Why two are not enough: degradation of p-toluenesulfonate by a bacterial community from a pristine site in Moorea, French Polynesia


  • Tewes Tralau,

    1. Department of Product Safety, Federal Institute for Risk Assessment, Berlin, Germany
    2. Department of Biology, University of Konstanz, Konstanz, Germany
    3. Manchester Interdisciplinary Biocentre, University of Manchester, Manchester, UK
    Search for more papers by this author
  • Eun Chan Yang,

    1. Dunstaffnage Marine Laboratory, Scottish Association for Marine Science, Scottish Marine Institute, Argyll, Scotland, UK
    Search for more papers by this author
  • Carola Tralau,

    1. Department of Biology, University of Konstanz, Konstanz, Germany
    2. Manchester Interdisciplinary Biocentre, University of Manchester, Manchester, UK
    Search for more papers by this author
  • Alasdair M. Cook,

    1. Department of Biology, University of Konstanz, Konstanz, Germany
    Search for more papers by this author
  • Frithjof C. Küpper

    1. Department of Biology, University of Konstanz, Konstanz, Germany
    2. Dunstaffnage Marine Laboratory, Scottish Association for Marine Science, Scottish Marine Institute, Argyll, Scotland, UK
    Search for more papers by this author

  • Editor: Paolina Garbeva

Correspondence: Frithjof C. Küpper, Dunstaffnage Marine Laboratory, Scottish Association for Marine Science, Scottish Marine Institute, Oban, Argyll PA37 1QA, Scotland, UK. Tel.: +44 1631 559 216; fax: +44 1631 559 001; frithjof.kuepper@sams.ac.uk


In previous work, only one culture (strain TA12) from a pristine site was reported to utilize the xenobiotic compound p-toluenesulfonate (TSA) as a sole source of carbon and energy for aerobic growth. ‘Strain TA12’ has now been recognized as a community of three bacteria: Achromobacter xylosoxidans TA12-A, Ensifer adhaerens TA12-B and Pseudomonas nitroreducens TA12-C. Achromobacter xylosoxidans TA12-A and E. adhaerens TA12-B were identified as the TSA degraders. These two organisms contain several tsa genes from the Tntsa cluster described previously in Comamonas testosteroni T-2 and use the tsa pathway. Apparently, due to vitamin auxotrophy, the growth of the pure cultures with TSA was markedly slower than the growth of the community with TSA. The third bacterium (P. nitroreducens) TA12-C is, then, a provider of essential vitamins for the TSA degraders and occurs at a low frequency.


p-Toluenesulfonate (TSA) (Fig. 1a) is a xenobiotic arylsulfonate that is widely used in industry and that is found in seepage from landfills (Riediker et al., 2000). Biodegradation of TSA has been explored as a sole source of carbon and energy for bacteria for over 60 years (e.g. Czekalowski & Skarzynski, 1948), and three different pathways have been discovered (Focht & Williams, 1970; Locher et al., 1989; Junker et al., 1994), the best characterized of which is the tsa system in Comamonas testosteroni T-2 (Fig. 1b) (Cook et al., 1999; Providenti et al., 2001; Tralau et al., 2001, 2003a, b; Mampel et al., 2004; Monferrer et al., 2010). The overall objective of this project was not only to elucidate the enzymatic reactions involved in TSA degradation but also to evaluate their evolutionary origin and potential ecological significance in natural environments. Earlier work showed the world-wide occurrence of TSA degradation, including the tsa operon, but, with one exception, all isolates were from contaminated sites, for example sewage works: the exception is ‘strain TA12’ from Moorea, an island neighboring Tahiti, French Polynesia (Tralau et al., 2001) – none of the other samples from pristine sites elsewhere in Moorea, in the coastal and marine environments (with varying human impact) of Roscoff (Brittany, France), Carna and Mace Head Co. (Galway, Ireland), Aspropyrgos (Greece) or in the pristine peat bog of Murnauer Moos (Bavaria, Germany) yielded any isolates growing on TSA (Tralau et al., 2001).

Figure 1.

