Adsorption and Desorption Behaviors of DNA with Magnetic

ARTICLE pubs.acs.org/Langmuir

Adsorption and Desorption Behaviors of DNA with Magnetic Mesoporous Silica Nanoparticles Xu Li, Jixi Zhang, and Hongchen Gu* Nano Biomedical Research Center, Med-X Research Institute, Shanghai Jiao Tong University, 1954 Huashan Road, Shanghai, 200030, China

bS Supporting Information ABSTRACT: The interaction between DNA and mesopores is one of the basic concerns when mesoporous silica nanoparticle (MSN) is used as a DNA carrier. In this work, we have synthesized a type of mesoporous silica nanoparticle that has a Fe3O4 inner core and mesoporous silica shell. This magnetic mesoporous silica nanoparticle (denoted as M-MSN) offers us a convenient platform to manipulate the DNA adsorption and desorption processes as it can be easily separated from solution by applying a magnetic field. The DNA adsorption behavior is studied as a function of time in chaotropic salt solution. The maximum amount of adsorbed DNA is determined as high as 121.6 mg/g. We have also developed a method to separate the DNA adsorbed onto the external surface and into the mesopores by simply changing temperature windows. The desorption results suggest that, within the whole adsorbed DNA molecules, about 89.5% has been taken up by M-MSN mesopores. Through the dynamic light scattering experiment, we have found that the hydrodynamic size for M-MSN with DNA in its mesopores is higher than the naked M-MSN. Finally, the preliminary result of the adsorption mechanism study suggests that the DNA adsorption into mesopores may generate more intermolecular hydrogen bonds than those formed on the external surface.

I. INTRODUCTION After they were first reported by T. Kresge and co-workers in 1992,1 mesoporous silica materials have been intensively studied from synthesis to applications.2
4 Among the numerous explored application areas, a drug delivery system based on the mesoporous structure is an emerging area under development recently.5 This new application was triggered by the success of synthesis of mesoporous silica nanoparticles (MSNs) in the early 2000s.6,7 Compared to the former irregularly shaped or bulky mesoporous material, the MSN has a uniform and tunable particle size typically below 100 nm while retaining the various features such as the large surface area, the ordered pore structure, and the super high specific pore volume, which may allow well controllable loading and release of therapeutical medicines.8
10 The nanometer size of MSN makes the particle potentially suitable for mammalian cells' engulfment and systematic administration for in vivo applications.11 To date, many efforts have been devoted to exploring the possibilities of MSN as carriers for biomedical purposes, including drug delivery,12
14 enzyme and protein immobilization,15
17 bioseparation,18
20 and so on. It is worth noting that the guest molecules used in these current researches have been mostly focused on small molecules14 and proteins.16 Not enough attention has been paid to DNA, a key biomolecule, with regard to its ability lodge inside MSN. To explore the possibility of MSN as a gene delivery vehicle, researchers pretreated MSN with amino silane21 or cationic polymers.22
24 With amino functionalized surface, the r 2011 American Chemical Society

MSN could interact with DNA through electrostatic force. In further cell experiments,23,24 it was found that the DNA
MSN complex were efficient in gene transfection. Unfortunately, DNA adsorption only took place on the external surface of MSN in the above-mentioned studies, whereas the surface area inside mesopores was not utilized. To date, only a small amount of work has been done on DNA adsorption within mesopores. Fujiwara and co-workers25 used nitrogen sorption to estimate the pore volume of mesoporous materials before and after DNA adsorption and found that the pore volume of mesoporous silica remarkably decreased after DNA adsorption. They suggested this decrease should be owed to the occupancy of mesopores' inner space by DNA molecules. However, for DNA in mesopores, the decrease of pore volume might not be sufficient because it might also be caused by the residual ions in mesopores. Solberg and Landry26 observed fluorescently labeled DNA within the pores of their mesoporous silica materials through confocal microscopy. This method tried to observe DNA in mesopores directly. Nevertheless, the resolution of optical microscopy was limited at the micrometer level, thus making it impossible to get clear photographs of nanomaterials. Gao and co-workers27 conducted methylation treatment on the external surface of a surfactantcontaining mesoporous material, followed by functionalization Received: November 23, 2010 Revised: March 27, 2011 Published: April 13, 2011 6099

dx.doi.org/10.1021/la104653s | Langmuir 2011, 27, 6099–6106

Langmuir

ARTICLE

Figure 1. (A) TEM micrograph for M-MSN. The bar represented 100 nm. (B) Nitrogen sorption Isotherms for M-MSN. The insert showed the pore size distribution plot for MSN.

