RESEA
RCH
ARTICL
E
Functional
and
structural
similarity
of
human
DNA
primase
[4Fe4S]
cluster
domain
constructs
Marilyn
E.
Holt
ID
1
, Lauren
E.
Salay
1
, Elizabeth
O’Brien
2
, Jacqueline
K.
Barton
2
, Walter
J. Chazin
ID
1
*
1
Department
s of Biochemi
stry
and
Chemistry,
and
Center
for
Structur
al Biology,
Vanderbil
t University,
Nashvil
le,
Tennessee,
United
States
of America,
2
Division
of Chemistry
and
Chem
ical
Engineeri
ng,
Californi
a Institute
of Technolo
gy,
Pasade
na,
California,
United
States
of America
*
walter.c
hazin@vande
rbilt.edu
Abstract
The
regulatory
subunit
of human
DNA
primase
has
a C-terminal
domain
(p58C)
that
contains
a [4Fe4S]
cluster
and
binds
DNA.
Previous
electrochem
ical
analysis
of a p58C
construct
revealed
that
its
affinity
for
DNA
is sensitive
to the
redox
state
of the
[4Fe4S]
cluster.
Con-
cerns
about
the
validity
of this
conclusion
have
been
raised,
based
in part
on
differences
in
X-ray
crystal
structures
of the
p58C
272-464
construct
used
for
that
study
and
that
of a N-termi-
nally
shifted
p58C
266-456
construct
and
consequently,
an
assumption
that
p58C
272-464
has
abnormal
physical
and
functional
properties.
To
address
this
controversy,
a new
p58C
266-464
construct
containing
all
residues
was
crystallized
under
the
conditions
previously
used
for
crystallizing
p58C
272-464
, and
the
solution
structures
of both
constructs
were
assessed
using
circular
dichroism
and
NMR
spectroscopy.
In the
new
crystal
structure,
p58C
266-464
exhibits
the
same
elements
of secondary
structure
near
the
DNA
binding
site
as
observed
in the
crystal
structure
of p58C
272-464
. Moreover,
in solution,
circular
dichroism
and
15
N,
1
H-hetero-
nuclear
single
quantum
coherence
(HSQC)
NMR
spectra
show
there
are
no
significant
differences
in the
distribution
of secondary
structures
or in the
tertiary
structure
or the
two
constructs.
To
validate
that
the
two
constructs
have
the
same
functional
properties,
binding
of a primed
DNA
template
was
measured
using
a fluorescence-bas
ed
DNA
binding
assay,
and
the
affinities
for
this
substrate
were
the
same
(3.4
±
0.5
μ
M
and
2.7
±
0.3
μ
M,
respec-
tively).
The
electrochemical
properties
of p58C
266-464
were
also
measured
and
this
p58C
construct
was
able
to engage
in redox
switching
on
DNA
with
the
same
efficiency
as
p58C
272-464
. Together,
these
results
show
that
although
p58C
can
be
stabilized
in different
conformations
in the
crystalline
state,
in solution
there
is effectively
no
difference
in the
struc-
ture
and
functional
properties
of p58C
constructs
of different
lengths.
Introduction
DNA
synthesis
at
the
replication
fork
begins
with
the
formation
of
8–12
nucleotide
(nt)
RNA
primers
on
the
single-stranded
DNA
template
[1,
2].
In
eukaryotes,
primers
are
generated
by
PLOS
ONE
| https://doi.or
g/10.137
1/journal.po
ne.02093
45
December
18,
2018
1 / 17
a1111111111
a1111111111
a1111111111
a1111111111
a1111111111
OPEN
ACCESS
Citation:
Holt ME, Salay
LE, O’Brien
E, Barton
JK,
Chazin
WJ (2018)
Functional
and structural
similarity
of human
DNA primase
[4Fe4S]
cluster
domain
constr
ucts.
PLoS
ONE 13(12)
: e0209345.
https://do
i.org/10.1371/j
ournal.pone
.0209345
Editor:
Albert
Jeltsch,
Univers
ita ̈t Stuttgart,
GERMANY
Received:
May 25, 2018
Accepted:
November
14, 2018
Published:
December
18, 2018
Copyright:
©
2018
Holt et al. This is an open
access
article
distributed
under
the terms
of the
Creative
Commons
Attribution
License,
which
permits
unrestricte
d use, distribu
tion, and
reproduction
in any medium,
provided
the original
author
and source
are credited.
