Langmuir 2004, 20, 5539-5543
5539
Optical Properties of an Immobilized DNA Monolayer from 255 to 700 nm Selim Elhadj, Gaurav Singh, and Ravi F. Saraf* Department of Chemical Engineering, Virginia Tech, Blacksburg, Virginia 24061 Received February 9, 2004. In Final Form: April 23, 2004 The real (n) and imaginary (k) refractive indices of an immobilized monolayer of 27 nucleotide (nt) single stranded DNA (ssDNA) and the corresponding double stranded DNA (dsDNA) are measured in the 255700 nm range. Multiple techniques are used to obtain consistent estimation. The coverage is ∼6.5% with an average interchain distance of tethered ssDNA molecules of ∼11.8 nm, which is significantly larger than the “footprint” of the chain on the surface. The measured increase in n by ∼5% between the ssDNA and the dsDNA is 20% smaller than the expected change due to doubling of the molecular weight. The change in k is not significant, indicating that the electron delocalization effect expected in dsDNA due to base pair stacking is not important at optical frequencies.
Introduction The optical properties of immobilized DNA monolayers are of great importance as microarray methods, also called DNA chip methods, become increasingly pervasive in biosciences and biotechnology.1-7 However, the first claimed estimate of the refractive index of an immobilized DNA monolayer was surprisingly recent, where a refractive index of n ) 1.462 of an immobilized single stranded DNA (ssDNA) monolayer of thickness d ∼ 3.2 nm was estimated at a fixed wavelength of 633 nm using ellipsometry.8 The real and imaginary refractive indices, n and k, respectively, of ssDNA or double stranded DNA (dsDNA) in the full UV-vis spectrum, where most of the optical probes operate, are not reported.9 In a DNA chip, probe molecules of a known ssDNA sequence are tethered (i.e., immobilized) on an array of 100-500 µm size spots on a solid substrate called the chip. By detecting specific binding between the probes and unknown DNA/RNA fragments (target) from a biosystem, one can perform a combinatorial sequence analysis of the latter. Typically, ∼100 nt probes are used for gene expression and ∼25 nt probes for sequence analysis.10,11 The pervasive method used to detect the binding event is done by tagging the target with a fluorescent dye.4,7,12 However, the labels are expensive with 60-80% of material cost, their performance deteriorates due to photobleaching and photochemical degradation, and they can significantly alter the specificity of probe-target binding.13 A promising emerging alterna* To whom correspondence should be addressed. (1) Duggan, D. J.; Bittner, M.; Chen, Y. D.; Meltzer, P.; Trent, J. M. Nat. Genet. 1999, 21, 10-14. (2) Hacia, J. G. Nat. Genet. 1999, 21, 42-47. (3) Lipshutz, R. J.; Fodor, S. P. A.; Gingeras, T. R.; Lockhart, D. J. Nat. Genet. 1999, 21, 20-24. (4) Lockhart, D. J.; Winzeler, E. A. Nature 2000, 405, 827-836. (5) Marshall, A.; Hodgson, J. Nat. Biotechnol. 1998, 16, 27-31. (6) Schena, M.; Shalon, D.; Heller, R.; Brown, P. O.; Davis, R. W. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 10614-10619. (7) Southern, E. M. Trends Genet. 1996, 12, 110-115. (8) Gray, D. E.; Case-Green, S. C.; Fell, T. S.; Dobson, P. J.; Southern, E. M. Langmuir 1997, 13, 2833-2842. (9) Kricka, L. J. Ann. Clin. Biochem. 2002, 39, 114-129. (10) Lockhart, D. J.; et al. Nat. Biotechnol. 1996, 14, 1675-1680. (11) Schena, M.; Shalon, D.; Davis, R. W.; Brown, P. O. Science 1995, 270, 467-470. (12) Chee, M.; et al. Science 1996, 274, 610-614. (13) Naef, F.; Lim, D. A.; Patil, N.; Magnasco, M. Phys. Rev. E 2002, 65, 040902.
