Insights on Virulence and Antibiotic Resistance: A Review of the Accessory Genome of Staphylococcus aureus

Staphylococcus aureus has been known as a causative agent of infection since 1882 when Alexander Ogston identified its role in sepsis and abscess formation.¹ It has continued to be one of the most recognized human pathogens throughout the community and hospital settings. S aureus is an opportunistic bacterium, which is frequently part of the human microflora, causing disease when the immune system becomes compromised. Although S aureus can be found in different parts of the body, the anterior nares are the primary ecological niche in humans. Nasal carriage differs between individuals and is a significant risk factor for S aureus infection.² In healthy populations, approximately 20% of individuals carry S aureus persistently, while close to 60% carry the infection intermittently, and about 20% do not carry it at all.² From the nares, staphylococci may spread to the skin, surgical wounds, foreign bodies (eg, tracheostoma, external fixation devices), burns, and the upper respiratory tract. Moreover, about 80%–100% of patients with eczema and atopic dermatitis are known to be colonized with the organism—bacterial colonization is an important factor in aggravating skin lesions.³ Another mode of transient transmission is via colonized hands of healthcare workers who acquire the organism after close contact with colonized patients or contaminated equipment.^4,5 S aureus can cause a wide range of infections that include 1) skin and soft tissue infections, such as bullous and nonbullous impetigo, folliculitis, cellulitis, surgical and wound infections, and mastitis; 2) systemic and life-threatening conditions, such as endocarditis, osteomyelitis, pneumonia, brain abscesses, meningitis, and bacteremia; 3) toxinoses, such as food poisoning, scalded skin syndrome, and toxic shock syndrome.^6,7

There has been growing concern regarding the increasing level of antibiotic resistance in S aureus,especially the severeconsequences of hospital and community-acquired methicillin-resistant S aureus (MRSA).^8–11 The advent of quinupristin-dalfopristin, (United States Food and Drug Administration [FDA]-approved in 1999), and antibiotics like linezolid (oxazolidinone, FDA approved in 2000) and daptomycin (lipopeptide, FDA approved in 2003) with their novel mechanisms of action against MRSA, offered some hope for the treatment of S aureus infection. However linezolid and daptomycin- resistant S aureus have been reported,^12–15 reinforcing the need to control MRSA in the hospital and community settings. The menace of S aureus infections have therefore led to a great interest in sequencing the genome of this pathogen in order to provide detailed insight into how the organism generates a variety of virulence factors and develops resistance to so many antibiotics.¹⁶ It has also stimulated research in identifying targets on whose genes are essential for in-vivo and in-vitro growth,thereby exploiting antibiotic target identification and selection.¹⁷ To date, 8 complete genome sequencing of S aureus have been conducted (Table 1) and 7 have been published.^18–21 The remaining sequenced strain is NCTC 8325.²² This review examines the accessory genome of sequenced S aureus strains and provides information on how this pathogen causes disease and exhibits resistance to a wide range of antibiotics.

Comparative Genomics of S aureus

Phylogenetic classification places S aureus in the Bacillus/Staphylococcus group. Accordingly, up to 52%of predicted proteins encoded by the N315 genome are similar to those in Bacillus substilis and B halodurans. They typically contain housekeeping genes involved in essential functions of the vegetative life of the bacteria, such as DNA replication, protein synthesis, and carbohydrate metabolism.¹⁸ The 8 S aureus sequenced genomes range in size from 2.820Mb to 2.903Mb^16,21,24 and is composed of core and auxiliary (accessory) genes.²³ The majority of genes comprising the core genome are those associated with central metabolism and other housekeeping functions. Approximately 75% of a S aureus genome comprises a core component of genes present in all of the strains.²⁴ Other genes within the core genome that are not essential for growth and survival include virulence genes that are not carried by other staphylococcal species, surface binding proteins, toxins, exoenzymes, and the capsule biosynthetic cluster.²³ Comparison of the percentage similarities of the DNA of the core genomes showed that N315, Mu50, MW2, and MSSA476 are closely related. The hospital strains N315, Mu50, and the community-acquired MW2 and MSSA476 belong to identical but separate sequence types (STs) by multilocus sequence typing (MLST, [ST5 and ST1, respectively]), while COL and NCTC8325 belong to closely related STs (ST250 and ST8).¹⁶ Only 678 single nucleotide polymorphisms (SNPs) were identified in comparing the core chromosome of FPR3757 with that of COL, confirming that they are related by vertical descent from a common ancestor.²¹ MRSA 252 is the most divergent of the sequenced strains (ST36) and it is a representative of the EMRSA-16 epidemic clonal group.

