Article pubs.acs.org/ac
Trace Detection of Specific Viable Bacteria Using TetracysteineTagged Bacteriophages Lina Wu, Tian Luan, Xiaoting Yang, Shuo Wang, Yan Zheng, Tianxun Huang, Shaobin Zhu, and Xiaomei Yan* The MOE Key Laboratory of Spectrochemical Analysis & Instrumentation, The Key Laboratory for Chemical Biology of Fujian Province, Department of Chemical Biology, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen, Fujian 361005, China ABSTRACT: Advanced methods are urgently needed to determine the identity and viability of trace amounts of pathogenic bacteria in a short time. Existing approaches either fall short in the accurate assessment of microbial viability or lack specificity in bacterial identification. Bacteriophages (or phages for short) are viruses that exclusively infect bacterial host cells with high specificity. As phages infect and replicate only in living bacterial hosts, here we exploit the strategy of using tetracysteine (TC)-tagged phage in combination with biarsenical dye to the discriminative detection of viable target bacteria from dead target cells and other viable but nontarget bacterial cells. Using recombinant M13KE-TC phage and Escherichia coli ER2738 as a model system, distinct differentiation between individual viable target cells from dead target cells was demonstrated by flow cytometry and fluorescence microscopy. As few as 1% viable E. coli ER2738 can be accurately quantified in a mix with dead E. coli ER2738 by flow cytometry. With fluorescence microscopic measurement, specific detection of as rare as 1 cfu/mL original viable target bacteria was achieved in the presence of a large excess of dead target cells and other viable but nontarget bacterial cells in 40 mL artificially contaminated drinking water sample in less than 3 h. This TC-phage-FlAsH approach is sensitive, specific, rapid, and simple, and thus shows great potential in water safety monitoring, health surveillance, and clinical diagnosis of which trace detection and identification of viable bacterial pathogens is highly demanded.
S
believed that EMA can only penetrate cells with compromised membranes, and its covalent binding to DNA through photoactivation inhibits subsequent PCR amplification. However, the general application of EMA is hampered by the fact that EMA at high concnetrations may penetrate into viable cells, resulting in underestimates of viable cell numbers.11,12 Moreover, the reliance on cell membrane integrity renders the EMA-qPCR approach not suitable for certain disinfection methods that do not cause sufficient membrane damage, such as UV or chloramine treatment.12 LIVE/DEAD BacLight Bacterial Viability Kits (Molecular Probes, Invitrogen) have been extensively used for bacterial viability assessment.13−15 They employ two nucleic acid stains, green-fluorescent SYTO 9 and red-fluorescent propedium iodide (PI) to label all the bacteria and dead bacteria with damaged membranes, respectively. However, a recent report shows that up to 40% of the gram negative Sphingomonas sp. LB126 and the gram positive Mycobacterium frederiksbergense LB501T can be stained by PI during early exponential growth as compared to the 2−
pecific detection of infectious agents is important for healthcare, food safety, water quality control, and antibioterrorism.1,2 In environmental and clinical settings, viable pathogens often coexist with dead pathogenic bacteria and other viable but harmless microbes.3 Therefore, it is a prerequisite to differentiate viable pathogens from nonhazardous bacteria for an accurate virulence assessment. On the other hand, an infectious dose as low as 10 viable pathogens can lead to disease or even death, such as Shigella, Salmonella, E. coli O157:H7, and E. coli O104:H4.4−6 Therefore, it is vital to develop specific and sensitive methods that are fast enough to detect viable bacterial pathogens before they multiply and become a severe heath risk. Current methods for the specific detection of pathogenic bacteria are often based on polymerase chain reaction (PCR) amplification of specific DNA sequences or on antibody− antigen interaction. However, both the DNA-based molecular techniques and immunoassays cannot differentiate whether positive signals originate from live or dead bacterial targets.7,8 While plate counting is the gold standard for bacterial viability testing, it is tedious and time-consuming. Recently, ethidium bromide monoazide (EMA) in combination with quantitative PCR (EMA-qPCR) has been described as a promising method to distinguish viable taget bacteria from dead cells.9,10 It is © XXXX American Chemical Society
Received: November 5, 2013 Accepted: December 4, 2013
A
dx.doi.org/10.1021/ac403572z | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
10 g NaCl per liter) at 37 °C in baffled flasks with rotary aeration until the exponential growth phase was reached. Heatkilled bacteria were used as the model of dead bacteria. When E. coli ER2738 cells were heated at 80 °C for 30 min, the cells can be completely killed as confirmed by plate counting. Throughout this work, colony forming units (cfu/mL) were used to quantify viable bacteria, and the unit of cells/mL was applied to quantify dead bacteria and the mixture of live and dead bacteria. Preparation of M13KE-TC Reagent Phages. A small amount of M13KE-TC phage stock was added to 20 mL of log phase E. coli ER2738 and incubated with vigorous shaking at 37 °C for 4.5 h. The culture was centrifuged twice (10000 rpm, 10 min, 4 °C) to remove cell debris. The supernatant was purified by 1/4 volume of PEG/NaCl [20% (w/v) polyethylene glycol8000, 2.5 M NaCl] precipitation as described as follows. The upper 80% of the supernatant was transferred to a fresh tube and to which 1/4 volume of PEG/NaCl was added. The phage was precipitated at 4 °C overnight and centrifuged (18000g, 15 min, 4 °C). The pellet was resuspended in 1 mL of TBS (50 mM Tris-HCl, 150 mM NaCl, pH 7.5). The suspension was transferred to a 1.5 mL microcentrifuge tube and centrifuged (10000 rpm, 5 min, 4 °C) to pellet residual cells. The suspension was transferred to a 1.5 mL tube and stored at 4 °C. Phage Infection and FlAsH Labeling. One milliliter of bacterial culture in the exponential growth phase was inoculated with M13KE-TC phage particles at a multiplicity of infection of 100. After incubation at 37 °C for 1 h with vigorous shaking (250 rpm), 100 μL of the mixture was centrifuged and washed with HBSS. The bacterial cells were resuspended in FlAsH-EDT2 (20 μL, 5 μM) while being kept in dark for 60 min at 37 °C. After incubation, the cells were washed twice with HBSS and then resuspended in 100 or 20 μL HBSS for flow cytometry or fluorescence microscopy analysis, respectively. Prior to flow cytometric analysis, the sample was further