Department of Microbiology and Infectious Diseases, Faculty of Medicine, University of Calgary, Heritage Medical Research Building, 3300 Hospital Dr. NW, Calgary, Alberta, Canada T2N 4N1
Correspondence
Anthony B. Schryvers
schryver{at}ucalgary.ca
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Once the ferric ion has crossed the outer membrane it is transported into the cytoplasm by an ABC (ATP-binding cassette) (Doige & Ames, 1993) import system. This system comprises a periplasmic iron-binding protein, FbpA, and an inner-membrane complex consisting of FbpB, a predicted transmembrane channel-forming protein, and FbpC, a predicted ATPase (Adhikari et al., 1995
, 1996
). Isogenic mutants lacking FbpA are defective in utilizing several different sources of ferric ion for growth (Khun et al., 1998
; Kirby et al., 1997
), suggesting that the FbpABC pathway represents a convergence point for non-siderophore-mediated outer-membrane ferric ion transport systems.
In view of its critical role in iron acquisition from transferrin, it is not surprising that FbpABC pathways have been identified in species that possess transferrin receptors such as Neisseria gonorrhoeae (Adhikari et al., 1996), Haemophilus influenzae (Adhikari et al., 1995
), Mannheimia (Pasteurella) haemolytica (Kirby et al., 1998
) and Actinobacillus pleuropneumoniae (Chin et al., 1996
). Functional homologues of FbpABC are also found in species such as Serratia marscesens (Angerer et al., 1990
) and Yersinia spp. (Gong et al., 2001
) that lack the transferrin receptor-mediated uptake pathway. The role of the FbpABC pathway in these species is less clear, as they possess alternative means of transporting iron such as siderophore-mediated uptake systems.
Production of functional FbpA is dependent upon export into the periplasmic space, a process initiated by the presence of a leader (signal) peptide at the N-terminus responsible for the targeting of exported proteins (Swidersky et al., 1990). Leader peptides have been shown to interact with SecA, a protein that is found in cytoplasmic and membrane-associated forms (Akita et al., 1990
). SecA mediates targeting of the nascent protein chain to the membrane by binding to the translocation channel composed of SecYEG (Stathopoulos et al., 2000
). The leader peptide associates with the SecYEG complex and is cleaved by the signal peptidase, LepB, located on the periplasmic face of the membrane prior to translocation of the remainder of the polypeptide chain (Danese & Silhavy, 1998
).
Heterologous genetic exchange experiments were initiated in H. influenzae to probe the interaction between FbpA and other pathway components. In this study we demonstrate that the functional expression of foreign FbpAs was defective and that the defect was attributable to the leader peptide region of the foreign fbpA gene. A series of hybrid genes with foreign leader peptides were prepared and expressed in both the Escherichia coli and H. influenzae backgrounds to demonstrate that the defect in functional expression was directly attributable to the presence of the foreign leader peptide.
![]() |
METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Construction of hybrid fbpA and blaP genes with different signal sequences.
A series of hybrid genes with foreign leader peptides were prepared as illustrated in Fig. 1. The amino acid sequences of the leader peptides are shown in part A, the first 20 residues of the mature protein of the hybrid genes are shown in part B and the nomenclature used for the hybrid proteins is illustrated in part C. The overall strategy was to amplify the mature coding sequence of H. influenzae FbpA, the N. gonnorhoeae FbpA and the TEM-1
-lactamase and ligate them into vectors containing the leader sequences of H. influenzae fbpA, the N. gonnorhoeae fbpA genes or simply a start codon. For this set of experiments, the ColE1 origin of replication in the gene replacement vector was replaced by the lower copy number p15a origin. Vectors with the appropriate inserts were used as templates for inverse PCR reactions with the appropriate reverse primers (1013, 1090 or 1016) and ligated with the amplified mature coding sequences. The first codon of the mature FbpA protein that encoded an aspartate residue was also incorporated into the 5' end of primers 1013 and 1090 so that an extra aspartate residue was added to the mature proteins. The correct orientation of the resulting plasmids was confirmed by colony PCR and sequencing was performed to confirm the junctional sequences. The selected plasmids were linearized and used to transform an H. influenzae strain with a chloramphenicol-resistance cassette replacing the fbpA gene (h306) essentially as described above.
|
Iron starvation of H. influenzae and anaerobic growth assays.
