| dc.description.abstract | Mitochondria are semiautonomous organelles in a cell as they have their own
genome that replicates independently. It plays a major role in oxidative phosphorylation due
to which mitochondrial DNA (mtDNA) is frequently exposed to oxidative damages. Factors
such as ionizing radiation, radiomimetic drugs and replication fork stalling can lead to point
mutations, deletions and other rearrangements resulting in mitochondrial genome fragility.
Since a distinct feature of the mitochondrial genome is the absence of noncoding regions,
mutations can severely affect the mitochondrial functions. In many cases, deletions in
human mtDNA are flanked by short direct repeats; however, the mechanism by which they
arise is still unknown. Mitochondria from patient samples show frequent genomic aberrations
such as point mutations, insertions and large-scale deletions that could possibly account for
mitochondria associated disease pathogenesis including cancer.
Previous studies have shown the association of several mitochondrial mutations and
deletions with ageing and human disorders such as myopathies, dystonia and hepatocellular
carcinoma. Bioinformatic analysis revealed that among the deletion breakpoints from
patients with mitochondrial disorders, majority (87%) are located at G4 DNA motifs.
Interestingly, among this, ~50% of the break points were due to a deletion at base pair
position 8271-8281, ~35% were due to deletion at 12362-12384 and ~12% due to deletion at
15516-15545. In first part of my thesis, I have investigated the molecular basis for the
occurrence of mitochondrial DNA deletions.
Firstly, different non-B DNA structure forming motifs were investigated in
mitochondrial genome by using non-B DNA prediction tools such as non-B DB database and
QGRS mapper. Results revealed the presence of five G-quadruplex forming motifs and
several inverted and direct repeats. Formation of G-quadruplex DNA structures at the
mitochondrial fragile regions were characterized by using various biochemical assays like
electromobility shift assay, primer extension, Taq polymerase stop assay, DMS protection
assay and biophysical assays such as circular dichroism (CD). Interestingly, the site of 9 bp
2
deletions (8271-8281 bp) mapped to one of the G-quadruplex motifs (Region I) and two
direct repeats. Detailed investigation revealed formation of intramolecular parallel Gquadruplex
in presence of KCl when the region was present on a shorter DNA. Bisulfite
modification assay when performed on a plasmid harboring Region I suggests the single
strandedness of this region. Consistent results were also observed when mitochondrial
genome was analyzed for non-B DNA structures by bisulfite modification assay. Further,
immunocolocalization of BG4, an antibody that binds to G-quadruplexes with mitochondria in
conjunction with ChIP assay using BG4, confirmed the formation of G-quadruplex in the
mitochondrial genome.
Mitochondrial extracts isolated from the rat tissues were used for investigating the
cleavage activity at Region I of mitochondrial genome. Strong pause sites were observed
when the primer extension was performed and mapping of this cleavage site with the help of
DNA sequencing ladder as marker, confirmed that the pause sites matched with the position
corresponding to G-quadruplex forming region.
Next, an endonuclease screening assay was performed to investigate the enzyme
that is responsible for induction of breakage at mitochondrial region I. It was observed that
purified Endonuclease G (Endo G), a mitochondrial nuclease, can cleave mitochondrial DNA
at a site corresponding to 9 bp deletion. However, incubation with endonucleases such as
CtIP, MRE11, FEN1, RAGs did not result in any significant cleavage. Importantly, a mutation
at the G-quadruplex motif abrogated cleavage by Endonuclease G. Immunodepletion of
Endonuclease G in mitochondrial protein extracts and shRNA mediated knockdown of the
same within human cells significantly reduced efficiency of cleavage at the mitochondrial Gquadruplex
Region I. Further, immunofluorescence was performed to understand the
colocalization of Endonuclease G with a G-quadruplex specific antibody, BG4. Results
revealed the binding of Endonuclease G to G-quadruplexes formed in the mitochondrial
genome within cells. This was further investigated by performing ChIP DNA sequencing.
Results revealed specific binding of Endonuclease G to different G-quadruplex DNA
3
structures formed in mitochondrial DNA (when present on a plasmid or in the mitochondrial
genome), when it was done both in vitro and inside the cells. Therefore, this study provides
novel insights into the mechanism of fragility of the mitochondrial genome associated with
aging and different human diseases.
Compared to repair of nuclear DNA, repair of various DNA damages is less
understood in the mitochondria. Base excision repair (BER) is one of the most, well-studied
DNA repair pathways in mitochondria. However, inability of mitochondrial extracts to remove
UV induced damage confirmed the absence of Nucleotide Excision repair (NER). In a
previous study, we showed that classical nonhomologous DNA end-joining (c-NHEJ), the
predominant double-strand break (DSB) repair pathway in the nucleus is undetectable in
mitochondria. Further, it was seen that DSB repair through microhomology mediated endjoining
(MMEJ) could explain DNA deletions often seen in the mitochondrial genome.
Studies conducted in this thesis, reveal that breaks generated by Endonuclease G were
repaired by MMEJ using direct repeats (microhomology) that are present in the immediate
vicinity of the broken ends. Further, efforts were made to reconstitute the whole breakage
and joining process on a mitochondrial DNA either using purified Endonuclease G or
mitochondrial protein extracts. Thus, the results presented in this thesis establish the
molecular basis behind fragility of the most common mitochondrial deletion, associated with
multiple human diseases.
Although mitochondrial disorders due to error prone repair of mtDNA are well
established, integrity of mitochondrial genome is well maintained in normal mammalian cells,
including humans. In the last chapter of the thesis, I have investigated the presence of
homologous recombination (HR) in mitochondria and its potential role in maintenance of
mitochondrial genomic integrity. Biochemical studies revealed that HR mediated repair is
more efficient in the mitochondria of testes as compared to brain, kidney and spleen.
Interestingly, a significant increase in the efficiency of HR was observed when a DSB was
introduced. Analyses of the clones suggest that most of the recombinants obtained from the
4
reaction products of brain, kidney and spleen were generated through reciprocal exchange,
while 30% of recombinants were due to gene conversion in testicular extracts. Colocalization
and immunoblotting studies showed the presence of RAD51 and MRN complex proteins in
the mitochondria. To study the role of HR proteins in mitochondria, we also performed
immunodepletion of MRE11, RAD51 or NIBRIN. Results of these studies showed reduced
HR mediated repair following immunodepletion of HR associated proteins, while depletion of
a control protein SMAC/DIABLO, did not affect the frequency of HR. Thus, the results from
second part of the study provide the evidence for occurrence of homologous recombination
in mitochondria for maintaining genome stability.
In summary, G-quadruplex structures in mitochondria are substrates for enzymes like
Endonuclease G leading to breaks in Region I of mitochondria. When such breaks are
generated adjacent to direct repeats, microhomology-mediated end joining takes over as the
repair mechanism leading to a 9 bp deletion (as one of the direct repeats is lost). Patients
with mitochondrial dysfunction were reported to have such deletions frequently. This study
thus demonstrates the mechanism of fragility in one of the most commonly deleted regions
of the mitochondrial genome. The second part of the study demonstrates that the
maintenance of mitochondrial genome integrity is mediated through homologous
recombination mediated repair and is dependent on RAD51 and MRN complex. Thus, the
present study identifies the role of non-B DNA structures and DNA double-strand break
repair in mitochondrial DNA transactions. | en_US |
