Horizontal gene transfer (HGT) as a mechanism of disseminating RDX-degrading activity among Actinomycete bacteria

Journal of Applied Microbiology ISSN 1364-5072

ORIGINAL ARTICLE

Horizontal gene transfer (HGT) as a mechanism of disseminating RDX-degrading activity among Actinomycete bacteria C.M. Jung, F.H. Crocker, J.O. Eberly and K.J. Indest Environmental Laboratory, US Army Engineer Research and Development Center, Vicksburg, MS, USA

Keywords conjugation, degradation, Gordonia, hexahydro-1,3,5-trinitro-1,3,5,-triazine, horizontal gene transfer, pGKT2. Correspondence Karl J. Indest, Environmental Laboratory, CEERDC-EP-P, US Army Engineer Research and Development Center, 3909 Halls Ferry Road, Vicksburg, MS 39180, USA. E-mail: [email protected]

2010 ⁄ 1564: received 7 September 2010, revised 1 February 2011 and accepted 5 March 2011 doi:10.1111/j.1365-2672.2011.04995.x

Abstract Aims: Hexahydro-1,3,5-trinitro-1,3,5,-triazine (RDX) is a cyclic nitramine explosive that is a major component in many high-explosive formulations and has been found as a contaminant of soil and groundwater. The RDX-degrading gene locus xplAB, located on pGKT2 in Gordonia sp. KTR9, is highly conserved among isolates from disparate geographical locations suggesting a horizontal gene transfer (HGT) event. It was our goal to determine whether Gordonia sp. KTR9 is capable of transferring pGKT2 and the associated RDX degradation ability to other bacteria. Methods and Results: We demonstrate the successful conjugal transfer of pGKT2 from Gordonia sp. KTR9 to Gordonia polyisoprenivorans, Rhodococcus jostii RHA1 and Nocardia sp. TW2. Through growth and RDX degradation studies, it was demonstrated that pGKT2 conferred to transconjugants the ability to degrade and utilize RDX as a nitrogen source. The inhibitory effect of exogenous inorganic nitrogen sources on RDX degradation in transconjugant strains was found to be strain specific. Conclusions: Plasmid pGKT2 can be transferred by conjugation, along with the ability to degrade RDX, to related bacteria, providing evidence of at least one mechanism for the dissemination and persistence of xplAB in the environment. Significance and Impact of Study: These results provide evidence of one mechanism for the environmental dissemination of xplAB and provide a framework for future field relevant bioremediation practices.

Introduction Many military ranges in the United States and Canada are contaminated with explosive and propellant residues (Jenkins et al. 1998, 2001; Pennington et al. 2001, 2005; Walsh et al. 2001). Of the nitroaromatic and nitramine explosives contaminants on these sites, hexahydro-1,3,5trinitro-1,3,5,-triazine (RDX) is known to migrate easily through soil into surface water and groundwater (Sheremata et al. 2001; Clausen et al. 2004; Davis et al. 2004). Bioremediation of RDX in soils and groundwater has been demonstrated (Davis et al. 2004; Fuller et al. 2005; Shrout et al. 2005), and a number of micro-organisms have been characterized that can degrade or transform RDX

(Coleman et al. 1998; Seth-Smith et al. 2002; Thompson et al. 2005; Zhao et al. 2005; Ronen et al. 2008). Gordonia sp. KTR9, which can utilize RDX as a sole source of nitrogen for aerobic growth (Thompson et al. 2005), contains the cytochrome P450-like enzyme, XplA, responsible for RDX degradation (Seth-Smith et al. 2002, 2008). XplA is encoded by the second gene of the xplAB genetic locus that resides on a 182-kb plasmid, pGKT2, in Gordonia sp. KTR9 (Indest et al. 2010). XplA is an unusual microbial cytochrome P450 in that the flavodoxin domain is fused to a C-terminal cytochrome P450 domain. XplB serves as a partner NADH-utilizing flavodoxin reductase. Together, these enzymes catalyse the denitration of RDX to yield 4-nitro-2,4-diazabutanal (NDAB), nitrous oxide,

Journal of Applied Microbiology 110, 1449–1459 ª 2011 The Society for Applied Microbiology No claim to US Government works

1449

Genetic transfer of RDX degradation

C.M. Jung et al.

