12228
Langmuir 2008, 24, 12228-12234
Size-Controlled, One-Pot Synthesis, Characterization, and Biological Applications of Epoxy-Organosilica Particles Possessing Positive Zeta Potential Michihiro Nakamura* and Kazunori Ishimura Department of Anatomy and Cell Biology, Medical Informatics, Institute of Health Biosciences, The UniVersity of Tokushima Graduate School, 3-18-15 Kuramoto, Tokushima 770-8503, Japan ReceiVed June 21, 2008. ReVised Manuscript ReceiVed August 22, 2008 Epoxy-organosilica particles made from 2-(3,4-epoxycyclohexyl)ethyltrimethoxysilane (EpoMS) as a single silica source were synthesized by means of a one-pot method. We evaluated three sets of synthesis conditions, including traditional Sto¨ber conditions and two variations. Although the traditional conditions did not afford EpoMS particles, the variations did. The size distributions of the particles were evaluated by means of transmission electron microscopy. The mean diameters and size distributions of the particles depended on the EpoMS concentration, and the best coefficient of variation for the size distribution was 5.9%. The surface of the particles had unique properties, such as a positive zeta potential. The particles bound strongly to proteins as well as to DNA. The particles made from EpoMS, allowing particles internally functionalized with fluorescent dye to be prepared by means of a one-pot synthesis. EpoMS particles doped and tuned with fluorescent dye showed strong fluorescence signals and distinct peaks on flow cytometry, and the fluorescent particles could be used to label cells. The labeled cells showed clear fluorescence under a fluorescence microscope, and electron microscopy showed many particles in the cytoplasm. This is the first report describing the synthesis of epoxy-organosilica particles with a positive zeta potential and describing differences in the characteristics of particle formations due to changes in synthesis conditions. We also discuss the advantages of EpoMS particles, as well as the potential biological applications of these particles.
Introduction Particles are increasingly being used for biomedical applications including bioanalysis, multitarget detection systems, in vitro and in vivo imaging, and nanomedicine.1-14 Multifunctionalized particles are also being developed. Multifunctionalized particles for biomedical applications must have strong and clear signals, bind targets with high affinity and specificity, and be adaptable for use in systems for controlled release of drugs or for the exhibition of a specific pharmacological effect. To achieve these ends, silica particles are becoming important as core particles because they are easy to prepare and separate, their surfaces can be modified or labeled, they are highly hydrophilic, and they are safe. * To whom correspondence should be addressed. Phone: +81-88-6339220. Fax: +81-88-633-9426. E-mail:
[email protected]. (1) Gao, X.; Cui, Y.; Levenson, R. M.; Chung, L. W. K.; Nie, S. Nat. Biotechnol. 2004, 22, 969–976. (2) Roy, I.; Ohulchanksy, T. Y.; Bharali, D. J.; Pudavar, H. E.; Mistretta, R. A.; Kaur, N.; Prasad, P. N. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 279–284. (3) Wang, L.; Yang, C. Y.; Tan, W. H. Nano Lett. 2005, 5, 37–43. (4) Wang, L.; Tan, W. H. Nano Lett. 2006, 6, 84–88. (5) Zhao, X. J.; Hilliard, L. R.; Mechery, S. J.; Wang, Y.; Bagwe, R.; Jin, S.; Tan, W. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 15027–15032. (6) Santra, S.; Zhang, P.; Wang, K.; Tapec, R.; Tan, W. Anal. Chem. 2001, 73, 4988–4993. (7) Zhao, X.; Dytioco, R.; Tan, W. J. Am. Chem. Soc. 2003, 125, 11474– 11475. (8) Corot, C.; Robert, P.; Ide´e, J. M.; Port, M. AdV. Drug DeliVery ReV. 2006, 58, 1471–1504. (9) He, X. X.; Wang, K. M.; Tan, W. H.; Liu, B.; Lin, X.; He, C. M.; Li, D.; Huang, S. S.; Li, J. J. Am. Chem. Soc. 2003, 125, 7168–7169. (10) Hu, M.; Chen, J.; Li, Z. Y.; Au, L.; Hartland, G. V.; Li, X.; Marquez, M.; Xia, Y. Chem. Soc. ReV 2006, 35, 1084–1094. (11) Santra, S.; Bagwe, R. P.; Dutta, D.; Stanley, J. T.; Walters, G. A.; Tan, W. H.; Moudgil, B. J.; Mericle, R. A. AdV. Mater. 2005, 17, 2165–2169. (12) Tan, W.; Wang, K.; He, X.; Zhao, J.; Drake, T.; Wang, L.; Bagwe, R. P. Med. Res. ReV. 2004, 24, 621–638. (13) Wang, L.; Wang, K. M.; Santra, S.; Zhao, X. J.; Hilliard, L. R.; Smith, J. E.; Wu, J. R.; Tan, W. H. Anal. Chem. 2006, 78, 646–654. (14) Liu, Y.; Miyoshi, H.; Nakamura, M. Int. J. Cancer 2007, 120(12), 2527– 2537.
