Pertactin contributes to shedding and transmission of Bordetella bronchiseptica.

Whooping cough is resurging in the United States despite high vaccine coverage. The rapid rise of Bordetella pertussis isolates lacking pertactin (PRN), a key vaccine antigen, has led to concerns about vaccine-driven evolution. Previous studies showed that pertactin can mediate binding to mammalian cells in vitro and act as an immunomodulatory factor in resisting neutrophil-mediated clearance. To further investigate the role of PRN in vivo, we examined the functions of pertactin in the context of a more naturally low dose inoculation experimental system using C3H/HeJ mice that is more sensitive to effects on colonization, growth and spread within the respiratory tract, as well as an experimental approach to measure shedding and transmission between hosts. A B. bronchiseptica pertactin deletion mutant was found to behave similarly to its wild-type (WT) parental strain in colonization of the nasal cavity, trachea, and lungs of mice. However, the pertactin-deficient strain was shed from the nares of mice in much lower numbers, resulting in a significantly lower rate of transmission between hosts. Histological examination of respiratory epithelia revealed that pertactin-deficient bacteria induced substantially less inflammation and mucus accumulation than the WT strain and in vitro assays verified the effect of PRN on the induction of TNF-α by murine macrophages. Interestingly, only WT B. bronchiseptica could be recovered from the spleen of infected mice and were further observed to be intracellular among isolated splenocytes, indicating that pertactin contributes to systemic dissemination involving intracellular survival. These results suggest that pertactin can mediate interactions with immune cells and augments inflammation that contributes to bacterial shedding and transmission between hosts. Understanding the relative contributions of various factors to inflammation, mucus production, shedding and transmission will guide novel strategies to interfere with the reemergence of pertussis.

Introduction

Pertussis, or whooping cough, is an acute respiratory disease caused by the gram-negative bacterial pathogen Bordetella pertussis. Historically, pertussis had a high morbidity and mortality rate and was the predominant childhood killer before the introduction of whole cell vaccines (WCV) that greatly reduced its prevalence [1]. In highly vaccinated populations, the reduced incidence of severe disease led to attention being focused on the moderate side effects associated with WCV, including common local reactions (swelling and pain), uncommon systemic reactions (fever, irritability, drowsiness, loss of appetite and vomiting), and very rare neurological reactions (acute encephalopathy in newborns) [2]. To address these concerns, less reactive pertussis acellular vaccines (ACV) containing up to five purified, detoxified B. pertussis proteins, including pertussis toxoid, pertactin, type 2 and 3 fimbriae, and filamentous hemagglutinin, were increasingly used in the 1980s and 1990s, and eventually replaced the WCV in most developed countries [3]. Coincident with the switch to ACV, the incidence of whooping cough has been increasing in all age groups, from infants to adults [4]. Moreover, in recent years several countries have seen an increased percentage of clinical isolates that fail to express pertactin (PRN), a prominent outer membrane protein [5,6]. The loss of PRN, a prominent ACV component, has led to concern that ACV vaccines are driving the evolution of B. pertussis to evade vaccine-induced immunity [7]. The long-term consequences of this possibility are difficult to predict as we have limited knowledge of the biological functions of PRN.

PRN is an autotransporter protein that appears on the outer membrane of B. pertussis [8]. This protein has been included in acellular vaccines in several countries, based on evidence that pertactin specific antibodies contribute to protection against pertussis. Previous studies revealed that PRN exerts cell binding through its RGD motif and contributes to adherence and invasion of mammalian cells [9-12]. However, the adhesion functions of pertactin were not detected in some other studies [13-17]. In vivo studies have shown an immunomodulatory role of PRN [17,18], in the context of severe pneumonic infection initiated by super-natural challenge with roughly a million CFU of B. pertussis or B. bronchiseptica delivered deep into the lungs of mice where they caused severe lung pathology. This experimental system was established to study the most extreme form of disease and has enabled the development and testing of vaccines and therapeutics to prevent and treat severe disease in the lower respiratory tract. But B. pertussis is highly infectious, suggesting very small initial inocula, and natural infections begin with the weeks-long catarrhal stage, which is limited to the nasopharyngeal region and is a highly contagious period. Thus, colonization with low initial numbers followed by weeks of growth within and shedding from the upper respiratory tract are critical aspects of the infectious cycle of B. pertussis. However, these aspects are not simulated in the standard inoculation approach that delivers as many as a million bacteria deep into the lungs and measures the consequences of the subsequent extreme pneumonic pathology [19-21].

We have focused on simulating the progression of natural infections in order to study the contributions of individual bacterial factors and host immune functions [22-25]. We use B. bronchiseptica because it naturally and highly efficiently colonizes, grows, causes pathology, is shed, and transmits between mice. Importantly, it can efficiently colonize mice when delivered to the external nares in very small numbers (<5 CFU), allowing the entire natural progression of the infection to be studied, including many aspects that are obviated by the conventional approach of washing millions of bacteria deep into the lungs [22-25]. B. pertussis and B. bronchiseptica are very closely related and shared genes are ~ 98% identical at the nucleotide level, but differ in some notable regards. While most of the best studied factors are shared, each expresses a somewhat different subset of factors [5,26-31], that are likely to explain their differences in host range and/or the diseases they cause: while B. pertussis cause an acute and severe disease in humans, B. bronchiseptica causes infection that can range in severity and often persists in the upper respiratory tract of infected animals. However, similarities far outbalance differences. Both species infect mammals and, with the notable exception of expression of the pertussis toxin, share nearly all genes implicated in interacting with their hosts, including the vaccine antigens fimbriae (FIM), filamentous hemagglutinin (FHA) and both species express PRN orthologs that are 92% similarity of PRN at the amino acid level and are likely to perform similar functions [31].

The study of the true in vivo functions of any of Bordetella spp. factors requires experimental systems in which their effects can be measured. Unfortunately, B. pertussis poorly colonizes mice, making studies of the details of natural infections challenging. Unlike B. pertussis, B. bronchiseptica efficiently colonizes mice with an infectious dose less than 5 CFU with infections progressing efficiently, allowing the study of the mechanistic basis for their complex interactions in the context of naturally progressing infection in the natural host [22-25]. We also previously described B. bronchiseptica efficiently transmitting among TLR4-deficient C3H/HeJ mice. B. bronchiseptica grows and is shed in higher numbers in these mice, allowing more efficient transmission [22-25]. This experimental model allows all the bacterial components necessary for efficient transmission to be experimentally manipulated to better understand bacterial mechanisms involved in the transmission process.

In this study, to investigate the biological role(s) of PRN in the infectious process and pathogenicity of Bordetella species, we used a mutant of B. bronchiseptica with an in-frame deletion in the prn gene which was generated by allelic exchange as described by Inatsuka et al. [17]. In vivo experiments showed that prn was not necessary for efficient colonization or early growth in the nasopharynx, but was required for efficient bacterial shedding and transmission between mice. Furthermore, wild-type B. bronchiseptica induced higher levels of inflammation and more mucus secretion than the Δprn mutant, revealing a role for PRN in promoting inflammation and mucus secretion, which was supported by in vitro assays showing higher secretion levels of the pro-inflammatory cytokine TNF-α from WT infected compared to Δprn mutant infected murine macrophages. Moreover, the WT strain, but not the prn deletion strain, was recovered from splenocytes of infected mice, indicating that PRN contributes to to systemic dissemination that involves intracellular survival. Together these data suggest that PRN promotes the induction of inflammation and mucus production, mediates shedding and transmission to new hosts and is involved in the intracellular survival and systemic dissemination of the pathogen.

