Structure of the catalytic core of S. cerevisiae DNA polymerase eta: implications for translesion DNA synthesis.
ABSTRACT DNA polymerase eta is unique among eukaryotic polymerases in its proficient ability to replicate through a variety of distorting DNA lesions. We report here the crystal structure of the catalytic core of S. cerevisiae DNA polymerase eta, determined at 2.25A resolution. The structure reveals a novel polydactyl right hand-shaped molecule with a unique polymerase-associated domain. We identify the catalytic residues and show that the fingers and thumb domains are unusually small and stubby. In particular, the unexpected absence of helices "O" and "O1" in the fingers domain suggests that openness of the active site is the critical feature which enables DNA polymerase eta to replicate through DNA lesions such as a UV-induced cis-syn thymine-thymine dimer.
- [Show abstract] [Hide abstract]
ABSTRACT: Quantitating relative (32)P-band intensity in gels is desired, e.g., to study primer-extension kinetics of DNA polymerases (DNAPs). Following imaging, multiple (32)P-bands are often present in lanes. Though individual bands appear by eye to be simple and well-resolved, scanning reveals they are actually skewed-Gaussian in shape and neighboring bands are overlapping, which complicates quantitation, because slower migrating bands often have considerable contributions from the trailing edges of faster migrating bands. A method is described to accurately quantitate adjacent (32)P-bands, which relies on having a standard: a simple skewed-Gaussian curve from an analogous pure, single-component band (e.g., primer alone). This single-component scan/curve is superimposed on its corresponding band in an experimentally determined scan/curve containing multiple bands (e.g., generated in a primer-extension reaction); intensity exceeding the single-component scan/curve is attributed to other components (e.g., insertion products). Relative areas/intensities are determined via pixel analysis, from which relative molarity of components is computed. Common software is used. Commonly used alternative methods (e.g., drawing boxes around bands) are shown to be less accurate. Our method was used to study kinetics of dNTP primer-extension opposite a benzo[a]pyrene-N(2)-dG-adduct with four DNAPs, including Sulfolobus solfataricus Dpo4 and Sulfolobus acidocaldarius Dbh. Vmax/Km is similar for correct dCTP insertion with Dpo4 and Dbh. Compared to Dpo4, Dbh misinsertion is slower for dATP (∼20-fold), dGTP (∼110-fold) and dTTP (∼6-fold), due to decreases in Vmax. These findings provide support that Dbh is in the same Y-Family DNAP class as eukaryotic DNAP κ and bacterial DNAP IV, which accurately bypass N(2)-dG adducts, as well as establish the scan-method described herein as an accurate method to quantitate relative intensity of overlapping bands in a single lane, whether generated from (32)P-signals or by other means (e.g., staining). Copyright © 2014 Elsevier B.V. All rights reserved.DNA Repair 11/2014; 25C:97-103. · 3.36 Impact Factor
- EcoSal Plus. 08/2009; 2009.
- [Show abstract] [Hide abstract]
ABSTRACT: Protein-DNA recognition is a central biological process that governs the life of cells. A protein will often undergo a conformational transition to form the functional complex with its target DNA. The protein conformational dynamics are expected to contribute to the stability and specificity of DNA recognition and therefore may control the functional activity of the protein-DNA complex. Understanding how the conformational dynamics influences the protein-DNA recognition is still challenging. Here, we developed a two-basin structure-based model to explore functional dynamics in Sulfolobus solfataricus DNA Y-family polymerase IV (DPO4) during its binding to DNA. With explicit consideration of non-specific and specific interactions between DPO4 and DNA, we found that DPO4-DNA recognition is comprised of first 3D diffusion, then a short-range adjustment sliding on DNA and finally specific binding. Interestingly, we found that DPO4 is under a conformational equilibrium between multiple states during the binding process and the distributions of the conformations vary at different binding stages. By modulating the strength of the electrostatic interactions, the flexibility of the linker, and the conformational dynamics in DPO4, we drew a clear picture on how DPO4 dynamically regulates the DNA recognition. We argue that the unique features of flexibility and conformational dynamics in DPO4-DNA recognition have direct implications for low-fidelity translesion DNA synthesis, most of which is found to be accomplished by the Y-family DNA polymerases. Our results help complete the description of the DNA synthesis process for the Y-family polymerases. Furthermore, the methods developed here can be widely applied for future investigations on how various proteins recognize and bind specific DNA substrates.PLoS Computational Biology 09/2014; 10(9):e1003804. · 4.83 Impact Factor
Molecular Cell, Vol. 8, 417–426, August, 2001, Copyright 2001 by Cell Press
Structure of the Catalytic Core
of S. cerevisiae DNA Polymerase ?:
Implications for Translesion DNA Synthesis
son et al., 2000a, 2000b; Ohashi et al., 2000; Reuven et
al., 1999; Tang et al., 1999; Wagner et al., 1999; Tissier
et al., 2000). The sequence of these DNA polymerases
is unrelated to that of classical polymerases (Pol I–III in
prokaryotes and Pol ?–? in eukaryotes; Johnson et al.,
The discovery of Pol? has gained added significance
with the subsequent finding that mutations in Pol? are
responsible for an inherited disorder, the variant form of
xeroderma pigmentosum (XP-V; Johnson et al., 1999a;
Masutani et al., 1999). Xeroderma pigmentosum (XP)
patients arehypersensitive tosunlight, andsuffer froma
high incidence of skin cancers. In most of these patients
(belonging to groups XP-A to XP-G), the disease results
otide excision repair (NER; Freidberg et al., 1995). How-
but they are defective in their ability to replicate UV-
damaged DNA (Lehmann et al., 1975; Cordeiro-Stone
et al., 1997). In the majority of cell lines derived from
XP-V patients, Pol? is severely truncated (Johnson et
al., 1999a; Masutani et al., 1999), resulting in a protein
with no polymerase activity. Pol?, thus, is the first DNA
polymerase demonstrated to act as a tumor suppressor
T-T dimer with the same efficiency and fidelity as on
undamaged DNA. Both polymerases insert A’s opposite
the two T’s of the dimer, and on damaged as well as
undamaged DNA, they incorporate wrong nucleotides
with the same frequency of ?10?2–10?3(Washington et
al., 1999, 2000; Johnson et al., 2000c). Yeast Pol? can
also efficiently and accurately replicate DNA containing
7,8-dihydro-8-oxoguanine (8-oxoG) adducts formed by
oxidative damage (Haracska et al., 2000b). Eukaryotic
replicative DNA polymerases tend to insert an A oppo-
site the lesion, as a consequence of which 8-oxoG is
highly mutagenic and causes G:C to T:A transversions.