The ability to functionalize both the interior and the surface of silica particles is necessary for various applications. The internal functionalization of silica particles with signal molecules, such as fluorescent dyes, has been thoroughly investigated.15-26 Fluorescent silica particles possess several advantages over other fluorescent particles such as quantum dots and latex beads, including high fluorescence intensity, good photostability due to the exclusion of oxygen by silica encapsulation, good potential for surface modification with various biomolecules, and low toxicity. Surface functionalization of silica particles with various biomolecules such as proteins, enzymes, peptides, and DNA is an important technology for various applications including highthroughput assays, imaging systems, and drug-delivery systems. The type and extent of surface functionalization are largely determined by the surface properties of the as-prepared particles. Traditionally, silica particles have been prepared from inorganic silicates such as tetraethoxysilicate (TEOS).27,28 Various methods for modifying the surface of TEOS particles with organic (15) Vanzo, E. U.S. Patent 4,077,804, 1978. (16) Peterson, J. I. U.S. Patent 4,194,877, 1980. (17) Yabuuchi, N.; Otsuka, C.; Kashihara, A. U.S. Patent 5,367,039, 1994. (18) Van Blaaderen, A.; Vrij, A. Langmuir 1992, 8, 2921–2931. (19) Verhaegh, N. A. M.; van Blaaderen, A. Langmuir 1994, 10, 1427–1438. (20) Imhof, A.; Megens, M.; Engelberts, J. J.; de Lang, D. T. N.; Sprik, R.; Vos, W. L. J. Phys. Chem. B 1999, 103, 1408–1415. (21) Santra, S.; Zhang, P.; Wang, K.; Tapec, R.; Tan, W. Anal. Chem. 2001, 73, 4988–4993. (22) Zhao, X.; Bagwe, R. P.; Tan, W. AdV. Mater. 2004, 16, 173–176. (23) Lian, W.; Litherland, S. A.; Badrane, H.; Tan, W.; Wu, D.; Baker, H. V.; Gulig, P. A.; Lim, D. V.; Jin, S. Anal. Biochem. 2004, 334, 135–144. (24) Rossi, L. M.; Shi, L.; Quina, F. H.; Rosenzweig, Z. Langmuir 2005, 21(10), 4277–4280. (25) Ow, H.; Larson, D. R.; Srivastava, M.; Baird, B. A.; Webb, W. W.; Wiesner, U. Nano Lett. 2005, 5, 113–117. (26) Nakamura, M.; Shono, M.; Ishimura, K. Anal. Chem. 2007, 79, 6507– 6514. (27) Sto¨ber, W.; Fink, A.; Bohn, E. J. Colloid Interface Sci. 1968, 26, 62–69. (28) Yanagi, M.; Asano, Y.; Kandori, K.; Kon-no, K. Abstracts of the 39th Symposium of the DiVision of Colloid Interface Chemistry, Chemical Society of Japan, Tokyo, 1986, 386.
10.1021/la801950q CCC: $40.75 2008 American Chemical Society Published on Web 09/30/2008
Epoxy-Organosilica Particles
Figure 1. Molecular structures of tetraethoxysilicate (TEOS), 3-mercaptopropyltrimethoxysilane (MPMS), and 2-(3,4-epoxycyclohexyl)ethyltrimethoxysilane (EpoMS).
functional groups using organosilicates containing a thiol or an amino group are also being developed for use in various applications.29-31 Recently, we prepared thiol-organosilica particles from a single thiol-organosilicate by means of a one-pot synthesis.32,33 The particles bear surface and internal thiol groups, which allow for surface functionalization through adsorption, thiol-exchange reactions, and maleimide coupling. In addition to surface functional groups, surface charge is also an important factor for the surface modification and for particle function. The surface charge can affect adsorption with biomolecules such as DNA and proteins and can also affect the particles’ in vivo metabolism, distribution, and internalization into and distribution in cells.35,36 Silica particles prepared from TEOS or thiolorganosilicates have a negative surface charge, and silica particles with a positive surface charge made from a single silica source have not yet been reported. In this study, we prepared novel epoxy-organosilica particles with a positive surface charge by means of a one-pot synthesis using 2-(3,4-epoxycyclohexyl)ethyltrimethoxysilane (EpoMS; Figure 1) as a single silica source. We evaluated three sets of synthesis conditions, including the traditional Sto¨ber conditions,27 to prepare the particles. The particles were internally functionalized with fluorescent dye by means of a one-pot synthesis, and the surface of the particles could be modified not only with proteins but also with DNA. We also demonstrate the promise of EpoMS particles for biological applications.
