Staphylococcus aureus Colonization of the Human Nose and Interaction with Other Microbiome Members

ABSTRACT

Staphylococcus aureus is usually regarded as a bacterial pathogen due to its ability to cause multiple types of invasive infections. Nevertheless, S. aureus colonizes about 30% of the human population asymptomatically in the nares, either transiently or persistently, and can therefore be regarded a human commensal as well, although carriage increases the risk of infection. Whereas many facets of the infection processes have been studied intensively, little is known about the commensal lifestyle of S. aureus. Recent studies highlight the major role of the composition of the highly variable nasal microbiota in promoting or inhibiting S. aureus colonization. Competition for limited nutrients, trace elements, and epithelial attachment sites, different susceptibilities to host defense molecules and the production of antimicrobial molecules by bacterial competitors may determine whether nasal bacteria outcompete each other. This chapter summarizes our knowledge about mechanisms that are used by S. aureus for efficient nasal colonization and strategies used by other nasal bacteria to interfere with its colonization. An improved understanding of naturally evolved mechanisms might enable us to develop new strategies for pathogen eradication.

INTRODUCTION

The human body offers several distinct niches for specific microbiomes, variable consortia of bacterial communities. The skin, for instance, is primarily colonized by members of the genera Propionibacterium (now Cutibacterium), Corynebacterium, and Staphylococcus (1). A similar pattern of genera is found in the human nose, which is regarded as a transition zone from the dry skin to the moist, mucoid airways (2). This rather confined area, more specifically, the region from the anterior nasal vestibule to the posterior nasopharyngeal cavity, is the favored colonization site of Staphylococcus aureus. While S. aureus is a member of the normal nasal microbiome in ca. 30% of the human population, it can also become an aggressive, life-threatening pathogen (3). Nasal carriage is a major risk factor for S. aureus infections. Accordingly, S. aureus is eradicated from the nose in at-risk patients by treatment with the antibiotic mupirocin. Interestingly, a significant percentage of humans seem never to be colonized by S. aureus, for reasons that are currently unknown. Increasing evidence suggests that the composition of the nasal microbiome is an important factor for the exclusion of S. aureus from the nose (2).

This article summarizes our knowledge of S. aureus nasal colonization and its interaction with other bacterial species that share the same habitat. Understanding naturally occurring S. aureus elimination strategies might enable the development of new eradication approaches in the future.

S. AUREUS AS A HUMAN COMMENSAL AND PATHOGEN

The preferred habitat of S. aureus is the anterior nares. Approximately 20% of the human population is permanently colonized by S. aureus, and another 20% are considered to be noncarriers. The remaining 60% of the population belong to the group of intermittent carriers (4). Although several factors that affect the S. aureus carrier status have been described, why only certain people are colonized remains elusive. In general, S. aureus occurs particularly frequently in men (5) and in women using hormonal contraception (6). Moreover, diabetic (7), hospitalized, and dialysis patients (4) are more often colonized than healthy humans, and there is a higher carriage rate during childhood compared to adulthood (8). A negative correlation exists between smoking and S. aureus carriage (9). Seasonal differences, such as the transition from winter to spring, and temperature and pollen or dust levels, could also affect the nasal microbiome composition (10). Host genetics appear to have only a moderate impact on the shape of the microbiome (11).

S. aureus possesses an extensive arsenal of virulence factors and, therefore, the clinical manifestations of S. aureus diseases are diverse and range from rather mild soft skin and tissue infections to severe and life-threatening infections, such as pneumonia, osteomyelitis, endocarditis, and sepsis (12). Although nasal colonization by S. aureus is asymptomatic, it increases the risk of invasive infections, especially in immunocompromised and hospitalized patients (4, 13). Analysis of infecting strains revealed that it is usually the nasal strain of a patient that is responsible for subsequent infections (13, 14). Therefore, decolonization of nasal S. aureus serves as a very efficacious measure for prevention of invasive infections (13). Since S. aureus transmission from the nose to other parts of the body occurs mainly via hand contact, S. aureus usually disappears also from other body sites when it is eliminated from the nose (15, 16).