In contrast, yeast Pol? inserts a C opposite 8-oxoG
(Haracska et al., 2000b).
DNA polymerases with known structures include mem-
bers of the PolI family in prokaryotes, homologs of Pol?
in bacteriophage RB69 and archaebacteria (Wang et al.,
1997; Hopfner et al., 1999; Zhao et al., 1999; Rodriguez
et al., 2000; Hashimoto et al., 2001), and eukaryotic Pol?
(Pelletier et al., 1994). Members of the PolI family include
aquaticus (Taq) DNA polymerase, and phage T7 DNA
polymerase (Ollis et al., 1985; Beese et al., 1993; Kim et
al., 1995a; Korolev et al., 1995; Eom et al., 1996; Doublie
et al., 1998; Kiefer et al., 1998; Li et al., 1998). All of
these DNA polymerases share a similar architectural
plan that resembles a partially opened right hand with
“thumb,” “fingers,” and “palm” domains (Steitz, 1999).
Currently, there is no structural information on Pol? or
any other translesion synthesis DNA polymerase. Con-
sequently, many important questions about the archi-
tecture and the mechanism of these novel polymerases
remain unanswered. Does Pol? have the palm, fingers,
and thumb geometry of classical polymerases? Which
Jose Trincao,1Robert E. Johnson,2
Carlos R. Escalante,1Satya Prakash,2
Louise Prakash,2and Aneel K. Aggarwal1,3
1Structural Biology Program
Department of Physiology and Biophysics
Mount Sinai School of Medicine
New York, New York 10029
2Sealy Center for Molecular Science
University of Texas Medical Branch
Galveston, Texas 77555
DNA polymerase ? is unique among eukaryotic poly-
merases in its proficient ability to replicate through a
variety of distorting DNA lesions. We report here the
crystal structure of the catalytic core of S. cerevisiae
DNA polymerase ?, determined at 2.25A˚resolution.
The structure reveals a novel polydactyl right hand-
shaped molecule with a unique polymerase-associ-
ated domain. We identify the catalytic residues and
show that the fingers and thumb domains are unusu-
ally small and stubby. In particular, the unexpected
absence of helices “O” and “O1” in the fingers domain
suggests that openness of the active site is the critical
feature which enables DNA polymerase ? to replicate
through DNA lesions such as a UV-induced cis-syn
The survival of organisms depends critically on the abil-
ity to faithfully replicate DNA. However, cellular DNA is
continually subjected to damaging agents such as UV
and ionizing radiation, as well as oxidation and hydroly-
sis. A variety of DNA repair pathways has evolved to
repair the resulting lesions, but some lesions escape
repair andare encountered by thereplication machinery
(Freidberg et al., 1995). How cells bypass these lesions
during DNA replication has been a key question in the
areas of DNA replication, mutagenesis, and carcino-
The clearest answer to this longstanding puzzle has
come with the discovery of DNA polymerase ? (Pol?),
the product of the RAD30 gene in Saccharomyces cere-
visiae(Johnson etal.,1999b). Unlikeclassical DNApoly-
merases that become stalled at a UV-induced cis-syn
cyclobutane thymine-thymine (T-T) dimer, Pol? can effi-
ciently and accurately replicate past this common sun-
light-induced lesion (Johnson et al., 1999b). Pol? is able
to replicate through a variety of other distorting DNA
lesions as well (Haracska et al., 2000a, 2000b; Minko et
al., 2001). Pol? is a member of a new family of DNA
polymerases (Johnson et al., 1999c; Goodman and
Tippin, 2000) which includes Pol?/Pol? and Pol? in hu-
mans and DinB (PolIV) and UmuC (PolV) in E. coli (John-
Table 1. Data Collection Phasing and Refinement Statistics
Data CollectionSe-edgeSe-peakSe-remote Native
Number of reflections measured
Data coverage (%)
MAD phasing statistics
Number of sites
FoM (centric/acentric) 3.2A˚c
FoM (DM) 2.25A˚d
Resolution range (A˚)
Reflections, F ? 2? (F)
Average B factor (A˚2)
aValues for outermost shell are given in parentheses.
bRmerge? ? |I ? ?I?|/?|, where I is the integrated intensity of a given reflection.
cFoM ? Mean figure of merit computed to 3.2A˚.
dFoM ? Overall mean figure of merit at 2.25A˚after density modification.
eRcryst? ? ||Fo| ? |Fc||/? |Fo|.
fRfreewas calculated using 10% of data excluded from refinement.
are the putative active site residues? How does the
enzyme replicate past DNA lesions? To address these
questions, we undertook structural analysis of a yeast
Pol? fragment that retains the DNA polymerase and
damage bypass activities of the full-length enzyme. The
structure provides an in-depth look at the geometry of
this important translesion synthesis DNA polymerase
and offers new insights into the mechanism of transle-
sion DNA synthesis.
data, and the phases then extended to 2.25A˚with sol-
vent flattening. An electron density map calculated at
that resolution (2.25A˚) was of excellent quality, allowing
the construction of both copies of Pol? in the crystallo-
graphic asymmetric unit (molecules A and B), without
the need for noncrystallographic symmetry averaging.