 (a) Structure of TSA. (b) Degradation of TSA and TCA by the original isolate of Comamonas testosteroni T-2, which involves four regulons (R1–R4), in a scheme modified from Tralau et al. (2003a, b). Degradation of TSA and its analogue TCA by the original isolate of Comamonas testosteroni T-2, which involves four regulons. Regulons 1 and 4 are essential for the degradation of both substrates while regulon 2 and regulon 3 are specific for TSA or TCA, respectively. Reactions catalyzed by enzymes that are encoded by chromosomal genes are framed. R1 and R3 are plasmid encoded. A sketch of the tsa transposon is shown underneath the pathway. Arrowheads indicate the direction of transcription. p-sulfobenzylalcohol (SOL), p-sulfobenzaldehyde (SYD); p-sulfobenzoate (PSB), protocatechuate (PCA), 4-carboxy-2-hydroxymuconate semialdehyde (CHS), carboxybenzylalcohol (COL), carboxybenzaldehyde (CYD), terephthalate (TER), 1,2-dihydroxy-3,5-cyclohexadien-1,4-dicarboxylic acid (DCD). Enzymes: Toluenesulfonate methylmonooxygenase (TsaMB), p-sulfobenzylalcohol dehydrogenase (TsaC), p-sulfobenzaldehyde dehydrogenase (TsaD), p-sulfobenzoate-3,4-dioxygenase (PsbA(C)), protocatechuate-4,5-oxygenase (PmdAB), terephthalate dioxygenase (TphA23), 1,2-dihydroxy-3,5-cyclohexadiene-1,4-dicarboxylic acid dehydrogenase (TphB) (Saller et al., 1995; Wang et al., 1995; Cook et al., 1999; Tralau et al., 2003a; Mampel et al., 2004).

Preliminary analyses of the genomes of C. testosteroni KF1 and Delftia acidovorans SPH1, together with Integrated Microbial Genomes software (http://img.jgi.doe.gov/cgi-bin/pub/main.cgi), indicate the widespread nature of regulons R2 (Wang et al., 1995, J. Ruff & A.M. Cook, unpublished data) and R4 (Providenti et al., 2001) of the tsa system in Fig. 1b (D. Schleheck & A.M. Cook, unpublished data). Furthermore, an analogue of TSA, p-toluenecarboxylate (TCA), can be considered to occur naturally in turpentine (Cahours, 1850), and the initial reaction steps in the degradation of TCA involve the same enzymes required for TSA (Junker et al., 1996).

At the onset of this study, considerable uncertainty prevailed as to the identity of isolate ‘TA12.’ In order to clarify its taxonomic affiliation and the TSA-degrading pathway of this culture unambiguously, we conducted a combination of reisolations, growth and biochemical experiments as well as sequencing of 16S rRNA genes.

Materials and methods

Bacteria and growth conditions

‘Strain TA12’ was obtained in earlier work, and the same complex or carbon-limited salt media were used here (Tralau et al., 2001). Isolated organisms were grown at least as six biological replicates at 28 °C in 150-μL cultures in 300-μL wells of 96-well plates in a plate reader (Synergy HT, Biotek), and all measurements were performed as 10-fold technical replicates. The results of the growth curves were found to be statistically significant within a confidence interval of at least 3%. Carbon sources were provided as 6 mM [succinate, TSA, TCA, p-sulfobenzoate (PSB) and terephthalate (TER)] or 3 mM [protocatechuate (PCA)]. Vitamin supplements for minimal media were prepared as described elsewhere (Pfennig, 1978). Soy broth was purchased from Sigma-Aldrich.

Chemical analyses

The fatty acid composition of cell membranes was determined under contract by the German Collection of Microorganisms (DSMZ).

