Molecular Beacon Modified Sensor Chips for Oligonucleotide

Nov 2, 2014 - Citation data is made available by participants in Crossref's Cited-by Linking service. For a more comprehensive list of citations to th...
0 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/Langmuir

Molecular Beacon Modified Sensor Chips for Oligonucleotide Detection with Optical Readout Qiang Su,† Daniel Wesner,‡ Holger Schönherr,‡ and Gilbert Nöll*,† †

Nöll Junior Research Group, Organic Chemistry, Department of Chemistry and Biology, Faculty IV, Siegen University, Adolf-Reichwein-Strasse 2, 57068 Siegen, Germany ‡ Physical Chemistry I, Department of Chemistry and Biology, Faculty IV, Siegen University, Adolf-Reichwein-Strasse 2, 57076 Siegen, Germany S Supporting Information *

ABSTRACT: Three different surface bound molecular beacons (MBs) were investigated using surface plasmon fluorescence spectroscopy (SPFS) as an optical readout technique. While MB1 and MB2, both consisting of 36 bases, differed only in the length of the linker for surface attachment, the significantly longer MB3, consisting of 56 bases, comprised an entirely different sequence. For sensor chip preparation, the MBs were chemisorbed on gold via thiol anchors together with different thiol spacers. The influence of important parameters, such as the length of the MBs, the length of the linker between the MBs and the gold surface, the length and nature of the thiol spacers, and the ratio between the MBs and the thiol spacers was studied. After hybridization with the target, the fluorophore of the longer MB3 was oriented close to the surface, and the shorter MBs were standing more or less upright, leading to a larger increase in fluorescence intensity. Fluorescence microscopy revealed a homogeneous distribution of the MBs on the surface. The sensor chips could be used for simple and fast detection of target molecules with a limit of detection in the larger picomolar range. The response time was between 5 and 20 min. Furthermore, it was possible to distinguish between fully complementary and singly mismatched targets. While rinsing with buffer solution after hybridization with target did not result in any signal decrease, complete dehybridization could be carried out by intense rinsing with pure water. The MB modified sensor chips could be prepared in a repeatable manner and reused many times without significant decrease in performance.

■

INTRODUCTION For health-care applications, there is a growing demand for analytical techniques that allow the simple and fast detection of specific oligonucleotide sequences (DNA or RNA). DNA detection is an essential element of genetic screening,1 medical diagnosis,2,3 forensic analysis,4,5 and single-nucleotide polymorphism (SNP) profiling.6,7 Recently, there has also been an increasing interest in the detection of (relatively short) microRNA (mi-RNA) sequences in medical diagnosis. The detection of strongly enhanced or decreased levels of characteristic miRNA (comprising 18−24 bases) can be used for the early stage diagnosis of severe diseases, such as cardiac diseases or different types of cancer.8,9 Among different analytical approaches for label-free detection of oligonucleotides in solution, the use of DNA hairpins as molecular beacons (MBs) has proven to be a useful method.10−13 Common MBs are DNA hairpin structures comprising a single-stranded loop with a sequence complementary to the target as well as a double-stranded stem region equipped with a fluorophore and a quencher in close proximity, causing the fluorescence to be quenched by energy transfer. Upon hybridization with target, the stem opens, and fluorophore and quencher become separated. The light (fluorescence) is switched on. MBs have been found to exhibit © 2014 American Chemical Society

extraordinary stability, better selectivity, and higher specificity than similar assays using linear single-stranded DNA.11,14,15 Besides their use in solution, MBs have been immobilized on different types of solid substrates.16−19 While in most studies, the applied substrates were used only as an anchor for the entire MBs comprising fluorophore and quencher, in some studies the substrate itself has been employed as the quenching unit.20−22 As an alternative to optical detection (detection of the fluorescence increase upon hybridization), the MB concept has been adopted for electrochemical sensors.23−26 For this purpose, the fluorophore has been replaced by a redox probe. After hybridization with target DNA, the distance between redox probe and gold electrode (i.e. the solid support) will change. Depending on the design of the MBs including the number of stem-loop structures, the distance between redox probe and Au electrode may increase or decrease upon hybridization, leading to a decrease or increase in the current for reduction and (re)oxidation of the redox probe.24,27,28 Recently, we have introduced surface plasmon fluorescence spectroscopy (SPFS) as a new optical readout method for Received: October 25, 2014 Published: November 2, 2014 14360

dx.doi.org/10.1021/la504105x | Langmuir 2014, 30, 14360−14367

Langmuir

Article

(serving as spacer between MB and the surface) followed by three dithiane rings (for chemisorption at the gold surface) were attached. MB2 comprised the same sequence as MB1, but the two thymidines next to the dithiane rings were left out in order to decrease the distance between fluorophore and gold surface and thereby increase the fluorescence quenching efficiency for the MB in the closed state. The synthesis of the longer MB3 started from a thymidine at the controlled pore glass support (3′ end). After three dithiane rings were introduced, the beacon with a total length of 56 bases (9 base pairs stem and 38 bases loop) plus a Cy5 dye at the 5′ end was attached. Complementary target oligonucleotides were purchased from Eurogentec or Metabion GmbH, (Martinsried, Germany). For MB1 and MB2, apart from the fully complementary target, targets also bearing a single mismatch at two different positions were investigated. The sequences of the MBs and target oligonucleotides are shown in the Supporting Information (SI) Table S1. Further experimental details about materials, instrumentation, the preparation of the gold substrates, the sensor chip preparation, the DNA hybridization and dehybridization procedure, and the confocal fluorescence microscopy experiments are given in the SI. Surface Coverage of the MB Capture Probe Layer. From the shift in the SPR angular scan curves collected before and after adsorption of MB and short thiols, the optical thickness dA·nA of the MB/thiol layer (and thus a value for dA, if a value for nA is assumed) can be calculated. For this purpose, the SPR software Winspall (version 3.0.2.0, Max Planck Institute of Polymer Research, Mainz, Germany), which is based on the Fresnel equations, was used assuming a value of nA = 1.5 for the refractive index of DNA.29 With knowledge of dA and nA, the mass of the film (in ng·mm−2) can be calculated by Feijter’s eq 1, if values for the refractive index increment dn/dc of the adsorbed layer and the refractive index of the buffer nsol are known as well.30 n − nsol M = dA A (1) dn/dc

