INFECriON AND IMMUNrrY, Nov. 1992, p. 4578-4585

Vol. 60, No. 11

0019-9567/92/114578-08$02.00/0 Copyright © 1992, American Society for Microbiology

Uptake and Intracellular Survival of Bordetella pertussis in Human Macrophages RICHARD L. FRIEDMAN,"* KATHRYN NORDENSSON,1 LINDA WILSON,1 EMMANUEL T. AKPORIAYE,l AND DAVID E. YOCUM2 Departments ofMicrobiology and Immunology' and Internal Medicine,2 College of Medicine, University of Arizona, Tucson, Arizona 85724 Received 17 July 1992/Accepted 27 August 1992

Recent reports have demonstrated that Bordetella pertussis has invasive behavior in vivo and in vitro. In this study, we investigated the ability of a virulent strain, avirulent mutants, and mutants deficient in specific virulence factors to enter and survive intracellularly in human macrophages in vitro. Uptake of virulent B. pertussis was dose dependent and occurred in the absence of serum or specific antibody, with entry occurring via a microfilament-dependent phagocytic process. The virulent wild-type parental strain was internalized and persisted intracelularly over the 3 days of experiments, as determined by transmission electron microscopy and by recovery of viable plate counts. This is the first report of long-term survival of B. pertussis in human macrophages. Avirulent mutants entered macrophages, but at only an average of 1.5% of virulent parental levels, and did not survive intracellularly. Mutants which did not express adenylate cyclase toxin, filamentous hemagglutinin, or pertussis toxin had decreased abilities to enter and to survive inside macrophages. The results suggest that the internalization process, as well as intracellular survival, is virulence dependent and that mutations which inactivate expression of virulence factors may affect both. The ability of B. pertussis to enter and persist inside macrophages may be important not only for survival of the bacteria but also in the pathogenesis of whooping cough. The human pathogen Bordetella pertussis causes an acute and chronic respiratory infection which is localized to the tracheobronchial tree, causing the major childhood disease whooping cough (17, 32, 59). B. pertussis is typically viewed as a noninvasive microorganism which produces a large array of potential virulence factors that may play a role in the pathogenesis of pertussis. These factors include pertussis toxin (PT) (40, 46), filamentous hemagglutinin (FHA) (32, 45), endotoxin (17, 32, 59), adenylate cyclase toxin (AC) (8, 19), heat-labile toxin (65), tracheal cytotoxin (21), agglutinogens (55), and pertactin (6, 29). Previous reports suggest that B. pertussis may have invasive capabilities and the ability to survive intracellularly in mammalian cells in vitro and in vivo. Crawford and Fishel were the first, in 1959, to report that B. pertussis could invade and persist in HeLa and mouse kidney tissue culture cell lines (12). They suggested that the intracellular localization of B. pertussis may be an important survival mechanism in the disease process. Later work by Gray and Cheers, using the mouse intranasal infection model, reported that B. pertussis caused a persistent low-level lung infection that lasted 4 weeks, which they called immunological complaisance (22). In further investigations, they found B. pertussis present in mouse lung macrophages early in the infection, and the organism persisted through the complaisant period (7). Woods et al. investigated the ability of virulent and avirulent strains to survive and persist in a rat model for respiratory infection with B. pertussis (64). Avirulent B. pertussis was not recovered after 3 days from lungs of infected rats, while virulent B. pertussis was recovered on days 3 and 7 after inoculation but not on days 10 and 14. Recovery of viable B. pertussis from lungs of infected rats *

Corresponding author.

was again possible on day 21. Electron micrographs of lungs of infected animals revealed that the lung cells contained B. pertussis which appeared viable at 14 days. The authors propose that B. pertussis was not recoverable on days 10 and 14 because the microbe was residing intracellularly inside lung cells, perhaps alveolar macrophage (64). Ewanowich et al. recently reported that Bordetella parapertussis has the ability to invade both HeLa cells and human respiratory epithelial tissue in vitro (15). The bacteria persisted but did not multiply in the cells over the 7-h period of the study. Ewanowich and associates reported similar observations of invasion and persistence in HeLa cells with use of B. pertussis (14). Lee et al. also reported invasion and intracellular survival of virulent B. pertussis in a HeLa cell model (27), while Mouallem et al. reported the ability of B. pertussis to invade Chinese hamster ovary tissue culture cells (36). Roberts et al. also have reported the ability of B. pertussis to be internalized into HEp-2 tissue culture cells (43). Studies investigating the interaction of B. pertussis with human polymorphonuclear leukocytes (PMN) have revealed that the microbe survives intracellularly within the phagocytes (54). Intracellular survival appears to be due in part to inhibition of phagosome-lysosome fusion in PMN, yet respiratory burst activity of the phagocytes occurs at normal levels (53). Several recent studies have shown that B. pertussis can also persist intracellularly within human macrophages (4, 18, 47). Cumulatively, these results are intriguing because they suggest that Bordetella spp. have the ability to enter and persist in mammalian cells. In this study, we further investigated the interaction of virulent and mutant strains of B. pertussis with human blood macrophages and the role of the microbe's virulence factors in uptake and intracellular survival. The results demonstrate 4578

VOL. 60, 1992

SURVIVAL OF B. PERTUSSIS IN MACROPHAGES TABLE 1. Bacterial strains used

Strain (genotype)'

BP338 (parental) BP537 (vir)b BP348 (Tn5:cyaA) BP347 (vir-l::TnS) BP-TOX6 (ptxA6) BP101 (fhaBAIOI) BP101-TOX6 (faBA101 ptrA6) BC67 (cyaE::Tn5tacl)

Relevant characteristics

Virulent wild type Avirulent ACAvirulent

PTAFHAC AFHAC PTRegulates AC expression

Reference 60 41 60 60 41 41 42 9, 10

a All strains are isogenic derivatives of BP338. See references for details of strain construction. b Spontaneous avirulent variant. c BP101 is an internal in-frame deletion inJhaB which produces a truncated

FHA.

