Superantigens of Staphylococcus aureus From Patients With Diabetic Foot Ulcers

Abstract

Background

Diabetic foot ulcer (DFU) infections are challenging. Staphylococcus aureus is the most commonly isolated pathogen in DFUs. Superantigens (SAgs) are causative in many S. aureus infections. We hypothesized both that DFU S. aureus will produce large SAg numbers, consistent with skin infections, and that certain SAgs will be overrepresented. We assessed the SAg and α-toxin profile of isolates from patients with DFU, compared with profiles of isolates from other sources.

Materials

Twenty-five S. aureus isolates from patients with DFU were characterized. Polymerase chain reaction was used to detect genes for methicillin-resistance and SAgs. Some SAgs and the α-toxin were quantified. We compared the SAg profile of DFU isolates with SAg profiles of S. aureus isolates from skin lesions of patients with atopic dermatitis and from vaginal mucosa of healthy individuals.

Results

Most DFU isolates were methicillin susceptible (64%), with USA100 the most common clonal group. The SAg gene profile of DFU isolates most closely resembled that of isolates from patients with atopic dermatitis, with the highest number of different SAg genes per isolate and a high prevalence of staphylococcal enterotoxin D and the enterotoxin gene cluster. DFU isolates also had a high prevalence of staphylococcal enterotoxin-like X.

Conclusions

Comparison of the SAg profile of DFU isolates to SAg profiles of skin lesion isolates and vaginal mucosa isolates revealed that the SAg profile of DFU isolates was more similar to that of skin lesion isolates. SAgs offer selective advantages in facilitating DFU infections and suggest that therapies to neutralize or reduce SAg production by S. aureus may be beneficial in management of patients with DFU.

Diabetes mellitus affects approximately 5% of the US population. Among diabetic patients, 5%–25% will develop foot ulceration during their lifetimes [1]. Nearly 1 in 6 patients with diabetic foot ulcer (DFU) infections eventually require amputation, and >50% of those amputees have survival rates of <5 years [2, 3]. A recent study estimated that the annual healthcare cost for DFU infections in the United States exceeds $10 billion, most of which is related to inpatient stay and disease progression [4, 5].

Staphylococcus aureus is the most common pathogen in infections of diabetic foot wounds [6, 7]. Although S. aureus is considered a commensal because of its colonization of up to 30% of the population, it causes a wide variety of human illnesses, ranging from relatively benign soft-tissue infections to life-threatening toxic shock syndrome (TSS), necrotizing pneumonia, and infective endocarditis [8].

Among S. aureus virulence factors, superantigen (SAg) exotoxins contribute significantly to major illnesses [9]. The SAg family consists of TSS toxin- 1 (TSST-1), the cause of both menstrual and non-menstrual TSS; staphylococcal enterotoxins A–E (SEA-SEE) and SEG, classically the common causes of food poisoning and nonmenstrual TSS; and staphylococcal enterotoxin-like (SEl) H–X, which either lack or have not been tested for emetic activity [10]. Of the SEl molecules, only SEl-X has thus far been associated with disease [11].

A defining characteristic of SAgs is their abilities to overinduce cytokine production from both T lymphocytes and macrophages. SAgs cross-link the variable region of the T-cell receptor β-chain (Vβ-TCR) and the α and/or β chain of major histocompatibility complex class II molecules on antigen-presenting cells. This interaction causes a cytokine storm with immune dysregulation that mediates the severe manifestations of SAg illnesses [12, 13].

In a recent study of S. aureus strains from human mucosal surfaces and skin lesions due to atopic dermatitis (AD), >70% of the isolates produced 1 or more of TSST-1, SEB, and SEC, the 3 SAgs associated with nearly all cases of TSS [14]. In that study, S. aureus isolates from patients with steroid-resistant AD were also selected for their greater capacity to produce SAgs, which further suggests a specific adaptation mechanism of the organism to the illness. We hypothesized that similar mechanisms could also apply to S. aureus isolates from patients with DFU, since the manifestations of DFUs are usually characterized by proinflammatory cytokine release at the sites of infection, following localization of neutrophils, lymphocytes, and macrophages. We compared S. aureus isolates from patients with DFU with isolates from human mucosal surfaces and skin lesions due to AD for their capabilities to encode SAgs.