Results

PRN is not necessary for B. bronchiseptica to efficiently colonize the host and grow within the respiratory tract

To evaluate the contribution of pertactin to various aspects of the biology of B. bronchiseptica, we used an isogenic prn gene deficient mutant (BbΔprn) of B. bronchiseptica [17]. The parental strain RB50 (Bb WT) is well-established as being highly efficient in colonizing, persisting, and transmitting among mice [22-25]. To confirm the deletion was clean and not complicated by other changes, we re-sequenced the whole genome and confirmed the in-frame deletion of prn (S1 Fig). The prn mutant strain showed no defect in in vitro growth, adherence to human alveolar epithelial cells or cytotoxicity to murine RAW 264.7 macrophages (S2–S4 Figs). Using the conventional pneumonic infection model, C57BL/6 mice were inoculated intranasally with 50 μL PBS buffer containing 5x105 CFU of either WT or BbΔprn bacteria. Bacterial numbers in nasal cavities, trachea and lungs harvested at 7-, 14- and 28-days post-inoculation (dpi) did not differ significantly between mice infected with Bb WT or BbΔprn, (S5 Fig). Interestingly, BbΔprn showed higher colonization levels in the nasal cavity and lungs at 14 dpi, but these differences were not observed in later timepoints. Histopathological analysis of nasal cavities at 7 dpi and 14 dpi detected mild suppurative inflammation in both groups with similar incidence and severity. Thus, the standard high dose pneumonic infection experimental system did not reveal a significant role for PRN in the pathogenesis of B. bronchiseptica. Rather than extending the use of this conventional experimental system that generates extreme lung pathology but poorly simulates natural infection, we considered newer assays and approaches we have developed.

PRN contributes to transmission

We recently developed an experimental system to study various aspects of the process of transmission between hosts using C3H/HeJ mice [22-25,32]. To relate shedding and transmission to other aspects of the infectious process often assumed to affect shedding, such as bacterial load, we first assessed the time course of infection in these mice. C3H/HeJ mice were inoculated with 150 CFU of either Bb WT or BbΔprn in 5 μL of PBS, a volume that deposits the bacterial inoculum only into the nasal cavity. Respiratory organs were harvested at 7-, 14- and 28-dpi. Both bacteria grew over time, with similar numbers observed in nasal cavity, trachea and lung (S6 Fig), suggesting that PRN has no critical role in colonization, growth, and spread within the respiratory tracts of these mice.

To assess the contribution of PRN to transmission, pairs of C3H/HeJ mice were inoculated with 150 CFU of either Bb WT or BbΔprn strain (in index mice) and then co-housed with two naïve C3H/HeJ mice (uninfected recipient mice) in cages of four. To evaluate bacterial transmission, the nasal cavities of 12 such co-housed naïve mice per strain (from 6 cages) were assessed for bacterial colonization after 21 days of co-housing with infected mice. In the Bb WT group, all 12 co-housed naïve mice became colonized, while BbΔprn bacteria were only transmitted to 5 out of 12 co-housed naïve mice (Fig 1A), suggesting that PRN contributes to bacterial transmission between hosts.

Fig 1

B. bronchiseptica PRN contributes to bacterial transmission and shedding.

A) Transmission rates among Bb WT (blue) or BbΔprn (orange) inoculated mice. B) Number of bacteria shed from the external nares of Bb WT infected mice (blue) and BbΔprn infected (orange) mice. Twelve mice were utilized in each time point per group. The dashed lines show the detection limit in the experiment. Error bars show the standard error of mean. Statistical significance was calculated by using chi-square test in panel A and two-way ANOVA in panel B. **p < 0.01.

PRN contributes to bacterial shedding

The deficiency of BbΔprn in transmission could be due to either a defect in shedding from infected hosts or colonization of exposed hosts. To test the efficiency of colonization, we inoculated C3H/HeJ mice with decreasing doses of bacteria. Both Bb WT and BbΔprn efficiently colonized the respiratory tracts of all mice with a calculated inoculation dose of ~5 CFU (S7 Fig) indicating that PRN is not required for efficient colonization of the respiratory tract.

Since PRN contributes to transmission but is not required for efficient colonization, we investigated its effect on shedding from the noses of challenged mice. C3H/HeJ mice (n = 12) were inoculated with 150 CFU of either Bb WT or BbΔprn, and bacteria shed from the nose were collected by gently swabbing the external nares with Dacron-polyester tipped swab every two- or three-days post inoculation. Throughout the experiment, WT bacteria were shed from the nares at high numbers, > 1000 CFU at multiple consecutive timepoints, indicating prolonged, high level shedding. In contrast, the BbΔprn strain was shed at 10 to 100-fold lower numbers from the first day of shedding to the end of the experiment at day 21 (Fig 1B), revealing a critical role for PRN in efficient shedding from the nose to the environment.

PRN contributes to inflammation in the nasal cavity

The striking difference in shedding, without a substantially different load of bacteria in the respiratory tract, led us to speculate that PRN might induce shedding by altering the inflammatory state and/or mucus production. To test this hypothesis, C3H/HeJ mice infected with Bb WT or BbΔprn were collected at 7 and 14 dpi for histopathological analysis. Histopathology showed that in the WT group, all 10 analyzed mice had inflammation in the nasal cavity ranging from mild to severe (Fig 2G–I). In contrast, of the BbΔprn infection group, 2 out of 10 had no apparent inflammation and the other 8 had severity scores which varied from minimal to mild (Fig 2D–F). These data indicate that PRN is involved in induction of inflammation in the nasopharynx (Fig 3A and S1 Table). To probe the effect of PRN on immune cell recruitment, flow cytometry was used to analyze immune cell populations in nasal cavities 14 days after inoculation with Bb WT or BbΔprn (Fig 4). Bb WT induced a significant increase in numbers of neutrophils, NK cells, macrophages, and B cells in the nasal cavities (Fig 4C–F). In contrast, BbΔprn induced an increase in macrophages, but no significant increase of other cell types compared to the uninfected control (Fig 4). These results indicate that PRN plays a role(s) in activation and recruitment of neutrophils, NK cells, and B cells, leading to higher inflammation levels in the nasal cavity.

Fig 2

B. bronchiseptica PRN induces inflammation and mucus secretion.

(A-I) Coronal sections of the nasal cavity (level 4) of control or B. bronchiseptica (BbΔprn or Bb WT strains) infected C57BL/6J mice at 14 dpi with stained with Alcian blue–Periodic acid-Schiff hematoxylin (AB-PASH) stain. A) Image of normal nasal cavity, level 4, from a PBS-inoculated control mouse. At this level, a layer of olfactory epithelium (arrowheads) lines the nasal mucosa within the dorsal and middle nasal meatus. Scale bar = 500μm. B) Higher magnification of Fig 2A (dashed rectangle) showing ethmoturbinates (E) and septum (S) lined by a layer of olfactory epithelium (arrowhead) within the middle meatus (MM). Scale bar = 200 μm. C) Higher magnification of Fig 2B (dashed rectangle) showing an ethmoturbinates lined by olfactory epithelium (arrow). Arrowhead points to a mucus producing (alcian blue-positive) gland within the lamina propria. Scale bar = 20 μm. D) Image of the nasal cavity, level 4, from a mouse inoculated with BbΔprn. The arrow points to a moderate accumulation of mucopurulent within the middle meatus. Scale bar = 500 μm. E) Higher magnification of Fig 2D (dashed rectangle) showing deposition of mucopurulent exudate (arrow) covering the olfactory epithelium (arrowhead) within the middle meatus (MM). There is thinning of the olfactory epithelium (arrowheads) along the septum (S) and ethmoturbinate (E). Scale bar = 200 μm. F) Higher magnification of Fig 2E (dashed rectangle) showing an ethmoturbinate. There is loss of olfactory epithelium and replacement by ciliated epithelium covered by a thin layer of mucus (arrow). There is loss of mucus producing glands within the lamina propria. Scale bar = 20 μm. G) Image of the nasal cavity, level 4, from a B. bronchiseptica (RB50)-inoculated mouse. The arrow points to a severe accumulation of mucopurulent within the dorsal (DM) and middle meatus (MM). Scale bar = 500 μm. H) Higher magnification of Fig 2G (dashed rectangle) showing deposition of mucopurulent exudate (arrow) covering the olfactory epithelium within the middle meatus (MM). There is thinning and loss of the olfactory epithelium (arrowheads) along the septum (S) and ethmoturbinate (E). Scale bar = 200 μm. I) Higher magnification of Fig 2H (dashed rectangle) showing an ethmoturbinate. Large numbers of PAS-positive neutrophils cover the ethmoturbinate. There is total loss of olfactory epithelium and replacement by non-ciliated epithelium (arrow). There is loss of mucus producing glands within the lamina propria. Scale bar = 20 μm. Abbreviations: Dorsal Meatus (DM); Ethmoturbinate (E); Harderian Gland (HG); Maxillary Sinus (MS); Middle Meatus (MM); Nasopharyngeal Meatus (NPM); Olfactory Bulb (OB); Septum (S). There were 5 mice in each time point per group.

