Open Access Original article

Thymine/adenine diblock-oligonucleotide monolayers and hybrid brushes on gold: a spectroscopic study

Caitlin Howell123*, Hicham Hamoudi24 and Michael Zharnikov2*

Author Affiliations

1 Institute of Toxicology and Genetics, Karlsruhe Institute of Technology, Hermann-von-Helmholtz-Platz 1, 76344, Eggenstein-Leopoldshafen, Germany

2 Applied Physical Chemistry, University of Heidelberg, Im Neuenheimer Feld 253, 69120, Heidelberg, Germany

3 Present address: School of Engineering and Applied Sciences, Harvard University, 29 Oxford St, Cambridge, MA 02138, USA

4 Present address: International Center for Materials Nanoarchitectonics, Institute for Materials Science, Namiki 1-1, Tsukuba, Ibaraki, Japan

For all author emails, please log on.

Biointerphases 2013, 8:6  doi:10.1186/1559-4106-8-6


The electronic version of this article is the complete one and can be found online at: http://www.biointerphases.com/content/8/1/6


Received:14 December 2012
Accepted:13 February 2013
Published:27 February 2013

© 2013 Howell et al; licensee Springer.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Background

The establishment of spectroscopic analysis techniques for complex, surface-bound biological systems is an important step toward the further application of these powerful experimental tools to new questions in biology and medicine.

Methods

We use a combination of the complementary spectroscopic techniques of X-ray photoelectron spectroscopy, Infrared reflection-absorption spectroscopy, and near-edge x-ray absorption fine structure spectroscopy to monitor the composition and molecular orientation in adenine/thymine diblock oligonucleotide films and their hybridized brushes on gold.

Results

We demonstrate that the surface-bound probe molecules, consisting of a binding adenine block, d(A), and a sensing thymine block, d(T), deviate from the ideal L-shape model due to the internal intra- and intermolecular hybridization. This effect becomes more pronounced with increasing length of the d(A) block. Nevertheless, these films were found to hybridize well with the complementary target d(A) strands, especially if they were treated in advance to reduce internal interaction between the molecules. In spite of the structural complexity of these films, the hybridization efficiency correlated well with the potential accessibility of the sensing d(T) blocks, defined by their lateral spacing.

Conclusions

These findings are a good demonstration of the strength of multi-technique spectroscopic analysis when applied to assemblies of biological molecules intrinsically prone to complex interactions.

Keywords:
ssDNA film; Diblock-oligonucleotide; HRXPS; NEXAFS; IRRAS

Background

Understanding the properties of and processes involving biological molecules on surfaces and interfaces is at the heart of many critical issues in biology and medicine [1-3]. The organization and hybridization of surface-bound DNA in particular is an important component in many established, new, and emerging medical and other biology-based technologies, including microarrays [3-5], DNA-decorated nanoparticles [6,7], and DNA computing [8,9]. In systems such as these, the tethering of the DNA strands to a solid surface is necessary for greater control over density, orientation, and hybridization processes. Nevertheless, precise measurement or reconstruction of these parameters and processes through conventional means such as microscopy and fluorescent labeling [10] or surface plasmon resonance (SPR) spectroscopy [11,12] can be challenging due to the nanoscale nature of the DNA molecules involved as well as the non-specific character of these techniques.

Different types of spectroscopy have been shown to be useful for such nanoscale analysis, as they inherently give specific, label-free, and molecular-level information. Techniques such as X-ray photoelectron spectroscopy (XPS) [13,14], near-edge X-ray absorption fine structure (NEXAFS) spectroscopy [15-18], time-of-flight secondary ion mass spectrometry (ToF-SIMS) [19-21], sum-frequency-generation (SFG) spectroscopy [22,23], and infrared reflection-absorption spectroscopy (IRRAS) [24,25] can be used to determine composition, content, molecular orientation and hybridization behavior of DNA assemblies. Yet despite the large amounts of potential information obtainable through use of these methods, the interpretation of spectra from biological samples, including surface-bound DNA, can be difficult, often resulting in the restriction of the use of these techniques to simplified or relatively straightforward systems. In order to expand the use of spectroscopic methods, comparative studies on biological molecules capable of acting in increasingly complex ways are needed.

In this context, Opdahl and co-workers [24] recently reported a method of controlling the grafting density in films of single-stranded DNA (ssDNA) on gold substrates by exploiting the fact that adenine nucleotides have a higher affinity for gold than thymine nucleotides. They created films in which DNA molecules consisting of a block of adenine nucleotides connected to a block of thymine nucleotides adsorbed in a (presumably) L-shaped configuration on a gold surface. In an ideal version of this system, illustrated in Scheme 1, the d(A) block serves as the binding unit which can be shortened or lengthened to increase or decrease the density, respectively. The remaining d(T) block, either acting as or connected to the sensing portion of the molecule, is free to protrude into ambient, associate with the surface-bound d(A) block, or hybridize with a complimentary target [11,12]. In reality, a systematic lengthening of the d(A) block also brings increasing complexity to the system, changing the layer from being relatively tightly-packed and well-defined to one in which more space exists between the protruding d(T) blocks. This allows both more freedom of movement and a greater probability of the formation of stable inter- or intra-molecular hybrids between the complimentary d(A) and d(T) portions of the molecules. Such systematic changes may affect the number and quality of any hybrids that are then formed by introduction of a complimentary target strand, perhaps the most important consideration in the application of these molecules to biotechnology and medicine.

Scheme 1

Idealized representation of the (dA)n-(dT)15 monolayers (n = 5, 10, 15) before and after hybridization with the A15 target. The molecules used in this study are taken as representative examples of adenine-thymine diblock-oligonucleotide films.