of the silanols on the internal surface with aminopropyl groups upon removing the template. Finally, this mesoporous material could bind DNA effectively. The result clearly demonstrated that the binding process only occurred within the mesopores. Although this evidence was convincible for DNA adsorption in mesopores, the studying method only applied to the mesoporous materials with large pore size (20 nm) and was not suitable for those with smaller pore size. In summary, although the foregoing studies realize the adsorption for DNA within MSN mesopores, the proofs are still insufficient. The interaction between DNA and MSN is far from clear. Hence, it is necessary to find additional evidence about DNA adsorption within mesopores. In this work, we synthesize a type of magnetic mesoporous silica nanoparticle (denoted as M-MSN) with a Fe3O4 core and mesoporous silica shell. With M-MSN, the study on the hydrodynamic size and the behaviors of DNA adsorption and desorption for M-MSN clearly demonstrate that DNA is taken up into M-MSN mesopores. We also develop a method to distinguish the DNA adsorbed onto the surface of M-MSN and DNA into the mesopores by simply changing temperature windows. The mechanism of DNA adsorption into M-MSN is also analyzed based on our experimental data.

2. EXPERIMENTAL SECTION Materials. The mesoporous silica nanoparticle (MSN) used throughout this work was a type of composite nanosphere with a core
shell structure. The core was a single Fe3O4 nanoparticle and was covered by a silica layer with radical mesoporous structure. Because of the superparamagnetic property of Fe3O4 nanoparticles, we denoted this composite particle as magnetic mesoporous silica nanoparticle (M-MSN). Because of the successful entrapment of a single magnetic nanoparticle inside the mesoporous shell, the M-MSN could be separated from solution by applied magnetic field instead of centrifugation, which provided a convenient platform to study the DNA adsorption and desorption behaviors of mesoporous materials. Double-stranded salmon DNA was purchased from Sigma and used as supplied. As stated by the manufacturer, the DNA was cleaved into short segments by sonication and extracted in saturated NaCl solution. The length of DNA was determined by 0.7% agarose gel electrophoresis (shown lanes 4 and 5 of Figure S1 of the Supporting Information). The salmon DNA had a distribution from 20 to 250 bp with an average chain length of ∼50 nm.

The binding buffer was a commercial kit used for DNA adsorption on silica surface, which was purchased from Shanghai Allrun Nano New Material Technology Company. The major ingredient of this buffer was 4 M chaotropic salt (Guanidine Hydrochloride) with pH 5.2. The elution buffer was 150 mM NaCl solution. All of these buffer solutions were autoclaved at 120 °C for 2 h before use. Synthesis of M-MSN. M-MSN was synthesized according to the reported literature by Hyeon’s group.28 In summary, uniform Fe3O4 nanocrystals were synthesized through the traditional coprecipitation method and stabilized with oleic acid as reported by our groups.29 Then, the magnetic nanoparticles were transferred to an aqueous phase using cationic surfactant CTAB (purchased from Sigma). A typical formation process of mesoporous silica coating was performed by self-assembly of the surfactant template CTAB and silica precursor TEOS (also purchased from Sigma) in basic solution. Finally, the template was removed to generate mesopores by a commonly used extraction procedure using acetone as solvent.

Characterization of M-MSN and DNA Concentration Measurement. M-MSN was washed by acetone and then dried in vacuum

at 70 °C for 4 h to get a deep-yellow powder. Nitrogen adsorption and desorption isotherms were obtained on ASAP2010 analyzer (Micromeritcs, USA). Surface area and pore size distribution were calculated by BET and BJH methods, respectively. Samples of M-MSN for transmission electron microscopy (TEM) 2100 (JEOL, JAPAN) were dispersed in ethanol and vortexed for 30s, and then the suspended solution was transferred to carbon-coated copper grids. The DNA concentration in solution was determined by using NANODROP 1000 spectrophotometer (Thermo Scientific, USA) and the final result of concentration was an average of duplicate samples and was calibrated by the standardization plot method which has been reported in our previous work.35 DNA Adsorption into M-MSN. Adsorption isotherm was obtained by preparing a series of salmon DNA solutions with initial concentrations ranging from 50 to 2000 μg/mL in ddH2O at pH 7. In each experiment, 100 μL of the DNA solution was added into a 2 mL centrifuge tube with 0.5 mg M-MSN and 100 μL ddH2O prepared before, and then 200 μL of binding buffer was added. The mixture was well-dispersed by vortex for 30 s and then was continuously shaken with 270 rpm at 25 °C for 20 h. The final solution was placed under magnetic field for 2 min to separate M-MSN and supernatant liquid absolutely. The amount of DNA adsorbed by M-MSN was calculated from the differences of DNA concentration in solutions before and after the adsorption process. Adsorption dynamic curve was obtained through measuring DNA concentration in supernatant at premeditated time points from 2 to 1200 min after mixing M-MSN with DNA solution. The adsorption experimental 6100