The high risk of S. aureus carriers to infection justifies widely applied mupirocin treatment to eradicate S. aureus from the nares of patients. Mupirocin interferes with the synthesis of bacterial proteins by reversibly binding to bacterial isoleucyl-tRNA (17). Application of a mupirocin ointment twice a day for 5 consecutive days typically leads to eradication of S. aureus from the nose and to a reduced risk of invasive S. aureus infections (14, 17). The success rate of mupirocin treatment reaches approximately 90%, yet resistance is increasing, reaching up to 30% in some parts of the United States. Hence, new antimicrobial compounds for S. aureus decolonization are urgently needed (17, 18). Recolonization by S. aureus frequently occurs a few weeks after the end of mupirocin treatment, probably as a result of S. aureus cells using the posterior vestibula as a hidden reservoir, which can hardly be reached by mupirocin treatment (19–21).

Resistance to several common antibiotics has increased in S. aureus in the past decades. Resistance to the entire class of β-lactam antibiotics, including methicillin, characterizes the epidemic methicillin-resistant S. aureus (MRSA) strains. However, multidrug-resistant MRSA strains, which have acquired resistance to additional antibiotics, such as erythromycin, clindamycin, ciprofloxacin, tetracycline, or mupirocin, have also been observed (22). MRSA infections used to be acquired in hospitals by patients and are called hospital-associated MRSA infections. However, in the past decade, increasing numbers of MRSA infections have been acquired outside the health care system by emerging MRSA clones referred to as community-associated MRSA. Many community-associated MRSA clones are much more virulent and transmissible than hospital-associated MRSA clones (23).

DIVERSITY OF THE HUMAN NASAL MICROBIOME

In addition to the genus Staphylococcus, which belongs to the phylum Firmicutes, the nasal microbiome also contains members of other phyla, in particular, Actinobacteria and Proteobacteria. Within these phyla, the genera Corynebacterium, Propionibacterium, Staphylococcus, and Moraxella are most common in the human nose (24). Other genera are less frequently found (11). Interestingly, the nasal cavities are also colonized by various anaerobic species (19).

To classify different microbiome compositions, seven community state types (CSTs) were defined according to the dominance of specific nasal species, genera, or families. CST1 is dominated by S. aureus, and CST2 is defined by the major occurrence of Enterobacteriaceae, such as Escherichia spp., Proteus. spp., and Klebsiella spp. Staphylococcus epidermidis dominates CST3, whereas Propionibacterium spp. are major colonizers in CST4. Corynebacterium is the dominating genus of CST5, and Moraxella spp. are most prevalent in CST6, whereas CST7 is defined by the predominance of Dolosigranulum pigrum. The most common nasal CST is CST4, followed by CST3 and CST1, whereas CST6 is the least prevalent (Table 1). Although the conspicuous presence of various species or genera defines the CSTs, these species can also be identified at lower abundances in other CSTs (11). Comparing the microbiome compositions at various body sites, it is obvious that the nasal bacterial composition overlaps with those of the oral cavity and skin and can therefore be regarded as a bridge between the microbiomes of these two body sites (24, 25).

TABLE 1.

Community state types of the human nose^a

Community state type	Dominated by	Prevalence (%)
CST1	S. aureus	12.4
CST2	Enterobacteriaceae, e.g., Escherichia spp. (11), Proteus spp. (11), Klebsiella spp. (11)	9.0
CST3	S. epidermidis	22.5
CST4	Propionibacterium spp., e.g., P. acnes (88)	28.7
CST5	Corynebacterium spp., e.g., C. accolens (88), C. pseudodiptheriticum (88), C. propinquum (88)	11.2
CST6	Moraxella spp., e.g., M. lacunata (88), M. nonliquefaciens (88)	5.6
CST7	Dolosigranulum pigrum	10.7

^a Seven major community state types were defined according to the abundance of dominant nasal bacterial species, genera, or families (11).

Human skin areas with mostly dry, sebaceous, or moist properties differ in the abundance of bacterial microbiome members, while the core species are largely the same (1). It remains unclear if the individual differences in the skin microbiomes can also be attributed to specific CSTs. S. epidermidis, Staphylococcus capitis, Staphylococcus hominis, and Staphylococcus warneri are the most abundant coagulase-negative staphylococci (CoNS) on the skin (1). S. aureus also occurs transiently on the skin of healthy humans, especially in skin areas of the axillae and perineum (26, 27). Notably, the skin of atopic dermatitis patients is often permanently colonized by S. aureus on inflamed and noninflamed skin parts (28).

