Correspondence: Harsh Vardhan Batra, Microbiology Division, Defence Food Research Laboratory, Siddarthanagar, Mysore, Karnataka 570011, India. Tel/ Fax.: +91821-2579435; e-mail: firstname.lastname@example.org
Staphylococcus aureus and Clostridium perfringens are two major bacteria that infect open wounds and delay the healing process. The rapid and progressive deterioration of soft tissue during S. aureus and C. perfringens coinfections is due to analogous necrotic alpha toxins produced by the two organisms. The aim of this study was to determine the alpha toxins of S. aureus and C. perfringens by duplex PCR. The PCR assay employed two sets of primers: hlaf/r to amplify staphylococcal alpha toxin gene hla (274 bp) and cpaf/r to amplify clostridial alpha toxin gene cpa (398 bp) along with a competitive internal amplification control (608 bp), simultaneously. Optimization of the duplex PCR assay was achieved by a modified Taguchi method, an engineering optimization process, in a nine-tube combinatorial array. The detection level of the duplex PCR was found to be 10 pg of purified DNA or 103 CFU mL−1 of S. aureus and 100 pg of purified DNA or 104 CFU mL−1 of C. perfringens. Other bacteria routinely found in tissue infections were tested for cross-reactivity and the duplex PCR turned out to be highly specific. This duplex PCR assay provides a rapid, robust and reliable alternative to the existing conventional techniques in establishing the aetiology of S. aureus and C. perfringens in soft tissue infections.
Acute soft tissue infections such as cutaneous abscesses, traumatic wounds and necrotizing infections are dangerous forms of bacterial infections with high morbidity and occasional mortality (Bowler et al., 2001). The austerity and progression of these infections are known to be affected significantly by polymicrobial aerobic–anaerobic interactions of varied aetiology (Sun et al., 2009). The most frequently isolated aerobic (facultative) and anaerobic bacteria from these infections are Staphylococcus aureus and Clostridium perfringens (Titball et al., 2002; Harrington et al., 2010), respectively. Growth of S. aureus and other aerobic microbes in soft tissue infections induces a hypoxic/anoxic environment leading to ischaemia and subsequent colonization by a wide variety of endogenous anaerobic bacteria, most common among these being C. perfringens (Bowler et al., 2001). The amalgam of toxins secreted during the coinfections of these lethal organisms causes gangrenous ischaemia, necrotizing fasciitis, etc., in animals and humans (Ribeiro et al., 2007).
Staphylococcus aureus is a Gram-positive, facultative anaerobe commonly inhabiting food, soil, skin and any damaged tissue (Franklin & Lowy, 1998). The bacterium produces a huge arsenal of exotoxins (enzymes such as nucleases, lipases and proteases; cytotoxins such as alpha, beta, gamma and delta haemolysins; and enterotoxins such as staphylococcal enterotoxins A and B) that contribute to its pathogenicity (Dinges et al., 2000). Alpha toxin, the major cytotoxic agent elaborated by S. aureus, forms heptameric pores by nonspecifically binding onto membranes of erythrocytes and other nucleated cells (Bernheimer et al., 1972) and causes many diseases such as dermonecrosis, staphylococcal pneumonia and mastitis (Bhakdi & Tranum-Jensen, 1991). This toxin, which is encoded by the chromosomally located hla gene, is the first bacterial toxin to be identified as a pore former. It induces arachidonic acid metabolism in mammals, leading to vasoconstriction and cell death resulting in tissue necrosis (Dinges et al., 2000).
Clostridium perfringens is a Gram-positive, spore-forming anaerobe found in food, vegetation, soil, and the gastrointestinal tracts of humans and animals. The organism is classified into five toxinotypes (A, B, C, D and E) based on the production of four major toxins, namely alpha, beta, epsilon and iota toxins (Rood, 1998). Alpha toxin, produced by all types of C. perfringens, is responsible for gas gangrene disease, a condition where necrotizing tissue is associated with gas production. This is the first bacterial toxin demonstrated to be an enzyme (MacFarlane & Knight, 1941). Clostridial alpha toxin, encoded by the chromosomally located cpa gene, possesses various biological activities, including haemolytic, dermonecrotic, phospholipase C and sphingomyelinase activities (Nagahama et al., 1995). It plays a major role in the pathogenesis of gangrene disease by mistrafficking the neutrophils, preventing their entry into infected tissues and triggering vasoconstriction and platelet aggregation, thus reducing blood supply to the infected tissues (Titball et al., 1999).