Here, we use the abovementioned diblock-oligonucleotides to systematically increase the degree of complexity in DNA hybrid brushes. Using a combination of high-resolution XPS, IRRAS, and NEXAFS spectroscopy to gather complimentary information on molecular density, content, and orientation, we present a molecular-level picture of the effects of increasing complexity on these parameters. We find that as the DNA hybrid systems become capable of more complex interactions the predictability of orientation becomes increasingly difficult, even as the molecular content continues to change as expected. These results highlight the importance of using multiple spectroscopic techniques to gain a complete picture of parameters of and processes involving biological systems on the molecular scale, especially as the level of complexity increases.

Methods

Materials

Gold substrates were produced by using radio frequency magnetron sputtering to coat silicon wafers (Si-Mat Silicon Materials, Kaufering, Germany) with a chromium adhesion later, followed by a gold layer. The result was a 100 nm-thick polycrystalline gold film with a predominant (111) orientation. The coated wafers were cut into appropriate sizes, cleaned by exposure to UV/ozone (UV cleaner 42–220, Jelight) for 2.5 hrs, and rinsed thoroughly with HPLC-grade water immediately prior to film deposition.

High performance liquid chromatography-purified DNA sequences were purchased from Sigma–Genosys (Steinheim, Germany): 5’-(dA)5-(dT)15-3’, 5’-(dA)10-(dT)15-3’, 5’-(dA)15-(dT)15-3’, probe sequences, as well as a complimentary (dA)15 target. These will be referred to hereafter as A5T15, A10T15, A15T15, and A15, respectively (see Scheme 1). Block-DNA containing only adenine and thymine was chosen over block-DNA containing a random sequence of all four nucleobase types in order to facilitate interpretation of the spectra. The Tris–HCl, EDTA, and NaCl used in the deposition and hybridization buffers were obtained from Sigma-Aldrich (Seelze, Germany).

DNA Immobilization and hybridization

DNA brushes were formed by incubating gold wafers for 40 h at 37°C in 1 M CaCl2–TE buffer (10 mM Tris–HCl and 1 mM EDTA, pH=7.08) containing a 3 μM DNA solution. After incubation, unhybridized samples were rinsed for 1 min under flowing deionized water (resistivity >18.2 MΩ cm) to remove excess DNA and dried with N2[13].

Samples intended for hybridization were rinsed for either one full minute under deionized water or for 30 seconds under deionized water, followed by 30 seconds under flowing 0.1 M NaOH, followed again by 30 seconds under deionized water. The NaOH rinse was added in an effort to destabilize any partial inter-molecular hybrids and improve overall hybridization and orientation [12]. After rinsing, these samples were placed in a 1M NaCl buffer containing 3 μM of the (dA)15 target sequence for 7 h at room temperature. Following incubation, samples were rinsed with 1 M NaCl buffer for 1 min, briefly dipped in deionized water to remove excess salts [26], and finally dried under flowing N2 according to previously described procedures [11]. It should be noted that exposing surface-bound DNA hybrids to pure water, while necessary to prepare them for HRXPS and NEXAFS spectroscopy measurements, is known to affect their stability [11,26]. However, a recent systematic study on the subject has shown that when the samples are rinsed with small volumes of water (~0.5 mL), such as those used here, breakup of such hybrids is minimal [26].

High Resolution X-ray Photoelectron Spectroscopy

HRXPS measurements were performed at the HE-SGM beamline of the synchrotron storage ring BESSY II in Berlin, Germany. The measurements were conducted under UHV conditions at a base pressure better than 1.5 × 10–9 mbar and at room temperature. The spectral acquisition time was selected such that no noticeable damage by the primary X-rays occurred during the measurements [27-30]. The acquisition of the spectra was carried out in normal emission geometry with an energy resolution of ~0.4 eV. The energy scale was referenced to the Au 4f7/2 peak at 83.95 eV [31].

The spectra were fitted with symmetric Voigt functions and either a Shirley or linear background. The fits were carried out self-consistently, i.e. the same peak parameters were used for identical spectral regions. The film thickness was determined on the basis of the intensity ratios of the C 1s and Au 4f emissions [32], assuming a standard exponential attenuation of the photoelectron signal and using the attenuation lengths reported by Lamont and Wilkes [33].

The thickness was then used to calculate the density of the surface-adsorbed DNA molecules in the A5T15/Au film according to the procedures developed by Petrovykh et al. [13,14]. Due to minor carbon contamination in the A10T15/Au sample, the density values for this and the A15T15/Au films were calculated on the basis of the averaged intensity of the characteristic N 1s, P 2p, and O 1s signals and corrected for molecular composition. Density values of 1.40 × 1013, 1.02 × 1013, and 8.96 × 1012 molecules cm-2 were obtained using this method for the A5T15/Au, A10T15/Au, and A15T15/Au films, respectively. These values are comparable to those reported for similar films [24], although somewhat lower. This is most likely due to the fact that the C1s/Au4f intensity ratio was used to calculate thickness rather than the intensity of the Au 4f emission from sputter-cleaned gold [13,14].

Infrared reflection-adsorption spectroscopy

IRRAS was performed on a Vertex 80 Infrared spectrometer (Bruker Optics, Ettlingen, Germany) using an A518 IRRAS sampling stage with an 80° fixed-angle gracing incidence. The signal was recorded by a D313 narrow-band MCT detector cooled by liquid N2. A gold substrate covered with a perdeuterated dodecanethiol self-assembled monolayer was used as a reference. Background measurements were performed with 1024 scans, and suppression of the water signal in samples was usually achieved between 1000 and 3000 scans. Baseline correction and data evaluation were performed using the Opus 6.5 IR software package (Bruker Optics).

