Research in the Lee lab endeavors to gain fundamental insights into damage, repair,
and enzymatic modifications of nucleic acids. In addition, we are striving to discover
potent chemotherapeutics that selectively inhibit DNA-modifying enzymes. We are
integrating tools of protein biochemistry, molecular biology, synthetic organic chemistry,
and X-ray crystallography to achieve these goals. Our current research interests
can further be divided into base-excision DNA repair enzymes, DNA-modifying enzymes,
and inhibitors of DNA-modifying enzymes.
1. Base-excision DNA repair enzymes:
The genomes of all living organisms are persistently exposed to threats imposed by
DNA damaging agents of exogenous and endogenous origins, leading to production of
various types of DNA lesions. These lesions if left unrepaired can cause genotoxic
mutations and ultimately tumors. To counteract the deleterious effects of the DNA
lesions, all the organisms employ DNA repair enzymes that search and remove the
lesions in DNA.
Of the enzymes, base-excision DNA repair (BER) proteins deal with the majority of
DNA lesions, playing a critical role in maintaining the genome integrity. Each cell
has thousands of BER enzymes that search, remove, and repair their substrate DNA
lesions. This repair process mostly occurs in a remarkably lesion-specific manner,
as seen in an A:oxoG-specific DNA repair enzyme MutY (Fig. 1). We are particularly
interested in how base-excision repair enzymes recognize and catalyze their substrate
nucleobase lesions in millions-fold excess of normal bases.
Fig. 1. X-ray Structure of an adenine-excision DNA repair protein MutY bound to oxoG:A-containing
duplex DNA. Substrate adenine (colored in blue) is flipped out of the duplex DNA
and presented into catalytic site of the enzyme forming lesion-specific hydrogen
bonding contacts. The estranged oxoG (in red) makes delicate hydrogen bonding interactions
with MutY. (Lee et al. PNAS 2009, 106, 18497)
<Methyl-CpG-binding domain proteins>
One protein of our interests is methyl-CpG-binding domain protein 4 (MBD4), which
is known to bind the methylated DNA of the CpG repeating regions, and also to remove
T from T:G mismatched DNA (Fig. 2). Unlike thymine DNA glycosylase (TDG) which is
a strict glycosylase, MBD4 possesses C-term glycosylase domain and N-term methyl-binding
domain. It has been shown that MBD4 suppresses CpG mutability and tumorigenesis in
vivo. Its potential involvement in epigenetic DNA demethylation has also been suggested.
Currently how the enzyme recognizes and removes its substrate base in methylated
CpG islands is completely unknown. Structure determination of this enzyme bound
to methylated DNA and T:G mismatch-containing DNA thus should drastically further
our understanding on the substrate recognition and catalysis of this important enzyme.
Fig. 2. DNA repair reaction catalyzed by methyl-CpG-binding domain protein 4 (MBD4).
The enzyme removes T mispaired with G at the CpG islands to generate an abasic site
and thymine. The abasic product is further processed by base-excision DNA repair
pathway to install the correct base cytosine.
2. DNA-modifying enzymes:
Whereas aberrant DNA modifications by DNA damaging agents are generally harmful,
enzymatic DNA modifications are essential for cellular life. The enzymatic modifications
include 5-methylation of cytosine and demethylation of 5-methylcytosine at the CpG
repeating regions, and deamination of adenine and cytosine in DNA. These enzyme-catalyzed
covalent modifications play important roles in key biological processes, such as
chromatin remodeling, X-chromosome inactivation, gene activation and inactivation,
antibody production, and anti-viral activity. DNA methyl transferase, methyl-CpG-binding
domain proteins, and the family of APOBEC proteins are among the DNA-modifying enzymes.
Despite the biological significance of the enzymatic transformations, the molecular
mechanisms by which the enzymes search, recognize, and catalyze their substrate DNA
are still poorly understood. Our studies focus on providing important insights into
these mechanisms using integrative approaches of X-ray crystallography and chemical
biology.
Fig. 3. Modification of cytosine by various DNA-modifying enzymes.
3. Chemotherapeutics targeting DNA-modifying enzymes:
<Inhibitors of DNA Methyltransferase 1>
Many cancer cells epigenetically inactivate tumor-suppressor genes by hypermethylating
the CpG repeating regions nearby tumor-suppressor genes (Fig. 4). Due to the reversibility
of this epigenetic mutations, cancer cells must actively maintain hypermethylation
state of the CpG islands, which are catalyzed by DNA methyltransferase 1 (DNMT1).
Inhibiting DNMT1 is known to reverse the hypermethylation, leading to reactivation
of the tumor suppressor genes and thus killing tumor cells. These observations prompted
many research groups to work on targeting DNMT1, decitabine being the first epigenetic
therapeutic approved by the FDA. Decitabine inactivates the enzyme by forming a
covalent adduct with DNMT1 (Fig. 4), effectively treats myelodysplastic syndrome,
and has shown promising response rates against leukemia. Unfortunately, the several
toxic side effects of decitabine, which has been suggested to relate with the chemical
instability of the DNMT1-decitabine adduct, limit its usefulness. Our research seeks
to develop powerful DNMT1 inhibitors with minimal side effects. To this end, we are
preparing various cytidine analogs that can inhibit DNMT1 by generating DNMT1-inhibitor
covalent adducts with enhanced chemical stability. Because these new inhibitors will
not release DNMT1, their ability to combat cancers will be more potent, long-lasting,
and free of side effects than the current drug on the market. Once potent DNMT1 inhibitors
are identified, we will determine the structure of the DNMT1-inhibitor adduct, which
would unveil a molecular basis for designing and developing epigenetic therapeutics
with an improved efficacy.
Fig. 4. (top) Gene silencing is initiated by the cytosine methylation that is catalyzed
by DNA methyltransferase 1 (DNMT1). Small molecule inhibitors of DNMT1 could reactivate
the silenced genes, especially various tumor suppressor genes in cancers ; (bottom)
Chemotherapeutics targeting DNMT1. Note that decitabine is a cytidine analog that
has nitrogen atom in place of carbon at the 5-position. The catalytic cysteine residue
of DNMT1 can attack the cytosine analog but the subsequent methyl transfer reaction
cannot occur, suicidally inhibiting DNMT 1 (MBD: methyl-CpG binding domain protein;
HMT: histone methyltransferase; HDAC: histone deacetylase).
