Pertussis toxin inhibits neutrophil recruitment to delay antibody-mediated clearance of Bordetella pertussis

Whooping cough is considered a childhood disease, although there is growing evidence that children are infected by adult carriers. Additionally, increasing numbers of vaccinated adults are being diagnosed with Bordetella pertussis disease. Thus it is critical to understand how B. pertussis remains endemic even in highly vaccinated or immune populations. Here we used the mouse model to examine the nature of sterilizing immunity to B. pertussis. Antibodies were necessary to control infection but did not rapidly clear B. pertussis from the lungs. However, antibodies affected B. pertussis after a delay of at least a week by a mechanism that involved neutrophils and Fc receptors, suggesting that neutrophils phagocytose and clear antibody-opsonized bacteria via Fc receptors. B. pertussis blocked migration of neutrophils and inhibited their recruitment to the lungs during the first week of infection by a pertussis toxin–dependent (PTx-dependent) mechanism; a PTx mutant of B. pertussis induced rapid neutrophil recruitment and was rapidly cleared from the lungs by adoptively transferred antibodies. Depletion of neutrophils abrogated the defects of the PTx mutant. Together these results indicate that PTx inhibits neutrophil recruitment, which consequently allows B. pertussis to avoid rapid antibody-mediated clearance and therefore successfully infect immune hosts.

The widespread use of vaccines in developed nations has decreased the incidence of whooping cough (1–3). However, recent surveys reveal that a majority of individuals in a vaccinated population are transiently infected with the causative agent, Bordetella pertussis, and that it is widespread and endemic (4–8). In vaccinated populations, however, the bacterium induces a mild form of the disease that often goes undiagnosed (9–11). Although severe disease may be the greatest public health concern, undiagnosed pertussis poses an ongoing pervasive risk to very young (pre-vaccine), unvaccinated, and immune-compromised populations. In fact, childhood disease predates the age at which children extensively socialize with each other and appears to commonly have as its source an adult, non- or mildly symptomatic carrier (10–14). The ability of B. pertussis to circulate in vaccinated and immune populations has been known clinically for years but has not been well studied experimentally. Although experimental infection of naive mice may simulate disease, infection of vaccinated or convalescent animals with waning immunity may be more relevant to the biology of the bacterium in a vaccinated population.

Current pertussis vaccines induce a strong serum antibody response that has been shown to be critical for protection from the disease (15, 16). However, their efficacy against subclinical infection is doubtful, as the majority of vaccinated populations test positive for subsequent infection (10, 17), suggesting that the bacterium successfully infects immune and/or vaccinated individuals. Using animal models, we and others have previously shown that although B cells are necessary for B. pertussis clearance from the respiratory tract (18–20), adoptively transferred serum antibodies have little or no effect on bacterial numbers for the first 7 days after inoculation (18–21) but begin to control and clear the bacteria thereafter. The ability to resist rapid antibody-mediated clearance may increase the duration and intensity of infection, both of which facilitate the transmission of the bacteria and would therefore be critical to the endemism of B. pertussis in vaccinated populations.

B. pertussis is thought to have emerged from a B. bronchiseptica–like progenitor (22, 23). These closely related subspecies share a similar set of virulence determinants but a different host range (24). Interestingly, while both require B cells for their clearance from the respiratory tract, only B. bronchiseptica is rapidly (within 3 days) cleared by adoptively transferred serum antibodies (20). We previously elucidated the mechanism of antibody-mediated clearance of B. bronchiseptica in order to determine the pathway that is presumably inhibited by B. pertussis (25, 26). Serum antibody–mediated clearance of B. bronchiseptica requires a TLR4-induced early recruitment of neutrophils that phagocytose bacteria via Fcγ receptors (FcγRs) and CR3. We hypothesized that serum antibody–mediated clearance of B. pertussis also requires neutrophils and that it may resist rapid serum antibody–mediated clearance by inhibiting neutrophil recruitment, presumably via a mechanism not shared by B. bronchiseptica.

Pertussis toxin (PTx), which is only expressed by B. pertussis, is an A-B type toxin known to inhibit G protein signaling pathways that involve Giα, interfering with a class of receptor that includes the majority of the chemokine receptors (27, 28). Various in vitro and in vivo studies have demonstrated its ability to inhibit the chemotaxis of neutrophils, lymphocytes, and macrophages (29–31). Although PTx has been proposed to be involved in the pathogenesis of whooping cough (32), its exact role in vivo during B. pertussis infection is not yet understood. Addition of inactivated PTx in the pertussis vaccine preparations has helped improve the vaccine’s efficacy, suggesting that anti-PTx antibodies are important for protection against B. pertussis disease (15, 33). Furthermore, a serological study indicated a positive correlation between anti-PTx antibody levels and protection from disease (34).

In the current analysis of the mechanism of antibody-mediated bacterial clearance, we observed that B. pertussis clearance was similar to that of B. bronchiseptica: antibody-facilitated clearance of both bacteria required FcγRs and neutrophils. A significant difference, however, was observed in the kinetics of clearance of B. pertussis. The delayed elimination of bacteria correlated with the delayed recruitment of neutrophils to the lungs. Consequently, a PTx mutant of B. pertussis was rapidly cleared by adoptively transferred serum antibodies, suggesting that PTx inhibits early neutrophil recruitment and thereby contributes to the delayed antibody-mediated bacterial clearance. Expression of PTx may be an adaptation strategy of B. pertussis to cause an acute infection and extend its infectious period in immune hosts, facilitating its persistence in immune human populations.

