Vol. 171, No. 10JOURNAL OF BACTERIOLOGY, OCt. 1989, p. 5759-57610021-9193/89/105759-03$02.00/0Copyright © 1989, American Society for Microbiology

Tn5 Mutagenesis of Anabaena sp. Strain PCC 7120: Isolation of aNew Mutant Unable To Grow without Combined Nitrogen

DULAL BORTHAKURt AND ROBERT HASELKORN*Department of Molecular Genetics and Cell Biology, University of Chicago, 920 East 58th Street, Chicago, Illinois 60637

Received 19 May 1989/Accepted 29 June 1989

Methylation by Ava methylases in Escherichia coli increases the efficiency to transfer of TnS in pBR322bla::TnS from E. coli to Anabaena sp. strain PCC 7120 by conjugation. Following conjugation, TnS but not pBR322sequences were found at many different positions in the Anabaena chromosome. This procedure was used tomutagenize, tag, and clone a previously unrecognized gene required for nitrogen fixation in this Anabaena sp.

Anabaena sp. strain PCC 7120 is a filamentous cyanobac-terium capable of fixing atmospheric nitrogen in specializedcells called heterocysts. The molecular genetic techniquesfor Anabaena are not yet as well developed as for othernitrogen-fixing bacteria such as Klebsiella, Azotobacter,Rhodobacter, and Rhizobium spp. Recently, techniqueshave been developed for transferring cloned genes intoAnabaena spp. (14). However, the restriction enzymes AvaIand AvaIl present in Anabaena cells severely reduce thefrequency of stable transfer of cloned DNA containing therecognition sites for these enzymes when that DNA istransferred by conjugation. More recently, Elhai and Wolkdeveloped a technique for methylating cloned DNA withAval and AvaII methylases in Escherichia coli before trans-fer into Anabaena spp. (3). Although mutants have beenisolated in Anabaena spp. (2, 6, 7, 11, 12) and drug resis-tance cassettes have been used to generate mutants (5; D.Borthakur, M. Basche, W. J. Buikema, P. B. Borthakur, andR. Haselkorn, submitted for publication), transposition oftransposons has not yet been used successfully to isolatemutants. We describe here random TnS mutagenesis ofAnabaena sp. strain PCC 7120 and one example of cloning ofthe interrupted DNA by TnS tagging, using the methylationsystem to increase the efficiency of TnS transfer.

Construction of pBR322::TnS. For random TnS mutagen-esis of our Anabaena strain, a suicide plasmid, pBR322::TnS, was constructed by isolating a TnS insertion in thef-lactamase (bla) gene in pBR322. Although pBR322 can bemobilized into Anabaena sp. strain PCC 7120, it cannotreplicate in Anabaena spp. (14). E. coli MC1061 containingpBR322 was infected with X::TnS, selecting for kanamycinresistance. Plasmid DNA isolated from the kanamycin-resistant colonies were used to transform E. coli, selectingfor kanamycin resistance. From the transformants, oneampicillin-sensitive transformant in which TnS was insertedinto the bla gene in pBR322 was isolated.TnS mutagenesis ofAnabaena sp. TnS has four Aval sites at

which the plasmid pBR322::TnS might be cleaved by thenative AvaI restriction enzyme upon being transferred intoour Anabaena sp. Therefore, pBR322: :TnS was first used totransform E. coli HB101(pRL528) containing AvaI and AvaIImethylase activity (3). The plasmid pBR322::TnS was thentransferred into Anabaena sp. strain PCC 7120 by conjuga-tion with RP4 as the helper plasmid (3, 13). A triparental

* Corresponding author.t Present address: Biotechnology Program, University of Hawaii,

Honolulu, HI 96822.

cross was made on a Millipore HATF filter on a platecontaining BGll agar supplemented with 5% LB agar (3).After 2 days of incubation under low light (60 microeinsteinsm-2 s-1) at 30°C, the filter was transferred to a BG11 platecontaining neomycin (30 ,ug/ml). The plate was furtherincubated under low light for 2 days and then transferred toan incubator at 125 microeinsteins m-2 s-' and sparged witha mixture of CO2 (1%) and air. The light blue-green color ofthe filter was almost completely bleached after 10 days, atwhich time the filter was transferred to a fresh BG11 agarplate containing neomycin (30 ,ug/ml). At the end of the thirdweek, several small blue-green colonies appeared on thefilter, the number of colonies increasing to about 200 to 250by the fourth week. When the colonies became at least 1 mmin diameter, they were transferred individually to a freshBG11 agar plate containing neomycin (30 ,ug/ml). SincepBR322::TnS cannot replicate in Anabaena spp., these neo-mycin-resistant colonies should have arisen through trans-position of TnS into the Anabaena genome.

