Development of a magnetic separation method to ...

Journal of Microbiological Methods 128 (2016) 96–101

Contents lists available at ScienceDirect

Journal of Microbiological Methods journal homepage: www.elsevier.com/locate/jmicmeth

Development of a magnetic separation method to capture sepsis associated bacteria in blood Ana Luisa Kalb Lopes a,⁎, Josiane Cardoso a, Fernanda Roberta Correa Cleto dos Santos a, Ana Claudia Graziani Silva a, Maria Isabel Stets a, Nilson Ivo Tonin Zanchin b, Maurilio José Soares c, Marco Aurélio Krieger a a b c

Instituto de Biologia Molecular do Paraná, Department of Research and Development, Prof. Algacyr Munhoz Mader Street 3775, 81350-010 Curitiba, PR, Brazil Laboratory of Proteomic and Protein Engineering, Carlos Chagas Institute, Fiocruz, Prof. Algacyr Munhoz Mader Street 3775, 81350-010 Curitiba, PR, Brazil Laboratory of Cell Biology, Carlos Chagas Institute, Fiocruz, Prof. Algacyr Munhoz Mader Street 3775, 81350-010 Curitiba, PR, Brazil

a r t i c l e

i n f o

Article history: Received 30 May 2016 Received in revised form 14 July 2016 Accepted 15 July 2016 Available online 16 July 2016 Keywords: Sepsis S. aureus Lysozyme Magnetic separation

a b s t r a c t Bloodstream infections are important public health problems, associated with high mortality due to the inability to detect the pathogen quickly in the early stages of infection. Such inability has led to a growing interest in the development of a rapid, sensitive, and specific assay to detect these pathogens. In an effort to improve diagnostic efficiency, we present here a magnetic separation method for bacteria that is based on mutated lysozyme (LysE35A) to capture S. aureus from whole blood. LysE35A-coated beads were able to bind different MSSA and MRSA isolates in the blood and also other six Gram-positive and two Gram-negative species in whole blood. This system was capable to bind bacteria at low concentrations (10 CFU/ml) in spiked blood. Samples captured with the mutated lysozyme showed more responsive amplification of the 16S gene than whole blood at concentrations of 103–105 CFU. These data demonstrate detection of S. aureus directly in blood samples, without in vitro cultivation. Our results show that capture with LysE35A-coated beads can be useful to develop a point of care diagnostic system for rapid and sensitive detection of pathogens in clinical settings. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Bacterial pathogens have a significant impact on human health, especially those causing bloodstream infection, a relevant cause of sepsis. This condition usually requires at least one day or more for precise diagnosis, increasing the chances of patient mortality (Liesenfeld et al., 2014). The extremely high mortality by blood infections is due, in part, to the inability to rapidly detect, identify and thus treat patients with appropriate antibiotics in the early stages of infection (Carrigan et al., 2004). Consequently, considerable effort has been devoted to the development of rapid, sensitive, and specific assays to detect these pathogens. Amplification-based molecular diagnosis methods such as PCR can reduce the assay time to hours. However, this methodology is often not sensitive enough to detect low concentrations of bacteria, thus needing additional steps such as initial enrichment (D.K. Kang et al., 2014), as well as dilution, inactivation or removal of inhibitors from the sample (Fredricks and Relman, 1998; Yamamoto, 2002). Magnetic separation is an alternative for the isolation of target cells directly ⁎ Corresponding author at: Instituto de Biologia Molecular do Paraná, Department of Research and Development, 3375 Professor Algacyr Munhoz Mader Street, Curitiba, Paraná, Brazil. E-mail addresses: [email protected], [email protected] (A.L.K. Lopes).

http://dx.doi.org/10.1016/j.mimet.2016.07.012 0167-7012/© 2016 Elsevier B.V. All rights reserved.

from samples, by eliminating the components that interfere with the PCR and related techniques (Olsvik et al., 1994; Šafařı́k and Šafařı́ková, 1999). Magnetic separation has several applications for the detection of pathogenic microorganisms, especially in food (Tomoyasu, 1992; Opsteegh et al., 2010), clinical (Kassimi et al., 2002; Nam et al., 2013), veterinary (Coklin et al., 2011; Isaksson et al., 2014) and environmental microbiology (Yakub and Stadterman-Knauer, 2004; Sierra et al., 2014). This methodology uses small magnetic particles coated with antibodies, peptides or oligonucleotides to bind cell surfaces and has been shown to be efficient for the isolation of different organisms (Šafařı́k and Šafařı́ková, 1999). Magnetic separation of bacteria using mutated lysozyme coupled to magnetic beads allows specific capture from cell suspension, by recognition of bacterial cell wall peptidoglycans by the lysozyme-beads complex (Diler et al., 2011). Lysozyme catalyses the breakdown of peptidoglycans of bacterial cell wall, lysing sensitive bacteria (Benkerroum, 2008). Two amino acids are essential for this lytic activity: glutamate at position 35 (E35) and aspartate at position 52 (D52) (Malcolm et al., 1989). Mutation of glutamate-35 to its corresponding amide alanine (LysE35A) completely abolished bacteriolytic activity, whereas the affinity for target structures is maintained (Diler et al., 2011). Here we used Staphylococcus aureus as a model pathogen to demonstrate the feasibility of magnetic separation for bacteria in whole blood.

