Open Access
Add to BibSonomyAbstract
We developed a rapid bacterial counting method based on the photothermal assembling (PTA). Based on the laser-induced PTA in fluid medium, an initial bacterial concentration was estimated from the number of assembled bacteria. The measuring time of our method is 90 s, which is more rapid than the conventional cultivation method requiring several days at longest. Furthermore, the difference between the estimated concentrations by our method and by the cultivation method is less than 10%, which sufficiently guarantees the availability. The clarified principle will pave the way to a rapid and high throughput bacterial assay useful for medical care and food safety.
© 2016 Optical Society of America
1. Introduction
Bacterial counting methods are required in many fields such as food safety and medical care. In the case of food safety, environmental monitoring and bacterial analysis during manufacturing processes are important to prevent bacterial contamination of food products. In addition, clinical tests and prevention of nosocomial infections are necessary in the field of medical care. In both cases, a highly precise, rapid, and low-cost bacterial counting method is required. Cultivation, fluorescent staining, and dielectrophoretic methods are used as the traditional and mainstream bacterial counting methods. In particular, the cultivation method is used widely in the laboratory due to its low cost; however, the measuring time is on the order of several days [1,2 ]. The measuring time is shorter for the fluorescent staining method, but this method has lower precision and requires expensive fluorescent dyes and highly skilled professionals [3–5 ]. In the dielectrophoretic method, bacteria are gathered onto an electrode and the number determined from the correlation between impedance change and concentration [6–8 ]. This method requires only a few minutes to measure the bacterial concentration and has high precision, but its application is limited to particular bacteria depending on their electric properties. Thus, there are many targets for improving bacterial counting to satisfy the need for a rapid, label free, accurate, and versatile assay.
Photothermal assembling (PTA) enables the assembly of various floating objects into a desired position in a liquid. This method assembles small objects by using laser light and light absorptive material. When the light absorptive material is irradiated with laser light, the light energy is converted into heat. The localized heat generates convection flow and a bubble in the liquid. The flow then assembles floating objects toward the bubble and traps them. Recently, PTA has attracted the interests of many researchers because of its many potential applications. For example, PTA can detect a very small amount of protein [9] or the local assembly of microparticles or nanoparticles [10–13 ], and can be used for the colloidal lithography [14].
Here, inspired by these previous studies, we estimated an initial concentration of bacteria by PTA from the number of assembled targets using a flat metallic thin film as the heat source. Target bacteria were collected from a wide area within the liquid by local heating of the thin film by laser irradiation. In a preliminary experiment, we performed PTA using known concentrations of polystyrene microbeads, and investigated the assembling number to the total number of microbeads. Furthermore, we determined the optimum condition for reproducibility of this technique by changing the several parameters of laser. Subsequently, in the main experiments, we used Pseudomonas aeruginosa as the target bacteria and estimated their concentrations using PTA and compared with the conventional cultivation method.
2. Experimental
2.1 Sample preparation
Samples for the preliminary experiment were prepared using two types of concentrations of 1.0 μm polystyrene microbeads (Fluoresbrite Carboxylate Microspheres (2.5% Solids-Latex), 1.0 µm BB; Polysciences, USA): 4.55 × 108 and 4.55 × 105 particles/mL. Ultrapure water was used as the dispersant medium.
For the preliminary experiment with bacteria, Escherichia coli (E. coli), the typical bacteria that are easy to cultivate, were gathered by PTA, where SYTO9 (Green Fluorescent Nucleic Acid Stain S34854; Thermo Fisher, USA) was used as a fluorescent dye emitting green fluorescence attached to these bacteria. In the main bacterial counting experiments, a dispersion of Pseudomonas aeruginosa (P. aeruginosa), a Gram-negative aerobic bacillus, was used. Both types of bacteria were cultured in 30 mL of nutrient broth, and then dispersed in sterilized ultrapure water. These bacteria have a circular columnar shape, and E. coli has a diameter of 0.68 ± 0.12 μm and length of 1.69 ± 0.49 μm, and P. aeruginosa has a diameter of 0.59 ± 0.09 μm and length of 1.72 ± 0.36 μm as shown in our previous work [15]. The bacteria were washed three times with sterilized ultrapure water by centrifuging at 3502 × g for 5 min (6000 rpm, Max. radius is 8.7 cm). Sample liquids containing P. aeruginosa were then prepared with different concentrations. In the cultivation method, we used Petrifilm (PetrifilmTM; 3M Health Care, USA) and estimated the concentration of bacteria three times.
