- Email: [email protected]
Polypeptide binding to mesostructured titania films
Microporous and Mesoporous Materials 142 (2011) 1–6
Contents lists available at ScienceDirect
Microporous and Mesoporous Materials journal homepage: www.elsevier.com/locate/micromeso
Polypeptide binding to mesostructured titania films Stefania Mura a, Gianfranco Greppi b, Anna Maria Roggio c, Luca Malfatti d, Plinio Innocenzi d,⇑ a
Dipartimento di Scienze Molecolari e Agroalimentari, Università di Milano, via Celoria 2, 20133 Milano, Italy Dipartimento di Scienze Zootecniche, Università di Sassari, via De Nicola 9, 07100 Sassari, Italy c Porto Conte Ricerche, strada provinciale 55 Porto Conte/Capo Caccia, km 8.400 località Tramariglio, 07041 Alghero (SS), Italy d Materials Science and Nanotechnology Laboratory, D.A.P., CR-INSTM, Università di Sassari, Palazzo Pou Salit, Piazza Duomo 6, 07041 Alghero (SS), Italy b
a r t i c l e
i n f o
Article history: Received 18 January 2010 Received in revised form 21 September 2010 Accepted 31 October 2010 Available online 4 November 2010 Keywords: Self-assembly Mesoporous Films Peptide
a b s t r a c t Mesoporous titania films have been prepared via evaporation induced self-assembly and used as a matrix to bind linear pentapeptides. The mesoporous surface has been functionalized with amine groups by immersing the samples in a solution of aminopropyltriethoxysilane in toluene. Different pentapeptides have been synthesized for binding to the mesoporous films: H–Ile-Gln-Asp-Leu-Phe–COOH, H–ValGln-Asp-Leu-Phe–COOH and Fmoc-Phg-Gln-Asp-Leu-Phe–COOH. The peptides have been bonded to the amine functionalized surface of the titania mesoporous films with an impregnation process. The H–Val-Gln-Asp-Leu-Phe–COOH peptide has been successfully bonded to the titania matrix, while the other peptides have shown not be suitable for the process because of lower solubility and sterical hindrance. The functionalization with aminopropyltriethoxysilane and peptide binding has been studied by Fourier transform infrared spectroscopy and fluorescence spectroscopy. A fluorescent marker, fluorescein isothiocyanate, has been used to confirm the incorporation of the peptides into the titania matrix. The process is reliable and robust, after several washing cycle of the samples, the peptides are still well bonded to the titania mesoporous films. Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction Mesostructured transition metal oxide films have a variety of potential applications as magnetic, photocatalysis [1], photovoltaic materials [2], and for the incorporation of other molecules by physisorption [3]. Titania mesoporous films exhibit different mesostructures, pore symmetry and connectivity that can be tailored using different synthesis strategies [4,5]. A great number of active sites (or functions) can be concentrated on the large pore surface area, enabling their possible applications as biosensors [6,7]. Titania mesostructured films can be produced by evaporation induced self-assembly [8] (EISA) using alcoholic solutions of metal chlorides as the inorganic source and ionic surfactants or block copolymers as the structure-directing agents [9]. In particular, the high surface area, chemical durability and optical transparency suggest that titania mesoporous films could be used as support for bioanalytical applications and for biosensors; nanoporous TiO2 films have been used for protein immobilization because of the good biocompatibility of this material [10]. Mesoporous materials have appeared, in fact, as a feasible matrix for encapsulation of proteins and peptides immobilization [11–13]. The main advantage of mesoporous titania films is the high hydrolytic stability in comparison ⇑ Corresponding author. Tel.: +39 079 998630; fax: +39 079 9720420. E-mail address: [email protected] (P. Innocenzi). 1387-1811/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.micromeso.2010.10.047
with mesoporous silica samples which have a severe limit for applications in biotechnologies due to their very low stability in water and physiological solutions [14]. In the present work, we have used mesoporous titania films as supports for the incorporation of biomolecules such as peptides; for this purpose TiO2 films have been modified with a functional precursor that leaves an amino-terminal group on the pore surface for the linking with the bio-element (peptide). In general, functionalization of mesoporous films with organic groups can be obtained using two main approaches: co-condensation with a functional precursor during film formation, or post-functionalization of the pore surface after film deposition. Each route has certain advantages: co-condensation allows a homogeneous distribution of organic groups over the pore surfaces, the post-grafting method presents more protruding (thus more accessible) functions and better defined pore sizes [5]. Moreover, it is difficult to add organic groups by co-condensation in non-silica frameworks because of the instability of most G–Ti bonding (G = grafting group) towards hydrolysis in the very acidic synthesis condition used for films preparation [1] while postfunctionalization ensures that molecules are grafted to the pore surface rather than incorporated within the oxide framework [1]. Films obtained by post-grafting show functional groups, for instance amines, only on the pore surface whereas films obtained by co-condensation should present functional groups also within the pore walls besides those present on the pore surfaces. This is the
2
S. Mura et al. / Microporous and Mesoporous Materials 142 (2011) 1–6
reason why using post-functionalization of mesoporous materials more amine groups are potentially available for reaction than in the co-condensed films [4,5]. 3-Aminopropyltriethoxysilane (APTES) is commonly used as coupling agent for the modification of silica surfaces to promote protein adhesion, cell growth, to attach metal nanoparticles to silica substrates, for biological implants and in lab on chip applications [15]. The hydrolysis and condensation of silanes drives the bonding of APTES to the substrate via formation of siloxane bonds on the surface [16]. The initial hydrolysis step can occur either in solution or at the substrate surface depending on the amount of water present in the system. An overabundance of water will result in excessive polycondensation in the solvent phase, while a deficiency of water will result in the formation of an incomplete monolayer [9]; solvent, concentration, reaction time, and reaction temperature are other parameters that affect the grafting kinetics [9]. In titania materials the functionalization via APTES is realized by condensation of Ti–OH with Si–OH to form mixed Ti–O–Si bonds [17]. Silanization experiments with APTES, to add amino functional groups for the simple immobilization of biomolecules, have been performed to elucidate their use in protein chips and biosensors [11]. In the present work, we have fabricated a titania mesostructured support to immobilize a peptide with potentially selective properties with respect to dioxins for the future development of a biosensor. Dioxin is a toxic chemical species in the environment that must be strictly monitored; it is mainly generated from waste incineration or pesticide.1 The method to determine the accurate environmental concentration of dioxin is chemical analysis by high-resolution gas chromatography mass spectrometry (HR-GC/MS) [13]. Recently, simple and rapid methods, for instance a gene assay and ELISA, using biological species for capturing dioxins have been accepted as official and legal methods to detect dioxins instead of time-consuming GC/MS. Antibodies or receptors, which consist of natural amino acids have the potential to recognize dioxin molecules [11,13]. However, denaturation of the antibodies caused by the presence of organic solvents necessary to dissolve hydrophobic dioxin in aqueous solution is a serious problem in the use of antibodies in dioxin assays. If the concentration of organic solvents in the assay solution is lowered, high concentrations of dioxin cannot be dissolved, which affects the precision of the analysis [13]. If we could use a synthetic oligopeptide instead of a natural immuno antibody, the problem of denaturation in organic solvent would be solved. Moreover, quality control procedures for a synthetic peptide would be easier than for a natural antibody and regeneration and reuse by washing out bound target would be feasible [13]. In previous studies, peptides that specifically bind dioxin have been screened from a combinatorial pentapeptide library and two sequences, Phe-Leu-Asp-Gln-Ile and Phe-Leu-Asp-Gln-Val, have been obtained [18]. Their design is based on the binding pocket of aryl-hydrocarbon receptor or CDR domain of monoclonal anti-dioxin antibody. In addition, Phe-Leu-Asp-Gln-Phg (5 Phg), the peptide substituted at the fifth position by a phenyl glycine (Phg) with better performance than the original, has been discovered from a one-amino-acid substituted library containing synthetic peptides [13]. In the present work, we have tried to develop a suitable mesoporous titania film
1 The term dioxin refers to a class of structurally and chemically related halogenated aromatic hydrocarbons that includes polychlorinated dibenzodioxins (PCDDs or dioxins), polychlorinated dibenzofurans (PCDFs or furans) and the ‘‘dioxinlike’’ polychlorinated biphenyls (PCBs). Because of their chemistry, dioxins are both toxic and persistent in the environment. Dioxins and furans are included in the United Nations Environment Programme ‘‘Dirty Dozen’’. In this paper, the term dioxin is used in a toxicological meaning to designate the PCDDs, the PCDFs and the coplanar (‘‘dioxin-like’’) PCBs, since these classes of compounds show the same type of toxicity. Because of the large number of congeners, relevant individual congeners are assigned with a toxic equivalency factor (TEF) that relate their toxicity to that of tetrachlorodibenzo-p-dioxin (TCDD) (2,3,7,8-TCDD) and are to be evaluated as dioxins.
