| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Microbiology and Immunology, Otago School of Medical Sciences, University of Otago, PO Box 56, Dunedin, New Zealand
Correspondence
Gregory M. Cook
greg.cook{at}stonebow.otago.ac.nz
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Wanner, 1996), Acinetobacter johnsonii (van Veen, 1997), Sinorhizobium meliloti (Bardin et al., 1998; Voegele et al., 1997; Yuan et al., 2006) and Bacillus subtilis (Allenby et al., 2004; Qi et al., 1997). Pi transporters have also been identified in a number of other bacteria, and there are noticeable parallels between the organizations of transport systems in different species.
Under conditions of excess Pi in the environment, bacteria utilize a low-affinity transporter, the Pit system (phosphate inorganic transport). This transporter is constitutively expressed, recognizes metal phosphate as its substrate and is driven by the proton-motive force (
µH+). The Pit system has been studied in the most detail in E. coli (Hoffer et al., 2001; van Veen et al., 1994b), A. johnsonii (van Veen et al., 1993, 1994a) and S. meliloti (Bardin et al., 1998; Voegele et al., 1997).
Bacteria also contain an inducible, high-affinity Pi transporter, which operates under Pi limitation. This is usually the ATP-binding cassette (ABC)-type transport system Pst (phosphate specific transport). It consists of four components, PstSCAB, where PstS is the substrate-binding protein, PstC and PstA are transmembrane proteins, and PstB is the ATPase. Pst recognizes Pi, rather than metal phosphate, as its substrate. Well characterized Pst systems include those of E. coli, S. meliloti and B. subtilis (Allenby et al., 2004; Qi et al., 1997; Wanner, 1996; Yuan et al., 2006).
Additionally, many bacteria are able to utilize phosphonates or phosphite as alternative sources of phosphorus. Phosphonates are compounds that contain a direct carbon–phosphorus link, and they can be degraded by E. coli (Metcalf & Wanner, 1991), Klebsiella aerogenes (Imazu et al., 1998), Pseudomonas stutzeri (White & Metcalf, 2004b), S. meliloti (Bardin et al., 1996; Voegele et al., 1997), some streptomycete isolates (Obojska et al., 1999), and other micro-organisms (reviewed by Kononova & Nesmeyanova, 2002). In E. coli and K. aerogenes, phosphonates and phosphite are taken up into the cells by the ABC-type Phn system, consisting of the three components PhnCDE, where PhnC is the ATPase, PhnD the substrate-binding protein, and PhnE the permease. Pi is a non-specific substrate for this system (Metcalf & Wanner, 1991). S. meliloti utilizes a high-affinity ABC-type transport system, PhoCDET, for the uptake of both Pi and phosphonates. PhoC and PhoD are homologous to PhnC and PhnD, respectively, while PhoE and PhoT both encode transmembrane proteins homologous to PhnE (Bardin et al., 1996; Voegele et al., 1997).
The genes encoding high-affinity Pi transport systems, such as the pst genes, or for transport systems for alternative phosphorus sources, such as the phn genes, are generally induced when the organism encounters Pi-limited conditions (Bardin & Finan, 1998; Qi et al., 1997; Sola-Landa et al., 2005; Wanner, 1996; White & Metcalf, 2004a). This induction is mediated by the Pi-responsive two-component regulatory system, PhoBR in Gram-negative micro-organisms (Bardin & Finan, 1998; von Kruger et al., 1999; Wanner & Chang, 1987), or PhoPR in Gram-positives (Hulett et al., 1994; Perez et al., 2001; Sola-Landa et al., 2003). In both cases, PhoR is a membrane-bound sensor and PhoP or PhoB are the response regulators. In E. coli and Mycobacterium smegmatis, repression of the Pho response requires PstS, and mutations in pstS lead to constitutive expression of Pho regulon genes (Kriakov et al., 2003; Wanner, 1996).
Mycobacterium tuberculosis contains several copies of genes for components of a Pst system, encoding three different PstS, two PstC and PstA each, and one PstB protein (Braibant et al., 1996a). Other slow-growing species of mycobacteria also contain several copies of pst genes (Braibant et al., 1996a; Lefèvre et al., 1997). Mutants in some of these genes have been created and they were defective in Pi transport and had reduced virulence in vivo (Collins et al., 2003; Peirs et al., 2005). The reason for the presence of several high-affinity Pi transport systems in pathogenic mycobacteria is not understood. The fast-growing M. smegmatis contains only a single copy of the pst genes, organized as a pstSCAB operon, and mutants in these genes display reduced Pi uptake and an increased requirement for Pi for growth (Bhatt et al., 2000; Kriakov et al., 2003). Pi transport by mycobacteria is poorly characterized, and no detailed studies are available on the requirements for Pi for the growth of these bacteria. In a previous study by our group, a three-gene operon (phnDCE) of M. smegmatis with homology to the Phn phosphonate/phosphate transport systems of other bacteria was identified, but no function was assigned to these genes (Tran et al., 2005).
