Article pubs.acs.org/Langmuir
Dendritic Silica Nanomaterials (KCC-1) with Fibrous Pore Structure Possess High DNA Adsorption Capacity and Effectively Deliver Genes In Vitro Xiaoxi Huang,† Zhimin Tao,*,†,‡ John C. Praskavich, Jr.,¶ Anandarup Goswami,†,‡ Jafar F. Al-Sharab,⊥ Tamara Minko,¶ Vivek Polshettiwar,§ and Tewodros Asefa*,†,‡ †
Department of Chemistry and Chemical Biology, Rutgers, The State University of New Jersey, 610 Taylor Road, Piscataway, New Jersey 08854, United States ‡ Department of Chemical and Biochemical Engineering, Rutgers, The State University of New Jersey, 98 Brett Road, Piscataway, New Jersey 08854, United States ¶ Department of Pharmaceutics, Ernest Mario School of Pharmacy, Rutgers, The State University of New Jersey, 160 Frelinghuysen Road, Piscataway, New Jersey 08854, United States § Nanocatalysis Laboratory (NanoCat), Department of Chemical Sciences, Tata Institute of Fundamental Research (TIFR), Homi Bhabha Road, Mumbai 400005, India ⊥ Department of Mechanical & Aerospace Engineering, Polytechnic Institute of New York University (NYU-Poly) Six MetroTech Center, Brooklyn, New York 11201, United States S Supporting Information *
ABSTRACT: The pore size and pore structure of nanoporous materials can affect the materials’ physical properties, as well as potential applications in different areas, including catalysis, drug delivery, and biomolecular therapeutics. KCC-1, one of the newest members of silica nanomaterials, possesses fibrous, large pore, dendritic pore networks with wide pore entrances, large pore size distribution, spacious pore volume and large surface area structural features that are conducive for adsorption and release of large guest molecules and biomacromolecules (e.g., proteins and DNAs). Here, we report the results of our comparative studies of adsorption of salmon DNA in a series of KCC-1-based nanomaterials that are functionalized with different organoamine groups on different parts of their surfaces (channel walls, external surfaces or both). For comparison the results of our studies of adsorption of salmon DNA in similarly functionalized, MCM-41 mesoporous silica nanomaterials with cylindrical pores, some of the most studied silica nanomaterials for drug/gene delivery, are also included. Our results indicate that, despite their relatively lower specific surface area, the KCC-1-based nanomaterials show high adsorption capacity for DNA than the corresponding MCM-41based nanomaterials, most likely because of KCC-1’s large pores, wide pore mouths, fibrous pore network, and thereby more accessible and amenable structure for DNA molecules to diffuse through. Conversely, the MCM-41-based nanomaterials adsorb much less DNA, presumably because their outer surfaces/cylindrical channel pore entrances can get blocked by the DNA molecules, making the inner parts of the materials inaccessible. Moreover, experiments involving fluorescent dye-tagged DNAs suggest that the amine-grafted KCC-1 materials are better suited for delivering the DNAs adsorbed on their surfaces into cellular environments than their MCM-41 counterparts. Finally, cellular toxicity tests show that the KCC-1-based materials are biocompatible. On the basis of these results, the fibrous and porous KCC-1-based nanomaterials can be said to be more suitable to carry, transport, and deliver DNAs and genes than cylindrical porous nanomaterials such as MCM-41.
