Characterization of A Polymeric Adsorbed Coating for DNA Microarray

Jan 24, 2004 - molecules onto the surface of glass substrates such as microwell plates, beads, tubes, and microscope slides, for hybridization assays ...
0 downloads 0 Views 145KB Size
Download PDF
Anal. Chem. 2004, 76, 1352-1358

Characterization of A Polymeric Adsorbed Coating for DNA Microarray Glass Slides Giovanna Pirri,† Francesco Damin,† Marcella Chiari,*,† Elza Bontempi,‡ and Laura E. Depero‡

Istituto di Chimica del Riconoscimento Molecolare, CNR, Milano, Italy, and INSTM and Laboratorio di Strutturistica Chimica, Universita` di Brescia, Brescia, Italy

A new method was developed to covalently attach target molecules onto the surface of glass substrates such as microwell plates, beads, tubes, and microscope slides, for hybridization assays with fluorescent targets. The innovative concept introduced by this work is to physically adsorb onto underivatized glass surfaces a functional copolymer, able to graft amino-modified DNA molecules. The polymer, obtained by radical copolymerization of N,Ndimethylacrylamide, N-acryloyloxysuccinimide, and 3-(trimethoxysilyl)propyl methacrylate, copoly(DMA-NASMAPS), self-adsorbs onto the glass surface very quickly, typically in 5-30 min. The film, formed on the surface, bears active esters, which react with amino-modified DNA targets. The surface layer is stable in an aqueous buffer containing various additives (SDS, urea, salt), even at boiling temperature. It should be emphasized that the coating is formed by the immersion of glass slides in a diluted aqueous solution of the polymer. Therefore, the procedure is fast, inexpensive, robust, and reliable, and it does not require time-consuming glass pretreatments. Slides, coated with copoly(DMA-NAS-MAPS), were profitably used as substrates for the preparation of low-density DNA microarrays. The density and the thickness of the films were evaluated by X-ray reflectivity measurements whereas the extent of reaction of functional groups with DNA molecules was determined by a functional test. The experiments indicate that half of the active groups present on the surface reacts with oligonucleotide probes. In recent years, considerable advancements in speed and throughput of genetic analyses have been obtained by the introduction of DNA microarray technology.1-3 The technique has been widely used in different molecular biology applications including gene expression analysis, DNA sequencing, genome mapping, mutation detection, and gene discovery.4-7 The power * Corresponding author. Tel: 0039 02 28500035. Fax: 0039 02 28500036. E-mail: [email protected]. † Istituto di Chimica del Riconoscimento Molecolare, CNR. ‡ Universita ` di Brescia. (1) Watson, A.; Mazumder, A.; Stewart, M.; Balasubramanian, S. Curr. Opin. Biotechnol. 1998, 9, 609-614. (2) Duggan, D. J.; Bittner, M.; Chen, Y.; Meltzer, P.; Trent, J. M. Nat. Genet. (Suppl.) 1999, 21, 10-14. (3) Graves, D. J. Trends Biotechnol. 1999, 17, 127-134. (4) Pollack, J. R.; Perou, C. M.; Alizadeh, A. A.; Eisen, M. B.; Pergamenschikov, A.; Williams, C. F.; Jeffrey, S. S.; Botstein, D.; Brown, P. O. Nat. Genet. 1999, 23, 41-46.

