Convective Assembly of Bacteria for Surface-Enhanced Raman

Jarvis et al. prepared the samples by simple mixing of the bacteria with the silver ...... Jean C. Lee , Milos Miljkovic , Catherine M. Klapperich , A...
0 downloads 0 Views 1MB Size
894

Langmuir 2008, 24, 894-901

Convective Assembly of Bacteria for Surface-Enhanced Raman Scattering Mehmet Kahraman, M. Mu¨ge Yazıcı, Fikrettin S¸ ahin, and Mustafa C¸ ulha* Yeditepe UniVersity, Faculty of Engineering and Architecture, Genetics and Bioengineering Department, Kayisdagi, Istanbul, Turkey ReceiVed July 24, 2007. In Final Form: October 22, 2007 A sample preparation method based on convective assembly for “whole-microorganism” identification using surfaceenhanced Raman scattering (SERS) is developed. With this technique, a uniform sample can easily be prepared with silver nanoparticles. During the deposition process, bacteria and nanoparticles are assembled to form a unique wellordered structure with great reproducibility. The SERS spectra acquired from the samples prepared with this technique have better quality and improved reproducibility for SERS spectra obtained from the same sample and limited variation due to the consistent sample preparation. E. coli, a Gram-negative bacilli, and Staphylococcus cohnii, a Gram-positive coccus, are studied as model bacteria.

Introduction The fast identification of microorganisms is critical for the survival of infected individuals. Today, many conventional methods are available to clinicians for microbial identification. However, almost all are still laborious and expensive. In recent years, the desire to bypass the time-consuming procedures has resulted in the idea to utilize spectroscopic methods for “wholemicroorganism” identification. Mass spectroscopy (MALDI and PyMS),1-3IR, and Raman spectroscopy are the examples reported to date.4-9 The greatest advantage of these techniques is to use a whole microorganism with a limited number of microbial cells. Among these techniques, Raman spectroscopy has drawn considerable attention due to its compatibility with the biological samples and easy sample preparation. The only drawback of this technique is the weak signal that is the nature of Raman scattering phenomena. This inefficient scattering requires longer signal collection times and the use of higher laser power for the acquisition of Raman spectra, which may result in damage to the sample. The weak Raman scattering can be enhanced by many orders of magnitude when a molecule of interest is brought close to the surface of a noble metal such as gold or silver.10-12 This * Corresponding author. Telephone: +90 (216) 578 1587. Fax : +90 (216) 578 0829. E-mail: [email protected]. (1) Claydon, M. A.; Davey, S. N.; Edwards-Jones, V.; Gordon, D. B. Nat. Biotechnol. 1996, 14, 1584-1586. (2) Goodacre, R.; Hiom, S. J.; Cheeseman, S. L.; Murdoch, D.; Weightman, A. J.; Wade, W. G. Curr. Microbiol. 1996, 32, 77-84. (3) Haag, A. M.; Taylor, S. N.; Johnston, K. H.; Cole, R. B. J. Mass. Spectrom. 1998, 33, 750-756. (4) Naumann, D.; Helm, D.; Labischinski, H.; Giesbrecht, P. The characterization of microorganisms by Fourier transorm infrared spectroscopy (FT-IR). In Modern Techniques for Rapid Microbiological Analysis, Nelson, W. H., Ed.; VCH: New York, 1991. (5) Naumann, D. Proceedings of SPIE - The International Society for Optical Engineering 1998, 3257, 245-257. (6) Dalterio, R. A.; Nelson, W. H.; Britt, D.; Sperry, J. F.; Purcell, F. J. Appl. Spectrosc. 1986, 40, 271-272. (7) Dalterio, R. A.; Beak, M.; Nelson, W. H.; Britt, D.; Sperry, J. F.; Purcell, F. J. Appl. Spectrosc. 1987, 41, 241-244. (8) Puppels, G. J.; De Mul, F. F. M.; Otto, C.; Greve, J.; Robert-Nicoud, M.; Arndt-Jovin, D. J. Nature (London) 1990, 347, 301-303. (9) Puppels, G. J.; Garritsen, H. S. P.; Segers-Nolten, G. M. J.; De Mul, F. F. M.; Greve, J. Biophys. J. 1991, 60, 1046-1056. (10) Chang, R. K.; Furtak, T. E. Surface Enhanced Raman Scattering; Plenum: New York, 1982. (11) Otto, A. Light Scattering. In Solids IV, Topic in Applied Optics, Cardona, M., Guentherodt, G., Eds.; Springer: Heidelberg, 1984; p 289.