that once virulent B. pertussis is internalized into macrophages, it can survive and persist intracellularly. MATERIALS AND METHODS Bacterial strains. B. pertussis strains used for this work are listed in Table 1. Stock cultures of B. pertussis were stored at -70°C as a suspension in Stainer-Scholte medium containing 50% glycerol (54). Bacteria were inoculated onto charcoal agar (Difco Laboratories, Detroit, Mich.) supplemented with 10% horse blood, grown at 37C for 2 to 3 days, and used in experiments. In experiments using the conditional TnStacl AC mutant BC67, the mutant was grown on charcoal plates with or without the addition of 1 mM ,-D-thiogalactopyranoside (IPTG; U.S. Biochemical Corp., Cleveland, Ohio). In the presence of IPTG, BC67 expresses wild-type levels of AC, while in its absence, no AC is

expressed (9, 10). Macrophage isolation. Mononuclear cells were isolated from venous blood treated with EDTA by Ficoll-Hypaque gradient centrifugation as previously described (3). The mononuclear cell layer was washed and resuspended in RPMI 1640 with L-glutamine (Sigma Chemical Co., St. Louis, Mo.) containing 15% autologous normal human serum (NS) and adjusted to 106 cells per ml. Appropriate volumes of mononuclear cells (to give 2 x 105 monocytes per well) were incubated in wells of a 24-well tissue culture plate (Costar, Cambridge, Mass.) and allowed to adhere for 2 h at 37°C in a CO2 incubator. Nonadherent cells were removed by washing the chambers with RPMI, with adherent monocytes remaining. The number of macrophages in each chamber was verified by use of a 10-mm calibrated grid and an inverted microscope. The adhered cells were >90% macrophages, as verified by staining with Wright stain, and viability monitored by trypan blue exclusion was >95%. Macrophages were incubated overnight in RPMI-15% NS and used the next day in infection assays. Ten different blood donors were used in these studies as sources of macrophages and of serum. Normally, macrophages and serum from the same donor were used. When needed, pooled human serum was also used, with no differences observed in the viability of macrophages or in the ability of RPMI-15% NS to kill extracellular B. pertussis (see below). Macrophage infection assay. Cultures of B. pertussis, grown on charcoal agar plates, were recovered by use of sterile swabs, suspended in Stainer-Scholte medium (pH 7.4), and diluted to an appropriate optical density at 650 unm to give a concentration of 7 x 10 CFU/ml. Macrophage

4579

monolayers, washed and incubated in RPMI without serum, were infected with B. pertussis by using an infection (bacteria/macrophage) ratio of 10:1, 50:1, 100:1, or 400:1. Infected monolayers were incubated statically at 37°C in a CO2 incubator for 2 h. Monolayers were then washed three times and incubated for 1 h with RPMI-15% NS to kill the remaining extraceilular bacteria via complement activation. Serum killing was used instead of killing by gentamicin (14) because of the concern that over the extended incubation period of 3 days, the antibiotic would penetrate into the macrophages. Hand and King-Thompson have reported that gentamicin can indeed penetrate phagocytes and inactive intracellular bacteria (24). In initial experiments, gentamicin (100 ,ug/ml) was used to kill extracellular bacteria, but upon extended incubation (1 to 3 days), gentamicin appeared to penetrate into the macrophages at levels that killed the intracellular B. pertussis (data not shown). Thus, the alternate method of complement serum killing was used to circumvent this problem. Incubation of B. pertussis alone in the presence of RPMI15% NS for 1 h killed >99.9995% of the microbes (reducing numbers of viable bacteria from 2 x 107 to 2 x 103). Upon extended incubation in RPMI-15% NS for 1 day, no viable B. pertussis was recovered. Further control experiments were also done to demonstrate the requirement for viable macrophages for B. pertussis survival in the macrophage infection assay. Macrophage sonic extracts (from 2 x 105 macrophages) were added to RPMI-15% NS containing B. pertussis, and viability was monitored over 3 days. Results of these studies demonstrated that B. pertussis did not replicate or survive under these conditions, with no viable counts recovered by day 1. Therefore, in the macrophage infection model, the presence of viable B. pertussis requires internalization into viable macrophages to evade serum killing. All mutants used in these studies (Table 1) were tested and found to be susceptible to complement killing at levels observed for BP338. After extracellular killing of bacteria by 15% NS, monolayers were washed, fresh RPMI-15% NS was added, and cultures were incubated for 0, 1, 2, or 3 days. RPMI with 15% NS was observed to maintain its killing activity over the 3 days of the studies. In control experiments, uninfected macrophage monolayers were incubated with RPMI-15% NS for 1, 2, or 3 days, with the spent medium being recovered and tested for its ability to kill BP338 (3 x 106 bacteria per assay). It was found that spent medium, from all time points, killed BP338 and reduced viable numbers to 99.995% after 1 h of incubation. With extended incubation, no viable counts were recovered. B. pertussis killing was due to complement-mediated killing, since the addition to media of 10 mM EDTA, which chelates Ca2+ and blocks complement activation, eliminated all serum killing (data not

shown). After 0, 1, 2, or 3 days, monolayers were lysed by the addition of sterile distilled water containing 10 mM EDTA (to inactivate complement-mediated killing), and dilution plate counts were made on charcoal agar plates for detection of viable bacteria. Treatment of macrophage monolayers with inhibitors. In some experiments, macrophage monolayers were treated either with cytochalasin D (CD) at 1.0 and 2.5 F±g/ml or with monodanyslcadaverine (MDC) at 25 and 50 ,uM (both purchased from Sigma). Monolayers were preincubated for 1 h and during the infection assay with the inhibitors (15, 49). The levels of inhibitors used did not decrease viability of

4580

FRIEDMAN ET AL.

IN?FECT. IMMUN.