diluted 200-fold with HBSS. Flow Cytometry and Fluorescence Microscopy Analysis. FlAsH-labeled samples were analyzed with a Becton Dickinson FACSVerse flow cytometer using a 488 nm excitation laser and the FL1 (527/32 nm band-pass filter) detector. The sample flow rate was kept constant throughout the analysis. All the bacteria samples were detected in the log scale, and events were triggered on side scattering. A total of 5000 events falling in the gated bacterial region were collected for each sample. Data acquisition and analysis were carried out by using BD FACSuite software. The data were analyzed by FlowJo 7.6.1 software (Tree Star, Inc., Ashland, OR). Fluorescence images were obtained with an Axio Imager A1 fluorescence microscope (Carl Zeiss) using an Alpha Apochromatic Plan oil-immersion objective lens (100×). Bacterial Enumeration with External Standard Method. In order to determine the reproduction rate of viable E. coli ER2738 upon TC-Phage infection, bacterial enumeration was carried out as described below. One milliliter of E. coli ER2738 in the exponential growth phase (optical density of 0.60 at 600 nm) or its serial dilutions were inoculated with 100 μL M13KETC phage particles (5 × 1011 pfu/mL). After 60 min of incubation at 37 °C with vigorous shaking (250 rpm), all the samples were subjected to heat treatment to kill the bacteria. Then the samples were centrifuged at 10000 rpm for 5 min. Supernatant was carefully removed and the cell pellet was resuspended in 1 mL normal saline. These samples were then diluted to an appropriate bacteria concentration of ∼1 × 105
5% of cells in the early stationary phase of growth.16 These results pose a query regarding the reliability of using PI-based assessment of membrance integrity as a universal indicator of bacterial viability. Additionally, SYTO 9/PI staining cannot determine bacterial identity. Although some new approches have been developed recently for bacterial viability assessment,17,18 they entail the usage of special instruments or complex procedures and are of limited specificity in bacterial detection. Bacteriophages (or phages for short) are viruses that infect a broad or narrow range of host bacteria with unique specificity. Naturally occurring phages are ubiquitous and thus provide a resourceful pool of specific detection reagents for the surveillance of pathogenic bacteria in food, water, and clinical samples.19−24 Particularly, protein or peptide tag can be displayed on the surface of phages by splicing its gene to a coat protein gene. If bacteria susceptible to the particular recombinant phage are present in a sample, then a productive infection will take place with every newly formed progeny phages displaying the tag on its surface. Many reporter genes have been constructed into phages, such as β-galactosidase (lacZ), bacterial (lux) and firefly (luc) luciferase, and GFP and its variants, to facilitate colorimetric, bioluminescent, and fluorescent detection of phage-based assays, respectively.20,25−28 Recently, streptavidin-coated quantum dots have been used to tag progeny phages bearing biotinylation peptides on their capsid proteins for highly sensitive bacteria detection.19,29 In 2011, we developed a new strategy for sensitive and selective bacteria detection by utilizing a tetracysteine (TC)-tagged phage in conjunction with membrane permeable fluorescein arsenical helix binder (FlAsH).30 This methodology integrates the natural specificity of phages and rapid phage replication within their bacterial hosts with the highly sensitive fluorescence labeling of biarsenical dye to TCtagged proteins. The small size of the TC-peptide tag (approximately 36 base pairs) renders an easy incorporation of the reporter gene into the phage genome and minimal interference with phage infection and assembly. Because phages propagate only in living host cells, in the present study the TCphage-FlAsH strategy is exploited to the discriminative detection of viable target bacteria from dead target cells and other viable but nontarget bacterial strains. E. coli ER2738 and its specific phage M13KE are chosen as a model system.
■
EXPERIMENTAL SECTION Reagents and Chemicals. The TC-FlAsH In-Cell Tetracysteine Tag Detection Kit was purchased from Molecular Probes of Invitrogen, Inc. (Eugene, OR). The recombinant M13KE-TC phage was constructed in our laboratory.30 QuantiT PicoGreen dsDNA reagent and yellow-green FluoSpheres of 0.5 μm were obtained from Molecular Probes (Eugene, OR). The manufacture-reported particle concentration of 0.5 μm FluoSpheres is 2.9 × 1011 particles/mL. The wash buffer, Hanks balanced saline solution (HBSS), was purchased from Sangon Biotech Company, Ltd. (Shanghai, China). All other chemicals for buffer preparation were obtained from Sigma (St. Louis, MO). Distilled, deionized water supplied by a Milli-Q RG unit (Millipore, Bedford, MA) was filtered through a 0.22 μm filter and used in the preparation of buffer solutions. Bacterial Cell Culturing. E. coli ER2738 and four M13KE insensitive bacterial strains: E. coli ER2566, E. coli BL21, Pseudomonas aeruginosa, and Salmonella sp. were grown in Luria−Bertani (LB) broth (10 g tryptone, 5 g yeast extract, and B
dx.doi.org/10.1021/ac403572z | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
Figure 1. Schematic representation of the scheme for the trace detection of specific viable bacteria using the TC-phage-FlAsH strategy. E. coli ER2738 and its specific phage M13KE are used as a model system.
Figure 2. Flow cytometric differentiation between viable and dead target bacteria with the TC-phage-FlAsH strategy. (a and b) Bivariate dot-plots of bacterial side scattering versus fluorescence intensity for the (a) dead E. coli ER2738 and (b) viable E. coli ER2738 inoculated with TC-tagged recombinant phage M13KE-TC. Events falling inside regions P1 or P2 are considered as dead or viable cells, respectively. (c) Bacterial fluorescence distribution histograms obtained for the dead and viable E. coli ER2738.
pellet was transferred to a 1.5 mL Eppendorf tube followed by three washes with 500 μL sterile PBS each to remove residual bacterial cells attached to the surface of the 50 mL tube. The washing solutions were all collected to the same 1.5 mL microcentrifuge tube. Then the bacteria suspension was centrifuged at 10000 rpm for 5 min and resuspended with 1.0 mL LB broth. That is, the 40 mL spiked purified water was 40-fold concentrated before being assayed by the TC-phageFlAsH approach. Then, every artificially spiked drinking water sample was inoculated with 100 μL M13KE-TC (5 × 1011 pfu/ mL) and incubated at 37 °C for 1 h with vigorous shaking (250 rpm). The sample was centrifuged, washed with HBSS, labeled with FlAsH, and then analyzed by fluorescence microscopy.
