Abstract
Organically modified silicate (ORMOSIL) nanoparticles (NPs) doped with rhodamine 6G and rhodamine B (RB) dyes were synthesized by Stöber method from methyltriethoxysilane CH3Si(OCH3)3 precursor (MTEOS). The NPs are surface functionalized by cationic amino groups. The optical characterization of dye-doped ORMOSIL NPs was studied in comparison with that of free dye in solution. The synthesized NPs were used for labeling bacteria E. coli O157:H7. The number of bacteria have been counted using the fluorescent spectra and microscope images of labeled bacteria. The results show the ability of NPs to work as biomarkers.
Export citation and abstract BibTeX RIS
Content from this work may be used under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
1. Introduction
Organically modified silicates (ORMOSILs) or silsesquioxanes are organic–inorganic hybrid materials in which the organic components are bonded to a siloxane or silica backbone [1]. In recent years, ORMOSIL nanoparticles (NPs) doped with organic dyes have been widely used in many applications such as gene delivery, photodynamic therapy and other photonic areas [2–12]. Fluorescent dye-doped NPs (DDNPs) represent an important class of nanomaterials for optical bioimaging [13–16]. The most prominent materials among them are silica and ORMOSIL, which have several favorable properties such as optical transparency, non-antigenicity and rich surface chemistry for facile bioconjugation [15, 16]. The silica/ORMOSIL host matrix protects the contained dye molecules from photobleaching and prevents their interaction with the biological environment. Recently, ORMOSIL NPs conjugated with a near-infrared (NIR) fluorophore DY776 and radiolabeled with iodine−124 were fabricated [17]. The biodistribution demonstrates the use of these NIR dye and iodine−124 conjugated ORMOSIL NPs as promising probes for safe in vivo bioimaging. These NPs facilitate optical bioimaging in the NIR window, with maximum tissue penetration of light and minimum background signal [18, 19]. The in vivo studies in drosophila indicate that these novel silica-based NPs are biocompatible and not toxic to whole organisms, and have potential for the development of in vivo and long-term applications [20].
Although many studies on photophysical properties of dye molecules encapsulated in silica-based NPs have been done for pure silica NPs [21–24], there are few works for ORMOSILs. Herein, we report the results of ORMOSIL NP synthesis by modified Stöber method from trimethoxysilane CH3Si(OCH3)3 (MTEOS) precursor, both empty and doped with fluorescent dyes, in the non-polar core of dioctyl sodium sulfosuccinate (aerosol-OT)/DMSO/water microemulsions. The amino-functionalized NPs were also synthesized by a synchronous hydrolysis of MTEOS and 3-aminopropyltriethoxysilane (APTEOS). By varying the concentrations of aerosol-OT, monodispersed NPs of various sizes (between 20 and 100 nm) have been synthesized. The photophysical properties of incorpored dye molecules were studied and compared with those of free dyes in solutions. The synthesized NPs were used for labeling bacteria E. coli O157:H7.
2. Experimental
2.1. Synthesis
2.1.1. Materials.
Methyltrimethoxysilane (MTEOS), aminopropyltriethoxysilane (APTEOS), dimethylsulfoxide (DMSO), clorotrimethylsilane (CTMES), aqueous ammonia solution 25%, co-surfactant butanol-1 were purchased from Merck. Rhodamine 6G (Rh6G) and rhodamine B (RB) dyes were obtained from Exiton. Surfactant aerosol-OT (AOT) (96%) was purchased from Fluka. Dialysis tubing with molecular weight cut-off (MWCO) of 10000 D was purchased from Sigma-Aldrich.
