Introduction

Chloroanilines constitute a group of xenobiotics that have been in industrial use for a long time in the production of paints, pesticides, plastics, and pharmaceuticals (Meyer 1981). This group of compounds is considered as important environmental pollutants and accumulation in the long run may be detrimental to human health as a result of their persistence, toxicity, and transformation into hazardous intermediates (Pieper and Reineke 2004; Pizon et al. 2009). A number of bacterial species, capable of utilizing chloroanilines as sole source of nutrients, have been isolated from various environments such as a bioreactor (Kaminski et al. 1983), activated sludge (Boon et al. 2001; Ren et al. 2005) and soil (Vangnai and Petchkroh 2007). In bacteria, aerobic catabolic pathways for aromatic hydrocarbon degradation can schematically be divided into two major biochemical steps. First, early reactions, the so-called upper pathways or peripheral routes, channel the hydrocarbons towards the formation of partially oxidized aromatic intermediates. Then, dihydroxylated aromatic molecules that can undergo the cleavage of the ring are produced and further processed to give compounds that can enter the tricarboxylic acid (TCA) cycle. Although a wide variety of very different peripheral routes for the oxidation of many different aromatic hydrocarbons exist, only a limited number of dihydroxylated compounds that can be cleaved and productively processed to enter the TCA cycle are known. However, until now, the peripheral routes for the bacterial aerobic degradation of aniline and substituted anilines have been found only to be initiated by the reaction catalyzed by aniline dioxygenase among those bacteria, such as Frateuria species ANA-18 (Murakami et al. 2003), Delftia acidovorans strain 7N (Urata et al. 2004), D. tsuruhatensis AD9 (Liang et al. 2005) and Delftia sp. AN3 (Zhang et al. 2008); and never by any other peripheral routes such as hydroxylation, which has normally been found in the degradation of phenol and toluene (Divari et al. 2003; Griva et al. 2003). The reactions for oxygenative ring fission of catechol and the subsequent conversion to TCA intermediates are limited to one of two metabolic alternatives: i.e., (1) intradiol (ortho and modified ortho)-cleavage, and (2) extradiol (meta)-cleavage. The use of these two cleavage routes is dependent upon the microbial species and/or the nature of the growth substrate. Ortho-cleavage is the dominant cleavage mechanism in the degradation of chlorinated compounds, as extradiol cleavage of halocatechols may produce dead-end or suicide metabolites (Klecka and Gibson 1981; Surovtseva et al. 1980).

Here we report gene cloning, DNA sequencing, and partial functional analysis of the complete chloroaniline-degrading cluster (pca), from Diaphorobacter sp. PCA039; and homology analysis and enzyme assays indicate that the degradation of p-chloroaniline by strain PCA039 is initiated by hydroxylation instead of dioxygenation.

Materials and methods

Chemicals

Aniline, p-aminophenol, phenol, and p-chloroaniline were from Sinopharm Chemical Reagent Beijing Co., Ltd. Catechol and 4-chlorocatechol were from Sigma–Aldrich® Co.

Bacterial strains, plasmids, and culture conditions

The bacterial strains and plasmids used in this study are listed in Table 1. All Escherichia coli strains were grown in Luria–Bertani (LB) medium (Sambrook and Russell 2001). Diaphorobacter sp. PCA039 was able to utilize p-chloroaniline as sole carbon, nitrogen and energy source for growth and was grown at 30°C in mineral salt medium (Ren et al. 2005).

Table 1 Bacterial strains and plasmids
Full size table

DNA preparation and manipulation

Escherichia coli was grown aerobically in a Luria–Bertani medium, and chemically competent cells were transformed by heat shock. Extraction of genomic DNA was carried out following the procedures of Sambrook and Russell (2001). Recombinant plasmid DNA was isolated with a Tian-prep Mini kit (TianGen Biotech CO., Ltd). Restriction enzymes, T4 DNA ligase and calf intestinal alkaline phosphatases were from New England Biolabs (USA) or Takara (Tokyo, Japan).

Construction of genomic DNA library of strain PCA039

Genomic DNA of Diahorobacter sp. PCA039 was partially sheared with a Hamilton syringe to obtain approximately 40 kb-size fragments. The partially treated DNA fragments were end-repaired by End-repair enzyme mix of Copycontrol™ Fosmid Library Production Kit (Cat. No. CCFOS059) and then ligated to the cloning-ready Copycontrol pCC2FOS vector, and the recombinant molecules were packaged into λ phage followed by phage transfection to E. coli EPI300 strain by using protocols described in the MaxPlax™ Lambda packaging kit (Epicentre Biotechnologies, Madison, Wisconsin, USA).