The aim of this study was to determine whether the putative PhnDCE transport system of M. smegmatis has a role in the uptake of Pi and phosphonates, and to compare the characteristics of the putative Phn system to those of the Pst system. The expression of the phn and pst operons in response to Pi limitation was investigated using transcriptional fusions to lacZ. Mutants of both systems were created and characterized regarding their growth at different Pi concentrations, as well as kinetics of Pi uptake into whole cells.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
. E. coli strains were grown in Luria–Bertani (LB) medium at 37 °C with agitation (200 r.p.m.). M. smegmatis strain mc2155 (Snapper et al., 1990) and derived strains were routinely grown at 37 °C, 200 r.p.m. in LB containing 0.05 % (w/v) Tween 80 (LBT) or in Middlebrook 7H9 medium (Difco) supplemented with 10 % albumin-glucose [dextrose]-catalase enrichment (ADC; Becton Dickinson) and 0.05 % (w/v) Tween 80. M. smegmatis transformants were grown at 28 °C for propagation of temperature-sensitive vectors and at 40 °C for allelic-exchange mutagenesis. For Pi limitation studies, M. smegmatis was grown in modified Sauton's medium (ST). The composition of this medium per litre was as follows: 0.5 g MgSO4, 2 g citric acid, 1 g L-asparagine, 0.3 g KCl, 1 g glycerol, 0.5 g Tween 80, 320 µl 0.5 M FeCl3 and 100 µl 1 M NH4Cl. High-phosphate medium contained 10 mM or 100 mM K2HPO4. For low Pi concentrations (1 mM and less), Pi was added to Pi-free medium from a 0.2 M stock of KH2PO4. As alternative sources of phosphorus, methylphosphonate or phosphite were added to Pi-free medium to final concentrations of 1 mM or 5 mM, and phosphonoacetate was added to final concentrations of 1 mM or 2 mM. Cells used as inocula for low-phosphate media were washed in 2 vols saline/Tween 80 (0.85 % NaCl, 0.05 % (w/v) Tween 80). Selective media contained kanamycin (50 µg ml–1 for E. coli; 20 µg ml–1 for M. smegmatis), gentamicin (20 µg ml–1 for E. coli; 5 µg ml–1 for M. smegmatis) or hygromycin (200 µg ml–1 for E. coli; 50 µg ml–1 for M. smegmatis). Solid media contained 1.5 % agar.
|
DNA manipulation and cloning of constructs.
All molecular biology techniques were carried out according to standard procedures (Sambrook et al, 1989). Restriction or DNA-modifying enzymes and other molecular biology reagents were obtained from Roche Diagnostics or New England Biolabs.
Genomic DNA of M. smegmatis was isolated by a modification of the method of Gonzalez-y-Merchand et al. (1996). In brief, cells grown on LBT agar were resuspended in 200 µl lysis buffer (4 M guanidine thiocyanate, 1 mM 
-mercaptoethanol, 10 mM EDTA, 0.1 %, w/v, Tween 80), snap-frozen in solid CO2/ethanol and heated to 65 °C for 10 min. Snap-freezing/heating was repeated and the cells were then cooled on ice for 5 min. The aqueous phase was extracted twice with chloroform and DNA precipitated by 2-propanol. The pellet was washed once in 70 % ethanol, air-dried and dissolved in deionized water.
To create a transcriptional fusion of the phn operon, a 580 bp PCR product, encompassing approximately 500 bp of DNA upstream of phnD plus 85 bp of its coding region, was amplified using primers PphnF (5'-AAATTTGGGCCCGCACTCAATCCCTGGGCCTT-3') and PphnR (5'-AAATTTGGTACCGCTTGTCGGAGCCCGAACAG-3'). The product was cloned into the ApaI and Asp718 sites of pJEM15 (Timm et al., 1994), creating plasmid pSG10. To create a transcriptional fusion of the pst operon, an 800 bp PCR product, encompassing approximately 560 bp of DNA upstream of pstS plus 200 bp of its coding region, was amplified using primers PpstF (5'-AAATTTGGTACCGTCAACAACGCCGCGTTCTCGTG-3') and PpstR (5'-AAATTTGGTACCCTCGGCGATCGCGTTCTGTTGC-3'). The product was cloned into the Asp718 site of pJEM15, creating plasmid pSG42.
To create a construct for the deletion of phnD, the kanamycin resistance cassette (Kmr), encoded by aphA-3, was amplified from pUC18K (Menard et al., 1993) by PCR using the primers 3'mcspUC (5'-GTTTTCCCAGTCACGACGTT-3') and 5'mcspUC (5'-CACACAGGAAACAGCTATGA-3'). The resulting 850 bp product was digested with SacI and HindIII. PCR products of approximately 1.2 kb flanking the phnD gene of M. smegmatis on either side were amplified using the primers phnDKO1 (5'-AAATTTACTAGTGGGGCTGGCTAACGAATGT-3') with phnDKO2 (5'-AAATTTGAGCTCGCTTGTCGGAGCCCGAACAG-3') (left flank) and phnDKO3 (5'-AAATTTAAGCTTACGGGCTCCGAGAAGTGCAA-3') with phnDKO4 (5'-AAATTTACTAGTGAGAACACCTGGGAGCACA-3') (right flank). The left flank PCR product was digested with SpeI and SacI, the right flank PCR product with HindIII and SpeI. Both flanks and the kanamycin cassette were ligated into the SpeI site of pBluescript II KS (Stratagene). The resulting assembly, left flank/Kmr/right flank, was subcloned as a SpeI fragment into pX33 [pPR23 (Pelicic et al., 1997) carrying a constitutive xylE marker], generating plasmid pSG3. The expected double-crossover event would result in a non-polar deletion–insertion at the phnD locus, eliminating 85 % of the phnD coding sequence in exchange for the kanamycin resistance marker.