Near-edge X-ray absorption fine structure spectroscopy

NEXAFS and HRXPS spectra were collected at the same beamline. The acquisition of the NEXAFS spectra was performed at the C, N, and O K-edges in the partial electron yield (PEY) mode with retarding voltages of −150, –300, and −350 V, respectively. Linear polarized synchrotron light with a polarization factor of ~0.91 was used. The energy resolution of the whole setup was estimated to be on the order of ~0.3 eV, slightly dependent on the photon energy. To monitor the orientational order in the films, the incidence angle of the light was varied from 90° (E-vector in the surface plane) to 20° (E- vector near the surface normal) in steps of 10-20° (the angles are defined with respect to the surface plane) [34]. This approach is based on the dependence of the cross-section of the resonant photoexcitation process on the orientation of the electric field vector of the synchrotron light with respect to the molecular orbital of interest (so-called linear dichroism in X-ray absorption) [34]. This effect results in a characteristic dependence of an adsorption resonance intensity on the incidence angle of X-rays, as long as there is some orientational order present in the probed system. The photon energy (PE) scale was referenced to the pronounced π* resonance of highly oriented pyrolytic graphite at 285.38 eV [35].

Results and discussion

Unhybridized DNA brushes

Unhybridized films were examined to verify the quality of the deposited DNA layers and to gather an initial picture of the system before the formation of hybrids. Figure 1 (left) shows the N 1s HRXPS spectra of the initial unhybridized A5T15, A10T15, and A15T15 films. All three spectra have been decomposed into the components associated with thymine and adenine, with thymine appearing as a single peak with a binding energy of ~400.5 eV and adenine as a doublet with individual peaks at ~398.7 and ~400.5 eV and an intensity ratio of 2:1 [11,18,24]. The expected increase in the adenine content with increasing length of the d(A) block can be clearly seen, in agreement with previous reports on similar films [11,24]. The relative intensities of the thymine and adenine signals are presented in Figure 1 (right). These parameters exhibit the expected trend with the increasing length of the d(A) block but, after normalization to the number of nitrogen atoms present in thymine (2) and adenine (5), give a much higher proportion of thymine than can expected from the molecular composition alone. This seeming disagreement can be explained by the stronger attenuation of the photoelectron signal from the d(A) block serving as the molecular anchor and located, partly or almost completely, underneath the d(T) block. Note that the attenuation is especially strong at the kinetic energy of the N 1s photoelectrons (~180 eV; an attenuation length of 0.84 nm [33]), which explains why this effect was less pronounced in previous work relying on laboratory XPS systems [13,24,25] which have much higher excitation energy and, subsequently, higher photoelectron kinetic energies. Note also that the respective attenuation factor (~2.5) does not change significantly with the increasing length of the d(A) block, which contradicts the idea of the L-shape adsorption model and suggests that the exact adsorption geometry of the AnT15 molecules depends on the molecular composition. Nevertheless, the presence of such a strong attenuation supports the general view of the d(A) and d(T) blocks in the AnT15 films as binding and sensing units, respectively.

thumbnailFigure 1. Left: N 1s HRXPS spectra of unhybridized DNA brushes. The spectra are decomposed into the components assigned to thymine (blue) and adenine (red). Right: Relative intensities of the adenine (red) and thymine (blue) components.

In Figure 2, the IRRAS spectra of the AnT15 films in the region from 1850 to 1350 cm-1 are displayed. All three films show a strong peak at ~1710 cm-1. This peak has been attributed to stretching vibrations of the C=O bond in the thymine nucleotides which are not in direct contact with gold [15], and indicates that the d(T) portions of the DNA molecules were indeed extending into the ambient as predicted, rather than associating with the substrate. The spectra also display features typical of adenine nucleobases [36], with a peak at ~1650 cm-1, which has previously been assigned to NH2 bending modes, and another at ~1600 cm-1, assigned to NH2 bending modes or their combination with C=N stretching modes [25]. With the increasing length of the d(A) block, the corresponding adenine features increase in intensity relative to the thymine features. This is in agreement with the HRXPS results as well as with IRRAS results on similar samples presented in previous studies [24].

thumbnailFigure 2. IRRAS spectra of the unhybridized AnT15 films in the region between 1850 and 1350 cm-1. Dotted lines indicates the contributions from the thymine (T) and adenine (A) portions of the oligonucleotides. The signals from the A portion increase relative to the T portion with increasing length of the d(A) block.

NEXAFS spectroscopy provides information on the chemical composition of a film by sampling the electronic structure of the unoccupied molecular orbitals, as well as on the orientational order of the adsorbed molecules through examination of the angular dependence of the resonant excitations [34]. The chemical composition is usually probed by collecting spectra at the so-called magic angle of X-ray incidence (55˚). These spectra are not affected by orientational effects and are therefore solely representative of the film composition. Such spectra of the AnT15 films collected at the N K-edge are shown in left-hand panel of Figure 3. The spectra of all films exhibit a superposition of two characteristic resonances of thymine at 401.1 eV (corresponding to the N 1s →LUMO transition) and 402.0 eV (corresponding to the N1s →LUMO+1 transition), and two resonances from adenine at 399.4 eV (N 1s →LUMO transition) and 401.3 eV (N 1s →LUMO + 2 transition) [37]. The resonances of thymine are merged, with the one at 401.1 eV being more intense, while the resonances of adenine are well separated, with the 399.4 eV peak being much more intense than the one at 401.3 eV [37]. Accordingly, only a slightly asymmetric resonance of thymine at 401.1 eV (marked by blue) and a sharp resonance of adenine at 399.4 eV (marked by red) are clearly present in the spectra of the AnT15 films in Figure 3 (left). Since the overlap between these resonances is negligible, it is possible to identify unique adenine and thymine contributions in the spectra. As seen in Figure 3, the relative intensity of the adenine contribution increases with the increasing length of the d(A) block while that of thymine decreases, in agreement with the expected film composition and the HRXPS results.

thumbnailFigure 3. N K-edge NEXAFS spectra of the unhybridized AnT15 films. Left: Spectra acquired at an X-ray incidence angle of 55°. The positions of the most pronounced, π*-like absorption resonances of thymine (T) and adenine (A) are marked by the thin vertical blue (T) and red (A) dotted lines. Right: Difference of the spectra acquired at X-ray incidence angles of 90° and 20°. Adenine and thymine contributions are indicated by the red or blue fill (respectively), with an upward or downward orientation indicating a mostly horizontal or vertical orientation (respectively) of the nucleotides relative to the surface. Horizontal dashed lines correspond to zero for the individual difference spectra.

