JOURNAL OF BACTERIOLOGY, Mar. 2003, p. 1745–1748
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Vol. 185, No. 5
Transposition of Tn5367 in Mycobacterium marinum, Using a
Conditionally Recombinant Mycobacteriophage
Jan Rybniker, Martina Wolke, Christiane Haefs, and Georg Plum*
Institute for Medical Microbiology, Immunology and Hygiene, University of Cologne, Cologne, Germany
Received 27 September 2002/Accepted 6 December 2002
Mycobacterium marinum is a close relative of the obligate human pathogen Mycobacterium tuberculosis. As
with M. tuberculosis, M. marinum causes intracellular infection of poikilothermic vertebrates and skin infection
in humans. It is considered a valid model organism for the study of intracellular pathogenesis of mycobacteria.
Low transformation efficiencies for this species have precluded approaches using mutant libraries in patho-
genesis studies. We have adapted the conditionally replicating mycobacteriophage phAE94, originally devel-
oped as a transposon mutagenesis tool for M. tuberculosis, to meet the specific requirements of M. marinum.
Conditions permissive for phage replication in M. tuberculosis facilitated highly efficient transposon delivery in
M. marinum. Using this technique we succeeded in generating a representative mutant library of this species,
and we conclude that TM4-derived mycobacteriophages are temperature-independent suicide vectors for M.
Mycobacterium marinum is a pathogenic, slow-growing
member of the genus Mycobacterium (18). It causes fatal in-
fection in salt- and freshwater fish as well as in amphibians. In
humans, M. marinum is the causative agent of a disease called
swimming pool granuloma (2, 8, 23), a chronic skin infection of
the extremities that can convert into a systemic infection in
immunocompromised patients such as those with AIDS (10,
Though classified as a slow-growing mycobacterium, M. ma-
rinum has a relatively short generation time of 4 to 6 h com-
pared to 20 h for Mycobacterium tuberculosis. The optimal
growth temperature range for most M. marinum isolates is 25
Like M. tuberculosis and other virulent mycobacteria, M.
marinum survives and replicates in host macrophages, where it
prevents phagosome maturation (1, 4, 7, 12). There are also
significant genotypic and phenotypic similarities between these
species, which implies that M. marinum is a potent model
system for the study of mycobacterial pathogenesis (4). The
major advantages over M. tuberculosis are the faster growth
and the safer handling of M. marinum in the laboratory. Work-
ing with it requires only common laboratory precautions (bio-
safety level 2). The complete genome of M. marinum (strain M,
human isolate) is currently being sequenced by the Sanger
Institute. The available coverage of the genome was 99.99%
complete at the time that this paper was submitted.
The mycobacteriophages phAE77 and phAE94 have been
used as powerful tools for transposition mutagenesis in several
fast- and slow-growing mycobacterial species (3, 11). Both
phages are highly efficient in delivering the mycobacterial
transposon Tn5367, while they are incapable of replicating at
37°C due to the presence of temperature-sensitive mutations.
Tn5367 is an IS1096-derived insertion element containing a
kanamycin resistance gene as a selectable marker (14).
PhAE77 is a derivative of the well-described lytic mycobacte-
riophage D29, whereas the progenitor of phAE94 is mycobac-
teriophage TM4, which was isolated as a temperate phage of
Mycobacterium avium (22). Both phages are propagated in
Mycobacterium smegmatis at the replication-permissive tem-
perature of 32°C and are used for transposon delivery at 37°C.
The transposable element Tn5367 has been shown to insert
randomly into the genome of M. marinum isolate ATCC 927
after transformation using the electroporation method. How-
ever, the transposition frequency was too low for the genera-
tion of a representative mutant library and it was suggested
that the low transposition frequency derives from a low trans-
formation frequency (20). In this study we have overcome this
phAE94 as a delivery vector for Tn5367. We were able to
create a comprehensive bank of kanamycin-resistant M. mari-
num mutants, showing that a TM4-derived vector is a potent
tool for the transduction of M. marinum.
