Five-Member Gene Family of Bartonella quintana
Division of Biological Sciences, The University of Montana, Missoula, Montana 59812,1 Department of Molecular Evolution, Evolutionary Biology Center, Uppsala University, Uppsala, Sweden,2 Microscopy Branch, Rocky Mountain Laboratories, National Institute of Allergy and Infectious Diseases, Hamilton, Montana 598403e#rg, 百拇医药
Received 16 September 2002/ Returned for modification 10 October 2002/ Accepted 1 November 2002e#rg, 百拇医药
ABSTRACTe#rg, 百拇医药
Bartonella quintana, the agent of trench fever and an etiologic agent of bacillary angiomatosis, has an extraordinarily high hemin requirement for growth compared to other bacterial pathogens. We previously identified the major hemin receptor of the pathogen as a 30-kDa surface protein, termed HbpA. This report describes four additional homologues that share approximately 48% amino acid sequence identity with hbpA. Three of the genes form a paralagous cluster, termed hbpCAB, whereas the other members, hbpD and hbpE, are unlinked. Secondary structure predictions and other evidence suggest that Hbp family members are ß-barrels located in the outer membrane and contain eight transmembrane domains plus four extracellular loops. Homologs from a variety of gram-negative pathogens were identified, including Bartonella henselae Pap31, Brucella Omp31, Agrobacterium tumefaciens Omp25, and neisserial opacity proteins (Opa). Family members expressed in vitro-synthesized proteins ranging from ca. 26.5 to 35.1 kDa, with the exception of HbpB, an ~
55.9-kDa protein whose respective gene has been disrupted by a ~o8?, 百拇医药
510 GC-rich element containing variable-number tandem repeats. Transcription analysis by quantitative reverse transcriptase-PCR (RT-PCR) indicates that all family members are expressed under normal culture conditions, with hbpD and hbpB transcripts being the most abundant and the rarest, respectively. Mutagenesis of hbpA by allelic exchange produced a strain that exhibited an enhanced hemin-binding phenotype relative to the parental strain, and analysis by quantitative RT-PCR showed elevated transcript levels for the other hbp family members, suggesting that compensatory expression occurs.o8?, 百拇医药
INTRODUCTIONo8?, 百拇医药
Bartonella quintana is the bacterial agent of trench fever and is transmitted between humans by the bite of the human body louse (Pediculus humanis) (5). B. quintana was a major cause of morbidity during World Wars I and II and is reemerging as a secondary infectious agent primarily in immunocompromised patients and a cause of "urban trench fever" in homeless, inner-city dwellers (11, 12). B. quintana infections can manifest as typical trench fever (11), bacillary angiomatosis (13), chronic bacteremia (24), endocarditis (23), lymphadenpathy (13), or infections of the nervous and skeletal systems (13, 18). These conditions can occur concurrently and may be life-threatening.
Very little is known of B. quintana's virulence determinants, epidemiology, or the reason for its reemergence. It is known that this bacterium has a high requirement for hemin. Investigation into this extraordinary hemin requirement led to our identification and characterization of a major hemin-binding protein, hemin-binding protein A (HbpA), from B. quintana (6). HbpA is a heat-modifiable outer surface protein that retains its ability to bind hemin after sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).m, http://www.100md.com
We have determined that hbpA belongs to a gene family consisting of five genes. Here we describe the four additional members of this multigenic family, and report the first successful site-directed mutagenesis and trans-complementation in this Bartonella species using hbpA as the gene target. We use quantitative reverse transcription-PCR (RT-PCR) to determine the relative transcript level of each hbp gene family member and compare transcript levels in the wild-type to that of the hbpA mutant.
MATERIALS AND METHODS8, 百拇医药
Bacterial strains and culture conditions. B. quintana was grown on heart infusion agar blood plates (HIAB [6]) at 37°C in 5% CO2 and 100% relative humidity and was harvested at 3 days postinoculation (approximately mid-log phase [32]). E. coli was grown in Luria-Bertani medium for 16 h at 37°C. When required, antibiotic supplements were added to the medium at standard concentrations (21). B. quintana and Escherichia coli strains used or generated in this study are summarized in Table 1.8, 百拇医药
fig.ommitteed8, 百拇医药
TABLE 1. Bacterial strains and plasmids used in this study8, 百拇医药
Preparation and manipulation of nucleic acids. Plasmids used or generated in this study are listed in Table 1. Plasmids intended for routine use were purified from E. coli with a Perfectprep kit (Eppendorf Scientific, Westbury, N.Y.). Plasmids used in sequencing, in vitro transcription-translation, or electroporation were prepared with a Midi-Prep kit (Qiagen, Valencia, Calif.). Genomic DNA was prepared by a hexadecyltrimethyl ammonium bromide technique (2). RNA was isolated using an Ultraspec-II RNA isolation system (Biotex, Huston, Tex.) and contaminating DNA was removed using DNA-Free from Ambion (Austin, Tex.). RNA purity was assayed by a standard protocol (2). Clonings, ligations, and transformation reactions with E. coli were performed as previously described (6).
Nucleotide sequencing and analysis. Both DNA strands were sequenced using a BigDye Terminator Cycle Sequencing Ready Reaction Kit (Applied Biosystems, Inc. [ABI]/Roche, Branchburg, N.J.) and an automated DNA sequencer (ABI; model 377). Sequence data were compiled and analyzed with Seqweb version 2.0 (Accelrys, San Diego, Calif.) or the National Center for Biotechnology Information website. BLAST 2.0 (1) was used for database searches, whereas sequence alignments were done with FASTA 2.0 (19), CLUSTALW 1.6 (26), and BOXSHADE 3.21 .f!w, 百拇医药
SDS-PAGE, immunoblots, and N-terminal sequencing. Protein samples (20 µg total) were separated on SDS-PAGE gels (12.5% [wt/vol] acrylamide) prepared by standard protocol (2). For immunoblots, gels were transferred overnight to supported nitrocellulose (0.45-µm pore size; Osmonics, Minnetonka, Minn.) by the general methods of Towbin et al. (27). The resulting blot was probed by using rabbit anti-HbpA antiserum prepared as before (6) and subsequently developed by using horseradish peroxidase-conjugated goat anti-rabbit immunoglobulin G antibody (Sigma), H2O2, and 4-chloronapthol as previously detailed (22). For N-terminal sequencing, a Triton X-114 precipitate was prepared from B. quintana as previously described (6) and transferred to polyvinylidene difluoride (16). The polyvinylidene difluoride was stained for 15 min with 0.05% Coomassie blue and rinsed briefly, and then the HbpD protein band excised, dried, and subjected to Edman degradation by using an ABI 431A automated peptide sequencer. Sequencing was performed on two separate samples.
Mutagenesis and trans-complementation of hbpA. Mutagenesis of B. quintana was done by using a strategy we previously described for Bartonella bacilliformis (3). Briefly, a transformation-competent strain of B. quintana (L200) was prepared by electroporating wild-type B. quintana (OK 90-268) with pEST. Transformants were subsequently selected on HIAB-Kan plates (i.e., HIAB plus 25 µg of kanamycin/ml) and one resulting strain (LS100) was cured of pEST by six serial culture passages on HIAB. The resulting strain, LS200, was verified as cured of pEST by PCR analysis and kanamycin sensitivity as previously described (3). LS200 was subsequently electroporated with a suicide vector containing a 240-bp internal fragment of hbpA (nucleotides 344 to 583), termed pHBPA'. Mutants were selected on HIAB-Kan. One resulting strain, LS300, was verified as a hbpA mutant by PCR, SDS-PAGE, and immunoblotting. Finally, LS300 was complemented in trans by transforming with the shuttle vector, pBBR1-HBPA, to generate strain LS400.
Hemin-binding assay. Hemin binding by intact B. quintana cells was assayed in vitro essentially as before (6). Briefly, eight plates of B. quintana were harvested into 1.0 ml of 100 mM Tris (pH 8.0) and washed four times by centrifuging the suspension for 5 min at 2,940 x g and resuspending the resulting pellet into 1.0 ml of 100 mM Tris (pH 8.0). The final pellet was resuspended to an optical density at 600 nm (OD600) of 1.0, and four 1-ml aliquots were obtained from each strain. A total of 5 µg of hemin (5 µl of a fresh 1-mg/ml hemin stock solution in 0.02 N NaOH) were added to each tube, gently mixed, and incubated open for 1 h at 37°C at 5% CO2, with gentle mixing at 15-min intervals. Four negative control tubes without B. quintana cells were prepared and incubated as well. After incubation, the suspensions were pelleted by centrifuging for 2 min at 16,000 x g, and the resulting supernatants clarified twice by transferring to new microcentrifuges tube and centrifuging again at 16,000 x g. The OD400 of the final supernatants was assayed, and hemin binding was determined by comparing the reduction in OD400 to negative controls.
