INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

In Escherichia coli, sulfate starvation causes increased synthesis of several proteins involved in scavenging sulfur from alternative sulfur sources (15). Recently, two sets of genes whose expression is derepressed in the absence of sulfate or cysteine were identified. The tauABCD gene cluster, located at 8.5 min on the E. coli chromosome, encodes a sulfonate-sulfur utilization system that is specifically involved in the utilization of taurine (2-aminoethanesulfonic acid) as a source of sulfur. Disruption of tauB, tauC, or tauD resulted in the loss of the ability to utilize taurine as a source of sulfur but did not affect the utilization of a range of other aliphatic sulfonates (21). The TauD protein is an -ketoglutarate-dependent taurine dioxygenase (3), and the TauABC proteins exhibit similarity to ATP-binding cassette (ABC)-type transport systems (21). A second set of genes, the ssuEADCB gene cluster, located at 21.4 min on the chromosome, enables E. coli to utilize aliphatic sulfonates other than taurine as a source of sulfur. Deletion of ssuEADCB caused an inability to utilize alkanesulfonates but did not affect the utilization of taurine (24). SsuD is a monooxygenase that catalyzes the desulfonation of a wide range of sulfonated substrates other than taurine, including C₂ to C₁₀ unsubstituted linear alkanesulfonates, substituted ethanesulfonic acids and the buffer substances HEPES, MOPS (morpholinepropanesulfonic acid), and PIPES [piperazine-N,N'-bis(2-ethanesulfonic acid)]. This monooxygenase is dependent on reduced flavin mononucleotide (FMN) which is provided by the SsuE protein, an NAD(P)H:FMN oxidoreductase (4). The SsuABC proteins also appear to constitute an ABC transport system (24). Neither the TauD enzyme nor the two-component SsuD-SsuE monooxygenase desulfonated methanesulfonic acid, cysteic acid, or aromatic sulfonates (4), compounds that are unable to satisfy the sulfur requirement of E. coli EC1250. These two enzyme systems thus cover the full range of desulfonation activities in this E. coli strain. They convert alkanesulfonates to the corresponding aldehyde and sulfite, which has been shown to enter the sulfite reduction pathway to cysteine (20).

In the present study we investigated the role of the tauABC and ssuABC genes in the utilization of taurine and alkanesulfonates as sulfur sources. The tauA and ssuA genes encode putative signal sequences, indicating that their products probably function as periplasmic binding proteins. The sequences of TauB and SsuB and of TauC and SsuC are significantly similar to those of ATP-binding proteins and integral membrane components, respectively, of members of the ABC transporter superfamily (6). By analogy to known binding-protein-dependent ABC transporters (2), it is inferred that these systems are composed of a homodimeric membrane protein and a homodimeric ATP-binding protein. A pairwise comparison of the components of the TauABC and SsuABC transporters revealed sequence identities of 22.7% for TauA and SsuA, 40.4% for TauB and SsuB, and 34.5% for TauC and SsuC. Using a genetic approach, we explored to what extent the substrate specificity of the TauD and SsuD-SsuE desulfonation systems is reflected in the substrate range of the corresponding transport systems and whether components of the two transport systems are functionally exchangeable.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals. All chemicals used as sulfur sources were of the highest quality available and were obtained from Fluka, except N-phenyltaurine and 4-phenyl-1-butanesulfonate (Sigma), isethionic acid (Aldrich), and 3-aminopropanesulfonate (Acros Organics). Oligonucleotides were purchased from Microsynth (Balgach, Switzerland). Restriction endonucleases, T4 DNA ligase, and Taq DNA polymerase were obtained from MBI Fermentas. Pfu DNA polymerase was from Promega.

E. coli strains and growth conditions. E. coli strain DH5 (16), used for cloning purposes, was grown with constant shaking (180 rpm) at 37 or 30°C in Luria-Bertani (LB) medium (16). Solid media were prepared by addition of 1.5% (wt/vol) agar. When appropriate, the following additions were made: ampicillin, 100 µg/ml; kanamycin, 50 µg/ml; chloramphenicol, 35 µg/ml; isopropyl--D-1-thiogalactopyranoside (IPTG), 0.5 mM; 5-bromo-4-chloro-3-indolyl galactoside (X-Gal), 80 µg/ml; and sucrose, 5% (wt/vol). For plasmid isolation, restriction enzyme digestion, and transformation of E. coli, standard procedures were used (1). DNA for sequencing was prepared using either the QiaSpin Miniprep kit from Qiagen or the Jetstar Midiprep kit from Genomed.