Fig 3

PRN induces inflammation in the nasal cavity.

A) Inflammation level in the nasal cavities of mice inoculated with either Bb WT (blue) or Bb Δprn (orange) at 7-and 14-dpi. B) Mucus accumulation in the nasal cavity of Bb WT (blue) or Bb Δprn (orange) inoculated mice at day 7 pi. Individual samples from either group are highlighted with dashed circles. The veterinary pathologist conducting the inflammation level analysis (UBM) was blinded to the sample source. There were 5 mice in each time point per group. Error bar shows the standard error of mean. Statistical significance was calculated using Two-way ANOVA. *p < 0.05, **p < 0.01, ***p < 0.001.

Fig 4

B. bronchiseptica PRN contributes to the recruitment of leukocytes, neutrophils and B cells in the nasal cavity.

A) Total cells recruited in nasal cavities. B) Comparison of CD11b+ cells recruited in nasal cavities. C) Neutrophils recruited in nasal cavities. D) NK cells recruited in nasal cavities. E) Macrophages recruited in nasal cavities. F) B cells recruited in nasal cavities. There were 5 mice in each time point per group. Error bar shows the standard error of mean. Statistical significance was calculated using One-way ANOVA. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

PRN increases mucus secretion in acute inflammation

The dramatically higher number of bacteria shed from mice infected with Bb WT than those given BbΔprn led us to examine whether increased mucus production might be involved. To compare mucus secretion in the nasal cavity of Bb WT infected and BbΔprn infected mice, Alcian blue-Periodic acid Schiff (PAS) staining was performed on tissue sections. At 7 dpi all 5 Bb WT infected mice displayed moderate mucus secretion. In contrast, only 2 out of 5 BbΔprn infected mice displayed some mucus accumulation and only in a small area, while the other 3 showed no mucus accumulation at all (Fig 3B). The substantially higher mucus secretion in Bb WT infected mice compared to BbΔprn infected mice indicates that PRN plays a role in the induction of mucus production during the acute phase of infection.

PRN induces secretion of pro-inflammatory cytokines

Because of the very different numbers of neutrophils we observed in the noses of the two groups of mice, we hypothesized that PRN may affect the induction of two pro-inflammatory cytokines that recruit and activate neutrophils, IL-1 and TNF-α [33-35]. We harvested the nasal cavities of mice inoculated with Bb WT or BbΔprn at 24- and 48- hpi and observed somewhat higher levels of IL-1β and TNF-α in the former, albeit without statistical significance (S8 Fig). To test the effect of PRN on cytokine secretion from macrophages, we exposed macrophages to Bb WT or BbΔprn and observed higher levels of TNF- α in the former group, suggesting that PRN promotes secretion of TNF-α, but did not affect IL-1β (Fig 5 and S9 Fig).

Fig 5

PRN induces the secretion of TNF-α from RAW 264.7 macrophages.

TNF-α secretion from macrophages when exposed with Bb WT (blue) or BbΔprn (orange) for 1 hour with MOI at 10 (A) or MOI at 100 (B). Error bar shows the standard error of mean. Statistical significance was calculated using unpaired T test. *p < 0.05, **p < 0.01, ***p < 0.001.

PRN contributes to intracellular survival and systemic dissemination

Bb has been shown to be able to invade and persist in immune cells, which can have profound effects on the local and systemic immune responses [36-38]. Since PRN has been implicated in interactions with immune cells, we tested its effects on access to and invasion of immune organs [17,39]. We harvested spleens from mice inoculated with 150 CFU of either Bb WT or BbΔprn at 7-, 14- and 28-dpi. Bb WT was recovered from 3 of 4 spleens and 5 of 8 spleens of infected mice on days 7 and 14, respectively, while no bacteria were found in the spleens of any BbΔprn inoculated mice (Fig 6A). To examine whether Bb WT in the spleen were surviving intracellularly, single cell suspension of splenocytes were treated with gentamicin treatment to kill the extracellular bacteria. Gentamicin killed 100% of control bacteria from culture (without host cells present), but did not kill all of those recovered from spleen, indicating that some were within host cells (Fig 6B), indicating that Bb WT expressing PRN can survive within the splenocytes. Additionally, we detected a higher IgG titer in Bb WT infected mice compared to that in BbΔprn infected mice (S10 Fig), suggesting that a strong immune response was induced to against this systemically disseminated infection.

Fig 6

PRN contribute to systemic dissemination and intracellular survival of B. bronchiseptica.

A) Bb WT were recovered from spleen of infected mice at 7- and 14 dpi, while there was no bacterial recovery from spleen in BbΔprn-infected mice in both time points. There were 4 mice for assays performed 7- and 28 dpi, and 8 mice for assays performed at 14 dpi per group. B) Bb WT were recovered from isolated splenocytes of infected mice at 7 dpi. The brown circles represent bacteria recovered from isolated splenocytes without gentamicin treatment while purple squares represent bacteria recovered from splenocytes post gentamicin treatment. The dashed lines show the detection limit in both experiments. Error bar shows the standard error of mean. Statistical significance was calculated by using Two-way ANOVA. *p < 0.05, **p < 0.01, ***p < 0.001.

Discussion

Previous studies have used different and unrelated clinical isolates of B. pertussis with or without PRN, demonstrating, for example, that an isolate lacking PRN sustained a longer infection in mice immunized with an acellular pertussis vaccine [40]. Such differences cannot be specifically attributed to the loss of PRN, since the genomes of these two separate isolates differ in many other genes. Moreover, because B. pertussis poorly colonizes, sheds, and transmits between mice, it has been difficult to assess the role of PRN in these critical aspects of the infectious cycle. To address these problems, we and Inatsuka et al. used an in-frame deletion of prn in B. bronchiseptica, a close evolutionary progenitor of B. pertussis that naturally infects mice, to test the roles of PRN in infections of a natural host. In the conventional Bordetella spp. experimental system in which a very high inoculation dose (5 x 105 CFU) delivered in a large volume (50 μL) to deposit bacteria deep in the lungs, PRN had a modest effect in BALB/c mice in the prior study [17]. In our high dose pneumonic challenge experiment, using different mice (C57BL/6) assessed at different time points, we did not observe measurable defects of the prn mutant. Since BALB/c and C57BL/6 mice have different immune responses to B. pertussis infection and vaccination [41-43], the different genetic backgrounds of two mouse strains and/or the different time points may explain the different observations in these two studies.

Rather those extensive comparisons between experimental conditions using the conventional experimental system, which is an extreme model of severe pneumonic disease, we directed our efforts toward assays that are more sensitive to various aspects of the infection and transmission processes. To better simulate a more natural, gradually progressing Bordetella spp. infection, we used a low dose inoculation model in which an experimental dose as low as 5 CFU is delivered into the external nares of mice in a 5μl droplet that deposits bacteria only in the nose. In this more natural experimental system, PRN was not required for efficient colonization and spread within inoculated hosts. When delivered only into the nasal cavity, rather than deep into the lungs, Bb WT-inoculated mice shed significantly more bacteria and transmitted between mice much more efficiently than BbΔprn-inoculated mice (Fig 1). Bb WT also induced significantly more inflammation and mucus production than RB50Δprn. Inflammation and mucus production contribute to the rhinorrhea that facilitates shedding and transmission, as noted previously with Streptococcus pneumonia transmission [44]. The significantly higher number of neutrophils were observed in the nares of Bb WT infected mice, suggesting that PRN may play critical roles in recruitment and activation of neutrophils and/or in resistance to neutrophils-mediated clearance, as was observed by Inatsuka CS et al [17]. PRN induced secretion of TNF-α from RAW 264.7 macrophages in in-vitro, which may explain the higher inflammation levels observed in mice infected with Bb WT. The role of TNF-α in the transmission of B. bronchiseptica deserves more attention in future studies as this factor may work as a potential therapeutic target for blocking transmission of Bordetella spp. and other pathogens.