1352 Analytical Chemistry, Vol. 76, No. 5, March 1, 2004

of microarrays comes from the miniaturized format that offers significant advantages in terms of cost, speed, automation, and operational simplicity. Two are the approaches for conducting DNA microarrays: (a) direct on-surface synthesis of oligonucleotides8-11 and (b) immobilization of presynthesized DNA molecules, the so-called deposition method.12-18 Both approaches present advantages and disadvantages. Concerning the direct synthesis of oligonucleotides on a solid surface, the best-known method is the one based on photolithography on glass, developed by Fodor et al.19 The use of the in situ synthesis, combined with a combinatorial approach, leads to the production of high-density oligonucleotide microarrays with relatively few coupling steps. This method is expensive and does not offer very much flexibility of design. Alternatively, DNA microarrays can be fabricated by deposition on glass of oligonuclotides or cDNA, followed by on-chip immobilization, using high-speed robotics. Depositioning technologies are very flexible in terms of spot disposition and volumes, although the array density is significantly lower than that provided by in situ synthesis. In addition, deposition is the only method to attach onto a glass surface large molecules such as PCR fragments and cDNAs. (5) De Saizieu, A.; Certa, U.; Warrington, J.; Gray, C.; Keck, W.; Mous, J. Nat. Biotechnol. 1998, 16, 45-48. (6) Johnston, M. Curr. Biol. 1998, 8, 171-174. (7) O’ Donnell-Maloney, M. J.; Little, D. P. Genet. Anal. 1996, 13 (6), 151157. (8) Lipshutz, R. J.; Fodor, S. P.; Gingeras, T. R.; Lockhart, D. J. Nat. Genet. 1999, 21, 20-24. (9) McGall, G.; Labadie, J.; Brock, P.; Wallraff, G.; Nguyen, T.; Hinsberg, W. Proc. Natl., Acad. Sci., U.S.A. 1996, 93, 13555-13560. (10) Singh-Gasson, S.; Green, R. D.; Yue, Y.; Nelson, C.; Blattner, F.; Sussman, M. R.; Cerrina, F. Nat. Biotechnol. 1999, 17, 974-978. (11) Milner, N.; Mir, K. U.; Southern, E. M. Nat. Biotechnol. 1997, 15, 537541. (12) Proudnikov, D.; Timofeev, E.; Mirzabekov, A. Anal. Biochem. 1998, 259, 34-41. (13) Chrisey, L. A.; Lee, G. U.; O’Ferrall, C. E. Nucleic Acids Res. 1996, 24, 3031-3039. (14) Rasmussen, S. R.; Larsen, M. R.; Rasmussen, S. E. Anal. Biochem. 1991, 198, 138-142. (15) Salo, H.; Virta, P.; Hakala, H.; Prakash, T. P.; Kawasaki, A. M.; Manoharan, M.; Lonnberg, H. Bioconjugate Chem. 1999, 10, 815-823. (16) Beier, M.; Hoheisel, J. D. Nucleic Acids Res. 1999, 27, 1970-1977. (17) Cheung, V. G.; Morley, M.; Aguilar, F.; Massimi, A.; Kucherlapati, R.; Childs, G. Nat. Genet. 1999, 21, 15-19. (18) Morozov, V. N.; Morozova, T. Ya. Anal. Biochem. 1999, 71, 3110-3117. (19) Fodor, S. P.; Read, J. L.; Pirrung, M. C.; Stryer, L.; Lu, A. T.; Solas, Science 1991, 251, 767-773. 10.1021/ac0352629 CCC: $27.50