In vitro transcription-translation. Plasmids containing individual hbp genes plus their respective promoters were directionally cloned into pBluescript SK(+/-) in opposite orientation to the lacZ promoter. The resulting plasmids (Table 1) were used as templates for a S30 extract kit for circular DNA per the manufacturer's instructions (Promega, Madison, Wis.). Proteins were radiolabeled with a [35S]cysteine-methionine mix (Express; New England Nuclear, Boston, Mass.). Translation products were separated on SDS-PAGE (12.5% [wt/vol] acrylamide) and visualized by exposure of the dried gel to X-ray film overnight.4;, 百拇医药
RT-PCR quantification of hbp family transcripts. Quantitative RT-PCR was performed using TaqMan One-Step RT-PCR master mix from ABI. Ten-microliter reactions were performed in triplicate in a 384-well format, and reactions contained 500 nM concentrations of each primer, 100 nM probe, Master Mix and MultiScribe, and RNase inhibitor Mix to 1x (ABI); RNA was then serially diluted twofold from 5 ng per reaction to 0.153 ng per reaction. Probes consisted of an oligonucleotide labeled at the 5' end with the reporter dye 5-carboxyfluorescein and at the 3' end with the quencher N,N',N'-tetramethyl-6-carboxyrhodamine. Primers and probes used in this study are listed in Table 2. Quantitative RT-PCR conditions were as follows: 1 cycle at 50°C for 30 min, 1 cycle at 95°C for 10 min, and 40 cycles of 95°C for 15 s and of 60°C for 60 s. Changes in fluorescence were monitored by using an ABI 7900HT sequence detection system and raw data were analyzed by SDS software version 2.0 (ABI). Chromosomal DNA from B. quintana was used as a control to ensure that the primers or probe from each gene were binding at similar efficiencies. The efficiency of primers or probe binding was determined by linear regression by plotting the cycle threshold (CT) value versus the log of the RNA dilution. The slopes for all reactions were determined to be similar, indicating similar reaction efficiencies. Relative quantification of transcript was determined using the comparative CT method calibrated to 16S rRNA (14). Quantitative RT-PCR experiments were performed multiple times independently with comparable results.
fig.ommitteed#*qiq, http://www.100md.com
TABLE 2. Primers and probes designed for TaqMan analysis of the hbp gene family#*qiq, http://www.100md.com
Statistical analysis. Student's t test results, standard error of the mean values, and graphs were generated with SigmaPlot 3.0 software (Jandel Scientific, San Rafael, Calif.). A P value of <0.05 was considered significant.#*qiq, http://www.100md.com
Nucleotide sequence accession numbers. GenBank accession numbers for the sequence data reported in this paper include: hbpA (AF266281), the hbpCAB locus (AY126673), hbpD (AY126674), and hbpE (AY126675).#*qiq, http://www.100md.com
RESULTS#*qiq, http://www.100md.com
Discovery of the hbp gene family. Sequence analysis of DNA flanking the hbpA gene revealed two closely linked homologs in the same orientation, designated hbpB and hbpC. This paralogous cluster of genes forms the hbpCAB locus . In addition to hbpCAB, two unlinked homologs, termed hbpD and hbpE, were subsequently identified using sequence data from the B. quintana genome project (O. Karlberg, B. Legault, K. Naslund, A. S. Eriksson, B. Lascola, M. Holmberg, and S. G. E. Andersson, unpublished data), bringing the hbp family membership to five genes.
fig.ommitteedy#, 百拇医药
FIG. 1. (A) Linkage map of the hbpCAB locus of B. quintana. Arrows indicate the positions of the open reading frames in the gene cluster. The gray box in hbpB indicates the position of the 510-bp insert with its nested 126-bp tandem repeat region shown as a white box (see Fig. 1B). (B) Variable-number tandem repeat region of hbpB. The number of repeats and their respective sequences are indicated.y#, 百拇医药
The hbpC, hbpD, and hbpE genes are 831, 882, and 897 bp in length, respectively; values that are similar to the 816-bp hbpA gene (6). In contrast, the hbpB gene is 1,359 bp in length due to a ~y#, 百拇医药
510-bp insert near the center of its open reading frame. This insert contains a nested 126-bp variable-number tandem repeat region comprised of 14 in-frame repeats . Without this insert, hbpB would be similar in length to other hbp family members. It is also interesting that the insert has an elevated G+C content relative to other genes in the B. quintana genome (~
50% versus 39% G+C [28]), including the other hbp genes. In addition, hbpB sequences that flank the insert have a typical G+C content for a B. quintana gene.i, http://www.100md.com
A multiple sequence alignment of the predicted proteins encoded by the hbp gene family reveals a high degree of amino acid sequence conservation (Fig. 2). The average sequence identity between Hbp family members is 48% (excluding HbpB). Each protein contains a predicted secretory signal sequence (see Fig. 2) as described for HbpA (6) and HbpD (this study) and contains a terminal phenylalanine. The prominent 36-kDa protein previously shown to copurify with HbpA in Triton X-114 extracts of B. quintana (6) was identified as HbpD by N-terminal sequencing. The N-terminal sequence of mature HbpD was determined to be ADVIIPEQPESVVAVPAFS, a perfect match to the predicted HbpD sequence shown in Fig. 2.i, http://www.100md.com
fig.ommitteedi, http://www.100md.com
FIG. 2. Multiple sequence alignment of the five Hbp family members. Identical residues are shaded in black; conserved residues are shaded in gray. Predicted ß-strand transmembrane domains are boxed and numbered. Transmembrane domains 5, 6, 7, and 8 are nearly identical to those in neisserial Opa proteins (15). The secretory signal sequence cleavage sites (as determined for HbpA [6] and HbpD [this study]) are indicated by an arrowhead.
Homologs from other bacteria. BLAST searches revealed that the closest homologs in the database (greatest to least) include B. henselae PAP31 (phage-associated/membrane protein [4]), Brucella OMP31 (putative porin [29, 30]), and A. tumefaciens OMP25 (immunogenic surface protein [9]). Sequence identities to HbpA range from 32% (OMP31) to 58% (Pap31). The function(s) of these related surface proteins has not been fully elucidated, and their widespread occurrence in several plant and animal pathogens within the -proteobacteria warrants further investigation. BLASTp searches with Hbps also generated numerous "hits" on the neisserial opacity (Opa) proteins. Although the overall sequence identity value between Hbps and Opa is only about 25%, the Hbp family members are approximately 40% similar to Opa, and considerable identity to Opa is observed in the last quarter of the Hbp molecule indicated by a Opa conserved domain (not shown).x{uz, 百拇医药
Secondary structure model for HbpA. Given the similarity between Opa and Hbps and the fact that the last four predicted transmembrane domains of HbpA to -E (see boxes 5, 6, 7, and 8 in Fig. 2) are nearly identical to those predicted in the two-dimensional model for Opa (15), a two-dimensional model of HbpA was generated using the methodology applied to Opa (15). The resulting ß-barrel model for HbpA contains eight transmembrane domains, the last terminating with a phenylalanine (Fig. 3). As is typical of outer membrane protein families, the predicted transmembrane strands for the Hbps correspond to the most-conserved sequences among the family members (Fig. 2) and likely serve as framework regions (10). The two-dimensional model for HbpA also indicates that the largest loops of the protein are extracellular, whereas the short loops are intracellular. An excellent example of this is the unusually large L2 loop that is predicted for HbpB, corresponding to the 510-bp GC-rich insert containing the variable number tandem repeat (Fig. 1). The predicted transmembrane domains of HbpA are antiparallel amphipathic ß strands as found in porins (8, 31) and in Opa (15). Many of the predicted ß strands of Hbp (Fig. 2 and 3) are flanked by aromatic residues; a characteristic of ß strands that span outer membranes (8, 31). Similar models can also be generated from the other Hbp family members (not shown), implying a conserved structure.