Serum antibodies do not rapidly clear B. pertussis. We and others have previously shown that B cells are necessary for the clearance of B. pertussis from the lungs (18–20). Therefore we sought to investigate the role of serum antibodies in bacterial clearance. Groups of wild-type mice were inoculated with 5 × 10⁵ CFU of B. pertussis in 50 μl of PBS i.n., which has previously been shown to consistently deliver bacteria into the lungs. Immediately after inoculation, 200 μl of either naive or convalescent-phase serum were injected i.p. Mice were sacrificed on days 1, 3, 7, 10, and 14 after inoculation, and lungs were harvested to enumerate the number of bacteria. Numbers of bacteria recovered from the lungs of either naive serum– or convalescent-phase serum–treated mice did not differ significantly on days 1, 3, and 7 after inoculation, suggesting that serum antibodies have minimal effect on bacterial clearance during the first 7 days after inoculation (Figure 1). However, convalescent-phase serum–treated mice harbored only approximately 1,000 bacteria, while naive serum–treated mice harbored approximately 10⁵ bacteria on day 10 after inoculation (P < 0.001). Similarly, no bacteria could be detected in the lungs of convalescent-phase serum–treated mice on day 14 after inoculation, while approximately 10⁵ CFU of B. pertussis were recovered from the lungs of naive serum–treated mice (P < 0.001). These data suggest that serum antibodies were able to clear B. pertussis from the lungs only after 1 week of infection. Apparently B. pertussis resists the effect of antibodies during the first week of infection, probably by inhibiting 1 or more antibody effector function.

Figure 1

Serum antibodies clear B. pertussis after 7 days of inoculation. C57BL/6 mice were inoculated with 5 × 10⁵ CFU of B. pertussis and adoptively transferred with 200 μl of naive serum (circles) or convalescent-phase serum (triangles). Lungs were harvested on the indicated days, and bacterial numbers were enumerated and expressed as geometric mean ± SEM. Dotted line denotes limit of detection. n = 4 per group. *P < 0.001.

FcγRs, but not C3, are required for serum antibody–mediated clearance of B. pertussis. Antibodies may clear bacteria by neutralization, complement-mediated lysis, or opsonization for FcγR-mediated phagocytosis. We have previously shown that serum antibodies rapidly clear B. bronchiseptica from the lungs of mice and that the mechanism involved both complement and FcγRs (25, 26). Since B. pertussis is very closely related to B. bronchiseptica but is cleared much more slowly by serum antibodies, we predicted that B. pertussis has some mechanism to resist 1 or more of these antibody effector functions early in the infection. In order to test the importance of complement in serum antibody–mediated bacterial clearance, we used mice lacking the central complement component C3, required for both classical and alternative complement cascades. Groups of wild-type C57BL/6 and congenic C3^–/– mice were inoculated with B. pertussis as described above. Immediately after inoculation, 200 μl of naive or B. pertussis–induced convalescent-phase serum was injected i.p. Fourteen days after inoculation, mice were sacrificed, and lungs were harvested to enumerate B. pertussis CFU. Naive serum had no effect on bacterial numbers in the lungs of wild-type or C3^–/– mice (Figure 2). Convalescent-phase serum completely cleared the bacteria from the lungs of wild-type and C3^–/– mice by day 14 (P < 0.001). These data indicate that C3 is not required for serum antibody–mediated clearance of B. pertussis.

Figure 2

Adoptively transferred antibodies clear B. pertussis in the lungs of wild-type and C3^–/–, but not FcγR^–/–, mice. C57BL/6 (WT), C3^–/–, and FcγR^–/– mice were inoculated with 5 × 10⁵ CFU of B. pertussis and adoptively transferred 200 μl of either naive serum (white bars) or convalescent-phase serum (black bars). On day 14 after inoculation, mice were sacrificed, and bacteria in the lungs were enumerated and expressed as geometric mean ± SEM. Dotted line denotes limit of detection. n = 4 per group. *P < 0.001.

In order to test the importance of FcγRs in antibody-mediated clearance of B. pertussis, we used FcγR^–/– mice, which lack all 3 Fc receptors for IgG and 1 for IgE (35). Groups of FcγR^–/– mice were inoculated with B. pertussis and adoptively transferred naive serum or convalescent-phase serum as described above. As opposed to wild-type mice, in which convalescent-phase serum completely cleared bacteria from the lungs within 14 days, the same immune serum had no effect on bacterial numbers in the lungs of FcγR^–/– mice (P < 0.001; Figure 2). These data indicate that FcγRs are required for serum antibody–mediated clearance of B. pertussis. Together, these data showed that antibodies were not functioning by neutralization alone or via complement-mediated lysis. Antibodies appeared to clear B. pertussis from the lungs via phagocytosis by recruited FcγR-bearing cells.

Neutrophils are required for antibody-mediated clearance of B. pertussis. We have previously shown that FcγR-bearing neutrophils are the primary inflammatory cells recruited to the lungs following infection with B. bronchiseptica and are required for rapid antibody-mediated bacterial clearance (26, 36). Since B. pertussis clearance required FcγRs, we sought to determine whether neutrophils are required for antibody-mediated clearance of this organism. Groups of C57BL/6 mice were injected i.p. with PBS or 1 mg anti–Ly-6 monoclonal antibodies (RB6-8C5). This treatment has been previously shown to deplete neutrophils for at least 2 weeks with no apparent effect on other cells such as macrophages or dendritic cells (37, 38). Following treatment, mice were inoculated with B. pertussis and adoptively transferred naive or immune serum as described above. The mice were sacrificed on day 14 after inoculation, and bacterial numbers in the lungs were enumerated (Figure 3). Naive serum had no effect on bacterial numbers in the lungs of mice treated with PBS or RB6-8C5. However, immune serum completely cleared bacteria from the lungs of mice treated with PBS, but not those treated with RB6-8C5 (P < 0.001), indicating that neutrophils are required for antibody-mediated bacterial clearance. The above-mentioned observations were verified using a recently isolated strain of B. pertussis, 6068 (a kind gift from Jeff Miller, UCLA, Los Angeles, California, USA).

Figure 3

Depletion of neutrophils abrogates antibody-mediated clearance of B. pertussis. C57BL/6 mice were treated with PBS or 1 mg of neutrophil-depleting mAb RB6-8C5. Mice were subsequently inoculated with 5 × 10⁵ CFU of B. pertussis and adoptively transferred 200 μl of either naive (white bars) or convalescent-phase serum (black bars). On day 14 after inoculation, mice were sacrificed, and bacteria in the lungs were enumerated and expressed as geometric mean ± SEM. Dotted line denotes limit of detection. n = 4 per group. *P < 0.001.