Cloning of the catechol 2, 3-dioxygenase gene from strain PCA039 and library screening

DNA sequences for catechol 2, 3-dioxygenase (C23O) gene from Acidovorax sp. JS42 (accession number YP_984548) and other bacteria were used to design a pair of degenerate primers, CDf65: 5′-AACCAYGTDGCWTACAAGGT-3′ and CDr235: 5′- GHCTTGAGBACRTCRTGCCA-3′ (S=C, G; Y=C, T; B=C, G, T; H=A, C, T; R=A, G; W=A, T; D=A, G, T) for PCR amplification of a part (510 bp) of C230 gene from genomic DNA of Diahorobacter sp. PCA039, and it was confirmed by sequencing and BLAST search. To screen the positive clones, colony PCR was performed using this pair of primers. Recombinant E. coli strains with PCR-positive were further confirmed by spraying catechol solution (10 mM) onto the colonies for detection of C23O activity.

DNA sequencing and sequence analysis

DNA sequencing of the recombinant Fosmid was performed by the shotgun method by SinoGenoMax Co., Ltd (Chinese National Human Genome Center, Beijing). The ORFs were analysed using DNAstar (Lynnon Biosoft) and Vector NTI 10.0 software (Invitrogen, USA); homology search for protein sequences was carried out using the BLAST and FASTA programs (Altschul et al. 1990; Pearson 1990). The phylogenetic tree were generated using the neighbor joining method of Saitou and Nei (1987) with MEGA 4.0 software (Tamura et al. 2007), and multiple sequence alignment was done using Clustal_X (Thompson et al. 1997).

Preparation of intact cell suspension, crude enzymes and assays of enzyme activity

Recombinant strains (E. coli TOP10-S201, E. coli TOP10-S202 and E. coli TOP10-S203, Table 1) were grown to a density of A 600 about 1.0–1.5, and then the cells were harvested by centrifugation at 10,000×g for 10 min. The cell pellets were washed twice in 50 mM sodium phosphate buffer (pH 7.6) containing 1 mM 2-mercaptoethanol, and resuspended in the same buffer. Cells were disrupted sonically by twenty 4-s 200-W bursts on ice with a Braun-Sonic 1510 apparatus. Cellular debris was removed by centrifugation at 120,000×g at 4 °C for 15 min, and the supernatant was used immediately for enzyme assays.

Phenol hydroxylase (EC 1.14.13.7) (PH) activity was assayed by measuring the decrease in A 340, using NADPH as the co-substrate (Cafaro et al. 2004). One unit of enzyme activity is defined as the amount of enzyme caused the oxidation of 1 μmol of NADPH per min.

Dioxygenase activity was determined with a Clark oxygen electrode (YSI, Ohio, USA), according to the method of Fukumori and Saint (1997). To estimate the endogenous respiratory rate, 0.3 ml of sterile distilled water instead of substrate solution was used in a parallel experiment.

Catechol 2,3-dioxygenase (EC 1.13.11.2) (C23O) activity was measured according to Nakazawa and Yokota (1973) using a DU-7 spectrophotometer (Beckman, USA). The absorption coefficient of 2-hydroxymuconic semialdehyde was 12,000 l mol−1 cm−1. Enzyme specific activities are reported as μmol of 2-hydroxymuconic semialdehyde produced per minute per mg of protein.

Protein was determined according to the Bradford method with bovine serum albumin as the standard.

Transcription detection of genes in degradation of four substrates by RT-PCR

Diaphorobacter sp. PCA039 was grown on chloroaniline, aniline, phenol and 4-aminophenol for 72 h, respectively. Then the cells were harvested by centrifugation and frozen by immersion in liquid N2, respectively. Total RNA was prepared by using TRIZOL kit (Invitrogen) according to the manufacturer’s instructions, further purified by Axyprep™ Multisource total RNA miniprep kit (Axygen) according to the manufacturer’s instructions, and used as template for gene expression analysis by reverse transcriptase-PCR (RT-PCR) according to the methods of Sambrook and Russell (2001) and the primers as listed in Table 2. The RT-PCR products were then authenticated by sequencing.

Table 2 Primers for RT-PCR
Full size table

Nucleotide sequence accession number

The nucleotide sequence of the p-chloroaniline degradation gene cluster (pca) of Diaphorobacter sp. PCA039 has been deposited in the DDBJ/EMBL/GenBank under the accession number of FJ601374.