ammonium, formaldehyde and carbon dioxide (Fournier et al. 2002; Jackson et al. 2007). The xplAB genes show a high level of conservation (>98%), at the amino acid sequence level, despite that the xplAB complex has been independently isolated from disparate geographical locations. For example, Gordonia sp. KTR9 (Thompson et al. 2005) and Microbacterium sp. MA1 (Andeer et al. 2009) were isolated in North America, CA and TN, respectively. Rhodococcus sp. DN22 (Coleman et al. 1998), Rhodococcus rhodochrous 11Y (Seth-Smith et al. 2002) and Rhodococcus sp. YH1 (Nejidat et al. 2008) were isolated from Australia, England and Israel, respectively. Lateral transfer of xplAB has been implied from the analysis of DNA sequencing data of this gene locus and associated flanking region (Andeer et al. 2009). The most likely explanation for this conservation is that the RDX-degrading gene locus was the result of an isolated evolutionary event that was subsequently disseminated over large geographical distances. Similarly, there have been other accounts of genes involved in halogenated aliphatic and aromatic degradation that are both highly conserved and widely distributed geographically (Janssen et al. 2005). For example, nearly identical genes coding for atrazine chlorohydrolase (AtzA), the first step in atrazine degradation, have been isolated from different bacterial sources from multiple locations around the world (Janssen et al. 2005). Similarly, genes involved in pentachlorophenol degradation, specifically pcpB, were found to be highly conserved over numerous bacterial clades isolated from various sources (Crawford et al. 2007). In most instances, these conserved catabolic genes are associated with insertion sequences and transposable elements (Janssen et al. 2005). Thus far, homologs of xplAB have been found on large plasmids in various Actinomycetes including Rhodococcus sp. DN22 (Coleman et al. 2002), R. rhodochrous 11Y (Seth-Smith et al. 2002) and Microbacterium sp. MA1 (Andeer et al. 2009). Limited information is available on the incidence of conjugal transfer for large catabolic plasmids in the Actinomycetes bacterial group (Bro¨ker et al. 2008). In an attempt to demonstrate HGT of pGKT2, conjugal matings between Gordonia sp. KTR9 and other Actinomycetes as well as more distantly related bacteria were carried out. Resultant transconjugants were assessed for the ability to degrade and utilize RDX as a sole source of nitrogen. The inhibitory effects of exogenous inorganic nitrogen sources on RDX degradation in each transconjugant compared to the donor Gordonia sp. KTR9 were also evaluated. Overall, our results demonstrate that pGKT2 can be successfully transferred between select Actinobacteria, conferring upon the host the capacity to degrade RDX and the unique ability to utilize RDX as a sole source of nitrogen. 1450

Materials and methods Construction of a traceable marker and conjugation A predicted nonessential coding region of the Gordonia sp. KTR9 plasmid pGKT2, found on ORF pGKT2_4819 (Genbank Acc. No. CP002112), with no significant matches in the Genbank database, was targeted for homologous recombination and insertion of a kanamycin resistance (Kmr) marker (NP_478145.1). This site is approx. 40 kb downstream from the xplAB gene complex. The Kmr gene, flanked by 1-kb regions specific to pGKT2, was synthesized by Celtek Biosciences, LLC (Nashville, TN, USA), into a pCR2Æ1 vector (Invitrogen, Carlsbad, CA, USA), (Fig. 1). The Kmr construct containing the KTR9 flanking regions (2Æ9 kb) was liberated from the plasmid by digestion with BamHI (New England BioLabs, Inc., Ipswich, MA, USA) and gel purified with a Wizard SV Gel and PCR Clean-up System Kit (Promega, Madison, WI, USA). The purified 2Æ9-kb fragment was ligated into a BamHI-cut mobilizable vector, pK18mobsacB (Scha¨fer et al. 1994; van der Geize et al. 2001)(Fig. 1), and transformed into One Shot Top10 Chemically Competent Escherichia coli cells (Invitrogen) that were selected for on Luria–Bertani plates with 50 lg ml)1 kanamycin. Recombinant pK18mobsacB was introduced from Top10 E. coli cells into Gordonia sp. KTR9 based on the conjugation strategy presented by van der Geize et al. (2001). Double crossover transconjugants of the Kmr marker from pK18mobsacB into pGKT2 via homologous flanking regions were selected for on LBP (per L: 10 g peptone, 10 g NaCl, 5 g yeast extract) with 50 lg ml)1 kanamycin and 10% sucrose. Each step was screened by PCR (Table 1) targeting the Kmr construct and various locations of the pGKT2 and pK18mobsacB plasmids (Fig. 1) to confirm their presence or, in the case of pK18mobsacB, its absence. All PCR amplification conditions consisted of a single step at 94C for 5 min followed by 30 cycles at 94C for 30 s, with annealing and elongation times and temperatures for each primer set as listed in Table 1, and a final step at 72C for 5 min. Conjugation reactions between the kanamycin resistant, orange-pigmented Gordonia sp. KTR9 donor strain and recipient bacteria were carried out by the filter mating method of Lessard et al. (2004). Resultant transconjugants that were capable of growth on LBP with 100 lg ml)1 kanamycin were then screened by PCR (Table 1) targeting various locations of pGKT2 (Fig. 1) to confirm its presence. Recipient strains were chosen based on colony morphology, ability to grow at 30C on LBP medium, sensitivity to kanamycin and inability to utilize RDX as a sole nitrogen source. Recipients listed as follows: Arthrobacter aurescens TW17 (Hanne et al. 1993),

