Dechlorination of Chloral Hydrate Is Influenced by the Biofilm Adhesin Protein LapA in Pseudomonas putida LF54

Abstract

LapA is the largest surface adhesion protein of Pseudomonas putida that initiates biofilm formation. Here, by using transposon insertion mutagenesis and a conditional lapA mutant, we demonstrate for the first time that LapA influences chloral hydrate (CH) dechlorination in P. putida LF54.

TEXT

Chloral hydrate (CH) is a potent genotoxic and carcinogenic compound that has adverse effects on health (1, 2). Although bacterial degradation of CH has been known for several decades, the mechanism of CH degradation in bacteria is not well understood. We previously isolated Pseudomonas sp. LF54 (which was renamed Pseudomonas putida LF54; see the supplemental material), the first bacterium found to use CH as the sole carbon source via an assimilation pathway (3). This strain transforms CH to 2,2,2-trichloroethanol (TCAol), which is dechlorinated to 2,2-dichloroethanol (DCAol) with CO₂ as the end product (Fig. 1).

Fig 1.

Proposed chloral hydrate biodegradation pathway in Pseudomonas putida LF54. CH, chloral hydrate; TCAol, trichloroethanol; DCAol, dichloroethanol; MCAol, monochloroethanol.

Here, we generated transposon (Tn) mutants of LF54 to identify genes related to CH degradation. First, LF54 cells were rendered competent by a microcentrifuge-based procedure (4). Then LF54 Tn mutants were generated using the EZ-Tn5<KAN-2> Tnp transposome kit (Epicentre Biotechnologies, Madison, WI) and selected on LB agar containing 20 μg ml⁻¹ kanamycin. The insertion fragment was verified by colony PCR by using the primers EZTN-F and EZTN-R (Table 1). Next, CH dechlorination-defective Tn mutants were screened by 2 methods. For these assays, the cells were grown in LB medium (Lennox; Nacalai Tesque, Kyoto, Japan) at 30°C for 20 h, harvested by centrifugation, washed twice with chloride-free MS medium (3), and resuspended in a 25-ml serum bottle containing 5 ml of MS medium at the indicated cell densities. Next, 1 mM CH was added and the cultures were incubated at 30°C for 18 h with shaking at 150 rpm.

Table 1.

Strains, plasmids, and primers used in the study

Strain, plasmid, or primer	Relevant characteristics or sequencea	Source, reference, and/or PCR program
Strains
LF54	Wild-type Pseudomonas sp. LF54; CH-degrading bacterium	3
PpY101	Wild-type P. putida PpY101; met; Nalr	7
F1	Wild-type P. putida F1; toluene-degrading bacterium	6
Tn-mus01	LF54 lapA inactivated by transposon; Kanr	This study
Tn-mus03	LF54 lapA inactivated by transposon; Kanr	This study
Tn-mus04	LF54 lapA inactivated by transposon; Kanr	This study
Tn-mus11	LF54 lapA inactivated by transposon; Kanr	This study
Tn-mus13	LF54 lapA inactivated by transposon; Kanr	This study
Tn-mus14	LF54 lapA inactivated by transposon; Kanr	This study
Tn-mus15	LF54 lapA inactivated by transposon; Kanr	This study
LF54-lapA	Conditional lapA mutant in LF54; lapA inactivated by a single-crossover knockout with pSC200-lapA; Genr	This study
E. coli S17-1 λ pir	recA thi pro ΔhsdR M+ RP4-2-Tc::Mu-Km::Tn7 λpir	15
Plasmids
pSC200	Broad-host-range vector; Genr	13, 14
pSC200-lapA	lapA fragment inserted into pSC200 for single-crossover knockouts; Genr	This study
Primers
EZTN-F	5′-TCTTGCTCGAGGCCGCG-3′	This study
EZTN-R	5′-TTGCATGCCTGCAGGTCG-3′	95°C for 3 min; 95°C for 45 s, 60°C for 45 s, and 72°C for 1.5 min (28 cycles); 72°C for 10 min
KAN-2 FP-1	5′-ACCTACAACAAAGCTCTCATCAACC-3′	Epicentre Biotechnologies; 8
KAN-2 RP-1	5′-GCAATGTAACATCAGAGATTTTGAG-3′	95°C for 4 min; 95°C for 30 s and 60°C for 4 min (60 cycles)
lapA-F1	5′-GGAATTCCATATGATGAGCAGCGTTGTAGCCA-3′	This study
lapA-R1	5′-GGGGTACCTTATCGAGTACGCCCGCAAA-3′	95°C for 3 min; 95°C for 45 s, 56°C for 45 s, and 72°C for 1.5 min (28 cycles); 72°C for 10 min
lapA-upstream-F	5′-TATCCAGCAGGGGATCGTCA-3′	This study
lapA-R2	5′-TTATCGAGTACGCCCGCAAA-3′	95°C for 3 min; 95°C for 45 s, 58°C for 45 s, and 72°C for 1.5 min (28 cycles); 72°C for 10 min
pSC200-F	5′-CGATAGGGCGTCTGCATCC-3′	This study
lapA-R3	5′-CGATAGGGCGTCTGCATCC-3′	95°C for 3 min; 95°C for 45 s, 60°C for 45 s, and 72°C for 1.5 min (28 cycles); 72°C for 10 min

