In-Line Monitoring of Bacteria in Drinking Water By Infrared Spectroscopy and Micro-Flow Cytometry

Water from a municipal network water supply is stored in tanks and then pumped to tanks in tall buildings, which is so called “second water supply”. However, the water stored in tanks  provides the potential  environment for bacteria to grow and multiply. It is labor and time-consuming to routinely and frequently sample water from the water tanks distributed in a city, and to test all samples by biological methods in labs, which thus restricts the detection and prevention of water-borne diseases. Therefore, a rapid and low-cost method^1,2 is urgently needed to routinely and automatically monitor and report the bacterial level in water tanks, allowing for taking appropriate actions in time.

    Infrared (IR) spectroscopy is a powerful tool for characterization and identification of micro-organisms because of its sensitivity and specificity, speed, and potential for automation. Microorganisms have unique “fingerprint” spectral features in the mid-infrared (MIR) range. Therefore, IR spectroscopy is a feasible technique for the identification of unknown bacterial samples based on the pre-established bacterial spectrum libraries.

    There are mainly two modes in IR spectroscopy to obtain the IR spectrum of bacteria in water - attenuated total reflection (ATR) and transmission modes. The first mode has been used by many researchers previously because it could avoid the influence of water on the bacterial spectrum, which has three strong absorption bands in the range of 400 to 4000 cm^-1. However, the ATR technique lacks the ability to provide spectral data for quantitative analysis because the spectrum acquired in ATR mode is dependent on the thickness and density of the bacterial film formed on the outer surfaces of the ATR crystal. Therefore the transmission mode of FTIR is used in this research.

    Three strains of two species, E.coli K-12 MG1655 and E.coli K-12 HS2210 from Ecoli and Staphylococcus aureus (ATCC 25923) were placed directly into double distilled water. Then, the CaF₂ cuvette containing bacteria suspention with different concentrations were used to obtain the IR spectra using a Vertex v80 FTIR spectrometer. Ultra-high concentrations of bacteria over 10¹⁰ cells/ml are used to guarantee enough changes in absorption between double distilled water and samples. Time-resoved IR spectra of E.coli K-12 MG1655 shows that its spectrum varies slightly within one hour. Obvious differences were observed in the spectra of E.coli K-12 MG1655 and Staphylococcus aureus (ATCC 25923), which shows the specificity on the IR spectrum of different species. However, similar differences were not found between E.coli K-12 MG1655 and HS2210. In addition, the intensity of the absorption peaks in the spectra of all the three strains is proportional to the concentration of the samples.

Because of limitation of FTIR spectroscopy on quantifying bacteria accurately, a photonic-microfluidic integrated device based flow cytometer^3-5 is incorporated into the monitoring system to ensure reliable estimation of bacterial concentration. The functionality layer of the device including input and output waveguides, lens systems and microfluidic channels is fabricated by patterning SU-8 photoresist using photolithography, which is actually supported by a pyrex glass substrate and covered by a PDMS covering layer with holes aligned with all the inlets and outlet. Light from a laser is directed into the input waveguide of the device through an optic fiber, then focused to the center of the microfluidic channel by a lens system, where the bacteria will be interrogated. Bacterial suspension in a syringe is injected to the sample inlet by a syringe pump and hydro-focused into the center of microfluidic channel by the sheath flows at both sides. Scattered light produced by bacteria passing through the focused light beam will be collected by the objective and filtered by an optic filter with passband of 525-535 nm. A photomultiplier tube is used to detect, convert and amplify the light signals into current signals, which will be further amplified into votage singals by a current to voltage amplifier. Voltages signals will be digitized by a data acqusition card and recorded by a custom LabView program for data analysis.

The performance of flow cytometer based on the integrated devices with different beam shaping lens systems was calibrated by detecting scattered light from 1µm-diameter blank polystyrene beads running through the device. Due to uniform geometry, size and material, the device yielding minimal coefficient of variation (CV) will be the best for applications.

Enumeration of E.coli in phosphate-buffered saline was carried out to validate the accuracy of counting bacteria using this micro-chip based flow cytometer. A 92% of detection efficiency was achieved. Expanded population of distribution of E. coli counting than beads counting was found because of the more flexibility of scattering light by rod-shape of E.coli than sphere-shape of beads as well as the natural size variation of E. coli. Tests on mixture of E.coli and 2µm, 4 µm beads show distinct separation of the population distribution, confirming the ability to differentiate E.coli from larger size inorganic particles by scattering intensity.

References:

1. M. W. Shinwari, M. J. Deen and D. Landheer, “Study of the Electrolyte-Insulator-Semiconductor Field-Effect Transistor (EISFET) with Applications in Biosensor Design,” Microelectronics Reliability,  47(12), 2025-2057 (December 2007). 
2. Z. Li, M.J. Deen, Q. Fang, P.R. Selvaganapathy, "Design of a flat field concave-grating-based micro-Raman spectrometer for environmental applications"  Appl Optics, 51, 6855-6863 (2012).
3. B. Watts, Z. Zhang, C-Q. Xu, X. Cao, and M. Lin, “A photonic-microfluidic integrated device for reliable fluorescence detection and counting,” Electrophoresis 33(21), 3236-3244 (2012).
4. B. R. Watts, T. Kowpak, Z. Zhang, C-Q. Xu, S. Zhu, X. Cao, and M. Lin, “Fabrication and Performance of a Photonic-Microfluidic Integrated Device,” Micromachines 3(1),  62-77 (2012).
5. B. R. Watts, Z. Zhang, C-Q. Xu, X. Cao, and M. Lin, “Integration of optical components on-chip for scattering and fluorescence detection in an optofluidic device,” Biomed. Optics Exp. 3(11), 2784-2793 (2012).