Figure 1. Convective assembly process of bacteria (A) and a photograph of E. coli cells assembled on a glass slide (B).

phenomenon is called surface-enhanced Raman scattering (SERS) and has found many application since its discovery.13 Either a surface or a colloidal solution of gold or silver is used for the SERS experiments. For bacterial SERS studies, a gold or silver colloidal solution was preferred over a noble metal surface most of the time.14-22 Only a few reports concerning the detection of a biomarker belonging to photogenic bacterial species on a noble metal surface have appeared.23,24 (12) Moskovits, M. ReV. Mod. Phys. 1985, 57, 783. (13) Fleischmannn, M.; Hendra, P. J.; Mac Quillan, A. J. Chem. Phys. Lett. 1974, 26, 163. (14) Efrima, S.; Bronk, B. V. J. Phys. Chem. B 1998, 102, 5947-5950. (15) Zeiri, L.; Bronk, B.; Shabtai, V. Y.; Eichler, J.; Efrima, S. Appl. Spectrosc. 2004, 58, 33-40. (16) Zeiri, L.; Efrima, S. J. Raman Spectrosc. 2005, 36, 667. (17) Sengupta, A.; Laucks, M. L.; Davis, E. J. Appl. Spectrosc. 2005, 59, 1016-1023. (18) Sengupta, A.; Mujacic, M.; Davis, E. J. Anal. Bioanal. Chem. 2006, 386, 1379. (19) Laucks, M. L.; Sengupta, A.; Junge, K.; Davis, E. J.; Swanson, B. D. Appl. Spectrosc. 2005, 59, 1222. (20) Jarvis, R. M.; Goodacre, R. Anal. Chem. 2004, 76, 40-47. (21) Jarvis, R. M.; Brooker, A.; Goodacre, R. Anal. Chem. 2004, 76, 51985202. (22) Premasiri, W. R.; Moir, D. T.; Klempner, M. S.; Krieger, N.; Jones, G., II; Ziegler, L. D. J. Phys. Chem. B 2005, 109, 312-320.

10.1021/la702240q CCC: $40.75 © 2008 American Chemical Society Published on Web 01/08/2008

ConVectiVe Assembly of Bacteria Via SERS

Langmuir, Vol. 24, No. 3, 2008 895

Figure 2. SEM image of a sample of E. coli generated by a simple mixing of 1× silver colloidal suspension.

Figure 3. SERS spectra of E. coli from a sample generated by a simple mixing of 1× silver colloidal suspension.

Since the report of the SERS of photosynthetic bacterium by Holt and Cotton,25 several reports have described the feasibility of using the SERS spectra for identification and classification.14-22 Efrima and Bronk and Zeiri et al. studied the cell wall and its components with SERS. They covered the bacterial cell wall with silver nanoparticles by reducing Ag+ in the presence of bacteria in solution.14-16 They also demonstrated that the SERS spectra obtained from the bacterium cell wall and inside the bacterium were different by introducing the silver nanoparticles inside of the bacterium cell.13 The SERS of bacteria and pollens were studied by Sengupta et al. and Laucks et al.17-19 They prepared the bacterial samples by placing the bacteria in the colloidal solution until the colloids were adsorbed on the bacterial (23) Zhang, X.; Young, M. A.; Lyandres, O.; Van Duyne, P. R. J. Am. Chem. Soc. 2005, 127 (12), 4484 -4489. (24) Yan, F.; Vo-Dinh, T. Sens. Actuators, B 2007, 121 (1), 61-66. (25) Holt, R. E.; Cotton, T. M. J. Am. Chem. Soc. 1989, 111, 2815.

cell wall. The influence of colloid reproducibility, particle size, and aggregation on SERS spectra in solution was successfully investigated. They stated that the bacterial SERS spectra evolve over time and stabilize within a few hours, which is an indicator of the adsorption of colloidal silver particles on the bacterial cell wall. Jarvis et al. prepared the samples by simple mixing of the bacteria with the silver colloidal solution and acquired the SERS spectra from dried sample at room temperature.20,21 Premasiri et al. first attached the gold nanoparticles on silica particles and then mixed with bacterial cells.22 Although the prepared sample was acknowledged as the source of irreproducibility in SERS spectra in all these studies, a uniform sample preparation method was not pursued. Simple mixing is a great method for fast sample preparation. However, the generated sample is not a uniform mixture of nanoparticles and bacterial cells. The use of such a sample results in large irreproducibility in the SERS spectra due to the requirement that the nanoparticles and bacterial cells must be under the impinging laser beam. The other factor contributing to the variation in SERS spectra could be the nonuniform biochemical structure of the bacterial cell wall and the interaction of silver nanoparticles with localized biochemical structures on the bacterial cell wall at the lower colloidal concentrations.18 However, with the increased colloidal solution concentration, the interaction of nanoparticles with the bacterium cell seems random due to the formation of larger aggregates and the saturation of localized structures on the bacterial cell wall. In our previous studies, we have demonstrated the improvement in SERS spectra quality and reproducibility with increasing colloidal solution concentration,26 which supports the idea that the bacterial surface (26) Kahraman, M.; Yazıcı, M. M.; Sahin, F.; Bayrak, O. F.; Culha, M. Appl. Spectrosc. 2007, 61 (5), 479-485.