PTx inhibits antibody-mediated clearance of B. pertussis. B. bronchiseptica and B. pertussis are very closely related and share a majority of the known virulence genes (22, 23). These 2 closely related subspecies both required similar antibody effector functions and phagocytic cells for their clearance from the lungs of mice (Figure 2 and ref. 26). Yet immune serum substantially reduces the numbers of B. bronchiseptica in the lungs within 1 day (by more than 99%), whereas it has no effect on the numbers of B. pertussis for at least 7 days after inoculation (20). These data suggested that some B. pertussis–specific virulence gene(s) delay antibody-mediated bacterial clearance. PTx is a B. pertussis–specific virulence factor that has been extensively studied in vitro and shown to inhibit the downstream effects of G protein–coupled receptors, including many chemokine receptors (27, 28). Furthermore, in vitro and in vivo experiments using purified PTx have shown that PTx can inhibit recruitment of cells such as neutrophils and lymphocytes (29–31). One plausible explanation for the delayed bacterial clearance may be that secreted PTx inhibits early neutrophil recruitment, which is necessary for antibody-mediated clearance. In support of this argument, we have previously observed that B. bronchiseptica induces significantly more neutrophil infiltration than B. pertussis on day 3 after inoculation (36). Therefore, we predicted that in the absence of PTx, B. pertussis would be more susceptible to antibody-mediated clearance. In order to test this hypothesis, we inoculated mice with 5 × 10⁵ CFU of B. pertussis or the PTx mutant B. pertussisΔPTx as described above and adoptively transferred either naive or convalescent-phase serum. Lungs were harvested from these mice on days 3 and 7 after inoculation, and bacterial numbers in the lungs were enumerated. Convalescent-phase serum rapidly reduced numbers of B. pertussisΔPTx in the lungs as early as 3 days after inoculation (P < 0.05) and completely cleared this bacterium within 7 days (P < 0.05), but had no effect on the numbers of B. pertussis in this time frame (Figure 4). This was not due to the higher number of wild-type bacteria, since we have previously observed that adoptively transferred antibodies clear more than 10⁶ bacteria when PTx is not present (G.S. Kirimanjeswara and E.T. Harvill, unpublished observations). These data suggest that PTx is involved in inhibiting antibody-mediated clearance of B. pertussis, perhaps by inhibiting neutrophil migration to the lungs. Additionally, analysis of the antibody-mediated clearance of B. pertussis in the trachea indicated a similar mechanism as that of the lungs was involved, and PTx inhibited rapid bacterial clearance in this organ as well. However, no significant effect of adoptively transferred serum antibodies was observed on the bacterial numbers in the nasal cavity on any tested days.

Figure 4

Serum antibodies rapidly clear B. pertussisΔPTx but not B. pertussis. C57BL/6 mice were inoculated with either B. pertussis (B.p) or B. pertussisΔPTx (B.pΔPTx). Subsequently, these mice were adoptively transferred 200 μl of either naive serum (NS) or convalescent-phase serum (IS). Mice were sacrificed on the indicated days, and bacterial numbers in the lungs were determined and expressed as geometric mean ± SEM. Dotted line denotes limit of detection. n = 4 per group. *P < 0.001; **P < 0.05.

PTx reduces neutrophil recruitment to the lungs. B. bronchiseptica induces the recruitment to the lungs of substantially higher numbers of neutrophils than does B. pertussis during the first week of infection (20, 26, 36). Since neutrophils are required for antibody-mediated clearance of both B. bronchiseptica and B. pertussis, we predicted that B. pertussis has mechanisms to inhibit neutrophil recruitment to the lungs in order to resist the effect of serum antibodies. Additionally, as previous studies have shown that PTx reduces the proportion of neutrophils recovered in the broncheoalveolar lavage fluid (39), we hypothesized that PTx inhibits the migration of neutrophils to the lungs, decreasing their numbers and thereby inhibiting antibody-mediated bacterial clearance. To test these hypotheses, groups of mice were inoculated with B. pertussis or B. pertussisΔPTx as described above, and leukocytes recruited to the lungs were enumerated. Approximately 5–6 × 10⁵ leukocytes were observed on day 1 after inoculation, and 7–8 × 10⁵ on days 3 and 7, in the lungs of mice inoculated with B. pertussis (Figure 5). On days 1, 3, and 7 after inoculation, the majority of leukocytes in the lungs of mice inoculated with B. pertussis were macrophages (∼4 × 10⁵), a small percentage of cells were lymphocytes (∼2 × 10⁵), and the rest were neutrophils (∼1 × 10⁵). In contrast, the lungs of mice inoculated with B. pertussisΔPTx harbored approximately 10–12 × 10⁵ leukocytes on days 1, 3 and 7 after inoculation, and the majority of these cells were neutrophils (5–6 × 10⁵). The absolute number of macrophages and lymphocytes recruited were similar between the 2 groups (P < 0.05). These data indicate that the higher proportion of neutrophils previously reported to be recruited by B. pertussisΔPTx (39) may be attributed to a large increase in absolute numbers of neutrophils within the lungs. Thus, we concluded that PTx appears to inhibit neutrophil recruitment to the lungs during the first week of infection. As reported by other investigators (40), the number of leukocytes, particularly neutrophils, began to increase in the lungs of mice infected with B. pertussis on day 7 after inoculation, reaching a peak at day 10 after inoculation. On day 10, the lungs of mice inoculated with B. pertussis harbored approximately 10 × 10⁵ leukocytes, of which 5–6 × 10⁵ cells were neutrophils and 2–3 × 10⁵ were lymphocytes, indicating that the effects of PTx are ultimately overcome and neutrophils are recruited to the lungs later during infection. Interestingly, this delayed neutrophil recruitment correlates with the time frame in which antibodies begin to be effective against B. pertussis.

Figure 5

PTx inhibits recruitment of neutrophils to the lungs. C57BL/6 mice were inoculated with 5 × 10⁵ CFU of B. pertussis (circles) or B. pertussisΔPTx (squares). Mice were sacrificed on the indicated days and their lungs perfused with 5 ml cold PBS, and total leukocytes per whole lung were enumerated. Individual cell types were determined by modified Giemsa staining of cells. n = 4 per group. *P < 0.001; **P < 0.05.