Strains. A fish isolate of M. marinum (ATCC 927) was
grown at 32°C in Middlebrook 7H9 broth enriched with 10%
oleic acid–albumin–dextrose complex (without Tween) in a
stirring bottle. Using the double-agar-layer method, phAE77
and phAE94 were propagated in M. smegmatis mc2155 at 32°C
(13). After infection, M. marinum cells were plated on Middle-
brook 7H10 agar supplemented with 0.1% Tween 80, 0.4%
Casamino Acids, 40 ?g of tryptophan/ml, and 40 ?g of kana-
Transposon mutagenesis. M. marinum cultures were grown
for 10 days to approximately 2 ? 108CFU/ml (at an optical
density at 600 nm of 0.8). A total of 10 ml of this culture was
concentrated by centrifugation and resuspended in 1 ml of MP
buffer (50 mM Tris-HCl [pH 7.6], 150 mM NaCl, 2 mM CaCl2)
(13). Then, 1010PFU of phAE77 or phAE94 was added and
the mixture was incubated at the nonpermissive temperature
(37°C) for 5 h in a shaking incubator to inhibit a possible lytic
* Corresponding author. Mailing address: Institut fu ¨r Medizinische
Mikrobiologie, Immunologie und Hygiene, Klinikum der Universita ¨t
zu Ko ¨ln, Goldenfelsstr. 21, 50935 Cologne, Germany. Phone: 49-221-
478-3066. Fax: 49-221-478-3094. E-mail: firstname.lastname@example.org.
or lysogenic cycle of the phage. Adsorption stop buffer (20 mM
sodium citrate and 0.2% Tween 80) was added to prevent
further phage infections. One-third of this mixture was plated
immediately on 7H10 agar with kanamycin and incubated at
32°C. The remaining cells were kept on 37°C for 6 or 12 h
before being plated to prevent a lytic or lysogenic cycle of the
phage. Transposition frequency was expressed as the number
of kanamycin-resistant (Kanr) colonies per milliliter of input
Analysis of the Kanrmutants. Kanamycin-resistant M. ma-
rinum colonies were picked and grown in 7H9 broth containing
40 ?g of this aminoglycoside/ml. Mycobacterial DNA was iso-
lated as previously described (15). To reveal random transpo-
sition of the marker gene, the restriction enzyme BamHI was
chosen, as it cuts in the flanking genomic sequences but not
within the transposon itself. Digested genomic DNA was blot-
ted onto nylon membrane and hybridized to a non-radioactive-
ly-labeled IS1096 probe (RPN3000; Amersham Pharmacia).
The digested DNA of M. smegmatis mc2155, the natural host of
IS1096, was used as a positive control.
To determine whether the prophage or parts of the pro-
phage had integrated into the genome of the KanrM. marinum
mutants, DNAs of eight different mutants as well as phAE94
genomic DNA as a positive control were digested with PstI,
which cuts more frequently in phAE94 than BamHI. Using the
system described above, Southern blotting was performed; two
PCR products of essential genes of phAE94 were labeled as a
The transposon mutants were further characterized by se-
quencing randomly primed PCR products (5, 16). This method
uses a specific primer from a region within Tn5367 together
with an arbitrary primer in a first round of PCR. The product
is used as a template for a second, nested PCR using a specific
primer also derived from Tn5367 but closer to the flanking
region and a subprimer of the arbitrary primer as a second
primer. This second-round PCR product can be sequenced
using the second specific primer. The arbitrary primers were
NN-3?) and ARB2 (5?GGCCACGCGTCGACTAGTAC-3?).
Specific primers were RPCRa1 (5?-CTTGCTCTTCCGCTTC
TTCTC-3?) and RPCRa2 (5? CTCTACACCGTCAAGTGCG
AAGAG-3?) as well as RPCRb1 (5?-CAGGCACGTCGAGG
TCTTTC-3?) and RPCRb2 (5?-CTTTCAGATGGATGGCGT
AG-3?) for the opposite side of the transposon. First- and
second-round conditions were the same as those previously
described (5). Reaction procedures were performed in a Pel-
tier thermal cycler model PTC-200 (MJ Research, Waltham,
Mass). Second-round products were purified using a PCR pu-
rification kit (Qiagen) and sequenced using the Big-Dye ter-
minator cycle sequencing ready reaction kit (Applied Biosys-
tems) and an ABI Prism 310 genetic analyzer (Applied
Our goal was to adapt the conditionally replicating myco-
bacteriophage system as a tool for transposon mutagenesis in
M. marinum. Using phAE94, the yield of KanrM. marinum was
very high, with up to 105colonies per ml of transduced culture
in a single experiment. In contrast, phAE77 yielded signifi-
cantly fewer transposon mutants under the same conditions
(103per ml of transduced culture), so further experiments and
the generation of the mutant library were done with phAE94.