A.L.K. Lopes et al. / Journal of Microbiological Methods 128 (2016) 96–101

This Gram-positive bacterium is a common causative pathogen of bloodstream infections and is associated with high morbidity and mortality (del Rio et al., 2009; Corey, 2009). The occurrence of drug-resistant and/or highly virulent strains reinforces their public health threat. Methicillin-resistant Staphylococcus aureus (MRSA) is resistant to all existing penicillin and lactam-based antimicrobial drugs and has become one of the most prevalent antibiotic-resistant pathogens in hospitals (Pray, 2008), which emphasizes the importance of rapid initiation of appropriate treatment. Thus, we present here a magnetic separation method for bacteria that uses LysE35A to capture S. aureus from whole blood that can be used as an enrichment step for PCR. The mutated lysozyme was expressed in Pichia pastoris and immobilized in magnetic beads after purification by cation-exchange chromatography. Our data demonstrate detection of S. aureus directly in blood samples, without in vitro cultivation. 2. Material and methods 2.1. Microorganisms The methylotrophic Pichia pastoris strain GS115 (his4, mut+) and the control strain for secreted expression GS115 albumin (HIS4, murS) were used for protein expression. The DH5α strain of Escherichia coli was used for plasmid selection/propagation. Both strains were purchased from Invitrogen (Karlsruhe, Germany). pPicZα vector (Invitrogen, Carlsbad, CA, USA) was used for secreted expression. Pichia pastoris was cryopreserved at − 80 °C in 15% (v/v) glycerol and cultivated on YPD medium [1% (w/v) yeast extract, 2% (w/v) peptone, 2% (w/v) glucose and 2% (w/v) agar]. For protein expression we used medium BMGY [1% yeast extract, 2% peptone, 100 mM potassium phosphate, pH 6.0, 1.34% (w/v) YNB (Yeast Nitrogen Base, Invitrogen, USA), 4 × 10−5% (w/v) biotin and 1% (v/v) glycerol]. Alternatively, BMMY medium was used (the same as BMGY, except that glycerol was replaced by 1% methanol). Staphylococcus aureus, Staphylococcus haemolyticus, Staphylococcus epidermidis, Listeria monocytogenes, Enterococcus faecium, Enterococcus faecalis, Streptococcus agalactiae, Escherichia coli and Klebsiella pneumoniae were isolated from various clinical samples (such as blood, cerebrospinal fluid, urine and pus) collected in Hospital de Clínicas-UFPR, Curitiba, Brazil. Six S. aureus clinical isolates were obtained from blood cultures, being 3 susceptible (MSSA) and 3 resistant (MRSA) to methicillin. S. aureus ATCC 29213 was used as a reference strain in all capture studies. 2.2. Plasmid construction A synthetic lysozyme gene containing a mutation at position 35 (LysE35A) was synthesized using the services of a commercial provider (GenScript, Piscataway, NJ, USA). This gene and the pPicZα vector were digested by XhoI and NotI and then ligated by T4 DNA ligase. The resulting plasmid was transformed into DH5α cells by heat shock. After confirmation by sequencing (Macrogen Inc. Seoul, Korea), the pPicZα vector containing LysE35A was transformed to GS115 yeast cells by the lithium chloride method (Mount et al., 1996). For PCR control of gene insertion into the AOX1 locus of the host genome, colonies were picked from YPD agar plates and suspended in 10 μl H2O. After addition of 25 U lyticase (Sigma-Aldrich, St. Louis, MO, USA), the suspension was incubated for 10 min at 30 °C, 10 min at − 80 °C and then used in PCR. 2.3. Protein expression, purification and analysis Expression of LysE35A was obtained based on the procedure outlined in Invitrogen's EasySelect Pichia Expression Kit manual, using the MutS strain. A single colony of transformed yeast GS115 cells was inoculated into 40 ml of BMGY medium and incubated at 28–30 °C in a