fig.ommitteed*sd*, http://www.100md.com
FIG. 3. Predicted two-dimensional ß-barrel structure for HbpA. The inner and outer faces of the outer membrane are indicated. The bold residues (shifted to the right) indicate the nonpolar side of the eight transmembrane ß strands. The transmembrane domains correspond to those boxed in Fig. 3.*sd*, http://www.100md.com
Expression of hbp family members. In vitro transcription-translation of cloned hbp genes shows that the apparent molecular masses for the Hbp proteins as determined by SDS-PAGE and/or autoradiographic analysis (not shown) are in close agreement with values from predicted amino acid sequences, with values of ~*sd*, http://www.100md.com
30 and 29.3 kDa (HbpA), 55.9 and 47.1 kDa (HbpB), 28.6 and 30.1 kDa (HbpC), 26.5 and 32.7 kDa (HbpD), and 35.1 and 33 kDa (HbpE), respectively. We attribute the discrepancy in HbpB values to aberrant SDS-PAGE migration resulting from its large and unusual insert sequence. Another discrepancy was observed between mature HbpD produced in vivo in B. quintana (~
36 kDa) (6) and recombinant, immature HbpD produced in vitro (26.5 kDa). The expected value for immature HbpD should be ca. 2.4 kDa greater than the mature protein by virtue of its intact signal sequence.j$ns}, 百拇医药
RT-PCR was performed to quantify relative expression levels of the hbp family members during bacterial growth on standard medium (HIAB). The data clearly show that all hbp family members are expressed under routine culture conditions. Further, a comparison of relative expression levels based on CT values reveals that hbp transcripts are produced in the following order (most abundant to rarest): hbpD, hbpA, hbpC, hbpE, and hbpB (Table 3).j$ns}, 百拇医药
fig.ommitteedj$ns}, 百拇医药
TABLE 3. Real-time PCRa of hemin-binding protein gene family in wild-type and hbpA mutant in B. quintanaj$ns}, 百拇医药
Effect of hbpA mutagenesis on the hemin-binding phenotype. B. quintana hbpA mutants were generated via allelic exchange after electroporation of pHBPA' suicide vector into strain LS200. One resulting mutant, strain LS300, was isolated from HIAB-Kan plates, and its mutation was verified by PCR analysis of hbpA (not shown) as previously described (3, 7). The protein profile for LS300 was also examined by SDS-PAGE, and immunoblots were developed with rabbit anti-HbpA, as previously described (6). The data clearly show that the prominent 30-kDa HbpA band in LS200 is absent in the protein profile of LS300 (Fig. 4A, lanes 2 and 3, respectively). Further, HbpA or a truncated version of the protein cannot be detected in the LS300 lysate by immunoblot analysis . Complementation of LS300 in trans reestablished synthesis of the HbpA protein to levels ca. 5% of the parental strain, LS200, and detection was limited to immunoblots (Fig. 4B, lane 4). This is the first report of site-directed mutation and trans-complementation in B. quintana.
fig.ommitteed':i*, http://www.100md.com
FIG. 4. Analysis of HbpA synthesis in a B. quintana hbpA mutant and trans-complemented strain. (A) Coomassie blue-stained SDS-PAGE gel containing cell lysates of the LS200 parental strain (lane 2), the hbpA mutant LS300 (lane 3) and the trans-complemented strain LS400 (lane 4). (B) Corresponding immunoblot. The HbpA protein is indicated by an arrowhead. Molecular mass standards (lane 1) are indicated to the left in kilodaltons.':i*, http://www.100md.com
To gauge the effect of the hbpA mutation on the hemin-binding phenotype of B. quintana, an in vitro hemin-binding assay was done as previously described (6). Much to our surprise, data from this assay showed that the LS300 hbpA mutant actually exhibits a significantly higher level of hemin-binding relative to the LS200 parental strain (16 ± 0.5 versus 10.3 ± 0.25 µg/mg of protein, respectively), whereas the LS400 trans-complemented strain exhibits an intermediate phenotype (12.4 ± 0.5 µg/mg of protein) (Fig. 5).':i*, http://www.100md.com
fig.ommitteed
FIG. 5. Hemin-binding assay of parental (LS200), hbpA mutant (LS300), and trans-complemented (LS400) strains of B. quintana. Data are expressed as the means plus the standard error of the mean for eight assays.cv|d[9!, 百拇医药
Quantitative RT-PCR analysis of wild-type versus mutant. The relative transcript levels of the hbp gene family in the wild-type and hbpA mutant in B. quintana were assessed by quantitative RT-PCR. The results are summarized in Table 3. When CT values were normalized to 16S rRNA, we determined that the transcription levels of hbpB and hbpC were increased 3.58- and 3.48-fold, respectively, in the mutant relative to the wild type. Even more striking was the observation that hbpE and hbpD were increased 5.94- and 13.74-fold, respectively, in the mutant compared to the wild type. If these homologs share receptor function, the significant increase in the level of transcription by the other four gene family members may explain the enhancement in hemin binding seen in the hbpA mutant.
RT-PCR also showed that hbpA expression was apparently decreased by 3.33-fold in the mutant (Table 3). We ascribe this moderate level of expression to two observations. First, Northern blot analysis using RNA from wild-type and mutant strains reveals a readthrough transcript of approximately 850 bp in the mutant versus a 1,000-bp message in the wild type (data not shown). It is likely that a promoter on the suicide vector-based insert is responsible. Second, the RT-PCR target region (nucleotides 647 to 718) is located 3' to the mutational target (nucleotides 344 to 583), making this transcript detectable.(, http://www.100md.com
DISCUSSION(, http://www.100md.com
B. quintana has the highest reported hemin requirement for bacterial growth in vitro (20 to 40 µg per ml of medium [17]). However, the reason for this extraordinary hemin requirement and the mechanisms involved in its acquisition are poorly characterized. In a previous study, we identified eight membrane-associated proteins in B. quintana that bind hemin in vitro (6). Of these, the most prominent was a 30-kDa outer membrane protein, HbpA, that appears to play a role as a hemin receptor for the pathogen.
Chromosomal walking of sequences that flank hbpA led to the discovery of two paralagous genes, which we termed hbpB and hbpC (Fig. 1). Sequence data obtained from the B. quintana genome project (University of Uppsala, Uppsala, Sweden) identified two additional, unlinked homologues termed hbpD and hbpE. Each of the five hbp genes, including the closely linked hbpCAB genes, possesses endogenous promoter regions, as demonstrated by expression in vitro from plasmids containing directionally cloned inserts in opposition to the lacZ promoter of the multiple cloning site. Further, these genes contain potential fur regulatory elements as previously described for hbpA (6), suggesting that they may be regulated by Fur. This is the first report of a multigenic family in Bartonella.a+}, 百拇医药
Members of the hbp gene family encode homologs that share approximately 50% amino acid sequence identity (Fig. 2), suggesting that structure and function may also be conserved. We hypothesize that all Hbp family members are located in the outer membrane based upon: (i) their considerable homology to HbpA, a known outer membrane protein (6); (ii) possession of a C-terminal phenylalanine (25); and (iii) a predicted (HbpB, -C, and -E) or verified (HbpA and -D) secretory signal sequence (Fig. 2).
Although the majority of Hbp family members are approximately 30 kDa, the HbpB protein is nearly 56 kDa due to an ~t, 百拇医药
510-bp insert in its respective gene. The insert is interesting by virtue of its 14 nested variable-number tandem repeats (Fig. 1B) and its aberrantly high GC richness compared to the B. quintana genome (~t, 百拇医药
50 versus 39% G+C; [28]). These features suggest that the element was derived from a foreign source such as a phage or transposable element. A second incongruity was observed between the molecular mass of recombinant, immature HbpD produced in vitro with that observed for mature HbpD isolated from Triton X-114 extracts of B. quintana (36 kDa [6]). It is possible that this disparity results from posttranslational modification of HbpD in Bartonella.t, 百拇医药
Numerous homologs of the Hbp proteins were identified by using BLAST searches, and many of these are surface proteins from closely related pathogenic -proteobacteria (e.g., Brucella and Agrobacterium). It is tempting to speculate that these homologs, together with the Hbp family, comprise a superfamily of related outer membrane proteins that may share at least some functions. One interesting homolog identified by BLAST searches was the neisserial Opa. A secondary structure prediction for HbpA reveals a potential ß-barrel structure containing eight transmembrane domains, four extracellular loops, and three intracellular loops. The predicted transmembrane domains are highly conserved in all Hbp family members (Fig. 2), and similar predictions can be made with other Hbp proteins (data not shown). Studies to verify this predicted topology are currently under way in our laboratory.
The existence of a multigene Hbp family in Bartonella might provide the pathogen with redundant "backup" systems for facilitating hemin acquisition or some other unknown function(s). Undoubtedly, many of the conserved regions of these molecules may be more closely related to structure than function. For example, the predicted conserved transmembrane domains may simply serve to anchor the extracellular loop domains for their designed activity.zy?, 百拇医药
Mutagenesis of hbpA was done in order to examine the effect of mutation on the hemin-binding phenotype of B. quintana and to establish a system of genetic manipulation for this bacterium. Using allelic exchange, we successfully mutagenized hbpA with a suicide vector. In addition, trans-complementation of the mutation with a shuttle vector was accomplished. The resulting strains were subsequently analyzed for their hemin-binding phenotype in vitro. We discovered that mutagenesis of hbpA actually rendered a strain that bound 56% more hemin than the parental strain, whereas reestablishment of hbpA expression in trans (to levels ca. 5% that of the parental strain) provided an intermediate phenotype (Fig. 5).
