During bacterial infection, professional phagocytes are attracted to the site of infection, where they constitute a first line of host cell defense. Their function is to engulf and destroy the pathogens. Thus, bacteria must withstand the bactericidal activity of professional phagocytes, including macrophages to counteract the host immune system. Bacillus cereus infections are characterized by bacteremia despite the accumulation of inflammatory cells at the site of infection. This implies that the bacteria have developed means of resisting the host immune system. Bacillus cereus spores survive, germinate, and multiply in contact with macrophages, eventually producing toxins that kill these cells. However, the exact mechanism by which B. cereus evades immune attack remains unclear. This review addresses the interaction between B. cereus and macrophages, highlighting, in particular, the ways in which the bacteria escape the microbicidal activities of professional phagocytes.
During bacterial infection, professional phagocytes, such as monocytes, macrophages, dendritic cells, and polymorphonuclear leukocytes, are attracted to the site of infection, where they constitute a first line of host cell defense. Their function is to engulf the infectious agents, internalizing, and destroying them (Underhill & Goodridge, 2012; Fig. 1). The uptake of infectious agents by phagocytic cells involves the binding of the pathogen to receptors on the cell surface. Phagocytes can recognize pathogens directly, through specific motifs called pathogen-associated molecular patterns (PAMPs) or after opsonization, a process in which the pathogen is coated with antibodies or complement factors (Kumar et al., 2011). Interaction of the pathogen with the receptor triggers an intracellular cascade, leading to the reorganization of the actin machinery into a pseudopod that engulfs the pathogen. The internalized pathogens are then held within a vacuole, the phagosome. The phagosome undergoes a series of transformations by sequential fusion with the endosome and lysosome, culminating in the formation of a mature degradative phagolysosome (Fairn & Grinstein, 2012). This organelle has diverse microbicidal weapons, including an acidic environment, oxidative and nitrative responses, and the production of digestive enzymes and several anti-microbial peptides. These elements eventually eliminate the pathogen.
Thus, if pathogens are to survive, to replicate, and to disseminate within the host, they must adapt to the highly hostile environment created by the immune response. Interactions between host and pathogen should not be seen as a static phenomenon, but as an ‘arms race’, in which each opponent tries to respond as effectively as possible (Sarantis & Grinstein, 2012). This has led to pathogens evolving highly sophisticated strategies for overcoming the host immune response. As such, phagocytosis may be seen as an opportunity or an obstacle for microbial pathogens, depending on the lifestyle of the pathogens concerned. Intracellular bacteria have developed many sophisticated strategies for entering phagocytic cells and surviving within them, whereas other bacterial pathogens have evolved mechanisms for preventing phagocytosis or escaping this process.
In this review, we consider an example of the bacterial mechanisms used to counteract the host immune system. We address the interaction between Bacillus cereus and macrophages, focusing particularly on the ways in which the bacteria escape the microbicidal activities of these professional phagocytes.
B. cereus uptake and release by host macrophages
Bacillus cereus belongs to the Bacillus cereus group, which contains seven species of diverse sporulating Gram-positive bacteria (i.e. Bacillus anthracis, Bacillus thuringiensis, and Bacillus cereus). These highly related bacilli are ubiquitous pathogenic bacteria, able to colonize hosts as diverse as insects and mammals. They differ in terms of their plasmid-encoded factors: a capsule and toxins causing anthrax for B. anthracis, and insecticidal crystal proteins for B. thuringiensis (Schnepf et al., 1998; Mock & Fouet, 2001). Apart from the specific genes borne by plasmids, the genomes of the three species, B. anthracis, B. thuringiensis, and B. cereus, are very similar and the genetic determinants required for nonspecies-specific aspects of infection may be common to all the bacteria of the B. cereus group (Ivanova et al., 2003).
Bacillus cereus is an emerging human pathogen initially characterized as a causal agent of gastroenteritis. It is the third most important cause of collective food poisoning infections, after Salmonella and Staphylococcus aureus (Anonymus, 2009). Bacillus cereus food poisoning is generally mild, but bloody diarrhea and emetic poisoning, which may be fatal in some cases, have been reported (Mahler et al., 1997; Lund et al., 2000; Dierick et al., 2005; Decousser et al., 2013). The number of cases of B. cereus foodborne infections is probably largely underreported, as the reporting of such cases is not mandatory. Bacillus cereus is also an opportunistic pathogen causing severe local and systemic infections in humans (Auger et al., 2009; Bottone, 2010; Cadot et al., 2010; Ramarao, 2012). The increasing number of opportunistic infections described highlights the importance of studying this emerging pathogen. The most frequently described conditions are endophthalmitis (Callegan et al., 2006), necrotizing infections (Darbar et al., 2005), endocarditis (Abusin et al., 2008), bacteremia (Hernaiz et al., 2003), osteomyelitis (Schricker et al., 1994), septicemia (Matsumoto et al., 2000), liver abscesses (Latsios et al., 2003), pneumonia, and meningitis, particularly in neonates, leading to death of the infant within days (Miller et al., 1997; Gray et al., 1999; Evreux et al., 2007). These infections are characterized by bacteremia despite the accumulation of inflammatory cells at the site of infection (Hernandez et al., 1998). This implies that the bacteria have developed by means of resisting the host (Bouillaut et al., 2005; Ramarao & Lereclus, 2006; Gilois et al., 2007; Tran et al., 2010; Kamar et al., 2013) and, in particular, the action of the inflammatory cells, enabling them to escape from the host immune system. Macrophages, through their strategic position throughout the body, are key actors in immune surveillance. They play an essential role in the sensing and elimination of invasive microorganisms, but also orchestrate the adaptive immunity. Thus, macrophages form an essential barrier that pathogens must overcome to be successful.