surface bound MBs, which use the gold surface as the quenching unit.29 The working principle of this MB-SPFS sensor is shown in Scheme 1. Rather close to the surface the Scheme 1. Working Principle of the MB-SPFS Sensora

a

The MB has been adsorbed at a gold coated glass slide, together with short thiols serving as spacer (for clarity, the short thiols are not shown). A prism is used to excite surface plasmons with p-polarized laser light (λ = 633 nm). This results in the formation of an enhanced evanescent field as shown on the left. The intensity of this evanescent field decays exponentially with distance from the surface. In the closed state, the MB fluorophore is localized in close vicinity to the surface, and the fluorescence is quenched by energy transfer to the gold surface. After hybridization, the MB hairpin structure opens and the fluorophore becomes separated from the surface. The fluorescence is switched on.

fluorescence is quenched by energy transfer to the gold surface (the MB is in the closed state). Upon hybridization with the target, the distance between the gold surface and the fluorophore increases, and the fluorophore can be excited by the enhanced evanescent field caused by the surface plasmons. However, the intensity of the evanescent field decays exponentially with distance. Hence, there is an optimum distance for fluorescence detection for the MB in the open state, at which strong excitation without significant quenching can be achieved. For that reason, the length of the MB in the open state has to be optimized. In a first study, we have developed an MB-SPFS sensor for the detection of target oligonucleotides with a total length of 24 bases.29 The performance of this sensor in terms of sensitivity, selectivity (i.e., single mismatch discrimination ratio), and reusability was in the same range as for an electrochemical DNA pseudoknot sensor comprising two stem-loop structures, which was designed for the detection of targets with a length of only 17 bases.28 In the current study, we present our efforts to improve the performance of the MB-SPFS sensor by varying important experimental parameters, such as the type and physical properties of short thiol molecules, which are coadsorbed with the MB as spacer molecules, the length of the MB loop, the length of the linker between the MB stem, and the disulfide anchor for surface attachment, the local density of MB, as well as the type and position of the single mismatch in hybridization studies (for comparison with fully complementary target).

■

For tris buffer, a refractive index of nsol = 1.35 was determined experimentally by fitting the angular scan curve measured for the bare gold layer in buffer solution. For DNA, a dn/dc value of 0.175 cm3·g−1 has been reported.31−33 From the fitted average thickness (1.0 nm) a surface coverage of 6.5 × 10−12 mol·cm−2 (85.7 ng·cm−2) was calculated for MB1 on the gold surface, when a mixture of MCB and MPA (1:1) was used as thiol spacers. According to average thicknesses measured for the other MBs, surface coverages of 7.5 × 10−12 mol· cm−2 (94.3 ng·cm−2) and 6.9 × 10−12 mol·cm−2 (128.6 ng·cm−2) were calculated for MB2 and MB3, respectively, when a mixture of MCB and MPA (1:1) was used as thiol spacers. During the calculation, the contribution of the thiol spacer molecules is negligible, because the molecular weight of MBs (MB1:13123.4 g/mol; MB2:12515.0 g/mol; MB3:18629.6 g/mol) is much higher than that of the short thiols (e.g., MCB: 106.19 g/mol).

■

RESULTS AND DISCUSSION Preparation and Performance of the MB-SPFS Sensors. The surface plasmon resonance (SPR) and the SPFS kinetic scan curves collected at a fixed angle of 67.1° during the preparation of an MB-SPFS sensor surface are shown in Figure 1A for MB1, in combination with a mixture of mercaptobutanol and mercaptopropionic acid (MCB/MPA) as spacers. Directly after the addition of the MB-containing solution, the fluorescence strongly increases. This increase is followed by a minor decrease during the formation of the MBmonolayer (probably because the fluorescence of the MBs, which are adsorbed at the surface, is going to be quenched) before a stable fluorescence background is reached. After additional incubation of a mixture of MCB and MPA in water, the fluorescence background signal is decreased. In Figure 1B, the SPR and SPFS angular scan curves (showing the angledependent reflectivity in % and the fluorescence intensity in