Genetic methods and software used

16S-rRNA gene sequences of about 1500 bp were generated following standard procedures. PCR and sequencing were performed using bacterial 16S primers, 27f and 1492r (27f: AGR GTT TGA TCM TGG CTC AG; 1492r: CGG KTA CCT TGT TAC GAC TT) (Weisburg et al., 1991). PCR amplification was performed using the Taq PCR Master Mix Kit (Qiagen) in a total volume of 50 μL, containing 0.5 U μL−1 of Taq DNA polymerase, 1 × Qiagen PCR buffer, 1.5 mM MgCl2 and 200 μM of each dNTP, 10 μM of each primer and 1–10 ng of template DNA. PCR was carried out with an initial denaturation at 95 °C for 3 min, followed by 30 cycles of amplification (denaturation at 95 °C for 1 min, annealing at 55 °C for 2 min and extension at 68 °C for 3 min), with no final extension. Amplified DNA was purified using the QIAquick PCR Purification Kit (Qiagen) and sent to the NERC Biomolecular Analysis Facility (Edinburgh) for sequencing. The alignment of each gene sequence was refined by eye using se-al v2.0a11 (Sequencing Alignment Editor Version 2.0 alpha 11; software available at http://tree.bio.ed.ac.uk/software/seal/).

The primer pairs and conditions for PCR mapping of tsa genes were described elsewhere (Tralau et al., 2001, 2003a, b; Mampel et al., 2004). Sequence data were analyzed using standard software (chromas from Technelysium and dna-star package from Lasergene).

Results and discussion

Separation of ‘strain TA12’ into three pure cultures

Under the light microscope, cultures of ‘strain TA12’ appeared to be homogeneous, motile rods (2 μm long and 1 μm wide). Selective plates (TSA-salts medium) supported the growth of small (<0.5 mm), homogeneous colonies whose surface was opalescent ochre. However, attempts to sequence the 16S rRNA gene led to reads of poor quality, and the analyses of fatty acids indicated no logical identification; hence, a mixed culture was suspected. Colonies picked from one passage on complex medium required a month to grow to the stationary phase in selective medium, and colonies picked after several passages on complex agar no longer grew in selective medium. Moreover, long incubation on plates of complex medium yielded colonies with different types of edges. Nonetheless, sequencing still indicated mixed cultures. With the assistance of the DSMZ, these mixtures were separated by alternate cultivation on soy agar and in selective medium supplemented with a vitamin solution. As colonies on soy agar looked alike, colonies were picked at random and transferred to a full soy broth and selective medium with vitamins, respectively. Picked colonies were repeatedly subjected to partial 16S sequencing for identification. This resulted in three pure isolates, two of which grew in TSA minimal medium supplemented with a vitamin solution and one that occurred at a low frequency (<5%) and that grew only in soy broth. The latter isolate had a slightly smaller colony phenotype, while the two TSA-degrading organisms appeared indistinguishable.

Analysis of the fatty acid composition of the two TSA-degrading organisms (Table 1) did not result in an identification. On hindsight, it became clear that E. adhaerens and Achromobacter xylosoxidans lacked reference entries in the corresponding MIDI database (version 5.0).

Table 1.   Fatty acid composition of Achromobacter xylosoxidans TA12-A and Ensifer adhaerens TA12-B
Fatty acidA. xylosoxidans
TA12-A (%)
E. adhaerens
TA12-B (%)
  1. Nomenclature follows the format used in the MIDI databases. ‘Methyl’ indicates a methyl group and ‘c’ the positioning of substituents in the cis position.

10:0 3OH0.123.06
11:0 iso 3OH00.09
13:0 iso00.06
12:0 2OH1.412.3
12:0 3OH0.221.64
15:1 ω6c0.130
15:0 iso00.1
14:0 3OH/16 : 1 iso 15.140.26
16:1 ω9c0.420
16:1 ω7c/15 iso 2OH13.7812.94
15:0 iso 2OH/16 : 1ω7c05.97
16:1 ω5c0.180
17:0 iso00.21
17:0 cyclo19.911.97
16:1 2OH00.22
16:0 2OH00.47
16:0 3OH0.290
18:1 ω9c0.850
18:1 ω7c21.8617.62
18:1 ω6c0.1910.72
11 methyl 18:1 ω7c0.130
19:0 iso0.180
19:0 cyclo ω8c0.861.38
18:0 3OH0.210

The sequence of the complete 16S-rRNA gene of the isolated TSA-degrading organisms shared 99.0% and 99.6% identity with those of the type strains of the betaproteobacterium A. xylosoxidans DSM 10346 (Y14908) and the alphaproteobacterium E. adhaerens LMG 9954 (AM181735), respectively. The 16S-rRNA of the third strain had a 99% sequence identity with the type strain of the gammaproteobacterium P. nitroreducens DSM 14339 (AM088474). This organism was found to accelerate the growth of E. adhaerens on TSA alone as well as in combination with A. xylosoxidans (Table 2).