For iron starvation of H. influenzae, colonies from a fresh cultured chocolate plate were used to inoculate prewarmed BNP [BHI, 3·32 µg NAD+ ml-1 and 20 µg PPIX (protoporphyrin IX, Sigma) ml-1] broth. The culture was grown at 37 °C with shaking to an OD600 of 0·2 and was then supplemented with EDDHA to a final concentration of 40 µg ml-1 (BNPE). The culture was grown for an additional 2 h, diluted to an OD600 of 0·05 in BNPE broth, and 200 µl of this suspension was applied with sterile cotton swabs onto BNE agar plates (BNPE minus PPIX).
Anaerobic growth assays were conducted as previously described (Kirby et al., 1997) to test the ability of H. influenzae strains to utilize different iron sources. Sterile solutions of fully iron-saturated human transferrin and human lactoferrin (Sigma) were prepared at a stock concentration of 8 mg ml-1 (100 µM) in 50 mM Tris. A stock solution of haemin at a concentration of 100 µM was also prepared and filter-sterilized. Twenty-five microlitres (equivalent to 2·5 nmol) of the above iron supplements were applied onto sterile 0·25 inch (0·64 cm) concentration disks (Difco). The plates were incubated overnight anaerobically at 37 °C and growth around the disk was evaluated.
Nitrocefin assay.
Nitrocefin is a chromogenic cephalosporin that undergoes a distinctive colour change from yellow to red upon hydrolysis by -lactamases (O'Callaghan et al., 1972
). The level of TEM-1
-lactamase production in different bacterial strains can be determined by measuring the steady-state rate of nitrocefin hydrolysis in the cell lysates. To prepare cell lysate, 1 ml of culture with an OD600 of 1·0 was centrifuged at 14 000 r.p.m. for 1 min. The cell pellet was resuspended in 150 µl B-per detergent (Pierce) and incubated on ice for 1 min. The cell lysate was then vigorously vortexed for 30 s. The cell lysate was centrifuged at 14 000 r.p.m. for 1 min and the supernatant was removed for assay.
In a 1 ml cuvette, 470 µl phosphate buffer (pH 7·2), 20 µl cell extract and 10 µl nitrocefin (500 µg ml-1) were mixed. The nitrocefin solution was prepared by dissolving nitrocefin powder in a few drops of DMSO, made up to the final volume with phosphate buffer, and stored in the dark at -20 °C. After the cell lysate was mixed with the substrate, the increase in A482 was followed for 1 or 2 min. For use as a blank, 490 µl buffer and 10 µl nitrocefin were mixed. The rate (A482 min-1) was determined by plotting absorbance over time.
Bioinformatic analyses.
The periplasmic binding proteins (PBPs) of the ABC transporters in E. coli and H. influenzae were selected from the ABCdb database (http://ir2lcb.cnrs-mrs.fr/ABCdb/presentation.html) and from a genomic transport analysis page that provided tabulations of possible ABC transporters annotated from complete genomes (http://www-biology.ucsd.edu/msaier/transport/). The signal peptides were identified using SignalP V2.0.b2 on the signalP server (http://www.cbs.dtu.dk/services/SignalP-2.0/), which also provided predictions of the length of the different regions of the signal peptide (n-, h- and c-regions). The net charges of the n-regions were assessed by determining the number of cationic amino acids.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
To evaluate functional reconstitution of the FbpABC pathway, the recombinant H. influenzae strains were tested for their ability to acquire iron from human transferrin (hTf) and ferric citrate on an iron-limited medium (Kirby et al., 1997). Under these assay conditions none of the strains containing intact foreign fbpA genes were capable of using hTf as a source of iron for growth (data not shown), suggesting a defect in the pathway. The control strain with the H. influenzae fbpA gene and kanamycin-resistance cassette grew as well as the wild-type strain, indicating that the deficiency was not due to a polar effect on expression of the fbpBC genes. All of the strains grew on haemin as an iron source, indicating that the defect was specific to the FbpABC pathway.
