Preparation and Characterization of Imogolite/DNAHybrid HydrogelsNattha Jiravanichanun,† Kazuya Yamamoto,† Kenichi Kato,‡ Jungeun Kim,‡ Shin Horiuchi,§

Weng-On Yah,∥ Hideyuki Otsuka,†,∥ and Atsushi Takahara*,†,∥

†Institute for Materials Chemistry and Engineering, Kyushu University, 744 Motooka, Nishi-ku, Fukuoka 819-0395, Japan‡Japan Synchrotron Radiation Research Institute, 1-1-1 Kouto, Sayo-cho, Sayo-gun, Hyogo 679-5198, Japan§Nanotechnology Research Institute, National Institute of Advanced Industrial Science and Technology (AIST), 1-1-1 Higashi,Tsukuba, Ibaraki 305-8565, Japan∥Graduate School of Engineering, Kyushu University, 744 Motooka, Nishi-ku, Fukuoka 819-0395, Japan

ABSTRACT: Imogolite is one of the clay minerals containedin volcanic ash soils. The novel hybrid hydrogels were preparedfrom imogolite nanofibers and DNA by utilizing stronginteraction between the aluminol groups on imogolite surfaceand phosphate groups of DNA. The hybrid hydrogels ofimogolite and DNA were prepared in various feed ratios, andtheir physicochemical properties and molecular aggregationstates were investigated in both dispersion and gel states. Themaximum DNA content in the hybrid gels was shown inequivalent molar ratio of imogolite and DNA. The physicalproperties of the hybrid gels were changed by varying DNA blend ratios. In the dispersion state, the hybrid gels showed a fibrousstructure of imogolite, whereas a continuous network structure was observed in pure imogolite, indicating that the hybrid with DNAenhanced the dispersion of imogolite. In the gel state, DNA and imogolite nanofibers formed a 3D network structure.

1. INTRODUCTIONAluminol (AlOH) groups have been known to have greataffinity for phosphonic acid groups. The adsorption and bindingof nucleic acid and clay minerals has attracted great interestbecause the complexes provide protection against the degrada-tion by nucleases but maintain their biological activity such asgene delivery.1−4 The interaction of clays such as montmor-illonite with important biological polymers such as DNA andRNA has been investigated.5−7 However, the study on the inter-action between imogolite and DNA has not yet been reported.Imogolite is one of the important clay minerals contained in

volcanic ash soils.8,9 Imogolite is a hydrous aluminosilicatenanofiber having outer surface of AlOH groups. The diameter ofthe nanofiber is ∼2 nm. The AlOH groups on the outer surfaceof imogolite are pH-dependent, possessing positive charge atlow pH and negative charge at high pH. Point of zero charge(pzc) for imogolite was reported to be about pH 6−9.10Additionally, imogolite can also be synthesized chemically fromtetraethoxysilane and aluminum chloride.11−13 Because of theseunique physicochemical properties, imogolite is categorized asan interesting nanomaterial. The self-assembly of polyelectrolytehomopolymers with imogolite by solution mixing was reported.14

Hydrogen bonding and ionic interactions between imogolite andpolyelectrolyte contribute to the self-assembly.Double-helix DNA molecule has diameter of ∼2 nm similar

to imogolite nanofiber. Both nanomaterials can be expected toexhibit good compatibility in aqueous media. Importantly,

phosphate groups on the outside of DNA double helix havestrong interaction with aluminol groups on the imogolitesurface. However, DNA could not be inserted into the innerpore of imogolite because the inner diameter of imogolite is ∼1 nm,which is smaller than that of DNA. These reasons stimulatedour interest in the combination of DNA and imogolite.Furthermore, this hybridization can be performed without surfacemodification; therefore, the preparation process is very simple.In the present paper, we report preparation of the imogolite/DNAhybrid gels and investigation of their characteristics. The optimumadsorption of DNA on imogolite was determined, and the massratio of imogolite-to-DNA in the hybrid gels was examined. Toreveal the structure−property relationships of the hybrid gels,dynamic viscoelastic measurement, transmission electron micro-scopy (TEM) observation, and wide-angle X-ray diffraction(WAXD) measurement were performed.