CoNS not only colonize the human skin; they can also occasionally be detected in the nose (29, 30). Only S. epidermidis is a core microbiome member of both the skin and nose (29). In general, the nasal microbiome is less dense and less diverse than, for instance, the gut microbiome (31, 32). S. epidermidis is particularly prevalent on moist areas of the body, such as the axillae, inguinal and perineal areas, toe tissue, and conjunctiva, and is also a common resident of the anterior nares (29). In addition, a variety of other CoNS occupy specific regions of the skin. S. hominis and Staphylococcus haemolyticus are often found in the axillae and pubic areas (29). S. capitis can be preferentially isolated from the sebaceous glands on the forehead and scalp (29) and is also found on the skin and in the nares (33). Whereas Staphylococcus auricularis is part of the external ear microbiome, Staphylococcus saprophyticus often colonizes the rectum and genitourinary tract (29). Staphylococcus lugdunensis is also part of the human skin flora and is particularly prevalent in the pelvic and perineal regions, the groin, the axillae, and in the nail bed of the first toe. S. lugdunensis has been isolated from the human nose at an incidence rate of 10 to 26% (19, 29, 34, 35). S. warneri is typically found at a lower percentage than other CoNS in the nares and on the skin, preferably on the head, arms, and legs (33). Compared to the other CoNS, S. lugdunensis and S. saprophyticus are considered species with slightly higher pathogenicity (29).

Compared to species such as S. epidermidis, which can be present in the nose with multiple strains simultaneously, nasal S. aureus isolates typically belong only to one specific clone (36, 37). Notably, the number of S. aureus cells, which can be cultivated from carriers by nasal swabbing, varies from only a few to more than 10⁷ per swab (38). Attempts to distinguish persistent from intermediate carriers by a single nasal swab indicated that the detection of more than 10³ S. aureus organisms CFU/swab corresponds to a high probability of persistent carriage (39).

The anterior nares are regarded as the primary habitat of S. aureus (4), but a recent study showed that this species can be isolated with an even higher incidence from deeper (posterior) areas of the nasal cavity, suggesting that the posterior vestibule or the entire nasal vestibule may be the principle habitat of S. aureus (19). Further studies indicated that S. aureus can penetrate into the nasal tissue of healthy humans, and individual cells could be visualized even at the stratum basale of the nasal epithelium, which differs in structure between the anterior and posterior areas (40, 41). While the anterior nares are characterized by a keratinized, stratified squamous epithelium, the posterior nares are defined by pseudostratified, columnar ciliated epithelial cells (2, 41). In the anterior vestibule, the moist squamous epithelium on the septum adjacent to the nasal ostium harbors the highest number of S. aureus cells. By contrast, CoNS prefer the skin of the nasal septum and the anterior, hair-covered epidermal portion of the lateral wall (5).

METABOLISM IN THE HUMAN NOSE

The nose is an environment with a very low nutrient supply, which might be the reason for the low species diversity of this habitat. Human nasal secretions contain sodium chloride at concentrations found also in other body fluids and small amounts of potassium, magnesium, and phosphate, but they are low in potential nutrients such as sugars, amino acids, and other major building blocks (30). Based on the composition of nasal secretions, a synthetic nasal medium (SNM3) was established that enabled the efficient growth of S. aureus, whereas CoNS did not steadily grow in SNM3 (30). This finding indicates that S. aureus is better adapted to life in the human nose than many CoNS, which may use the human nose only as a temporary, but not preferred, habitat. S. aureus and many other nasal bacteria, such as S. epidermidis, Finegoldia magna, Propionibacterium acnes, and Streptococcus pyogenes, secrete proteases that degrade human proteins, such as albumin, lactoferrin, mucin, cytokeratin 10, and hemoglobin. These proteins are present in considerable amounts in human nasal secretions and might serve as nutrient sources for various bacteria (42–44). It is likely that secretory proteases produced by a specific bacterial strain generate peptides and amino acids that can be utilized by many other microbiome members and may have a broad impact on microbiome metabolism.

