| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Microbiology, University of Stellenbosch, Private Bag X1, Matieland 7602, South Africa
Correspondence
Douglas E. Rawlings
der{at}sun.ac.za
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The GenBank/EMBL/DDBJ accession number for the sequence of the insert of pAtcars4 reported in this paper is DQ810790.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Together with the iron-oxidizing bacterium Leptospirillum ferriphilum, it has been reported to dominate the microbial populations in commercial arsenopyrite concentrate bio-oxidation tanks that operate at 40 °C and pH 1.6 (Rawlings et al., 1999). These vigorously aerated tanks are used in a pretreatment process to decompose and open up the structure of gold-bearing arsenopyrite concentrates to facilitate the extraction of the gold by cyanide (Rawlings et al., 2003).
Southern hybridization experiments using the arsB genes from the Escherichia coli plasmid R773, or the arsC gene of Acidithiobacillus ferrooxidans (Butcher et al., 2000) as probes have indicated that all strains of At. caldus tested contain a set of chromosomal arsenic resistance genes (Dopson et al., 2001; Tuffin et al., 2004). Studies on At. caldus strain KU have shown that arsenic resistance is inducible by arsenite [As(III)], arsenate [As(V)] and antimony, resulting in an energy-dependent decrease in accumulation of these toxic ions (Dopson et al., 2001). This, together with the demonstration of arsenate reductase activity, suggests that At. caldus contains an arsenic resistance mechanism similar to those described for other bacteria. However, the chromosomal arsenic resistance genes have not been isolated.
The large quantities of toxic arsenic compounds released during the bio-oxidation of arsenopyrite concentrates (Dew et al., 1997) result in the acquisition, by the At. caldus isolates found in these tanks, of an additional set of arsenic resistance genes that are not present in the less arsenic-resistant isolates of the same species (Tuffin et al., 2005). These additional genes are located on an unusual Tn21-like transposon, TnAtcArs, which is transpositionally active in laboratory strains of E. coli, and appears to be dedicated to the spread of arsenic resistance.
Here, we report the cloning and characterization of the chromosomal arsenic resistance genes present in all strains of At. caldus. We examined the levels of arsenic resistance conferred by both the chromosomal and TnAtcArs ars genes when cloned in laboratory strains of E. coli, independently and when present together in the same host. Furthermore, we tested whether there was any interaction between the two arsenic resistance operons at the regulation level. We transferred the arsenic resistance genes present on TnAtcArs by conjugation from E. coli to At. caldus, and demonstrated increased arsenic resistance in the recipient At. caldus. This is believed to be the first report of the transfer of genes to At. caldus, and could serve as a basis for the development of a genetic system for the bacterium.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
. pEcoR252 (Zabeau & Stanley, 1982), pBluescript SK (Stratagene), pGem-T (Promega), pGL10 (Kmr, RK2/RP4 replicon, cloning vector; A. Toukdarian, University of California, San Diego), pUCBM21 (Boehringer Mannheim), pMC1403 (Casadaban et al., 1983), pKK223-3 (Pharmacia), pSa (Tait et al., 1982) and pEcoBlunt (Tuffin et al., 2005) have been described previously. E. coli strains (DH5
, Promega; ACSH50Iq, Butcher & Rawlings, 2002; HB101, Bolivar et al., 1977; CSH56, Cold Spring Harbor Laboratory) were grown in Luria–Bertani (LB) broth medium (Sambrook et al., 1989), with ampicillin (100 µg ml–1) or kanamycin (100 µg ml–1) added as required. At. caldus strains were grown at 37 °C in tetrathionate medium (5 mM), sterilized and adjusted to pH 2.5 (Rawlings et al., 1999); sodium thiosulphate medium (5 mM), sterilized and adjusted to pH 4.6 (sodium thiosulphate replaced the tetrathionate); or elemental sulphur (S0) medium, sterilized and adjusted to pH 4.6. For the S0 medium, elemental sulphur was sterilized by adding 0.5 g S0 to 5 ml water, and heating to 105 °C for 1 h on two successive days. The sterilized sulphur was then added to the basal medium (0.5 g l–1).
