It's a proof of concept," says Donata Vercelli,
University of Arizona, Tucson. "That's why it's so important."
According to Vercelli, the environmental susceptibility of epigenetics
probably explains why genetically identical organisms such as twins can
have dramatically different phenotypes in different environments. She
points to the agouti mice, as well as another recent cluster of studies
on a heat shock protein, Hsp90, in Drosophila melanogaster, as "model
systems that have very eloquently demonstrated" the critically important
role that epigenetic inheritance plays in this kind of
gene-by-environment interaction.

Hsp90 regulates developmental genes during times of stress by releasing
previously hidden or buffered phenotypic variation. Douglas Ruden of the
University of Alabama, Tuscaloosa, says he noticed some weird fruit fly
phenotypes--things like appendage-like organs sticking out of their
eyes--at about the same time that a paper appeared in Nature connecting
Hsp90 activity in Drosophila to genetic variation.8 In that paper,
Suzanne Rutherford and Susan Lindquist, then at the University of
Chicago, presented compelling evidence that Hsp90 serves as an
"evolutionary capacitor," a genetic factor that regulates phenotypic
expression by unleashing "hidden" variation in stressful conditions.8
Even after restoring normal Hsp90 activity, the new phenotypes responded
to ten or more generations of selection. The scientists concluded that,
once released, even after normal Hsp90 activity was restored, the
previously buffered variation persisted in a heritable manner,
generation after generation.
When the Lindquist paper came out, Ruden says he thought, "Ah, I'm
probably seeing the same thing." He was doing some crosses, "and I
started to see this weird phenotype." But Ruden and collaborators
concluded that their strange eye phenotype was due to something other
than, or in addition to, the sudden unleashing of hidden genetic
variation.9 Indeed, the researchers used a strain of flies that had
little genetic variation, and yet was still capable of responding to 13
generations of selection even after normal Hsp90 activity was restored.
Because of the genomic homogeneity of their flies, combined with
observations that mutations encoding chromatin-remodeling proteins
induced the same abnormal eye phenotype, the investigators concluded
that reduced levels of Hsp90 affected the phenotype by epigenetically
altering the chromatin.
Courtesy of Douglas Ruden

 MORPHOLOGICAL EVOLUTION THROUGH AN EPIGENETIC MECHANISM: Light
photomicrograph of the head of a fruitfly with epigenetically-induced
ectopic bristles in the eyes.

Although it is hard to imagine that an appendage-like structure sticking
out of the eye would be adaptive in times of stress, Vercelli says that
epigenetic change clearly can be environmentally induced in a heritable
manner, in this case by alterations to Hsp90. The morphological
variations in the eye were probably only the most obvious of many
phenotypic differences caused by the chromatin changes.

In a written commentary, evolutionary biologist Massimo Pigliucci said
that Ruden's experiment was "one of the most convincing pieces of
evidence that epigenetic variation is far from being a curious nuisance
to evolutionary biologists."10 Pigluicci doesn't go so far as to say
that the heritable changes caused by Hsp90 alterations are Lamarckian,
but Ruden does. "Epigenetics has always been Lamarckian. I really don't
think there's any controversy," he says.

Not that Mendelian genetics is wrong; far from it. The increased
understanding of epigenetic change and the recent evidence indicating
its role in inheritance and development doesn't give epigenetics greater
importance than DNA. Genetics and epigenetics go "hand in hand," says
Ohlsson. In the case of disease, says Reik, "there are clearly genetic
factors involved, but there are also other factors involved. My
suspicion is that it will be a combination of genetic and epigenetic
factors, as well as environmental factors, that determine all these
diseases."

Leslie Pray (lpray@nasw.org) is a freelance writer in Holyoke, Mass.
References
1. L.H. Lumey, "Decreased birthweights in infants after maternal in
utero exposure to the Dutch famine of 1944-1945," Paediatr Perinat Ep,
6:240-53, 1992.
2. G. Kaati et al., "Cardiovascular and diabetes mortality determined by
nutrition during parents' and grandparents' slow growth period," Eur J
Hum Genet, 10:682-8, 2002.
3. R.A. Waterland, R.A. Jirtle, "Transposable elements: targets for
early nutritional effects on epigenetic gene regulation," Mol Cell Biol,
23:5293-300, 2003.
4. J. Lederberg, "The meaning of epigenetics," The Scientist, 15[18]:6,
Sept. 17, 2001.
5. H. Tamaru, E.U. Selker, "A histone H3 methyltransferase controls DNA
methylation in Neurospora crassa," Nature, 414:277-83, 2001.
6. T.A. Volpe et al., "Regulation of heterochromatic silencing and
histone H3 lysine-9 methylation by RNAi," Science, 297:1833-7, 2002.
7. A.P. Feinberg, B. Vogelstein, "Hypomethylation distinguishes genes of
some human cancers from their normal counterparts," Nature, 301:89-92,
1983.
8. S.L. Rutherford, S. Lindquist, "Hsp90 as a capacitor for
morphological evolution," Nature, 396:336-42, 1998.
9. V. Sollars et al., "Evidence for an epigenetic mechanism by which
Hsp90 acts as a capacitor for morphological evolution," Nat Genet,
33:70-4, 2003.
10. M. Pigliucci, "Epigenetics is back!" Cell Cycle, 2:34-5, 2003.

"It's uncertain whether intelligence has any long term survival value.
Bacteria do quite well without it."
 Stephen Hawking