Table 2.   Growth rate μ (h−1) of the mixed cultures with different substrates in a medium without supplemented vitamins
  1. Data for the combination Achromobacter xylosoxidans with Pseudomonas nitroreducens are not shown because this combination never showed any growth on TSA. Lysogeny broth (LB) was used for comparing growth on a nonlimiting, nutritionally rich medium.

 A. xylosoxidans TA12-A and E. adhaerens TA12-BE. adhaerens TA12-B and P. nitroreducens TA12-C
TSA0.015 ± 0.0020.01 ± 0.001
PCA0.15 ± 0.0080.044 ± 0.001
LB0.3 ± 0.0140.315 ± 0.013
 A. xylosoxidans TA12-A, E. adhaerens TA12-B and P. nitroreducens TA12-C 
TSA0.033 ± 0.001 
LB0.301 ± 0.0001 

The three newly recognized organisms (based on their 16S-rRNA sequences) have been deposited with the German Culture Collection (DSMZ, Braunschweig, Germany) as A. xylosoxidans TA12-A (DSM 22913) and E. adhaerens TA12-B (DSM 23677). The TSA nondegrader, P. nitroreducens TA12-C, was also deposited (DSM 23662).

Growth of strains A. xylosoxidans TA12-A and E. adhaerens TA12-B in pure culture and as a community including P. nitroreducens TA12-C

While ‘strain TA12’ utilized TSA relatively rapidly (growth rate μ=0.09 h−1) without any additives, the growth of the pure cultures of A. xylosoxidans TA12-A and E. adhaerens TA12-B was slower and required the addition of vitamins in order to grow (Table 3). The addition of biotin was subsequently found to be sufficient to restore a slow growth (μ=0.01–0.015 h−1) of pure cultures of A. xylosoxidans TA12-A and E. adhaerens TA12-B, hence identifying it to be the most essential vitamin.

Table 3.   Growth rate μ (h−1) of Achromobacter xylosoxidans TA12-A, Ensifer adhaerens TA12-B and Pseudomonas nitroreducens TA12-C with supplemented vitamins
A. xylosoxidans
E. adhaerens
P. nitroreducens
  1. Again, lysogeny broth (LB) was used for comparing growth on a nonlimiting, nutritionally rich medium.

TSA0.033 ± 0.0020.082 ± 0.005No growth
PSB0.035 ± 0.0010.089 ± 0.007No growth
PCA0.069 ± 0.0040.12 ± 0.0020.14 ± 0.006
TCA0.008 ± 0.002No growthNo growth
TER0.036 ± 0.0010.078 ± 0.004No growth
Succinate0.12 ± 0.0020.12 ± 0.0020.11 ± 0.01
LB0.18 ± 0.0140.15 ± 0.0130.19 ± 0.011