Accessory genome. The accessory genome of the S aureus genome encodes a diverse range of nonessential functions ranging from virulence, drug and metal resistance, to substrate utilization and miscellaneous metabolism.²⁴ They usually consist of mobile genetic elements (MGEs) that transfer horizontally between strains. These elements include bacteriophages, pathogenicity islands, chromosomal cassettes, genomic islands, and transposons. The identification and characterization of these elements has provided valuable information into how S aureus causes disease and its relative diversity.²³

Bacteriophages. Bacteriophage transduction appears to be the main method for gene transfer in S aureus of the 3 mechanisms for horizontal gene transfer in bacteria.²⁴ Temperate bacteriophages are common in all the sequenced strains in which they carry at least 1 phage. A unique feature of the 2 community-acquired MRSA (MW2 and FPR3757) chromosomes is the φSa2mw and φSa3usa, which carries the lukF-PV and lukS-PV genes that encode the Panton-Valentine (PV) leukocidin components (Table 2). It has a potent toxic effect on human white blood cells and is strongly associated with severe forms of pneumonia (necrotic pneumonia) caused by community-acquired S aureus strains.²⁵ The prophage φSa2usa is remarkably similar to the φSa2mw in that they are both integrated into the same conserved 29bp attachment site on the core chromosome.²¹ A unique feature in S aureus COL is a L54-like phage named φCOL.20 All sequenced S aureus strains with the exception of COL have the prophage φSa3 (Table 2). The prophage contains the staphylokinase gene (sak) gene, which is a potent plasminogen activator that could facilitate bacterial spreading through its fibrin-specific blood clotting activities.²¹ In addition, the sequenced N315 genome carries a different enterotoxin gene (designated sep), that encodes a protein with only 77% amino acid similarity to enterotoxin A.¹⁸ N315, MRSA252, and FPR3757 also possess the virulence chp (chemotaxis inhibitory protein) gene.^21,24 Two additional enterotoxin gene alleles, seg2, and sek2, which encode putative enterotoxin G and K homologues, were observed in MW2 and MSSA476.^16,19

Genomic islands. Two types of genomic islands (vSaα and vSaβ) have been identified in all the sequenced genomes of S aureus. The genomic island vSaα is distinctive because it carries many putative staphylococcal exotoxin (set) and lipoproteins (lpl) genes (Table 2). They are capable of inducing proinflammatory cytokine production by human peripheral blood mononuclear cells.²⁶ The structure of vSaβ differs from strain to strain with the presence or absence of some genes. The superantigen gene cluster (composed of 6 enterotoxin genes) carried by islands of N315, Mu50, and MRSA252 are missing from vSaβ of MW2, MSSA476, and FPR3757. Instead, vSaβ island of MW2, MSSA476,COL,and FPR3757 possess a gene cluster (designated bsa), which encodes a putative bacteriocin.19,21 This may indicate that bacteriocin,which is useful in the competition with other bacterial species for colonization in humans, is not needed in the hospital environment for multidrug resistant hospital-acquired MRSA.²⁷ The staphylococcal serine proteases (spl) and lipoproteins (lpl) genes were also present in all sequenced strains. The leukocidin DE gene (lukDE) was absent only in MRSA252, which carried a hyaluronate lyase (hysA) gene (Table 2).

S aureus pathogenicity islands. S aureus pathogenicity islands (SaPIs) often carry superantigen genes, such as toxic shock syndrome (tst) and enterotoxins B and C, implicated in toxic shock and food poisoning. These islands carry approximately one-half of the S aureus toxins or virulence factors, and allelic variation of these genes, which contribute to the pathogenic potential of S aureus.²⁰ They can integrate at specific sites, and can transfer horizontally at very high frequency with the help of specific bacteriophages.²⁸ The N315 and Mu50 genome possess the superantigen genes se1, se3, and tst (Table 2). A feature of the toxic shock syndrome toxin (TSST) island family is its close linkage to prophages φN315/φMu50A and φMu50B, respectively, which are integrated in close proximity to these islands. It has been suggested that these phages may be involved in the horizontal transfer of the islands.18 A unique feature of Mu50 is the carriage of the fhuD gene that possibly encodes a ferrichrome-binding ABC transporter (Table 2).This iron transporter might confer a selective advantage to Mu50 in human tissue.¹⁸ The MW2 also carries 2 allelic forms of enterotoxin sel2 and sel4 that are unique and contribute to its increased virulence.^19,20,23 The MSSA476 does not have a pathogenicity island in its genome, while the MRSA252 carries a SaPI-like element, SaPI4.¹⁶ SaPI5, a novel staphylococcal pathogenicity island, was identified in FPR3757 and 2 enterotoxins (seq and sek) that are closely related to the genes observed in COL.²¹