© 2004 American Chemical Society Published on Web 01/24/2004

The way DNA molecules bind to a solid support is crucial for the success of any experiment on microarrays. In particular, the hybridization step on DNA microarrays is strongly influenced by (a) the characteristics of the immobilized probes, (b) the nature of targets, (c) the quality of the fluorophores for labeling targets, and (d) the surface characteristics of the microarray support. In particular, the latter plays a crucial role in providing quality results, high throughput, and low analysis cost. Immobilization of relatively long pieces of DNA (PCR fragments, cDNAs, etc.) can easily be obtained by electrostatic interactions between the negatively charged backbone of the DNA and the positive charges of the poly(L-lysine), electrostatically adsorbed onto glass.20 Alternatively, DNA with a 5′ modification can be covalently bound to surfaces modified with appropriate functional groups.21 Glass is one of the most widely used substrates for the immobilization of biomolecules. Organosilanization reactions are commonly performed to graft epoxy, amino, aldehyde, or other functional groups onto the surface. However, successful and reproducible deposition of a monolayer on a glass surface requires a strict control of operative parameters. The surface must be cleaned to remove contamination, to create surface attachment sites, and to control surface roughness. Numerous cleaning methods exist for glass substrates and these include gas plasmas, as well as combinations of acids, bases, and organic solvents that are allowed to react at different temperatures. Surfaces can also be smoothed or roughened using various techniques such as chemical deposition, grinding, polishing, and chemical etching.22 Despite the significant amount of investigation into it and because of the great number of variables involved, the silanization process is neither fully understood nor highly reproducible.23-26 Over the past few years, polymeric coatings have been developed based on polyacrylamide or poly(dimethylacrylamide) gels for regioselective immobilization by either the 3′ or 5′ end of short oligonucleotides.27-29 However, these coatings require glass silanization and careful control of operative parameters. Polymeric coatings have found wide application in biosensors. In the BIAcore system, for instance, the sensor surface is prepared by coupling the probe to dextran, using standard chemistries such as amine and thiol coupling. The molecule of interest is attached to carboxymethyl dextran polymers, thus lending a certain degree of mobility to the ligand. A simple and robust method that allows one to modify glass surfaces is of great interest. The present paper describes a method for the covalent attachment of target molecules onto the surface of a glass substrate such as beads, tubes, and microscope slides, (20) Schena, M.; Shalon, D.; Heller, R.; Chai, A.; Brown, P. O.; Davis, R. W. Proc. Natl., Acad. Sci. U.S.A. 1996, 93, 10614-10619. (21) Proudnikov, D.; Timofeev, E.; Mirzabekov, A. Anal. Biochem. 1998, 259, 34-41. (22) Henke, L.; Nagy, N.; Krull, U. J. Biosens. Bioelectron. 2002, 17, 547-555. (23) Haller, I. J. Am. Chem. Soc. 1978, 100, 8050-8055. (24) Nawrocki, J. Chromatographia 1991, 31, 177-192. (25) Silberzan, P.; Leger, L.; Aussere, D.; Benattar, J. J. Langmuir 1991, 7, 1647151. (26) Van Der Voort, P.; Vansant, E. F. J. Liq. Chromatogr. Relat. Technol. 1996, 19, 2723-2752. (27) Timofeev, E. N.; Kochetkova, S. V.; Mirzabekov, A. D.; Florentiev, V. L. Nucleic Acids Res. 1996, 24, 3142-314828 F. N. (28) Rehman, M. A.; Abrams, E. S.; Hammond, P. W.; Kenney, M.; Boles, T. C. Nucleic Acids Res. 1999, 27, 649-65529. (29) Guschin, D.; Yershov, G.-; Zaslavsky, A.; Gemmell, A.; Shick, V.; Proudnikov, D.; Arenkov, P.; Mirzabekov, A. Anal. Biochem. 1997, 250, 203-211.

allowing for a hybridization assay with fluorescent targets. This can be performed by the adsorption of copolymers with one of the monomers interacting directly with the surface and playing the role of an anchor and the other one extending in the solution and binding to the DNA target molecule. These polymers form a highly hydrophilic coating with accessible functionalities to which modified DNA can be covalently grafted. The coating was achieved by combining the adsorptive properties of poly(dimethylacrylamide) onto glass with a reaction between the silanol groups and electrophilic groups, pending from the polymer backbone. Copolymer composition was optimized to ensure an optimal combination of properties such as the following: (1) complete and uniform surface coverage of the surface with an ultrathin film, (2) a hydrophilic surface having minimum nonspecific attraction for biomolecules, (3) sufficient stability for use as the substrate for DNA microarray experiments, and (4) ease and reproducibility of the coating process. The structure of the polymeric film deposited on the surface was characterized by means of X-ray reflectivity (XRR), which is a surface-sensitive technique that provides information on mass density, thickness, and roughness of very thin films that are deposited on flat substrates. This measurement is based on the specular reflection of X-rays from planar surfaces. The reflected intensities show fringes that depend on the film thickness, and different modulation lengths correspond to the existence of different layers. The critical angle of total reflection is related to the mass density.30 MATERIALS AND METHODS Chemicals. N,N-Dimethylacrylamide, ethanolamine, 3-(trimethoxysilyl)propyl methacrylate, and sodium dodecyl sulfate (SDS) were from Merck (Darmstadt, Germany). Ammonium sulfate was from Sigma (St. Louis, MO). Tris(hydroxymethyl)aminomethane (Tris) was from Promega (Madison, WI). NaCl and sodium citrate were from Serva (Heidelberg, Germany). N-Acryloyloxysuccinimide was from Polysciences (Warrington, PA). Untreated glass microscope slides (25 × 75 mm) were purchased from Sigma. Oligonucleotides, 23, and 20 amino modified at the 5′ teminus, and a 24-mer Cy3-oligonucleotide labeled at the 5′ terminus were synthesized by MWG-Biotech AG (Ebevsberg, Germany). Synthesis of Copoly(N,N-dimethylacrylamide (DMA)Acryloyloxysuccinimide (NAS)) and Copoly(DMA-NAS-3(Trimethoxysilyl)propyl methacrylate) (MAPS)). The monomers were dissolved in 6 mL of dried tetrahydrofuran (THF) in a 25-mL, round-bottomed flask, equipped with condenser, magnetic stirring, and nitrogen connection. The solution was degassed by alternating a nitrogen purge with a vacuum connection, over a 30-min period. R,R′-Azoisobutyronitrile (AIBN) was added to the solution, which was then warmed to 50 °C and maintained at this temperature under a slightly positive pressure of nitrogen for 24 h. After the polymerization was completed, the solution was evaporated using a rotary evaporator and the white solid was dissolved in chloroform and precipitated by adding petroleum ether. The supernatant was eliminated, and the whole procedure (30) Bontempi, E.; Depero, L. E.; Sangaletti, L. Philos. Mag. B 2000, 80, 623633.