Data
Availabilit
y Statement:
All crystallo
graphy
files are available
from
the Protein
Data Bank
archive
(accession
number
6DHW).
All other
relevant
data are within
the paper
and its
Supporting
Information
files.
Funding:
This research
was supported
by National
Institutes
of Health
(https://www
.nih.gov/)
grants
R35 GM1180
89 (W.J.C.),
R01 GM1269
04 (J.K.B.),
T32 GM80320
(M.E.H,
L.E.S.)
and T32 GM0761
6
(E.O.B.)
with additional
support
from
the Moore
Foundation
(https://www
.moore.org/
) (J.K.B.)
and
a Ralph
M. Parsons
fellowship
(http://rm
pf.org/)
the
heterotetrameric
DNA
polymerase
α
-primase
(pol-prim)
complex,
which
possesses
two
enzymatic
activities
in
two
distinct
active
sites
[3–5].
Primase,
a DNA-dependent
RNA
poly-
merase,
generates
the
initial
hybrid
RNA-DNA
primed
substrate,
which
is then
handed
off
to
DNA
polymerase
α
(pol
α
)
to
extend
the
initial
primer
by
approximately
twenty
DNA
nts.
The
extended
primed
substrates
are
in
turn
handed
off
to
the
processive
polymerases
ε
and
δ
,
which
synthesize
the
bulk
of
nascent
DNA
on
the
leading
and
lagging
strands,
respectively
[6–
8].
Human
DNA
primase
is composed
of
catalytic
(p48)
and
regulatory
(p58)
subunits.
The
regulatory
subunit
has
a C-terminal
domain
(p58C)
that
is unique
to
higher
eukaryotes
and
contains
a [4Fe4S]
cluster
[9–11].
This
domain
regulates
the
catalytic
efficiency
of
primase,
a
function
attributed
to
the
ability
to
bind
nucleotides,
DNA
template,
and
primed
substrate
[7,
9,
10,
12–18].
We
have
recently
proposed
that
[4Fe4S]
redox
control
of
DNA
binding
affinity
may
serve
as
a mechanism
to
drive
handoff
of
the
RNA
primed
template
from
the
primase
to
the
pol
α
subunits
of
human
pol-prim
[19].
Skepticism
about
some
of
the
reported
results
have
been
expressed
and
debated
[20–22],
much
of
which
was
related
to
differences
in
crystal
struc-
tures
obtained
from
p58C
constructs
with
different
N-termini
[23].
The
p58C
domain
of
human
primase
has
been
crystallized
under
two
different
conditions
and
these
have
generated
structures
that
have
the
same
global
fold
but
localized
differences
in
secondary
structures
[14,
15].
Fig
1 shows
a best-fit
superposition
of
the
high-resolution
X-ray
crystal
structures
of
the
two
p58C
272-464
and
p58C
266-456
constructs.
The
two
structures
are
clearly
very
similar
except
for
residues
Leu318-His351,
which
are
positioned
near
the
DNA
binding
site.
In
the
crystal
structure
of
p58C
272-464
, these
residues
occupy
a
β
-sheet-like
struc-
ture
that
is stabilized
by
cross-strand
interactions
with
another
molecule
in
the
unit
cell
[15].
In
addition,
a disulfide
cross-link
is formed
during
crystallization
between
the
Cys449
residues
of
adjacent
p58C
molecules.
In
contrast,
when
p58C
266-456
was
crystallized
under
a different
set
of
conditions,
this
β
-type
interaction
is not
observed
and
instead
these
residues
occupy
a
helical
hairpin
[14].
It has
been
proposed
that
Ile271
is critical
for
stabilizing
this
helical
motif
and
therefore
the
absence
of
this
residue
explains
why
p58C
266-456
has
a different
structure
[20,
22].
Many
of
the
concerns
about
our
work
were
attributed
to
the
fact
that
the
structural
differ-
ences
between
the
two
p58C
constructs
occurred
in
a region
that
contains
residues
interacting
Fig
1.
Compariso
n of
the
structures
of
p58C
272-464
and
p58C
266-456
.