2.1.2. Synthesis.
The NPs, both void and dye-doped, were synthesized by modified Stöber method as in previous work [25]. The micelles were prepared by dissolving a fixed amount of aerosol-OT and 2-butanone in 20 ml of double-distilled water by vigorous magnetic stirring. A 100 μl sample of the dye in DMSO (10 mM) was dissolved by magnetic stirring in the resulting clear solution. For void NPs, 100 μl of DMSO without the dye was added. After that, 200 μl of neat MTEOS was added to the micellar system, and the resulting solution was stirred for 30 min. Next, ORMOSIL NPs were precipitated by adding 40 μl APTEOS or aqueous ammonia solution with continous stirring. For NPs without amino group on the surface, another catalyst was replacing APTEOS. After 1 h of reaction, a 20 μl of clorotrimethylsilane was added to quench the remaining silanol groups on the surface of NPs. Next, this system was stirred for 20 h at room temperature. After the formation of the NPs, the solution was dialyzed in a 10000 D MWCO dialysis tubing against water for a week in order to remove the remaining chemical agents and all the surfactant AOT and butanol-1. The dialyzed solution then was filtered through a 0.2 μm cutoff membrane filter and kept in the dark at 4 °C; NPs are stable for several months.
2.1.3. Apparatus.
Transmission and scanning electron microscopes (TEM, JEM 1011 and SEM, Hitachi S-480) were used to determine the shape, size and surface of particles. The chemical structure of NPs was studied using a micro-Raman spectrograph LAMBRAM-1 with a laser 632.8 nm He–Ne as excitation source at room temperature and Impact 410 Nicolet Fourier transform infrared (FTIR) spectrophotometer. Absorption spectra were measured using JASCO-V570-UV–Vis–NIR spectrometer. The fluorescence spectra were recorded on a Cary Eclipse spectrofluorometer (Varian).
The two-photon excitation fluorescence lifetime was detected using a Ti-sapphire laser (Mai Tai, Spectra Physics, 900 nm, 80 fs) with a repetition rate reduced to 8 MHz. A time-correlated single-photon counting SPC-430 card (Becker-Hickl GmbH) was used for the acquisition. The monochromator (Jobin-Yvon H10), microchannel plate photomultiplier (Hamamatsu R1564U-06) and an amplifier (Phillips Scientific 6954) were used as the detection system. The microcell (volume 50 μl) was thermostated with a Haake type-F3 circulating bath T =20 °C.
2.2. Cell labeling
In order to test the biomarker ability of prepared DDNPs, a specific E. coli O157:H7 bacteria label was carried out. The E. coli O157:H7 bacteria were provided from Institute of Biotechnology (VAST). The direct conjugation of DDNPs to anti E. coli O157:H7 antibodies (Abs) through amine-carboxylic acid coupling was used with N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide hydrochloride (EDC) as catalyzer.
- Mix DDNPs (5 μl, 4.77 mg ml−1) with antibodies (80 μl, 106–8 colony-forming unit (CFU) in 2-(N-morpholino) ethanesulfonic acid (MES) buffer, add 1200 μl EDC and then shake the mix at 200 rpm for 3 h maintaining a 30 °C.
- Separate DDNP–Ab conjugates from excess free antibodies and DDNPs via centrifugation of 35 000 rpm at 4 °C temperature (twice). The concentrated DDNP–Ab complex was rediluted in 100 μl phosphate buffered saline (PBS) buffer.
- Mix DDNP–Ab complex (50 μl) with bacteria (5 μl, 104 CFU ml−1) while shaking for 10 min at room temperature. The sample was incubated for 30 min at 37 °C and then centrifuged at 12 000 rpm to remove the excess complex NP–Ab that did not bind to the bacteria. The pellet was diluted into 100 μl of 0.1 M PBS buffer, then was washed again to remove all unbound to bacteria complex. At the end, 100 μl of PBS buffer was added to the sample. The control sample (without NPs) was also carried out by the above route.
2.2.1. Cell imaging.
The labeled cells-bacteria were imaged by optical, transmission and scanning electron microscopy. The cell-NP suspensions were imaged on a Nikon Ti-E inverted microscope, objective 60 × NA1.4. Confocal images were taken by depositing 10 μl of the suspension onto a thin glass slide placed above an oil immersion (CFI plan Apo VC 60 × NA1.4) objective on the confocal microscope. Fluorescence imaging was conducted on a confocal microscope setup consisting of a Nikon Ti-E inverted microscope with a Nikon C1plus confocal scanning system and three lasers, with three separate photomultiplier tubes for detection. The fluorescent NPs were excited with the 543 nm line of the green He–Ne (Melles Griot) laser, and emission was detected using a HQ590/50 nm bandpass filter. Pixel format was 1024 × 1024.