The current model includes residues 1–509 for mole-
cules A and B, and 318 water molecules (Table 1).
Palm, Fingers, Thumb, and PAD
Pol? has the shape of a polydactyl right hand, in which
a novel polymerase-associated domain (PAD) mimics
is thus defined by four domains: palm, fingers, and
and the PAD that packs alongside the fingers (Figure
ases and carries the active site residues that catalyze
the nucleotidyl transfer reaction. The fingers and thumb
domains are radically different from those in other DNA
polymerases (Figure 2A).
The palm can be divided into large and small subdo-
mains. The large subdomain contains a mixed 6-stranded
? sheet (?1, ?7, ?8, ?9, ?10, and ?11) flanked by two
long ? helices (?F and ?J) from one side and a short ?
helix (?K) from the other. The side of the ? sheet with
the long ? helices forms part of the hydrophobic core,
while the other side is largely solvent exposed and con-
stitutes the floor of the DNA binding groove (Figure 1).
on the T7 DNA polymerase (T7 Pol) palm domain
Results and Discussion
We have previously shown that yeast Pol? containing
residues 1–513 has the same DNA polymerizing and
damage bypass activities as the full-length enzyme of
632 residues (Kondratick et al., 2001). We chose a dele-
tion from the C terminus because these residues are
the most divergent among translesion synthesis DNA
polymerases (Johnson et al., 1999c). For the structural
work described here, we expressed the C-terminally
truncated yeast Pol? containing residues 1–513 as a
GST fusion and purified the protein from yeast cells.
The GST portion was subsequently cleaved off and te-
tragonal crystals were obtained from solutions con-
taining polyethylene glycol and ammonium acetate, dif-
fracting to 2.25A˚resolution with synchrotron radiation.
The structure was solved by the multiwavelength anom-
alousdiffraction (MAD)method(Hendrickson, 1991),us-
ing selenomethionine-labeled Pol? expressed in E. coli.
The initial MAD phases (3.2A˚) were applied to native
Crystal Structure of DNA Polymerase ?
Figure 1. Structure of Pol? (Residues 1–513)
(A) A ribbon drawing showing the polydactyl right-hand shape of Pol?. Pol? is composed of palm (blue and red), fingers (yellow), and thumb
(orange) domains, and a unique PAD (green). For clarity, the palm ? sheet is drawn in red and the ? helices in blue. Also shown are the active
site residues (Asp30, Asp155, and Glu156) in a ball-and-stick representation. The ? helices (?A to ?S) and ? strands (?1 to ?15) are labeled
sequentially from the N to the C terminus.
(B) The secondary structure and domain topology of Pol?. The secondary structure elements were defined using PROCHECK (Laskowski et
al., 1993). Also indicated are the active site residues Asp30 (as D on strand ?1) and the consecutive Asp155 and Glu156 (as DE on strand
?8). The coloring scheme is the same as in (A).
(Doublie et al., 1998), with strands ?1, ?7, ?8, and ?10
and helices ?F and ?J overlapping onto strands ?9,
?12, ?13, and ?14 and helices ?R and ?Q in T7 Pol,
(rmsd) for the superimposed ?/? substructures in the
two polymerases is 2.1A˚(64 C?’s). The palm domains
of other polymerases can be similarly superimposed,
with rmsd’s ranging from ?1.8A˚ (59 C?’s) for phage
Figure 2. Comparison between Pol? and T7 DNA Polymerase
(A) Pol? (left) and T7 polymerase (right) are aligned based on a superposition of their palm domains. The view differs from that in Figure 1A
by a ?180? rotation about the vertical axis. The protein domains are colored as in Figure 1A. The Pol? fingers and thumb domains are smaller
than the equivalent domains in T7 polymerase. Note also that Pol? fingers domain lacks the equivalent of helices O and O1 (labeled on T7
(B) Comparison between a portion of the palm domain in Pol? (left) and T7 polymerase (right). The colored segments (red for ? strands and
blue for ? helices) superimpose with an rmsd of ?2.1A˚. Also shown are the active site residues, Asp30, Asp155, and Glu156 in Pol? and
Asp475, Asp654, and Glu655 in T7 polymerase.
RB69 Pol? to ?2.4A˚(59 C?’s) for Taq DNA polymerase
(Wang et al., 1997; Li et al., 1998). These superpositions
establish Asp30, Asp155, and Glu156 as the active site
residues in Pol?, aligning, for instance, with Asp475,
Asp654, and Glu655 in T7 Pol (Doublie et al., 1998). As
in T7 Pol, the first carboxylate (Asp30) of this catalytic
triad in Pol? emanates from a ? strand (?1) in the palm
domain that leads into the fingers domain, while the
second and third carboxylates stem from a neighboring
? hairpin (?7 and ?8; Figure 2). The small subdomain is
a cluster of helices (?A, ?B, ?G, ?H, and ?I), whose
location at the base of the palm gives the impression
of a “wrist” to the yeast Pol? hand (Figure 1). Curiously,
main, based on sequence alignment (Figure 3).
