Chimeric RNA–DNA Molecular Beacons for Quantification of Nucleic

Apr 1, 2013 - The chimeric RNA–DNA MB proves to be a more sensitive and selective tool for the quantification of nucleic acids, DNA mismatches, and ...
0 downloads 0 Views 511KB Size
Article pubs.acs.org/ac

Chimeric RNA−DNA Molecular Beacons for Quantification of Nucleic Acids, Single Nucleotide Polymophisms, and Nucleic Acid Damage Amira F. El-Yazbi and Glen R. Loppnow* Department of Chemistry, University of Alberta, Edmonton, AB T6G 2G2 Canada S Supporting Information *

ABSTRACT: Single nucleotide polymorphisms (SNPs) are the main cause for variations in the human genome. DNA lesions, such as cyclobutane pyrimidine dimers (CPDs), [6−4] pyrimidine-pyrimidinones, dewar pyrimidinones, and photohydrates, can subsequently lead to mutagenesis, carcinogenesis, and cell death. Much effort has focused on methods for detecting DNA, SNPs, or damaged nucleic acids. However, almost all of the proposed methods consist of multistep procedures, are limited to specific types of damage, some of these methods require expensive instruments, and some suffer from a high level of interferences. In this paper, we present a novel, simple, mix-and-read assay for the detection of nucleic acids that is general for all types of SNPs and nucleic acid damage. This method uses a chimeric RNA−DNA molecular beacon (chMB). The calibration curve of the chMB for detecting single base mismatch and ultraviolet (UV)-induced DNA damage shows good linearity (R2 = 0.981 and 0.996, respectively) and limits of detection of 10.4 ± 2.2 and 8.64 ± 1.2 nM, respectively. The chimeric RNA−DNA MB proves to be a more sensitive and selective tool for the quantification of nucleic acids, DNA mismatches, and UV-induced DNA damage than DNA MBs.

S

electrophoresis,11−13 electrochemical,12,13 mass spectrometric, 1 4 − 1 7 high-performance liquid chromatography (HPLC),15,18 and polymerase chain reaction (PCR) amplification13,19 methods. While useful, these techniques are destructive, elaborate, and time-consuming. In addition, electrophoretic and chromatographic techniques need to isolate the probe−target hybrids from an excess of unhybridized probes, which include additional steps that may introduce more lesions.9 Other spectroscopic-based techniques have recently been used to characterize nucleic acid damage. Typically, fluorescent methods offer enhanced sensitivity and the potential for use in situ or in vivo. Differences in the fluorescence lifetime of a dye intercalated in undamaged and damaged DNA have been used to detect DNA damage.20 Fluorescently labeled antibodies provide a highly selective probe of particular damage photoproducts, such as thymine cyclobutyl photodimers.21 More recently, hybridization probes22−31 have been used to detect single nucleotide polymorphisms (SNPs), gene mutations, and DNA damage. The inherent sensitivity of fluorescence makes these hybridization probes a sensitive tool for the detection of nucleic acid damage. The most studied hybridization probes are binary probes28−31 and molecular beacons (MB).22−27 Binary probes

ingle nucleotide polymorphisms (SNPs) and nucleic acid damage are two important factors that determine variability and change in the human genome. SNPs are a substitution, insertion, or deletion at a single base position on a DNA strand. There is expected to be, on average, one SNP for every 1000 bases of the human genome.1 Some variations located in genes are suspected to alter both the protein structure and the expression level. SNPs are the most abundant form of DNA sequence variation in the human genome and contribute to phenotypic diversity, influencing an individual’s anthropometric characteristics, risk of certain disease, and variable response to drugs and the environment.1 Due to their dense distribution across the genome, SNPs are used as markers in medical diagnosis and in personalized medicine. Exposure of nucleic acids to a variety of internal and external agents can lead to another source of variability in the genome: damage and mutation. Nucleic acid damage includes genotoxic molecular-scale lesions such as cyclobutane photodimers (CPDs), [6−4] pyrimidine pyrimidinones, dewar pyrimidinones, and photohydrates.2 Other damage includes single- and double-strand breaks, 8-oxoguanosine, other oxidation products, and cross-links. All these damage products lead to miscoding during DNA replication and will result in mutagenesis, carcinogenesis, aging, and cell death.3−6 The effects of SNPs and nucleic acid damage on the molecular origin of mutagenesis, carcinogenesis7,8, and the need for a sensitive and precise measurement of SNPs and damaged DNA led to the development of a number of techniques for their detection. These include gel electrophoresis,9,10 capillary © 2013 American Chemical Society

Received: June 27, 2012 Accepted: April 1, 2013 Published: April 1, 2013 4321

dx.doi.org/10.1021/ac301669y | Anal. Chem. 2013, 85, 4321−4327

Analytical Chemistry

Article

strand has unique advantages specific to the application. For instance, the DNA in the hybridizing arms in gene targeting enhances the chemical cleavage step, while the RNA loop did not affect its enzymatic activity.55 Furthermore, an uninterrupted stretch of DNA bases within the chimera is known to be active in sequence alteration while RNA residues aid in complex stability in gene repairing.57 The chMB used in this work is designed so that the RNA bases are complementary to the oligonucleotide target, while DNA bases mainly constitute the stem of the MB. Typically, the thermal stability of RNA duplexes and RNA−DNA hybrid duplexes is greater than that of DNA duplexes when their sequences are identical.61 Because of this, the chMB-DNA hybrid is expected to have a higher fluorescence signal than typical MB-DNA hybrids. The DNA stem is kept in the chMB to provide a reasonable melting temperature that allows the stem to open in the presence of the undamaged oligonucleotide target. To our knowledge, this study represents the first use of chMBs for the quantification of nucleic acids, mismatches, and DNA damage. The sensitivity and selectivity of the chMB to nucleic acids, single base mismatch, and UVC-induced DNA damage were examined and found to be superior to conventional DNA MBs.