2.2 Experimental design for PTA
A schematic of the experimental setup for PTA is shown in Fig. 1(a) . An inverted microscope equipped with a laser light source of 1064 nm (Eclipse Ti-U; Nikon, Japan) was used. The diameter of the laser spot was set to 2.5 μm by using a 40 × objective (N. A. = 0.60). The laser power after penetrating the cover glass was measured by a power meter (UP17P-6S-H5 with tuner; Gentec Electro-Optics, Canada) and this value was used in these experiments. Gold film (10 nm) was deposited on a cover glass as the light absorptive substrate, which was fabricated by a sputtering instrument (E-1010; Hitachi Ltd., Japan) and the thickness was measured by a profilometer (Dektak 150; Takaoshin Ltd., Japan).
Fig. 1 (a) Schematic of the experimental setup. (b) Schematic of the PTA process. (c) Fluorescence images of preliminary experiments of PTA using SYTO9-stained E. coli of different concentrations.
2.3 PTA procedures
A 2.5 μL sample was placed onto the substrate and then irradiated with laser light at the center of the droplet while focusing on the gold film. Initially, PTA was performed using two types of sample liquids containing fluorescently dyed E. coli. We then confirmed that bacteria could be assembled into the space between the bubble and substrate through the mechanism shown in Fig. 1(b). As preliminary experiments, the E. coli with fluorescent dye were observed after assembling by 30 s laser irradiation under dark field conditions using a cooled charge-coupled device (CCD) camera for different concentrations (left: 5.2 × 108, right: 5.2 × 106 estimated by Petrifilm) (Fig. 1(c)).
Next, we optimized the PTA conditions for polystyrene microbeads by changing the laser power from 100 to 300 mW (in 50 mW increments) and the irradiation time from 30 to 90 s (in 30 s increments) (Fig. 2 ). The dynamics of assembly during laser irradiation under bright field conditions were observed using a cooled CCD camera. After completing laser irradiation, images were recorded under focusing on both the substrate and the equatorial plane of the bubble. In the main experiments with P. aeruginosa, we used the optimum condition obtained by the above experiment with microbeads for PTA without the fluorescent dye to develop the label-free bacteria counting method (Fig. 3 and 4 ). The first sample contained a high bacterial concentration (~108 cells/mL) and the second contained a low concentration (~105 cells/mL)
Fig. 2 (a) Optical transmission images of microbeads assembled by PTA (irradiation time is shown in white text). (b) Equatorial plane of a bubble. (c) Assembling percentage of polystyrene microbeads. In each curve, the average of five time measurements was taken. (left: high concentration, right: low concentration. The error bars mean standard deviation (SD).).
Fig. 3 Optical transmission images of bacteria assembled by PTA (left: high concentration, right: low concentration. The scale bar in the inset indicates 5.0 μm).
Fig. 4 Estimated bacterial concentrations (conc.) by the PTA and cultivation method (Error bars: SD).
3. Results and discussion
The first PTA experiment was performed using two concentrations of polystyrene microbeads. In this experiment, the polystyrene microbeads were assembled around the bubble by PTA. The number of assembled microbeads was calculated and the assembling percentage estimated by dividing this value by the total number of microbeads in the sample liquid. We performed this estimation for each condition and selected the optimum condition as the one that had the greatest reproducibility of the assembling percentage. We also used different microbeads concentrations to investigate how this affected the assembling percentage.