The fingers domain is stubby (25A˚ ? 26A˚ ? 33A˚),
containing two small ? sheets (?2, ?3, and ?4; ?5 and
?6) and three short ? helices (?C, ?D, and ?E; Figure
Figure 3. Comparison of Sequences within the Translesion Synthesis DNA Polymerase Family
Included in the comparison are S. cerevisiae Pol? (yPol?), human Pol? (hPol?), human Pol? (hPol?), human Pol? (hPol?), E. coli DinB (ecDinB),
and E. coli UmuC (ecUmuC). Shown above the alignment is the relative location of ? helices and ? strands in the Pol? structure. These
secondary structure elements are colored according to which domain they belong to: palm (blue and red), fingers (yellow), thumb (orange),
and the PAD (green). Also shown above the alignment are the conserved sequence motifs, designated as I–V (Johnson et al., 1999c).
Figure 4. Putative Interactions with Template-Primer
(A) The DNA coordinates (dark blue) were obtained following superposition of the Pol? palm domain onto the equivalent domain in the T7
Pol/template-primer/ddGTP ternary complex (cf. Figure 2B; Doublie et al., 1998). Pol? is shown in the same orientation as in Figure 1A.
(B) A T-T dimer (red) is modeled in the active sites of Pol? (left) and Taq DNA polymerase in the open state (right). The incoming nucleoside
triphosphate is drawn in light blue, and the rest of the template and the primer is in dark blue. Pol? readily accommodates the 5? T of the
T-T dimer, whereas in Taq DNA polymerase, it faces severe clashes.
1). In contrast, the domain in most other DNA polymer-
ases is larger and composed mostly of ? helices. T7
32A˚ ? 42A˚) that contains eight ? helices (Figure 2A;
Doublie et al., 1998), while RB69 Pol? has a domain
characterized by two long ? helices that protrude ?50A˚
from the palm (Wang et al., 1997). However, the most
surprising aspect of the Pol? fingers domain is the lack
of equivalent of helices “O” and “O1” that play a central
role in closing off the active site and in the fidelity of
PolI DNA polymerases (Figure 2A; Doublie et al., 1998,
1999; Li et al., 1998; Suzuki et al., 2000). Instead, a
small loop between helices D and E partially grazes the
entrance to the active site in Pol?.
The thumb is similarly small and stubby (22A˚? 24A˚?
25A˚), comprised of a 90-residue stump at the palm C
terminus (Figure 1). In contrast, the domain in T7 Pol
extends ?40A˚from the base of the palm and, like all
PolI polymerases, is encoded as a large insertion within
of six ? helices (?L, ?M, ?N, ?O, ?P, and ?Q) that are
structurally unrelated to helices in other DNA polymer-
ases (Figures 1 and 2A). The DNA binding surface area
enclosed by the palm, fingers, and thumb domains in
Pol? (675A˚2) is substantially less than in T7 Pol (1630A˚2)
or RB69 Pol? (1135A˚2). This could explain why a Pol?
construct (residues 1–398) containing only the palm,
fingers, and thumb domain is unable to bind and poly-
merize DNA efficiently (Kondratick et al., 2001).
The size of the Pol? hand is augmented by an extra
Crystal Structure of DNA Polymerase ?
domain, the PAD (residues 393–508). The PAD is joined
to the thumb by a flexible tether that traverses the DNA
binding groove from the thumb to the fingers side, a
distance of over 30A˚(Figure 1). The PAD bears uncanny
resemblance to the palm in containing a mixed ? sheet
(?R and ?S) from one side. The two ? sheets are roughly
perpendicular to each other, and are the principal ele-
ments defining the floor and the wall of the DNA binding
groove (Figure 1). Most importantly, the inclusion of the
PAD (13A˚? 15A˚? 49A˚) increases the potential DNA
binding surface of Pol? from 675A˚2to 1113A˚2, compara-
ble to that observed in other DNA polymerases (see
Conserved Motifs in Translesion Synthesis
The Pol? sequence is unrelated to that of classical poly-
merases (Pol I–III in prokaryotes and Pol ?–? in eukary-
otes), but shows significant homology to Rev1 (a deoxy-
cytidyl transferase) in yeast and DinB (Pol IV) and UmuC
(Pol V) in E. coli (Johnson et al., 1999c). Other Pol?-
related proteins have been purified within the last two
son et al., 2000a, 2000b; Ohashi et al., 2000; Tissier et
al., 2000) and E. coli umuC and dinB genes (Reuven et
al., 1999; Tang et al., 1999; Wagner et al., 1999). The
alignment of these sequences reveals five conserved
sequence motifs, designated I–V (Figure 3; Johnson et
ture of these novel DNA polymerases.
Motif I in our structure encodes the ? strand (?1) car-
rying the first catalytic residue (Asp30), while motif III
encodes the ? hairpin (?7 and ?8) carrying the second
and third catalytic (Asp155 and Glu156) residues. Motifs
I and III are the exact structural analogs of conserved
motifs A and C in PolI and Pol? DNA polymerases that
contain the invariant active site residues (Delarue et
al., 1990). Motif II maps to the fingers domain and is
characterized by a conserved YxAR sequence (Figure
(Delarue et al., 1990), with the conserved Tyr and Arg
residues mimicking residues in T7 Pol (such as Arg518
and His506) that interact with the incoming nucleoside
triphosphate (Doublie et al., 1998). Motif IV is marked by
several conserved basic residues, two of which (Arg249
and Lys268) play a structural role in packing helices J
and K against the palm ? sheet, while another two
(Lys272 and Lys279) are in a position to contact the
primer DNA strand, analogous to Arg452 and His704 in
T7 Pol (Doublie et al., 1998). Motif V maps to a region
of the thumb domain facing the DNA binding cleft. The
mappingof theseconservedmotifsto strategicportions
of Pol? (Figure 3) suggests a similar basic structure for
the other related translesion synthesis DNA polymer-
to polymerize DNA and to bypass DNA lesions, we ex-
pect these polymerases to differ from one another in
Figure 5. A Model Comparing the Replication Mechanism between
Replicative DNA Polymerases and Pol?