Journal of Applied Microbiology 110, 1449–1459 ª 2011 The Society for Applied Microbiology No claim to US Government works

C.M. Jung et al.

Genetic transfer of RDX degradation

Bam HI

C

on

st

ru

Cons 40788 truct f-4120 1r

Kmr construct

BamH

I

hII cBap mbs f-5596r Ba mH 4 5 52 I

pGKT2::Kmr

uct

8642 bp

Ba

I mH

SacB

pK18mobsacB::Kmr

f 5a 56 20 48ar 1 s 2 la 121 9p Ktr 9plas Ktr

182454 bp nstr r Km co

mbs c 3801 BsacB f-422 3r

xp/BA region Ktr9 Ktr9 plas14 pl a 9 s15 923b f 070 1br

Kanco ns 363f-8 rt 25r

m 21 bs 87 cB f -2 pla 64 s 1r

Figure 1 Construction of plasmid vectors for insertion of a traceable kanamycin resistance marker in Gordonia sp. KTR9 pGKT2. (a) Original construct harbouring the Kmr marker, flanked by regions of homology to pGKT2 and with BamHI ends. (b) pK18mobsacB mobilizable suicide vector digested with BamHI and ligated to the BamHI-liberated Kmr fragment. (c) pGKT2::Kmr recombined via conjugal transfer from Escherichia coli Top 10 competent cells into the Gordonia sp. KTR9 host. PCR primers used to verify the each step of the process are marked on the plasmids and are listed in Table 1.

Kmr

pCR2.1 6834 bp

ct

Table 1 Primers and amplification conditions employed for screening the construction of the traceable Kmr marker and putative transconjugants Primer

Target plasmid*

Sequence (5¢–3¢)

PCR parameters§

Amplicon size

ktr9plas120565af ktr9plas121248ar ktr9plas149923bf ktr9plas150701br construct40788f construct41201r mbscBaphII5254f mbscBaphII5596r mbscBsacB3801f mbscBsacB4223r kanconsrt363f kanconsrt825r mbscBplas2187f mbscBplas2641r GntR5448f GntR5918r P4507204f P450795r xpla2239(11220)f xpla12062r

c

cgaagggcttctcacttcac gacgagacctccttgcagac tggcgaactgagaccttctt gcattctgtccttgctgtca atcgtcaggtcttcgagcatg cacctcacaccgttgaataa atactttctcggcaggagca cgcatgattgaacaagatgg aaatgtggctgaacctgacc gcgaaagaaacgaaccaaaa gcctgagcgagacgaaatac attcaacgggaaacgtcttg gctctttggcatcgtctctc ctcttcgttcgtctggaagg tgagtgacgccatcacgatc tctctgcctgaagagtgacc tcgcgccgaaactcatcaac aacgggccacgaaatgacctac gatgaccgctgctgcgtccat aggaaaccgcaggaaagac

63C, 40 s; 72C, 40 s

683

c a,b,c b b a,b,c b c c c

802 63C, 40 s; 72C, 1 min

413 ⁄ 1330

60C, 30 s; 72C, 30 s

343 423 446 455

64C, 30 s; 72C, 40 s

470

65C, 1 min; 72C, 1 min

755 842

*Target plasmid refers to the following: (a) original construct harbouring the Kmr marker, (b) pK18MobSacB mobilizable suicide vector, (c) pGKT2 with the Kmr marker. Primers are specific for pGKT2 flanking regions. The expected size for the Kmr insert within this region is 1330 bp, and without the Kmr, it is 413 bp. xplAB region of pGKT2. §Annealing; elongation conditions for PCRs.