97

shaking incubator (250 rpm) until the culture reached OD600 ranging from 2 to 6. The cells were harvested by centrifugation and the pellet was resuspended in four liters of BMGY medium, with incubation at 28–30 °C (shaking at 250–300 rpm) until the culture reached OD600 ranging from 2 to 6. Protein expression was induced by resuspending the cells in one liter of BMMY medium, followed by incubation at 28– 30 °C for 120 h in a shaking incubator (250 rpm). To induce and maintain the recombinant protein expression, methanol was added daily to a final concentration of 1%. One sample per day was collected to analyze expression levels by Silver-stained SDS-PAGE and immunoblotting. The recombinant protein was concentrated with 2.6 M ammonium sulfate and the precipitate was dissolved in Tris-HCl buffer pH 8/ 50 mM NaCl/7 mM β-mercaptoethanol. Purification of LysE35A was performed by ion exchange chromatography in AKTA system (GE Healthcare Life Sciences). The resin used was SP Sepharose and the bound protein was eluted with a gradient of 0–1 M NaCl in Tris-HCl buffer pH 8. Analysis of LysE35A expression levels was performed with 17% polyacrylamide-containing SDS gels. For immunoblotting analysis, protein was separated by SDS-PAGE using 17% polyacrylamide gels and the protein bands transferred onto a nitrocellulose membrane (HybondC, Amersham Biosciences, England) according to standard protocols (Sambrook et al., 1989). Nonspecific binding sites were blocked by incubating the membrane for 1 h with 5% nonfat milk and 0.05% Tween-20 in PBS, pH 8.0. The membrane was then incubated for 1 h with polyclonal commercial hen egg white lysozyme (HEWL) antibody (Thermo Scientific, Rockford, IL, USA) (1:10,000 dilution), washed three times with 0.05% Tween-20 in PBS and then incubated for 45 min with anti-rabbit horseradish peroxidase-conjugated IgG (Amersham Biosciences, England). Bound antibodies were detected with the ECL Western blotting analysis system (Amersham Biosciences, England), according to the manufacturer's instructions. The activity of the recombinant proteins was determined by lysoplate assay, as previously described (Maeda et al., 1980). The method was carried out with 250 μg of commercial (positive control) and mutated lysozyme. The plates containing Micrococcus luteus and lysozyme were incubated and examined for clear zones surrounding the disc. 2.4. Cell capture Purified recombinant lysozyme molecules were coupled to Dynabeads M-280 Tosylactivated (Invitrogen) according to the manufacturer's instructions. EDTA treated blood from human volunteers was obtained from the Biobanco of Hospital de Clínicas-UFPR after approval by the Ethics Committee (Protocol number #42012). Fifty microliter of coated or uncoated beads (negative control) were added to 1 ml of a desired concentration (CFU/ml) of bacteria in PBS. Alternatively, whole blood was used. In this case, prior to capture the red cells were removed from the whole blood. Two lysing buffers were tested: Triton/Tris-EDTA (TTE: 20 mM Tris-HCl, pH 8.3, 1 mM EDTA, 1% Triton X-100) and ammonium chloride (NH4CL:155 mM NH4Cl, 10 mM KHCO3, 0.1 mM EDTA pH 7.3). The whole blood was diluted at 1:5 in TTE buffer or 1:1 in NH4Cl buffer. Each tube was vortexed immediately after adding the lysing solution and incubated at room temperature for 10 min. PBS and blood samples were incubated at room temperature for 90 min under stirring. Thereafter, they were placed in a magnetic separator rack and resuspended in 100 μl PBS, plated on PCA medium and incubated for 24 h. Colony counts (CFU per milliliter) were determined for each bacterium after magnetic separation and the capture efficiency was calculated. For qPCR detection, samples were placed in a magnetic separator rack and resuspended in 1 ml PBS, from which 100 μl was plated on PCA medium for quality control and the remaining was used for DNA extraction and PCR. All experiments were performed in triplicate.

98


2.5. DNA extraction and detection by qPCR Bacterial DNA from whole blood and captured cell were isolated with the MolYsis complete 5 kit (Molzym, Germany). DNA eluates were measured using a NanoDrop 8000 spectrophotometer (Thermo Scientific, Rockford, IL). The DNA was submitted to qPCR analysis, as follows: SYBR Green based qPCR (Applied Biosystems, USA), using a Applied Biosystems 7500 system (Applied Biosystems) with a reaction volume of 20 μl composed by 10 μl SYBR green master mix, 1 μl of each of the 16S (5 μM) forward (5′ TCCTACGGGAGGCAGCAG 3′) (Nadkarni et al., 2002) and reverse (5′ GTATTACCGCGGCTGCTGG 3′) (Negoro et al., 2013) primers, and 5 μl of DNA sample. Thermocycling was initiated at 95 °C for 1 min, followed by 40 cycles of 20 s denaturation at 95 °C, 30 s annealing at 60 °C and 30 s extension at 72 °C; TaqMan (Applied Biosystems, USA), using a Applied Biosystems 7500 system with a reaction volume of 20 μl composed by10 μl Taqman master mix, 1 μl of each of the 16S forward and reverse primers (8 μM), 1 μl of 16 μM probe RNAse P (5′ ATTO550-TTCTGACCTGAAGGCTCTGCGCG 3′), 1 μl of each of the RNAseP forward (5′ AGATTTGGACCTGCGAGCG 3′) and reverse (5′ GAGCGGCTGTCTCCACAAGT 3′) primers (Emery et al., 2004), 1 μl of 16 μM probe Gram positive bacteria (5′ FAMCTGAYSSAGCAACGCCGCG 3′), 3 μl water and 1 μl of DNA sample. Thermocycling was initiated at 95 °C for 10 min, followed by 40 cycles of 15 s denaturation at 95 °C, 30 s annealing/extension at 60 °C.

