|Home | About | Journals | Submit | Contact Us | Français|
The tcpRXABCYD operon of Cupriavidus necator JMP134 is involved in the degradation of 2,4,6-trichlorophenol (TCP). All of the gene products except TcpY have assigned functions in TCP metabolism. Sequence comparison identified TcpY as a member of COG4313, a group of hypothetical proteins. TcpY has a signal peptide, indicating it is a membrane or secreted protein. Secondary structure and topology analysis indicated TcpY as a β-barrel outer membrane protein, similar to the Escherichia coli outer membrane protein FadL that transports hydrophobic long-chain fatty acids. Constitutive expression of tcpY in two C. necator strains rendered the cells more sensitive to TCP and other polychlorophenols. Further, C. necator JMP134 expressing cloned tcpY transported more TCP into the cell than a control with the cloning vector. Thus, TcpY is an outer membrane protein that facilitates the passing of polychlorophenols across the outer membrane of C. necator. Similarly, other COG4313 proteins are possibly outer membrane transporters of hydrophobic aromatic compounds.
2,4,6-Trichlorophenol (TCP), a major environmental pollutant, has been used extensively as a preservative and a biocide (Czaplicka 2004). The gram negative bacterium Cupriavidus necator JMP134, a versatile degrader of aromatic compounds, can use TCP as a carbon and energy source (Clement et al. 1995; Padilla et al. 2000). A gene cluster, tcpRXABCYD, is involved in TCP degradation (Louie et al. 2002; Matus et al. 2003; Sanchez and Gonzalez 2007), and all the encoded proteins have been characterized except TcpY (Fig. 1). First, TcpA (TCP 4-monooxygenase) oxidizes TCP to 2,6-dichloro-p-benzoquinone and then hydrolyzes the latter to 6-chloro-2-hydroxy-p-benzoquinone (Xun and Webster 2004). Since TcpA is a reduced flavin adenine dinucleotide (FADH2)-dependent monooxygenase, TcpX (NADH:FAD oxidoreductase) supplies FADH2 to TcpA in coupled reactions (Belchik and Xun 2008). Second, TcpB (NADH:quinone oxidoreductase) reduces the quinone to 6-chloro-2-hydroxy-p-hydroquinone (Belchik and Xun 2008). Third, TcpC (dioxygenase) catalyzes the ring-cleavage of the latter to produce 2-chloromaleylacetate (Louie et al. 2002). Fourth, TcpD (maleylacetate reductase) removes the last chlorine and then converts maleylacetate to 3-ketoadipate, a common metabolic intermediate of microbial degradation of aromatic compounds (Harwood and Parales 1996). The tcpR gene codes for a regulatory protein controlling the expression of tcpXABCYD (Sanchez and Gonzalez 2007).
The tcpY gene is located in the middle of the tcp gene cluster, but the function of TcpY remains unknown. TcpY does not share significant sequence similarity to any known proteins, and a tcpY mutant is normal in TCP degradation (Sanchez and Gonzalez 2007). However, a structural homology search described below indicated TcpY displayed significant homology with FadL, an Escherichia coli outer membrane transporter protein required for long-chain fatty acid (LCFA) transport (van den Berg et al. 2004). Gram-negative bacteria have hydrophilic outer membranes that act as a barrier against hydrophobic molecules and thus require transporters to facilitate diffusion into the cytoplasm (van den Berg 2005). Like LCFAs, TCP is a hydrophobic molecule. Therefore, TcpY may function as a transporter to facilitate TCP diffusion into the cell. Here, we report evidence supporting TcpY as an outer membrane transporter that facilitates the diffusion of polychlorophenols into C. necator.
Reagents used were purchased from Sigma Chemical Co. (St. Louis, MO) or Aldrich Chemical Co. (Milwaukee, WI). Enzymes were purchased from Invitrogen (Carlsbad, CA) or New England Biolabs (Beverly, MA).
Cupriavidus necator JMP134 and C. necator H16 were grown in a mineral salt medium with 0.2% glutamate as the carbon source at 30°C (Louie et al. 2002). E. coli strains were grown in Luria–Bertani (LB) medium at 37°C. Kanamycin was used at a final concentration of 30 μg ml−1.
The BLASTP program (Altschul et al. 1997) was used to search sequence similarity. Signal sequence of membrane proteins was predicted by using SignalP http://www.cbs.dtu.dk/services/SignalP/) (Emanuelsson et al. 2007).
The likelihood of proteins to form β-strands in the outer membrane of gram-negative bacteria was calculated using the PRED-TMBB (http://bioinformatics.biol.uoa.gr/PREDTMBB) (Bagos et al. 2004). The three-dimensional model of TcpY was constructed using Phyre version 0.2 (http://www.sbg.bio.ic.ac.uk/phyre/) (Bennett-Lovsey et al. 2008). Phyre calculates the best fit towards a solved crystal structure with estimated precision predicting likelihood of fit.