Although the hemin-binding assay measures both receptor and non-receptor-mediated binding, enhanced hemin binding by the hbpA mutant led us to hypothesize that alterations in the expression of the other hbp family members might be responsible for this phenotype. To investigate this possibility, we performed quantitative RT-PCR with RNA extracted from the wild type and the hbpA mutant. The results summarized in Table 3 suggest that hbpB, hbpC, hbpD, and hbpE are all upregulated in the mutant relative to the wild-type strain, even as high as 13.74-fold in the case of hbpD. Although it is possible that altered surface characteristics (e.g., hydrophobicity or charge) or other non-receptor-mediated hemin binding may be responsible for this observation (Fig. 5), the clear and significant upregulation of hbp family members suggests that a compensatory expression is taking place. This observation, coupled with the family's conservation in predicted structure and possession of putative Fur regulatory elements, suggests that other hbp gene products may also serve as hemin receptors. This hypothesis is currently under investigation in our laboratory.
This is the first multigene family described from a Bartonella species. As such, it presents a unique opportunity to investigate differential gene regulation in this poorly characterized bacterium. In addition, the hypothesized role for additional hbp family members in hemin binding underscores the potential importance of this gene family for virulence.95-1vv2, http://www.100md.com
ACKNOWLEDGMENTS95-1vv2, http://www.100md.com
We thank Patty McIntire at The University of Montana Murdock Molecular Biology Facility for sequence analyses, Mark Garfield at the Twinbrook II Facility at NIH for peptide sequencing, and Russ Regnery for B. quintana.95-1vv2, http://www.100md.com
M.F.M. was supported by Public Health Service grant AI45534 and American Heart Established Investigator grant 9940002N. J.A.C. was supported through a NIH-NIAID Intramural Research Training Award. K.R.S was supported through an NSF-EPSCoR undergraduate fellowship.95-1vv2, http://www.100md.com
REFERENCES95-1vv2, http://www.100md.com
Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic local alignment search tool. J. Mol. Biol. 215:403-410.
Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl. 1995. Current protocols in molecular biology. John Wiley & Sons, Inc., New York, N.Y.m[q-h+], 百拇医药
Battisti, J. M., and M. F. Minnick. 1999. Development of a system for genetic manipulation of Bartonella bacilliformis. Appl. Environ. Microbiol. 65:3441-3448.m[q-h+], 百拇医药
Bowers, T. J., D. Sweger, D. Jue, and B. Anderson. 1998. Isolation, sequencing and expression of the gene encoding a major protein from the bacteriophage associated with Bartonella henselae. Gene 206:49-52.m[q-h+], 百拇医药
Byam, W. 1919. Trench fever, p. 120-130. In L. L. Lloyd (ed.), Lice and their menace to man. Oxford University Press, Oxford, United Kingdom.m[q-h+], 百拇医药
Carroll, J. A., S. A. Coleman, L. S. Smitherman, and M. F. Minnick. 2000. Hemin-binding surface protein from Bartonella quintana. Infect. Immun. 68:6750-6757.m[q-h+], 百拇医药
Coleman, S. A., and M. F. Minnick. 2001. Establishing a direct role for the Bartonella bacilliformis invasion-associated locus B (IalB) protein in human erythrocyte parasitism. Infect. Immun. 69:4373-4381.
Cowan, S. W., T. Schirmer, G. Rummel, M. Steiert, R. Ghosh, R. A. Pauptit, J. N. Jansonius, and J. P. Rosenbusch. 1992. Crystal structures explain functional properties of two E. coli porins. Nature 358:727-733.l1flhu/, http://www.100md.com
Goodner, B., G. Hinkle, S. Gattung, N. Miller, M. Blanchard, B. Qurollo, et al. 2001. Genome sequence of the plant pathogen and biotechnology agent Agrobacterium tumefaciens. Science 294:2323-2328.l1flhu/, http://www.100md.com
Jeanteur, D., J. Lakey, and F. Pattus. 1994. The porin superfamily: diversity and common features, p. 363-380. In J. M. Ghuysen and R. Hakenbeck (ed.), Bacterial cell wall. Elsevier, Amsterdam, The Netherlands.l1flhu/, http://www.100md.com
Jackson, L. A., and D. H. Spach. 1996. Emergence of Bartonella quintana infection among homeless persons. Emerg. Infect. Dis. 2:141-144.l1flhu/, http://www.100md.com
Jackson, L. A., D. H. Spach, D. A. Kippen, N. K. Sugg, R. L. Regnery, M. H. Sayers, and W. E. Stamm. 1996. Seroprevalence to Bartonella quintana among patients at a community clinic in downtown Seattle. J. Infect. Dis. 173:1023-1026.
Koehler, J. E., F. D. Quinn, T. G. Berger, P. E. LeBoit, and J. W. Tappero. 1992. Isolation of Rochalimaea species from cutaneous and osseous lesions of bacillary angiomatosis. N. Engl. J. Med. 327:1625-1631.s'/, 百拇医药
Livak, K. J., and T. D. Schmittgen. 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2method. Methods 25:402-408.s'/, 百拇医药
Malorny, B., G. Morelli, B. Kusecek, J. Kolberg and M. Achtman. 1998. Sequence diversity, predicted two-dimensional protein structure and epitope mapping of neisserial Opa proteins. J. Bacteriol. 180:1323-1330.s'/, 百拇医药
Matsudaira, P. 1987. Sequence from picomole quantities of proteins electroblotted onto polyvinylidene difluoride membranes. J. Biol. Chem. 262:10035-10038.s'/, 百拇医药
Myers, W. F., L. D. Cutler, and C. L. Wisseman. 1969. Role of erythrocytes and serum in the nutrition of Rickettsia quintana. J. Bacteriol. 97:663-666.s'/, 百拇医药
Parrott, J. H., L. Dure, W. Sullender, W. Buraphacheep, T. A. Frye, C. A. Galliani, E. Marston, D. Jones, and R. Regnery. 1997. Central nervous system infection associated with Bartonella quintana: a report of two cases. Pediatrics 100:403-408.
Pearson, W. R. 1990. Rapid and sensitive sequence comparison with FASTP and FASTA. Methods Enzymol. 183:63-98.a, 百拇医药
Reschke, D. K., M. E. Frazier, and L. P. Mallavia. 1990. Transformation of Rochalimaea quintana, a member of the family Rickettsiaceae. J. Bacteriol. 172:5130-5134.a, 百拇医药
Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.a, 百拇医药
Scherer, D. C., I. DeBuron-Connors, and M. F. Minnick. 1993. Characterization of Bartonella bacilliformis flagella and effect of antiflagellin antibodies on invasion of erythrocytes. Infect. Immun. 61:4962-4971.a, 百拇医药
Spach, D. H., K. P. Callis, D. S. Paauw, Y. B. Houze, F. D. Schoenknecht, D. F. Welch, H. Rosen, and D. J. Brenner. 1993. Endocarditis caused by Rochalimaea quintana in a patient infected with human immunodeficiency virus. J. Clin. Microbiol. 31:692-694.a, 百拇医药
Spach, D. H., A. S. Kanter, M. J. Dougherty, A. M. Larson, M. B. Coyle, D. J. Brenner, B. Swaminathan, G. M. Matar, D. F. Welch, R. K. Root, and W. E. Stamm. 1995. Bartonella (Rochalimaea) quintana bacteremia in inner-city patients with chronic alcoholism. N. Engl. J. Med. 332:424-428.
