Potential Contributions of Anaerobes in Cystic Fibrosis Airways

Cystic fibrosis (CF) is the most common, lethal genetic disease among the Caucasian population. The leading cause of mortality is recurrent acute exacerbations resulting in chronic airway inflammation and subsequent downward progression of pulmonary function. Traditionally, these periods of clinical deterioration have been associated with several principal pathogens. However, a growing body of literature has demonstrated a polymicrobial lower respiratory community compromised of facultative and obligate anaerobes.

ABSTRACT

Cystic fibrosis (CF) is the most common, lethal genetic disease among the Caucasian population. The leading cause of mortality is recurrent acute exacerbations resulting in chronic airway inflammation and subsequent downward progression of pulmonary function. Traditionally, these periods of clinical deterioration have been associated with several principal pathogens. However, a growing body of literature has demonstrated a polymicrobial lower respiratory community compromised of facultative and obligate anaerobes. Despite the understanding of a complex bacterial milieu in CF patient airways, specific roles of anaerobes in disease progression have not been established. In this paper, we first present a brief review of the anaerobic microorganisms that have been identified within CF lower respiratory airways. Next, we discuss the potential contribution of these organisms to CF disease progression, in part by pathogenic potential and also through synergistic interaction with principal pathogens. Finally, we propose a variety of clinical scenarios in which these anaerobic organisms indirectly facilitate principal CF pathogens by modulating host defense and contribute to treatment failure by antibiotic inactivation. These mechanisms may affect patient clinical outcomes and contribute to further disease progression.

INTRODUCTION

Cystic fibrosis (CF), a multisystem disorder, is the most common and fatal genetic disease among the Caucasian population. Despite the multitude of symptoms that are present with the disease, the hallmark characteristic of adult CF patients is the clinical manifestation affecting lung function. In healthy individuals, mucociliary clearance functions to constantly clear mucus from the lower respiratory tract upwards past the bronchi, keeping the lower respiratory tract effectively sterile. However, those with CF have impaired clearance and deficient innate immune responses in the lower respiratory tract, leading to chronic airway colonization and associated inflammation. Respiratory infections in CF patients begin in infancy, with earlier infections related to subsequent declining lung function (1). Lung function in CF patients follows a downward progression during the individual’s lifetime, with alternating periods of clinical stability and acute exacerbations. These exacerbations are typically brought on by overreactive inflammation by the host, presumably in response to changing microbiology. The resolution of these exacerbations is achieved through aggressive clinical intervention, including antibiotic therapy, often with limited efficacy as the disease progresses. Pulmonary failure, as a result of chronic respiratory infections and inflammation, is the leading cause of death in CF patients (2, 3).

While CF respiratory microbiology traditionally focuses on several specific pathogens, studies over the last 2 decades have demonstrated the lower respiratory tract in CF patients is colonized by a complex polymicrobial community (1, 4). In addition to traditional pathogens, this community is comprised largely of facultative and obligate anaerobes that normally reside within the oral and upper respiratory tract. Within the thick mucus secretions characteristic of CF lungs, hypoxic and anaerobic niches exist that support the growth of anaerobic species. Despite the emerging knowledge of the complex polymicrobial milieu of CF patient airways, the specific role of anaerobes in disease progression has not been carefully examined. The focus of this review will be to summarize recent literature as well as highlight avenues of research required to further discern these relationships.

CF MICROBIOLOGY AS A POLYMICROBIAL COMMUNITY

Traditionally, a narrow spectrum of organisms has been associated with CF airway infections. These include Pseudomonas aeruginosa, Staphylococcus aureus, Haemophilus influenzae, and the Burkholderia cepacia complex (1). Emerging pathogens such as Stenotrophomonas maltophilia, Achromobacter xylosoxidans, and nontuberculosis mycobacteria (NTM) have also been implicated (1). The vast majority of CF lower airway infections are now established as polymicrobial, with the organisms present requiring substantially different growth and nutritional conditions for culturing (5). Both culture-dependent and culture-independent molecular approaches have been used to characterize the microbiology of CF airways (3, 4, 6–11).

