 | |
The term epigenetic changes of chromatin is used to describe changes in the
regulation of gene activity and expression, that are not dependent on gene sequence.
The most studied epigenetic changes are histone post-translational modifications and
DNA methylation. These studies have shown that epigenetic changes play an
important role in a broad range of biological functions and also that their deregulation
is involved in many disease states. Moreover, a large body of data has correlated
epigenetic changes with the aging process. However, the mode by which these
changes exert their effects on gene expression and their mechanisms of action still
need further clarification.
The first part of the present study was focused on the epigenetic changes of chromatin
that occur in age-related genes. The first candidate was H1.0, a gene that encodes for
a DNA linker histone variant that has been shown to be related to both in vivo and in
vitro aging. In the present study, H1.0 mRNA leve ...
Όλα τα τεκμήρια στο ΕΑΔΔ προστατεύονται από πνευματικά δικαιώματα.
The term epigenetic changes of chromatin is used to describe changes in the
regulation of gene activity and expression, that are not dependent on gene sequence.
The most studied epigenetic changes are histone post-translational modifications and
DNA methylation. These studies have shown that epigenetic changes play an
important role in a broad range of biological functions and also that their deregulation
is involved in many disease states. Moreover, a large body of data has correlated
epigenetic changes with the aging process. However, the mode by which these
changes exert their effects on gene expression and their mechanisms of action still
need further clarification.
The first part of the present study was focused on the epigenetic changes of chromatin
that occur in age-related genes. The first candidate was H1.0, a gene that encodes for
a DNA linker histone variant that has been shown to be related to both in vivo and in
vitro aging. In the present study, H1.0 mRNA levels were found to be elevated in both
Go phase and PHA-activated lymphocytes of old donors as compared to young. These
observations are in agreement with results for the levels of histone acetylation that
were found to be higher in the old donor samples as compared to those of the young
samples, for the entire H1.0 gene region and especially around the promoter region of
the gene. Additionally, the same histone acetylation changes were also found in
fibroblasts from young and old donors. Histone H3K4 trimethylation didn’t seem to
have important age-related changes in lymphocytes, mainly because the differences
observed didn’t correlate with the H1.0 mRNA levels. Lastly, neither H1.0 mRNA
levels nor changes in histone modification levels (acetylation and H3K4me3) were
found to occur as a function of age in the third aging cell system used in this study,
i.e., monocytes and in vitro differentiated dendritic cells.
The second gene that was analyzed was DFNA5, a gene whose expression levels were
found to change during development from previous studies in mice. In the present
thesis work, microarray and real time PCR analyses of monocytes and in vitro
differentiated dendritic cells from donors of different age groups showed that this
gene’s mRNA levels change, both as a function of differentiation of monocytes to
dendritic cells, and as a function of age. More particularly, DFNA5 expression levels
were found to be higher in dendritic cells as compared to monocytes. With respect to
age, DFNA5 expression levels were higher in dendritic cells of newborns as compared
213
to the dendritic cells of adult donors. The study of histone post-translational
modifications revealed that there are age-related epigenetic changes around the
DFNA5 gene region, but different to those observed in the H1.0 gene. H3K4
trimethylation (Η3Κ4me3), a histone modification associated with active genes, was
found to be higher in the dendritic cells derived from newborns as compared to those
from adult donors. On the contrary, H3K27 trimethylation, an epigenetic mark of
silenced genes, was higher in the adult samples. These observations concur with the
age-related differences in the expression levels of DFNA5 .
Both genes, H1.0 and DFNA5, were observed to have age-related chromatin
epigenetic changes which are consistent with the changes found in their mRNA
levels. However, the type of histone epigenetic change that occurred as a function of
age differed amongst the two genes. In the case of H1.0, histone acetylation was
found to be the major determinant in the gene’s expression levels, whereas histone
methylation levels were central to the DFNA5 gene. If one takes into account that
there are major structural differences between the two genes, then specific sequence
characteristics may dictate specific types of epigenetic changes during aging. This
may mean that there are specific DNA sequence structures more susceptible to agerelated
epigenetic changes.
In line with this latter hypothesis, we continued the present thesis work in the
direction of finding specific DNA sequence characteristics and structures which may
be more susceptible to chromatin epigenetic structural changes during aging. In order
to carry out this line of work, a whole genome scale technique and a well-studied
epigenetic mark were chosen. DNA methylation was chosen, since it is an epigenetic
mark that has been well-studied during the past years and has also been linked to
development and aging. Methylated DNA immunoprecipitation was performed,
followed by analysis of the immunoprecipitated DNA with next generation
sequencing (MeDIP-Seq). This technique allows one to study the methylome on a
whole-genome scale without bias or focus on specific gene regions.
The results presented in this thesis work were from 3 different experiments. In each
experiment, comparisons were carried out between two age groups, newborns and
adults, with pooled DNA samples (10 donors from each age-group), so as to avoid
differences due to individual variation. In the first experiment DNA was pooled from
monocytes of each age-group. In the second and third experiments, DNA was pooled
214
from both monocytes and dendritic cells, respectively, from 10 donors of each agegroup.
Initially, macroscopic analysis of the first two experiments proved the credibility of
the technique that was used, but didn’t reveal any wide-range age-related differences
in the methylome. Next, the whole genome was scanned at a better resolution level
(scanning window: 100kb, 2kb etc.). The results from this line of work showed that
349 regions appeared to have at least a 1.5 fold difference between newborns and
adults. The main characteristic of these regions was that they were GC-rich and
contained multiple small CpG islands (smaller than usually described) which we
named CpG “micro-clusters”. Further analysis of the results revealed another
sequence characteristic susceptible to age-related changes in methylation, regions rich
in sequences that are at least partially repeated more than once in the genome (not
classic repeat sequences) and are excluded by the repeat masker program (Multiple
Aligned Repeat-Masker eXcluded sequences, MAReX). From the analysis of all three
experiments, a number of genes contained, or were adjacent to, MAReX sequences
and also had age-related differences. These genes were members of the PCDHG and
FAM90A gene families and the HRNR and PROP1 genes. Finally, we analyzed genes
with at least 2 fold age-related differences in DNA methylation that had all of the
above characteristics. These genes were FZD1, FZD7, FGF17 and the pseudogene
ECEL1P2. All of these genes appeared to be implicated in cell cycle-related events
and early mammalian development. These sites were also enriched in H3K27
trimethylation at exactly the same positions as those where age-related DNA
methylation differences were observed. However, in this case the age-related changes
were in opposite directions.
The conclusions drawn from the second part of this thesis work are that there are
some structures in the genome more susceptible to age-related epigenetic changes.
These structures seem to be regions rich in CpG “micro-clusters” and MAReX
sequences.
