We propose a bacterial detection scheme which uses no biochemical markers and can be applied in a Point-of-Care setting. The detection scheme aligns asymmetric bacteria with an electric field and detects the optical scattering.
© 2006 Optical Society of America
The ability to align asymmetric biological particles immersed in a solution of different permittivity with an alternating electric field has been well characterized and applied to the study of several types of elongated cells such as yeast, erythrocytes, bacteria, and algae [1–4]. The ability to measure the orientation and biophysical characteristics of individual bacteria through optical diffraction and scattering techniques has also been well characterized and applied to the study of several species of bacteria [5, 6]. In this paper, we demonstrate a bacteria detection scheme that measures the optical scattering from aligned asymmetric bacteria in the presence of an applied alternating electric field by exploiting the fact that most bacteria or bacteria aggregates of interest are asymmetric. We also observed that the alignment occurs only in living bacteria with functional cellular membranes since the cellular membrane becomes permeable when the bacteria die .
The optical setup, shown in Fig. 1, consists of a 5 mW HeNe laser which illuminates a transparent specimen holder consisting of a 1 mm thick glass plate with a conductive electrode pattern in indium tin oxide (ITO) and a thin cover glass, separated by a 20 µm spacer. The void between the glass plates holds approximately 2 µL of test specimen. The laser beam has a diameter of approximately 1.5 mm resulting in an interaction volume of approximately 150 nL. An array of photodiodes at different angles or a simple optical power detector (UDT S370) at a continuous variable angle is used to measure the optical scattering. The specimen holder is connected to a signal generator to provide an alternating voltage of ±10 V.
An alternating voltage was chosen to eliminate electrolysis in the test specimen at the electrodes. Electrolysis causes bubbles of hydrogen and oxygen gas to form at the boundary of the electrode and liquid. To avoid the creation of bubbles without decreasing the applied voltage, a frequency greater than 1 MHz was necessary. Under observation through the microscope, the asymmetric bacteria aligned to the electric field most efficiently at 10 MHz.
The specimen holder, as shown in Fig. 2, is fabricated using contact photo lithography. First, we obtain ITO coated glass plates (Aldrich) rated at 30–60 Ω/square. Next, positive photoresist (S1813) is spincoated onto the plates. A pattern is exposed using a UV mask aligner and developed. Finally, the ITO is etched to create the electrode pattern.
The conductive electrode pattern is interdigitated as shown in Fig. 2. In determining the optimal electrode spacing and width for optical scattering, there are three factors to consider. First, it has been observed that bacteria do not align over the electrodes. Therefore, the electrode width should be minimized. Second, the spacing between electrodes should be minimized to increase the electric field strength. Third, the electrode width should be maximized to decrease the resistance of the electrode. By testing the performance of different electrode spacing and widths from 10 microns to 500 microns, the optimal electrode spacing and width was determined to be 200 microns and 100 microns, respectively.
To visualize the effects of the electric field on asymmetric bacteria in solution, the specimen holder is placed under a microscope. Our sample of bacteria is E. coli (K12), which are rod-shaped bacteria. When no electric field is applied, live E. coli move randomly and are aligned randomly as shown in Fig. 3(a). When the electric field is applied, live E. coli align to the field as shown in Fig. 3(b); we observe the shorter E. coli aligning rapidly and the longer E. coli aligning slowly. Even at the corners of the electrodes, E. coli align to the field lines, as shown in the inset of Fig. 2. When the electric field is turned off, live E. coli move to orient themselves randomly. Note that dead E. coli and other symmetric particles do not move or change orientations with respect to the electric field and appear stuck to either the glass plate or cover slide.
The orientation of the bacteria need not be parallel to the image plane. The orientation of the bacteria may be in and out of the image. By stacking two glass plates with ITO as shown in Fig. 4, we were able to align bacteria perpendicular to the plane of the device. Bacteria that are located just outside of the electrode boundary do not align and remain in random orientations.
The ability to detect bacteria through an automated process without expensive and bulky equipment such as an optical microscope would be beneficial for many reasons. First, untrained technicians could operate the equipment. Second, the results may be obtained rapidly. Third, the system could be made small and inexpensive.
To eliminate the need for an imaging system, we have investigated optical scattering as a means to detect the presence of live bacteria. Randomly oriented rod-shaped bacteria have optical scattering measurements with peaks as a function of scattering angle corresponding to the bacteria’s radius and length [5, 6]. Aligned rod-shaped bacteria have optical scattering measurements with peaks as a function of scattering angle corresponding to either the bacteria’s radius or length. In our scheme, the automated detection compares the optical scattering of randomly oriented bacteria versus aligned bacteria.