to immobilize the peptides via a post-synthesis functionalization grafting route. 2. Experimental section 2.1. Chemicals All commercially available solvents and reagents were used without further purification. TiCl4 (reagent grade, >98%), anhydrous ethanol (reagent grade, >99.9%), bidistilled water, acetone (reagent grade, >99.8%), toluene (reagent grade, >99.5%), were purchased from Carlo Erba. Pluronic F-127 (cell culture test), 3-(aminopropyl)triethoxysilane (reagent grade, >98%), were purchased from Sigma–Aldrich. N,N0 -dimethylformamide (puriss. p.a., >99.8%) (DMF), 1-methyl-2-pyrrolidone (for peptide synthesis, >99.8%) (NMP), dichloromethane (reagent grade, >99.5%) (DCM), 2-propanol (reagent grade, >99.5%), trifluoroacetic acid (reagent grade, >98%) (TFA), piperidine (reagent grade, >98%), triisopropylsilane (99%) (TIS) and N-[(dimethylamino)-1H-1,2,3-triazolo[4,5-b]pyridin-1-ylmethylene]-N-methylmethanaminium hexafluorophosphate N-oxide (99%) (HATU) were purchased from Sigma–Aldrich. Fmoc-Pal-Peg resin (substitution 0.21 mmol g1) was obtained from Applied Biosystems (Applied Biosystems Inc., USA). The Fmoc-amino acids and 1-hydroxybenzotriazole hydrate (HOBt) were purchased from Novabiochem (Novabiochem, Switzerland). a-Cyano-4-hydroxycynnamic acid from Fluka was used as matrix for MALDI TOF experiments. All solvents used in the chromatography section were of the gradient grade for HPLC, and were purchased from Merck-VWR (D-Darmstadt, Germany). Fluorescein isothiocyanate (FITC), isomer I (90% purity) was purchased from ACROS organics and used without further purification. 2.2. Synthesis of peptides Three different linear pentapeptides have been synthesized: H–Ile-Gln-Asp-Leu-Phe–COOH, H–Val-Gln-Asp-Leu-Phe–COOH and Fmoc-Phg-Gln-Asp-Leu-Phe–COOH. The detailed description of the peptides synthesis is reported in the Supplementary data. 2.3. Synthesis of mesoporous titania films The precursor solution was prepared in different steps. In the first one Pluronic F127 was dissolved in ethanol; the solution was stirred for 2 h at 25 °C in a closed vessel. TiCl4 was slowly added to the ethanolic solution of Pluronic F127, under stirring for 20 min in an ice bath. Finally water was added to this mixture that was left to react under stirring for 3 h at 25 °C in a closed vessel. The addition of water causes the hydrolysis of TiCl4 producing EtOH and HCl, responsible of the high stability of the sol that is stable for several months at 25 °C. The molar ratios in solution were TiCl4:Pluronic F127:H2O:EtOH = 1:0.005:10:40. Titania thin films were deposited by dip-coating silicon wafers in the precursor solution at 25 °C and at a withdrawal rate of 15 cm min1. The relative humidity (RH) inside the dip-coater chamber was maintained between 18% and 25% to obtain good optical quality and high structural order in the titania films. The as deposited films were aged at room temperature (RH = 50%) for 24 h. To increase the inorganic polycondensation and stabilize the mesophase the films were submitted to different firing steps: 60, 120 and 200 °C for 24 h at each temperature in an oven with heating rate of 10 °C min1. This treatment is important to allow the formation of highly organized, high surface area mesoporous coatings, exhibiting crystalline frameworks. The final calcination process to remove the organic template of these stabilized coatings was done at 350 °C for 3.5 h in air under static conditions with a heating rate
S. Mura et al. / Microporous and Mesoporous Materials 142 (2011) 1–6
of 10 °C min1. Calcined mesoporous films at 350 °C were used as the matrix for incorporation of the pentapeptides. The functionalization process of titania films with amine groups was studied in different solvents (EtOH, toluene) and at different temperatures (20–80 °C), for different concentrations of APTES (0.002–0.2 M) and for different times (1–24 h). Based on these results we have selected the following protocol: immersion of 350 °C treated film in a 0.2 M solution of APTES in toluene for 24 h under stirring at 25 °C. The amine grafted films were carefully washed by toluene with several washing cycles and finally dried in air under static conditions. To identify the presence of the peptide in the film by fluorescence spectroscopy, we prepared a solution of fluorescein isothiocyanate (FITC) in dimethyl sulfoxide (DMSO) at the concentration of 5 mg ml1. We dissolved 1 mg of peptide in 1 ml of toluene, immediately after 20 lL of FITC–DMSO were added to this solution. Carbonate buffer 0.1 M and sodium carbonate 0.1 M solutions were also tested as possible solvents, but the samples immersed in these solutions were not able to give fluorescence and were not further used. The FITC–DMSO–toluene solution was left under stirring for 3 h; after this step the solution was diluted by adding 20 ml of toluene and used to immerse the titania films functionalized with APTES. The samples were left in the solution for 24 h in the dark; after this step the films were washed several times with fresh ethanol and dried in air. 2.4. Film characterization Fourier transform infrared (FTIR) analysis was performed using a Bruker Vertex 70v spectrophotometer. The optical bench and the sample compartment were kept in vacuum during the measurement at pressure lower than 0.5 hPa. The measurements in the middle infrared (MIR) region were performed using a Globar source, a KBr beamsplitter, and a RT-DLaTGS detector averaging 256 scans with 4 cm1 of resolution. The measurements in the far infrared range were done using a Globar source, a Si beamsplitter, and a RT-DTGS-FIR detector. The spectra were recorded in transmission, in the 600–100 cm1 range by averaging 32 scans with 4 cm1 of resolution. A silicon wafer was used as the substrate to measure the background; the baseline was calculated by a rubberband algorithm (OPUS 7 software). Fluorescence analysis was done using a FluoroMax-3 Horiba Jobin Yvon spectrofluorometer. The probing beam was set to impinge on one side of the sample (silicon substrate, incidence angle of 2–3°) so that the sample acted as a waveguide for the incident light wave, while the luminescence was collected at 90° with respect to the incident beam. This configuration enhanced the signal-to-noise ratio and avoided reflection effects. Each acquisition is the average of three different accumulations. An excitation wavelength of kex = 490 nm was used for acquiring the emission spectra. The organization of the porous structure was investigated by 2D grazing incidence small angle X-ray scattering (GISAXS) at the Austrian SAXS beamline of ELETTRA synchrotron facility (Trieste, Italy). An incident energy of 8 keV (wavelength 1.54 Å) was used; the instrumental glancing angle between the incident radiation and the sample was set slightly above the critical angle (grazing incidence). A two-dimensional CCD detector (Photonic Science, UK) was used to acquire the scattering patterns; each measurement consisted typically on the average of 10 acquisitions with integration time of 6 s. Pore arrangements were also studied by transmission electron microscopy (TEM). TEM micrographs were obtained in bright field mode on a JEOL 200CX microscope equipped with a tungsten cathode operating at 200 kV. Finely ground films scratched from the substrate were dispersed in n-octane by sonication, and then they were dropped on a carbon-coated copper grid and dried for TEM
3
observations. Center-to-center interpore distance was evaluated by line profile analysis on a set of representative TEM images as the average FWHM of the intensity distribution along a line passing through the pore centers. Film thicknesses and refractive indexes were measured by spectroscopic ellipsometry using an a-SE instrument (J.A. Woollam, USA) working in the 400–850 nm range.
3. Discussion and results 3.1. Mesostructured titania film synthesis To obtain mesostructured titania films we have used a well established and highly reproducible synthesis that has been reported by Soler-Illia et al. in previous publications [7,8]. This synthesis allow getting organized mesostructures after removal of the templating micelles via thermal calcination; after firing the films still present high organization, excellent optical quality and hydrolytic stability. The organization of the porous structures on films has been studied by GISAXS; Fig. 1 shows the typical pattern of an ordered mesoporous titania film after thermal treatment at 350 °C. The pattern appears composed of three peaks that can be 0) reflections of a body centered indexed as (1 1 0), (1 0 1), and (1 1 in the space group) structure with domains preferencubic (Im3m tially oriented with the [1 1 0] direction normal to the surface, and allowed by the circular permutation [19]. The lower intensity of the (1 1 0) peak is due to the incident angle that reduces the quantity of domains having their in-plane diffraction in the Bragg conditions. As a consequence of the thermal treatment, the cubic unit cell is contracted in the out-of-plane [1 1 0] direction (normal to the substrate), which will be referred to as d110, whereas the inplane (parallel to the substrate) cell constant will be referred to as d110 . Following this attribution, the lattice parameters have been calculated to be d110 = 7.0 ± 0.1 nm and d110 = 3.1 ± 0.1 nm. The synthesis protocol has shown to be very robust and highly reproducible; we have performed systematic GISAXS analysis on different batches of samples (more than 20 samples have been analyzed by GISAXS) prepared in various times and we have always observed cubic organization of the porous structure [17]. A direct observation of the pore arrangement has been obtained by transmission electron microscopy to support the information provided by SAXS data. Bright field TEM images were collected
Fig. 1. GISAXS pattern of an ordered mesoporous titania film treated at 350 °C.