M9- Day Zero 108 1 Day One El Day Two ,_ 1070 Day Three co .106

-

: 105-_

102 1

2

101 100:1 lU:li0:1 Infection Ratios

400:1

FIG. 1. Levels of uptake and survival of virulent B. pertussis BP338 in macrophages. Macrophage monolayers were infected with BP338 at infection ratios of 10:1, 50:1, 100:1, and 400:1. Extracellular bacteria were killed by the addition of 15% NS, and cultures were washed and incubated with fresh tissue culture medium at 37°C in a CO2 incubator. At 0, 1, 2, and 3 days, monolayers were lysed and viable plate counts (CFU per monolayer) were determined. Results are presented as the means ± standard deviations of at least six experiments done in duplicate.

either the macrophages or the B. pertussis used in the experiments (data not shown). TEM. Macrophage monolayers were adhered to plastic coverslips in 24-well tissue culture plates (4 x 106 macrophage per well) and infected with BP338 at an infection ratio of 200:1 as described above. This ratio was used to assist in the ability to observe B. pertussis in the transmission electron microscopy (EM) preparations. At various time points, coverslips were washed and fixed in 0.1 M phosphate buffer (pH 7.2) containing 2% glutaraldehyde for 1 h at room temperature. Specimens were postfixed in 1% osmium tetroxide, dehydrated via graded alcohol steps, and embedded in EMbed 812 (Electron Microscopy Sciences, Fort Washington, Pa.). Sections were cut, stained with uranyl acetate and lead citrate, and viewed on a JEOL 1200EX II transmission electron microscope. Statistical analysis. The significance of differences between results was calculated by analysis of variance. All significant differences between groups identified by two-way analyses of variance at the 95% confidence level or greater were confirmed by post hoc testing. Statistical analysis was performed by Biostatistical Services, University of Arizona Health Sciences Center.

RESULTS Uptake and survival of virulent B. perussis in human macrophages. Figure 1 shows the results of uptake and survival of BP338 at infection ratios of 10:1, 50:1, 100:1, and 400:1. With increasing infection ratios, increasing numbers of B. pertussis were internalized into the macrophages. At the 10:1 infection ratio, approximately 6 x 104 B. pertussis cells were internalized on day 0, and over 3 days, viable counts steadily decreased. At the 50:1 infection ratio, higher numbers were internalized (approximately 3.5 x 105), but viable counts decreased over the 3 days. At the 100:1 and 400:1 infection ratios, significantly higher numbers of microbes (2.7 x 106 and 4.4 x 106, respectively) (P < 0.05) were internalized by the macrophages. These values represent approximately 13 microbes per macrophage at the 100:1 ratio and 22 microbes per macrophage at the 400:1 ratio. Viable counts remained constant over the 3 days of the study

at these infection ratios. While graphically, viable numbers appeared to increase (Fig. 1), statistically there was no significant difference between viable counts at days 0 to 3 at both 100:1 and 400:1 infection ratios (P < 0.05). The percentage of initial inoculum which was internalized by the macrophage monolayer at time zero peaked at the 100:1 infection ratio, with 12.4% of the infectious dose being internalized. Viability of the macrophage monolayers over the 3 days of the studies was monitored via trypan blue exclusion. Monolayers infected at a 10:1, 50:1, or 100:1 infection ratio had levels of viability comparable to those observed with control uninfected macrophages (91 to 98% viability over 3 days). Only at the 400:1 infection ratio was a decrease in macrophage viability observed, with 90 and 80% viability observed by days 2 and 3, respectively (data not shown). EM examination of macrophages infected with B. pertussis. EM was performed to verify that B. pertussis BP338 had indeed been internalized by macrophages in the infection assay and was present intracellularly. Figure 2 shows typical observations of intracellular B. pertussis in the cytoplasm of macrophages on days 0 and 3. On day 0 (Fig. 2A and B) after initial uptake, numerous B. penussis organisms were present in tightly fitting phagosomes in the cytoplasm of the macrophages. The microbes were always observed to be contained in phagosomes and not free in the cell cytoplasm. On day 3, microbes were still present singly in phagosomes, suggesting that no intracellular multiplication of the microbes had occurred (Fig. 2C and D). Similar EM results were obtained on days 1 and 2 of these studies (data not shown). Mechanism of B. pertussis entry into macrophages. Two basic mechanisms are used by classic intracellular parasites for internalization into mammalian cells: microfilament-dependent phagocytosis and receptor-mediated endocytosis (37). Studies were done with CD, which inhibits microfilament formation and thus blocks phagocytosis (16), and MDC, which inhibits receptor-mediated endocytosis (49). Macrophage monolayers were preincubated with the inhibitors for 1 h before and during the infection assays. The levels of inhibitors used did not decrease the viability of either the macrophages or B. pertussis BP338 (data not shown). CD at 1.0 and 2.5 ,ug/ml inhibited B. pertussis uptake into macrophage to 8 and 1.7% of control BP338 levels (P < 0.01) (Fig. 3). MDC treatment had no significant effect on levels of uptake at either 25 or 50 ,uM. These results demonstrate that B. pertussis uptake into macrophages occurs primarily via a microfilament-requiring process, i.e., by phagocytosis and not by receptor-mediated endocytosis. Uptake and survival in macrophages of mutant B. pertussis. Studies using various mutants of B. pertussis were performed to investigate the role of B. pertussis virulence factors in the microbe's ability to enter and to survive inside macrophages. Table 2 shows the levels of internalization of B. pertussis mutants into macrophage monolayers compared with values for the parental strain BP338. Avirulent mutants BP537 (spontaneous vir mutant) and BP347 (TnS vir mutant) were able to enter macrophages but only at 0.5 and 2.5%, respectively, of parental levels. The AC mutant BP348 (TnS AC- mutant) was able to enter macrophages but only to 38% of parental levels. BC67 is an AC conditional expression mutant which contains TnStacl inserted in the cyaE gene of the AC operon, which regulates AC secretion (10). This transposon contains a strong outward-facing promoter (Ptac) and a lacI repressor gene which represses this promoter. In the presence of the inducer IPTG, the promoter is derepressed and

