TEXT

Top Abstract Text References

Ralstonia eutropha JMP134(pJP4) can utilize 2,4-dichlorophenoxyacetate (2,4-D) as the sole carbon and energy source for growth (5). The tfd genes encoding the 2,4-D pathway are located on plasmid pJP4 (4) and have been studied extensively. The genes tfdA (24), tfdB, and tfdCDEF (21) encode enzymes that are responsible for the conversion of 2,4-D via 2,4-dichlorophenol (2,4-DCP) and 3,5-dichlorocatechol to the central metabolite 3-oxoadipate. In addition, there are two genes with a regulatory function, namely, tfdR and tfdS (15, 18, 30).

Lately, our laboratory has been interested in ISJP4, an insertion sequence which is located in monocopy on plasmid pJP4 (16). This element seems to have had quite an impact on the genetic organization of the plasmid (Fig. 1). Firstly, it inserted itself into the tfdT gene and thereby inactivated the gene product as a regulator of tfdCDEF expression (15). Secondly, the presence of a small piece of ISJP4 near tfdS may be reminiscent of the element's involvement in the duplication of tfdR and tfdS, which are actually two identical copies of the same gene (30). Thirdly, the copy of ISJP4 which is inserted in tfdT and its small remnant near tfdS can form a composite transposon and together are able to mobilize intervening DNA sequences (16). It is quite possible that the approximately 11-kb fragment that lies trapped between the two pieces of ISJP4 was actually mobilized onto an early version of pJP4 (16).

View larger version (47K): [in this window] [in a new window]   FIG. 1.   (Top) Overview of the genetic organization on plasmid pJP4. Arrows, positions and orientations of the various tfd genes; black boxes, ISJP4 DNA. (Bottom) Restriction maps of plasmids used in this study. Plasmid pCBA94 contains a 1.9-kb NruI fragment of pJP4 in pUC19 (29). The sequence of this DNA fragment is presented in Fig. 2. Plasmid pCBA78 contains a 1.1-kb PstI-EcoRV fragment of tfdK in pUC19. A 1.7-kb SalI fragment carrying the kanamycin cassette from plasmid pUT/Km (12) was inserted into the XhoI site of pCBA78, resulting in plasmid pCBA79. The latter was used to construct a tfdK mutant of R. eutropha JMP134(pJP4). pUC-derived plasmids were propagated in cells of E. coli DH5 (23) cultivated at 37°C on Luria broth medium (23) in the presence of 100 µg of ampicillin per ml.

The latter idea made us curious about the information content of this mobile-mobilized DNA fragment. Therefore, we decided to determine its nucleotide sequence and discovered a whole new additional set of tfd genes (Fig. 1), i.e., tfdD_IIC_IIE_IIF_II, tfdB_II, and tfdK. The tfdK gene, its product, and its putative function are described in the present communication. The resemblance of the gene product to various transport proteins led us to investigate 2,4-D uptake by R. eutropha JMP134 and the possible role of tfdK in that process.

The tfdK gene encodes a member of the major facilitator superfamily of transporter proteins. The sequence of the tfdK gene and its flanking regions on plasmid pJP4 is presented in Fig. 2. The gene starts with an ATG codon, ends with TAG, and is 1,380 bp long. Interestingly, the stop codon (TAG) overlaps with the TTA duplicated target site of ISJP4 and its right-hand inverted repeat (GAGACTGTTTCAAAAAGA) (16). The tfdK gene is preceded by an AGGAG ribosome binding site. The polypeptide that it putatively encodes is 460 amino acids (aa) in length, with a predicted molecular mass of 47.6 kDa. More than half (54%) of the amino acids encoded by tfdK are hydrophobic (A, I, L, F, W, and V). Comparison of TfdK to entries in various protein sequence databases revealed homology to members of the so-called major facilitator superfamily (MFS) (17). This family consists of eubacterial, archaeal, and eukaryotic membrane proteins which facilitate the transport of various compounds such as sugars and antibiotics. Typical for members of the MFS is a structure of 12 membrane-spanning -helices, which were also identified for TfdK (Fig. 2). Furthermore, TfdK contained the GXXX[D/E][R/K]XG[R/K][R/K] motif between the second and third transmembrane domains (GYLADRIGRR; Fig. 2) and a partial repetition of this motif between domains 8 and 9 (GLFASRLIDR), which is another characteristic of MFS members (13).

