Community-associated methicillin-resistant Staphylococcus aureus immune evasion and virulence

Abstract

Staphylococcus aureus is a significant cause of human infections globally. Methicillin-resistant S. aureus (MRSA) emerged in the early 1960s and is now endemic in most healthcare facilities. Although healthcare-associated MRSA infections remain a major problem in most industrialized countries, those caused by community-associated MRSA (CA-MRSA) are now the most abundant cause of bacterial infections in the community in some parts of the world, such as the United States. The basis for the emergence and subsequent success of CA-MRSA is incompletely defined. However, the ability of the pathogen to cause disease in otherwise healthy individuals is likely attributed, in part, to its ability to circumvent killing by the innate immune system, which includes survival after phagocytosis by neutrophils. In this review, we discuss the role of neutrophils in host defense against S. aureus and highlight progress made toward understanding mechanisms of CA-MRSA virulence and pathogenesis.

Introduction

Staphylococcus aureus is a leading cause of bacterial infections worldwide. The pathogen has capacity to cause a wide range of syndromes of varied severity. For example, S. aureus is the most abundant cause of skin and soft tissue infections in most industrialized countries, including the United States [1]. Such infections are often relatively minor. On the other hand, S. aureus is a leading cause of bloodstream infections in the healthcare setting, many of which are fatal [2]. Given the high prevalence of S. aureus infections, it is, perhaps, not surprising that ∼30% of non-institutionalized individuals are permanently colonized with the organism [3, 4]. Therefore, S. aureus may be considered a component of normal human flora.

S. aureus is widely known for its ability to rapidly acquire resistance to antibiotics. For instance, although penicillin was initially highly effective in the treatment of staphylococcal infections [5], by the late 1940s, penicillin resistance had increased such that the antibiotic was no longer an effective agent for treatment of S. aureus infections [6]. Similarly, methicillin-resistant S. aureus (MRSA) were reported 2 years after methicillin was introduced for treatment of penicillin-resistant S. aureus infections [7]. MRSA has since spread worldwide and is now endemic in hospitals and healthcare settings of most industrialized countries (reviewed in [8]). Notably, MRSA may be the leading cause of mortality by any single infectious agent in the United States ([2] and Fig. 1 in [9]).

Community-associated MRSA (CA-MRSA) infections were first reported in indigenous populations of Western Australia in the early 1990s [10] and subsequently among children in the Midwestern region of the United States [11, 12]. Unexpectedly, CA-MRSA disseminated rapidly and is now epidemic in the United States—it is the most abundant cause of bacterial infections in the community [13]. In contrast to healthcare-associated MRSA (HA-MRSA) infections, for which there are predisposing risk factors or underlying susceptibilities, CA-MRSA infections occur in otherwise healthy individuals with no exposure to the healthcare setting. Thus, anyone is at risk for infection. These observations imply that CA-MRSA strains have enhanced virulence compared with traditional HA-MRSA strains, and recent studies have addressed this notion in animal infection models or experiments with human leukocytes in vitro (see below). Although the vast majority of CA-MRSA infections are of skin and soft tissue, the pathogen has capacity to cause severe invasive disease [14, 15]. Direct contact with an infected or colonized individual is likely the primary means of transmission (see http://www.cdc.gov/ncidod/dhqp/ar_mrsa_ca_clinicians.html#8).

CA-MRSA strains have a unique methicillin resistance element, known as staphylococcal cassette chromosome mec type IV or V (SCCmecIV or V), whereas traditional HA-MRSA strains contain a larger SCCmec element (types I–III) ([16] and reviewed in [8]). Initially, the presence of genes encoding Panton–Valentine leukocidin (PVL, encoded by lukS-PV and lukF-PV), a two-component cytolytic toxin, and SCCmecIV was associated with emergent CA-MRSA strains worldwide [17]. Although all CA-MRSA strains contain SCCmecIV or V, recent reports indicate some CA-MRSA strains lack genes encoding PVL [18–21]. The molecular basis for enhanced virulence and transmission of CA-MRSA, which includes understanding interaction of this pathogen with the innate immune system, remains an intense area of investigation.

Here, we review our current knowledge of host defense against S. aureus, with emphasis on the enhanced ability of CA-MRSA to circumvent killing by neutrophils, and provide an update on CA-MRSA virulence mechanisms.