METHODS

Staphylococcal Isolates

The DFU isolates were obtained from 25 patients with nonischemic, neuropathic plantar DFUs that were free of clinical signs of infections and were not being treated with topical or systemic antibiotics. The DFUs in this sample had been present from 1 to 156 weeks, with a mean duration (±SD) of 29.1 ± 32.8 weeks. Specimens of microbes were obtained from DFUs at baseline by using Levine's technique and established study protocols [15]. Levine's technique is different from other swab specimen techniques in that fluid samples are obtained from deep tissue layers. The wounds were cleansed with saline, and Amies with charcoal transport swabs (Copan, Italy) were rotated over a 1-cm² area of viable, nonnecrotic wound tissue for 5 seconds, using sufficient pressure to extract wound tissue fluid. Levine's technique had an accuracy (ie, area under the receiver operating characteristic curve) of 0.80 when compared to wound tissue specimens, the gold standard [15, 16]. Ulcer swab specimens were immediately transported to the dedicated microbiological research laboratory of one of the authors (D. J. D.).

DFU swab specimens were plated on Columbia blood agar (Remel, Lenexa, KS) and CHROMagar MRSA (BD, Franklin Lakes, NJ). The Columbia plates were incubated in 5% CO₂ at 37°C for 48 hours, and MRSA plates were incubated aerobically at 37°C for 48 hours. The plates were examined for S. aureus growth and colony characteristics. Mauve-colored colonies on the CHROMagar MRSA plates were considered indicative of MRSA and were subcultured for identification. Differing colony morphological characteristics on the Columbia plates were subcultured for identification. All S. aureus isolated were identified using standard culture-based assays and 16S sequencing [7].

Additionally, previous published SAg profiles of S. aureus isolates from skin lesions due to AD and normal vaginal mucosa sources [14] were also included for cross-comparison.

SAg Gene Amplification and Exotoxin Assays

DNA was extracted from organisms cultured overnight at 37°C with shaking (200 revolutions per minute) in 10 mL of Todd-Hewitt broth (Difco Laboratories, Detroit, MI) [17]. The primers used for polymerase chain reaction (PCR) amplification of SAg genes are listed in Table 1. Genes were amplified by Taq polymerase (Qiagen, Valencia, CA).

Table 1.

Polymerase Chain Reaction Primer Sequence and Expected Product Size of 22 Staphylococcal Superantigen Genes and the Methicillin Resistance Gene (mecA)