EXPERIMENTAL SECTION

MBs and Targets. All MBs used in the current study were purchased from Eurogentec (Seraing, Belgium). The sequence of MB1 was the same as that investigated previously.22,29 The stem and the loop of MB1 comprised 9 base pairs and 18 bases, respectively. At the 3′ end, a Cy5 dye was attached, while at the 5′ end, two thymidines 14361

dx.doi.org/10.1021/la504105x | Langmuir 2014, 30, 14360−14367

Langmuir

Article

corresponding SPFS angular scan curve was collected), the SPFS signal did not change significantly. When the SPFS signal collected during rinsing with buffer was fitted to a monoexponential decay, a very small time constant was obtained (for details, see SI Figure S1). The same observation has been made also during the investigation of the other MBs in this study. Comparison of Individual MBs in Combination with Different Spacer Molecules. In order to compare the individual MBs, MB1, MB2, and MB3 were investigated before and after hybridization with 200 μL of a solution containing the fully complementary target at a concentration of 1 μM (see Figure 2A−C). For all three MBs, different thiol molecules serving as spacer between individual MB molecules were evaluated. The thiols mercaptohexanol (MCH), a mixture of mercaptobutanol and mercaptopropionic acid (MCB/MPA), mercaptoundecyl dihydrogen phosphate (MDPA), and mercaptopropanesulfonic acid sodium salt (MPS) were used in the present study. All thiols were coadsorbed for 2 h with the individual MBs (2 μM) in a total thiol concentration of 10 μM. After rinsing with water, a solution of the thiols at a total concentration of 1 mM in pure water was incubated for 1 h to remove any nonspecifically adsorbed hairpin probes from gold surface. Thereafter, the MB modified sensor chips equilibrated with buffer solution overnight prior to use. During SPFS measurements, two opposite effects have to be considered. While the intensity of the surface plasmon enhanced optical field with maximum intensity at the surface decays exponentially with distance, close to the surface, the fluorescence will be quenched. An optimum distance for maximum fluorescence intensity has been estimated for a distance of about 20 nm orthogonal to the surface.34 With respect to this estimation, the highest fluorescence intensity would be expected for MB3 in the current study if all three MBs were oriented orthogonal to the surface. As can be seen from Figure 2, the best performance was obtained for MB1 and MB2 (Figure 2A,B), both designed for the detection of the same target with a length of 24 bases, whereas the increase in the fluorescence signal for the longer MB3 designed for the detection of a target with 39 complementary bases was much lower. This was a general trend independent of the different type of thiol spacers. One explanation for this observation could be that all three MBs are oriented on average more or less upright with respect to the surface and the fluorophore of the longer MB3 comprising 56 bases is localized further from the surface at a distance at which the evanescent field is less intense. This would suppose that the fluorescence of MB1 and MB2 with a total length of 38 and 36 bases is not significantly quenched in the open state, even though MB1 and MB2 are rather short (the length of fully hybridized dsDNA of 56 base pairs is about 20 nm and that for dsDNA of 38 base pairs is about 14 nm, respectively). However, this explanation would be in opposition to the assumption that an optimum distance for maximum fluorescence intensity is reached at a distance of about 20 nm orthogonal to the surface.34 A different more reasonable explanation is that the fluorescence of MB3 is partially quenched by energy transfer to the gold surface because its fluorophore (all three MBs comprise Cy5 as fluorophore) is on average oriented closer to the surface than those of MB1 and MB2. Since MB3 is much longer than MB1 and MB2, this would imply that MB3 is either lying partially at the surface or at least its fluorophore labeled

Figure 1. (A) Representative kinetic scan (collected at a fixed angle of 67.1°) during the preparation of the sensor chip. The reflectivity (SPRsignal) and the fluorescence intensity (SPFS-signal) are shown. First, a mixture of MCB and MPA (5 μM each) were coadsorbed with MB1 (2 μM) in tris buffer for a period of 2 h. A zoom in the adsorption kinetics is shown in the inset. After adsorption of the mixture of MB1 and short thiols for 120 min, the surface was rinsed with pure water and a mixture of MCB and MPA (0.5 mM each) in water was incubated (at t = 140 min) for 60 min. Next, the surface was rinsed with tris buffer again and incubated overnight (not shown). (B) Angular scan curves showing the SPR and the SPFS signal of the bare gold surface, the gold surface after modification with MB1 and thiol spacers (before hybridization), and after hybridization with 1 μM complementary target (T1) in tris buffer. In the inset, the changes in the SPR-signals around the minimum angle are also shown.

counts per second, cps) collected for the bare gold surface, for the MB1 sensor surface, and for the sensor surface after hybridization with a fully complementary target at a concentration of 1 μM for 20 min and subsequent rinsing with 10 mL buffer are shown. While for the bare gold surface in the entire scan range the fluorescence signal is rather low, for the MB1 modified sensor chips a significant amount of background fluorescence with a maximum of 1.14 × 105 cps at an angle of 67.1° is detected. After hybridization with target, the fluorescence increases to a maximum value of 2.09 × 106 cps at the same angle. Hence, by hybridization with 1 μM fully complementary target, the fluorescence increases by one order of magnitude. In Figure 1B (inset), the changes in the angle dependent reflectivity scans (SPR scan curves) around the minimum angle are also shown. While the minimum in reflectivity changes by 0.2° after preparation of the MB1 sensor surface, the change after hybridization with 1 μM fully complementary target is only marginal. Thus, as described previously,34 by SPFS, a much higher sensitivity can be reached than by common SPR measurements. It is also noteworthy that when the sensor chips were rinsed with buffer solution after hybridization (before the 14362