cells/mL with 200-fold diluent of PicoGreen dsDNA stock solution. After 15 min of incubation at 80 °C in the dark, the stained samples were analyzed by flow cytometry. Bacterial concentration was determined by the standard curve generated by the 0.5 μm FluoSpheres at the same instrument conditions. The reproduction rate was calculated by comparing the concentrations measured before and after 1 h of cultivation upon phage inoculation. Preparation of Bacteria-Spiked Drinking Water Samples. Bottled pure water was purchased from a local supermarket and filtered through a 0.22 μm membrane. Bacteria spiked drinking water samples were prepared by adding 100 μL of 1.0 × 109, 1.0 × 107, 1.0 × 105, 4.0 × 103, 4.0 × 102, and 0 cfu/mL viable E. coli ER2738 into 40 mL of the filtered pure water contained in a 50 mL Eppendorf tube, respectively. The final concentrations of viable E. coli ER2738 were 2.5 × 106, 2.5 × 104, 2.5 × 102, 10, 1, and 0 cfu/mL, respectively. Then a cocktail containing equal proportions of E. coli ER2566, E. coli BL21, Pseudomonas aeruginosa, Salmonella sp., and heat-killed E. coli 2738 was added to each of the latter four tubes to make the total cell concentration of every tube 2.5 × 106 cells/mL. These samples were then centrifuged at 8000 rpm for 30 min. Supernatant was carefully removed and the cell
■
RESULTS AND DISCUSSION Principle for the Trace Detection of Specific Viable Bacteria. The scheme for the specific detection of viable target bacteria is shown in Figure 1. First, the “reagent” phage (TCtagged recombinant phage) is used to inoculate the bacteria sample. If viable bacteria sensitive to this particular phage are present in the sample then a productive infection will occur, and a large number of progeny phages will be produced with TC peptide displaying on their surface. Whereas for dead target C
dx.doi.org/10.1021/ac403572z | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
cytometry, the fluorescence signal produced by dead target cell cannot be distinguished from the background in the microscopic images. On the contrary, bright fluorescence was observed from viable target cells infected with M13KE-TC phages, and single bacterial cells can be readily identified. Detection of Viable Target Bacteria in a Mix with Dead Target Cells. It is well-known that viable bacteria always coexist with dead ones in environmental or clinical settings, and the percentage of live target bacteria can vary largely.32 We examined whether the coexistence of dead target bacteria would interfere with the detection of viable target cells by flow cytometric analysis. A series of bacterial samples containing different proportions of viable target cells (0% to 100%) were prepared by mixing viable E. coli ER2738 with dead E. coli ER2738 cells. The mixed samples were inoculated with M13KE-TC and stained with FlAsH-EDT2. The results obtained by flow cytometry are shown in Figure 4. Similar to the data of Figure 2, dead target bacteria yielded very weak fluorescence, whereas viable cells (marked by region P2) exhibited strong fluorescence enhancement. With the increase of the percentage of viable target cells from 0% to 100%, a gradual population increase in the P2 region was observed. When the initial percentages of viable cells were 0, 1, 5, 10, 50, and 100% in the mixed samples, the detected percentages of cells falling inside the P2 region were 0.1, 2.1, 12.3, 20.3, 75.8, and 99.3%, respectively. For the TC-Phage-FlAsH strategy, a short period (1 h) of cultivation is carried out to provide an optimal environment for phage infection and amplification. It should be aware that viable bacteria themselves also replicate in the meantime, which results in an increase of cell number and change of viable bacteria concentration. Within 1 h of cultivation upon phage inoculation, the measured cell reproduction rate was 2.61 ± 0.05 fold for viable E. coli ER2738 (concentration range 5.8 × 106/mL ∼ 2.9 × 108/mL). On the basis of this data, the calculated percentages of viable E. coli ER2738 are 0, 2.6, 12.1, 22.5, 72.3, and 100% for the mixtures of which the mixed percentages of viable target bacteria are 0, 1, 5, 10, 50, and 100%, respectively. Plotting the detected percentages of viable E. coli ER2738 cells versus the calculated ratios, a very linear correlation (R2 of 0.9977) was obtained (Figure 4g). These data suggest that TC-phage-FlAsH approach combined with flow cytometric analysis is suitable for the quantitative and discriminative analysis of viable target bacteria in a mix with dead target cells. Trace Detection of Specific Viable Bacteria in High Background of Other Bacteria and Dead Target Cells in Artificially Contaminated Drinking Water Samples. Water is the essential resource for life and also accounts for the main carrier of infection transmission. Bacteria are indigenous in drinking water, and high concentrations of bacterial cells (in the range of 104−105 cells/mL) of diverse populations are commonly found in both bottled mineral water and tap water at the time of consumption.33,34 While most of them are nonhazardous bacteria, their abundant existence could interfere with the detection of trace amounts of pathogenic bacteria. To examine the feasibility of detecting a small number of viable target bacteria among a large excess of viable but nontarget bacteria and dead target cells, different quantities of viable E. coli ER2738 cells (from 0 to 2.5 × 106 cfu/mL) were mixed with a cocktail containing equal proportions of four different M13 phage insensitive bacterial strains and dead E. coli ER2738 cells to make the total cell concentration of every tube 2.5 × 106 cells/mL. Following the TC-phage incubation and
bacteria, phages can only adsorb on the surface but no phage propagation can proceed. Then a cell-membrane-permeable biarsenical dye FlAsH-EDT2 (EDT = 1,2-ethanedithiol) is used to fluorescently label the TC-tagged phages. The significant fluorescence enhancement attained for viable target cell provides an accurate identification of specific viable bacteria. Discrimination between Viable and Dead Target Bacteria. Fluorescently labeled bacteriophages have been mainly used for the discriminative detection of host bacterial strains from other insensitive cells. Because dead host cells also bear receptors for phage binding, the possibility of applying the proposed TC-phage-FlAsH strategy for the differentiation between live and dead target bacteria was initially examined by flow cytometry. Figure 2 (panels a and b) shows the dotplots of side scattering intensity versus fluorescence intensity obtained from dead and viable E. coli ER2738, respectively. For dead target bacteria in which phages cannot replicate, only a very weak fluorescence signal was detected. By contrast, phage propagation within viable E. coli ER2738 produced a bright green fluorescence, which resulted in a remarkable right-shift in fluorescence intensity (Figure 2b). It has been reported that each host cell can produce approximately one thousand M13 phages within an hour of phage infection.31 To facilitate an easy discrimination between live and dead target cells, regions representing dead and viable target bacteria were drawn and designated as P1 and P2, respectively. For the freshly prepared viable bacterial sample, approximately 99.2% of cells fall in the P2 region, whereas for the heat-killed bacterial sample, only 0.3% of the cells did so. The fluorescence distribution histograms of these two samples are plotted in Figure 2c, of which near baseline separation between viable and dead target cells can be observed. The median fluorescence intensities of P1 and P2 regions are 279 and 6929 for the dead and viable target cells, respectively. Due to the effective infection and amplification of the engineered M13KE-TC phages within live host cells, a clear differentiation between viable and dead target bacteria was achieved. The feasibility of using TC-phage-FlAsH strategy to differentiate viable and dead target bacteria was verified by fluorescence microscopy as well. Figure 3 (panels a−b) shows the bright-field and fluorescence micrographs of dead and viable E. coli ER2738 cells infected with M13KE-TC, respectively. In agreement with data obtained by flow
Figure 3. Bright-field (left) and fluorescence (right) microscopy images of (a) dead and (b) viable E. coli ER2738 inoculated with TCtagged recombinant phage M13KE-TC. Scale bar: 5 μm. D
dx.doi.org/10.1021/ac403572z | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
Figure 4. Flow cytometric analysis of viable target bacteria in a mix with dead target cells using TC-phage-FlAsH strategy. (a−f) Percentages of viable E. coli ER2738 cells are 0, 1, 5, 10, 50 and 100%, respectively. (g) Plot of the detected percentages of viable target cells versus the calculated percentages based on cell reproduction rate.
Figure 5. Trace detection of specific viable bacteria in high background of other nontarget bacteria and dead target cells using TC-phage-FlAsH strategy. Concentrations of viable E. coli ER2738 cells are (a) 2.5 × 106, (b) 2.5 × 104, (c) 2.5 × 102, (d) 10, (e) 1, and (f) 0 cfu/mL. Total bacterial concentration is 2.5 × 106 cells/mL for each sample. Scale bar: 5 μm.
nontarget cells by fluorescence microscopic examination. Phages are robust reagents that can easily be grown into a large quantity. With more and more phages being identified and sequenced, the TC-phage-FlAsH methodology holds great potential in the trace detection of specific viable pathogenic bacteria.