|
). Total DNA was extracted from At. caldus strain #6, which had been isolated from the arsenopyrite bio-oxidation plant at the Fairview Mine, Barberton, South Africa. For the construction of the At. caldus strain #6 gene bank containing large inserts, chromosomal DNA was isolated as follows. At. caldus cells were harvested by centrifugation, washed three times in acidified water (pH 1.8), and resuspended in Tris/EDTA (TE) buffer, pH 7.6. Lysis was with 1 % SDS in the presence of proteinase K (1 mg ml–1) at 37 °C. Proteins were precipitated by centrifugation in the presence of 2 M ammonium acetate before DNA was precipitated with ethanol, washed twice in 70 % ethanol, and resuspended in TE buffer (pH 7.6). This DNA was partially digested with Sau3A, the fragments separated using a sucrose gradient, and fragments in the 10–25 kb size range were ligated into the BglII site of the positive-selection cloning vector pEcoR252. Approximately 9600 colonies were obtained by transforming the ligation mixture into E. coli DH5, and selecting for growth on Luria agar (LA) plus ampicillin (100 µg ml–1). These colonies were scraped from an LA plate, and used to prepare the At. caldus gene bank. Sequencing was by the dideoxy chain-termination method, using an ABI PRISM 377 automated DNA sequencer, and the sequence was analysed using a variety of software programs, but mainly a combination of the Glimmer 2 (www.tigr.org/softlab; Delcher et al., 1999) and DNAMAN (Lynnon BioSoft) programs. Comparison searches were performed using the gapped-BLAST program at the National Centre for Biotechnology Information (NCBI, http://www.ncbi.nlm.nih.gov; Altschul et al., 1997). The nucleotide sequence for the insert of pAtcars4 was deposited in the GenBank database under the accession number DQ810790.
PCR.
PCR was performed using the primers described in Table 2, with 50 ng plasmid DNA in a 50 µl volume containing 2 mM MgCl2, 0.25 µM each primer, 200 µM each dNTP, and 1 U TaqI polymerase. Reactions were carried out in a Hybaid Sprint thermocycler, with an initial denaturation at 94 °C for 60 s, followed by 25 cycles of denaturation (30 s at 94 °C), an annealing step of 30 s, and a variable elongation step at 72 °C. Annealing temperatures and elongation times were altered as required.
|
). Assays performed in E. coli ACSH50Iq containing plasmids were carried out in LB medium containing appropriate antibiotics and various concentrations of sodium arsenite. Growth assays to determine the resistance to arsenate were performed in low-phosphate medium (Oden et al., 1994) supplemented with 2 mM K2HPO4. Overnight cultures were diluted 100-fold into fresh medium, incubated at 37 °C for 5 h, and the OD600 was determined. The incubation time used corresponded to the middle of the exponential growth phase of controls under the same conditions. In all cases, the resistance was expressed as the percentage OD600 compared with that of the control culture with no added arsenic.
Construction of the promoter–lacZ fusions.
The promoter regions for arsR and arsB were amplified by PCR using the primer pairs ChArsRLacZF/ChArsRLacZR for arsR, and ChArsBLacZF/ChArsBLacZR for arsB (Table 2
). The PCR products were digested with BamHI/EcoRI, and ligated to the promoterless lacZ reporter gene of pMC1403. Fusions were confirmed by DNA sequencing.
-Galactosidase assays.
Overnight cultures were diluted 1 : 200 into fresh medium containing the appropriate antibiotics, 0.4 mM IPTG, sodium arsenate or sodium arsenite (25 µM), when indicated, and were incubated at 30 °C to OD600 0.5. The -galactosidase activity was measured using the method of Miller (1972).
RNA analysis and RT-PCR.
Total RNA was isolated, as described by Trindade et al. (2003), from 50 ml mid-exponential-phase cultures of E. coli ACSH50Iq carrying various plasmids, grown in LB medium containing 0.1 mM arsenate, 0.1 mM arsenite or no arsenic, and with antibiotic selection. RNA was also isolated from At. caldus #6. Cells were first grown in tetrathionate medium without arsenate, and then diluted 100-fold into fresh medium containing 0.1 mM arsenite or no arsenite. RNA was isolated using the same method as for E. coli. RNA was separated on a 1 % formaldehyde denaturing gel according to standard procedures (Sambrook et al., 1989), transferred to Hybond-N+ nylon membranes (Amersham), and hybridized according to the manufacturer's instructions, using DIG-labelled DNA probes specific for the arsA and tnpA transcripts. The probes were synthesized using either the PCR DIG Probe Synthesis kit or the Random Primed DNA Labelling kit (Roche), using the primer combination RNAarsAF/RNAarsAR for arsA, and TnpAF/pEcoF for tnpA (Tuffin et al., 2005).