dx.doi.org/10.1021/la104653s |Langmuir 2011, 27, 6099–6106

Langmuir

ARTICLE

Figure 2. (A) Descending curve for salmon DNA concentration in binding solution changing with time. The concentration data were measured at premeditated moments in one adsorption process at 25 °C. (B) Ascending adsorption isotherm for salmon DNA into M-MSN as a function of DNA equilibrium concentration at 25 °C. A Langmuir fit was used for the data (solid line). method was the same as demonstrated above with the initial concentration at 467.8 μg/mL. DNA Desorption from M-MSN. All of the M-MSN samples were saturatedly adsorbed with DNA before desorption study throughout this work. After adsorption for 20 h, the M-MSN sample was separated and washed with ethanol twice to ensure that residual chaotropic salts were removed. Then the M-MSN sample was directly suspended in 400 μL elution buffer in a 2 mL centrifuge tube and incubated at certain temperatures immediately. To obtain the DNA desorption dynamic curves, we measured the DNA concentration in supernatant at different time points after the M-MSN sample mixing with the elution buffer. The whole desorption process would continue until the variation of measured DNA concentration in supernatant became negligible. Determination of Hydrodynamic Size. The hydrodynamic size was determined by Dynamic lighter Scattering (DLS) and measured with high-performance particle sizer (HPPS, Malvern Instrument). The samples included naked M-MSN, M-MSN (after DNA adsorption), M-MSN (after desorption under 0.5 °C), and M-MSN (after desorption under 37 °C) with the concentration of M-MSN about 20 μg/mL. Every sample was dissolved in ethanol and underwent sonication for 5 min before measured. The measurement was carried out at 20 °C.

3. RESULT AND DISCUSSION Characterization of the M-MSN. We have synthesized M-MSN and used various methods to characterize these nanoparticles. From the TEM micrograph of M-MSN (part A of Figure 1), the average diameter of M-MSN was determined to be 70 ( 20 nm. The wormholelike mesopores with diameters of around 2
3 nm were observed in the silica shells. The magnetic core inside the mesoporous silica matrix had an average diameter of ∼15 nm. Part B of Figure 1 showed the nitrogen sorption isotherm. The mean pore size was calculated to be 2.7 nm, as shown in the insert, which was similar with that recorded by TEM. The obvious nitrogen condensation step at p/p0 = 0.25
0.4 and the extremely narrow distribution indicated that the M-MSN had a well-developed pore structure. At a relative pressure of 0.9 < p/p0 < 1, a steep increase occurred and a desorption hysteresis loop appeared, which demonstrated another large pore with the average pore size around 25 nm (in the insert of part B of Figure 1). This large pore should be owed to the interparticle spaces among the nanoparticles, which had also been confirmed by the other researchers.33,34 The BET surface area was as high as 696.29 m2/g and the pore volume was 1.72 cm3/g.

Adsorption and Desorption Behaviors of DNA with M-MSN. To determine the DNA adsorption equilibrium time