Beyond the chemical composition, information on average molecular orientation can be obtained from NEXAFS data by considering the angular dependence of the resonance intensity [34]. A suitable way to monitor such dependence is to plot the spectra acquired at the normal and gracing incidence angles of the X-ray beam. The orientational order present in the probed system will then manifest itself as difference peaks at the positions of the characteristic absorption resonances. The sign and intensity of these peaks will then provide information on the exact molecular geometry and degree of the orientational order. To this end, the difference spectra of the AnT15 films and derived brushes of DNA hybrids were considered in detail.

Generally, the nucleobases are assumed to be oriented perpendicularly to the ssDNA backbone, even though they possess some flexibility in orientation due to rotation relative to the backbone and from irregular bends in the backbone. But even taking into account this flexibility, a very strong positive peak would arise in the difference NEXAFS spectra from a film in which the DNA strands are parallel to the surface (nucleobases would then be predominantly oriented perpendicular to the substrate), and a very strong negative peak would be seen when all DNA molecules are oriented perpendicular to the surface (the nucleobases would then be predominantly oriented parallel to the substrate) [15,17,38]. In the case of ideal L-shape diblock-oligonucleotides [24,36], we would therefore expect to see a very strong positive peak from the parallel surface-binding adenine block and a very strong negative peak from the perpendicular sensing thymine block. In the case of deviations from the ideal parallel or perpendicular alignment, as presumably occurs in a real diblock-oligonucleotide system [15], we would expect to see less intense (but still clearly visible) positive adenine peaks and negative thymine peaks. In the ultimate case of partial or complete disordering of either the d(A) or d(T) blocks, no respective difference peaks would be perceptible.

The right-hand panel of Figure 3 shows the difference between the spectra of the AnT15 films collected at 90˚ and 20˚. The A5T15 film shows a strong positive adenine peak and a very weak negative thymine peak indicating, as expected, a predominantly parallel-to-the-substrate alignment of the adenine blocks of the DNA strands. Unexpectedly, however, this also indicates a near lack of any sort of orientational order in the thymine blocks. The most probable reason for this behavior is the interaction of the d(T) blocks with the surface-bound d(A) nucleotides [12], which can be explained by partial intramolecular and intermolecular hybridization. These hybrids could have formed either in solution prior to deposition, or after contact with the substrate. More sophisticated explanations are possible as well, including complex hybridization scenarios involving several molecules [26,39,40]. However, which of these configurations is dominant is not something that can be directly determined via NEXAFS spectroscopy but, presumably, requires its combination with theoretical calculation of most energetically favorable structural motifs.

The A10T15 film, in contrast, shows a predominantly perpendicular alignment of the thymine block as well as a predominantly parallel orientation of the adenine block. However, the intensities of the difference peaks are small, indicating that the extent of nucleobase alignment is small as well. As the A10T15 molecules have the potential to form twice the number of base pairs compared to the A5T15 molecules, it would be expected that any intermolecular hybrids formed in solution prior to deposition would be more stable. This may be one explanation for the results seen here, as a greater number of intermolecular hybrids would result in a smaller overall parallel ordering of the adenine nucleotides due to the presence of d(A) blocks with perpendicular orientation. In turn, the greater stability of these remaining inter- and intramolecular hybrids, compared to those of the A5T15 molecules, would result in a more stable perpendicular order in the thymine nucleotides [23]. However, the still relatively low maximum number of base pairs overall would likely mean that these types of intermolecular hybrids are not very stable [41,42] and thus unlikely to be the most common species in the A10T15/Au film. This would explain the continued dominance of the parallel orientation in the adenine nucleotides in this sample seen in the NEXAFS data. Note that the molecular orientations implied by the 90˚–20˚ difference spectra in Figure 3 for the A5T15/Au and A10T15/Au films are in full agreement with previous NEXAFS results [18].

The d(A) and d(T) blocks of the A15T15 molecules, in contrast to the A5T15 and A10T15 species, are fully complimentary in length, which should create the most stable inter- and intramolecular hybrids in solution. The most stable configuration of these molecules should be the full 30-base pair double helix formed from aligning two A15T15 strands. Upon exposure to the gold surface and subsequent rinsing and drying, the NEXAFS data reveal that both the adenine and thymine portions of these molecules are preferentially oriented perpendicular to the surface, suggesting that a large portion of the thymine blocks and at least a part of the adenine blocks are remaining upright. One molecular configuration which would be consistent with these data would be the attachment of the DNA molecules in complete hybrid form via only a portion of the d(A) block. Other arrangements could also exist, including an L-shaped configuration in which a second molecule is hybridized only with the d(T) portion, however the intensity of the adenine peak in the NEXAFS 90˚–20˚ data suggests that on average in this film more d(A) blocks are oriented upright than lying flat, in clear contrast to the A5T15/Au and A10T15/Au layers.

Summarizing, the HRXPS, IRRAS, and NEXAFS composition data give expected results for the unhybridized AnT15 films, confirming their quality and composition compared to previously tested films of this type [24]. The NEXAFS results presented here support the theory of complex hybridization interactions affecting these layers, as the simplest molecular arrangements necessary to produce the observed spectra become more complex from the A5T15/Au, to A10T15/Au, to the A15T15/Au film.

DNA hybrid brushes

Hybridization was performed using two approaches: first, by rinsing the initial unhybridized films with only pure water, and second, by rinsing the films with water, followed by a 0.1 M solution of NaOH, then water a final time. The latter approach was intended to further destabilize any inter- or intra-molecular hybrids in the probe film with the aim of improving both the number and quality of the hybrids formed upon exposure to the complimentary A15 target [11].