dx.doi.org/10.1021/la504105x | Langmuir 2014, 30, 14360−14367

Langmuir

Article

The shorter MB1 and MB2 comprise the same beacon sequence, which has also been used in a previous study of surface bound MBs using fluorescence microscopy as a readout technique.22 MB1 and MB2 differ only by two thymidines, which are inserted in MB1 as a spacer between the three dithiane rings at the 3′ end and the beacon sequence. These two thymidines, which were supposed to introduce additional flexibility and thereby enhance the hybridization efficiency, are left out in MB2. When comparing the fluorescence intensity of MB1 and MB2 after hybridization with the target (see Figure 2A,B), it seems that these two thymidines have no significant effect on the performance when a mixture of the relatively short molecules MCB and MPA was used as thiol spacers. Among all thiol spacers for a mixture of MCB and MPA, the highest fluorescence intensity could be obtained after hybridization with target. For MB1, mercaptohexanol (MCH) and mercaptopropanesulfonic acid (MPS) also resulted in relatively high fluorescence signals, whereas the fluorescence signal was less intense when the rather long 11-Mercapto-1-undecyl dihydrogen phosphate (MDPA) was used. The fluorescence increase upon hybridization was lower for MB2 in comparison to MB1 when MPS or MDPA were used as spacer. Apparently bulky or long thiol spacers hamper the hybridization efficiency for MB2, which is attached to the surface by a rather short tether. In MB1, the surface tether has been extended by two thymidines, which improves the hybridization efficiency when rather long thiols or thiols with bulky end groups are present. The same influence of the thiol spacer length on the sensor performance has been observed previously for surface bound electrochemical DNA sensors.37 However, the fluorescence background signal for the MBs in the closed state is in general somewhat lower for MB2 than for MB1. As a main strategy to further improve the performance of the MBs, the dynamic range, i.e., the difference in the fluorescence signal between open and closed state should be increased. This could be reached by MB structures, in which in the closed state, the fluorophore is oriented at the surface as closely as possible (as shown in this study for MB2). In order to reach optimum performance, such MBs have to be used in combination with short thiol spacers. Influence of the Thiol Spacer on the Ability of the MBs to Detect SNP. In order to prove whether there is an influence of the thiol spacer on the ability of the MB modified sensor chips to detect individual SNPs, MB1 was evaluated together with a mixture of MCB and MPA and with MCH as thiol spacers. MB1 was chosen for this study because MB1 showed good performance upon hybridization with a fully complementary target when both thiol spacers were used (either a mixture of MCB and MPA or MCH, see Figure 2A). In order to investigate the influence of SNPs, the fully complementary target was permutated at positions 6 and 20, respectively. The bases C and G, which are present in the fully complementary target at these positions respectively, were exchanged with each of the other nonmatching DNA bases. As it is shown in Figure 3A,B, in all cases there is a clear difference in the fluorescence increase for hybridization with a fully complementary target or targets bearing a single base mismatch. In order to quantify the influence of the SNPs, the single mismatch discrimination ratios were calculated, which are defined as the increase in fluorescence intensity after hybridization with a fully complementary target divided by the increase in fluorescence intensity after hybridization with targets bearing SNPs. For the MB1/MCB/MPA modified

Figure 2. Comparison of the molecular beacons MB1 (A), MB2 (B), and MB3 (C) with the thiol spacers MCH (black), MCB/MPA (red), MDPA (blue), and MPS (purple). The SPFS angular scan curves are shown for the MB modified sensor chips before hybridization (open) and after 20 min hybridization (filled symbols) with 1 μM fully complementary target DNA and subsequent rinsing with 10 mL of buffer solution.

end is dynamically approaching the surface, e.g., by elastic bending or rotational motion,35,36 whereas MB1 and MB2 are standing at the surface in a more or less upright orientation. This second explanation is supported by the general trend, which is found for the different thiol spacers. For MB3, the highest fluorescence is obtained with MDPA, which is the longest thiol spacer in the experimental study. Thus, it seems likely that in contrast to MB1 and MB2 for the open MB3, the fluorophore labeled end is oriented closer to the gold surface down to a distance limited by the thickness of the selfassembled layer formed by the thiol spacers. 14363

dx.doi.org/10.1021/la504105x | Langmuir 2014, 30, 14360−14367

Langmuir

Article

is similar to the ratio of 3.2 (1195038cps/373450cps) with MCB/MPA thiol spacer, and the ratio of 2.4 (1146334cps/ 479862cps) for SNP (G to A) with MCH is slightly lower than the ratio 2.6 (1195039cps/459630cps) with MCB/MPA. Even though there are some minor differences, in general the influence of the thiol spacers on the single mismatch discrimination ratios seems to be relatively low. Influence of the Ratio between MB and Thiol Spacer on the Performance of the Sensor Chips. The density at which the MBs are immobilized on the gold surface may strongly affect the performance of the proposed sensor platform. In order to study the influence of the surface coverage, MB1 was adsorbed for 2 h in combination with a 1:1 mixture of MCB and MPA as thiol spacers in different ratios. While the concentration of MB1 was kept constant, the concentration of the thiol spacers was varied. Ratios between MB and thiols between 2:0 and 2:200 were examined. As described before, after adsorption of the MB1 and a thiol spacer containing solution, the surface was rinsed with water, followed by incubation of the thiol spacers in water (MCB and MPA, 0.5 mM each) for 1 h, and subsequent rinsing with buffer solution for equilibration. The average layer thickness dA calculated from the shift of the SPR angular scan curves, the average surface coverage calculated from dA using Feijter’s equation (eq 1), and the increase in the fluorescence intensity after hybridization with a fully complementary target T1 at a concentration of 1 μM for 20 min for the individual ratios between MB1 and the thiol spacers are shown in the SI Table S2. The highest fluorescence increase after hybridization with target was obtained for a ratio of (2:10) between MB1 and thiol spacers (i.e., 2 μM MB1, 5 μM MCB, and 5 μM MPA). A similar ratio (1:10) between MB and thiols yielded the highest fluorescence increase during previous studies of surface-immobilized MBs using fluorescence microscopy as the readout technique.22 Even though in the current study the ratio between MB and thiol spacers was varied over a broad range, there was only a minor influence of this variation on the resulting average layer thickness and thus on the surface coverage (values between 0.7 and 1.0 nm were calculated for the average layer thickness). This low variation could be explained by the second incubation step, in which only the thiol spacers in water (MCB and MPA, 0.5 mM each) were incubated. Possibly during this second incubation step, some of the MBs get displaced, and therefore the MB concentration will be leveled out to some extent. However, the second incubation step turned out to be very important for the proper performance of the MB-sensor chips. When MB sensor chips were prepared without the second incubation step, it was not possible to dehybridize after the first hybridization step, and the sensor chips could not be reused. As it will be shown below, MB sensor chips prepared in a proper way can be used many times (i.e. for many hybridization/ dehybridization steps) without significant decrease in performance. Homogeneity of the surface coverage. During SPFS a laser spot with a diameter of 1.2 mm is used, and the signal response is averaged over an area ≥1.1 mm2 (depending on the scan angle). In order to estimate the homogeneity of the sensor chip surface within this area, the MB2/MCB/MPA modified sensor chips were investigated by confocal fluorescence microscopy after hybridization with a fully complementary target. As shown in Figure 4, the fluorescence intensity is evenly distributed on a larger scale demonstrating a homogeneous