Figure 5 shows the optical scattering for samples of 5·107 colony forming units (CFU)/mL live E. coli in urine and no live E. coli in urine. When the electric field is turned on (between the dashed green lines), we notice that the sample of 5·107 CFU/mL live E. coli in urine has a scattered power increase of 20%. A control experiment with filtered and sterilized urine that contained dead E. coli showed no change in the amount of scattered power with an applied electric field.
The optical scattering measurement shows different rise and fall times when the electric field is turned on and off. We noticed that bacteria of different sizes align at different speeds to the electric field. For our particular sample of 5·107 CFU/mL live E. coli, we have a rise time measurement of approximately 500 milliseconds and a fall time measurement of approximately 1.5 seconds as shown in Fig. 6. The different rise and fall times of bacteria may be due to their different sizes, dielectric constant or polarizability. Therefore, through calibration, we may be able to distinguish the types of different bacteria in our test sample.
Figure 7 shows optical scattering measurements with varying bacteria concentration and angle of the detector. Figure 7(a) shows similar measurements as shown in Fig. 5. On average, the scattering measurements indicate that there is about a 20% increase when the electric field is turned on for 5·107 CFU/mL live E. coli in urine. Figure 7(b) shows measurement data for 5·106 CFU/mL live E. coli in urine. There is about a 5% increase when the electric field is turned on. We note that the threshold of the optical scattering measurement for detecting bacteria occurs at a concentration which is on the order of magnitude 106 CFU/mL live E. coli in urine.
It has been demonstrated previously that bacteria may be concentrated using a microfluidic circuit that takes advantage of dielectrophoresis (DEP) to redirect live bacteria in to a reservoir . This device has the ability to concentrate live bacteria at four to five orders of magnitude higher than the original concentration. It is fabricated in a similar technique as our device and can be directly integrated with the bacteria detector we describe. Combining the bacteria concentrator with our detection scheme could be useful in diagnosing bacteria infections.
Current bacteria detection methods include urine cultures  and biosensor techniques that detect antigen-antibody, enzyme-substrate, or receptor-ligand complexes by measuring fluorescent light, surface reflection, and electrical properties [10–13]. These methods can be time and labor consuming (12–24 hours for urine culture), expensive, or complex to operate. We have demonstrated a bacteria detection scheme that uses no biochemical markers and has the potential to be small, inexpensive, easy to use for untrained technicians, and provides rapid results.
The authors thank Harvey L. Kasdan and Iris Diagnostics for providing samples of E. coli and acknowledge the Caltech Watson lab. The research was supported by the DARPA center for optofluidic integration.
References and links
1. T. B. Jones, “Basic theory of dielectrophoresis and electrorotation,” IEEE EMBS 22, 33–42 (2003).
3. M. Kriegmaier, M. Zimmermann, K. Wolf, U. Zimmermann, and V. L. Sukhorukov, “Dielectric spectroscopy of Schizosaccharomyces pombe using electro rotation and electro orientation,” Biochim. Biophys. Acta. 1568, 135–146 (2001). [CrossRef] [PubMed]
4. O. V. Ignatov, O. I. Guliya, S. Y. Shchyogoleva, V. D. Buninb, and V. V. Ignatova, “Effect of p-nitrophenol metabolites on microbial cell electro-optical characteristics,” FEMS Microbiol. Lett. 214, 81–86 (2002). [CrossRef] [PubMed]
6. A. Katz, A. Alimova, M. Xu, E. Rudolph, M. K. Shah, H. E. Savage, R. B. Rosen, S. A. McCormick, and R. R. Alfano, “Bacteria size determination by elastic light scattering,” IEEE J. Quantum Electron. 9, 277–287 (2003). [CrossRef]
8. R. Gomez-Sjoberg, D. T. Morisette, and R. Bashir, “Impedance microbiology-on-a-chip: microfluidic bioprocessor for rapid detection of bacterial metabolism,” IEEE J. MEMS 14, 829–838 (2005).
9. A. K. Bhunia and A. Lathrop, Pathogen Detection, Food-borne (McGraw- Hill, New York, NY, 2003).
10. J. Homola, S. S. Yee, and G. Gauglitz, “Surface plasmon resonance sensors: review,” Sens. Actuators B 54, 3–15 (1999). [CrossRef]
11. C. Gauthier, Y. St-Pierre, and R. Villemur, “Rapid antimicrobial susceptibility testing of urinary tract isolates and samples by flow cytometry,” J. Med. Microbiol. 51, 192–200 (2002). [PubMed]
12. K. Roberts, M. Parameswaran, M. Moore, and R. S. Muller, “A silicon microfabricated aperture for counting cells using the aperture impedance technique,” Can. J. Elect. Comput. Eng. 24, 109–113 (1999).
13. M. Pezzlo, “Detection of urinary tract infections by rapid methods,” Clin. Microbiol. Rev. 1, 268–280 (1988). [PubMed]