FlAsH-EDT2 staining, these samples were examined by fluorescence microscopy. Clearly, only viable E. coli ER2738 cells gave out strong green fluorescence (Figure 5) and the fluorescence signals generated by dead E. coli ER2738 or nonhost cells could not be differentiated from the background. With the gradual decrease in the concentration of viable E. coli ER2738 cells in the mixture, fewer and fewer fluorescent bacteria can be visualized in each microscopic field. Still, viable target bacteria as low as 1 cfu/mL in 40 mL water sample can be recovered and detected by TC-phage-FlAsH strategy above the background of 2.5 × 106 times more dead target cells and viable but nonhost cells (Figure 5e). Noting that although the reproduction rate can vary depending on the bacterial strains and their corresponding phages, replication of viable bacteria within 1 h of cultivation upon phage infection also aids the trace detection of viable target bacteria.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail: [email protected]. Tel: 86-592-2184519. Fax: 86592-2189959. Notes
The authors declare no competing financial interest.
■
■
ACKNOWLEDGMENTS This work was supported by the National Key Basic Research Program of China (Grant 2013CB933703), National Natural Science Foundation of China (Grants 20975087, 90913015, 21027010, 21105082, 21225523, and 91313302), Research Funds for the Doctoral Program of Higher Education of China (Grants 20090121120008 and 20090121110009), and the NFFTBS (Grant J1310024), for which we are most grateful.
CONCLUSION In summary, we have demonstrated that tetracysteine-tagged phage combined with FlAsH fluorescence staining is specific and sensitive enough for the trace detection of viable host bacteria. Particularly, as few as 1 cfu/mL of specific viable bacteria can be detected in 40 mL of a drinking water sample containing a large excess of dead target cells and viable but E
dx.doi.org/10.1021/ac403572z | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
■
Article
REFERENCES
(1) Binder, S.; Levitt, A. M.; Sacks, J. J.; Hughes, J. M. Science 1999, 284, 1311−1313. (2) Batt, C. A. Science 2007, 316, 1579−1580. (3) Luna, G. M.; Manini, E.; Danovaro, R. Appl. Environ. Microbiol. 2002, 68, 3509−3513. (4) Ivnitski, D.; Abdel-Hamid, I.; Atanasov, P.; Wilkins, E.; Stricker, S. Electroanalysis 2000, 12, 317−325. (5) Bilinski, P.; Kapka-Skrzypczak, L.; Posobkiewicz, M.; Bondaryk, M.; Holownia, P.; Wojtyla, A. Ann. Agric. Environ. Med. 2012, 19, 3− 10. (6) Leggett, H. C.; Cornwallis, C. K.; West, S. A. PLoS Pathog. 2012, 8, e1002512. (7) Zhao, X.; Hilliard, L. R.; Mechery, S. J.; Wang, Y.; Bagwe, R. P.; Jin, S.; Tan, W. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 15027−15032. (8) Mull, B.; Hill, V. R. Appl. Environ. Microbiol. 2009, 75, 3593− 3597. (9) Nocker, A.; Camper, A. K. Appl. Environ. Microbiol. 2006, 72, 1997−2004. (10) Lee, J. L.; Levin, R. E. J. Microbiol. Methods 2009, 76, 93−96. (11) Flekna, G.; Stefanic, P.; Wagner, M.; Smulders, F. J.; Mozina, S. S.; Hein, I. Res. Microbiol. 2007, 158, 405−412. (12) Shi, H.; Xu, W.; Luo, Y.; Chen, L.; Liang, Z.; Zhou, X.; Huang, K. J. Appl. Microbiol. 2011, 111, 1194−1204. (13) Suller, M. T.; Lloyd, D. Cytometry, Part A 1999, 35, 235−241. (14) Berney, M.; Hammes, F.; Bosshard, F.; Weilenmann, H. U.; Egli, T. Appl. Environ. Microbiol. 2007, 73, 3283−3290. (15) Krause, M.; Rosch, P.; Radt, B.; Popp, J. Anal. Chem. 2008, 80, 8568−8575. (16) Shi, L.; Gunther, S.; Hubschmann, T.; Wick, L. Y.; Harms, H.; Muller, S. Cytometry, Part A 2007, 71, 592−598. (17) Nicolaou, N.; Xu, Y.; Goodacre, R. Anal. Chem. 2011, 83, 5681− 5687. (18) Xu, S.; Mutharasan, R. Anal. Chem. 2011, 83, 1480−1483. (19) Edgar, R.; McKinstry, M.; Hwang, J.; Oppenheim, A. B.; Fekete, R. A.; Giulian, G.; Merril, C.; Nagashima, K.; Adhya, S. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 4841−4845. (20) Smartt, A. E.; Xu, T.; Jegier, P.; Carswell, J. J.; Blount, S. A.; Sayler, G. S.; Ripp, S. Anal. Bioanal. Chem. 2012, 402, 3127−3146. (21) Liébana, S.; Spricigo, D. A.; Cortés, M. P.; Barbé, J.; Llagostera, M.; Alegret, S.; Pividori, M. I. Anal. Chem. 2013, 85, 3079−3086. (22) Poshtiban, S.; Javed, M. A.; Arutyunov, D.; Singh, A.; Banting, G.; Szymanski, C. M.; Evoy, S. Analyst 2013, 138, 5619−5626. (23) Shabani, A.; Marquette, C. A.; Mandeville, R.; Lawrence, M. F. Analyst 2013, 138, 1434−1440. (24) Tlili, C.; Sokullu, E.; Safavieh, M.; Tolba, M.; Ahmed, M. U.; Zourob, M. Anal. Chem. 2013, 85, 4893−4901. (25) Awais, R.; Fukudomi, H.; Miyanaga, K.; Unno, H.; Tanji, Y. Biotechnol. Prog. 2006, 22, 853−859. (26) Namura, M.; Hijikata, T.; Miyanaga, K.; Tanji, Y. Biotechnol. Prog. 2008, 24, 481−486. (27) Schofield, D. A.; Molineux, I. J.; Westwater, C. J. Clin. Microbiol. 2009, 47, 3887−3894. (28) Derda, R.; Lockett, M. R.; Tang, S. K.; Fuller, R. C.; Maxwell, E. J.; Breiten, B.; Cuddemi, C. A.; Ozdogan, A.; Whitesides, G. M. Anal. Chem. 2013, 85, 7213−7220. (29) Yim, P. B.; Clarke, M. L.; McKinstry, M.; De Paoli Lacerda, S. H.; Pease, L. F., 3rd; Dobrovolskaia, M. A.; Kang, H.; Read, T. D.; Sozhamannan, S.; Hwang, J. Biotechnol. Bioeng. 2009, 104, 1059−1067. (30) Wu, L.; Huang, T.; Yang, L.; Pan, J.; Zhu, S.; Yan, X. Angew. Chem., Int. Ed. 2011, 50, 5873−5877. (31) Derda, R.; Tang, S. K.; Whitesides, G. M. Angew. Chem., Int. Ed. 2010, 49, 5301−5304. (32) Nocker, A.; Richter-Heitmann, T.; Montijn, R.; Schuren, F.; Kort, R. Int. Microbiol. 2010, 13, 59−65. (33) Leclerc, H.; Moreau, A. FEMS Microbiol. Rev. 2002, 26, 207− 222. (34) Hammes, F.; Berney, M.; Wang, Y.; Vital, M.; Koster, O.; Egli, T. Water Res. 2008, 42, 269−277. F
dx.doi.org/10.1021/ac403572z | Anal. Chem. XXXX, XXX, XXX−XXX
Trace Detection of Specific Viable Bacteria Using TetracysteineTagged Bacteriophages Lina Wu, Tian Luan, Xiaoting Yang, Shuo Wang, Yan Zheng, Tianxun Huang, Shaobin Zhu, and Xiaomei Yan* The MOE Key Laboratory of Spectrochemical Analysis & Instrumentation, The Key Laboratory for Chemical Biology of Fujian Province, Department of Chemical Biology, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen, Fujian 361005, China ABSTRACT: Advanced methods are urgently needed to determine the identity and viability of trace amounts of pathogenic bacteria in a short time. Existing approaches either fall short in the accurate assessment of microbial viability or lack specificity in bacterial identification. Bacteriophages (or phages for short) are viruses that exclusively infect bacterial host cells with high specificity. As phages infect and replicate only in living bacterial hosts, here we exploit the strategy of using tetracysteine (TC)-tagged phage in combination with biarsenical dye to the discriminative detection of viable target bacteria from dead target cells and other viable but nontarget bacterial cells. Using recombinant M13KE-TC phage and Escherichia coli ER2738 as a model system, distinct differentiation between individual viable target cells from dead target cells was demonstrated by flow cytometry and fluorescence microscopy. As few as 1% viable E. coli ER2738 can be accurately quantified in a mix with dead E. coli ER2738 by flow cytometry. With fluorescence microscopic measurement, specific detection of as rare as 1 cfu/mL original viable target bacteria was achieved in the presence of a large excess of dead target cells and other viable but nontarget bacterial cells in 40 mL artificially contaminated drinking water sample in less than 3 h. This TC-phage-FlAsH approach is sensitive, specific, rapid, and simple, and thus shows great potential in water safety monitoring, health surveillance, and clinical diagnosis of which trace detection and identification of viable bacterial pathogens is highly demanded.