To determine whether the impairment in transport was due to a deficiency in expression of the foreign FbpAs, the recombinant H. influenzae strains were tested for the expression of the foreign fbpA genes by Western blot analysis using species-specific antisera (Fig. 2, column I). Relative to the negative control strain (lane 1, H263), there was strong expression of the H. influenzae FbpA (Fig. 2A
, lane 2), low-level expression of the N. gonorrhoeae FbpA (Fig. 2B
, lane 2) and M. haemolytica FbpA (Fig. 2D
, lane 2) but no detectable expression of the Y. enterocolitica FbpA (Fig. 2C
, lane 2) or the A. pleuropneumoniae FbpA (Fig. 2E
, lane 2). To evaluate whether the lack of expression of the foreign fbpA gene in H. influenzae was due to alterations during homologous recombination, each hybrid operon was recovered from the chromosome by PCR amplification and cloned into an expression plasmid in E. coli. Using an upstream primer specific to each fbpA gene and the downstream primer designed to the H. influenzae fbpC gene, the hybrid operons were amplified, ligated into the expression vector and transformed into an entA E. coli strain (E573). When the resulting strains were tested for expression by Western blot analysis, all of the FbpA proteins were present at readily detectable levels (Fig. 2
, lane 5), indicating that there were no major defects in the respective fbpA genes. The additional bands present in Fig. 2D
, column II that are also present in the vector control (lane 4), are due to spurious cross-reactivity by the antiserum against M. haemolytica FbpA.
|
|
The leader peptides from the H. influenzae and N. gonorrhoeae FbpAs were PCR amplified along with a control containing only a start codon and ligated to the mature coding sequences of the H. influenzae FbpA, N. gonnorhoeae FbpA and TEM-1 -lactamase (Fig. 1
). The hybrid genes were assembled in a vector flanked by the region from H. influenzae upstream of the fbpA gene and a kanamycin-resistance marker followed by a partial fbpB gene from H. influenzae to facilitate incorporation into the native fbpA locus in H. influenzae through homologous recombination. The recombinant plasmids were linearized by digestion with SacII and used to transform the parent H. influenzae strain. In order to avoid the complication of unexpected crossover events within the fbpA locus, all transformations were performed with an H. influenzae strain that had a chloramphenicol-resistance cassette in place of the native H. influenzae fbpA gene (strain H306). KanR/CmS colonies were screened to confirm the presence of the modified genes by PCR analysis.
The activity of the hybrid -lactamase genes in E. coli could be tested directly with the p15a hybrid plasmids but an intact fbp pathway was required for evaluating the hybrid fbpA genes. Thus, the hybrid fbpA genes were subcloned into a complete fbpABC operon for assessment of function. The reconstitution of hybrid operons in E. coli was achieved by directional cloning with BamHI and SacII double digestion of the vector p15aHiUPFBC and the replacement vectors that harboured the hybrid genes. BamHI and SacII subfragments containing the hybrid fbpA genes and the H. influenzae fbpBC genes were ligated together and the resultant plasmids were transformed into an E. coli entA strain for expression and functional analysis.
