|
|
|
Journal of Bacteriology, November 2003, p . 6425-6433, Vol . 185, No . 21 Isolation and Characterization of Mutants of the Bacillus subtilis Oligopeptide Permease with Altered Specificity of Oligopeptide Transport
Jonathan Solomon, Department of Biology, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 Received 2 July 2003/ Accepted 8 August 2003
The B . subtilis opp operon encodes five proteinsOppA, OppB, OppC, OppD, and OppF . OppB and OppC are membrane-spanning proteins that create the transmembrane pore through which oligopeptides are imported (30, 35) . OppD and OppF are ATP binding proteins that link transport to the hydrolysis of ATP . OppA is tethered to the extracellular face of the cell membrane via a lipid anchor . OppA binds to the peptide substrate and facilitates its uptake by interacting with and delivering the peptide to the OppBCDF complex . Mutations in oppA, oppB, oppC, and oppD completely block the ability to take up oligopeptides (30), whereas mutations in oppF cause a partial defect in transport (19) . This partial defect causes resistance to low levels of a toxic tripeptide and a delay in competence gene expression (19, 35) . Oligopeptide permeases can transport a wide variety of peptide substrates, in contrast to many other transporters, which have a limited number of substrates . The structure of substrate binding protein OppA of Salmonella enterica serovar Typhimurium (36, 39, 40) and mutational analyses of OppA of L . lactis, based on the structure of OppA of Salmonella serovar Typhimurium (32), have provided insight into this unusual feature . When an oligopeptide binds in the cleft between two lobes (domains I and III) of OppA, the lobes close up, capturing the peptide . There are important charge interactions with the N and C termini of the bound peptide and weak bond interactions with the peptide backbone . The spacious cavity within the binding pocket of OppA can accommodate a wide variety of amino acid side chains, resulting in very little sequence specificity for binding peptides . In B . subtilis, the oligopeptide permease also functions to regulate two developmental processes, sporulation and genetic competence . In fact, B . subtilis opp was first identified as a sporulation locus, named spo0K, because null mutations caused decreased sporulation efficiency (30, 35) . The B . subtilis oligopeptide permease contributes to the regulation of sporulation and competence, at least in part, by importing specific signaling peptides derived from phr gene products (16, 17, 27, 33) . Seven phr genes have been recognized in the B . subtilis genome . Each encodes the precursor of a putative signaling peptide that is secreted and imported via the oligopeptide permease . Once inside the cell, a specific Phr peptide antagonizes the activity of its cognate regulatory protein(s) of the Rap family . The competence and sporulation-stimulating factor (CSF) is a pentapeptide derived from the phrC gene product (37) . It is imported via the oligopeptide permease and, once inside the cell, stimulates the expression of genes (e.g., srfA) activated by transcription factor ComA . CSF most likely stimulates the activity of ComA by inhibiting RapC, a negative regulator of ComA (18, 37) . At least three Rap proteins, RapA, RapB, and RapE, contribute to the regulation of sporulation (14, 28, 29) . All three are response regulator aspartyl phosphate phosphatases and act to dephosphorylate Spo0F, thereby inhibiting sporulation . Pentapeptides derived from phrA, phrC, and phrE are imported via the oligopeptide permease and inhibit the activity of RapA, RapB, and RapE, respectively (18, 25), thereby stimulating sporulation . In this article we describe the isolation of rare, partly functional mutations in B . subtilis opp . The identification of these mutations was relatively simple because of the role of B . subtilis opp in regulating competence and sporulation . The mutants were deficient in the transport of a specific toxic peptide but still retained the ability to sporulate and/or become competent . The mutations, mostly in oppA, affected residues whose alteration appears to change the specificity of oligopeptide transport .
The B . subtilis strains used are listed in Table 1 . All are derived from strain JH642 and contain trpC2 and pheA1 mutations (31) . The metC85::Tn917 allele (41) is from Bacillus Genetic Stock Center strain 1A607 .
