Dechlorination of Chloral Hydrate by Pseudomonas putida LF54 which Possesses Biofilm Adhesin Protein LapA

Because of the lack of enzymes in critical steps of catabolic pathways, low-molecular-weight halogenated compounds are often recalcitrant to biodegradation. In our previous study, we isolated Pseudomonas sp. LF54 (LF54), the first bacterium that has been shown to use chloral hydrate (CH) as sole carbon source by an assimilation pathway in which dechlorination is the critical step. In this study, we identified a transposon (Tn) mutant that can render LF54 defective in CH dechlorination. The molecular characterization of Tn mutants revealed that the transposon insertion sites map to lapA . Sequence analyses verified the existence of lapA in LF54. Additionally, induced expression of lapA in the conditional lapA mutant of LF54 further verified that defective lapA expression renders LF54 defective in dechlorination. Recent studies have revealed that the largest cell-surface-associated protein LapA, a biofilm adhesin, is able to initiate biofilm formation. This function was also verified in the induced conditional lapA mutant and in LF54. Furthermore, we also found out that the defective lapA mutant rendered the variation of bacterial motility. LapA, the largest biofilm adhesin protein of P. putida , which influences CH dechlorination and flagella motility, is a novel discovery not previously reported. Recent studies have revealed that LapA is able to initiate biofilm formation and achieve stability [24, 26, 27]. However, no study of lapA in connection with the biodegradation of CH has been reported to date. We evaluated whether this function is conserved in LF54. The cultures were started at an optical density of 600 nm (OD 600 ) of 0.01 in glass tubes. After 20 h of growth, the tubes were washed with distilled water and stained with 0.1% crystal violet (Sigma, St. Louis, MO, USA). Following the staining, the tubes were washed twice with distilled water. The tubes were subsequently dried and the crystal violet was dissolved in 96% ethanol for quantification by spectrometry at 595 nm [38]. Biofilm formation in LF54- lapA and the Tn-mus13 strain was significantly lower than that in the wild-type strain (shown in Figure 5). The ability to initiate biofilm formation in LF54- lapA was restored when lapA was induced by rhamnose. These results indicate the relationship between biofilm formation and CH degradation.


Introduction
Chloral hydrate (CH) is synthesized by the chlorination of ethanol. As a sedative and hypnotic drug, CH is used in human and veterinary medicine. Its chemical formula is C 2 H 3 Cl 3 O 2 . The anhydrous chemical, chloral, is used as an intermediate in the production of insecticides and herbicides such as dichlorvos, naled, methoxychlor, trichlorfon, and trichloroacetic acid [1].
Drinking water is the major exposure route of CH to the general public, as CH is the third byproduct formed when drinking water is disinfected with chlorine [2,3]. Therefore, environments near factories that produce CH and the above-mentioned pesticides might be contaminated. CH is irritating to the skin and mucous membranes. It often causes some symptoms even at the recommended clinical dose [1]. According to the World Health Organization (WHO) guidelines, CH has adverse effect on health and should be limited in drinking water [4]. In addition, CH was reported as a potent genotoxic and carcinogenic compound [3]. Therefore, more efforts should be made to minimize further CH release and contamination in the environment.
Studies of CH were concentrated on animal metabolism, as CH was used as a sedative and hypnotic drug. In animal, the major metabolites are trichloroethanol (TCAol), TCAol glucuronide, trichloroacetic acid (TCA), and a small amount of dichloroacetic acid (DCA) [5][6][7][8]. The major excretion pathway of CH metabolite is elimination in the urine. With regard to the bacterial degradation of CH, several cometabolism processes have been reported; methanotrophic-degrading bacteria Methylocystis sp. M and Methylosinus trichosporium OB3b can transform trichloroethylene into CH, which is further degraded to TCAol and TCA [9,10]. In our previous study, we successfully isolated Pseudomonas sp. LF54 (LF54], the first bacterium reported to date that can use CH as the sole carbon source via an assimilation pathway [11]. This strain transforms CH to TCAol, which is then dechlorinated to 2, 2-dichloroethanol (DCAol), and CO 2 was detected as the end product ( Figure 1). In general, halohydrin dehalogenation is intramolecular substitution that enzymes convert vicinal halohydrins to an epoxide, a proton and a halide ion [12][13][14]. To detoxify the compounds, halohydrin dehalogenases remove halogen substituents and put them to enter central metabolism. In view of halohydrin dehalogenases properties, the cleavage carbon-halogen bonds started with cloning and expression of three different genes that encode these enzymes [14]. However, the dechlorination in LF54 is reduction that chlorine (Cl) was substituted by hydrogen (H). This is a novel pathway of halohydrin dehalogenation has not been reported to date. Therefore, the further study of LF54 would advantage CH dechlorination and bioreme-diation. The aim of this Chapter is to identify genes related to CH degradation and to investigate mechanism of CH degradation in P. putida LF54 with the method of molecular biology. CH, chloral hydrate; TCAol, Trichloroethanol; DCAol, Dichloroethanol; MCAol, Monochloroethanol.