View larger version (77K): [in this window] [in a new window]   FIG. 2.   Nucleotide sequence of the 1,846-bp NruI fragment of plasmid pJP4 containing the tfdK gene. DNA sequencing of subclones of this fragment was performed in a Thermosequenase reaction (Amersham, Little Chalfont, United Kingdom) with IRD-41-labeled primers (MWG Biotech, Ebersberg, Germany). DNA fragments were separated and analyzed on a LiCOR Automated Sequencer (model 400L; LiCOR, Lincoln, Nebr.). Relevant restriction sites are indicated in italics. A putative ribosome binding site is boxed. ISJP4 DNA is underlined, and its border is indicated by a black arrow (IR-R). The translation product of tfdK is given below the nucleotide sequence. Two characteristic MFS motifs are circled (13); underlined within these are conserved amino acid residues. Arrows on the amino acid sequence indicate predicted transmembrane -helices (see text); arrows that point in the same direction as the tfdK reading frame represent helices with an inside-to-outside orientation, oppositely pointed arrows represent outside-to-inside membrane helices. Eleven such helices were identified by the TMpred program of the ISREC Bioinformatics Group (Epalinges, Switzerland). The position of a 12th hydrophobic -helix is also shown (open arrow), but this helix was not recognized by the program as a membrane-spanning one.

Uptake of 2,4-D by wild-type cells of R. eutropha JMP134(pJP4). Wild-type JMP134(pJP4) cells that had been grown on 5 mM 2,4-D took up 2,4-D (Fig. 3A). Uptake rates were derived from the slope of the lines through each set of data points. The rates of 2,4-D uptake showed saturation kinetics with 2,4-D concentrations of as much as 60 µM (Fig. 3C), which suggests the involvement of a protein in the transport process. The uptake of 2,4-D was inducible; when cells growing on 10 mM fructose were spiked with 0.2 mM 2,4-D at mid-log phase and were allowed to continue growing for another 45 min before harvesting, we found uptake rates comparable to those of 2,4-D grown cells, whereas fructose-grown cells showed no significant uptake activity in the assay (not shown). Furthermore, the uptake was sensitive to incubation with metabolic inhibitors. When cells were preincubated with 0.5 mM potassium cyanide (Merck, Darmstadt, Germany), which prevents the formation of a proton motive force, uptake (at an initial 2,4-D concentration of 9.8 µM) was reduced to 17% of normal activity. Preincubation with the protonophore carbonylcyanide-m-chlorophenylhydrazone (CCCP) at 20 µM (Fluka, Buchs, Switzerland) or -dinitrophenol (DNP) (Merck) at 0.5 mM reduced uptake activities to 20 or 4%, respectively. The effect of 1 mM tetraphenylphosphonium (Fluka), which dissipates the membrane potential, was less severe; about half of the activity (44%) remained.