To create a deletion construct for pstS, the gentamicin resistance cassette (Gmr), aacC-1, from pBSL141 (Alexeyev et al., 1995) was excised from the plasmid using PstI and NotI. PCR products flanking the pstS gene of M. smegmatis for approximately 780 bp (left flank) and 1 kb (right flank) were amplified using the primers pstSKO1 (5'-AAATTTGGATCCGTCAACAACGCCGCGTTCTCGTG-3') with pstSKO2 (5'-AAATTTCTGCAGCTCGGCGATCGCGTTCTGTTGC-3') and pstSKO3 (5'-AAATTTGCGGCCGCCTTCAAAGAGCGTTTGCTCAC-3') with pstSKO4 (5'-AAATTTTCTAGAACTGGTCGTTGAATTCGCTTG-3'), respectively. The left flank PCR product was digested with BamHI and PstI, the right flank PCR product with NotI and XbaI. Both flanks and the gentamicin cassette were ligated into the BamHI and XbaI sites of pBluescript II KS (Stratagene). The resulting assembly, left flank/Gmr/right flank, was subcloned as a BamHI/XbaI fragment into pPR23, resulting in plasmid pSG30. The expected double-crossover event would result in a deletion–insertion at the pstS locus, eliminating 76 % of the pstS coding sequence in exchange for the gentamicin resistance marker.
Allelic replacement of phnD and pstS was essentially carried out as described by Pelicic et al. (1997). Deletion of pstS was achieved by growing a culture of M. smegmatis carrying pSG30 at 28 °C with agitation (200 r.p.m.) to an OD600 of approximately 0.6 to 0.8, followed by plating onto low-salt LBT plates (2 g NaCl l–1) containing gentamicin and 10 % sucrose at 40 °C, selecting for double-crossover events. Replacement of pstS with the gentamicin marker created strain SG95 (pstS : : aacC-1). For the deletion of phnD, a two-step protocol was used. A culture of M. smegmatis carrying pSG3 was grown at 28 °C as described above and then plated onto LBT solid medium containing kanamycin. Incubation at 40 °C in the presence of kanamycin selected for integration of the whole deletion construct into the chromosome of M. smegmatis via a single-crossover event at either the left or right flank. Colonies that formed a yellow product when exposed to 250 mM catechol due to the presence of the xylE marker were screened by Southern hybridization analysis for correct integration of the construct. One integrant was chosen and grown in 2 ml 7H9/ADC containing kanamycin at 37 °C, 200 r.p.m. to an OD600 of 0.7. Aliquots of this culture were then plated onto low-salt LBT plates containing kanamycin and 10 % sucrose and incubated at 40 °C, to select for a second crossover event resulting in loss of the plasmid and replacement of phnD with the kanamycin marker. Colonies that did not form the yellow product after exposure to catechol were picked onto LBT plates containing kanamycin, and counter-picked onto LBT plates containing gentamicin, to confirm loss of the plasmid backbone. Candidate clones (Kmr, Gms) were screened by Southern hybridization analysis for correct deletion of phnD. Replacement of phnD with the kanamycin marker created strain SG34 (phnD : : aphA-3).
To create a phnD pstS double deletion mutant, strain SG34 was transformed with pSG30 and the mutant was created using a two-step protocol as described above. Gentamicin was used for selection in all steps, and no final counter-selection for the loss of the plasmid backbone was carried out. Replacement of pstS with the gentamicin marker in the background of the phnD deletion strain created strain SG118 (phnD : : aphA-3 pstS : : aacC-1).
For Southern hybridization analysis, SmaI- or EcoRI-digested genomic DNA of putative mutants was separated on a 1 % agarose/TAE gel and transferred onto a nylon membrane (Hybond-N+, Amersham) by vacuum blotting. Probes were labelled by random priming using [
-32P]dCTP (Amersham) and Ready-To-Go DNA-labelling beads (Amersham).
Constructs used for complementation of allelic-exchange mutants were cloned into the integrative E. coli/mycobacteria shuttle vector pUHA267 (Lee et al., 1991). For complementation of the phnD mutation, a 1.5 kb PCR product encompassing the phnD gene plus 500 bp of upstream DNA was amplified by PCR using primers cphnDF (5'-AAATTTGGTACCGCACTCAATCCCTGGGCCTT-3') and cphnDR (5'-AAATTTGGTACCATGACAACGACGTCATCACC-3') and cloned into the Asp718 site of pUHA267, creating plasmid pSG38. For complementation of the pstS mutation, a 1.7 kb PCR product encompassing the pstS gene plus 560 bp upstream DNA was amplified by PCR using primers PpstF and cpstSR (5'-AAATTTGGTACCAAATCCAGCTGACTAGCTGATCG-3') and cloned into the Asp718 site of pUHA267, creating plasmid pSG43.
-Galactosidase and inorganic phosphate assays.
To determine the threshold Pi concentration leading to induction of the phn and pst operon genes, strains carrying pSG10 or pSG42 were grown in medium containing an intermediate Pi concentration (200 µM). The medium was modified from the standard ST medium used in this study to contain higher concentrations of the carbon source (5 g glycerol l–1) and nitrogen source (4 g L-asparagine l–1), providing both nutrients in excess. At various time points, 2 ml to 4 ml of the culture where removed to determine OD600. Cells were then pelleted and cell pellets and supernatants were stored at –20 °C. -Galactosidase activity for the 0 h time point (t0) was determined in cells of the inoculum culture.
To measure -galactosidase activities, cell pellets were resuspended in 1 ml Z-buffer (61 mM Na2HPO4, 39 mM NaH2PO4, 10 mM KCl, 1mM MgSO4) containing 0.05 % (w/v) Tween 80, and OD600 was adjusted to 0.2 in 1.5 ml Z-buffer. Of this suspension, 200, 400 and 800 µl were used in each assay and the volumes increased to 1 ml with Z-buffer. Cells were then permeabilized by adding 40 µl chloroform and 20 µl 0.1 % (w/v) SDS and vortexing for 5 s. Tubes were rested at room temperature for 5 to 10 min and then the reaction was started by the addition of 200 µl 4 mg ml–1 stock-solution of ONPG. Reactions were incubated at room temperature until the assay with the highest activity of -galactosidase showed a pale yellow colour (approx. 8 to 20 min). The reaction was stopped by the addition of 500 µl 1 M sodium carbonate. Cells were removed by centrifugation at 16 100 g for 3 min and the assay quantified spectrophotometrically at a wavelength of 420 nm (A420). A blank control consisted of 1 ml Z-buffer in the assay. Activity of -galactosidase was expressed as Miller units (MU) and was calculated as the increase in A420 per minute, per 1 ml cell suspension (normalized to an OD600 of 1.0) used, multiplied by a factor of 1000.