PRN, as an outer surface protein with a demonstrated ability to affect adhesion to mammalian cells in vitro [9-11], is generally considered an “adhesin”. However, it is not clear that PRN contributes to adhesion to either bronchial or laryngeal cells in vivo. In this study, deleting PRN did not affect the ability of B. bronchiseptica to efficiently colonize mice with a remarkably small inoculum, indicating that PRN is not required for the efficient adherence involved in initial colonization. However, Bb WT showed significantly higher recovery from spleens of infected hosts, indicating that PRN affects the ability to get to and/or survive in the spleen. The observation of substantial numbers of Bb WT, but not BbΔprn, within splenocytes (Fig 6), suggests PRN might affect invasion of immune cells and/or phagocytic killing within splenocytes. Upon encountering invading bacteria, immune cells trigger cascades of inflammatory responses by secreting cytokines, chemokines, small lipid mediators (SLM), and antimicrobial peptides (AMPs) that can contribute to increased phagocytic capacity and bacterial clearance [45], amongst other potential effects. After phagocytosis, peripheral immune cells can carry engulfed bacteria to deeper tissues including draining lymph nodes for T cell priming [46]. Thus, systemic dissemination of Bb WT, but not BbΔprn, may result in a stronger immune response against the former, consistent with the histopathology analysis and IgG antibody titers observed (Fig 2 and S10 Fig). In addition, the role of PRN in inducing TNF-α secretion may also contribute to the higher inflammation observed in the nasal cavity of Bb WT infected mice (Fig 5).

The increasing prevalence of PRN-deficient B. pertussis strains raises questions about both positive and negative (purifying) selection. It appears increasingly likely that PRN-containing ACVs select for PRN-deficient B. pertussis. It is also likely that the epidemiology of pertussis has changed as host behavior, population density, and worldwide travel affect the network of connected hosts. Even though PRN-deficient B. pertussis isolates may be shed less, a dense and well-connected population of vaccinated hosts in countries like the USA may be sufficient for a successful chain of transmission. Alternatively, the loss of PRN may have less cost in B. pertussis due to compensation by other genes. There are 15 other autotransporter genes in the genome of B. pertussis [5,31], one or more of which may compensate for the loss of PRN. Since PRN plays roles in systemic dissemination and induction of inflammation in murine transmission models (Figs 2–6), other factors facilitating systemic dissemination and promoting inflammation may compensate for its loss. Alternatively, C3H/HeJ mice may fail to mimic immunocompetent hosts in every aspect of pathogenesis or B. pertussis may be different from B. bronchiseptica in various aspects, and it is possible that the functions of PRN in these two species may differ to some extent. However, to test this, an efficient B. pertussis transmission model is needed. In addition, further studies are needed to identify factors that might compensate for the loss of PRN in current circulating strains. This might lead to the identification of alternative antigens that could be included in next generation vaccines against B. pertussis.

Material and methods

Ethics statement

Mouse experiments used in this study were performed in strict accordance to recommendations outlined in the Guide for Care and Use of Laboratory Animals of the National Institute of Health. Protocols were approved by the Institutional Animal Care and Use Committee at University of Georgia (A2016 02-010-Y2-A3, Bordetella-Host Interactions). Mice were closely monitored during experiments and any mouse found moribund was euthanized using CO2 inhalation to prevent unnecessary suffering.

Bacterial strains and growth

B. bronchiseptica WT strain RB50 and isogenic pertactin knockout strain BbΔprn (SP5) have been previously described [11,17]. To allow mutant and WT to be distinguished in “competition assays”, SM10λpir cells carrying allelic exchange vector pEH10 was used to generate gentamicin-resistant Δprn mutant strain. The generation of pEH10 was described previously [29]. Liquid cultures were prepared using Stainer Scholte (SS) medium supplemented with 0.5% (w/v) Heptakis (2,6-di-O-methyl)-β-cyclodextrin) (Sigma H0513). Plate cultures were grown on Bordet Gengou (BG) agar supplemented with 10% (v/v) defibrinated sheep blood (Hemstat, Hemostat Laboratories) and streptomycin (20 μg/mL). Comparative growth curves were generated from triplicate cultures of bacteria grown 48 hours in SS medium at 37°C and shaking at 200 rpm.

Adherence assay

Adherence assays were conducted following protocols described earlier [12,47]. In brief, human epithelial lung A459 cells were seeded in triplicate in 24-well plates at a density of 2.5 × 105 cell/well in Dulbecco’s modified Eagle’s medium (DMEM) (supplemented with 10% fetal bovine serum, 10 mM glutamine, 25 mM sodium pyruvate, 10 mM HEPES). Log-phase bacteria were suspended in warm DMEM medium and added to each well at a multiplicity of infection of 10:1 (bacteria: eukaryotic cells). The plate was centrifuged at 300Xg for 10 minutes to synchronize infection and the assay plates were incubated for 5 minutes at 37°C. Unattached bacteria were then removed by washing the cells 4 times with 1 mL phosphate-buffered saline (PBS). A459 cells were lysed with 100 μL of 0.1% sodium deoxycholate for 5 minutes and released bacteria suspended in 900 μL of PBS. The bacteria were enumerated by dilution plating on BG agar plates. RB50 ΔfhaB [12], a mutant strain deleted of the gene encoding the filamentous hemagglutinin and known to be defective in adherence, was used as the negative control.

Cytotoxicity assay

Cytotoxicity assays were conducted on RAW 264.7 cells, using the CytoTox 96 Nonradioactive Cytotoxicity Assay Kit (Promega) following manufacturer’s protocols. In brief, 100 μL of 2.5 × 104 macrophages were seeded in triplicate in a 96-well plate followed by adding bacteria at a multiplicity of infection of 10:1. The assay plate was centrifuged at 300Xg for 10 minutes. Bacteria were incubated with the macrophages for 4 hours, following which the plate was centrifuged for 5 minutes (300Xg). 50 μL of the supernatant was placed into a fresh flat-bottomed 96-well plate and was calorimetrically assayed for lactate dehydrogenase. A noninfected group was used as a negative control.

Cytokine test

To assess secretion of TNF-α and IL-1β from RAW 264.7 macrophages, the supernatant of infected cells (MOI = 10 or MOI = 100) was collected at 1-or 4 hpi. To assess the levels of TNF-α and IL-1β in the noses of mice infected with Bb WT or BbΔprn, noses were collected in PBS and homogenized using a bead tissue disruptor. Cytokine levels were determined using the commercially available ELISA kits R&D Systems Mouse IL-1 beta DuoSet ELISA and TNF-alpha DuoSet ELISA following the manufacturer’s instructions.

Mouse infections

All work with mice was conducted following institutional guidelines. Six-week-old female C57 BL/6 or C3H/HeJ mice (Jackson Laboratories, Bar Harbor, Maine) were used for assessing the colonization and the progress of infection of the respiratory tract. As required, 5–150 CFU of bacteria was delivered in 5 μL of PBS to the nares of mice anesthetized with isoflurane/oxygen. For the colonization profiles, groups of 4 mice were inoculated with WT or mutant bacteria, and at the indicated days 4 mice of each group were euthanized with carbon dioxide (CO2) and the nasal cavity, trachea, and lungs were collected in PBS and homogenized using a bead tissue disruptor. Bacterial load was enumerated by dilution plating.

Transmission and shedding assay

Transmission assays were conducted using the transmission permissive C3H/HeJ (TLR4 deficient) strain of mouse (Jackson Laboratory) whereby infected (index) mice were co-housed with uninfected (naive) mice [25]. In brief, mice were lightly anaesthetized with isoflurane/oxygen and inoculated with 150 CFU of bacteria delivered in 5 μL of PBS onto the nares. Inoculated (index) mice were then placed in cages with 2 uninfected (naive) mice. Transmission of B. bronchiseptica was assessed after 3 weeks of co-housing by enumerating the bacterial load in the nasal cavities of the naive mice. To monitor shedding, the external nares of the index mice were swabbed (32 swipes) with a dry Dacron polyester tipped swab at the indicated times. The swab was vortexed vigorously in 1 mL PBS for 30 seconds and bacteria enumerated on BG agar plates.