Primer	Sequence	Size, bp	Reference
SEA forward	GATTCACAAAGGATATTGTTGATAAATAT	400	[18]
SEA reverse	GTCCTTGAGCACCAAATAAATC
SEB forward	GTATGATGATAATCATGTATCAGCAA	625	[18]
SEB reverse	CGTAAGATAAACTTCAATCTTCACAT
SEC forward	GAGTCAACCAGACCCTATGCC	650	[18]
SEC reverse	CGCCTGGTGCAGGCATC
SED forward	GCATTACTCTTTTTTACTAGTTTGGTA	530	[18]
SED reverse	CCTTGCTTGTGCATCTAATTC
SEE forward	CTGAATTACAAAGAAATGCTTTAAGC	420	[18]
SEE reverse	GCCTTGCCTGAAGATCTA
TSST-1 forward	GAAATTTTTCATCGTAAGCCCTTTGTTG	655	[18]
TSST-1 reverse	TTCATCAATATTTATAGGTGGTTTTTCA
SEG forward	GGGAACTATGGGTAATGTAATGAATC	430	This study
SEG reverse	TGAGCCAGTGTCTTGCTTTG
SEl-H forward	TCACATCATATGCGAAAGCAG	357	[18]
SEl-H reverse	TAGCACCAATCACCCTTTCC
SEl-I forward	GCTCAAGGTGATATTGGTGTAGG	572	This study
SEl-I reverse	CTTACAGGCAGTCCATCTCC
SEl-J forward	CAGCGATAGCAAAAATGAAACA	450	[18]
SEl-J reverse	CCCTCTTCTAGCGGAACAAC
SEl-K forward	TGGATCAATGGAAATCAACAAAA	420	[18]
SEl-K reverse	CGGGCTACCCGAAAAATAAT
SEl-L forward	CTGTTTGATGCTTGCCATTG	370	[18]
SEl-L reverse	GCGATGTAGGTCCAGGAAC
SEl-M forward	CGGTGGAGTTACATTAGCAGGT	320	This study
SEl-M reverse	TTTCAGCTTGTCCTGTTCCA
SEl-N forward	GCTTATACGGAGGAGTTACG	298	This study
SEl-N reverse	GCTCCCACTGAACCTTTTACG
SEl-O forward	GGAATTTAGCTCATCAGCGATT	390	This study
SEl-O reverse	TGCTCCGAATGAGAATGAAA
SEl-P forward	ACCAACCGAATCACCAGAAG	400	[18]
SEl-P reverse	GTTCAAAAGACACCGCCAAT
SEl-Q forward	GATGTAGGGGTAATCAACCTTAG	500	[18]
SEl-Q reverse	CTCTCTGCTTGACCAGTTCC
SEl-R forward	TACTATGGGGAATGTTGAATCC	558	[18]
SEl-R reverse	GGTATAAAGGGAACCAAATCC
SEl-S forward	CTAACTCTTGAATTGTAGGTTCC	332	[18]
SEl-S reverse	CTCCACACAACTATTATCAAACG
SEl-T forward	TCGGGTGTTACTTCTGTTTGC	170	[18]
SEl-T reverse	GGTGATTATGTAGATGCTTGGG
SEl-U forward	GCAGCTTACTATTTATGTTAAATGGC	390	This study
SEl-U reverse	CTATTTGATTTCCATCATGCTCGG
SEl-X forward	TCTATGGGGGAACATTTGGA	420	[18]
SEl-X reverse	CCGCCATCTTTTGTATTTATGA
mecA forward	TGCTATCCACCCTCAAACAGG	280	This study
mecA reverse	AACGTTGTAACCACCCCAAGA

Abbreviations: SE, staphylococcal enterotoxin; SEl, staphylococcal enterotoxin-like toxin; TSST, toxic shock syndrome toxin.

Selected SAg proteins were detected using antibody-based assays [19]; the SAg proteins evaluated included TSST-1, SEA–D, and SEl-X. Organisms were cultured in Todd-Hewitt broth at 37°C with shaking overnight. The resultant supernates were concentrated by SAg precipitation with 4 volumes of 100% ethanol overnight at 4°C, centrifuged at 4000 × g for 10 minutes, and resolubilized in distilled water to 5% of the original volume. Western immunoblotting with specific polyclonal rabbit antisera prepared against the designated SAgs was used to quantify the amounts of SAg produced by each isolate [14, 20]. Purified SAgs were used to generate standard curves for quantification.

Pulsed-Field Gel Electrophoresis (PFGE) for Clonal Typing of DFU Isolates

All DFU isolates were characterized by PFGE [21]. Whole chromosomal DNA in agarose was digested with SmaI (Sigma-Aldrich, St. Louis, MO), and the restriction fragments were separated on a CHEF DRII apparatus (Bio-Rad Laboratories, Hercules, CA). After electrophoresis, the gels were stained with ethidium bromide, illuminated by UV light, and photographed. PFGE patterns were analyzed using Bionumerics software (Applied Maths, Kortrijk, Belgium). The unweighted pair group method with arithmetic averages and DICE coefficient (0.5% optimization, 1.0% position tolerance) were used for dendrogram construction. A similarity coefficient of 0.8 was used to define PFGE types, and isolates with indistinguishable banding patterns were considered to belong to the same subtype. Isolate patterns from this study were compared to those of the Centers for Disease Control and Prevention's USA type strains [22].

tstH and selx Location on the S. aureus Chromosome

SAg gene profiling of S. aureus from DFUs identified 1 isolate that carried the genes for both TSST-1 and SEl-X, which appear to exclude each other in all strains tested thus far. We evaluated the position of the 2 genes in the chromosome. The primers used for PCR amplification of upstream and/or downstream regions of tstH and selx genes are listed in Table 2. The PCR products were subsequently sequenced and compared with information in the published database.

