PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of iaiPermissionsJournals.ASM.orgJournalIAI ArticleJournal InfoAuthorsReviewers
 
Infect Immun. 2004 September; 72(9): 5365–5372.
PMCID: PMC517454

Analysis of Subassemblies of Pertussis Toxin Subunits In Vivo and Their Interaction with the Ptl Transport Apparatus

Abstract

Pertussis toxin (PT) has an AB5 structure that is typical of many bacterial protein toxins; however, this toxin is more complex than many toxins since it is composed of five different subunit types, subunits S1 to S5. Little is known about how PT assembles in vivo and how and when it interacts with its secretion apparatus, known as the Ptl transporter. In order to better understand these events, we expressed subsets of the genes encoding the S1, S2, and/or S4 subunits of PT in strains of Bordetella pertussis that either did or did not produce the Ptl proteins. We found evidence to suggest that certain subassemblies of the toxin, including subassemblies consisting of the S1 subunit and incomplete forms of the B oligomer, can form in vivo, at least transiently. These results suggest that the B oligomer of the toxin does not need to completely form before interactions between the S1 subunit and B-oligomer subunits can occur in vivo. All subassemblies localized primarily to the membrane fraction of the cell. Moreover, we found that Ptl-mediated secretion occurs in a strain that produces S1 and an incomplete complement of B-oligomer subunits. These results indicate that subassemblies of the toxin consisting of the S1 subunit and a partial B oligomer can interact with the Ptl system.

Bacterial toxins are often large multisubunit proteins that, after synthesis in the bacterium, must cross the bacterial membrane barriers to gain access to their eukaryotic cell targets. Biogenesis of the toxin molecule is complex in that proper folding, assembly, and secretion of the proteins must occur in the appropriate sequence. The molecular events that take place during toxin biogenesis remain elusive for many large bacterial protein toxins.

One such bacterial protein toxin is pertussis toxin (PT). PT plays an important role in the pathogenesis of Bordetella pertussis (26), the causative agent of whooping cough (pertussis). PT is typical of many bacterial protein toxins in that it has an A-B structure that comprises an enzymatically active S1 subunit and five subunits that make up the binding component or B oligomer (25). The B oligomer, composed of one copy each of subunits S2, S3, and S5 and two copies of S4, is more complex than the binding components of many other AB5 toxins in that not all of the subunits are identical.

After synthesis, the individual toxin subunits are thought to be secreted via a Sec-like system across the inner membrane since each subunit is synthesized with its own signal sequence (18, 19). The toxin is secreted across the outer membrane of B. pertussis by a type IV secretion system, known as the Ptl system (27), which has striking homology to other type IV transporters (3, 28). The Ptl apparatus is composed of nine proteins, PtlA to PtlI (6, 27). As shown in Fig. Fig.1A,1A, the genes encoding the Ptl proteins are located within the same operon and directly downstream from the ptx genes that encode the toxin subunits (15, 27).

FIG. 1.
ptx-ptl region and vector inserts used in this study. (A) Genetic organization and nucleotide numbering system (19) of the ptx-ptl region. Pr indicates the location of the ptx-ptl promoter. (B) Vector inserts used in this study that are capable of expressing ...

It has been demonstrated previously that neither the individual S1 subunit nor the individual B oligomer is secreted from B. pertussis via Ptl-mediated transport (8). While these results suggest that individual subunits of the toxin are not secreted by the Ptl system, we know little about the actual sequence of events in the assembly and secretion of PT. Since the organism can be genetically manipulated such that only a subset of the genes encoding the toxin subunits are expressed, production of toxin subunits can be limited to individual subunits or specific combinations of subunits. Partial assemblies of the toxin that form can then be characterized. In this study, we expressed combinations of the genes encoding subunits S1, S2, and S4 in strains of B. pertussis that either do or do not produce the Ptl proteins. We examined the stability of subunits within these subassemblies, the localization of subunits within the bacterial cell, and their secretion in the hope of better understanding the sequence of events that occur during the biogenesis of PT.

MATERIALS AND METHODS

Bacterial strains, growth conditions, and plasmids.

The strains of B. pertussis and Escherichia coli and the plasmids used in this study are listed in Table Table1.1. B. pertussis strains were grown at 37°C on Bordet-Gengou (BG) agar.

TABLE 1.
Strains and plasmids used in this study

B. pertussis mutants.

Construction of BP536Δptx and BP536Δptxptl has been described previously (8). BP536Δptx has an in-frame deletion in the PT structural genes from nucleotide 936 through nucleotide 3514, corresponding the second half of ptxS1, all of ptxS2, ptxS4, and ptxS5, and 84% of ptxS3 (Fig. (Fig.1A1A shows a map of the ptx-ptl region). This strain does not produce PT but does produce the Ptl proteins. It should be noted that the in-frame deletion would allow expression of a fusion protein consisting of the signal sequence of the S1 subunit, the first 109 amino acids of the coding sequence of the S1 subunit, and the last 37 amino acids of the S3 subunit. BP536Δptxptl has a deletion extending from nucleotide 425 through nucleotide 11971 which spans virtually the entire ptx-ptl region. This strain does not produce PT or the Ptl proteins.

