124
Recently determined crystal structures of PcrA helicase
complexed with a DNA substrate have revealed details of the
helicase mechanism. PcrA and UvrD helicases have been
shown to be functional as monomers, challenging previous
suggestions that all helicases are required to be oligomeric.
Crystal structures of the hexameric helicases RepA and T7
gene 4 explain the formation of hexameric assemblies from
identical monomers with RecA-like folds, but their molecular
mechanism remains elusive.
Addresses
Sir William Dunn School of Pathology, University of Oxford,
South Parks Road, Oxford OX1 3RE, UK
*e-mail: wigley@eric.path.ox.ac.uk
Current Opinion in Structural Biology 2000, 10:124–128
0959-440X/00/$ — see front matter © 2000 Elsevier Science Ltd.
All rights reserved.
Abbreviations
HCV hepatitis C virus
NTP nucleotide triphosphate
Introduction
DNA helicases are ubiquitous enzymes involved in almost all
aspects of DNA metabolism [1]. They use energy from NTP
hydrolysis to unwind DNA. Their functional diversity is
demonstrated by the broad spectrum of clinical abnormalities
of inherited human disorders in which known or predicted
DNA helicases are mutated [2]. Despite the large number of
helicases that have been purified and characterised biochem-
ically, their precise molecular mechanism of action has
remained elusive and it is only in the past two to three years
that emerging structural and biochemical information has
allowed us to start piecing together details of this mechanism.
Structural information
Hexameric helicases
Although there is an abundance of information about hexa-
meric helicases from electron microscopy [3],
high-resolution structural information has been more
scant. More structural information has emerged in the past
year, however, and this is reviewed below.
High-resolution X-ray [4
•
] and solution NMR [5
•
] structures
have been reported for the N-terminal domain (N-domain)
of the bacterial replicative hexameric DNA helicase DnaB.
Both structures reveal that the N-domain dimerises, a fea-
ture that is suggested to be relevant in relation to the C
3
quaternary state of the hexamer. The C
3
quaternary state is
thought to be a trimer of dimers, whereas the C
6
quaternary
state is a hexamer of identical monomers. The proposed C
3
to C
6
transition of the hexamer would require large rotations
of the N-domains by at least 120° relative to the C-terminal
domains (C-domains). The crystal structure of the
N-domain of another hexameric helicase, Rho, complexed
to an RNA oligonucleotide shows two molecules bound to
the same RNA molecule [6]. In this protein, the C
3
to C
6
transition is proposed to be induced by the pairing of the N-
domains, which is induced upon binding RNA. This primes
three ATP-binding sites and enhances ATP binding to
these sites (primary sites). ATP hydrolysis is then stimulat-
ed by subsequent RNA binding to sites in the C-domain
(secondary sites), inside the hexameric ring. The enzyme
could then track along the RNA while remaining tethered to
the primary RNA sites of the N-domain.
The crystal structure of a fragment of the hexameric bac-
teriophage T7 gene 4A protein shows six molecules, each
with a RecA-like fold, forming a helical filament [7
••
]. The
helical arrangement of the six molecules in the structure is
unexpected but, by looking down the screw axis, a model
is presented for how a hexameric ring might form
(Figure 1a). The crystal structure of another hexameric
helicase, RepA, reveals how an annular hexameric ring is
formed from monomers with RecA-like folds (W Saenger
et al., personal communication) (Figure 1b). The central
hole is large enough to accommodate ssDNA, but not
dsDNA. The structural homology with RecA had been
predicted, as the RecA hexamer is thought to be a struc-
tural homologue of the hexameric helicases [8,9].
Monomeric and dimeric helicases
Crystal structures of the homologous PcrA [10] and Rep [11]
helicases revealed monomeric enzymes consisting of two
domains [1,2], with each domain comprising two subdomains
(1A, 1B and 2A, 2B) (Figure 1c). In addition, the Rep struc-
ture shows two monomers in different conformations (‘open’
and ‘closed’), both bound to a 16-mer ssDNA. The ‘open’
conformation is also adopted in the apo form, whereas the
‘closed’ conformation is induced by binding to ssDNA. The
significance of these conformations became apparent from
recently determined crystal structures of PcrA complexed
with a single-strand-tailed DNA duplex [12
••
]. One of the
structures is a complex with ADPNP (a nonhydrolysable
analogue of ATP) and Mg
2+
, representing a ‘substrate’ com-
plex (Figure 1e). The second structure contains a sulfate ion
situated in a position equivalent to that occupied in the
active site by the phosphate ion produced after ATP hydrol-
ysis and is therefore mimicking a ‘product’ complex
(Figure 1d). These structures provided us with snapshots of
steps along the catalytic cycle and comparisons between
these structures and that of the apo form have given us
important clues about the molecular mechanism of helicases.
Molecular mechanism of PcrA: the ‘Mexican
wave’ model
The ssDNA-binding region in PcrA (and in Rep) compris-
es a cleft across the top surface of subdomains 1A and 2A.
DNA helicases: ‘inching forward’
Panos Soultanas and Dale B Wigley*
SBA102.QXD 02/17/2000 12:12 Page 124