The expression of the H. influenzae FbpA and the N. gonorrhoeae FbpA from genes containing different leader peptides in both the E. coli and H. influenzae backgrounds was analysed by Western blotting using polyclonal antisera (Figs 4 and 5
). There was no detectable FbpA in recombinant strains that contained the leaderless fbpA genes regardless of the host species (lanes A4, B8, C4 in Fig. 4
and lanes A3, B8, C4 in Fig. 5
), suggesting that protein expressed in the cytoplasm is misfolded and degraded. The hybrid gene constructed with the H. influenzae leader and mature coding sequence [HIFA(L)HIFA] differed from the wild-type H. influenzae fbpA gene (WT HIFA) only by having an extra aspartate residue at the beginning of the mature protein and both were shown to express at a similar level in E. coli (compare lanes A2 and A3, and C3 and C5 in Fig. 4
). The most significant observation concerning H. influenzae fbpA expression was that the gene containing the N. gonorrhoeae leader sequence [NGFA(L)HIFA] was expressed in E. coli strains (lanes A5 and C6, Fig. 4
) but not in H. influenzae (lane B9, Fig. 4
). The wild-type N. gonorrhoeae fbpA and the gene containing the H. influenzae fbpA leader [HIFA(L)NGFA] were expressed at similar levels in E. coli strains (lanes A2, A4, C3 and C5 in Fig. 5
), whereas the expression of wild-type N. gonorrhoeae fbpA in H. influenzae was greatly reduced, compared to that of the gene containing the H. influenzae fbpA leader [HIFA(L)NGFA] (lanes B6 and B9 in Fig. 5
, respectively). The hybrid Haemophilus/Neisseria FbpA in strain H292 which arose spontaneously during homologous recombination was essentially the same as genetically reconstituted hybrid HIFA(L)NGFA in strain H354 in terms of amino acid sequence and level of expression (lanes B7 and B9, Fig. 5
).
|
|
|
Analyses of -lactamase with different signal peptides
In the absence of available antiserum against -lactamase, expression of hybrid TEM-1
-lactamases was monitored functionally: by growth on ampicillin-containing agar and by assaying enzyme activity using the chromogenic substrate nitrocefin. Except when the leader peptide was absent, expression of all of the
-lactamases resulted in growth on LB agar with 150 µM ampicillin in E. coli and on BNH agar with 100 µM ampicillin in H. influenzae (Fig. 6B
). Since it was probable that even a low level of functional
-lactamase expression was sufficient to confer ampicillin resistance upon the host, a more sensitive and quantitative measure of
-lactamase was required. Thus an enzyme assay using nitrocefin as a substrate was employed for quantitative analysis. The rate of nitrocefin hydrolysis was measured by the change of A482. No activity was detected for the strain of H. influenzae expressing
-lactamase lacking a leader peptide. The highest rate of nitrocefin hydrolysis was observed for the gene containing the H. influenzae leader peptide [HIFA(L)BlaP], which was more than 10-fold higher than that from the gene containing the N. gonorrhoeae leader peptide [NGFA(L)BlaP] and >180-fold higher than that of the wild-type
-lactamase gene (Fig. 6
). These results indicate that even though the N. gonorrhoeae leader peptide and native
-lactamase leader peptide resulted in substantially reduced levels of export relative to the H. influenzae leader peptide, there was sufficient enzyme to support growth in the presence of 100 µM ampicillin.
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The lack of function of the leaderless FbpAs and -lactamase in either host (Fig. 6
) confirms that the leader peptides are required to mediate the export of these proteins, as has been proposed for most extra-cytoplasmic proteins (Danese & Silhavy, 1998
). Signal peptides consist of three regions: (i) a positively charged, hydrophilic, N-terminal region (n-region), (ii) a largely hydrophobic core region of 915 amino acids postulated to form an
-helix (h-region), and (iii) a 46 amino acid C-terminal region (c-region). These regions of the signal peptide primarily interact with different components of the export apparatus and thus are implicated in different stages of the export process. In this study we do not provide direct experimental evidence to determine what segments of the signal peptide are responsible for the recognition specificity by the H. influenzae secretion apparatus or to determine what stage of the export process is impaired. However, some insights might be gleaned from a comparison of the signal peptides from H. influenzae FbpA, N. gonorrhoeae FbpA and the TEM-1
-lactamase (Fig. 1
) or a more extensive analysis of signal peptides from periplasmic proteins from H. influenzae and E. coli.
