recombination-dependent concatemeric viral dna replication

R

R

AD

a

ARRAA

KBPRC

C

MtD

0d

Virus Research 160 (2011) 1–14

Contents lists available at ScienceDirect

Virus Research

journa l homepage: www.e lsev ier .com/ locate /v i rusres

eview

ecombination-dependent concatemeric viral DNA replication

mbra Lo Piano, María I. Martínez-Jiménez, Lisa Zecchi, Silvia Ayora ∗

epartamento de Biotecnología Microbiana, Centro Nacional de Biotecnología, CSIC, Campus de la Universidad Autónoma de Madrid, C/Darwin 3, Cantoblanco, 28049 Madrid, Spain

r t i c l e i n f o

rticle history:eceived 28 April 2011eceived in revised form 7 June 2011ccepted 10 June 2011vailable online 17 June 2011

eywords:reak-induced replicationrimosome assembly

a b s t r a c t

The initiation of viral double stranded (ds) DNA replication involves proteins that recruit and load thereplisome at the replication origin (ori). Any block in replication fork progression or a programmedbarrier may act as a factor for ori-independent remodelling and assembly of a new replisome at thestalled fork. Then replication initiation becomes dependent on recombination proteins, a process calledrecombination-dependent replication (RDR). RDR, which is recognized as being important for replicationrestart and stability in all living organisms, plays an essential role in the replication cycle of many dsDNAviruses. The SPP1 virus, which infects Bacillus subtilis cells, serves as a paradigm to understand the linksbetween replication and recombination in circular dsDNA viruses. SPP1-encoded initiator and replisome

eplication fork reversaloncatemeric DNA synthesis

assembly proteins control the onset of viral replication and direct the recruitment of host-encoded repli-somal components at viral oriL. SPP1 uses replication fork reactivation to switch from ori-dependent�-type (circle-to-circle) replication to �-type RDR. Replication fork arrest leads to a double strand breakthat is processed by viral-encoded factors to generate a D-loop into which a new replisome is assembled,leading to �-type viral replication. SPP1 RDR proteins are compared with similar proteins encoded byother viruses and their possible in vivo roles are discussed.

© 2011 Elsevier B.V. All rights reserved.

ontents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22. Viral recombination proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

2.1. Recombinase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32.1.1. T4-UvsX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32.1.2. T7-gp2.5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32.1.3. RecT-like recombinases (RecT, Red�, G35P) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32.1.4. HSV-1-ICP8 and baculovirus LEF-3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42.1.5. Vaccinia-E9L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2.2. Recombinase mediators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42.3. Exonuclease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2.3.1. T4-gp46/gp47 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42.3.2. T7-gp6 exonuclease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52.3.3. RecE and related exonucleases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52.3.4. HSV-1-UL12. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2.4. SSB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.4.1. T4-gp32 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2.4.2. SPP1-G36P . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2.4.3. Vaccinia-I3L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Abbreviations: DA, DNA arrested; DSB, Double Strand Break; dsDNA, double-strre11–Rad50–Nbs1 complex; ori, origin of replication; OB-fold, oligosaccharide/oligonucl

ion dependent replication; SaPI, Staphylococcus aureus pathogenicity island; SSA, single-sNA; TR, terminal repeat; VLF-1, baculovirus very late expression factor 1.∗ Corresponding author. Tel.: +34 91585 5450; fax: +34 91585 4506.

E-mail address: [email protected] (S. Ayora).

168-1702/$ – see front matter © 2011 Elsevier B.V. All rights reserved.oi:10.1016/j.virusres.2011.06.009

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

anded DNA; HJ, Holliday junction; HSV-1, Herpes-simplex virus type 1; MRN,eotide binding fold; pac, packaging site; PAI, Pathogenicity Island; RDR, recombina-

trand annealing; SSB, single-stranded DNA binding protein; ssDNA, single-stranded

dx.doi.org/10.1016/j.virusres.2011.06.009

http://www.sciencedirect.com/science/journal/01681702

http://www.elsevier.com/locate/virusres

mailto:[email protected]

dx.doi.org/10.1016/j.virusres.2011.06.009

2 A. Lo Piano et al. / Virus Research 160 (2011) 1–14

2.5. HJ-resolving enzymes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62.5.1. T4-Endonuclease VII . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62.5.2. T7-Endonuclease I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62.5.3. Lambdoid Rap/RusA endonucleases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62.5.4. Vaccinia virus HJ resolvase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

2.6. DNA helicases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73. Biological significance of RDR. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

3.1. Replisome assembly via RDR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73.2. RDR of viruses with linear genomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83.3. RDR of viruses with circular genomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

3.3.1. Shift from � to � by replication re-start in phage SPP1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93.3.2. Shift in phage � as a prototype of Proteobacteria infecting viruses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93.3.3. RDR in HSV-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

3.4. Formation of transducing plasmid molecules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113.5. Amplification of pathogenicity islands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

4. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

. . . . . .

1

gcomtaedsmaae(fiSltoiZcD(ut[Dl2�iav(MapDptwm

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. Introduction

DNA replication is established from a single origin in bacterialenomes and in many double-stranded (ds) DNA viruses with cir-ular genomes, but from multiple origins in linear chromosomesf some viruses, or in eukaryotic and archaeal cells. In bacteriaulti-level control mechanisms ensure the assembly of a replica-

ion fork at the origin region, only once per cell cycle, and suchcontrol does not exist in lytic viruses (Zakrzewska-Czerwinska

t al., 2007; Kaguni, 2006; Weigel and Seitz, 2006). Initiation of ori-ependent � (circle-to-circle) DNA replication has been extensivelytudied in �-Proteobacteria, with Escherichia coli and its viruses asodel systems (Kaguni, 2006; Kornberg and Baker, 1992; Weigel

nd Seitz, 2006). It has been shown that the key steps identifiedre common in many bacteria and phages, although with differ-nt number of players. Initiation of DNA replication in Firmicutese.g., Bacillus subtilis) and their viruses (e.g., virus SPP1) requiresrst the wrapping of the origin region (oriC in bacteria or oriL inPP1) by the origin recognition protein (e.g., DnaA or G38P) fol-owed by localized unwinding of the adjacent AT-rich region, wherehe replicative hexameric DNA helicase (e.g., DnaC in B. subtilisr G40P in SPP1) bound to its loader (e.g., DnaD-DnaB and DnaIn B. subtilis or G39P in SPP1) is delivered (Ayora et al., 1999;akrzewska-Czerwinska et al., 2007). Then the activated helicaseoordinates all events at the fork. The viral hexameric replicativeNA helicase G40P directly loads the host encoded DnaG primase

Ayora et al., 1998; Barcena et al., 1998; Lecointe et al., 2007), andpon interaction with the � subunit of the clamp loader it loadshe polymerase (Martínez-Jiménez et al., 2002). This holoenzymePolCDnaE(�2)(�4��′)] is composed by two polymerases [PolC andnaE, (Dervyn et al., 2001)], the sliding clamp (�2), and the clamp

oader complex [including �, �, and �′ subunits; Sanders et al.,010]. Bacterial and SPP1 ori-dependent replication proceed by the(circle-to-circle) mechanism. In general, these steps of replication

nitiation are conserved in other circular viruses that replicate by� mechanism, just the number of virus-encoded functions may

ary. As an example, the genome of herpes simplex virus type 1HSV-1) encodes seven essential replication proteins (reviewed by

uylaert et al., 2011). These include an ori binding protein (UL9),nd six core replication proteins including a DNA polymerase com-osed by two subunits (UL30 and UL42), ICP8, a single-strandedNA binding protein (SSB), and a helicase/primase complex com-
osed of three subunits (the helicase UL5, the UL52 primase andhe accessory protein UL8). The UL8 accessory protein interactsith the UL5/52 complex, and also with UL9, ICP8 and DNA poly-erase (Muylaert et al., 2011). UL9 binds to the ori, distorts it, and
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

recruits additional factors, as ICP8 to assemble the replisome. Sub-sequently, UL9, through its interaction with the UL8 loading proteinmay recruit the viral helicase–primase to initiate primer synthe-sis and to establish a bi-directional replication fork. The viral DNApolymerase may be recruited to the site of initiation either by theinteraction of its catalytic subunit (UL30) with UL8, or via an inter-action between its processivity factor (UL42) and UL9 (reviewed inBoehmer and Nimonkar, 2003).

Initiation of replication of dsDNA viruses with linear genomesdoes not completely follow these general rules. In phage T4, noorigin binding protein exists. In addition, T4 contains several repli-cation origins (Brister and Nossal, 2007). Though few obvioussequence characteristics are shared between them, all of the T4origins are thought to facilitate formation of RNA primers usedto initiate leading strand DNA synthesis. The presence of an R-loop presumably holds the origin duplex in an open conformation,giving the T4 gp41/61 primosome complex access to allow exten-sive parental DNA unwinding and priming (reviewed in Kreuzerand Brister, 2010). Several viral proteins, in addition to the heli-case/primase are required for significant replication of these R-loopsubstrates: DNA polymerase (gp43), polymerase clamp (gp45),clamp loader (gp44/62), and SSB (gp32). While the gp41 helicasecan be loaded onto DNA without the helicase loader (gp59), thepresence of gp59 greatly accelerates the process. The phage SSB,gp32, seems to help also in the recruitment of the gp41 helicase,but also recruits, by direct protein–protein interaction another heli-case, Dda (Ma et al., 2004). The Dda helicase, seems to be alsoinvolved in initiation of replication, although it is not an essentialcomponent (Jongeneel et al., 1984).

During the replication process there is a trade-off between tran-scription, replication, and other DNA transactions that could leadto the stall or collapse of the replication machinery. In bacteria,re-start of replication from a fortuitous or programmed replicationfork barrier might simply involve resumption of DNA synthesis bya PriA-mediated replisome reassembly system without necessar-ily invoking a recombination event and a switch in the replicationmode (reviewed in Cox et al., 2000; Gabbai and Marians, 2010;Heller and Marians, 2006; Kreuzer, 2000; Michel et al., 2004). Inviruses, recombination-dependent reassembly of the replisome isneeded to resuscitate a replication fork with a switch from � toRDR in circular genomes, which is visualized as a Greek “�” let-ter by electron microscopy, or from early to late replication in
linear viruses. Both, � type or late viral replication, proceed viaRDR. This replication mode generates the DNA substrate (con-catemeric viral DNA) to be encapsidated into an empty proheadto generate a mature viral particle. Functions usually involved in

A. Lo Piano et al. / Virus Research 160 (2011) 1–14 3

Table 1Common RDR factors present in some dsDNA viruses.

Virus Recombinase Mediatora End resectionb SSBa HJ resolvasea DNA helicase Replication typec

T4 UvsX UvsY gp46/gp47 gp32 EndoVII (gp49) gp41, UvsW, Dda From liner DNAT7 gp2.5 gp6 gp2.5 Endo I (gp3) gp4 From linear DNA� Red� Orf? Red Ea10? Rap (NinG) � → �Rac RecT RecE ?SPP1 G35P G34.1P G36P G44P G40P � → �HSV-1 ICP8 UL12 ICP8 UL5/8/52 � → �?Vaccinia E9L E9L I3L A22R ?Baculovirus LEF-3 Alkaline nuclease LEF-3 VLF-1? ?

a The enzymes annotated with? have only been proposed to have this activity.es ar

.

Dvs(tatHttpusc

2

2

trrpT2ibbvr

2

oesrSeao

riiPm(mb

b gp46/gp47 and E9L are 3′ → 5′ exonucleases, the other viral end resection enzymc ? denotes that it is unknown if replication starts from a circular or a linear DNA

NA recombination provide the means to generate concatemericiral DNA (Table 1). Bacterial viruses are excellent simple modelystems for genetic and biochemical studies of these processesAdhya et al., 2005; Mosig et al., 1995; Viret et al., 1991). Inhis review, we will analyze the recombination proteins whichre usually present in many dsDNA viruses, focusing on viruseshat replicate via circular replication intermediates (e.g., SPP1, �,SV-1) or from a linear genome (e.g., T4 and T7) where most of

he biochemical studies have been performed. In a second part,he different scenarios where the recombination proteins couldarticipate are discussed, focusing mainly on the mechanism(s)nderlying RDR, with special emphasis in the poorly understoodwitch from the origin-mediated circle-to-circle � type to rollingircle or � type RDR; specially in the SPP1 virus.

. Viral recombination proteins

.1. Recombinase

Many dsDNA viruses encode a recombinase (Table 1), which ishought to be required for the circularization of genomes, for DNAepair and/or for RDR. So far, three major superfamilies of phageecombinases emerged, either related to Rad51 (T4-UvsX, and theoorly characterized Sak4), to T7 gp2.5, or to Rad52 (RecT-like).hey are present in both, virulent and temperate phages (Iyer et al.,002; Lopes et al., 2010). Eukaryotic viral single-strands anneal-

ng (SSA) proteins have been also described, but they have not yeteen placed in any of these recombinase superfamilies. We willriefly summarize their differences and interactions with otheriral recombination components, because their spread has beenecently described (Lopes et al., 2010).

.1.1. T4-UvsXThe T4 UvsX protein is a member of the ATP-dependent class

f general recombination enzymes typified by bacterial RecA orukaryotic Rad51. Like RecA, UvsX forms presynaptic filaments onsDNA, which are the obligatory nucleoprotein intermediates inecombination, and filament formation is inhibited by a cognateSB (Ando and Morrical, 1998; Carrasco et al., 2008). UvsX, as RecA,xhibits both ssDNA-dependent ATPase and DNA strand exchangectivities, and the crystal structure closely resembles the structuref RecA (Gajewski et al., 2011).

UvsX exhibits a strict requirement for the mediator UvsY, thatecruits UvsX onto gp32-coated ssDNA (Bleuit et al., 2001). Thiss a contrast to the other phage recombinases, which are ATP-ndependent, and/or lack a recombination mediator (Table 1).rotein interactions have been described not only with the UvsY
ediator, but also with the T4 branch migration helicase, UvsW,
Gajewski et al., 2011), and with Endo VII (the HJ resolvase). Endo VIIediated HJ-resolution is inhibited by the UvsX and UvsY proteins

y a direct protein–protein interaction (Birkenkamp-Demtroder

e 5′ → 3′ exonucleases.

et al., 1997). The UvsW helicase promotes the formation of a nickedcircular product by UvsX and gp32 in the absence of UvsY (Gajewskiet al., 2011). Thus, the efficiency of the strand exchange reactiondepends on the balance in the concentrations of different proteinplayers.