Neutrophils and bacterial immune evasion

Neutrophils or polymorphonuclear leukocytes (PMNs) are the most abundant of all leukocytes and are essential for host defense against invading pathogenic microbes. The importance of PMNs in host defense against bacterial pathogens is exemplified by pronounced susceptibility to infection in patients with primary neutrophil defects. For example, individuals with severe congenital neutropenia, leukocyte adhesion deficiency, chronic granulomatous disease, and neutrophil-specific granule deficiency are at increased risk for developing recurrent S. aureus infections [22]. PMNs are terminally differentiated innate immune cells and following maturation in bone marrow are released into circulation. Neutrophils are rapidly recruited to infected tissue through host chemotactic signals delivered by cytokines and chemokines, as well as bacteria-derived products such as N-formyl peptides. PMNs recognize and ingest bacterial pathogens at sites of infection though a process known as phagocytosis, which is facilitated by opsonization of the microbial surface with host serum proteins such as antibody and complement. Phagocytosis of bacteria induces production of microbicidal reactive oxygen species (ROS) such as superoxide radicals, hydrogen peroxide, hydroxyl radicals, and chloramines [23]. In addition to ROS, bacteria are exposed to antimicrobial peptides and proteases following fusion of cytoplasmic granules with pathogen-containing phagosomes [24]. Together, oxygen-dependent and oxygen-independent neutrophil killing mechanisms are sufficient to kill most invading microorganisms. Notwithstanding, bacterial pathogens such as S. aureus and CA-MRSA, in particular, survive following neutrophil phagocytosis and thereby cause disease [25–27]. Inasmuch as PMNs are essential to host defense, S. aureus evasion of the innate immune system represents an important mechanism for pathogenesis. Some of the specific immune evasion strategies employed by S. aureus are considered below.

Resolution of infection

Influx to and accumulation of neutrophils at sites of infection is a critical component of the acute inflammatory response and required for resolution of bacterial disease. Notwithstanding, the degradative enzymes and cytotoxic molecules that contribute to PMN bactericidal activity are potentially harmful to host tissue should the neutrophil lyse. For instance, neutrophils are intimately associated with host inflammatory disorders such as rheumatoid arthritis. Thus, circumspect removal of activated PMNs from infection sites is paramount to the resolution of inflammation [28]. Neutrophils undergo either accelerated apoptosis following phagocytosis to resolve the inflammatory process or a slower spontaneous apoptosis to maintain immune system homeostasis. In either scenario, macrophages are essential to recognize and remove effete PMNs through a process called efferocytosis [28]. Several bacterial pathogens alter this conventional innate immune process as a mechanism for pathogenesis (Fig. 1). For example, the pathogen Anaplasma phagocytophilum inhibits neutrophil apoptosis to facilitate an obligate intracellular lifecycle. By contrast, pathogens such as S. aureus promote neutrophil lysis and thereby survive and cause disease. Although most S. aureus strains eventually cause PMN lysis, it is noteworthy that prominent CA-MRSA strains are particularly adept at inducing rapid neutrophil lysis (Fig. 1) [26]. Thus, the ability of prominent USA300 and USA400 to cause rapid PMN lysis may contribute to increased virulence of these strains.

Fig. 1.

Two possible outcomes of the interaction of PMNs with bacteria. On one hand, bacteria (pink spheres) are ingested, activate PMNs, and are killed, and this process ultimately triggers PMN apoptosis and the removal of these cells by macrophages (Efferocytosis). This phenomenon leads to the resolution of the inflammatory response (RESOLUTION). Alternatively, bacterial pathogens (yellow spheres or red particles) are ingested and either delay normal PMN apoptosis and turnover or cause PMN lysis. Typically, these processes facilitate survival of bacteria and thereby promote disease (DISEASE)

CA-MRSA virulence and immune evasion

S. aureus produces numerous surface molecules and freely secreted toxins that have potential to alter or inhibit function of innate immune cells (reviewed in [29]). In addition to the large repertoire of putative S. aureus immune evasion factors, the significant redundancy of function among many of these factors suggests that no single molecule can uniquely explain the ability of S. aureus to cause human disease. For example, S. aureus has capacity to generate multiple molecules that inhibit neutrophil chemotaxis and complement function, sequester host antibody, promote resistance to antimicrobial peptides or reactive oxygen species, and cause host cell lysis [29]. Here, we focus our discussion on the processes and molecules implicated in the enhanced virulence phenotype of CA-MRSA.

CA-MRSA strains are highly virulent in mouse infection models

S. aureus stains known as USA300 [30, 31] and USA400 [32] are the most abundant cause of CA-MRSA infections in the United States, although the number of USA300 infections now far exceeds those caused by USA400 [13, 33, 34]. Voyich et al. were the first to compare virulence of USA300 (strain LAC) and USA400 (strains MW2 and MnCop) with early or representative HA-MRSA strains, COL, and MRSA252 [26]. The authors found that CA-S. aureus strains, which included CA-MRSA and CA-MSSA (MnCop) strains, were significantly more virulent than either COL or MRSA252 in a mouse sepsis model, and follow-up studies with an extended time course confirmed the original findings [35]. Although neutrophil phagocytosis and activation were similar among all strains tested in vitro, survival of USA300 and USA400 was significantly greater after phagocytosis compared with COL or MRSA252. The molecular basis of enhanced survival of these CA-MRSA strains after phagocytosis is currently under investigation, but, as noted above, it ultimately results in destruction of neutrophils. In addition, the observation that MW2 and MnCop have comparable virulence in animal models suggests that methicillin resistance per se has no impact on virulence. This idea was confirmed in subsequent studies by Diep et al. [36].