2. EXPERIMENTAL SECTIONMaterials. DNA (double strand) from herring testes in sodium

salt form was purchased from Sigma Aldrich. The molecular weightwas polydisperse between 400 and 1000 base pairs (bp) with center ofdistribution at ca. 700 bp. Imogolite was synthesized according toa previously reported method.13 HCl, NaHCO3, acetate buffer, trisbuffer, and NaHCO3/NaOH buffer were prepared as buffer solutions

Received: November 16, 2011Revised: December 13, 2011Published: December 13, 2011

Article

pubs.acs.org/Biomac

© 2011 American Chemical Society 276 dx.doi.org/10.1021/bm201616m | Biomacromolecules 2012, 13, 276−281

of pH 4−11. All other reagents were chemical grade and used asreceived.Preparation of Hybrid Gels in Varying Imogolite/DNA Blend

Ratio. Hybrid gels of imogolite and DNA were prepared by mixing1 mg/mL solutions of imogolite and DNA, which were prepared andadjusted to pH 4 by diluted HCl (Scheme 1). The several hybrid gels

were prepared by varying DNA to imogolite feed ratio, as listed inTable 1, which shows the concentration of imogolite and DNA in wt %.Solvent contents were the same for all samples. Adsorption of DNA onimogolite was rapidly observed visually as gel formation. The mixture wasfurther shaken at 37 °C overnight. Then, all mixtures were centrifugedto separate the hybrid gels and supernatants. The collected hybrid gelswere washed with water by centrifugation at 3000 rpm and repeatedthree times until no more DNA was desorbed.Characterization of Adsorbed DNA in Hybrid Gels. The DNA

concentration in supernatants was measured by a UV−vis spectro-meter, Lamda 35 (Perkin-Elmer Japan) at a wavelength region of190 to 450 nm. DNA absorbs UV-light at λmax 260 nm. The remainedDNA in supernatant was calculated based on the initial amount ofDNA. The collected hybrid gels were freeze-dried and weighed tocalculate % yield based on the initial weight of imogolite and DNA.FT-IR measurement of the hybrid gels was carried out as KBr pelletby a Spectrum 100 spectrometer (PerkinElmer Japan). The amount ofbound DNA on imogolite was estimated by the following experiment.The known-weight gels were dissolved in concentrated HCl for1 week, neutralized with saturated NaHCO3, integrated volume, andmeasured DNA amount by UV−vis spectroscopy. The DNA content(wt %) in the hybrid gel was calculated by dividing the DNA amountin HCl solution by the initial weight of DNA. The measurement foreach sample was repeated three times for confirmation.Water Content and Swelling Degree. Water content in the

hybrid gel was estimated by drying the collected gel and was cal-culating by subtracting the weight of dried gel from the initial weightof wet gel. A swelling experiment was done by equilibrating the freeze-dried gel in water for 24 h. Then, swelling degrees (Q) were estimated

according to the equation. Q = (Wwet − Wdry)/Wdry, where Wwet andWdry are the weights of the wet gel and dried gel, respectively.

Desorption of DNA from Hybrid Gels. The protection ofDNA by the hybrid gel was tested by incubated the hybrid gel undervarious conditions such as different pH, salt concentration, andtemperature. Acetate buffer, tris buffer, and NaHCO3/NaOH bufferwere applied for pH 4, 7−9, and 10−11, respectively. The freeze-driedhybrid gels having DNA content of 70 wt % were incubated undergiven buffer solution, salt concentration, and temperature. Theamounts of DNA release from the hybrid gels were monitored byUV−vis spectroscopy.

Morphology and Structure of Hybrid Gels. Transmissionelectron microscopic (TEM) images were observed with H-7500(Hitachi High-Technologies Corporation, Tokyo, Japan) at 100 kV ofacceleration voltage and 10 μA of beam current. Wide-angle X-raydiffraction (WAXD) measurements were carried out at roomtemperature on a large Debye−Scherrer camera at the BL02B2beamline of SPring-8. The incident beam from the bending magnetwas monochromatized to λ = 0.1 nm. The dried powder samples werepacked into glass capillary tubes with an outer diameter of 0.5 mm, andthe tubes were rotated during the measurements. The data werecollected in a 2θ range from 1 to 75° with a step interval of 0.01°. Therheological properties of hybrid gels were measured with MCR101rheometer. A plate with diameter 12 mm was applied, and measuredtemperature was controlled at 25 °C. Each sample was measured threetimes to get an average value.

Scheme 1. Schematic Representation for the Preparation ofHybrid Gel from Imogolite and DNA

Table 1. Concentration of Imogolite and DNA Feeds for theHybrid Gel Preparation

gel no. 1 2 3 4 5 6 7 8

DNA content (wt %) 9 17 29 38 50 63 71 83Imogolite content (wt %) 91 83 71 62 50 37 29 17

Figure 1. Amount of DNA remained in supernatant (blue plots) andyield of the obtained hybrid gels (green plots) as a function of DNAfeed content.