(A)
Best-fit
superposition
over
all
backbone
atoms
of
the
crystal
structur
es
of
the
p58C
272-464
(PDB
ID:
3L9Q)
and
p58C
266-456
(PDB
ID:
5F0S)
constructs.
The
substrate
in
the
p58C
266-456
structure
is removed
for
clarity.
Inset:
Best-fit
superposition
over
all
backbone
atoms
of
the
DNA
binding
region
with
the
primed
substrat
e shown
in
the
p58C
266-456
structur
e. p58C
266-456
is colored
orange,
p58C
272-464
blue,
and
the
RNA-prime
d DNA
substrate
in
grey
(RNA
in
sticks,
DNA
in
slabs).
In
both
panels,
Leu318-Hi
s351
are
colored
yellow
in
the
p58C
272-464
structure
and
pink
in
the
p58C
266-456
structur
e. Connectivi
ty
between
residues
where
electron
density
is missing
is indicated
by
dashed
lines.
https://d
oi.org/10.1371/j
ournal.pon
e.0209345.g0
01
Character
ization
of primase
C-termina
l domain
constructs
PLOS
ONE
| https://doi.or
g/10.137
1/journal.po
ne.02093
45
December
18,
2018
2 / 17
(E.O.B.).
This research
used
resources
of the
Advanced
Photon
Source,
a U.S. Departme
nt of
Energy
(DOE)
Office
of Science
User Facility
operated
for the DOE Office
of Science
by Argonne
National
Laboratory
under
Contract
No. DE-AC02-
06CH11357.
Use of the LS-CAT
Sector
21 was
supported
by the Michigan
Economic
Developmen
t
Corporation
and the Michigan
Technology
Tri-
Corridor
(Grant
085P10008
17). NMR
data
collection
was supporte
d in part by instrumentat
ion
grants
from
the National
Science
Foundation
(https://w
ww.nsf.gov/)
(0922862)
and Nationa
l
Institutes
of Health
(S10 RR025677)
, and
matching
funds
from
Vanderb
ilt University
(https://
www.vande
rbilt.edu/).
Fluoresce
nce data collection
was supported
in part by an instrumen
tation
grant
from
the National
Institutes
of Health
(S10
OD021483)
. Molecular
graphics
and analyses
were
performed
with the UCSF
Chimera
package.
Chimera
is develope
d by the Resource
for
Biocompu
ting, Visualization,
and Informatics
at the
University
of California,
San Francisco
(supported
by NIGMS
P41-GM1033
11). The funders
had no
role in study
design,
data collection
and analysis,
decision
to publish,
or preparation
of the
manuscript.
Competing
interests
:
The authors
have declared
that no competing
interests
exist.
with
DNA
substrates.
However,
it is well
known
that
differences
between
crystal
structures
arise
when
crystals
are
formed
under
different
crystallization
conditions.
In
order
to
resolve
any
controversy
and
directly
address
the
concerns
raised
about
our
findings,
we
report
here
a
comprehensive
set
of
studies
of
a p58C
266-464
construct.
Comparisons
of
the
crystal
structure
of
this
construct
are
made
to
previously
reported
crystal
structures.
In
addition,
the
structure
was
analyzed
in
solution
by
circular
dichroism
and
NMR,
along
with
assays
of
DNA
binding
affinities
and
electrochemical
properties.
These
results
show
that
the
structure
of
Leu318-
His351
varies
in
accord
with
crystallization
conditions,
whereas
the
structure
in
solution
and
biochemical
properties
of
different
p58C
constructs
are
effectively
the
same.
Methods
p58C
266-464
construct
generation
The
p58C
266-464
construct
was
created
with
a Q5
site-directed
mutagenesis
kit
from
New
England
Biolabs,
using
the
p58C
272-464
construct
plasmid
[15]
as
the
template,
5’- GGCAA
GATTTCCTTAGATCAGATTGATTTGCTTTCTACC—3’
for
the
forward
primer,
and
5’–CA
CATTTCCgggcccctggaacagaac—3’
for
the
reverse
primer.
10
ng
of
p58C
272-464
plasmid
was
used
in
the
exponential
amplification,
which
was
completed
as
described
in
the
Q5
Site-
Directed
Mutagenesis
Kit
manual
(New
England
Biolabs),
except
that
the
final
extension
time
was
extended
to
5 minutes.