Results

Screening for C230 gene from genomic library of strain PCA039

A genomic library of strain PCA039 was constructed with more than 13, 000 recombinant strains of E. coli EPI300 harboring about 40 kb-size genomic DNA from Diaphorobacter sp. PCA039. From three of them, the specific fragment (510 bp) could be amplified, showing that they might carry the C23O gene from strain PCA039. A recombinant strain, E. coli EPI300-PCA1, contained a recombinant Fosmid, named FosB12, which carried a 28 kb DNA fragment from strain PCA039, was obtained (Table 1).

The pca cluster and the genes involved in p-chloroaniline degradation

We obtained the DNA sequence of the ecombinant Fosmid, FosB12, identified 29 ORFs in a cluster spanning ~28 kb (Fig. 1) and named it as the pca cluster. Sequence and ORF analyses of this total 28 kb fragment indicated that there were 29 intact ORFs (Fig. 1a; Table 3), and the deduced products showed significant identity (73–100%, Table 3) to functionally identified proteins from other bacteria, especially to those from Acidovorax sp. JS42 (Copeland et al. 2006), a phenol-degrader. The high identity might explain the functions of these orfs. Among them, at least 20 (pcaRKLMNOPR 1 D 1 C I U 1 R 2 D 2 C 2 U 2 EFGJI 1 I 2 ) were expected to contribute towards complete metabolism of p-chloroaniline to TCA-cycle intermediates via the meta-cleavage pathway as shown in Fig. 1b and summarized in Table 3. Homology analysis of the pca cluster revealed that it consists at least three parts as follows:

Fig. 1
figure 1

Genetic organization of pca cluster (a) and putative chloroaniline degradation pathway (c) of Diaphorobacter sp. PCA039. The sequenced region of 28.4 kb is shown. Black arrows indicate PH genes, open and grey arrows indicate meta-cleavage pathway genes. b Details of plasmid construction are provided in Table 1. The function of other genes was explained in Table 3. Abbreviations: PH Phenol hydroxylase, AD aniline dioxygenase, C23O catechol 2, 3-dioxygenase, HMSD 2-hydroxymuconic semialdehyde dehydrogenase (EC), HMSH 2-hydroxymuconic semialdehyde hydrolase (EC 3.7.1.9), OCD 4-oxalocrotonate decarboxylase (EC 4.1.1.77), OEH 2-oxopent-4-dienoate hydratase, HOA 4-hydroxy-2-oxovalerate aldolase (EC 4.1.3.39), ADA acetaldehyde dehydrogenase. + Positive, − negative, S Sau 3AI site

Full size image
Table 3 Analytical data on pca cluster and related genes from other bacteria
Full size table

Degradative genes of the pca cluster

The products of pcaKLMNOP, PcaKLMNOP, showed significant sequence homology (73–94%) with a multicomponent phenol hydroxylase (PH), present in phenol-degrading strains such as Alcaligenes faecalis IS-46 (Zhu et al. 2008) and Acidovorax sp. JS42 (Copeland et al. 2006), especially with those from Acidovorax sp. JS42. The remaining 14 gene products (PcaR1D1CIU1R2D2C2U2EFGJI1I2) exhibited considerable amino acid identity (73–100%) to enzymes of central catechol metabolism via the meta-cleavage pathway found in aniline-degrading and other aromatic-degrading bacteria. Interestingly, in the pca cluster, there were two C23O sets, PcaR1D1CIU1 and PcaR2D2C2U2. Phylogenetic analysis (data not shown) indicated that PcaC1 and PcaC2 belonged to two different branches in the phylogenetic tree, and shared only 41.9% identity and 57.5% similarity between themselves.