2.2.2. Cell detection.
The number of bacteria in the obtained sample after above incubation procedure was verified by plate counting technique, a golden standard method in bacterium counting in microbiology and cell biology [26]. The obtained sample was divided into two parts. The first half of the sample was grown on agar plate for 16–18 h in a 37 °C incubator to obtain an accurate number of bacteria by CFU counts. The second half of the sample was used for bacterial cell determination by using the spectrofluorometer.
The samples with different bacteria concentration from 1 × 104 CFU to 1 × 102 CFU were prepared with a step of 25% concentration less than that of the previous sample: 1 × 104, 7.5 × 103, 5 × 103, 2.5 × 103 CFU and so on. The fluorescence intensity in each sample was detected with 532-nm excitation using a Cary Eclipse Spectrofluorometer. Control samples were those of the same bacteria concentrations but without NPs. The fluorescence curves of the controls were considered the background.
3. Results and discussion
3.1. Size, shape and chemical structure
The TEM image of DDNPs is shown in figure 1(a). It shows that the particle shape is spherical with the average diameter of about 40 nm with high monodispersion.
Figure 1. (a) TEM image of DDNPs. (b) Raman spectra of void NPs; curve 1: spectrum of NPs prepared without APTEOS, curve 2: spectrum of NP prepared with APTEOS. (c) FTIR spectra of NPs; upper curve: spectrum of NPs prepared without APTEOS, lower curve: spectrum of NPs prepared with APTEOS.
Download figure:
Standard imageThe chemical structure of NPs is determined by analyzing the micro-Raman and Fourier transform infrared (FTIR) spectra of both void ORMOSIL NPs prepared from MTEOS precursor with and without APTEOS catalyzer. As is shown in figures 1(b) and (c), the Raman and FTIR spectra of NPs prepared without APTEOS are composed from two principal groups: the vibration bands of SiO2 network and that of methyl group bound to silicon atom (Si-CH3). The Raman and FTIR spectra of NPs made with APTEOS are mainly similar to those of non-APTEOS but in addition there are the bands of vibration of amino groups NH2. So, it is clear that the APTEOS catalyzer forms the amino group bound to silicon atom on the surface of NPs. This amino group will play the role of a biocompatible agent.
3.2. Photophysical properties
3.2.1. Fluorescence spectra and lifetime.
In order to compare the optical properties of the dye doped to NPs and bare dye, the Rh6G and RB dyes were diluted in water with 0.2% DMSO (v/v) such that its intensity of absorption is the same as in NPs. The absorption and fluorescence spectra of RB and Rh6G in water and in NPs are presented in figure 2. From this figure it is seen that the absorption and fluorescence spectra of Rh6G and RB in water and in NPs are similar but there is a little red-shift (∼ 5 nm) of the spectral maxima of dyes in NPs in comparison with those of dyes in water. This means that the interactions between dye molecules and host matrix are weak.
Figure 2. Absorption and fluorescence spectra of Rh6G (a) and RB (b).
Download figure:
Standard imageFrom figure 3, the lifetime was calculated as 2.3, 2.7 and 3.5 ns for NPs with size 20, 40 and 50 nm, and as 1.5 ns for free dyes in ethanol. So it is clear that the fluorescence efficiency of RB dye in the ORMOSIL NPs host is higher than that of free dye in ethanol. This phenomenon can be explained as follows: because the dye molecules are located in the pores of ORMOSIL porous matrix, so their monodispersion is improved in comparison with that of free dyes in ethanol whose emission efficiency is reduced due to the collision between dye molecules. The dependence of the NPs' lifetime on their size will be discussed in other work.
Figure 3. Fluorescence decays under two photon excitation (Ti: sapphire laser at 900 nm, 80 fs) at room temperature of the RB doped ORMOSIL NPs of different sizes: (1) 20 nm, (2) 40 nm, (3) 50 nm and (4) free RB molecules in ethanol.
Download figure:
Standard image3.2.2. Photostability.