SURVIVAL OF B. PERTUSSIS IN MACROPHAGES

VOL. 60, 1992

4581

A

C

P~~~~~~~~P

FIG. 2. Electron micrographs demonstrating intracellular B. pertussis within macrophages on day 0 (A and B) and day 3 (C and D) after infection. (A) After initial uptake (day 0), several B. pertussis BP338 microbes (arrowheads) are present in tight-fitting vacuoles in the macrophage cytoplasm. Magnification, x 10,080. (B) Higher magnification of an area in panel A (note that the orientation has been changed) showing the membrane-bound vacuoles containing B. pertussis (arrowheads). Magnification, X40,320. (C) Macrophages containing intracellular B. pertussis 3 days after infection (arrowheads). Magnification, X7,837. (D) Higher magnification of an area of panel C with B. pertussis-containing vacuoles (arrowheads). Magnification, X34,381.

transcription of adjacent genes can occur. Therefore, TnStacl produces insertion mutants with conditional phenotypes regulated by IPTG. In the absence of IPTG, BC67 does not express or secrete AC; in the presence of IPTG, BC67 expresses and secretes wild-type levels of AC (9, 10). BC67 was grown either with (BC67 + IPTG) or without (BC67 IPTG) IPTG, and the ability of BC67 to enter macrophages was determined. In these experiments, monolayers were cultured either with or without the addition of 20 mM IPTG to maintain expression of AC by BC67 when required. BC67 + IPTG entered macrophages at levels not statistically different from those of virulent BP338, while uptake of BC67 - IPTG dropped to levels similar to those of BP348 (AC-)

(Table 2). Uptake of BC67 + IPTG was significantly greater than uptake of BC67 - IPTG and BP348 (P < 0.05). Strains BP101 (AFHA, with a 2.4-kb deletion in the FHA gene), BP-TOX6 (PT-, with a 4.7-kb deletion in the PT gene), and BP101-TOX6 (AFHA PT-) were used. BP101, BP-TOX6, and BP101-TOX6 had decreased abilities to enter macrophages (Table 2). The internalization levels for BP101, BP-TOX6, and BP101-TOX6 were 7, 35, and 14%, respectively, of parental levels. The level of BP338 internalization was significantly higher (P < 0.05) than those of all mutants except BC67 + IPTG. Figure 4 shows the ability of the various B. pertussis mutants to survive intracellularly in macrophages over 3

4582

FRIEDMAN ET AL. M FRPMI

160 -

CJ

RPMI

INFECT. IMMUN.

+1.0 mg/ml CD _ +2.5 +25 pM MDC lIEU +50

mg/ml CD

gU

107 _

MM MDC

10

cc 120-

Day One Day Two Day Three

0

80

OD

0 U.

40

103

=3_

CD

-

MDC

Treatment

FIG. 3. Effects of CD and MDC on B. pernussis entry into

338 Vir+

537 Vir-

347 Vir-

348 167- 67+ 101 TOX6 101TOX6 AC- I-AC- JI FHA- PT- FHA/PT-

macrophages. Monolayers were incubated with CD (at 1.0 and 2.5 jg/ml) and MDC (at 25 and 50 p,M) both before and during infection of the phagocytes; a 100:1 infection ratio of BP338 was used. After infection, monolayers were washed, macrophages were lysed, and viable plate counts were determined. Results are presented as percentages of control levels of entry (RPMI control equals 100%) compared with levels of entry in inhibitor-treated phagocytes. Each assay was performed at least six times in duplicate.

FIG. 4. Intracellular survival of virulent and mutant B. pertussis in macrophages. Monolayers were infected with virulent and mutant strains at an infection ratio of 100:1 as described in the text. On days 1, 2, and 3, monolayers were lysed and viable plate counts (CFU per monolayer) were determined. BC67 was grown with (67+) or without (67-) IPTG. Results presented are the means ± standard deviations of at least six experiments done in duplicate.

days. Viable counts dropped by greater than 3 log units over the 3 days of the study with the vir strains. By 3 days, no viable BP537 was detected and only 10 CFU of BP347 per monolayer was detected. AC mutants BP348 and BC67 IPTG persisted better than the other B. pertussis mutants tested, while BC67 + IPTG survived at levels comparable to those of BP338. The PT and FHA mutants were found to have decreased abilities to survive intracellularly in macrophages, with viable counts dropping over 3 days (Fig. 4). On days 1, 2, and 3 of the studies, levels of viable intracellular BP338 (at a 100:1 infection ratio) were significantly higher than levels for all mutants studied except BC67 + IPTG (P < 0.05).

macrophage monolayers (Fig. 2). The level of uptake of B. pertussis at the 100:1 infection ratio (12.4% of the infectious dose was internalized) was comparable to levels reported for such invasive microbes as Yersinia pseudotuberculosis, Salmonella typhimurium, and Shigella flexneri (16, 51). Ewanowich et al. reported comparable levels of invasion by B. pertussis versus S. flemneri and Salmonella hadar in their HeLa cell model (14). This is the first report of long-term survival of B. pertussis in human macrophages in vitro. Saukkonen et al. reported survival in human macrophages in short-term culture for a period of 6 h (47). Studies utilizing HeLa cells as an internalization model have been done for 7 h and up to 9 days (12, 14, 27). The direct entry into macrophages, without opsonization by complement or antibody, is in contrast to the strict requirement for opsonization by antipertussis antibody to induce uptake (phagocytosis) of B. pertussis by human PMN (53, 54). Without antibody opsonization, no internalization of the bacteria into PMN occurs. While virulent B. pertussis survived inside macrophages after uptake at the 100:1 and 400:1 infection ratios, no intracellular multiplication was observed (Fig. 1 and 2). At lower infection ratios of 10:1 and 50:1, B. pertussis was able to be internalized, but numbers of viable intracellular bacteria dropped significantly over the 3 days of the studies, suggesting that killing occurred. These results suggest that for B. pertussis to persist inside macrophages, a critical number of virulent microbes must be present intracellularly. This requirement is also reflected in the results of experiments using mutant strains. Since uptake levels are so low for the vir mutants and the other mutant strains, the intracellular number of bacteria required for survival cannot be attained. It is possible that if avirulent mutants could attain a critical intracellular number, they would be able to survive. Another possibility is that this critical intracellular bacterial number is required so that levels of exotoxins (AC, PT, or others factors) high enough to inhibit normal macrophage bacterial killing or metabolic processes are produced. PT and AC have been demonstrated to have inhibitory effects on phagocytes. Investigators have shown that human PMN chemiluminescence and chemotaxis, as well as macrophage bactericidal activity, are inhibited by AC (8, 19, 20, 30). PT also has been reported to inhibit chemotaxis, lysosomal enzyme release, and superoxide production in PMN (2, 52).