Analytical Chemistry, Vol. 76, No. 5, March 1, 2004

1353

was repeated two times. The polymer was dried under vacuum for 24 h at room temperature and stored at 4 °C. 13C NMR (DMSO), δ (ppm): 174.6 (backbone carbonyl), 166 (succinimide carbonyl) 40-30 (methylene carbons). The degree of succinimide insertion was determined from the ratio of the integrals of backbone and succinimide carbons, and the NAS molar fraction was found to be 0.015. The average molecular masses (Mw) and the polydispersity index of the different polymers were determined by gel permeation chromatography. Samples and standards were run in a 802.5, 804, 806 OH-PAK columns in series at 40 °C (Shodex) connected with UV Bruker detectors injecting 0.5% w/v sample solutions. The mobile phase was 100 mM NaCl, 50 mM NaH2PO4. Five polyacrylamide standards with Mw of 21 900, 58 400, 79 900, 400 000, and 600 000 were used to calibrate the GPC (Polysciences). Glass Slide Coating. The coating of glass slides requires two steps: (a) surface pretreatment and (b) adsorption of the polymer. In the first step, the slides were treated with 1 M NaOH for 30 min and 1 M HCl for 1 h, washed with water, and dried. In the second step, pretreated glass slides were immersed for 30 min in a solution of the polymer, (0.2-1% w/v in a water solution of ammonium sulfate at a 20% saturation level). The slides were then washed extensively with water and dried under vacuum at 80 °C. Oligonucleotides Immobilization. Custom synthesized 5′amine-modified oligonucleotides (100 nM/mL stock solution) were dissolved in 150 mM sodium phosphate buffer pH 8.5. Solutions of oligonucleotides at different concentrations were printed on coated slides to form microarrays using an Arrayt SpotBot spotter from Telechem. Printed slides were placed in an uncovered storage box placed in a sealed chamber, saturated with NaCl, and incubated at room temperature from 4 h to overnight. Fluorescent Determination of Nucleic Acid Attachment and Hybridization Density. Attachment Density. Twelve subarrays of 5′-amino-modified oligonucleotides labeled with Cy3 and dissolved in 150 mM sodium phosphate, pH 8.5, ranging in concentration from 2.5 to 25 µM were patterned using an Arrayt SpotBot spotter from Telechem. In each printed array, the diameter of the spots was ∼150 µm. The slides were placed in an uncovered storage box placed in a sealed chamber, saturated with NaCl, and incubated at room-temperature overnight. The slides were then shaken for 10 min in 2× SSC/0.1% SDS buffer, at 65 °C. After a brief rinse with 0.2× SSC and 0.1× SSC, the slides were scanned for florescence. Hybridization Density. After spotting the slides with oligonucleotide probes as reported above, the residual reactive groups of the coating were blocked by dipping the printed slides in 50 mM ethanolamine/0.1% SDS/0.1 M Tris pH 9.0 at 50 °C for 15 min. After discarding the blocking solution, the slides were rinsed two times with water and shaken for 15 min in 4× SSC/0.1% SDS buffer, prewarmed at 50 °C, and briefly rinsed with water. Oligonucleotide target (2.5 µL/cm2 of coverslip) was dissolved in the hybridization buffer (5× SSC/0.1% SDS/0.02% BSA) and immediately applied to microrrays. The slides, placed in hybridization chambers, were transferred to a humidified incubator at a temperature of 65 °C for 4 h. After hybridization, the slides were first washed with 4× SSC at room temperature to remove the coverslip and then with 2× SSC/0.1% SDS at hybridization temperature for 5 min. This 1354