Replicative DNA polymerases (top) are postulated to contain a tight
active site that accommodates only a single template base. Pol?
(bottom) is shown with a more open active site that can potentially
accommodate two template bases.
important for DNA polymerase and T-T dimer bypass
activities (Kondratick et al., 2001). The fourth acidic resi-
due (Glu39) identified in the mutational analysis appears
to play more of a structural role in maintaining the integ-
rity of the fingers domain. Residues Asp30, Asp155,
and Glu156 are conserved in all Pol?-related translesion
synthesis DNA polymerases and comprise the active
site, aligning with Asp475, Asp654, and Glu655 in T7
Pol (Figure 2B). Based on this structural homology to
T7 Pol, Asp30 and Asp155 are expected to coordinate
two divalent metal ions in the active site, while Glu156
is expected to play a modest role in catalysis. Accord-
ingly, the E156A mutation in Pol? shows a decrease in
catalysis but is not completely inactive like the D30A
and D155A mutant proteins (Kondratick et al., 2001).
Similar results were obtained in a mutagenesis study of
the equivalent catalytic residues in the Klenow fragment
critical for catalysis than Glu883 (Polesky et al., 1990,
1992). Taken together, these structural and biochemical
similarities suggest a common metal-assisted mecha-
nism of catalysis among replicative and translesion syn-
thesis DNA polymerases.
Putative Interactions with Template-Primer
The similarity between the palm domain of Pol? and
that of other DNA polymerases allows both a template-
primer and an incoming nucleoside triphosphate (NTP)
to be modeled into the Pol? DNA binding cleft (Figure
4A). Thus, a superposition with the T7 Pol/template-
primer/ddGTP ternary complex (Doublie et al., 1998) re-
sults in positioning ddGTP in the Pol? active site and
the primer 3? end in the joint between the palm and
fingers domains. The thumb and the PAD straddle the
duplex portion of the modeled template-primer, con-
nected by a long loop that cradles the underside of the
Asp30, Asp155, and Glu156 are three of the four acidic
residues identified in a mutational analysis of Pol? as
to secure the template-primer, with the thumb making
contacts in the minor groove and the PAD interacting
in the major groove. The shape of the extended PAD ?
sheet matches remarkably well to the contour of the
major groove surface, compatible with a role for the
PAD in stabilizing the Pol?/DNA complex.
The role of the PAD may be analogous to that of E.
coli thioredoxin in T7 DNA replication. T7 Pol recruits
thioredoxin to form a tight one-to-one complex that pre-
vents the dissociation of the template-primer during
DNA synthesis (Modrich and Richardson, 1975; Huber
et al., 1987). Thioredoxin binds an extended, flexible
loop within the T7 Pol thumb domain (Doublie et al.,
1998) and—like the PAD—it could swing over to the
fingers side to encircle the template-primer.
tive DNA polymerases usually insert an A opposite the
lesion. This is probably because 8-oxoG, in the absence
formation, favoring the formation of a Hoogsteen base
pair with an adenine. However, it is tempting to specu-
late that the accommodation of an extra unpaired tem-
plate base in the Pol? active site imposes sufficient
backbone constraint or stacking interactions on 8-oxoG
to favor anti over syn conformation, leading to the incor-
porationof Crather thanA. ThePol? structureisthe first
step toward defining the architecture and mechanism
of this remarkable DNA polymerase. In particular, the
“openness” of the active site appears to be the critical
feature which distinguishes Pol ? from replicative poly-
merases, enabling the former to bypass DNA lesions
Mechanism for Bypassing DNA Lesions
One of the most intriguing features of Pol? to emerge
from this DNA modeling is the paucity of putative con-
tacts to the template 5? end. The unpaired bases of the
modeled template 5? end are relatively unhindered in
continuing a helical passage across the Pol? fingers
domains. In contrast, only a single unpaired template
base is held in the active site of T7 polymerase or in
Taq or Bacillus DNA polymerase I, while the preceding
5? unpairedtemplate base(s) isdirected out ofthe active
site at a 90? angle (Doublie et al., 1998; Kiefer et al.,
1998; Li et al., 1998). This steric block comes primarily
from helices O and O1 of the fingers domain (Figure 2A)
and, in the case of Taq polymerase, it is true for both
the closed and open states of the enzyme (Li et al.,
1998). (The Taq open state was obtained by soaking out
the NTP and has a configuration similar to that of apo
enzyme.) Because the 5? T of a T-T dimer (T-T) cannot
be flipped out of the active site due to its covalent cis-
syn cyclobutane linkage to the 3? T (T-T), we suggest
that this may be the reason why replicative polymerases
such as Taq or T7 become stalled at this common UV-
induced lesion. On the other hand, Pol? lacks the O and
O1 helices (Figure 2A), and its active site is much less
restricted in accommodating the 5? T of the T-T dimer
bone around a T-T dimer is relatively undistorted, and
well as their Watson-Crick hydrogen bonding potential
(Kemmink et al., 1987; Kim et al., 1995b). Thus, we pro-
pose that by accommodating two rather than only a
single unpaired template base in the active site, Pol?
can replicate a T-T dimer without becoming stalled. A
tight active site allows replicative DNA polymerases to
better sense the geometry of the nascent base pair, and
thereby achieve fidelities surpassing those from correct
Watson-Crick hydrogen bonding. Pol? incorporates
wrong nucleotides at a substantially higher error rate
(10?2–10?3) than a eukaryotic DNA polymerase such as
Pol? (10?5; Washington et al., 1999, 2000; Johnson et
al., 2000c; Matsuda et al., 2000). The low fidelity of Pol?
is consistent with a more open active site, which is less
specific but better able to accommodate DNA lesions.