Identification of genes related to CH degradation
DNA transposition is an important biological phenomenon and also offers versatile tools for genetic analysis of a variety of biological processes. It is applied in creating insertional knockout mutations, providing genetic or physical landmarks for adjacent DNAs, generating the fusions of gene-operon to reporter functions, and locating primer binding sites for the analysis of DNA sequence [15]. Tn5 is one of the simplest and best characterized transposons; there are only Tn5 transposase, transposon DNA flanked by Tn5 inverted repeats, and target DNA are required in vitro transposition [16]. Then electroporation of transposon into a bacterial cell, the reaction proceeds and results in insertion of the transposon DNA at random into the genome to create gene "knockouts" which can block metabolic process.
The transposon used in this chapter was generated with EZ-Tn5 TM <KAN-2> Tnp Transposo-meTM Kit carrying a kanamycin resistance marker. The transposon can be electroporated into living cells where the transposase is activated by Mg 2+ in the host's cellular environment, resulting in random insertion of transposon into the host genomic DNA. The transposonspecific primers can be used for bidirectional DNA sequencing from genomic DNA without cloning or locating the transposon insertion sites in target genomic DNA.
In this chapter, in order to identify genes related to CH degradation, a transposon mutant library was generated and Tn mutants (CH degradation-defective) were screened by two screening steps. Transposon insertion sites were sequenced bidirectionally using special sequencing primers from the two sides of the transposon and located using BlastN programs.

Transposon mutant (Tn mutant) library
LF54 cells were made competent utilizing a microcentrifuge-based procedure [17]. The cells grown in LB mediun were distributed into microcentrifuge and harvested by centrifugation at room temperature for 2 min at 16,000 × g. The cell pellet was washed twice at room temperature and resuspended in 100 μl 300 mM sucrose, which contained average 10 9 -10 10 viable bacteria.
For electroporation, 10 ng of pSUP104 [18] and 20 ng Transposome was mixed with 100 μl electrocompetent cells, and the mixture transferred to 2 mm gap width electroporation cuvette. After applying a pulse (settings: 25 μF; 200 Ω; 2.5 kV on a Bio-Rad GenepulserXcell TM , Bio-Rad) [17], the mixture was transferred into 1 mL of room temperature SOC and shaken for 1 h at 30°C; 100 μl of the pSUP104 mixture was plated on Tc plates, and each 100 μl of Transposome mixture was plated on Kana plates. The plates were incubated at 30°C within 24 h.
Controls included cells that were pulsed with added H 2 O and were planted on both Tc and Kana plates. Mutagenized cells (Tn mutants) were selected by plating on Kana plate. The strains were stored in 20% glycerin LB media at -80°C.

Tn mutants screening strategy
Tn mutants (CH dechlorination defective) were screened by two screening steps. The cells were harvested by centrifugation, and washed twice with chloride-free MS medium and were resuspended in a 25 mL serum bottle containing 5 mL of MS medium in proper cells concentration as indicated below. After addition of 1mM CH, the cultures were incubated at 30°C and shaken at 150 rpm for 18 h.