Protein sequences were aligned by using CLUSTALW (Larkin et al. 2007), and a phylogenetic tree was generated by using MEGA version 4.0.2 (Tamura et al. 2007). The evolutionary distances were computed using the Jones et al. (1992) method and are in units of the number of amino acid substitutions per site.
A 493-bp internal fragment of tcpY was amplified from C. necator JMP134 DNA by PCR using primer pair TcpY-internalF (gtggtaccgatcgtgtctgc) and TcpY-internalR (actgatactgggtggcttcg). The PCR products were cloned into pCR2.1-TOPO, forming plasmid pKOY. The plasmid was amplified in E. coli and isolated. The purified plasmid DNA (50 ng) was electroporated into C. necator JMP134 cells (Louie et al. 2002). Recombinant strains were selected on LB agar plates containing kanamycin.
The full length tcpY gene was cloned in the plasmid pTrc99a (Pharmacia, Piscataway, NJ), where tcpY was under the control of the trc promoter that contains a LacI binding site (Amann et al. 1988). The tcpY gene was PCR amplified with primers TcpYR (gctttcatcgaagcttctcccgtc) and TcpYF-pTrc99 (gttcgacagatgccatggaggtcc), digested with HindIII and NcoI, and ligated into pTrc99a to generate pTrc99a-tcpY. As the pTrc99a-tcpY plasmid cannot replicate in C. necator JMP134, the cloned tcpY is moved to pBBR1MCS2 that can replicate in a broad host range (Kovach et al. 1994, 1995). The pTrc99a-tcpY plasmid was used as a template for PCR using pTrc99-F (gttctggataagcttttttgcgcc) with a HindIII site and pTrc99-R (atcttctctcatccgcca). This PCR product was digested with HindIII and ligated into pBBR1MCS2 to create pTcpY, in which tcpY is under the control of the trc promoter from pTrc99a. Electroporation was utilized to transfer pTcpY into C. necator JMP134 and C. necator H16. In C. necator, the trc promoter is constitutively expressed due to the lack of repressor, LacI. pBBR1MCS2 was also electroporated into these strains as the empty vector control.
To determine the effects of constitutive tcpY expression when TCP is present, we utilized mineral salt medium agar plates with 0.2% glutamate and varying amounts of TCP (from 0 to 400 μM). Glutamate represses the tcp operon in C. necator JMP134, so no degradation occurs (Louie et al. 2002). Briefly, cells were cultured overnight in LB medium, harvested and resuspended in the mineral salt medium without carbon source at a turbidity of 1.0 at 600 nm (~1 × 107 cells ml−1). Serial dilutions were performed in the medium, and 3 μl was spotted onto the agar plates. The plates were incubated for 3 days at 30°C and pictures were taken. If the cells are sensitive to the TCP concentrations, the cells at higher dilutions would have small or no colonies due to slower growth.
Several polychlorophenols were tested to determine if TcpY was specific to TCP. C. necator JMP134 was cultured overnight in LB, harvested, and resuspended in the mineral salt medium with no carbon source to a turbidity of 1.0 at 600 nm. A 1% inoculum was then transferred to the mineral salt medium containing 0.2% glutamate and 100 μM 2,6-dichlorophenol (DiCP), TCP, 2,3,4,6-tetrachlorophenol (TeCP), 2,3,5,6-TeCP, or pentachlorophenol (PCP). Cultures were shaken for 18 h at 30°C and a final turbidity of 600 nm was measured.
Cells of C. necator JMP134 carrying pTcpY or pBBR1MCS2 were cultured for 5 h to a turbidity of 2 at 600 nm in LB medium, harvested and washed with mineral salt medium containing 0.2% glutamate, and resuspended in the same medium to a turbidity of 2.0 at 600 nm. Protein concentration was determined to be about 0.4 mg per ml by using modified Lowry protein assay (Pierce Biotechnology, Rockford, IL). One ml of culture was transferred to 1.7-ml micro-fuge tube, to which varying amounts of TCP (12.5–100 μM TCP) was added. The tubes were shaken for 1 min at room temperature, and then were centrifuged at 12,000×g for 1 min. First, the pellets were washed with 1 ml of 50 mM KPi (pH 7.0) without suspending the cells. Then, they were washed twice with 1 ml of 50 mM KPi (pH 7.0) by suspension and centrifugation. After the washing, each pellet was resuspended in 100 μl of the KPi buffer with 1% sodium dodecyl sulfate. The sample was vortexed and incubated at 50°C for 5 min. About 100 μl of hexane:1-propanol (vol/vol, 4:1) was added, and the mixture was vortexed for 1 min. After centrifugation, TCP in the organic phase was analyzed with high performance liquid chromatography equipped with a C-18 column and photodiode array detector, as previously described (Louie et al. 2002). The volume of the organic phase was assumed to be 100 μl for calculations.