The n-region is important for recognition by SecA, the first step in the export process (Akita et al., 1990). The n-region of the leader peptides used in this study did vary (Fig. 1
) and comparisons between leader peptides from H. influenzae and E. coli ABC transporter periplasmic binding proteins (PBPs) revealed some differences. A majority (82 %) of the signal peptides from the H. influenzae periplasmic proteins had two or more lysines at the N-terminus, and unlike the situation in E. coli, arginine was rarely present. There were no obvious differences in the
-helical h-region, particularly between the H. influenzae and N gonorrhoeae leader peptides (Fig. 1
). Likewise, there were no obvious differences in the -3 and -1 positions of the leader (von Heijne, 1990
) that are required for recognition by the signal peptidase. It is interesting to note that both the N. gonorrhoeae FbpA leader and the TEM-1
-lactamase leader have a proline in the -4 position (Fig. 1
), implying that there may be structural differences in the junction between the transmembrane
-helical h-region and the c-region segment (Barkocy-Gallagher et al., 1994
).
The sequence comparisons suggest that the n-region or junctional regions of the leader peptides and their interaction with the SecA or leader peptidase components are most likely responsible for the observed barrier to functional expression of periplasmic proteins in H. influenzae. Thus it would be logical to target these regions and genes in subsequent studies designed to further probe the novel features of the H. influenzae export apparatus revealed in this study. The availability of a simple screening procedure (large vs small colony) and a high-efficiency natural transformation system should facilitate genetic approaches for probing the interactions between the leader peptide and export apparatus. Amino acid alignments of various signal peptidases reveal that the H. influenzae signal peptidase contains an extra 26 amino acids predicted to form an extended loop near to the S3 substrate-binding site (Paetzel et al., 1998) (data not shown) and thus genetic experiments should be included to evaluate the role of this region.
The specificity of the H. influenzae export apparatus constitutes a barrier to functional expression of foreign FbpAs (Fig. 6) since low levels of expression (Fig. 2B
E, lane 2) are not sufficient to mediate iron acquisition from transferrin. Although the levels of production of TEM-1
-lactamase were also dramatically reduced (Fig. 6
), very low levels of functional enzyme were sufficient to retain function, resistance to antibiotics in the medium. Thus it seems that specificity of the Sec-dependent pathway can constitute a barrier to functional expression of foreign periplasmic proteins in Gram-negative bacteria but depends upon the level of periplasmic protein required for function. Since the Sec-dependent pathway also constitutes the first phase in the export and assembly of many outer-membrane proteins and some secreted proteins, the specificity of this process may have much broader significance.
![]() |
ACKNOWLEDGEMENTS |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Adhikari, P., Berish, S. A., Nowalk, A. J., Veraldi, K. L., Morse, S. A. & Mietzner, T. A. (1996). The fbpABC locus of Neisseria gonorrhoeae functions in the periplasm-to-cytosol transport of iron. J Bacteriol 178, 21452149.[Abstract]
Akita, M., Sasaki, S., Matsuyama, S. & Mizushima, S. (1990). SecA interacts with secretory proteins by recognizing the positive charge at the amino terminus of the signal peptide in Escherichia coli. J Biol Chem 265, 81648169.
Angerer, A., Gaisser, S. & Braun, V. (1990). Nucleotide sequences of the sfuA, sfuB, and sfuC genes of Serratia marcescens suggest a periplasmic-binding-protein-dependent iron transport mechanism. J Bacteriol 172, 572578.[Medline]
Barcak, G. J., Chandler, M. S., Redfield, R. J. & Tomb, J.-F. (1991). Genetic systems in Haemophilus influenzae. Methods Enzymol 204, 321342.[Medline]
Barkocy-Gallagher, G. A., Cannon, J. G. & Bassford, P. J., Jr (1994). Beta-turn formation in the processing region is important for efficient maturation of Escherichia coli maltose-binding protein by signal peptidase I in vivo. J Biol Chem 269, 1360913613.