First screening
The first screening was performed by multistep colorimetric method, as CH degradation is a multistep reaction ( Figure 1). The concentration of resuspended cells in this step is 4 × 10 9 cells mL -1 . In weak alkali conditions, 3-methyl-1-phenyl-5-pyrazolone reacts with CH and a brownish red compound is formed and in proportion to CH in 480 nm [19] measured by Varioskan Flash spectral scanning multimode reader (Thermo Fisher Scientific Inc., Waltham, Massachusetts, USA). The chloridion was mixed with a reagent consisting of Hg 2+ , SCN -, and Fe 3+ ; a new yellow complex was formed which was measured at 463 nm [20].

Second screening
The second screening was performed by GC [11]. The concentration of resuspended cells is 2.5 × 10 10 cells mL -1 . The samples were 10 times concentrated by extraction with t-butyl methyl ether (Wako, Osaka, Japan), and 1,2,3-trichloropropane (Wako) was used as an internal standard. The oven temperature was programmed as follows: 35°C (isothermal) sustained for 7 min then increased to 250°C at a rate of 20°C min -1 . The injector and detector temperatures were 270°C.

Transposon location
Transposon insertion sites were sequenced bidirectionally using sequencing primers, KAN-2 FP-1 and KAN-2 RP-1 (Table 1), specific for the ends of the inserted transposon [21]. The DNA sequencing reactions (two microgram of bacterial genomic DNA and 5 pmoles of primer) were performed using a BigDye™ Terminator Cycle Sequence Kit (Applied Biosystems, Perkin-Elmer, Foster City, CA, USA). The conditions of sequencing reactions are not shown [21]. Sequencing reactions were purified by ethanol precipitation, and resuspended in 10 μl of formamide (Wako, Osaka, Japan). After denaturing at 95°C for 2 min, all products were analyzed with an ABI PRISM ® 3130 Genetic Analyzer (Applied Biosystems).
The sequences were assembled and the genomic transposition sites were located using BlastN programs maintained at National Center for Biotechnology Information website (http:// blast.ncbi.nlm.nih.gov/).

Results
The transformation efficiency is shown in Table 2. The efficiency of pSUP104 is 1.92 × 10 -6 ; and the efficiency of transposon is 7.61 × 10 -5 . The efficiency of the transposome is almost the same as the plasmid. A library of mutants was generated using Tn mutants in LF54 and stored at -80°C.  The first screening for the degradation ability defect was performed by multistep colorimetric method. CH solution (MS medium) as the negative control and LF54 as the positive control were monitored. The results are not shown, as the data are excessive. Disappearance of CH was detected among Tn mutant library; we did not find Tn mutant, in which the process of CH transformed into TCAol was inhibited. According to chloridion release, 96 Tn mutants were picked up (the data are not shown).
The second screening was confirmed using GC; there were 7 Tn mutants that obviously produced less DCAol than LF54 (Figure 2). These strains were designated Tn-mus01, Tn-mus03, Tn-mus04, Tn-mus11, Tn-mus13, Tn-mus14, and Tn-mus15. P. putida F1 [22] and PpY101 [23] as the negative control was monitored. All the strains can transform the whole CH into TCAol after 18 h. For strain LF54, approximately 25% TCAol was dechlorinated into DCAol and part of them was converted into CO 2 as the end product [11]. F1 and PpY101 transformed CH into TCAol, but did not subsequently dechlorinate it. However, the DCAol levels in Tn mutants were obviously lower than those in LF54 and closer to F1 and PpY101.

Assay of lapA in LF54
Recently, some studies have revealed that the largest cell-surface-associated protein LapA, a biofilm adhesin, is able to initiate biofilm formation and achieve stable, "irreversible" binding to a large variety of surfaces in P. fluorescens and P. putida [24,26,27]. However, no study of lapA in connection with the biodegradation of low-molecular-weight chlorinated compounds has been reported to date. The lap genes were conserved among environmental pseudomonads such as P. putida and P. fluorescens, but absent from pathogenic pseudomonads such as P. syringae and P. aeruginosa [24,27]. In P. putida KT2440 (  In this chapter, in order to verify the existence of lapA in LF54, we sequence part of lapA in LF54. Because the entire lapA sequence of LF54 was not acquired from the whole genome sequence analysis, strains defective in lapA were assessed to define further the role of lapA in CH dechlorination. This strategy created LF54-lapA, in which the expression of lapA depended on the rhamnose concentration in the medium.