Southern hybridization analysis. Total genomic DNA pre-pared from a few neomycin-resistant isolates was analyzedby Southern hybridization with a 2.0-kilobase (kb) internalregion of TnS (Fig. 1). It appeared that TnS transposed intodifferent DNA fragments in these mutants. When pBR322was used as the probe in Southern hybridization with theDNA isolated from these mutants, no hybridization wasdetected (data not shown). This result suggests that theneomycin-resistant mutants arose due to random transposi-tion of Tn5 and not due to integration of pBR322::TnS intothe Anabaena genome.

In one of the mutants (lane A, Fig. lc), there appeared tobe six or seven HindIII bands in addition to the 3.3-kbinternal HindIII band of Tn5, suggesting that there aremultiple copies of TnS in this isolate. All these fragments,including the 3.3-kb HindIII fragment of TnS in this isolate,can be seen as distinct visible bands in the stained gel (laneA, Fig. lb). No band of such intensity was seen in theHindlll digest of wild-type Anabaena DNA (lane K, Fig.lb). One explanation for the dark bands in this mutant is thatTnS transposed into a plasmid in the Anabaena strain andthe copy number of the plasmid increased as a result of theTnS insertion and drug selection. TnS transposes by aconservative mechanism that requires a Tn5-encoded trans-posase and several host factors (1). It is possible that, in thiscase, the original TnS transposed further, either within thesame plasmid or to other plasmids similar in size andfrequency of HindIII sites. This result was not investigatedfurther.

5759

on Novem

ber 6, 2017 by guesthttp://jb.asm

.org/D

ownloaded from

5760 NOTES

a~~~~-.-----6;

0..eFa~~~~~~~~~~~~~~~~~~~~~~~

4 k r:-

Ij"I k j j. (I 1.i i 'i K i P,kh \ l; 4

.24

-- f, 36

_. ..2

......_

.....__ as

I,

4'

FIG. 1. Southern analysis of Anabaena strains. (a) Map of TnS (8) showing the position of the neomycin phosphotransferase (neo) gene,the two inverted repeats (thick line), and the 2-kb region used as the probe. The 2-kb region of TnS shown above was cut out as a BamHIfragment from plasmid pSK101 (10). There is another PstI (P) site in the neo gene which is not shown here. TnS has two Hindlll (H) sites3.3 kb apart in the inverted repeats. (b) Approximately 2 ,ug of genomic DNA from Anabaena sp. strain PCC 7120 (lane K) and itsneomycin-resistant derivatives (lanes A to J) was digested with Hindlll, and the fragments were separated on a 0.8% agarose gel.BstII-digested lambda size markers are shown in lane L. (c) The fragments from this gel were transferred to a nylon membrane and hybridizedwith the labeled 2-kb DNA from TnS. Lane A was exposed for 4 h; lanes B to L were exposed for 12 h. The 3.3-kb band present in lanes Ato J of panel c corresponds to the internal HindlIl fragment of TnS. Each lane B to J contains, in addition, two HindlIl fragments identifiedby the PstI-HindIII fragment of TnS. These correspond to the portions of the Anabaena HindlIl fragment into which TnS transposed, togetherwith 1.2 kb from the inverted repeats at the termini of TnS. Lane A is discussed in the text.