Bruns, C. M., Norwalk, A. J., Avrai, A. S., McTigue, M. A., Vaughan, K. A., Mietzner, T. A. & McRee, D. E. (1997). Structure of Haemophilus influenzae Fe3+-binding protein reveals convergent evolution within a superfamily. Nat Struct Biol 4, 919924.[Medline]
Chin, N., Frey, J., Chang, C. F. & Chang, Y. F. (1996). Identification of a locus involved in the utilization of iron by Actinobacillus pleuropneumoniae. FEMS Microbiology Lett 143, 16.[CrossRef][Medline]
Danese, P. N. & Silhavy, T. J. (1998). Targeting and assembly of periplasmic and outer-membrane proteins in Escherichia coli. Annu Rev Genet 32, 5994.[CrossRef][Medline]
Doige, C. & Ames, G.-L. (1993). ATP-dependent transport in bacteria and humans: relevance to cystic fibrosis and multidrug resistance. Annu Rev Microbiol 47, 291319.[CrossRef][Medline]
Gong, S., Bearden, S. W., Geoffroy, V. A., Fetherston, J. D. & Perry, R. D. (2001). Characterization of the Yersinia pestis Yfu ABC inorganic iron transport system. Infect Immun 69, 28292837.
Gray-Owen, S. D. & Schryvers, A. B. (1996). Bacterial transferrin and lactoferrin receptors. Trends Microbiol 4, 185191.[CrossRef][Medline]
Jarosik, G. P., Maciver, I. & Hansen, E. J. (1995). Utilization of transferrin-bound iron by Haemophilus influenzae requires an intact tonB gene. Infect Immun 63, 710713.[Abstract]
Khun, H. H., Kirby, S. D. & Lee, B. C. (1998). A Neisseria meningitidis fbpABC mutant is incapable of using nonheme iron for growth. Infect Immun 66, 23302336.
Kirby, S. D., Gray-Owen, S. D. & Schryvers, A. B. (1997). Characterization of a ferric binding protein mutant in Haemophilus influenzae. Mol Microbiol 25, 979987.[Medline]
Kirby, S. D., Lainson, F. A., Donachie, W., Okabe, A., Tokuda, M., Hatase, O. & Schryvers, A. B. (1998). The Pasteurella haemolytica 35 kDa iron regulated protein is an FbpA homologue. Microbiology 144, 34253436.[Abstract]
O'Callaghan, C. H., Morris, A., Kirby, S. M. & Shingler, A. H. (1972). Novel method for detection of beta-lactamases by using a chromogenic cephalosporin substrate. Antimicrob Agents Chemother 1, 283288.[Medline]
Paetzel, M., Dalbey, R. E. & Strynadka, N. C. (1998). Crystal structure of a bacterial signal peptidase in complex with a beta-lactam inhibitor. Nature 396, 186190.[CrossRef][Medline]
Postle, K. (1993). TonB protein and energy transduction between membranes. J Bioenerg Biomembr 25, 591601.[Medline]
Ratledge, C. & Dover, L. G. (2000). Iron metabolism in pathogenic bacteria. Annu Rev Microbiol 54, 881941.[CrossRef][Medline]
Stathopoulos, C., Hendrixson, D. R., Thanassi, D. G., Hultgren, S. J., St Geme, J. W., 3rd & Curtiss, R., 3rd (2000). Secretion of virulence determinants by the general secretory pathway in gram-negative pathogens: an evolving story. Microbes Infect 2, 10611072.[CrossRef][Medline]
Stojiljkovic, I. & Srinivasan, N. (1997). Neisseria meningitidis tonB, exbB, and exbD genes: Ton-dependent utilization of protein-bound iron in Neisseriae. J Bacteriol 179, 805812.[Abstract]
Swidersky, U. E., Hoffschulte, H. K. & Muller, M. (1990). Determinants of membrane-targeting and transmembrane translocation during bacterial protein export. EMBO J 9, 17771785.[Abstract]
von Heijne, G. (1990). The signal peptide. J Membr Biol 115, 195201.[Medline]
Zimmermann, L., Angerer, A. & Braun, V. (1989). Mechanistically novel iron(III) transport system in Serratia marcescens. J Bacteriol 171, 238243.[Medline]
Received 9 April 2003;
revised 2 July 2003;
accepted 15 July 2003.
HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
J MED MICROBIOL | ALL SGM JOURNALS |