tive solution that does not require tags is to detect the probe-target binding event on the basis of techniques that rely on the change in the refractive index due to hybridization. Surface plasmon resonance (SPR) spectroscopy14 is perhaps the most popular method that uses the optical property change, primarily due to its long history as an analytical tool in life sciences.15 SPR spectroscopy is also demonstrated to determine number mismatches by measuring the kinetics of binding.16 The ability to measure kinetics is a feature that the fluorescence method does not have. Ellipsometry17,18 and optical reflectivity19 are other label-free detection methods that also rely on the change in the optical property upon probetarget binding. The change in the absolute n value as the immobilized monolayer transforms from ssDNA to dsDNA is expected to be significant due to the change in the monolayer density and polarizability of the molecule. Densification occurs due to the increase in molecular mass that occurs when the DNA transforms from ssDNA to dsDNA. Perhaps, a more interesting effect is the significance of the possible onset of electron delocalization as the DNA transforms from ssDNA to dsDNA. Ever since the first suggestion of long range electron transport along the dsDNA chain,20 several studies have reported semiconducting- to conducting-like electronic behavior in dsDNA.21-26 The interpretations of these measurements are not fully realized because they are complicated by contact resistance, conformation of the DNA, the possibility of trapped ions, (14) Thiel, A. J.; Frutos, A. G.; Jordan, C. E.; Corn, R. M.; Smith, L. M. Anal. Chem. 1997, 69, 4948-4956. (15) Englebienne, P.; Van Hoonacker, A.; Verhas, M. Spectrosc.: Int. J. 2003, 17, 255-273. (16) Heaton, R. J.; Peterson, A. W.; Georgiadis, R. M. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 3701-3704. (17) Gray, D. E.; Case-Green, S. C.; Fell, T. S.; Dobson, P. J.; Southern, E. M. Langmuir 1997, 13, 2833-2842. (18) Ostroff, R. M.; et al. Clin. Chem. 1998, 44, 2031-2035. (19) Jenison, R.; Yang, S.; Haeberli, A.; Polisky, B. Nat. Biotechnol. 2001, 19, 62-65. (20) Murphy, C. J.; et al. Science 1993, 262, 1025-1029. (21) Bezryadin, A.; Dekker, C.; Schmid, G. Appl. Phys. Lett. 1997, 71, 1273-1275. (22) Fink, H. W.; Schonenberger, C. Nature 1999, 398, 407-410. (23) Fink, H. W. Cell. Mol. Life Sci. 2001, 58, 1-3. (24) Porath, D.; Bezryadin, A.; de Vries, S.; Dekker, C. Nature 2000, 403, 635-638. (25) Watanabe, H.; Manabe, C.; Shigematsu, T.; Shimotani, K.; Shimizu, M. Appl. Phys. Lett. 2001, 79, 2462-2464. (26) Yoo, K. H.; et al. Phys. Rev. Lett. 2001, 8719, article no. 198.
10.1021/la049653+ CCC: $27.50 © 2004 American Chemical Society Published on Web 05/26/2004
5540
Langmuir, Vol. 20, No. 13, 2004
and bound water. Nevertheless, several interesting molecular electronic devices, such as nanowires,27-29 transistors,25,30,31 and magnetic spin valves,32 have been suggested and attempted on the basis of DNA’s charge transport properties. One of the more discussed models prescribes that the stacking of the base pairs at the core of the double helix forms a “conduit” for the charges to propagate.33 The various models suggest that the transport occurs by sequential or coherent tunneling of G+, that is, hole carriers over consecutive stacks of A-T base bridges,34-36 where A, T, G, and C are adenine, thymine, guanine, and cytosine, respectively, the four bases that hold the two strands of DNA together via hydrogen bonding. In other words, dsDNA is a p-type semiconductor. Low temperature studies indicate that the conduction can even lead to superconductivity.37 The exact nature and mechanism of conduction are still not fully resolved; however, it is accepted in the literature that the optical and electronic properties of immobilized/self-assembled ssDNA and dsDNA are central to the next generation, for molecular electronics and possibly spintronics devices using DNA. Experimental Section DNA Grafting and Hybridization on a Si Substrate with a Native Oxide. DNA was grafted on a SiO2 (native)/Si substrate in a four-step process: (i) The Si wafer is treated in a H2SO4/ H2O2 (3:1) “piranha” solution to form a Si-OH functionalized surface. (ii) Next, a monolayer of 3-aminopropyl trimethoxysilane (UCT) is immobilized on the substrate from a 0.1% v/v solution in DI water for 30 min followed by curing at 105 °C. (iii) Subsequently, the amino end of the silane is reacted with a bifunctional cross-linker called Sulfo-SMCC (Pierce) to form an amide linkage. (iv) Finally, a sulfhydryl-terminated ssDNA (Integrated DNA Technologies, IL) probe is covalently bonded to the SMCC. The ssDNA probe had a six-carbon length