Experimental Section Materials. 2-(3,4-Epoxycyclohexyl)ethyltrimethoxysilane (EpoMS), 3-mercaptopropyltrimethoxysilane (MPMS), tetraethoxysilane (TEOS), rhodamine B, tris(dichlororuthenium)(II) hexahydrate (Ru(bpy)), and bovine serum albumin (BSA) were purchased from SigmaAldrich Chemical Co. (St. Louis, MO). Fluorescein-labeled oligonucleotide (fluorescein-5′-GGCAGATCGTCAGTCAGTCAC-3′) was purchased from Invitrogen (Carlsbad, CA). Green fluorescent protein (GFP) was purchased from Upstate (Lake Placid, NY). Fetal bovine serum was perchased from HANA-NESCO BIO corp. (Osaka, Japan). Ethyl alcohol and 30% NH4OH were purchased from Wako Fine Chemicals Inc. (Osaka, Japan). One-Pot Synthesis of Organosilica Particles and Internally Functionalized Fluorescent Organosilica Particles. Particles were prepared using three synthesis conditions, referred to herein as A, B, and C, as in a previous report.33 Solutions of EpoMS were prepared in 2-propanol at concentrations of 12,5, 25, 50, 100, and 200 mM. Under condition A (traditional Sto¨ber conditions), 25.1 µL of EpoMS solution was mixed with 325 µL of ethyl alcohol and 36 µL of 28% NH4OH, and the final volume was brought to 500 µL with distilled water. Under condition B, 25.1 µL of EpoMS solution was mixed with (29) Qhobosheane, M.; Santra, S.; Zhang, P.; Tan, W. Analyst 2001, 126, 1274–1278. (30) Hilliard, L.; Zhao, X.; Tan, W. Anal. Chim. Acta 2002, 470, 51–56. (31) Bagwe, R. P.; Hilliard, L. R.; Tan, W. Langmuir 2006, 22, 4357–4362. (32) Nakamura, M.; Ishimura, K. J. Phys. Chem. C 2007, 111, 18892–18898. (33) Nakamura, M.; Ishimura, K. Langmuir 2008, 24(9), 5099–5108. (35) Harush-Frenkel, O.; Debotton, N.; Benita, S.; Altschuler, Y. Biochem. Biophys. Res. Commun. 2007, 353, 26–32. (36) Panyam, J.; Zhou, W. Z.; Prabha, S.; Sahoo, S. K.; Labhasetwar, V. FASEB J. 2002, 16, 1217–1226.
Langmuir, Vol. 24, No. 21, 2008 12229 36 µL of 28% NH4OH, and the final volume was brought to 500 µL with distilled water. Under condition C, 25.1 µL of EpoMS was diluted to a volume of 500 µL with 28% NH4OH. All the synthesis solutions were mixed well and then incubated at 25 °C for 1 or 3 days. Internally functionalized fluorescent EpoMS particles were also prepared by means of a one-pot method. Various concentrations of Rhodamine B or Ru(bpy) were mixed with a condition B solution containing 12.5 mM or 200 mM EpoMS, and the reaction mixture was then incubated for 3 days. After incubation, the mixture was centrifuged to remove unbound reagents. The particles were washed extensively with 70% ethyl alcohol and water. Electron Microscopy of Silica Particles. The particles were fixed on a 400-mesh copper grid coated with nitrocellulose, and transmission electron microscopy images were obtained with a Hitachi H500 or H7650 electron microscope (Hitachi, Tokyo, Japan) or a JEOL JEM-1200EXII electron microscope (JEOL Ltd., Tokyo, Japan). Preparation of Surface-Modified Organosilica Particles with Proteins or DNA. To modify the surface of the silica particles with biomolecules, 1 µL of EpoMS particles made using condition B (200 mM EpoMS) or MPMS particles were reacted with 1 µL of 250 µg/ml GFP in phosphate-buffered saline pH 7.4 (PBS) or with 1 µL of 2 µM fluorescein-labeled oligonucleotide in PBS. Thirty microliters of EpoMS particles made using condition B were reacted with 30 µL of various concentrations of BSA in water at room temperature and diluted with 2940 µL of water for zeta-potential analysis. Flow Cytometry. Flow cytometry analysis was performed on FACSCalibur flow cytometers (Becton Dickinson, San Jose, CA) with 488- and 635-nm excitation lasers. Fluorescences were detected on the FL1 channel (530/30-nm band-pass filter), on the FL2 channel (585/42-nm band-pass filter), and on the FL3 channel (670 nm longpass filter). Surface-modified particles as described above and the EpoMS particles doped and tuned with fluorescent dye were diluted and analyzed. All data were obtained without compensation. Zeta-Potential Analysis. The zeta potentials of particles were determined with a NICOMP Submicron particle sizer, model 380/ ZLS (Nicomp Particle Sizing Systems, Santa Barbara, CA) at room temperature. Electrodes were dipped directly in the solution containing freshly prepared silica particles or EpoMS particles surface-modified with BSA. Based on the principles of electrophoretic light scattering, quantitative measures of the charge on colloidal particles in liquid suspension were performed. Fluorescence and Light Microscopic Analysis of Fluorescent Silica Particles. Aliquots of the particle solutions were placed on glass slides. We analyzed the particles attached to the glass slides by means of a fluorescence and light microscopy system consisting of an inverted fluorescence microscope (TE 2000, Nikon, Kanagawa, Japan) equipped with a 100-W mercury lamp as a light source and a CCD camera (Digital Sight DS-L1, Nikon, Kanagawa, Japan) or a CCD camera (Rolera-XR Mono Fast 1394 Cooled, Qimaging, Burnaby, BC, Canada) with continuous excitation with Image-Pro Plus software (MediaCybernetics, Silver Spring, MD). Preparation and Transmission Electron Microscopy of Mice Peritoneal Macrophages Containing Fluorescent Silica Nanoparticles. Male C57 BL/6J mice were injected intraperitoneally with 0.1 mL of a solution containing 10 mg/mL of fluorescent EpoMS particles containing rhodamine B. After 1 day, mice were sacrificed, and the peritoneal cells were harvested and fixed. A portion of the cells was used for fluorescence and light microscopy. Mice peritoneal cells were fixed with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) for 4 h and then rinsed sequentially with PBS containing 10, 15, and 20% sucrose for 4 h. The cells were washed, treated with 2% OsO4 and uranyl acetate, dehydrated in a graded series of ethanol solutions, and embedded in Epon epoxy resin. Ultrathin sections (thickness, 80 nm) were cut with a Reichert Ultracut E ultramicrotome (Leica Microsystems, Wetzlar, Germany) and examined with a Hitachi H500 or H7650 electron microscope (Hitachi, Tokyo, Japan).