The Pi concentration in the supernatant was determined by inorganic phosphate assay as described previously (Monk et al., 1991).
Phosphate transport assays.
M. smegmatis was grown in high-phosphate ST medium to an OD600 of approximately 1.0, harvested by centrifugation (7700 g, 20 min), washed in 4 vols buffer (50 mM MOPS pH 7.5, 5 mM MgCl2) containing 0.05 % (w/v) Tween 80, and resuspended to an OD600 of 0.7 in Pi-free ST medium. Cells were then incubated at 37 °C, 200 r.p.m. for 2 h to deplete intracellular stores of Pi and to induce expression of high-affinity Pi transport systems as described previously (Bhatt et al., 2000). The presence of an appropriate energy source in the medium ensured that the cells were energized at the time of the assay. As a control, cells were resuspended to an OD600 of 1.0 and starved of Pi in the presence of 200 µg chloramphenicol ml–1. The final OD600 of each culture was measured prior to the uptake assay. Each assay contained 200 µl cell suspension, 2 µl radioactive Pi stock ([33P]orthophosphate; >92.5 TBq mmol–1; Amersham), and 5 µl unlabelled Pi stock to give a final concentration of 5 µM to 200 µM Pi in the assay mix. 33P was added to give approximately 150,000 to 300,000 c.p.m., resulting in a final concentration of 165 pM to 1.65 nM of radioactive Pi in the assay, depending on the age of the radioactive stock. The cell suspension aliquots (200 µl) were pre-incubated at 37 °C in 5 ml polystyrene test-tubes. Transport was started by addition of 7 µl of the appropriate mix of 33P and unlabelled Pi, and the tubes were incubated again at 37 °C for 0 to 30 seconds. The reactions were stopped by the addition of 2 ml cold 0.1 M LiCl and rapid filtration through 0.45 µm HA MF membrane filters (Millipore), using a vacuum manifold (Millipore) with an applied vacuum of approximately 80 p.s.i. Filters were washed again with 2 ml 0.1 M LiCl, air-dried in 4 ml scintillation vials and covered in 2 ml scintillation fluid (Amersham). The amount of radioactivity taken up by the cells was determined with a 1214 Rackbeta liquid scintillation counter (LKB Wallac) using the 14C-window and counting each vial for 1 min. The amount of Pi taken up by the cells was calculated from the counts on the filters of each time-point relative to a t0 control (i.e. simultaneous addition of 33P and LiCl), the total counts initially added to the assay, and the excess of unlabelled Pi over 33P. Rates of Pi uptake were expressed as nmol Pi min–1 (mg protein)–1. Controls for non-specific binding of 33P were carried out by pre-incubating cells with nigericin and valinomycin (15 µM each) for 30 min as previously described (Rao et al., 2001).
The protein concentration in suspensions of whole cells of M. smegmatis was determined as described previously (Meyers et al., 1998). A cell suspension of OD600 1.0 had a protein content of 64 µg ml–1, and this value was the same for the wild-type and all mutants. Vmax values of Pi transport were determined from Michaelis–Menten plots; Km values were determined from double-reciprocal plots of the same data.
DNA and protein sequence analysis.
Preliminary sequence data for M. smegmatis mc2155 were obtained from the Institute for Genomic Research website at http://www.tigr.org. phnDCE correspond to the loci MSMEG0640 to MSMEG0638, and pstSCAB correspond to MSMEG5750 to MSMEG5753.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). OD600 was a reliable indicator of cell yield of cultures of M. smegmatis as determined by viable cell counts. Increasing Pi concentrations to above 100 µM led to no further increase in growth rate or cell yield (data not shown). Because the substrate concentration in the growth medium can be expected to fall substantially during growth of M. smegmatis, growth rates were determined in the initial phase of exponential growth, i.e. between 14 h and 20 h of incubation. In Fig. 1(b), µ is plotted as a function of Pi concentration, according to the Monod equation. The same data plotted as µ versus µ [Pi]–1, in analogy to an Eadie–Hofstee plot of enzyme kinetics, showed biphasic growth kinetics, indicating a change in the kinetics of Pi utilization of M. smegmatis when Pi concentrations drop below approximately 30 µM. At concentrations above 30 µM, the apparent saturation constant, which is the Pi concentration that supports half-maximal growth rate (Pirt, 1975), was 17 µM, and the maximal growth rate was 0.3 h–1 (2.3 h doubling time). At lower Pi concentrations, the apparent saturation constant was 3.4 µM, and the maximal growth rate was 0.19 h–1 (3.7 h doubling time) (Fig. 1c).