Corynebacterium glutamicum (Jung et al. 2009), Enterobacter cloacae ATCC 13047, Enterococcus faecium ATCC 6569, E. coli BHB2600, Gordonia amarae DSMZ 43392,

G. desulfuricans DSMZ 44981, G. hirsuta DSMZ 44140, G. hydrophobica DSMZ 44015, G. polyisoprenivorans DSMZ 44302, Gordonia sp. KTC13 (Crocker et al. 2006),

Journal of Applied Microbiology 110, 1449–1459 ª 2011 The Society for Applied Microbiology No claim to US Government works

1451

Genetic transfer of RDX degradation

C.M. Jung et al.

Micrococcus luteus ATCC 4698, Mycobacterium frederikbergense FAn9, Myco. gilvum PYR-GCK ATCC 43909, Myco. smegmatis PAH-100, Myco. smegmatis MC2155, Nocardia sp. TW2 (Hanne et al. 1993), Nocardioides luteus BAFB (Jung et al. 2002), Propionibacteria acnes ATCC 6919, Proteus mirabilis ATCC 7002, Pseudomonas aeruginosa (Jung et al. 2009), Ps. putida ATCC 23974, R. jostii RHA1 (Seto et al. 1995), Salmonella typhimurium ATCC 13076, Staphylococcus aureus ATCC 25923 and Staph. epidermidis ATCC 12228.

were spread on LBP plates. Resultant colonies were replica plated (50 for each sample) on LBP and LBP with 100 lg ml)1 kanamycin. The number of generations per day (24 h) for each recipient strain was previously determined by growth curve measurements and colony plate counts. Confirmatory PCR amplification with pGKT2specific primers (Table 1, Fig. 1) was used on each of the samples incapable of growth on kanamycin to check for plasmid loss, along with a random sampling of those capable of growth, as a positive control check.

Degradation of RDX by transconjugants

Pulsed-field gel electrophoresis and Southern hybridization

)1

For RDX degradation experiments, LBP with 25 mg l RDX (approx. 115 lmol l)1) or mineral salt medium (MSM) (Thompson et al. 2005) containing 5 mmol l)1 sodium succinate, 5 mmol l)1 glucose, 10 mmol l)1 glycerol and 4 mmol l)1 NH4Cl or 25 mg l)1 RDX as the sole nitrogen source was used as a growth medium. For addition of RDX to medium, acetone from a concentrated RDX stock solution was allowed to evaporate before the addition of medium to the bottle or flask. Following 2-day incubation from a single colony at 30C, donor, transconjugant, and wild-type recipient cells were transferred into either fresh LBP with RDX or MSM devoid of any carbon or nitrogen source at an initial OD600 of 0Æ05. The MSM-grown cells were incubated for 24 h to ensure that cellular nitrogen reserves would be depleted and then transferred into MSM with the carbon sources listed above and RDX. Cultures were incubated in screw-cap test tubes containing 5 ml media and agitated on a rotating test tube rack at 28 rev min)1. Cell growth was monitored spectrophotometrically at OD600 every 24 h, and samples were taken for analysis by HPLC. All studies were performed in triplicate. The effects of alternative nitrogen sources on degradation of RDX were tested in each transformant with the final RDX-containing MSM either unsupplemented or supplemented with KNO3, KNO2 or (NH4)2SO4 (each at 4 mmol l)1). Cells were starved for 24 h as in the growth and RDX degradation study, but the initial inoculum was at an OD600 of 0Æ01. Nitrogen inhibition studies were carried out in triplicate, and cell growth and RDX degradation were monitored as reported above. Plasmid stability Plasmid stability in transconjugants was assessed following the method of De Gelder et al. (2007). Briefly, three colonies were separately inoculated from LBP agar with 100 lg ml)1 kanamycin into 5 ml LBP broth and incubated at 30C for 24 h. The cultures were transferred daily (60 ll into 6 ml LBP), and appropriate dilutions 1452