Figure 4 shows the fluorescence intensity versus time curves of Rh6G and RB dye molecules in water and in NPs upon a He–Ne laser irradiation at 543 nm and 3.2 mW cm−2. The fluorescence intensity of dyes in water was down to half after about 90 min of irradiation while that of dyes in NPs remains unchanged after 140 min of lighting up.
Figure 4. Emission intensity of Rh6G (a) and RB (b) in water and in NPs versus time. Excitation by He–Ne laser at 543 nm, 3.2 mW cm−2.
Download figure:
Standard image3.2.3. Effect of surfactant concentration.
Table 1 shows the effect of surfactant concentration on particle size. As the results show, the particle size increases from 20 nm (sample 2SB20) to 90 nm (sample 6SB20) proportionally to the surfactant quantity. At low surfactant values, the microemulsion droplet size is smaller, the number of MTEOS molecules in the droplet is fewer. Hence, there is a smaller number of formed monomers and nuclei, so the final particle size is smaller. When the surfactant value increases, the microemultion droplet size increases, the MTEOS molecule number in the droplet increases, more monomers and nuclei are formed, and the resultant particle size is larger [27].
Table 1. Characterization of RB dye-doped ORMOSIL NPs with different quantities of surfactant agents.
| Sample | AOT (g) | Bu-2 (μl) | Size (nm) | Concentration of NPs (particles m−1) | Concentration of RB dye in solution (10–5 mol l−1) | Concentration of RB dye in each NP (10–2 mol l−1) | Number of dye molecules per particle | Brightnessa |
|---|---|---|---|---|---|---|---|---|
| 2SB20 | 0.22 | 400 | 20 ± 5 | 4.60×1014 | 3.09 | 1.59 | 40 | 35 |
| 4SB20 | 0.44 | 800 | 40 ± 5 | 5.20×1013 | 2.43 | 1.39 | 281 | 316 |
| 5.3SB20 | 0.583 | 1060 | 70 ± 5 | 7.84×1012 | 2.23 | 1.58 | 1710 | 2270 |
| 6SB20 | 0.66 | 1200 | 90 ± 10 | 3.43×1012 | 2.24 | 1.72 | 3940 | 5600 |
| RB/ethanol | – | – | 0.5 | – | 1.67 | – | – | 1 |
aThe brightness of one dye molecule is considered as the unit.
The particle concentration, dye concentration and number of dye molecules in each NP were estimated for each sample. The number of dye molecules is ∼40 in 20 nm-particles and ∼3940 in 90 nm-particles. So, it is clear that each NP contains from a few tens to thousands of dye molecules depending on its size. The dye concentration in the whole solution is ∼10−5 M l−1 (figure 5(a)), but in each particle this parameter increases to ∼10−2 M l−1, a very high concentration. The forms of absorption spectra of dyes in NP solution are the same as those of free RB dyes at 1.67 × 10−5 M l−1 concentration, but with a little shift due to the interactions of dyes with the solid host. There are no effects of dimerization of dyes in NPs, even at ∼ 10−2 M l−1 concentration (figure 5(b)). At this concentration the fluorescence of RB molecules in ethanol is totally quenched due to the collision (data not shown), but there is no quenching effect in the fluorescence spectra of NP solutions (figure 5(c)). These results can be explained as follows: the dye molecules in NPs are located in the pores of silica porous matrix, so they are monodispersive and therefore there is no collision between them, hence no quenching in their fluorescence.
Figure 5. Absorption and fluorescence spectra of RB molecules in ethanol and in NPs.
Download figure:
Standard imageConsidering the fluorescence intensity of one free RB molecule in solution as the unit, we can estimate the brightness of each NP from its fluorescence spectra. The results show that the brightness of NPs is much higher than that of free dye molecules (table 1), depending on their size. Approximately 316 times brighter fluorescence was observed from the 40-nm NPs when compared to that obtained from the aqueous dye solution of the same concentration. The fluorescence of one 90-nm ORMOSIL particle is 5600 times brighter than that of one free dye molecule. We can see in table 1 that the brightness of NPs is proportional with the dye number contained in particles, but the brightness of each particle is higher than the dye number inside. As was explained in section 3.2.1, the lifetime of dye molecules in NPs is longer than that in ethanol, e.g. the fluorescence efficiency of dyes in NPs is improved, so their brightness is improved compared with that of free dyes in ethanol.