Interestingly, keeping the infected cells at 37°C for 6 and 12 h
had no influence on the number of transposon mutants. Even
when the incubation temperature of the cell-phage mixture
was decreased to 32°C throughout the whole experiment, the
yield of mutants was the same as for the 37°C trial. Spotting 5
?l of a phAE94 dilution on 7H10 top agar plates containing M.
marinum did not produce any clear or turbid plaques that
would indicate lytic growth or a lysogenic state of the phage in
M. marinum. This implies that TM4-derived mycobacterio-
phages are temperature-independent suicide vectors for M.
marinum, which might be an important feature for future ex-
periments with this model bacterium.
To determine whether Tn5367 inserted in a random fashion
into the M. marinum genome, eight mutants were picked and
analyzed by Southern hybridization (Fig. 1). Seven of eight
screened mutants showed a single hybridizable band in differ-
ent chromosomal locations, indicating a random insertion of
the transposon. One transposon mutant displayed two bands of
equal intensity, which might have been the result of two trans-
position events in the chromosome or of the presence of a
mixed colony containing two individual mutants. The positive
control showed several bands, as there are multiple copies of
IS1096 in M. smegmatis (6).
Using the randomly primed PCR method, sequence analysis
of the Tn5367 insertion junctions yielded good-quality se-
quences for all examined mutants. A BLAST analysis of the
sequences revealed different results for the chromosomal DNA
adjacent to the ends of the transposon. The sequences ob-
tained using primer RPCRa2 showed the expected GC-rich
mycobacterial genomic DNA in all mutants. The majority had
a high level of homology to M. marinum and M. tuberculosis.
The opposite side of the transposon (sequencing primer
RPCRb2) showed an insertion of parts of the cloning vector
pYUB552 that flanks Tn5367 in the phAE94 genome. In 2 out
of 20 examined mutants, small parts of TM4 genomic DNA
adjacent to this cloning vector had additionally integrated into
the chromosome. However, further analysis of these sequences
revealed that this cointegration of vector and phage DNA
stops at different points for all examined mutants and merges
FIG. 1. Southern blot analysis of M. marinum Tn5367 mutants.
Lanes 1 to 10 show BamHI digests of total chromosomal DNA hybrid-
ized to a labeled IS1096 probe. Lane L, DNA markers (sizes in kilo-
base pairs are indicated on the left); lane 1, wild type; lanes 2 to 9, Kanr
insertion mutants; lane 10, M. smegmatis mc2155.
into the mycobacterial genome. Southern hybridization and
PCR analysis with several primers distributed over the TM4
genome were performed to detect residual phAE94 sequences
in the M. marinum mutants. All examined mutants were neg-
ative for the presence of the sequences (data not shown). All
insertions were flanked by a unique 8-bp target duplication
such as was described previously for the transposition of
Tn5367 in other mycobacterial strains (3, 11). The sequenced
chromosomal DNA produced high-scoring segment pairs when
analyzed using the BLASTN program (BLASTN 2.0 MP
[Washington University] [http://blast.wustl.edu]) with the al-
most completed M. marinum genome data (Table 1) at the
Sanger Institute website (http://www.sanger.ac.uk/Projects/
M_marinum/blast_server.shtml). An additional BLAST analy-
sis using the National Center for Biotechnology Information
that the sequences of some of these mutants showed a high
percentage of similarity to those of M. tuberculosis genes (Ta-
ble 1). These data imply that Tn5367 transposes with no se-
quence specificity into the genome of M. marinum after the
highly efficient vector phAE94 is used for transfection of the
Interestingly, using the same method to try to mutagenize
Mycobacterium ulcerans, a very closely related slow-growing
mycobacterium with an optimal growth temperature of 32°C,
also resulted in a high number of mutants. Unfortunately,
these mutants showed an extremely low growth rate, with an
estimated generation time of 60 h (data not shown). Due to the
slow growth, it was not possible to isolate sufficient amounts of
DNA to perform a Southern blot analysis. However, by se-
quencing randomly primed PCR products we were able to
identify M. ulcerans sequences adjacent to the transposon. The
reason for the inhibition of growth remained unclear, and
attempts to colonize the mutants on agar without kanamycin
did not accelerate the growth of the bacteria. In contrast to M.