Figure 3. Fluorescence increase for MB1 at different time points measured at a fixed angle of (67.2°) after hybridization with fully complementary target T1 (blue), targets containing an SNP at nucleotide position 6 (from the 5′ end of the T1 strands) with C to G (red), C to A (pink), and C to T (magenta), and for targets containing an SNP at nucleotide position 20 (from the 5′ end of the T1 strands) with G to C (black), G to A (orange), and G to T (gray). While in (A) MB1 was coadsorbed with a mixture of MCB/MPA, in (B) MCH was used as thiol spacer. The SPFS signal shows the average of three measurements on the same sensor chip. The error bars show the confidence interval for P = 0.95.

chip, the increase in fluorescence intensity within the first 5 min of hybridization after adding the T1 target is about 4.5 times as high as that for SNP (C to A) target (single mismatch discrimination ratio of 4.5 = 1195039 cps/264499 cps), and after 15 min, the single mismatch discrimination ratio is still about 3.7 (1705891 cps/618217 cps), as shown in Figure 3A. In comparison, for the chip modified with MB1 and MCH, the single mismatch discrimination ratios for the SNP (C to A) target after 5 and 15 min were 6.7 (1146334cps/171517cps) and 4.1 (1425406cps/349583cps), as shown in Figure 3B. The decrease in discrimination ratio with extended hybridization time could be explained by mass transfer limitation (in the optical flow cell, the target has to reach the MB-modified surface by mass transfer, mainly diffusion). Furthermore, a comparison of the two types of chips indicates that the discrimination ratio (6.7) with MCH as thiol spacers is higher than the ratio (4.5) with MCB/MPA after 5 min of incubation for analyzing the SNP C to A at the target position 6. In contrast, for other SNPs, the single mismatch discrimination ratio with MCH is similar to or slightly lower than with the MCB/MPA thiol spacers; for instance, the ratio of 3.1 (1146334cps/372729cps) for the SNP (C to T) with MCH 14364

dx.doi.org/10.1021/la504105x | Langmuir 2014, 30, 14360−14367

Langmuir

Article

dehybridization step.22 The reason for the high reusability for the MB sensor chips used in this study might be strong attachment of the MBs to the gold surface via the three dithiane rings used as thiol anchors, which are able to form six sulfur− gold bonds.38 An additional reason might be the gentle dehybridization procedure, i.e., rinsing with pure water,29,37 which is probably less harmful to the MB/thiol layer than thermal hybridization22,33 or the use of a sodium hydroxide solution.39 Determination of the Detection Limit for the MB Modified Sensor Chips and Repeatability of the Sensor Chip Preparation. In order to determine the detection limit (blank signal plus three times the standard deviation), MB1 was investigated together with a mixture of MCB and MBA as thiol spacer, since this combination revealed the best sensitivity for MB1 in Figure 2A. For each sensor chip, the increase in the fluorescence signal was investigated during the first 20 min of hybridization with fully complementary target. The target concentration was varied stepwise. After starting at high target concentration, the concentration was subsequently lowered. In between individual hybridization steps, dehybridization was achieved by rinsing with pure water. In Figure 5A, the increase in fluorescence intensity for the MB1/MCB/MBA modified sensor chip with target at concentrations from 500 nM to 1 nM (and blank) is shown. For this MB modified sensor chip, the detection limit was at a concentration of 1 nM. In Figure 5B, the increase in fluorescence intensity with respect to the target concentration is plotted for 5, 10, and 20 min of hybridization. As it can be seen from Figure 5B, within the first minutes of hybridization, the fluorescence intensity increases almost linearly with target concentration, whereas after 20 min and also after 10 min, there are larger deviations from linearity especially at higher target concentrations. This deviation can be explained by mass transfer limitation as discussed before. While for MB1, a detection limit of 1 nM was determined, an even lower detection limit is expected for MB2, since for MB2, the highest dynamic range and the lowest background fluorescence was observed (see Figure 2). Furthermore, we were interested in the repeatability of the sensor chip preparation procedure. Therefore, we determined the detection limit of MB2 from three independently prepared MB2/MCB/ MPA modified sensor chips, as shown in SI Figure S3. In SI Figure S3, the increase in fluorescence intensity for three MB2/ MCB/MBA modified sensor chips with blank and target concentrations starting from 100 pM and increasing up to 500 nM is shown. The three MB2/MCB/MBA modified sensor chips show a similar response in terms of background fluorescence, sensitivity, reusability, and detection limit. As expected for this type of sensor chip, a somewhat lower detection limit of 500 pM was reached than that for the MB1/ MCB/MBA modified sensor chip. A detection limit in the lower nM or even in the higher pM range as reached in the current study is in the same order of magnitude as for other optical or electrochemical label-free assays based on DNA hairpin structure modified sensor chips.19,21,22,24,40