BP536Δptl was constructed as follows. A 7.4-kb segment of the ptx-ptl region, extending from nucleotide 3626 through nucleotide 11039 and corresponding to ptlA, ptlB, ptlC, ptlD, ptlI, ptlE, ptlF, ptlG, and the 5′ end of ptlH, was deleted from the B. pertussis chromosome by homologous recombination as follows. Plasmid pSZH4, which has been described previously (11), was digested with BlpI. The 8.3-kb fragment, which contained the ptx-ptl promoter, the ptx genes, the truncated ptlH gene, and the pUC19 vector backbone, was isolated and self-ligated. The resulting plasmid (pAMC109) was digested with EcoRI and HindIII, and the 5.6-kb fragment was inserted into pSS1129. This plasmid was used for allelic exchange in strain BP536 as previously described (23). The 7.4-kb deletion in the ptx-ptl region was verified by PCR.

Plasmid construction. (i) Construction of pDMC36.

Plasmid pDMC36, which carries ptxS1 controlled by the ptx-ptl promoter in pUFR047, has been described previously (8).

(ii) Construction of pSPF21.

Plasmid pSPF21, which carries ptxS1, ptxS2, and ptxS4 controlled by the ptx-ptl promoter region (nucleotides 1 to 2504 of the ptx-ptl region), was constructed as follows. The SstI-NsiI fragment of pSZH4 (11) was inserted into pGEM-11Zf(+). The KpnI-XbaI fragment of the resultant vector (pSPF13) was excised and replaced with the KpnI-XbaI fragment of pDMC36. The 2.5-kb SstI-HindIII fragment of the resultant vector (pSPF20) was inserted into the broad-host-range vector pUFR047 to obtain pSPF21.

(iii) Construction of pSPF25.

Plasmid pSPF25, which carries ptxS1 and ptxS2 controlled by the ptx-ptl promoter (nucleotides 1 to 2064 of the ptx-ptl region), was constructed as follows. The 2.1-kb SstI-SmaI fragment of pSPF20 (described above) was inserted into pUC19. The SstI-HindIII fragment of the resulting vector (pSPF24) was inserted into pUFR047 to obtain pSPF25.

(iv) Construction of pSPF27.

Plasmid pSPF27, which carries the ptx-ptl promoter, ptxS1, and ptxS4 (nucleotides 1 to 2504 with an in-frame deletion from nucleotide 1613 to nucleotide 1924, encompassing more than 50% of the coding region for the mature form of the S2 subunit), was constructed as follows. First, by using PCR, a DNA fragment was generated which consisted of nucleotides 1301 to 1930 of the ptx-ptl region containing a deletion from nucleotide 1613 to nucleotide 1924. The upstream primer was 5′-CGCGTATTCGTTCTAGACCTGGCCC (nucleotides 1301 to 1325 of the ptx-ptl region, which contained the underlined XbaI site). The downstream primer was 5′-CTCCTCACGCGTTCGTAGAGCGCAAATATTGACCAGCC (six nucleotides followed by the underlined MluI site [nucleotides 1930 to 1925 of the ptx-ptl region] followed by nucleotides 1612 to 1586 of the ptx-ptl region). The resulting fragment was inserted into the 5.1-kb XbaI-MluI fragment of pSPF20 (constructed as described above). The resulting plasmid (pSPF26) was digested with SstI and HindIII, and the 2.2-kb fragment was inserted into pUFR047 to obtain pSPF27.

(v) Construction of pSPF19.

Plasmid pSPF19, which carries the ptx-ptl promoter, ptxS2, and ptxS4 (nucleotides 1 to 2504 of the ptx-ptl region with an in-frame deletion from nucleotide 784 to nucleotide 1311, encompassing more than 75% of the coding region for the mature form of the S1 subunit), was constructed as follows. First, by using PCR, a DNA fragment was generated which consisted of nucleotides 411 to 1317 of the ptx-ptl region with a deletion from nucleotide 784 to nucleotide 1311. The upstream primer was 5′-CGCGATGGTACCGGTCACCGTCCG (nucleotides 411 to 434 of the ptx-ptl region, which contained the underlined KpnI site). The downstream primer was 5′-CTCTCTAGAAGCGCCGGCTGCTGCTGGTGGAGA (three nucleotides followed by the underlined XbaI site [nucleotides 1317 to 1312 of the ptx-ptl region] followed by nucleotides 783 to 760 of the ptx-ptl region). The resulting KpnI-XbaI fragment was inserted into the 4.8-kb fragment of pSPF13 (constructed as described above) digested with the same enzymes. The 2.0-kb SstI-HindIII fragment of the resulting plasmid (pSPF17) was inserted into pUFR047 to obtain pSPF19.