|
Induction of phnD–lacZ and pstS–lacZ occurs under Pi-limited conditions
In order to investigate how expression of the genes encoding the putative phosphonate/phosphate transport system Phn (phnDCE) and the high-affinity Pi transport system Pst (pstSCAB) responds to the Pi concentration in the growth medium, strains of M. smegmatis carrying transcriptional lacZ-fusion constructs of the phn operon (pSG10) and of the pst operon (pSG42) were grown in modified ST medium as described in Methods. The Pi concentration, -galactosidase activities and OD600 were monitored throughout growth (Fig. 2). The Pi concentration in the medium decreased rapidly as the strains entered exponential growth, and Pi was depleted from the medium completely as the cultures reached an OD600 between 1.5 and 2.5. Stationary-phase cultures had an OD600 of 5, suggesting that the bacteria had accumulated considerable intracellular stores of Pi to maintain growth after Pi had been depleted from the medium. For both promoter fusions, -galactosidase activities increased significantly once the Pi concentration in the medium dropped below values of approximately 40 µM. Activities of the phnD–lacZ fusion in pSG10 increased rapidly from values of 4 to 7 MU to reach a maximum value of 185 MU (26-fold induction over t0) in late exponential phase, with a final value of approximately 100 MU in stationary phase (Fig. 2a). Activities of the pstS–lacZ fusion in pSG42 increased from 8 to 16 MU under phosphate-replete conditions to a maximum of 136 MU (17-fold induction over t0) in late exponential phase, with approximately 110 MU in stationary phase (Fig. 2b). These data strongly suggest that both the Phn and Pst systems have a role in Pi transport under Pi-limited growth conditions.
|
were unsuccessful, and we therefore decided to take a two-step approach of plasmid integration and excision (Fig. 3a). Colonies displaying the expected SmaI band pattern for integration of the plasmid via the left flank were readily obtained (Fig. 3b, lane 2) and one such colony was chosen to select for a second crossover event, resulting in the deletion of phnD and creating strain SG34 (Fig. 3b, lane 4). No colonies were obtained that displayed the correct band pattern for integration of the plasmid via the right flank, and the remaining colonies contained a single 8.3 kb band (Fig. 3b, lane 1), suggesting the loss of one copy of the left flank.
|
; Kriakov et al., 2003), a deletion mutant in pstS, encoding the substrate-binding protein of the Pst system, was created. The construct for the replacement of pstS with a gentamicin marker was cloned into pPR23 (Pelicic et al., 1997) as described in Methods, transformed into M. smegmatis mc2155, and knockout mutants were selected as described previously (Pelicic et al., 1997). Replacement of pstS with the gentamicin marker led to the introduction of an additional EcoRI restriction site, resulting in a band shift from 3.5 kb in the wild-type to 1.7 kb in the deletion mutant (strain SG95) in Southern hybridization analysis of EcoRI-digested genomic DNA probed with a radiolabelled PCR product of the left flank of the deletion construct (Fig. 3c
). This hybridization result further confirmed that the genome of M. smegmatis contains only a single copy of the pst operon.
Additionally, a phnD pstS double deletion mutant was created to determine whether M. smegmatis harboured other genes that encode high-affinity Pi transport systems. Strain SG34 was transformed with the pstS deletion construct (pSG30), and selection for the desired mutation was carried out using the two-step approach as described above. Colonies displaying the expected EcoRI band pattern for integration of the plasmid via the left flank (4.3 kb and 1.7 kb) or via the right flank (3.5 kb and 2.3 kb) when probed with a radiolabelled left flank PCR product, were both readily obtained (Fig. 3d, lanes 1 and 2). One such colony was chosen to select for a second crossover event, resulting in strain SG118, which carried deletions of both phnD and pstS (Fig. 3d, lane 4).
Mutants in the Phn and Pst systems require high concentrations of phosphate for growth in minimal medium
Transposon mutants of M. smegmatis in pstS, pstA and pstC have been described previously, and all of these mutants failed to grow on 10 mM Pi in minimal medium (Kriakov et al., 2003). We therefore conducted growth experiments in ST medium to confirm that the non-polar deletion mutant in pstS created in the present study, strain SG95, was also unable to grow on 10 mM Pi in a minimal medium (Fig. 4a). Furthermore, strains SG34 (Fig. 4a) and SG118 (data not shown) also failed to grow under these conditions. In contrast, all mutants showed the same growth characteristics as the wild-type when grown in more complex media such as LBT or ADC-enriched 7H9 medium (data not shown), which is also consistent with previous results obtained for Pi transporter mutants of mycobacteria (Kriakov et al., 2003; Peirs et al., 2005). Transformation of SG34 with pSG38, which supplied a single copy of phnD to the cell, resulted in strain SG106. Transformation of strain SG95 with pSG43, supplying a single copy of pstS, resulted in strain SG120. Both complemented strains, SG106 and SG120, displayed a fully restored wild-type phenotype and were able to grow in ST medium containing as little as 0.1 mM Pi (Fig. 4b), proving the role of phnD and pstS, respectively, in the observed growth characteristics of the mutants.
|
) and to induce the Phn and Pst transport systems (Fig. 2). Pi transport was measured by following the accumulation of [33P]orthophosphate (33P) inside whole cells of M. smegmatis over time. As shown for strain mc2155 in Fig. 5(a), 33P accumulation was linear over a period of approximately 30 s, and no more than 15 % of the total radioactivity added to the reaction was taken up into the cells in that time. In all further experiments, initial rates of transport were therefore determined between 0 and 30 s. Cells that were grown in high-phosphate medium (100 mM Pi) and subjected to Pi starvation in the presence of 200 µg chloramphenicol ml–1 to prevent synthesis of inducible transport systems, showed no measurable uptake of Pi (Fig. 5a), indicating that the rates determined for induced cells were indeed due to the induction and activity of high-affinity transport systems. This further confirms the data obtained for the transcription of the phn and pst operons shown in Fig. 2.
|
), inhibited transport by approximately 86 % compared to an untreated control, demonstrating that accumulation of 33P in M. smegmatis was via an energy-dependent mechanism (Fig. 5b).