For RT-PCR, the First Strand cDNA Synthesis kit (AMV; Roche) was used for cDNA synthesis and product detection. The protocol of the manufacturer was used for the reverse transcriptase reaction. The PCR was performed as described above, using 2 of the 20 µl (total volume) of the reverse transcriptase reaction, and the extension times were altered as required for the different primer pairs. Primer combinations used for cDNA synthesis are shown in Fig. 1. To detect DNA contamination in the mRNA extracts, reactions were performed with each primer pair without any AMV reverse transcriptase.
|
); and (b) 1.5 % agar; these were combined in equal volumes after autoclaving and cooling to 50 °C, after which, 2 % (w/v) sodium thiosulphate was added. The solid medium was supplemented with 0.05 % yeast extract when used as a mating medium for E. coli and At. caldus. A 0.22 µm filter (Osmonics) was placed on the solid medium, and 200 µl donor/recipient mixture was spotted onto the filter. The plate containing the filter and mixture was incubated at 37 °C for up to 11 days. The filter was washed in 5 ml 0.8 % NaCl solution, and vigorously shaken to dislodge the cells. Cells were pelleted and resuspended in 1 ml 0.8 % NaCl solution. For the E. coli/E. coli matings, the solution was plated onto LA medium, with 30 µg kanamycin ml–1 and 25 µg nalidixic acid ml–1 to select for transconjugant cells, while the E. coli/At. caldus mating solutions were plated onto sodium thiosulphate solid medium, without yeast extract, plus 30 µg kanamycin ml–1. The E. coli/At. caldus mating solutions were also inoculated into sodium thiosulphate liquid medium with 100 µg kanamycin ml–1 to select for transconjugant cells. |
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2005) previously isolated from At. caldus. Several of these plasmids were mapped in detail, and two, pAtcars4 and pAtcars6 containing 9.7 and 8.6 kb inserts, respectively, were chosen for further study (Fig. 1). Confirmation that the insert originated from At. caldus was obtained by Southern hybridization. An internal 1.3 kb partial NcoI fragment (Fig. 1) was labelled and used to probe the chromosomal DNA of At. caldus strain #6 digested with StuI, BamHI, NcoI and SalI. Hybridization signals for 0.4 and 0.7 kb BamHI fragments, a 4.7 kb StuI fragment, 0.4 and 0.85 kb NcoI fragments, and a 9.3 kb SalI fragment were obtained for both the chromosomal DNA of At. caldus strain #6 and plasmid insert (data not shown). This indicates that the insert DNA originated from At. caldus strain #6, that it was present in a single copy, and that no rearrangements in the region over which restriction endonuclease sites extended had occurred during cloning.
Sequence analysis of the At. caldus chromosomal ars genes
The entire insert of pAtcars4 was sequenced in both directions, and 10 ORFs were identified (Fig. 1). Three genes were found whose products were clearly related to previously identified arsC, arsR and arsB gene products. Unlike most ars operons, in which the arsR and arsC genes are transcribed together with the arsB gene, the At. caldus chromosomal arsRC and arsB genes were divergently transcribed in a manner previously found for only one other bacterium, At. ferrooxidans (Butcher et al., 2000). Alignments with other proteins in the NCBI database revealed that At. caldus chromosomal ArsB, ArsC and ArsR were most closely related to those of the At. ferrooxidans ars operon (85, 78 and 74 % amino acid sequence identity, respectively). The structure of the At. caldus chromosomal ars operon differed substantially from that of the TnAtcArs transposon isolated from the same bacterium, which contained ars genes R, C, D1, A1, D2, A2, ORF7, ORF8 and B, all transcribed in the same direction. The amino acid sequences of At. caldus ArsB, ArsC and ArsR chromosomal gene products and the equivalent proteins from the TnAtcArs were substantially different, with 60, 72 and 45 % amino acid sequence identity, respectively. In addition to the ars genes, seven other ORFs were identified using the Glimmer 2 program. The predicted product of ORF1 had high amino acid sequence identity (76 %) to a hypothetical protein from Polaromonas naphthalenivorus, and was transcribed in the same direction as arsRC, with only 10 bp between the stop codon of ArsC and the predicted start codon of ORF1. Likewise, ORFs 5 and 6 were transcribed in the same direction as arsB, with 30 bp between the stop codon of ArsB and the predicted start codon of ORF5, and 35 bp between ORFs 5 and 6. Although the predicted ORF5 product was most closely related to a hypothetical protein from Legionella pneumophila, the two proteins were of very different sizes (130 and 402 aa, respectively), and the region of high identity (60 %) was only 41 aa in size. The predicted amino acid sequence of ORF6 was 43 % identical to that of a hypothetical protein of Alkalilimnicola ehrlichei of equivalent size. ORFs 7–10 appeared to form an operon consisting of a transcriptional regulator and three other genes (Fig. 1), but as deletion of this region did not have an effect on arsenic resistance, they were not studied further.