Molecules implicated in CA-MRSA virulence

Inasmuch as the genes encoding PVL are present with many CA-MRSA strains, it has been suggested over the past several years that the toxin might be responsible for the emergence, transmission, and/or enhanced virulence of CA-MRSA. This notion seems logical, since PVL is a known virulence factor of S. aureus and unique among the earliest CA-MRSA strains. In addition, a previous pandemic of community-associated infections was caused by a PVL-positive strain known as phage type 80/81 [37]. As a first step toward testing this notion, Voyich et al. constructed USA300 and USA400 isogenic lukS-PV and lukF-PV-negative strains (Δpvl) and compared virulence with the parental wild-type strains in mouse skin infection and sepsis models [35]. Unexpectedly, the wild-type and isogenic Δpvl strains had comparable ability to cause disease in these mouse models of CA-MRSA infection. In addition, survival of these wild-type and Δpvl strains following phagocytosis by human neutrophils was comparable [35]. Bubeck Wardenburg et al. were the first to evaluate wild-type and isogenic Δpvl CA-MRSA strains in a mouse pneumonia model and found comparable mortality between wild-type and Δpvl strains at all time points tested [38]. Since the first studies with USA300 and USA400 strains were published, there have been several other reports investigating the role of PVL in virulence using mouse, rat, or rabbit infection models [39–44]. Conclusions from the majority of studies with rodent models concur [35, 38, 39, 41, 43] or variance can be explained by the model used [39] or an unintended genetic mutation [44, 45], with one exception [40]. The work by Brown et al. [40] used USA300 strains and animal infection models that were either identical or virtually identical to those of Bubeck Wardenburg et al. [43], so, the basis for the differences in experimental outcome is not clear. It is important to note that PVL is epidemiologically linked to certain types of disease, such as severe necrotizing pneumonia and some types of severe skin infection [46, 47]. Therefore, the findings in animal models published to date may simply indicate that PVL has little or no significant role in most CA-MRSA infections but can contribute to some types of severe infection, which are typically infrequent and may involve an underlying human susceptibility factor. That said, one must consider the limitations of animal models in general for predicting the role of virulence factors in human infection, as there is increasing evidence for the host-specific activity of staphylococcal virulence factors [48–50].

There has been concerted effort over the past few years to determine the molecular basis of enhanced virulence and transmission of USA300. During the course of such work, Wang et al. identified novel S. aureus peptides with limited homology to the phenol-soluble modulins (PSMs) of Staphylococcus epidermidis [51]. A subset of these peptides, known as α-type PSMs (PSMα), is ∼20 amino acids and recruits and lyses neutrophils. In accordance with these in vitro data, USA300 and USA400 isogenic Δpsmα deletion strains have significantly reduced virulence in mouse sepsis and skin infection models, indicating PSMα peptides are major determinants of CA-MRSA virulence [51]. It is also notable that PSMα peptides are produced in vitro at significantly higher concentrations by USA300 compared with the most prominent HA-MRSA pulsed-field types, USA100 and USA200 [52]. The relatively high expression of PSMα peptides by USA300 is, in part, related to the high activity of accessory gene regulator (agr) in this strain [52]. An intriguing idea proposed by Li et al. [52] is that the enhanced virulence phenotype of USA300 is based largely on differential expression of core-genome encoded virulence factors such as PSMα peptides and alpha-hemolysin (alpha-toxin, Hla) rather than acquisition of mobile genetic elements [52].

Recent work by Bubeck Wardenburg et al. revealed that Hla plays prominent role in pathogenesis of S. aureus pneumonia [38, 53, 54]. An isogenic hla-deficient strain of USA300 was essentially avirulent in the mouse pneumonia [38, 43] and active or passive immunization against Hla protected animals from death [53]. Although this cytolytic toxin has long been known to be a virulence factor of S. aureus, these studies provide strong support to the idea that Hla is one of the primary determinants of (if not the major determinant) virulence in S. aureus pneumonia. These findings are underscored by recent observations that indicate expression of Hla by USA300 is relatively high compared with other strains of the same clonal complex [52]. Work by Bartlett et al. investigated the mechanism of action of Hla in vivo and found that the toxin elicits host CXC chemokine production in a mouse S. aureus pneumonia model [55]. These chemokines promote influx of neutrophils, which, in turn, may be largely responsible for lung injury during infection. Such a process would be consistent with the host–pathogen model described above, whereby lysis of neutrophils after phagocytosis of S. aureus could contribute to host tissue destruction.

Concluding remarks

S. aureus continues to be a leading cause of bacterial infections in healthcare settings worldwide. However, the recent emergence of highly virulent CA-MRSA strains has changed the epidemiology of severe S. aureus infections such that all populations can be considered at risk. As discussed in this review, it is increasingly clear that prominent CA-MRSA strains are particularly adept at evasion of the innate immune response—a feature that likely contributes to enhanced virulence. Although progress has been made towards identifying factors involved in CA-MRSA virulence, several key questions remain unanswered. For example, the molecular basis of enhanced USA300 transmission remains incompletely determined. Understanding the enhanced transmission phenotype will likely provide new insight into the basis for emergence of epidemic S. aureus clones. Also, our knowledge of the host genetics of susceptibility to severe S. aureus infections is limited and must be addressed. A complete understanding of molecules that impact evolution of epidemic CA-MRSA strains will be essential to guide development of novel therapeutics for prevention and treatment of severe S. aureus disease.