Всероссийская мультиконференция с международным участием «Биотехнология – медицине будущего»
29 июня - 2 июля 2019 г., г. Новосибирск, Россия
136
Molecular dynamics pipeline for predicting the effects of cancer-associated
amino acid substitutions: Identification of new OGG1 somatic variants
with low activity
Popov A.V.
1,2
, Endutkin A.V.
1,2
, Barmatov A.E.
1
, Makasheva K.A.
2
, Yatsenko D.D.
2
,
Raspopova D.Yu.
2
, Zharkov D.O.
1,2
1
SB RAS Institute of Chemical Biology and Fundamental Medicine, Novosibirsk, Russia
2
Novosibirsk State University, Novosibirsk, Russia
Progress in modern sequencing methods has allowed increasingly personalized antitumor
therapy. To enhance
the power of personalization, data on the functionality of tumor-
associated variants of proteins involved in the response to DNA damage are of considerable
interest. Particular attention today is paid to predictions of the functionality of protein variants
by computational methods. However, the vast majority of such
methods are presently based
on the phylogenetic information and barely use the modern possibilities of computational
analysis of the structure and dynamics of proteins.
We have employed molecular dynamics simulation to model the structures of mutant
variants, observed in clinical tumor samples, of human 8-oxoguanine-DNA glycosylase
(OGG1), a protein belonging to the base excision repair system.
This enzyme reduces the
cytotoxicity of several classes of anticancer drugs, including thiotepa, bleomycin, cisplatin, etc.
The enzymatic activity of these mutant variants has been determined in parallel.
We have developed a software pipeline to automate the retrieval
and preparation of many
similar structures differing only by amino acid substitutions, for feeding into molecular
dynamics simulation programs. Trajectories of a random sample of mutant variants of OGG1
were obtained. The predicted functionality of all known mutant OGG1 variants found in
human cancers was classified according to the results of four algorithms using only phyloge-
netic information (SIFT, FATHMM, MutationTaster and PROVEAN),
and compared with the
classification of OGG1 mutants based on the results of molecular dynamics. We have purified
a series of mutant OGG1 proteins and characterized them biochemically, establishing that
OGG1 I145M and R161W variants found in esophageal squamous
cell carcinoma and colon
adenocarcinoma, respectively, have a significantly reduced enzymatic activity. Within the set
of 11 experimentally characterized mutant variants, molecular
dynamics analysis showed
a better correlation with experimental results than did predictions by phylogenetic methods.
This study was supported by RSF (grant 18-74-00052).