Struyve, M., M. Moons, and J. Tommassen. 1991. Carboxy-terminal phenylalanine is essential for the correct assembly of a bacterial outer membrane protein. J. Mol. Biol. 218:141-148.601], http://www.100md.com
Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673-4680.601], http://www.100md.com
Towbin, H., T. Staehelin, and J. Gordon. 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 76:4350-4354.601], http://www.100md.com
Tyeryar, F. J., E. Weiss, D. B. Millar, F. M. Bozeman, and R. A. Ormsbee. 1973. DNA base composition of rickettsiae. Science 180:415-417.601], http://www.100md.com
Vizcaino, N., A. Cloeckaert, M. S. Zygmunt, and G. Dubray. 1996. Cloning, nucleotide sequence, and expression of the Brucella melitensis omp31 gene coding for an immunogenic major outer membrane protein. Infect. Immun. 64:3744-3751.601], http://www.100md.com
Vizcaino, N., J. M. Verger, M. Grayon, M. S. Zygmunt, and A. Cloeckaert. 1997. DNA polymorphism at the omp-31 locus of Brucella spp.: evidence for a large deletion in Brucella aborutus, and other species-specific markers. Microbiology 143:2913-2921.601], http://www.100md.com
Weiss, M. S., U. Abele, J. Weckesser, W. Welte, E. Schiltz, and G. E. Schulz. 1991. Molecular architecture and electrostatic properties of a bacterial porin. Science 254:1627-1630.601], http://www.100md.com
Weiss, E., and G. A. Dasch. 1982. Differential characteristics of strains of Rochalimaea: Rochalimaea vinsonii sp. nov., the Canadian vole agent. Int. J. Syst. Bacteriol. 32:305-314.(Michael F. Minnick Kate N. Sappington Laura S. Smitherman Siv G. E. Andersson Olof Karlberg and Jame)
Received 16 September 2002/ Returned for modification 10 October 2002/ Accepted 1 November 2002e#rg, 百拇医药
ABSTRACTe#rg, 百拇医药
Bartonella quintana, the agent of trench fever and an etiologic agent of bacillary angiomatosis, has an extraordinarily high hemin requirement for growth compared to other bacterial pathogens. We previously identified the major hemin receptor of the pathogen as a 30-kDa surface protein, termed HbpA. This report describes four additional homologues that share approximately 48% amino acid sequence identity with hbpA. Three of the genes form a paralagous cluster, termed hbpCAB, whereas the other members, hbpD and hbpE, are unlinked. Secondary structure predictions and other evidence suggest that Hbp family members are ß-barrels located in the outer membrane and contain eight transmembrane domains plus four extracellular loops. Homologs from a variety of gram-negative pathogens were identified, including Bartonella henselae Pap31, Brucella Omp31, Agrobacterium tumefaciens Omp25, and neisserial opacity proteins (Opa). Family members expressed in vitro-synthesized proteins ranging from ca. 26.5 to 35.1 kDa, with the exception of HbpB, an ~
55.9-kDa protein whose respective gene has been disrupted by a ~o8?, 百拇医药
510 GC-rich element containing variable-number tandem repeats. Transcription analysis by quantitative reverse transcriptase-PCR (RT-PCR) indicates that all family members are expressed under normal culture conditions, with hbpD and hbpB transcripts being the most abundant and the rarest, respectively. Mutagenesis of hbpA by allelic exchange produced a strain that exhibited an enhanced hemin-binding phenotype relative to the parental strain, and analysis by quantitative RT-PCR showed elevated transcript levels for the other hbp family members, suggesting that compensatory expression occurs.o8?, 百拇医药
INTRODUCTIONo8?, 百拇医药
Bartonella quintana is the bacterial agent of trench fever and is transmitted between humans by the bite of the human body louse (Pediculus humanis) (5). B. quintana was a major cause of morbidity during World Wars I and II and is reemerging as a secondary infectious agent primarily in immunocompromised patients and a cause of "urban trench fever" in homeless, inner-city dwellers (11, 12). B. quintana infections can manifest as typical trench fever (11), bacillary angiomatosis (13), chronic bacteremia (24), endocarditis (23), lymphadenpathy (13), or infections of the nervous and skeletal systems (13, 18). These conditions can occur concurrently and may be life-threatening.
Very little is known of B. quintana's virulence determinants, epidemiology, or the reason for its reemergence. It is known that this bacterium has a high requirement for hemin. Investigation into this extraordinary hemin requirement led to our identification and characterization of a major hemin-binding protein, hemin-binding protein A (HbpA), from B. quintana (6). HbpA is a heat-modifiable outer surface protein that retains its ability to bind hemin after sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).m, http://www.100md.com
We have determined that hbpA belongs to a gene family consisting of five genes. Here we describe the four additional members of this multigenic family, and report the first successful site-directed mutagenesis and trans-complementation in this Bartonella species using hbpA as the gene target. We use quantitative reverse transcription-PCR (RT-PCR) to determine the relative transcript level of each hbp gene family member and compare transcript levels in the wild-type to that of the hbpA mutant.
MATERIALS AND METHODS8, 百拇医药
Bacterial strains and culture conditions. B. quintana was grown on heart infusion agar blood plates (HIAB [6]) at 37°C in 5% CO2 and 100% relative humidity and was harvested at 3 days postinoculation (approximately mid-log phase [32]). E. coli was grown in Luria-Bertani medium for 16 h at 37°C. When required, antibiotic supplements were added to the medium at standard concentrations (21). B. quintana and Escherichia coli strains used or generated in this study are summarized in Table 1.8, 百拇医药
fig.ommitteed8, 百拇医药
TABLE 1. Bacterial strains and plasmids used in this study8, 百拇医药
Preparation and manipulation of nucleic acids. Plasmids used or generated in this study are listed in Table 1. Plasmids intended for routine use were purified from E. coli with a Perfectprep kit (Eppendorf Scientific, Westbury, N.Y.). Plasmids used in sequencing, in vitro transcription-translation, or electroporation were prepared with a Midi-Prep kit (Qiagen, Valencia, Calif.). Genomic DNA was prepared by a hexadecyltrimethyl ammonium bromide technique (2). RNA was isolated using an Ultraspec-II RNA isolation system (Biotex, Huston, Tex.) and contaminating DNA was removed using DNA-Free from Ambion (Austin, Tex.). RNA purity was assayed by a standard protocol (2). Clonings, ligations, and transformation reactions with E. coli were performed as previously described (6).
Nucleotide sequencing and analysis. Both DNA strands were sequenced using a BigDye Terminator Cycle Sequencing Ready Reaction Kit (Applied Biosystems, Inc. [ABI]/Roche, Branchburg, N.J.) and an automated DNA sequencer (ABI; model 377). Sequence data were compiled and analyzed with Seqweb version 2.0 (Accelrys, San Diego, Calif.) or the National Center for Biotechnology Information website. BLAST 2.0 (1) was used for database searches, whereas sequence alignments were done with FASTA 2.0 (19), CLUSTALW 1.6 (26), and BOXSHADE 3.21 .f!w, 百拇医药
SDS-PAGE, immunoblots, and N-terminal sequencing. Protein samples (20 µg total) were separated on SDS-PAGE gels (12.5% [wt/vol] acrylamide) prepared by standard protocol (2). For immunoblots, gels were transferred overnight to supported nitrocellulose (0.45-µm pore size; Osmonics, Minnetonka, Minn.) by the general methods of Towbin et al. (27). The resulting blot was probed by using rabbit anti-HbpA antiserum prepared as before (6) and subsequently developed by using horseradish peroxidase-conjugated goat anti-rabbit immunoglobulin G antibody (Sigma), H2O2, and 4-chloronapthol as previously detailed (22). For N-terminal sequencing, a Triton X-114 precipitate was prepared from B. quintana as previously described (6) and transferred to polyvinylidene difluoride (16). The polyvinylidene difluoride was stained for 15 min with 0.05% Coomassie blue and rinsed briefly, and then the HbpD protein band excised, dried, and subjected to Edman degradation by using an ABI 431A automated peptide sequencer. Sequencing was performed on two separate samples.
Mutagenesis and trans-complementation of hbpA. Mutagenesis of B. quintana was done by using a strategy we previously described for Bartonella bacilliformis (3). Briefly, a transformation-competent strain of B. quintana (L200) was prepared by electroporating wild-type B. quintana (OK 90-268) with pEST. Transformants were subsequently selected on HIAB-Kan plates (i.e., HIAB plus 25 µg of kanamycin/ml) and one resulting strain (LS100) was cured of pEST by six serial culture passages on HIAB. The resulting strain, LS200, was verified as cured of pEST by PCR analysis and kanamycin sensitivity as previously described (3). LS200 was subsequently electroporated with a suicide vector containing a 240-bp internal fragment of hbpA (nucleotides 344 to 583), termed pHBPA'. Mutants were selected on HIAB-Kan. One resulting strain, LS300, was verified as a hbpA mutant by PCR, SDS-PAGE, and immunoblotting. Finally, LS300 was complemented in trans by transforming with the shuttle vector, pBBR1-HBPA, to generate strain LS400.
Hemin-binding assay. Hemin binding by intact B. quintana cells was assayed in vitro essentially as before (6). Briefly, eight plates of B. quintana were harvested into 1.0 ml of 100 mM Tris (pH 8.0) and washed four times by centrifuging the suspension for 5 min at 2,940 x g and resuspending the resulting pellet into 1.0 ml of 100 mM Tris (pH 8.0). The final pellet was resuspended to an optical density at 600 nm (OD600) of 1.0, and four 1-ml aliquots were obtained from each strain. A total of 5 µg of hemin (5 µl of a fresh 1-mg/ml hemin stock solution in 0.02 N NaOH) were added to each tube, gently mixed, and incubated open for 1 h at 37°C at 5% CO2, with gentle mixing at 15-min intervals. Four negative control tubes without B. quintana cells were prepared and incubated as well. After incubation, the suspensions were pelleted by centrifuging for 2 min at 16,000 x g, and the resulting supernatants clarified twice by transferring to new microcentrifuges tube and centrifuging again at 16,000 x g. The OD400 of the final supernatants was assayed, and hemin binding was determined by comparing the reduction in OD400 to negative controls.