1. INTRODUCTION
required. In other words, for the success of gene therapy, gene delivery systems that can carry large payloads of genes and allow controlled release of their contents on demand are vital. Over the
1
Ever since its inception, gene therapy is regarded as a powerful therapeutic method for treating various complex medical problems, such as cancer, monogenic disorders, cardiovascular diseases, etc. For gene therapy to work effectively toward curing these diseases though, gene delivery systems capable of achieving the desired therapeutic effects efficiently as well as safely are © XXXX American Chemical Society
Received: April 15, 2014 Revised: August 18, 2014
A
dx.doi.org/10.1021/la501435a | Langmuir XXXX, XXX, XXX−XXX
Langmuir
Article
past few years, research efforts to find such gene delivery systems have yielded some interesting materials, including various polymer-based (nano)materials, liposomes and inorganic nanoparticles.2−6 Among these, silica-based nanoparticles (both porous and nonporous types) stand out as some of the most promising materials for gene delivery because of their biocompatibility, large surface area, tunable pore size and conducive surface properties for further functionalization.7 In fact, a number of them have already been successfully demonstrated to deliver nucleic acids (DNAs and siRNAs) into cells. For example, mesoporous silica nanoparticles (MSNs) modified with polyamidoamine dendrimers were shown to have high adsorption capacity for plasmid DNA, and thereby greater ability to transfect HeLa cells.8 In another example, MSNs functionalized with polyethylenimine-polyethylene glycol copolymers were successfully applied to codeliver anticancer drug and siRNA into breast cancer cells, reducing the cells’ ability to develop drug resistance.9 However, in typical nanoparticle-based gene delivery systems, including those in the aforementioned two examples, the nucleic acids were carried exclusively by the exterior surfaces of the nanoparticles. Conversely, the inner pore spaces of the nanoparticles are unable to carry the nucleic acids, resulting in the nanoparticles’ low loading capacity for nucleic acids.8,9 This is often to do with the fact that the pores of these nanoparticles are too small and/or cylindrical in shape to accommodate the relatively large size nucleic acid molecules. Hence, finding nanoporous materials whose pores are large enough (or more suitable) to hold nucleic acids (genes), and thereby result in high adsorption capacity for nucleic acid molecules, is desirable. Needless to say such types of large pore nanoporous materials can also protect the large payloads of DNA or siRNA molecules hosted inside their pores from undergoing undesired enzymatic degradation, as previously demonstrated for some large pore mesoporous silica materials with potential applications for gene delivery.10−13 While the surface area and pore volume of nanoporous materials can undoubtedly play major roles in the materials’ ability to adsorb guest molecules including nucleic acids,14 the pore geometry of the materials can also profoundly affect the materials’ adsorption and release properties toward different molecular cargoes. For example, Mika and coworkers studied how ibuprofen is adsorbed and released by a series of ordered microporous and mesoporous silica materials possessing different pore size, pore connectivity and pore geometry, and they found that the materials’ pore connectivity as well as pore geometry can dictate how ibuprofen is released from the materials.15 Specifically, they showed that the material with onedimensional and cage-like pores is the one that releases the drug most slowly. In a different study, Sung and coworkers synthesized porous materials with cylindrical and lamellar nanoarrays for drug delivery; they found that the drug release rates from the materials could easily be controlled by altering the size, geometry or depth of the nanoscale pores in the materials.16 Similar observations on the effect of pore geometry of porous materials on the materials’ adsorption/release profiles toward molecules other than drugs have also been documented. For instance, Yang and coworkers reported that the pore geometry of metal-organic frameworks is one of the crucial factors that affect the materials’ adsorption behavior towards thiophenic sulfur from diesel oil.17 In another example, Carrott and coworkers evaluated the effect of pore size, geometry and surface chemistry of mesoporous silica and mesoporous organosilica materials on
the materials’ adsorption property for hydrocarbons; their results indicated that the materials’ pore size as well as pore geometry plays important roles in how hydrocarbons adsorb in these porous materials.18 In contrast to conventional porous silica-based nanomaterials with cylindrical pores (such as MCM-41 or SBA-15), KCC-1,19 which is a new type of silica nanomaterial, has fibrous, dendriticlike pore structures. Moreover, KCC-1 possesses high surface area (typically >700 m2/g), large pore sizes, and broad pore size distribution. These unique structural features of KCC-1 have already been shown to render KCC-1 potential applications in areas such as heterogeneous catalysis20 and CO2 capture.21 These pore structures could also be expected to make KCC-1 an ideal host material for effectively carrying bulky guest molecules such as DNAs, RNAs and genes and delivering the latter to cells. This is actually a reasonable expectation especially considering the report by Gai et al.,22 which showed that fibrous mesoporous silica nanomaterials have improved properties for delivery of the small molecule anticancer drug doxorubicin. The hype on KCC1's potential applications for drug delivery can be even more so, when considering the recent work by Du et al.,23 which showed that dendritic-type silica nanoparticles possessing some large pores around their external surfaces and many small pores throughout their structure can codeliver antitumor drug and genes with improved therapeutic effect. Herein, we report detailed studies of the adsorption properties and adsorption capacity of KCC-1-based silica nanomaterials with extensive large pores and fibrous structure throughout the nanoparticles for macromolecular salmon DNA. We also show that the adsorption capacities and release properties of KCC-1based materials are better compared with the corresponding mesoporous materials with cylindrical pores such as MCM-41, which have been widely studied as host materials for various guest molecules. Our comparative studies include DNA adsorption on different surfaces of KCC-1 and MCM-41-based materials possessing two different types of organoamine groups on their different surfaces: (i) only on the pore walls, (ii) only on the external surfaces, and (iii) on both surfaces. The results generally show that: (1) grafting organoamine groups on the surfaces of KCC-1 augments KCC-1’s adsorption capacity for DNAs and (2) compared with the MCM-41-based materials, the amine-grafted KCC-1 materials are more effective and advantageous for transporting and delivering DNAs loaded on the materials’ surfaces into cells in vitro. The effects of organoamine modification and surfactant removal methods (extraction versus calcination) on the cytotoxicity/biocompatibility of the resulting materials are also described. Overall, our studies and results reveal that the fibrous, large pore KCC-1-based materials are more suitable to carry, transport, and deliver DNAs and genes than materials with small, cylindrical pores, such as MCM-41.
