JOURNAL OF BACTERIOLOGY,0021-9193/01/$04.0010 DOI: 10.1128/JB.183.17.5206–5208.2001

Sept. 2001, p. 5206–5208 Vol. 183, No. 17

Copyright © 2001, American Society for Microbiology. All Rights Reserved.

Lambda Excision Revisited: Testing a Model forSynapsis of Prophage Ends

MARTIN L. PATO*

Department of Microbiology, University of Colorado Health Sciences Center,Denver, Colorado 80262

Received 29 March 2001/Accepted 6 June 2001

Excision of lambda prophage was reexamined to test a model for prophage end synapsis. The model proposesthat, during in situ prophage replication, following induction, the diverging replication forks are held together.Consequently, prophage DNA is spooled through the replication machinery, drawing the prophage endstogether and facilitating synapsis. The model predicts that excision will be slowed if in situ lambda replicationis inhibited, and the predicted low rate of excision of a nonreplicating prophage was observed after thermoin-duction. However, excision was rapid if additional Int protein was supplied or if the temperature was reducedafter induction, showing that (i) Int is partially thermosensitive for excision at 42°C and (ii) in situ replicationis not required for rapid excision, a finding that is inconsistent with the model.

Integration of the genome of bacteriophage lambda into thechromosome of its host and excision of the prophage form oflambda have long been studied as the paradigm of site-specificrecombination. Recombination between the attP site on thecircularized, 48-kDa viral genome and the attB site on thechromosome results in integration, whereas recombination be-tween the attL and attR sites at the prophage junctions resultsin excision. While the earliest studies focused on the intactviral genome, by the 1970s systems were developed for study-ing integration and excision both in vivo and in vitro usingartificial substrates such as the latt2 phage which carry tworecombining sites on a single virus genome and plasmids car-rying cloned att sites. Integration was found to require thephage-encoded Int and the host-encoded integration host fac-tor proteins, whereas excision required both these proteins andthe phage-encoded Xis and host-encoded factor for inversionstimulation (FIS) proteins (Fig. 1) (for a review, see reference7).

My interest in lambda excision was piqued by an earlier workon replicative transposition of bacteriophage Mu, in particular,the requirement for a special mechanism for promoting therapid synapsis of Mu prophage ends (6). The question arose asto whether a mechanism exists to promote the long-rangeDNA interactions required for synapsis of attL and attR duringexcision. The lambda prophage, like the Mu prophage, is large(about 10 kb larger than Mu) and resides within the complexstructure of the host nucleoid. However, synapsis of thelambda prophage ends is not subject to the topological re-straints imposed on synapsis of Mu ends. Lambda recombina-tion can occur between att sites in direct or indirect orientationand as an intramolecular or intermolecular reaction. Synapsisof Mu ends requires that they be plectonemically interwoundin only one orientation.

If a special mechanism is required for rapid synapsis of

lambda prophage termini, an appealing model is the following.Upon induction, in situ, bidirectional replication of the pro-phage from an internal origin is initiated, with replicationproceeding beyond the prophage ends into adjacent bacterialDNA, resulting in an “onionskin” of multiple prophage copies(3). If, rather than moving apart on the DNA, the replicationforks are held together, perhaps at a membrane site, then thereplicating DNA will be spooled through the replication ma-chinery and the termini will be drawn together, perhaps assist-ing in synapsis. The model predicts that excision should beslowed in the absence of lambda-specific replication. Whileevidence exists which shows that excised lambda can be ob-served as early as 15 min after thermoinduction of a cI857lysogen (5), a careful comparison with the kinetics of excisionof nonreplicating prophage is lacking. In this study I reexam-ined excision of lambda prophages using a sensitive, new assayto address the predictions of this model.

Recombination between attL and attR during excision re-stores attB and attP (Fig. 1). A primer derived from a sequencein the bacterial DNA adjacent to attB can be used to assess therelative amounts of attL and attB and the percent excision.DNA is isolated at intervals after induction, purified by phenolextraction, and cleaved with selected restriction enzymes (AvaIand HaeII in these experiments). The cleaved DNA is usedtogether with a 32P-end-labeled primer in 30 cycles of primerextension in a PCR apparatus. Two fragments can be gener-ated: a 219-bp attL-associated fragment from the primer site tothe AvaI site and a 190-bp attB-associated fragment from theprimer site to the HaeII site. The former is present beforeexcision and the latter is present after excision. The fragmentsare separated on a sequencing gel and quantified with a phos-phorimager.

For the initial experiment, cultures of Escherichia coli N99sup0 lcI857P1 and lcI857Psus80 lysogens were grown at 30°Cand induced by shifting to 42°C. Samples were removed atintervals into a sodium dodecyl sulfate lysis mixture at 80°Cand incubated for 10 min, and DNA was isolated and pro-cessed as described above. The resulting data are shown in Fig.2, and the calculated percent excision is plotted against time

* Mailing address. Department of Microbiology, University of Col-orado Health Sciences Center, 4200 E. 9th Ave., Denver, CO 80262.Phone: (303) 315-7213. Fax: (303) 315-6785. E-mail: martin.pato@uchsc.edu.

5206

on October 20, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

after induction in Fig. 3A. More-rapid excision of the wild-typeprophage (about 50% by 20 min) than of the nonreplicating P2

prophage (about 10% by 20 min) was observed. The differencein the actual number of excision events was even greater, sincethe wild-type prophage copy number increases after induction(3).