Isolation of a mutant that is unable to grow without com-bined nitrogen. A total of 400 colonies isolated from fourconjugation plates were streaked on BG11 agar plates withand without combined nitrogen. One isolate, called T-123,

A B C A B C D

was found to be unable to grow on medium without com-bined nitrogen (Fix-). Strain T-123 formed heterocystswhich were indistinguishable in the light microscope fromthose of wild-type Anabaena sp. strain PCC 7120. Southernhybridization of total genomic digests of T-123 with the2.0-kb internal fragment of TnS (Fig. 1) as the probe showeda 9.2-kb EcoRI band and three HindIll fragments of 3.3, 2.3,and 1.9 kb (Fig. 2a), suggesting that the TnS was inserted ina 3.6-kb EcoRI fragment (Tn5 is 5.6 kb in size and has noEcoRI site) or in a 1.8-kb HindIll fragment (the 3.3-kb bandis the internal HindIII fragment of Tn5; both the 2.3- and the1.9-kb fragments contain a 1.2-kb region of DNA from thetwo inverted repeats of Tn5).The 9.2-kb EcoRI fragment containing the TnS and flank-

ing sequences from strain T-123 was cloned as follows. Acosmid library was made from EcoRI-digested DNA of

4m,. .r

u

am ,s

(a)

FIG. 2. Southern hybridization analysis of strain T-123 and plasmidpDB25. (a) Genomic DNA of strain T-123 digested with EcoRI (laneA) and HindlIl (lane B) and plasmid pDB25 digested with HindlIl(lane C) and probed with the 2-kb internal fragment of TnS (see Fig.1). (b) EcoRI-digested genomic DNA of strain PCC 7120 (lane A)and strain T-123 (lane B) and HindlIl-digested genomic DNA ofstrain PCC 7120 (lane C) and plasmid pDB25 (lane D) were probedwith the 1.9-kb HindlIl fragment of plasmid pDB25. Conditions ofelectrophoresis and DNA digestion are as described in the legend toFig. 1; approximately 100 ng of DNA was loaded in each lane forplasmid pDB25.

J. BACTERIOL.

go

00 - " .-,.

(b}""Ir

on Novem

ber 6, 2017 by guesthttp://jb.asm

.org/D

ownloaded from

NOTES 5761

A B C D

-9.2 kb

-3.6 kb

_ -2.3 Rb- _-1.9 kb-MW' -1.8 kb

(b)

FIG. 3. Southern analysis of the wild-type Anabaena DNA frag-ment cloned in X26. (a) The 1.9-kb HindIll fragment of pDB25 was

used as the probe. Lane A, EcoRI-digested genomic DNA ofstrain PCC 7120. Lane B, X26 DNA digested with EcoRI. Lane C,HindIll-digested genomic DNA of strain PCC 7120. Lane D,HindIll-digested X26. (b) The 1.8-kb HindIll fragment of X26 was

used to probe the genomic DNA of strains PCC 7120 and T-123.Lane A, PCC 7120, EcoRI; lane B, T-123, EcoRI; lane C, PCC 7120,HindIll; lane D, T-123, HindIll. Conditions of electrophoresis andDNA digestion are as described in the legend to Fig. 1; approxi-mately 50 ng of X26 DNA was loaded in lanes B and D of panel a.

strain T-123 in the cosmid vector pLAFR1 (4). One cosmidclone, called pDB25, containing the 9.2-kb EcoRI fragmentwith the Tn5 insert together with an unrelated 10.5-kbfragment was isolated by plating the cosmid library on LBagar containing kanamycin and tetracycline. The 9.2-kbEcoRI fragment in pDB25 was found to be the same size asthe EcoRI fragment of strain T-123 that hybridized with the2.0-kb internal fragment of TnS (data not shown). The threeHindlll fragments (3.3, 2.3, and 1.9 kb) of strain T-123 thathybridized with this probe were present in the Hindlll digestof pDB25 and hybridized with the same probe (lane C in Fig.2a).The 1.9-kb HindlIl fragment of pDB25 containing a 1.2-kb

region from TnS and 0.7 kb of flanking sequences was usedto probe genomic digests of Anabaena DNA (Fig. 2b). Asexpected, it hybridized with a 3.6-kb EcoRI and a 1.8-kbHindIll fragment of strain PCC 7120. The same probe wasthen used to screen a lambda Charon4 library of EcoRI-digested DNA from Anabaena sp. strain PCC 7120 (9), andone positive clone (A26) that hybridized with the probe was

identified. X26 contained a 3.6-kb EcoRI fragment and a

1.8-kb HindlIl fragment that hybridized with the 1.9-kbHindlll fragment of pDB25 (Fig. 3a). The 1.8-kb Hindlllfragment of X26 was then used as a probe of the genomicdigests of strains PCC 7120 and T-123 (Fig. 3b). It hybridizedwith a 3.6-kb EcoRI and a 1.8-kb Hindlll fragment of strainPCC 7120 and a 9.2-kb EcoRI fragment and two Hindlll

fragments (2.3 and 1.9 kb) of strain T-123, indicating that A26contained the wild-type fragment of DNA that was inter-rupted by TnS in strain T-123.