This paradigm shift in viewing CF infection from single principal pathogenic organisms to complex polymicrobial community dynamics has moved to the forefront of CF microbiology. One of the first studies to address this notion was by Rogers et al. (11), using terminal-restriction fragment length polymorphism (T-RFLP) analysis to characterize the diversity of bacterial communities in 71 CF sputum samples. They identified the mean number of bands (each representing a specific bacterial species) per CF patient was 13.4 (±6.7), which included 8 predicted anaerobes (Bacteroides gracilis, Eubacterium brachy, Mycoplasma salivarium, Porphyromonas sp., Porphyromonas salivae, Prevotella melaninogenica, Prevotella species oral clone, and Veillonella atypica). Clinically, this was the first study to highlight a diverse and complex polymicrobial community in an area of the body previously thought to be sterile in the healthy individual under normal circumstances. This original culture-independent molecular method has been replaced with partial 16S amplicon sequencing to profile these microbial communities, reproducing earlier findings with greater sensitivity (12, 13). Importantly, and often overlooked, is that these culture-independent approaches are supported by extensive culture data from numerous groups demonstrating viable and robust populations of anaerobes, often at cell densities (defined using number of CFU, CFU per gram, or milliliter of sputum or bronchoalveolar lavage sample) comparable to those of traditional CF pathogens (6, 14–16).

WHO ARE THE PLAYERS?

The development of molecular detection techniques has made the identification of the CF microbiome, at least for bacteria, routine. Notably, molecular techniques over the last decade have highlighted the CF microbiome as a diverse polymicrobial community (12, 13). These molecular methods are not without limitations and have been less insightful into disease progression (17). Anaerobes exist within the lower respiratory tract as a function of the surrounding environment, making it opportune for metabolism and growth. The oxygen gradient within the CF airways is heterogeneous, in part due to diffuse areas of bronchiectasis and mucus plugging (18, 19). The environment is further influenced by principal pathogens and changing metabolism, as observed with P. aeruginosa, and the formation of biofilms with subsequent nitrate utilization that supports anaerobic growth (19). As a result of infection, neutrophils infiltrate the environment with the breakdown of surrounding cellular content, consequently contributing DNA and other debris to increase the volume of sputum plugs, thereby perpetuating a vicious inflammatory cycle. The relative deficit in oxygen, as well as reduced mucociliary clearance, provides a niche for anaerobic bacteria to thrive.

Quantitative PCR methodology (20) and culture approaches (6, 14–16) have demonstrated the prevalence of anaerobes across patients to be similar to traditional CF pathogens in sputum samples, accounting for 46% of the molecular signatures across 6 patients using culture-dependent techniques (6). Common anaerobic isolates from CF airways are listed in Table 1. Prevotella are aerotolerant anaerobes that are common in the oral and upper respiratory tract, representing the most prevalent and numerically abundant anaerobes in the CF airways. Bronchoalveolar lavage fluid analysis using culture-independent methods supports these results, with the identification of several anaerobic and facultative anaerobic species, primarily Prevotella and Streptococcus species (21). Tunney et al. (16) recovered anaerobic species from CF sputum using agar plates under anaerobic conditions in significant quantities (defined as 10⁴ ≤ n ≤ 9 × 10⁷ CFU/g of sputum), including 14 genera of obligate anaerobes from 42 of 66 sputum samples (64%), representing 33 of 50 patients (66%). These included bacteria from the genera Prevotella, Veillonella, Propionibacterium (Cutibacterium), and Actinomyces, with isolated organisms in CFU numbers (CFU/ml) comparable to those of P. aeruginosa (10⁴ to 10⁷). Interestingly, patients colonized with P. aeruginosa were significantly more likely to have recovery of anaerobic species from sputum (1). Subsequent studies have shown the most common genera identified to be Prevotella, Veillonella, Propionibacterium, Peptostreptococcus, and Clostridium (22). Recently, a core CF airway microbiome from an analysis of 299 patients was identified, with anaerobic species representing 5 of 13 species (Prevotella, Veillonella, and Porphyromonas) (23). Another study, focused on Prevotella, found that they represent 70% of anaerobic bacteria by genera isolated from adult sputum samples. In addition, these organisms were isolated from over three-quarters of adult CF patients studied, with an average of one to four Prevotella species per patient and CFU counts in the range of 10⁴ to 10⁸ (15). Finally, a culture-based study identified 17 genera of anaerobic bacteria representing 30% of all genera identified (14).

TABLE 1.

Common obligate anaerobes recovered from CF airways^a

Genus	Species
Prevotella	P. melaninogenica, P. denticola, P. oris, P. histicola, P. pallens, P. tannerella, others
Veillonella	V. atypica, V. parvula, others
Fusobacterium	F. nucleatum
Propionibacterium (now Cutibacterium)	P. acnes, P. acidifaciens
Porphyromonas	P. gingivalis, others
Atopobium	A. parvulum

^aOnly the more prevalent anaerobes from CF airways are listed.