of M-MSN, we plotted the adsorption dynamic curve for M-MSN. Part A of Figure 2 described the decline of DNA concentration in adsorption supernatant over time. The initial concentration of DNA in binding solution was 467.8 μg/mL. In the first 20 min, the DNA concentration underwent a remarkable decrease, and then this decrease slowed down sharply. In 600 min, the changes of DNA concentration in the supernatant became negligible, suggesting the DNA adsorption had reached equilibrium. Therefore, we adopted the adsorption time of 1200 min for isotherm experiments to ensure DNA adsorption to reach equilibrium. Interestingly, the time for DNA adsorption into M-MSN to reach equilibrium was much longer than the time for DNA adsorption onto a common silica nanoparticle surface to reach equilibrium in our previous experiments35 (within 20 min). Part B of Figure 2 showed the DNA adsorption isotherm of M-MSN. These concentration data fitted well with Langmuir adsorption model, written as Γ = ΓmaxKLCs/(1þ KLCs), where Γ was the amount of DNA adsorbed on to the silica, Γmax was the maximum adsorbed capacity of this material, Cs was the equilibrium concentration of DNA in solution, and KL was the Langmuir constant. From this curve, the Γmax was determined to be 121.6 mg DNA/g M-MSN. To compare the adsorption capacity of M-MSN with common silica nanoparticles, we synthesized a type of silica nanoparticle through the Stober method36 with a diameter of 80 ( 15 nm (denoted as S_80 and the corresponding TEM micrograph was shown in Figure S2 of the Supporting Information), which was close to the diameter of our M-MSN. Then we determined the DNA adsorption capacity of S_80 under the same adsorption solution conditions. The adsorption isotherm was plotted in part A of Figure S3 (Supporting Information). From the Langmuir fit model, the maximum adsorption capacity for S_80 was calculated to be 48.3 mg/g. Obviously, M-MSN had a much higher DNA adsorption capacity than the common silica nanoparticle. Because M-MSN and S_80 had a similar particle size, the large difference in DNA loading capacity should be attributed to the existence of the mesoporous structure in M-MSN. We also investigated the adsorption capacity of M-MSN with CTAB templates in the mesopores. The adsorption isotherm was shown in part B of Figure S3 (Supporting Information) and the maximum adsorption capacity was calculated to be 7.6 mg/g. As the mesopores were filled with CTAB templates, the adsorbed 6101

dx.doi.org/10.1021/la104653s |Langmuir 2011, 27, 6099–6106

Langmuir

ARTICLE

Figure 3. (A) Desorption dynamic curves for salmon DNA from M-MSN at 0.5 °C (circle) and 37 °C (solid circle), respectively. For the DNA desorption dynamic curves at 37 °C, the M-MSN sample with adsorbed DNA first underwent desorption step at 0.5 °C several times until the DNA concentration in elution buffer was nearly zero. Then the M-MSN was put into fresh elution buffer at 37 °C to plot the desorption dynamic curve. (B) The accumulated DNA concentration in elution buffer at different temperatures.

DNA should exist on the M-MSN external surface. This result further confirmed that the important role of mesopores in DNA adsorption with M-MSN. After adsorption, we chose the sample of DNA-loaded M-MSN with the adsorbed amount around 90 mg/g (close to saturated adsorption) to investigate the desorption behaviors. Interestingly, the DNA desorption behaviors for M-MSN was strongly temperature dependent. We first put M-MSN sample in to elution buffer under 0.5 °C, the DNA concentration in elution buffer fluctuated slightly near 7.6 μg/mL during the whole desorption process (data shown as a circle in part A of Figure 3). Then we separated the M-MSN samples out by magnet and put them into fresh elution buffer under the same temperature. The above process was carried out repeatedly until the DNA concentration in the desorption supernatant was nearly zero. The total amount of desorbed DNA under 0.5 °C was calculated to be 8.7 mg/g. After desorption under 0.5 °C, the M-MSN sample was put into fresh elution buffer at 37 °C to plot the desorption dynamic curve (data shown as a solid circle in part A of Figure 3). The DNA concentration grew up gradually and the desorption process took nearly 60 min to get to equilibrium. The final equilibrium concentration in solution was ∼90 μg/mL and the amount of desorbed DNA at 37 °C was determined to be 71.1 mg/g, which was 89.5% of the whole adsorbed amount. As majority of DNA was still adsorbed in M-MSN in the desorption step at 0.5 °C, we carried out a desorption experiment under different temperatures to obtain the temperature window for the release of this part of DNA. After desorption several times in elution buffer at 0.5 °C, the desorbed temperature was raised up to 5 °C in 2 min and the desorption proceeded for 60 min. The accumulated DNA concentration in the solution was measured before temperature increased. This process was repeated by increasing temperature 5 °C in each round until the final desorption temperature was raised to 45 °C (the results were shown in part B of Figure 3). We noticed that DNA concentration was negligible in solution before the temperature rises to 15 °C, whereas the DNA concentration increased sharply and steadily from 15 to 40 °C. Finally, the DNA concentration almost did not change from 40 to 45 °C.