4
S. Mura et al. / Microporous and Mesoporous Materials 142 (2011) 1–6
Fig. 2. Bright-field TEM images of mesoporous titania films treated at 350 °C.
Absorbance
0.08 0.06 0.04 0.02 0.00 4000
3750
3500
3250
3000
2750
2500
Wavenumber / cm-1 Fig. 3. FTIR absorption spectra in the 4000–2500 cm1 interval of as deposited (solid line) and calcined (dash line) mesostructured titania films; calcination has been done in air at 350 °C.
for samples fired at 350 °C. Fig. 2 shows representative direct images of the deposited films with some projections of a slightly distorted cubic structure. Even if an unambiguous identification would requires cross-sectional TEM, according to the literature, we have attributed this images to the (1 1 0) projection [8]. Under these conditions, in fact, the {1 1 0} face appears as a sequence of channels because the depth of field of the electron microscopy merges several spherical pore planes. Starting from this consideration, the distance among pores has been calculated by a line profile analysis on representative TEM images. The center-to-center interpore distance was estimated to be 8.5 ± 1 nm. The effect of the mesoporous film functionalization was also followed by spectroscopic ellipsometry. The data were obtained assuming a fitting model based on the assumption of transparent films on silicon (Cauchy dispersion relation).2 After calcination at 350 °C in air the titania films show an average thickness of 187 ± 5 nm with a refractive index of 1.72 measured at k = 632.8 nm. After functionalization with APTES the film thickness slightly increases to a value of around 218 ± 37 nm with a significant changes of the refractive index to a value of 1.90. Finally, after peptide binding, the films reach a thickness of around 243 ± 30 nm with an average refractive index of 1.79. This value, which is lower than that measured on the amino-functionalized films, can be attributed to the formation of an overlayer with a low refractive index made of peptides bonded to mesoporous film surface. Fig. 3 shows the FTIR absorption spectra in the 4000–2500 cm1 interval of as deposited (solid line) and calcined (dash line)
0.10 0.08
Absorbance
0.10
mesostructured titania films; calcination has been done in air at 350 °C. The spectra of the as deposited film show the presence of a wide intense band peaking around 3250 cm1 which is assigned to O–H stretching mode, due to Ti–OH species and absorbed water (overlapped band around 3200 cm1) [20]. The presence of the organic template in the as deposited film is indicated by the C–H stretching bands in the 3000–2800 cm1 interval [21]. After calcination of the film at 350 °C the spectrum appears as a flat curve with only a very small signal around 2900 cm1; this indicates that the thermal treatment removes almost completely the surfactant and at the same time promotes the densification of the titania network, through condensation reactions of Ti–OH species. We have also checked the effect of the thermal treatment on the structure of the titania pore walls; in the as deposited films the titania is in the amorphous state and the analysis performed by glancing incidence XRD has not shown the formation of titania crystalline phase even after calcination. However a more sensitive analysis by FT–FIR absorption spectroscopy has been done on the 350 °C calcined titania films; Fig. 4 shows the absorption spectra in the 600–100 cm1 interval of a titania mesostructured film after calcination at 350 °C. The spectrum shows two intense and well defined absorption bands peaking around 442 and 272 cm1. These bands are assigned to transverse optical (TO) Eu phonons in tetragonal anatase with two TiO2 units per primitive cell [22,23]. The spectrum indicates, therefore, that the titania is partially crystallized into the anatase phase, as small crystallites in an amorphous matrix; the dimension of the crystallites is likely very small because it is under the detection limit by X-ray diffraction. The calcined titania films have been used for post-synthesis grafting by amine groups with APTES. The result of the grafting process is reported in Fig. 5, which shows the FTIR absorption spectra in the 1800–600 cm1 of a mesoporous titania film after calcination at 350 °C (solid line, a) and functionalization with APTES
0.06 0.04 0.02 0.00 600
2
2
4
The Cauchy dispersion equation, n(k) = An + Bn/k + Cn/k , allows calculating the refractive index as a function of the wavelength. An is a parameter related to the average refractive index of the material, whereas Bn and Cn are parameters that provide the shape or curvature of the n(k) curve.
500
400
300
200
100
Wavenumber / cm-1 Fig. 4. FT–FIR absorption spectra in the 600–100 cm1 interval of a titania mesostructured film after calcination at 350 °C.
5
S. Mura et al. / Microporous and Mesoporous Materials 142 (2011) 1–6
0.04
a b c
Absorbance
0.03
0.02
0.01
0.00 1800
1600
1400
1200
1000
800
600
Wavenumber / cm-1 Fig. 5. FTIR absorption spectra of a mesoporous titania film in the 1800–600 cm1 after firing at 350 °C (solid line, a) and functionalization with APTES (dash line, b). The reference spectrum of APTES is the dot line, c.
(dash line, b); the reference spectrum of APTES (liquid film) is the dot curve, c (the full spectra are reported in SI). The spectrum of the functionalized titania film clearly shows the signature of the APTES [24], which can be easily detected because the spectrum of 350 °C calcined titania film has not absorption bands in the middle infrared range. This material has been then used as the host matrix for a further grafting step to bond specific peptides.
film in a physiological solution containing the peptides. The titania thin films functionalized with APTES (vide supra) were immersed in a solution of 1 mg/10 ml of peptide for different times (1–60 h) and solvents (phosphate buffer solution (PBS), toluene, ethanol). We have finally found that the best functionalization conditions are realized immersing the host porous films in a toluene solution, under shaking at room temperature for around 24 h. Not all the peptides have been successfully grafted, in fact, only H–Val-GlnAsp-Leu-Phe–COOH has appeared to be strongly bonded even after different washing steps with toluene and drying in air. The H–IleGln-Asp-Leu-Phe–COOH peptide resulted only weakly bonded while in the case of the peptide Fmoc-Phg-Gln-Asp-Leu-Phe–COOH the solubility is low and the binding to the matrix resulted very poor; likely the great molecular weight of the FMOC protector group hampers the penetration into the mesopores. We have measured the FTIR spectra before and after washing to evaluate the quality of grafting, only the samples that have not shown a significant difference have been considered as successfully grafted. Fig. 6a shows the FTIR absorption spectra in the 1800–600 cm1 range (full spectra are shown in the SI) of a mesoporous titania film after calcination at 350 °C and functionalization by APTES (dash line) and after binding with the peptide (solid line); the reference spectrum of the H–Val-Gln-Asp-Leu-Phe–COOH
0.09 0.06
We have synthesized, as described in Section 2, three different linear pentapeptides that are potential candidates for developing nano-biosensors of dioxins based on thin mesoporous films technologies. The pentapeptides are: H–Ile-Gln-Asp-Leu-Phe–COOH; H–Val-Gln-Asp-Leu-Phe–COOH; and Fmoc-Phg-Gln-Asp-Leu-Phe– COOH and are shown in Scheme 1. Different binding methods for linking the pentapeptides have been tested; we have tried two different routes: direct grafting via formation of covalent bonds and electrostatic grafting. In the first case to obtain a covalent linking between the APTES and the peptides, we have employed glutaraldehyde which is commonly used as bifunctional crosslinking reagent to covalently couple proteins with various surfaces. We have changed different synthesis parameters, but after washing the samples, we never observed the signature of glutaraldehyde in the mesoporous films by FTIR or Raman spectroscopy. We have attributed this failure to the difficulty of realizing the reaction between the amines of grafted APTES with glutaraldehyde in a nanoconfined environment. Therefore, we have directly immersed the
0.03
Absorbance
3.2. Grafting of peptides
a)
0.00 1.5
b)
1.0 0.5 0.0 1800
1600
1400
1200
Wavenumber /
1000
800
600
cm-1
Fig. 6. FTIR absorption spectra range of a mesoporous titania film: (a) after calcination at 350 °C and functionalization by APTES (dash line) and after binding with the peptide (solid line). The reference spectrum of the H–Val-Gln-Asp-LeuPhe–COOH peptide is reported in the graph (b).