Antibody-mediated clearance of B. pertussis δPTx requires FcγRs and neutrophils. The above data reveal a strong correlation between neutrophil recruitment to the lungs and serum antibody–mediated bacterial clearance and suggest that neutrophils eliminate B. pertussis via FcγR-mediated opsonization and phagocytosis. However, it is possible that B. pertussisΔPTx may be cleared by serum antibodies by a mechanism different from that of wild-type B. pertussis. Therefore, we sought to determine the mechanism of antibody-mediated clearance of B. pertussisΔPTx. Groups of C57BL/6 and FcγR^–/– mice were inoculated with B. pertussisΔPTx and adoptively transferred 200 μl of either naive or immune serum as described above. Naive serum had no significant effect on the bacterial numbers in the lungs of wild-type or FcγR^–/– mice on either day 3 or day 7 (Figure 6A). In wild-type mice, immune serum rapidly reduced bacterial numbers in the lungs by day 3 after inoculation (P < 0.001) and completely cleared bacteria by day 7 (P < 0.001). However, no significant reduction in the numbers of bacteria was observed in the lungs of immune serum–treated FcγR^–/– mice on days 3 and 7 after inoculation, indicating that FcγRs are required for antibody-mediated clearance of B. pertussisΔPTx.

Figure 6

FcγRs and neutrophils are required for serum antibody–mediated clearance of B. pertussisΔPTx. (A) C57BL/6 or FcγR^–/– mice were inoculated with 5 × 10⁵ CFU of B. pertussisΔPTx and adoptively transferred 200 μl of either naive or convalescent-phase serum. (B) C57BL/6 mice also received either PBS or 1 mg of mAb RB6-8C5 (n = 4 per group). Lungs were harvested on the indicated days, and bacteria were enumerated and expressed as mean ± SEM. Dotted line denotes limit of detection. n = 4 per group. *P < 0.001; **P < 0.05.

Based on these results, we tested the hypothesis that neutrophils are involved in rapid antibody-mediated clearance of B. pertussisΔPTx. Groups of C57BL/6 mice were inoculated with B. pertussisΔPTx and adoptively transferred 200 μl of either naive or immune serum as described above. Half of the mice in each group were also given 1 mg RB6-8C5 injected i.p. to deplete neutrophils, and the other half were given PBS as control. Among the mice given naive serum, RB6-8C5 treatment resulted in approximately 10- to 100-fold higher numbers of B. pertussisΔPTx in the lungs compared with PBS treatment (Figure 6B). These numbers were indistinguishable from the numbers of wild-type B. pertussis in the lungs at this time point (compare to Figure 4), indicating that PTx is not required for efficient colonization when neutrophils are depleted. These data strongly suggest that PTx enables B. pertussis to colonize the lung by inhibiting neutrophil recruitment, as proposed by Carbonetti et al. (39). Immune serum rapidly reduced the number of B. pertussisΔPTx in the lungs of mice treated with PBS on day 3 (P < 0.001) and completely cleared bacteria on day 7 after inoculation (P < 0.001). However, immune serum failed to reduce the number of bacteria on days 3 and 7 after inoculation in mice treated with RB6-8C5, indicating that neutrophils are required for rapid serum antibody–mediated clearance of B. pertussisΔPTx. These data indicate that serum antibodies clear both B. pertussis and B. pertussisΔPTx by a similar mechanism involving FcγR-dependent phagocytosis by neutrophils. These data were verified using an independently created PTx mutant of B. pertussis (a kind gift from N. Carbonetti, University of Maryland, College Park, Maryland, USA).

PTx does not modulate chemokine and cytokine production by alveolar macrophages. PTx could affect neutrophil recruitment in various ways. It is known to influence chemokine and cytokine responses (41), interfere with G protein–coupled chemokine receptor signaling (42), and downregulate adhesion molecule expression on endothelial cells (43). To investigate whether PTx modulates chemokine expression by alveolar macrophages, a confluent layer of murine alveolar macrophage (MH-S) cells was incubated with B. pertussis or B. pertussisΔPTx at an MOI of 10 for 12 hours. The accumulated amounts of various chemokines, such as macrophage inflammatory protein-1α (MIP-1α), monocyte chemoattractant protein 1 (MCP-1), chemokine RANTES (RANTES), and chemokine KC (KC), were then analyzed by ELISA. Figure 7 shows that macrophages produced significant amounts of these chemokines when treated with 100 ng/ml of E. coli LPS as a positive control or infected with B. pertussis or B. pertussisΔPTx. However, no significant differences were observed between the number of chemokines produced by macrophages incubated with these 2 strains. These data suggest that PTx does not inhibit neutrophil recruitment by blocking chemokine production.

Figure 7

PTx does not modulate chemokine production by alveolar macrophages. MH-S cells were incubated with B. pertussis (black bars) or B. pertussisΔPTx (dark gray bars) at MOI of 10 for 12 hours. Also shown are medium controls (white bars) and positive controls (100 ng/ml E. coli LPS, light gray bars). Culture supernatants were collected and analyzed for chemokines by ELISA and expressed as mean ± SEM.

PTx inhibits neutrophil migration. In order to determine whether PTx acts directly on neutrophils or endothelial cells to inhibit neutrophil recruitment to the lungs, we used an in vitro neutrophil migration assay to determine the number of neutrophils that migrate across endothelial cells toward chemokines produced by macrophages incubated with either B. pertussis or B. pertussisΔPTx at an MOI of 10 for 12 hours. The average number of neutrophils per field that migrated toward a media negative control was approximately 35 (Figure 8). More than 500 neutrophils per field were found to migrate across an endothelial cell barrier toward the supernatant from macrophages treated with E. coli LPS. The number of neutrophils that migrated toward supernatant from macrophages incubated with B. pertussis was slightly higher than the negative control at approximately 45 per field. However, the number of neutrophils that migrated toward supernatant from macrophages incubated with B. pertussisΔPTx was approximately 400 per field (P < 0.001). Since our previous data indicated that PTx does not modulate chemokine production by alveolar macrophages, we predicted that PTx in supernatants from macrophages incubated with B. pertussis may alter chemotaxis of neutrophils. In order to address this possibility, equal amounts of supernatants from cells incubated with B. pertussis and B. pertussisΔPTx were mixed and used for the in vitro neutrophil migration assay. The number of neutrophils that migrated toward this mixture was approximately 50 per field, similar to the number of neutrophils that migrated toward supernatant from cells incubated with B. pertussis alone. These data suggest that culture supernatants from cells incubated with B. pertussis have some components that inhibit chemotaxis of neutrophils toward supernatant from B. pertussisΔPTx–exposed macrophages.