The above different desorption behaviors thus divided the adsorbed DNA in M-MSN into two parts. One part of DNA would be released out at low temperature (0.5 °C) easily (we denoted this part of DNA as the LT-DNA). In comparison, the other part of DNA would still be adsorbed on M-MSN at 0.5 °C but released out at higher temperature (at least 15 °C) gradually (we denoted it as the HT-DNA). The phenomenon could probably be interpreted as the different barriers within the DNA desorption processes from the external surface or from the mesopores. Interestingly, almost all of the adsorbed DNA on S_80 could be released out under 0.5 °C (data not shown), of which the desorption behavior was the same as the LT-DNA in M-MSN. According to the above results, it seemed that the LT-DNA should be the part of DNA on external surface and the HT-DNA should be the part of DNA in mesopores. Considering the existent status of DNA adsorption on the M-MSN external surface, the DNA adhered by side-on conformation to generate hydrogen bonds, which was the same as the manner of DNA adsorption on pure silica nanoparticle surface. Consequently, we measured the hydrodynamic sizes of S_80 with different statuses in ethanol. Before DNA adsorption, the hydrodynamic size of naked S_80 was about 139.8 nm (part A of Figure 4). After DNA adsorption (part B of Figure 4), the hydrodynamic size increased to 150 nm, which was a little larger than the size before DNA adsorption. Then, after DNA desorption (part C of Figure 4), the hydrodynamic size dropped to about 135 nm. The corresponding sketches of this nanoparticle were shown in parts D, E, and F respectively of Figure 4. Comparatively, we also measured the hydrodynamic sizes of M-MSN in ethanol. As shown in part A of Figure 5, before DNA adsorption, the hydrodynamic size of naked M-MSN was around 130 nm. Then, after DNA adsorption (part B of Figure 5), it was noticeable that the hydrodynamic size increased sharply to about 220 nm. After desorption under 0.5 °C to eliminate the LT-DNA (part C of Figure 5), the hydrodynamic size of M-MSN still maintained above 220 nm. Then after release out of the residual HT-DNA under 37 °C (part D of Figure 5), the hydrodynamic size of M-MSN decreased back to about 130 nm. The sketches of M-MSN under those statuses were depicted from part E of Figure 5 to part H of Figure 5. 6102

dx.doi.org/10.1021/la104653s |Langmuir 2011, 27, 6099–6106

Langmuir

ARTICLE

Figure 4. The hydrodynamic distribution of pure silica nanoparticle S_80 in ethanol: (A) before DNA adsorption; (B) after DNA adsorption; (C) after DNA desorption. The hydrodynamic size for them was calculated to be 139.8, 149.6, and 134.7 nm. The sketches in (D) to (F) described the statues of M-MSN for (A) to (C), respectively. The white circle represented the silica nanoparticle. The double helix bars in (E) represented the DNA molecules.

As the DNA on M-MSN external surface only slightly increased the hydrodynamic size, there was no doubt that the intensive increase of hydrodynamic size should be attributed to the DNA adsorption into mesopores. As the elimination of LTDNA only slightly influenced the hydrodynamic size, we confirmed that the HT-DNA should be the part of DNA entrapped in the mesopores. Considering the average chain length of DNA was calculated to be ∼50 nm (Figure S1 of the Supporting Information) and the depth of mesopore channel was about 20
30 nm (part A of Figure 1), it was quite possible that only part of the DNA chain was inset into the pore channel with the

other part stretched outside like a tail. This tail might also be the reason for the great increase of hydrodynamic size of M-MSN after DNA adsorption. A sectional drawing for DNA in M-MSN mesopores was shown in Figure 6. Mechanism for DNA Adsorption into Mesopores. Under common conditions, the negative charges on DNA and M-MSN surfaces prohibit them from getting close. In the mechanism study for DNA adsorption onto silica surface in chaotropic salt solution, Melzak and co-workers30 generally categorized the main driving force for adsorption into three parts: the high concentration of ions decrease the Debye Length in binding solution,37which effectively 6103