2. MATERIALS AND METHODS 2.1. Materials. Tetraethyl orthosilicate (TEOS), (3-aminopropyl)trimethoxysilane (APTMS), hexamethyldisilazane (HMDS), deoxyribonucleic acid sodium salt from salmon testes, anhydrous toluene, N,Ndimethylformamide (DMF), cetylpyridinium bromide (CPB), and cetyltrimethylammonium bromide (CTAB) were all obtained from Sigma-Aldrich. N-(2-aminoethyl)-3-aminopropyltrimethoxysilane (AAPTMS) was acquired from Gelest, Inc. Cell counting kit-8 was purchased from Dojindo Laboratory. Phosphate buffered saline (PBS) was obtained from GIBCO. Oligonucleotide DNA conjugated with cyanine dye (Cy-3) (Oligo-DNA-Cy3) was purchased from Gene Link. All the reagents were used as received. B
dx.doi.org/10.1021/la501435a | Langmuir XXXX, XXX, XXX−XXX
Langmuir
Article
Scheme 1. Schematic Illustration of the Synthetic Procedures Employed for Preparation of KCC-NH2, KCC-NH2-Ext, KCC-NH2E, and KCC-NH2-I
with distilled water and then ethanol and finally dried in oven overnight at 80 °C, giving as-synthesized CTAB-containing mesostructured silica (denoted here as MCM-41-CTAB). The CTAB surfactant templates from the material were removed by calcination at 550 °C in air for 5 h to afford MCM-41 mesoporous silica, labeled hereafter as MCM-41. To graft organoamine groups on MCM-41, 400 mg of MCM-41 was first mixed with anhydrous toluene (25 mL) and sonicated for 30 min. Then, APTMS (0.8 mmol, 140 μL) was added into the mixture under stirring. After stirring the mixture for 6 h at 80 °C, the solid product was collected by centrifugation and washed with ethanol three times (3 × 20 mL). The material was dried in oven overnight at 50 °C giving MCMNH2, which has organoamine groups throughout its surfaces (i.e., both on its inner channel walls as well as external surfaces). 2.3. Synthesis of KCC-1 and MCM-41 with Amine Groups only on External Surfaces using Solvent-Extracted KCC-1 and MCM41 Materials (KCC-NH2-E and MCM-NH2-E). To prepare KCC-1 and MCM-41 nanoparticles with organoamine groups only on the external surfaces, the postgrafting reaction of aminorganosilanes was carried out with the as-synthesized materials containing surfactant templates. In a typical synthesis, the as-synthesized, surfactant-containing KCC-CPB or MCM-41-CTAB (200 mg) was dispersed in 10 mL anhydrous toluene and sonicated for 30 min. APTMS (0.2 mmol, 35 μL) was then added into the mixture. After stirring the mixture for 6 h at 80 °C, the solid product was separated by centrifugation, copiously washed with ethanol, and then dried at 50 °C overnight. The resulting material, denoted as KCC-CPB-NH2 or MCM-CTAB-NH2, contained organoamine groups only on its external surfaces and CPB or CTAB templates, respectively,
2.2. Synthesis of Various Amine-Grafted KCC-1 and MCM-41 using Calcined KCC-1 and MCM-41 (KCC-NH2, KCC-NH-NH2, and MCM-NH2). (Note: Please refer to Scheme 1 and Scheme S1 and Tables S1−S4 in Supporting Information for synthesis flowchart and the list of materials we prepared and investigated.) First KCC-1 was synthesized by following previously reported procedures.19 Prior to the postgrafting reaction, the KCC-1 nanoparticles were dried in oven at 50 °C to remove physisorbed water that may be present on their surfaces. This procedure was followed also for the syntheses of the other organoaminemodified nanomaterials described below. Then 400 mg of the dried KCC-1 was placed in dried glassware and mixed with anhydrous toluene (25 mL). The mixture was sonicated for 30 min to disperse the KCC-1 microspheres. Into the resulting dispersion, APTMS (0.8 mmol, 140 μL) was added, and the mixture was stirred at 80 °C for 6 h. After this, the solid product was isolated by centrifugation and washed with ethanol (3 × 20 mL). The product was dried in oven overnight at 50 °C, resulting in 3-aminopropyl-grafted KCC-1, labeled as KCC-NH2. Similarly, KCC-NH-NH2 was synthesized using AAPTMS (0.8 mmol), instead of APTMS, during the grafting step. For the synthesis of MCM-NH2, first MCM-41-type mesostructured material was synthesized according to previously reported methods.24 In a typical synthesis, 480 mL distilled water was mixed with CTAB (5.5 mmol, 2 g) and NaOH solution (2.0 M, 7 mL). The mixture was stirred at 80 °C for 30 min to completely dissolve the CTAB. Then, TEOS (44.8 mmol, 10 mL) was added dropwise into the solution over 3 min, and the solution was stirred for 2 h at 80 °C. The resulting solid product was collected via filtration of the mixture. The solid product was washed C
dx.doi.org/10.1021/la501435a | Langmuir XXXX, XXX, XXX−XXX
Langmuir
Article