Determination of Pi transport kinetics in the wild-type and pstS and phnD mutant strains
In previous studies of Pi transport in mycobacteria, uptake of radioactive Pi isotopes into the cells was measured over 8 to 20 min (Bhatt et al., 2000; Peirs et al., 2005), and most likely reflected steady-state levels of both transport and utilization of Pi. To further characterize Pi transport in M. smegmatis, we therefore chose to determine the kinetic parameters of Pi uptake at different external Pi concentrations. Initial rates of transport were determined over a range of Pi concentrations (5 µM to 200 µM), and plots of rates of transport versus Pi concentration showed that uptake followed Michaelis–Menten kinetics (Fig. 6a). The apparent Km and Vmax values of the wild-type were 40 µM Pi and 170 nmol Pi min–1 (mg protein)–1, respectively. An Eadie–Hofstee plot of the same data was linear (data not shown). We determined the kinetic parameters of Pi transport in the phnD deletion strain, SG34 (Fig. 6b), the pstS deletion strain, SG95 (Fig. 6c), and the double deletion strain, SG118 (Fig. 6d). No significant differences were observed between these mutant strains and the wild-type, i.e. the Km values were in the range 40 to 90 µM Pi, and Vmax values were between 100 and 200 nmol Pi min–1 (mg protein)–1. Moreover, Eadie–Hofstee plots of the same data were linear (data not shown). These results suggest that the Phn and Pst systems have similar affinities for Pi, and that the loss of one of these systems can be compensated for by the remaining system.
|
The phosphate transport systems of M. smegmatis are not competitively inhibited by phosphonates or phosphite
The Phn-like systems characterized to date recognize a variety of phosphonates, as well as phosphite, as substrates and are thought to transport Pi non-specifically (Imazu et al., 1998; Metcalf & Wanner, 1991; Metcalf & Wolfe, 1998; Voegele et al., 1997; White & Metcalf, 2004b). The substrate specificity of the high-affinity Pi transport systems of M. smegmatis was determined by testing the ability of alternative substrates (i.e. phosphonates or phosphite) to competitively inhibit Pi uptake at a concentration (1 mM) 10-fold above that of radiolabelled Pi (100 µM) as described previously (Voegele et al., 1997). At a 10-fold excess, any compound recognized as a substrate besides Pi should out-compete radiolabelled Pi for binding by the transport system and therefore strongly reduce the uptake of 33P into the cell. Excess unlabelled Pi (31P) as well as arsenate, a structural analogue of Pi, were included in the assay as positive controls for competitive inhibition. In addition to the wild-type, the mutant strains were also tested for their substrate range. We suggested above that the loss of one or two Pi transport systems can be compensated for by increased activity of the remaining systems. It is therefore conceivable that differences in the substrate specificity of the high-affinity transport systems are more easily detected in strains defective in one or more of these systems. The results are summarized in Table 2. Phosphonoacetate, methylphosphonate or phosphite did not inhibit the initial rate of Pi uptake as compared to the control rate in any of the strains tested and are therefore probably not recognized by any of the high-affinity Pi transport systems. Arsenate inhibited the rate of 33P transport by 86 % to 90 %, and excess unlabelled Pi inhibited transport by 87 % to 92 %, in good agreement with the theoretical value of 90 % inhibition expected for 10-fold dilution of the radiolabelled substrate with unlabelled substrate.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Expression analysis of transcriptional lacZ-fusions of the phnDCE and pstSCAB promoter areas revealed that both promoters under investigation were induced once the Pi concentration in the medium dropped to a value below approximately 40 µM. This Pi concentration is consistent with the threshold value (30 µM Pi) for Pi limitation, as determined from the growth kinetics of M. smegmatis. While Pi starvation induced expression of genes occurs at a similar concentration in B. subtilis (between 50 µM and 80 µM Pi) (Allenby et al., 2005; Hulett & Jensen, 1988), the threshold for induction in Gram-negative bacteria appears to be considerably lower, at about 4 µM Pi in E. coli (Wanner, 1996), and below 10 µM in A. johnsonii (van Veen et al., 1994a).
We created mutants of M. smegmatis, carrying deletions in the genes encoding substrate-binding proteins of both the Phn system (phnD) (Tran et al., 2005), and the Pst system (pstS) (Bhatt et al., 2000; Kriakov et al., 2003), as well as a phnD pstS double mutant. All mutants were unable to grow in ST medium containing 10 mM Pi, while they were not impaired in growth in more complex media such as 7H9/ADC or LBT. Kriakov et al. (2003) isolated M. smegmatis transposon mutants in pstS, pstC and pstA, and these also failed to grow in minimal medium (‘home-prepared’ 7H9) with 10 mM Pi. Mutants of M. tuberculosis in pstS1 and pstS2 were not tested for growth in minimal medium, but had no growth defect in ADC-enriched 7H9 medium (Peirs et al., 2005). These data are difficult to interpret, as no two groups used the same media for characterization of their mutants. It appears that growth of the mutants depends strongly on the complexity of the medium and that the growth defects are not observed in more complex media. This growth defect in minimal medium cannot be explained by a lack of high-affinity Pi transport, because the mutants were not impaired in Pi uptake, as discussed below. However, mutations in genes of the pst operon have been shown to cause constitutive expression of Pho regulon genes in M. smegmatis (Kriakov et al., 2003). Constitutive activation of the Pho response leads to induction of stationary-phase genes in E. coli (Ruiz & Silhavy, 2003) and to initiation of sporulation in B. subtilis (Hulett, 1995). If a similar mechanism exists in M. smegmatis, it could be speculated that hyper-induction of the Pho response due to deletion of a high-affinity Pi transport system causes a stationary-phase response and thereby prevents growth of the bacterium, unless such a response is overridden by components of a complex growth medium. Future work will be directed at investigating this hypothesis. Complementation of the pstS and phnD mutants completely restored the wild-type phenotype, confirming that the observed growth defect was indeed due to the deletion of the target gene, and not to the presence of the antibiotic markers used.
