48
A. Jayaraman et al.
Fig. 4 DNA melting curves for a d(GGGGGGGG), b d(GCGCGCGC), c d(AAAAAAAA) and
d d(ATATATAT) at 6 μM DNA duplex concentration and 1 mM salt concentration. Symbols are
our simulation data and lines are prediction of NN methods: solid, short-dashed, long-dashed and
dash-dotted lines represent melting curves from the work of Blake et al. [80], Huguet et al. [81],
Sugimoto et al. [41], and SantaLucia [42], respectively. Reproduced from Ref. [19] with permission
from The Royal Society of Chemistry]
model captures the right trends in DNA melting with changing G-C content and base
sequence.
As done above for DNA, one could also tune the CG model (bead sizes and/or
bonded and non-bonded interaction parameters) to mimic other specific ONA
chemistries, like the peptide nucleic acids (PNA), locked nucleic acids (LNAs), etc.
As NN methods do not capture the melting trends for these synthetic ONAs, in the
work of Ghobadi and Jayaraman [19], the above model parameters were changed
to mimic the physical changes one would expect upon changing ONA backbone
chemistries (e.g., increasing/decreasing backbone flexibility, intra-strand nucleobase
spacing, charged/neutral backbone). The impact of these physical changes on ONA
melting behavior was also quantified to guide future synthesis of ONA with the
appropriate backbone chemistry to achieve desired melting curve/ONA stability.
Effect of Polymer Conjugation on ONA Melting Curves: To understand
the effect of polymer conjugation on ONA melting behavior, we present some
results published recently [67] for 40-mer and 100-mer flexible polymer conjugated to 8-mer GGGGCCCC ONA either with electrostatically neutral and flexible
(k
BB
angle = 10ε/rad
2 ) backbone, like PNAs or negatively charged and semi-flexible
(k
BB
angle = 30ε/rad
2 ) backbone, like DNA. The (implicit) solvent quality for ONA is
maintained to be a good solvent while the (implicit) solvent quality of polymer is
varied through the choice of ε
PP equal to 1ε and 5ε in Eq. (6). Using the units of energy
and temperature described in the CG model above, ε
PP
= 5ε and ε
HB
G−C = 61ε are
approximately equal to 0.84kT and 10.1kT, respectively. Thus, the polymer solvophobicity can be considered to be weak in comparison with the h-bonding interactions
and in this solvent the polymer should not precipitate out of solution but cause some
polymer aggregation. For comparison purposes, we also show results for analogous
systems with ONA not conjugated to any polymer denoted as unconjugated ONA.
We first present the results for neutral and flexible ONAs (e.g., PNAs) which
are conjugated to solvophobic polymer (Fig. 5a, b). When the ONA is conjugated
with polymer and ε
P P
= 1ε, the polymer conjugation has no effect on melting
A. Jayaraman et al.
Fig. 4 DNA melting curves for a d(GGGGGGGG), b d(GCGCGCGC), c d(AAAAAAAA) and
d d(ATATATAT) at 6 μM DNA duplex concentration and 1 mM salt concentration. Symbols are
our simulation data and lines are prediction of NN methods: solid, short-dashed, long-dashed and
dash-dotted lines represent melting curves from the work of Blake et al. [80], Huguet et al. [81],
Sugimoto et al. [41], and SantaLucia [42], respectively. Reproduced from Ref. [19] with permission
from The Royal Society of Chemistry]
model captures the right trends in DNA melting with changing G-C content and base
sequence.
As done above for DNA, one could also tune the CG model (bead sizes and/or
bonded and non-bonded interaction parameters) to mimic other specific ONA
chemistries, like the peptide nucleic acids (PNA), locked nucleic acids (LNAs), etc.
As NN methods do not capture the melting trends for these synthetic ONAs, in the
work of Ghobadi and Jayaraman [19], the above model parameters were changed
to mimic the physical changes one would expect upon changing ONA backbone
chemistries (e.g., increasing/decreasing backbone flexibility, intra-strand nucleobase
spacing, charged/neutral backbone). The impact of these physical changes on ONA
melting behavior was also quantified to guide future synthesis of ONA with the
appropriate backbone chemistry to achieve desired melting curve/ONA stability.
Effect of Polymer Conjugation on ONA Melting Curves: To understand
the effect of polymer conjugation on ONA melting behavior, we present some
results published recently [67] for 40-mer and 100-mer flexible polymer conjugated to 8-mer GGGGCCCC ONA either with electrostatically neutral and flexible
(k
BB
angle = 10ε/rad
2 ) backbone, like PNAs or negatively charged and semi-flexible
(k
BB
angle = 30ε/rad
2 ) backbone, like DNA. The (implicit) solvent quality for ONA is
maintained to be a good solvent while the (implicit) solvent quality of polymer is
varied through the choice of ε
PP equal to 1ε and 5ε in Eq. (6). Using the units of energy
and temperature described in the CG model above, ε
PP
= 5ε and ε
HB
G−C = 61ε are
approximately equal to 0.84kT and 10.1kT, respectively. Thus, the polymer solvophobicity can be considered to be weak in comparison with the h-bonding interactions
and in this solvent the polymer should not precipitate out of solution but cause some
polymer aggregation. For comparison purposes, we also show results for analogous
systems with ONA not conjugated to any polymer denoted as unconjugated ONA.
We first present the results for neutral and flexible ONAs (e.g., PNAs) which
are conjugated to solvophobic polymer (Fig. 5a, b). When the ONA is conjugated
with polymer and ε
P P
= 1ε, the polymer conjugation has no effect on melting