Bacterial strains and culture conditions
The following strains of Comamonas testosteroni were used: DSM 38, 1455, 1622, 6781, 11414, 12678, 50241, 50242, 50244, LMG 19554, SB3  and SB4 . In addition, C. aquatica ATCC 11330, C. nitrativorans DSM 13191, C. koreensis TK17 , C. terrigena TK41 , Delftia acidovorans ATCC 15668, Delftia tsuruhatensis SB5 , Pseudomonas putida DSM 291, Pseudomonas aeruginosa PAO1 , Novosphingobium capsulatum DSM 30196, Aeromonas hydrophila DSM 30187, Escherichia coli DH5α, Serratia ficaria ATCC 33105, Acinetobacter calcoaceticus BD413 , Ralstonia eutropha DSM 531, and Burkholderia cepacia H111  were included as non-target organisms. All cultures were maintained on nutrient broth at 30°C.
Enrichment of Comamonas testosteroni from activated sludge
Activated sludge was obtained from the aeration basin of the municipal wastewater treatment plant Garching (Germany). A sample representing approximately 1 mg of dry weight was inoculated into 25 ml of MMO medium  containing 0.05% testosterone (added as a 10% solution in DMSO) as the sole carbon source and incubated for 4 d (30°C, 150 rpm), until the disappearance of the testosterone crystals indicated degradation of the compound. 250 μl of this culture was inoculated into 25 ml of fresh medium and again incubated for 2 d, and a sample was spread onto MMO plates with testosterone which were incubated for 7 d at 30°C.
Primer and probe design and evaluation
A multiple sequence alignment and calculation of a phylogenetic tree of a selection of 16S rRNA gene sequences of the genus Comamonas was done using ClustalX 1.81  (see Fig. 1 for the sequences included in the analysis). Sequence stretches showing identities within C. testosteroni clade A sequences, but having mismatches to the remaining sequences, were used to design species-specific oligonucleotides. The sequences of the oligonucleotides which were finally used in this study are shown in Table 1.
The specificities of the primers and probes developed in this study as well as those of the probes COM1424, CTE, and PPT were checked using BLAST within the NCBI website  as well as the ProbeMatch tool in the RDP-II .
For PCR from pure cultures, cells suspended in water were used as template DNA. DNA from activated sludge and the enrichment cultures was isolated as described previously .
Conventional PCR was carried out using a Mastercycler gradient (Eppendorf, Hamburg, Germany) and Qiagen HotStarTay DNA Polymerase (Qiagen, Hilden, Germany). A single reaction contained 0.2 μM of each primer CteA1-for and CteA1-rev, 0.2 mM of each dNTP, 1.5 mM MgCl2, 0.02 U/μl of Taq Polymerase in a volume of 20 μl. Either 1 μl of cell suspension or 10 ng of DNA was added as template DNA. The PCR protocol consisted of an initial denaturation of 15 min at 95°C followed by 30 cycles of 30 sec at 95°C, 1 min at 58°C and 2 min at 72°C and was concluded by a final extension of 8 min at 72°C. To exclude the possibility of PCR inhibition, parallel positive control reactions using the general eubacterial primers 341F and 534R were conducted for each template. The 16S rRNA genes of isolates were partially amplified using primers 27F and 1492R  and custom sequenced using primer 341F by MWG Biotech (Ebersberg, Germany). PCR products were analysed on 2% TAE agarose gels.
Real-time PCR was carried out in an ABI GeneAmp 5700 device (Applied Biosystems, Foster City, USA) using the Qiagen Quantitect SYBR Green PCR Kit. Each reaction was performed in triplicate and contained 12.5 μl of PCR-Mastermix, 0.2 μM of either each primer CteA2-for and CteA2-rev or 341f and 534r, 5 mM MgCl2, and 5 μl of template DNA in a final volume of 25 μl. The thermal protocol consisted of 15 min at 95°C followed by 40 cycles of 15 sec at 95°C and 60 sec at 60°C. The same serial 10fold dilutions of genomic DNA of Comamonas testosteroni DSM 50244T were used to generate standard curves for both primer pairs.
FISH and image analysis
FISH was performed as described , using paraformaldehyde as a fixative. For elucidation of the hybridization conditions for the probe CteA, hybridization temperature was kept constant at 46°C. The formamide concentration in the hybridization buffer was varied by 10% increments. C. testosteroni LMG 19554 and C. aquatica ATCC 11330 were used as positive and negative controls, respectively. The formamide concentration which resulted in clear strong signals for C. testosteroni and no signals for C. aquatica was used in subsequent hybridizations. Probe CteA was labelled with the cyanine dye CY3, whereas the EUB probe mix (an equimolar mixture of probes EUB338, EUB338-II, and EUB338-III) was labelled with the cyanine dye CY5. All hybridizations were performed as simultaneous dual colour hybridizations where the EUB mix served the role of a general counter stain for total bacterial cell number. Images were recorded using a Zeiss LSM 510 Meta (Carl Zeiss, Jena, Germany) confocal laser scanning microscope. For quantitative analyses, the percentages of area coverage of signals from the CteA probe and the EUB mix were calculated using the Quantimet Q500W (Leica, Cambridge, England) image analysis system. The abundance of C. testosteroni was calculated as the ratio of the area covered by biomass stained simultaneously with both CteA and EUB probes to the area covered by EUB-stained biomass only. For each sample, 20 microscopic fields (92 × 92 μm) were analysed.
The 16S rRNA sequences determined in this study have been deposited in the EMBL nucleotide sequence database  under the accession numbers [EMBL:AM113739-AM113745].