Characterization of Pi transport in the wild-type and Phn and Pst system mutant strains, SG35 and SG95, showed no significant differences in the kinetic parameters between these strains. This suggests that the Phn and Pst systems have similar affinities for Pi and that loss of one system can be compensated for by increased activity of the remaining system. The high-affinity Pi transport system of the closely related Streptomyces granaticolor has an apparent Km of 60 µM Pi (Licha et al., 1997), which is similar to the values obtained here for M. smegmatis. The Pst systems that have been characterized from Gram-negative bacteria and the PhoCDET system of S. meliloti have much lower apparent Km values between 0.2 and 0.4 µM Pi (Voegele et al., 1997; Willsky & Malamy, 1980; Yuan et al., 2006). This is in good agreement with the lower Pi concentration at which these systems are induced in Gram-negative bacteria as compared to Gram-positives, as discussed above.
A strain with deletions in both the Phn and the Pst systems still exhibited high-affinity Pi uptake. The difficulties encountered when deleting pstS in the phnD mutant background, while deletion of the gene in the wild-type was possible in a single double-crossover attempt, may have suggested that the presence of at least one of these systems was essential for M. smegmatis and that Pi transport in the double mutant might be attributed to a suppressor mutation. Such suppressor mutations, caused by gene amplification or increased expression of the low-affinity, constitutive Pit system, have been isolated for an E. coli pstS pitA double mutant (Hoffer et al., 2001) and an S. meliloti phoC mutant (Voegele et al., 1997). Increased expression of pitA in the phoC mutant of S. meliloti restored the ability of the mutant to grow in the presence of 2 mM Pi and to transport Pi at the same levels as the wild-type (Bardin et al., 1998). However, pitA in S. meliloti is repressed under conditions of Pi limitation (Bardin et al., 1998), whereas Pi transport in the phnD pstS double mutant of M. smegmatis described here was clearly inducible by Pi starvation. It is therefore unlikely that the Pi transport activity of the double mutant strain is due to increased synthesis of a Pit-like low-affinity transport system, but could rather be due to a further, as yet unidentified inducible Pi transport system. The presence of three different, inducible transport systems with Km values between 40 µM and 90 µM Pi should enable M. smegmatis to easily adapt to changes in the availability of Pi in its environment as has been hypothesized previously for the presence of multiple Pst systems in M. tuberculosis (Lefèvre et al., 1997).
Phn-type transport systems are primarily used for the uptake of alternative phosphorus compounds and generally recognize a wide variety of phosphonates, as well as phosphite (Imazu et al., 1998; Metcalf & Wanner, 1991; Metcalf & Wolfe, 1998; Voegele et al., 1997; White & Metcalf, 2004b). They are thought to transport Pi non-specifically. One of the most commonly utilized classes of phosphonates are the alkylphosphonates. These are degraded by the C–P lyase pathway, which is encoded by the phn genes of E. coli and other bacteria (Chen et al., 1990). Phosphonoacetate, degraded by phosphonoacetate hydrolase (McMullan & Quinn, 1994), can be utilized as a phosphorus source by some micro-organisms, including two streptomycete isolates (Obojska et al., 1999). We therefore tested two phosphonate compounds as well as phosphite for their ability to act as competitive inhibitors for Pi transport in this organism, and none of these compounds inhibited the uptake of 33P. These data strongly suggest that the PhnDCE transport system of M. smegmatis, in contrast to the previously identified Phn-type systems of Gram-negative bacteria, is specific for Pi. This finding is supported by the inability of the same compounds to support growth of M. smegmatis as the sole phosphorus source.
In summary, we demonstrate that both PhnDCE and PstSCAB of M. smegmatis are high-affinity (Km values between 40 µM and 90 µM Pi) ABC-type transport systems, which recognize Pi, but not phosphonates or phosphite, as substrates. The Phn system of M. smegmatis is the first Phn-like system described to date that appears to be specific for Pi. The presence of multiple high-affinity Pi transport systems has so far only been shown in pathogenic species of mycobacteria (Braibant et al., 1996a, b; Lefèvre et al., 1997; Peirs et al., 2005), while M. smegmatis was thought to contain only a single Pst system (Bhatt et al., 2000; Kriakov et al., 2003). The findings presented here suggest that both slow- and fast-growing mycobacteria require several such transport systems for growth. Furthermore, M. smegmatis contains a third, as yet unidentified system for the uptake of Pi, that is active when the Phn and Pst systems are absent.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Allenby, N. E., O'Connor, N., Pragai, Z., Carter, N. M., Miethke, M., Engelmann, S., Wipat, A., Ward, A. C. & Harwood, C. R. (2004). Post-transcriptional regulation of the Bacillus subtilis pst operon encoding a phosphate-specific ABC transporter. Microbiology 150, 2619–2628.
Allenby, N. E., O'Connor, N., Pragai, Z., Ward, A. C., Wipat, A. & Harwood, C. R. (2005). Genome-wide transcriptional analysis of the phosphate starvation stimulon of Bacillus subtilis. J Bacteriol 187, 8063–8080.
Bardin, S. D. & Finan, T. M. (1998). Regulation of phosphate assimilation in Rhizobium (Sinorhizobium) meliloti. Genetics 148, 1689–1700.