12230 Langmuir, Vol. 24, No. 21, 2008
Nakamura and Ishimura
Figure 2. TEM images of epoxy-organosilica particles prepared from EpoMS as a function of EpoMS concentration. EpoMS particles prepared under condition A (A), condition B (B), and condition C (C) were observed after incubation periods of 1 day. Scale bars: 1000 nm. Table 1. Formation of EpoMS Particles under Conditions A, B, and C with Incubation Periods of 1 Daya concn (mM)
condition A
condition B
condition C
200 100 50 25 12.5 6.25
× × × × × ×
O O O O O ×
O O O O O ×
a Formations and incomplete formations of EpoMS particles are indicated as O and ×, respectively.
Results and Discussion 1. Synthesis of EpoMS Particles. Varying concentrations of EpoMS were used in three kinds of synthetic conditions to prepare particles from a single silicate source. The first condition, condition A, followed traditional Sto¨ber methods27 (65% ethanol, 33% water, and 2% ammonium hydroxide). The second condition, condition B, contained 98% water and 2% ammonium hydroxide but did not contain ethanol. The third condition, condition C, contained a higher concentration of ammonium hydroxide (73% water and 27% ammonium hydroxide). Conditions B and C were varied only slightly from the Sto¨ber method in efforts to determine the minimum solution components necessary to form EpoMS particles. The synthesized particles were evaluated by transmission electron microscopy (TEM), as shown in Figure 2. Particles were labeled as being completely or incompletely formed by analyzing these TEM images, as summarized in Table 1. The formation of a given batch of particles was deemed complete if unaggregated particles of >10 nm diameter with clearly visible edges were observed in TEM images.
As shown in Figure 2, no EpoMS particles formed under condition A at any EpoMS concentration, even after incubation for 3 days. Under conditions B and C, particles were clearly observed after 1 day at all EpoMS concentrations except the lowest (6.25 mM); even after 3 days, no particles were observed at the lowest concentration. These results differed from the results obtained in a previous study33 of particle formation from TEOS and three thiol-organosilica sources, 3-mercaptopropyltrimethoxysilane (MPMS), 3-mercaptopropyltriethoxysilane (MPES), and 3-mercaptopropylmethyldimethoxysilane (MPDMS), under similar sets of conditions. Under condition A, particles formed at all tested concentrations of TEOS and nearly all tested concentrations of MPMS and MPES, whereas particles did not form from MPDMS, even after 3 days. Under condition B, TEOS solutions did not form particles, even after 3 days. High concentrations of MPMS and MPES formed particles, but low concentrations of MPMS and MPES did not after 1 day. All concentrations of MPMS and MPES formed particles after 3 day, but no MPDMS particles formed, even after 3 days. Under condition C, high concentrations of TEOS formed particles. Low concentrations of MPMS, MPES, and MPDMS formed particles, but high concentrations of MPMS, MPES, and MPDMS did not. In the current study, no EpoMS particles formed in the presence of ethanol (condition A). This trend was similar to those of MPDMS, except that under condition B, EpoMS particles formed but MPDMS particles did not. In the absence of ethanol (conditions B and C), EpoMS particles did form at a wide range of EpoMS concentrations (12.5-200 mM for both conditions). These findings suggest that the formation trend for EpoMS particles differed from the trends for TEOS particles and thiol-organosilica particles. Previously we proposed the following mechanism for the formation of thiol-organosilica particles: micelles of thiolorganosilicate form first, and then ammonium hydroxidecatalyzed hydrolysis and condensation result in the formation of particles.32,33 The concentration of ammonium hydroxide, the formation and the structure of the thiol-organosilicate micelles, and the kinetics of the hydrolysis and condensation reactions might be important in considering the formation of thiolorganosilica particles. In the case of EpoMS, the presence or absence of ethanol