6
2.5x10
a b c d
6
2.0x10
Intensity
6
1.5x10
6
1.0x10
5
5.0x10
0.0 520
560
600
640
680
Wavelength / nm
Scheme 1.
Fig. 7. Fluorescence emission spectra (kex = 490 nm) of toluene (solid line, a), toluene–H–Val-Gln-Asp-Leu-Phe–COOH peptide (dash line, b), toluene–FITC (dot line, c), toluene–FITC–Val-Gln-Asp-Leu-Phe–COOH peptide (dash-dot line, d).
6
S. Mura et al. / Microporous and Mesoporous Materials 142 (2011) 1–6
16000
a b c
14000 12000
Intensity
10000 8000 6000 4000 2000 0 510
540
570
600
630
660
690
any emission band (solid curve) and their spectrum is very similar to that obtained from titania–APTES mesoporous films before immersion (dot curve). From the spectra of Fig. 8 it is clear that the APTES amino groups are not effective to bind FITC while a chemical bond between functionalized film surface and fluorescent dye is only obtained via chemical reaction with the peptide. As a matter of fact, only the samples formed by TiO2–APTES–peptide– FITC show a strong increase of the emission properties. The FTIR and fluorescence data well support, therefore, the successful binding of the peptide to the titania mesoporous films. This process is robust and the samples even after washing in different solvents still show evidence of the presence of the peptide that is not affected by the washing cycle.
Wavelength / nm Fig. 8. Fluorescence emission spectra (kex = 490 nm) of the mesoporous APTES– titania after immersion in a solution of toluene–FITC (solid line, a), and toluene–FITC– Val-Gln-Asp-Leu-Phe–COOH peptide (dash line, b). The spectrum of APTES–titania films before immersion (dot line, c) is also reported as the reference.
peptide is reported in Fig. 6b. The FTIR spectra show that the absorption of the peptide at the end of the grafting process is very effective; the signature of the peptide can be clearly observed in the titania–APTES mesoporous film even after careful washing of the sample (Fig. 6a). We have used fluorescence spectroscopy to get a better insight into the functionalization process. This approach is based on using a fluorescent marker, such as fluorescein isothiocyanate (FITC), which after reaction with amines shows an intense fluorescent emission revealing the presence of the peptides. Fig. 7 shows the emission fluorescence spectra (kex = 490 nm) of toluene (solid line), toluene–H–Val-Gln-Asp-Leu-Phe–COOH peptide (dash line), toluene–FITC (dot line), toluene–FITC–Val-Gln-Asp-Leu-Phe–COOH peptide (dash-dot line). The fluorescence of toluene and H–ValGln-Asp-Leu-Phe–COOH peptide in toluene is zero with the exception of a small band of very weak intensity peaking around 570 nm; the two spectra are completely overlapped which reveals that H–Val-Gln-Asp-Leu-Phe–COOH peptide in toluene is not fluorescent. The fluorescence spectrum of FITC in toluene shows a wide band of low intensity around 530 nm; after binding with the peptide the intensity of this band is highly enhanced, an intense emission band peaking around 530 is, in fact, detected (dot line in Fig. 7). We can clearly attribute the increased fluorescence in the toluene–FITC–H–Val-Gln-Asp-Leu-Phe–COOH peptide solution to the reaction of fluorescein isothiocyanate with the peptide because the other solutions are not fluorescent. Fluorescein isothiocyanate is commonly used to attach a fluorescent marker to proteins via reaction of isothiocyanate with the primary and terminal amines in proteins [25]. We can use, therefore, this solution for postfunctionalization of the titania mesoporous film and as a marker of the process. On the other hand we have also tried to link the fluorescent marker FITC directly to the peptide that has already been immobilized on the film but this process has revealed not to be effective and not fluorescence emission has been observed. We have used the fluorescent solution of FITC–peptide to bind the fluorescent marker to the surface of the titania mesopores; a toluene–FITC solution (Fig. 8) has been also used to observe if any specific effect could be produced upon immersion of the samples in this solution. After immersion in the solutions the films have been analyzed by fluorescence spectroscopy; the titania films immersed in the FITC–peptide solution show a strong fluorescence with an intense band peaking around 530 nm (dash curve), while the samples immersed in the toluene–FITC solution do not show
4. Conclusions Mesostructured titania films with cubic pore organization have been used as a porous matrix for hosting linear pentapeptides; the functionalization of the pore surface with amine groups by 3aminopropyltriethoxysilane has shown to be an effective route for binding the peptides. The peptide has been introduced using an impregnation process that resulted depend on solvent type and time of immersion of the sample in the solution. The H–Val-GlnAsp-Leu-Phe–COOH pentaptide has been successful bonded to the film and resulted very tightly bind to the host matrix. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.micromeso.2010.10.047. References [1] Y. Sakatani, D. Grosso, L. Nicole, C. Boissiere, G. Soler-Illia, C. Sanchez, J. Mater. Chem. 16 (2006) 77–82. [2] L. Malfatti, P. Falcaro, H. Amenitsch, S. Caramori, R. Argazzi, C.A. Bignozzi, S. Enzo, M. Maggini, P. Innocenzi, Micropor. Mesopor. Mater. 88 (2006) 304–311. [3] P.C. Angelomè, G. Soler-Illia, Chem. Mater. 17 (2005) 322–331. [4] P.C. Angelomè, L. Andrini, M.E. Calvo, F.G. Requejo, S. Bilmes, G. Soler-Illia, J. Phys. Chem. C 111 (2007) 10886–10893. [5] L. Malfatti, M.G. Bellino, P. Innocenzi, G. Soler-Illia, Chem. Mater. 21 (2009) 2763–2769. [6] X. Chen, S. Dong, Biosens. Bioelectron. 18 (2003) 999–1004. [7] P.C. Angelomè, G. Soler-Illia, J. Mater. Chem. 15 (2005) 3903–3912. [8] E.L. Crepaldi, G. Soler-Illia, D. Grosso, F. Cagnol, F. Ribot, C. Sanchez, J. Am. Chem. Soc. 125 (2003) 9770–9786. [9] G. Soler-Illia, P. Innocenzi, Chem.: Eur. J. 12 (2006) 4478–4494. [10] E. Topoglidis, A.E.G. Cass, G. Gilardi, S. Sadeghi, N. Beaumont, J. Durrant, Anal. Chem. 70 (1998) 5111–5113. [11] Q. Gao, W. Xu, Y. Xu, D. Wu, Y. Sun, F. Deng, W. Shen, J. Phys. Chem. B 112 (2008) 2261–2267. [12] M. Hartmann, Chem. Mater. 17 (2005) 4577–4593. [13] S.P. Hudson, S. White, D. Goradia, H. Essa, B. Liu, L. Qiao, Y. Liu, J.C. Cooney, B.K. Hodnett, E. Magner, Biomol. Catal., in: ACS Symposium Series, vol. 986, 2008, pp. 49–60 (Chapter 2). [14] J.D. Bass, D. Grosso, C. Boissiere, E. Belamie, T. Coradin, C. Sanchez, Chem. Mater. 19 (2007) 4349–4356. [15] L.S. Jang, H.J. Liu, Biomed. Microdev. 11 (2009) 331–338. [16] J. Howarter, J.P. Youngblood, Langmuir 22 (2006) 11142–11147. [17] A. Calvo, P.C. Angelomé, V.M. Sanchez, D.A. Scherlis, F.J. Williams, G. Soler-Illia, Chem. Mater. 20 (2008) 4661–4668. [18] Y. Inuyama, C. Nakamura, T. Oka, Y. Yoneda, I. Obataya, N. Santo, J. Miyake, Biosens. Bioelectron. 22 (2007) 2093–2099. [19] M.P. Tate, V.N. Urade, J.D. Kowalski, T. Wei, B.D. Hamilton, B.W. Eggiman, H.W. Hillhouse, J. Phys. Chem. B 110 (2006) 9882. [20] P. Innocenzi, J. Non-Cryst. Solids 316 (2003) 309. [21] P. Falcaro, D. Grosso, H. Amenitsch, P. Innocenzi, J. Phys. Chem. B 108 (2004) 10942. [22] R.J. Gonzalez, R. Zallen, H. Berger, Phys. Rev. 55 (1997) 7014. [23] C. Pecharromàn, F. Gracìa, J.P. Holdago, M. Ocana, A.R. Gonzalez-Elipe, J. Bassas, J. Santiso, A. Figueras, J. Appl. Phys. 93 (2003) 4634. [24] R. Pena-Alonso, F. Rubio, J. Rubio, J.L. Oteo, J. Mater. Sci. 42 (2007) 595–603. [25] A.B. Schreiber, J. Haimovic, Methods Enzymol. 93 (1983) 147–155.