View larger version (25K): [in this window] [in a new window]   FIG. 3.   Uptake of 2,4-D by 2,4-D-cultivated R. eutropha JMP134 cells harboring pJP4 or pJP4::cba79 at different substrate concentrations. (A and B) Uptake of 2,4-D (in nanomoles · milligram of protein1) for wild type and the tfdK mutant, respectively. Numbers accompanying a set of data points represent the 2,4-D concentration (micromolar) in that particular uptake assay. Data points were corrected for aspecific binding of 2,4-D to the filters by control experiments in the absence of cells. The slope of the line through a set of data points was taken as the uptake rate. For higher 2,4-D concentrations, lines extrapolated to time zero did not pass through a zero 2,4-D concentration. This phenomenon was observed only with cells grown on 2,4-D; perhaps such cells have an altered cell surface that stimulates adsorption of 2,4-D. The uptake rates were plotted versus corresponding 2,4-D concentrations in the assay as shown in panels C (0 to 60 µM) and D (0 to 2 mM). Bars in panel D represent standard errors of the measurements; error bars at lower concentrations (panel C) fell within the symbols. Protein concentrations of the original cell suspensions were determined by the Bradford method (2) after boiling cells for 10 min in 0.1 N NaOH. Uptake experiments with 2,4-D-grown wild-type or tfdK mutant cells were repeated three times with independently cultured cells; uptake data for one culture only are shown. Solid lines and dotted lines in panels C and D represent modelled uptake behavior of wild-type cells and tfdK mutant cells, respectively (see text). The following parameters were used: 2,4-D permeability constant, 1.8 × 104 liters · min1 · mg of protein1 (both strains); Vmax,TfdA, 46 (wild-type) or 41 (tfdK mutant) nmol · min1 · mg of protein1; Vmax,TfdK of active transport, 20 (wild type) or 0 (tfdK mutant) nmol · min1 · mg of protein1; Km,TfdK, 5 µM (wild type only; not applicable for the tfdK mutant).

Uptake of 2,4-D by the tfdK mutant. Interruption of the tfdK gene had a clearly negative effect on the ability to transport 2,4-D in the micromolar range (Fig. 3B). Rates of uptake of 2,4-D-grown tfdK mutant cells were up to 10 times lower than those of the wild type and increased linearly with the 2,4-D concentration in this range (Fig. 3C). This suggests diffusion as a mechanism for 2,4-D transport. To substantiate this, we calculated the permeability coefficient for 2,4-D from the data obtained with the tfdK mutant, assuming diffusion as the sole transport mode. Uptake by simple diffusion (in micromoles · minute¹ · milligram of protein¹) occurs at a rate (V_diff) which is equal to

(1)

where P is the permeability coefficient (in liters · minute¹ · milligram of protein¹) and [2,4-D]_out and [2,4-D]_in are the external and intracellular concentrations (micromolar) of 2,4-D, respectively (22). Since intracellular concentrations are kept low in 2,4-D-grown cells because of metabolic activity, we also included a term for the activity of TfdA, the first enzyme of the 2,4-D pathway (8, 9, 24). We measured TfdA activities of 41 nmol · min¹ · mg of protein¹ in extracts of 2,4-D-grown tfdK mutant cells, as determined by the method described by Nickel et al. (20) and 46 nmol · min¹ · mg of protein¹ in extracts of wild-type cells. The activity of the TfdA enzyme (V_TfdA) (in micromoles · minute¹ · milligram of protein¹) equals

(2)

where V_max,TfdA is the maximum rate of conversion (in this case, 0.041 µmol · min¹ · mg of protein¹, see above) and K_m,TfdA is the enzyme's half-saturation constant (17.5 µM) (9). When diffusion and conversion of 2,4-D are in equilibrium (i.e., V_diff = V_TfdA), equations 1 and 2 can be combined, and the uptake rate (V_upt) (in micromoles · minute¹ · milligram of protein¹), which is equal to V_diff and V_TfdA, can be expressed as a function of the extracellular 2,4-D concentration as follows (1):

(3)

The 2,4-D uptake rates of the tfdK mutant were fitted in equation 3, and we could calculate an apparent permeability coefficient (P) for 2,4-D of 1.8 × 10⁴ liters · min¹ · mg of protein¹. By substituting this value in equation 1, one can calculate for any given extracellular 2,4-D concentration the highest possible uptake rate, i.e., V_upt = P · [2,4-D]_out (with [2,4-D]_in = 0), through diffusion alone. For example, at an extracellular 2,4-D concentration of 1 µM and with approximately 0.025 mg of protein per 0.4 × 10⁹ cells (determined in our assays), and an estimated surface area of 6 × 10⁸ cm² per cell (cells were monad shaped, 2 µm long, and 1 µm in diameter), the maximal number of 2,4-D molecules that can enter a single cell per second is 110. At high extracellular 2,4-D concentrations, the diffusion model predicts saturation of uptake to rates that approximate the V_max of TfdA (Fig. 4A); this is also what we found experimentally (Fig. 3D). At low 2,4-D concentrations, the rate of transport is indirectly driven by TfdA activity but is limited by the permeability of 2,4-D across the membrane. From these model considerations, we conclude that uptake of 2,4-D by the tfdK mutant most likely occurs through simple diffusion.