Figure 2. IR spectra of imogolite, DNA, and the hybrid gels variedDNA/imogolite feed ratios.

Biomacromolecules Article

dx.doi.org/10.1021/bm201616m | Biomacromolecules 2012, 13, 276−281277

3. RESULTS AND DISCUSSION

3.1. Preparation of DNA/Imogolite Hybrid Gels. Theresidual in supernatant represents the fraction of unbound DNA.The fraction of remained DNA in supernatants with varyingDNA/imogolite feed ratio is shown in Figure 1 (blue plots). WhenDNA feed content was lower than 50 wt %, DNA moleculeswere completely immobilized on imogolite to form the hybrid gels.In contrast, when DNA feed content was higher than 50 wt %,the amount of unbound DNA started to increase sharply. Theyield of the obtained imogolite/DNA hybrid gels is shown inFigure 1 (green plots). The amount of the obtained hybrid gelsincreased as increasing DNA feed content. The maximum amountof the hybrid gel was obtained at DNA to imogolite feed ratio63/37 (gel 6). Above this point, the amount of obtained gelsdecreased because the amount of imogolite decreased.Figure 2 shows FT-IR spectra of the hybrid gels with varied

DNA/imogolite feed ratio. The characteristic peaks of imogolitewere observed at 930 and 960 cm−1 corresponding to thestretching of Al−O−Si groups. The prominent bands at 1090,1236, and 1700 cm−1 correspond to PO2

− and P−O groups inDNA. These results confirmed the existence of imogolite andDNA in the hybrid gels. When DNA feed content increased,there is an increase in the intensity band of P−O and PO2

−

stretching. The FT-IR spectra provided an average quantitativeDNA amount in the hybrid gels.The bound DNA into imogolite in the hybrid gel was estimated

as shown in Figure 3a. UV−vis spectra of DNA adsorption foreach hybrid gel are shown in Figure 3b. Increase in the DNA feedcontent led to an increase of DNA content in the hybrid gels. Gel7 showed the highest DNA content in the hybrid gel (75%),which was made from DNA/imogolite feed ratio of 71/29. DNA/Imogolite weight ratio at 75/25 could calculate to be 1:1 molarratio. Above this point, DNA content decreased, which may becaused by electrostatic repulsion generated by the excessivenegative charges of DNA hindered the formation of hybrid gels.

Figure 3. (a) DNA content in the hybrid gels and (b) normalizedUV−vis absorption spectra showing an increase in DNA content in thehybrid gel against DNA feed content.

Figure 4. Photographs of the hybrid gels 1, 3, 5, and 7; (a) front view and (b) up view and their model structures. DNA/imogolite feed ratio for eachgels is shown in Table 1.

Biomacromolecules Article

dx.doi.org/10.1021/bm201616m | Biomacromolecules 2012, 13, 276−281278

3.2. Physical Properties of DNA/Imogolite HybridGels. The aggregation occurred when imogolite and DNAsolutions were mixed. The initial state of the obtained gels aftercollection by centrifugation exhibited a difference in physicalproperties depending on DNA feed content, as shown in Figure 4.Gels 1, 3, 5, and 7 were prepared by varied DNA to imogolitefeed ratio as shown in Table 1.The frequency dependence of dynamic viscoelasticity of the

hybrid gels is shown in Figure 5. Gels 1 and 3 were gel-like andhigh viscosity, which did not fall down even upturn the vials.Lower modulus in gel 1 than gel 3 is probably due to the lowerdensity of cross-linking point in hybrid network. Gels 5 and 7were fluid-like and lower viscosity than gels 1 and 3. This isbecause of the incompleteness of network structure. Figure 4bshows photographs of the gel surface (up side view), and the

smoothness of gels was ordered from roughness to smoothness;gels 3 > 1 > 5 > 7. The different amount of DNA in the hybridgels influenced the gel viscosity and physical properties. Thealuminol groups of imogolite can be both positively andnegatively charged depending on the pH in the aqueousenvironment. In the cases of gels 1 and 3, which have higherimogolite content, the entanglement of imogolite themselves andimogolite with DNA molecules like a spontaneous aggregationmight lead to highly viscous gels. In gels 5 and 7, which havehigher DNA content, coverage of imogolite−DNA networksurface by DNA molecules induced repulsion between smallaggregations and led to low viscous liquid-like gels.