In the second experiment, we performed PTA to estimate the concentration of bacteria using the optimum condition obtained by the microbeads experiment and following Eq. (1),
where C f indicates the initial bacterial concentration in the sample liquid (/mL), N indicates the number of assembled objects, D indicates the dilution, α indicates the assembling percentage, and V indicates the volume of droplet (mL). We determined α for the polystyrene beads and used this value in the main experiments with bacteria. We also compared the bacterial concentrations obtained by PTA with that of the cultivation method.3.1 Assembling percentage and optimum condition obtained by the microbeads experiment
Polystyrene microbeads (high concentration) were assembled around the bubble by irradiating the gold film with the laser light (300 mW) for 30, 60, or 90 s (Fig. 2(a)). Longer laser irradiation times allowed the assembly of a larger number of microbeads around the bubble (Fig. 2(c), left panel). In the case of low microbeads concentration (Fig. 2(c), right panel), the assembling percentage seems to non-monotonically increase due to the variability in counting by human eyes depending on the observation region. When irradiating the gold film with laser light, a bubble forms and convection flow occurs with assembly dispersions around the bubble (Fig. 1(b)). During laser irradiation, the bubble gradually enlarges and the number of assembled microbeads is also increased. When laser irradiation ends, the bubble remains attached to the gold film. The equatorial plane of the bubble, after finishing laser irradiation, is shown in Fig. 2(b). From Figs. 2(a) and 2(b), we obtained the radius range where microbeads assembled (d), the radius of the plane where the bubble attached to the gold film (r), and the radius of the bubble (R). From r and R, we calculated the height from the gold film to the equatorial plane of the bubble (h). We then calculated the number of assembled microbeads for a high concentration by using these values and the following equation,
where the volume of the space filled with microbeads between the substrate and the bubble was divided by the volume of a microbead V particle. Assuming that this space is densely filled with microbeads, and multiplying N obtained from Eq. (2) with the filling rate of 0.74 under the close packing condition, we determined the assembling percentage α from Eq. (1). These calculations were performed for the high microbeads concentration. For the low microbeads concentration, the number of assembled microbeads was counted from the image obtained by optical microscopy. The assembling percentage of polystyrene microbeads for each condition is shown in Fig. 2(c). When laser irradiation occurred with high power or for a long time, the number of assembled microbeads tended to increase. In order to investigate the optimum condition to improve reproducibility, the relative error of the assembling percentage was evaluated under each condition. The laser power was 250 mW and the irradiation time 90 s, which was the optimum condition within the present experiment. When the laser power was higher, the growth process of the bubble was unstable and the convection flow became intense. Thus, to improve reproducibility, the laser power must remain relatively low to avoid such an instability. When the irradiation time was very long, the radius where microbeads assembled (d) became larger than the radius of the bubble (R). Under the above-mentioned condition taking into account these factors, the assembling percentage with high and low microbeads concentrations was 1.74% and 1.77%, respectively. These results indicate that the PTA method need not be regulated individually for the sample liquids at different concentrations.3.2 Estimated concentrations of bacteria by PTA and comparison with the cultivation method
For bacterial counting, we used PTA at high and low bacterial concentrations under the optimum condition determined with polystyrene microbeads (Fig. 3). There was little difference in the assembling processes between microbeads and bacteria. However, after completing laser irradiation, we found that some bacteria escaped from the bubble. Therefore, we recorded images on the substrate just before finishing laser irradiation and used these images to calculate the number of assembled bacteria. This differed from the experiment with microbeads where we recorded the image after finishing laser irradiation. When calculating the number of assembled bacteria, the denominator of Eq. (2) was changed into bacterial volume V bacterium. The shape of these bacteria were assumed to be columnar with a radius of 0.35 μm and length of 2.0 μm, and a filling rate of 1.0. Consequently, the number of assembled bacteria was 2.6 × 104 and 37 cells for high and low concentrations, respectively. From these results, and the assembling percentage obtained by the first experiment, the initial concentrations of bacteria were estimated to be 6.1 × 108 and 8.4 × 108 cells/mL from the high and low concentration conditions, respectively (Fig. 4). For comparison, we measured the bacterial concentration three times by using Petrifilm. Using this method, the bacterial concentration was estimated to be 5.6 × 108 cells/mL. In each bar graph for PTA, the average of five time measurements was taken. This means that the initial concentrations estimated by PTA were similar to that of the cultivation method, and the result from high concentration condition is very close to the value estimated by Petrifilm (only 8.9% difference in concentration). The large difference with the low bacterial concentration by PTA can be attributed to variations in the stirring and dilution processes.
4. Conclusion
The bacterial concentrations obtained by PTA were highly correlated with the cultivation method. Therefore, the PTA method was useful to count the number of bacteria floating in a liquid. The PTA method enabled us to measure the bacterial concentration in only 90 s while the measuring time was 24-48 h in the cultivation method, drastically shortening the measuring time. Furthermore, this method may be applicable to various kinds of bacteria because differences of shape and electric properties do not almost affect the assembling percentage in PTA. Establishing this technique for routine use could have a remarkable impact on bacteria counting in the fields of food safety, medical care, and laboratory analyses. In the future, we will investigate the effective area needed to assemble small objects by PTA in a sample liquid, and other bacteria to confirm the broad applicability of this method.
Acknowledgments
The authors would like to thank Prof. S. Ito, Prof. I. Nakase, and Prof. K. Imura for their helpful discussions. We also thank Prof. H. Shiigi, Prof. T. Nagaoka, and Prof. H. Ishihara for their encouragement. This work was supported by Grants-in-Aid for Scientific Research (B) (No. 15H03010, No. 26286029), Grants-in-Aid for Exploratory Research (No. 15K14697, No. 24654091) from JSPS, The Canon Foundation, SENTAN-JST, Center for the Promotion of Interdisciplinary in OPU, and a Grant-in-Aid for JSPS Fellows (No. 15J09975), and the Cooperative Research Program of “Network Joint Research Center for Materials and Devices”.