2.1.2. T7-gp2.5The T7-encoded gp2.5 recombinase shares functional and struc-

tural homology with SSB. The structure of gp2.5 shows a conservedOB-fold (oligosaccharide/oligonucleotide binding fold) that is welladapted for interactions with ssDNA (Hollis et al., 2001). However,in contrast to SSB it also provides a recombinase function throughits SSA activity (Rezende et al., 2003). gp2.5, together with thehelicase/primase gp4, mediates homologous DNA strand exchange(Kong and Richardson, 1996). The COOH-terminal 21 amino acidsof gp2.5 are essential for specific protein–protein interactions withgp1 (T7 DNA polymerase) and with the helicase/primase, gp4(Ghosh et al., 2010; He et al., 2003; Kong and Richardson, 1998;Marintcheva et al., 2006).

2.1.3. RecT-like recombinases (RecT, Redˇ, G35P)The RecT family of viral recombinases is related to eukaryotic

Rad52 (Iyer et al., 2002; Lopes et al., 2010). They have an ATPindependent annealing activity and no high single-strand bind-ing affinity. The best-studied representatives are RecT from theE. coli Rac defective prophage, Red� from phage �, and G35P fromSPP1. Although placed in the same superfamily, RecT, Red�, andG35P share almost no sequence identity. The proteins are usuallyencoded in operons together with an interacting 5′ → 3′ exonucle-ase (Datta et al., 2008; Muyrers et al., 2000). No crystal structureis known for any of the members, but several electron microscopystudies have shown that this family of proteins, similarly to Rad52,forms rings and filaments alone and/or with ssDNA (Ayora et al.,2002; Passy et al., 1999; Thresher et al., 1995). By atomic forcemicroscopy studies it has been recently shown that Red� doesnot form rings but a shallow helix, which is disrupted by ssDNA,and a stable filament when annealing two complementary ssDNAs(Erler et al., 2009). In contrast, G35P forms stable filaments withssDNA (Ayora et al., 2002). Nevertheless, it seems that in all thecases filament formation is an important step in the recombinationreaction, and despite of its independency of ATP, their mechanismof recombination may have similarities with the one used by therecombinases of the RecA-Rad51 superfamily. In support of thisidea, it was shown that RecT binding to ssDNA mediates unstackingof the bases. This property is a hallmark of homology recognitionby RecA-like proteins (Noirot et al., 2003).

Except the described interaction of the recombinase with the
5′ → 3′ exonuclease, which seems to be conserved in all the family,other protein–protein interactions are poorly known in this fam-ily of recombinases. In SPP1, G35P was shown to interact physicallywith the phage SSB (G36P), and with the replicative helicase (G40P),

4 s Rese

bebu(

2

i2fHapiroIballMc

wIcipptstb(ftm2

trmtRaoSshai

2

hPhdjt3p

A. Lo Piano et al. / Viru

ut the significance of these interactions remains unknown (Ayorat al., 2002). Although no direct protein–protein interaction haseen described with the exonuclease, it was shown that G35P stim-lates G34.1P exonuclease activity on dsDNA, but not on ssDNAMartinez-Jimenez et al., 2005).

.1.4. HSV-1-ICP8 and baculovirus LEF-3The recombinases of eukaryotic viruses have not been classified

nto any of the previously mentioned superfamilies (Iyer et al.,002), but they share many of the features of the recombinasesrom phages (Muylaert et al., 2011). The best characterized is theSV-1 ICP8 protein, which as T7-gp2.5, has both ssDNA bindingnd SSA activities (Makhov and Griffith, 2006). ICP8 is a 128 kDarotein encoded by the HSV-1 UL29 gene. It was identified orig-

nally as a SSB, and, in this capacity, facilitates HSV-1 replicationeactions in vitro by its ability to destabilize duplex DNA duringrigin unwinding (Boehmer and Lehman, 1993a). Furthermore,CP8 catalyzes invasion of ssDNA into dsDNA and strand transferetween ssDNA and homologous tailed duplex DNA (Nimonkarnd Boehmer, 2003). As shown for other recombinases, ICP8 formsong left-handed helical filaments (Makhov et al., 2009). Simi-arly, the baculovirus SSB, LEF-3, forms oligomers, and promotes

g2+-independent unwinding of DNA duplexes and annealing ofomplementary DNA strands (Mikhailov et al., 2006).

The HSV-1 ori recognition protein, UL9, interacts specificallyith ICP8 (Boehmer et al., 1994; Boehmer and Lehman, 1993b), and

CP8 facilitates replication in vitro by the HSV-1 DNA polymeraseomplex UL30/42 (Nimonkar and Boehmer, 2004). A model for RDRn HSV-1 based on the ability of ICP8 to anneal ssDNA with com-lementary duplex DNA resulting in the formation of D-loops wasroposed. These D-loop intermediates nucleate the assembly ofhe viral replisome (Nimonkar and Boehmer, 2003). In vitro, it washown that a heterologous SSB could substitute for ICP8 in the syn-hesis, indicating that species-specific protein–protein interactionsetween the SSB and other replisome components are not crucialNimonkar and Boehmer, 2004). An acidic C-terminal tail, essentialor protein–protein interaction, and present in many SSBs, seemso be absent in ICP8. At least, a 60-amino acid C-terminal deletion

utant of ICP8 also binds very strongly to UL9 (Manolaridis et al.,009).

ICP8 is also able to bind RNA and catalyze R-loop forma-ion, which suggests that viral transcripts may be used to initiateeplication by R-loop formation, in a recombination-dependentechanism (Boehmer, 2004; Muylaert et al., 2011). It also explains

he observed role of ICP8 in late viral gene expression. In vitroNA binding and R-loop formation has not been described forny other viral or phage recombinase, but phage T4 initiatesrigin-dependent replication via an R-loop mechanism in vivo (seeection 3.2; Dudas and Kreuzer, 2005). ICP8 has also been demon-trated to promote DNA strand exchange in concert with the HSV-1elicase–primase complex (Makhov and Griffith, 2006; Nimonkarnd Boehmer, 2003). A possible role could be to promote RDR dur-ng repair of DSBs (Nimonkar and Boehmer, 2002).

.1.5. Vaccinia-E9LPoxviruses are large dsDNA viruses that are subjected to very

igh frequencies of genetic recombination during viral replication.oxvirus recombination reactions show all the molecular geneticallmarks of simple SSA reactions, including the transient pro-uction of abundant hybrid DNA and evidence of processing of

oint regions by a 3′ → 5′ exonuclease, that promotes recombina-ion (Gammon and Evans, 2009). This is done in this case by the′–5′ proofreading exonuclease domain of the vaccinia virus DNAolymerase (E9L; Hamilton et al., 2007). E9L can catalyze SSA reac-

arch 160 (2011) 1–14

tions in vitro, and this reaction is stimulated by the vaccinia SSB(I3L;Willer et al., 2000).

2.2. Recombinase mediators

T4 UvsY is a recombination mediator that promotes the assem-bly of the UvsX-ssDNA filament. The action of UvsY has parallels inother systems, as the bacterial RecO/RecOR mediator system thathelps RecA to overcome the inhibitory effect exerted by SsbA (Ayoraet al., 2011). UvsY helps UvsX to displace T4-gp32 from ssDNA, areaction necessary for proper formation of the presynaptic filament(Pant et al., 2008). UvsY binds dsDNA and ssDNA, but preferentiallyloads UvsX onto ssDNA under physiological conditions (Xu et al.,2010). The formation of competent UvsX-ssDNA filaments requiresa hand-off of ssDNA from UvsY to UvsX, with the efficiency of thishand-off controlled by the relative ssDNA-binding affinities of thetwo proteins (Farb and Morrical, 2009). In addition to its interac-tion with gp32 and UvsX, an interaction of UvsY with the gp46/gp47ATPase/exonuclease complex has been also described (Bleuit et al.,2001).

Phage � Orf substitutes for the activities of the E. coli RecFOR pro-teins in vivo (Sawitzke and Stahl, 1992) and is therefore implicatedas a recombination mediator, encouraging the assembly of bacte-rial RecA onto SSB-coated ssDNA. Orf associates with E. coli SSB(Curtis et al., 2011). However, if Orf plays any role in Red-Red�recombination has not been shown.

A protein with a genuine recombinase mediator activity seemsto be absent in the other viruses. In some cases, it seems that therecombinase is also the SSB (as gp2.5, ICP8 and LEF-3), or that therecombinase is directly loaded onto the naked ssDNA generated bythe action of its partner 5′ → 3′ exonuclease. In SPP1, it seems toexist a direct connection between the recombinase (G35P) and theviral SSB (G36P), so that a mediator function would be unnecessary(see Section 2.4.2).

2.3. Exonuclease

Many recombinases of the RecT-superfamily are geneticallylinked with a 5′ → 3′ exonuclease (Datta et al., 2008). In the case ofother recombinases, as the ones of the T4 and T7 viruses, their genesare not in an operon with a gene coding for an exonuclease, but itexists in the genome an exonuclease that could help the recom-binase by creating the necessary substrate for the recombinationreaction.

2.3.1. T4-gp46/gp47gp46/gp47 is a member of the Rad50/Mre11 family of

ATPases/exonucleases which includes the SbcCD nuclease fromE. coli (Connelly and Leach, 2002). gp46/47 is required for recombi-nation and host-DNA degradation. The complex, which is essentialfor phage amplification, was found to be membrane associated,what hampered its purification and characterization for longtime (Mickelson and Wiberg, 1981; Mosig, 1998). It was initiallyreported that gp46 may be a 5′ → 3′ exonuclease that is stimu-lated by gp47 and that the gp46/gp47 complex exhibits specificprotein–protein interactions with the UvsY mediator. This interac-tion was thought to recruit UvsY to the ssDNA generated by thegp46/gp47 complex, so that the UvsX recombinase is loaded on it(Bleuit et al., 2001). The recent success in the purification of bothcomponents has shown that the complex has many biochemicalsimilarities with the eukaryotic Rad50/Mre11 complex: T4-gp47(Mre11) is a gp46 (Rad50)-dependent and Mn2+-dependent dsDNA
exonuclease and ssDNA endonuclease. gp46 (Rad50) is a relativelyinefficient ATPase, but the presence of gp47 (Mre11) and dsDNAincreases ATP hydrolysis by 20-fold (Herdendorf et al., 2011). Direc-tionality assays indicate that the prevailing activity is a 3′ → 5′

s Rese

dRaaerbc

2

mt1gt1acdthgi

2

RgtppMholas2

i(rd5iM

2

htbtStsPbub(tIbv


sDNA exonuclease, which is incompatible with its proposed role inDR. However, the authors have found that in the presence of UvsY,nd the phage SSB (gp32), gp47 uses Mg2+ as a cofactor for its nucle-se activity and degrades dsDNA endonucleolitically (Herdendorft al., 2011). It was shown that the T4 origin fork undergoes forkegression in vivo and that this regression creates a double strandreak (DSB) where the gp46/gp47 complex would act to allow repli-ation fork reactivation (Long and Kreuzer, 2008).

.3.2. T7-gp6 exonucleaseAlthough early studies suggested that gp6 5′ → 3′ exonuclease

ight be involved in phage T7 replication and/or recombina-ion, its precise role remains unknown (Engler and Richardson,983; Sadowski and Vetter, 1976). Phage T7 packages its dsDNAenome from a head-to-tail concatemer, and gp6 is required forhe generation of the packaging substrate (White and Richardson,987). The gp6 enzyme stimulates concatemerization of T7 DNAnd the mechanism of stimulation could be by the generation ofomplementary single-stranded T7 DNA ends by exonucleolyticigestion (Serwer et al., 1990; Son and Serwer, 1992). In vitro,he strand exchange mediated by the gp2.5 recombinase and gp4elicase/primase was enhanced by the ressecting activity of thep6 exonuclease (Kong and Richardson, 1996). No protein–proteinnteraction has been described.

.3.3. RecE and related exonucleasesThe � Red system, constituted by the Red exonuclease and the

ed� or � recombinase, is essential for phage growth in a recA back-round. The Red system, alone or in concert with RecA, is requiredo generate linear concatemers of the � genome. Red is a highlyrocessive 5′ → 3′ exonuclease that degrades dsDNA, and this highrocessivity is probably due to its toroidal structure (Kovall andatthews, 1997; Subramanian et al., 2003). Although no sequence

omology exists between the exonucleases present in RecET-likeperons, the overall mechanism of action could be conserved. Ateast the structure of RecE from the E. coli Rac prophage also showstoroidal tetramer that forms a central channel of similar size and

hape as the one seen in the � exonuclease trimer (Zhang et al.,009).

RecE and � exonuclease each form a specific protein–proteinnteraction with their respective SSA protein, RecT or � proteinMuyrers et al., 2000). The SPP1 G34.1P exonuclease facilitates G35Pecombinase action. G34.1P, as RecE and � exonuclease, is a Mg2+-ependent enzyme that processively digests the dsDNA to form′-mononucleotides and 3′-ended ssDNA tails. The maximum activ-

ty occurs at alkaline pH (Martinez-Jimenez et al., 2005; Vellani andyers, 2003).

.3.4. HSV-1-UL12UL12 of HSV-1 is an alkaline 5′ → 3′ exonuclease that shares

omology with � exonuclease (Reuven et al., 2003). The pheno-ype of UL12 deficient mutants is complex, but suggests that it maye involved in processing viral replication intermediates, consis-ent with a role in recombination (Wilkinson and Weller, 2003).imilarly to the action of G35P on G34.1P, ICP8 inhibited UL12 diges-ion of ssDNA, but stimulated digestion of dsDNA to catalyze DNAtrand exchange (Reuven et al., 2003; Reuven and Weller, 2005).urified UL12 and ICP8 proteins interact (Thomas et al., 1992),ut the interaction of UL12 with other viral components remainsnknown (Muylaert et al., 2011). Recently, a specific interactionetween UL12 and components of the cellular Mre11–Rad50–Nbs1MRN) complex has been observed. The MRN complex is an impor-
ant factor in the decision of repair by homologous recombination.t recognizes a DSB and resects the ends, preparing them for recom-ination. It was also shown that Rad50 is important for efficientirus growth (Balasubramanian et al., 2010). However, if UL12 may
arch 160 (2011) 1–14 5

guide the MRN complex towards the viral genome resulting in endresection, or the UL12–MRN interaction acts to regulate MRN, toinfluence the repair pathway choice most beneficial for HSV-1,remains unknown.