The Bio-Rad CHEF DRII system with the CHEF Bacterial Genomic DNA Plug Kit (Bio-Rad, Hercules, CA, USA) was employed for pulsed-field gel electrophoresis (PFGE). Cultures of Gordonia sp. KTR9, transconjugants, and wild-type recipients were grown in LBP medium and harvested from late exponential phase. Cell plugs were prepared from approx. 1 · 108 cells per plug in accordance with the manufacturer’s instructions, with the addition of 180 U of mutanolysin (Sigma Aldrich, St Louis, MO, USA) to increase cell lysis. G. polyisoprenivorans was resistant to lysis so an alternative lysis method was employed in which G. polyisoprenivorans was incubated in 30 ml LBP for 48 h at 30C, washed and resuspended in fresh LBP with 20 lg ml)1 ampicillin and incubated for an additional 24 h (Liu et al. 1996). The culture was then washed with resuspension buffer, pelleted, frozen at )70C for 5 min, thawed at 60C and resuspended in 0Æ5-ml resuspension buffer. Lysostaphin (100 U) was also added to the lysis buffer to facilitate cell lysis. Electrophoresis parameters consisted of 0Æ5 · TBE at 5Æ5 v cm)1 with a switch time of 40–90 s at a 120 angle for 21 h at 14C. The lambda ladder marker (Bio-Rad) was used as a size standard. A 683-bp probe specific to pGKT2 was generated by PCR amplification with primers KTR9plas120505aF and KTR9plas121248aR (Table 1), and hybridization of the PFGE and Southern analysis was performed as specified in Indest et al. (2010). HPLC analysis RDX concentrations were measured every 24 h by taking 200 ll samples that were combined with 200 ll acetonitrile and filtered through 0Æ45-lm PTFE filters (National Scientific, Rockwood, TN, USA). Sample analysis was performed with an Agilent 1100 Series HPLC (Palo Alto, CA, USA) equipped with a diode array UV absorbance detector (254 nm). An Agilent ODS-Hypersil C18 LC-18 reversed-phase column (100 · 4Æ6 mm · 5 lm) was used with a similar guard column (20 · 4Æ0 mm · 5 lm), and

Journal of Applied Microbiology 110, 1449–1459 ª 2011 The Society for Applied Microbiology No claim to US Government works

C.M. Jung et al.

Genetic transfer of RDX degradation

the system was operated as stated in Crocker et al. (2005).

(a) L 1 2 3

L 4 5 6

(b) 1 2 3

4

5 6

Results Uptake of KTR9 plasmid A kanamycin resistance marker (Kmr) was inserted, via a targeted, site-specific recombination event downstream of the xplAB gene locus to create a traceable marker with which to monitor the movement of pGKT2 from donor to recipient strains. With Gordonia sp. KTR9 containing pGKT2::Kmr as the donor strain, a variety of closely and distantly related bacteria were tested for their ability to serve as recipients in diparental matings. Of the 26 bacteria tested, G. polyisoprenivorans, R. jostii RHA1 and Nocardia sp. TW2 acquired pGKT2 by a nonselective, filter-based method of conjugation at frequencies of 5 · 10)4, 4 · 10)5 and 7 · 10)6 transconjugants per recipient, respectively. Transfer of pGKT2 was verified by PCR analysis targeting various areas of pGKT2 (Fig. 1, Table 1; data not shown). As further confirmation of the transfer, PFGE, in conjunction with Southern analysis, using a pGKT2-specific probe was conducted with the donor Gordonia sp. KTR9 strain, transconjugants, and associated wild-type strains. An expected band for pGKT2 of approx. 182 kb was observed on the Southern blots of KTR9 and the transconjugants, but was absent in all wild-type strains (Fig. 2).