The
KLD
reaction
was
also
completed
as
described
in
the
manual,
except
that
4
μ
L
of
PCR
product
was
used
in
the
reaction
and
the
incubation
time
was
increased
to
20
minutes.
10
μ
L
of
the
KLD
reaction
product
was
transformed
into
XL1-Blue
cells.
DNA
was
extracted
from
individual
colonies
with
a Qiagen
QIAprep
Spin
Miniprep
Kit.
Appropriate
insertion
of
residues
266–271
was
confirmed
through
plasmid
sequencing
(GEN-
EWIZ,
LLC).
Protein
expression
and
purification
p58C
constructs
were
expressed
and
purified
as
previously
described
[19,
23].
In
short,
plas-
mid
DNA
was
transformed
into
BL21
(DE3)
cells
(Novagen)
and
cultured
in
Terrific
Broth
media
at
37 ̊C
to
an
OD
600
of
0.6–0.8,
when
flasks
were
moved
to
a 21 ̊C
incubator
with
shak-
ing.
After
30
minutes,
ferric
citrate
and
ammonium
ferrous
citrate
were
added
to
a final
con-
centration
of
0.1
mg/mL,
and
isopropyl
1-thio-
β
-D-galactopyranosi
de
was
added
to
a final
concentration
of
0.5
mM.
Protein
was
expressed
at
21 ̊C
overnight.
The
same
growth
protocol
was
used
to
generate
15
N-labeled
p58C,
except
that
cells
were
cultured
in
M9
media
supple-
mented
with
15
N-labled
ammonium
chloride
(Cambridge
Isotopes)
and
expressed
overnight
at
25 ̊C.
The
same
purification
scheme
was
used
for
both
unlabeled
and
15
N-labeled
protein
[23].
In
short,
protein
was
first
purified
by
nickel
affinity
chromatography
(Amersham
Biosci-
ences).
The
6xHis
tag
was
cleaved
with
H3C
protease
and
the
protein
was
dialyzed
into
a low-
imidazole
buffer
[19,
23].
The
protein
was
repassed
over
the
nickel
column
to
remove
the
H3C
protease
and
uncleaved
protein.
A heparin
column
was
used
as
the
final
purification
step
to
remove
residual
contaminants
[19,
23].
Crystallization
and
structure
determination
The
structure
of
p58C
266-464
was
determined
as
previously
described
for
p58C
272-464
[15,
23].
p58C
266-464
crystals
were
grown
by
hanging
drop
vapor
diffusion
at
16 ̊C
from
a drop
com-
posed
of
equal
volumes
of
50
mg/ml
protein
in
20
mM
MES
(pH
6.5)
and
75
mM
NaCl
and
reservoir
solution
containing
100
mM
Tris
(pH
8.5),
400
mM
Li
2
SO
4
and
18%
(v/v)
PEG
3350.
Prior
to
data
collection,
crystals
were
transferred
to
a drop
containing
100
mM
Tris
(pH
8.5),
Character
ization
of primase
C-termina
l domain
constructs
PLOS
ONE
| https://doi.or
g/10.137
1/journal.po
ne.02093
45
December
18,
2018
3 / 17
400
mM
Li
2
SO
4
,18%
(v/v)
PEG
3350,
and
20%
(v/v)
glycerol
for
five
seconds.
The
crystals
were
looped
and
flash
frozen
in
liquid
nitrogen.
X-ray
data
were
collected
at
beamline
21ID-D
(Life
Sciences
Collaborative
Access
Team)
of
the
Advanced
Photon
Source
at
Argonne
National
Laboratory
at
11.5
kEV.
All
data
were
processed
by
HKL2000
[24].
The
structure
was
determined
using
molecular
replacement
(PHASER-MR)
with
PDB
entry
3L9Q,
residues
274–
316
and
360–457,
as
the
search
model.
Manual
model
building
for
the
structure
was
performed
using
Coot
model
building
software,
and
waters
were
placed
with
the
Coot
routine,
Find
Waters
[25].
The
final
model
was
obtained
by
iterative
cycles
of
model
building
in
Coot
and
structure
refinement
using
Phenix.refine
in
the
Phenix
suite
of
programs
[26,
27].
Structures
were
superimposed
and
RMSD
calculated
in
Chimera
with
the
Matchmaker
algorithm
[28].