affected particle formation, but neither the EpoMS concentration nor the ammonium hydroxide concentration had any effect. The EpoMS reaction mixtures became cloudy within a few hours under condition B and condition C, but under condition A, the mixtures did not become cloudy, even after 3 days, which indicates that micelles did not form. EpoMS particle formation may involve the formation of EpoMS micelles and subsequent ammonium hydroxide-catalyzed hydrolysis and polymerization of EpoMS, resulting in the formation of particles like those of thiol-organosilica particles. Under condition A, EpoMS micelle formation may have been prevented by the presence of the ethanol, owing to the solubility of EpoMS in ethanol. The ranges of EpoMS concentrations at which particles formed (12.5-200 mM) and at which ones could not form (6.25 mM) were the same for condition B and condition C. These results differs from the results for the thiol-organosilica particles, because the formation trends for thiol-organosilica particle differed between condition B and condition C. High concentrations of MPMS and MPES under condition B, and low concentrations of MPMS, MPES, and MPDMS under condition C, could form particles after 1 day.33 These results indicated that low concentration of EpoMS affected particle formations. It is possible that low concentrations of EpoMS formed EpoMS micelles with unstable structure to form particles or could not
Epoxy-Organosilica Particles
Langmuir, Vol. 24, No. 21, 2008 12231
Table 2. Size Evaluation of EpoMS Particles Prepared Using Conditions B and C condition
concn (mM)
diameter (nm)
CV (%)
B B B B B C C C C C
200 100 50 25 12.5 200 100 50 25 12.5
1630 1407 824 512 362 1436 1139 879 533 92
5.7 8.1 9.5 21.2 22.3 10.9 13.2 17.0 33.4 46.2
form EpoMS micelles under condition B and condition C. Therefore, the hydrolysis and condensation reactions could not be effective to form EpoMS particles. It is possible that the mechanism for the formation of EpoMS particles is similar to those of thiol-organosilica particles: micelles of EpoMS form first, and then ammonium hydroxide-catalyzed hydrolysis and condensation result in the formation of particles. 2. Characterization of EpoMS Particles. 2.1. Size Distribution. The size distribution of the EpoMS particles was evaluated by TEM (Table 2). The mean particle diameters and standard deviations are plotted against EpoMS concentration in Figure 3. Under condition B and condition C, the size distributions were concentration dependent. A decline in the concentration dependence was observed at concentrations above 100 mM EpoMS under both condition B and condition C. No decline in concentration dependence was observed for TEOS, MPMS, MPES, or MPDMS particles under condition C.33 The mean diameters of EpoMS particles obtained under condition B and condition C were similar at all EpoMS concentrations except 6.12 mM. The mean diameters of the EpoMS particles were larger than those of the MPMS and MPES particles;33 under condition B, the diameter of the EpoMS particles was about double the diameters of the corresponding MPMS and MPES particles at the higher concentrations (50-200 mM).33 There was difference about size of particle between EpoMS particles and TEOS and thiol-organosilica particles at the same concentrations. Under conditions B and C, we were able to prepare EpoMS particles with narrow size distributions (best coefficient of variation, 5.7%; Table 2). The coefficient of variation (CV) varied with EpoMS concentration: at higher concentrations (50-200 mM), the CV was under 20%. The best CV values were obtained at the highest concentrations under both condition B and condition C (Figure 3). In our previous study,33 we were unable to synthesize thiol-organosilica and TEOS particles with narrow size distributions (CV < 20%). There were also substantial differences between the size distribution of the EpoMS particles and the
Figure 3. Mean particle diameter and standard deviation of size distribution as a function of EpoMS concentration under condition B (black solid circles) and condition C (red open squares).
