40
A. Jayaraman et al.
DNA is called a nucleotide which consists of a phosphate, a sugar, and a nucleobase.
There are four nucleobases found in DNA, namely, adenine (A), thymine (T), cytosine (C), and guanine (G). These nucleobases contain h-bonding donor and acceptor
atoms which enable DNA strands to hybridize with other DNA strands. According
to the Watson–Crick base pairing [24] rule, nucleobase A forms two h-bonds with
T, and G forms three h-bonds with C. This h-bonding interaction between complementary bases on two DNA strands and the intra-strand base–base stacking drive
the hybridization of two complementary DNA strands leading to a DNA duplex.
DNA duplexes or double strands are stable at low temperatures and melt/dissociate
into single strands above their melting temperature. The hybridization of complementary DNA strands and the thermo-reversible melting/hybridization of the DNA
duplexes make these polymers useful in several bio- and nano-technologies such as
bio-sensing [25, 26], gene sequencing [27, 28] and anti-sense therapy [29, 30]. Thus,
it is valuable to identify ways to controllably manipulate the melting temperature of
DNA for specific applications.
Ways to tune DNA melting temperatures include varying DNA strand length and
sequence [31], altering the salt concentration and/or solvent quality [32], conjugating
DNA with other macromolecules [33], etc. Going beyond DNA design, searching
for alternative chemistries that mimic DNA-like behavior, researchers have synthesized peptide nucleic acids (PNA) [34], click chemistry-based nucleic acids (CNA)
[35], locked nucleic acids (LNA) [36], threose nucleic acid (TNA) [37], etc. In most
of these examples, the main difference between DNA and their alternatives is in the
backbone chemistry, which in turn impacts the thermodynamic driving forces that
affect the duplex stability [19]. For example, PNA has an electrostatically neutral
and more flexible backbone as compared to DNA [38], while TNA has an additional
carbon atom in the sugar backbone as compared to DNA which alters the spacing
between nucleotides [39]. Despite synthetic advances in creating novel oligonucleic
acids or ONAs, there are only a few fundamental studies depicting the effect of
physical characteristics of ONA such as backbone flexibility and backbone electrostatics on the thermodynamics of hybridization. This may be due to the extensive
amount of time, cost and effort involved in the systematic synthesis of a range of
ONAs with chemical modifications and sophisticated experimental characterizations.
Thus, computational studies are valuable in this regard.
Empirical mathematical models such as nearest neighbor (NN) methods have been
used to predict DNA melting profiles [40–42]. Software packages (or Web servers)
like the m-fold server [43] utilize the thermodynamic information from NN models
to predict folded structures and melting temperatures of DNA and RNA. While such
methods can predict melting profiles of naturally occurring nucleic acids such as RNA
and DNA, they cannot be used for new, synthetic nucleic acid chemistries. As a result,
only sparse literature is available for synthetic ONAs [44, 45]. Molecular dynamics
(MD) simulations have also been used to elucidate the thermodynamics and kinetics
of the DNA hybridization process. A few studies have been done using all-atom MD
simulations which are chemically detailed but in turn are computationally expensive
to reach experimentally relevant time and length scales [46–58]. In addition, they
also suffer from the issue of lack of available force fields for novel synthetic ONAs
A. Jayaraman et al.
DNA is called a nucleotide which consists of a phosphate, a sugar, and a nucleobase.
There are four nucleobases found in DNA, namely, adenine (A), thymine (T), cytosine (C), and guanine (G). These nucleobases contain h-bonding donor and acceptor
atoms which enable DNA strands to hybridize with other DNA strands. According
to the Watson–Crick base pairing [24] rule, nucleobase A forms two h-bonds with
T, and G forms three h-bonds with C. This h-bonding interaction between complementary bases on two DNA strands and the intra-strand base–base stacking drive
the hybridization of two complementary DNA strands leading to a DNA duplex.
DNA duplexes or double strands are stable at low temperatures and melt/dissociate
into single strands above their melting temperature. The hybridization of complementary DNA strands and the thermo-reversible melting/hybridization of the DNA
duplexes make these polymers useful in several bio- and nano-technologies such as
bio-sensing [25, 26], gene sequencing [27, 28] and anti-sense therapy [29, 30]. Thus,
it is valuable to identify ways to controllably manipulate the melting temperature of
DNA for specific applications.
Ways to tune DNA melting temperatures include varying DNA strand length and
sequence [31], altering the salt concentration and/or solvent quality [32], conjugating
DNA with other macromolecules [33], etc. Going beyond DNA design, searching
for alternative chemistries that mimic DNA-like behavior, researchers have synthesized peptide nucleic acids (PNA) [34], click chemistry-based nucleic acids (CNA)
[35], locked nucleic acids (LNA) [36], threose nucleic acid (TNA) [37], etc. In most
of these examples, the main difference between DNA and their alternatives is in the
backbone chemistry, which in turn impacts the thermodynamic driving forces that
affect the duplex stability [19]. For example, PNA has an electrostatically neutral
and more flexible backbone as compared to DNA [38], while TNA has an additional
carbon atom in the sugar backbone as compared to DNA which alters the spacing
between nucleotides [39]. Despite synthetic advances in creating novel oligonucleic
acids or ONAs, there are only a few fundamental studies depicting the effect of
physical characteristics of ONA such as backbone flexibility and backbone electrostatics on the thermodynamics of hybridization. This may be due to the extensive
amount of time, cost and effort involved in the systematic synthesis of a range of
ONAs with chemical modifications and sophisticated experimental characterizations.
Thus, computational studies are valuable in this regard.
Empirical mathematical models such as nearest neighbor (NN) methods have been
used to predict DNA melting profiles [40–42]. Software packages (or Web servers)
like the m-fold server [43] utilize the thermodynamic information from NN models
to predict folded structures and melting temperatures of DNA and RNA. While such
methods can predict melting profiles of naturally occurring nucleic acids such as RNA
and DNA, they cannot be used for new, synthetic nucleic acid chemistries. As a result,
only sparse literature is available for synthetic ONAs [44, 45]. Molecular dynamics
(MD) simulations have also been used to elucidate the thermodynamics and kinetics
of the DNA hybridization process. A few studies have been done using all-atom MD
simulations which are chemically detailed but in turn are computationally expensive
to reach experimentally relevant time and length scales [46–58]. In addition, they
also suffer from the issue of lack of available force fields for novel synthetic ONAs