In vitro transcription-translation. Plasmids containing individual hbp genes plus their respective promoters were directionally cloned into pBluescript SK(+/-) in opposite orientation to the lacZ promoter. The resulting plasmids (Table 1) were used as templates for a S30 extract kit for circular DNA per the manufacturer's instructions (Promega, Madison, Wis.). Proteins were radiolabeled with a [35S]cysteine-methionine mix (Express; New England Nuclear, Boston, Mass.). Translation products were separated on SDS-PAGE (12.5% [wt/vol] acrylamide) and visualized by exposure of the dried gel to X-ray film overnight.4;, 百拇医药
RT-PCR quantification of hbp family transcripts. Quantitative RT-PCR was performed using TaqMan One-Step RT-PCR master mix from ABI. Ten-microliter reactions were performed in triplicate in a 384-well format, and reactions contained 500 nM concentrations of each primer, 100 nM probe, Master Mix and MultiScribe, and RNase inhibitor Mix to 1x (ABI); RNA was then serially diluted twofold from 5 ng per reaction to 0.153 ng per reaction. Probes consisted of an oligonucleotide labeled at the 5' end with the reporter dye 5-carboxyfluorescein and at the 3' end with the quencher N,N',N'-tetramethyl-6-carboxyrhodamine. Primers and probes used in this study are listed in Table 2. Quantitative RT-PCR conditions were as follows: 1 cycle at 50°C for 30 min, 1 cycle at 95°C for 10 min, and 40 cycles of 95°C for 15 s and of 60°C for 60 s. Changes in fluorescence were monitored by using an ABI 7900HT sequence detection system and raw data were analyzed by SDS software version 2.0 (ABI). Chromosomal DNA from B. quintana was used as a control to ensure that the primers or probe from each gene were binding at similar efficiencies. The efficiency of primers or probe binding was determined by linear regression by plotting the cycle threshold (CT) value versus the log of the RNA dilution. The slopes for all reactions were determined to be similar, indicating similar reaction efficiencies. Relative quantification of transcript was determined using the comparative CT method calibrated to 16S rRNA (14). Quantitative RT-PCR experiments were performed multiple times independently with comparable results.
fig.ommitteed#*qiq, http://www.100md.com
TABLE 2. Primers and probes designed for TaqMan analysis of the hbp gene family#*qiq, http://www.100md.com
Statistical analysis. Student's t test results, standard error of the mean values, and graphs were generated with SigmaPlot 3.0 software (Jandel Scientific, San Rafael, Calif.). A P value of <0.05 was considered significant.#*qiq, http://www.100md.com
Nucleotide sequence accession numbers. GenBank accession numbers for the sequence data reported in this paper include: hbpA (AF266281), the hbpCAB locus (AY126673), hbpD (AY126674), and hbpE (AY126675).#*qiq, http://www.100md.com
RESULTS#*qiq, http://www.100md.com
Discovery of the hbp gene family. Sequence analysis of DNA flanking the hbpA gene revealed two closely linked homologs in the same orientation, designated hbpB and hbpC. This paralogous cluster of genes forms the hbpCAB locus . In addition to hbpCAB, two unlinked homologs, termed hbpD and hbpE, were subsequently identified using sequence data from the B. quintana genome project (O. Karlberg, B. Legault, K. Naslund, A. S. Eriksson, B. Lascola, M. Holmberg, and S. G. E. Andersson, unpublished data), bringing the hbp family membership to five genes.
fig.ommitteedy#, 百拇医药
FIG. 1. (A) Linkage map of the hbpCAB locus of B. quintana. Arrows indicate the positions of the open reading frames in the gene cluster. The gray box in hbpB indicates the position of the 510-bp insert with its nested 126-bp tandem repeat region shown as a white box (see Fig. 1B). (B) Variable-number tandem repeat region of hbpB. The number of repeats and their respective sequences are indicated.y#, 百拇医药
The hbpC, hbpD, and hbpE genes are 831, 882, and 897 bp in length, respectively; values that are similar to the 816-bp hbpA gene (6). In contrast, the hbpB gene is 1,359 bp in length due to a ~y#, 百拇医药
510-bp insert near the center of its open reading frame. This insert contains a nested 126-bp variable-number tandem repeat region comprised of 14 in-frame repeats . Without this insert, hbpB would be similar in length to other hbp family members. It is also interesting that the insert has an elevated G+C content relative to other genes in the B. quintana genome (~
50% versus 39% G+C [28]), including the other hbp genes. In addition, hbpB sequences that flank the insert have a typical G+C content for a B. quintana gene.i, http://www.100md.com
A multiple sequence alignment of the predicted proteins encoded by the hbp gene family reveals a high degree of amino acid sequence conservation (Fig. 2). The average sequence identity between Hbp family members is 48% (excluding HbpB). Each protein contains a predicted secretory signal sequence (see Fig. 2) as described for HbpA (6) and HbpD (this study) and contains a terminal phenylalanine. The prominent 36-kDa protein previously shown to copurify with HbpA in Triton X-114 extracts of B. quintana (6) was identified as HbpD by N-terminal sequencing. The N-terminal sequence of mature HbpD was determined to be ADVIIPEQPESVVAVPAFS, a perfect match to the predicted HbpD sequence shown in Fig. 2.i, http://www.100md.com
fig.ommitteedi, http://www.100md.com
FIG. 2. Multiple sequence alignment of the five Hbp family members. Identical residues are shaded in black; conserved residues are shaded in gray. Predicted ß-strand transmembrane domains are boxed and numbered. Transmembrane domains 5, 6, 7, and 8 are nearly identical to those in neisserial Opa proteins (15). The secretory signal sequence cleavage sites (as determined for HbpA [6] and HbpD [this study]) are indicated by an arrowhead.
Homologs from other bacteria. BLAST searches revealed that the closest homologs in the database (greatest to least) include B. henselae PAP31 (phage-associated/membrane protein [4]), Brucella OMP31 (putative porin [29, 30]), and A. tumefaciens OMP25 (immunogenic surface protein [9]). Sequence identities to HbpA range from 32% (OMP31) to 58% (Pap31). The function(s) of these related surface proteins has not been fully elucidated, and their widespread occurrence in several plant and animal pathogens within the -proteobacteria warrants further investigation. BLASTp searches with Hbps also generated numerous "hits" on the neisserial opacity (Opa) proteins. Although the overall sequence identity value between Hbps and Opa is only about 25%, the Hbp family members are approximately 40% similar to Opa, and considerable identity to Opa is observed in the last quarter of the Hbp molecule indicated by a Opa conserved domain (not shown).x{uz, 百拇医药
Secondary structure model for HbpA. Given the similarity between Opa and Hbps and the fact that the last four predicted transmembrane domains of HbpA to -E (see boxes 5, 6, 7, and 8 in Fig. 2) are nearly identical to those predicted in the two-dimensional model for Opa (15), a two-dimensional model of HbpA was generated using the methodology applied to Opa (15). The resulting ß-barrel model for HbpA contains eight transmembrane domains, the last terminating with a phenylalanine (Fig. 3). As is typical of outer membrane protein families, the predicted transmembrane strands for the Hbps correspond to the most-conserved sequences among the family members (Fig. 2) and likely serve as framework regions (10). The two-dimensional model for HbpA also indicates that the largest loops of the protein are extracellular, whereas the short loops are intracellular. An excellent example of this is the unusually large L2 loop that is predicted for HbpB, corresponding to the 510-bp GC-rich insert containing the variable number tandem repeat (Fig. 1). The predicted transmembrane domains of HbpA are antiparallel amphipathic ß strands as found in porins (8, 31) and in Opa (15). Many of the predicted ß strands of Hbp (Fig. 2 and 3) are flanked by aromatic residues; a characteristic of ß strands that span outer membranes (8, 31). Similar models can also be generated from the other Hbp family members (not shown), implying a conserved structure.