curve was then used to determine the concentration of DNA in any of the other solutions. In a typical adsorption experiment, 1 mg/mL dispersion of nanoparticles containing 150 μg/mL DNA was prepared and stirred at room temperature for 48 h. At certain intervals of time, the mixture was taken and centrifuged, and the UV−vis spectrum of the supernatant was then measured. The amounts of DNA adsorbed on the nanoparticles was calculated by subtracting the amount of DNA in the supernatants from the initial amount of DNA in the original solutions, and then normalizing the values with the weight of the nanoparticles. 2.7. Cell Culture and in Vitro Toxicity Assay. A2780 cell lines derived from ovarian carcinoma were used to test the toxicity/ biocompatibility of the nanomaterials. Typically, cells were cultured in RPMI 1640 media containing 10% fetal bovine serum at 37 °C under 5% CO2 incubator. Then, 100 μL/well cell suspension was distributed into a 96-well plate containing 5,000 A2780 cells per well. After adding each type of nanomaterial synthesized above the cell culture medium with final concentrations of 25, 50, and 100 μg/mL, the cells were incubated at 37 °C for 48 h. Then, 10 μL WST-8 (Dojindo Laboratory) was added into each well, and incubation of the cells was continued for another 2 h. The absorbance (Abs) of the solutions at 450 nm was measured using a microplate reader, and cell viability was calculated with [(Abstreated − Absmedia)/(Absuntreated − Absmedia) ± standard deviation]%.24,27 2.8. Oligonucleotide DNA-Cy3 Cell Uptake Assay. Cell uptake assays were carried out in 24-well cell culture plates. A2780 cells were seeded in plates at 5 × 104 cells/well density and then incubated at 37 °C with 5% CO2 for 24 h. At the same time, 1000 μg/mL suspensions of nanoparticles were prepared in PBS, both for MCM-NH2 and KCCNH2 under sonication for 1 h. Then, 5 μL of the resulting dispersion was taken and mixed with 1 μL Oligo-DNA-Cy3 solution (2 ng/μL). The mixture was put into 4 °C refrigerator to allow the binding of DNAs with the materials. After this, 1 mL of fresh culture medium was added into the aforementioned mixture of nanoparticles and DNA solution. After removing the cell culture medium, a fresh culture medium containing 5 μg/mL nanoparticles and 2 ng/mL Oligo-DNA-Cy3 was added into it. For control experiments, A2780 cells without Oligo-DNA-Cy3 and A2780 cells incubated only with Oligo-DNA-Cy3 (2 ng/mL) (i.e., without nanoparticles) were used. After incubation for another 30 min, the culture medium was removed, and the cells were carefully washed with PBS solution and transferred to a fluorescence microscope for observation. 2.9. Incubation of Cells with KCC-1 Nanoparticles and Their Observation by SEM. A2780 cells were seeded in 12 well plate (where a small glass slide was placed at the bottom of the well so that cells can grow on the surface of the glass) for 48 h at the density of 100 000 cell per well. Then a KCC-1 stock solution in PBS was prepared and sonicated for 1 h. Into 2 mL cell culture media that was present in each well, the KCC-1 stock solution was then added to a final concentration of KCC-1 50 μg KCC/mL. The cells were incubated with the KCC-1 nanoparticles for 1 h at room temperature. After this, the cell media were carefully removed, and the cells on the glass sheet were washed three times with PBS and subjected to fixation with 2.5% glutaraldehyde (50% glutaraldehyde diluted 20 times in 1 × PBS) for 3 h. The cells along with the glass slides were then washed with PBS three times, and dehydrated in 30%, 50%, 70%, 80%, and 90% (v/v) ethanol solutions for 5 min each. Finally, the glass slides were soaked in 100% ethanol three times for 5 min each. After completely letting them to dry under a flow of air, the cells/glass slides were coated with 10 nm thick gold to be analyzed by SEM. 2.10. Instrumentation and Materials Characterizations. Scanning electron microscopic (SEM) images were taken on a Zeiss Sigma Field Emission SEM. Transmission electron microscopy (TEM) images were obtained with a Topcon 002B TEM microscope operating at 200 kV. The UV−vis spectra were recorded on a Lambda 850 spectrophotometer (PerkinElmer). Nitrogen adsorption/desorption isotherms were measured using a Micromeritics Tristar-3000 instrument. Prior to each measurement, the sample was degassed at 323 K overnight with nitrogen flow. Using the isotherms, the surface area of the materials was obtained by the Brunauer−Emmett−Teller (BET) method and their pore size distributions were determined with the Barrett−Joyner−Halenda (BJH) method. Thermogravimetric analyses