S
believed that EMA can only penetrate cells with compromised membranes, and its covalent binding to DNA through photoactivation inhibits subsequent PCR amplification. However, the general application of EMA is hampered by the fact that EMA at high concnetrations may penetrate into viable cells, resulting in underestimates of viable cell numbers.11,12 Moreover, the reliance on cell membrane integrity renders the EMA-qPCR approach not suitable for certain disinfection methods that do not cause sufficient membrane damage, such as UV or chloramine treatment.12 LIVE/DEAD BacLight Bacterial Viability Kits (Molecular Probes, Invitrogen) have been extensively used for bacterial viability assessment.13−15 They employ two nucleic acid stains, green-fluorescent SYTO 9 and red-fluorescent propedium iodide (PI) to label all the bacteria and dead bacteria with damaged membranes, respectively. However, a recent report shows that up to 40% of the gram negative Sphingomonas sp. LB126 and the gram positive Mycobacterium frederiksbergense LB501T can be stained by PI during early exponential growth as compared to the 2−
pecific detection of infectious agents is important for healthcare, food safety, water quality control, and antibioterrorism.1,2 In environmental and clinical settings, viable pathogens often coexist with dead pathogenic bacteria and other viable but harmless microbes.3 Therefore, it is a prerequisite to differentiate viable pathogens from nonhazardous bacteria for an accurate virulence assessment. On the other hand, an infectious dose as low as 10 viable pathogens can lead to disease or even death, such as Shigella, Salmonella, E. coli O157:H7, and E. coli O104:H4.4−6 Therefore, it is vital to develop specific and sensitive methods that are fast enough to detect viable bacterial pathogens before they multiply and become a severe heath risk. Current methods for the specific detection of pathogenic bacteria are often based on polymerase chain reaction (PCR) amplification of specific DNA sequences or on antibody− antigen interaction. However, both the DNA-based molecular techniques and immunoassays cannot differentiate whether positive signals originate from live or dead bacterial targets.7,8 While plate counting is the gold standard for bacterial viability testing, it is tedious and time-consuming. Recently, ethidium bromide monoazide (EMA) in combination with quantitative PCR (EMA-qPCR) has been described as a promising method to distinguish viable taget bacteria from dead cells.9,10 It is © XXXX American Chemical Society
Received: November 5, 2013 Accepted: December 4, 2013
A
dx.doi.org/10.1021/ac403572z | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
10 g NaCl per liter) at 37 °C in baffled flasks with rotary aeration until the exponential growth phase was reached. Heatkilled bacteria were used as the model of dead bacteria. When E. coli ER2738 cells were heated at 80 °C for 30 min, the cells can be completely killed as confirmed by plate counting. Throughout this work, colony forming units (cfu/mL) were used to quantify viable bacteria, and the unit of cells/mL was applied to quantify dead bacteria and the mixture of live and dead bacteria. Preparation of M13KE-TC Reagent Phages. A small amount of M13KE-TC phage stock was added to 20 mL of log phase E. coli ER2738 and incubated with vigorous shaking at 37 °C for 4.5 h. The culture was centrifuged twice (10000 rpm, 10 min, 4 °C) to remove cell debris. The supernatant was purified by 1/4 volume of PEG/NaCl [20% (w/v) polyethylene glycol8000, 2.5 M NaCl] precipitation as described as follows. The upper 80% of the supernatant was transferred to a fresh tube and to which 1/4 volume of PEG/NaCl was added. The phage was precipitated at 4 °C overnight and centrifuged (18000g, 15 min, 4 °C). The pellet was resuspended in 1 mL of TBS (50 mM Tris-HCl, 150 mM NaCl, pH 7.5). The suspension was transferred to a 1.5 mL microcentrifuge tube and centrifuged (10000 rpm, 5 min, 4 °C) to pellet residual cells. The suspension was transferred to a 1.5 mL tube and stored at 4 °C. Phage Infection and FlAsH Labeling. One milliliter of bacterial culture in the exponential growth phase was inoculated with M13KE-TC phage particles at a multiplicity of infection of 100. After incubation at 37 °C for 1 h with vigorous shaking (250 rpm), 100 μL of the mixture was centrifuged and washed with HBSS. The bacterial cells were resuspended in FlAsH-EDT2 (20 μL, 5 μM) while being kept in dark for 60 min at 37 °C. After incubation, the cells were washed twice with HBSS and then resuspended in 100 or 20 μL HBSS for flow cytometry or fluorescence microscopy analysis, respectively. Prior to flow cytometric analysis, the sample was further diluted 200-fold with HBSS. Flow Cytometry and Fluorescence Microscopy Analysis. FlAsH-labeled samples were analyzed with a Becton Dickinson FACSVerse flow cytometer using a 488 nm excitation laser and the FL1 (527/32 nm band-pass filter) detector. The sample flow rate was kept constant throughout the analysis. All the bacteria samples were detected in the log scale, and events were triggered on side scattering. A total of 5000 events falling in the gated bacterial region were collected for each sample. Data acquisition and analysis were carried out by using BD FACSuite software. The data were analyzed by FlowJo 7.6.1 software (Tree Star, Inc., Ashland, OR). Fluorescence images were obtained with an Axio Imager A1 fluorescence microscope (Carl Zeiss) using an Alpha Apochromatic Plan oil-immersion objective lens (100×). Bacterial Enumeration with External Standard Method. In order to determine the reproduction rate of viable E. coli ER2738 upon TC-Phage infection, bacterial enumeration was carried out as described below. One milliliter of E. coli ER2738 in the exponential growth phase (optical density of 0.60 at 600 nm) or its serial dilutions were inoculated with 100 μL M13KETC phage particles (5 × 1011 pfu/mL). After 60 min of incubation at 37 °C with vigorous shaking (250 rpm), all the samples were subjected to heat treatment to kill the bacteria. Then the samples were centrifuged at 10000 rpm for 5 min. Supernatant was carefully removed and the cell pellet was resuspended in 1 mL normal saline. These samples were then diluted to an appropriate bacteria concentration of ∼1 × 105