Figure 8

PTx inhibits neutrophil migration. In 3-μM transwell chambers, 10⁵ PMNs were allowed to migrate across endothelial cells toward supernatant from alveolar macrophages (Mφ) incubated with B. pertussis or B. pertussisΔPTx. Purified PTx or catalytically inactive PTx (ΔPTx) were added to the supernatant as indicated. The number of neutrophils that migrated to the lower chamber were enumerated by observing 10 fields and expressed as mean ± SEM. The experiments were repeated at least 3 times. *P < 0.001.

Since PTx has been shown to affect G protein–coupled receptor signaling pathways that may influence chemotaxis, we hypothesized that the PTx present in the supernatant from macrophages incubated with B. pertussis may directly act on neutrophils to decrease their migration across endothelial cells. To test this, we mixed various amounts of purified PTx (obtained from N. Carbonetti) with the supernatant from macrophages incubated with B. pertussisΔPTx. Neutrophil migration toward supernatants from cells incubated with B. pertussisΔPTx was reduced by 90% when purified PTx was added (P < 0.001; Figure 8). However, adding the same amount of catalytically inactivated PTx (a kind gift of N. Carbonetti; described in refs. 39, 44–46) had no effect on the migration of neutrophils across endothelial cells. Mixing purified PTx with supernatant from macrophages treated with E. coli LPS also resulted in a significant reduction in neutrophil migration (P < 0.05). These data indicate that PTx inhibits the migration of neutrophils across endothelial cells and that this inhibition is dependent on its enzymatic activity.

The inhibitory effect of PTx may be due to its action on endothelial cells or on neutrophils. In order to differentiate between these 2 possibilities, endothelial cells growing on the upper chamber of the transwell system were preincubated with B. pertussis or B. pertussisΔPTx. Subsequently, neutrophils were allowed to migrate across them toward supernatants from macrophages incubated with B. pertussisΔPTx. Preincubation of endothelial cells with either B. pertussis or B. pertussisΔPTx did not alter neutrophil migration (data not shown). In fact, neutrophil migration assays carried out in the absence of endothelial cells showed that adding PTx to supernatant from macrophages incubated with B. pertussisΔPTx inhibited migration of neutrophils (data not shown). These data suggest that PTx acts directly on neutrophils and affects their chemotaxis across a membrane and their migration across endothelial cells.

PTx has been studied extensively for its adjuvant properties (47) and inhibition of GTPase activities (48, 49). Although it has many effects on various cell types in vitro, its contributions to B. pertussis infection and pathogenesis have not yet been determined. Recently, Carbonetti et al. reported that PTx is required for efficient early colonization by B. pertussis in the lungs of mice (39). Our present data indicate that PTx facilitates B. pertussis colonization by inhibiting early recruitment of neutrophils to the lungs. However, we also noted a much stronger effect of PTx on the ability of B. pertussis to resist antibody-mediated clearance that may be critical to the biology of this endemic human pathogen.

Based on its known in vitro activities, PTx could inhibit neutrophil recruitment by modulating chemokine or cytokine production by epithelial cells and macrophages, directly interfering with the chemokine receptor signaling, and/or altering adhesion molecule expression that may interfere with the diapedesis of blood leukocytes (30, 41, 45, 50, 51). In our study, both B. pertussis and B. pertussisΔPTx induced production of similar levels of a range of chemokines by alveolar macrophages. However, only the supernatant from macrophages incubated with B. pertussisΔPTx facilitated the migration of neutrophils across endothelial cells. Additionally, purified PTx, but not catalytically inactive PTx, was able to inhibit the migration of neutrophils in this assay. Together, these data indicate that PTx inhibits polymorphonuclear neutrophil (PMN) recruitment by acting directly on chemokine receptor signaling of neutrophils. However, other effects of PTx may also contribute to the inhibition of neutrophil recruitment in vivo. The migration of neutrophils across airway epithelium requires PTx-sensitive signals (52), but the nature of these signals is difficult to delineate during an active infection with B. pertussis, not only due to the multiple possible inhibitory effects of PTx, but also due to the variety of pro- and antiinflammatory factors induced and released by the bacteria. Interestingly, the number of neutrophils recruited to the lungs increased after day 7 after inoculation in spite of the presence of PTx. This increase in the number of neutrophils correlates with the kinetics of antibody-mediated bacterial clearance. We have observed that a T cell–dependent response was required to overcome the effect of PTx and allow antibody-mediated bacterial clearance during the second week of infection (data not shown). The implication of an inhibitory effect of PTx on neutrophil recruitment is 2-fold: it allows B. pertussis to evade innate immunity early in colonization and also provides a means to resist the effect of serum antibodies in the lungs. Although the former effect may be important to primary infection in naive, unvaccinated hosts, the latter may be more relevant to the persistence of B. pertussis as an endemic pathogen even within vaccinated populations.

Acute, highly contagious, immunizing pathogens face the significant epidemiological challenge of long-term persistence within the host population. Immunity results in depletion of susceptible hosts through the course of each epidemic; thereafter host replenishment requires births of susceptible individuals or loss of immunity — which is why pathogens that convey strong, long-lasting immunity result in “childhood diseases” (53). Rapid contagion, in turn, results in fast transmission among hosts, which is a short-term evolutionary benefit to the pathogen (32, 54). However, it also results in large-amplitude epidemics with intervening deep epidemic troughs. In small and medium host populations, the chain of transmission will be broken in the troughs so that the pathogen will go extinct. The most relevant theoretical models for childhood infections, the so-called realistic age-structured models (55, 56), predict an endemic threshold of around half a million hosts in order for transmission to be sustained through the epidemic troughs of acute, immunizing infections. This prediction is closely matched by epidemiological surveillance data (57, 58). Previous theoretical studies have highlighted 2 key adaptations that increase the height of the epidemic troughs to allow long-term endemism within smaller host communities: (a) reinfection of previously immunized hosts and adult carriers and (b) prolongation of the infectious period (7). Our study is of wide epidemiological significance in showing that B. pertussis, through expression of PTx, slows migration of neutrophils and thereby extends the infection period (relative to B. pertussisΔPTx strains) and allows for transient reinfection of previously immunized hosts. PTx expression may, therefore, be a key adaptation by B. pertussis for interacting with the unique population dynamics of its human host.