in its channels. The KCC-CPB-NH2 or MCM-CTAB-NH2 (130 mg) was then stirred in a solution of ethanol (15 mL) and concentrated HCl (0.6 mL) to remove the CPB or CTAB templates. After stirring for 5 h at 60 °C, the mixture was centrifuged, and the solid product was recovered and dispersed in fresh ethanol/concentrated HCl (15 mL/0.6 mL) solution to remove any residual templates that may be present in the materials. The solid product was collected once again via centrifugation, and then washed thoroughly with 0.1 M NaHCO3 to remove any residual acid on the material.25 Finally, the product was washed with distilled water, then ethanol, and dried in oven at 50 °C overnight, giving KCC-NH2-E or MCM-NH2-E, respectively, that contain organoamine groups only on its external surfaces. 2.4. Synthesis of KCC-1 and MCM-41 with Amine Groups in their Interior Surfaces using Solvent-Extracted KCC-1 and MCM-41 Materials (KCC-NH2-I and MCM-NH2-I). To graft organoamine groups only on the inner channel walls of the materials, first the external surfaces of KCC-CPB or MCM-41-CTAB were modified with trimethylsilyl (−SiMe3) groups by stirring the as-synthesized KCC− CPB or MCM-41-CTAB (600 mg) in a solution of HMDS/toluene (1.125 mL/9 mL) for 18 h at room temperature. This made the external surface silanols of the as-synthesized KCC-CPB or MCM-41-CTAB to be protected with -SiMe3 groups, yielding materials denoted here as KCC-CPB-Me and MCM-CTAB-Me. The CPB or CTAB templates were then removed by stirring 500 mg of KCC-CPB-Me and MCMCTAB-Me in HCl/ethanol (50 mL/2 mL) solutions for 5 h at 60 °C. After collecting the solid products by centrifugation, the extraction process was repeated one more time. Finally, the solid materials were recovered via centrifugation, washed with ethanol and dried in oven at 50 °C overnight, producing surfactant-extracted KCC-Me-Ext and MCM-Me-Ext, respectively. To graft the amine groups into their channel walls, the KCC-Me-Ext or MCM-Me-Ext (200 mg) was dispersed in 12.5 mL anhydrous toluene via sonication for 30 min, and into the mixture APTMS (0.4 mmol, 70 μL) was added. After stirring the mixture at 80 °C for 6 h, the solid product was collected via centrifugation of the mixture. The solid products were then washed with ethanol three times (3 × 10 mL), resulting in materials labeled as KCCNH2-I and MCM-NH2-I, respectively, which contain organoamine groups only on their inner channel walls and −SiMe3 groups on their external surfaces. 2.5. Synthesis of KCC-1 and MCM-41 with Amines Groups both on Exterior and Interior Surfaces using Solvent-Extracted KCC-1 and MCM-41 (KCC-NH2-Ext and MCM-NH2-Ext). First, the surfactant templates in KCC-CPB or MCM-41-CTAB (200 mg) were removed by stirring the materials in HCl/ethanol (1 mL/20 mL) solution at 60 °C for 5 h. The extraction was performed twice to remove the surfactant templates as much as possible. The solid product was then collected via centrifugation and washed with distilled water, followed by ethanol extensively. After drying it in oven at 50 °C overnight, 80 mg of the resulting material (KCC-Ext or MCM-41-Ext) was taken and dispersed in anhydrous toluene (5 mL). After sonicating the mixture for 30 min, APTMS (0.17 mmol, 30 μL) was added into it. The mixture was stirred at 80 °C for 6 h, and the solid product was isolated from it by centrifugation. The solid product was then washed with ethanol three times (3 × 10 mL) and dried in oven overnight at 50 °C. This finally gave KCC-NH2-Ext or MCM-NH2-Ext, respectively, which possesses organoamine groups both on its internal and external surfaces. 2.6. Adsorption of DNAs in the Nanoparticles. The amount of DNA adsorbed on the materials as a function of time was monitored by mixing the materials with salmon DNA solutions and then measuring the concentrations of DNA in the supernatants at various time intervals over 48 h. To determine the concentration of DNA in the solutions, first a standard curve of absorbance versus concentration was obtained by using different known concentrations (1, 5, 25, 50, 75, 100, and 150 μg/ mL) of DNA in phosphate buffered saline (PBS), whose absorption spectra in the range of 200−350 nm were acquired. The absorbance at 260 nm on the spectra (which corresponds to the DNA absorption26) versus DNA concentration was fitted into a linear function according to the Beer−Lambert’s law, giving the equation: [DNA] = 71.6 × Abs260 − 4.08 (R2 = 0.993), where [DNA] stands for the DNA concentration and Abs260 denotes for the absorbance at 260 nm. The resulting calibration D
dx.doi.org/10.1021/la501435a | Langmuir XXXX, XXX, XXX−XXX
Langmuir
Article