DISCUSSION

The results of this study demonstrate that B. pertussis has the ability to enter, survive, and persist intracellularly in human macrophages. Virulent B. pertussis was able to enter macrophages, in the absence of serum, via a microfilamentdependent phagocytic process which was inhibited by CD (Fig. 1 and 3). Intracellular survival and persistence of B. pertussis over the 3 days of the studies were confirmed by viable plate counts and by EM examination of infected TABLE 2. Levels of uptake by macrophages of virulent and

mutant B. pertussis strainsa

Strain

BP338 ............................................... BP537 ........................................ BP347 ........................................ BP348 ........................................ BC67 - IPTG ........................................ BC67 + IPTG ........................................

BP101 ........................................ BP-TOX6 ........................................

BP101-TOX6 ........................................

% Uptake b 100 ± 21.8 0.5 ± 0.5 2.5 ± 2.4 38 ± 21 26.1 ± 1.0 94.9 ± 15 7 +6.7 35 ± 25 14 ± 17.4

a Monolayers were infected with virulent strain BP338 and various B. pertussis mutants at a 100:1 infection ratio. After infection (2 h), monolayers were incubated for 1 h with 15% NS to kill extracellular bacteria and washed, macrophages were lysed, and viable plate counts were determined. b Percent uptake of B. pertussis strains compared with levels of BP338 uptake (100%) standard deviation. Each experiment was performed at least six times in duplicate.

4583

VOL. 60, 1992

SURVIVAL OF B. PERTUSSIS IN MACROPHAGES

While virulent B. pertussis was able to enter and persist inside macrophages, vir mutants had significantly decreased abilities to be internalized (on the average, 1.5% of parental levels) and either did not survive intracellularly (BP537) or survived at only 10 CFU per monolayer (BP347) (Table 2 and Fig. 4). These results demonstrate that the internalization process, as well as intracellular survival, is vir dependent. Uptake and intracellular survival in the HeLa cell model has also been reported to be vir dependent (14, 27). The products of the vir, or bvgAS, locus coordinately regulate expression of most virulence factors of B. pertussis (13, 23, 26, 35, 44, 48, 56). The vir locus is a member of a family of twocomponent bacterial sensory transduction systems which senses environmental signals and responds via regulation of virulence gene expression (10, 30, 31, 51). The AC mutants used in this study were found to have significantly decreased abilities to enter and to survive intracellularly in macrophages (Table 2 and Fig. 4). In experiments using the AC conditional expression mutant BC67, in the presence of IPTG, AC was expressed and the mutant was able to enter and persist in monolayers at parental levels. Without IPTG, AC was not expressed and levels of BC67 internalization and persistence dropped to those obtained with the AC- mutant BP348. These observations may suggest that AC is important in the uptake process and in intracellular survival. The role of AC in the uptake into HeLa cells is a point of controversy. Lee et al. reported that AC mutants entered HeLa cells as well did wild-type strains (27), while Ewanowich et al. reported up to a twofold increase in internalization levels with AC mutants (14). Purified AC has been demonstrated to induce increases in intracellular cyclic AMP (cAMP) levels in phagocytes (8, 19, 20), and increased levels of intracellular cAMP have been shown to impede phagosome-lysosome fusion in PMN and in macrophages (31, 61). Thus AC, by increasing intracellular levels of cAMP in macrophages, may block phagosomelysosome fusion, thereby allowing intracellular survival. In studies of PMN-B. pertussis interactions, intracellular B. pertussis inhibited phagosome-lysosome fusion and the bacteria were not killed (54). Results of experiments using PT and FHA mutants suggest the requirement of these factors for optimal uptake and survival of B. pertussis in macrophages (Table 2 and Fig. 4). Both FHA and PT mutants had significantly decreased abilities to be internalized by macrophages. This result is not surprising, since studies by other investigators have demonstrated that B. pertussis adheres to macrophages and other mammalian cell types via FHA and PT (41, 42, 47, 57, 58). PT can function as an adhesion, by interaction with certain carbohydrates on cell surfaces (47). FHA interacts with macrophages via two mechanisms: by binding to galactosecontaining glycoconjugates on the phagocyte and by the interaction of an Arg-Gly-Asp (RGD) sequence of FHA with the macrophage integrin complement receptor 3 (CR3) (41, 42, 47). FHA binding to CR3 of macrophages induces internalization of the microorganism into the phagocyte (47). These observations are intriguing because many intracellular pathogens interact with macrophage CR3 to induce phagocytosis. CR3 has been found to mediate uptake of Legionella pneumophila, Mycobacterium leprae, Histoplasma capsulatum, Leishmania donovani, and Y. pseudotuberculosis into mammalian cells (5, 25, 39, 50, 63). Lower levels (CFU per monolayer) of the double mutant BP101-TOX6 than of the single FHA and PT mutants were recovered by day 3, suggesting a synergistic effect of the loss