194 145·5 97 48·5

(c)

Growth and RDX degradation Growth of G. polyisoprenivorans and R. jostii RHA1 transconjugants was similar in rich LBP medium containing RDX to their respective wild-type strains, while the doubling time (Dt) for the Nocardia sp. TW2 transconjugant

1

2

3

(e) L

1

2

3

4

5

6

(d)

1

2

3

4

5

6

5

6

194 145·5 97 48·5

Plasmid stability The relative stability of pGKT2 in each transconjugant was determined by first calculating the number of generations for each per day (24 h). G. polyisoprenivorans, R. jostii RHA1 and Nocardia sp. TW2 transconjugants produce 7Æ4, 8Æ3 and 5Æ5 generations per day, respectively, in LBP at 30C with shaking. The G. polyisoprenivorans and R. jostii RHA1 transconjugants showed less than a 10% plasmid loss by 100 generations. In contrast, the plasmid was stable in Nocardia sp. TW2 up to 50 generations and by 100 generations only 43% of the cells analysed still contained the pGKT2. Interestingly, each of the Nocardia sp. TW2 sample replicates behaved differently starting at 50 generations, indicated by the large error bars (Fig. 3).

L

4

5

6 (f) 1

2

3

4

194 145·5 97 48·5

Figure 2 Hybridization of a pGKT2-specific probe (ktr9plas120565afktr9plas121248ar) to uncut (lanes 1–3) and XbaI cut (lanes 4–6) Gordonia sp. KTR9 (lanes 1, 4), wild type (lanes 2, 5) and associated transconjugant (lanes 3, 6) DNA as resolved by pulsed-field gel electrophoresis (a, c, e) and by Southern analysis (b, d, f). (a, b) Gordonia polyisoprenivorans, (c, d) Rhodococcus jostii RHA1, (e, f) Nocardia sp. TW2; Lane L, Lambda ladder.

Journal of Applied Microbiology 110, 1449–1459 ª 2011 The Society for Applied Microbiology No claim to US Government works

1453

Genetic transfer of RDX degradation

C.M. Jung et al.

RDX was completely degraded by the R. jostii RHA1 transconjugant within 24 h, the KTR9 donor strain and Nocardia sp. TW2 transconjugant was within 48 h, and RDX was degraded by the G. polyisoprenivorans transconjugant within 72 h. Degradation rate and growth rate were set as a ratio, and each transconjugant was compared against the Gordonia sp. KTR9 donor strain. There was no significant difference in the RDX degradation abilities between each transconjugant and the donor (T-test, P > 0Æ01). None of the wild-type strains were capable of degrading RDX, demonstrating the inability of the wildtype strains to utilize RDX as a nitrogen source for growth (Fig. 4).

100

% Plasmid retnetion

80

60

40

20

0

0

25

50 75 Generations

Nitrogen inhibition

100

Figure 3 Percentage of transconjugant populations retaining pGKT2 over 100 generations without selective pressure. Symbols; Gordonia polyisoprenivorans (h), Rhodococcus jostii RHA1 (D), Nocardia sp. TW2(e).

was slowed to about 40% of the wild-type Dt (data not shown). All of the transconjugants grew in minimal MSM medium with RDX as the sole source of nitrogen, while the wild-type strains showed no signs of growth over 96 h (data not shown). To assess their ability to degrade and utilize RDX for growth, transconjugants and wild-type strains were grown in MSM with RDX as a sole source of nitrogen (Fig. 4). 30

RDX (mg l–1)

25 20

Transconjugants were grown in MSM with either RDX as a sole source of nitrogen or in concert with an alternative nitrogen source in the form of KNO3, KNO2 or (NH4)2SO4. The effect of inorganic nitrogen on RDX degradation for each transconjugant was variable. R. jostii RHA1 was most similar to Gordonia sp. KTR9, showing a slower rate of growth when given RDX alone as a nitrogen source, with complete degradation of RDX occurring by 48 h (Fig. 5b). After 96 h, incomplete degradation of RDX (approx. 50%) was seen in R. jostii RHA1 when NO3 and NO2 were added, and less than 10% RDX degradation occurred in the presence of NH4. Nocardia sp. TW2 showed similar trends in growth and degradation to R. jostii RHA1 and Gordonia sp. KTR9 but with complete degradation of RDX in the presence of NO3 and NO2 (Fig. 5c). Gordonia polyisoprenivorans, however, exhibited little difference in growth or RDX degradation between cultures given only RDX or RDX with NO3 or NH4 (Fig. 5a). Growth and RDX degradation by G. polyisoprenivorans were limited when NO2 was presented as an additional nitrogen source.