The staphylococcal and clostridial alpha toxins both have haemolytic and myonecrotic activity with remarkably similar pathology. During S. aureus and C. perfringens coinfections, the rapid deterioration of tissue and gas production by clostridia occasionally masks the staphylococcal infection. Consequently, use of selective culture media to specifically isolate aerobic or anaerobic microbes tends to bias the results. On the other hand, many strains of S. aureus are nonpathogenic and are ubiquitous in nature. Hence, determining the mere presence of S. aureus in poly/monoculture infections does not necessarily establish it as the aetiologic agent, unless proven to be toxigenic. Current diagnostic systems used to perform microbiological analysis in hospitals solely determine the presence of bacteria in the tissue specimens rather than their virulence potential and, often, reporting a ‘mixed culture’ is misinterpreted as being associated with poor sampling or natural contamination rather than emphasizing the significance of microbial synergy (Bowler et al., 2001). Accordingly, we previously developed a Western blot/dot enzyme-linked immunosorbent assay (ELISA)-based detection system for concurrent diagnosis of alpha toxins of S. aureus and C. perfringens (Uppalapati et al., 2012). In the current study, a duplex PCR was standardized to simultaneously detect alpha toxins of S. aureus and C. perfringens by employing a modified Taguchi process optimization method. This duplex PCR was found to be robust and sensitive with a minimum detection level of 10 pg DNA/103 cells of S. aureus and 100 pg DNA/104 cells of C. perfringens.
Materials and methods
Bacterial strains, culture conditions and culture media
All the bacterial strains used in this study are listed in Table 1. Standard strains were procured from bacterial collection centres such as American Type Culture Collection (ATCC), USA, Microbial Type Culture Collection (MTCC) and National Collection of Industrial Microorganisms (NCIM), India. DFR strains of S. aureus and C. perfringens were isolated at the Defence Food Research Laboratory, Mysore, from different food and environmental samples. Twelve S. aureus and seven C. perfringens clinical strains were kindly provided by Microbiology Department, SDM Medical College, Dharwad, India, and Government Veterinary College, Mysore, India, respectively. The media and media components were procured from Himedia (Mumbai, India). All the bacterial cultures were maintained in brain heart infusion (BHI) broth and incubated at 37 °C with shaking, except Clostridia, which were maintained in Fluid Thioglycollate broth and incubated in anaerobic jars (Himedia) at 37 °C.
For all standardization experiments, total DNA of bacterial strains was isolated following Marmur's protocol (Marmur, 1961) with minor modifications. Briefly, 3 mL overnight grown bacterial culture pellet was resuspended in 300 μL lysis buffer (50 mM Tris-HCl, 50 mM EDTA, 1% sodium dodecyl sulphate, 10 mM NaCl, pH 8.0) and incubated at room temperature for 45 min. DNA was isolated by phenol-chloroform extraction followed by ethanol precipitation. The DNA pellet was dissolved in 100 μL distilled water and stored at −20 °C until further use. DNA from broth cultures was also extracted by thermal lysis to verify the robustness of the PCR. Bacterial pellets from overnight cultures or artificially contaminated (spiked) samples were resuspended in distilled water and boiled for 10 min at 95 °C. The debris was pelleted and 3 μL/5 μL supernatant was used as the source of template for amplification. All standardization experiments were performed with DNA/cultures of strains S. aureus ATCC 6538P and C. perfringens ATCC 13124.
Primer and internal amplification control design
The primers used in the study were custom synthesized from Eurofins, Bangalore, India. Nucleotide sequences of hla (accession number NC_009641) and cpa (accession number NC_X13608) genes were retrieved from the NCBI database and corresponding primers were designed following imperatives set by Sambrook et al. (1989) using Primer 3 software (http://simgene.com/Primer3). The hla gene (274 bp) was amplified using primers hlaf (5′-GTACTACAGATATTGGAAGC-3′) and hlar (5′-GTAATCAGATATTTGAGCTAC-3′) and the cpa gene (398 bp) was amplified by primers cpaf (5′-ATGTTACTGCCGTTGATAGC-3′) and cpar (5′-TCCAGCATCTTTCTCACCAC-3′). Multiple sequence alignment of staphylococcal haemolysin (hla) gene sequences retrieved from the NCBI-GenBank database was performed and primers were designed complementary to the conserved region (Supporting Information, Data S1). Primers for the clostridial cpa gene were designed such that they did not incorporate any of the 71 polymorphic sites described recently (Siqueira et al., 2012). To account for false negative results, a competitive internal amplification control (IAC) was synthesized using the composite primer technique (Siebert & Larrick, 1992) with pUC 19 DNA fragment flanked by cpa primer sequences as IAC and cpa gene as its competitive partner. The size of the IAC (608 bp) was made larger than the cpa gene amplicon, to ensure the competitive edge (Hoorfar et al., 2004).