(TGA) of the materials were performed with PerkinElmer TGA7 by heating the samples at a rate of 10 °C/min under a constant flow of nitrogen at a rate of 20 mL/min. Solid state NMR spectra 29Si (99.35 MHz) and 13C (125.75 MHz) were acquired using a Brüker AVANCE 500 NMR spectrometer. For the 29Si CP-MAS NMR experiments, 10.0 kHz spin rate, 5 s recycle delay, 4 ms contact time, 1H π/2 pulse width of 2.9 μs, and 1024 scans with 100 SWf-TPPM28 1H decoupling were employed. For the 13C CP-MAS NMR experiments, 10.0 kHz spin rate, 5 s recycle delay, 1 ms contact time, 1H π/2 pulse width of 2.9 μs, and 1600 scans using 100 kHz SWf-TPPM28 1H decoupling were employed. Scanning electron microscope (SEM) images of cells incubated with nanoparticles were obtained using a Shimadzu SS 550 SEM.
organoamines in them. This is further confirmed by elemental analysis, which shows the amount of nitrogen in KCC-NH2 and KCC-NH-NH2 to be 1.54 and 2.83 mmol N/g, respectively. Nitrogen adsorption/desorption studies of the materials further reveals the successful modification of the materials with the organoamine groups (Table S1). Specifically, while the surface area of KCC-CPB is relatively small (179 m2/g), as expected, that of the surfactant-free KCC-1 is significantly higher (737 m2/g). The organoamine-grafted samples, KCC-NH2 and KCC-NH-NH2, on the other hand, have smaller surface areas (490 and 393 m2/g, respectively) and smaller pore volumes (1.12 and 0.98 cm3/g, respectively) (cf. KCC-1, whose surface area and pore volume are 737 m2/g and 1.58 cm3/g, respectively) (Table S1). These results further confirm the presence of organoamine groups in KCC-NH2 and KCC-NH-NH2 (or their absence in KCC-1). The gas adsorption isotherms of all the KCC-1 are found to be of type IV, with a clear H3-type hysteresis loop at relatively high pressure (Figure S2a). This suggests that the pores of this material are generally slit-like and/or mesoporous consisting of large pores. Their pore sizes were calculated from the N2 adsorption isotherm with the BJH method and the Kelvin model of pore filling.29 The results show bimodal pore size distribution curves (Figure 2c), consisting of monodisperse mesopores centered at ca. 3 nm and large pores with broad pore distribution ranging from ca. 10-80 nm. Its organoaminemodified counterparts also have bimodal pore distributions, but the size of the smaller pores of the particles is slightly decreased after amine modification, as expected, due to filling up of their pores with the organoamine groups (see inset in Figure 2C and also Figure S3 in Supporting Information). Similar decrease in pore size for the larger pores was, however, not evident due to the negligible length of the grafted organoamine groups compared with the dimension of the large pores of the materials. Thus, overall, the above results confirm that the desired organoamine groups are successfully grafted on the surfaces of KCC-1. In parallel, MCM-41 type MSNs with well-ordered mesoporous structure and their amine-functionalized counterparts were synthesized using methods reported previously.24 The TEM image in Figure S1 shows that MCM-41 possesses particles with sizes in the range of 400−600 nm in diameter, with onedimensional well-ordered cylindrical channel pores of 2−3 nm in diameter. MCM-NH2 was synthesized by allowing MCM-41 nanoparticles to react with APTMS. The presence of organoamine groups in MCM-NH2 is confirmed by TGA shown in Figure 2b. While MCM-41-CTAB (i.e., the surfactant-containing, as-synthesized MCM-41) shows 45.4% weight loss in the range of room temperature to 800 °C associated with the loss of CTAB, the calcined MCM-41 gives only ∼1.9% weight loss, indicating that the latter has no CTAB templates in it as expected. On the other hand, the amine-functionalized MCM-NH2 exhibits a higher weight loss (13.4%) compared with MCM-41, which is due to the loss of grafted organoamine groups on the former. Furthermore, elemental analysis shows the presence of substantial amount of nitrogen (1.46 mmol N/g) in MCM-NH2. The data obtained with nitrogen adsorption/desorption measurements, as compiled in Table S1, also indicate the successful modifications of MCM-41 with organoamine groups. Not surprisingly, while MCM-41-CTAB has a very small surface area (17 m2/g) and negligible pore volume (0.04 cm3/g), the calcined MCM-41 has very high surface area (1,086 m2/g) and large pore volume (0.82 cm3/g). These results indicate the
3. RESULTS AND DISCUSSION 3.1. Synthesis of Calcined KCC-1 and MCM-41 Materials with Different Organoamines and their Comparative DNA Adsorption Properties towards DNA. KCC-1 was synthesized according to previously reported microwave-assisted synthetic procedure,19 starting with KCCCPB and then removing its CPB templates via calcination. Its transmission electron microscopy (TEM) and scanning electron microscopy (SEM) images (Figure 1) show that KCC-1 is
Figure 1. (a) TEM and (b) SEM images of KCC-1 microspheres.