Existence of lapA in LF54
The complete genome sequence of LF54 was performed by Genome Analyzer IIx (Illumina, San Diego, CA) provided by Hokkaido System Science Co., Ltd. The genomic DNA sequence was queried the database using genomic blast programs maintained at National Center for Biotechnology Information website (http://www.ncbi.nlm.nih.gov/sutils/genom_tree.cgi). The upstream fragment includes part of 5'region of lapA, amplified by PCR with primers lapA-upstream-F, and lapA-R2 (Table 1). Primers were synthesized based on the nucleotide sequence from the Pseudomonas genome database (http://www.pseudomonas.com). The purified fragment was sequenced using a BigDye™ Terminator Cycle Sequence Kit. The product was analyzed with an ABI PRISM® 3130 Genetic Analyzer (Applied Biosystems). The sequence was used to query the GeneBank database.

Conditional lapA mutant
To construct a conditional lapA mutant in LF54 (LF54-lapA), the 5-region fragment (461 bp) and the upstream fragment including part of the 5-region of lapA amplified by PCR, using the primers lapA-F1, lapA-R1, lapA-upstream-F, and lapA-R2 (Table 1) and LF54 genomic DNA as the template were used. Then, the 5-region fragment (461 bp) was digested with NdeI/KpnI and ligated into the multiple-cloning site of pSC200 to yield pSC200-lapA [28,29].
The plasmid pSC200-lapA was introduced into the mobilizer strain E. coli S17-1 λ pir and was then transferred to LF54 by conjugation [30]. Single crossover events were selected on Gm plates to obtain the conditional lapA mutant (LF54-lapA). The correct insertions were verified by colony PCR with the primers of pSC200-F and lapA-R3 (Table 1). DNA fragments were confirmed with electrophoresis on 1.2% agarose gel.
The protein expression of lapA was inducted by 0.1% (wt/vol) rhamnose in the medium. CH degradation ability of conditional lapA mutant was affirmed by GC method mentioned.

Results
In this study, the complete genome sequence of LF54 was performed by Genome Analyzer IIx containing 222 contigs. Total contig length is 5,632,841 bp (the data are not shown). The complete genome sequence of LF54 was blasted with 34 Pseudomonas genomes databases by Genomic blast program maintained at NCBI website. The almost complete genome sequence of LF54 shows high similarity to P. putida strains, the coverage of 86% to P. putida F1 (total score = 9.643e +06 ), and 85% to P. putida KT2440 (total score = 9.132e +06 ).
As domains 2 and 3 are an extensive repetitive region, at this stage of the complete genome sequence of LF54 it has not been possible to assemble the complete lapA gene. Among the 7 CH dechlorination defective Tn mutants, Tn-mus13 mapped to the end of LapA domain 3, and was also located at 5 end of the NODE_189 contig (294,102 bp, AOUR00000000). Therefore, NODE_189 included a part of domain 3 and the entire domain 4 (Figure 3, light gray line); the sequence identity was 93% to lapA in P. putida KT2440. In addition, the sequence identity of the upstream fragment including a part of the 5 region of lapA (KC686680) was 89% ( Figure  3, dark gray line). Combining with sequence flanking the transposon insertions (corresponding to domain 3, Figure 3, black line), the structure of the protein appears to be very similar to its P. putida counterpart.
CH degradation ability of LF54-lapA was detected by GC analysis (Figure 4). LF54-lapA with glucose was used as the negative control, and LF54 was the positive control. LF54-lapA can completely transform CH into TCAol after 18 h, irrespective of the presence of rhamnose or glucose. CH dechlorination was inhibited in LF54-lapA in the absence of rhamnose, and recovered by rhamnose induction. The sample without rhamnose induction was similar to the glucose control and to the 7 Tn mutants that showed inhibited CH dechlorination.