(vi) Construction of pSPF23.

Plasmid pSPF23, which carries the ptx-ptl promoter and ptxS2 (nucleotides 1 to 2064 with an in-frame deletion from nucleotide 784 to nucleotide 1311, encompassing more than 75% of the coding region for the mature form of the S1 subunit), was constructed as follows. The SstI-SmaI fragment of pSPF17 (constructed as described above) was inserted into pUC19. The SstI-HindIII fragment of the resulting plasmid (pSPF22) was inserted into pUFR047 to obtain pSPF23.

(vii) Construction of pSPF18.

Plasmid pSPF18, which carries the ptx-ptl promoter and ptxS4 (nucleotides 1 to 2504 with an in-frame deletion from nucleotide 794 to nucleotide 1924 such that the first 35% of the coding region of ptxS1 is fused to approximately 15% of the coding region of ptxS2), was constructed as follows. First, by using PCR, a DNA fragment was generated which consisted of nucleotides 411 to 1930 of the ptx-ptl region with a deletion from nucleotide 794 to nucleotide 1924. The upstream primer was 5′-CGCGATGGTACCGGTCACCGTCCG (nucleotides 411 to 434 of the ptx-ptl region, which contained the underlined KpnI site). The downstream primer was 5′-CTCCTCACGCGTACCTCGGTATAGCGCCGGCTGCTG (six nucleotides followed by the underlined MluI site [nucleotides 1930 to 1925 of the ptx-ptl region] followed by nucleotides 793 to 770 of the ptx-ptl region). The resulting KpnI-MluI fragment was inserted into a 4.2-kb KpnI-MluI fragment isolated from pSPF13 (constructed as described above). The 1.3-kb SstI-HindIII fragment of the resulting plasmid (pSPF16) was inserted into pUFR047 to obtain pSPF18.

Introduction of plasmids into B. pertussis.

Introduction of plasmids with a pUFR047 backbone into strains of B. pertussis has been described previously (8).

Preparation of cell extracts from plate cultures.

Cell extracts of B. pertussis were prepared by suspending cells grown on BG agar for approximately 72 h in 3 to 5 ml of phosphate-buffered saline (pH 7.4) to an A550 of 2. Samples of cells (50 μl) were precipitated with an equal volume of 20% trichloroacetic acid. After centrifugation, the precipitates were suspended in 20 μl of sodium dodecyl sulfate (SDS) sample buffer containing dithiothreitol.

Localization of toxin subunits.

Cell extracts of B. pertussis were prepared by suspending cells grown on BG agar in 3 to 5 ml of phosphate-buffered saline to an A550 of 2. The cells were then disrupted by sonication, and soluble (cytoplasmic and periplasmic) and insoluble (membrane) fractions were then prepared as previously described (7). Portions of the soluble fraction (150 μl) and the membrane fraction (150 μl) were precipitated with an equal volume of 20% trichloroacetic acid. The precipitates were collected after centrifugation and suspended in 20 μl of the sample buffer used for SDS-polyacrylamide gel electrophoresis.

Analysis of secretion of toxin subunits.

In order to examine secretion of PT and its subassemblies, we grew B. pertussis using a modification of the method described previously (4). Briefly, BG agar plates were overlaid with 6 ml of Stainer-Scholte medium (SS medium) and incubated at 37°C for 2 h. Conditioned SS medium was collected from each plate and inoculated with B. pertussis strains at an initial A550 of 0.2 to 0.25. Cultures were grown in conditioned SS medium at 37°C with shaking (170 to 180 rpm) for 24 h to an A550 of 2 to 2.2. Cultures were harvested and centrifuged. The cells were suspended in a volume of phosphate-buffered saline equal to the volume of the culture supernatant. Samples of supernatant (100 μl) and cells (50 μl) were precipitated with ethanol and 20% trichloroacetic acid, respectively. Pellets were suspended in SDS sample buffer and loaded onto 4 to 20% polyacrylamide gradient gels for electrophoresis and immunoblot analysis.

The percentage of secretion of PT subunits was quantified by densitometry and was calculated as follows: amount of subunit found in the supernatant/total amount of subunit produced (cell associated + supernatant) × 100.

Immunoblot analysis.

Samples were subjected to SDS-polyacrylamide gel electrophoresis performed essentially as described by Laemmli (16) by using 4 to 20% polyacrylamide gradient gels (for visualizing subunits S1 and S2) and 16.5% Tris-Tricine gels (for visualizing subunit S4) obtained from Bio-Rad Laboratories, Hercules, Calif. Immunoblot analysis was performed essentially as previously described (1); monoclonal antibody 3CX4 (13, 14) was used to visualize the S1 subunit, monoclonal antibody P11B10 (9) was used to visualize the S2 subunit, and monoclonal antibody 6DX3 (13) was used to visualize the S4 subunit. Where indicated below, immunoblots were scanned with a Bio-Rad GS-800 densitometer and were analyzed by using Quality One software.