896 Langmuir, Vol. 24, No. 3, 2008

Kahraman et al.

Figure 4. SEM image of a sample prepared with a simple mixing of E. coli cells and 8× colloidal suspension.

Figure 5. SERS spectra of E. coli obtained from the sample seen in Figure 4.

must be brought into contact with as many silver nanoparticles as possible to improve the reproducibility of the SERS spectra. The convective assembly is a technique in which a controlled assembly of two-dimensional (2D) or three-dimensional (3D) structures from nano- and micro-particles can be prepared. The mechanism of the convective assembly can be explained as the self-assembly of colloidal particles in thin evaporating films.27,28 The technique was used to prepare coatings from sterically protected and silica-encapsulated nanoparticles and porous 3D gold structures for SERS.29-33 The porous 3D SERS substrate was prepared with the convective assembly of nano- and (27) Denkov, N. D.; Velev, O. D.; Kralchevsky, P. A.; Ivanov, I. B.; Yoshimura, H.; Nagayama, K. Langmuir 1992, 8, 3183-3190. (28) Denkov, N. D.; Velev, O. D.; Kralchevsky, P. A.; Ivanov, I. B.; Yoshimura, H.; Nagayama, K. Nature (London) 1993, 361, 26-26. (29) Brust, M.; Bethell, D.; Kiely, C. J.; Schiffrin, D. J. Langmuir 1998, 14, 5425-5429. (30) Ung, T.; Liz-Marzan, L. M.; Mulvaney, P. Colloid Surf., A 2002, 202, 119-126. (31) Tessier, P. M.; Velev, O. D.; Kalambur, A. T.; Lenhoff, A. M.; Rabolt, J. F.; Kaler, E. W. AdV. Mater. 2001, 13, 396.

Figure 6. SERS spectra of S. cohnii from a sample generated by a simple mixing of 1× silver colloidal suspension.

microsized latex and concentrated gold nanoparticles. During the assembly process, gold or silver nanoparticles fill the voids around the latex particles, forming a colloidal crystal. After washing the latex particles from the structure with an organic solvent, the remaining porous three-dimensional structure can be used as a SERS substrate.33 In this study, we report the development of a uniform bacterial sample preparation method based on the convective assembly, which is already an established technique for the preparation of thin films. First, we demonstrate the influence of colloidal suspension concentration and, then, the ordered surface on SERS spectra. This method deposits bacteria and silver nanoparticles on a glass slide as a thin film in an ordered structure. The reported (32) Tessier, P. M.; Velev, O. D.; Kalambur, A. T.; Rabolt, J. F.; Lenhoff, A. M.; Kaler, E. W. J. Am. Chem. Soc. 2000, 122, 9554. (33) Tessier, P. M.; Christesen, S. D.; Ong, K. K.; Clemente, E. M.; Lenhoff, A. M.; Kaler, E. W.; Velev, O. D. Appl. Spectrosc. 2002, 56, 1524.

ConVectiVe Assembly of Bacteria Via SERS

Langmuir, Vol. 24, No. 3, 2008 897

Figure 7. SEM image of S. cohnii mixed with 25× silver colloidal suspension.

method eliminates the spot-to-spot variations and generates reproducible and higher-quality SERS spectra. Because the shape, size, and biochemical structure of the bacterial cell wall may influence the convective assembly process, E. coli, a bacilliGram-negative bacterium, and S. cohnii, Gram-positive bacterium, are chosen for the study. Materials and Methods Chemicals. AgNO3 (99.5%) and nutrient agar are purchased from Fluka (Seelze, Germany). Sodium citrate (99%) is purchased from Merck (Darmstadt, Germany). All chemicals are used as received without further purification. Preparation of Bacteria Samples. E. coli (35218 ATCC) and Staphylacoccus cohnii were obtained from our microorganism collection (Yeditepe University, Genetics and Bioengineering Department). Both bacteria were verified by the Microbial Identification System (MIDI) before their use and were axenically and aerobically subcultered three times for 24 h at 37 °C on 20 mL nutrient agar. The bacteria were collected with sterile plastic inoculating loops on a solid culture plate. The collected samples were added into 1 mL d.i. water, vortexed and centrifuged for 5 min at 7500 rpm. The supernantent was discarded. This procedure was repeated three times. Bacterial cell count was performed using a UV spectrometer, Biolog Microsystem (Bio-Tek Ins. Inc., Winooski, VT), as percent transmittance (O.D. ) 2-log of % transmittance at 600 nm). 4 µL of each bacterial sample was serially diluted and the absorbance measurement was performed from the final dilution before culturing for colony count. The number of bacteria in the original sample was estimated by multiplication of the dilution factor by the number of colonies counted on the solid culture media.