Besides a T-T dimer, yeast Pol? can also efficiently and
accurately replicate DNA containing 8-oxoG adducts
(Haracska et al., 2000b). In contrast, eukaryotic replica-
Protein Expression and Purification
The GST-Pol? (residues 1–513) fusion protein (Kondratick et al.,
2001) was expressed in yeast from plasmid pBJ847. This fusion
protein contains a PreScission protease recognition sequence,
LEVLFQGP, which is cleaved specifically between the glutamine
and glycine residues and is located 7 amino acids N-terminal to the
first methionine of Pol?. Yeast strain BJ5464 harboring plasmid
pBJ847 was grown in synthetic complete medium lacking leucine
and induced with galactose as described (Johnson et al., 2000c).
GST-Pol?(1–513) protein was purified as described previously for
the full-length protein (Johnson et al., 1999b) with the following
modifications: prior to affinity purification on glutathione-Sepha-
rose, protein was precipitated from yeast cell extract using 35%–
50% ammonium sulfate. The pellets were then solubilized and
passed over a glutathione-Sepharose column. The Pol?(1–513) pro-
tein lacking the GST tag was eluted from the column by treatment
with PreScission protease (Amersham Pharmacia) and was further
To express the GST-Pol?(1–513) protein in E. coli, the EcoNI/SalI
fragment from pBJ847, containing the GST-Pol?(1–513) fusion, was
used to replace the GST gene in plasmid pGEX-6P-3 (Amersham
Pharmacia), generating plasmid pBJ875. To prepare selenomethio-
nine-labeled GST-Pol?(1–513) protein, plasmid pBJ875 was trans-
formed into an E. coli B834 methionine auxotrophic strain, and cells
were grown in M9 minimal medium supplemented with all amino
acids, except that selenomethionine replaced methionine. Se-Met-
labeled protein was purified in a manner similar to the one used for
purification from yeast, involving affinity purification over a glutathi-
one-Sepharose column and proteolysis with PreScission protease,
followed by a Mono Q column.
Yeast Pol? crystallizes in two crystal forms: orthorhombic and te-
tragonal. We first obtained the orthorhombic crystals from solutions
containing 8% PEG 4K and 700 mM ammonium acetate (pH 6.5),
at 20?C. The crystals belong to space group P212121, with unit cell
dimensions of a ? 86.3A˚, b ? 106.0A˚, c ? 167.6A˚, and ? ? ? ? ? ?
90?. Although these crystals are fairly large (up to 1.5 ? 0.2 ? 0.2
mm), they are hollow and have a diffraction limit of 2.8A˚at home.
The tetragonal crystals were obtained from solutions containing 6%
PEG 20K and 600 mM ammonium acetate, at 4?C. The crystals
belong to space group P41212 with unit cell dimensions of a ? b ?
104.8A˚, c ? 292.3A˚, and ? ? ? ? ? ? 90?. These crystals are smaller
(usually 0.2 ? 0.2 ? 0.05 mm) than the orthogonal form, but they
diffract better and were used for the subsequent structure determi-
Data Collection, Structure Determination, and Refinement
The MAD data were measured at the Advanced Photon Source
(APS, beamline 31-ID), at wavelengths corresponding to the edge
and peak of the selenium K edge absorption profile plus at two
Crystal Structure of DNA Polymerase ?
remote points (Table 1). The positions of the selenium atoms and
the experimental phases were computed with CNS (Brunger et al.,
1998). The initial experimental phases (3.2A˚) were applied to native
data measured at the National Synchrotron Light Source (beamline
X4A), and the phases were then extended to 2.25A˚ with solvent
flattening. This yielded an experimental electron density map that
was readily interpretable without the need for noncrystallographic
averaging. The model for both Pol? molecules (A and B) was built
into this map. The initial model had an R factor of 42.5% (Rfree?
42%), which quickly converged to 22.6% (Rfree? 24.9%) after itera-
tive rounds of refinement with CNS, model building with O (Jones
et al., 1991), and water picking. The final model includes residues
1–509 for molecules A and B, and 318 water molecules (Table 1).
The model has good stereochemistry (Table 1), with 87.6% of the
residues in the most favored conformation in a Ramachandran plot
and only 0.3% in the disallowed regions.
type B DNA polymerase from Thermococcus gorgonarius. Proc.
Natl. Acad. Sci. USA 96, 3600–3605.
Huber, H.E., Tabor, S., and Richardson, C.C. (1987). Escherichia coli
ase and primed templates. J. Biol. Chem. 262, 16224–16232.
Johnson, R.E., Kondratick, C.M., Prakash, S., and Prakash, L.
(1999a). hRAD30 mutations in the variant form of xeroderma pig-
mentosum. Science 285, 263–265.