Figure 4 shows the HRXPS data for the resulting brushes of DNA hybrids, with the two left-hand panels displaying the N 1s spectra, and the vertically arranged right-hand panels presenting the relative intensities of the adenine and thymine signals for each particular spectrum. The intensity of the adenine signal is clearly higher in these spectra as compared to the unhybridized AnT15 films (Figure 1, right panel), as expected due to the presence of the A15 target. A comparison between the intensity values for the films rinsed with alternating H2O and NaOH to those rinsed with only H2O shows that the addition of NaOH increases the amount of adenine in all cases, also as predicted.

thumbnailFigure 4. Left two panels: N 1s HRXPS spectra of the AnT15 films hybridized after rinsing with H2O (left) or with the H2O-NaOH rinsing procedures (right). The adenine portion of the decomposed spectra (red fill) can be seen to increase relative to the thymine portion (blue fill) in the films rinsed with H2O and NaOH relative to those rinsed with only H2O. Rightmost panels: Relative percentages of nitrogen signal stemming from the adenine (red) and thymine (blue) components of the films rinsed with H2O (top panel) and those rinsed with H2O and NaOH (bottom panel).

Beyond this qualitative conclusion, it is possible to perform a numerical analysis of the intensity values in Figure 4 to get at least a coarse estimate of the degree of hybridization. Normalizing the intensities to the number of nitrogen atoms in adenine (5) and thymine (2) and assuming the same attenuation of the photoemission signal for the d(A) anchor of the AnT15 probe species (a factor of 2.5; see above), we obtained the following values for the extent of hybridization in the H2O-rinsed AnT15/Au:A15 brushes: 24%, 46%, and 87% for n = 5, 10, and 15, respectively. For the films rinsed with alternating H2O and NaOH, these values were calculated to be 39%, 57%, and ~100% for n = 5, 10, and 15, respectively. Note that these values can be considered as coarse estimates only, since the attenuation factor for the photoemission signal from the d(A) anchor of the AnT15 probe can change after the hybridization event. It is most likely that this factor would decrease as the inter- and intramolecular (probe) hybrids are disrupted and upright target-probe hybrids are formed, resulting in a partial uncovering of the d(A) anchors. Thus, the values of the hybridization efficiency calculated here are presumably the upper limits of the real values. Note also that these values represent coarse estimates only since they are based on a single sample for each particular system. However, qualitatively similar differences between the outcomes of the H2O and H2O-NaOH rinsing procedures were observed for all AnT15 samples, which strengthens the reliability of the conclusions.

Apart from this uncertainty, two tendencies are apparent. First, the H2O-NaOH rinsing procedure results in a greater number of hybrids than the H2O rinse alone. Second, the number of hybrids in the AnT15 films increases with increasing length of the d(A) block. This is expected, as longer d(A) blocks increase the potential for a larger separation between the sensing d(T) blocks and thus greater accessibility for the target strands.

In the case of the A15T15, we expect that in its as-deposited state this film has a rather high degree of intramolecular hybridization. However, any molecules that remain after the post-deposition 1 min rinsing with pure water would have to be either at least partially attached to the surface via the d(A) block or hybridized to a surface-bound molecule. Previous work examining the stability of surface-bound DNA hybrids has shown them to be quite unstable and readily broken apart and replaced by a fully complimentary, non-surface-bound target molecule [12], although it should be noted that those hybrids were not formed first in solution. In diblock-oligonucleotide system used here, the lowest energy state will be with the d(A) portion of the molecule fully adsorbed to the gold substrate and the d(T) portion fully hybridized. Thus, it is expected that the introduction of the A15 targets would break apart most of the inter or intra-molecular hybrids in the A15T15 layer.

The change in adenine content made clear by HRXPS was also observed with IRRAS. The spectra from A15T15/Au before and after hybridization are shown in Figure 5, in which the intensity of the adenine-specific peak at 1650 cm-1 increases relative to the thymine-specific peak at 1707 cm-1 from the unhybridized films to those rinsed with H2O prior to hybridization. This peak increases further in the film treated with the H2O-NaOH rinsing procedure.

thumbnailFigure 5. IRRAS spectra of the A15T15 film in an unhybridized state (lower spectrum), after rinsing with only H2O and hybridizing (middle spectrum), and after rinsing with H2O-NaOH and hybridizing (upper spectrum). The adenine (A) portion can be seen to increase relative to the thymine (T) portion from the unhybridized to the H2O-rinsed, and from the H2O-rinsed to the H2O and NaOH-rinsed film.

The NEXAFS data for the hybrid brushes are presented in Figure 6. The N K-edge spectra collected at 55˚ (left panels) showed an increase in the intensity of the adenine-assigned resonance at 399.4 eV relative to the thymine contribution at 401.6 eV after hybridization compared to the analogous spectra for the unhybridized films (Figure 3, left), supporting the results of the other techniques. Furthermore, the spectra exhibit an increase in the intensity of the adenine-assigned resonance relative to the thymine contribution for all films rinsed with alternating H2O and NaOH over those rinsed with just H2O prior to hybridization, an effect which is especially pronounced for A5T15/Au. This indicates a higher adenine content in the NaOH-treated layers, in agreement with both the HRXPS (Figure 4) and IRRAS (Figure 5) results. The information gathered from the 90˚–20˚ difference spectra reveals significant changes in orientation in these films relative to their unhybridized counterparts, especially in the case of A5T15/Au (Figure 3). The A5T15/Au sample, which unhybridized had shown a positive adenine difference peak and nearly no thymine difference peak, now clearly displays negative peaks for both nucleobase types. This indicates that the predominant orientation of both the d(A) and d(T) blocks present in this film is perpendicular to the surface. This agrees qualitatively with the idealized model (Scheme 1), with the A5 block of the A5T15 probes adhering parallel to the substrate and the T15 block protruding from the surface and hybridizing with the A15 target. In spite of the limited extent of hybridization (see above), the spectral contribution of the A15 targets is clearly higher than that of the A5 anchors. This is related to the greater number of adenine nucleotides in the A15 targets versus the A5T15 molecules (as far as one compares the individual species) as well as to a stronger attenuation of the PEY signal from the A5 anchors as compared to the perpendicularly-oriented A15 probes (this effect is smaller as compared to the photoelectrons but still applicable [43]). Interestingly, the difference peaks increase in intensity for the film rinsed with the H2O-NaOH rinsing procedure, indicating a slightly improved orientation of the hybrids in this sample. This means that a rinse including NaOH is more efficient for the disruption of inter- or intramolecular hybrids in the probe film before the hybridization, making the T15 strands of assembled A5T15 more easily accessible to the target molecules.