Baculovirus encodes a related alkaline exonuclease, and nullmutants in this gene display similar defects in DNA processing,resulting in aberrant encapsidation and nuclear egress (Okano et al.,2007). Thus, the role of viral alkaline exonucleases in DNA repli-cation may be evolutionarily conserved among different dsDNAviruses.

2.4. SSB

As already mentioned, some SSA recombinases are also SSBs,and as such have been already described. We will describe herebriefly the properties of the genuine helix destabilizing SSBs thatare encoded by the viruses.

2.4.1. T4-gp32Gene 32 is an essential gene. gp32 not only binds to ssDNA,

but also to RNA, and as such is one of the three T4 characterizedtranslational repressors (Miller et al., 2003). It contains three mod-ular domains, “A domain,” “B domain,” and “core.” The central coredomain contains the ssDNA binding site consisting of an OB-fold,a motif conserved in many SSBs (Shamoo et al., 1995). The acidic,C-terminal A domain is the site of essential protein–protein interac-tions with several T4 DNA replication and recombination proteins,including gp59, Dda, gp43, and gp61 (Formosa and Alberts, 1986).gp32 interacts with UvsY, what facilitates the loading of UvsXon ssDNA (Formosa et al., 1983). Similarly to E. coli SSB, gp32exerts both positive and negative effects on filament assembly byUvsX: positive by denaturing ssDNA secondary structures, and neg-ative by competing with UvsX for ssDNA binding (Liu et al., 2006).gp32 interaction with gp59 reduces the affinity of gp32 for ssDNA,enabling the loading of the replicative helicase, gp41, onto DNA(Ma et al., 2004; Trakselis et al., 2001). Stimulation of DNA synthe-sis by the T4 Dda helicase requires direct gp32-Dda interactions(Ma et al., 2004). The dependence of the role of Dda helicase ininitiation of replication on protein–protein interactions with gp32appears to parallel the relationship between the UL9 helicase andICP8 of HSV-1. The UL9–ICP8 complex has been implicated in theunwinding of HSV-1 origins of replication. Not all the interactionsof gp32 are restricted to its C-terminus, and sometimes they havenot been mapped. Recently, direct associations between gp32 andE. coli NDP kinase, T4 ribonucleotide reductase, and thymidylatesynthase were also described (Kim et al., 2005). This suggests that,as E. coli SSB, gp32 can also act as an organizer/mobilizer of genomemaintenance complexes (Shereda et al., 2008).

2.4.2. SPP1-G36PUnder the control of the early promoter 3 (PE3), and immedi-

ately downstream of G35P, there is a gene whose product sharessubstantial homology (from 32 to 41% identity) with E. coli and B.subtilis SSBs (Alonso et al., 1997). The E. coli and B. subtilis SSBs inter-act with at least 14 and 12 proteins involved in DNA metabolism,respectively, among them with some replisome components asDnaG primase, and the subunit of the DNA pol III in E. coli, or withDnaE and PriA in B. subtilis (Costes et al., 2010; Shereda et al., 2008).Most, if not all, of these interactions are mediated by the amphi-pathic C-terminus of SSB. G36P also has an acidic C-terminal tail, butits role in protein–protein interactions has not been addressed. Due
to the lack of conditional-lethal B. subtilis and SPP1 SSB mutants,the essentially of SsbA or G36P for phage growth has not beenaddressed (Weise et al., 1994). G36P has been reported to inter-act with the recombinase G35P (Ayora et al., 2002). The SSA and

6 s Rese

s(

lSmawu

2

leai(ieh

2

mnpLc2osbip2ttp

2

opdmErVmTpaef

2

b(EidrRa


trand exchange activity catalyzed by G35P are stimulated by G36PMartinez-Jimenez et al., 2005).

In the genomes of other studied viruses, which code for RecT-ike recombinases (i.e., in the Rac and � phages), the existence of aSB is not apparent. Ea10 from � is the most abundant early proteinade by the phage (Drahos and Hendrix, 1982), and is annotated

s “putative SSB”. Ea10 does not share homology with any otherell-characterized phage SSBs and if Ea10 binds to ssDNA remainsnknown.

.4.3. Vaccinia-I3LThe I3L gene product is a protein made early in infection that

ocalizes to the cytoplasmic sites of viral DNA replication and pref-rentially binds ssDNA. A domain containing conserved aromaticnd charged residues thought to mediate DNA binding, and presentn many SSBs, appears to be present in the I3L protein as wellRochester and Traktman, 1998). Aggregation of ssDNA by gpI3Ln the presence of MgCl2 could facilitate the SSA activity (Tsengt al., 1999). Other viral proteins with pure ssDNA binding activityave not been described.

.5. HJ-resolving enzymes

A four-way DNA or Holliday junction (HJ) is the central inter-ediate in homologous recombination. HJ-resolving enzymes are

ucleases that are highly selective for branched DNA, and areresent in all organisms from bacteria to humans (Declais andilley, 2008; Sharples, 2001). The bacterial resolvases work in con-ert with the RuvAB branch migration translocase (Canas et al.,008). The viral HJ-resolving enzymes, which do not have an obvi-us connection with a branch migration helicase, lack cleavageequence specificity. The best characterized are T4 Endo VII (whichelongs to the HNH endonuclease family), T7 Endo I (belong-

ng to the endonuclease superfamily), and RusA (from the DLP12rophage, with homologs in many phages) (Declais and Lilley,008; Lilley and White, 2001). In spite of their differences in struc-ure, their mechanism of action is similar. The enzymes are dimershat specifically recognize the HJ intermediate, and distort it in arevious step necessary for cleavage (Declais and Lilley, 2008).

.5.1. T4-Endonuclease VIIEndo VII (also termed gp49) is not essential for RDR but for res-

lution of branches from newly synthesized DNA, a step prior toackaging into preformed heads. In vivo the T4 packaging motoreals with Y- or X-structures present in the replicative concate-er. It was recently shown in vitro, by employing a portal bound

ndo VII, that its resolution activity trims and releases these DNAoadblocks to make packaging substrates (Dixit et al., 2011). EndoII was the first enzyme shown to resolve HJs into duplex DNAolecules by introducing symmetrical nicks in equivalent strands.

he cleavage of HJs by Endo VII is inhibited by the UvsX and UvsYroteins. The inhibition is due to a direct protein–protein inter-ction between Endo VII, UvsX and UvsY (Birkenkamp-Demtrodert al., 1997). Association of Endo VII with portal vertex protein gp20rom the prohead was also demonstrated (Golz and Kemper, 1999).

.5.2. T7-Endonuclease IThe phage T7 HJ-resolving enzyme (Endo I, also termed gp3)

elongs to the same superfamily as the Firmicutes RecU resolvaseAyora et al., 2004), or the archaeal HJ resolvases (White, 2011).ndo I plays the important role of resolving HJs that occur dur-ng homologous recombination of T7 DNA. Endo I also promotes
egradation of host chromosomal DNA in order to supply the T7eplication system with nucleotides (Sadowski, 1974). AlthoughDR has not been demonstrated in the T7 replication system, gp3,nd RDR could play a role in replication, at least all the RDR enzymes
arch 160 (2011) 1–14

are coded by the T7 genome (Table 1). Defects in processivity of theT7 DNA polymerase can be suppressed by reducing the endonucle-ase activity of gp3 (Lee et al., 2009). If the enzyme interacts withother replication and/or recombination proteins remains unknown.

2.5.3. Lambdoid Rap/RusA endonucleasesLambdoid phages have often homologs of the RusA HJ resolvase

or of the Rap endonuclease. RusA was identified through the iso-lation of spontaneous mutations capable of suppressing the DNArepair defect of E. coli ruv mutants. The mutations were foundto up-regulate expression of the rusA gene located in the crypticDLP12 prophage (Mahdi et al., 1996). RusA homologs are presentin numerous phages as phi82, HK97, HK022 and r1t (Sharples, 2001;Sharples et al., 2002). RusA-like enzymes show a high level of selec-tivity for DNA junctions, and it was also shown that they haveappreciable affinity for duplex DNA. However, the DLP12-encodedRusA enzyme does not show DNA cleavage activity with duplexsubstrates (Macmaster et al., 2006).

In phage SPP1, it was observed a small polypeptide, G45P, withlow, but significant degree of identity with RusA (Bolt et al., 2000).The DNA region was re-sequenced and a gene product with homol-ogy with RusA was identified, now called G44P (Lo Piano, Zecchi andAyora, unpublished results). The biology of the RusA-like enzymesremains unknown, but at least in phage SPP1 G44P is not essentialfor replication, as evidenced by a series of SPP1 deletion mutantsidentified, which carried a deletion of the SPP1 resolvase. How-ever, although the poor growth phenotype associated with thesemutants was attributed to defects in packaging (Chai et al., 1993),it could be due also to the lack of the resolvase.

The rap (ninG) gene of phage �, located in the non-essentialninR region, has also been linked with phage recombination. Raphomologs can be found in the nin region of other coliphages as 21,H-19B, and P22 (Sharples, 2001). Rap+ phages show an increase inrecombination between � and a plasmid (Hollifield et al., 1987). TheRap function is also active in the Red pathway of recombination(Tarkowski et al., 2002). Rap is a structure-specific endonucleasetargeted to the branch points of DNA molecules. It cleaves a vari-ety of branched structures as HJs, D-loops, forks and flap branchedstructures (Sharples et al., 1998; Sharples et al., 1999). The cleav-age was observed around the central core of the junctions withno obvious sequence specificity and no symmetrical cleavage, incontrast with HJ resolvases like RuvC or RusA. However, when Rapwas incubated with larger HJs the major products observed weresymmetrical cleavages (Sharples et al., 2004).

2.5.4. Vaccinia virus HJ resolvaseThe vaccinia virus A22R gene encodes a resolvase that has some

homology with the bacterial RuvC resolvase. A22R is involved in thereplication and processing of viral DNA into unit-length genomes;in fact an A22R mutant exhibited a conditional replication defect.Late-stage viral DNA replication was reduced, and most of thenewly synthesized viral DNA remained in a branched or concate-meric form (Garcia and Moss, 2001). The A22R resolvase seems tobe targeted specifically to HJs, although nicking of a three-strandedversion of a HJ was also detected. No endonuclease activity was seenwith a Y-junction and certain other branched substrates (Garciaet al., 2000). Similarly to phage resolving enzymes, the poxvirusenzyme exhibited little cleavage sequence specificity (Garcia et al.,2006).

A22R is the only eukaryotic viral resolvase characterized so far.In cells transfected with a baculovirus very late expression fac-tor 1 (VLF-1) knockout mutant, aberrant tubular structures were
observed, suggesting that this mutant was defective in producingmature capsids. It was proposed that VLF-1 could function as a DNAresolvase on branched DNA structures generated during the DNAreplication process, and that this activity would be required for

s Research 160 (2011) 1–14 7

pa

2

r(chtBrsnra

pUiifDi(rrsmiafssibpa2

iu2srs1RiUoehdoot

3

3

b

Fig. 1. Restart of DNA replication by fork reversal and Holliday Junction (HJ) for-mation. (1) The replisome (ovals) stalls by a block (red box). (2) The replisomedisassembles and the fork reverses. (3) HJ resolvases (double arrows) recognise the4-strand intermediate and cleave it. (4) The lesion/block is removed, and the dsDNAend resected. (5) The resulting end invades a homologous duplex (by the actionof a recombinase). (6) The 4-strand recombination intermediate is resolved by HJresolvases. (7) The replisome assembles on the new fork. The arrows in DNA indicate


rocessing during packaging (Vanarsdall et al., 2006), but such anctivity has not been demonstrated.

.6. DNA helicases

Some viruses as T4, T7, SPP1 and HSV-1 encode for their owneplicative DNA helicase (T4 gp41, SPP1 G40P) or helicase–primaseT7 gp4, HSV-1 UL5/8/52), whose main role seems to be DNA repli-ation, but that could also participate in RDR. At least in vitro, itas been shown that they stimulate the strand exchange reac-ion (Kodadek and Alberts, 1987; Kong et al., 1997; Nimonkar andoehmer, 2002; Salinas and Kodadek, 1995). The phage encodedeplicative helicases have been studied extensively, because theyerve as model systems for more complex helicases. Their mecha-ism of action, and their roles in replication have been previouslyeviewed (Enemark and Joshua-Tor, 2008; Gai et al., 2010; Patelnd Picha, 2000) and are beyond the scope of this review.

Phage T4 also encodes for two other DNA helicases that couldarticipate more directly in DNA recombination: the Dda andvsW proteins. Dda, gp41, and UvsW are all capable of stimulat-

ng branch migration, but with clearly different biological rolesn T4 recombination. Complex interplays between the three dif-erent DNA helicases are likely to modulate many aspects of T4NA metabolism. Dda stimulates leading strand DNA synthesis

n vitro, but is non-essential for phage DNA replication in vivoMa et al., 2004). However, it is believed to be involved in T4eplication initiation (Perumal et al., 2010). Additionally, it mayegulate recombination both positively and negatively at differenttages: presynaptic filament formation and postsynaptic branchigration. Dda inhibits UvsX-mediated strand pairing reactions

n vitro (Kodadek et al., 1988). However, the late addition of Ddafter synapsis stimulates the rate of branch migration more thanour-fold (Kodadek and Alberts, 1987). Furthermore, a species-pecific protein–protein interaction might be important for thistimulation, because Dda cannot stimulate RecA. Protein–proteinnteractions between Dda and the C-terminal domain of gp32 haveeen described (Ma et al., 2004). Furthermore, Dda is capable of dis-lacing protein blocks in the path of translocation of the enzyme,nd this could be also important for recombination (Perumal et al.,010).