Programs
used
for
structure
determination
and
refinement
were
accessed
through
SBGrid
[29].
Statistics
for
data
collection
and
refinement
are
shown
in
Table
1.
Table
1.
Crystallogra
phic
data
collecti
on
and
refinemen
t statistics.
Data
collection
Space
Group
C2
Cell
Dimensions
a, b,
c (
Å
)
110.19,
52.56,
88.77
α
,
β
,
γ
( ̊)
90,
115.08,
90
Temperat
ure
(K)
100
Wavelengt
h (
Å
)
1.08
Resoluti
on
(
Å
)
50.00–2.01
(2.08–2.01)
Unique
Reflect
ions
27193
R
meas
(%)
0.14
(0.79)
I/σI
10.06
(2.12)
Complete
ness
(%)
88.6
(84.4)
Redundanc
y
3.6
(3.3)
Refinement
Resoluti
on
(
Å
)
50.00–2.01
(2.08–2.01)
No.
reflections
27153
R
work
/R
free
0.18/0.21
(0.23/0.2
4)
No.
molecules/AS
U
2
No.
atoms
2872
Protein
2689
Ligand/ion
41
Water
52
B-factors
Mean
39.1
Protein
38.7
Ligand/ion
66.9
R.m.s.
deviations
Bond
lengths
(
Å
)
0.003
Bond
angles
( ̊)
0.62
Ramach
andran
Favored
308
(99.4%)
Allowed
2 (0.6%)
Disallow
ed
0 (0%)
Values
in
parentheses
are
for
the
highest-res
olution
shell.
https://d
oi.org/10.1371/j
ournal.pon
e.0209345.t00
1
Character
ization
of primase
C-termina
l domain
constructs
PLOS
ONE
| https://doi.or
g/10.137
1/journal.po
ne.02093
45
December
18,
2018
4 / 17
Circular
dichroism
(CD)
spectroscopy
Samples
were
buffer
exchanged
into
20
mM
K
2
HPO
4
(pH
7.2)
and
diluted
to
a concentration
of
0.3
mg/mL.
The
far-UV
CD
spectrum
over
the
range
190–260
nm
was
acquired
at
room
temperature
using
a Jasco
J-810
spectrophotometer.
Each
spectrum
is the
average
of
three
scans
acquired
with
a scanning
rate
of
50
nm/min
and
data
pitch
of
1 nm.
Prior
to
generating
the
overlay
in
Graphpad
Prism
7,
the
p58C
266-464
spectrum
was
scaled
to
the
p58C
272-464
spectrum
by
averaging
the
values
of
the
CD
208
(p58C
272-464
)/CD
208
(p58C
266-464
) and
CD
222
(p58C
272-464
)/CD
222
(p58C
266-464
) ratios
to
generate
a scaling
factor
(~0.86),
then
multi-
plying
the
entire
p58C
266-464
spectrum
by
this
scaling
factor.
NMR
spectroscopy
Spectra
were
acquired
using
a Bruker
AV-III
800
MHz
spectrometer
equipped
with
a CPTCI
single-gradient
cryoprobe.
15
N-enriched
p58C
constructs
at
a concentration
of
200
μ
M
were
prepared
in
a buffer
containing
20
mM
MES
(pH
6.5),
50
mM
NaCl,
2 mM
DTT,
and
5%
(v/v)
D
2
O.
Two-dimensional
15
N-
1
H
heteronuclear
single
quantum
coherence
(HSQC)
spectra
were
acquired
at
25 ̊C
with
2,048
and
128
points
in
the
1
H
and
15
N
dimensions,
respectively.
64
scans
were
recorded
in
the
direct
(
1
H)
dimension
for
each
point
sampled
in
the
indirect
(
15
N)
dimension.
Data
were
processed
by
Topspin
(Bruker)
and
analyzed
with
Sparky
(Uni-
versity
of
California).
RNA
primer
generation
Triphosphorylated
RNA
primer
was
transcribed
with
T7
RNA
polymerase
[30]
and
purified
on
a 25%
TBE-polyacrylamide
gel
supplemented
with
8 M
urea
according
to
standard
meth-
ods.
Dried
RNA
pellet
was
resuspended
in
RNAse-free
H
2
O
and
aliquoted
prior
to
further
purification.