Johnson, R.E., Washington, M.T., Prakash, S., and Prakash, L.
(1999c). Bridging the gap: a family of novel DNA polymerases that
replicate faulty DNA. Proc. Natl. Acad. Sci. USA 96, 12224–12226.
Johnson, R.E., Prakash, S., and Prakash, L. (2000a). The human
DINB1 gene encodes the DNA polymerase Pol?. Proc. Natl. Acad.
Sci. USA 97, 3838–3843.
Johnson, R.E., Washington, M.T., Haracska, L., Prakash, S., and
Prakash, L. (2000b). Eukaryotic polymerases ? and ? act sequentially
to bypass DNA lesions. Nature 406, 1015–1019.
Johnson, R.E., Washington, M.T., Prakash, S., and Prakash, L.
(2000c). Fidelity of human DNA polymerase ?. J. Biol. Chem. 275,
Jones, T.A., Zou, J.-Y., and Cowan, S.W. (1991). Improved methods
for building models in electron density maps and the location of
errors in these models. Acta Crystallogr. A 47, 110–119.
J.H., and Kaptein, R. (1987).
protons of the duplex d(GCGTTGCG).d(CGCAACGC) containing a
thymine photodimer. Nucleic Acids Res. 15, 4645–4653.
Kiefer, J.R., Mao, C., Braman, J.C., and Beese, L.S. (1998). Visualiz-
crystal. Nature 391, 304–307.
Kim, Y., Eom, S.H., Wang, J., Lee, D.S., Suh, S.W., and Steitz, T.A.
(1995a). Crystal structure of Thermus aquaticus DNA polymerase.
Nature 376, 612–616.
Kim, J.K., Patel, D., and Choi, B.S. (1995b). Contrasting structural
impacts induced by cis-syn cyclobutane dimer and (6–4) adduct in
DNA duplex decamers: implication in mutagenesis and repair activ-
ity. Photochem. Photobiol. 62, 44–50.
Kondratick, C.M., Washington, M.T., Prakash, S., and Prakash, L.
(2001). Acidic residues critical for the activity and biological function
of yeast DNA polymerase ?. Mol. Cell. Biol. 21, 2018–2025.
Korolev, S., Nayal, M., Barnes, W.M., Di Cera, E., and Waksman, G.
(1995). Crystal structure of the large fragment of Thermus aquaticus
bility. Proc. Natl. Acad. Sci. USA 92, 9264–9268.
Laskowski, R.A., MacArthur, M.W., Moss, D.S., and Thornton, J.M.
(1993). PROCHECK: a program to check the stereochemical quality
of protein structures. J. Appl. Crystallogr. A47, 110–119.
Lehmann, A.R., Kirl-Bell, S., Arlett, C.F., Paterson, M.C., Lohman,
P.H.M., de Weerd-Kastelein, E.A., and Bootsma, D. (1975). Xero-
derma pigmentosum cells with normal levels of excision repair have
a defect in DNA synthesis after UV-irradiation. Proc. Natl. Acad. Sci.
USA 72, 219–223.
Li, Y., Korolev, S., and Waksman, G. (1998). Crystal structures of
open and closed forms of binary and ternary complexes of the large
fragment of Thermus aquaticus DNA polymerase I: structural basis
for nucleotide incorporation. EMBO J. 17, 7514–7525.
Masutani, C., Kusumoto, R., Yamada, A., Dohmae, N., Yokoi, M.,
Yuasa, M., Araki, M., Iwai, S., Takio, K., and Hanaoka, F. (1999). The
XPV (xeroderma pigmentosum variant) gene encodes human DNA
polymerase ?. Nature 399, 700–704.
Matsuda, T., Bebenek, K., Masutani, C., Hanaoka, F., and Kunkel,
T.A. (2000). Low fidelity DNA synthesis by human DNA polymerase
?. Nature 404, 1011–1013.
Minko, I.G., Washington, M.T., Prakash, L., Prakash, S., and Lloyd,
R.S. (2001). Translesion DNA synthesis by yeast DNA polymerase
We are grateful to K. D’Amico and C. Ogata for facilitating X-ray
data collection at APS and NSLS, respectively. We thank L. Shapiro
for help with data collection and comments on the manuscript. This
work was supported by NIH grants GM44006 (A.K.A.) and GM19261
(L.P.) and institutional funds (A.K.A.). J.T. is supported by Praxis XXI
Received June 8, 2001; revised July 3, 2001.
1H NMR study of the exchangeable
Beese, L.S., Derbyshire, V., and Steitz, T.A. (1993). Structure of DNA
polymerase I Klenow fragment bound to duplex DNA. Science 260,
Brunger, A.T., Adams, P.D., Clore, G.M., Delano, W.L., Gros, P.,
Grosse-Kunstleve, R., Jiang, W., Kuszewski, J., Nilges, M., Pannu,
N.S., et al. (1998). Crystallography & NMR system: a software suite
for macromolecular structure determination. Acta Crystallogr. D54,
Cordeiro-Stone, M., Zaritskaya, L.S., Price, L.K., and Kaufmann,
W.K. (1997). Replication fork bypass of a pyrimidine dimer blocking
leading strand DNA synthesis. J. Biol. Chem. 272, 13945–13954.
Delarue, M., Poch, O., Tordo, N., Moras, D., and Argos, P. (1990).
An attempt to unify the structure of polymerases. Protein Eng. 3,
Doublie, S., Tabor, S., Long, A.M., Richardson, C.C., and Ellen-
berger, T. (1998). Crystal structure of a bacteriophage T7 DNA repli-
cation complex at 2.2 A resolution. Nature 391, 251–258.