The T4 UvsW helicase has been shown to play a crucial rolen the switch from ori-dependent replication to RDR through thenwinding of R-loop initiation intermediates (Dudas and Kreuzer,001). UvsW also unwinds branched DNA substrates as HJs, and Y-ubstrates (Webb et al., 2007). A null uvsW deletion mutant showededuced overall phage–phage recombination, and displayed highensitivity to hydroxyurea and UV irradiation (Derr and Kreuzer,990). UvsW has been described as a functional analog of E. coliecG, because UvsW could complement some of the defects found

n E. coli recG mutants (Carles-Kinch et al., 1997). However, thevsW structure reveals a dynamic four-helix bundle with homol-gy to the nucleic acid binding module of RecQ-like helicases (Kerrt al., 2007). Consistent with this, UvsW is a 3′–5′ helicase that alsoas SSA activity, as RecQ-like helicases have. It was also recentlyemonstrated a structural similarity between UvsW and eukary-tic Rad54 protein, suggesting that the core recombinatorial triadf Rad51/Rad52/Rad54 in eukaryotic cells is functionally equivalento the UvsX/UvsY/UvsW triad in T4 (Kerr et al., 2007).

. Biological significance of RDR

.1. Replisome assembly via RDR

The progress of replication forks is often threatened in vivo, bothy DNA damage and by proteins bound to the template. Blocked

the 3′-ends. Newly synthesized DNA is shown as thin lines. Parental DNA is shownas thick lines.

forks must be reactivated, and the original blockage cleared, inorder to complete genome duplication, implying that blocked forkprocessing may be critical for genome stability. Recent reviewsexist about the many recombination mechanisms proposed thatcould lead to replication restart after replication fork stalling(Atkinson and McGlynn, 2009; Cox et al., 2000; Masai et al., 2010;Michel et al., 2004) and we will only discuss briefly here the oneswhich resemble viral RDR. One possible pathway is replicationfork reversal (Fig. 1, steps 1 and 2, Michel et al., 2004). It involvesthe unwinding of blocked forks to form four-stranded structuresresembling HJs. The cleavage of the four-stranded structure by a HJresolvase would generate a dsDNA end (Fig. 1, step 3). Then, thisend could be processed by exonucleases to promote the loadingof strand exchange proteins. This would result in D-loop forma-tion with the intact sister duplex (Fig. 1, steps 4 and 5). Loading ofthe replication machinery onto this D-loop and resolution of theconnected HJ would restore an intact replication fork, and thenreplication fork progression could resume, assuming the originalblock was cleared (Fig. 1, steps 5 and 7). In � Proteobacteria (E. colias prototype), there are several pathways for primosome assemblyoutside oriC, depending on PriA, and at least one PriA-independentpathway, but dependent on PriC (Heller and Marians, 2006; Gabbaiand Marians, 2010). PriA has two major roles in replication restart.The first is as a 3′ → 5′ DNA helicase that remodels stalled fork struc-tures to generate a ssDNA region on the lagging-strand templatefor replicative helicase, DnaB, loading. The second is to orchestrateori-independent replisome loading at the stalled fork (Sandler et al.,
1999). In B. subtilis replication re-start depends on PriA, DnaB (notto be confused with E. coli DnaB replicative helicase), DnaD andDnaI proteins. In this bacterium, the presence of PriB, PriC and DnaT

8 s Research 160 (2011) 1–14

frarci

3

ianiroah2taTinnsasoUaflcoprhwct2

mmlTpRDttaebsc(Tetuile(

Fig. 2. RDR of linear viral molecules. (A) Model for T4 RDR. (1) The early mode ofT4 replication amplifies viral molecules shorter than unit length. To generate a fullmolecule, T4 uses strand invasion in a homologous region of another phage moleculeto generate a D-loop. (2) Semiconservative replication is initiated at the D-loop.(3) Branch migration and ligation create a Holliday junction, which can be cleavedin either of two orientations (only one is shown). (4) The replication of the fullmolecule is completed. The invasion into a second molecule, associated with DNAreplication, generates long branched molecules, and upon resolution a concatemericDNA molecule (not shown). (B) Model for RDR in T7. (1) The replication of the linearT7 DNA is inherently incomplete. (2) Owing to the presence of ∼160 bp long directTRs (yellow boxes), replication intermediates with 3′-overhangs can hybridize toeach other, forming dimers with one single TR in the center. Dimers could initiatesecondary initiation events, resulting in the cyclical growth of concatemer lengths(not shown). (3) A single nick is introduced at the left of the central TR. (4) Thegenerated 3′-OH is used for strand displacement synthesis, and a DSB is made at theother site of the TR. (5) One full phage molecule is produced, and the other end isprocessed by the gp6 exonuclease that degrades the newly synthesized DNA. (6) Theresulting 3′-end is then extended to replicate the TR and to complete concatemer


unctions are not obvious (Lecointe et al., 2007). Unlike the bacte-ial system, SPP1 replication is independent of PriA, DnaB, DnaDnd DnaI functions (Ayora et al., 2002). It is likely that the phageeplication restart mechanism used for the switch from � to � repli-ation is composed exclusively by phage encoded factors, at leastt is independent of the PriA-replisome loading machinery.

.2. RDR of viruses with linear genomes

The evidence that RDR takes place at late times during T4nfection comes from the isolation of viral mutants with a “DNA-rrested” (DA) phenotype. In these mutants, replication initiatesormally from a linear genome but then ceases. In addition to that

t was found that mutations in the same genes could reduce botheplication and recombination (Mosig, 1998). T4 DNA synthesisriginates from at least five discrete loci near oriA, oriC, oriE, oriF,nd oriG (Brister and Nossal, 2007) by using the 3′OH end of theost RNA polymerase transcript (R-loop) as a primer (Jones et al.,001; Weigel and Seitz, 2006). Early replication is performed byhe T4 encoded enzymes and has been reviewed elsewhere (Dudasnd Kreuzer, 2005; Kornberg and Baker, 1992; Miller et al., 2003;rakselis et al., 2001). The switch between ori-dependent and ori-ndependent RDR relies on the number of infecting viruses, and isot completely clear (Kreuzer and Brister, 2010). Due to the linearature of the DNA, when the assembled fork reaches the chromo-omal end, incomplete linear duplex molecules with 3′-ssDNA tailsccumulate (Dudas and Kreuzer, 2005; Mosig et al., 1995). ThissDNA region is coated by gp32. UvsY facilitates the nucleationf UvsX onto gp32-coated viral molecules (see Section 2.2). ThevsX-ssDNA filament invades a homologous duplex DNA formingD-loop that can be converted into a fully functional replication

ork (Fig. 2A, step 1), by the interaction of gp32 with the helicaseoader, gp59 (Arumugam et al., 2009), that loads the gp41 heli-ase (Fig. 2A, step 2). This D-loop may be within the interior regionf a co-infecting T4 DNA molecule, because T4 DNA is circularlyermuted. After multiple rounds of RDR the newly replicated DNAesults in a very complex branched network rather than a simpleead-to-tail linear molecule (Mosig et al., 1995). The branched net-ork DNA may be de-branched by the Endo VII HJ resolvase in

oncert with the HJ translocase, UvsW, to generate a linear head-o-tail concatemer (Kreuzer, 2000; Mosig et al., 1995; Mosig et al.,001) (Fig. 2A, steps 3 and 4).

T7 is another phage whose replication starts from a linearolecule. If RDR takes place in phage T7 is unkonwn, but T7 hasany if not all of the proteins required (Molineux, 2006). Simi-

arly to T4, T7 starts origin-mediated DNA replication when the7-encoded DNA polymerase, gp5, displaces the T7-encoded RNAolymerase (gp1), that has initiated transcription (Hamdan andichardson, 2009; Molineux, 2006). gp2.5 binds to the displacedNA strand at the R-loop, and the gp4A helicase/primase facili-

ates the loading of gp5, that works in concert with host-encodedhioredoxin (gp5/thioredoxin) at the R-loop (He et al., 2003; Hend Richardson, 2004; Sugimoto et al., 1987). Replication of a lin-ar molecule that has initiated at any internal origin is incompleteecause the ends of the template strand cannot replicate. To avoidhortening of the ends after several replication rounds and to form aoncatemer, T7 uses the presence of ∼160-bp long terminal repeatsTR) (Dreiseikelmann et al., 1980) (see Fig. 2B, steps 1 and 2).he 3′-tailed TRs of the first round of replication can hybridize toach other, forming a dimer via complementary base-pairing ofhe ends. If gp2.5 participates in this annealing reaction remainsnknown. These substrates are subject to RDR or to secondary

nitiation events, resulting in the cyclical growth of concatemerengths (Molineux, 2006). The result of this process would be a lin-ar oligomer with only one single TR copy between two genomesFig. 2B, step 3). How the single TR between genomes is duplicated

processing. Newly replicated DNA is shown in dashed black. Sites of endonucleasecleavage are shown by white arrows.

to produce mature progeny is still poorly understood. A covalentlyclosed hairpin containing one copy of the TR, might be involved inthe duplication of the TR (Chung et al., 1990). Alternatively, pausingor terminating gp1 transcription allows the non-template strand tobe nicked by the DNA terminase, and strand displacement synthesisallows duplication of the TR (Fig. 2B, steps 3 and 4). The large head-to-tail concatemers are then processed to be encapsidated into anempty prohead (Molineux, 2006). The gp6 exonuclease is knownto stimulate concatemerization and the mechanism of stimulationis believed to be the generation of complementary ssDNA ends byexonucleolytic digestion (Fig. 2B, step 5) (Serwer et al., 1990; Sonand Serwer, 1992). These ssDNA tails could anneal to form branchedDNA molecules, in a process similarly to T4 RDR. Branched struc-tures are not appropriate substrate for DNA packaging into emptyproheads, and concatemer formation may require additionally theT7 EndoI enzyme (Chung et al., 1990). In fact, T7 mutants in gp3 orgp6 show a premature replication shut off. However if this is dueto a direct participation in a RDR mechanism, or if this is becausethese enzymes are providers of the nucleotide pool by host DNAdegradation remains unknown.

3.3. RDR of viruses with circular genomes

Upon infection, the virus linear chromosome either remains inits linear form (see above) or circularizes either by annealing of the

s Rese

cdCmbtpie(iai2S

3

mechcniualaiPaaas1o(atre1(mocoe�t

mgS1ttream(tTi


ohesive ends (e.g., � cos sites), by the use of the terminal redun-ancy (e.g., SPP1) or by direct end-to-end ligation (e.g., HSV-1).ircular viruses with terminal redundancy, as SPP1, initiate origin-ediated replication by the � (circle-to-circle) mode (see above),

ut at some point, after at least one round of � replication, switcho RDR or � type. This replication mode is required to producerogeny in high yield, and to generate a linear concatemer that

s the obligatory intermediate used by the DNA packing machin-ry to encapsidate a mature DNA molecule into an empty proheadAyora et al., 2002; Taylor and Wegrzyn, 1995). Circularization andnitiation by � type replication followed by RDR might not only bective in �, SPP1 and many other phages, but also in HSV-1 virus andn mitochondrial and chloroplast DNA (Skalka, 1977; Szczepanska,009; Strang and Stow, 2005; Pohjoismaki et al., 2009; Ling andhibata, 2004; Shutt and Gray, 2006; Oldenburg and Bendich, 2004).

.3.1. Shift from � to � by replication re-start in phage SPP1Once SPP1 is adsorbed to the B. subtilis surface it injects its linear

ature DNA, which consists of 104% of the genome. Transcription ofarly genes helps to finish with viral internalization. SPP1 DNA thenircularizes due to annealing of the terminal redundant ends and aost-encoded DNA ligase seals the ends. This annealing of the endsan be performed by redundant functions, because the absence ofo single host- or SPP1-encoded recombination function impairs

ts circularization (Alonso et al., 2006). SPP1 replication proceedsnidirectionally from an origin (Ganesan et al., 1976; McIntosh etl., 1978). Analysis of SPP1 conditional-lethal replication mutantsed to the identification of several complementation groups thatre transcribed from the five early promoters (PE1 to PE5). Mutantsn genes 38, 39 and 40, which are located under the control ofE2 showed a block in DNA replication (D0 phenotype), whereasmutant in gene 35, located under the control of PE5, showed a DArrest phenotype (Alonso et al., 2006). SPP1 replication begins soonfter the early operons are expressed, but is independent from tran-cription of the early PE2 promoter into the oriL region (Pedre et al.,994). SPP1 oriL, which lies within gene 38, shows a high degreef identity with oriR that is located in a late non-coding regionPedre et al., 1994). G38P binds to oriR albeit with 3- to 4-fold lowerffinity than to oriL (Missich et al., 1997). G38P, G39P and G40P arehe only SPP1-encoded functions necessary and sufficient to driveing-to-ring � replication from the cis-acting oriL region in an oth-rwise non-replicative element in B. subtilis cells (Missich et al.,997). SPP1 � type replication also requires host-encoded: DnaGDNA primase), DNA polymerase holoenzyme, and DNA topoiso-

erases (Pedre et al., 1994). During SPP1 replication accumulationf daughter circles from the initial circular template (SPP1 � repli-ation) and/or branched replication intermediates have not beenbserved (Alonso et al., 2006). It is likely, therefore, that even atarly times a switch from � to � (RDR) replication takes place, andreplication leads to the generation of SPP1 genome concatemers

hat are the substrate for DNA packaging (Alonso et al., 2006).Electron microscopy analysis of SPP1 replication intermediates

apped the SPP1 initiation site at an area approximately 0.16enome length from the left terminus (Ganesan et al., 1976). ThePP1 oriL region was mapped within this region (Missich et al.,997). However, by reversible inhibition of SPP1 DNA synthesis,he selective labeling with radioactive thymidine located the ini-iation site at an area approximately 0.2 genome length from theight terminus (McIntosh et al., 1978), where oriR is located (Alonsot al., 1997). Hence, the oriL and oriR regions, that map at the left andt the right end of the packaging initiation (pac) site respectively,ight correspond to the previously mapped SPP1 origin regions

Ganesan et al., 1976; McIntosh et al., 1978). These data suggesthat in vivo SPP1 may use both origins for initiation of replication.his is consistent with electron microscopy data that showed thatn a plasmid molecule that contained both origins, G38P was found

arch 160 (2011) 1–14 9

to bind simultaneously to both origins promoting DNA pairing (LoPiano and Ayora, unpublished results).