5% of cells in the early stationary phase of growth.16 These results pose a query regarding the reliability of using PI-based assessment of membrance integrity as a universal indicator of bacterial viability. Additionally, SYTO 9/PI staining cannot determine bacterial identity. Although some new approches have been developed recently for bacterial viability assessment,17,18 they entail the usage of special instruments or complex procedures and are of limited specificity in bacterial detection. Bacteriophages (or phages for short) are viruses that infect a broad or narrow range of host bacteria with unique specificity. Naturally occurring phages are ubiquitous and thus provide a resourceful pool of specific detection reagents for the surveillance of pathogenic bacteria in food, water, and clinical samples.19−24 Particularly, protein or peptide tag can be displayed on the surface of phages by splicing its gene to a coat protein gene. If bacteria susceptible to the particular recombinant phage are present in a sample, then a productive infection will take place with every newly formed progeny phages displaying the tag on its surface. Many reporter genes have been constructed into phages, such as β-galactosidase (lacZ), bacterial (lux) and firefly (luc) luciferase, and GFP and its variants, to facilitate colorimetric, bioluminescent, and fluorescent detection of phage-based assays, respectively.20,25−28 Recently, streptavidin-coated quantum dots have been used to tag progeny phages bearing biotinylation peptides on their capsid proteins for highly sensitive bacteria detection.19,29 In 2011, we developed a new strategy for sensitive and selective bacteria detection by utilizing a tetracysteine (TC)-tagged phage in conjunction with membrane permeable fluorescein arsenical helix binder (FlAsH).30 This methodology integrates the natural specificity of phages and rapid phage replication within their bacterial hosts with the highly sensitive fluorescence labeling of biarsenical dye to TCtagged proteins. The small size of the TC-peptide tag (approximately 36 base pairs) renders an easy incorporation of the reporter gene into the phage genome and minimal interference with phage infection and assembly. Because phages propagate only in living host cells, in the present study the TCphage-FlAsH strategy is exploited to the discriminative detection of viable target bacteria from dead target cells and other viable but nontarget bacterial strains. E. coli ER2738 and its specific phage M13KE are chosen as a model system.
■
EXPERIMENTAL SECTION Reagents and Chemicals. The TC-FlAsH In-Cell Tetracysteine Tag Detection Kit was purchased from Molecular Probes of Invitrogen, Inc. (Eugene, OR). The recombinant M13KE-TC phage was constructed in our laboratory.30 QuantiT PicoGreen dsDNA reagent and yellow-green FluoSpheres of 0.5 μm were obtained from Molecular Probes (Eugene, OR). The manufacture-reported particle concentration of 0.5 μm FluoSpheres is 2.9 × 1011 particles/mL. The wash buffer, Hanks balanced saline solution (HBSS), was purchased from Sangon Biotech Company, Ltd. (Shanghai, China). All other chemicals for buffer preparation were obtained from Sigma (St. Louis, MO). Distilled, deionized water supplied by a Milli-Q RG unit (Millipore, Bedford, MA) was filtered through a 0.22 μm filter and used in the preparation of buffer solutions. Bacterial Cell Culturing. E. coli ER2738 and four M13KE insensitive bacterial strains: E. coli ER2566, E. coli BL21, Pseudomonas aeruginosa, and Salmonella sp. were grown in Luria−Bertani (LB) broth (10 g tryptone, 5 g yeast extract, and B
dx.doi.org/10.1021/ac403572z | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
Figure 1. Schematic representation of the scheme for the trace detection of specific viable bacteria using the TC-phage-FlAsH strategy. E. coli ER2738 and its specific phage M13KE are used as a model system.
Figure 2. Flow cytometric differentiation between viable and dead target bacteria with the TC-phage-FlAsH strategy. (a and b) Bivariate dot-plots of bacterial side scattering versus fluorescence intensity for the (a) dead E. coli ER2738 and (b) viable E. coli ER2738 inoculated with TC-tagged recombinant phage M13KE-TC. Events falling inside regions P1 or P2 are considered as dead or viable cells, respectively. (c) Bacterial fluorescence distribution histograms obtained for the dead and viable E. coli ER2738.
pellet was transferred to a 1.5 mL Eppendorf tube followed by three washes with 500 μL sterile PBS each to remove residual bacterial cells attached to the surface of the 50 mL tube. The washing solutions were all collected to the same 1.5 mL microcentrifuge tube. Then the bacteria suspension was centrifuged at 10000 rpm for 5 min and resuspended with 1.0 mL LB broth. That is, the 40 mL spiked purified water was 40-fold concentrated before being assayed by the TC-phageFlAsH approach. Then, every artificially spiked drinking water sample was inoculated with 100 μL M13KE-TC (5 × 1011 pfu/ mL) and incubated at 37 °C for 1 h with vigorous shaking (250 rpm). The sample was centrifuged, washed with HBSS, labeled with FlAsH, and then analyzed by fluorescence microscopy.