Histopathology

Following fixation in neutral-buffered, 10% formalin solution and subsequent decalcification in Kristensen’s solution, coronal sections were made through the nose. Tissues were subsequently processed, embedded in paraffin, sectioned at approximately 5μm, and stained with hematoxylin and eosin. Histopathological examination consisted of evaluation of the nose for the incidence (presence or absence), severity, and distribution of inflammation. Histopathologic severity scores were assigned as grades 0 (no significant histopathological alterations); 1 (minimal); 2 (mild); 3 (moderate); or 4 (severe) based on an increasing extent and/or complexity of change, unless otherwise specified. Lesion distribution was recorded as focal, multifocal, or diffuse, with distribution scores of 1, 2, or 3, respectively [48].

Flow cytometry

Five mice per experimental group were euthanized by CO2 inhalation. Nasal cavities were harvested, placed in 1mL of RPMI 1640 medium and homogenized via a syringe plunger against a 40μm cell strainer. Cell suspensions were centrifuged at 1,500rpm for 10 minutes and remaining red blood cells were lysed with ACK lysing buffer. After washing with PBS, the cells were incubated with 1μl Zombie aqua (Biolegend) for 20 minutes, washed again, and incubated with 1μl Fc Block (Biolegend) for 30 minutes. Surface marker staining was added to each sample from a master mix of antibodies. Cells were fixed, washed, and resuspended in 250μl FACS buffer. Flow cytometry (Acea Novocyte Quanteon) was performed was used to sort neutrophils (CD11b-Pe-Cy7, CD115-APC, Ly6G-AF488) and macrophages (CD11b-PE-Cy7 CD115-APC, Ly6G-AF488, F4/80-PE). In a separate panel, T cells (CD45-AF700, CD3-APC), B cells (B220-PE-Cy7), and NK cells (NK1.1-PE) were sorted via a separate gating strategy. Viable cells were gated as Zombie aqua-negative cells. The data were analysed with FlowJo 10.0. (S11 Fig).

Cell isolation for intracellular survival test

The spleen organs harvested from WT or mutant infected Hej mice was cut into 3–4 pieces using sterile scissors. Spleen pieces were mashed with the rough surfaces of two sterile frosted microscope slides and slide surfaces were rinsed with 1 mL (FBS-free) DMEM medium. The samples were then passed through a 70 μm cell strainer to obtain a single-cell suspension. A 100μl sample was plated on BG agar containing 20 μg/ml streptomycin to estimate the number of total bacteria. Cells were washed twice with PBS and then incubated for 1 h with 300 μg/ml gentamicin to eliminate extracellular bacteria. Cells were washed twice with PBS and then lysed with 0.1% triton X-100 for 15 minutes at room temperature. CFU counting was performed on cell lysates by plating 10-fold serial dilutions onto BG agar plates containing 20 μg/ml streptomycin to estimate the number of intracellular bacteria. Control experiments to assess the efficacy of gentamicin were performed in parallel [49]. Briefly, samples of 103, 104 or 105 bacteria were incubated with 300 μg/ml gentamicin for 1 hour at 37°C and plated on BG agar. There were 3 replicates for each group. The results showed that more than 99.9% bacteria were killed.

Statistical analysis

Statistical analysis of differences between the WT and mutant groups was performed using the Unpaired Student 2-tailed t test, One-way ANOVA and Two-way ANOVA test, as appropriate. P value less than 0.05 shown as *; P value less than 0.01 shown as **; P value less than 0.001 shown as ***; P value less than 0.0001 shown as ****.

Schematic of in-frame deletion of pertactin gene.

(TIF)

Click here for additional data file.

Similar laboratory growth in vitro of Bb WT (blue) and BbΔprn (orange) bacteria.

There were 3 replicates in each time point per group. Error bar shows the standard error of mean. Statistical significance was calculated using Two-way ANOVA.

(TIF)

Click here for additional data file.

No effect of pertactin on bacterial adhesion to human alveolar epithelial cells.

Adherence to A549 lung epithelial cells of Bb WT (blue), BbΔprn (orange) and BbΔfha (yellow) that was previously shown to be impaired in its ability to adhere to epithelial cells. There were 3 replicates in each group. Error bar shows the standard error of mean. Statistical significance was calculated using One-way ANOVA. ***p < 0.001.

(TIF)

Click here for additional data file.

No significant difference in cell cytotoxicity of Bb WT and BbΔprn to RAW 264.7 macrophages.

There were 3 replicates in each group. Error bar shows the standard error of mean. Statistical significance was calculated using One-way ANOVA. ****p < 0.0001.

(TIF)

Click here for additional data file.

Comparative colonization profiles of Bb WT and BbΔprn in C57 BL/6 mice.

Bacterial CFU recovered on days 3-, 7-, 14-, and 28 dpi from the nasal cavities, trachea, and lungs of mice inoculated with either wild-type (blue) or mutant (orange) bacteria. There were 4 mice in each time point per group. Error bar shows the standard error of mean. Statistical significance was calculated by Two-way ANOVA. *p < 0.05, **p < 0.01, ***p < 0.001.

(TIF)

Click here for additional data file.

The role of PRN in colonization of B. bronchiseptica in C3H/HeJ mice.

Comparative colonization profiles of Bb WT and BbΔprn in C3H/HeJ mice. Number of colony-forming units (CFU) recovered on days 3-, 7-, 14-, and 28 pi from the nasal cavities, trachea, and lungs of mice infected with either wild-type (blue) or mutant (orange) bacteria. There were 4 mice in each time point per group. Error bar shows the standard error of mean. Statistical significance was calculated by using Two-way ANOVA.

(TIF)

Click here for additional data file.

B. bronchiseptica PRN is not required for efficient colonization.

ID50 test of Bb WT and BbΔprn showing bacterial numbers at 7 dpi in respiratory organs of C3H/HeJ mice inoculated with incrementally increasing doses of 5 CFU, 25 CFU and 125 CFU. Number of CFU recovered from nasal cavities, trachea, and lungs revealed no difference in colonization between wild-type and mutant. There were 4 mice in each time point per group. Error bar shows the standard error of mean. Statistical significance was calculated by using Unpaired t-test.

(TIF)

Click here for additional data file.

TNF-α and IL-1β levels in noses of mice infected with Bb WT or BbΔprn at 24hpi or 48hpi.

The levels of TNF-α in noses of mice infected with Bb WT (blue) or BbΔprn (orange) at 24hpi (A) or 48hpi (C). The levels of IL-1β in noses of mice infected with Bb WT (blue) or BbΔprn (orange) at 24hpi (B) or 48hpi (D). Error bar shows the standard error of mean. Statistical significance was calculated using unpaired T test.

(TIF)

Click here for additional data file.

PRN does not induce the secretion of IL-1β from RAW 264.7 macrophages.

A) The IL-1β detected in the supernatant when RAW macrophages were exposed with Bb WT (blue) or BbΔprn (orange) with a MOI at 10 for 1 hour. B) The IL-1β detected in the supernatant when RAW macrophages were exposed with Bb WT (blue) or BbΔprn (orange) with a MOI at 100 for 1 hour. Error bar shows the standard error of mean. Statistical significance was calculated using unpaired T test.

(TIF)

Click here for additional data file.

PRN contributed to the generation of anti-B. bronchiseptica IgG antibodies.

IgG antibody titers against B. bronchiseptica were determined in sera of C3H/HeJ mice infected with either Bb WT (blue) or BbΔprn (orange) at 28 dpi. There were 4 mice per group. Error bars show the standard error of mean. Statistical significance was calculated by using One-way ANOVA. *p < 0.05, **p < 0.01, ***p < 0.001.

(TIF)

Click here for additional data file.

Gating strategy for flow cytometry.

A) Gating strategy for myeloid cell types. B) Gating strategy for lymphoid cell types.

(TIF)

Click here for additional data file.

Scoring standard for histopathology analysis.

(TIF)

Click here for additional data file.