Defined mixed cultures of E. adhaerens TA12-B with A. xylosoxidans TA12-A and E. adhaerens TA12-B with P. nitroreducens TA12-C were able to grow on TSA without the addition of vitamins, but growth remained slow. This shows a partial vitamin auxotrophy of the two TSA degraders. Growth rates in the absence of vitamin supplement could be increased up to threefold (μ=0.033 h−1) by cultivating all three pure strains as a mixture (Table 2). The results show that A. xylosoxidans TA12-A and E. adhaerens TA12-B can complement each other with regard to auxotrophy for vitamins and identified biotin as the lacking essential vitamin. However, a notable increase in the growth rate requires the presence of all three strains, indicating that P. nitroreducens TA12-C complements a supply of limiting vitamins. The corresponding mixed cultures were started with equal amounts of all three strains. Doubling the amount of P. nitroreducens TA12-C at the time of inoculation resulted in a slightly reduced growth rate on TSA (0.030 ± 0.001 h−1) and plating of the corresponding late exponential culture showed P. nitroreducens TA12-C to occur in a similar low frequency (<5%) as observed in the original isolate TA12. These results clearly indicate that the TSA degraders have multiple vitamin deficiencies and that the addition of P. nitroreducens TA12-C alone is not sufficient to alleviate the deficit. This is supported by the fact that the combination of A. xylosoxidans TA12-A with P. nitroreducens TA12-C fails to produce growth, while A. xylosoxidans TA12-A will grow readily on TSA in the presence of supplemented vitamins or E. adhaerens TA12-B (Tables 2 and 3). The phenomenon of transient excretion of p-sulfobenzylalcohol (SOL) and p-sulfobenzoate (PSB), known in, for example C. testosteroni T-2 (Junker, 1996), was shown to occur for ‘strain TA12’ (Tralau et al., 2001) and the quantitative recovery of the sulfonate moiety as sulfate was obtained, which indicates the metabolism of TSA via the gene products of the tsa operon. The P. nitroreducens TA12-C does not utilize TSA or any of the excreted PSB. Thus, it is unclear whether this organism benefits from cross-feeding of vitamins or whether metabolites from aromatic metabolism (e.g. PCA, Table 2) are being cross-fed, as observed elsewhere (Feigel & Knackmuss, 1993; Pelz et al., 1999).

The tsa genes in A. xylosoxidans TA12-A and E. adhaerens TA12-B

The substrate utilization patterns of the original mixed culture TA12 (Tralau et al., 2001) shared the growth substrates TSA and p-sulfobenzoate (Fig. 1b); thus, some tsa genes were predicted in both strains. PCR mapping in each organism indicated that E. adhaerens TA12-B contained tsaMBCD2, tsaSR and tsaMBCD1. Transporter tsaT could not be detected directly, indicating a modified tsaT gene in between the duplicated tsa operon. In contrast, A. xylosoxidans TA12-A contained only the cluster tsaTSRMBCD (Fig. 2). Partial sequencing of tsaM in each strain yielded identical sequences for both organisms, corresponding to the active TsaM encoded in C. testosteroni T-2 (Tralau et al., 2001): tsaMBCD2 are not transcribed in strain C. testosteroni T-2 (Tralau et al., 2001); thus, their absence in strain A. xylosoxidans TA12-A should not be a disadvantage. No tsaQ, which encodes a regulator in C. testosteroni T-2 (Tralau et al., 2003a, b), was detected in E. adhaerens TA12-B or in A. xylosoxidans TA12-A. In C. testosteroni T-2, TSA is transported into the cell using the gene products of tsaST (Mampel et al., 2004). The apparent absence of tsaT from E. adhaerens TA12-B indicates the outer membrane pore of the TSA transporter to be replaceable.

Figure 2.

 PCR-analysis of the tsa genes in Achromobacte xylosoxidans TA12-A and Ensifer adhaerens TA12-B. The sketch shows Tntsa in Comamonas testosteroni T-2; the bars indicate the corresponding regions detected in A. xylososixidans TA12-A and E. adhaerens TA12-B. (a) Duplication of tsaMBCD. Because of the reverse orientation of the duplicated tsa operon, any primer binding to the reverse strand of tsaMBCD will amplify the corresponding sequence in between both operons. A duplication of tsaMBCD was only detected in E. adhaerens TA12-B as indicated by the 6-kb PCR fragment reaching from tsaB1 to tsaB2 (gel and bracket above the sketch). (b) Mapping of the tsa genes. Stretches of the tsa locus in both isolates were amplified using primers for the first and second half, respectively (brackets below the sketch). Comamonas testosteroni T-2 was used as a positive control. The tsa operon including its regulator tsaR and putative secondary transporter tsaS was detected in both isolates (gel below the sketch). The gene for the outer membrane porin (tsaT) could only be detected in Achromobacte xylosoxidans TA12-A and the regulatory genes for the transport of TSA (tsaQ1 and tsaQ2) were absent in both isolates (data not shown).