Statistical analysis.

Statistical significance was determined by using Student's t test. Differences were considered significant when the P value was less than 0.05.

RESULTS

In an attempt to identify interactions between individual PT subunits that occur after synthesis within the cell but before secretion, we constructed strains of B. pertussis in which ptxS1, ptxS2, and ptxS4 were expressed alone or in combination with each other either in the presence or in the absence of the ptl genes. We then examined the PT subunits within the resulting strains. In order to do this, we first constructed six plasmids (Table (Table11 and Fig. Fig.1B):1B): pSPF23 (containing the ptx-ptl promoter and ptxS2), pSPF18 (containing the ptx-ptl promoter and ptxS4), pSPF25 (containing the ptx-ptl promoter, ptxS1, and ptxS2), pSPF27 (containing the ptx-ptl promoter, ptxS1, and ptxS4), pSPF19 (containing the ptx-ptl promoter, ptxS2, and ptxS4), and pSPF21 (containing the ptx-ptl promoter, ptxS1, ptxS2, and ptxS4). We then introduced these plasmids into BP536Δptx, a strain that has an in-frame deletion in the ptx region such that it does not produce PT but does produce a functional Ptl transporter (8), or into BP536Δptxptl, a strain in which the entire ptx-ptl region is deleted such that it produces neither PT nor the Ptl proteins. We reasoned that if subassemblies formed or if the Ptl proteins interacted with individual subunits or subassemblies, the stabilities of the individual subunits might be altered. When we examined the protein profiles of the resulting strains, we were able to detect certain differences in the stabilities of the subunits which were dependent on coexpression of other ptx genes.

We first examined the stability of the S2 subunit in whole-cell extracts of the strains producing only S2 or producing S2 in combination with S1 and/or S4. As shown in Fig. Fig.2,2, no S2 subunit was observed in whole-cell extracts of the strains producing only the S2 subunit [BP536Δptx(pSPF23) and BP536Δptxptl(pSPF23)], regardless of whether the Ptl proteins were present (Fig. (Fig.2,2, lanes 1 and 2). When whole-cell extracts of strains producing S1 and S2 [BP536Δptx(pSPF25) and BP536Δptxptl(pSPF25)] were examined, a small amount of S2 subunit was apparent (lanes 3 and 4), consistent with the idea that the S2 subunit exhibits slightly greater stability in the presence of the S1 subunit. When we examined whole-cell extracts of strains producing both the S2 and S4 subunits [BP536Δptx(pSPF19) and BP536Δptxptl(pSPF19)], it was apparent that the stability of the S2 subunit was greatly enhanced in the presence of the S4 subunit (lanes 5 and 6). No significant difference in the stability of the S2 subunit was observed when we compared strains producing the S2 and S4 subunits and strains producing the S1, S2, and S4 subunits [BP536Δptx(pSPF21) and BP536Δptxptl(pSPF21)] (Fig. (Fig.2,2, compare lanes 5 and 6 to lanes 7 and 8). The presence of the Ptl proteins did not significantly affect the stability of S2 in these strains.

FIG. 2.
Immunoblot analysis of cell extracts of B. pertussis strains producing subsets of PT subunits that include the S2 subunit. Samples of cell extracts (50 μl) from different strains were prepared as described in Materials and Methods and were subjected ...

We next examined whole-cell extracts of strains that produce the S4 subunit either alone or in combination with the S1 subunit and/or the S2 subunit (Fig. (Fig.3).3). When the S4 subunit was produced in the absence of other PT subunits in strains BP536Δptx(pSPF18) and BP536Δptxptl(pSPF18), only a minute quantity of the mature form of the S4 subunit was observed (Fig. (Fig.3,3, lanes 1 and 2). We did observe a protein species that reacted with our anti-S4 subunit monoclonal antibody and which migrated more slowly than the mature species on SDS-polyacrylamide gels. This species may have represented the S4 subunit which did not have its signal sequence cleaved. The S4 protein profile of strains producing only the S4 subunit was not significantly affected by the presence of the Ptl proteins. When we examined strains producing both the S1 and S4 subunits [BP536Δptx(pSPF27) and BP536Δptxptl(pSPF27)], we again observed very little of the mature form of the S4 subunit (lanes 3 and 4). In contrast, when the S2 subunit was coproduced with the S4 subunit, the stability of the S4 subunit was strikingly enhanced (lanes 5 and 6). A small amount of the higher-molecular-weight species was again observed, which might have represented the S4 subunit which retained its signal sequence. Coproduction of the S1 subunit with the S2 and S4 subunits did not significantly increase the stability of the S4 subunit (lanes 7 and 8). Again, we saw no effect of the Ptl proteins.