From the available biochemical information obtained by in vitroanalysis of separated proteins and their interactions, a model wasproposed for switch from � to � replication (RDR) and the genera-tion of concatemeric DNA (Martinez-Jimenez et al., 2005) (Fig. 3).First a DSB is produced when the replication fork encounters G38Pbound to oriR (Fig. 3A, step 1). The stalled SPP1 replication forkmight collapse spontaneously, or by the action of an uncharac-terized SPP1-encoded product. Fork reversal and HJ formationgenerate a DSB, probably by G44P cleavage, as described in Section3.1 (Fig. 3A, steps 2 and 3), which is processed by the G34.1P exonu-clease to generate the proper substrate for loading G35P (Fig. 3A,step 4). For the generation of concatemeric DNA two models wereproposed. In both models, the first event (Fig. 3B, step 1) will bethe G35P promoted DNA strand invasion on another supercoiledDNA phage molecule, as it has been documented in vitro (Ayoraet al., 2002). In one of the models, after Skalka (Skalka, 1977) aD-loop specific endonuclease will process the recombination inter-mediate, and the resulting ends will be ligated (Fig. 3B, steps 2–4).The enzyme responsible for this endonuclease activity could be theG44P HJ resolvase. G35P will recruit, by protein–protein interac-tion G36P, which protects the ssDNA, as well as the replicative DNAhelicase, G40P (Ayora et al., 2002). Alternatively, G35P and G38P-bound to oriR (Ayora et al., 1999), in concert load G39P-G40P-ATP tothe D-loop. In a second step, G40P bound to the ssDNA recruits, bydirect protein–protein interaction the replication proteins from thehost, and concatemeric DNA synthesis starts in a � type mechanismsimilar to the plasmid rolling circle mechanism.

In model II (Fig. 3B, steps 5 and 6), a mechanism that does notinvoke the activity of a D-loop specific endonuclease and that sharessome features with an early model was proposed (Ayora et al., 2002;Formosa and Alberts, 1986; George et al., 2001). Here, similarly asdescribed above, G35P and G40P, by protein–protein interactions,will recruit the encoded host components, so that the full replisomeand unidirectional replication is re-established. In both models, thenewly synthesized concatemeric DNA, up to ∼ 4 genome equiv-alents (Chai et al., 1995), contains the pac signal in the properorientation to initiate DNA encapsidation. This event will coupleconcatemeric DNA replication and DNA packaging. Model I is mech-anistically similar to replication fork re-activation after host DSBrepair, but leads to the accumulation of SPP1 concatemers (Alonsoet al., 2006; Ayora et al., 2002). Only SPP1 proteins participate, sincethe switch to � replication is independent of host-encoded compo-nents of the B. subtilis replisome assembly machinery (e.g., PriA,DnaB, DnaD) (Ayora et al., 2002; Lecointe et al., 2007), or the hostrecombination apparatus [namely, RecA, RecF, RecO, RecG, RuvAB,AddAB (counterpart of E. coli RecBCD) and RecU (counterpart ofE. coli RuvC) proteins (Ayora, S., unpublished results)].

3.3.2. Shift in phage � as a prototype of Proteobacteria infectingviruses.

After linear � DNA injection into E. coli cells, the DNA moleculecircularizes due to spontaneous pairing of the cohesive ends (12-bp5′-overhangs), and the action of the host DNA ligase. Initiation of�-type DNA replication in phage � relies on a cis-acting replicationorigin (�-ori) region, and on transcription from the PR promoterof the early right operon that encodes for two viral products (areplisome organiser, �-O, and the helicase loader, �-P) (Taylor andWegrzyn, 1995; Wegrzyn and Wegrzyn, 2005; Weigel and Seitz,2006). The rest of the replication apparatus is host-encoded andrecruited to the �-ori directly or indirectly by the �-O and �-P
proteins (Taylor and Wegrzyn, 1995; Wegrzyn and Wegrzyn, 2005).
Replication of � DNA begins by transcription from the PR pro-moter onto �-ori and the assembly of �-O to the region (O-some).�-P binds to the host DnaB helicase (Dodson et al., 1986) and this

10 A. Lo Piano et al. / Virus Research 160 (2011) 1–14

Fig. 3. Model for the shift from � to RDR in SPP1, and the generation of concatemeric linear DNA. (A) The generation of a DSB. (1) G38P (grey oval) bound to oriL or oriR mighthinder the progression of the replisome (white ovals). (2) The fork reverses and a HJ is formed. (3) The HJ resolvase (G44P) cleaves this substrate and a DSB is formed. (4) Theend is processed by the 5′–3′ exonuclease (G34.1P, not shown) and the G35P recombinase (red ovals) will filament in the generated 3′ ssDNA. (B) Generation of concatemericDNA. (1) G35P promotes strand invasion on another supercoiled SPP1 molecule adjacent to oriR. Model I: in steps 2–4, a D-loop specific endonuclease (G44P), together withD rationg us bloo

cbG(rtpperpaptWtDco�itCedr

vfT

NA ligase, will generate the proper substrate for � type DNA replication and geneeneration of concatemeric DNA is proposed. The DNA that comes from the previof endonuclease cleavage are shown by white arrows.

omplex is recruited to the O-some by protein–protein interactionetween O and P. This complex is re-modelled by DnaK, DnaJ andrpE chaperones to release �-P from the nucleoprotein complex

Liberek et al., 1988), so that the DnaB helicase is activated. Initiallyeplication proceeds both bidirectionally and unidirectionally byhe � mechanism, whereas in bacteria devoid of a functional DnaArotein, replication from �-ori is predominantly unidirectional, toroduce daughter circles from the parental substrate during thearly stages of infection. At later times (10–15 min) after infectioneplication shifts to the � mode (Taylor and Wegrzyn, 1995) toroduce the linear concatemers (Narajczyk et al., 2007). The mech-nism of this switch to � replication remains to be elucidated. It isroposed that DnaA concentration may be an important factor forhe switch from � to � replication mode (Taylor and Wegrzyn, 1995;

egrzyn and Wegrzyn, 2005). Amplification of � DNA, which con-ains multiple DnaA-binding sites, might cause a depletion of freenaA protein that would impair transcription from PR. Less effi-ient transcriptional activation of ori� might then allow the loadingf only one DnaB helicase complex, leading to unidirectional and-type replication. Another model suggested that the metabolic

nstability of �-O may play a relevant role in the switch from � to �-ype replication. This hypothesis assumed transient impairment oflpP/ClpX-mediated proteolysis of the �-O initiator protein. How-ver, mutations in clpP and clpX genes had little influence on bothirectionality of lambda DNA replication and appearance of sigmaeplication intermediates (Narajczyk et al., 2007).

We hypothesize that �-O protein bound to the �-ori might disfa-or replication and create a physical barrier that leads to replicationork stalling, with a subsequent change to the � type replication.hen, the shift could occur in a similar way as the SPP1 proposed

of concatemeric DNA. Model II: in steps 5 and 6, a bubble migration model for thecked replication fork is drawn in blue, and newly synthesized DNA in yellow. Sites

model (Fig. 3). In fact, a strand invasion model was early postulatedfor � RDR (Bastia and Sueoka, 1975; Better and Freifelder, 1983;Dodson et al., 1986). Recently, a strand invasion activity has beendemonstrated in vitro for the � protein (Rybalchenko et al., 2004).

3.3.3. RDR in HSV-1HSV-1 linear duplex viral genome circularizes, by direct end-

to-end ligation, upon entering the host nucleus (Strang and Stow,2005). The genome of HSV-1 contains three origins of replication:one copy of oriL and two copies of oriS (Boehmer et al., 1993;Lockshon and Galloway, 1988). oriL is located in the center of theunique long sequences, whereas the two copies of oriS, which arestructurally different from oriL, are within the reiterated sequencesflanking the unique short sequences. Functional differences existbetween oriL and oriS and reveal a prominent role for oriL in HSV-1pathogenesis (Balliet and Schaffer, 2006). oriL consists of a perfect144-bp large palindrome that has in the center an AT-rich region of20-bp which is thought to be unwound during the initiation of DNAreplication (Hardwicke and Schaffer, 1995). On the other hand, theoriS core origin is a 90-bp sequence that includes a 45-bp imper-fect palindrome and an AT-rich region of 18-bp. The deletion ofeither oriL or one or even both copies of oriS has no effect on viralDNA replication, suggesting that one copy of an ori sequence may besufficient for the replication of HSV-1 (Igarashi et al., 1993; Polvino-Bodnar et al., 1987). Previously it was thought that upon infectionHSV-1 origin-mediated replication begins by the theta mode from
the circular genome, then a switch to RDR occurs and head-to-tailconcatemers are produced via replication, in a way similar to SPP1replication. Nevertheless new evidences show that another possi-bility is that a non-linear, perhaps branched structure, is produced

s Rese

tnotflig

3

a(dfiamfinhtptc

o1Dig1bpitp

3

raPwmuomsrrtibaePAoaabr


hat will be resolved by the viral and cellular proteins, by a mecha-ism similar T4 RDR (Wilkinson and Weller, 2003). Independentlyf the mechanism used, it is clear that recombination is essential inhe viral cycle, and consistent with this, ICP8 is an essential proteinor HSV DNA replication and possesses SSA activities, and the alka-ine nuclease, encoded by the UL12 gene, plays an important rolen HSV-1 replication, as a null mutant of UL12 displays a severerowth defect (Reuven et al., 2004).

.4. Formation of transducing plasmid molecules

Bacteriophages contribute to horizontal gene transfer by gener-lized transduction of non-viral DNA that can be from chromosomalchromosomal transduction) or plasmidic origin (plasmid trans-uction) (Canchaya et al., 2003; Viret et al., 1991). Bacterial DNAragment ends can be erroneously recognized by the viral packag-ng machinery, and thereby stretches of the long chromosomal DNAre encapsidated into empty proheads. The transduced chromoso-al DNA is then integrated into recipient host in a RecA-dependent

ashion. The transduction of plasmid DNA or of pathogenicityslands (see below) is a RecA-independent process that requires deovo synthesis and amplification of the transducing DNA to render aead-to-tail plasmid concatemer (Viret et al., 1991). Alternatively,he transducing particle is made by reversible integration of theathogenicity island or the plasmid DNA into the genome of theransducing virus, rendering an inactive virus due to the packagingonstrains (Orbach and Jackson, 1982).

In SPP1 virus, the packaging of amplified non-viral plasmid DNAccurs at a low frequency (in <0.001% of infected cells) (Viret et al.,991). However, if homology between the plasmid and the viralNA is present, a hybrid concatemeric molecule appears via RecA-

ndependent recombination, and the frequency of transduction isreatly increased, even by homology as short as 50-bp (Alonso et al.,986; Deichelbohrer et al., 1985). This increment was observed toe irrespective of whether the fragment contained or not the SPP1ac site (Bravo et al., 1990). The role of viral recombination proteinsn plasmid transduction has not been addressed, but it is temptingo speculate that they are involved in generalized transduction byromoting concatemeric plasmid RDR (Bravo and Alonso, 1990).

.5. Amplification of pathogenicity islands

The pathogenicity islands (PAIs) are genetic elements incorpo-ated into the genome of pathogenic bacteria, wich are usuallybsent in non-pathogenic organisms of the same species. TheAIs are characterized by: (i) disparate dG + dC content comparedith the rest of the genome, (ii) repeats flanking the genetic ele-ent, (iii) frequent association with tRNA-encoding genes (which

sually represent bacteriophage attachment sites), (iv) carriagef one or more virulence factors, (v) functional genes of otherobile elements such as phages and plasmids, and (vi) their large

ize (> 10-Kb) (Hacker and Kaper, 2000). The PAIs, which rep-esent unstable DNA regions, can be mobilized to other suitableecipients by horizontal transfer (transformation, conjugation andransduction). In Firmicutes the PAIs are frequently excised dur-ng induction of temperate phages, and transferred predominantlyy encapsidation of concatemeric DNA in the helper phage, inmechanism that is independent of the RecA protein (Lindsay

t al., 1998; O’Shea and Boyd, 2002). In Staphylococcus aureus,AIs (known as SaPIs) are very common (Novick et al., 2010).fter excision, SaPI replication may be initiated on either linearr circular DNA, and generates concatemeric DNA. SaPIs are pack-
ged in particles composed exclusively of viral proteins. In thebsence of a replicating phage, the incoming SaPI DNA circularizesefore integration. This circularization presumably occurs throughecombination between the redundant termini, and it must involve
arch 160 (2011) 1–14 11

an unknown recombinase, as SaPI transfer is RecA-independent(Novick et al., 2001).

Helper phages are specific to the different SaPIs. This suggeststhat homology might exist between the helper phage and the SaPI,and in fact, this has been shown to exist between SaPI1 and itsstaphylococcal temperate phage 80 (Ruzin et al., 2001). The effi-cient transfer of a SaPI by a specific helper phage reminds theefficient transduction by phages of plasmids sharing regions ofhomology described in the previous section. It is likely that PAIsmobilization, and the generation of phage concatemeric DNA byRDR, share many features in common.

4. Conclusions

The present review aims at summarizing our knowledge of theproteins and mechanisms underlying RDR and more specificallythe mechanisms that lead to the amplification and accumulationof viral linear concatemeric DNA. Many viral genomes code fora recombinase, and for other recombination proteins. How is theproduction of the viral concatemer depends mostly of the initialDNA molecule, whether it is a linear or a circular dsDNA molecule.Linear viral molecules may accomplish RDR by multiple replica-tion and recombination cycles as described for T4 and T7 phages,whereas circular molecules may switch from the initial � to the� replication mode as described for the SPP1 phage, where thegeneration of a DSB by fork reversal may trigger RDR. At present,many steps of the RDR reactions remain obscure, and there is aneed for simple models, as bacterial viruses, to extend our knowl-edge to more complex systems, such as the replication of eukaryoticviruses, mitochondrial or chloroplast DNA, and the amplification ofPAIs.

Acknowledgements

We thank Juan C. Alonso for reading the manuscript. This workwas partially supported by grant BFU2009-09520 from MICINN toS.A. A.L.P. received a contract from Community of Madrid. L.Z. wasa recipient of a EU Leonardo grant.

References

Adhya, S., Black, L., Friedman, D., Hatfull, G., Kreuzer, K., Merril, C., Oppenheim, A.,Rohwer, F., Young, R., 2005. 2004 ASM Conference on the New Phage Biology:The ‘Phage Summit’. Mol. Microbiol. 55, 1300–1314.

Alonso, J.C., Luder, G., Stiege, A.C., Chai, S., Weise, F., Trautner, T.A., 1997. Thecomplete nucleotide sequence and functional organization of Bacillus subtilisbacteriophage SPP1. Gene 204, 201–212.