15

Discussion 10 5 0

0

20

40

60 Time (h)

80

100

Figure 4 Degradation of RDX in wild-type (dashed lines) and transconjugant (solid lines) Gordonia polyisoprenivorans ( h), Rhodococcus jostii RHA1 ( D), Nocardia sp. TW2 (¤ e) and the Gordonia KTR9 donor strain (s) in MSM with RDX (25 mg l)1) added as a sole nitrogen source. No degradation was seen in the wild-type strains. Sterile control (d). Error bars represent the standard error for three samples.

1454

Horizontal gene transfer (HGT), the lateral or nonheritable movement of genetic material between bacteria, contributes to bacterial genome plasticity, which in turn allows bacteria to rapidly adapt to new environments. There are numerous accounts of HGT among bacteria in the literature (Sørensen et al. 2005); however, most of these studies have focused primarily on dissemination of antimicrobial resistance and pathogenicity factors in a wastewater treatment plant setting. More limited are HGT studies demonstrating the transfer of genes involving degradation of xenobiotic compounds in an open environmental system such as a field site (Christensen et al. 1998; Top et al. 1998; Dejonghe et al. 2000; Aspray

Journal of Applied Microbiology 110, 1449–1459 ª 2011 The Society for Applied Microbiology No claim to US Government works

C.M. Jung et al.

30

0·100

20

0·010

10

10·000

20

60 40 Time (h)

80

40

(b)

1·000

30

0·100

20

0·010

10

0·001 0

20

40 60 Time (h)

80

10·000 (c)

OD (600 nm)

0 100

0 100

40

1·000

30

0·100

20

0·010

10

0·001

0

RDX (mg l–1)

0

RDX (mg l–1)

1·000

0·001

OD (600 nm)

40

(a)

20

40 60 Time (h)

80

0 100

RDX (mg l–1)

OD (600 nm)

10·000

Genetic transfer of RDX degradation

et al. 2005; Overhage et al. 2005; Musovic et al. 2010). Perhaps, the best studied, environmentally relevant HGT model system has been the 2,4-dichlorophenoxyacetic acid (2,4-D) herbicide-degradative plasmids that have been successfully transferred in contaminated soil systems to expedite herbicide degradation rates (see references in (Top et al. 2002)). In contrast to 2,4-D which is often found in rich, agricultural soils, RDX is generally found in military and manufacturing settings. RDX-degrading bacteria have been isolated on military installations such as China Lake, CA, which are often arid environments with high sand content and low organic carbon (Crocker et al. 2005). Soil disruption through vehicular transport or explosives detonation on such sites may be important in physical relocation of bacteria. Furthermore, the bacteria associated with HGT events of agricultural chemicals and contaminants such as 2,4-D have been typically fast-growing Proteobacteria (Top et al. 2002), while the primary bacteria responsible for degradation of RDX are often the slower-growing Actinomycetes. The potential for lateral transfer of RDX-degrading genes in the Actinomycetes has been suggested based on the extreme conservation of the xplAB gene complex, the preponderance of mobile elements flanking the xplAB coding region and the overall synteny of this region between Microbacterium sp. MA1 and R. rhodochrous 11Y (Andeer et al. 2009). In this study, we demonstrate that only Actinomycete bacteria were able to take up and express the Kmr gene marker of pGKT2. Variable but low transfer frequencies were observed from Gordonia sp. KTR9 to G. polyisoprenivorans, R. jostii RHA1 and Nocardia sp. TW2 that were 5 · 10)4, 4 · 10)5 and 7 · 10)6 transconjugants per recipient, respectively. A comparison of HGT frequencies within and between Actinomycete bacteria is as varied as the physiologies of the organisms themselves (Desomer et al. 1988; Bro¨ker et al. 2008; Doroghazi and Buckley 2010). For example, Doroghazi and Buckley (2010) found that many of the Streptomyces species are extremely promiscuous (frequencies of 10)2 to 10)3 transconjugants per recipient) within and between species and at transfer frequencies far exceeding those found in our study. While another study showed that frequencies of R. fascians intraspecies transfer of the same plasmid varied from 10)2 to