FIG. 3.
Immunoblot analysis of cell extracts of B. pertussis strains producing subsets of PT subunits that include the S4 subunit. Samples of cell extracts (50 μl) from different strains were prepared as described in Materials and Methods and were subjected ...

We next examined the stability of the S1 subunit of strains producing the S1 subunit with the S2 and/or S4 subunit (Fig. (Fig.4).4). As previously noted, the protein profile of the S1 subunit produced by BP536Δptxptl(pDMC36), which produces the S1 subunit but none of the B-oligomer subunits, consists of two major species (Fig. (Fig.4,4, lane 1). The more slowly migrating of these species represents the mature full-length form of the S1 subunit, whereas the species that migrates faster represents a well-described degradation product (2, 8). In certain instances, a third species that migrates slightly more slowly than the full-length form is apparent. This species likely represents the S1 subunit with its signal sequence intact (8). The S1 protein profiles of strains producing only the S1 subunit or the S1 subunit in combination with S2 and/or S4 are shown in Fig. Fig.4.4. The S1 protein profiles of strains producing only the S1 subunit, the S1 and S4 subunits, and the S1 and S2 subunits did not differ greatly from each other. However, in triplicate experiments, densitometric analysis of the immunoblots indicated that strains producing all three subunits reproducibly produced amounts of the mature form of the S1 subunit that were significantly larger than the amounts produced by strains producing only the S1 and S2 subunits (P < 0.05) (e.g., compare lanes 6 and 7 to lanes 4 and 5 in Fig. Fig.4),4), suggesting possible formation of an S1-S2-S4 subassembly in vivo.

FIG. 4.
Immunoblot analysis of cell extracts of B. pertussis strains producing subsets of PT subunits that include S1. Samples of cell extracts (50 μl) from different strains were prepared as described in Materials and Methods and were subjected to SDS-polyacrylamide ...

We next asked where the subassemblies might be localized within the bacterial cell. It has been demonstrated previously that the S1 subunit, either alone or in combination with the B oligomer, localizes primarily to the membrane fraction of the cell (7). In order to localize the S2 subunit in strains producing various subsets of toxin subunits, we separated the bacterial cells into a soluble fraction (consisting of the cytoplasm and periplasm) and an insoluble (membrane) fraction. As shown in Fig. Fig.5,5, the S2 subunit of strains BP536Δptx(pSPF25) and BP536Δptxptl (pSPF25), which produced the S1 and S2 subunits, partitioned primarily to the membrane fraction. The S2 subunit of BP536Δptx(pSPF19) and BP536Δptxptl(pSPF19), which produced the S2 and S4 subunits, also partitioned primarily to the membrane fraction, although some S2 was also found in the soluble fraction. The majority of the S2 subunit of BP536Δptx(pSPF21) and BP536Δptxptl(pSPF21), which produced S1, S2, and S4, partitioned to the membrane fraction, although a significant amount was observed in the soluble fraction. Since the S2 subunit that was observed in the soluble fraction had its signal sequence cleaved, it must have been located in the periplasm (21).

FIG. 5.
Immunoblot analysis of the S2 subunit in cell fractions of strains producing subsets of PT subunits. Different strains were subjected to cell fractionation as described in Materials and Methods. Samples (150 μl) were subjected to SDS-polyacrylamide ...

Finally, we asked whether subassemblies of the toxin can interact with the Ptl proteins. We did this by examining secretion of subassemblies from B. pertussis. As shown in Fig. Fig.6A,6A, PT (as visualized by the S2 subunit) was secreted from the wild-type strain but not from the strain containing a deletion of the ptl genes (Fig. (Fig.6A,6A, lanes 1 to 4). Note that in all experiments whose results are shown Fig. Fig.6,6, we loaded two times more culture supernatant than cellular material on the SDS-polyacrylamide gel used for the immunoblot in order to optimize our ability to detect subunits in the culture supernatant. As previously reported (8), secretion from the wild-type strain was not efficient since only an average of 28% of the toxin was found to be secreted in a set of four replicate experiments performed in this study. Strains producing only S2 and S4 did not secrete any S2 subunit regardless of whether the Ptl proteins were present (Fig. (Fig.6A,6A, lanes 5 to 8). When we examined strains producing S1, S2, and S4 (Fig. (Fig.6B),6B), we found that Ptl-mediated secretion did occur since, reproducibly, a higher percentage of the total amount of S2 subunit produced was released into the supernatant by the strain producing the Ptl proteins than by the strain that did not produce the Ptl proteins (lanes 3 to 6). In a set of four replicate experiments performed in this study, we obtained results similar to those shown in Fig. Fig.6.6. When the immunoblots were scanned by using densitometry, an average of 23% of the S2 subunit was released from BP536Δptx(pSPF21), the strain producing S1, S2, S4, and the Ptl proteins, whereas an average of 11% of the S2 subunit was released from BP536Δptxptl(pSPF21), the strain producing S1, S2, and S4 but no Ptl proteins (Fig. (Fig.6C).6C). The difference was determined to be statistically significant (P < 0.05). Notably, while we did not observe significant levels of S2 released from BP536Δptl and BP536Δptxptl(pSPF19), strains that produced either the holotoxin or the S2-S4 subassembly but no Ptl proteins, we always observed a small amount of S2 released from the strain that produced S1, S2, and S4 but did not produce Ptl proteins. Such Ptl-independent release has been observed for strains producing the S1 subunit but no B-oligomer subunits (8). It is not clear to what this Ptl-independent release is due.