cells/mL with 200-fold diluent of PicoGreen dsDNA stock solution. After 15 min of incubation at 80 °C in the dark, the stained samples were analyzed by flow cytometry. Bacterial concentration was determined by the standard curve generated by the 0.5 μm FluoSpheres at the same instrument conditions. The reproduction rate was calculated by comparing the concentrations measured before and after 1 h of cultivation upon phage inoculation. Preparation of Bacteria-Spiked Drinking Water Samples. Bottled pure water was purchased from a local supermarket and filtered through a 0.22 μm membrane. Bacteria spiked drinking water samples were prepared by adding 100 μL of 1.0 × 109, 1.0 × 107, 1.0 × 105, 4.0 × 103, 4.0 × 102, and 0 cfu/mL viable E. coli ER2738 into 40 mL of the filtered pure water contained in a 50 mL Eppendorf tube, respectively. The final concentrations of viable E. coli ER2738 were 2.5 × 106, 2.5 × 104, 2.5 × 102, 10, 1, and 0 cfu/mL, respectively. Then a cocktail containing equal proportions of E. coli ER2566, E. coli BL21, Pseudomonas aeruginosa, Salmonella sp., and heat-killed E. coli 2738 was added to each of the latter four tubes to make the total cell concentration of every tube 2.5 × 106 cells/mL. These samples were then centrifuged at 8000 rpm for 30 min. Supernatant was carefully removed and the cell
■
RESULTS AND DISCUSSION Principle for the Trace Detection of Specific Viable Bacteria. The scheme for the specific detection of viable target bacteria is shown in Figure 1. First, the “reagent” phage (TCtagged recombinant phage) is used to inoculate the bacteria sample. If viable bacteria sensitive to this particular phage are present in the sample then a productive infection will occur, and a large number of progeny phages will be produced with TC peptide displaying on their surface. Whereas for dead target C
dx.doi.org/10.1021/ac403572z | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
cytometry, the fluorescence signal produced by dead target cell cannot be distinguished from the background in the microscopic images. On the contrary, bright fluorescence was observed from viable target cells infected with M13KE-TC phages, and single bacterial cells can be readily identified. Detection of Viable Target Bacteria in a Mix with Dead Target Cells. It is well-known that viable bacteria always coexist with dead ones in environmental or clinical settings, and the percentage of live target bacteria can vary largely.32 We examined whether the coexistence of dead target bacteria would interfere with the detection of viable target cells by flow cytometric analysis. A series of bacterial samples containing different proportions of viable target cells (0% to 100%) were prepared by mixing viable E. coli ER2738 with dead E. coli ER2738 cells. The mixed samples were inoculated with M13KE-TC and stained with FlAsH-EDT2. The results obtained by flow cytometry are shown in Figure 4. Similar to the data of Figure 2, dead target bacteria yielded very weak fluorescence, whereas viable cells (marked by region P2) exhibited strong fluorescence enhancement. With the increase of the percentage of viable target cells from 0% to 100%, a gradual population increase in the P2 region was observed. When the initial percentages of viable cells were 0, 1, 5, 10, 50, and 100% in the mixed samples, the detected percentages of cells falling inside the P2 region were 0.1, 2.1, 12.3, 20.3, 75.8, and 99.3%, respectively. For the TC-Phage-FlAsH strategy, a short period (1 h) of cultivation is carried out to provide an optimal environment for phage infection and amplification. It should be aware that viable bacteria themselves also replicate in the meantime, which results in an increase of cell number and change of viable bacteria concentration. Within 1 h of cultivation upon phage inoculation, the measured cell reproduction rate was 2.61 ± 0.05 fold for viable E. coli ER2738 (concentration range 5.8 × 106/mL ∼ 2.9 × 108/mL). On the basis of this data, the calculated percentages of viable E. coli ER2738 are 0, 2.6, 12.1, 22.5, 72.3, and 100% for the mixtures of which the mixed percentages of viable target bacteria are 0, 1, 5, 10, 50, and 100%, respectively. Plotting the detected percentages of viable E. coli ER2738 cells versus the calculated ratios, a very linear correlation (R2 of 0.9977) was obtained (Figure 4g). These data suggest that TC-phage-FlAsH approach combined with flow cytometric analysis is suitable for the quantitative and discriminative analysis of viable target bacteria in a mix with dead target cells. Trace Detection of Specific Viable Bacteria in High Background of Other Bacteria and Dead Target Cells in Artificially Contaminated Drinking Water Samples. Water is the essential resource for life and also accounts for the main carrier of infection transmission. Bacteria are indigenous in drinking water, and high concentrations of bacterial cells (in the range of 104−105 cells/mL) of diverse populations are commonly found in both bottled mineral water and tap water at the time of consumption.33,34 While most of them are nonhazardous bacteria, their abundant existence could interfere with the detection of trace amounts of pathogenic bacteria. To examine the feasibility of detecting a small number of viable target bacteria among a large excess of viable but nontarget bacteria and dead target cells, different quantities of viable E. coli ER2738 cells (from 0 to 2.5 × 106 cfu/mL) were mixed with a cocktail containing equal proportions of four different M13 phage insensitive bacterial strains and dead E. coli ER2738 cells to make the total cell concentration of every tube 2.5 × 106 cells/mL. Following the TC-phage incubation and
bacteria, phages can only adsorb on the surface but no phage propagation can proceed. Then a cell-membrane-permeable biarsenical dye FlAsH-EDT2 (EDT = 1,2-ethanedithiol) is used to fluorescently label the TC-tagged phages. The significant fluorescence enhancement attained for viable target cell provides an accurate identification of specific viable bacteria. Discrimination between Viable and Dead Target Bacteria. Fluorescently labeled bacteriophages have been mainly used for the discriminative detection of host bacterial strains from other insensitive cells. Because dead host cells also bear receptors for phage binding, the possibility of applying the proposed TC-phage-FlAsH strategy for the differentiation between live and dead target bacteria was initially examined by flow cytometry. Figure 2 (panels a and b) shows the dotplots of side scattering intensity versus fluorescence intensity obtained from dead and viable E. coli ER2738, respectively. For dead target bacteria in which phages cannot replicate, only a very weak fluorescence signal was detected. By contrast, phage propagation within viable E. coli ER2738 produced a bright green fluorescence, which resulted in a remarkable right-shift in fluorescence intensity (Figure 2b). It has been reported that each host cell can produce approximately one thousand M13 phages within an hour of phage infection.31 To facilitate an easy discrimination between live and dead target cells, regions representing dead and viable target bacteria were drawn and designated as P1 and P2, respectively. For the freshly prepared viable bacterial sample, approximately 99.2% of cells fall in the P2 region, whereas for the heat-killed bacterial sample, only 0.3% of the cells did so. The fluorescence distribution histograms of these two samples are plotted in Figure 2c, of which near baseline separation between viable and dead target cells can be observed. The median fluorescence intensities of P1 and P2 regions are 279 and 6929 for the dead and viable target cells, respectively. Due to the effective infection and amplification of the engineered M13KE-TC phages within live host cells, a clear differentiation between viable and dead target bacteria was achieved. The feasibility of using TC-phage-FlAsH strategy to differentiate viable and dead target bacteria was verified by fluorescence microscopy as well. Figure 3 (panels a−b) shows the bright-field and fluorescence micrographs of dead and viable E. coli ER2738 cells infected with M13KE-TC, respectively. In agreement with data obtained by flow
Figure 3. Bright-field (left) and fluorescence (right) microscopy images of (a) dead and (b) viable E. coli ER2738 inoculated with TCtagged recombinant phage M13KE-TC. Scale bar: 5 μm. D
dx.doi.org/10.1021/ac403572z | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
Figure 4. Flow cytometric analysis of viable target bacteria in a mix with dead target cells using TC-phage-FlAsH strategy. (a−f) Percentages of viable E. coli ER2738 cells are 0, 1, 5, 10, 50 and 100%, respectively. (g) Plot of the detected percentages of viable target cells versus the calculated percentages based on cell reproduction rate.
Figure 5. Trace detection of specific viable bacteria in high background of other nontarget bacteria and dead target cells using TC-phage-FlAsH strategy. Concentrations of viable E. coli ER2738 cells are (a) 2.5 × 106, (b) 2.5 × 104, (c) 2.5 × 102, (d) 10, (e) 1, and (f) 0 cfu/mL. Total bacterial concentration is 2.5 × 106 cells/mL for each sample. Scale bar: 5 μm.
nontarget cells by fluorescence microscopic examination. Phages are robust reagents that can easily be grown into a large quantity. With more and more phages being identified and sequenced, the TC-phage-FlAsH methodology holds great potential in the trace detection of specific viable pathogenic bacteria.