Table 3. Zeta Potential of Organosilica Particles particles MPMS NPs EpoMS NPs EpoMS NPs EpoMS NPs EpoMS NPs EpoMS NPs EpoMS NPs
containing rhodamine B containing Rubpy + 0.04 mg/mL BSA + 0.2 mg/mL BSA + 1 mg/mL BSA
zeta potential (mV) -49.4 +46.1 +44.8 +46.1 +26.0 +9.8 -2.05
distributions of the particles prepared in our previous study under the same conditions.33 2.2. Surface Characterization and Functionalization. We analyzed the zeta potentials of the EpoMS particles to characterize the surface charge of the particles. As shown in Table 3, the zeta potentials of the EpoMS particles were positive and were different from thiol-organosilica particles and TEOS particles showing the negative zeta potentials.32,33 The epoxy group is highly reactive with amines and forms a thermosetting epoxide polymer when mixed with a catalyzing agent such as amines. It is possible that epoxy group on the surface of EpoMS particles reacted with ammonium hydroxide and then formed a positively charged amino group. The clear reason why EpoMS particles showed a positive surface charge and more understanding of the molecular structure of EpoMS particles require further investigations. The zeta potentials of EpoMS particles containing rhodamine B or Ru(bpy) in their interiors were almost the same as for particles without dye, which indicates that incorporation of the dyes did not affect the zeta potential. It is reported that TEOS particles are silanized with N1-[3-(trimethoxysilyl)propyl]diethylenetriamine or aminopropyl triethoxysilane to prepare amino-modified TEOS nanoparticles.12 Amino-modified silica nanoparticles are also synthesized by using the synchronous hydrolysis of tetraethoxysilane and N-γ-aminopropyltriethoxysilane.34 These silica particles are positively charged due to the cationic amino groups on the surface. However, EpoMS particles were made from just EpoMS as a single silica source and showed a positive surface charge. These EpoMS particles are the first silica particles reported to have positive zeta potential as prepared from a single silica source. Biomolecules, including DNA and most proteins, are anionic and can interact with the EpoMS particles with positive zeta potential. Biomolecules have amino groups that can form covalent linkages with the epoxy group. Therefore, EpoMS particles with positive zeta potential have great potential for use in surface functionalization with biomolecules. We used flow cytometry to investigate the surface functionalization of EpoMS particles with proteins and DNA. The particle solutions were mixed with protein solutions containing either green fluorescent protein (GFP) or with DNA solutions containing fluorescein-labeled oligonucleotides. After the particle solutions were mixed with GFP or the labeled oligonucleotides, the flow cytometry peaks for EpoMS particles shifted markedly to the right owing to fluorescence from bound GFP or bound fluorescein-labeled oligonucleotide on particles (Figure 4a,c). As a control experiment, MPMS particles were treated with GFP, and after treatment, the peak for the MPMS particles shifted markedly to the right owing to fluorescence from GFP (Figure 4b). As previously reported,32,33 MPMS particles bind with proteins much more strongly than do TEOS particles. The flow cytometry results indicate that both thiol-organosilica particles and EpoMS particles bind strongly with GFP. However, when the MPMS particles were treated with fluorescein-labeled oligonucleotide, no substantial peak shift (34) He, X. X.; Wang, K.; Tan, W.; Liu, B.; Lin, X.; He, C.; Li, D.; Huang, S.; Li, J. J. Am. Chem. Soc. 2003, 125, 7168–7169.
12232 Langmuir, Vol. 24, No. 21, 2008
Nakamura and Ishimura
Figure 4. Flow cytometry analysis of silica particles surface-modified with protein and DNA. EpoMS particles (a, c) and MPMS particles (b, d) modified with GFP (a, b) or fluorescein-labeled oligonucleotide (c, d) were analyzed. Red lines and green lines indicated before and after reaction, respectively.
was observed (Figure 4d). These findings indicate that EpoMS particles, unlike MPMS particles, bound both with proteins and with DNA. Thus, the ability to bind to DNA appears to be unique to EpoMS particles. The difference between EpoMS particles and MPMS particles to the ability to bind to DNA could be the difference of their surface charge. The DNA might bind electrostatically with the positive charged surface of EpoMS particles. We used fluorescence microscopy to confirm the surface modification of EpoMS particles with protein or DNA and to evaluate the dispersion of the surface-modified particles. Solutions of EpoMS particles mixed with GFP or DNA showed welldispersed particles with distinct fluorescence (data not shown). In addition, the flow cytometry peaks for EpoMS particles surfacemodified with GFP and fluorescein-labeled oligonucleotides were very sharp, and peaks indicating aggregates were minor. These findings indicate that EpoMS particles retained good dispersion after surface modification. When EpoMS particles were surface-treated with BSA, the zeta potential was reduced owing to the negative charge of the BSA, and the amount of the reduction increased with increasing BSA dose (Table 3). Our results indicate that the positive charge of the EpoMS particles could be regulated by surface modification with negatively charged biomolecules. EpoMS particles with surface positive charge have high potential for surface functionalization using biomolecules including not only proteins but also DNA. In addition the regulation of the surface charge of EpoMS particles using biomolecules such as BSA might be useful to reduce the nonspecific bindings with biomolecules and cells during cell labeling. Further investigations and characterizations of the surface of EpoMS particles are required to make it possible to prepare useful particle for various applications including bioassay, imaging, and drug delivery systems. 2.3. Internal Functionalization. We used the one-pot method to synthesize fluorescent EpoMS particles doped and tuned with fluorescent dyes. Various concentrations of rhodamine B or Ru(bpy) were mixed with a condition B solution containing 200 mM EpoMS. Fluorescent-tuned EpoMS particles were analyzed by means of flow cytometry. Fluorescent-tuned EpoMS particles