Figure 4. Confocal fluorescence images at larger (A) and smaller scale (B) of a MB2/MCB/MPA modified sensor chip surface after hybridization with fully complementary target (1 μM) for 20 min. The dark area (arrow) in panel (A) is partially bleached by the laser before imaging. The corresponding cross sections (C and D) show minor intensity changes (apart from the bleached area). In (B), the lateral resolution and the pixel size were 220 and 20 nm, respectively. The averaged profile in (D) is shown in red.

surface coverage of the beacons. Partial bleaching the dye molecules by the laser (arrow in Figure 4A) proves that the intensity is due to fluorescence. On areas of the gold coated glass slide, which are not modified with MB2 and thiols (bare gold surface), no significant background signal from the surface due to other effects, such as scattering, is detected (data not shown). Local variations in the fluorescence intensity of up to 16% on the larger and 5.3% on the smaller scale along the averaged profile (averaged over 5 pixels, since the pixel size of 20 nm is 10x smaller than the lateral resolution of 220 nm) indicate minor heterogeneities in the surface coverage of the beacon or the ability to bind the complementary DNA strands. These heterogeneities are at a length scale of around 1 μm (Figure 4D) and thus far below the spot size of the laser. Therefore, the surface of the sensors can be regarded as homogeneously covered for SPFS measurements. Reusability of the Sensor Platform. In order to prove the reusability of the MB modified sensor chips, ten hybridization dehybridization cycles were carried out on a MB1/MCB/MPA modified sensor chip (see SI Figure S2). Regeneration of the sensor chip (dehybridization) was achieved by intense rinsing with pure water followed by equilibration in buffer solution. For each hybridization step, the fluorescence signal was collected during the first 20 min of incubation with fully complementary target T1. In the first two hybridization steps, the target was added at a concentration of 1 μM, and in the latter eight at a concentration of 100 nM. For the latter eight hybridization steps, the relative standard deviation (RSD) of the fluorescence increase was 3%. This low RSD suggests a good reusability over a period of almost 2 days of performance. The reusability of the sensor chip is better than that reported for a different approach, in which the fluorescence intensity of an MB-modified gold sensor chip decreased by 40% after the first (thermal)

■

CONCLUSIONS In the current study, three different surface bound MBs were compared using SPFS as optical readout technique. The MBs were chemisorbed on gold, together with different types of thiols serving as spacers between individual MBs. While MB1 and MB2, both consisting of 36 bases (9 base pairs stem plus 14365

dx.doi.org/10.1021/la504105x | Langmuir 2014, 30, 14360−14367

Langmuir

Article

in length, are required to answer this question and to establish general design rules for surface bound molecular beacons. By comparison of MB1 and MB2, we could show that a shorter linker between the MB stem and the gold surface diminishes the background fluorescence and thereby increases the dynamic range for concentration dependent target detection. MBs attached by rather short linkers (leading to low background fluorescence as shown for MB2) have to be combined with short thiol spacers, since longer thiol spacers would sterically hinder the hybridization of a complementary target. For MB1 and MB2, a detection limit in the lower nM or higher pM range could be reached, respectively. By lowering the background fluorescence and/or optimizing the distance between dye and gold surface in the open state, the sensitivity and the detection limit could be further improved. The MB and MB2 modified sensor chips could be prepared in a highly repeatable manner and used for several detection events at the same chip without significant decrease in performance with an increasing number of (de)hybridization steps. Fluorescence microscopy measurements revealed that with respect to the area of the laser spot used for readout, the MBs are homogeneously distributed over the surface. Using MB1, the ability to detect single nucleotide polymorphism has also been evaluated. In all cases, the fluorescence increase for a singly mismatched target was significantly lower than that for the fully complementary target. Depending on the individual mismatch single mismatch discrimination, ratios between 2.3 and almost 7 were obtained after 5 min of hybridization. In this context, it is noteworthy that if an MB modified sensor chip were used for the detection of specific oligonucleotides, such as mi-RNA in real samples, it is not very likely that, instead of fully complementary target, a singly mismatched target is present at high concentration. Furthermore, after detection by SPFS, the target could be isolated upon dehybridization by rinsing with pure water and subsequently used for amplification and additional analytical studies. Hence, the MB modified sensor chips combine the advantage of optical label-free detection with the possibility to perform an affinity chromatography type of purification.

Figure 5. (A) Determination of the detection limit for a MB1/MCB/ MPA modified sensor chip. Fully complementary target T1 was incubated at concentrations of 500 nM to 1 nM in buffer solution for 20 min. After each hybridization step, dehybridization was achieved by rinsing with pure water. The short dashed blue line represents the fluorescence background. The inset shows a zoom in the SPFS signals collected for hybridization with target solution at concentrations of 10 nM to 1 nM and blank. (B) Plot of the fluorescence increase after 5, 10, and 20 min of hybridization with fully complementary target for concentrations between 1 nM and 500 nM. The inset shows a zoom in the lower concentration range up to 50 nM.