Alonso, J.C., Luder, G., Trautner, T.A., 1986. Requirements for the formation ofplasmid-transducing particles of Bacillus subtilis bacteriophage SPP1. EMBO J.5, 3723–3728.

Alonso, J.C., Tavares, P., Lurz, R., Trautner, T.A., 2006. Bacteriophage SPP1, second ed.Oxford University Press, Oxford/New York.

Ando, R.A., Morrical, S.W., 1998. Single-stranded DNA binding properties of the UvsXrecombinase of bacteriophage T4: binding parameters and effects of nucleotides.J. Mol. Biol. 283, 785–796.

Arumugam, S.R., Lee, T.H., Benkovic, S.J., 2009. Investigation of stoichiometry of T4bacteriophage helicase loader protein (gp59). J. Biol. Chem. 284, 29283–29289.

Atkinson, J., McGlynn, P., 2009. Replication fork reversal and the maintenance ofgenome stability. Nucleic Acids Res. 37, 3475–3492.

Ayora, S., Carrasco, B., Cardenas, P., César, C.E., Canas, C., Yadav, T., Marchisone, C.,Alonso, J.C., 2011. DNA double strand break repair in bacteria: a view from Bacil-lus subtilis. FEMS Microbiol. Rev., doi: 10.1111/j.1574-6976.2011.00272.x. [Epubahead of print].

Ayora, S., Carrasco, B., Doncel, E., Lurz, R., Alonso, J.C., 2004. Bacillus subtilis RecUprotein cleaves Holliday junctions and anneals single-stranded DNA. Proc. Natl.Acad. Sci. U. S. A. 101, 452–457.

Ayora, S., Langer, U., Alonso, J.C., 1998. Bacillus subtilis DnaG primase stabilises thebacteriophage SPP1 G40P helicase–ssDNA complex. FEBS Lett. 439, 59–62.

Ayora, S., Missich, R., Mesa, P., Lurz, R., Yang, S., Egelman, E.H., Alonso, J.C., 2002.Homologous-pairing activity of the Bacillus subtilis bacteriophage SPP1 replica-tion protein G35P. J. Biol. Chem. 277, 35969–35979.

1 s Rese

A

B

B

B

B

B

B

B

B

B

B

B

B

B

B

B

B

B

C

C

C

C

C

C

C

C

C

C

C

D

D

2 A. Lo Piano et al. / Viru

yora, S., Stasiak, A., Alonso, J.C., 1999. The Bacillus subtilis bacteriophage SPP1 G39Pdelivers and activates the G40P DNA helicase upon interacting with the G38P-bound replication origin. J. Mol. Biol. 288, 71–85.

alasubramanian, N., Bai, P., Buchek, G., Korza, G., Weller, S.K., 2010. Physical inter-action between the herpes simplex virus type 1 exonuclease, UL12, and the DNAdouble strand break sensing MRN complex. J. Virol. 84, 10.

alliet, J.W., Schaffer, P.A., 2006. Point mutations in herpes simplex virus type 1 oriL,but not in oriS, reduce pathogenesis during acute infection of mice and impairreactivation from latency. J. Virol. 80, 440–450.

arcena, M., Martin, C.S., Weise, F., Ayora, S., Alonso, J.C., Carazo, J.M., 1998. Poly-morphic quaternary organization of the Bacillus subtilis bacteriophage SPP1replicative helicase (G40P). J. Mol. Biol. 283, 809–819.

astia, D., Sueoka, N., 1975. Studies on the late replication of phage lambda: rolling-circle replication of the wild type and a partially suppressed strain, Oam29Pam80. J. Mol. Biol. 98, 305–320.

etter, M., Freifelder, D., 1983. Studies on the replication of Escherichia coli phagelambda DNA. I. The kinetics of DNA replication and requirements for the gener-ation of rolling circles. Virology 126, 168–182.

irkenkamp-Demtroder, K., Golz, S., Kemper, B., 1997. Inhibition of Holliday struc-ture resolving endonuclease VII of bacteriophage T4 by recombination enzymesUvsX and UvsY. J. Mol. Biol. 267, 150–162.

leuit, J.S., Xu, H., Ma, Y., Wang, T., Liu, J., Morrical, S.W., 2001. Mediator proteinsorchestrate enzyme-ssDNA assembly during T4 recombination-dependent DNAreplication and repair. Proc. Natl. Acad. Sci. U. S. A. 98, 8298–8305.

oehmer, P.E., 2004. RNA binding and R-loop formation by the herpes simplexvirus type-1 single-stranded DNA-binding protein (ICP8). Nucleic Acids Res. 32,4576–4584.

oehmer, P.E., Craigie, M.C., Stow, N.D., Lehman, I.R., 1994. Association of origin bind-ing protein and single strand DNA-binding protein, ICP8, during herpes simplexvirus type 1 DNA replication in vivo. J. Biol. Chem. 269, 29329–29334.

oehmer, P.E., Dodson, M.S., Lehman, I.R., 1993. The herpes simplex virus type-1origin binding protein. DNA helicase activity. J. Biol. Chem. 268, 1220–1225.

oehmer, P.E., Lehman, I.R., 1993a. Herpes simplex virus type 1 ICP8: helix-destabilizing properties. J. Virol. 67, 711–715.

oehmer, P.E., Lehman, I.R., 1993b. Physical interaction between the herpes simplexvirus 1 origin-binding protein and single-stranded DNA-binding protein ICP8.Proc. Natl. Acad. Sci. U. S. A. 90, 8444–8448.

oehmer, P.E., Nimonkar, A.V., 2003. Herpes virus replication. IUBMB Life 55,13–22.

olt, E.L., Sharples, G.J., Lloyd, R.G., 2000. Analysis of conserved basic residues asso-ciated with DNA binding (Arg69) and catalysis (Lys76) by the RusA Hollidayjunction resolvase. J. Mol. Biol. 304, 165–176.

ravo, A., Alonso, J.C., 1990. The generation of concatemeric plasmid DNA in Bacillussubtilis as a consequence of bacteriophage SPP1 infection. Nucleic Acids Res. 18,4651–4657.

ravo, A., Alonso, J.C., Trautner, T.A., 1990. Functional analysis of the Bacillus subtilisbacteriophage SPP1 pac site. Nucleic Acids Res. 18, 2881–2886.

rister, J.R., Nossal, N.G., 2007. Multiple origins of replication contribute to a discon-tinuous pattern of DNA synthesis across the T4 genome during infection. J. Mol.Biol. 368, 336–348.

anchaya, C., Fournous, G., Chibani-Chennoufi, S., Dillmann, M.L., Brussow, H., 2003.Phage as agents of lateral gene transfer. Curr. Opin. Microbiol. 6, 417–424.

anas, C., Carrasco, B., Ayora, S., Alonso, J.C., 2008. The RecU Holliday junctionresolvase acts at early stages of homologous recombination. Nucleic Acids Res.36, 5242–5249.

arles-Kinch, K., George, J.W., Kreuzer, K.N., 1997. Bacteriophage T4 UvsW pro-tein is a helicase involved in recombination, repair and the regulation of DNAreplication origins. EMBO J. 16, 4142–4151.

arrasco, B., Manfredi, C., Ayora, S., Alonso, J.C., 2008. Bacillus subtilis SsbA anddATP regulate RecA nucleation onto single-stranded DNA. DNA Repair (Amst.)7, 990–996.

hai, S., Lurz, R., Alonso, J.C., 1995. The small subunit of the terminase enzyme ofBacillus subtilis bacteriophage SPP1 forms a specialized nucleoprotein complexwith the packaging initiation region. J. Mol. Biol. 252, 386–398.

hai, S., Szepan, U., Luder, G., Trautner, T.A., Alonso, J.C., 1993. Sequence analy-sis of the left end of the Bacillus subtilis bacteriophage SPP1 genome. Gene129, 41–49.

hung, Y.B., Nardone, C., Hinkle, D.C., 1990. Bacteriophage T7 DNA packaging. III. A“hairpin” end formed on T7 concatemers may be an intermediate in the pro-cessing reaction. J. Mol. Biol. 216, 939–948.

onnelly, J.C., Leach, D.R., 2002. Tethering on the brink: the evolutionarily conservedMre11–Rad50 complex. Trends Biochem. Sci. 27, 410–418.

ostes, A., Lecointe, F., McGovern, S., Quevillon-Cheruel, S., Polard, P., 2010. The C-terminal domain of the bacterial SSB protein acts as a DNA maintenance hub atactive chromosome replication forks. PLoS Genet. 6, e1001238.

ox, M.M., Goodman, M.F., Kreuzer, K.N., Sherratt, D.J., Sandler, S.J., Marians,K.J., 2000. The importance of repairing stalled replication forks. Nature 404,37–41.

urtis, F.A., Reed, P., Wilson, L.A., Bowers, L.Y., Yeo, R.P., Sanderson, J.M., Walmsley,A.R., Sharples, G.J., 2011. The C-terminus of the phage lambda Orf recombinaseis involved in DNA binding. J. Mol. Recognit. 24, 333–340.

atta, S., Costantino, N., Zhou, X., Court, D.L., 2008. Identification and analysis ofrecombineering functions from Gram-negative and Gram-positive bacteria andtheir phages. Proc. Natl. Acad. Sci. U. S. A. 105, 1626–1631.

eclais, A.C., Lilley, D.M., 2008. New insight into the recognition of branched DNAstructure by junction-resolving enzymes. Curr. Opin. Struct. Biol. 18, 86–95.

arch 160 (2011) 1–14

Deichelbohrer, I., Alonso, J.C., Luder, G., Trautner, T.A., 1985. Plasmid transduction byBacillus subtilis bacteriophage SPP1: effects of DNA homology between plasmidand bacteriophage. J. Bacteriol. 162, 1238–1243.

Derr, L.K., Kreuzer, K.N., 1990. Expression and function of the uvsW gene of bacte-riophage T4. J. Mol. Biol. 214, 643–656.

Dervyn, E., Suski, C., Daniel, R., Bruand, C., Chapuis, J., Errington, J., Janniere, L.,Ehrlich, S.D., 2001. Two essential DNA polymerases at the bacterial replicationfork. Science 294, 1716–1719.

Dixit, A., Ray, K., Lakowicz, J.R., Black, L.W., 2011. Dynamics of the T4 bacterio-phage DNAa packasome motor: endo VII resolvase release of arrested Y-DNAsubstrates. J. Biol. Chem..

Dodson, M., Echols, H., Wickner, S., Alfano, C., Mensa-Wilmot, K., Gomes, B.,LeBowitz, J., Roberts, J.D., McMacken, R., 1986. Specialized nucleoprotein struc-tures at the origin of replication of bacteriophage lambda: localized unwindingof duplex DNA by a six-protein reaction. Proc. Natl. Acad. Sci. U. S. A. 83,7638–7642.

Drahos, D.J., Hendrix, R.W., 1982. Effect of bacteriophage lambda infection on syn-thesis of groE protein and other Escherichia coli proteins. J. Bacteriol. 149,1050–1063.

Dreiseikelmann, B., Steger, U., Wackernagel, W., 1980. Length determination of theterminal redundant regions in the DNA of phage T7. Mol. Gen. Genet. 178,237–240.

Dudas, K.C., Kreuzer, K.N., 2001. UvsW protein regulates bacteriophage T4origin-dependent replication by unwinding R-loops. Mol. Cell. Biol. 21,2706–2715.

Dudas, K.C., Kreuzer, K.N., 2005. Bacteriophage T4 helicase loader protein gp59 func-tions as gatekeeper in origin-dependent replication in vivo. J. Biol. Chem. 280,21561–21569.

Enemark, E.J., Joshua-Tor, L., 2008. On helicases and other motor proteins. Curr. Opin.Struct. Biol. 18, 243–257.

Engler, M.J., Richardson, C.C., 1983. Bacteriophage T7 DNA replication. Synthesis oflagging strands in a reconstituted system using purified proteins. J. Biol. Chem.258, 11197–11205.

Erler, A., Wegmann, S., Elie-Caille, C., Bradshaw, C.R., Maresca, M., Seidel, R., Haber-mann, B., Muller, D.J., Stewart, A.F., 2009. Conformational adaptability of Redbetaduring DNA annealing and implications for its structural relationship withRad52. J. Mol. Biol. 391, 586–598.

Farb, J.N., Morrical, S.W., 2009. Functional complementation of UvsX and UvsY muta-tions in the mediation of T4 homologous recombination. Nucleic Acids Res. 37,2336–2345.

Formosa, T., Alberts, B.M., 1986. DNA synthesis dependent on genetic recombi-nation: characterization of a reaction catalyzed by purified bacteriophage T4proteins. Cell 47, 793–806.

Formosa, T., Burke, R.L., Alberts, B.M., 1983. Affinity purification of bacteriophageT4 proteins essential for DNA replication and genetic recombination. Proc. Natl.Acad. Sci. U. S. A. 80, 2442–2446.

Gabbai, C.B., Marians, K.J., 2010. Recruitment to stalled replication forks of the PriADNA helicase and replisome-loading activities is essential for survival. DNARepair (Amst.) 9, 202–209.

Gai, D., Chang, Y.P., Chen, X.S., 2010. Origin DNA melting and unwinding in DNAreplication. Curr. Opin. Struct. Biol. 20, 756–762.

Gajewski, S., Webb, M.R., Galkin, V., Egelman, E.H., Kreuzer, K.N., White, S.W., 2011.Crystal structure of the phage T4 recombinase UvsX and its functional interac-tion with the T4 SF2 helicase UvsW. J. Mol. Biol. 405, 65–76.

Gammon, D.B., Evans, D.H., 2009. The 3′-to-5′ exonuclease activity of vaccinia virusDNA polymerase is essential and plays a role in promoting virus genetic recom-bination. J. Virol. 83, 4236–4250.

Ganesan, A.T., Andersen, J.J., Luh, J., Effron, M., 1976. Microbiology 1976. AmericanSociety of Microbiology, Washinton, DC, pp. 319–325.

Garcia, A.D., Aravind, L., Koonin, E.V., Moss, B., 2000. Bacterial-type DNA Holli-day junction resolvases in eukaryotic viruses. Proc. Natl. Acad. Sci. U. S. A. 97,8926–8931.