dx.doi.org/10.1021/la104653s |Langmuir 2011, 27, 6099–6106

Langmuir

ARTICLE

Figure 5. Hydrodynamic distribution of M-MSN in ethanol: (A) before DNA adsorption, (B) after DNA adsorption, (C) after DNA desorption under 0.5 °C, and (D) after DNA desorption under 37 °C. The hydrodynamic size for them was calculated to be 131.3, 223.4, 236.6, and 137.6 nm, respectively. The sketches in (E) to (H) described the statues of M-MSN for (A) to (D), respectively. The black circle represented the magnetic core of M-MSN and the white circle with heterogeneous surface represented the silica shell of M-MSN. The double helix bars in (F) and (G) represented the DNA molecules. 6104

dx.doi.org/10.1021/la104653s |Langmuir 2011, 27, 6099–6106

Langmuir

Figure 6. (A) Sectional drawing for existent status of DNA in M-MSN mesopores (the empty mesopores were omitted and the DNA chain inside or outside the mesopore was not depicted accurately). Arrow a represented the magnetic core, arrow b represented the mesopore, arrow c represented DNA molecule. (B) An enlarged image for DNA molecule in mesopore.

shields the negative charges and intensively weakens the repulsive electrostatic force between DNA and silica; the guanidine cations can strongly capture the water molecules, which offers the dehydration effect on the DNA and silica surface; a large number of phosphorus and silanol groups, especially the geminal silanol groups38
39 (the pKa for the deprotonation step of the first one of the silanol group is about 840), are protonated under this acidic condition (such as at pH 5), which is propitious to the generation of intermolecular hydrogen bonds. To sum up, the main driving forces for DNA adsorption onto silica surface are shielded intermolecular electrostatic force, dehydration effect, and intermolecular hydrogen bonds. The process of DNA adsorption into M-MSN is also controlled by those driving forces thermodynamically. In addition, considering the mesoporous structure of M-MSN, the diffusion kinetics and the generation of intermolecular hydrogen bonds should be different for DNA on external surface and internal surface. According to the classic worm-chain model,31 the linear duplex DNA molecule is usually semistiff and the chain will be absolutely rigid in its persistent length. According to the work of Sobel and Harpst,32 in 1M-3 M salt solution (which is similar with our adsorption condition), the persistent length of DNA is around 40 nm, thus, the salmon DNA used in our case (chain length is about 50 nm) can hardly bend and may behave as a very short rod. As diffusion begins, driven by the thermodynamic driving force, the DNA molecules meet the appropriate pore entrances randomly. Because the mean pore size of M-MSN is 2.7 nm, which is only a little larger than the DNA diameter. Obviously, the small entrance area will decrease the chance of DNA contacting with pore entrance. As diffusion proceeds, both the available DNA molecules in solution and the empty pores in M-MSN are decreasing, which make the contact chance get lower further and prolong the diffusion time to reach equilibrium. From this discussion, it seems that the entrance of longer chain DNA inside mesopores will be more difficult not only because of the low diffusivity but also the wormlike DNA chain conformation. This conclusion matches well with our result about the calf thymus DNA (20 kbp, lane 2 in Figure S1 of the Supporting Information) adsorption into M-MSN, whose maximum adsorption capacity is only ∼22 mg/g (Figure S4 of the Supporting Information). In the situation of DNA binding onto the external surface of M-MSN, not all of its phosphate groups can participate in hydrogen bonds' generation between DNA and the external surface because of

ARTICLE

the double helix structure. DNA in B form usually has 10 bases in every helix turn,31 which means only one or two phosphate groups per turn can make contact with the silica surface; the others are too far to form hydrogen bonds. However, for DNA binding in M-MSN mesopores, both DNA and mesopores have a cylinder-like structure and the pore size in M-MSN is only slightly larger than the DNA diameter. Consequently, it is reasonable that every phosphorus group on DNA gets closer to silanol groups on the internal surface of mesopores and all of these phosphorus groups are available in the formation of hydrogen bonds. According to our DNA desorption results in parts A and B of Figure 3, the DNA molecules released out from mesopores need a higher temperature to overcome the energy barrier in the desorption process. This barrier should be related to the intermolecular hydrogen bonds between DNA and the silica surface. The more hydrogen bonds generated in the DNA molecules, the higher the energy barrier exists in the DNA desorption process and a corresponding higher temperature is required to break these bonds. Thus, it seems that DNA may generate more hydrogen bonds in mesopores than on the external surface. We notice that Fujiwara et.al25 has also focused on the status of hydrogen bonds during DNA adsorption within mesopores. Our study on the DNA adsorption and desorption behaviors with M-MSN have provided more evidence.