Staphylococcal cassette chromosomes. Staphylococcal cassette chromosome (SCC) elements are mobile genetic elements that integrate at the unique site (attBscc) in the chromosome of MRSA.²⁹ It is located near the replication origin of S aureus, which is 10 kb downstream of purA and 66 kb–89 kb upstream of spa gene depending on the size of the integrated copy of SCCmec.^18,19,30 The ccr gene complex contains 2 site-specific recombinase genes, ccrA and ccrB, which are responsible for the mobility of SCCmec.^31,32 The region other than the mec and ccr gene complexes is designated the J (junkyard) region. Each SCCmec type is further classified into subtypes on the basis of the J-region sequence.³³ The J regions contain various genes or pseudo genes whose presence does not appear essential or useful for the bacterial cell; the notable exceptions are resistance genes for nonβ-lactam antibiotics or heavy metals, some of which are derived from plasmids or transposons.³⁰ At least 5 different versions of SCCmec are found in S aureus— SCCmec types I to V.^32,34,35 Type-II SCCmec of N315, Mu50, and MRSA252 contain an integrated copy of plasmid pUB110 with bleomycin and kanamycin resistance genes and transposon Tn554 carrying erythromycin and spectinomycin resistance in the J region.^16,18,36 The community-acquired MRSA strain (MW2) has a type IVa element, which is identical to that of FPR3757; its structure is smaller in size than that of hospital-acquired MRSA strains.^19,21 Type IVa SCCmec of MW2 is made up of 2 allelic elements—class B mec-gene complex (mecA and its regulatory genes), and type-2 ccr A and B genes. Type I SCCmec in COL is composed of class B mec gene complex and type 1 ccr gene complex.^29,30 Types I, IVa, and V carry only the mecA gene.^30,37 Notably, the methicillin- susceptible MSSA476 contained a novel SCC-like element (SCC₄₇₆), which showed the greatest similarity to a previously described S hominis non-mec SCC element (SCC₁₂₂₆₃).³⁸ SCC₄₇₆ shares the same left and right boundaries (attL and attR) and similar inverted repeat sequences as SCCmec elements, but does not carry the mecA gene.³¹ However, it carries a novel gene with homology to the fusidic acid resistance gene far1.³⁹ A striking feature of the FPR3757 genome is the 30.9 kb DNA that contains a cluster of 6 genes—the arc gene cluster, which encodes a arginine deiminase pathway that converts L-arginine to carbon dioxide, ATP, and ammonia.²¹ The newly identified ACME (Arginine Catabolite Mobile Element) has been proposed as a new member of the SCC family because it integrates into the open reading frame (ORF) orfX on the chromosome using the same attachment site as SCC element.²¹ A recent study also confirmed that ACME is currently restricted to related strains of FPR3757 (USA300) and a limited number of USA100 isolates with SCCmec type IVa.⁴⁰

Plasmids

S aureus isolates, and particularly those from hospitals, often carry 1 or more free or integrated plasmids.All types of S aureus plasmids frequently harbor genes that encode resistance to antibiotics, heavy metals, or antiseptics. Some virulence genes are also reported to be carried on plasmid, such as exfoliative toxin B and some superantigens.^41,42 The MW2 strain contained no antibiotic resistance genes apart from blaZ (encoding penicillinase) and a cadD (cadmium resistance gene) on plasmid pMW2 (20654bp).^19,23 The N315 plasmid pN315 (24653bp) contained a cadmium resistance determinant cadDX, an arsenate-resistant determinant arsRBC, and aTn552-related transposon that harbors the penicillinresistant gene blaZ.¹⁸ On the Mu50 plasmid pMu50 (25107bp) there is a copy of Tn4001 that carries aacaphD (encoding aminoglycoside resistance). Others include the qacA genes that encode resistance to quaternary ammonium compounds (eg, diamidines, chlorhexidine, and intercalating dyes). Strain MRSA252 carries an integrated plasmid that confers resistance to heavy metals arsenic (arsBC) and cadmium (cadAC) (Table 2). The blaZ and cadD genes are the only apparent resistance genes within the MSSA476 genome located on a pSAS1 plasmid.¹⁶ Strains N315, Mu50, and MRSA252 also carry a pUB110 plasmid with bleomycinand kanamycin-resistant genes.^16,18 The plasmid (pT181) encoding tetracycline resistance was identified in COL.^20,23 Three plasmids were noted in the recently sequenced FPR3757 genome.¹⁰ They include pUSA02, which encodes resistance to tetracycline (tetK), pUSA03, which is a conjugative plasmid that carries ermC and ileS genes and encodes constitutive resistance to macrolides-lincosamides-streptogramins B and highlevel resistance to mupirocin, respectively. The third plasmid pUSA01 has no identifiable function.²¹