In the case of systemic disease or after crossing the intestinal barrier upon ingestion, it is likely as spore that B. cereus first interacts with the immune cells. Previous studies have shown that B. cereus spores can survive within macrophages, subsequently escaping from this hostile environment (Ramarao & Lereclus, 2005). Spores are composed of structured layers. The outermost layer of B. cereus spores is called the exosporium; it forms a loose balloon-like structure around the spore that may contribute to bacterial resistance to its host (Henriques & Moran, 2007). The metalloprotease InhA1 is secreted into the extracellular medium and is also a major proteinaceous component of the spore exosporium (Charlton et al., 1999). We have previously demonstrated that InhA1 is involved in the escape of B. cereus from macrophages, as B. cereus strains deleted in inhA1 remain transiently blocked within the cell (Ramarao & Lereclus, 2005). Moreover, heterologous production of the protein in Bacillus subtilis is sufficient to promote escape from macrophages. Interestingly, the B. cereus inhA1 mutant can germinate even when trapped within the macrophage, providing support for the notion that escape and germination are independent events. It seems likely that germination begins after uptake, as the bacteria are sensitive to heat even in the presence of an intact external structure. However, complete germination does not seem to be required for escape, as germinating bacteria are found extracellularly and vegetative bacteria are also able to escape from macrophages after engulfment (Ramarao & Lereclus, 2005).
B. cereus-induced cell toxicity
It remains unclear whether B. cereus escapes macrophage by ‘hijacking’ an active cellular process or by causing the lysis of the cells, either through the action of a cytotoxic factor or mechanically, due to the accumulation of growing numbers of intracellular bacteria. The lifestyle of B. cereus suggests it is unlikely to remain intracellular. Once the vegetative bacteria are released from the cell, they must avoid reuptake by the in situ phagocytes to escape the immune response. It is tempting to speculate that InhA1 is involved in the lysis of macrophages, as its production in B. subtilis is sufficient to induce toxicity in macrophages although by an unknown mechanism (Ramarao & Lereclus, 2005). However, a B. cereus inhA1-deficient mutant although avirulent (Guillemet et al., 2010) can nevertheless kill macrophages, indicating that other cytotoxic factors are involved. Pathogens frequently destroy macrophages by inducing apoptosis (Navarre & Zychlinsky, 2000). Apoptosis is a programmed multistep cell death pathway that may be activated in several ways (Elmore, 2007). Physiological apoptosis plays an essential role in development, differentiation, and tissue homeostasis. Apoptosis can also occur as a defense mechanism, when cells are damaged by an external agent. This ability to alter inflammatory responses within phagocytic cells may confer significant advantages on the pathogen. The only B. cereus factor known to induce apoptosis is hemolysin II (HlyII; Tran et al., 2011a). HlyII is an oligomeric β-barrel pore-forming toxin. Other toxins of this group include the α-toxin of S. aureus, β-toxin of Clostridium perfringens, and B. cereus cytotoxin K (CytK; Ramarao & Sanchis, 2013). HlyII forms heptameric transmembrane pores in erythrocytes and artificial membranes (Andreeva et al., 2006, 2007; Andreeva-Kovalevskaya et al., 2008). It induces the apoptosis of host monocytes and macrophages in vivo, in a death receptor-dependent pathway (Tran et al., 2011a). Cell death occurs in two steps: HlyII binds to dendritic cells and/or macrophages and induces the formation of a pore, leading to transient membrane permeability (Tran et al., 2011b). The formation of this pore eventually leads to the induction of apoptosis in the cells. It remains unclear how HlyII interacts with macrophage surfaces to form pores, and it has been reported that HlyII has no specific receptor in erythrocytes (Andreeva et al., 2006). However, there is some evidence for the existence of a specific receptor for HlyII. First, the susceptibility of cells to HlyII depends strongly on cell type. HlyII activity may even be specific within a particular family of cell types, as a previous study reported an effect on Caco2 cells (Andreeva et al., 2006), whereas our data suggest that the purified toxin does not induce permeability or apoptosis in HeLa cells, another type of epithelial cell (Tran et al., 2011a). Moreover, the related α-toxin of S. aureus appears to bind to the host cell receptor phosphocholine (Valeva et al., 2006; Liang & Ji, 2007). This receptor binding may allow the protein to accumulate locally in microdomains enriched in cholesterol and sphingolipids (lipid rafts). This results in high local concentrations, favoring toxin oligomerization and, thus, stable membrane-anchored binding to target host cells, suggesting that certain cell types have high-affinity toxin-binding sites (Valeva et al., 2006). Thus, HlyII may bind to a specific receptor possibly present in a lipid-rich microdomain.