^aRestriction enzyme sites introduced for the subsequent cloning of DNA fragments are underlined. met, methionine auxotroph; Nal^r, nalidixic acid resistant; Kan^r, kanamycin resistant; Gen^r, gentamicin resistant.

For the first screening, a previously described chloride ion colorimetric method (5) was used. The concentration of resuspended cells was 4 × 10⁹ cells ml⁻¹. A noncell control and wild-type (wt) LF54 were monitored as negative and positive controls, respectively. Approximately 100 Tn mutants released fewer chloride ions than wt LF54.

For the second screening, a previously described capillary gas chromatography (GC) method (3) was used. The concentration of resuspended cells was 2.5 × 10¹⁰ cells ml⁻¹. Culture samples were centrifuged, and then the supernatants were concentrated 10-fold by extraction with t-butyl methyl ether. 1,2,3-Trichloropropane (Wako, Japan) was used as an internal standard. P. putida F1 (6) and P. putida PpY101 (7) were also monitored as references. Seven Tn mutants produced observably less DCAol than did LF54, and the levels of DCAol produced by all 7 of these mutants were similar. A representative mutant, Tn-mus13, is shown in Fig. 2. Strain LF54 transformed all the CH into TCAol after 18 h and dechlorinated approximately 25% of the TCAol into DCAol. F1 and PpY101 transformed CH into TCAol but did not subsequently dechlorinate it. However, DCAol levels in the Tn mutants were obviously lower than those in LF54 and were closer to those in F1 and PpY101, a finding which indicates that the TCAol transformation step in CH degradation was inhibited.

Fig 2.

Chloral hydrate (CH) degradation. (A) CH concentration. (B) TCAol concentration. (C) DCAol concentration. The samples were incubated at 30°C and shaken at 150 rpm for 18 h. The initial concentration of CH was 1 mM. P. putida F1, PpY101, LF54-lapA with glucose, and MS medium from a noncell control with CH were used as negative controls. LF54 was used as the positive control.

The transposon insertion sites were sequenced bidirectionally using the sequencing primers KAN-2 FP-1 and KAN-2 RP-1 (Table 1) (8) and analyzed using an ABI Prism 3130 genetic analyzer (Life Technologies, Foster City, CA). Genomic transposition sites were located using the NCBI BLASTN program (http://blast.ncbi.nlm.nih.gov/). Using primers from each end of the transposon, nearly 1 kb of sequence from each Tn mutant was assembled. All 7 sequences flanking the transposon insertion sites (GenBank accession numbers KC686681 to KC686687) mapped to LapA in P. putida KT2440 (9, 10).

The lap genes are conserved among environmental pseudomonads, such as P. putida and Pseudomonas fluorescens, but are absent from pathogenic pseudomonads, such as Pseudomonas aeruginosa and Pseudomonas syringae (9, 11). In P. putida KT2440, LapA is one of the largest proteins, at 8,682 amino acids and an estimated molecular mass of 888 kDa (9). It contains 4 domains, 2 of which (domains 2 and 3) are composed of long multiple repeats (Fig. 3). Recently, many studies have shown that LapA can initiate biofilm formation and achieve stable, “irreversible” binding of P. fluorescens and P. putida to a wide variety of surfaces (9, 11, 12). In this study, 7 sequences flanking transposon insertion sites were found to be located in the middle of lapA (a region corresponding to domain 3) (Fig. 3, black line); these sequences showed 91 to 96% identities.

Fig 3.

LapA protein in P. putida KT2440 and P. putida LF54. In P. putida KT2440, LapA contains an N-terminal transmembrane region (domain 1), an extensive repetitive region consisting of 9 repeats of 100 amino acids (domain 2) and 29 repeats of 218 to 225 amino acids (domain 3), and several conserved motifs at the C terminus (domain 4). ↓ and ↑ indicate transposon insertion sites and directions. ↓, forward direction; ↑, backward direction. The bold line indicates lapA sequenced in the LF54 chromosome. The dark gray line indicates an upstream fragment including a part of the 5′ region of lapA; the sequence identity was 89%. The black line indicates the transposon insertion sites flanking the sequence; the sequence identities were 91 to 96%. The light gray line indicates the 5′ end sequence of the NODE_189 contig; the sequence identity was 93%. The hollow line indicates the 5′ region of lapA (461 bp), which was used for constructing the conditional lapA mutant.