marinum, spotting a dilution of phAE94 on top agar containing
M. ulcerans and incubating at 32°C leads to turbid plaques,
implying that stable lysogens are formed. This shows that
phAE94 is a temperate phage of M. ulcerans. Though incuba-
tion was at 37°C for several hours during the transposon mu-
tagenesis, an integration of the phAE94 prophage together
with phage-derived repressor genes in the Kanrmutants might
have occurred, leading to the observed growth inhibition.
As additional evidence for successful transposon mutagen-
esis in M. marinum, a library of approximately 5 ? 104KanrM.
marinum mutants was pooled and plated on 7H10 agar plates
containing 40 ?g of ethionamide/ml. This compound is a pro-
drug that requires activation by bacterial enzymes. Drug resis-
tance results from genetic mutation of the enzymes of the
activation pathway (9, 19). As expected, the number of ethio-
namide-resistant colonies was significantly larger for the trans-
poson mutant library than for the wild-type strain. We ob-
served 45 ethionamide-resistant cells per 105plated bacteria
for the library compared to 8.4 cells per 105plated mycobac-
teria of the M. marinum wild-type strain, i.e., an increase by a
factor of 5.3.
This work was supported by a studentship of Stiftung Maria Pesch.
We thank W. R. Jacobs, Jr., for providing phAE77 and phAE94. We
also thank F. Portaels for the M. marinum and M. ulcerans strains used
in this work and for the use of laboratory facilities. We thank P. L. C.
Small for the use of laboratory facilities and Brian Ranger for proof-
1. Armstrong, J. A., and P. D. Hart. 1971. Response of cultured macrophages
to Mycobacterium tuberculosis, with observations on fusion of lysosomes with
phagosomes. J. Exp. Med. 134:713–740.
2. Aubry, A., O. Chosidow, E. Caumes, J. Robert, and E. Cambau. 2002.
Sixty-three cases of Mycobacterium marinum infection: clinical features,
treatment, and antibiotic susceptibility of causative isolates. Arch. Intern.
3. Bardarov, S., J. Kriakov, C. Carriere, S. Yu, C. Vaamonde, R. A. McAdam,
B. R. Bloom, G. F. Hatfull, and W. R. Jacobs, Jr. 1997. Conditionally
replicating mycobacteriophages: a system for transposon delivery to Myco-
bacterium tuberculosis. Proc. Natl. Acad. Sci. USA 94:10961–10966.
4. Barker, L. P., K. M. George, S. Falkow, and P. L. Small. 1997. Differential
trafficking of live and dead Mycobacterium marinum organisms in macro-
phages. Infect. Immun. 65:1497–1504.
5. Chen, T., H. Dong, R. Yong, and M. J. Duncan. 2000. Pleiotropic pigmen-
tation mutants of Porphyromonas gingivalis. Microb. Pathog. 28:235–247.
6. Cirillo, J. D., R. G. Barletta, B. R. Bloom, and W. R. Jacobs, Jr. 1991. A
novel transposon trap for mycobacteria: isolation and characterization of
IS1096. J. Bacteriol. 173:7772–7780.
7. Clemens, D. L., and M. A. Horwitz. 1995. Characterization of the Mycobac-
terium tuberculosis phagosome and evidence that phagosomal maturation is
inhibited. J. Exp. Med. 181:257–270.
8. Collins, C. H., J. M. Grange, W. C. Noble, and M. D. Yates. 1985. Mycobac-
terium marinum infections in man. J. Hyg. (London) 94:135–149.
9. DeBarber, A. E., K. Mdluli, M. Bosman, L. G. Bekker, and C. E. Barry III.
2000. Ethionamide activation and sensitivity in multidrug-resistant Mycobac-
terium tuberculosis. Proc. Natl. Acad. Sci. USA 97:9677–9682.
10. Hanau, L. H., A. Leaf, R. Soeiro, L. M. Weiss, and S. S. Pollack. 1994.
Mycobacterium marinum infection in a patient with the acquired immuno-
deficiency syndrome. Cutis 54:103–105.