8 Dec 2020

Dear Mr. Ma,

Thank you very much for submitting your manuscript "Pertactin contributes to shedding and transmission of Bordetella bronchiseptica" for consideration at PLOS Pathogens. As with all papers reviewed by the journal, your manuscript was reviewed by members of the editorial board and by several independent reviewers. In light of the reviews (below this email), we would like to invite the resubmission of a significantly-revised version that takes into account the reviewers' comments.

While this manuscript does provide some novel findings the manuscript as it currently stands does not reach the threshold for publication in PlosPathogen. Increased mechanistic insight is required to better prove the role of Prn and most importantly a better attempt must be made to translate the relevance of the model to B. pertussis infection in humans.

We cannot make any decision about publication until we have seen the revised manuscript and your response to the reviewers' comments. Your revised manuscript is also likely to be sent to reviewers for further evaluation.

When you are ready to resubmit, please upload the following:

[1] A letter containing a detailed list of your responses to the review comments and a description of the changes you have made in the manuscript. Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

[2] Two versions of the revised manuscript: one with either highlights or tracked changes denoting where the text has been changed; the other a clean version (uploaded as the manuscript file).

Important additional instructions are given below your reviewer comments.

Please prepare and submit your revised manuscript within 60 days. If you anticipate any delay, please let us know the expected resubmission date by replying to this email. Please note that revised manuscripts received after the 60-day due date may require evaluation and peer review similar to newly submitted manuscripts.

Thank you again for your submission. We hope that our editorial process has been constructive so far, and we welcome your feedback at any time. Please don't hesitate to contact us if you have any questions or comments.

Sincerely,

Rachel M McLoughlin, PhD

Associate Editor

PLOS Pathogens

David Skurnik

Section Editor

PLOS Pathogens

Kasturi Haldar

Editor-in-Chief

PLOS Pathogens

​orcid.org/0000-0001-5065-158X

Michael Malim

Editor-in-Chief

PLOS Pathogens

orcid.org/0000-0002-7699-2064

***********************

While this manuscript does provide some novel findings the manuscript as it currently stands does not reach the threshold for publication in PlosPathogen. Increased mechanistic insight is required to better prove the role of Prn and most importantly a better attempt must be made to translate the relevance of the model to B. pertussis infection in humans.

Reviewer's Responses to Questions

Part I - Summary

Please use this section to discuss strengths/weaknesses of study, novelty/significance, general execution and scholarship.

Reviewer #1: There are some novel and potentially interesting findings in this study, but unfortunately it is very thin, lacking detail and mechanistic insight. The authors need to do a good deal more work before this story is complete.

Reviewer #2: The paper addresses a long standing query in the Bordetella field: the true role of Prn in infection biology. Early studies suggested Prn to be an adhesin, and the presence of an RGD motif supports this. However, clear evidence for this has been obtained from in vitro studies only, and in vivo studies demonstrated efficient colonisation in the absence of Prn.

Here, careful studies using very small inoculae very clearly demonstrate that BB prn mutants can very efficiently colonise and persist in the respiratory tracts of mice, demonstrating that Prn is not required for these processes in BB.

Instead, data is presented to argue for a role in intracellular survival, and in inducing inflammatory responses to BB infection. Of particular interest is the demonstration of reduced shedding of BB prn mutants from the respiratory tract and reduced transmission between index and recipient mice. The authors combine these findings to suggest a role for Prn in inducing inflammation, that facilitates bacterial shedding and onward transmission.

This story makes a telling contribution to defining the infection biology of the bordetellae, and adds to understanding the role of Prn in this. Prn is of particular importance as it is one of the components of many of the acellular Pertussis vaccines. Importantly, in countries experiencing resurgence of pertussis, there is a stark increase in the frequency of isolates that are Prn-deficient, and there are multiple lines of evidence to suggest that Prn-deficient B. pertussis strains have a fitness advantage in populations that are highly vaccinated with the acellular vaccines.

In the light of this, the study is important, as it addresses the role of pertactin in infection which is crucial to understanding the consequences of Prn deficiency for B. pertussis fitness.

Reviewer #3: Using a TLR4 deficient mouse model, Ma et al explore the role of Pertactin (PRN) of B. bronchiseptica in colonization, growth, spread within the respiratory tract, shedding, and transmission. It was demonstrated that while PRN deletion mutants showed no change in colonization of the trachea and lungs of mice compared to the WT strain, shedding and transmission was significantly reduced. The respiratory epithelia of mice infected with the PRN deletion mutant exhibited significantly reduced inflammation and mucus accumulation compared to the WT strain. Although the results suggest an important role for pertactin in B. bronchiseptica, the authors overstate what their model shows with regards to B. pertussis infections.

The argument that studying the B. bronchiseptica in one of its natural hosts provides valuable insights into B. pertussis infections in humans would only be valid if the diseases caused by the two pathogens in their natural hosts were comparable. Although the bacteria are genetically similar and share some virulence factors, the diseases caused by B. bronchiseptica and B. pertussis are very different, with different tropisms (humans/NHP vs. many mammals except humans) and clinical presentations (acute/weeks vs prolonged/chronic). Additionally, pertussis toxin is a key toxin and essential virulence factor in pertussis infections (PT-only vaccines prevent disease), yet it is not expressed by B. bronchiseptica—highlighting a major difference in disease mechanisms. Bb and Bp also express different lipid A structures and polysaccharide chains (LPS vs LOS), with Bb having a more inflammatory lipid A/LPS (2). Using a TLR4-deficient mouse strain to allow disease transmission further confounds the applicability of these results to B. pertussis infections, since TLR4 appears to have different roles in Bb and Bp infections (3). PRN’s contribution to inflammation, which the researchers suggest may influence mucus production, shedding and transmission, may be insignificant in a TLR4-competent mouse (a key innate immune recognition molecule for endotoxin and host DAMPs). It is possible the effects observed are artifacts of the experimental model.

B. pertussis survives very poorly outside the host whereas B. bronchiseptica has been shown to survive for prolonged periods of time in water. B bronchiseptica can be efficiently transferred by contact and by sharing of water sources. This is not true of B pertussis which is transmitted by respiratory droplets and aerosols.

**********

Part II – Major Issues: Key Experiments Required for Acceptance

Please use this section to detail the key new experiments or modifications of existing experiments that should be absolutely required to validate study conclusions.

Generally, there should be no more than 3 such required experiments or major modifications for a "Major Revision" recommendation. If more than 3 experiments are necessary to validate the study conclusions, then you are encouraged to recommend "Reject".

Reviewer #1: 1) The authors justify their research on pertactin (PRN) from Bordetella on the basis of the observations that pertactin-deleted mutants of Bordetella pertussis have emerged in populations immunized with acellular pertussis vaccine that include PRN as an antigen. However, they carry out their research with B. bronchiseptica and the justification is that they have a mouse transmission model with this bacteria, which they don’t have with B. pertussis. However, many aspects of the study, local inflammation, mucus secretion and intracellular survival could have been done with B. pertussis, where they could have used human clinical isolates, rather than lab-generated mutants of B. bronchiseptica.

2) The biggest issue with this report is that it lacks depth, the essential data could be condensed into 2 figures. Do we need a whole figure showing only two bars on a bar graph (Fig. 3) These bars don’t have error bars, making less of case for including them in a figure at all. The first two figures of negative data (Fig 1 and 2) that could be one supplementary figure. The 3 interesting observations are a) transmission and shedding is lower in mice infected with the PRN-neg mutant, b) there is less inflammation/mucus secretion in nasal cavity of mice infected with the PRN-neg mutant c) intracellular survival in macrophages in vitro is lower with the PRN-neg mutant. However, none of these observations are linked. The authors speculate on immune responses being stronger in mice infected with WT compared with PRN-negative mutants of B. bronchiseptica, but have not examined this. They speculate about higher inflammasome activation with WT bacteria, but have not assessed this. There is not data linking either intracellular survival or inflammation with transmission.

3) The authors should consider some studies with purified PRN where they look at its effect on inflammatory responses of macrophages in vitro. There is some evidence that PRN works with FHA to suppress LPS-induced inflammatory responses by macrophages. It is not clear how those observation would fit with the suggestion in the current study that PRN may promote inflammation. Either way, it would worth testing the effect of PRN on induction or inhibition of inflammasome activation? Studies such as these might help to expand on their theories around the inflammasome.