Regulatory genes

There are at least three regulatory genes (pcaR, pcaR 1 and pcaR 2 ) and one potential candidate orfZ in the pca cluster. The deduced product (PcaR) of pcaR is a large protein with 596 amino acid residues. PcaR shows as high as 94% sequence identity (Table 3) to a Sigma 54 specific transcriptional regulator from Acidovorax sp. JS42 (Copeland et al. 2006). A conserved domain (CD) search reveals that a region of 180 residues at the C terminus homologous to the DNA-binding domains of various regulators (Calogero et al. 1994; Canellakis et al. 1993), in addition to a region corresponding to a proteobacterial transcriptional activator domain (PAD) at the N terminus and the V4R (vinyl 4 reductase) domain predicted to bind small molecules such as hydrocarbons. Sigma 54 specific regulator is a transcriptional activator found in the aromatic-compound-degrading cluster from bacteria such as Acidovorax sp. JS42. The predicted products of pcaR 1 and pcaR 2 show 98 and 93% sequence identity, respectively, to a LysR family transcriptional regulator found in aromatic compound degradation from Comamonas sp. JS765 and other homologous activators found in a variety of Proteobacteria. It is likely that PcaR1 and PcaR2 function by regulating two C23O systems. OrfZ shows high identity (55–100%) to the GntR family transcriptional regulator found in proteobacterial strains such as Acidovorax sp. JS42 (Copeland et al. 2006) and R. eutropha JMP134 (Kim et al. 1996).

Transposase genes and genes with unknown function

Besides the degradative genes and regulatory genes, in the pca cluster, orfU1 and orfU2 are located in the two C23O systems, which is a common phenomenon in the meta-cleavage pathway, though their exact function is unknown. Another gene, orfY, encodes a putative membrane transport protein similar to aromatic aminotransferase and transmembrane transport proteins found in other bacteria such as Delftia sp. AN3 (Zhang et al. 2008). Moreover, on the up- and down-stream of the pca cluster were located the transposases and hypothesized exodeoxyribonuclease III Xth, which were encoded by pcaT and pcaQ, respectively. These gene products are believed to function in horizontal gene transfer and protection among the strains. In addition, two orfs (orfX, pcaQ) also found with high similarity to those existed in the meta degradation pathway of an aniline degrader, D. tsuruhatensis AD9 (Liang et al. 2005) and a phenol degrader, Acidovorax sp. JS42 (Copeland et al. 2006), in addition to P. putida UCC22 (Fukumori and Saint 1997). However, the functions of their products have still not been identified.

Function of the genes pcaKLMNOP

As revealed in the homologous analysis, the products of genes pcaKLMNOP exhibited significant homology (Table 3) to those of multicomponent PH found in phenol degraders such as Alcaligenes faecalis IS-46 (Zhu et al., the afp products). However, they were not homologous to those of multicomponent aniline dioxygenase (AD) commonly found in aniline and substituted aniline degraders such as D. tsuruhatensis AD9 (Liang et al. 2005), Delftia sp. AN3 (Zhang et al. 2008) and D. acidovorans strain 7N(Urata et al. 2004). This might infer that the degradation of p-chloroaniline by strain PCA039 is unusually initiated by hydroxylation, as the degradation of phenol, not initiated by dioxygenation as the degradation of aniline and substituted anilines. To date, the degradation of aniline and substituted anilines was found only to be started with the peripheral reaction catalyzed by AD (Fujii et al. 1997; Liang et al. 2005; Murakami et al. 2003; Urata et al. 2004; Zhang et al. 2008); never started with any other peripheral steps. In order to confirm the function of the genes pcaKLMNOP, they, as a whole, were subcloned into plasmid pUC118 (Table 1) and transformed into E. coli TOP10. A recombinant strain, E. coli TOP10-S201, was obtained. Then recombinant strain E. coli TOP10-S201 was used for PH and AD assays.

As shown in Figs. 1 and 2, recombinant strain E. coli TOP10-S201 did only exhibit PH activity on substrates such as aniline, phenol, 4-aminophenol and chloroaniline, especially on aniline. However, it never exhibited AD activity on any substrates (Fig. 1). This result suggested that the degradation of p-chloroaniline by Diaphorobacter sp. PCA039 was initiated by hydroxylation instead of normal dioxygenation. After the regulatory gene pcaR knock-out, the mutant strain indeed lost the ability of degrading chloroaniline (data not shown), also suggesting the products of genes pcaKLMNOP are necessary for chloroaniline degradation. This is a novel peripheral route for the degradation of chloroanilines that has never been reported before.

Fig. 2
figure 2

Phenol hydroxylase assay of recombinant strain E. coli TOP10-S201 on various substrates, as expressed by absorption at 340 nm (A 340) due to the consumption of NADPH. A 340 was recorded every 30 s and NADPH was added to all samples. control l, E. coli TOP10/PH only; control 2, E coli TOP10 + aniline; E. coli TOP10/PH + 4-aminophenol; E. coli TOP10/PH + chloroaniline; E. coli TOP10/PH + phenol; E. coli TOP10/PH + aniline

Full size image

Prelude to gene expression and transcripts analysis of pca cluster when strain PCA039 grew on different substrates.