GS115 P. pastoris cells by lithium chloride transformation, as evaluated by PCR with the primer pair 5′AOX1 and 3′AOX1. Expression of LysE35A was confirmed by SDS-PAGE and Western blot (Fig. 1). The protein was enriched by 2.6 M ammonium sulfate precipitation from the expression medium and purified by cation-exchange chromatography. To verify the catalytic activity of the recombinant protein, a lysoplate assay was carried out. Commercial hen egg white lysozyme was used as a positive control and showed a lysis zone that was not observed for the mutated lysozyme, demonstrating that mutation of glutamate 35 led to catalytic inactivity. 3.2. Capture assays

3. Results

To isolate intact bacteria from different matrices, magnetic beads were coated with the mutated lysozyme. Efficiency and specific capture of intact bacteria by LysE35A-coated beads was initially examined in PBS containing 103 CFU and compared to negative controls consisting of uncoated beads (Fig. 2). LysE35A-coated beads were able to capture all six Gram positive bacteria (S. aureus, S. haemolyticus S. epidermidis, L. monocytogenes, E. faecium and E. faecalis) tested. Best results were obtained for S. aureus and E. faecium, with binding of nearly 100% of the cells (Fig. 2). The highest capture efficiency for S. aureus in blood was obtained with 1.5 mg of beads coupled to 30 μg LysE35A in blood diluted five times in TTE buffer, after 90 min of incubation with the bacteria (Fig. S2). Under these conditions, different clinical isolates of Methicillin-susceptible (MSSA) or resistant (MRSA) S. aureus (103 CFU) were used and the capture efficiency was compared to an ATCC strain. Our data indicated that LysE35A coated beads were able to bind MSSA and MRSA isolates in blood, but showed best efficiency with the ATCC strain (Fig. 3A). Next we tested the capture of different bacteria species to verify whether LysE35A-coated beads were able to bind other species under the same conditions. Six Gram-positive and two Gram-negative bacteria species (Escherichia coli and Klebsiella pneumoniae) commonly implicated in sepsis (Angus and van der Poll, 2013) were tested. LysE35A-coated beads were able to bind to all bacteria, with significant results with Listeria monocytogenes and Klebsiella pneumoniae (Fig. 3B). A 10-fold dilution (10–10,000) series of S. aureus was incubated with LysE35A-coated beads to determine if coated beads were able to bind to low bacteria concentration in blood. The results showed the ability of LysE35A to bind to all tested concentrations, with best results with the highest concentration (Fig. 4). There was no statistical difference between the tested concentrations.

3.1. Molecular cloning and LysE35A expression

3.3. Scanning electron microscopy

A synthetic gene coding for lysozyme with mutation of glutamate 35 was used, where glutamic acid was replaced by alanine (LysE35A). DNA sequencing confirmed that the LysE35A gene was inserted into the yeast expression vector pPicZαA in the correct orientation (Fig. S1). The vector with the desired sequence was successfully introduced into

To investigate the interaction between LysE35A-coated beads and bacteria, the capture of S. aureus (106 CFU) was performed in PBS buffer and observed by scanning electron microscopy. Attachment of S. aureus to LysE35A-coated beads was further confirmed (Fig. 5). Some beads were found connected to each other in clusters that could facilitate

2.6. Scanning electron microscopy S. aureus samples were serially diluted to 106 CFU/ml in PBS and added to 50 μl of coated beads. The samples were incubated for 90 min at room temperature under stirring and then placed in a magnetic separator rack and resuspended in 100 μl 2.5% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.2). The samples were adhered to glass coverslips coated with 0.1% poly-L-lysine, dehydrated in graded acetone series, critical point dried and then coated with a 20 nm thick gold layer in a sputtering device. The samples were then examined in a JEOL Scanning Electron Microscope (SEM), model JSM 6010PLUS-LA. 2.7. Statistical analysis Data represent the mean and standard deviation calculated from experiments. Statistic comparison analyses were made using t-test, considering P b 0.05 as statistically significant.

Fig. 1. Silver-stained electrophoresis gel of LysE35A expression during 120 h. M: Benchmark marker; C1: pPicZα vector at 0 and 120 h; C2: GS115/albumin at 0 and 120 h; H: commercial hen egg white lysozyme; (B) Western blot analysis of purified protein with HEWL polyclonal antibody. Proteins were eluted with a linear concentration gradient of 0–1 M NaCl in Tris-HCl buffer and 1 ml fractions were collected from each tube (1–9).


99

Fig. 4. 10-fold dilution series of S. aureus captured from blood by LysE35A. Bacteria were incubated in whole blood diluted in TTE buffer with LysE35A-coated magnetic beads or magnetic beads without LysE35A (control), and the capture efficiency was determined by colony counts. Data are the mean ± SD of assays performed in triplicate. *Statistically significant difference (P b 0.05) as compared with the control.