3.3. Cell labeling
3.3.1. Cell imaging.
Figure 5 presents the transmission and fluorescence images of E. coli O157:H7 bacteria after incubation with the DDNP–Ab complex (figures 5(a) and (b)). It is clear from figure 5 that the bacterial cells are labeled with NPs. The bacteria incubated with specific E. coli O157:H7 antibody do not fluoresce (figure 5(c)).
There are about 300 RB dye molecules encapsulated within each particle (data not shown); high signal amplification was achieved when the antibody-conjugated NPs were bound to antigens on the surface of the bacteria (figures 6(d) and (e)). The scanning electron microscope image of the E. coli O157:H7 cell after incubation with the DDNP–Ab complex shows that there were thousands of antibody-conjugated NPs bound to a single bacterium, providing significant fluorescent signal amplification as compared with a single dye molecule. The NP-based signal amplification can be easily seen in a fluorescent image, as shown in figure 6(e) inset. Due to the bound NPs, the dimension of bacterium is a size up after incubation (figures 6(d) and (f)).
Figure 6. Images of E. coli O157:H7 bacterial cells. Tranmission (a) and fluorescence (b) microscope images of cells after incubated with antibody-conjugated NPs and fluorescence microscope image of cells before they are incubated with antibody-conjugated NPs (c) (the size of images is 46 μm × 46 μm). (d) SEM image of cell incubated with DDNP–Ab complex. (e) Fluorescence confocal microscope image of bacterial cell after incubation with DDNP–Ab complex. The NP-based fluorescence signal amplification can be easily seen in a fluorescent image ((e) inset). The bacterial size after (d) is much larger than that before (f) incubation, due to the bound NPs.
Download figure:
Standard image3.3.2. Cell detection.
Figure 7 presents the fluorescence curves of samples with different bacterial concentrations. As is shown in figure 6, the fluorescence intensities of samples are proportional to the numbers of bacteria in the sample. So we can use the curve of fluorescence intensity versus bacterial number as a calibration curve to determine the bacterial number in the sample under analysis [23]. The total time needed for this quantitative analysis is about 3–4 h.
Figure 7. (a) The fluorescence spectra of different bacterial concentration samples; (b) The fluorescence intensity versus the bacterial number (red curve), fit line (blue line).
Download figure:
Standard imageThe fluorescence based bacterial detection results (7.8 × 104 CFU) were compared with those of the plate counting method (8.1 × 104 CFU). The results show the high correlation between the two methods (figure 8), confirming the ability of spectrofluorometer method for quantitative bacterial detection. In this experiment, the detected number of bacteria is limited at 102 CFU, or the limit of spectrofluorometer method is 102 CFU. The limit of this method will be improved by using NPs with larger size, which contain thousands of dye molecules in each particle, so this method will be able to detect a single bacterium [23].
Figure 8. Comparison of bacteria detection with the plate-counting method versus the spectrofluorometer method with antibody-conjugated NPs.
Download figure:
Standard image4. Conclusion
Aqueous Rh6G and RB doped ORMOSIL NPs have been synthesized by modified Stöber method. The NPs are uniform spheres of diameter in the range 20–90 nm, mono-dispersed, with the amino NH2 group on the surface. The photophysical properties of the dyes inside NPs such as photostability, brightness and lifetime are improved in comparison with those of bare dyes in water. Especially, the fluorescence of one 90-nm ORMOSIL particle is 5600 times brighter than that of one free dye molecule. The dye-doped silica based NPs were used for immunolabeling the bacteria E. coli O157:H7. The results show that due to the brightness of NPs, the fluorescence microscope signal of bacteria is amplified. A spectrofluorometer method for rapid quantitative detection of bacteria was proposed. The dye-doped ORMOSIL NPs show their ability to work as biomarkers.
Acknowledgment
This work was supported by Vietnam National Foundation for Science and Technology Development (NAFOSTED) No 103.06.101.09.