Table 2.

Polymerase Chain Reaction Primers for Upstream and/or Downstream Regions of tstH and selx

Primer	Sequence	Size, bp
SAPI1 forward	GTAGGTTCAGGTTTTCATTCTTC	2054
SAPI1 reverse	CCGTCATTCATTGTTATTTTCC
SAPI2 forward	ATTTTACATCATTCCTGGCAT	3500
SAPI2 reverse	CCGTCATTCATTGTTATTTTCC
SAPIn1 forward	CGCCTGGTGCAGGCATC	3500
SAPIn1 reverse	CCGTCATTCATTGTTATTTTCC
Upstream selx forward	CGCTATTTGCTCATCAAAATTGAG	1487
Upstream selx reverse	GACTCTAATGTATATTTACCGCCATC
Downstream selx forward	CGGTAAATATACATTAGAGTCGCATAAAG	1085
Downstream selx reverse	CGCTGACATTGATAACACTGC

Abbreviation: SAPI1, Staphylococcus aureus pathogenicity island 1.

RESULTS

PFGE Clonal Types of S. aureus Isolates From DFUs

Among 25 collected DFU isolates, PFGE clonal types encompassed USA100 to USA1000 strains (Figure 1). USA100 was the most prevalent clonal group, accounting for 24% of the DFU strains. USA200, USA700, and USA900 clonal groups were distributed equally (at 11%) among the strains. Other clonal groups were represented in <9% of the strains. There were no representatives of the USA500 clonal group.

Figure 1.

USA clonal types of Staphylococcus aureus isolates recovered from patients with diabetic foot ulcers (DFU). Pulsed-field gel electrophoresis was used to clonally type 25 DFU isolates.

mecA and SAg Gene Profile of S. aureus Isolates From DFUs

We screened all DFU isolates for the presence of mecA and the SAg genes sea–e, tstH, seg, and selh–x by conventional PCR. Several important observations were made. Nine isolates (36%) contained mecA (Table 3), consistent with previous studies suggesting that the prevalence of methicillin-resistant S. aureus (MRSA) among DFU isolates is 15%–30% [23]. Ten isolates (40%) encoded SED (Figure 2 and Table 4). SED is plasmid encoded, providing evidence for the high prevalence of the sed-containing plasmid pIB485 among DFU isolates. SEl-X is the most recently identified S. aureus SAg and has been shown to play a significant role in necrotizing pneumonia caused by the USA300 strain LAC [11]. Twenty-two isolates (88%) carried the SEl-X gene, and the remaining 3 isolates contained the gene for TSST-1, indicating that all isolates contained the gene for one of the 2 group I SAgs (Figure 2 and Table 4). One additional isolate harbored the genes for SEl-X and TSST-1, which appear to exclude each other in all other strains tested to date (Table 3).

Table 3.

mecA Element Distribution and Selected Superantigen Combinations in Staphylococcus aureus Isolates From Patients With Diabetic Foot Ulcers

Superantigen(s)	Isolates, No. (%)
mecA	9 (36)
selo, selm, seli, seln, seg, and selu,	16 (61)
selo, selm, seli, seln, and seg	1 (4)
selm, seli, and seln	2 (8)
seg and selu	2 (8)
TSST-1 and SEC	1 (4)
TSST-1 and SEl-X	1 (4)
TSST-1, SEB, or SEC	6 (24)

Abbreviations: SE, staphylococcal enterotoxin; SEl, staphylococcal enterotoxin-like toxin; TSST, toxic shock syndrome toxin.

Figure 2.

Superantigen (SAg) gene profile of diabetic foot ulcer isolates. Conventional polymerase chain reaction and specific primers for 22 SAg genes were used to identify the SAg gene profile.

Table 4.