The srfA-lacZ fusion (srfA-lacZ Media. Routine growth and transformation of B . subtilis were done by using standard techniques and media (12) . Liquid minimal medium was defined S7 medium (42), except that morpholinepropanesulfonic acid (MOPS) buffer was used at 50 mM instead of 100 mM . Minimal medium in agar plates contained Spizizen's minimal salts (12) . Minimal medium was supplemented with 1% glucose (0.1% glucose for minimal sporulation plates), 0.1% glutamate, and essential amino acids . Nutrient sporulation medium was DS medium (12) . Competence indicator plates contained SpII medium (12) . Antibiotics were used at the following concentrations: ampicillin, 100 µg/ml; chloramphenicol, 5 µg/ml; spectinomycin, 100 µg/ml; neomycin, 5 µg/ml; and erythromycin, 0.5 µg/ml, and lincomycin, 12.5 µg/ml, together to select for the mls gene . 5-Bromo-4-chloro-3-indolyl-ß-D-galactopyranoside (X-Gal) was added to indicator plates from 30 to 120 µg/ml . Bialaphos (phosphinothricyl-alanyl-alanine) was a generous gift from Meiji Seika Kaisha, Ltd., Pharmaceutical Research Center, Yokohama, Japan, and was routinely used at 5 µg/ml unless otherwise indicated . Mutagenesis and mutant isolation. Cells were grown in DS medium to mid-exponential phase, pelleted, and resuspended in Spizizen salts-1 mM MgSO4 . Cells were then shaken at 37° for 30 min in the presence of ethyl methanesulfonate (EMS) (1.5% final concentration) . EMS-treated cells were pelleted, washed twice in DS medium to remove the EMS, separated into 10 aliquots, and grown to confluence in Luria-Bertani medium . Mutants resistant to 5 µg of bialaphos/ml were selected either on competence plates (SpII medium) containing X-Gal or on minimal sporulation plates . Bialaphos-resistant colonies that appeared darker blue (comG-lacZ; Com+) or more opaque (Spo+) than opp-null mutants were purified, and chromosomal DNA was used for backcrossing . After backcrossing, with selection for a marker (yjbB::spc) linked to opp, 13 candidates retained bialaphos resistance and developmental phenotypes and were characterized further . Sequencing of mutations. We determined in which gene of the opp operon each mutation was located by complementation analysis . Wild-type copies of oppA and oppBCDF were introduced at the amyE locus, and the partial diploids were tested for mutant phenotypes (19) . All but two mutations were complemented by oppA . The other two mutations were further tested by complementation with oppBC, oppD, and oppF at amyE (19) . Each mutant allele was sequenced by using an fmol DNA sequencing kit (Promega) . The DNAs for the sequencing reactions were PCR products derived from the chromosomal DNAs of the backcrossed mutants . PCR products were purified by using a Qiagen PCR purification kit . Bialaphos resistance assays. Cells were grown for at least three doublings in liquid minimal medium . The cultures were diluted 1/10 into minimal medium supplemented with various concentrations of bialaphos . The diluted cultures were grown for 3 h, after which the optical density at 600 nm (OD600) was determined . ß-Galactosidase assays. ß-Galactosidase specific activity was measured essentially as described previously (21) and is presented as 1,000 times the change in the A420 per minute per milliliter of culture per OD600 unit . Response to CSF. CSF is the pentapeptide ERGMT (37) and was synthesized by and purchased from Genemed Synthesis, Inc . (S . San Francisco, Calif.) . CSF activity was measured essentially as described previously (20, 37, 38) . Briefly, 0.25 ml of cells (containing an srfA-lacZ fusion) at an OD600 of 0.05 to 0.1 was added to 0.25 ml of medium containing various concentrations of CSF and bovine serum albumin (50 µg/ml) to prevent a nonspecific loss of activity . The mixture was incubated at 37°C for 70 min and then assayed for ß-galactosidase specific activity . Sporulation assays. Cells were grown in DS medium at 37°C, and sporulation was assayed 18 to 24 h after the end of exponential growth . Percent sporulation is 100 times the number of spores per milliliter (heat-resistant CFU after treatment at 80°C for 20 min) divided by the number of viable cells per milliliter . OppA structural modeling. OppA (SwissProt accession number P24141) from B . subtilis was modeled by using the First Approach Mode at the Swiss-Model protein structure homology modeling server (http://us.expasy.org) . B . subtilis OppA was modeled by using 33 structural templates of Salmonella serovar Typhimurium OppA and 5 structural models of E . coli dipeptide permease binding protein DppA (7) . Amino acids 34 to 526 (out of 545) of B . subtilis OppA were included in the model . The graphics in Fig . 1 were produced by using Swiss-Pbd Viewer (9, 23, 24) . The model is very similar to the solved crystal structure of Salmonella serovar Typhimurium OppA (36, 39, 40) .