The degradation of TSA via the tsa operon normally involves the transient excretion of SOL, PSB and PCA, whereas TCA is degraded to TER (an analogue of PSB), which is then converted to PCA via 1,2-dihydroxy-3,5-cyclohexadien-1,4-dicarboxylic acid (DCD) (see Fig. 1b). Cultures of E. adhaerens TA12-B and A. xylosoxidans TA12-A were found to grow with PSB, TER and PCA, but only strain A. xylosoxidans TA12-A would grow with TCA and did so very slowly (μ=0.008 h−1) (Table 2). The degradation of TCA is inherently linked to the tsa operon (Fig. 1b). Thus, the lack of growth with TCA is most likely explained by a lack or a severe impairment of transport for TCA in both organisms, E. adhaerens TA12-B and A. xylosoxidans TA12-A. Moreover, the transport of TSA, PSB and TER is potentially impaired in A. xylosoxidans TA12-A as this organism grows slowly with these substrates, but faster with PCA, succinate or full broth. In C. testosteroni T-2, two regulators, TsaR and TsaQ, are known to be essential for the degradation of TSA. TsaR was found to regulate the transcription of the tsa operon and, together with TsaQ, the transcription of the transporter TsaT (Tralau et al., 2003a, b). The degradation of TSA by E. adhaerens TA12-B and A. xylosoxidans TA12-A apparently proceeds without tsaQ; hence, TSA transport must be regulated differently. Nevertheless, as a knockout of tsaQ severely impaired growth on TCA and PSB in C. testosteroni T-2 (Tralau et al., 2003a), the absence of tsaQ might well explain the difficulties of growing with PSB or TCA.

Significance of the findings

We now report that the unusual isolate from a pristine site, ‘strain TA12’, is actually a community of three bacteria, which have been identified. Two of these organisms utilize TSA, but are partially auxotrophic for the essential biotin, whereas the third partner occurs at a low frequency and provides further supply of growth rate-limiting vitamins. Thus, growth in co-culture is faster than that in a pure culture. Both Achromobacter spp. and Ensifer spp. are reported to degrade xenobiotic compounds (e.g. Song et al., 2000; Erdlenbruch et al., 2001; Hinteregger & Streichsbier, 2001), to be associated with root rhizospheres and to promote plant growth (e.g. Bertrand et al., 2000; Rogel et al., 2001). Given the natural occurrence in wood extracts of p-methyltoluene (Cahours, 1850), which is degraded via TCA (e.g. Dagley, 1971), one can speculate that the tsa genes in this pristine site represent a simple development from genes encoding TCA degradation. This notion is supported by the partial absence of the TSA transporter TsaT (in E. adhaerens TA12-B) and the lack of its regulator TsaQ in both organisms. Nevertheless, TCA failed to be a substrate for the community as well as for E. adhaerens TA12-B and was used only very slowly by A. xylosoxidans TA12-A. This is most likely due to the absence of an efficient TCA transport system, as the degradation of TCA is inherently linked to the tsa pathway and the ability to use TER. Previous studies found the tsa operon to be part of a transposon, Tntsa, allowing easy excision under stress (Tralau et al., 2001). The rapid loss of the TSA-degrading phenotype under nonselective growth conditions shows that the tsa genes of both organisms are indeed readily lost. We thus postulate that selective pressure maintains these genes at the original isolation site in French Polynesia. We further postulate that this pressure is exerted by TCA as a metabolite from the breakdown of more complex wood components within the cell.


We are grateful to Elke Lang at the DSMZ for her help and substantial input regarding the separation of the isolates and to David H. Green and Mark Hart (SAMS) for useful discussions and advice. Research was funded by the German Research Foundation, the University of Konstanz, the Boehringer Ingelheim Fonds (for a travel grant to F.C.K.), the UK Natural Environment Research Council (sequencing grant MGF-154 to F.C.K.) and the Biotechnology and Biological Sciences Research Council. We would also like to thank Laurent Meijer (CNRS, Roscoff), George R. Pettit and Robin K. Pettit (Cancer Research Institute, Arizona State University) for conducting the expedition to Moorea and for sharing soil and sediment samples.


The sequences reported in this paper for the 16S-rRNA genes of Achromobacter xylosoxidans TA12-A, Ensifer adhaerens TA12-B and Pseudomonas nitroreducens TA12-C have been deposited in the GenBank database (accession numbers HM219615, HM219616 and HM219617, respectively).