Contents lists available at ScienceDirect
Microporous and Mesoporous Materials journal homepage: www.elsevier.com/locate/micromeso
Polypeptide binding to mesostructured titania films Stefania Mura a, Gianfranco Greppi b, Anna Maria Roggio c, Luca Malfatti d, Plinio Innocenzi d,⇑ a
Dipartimento di Scienze Molecolari e Agroalimentari, Università di Milano, via Celoria 2, 20133 Milano, Italy Dipartimento di Scienze Zootecniche, Università di Sassari, via De Nicola 9, 07100 Sassari, Italy c Porto Conte Ricerche, strada provinciale 55 Porto Conte/Capo Caccia, km 8.400 località Tramariglio, 07041 Alghero (SS), Italy d Materials Science and Nanotechnology Laboratory, D.A.P., CR-INSTM, Università di Sassari, Palazzo Pou Salit, Piazza Duomo 6, 07041 Alghero (SS), Italy b
a r t i c l e
i n f o
Article history: Received 18 January 2010 Received in revised form 21 September 2010 Accepted 31 October 2010 Available online 4 November 2010 Keywords: Self-assembly Mesoporous Films Peptide
a b s t r a c t Mesoporous titania films have been prepared via evaporation induced self-assembly and used as a matrix to bind linear pentapeptides. The mesoporous surface has been functionalized with amine groups by immersing the samples in a solution of aminopropyltriethoxysilane in toluene. Different pentapeptides have been synthesized for binding to the mesoporous films: H–Ile-Gln-Asp-Leu-Phe–COOH, H–ValGln-Asp-Leu-Phe–COOH and Fmoc-Phg-Gln-Asp-Leu-Phe–COOH. The peptides have been bonded to the amine functionalized surface of the titania mesoporous films with an impregnation process. The H–Val-Gln-Asp-Leu-Phe–COOH peptide has been successfully bonded to the titania matrix, while the other peptides have shown not be suitable for the process because of lower solubility and sterical hindrance. The functionalization with aminopropyltriethoxysilane and peptide binding has been studied by Fourier transform infrared spectroscopy and fluorescence spectroscopy. A fluorescent marker, fluorescein isothiocyanate, has been used to confirm the incorporation of the peptides into the titania matrix. The process is reliable and robust, after several washing cycle of the samples, the peptides are still well bonded to the titania mesoporous films. Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction Mesostructured transition metal oxide films have a variety of potential applications as magnetic, photocatalysis [1], photovoltaic materials [2], and for the incorporation of other molecules by physisorption [3]. Titania mesoporous films exhibit different mesostructures, pore symmetry and connectivity that can be tailored using different synthesis strategies [4,5]. A great number of active sites (or functions) can be concentrated on the large pore surface area, enabling their possible applications as biosensors [6,7]. Titania mesostructured films can be produced by evaporation induced self-assembly [8] (EISA) using alcoholic solutions of metal chlorides as the inorganic source and ionic surfactants or block copolymers as the structure-directing agents [9]. In particular, the high surface area, chemical durability and optical transparency suggest that titania mesoporous films could be used as support for bioanalytical applications and for biosensors; nanoporous TiO2 films have been used for protein immobilization because of the good biocompatibility of this material [10]. Mesoporous materials have appeared, in fact, as a feasible matrix for encapsulation of proteins and peptides immobilization [11–13]. The main advantage of mesoporous titania films is the high hydrolytic stability in comparison ⇑ Corresponding author. Tel.: +39 079 998630; fax: +39 079 9720420. E-mail address: [email protected] (P. Innocenzi). 1387-1811/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.micromeso.2010.10.047
with mesoporous silica samples which have a severe limit for applications in biotechnologies due to their very low stability in water and physiological solutions [14]. In the present work, we have used mesoporous titania films as supports for the incorporation of biomolecules such as peptides; for this purpose TiO2 films have been modified with a functional precursor that leaves an amino-terminal group on the pore surface for the linking with the bio-element (peptide). In general, functionalization of mesoporous films with organic groups can be obtained using two main approaches: co-condensation with a functional precursor during film formation, or post-functionalization of the pore surface after film deposition. Each route has certain advantages: co-condensation allows a homogeneous distribution of organic groups over the pore surfaces, the post-grafting method presents more protruding (thus more accessible) functions and better defined pore sizes [5]. Moreover, it is difficult to add organic groups by co-condensation in non-silica frameworks because of the instability of most G–Ti bonding (G = grafting group) towards hydrolysis in the very acidic synthesis condition used for films preparation [1] while postfunctionalization ensures that molecules are grafted to the pore surface rather than incorporated within the oxide framework [1]. Films obtained by post-grafting show functional groups, for instance amines, only on the pore surface whereas films obtained by co-condensation should present functional groups also within the pore walls besides those present on the pore surfaces. This is the
2
S. Mura et al. / Microporous and Mesoporous Materials 142 (2011) 1–6
reason why using post-functionalization of mesoporous materials more amine groups are potentially available for reaction than in the co-condensed films [4,5]. 3-Aminopropyltriethoxysilane (APTES) is commonly used as coupling agent for the modification of silica surfaces to promote protein adhesion, cell growth, to attach metal nanoparticles to silica substrates, for biological implants and in lab on chip applications [15]. The hydrolysis and condensation of silanes drives the bonding of APTES to the substrate via formation of siloxane bonds on the surface [16]. The initial hydrolysis step can occur either in solution or at the substrate surface depending on the amount of water present in the system. An overabundance of water will result in excessive polycondensation in the solvent phase, while a deficiency of water will result in the formation of an incomplete monolayer [9]; solvent, concentration, reaction time, and reaction temperature are other parameters that affect the grafting kinetics [9]. In titania materials the functionalization via APTES is realized by condensation of Ti–OH with Si–OH to form mixed Ti–O–Si bonds [17]. Silanization experiments with APTES, to add amino functional groups for the simple immobilization of biomolecules, have been performed to elucidate their use in protein chips and biosensors [11]. In the present work, we have fabricated a titania mesostructured support to immobilize a peptide with potentially selective properties with respect to dioxins for the future development of a biosensor. Dioxin is a toxic chemical species in the environment that must be strictly monitored; it is mainly generated from waste incineration or pesticide.1 The method to determine the accurate environmental concentration of dioxin is chemical analysis by high-resolution gas chromatography mass spectrometry (HR-GC/MS) [13]. Recently, simple and rapid methods, for instance a gene assay and ELISA, using biological species for capturing dioxins have been accepted as official and legal methods to detect dioxins instead of time-consuming GC/MS. Antibodies or receptors, which consist of natural amino acids have the potential to recognize dioxin molecules [11,13]. However, denaturation of the antibodies caused by the presence of organic solvents necessary to dissolve hydrophobic dioxin in aqueous solution is a serious problem in the use of antibodies in dioxin assays. If the concentration of organic solvents in the assay solution is lowered, high concentrations of dioxin cannot be dissolved, which affects the precision of the analysis [13]. If we could use a synthetic oligopeptide instead of a natural immuno antibody, the problem of denaturation in organic solvent would be solved. Moreover, quality control procedures for a synthetic peptide would be easier than for a natural antibody and regeneration and reuse by washing out bound target would be feasible [13]. In previous studies, peptides that specifically bind dioxin have been screened from a combinatorial pentapeptide library and two sequences, Phe-Leu-Asp-Gln-Ile and Phe-Leu-Asp-Gln-Val, have been obtained [18]. Their design is based on the binding pocket of aryl-hydrocarbon receptor or CDR domain of monoclonal anti-dioxin antibody. In addition, Phe-Leu-Asp-Gln-Phg (5 Phg), the peptide substituted at the fifth position by a phenyl glycine (Phg) with better performance than the original, has been discovered from a one-amino-acid substituted library containing synthetic peptides [13]. In the present work, we have tried to develop a suitable mesoporous titania film
1 The term dioxin refers to a class of structurally and chemically related halogenated aromatic hydrocarbons that includes polychlorinated dibenzodioxins (PCDDs or dioxins), polychlorinated dibenzofurans (PCDFs or furans) and the ‘‘dioxinlike’’ polychlorinated biphenyls (PCBs). Because of their chemistry, dioxins are both toxic and persistent in the environment. Dioxins and furans are included in the United Nations Environment Programme ‘‘Dirty Dozen’’. In this paper, the term dioxin is used in a toxicological meaning to designate the PCDDs, the PCDFs and the coplanar (‘‘dioxin-like’’) PCBs, since these classes of compounds show the same type of toxicity. Because of the large number of congeners, relevant individual congeners are assigned with a toxic equivalency factor (TEF) that relate their toxicity to that of tetrachlorodibenzo-p-dioxin (TCDD) (2,3,7,8-TCDD) and are to be evaluated as dioxins.