of both PT and FHA expression (Fig. 4). The role that FHA plays in intracellular survival is not known. In mammalian cells, PT, as demonstrated for AC, can induce increases in intracellular cAMP levels which could impede phagosomelysosome fusion and other phagocyte functions (8, 19, 20, 31, 61). While FHA mutants, PT mutants, or double mutants had significantly decreased abilities to enter and survive in macrophages, the mutants could still localize within phagocytes, suggesting that other ligands can induce entry of B. pertussis. One such ligand may be the recently identified B. pertussis surface protein pertactin (6, 29). This outer membrane protein has been cloned and sequenced and found to contain an RGD sequence that promotes adherence of B. pertussis to tissue culture cells (6, 28, 29). Studies using HEp-2 cells demonstrated that pertactin plays a role in the adherence and entry process (43). Recent studies by Leininger et al. found that the RGD sequence of pertactin mediated uptake into HeLa cells, while the RGD sequence of FHA had no effect (28). The role of pertactin in attachment and entry into macrophages needs further investigation. The results of these studies suggest that B. pertussis has multiple methods for uptake into mammalian cells. These multiple bacterial ligands which induce entry may act synergistically, or they may allow the microbe to enter various target cell types, depending on the presence of appropriate receptors. The reported ability of viable B. pertussis to enter and survive in macrophages (18, 47), PMN (53, 54), and nonphagocytic cells (12, 14, 15, 27) and the ability of its exoproducts to suppress various macrophage functions strongly suggest that B. pertussis is invasive and that host macrophage responses may play an important role in the defense against pertussis. The macrophage response (cellmediated immunity) plays a critical role in the control of infections caused by classical intracellular pathogens such as Listeria monocytogenes, L. pneumophila, Mycobacteriun tuberculosis, and others (37). In support of the potential importance of cell-mediated immunity in whooping cough are the reported cases of both adults and children with human immunodeficiency virus infection who developed pertussis infections (1, 4, 38). Most intriguing is the paper of Bromberg et al., which reported the presence of intracellular B. pertussis within pulmonary alveolar macrophage recovered from bronchoalveolar lavage specimens from pediatric patients infected with human immunodeficiency virus (4). This is the first direct evidence that in vivo intracellular survival of B. pertussis may occur in human macrophages. The potential role that macrophages or cell-mediated immunity may play in pertussis immunity is a vital area for further investigation. ACKNOWLEDGMENTS We thank Marilee Sellers for expert technical assistance with the EM studies. We thank S. Falkow, D. Relman, and W. Goldman for providing the bacterial strains used in this study. Special thanks are extended to M. Peppler and W. Goldman for critical reviews of the manuscript. This work was supported by a grant from the Arizona Disease Control Research Commission to R.L.F. and D.E.Y. REFERENCES 1. Adamson, P. C., T. C. Wu, B. D. Meade, M. R. Rubin, C. R.

Manclark, and P. A. Pizzo. 1989. Pertussis in a previously immunized child with human immunodeficiency virus infection. J. Pediatr. 115:589-591.

4584

FRIEDMAN ET AL.

2. Becker, E. L., J. C. Kermode, P. H. Naccache, R. Yassin, J. J. Munoz, M. L. Marsh, C. K. Huang, and R. I. Sha'afi. 1986. Pertussis toxin as a probe of neutrophil activation. Fed. Proc.

45:2151-2155. 3. Boyum, A. 1968. Isolation of mononuclear cells and granulocytes from human blood. Scand. J. Clin. Lab. Invest. 97:7789. 4. Bromberg, K., G. Tannis, and P. Steiner. 1991. Detection of Bordetella pertussis associated with the alveolar macrophages of children with human immunodeficiency virus infection. Infect. Immun. 59:4715-4719. 5. Bullock, W. E., and S. D. Wright. 1987. Role of the adherencepromoting receptors, CR3, LFA-1, and p150,95, in binding of Histoplasma capsulatum by human macrophages. 1987. J. Exp. Med. 165:195-210. 6. Charles, I. G., G. Dougan, D. Pickard, S. Chatfield, M. Smith, P. Novotny, P. Morrisey, and N. F. Fairweather. 1989. Molecular cloning and characterization of protective outer membrane protein P. 69 from Bordetella pertussis. Proc. Natl. Acad. Sci. USA 86:3554-3558. 7. Cheers, C., and D. F. Gray. 1969. Macrophage behavior during the complaisant phase of murine pertussis. Immunology 17:875887. 8. Confer, D. L., and J. W. Eaton. 1982. Phagocyte impotence caused by an invasive bacterial adenylate cyclase. Science 217:948-950. 9. Cookson, B. T., D. E. Berg, and W. E. Goldman. 1990. Mutagenesis of Bordetella pertussis with transposon TnStacl: conditional expression of virulence-associated genes. J. Bacteriol. 172:1681-1687. 10. Cookson, B. T., T. Tomcsanyi, D. E. Berg, and W. E. Goldman. 1990. A strategy for studying pathogenic mechanisms: transposon insertion mutations with conditional phenotypes, p. 243250. In C. R. Manclark (ed.), Proceedings of the 6th International Symposium on Pertussis. DHHS publication 90-1164. U.S. Public Health Service, Bethesda, Md. 11. Coote, J. G. 1991. Antigenic switching and pathogenicity: environmental effects on virulence gene expression in Bordetella pertussis. J. Gen. Microbiol. 137:2493-2503. 12. Crawford, J. G., and C. W. Fishel. 1959. Growth of Bordetella pertussis in tissue culture. J. Bacteriol. 77:465-474. 13. Domenighini, M., D. Relman, C. Capiau, S. Falkow, A. Prugnola, V. Scarlato, and R. Rappuoli. 1990. Genetic characterization of Bordetella pertussis filamentous hemagglutinin: a protein processed from an unusually large precursor. Mol. Microbiol. 4:787-800. 14. Ewanowich, C. A., A. R. Melton, A. A. Weiss, R. K. Sherburne, and M. S. Peppler. 1989. Invasion of HeLa 229 cells by virulent Bordetella pertussis. Infect. Immun. 57:2698-2704. 15. Ewanowich, C. A., R. K. Sherburne, S. F. P. Man, and M. S. Peppler. 1989. Bordetella parapertussis invasion of HeLa 229 cells and human respiratory epithelial cells in primary culture. Infect. Immun. 57:1240-1247. 16. Finlay, B. B., and S. Falkow. 1988. Comparison of the invasion strategies used by Salmonella cholera-suis, Shigella fleeri, and Yersinia enterocolitica to enter cultured animal cells: endosome acidification is not required for bacterial invasion or intracellular replication. Biochimie 70:1089-1099. 17. Friedman, R. L. 1988. Pertussis: the disease and new diagnostic methods. Clin. Microbiol. Rev. 1:365-376. 18. Friedman, R. L., and P. Z. Detsky. 1989. Multiplication of Bordetella pertussis in human monocytes, abstr. D-128, p. 103. Abstr. 89th Gen. Meet. Am. Soc. Microbiol. 1989. American Society for Microbiology, Washington, D.C. 19. Friedman, R. L., R. L. Fiederlein, L. Glasser, and J. N. Galgiani. 1987. Bordetella pertussis adenylate cyclase: effects of affinitypurified adenylate cyclase on human polymorphonuclear leukocyte functions. Infect. Immun. 55:135-140. 20. Galgiani, J. N., E. L. Hewlett, and R. L. Friedman. 1988. Effects of adenylate cyclase toxin from Bordetella pertussis on human neutrophil interactions with Coccidioides immitis and Staphylococcus aureus. Infect. Immun. 56:751-755. 21. Goldman, W. E., D. G. Klapper, and J. B. Baseman. 1982.