Bardin, S., Dan, S., Osteras, M. & Finan, T. M. (1996). A phosphate transport system is required for symbiotic nitrogen fixation by Rhizobium meliloti. J Bacteriol 178, 4540–4547.
Bardin, S. D., Voegele, R. T. & Finan, T. M. (1998). Phosphate assimilation in Rhizobium (Sinorhizobium) meliloti: identification of a pit-like gene. J Bacteriol 180, 4219–4226.
Bhatt, K., Banerjee, S. K. & Chakraborti, P. K. (2000). Evidence that phosphate specific transporter is amplified in a fluoroquinolone resistant Mycobacterium smegmatis. Eur J Biochem 267, 4028–4032.[Medline]
Braibant, M., Lefèvre, P., de Wit, L., Ooms, J., Peirs, P., Huygen, K., Wattiez, R. & Content, J. (1996a). Identification of a second Mycobacterium tuberculosis gene cluster encoding proteins of an ABC phosphate transporter. FEBS Lett 394, 206–212.[CrossRef][Medline]
Braibant, M., Lefèvre, P., de Wit, L., Peirs, P., Ooms, J., Huygen, K., Andersen, A. B. & Content, J. (1996b). A Mycobacterium tuberculosis gene cluster encoding proteins of a phosphate transporter homologous to the Escherichia coli Pst system. Gene 176, 171–176.[CrossRef][Medline]
Chen, C. M., Ye, Q. Z., Zhu, Z. M., Wanner, B. L. & Walsh, C. T. (1990). Molecular biology of carbon-phosphorus bond cleavage. Cloning and sequencing of the phn (psiD) genes involved in alkylphosphonate uptake and C–P lyase activity in Escherichia coli B. J Biol Chem 265, 4461–4471.
Collins, D. M., Kawakami, R. P., Buddle, B. M., Wards, B. J. & de Lisle, G. W. (2003). Different susceptibility of two animal species infected with isogenic mutants of Mycobacterium bovis identifies phoT as having roles in tuberculosis virulence and phosphate transport. Microbiology 149, 3203–3212.
Gonzalez-y-Merchand, J. A., Estrada-Garcia, I., Colston, M. J. & Cox, R. A. (1996). A novel method for the isolation of mycobacterial DNA. FEMS Microbiol Lett 135, 71–77.[CrossRef][Medline]
Hanahan, D., Jessee, J. & Bloom, F. R. (1991). Plasmid transformation of Escherichia coli and other bacteria. Methods Enzymol 204, 63–113.[Medline]
Hoffer, S. M., Schoondermark, P., van Veen, H. W. & Tommassen, J. (2001). Activation by gene amplification of pitB, encoding a third phosphate transporter of Escherichia coli K-12. J Bacteriol 183, 4659–4663.
Hulett, F. M. (1995). Complex phosphate regulation by sequential switches in Bacillus subtilis. In Two-Component Signal Transduction, pp. 298–302. Edited by J. A. Hoch & T. J. Silhavy. Washington, DC: American Society for Microbiology.
Hulett, F. M. & Jensen, K. (1988). Critical roles of spo0A and spo0H in vegetative alkaline phosphatase production in Bacillus subtilis. J Bacteriol 170, 3765–3768.
Hulett, F. M., Lee, J., Shi, L., Sun, G., Chesnut, R., Sharkova, E., Duggan, M. F. & Kapp, N. (1994). Sequential action of two-component genetic switches regulates the PHO regulon in Bacillus subtilis. J Bacteriol 176, 1348–1358.
Imazu, K., Tanaka, S., Kuroda, A., Anbe, Y., Kato, J. & Ohtake, H. (1998). Enhanced utilization of phosphonate and phosphite by Klebsiella aerogenes. Appl Environ Microbiol 64, 3754–3758.
Kononova, S. V. & Nesmeyanova, M. A. (2002). Phosphonates and their degradation by microorganisms. Biochemistry 67, 184–195.
Kriakov, J., Lee, S. & Jacobs, W. R., Jr (2003). Identification of a regulated alkaline phosphatase, a cell surface-associated lipoprotein, in Mycobacterium smegmatis. J Bacteriol 185, 4983–4991.
Lee, M. H., Pascopella, L., Jacobs, W. R., Jr & Hatfull, G. F. (1991). Site-specific integration of mycobacteriophage L5: integration-proficient vectors for Mycobacterium smegmatis, Mycobacterium tuberculosis, and bacille Calmette-Guerin. Proc Natl Acad Sci U S A 88, 3111–3115.
Lefèvre, P., Braibant, M., de Wit, L., Kalai, M., Roeper, D., Grotzinger, J., Delville, J. P., Peirs, P., Ooms, J., Huygen, K. & Content, J. (1997). Three different putative phosphate transport receptors are encoded by the Mycobacterium tuberculosis genome and are present at the surface of Mycobacterium bovis BCG. J Bacteriol 179, 2900–2906.
Licha, I., Benes, I., Janda, S., Host'alek, Z. & Janacek, K. (1997). Characterization of phosphate transport in Streptomyces granaticolor. Biochem Mol Biol Int 41, 431–437.[Medline]
), 60 µM (
), 40 µM (
), 10 µM (
), 2.5 µM (
) Pi. (b) Growth rates (µ) plotted as a function of Pi concentration. (c) Data from (b) plotted as µ versus µ divided by Pi concentration in µM. Results were reproducible in two independent experiments. Panels (b) and (c) show the data from both experiments.
), and 
). (a) Cells carrying pSG10. (b) Cells carrying pSG42. Experiments were carried out in duplicate; representative results are shown.
) in ST medium containing 0.1 mM Pi. Experiments were carried out in triplicate and representative results are shown.