to immobilize the peptides via a post-synthesis functionalization grafting route. 2. Experimental section 2.1. Chemicals All commercially available solvents and reagents were used without further purification. TiCl4 (reagent grade, >98%), anhydrous ethanol (reagent grade, >99.9%), bidistilled water, acetone (reagent grade, >99.8%), toluene (reagent grade, >99.5%), were purchased from Carlo Erba. Pluronic F-127 (cell culture test), 3-(aminopropyl)triethoxysilane (reagent grade, >98%), were purchased from Sigma–Aldrich. N,N0 -dimethylformamide (puriss. p.a., >99.8%) (DMF), 1-methyl-2-pyrrolidone (for peptide synthesis, >99.8%) (NMP), dichloromethane (reagent grade, >99.5%) (DCM), 2-propanol (reagent grade, >99.5%), trifluoroacetic acid (reagent grade, >98%) (TFA), piperidine (reagent grade, >98%), triisopropylsilane (99%) (TIS) and N-[(dimethylamino)-1H-1,2,3-triazolo[4,5-b]pyridin-1-ylmethylene]-N-methylmethanaminium hexafluorophosphate N-oxide (99%) (HATU) were purchased from Sigma–Aldrich. Fmoc-Pal-Peg resin (substitution 0.21 mmol g1) was obtained from Applied Biosystems (Applied Biosystems Inc., USA). The Fmoc-amino acids and 1-hydroxybenzotriazole hydrate (HOBt) were purchased from Novabiochem (Novabiochem, Switzerland). a-Cyano-4-hydroxycynnamic acid from Fluka was used as matrix for MALDI TOF experiments. All solvents used in the chromatography section were of the gradient grade for HPLC, and were purchased from Merck-VWR (D-Darmstadt, Germany). Fluorescein isothiocyanate (FITC), isomer I (90% purity) was purchased from ACROS organics and used without further purification. 2.2. Synthesis of peptides Three different linear pentapeptides have been synthesized: H–Ile-Gln-Asp-Leu-Phe–COOH, H–Val-Gln-Asp-Leu-Phe–COOH and Fmoc-Phg-Gln-Asp-Leu-Phe–COOH. The detailed description of the peptides synthesis is reported in the Supplementary data. 2.3. Synthesis of mesoporous titania films The precursor solution was prepared in different steps. In the first one Pluronic F127 was dissolved in ethanol; the solution was stirred for 2 h at 25 °C in a closed vessel. TiCl4 was slowly added to the ethanolic solution of Pluronic F127, under stirring for 20 min in an ice bath. Finally water was added to this mixture that was left to react under stirring for 3 h at 25 °C in a closed vessel. The addition of water causes the hydrolysis of TiCl4 producing EtOH and HCl, responsible of the high stability of the sol that is stable for several months at 25 °C. The molar ratios in solution were TiCl4:Pluronic F127:H2O:EtOH = 1:0.005:10:40. Titania thin films were deposited by dip-coating silicon wafers in the precursor solution at 25 °C and at a withdrawal rate of 15 cm min1. The relative humidity (RH) inside the dip-coater chamber was maintained between 18% and 25% to obtain good optical quality and high structural order in the titania films. The as deposited films were aged at room temperature (RH = 50%) for 24 h. To increase the inorganic polycondensation and stabilize the mesophase the films were submitted to different firing steps: 60, 120 and 200 °C for 24 h at each temperature in an oven with heating rate of 10 °C min1. This treatment is important to allow the formation of highly organized, high surface area mesoporous coatings, exhibiting crystalline frameworks. The final calcination process to remove the organic template of these stabilized coatings was done at 350 °C for 3.5 h in air under static conditions with a heating rate
S. Mura et al. / Microporous and Mesoporous Materials 142 (2011) 1–6
of 10 °C min1. Calcined mesoporous films at 350 °C were used as the matrix for incorporation of the pentapeptides. The functionalization process of titania films with amine groups was studied in different solvents (EtOH, toluene) and at different temperatures (20–80 °C), for different concentrations of APTES (0.002–0.2 M) and for different times (1–24 h). Based on these results we have selected the following protocol: immersion of 350 °C treated film in a 0.2 M solution of APTES in toluene for 24 h under stirring at 25 °C. The amine grafted films were carefully washed by toluene with several washing cycles and finally dried in air under static conditions. To identify the presence of the peptide in the film by fluorescence spectroscopy, we prepared a solution of fluorescein isothiocyanate (FITC) in dimethyl sulfoxide (DMSO) at the concentration of 5 mg ml1. We dissolved 1 mg of peptide in 1 ml of toluene, immediately after 20 lL of FITC–DMSO were added to this solution. Carbonate buffer 0.1 M and sodium carbonate 0.1 M solutions were also tested as possible solvents, but the samples immersed in these solutions were not able to give fluorescence and were not further used. The FITC–DMSO–toluene solution was left under stirring for 3 h; after this step the solution was diluted by adding 20 ml of toluene and used to immerse the titania films functionalized with APTES. The samples were left in the solution for 24 h in the dark; after this step the films were washed several times with fresh ethanol and dried in air. 2.4. Film characterization Fourier transform infrared (FTIR) analysis was performed using a Bruker Vertex 70v spectrophotometer. The optical bench and the sample compartment were kept in vacuum during the measurement at pressure lower than 0.5 hPa. The measurements in the middle infrared (MIR) region were performed using a Globar source, a KBr beamsplitter, and a RT-DLaTGS detector averaging 256 scans with 4 cm1 of resolution. The measurements in the far infrared range were done using a Globar source, a Si beamsplitter, and a RT-DTGS-FIR detector. The spectra were recorded in transmission, in the 600–100 cm1 range by averaging 32 scans with 4 cm1 of resolution. A silicon wafer was used as the substrate to measure the background; the baseline was calculated by a rubberband algorithm (OPUS 7 software). Fluorescence analysis was done using a FluoroMax-3 Horiba Jobin Yvon spectrofluorometer. The probing beam was set to impinge on one side of the sample (silicon substrate, incidence angle of 2–3°) so that the sample acted as a waveguide for the incident light wave, while the luminescence was collected at 90° with respect to the incident beam. This configuration enhanced the signal-to-noise ratio and avoided reflection effects. Each acquisition is the average of three different accumulations. An excitation wavelength of kex = 490 nm was used for acquiring the emission spectra. The organization of the porous structure was investigated by 2D grazing incidence small angle X-ray scattering (GISAXS) at the Austrian SAXS beamline of ELETTRA synchrotron facility (Trieste, Italy). An incident energy of 8 keV (wavelength 1.54 Å) was used; the instrumental glancing angle between the incident radiation and the sample was set slightly above the critical angle (grazing incidence). A two-dimensional CCD detector (Photonic Science, UK) was used to acquire the scattering patterns; each measurement consisted typically on the average of 10 acquisitions with integration time of 6 s. Pore arrangements were also studied by transmission electron microscopy (TEM). TEM micrographs were obtained in bright field mode on a JEOL 200CX microscope equipped with a tungsten cathode operating at 200 kV. Finely ground films scratched from the substrate were dispersed in n-octane by sonication, and then they were dropped on a carbon-coated copper grid and dried for TEM
3
observations. Center-to-center interpore distance was evaluated by line profile analysis on a set of representative TEM images as the average FWHM of the intensity distribution along a line passing through the pore centers. Film thicknesses and refractive indexes were measured by spectroscopic ellipsometry using an a-SE instrument (J.A. Woollam, USA) working in the 400–850 nm range.
3. Discussion and results 3.1. Mesostructured titania film synthesis To obtain mesostructured titania films we have used a well established and highly reproducible synthesis that has been reported by Soler-Illia et al. in previous publications [7,8]. This synthesis allow getting organized mesostructures after removal of the templating micelles via thermal calcination; after firing the films still present high organization, excellent optical quality and hydrolytic stability. The organization of the porous structures on films has been studied by GISAXS; Fig. 1 shows the typical pattern of an ordered mesoporous titania film after thermal treatment at 350 °C. The pattern appears composed of three peaks that can be 0) reflections of a body centered indexed as (1 1 0), (1 0 1), and (1 1 in the space group) structure with domains preferencubic (Im3m tially oriented with the [1 1 0] direction normal to the surface, and allowed by the circular permutation [19]. The lower intensity of the (1 1 0) peak is due to the incident angle that reduces the quantity of domains having their in-plane diffraction in the Bragg conditions. As a consequence of the thermal treatment, the cubic unit cell is contracted in the out-of-plane [1 1 0] direction (normal to the substrate), which will be referred to as d110, whereas the inplane (parallel to the substrate) cell constant will be referred to as d110 . Following this attribution, the lattice parameters have been calculated to be d110 = 7.0 ± 0.1 nm and d110 = 3.1 ± 0.1 nm. The synthesis protocol has shown to be very robust and highly reproducible; we have performed systematic GISAXS analysis on different batches of samples (more than 20 samples have been analyzed by GISAXS) prepared in various times and we have always observed cubic organization of the porous structure [17]. A direct observation of the pore arrangement has been obtained by transmission electron microscopy to support the information provided by SAXS data. Bright field TEM images were collected
Fig. 1. GISAXS pattern of an ordered mesoporous titania film treated at 350 °C.