At. caldus arsCRB confers arsenic resistance to E. coli that is not enhanced by co-transcribed ORFs 1 and 5
Because of the location and direction of transcription of ORFs 1, 5 and 6, we tested whether they contributed to arsenic resistance in E. coli. Constructs pAtcars4, pAtcars6, pAtcarsCRB5, pAtcarsCRB, pAtcarsRB, pAtcarsRB5 and pEcoBlunt (vector control) were transformed into the E. coli arsenic mutant ACSH50Iq, and transformants were tested for their ability to grow in LB medium plus 0.25 mM arsenite and 0.5 mM arsenate. The chromosomal ars operon of At. caldus (pAtcars4) conferred marked resistance to the E. coli ars mutant ACSH50Iq in both arsenate and arsenite compared to the negative control pEcoBlunt (Fig. 2A, B). Constructs pAtcars6, pAtcarsRB and pAtcarsRB5 conferred lower resistance to arsenate than did constructs containing arsC, as a functional arsC is required for the conversion of arsenate to arsenite. There were no consistent differences in arsenic resistance in cells harbouring only arsCRB and ORF5 (pAtcarsCRB5) or arsCRB (pAtcarsCRB) compared with pAtcars4, suggesting that neither ORF1 nor ORFs 6–10 are required for arsenic resistance in E. coli ACSH50Iq.
|
), a 522 bp fragment was obtained from the RNA samples extracted from E. coli ACSH50Iq grown in the presence or absence of arsenite and arsenate (data not shown). This indicates that ORF1 was co-transcribed with arsC. When using the ArsC-RT_Rev/ArsR-RT_Fw primer combination, a 423 bp fragment was obtained, indicating that arsC was co-transcribed with arsR (data not shown). A 451 bp product was obtained when primers ArsB-RT_Fw/ORF5-RT_Rev were used, indicating that arsB and ORF5 were co-transcribed (data not shown). However, no 473 bp product was obtained using primers ORF5-RT_Fw/ORF6-RT_Rev, indicating that co-transcription of ORF5 and ORF6 did not occur (data not shown). No products were obtained in the absence of AMV reverse transcriptase, indicating that the mRNA was DNA free. This suggests that ORF1 and ORF5 were co-transcribed with arsRC and arsB, respectively, even though no changes in resistance to arsenate or arsenite could be detected in E. coli when ORF1 or ORF5 were deleted. Similarly, RNA was extracted from At. caldus strain #6 grown in the presence or absence of arsenite, and ORF1–arsRC and ORF5–arsB co-transcription was confirmed. The products of ORF1 and ORF5 may play a role in arsenic resistance in the natural host, but in the absence of an effective genetic system for At. caldus, this has not been tested.
Regulation of divergent arsR and arsB of At. caldus and cross-regulation by TnAtcArs
Using the promoterless lacZ gene on pMC1403, we constructed in-frame translational arsR–lacZ (pChArsRLacZ) and arsB–lacZ (pChArsBLacZ) fusions to serve as reporter constructs for measuring the transcription of the bi-directional At. caldus chromosomal ars operon. When transformed into the E. coli ars mutant ACSH50Iq, pChArsRLacZ and pChArsBLacZ gave -galactosidase activity averaging 12 and 48 miller units, respectively (Fig. 3A, B). This suggests that transcription of the cloned genes from the promoter region was stronger in the direction of arsB than that of arsRC. The effect of the arsR gene product on the expression of the arsR–lacZ and arsB–lacZ fusions was determined by cloning the arsR gene behind a tac promoter in the compatible vector ptacGL (tac promoter of pKK223.3 cloned into pGL10). This construct (ptacChArsR) was placed in trans with fusions pChArsRLacZ and pChArsBLacZ in the E. coli ars mutant ACSH50Iq. To determine the level of arsenic required for induction of the two reporter gene constructs, -galactosidase assays were performed without arsenic, or with 5, 25 or 100 µM arsenate or arsenite added (data not shown). Addition of 25 µM of either arsenite or arsenate was generally sufficient for full induction, while at 100 µM, cell growth was negatively affected, probably due to the absence of a functional ars operon. No significant difference in induction between arsenite and arsenate was observed. In low-phosphate medium, arsenate was able to induce the ars operon more efficiently at lower concentrations (5 µM) than it did in LB medium (data not shown). It is known that the transport of arsenate into bacterial cells is via some of the phosphate transport systems (Sanders et al., 1997), and in low-phosphate medium, arsenate is likely to be taken up more effectively, due to reduced competition by phosphate.