Superantigen (SAg) Genes in Staphylococcus aureus Isolates From Diabetic Foot Ulcers (DFUs), Atopic Dermatitis (AD) Lesions, and Vaginal Mucosa (VM)

SAg	Isolate Source, No. (%)			P
	DFU	AD	VM	DFU vs AD	DFU vs VM
SEA	6 (24)	41 (41)	8 (27)	NS	NS
SEB	2 (8)	35 (35)	3 (10)	.007	NS
SEC	4 (16)	25 (25)	9 (30)	NS	.02
SED	10 (40)	39 (39)	4 (13)	NS	≤.001
SEE	2 (8)	60 (60)	6 (20)	≤.001	.02
TSST-1	2 (8)	38 (38)	12 (40)	.003	≤.001
SEl-G	16 (64)	55 (55)	9 (30)	NS	≤.001
SEl-H	0 (0)	43 (43)	3 (10)	≤.001	.04
SEl-I	17 (68)	48 (48)	10 (33)	NS	≤.001
SEl-J	3 (12)	69 (69)	8 (27)	≤.001	.01
SEl-K	7 (28)	49 (49)	23 (77)	NS	≤.001
SEl-L	4 (16)	29 (29)	29 (97)	NS	≤.001
SEl-M	17 (68)	64 (64)	7 (23)	NS	≤.001
SEl-N	17 (68)	51 (51)	5 (17)	NS	≤.001
SEl-O	15 (60)	37 (37)	2 (7)	.04	≤.001
SEl-Q	7 (28)	40 (40)	18 (60)	NS	≤.001
SEl-R	2 (8)	ND	ND	…	…
SEl-S	0 (0)	ND	ND	…	…
SEl-T	2 (8)	ND	ND	…	…
SEl-U	15 (60)	ND	ND	…	…
SEl-X	22 (88)	ND	ND	…	…

Abbreviations: ND, not determined; NS, not significant; SE, staphylococcal enterotoxin; SEl, staphylococcal enterotoxin-like toxin; TSST, toxic shock syndrome toxin.

The genes encoding SEl-O, SEl-M, SEl-I, SEl-U, SEl-N, and SEG are, in their order in the genome, often found together as an enterotoxin gene cluster (egc). Sixteen isolates (64%) contained all 6 genes of an intact egc (Figure 2), while 5 additional isolates (20%) contained fragments of the egc (Table 3). Thus, all or part of the egc cluster was present in 21 of 25 isolates (84%). We did not detect genes encoding SEE, SEl-H, and SEl-S in the 25 DFU isolates. TSST-1, SEB, and SEC are the causative agents of most S. aureus TSS cases. Six isolates (24%) were positive for at least one of these toxins (Table 3). Only 1 isolate carried genes for both TSST-1 and SEC. No isolate carried the genes for SEC and SEB together or TSST-1 and SEB together, which is often the case in S. aureus SAg production.

Chromosomal Localization of tstH and selx in S. aureus Isolates With Both Genes

The TSST-1 gene (tstH) is located in mobile genomic elements called S. aureus pathogenicity islands (SaPIs). Among the human S. aureus isolates, tstH is carried on SaPI1, SaPI2, and SaPIn1 (or SaPIm1) [24]. Unlike tstH and the rest of the SAg genes, selx resides in the S. aureus core chromosome, where it is flanked by 2 hypothetical proteins of unknown function [11]. So far, strains carrying tstH (belonging to the USA200, CC30 clonal groups) exclude selx by an unknown mechanism [11]. Hence, we examined the location of tstH and selx in the DFU isolate carrying both genes. Amplification and sequencing of the upstream and downstream regions of both genes indicated normal positioning of tstH and selx within the chromosome. tstH was located on SAPIn1, which contains both tstH and sec, and selx was located within the core genome cluster specific for S. aureus [11].

Comparison of SAg Gene Profiles Among DFU and Other Skin/Mucosal Isolates

S. aureus SAgs have significant roles in many diseases [9]. However, the SAg gene profile is not conserved among S. aureus isolates, and different diseases, such as TSS and food poisoning, are associated with distinct subsets of SAg genes [9]. Thus, we compared the DFU SAg gene profile with the profiles of 2 groups previously published: 100 isolates from the skin of patients with AD and 30 vaginal mucosal isolates from healthy women.