4
S. Mura et al. / Microporous and Mesoporous Materials 142 (2011) 1–6
Fig. 2. Bright-field TEM images of mesoporous titania films treated at 350 °C.
Absorbance
0.08 0.06 0.04 0.02 0.00 4000
3750
3500
3250
3000
2750
2500
Wavenumber / cm-1 Fig. 3. FTIR absorption spectra in the 4000–2500 cm1 interval of as deposited (solid line) and calcined (dash line) mesostructured titania films; calcination has been done in air at 350 °C.
for samples fired at 350 °C. Fig. 2 shows representative direct images of the deposited films with some projections of a slightly distorted cubic structure. Even if an unambiguous identification would requires cross-sectional TEM, according to the literature, we have attributed this images to the (1 1 0) projection [8]. Under these conditions, in fact, the {1 1 0} face appears as a sequence of channels because the depth of field of the electron microscopy merges several spherical pore planes. Starting from this consideration, the distance among pores has been calculated by a line profile analysis on representative TEM images. The center-to-center interpore distance was estimated to be 8.5 ± 1 nm. The effect of the mesoporous film functionalization was also followed by spectroscopic ellipsometry. The data were obtained assuming a fitting model based on the assumption of transparent films on silicon (Cauchy dispersion relation).2 After calcination at 350 °C in air the titania films show an average thickness of 187 ± 5 nm with a refractive index of 1.72 measured at k = 632.8 nm. After functionalization with APTES the film thickness slightly increases to a value of around 218 ± 37 nm with a significant changes of the refractive index to a value of 1.90. Finally, after peptide binding, the films reach a thickness of around 243 ± 30 nm with an average refractive index of 1.79. This value, which is lower than that measured on the amino-functionalized films, can be attributed to the formation of an overlayer with a low refractive index made of peptides bonded to mesoporous film surface. Fig. 3 shows the FTIR absorption spectra in the 4000–2500 cm1 interval of as deposited (solid line) and calcined (dash line)
0.10 0.08
Absorbance
0.10
mesostructured titania films; calcination has been done in air at 350 °C. The spectra of the as deposited film show the presence of a wide intense band peaking around 3250 cm1 which is assigned to O–H stretching mode, due to Ti–OH species and absorbed water (overlapped band around 3200 cm1) [20]. The presence of the organic template in the as deposited film is indicated by the C–H stretching bands in the 3000–2800 cm1 interval [21]. After calcination of the film at 350 °C the spectrum appears as a flat curve with only a very small signal around 2900 cm1; this indicates that the thermal treatment removes almost completely the surfactant and at the same time promotes the densification of the titania network, through condensation reactions of Ti–OH species. We have also checked the effect of the thermal treatment on the structure of the titania pore walls; in the as deposited films the titania is in the amorphous state and the analysis performed by glancing incidence XRD has not shown the formation of titania crystalline phase even after calcination. However a more sensitive analysis by FT–FIR absorption spectroscopy has been done on the 350 °C calcined titania films; Fig. 4 shows the absorption spectra in the 600–100 cm1 interval of a titania mesostructured film after calcination at 350 °C. The spectrum shows two intense and well defined absorption bands peaking around 442 and 272 cm1. These bands are assigned to transverse optical (TO) Eu phonons in tetragonal anatase with two TiO2 units per primitive cell [22,23]. The spectrum indicates, therefore, that the titania is partially crystallized into the anatase phase, as small crystallites in an amorphous matrix; the dimension of the crystallites is likely very small because it is under the detection limit by X-ray diffraction. The calcined titania films have been used for post-synthesis grafting by amine groups with APTES. The result of the grafting process is reported in Fig. 5, which shows the FTIR absorption spectra in the 1800–600 cm1 of a mesoporous titania film after calcination at 350 °C (solid line, a) and functionalization with APTES
0.06 0.04 0.02 0.00 600
2
2
4
The Cauchy dispersion equation, n(k) = An + Bn/k + Cn/k , allows calculating the refractive index as a function of the wavelength. An is a parameter related to the average refractive index of the material, whereas Bn and Cn are parameters that provide the shape or curvature of the n(k) curve.
500
400
300
200
100
Wavenumber / cm-1 Fig. 4. FT–FIR absorption spectra in the 600–100 cm1 interval of a titania mesostructured film after calcination at 350 °C.
5
S. Mura et al. / Microporous and Mesoporous Materials 142 (2011) 1–6
0.04
a b c
Absorbance
0.03
0.02
0.01
0.00 1800
1600
1400
1200
1000
800
600
Wavenumber / cm-1 Fig. 5. FTIR absorption spectra of a mesoporous titania film in the 1800–600 cm1 after firing at 350 °C (solid line, a) and functionalization with APTES (dash line, b). The reference spectrum of APTES is the dot line, c.
(dash line, b); the reference spectrum of APTES (liquid film) is the dot curve, c (the full spectra are reported in SI). The spectrum of the functionalized titania film clearly shows the signature of the APTES [24], which can be easily detected because the spectrum of 350 °C calcined titania film has not absorption bands in the middle infrared range. This material has been then used as the host matrix for a further grafting step to bond specific peptides.
film in a physiological solution containing the peptides. The titania thin films functionalized with APTES (vide supra) were immersed in a solution of 1 mg/10 ml of peptide for different times (1–60 h) and solvents (phosphate buffer solution (PBS), toluene, ethanol). We have finally found that the best functionalization conditions are realized immersing the host porous films in a toluene solution, under shaking at room temperature for around 24 h. Not all the peptides have been successfully grafted, in fact, only H–Val-GlnAsp-Leu-Phe–COOH has appeared to be strongly bonded even after different washing steps with toluene and drying in air. The H–IleGln-Asp-Leu-Phe–COOH peptide resulted only weakly bonded while in the case of the peptide Fmoc-Phg-Gln-Asp-Leu-Phe–COOH the solubility is low and the binding to the matrix resulted very poor; likely the great molecular weight of the FMOC protector group hampers the penetration into the mesopores. We have measured the FTIR spectra before and after washing to evaluate the quality of grafting, only the samples that have not shown a significant difference have been considered as successfully grafted. Fig. 6a shows the FTIR absorption spectra in the 1800–600 cm1 range (full spectra are shown in the SI) of a mesoporous titania film after calcination at 350 °C and functionalization by APTES (dash line) and after binding with the peptide (solid line); the reference spectrum of the H–Val-Gln-Asp-Leu-Phe–COOH
0.09 0.06
We have synthesized, as described in Section 2, three different linear pentapeptides that are potential candidates for developing nano-biosensors of dioxins based on thin mesoporous films technologies. The pentapeptides are: H–Ile-Gln-Asp-Leu-Phe–COOH; H–Val-Gln-Asp-Leu-Phe–COOH; and Fmoc-Phg-Gln-Asp-Leu-Phe– COOH and are shown in Scheme 1. Different binding methods for linking the pentapeptides have been tested; we have tried two different routes: direct grafting via formation of covalent bonds and electrostatic grafting. In the first case to obtain a covalent linking between the APTES and the peptides, we have employed glutaraldehyde which is commonly used as bifunctional crosslinking reagent to covalently couple proteins with various surfaces. We have changed different synthesis parameters, but after washing the samples, we never observed the signature of glutaraldehyde in the mesoporous films by FTIR or Raman spectroscopy. We have attributed this failure to the difficulty of realizing the reaction between the amines of grafted APTES with glutaraldehyde in a nanoconfined environment. Therefore, we have directly immersed the
0.03
Absorbance
3.2. Grafting of peptides
a)
0.00 1.5
b)
1.0 0.5 0.0 1800
1600
1400
1200
Wavenumber /
1000
800
600
cm-1
Fig. 6. FTIR absorption spectra range of a mesoporous titania film: (a) after calcination at 350 °C and functionalization by APTES (dash line) and after binding with the peptide (solid line). The reference spectrum of the H–Val-Gln-Asp-LeuPhe–COOH peptide is reported in the graph (b).