Garcia, A.D., Moss, B., 2001. Repression of vaccinia virus Holliday junction resolvaseinhibits processing of viral DNA into unit-length genomes. J. Virol. 75,6460–6471.

Garcia, A.D., Otero, J., Lebowitz, J., Schuck, P., Moss, B., 2006. Quaternary structureand cleavage specificity of a poxvirus Holliday junction resolvase. J. Biol. Chem.281, 11618–11626.

George, J.W., Stohr, B.A., Tomso, D.J., Kreuzer, K.N., 2001. The tight linkage betweenDNA replication and double-strand break repair in bacteriophage T4. Proc. Natl.Acad. Sci. U. S. A. 98, 8290–8297.

Ghosh, S., Hamdan, S.M., Richardson, C.C., 2010. Two modes of interaction ofthe single-stranded DNA-binding protein of bacteriophage T7 with the DNApolymerase–thioredoxin complex. J. Biol. Chem. 285, 18103–18112.

Golz, S., Kemper, B., 1999. Association of Holliday-structure resolving endonucle-ase VII with gp20 from the packaging machine of phage T4. J. Mol. Biol. 285,1131–1144.

Hacker, J., Kaper, J.B., 2000. Pathogenicity islands and the evolution of microbes.Annu. Rev. Microbiol. 54, 641–679.

Hamdan, S.M., Richardson, C.C., 2009. Motors, switches, and contacts in the repli-some. Annu. Rev. Biochem 78, 205–243.

Hamilton, M.D., Nuara, A.A., Gammon, D.B., Buller, R.M., Evans, D.H., 2007. Duplexstrand joining reactions catalyzed by vaccinia virus DNA polymerase. NucleicAcids Res. 35, 143–151.

s Rese

H

H

H

H

H

H

H

I

I

J

J

K

K

K

K

K

K

K

K

K

K

K

K

L

L

L

L

L

L

L


ardwicke, M.A., Schaffer, P.A., 1995. Cloning and characterization of herpes simplexvirus type 1 oriL: comparison of replication and protein–DNA complex formationby oriL and oriS. J. Virol. 69, 1377–1388.

e, Z.G., Rezende, L.F., Willcox, S., Griffith, J.D., Richardson, C.C., 2003. The carboxyl-terminal domain of bacteriophage T7 single-stranded DNA-binding proteinmodulates DNA binding and interaction with T7 DNA polymerase. J. Biol. Chem.278, 29538–29545.

e, Z.G., Richardson, C.C., 2004. Effect of single-stranded DNA-binding proteins onthe helicase and primase activities of the bacteriophage T7 gene 4 protein. J. Biol.Chem. 279, 22190–22197.

eller, R.C., Marians, K.J., 2006. Replication fork reactivation downstream of ablocked nascent leading strand. Nature 439, 557–562.

erdendorf, T.J., Albrecht, D.W., Benkovic, S.J., Nelson, S.W., 2011. Biochemical char-acterization of bacteriophage T4 Mre11–Rad50 complex. J. Biol. Chem. 286,2382–2392.

ollifield, W.C., Kaplan, E.N., Huang, H.V., 1987. Efficient RecABC-dependent, homol-ogous recombination between coliphage lambda and plasmids requires a phageninR region gene. Mol. Gen. Genet. 210, 248–255.

ollis, T., Stattel, J.M., Walther, D.S., Richardson, C.C., Ellenberger, T., 2001. Struc-ture of the gene 2.5 protein, a single-stranded DNA binding protein encoded bybacteriophage T7. Proc. Natl. Acad. Sci. U. S. A. 98, 9557–9562.

garashi, K., Fawl, R., Roller, R.J., Roizman, B., 1993. Construction and properties of arecombinant herpes simplex virus 1 lacking both S-component origins of DNAsynthesis. J. Virol. 67, 2123–2132.

yer, L.M., Koonin, E.V., Aravind, L., 2002. Classification and evolutionary historyof the single-strand annealing proteins, RecT, Redbeta, ERF and RAD52. BMCGenomics 3, 8.

ones, C.E., Mueser, T.C., Dudas, K.C., Kreuzer, K.N., Nossal, N.G., 2001. BacteriophageT4 gene 41 helicase and gene 59 helicase-loading protein: a versatile couplewith roles in replication and recombination. Proc. Natl. Acad. Sci. U. S. A. 98,8312–8318.

ongeneel, C.V., Bedinger, P., Alberts, B.M., 1984. Effects of the bacteriophage T4 ddaprotein on DNA synthesis catalyzed by purified T4 replication proteins. J. Biol.Chem. 259, 12933–12938.

aguni, J.M., 2006. DnaA: controlling the initiation of bacterial DNA replication andmore. Annu. Rev. Microbiol. 60, 351–375.

err, I.D., Sivakolundu, S., Li, Z., Buchsbaum, J.C., Knox, L.A., Kriwacki, R., White, S.W.,2007. Crystallographic and NMR analyses of UvsW and UvsW.1 from bacterio-phage T4. J. Biol. Chem. 282, 34392–34400.

im, J., Wheeler, L.J., Shen, R., Mathews, C.K., 2005. Protein-DNA interactions inthe T4 dNTP synthetase complex dependent on gene 32 single-stranded DNA-binding protein. Mol. Microbiol. 55, 1502–1514.

odadek, T., Alberts, B.M., 1987. Stimulation of protein-directed strand exchange bya DNA helicase. Nature 326, 312–314.

odadek, T., Wong, M.L., Alberts, B.M., 1988. The mechanism of homologous DNAstrand exchange catalyzed by the bacteriophage T4 uvsX and gene 32 proteins.J. Biol. Chem. 263, 9427–9436.

ong, D., Griffith, J.D., Richardson, C.C., 1997. Gene 4 helicase of bacteriophage T7mediates strand transfer through pyrimidine dimers, mismatches, and nonho-mologous regions. Proc. Natl. Acad. Sci. U. S. A. 94, 2987–2992.

ong, D., Richardson, C.C., 1996. Single-stranded DNA binding protein and DNA heli-case of bacteriophage T7 mediate homologous DNA strand exchange. EMBO J.15, 2010–2019.

ong, D., Richardson, C.C., 1998. Role of the acidic carboxyl-terminal domainof the single-stranded DNA-binding protein of bacteriophage T7 in specificprotein–protein interactions. J. Biol. Chem. 273, 6556–6564.

ornberg, A., Baker, T., 1992. DNA Replication, second ed. W.H. Freeman and Co.,New York.

ovall, R., Matthews, B.W., 1997. Toroidal structure of lambda-exonuclease. Science277, 1824–1827.

reuzer, K.N., 2000. Recombination-dependent DNA replication in phage T4. TrendsBiochem. Sci. 25, 165–173.

reuzer, K.N., Brister, J.R., 2010. Initiation of bacteriophage T4 DNA replication andreplication fork dynamics: a review in the Virology Journal series on bacterio-phage T4 and its relatives. Virol. J. 7, 358.

ecointe, F., Serena, C., Velten, M., Costes, A., McGovern, S., Meile, J.C., Errington,J., Ehrlich, S.D., Noirot, P., Polard, P., 2007. Anticipating chromosomal replica-tion fork arrest: SSB targets repair DNA helicases to active forks. EMBO J. 26,4239–4251.

ee, S.J., Chowdhury, K., Tabor, S., Richardson, C.C., 2009. Rescue of bacteriophage T7DNA polymerase of low processivity by suppressor mutations affecting gene 3endonuclease. J. Virol. 83, 8418–8427.

iberek, K., Georgopoulos, C., Zylicz, M., 1988. Role of the Escherichia coli DnaKand DnaJ heat shock proteins in the initiation of bacteriophage lambda DNAreplication. Proc. Natl. Acad. Sci. U. S. A. 85, 6632–6636.

illey, D.M., White, M.F., 2001. The junction-resolving enzymes. Nat. Rev. Mol. CellBiol. 2, 433–443.

indsay, J.A., Ruzin, A., Ross, H.F., Kurepina, N., Novick, R.P., 1998. The gene for toxicshock toxin is carried by a family of mobile pathogenicity islands in Staphylo-coccus aureus. Mol. Microbiol. 29, 527–543.

ing, F., Shibata, T., 2004. Mhr1p-dependent concatemeric mitochondrial DNA for-
mation for generating yeast mitochondrial homoplasmic cells. Mol. Biol. Cell 15,310–322.
iu, J., Qian, N., Morrical, S.W., 2006. Dynamics of bacteriophage T4 presynapticfilament assembly from extrinsic fluorescence measurements of Gp32-single-stranded DNA interactions. J. Biol. Chem. 281, 26308–26319.

arch 160 (2011) 1–14 13

Lockshon, D., Galloway, D.A., 1988. Sequence and structural requirements of a herpessimplex viral DNA replication origin. Mol. Cell. Biol. 8, 4018–4027.

Long, D.T., Kreuzer, K.N., 2008. Regression supports two mechanisms of fork pro-cessing in phage T4. Proc. Natl. Acad. Sci. U. S. A. 105, 6852–6857.

Lopes, A., Amarir-Bouhram, J., Faure, G., Petit, M.A., Guerois, R., 2010. Detection ofnovel recombinases in bacteriophage genomes unveils Rad52, Rad51 and Gp2.5remote homologs. Nucleic Acids Res. 38, 3952–3962.

Ma, Y., Wang, T., Villemain, J.L., Giedroc, D.P., Morrical, S.W., 2004. Dual functions ofsingle-stranded DNA-binding protein in helicase loading at the bacteriophageT4 DNA replication fork. J. Biol. Chem. 279, 19035–19045.

Macmaster, R., Sedelnikova, S., Baker, P.J., Bolt, E.L., Lloyd, R.G., Rafferty, J.B., 2006.RusA Holliday junction resolvase: DNA complex structure – insights into selec-tivity and specificity. Nucleic Acids Res. 34, 5577–5584.

Mahdi, A.A., Sharples, G.J., Mandal, T.N., Lloyd, R.G., 1996. Holliday junctionresolvases encoded by homologous rusA genes in Escherichia coli K-12 and phage82. J. Mol. Biol. 257, 561–573.

Makhov, A.M., Griffith, J.D., 2006. Visualization of the annealing of complemen-tary single-stranded DNA catalyzed by the herpes simplex virus type 1 ICP8SSB/recombinase. J. Mol. Biol. 355, 911–922.

Makhov, A.M., Sen, A., Yu, X., Simon, M.N., Griffith, J.D., Egelman, E.H., 2009. Thebipolar filaments formed by herpes simplex virus type 1 SSB/recombinationprotein (ICP8) suggest a mechanism for DNA annealing. J. Mol. Biol.386, 273–279.

Manolaridis, I., Mumtsidu, E., Konarev, P., Makhov, A.M., Fullerton, S.W., Sinz, A.,Kalkhof, S., McGeehan, J.E., Cary, P.D., Griffith, J.D., Svergun, D., Kneale, G.G.,Tucker, P.A., 2009. Structural and biophysical characterization of the proteinsinteracting with the herpes simplex virus 1 origin of replication. J. Biol. Chem.284, 16343–16353.

Marintcheva, B., Hamdan, S.M., Lee, S.J., Richardson, C.C., 2006. Essential residues inthe C terminus of the bacteriophage T7 gene 2.5 single-stranded DNA-bindingprotein. J. Biol. Chem. 281, 25831–25840.

Martinez-Jimenez, M.I., Alonso, J.C., Ayora, S., 2005. Bacillus subtilis bacteriophageSPP1-encoded gene 34.1 product is a recombination-dependent DNA replicationprotein. J. Mol. Biol. 351, 1007–1019.

Martínez-Jiménez, M.I., Mesa, P., Alonso, J.C., 2002. Bacillus subtilis tau subunit ofDNA polymerase III interacts with bacteriophage SPP1 replicative DNA helicaseG40P. Nucleic Acids Res. 30, 5056–5064.

Masai, H., Tanaka, T., Kohda, D., 2010. Stalled replication forks: making ends meetfor recognition and stabilization. Bioessays 32, 687–697.

McIntosh, P.K., Dunker, R., Mulder, C., Brown, N.C., 1978. DNA of Bacillus subtilis bac-teriophage SPP1: physical mapping and localization of the origin of replication.J. Virol. 28, 865–876.

Michel, B., Grompone, G., Flores, M.J., Bidnenko, V., 2004. Multiple pathways processstalled replication forks. Proc. Natl. Acad. Sci. U. S. A. 101, 12783–12788.

Mickelson, C., Wiberg, J.S., 1981. Membrane-associated DNase activity controlled bygenes 46 and 47 of bacteriophage T4D and elevated DNase activity associatedwith the T4 das mutation. J. Virol. 40, 65–77.

Mikhailov, V.S., Okano, K., Rohrmann, G.F., 2006. Structural and functional analysisof the baculovirus single-stranded DNA-binding protein LEF-3. Virology 346,469–478.

Miller, E.S., Kutter, E., Mosig, G., Arisaka, F., Kunisawa, T., Ruger, W., 2003. Bacterio-phage T4 genome. Microbiol. Mol. Biol. Rev. 67, 86–156 (table of contents).

Missich, R., Weise, F., Chai, S., Lurz, R., Pedre, X., Alonso, J.C., 1997. The replisomeorganizer (G38P) of Bacillus subtilis bacteriophage SPP1 forms specialized nucle-oprotein complexes with two discrete distant regions of the SPP1 genome. J. Mol.Biol. 270, 50–64.

Molineux, I.J., 2006. The T7 Group, second ed. Oxford University Press, Oxford/NewYork.

Mosig, G., 1998. Recombination and recombination-dependent DNA replication inbacteriophage T4. Annu. Rev. Genet. 32, 379–413.

Mosig, G., Colowick, N., Gruidl, M.E., Chang, A., Harvey, A.J., 1995. Multiple initiationmechanisms adapt phage T4 DNA replication to physiological changes duringT4’s development. FEMS Microbiol. Rev. 17, 83–98.

Mosig, G., Gewin, J., Luder, A., Colowick, N., Vo, D., 2001. Two recombination-dependent DNA replication pathways of bacteriophage T4, and their roles inmutagenesis and horizontal gene transfer. Proc. Natl. Acad. Sci. U. S. A. 98,8306–8311.

Muylaert, I., Tang, K.W., Elias, P., 2011. Replication and recombination of Herpessimplex virus DNA. J. Biol. Chem..