In recent years, there has been considerable effort to characterize the healthy lung microbiome (24). Numerous studies have demonstrated that the lower airways in healthy individuals are not sterile and harbor a bacterial community that reflects the oropharynx (most similar to the supraglottic microbiota) and is likely deposited by microaspiration; however, there is conflicting evidence concerning the longitudinal preservation of these communities (25). The similar bacterial compositions between the healthy lung and oropharynx communities and the overall low bacterial load are consistent with transient but frequent colonization of the lower airways. The lungs in chronic airways disease such as CF contain many of the same species. However, their composition (relative proportions of each species) is distinct, with different species thriving, indicating active growth in the lungs. Moreover, the bacterial load of these organisms (independent of the traditional pathogens) can be several orders of magnitude greater than those measured in non-CF patients (10⁵ to 10⁷ versus 10⁴ CFU/g) (26).

Analyses of CF airway samples have consistently demonstrated the repeated isolation of anaerobic microbes from several time points within and among patients, suggesting the persistence of these organisms rather than merely as transient passengers (27). Furthermore, the abundance of anaerobic organisms found using culture-independent (by copies of amplicon) methods and by quantitative culture demonstrates similar abundances to principal pathogens. Often a limitation of anaerobic culture comments on the natural environment predicated to reside within the oropharynx, suggesting that anaerobic organisms identified in CF sputum samples are a contaminant (21). Nevertheless, culture-independent analysis comparing mouthwash and sputum samples shows various compositions of anaerobes, implying they do truly originate from the lower respiratory tract (8). Several studies have further demonstrated that bacteria from oral secretions and sputum specimens are dissimilar, highlighting anaerobes in CF as unlikely to be purely contamination but rather true members of the lower respiratory tract microbiome (16, 27, 28). The relatively low number of anaerobic organisms in throat swabs compared to sputum further implies proliferation in the lower respiratory tract (28).

ANAEROBES AND CF DISEASE?

A well-defined role for anaerobes in disease progression in CF patients (or other chronic airway diseases) has not been firmly established. As noted above, there is ample data demonstrating that these organisms can colonize the airways of CF patients. Despite this, the evidence in the literature on how anaerobes may be contributing in disease states is controversial. Fodor et al. assessed patient sputum samples pre- and postacute exacerbation with the observation of highly similar microbial composition between these time points, suggesting exacerbation represents intrapulmonary spread of infection rather than changes in the microbial community (26). Zemanick et al. found anaerobes within sputum at times of clinical pulmonary exacerbation were associated with reduced inflammation and improved lung function compared to those sputum samples containing P. aeruginosa (29). Subsequently, the question of the potential role of anaerobes within CF disease progression was highlighted as an “innocent bystander” as opposed to a true pathogen. The altered lung environment in CF airways, specifically the anaerobic pockets that form within mucous secretions, form a niche in which these organisms may fill, and subsequent growth within the airways may be evidence of contribution to disease. The growth of anaerobes may be a marker of disease progression without contributing to disease. However, there are a number of postulated mechanisms by which these organisms could play a role in disease progression (Fig. 1). While these are not necessarily exclusive to obligate anaerobes, the examples presented here will focus on the obligate anaerobes commonly found in CF airways.

FIG 1.

Mechanisms by which anaerobes may contribute to disease progression in cystic fibrosis. (a) Mechanisms by which anaerobes may promote disease progression include directly as pathogens, since many of these anaerobes are frequently associated with infections at other body sites and may have pathogenic potential (e.g., Prevotella sp. and Fusobacterium sp.). (b to d) They may contribute more indirectly by degradation of antibiotics directed against the primary pathogens (b), by acting as synergens that lead to enhanced virulence gene expression on the primary pathogens (c), and by attenuating or dysregulating host defense (d). Mechanisms by which anaerobes may attenuate disease progression (green) include direct competition with the pathogen for resources or production of antimicrobial activities against the primary CF pathogens (e) and stimulating host antimicrobial response (f). We postulate that many of the properties will be species- or strain-specific phenotypes and not necessarily a generic feature of anaerobes. These activities will fluctuate with the bacterial load and growth of the specific strains.

ANAEROBES AS PATHOGENS

The concept of what a pathogen is is complex, and in a simple generalization, two types of pathogens are typically described: primary (disease causing in healthy individuals) and opportunistic (disease causing in those who are compromised in innate or humoral immunity) (30). Commensal organisms are those that reside within the host, are generally considered benign, and are thought to be protective against invading pathogen colonization (31). However, there are several examples where commensal organisms are virulent toward the host (31), and a more complex view of commensals is required in the context of polymicrobial communities (4). The anaerobes found in CF airways are frequently found in polymicrobial infections at other body sites, particularly soft-tissue infections.