Analytical Chemistry, Vol. 76, No. 5, March 1, 2004

operation was repeated twice and was followed by two washing steps with 0.2× SSC and 0.1× SSC for 1 min at room temperature. The slides were scanned with a Scan Array Express scanner from Packard Bioscience (Boston, MA). In both attachment and hybridization density measurements, row data of spot fluorescence intensity were converted to molecules per square centimeter using the standard curve of fluorescence made from a serial dilution of known amounts of oligonucleotides. The data reported are replicate of 16 spots for each oligonucleotide concentration. Model Experiment of Gene Expression. Arrays were produced using an Arrayt SpotBot spotter. Probes (70-mer oligoucleotides amino modified in 5′) for insulin, leptin, resistin, AP-2, and adipsin were printed. Each probe was spotted five times. Target cDNAs, labeled with Cy3 and Cy5, were independently generated as described by Amersham/Molecular Dynamics. A 25-µg amount of RNA (or 2.5 µg of polyA+ mRNA) and 1 µg of Oligo dT (Amersham), were used in each reaction. The resultant cDNA targets labeled with Cy3 and Cy5 were then purified using a PCR purification kit (Qiagen), as described by the manufacturer and mixed together just before hybridization. RESULTS AND DISCUSSION Polymer Characteristics. This work describes a hydrophilic surface environment made of a thin layer of a hydrophilic linear copolymer that self-adsorbs to a glass surface from a poor solvent. Two copolymers were synthesized, characterized, and tested for their ability to form films by adsorption on glass substrates. The density and the thickness of the films obtained with both polymers were evaluated by XRR. One copolymer, copoly(DMA-NAS), was obtained by radical copolymerization of dimethylacrylamide with acryloyloxysuccinimide in THF, catalyzed by AIBN. The major polymer constituent, responsible for glass self-adsorption, is dimethylacrylamide, while the second monomer, acryloyloxysuccinimide, is responsible for covalent binding of modified DNA molecules. The other polymer, copoly(DMA-NAS-MAPS), is a ter-copolymer, containing a third constituent, the 3-(trimethoxysilyl)propyl methacrylate, which increases the strength of the binding of the polymer with the glass. The layer of the adsorbed copolymer on the glass surface has a double function: to block the glass surface sites by making hydrogen bonds and to control the interactions with the external medium by bearing specific chemical functions. Both copolymers form films on glass surfaces when adsorbed from an aqueous solution. However, functional and structural characterization of the copoly(DMA-NAS-MAPS) film showed that this coating is more suitable for application in DNA microarrays, because of its ability to form more stable and thicker films. Indeed, the silane functionalities pending from the backbone induce, upon adsorption, silylation of the silanol groups on the glass surface, thus increasing the strength of the binding with the glass (Figure 1). This results in the formation of a layer with increased stability. The molar mass distribution and the polydispersity index (Id) of the two polymers are reported in Table 1. The presence oforganic solvents as the polymerization medium, which is necessary to prevent the hydrolysis of the active ester monomer, and the high initiation temperature lead to the formation of polymers with a relatively low molecular weight. However, to

Figure 1. Schematic representation of the copoly(DMA-NAS-MAPS) coating.