containing rhodamine B were detected as six distinct peaks with different fluorescence intensities depending on the concentration between 6.25 µM to 200 µM of rhodamine B used to prepare the particles (emission at 585/42 nm, FL-2; Figure 5A). Fluorescent-tuned EpoMS particles containing Ru(bpy) were also detected as six distinct peaks with different fluorescence intensities depending on the concentration between 6.25 µM to 200 µM of Ru(bpy) used to prepare the particles (emission at above 670 nm, FL-3; Figure 5B). The peaks of fluorescent-tuned EpoMS particles containing rhodamine and fluorescent-tuned EpoMS particles containing Ru(bpy) could be distinguished clearly as 12 spots in a plot of FL-3 versus FL-2 depending on the fluorescence intensities (Figure 5C). These results indicate that the fluorescence intensities of EpoMS particles can be tuned and can be detected as distinct peaks and spots clearly by flow cytometry analysis, and the results also indicate that the sizes of fluorescent-tuned EpoMS particles were well controlled because the shapes of the peaks were sharp. To evaluate the stabilities of fluorescent dye in the EpoMS particles we performed flow cytometric assay again using the same particles as before. These particles remained stable 6 months after the preparation. The fluorescent intensities of fluorescenttuned EpoMS particles containing rhodamine and fluorescenttuned EpoMS particles containing Ru(bpy) were almost the same as those of the first experiment (data not shown). These results indicated that dye leakages of the fluorescent-tuned EpoMS particles were almost none and EpoMS particles had good stabilities to keep the fluorescent dye within the particles. Using fluorescence microscopy, we evaluated photostabilities of EpoMS particles containing rhodamine in PBS. Fluorescent intensities of fluorescent EpoMS particles remained above 60% of their initial values after 250 s under continuous excitation (data not shown). In our previous work, fluorescent intensities of MPMS particles containing rhodamine remained above 50% under the same conditions.32 These results indicated that fluorescent EpoMS particles had good photostabilities as the same as in the case of fluorescent MPMS particles. In addition fluorescent EpoMS particles in FBS exhibited also high fluorescent intensity and photostability; fluorescent intensities
Epoxy-Organosilica Particles
Langmuir, Vol. 24, No. 21, 2008 12233
Figure 6. Fluorescence microscopy and electron microscopy of mouse peritoneal cells containing fluorescent EpoMS particles containing rhodamine B: peritoneal cells labeled with fluorescent EpoMS particles containing rhodamine B were observed with excitation at 520/15 nm (a) or under a bright field (b). (c) View shows the merged image. Electron microscopy images of mouse peritoneal cells containing fluorescent EpoMS particles: (d) macrophage containing EpoMS particles in the cytoplasm; the particles were observed in endosomal membranes (arrow heads) (e, f, and g) or the cytosol (g). Scale bars are 500 nm (d), 200 nm (e), and 100 nm (f and g). Figure 5. Flow cytometry analysis of fluorescent-tuned EpoMS particles containing rhodamine B (A) and Ru(bpy) (B). Fluorescent-tuned EpoMS particles were prepared with 200 µM (green lines), 100 µM (red lines), 50 µM (light blue lines), 25 µM (orange lines), 12.5 µM (blue lines), and 6.25 µM (yellow lines) of rhodamine B or Ru(bpy), respectively. View C is a plot of FL-3 against FL-2 for EpoMS particles containing rhodamine B (red) and EpoMS particles containing Ru(bpy) (blue).
remained above 60% of their initial values after 250 s. These results indicated that high possibilities of fluorescent EpoMS particles for imaging in vivo. In addition to EpoMS particles doped with fluorescent dye, we prepared fluorescent EpoMS particles using chemical crosslinking reagents, such as maleimides and succinimidyl esters, as described in a previous report.26,32,33 EpoMS particles containing various kinds of fluorescent dye-(3-aminopropyl)trimethoxysilane conjugates and fluorescent dye-MPMS conjugates showed high fluorescence (data not shown). Fluorescent particles could be prepared from EpoMS both by using cross-linking reagents containing various kinds of dyes and by using fluorescent dye doping, which is less expensive. Fluorescent-tuned EpoMS particles and EpoMS particles with surfaces modified by proteins and DNA were clearly detected as distinct peaks during flow cytometry. Recently, the multiplexed flow cytometric assay, which is a rapid, highly sensitive, and accurate assay for multiple samples of small volume in the same tube, has become popular.37,38 The ability to functionalize EpoMS
particles both internally and on the surface makes them potentially advantageous for use as microbeads in the multiplexed flow cytometric assay, not only for proteins but also for DNA. The development of a high-throughput bead assay using flow cytometry with EpoMS particles and thiol-organosilica particles for biological investigation is under investigation. 3. In Vivo Injection of EpoMS Particles and Histological Analysis. For in vivo injection, fluorescent EpoMS particles were prepared from a condition B solution containing 12.5 mM EpoMS and 200 µM of rhodamine B. The average sizes of the formed particles was about 390 nm. Mice were intraperitoneally injected with a large amount of fluorescent EpoMS particles containing rhodamine B (up to 1 mg/mouse). Injection resulted in no apparent toxicity, abnormal changes, or death for up to 1 month. Peritoneal cells were harvested 1 day after injection and examined by fluorescence and light microscopy. Cells showing fluorescence were observed clearly (Figure 6a-c). Some fluorescence-positive cells showed a cluster-like appearance of fluorescence. The peritoneal cells were examined by electron microscopy. As shown in Figure 6d, cells containing many particles in the cytoplasm were observed. The cells containing particles had phagocytic vacuoles in the cytoplasm and an irregular surface with pleats and protrusions, a well-developed Golgi (37) Kellar, K. L.; Iannone, M. A. Exp. Hematol. 2002, 30, 1227–1237. (38) Nolan, J. P.; Mandy, F. Cytometry, Part A 2006, 69, 318–325.