FlAsH-EDT2 staining, these samples were examined by fluorescence microscopy. Clearly, only viable E. coli ER2738 cells gave out strong green fluorescence (Figure 5) and the fluorescence signals generated by dead E. coli ER2738 or nonhost cells could not be differentiated from the background. With the gradual decrease in the concentration of viable E. coli ER2738 cells in the mixture, fewer and fewer fluorescent bacteria can be visualized in each microscopic field. Still, viable target bacteria as low as 1 cfu/mL in 40 mL water sample can be recovered and detected by TC-phage-FlAsH strategy above the background of 2.5 × 106 times more dead target cells and viable but nonhost cells (Figure 5e). Noting that although the reproduction rate can vary depending on the bacterial strains and their corresponding phages, replication of viable bacteria within 1 h of cultivation upon phage infection also aids the trace detection of viable target bacteria.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail: [email protected]. Tel: 86-592-2184519. Fax: 86592-2189959. Notes
The authors declare no competing financial interest.
■
■
ACKNOWLEDGMENTS This work was supported by the National Key Basic Research Program of China (Grant 2013CB933703), National Natural Science Foundation of China (Grants 20975087, 90913015, 21027010, 21105082, 21225523, and 91313302), Research Funds for the Doctoral Program of Higher Education of China (Grants 20090121120008 and 20090121110009), and the NFFTBS (Grant J1310024), for which we are most grateful.
CONCLUSION In summary, we have demonstrated that tetracysteine-tagged phage combined with FlAsH fluorescence staining is specific and sensitive enough for the trace detection of viable host bacteria. Particularly, as few as 1 cfu/mL of specific viable bacteria can be detected in 40 mL of a drinking water sample containing a large excess of dead target cells and viable but E
dx.doi.org/10.1021/ac403572z | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
■
Article
REFERENCES
(1) Binder, S.; Levitt, A. M.; Sacks, J. J.; Hughes, J. M. Science 1999, 284, 1311−1313. (2) Batt, C. A. Science 2007, 316, 1579−1580. (3) Luna, G. M.; Manini, E.; Danovaro, R. Appl. Environ. Microbiol. 2002, 68, 3509−3513. (4) Ivnitski, D.; Abdel-Hamid, I.; Atanasov, P.; Wilkins, E.; Stricker, S. Electroanalysis 2000, 12, 317−325. (5) Bilinski, P.; Kapka-Skrzypczak, L.; Posobkiewicz, M.; Bondaryk, M.; Holownia, P.; Wojtyla, A. Ann. Agric. Environ. Med. 2012, 19, 3− 10. (6) Leggett, H. C.; Cornwallis, C. K.; West, S. A. PLoS Pathog. 2012, 8, e1002512. (7) Zhao, X.; Hilliard, L. R.; Mechery, S. J.; Wang, Y.; Bagwe, R. P.; Jin, S.; Tan, W. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 15027−15032. (8) Mull, B.; Hill, V. R. Appl. Environ. Microbiol. 2009, 75, 3593− 3597. (9) Nocker, A.; Camper, A. K. Appl. Environ. Microbiol. 2006, 72, 1997−2004. (10) Lee, J. L.; Levin, R. E. J. Microbiol. Methods 2009, 76, 93−96. (11) Flekna, G.; Stefanic, P.; Wagner, M.; Smulders, F. J.; Mozina, S. S.; Hein, I. Res. Microbiol. 2007, 158, 405−412. (12) Shi, H.; Xu, W.; Luo, Y.; Chen, L.; Liang, Z.; Zhou, X.; Huang, K. J. Appl. Microbiol. 2011, 111, 1194−1204. (13) Suller, M. T.; Lloyd, D. Cytometry, Part A 1999, 35, 235−241. (14) Berney, M.; Hammes, F.; Bosshard, F.; Weilenmann, H. U.; Egli, T. Appl. Environ. Microbiol. 2007, 73, 3283−3290. (15) Krause, M.; Rosch, P.; Radt, B.; Popp, J. Anal. Chem. 2008, 80, 8568−8575. (16) Shi, L.; Gunther, S.; Hubschmann, T.; Wick, L. Y.; Harms, H.; Muller, S. Cytometry, Part A 2007, 71, 592−598. (17) Nicolaou, N.; Xu, Y.; Goodacre, R. Anal. Chem. 2011, 83, 5681− 5687. (18) Xu, S.; Mutharasan, R. Anal. Chem. 2011, 83, 1480−1483. (19) Edgar, R.; McKinstry, M.; Hwang, J.; Oppenheim, A. B.; Fekete, R. A.; Giulian, G.; Merril, C.; Nagashima, K.; Adhya, S. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 4841−4845. (20) Smartt, A. E.; Xu, T.; Jegier, P.; Carswell, J. J.; Blount, S. A.; Sayler, G. S.; Ripp, S. Anal. Bioanal. Chem. 2012, 402, 3127−3146. (21) Liébana, S.; Spricigo, D. A.; Cortés, M. P.; Barbé, J.; Llagostera, M.; Alegret, S.; Pividori, M. I. Anal. Chem. 2013, 85, 3079−3086. (22) Poshtiban, S.; Javed, M. A.; Arutyunov, D.; Singh, A.; Banting, G.; Szymanski, C. M.; Evoy, S. Analyst 2013, 138, 5619−5626. (23) Shabani, A.; Marquette, C. A.; Mandeville, R.; Lawrence, M. F. Analyst 2013, 138, 1434−1440. (24) Tlili, C.; Sokullu, E.; Safavieh, M.; Tolba, M.; Ahmed, M. U.; Zourob, M. Anal. Chem. 2013, 85, 4893−4901. (25) Awais, R.; Fukudomi, H.; Miyanaga, K.; Unno, H.; Tanji, Y. Biotechnol. Prog. 2006, 22, 853−859. (26) Namura, M.; Hijikata, T.; Miyanaga, K.; Tanji, Y. Biotechnol. Prog. 2008, 24, 481−486. (27) Schofield, D. A.; Molineux, I. J.; Westwater, C. J. Clin. Microbiol. 2009, 47, 3887−3894. (28) Derda, R.; Lockett, M. R.; Tang, S. K.; Fuller, R. C.; Maxwell, E. J.; Breiten, B.; Cuddemi, C. A.; Ozdogan, A.; Whitesides, G. M. Anal. Chem. 2013, 85, 7213−7220. (29) Yim, P. B.; Clarke, M. L.; McKinstry, M.; De Paoli Lacerda, S. H.; Pease, L. F., 3rd; Dobrovolskaia, M. A.; Kang, H.; Read, T. D.; Sozhamannan, S.; Hwang, J. Biotechnol. Bioeng. 2009, 104, 1059−1067. (30) Wu, L.; Huang, T.; Yang, L.; Pan, J.; Zhu, S.; Yan, X. Angew. Chem., Int. Ed. 2011, 50, 5873−5877. (31) Derda, R.; Tang, S. K.; Whitesides, G. M. Angew. Chem., Int. Ed. 2010, 49, 5301−5304. (32) Nocker, A.; Richter-Heitmann, T.; Montijn, R.; Schuren, F.; Kort, R. Int. Microbiol. 2010, 13, 59−65. (33) Leclerc, H.; Moreau, A. FEMS Microbiol. Rev. 2002, 26, 207− 222. (34) Hammes, F.; Berney, M.; Wang, Y.; Vital, M.; Koster, O.; Egli, T. Water Res. 2008, 42, 269−277. F
dx.doi.org/10.1021/ac403572z | Anal. Chem. XXXX, XXX, XXX−XXX