Mutants resistant to 5 µg of bialaphos/ml were isolated at a frequency of approximately 10-3 following mutagenesis with EMS (see Materials and Methods) . Of these, slightly more that 1 in 104 appeared to have levels of comG-lacZ expression or sporulation that were higher than those of opp-null mutants . We isolated 21 bialaphos-resistant mutants that appeared to retain the expression of comG-lacZ and 10 mutants that still appeared to sporulate . We isolated chromosomal DNA from these mutants and transferred (by transformation) the opp region to nonmutagenized cells by selecting for a marker, yjbB::spc, that is closely linked to opp . After backcrossing, 13 candidates retained their phenotypes . All further characterization was done with these strains or their derivatives . Mapping and sequencing of mutations. We determined in which gene of the opp operon each mutation was located by complementation analysis and DNA sequencing (see Materials and Methods) . Most of the mutations (Table 2) were in oppA, encoding the substrate binding protein . Six different missense mutations in oppA were identified; one allele (oppA-E259K) was isolated four times (three independent isolates), and another (oppA-G337R) was isolated independently three times . One mutation was in oppC, encoding one of the two integral membrane proteins that make the pore . One mutation was in oppD, encoding one of the two ATP binding subunits . The oppD mutation is not in the conserved ABC region . That only two of the alleles were isolated multiple times indicates that the screen was probably not saturated .
Five of the partly functional mutations in oppA are in domain III and likely affect peptide binding directly, and one mutation is in domain I . Two mutations, D449N and R443H, affect amino acids (Fig . 1, aqua balls) in domain III that are predicted to form important charge interactions with the N and C termini of at least some tri- and tetrapeptides (36, 39, 40) . Two mutations, G337R and G466E, alter residues (Fig . 1, red balls) in domain III that are predicted to line the peptide binding pocket . These changes are likely to interfere directly with the binding of some oligopeptides . The G299E mutation also may affect peptide binding directly . This mutation alters an amino acid (Fig . 1, black balls) that is at the crossover region between domains III and I and is predicted to be partially exposed in the peptide binding pocket . The E259K mutation affects a residue (Fig . 1, yellow balls) in domain I that is predicted to be outside the peptide binding pocket . We suspect that E259K affects peptide binding indirectly . Of the amino acids altered in the mutants, only two, D449 and R443, are conserved between B . subtilis OppA and Salmonella serovar Typhimurium OppA . In Salmonella serovar Typhimurium OppA, these two amino acids form charge interactions with the N and C termini of at least some tri- and tetrapeptides . That we isolated mutations affecting these residues in B . subtilis OppA is consistent with the notion that they affect peptide binding directly . These residues do not appear to be conserved in OppA of L . lactis, which binds larger oligopeptides . The precise role of the various amino acids of B . subtilis OppA in peptide binding ultimately will be elucidated when the structure is determined .
Bialaphos transport.
The eight opp mutations impaired but did not completely abolish bialaphos transport, as determined by resistance to various concentrations of the toxic tripeptide . We tested for resistance to bialaphos by measuring growth in liquid medium . Cells in exponential growth were diluted into various concentrations of bialaphos, and the change in the optical density of each culture was determined after 3 h and compared to that of a parallel culture without bialaphos (Fig . 2) . The concentration of bialaphos at which growth was
Sporulation. The partly functional opp mutations had a range of effects on sporulation . We measured sporulation frequencies for each of the missense mutants in DS medium and compared the frequencies to those of the wild type and an opp-null mutant (Table 2) . The oppC-A266V and oppD-E239K mutants had large defects in sporulation, comparable to the defect in the oppA-null mutant . The oppD-E239K mutant also had a high level of bialaphos resistance, but the oppC-A266V mutant had a very low level of bialaphos resistance (Table 2 and Fig . 2) . Thus, there is no correlation between a defect in bialaphos resistance (transport) and sporulation . The oppA alleles caused a range of sporulation defects and, again, the severity of the defect did not correlate with the level of resistance to bialaphos (Table 2 and Fig . 2) . oppA-G337R (in the peptide binding pocket) and oppA-E259K (in domain I, apparently located away from the binding pocket) resulted in approximately wild-type levels of sporulation . In contrast, oppA-R443H (predicted to affect an interaction with the C terminus of some peptides), resulted in a low level of sporulation . The other three oppA mutants had levels of sporulation intermediate between those of the wild type and the oppA-null mutant . Competence gene expression. The partly functional opp mutants also had a range of effects on competence gene expression, from causing only slight delays to causing severely reduced expression (Fig . 3 and Table 2) . We measured the expression of srfA-lacZ and comG-lacZ in each of the mutants grown in defined minimal medium . srfA is expressed in all cells and is controlled by the response regulator transcription factor ComA (6, 17; see also references therein) . ComA acquires phosphate from the membrane-bound receptor kinase ComP, and the activity of ComP is stimulated by ComX pheromone . The activity of ComA is also modulated by CSF, the pentapeptide derived from the phrC gene product . CSF (and perhaps peptides derived from other phr genes) is transported into the cell via the oligopeptide permease, where it apparently inhibits the activity of RapC, a negative regulator of ComA . The expression of comG depends on the expression of comS, a small open reading frame in the srfA operon (5, 11), and comG expression is a good indicator of the level of competence development (6; see also references therein) .