Preparation of Silver Colloids and Bacterial Samples. Ag colloid was prepared by the method reported by Lee.34 Briefly, 90 mg AgNO3 was dissolved in 500 mL water. This solution was heated to a boil. A 10 mL aliquot of 1% sodium citrate was added into the solution and boiling maintained until the volume was reduced to half of the initial volume. The maximum of its absorption was recorded at 420 nm. This colloidal silver solution is referred to as 1× in the text. The silver colloidal suspension prepared with this method generates a mixture of several sizes and shapes of silver nanoparticles with an average diameter of 50 nm. The colloidal solution was concentrated by centrifugation at 5500 rpm for 30 min, discarding a portion of the supernatant and bringing the final concentration to 8× for E. coli and 25× for S. cohnii higher concentration of the initial concentration. For simple mixing samples, a 5 µL aliquot of the each washed bacterium was added into a specified concentration of colloid solution with a volume of 100 µL. Then, it was mixed with the pipet tip to create a mixture as homegenous as possible, and a 5 µL aliquot of this mixture was immediately transferred onto a CaF2 slide and dried at room temperature before analysis. Convective Assembly of Bacteria. The details of the method are described somewhere else.35 In contrast to the reported setup, the bottom slide was attached on a stage rather than the slide with an angle was mobile. The bacterial assembly process is depicted in Figure 1. First, the speed of the moving stage and the mixing ratio were first optimized for E. coli. It was found that a 4 µL/36 µL washed bacteria: 8× concentrated silver colloidal suspension ratio was optimum. The number of bacteria in 4 µL were estimated to be about 5.2 × 109 and 1.5 × 1010 for E. coli and S. cohnii, respectively, from the absorbance measurements. The mixture was spotted at the junction of two slides with a micropipette. The angle between two slides was about 24°. The velocity of the bottom stage was set to 1.0 µm/s for all measurements. An area of 0.5 cm × 2.5 (34) Lee, P. C.; Meisel, D. J. Phys. Chem. 1982, 88, 3391. (35) Prevo, B. G.; Velev, O. D. Langmuir 2004, 20, 2099.

898 Langmuir, Vol. 24, No. 3, 2008 cm was coated with bacteria and silver nanoparticles. The room temperature and relative humidity were 22-25 °C and 40-50% during the experiments, respectively. The average dimensions of E. coli cells were determined from SEM images as 0.90 µm × 1.8 µm and 0.90 µm for E. coli and S. cohnii, respectively. Raman Instrumentation. All measurements were performed using a completely automated Renishaw InVia Reflex Raman Microscopy System (Renishaw Plc., New Mills, Wotton-under-Edge Gloucestershire, U.K.) equipped with an 830 nm diode and an 514 nm Ar+ laser. The performance of the 830 nm diode laser was found to be superior for bacterial SERS studies with the use of aggregated silver nanoparticles.26 Thus, in all experiments performed in this study, an 830 nm diode laser was used. The laser power was in the range 0.2-6 mW, and the exposure time was 10 s. A 50× objective was used. The wavelength of the instrument was automatically calibrated using an internal silicon wafer and the spectrum was centered at 520 cm-1. Scanning Electron Microscope. The prepared bacterial samples were spotted and dried on SEM specimen stub. A Karl Zeiss EVO 40 model SEM instrument was used. The accelerating voltage was 10 kV.

Results and Discussion In our previous study, we demonstrated the change of spectral features in SERS spectra of E. coli as the colloidal suspension concentration increased.26 During the course of the present study, we noticed that a number of parameters such as method of mixing bacteria sample and colloidal suspension, using a vortex or simply stirring with a pipet tip, waiting time after mixing, and centrifugation time during the concentration process of the colloidal suspension could influence the SERS spectra of E. coli. Mixing vigorously and waiting longer before SERS acquisition after mixing seem to influence the SERS spectra due to possible penetration of smaller silver nanoparticles into the bacterial cell wall structure and perhaps increased aggregation tendency of nanoparticles. The centrifugation is used to concentrate colloidal suspension, and the centrifugation time might have an impact on the size distribution on the concentrated silver nanoparticles. As the centrifugation time is increased, the possibility of transferring a greater number of smaller silver nanoparticles (