S
1
A Redox Role for the [4Fe
4S] Cluster
of Yeast DNA Polymerase δ
Phillip L. Bartels
1
, Joseph L. Stodola
2
,
Peter M.J. Burgers
2
,*
, and Jacqueline K. Barton
1
,*
Supporting Information
DNA synthesis and purification for electrochemistry
.
Thiol
-
modified DNA
sequences were
prepared by standard phosphoramidite chemistry on a DNA synthesizer (Applied Biosystems) using A,
G, C, T phosphoramidites and the 3
′
-
Thiol
-
Modifier 6
-
S
-
S CPG as purchased from Glen research. DNA
substrates were cleaved and deprotected by 8
-
hour incubation in NH
4
OH (Sigma
-
Aldrich) at 65 °C.
Deprotected DNA was separated from truncation products by reverse
-
phase HPLC (Agilent PLRPS
column, gradient of 5
–
75% ACN/95
-
25% 50 mM NH
4
Ac over 30 minutes at a 2 mL/min flow rate).
Thiol
-
modified DNA
was reduced by dissolving in 50 μL Tris, pH 8.0 (Qiagen elution buffer), adding
excess DTT (Sigma
-
Aldrich), and shaking for 45 minutes. DTT was removed by filtration through a
NAP
-
5 column (GE Healthcare) prior to a final round of HPLC purification (gradie
nt of 5
–
15%
ACN/95
–
85% 50 mM NH
4
Ac over 35 minutes at 2 mL/min). Lastly, single
-
stranded DNA was
desalted by standard ethanol precipitation (100 μl water, 1 mL 100% EtOH, 130 mM NaCl) and the
identity of the substrate was confirmed by MALDI
-
TOF. Unmodi
fied oligomers were ordered from IDT
and purified by the DMT
-
free HPLC method. Desalted DNA was dissolved in a phosphate storage
S
2
buffer (5 mM sodium phosphate, 50 mM NaCl, pH 7.0) and concentrations were determined by UV
-
visible spectroscopy using ε
260
val
ues estimated by Integrated DNA Technologies (IDT). Equimolar
concentrations of single stranded DNA were then degassed and annealed (rapid heating to 95
o
C, 5
-
minute incubation, and 1.5 hour cooling to 20
o
C).
Preparation of DNA
-
modified Gold Electrodes.
Multiplexed chips containing 16 Au
electrodes (0.015 cm
2
area) were prepared as described previously (
64
). Self
-
assembled monolayers
(SAMs) were formed by incubating 25 μL of 25 μM duplexed DNA on the electrode overnight, after
which electrodes were rins
ed 3
-
5 times in phosphate buffer (5 mM sodium phosphate, pH 7.5, 50 mM
NaCl) and backfilled for 45 minutes with 1 mM 6
-
mercapto
-
1
-
hexanol (Sigma
-
Aldrich) in the same
buffer containing 5% (v/v) glycerol. Electrodes were then extensively rinsed in phosphate
buffer
followed by protein storage buffer (30 mM HEPES, pH 7.4, 350 mM NaAc, 1 mM DTT, 0.1 mM
EDTA, 10% glycerol, and 0.01% decaethylene glycol monododecyl ether). Lastly, the absence of
electroactive impurities was confirmed by scanning the surface with c
yclic voltammetry (CV).
Bulk electrolysis experiments were undertaken by droplet electrochemistry (30
-
40 μL solution)
on Au rod electrodes of 0.0314 cm
2
electrode area (Pine Research Instrumentation). Electrodes were
cleaned as previously described (
65
), a
nd monolayers formed using the same procedure as the
multiplexed chip.
S
3
Figure
S1
.
Scan rate dependence of the CV current in 500 nM Pol δ DV incubated with 5.0 μM PCNA,
80 μM dATP, and 8.0 mM MgAc
2
.
(
a
)
The maximum peak current incre
ases with increasing scan
rate,
coupled with an increase in peak splitting. (
b
) The current
exhibits
a linear dependence on th
e square
root of the scan rate,
characteristic of a diffusive rather than adsorbed species. The scan rates included
are 20, 50, 80, 100, 200, and 500 mV/s.
The line was fit to data averaged from 8 separate experiments,
and the fit is I = 7.7559ν
1/2
+ 0.5725 with an R
2
value of 0.9828.
S
4
Figure
S2
.
Pol δ and EndoIII electrochemistry compared. 1.5 μM EndoIII (stored in 20 mM sodium
phosphate, pH 7.5, 150 mM
NaCl, 1 mM EDTA) was exchan
ged into Pol δ storage buffer (3
0 mM
HEPES, pH 7.4, 350 mM NaAc, 1 mM DTT, 0.1 mM EDTA, 10% v/v glycerol, 0.01% decaethylene
glycol monododecyl ether w/v) and added to a multiplexed chip containing unmodified Pol δ DNA
(49:58
-
mer
substrate).
(
a
)
UV
-
visible spectra taken before and after buffer exchange confirm the stability
of E
ndoIII in a HEPES
-
based buffer
.
(
b
)
The midpoint potential as measured by CV is 113 ± 3 mV,
virtually indistinguishable from Pol
δ DV at 113 ± 5 mV versus N
HE
.
S
5
Figure
S3
.