Table 1. Characteristics of Polymers

MW polydispersity

poly (DMA-co-NAS-coMAPS)

poly (DMA-co-NAS)

150 000 3.2

120 000 3.4

exploit the self-adsorbing properties of poly(dimethylacrylamide), the molecular weight (Mw) cannot be lower than 120 000. Both polymers, dissolved in an aqueous solution, quickly adsorb onto a glass surface, typically in 5-30 min, generating a layer that is stable in an aqueous buffer containing various additives (SDS, urea, salts) even at high temperatures. The procedure requires the immersion of the glass slide into the polymer dilute solution. The only pretreatment required for the glass is the alkaline activation of surface silanols, which maximizes the number of hydrogen bonds with carbonyl groups on the polymer. Glass substrates coated with the two copolymers were used as substrates for the preparation of low-density DNA microarrays. The functional tests revealed differences in terms of DNA binding capacity and probe accessibility in the two cases. The most suitable environment for hybridization was the one provided by the film formed with copoly(DMA-NAS-MAPS). The coatings were characterized by X-ray reflectivity, a technique based on the reflection process of X-rays from flat

surfaces, that takes place close to a total reflection angle. Thanks to the very small penetration depth of the X-rays obtained in these conditions, this technique is very sensitive to the surface. The XRR patterns for thin films exhibit interference fringes that originate from the multiple reflection of a X-ray at the film substrate interface. These fringes, beyond the region for total reflection, are related to the thickness of the film and surface and interface roughness. Moreover, from the value of the critical angle, the mass density can be evaluated. Influence of Solvent Type and Polymer Concentration. The influence of the polymer solvent on coating formation was assessed. Polymer solutions were prepared at 1% w/v concentration in water/aqueous ammonium sulfate at a 20% saturation level and THF. Glass slides coated with these polymer solutions as described above were then used in microarray experiments. The slides coated with the polymer dissolved in aqueous ammonium sulfate provided a better performance with highly fluorescent spots, whereas the use of organic solvents dramatically reduced the amount of polymer adsorbed onto the surface. Adsorption of the polymer involves a nonspecific interaction with the surface including hydrogen bonding, hydrophobic interaction, and van der Waals forces. The affinity of a given polymer for the surface depends on the nature of the solvent from which adsorption takes place. The use of ammonium sulfate facilitates removal of water molecules from the vicinity of the Analytical Chemistry, Vol. 76, No. 5, March 1, 2004

1355

Figure 2. Attachment density of oligonucleotides onto coated slides as a function of the polymer concentration in the coating solution. Quantitations of the attachment density are presented as the fluorescence intensity detected after washing the slides at different concentrations of spotted oligonucleotide and detected with a fluorescence scanner as described under Materials and Methods. Cy3-5′ amino-modified oligonucleotides were attached to the surface at different concentrations as described under Materials and Methods.

surface and promotes the formation of hydrophobic interactions between glass and polymer segments. On the contrary, THF disfavors adsorption as the affinity of the polymer for the solvent is higher than that of the glass substrate. In view of these results, the density of the oligonucleotide attachment and the X-ray reflectivity measurements were performed on slides coated with 1% w/v polymer in a water solution of ammonium sulfate at a 20% saturation level. The influence of the coating solution concentration on the surface binding capacity was then investigated. 5′-Amino-modified oligonucleotides labeled with Cy3, were immobilized on the slides coated with solutions of the two polymers dissolved in ammonium sulfate at a 20% saturation level, at three different concentrations, 1, 2, and 5% w/v. The slides were then washed and scanned. Figure 2 shows the effect of copoly(DMA-NAS) concentration on the attachment of oligonuceotides. Similar results were also obtained with copoly(DMA-NAS-MAPS), (data not shown). One percent w/v polymer concentration provided spots with a more intense fluorescence signal. In view of this result, in all the experiments that follow, a 1% w/v polymer was used to coat the slides. Density of Oligonuclotide Attachment. We have tested the immobilization of oligonucleotides through a reaction between amino groups and active esters (Figure 3) with combinations of different oligonucleotide concentrations (ranging from 1 to 25 µM). The number of probe molecules per unit area within an individual probe site was determined by attaching to coated slides an amino-modified oligonucleotide (23 mer), fluorescently labeled with Cy3 as described in the materials and methods section. Quantitative measurements of the attachment density were obtained using a LIF scanner as reported in the Materials and Methods section. 1356

Analytical Chemistry, Vol. 76, No. 5, March 1, 2004

Figure 3. Attachment density of oligonucleotides onto (a) copoly(DMA-NAS-MAPS) and (b) copoly(DMA-NAS) coated slides. Quantitations of the attachment density as in Figure 2.