fig.ommitteed*sd*, http://www.100md.com
FIG. 3. Predicted two-dimensional ß-barrel structure for HbpA. The inner and outer faces of the outer membrane are indicated. The bold residues (shifted to the right) indicate the nonpolar side of the eight transmembrane ß strands. The transmembrane domains correspond to those boxed in Fig. 3.*sd*, http://www.100md.com
Expression of hbp family members. In vitro transcription-translation of cloned hbp genes shows that the apparent molecular masses for the Hbp proteins as determined by SDS-PAGE and/or autoradiographic analysis (not shown) are in close agreement with values from predicted amino acid sequences, with values of ~*sd*, http://www.100md.com
30 and 29.3 kDa (HbpA), 55.9 and 47.1 kDa (HbpB), 28.6 and 30.1 kDa (HbpC), 26.5 and 32.7 kDa (HbpD), and 35.1 and 33 kDa (HbpE), respectively. We attribute the discrepancy in HbpB values to aberrant SDS-PAGE migration resulting from its large and unusual insert sequence. Another discrepancy was observed between mature HbpD produced in vivo in B. quintana (~
36 kDa) (6) and recombinant, immature HbpD produced in vitro (26.5 kDa). The expected value for immature HbpD should be ca. 2.4 kDa greater than the mature protein by virtue of its intact signal sequence.j$ns}, 百拇医药
RT-PCR was performed to quantify relative expression levels of the hbp family members during bacterial growth on standard medium (HIAB). The data clearly show that all hbp family members are expressed under routine culture conditions. Further, a comparison of relative expression levels based on CT values reveals that hbp transcripts are produced in the following order (most abundant to rarest): hbpD, hbpA, hbpC, hbpE, and hbpB (Table 3).j$ns}, 百拇医药
fig.ommitteedj$ns}, 百拇医药
TABLE 3. Real-time PCRa of hemin-binding protein gene family in wild-type and hbpA mutant in B. quintanaj$ns}, 百拇医药
Effect of hbpA mutagenesis on the hemin-binding phenotype. B. quintana hbpA mutants were generated via allelic exchange after electroporation of pHBPA' suicide vector into strain LS200. One resulting mutant, strain LS300, was isolated from HIAB-Kan plates, and its mutation was verified by PCR analysis of hbpA (not shown) as previously described (3, 7). The protein profile for LS300 was also examined by SDS-PAGE, and immunoblots were developed with rabbit anti-HbpA, as previously described (6). The data clearly show that the prominent 30-kDa HbpA band in LS200 is absent in the protein profile of LS300 (Fig. 4A, lanes 2 and 3, respectively). Further, HbpA or a truncated version of the protein cannot be detected in the LS300 lysate by immunoblot analysis . Complementation of LS300 in trans reestablished synthesis of the HbpA protein to levels ca. 5% of the parental strain, LS200, and detection was limited to immunoblots (Fig. 4B, lane 4). This is the first report of site-directed mutation and trans-complementation in B. quintana.
fig.ommitteed':i*, http://www.100md.com
FIG. 4. Analysis of HbpA synthesis in a B. quintana hbpA mutant and trans-complemented strain. (A) Coomassie blue-stained SDS-PAGE gel containing cell lysates of the LS200 parental strain (lane 2), the hbpA mutant LS300 (lane 3) and the trans-complemented strain LS400 (lane 4). (B) Corresponding immunoblot. The HbpA protein is indicated by an arrowhead. Molecular mass standards (lane 1) are indicated to the left in kilodaltons.':i*, http://www.100md.com
To gauge the effect of the hbpA mutation on the hemin-binding phenotype of B. quintana, an in vitro hemin-binding assay was done as previously described (6). Much to our surprise, data from this assay showed that the LS300 hbpA mutant actually exhibits a significantly higher level of hemin-binding relative to the LS200 parental strain (16 ± 0.5 versus 10.3 ± 0.25 µg/mg of protein, respectively), whereas the LS400 trans-complemented strain exhibits an intermediate phenotype (12.4 ± 0.5 µg/mg of protein) (Fig. 5).':i*, http://www.100md.com
fig.ommitteed
FIG. 5. Hemin-binding assay of parental (LS200), hbpA mutant (LS300), and trans-complemented (LS400) strains of B. quintana. Data are expressed as the means plus the standard error of the mean for eight assays.cv|d[9!, 百拇医药
Quantitative RT-PCR analysis of wild-type versus mutant. The relative transcript levels of the hbp gene family in the wild-type and hbpA mutant in B. quintana were assessed by quantitative RT-PCR. The results are summarized in Table 3. When CT values were normalized to 16S rRNA, we determined that the transcription levels of hbpB and hbpC were increased 3.58- and 3.48-fold, respectively, in the mutant relative to the wild type. Even more striking was the observation that hbpE and hbpD were increased 5.94- and 13.74-fold, respectively, in the mutant compared to the wild type. If these homologs share receptor function, the significant increase in the level of transcription by the other four gene family members may explain the enhancement in hemin binding seen in the hbpA mutant.
RT-PCR also showed that hbpA expression was apparently decreased by 3.33-fold in the mutant (Table 3). We ascribe this moderate level of expression to two observations. First, Northern blot analysis using RNA from wild-type and mutant strains reveals a readthrough transcript of approximately 850 bp in the mutant versus a 1,000-bp message in the wild type (data not shown). It is likely that a promoter on the suicide vector-based insert is responsible. Second, the RT-PCR target region (nucleotides 647 to 718) is located 3' to the mutational target (nucleotides 344 to 583), making this transcript detectable.(, http://www.100md.com
DISCUSSION(, http://www.100md.com
B. quintana has the highest reported hemin requirement for bacterial growth in vitro (20 to 40 µg per ml of medium [17]). However, the reason for this extraordinary hemin requirement and the mechanisms involved in its acquisition are poorly characterized. In a previous study, we identified eight membrane-associated proteins in B. quintana that bind hemin in vitro (6). Of these, the most prominent was a 30-kDa outer membrane protein, HbpA, that appears to play a role as a hemin receptor for the pathogen.
Chromosomal walking of sequences that flank hbpA led to the discovery of two paralagous genes, which we termed hbpB and hbpC (Fig. 1). Sequence data obtained from the B. quintana genome project (University of Uppsala, Uppsala, Sweden) identified two additional, unlinked homologues termed hbpD and hbpE. Each of the five hbp genes, including the closely linked hbpCAB genes, possesses endogenous promoter regions, as demonstrated by expression in vitro from plasmids containing directionally cloned inserts in opposition to the lacZ promoter of the multiple cloning site. Further, these genes contain potential fur regulatory elements as previously described for hbpA (6), suggesting that they may be regulated by Fur. This is the first report of a multigenic family in Bartonella.a+}, 百拇医药
Members of the hbp gene family encode homologs that share approximately 50% amino acid sequence identity (Fig. 2), suggesting that structure and function may also be conserved. We hypothesize that all Hbp family members are located in the outer membrane based upon: (i) their considerable homology to HbpA, a known outer membrane protein (6); (ii) possession of a C-terminal phenylalanine (25); and (iii) a predicted (HbpB, -C, and -E) or verified (HbpA and -D) secretory signal sequence (Fig. 2).
Although the majority of Hbp family members are approximately 30 kDa, the HbpB protein is nearly 56 kDa due to an ~t, 百拇医药
510-bp insert in its respective gene. The insert is interesting by virtue of its 14 nested variable-number tandem repeats (Fig. 1B) and its aberrantly high GC richness compared to the B. quintana genome (~t, 百拇医药
50 versus 39% G+C; [28]). These features suggest that the element was derived from a foreign source such as a phage or transposable element. A second incongruity was observed between the molecular mass of recombinant, immature HbpD produced in vitro with that observed for mature HbpD isolated from Triton X-114 extracts of B. quintana (36 kDa [6]). It is possible that this disparity results from posttranslational modification of HbpD in Bartonella.t, 百拇医药
Numerous homologs of the Hbp proteins were identified by using BLAST searches, and many of these are surface proteins from closely related pathogenic -proteobacteria (e.g., Brucella and Agrobacterium). It is tempting to speculate that these homologs, together with the Hbp family, comprise a superfamily of related outer membrane proteins that may share at least some functions. One interesting homolog identified by BLAST searches was the neisserial Opa. A secondary structure prediction for HbpA reveals a potential ß-barrel structure containing eight transmembrane domains, four extracellular loops, and three intracellular loops. The predicted transmembrane domains are highly conserved in all Hbp family members (Fig. 2), and similar predictions can be made with other Hbp proteins (data not shown). Studies to verify this predicted topology are currently under way in our laboratory.
The existence of a multigene Hbp family in Bartonella might provide the pathogen with redundant "backup" systems for facilitating hemin acquisition or some other unknown function(s). Undoubtedly, many of the conserved regions of these molecules may be more closely related to structure than function. For example, the predicted conserved transmembrane domains may simply serve to anchor the extracellular loop domains for their designed activity.zy?, 百拇医药
Mutagenesis of hbpA was done in order to examine the effect of mutation on the hemin-binding phenotype of B. quintana and to establish a system of genetic manipulation for this bacterium. Using allelic exchange, we successfully mutagenized hbpA with a suicide vector. In addition, trans-complementation of the mutation with a shuttle vector was accomplished. The resulting strains were subsequently analyzed for their hemin-binding phenotype in vitro. We discovered that mutagenesis of hbpA actually rendered a strain that bound 56% more hemin than the parental strain, whereas reestablishment of hbpA expression in trans (to levels ca. 5% that of the parental strain) provided an intermediate phenotype (Fig. 5).