MECHANISMS OF NASAL EPITHELIAL ATTACHMENT

Efficient attachment mechanisms are a prerequisite for bacterial nasal colonizers to remain in tight contact with epithelial cells and resist clearance by mucociliar movement. S. aureus binds to fully keratinized, dead desquamated cells of the anterior nasal cavity and adheres also to live ciliated cells in the posterior nasal cavity. S. aureus uses different adhesion mechanisms depending on the specific epithelial characteristics at different parts of the nares (45–48).

The cell wall glycopolymer wall teichoic acid (WTA) mediates the initial attachment of S. aureus to epithelial cells, and WTA is crucial for S. aureus nasal colonization (41, 47, 49). WTA-deficient S. aureus mutants are limited in their ability to bind to nasal epithelial cells. In addition, mutations in the dltABCD operon, leading to a loss of d-alanine modification of WTA, also exhibit diminished epithelial binding capacities (47, 49). Similarly, the modification of WTA with N-acetylglucosamine (GlcNAc) in α- or β-configuration is essential for efficient binding to nasal epithelial cells, and a lack of WTA glycosylation significantly abolished the ability of S. aureus to colonize cotton rat nares in vivo (50).

The scavenger receptor class-F member 1 (SREC1) on nasal epithelia in the posterior nasal cavity is a target for the WTA of S. aureus and allows S. aureus binding to epithelial cells (46). The charged zwitterionic properties of the ribitol-phosphate repeating units of WTA are essential in SREC1 binding (51). The loss of the d-alanine modification of the WTA significantly reduces the SREC1 binding.

The adherence of S. aureus to the nasal epithelium is also mediated by cell wall-anchored proteins, which are involved in long-term persistence within the anterior nasal cavity. S. aureus expresses a variety of cell wall proteins, which belong to the group of MSCRAMMs (microbial surface components recognizing adhesive matrix molecules) (41). MSCRAMMs such as clumping factor B (ClfB), iron-regulated surface determinant A (IsdA), and serine-aspartate repeat-containing protein D (SdrD) play an important role in nasal colonization by S. aureus.

ClfB is a fibrinogen-binding protein produced by S. aureus that binds to the matrix proteins cytokeratin 10 and loricrin. These proteins are exposed on human squamous epithelial cells (48, 52, 53). IsdA, which is involved in heme uptake and iron acquisition (54), is expressed during human infections and is responsible for nasal colonization and survival of S. aureus on the skin. Similar to ClfB, IsdA interacts with cytokeratin 10 and loricrin, as well as with the extracellular matrix protein involucrin (48, 55). SdrD mediates adhesion to the human squamous epithelium by binding to desmoglein 1, a desmosomal cadherin that is mainly expressed by the epidermis and mucosa (56, 57) (Fig. 1). The connective function of desmoglein 1 is required for the epidermis to maintain its integrity and structure (57). In addition, the surface proteins SdrC and SasG promote adhesion to squamous epithelium cells, but their binding partners are still unknown (45, 58). Interfering with glycopolymer-receptor interaction might become a new strategy for controlling S. aureus colonization in the nose (46). Furthermore, ClfB could represent an ideal target molecule for new decolonization strategies (53). In this context, ClfB can be considered a promising component for the development of a vaccine that would also reduce nasal colonization by S. aureus (59).

FIGURE 1.

Attachment mechanisms of S. aureus in the human nasal cavity. The anterior and posterior parts of the human nose are lined by different types of epithelial cell, which require alternative bacterial adhesion mechanisms. For the keratinized stratified squamous epithelium in the anterior nasal cavity, S. aureus predominantly uses cell wall-attached surface proteins (MSCRAMMs) (2, 41). In contrast, the primary attachment in the posterior area, composed of a pseudostratified columnar ciliated epithelium, is mediated by specific interaction of the cell-wall linked wall teichoic acid (WTA) with the scavenger receptor class F member 1 (SREC1) (41, 46). The corneocytes (desquamated epithelial cells) in the anterior nasal cavity contain high levels of the proteins loricrin, cytokeratin 10, and involucrin (55). S. aureus can express a variety of cell wall proteins, which bind to these matrix proteins. The S. aureus adhesin clumping factor B (ClfB) binds to cytokeratin 10 and loricrin (48, 52), whereas the iron-regulated surface determinant A (IsdA) can also interact with involucrin (55). In addition, the S. aureus serine-aspartate repeat-containing protein D (SdrD) mediates adhesion to human squamous epithelial cells by binding to desmoglein 1 (56). Some S. epidermidis isolates secrete an extracellular serine protease (Esp), which inhibits S. aureus colonization by degradation of the surface proteins IsdA and SdrD and host receptors (77, 78). In addition, S. lugdunensis can prevent nasal colonization of S. aureus by producing the cyclic thiazolidine-containing peptide antibiotic lugdunin (35).