Construction of chromosomal in-frame deletion mutants. The DNA (at least 500 bp) flanking the gene or group of genes to be deleted was amplified using Pfu DNA polymerase. Oligonucleotide primers were designed to introduce adequate restriction sites for subsequent cloning purposes (Table 1). Their approximate locations in the tau and ssu operons are shown in Fig. 1. Identical restriction sites were introduced at the 5' end (around 20 bp downstream of the start codon) and at the 3' end (30 to 40 bp before the stop codon) of the gene or group of genes to be deleted. The external primers used for PCR of the flanking regions introduced restriction sites available in plasmid pBluescript II KS (Stratagene). After digestion with the appropriate restriction enzymes, both PCR products were ligated together into pBluescript. The inserts of the resulting plasmids were sequenced to confirm that in-frame ligation had occurred and that no changes in the DNA sequence were introduced during PCR. Subsequently the deletion inserts were subcloned in plasmid pKO3 (10). All plasmids used for the construction of chromosomal E. coli EC1250 in-frame deletion mutations of tau and ssu genes are listed in Table 2.

View larger version (24K): [in this window] [in a new window]   FIG. 1.   Organization of the ssuEADCB (A) and tauABCD (B) operons, showing the approximate positions of the oligonucleotide primers used for the construction of in-frame deletion inserts and the plasmids used for complementation analysis of deletion mutants.

Construction of tauD and ssuD tauD mutants. Since efforts to obtain an in-frame deletion of tauD remained unsuccessful, a tauD mutant of E. coli EC1250 was constructed by P1 transduction (11). The tauD::lacZ fusion of the E. coli MC4100 derivative MW108 [(tauD::lacZ)(placMu9)] (21) was transduced into E. coli EC1250. Kanamycin-resistant transductants were screened for inability to grow on M63 minimal medium with taurine as a sulfur source and for the formation of blue colonies on M63 minimal medium supplemented with glutathione as a sulfur source and X-Gal to confirm the presence of the lacZ fusion. To obtain an ssuD tauD double mutant, an EC1250 mutant with ssuD deleted was used as a recipient.

Complementation analysis of in-frame deletion mutants. Restriction sites which are unique within the tauABCD and ssuEADCB operons were used to construct the plasmids shown in Fig. 1 and listed in Table 2. Plasmids for the complementation analysis of tau deletions were constructed in pUC19 (26), starting from plasmid pUC18ALA4 (14). Plasmids required for the complementation analysis of ssu deletions were constructed in pBluescript II KS, starting with plasmid pME4221, which carries the entire ssu operon on a 5.6-kb EcoRI fragment (24). Growth of deletion mutants transformed with the appropriate plasmid was measured in 5 ml of sulfur-free M63 minimal medium supplemented with taurine or butanesulfonate as a source of sulfur. Cultures were incubated at 37°C overnight, and the optical density at 600 nm was recorded on the following day. Alternatively, when growth of strains was investigated over a longer period of time, single colonies of the deletion mutant as well as of the strain transformed with the appropriate plasmid were resuspended in 50 µl of sulfur-free M63 minimal medium, from which 20 µl was spotted on an M63 minimal medium plate containing taurine or butanesulfonate as a sulfur source. Sulfur-limited solid media were prepared by the addition of 1.5 to 2% SeaPlaque agarose (FMC BioProducts).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Construction of in-frame deletions in the tau and ssu operons. Figure 2 shows the functional transport components in deletion mutants that were constructed to probe for the requirement for the TauABC and SsuABC proteins for growth on alkanesulfonate compounds. A first group of mutants lacked one of the putative periplasmic binding proteins (Fig. 2A and B) or the membrane-associated components TauBC or SsuCB (Fig. 2D and E). A second group included constructs that were defective in the corresponding components of both transport systems at the same time. A mutant devoid of both periplasmic binding proteins (Fig. 2C) allowed exploration of whether or not certain alkanesulfonates can enter the cell in the absence of the periplasmic components. Similarly, a mutant lacking the membrane-associated components of both systems (Fig. 2F) was expected to be defective in the uptake of all sulfonate sulfur sources transported via the TauABC and SsuABC systems. A third group of mutants carried various combinations of deletions that abolished the function of a periplasmic component, an integral membrane component, and an ATP-binding protein of either transport system. Six out of the eight possible combinations of multiple deletion mutants were obtained (Fig. 2G to L), and the growth properties of these mutants were expected to provide information on the exchangeability of components between the TauABC and SsuABC transport systems. Attempts to obtain the multiple deletion tauABC and ssuA ssuB tauC strains by using the chromosomal mutagenesis system described by Link et al. (10) were unsuccessful, and their construction was not pursued further.

View larger version (86K): [in this window] [in a new window]   FIG. 2.   Active components of the TauABC and SsuABC transport systems in the deletion mutants.