RNAs
used
for
binding
assays
were
HPLC
purified
on
a Luna
5
μ
M
C18(2)
100
Å
250X4.6
mm
column
(Phenomenex).
Buffer
A:
0.1
M
ammonium
formate;
Buffer
B:
acetoni-
trile;
flow
rate:
1.5
mL/min.
Purification
program:
1–5%
Buffer
B over
three
minutes,
5–8%
Buffer
B over
22
minutes,
then
80%
Buffer
B for
five
minutes.
RNA
typically
eluted
around
11
minutes.
HPLC-purified
RNA
was
further
validated
with
mass
spectrometry,
which
confirmed
that
a product
of
the
expected
mass
had
been
generated.
Substrate
binding
assays
The
fluorescence
intensity
(FI)
assay
was
performed
with
a Monolith
NT.115
series
microscale
thermophoresis
(MST)
instrument
(NanoTemper)
at
25 ̊C.
Cy5-labeled
18
nt
DNA
template
was
purchased
from
Sigma-Aldrich.
A 1.1:1
ratio
of
8 nt
RNA
and
this
DNA
template
was
annealed
in
annealing
buffer
(20
mM
MES
(pH
7.0),
75
mM
NaCl),
resulting
in
a 25
μ
M
stock
of
annealed,
primed
substrate.
This
stock
was
diluted
to
100
nM
with
MST
buffer
(20
mM
MES
(pH
6.5),
50
mM
NaCl,
2 mM
DTT,
0.05%
Tween).
The
primed
substrate
was
mixed
with
p58C
and
allowed
to
incubate
in
the
dark
at
room
temperature
for
15
min.
Samples
with
a final
substrate
concentration
of
50
nM
were
then
loaded
into
MO-K003
Monolith
NT.115
hydrophobic
capillaries
(NanoTemper)
and
fluorescence
was
measured
at
20%
LED
and
40%
MST
power.
Final
K
D
values
were
calculated
using
the
one-site
total
binding
equation
in
GraphPad
Prism
7.
Titrations
were
completed
after
running
an
SD
test
to
ensure
that
the
con-
centration-dependent
changes
in
fluorescence
intensity
were
not
due
to
adsorption
to
the
capillaries
or
aggregation
of
the
fluorophore
[31].
MST
RNA
primer:
5’-PPP-GGCUCGGA-3’
MST
DNA
template:
5’-Cy5-AAACTCCGAGCCAACATA-3’
Character
ization
of primase
C-termina
l domain
constructs
PLOS
ONE
| https://doi.or
g/10.137
1/journal.po
ne.02093
45
December
18,
2018
5 / 17
Fluorescence
anisotropy
(FA)
was
measured
with
a SpectraMax
M5
microplate
reader
(Molecular
Devices).
A 6FAM-labeled
22
nt
DNA
template
was
purchased
from
Sigma-
Aldrich.
A 1.1:1
ratio
of
12
nt
RNA
and
this
DNA
template
was
annealed
in
annealing
buffer
(20
mM
MES
(pH
7.0),
75
mM
NaCl),
resulting
in
a 25
μ
M
stock
of
annealed,
primed
sub-
strate.
The
stock
was
diluted
to
800
nM
with
DNA
binding
buffer
(20
mM
MES
(pH
6.5),
50
mM
NaCl,
2 mM
DTT).
This
primed
substrate
was
mixed
with
p58C
and
allowed
to
incubate
in
the
dark
at
room
temperature
for
15
min.
Samples
with
a final
substrate
concentration
of
50
nM
were
then
loaded
into
a 384-well
plate
and
polarized
fluorescence
intensities
were
mea-
sured
using
excitation
and
emission
wavelengths
of
485
nm
and
520
nm.
The
fluorescein
con-
trol
experiments
were
performed
with
25
nM
fluorescein
(Sigma
Aldrich)
dissolved
in
DNA
binding
buffer
containing
0.016%
DMSO,
which
was
then
mixed
with
p58C
and
incubated
in
the
dark
at
room
temperature
for
15
min
prior
to
determining
fluorescence
anisotropy
in
the
same
way
as
for
the
p58C-DNA
titrations.
Three
replicates
were
collected
for
each
titration.