BLASTP search with TcpY sequence returned only hypothetical proteins that are grouped together in the cluster of orthologous groups (COG) 4313 (Marchler-Bauer et al. 2007). Although no experimental data exists for this particular COG, members are thought to be involved in the meta-pathway of phenol degradation due to their gene location with genes responsible for phenol degradation (Xu et al. 2003). SignalP indicated that TcpY had a signal peptide and a predicted cleavage site between amino acid residues 38 and 39. The estimated molecular mass of the mature polypeptide was 31,715.24 Da. The signal peptide suggested that TcpY was a membrane or secreted protein. C. necator has both cytoplasmic and outer membranes. Integral membrane proteins of cytoplasmic membranes usually consist of α-helices for the transmembrane portions; whereas, the outer membrane proteins are composed of antiparallel β-strands for transmembrane segments that form a barrel in the outer membrane, e.g., porins (Schulz 2002). PRED-TMBB, a web-server capable of predicting and discriminating β-barrel outer membrane proteins, was utilized to analyze TcpY (Bagos et al. 2004). PRED-TMBB scored TcpY with a value of 2.877 that is less than the threshold value of 2.965. Values below this threshold correctly predict β-barrel membrane proteins with 95% accuracy using a database of known β-barrel outer membrane proteins (Bagos et al. 2004). PRED-TMBB predicted the topology of TcpY with fourteen transmembrane β-strands. Structural homology modeling using Phyre predicted TcpY structure with an estimated 95% precision related to the three-dimensional structure of FadL from E. coli. FadL is a transporter that facilitates the diffusion of hydrophobic LCFAs across hydrophilic outer membranes of E. coli and related gram-negative bacteria (van den Berg et al. 2004). Thus, bioinformatic analysis indicated TcpY as a β-barrel outer membrane protein, possibly involved in TCP uptake.
Other structural homologs of FadL have been linked to the uptake of hydrophobic molecules for biodegradation (van den Berg 2005). XylN is identified as an outer membrane transporter for the uptake of xylene (Kasai et al. 2001); TodX and TbuX are for toluene uptake (Wang et al. 1995; Kahng et al. 2000); StyE is for styrene uptake (Mooney et al. 2006). The resolved structures of TodX and TbuX are similar to FadL (Hearn et al. 2008). To determine how TcpY is related to these proteins, a multiple alignment was utilized. The proteins are listed in Table 1 and the phylogenetic tree created from the alignment is shown in Fig. 2. Bootstrap values of 100 indicated the proteins fell into three separate clades: LCFA transporters, xenobiotic transporters, and COG4313 hypothetical proteins. TcpY was grouped with the COG4313 proteins. The substrates for the three groups are also different: hydrophobic LCFA, aromatic compounds without hydrophilic groups, and aromatic compounds with hydroxyl groups, respectively.
The COG4313 proteins used for phylogenetic analysis (Fig. 2) were further analyzed (Table 2). Like TcpY, the four hypothetical proteins all contained signal peptides, scored as β-barrel outer membrane proteins, and shared high structural homology with FadL. COG4313 proteins are within gene clusters responsible for biodegradation of phenolic compounds, which are likely the substrates of COG4313 proteins. The orf8 gene is located near meta-pathway of phenol degradation genes in Acinetobacter calcoaceticus PHEA-2 (Xu et al. 2003). The P. putida GB-1-Hypo protein gene is adjacent to phenol degradation genes.
The tcpY gene was disrupted via homologous integration of a suicidal plasmid that carried an internal fragment of the tcpY gene in C. necator JMP134. The integration of pKOY resulted in a kanamycin resistant mutant that contained two truncated copies of tcpY on the chromosome. Integration of pKOY was confirmed by PCR using one primer located on the chromosome TcpYKOF (gtattccggcctgatgttgc) and one located on the plasmid pCR-R (gttttcccagtcacgacgtt) (data not shown). The tcpY mutant and C. necator JMP134 wild type degraded TCP at comparable rates using TCP concentrations from 10 to 400 μM. Also, the mutant and wild type showed similar sensitivity to TCP (100–400 μM) as well as other polychlorophenols both in the mineral salt liquid medium and on agar plates (data not shown). Sanchez and Gonzalez have also reported that a tcpY knockout mutant can grow at the same rate as the wild type with TCP as the sole carbon source (Sanchez and Gonzalez 2007). Thus, TCP is able to pass the outer membrane of C. necator JMP134 without the aid of TcpY under the culturing conditions. However, it is unknown if TcpY affects TCP transport in the environment, as different growth conditions alter the permeability of the outer membrane of gram-negative bacteria (Nikaido 2003).