INFECT. IMMUN.

Detection, isolation, and analysis of a related Bordetella pertussis product toxic to cultured tracheal cells. Infect. Immun. 36:782-794. 22. Gray, D. F., and C. Cheers. 1967. The steady state in cellular immunity. II. Immunological complaisance in murine pertussis. Aust. J. Exp. Biol. Med. Sci. 45:417-426. 23. Gross, R., B. Arico, and R. Rappuoli. 1989. Genetics of pertussis toxin. Mol. Microbiol. 3:119-124. 24. Hand, W. L., and N. L. King-Thompson. 1986. Contrasts between phagocyte antibiotic uptake and subsequent intracellular bactericidal activity. Antimicrob. Agents Chemother. 29: 135-140. 25. Isberg, R. R., and J. M. Leong. 1990. Multiple 31 chain integrins are receptors for invasin, a protein that promotes bacterial penetration into mammalian cells. Cell 60:861-871. 26. Laoide, B., and A. Ullmann. 1990. Virulence dependent and independent regulation of the Bordetella pertussis cya operon. EMBO J. 9:999-1005. 27. Lee, C. K., A. L. Roberts, T. M. Finn, S. Knapp, and J. J. Mekalanos. 1990. A new assay for invasion of HeLa 229 cells by Bordetella pertussis: effect of inhibitors, phenotypic modulation, and genetic alterations. Infect. Immun. 58:2516-2522. 28. Leininger, E., C. A. Ewanowich, A. Bhargava, M. S. Peppler, J. G. Kenimer, and M. J. Brennan. 1992. Comparative roles of the Arg-Gly-Asp sequence present in the Bordetella pertussis adhesins pertactin and filamentous hemagglutinin. Infect. Immun. 60:2380-2385. 29. Leininger, E., M. Roberts, J. G. Kenimer, I. G. Charles, N. Fairweather, P. Novotny, and M. J. Brennan. 1991. Pertactin, an Arg-Gly-Asp-containing Bordetella pertussis surface protein that promotes adherence of mammalian cells. Proc. Natl. Acad. Sci. USA 88:345-349. 30. Leusch, M. S., S. Paulaitis, and R. L. Friedman. 1990. Adenylate cyclase toxin of Bordetella pertussis: production, purification, and partial characterization. Infect. Immun. 58:3621-3626. 31. Lowrie, D. B., V. R. Moer, and P. S. Jackett. 1979. Phagosomelysosome fusion and cyclic adenosine 3'-5'monophosphate in macrophages infected with Mycobacterium microti, Mycobacterium bovis BCG, or Mycobacterium lepraemurium. J. Gen. Microbiol. 110:431-441. 32. Manclark, C. R., and J. L. Cowell. 1984. Pertussis, p. 60-106. In R. Germanier (ed.), Bacterial vaccines. Academic Press, Inc., Orlando, Fla. 33. Mekalanos, J. J. 1992. Environmental signals controlling expression of virulence determinates in bacteria. J. Bacteriol. 174:1-7. 34. Miller, J. F., J. J. Mekalanos, and S. Falkow. 1989. Coordinate regulation and sensory transduction in the control of bacterial virulence. Science 243:916-922. 35. Miller, J. F., C. R. Roy, and S. Falkow. 1989. Analysis of Bordetella pertussis virulence and gene regulation by use of transcriptional fusions in Escherichia coli. J. Bacteriol. 171: 6345-6348. 36. Mouallem, M., Z. Farfel, and E. Hanski. 1990. Bordetella pertussis adenylate cyclase toxin: intoxication of host cells by bacterial invasion. Infect. Immun. 58:3759-3764. 37. Moulder, J. W. 1985. Comparative biology of intracellular parasitism. Microbiol. Rev. 49:298-337. 38. Ng, V. L., M. York, and W. K. Hadley. 1989. Unexpected isolation of Bordetella pertussis from patients with acquired immunodeficiency syndrome. J. Clin. Microbiol. 27:337-338. 39. Payne, N. R., and M. A. Horwitz. 1987. Phagocytosis of Legionella pneumophila is mediated by human monocyte complement receptors. J. Exp. Med. 166:1377-1389. 40. Pittman, M. 1984. The concept of pertussis as a toxin-mediated disease. Pediatr. Infect. Dis. 3:467-485. 41. Relman, D. A., M. Domenighini, E. Tuomanen, R. Rappuoli, and S. Falkow. 1989. Filamentous hemagglutinin of Bordetella pertussis: nucleotide sequence and crucial role in adherence. Proc. Natl. Acad. Sci. USA 86:2637-2641. 42. Relman, D. A., E. Tuomanen, S. Falkow, D. T. Golenbock, K. Saukkonen, and S. D. Wright. 1990. Recognition of a bacterial adhesion by an integrin: macrophage CR3 (aMI32, CD11b/CD18)