FIG. 6.
Immunoblot analysis of the S2 subunit in cell extracts and culture supernatants of B. pertussis strains. Samples of culture supernatants (100 μl) and cell extracts (50 μl) were prepared as described in Materials and Methods. Samples were ...

DISCUSSION

Previously, the stability and localization of the S1 subunit of PT were examined when it was produced either in the presence or in the absence of all of the subunits of the B oligomer. It was found that the presence of the B oligomer had a significant effect on the stability of the S1 subunit within the cell, suggesting that assembly of the toxin occurs within the bacterium (7). Moreover, evidence that the S1 subunit localized to the outer membrane of the cell was obtained. No evidence of an interaction between the S1 subunit and the Ptl transporter was found (8).

In this study, we continued the examination of the assembly of the toxin within the bacterial cell and the interaction of the toxin with the Ptl proteins by expressing subsets of the ptx genes either in the presence or in the absence of expression of the ptl genes. Since the crystal structure of PT has been determined (22), the interactions between subunits within the holotoxin molecule have been defined. Figure Figure77 shows the crystal structure of the holotoxin and also the interactions that occur among subunits S1, S2, and S4 in the holotoxin molecule.

FIG. 7.
PT and subassemblies. The crystal structure of PT is shown along with the structures of subassemblies containing different subunits. The structures were derived from the structures of the individual subunits in the crystal structure of the holotoxin molecule. ...

We examined four potential subassemblies of PT: S1-S2, S1-S4, S2-S4, and S1-S2-S4. We found evidence that is consistent with the existence of the S1-S2, S2-S4, and S1-S2-S4 subassemblies in vivo since we observed alterations in the stabilities of individual subunits when they were coproduced with the subassembly partner(s). Although they are indirect measures of assembly, changes in the stability of individual subunits are suggestive of complex formation. To our knowledge, this is the first evidence supporting the existence of subassemblies of the toxin in vivo. We did not find evidence for the existence of the S1-S4 subassembly; however, the lack of evidence does not preclude the existence of this structure.

The S2-S4 subassembly appears to be the most stable of the subassemblies. Evidence for stable association of the S2 and S4 subunits in vitro has been known for years. Tamara and coworkers showed that S2-S4 dimers formed when the individual subunits were mixed together in vitro (25). The evidence presented here suggests that such an interaction can also occur in vivo. Examination of the crystal structure of the toxin (Fig. (Fig.7)7) demonstrates that interactions between these two subunits in the holotoxin occur mainly through antiparallel β-sheet interactions (22). When we examined the localization of the S2-S4 subassembly within the cell, we found that it partitioned primarily to the membrane fraction of the cell, suggesting that the subunits of the B oligomer, like the S1 subunit, can localize to the membrane.

Our data are consistent with the existence of a subassembly consisting of the S1 and S2 subunits. When we examined the crystal structure of S1 and S2 in the holotoxin (Fig. (Fig.7),7), the interactions between the two subunits appeared not to be extensive. Nonetheless, our results suggest that the complex can exist, at least transiently. The finding that the S2 subunit can interact with the S1 subunit in the absence of other toxin subunits suggests the possibility that the subunits of the B oligomer of the toxin may assemble with S1 as they are produced, or perhaps the S2 subunit interacts with the S1 subunit and then acts as a nucleus for assembly of the remaining B-oligomer subunits. In any case, there does not appear to be an absolute requirement for complete assembly of the B oligomer before interaction with the S1 subunit.

We found evidence which is consistent with the existence of the S1-S2-S4 subassembly in that the stability of the S1 subunit was increased when both the S2 and S4 subunits were coproduced along with the S1 subunit. However, care must be taken in interpreting the data in this manner since both the S2 and S4 subunits are significantly more stable and therefore found in larger quantities when they are coproduced. The simple fact that higher levels of each of the subunits are present in such strains may account for the increase in the stability of the S1 subunit that we observed.