Monoplex PCR amplification
The monoplex PCRs for hla, cpa and IAC genes were performed in Mastercycler Pro thermalcycler (Eppendorf, Germany) in a 20-μL reaction mix containing 50 ng template DNA, 1 × PCR buffer (with 1.5 mM MgCl2), 200 μM each dNTP, 1 μM each primer and 1 unit Taq polymerase (Fermentas, New Delhi, India). The PCR conditions were as follows: initial denaturation at 94 °C for 4 min, 32 cycles of denaturation at 94 °C for 30 s, annealing at 56 °C for 30 s and extension at 72 °C for 30 s, and a final extension at 72 °C for 4 min. The PCR amplicons were electrophoresed in 1.2% agarose gel, stained with ethidium bromide and visualized under UV transillumination (G-box, Syngene, India).
Modified Taguchi method (standardization of duplex PCR)
Duplex PCR was standardized following the modified Taguchi method described by Cobb & Clarkson (1994). In this method, critical factors that affect a process were identified and compositions of each factor were varied in a balanced orthogonal array for standardization. Four critical factors were selected that may affect the amplification during duplex PCR: concentrations of IAC template, competitive DNA template (C. perfringens DNA), and cpa and hla gene primers. Each factor had three levels or concentrations (IAC: 10, 1 and 0.1 pg; competitive DNA: 10, 1 and 0.1 ng; cpa and hla primer pairs: 1.5, 1 and 0.5 μM) and all the parameters were distributed in a balanced orthogonal array of nine experimental combinations (Supporting Information, Data S2) as described by Cobb & Clarkson (1994). The PCR conditions were same as described above. The PCR amplicons were electrophoresed in 1.5% agarose gel, stained with ethidium bromide, documented (G-Box, Syngene, India) and electropherograms (Supporting Information, Data S3) were constructed by GeneSys image acquisition software (Genetools, Syngene, India).
Sensitivity of duplex PCR
Two different strategies were used to determine the sensitivity of the duplex PCR assay. In the first strategy, equal concentrations (100 ng each) of purified DNA from S. aureus and C. perfringens were combined and 10-fold serially diluted in water to obtain amounts ranging from 100 to 1 pg each DNA, and used in duplex PCR assay. In the second strategy, to determine the detection sensitivity of pure cultures, 8–12 h grown cultures of S. aureus and C. perfringens were combined in equal concentrations (106 cells each) and 10-fold serially diluted in nutrient broth. Bacterial cells from 3 mL of each dilution were centrifuged at 12 000 g for 5 min, resuspended in 50 μL distilled water and DNA was extracted by thermal lysis. For analysis, 3 μL of each sample was used in duplex PCR assay.
Specificity of duplex PCR
Specificity of the primers was confirmed by performing standardized duplex PCR with other nonspecific bacterial DNA. Also, to assess interference of this nonspecific bacterial DNA on the sensitivity of the alpha toxin gene duplex PCR, 100 ng of purified DNA from Bacillus cereus ATCC 14579, Escherichia coli DH5α, Streptococcus faecalis NCIM 2405, Staphylococcus epidermidis MTCC 435 and Clostridium difficile ATCC 9689 were added to the detectable amounts of the S. aureus DNA (10 pg) and C. perfringens (100 pg) DNA and the ability of the duplex PCR to amplify the products was assessed. As nonspecific DNA negative control, a PCR reaction was performed only with S. aureus and C. perfringens DNA.