|
10 % of that of the fully induced operon. In contrast, the arsR–lacZ fusion was regulated less effectively by ArsR, with expression being reduced to 50 % of fully induced levels in the absence of arsenic (Fig. 3A, B).
As described earlier, the amino acid sequences of the At. caldus chromosomal ArsR and the TnAtcArs ArsR shared only 45 % sequence identity, and highly arsenic-resistant strains of At. caldus have both arsenic resistance systems. We determined how efficiently the heterologous TnAtcArs ArsR was able to regulate the arsR–lacZ (pChArsRLacZ) and arsB–lacZ (pChArsBLacZ) fusions in E. coli. The more strongly expressed arsB–lacZ fusion was repressed approximately 8–12-fold by its own ArsR in the absence of arsenite or arsenate, whereas it was not repressed in the presence of the heterologous ArsR (Fig. 3A). The more weakly expressed arsR–lacZ fusion was repressed approximately twofold by its own ArsR, but not by the heterologous ArsR (Fig. 3B).
TnAtcArs is an operon transcribed in one direction from arsR (Tuffin et al., 2005), and a TnAtcArs arsR–lacZ fusion was available (pTnArsRLacZ) from that study, which enabled us to carry out the reverse experiment. The transposon arsR–lacZ fusion was expressed at a very much higher level (>1200 units) than either of the chromosomal ars promoters (Fig. 3C), and was more effectively repressed by its own ArsR regulator (fourfold) than by the heterologous chromosomal ArsR (twofold).
Combined chromosomal ars and TnAtcArs arsenic resistance in E. coli and At. caldus
Highly resistant At. caldus strains have both chromosomal and transposon-located (TnAtcArs) ars genes in the same cell; therefore, we tested which system conferred the greater arsenic resistance, and whether placement of both ars systems in E. coli resulted in increased resistance to arsenic. The TnAtcArs genes were cloned into the low-copy-number vector pGL10 to produce pTnArs1GL, which is compatible with pEcoR252 used to construct pAtcarsCRB5. pTnArs1GL and pAtcarsCRB5 were transformed into the E. coli ars mutant ACSH50Iq, individually and together, and growth was assayed in the presence or absence of 1.5 mM arsenate, with antibiotic selection for plasmids (data not shown). In the absence of arsenate, growth of cells containing individual plasmids or the combination of plasmids was approximately the same, suggesting that the presence and selection of two types of plasmid in one cell did not inhibit cell growth. Growth in the presence of arsenate indicated that E. coli containing the transposon ars genes (pTnArs1GL) was more resistant than the strain containing the chromosomal ars genes (pAtcarsCRB5), in spite of the transposon genes being on a lower copy number vector. Arsenate resistance in E. coli cells containing both ars systems was intermediate between that in cells containing the two individual ars systems, although the small reduction in resistance was within the error associated with the assay.