Response to CSF (ERGMT). We tested the ability of each of the partly functional opp mutants to respond to CSF, a pentapeptide with the sequence ERGMT (18, 37) . Cells were assayed for the amounts of srfA-lacZ expression induced by various concentrations of CSF (Fig . 4) . As described previously (18, 37), in wild-type cells, the accumulation of ß-galactosidase specific activity from an srfA-lacZ fusion increases approximately threefold 70 min after the addition of 5 to 10 nM CSF (Fig . 4A) . Higher concentrations of CSF cause less stimulation until expression is inhibited (Fig . 4A) . This inhibitory effect is not understood but is dependent on the import of CSF and is independent of RapC, the target of the stimulatory effect (18, 37) . oppA-null mutants do not show any stimulation (or inhibition) of srfA expression in response to CSF (Fig . 4A) .
oppA-E259K, oppA-G299E, and oppA-G466E also impaired the ability to respond to CSF (Fig . 4D) but had little if any effect on the expression of srfA (Table 2 and Fig . 3) . These mutations did not alter the concentration of CSF at which a response is maximal but did alter the amplitude of the response . Finally, oppA-G337R and the mutations in the pore-forming protein, oppC-A226V, and the ATP binding protein, oppD-E239K, had little or no effect on the response to CSF (Fig . 4C) as well as little or no effect on the expression of srfA (Table 2 and Fig . 3) .
At least three peptides, PhrA, PhrE, and CSF (PhrC), are known to affect sporulation (15, 26, 27) . Several of the partly functional opp mutants (oppC-A266V, oppD-E239K, oppA-G299E, and oppA-R443H) are more severely affected for sporulation than for competence gene expression . These mutants likely are defective in the uptake of peptides important for sporulation but still able to import peptides that stimulate competence gene expression . The only Phr-derived peptide known to affect competence gene expression is CSF (PhrC) . However, the defect in competence caused by a phrC-null mutation is much less severe than that caused by an opp-null mutation (18, 37) . This difference implies that signaling peptides in addition to CSF might be imported by the oligopeptide permease to stimulate ComA activity and competence gene expression . Consistent with this notion, we found no direct correlation between the response to CSF and the defect in the expression of srfA in the opp mutants (Table 2 and Fig . 3 and 4) . We postulate that opp mutants (e.g., oppA-G299E and oppA-G466E) that are defective in the response to CSF but that have relatively normal expression of srfA (and comG) are able to import other signaling peptides that stimulate the activity of ComA . The identities of these peptides are presently not known . Two of the eight mutations that we characterized affect subunits other than the oligopeptide binding protein OppA . The membrane-spanning proteins of ABC transporters must interact with the transported molecule, and the mutation in oppC (oppC-A266V) could alter such an interaction . Alternatively, we suspect that the oppC-A266V and oppD-E239K mutations cause a more general defect in transport, perhaps affecting the ATPase cycle or communication between the substrate binding protein and the transport complex . If there is a more general defect in these mutants, then that defect has a much more severe effect on sporulation than on competence gene expression . The four mutations in oppA (oppA-G337R, oppA-R443H, oppA-D449N, and oppA-G466E) that change residues in the oligopeptide binding pocket probably directly alter the specificity of oligopeptide binding . The various effects of these mutations on bialaphos resistance, sporulation, and competence gene expression most likely reflect this altered specificity . The identification of additional signaling peptides affecting competence gene expression will help to clarify the effects of the various opp mutations . Furthermore, structural analysis of these mutant OppA proteins should be very useful for understanding the nature and specificity of peptide binding .
J.S . was supported in part by a Howard Hughes predoctoral fellowship . This work was supported in part by Public Health Service grant GM50895 to A.D.G . from the NIH .
What Is Dna?,
What Is Cell Biology?,
What Is Amino Acid?,
What Is Biotechnology?,
What Is Yeast?,
e,
Microorganism,
n,
Bacteria,
a,
Bacterium,
a,
Microbes,
o,
Microbiology,
i,
Pichia,
o,
Escherichia coli,
o,
Candida albicans,
i,
Pseudomonas aeruginosa,
i,
Bacteriological,
a,
Amikacin,
s,
Enterobacteriacea,
n,
Bacillus subtilis,
a,
Bacillus subtilis,
s,
Bacillus,
o,
Bacteriophages,
n,
Vibriosis,
n,
Microbial,
c,
Pasteurella,
r,
Escherichia coli,
c,
Escherichia coli,
n,
Haemophilus,
r,
S. cerevisiae,
c,
Bacillus,
e,
Microorganism,
r,
Antibiotic treatment
|
© 2005
Transgalactic Ltd (manufacturer of Bioscreen C software) |
Privacy Statement | P.O. Box
1393, 00101 Helsinki, Finland,
Last modified: May 25, 2005
| ||||||