To estimate the interference of tissue matrices on the sensitivity, detectable concentrations of pure cultures of S. aureus and C. perfringens (106 cells) were 10-fold serially diluted and 5 mL each dilution was artificially inoculated (spiked) onto 10 g minced sheep meat. Ten millilitres of sterile nutrient broth was added to the spiked samples, stomached thoroughly and incubated at 37 °C for 1 h. Later, debris-free supernatant was collected and centrifuged at 3000 g for 30 s. The supernatant was collected and centrifuged again at 12 000 g for 5 min. The pellet was resuspended in 50 μL distilled water and treated by thermal lysis to extract DNA. For analysis, 5 μL of each sample was included in duplex PCR assay. A PCR mixture containing water was used as no-template control. All the PCR products in sensitivity and specificity studies were electrophoresed on 1.2% agarose gel, documented and DNA bands were quantified as raw volume, based on densitometric scanning of the image by Genetools software. The raw volume of PCR products was calculated from the area of the fluorescent intensity peak of each band in the electropherograms.
All the experiments in the study were performed in triplicate and the values are given as means ± SD. Graphical illustrations were depicted using Microsoft Excel 2007.
Alpha toxin duplex PCR optimization and specificity
During the development of duplex PCR, optimal amplification conditions for each individual gene were construed initially by empirical permutations of annealing temperatures and MgCl2 concentrations. Although amplification of both hla and cpa genes without any specious products was observed in the majority of the combinatorial conditions, at 56 °C annealing temperature and 1.5 mM MgCl2 concentration, both amplicons had the highest yield. The same conditions were utilized for the amplification of IAC as well. Employing a modified Taguchi method, alpha toxin duplex PCR was standardized. Four critical factors were chosen in these experiment, namely concentrations of cpa primers, hla primers, IAC template and C. perfringens total DNA. The nine-tube standardization protocol ('Materials and methods') resulted in varied amplification patterns when the products were run on 1.2% agarose gel. To determine the relative concentrations of each band from the pattern, electropherograms were constructed by densitometric analysis of agarose gel images using Genetools software. The graphs (Data S3) demonstrated the competition between the IAC and cpa gene, and IAC concentration was found to be critical in determining the sensitivity of its competitive counterpart, the cpa gene. Optimally, it was found that IAC at a concentration of 0.1 pg resulted in highest amplification of all the PCR products. Primers hlaf/r did not interfere with amplification of the cpa gene or IAC in any combination. The optimal amplification conditions deduced from the modified Taguchi method were employed in further experiments. The agarose gel pattern of PCR amplicons is shown in Fig. 1. Specificity of the hla and cpa primers was assessed by performing alpha toxin duplex PCR on the genomic DNA of bacterial strains listed in Table 1. Agarose gel electrophoresis (Fig. 2a and b) confirmed the presence of amplicons at the expected molecular weights (274 bp for hla and 398 bp for cpa) when S. aureus and C. perfringens strains were used in the PCR assay. No spurious products were observed with other organisms. The IAC gene was amplified in all conditions, demonstrating that PCR mix and conditions were absolute. The presence of staphylococcal and clostridial haemolysins was also confirmed by observing the haemolysis zones around inoculated cultures on sheep blood agar plates (sata not shown). Four staphylococcal isolates (MYCSA 7, MYCSA 8, DFRSA 9 and DFRSA 12) were negative for the hla gene as well as haemolysis on sheep blood agar.
Sensitivity of duplex PCR
The minimum detection level of purified S. aureus and C. perfringens total genomic DNA by alpha toxin duplex PCR was observed to be 0.01 and 0.1 ng, respectively (Fig. 3a). When broth culture dilutions were used, the sensitivity of the PCR was 103 CFU mL−1 for S. aureus and 104 CFU mL−1 for C. perfringens (Fig. 3b). Although the cpa gene was amplified even with 103C. perfringens CFU/mL, the amplicon concentration was so low that band calling could be ambiguous. No amplicons except IAC were observed when 102 CFU mL−1 of either of the cultures was examined by PCR.
Interference of nonspecific bacterial DNA/tissue matrix
The robustness of a diagnostic PCR is dependent on its ability to amplify the target from a pool of non-specific DNA. In the current duplex PCR, even when 1 ng each S. aureus and C. perfringens DNA mixture was investigated in a background of 100 ng nonspecific DNA (B. cereus ATCC 14579, E. coli DH5α, Strep. faecalis NCIM 2405, S. epidermidis MTCC 435 and C. difficile ATCC 9689), the amplicon concentrations did not vary notably from the nonspecific DNA negative control (Fig. 4a). When the bacterial cultures were artificially inoculated onto meat particles and the DNA extracted from the particle suspension was analysed, a detectable sensitivity of 103 CFU mL−1 for both cultures was achieved by the duplex PCR (Fig. 4b).