composed of reasonably uniform microspheres with an average diameter of ∼500 nm possessing fibrous, dendritic-like porous structures. Moreover, the SEM image, shown in Figure 1b, indicates that KCC-1 microspheres have very wide pore entrances that might be accessible for large-size biomolecules (vide infra). The surfaces of the KCC-1 microspheres were then modified with organoamine groups. Modification of silica surfaces with amine groups enables the material’s surfaces to interact better with negatively charged molecules such as DNAs and have improved adsorption capacity for such guest molecules. The surface modification was achieved by grafting two different aminorganosilanes having different organoamine groups: APTMS and AAPTMS. This resulted in KCC-1 microspheres possessing surface organoamine or organodiamine moieties, denoted here as KCC-NH2 and KCC-NH-NH2, respectively. The resulting microspheres were characterized by various methods. The removal of surfactant templates by calcination and the subsequent grafting of the organoamine groups on both types of materials are confirmed by TGA (Figure 2a). While KCC-CPB shows ∼28% of weight loss in the range from room temperature to 800 °C on TGA corresponding to the large amount of CPB surfactant in it, the calcined KCC-1 loses only ∼l.3% of weight, indicating its being almost free of surfactant templates. On the other hand, KCC-NH2 and KCC-NH-NH2 loses 12.5% and 17.4%, respectively, of their weights in the same temperature range, which is clearly because of the loss of the grafted E
dx.doi.org/10.1021/la501435a | Langmuir XXXX, XXX, XXX−XXX
Langmuir
Article
Figure 2. (a) Thermogravimetric analysis (TGA) curves of KCC-CPB, KCC-1, KCC-NH2, and KCC-NH-NH2. (b) TGA curves of MCM-41-CTAB, MCM-41, and MCM-NH2. (c) Pore size distribution of KCC-CPB, KCC-1, KCC-NH2, and KCC-NH-NH2. (d) Pore size distribution of MCM-41CTAB, MCM-41, and MCM-NH2. (Inset shows the change in pore size for smaller pores (< 5 nm) of KCC-1 nanoparticles after their pore walls are grafted with organoamine groups). (e,f) 29Si and 13C CP-MAS spectra, respectively, of KCC-NH2-Ext. (g,h) 29Si and 13C CP-MAS spectra, respectively, of MCM-41-NH2-Ext.
(986 m2/g) and lower pore volume (0.42 cm3/g). The nitrogen adsorption/desorption curves (Figure S2b) of all MCM-41based materials showed type IV isotherm and H1-type hysteresis
presence of highly porous structures in the calcined MCM-41. However, as expected, the grafting of organoamine groups on the channel walls of the material leads to slightly lower surface area F
dx.doi.org/10.1021/la501435a | Langmuir XXXX, XXX, XXX−XXX
Langmuir
Article
Figure 3. Adsorption of salmon DNA as a function of time over 48 h in (a) KCC-1, KCC-NH2, and KCC-NH-NH2 and (b) MCM-41 and MCM-NH2.