Transposons

Transposon and insertion sequences can integrate themselves into any chromosome loci and are believed to contribute much to the adaptability of S aureus to an adverse environment.^19,43 Tn554 is a site-specific transposon that encodes resistance to spectinomycin and macrolide-lincosamide-streptogramin B antibiotics. Two copies of Tn554 were observed in N315, Mu50, and MRSA252 genomes, while 3 additional copies were found in the N315 genome.^16,18 A unique conjugative transposon,Tn5801,was found in Mu50, which carries a gene (tetM) encoding tetracycline and minocycline resistance.¹⁸ The MRSA252 chromosome also contained a Tn552 transposon that encodes the BlaI, BlaR, and BlaZ components of the inducible S aureus β-lactamase.¹⁶ There is also an element integrated into the chromosome that is similar to the Tn916 transposon in MRSA252 and COL; however, it does not appear to carry any obvious determinants of resistance.^16,23 Recorded transposons in the MW2 and MSSA476 genome are scarce.^16,19

Conclusion

The ability of a pathogen to exhibit a number of virulence factors and antibiotic resistance genes expands its ecological niche with comparative advantage over avirulent and antibiotic-susceptible strains. Since the emergence of penicillin- and methicillin-resistant S aureus in 1948 and 1961, respectively, it has evolved into a multidrug resistant microorganism, which is successful in transmitting and causing disease in the hospital and community environment. The substantial morbidity and mortality associated with hospital- and community- acquired MRSA continues to be a challenging public health problem. Although molecular epidemiological investigations support the view that the evolution of MRSA is predominantly clonal in nature, the horizontal transfer of MGEs encoding virulence and resistance genes play an important role in its progression to a fearsome pathogen. The multiple insertions of transposons and insertion sequences on the SCCmec in the hospital-acquired MRSA genomes, the unique feature of the Panton-Valentine leukocidin gene, and the size of the SCCmec in the 2 community-acquired MRSA (MW2 and FPR3757) chromosomes have revealed the outstanding ability of this pathogen to evolve and adapt in order to survive in the hospital and community settings. Furthermore, the horizontal acquisition of the ACME cluster in FPR3757 from S epidermidis and what appears to be a common origin of the conjugative plasmid pUSA03 and pLW103 (plasmid that carries the vanA gene in Enterococcus faecalis) suggest the possibility of gram-positive organisms to share in each others gene pool.A recent observation⁴⁴ that the PVL, sea, and chp genes were likely to be carried on unrelated phages apart from the φSa2 and the φSa3 suggest that sequencing of other dominant lineages could reveal more on the level of diversity of the pathogen in terms of virulence and the level of antibiotic resistance. Sequence data of S aureus strains and comparison with staphylococcal species could also open a vista of knowledge in the development of novel agents to combat this pathogen. Although a number of barriers for horizontal transfer of MGEs in S aureus have been identified,²⁴ it is evident that constant environmental challenges would enhance the fitness level and survivability of this pathogen.

Acknowledgement

The authors appreciate the support of the National Research Foundation, Republic of South Africa, and the Department of Microbiology, Faculty of Medicine, Kuwait University for Dr.A.O. Shittu during his PhD program.

Abbreviations: NI, virulence or drug resistance genes not presently identified; aacA-aphD, aminoglycoside resistance; arsBC, arsenic resistance genes; blaZ, penicillin resistance; bsa, bacteriocin biosynthetic genes; cadACDX, cadmium resistance genes; chp (chemotaxis inhibitory protein); ear, putative β-lactamase type protein; ermA, erythromycin resistance; far1, fusidic acid resistance; fhuD, siderophore transporter; qacA, quaternary ammonium compound (antiseptic) resistance;hysA, hyaluronate lyase; lpl, lipoproteins; lukDE, 2 components of the leukocidin DE toxins; lukSF-PV, 2 components of the Panton-Valentine leukocidin toxin; mecA, penicillin-binding protein 2a conferring resistance to methicillin; sak, staphylokinase; SCC, staphylococcal cassette chromosome; sea, enterotoxin A; sec3, enterotoxin C3; sec4, enterotoxin C4; seg2, enterotoxin G2; sek2, enterotoxin K2; sel, enterotoxin L; sel2, enterotoxin L2; sep, enterotoxin P; set, staphylococcal enterotoxins; spc, spectinomycin resistance; spl, staphylococcal serin proteases; tst, toxin shock syndrome toxin-1; tetM, tetracycline resistance.