The genes selp, selr–u, and selx were tested in the DFU isolates, but comparison of these gene frequencies with those in previous isolates was omitted since the gene sequences were not known at the time of publication of test results for the previous isolates, in 2008. The SAg gene profile of the DFU strains most closely resembled that of the AD skin isolates, with the exception of genes in the egc, which were overrepresented in the DFU isolates (Table 4). The SAg gene profile of DFU isolates differed significantly from vaginal mucosal isolates. DFU strains had high frequencies of sed and egc carriage but had a lower tendency to carry the rest of the SAg genes. Overall, S. aureus isolates from patients with DFUs and patients with AD carried the highest number of SAg genes, at an average of 6–7 genes per isolate, while vaginal mucosal isolates contained only 4 SAg genes per isolate (P < .05, by the Student t test).

TSST-1, SEB, SEC, and α-Toxin Production Levels

TSST-1, SEB, and SEC have been previously quantified in AD and vaginal mucosal isolates during growth in liquid culture and found to be produced in concentrations of 3−20 µg/mL for TSST-1, 25−80 µg/mL for SEB, and 40−80 µg/mL for SEC, with no statistical difference observed in SAg production across strains in the different groups [14]. We evaluated TSST-1, SEB, and SEC production in the DFU isolates in liquid culture to assess potential differences in SAg production. The ranges were consistent with those previously reported for AD and vaginal isolates: 3–39 µg/mL for TSST-1, 40–120 µg/mL for SEB, and 10–120 µg/mL for SEC. SED and SEl-X production was also detected by Western immunoblot in the DFU isolates (Supplementary Figure 1). Five randomly chosen DFU isolates produced approximately 10 µg/mL of SED. Three of the same DFU isolates produced 10–20 µg/mL of SEl-X; the remaining 2 isolates did not produce detectable SEl-X.

α-toxin is a significant virulence factor in S. aureus skin infections [25]. The toxin has been reported to synergize with SAgs in some disease settings [26]. α-toxin production was also evaluated among the 25 DFU isolates (Table 5). α-toxin levels varied widely, ranging from <5 µg/mL to 400 µg/mL, with an average of 123 µg/mL. USA300 isolates had the highest production, averaging 350 µg/mL (range, 320–400 µg/mL), whereas USA200 isolates produced the lowest levels, averaging 30 µg/mL (range, <5 to 66 µg/mL). These results are consistent with previous reports of low-level α-toxin production in most USA200 organisms due to the presence of a stop codon at position 113 [27]. USA100 and USA400 isolates produced mid-level amounts of α-toxin, with averages of 35 µg/mL and 27 µg/mL, respectively.

Table 5.

α-Toxin Production in 25 Staphylococcus aureus Isolates From Patients With Diabetic Foot Ulcers and in USA100, 200, 300, and 400 Clonal Groups

Isolates	α-Toxin Level, µg/mL, Range (Mean)
All	<5 to 400 (123)
USA 100	<5 to 187 (60)
USA 200	<5 to 66 (30)
USA 300	320–400 (350)
USA 400	42–180 (102)

DISCUSSION

S. aureus remains the most common isolate in DFU infections. Among S. aureus virulence factors, SAgs have major roles in many diseases. Although the prevalence of S. aureus SAgs has been investigated in illnesses such as endocarditis and AD [14, 28], there are no reports that assess these exotoxins in DFU infections. Hence, the present study is a comprehensive analysis of S. aureus SAg profile in patients with DFU.

Among S. aureus isolates from patients with DFUs, USA100 was the dominant clonal group. S. aureus USA100 is commonly known to colonize human anterior nares and skin [29, 30]. The high prevalence of this clonal group suggests the infecting strain was likely of endogenous origin. Our experience and the experience of others [7, 31] is that diabetic patients are heavily colonized with S. aureus strains, in quantities as high as 10¹³/person, making these organisms readily available for DFU infections.