Over 48 h of incubation time, the amount of DNA adsorbed onto the different materials, in μg DNA/mg of material, is 47.3 ± 1.3 for MCM-NH2, 109.3 ± 2.3 for KCC-NH2, and 131.5 ± 4.0 for KCC-NH-NH2. This means, the KCC-1-based materials containing diamine groups show the highest adsorption capacity for salmon DNA. In addition, the KCC-1-based materials generally show better adsorption capacity for DNAs than their corresponding MCM-41-based counterparts. For example, despite its significantly low surface area, KCC-NH2 adsorb more than twice salmon DNA per unit mass than MCM-NH2. Upon normalizing the amount of DNA adsorbed per surface area of the material, the values in micrograms of DNA per square meter of material become 48.0 ± 1.3, 222.9 ± 4.7, and 334.4 ± 10.2 for MCM-NH2, KCC-NH2, and KCC-NH-NH2, respectively (Supporting Information Figure S4). These results indicate that all the amine-functionalized KCC-1-based materials have significantly higher adsorption capacity for DNA compared with their corresponding MCM-41 counterparts. In other words, despite their lower surface areas, KCC-1-based materials render better adsorption capacity for DNA than their corresponding MCM-41 type materials. Moreover, the results indicate that the organodiamine-functionalized KCC-1 material shows higher adsorption capacity than the corresponding monoaminefunctionalized KCC-1. This is presumably due to the relatively more favorable surface properties of the former in rendering more H-bonding interaction with DNA molecules as a result of its two amine groups per grafted moiety. The salmon DNA used in our study consist of macromolecules with linear structures composed of 20 to 250 base pairs and have an average length of 50 nm and a diameter of ∼2 nm.30 The MCM-NH2 nanoparticles, whose average pore size is 2.3 nm, are therefore not big enough to host these types of molecules. Moreover, once DNA molecules are docked at the channel pore openings of these particles via adsorption, the channel pores would be blocked and become hardly available for adsorption of additional DNAs, as illustrated in Scheme 2. In contrast, their larger pores, pore openings, and dendritic structures render KCC-NH2 more accessible space for the DNA molecules to diffuse in and adsorb, because the first DNA molecules in contact with the surfaces of the materials can not physically prevent additional DNA molecules from entering into the pores of these materials. 3.2. Synthesis of Solvent-Extracted KCC-1- and MCM41-Based Materials Possessing Amine Groups on Different Parts of their Surfaces and their DNA Adsorption Properties. By using the MCM-41- or KCC-1-based nanoma-
loop, confirming the presence of mesoporous structure in the materials. The pore size distributions of the materials, displayed in Figure 2d, indicate that MCM-NH2 has a smaller average pore diameter (2.3 nm) than MCM-41 (2.8 nm), which is obviously due to the presence of organoamine groups in the channel pores of the former. Furthermore, the presence of covalently bonded organoamine groups on the surfaces of both types of materials, KCC-1 and MCM-41, is confirmed with 29Si and 13C CP-MAS NMR spectroscopy (Figure 2e−2h). Besides the typical peaks corresponding to silica materials’ Q4 and Q3 peaks at approximately −110 and −100 ppm, respectively, the 29Si CPMAS NMR spectra (Figures 2e and 2g) shows T3 and T2 peaks centered at approximately −67.0 and −60.0 ppm, respectively, corresponding to silicon atoms attached with the organoamine groups grafted onto the KCC-1 or MCM-41 surfaces. The 13C CP-MAS NMR spectra, on the other hand, shows peaks centered at approximately δ = 9.0, 24.5, and 42.0 ppm after postgrafting, corresponding to (C1) Si−CH2−CH2−CH2−NH2, (C2) Si− CH2−CH2−CH2−NH2, and (C3) Si−CH2−CH2−CH2−NH2 carbons, respectively (Figures 2f and 2h). Given their organoamine groups, and more importantly, their large pore size and dendritic pore structure, KCC-1 type porous silica nanomaterials can be expected to be more advantageous over other porous silica nanomaterials with small, cylindrical pores, such as MCM-41, for hosting large-size drug molecules and bioactive substances (e.g., DNAs and proteins). This was actually the initial conjecture, which led us to prepare the series of KCC-1-based materials, as well as MCM-41-based materials for comparison, and to investigate their adsorption properties towards DNA. Accordingly, the materials (KCC-1, KCC-NH2, KCC-NH-NH2, MCM-41, and MCM-NH2) were mixed with solutions of double-stranded salmon DNA, and their adsorption kinetics was investigated and their adsorption capacity was determined. The plots of amount of DNA adsorbed by the materials versus time for different KCC-1-based materials are depicted in Figure 3. The results clearly demonstrate that the different materials have distinctive adsorption behaviors toward salmon DNA over 48 h. Generally, the amine-functionalized KCC-1 nanoparticles exhibit higher adsorption capacity for DNA than their nonfunctionalized counterparts. The nonfunctionalized KCC-1 and MCM-41 shows much less adsorption capacity for DNA, affording only
