Addition of Bases to the 5-end of Human Telomeric DNA:

Influences on Thermal Stability and Energetics of Unfolding

Katherine L. Hayden
1
and David E. Graves
1,2

1
Department of Chemistry, University of Alabama at Birmingham,
2
UAB Comprehensive Cancer Center
901 14
th
Street South, Birmingham, AL 35294
Sequence R
2
n
W
error
22mer 0.9803 61.09 4.32
23mer 0.9855 73.25 4.44
24mer 0.9916 59.18 2.73
(log ())
(log [2])
= n
W
Table 2/Figure 7. DSC osmolyte perturbation results for hTel strands in BPEP with increasing glycerol. Strands
were prepared and annealed in BPEP at stock concentration of 1 mM. Samples of each strand were then
dialyzed in BPEP with appropriate glycerol percentage overnight and then prepared to 75uM single strand
concentration in appropriate buffer. Buffer and sample scans were ran from 5-105 C at a rate of 1 C/min.
Each sample was analyzed three times using fresh sample for each scan.
DSC Osmolyte Perturbation
PPC
22mer 23mer 24mer
Figure 6. Pressure perturbation calorimetry is a relatively new technique that can determine the thermal
expansion coefficients of a macromolecule in an aqueous solution as well as calculate changes in the partial
volume of the molecule as it unfolds. Unlike the constant pressure conditions of DSC, PPC is performed by
alternating between high and low pressures while raising the temperature. By changing the pressure above a
liquid, we are able to induce entropy and heat changes such that:

With a change in pressure and utilization of the Maxwell relationship:

We can derive the following expression for calculating the thermal expansion coefficient of the solute (
s
) as:

= /[]

Samples of each sequence were prepared at 1mM in BPEP and degassed thoroughly. Samples were analyzed
using a Microcal VP-DSC with PPC attachment from 5 to 120 C, where pulses of high pressure (70 psi) and
low pressure (0 psi) were performed at every 1 C with a 150 s delay between pulses. Scans of water against
water, buffer against water, buffer against buffer, and sample against buffer were all ran under the same
conditions.
Discussion
The addition of nucleotides (5-T, and 5-TT) to the 5 end of the core human telomere sequence, 5-
AGGG(TTAGGG)
3
-3, results in significant changes to the thermodynamic, structural and hydration properties as
compared to the core h-Tel 22. Overall, the G-quadruplex structures for the 23-mer and 24-mer maintain an
antiparallel conformation as indicated by the CD spectra; however,
1
H NMR analyses reveal significant changes
due to interactions of the single stranded 5 end with the terminal G-tetrad. DSC studies were carried out
under conditions of increasing osmolyte perturbation and reveal that both the 22-mer and 24-mer bind
approximately 60 water molecules upon denaturation. In contrast, the 23-mer was found to bind 73 water
molecules. This uptake of water upon unfolding seemingly conflicts with the DSC-pressure perturbation
calorimetry experiments where a negative peak upon unfolding indicates a decrease in the molecules partial
specific volume. Molecular dynamic investigations of each sequence have allowed us to investigate possible
interactions between the additional nucleotides with the terminal G-tetrad. Our results indicate that internal
and external single stranded regions of quadruplex structures play a more significant role in the overall
structure and stability than previously thought. The influences of these additional 5 nucleotides should also be
taken in to consideration in future research when using the human telomere G-quadruplex as a model target
for novel drug development.

=

(/)
-0.106 mL/mol -0.059 mL/mol -0.109 mL/mol
22mer
23mer
24mer
Molecular Dynamics
Figure 3. Molecular dynamics studies were performed using AMBER on the Alabama Super Computer
located in Huntsville, Al. Solvated simulations were performed on two different starting structures: Neidles
crystal structure with K
+
and Patels NMR structure with K
+
, and additional nucleotides were added to each
starting structure using Discovery Studios. Each structure was first minimized in vacuum before solvation
using a 10 octagon TIP3PBOX. The solvated structures then underwent a series of explicit MD simulations
to allow the surrounding waters to relax by slowly heating from 0 to 300 K while holding the nucleic acid
fixed with 5 kcal/mol*
2
harmonic constraint for 260ps. The harmonic constraint was then dropped to 0.5
kcal/mol*
2
for 60ps while holding the volume constant. Finally, the harmonic constraint was completely
removed for 100 ps during the final simulation.
1
H NMR
9, 10
3
11,17
22 15, 23
5, 16
21 4
24mer
23mer
22mer
Figure 4.
1
H NMR analysis of each sequence was performed on a 700 mHz Bruker Ultrashield NMR equipped
with a TCI cryoprobe. Each sequence was analyzed at 1 mM in 90 % BPEP with 10% D
2
O at 5 C using
excitation gated solvent suppression. Peak assignments for the 23mer in BPEP were adopted from Luu and
Patels NMR derived assignments reported in JACS, 2006
MDA-MB-231 Cell invasion
Figure 5. Cell invasion assays were performed using pre-casted Bio-Coat Matrigel Matrix multiwell plates
fromBD Biosciences. Each assay was performed in triplicate using 20,000 cells and 4 nmol of agonist per well.
After addition of agonist, each well was incubated for 24 hours at 37 C. After incubation, the invaded cells
were stained using Hema-3 staining kit from Sigma Aldrich and counted by eye using a microscope. ODN-M-
362 was used as a positive control for increased cellular invasion while no agonist was added to the vehicle
wells