Although the hemin-binding assay measures both receptor and non-receptor-mediated binding, enhanced hemin binding by the hbpA mutant led us to hypothesize that alterations in the expression of the other hbp family members might be responsible for this phenotype. To investigate this possibility, we performed quantitative RT-PCR with RNA extracted from the wild type and the hbpA mutant. The results summarized in Table 3 suggest that hbpB, hbpC, hbpD, and hbpE are all upregulated in the mutant relative to the wild-type strain, even as high as 13.74-fold in the case of hbpD. Although it is possible that altered surface characteristics (e.g., hydrophobicity or charge) or other non-receptor-mediated hemin binding may be responsible for this observation (Fig. 5), the clear and significant upregulation of hbp family members suggests that a compensatory expression is taking place. This observation, coupled with the family's conservation in predicted structure and possession of putative Fur regulatory elements, suggests that other hbp gene products may also serve as hemin receptors. This hypothesis is currently under investigation in our laboratory.
This is the first multigene family described from a Bartonella species. As such, it presents a unique opportunity to investigate differential gene regulation in this poorly characterized bacterium. In addition, the hypothesized role for additional hbp family members in hemin binding underscores the potential importance of this gene family for virulence.95-1vv2, http://www.100md.com
ACKNOWLEDGMENTS95-1vv2, http://www.100md.com
We thank Patty McIntire at The University of Montana Murdock Molecular Biology Facility for sequence analyses, Mark Garfield at the Twinbrook II Facility at NIH for peptide sequencing, and Russ Regnery for B. quintana.95-1vv2, http://www.100md.com
M.F.M. was supported by Public Health Service grant AI45534 and American Heart Established Investigator grant 9940002N. J.A.C. was supported through a NIH-NIAID Intramural Research Training Award. K.R.S was supported through an NSF-EPSCoR undergraduate fellowship.95-1vv2, http://www.100md.com
REFERENCES95-1vv2, http://www.100md.com
Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic local alignment search tool. J. Mol. Biol. 215:403-410.
Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl. 1995. Current protocols in molecular biology. John Wiley & Sons, Inc., New York, N.Y.m[q-h+], 百拇医药
Battisti, J. M., and M. F. Minnick. 1999. Development of a system for genetic manipulation of Bartonella bacilliformis. Appl. Environ. Microbiol. 65:3441-3448.m[q-h+], 百拇医药
Bowers, T. J., D. Sweger, D. Jue, and B. Anderson. 1998. Isolation, sequencing and expression of the gene encoding a major protein from the bacteriophage associated with Bartonella henselae. Gene 206:49-52.m[q-h+], 百拇医药
Byam, W. 1919. Trench fever, p. 120-130. In L. L. Lloyd (ed.), Lice and their menace to man. Oxford University Press, Oxford, United Kingdom.m[q-h+], 百拇医药
Carroll, J. A., S. A. Coleman, L. S. Smitherman, and M. F. Minnick. 2000. Hemin-binding surface protein from Bartonella quintana. Infect. Immun. 68:6750-6757.m[q-h+], 百拇医药
Coleman, S. A., and M. F. Minnick. 2001. Establishing a direct role for the Bartonella bacilliformis invasion-associated locus B (IalB) protein in human erythrocyte parasitism. Infect. Immun. 69:4373-4381.
Cowan, S. W., T. Schirmer, G. Rummel, M. Steiert, R. Ghosh, R. A. Pauptit, J. N. Jansonius, and J. P. Rosenbusch. 1992. Crystal structures explain functional properties of two E. coli porins. Nature 358:727-733.l1flhu/, http://www.100md.com
Goodner, B., G. Hinkle, S. Gattung, N. Miller, M. Blanchard, B. Qurollo, et al. 2001. Genome sequence of the plant pathogen and biotechnology agent Agrobacterium tumefaciens. Science 294:2323-2328.l1flhu/, http://www.100md.com
Jeanteur, D., J. Lakey, and F. Pattus. 1994. The porin superfamily: diversity and common features, p. 363-380. In J. M. Ghuysen and R. Hakenbeck (ed.), Bacterial cell wall. Elsevier, Amsterdam, The Netherlands.l1flhu/, http://www.100md.com
Jackson, L. A., and D. H. Spach. 1996. Emergence of Bartonella quintana infection among homeless persons. Emerg. Infect. Dis. 2:141-144.l1flhu/, http://www.100md.com
Jackson, L. A., D. H. Spach, D. A. Kippen, N. K. Sugg, R. L. Regnery, M. H. Sayers, and W. E. Stamm. 1996. Seroprevalence to Bartonella quintana among patients at a community clinic in downtown Seattle. J. Infect. Dis. 173:1023-1026.
Koehler, J. E., F. D. Quinn, T. G. Berger, P. E. LeBoit, and J. W. Tappero. 1992. Isolation of Rochalimaea species from cutaneous and osseous lesions of bacillary angiomatosis. N. Engl. J. Med. 327:1625-1631.s'/, 百拇医药
Livak, K. J., and T. D. Schmittgen. 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2method. Methods 25:402-408.s'/, 百拇医药
Malorny, B., G. Morelli, B. Kusecek, J. Kolberg and M. Achtman. 1998. Sequence diversity, predicted two-dimensional protein structure and epitope mapping of neisserial Opa proteins. J. Bacteriol. 180:1323-1330.s'/, 百拇医药
Matsudaira, P. 1987. Sequence from picomole quantities of proteins electroblotted onto polyvinylidene difluoride membranes. J. Biol. Chem. 262:10035-10038.s'/, 百拇医药
Myers, W. F., L. D. Cutler, and C. L. Wisseman. 1969. Role of erythrocytes and serum in the nutrition of Rickettsia quintana. J. Bacteriol. 97:663-666.s'/, 百拇医药
Parrott, J. H., L. Dure, W. Sullender, W. Buraphacheep, T. A. Frye, C. A. Galliani, E. Marston, D. Jones, and R. Regnery. 1997. Central nervous system infection associated with Bartonella quintana: a report of two cases. Pediatrics 100:403-408.
Pearson, W. R. 1990. Rapid and sensitive sequence comparison with FASTP and FASTA. Methods Enzymol. 183:63-98.a, 百拇医药
Reschke, D. K., M. E. Frazier, and L. P. Mallavia. 1990. Transformation of Rochalimaea quintana, a member of the family Rickettsiaceae. J. Bacteriol. 172:5130-5134.a, 百拇医药
Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.a, 百拇医药
Scherer, D. C., I. DeBuron-Connors, and M. F. Minnick. 1993. Characterization of Bartonella bacilliformis flagella and effect of antiflagellin antibodies on invasion of erythrocytes. Infect. Immun. 61:4962-4971.a, 百拇医药
Spach, D. H., K. P. Callis, D. S. Paauw, Y. B. Houze, F. D. Schoenknecht, D. F. Welch, H. Rosen, and D. J. Brenner. 1993. Endocarditis caused by Rochalimaea quintana in a patient infected with human immunodeficiency virus. J. Clin. Microbiol. 31:692-694.a, 百拇医药
Spach, D. H., A. S. Kanter, M. J. Dougherty, A. M. Larson, M. B. Coyle, D. J. Brenner, B. Swaminathan, G. M. Matar, D. F. Welch, R. K. Root, and W. E. Stamm. 1995. Bartonella (Rochalimaea) quintana bacteremia in inner-city patients with chronic alcoholism. N. Engl. J. Med. 332:424-428.
Struyve, M., M. Moons, and J. Tommassen. 1991. Carboxy-terminal phenylalanine is essential for the correct assembly of a bacterial outer membrane protein. J. Mol. Biol. 218:141-148.601], http://www.100md.com
Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673-4680.601], http://www.100md.com
Towbin, H., T. Staehelin, and J. Gordon. 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 76:4350-4354.601], http://www.100md.com
Tyeryar, F. J., E. Weiss, D. B. Millar, F. M. Bozeman, and R. A. Ormsbee. 1973. DNA base composition of rickettsiae. Science 180:415-417.601], http://www.100md.com
Vizcaino, N., A. Cloeckaert, M. S. Zygmunt, and G. Dubray. 1996. Cloning, nucleotide sequence, and expression of the Brucella melitensis omp31 gene coding for an immunogenic major outer membrane protein. Infect. Immun. 64:3744-3751.601], http://www.100md.com
Vizcaino, N., J. M. Verger, M. Grayon, M. S. Zygmunt, and A. Cloeckaert. 1997. DNA polymorphism at the omp-31 locus of Brucella spp.: evidence for a large deletion in Brucella aborutus, and other species-specific markers. Microbiology 143:2913-2921.601], http://www.100md.com
Weiss, M. S., U. Abele, J. Weckesser, W. Welte, E. Schiltz, and G. E. Schulz. 1991. Molecular architecture and electrostatic properties of a bacterial porin. Science 254:1627-1630.601], http://www.100md.com
Weiss, E., and G. A. Dasch. 1982. Differential characteristics of strains of Rochalimaea: Rochalimaea vinsonii sp. nov., the Canadian vole agent. Int. J. Syst. Bacteriol. 32:305-314.(Michael F. Minnick Kate N. Sappington Laura S. Smitherman Siv G. E. Andersson Olof Karlberg and Jame)