Susceptibility of humans to B. pertussis infection, particularly among people with no history of disease, has been shown to correlate with low levels of anti-PTx antibodies (59). Other studies have recognized that PTx is one of the critical components necessary for higher efficacy of acellular pertussis vaccines (34, 60). Although antibody titer to pertactin positively correlates with the resistance to diagnosed disease symptoms, addition of pertactin or filamentous hemagglutinin to PTx increases the efficacy of acellular pertussis vaccines (15, 33). Interestingly, antibody titers to PTx decrease much faster than those to other B. pertussis antigens (61). Since PTx allows B. pertussis to largely avoid the effects of antibodies to other antigens, the most effective vaccination strategy to B. pertussis may involve an increased focus on the induction of a long-lasting serum antibody response to PTx. Use of genetically inactivated mutant PTx in the current acellular vaccines, as proposed by Robbins et al. (62), may be a substantial improvement in achieving higher level of anti-PTx antibodies that reduce the rate of infection and severity of disease.

We thank all the members of E.T. Harvill’s lab, Avery August, Andrew Henderson, and the members of the Center for Infectious Disease Dynamics, The Pennsylvania State University, for critical reading of this manuscript and discussion of this study. We thank Nicholas Carbonetti (University of Maryland), Jeff Miller (UCLA), and Drusilla Burns (FDA) for providing various strains of B. pertussis and purified PTx. This work was supported by NIH grant 5-R01-A1053075-02.

Nonstandard abbreviations used: FcγR, Fcγ receptor; MH-S, murine alveolar macrophage (cell line); PMN, polymorphonuclear neutrophil; PTx, pertussis toxin; RB6-8C5, anti–Ly-6 monoclonal antibodies.

Conflict of interest: The authors have declared that no conflict of interest exists.