■

ASSOCIATED CONTENT

S Supporting Information *

18 bases loop), differed only in the length of the linker for surface attachment, the significantly longer MB3, consisting of 56 bases, (9 base pairs stem plus 38 bases loop), comprised an entirely different sequence. In contrast to MB1 and MB2, in which the thiol anchors were attached at the 5′ end, in MB3 the thiol anchor was introduced at the 3′ end. From the three MBs under investigation, only MB3 was sufficiently long to reach the optimum distance of 20 nm between fluorophore and gold surface for maximum fluorescence intensity during SPFS.34 Interestingly our measurements revealed that in comparison to MB1 and MB2, the fluorescence of MB3 was much lower after hybridization with target. Apparently, MB1 and MB2 are oriented in a more or less upright conformation at the surface, whereas the fluorophore of the longer MB3 is on average oriented closer to the surface. Since the sequence of MB3 was entirely different from that of MB1 and MB2, it is not yet clear whether this is a general tendency, i.e., the longer the MB structure, the higher the probability for the MB’s fluorophore to be oriented close to the surface after hybridization. Future measurements on a series of MBs with the same stem region and a loop region, which comprise a similar sequence but differ

Additional experimental details, as well as measurements regarding reusability, repeatability, and detection limit. This material is available free of charge via the Internet at http:// pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work has received funding from the European Research Council under the European Community’s Seventh Framework Programme (FP7/2007-2013) / ERC Grant Agreement No. 240544, ERC grant agreement no. 279202, the University of Siegen, the Deutsche Forschungsgemeinschaft (DFG Grant No. INST 221/87-1 FUGG), and from the country North Rhine-Westphalia. 14366

dx.doi.org/10.1021/la504105x | Langmuir 2014, 30, 14360−14367

Langmuir

■

Article

(24) Xiao, Y.; Lou, X.; Uzawa, T.; Plakos, K. J.; Plaxco, K. W.; Soh, H. T. An electrochemical sensor for single nucleotide polymorphism detection in serum based on a triple-stem DNA probe. J. Am. Chem. Soc. 2009, 131, 15311−15316. (25) Vallee-Belisle, A.; Ricci, F.; Uzawa, T.; Xia, F.; Plaxco, K. W. Bioelectrochemical switches for the quantitative detection of antibodies directly in whole blood. J. Am. Chem. Soc. 2012, 134, 15197− 15200. (26) Paleček, E. Surface-attached molecular beacons light the way for DNA sequencing. Trends Biotechnol. 2004, 22, 53−58. (27) Lubin, A. A.; Lai, R. Y.; Baker, B. R.; Heeger, A. J.; Plaxco, K. W. Sequence-specific, electronic detection of oligonucleotides in blood, soil, and foodstuffs with the reagentless, reusable E-DNA sensor. Anal. Chem. 2006, 78, 5671−5677. (28) Xiao, Y.; Qu, X.; Plaxco, K. W.; Heeger, A. J. Label-free electrochemical detection of DNA in blood serum via target-induced resolution of an electrode-bound DNA pseudoknot. J. Am. Chem. Soc. 2007, 129, 11896−11897. (29) Nöll, G.; Su, Q.; Heidel, B.; Yu, Y. A reusable sensor for the label-free detection of specific oligonucleotides by surface plasmon fluorescence spectroscopy. Adv. Healthcare Mater. 2014, 3, 42−6. (30) De Feijter, J.; Benjamins, d. J.; Veer, F. Ellipsometry as a tool to study the adsorption behavior of synthetic and biopolymers at the air− water interface. Biopolymers 1978, 17, 1759−1772. (31) Tumolo, T.; Angnes, L.; Baptista, M. S. Determination of the refractive index increment (dn/dc) of molecule and macromolecule solutions by surface plasmon resonance. Anal. Biochem. 2004, 333, 273−279. (32) Lee, L.; Johnston, A. P.; Caruso, F. Manipulating the salt and thermal stability of DNA multilayer films via oligonucleotide length. Biomacromolecules 2008, 9, 3070−3078. (33) De Paul, S. M.; Falconnet, D.; Pasche, S.; Textor, M.; Abel, A. P.; Kauffmann, E.; Liedtke, R.; Ehrat, M. Tuned graft copolymers as controlled coatings for DNA microarrays. Anal. Chem. 2005, 77, 5831−5838. (34) Neumann, T.; Johansson, M. L.; Kambhampati, D.; Knoll, W. Surface−plasmon fluorescence spectroscopy. Adv. Funct. Mater. 2002, 12, 575−586. (35) Anne, A.; Demaille, C. Dynamics of electron transport by elastic bending of short DNA duplexes. Experimental study and quantitative modeling of the cyclic voltammetric behavior of 3′-ferrocenyl DNA end-grafted on gold. J. Am. Chem. Soc. 2006, 128, 542−557. (36) Anne, A.; Demaille, C. Electron transport by molecular motion of redox-dna strands: unexpectedly slow rotational dynamics of 20-mer ds-DNA chains end-grafted onto surfaces via c6 linkers. J. Am. Chem. Soc. 2008, 130, 9812−9823. (37) Ricci, F.; Zari, N.; Caprio, F.; Recine, S.; Amine, A.; Moscone, D.; Palleschi, G.; Plaxco, K. W. Surface chemistry effects on the performance of an electrochemical DNA sensor. Bioelectrochemistry 2009, 76, 208−213. (38) Liepold, P.; Kratzmüller, T.; Persike, N.; Bandilla, M.; Hinz, M.; Wieder, H.; Hillebrandt, H.; Ferrer, E.; Hartwich, G. Electrically detected displacement assay (EDDA): A practical approach to nucleic acid testing in clinical or medical diagnosis. Anal. Bioanal. Chem. 2008, 391, 1759−1772. (39) Ramachandran, A.; Flinchbaugh, J.; Ayoubi, P.; Olah, G. A.; Malayer, J. R. Target discrimination by surface-immobilized molecular beacons designed to detect Francisella tularensis. Biosens. Bioelectron. 2004, 19, 727−736. (40) Qavi, A. J.; Mysz, T. M.; Bailey, R. C. Isothermal discrimination of single-nucleotide polymorphisms via real-time kinetic desorption and label-free detection of DNA using silicon photonic microring resonator arrays. Anal. Chem. 2011, 83, 6827−6833.