The results, summarized in Figure 3, represent average values of 16 spots for each concentration. Significant differences between the two coatings were observed. The copoly(DMA-NAS) film presents a maximal attachment density of 0.9 × 1013 molecules/ cm2, whereas the copoly(DMA-NAS-MAPS) shows a maximal density of 4 × 1013 molecules/cm2. No significant signal was detected for the non-amino-modified oligonucleotide control. The attachment density provided by copoly(DMA-NASMAPS) film is 1 order of magnitude higher than that obtained with classical 2D coatings involving silanization procedures.31 The number of probe molecules in a given spot is important as it controls their intramolecular interactions. Only 3D surface struc(31) Pirrung, C. M. Angew. Chem., Int. Ed. 2002, 41, 1276-1289.

Table 2. XRR Resultsa bulk

substrate sample 1 sample 2

layer 1

layer 2

layer 3

density (g/cm3)

density

thickness

density

thickness

density

thickness

2.37 2.37 2.37

2.14 2.14 2.14

2 0.72 0.6

1.1 1.8

2.06 3.45

1.12

1.8

a The substrate represents the results obtained by XRR simulation of Ultraclean Glass substrate; samples 1 and 2 refer to the XRR results obtained for copoly(DMA-NAS) and copoly(DMA-NAS-MAPS) film, respectively.

Figure 4. Efficiency of the hybridization of a synthetic target to immobilized probes on glass slides. Quantitations of the hybridization density are presented as the fluorescence intensity from the labeled target and measured with a fluorescence scanner as described under Materials and Methods. The probes were attached to the slides and hybridized as reported in Materials and Methods. The curve reports the hybridization of the target to various concentrations of probe oligonucleotide attached onto a glass slide surface coated with copoly(DMA-NAS-MAPS).

Figure 5. Hybridization density of a synthetic oligonucleotide when hybridized to a probe oligonucleotide with varying attachment densities on the glass surface.

tures made of acrylamide32 have a greater probe density because of the formation of a tridimensional matrix on the glass surface. However, the procedure for creating the matrix is complex and hard to control. In addition, changing buffers and accessing probes within the gel fibers is difficult. Another set of experiments was performed to evaluate the fluorescence signal produced in a oligo-oligo hybridization model experiment. This test is of great importance as the major concern in oligonucleotide arrays preparation is the accessibility of the probe for hybridization, as well as the specificity obtained in the interaction. A 22-mer probe oligonuclotide, ranging in concentration from 2.5 to 45 µM and a 20-mer negative control were spotted and hybridized with the complementary 24-mer Cy3-labeled oligonuclotide in a total volume of 25 µL at a 1 µM concentration. The results obtained with copoly(DMA-NAS-MAPS) are summarized in Figure 4, which reports the hybridization density as a function of probe concentration. The results in Figure 5 show that the hybridization efficiency is strongly related to the probe attachment density. On average, 80% of the surface-bound probe was hybridized to the target template. No detectable nonspecific hybridization signal was seen with the noncomplementary control oligonucleotide probe (data not shown). The results in Figure 6 show the variation of spot fluorescence intensity as a function of target concentration. The signal increases linearly up to a target concentration of 1 µM.

Finally, a hybridization with probes, typically used in gene expression experiments, was carried out in order to evaluate spot quality and background interference. Figure 7 shows an array of five oligonucleotide probes (70 mer), amino-modified in 5′, hybridized with complementary cDNA targets labeled with Cy3. The slide, containing five replicates of each probe, is characterized by a low background and by a consistent spot morphology. Characterization of Thin Film by X-ray Reflectivity. XRR measurements and simulations were performed on copoly(DMANAS) and copoly(DMA-NAS-MAPS) to evaluate the mass density and the thickness of the deposited film. Only few papers have reported on the analysis of films deposited on glass surfaces by XRR (see, for example, ref 33-35). In most cases, the structural

(32) Zlatanova, J.; Mirzabekov, A. Methods in Molecular Biology; Rampal. J. B., Ed.; Humana Press: Totowa, NJ, 2001; Vol. 170, pp 17-38.

(33) Stone, V. W.; Jonas, A. M.; Nysten, B.; Legras, R. Phys. Rev. B 1999, 60, 5883-5894.