Figure 5. (a) Storage, (b) loss moduli, and (c) complex viscosity(G′, G″, and η*, respectively) as a function of frequency for the hybridgels 1 (black), 3 (red), 5 (blue), and 7 (green).

Figure 6. Releasing of DNA from the hybrid gels; (a) varying pH7−11 (at 37 °C), (b) varying NaCl concentration (at pH 9, 37 °C),and (c) varying temperature (at pH 9).

Biomacromolecules Article

dx.doi.org/10.1021/bm201616m | Biomacromolecules 2012, 13, 276−281279

Water content of the hybrid gels was higher than 99%.However, once the hybrid gels were dried, their swelling degreewas not so high. The swelling degrees (Q) for gels 1, 3, 5, and 7with increasing DNA feed ratio were 14.7, 35.8, 15.2, and 12.3,respectively. The dried hybrid gels could adsorb water only10 to 40 times by weight. The drying process led to the strongaggregation, which made it difficult to redisperse the imogolitenanofibers. Gel 3 showed the highest swelling degree probablydue to the presence of effective cross-linking points. The chargeof imogolite surface is dependent on pH and has point of zerocharge (pzc) about pH 6−8.2 Theoretically, DNA is expected torelease from imogolite at pH higher than pzc because AlOH2

+

groups are deprotonated. From the experimental results, DNAwas not released from the hybrid gel at pH 4 because the surfaceof imogolite was positively charged. Figure 6a shows DNArelease in the buffer with pH of 7, 7.5, 8, 9, 10, and 11 monitoredfor 24 h. DNA release amount increased as pH increased.However, the DNA release amount was only 12% even at pH 11.This suggests that DNA exhibits a stronger interaction withimogolite. Figure 6b shows that the DNA release amountdecreased as NaCl concentration increased to 0.1 M. The resultsare opposed to the phenomena of salt-induced desorption ofDNA complexes with polycations.15 Probably, the added saltwould dissociate surface charge around the hybrid gel instead ofattacking the gel. These results indicate that the adsorption ofDNA on imogolite protected DNA from attack of salt ions.Figure 6c shows that DNA release amount decreased at lower

temperature. Upon temperature decrease, thermal fluctuationscontinuously reduced; therefore, DNA motion decreased, leadingto lower amount of DNA release. Under severe conditions suchas strong alkaline solutions or NaCl solutions, the hybridizationwith imogolite provided enough protection to DNA molecules.This is probably due to the very strong chemical interactionssuch as electrostatic interaction between aluminol groups ofimogolite and phosphate groups of DNA. The other reason isthat an entanglement of DNA molecules and imogolitenanofibers provided physical protection from the added salt oralkaline solution.

3.3. Molecular Aggregation States of DNA/ImogoliteHybrid Gels. The morphology of freeze-dried imogolite andthe hybrid gel was observed by TEM, as shown in Figure 7.Figure 7a shows imogolite bundles forming continuous networkstructure. Interestingly, imogolite presented different morphol-ogy in the hybrid gel (Figure 7b). TEM image of the hybridgel showed two characteristic of imogolite bundles. One is thefibrous morphology of imogolite bundles and another is anentanglement between imogolite fibers. First, a good dispersionof imogolite fibers is suggested so that DNA molecules wraparound a single imogolite nanofiber; this leads to a better fiberdispersion than that of pure imogolite. When DNA moleculesbridged between imogolite fibers, imogolite fibers were curledand entangled forming web-like structures. Good affinity ofAlOH groups of imogolite and phosphate functional groups ofDNA is a driving force for DNA adsorption on imogolite. Therewas no significant difference in morphology for all hybrid gels.WAXD measurement of freeze-dried imogolite, DNA, and the

hybrid gels was carried out by SR-powder diffractometer atBL02B2 in SPring-8. Figure 8 shows WAXD profiles of freeze-dried hybrid hydrogels. The WAXD pattern of DNA showed onebroad range at d spacing 5−30 nm. Imogolite exhibited peaks atd spacing 1.53, 0.92, and 0.66 nm assigned to the diffraction fromindividual tube16 and a peak at 2.2 nm associated with center-to-center tube distance of imogolite bundles.17 Two hybrid gelsshowed the WAXD pattern similar to imogolite, but there aresome differences at q = 2−5 and 11−18 nm−1. The shift of peakat q = 2−5 nm−1 region might be suggestive of distortion ofimogolite packing upon binding by DNA. This result agreed withobservation on TEM images. The higher intensity of the peak atq = 11−18 nm−1 indicated the difference of DNA content in thehybrid gels. The gel 7 with DNA feed content of 70 wt % clearlyshowed a higher intensity compared with other patterns.