Doublie, S., Sawaya, M.R., and Ellenberger, T. (1999). An open and
closed case for all polymerases. Structure 7, R31–R35.
Eom, S.H., Wang, J., and Steitz, T.A. (1996). Structure of Taq poly-
Freidberg, E.C., Walker, G.C., and Siede, W. (1995). DNA Repair and
Mutagenesis (Washington, DC: American Society for Microbiology).
Haracska, L., Prakash, S., and Prakash, L. (2000a). Replication past
O6-methylguanine by yeast and human DNA polymerase ?. Mol.
Cell. Biol. 20, 8001–8007.
Haracska, L., Yu, S.L., Johnson, R.E., Prakash, L., and Prakash, S.
(2000b). Efficient and accurate replication in the presence of 7,8-
dihydro-8-oxoguanine by DNA polymerase ?. Nat. Genet. 25,
Hashimoto, H., Nishioka, M., Fujiwara, S., Takagi, M., Imanaka, T.,
Inoue, T., and Kai, Y. (2001). Crystal structure of DNA polymerase
J. Mol. Biol. 306, 469–477.
Hendrickson, W.A. (1991). Determination of macromolecular struc-
tures from anomalous diffraction of synchrotron radiation. Science
Hopfner, K.P., Eichinger, A., Engh, R.A., Laue, F., Ankenbauer, W.,
? on templates containingN2-guanine adducts of 1,3-butadiene me-
tabolites. J. Biol. Chem. 276, 2517–2522.
bonucleic acid replication in vitro. A protein of Escherichia coli re-
quired for bacteriophageT7 DNA polymerase activity.J. Biol. Chem.
Ohashi, E., Ogi, T., Kusumoto, R., Iwai, S., Masutani, C., Hanaoka,
F., and Ohmori, H. (2000).Error-prone bypass of certain DNA lesions
by the human DNA polymerase ?. Genes Dev. 14, 1589–1594.
Ollis, D.L., Brick, P., Hamlin, R., Xuong, N.G., and Steitz, T.A. (1985).
Structure of large fragment of Escherichia coli DNA polymerase I
complexed with dTMP. Nature 313, 762–766.
Pelletier, H., Sawaya, M.R., Kumar, A., Wilson, S.H., and Kraut, J.
(1994). Structures of ternary complexes of rat DNA polymerase ?,
a DNA template-primer, and ddCTP. Science 264, 1891–1903.
Polesky, A.H., Steitz, T.A., Grindley, N.D., and Joyce, C.M. (1990).
Identification of residues critical for the polymerase activity of the
Klenow fragment of DNA polymerase I from Escherichia coli. J. Biol.
Chem. 265, 14579–14591.
Polesky, A.H., Dahlberg, M.E., Benkovic, S.J., Grindley, N.D., and
Joyce, C.M. (1992). Side chains involved in catalysis of the polymer-
ase reaction of DNA polymerase I from Escherichia coli. J. Biol.
Chem. 267, 8417–8428.
Reuven, N.B., Arad, G., Maor-Shoshani, A., and Livneh, Z. (1999).
The mutagenesis protein UmuC is a DNA polymerase activated by
J. Biol. Chem. 274, 31763–31766.
Rodriguez, A.C., Park, H.W., Mao, C., and Beese, L.S. (2000). Crystal
structure of a pol ? family DNA polymerase from the hyperthermo-
philic archaeon Thermococcus sp. 9 degrees N-7. J. Mol. Biol. 299,
Steitz, T.A. (1999). DNA polymerases: structural diversity and com-
mon mechanisms. J. Biol. Chem. 274, 17395–17398.
Thermus aquaticus DNA polymerase I mutants with altered fidelity.
J. Biol. Chem. 275, 32728–32735.
Tang, M., Shen, X., Frank, E.G., O’Donnell, M., Woodgate, R., and
Goodman, M.F. (1999). UmuD?2C is an error-prone DNA polymerase,
Escherichia coli polV. Proc. Nat. Acad. Sci. USA 96, 8919–8924.
Tissier, A., McDonald, J.P., Frank, E.G., and Woodgate, R. (2000).
pol?, a remarkably error-prone human DNA polymerase. Genes Dev.
Wagner, J., Gruz, P., Kim, S.-R., Yamada, M., Matsui, K., Fuchs,
R.P.P., and Nohmi, T. (1999). The dinB gene encodes a novel E. coli
DNA polymerase, DNA Pol IV, involved in mutagenesis. Mol. Cell 4,
Wang, J., Sattar, A.K., Wang, C.C., Karam, J.D., Konigsberg, W.H.,
and Steitz, T.A. (1997). Crystal structure of a pol ? family replication
DNA polymerase from bacteriophage RB69. Cell 89, 1087–1099.
Washington, M.T., Johnson, R.E., Prakash, S., and Prakash, L.
(1999). Fidelity and processivity of Saccharomyces cerevisiae DNA
polymerase ?. J. Biol. Chem. 274, 36835–36838.
Washington, M.T., Johnson, R.E., Prakash, S., and Prakash, L.
(2000). Accuracy of thymine-thymine dimer bypass by Saccharo-
myces cerevisiae DNA polymerase ?. Proc. Natl. Acad. Sci. USA 97,
Zhao, Y., Jeruzalmi, D., Moarefi, I., Leighton, L., Lasken, R., and
Kuriyan, J. (1999). Crystal structure of an archaebacterial DNA poly-
merase. Structure 7, 1189–1199.
The structure has been deposited in the Protein Data Bank with the
accession number 1JIH.