SQWV of 500 nM WT Pol δ and exonuclease
-
deficient Pol δ DV with and without 5.0 μM
PCNA. WT and exo
-
Pol δ DV share the same potential, and both generate a substantial signal on a
DNA
-
modified gold electrode; the smaller size of the WT si
gnal may be due in part to DNA degradation
by exonuclease activity. PCNA itself does not affect the potential, but its absence results in significantly
decreased signal size and lower stability over time. SQWVs were taken at 15 Hz frequency and 25 mV
ampli
tude, and electrochemistry was carried out in storage buffer (20 mM HEPES, pH 7.4, 350 mM
NaAc, 1 mM DTT, 0.1 mM EDTA, 10% glycerol v/v, 0.01% decaethylene glycol monododecyl ether
v/v) with 8.0 mM MgAc
2
and 80 μM dATP.
S
6
Figure S4
.
Pol δ electrochemistry on different DNA monolayer morphologies.
(
a
)
To find an optimal
DNA monolayer morphology for Pol δ signaling, we prepared multiplexed chips containing either
closely packed (assembled with 100 mM MgCl
2
) or loosely packed (no MgCl
2
) D
NA films. Two chips
were prepared for both morphologies, with one half of each chip consisting of well
-
matched (WM)
DNA
(dark blue)
and the other containing DNA with either an abasic site
(red)
or a CA mismatch
(orange)
6 nucleoti
des from the monolayer sur
face
.
(
b
)
On closely packed films, Pol δ SQWV signals
were highly
variable and showed 46 ± 33% attenuation on abasic DNA (solid SQWV traces)
but no
signi
ficant mismatch discrimination
(dashed SQWV traces)
.
(
c
)
In contrast, SQWV signals on loosely
packed fi
lms
were much more c
onsistent between electrodes, with a 44 ± 16% signal loss on abasic
DNA (solid traces)
and
46 ± 29%
signal loss with CA mismatch DNA (dashed traces).
To minimize the
effects of v
ariability between devices, all direct
comparisons were m
a
de on a single
chip
; scans that
were directly compared are denoted by either solid or dashed lines in the SQWV signals shown. The
SQWV traces shown are an
average of 6 individual electrodes on a single device
, with
scans taken at 15
Hz frequency and
25 mV
amplitude.
S
7
Figure
S5
.
Characterization of electrochemically oxidized Pol δ.
(
a
)
Bulk electrolysis potentials were
~200 mV beyond the major oxidative and reductive peaks at 0.412 V (oxidation) and
-
0.188 V
(reduction) versus NHE
.
(
b
)
Yields were
calculated by subtracting a background electrolysis (blue) from
one containing protein (red) and taking the area und
er the resultant curve (green)
. Electrolysis of 150 μL
of 2.74 μM Pol δ at 0.412 V
gave ~35% oxidation yield. (
c
) UV
-
visible spectra reveal
an increased
absorbance
from 300
-
400 nm
consistent with cluster oxidati
on
with
no evidence of protein aggregation
.
(
d
)
CW X
-
band EPR spectra at 10 K reveal the presence of both [4Fe4S]
3+
(g = 2.08) and [3Fe4S]
+
(g =
2.02) species in the oxidized sample, wi
th a residual amount of [3Fe4S]
+
cluste
r present in the native
sample
. These results are consistent with the formation of [4Fe4S]
3+
cluster after anaerobic bulk
electrolysis, with some degrading to form [3Fe4S]
+
cluster in the absence of DNA.
As slight sample loss
did occur following oxidation, the UV
-
visible spectrum of oxidized Pol δ has been normalized to native
absorbance at 280 nm to afford a more direct comparison.
EPR spectra were taken at 12.85 mW
microwave power,
2 G modulation amplit
ude, and a receiver gain of 5.02 x10
3
.
S
8
Figure S6
.
Complete alkaline agarose gels from Figure 2 and control lacking PCNA. The gels include
untreated and oxidized Pol δ DV with 5.0 nM PCNA (
a
), untreated and re
-
reduced Pol δ DV (
b
), and
untreated and oxid
ized Pol δ DV in the absence of PCNA (
c
). No DNA synthesis occurs in the absence
of PCNA, confirming that the observed activity in native and oxidized samples is processive.
S
9
Figure S7
.
Establishment of activity by oxidized Pol δ. To see if oxidized Po
l δ remained active or
stalled completely, 0.01% heparin was included in reactions to challenge synthesis and products were
analyzed on a 5% denaturing polyacrylamide gel to resolve DNA between 30 and 1000 b
p (left). Pol δ
remains
active
after oxidation
,
primarily forming intermediate
-
sized products
(red range on gel)
. Native
Pol δ is more sensitive to heparin, with more DNA close to primer length
(blue range)
, but when it does
associate with DNA, most products are around the maximum size
(orange range)
. T
hese results are
consistent with tighter binding and slower processive DNA synthesis by the oxidized form.
Gels were
quantified using ImageQuant software; as synthesis appears as smears at this resolution, the total amount
of background
-
subtracted radioact
ivity in each major range shown was compared between untreated and
oxidized Pol δ. Error bars are standard deviation of the mean (n = 3).
S
10
Fig
ure S8
.
AQ assay controls with 140 nM
E. coli
Klenow fragment exo
-
. UVA irradiation in the
presence of AQ
-
pr
imed DNA had no significant effect on DNA synthesis by Klenow fragment. The lack
of difference confirms that irradiation in the presence of AQ does not adversely affect polymerase
enzymes, and further supports the assignment of attenuated activity in Pol δ
under the same conditions
to [4Fe4S] cluster oxidation.
SI References
(64
)
Slinker, J.
D.; Muren, N. B.; Gorodetsky, A.A.
; Barton, J. K.
J
.
Amer. Chem. Soc.
2010
,
132
,
2769
–
2774
.
(65
)
Barton, J.K.; Bartels, P.L.; Deng, Y.; O’Brien, E.
Methods
Enzymol
.
2017
,
591
, 355
–
414.