View larger version (16K): [in this window] [in a new window]   FIG. 4.   Predictions of 2,4-D uptake rates by the diffusion model for the tfdK mutant and by the two models for TfdK-mediated transport in wild-type R. eutropha JMP134(pJP4) cells. (A) Influence of TfdA activity for different values of Vmax,TfdA (51, 46, 41, and 36 nmol · min1 · mg of protein1) on uptake rates by the tfdK mutant (dotted lines) and wild-type cells (solid lines) at different assay concentrations. For the wild type, the curves for active transport and facilitated diffusion overlap; therefore, only one is shown. (B) Intracellular accumulation of 2,4-D, i.e., [2,4-D]in/[2,4-D]out, in the tfdA mutant (Vmax,TfdA of 0 nmol · min1 · mg of protein1) and with an apparent permeability coefficient for 2,4-D of 1.8 × 104 liters · min1 · mg of protein1, assuming either active transport (Vmax,TfdK, 20 nmol · min1 · mg of protein1; Km,TfdK, 5 µM [equation 6]) or facilitated diffusion (Vmax,TfdK, 160 nmol · min1 · mg of protein1; Km,TfdK, 4 µM [equation 5]).

Modelling TfdK transport activity. Since the only distinction between the wild-type strain and the tfdK mutant was the presence of an intact versus interrupted tfdK gene, and since TfdA activities did not differ significantly between these two strains, the apparently higher rates of uptake by the wild type at lower concentrations must be due to a higher transport rate of 2,4-D across the membrane. Therefore, we propose that the product of the tfdK gene, TfdK, facilitates 2,4-D uptake in wild-type R. eutropha JMP134(pJP4). We tested two hypotheses for the function of TfdK: (i) TfdK is an active transporter protein capable of accumulating 2,4-D against a concentration gradient and (ii) TfdK is a facilitator of downhill, i.e., outside-to-inside, diffusion. Both functions were superimposed on the diffusion model (equation 3) and were fitted by numerical iteration (i.e., determining changes in extra- and intracellular 2,4-D concentrations and in the flux of 2,4-D over time intervals of 0.01 s during 30 s as a function of diffusion and TfdK and TfdA activity) to the observed 2,4-D uptake rates of the wild type.

(4)

1

1

1

1

(5)

1

1

1

2

(6)

View larger version (21K): [in this window] [in a new window]   FIG. 5.   2,4-D uptake by 2,4-D-induced tfdA mutant cells (A) and 2,4-DCP-induced (B) wild-type (black symbols) and tfdA mutant (grey symbols) cells of R. eutropha JMP134. Cells were grown on 10 mM fructose to mid-log phase, after which 2,4-D or 2,4-DCP was added to 0.4 or 0.1 mM, respectively, and the cells were allowed to continue growth for another 45 or 70 min, respectively, before harvesting, washing, and use in the uptake assay. Uptake assays were performed as described in the text. The 2,4-D concentration in a particular uptake assay is shown. Uptake values are given in picomoles of 2,4-D (one-time sample of 0.4 × 109 cells) and not in nanomoles per milligram of protein, since the protein content of the tfdA mutant was lower than that of wild-type cells. Thin lines in both panels indicate the predicted initial rates of uptake for tfdA mutant cells at the given assay concentrations either by diffusion (A; equation 1, with [2,4-D]in = 0) or by diffusion plus TfdK-mediated active transport (B; equations 1 plus 4, with [2,4-D]in = 0). Parameters: P = 1.8 × 104 liters · min1 · mg of protein1; Vmax,TfdK = 20 nmol · min1 · mg of protein1; Km,TfdK = 5 µM, with 0.025 mg of protein per filter or total cell surface area of 24 cm2 per filter.

Nucleotide sequence accession number. The sequence reported in this paper has been deposited with GenBank under accession no. U16782.