Growth of wild-type and deletion strains with alkanesulfonates as sulfur sources. Growth was tested in a medium containing one of the following sulfonates as the only sulfur source: taurine, isethionate (2-hydroxyethanesulfonic acid), 3-aminopropanesulfonate, HEPES, MOPS, PIPES, 2-(4-pyridyl)ethanesulfonate, N-phenyltaurine, 1,3-dioxo-2-isoindoline-ethanesulfonate, 4-phenyl-1-butanesulfonate, sulfoacetate, ethanesulfonate, propanesulfonate, butanesulfonate, pentanesulfonate, hexanesulfonate, octanesulfonate, decanesulfonate, and dodecanesulfonate. All of these compounds have been shown to be substrates for the TauD and/or the SsuD enzyme (3, 4). The qualitative assessment of growth with each of the 19 sulfonates tested was based on a comparison of the growth characteristics of each mutant with growth of the wild type. The wild type defined the upper level of growth, while the ssuD tauD double mutant served as a reference for zero growth with a particular sulfonate sulfur source. Growth of the deletion mutants on different sulfur sources was graded as described in Table 3, footnote a. The growth curves of the ssuA and the ssuCB mutants shown in Fig. 3 illustrate the four arbitrarily established growth categories. From Fig. 3 it is also evident that growth with sulfonate sulfur sources was diauxic, with a first, minor growth phase presumably supported by contaminating sulfate and a second, longer growth phase reflecting sulfonate utilization. Diauxic growth was not observed when strains were cultured with sulfate as a source of sulfur. Furthermore, mutants in which transporter genes were deleted grew faster than the wild-type strain on either sulfate or sulfonates as a sulfur source. However, depending on the sulfonate sulfur source tested, fluctuations in the growth rate and length of the lag phase were noted. Since strains lacking TauD or SsuD grew like the wild type, we presume that these phenomena have their origin at the level of the transport of alkanesulfonates.

View larger version (33K): [in this window] [in a new window]   FIG. 3.   Growth of E. coli EC1250 and deletion mutants. Cells were grown in a microtiter plate as described in Materials and Methods (150-µl cultures). The optical density at 600 nm (OD600 nm) was recorded with a microtiter plate reader every 5 min over 24 h. Every fifth measurement is shown. Only the growth profiles obtained with mutants grown on 1,3-dioxo-2-isoindolineethanesulfonate (A), butanesulfonate (B), PIPES (C), and MOPS (D) are shown. The optical densities obtained ranged from 0.25 to 0.5 and corresponded to optical densities of 1.0 to 2.0 when measured with a 1-cm light path in a Uvikon P-810 spectrophotometer.

Specificity of the TauABC and SsuABC transporters. Table 3 summarizes the growth phenotypes of the different E. coli constructs that carried one or more deletions in the genes for the putative TauABC and SsuABC transporters. The growth data indicate that taurine entered the cell exclusively via the TauABC transporter, whereas N-phenyltaurine, 4-phenyl-1-butanesulfonate, sulfoacetate, ethanesulfonate, propanesulfonate, hexanesulfonate, octanesulfonate, decanesulfonate, MOPS, and HEPES were transported only by the SsuABC transporter. Some sulfonates were taken up via both the TauABC and the SsuABC systems. These included PIPES, 2-(4-pyridyl)ethanesulfonate, isethionate, 1,3-dioxo-2-isoindolineethanesulfonate, 3-aminopropanesulfonate, butanesulfonate, and pentanesulfonate, 1,3-Dioxo-2-isoindolineethanesulfonate and 3-aminopropanesulfonate appeared to be transported with the same efficiency by both systems. Figure 4 summarizes these data. Comparison of this scheme with the previously reported substrate ranges of the TauD and SsuD desulfonation enzymes (3, 4) indicates that the range of substrates transported by the SsuABC system is congruent with the substrate specificity of its desulfonation enzyme. This rule does not apply to the TauABC system, whose desulfonation enzyme reacts with propanesulfonate, hexanesulfonate, and MOPS, compounds that enter the cell exclusively via the SsuABC transporter.

View larger version (53K): [in this window] [in a new window]   FIG. 4.   Substrate specificities of the E. coli TauABC and SsuABC transporters.

Complementation analysis of deletion mutants. For the construction of complementation plasmids, it was assumed that chromosomal in-frame deletions would not affect the expression of the genes located upstream of the deleted gene or group of genes. Therefore, the genetic information available on the complementation plasmids started with the regulatory sequences of the operon of interest and ended with the sequence of the gene that had been deleted on the chromosome (Fig. 1). In the case of mutants with multiple deletions, the complementation of each deletion was investigated separately, if necessary in an intermediary deletion mutant.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

The E. coli TauABCD and SsuEADCB systems for the assimilation of sulfur from taurine and other alkanesulfonates have previously been characterized at the level of the biochemistry of desulfonation (3, 4). Besides containing the structural genes directly involved in desulfonation, the tauABCD and ssuEADCB operons are thought to encode components of ABC transporters for sulfonates (21, 24). Based on the results from growth experiments and on sequence data, we postulate that TauABC and SsuABC form two distinct ABC transporters for uptake of alkanesulfonates. Sequence comparisons indicate that ABC-type transport systems involved in microbial desulfonation are also present in Bacillus subtilis (23) and Pseudomonas putida (25). Like in E. coli, these are encoded in the same transcription unit as the desulfonating enzyme. The deletion analysis of the E. coli taurine and alkanesulfonate transport systems presented here goes beyond amino acid sequence comparisons; it attempts to characterize these systems in terms of the substrate range and functional exchangeability of their components.