thumbnailFigure 6. N K-edge NEXAFS spectra of the AnT15 brushes after rinsing with only H2O (upper panels) or with H2O-NaOH (lower panels) followed by hybridization with an A15 target. Left panels: Spectra acquired at an X-ray incidence angle of 55˚. The positions of the most pronounced, π*-like absorption resonances of thymine (T) and adenine (A) are marked by the thin vertical blue (T) or red (A) dotted lines. Right panels: Difference of the spectra acquired at X-ray incidence angles of 90° and 20°. Adenine and thymine contributions are indicated by the red or blue fill (respectively), with an upward or downward direction indicating a preferentially horizontal or vertical orientation (respectively) of the DNA strand relative to the surface. Horizontal dashed lines correspond to zero for the individual difference spectra.

Hybridization of the A10T15/Au film also causes a previously positive adenine-assigned difference peak (Figure 3) to become negative, indicating a change in the orientation of the majority of the adenine strands from parallel to perpendicular to the surface after hybridization. This behavior also agrees qualitatively with the idealized model of the A10T15/Au:A15 hybrids (Scheme 1), assuming that the ideal L-shape is significantly distorted. Also in this case, the spectral contribution of the perpendicularly-oriented A15 probes is clearly higher than that of the parallel A10 anchors. This is similar to what was observed for the A5T15/Au:A15 film, and can likely be explained by the same reasoning. However, in contrast to A5T15/Au:A15, there is no significant change in either the orientation or intensity of these peaks between the H2O-rinsed samples and those subjected to the H2O-NaOH rinsing procedure. As previously mentioned, this may be due to the relative instability of hybrids with few base pairs [26,40]. The few intermolecular hybrids that were present in the film after rinsing but before hybridization may have been unstable enough to be replaced by a more energetically stable A10T15:A15 hybrid, regardless of whether the brush was exposed to NaOH or only H2O.

Both of the hybridized A15T15/Au samples show negative adenine- and thymine-assigned difference peaks, as was the case in the unhybridized layer. However, the dichroism becomes more pronounced after hybridization, which is especially obvious for the adenine-assigned difference peak. This increase in the dichroism can be associated with the contribution of the perpendicularly-oriented A15 targets, which have a particularly high spectral weight due to the high degree of hybridization (see above). The ideal L-shape model (Scheme 1) is likely also applicable in this case, at least qualitatively (under the assumption of distortion), given the high degree of hybridization. However, the comparably large spacing between the individual hybrids should diminish the orientational order, as usually occurs for loosely packed molecular films. This effect should be less pronounced in A10T15/Au:A15 and least pronounced in A5T15/Au:A15. Indeed, the latter film exhibits the highest extent of dichroism as compared to the two others, especially in the case of the film rinsed with alternating H2O and NaOH.

Returning to A15T15/Au:A15 and comparing the difference spectra for the cases of H2O and H2O-NaOH rinsing procedures, we can state that the addition of NaOH only slightly improves the orientational order in the hybrid films. Presumably, as mentioned above, the orientational order is limited by the low packing density and the improvement is solely related to the higher extent of hybridization in the case of the film treated with the H2O-NaOH rinsing procedure.

Conclusions

We have examined a series of adenine/thymine diblock-oligonucleotide films and their hybrid brushes using a combination of HRXPS, IRRAS, and NEXAFS spectroscopy to monitor sample composition as well as molecular orientation and ordering. As probe films, we used adenine-thymine diblock oligonucleotides composed of either 5, 10, or 15 adenine nucleotides [d(A)] connected to 15 thymine nucleotides [d(T)]. The quality and composition of these films was confirmed by the HRXPS, IRRAS, and NEXAFS composition data, while the binding function of the adenine block was confirmed by the HRXPS data. It was found that the orientational order in these films differed significantly from the ideal L-shape model, in which the binding d(A) block is oriented strictly parallel to the substrate while the d(T) portion of the molecule extends away and can be used to form hybrids with a complimentary target. In reality, the situation appeared to be more complex, a fact which was explained by inter- and intramolecular hybridization within the monomolecular assemblies. Deviations from the ideal model were observed in all probe films of the present study but became more pronounced with increasing length of the d(A) binding block and subsequently increased potential for stable inter- and intramolecular hybridization.

In spite of the complex molecular arrangements, all probe films performed well upon hybridization with the A15 target, exhibiting the expected correlation between the lateral density (i.e. accessibility) of the probe T15 strands and the extent of hybridization. Whereas this extent was relatively low for the most densely packed A5T15 films, it was found to increase the A10T15 samples, and was highest in the most loosely packed A15T15 monolayers. Furthermore, it was observed that the probe films rinsed with alternating H2O and NaOH exhibited higher hybridization efficiency than the films rinsed with H2O only. This confirms the ability of NaOH to reduce the number of surface-bound inter- and intramolecular hybrids, favoring more efficient hybridization.

The NEXAFS data for the hybrid AnT15/Au:A15 brushes do not contradict the ideal L-shape model but do not unequivocally support it. One can only say for sure that the T15:A15 hybrids have the expected predominantly upright orientation, with the highest orientational order present in the most densely packed A5T15/Au:A15 brush. It is, however, difficult if not impossible to monitor the orientation of the binding An blocks since their contributions to the spectra are masked by those of the of the A15 targets. However, a parallel-to-the-substrate orientation of the binding blocks can be assumed to be most likely in view of the extensive disruption of the inter- and intramolecular hybrids in the probe film upon rinsing and subsequent hybridization.