Transcripts of several genes of the pca cluster when strain PCA039 was grown on different substrates are represented in Fig. 3. This illustrates that all of the detected genes except pcaC2 were expressed when strain PCA039 grew on aniline, phenol, 4-aminophenol and chloroaniline; however, the gene pcaC 2 was only expressed when strain PCA039 grew on chloroaniline. These results confirmed the degradation of all above four substrates to be initiated with the “hydroxylation” peripheral route instead of the “dioxygenation” peripheral route, then followed the meta-cleavage pathway in strain PCA039. Both the expression of the genes pcaF, and pcaE and pcaH means that the two degrading branches (hydrolysis and dehydrogenation) are both necessary for complete degradation of these substrates. Furthermore, the two C23O sets (pcaC 1 and pcaC 2 ) in pca cluster, pcaC 1 were expressed in the degradation of all above four substrates, while pcaC2 was only expressed in the degradation of chloroaniline. It was strongly suggested that pcaC 2 might necessarily function in chlorocatechol cleavage, in turn PcaC1 was enough for catechol metabolism. In addition, PcaC1 and PcaC2 have distinct substrate specificity, PcaC2 showed relatively high activity on chlorocatechol (Catechol, 100%; chlorocatechol, 140%), while TdnC1 showed less activity on chlorocatechol (Catechol, 100%; chlorocatechol, 7.5%). According to these results, a hypothetic pathway for the degradation of chloroaniline by strain PCA039 was proposed (Fig. 1c).

Fig. 3
figure 3

RT-PCR detection of gene expression when strain PCA039 grew on different substrates. a RNA preparations after purification (RL10000, RNA markers, Takara) from cells grown on: 1 chloroaniline, 2 aniline, 3 phenol, 4 4-aminophenol. b RT-PCR products on 1.5% agarose gel. DL2000, DNA marker (Takara). Four substrates used for growing strain PCA039 were indicated. pcaL, phenol hydroxylase subunit L; pcaC1, catechol 2, 3-dioxygenase (C1); pcaC2, catechol 2,3-dioxygense (C2); pcaE, HMSD; pcaF, HMSH; pcaH, 4-oxalocrotonate decarboxylase; and pcaJ, 4-hydroxy-2-oxovalerate aldolase gene

Full size image

Discussion

Previous studies had pointed out that the first step in the degradation of aniline and substituted anilines was always dioxygenation, never degradation by any other peripheral routes (Fukumori and Saint 1997; Liang et al. 2005; Murakami et al. 2003; Urata et al. 2004; Zeyer et al. 1985; Zhang et al. 2008). Here, it can be concluded that the degradation of p-chloroaniline by Diaphorobacter sp. PCA039 is a novel peripheral route for the metabolism of anilines. Zeyer et al. deemed that the “normal” ortho ring fission pathway widely existed in the microbial degradation of aromatic compounds, chloroanilines, benzoate, phenol (Zeyer et al. 1985). And also, this led Janke et al. (1988, 1989) to propose that complete degradation of chloroaromatics should satisfy the following qualifications, i.e., (1) it should be a low specificity oxygenase system; (2) be short of, or with a blocked meta-pathway; and (3) have a modified ortho-cleavage pathway. However, productive meta-cleavage pathways in several strains have recently been shown to exist, capable of degrading chloroaromatic compounds via meta-cleavage systems (Surovtseva et al. 1980; Arensdorf and Focht 1995; Ren et al. 2005). The different phylogenetic position and low identity of PcaC1 and PcaC2 inferred that they might have evolved from different ancestors. In addition, it was reported that TdnC and TadC1 had relatively high activity on substituted catechols (methylcatechols), while TdnC2 and TadC2 have showed less activity on these substituted catechols (Fukumori and Saint 1997; Liang et al. 2005), and the authors assumed that it might be necessary for cells to acquire another C23O for these methylcatechols to expand the assimilation range for toluidines. In contrast, in strain PCA039, PcaC1 was produced on all four substrates, while PcaC2 was only produced on chloroaniline, as revealed by RT-PCR (Fig. 3). This might be due to only chloroaniline being able to induce the expression of pcaC 2 . Furthermore, PcaC2 showed very high activity on chlorocatechol, while PcaC1 had very low activity on chlorocatechol, suggesting that both PcaC1 and PcaC2 are necessary for the vigorous degradation of chloroaniline by strain PCA039.