4. CONCLUSIONS We have synthesized M-MSN as a platform, which can be easily manipulated to the study of carrier properties of mesoporous silica nanoparticles. On the basis of this platform, adsorption of short salmon DNA into M-MSN has been carried out in chaotropic salt solution and the maximum adsorption capacity was determined to be about 121.6 mg DNA/g M-MSN. We have developed an easy temperature windows method by manipulating the desorption temperatures to distinguish the adsorbed DNA on the M-MSN external surface or in its mesopores and further confirmed that about 89.5% of the adsorbed DNA was entrapped inside the mesopores. Through the dynamic light scattering experiment, we have found that the hydrodynamic size for M-MSN with DNA in its mesopores was higher than the naked M-MSN. To our knowledge, our work was the first study to give clear evidence for DNA adsorption into the mesopores by studying the behaviors of the adsorption and desorption processes. The preliminary result of the adsorption mechanism study suggested that the DNA adsorption into mesopores might generate more intermolecular hydrogen bonds than on an external surface and the short DNA molecule diffused into mesopores was much easier than the long DNA molecule. ’ ASSOCIATED CONTENT

bS

Supporting Information. Agarose gel photograph, TEM image, and adsorption isotherm figures. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Phone: þ86-21-62933176, e-mail: [email protected].

’ ACKNOWLEDGMENT This work was supported by National 863 High-Tech Program (2009AA03Z333), National Science and Technology Ministry, International Cooperation Project of P. R. China (2008DFA51860), 6105

dx.doi.org/10.1021/la104653s |Langmuir 2011, 27, 6099–6106

Langmuir Shanghai Nano Project (1052 nm01100), Shanghai STC Program (09DZ1904000,1052 nm01100), and SJTU funding (YG2009ZD203). The authors would like to thank the Instrumental Analysis Center of Shanghai Jiao Tong University for the characterization of materials. Especially thanks for Mrs. Ying Dai for graphic assistance and Dr. Weiliang Xia for word processing assistance.