Fig. 2. Capure efficiency of LysE35A-coated or uncoated (control) magnetic beads in 1 ml PBS containing 103 CFU of Gram-positive bacteria. Capture efficiency was assessed by counting the number of colonies. Data are the mean ± SD of assays performed in triplicate. *Statistically significant difference (P b 0.05), compared with the control.

bacterial connections, since occasionally one bacterium appeared linked to more than one bead. 3.4. Cell capture qPCR detection PCR-based methods have sufficient sensitivity to detect pathogens in blood. However, blood contains large amounts of PCR inhibitors

that can interfere with the reaction and greatly reduce the sensitivity of the assay (Yamamoto, 2002). To deal with this problem, PCR inhibitors must be diluted, inactivated or removed from the sample (Yamamoto, 2002). In this context, we tested two different methods for red blood cell lysis before capture: ammonium chloride (NH4Cl) or Triton-Tris-EDTA (TTE). After capture of bacteria with LysE35A in a magnetic system, the DNA was extracted and human DNA was detected by Taqman PCR with RNAse P as probe. Positive human DNA amplification in the samples bound to lysozyme-coated beads and in whole blood demonstrated that the two lysis methods had the same performance and the extraction process was not sufficient to degrade the human DNA in the samples. Capture of S. aureus was carried out by lysozyme-coated beads at different concentrations (100–106 CFU), followed by 16S gene amplification by qPCR. Captured samples (CT 29,7–29,1) showed more sensitive amplification of 16S gene than whole blood (CT 33,7–29,7) at concentrations of 103–105 CFU, respectively (Fig. 6). Concentrations lower than 100 CFU were below the qPCR detection limit. This result shows that LysE35A-beads capture in combination with real-time PCR to detect S. aureus in blood samples may potentially be used for rapid pathogens detection in blood infection.

4. Discussion

Fig. 3. Capure efficiency of LysE35A-coated or uncoated (control) magnetic beads in blood with different bacteria species (103 CFU). Bacteria were incubated in whole blood diluted in TTE buffer with LysE35A-coated magnetic beads or magnetic beads without LysE35A (control), and the capture efficiency was determined by colony counts. Data are the mean ± SD of assays performed in triplicate. A: MSSA and MRSA clinical isolates. B: Gram-positive and negative bacteria. *Statistically significant difference (P b 0.05) as compared with the control. #Statistically significant difference (P b 0.05) between captured samples.

The aim of this study was the specific capture, enrichment and separation of bacteria from blood with low contamination levels. We used a mutated lysozyme coupled to magnetic beads (LysE35A-bead) for the specific capture and separation of bacteria. Magnetic separation is efficient in removing inhibitory substances from blood and isolates the target cells directly from samples without the need of standard concentration steps (Olsvik et al., 1994). In addition, this method is non-destructive for the target cells, which allow subsequent detection of the bacteria. The system LysE35A-bead was incubated with diverse bacteria in PBS and blood (Staphylococcus aureus, S. haemolyticus, S. epidermidis, Listeria monocytogenes, Enterococcus faecium, Enterococcus faecalis, Streptococcus agalactiae, Escherichia coli and Klebsiella pneumoniae), allowing specific capture and separation of the microorganisms. Scanning electron microscopy showed the interaction between LysE35Acoated beads and S. aureus (ATCC strain). In PBS, LysE35A-beads were able to bind a wide range of bacteria, like S. aureus (94%) and E. faecium (81%). Immunomagnetic capture based on the presence of protein A on S. aureus bacterial cells obtained similar results, recovering 85% from a suspension of 103 CFU/ml (François et al., 2003). LysE35A-beads bound to 66% of L. monocytogenes, which was much higher than that found by Mendonça et al. (2012) using monoclonal antibodies against L. monocytogenes (CDC strain F4244), where magnetic capture recovered b15% from 103 CFU/ml. These data indicate that LysE35A-beads have high capacity to bind Gram-positive bacteria.

100


Fig. 5. Interaction between LysE35A-coated beads and S. aureus as shown by scanning electron microscopy. Bacteria (106 CFU) were incubated in 1 ml of PBS buffer with LysE35A-coated magnetic beads (108 particles). A = negative control. B and C = captured sample. Scale bar = 1 μm.