When we examined the cellular localization of the subassemblies, we found that the S2 subunits of the S1-S2, S2-S4, and S1-S2-S4 subsets each partitioned to both the soluble (cytoplasmic-periplamic) and insoluble (membrane) fractions. Other workers have previously reported that at least a portion of the S2 subunit partitions to the membrane fraction in cells producing the holotoxin (20). In these studies, in which only subsets of toxin subunits were produced, the S2 subunit was found to partition in large part to the membrane fraction. We found that the S2 subunit of the S1-S2 subassembly partitioned mainly to the membrane fraction, regardless of whether the Ptl proteins were present, which is consistent with the previous finding that the S1 subunit, when it is produced in the absence of the B-oligomer subunits, partitions primarily to the outer membrane fraction of the cell (7). Therefore, we suggest that the interaction between S1 and S2 subunits may take place at or in the outer membrane.

We found that the S2 subunit is secreted in a Ptl-mediated manner in BP536Δptx(pSPF21), a strain that produces the S1,S2, and S4 subunits, as well as the Ptl proteins. As described in Materials and Methods, deletion of the ptx genes in this strain was accomplished by creating an in-frame deletion from the 5′ end of ptxS1 to the 3′ end of ptxS3. Thus, this strain could produce a fusion protein containing the last 37 amino acids of the S3 subunit (the mature form of the S3 subunit comprises a total of 199 amino acids). For this reason, we cannot exclude the possibility that this portion of the S3 subunit might play a role in the secretion that was observed in this strain. Nonetheless, our results indicate that Ptl-mediated secretion can occur in the absence of a complete form of the B oligomer since this strain cannot produce the majority of the S3 subunit or any of the S5 subunit.

We did not observe any secretion of the S2 subunit with BP536Δptxptl(pSPF19), a strain producing only the S2 and S4 subunits and no Ptl proteins, which is consistent with the previous observation that strains producing the subunits of the B oligomer, but no S1 subunit, do not exhibit Ptl-mediated secretion (8). It has also been shown previously that strains of B. pertussis producing the S1 subunit and the Ptl proteins, but no B-oligomer subunits, do not exhibit Ptl-mediated secretion (8). These results, together with the observation presented here that a strain producing the S1, S2, and S4 subunits, as well as the Ptl proteins, exhibits Ptl-mediated secretion, indicate that Ptl-mediated secretion occurs only when the S1 subunit and at least a subset of the B-oligomer subunits are produced. These results suggest that the structure of PT that is recognized by the Ptl proteins is present only when assembly occurs. This structure may consist of regions from both the S1 subunit and subunits of the B oligomer; alternatively, a conformational change in one of the subunits may take place upon assembly, resulting in a structure that can be recognized by the Ptl proteins. The requirement for at least some assembly of the toxin before secretion takes place may ensure that energy is not wasted by the bacteria by producing subunits that are secreted before they can assemble into the toxic form of the molecule. If assembly is normally rapid compared to secretion, formation of the holotoxin would be assured.

Our findings are consistent with and extend the model that was proposed previously (7), in which PT assembles and then interacts with the Ptl proteins at the outer membrane. The Ptl proteins may interact directly with PT and then act in a piston-like manner to push the toxin out of the cell. Such a structure would resemble the pilus structure of the VirB transporter of Agrobacterium tumefaciens (10, 17), the prototypical type IV transporter. Alternatively, the Ptl transporter might form a channel around the toxin, allowing it to escape into the extracellular milieu.