1. Cherry, JD. The epidemiology of pertussis and pertussis immunization in the United Kingdom and the United States: a comparative study. Curr. Probl. Pediatr. 1984. 14:1-78.
   View this article via: PubMed Google Scholar
2. Cherry, JD, Brunell, PA, Golden, GS, Karson, DT. Report of the task force on pertussis and pertussis immunization-1988. Pediatrics. 1988. 81(Suppl.):939-984.
3. Cherry, J.D., and Heininger, U. 2004. Pertussis and other Bordetella infections. In Textbook of pediatric infectious diseases. W.B. Saunders. Philadelphia, Pennsylvania, USA. 1588–1608.
4. Gilberg, S, et al. Evidence of Bordetella pertussis infection in adults presenting with persistent cough in a French area with very high whole-cell vaccine coverage. J. Infect. Dis. 2002. 186:415-418.
   View this article via: PubMed Google Scholar
5. Hodder, SL, et al. Antibody responses to Bordetella pertussis antigens and clinical correlations in elderly community residents. Clin. Infect. Dis. 2000. 31:7-14.
   View this article via: PubMed Google Scholar
6. Nennig, ME, Shinefield, HR, Edwards, KM, Black, SB, Fireman, BH. Prevalence and incidence of adult pertussis in an urban population. JAMA. 1996. 275:1672-1674.
   View this article via: PubMed Google Scholar
7. Schmitt-Grohe, S, et al. Pertussis in German adults. Clin. Infect. Dis. 1995. 21:860-866.
   View this article via: PubMed Google Scholar
8. Wright, SW, Edwards, KM, Decker, MD, Zeldin, MH. Pertussis infection in adults with persistent cough. JAMA. 1995. 273:1044-1046.
   View this article via: PubMed Google Scholar
9. Deville, JG, et al. Frequency of unrecognized Bordetella pertussis infections in adults. Clin. Infect. Dis. 1995. 21:639-642.
   View this article via: PubMed Google Scholar
10. Deen, JL, et al. Household contact study of Bordetella pertussis infections. Clin. Infect. Dis. 1995. 21:1211-1219.
    View this article via: PubMed Google Scholar
11. Cherry, JD, et al. Comparison of values of antibody to Bordetella pertussis antigens in young German and American men. Clin. Infect. Dis. 1995. 20:1271-1274.
    View this article via: PubMed Google Scholar
12. Crowcroft, NS, et al. Severe and unrecognised: pertussis in UK infants. Arch. Dis. Child. 2003. 88:802-806.
    View this article via: PubMed Google Scholar
13. Baron, S, et al. Epidemiology of pertussis in French hospitals in 1993 and 1994: thirty years after a routine use of vaccination. Pediatr. Infect. Dis. J. 1998. 17:412-418.
    View this article via: PubMed Google Scholar
14. Bisgard, KM, et al. Infant pertussis: who was the source? Pediatr. Infect. Dis. J. 2004. 23:985-989.
    View this article via: PubMed Google Scholar
15. Cherry, JD, Gornbein, J, Heininger, U, Stehr, K. A search for serologic correlates of immunity to Bordetella pertussis cough illnesses. Vaccine. 1998. 16:1901-1906.
    View this article via: PubMed Google Scholar
16. Storsaeter, J, Hallander, HO, Gustafsson, L, Olin, P. Levels of anti-pertussis antibodies related to protection after household exposure to Bordetella pertussis. Vaccine. 1998. 16:1907-1916.
    View this article via: PubMed Google Scholar
17. Long, SS, Welkon, CJ, Clark, JL. Widespread silent transmission of pertussis in families: antibody correlates of infection and symptomatology. J. Infect. Dis. 1990. 161:480-486.
    View this article via: PubMed Google Scholar
18. Leef, M, Elkins, KL, Barbic, J, Shahin, RD. Protective immunity to Bordetella pertussis requires both B cells and CD4(+) T cells for key functions other than specific antibody production. J. Exp. Med. 2000. 191:1841-1852.
    View this article via: PubMed Google Scholar
19. Mahon, BP, Sheahan, BJ, Griffin, F, Murphy, G, Mills, KH. Atypical disease after Bordetella pertussis respiratory infection of mice with targeted disruptions of interferon-gamma receptor or immunoglobulin mu chain genes. J. Exp. Med. 1997. 186:1843-1851.
    View this article via: PubMed Google Scholar
20. Kirimanjeswara, GS, Mann, PB, Harvill, ET. Role of antibodies in immunity to Bordetella infections. Infect. Immun. 2003. 71:1719-1724.
    View this article via: PubMed Google Scholar
21. King, AJ, et al. Role of the polymorphic region 1 of the Bordetella pertussis protein pertactin in immunity. Microbiology. 2001. 147:2885-2895.
    View this article via: PubMed Google Scholar
22. Musser, JM, Hewlett, EL, Peppler, MS, Selander, RK. Genetic diversity and relationships in populations of Bordetella spp. J. Bacteriol. 1986. 166:230-237.
    View this article via: PubMed Google Scholar
23. van der Zee, A, Mooi, F, Van Embden, J, Musser, J. Molecular evolution and host adaptation of Bordetella spp.: phylogenetic analysis using multilocus enzyme electrophoresis and typing with three insertion sequences. J. Bacteriol. 1997. 179:6609-6617.
    View this article via: PubMed Google Scholar
24. Parkhill, J, et al. Comparative analysis of the genome sequences of Bordetella pertussis, Bordetella parapertussis and Bordetella bronchiseptica. Nat. Genet. 2003. 35:32-40.
    View this article via: PubMed Google Scholar
25. Pishko, EJ, et al. Antibody-mediated bacterial clearance from the lower respiratory tract of mice requires complement component C3. Eur. J. Immunol. 2004. 34:184-193.
    View this article via: PubMed Google Scholar
26. Kirimanjeswara, G.S., et al. 2005. The complex mechanism of antibody-mediated clearance of Bordetella from the lungs requires TLR4. J. Immunol. In press.
27. Moss, J, et al. Activation by thiol of the latent NAD glycohydrolase and ADP-ribosyltransferase activities of Bordetella pertussis toxin (islet-activating protein). J. Biol. Chem. 1983. 258:11879-11882.
    View this article via: PubMed Google Scholar
28. Murayama, T, Katada, T, Ui, M. Guanine nucleotide activation and inhibition of adenylate cyclase as modified by islet-activating protein, pertussis toxin, in mouse 3T3 fibroblasts. Arch. Biochem. Biophys. 1983. 221:381-390.
    View this article via: PubMed Google Scholar
29. Spangrude, GJ, Sacchi, F, Hill, HR, Van Epps, DE, Daynes, RA. Inhibition of lymphocyte and neutrophil chemotaxis by pertussis toxin. J. Immunol. 1985. 135:4135-4143.
    View this article via: PubMed Google Scholar
30. Brito, GA, et al. Role of pertussis toxin A subunit in neutrophil migration and vascular permeability. Infect. Immun. 1997. 65:1114-1118.
    View this article via: PubMed Google Scholar
31. Meade, BD, Kind, PD, Manclark, CR. Lymphocytosis-promoting factor of Bordetella pertussis alters mononuclear phagocyte circulation and response to inflammation. Infect. Immun. 1984. 46:733-739.
    View this article via: PubMed Google Scholar
32. Hewlett, EL. A commentary on the pathogenesis of pertussis. Clin. Infect. Dis. 1999. 28(Suppl. 2):S94-S98.
    View this article via: PubMed Google Scholar
33. Cherry, JD. Comparative efficacy of acellular pertussis vaccines: an analysis of recent trials. Pediatr. Infect. Dis. J. 1997. 16(4 Suppl.):S90-S96.
    View this article via: PubMed Google Scholar
34. Taranger, J, et al. Mass vaccination of children with pertussis toxoid--decreased incidence in both vaccinated and nonvaccinated persons. Clin. Infect. Dis. 2001. 33:1004-1010.
    View this article via: PubMed Google Scholar
35. Ravetch, JV. Fc receptors. Curr. Opin. Immunol. 1997. 9:121-125.
    View this article via: PubMed Google Scholar
36. Harvill, ET, Cotter, PA, Miller, JF. Pregenomic comparative analysis between bordetella bronchiseptica RB50 and Bordetella pertussis tohama I in murine models of respiratory tract infection. Infect. Immun. 1999. 67:6109-6118.
    View this article via: PubMed Google Scholar