Vehicle
ODN-M362
22mer hTel
23mer hTel
24mer hTel
0
50
100
150
200
250
I
n
v
a
d
e
d
c
e
l
l
c
o
u
n
t

Cell invasion assay using MDA-231
Abstract
Telomeric DNA has been intensely investigated for its role in chromosome protection, disease, cell aging and
apoptosis. These tandem repeats, such as (TTAGGG)n in humans, found at the ends of eukaryotic
chromosomes provide a novel target for the research and development of new drugs in the treatment of
cancer. Telomeric sequences show slight variations from species to species; however, each one contains a core
repeat of 3 to 4 guanines allowing the G-rich strand to fold into a compact and stable secondary structure
referred to as the G-quadruplex. Recently, our lab has examined the influences of 5- bases flanking the h-Tel
core sequence 5-AGGG(TTAGGG)
3
-3. Our studies reveal that the addition of bases to the 5- end of the core h-
Tel sequence results in significant changes to the thermodynamic properties associated with the thermal
denaturation of the G-quadruplex(es). Investigation of three sequences where additional nucleotides are
added to the 5 of the core h-Tel sequence using DSC, osmolyte perturbation, pressure perturbation
calorimetry, molecular dynamics, CD spectroscopy and NMR have allowed us to gain insights into the structural
and thermodynamic effects of these 5-end nucleotide additions. From these studies we hope further
understanding of the hTel G-quadruplex as well as the role of this structural motif in biological activity.
Blackburn EH, Gall JG. 1978. J Mol Biol 120: 33-53
Dragan AI, Russell DJ, Privalov PL. 2009. Biopolymers 91: 95-101
Luu KN, Phan AT, Kuryavyi V, Lacroix L, Patel DJ. 2006. J AmChem Soc 128: 9963-70
Parsegian VA, Rand RP, Rau DC. 2000. Proc Natl Acad Sci U S A 97: 3987-92
Petraccone L, Spink C, Trent JO, Garbett NC, Mekmaysy CS, et al. 2011. J AmChem Soc 133: 20951-61
Qu X, Chaires JB. 2001. J AmChem Soc 123: 1-7
Spink CH, Garbett N, Chaires JB. 2007. Biophys Chem126: 176-85
Williamson JR. 1994. Annu Rev Biophys Biomol Struct 23: 703-30
Burge S, Parkinson GN, Hazel P, Todd AK, Neidle S. 2006. Nucleic Acids Res 34: 5402-15
Parkinson GN, Lee MP, Neidle S. 2002. Nature 417: 876-80


CD
Figure 2. CD analysis of each sequence was performed on a JASCO J-815 spectrometer at 5 M concentrations
in BPEP from 225 to 325 nm with a 2 s response delay. Each sequence was also analyzed in BPEP buffer with 1
to 5% glycerol to ensure structural ruggedness throughout DSC osmolyte perturbation analysis (data not
shown).
22mer
23mer
24mer
Table 1/ Figure 1. DSC analysis of each hTel DNA sequence was performed on a Microcal VP-DSC at 100 uM
DNA in 10mM Bisphosphate, 1mMEDTA and 100mM KCl, pH 7.0. Each sample was analyzed from5 to 110 C at
60 C/hr. Results were cubic baseline adjusted and analyzed first by integration of the total area under the
curve and second by fitting to a non-2-state model using Origin DSC analysis software provided by Microcal.
Name Sequence Htot (kJ/mol) Htot Tmtot (C)
H1
(kJ/mol)

Tm1 (C)

H2
(kJ/mol)
Tm2 (C)
21mer (K) 5-GGG(TTAGGG)3-3 147.6 NA 69.05 69.2 59.5 80.5 70.38
5 22mer (K) 5-AGGG(TTAGGG)3-3 154.5 7.0 67.2 57.4 53.5 98.9 67.2
5 23mer (K) 5-TAGGG(TTAGGG)3-3 189.2 41.6 62.4 72.3 52.7 98.5 66.5
5 24mer (K) 5-TTAGGG(TTAGGG)3-3 229.7 82.1 61.3 60.0 46.2 174.3 62.4
3 22mer (K) 5-GGG(TTAGGG)3T-3 155.2 7.6 67.5 69.3 55.5 88.6 68.6
3 23mer (K) 5-GGG(TTAGGG)3TT-3 188.3 40.7 66.8 53.7 56.5 137.8 67.68
3 24mer (K) 5-GGG(TTAGGG)3TTA-3 199.1 51.5 65.9 63.7 55.2 135.2 66.6
DSC
References