The ability of HlyII to kill macrophages may account for the persistence and dissemination of B. cereus in the host. The induction of apoptosis by B. cereus may cause tissue damage and compromise the antimicrobial immune response, thereby promoting bacterial spread, leading to the signs and symptoms of disease. The importance of HlyII for virulence has been demonstrated in various models (Sineva et al., 2009; Tran et al., 2011a) and by the presence of a gene encoding this protein in several clinical isolates of B. cereus (Cadot et al., 2010).
Iron and glucose regulate expression of the hlyII gene, by activating the regulators Fur and HlyIIR, respectively (Sineva et al., 2012; Guillemet et al., 2013; Tran et al., 2013). Both iron and glucose are crucial for bacterial multiplication and, thus, for the capacity of the bacterium to colonize its host. HlyII probably induces host cell lysis to provide the bacteria with access to nutrients. As a model (Fig. 2), we suggest that when glucose is consumed by the bacteria and iron is sequestered by phagocytic cells as a natural host defense (Ratledge & Dover, 2000; Weinberg, 2009), the HlyIIR and Fur repressors become inactivated and hlyII expression is triggered. HlyII is then produced and secreted by the bacteria, triggering the death of hemocytes and macrophages (Tran et al., 2011a). The contents of the cell are then released into the environment, providing the bacteria with access to nutrients, allowing them to grow, and promoting a new cycle of hlyII gene inhibition/expression.
B. cereus response to ROS/NOS
Bacillus cereus encounters oxidants, including superoxide (), hydrogen peroxide, and nitric oxide (NO), when they germinate and grow in macrophages (MacMicking et al., 1997; Shatalin et al., 2008). The exposure of B. cereus to mild and lethal hydrogen peroxide concentrations modifies the expression of numerous genes (Ceragioli et al., 2010), including those involved in the common response to general stresses, such as groES, dnaK, and clp proteases. Genes encoding catalases, thioredoxin reductases, and peroxidases are also induced to remove hydrogen peroxide from the cells or from the extracellular environment. The induction of perR, iron, and manganese uptake systems suggests that iron and manganese are involved in the response of B. cereus to hydrogen peroxide. Tarasenko et al. (2008) showed that treatment with glycoconjugates increases the intracellular killing of B. cereus spores by inducing a dose-dependent increase in macrophage nitrogen derivatives (NO) production. Finally, the SOS response, which is activated when DNA is damaged, is induced, together with other DNA repair and protection mechanisms, by exposure to hydrogen peroxide, suggesting that exposure to oxygen or nitrogen derivatives leads to protein and DNA damage (Mols & Abee, 2011).
We have consistently shown that mutation frequency decline (Mfd), a bacterial protein known to be involved in DNA repair mechanisms (Savery, 2007), plays a crucial role in bacterial resistance to the host nitrogen response (N. Ramarao, unpublished data). Mfd is essential for specific resistance to the deleterious effects of the nitrogen stress imposed by host phagocytes. Moreover, a B. cereus mfd mutant is avirulent and unable to survive NO stress in vivo in two animal models (insects and mice). In Escherichia coli, Mfd is known to be required for DNA repair, after UV irradiation, for example. However, Mfd had never been implicated in bacterial pathogenicity or NO responses.
As Mfd is widely conserved in the bacterial kingdom, these data highlight a novel mechanism that may be used by a large spectrum of bacteria to overcome the host immune response, including the mutagenic properties of reactive nitrogen species. This might make it possible to develop new, potentially universal antimicrobial strategies.
Technological advances in the last decade have facilitated studies on the mechanisms of interaction between the pathogen and its host in the context of infection. During the establishment of infection, bacteria are confronted and must deal with the immune response of their host. Macrophages are one of the first actors in immunity. In addition to their sentinel activities, they degrade pathogens by phagocytosis and activate the appropriate immune response. The fine balance between the macrophage response and the ability of the pathogen to modulate the cellular response determines the outcome of infection, and the number of known examples of bacteria duping the immune system of their host is growing.
The interplay between pathogen and host should certainly not be seen as static. Although similar strategies are frequently used, the ways in which bacteria combine them, the timing of their use and their targets differ with the situation. Bacillus cereus has a genetic background very similar to that of B. anthracis, another pathogen of the B. cereus group. Many studies have investigated the way in which B. anthracis escapes the immune system. However, although the results of these studies can be used to guide similar research for B. cereus, it would be dangerous to assume that the immune evasion strategies of these two species are the same. On the cell side, it is essential to avoid the concept of a ‘bad’ or ‘good’ cellular response to the pathogen. The reality is far less black and white and seems to involve a subtle fine-tuning of a combination of connected cellular responses. This highlights the importance of using an appropriate model to study host–pathogen interactions.
Finally, the conversion of some of these advances into a true understanding of disease should make it possible to identify weak points in the immune response that could be corrected and the ‘Achilles’ heel’ of the pathogen, which could then be targeted by treatment. This next step promises to be a great challenge.