These results are a good illustration of the strength of spectroscopic analysis in its specific application to biological systems, demonstrating that the use of different techniques to observe multiple facets of the system can often reveal unexpected effects. In some cases, such effects may prove to be key factors in understanding biological properties and processes on surfaces and interfaces at the molecular level.

Competing interests

The authors declare that they have no competing interests.

Authors’ contributions

CH and MZ conceived of and designed the study, CH fabricated the samples and performed the IRRAS measurements and analysis, MZ and HH carried out the NEXAFS and HRXPS measurements and analysis. CH and MZ wrote the manuscript. All authors read and approved the final manuscript

Acknowledgement

We thank Prof. M. Grunze and Dr. P. Koelsch for support of this work, A. Nefedov and C. Wöll for the technical cooperation at BESSY II, the BESSY II staff for technical assistance during the experiments at the synchrotron, Dr. M. Bruns and V. Oberst for gold wafer fabrication, and S. Heissler for IR spectroscopy assistance. C.H. acknowledges support from a U.S. National Science Foundation Graduate Research Fellowship Award No. 0840595. This work has been supported by Deutsche Forschungsgemeinschaft (DFG) (Grants ZH 63/9-3 and ZH 63/14-1) and the Biointerfaces Programme.

References

  1. Castner DG, Ratner BD (2002) Biomedical surface science: Foundations to frontiers. Surf Sci. 500:28–60 Publisher Full Text OpenURL

  2. Höök F, Kasemo B, Grunze M, Zauscher S (2008) Quantitative biological surface science: Challenges and recent advances. ACS Nano 12:2428–2436 OpenURL

  3. Kasemo B (2002) Biological surface science. Surf Sci. 500:656–667 Publisher Full Text OpenURL

  4. Cosnier S, Mailley P (2008) Recent advances in DNA sensors. Analyst. 133:984–991 PubMed Abstract | Publisher Full Text OpenURL

  5. Hsiao SC, Shum BJ, Onoe H, Douglas ES, Gartner ZJ, Mathies RA, Bertozzi CR, Francis MB (2009) Direct cell surface modification with DNA for the capture of primary cells and the investigation of myotube formation on defined patterns. Langmuir. 25:6985–6991 PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  6. Mirkin CA, Letsinger RL, Mucic RC, Storhoff JJ (1996) A DNA-based method for rationally assembling nanoparticles into macroscopic materials. Nature. 382:607–609 PubMed Abstract | Publisher Full Text OpenURL

  7. Storhoff JJ, Mirkin CA (1999) What controls the optical properties of DNA-linked gold nanoparticle assemblies? Chem Rev. 99:1849–1862 PubMed Abstract | Publisher Full Text OpenURL

  8. Adleman LM (1994) Molecular computation of solutions to combinatorial problems. Science. 266:1021–1024 PubMed Abstract | Publisher Full Text OpenURL

  9. Liu Q, Wang L, Frutos AG, Condon AE, Corn RM, Smith LM (2000) DNA computing on surfaces. Nature. 403:175–179 PubMed Abstract | Publisher Full Text OpenURL

  10. Bally M, Halter M, Vörös J, Grandin HM (2006) Optical microarray biosensing techniques. Surf Int Anal. 38:1442–1458 Publisher Full Text OpenURL

  11. Schreiner SM, Shudy DF, Hatch AL, Opdahl A, Whitman LJ, Petrovykh DY (2010) Controlled and Efficient Hybridization Achieved with DNA Probes Immobilized Solely through Preferential DNA-Substrate Interactions. Anal Chem 2010(82):2803–2810 OpenURL

  12. Schreiner S, Hatch A, Shudy D, Howell C, Zhao J, Koelsch P, Zharnikov M, Petrovykh D, Opdahl A (2011) Impact of DNA–Surface Interactions on the Stability of DNA Hybrids. Anal Chem. 83:4288–4295 PubMed Abstract | Publisher Full Text OpenURL

  13. Petrovykh DY, Kimura-Suda H, Whitman LJ, Tarlov MJ (2003) Quantitative analysis and characterization of DNA immobilized on gold. J Am Chem Soc. 125:5219–5226 PubMed Abstract | Publisher Full Text OpenURL

  14. Petrovykh DY, Kimura-Suda H, Tarlov MJ, Whitman LJ (2004) Quantitative characterization of DNA films by X-ray photoelectron spectroscopy. Langmuir. 20:429–440 PubMed Abstract | Publisher Full Text OpenURL

  15. Petrovykh DY, Perez-Dieste V, Opdahl A, Kimura-Suda H, Sullivan JM, Tarlov MJ, Himpsel FJ, Whitman L (2006) Nucleobase orientation and ordering in films of single-stranded DNA on gold. J Am Chem Soc. 128:2–3 PubMed Abstract | Publisher Full Text OpenURL

  16. Lee CY, Gong P, Harbers GM, Grainger DW, Castner DG, Gamble LJ (2006) Surface coverage and structure of mixed DNA/alkylthiol monolayers on gold: characterization by XPS, NEXAFS, and fluorescence intensity measurements. Anal Chem. 78:3316–3325 PubMed Abstract | Publisher Full Text OpenURL

  17. Ballav N, Koelsch P, Zharnikov M (2009) Orientation and Ordering in Monomolecular Films of Sulfur-Modified Homo-oligonucleotides on Gold. J Phys Chem C. 113:18312–18320 Publisher Full Text OpenURL

  18. Howell C, Zhao J, Koelsch P, Zharnikov M (2011) Hybridization in ssDNA films—a multi-technique spectroscopy study. Phys Chem Chem Phys. 13:15512–15522 PubMed Abstract | Publisher Full Text OpenURL

  19. Lee CY, Canavan HE, Gamble LJ, Castner DG (2005) Evidence of impurities in thiolated single-stranded DNA oligomers and their effect on DNA self-assembly on gold. Langmuir. 21:5134–5141 PubMed Abstract | Publisher Full Text OpenURL