Whole blood is one of the most complex matrices due to its various components and high viscosity (Kwaan, 2010). Therefore, we tested a range of conditions to maximize binding of LysE35A-beads to bacteria. This system was able to bind some lower (10, 102, 103, 104 CFU/ml) concentrations of clinical isolates of S. aureus, including antibiotic-resistant MRSA in blood. This result is relevant because patients with clinically significant bacteremia are characterized by low numbers of circulating bacteria, typically in the range of 1–100 CFU/ml for adults (Yagupsky and Nolte, 1990) and b10 CFU/ml in neonates (Kellogg et al., 2000). Bacteremia caused by S. aureus is a serious infection associated with high morbidity, mortality and with complications such as infective endocarditis, vertebral osteomyelitis, and recurrent infection (Corey, 2009). S. aureus is pathogen with a proven ability to develop resistance and has many virulent strains (Levy and Marshall, 2004). The MRSA is one of the most prevalent antibiotic-resistant pathogens found in hospitals (Pray, 2008) and this has become an emerging problem in the community (Podschun and Ullmann, 1998; Gelatti et al., 2009). Moreover, sepsis caused by antibiotic-resistance bacteria is one of the most critical public health issues we face today (Levy and Marshall, 2004). As LysE35A-beads bind to clinical isolates of MRSA, this could offer a strategy for rapid diagnostic and initiation of appropriate treatment. We tested the ability of LysE35A-beads to bind different clinical species of Gram positive and Gram negative bacteria in blood, which are implicated in sepsis. Enterococcus faecalis, Listeria monocytogenes and Klebsiella pneumoniae showed similar efficiency to S. aureus. Klebsiella pneumoniae causes urinary tract infections, pneumonia and is the second Gram-negative bacterium implicated in nosocomial bloodstream infections (Linden, 2003). Moreover, virulence factors have provided new insights into the pathogenic strategies of these bacteria (Linden, 2003). Enterococci are the third most common nosocomial bloodstream pathogen and frequently are the causative pathogen(s) of intra-abdominal, genitourinary, surgical wound, endovascular, or other serious infections (Farber and Peterkin, 1991). Listeria monocytogenes has been implicated in several outbreaks of foodborne disease, which is a great concern to public health due to its clinical severity and high fatality rate, despite its low incidence (Leggieri et al., 2010). Blood is arguably one of the most complex fluids in the world, so the fact that LysE35A-

coated beads can capture bacteria in blood suggests that they may function in a wide range of other complex samples like food and environment. Several blood components (including heme, hematin, hemoglobin, lactoferrin, and IgG) can interfere with PCR and greatly reduce the sensitivity of the assay (Fredricks and Relman, 1998). This does not permit rapid detection of microorganisms, since culture enrichment and cleaning steps are still required. A robust sample preparation method to remove the majority of inhibitors found in blood is of great importance to make PCR a viable diagnostic technology, but no robust system has yet emerged. A sensitive and specific way to detect bacteria directly from blood is based on amplifying the 16S ribosomal RNA genes by PCR (Valiathan and Asthana, 2014). 16S rRNA genes contain high conserved regions and are universally present, which permit the identification of all bacteria (Regan et al., 2012). Based on this principle, we used this sequence to identify S. aureus captured from blood with the system LysE35A-coated beads. At concentrations of 103–105 CFU, captured samples have better results than whole blood. The CT values for 103 CFU/ml were 29.1 ± 0.1 for captured sample without incubation time and 33.7 ± 0.3 for blood sample. Similar result was obtained after 5 h of incubation of S. aureus in medium with blood followed by Taqman assay (Regan et al., 2012). The detection of MRSA was drastically affected without immunocapture on mixed culture with methicillinresistant Staphylococcus epidermidis (François et al., 2003). Ours results showed that capture with LysE35A-coated beads in combination with real-time PCR to detect S. aureus in blood samples is particularly attractive because of the potential to concentrate microorganisms. Magnetic separation in conjunction with PCR can be a sensitive method to detect low levels of DNA from pathogens in clinical samples and could be a helpful tool in the diagnosis of bloodstream infection. Recently, microfluidic and miniaturized lab-on-a-chip separation devices have been introduced to separate pathogenic bacteria from blood based on magnetic separation (Liu et al., 2004; Xia et al., 2006; Yung et al., 2009; J. H. Kang et al., 2014). The LysE35A-coated beads can be useful to develop a point of care diagnostic system using microfluidic lab-on-a-chip technologies for rapid and sensitive detection of pathogens in clinical settings. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.mimet.2016.07.012. Acknowledgements This work was supported by CNPq, FIOCRUZ and IBMP. The authors thank Luis G. Morello for the supply of 16S primers. References

Fig. 6. Real-time PCR amplification of the bacterial 16S gene. Comparison of DNA isolated from whole blood spiked with 10-fold dilution series of S. aureus and captured with LysE35A (■) or not (●). DNA was used in SYBR Green assays for the detection of S. aureus. Data are the mean ± SD of assays performed in triplicate.

Angus, D.C., van der Poll, T., 2013. Severe sepsis and septic shock. N. Engl. J. Med. 369, 840–851. Benkerroum, N., 2008. Antimicrobial activity of lysozyme with special relevance to milk. Afr. J. Biotechnol. 7, 4856–4867. Carrigan, S.D., Scott, G., Tabrizian, M., 2004. Toward resolving the challenges of sepsis diagnosis. Clin. Chem. 50, 1301–1314. Coklin, T., Farber, J.M., Parrington, L.J., Bin Kingombe, C.I., Ross, W.H., Dixon, B.R., 2011. Immunomagnetic separation significantly improves the sensitivity of polymerase chain reaction in detecting Giardia duodenalis and Cryptosporidium spp. in dairy cattle. J. Vet. Diagn. Investig. 23, 260–267.