References and links
1. J. L. Johnson, C. L. Brooke, and S. J. Fritschel, “Comparison of the BAX for Screening/E. coli O157:H7 Method with Conventional Methods for Detection of Extremely Low Levels of Escherichia coli O157:H7 in Ground Beef,” Appl. Environ. Microbiol. 64(11), 4390–4395 (1998). [PubMed]
2. V. Jasson, L. Jacxsens, P. Luning, A. Rajkovic, and M. Uyttendaele, “Alternative microbial methods: An overview and selection criteria,” Food Microbiol. 27(6), 710–730 (2010). [CrossRef] [PubMed]
3. H. Christensen, M. Hansen, and J. Sørensen, “Counting and size classification of active soil bacteria by fluorescence in situ hybridization with an rRNA oligonucleotide probe,” Appl. Environ. Microbiol. 65(4), 1753–1761 (1999). [PubMed]
4. A. Pascaud, S. Amellal, M. L. Soulas, and G. Soulas, “A fluorescence-based assay for measuring the viable cell concentration of mixed microbial communities in soil,” J. Microbiol. Methods 76(1), 81–87 (2009). [CrossRef] [PubMed]
5. R. Orth, N. O’Brien-Simpson, S. Dashper, K. Walsh, and E. Reynolds, “An efficient method for enumerating oral spirochetes using flow cytometry,” J. Microbiol. Methods 80(2), 123–128 (2010). [CrossRef] [PubMed]
6. A. P. Brown, W. B. Betts, A. B. Harrison, and J. G. O’Neill, “Evaluation of a dielectrophoretic bacterial counting technique,” Biosens. Bioelectron. 14(3), 341–351 (1999). [CrossRef] [PubMed]
7. B. H. Lapizco-Encinas, B. A. Simmons, E. B. Cummings, and Y. Fintschenko, “Dielectrophoretic concentration and separation of live and dead bacteria in an array of insulators,” Anal. Chem. 76(6), 1571–1579 (2004). [CrossRef] [PubMed]
8. T. Kikutani, F. Tamura, Y. Takahashi, K. Konishi, and R. Hamada, “A novel rapid oral bacteria detection apparatus for effective oral care to prevent pneumonia,” Gerodontology 29(2), e560–e565 (2012). [CrossRef] [PubMed]
9. Y. Nishimura, K. Nishida, Y. Yamamoto, S. Ito, S. Tokonami, and T. Iida, “Control of submillimeter phase transition by collective photothermal effect,” J. Phys. Chem. C 118(32), 18799–18804 (2014). [CrossRef]
10. S. Fujii, K. Kanaizuka, S. Toyabe, K. Kobayashi, E. Muneyuki, and M. A. Haga, “Fabrication and placement of a ring structure of nanoparticles by a laser-induced micronanobubble on a gold surface,” Langmuir 27(14), 8605–8610 (2011). [CrossRef] [PubMed]
11. H. Xin, H. Lei, Y. Zhang, X. Li, and B. Li, “Photothermal trapping of dielectric particles by optical fiber-ring,” Opt. Express 19(3), 2711–2719 (2011). [CrossRef] [PubMed]
12. H. Xin, X. Li, and B. Li, “Massive photothermal trapping and migration of particles by a tapered optical fiber,” Opt. Express 19(18), 17065–17074 (2011). [CrossRef] [PubMed]
13. H. Xin, D. Bao, F. Zhong, and B. Li, “Photophoretic separation of particles using two tapered optical fibers,” Laser Phys. Lett. 10(3), 036004 (2013). [CrossRef]
14. B. Roy, M. Arya, P. Thomas, J. K. Jürgschat, K. V. Rao, A. Banerjee, C. M. Reddy, and S. Roy, “Self-assembly of mesoscopic materials to form controlled and continuous patterns by thermo-optically manipulated laser induced microbubbles,” Langmuir 29(47), 14733–14742 (2013). [CrossRef] [PubMed]
15. S. Tokonami, Y. Nakadoi, M. Takahashi, M. Ikemizu, T. Kadoma, K. Saimatsu, Q. Dung, H. Shiigi, and T. Nagaoka, “Label-Free and Selective Bacteria Detection Using a Film with Transferred Bacterial Configuration,” Anal. Chem. 85(10), 4925–4929 (2013). [CrossRef] [PubMed]