S. epidermidis colonizes the human nose at a higher frequency than S. aureus. In contrast to S. aureus, S. epidermidis has a different WTA structure, and adhesion proteins related to ClfB, IsdA, SdrD, and SdrC, are absent. Therefore, it remains unclear how S. epidermidis accomplishes adhesion in the nose (2, 60).

BACTERIAL COMPETITION

Bacteria from habitats with a limited nutrient supply such as the skin and nose have developed mechanisms to increase their fitness in competition with other microbiome members. Competition between bacterial species can be direct or indirect. Direct inhibition can be achieved, for instance, through production of antimicrobials, whereas indirect inhibition may occur via competition for nutrients or modification of living conditions (61).

Many nasal Staphylococcus isolates produce antimicrobial substances against bacterial competitors at an unexpectedly high frequency (84%), with S. epidermidis as the most frequent producer of antimicrobial activity (62). Importantly, production of many of the antibacterial activities is strongly enhanced or exclusively detectable under specific environmental stress conditions which are present in the human nose, such as hydrogen peroxide release and iron limitation (62). Antimicrobial substances, also called bacteriocins, are categorized into various groups and subgroups (63). Many Staphylococcus isolates are producers of lantibiotics, ribosomally synthesized antimicrobial peptides characterized by the presence of the thioether amino acids lanthionine and methyllanthionine. A variety of lantibiotics have been described for Staphylococcus strains (64). Epidermin (S. epidermidis) (65), Pep5 (S. epidermidis) (66), epilancin K7 and 15X (S. epidermidis) (67, 68), epicidin 280 (S. epidermidis) (69), staphylococcin C55 (S. aureus) (70), various nukacins (S. warneri [71], S. epidermidis [62], and S. hominis [72]), and lantibiotic-α and -β (S. hominis) (73), were documented. In addition, putative lantibiotic-biosynthetic gene clusters were found in the genome of S. capitis, which share homology with the biosynthetic systems of epidermin/gallidermin and the nonlantibiotic bacteriocin epidermicin (74). Lantibiotics are usually exclusively active against Gram-positive bacteria but often show no activity against S. aureus (62, 71). It is rather uncommon that nasal S. aureus is severely affected by lantibiotic-producing CoNS. Interestingly, a systematic analysis of P. acnes isolates revealed a frequent capacity to inhibit S. epidermidis (75). No compound has been identified that could explain this inhibitory activity; however, the genomes of various P. acnes strains encode a putative thiopeptide-biosynthetic gene cluster that is similar to those for antimicrobially active siomycin and berninamycin, which might explain the inhibitory effect (76).

A recent study showed that secretion of the extracellular serine protease Esp by S. epidermidis can efficiently inhibit S. aureus colonization. Artificial inoculation of Esp-secreting S. epidermidis to the nasal cavities of human volunteers was sufficient to eradicate S. aureus (77). Nasal colonization of S. aureus is probably abolished by Esp via the degradation of both bacterial adhesive surface proteins IsdA and SdrD and host receptor proteins (78). Although most S. epidermidis isolates produce Esp and some are producers of lantibiotics, there is no clear correlation of the absence of S. aureus with the presence of S. epidermidis (20, 64, 77).