To test what change in arsenic resistance occurred when At. caldus gained a transposon-located ars system, we developed a conjugation system for At. caldus, as no genetic system for this bacterium has been reported. During experiments to demonstrate that TnAtcArs is transpositionally active in E. coli, a broad-host-range plasmid pSa, into which TnAtcArs had been transposed (pTn2), and that can be conjugated between strains of E. coli, has been isolated (Tuffin et al., 2005). E. coli HB101 cells containing either pTn2 or pSa were mated with At. caldus C-SH12 cells lacking the TnAtcArs system. After mating, At. caldus cells capable of growth in sodium thiosulphate liquid medium (without yeast extract) plus 100 µg kanamycin ml–1 were isolated from matings with E. coli containing either pTn2 or pSa. After subculturing in fresh sodium thiosulphate medium, At. caldus total DNA was extracted, and the TnAtcArs arsA gene was PCR-amplified using transposon arsA primers (RNAarsAF/RNAarsAR; Tuffin et al., 2005). The predicted 600 bp product was observed when using At. caldus C-SH12 (pTn2) total DNA as a template, but not with At. caldus C-SH12 (pSa) total DNA (data not shown). To confirm that the transconjugants were At. caldus and not some other thiosulphate-oxidizing bacterium, the 16S rRNA genes were amplified and gave the predicted diagnostic bands when digested with BamHI and StuI (Rawlings et al., 1999). To further confirm the presence of TnAtcArs in At. caldus C-SH12, Southern hybridization experiments were carried out using total DNA isolated from At. caldus strains C-SH12, C-SH12 (pTn2) and C-SH12 (pSa) (Fig. 4). When the arsD gene of TnAtcArs was used as a probe, only DNA from C-SH12 (pTn2) gave a positive hybridization signal, whereas, with pSa as a probe, DNA from both C-SH12 (pTn2) and C-SH12 (pSa) gave a positive signal.
|
30 mM arsenite (Fig. 5), while At. caldus C-SH12 (pTn2) grew at all concentrations tested. These results indicate that the addition of the transposon ars operon of At. caldus strain #6 to At. caldus C-SH12 dramatically increased its resistance to arsenite.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Tuffin et al., 2005). These ars genes and their predicted proteins were most similar to those of At. ferrooxidans, a mesophilic iron- and sulphur-oxidizing bacterium. The gene order and number of ars genes varies between ars operons, with two of the most commonly encountered sets of ars genes being the arsRBC genes, such as those found on the chromosome of E. coli (Carlin et al., 1995), and the arsRDABC genes, such as those found on plasmid R773 (Chen et al., 1986). Furthermore, transcription is usually unidirectional. The ars operons of both of these bacteria are unusual in that they are bi-directional, with the arsB gene (arsenite efflux pump) transcribed in the opposite direction to that of the arsR (regulator) and arsC (arsenate reductase) genes. This raises the possibility that transcription in the two directions might be regulated differently. Reporter gene studies in E. coli indicated that the level of transcription in the direction of the At. caldus arsB was four- to fivefold higher and subject to greater repression by ArsR than that in the direction of arsRC. However, there was no detectable difference in the ability of arsenite or arsenate to derepress the transcription of the ars genes in either direction, a finding that is similar to that of ArsR from At. ferrooxidans (Butcher & Rawlings, 2002). The amino acid sequences of the two ArsR repressors are 74 % identical, and both belong to a family of ArsR regulators that lack a consensus arsenite binding sequence (ELCVCDL) found in the more extensively studied ArsR family (Shi et al., 1994).
In spite of having bi-directional transcription and related arsRC and arsB genes, the At. caldus and At. ferrooxidans ars operons have some substantial differences. Unlike the At. ferrooxidans ars operon, no arsH was located downstream of the At. caldus arsB, and the flanking ORFs were different. Although ORF1 and ORF5 of the At. caldus operon were found to be transcribed together with the ars genes in both E. coli and At. caldus, no difference in arsenic resistance in E. coli was found when these genes were deleted. A similar inability to detect a difference is found when the arsH of At. ferrooxidans is deleted (Butcher et al., 2000). Recently, an increase in sensitivity to low levels of arsenite and, to a lesser extent, arsenate has been reported when the arsH of Sinorhizobium meliloti is deleted (Yang et al., 2005). Co-transcription of ORF1 and ORF5 indicates that they are associated with arsenic resistance, although their roles remain uncertain.
Strains of At. caldus from bacterial consortia that have been adapted to grow in the presence of high concentrations of arsenic have at least one additional set of arsenic resistance genes located on TnAtcArs (de Groot et al., 2003; Tuffin et al., 2005). The chromosomal and transposon ars operons are quite different, and although the ArsR regulators belong to the same family, they share only 45 % amino acid sequence identity. We investigated how the two systems were likely to interact when both were present in the same bacterium. Promoter fusion studies in E. coli indicated that each ArsR was a more effective repressor of the promoter of its own operon (both directions in the case of the chromosomal ars) than was the heterologous ArsR; however, although the chromosomal ArsR was able to partly regulate the transposon ars expression, the transposon ArsR was unable to regulate the chromosomal ars operon. The very much higher levels of reporter gene expression from the transposon ars operon (Fig. 3), and much less effective repression, might explain the higher levels of resistance conferred by the transposon ars operon in E. coli.