12234 Langmuir, Vol. 24, No. 21, 2008
complex, and a prominent rough endoplasmic reticulum. These findings indicate that the fluorescence-positive cells were peritoneal macrophages and that macrophages performed nonspecific phagocytosis of the fluorescent EpoMS particles and thus showed fluorescence from their cytoplasm. The fact that lymphocytes and eosinophils contained no vacuoles indicates that no endocytosis of EpoMS particles occurred. In addition, no indications of cell necrosis or apoptosis were observed histologically. Within these experiments we could not detect any clear toxic findings due to fluorescent EpoMS particles. However, there are various test to study the safety of chemicals, including chronic toxicity test, carcinogenesis test, mutagenicity test, neurotoxicity tests, immunotoxicity test, etc. Fluorescent EpoMS particles as well as other organosilica particles should be examined for toxicities using various tests according to their applications. In experiments similar to this study, fluorescent EpoMS particles will be useful for labeling cells. The intracellular distribution of EpoMS particles showed various patterns. As shown in Figure 6e, many EpoMS particles were contained in a large endosome. These findings indicate that the cluster-like appearance of fluorescence in the positive cells on fluorescence microscopy was due to the presence of a large endoplasm containing EpoMS particles. A single EpoMS particle in one endoplasm was also observed. In addition, EpoMS particles located in the cytosol were observed. As shown in Figure 6g, an endosomal membrane was observed around the smaller of the two particles in the figure (right), whereas no endosomal membrane was observed around the larger particle (left). These findings indicate that some EpoMS particles escaped from the endoplasm to the cytosol.39 The surface charge of particles strongly affects in vivo metabolism, distribution, internalization, and intracellular localization of the particles. The use of positively or negatively charged nanoparticles to target specific locations has been reported; positively charged nanoparticles made from polyethylene glycol (PEG) by PEGylation are internalized rapidly by means of the clathrin-mediated pathway, whereas negatively charged nanoparticles show a lower rate of endocytosis and do not utilize the clathrin-mediated endocytosis pathway.35 That is, the surface charge affects the rate of internalization into cells. (39) Yessine, M. A.; Leroux, J. C. AdV. Drug DeliVery ReV. 2004, 56, 999– 1021.
Nakamura and Ishimura
The surface charge can also be used to change the intracellular localization: poly(DL-lactide-co-glycolide) nanoparticles that are positively charged at endosomal pH (pH 4-5) escape from the endosomal compartment to the cytoplasm, whereas nanoparticles that are negatively charged at pH 4 are retained mostly in the endosomal compartment.36 The fluorescent EpoMS particles had a positive charge as prepared, and the surface charge of EpoMS could be regulated by modification with biomolecules. The relationships between the surface properties of particles (such as their aptitude for modification with biomolecules and their zeta potential) and the interaction of the particles with cells are of great importance for the development of drug-delivery systems. Organosilica particles, including EpoMS particles (positively charged) and thiol-organosilica particles (negatively charged), functionalized internally and on their surface, are expected to be useful for various applications.
Conclusion Epoxy-organosilica particles were synthesized in one pot from EpoMS in the absence of ethanol. The EpoMS particles were well-dispersed and showed narrow size distributions. The EpoMS particles could be internally modified, owing to the presence of interior carboxy chains and epoxy residues, which permitted the one-pot synthesis of fluorescent EpoMS particles. The particles showed positive zeta potentials and could be modified with both proteins and DNA. Fluorescent EpoMS particles did not exhibit death on in vivo injection nor was cytotoxicity found in labeled cells histologically. The EpoMS particles show high potential for use in various applications, such as biomedical analysis, chipbased technology, multitarget detection systems, in vitro and in vivo imaging, and nanomedicine. Acknowledgment. This work was supported in part by a Grantin-Aid for Younger Scientists (to M.N.), by a Grant for Practical Application of University R&D Results under the Matching Fund Method (to M.N.) from the New Energy and Industrial Technology Development Organization (NEDO) of Japan, and by a Grantin-Aid for Scientific Research (C) (to M.N.). We thank to Mr. Hiroyuki Doi (student, The University of Tokushima Faculty of Medicine) for his technical assistance. The work is under a patent pending (PCT/JP2007/61587). LA801950Q