While the low rate of excision of the P2 prophage is consis-tent with the model, it could have resulted from a limitation ofone of the proteins required for excision due to (i) synthesis ofInt and Xis from a single copy of the prophage rather thanfrom an amplified number of copies of the replicating pro-phage or (ii) thermolability of a protein at the inducing tem-

FIG. 1. Diagram of lambda prophage excision. Heavy dashed lines,chromosomal DNA; light lines, lambda DNA; horizontal arrow, prim-er (59 ATGTGTTCACAGGTTGCTCCG); short vertical arrows, re-striction enzyme sites.

FIG. 2. Excision of lambda prophage. Iterative primer extensionwas performed as described in the text at intervals after induction ofP1 and P2 lysogens at 42°C. Fragments associated with attL (219 bp)and attB (190 bp) are indicated. A sequencing lane is included for sizemarkers.

FIG. 3. Kinetics of prophage excision after induction. The percentexcision is calculated as the ratio of the attB-associated fragment to thetotal of the attB- and attL-associated fragments. (A) Excision of P2 (F)and P1 (E) prophage at 42°C. (B) Excision of P2 prophage at 42°C (F)and after shifting to 36°C (E) after 8 min at 42°C. (C) Excision of P2

prophage in the presence of plasmids expressing Int or Xis, designatedpInt or pXis, respectively, when IPTG is present. F, both plasmidswithout IPTG; E, both plasmids with IPTG; ■, pInt without IPTG; h,pInt with IPTG; Œ, pXis without IPTG; ‚, pXis with IPTG.

VOL. 183, 2001 NOTES 5207

on October 20, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

perature of 42°C. Integration, but not excision, has been re-ported to be temperature sensitive (2).

Thermoinduction of the P2 lysogen was repeated, and partof the culture was transferred to 36°C after 8 min at 42°C. Asshown in Fig. 3B, excision at 36°C ensued rapidly and pro-ceeded essentially to completion, indicating that some compo-nent of the excision machinery is thermolabile. To determine ifthermolability of Int or Xis is responsible for the observedinhibition of excision, compatible plasmids carrying the intgene or the xis gene under IPTG (isopropyl-b-D-thiogalacto-pyranoside)-inducible promoters were introduced, separatelyand in combination, into the P2 lysogen. Cultures of the plas-mid-containing strains were shifted to 42°C to express Int andXis from the prophage, and IPTG was added at the time of theshift to half of each culture. The results reported in Fig. 3Cshow that expression of Int from the plasmid allowed rapid andcomplete excision of the prophage at 42°C; i.e., elevated levelsof Int, along with Xis supplied either only from the prophageor also from a plasmid, are sufficient for efficient excision at42°C.

While I have not measured the actual amounts or the activityof the Int protein under the different conditions describedabove, I infer from the excision data that the Int protein showsreduced activity at the elevated temperature of 42°C. If the Intprotein displays reduced activity and is limiting in the excisionreaction at 42°C, then increasing the amount of the proteinshould increase the amount of excision observed. Therefore, areasonable interpretation of the above data is as follows. Lowlevels of excision are observed at 42°C when low levels of Intare synthesized from a single copy of a prophage such as thenonreplicating P2 prophage. Intermediate levels of excisionare observed at 42°C when intermediate levels of Int are sup-plied from the amplified copies of a replicating prophage. Highlevels of excision are observed at 42°C when high levels of Intare synthesized from a high-copy-number plasmid.

The results presented here, while not supporting the pro-posed model, do not exclude the possibility that the replicationforks are held together as proposed, as has been suggested forchromosomal forks (1, 4). In this context, the amplification ofprophage copies raises a very interesting question. Excisionfrom the onionskin of amplified copies requires that a partic-ular attL recombines with its partner attR and not with an attRfrom a different copy of the prophage. Such a transrecombi-nation would result in a tandem prophage dimer and a dele-tion, not in an excision. How the appropriate synapse is madeand whether or not inappropriate synapses are made remain tobe determined. Even though these experiments failed to ad-duce support for the synapsis model, the model nicely suggestshow appropriate att sites could be synapsed by the spooling ofthe distal att sites on the same prophage DNA through thefixed replication machinery.

This work was supported by a grant from the National ScienceFoundation.

I thank Frank Stahl for supplying the lambda phage and Anca Segallfor supplying the plasmids expressing Int and Xis.

REFERENCES

1. Dingman, C. W. 1974. Bidirectional chromosome replication: some topolog-ical considerations. J. Theor. Biol. 43:187–195.

2. Guarneros, G., and H. Echols. 1973. Thermal asymmetry of site-specificrecombination by bacteriophage lambda. Virology 52:30–38.

3. Imae, Y., and T. Fukasawa. 1970. Regional replication of the bacterial chro-mosome induced by derepression of prophage lambda. J. Mol. Biol. 54:585–597.

4. Lemon, K. P., and A. D. Grossman. 1998. Localization of bacterial DNApolymerase: evidence for a factory model of replication. Science 282:1516–1519.

5. Ljundquist, E., and A. Bukhari. 1977. State of prophage Mu DNA uponinduction. Proc. Natl. Acad. Sci. USA 74:3143–3147.

6. Pato, M. L., and M. Banerjee. 1996. The Mu strong gyrase-binding sitepromotes efficient synapsis of the prophage termini. Mol. Microbiol. 22:283–292.

7. Thompson, J. F., and A. Landy. 1989. Regulation of bacteriophage lambdasite-specific recombination, p. 1–22. In D. E. Berg and M. M. Howe (ed.),Mobile DNA. American Society for Microbiology, Washington, D.C.

5208 NOTES J. BACTERIOL.

on October 20, 2020 by guest

http://jb.asm.org/

Dow

nloaded from