REFERENCES

(1) Ferguson, J. A.; Boles, T. C.; Adams, C. P.; Walt, D. R. A fiberoptic DNA biosensor microarray for the analysis of gene expression. Nat. Biotechnol. 1996, 14, 1681−1684. (2) Lockhart, D. J.; Winzeler, E. A. Genomics, gene expression and DNA arrays. Nature 2000, 405, 827−836. (3) Giljohann, D. A.; Mirkin, C. A. Drivers of biodiagnostic development. Nature 2009, 462, 461−464. (4) Ge, J.; Budowle, B.; Planz, J. V.; Chakraborty, R. Haplotype block: A new type of forensic DNA markers. Int. J. Legal. Med. 2010, 124, 353−361. (5) Jobling, M. A.; Gill, P. Encoded evidence: DNA in forensic analysis. Nat. Rev. Genet. 2004, 5, 739−751. (6) Ramsay, G. DNA chips: State-of-the art. Nat. Biotechnol. 1998, 16, 40−44. (7) Syvänen, A.-C. Accessing genetic variation: Genotyping single nucleotide polymorphisms. Nat. Rev. Genet. 2001, 2, 930−942. (8) Jonstrup, S. P.; Koch, J.; Kjems, J. A microRNA detection system based on padlock probes and rolling circle amplification. RNA 2006, 12, 1747−1752. (9) Dangwal, S.; Bang, C.; Thum, T. Novel techniques and targets in cardiovascular microRNA research. Cardiovasc. Res. 2012, 93, 545− 554. (10) Broude, N. E. Stem-loop oligonucleotides: A robust tool for molecular biology and biotechnology. Trends Biotechnol. 2002, 20, 249−256. (11) Dubertret, B.; Calame, M.; Libchaber, A. J. Single-mismatch detection using gold-quenched fluorescent oligonucleotides. Nat. Biotechnol. 2001, 19, 365−370. (12) Tyagi, S.; Kramer, F. R. Molecular beacons: Probes that fluoresce upon hybridization. Nat. Biotechnol. 1996, 14, 303−308. (13) Heinlein, T.; Knemeyer, J.-P.; Piestert, O.; Sauer, M. Photoinduced electron transfer between fluorescent dyes and guanosine residues in DNA-hairpins. J. Phys. Chem. B 2003, 107, 7957−7964. (14) Williams, D. J.; Hall, K. B. Thermodynamic comparison of the salt dependence of natural RNA hairpins and RNA hairpins with nonnucleotide spacers. Biochemistry 1996, 35, 14665−14670. (15) Riccelli, P.; Merante, F.; Leung, K.; Bortolin, S.; Zastawny, R.; Janeczko, R.; Benight, A. S. Hybridization of single-stranded DNA targets to immobilized complementary DNA probes: Comparison of hairpin versus linear capture probes. Nucleic Acids Res. 2001, 29, 996− 1004. (16) Steemers, F. J.; Ferguson, J. A.; Walt, D. R. Screening unlabeled DNA targets with randomly ordered fiber-optic gene arrays. Nat. Biotechnol. 2000, 18, 91−94. (17) Liu, X.; Tan, W. A fiber-optic evanescent wave DNA biosensor based on novel molecular beacons. Anal. Chem. 1999, 71, 5054−5059. (18) Piestert, O.; Barsch, H.; Buschmann, V.; Heinlein, T.; Knemeyer, J.-P.; Weston, K. D.; Sauer, M. A single-molecule sensitive DNA hairpin system based on intramolecular electron transfer. Nano Lett. 2003, 3, 979−982. (19) Fang, X.; Liu, X.; Schuster, S.; Tan, W. Designing a novel molecular beacon for surface-immobilized DNA hybridization studies. J. Am. Chem. Soc. 1999, 121, 2921−2922. (20) Du, H.; Disney, M. D.; Miller, B. L.; Krauss, T. D. Hybridization-based unquenching of DNA hairpins on Au surfaces: prototypical “molecular beacon” biosensors. J. Am. Chem. Soc. 2003, 125, 4012−4013. (21) Peng, H.-I.; Strohsahl, C. M.; Leach, K. E.; Krauss, T. D.; Miller, B. L. Label-free DNA detection on nanostructured Ag surfaces. ACS Nano 2009, 3, 2265−2273. (22) Du, H.; Strohsahl, C. M.; Camera, J.; Miller, B. L.; Krauss, T. D. Sensitivity and specificity of metal surface-immobilized “molecular beacon” biosensors. J. Am. Chem. Soc. 2005, 127, 7932−7940. (23) Fan, C.; Plaxco, K. W.; Heeger, A. J. Electrochemical interrogation of conformational changes as a reagentless method for the sequence-specific detection of DNA. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 9134−9137. 14367

dx.doi.org/10.1021/la504105x | Langmuir 2014, 30, 14360−14367