37. Conlan, JW. Critical roles of neutrophils in host defense against experimental systemic infections of mice by Listeria monocytogenes, Salmonella typhimurium, and Yersinia enterocolitica. Infect. Immun. 1997. 65:630-635.
    View this article via: PubMed Google Scholar
38. Seiler, P, et al. Rapid neutrophil response controls fast-replicating intracellular bacteria but not slow-replicating Mycobacterium tuberculosis. J. Infect. Dis. 2000. 181:671-680.
    View this article via: PubMed Google Scholar
39. Carbonetti, NH, Artamonova, GV, Mays, RM, Worthington, ZE. Pertussis toxin plays an early role in respiratory tract colonization by Bordetella pertussis. Infect. Immun. 2003. 71:6358-6366.
    View this article via: PubMed Google Scholar
40. McGuirk, P, Mills, KH. A regulatory role for interleukin 4 in differential inflammatory responses in the lung following infection of mice primed with Th1- or Th2-inducing pertussis vaccines. Infect. Immun. 2000. 68:1383-1390.
    View this article via: PubMed Google Scholar
41. Belcher, CE, et al. The transcriptional responses of respiratory epithelial cells to Bordetella pertussis reveal host defensive and pathogen counter-defensive strategies. Proc. Natl. Acad. Sci. U. S. A. 2000. 97:13847-13852.
    View this article via: PubMed Google Scholar
42. Agarwal, RK, Sun, SH, Su, SB, Chan, CC, Caspi, RR. Pertussis toxin alters the innate and the adaptive immune responses in a pertussis-dependent model of autoimmunity. J. Neuroimmunol. 2002. 129:133-140.
    View this article via: PubMed Google Scholar
43. Shen, W, Bendall, LJ, Gottlieb, DJ, Bradstock, KF. The chemokine receptor CXCR4 enhances integrin-mediated in vitro adhesion and facilitates engraftment of leukemic precursor-B cells in the bone marrow. Exp. Hematol. 2001. 29:1439-1447.
    View this article via: PubMed Google Scholar
44. Pizza, M, et al. Mutants of pertussis toxin suitable for vaccine development. Science. 1989. 246:497-500.
    View this article via: PubMed Google Scholar
45. Carbonetti, NH, et al. Suppression of serum antibody responses by pertussis toxin after respiratory tract colonization by Bordetella pertussis and identification of an immunodominant lipoprotein. Infect. Immun. 2004. 72:3350-3358.
    View this article via: PubMed Google Scholar
46. Nencioni, L, et al. Characterization of genetically inactivated pertussis toxin mutants: candidates for a new vaccine against whooping cough. Infect. Immun. 1990. 58:1308-1315.
    View this article via: PubMed Google Scholar
47. Ausiello, CM, et al. Native and genetically inactivated pertussis toxins induce human dendritic cell maturation and synergize with lipopolysaccharide in promoting T helper type 1 responses. J. Infect. Dis. 2002. 186:351-360.
    View this article via: PubMed Google Scholar
48. Bokoch, GM, Gilman, AG. Inhibition of receptor-mediated release of arachidonic acid by pertussis toxin. Cell. 1984. 39:301-308.
    View this article via: PubMed Google Scholar
49. Katada, T, Tamura, M, Ui, M. The A protomer of islet-activating protein, pertussis toxin, as an active peptide catalyzing ADP-ribosylation of a membrane protein. Arch. Biochem. Biophys. 1983. 224:290-298.
    View this article via: PubMed Google Scholar
50. Boldrick, JC, et al. Stereotyped and specific gene expression programs in human innate immune responses to bacteria. Proc. Natl. Acad. Sci. U. S. A. 2002. 99:972-977.
    View this article via: PubMed Google Scholar
51. Smith, ML, Olson, TS, Ley, K. CXCR2- and E-selectin-induced neutrophil arrest during inflammation in vivo. J. Exp. Med. 2004. 200:935-939.
    View this article via: PubMed Google Scholar
52. Miller, LA, Usachenko, J, McDonald, RJ, Hyde, DM. Trafficking of neutrophils across airway epithelium is dependent upon both thioredoxin- and pertussis toxin-sensitive signaling mechanisms. J. Leukoc. Biol. 2000. 68:201-208.
    View this article via: PubMed Google Scholar
53. Lawrence, GL, MacIntyre, CR, Hull, BP, McIntyre, PB. Measles vaccination coverage among five-year-old children: implications for disease elimination in Australia. Aust. N. Z. J. Public Health. 2003. 27:413-418.
    View this article via: PubMed Google Scholar
54. Glass, K, Grenfell, BT. Waning immunity and subclinical measles infections in England. Vaccine. 2004. 22:4110-4116.
    View this article via: PubMed Google Scholar
55. Bolker, BM, Grenfell, BT. Chaos and biological complexity in measles dynamics. Proc. Biol. Sci. 1993. 251:75-81.
    View this article via: PubMed Google Scholar
56. Keeling, MJ, Grenfell, BT. Disease extinction and community size: modeling the persistence of measles. Science. 1997. 275:65-67.
    View this article via: PubMed Google Scholar
57. Bjornstad, ON, Peltonen, M, Liebhold, AM, Baltensweiler, W. Waves of larch budmoth outbreaks in the European alps. Science. 2002. 298:1020-1023.
    View this article via: PubMed Google Scholar
58. Finkenstadt, BF, Bjornstad, ON, Grenfell, BT. A stochastic model for extinction and recurrence of epidemics: estimation and inference for measles outbreaks. Biostatistics. 2002. 3:493-510.
    View this article via: PubMed Google Scholar
59. Storsaeter, J, Hallander, HO, Gustafsson, L, Olin, P. Low levels of antipertussis antibodies plus lack of history of pertussis correlate with susceptibility after household exposure to Bordetella pertussis. Vaccine. 2003. 21:3542-3549.
    View this article via: PubMed Google Scholar
60. Taranger, J, et al. Immunologic and epidemiologic experience of vaccination with a monocomponent pertussis toxoid vaccine. Pediatrics. 2001. 108:E115.
    View this article via: PubMed Google Scholar
61. Le, T, et al. Immune responses and antibody decay after immunization of adolescents and adults with an acellular pertussis vaccine: the APERT Study. J. Infect. Dis. 2004. 190:535-544.
    View this article via: PubMed Google Scholar
62. Robbins, JB, et al. The diphtheria and pertussis components of diphtheria-tetanus toxoids-pertussis vaccine should be genetically inactivated mutant toxins. J. Infect. Dis. 2005. 191:81-88.
    View this article via: PubMed Google Scholar
63. Hausman, SZ, Burns, DL. Use of pertussis toxin encoded by ptx genes from Bordetella bronchiseptica to model the effects of antigenic drift of pertussis toxin on antibody neutralization. Infect. Immun. 2000. 68:3763-3767.
    View this article via: PubMed Google Scholar
64. Circolo, A, et al. Genetic disruption of the murine complement C3 promoter region generates deficient mice with extrahepatic expression of C3 mRNA. Immunopharmacology. 1999. 42:135-149.
    View this article via: PubMed Google Scholar
65. Hodge, LM, Simecka, JW. Role of upper and lower respiratory tract immunity in resistance to Mycoplasma respiratory disease. J. Infect. Dis. 2002. 186:290-294.
    View this article via: PubMed Google Scholar
66. Mbawuike, IN, Herscowitz, HB. MH-S, a murine alveolar macrophage cell line: morphological, cytochemical, and functional characteristics. J. Leukoc. Biol. 1989. 46:119-127.
    View this article via: PubMed Google Scholar
67. Rubin-Bejerano, I, Fraser, I, Grisafi, P, Fink, GR. Phagocytosis by neutrophils induces an amino acid deprivation response in Saccharomyces cerevisiae and Candida albicans. Proc. Natl. Acad. Sci. U. S. A. 2003. 100:11007-11012.
    View this article via: PubMed Google Scholar
68. Laichalk, LL, et al. Intrapulmonary delivery of tumor necrosis factor agonist peptide augments host defense in murine gram-negative bacterial pneumonia. Infect. Immun. 1998. 66:2822-2826.
    View this article via: PubMed Google Scholar