Relationship between CH degradation and biofilm formation
In the past few years, bioremediation mechanisms by Pseudomonas for numerous halogenated pollutants have been described and many studies have reported the growth of planktonic cultures under controlled laboratory conditions [31][32][33][34]. However, bacteria are mostly found in multicellular communities known as biofilms in the environment, and not in the planktonic state [35,36]. Biofilm formation in Pseudomonas has been proposed to involve a series of regulated steps [37,26]. Firstly, utilizing flagellar-mediated motility, bacteria swim toward a certain surface and initiate reversible (or transient) attachment. Then, the attached bacteria become irreversibly attached to the surface to form small microcolonies. Finally, the microcolonies develop into a mature biofilm with an architecture separated by fluid-filled channels.
Recent studies have revealed that LapA is able to initiate biofilm formation and achieve stability [24,26,27]. However, no study of lapA in connection with the biodegradation of CH has been reported to date. We evaluated whether this function is conserved in LF54.
The cultures were started at an optical density of 600 nm (OD 600 ) of 0.01 in glass tubes. After 20 h of growth, the tubes were washed with distilled water and stained with 0.1% crystal violet (Sigma, St. Louis, MO, USA). Following the staining, the tubes were washed twice with distilled water. The tubes were subsequently dried and the crystal violet was dissolved in 96% ethanol for quantification by spectrometry at 595 nm [38].
Biofilm formation in LF54-lapA and the Tn-mus13 strain was significantly lower than that in the wild-type strain (shown in Figure 5). The ability to initiate biofilm formation in LF54-lapA was restored when lapA was induced by rhamnose. These results indicate the relationship between biofilm formation and CH degradation.
The tubes were subsequently dried and the crystal violet was dis quantification by spectrometry at 595 nm (38).
Biofilm formation in LF54-lapA and the Tn-mus13 strain was signific wild-type strain (shown in Figure 5). The ability to initiate biofilm fo restored when lapA was induced by rhamnose. These results indica biofilm formation and CH degradation.

LapA influences bacterial motility
Biofilm research has renewed our interest in bacterial motility [39]. Flagellar and twitching motility are necessary for biofilm development [40]. Biofilm formation of LF54 was verified, and in this chapter, these motilities were observed to verify the influences of LapA.
Twitching motility [41] was measured by stab-inoculating 1.0 % agar LB plates with a single bacterial colony. Plates were incubated at 37°C for 2-3days. Flagellar motility [38] including swimming and swarming motility was measured as the diameter of zone travelled by bacteria point-inoculated into 0.3% and 0.5% agar LB plates, respectively. The plates were incubated for 12 h at 30°C. CH was added in a series concentration: 0.1, 0.3, 0.5,1 mM.
Swimming and swarming motility of mutants was faster than LF54 (shown in Figure 6). Although, twitching motility of mutants was also faster than LF54, the difference was not much. Within limits the movement become slowly with the raise of concentration of the CH.  Figure 6. Swimming and swarming motility of mutant was faster than LF54. Twitching motility (pilus movement) of mutant was a little faster than LF54.
Chemotaxis is the movement of organisms toward or away from a chemical. In this study, the defective lapA in mutants of LF54 results in the increasing external press to cell. That is, the normal environment for the defective lapA mutant, it will amount to over nutrition, the mutant escaped rapidly like the negative chemotaxis. In the CH occasion, the motility appears to like the positive chemotaxis, and the speed was lower than the normal condition. Therefore, we supposed that the other reason of motility change maybe is the biofilm adhesion of mutant changed. The defective adhesion of mutant appears to makes flagellar and twitching motility easily than LF54 and results in faster motility.

Conclusion
So, for the first time we demonstrate dechlorination of CH by Pseudomonas putida LF54 which possesses Biofilm Adhesin Protein LapA. Although the lapA gene is conserved in P. putida strains [9], P. putida F1 and PpY101 were not able to dechlorinate CH ( Figure 3). Therefore, the LapA protein may have a unique function in LF54; conversely, dechlorination process may involve other factors. The latter possibility is more appealing to us because LapA does not share any similarity with enzymes possessing dehalogenation functions [17]. And we also found out that the defective lapA mutant rendered the variation of bacterial motility. For the future, an interesting challenge will be to verify the interactions between LapA and the CH dechlorination enzyme and to examine whether biofilm formation influences CH dechlorination; and to verify the relationship among the CH dechlorination and LapA and bacterial motility.
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.