Thus, we have shown that TnS transposes in Anabaenqsp. strain PCC 7120 and that TnS tagging may be a useful toolfor cloning and analyzing genes in Anabaena spp. The 3.6-kbEcoRI or the 1.8-kb Hindlll fragment may contain afix gene,which will require further analysis. The 3.6-kb EcoRI frag-ment identified and cloned in this study is not located in the39-kb nif region mapped earlier (9).

We thank J. Elhai for plasmid pRL528 and W. J. Buikema foruseful discussions.

This work was supported by Public Health Service research grantGM21823 from the National Institutes of Health.

LITERATURE CITED1. Berg, D. E. 1989. Transposon TnS, p. 185-210. In D. E. Berg

and M. M. Howe (ed.), Mobile DNA. American Society forMicrobiology, Washington, D.C.

2. Currier, T. C., J. F. Haury, and C. P. Wolk. 1977. Isolation andpreliminary characterization of auxotrophs of a filamentouscyanobacterium. J. Bacteriol. 129:1556-1562.

3. Elhai, J., and C. P. Wolk. 1988. Conjugal transfer of DNA tocyanobacteria. Methods Enzymol. 167:747-754.

4. Friedman, A. M., S. E. Brown, W. J. Buikema, and F. M.Ausubel. 1982. Construction of a broad host range cloningvector and its use in the genetic analysis of Rhizobium mutants.Gene 18:269-296.

5. Golden, J. W., and D. R. Wiest. 1988. Genome rearrangementand nitrogen fixation in Anabaena blocked by inactivation ofxisA gene. Science 242:1421-1423.

6. Grillo, J. F., P. J. Bottomley, C. Van Baalen, and F. R. Tabita.1979. A mutant of Anabaena sp. CA with oxygen-sensitivenitrogenase activity. Biochem. Biophys. Res. Commun. 89:685-693.

7. Haury, J. F., and C. P. Wolk. 1978. Classes of Anabaenavariabilis mutants with oxygen-sensitive nitrogenase activity. J.Bacteriol. 136:688-692.

8. Jorgensen, R. A., S. J. Rothstein, and W. S. Reznikoff. 1979. Arestriction enzyme cleavage map of TnS and location of a regionencoding neomycin resistance. Mol. Gen. Genet. 177:65-72.

9. Rice, D., B. Mazur, and R. Haselkorn. 1982. Isolation andphysical mapping of nitrogen fixing genes from the cyanobacte-rium Anabaena 7120. J. Biol. Chem. 257:13157-13163.

10. Shapira, S. K., J. Chou, F. V. Richaud, and M. J. Casadaban.1983. New versatile plasmid vectors for expression of hybridproteins coded by a cloned gene fused to lacZ gene sequencesencoding an enzymatically active carboxy-terminal portion ofP-galactosidase. Gene 25:71-82.

11. Spence, D. W., and W. D. P. Stewart. 1987. Heterocystlessmutants of Anabaena PCC7120 with nitrogenase activity.FEMS Microbiol. Lett. 40:119-122.

12. Wilcox, M., G. J. Mitchison, and R. J. Smith. 1975. Mutants ofAnabaena cylindrica altered in heterocyst spacing. Arch. Mi-crobiol. 103:219-223.

13. Wolk, C. P., Y. Cai, L. Cardemil, E. Flores, B. Hohn, M. Murry,G. Schmetterer, B. Schrautemeier, and R. Wilson. 1988. Isola-tion and complementation of mutants of Anabaena sp. straipPCC 7120 unable to grow aerobically on dinitrogen. J. Bacteriol.170:1239-1244.

14. Wolk, C. P., A. Vonshak, P. Kehoe, and J. Elhai. 1984. Con-struction of shuttle vectors capable to conjugative transfer froriEscherichia coli to nitrogen-fixing filamentous cyanobacteria.Proc. Natl. Acad. Sci. USA 81:1561-1565.

A B C D

-3.6 kb

(a)

VOL. 171, 1989

kb

on Novem

ber 6, 2017 by guesthttp://jb.asm

.org/D

ownloaded from