A.L.K. Lopes et al. / Journal of Microbiological Methods 128 (2016) 96–101 Corey, G.R., 2009. Staphylococcus aureus bloodstream infections: definitions and treatment. Clin. Infect. Dis. 48, S254–S259. del Rio, A., Cervera, C., Moreno, A., Moreillon, P., Miro, J.M., 2009. Patients at risk of complications of Staphylococcus aureus bloodstream infection. Clin. Infect. Dis. 48, S246–S253. Diler, E., Obst, U., Schmitz, K., Schwartz, T., 2011. A lysozyme and magnetic bead based method of separating intact bactéria. Anal. Bioanal. Chem. 401, 253–265. Emery, S.L., Erdman, D.D., Bowen, M.D., Newton, B.R., Winchell, J.M., Meyer, R.F., Tong, S., Cook, B.T., Holloway, B.P., McCaustland, K.A., Rota, P.A., Bankamp, B., Lowe, L.E., Ksiazek, T.G., Bellini, W.J., Anderson, L.J., 2004. Real-time reverse transcription-polymerase chain reaction assay for SARS-associated coronavirus. Emerg. Infect. Dis. 10, 311–316. Farber, J.M., Peterkin, P.I., 1991. Listeria monocytogenes, a food-borne pathogen. Microbiol. Rev. 55, 476–511. François, P., Pittet, D., Bento, M., Pepey, B., Vaudaux, P., Lew, D., Schrenzel, J., 2003. Rapid detection of methicillin-resistant Staphylococcus aureus directly from sterile or nonsterile clinical samples by a new molecular assay. J. Clin. Microbiol. 41, 254–260. Fredricks, D.N., Relman, D.A., 1998. Improved amplification of microbial DNA from blood cultures by removal of the PCR inhibitor sodium polyanetholesulfonate. J. Clin. Microbiol. 36, 2810–2816. Gelatti, L.C., Bonamigo, R.R., Becker, A.P., D'azevedo, P.A., 2009. Staphylococcus aureus resistentes à meticilina: disseminação emergente na comunidade. An. Bras. Dermatol. 84, 501–506. Isaksson, M., Hagström, Å., Armua-Fernandez, M.T., Wahlström, H., Ågren, E.O., Miller, A., Holmberg, A., Lukacs, M., Casulli, A., Deplazes, P., Juremalm, M., 2014. A semi-automated magnetic capture probe based DNA extraction and real-time PCR method applied in the Swedish surveillance of Echinococcus multilocularis in red fox (Vulpes vulpes) faecal samples. Parasit. Vectors 7, 583. Kang, D.K., Ali, M.M., Zhang, K., Huang, S.S., Peterson, E., Digman, M.A., Gratton, E., Zhao, W., 2014a. Rapid detection of single bacteria in unprocessed blood using integrated comprehensive droplet digital detection. Nat. Commun. 5, 5427. Kang, J.H., Super, M., Yung, C.W., Cooper, R.M., Domansky, K., Graveline, A.R., Mammoto, T., Berthet, J.B., Tobin, H., Cartwright, M.J., Watters, A.L., Rottman, M., Waterhouse, A., Mammoto, A., Gamini, N., Rodas, M.J., Kole, A., Jiang, A., Valentin, T.M., Diaz, A., Takahashi, K., Ingber, D.E., 2014b. An extracorporeal blood-cleansing device for sepsis therapy. Nat. Med. 20, 1211–1216. Kassimi, L.B., Gonzague, M., Boutrouille, A., Cruciere, C., 2002. Detection of encephalomyocarditis virus in clinical samples by immunomagnetic separation and one-step RTPCR. J. Virol. Methods 101, 197–206. Kellogg, J.A., Manzella, J.P., Bankert, D.A., 2000. Frequency of low-level bacteriemia in children from birth to fifteen years of age. J. Clin. Microbiol. 38, 2181–2185. Kwaan, H.C., 2010. Role of plasma proteins in whole blood viscosity: a brief clinical review. Clin. Hemorheol. Microcirc. 44, 167–176. Leggieri, N., Rida, A., François, P., Schrenzel, J., 2010. Molecular diagnosis of bloodstream infections: planning to (physically) reach the bedside. Curr. Opin. Infect. Dis. 23, 311–319. Levy, S.B., Marshall, B., 2004. Antibiotic resistance worldwide: cause, challenges and response. Nat. Med. 10, S122–S129. Liesenfeld, O., Lehman, L., Hunfeld, K.P., Kost, G., 2014. Molecular diagnosis of sepsis: new aspects and recent developments. Eur. J. Microbiol. Immunol. 4, 1–25. Linden, P., 2003. Can enterococcal infections initiate sepsis syndrome? Curr. Infect. Dis. Rep. 5, 372–378. Liu, R.H., Yang, J.N., Lenigk, R., Bonanno, J., Grodzinski, P., 2004. Self-contained, fully integrated biochip for sample preparation, polymerase chain reaction amplification, and DNA microarray detection. Anal. Chem. 76, 1824–1831. Maeda, K., Ito, K., Yamaguchi, N., 1980. A simple lysoplate method of lysozyme determination with samples dried on filter paper. Clin. Chim. Acta 100, 175–181. Malcolm, B.A., Rosenberg, S., Corey, M.J., Allen, J.S., de Baetselier, A., Kirsch, J.F., 1989. Sitedirected mutagenesis of the catalytic residues Asp-52 and Glu-35 of chicken egg white lysozyme. Proc. Natl. Acad. Sci. U. S. A. 86, 133–137.