4) The B. pertussis challenge dose for the early figures is 0.5 million CFU, but is only 150 CFU in the later studies, how can these be reconciled or compared in making conclusion on the course of infection versus inflammation or shedding etc.

Reviewer #2: A key message is that Prn contributes to the induction of inflammation by BB. As it stands, it is not possible to determine if the data supports this.

Ln342-351. More detail is required for the approach to histopathology scoring. Currently, it is not possible to understand how scores were determined. For example, grade 0- no significant alterations. What counts as significant? Grade 1 – minimal, which would be zero? Scoring based partly on ‘complexity’ of change. What does this mean? What area of the total tissue was scored? How was bias excluded from the selection of areas. Were observers blinded to the source in terms of WT or mutant? I appreciate that drawing quantitative measures from these sorts of data is difficult, but there needs to be greater clarity as to how quantitative measures were drawn from visual images for the conclusion regarding the role of Prn in inflammation to stand up to scrutiny.

Reviewer #3: Comments:

1. The authors state that RB50 transmits efficiently between mice (lines 143-144) but that only appears to be the case in an immuno-deficient mouse strain. In order to draw conclusions about the transmission of PRN-negative B. bronchiseptica strains, the authors should explore animal models of B. bronchiseptica transmission that don’t require the use of immunodeficient animals (e.g. the pig model)

2. The researchers should complement or repair the isogenic PRN KO, to confirm that unintentional, off-target effects are not contributing to their disease models (mice, cell culture).

3. The researchers should consider assessing inflammasome activation in their in vitro intracellular invasion system as they speculate that may explain differences in inflammation induced by the WT and PRN-deficient strains.

**********

Part III – Minor Issues: Editorial and Data Presentation Modifications

Please use this section for editorial suggestions as well as relatively minor modifications of existing data that would enhance clarity.

Reviewer #1: The preparation of manuscript, especially the figures, are not up to the standard required for a top-class journal. e.g. out of sequence figure call out, references to figures in the Introduction and discussion, poor quality figures; it is almost impossible to read the font on some figures. The Y axis scales on panels within figures are not consistent. Some Y axes are split when they don’t need to be, and many don’t start at zero.

I think many people in the field would disagree with the statement in the introduction that whole cell vaccine had “relatively minor side effects”.

Reviewer #2: There is a clear contradiction between Prn contributing to transmission and the striking increase in Prn-deficiency among B. pertussis. The authors mention possible compensation for loss of Prn from B. pertussis by other autotransporters, but there is no clear rationale for this. Could the authors comment on the difference between BB and B. pertussis in the ris system that has been linked to intracellular survival, and for which Chen and Stibbitz suggested a role in promoting the expression of Vrgs and aerosol transmission (Curr Opin Microbiol. 2019. DOI: 10.1016/j.mib.2019.01.002. The authors should also acknowledge that their findings could indicate that Prn has a different role in B. pertussis compared to BB. This does not question the findings reported here regarding the role for Prn in BB infection, but if a key motivation for this study was to address the question of Prn and Prn-deficiency among B. pertussis, then a more rounded discussion of this is required.

There are a number of other points that the authors could consider:

Ln 117. Describing BB and BP as 98% identical at the nucleotide level is misleading. Speciation of B. pertussis involved loss of over 1Mb of DNA compared to BB, describing identity as 98% infers there is only 2% difference between the two.

Ln119. ‘ref5’ suggests a citation issue.

Ln131. It is unusual to refer to results figures in the introductiomn.

Ln209-212. This is discussion not results.

Ln204. States RAW cells were infected with bacteria at an MOI of 10 and 1, materials states 100, 10 and 1. States extracellular bacteria were killed by 1 and 2 hours of gent treatment. But in Ln276-8, it suggests Prn mutant is gent resistant? Ln 315 states extracellular bacteria were killed with polymyxin but this was added 1 hour after addition of bacteria to the RAWs. There appears contradiction between these sections.

If polymyxin B was used, is the sensitivity of WT and the Prn mutant to PMB identical, particularly at low levels? Some PMB will cross the RAW membranes and an increased sensitivity of the Prn mutant would reduce numbers recovered. The cytotoxicity of BB for mammalian cells could compromise cell membranes. The same applies for Triton X-100 as this was used to lyse the RAW cells to recover the intracellular bacteria. It is important to show that different sensitivity was not the reason for the difference in recovered bacteria.

Figure 1. For transmission, index mice were co-housed with naïve mice for 21 days. Wt mice were cleared from the lungs of the mice sometime between the day 14 and day 28 time points. It is not clear when these WT were cleared, it might have been day 27, or day 15. Prn mutants were recovered from the lungs on day 28. Thus the colonisation profiles of WT and prn mutants are different, although not in the region where shedding would be expected to occur from (nasal cavity), and this wouldn’t explain reduced shedding of Prn mutants, but it’s not correct to say no difference.

Figure 2. It is very difficult to be precise in getting exactly 5 cfu per inoculum. Were the inoculae plated? If so, show data to show the likely range of inoculate the mice will have received for each of the 3 categories. The data suggest that none of the mice received zero cfu?

Ln189, In the text, Figure 5A and B are identified as showing histopathology induced by WT BB, but in the Fig, A-C are panels showing sections from control mice. This might stem from the odd numbering for Figure 5 which appears to be 3 quite separate figures, but are included in one Figure labelled at 5A-C. However, 5A then has panels A-I, so creates an odd labelling of 5AA, 5AB but this is not used in the text.

Fig 5A. It is notoriously difficult to demonstrate a clear difference between pathology using single panels from single samples. As mentioned above, some explanation of how histopathology was scored, and how cell sections were selected for analysis in an unbiased way. It is difficult for the reader to understand the data depicted in Figs 5B and C without this.

Figure 6. For the intracellular survival of BB in RAWs, the data is presented as a survival ratio, but there is no explanation of what this is, or how it was calculated. In Fig 6A data is shown for WT and Prn mutants at MOIs of 1 and 10 for 2 hours post internalisation. In Fig 6B data is shown for 1, 2, 4 and 24 hours post internalisation, but the data for 2 hours appear to be different to that in Fig 6A. Are these data from a different experiment, or a different MOI?

Reviewer #3: 1. Researchers did not indicate if their p-values for student T tests were corrected for multiple comparisons, which may impact the statistical validity of the results. It is possible that more mice are needed to reach statistical significance, and indeed may allow for some statistical comparisons between WT and PRN-KO mice in terms of dissemination to the spleen at day 7 and 14. Animal numbers and error bars (SD? SEM?) should be described in the figure legends.

2. The statement that natural B. pertussis infections begin in the nasopharyngeal region, progress slowly during a prolonged catarrhal stage and do not normally even reach the lungs (lines 63-65). Should be deleted or supported by primary references. The natural progression of B. pertussis colonization in humans is unknown and has not been studied. The data that’s available from human autopsy cases and NHP studies suggests B. pertussis penetrates deeply into the airway.

3. Line 235: “Bb WT inoculated mice shed significant more bacteria than BbΔprn inoculated mice” cite Fig. 4

2. Line 247-249: “In addition, Bb WT showed significantly higher intracellular survival in macrophages compared with the PRN deletion mutant in an in vitro assay” cite Fig. 6

3. Line 255-257: “Thus, increased intracellular survival may result in a stronger immune response against Bb WT than BbΔprn, consistent with the histopathology analysis in this study” cite Fig. 5

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

Figure Files:

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, . PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at .

Data Requirements:

Please note that, as a condition of publication, PLOS' data policy requires that you make available all data used to draw the conclusions outlined in your manuscript. Data must be deposited in an appropriate repository, included within the body of the manuscript, or uploaded as supporting information. This includes all numerical values that were used to generate graphs, histograms etc.. For an example see here on PLOS Biology: http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001908#s5.

Reproducibility:

To enhance the reproducibility of your results, PLOS recommends that you deposit laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see

24 Feb 2021

Submitted filename: Response to reviewers.docx

Click here for additional data file.