■ CONCLUSIONSThe imogolite/DNA hybrid gels were successfully prepared bya simple mixing method, utilizing aluminol functionality ofimogolite to bind to phosphate ester of DNA. Varying imogoliteand DNA feed ratio could control morphology and physicalproperties of the hybrid gels. Additionally, DNA could beprotected even under severe conditions by the hybridizationwith imogolite. The structures of the imogolite/DNA hybridgels were investigated using methods such as WAXD and TEM.In a dried state, TEM images of the hybrid gels showed fibrousimogolite morphology, which differed from a continuousnetwork structure in pure imogolite. Also, WAXD results wereconsistent with TEM observation. In the present Article, wehave demonstrated the specific interaction between imogolite andDNA for the first time. The results obtained here are importantto be the stage for further works to investigate the detailedproperties such as rheology and to develop the imogolite/DNAhybrid gels for various biomaterial applications.

Figure 8. WAXD profiles of freeze-dried imogolite, DNA, and thehybrid gels.

Figure 7. TEM images of (a) imogolite and (b) the hybrid gel.

Biomacromolecules Article

dx.doi.org/10.1021/bm201616m | Biomacromolecules 2012, 13, 276−281280

■ AUTHOR INFORMATIONCorresponding Author*E-mail: takahara@cstf.kyushu-u.ac.jp. Fax: +81-92-802-2518.Tel: +81-92-802-2517.

■ ACKNOWLEDGMENTSWe gratefully acknowledge the financial support of a Grant-in-Aid for Scientific Research (No. 19205031) and a Grant-in-Aidfor the Global COE Program, “Science for Future MolecularSystem” from the Ministry of Education, Culture, Science, Sports,and Technology of Japan. The synchrotron radiation experimentswere performed at BL02B2 in SPring-8 with the approval of theJASRI (Proposal No. 2010A1454).

■ REFERENCES(1) Lorenz, M. G.; Wackernagel, W. Appl. Environ. Microbiol. 1987,53, 2948−2952.(2) Lorenz, M. G.; Wackernagel, W.Microbiol. Rev. 1994, 58, 563−602.(3) Kawase, M.; Hayashi, Y.; Kinoshita, F.; Yamato, E.; Miyazaki, J.;Yamakawa, J.; Ishida, T.; Tamura, M.; Yagi, K. Biol. Pharm. Bull. 2004,27, 2049−2051.(4) Lina, F.-H.; Chena, C.-H.; Cheng, W. T. K.; Kuoc, T.-F. Biomaterials2006, 27, 3333−3338.(5) Alvarez, A. J.; Khanna, M.; Toranzos, G. A.; Stotzky, G. Mol. Ecol.1998, 7, 775−778.(6) Franchi, M.; Bramanti, E.; Bonzi, L. M.; Orioli, P. L.; Vettori, C.;Gallori, E. Origins Life Evol. Biospheres 1999, 29, 297−315.(7) Chen, B. Br. Ceram. Trans. 2004, 103, 241−249.(8) Yoshinaga, N.; Aomine, S. Soil Sci. Plant Nutr. 1962, 8, 6−13.(9) Yamamoto, K.; Otsuka, H.; Takahara, A. Polym. J. 2007, 39, 1−15.(10) Tsuchida, H.; Ooi, S.; Nakaishi, K.; Adachi, Y. Colloids Surf., B2005, 45, 131−135.(11) Farmer, V. C.; Fraser, A. R.; Tait, J. M. J. Chem. Soc., Chem.Commun. 1977, 13, 462−463.(12) Wada, S.-I.; Eto, A.; Wada, K. J. Soil Sci. 1979, 30, 347−355.(13) Yamamoto, K.; Otsuka, H.; Wada, S.-I.; Sohn, D.; Takahara, A.Soft Matter 2005, 1, 372−377.(14) Yang, H.; Chen, Y.; Su, Z. Chem. Mater. 2007, 19, 3087.(15) Dubas, S. T.; Schienoff, J. B. Macromolecules 2001, 34, 3736−3740.(16) Wada, S.-I.; Eto, S.; Wada, K. J. Soil Sci. 1979, 30, 347−355.(17) Farmer, V. C.; Adams, M. J.; Fraser, A. R.; Pamieri, F. ClayMiner. 1983, 18, 459−472.

Biomacromolecules Article

dx.doi.org/10.1021/bm201616m | Biomacromolecules 2012, 13, 276−281281