The complete genome of LF54 was sequenced using a Genome Analyzer IIx (Illumina, San Diego, CA) and was found to contain 222 contigs. The total contig length was 5,632,841 bp. Because domains 2 and 3 of LapA are extensively repetitive, assembling the complete lapA gene was not possible at this stage of genome sequencing. However, Tn-mus13 mapped to the end of LapA domain 3, which was located at the 5′ end of the NODE_189 contig (294,102 bp; GenBank accession number AOUR00000000). NODE_189 included part of domain 3 and all of domain 4 (Fig. 3, light gray line); it shared 93% sequence identity to P. putida KT2440 lapA. In addition, the upstream region including part of the 5′ end of lapA (GenBank accession number KC686680) was amplified by PCR using LF54 genomic DNA as a template and primers lapA-upstream-F and lapA-R2 (Table 1). The sequence identity this fragment shared with P. putida KT2440 lapA was 89% (Fig. 3, dark gray line). Therefore, the structure of the LapA protein of LF54 appears to be very similar to that of its counterpart in KT2440.

We subsequently constructed a conditional lapA mutant of LF54 (LF54-lapA). A 5′ fragment (Fig. 3, hollow line) was amplified by PCR using the primers lapA-F1 and lapA-R1 (Table 1). This amplified fragment was digested with NdeI/KpnI and ligated into pSC200 to yield pSC200-lapA (13, 14), which was introduced into the mobilizer strain Escherichia coli S17-1 λ pir and then transferred into LF54 by conjugation (15). Mutants generated as the result of single-crossover events were selected on gentamicin (Gen) plates (10 μg ml⁻¹) to obtain a conditional lapA mutant. Correct insertion was verified by colony PCR using the primers pSC200-F and lapA-R3 (Table 1). In this mutant strain, the native lapA promoter on the chromosome was replaced with a rhamnose-inducible promoter. LapA protein expression was induced by addition of 0.1% (wt/vol) rhamnose. The CH degradation ability of LF54-lapA was confirmed by GC (Fig. 2). LF54-lapA grown in glucose was used as the negative control. In the presence of either rhamnose or glucose, LF54-lapA can transform all the CH into TCAol after 18 h. In the absence of rhamnose, CH dechlorination was inhibited in LF54-lapA, and the ability was restored after induction with rhamnose. Without induction, the CH dechlorination levels in LF54-lapA were similar to those in the glucose control and the Tn mutants that showed CH dechlorination inhibition. In addition, when CH was provided as the sole carbon and energy source, the lapA mutants (the Tn mutants and the conditional lapA mutant) were not able to grow (data not shown) whereas LF54 was (3). These results clearly demonstrate that LapA is involved in CH dechlorination.

Because LapA is reported to initiate biofilm formation in other P. fluorescens and P. putida strains (9, 11), we evaluated whether this function is conserved in LF54. Biofilm formation was measured as previously described (16). Biofilm formation in LF54-lapA and the Tn-mus13 strain was significantly lower than that in the wild-type strain (Fig. 4). The ability to initiate biofilm formation in LF54-lapA was restored when lapA was induced by rhamnose.

Fig 4.

Biofilm formation. The cultures were started at an optical density at 600 nm (OD₆₀₀) of 0.01 in glass tubes and grown in LB medium with shaking at 150 rpm for 20 h. Cells attached to the glass tubes were stained with crystal violet and were quantified by solubilizing the crystal violet with ethanol; after that, the OD₅₉₅ was measured.

Here, we demonstrate for the first time that LapA, the biofilm adhesin protein of P. putida, influences CH dechlorination. Although the lapA gene is conserved in P. putida strains (9), P. putida F1 and PpY101 were not able to dechlorinate CH (Fig. 2). Therefore, the LapA protein may have a unique function in LF54; conversely, other factors may be involved in the dechlorination process. The latter possibility is more appealing to us because LapA does not share any similarity with enzymes possessing dehalogenation functions (17). An interesting challenge for the future will be to verify the interactions between the CH dechlorination enzyme and LapA and to examine whether biofilm formation influences CH dechlorination.

Nucleotide sequence accession numbers.

The nucleotide sequence data of P. putida LF54 have been submitted to NCBI GenBank under accession numbers KC686680 to KC686687 and AOUR00000000. The version described in this paper is the first version, AOUR01000000.