To evaluate whether TcpY facilitated the diffusion of TCP into C. necator JMP134 cells, we compared the sensitivity to TCP with and without the presence of constitutively expressed tcpY. If TcpY functions as an outer membrane transporter for TCP, constitutive expression of tcpY should produce TcpY at higher concentrations than in the wild type where the gene is repressed in the presence of glutamate (Louie et al. 2002). Without the operon expressing the genes required to degrade TCP, TCP would slow down cell growth. The decreased growth would produce small or no colonies on agar plates inoculated with serially diluted cells. On agar plates, C. necator JMP134 containing pBBR1MCS2 (empty vector) displayed decreased growth starting at 200 μM TCP but was still able to grow at 400 μM TCP (Fig. 3). On the other hand, the strain containing pTcpY, which constitutively expressed tcpY, exhibited decreased growth at 100 μM and almost a complete abolishment of growth at 400 μM TCP (Fig. 3). The results indicated that TcpY allowed more TCP to enter the cells.
To study the function of TcpY in a closely related bacterium, pBBR1MCS2 and pTcpY were electroporated into C. necator H16, a bacterium unable to degrade TCP. Overall, C. necator H16 had a higher sensitivity to TCP in comparison to C. necator JMP134. On agar plates, C. necator H16 control did not display decreased growth until 100 μM TCP while the strain carrying pTcpY started to exhibit decreased growth at 50 μM TCP and almost a complete inhibition of growth at 200 μM (data not shown). The results indicated that TcpY also brought more TCP into the cells and made them more sensitive to TCP in comparison to the control carrying the empty vector.
Since FadL transports various LCFAs, we tested the ability of tcpY to transport other polychlorophenols. The experiments were done by growing the cells with glutamate that suppresses the expression of the tcp operon. Further, the tcp operon is only induced by TCP and not by other polychlorophenols (Sanchez and Gonzalez 2007). We determined the sensitivity of C. necator JMP134 towards DiCP, TCP, 2,3,4,6-TeCP, 2,3,5,6-TeCP, and PCP. DiCP did not have any apparent effect on the growth of C. necator JMP134 carrying pTcpY compared to the empty vector control (Fig. 4), indicating it is likely not a substrate for TcpY. The two TeCPs and PCP all reduced growth when compared to without polychlorophenols in the medium for cells carrying the empty vector, demonstrating toxicity of the TeCPs and PCP. Moreover, C. necator JMP134 cells carrying pTcpY displayed more sensitivity to TCP, 2,3,4,6-TeCP, 2,3,5,6-TeCP, and PCP than the cells with the empty vector (Fig. 4).
2,4,6-Trichlorophenol uptake by JMP134 cells carrying pTcpY or pBBR1MCS2 was determined after 1 min incubation in the presence of glutamate. JMP134 cells constitutively expressing tcpY from the plasmid pTcpY transported more TCP into the cells than the cells carrying the vector pBBR1MCS2 at various TCP concentrations (Fig. 5). The method was taking advantage of TCP accumulation in lipid membranes so that quick washes did not result in any significant loss of the transported TCP. Further, glutamate inhibits TCP metabolism (Louie et al. 2002). Because centrifugation and wash both took time, the calculated rate may not be accurate. However, the measured rates reflected the difference mediated by TcpY. The data provided direct evidence that TcpY facilitated TCP transportation into C. necator cells.
β-Barrel outer membrane proteins share similar structures but limited sequence homology due to the lack of evolutionary constraints for loop regions that connect the β-sheets (Jeanteur et al. 1991). Some β-barrel outer membrane proteins can facilitate the transport of hydrophobic substrates across the outer membrane, which is often hydrophilic due to lipopolysaccharides (Nikaido 2003). TcpY is identified as a β-barrel outer membrane protein because of its structural similarity to characterized β-barrel outer membrane proteins, including FadL, TodX and TbuX (Wang et al. 1995; Kahng et al. 2000; van den Berg 2005). The toxicity and uptake studies support that TcpY facilitates the transport of polychlorophenols across the outer membrane of gram negative bacterium C. necator.
The research was supported by the US National Science Foundation grant MCB-0323167. Sara Belchik was also supported by an NIH Biotechnology Training Grant.
Sara Mae Belchik, School of Molecular Biosciences, Washington State University, Life Sciences Building, Room 202, 100 Dairy Road, Pullman, WA 99164-7520, USA.
Scott M. Schaeffer, School of Molecular Biosciences, Washington State University, Life Sciences Building, Room 202, 100 Dairy Road, Pullman, WA 99164-7520, USA.
Shelley Hasenoehrl, School of Molecular Biosciences, Washington State University, Life Sciences Building, Room 202, 100 Dairy Road, Pullman, WA 99164-7520, USA.
Luying Xun, School of Molecular Biosciences, Washington State University, Life Sciences Building, Room 202, 100 Dairy Road, Pullman, WA 99164-7520, USA.