Wound/devitalized tissue provides a favourable environment for microbial colonization and the variety of microoragnisms involved is influenced by many factors such as wound type (acute/chronic), depth, location, the level of tissue perfusion and the efficacy of the host immune response (Bowler et al., 2001). The relevance of culturing clinical tissue specimens for polymicrobial existence is important because definitive diagnosis of the pathogens greatly facilitates effective wound management and has been stressed by many investigators (Raahave et al., 1986; Lavery et al., 1994; Heggers, 1998). Conventional identification (Le Loir et al., 2003; Gracias & McKillip, 2004; Wu, 2008) of S. aureus and C. perfringens coinfections in soft tissues has its own limitations; merely reporting the presence does not completely establish the aetiology because of their ubiquity and variable virulence. A more definitive approach would be to identify the key toxins produced by them in wound infections. The current duplex PCR concurrently and specifically identified two analogous toxin-associated genes, namely staphylococcal and clostridial alpha toxins (hla and cpa genes), thereby also ascertaining the definitive involvement of S. aureus and C. perfringens with soft tissue infections.
The reproducibility of the PCR results is greatly influenced by the efficiency of the thermal cyclers and DNA polymerases, and by PCR contaminants and faulty PCR mixes (Hoorfar et al., 2003). These factors can be efficiently dealt with by integration of an IAC in the diagnostic PCRs. The IAC is independently amplified, irrespective of the sample type and nucleic acid load in the PCR master mix, thus accounting for the false negative results. Hoorfar et al. (2003) emphasized the importance of IAC in diagnostic PCRs, since when two major types of IACs were discovered: competitive and noncompetitive. Generally, competitive IAC is favourable over noncompetitive IAC as the latter requires standardizing extra reaction kinetics. On the other hand, competitive IAC amplifies with the same set of primers already used in the PCR mix under the same conditions. In the current study, a competitive IAC was designed from pUC 19 plasmid with overhangs corresponding to the cpa gene following the composite primer technique (Siebert & Larrick, 1992) and was successfully incorporated into the alpha toxin duplex PCR.
The PCR conditions were standardized by applying the modified Taguchi methods (Cobb & Clarkson, 1994) to successfully determine the optimal quantities of various components. Many traditional experimental design techniques have been described, including full factorial design, 2k factorial design, central composite, latin square and Plackett–Burman design, besides the Taguchi method (Morgan, 1991), but the last-named has found widespread use in PCR optimization experiments (Niens et al., 2005; Thanakiatkrai & Welch, 2012). Using this method, the optimal conditions for a particular process can be established using a minimum number of experiments by arranging the critical factors in an orthogonal array. The major critical factors in a monoplex PCR would logically be MgCl2 concentration, annealing time and temperature whereas in a multiplex PCR, primer mix, target DNA template and IAC concentrations are the key aspects that affect efficiency of PCR. We standardized alpha toxin duplex PCR in a nine-tube format with combinatorial concentrations of primer mix, target DNA template and IAC. During the PCR standardization, the competition by IAC lowered the sensitivity of the cpa gene. At 1 pg concentration of IAC, cpa gene detection sensitivity was observed to be only 10 ng, whereas the decrease in the quantity of IAC template (from 1 pg to 10 fg) significantly improved the sensitivity up to 0.1 ng (Data S3). This may be due to preferential amplification of IAC sequence over the cpa gene as annealing of primers to IAC is faster and easier than to the cpa gene from the whole genome of C. perfringens. On the other hand, when 100 ng C. perfringens genome is used as template in the PCR, faint amplification of IAC was observed (Fig. 3a). Often, the target DNA amplifies without IAC amplification, if the former is present in a ‘proportionally greater amount’ and when this occurs, the positive result is bona fide (Hoorfar et al., 2004). Also, in all cases, neither IAC nor C. perfringens genome concentrations interfered with the detection sensitivity of S. aureus. The modified Taguchi method therefore proved extremely useful in effective standardization of the duplex PCR for S. aureus and C. perfringens alpha toxin genes along with an appropriate IAC.
The currently developed alpha toxin duplex PCR is rapid, reliable and robust. This method, although not a substitute for classical experimental designs, if used as a supplementary strategy can make polymicrobial analysis of tissue infections efficient with respect to time, cost and accuracy. The modified Taguchi method employed in standardization of this duplex PCR is extremely useful for rapid designing of diagnostic mPCRs.
The authors state no conflict of interest. S.R.U. is supported by the Junior Research Fellowship of Council for Scientific and Industrial Research, Government of India.