In
an article, Ethan Watters expresses the epigenetic changes on DNA without
changing a single base pair. The article describes a study done on fat yellow
mice. The agouti gene was responsible for obesity and yellow fur, making the
mice prone to cancer and diabetes (1). The
offspring of the agouti mice were identical to their parents; large yellow “pincushions”
with a short life expectancy (1). With this in mind, researchers Jirtle and
Waterland designed a genetic experiment involving the pregnant agouti mice’s
diets. In the test group, they fed the mothers a diet that was rich in methyl
donors. The methyl groups can attach to a gene and “silence” it. After the
mothers gave birth, the pups were small and brown. The methyl groups had
entered the embryo’s chromosomes and effectively silenced the agouti gene. Now
here comes the big question: what do yellow fat mice have to do anything with
cancer?
Cancer
is the end result of many events that lead to tumors. Damaged DNA and mutations
are involved in the development of a tumor. For cancers that are associated
with a mutated gene, there is some hope. By methylating DNA (or inhibiting
methylation) that codes for the defective gene, we can either silence the gene
or contain tumor growth. Jirtle and Waterland’s experiment claims that
genistein is linked to cancer chemoprevention and low levels of adipose desposition
(2). Genistein is found in soy, which was included in the mothers’ diets (250
mg/kg). The researchers had hypothesized that there will be hypermethylation on
one of the six cytosine-guanine sites upstream transcritption of the Agouti
gene.
The
researchers had obtained virgin female mice that were assured to be
heterozygous for the Agouti gene through 200 generations of inbreeding. It is
believed that the Avy (Agouti) gene encodes for a single molecule
that affects the follicular melanocytes that results in a yellow phaeomelanin
pigment instead of a black eumelamin pigment (2). They fed the mice the
250mg/kg diet of genistein for about 2 weeks before mating with another heterozygous
male. DNA was collected on two
occasions: Day 21 when the offspring were still in the womb and Day 150 when
the offspring were already born and were growing. DNA samples were obtained at
Day 21 through tail samples and Day 150 through tail, liver, brain and kidney
samples. The researchers had classified the offspring by coat color, ranging
from yellow (<5% brown) to pseduoagouti (>95% brown). To determine which
site of cytosine-guanine sites, they amplified the regions in interest with PCR
and ran a 1.5% agarose gel (2).
Jirlte
and Waterland found a significant shirt in coat color and body weight with the
Genistein diet. The coat color was correlated to the hypermethylation of the
gene. Hypermethylation had decreased the expression of the Agouti gene which
protected the offspring from “adult-onset obesity” (2). The gel electrophoresis
revealed that site 4 had the most increased methylation that contributed to the
decreased production of the yellow phaeomelanin pigment to coat color. However,
Jirtle and Waterland noted in DNA collected at Day 21, hypermethylation was
found in all 3 germ layer tissues (ectoderm, mesoderm and endoderm), indicating
the change was early in embryonic development (2). The mothers’ appetite and
fur color did not change throughout the experiment. However some genes like
insulin growth factor 2 (IGF2) that increase the risks of developing colon
cancer were inherited or induced in the womb (2). If IGF2 gene was silenced by
introducing methyl groups to the embryo, it may prevent the development of
colon cancer.
In
the experiment, Jirtle and Waterland used both heterozygous (Avy/A)
parents to produce offspring. Would it be a factor if the offspring differed in
genotypes? Would the expression of the Agouti gene differ from a heterozygous
and a homozygous? Would the data change if the offspring with (A/A) were
considered, since they do not have the Agouti gene to begin with? Another
concern is the DNA data collection. At Day 21, only tail samples were
collected. Is the data enough to compare with Day 150 data of tail, liver…etc? The
researchers also mentioned that genistein is not a methyl donar, rather it
plays another role that leads to hypermethylation (since hypermethylation was
observed at site 4) (2). There are concerns about methylating DNA. If methyl
groups can silence “bad” genes, will it also silence “good” genes? It is
crucial for the baby to undergo normal development. If methyl groups were to
silence a gene that codes for the development of the eyes… well, that can’t be
good. (Dramatic music)
For the Cancer Project, I wanted to
concentrate on the prognosis of certain cancers by methylation patterns as
biomarkers. I plan to research on specific methyl groups that target cancerous cells
that can aid in prognosis. In one study by Szyf, DNA methylation was able to
knock out one of the enzymes in a pathway that was known to lead to tumors,
effectively preventing tumor development (1). Another study on the effects of green
tea by Fang and colleagues prevented the methylation of cancer fighting genes
and the growth of cancer in animal subjects (1). Methylation of DNA can also be
used as biomarkers; mutated genes will have different methylating patterns that
can allow the prognosis of certain cancers (1). This new tool is helpful with
diagnosis of difficult to detect cancers such as pancreatic cancer.
Reference Articles
Watters, E. (2006) DNA is Not Destiny. Discover Magazine. November Issue, page 32-39
Jirtle, R.L. (2006) Maternal Geinstein Alters Coat Color and Protects Avy Mouse Offspring from Obesity by Modifying the Fetal Epigenome. Envoronmental Health Perspectives. v114, pages 567-572