11. Harris, N. B., Z. Feng, X. Liu, S. L. Cirillo, J. D. Cirillo, and R. G. Barletta.
1999. Development of a transposon mutagenesis system for Mycobacterium
avium subsp. paratuberculosis. FEMS Microbiol. Lett. 175:21–26.
12. Hart, P. D., M. R. Young, A. H. Gordon, and K. H. Sullivan. 1987. Inhibition
of phagosome-lysosome fusion in macrophages by certain mycobacteria can
be explained by inhibition of lysosomal movements observed after phagocy-
tosis. J. Exp. Med. 166:933–946.
13. Jacobs, W. R., Jr., G. V. Kalpana, J. D. Cirillo, L. Pascopella, S. B. Snapper,
R. A. Udani, W. Jones, R. G. Barletta, and B. R. Bloom. 1991. Genetic
systems for mycobacteria. Methods Enzymol. 204:537–555.
14. McAdam, R. A., T. R. Weisbrod, J. Martin, J. D. Scuderi, A. M. Brown, J. D.
TABLE 1. Mapping and analysis of the sequences adjacent to Tn5367 insertions
% Sequence similarity to
M. marinum strain Ma
Protein or sequence of interrupted gene
(% sequence similarity to M. marinum wild type)b
PE-PGRS protein of M. tuberculosis H37Rv (73)
Hypothetical protein Rv2030c of M. tuberculosis H37Rv (72)
Putative DNA-binding protein of Streptomyces coelicolor (62)
M. tuberculosis sequence from clone y414b (82)
Hypothetical protein Rv3903c of M. tuberculosis H37Rv (62)
aData calculated using the M. marinum BLAST server (Sanger).
bData calculated using the National Center for Biotechnology Information database.
VOL. 185, 2003 NOTES1747
Cirillo, B. R. Bloom, and W. R. Jacobs, Jr. 1995. In vivo growth character- Download full-text
istics of leucine and methionine auxotrophic mutants of Mycobacterium bovis
BCG generated by transposon mutagenesis. Infect. Immun. 63:1004–
15. Mve-Obiang, A., M. Mestdagh, and F. Portaels. 2001. DNA isolation from
chloroform/methanol-treated mycobacterial cells without lysozyme and pro-
teinase K. BioTechniques 30:272–274, 276.
16. O’Toole, G. A., and R. Kolter. 1998. Initiation of biofilm formation in
Pseudomonas fluorescens WCS365 proceeds via multiple, convergent signal-
ling pathways: a genetic analysis. Mol. Microbiol. 28:449–461.
17. Parent, L. J., M. M. Salam, P. C. Appelbaum, and J. H. Dossett. 1995.
Disseminated Mycobacterium marinum infection and bacteremia in a child
with severe combined immunodeficiency. Clin. Infect. Dis. 21:1325–1327.
18. Rogall, T., J. Wolters, T. Flohr, and E. C. Bo ¨ttger. 1990. Towards a phylog-
eny and definition of species at the molecular level within the genus Myco-
bacterium. Int. J. Syst. Bacteriol. 40:323–330.
19. Sassetti, C. M., D. H. Boyd, and E. J. Rubin. 2001. Comprehensive identi-
fication of conditionally essential genes in mycobacteria. Proc. Natl. Acad.
Sci. USA 98:12712–12717.
20. Talaat, A. M., and M. Trucksis. 2000. Transformation and transposition of
the genome of Mycobacterium marinum. Am. J. Vet. Res. 61:125–128.
21. Tchornobay, A. M., A. L. Claudy, J. L. Perrot, V. Levigne, and M. Denis.
1992. Fatal disseminated Mycobacterium marinum infection. Int. J. Derma-
22. Timme, T. L., and P. J. Brennan. 1984. Induction of bacteriophage from
members of the Mycobacterium avium, Mycobacterium intracellulare, Myco-
bacterium scrofulaceum serocomplex. J. Gen. Microbiol. 130:2059–2066.
23. Wolinsky, E. 1992. Mycobacterial diseases other than tuberculosis. Clin.
Infect. Dis. 15:1–10.
1748NOTES J. BACTERIOL.