Final
K
D
values
were
calculated
using
the
one-site
specific
binding
equation
in
GraphPad
Prism
7;
prior
to
using
this
equation,
each
binding
curve
was
normalized
by
subtracting
the
fluorescence
anisotropy
value
of
the
zero
point
from
each
point
on
the
curve.
K
D
values
are
reported
as
the
mean
±
standard
deviation
of
three
measurements
for
each
variant.
FA
RNA
primer:
5’-PPP-GGACCTCCAGGA-3’
FA
DNA
template:
5’-6FAM-AAACTCCTGGAGGTCCAACATA-3’
Cyclic
voltammetry
(CV)
Sample
preparation
for
electrochemistry.
Multiplexed
chips
were
fabricated
as
described
previously
[19].
p58C
construct
samples
were
stored
prior
to
experiments
in
p58C
storage
buffer
(20
mM
Tris
(pH
7.2),
75
mM
NaCl).
All
p58C
constructs
were
buffer
exchanged
into
HEPES
electrochemistry
buffer
(20
mM
HEPES
(pH
7.2),
75
mM
NaCl)
using
Amicon
ultra
centrifugal
filters
(0.5
mL,
3 kDa
MWCO)
(Millipore
Sigma).
Protein
was
applied
to
the
filter
in
volumes
of
90–140
μ
L
and
centrifuged
for
15
minutes
at
14000
x g at
4 ̊C.
After
centrifuga-
tion,
400
μ
L
of
HEPES
electrochemistry
buffer
was
applied
to
the
filter
and
centrifuged
at
14000
x g for
20
minutes.
This
was
repeated
four
times
to
exchange
p58C
samples
into
HEPES
electrochemistry
buffer.
After
buffer
exchange
and
recovery
of
sample
by
centrifugation
(2
minutes,
1000
x g),
concentrations
of
[4Fe4S]
cluster-containing
p58C
or
mutants
were
mea-
sured
by
using
UV-Visible
spectroscopy
to
monitor
the
absorbance
of
the
[4Fe4S]
cluster
at
410
nm
(extinction
coefficient
= 17000
M
-1
cm
-1
) [19,
32].
Recovered
samples
(approx.
100–
150
μ
L
in
volume)
were
deoxygenated
for
2–3
minutes
with
argon.
Samples
were
then
trans-
ferred
into
the
anaerobic
chamber
(Coy
Laboratory
products).
Prior
to
deposition
onto
the
gold
electrode
surface,
p58C
266-464
samples
were
diluted
with
previously
deoxygenated
HEPES
electrochemistry
buffer
to
a molar
concentration
of
40
μ
M
[4Fe4S]
p58C.
Samples
were
initially
deposited
onto
multiplex
chip
quadrants
in
20
μ
L
volumes
and
the
remaining
sample
depos-
ited
in
a well
of
bulk
solution
above
the
chip
surface,
to
a final
volume
of
200–300
μ
L.
p58C
construct
electrochemistry.
All
electrochemistry
was
performed
using
a CHI620D
potentiostat
and
16-channel
multiplexer
(CH
Instruments)
in
an
anaerobic
glove
chamber.
Multiplex
gold
electrodes
were
part
of
a three
electrode
system
with
an
external
Ag/AgCl
refer-
ence
electrode
(Bioanalytical
Systems)
and
platinum
counter
electrode.
Cyclic
voltammetry
scans
were
performed
at
a scan
rate
of
100
mV/s
over
a potential
range
of
+0.412
V to
-0.288
V
vs.
NHE
or
+0.512
V to
-0.188
V vs
NHE.
Bulk
electrolysis
on
DNA
was
performed
at
an
applied
potential
of
+0.512
V vs.
NHE
for
all
electrochemical
oxidation
reactions.
The
oxidiz-
ing
potential
was
applied
for
at
least
8.33
minutes
for
single
oxidation
reactions
on
a surface,
and
6.67
minutes
for
the
iterative
oxidation
cycles
of
p58C
variants.
The
reducing
potential
Character
ization
of primase
C-termina
l domain
constructs
PLOS
ONE
| https://doi.or
g/10.137
1/journal.po
ne.02093
45
December
18,
2018
6 / 17
was
applied
for
8.33
minutes
in
all
electrochemical
reduction
reactions.