Notes

Editor: J. T. Barbieri

REFERENCES

1. Burnette, W. N. 1981. “Western blotting”: electrophoretic transfer of proteins from sodium dodecyl sulfate-polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated protein A. Anal. Biochem. 112:195-203. [PubMed]
2. Burns, D. L., S. Z. Hausman, W. Lindner, F. A. Robey, and C. R. Manclark. 1987. Structural characterization of pertussis toxin A subunit. J. Biol. Chem. 262:17677-17682. [PubMed]
3. Casales, E., and P. J. Christie. 2003. The versatile bacterial type IV secretion systems. Nat. Rev. Microbiol. 1:137-149. [PubMed]
4. Craig-Mylius, K. A., T. H. Stenson, and A. A. Weiss. 2000. Mutations in the S1 subunit of pertussis toxin that affect secretion. Infect. Immun. 68:1276-1281. [PMC free article] [PubMed]
5. DeFeyter, R., Y. Yang, and D. W. Gabriel. 1993. Gene-for-gene interactions between cotton R genes and Xanthomonas campestris pv. malvacearum avr genes. Mol. Plant-Microbe Interact. 6:225-237. [PubMed]
6. Farizo, K. M., T. G. Cafarella, and D. L. Burns. 1996. Evidence for a ninth gene, ptlI, in the locus encoding the pertussis toxin secretion system of Bordetella pertussis and formation of a PtlI-PtlF complex. J. Biol. Chem. 271:31643-31649. [PubMed]
7. Farizo, K. M., S. Fiddner, A. M. Cheung, and D. L. Burns. 2002. Membrane localization of the S1 subunit of pertussis toxin in Bordetella pertussis and implications for pertussis toxin secretion. Infect. Immun. 70:1193-1201. [PMC free article] [PubMed]
8. Farizo, K. M., T. Huang, and D. L. Burns. 2000. Importance of holotoxin assembly in Ptl-mediated secretion of pertussis toxin from Bordetella pertussis. Infect. Immun. 68:4049-4054. [PMC free article] [PubMed]
9. Frank, D. W., and C. D. Parker. 1984. Interaction of monoclonal antibodies with pertussis toxin and its subunits. Infect. Immun. 46:195-201. [PMC free article] [PubMed]
10. Fullner, K. J., J. C. Lara, and E. W. Nester. 1996. Pilus assembly by Agrobacterium T-DNA transfer genes. Science 273:1107-1109. [PubMed]
11. Hausman, S. Z., J. D. Cherry, U. Heininger, C. H. Wirsing von Konig, and D. L. Burns. 1996. Analysis of proteins encoded by the ptx and ptl genes of Bordetella bronchiseptica and Bordetella parapertussis. Infect. Immun. 64:4020-4026. [PMC free article] [PubMed]
12. Keen, N. T., S. Tamaki, D. Kobayashi, and D. Trollinger. 1988. Improved broad-host-range plasmids for DNA cloning in gram-negative bacteria. Gene 70:191-197. [PubMed]
13. Kenimer, J. G., K. J. Kim, P. G. Probst, C. R. Manclark, D. G. Burstyn, and J. L. Cowell. 1989. Monoclonal antibodies to pertussis toxin: utilization as probes of toxin function. Hybridoma 8:37-51. [PubMed]
14. Kim, K. J., W. N. Burnette, R. D. Sublett, C. R. Manclark, and J. G. Kenimer. 1989. Epitopes on the S1 subunit of pertussis toxin recognized by monoclonal antibodies. Infect. Immun. 57:944-950. [PMC free article] [PubMed]
15. Kotob, S. I., S. Z. Hausman, and D. L. Burns. 1995. Localization of the promoter for the ptl genes of Bordetella pertussis, which encode proteins essential for secretion of pertussis toxin. Infect. Immun. 63:3227-3230. [PMC free article] [PubMed]
16. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685. [PubMed]
17. Lai, E.-M., and C. I. Kado. 2000. The T-pilus of Agrobacterium tumefaciens. Trends Microbiol. 8:361-369. [PubMed]
18. Locht, C., and J. M. Keith. 1986. Pertussis toxin gene: nucleotide sequence and genetic organization. Science 232:1258-1264. [PubMed]
19. Nicosia, A., M. Perugini, C. Franzini, M. C. Casagli, M. G. Borri, G. Antoni, M. Almoni, P. Neri, G. Ratti, and R. Rappuoli. 1986. Cloning and sequencing of the pertussis toxin genes: operon structure and gene duplication. Proc. Natl. Acad. Sci. USA 83:4631-4635. [PubMed]
20. Rambow-Larsen, A. A., and A. A. Weiss. 2004. Temporal expression of pertussis toxin and Ptl secretion proteins by Bordetella pertussis. J. Bacteriol. 186:43-50. [PMC free article] [PubMed]
21. Randall, L. L., S. J. S. Hardy, and J. R. Thom. 1987. Export of protein: a biochemical view. Annu. Rev. Biochem. 41:507-541. [PubMed]
22. Stein, P. E., A. Boodhoo, G. D. Armstrong, S. A. Cockle, M. H. Klein, and R. J. Read. 1994. The crystal structure of pertussis toxin. Structure 2:45-57. [PubMed]
23. Stibitz, S. 1994. Use of conditionally counterselectable suicide vectors for allelic exchange. Methods Enzymol. 235:458-465. [PubMed]
24. Stibitz, S., and M.-S. Yang. 1991. Subcellular localization and immunochemical detection of proteins encoded by the vir locus of Bordetella pertussis. J. Bacteriol. 173:4288-4296. [PMC free article] [PubMed]
25. Tamura, M., K. Nogimori, S. Murai, M. Yajima, K. Ito, T. Katada, M. Ui, and S. Ishii. 1982. Subunit structure of islet-activating protein, pertussis toxin, in conformity with the A-B model. Biochemistry 21:5516-5522. [PubMed]
26. Weiss, A. A., and S. Falkow. 1983. Transposon insertion and subsequent donor formation promoted by Tn501 in Bordetella pertussis. J. Bacteriol. 153:304-309. [PMC free article] [PubMed]
27. Weiss, A. A., F. D. Johnson, and D. L. Burns. 1993. Molecular characterization of an operon required for pertussis toxin secretion. Proc. Natl. Acad. Sci. USA 90:2970-2974. [PubMed]
28. Winans, S. C., D. L. Burns, and P. J. Christie. 1996. Adaptation of a conjugal transfer system for the export of pathogenic macromolecules. Trends Microbiol. 4:64-68. [PubMed]

Articles from Infection and Immunity are provided here courtesy of American Society for Microbiology (ASM)