Our results demonstrate that the TauABC and the SsuABC systems complement each other with respect to the range of substrates transported. The substrate ranges of TauD and SsuD were to a large extent reflected in the substrate range of the corresponding ABC transporter. The substrate range of the SsuABC transporter fit that of the SsuD alkanesulfonate monooxygenase. This correlation is not maintained for the TauABC transport system, which was unable to transport HEPES, MOPS, and propanesulfonate, although these compounds are substrates for the TauD enzyme (3). Thus, two systems for alkanesulfonate uptake and metabolism are available in E. coli, with TauABCD being specifically involved in uptake and desulfonation of taurine and SsuEADCB being an uptake and desulfonation system for a wide range of alkanesulfonates other than taurine.

In the systems examined in this work, deletion strains that produced the substrate-binding protein of one transporter and the membrane component of the other did not grow with the sulfonates transported by either TauABC or SsuABC. Also, the formation of productive hybrid transporters in which the membrane component would be composed of the ATP-binding protein (SsuB or TauB) of one transporter and the membrane-spanning protein (SsuC or TauC) of the other transporter did not occur. The failure of hybrid transporters to support growth extended to alkanesulfonates that were common substrates of both wild-type transporters. It therefore appears to be due to the lack of interaction between the individual components of hybrid systems rather than to restrictions imposed by substrate specificity. An exception was noted when propanesulfonate or butanesulfonate was offered as a sulfur source. These compounds enabled growth, at respectable levels, of strains EC1250 tauA ssuA and EC1250 tauBC ssuCB as well as of mutants producing hybrid transporters (Table 3). In addition to being transported by the TauABC and/or the SsuABC system, propanesulfonate and butanesulfonate appear to enter the cell by an as-yet-unknown mechanism. Since both compounds are fully ionized under physiological conditions (8), their entrance into the cytoplasm by passive diffusion seems unlikely.

There are few reports on functional exchangeability of the components of ABC transporters in the literature. ATP-binding proteins of ABC transporters have domains in common that were shown to be functionally interchangeable in the recombinant hybrid protein HisP-MalK of Salmonella typhimurium. This protein was fully active in maltose transport but failed to complement a hisP deletion mutation (17). This observation indicated that HisP and MalK share domains that are exchangeable, whereas other domains confer system specificity by providing for interactions between the membrane-spanning protein and the ATP-binding protein (12, 17). Furthermore, complete ATP-binding proteins have successfully been exchanged between the Mal and Ugp transporters, which in E. coli transport maltose and sn-glycerol-3-phosphate, respectively (5). The N-terminal parts of the UgpC and MalK proteins exhibit approximately 60% sequence identity, which is much higher than the 35.6% sequence identity over the first 150 amino acids between the TauB and SsuB proteins.

Periplasmic binding proteins for sulfonates and sulfate esters from E. coli, B. subtilis, Pseudomonas aeruginosa, and P. putida share between 22 and 45% sequence identity (25). They are less than 15% identical to other solute-binding proteins and form a family of binding proteins separate from those previously defined by Tam and Saier (19). TauA and SsuA are members of this family, but they exhibit only 22.7% sequence identity and were found to be unable to substitute for each other (Table 3). This is in contrast to the E. coli uptake systems for sulfate and thiosulfate (18) and lysine-arginine-ornithine and histidine (13), where two binding proteins of overlapping substrate specificity interact with a unique membrane component. The sulfate- and thiosulfate-binding proteins show 45% sequence identity, and the lysine-arginine-ornithine- and histidine-binding proteins are 70% identical.

The proteins of the two postulated independent transport systems of E. coli for taurine on the one hand and alkanesulfonates on the other exhibit low sequence identity but share a common physiological role. As documented by the mechanisms regulating gene expression in these systems, they both are involved in scavenging sulfur for growth. In both systems gene expression is controlled by the transcriptional activator Cbl, whose synthesis is governed by CysB, the transcriptional activator of the cys regulon (22, 24). As we have shown, the two systems also transport structurally similar compounds and partially overlap with respect to substrate range. The genes of both the tau and ssu operons do not belong to those E. coli genes believed to be acquired by lateral transfer events during the past 100 million years (9). However, in view of the above-described considerations, the driving force for evolving two separate transporters with similar function in the same organism is not evident.