All
bulk
electrolysis
and
cyclic
voltammetry
was
performed
in
previously
deoxygenated
p58C
electrochemistry
buffer
(20
mM
HEPES
(pH
7.2),
75
mM
NaCl).
Charge
transfer
(nC)
in
the
cathodic
peak
of
oxidized
samples
CV
scans
was
assessed
using
the
area
under
the
current
wave
of
the
reduction
signal.
Charge
transfer
was
measured
for
oxidized
samples
using
CHI
software,
assessing
the
area
under
the
reductive
peak
in
CV
after
electrochemical
oxidation.
NTP-dependence
of
elec-
trochemical
signals
was
measured
by
pipetting
a small
volume
(1–3
μ
L)
of
a 0.1
M
ATP
stock
solution
into
each
quadrant
of
the
multiplexed
chip
setup.
Samples
were
added
by
quadrant,
as
physical
barriers
in
the
setup
prevent
diffusion
of
NTPs
between
electrode
quadrants.
After
the
volume
of
ATP
stock
was
deposited
onto
the
electrode
quadrant,
resulting
in
a 2.5
mM
or
5
mM
concentration
of
ATP
in
the
quadrant,
CV
scans
were
measured
(100
mV/s
scan
rate).
Charge
transfer
was
assessed
using
CHI
software;
charge
values
were
determined
by
calcula-
tion
of
the
area
under
the
reductive
and
oxidative
peak
curves.
Midpoint
potentials
of
NTP-
dependent
redox
signals
were
assessed
using
the
peak
selection
function
in
CHI
software.
Results
p58C
can
be
crystallized
in
different
conformations
To
test
if the
differences
between
the
p58C
266-456
and
p58C
272-464
structures
were
a by-product
of
the
differences
in
sequence,
we
produced,
purified,
and
crystallized
a p58C
construct
containing
residues
266–464
in
the
conditions
used
to
crystallize
p58C
272-464
[15,
19].
These
crystals
diffracted
to
1.6
Å
and
the
data
were
phased
using
molecular
replacement
with
the
structure
of
p58C
272-464
(3L9Q).
To
avoid
phase
bias,
residues
315–360
were
excluded
when
defining
the
search
model.
With
these
residues
omitted,
density
for
residues
in
extended
conformation
that
together
formed
a beta-type
interaction
were
clearly
evident
in
the
p58C
266-464
2Fo-Fc
map
(Fig
2).
This
region
was
re-built
manually
in
Coot
,
and
the
final
structure
was
refined
to
2.01
Å
res-
olution.
As
previously
observed
for
p58C
272-464
under
these
conditions,
crystallized
as
a dimer,
with
a disulfide
cross-link
between
the
two
Cys449
residues
and
several
stabilizing
interactions
between
symmetry-related
molecules
(S2
Fig).
In
p58C
266-464
, residues
330–340
and
353–360
in
chain
A and
330–345
and
353–359
in
chain
B are
missing
due
to
disorder.
Disordered
resi-
dues
in
the
same
regions
are
observed
in
the
structure
of
p58C
272-464
. A best-fit
superposition
of
the
two
structures
reveals
they
are
very
similar
(Fig
3),
with
a backbone
RMSD
of
only
0.23
Å
.
This
finding
shows
that
the
differences
in
residues
Leu318-His351
evident
from
comparing
the
previous
p58C
structures
[14,
15,
18]
are
not
intrinsic
to
the
differences
in
the
N-termini
of
the
constructs
but
rather
to
differences
in
the
crystallization
conditions.
Fig
2.
Electron
density
maps
revealing
β
-sheet-like
conform
ation
for
residues
Leu318-His
351.
The
ribbon
diagram
of
the
structure
with
Leu318-Hi
s351
displayed
in
stick
representati
on
is docked
into
the
starting
electron
density
map
around
the
β
-sheet-like
region.
Blue
mesh
represents
the
2F
o
-F
c
map
contoured
at
1
σ
,
Green
mesh
represents
the
F
o
-
F
c
map,
contoured
at
3
σ
.
Figure
made
in
Pymol
using
the
isomesh
command
[33].
https://d
oi.org/10.1371/j
ournal.pon
e.0209345.g0
02
Character
ization
of primase
C-termina
l domain
constructs
PLOS
ONE
| https://doi.or
g/10.137
1/journal.po
ne.02093
45
December
18,
2018
7 / 17