’ REFERENCES (1) Kresge, C. T.; Leonowicz, M. E.; Roth, W. J.; Vartuli, J. C.; Beck, J. S. Nature 1992, 359, 710. (2) Zhao, D.; Feng, J.; Huo, Q.; Melosh, N.; Fredrickson, G. H.; Chmelka, B. F.; Stucky, G. D. Science 1998, 279, 548–552. (3) Zhao, X. S.; (Max), Lu, G. Q.; Millar, G. J. Ind. Eng. Chem. Res. 1996, 35, 2075–2090. (4) Díaz, J. F.; Balkus, K. J., Jr. J. Mol. Catal. B: Enzym. 1996, 2, 115–126. (5) Vallet-Regi, M.; Ramila, A.; Del Real, R. P.; Perez-Pariente, J. Chem. Mater. 2001, 13, 308–311. (6) Nooney, R. I.; Thirunavukkarasu, D.; Yimei, C.; Josephs, R.; Ostafin, A. E. Chem. Mater. 2002, 14, 4721–4728. (7) Huh, S.; Wiench, J. W.; Yoo, J.-C.; Pruski, M.; Lin, V.S.-Y. Chem. Mater. 2003, 15, 4247–4256. (8) Slowing, I. I.; Vivero-Escoto, J. L.; Wu, C.-W.; Lin, V.S.-Y. Adv. Drug Delivery Rev. 2008, 60, 1278–1288. (9) Lai, C.-Y.; Trewyn, B. G.; Jeftinija, D. M.; Jeftinija, K.; Xu, S.; Jeftinija, S.; Lin, V.S.-Y. J. Am. Chem. Soc. 2003, 125, 4451–4459. (10) Sakamoto, T. W.; Kim, R.; Ryoo, O. T. Angew. Chem., Int. Ed. 2004, 43, 5231–5234. (11) Vallet-Regí, M.; Balas, F.; Arcos, D. Angew. Chem., Int. Ed. 2007, 46, 7548–7558. (12) Trewyn, B. G.; Giri, S.; Slowing, I. I.; Lin, V.S.-Y. Chem. Commun. 2007, 3236–3245. (13) Liong, M.; Lu, J.; Kovochich, M.; Xia, T.; Ruehm, S. G.; Nel, A. E.; Tamanoi, F.; Zink, J. I. ACS Nano 2008, 2, 889–896. (14) Trewyn, B. G.; Slowing, I. I.; Giri, S.; Chen, H.-T.; Lin, V.S.-Y. Acc. Chem. Res. 2007, 40, 846–853. (15) Park, H. S.; Kim, C. W.; Lee, H. J.; Choi, J. H.; Lee, S. G.; Yun, Y. P.; Kwon, I. C.; Lee, S. J.; Jeong, S. Y.; Lee, S. C. Nanotechnology 2010, 21, 225101. (16) Slowing, I. I.; Trewyn, B. G.; Lin, V.S.-Y. J. Am. Chem. Soc. 2007, 129, 8845–8849. (17) Wang, Y.; Caruso, F. Chem. Commun. 2004, 10, 1528–1529. (18) Sen, T.; Sebastianelli, A.; Bruce, I. J. J. Am. Chem. Soc. 2006, 128, 7130–7131. .; Boissiere, C. C. R. Chimie 2005, 8, 579–596. (19) Prouzet, E (20) Lee, H. S.; Chang, J. H. Proc. of SPIE 2009, 7270, 72701B–4. (21) Torney, F.; Trewyn, B. G.; Lin, V.S.-Y.; Wang, K. Nat. Nanotechnol. 2007, 2, 295–300. (22) Radu, D. R.; Lai, C.-Y.; Jeftinija, K.; Rowe, E. W.; Jeftinija, S.; Lin, V.S.-Y. J. Am. Chem. Soc. 2004, 126, 13216–13217. (23) Xia, T.; Kovochich, M.; Liong, M.; Meng, H.; Kabehie, S.; George, S.; Zink, J. I.; Nel, A. E. ACS Nano 2009, 3, 3273–3286. (24) Qin, F.; Zhou, Y.; Shi, J.; Zhang, Y. J. Biomed. Mater. Res. A 2009, 90, 333–338. (25) Fujiwara, M.; Yamamoto, F.; Okamoto, K.; Shiokawa, K.; Nomura, R. Anal. Chem. 2005, 77, 8138–8145. (26) Solberg, S. M.; Landry, C. C. J. Phys. Chem. B 2006, 110, 15261–15268. (27) Gao, F.; Botella, P.; Corma, A.; Blesa, J.; Dong, L. J. Phys. Chem. B 2009, 113, 1796–1804. (28) Kim, J.; Kim, H. S.; Lee, N.; Kim, T.; Kim, H.; Yu, T.; Song, I. C.; Moon, W. K.; Hyeon, T. Angew. Chem., Int. Ed. 2008, 47, 8438–8441. (29) Zhang, L.; He, R.; Gu, H. C. Appl. Surf. Sci. 2006, 253, 2611–2617.

ARTICLE

(30) Melzak, K. A.; Sherwood, C. S.; Turner, R. F. B.; Haynes, C. A. J. Colloid Interface Sci. 1996, 181, 635–644. (31) Lehninger, A.; Nelson, D. L.; Cox, M. M. Lehninger Principles of Biochemistry; W. H. Freeman: New York, 2008. (32) Sobel, E. S.; Harpst, J. A. Biopolymers 1991, 31, 1559–1564. (33) Suzuki, K.; Ikari, K.; Imai, H. J. Am. Chem. Soc. 2004, 126, 462–463. (34) Cai, Q.; Luo, Z. S.; Pang, W. Q.; Fan, Y. W.; Chen, X. H.; Cui, F. Z. Chem. Mater. 2001, 13, 258–263. (35) Qian, M.; Li, X.; Gu, H. C. Chinese J. Anal. Chem. 2010, 38, 963–967. (36) Stober, W.; Fink, A.; Bohn, E. J. Colloid Interface Sci. 1968, 26, 62–69. (37) Russel, W. B.; Saville, D. A.; Schowalter, W. R. Colloidal Dispersions; Cambridge University Press, 1989. (38) Luhmer, M.; D’Espinose, J. B.; Hommel, H.; Legrand, A. P. Magn. Reson. Imaging 1996, 14, 911–913. (39) Braga, P. R. S.; Costa, A. A.; De MacEdo, J. L.; Ghesti, G. F.; De Souza, M. P.; Dias, J. A.; Dias, S. C. L. Micropor. Mesopor. Mat. 2011, 139, 74–80. (40) Rosenholm., J. M.; Linden, M. J. Controlled Release 2008, 128, 157–164.

6106

dx.doi.org/10.1021/la104653s |Langmuir 2011, 27, 6099–6106