22 Mar 2021

Dear Mr. Ma,

Thank you very much for submitting your manuscript "Pertactin contributes to shedding and transmission of Bordetella bronchiseptica" for consideration at PLOS Pathogens. As with all papers reviewed by the journal, your manuscript was reviewed by members of the editorial board and by several independent reviewers.

While there is overall enthusiasm for this study there still remains a number of outstanding issues. These will need to be addressed in full prior to publication being considered.

In particular, the lack of mechanistic data to explain the role of PRN in inflammation. The reviewers had suggested assessing immune responses and inflammasome activation induced by WT versus PRN- mutant bacteria to explain the higher inflammation with the WT bacteria and although you have looked at immune cell recruitment in the respiratory tissue, which is useful, this is very superficial and not a readout of immune response. The inflammasome activation theory has not been supported by any data. Indeed, the new data suggests that the mutant bacteria induce more, not less, IL-18 production from macrophages which you suggest may be due to the cell type chosen. If this is the case then perhaps an alternative cell type should be considered.

Secondly the lack of data with purified PRN. In your response you mention an experiment to look at inflammasome activation via IL-1 and IL-18 production in the reply to reviewers but show no data because the experiment didn’t seem to work. This is may be because two signals for inflammasome activation have not been used. It will be important to definitively prove or disprove the theory on inflammasome activation.

Overall the insights into the biological action of pertactin are interesting but it's difficult to draw conclusions about the role of pertactin in B. pertussis infections since the rapid loss of pertactin from circulating strains indicates it is not required for infection or transmission of B. pertussis within the human population.

Finally, while it is agreed that models do not have to recreate every aspect of a disease system to be useful, but in order for a model to be useful, it’s strengths and weaknesses have to be considered and discussed. In the introduction for example the significant differences between B. bronchiseptica and B. pertussis infections are somewhat glossed over while emphasizing the similarities between the two strain.

We cannot make any decision about publication until we have seen the revised manuscript and your response to these comments. Your revised manuscript is also likely to be sent to reviewers for further evaluation.

When you are ready to resubmit, please upload the following:

[1] A letter containing a detailed list of your responses to the review comments and a description of the changes you have made in the manuscript. Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

[2] Two versions of the revised manuscript: one with either highlights or tracked changes denoting where the text has been changed; the other a clean version (uploaded as the manuscript file).

Important additional instructions are given below your reviewer comments.

Please prepare and submit your revised manuscript within 60 days. If you anticipate any delay, please let us know the expected resubmission date by replying to this email. Please note that revised manuscripts received after the 60-day due date may require evaluation and peer review similar to newly submitted manuscripts.

Thank you again for your submission. We hope that our editorial process has been constructive so far, and we welcome your feedback at any time. Please don't hesitate to contact us if you have any questions or comments.

Sincerely,

Rachel M McLoughlin, PhD

Associate Editor

PLOS Pathogens

David Skurnik

Section Editor

PLOS Pathogens

Kasturi Haldar

Editor-in-Chief

PLOS Pathogens

​orcid.org/0000-0001-5065-158X

Michael Malim

Editor-in-Chief

PLOS Pathogens

orcid.org/0000-0002-7699-2064

***********************

Dear Authors

While there is overall enthusiasm for this study there still remains a number of outstanding issues. These will need to be addressed in full prior to publication being considered.

In particular, the lack of mechanistic data to explain the role of PRN in inflammation. The reviewers had suggested assessing immune responses and inflammasome activation induced by WT versus PRN- mutant bacteria to explain the higher inflammation with the WT bacteria and although you have looked at immune cell recruitment in the respiratory tissue, which is useful, this is very superficial and not a readout of immune response. The inflammasome activation theory has not been supported by any data. Indeed, the new data suggests that the mutant bacteria induce more, not less, IL-18 production from macrophages which you suggest may be due to the cell type chosen. If this is the case then perhaps an alternative cell type should be considered.

Secondly the lack of data with purified PRN. In your response you mention an experiment to look at inflammasome activation via IL-1 and IL-18 production in the reply to reviewers but show no data because the experiment didn’t seem to work. This is may be because two signals for inflammasome activation have not been used. It will be important to definitively prove or disprove the theory on inflammasome activation.

Overall the insights into the biological action of pertactin are interesting but it's difficult to draw conclusions about the role of pertactin in B. pertussis infections since the rapid loss of pertactin from circulating strains indicates it is not required for infection or transmission of B. pertussis within the human population.

Finally, while it is agreed that models do not have to recreate every aspect of a disease system to be useful, but in order for a model to be useful, it’s strengths and weaknesses have to be considered and discussed. In the introduction for example the significant differences between B. bronchiseptica and B. pertussis infections are somewhat glossed over while emphasizing the similarities between the two strain.

Figure Files:

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, . PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at .

Data Requirements:

Please note that, as a condition of publication, PLOS' data policy requires that you make available all data used to draw the conclusions outlined in your manuscript. Data must be deposited in an appropriate repository, included within the body of the manuscript, or uploaded as supporting information. This includes all numerical values that were used to generate graphs, histograms etc.. For an example see here on PLOS Biology: http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001908#s5.

Reproducibility:

To enhance the reproducibility of your results, PLOS recommends that you deposit laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see

15 Jun 2021

Submitted filename: Response to reviewers.docx

Click here for additional data file.

21 Jun 2021

Dear Mr. Ma,

We are pleased to inform you that your manuscript 'Pertactin contributes to shedding and transmission of Bordetella bronchiseptica' has been provisionally accepted for publication in PLOS Pathogens.

Before your manuscript can be formally accepted you will need to complete some formatting changes, which you will receive in a follow up email. A member of our team will be in touch with a set of requests.

Please note that your manuscript will not be scheduled for publication until you have made the required changes, so a swift response is appreciated.

IMPORTANT: The editorial review process is now complete. PLOS will only permit corrections to spelling, formatting or significant scientific errors from this point onwards. Requests for major changes, or any which affect the scientific understanding of your work, will cause delays to the publication date of your manuscript.

Should you, your institution's press office or the journal office choose to press release your paper, you will automatically be opted out of early publication. We ask that you notify us now if you or your institution is planning to press release the article. All press must be co-ordinated with PLOS.

Thank you again for supporting Open Access publishing; we are looking forward to publishing your work in PLOS Pathogens.

Best regards,

Rachel M McLoughlin, PhD

Associate Editor

PLOS Pathogens

David Skurnik

Section Editor

PLOS Pathogens

Kasturi Haldar

Editor-in-Chief

PLOS Pathogens

​orcid.org/0000-0001-5065-158X

Michael Malim

Editor-in-Chief

PLOS Pathogens

orcid.org/0000-0002-7699-2064

***********************************************************

Reviewer Comments (if any, and for reference):

20 Jul 2021

Dear Mr. Ma,

We are delighted to inform you that your manuscript, "Pertactin contributes to shedding and transmission of Bordetella bronchiseptica," has been formally accepted for publication in PLOS Pathogens.

We have now passed your article onto the PLOS Production Department who will complete the rest of the pre-publication process. All authors will receive a confirmation email upon publication.

The corresponding author will soon be receiving a typeset proof for review, to ensure errors have not been introduced during production. Please review the PDF proof of your manuscript carefully, as this is the last chance to correct any scientific or type-setting errors. Please note that major changes, or those which affect the scientific understanding of the work, will likely cause delays to the publication date of your manuscript. Note: Proofs for Front Matter articles (Pearls, Reviews, Opinions, etc...) are generated on a different schedule and may not be made available as quickly.

Soon after your final files are uploaded, the early version of your manuscript, if you opted to have an early version of your article, will be published online. The date of the early version will be your article's publication date. The final article will be published to the same URL, and all versions of the paper will be accessible to readers.

Thank you again for supporting open-access publishing; we are looking forward to publishing your work in PLOS Pathogens.

Best regards,

Kasturi Haldar

Editor-in-Chief

PLOS Pathogens

​orcid.org/0000-0001-5065-158X

Michael Malim

Editor-in-Chief

PLOS Pathogens

orcid.org/0000-0002-7699-2064