Nasal S. lugdunensis can prevent colonization by S. aureus by producing an unusual antimicrobial compound termed lugdunin, which is a novel, cyclic thiazolidine-containing peptide antibiotic (35). The lugdunin gene cluster is encoded on the S. lugdunensis chromosome and is present in almost all S. lugdunensis strains. Lugdunin is generated by nonribosomal peptide synthetases. It has bactericidal activity against S. aureus and other Gram-positive bacteria, and it was shown to be active in vivo in animal models. Analysis of nasal microbiomes of hospitalized patients revealed that S. lugdunensis colonization is associated with a 6-fold reduced risk of S. aureus carriage in the nose, suggesting that lugdunin or lugdunin-producing commensals could become useful for the prevention of S. aureus colonization and infection.

Streptococcus spp. mainly occur in the oropharynx but can also be found at low frequency in the nostrils (79). Importantly, S. pneumoniae can modify the nasal habitat by releasing hydrogen peroxide, which induces the SOS response in S. aureus and activates DNA repair mechanisms as well as resident prophages. Subsequently, phage-produced lytic enzymes destroy S. aureus cells (80, 81). Hence, S. pneumonia is negatively correlated with nasal S. aureus carriage (82, 83).

An ambivalent correlation between Corynebacterium spp. and S. aureus was reported. While Corynebacterium accolens often occurs together with S. aureus, Corynebacterium pseudodiptheriticum is associated with the absence of S. aureus. A mutualistic relationship between C. accolens and S. aureus might rely on the joint mobilization of nutrients promoting the growth of these strains. In contrast, the competitive interaction between C. pseudodiptheriticum and S. aureus interferes with S. aureus colonization. Thus, the presence of C. accolens or C. pseudodiptheriticum might become a useful predictor of the propensity for nasal S. aureus carriage (20). A bacterial strain replacement study highlighted that nasal inoculation of persistent S. aureus carriers with a Corynebacterium sp. can lead to complete eradication of the pathogen in more than 70% of the probands (84).

Nostril- and skin-associated Propionibacterium spp. release coproporphyrin III, a porphyrin metabolite that promotes the aggregation and nasal colonization of S. aureus (85) (Fig. 2). A negative association with S. aureus carriage was reported for Simonsiella spp., D. pigrum, and F. magna, which are also regular members of the nasal microbiome (11, 31). The particular sensitivity of D. pigrum to a multitude of Staphylococcus isolates suggests that D. pigrum can only be present when no bacteriocin-producing Staphylococcus is in the same habitat (62).

FIGURE 2.

Established interactions between nasal bacteria. C. accolens and Propionibacterium spp. can promote the colonization by S. aureus (green boxes) by modulation of its adhesive capacities (20, 85) (blue arrows), while specific clones of S. epidermidis, S. lugdunensis, and Streptococcus spp. can lead to S. aureus growth inhibition (orange boxes). Some S. epidermidis isolates secrete high levels of extracellular serine protease (Esp), which inhibits S. aureus nasal colonization (77) (yellow arrow). In addition, S. epidermidis and S. lugdunensis can impede S. aureus colonization by producing antimicrobial molecules (35, 64) (black arrows). S. pneumoniae can release hydrogen peroxide, which leads to prophage activation in S. aureus along with phage-mediated lysis of S. aureus cells (80, 81) (red arrow).

COMPETITION BY INDUCTION OF HOST DEFENSE

In addition to the above-described competition scenarios, the human host can also impact the composition of the nasal microbiome. S. aureus expresses surfactant-like phenol-soluble modulin peptides, which mobilize proinflammatory lipoproteins from the staphylococcal cytoplasmic membrane. These lipoproteins activate the Toll-like receptor 2 and, consequently, lead to inflammation (86). The resulting inflammatory response in the nasal epithelium provokes the production of antimicrobial peptides. Although these hardly affect S. aureus, with its intrinsic immune evasion factors, they impair growth of other nasal commensal bacteria (5, 87).

CONCLUSION AND OUTLOOK

The mechanisms of microbiome-mediated exclusion of S. aureus from the human nose are probably multifactorial. It is likely that in addition to the processes described above, other factors are involved. It will be important to further investigate which other antimicrobial substances are produced by nasal commensals and which other strategies are used in their competition with S. aureus for nutrients and adhesion sites. New findings on nasal colonization may clarify why 20% of the human population is permanently colonized by S. aureus, while a similar percentage is never colonized. A better understanding could be helpful for developing new S. aureus eradication approaches that protect against recurrent S. aureus nasal colonization.