Muyrers, J.P., Zhang, Y., Buchholz, F., Stewart, A.F., 2000. RecE/RecT and Redal-pha/Redbeta initiate double-stranded break repair by specifically interactingwith their respective partners. Genes Dev. 14, 1971–1982.

Narajczyk, M., Baranska, S., Wegrzyn, A., Wegrzyn, G., 2007. Switch from theta tosigma replication of bacteriophage lambda DNA: factors involved in the processand a model for its regulation. Mol. Genet. Genomics 278, 65–74.

Nimonkar, A.V., Boehmer, P.E., 2002. In vitro strand exchange promoted by the her-pes simplex virus type-1 single strand DNA-binding protein (ICP8) and DNAhelicase–primase. J. Biol. Chem. 277, 15182–15189.

Nimonkar, A.V., Boehmer, P.E., 2003. The herpes simplex virus type-1 single-strand DNA-binding protein (ICP8) promotes strand invasion. J. Biol. Chem. 278,9678–9682.

Nimonkar, A.V., Boehmer, P.E., 2004. Role of protein–protein interactions duringherpes simplex virus type 1 recombination-dependent replication. J. Biol. Chem.279, 21957–21965.

1 s Rese

N

N

N

O

O

O

O

P

P

P

P

P

P

P

R

R

R

R

R

R

R

S

S

S

S

S

S

S

S

S

S

S

Regulation of the initiation of chromosomal replication in bacteria. FEMS Micro-biol. Rev. 31, 378–387.

Zhang, J., Xing, X., Herr, A.B., Bell, C.E., 2009. Crystal structure of E. coli RecE pro-tein reveals a toroidal tetramer for processing double-stranded DNA breaks.Structure 17, 690–702.

4 A. Lo Piano et al. / Viru

oirot, P., Gupta, R.C., Radding, C.M., Kolodner, R.D., 2003. Hallmarks of homol-ogy recognition by RecA-like recombinases are exhibited by the unrelatedEscherichia coli RecT protein. EMBO J. 22, 324–334.

ovick, R.P., Christie, G.E., Penades, J.R., 2010. The phage-related chromosomalislands of Gram-positive bacteria. Nat. Rev. Microbiol. 8, 541–551.

ovick, R.P., Schlievert, P., Ruzin, A., 2001. Pathogenicity and resistance islands ofstaphylococci. Microbes Infect. 3, 585–594.

’Shea, Y.A., Boyd, E.F., 2002. Mobilization of the Vibrio pathogenicity island betweenVibrio cholerae isolates mediated by CP-T1 generalized transduction. FEMSMicrobiol. Lett. 214, 153–157.

kano, K., Vanarsdall, A.L., Rohrmann, G.F., 2007. A baculovirus alkaline nucleaseknockout construct produces fragmented DNA and aberrant capsids. Virology359, 46–54.

ldenburg, D.J., Bendich, A.J., 2004. Most chloroplast DNA of maize seedlings in linearmolecules with defined ends and branched forms. J. Mol. Biol. 335, 953–970.

rbach, M.J., Jackson, E.N., 1982. Transfer of chimeric plasmids among Salmonellatyphimurium strains by P22 transduction. J. Bacteriol. 149, 985–994.

ant, K., Shokri, L., Karpel, R.L., Morrical, S.W., Williams, M.C., 2008. Modulation ofT4 gene 32 protein DNA binding activity by the recombination mediator proteinUvsY. J. Mol. Biol. 380, 799–811.

assy, S.I., Yu, X., Li, Z., Radding, C.M., Egelman, E.H., 1999. Rings and filaments ofbeta protein from bacteriophage lambda suggest a superfamily of recombinationproteins. Proc. Natl. Acad. Sci. U. S. A. 96, 4279–4284.

atel, S.S., Picha, K.M., 2000. Structure and function of hexameric helicases. Annu.Rev. Biochem. 69, 651–697.

edre, X., Weise, F., Chai, S., Luder, G., Alonso, J.C., 1994. Analysis of cis and transacting elements required for the initiation of DNA replication in the Bacillussubtilis bacteriophage SPP1. J. Mol. Biol. 236, 1324–1340.

erumal, S.K., Raney, K.D., Benkovic, S.J., 2010. Analysis of the DNA translocation andunwinding activities of T4 phage helicases. Methods 51, 277–288.

ohjoismaki, J.L., Goffart, S., Tyynismaa, H., Willcox, S., Ide, T., Kang, D., Suomalainen,A., Karhunen, P.J., Griffith, J.D., Holt, I.J., Jacobs, H.T., 2009. Human heart mito-chondrial DNA is organized in complex catenated networks containing abundantfour-way junctions and replication forks. J. Biol. Chem. 284, 21446–21457.

olvino-Bodnar, M., Orberg, P.K., Schaffer, P.A., 1987. Herpes simplex virus type 1oriL is not required for virus replication or for the establishment and reactivationof latent infection in mice. J. Virol. 61, 3528–3535.

euven, N.B., Staire, A.E., Myers, R.S., Weller, S.K., 2003. The herpes simplex virus type1 alkaline nuclease and single-stranded DNA binding protein mediate strandexchange in vitro. J. Virol. 77, 7425–7433.

euven, N.B., Weller, S.K., 2005. Herpes simplex virus type 1 single-strand DNA bind-ing protein ICP8 enhances the nuclease activity of the UL12 alkaline nuclease byincreasing its processivity. J. Virol. 79, 9356–9358.

euven, N.B., Willcox, S., Griffith, J.D., Weller, S.K., 2004. Catalysis of strand exchangeby the HSV-1 UL12 and ICP8 proteins: potent ICP8 recombinase activity isrevealed upon resection of dsDNA substrate by nuclease. J. Mol. Biol. 342, 57–71.

ezende, L.F., Willcox, S., Griffith, J.D., Richardson, C.C., 2003. A single-stranded DNA-binding protein of bacteriophage T7 defective in DNA annealing. J. Biol. Chem.278, 29098–29105.

ochester, S.C., Traktman, P., 1998. Characterization of the single-stranded DNAbinding protein encoded by the vaccinia virus I3 gene. J. Virol. 72, 2917–2926.

uzin, A., Lindsay, J., Novick, R.P., 2001. Molecular genetics of SaPI1 – a mobilepathogenicity island in Staphylococcus aureus. Mol. Microbiol. 41, 365–377.

ybalchenko, N., Golub, E.I., Bi, B., Radding, C.M., 2004. Strand invasion promotedby recombination protein � of coliphage �. Proc. Natl. Acad. Sci. U. S. A. 101,17056–17060.

adowski, P.D., 1974. Suppression of a mutation in gene 3 of bacteriophage T7 (T7endonuclease I) by mutations in phage and host polynucleotide ligase. J. Virol.13, 226–229.

adowski, P.D., Vetter, D., 1976. Genetic recombination of bacteriophage T7 DNA invitro. Proc. Natl. Acad. Sci. U. S. A. 73, 692–696.

alinas, F., Kodadek, T., 1995. Phage T4 homologous strand exchange: a DNA helicase,not the strand transferase, drives polar branch migration. Cell 82, 111–119.

anders, G.M., Dallmann, H.G., McHenry, C.S., 2010. Reconstitution of the B. sub-tilis replisome with 13 proteins including two distinct replicases. Mol. Cell 37,273–281.

andler, S.J., Marians, K.J., Zavitz, K.H., Coutu, J., Parent, M.A., Clark, A.J., 1999. dnaCmutations suppress defects in DNA replication- and recombination-associatedfunctions in priB and priC double mutants in Escherichia coli K-12. Mol. Microbiol.34, 91–101.

awitzke, J.A., Stahl, F.W., 1992. Phage lambda has an analog of Escherichia coli recO,recR and recF genes. Genetics 130, 7–16.

erwer, P., Watson, R.H., Son, M., 1990. Role of gene 6 exonuclease in the replicationand packaging of bacteriophage T7 DNA. J. Mol. Biol. 215, 287–299.

hamoo, Y., Friedman, A.M., Parsons, M.R., Konigsberg, W.H., Steitz, T.A., 1995. Crys-tal structure of a replication fork single-stranded DNA binding protein (T4 gp32)complexed to DNA. Nature 376, 362–366.

harples, G.J., 2001. The X philes: structure-specific endonucleases that resolve Hol-liday junctions. Mol. Microbiol. 39, 823–834.

harples, G.J., Bolt, E.L., Lloyd, R.G., 2002. RusA proteins from the extreme ther-
mophile Aquifex aeolicus and lactococcal phage r1t resolve Holliday junctions.Mol. Microbiol. 44, 549–559.
harples, G.J., Corbett, L.M., Graham, I.R., 1998. Lambda Rap protein is a structure-specific endonuclease involved in phage recombination. Proc. Natl. Acad. Sci. U.S. A. 95, 13507–13512.

arch 160 (2011) 1–14

Sharples, G.J., Corbett, L.M., McGlynn, P., 1999. DNA structure specificity of Rapendonuclease. Nucleic Acids Res. 27, 4121–4127.

Sharples, G.J., Curtis, F.A., McGlynn, P., Bolt, E.L., 2004. Holliday junction binding andresolution by the Rap structure-specific endonuclease of phage lambda. J. Mol.Biol. 340, 739–751.

Shereda, R.D., Kozlov, A.G., Lohman, T.M., Cox, M.M., Keck, J.L., 2008. SSB as an orga-nizer/mobilizer of genome maintenance complexes. Crit. Rev. Biochem. Mol.Biol. 43, 289–318.

Shutt, T.E., Gray, M.W., 2006. Bacteriophage origins of mitochondrial replication andtranscription proteins. Trends Genet. 22, 90–95.

Skalka, A.M., 1977. DNA replication-bacteriophage lambda. Curr. Top. Microbiol.Immunol. 78, 201–237.

Son, M., Serwer, P., 1992. Role of exonuclease in the specificity of bacteriophage T7DNA packaging. Virology 190, 824–833.

Strang, B.L., Stow, N.D., 2005. Circularization of the herpes simplex virus type 1genome upon lytic infection. J. Virol. 79, 12487–12494.

Subramanian, K., Rutvisuttinunt, W., Scott, W., Myers, R.S., 2003. The enzy-matic basis of processivity in lambda exonuclease. Nucleic Acids Res. 31,1585–1596.

Sugimoto, K., Kohara, Y., Okazaki, T., 1987. Relative roles of T7 RNA polymerase andgene 4 primase for the initiation of T7 phage DNA replication in vivo. Proc. Natl.Acad. Sci. U. S. A. 84, 3977–3981.

Szczepanska, A.K., 2009. Bacteriophage-encoded functions engaged in initiation ofhomologous recombination events. Crit. Rev. Microbiol. 35, 197–220.

Tarkowski, T.A., Mooney, D., Thomason, L.C., Stahl, F.W., 2002. Gene productsencoded in the ninR region of phage lambda participate in Red-mediated recom-bination. Genes Cells 7, 351–363.

Taylor, K., Wegrzyn, G., 1995. Replication of coliphage lambda DNA. FEMS Microbiol.Rev. 17, 109–119.

Thomas, M.S., Gao, M., Knipe, D.M., Powell, K.L., 1992. Association between the her-pes simplex virus major DNA-binding protein and alkaline nuclease. J. Virol. 66,1152–1161.

Thresher, R.J., Makhov, A.M., Hall, S.D., Kolodner, R., Griffith, J.D., 1995. Electronmicroscopic visualization of RecT protein and its complexes with DNA. J. Mol.Biol. 254, 364–371.

Trakselis, M.A., Mayer, M.U., Ishmael, F.T., Roccasecca, R.M., Benkovic, S.J., 2001.Dynamic protein interactions in the bacteriophage T4 replisome. TrendsBiochem. Sci. 26, 566–572.

Tseng, M., Palaniyar, N., Zhang, W., Evans, D.H., 1999. DNA binding and aggre-gation properties of the vaccinia virus I3L gene product. J. Biol. Chem. 274,21637–21644.

Vanarsdall, A.L., Okano, K., Rohrmann, G.F., 2006. Characterization of the role of verylate expression factor 1 in baculovirus capsid structure and DNA processing. J.Virol. 80, 1724–1733.

Vellani, T.S., Myers, R.S., 2003. Bacteriophage SPP1 Chu is an alkaline exonucleasein the SynExo family of viral two-component recombinases. J. Bacteriol. 185,2465–2474.

Viret, J.F., Bravo, A., Alonso, J.C., 1991. Recombination-dependent concatemeric plas-mid replication. Microbiol. Rev. 55, 675–683.

Webb, M.R., Plank, J.L., Long, D.T., Hsieh, T.S., Kreuzer, K.N., 2007. The phage T4protein UvsW drives Holliday junction branch migration. J. Biol. Chem. 282,34401–34411.

Wegrzyn, G., Wegrzyn, A., 2005. Genetic switches during bacteriophage lambdadevelopment. Prog. Nucleic Acid Res. Mol. Biol. 79, 1–48.

Weigel, C., Seitz, H., 2006. Bacteriophage replication modules. FEMS Microbiol. Rev.30, 321–381.

Weise, F., Chai, S., Luder, G., Alonso, J.C., 1994. Nucleotide sequence and complemen-tation studies of the gene 35 region of the Bacillus subtilis bacteriophage SPP1.Virology 202, 1046–1049.

White, J.H., Richardson, C.C., 1987. Processing of concatemers of bacteriophage T7DNA in vitro. J. Biol. Chem. 262, 8851–8860.

White, M.F., 2011. Homologous recombination in the archaea: the means justify theends. Biochem. Soc. Trans. 39, 15–19.

Wilkinson, D.E., Weller, S.K., 2003. The role of DNA recombination in herpes simplexvirus DNA replication. IUBMB Life 55, 451–458.

Willer, D.O., Yao, X.D., Mann, M.J., Evans, D.H., 2000. In vitro concatemerformation catalyzed by vaccinia virus DNA polymerase. Virology 278,562–569.

Xu, H., Beernink, H.T., Morrical, S.W., 2010. DNA-binding properties ofT4 UvsY recombination mediator protein: polynucleotide wrapping pro-motes high-affinity binding to single-stranded DNA. Nucleic Acids Res.38, 4821–4833.

Zakrzewska-Czerwinska, J., Jakimowicz, D., Zawilak-Pawlik, A., Messer, W., 2007.

recombination-dependent concatemeric viral dna replication

Documents