of 7
Supplemental Information for “A Multiplexed, Two-El
ectrode Platform for Biosensing
based on DNA-Mediated Charge Transport”
Ariel L. Furst,
1
Michael G. Hill,
1,2
and Jacqueline K. Barton
1,*
Methods and Materials
Preparation of Surfaces and First Alkanethiol Monol
ayers.
Gold surfaces were polished with 0.05 μm alumina sl
urries (Buhler) before
monolayer assembly. Mixed monolayers were then for
med on the substrate plate by self-
assembly of 100 mM 12-azidododecane-1-thiol (C
12
thiol azide) and 100 mM 11-
mercaptoundecyl-phosphoric acid from an ethanolic s
olution. Surfaces were incubated in
the thiol solution for 18-24 h, followed by rinsing
with ethanol and phosphate buffer (5
mM phosphate, 50 mM NaCl, pH 7.0).
DNA Synthesis and Purification.
Hexynyl-labeled oligonucleotides were synthesized o
n an Applied Biosystems
3400 DNA synthesizer, modified at the 5` end with a
C6-alkyne reagent purchased from
Glen Research, Inc. Complementary unmodified stran
ds were purchased from IDT.
DNA strands modified with Nile Blue at the 5` termi
nus were prepared as previously
reported.
S1
Briefly, DNA was synthesized with ultramild reage
nts (Glen Research, Inc)
to prevent Nile Blue degradation, and 5-[3-acrylate
NHS ester]-deoxy uridine was
incorporated as the 5` terminal base. With DNA on
the solid support, 10 mg/mL Nile
Blue perchlorate in 9:1
N,N-
dimethylformamide/
N,N
-diisopropylethylamine (Sigma
Aldrich) was added and allowed to shake for 24 h.
Beads were washed three times each
with
N,N-
dimethylformamide, methanol and acetonitrile. The
DNA was removed from
the solid support with 0.05 M potassium carbonate i
n methanol at ambient temperature
for 24 h.
Preparation of all oligonucleotides followed a repo
rted protocol. For non-
ultramild syntheses, DNA was deprotected and cleave
d from the solid support with
ammonium hydroxide (60° C for 12 h). Following a p
reliminary round of high-
performance liquid chromatography (HPLC) on a PLRP-
S column (Agilent),
oligonucleotides were treated with 80% acetic acid
in water for 20 minutes. Each
oligonucleotide was again purified by HPLC using a
gradient of acetonitrile and 50 mM
ammonium acetate. Oligonucleotides were then desalt
ed by ethanol precipitation and
quantified by ultraviolet-visible spectrophotometry
based on their extinction coefficients
at 260 nm (IDT Oligo Analyzer). Oligonucleotide ma
sses were verified by matrix-
assisted laser desorption (MALDI) mass spectrometry
. DNA duplexes were formed by
thermally annealing equimolar amounts of single-str
anded oligonucleotides in
deoxygenated phosphate buffer (5mM phosphate, 50 mM
NaCl, pH 7.0) at 90° C for 5
minutes followed by slowly cooling to 25° C.
The following sequences were prepared:
Well Matched
Alkyne: H-C
2
-(CH
3
)
6
-5’-GCT CAG TAC GAC GTC GA-3’
Complement: 3’-CGA GTC ATG CTG CAG CT-
5’
Mismatched
Alkyne: H-C
2
-(CH
3
)
6
-5’-GCT CAG TA
C
GAC GTC GA-3’
Complement: 3’-CGA GTC AT
A
CTG CAG CT-5’
TBP Binding Sequence:
Alkyne: H-C
2
-(CH
3
)
6
-5’-GGC GTC
TAT A
AA GCG ATC GCG A-3’
Complement: 3’-CCG CAG
ATA T
TT CGC TAC CGC T-5’
COPG Binding Sequence:
Alkyne: H-C
2
-(CH
3
)
6
-5’-AAC CG
T GCA
CTC AA
T GCA
ATC-3’
Complement:
3’-TTG GC
A CGT
GAG TT
A CGT
TAG-5’
The location of the mismatch is indicated in italic
s and with an underline, and the protein
binding sites are shown in bold.
TBP and CopG Experiments.
TATA-Binding Protein (TBP) was purchased from Prote
inOne, and CopG was
purchased from Origene. Both proteins were stored
at -80 °C until use. MicroBiospin 6
columns (BioRad) were used to exchange the shipping
buffer for Tris buffer (10 mM
Tris, 100 mM KCl, 2.5 mM MgCl
2
, 1 mM CaCl
2
, pH 7.6). Prior to electrochemical
measurements with CopG and TBP, electrodes were inc
ubated with 1 μM Bovine serum
albumin (BSA) for 30 min, followed by rinsing with
Tris buffer (10 mM Tris, 100 mM
KCl, 2.5 mM MgCl
2
, 1 mM CaCl
2
, pH 7.6). Protein solutions (4 μL) were added to e
ach
electrode and incubated for 20 minutes at ambient t
emperature prior to measurement.
References
S1. Gorodetsky, A. A.; Ebrahim, A.; Barton, J. K. E
lectrical Detection of TATA Binding
Protein at DNA-Modified Microelectrodes.
J. Am. Chem. Soc.
2008
,
130
, 2924-2925.
Figure Captions
Figure S1
: Electrochemistry of [Cu(phendione)
2
]
2+
. A cyclic voltammogram (CV) of
[Cu(phendione)
2
]
2+
was obtained in degassed Tris buffer (10 mM Tris,
100 mM KCl, 2.5
mM MgCl
2
, 1 mM CaCl
2
, pH 7.6) with a glassy carbon working electrode us
ing a scan
rate of 0.1 V/s against an AgCl/Ag reference electr
ode.
Figure S2:
Nyquist plots of electrochemical impedance spectros
copy of differentially
formed monolayers. Shown are results for a bare go
ld electrode (black), a mixed
monolayer of azide and phosphate-terminated thiols
(green), a DNA monolayer formed
from [Cu(phendione)
2
]
2+
catalyst activation from the secondary electrode (
blue), and a
DNA monolayer formed from the catalyst activation a
t the primary, substrate electrode
(red). Conditions used for impedance spectroscopy
were 400 μM ferrocyanide in
phosphate buffer (5 mM phosphate, 50 mM NaCl, pH 7.
0).
Figure S3:
Optimizing the spacer height. Eight Teflon spacers
of different heights were
tested for electrochemical signal and mismatch disc
rimination. The spacer between the
two electrode arrays establishes the gap between th
e two electrodes (left). The current
from constant current amperometry obtained as a fun
ction of spacer height (center) is
maximized with the 127 μm spacer (red asterisk). M
ismatch discrimination as a function
of spacer height (right) is reported as a ratio of
the mismatched signal to the well matched
signal, with maximal discrimination also observed w
ith the 127 μm spacer (red asterisk).
All electrochemistry was conducted in Tris buffer (
10 mM Tris, 100 mM KCl, 2.5 mM
MgCl
2
, 1 mM CaCl
2
, pH 7.6) with 4 μM methylene blue and 300 μM K
3
[Fe(CN)
6
]. 18-
mer well matched and mismatched DNA was used.
Figure S1
0.5 0.4 0.3 0.2 0.1 0.0 -0.1 -0.2 -0.3 -0.4 -0.5
-80
-60
-40
-20
0
20
40
60
Current (
μ
A)
Potential (V) v AgCl/Ag
Figure S2
Figure S3