101

Mendonça, M., Conrad, N.L., Conceição, F.R., Moreira, A.N., da Silva, W.P., Aleixo, J.A.G., Bhunia, A.K., 2012. Highly specific fiber optic immunosensor coupled with immunomagnetic separation for detection of low levels of Listeria monocytogenes and L. ivanovii. BMC Microbiol. 12, 275. Mount, R.C., Jordan, B.E., Hadfield, C., 1996. Transformation of lithium-treated yeast cells and the selection of auxotrophic and dominant markers. In: JM, W., Evans, I.H. (Eds.), Methods in Molecular BiologyYeast Protocols 53. Humana Press, Totowa, NJ, pp. 139–145. Nadkarni, M.A., Martin, F.E., Jacques, N.A., Hunter, N., 2002. Determination of bacterial load by real-time PCR using a broad-range (universal) probe and primers set. Microbiology 148, 257–266. Nam, J., Huang, H., Lim, H., Lim, C., Shin, S., 2013. Magnetic separation of malaria-infected red blood cells in various developmental stages. Anal. Chem. 85, 7316–7323. Negoro, E., Iwasaki, H., Tai, K., Ikegaya, S., Takagi, K., Kishi, S., Yamauchi, T., Yoshida, A., Urasaki, Y., Shimadzu, M., Ueda, T., 2013. Utility of PCR amplification and DNA microarray hybridization of 16S rDNA for rapid diagnosis of bacteremia associated with hematological diseases. Int. J. Infect. Dis. 17, 271–276. Olsvik, O., Popovic, T., Skjerve, E., Cudjoe, K.S., Hornes, E., Ugelstad, J., Uhlén, M., 1994. Magnetic separation techniques in diagnostic microbiology. Clin. Microbiol. Rev. 7, 43–54. Opsteegh, M., Langelaar, M., Sprong, H., den Hartog, L., De Craeye, S., Bokken, G., Ajzenberg, D., Kijlstra, A., van der Giessen, J., 2010. Direct detection and genotyping of Toxoplasma gondii in meat samples using magnetic capture and PCR. Int. J. Food Microbiol. 139, 193–201. Podschun, R., Ullmann, U., 1998. Klebsiella spp. as nosocomial pathogens: epidemiology, taxonomy, typing methods, and pathogenicity factors. Clin. Microbiol. Rev. 11, 589–603. Pray, L., 2008. Antibiotic resistance, mutation rates and MRSA. Nat. Educ. 1, 30. Regan, J.F., Furtado, M.R., Brevnov, M.G., Jordan, J.A., 2012. A sample extraction method for faster, more sensitive PCR-based detection of pathogens in blood culture. J. Mol. Diagn. 14, 120–129. Šafařı́k, I., Šafařı́ková, M., 1999. Use of magnetic techniques for the isolation of cells. J. Chromatogr. B Biomed. Sci. Appl. 722, 33–53. Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning: A Laboratory Manual. second ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Sierra, C., Martínez-Blanco, D., Blanco Jesús, A., Gallego, J.R., 2014. Optimisation of magnetic separation: a case study for soil washing at a heavy metals polluted site. Chemosphere 107, 290–296. Tomoyasu, T., 1992. Development of the immunomagnetic enrichment method selective for Vibrio parahaemolyticus serotype K and its application to food poisoning study. Appl. Environ. Microbiol. 58, 2679–2682. Valiathan, R., Asthana, D., 2014. Interpreting and reporting molecular diagnostics test. In: Kessler, H.H. (Ed.), Molecular Diagnostics of Infectious Diseases, third ed. Walter de Gruyter, Berlin, pp. 85–96. Xia, N., Hunt, T.P., Mayers, B.T., Alsberg, E., Whitesides, G.M., Westervelt, R.M., Ingber, D.E., 2006. Combined microfluidic-micromagnetic separation of living cells in continuous flow. Biomed. Microdevices 8, 299–308. Yagupsky, P., Nolte, F., 1990. Quantitative aspects of septicemia. Clin. Microbiol. Rev. 3, 269–279. Yakub, G.P., Stadterman-Knauer, K.L., 2004. Immunomagnetic separation of pathogenic organisms from environmental matrices. Methods Mol. Biol. 268, 189–197. Yamamoto, Y., 2002. PCR in diagnosis of infection: detection of bacteria in cerebrospinal fluids. Clin. Diagn. Lab. Immunol. 9, 508–514. Yung, C.W., Fiering, J., Mueller, A.J., Ingber, D.E., 2009. Micromagnetic–microfluidic blood cleansing device. Lab Chip 9, 1171–1177.