Anaerobic organisms have a variety of virulence factors in vitro that may play a role in clinical disease. Studies on Prevotella intermedia and Prevotella nigrescens have demonstrated the production of moderate protease activity (32). In CF, protease activity has been the focus of inquiry, as these enzymes break down connective tissues and molecules associated with host defense mechanisms, thereby providing an avenue for bacterial colonization and growth. P. intermedia has several proteases, including peptidase IV and cysteine protease (33). P. intermedia produces interparin A, a cysteine protease with several pathogenic properties. Primarily, interparin A has been associated with complement inhibition by degrading C3, affecting the subsequent downstream anti-inflammatory cascade as a defense mechanism and promoting bacterial biofilm formation. Interparin A can activate C1 complex in serum, causing the deposition of C1q and promotion of local inflammation (34). Taken together, these mechanisms could aid in evading both the classical and alternative pathways of complementation.

Pathogenicity by anaerobic species also has been implied by identifying specific host inflammatory responses. CF patients have been shown to have circulating antibodies against P. intermedia compared to healthy controls, suggesting a host response and immunogenicity (35). Finally, P. intermedia supernatant cells were observed to have direct cytotoxicity in respiratory epithelial cell line models, directly toxic to human-derived neutrophil cells and associated with increasing inflammatory cell recruitment into mouse lung tissue after inoculation (35).

ANTIBIOTIC RESISTANCE AND ANAEROBES

CF patients and their individual polymicrobial lung communities are under constant selection, given the numerous inhaled and systemic antibiotics prescribed in periods of acute exacerbation as well as maintenance chronic therapy. Pharmacokinetics plays a major role, especially in CF patients. Currently, there is no consensus on optimal antibiotic selection toward anaerobes in this population. Prior work following the identification of anaerobes in high abundance speculated that these organisms have been selected for after recurrent treatment, citing inherent aminoglycoside resistance as a plausible survival mechanism (14). However, antibiotic resistance mechanisms that act by drug inactivation may reduce the effectiveness of these therapies toward other bacteria and contribute indirectly to disease progression.

Prevotella species are one of the most frequent anaerobes isolated from CF samples, and many species are known to secrete potent beta-lactamases. Field et al. (15) demonstrated that 100% of P. melaninogenica (n = 9) and 75% of Prevotella histicola (n = 4) isolates from adult sputum produced in vitro produce these beta-lactamases, whereas prior study results ranged from 25 to 50% (36). This higher level of beta-lactamase production observed is likely a consequence of frequent exposure and may contribute to an overall antibiotic resistome in the CF patient (15). Prevotella species producing beta-lactamase were able to attenuate the killing of P. aeruginosa by ceftazidime in vitro (37), an often-utilized antimicrobial drug in CF exacerbation. Resistance mechanisms that inactivate the antibiotic can reduce the effective antibiotic concentrations in their environment such that other susceptible organisms nearby can survive.

POLYMICROBIAL INTERACTIONS THAT STIMULATE VIRULENCE GENE EXPRESSION IN PATHOGENS

Polymicrobial interactions in infections have not been extensively examined, but previous studies have demonstrated that the virulence of Pseudomonas aeruginosa can be increased in the presence of commensal bacteria (specific strains of Streptococcus, Staphylococcus, Neisseria, former Propinibacterium acnes [now Cutibacterium acnes], Actinomyces, and Rothia) isolated from CF sputum (38, 39). Coculturing or coinfection of these bacteria with the pathogen increased the expression of several P. aeruginosa virulence genes in vitro (38) and in vivo in animal models (39). These CF isolates were avirulent in monoinfection but enhanced P. aeruginosa pathogenesis and were subsequently described as synergens (4). Synergens may be defined as organisms not considered virulent on their own but rather enhance the virulence of pathogens they are found with in polymicrobial infections (4). Obligate anaerobes were excluded from the studies cited for reasons of experimental design. However, one mechanism of action of synergens is through the production of the cell-cell signaling metabolite AI-2, which is predicted to be produced by most anaerobes found in the CF airways and has been demonstrated for Prevotella isolates from CF patients (15). The coinfection of P. aeruginosa with the anaerobe Veillonella parvula in a murine in vivo model demonstrated more severe disease than P. aeruginosa monoinfection, and, notably, the recovery of P. aeruginosa following coinfection was significantly higher in abundance than that with monoinfection by number of CFU/gram (P = 0.016) (40). Metabolites such as 2,3-butanediol, produced by fermentation by a variety of bacteria, have been shown to increase virulence gene expression in P. aeruginosa (41), although this study was not done with obligate anaerobes. Fermentation products may also provide a nutritional resource to P. aeruginosa (42). P. aeruginosa virulence gene expression has also been shown to be induced in the presence of cell wall fragments from Gram-positive bacteria (43).