The conjugation of pSa from E. coli to At. caldus was attempted with At. caldus strains BC13 and KU (data not shown), in addition to At. caldus C-SH12. For some, as-yet-unknown reason, only the latter conjugation was successful. Furthermore, At. caldus C-SH12 formed indistinct colonies on solid selection medium so that it was not possible to calculate a conjugation frequency. No conjugation or any other genetic system has been established for At. caldus; therefore, experiments are currently in progress using this limited, but repeatable, successful plasmid transfer to optimize a conjugation system for At. caldus, using a range of conjugative plasmids.
Nevertheless, of more importance for our study was the demonstration that the transfer of TnAtcArs to At. caldus C-SH12, which previously contained only the chromosomal ars genes, resulted in a marked increase in resistance to arsenite. The increase in resistance was possibly even greater than shown, but arsenite precipitation occurred during extended aeration in the medium, at concentrations above 50 mM. No At. caldus strain lacking the chromosomal ars genes has been found, and we could therefore not determine the relative contributions of the two ars systems to arsenic resistance. However, the experiment clearly demonstrates the advantage to be gained by acquisition of TnAtcArs when At. caldus is required to grow in high concentrations of arsenic.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Bolivar, F., Rodriguez, R. L., Greene, P. J., Betlach, M. C., Heyneker, H. L. & Boyer, H. W. (1977). Construction and characterisation of new cloning vehicles. II. A multipurpose cloning system. Gene 2, 95–113.[Medline]
Butcher, B. G. & Rawlings, D. E. (2002). The divergent chromosomal operon of Acidithiobacillus ferrooxidans is regulated by an atypical ArsR protein. Microbiology 148, 3983–3992.
Butcher, B. G., Deane, S. M. & Rawlings, D. E. (2000). The chromosomal arsenic resistance genes of Thiobacillus ferrooxidans have an unusual arrangement and confer increased arsenic and antimony resistance to Escherichia coli. Appl Environ Microbiol 66, 1826–1833.
Carlin, A., Shi, W., Dey, S. & Rosen, B. P. (1995). The ars operon of Escherichia coli confers arsenical and antimonal resistance. J Bacteriol 177, 981–986.
Casadaban, M. J., Martinez-Arias, A., Shapira, S. K. & Chou, J. (1983). -Galactosidase gene fusions for analysing gene expression in Escherichia coli and yeast. Methods Enzymol 100, 293–308.[Medline]
Chen, C.-M., Misra, T., Silver, S. & Rosen, B. P. (1986). Nucleotide sequence of the structural genes for an anion pump; the plasmid encoded arsenical resistance operons. J Biol Chem 261, 15030–15038.
de Groot, P., Deane, S. M. & Rawlings, D. E. (2003). A transposon-located arsenic resistance mechanism from a strain of Acidithiobacillus caldus isolated from commercial, arsenopyrite biooxidation tanks. Hydrometallurgy 71, 115–123.[CrossRef]
Delcher, A. L., Harmon, D., Kasif, S., White, O. & Salzberg, S. L. (1999). Improved microbial gene identification with GLIMMER. Nucleic Acids Res 27, 4636–4641.
Dew, D. W., Lawson, E. N. & Broadhurst, J. L. (1997). The Biox® process for biooxidation of gold-bearing ores or concentrates. In Biomining: Theory, Microbes and Industrial Processes, pp. 45–80. Edited by D. E. Rawlings. Berlin: Springer.
Dopson, M., Lindstrom, E. B. & Hallberg, K. B. (2001). Chromosomally encoded arsenical resistance of the moderately thermophilic acidophile Acidithiobacillus caldus. Extremophiles 5, 247–255.[CrossRef][Medline]
Goebel, B. M. & Stackebrandt, E. (1994). Cultural and phylogenetic analysis of mixed microbial populations found in natural and commercial bioleaching environments. Appl Environ Microbiol 60, 1614–1621.