Our studies also show that individual S. aureus isolates from DFUs have the capacity to produce large numbers of SAgs. Interestingly, when compared to S. aureus isolates from other sources (AD and normal vaginal mucosa), the DFU isolate SAg gene profile most resembles that of AD isolates, both in the ability to produce large numbers of SAg types per organism and the distribution of each SAg. This suggests that the DFU isolates primarily originated from and were adapted to skin as opposed to mucosae, where fewer SAgs are produced. Each SAg activates a relatively unique subset of T cells expressing distinct Vβ-TCR regions. Hence, S. aureus strains producing a wide variety of SAgs would activate larger numbers of T cells to produce proinflammatory cytokines than strains that produce fewer SAgs. This could lead to chronic inflammatory states in DFUs, resulting in delayed or absent wound healing [32]. The high numbers of SAgs may also predispose the patients to more serious disease consequences. Studies have shown that approximately 20% of the adult population lacks antibodies to TSST-1 [33]; these individuals are susceptible to TSS [34]. If a significant part of the population lacks antibodies to any given SAg, production of large numbers of SAgs by individual organisms increases the likelihood that the infecting isolate can cause immune dysfunction without neutralization by existent antibodies.

Several important observations regarding the DFU isolate SAg gene profiles were made. First, 40% of the isolates contained the gene encoding SED. This is similar to findings for AD isolates but greater than the frequency among mucosal isolates. The SED gene is often located on a plasmid, and the active protein is structurally similar to SEA [12]. A recent study show that SEA could have a major role in AD by inducing the upregulation of adhesion molecules and eliciting inflammatory responses in endothelial cells and keratinocytes [35]. Thus, SED may be selected for in DFU isolates because of its enhanced ability to induce local inflammatory responses, similar to SEA.

SEl-X is a new member of the S. aureus SAg family, and it has been shown to have an important role in S. aureus necrotizing pneumonia infection caused by USA300 MRSA [11]. Interestingly, 88% of the DFU isolates carried the gene for SEl-X, the remainder contained the gene for TSST-1, and 1 isolate had genes for both SEl-X and TSST-1. Typically, S. aureus strains have the gene for either SEl-X or TSST-1. The reason for this apparent gene exclusion is unknown. It is interesting from an evolutionary perspective that we have identified at least 1 strain now that carried the genes for both SAgs. This strain contained both genes in their expected chromosomal locations. selx is located in the core chromosome of all S. aureus strains except USA200 strains, which encode TSST-1 within pathogenicity islands.

Comparison with findings from our 2008 AD study [14], it appears that TSST-1–expressing strains have become underrepresented in 2013 isolates, since fewer TSST-1–expressing strains were currently detected. Together, seg, seli, selm, seln, selo, and/or selu, are referred to as S. aureus egc. Approximately 84% of the DFU isolates encoded the egc genes. Additionally, among isolates carrying the egc, >90% had an intact cluster. Such findings differ in many respects with results of other studies in which the egc cluster in isolates from AD skin lesions [14], normal vaginal mucosa [36], blood [37], and food products [38] was fragmented; seg, sei, and seln were often most prevalent in those prior studies. The differences among the egc types [36] and the variance within each SAg [39] could be due to the screening primers used. However, it is more likely that an intact egc cluster is selected for in recent isolates, compared with prior isolates, because it confers a selective advantage in the DFU environment.

Collectively, the present study suggests that the current S. aureus isolates from patients with DFUs are selected for their greater capability to produce SAgs. These proteins are critical virulence factors for S. aureus strains and may be important for patient colonization and overall development of chronic DFU wounds. The data suggest that therapies to neutralize or reduce SAg production by S. aureus may be beneficial in the management of patients with DFU.

Supplementary Data

Supplementary materials are available at The Journal of Infectious Diseases online (http://jid.oxfordjournals.org). Supplementary materials consist of data provided by the author that are published to benefit the reader. The posted materials are not copyedited. The contents of all supplementary data are the sole responsibility of the authors. Questions or messages regarding errors should be addressed to the author.