Figure 6. Fluorescence intensity of spots containing 10 fmol of oligonucleotide probe and hybridized with different concentrations of target. Spotting, hybridization, and detection conditions are described in Materials and Methods.

Analytical Chemistry, Vol. 76, No. 5, March 1, 2004

1357

Figure 8. XRR experimental and simulated spectra obtained for copoly(DMA-NAS-MAPS) film.

Figure 7. Background and spot morphology in a gene expression experiment using oligonucleotide probes for insulin, leptin, resistin, AP-2 e, and adipsin (from the top to the bottom). Each lane contains five replicates of the same probe. The target molecules are a mixture of cDNA fragments labeled with Cy3 and Cy5.

characterization is performed on film deposited on silicon singlecrystal substrates, which are more flat in comparison to glass, and thus XRR measurements are easier to perform and to simulate. The characterization of film layers deposited on rough surfaces is a challenging task, since the surface can strongly influence their characteristics and thus the performance of the film. In the present case, Micromax Ultraclean glass slides from Perkin-Elmer, annealed at 400 °C for 3 h, were used as the substrate for film deposition. The XRR experimental spectrum shows a long modulation in the pattern that can be ascribed to one single superficial layer. This layer, of ∼1 nm, shows a mass density lower than that of glass (∼2 g/cm3), as expected, because of the smaller refraction index of amorphous silica.36 Two different types of samples were deposited on Micromax Ultraclean substrates according to the procedure described in the Materials and Methods section using in one case copoly(DMANAS) and in the second copoly(DMA-NAS-MAPS) copolymers. In the case of the XRR spectrum obtained from the copoly(DMA-NAS) film, it was necessary to also consider the layer found in the substrate in order to achieve good simulation (see Table 2). From the fitting, the thickness and the mass density of the film were 2 nm and 1.1 g/cm3, respectively. The XRR experimental and simulated spectra obtained for copoly(DMA-NAS-MAPS) film are shown in Figure 8. In this case, three layers must be considered to obtain a good simula(34) Weis, H.; Muggenburg, T.; Grosse, P.; Herlitze, L.; Friedrich, I.; Wuttig, M. Thin Solid Films 1999, 351, 184-189. (35) Peruzzi, S.; Bontempi, E.; Versace, C.; Depero, L. E. Mol. Cryst. Liq. Cryst. 2002, 372, 339-352. (36) Schalchli, A.; Benattar, J. J.; Licoppe, C. Europhys. Lett. 1994, 26, 271276.

1358 Analytical Chemistry, Vol. 76, No. 5, March 1, 2004

tion: the layer belonging to the substrate and two layers to describe the poly(DMA-NAS-MAPS) film. It is noteworthy that while the upper layer shows a mass density similar to that of the copoly(DMA-NAS) film, the bottom layer has a significant higher mass density (1.80 g/cm3). This may be ascribed to the efficient condensation of the adsorbed polymer chains with the silanols of the substrate. At the surface-air interface, the condensation is probably lower because of the loops and tails formed by the adsorbed polymer chains. CONCLUSIONS A new method was developed for the immobilization of DNA probes. The glass surface coated by the adsorbed polymers offers the following advantages: (a) high probe density and (b) reduction of nonspecific adsorption of interfering biomolecules onto the surface. The method is robust and fast and provides a cost-effective immobilization procedure for biological source materials. From the thickness and the mass density of the films, obtained by XRR analysis, it was possible to calculate the density of active esters on the surfaces. For copoly(DMA-NAS-MAPS) the number of active esters per square centimeter was found to be 8.9 × 1013. This value was calculated by taking into account film thickness and mass density and the molar fraction of NAS in the polymer. Following the same procedure, in the case of copoly(DMA-NAS), a density of 2.8 × 1013 molecules of NAS/cm2 was calculated. The number of functional groups per surface unit obtained from XRR data are in good agreement with those obtained by the oligonucleotide binding assay. Indeed, the functional test indicates that about half of the active groups present on the surface, 4 × 1013/cm2 for copoly(DMA-NAS-MAPS) and 0.9 × 1013 for copoly(DMA-NAS), reacts with the oligo probes, which indicates an excellent accessibility of surface functional groups. Received for review December 17, 2003. AC0352629

October

27,

2003.

Accepted