VOL. 60, 1992

SURVIVAL OF B. PERTUSSIS IN MACROPHAGES

binds filamentous hemagglutinin of Bordetella pertussis. Cell 61:1375-1382. 43. Roberts, M., N. F. Fairweather, E. Leininger, D. Pickard, E. L. Hewlett, A. Robinson, C. Hayward, G. Dougan, and I. G. Charles. 1991. Construction and characterization of Bordetella pertussis mutants lacking the vir-regulated P.69 outer membrane protein. Mol. Microbiol. 5:1393-1404. 44. Roy, C. R., and S. Falkow. 1991. Identification of Bordetella pertussis regulatory sequences required for transcriptional activation of thefhaB gene and autoregulation of the bvgAS operon. J. Bacteriol. 173:2385-2392. 45. Sato, Y., J. L. Cowell, H. Sato, and C. R. Manclark. 1983. Separation and purification of the hemagglutinin from Bordetella pertussis. Infect. Immun. 41:313-320. 46. Sato, Y., K. Izuiya, H. Sato, J. L. Cowell, and C. R. Manclark. 1981. Role of antibody to leukocytosis-promoting factor hemagglutinin and to filamentous hemagglutinin in immunity to pertussis. Infect. Immun. 31:1223-1231. 47. Saukkonen, K., C. Cabellos, M. Burroughs, S. Prasad, and E. Tuomanen. 1991. Integrin-mediated localization of Bordetella pertussis within macrophages: role in pulmonary colonization. J. Exp. Med. 173:1143-1149. 48. Scarlato, V., A. Pmgnola, B. Arico, and R. Rappuoli. 1990. Positive transcriptional feedback at the bvg locus controls expression of virulence factors in Bordetella pertussis. Proc. Natl. Acad. Sci. USA 87:6753-6757. 49. Schlegel, R, R B. Dickson, M. C. Willingham, and I. H. Pastan. 1982. Amantadine and dansylcadaverine inhibit vesicular stomatitis virus uptake and receptor-mediated endocytosis of a2macroglobulin. Proc. Natl. Acad. Sci. USA 79:2291-2295. 50. Schlensinger, L. S., and M. A. Horwitz. 1990. Phagocytosis of leprosy bacilli is mediated by complement receptors CR1 and CR3 on human monocytes and complement component C3 in serum. J. Clin. Invest. 85:1304-1314. 51. Small, P. L. C., R. R. Isberg, and S. Falkow. 1987. Comparison of the ability of enteroinvasive Escherichia coli, Salmonella typhimurium, Yersiniapseudotuberculosis, and Yersinia enterocolitica to enter and replicate within HEp-2 cells. Infect. Immun. 55:1674-1679. 52. Spangrude, G. J., F. Sacchi, H. R. Hill, D. E. Van Epps, and R. A. Daynes. 1985. Inhibition of lymphocyte and neutrophil chemotaxis by pertussis toxin. J. Immunol. 135:4135-4143. 53. Steed, L. L., E. T. Akporiaye, and R. L. Friedman. 1992.

Bordetella pertussis induces respiratory burst activity in human polymorphonuclear leukocytes. Infect. Immun. 60:2101-2105. 54. Steed, L. L., M. Setareh, and R. L. Friedman. 1991. Hostparasite interactions between Bordetella pertussis and human polymorphonuclear leukocytes. J. Leukocyte Biol. 50:321-330. 55. Steven, A. C., M. E. Baker, B. C. Trus, D. Thomas, J. M. Zhang, and J. L. Cowell. 1986. Helical structure of Bordetella petussis flimbriae. J. Bacteriol. 167:968-974. 56. Stibitz, S., A. A. Weiss, and S. Falkow. 1988. Genetic analysis of a region of the Bordetella pertussis chromosome encoding filamentous hemagglutinin and the pleiotropic regulatory locus vir. J. Bacteriol. 17p:2904-2913. 57. Tuomanen, E., and A. A. Weiss. 1985. Characterization of two adhesions of Bordetella pertussis for human ciliated respiratory epithelial cells. J. Infect. Dis. 152:118-125. 58. Urisa, A., J. L. Cowell, and C. R. Manclark. 1986. Filamentous hemagglutinin has a major role in mediating adherence of Bordetella pertussis to human WiDr cells. Infect. Immun. 52:695-701. 59. Weiss, A. A., and E. L. Hewlett. 1986. Virulence factors of Bordetella pertussis. Annu. Rev. Microbiol. 40:661-686. 60. Weiss, A. A., E. L. Hewlett, G. A. Meyers, and S. Falkow. 1983. Tn5-induced mutations affecting virulence factors of Bordetella pertussis. Infect. Immun. 42:33-41. 61. Weissman, G., I. Goldman, S. Hoffstein, G. Chauvet, and R. Robineaux. 1975 Yin/yang modulation of lysosomal enzyme release from polymorphonuclear leukocytes by cyclic nucleotides. Ann. N.Y. Acad. Sci. 256:222-231. 62. Willems, R., A. Paul, H. G. van der Heide, A. R. ter Avest, and F. R. Mooi. 1990. Fimbrial phase variation in Bordetella pertussis: a novel mechanism for ttanscriptional regulation. EMBO J. 9:2803-2809. 63. Wilson, M. E., and R. D. Pearson. 1988. Roles of CR3 and mannose receptors in the attachment and ingestion of Leishmania donovani by human mononuclear phagocytes. Infect. Immun. 56:363-369. 64. Woods, D. E., R. Franklin, S. J. Cryz, Jr., M. Ganss, M. Peppler, and C. Ewanowich. 1989. Development of a rat model for respiratory infection with Bordetella pertussis. Infect. Immun. 57:1018-1024. 65. Zhang, Y. L., and R. D. Sekura. 1991. Purification and characterization of heat-labile toxin of Bordetella pertussis. Infect. Immun. 59:3754-3759.

4585