Hallberg, K. B. & Lindström, E. B. (1994). Characterization of Thiobacillus caldus sp. nov., a moderately thermophilic acidophile. Microbiology 140, 3451–3456.[Abstract]
Miller, J. H. (1972). Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
Oden, K. L., Gladysheva, T. B. & Rosen, B. P. (1994). Arsenate reduction mediated by the plasmid-encoded ArsC protein is coupled to glutathione. Mol Microbiol 12, 301–306.[Medline]
Rawlings, D. E., Coram, N. J., Gardner, M. N. & Deane, S. M. (1999). Thiobacillus caldus and Leptospirillum ferrooxidans are widely distributed in continuous flow biooxidation tanks used to treat a variety of ores and concentrates. In Biohydrometallurgy and the Environment: Towards the Mining of the 21st Century. Part A, pp. 777–786. Edited by R. Amils & A. Ballester. Amsterdam: Elsevier.
Rawlings, D. E., Dew, D. & du Plessis, C. (2003). Biomineralization of metal-containing ores and concentrates. Trends Biotechnol 21, 38–44.[CrossRef][Medline]
Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
Sanders, O. I., Rensing, C., Kuroda, M., Mitra, B. & Rosen, B. P. (1997). Antimonite is accumulated by the glycerol facilitator GlpF in Escherichia coli. J Bacteriol 179, 3365–3367.
Shi, W., Wu, J. & Rosen, B. P. (1994). Identification of a putative metal binding site in a new family of metalloregulatory proteins. J Biol Chem 269, 19826–19829.
Silver, S., Budd, K., Leahy, K. M., Shaw, W. V., Hammond, D., Novick, R. P., Willsky, J. R., Malamy, M. H. & Rosenburg, H. (1981). Inducible plasmid-determined resistance to arsenate, arsenite, and antimony(III) in Escherichia coli and Staphylococcus aureus. J Bacteriol 172, 424–430.
Tait, R. C., Lundquist, R. C. & Kado, C. I. (1982). Genetic map of the crown gall suppressive IncW plasmid pSa. Mol Gen Genet 186, 10–15.[CrossRef][Medline]
Trindade, M., Abratt, V. R. & Reid, S. J. (2003). Induction of sucrose utilisation genes from Bifidobacterium lactis by sucrose and raffinose. Appl Environ Microbiol 69, 24–32.
Tuffin, M., de Groot, P., Deane, S. M. & Rawlings, D. E. (2004). Multiple sets of arsenic resistance genes are present within highly arsenic resistant industrial strains of the biomining bacterium, Acidithiobacillus caldus. International Congress Series. Int Cong Ser 1275, 165–172.[CrossRef]
Tuffin, I. M., de Groot, P., Deane, S. M. & Rawlings, D. E. (2005). A Tn21-like transposon containing an unusual ars operon is present in highly arsenic resistant strains of the biomining bacterium Acidithiobacillus caldus. Microbiology 151, 3027–3039.
Yang, H.-C., Cheng, J., Finan, T. M., Rosen, B. P. & Battacharajee, H. (2005). Novel pathway for arsenic detoxification in the legume symbiont Sinorhizobium meliloti. J Bacteriol 187, 6991–6997.
Zabeau, M. & Stanley, K. K. (1982). Enhanced expression of the cro--galactosidase fusion proteins under the control of the PR promoter of the bacteriophage lambda. EMBO J 1, 1217–1224.[Medline]
Received 26 June 2006;
revised 22 August 2006;
accepted 22 August 2006.
This article has been cited by other articles:
|
|
 |
|
  L. J. van Zyl, J. M. van Munster, and D. E. Rawlings Construction of arsB and tetH Mutants of the Sulfur-Oxidizing Bacterium Acidithiobacillus caldus by Marker Exchange Appl. Envir. Microbiol., September 15, 2008; 74(18): 5686 - 5694. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. J. van Zyl, S. M. Deane, L.-A. Louw, and D. E. Rawlings Presence of a Family of Plasmids (29 to 65 Kilobases) with a 26-Kilobase Common Region in Different Strains of the Sulfur-Oxidizing Bacterium Acidithiobacillus caldus Appl. Envir. Microbiol., July 15, 2008; 74(14): 4300 - 4308. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
| J MED MICROBIOL | ALL SGM JOURNALS | |
, pAtcars4; 
, pAtcars6; 
, pAtcarsCRB5; 
, pAtcarsCRB; 
, pAtcarsRB5; 
, pAtcarsRB; 
, pEcoBlunt. Each data point represents the results of three assays of three independent experiments. Error bars indicate SD.