6
2.5x10
a b c d
6
2.0x10
Intensity
6
1.5x10
6
1.0x10
5
5.0x10
0.0 520
560
600
640
680
Wavelength / nm
Scheme 1.
Fig. 7. Fluorescence emission spectra (kex = 490 nm) of toluene (solid line, a), toluene–H–Val-Gln-Asp-Leu-Phe–COOH peptide (dash line, b), toluene–FITC (dot line, c), toluene–FITC–Val-Gln-Asp-Leu-Phe–COOH peptide (dash-dot line, d).
6
S. Mura et al. / Microporous and Mesoporous Materials 142 (2011) 1–6
16000
a b c
14000 12000
Intensity
10000 8000 6000 4000 2000 0 510
540
570
600
630
660
690
any emission band (solid curve) and their spectrum is very similar to that obtained from titania–APTES mesoporous films before immersion (dot curve). From the spectra of Fig. 8 it is clear that the APTES amino groups are not effective to bind FITC while a chemical bond between functionalized film surface and fluorescent dye is only obtained via chemical reaction with the peptide. As a matter of fact, only the samples formed by TiO2–APTES–peptide– FITC show a strong increase of the emission properties. The FTIR and fluorescence data well support, therefore, the successful binding of the peptide to the titania mesoporous films. This process is robust and the samples even after washing in different solvents still show evidence of the presence of the peptide that is not affected by the washing cycle.
Wavelength / nm Fig. 8. Fluorescence emission spectra (kex = 490 nm) of the mesoporous APTES– titania after immersion in a solution of toluene–FITC (solid line, a), and toluene–FITC– Val-Gln-Asp-Leu-Phe–COOH peptide (dash line, b). The spectrum of APTES–titania films before immersion (dot line, c) is also reported as the reference.
peptide is reported in Fig. 6b. The FTIR spectra show that the absorption of the peptide at the end of the grafting process is very effective; the signature of the peptide can be clearly observed in the titania–APTES mesoporous film even after careful washing of the sample (Fig. 6a). We have used fluorescence spectroscopy to get a better insight into the functionalization process. This approach is based on using a fluorescent marker, such as fluorescein isothiocyanate (FITC), which after reaction with amines shows an intense fluorescent emission revealing the presence of the peptides. Fig. 7 shows the emission fluorescence spectra (kex = 490 nm) of toluene (solid line), toluene–H–Val-Gln-Asp-Leu-Phe–COOH peptide (dash line), toluene–FITC (dot line), toluene–FITC–Val-Gln-Asp-Leu-Phe–COOH peptide (dash-dot line). The fluorescence of toluene and H–ValGln-Asp-Leu-Phe–COOH peptide in toluene is zero with the exception of a small band of very weak intensity peaking around 570 nm; the two spectra are completely overlapped which reveals that H–Val-Gln-Asp-Leu-Phe–COOH peptide in toluene is not fluorescent. The fluorescence spectrum of FITC in toluene shows a wide band of low intensity around 530 nm; after binding with the peptide the intensity of this band is highly enhanced, an intense emission band peaking around 530 is, in fact, detected (dot line in Fig. 7). We can clearly attribute the increased fluorescence in the toluene–FITC–H–Val-Gln-Asp-Leu-Phe–COOH peptide solution to the reaction of fluorescein isothiocyanate with the peptide because the other solutions are not fluorescent. Fluorescein isothiocyanate is commonly used to attach a fluorescent marker to proteins via reaction of isothiocyanate with the primary and terminal amines in proteins [25]. We can use, therefore, this solution for postfunctionalization of the titania mesoporous film and as a marker of the process. On the other hand we have also tried to link the fluorescent marker FITC directly to the peptide that has already been immobilized on the film but this process has revealed not to be effective and not fluorescence emission has been observed. We have used the fluorescent solution of FITC–peptide to bind the fluorescent marker to the surface of the titania mesopores; a toluene–FITC solution (Fig. 8) has been also used to observe if any specific effect could be produced upon immersion of the samples in this solution. After immersion in the solutions the films have been analyzed by fluorescence spectroscopy; the titania films immersed in the FITC–peptide solution show a strong fluorescence with an intense band peaking around 530 nm (dash curve), while the samples immersed in the toluene–FITC solution do not show
4. Conclusions Mesostructured titania films with cubic pore organization have been used as a porous matrix for hosting linear pentapeptides; the functionalization of the pore surface with amine groups by 3aminopropyltriethoxysilane has shown to be an effective route for binding the peptides. The peptide has been introduced using an impregnation process that resulted depend on solvent type and time of immersion of the sample in the solution. The H–Val-GlnAsp-Leu-Phe–COOH pentaptide has been successful bonded to the film and resulted very tightly bind to the host matrix. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.micromeso.2010.10.047. References [1] Y. Sakatani, D. Grosso, L. Nicole, C. Boissiere, G. Soler-Illia, C. Sanchez, J. Mater. Chem. 16 (2006) 77–82. [2] L. Malfatti, P. Falcaro, H. Amenitsch, S. Caramori, R. Argazzi, C.A. Bignozzi, S. Enzo, M. Maggini, P. Innocenzi, Micropor. Mesopor. Mater. 88 (2006) 304–311. [3] P.C. Angelomè, G. Soler-Illia, Chem. Mater. 17 (2005) 322–331. [4] P.C. Angelomè, L. Andrini, M.E. Calvo, F.G. Requejo, S. Bilmes, G. Soler-Illia, J. Phys. Chem. C 111 (2007) 10886–10893. [5] L. Malfatti, M.G. Bellino, P. Innocenzi, G. Soler-Illia, Chem. Mater. 21 (2009) 2763–2769. [6] X. Chen, S. Dong, Biosens. Bioelectron. 18 (2003) 999–1004. [7] P.C. Angelomè, G. Soler-Illia, J. Mater. Chem. 15 (2005) 3903–3912. [8] E.L. Crepaldi, G. Soler-Illia, D. Grosso, F. Cagnol, F. Ribot, C. Sanchez, J. Am. Chem. Soc. 125 (2003) 9770–9786. [9] G. Soler-Illia, P. Innocenzi, Chem.: Eur. J. 12 (2006) 4478–4494. [10] E. Topoglidis, A.E.G. Cass, G. Gilardi, S. Sadeghi, N. Beaumont, J. Durrant, Anal. Chem. 70 (1998) 5111–5113. [11] Q. Gao, W. Xu, Y. Xu, D. Wu, Y. Sun, F. Deng, W. Shen, J. Phys. Chem. B 112 (2008) 2261–2267. [12] M. Hartmann, Chem. Mater. 17 (2005) 4577–4593. [13] S.P. Hudson, S. White, D. Goradia, H. Essa, B. Liu, L. Qiao, Y. Liu, J.C. Cooney, B.K. Hodnett, E. Magner, Biomol. Catal., in: ACS Symposium Series, vol. 986, 2008, pp. 49–60 (Chapter 2). [14] J.D. Bass, D. Grosso, C. Boissiere, E. Belamie, T. Coradin, C. Sanchez, Chem. Mater. 19 (2007) 4349–4356. [15] L.S. Jang, H.J. Liu, Biomed. Microdev. 11 (2009) 331–338. [16] J. Howarter, J.P. Youngblood, Langmuir 22 (2006) 11142–11147. [17] A. Calvo, P.C. Angelomé, V.M. Sanchez, D.A. Scherlis, F.J. Williams, G. Soler-Illia, Chem. Mater. 20 (2008) 4661–4668. [18] Y. Inuyama, C. Nakamura, T. Oka, Y. Yoneda, I. Obataya, N. Santo, J. Miyake, Biosens. Bioelectron. 22 (2007) 2093–2099. [19] M.P. Tate, V.N. Urade, J.D. Kowalski, T. Wei, B.D. Hamilton, B.W. Eggiman, H.W. Hillhouse, J. Phys. Chem. B 110 (2006) 9882. [20] P. Innocenzi, J. Non-Cryst. Solids 316 (2003) 309. [21] P. Falcaro, D. Grosso, H. Amenitsch, P. Innocenzi, J. Phys. Chem. B 108 (2004) 10942. [22] R.J. Gonzalez, R. Zallen, H. Berger, Phys. Rev. 55 (1997) 7014. [23] C. Pecharromàn, F. Gracìa, J.P. Holdago, M. Ocana, A.R. Gonzalez-Elipe, J. Bassas, J. Santiso, A. Figueras, J. Appl. Phys. 93 (2003) 4634. [24] R. Pena-Alonso, F. Rubio, J. Rubio, J.L. Oteo, J. Mater. Sci. 42 (2007) 595–603. [25] A.B. Schreiber, J. Haimovic, Methods Enzymol. 93 (1983) 147–155.