  20. Lee C-Y, Gamble LJ, Grainger DW, Castner DG (2006) Mixed DNA/Oligoethylene glycol functionalized gold surfaces improve DNA hybridization in complex media. Biointerphases 1:83–92 OpenURL

  21. Lee CY, Nguyen PCT, Grainger DW, Gamble LJ, Castner DG (2007) Structure and DNA hybridization properties of mixed nucleic acid/maleimide-ethylene glycol monolayers. Anal Chem. 79:4390–4400 PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  22. Stokes GY, Gibbs-Davis JM, Boman FC, Stepp BR, Condie AG, Nguyen ST, Geiger FM (2007) Making “Sense” of DNA. J Am Chem Soc. 129:7492–7493 PubMed Abstract | Publisher Full Text OpenURL

  23. Howell C, Jeyachandran YL, Koelsch P, Zharnikov M (2012) Orientation and Ordering in Sequence-and Length-Mismatched Surface-Bound DNA Hybrids. J Phys Chem C. 116:11133–11140 Publisher Full Text OpenURL

  24. Opdahl A, Petrovykh DY, Kimura-Suda H, Tarlov MJ, Whitman L (2007) Controlled and Efficient Hybridization Achieved with DNA Probes Immobilized Solely through Preferential DNA-Substrate Interactions. J Proc Natl Acad Sci USA. 104:9–14 Publisher Full Text OpenURL

  25. Saprigin AV, Thomas CW, Dulcey CS, Patterson CH, Spector MS (2005) Spectroscopic quantification of covalently immobilized oligonucleotides. Surf Int Anal. 37:24–32 Publisher Full Text OpenURL

  26. Howell C, Hamoudi H, Koelsch P, Zharnikov M (2011) Orientation changes in surface-bound hybridized DNA undergoing preparation for ex situ spectroscopic measurements. Chem Phys Lett. 513:267–270 Publisher Full Text OpenURL

  27. Zharnikov M, Grunze M (2002) Modification of thiol-derived self-assembling monolayers by electron and x-ray irradiation: Scientific and lithographic aspects. J Vac Sci Technol B. 20:1793–1807 Publisher Full Text OpenURL

  28. Heister K, Zharnikov M, Grunze M, Johansson LSO, Ulman A (2001) Characterization of X-ray induced damage in alkanethiolate monolayers by high-resolution photoelectron spectroscopy. Langmuir. 17:8–11 Publisher Full Text OpenURL

  29. Wirde M, Gelius U, Dunbar T, Allara DL (1997) Modification of self-assembled monolayers of alkanethiols on gold by ionizing radiation. Nucl Instrum Methods Phys Res B. 131:245–251 Publisher Full Text OpenURL

  30. Jäger B, Schürmann H, Müller HU, Himmel HJ, Neumann M, Grunze M, Wöll C (1997) X-ray and low energy electron induced damage in alkanethiolate monolayers on Au-substrates. Z Phys Chem. 202:263–272 Publisher Full Text OpenURL

  31. Surface chemical analysis – X-ray photoelectron spectrometers – Calibration of the energy scales. ISO p 15472 PubMed Abstract | Publisher Full Text OpenURL

  32. Thome J, Himmelhaus M, Zharnikov M, Grunze M (1998) Increased lateral density in alkanethiolate films on gold by mercury adsorption. Langmuir. 14:7435–7449 Publisher Full Text OpenURL

  33. Lamont CLA, Wilkes J (1999) Attenuation length of electrons in self-assembled monolayers of n-alkanethiols on gold. Langmuir. 15:2037–2042 Publisher Full Text OpenURL

  34. Stöhr J (1992) NEXAFS Spectroscopy, Springer Series in Surface Science 25 Berlin: Springer OpenURL

  35. Batson PE (1993) Carbon 1s near-edge-absorption fine structure in graphite. Phys Rev B: Condens Matter. 48:2608–2610 Publisher Full Text OpenURL

  36. Kimura-Suda H, Petrovykh DY, Tarlov MJ, Whitman LJ (2003) Base-dependent competitive adsorption of single-stranded DNA on gold. J Am Chem Soc. 125:9014–9015 PubMed Abstract | Publisher Full Text OpenURL

  37. Zubavichus Y, Shaporenko A, Korolkov V, Grunze M, Zharnikov M (2008) X-ray Absorption Spectroscopy of the Nucleotide Bases at the Carbon, Nitrogen, and Oxygen K-Edges. J Phys Chem B. 112:13711–13716 PubMed Abstract | Publisher Full Text OpenURL

  38. Samuel NT, Lee CY, Gamble LJ, Fischer DA, Castner DG (2006) NEXAFS characterization of DNA components and molecular-orientation of surface-bound DNA oligomers. J Electron Spectrosc Relat Phenom. 152:134–142 Publisher Full Text OpenURL

  39. Peterson AW, Wolf LK, Georgiadis RM (2002) Hybridization of mismatched or partially matched DNA at surfaces. J Am Chem Soc. 124:14601–14607 PubMed Abstract | Publisher Full Text OpenURL

  40. Hagan MF, Chakraborty AK (2004) Hybridization dynamics of surface immobilized DNA. J Chem Phys 10:4958–4968 OpenURL

  41. Shamsi MH, Kraatz HB (2011) The effects of oligonucleotide overhangs on the surface hybridization in DNA films: an impedance study. Analyst. 136:3107–3112 PubMed Abstract | Publisher Full Text OpenURL

  42. Lane MJ, Paner T, Kashin I, Faldasz BD, Li B, Gallo FJ, Benight AS (1997) The thermodynamic advantage of DNA oligonucleotide 'stacking hybridization'reactions: energetics of a DNA nick. Nucleic Acids Res. 25:611–616 PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  43. Zharnikov M, Frey S, Heister K, Grunze M (2002) An extension of the mean free path approach to X-ray absorption spectroscopy. J Electron Spectrosc Relat Phenom. 124:15–24 Publisher Full Text OpenURL