ATTENUATION OF HOST DEFENSE PATHWAYS

Bacteria produce proteins and metabolites that may impair immune responses and indirectly facilitate the growth of pathogens. Proteases degrade both cytokines and chemokines as well as cell surface receptors, thereby contributing to immune dysregulation. Degradative enzymes may also increase nutrient availability to other neighboring bacteria, indirectly supporting growth. This could include the liberation of free amino acids and sugars from the degradation of glycoproteins as well as the release of iron from intracellular sources, all of which may serve as nutritional reservoirs for the production of virulence factors.

Several Prevotella species produce exoenzymes involved in hemolysis and hemagglutination. Interparin A from P. intermedia breaks down hemoglobin with by-products that are anti-inflammatory within in vitro epithelial cells (44). P. melaninogenica mediates hemolysis through a lectin-binding protein that binds to erythrocytes to promote subsequent degradation. While first described in P. melaninogenica, similar mechanisms have been observed in P. intermedia and P. gingivalis (45). Clinically, this is relevant as a proinflammatory pathway, but there may be substantial nutritional benefit to the anaerobic organisms. Finally, short-chain fatty acids produced by anaerobes upregulate the expression of a short-chain fatty acid receptor, GPR41, subsequently upregulating interleukin-8 (IL-8), which may exacerbate inflammation in the host (46). It also should be noted that gut-derived short-chain fatty acids are important mediators of the gut-lung axis and contribute to healthy lung development (47).

A ROLE FOR ANAEROBES IN ATTENUATING DISEASE PROGRESSION

The emphasis in the preceding sections was on the potential role of anaerobes in promoting disease progression by direct and indirect processes. It is also possible that these organisms have beneficial effects leading to the attenuation of disease progression. Indeed, Zemanick et al. observed an inverse correlation between anaerobes and the severity of pulmonary exacerbations, whereby increased anaerobe levels were associated with reduced inflammation and improved lung function (29). Given that the bacterial load of anaerobes can be on par with that of the primary pathogens, competition for limiting resources will likely occur and may limit the proliferation of the pathogen under some circumstances; however, this has been under limited exploration. Furthermore, bacteria have evolved numerous mechanisms to compete with one another. These include cell contact-mediated killing via type VI and type VII secretion systems in Gram-negative and Gram-positive bacteria, respectively (48). Antimicrobials, including bacteriocins and related molecules produced (48) by these organisms, could inhibit pathogen growth, although this has been underexplored in the context of CF. Another mechanism by which anaerobes may reduce active disease is through immune modulation. While these organisms may contribute to immune dysregulation, some level of immune stimulation may increase innate defense to help contain pathogens. Inflammation is a double-edged sword, and acute exacerbations are characterized by overt immune activation. At the same time, during periods of relative stability the immune system does manage to contain large populations of bacteria within the lungs. Commensal organisms from the oropharynx and nasal mucosa, the source of most of the microbiota in the CF airways, play a role in immune balance at these sites and may contribute some beneficial immunomodulatory activity in the lower airways of chronic disease cases.

CONCLUSIONS

The role of anaerobes in CF disease pathogenesis is complex. While the prevalence of anaerobic species in CF lower airways has been well established, there is still uncertainty on their role in progressive lung disease. The anaerobes more commonly found in CF airways represent very different taxonomic groups, and a role for these in disease is most likely to be species or strain specific rather than a generic feature of anaerobes. The production of virulence factors and synergistic relationships suggests a degree of pathogenicity, but the extent to which it contributes to disease is unclear. It is evident that, with greater insight and knowledge into the microbial diversity of CF patients, integrating polymicrobial pathways into the design of optimal management of the complex microbiology would improve patient outcomes. Better characterization of species and strains may help identify the presence of potentially pathogenic or synergistic from more benign or avirulent strains. Many of the studies currently in the literature are in older patients who have had longer, complex infectious courses and may confound the role of anaerobes (49). There is a need for longitudinal studies in pediatric CF patients, before the establishment of chronic infection, to truly understand the natural history of anaerobic organisms in CF (49). Better methodologies rather than descriptive taxonomy of these polymicrobial communities are also needed to improve our understanding of airway infections and to develop treatment strategies that address mechanisms of infection that include more than just the primary pathogens. Ultimately, the clinician and researcher both need to be vigilant about treatment against unconventional pathogens and synergistic infections in CF given the ever-changing landscape of these complex environmental communities.