with ATA nucleotides as the spacer at the grafted end. The sequence for the 27 nt ssDNA is 5′-/5thioMC6/ATA GTT TTC GTT GCG TAA GCG TCT ATT/-3′, while the complementary target ssDNA sequence is 5′-CAA AAT AGA CGC TTA CGC AAC GAA AAC /-3′. The hybridization of the target ssDNA was performed overnight at 45 °C in a humidified incubator. The sample was subsequently washed with cleaning buffer and briefly with water to remove any unbound complementary DNA and salts. The resultant monolayer is subsequently referred to as the dsDNA monolayer. Measurement of the Number of Immobilized ssDNA Molecules per Square Centimeter and the Extent of Hybridization. Cy5 labeled DNA was used to determine the surface density of the grafted ssDNA monolayer. The labeled ssDNA with an identical sequence to that above is grafted on a Si substrate using the immobilization chemistry described above. By measuring the fluorescence and normalizing the intensity by emission per dye molecule, one can determine the coverage (i.e., the number of molecules per unit area). The emission per dye molecule is obtained by calibrating the fluorescent intensity over a range of concentrations in a phosphate buffer of pH 7.2. The (27) Braun, E.; Eichen, Y.; Sivan, U.; Ben Yoseph, G. Nature 1998, 391, 775-778. (28) Harnack, O.; Ford, W. E.; Yasuda, A.; Wessels, J. M. Nano Lett. 2002, 2, 919-923. (29) Richter, J.; Mertig, M.; Pompe, W.; Monch, I.; Schackert, H. K. Appl. Phys. Lett. 2001, 78, 536-538. (30) Ben Jacob, E.; Hermon, Z.; Caspi, S. Phys. Lett. A 1999, 263, 199-202. (31) Hermon, Z.; Caspi, S.; Ben Jacob, E. Europhys. Lett. 1998, 43, 482-487. (32) Zwolak, M.; Di Ventra, M. Appl. Phys. Lett. 2002, 81, 925-927. (33) Warman, J. M.; deHaas, M. P.; Rupprecht, A. Chem. Phys. Lett. 1996, 249, 319-322. (34) Berlin, Y. A.; Burin, A. L.; Ratner, M. A. J. Phys. Chem. A 2000, 104, 443-445. (35) Berlin, Y. A.; Burin, A. L.; Ratner, M. A. J. Am. Chem. Soc. 2001, 123, 260-268. (36) Yu, Z. G.; Song, X. Y. Phys. Rev. Lett. 2001, 86, 6018-6021. (37) Kasumov, A. Y.; et al. Science 2001, 291, 280-282.
Elhadj et al. Table 1. Structural Parameters for DNA Obtained from Fitting of the Ellipsometry Data and the Independently Measured Experimental Parameters Lorentz Model for Ellipsometry Data oscillator energy, En (eV) oscillator strength, Am (eV2) oscillator bandwidth, Br (eV) dielectric constant, (∞) film thickness (nm) kss and kds at 260 nm from ellipsometry fit
ssDNA
dsDNA
4.87 0.5 0.184 2.1 6.5 0.083
4.87 0.5 0.184 2.3 5.7 0.079
Experimentally Measured Parameters oscillator energy from UV-vis spectrum kss and kds at 260 nm from UV-vis spectrum
ssDNA
dsDNA
4.81 0.074
4.81 0.072
hybridization efficiency of the dsDNA monolayer is estimated by a similar method except in this case, the immobilized ssDNA has no tag, while the target ssDNA has the Cy5 label. For each set of experiments, the ssDNA is immobilized on two substrates simultaneously. The two samples are used to determine the immobilization density and the hybridization efficiency, respectively. Ellipsometry Measurements and Modeling of Optical Properties. A Woollam variable angle scanning ellipsometry (VASE) system38 was used to determine the optical properties and thicknesses of the ssDNA and dsDNA monolayers on the Si substrate. The 25 mm diameter Si substrates were examined under a Nomarski phase contrast microscope to ensure a large area of uniform coloration, indicating uniformity in monolayer coverage. Since the diameter of the ellipsometer beam is ∼1 mm, more than three spots on the uniform coverage region of the sample were analyzed. We performed an ellipsometry scan in the UV-vis range for an incident wavelength, λ ) 250-700 nm. The scans were performed at angles of incidence with respect to the film normal of 70, 72, and 74°, which are close to the to the Brewster angle of the Si substrate at ∼70°. The data were fit to the measured ellipticity (Ψ°) and phase (∆°) of the reflected “s” and “p” polarized light. The control samples of the blank substrate with only Piranha treatment and after SMCC cross-linker immobilization were characterized to obtain average thicknesses (with high reproducibility) of 2.3 and 1.8 nm for the native oxide and the cross-linker, respectively. We specifically note that, upon dehybridization, dsDNA samples reversibly change to the “original” immobilized ssDNA monolayer. The average fitting parameters are quoted in Table 1, and the corresponding curve is shown in Figure 1. Reflectivity Measurements and Modeling. An optical reflectometer is assembled to measure normal reflectivity from a solid surface as a function of λ. A position sensitive detector is used to record the complete spectrum of the reflected light in