are composed of two single-stranded oligonucleotides complementary to different parts of the oligonucleotide target and are labeled with a fluorophore at the 3′ or 5′ end. After adjacent hybridization of both strands to the complementary oligonucleotide, the labels are close enough to enable either FRET28 or strong fluorescence quenching31 to occur. An advantage of binary probes is that they do not yield false positive or nonspecific signals. However, one of their limitations is that the entropy of a binary probe system decreases more when bound to the target (three independent species becoming one hybrid), which reduces the equilibrium stability of the hybrid.32 Also, the hybridization kinetics of binary probes are slow because binary probe hybridization depends on the binding of two different components to the target. Molecular beacons22−25 are DNA hairpins labeled with a quencher on one end and a fluorophore on the other end. In the hairpin form, the quencher and fluorophore are in close proximity and the fluorescence is quenched. However, when the MB hybridizes to a complementary oligonucleotide target, a significant increase in fluorescence is detected. In MBs, the loop sequence is complementary to the target DNA sequence, and the resulting hybridization spatially separates the quencher and the fluorophore. In the presence of a single base mismatch or a single site of damage, the hybrid formed between the probe and the mismatch sequence is less stable, since this target is not perfectly complementary to the loop of the probe. Thus, the observed fluorescence intensity is lower for a sample containing mismatches or damaged DNA bases compared to one containing the perfectly complementary sequences. This inherent signal mechanism allows MBs to function as highly sensitive probes of nucleic acids, with their advantages of simplicity and high specificity. Consequently, a variety of sensitive assays have been developed using MBs constructed from DNA nucleotides to detect nucleic acid damage,25,26 SNPs,33,34 and gene mutations,35 to quantify the concentration of a target molecule during real-time or quantitative PCR36,37 and for intracellular imaging of nucleic acids.38 Alternative approaches to DNA-MBs have been developed in which MBs are constructed using modified DNA backbones, such as phosphorothioate,39 2′-O-methyl RNA bases,40−42 peptide nucleic acids (PNAs),43−46 and locked nucleic acids (LNAs).26,47,48 Backbone modifications have several attractive advantages, such as higher binding affinity, increased specificity, faster hybridization kinetics, and fluorescence background suppression because of a reduction in dynamic opening of the stem.49 However, there are some disadvantages and limitations, such as the toxicity occasionally associated with phosphorothioate-containing oligonucleotides.50 2′-O-Methylmodified MBs open up nonspecifically in cells, possibly due to protein binding.41,42 The neutral PNAs tend to self-aggregate51 and fold in a way that interferes with duplex formation.52 PNAs also change their physical properties substantially with small changes in sequence.53,54 While LNAs have excellent base mismatch discrimination capability,47,48 they are less selective in the detection of nucleic acid damage than DNA-MBs.26 In this paper, we present another type of backbone modification in which a chimeric RNA−DNA MB (chMB) is used for the quantification of base mismatches and UV-induced DNA damage. Recently, chimeric RNA−DNA strands have been widely used in gene targeting,55 as gene repair materials,56,57 as PCR primers,58 as a MB for the detection of ribonuclease activity,59 and as sequence-specific mediators of RNA interference.60 In each of these applications, the chimeric



EXPERIMENTAL SECTION The single-strand oligonucleotide targets, the chMB and the DNA MB probes (Table 1), were obtained from Integrated DNA Technologies Inc. (Coralville, IA). The oligonucleotide targets were purified by standard desalting, and the probes were purified by HPLC purification. The chemical reagents were obtained and used as described in the Supporting Information. UV irradiation of the target oligonucleotides and the Table 1. Oligonucleotide Sequences name chMB-A DNA MB-A chMB-B DNA MB-B chMB-C DNA MB-C Tm0 Tm1 Tm12 Tm17 TGG Trandom

nucleotide sequencea 5′-(6-FAM)-CCCrArArArArArArArArArArArArArArArArATTTGGG-(3-DAB)-3′ 5′-(6-FAM)-CCCAAAAAAAAAAAAAAAAATTTGGG-(3DAB)-3′ 5′-(6-FAM)-CCGrArArArArArArArArCrCrArArArArArArATTTCGG-(3-DAB)-3′ 5′-(6-FAM)-CCGAAAAAAAACCAAAAAAATTTCGG-(3DAB)-3′ 5′-(6-FAM)-CCCrArGrUrArArCrCrUrArArCrCrUrCrGrArCACTGGG-(3-DAB)-3′ 5′-(6-FAM)-CCCAGTAACCTAACCTCGACACTGGG-(3DAB)-3′ 3′-TTT TTT TTT TTT TTT TT-5′ 3′-TTT TTT TTA TTT TTT TT-5′ 3′-CTC GAA GCT AGG TTA CT-5′ 3′-AAA AAA AAA AAA AAA AA-5′ 3′-TTT TTT TTG GTT TTT TT-5′ 3′-TCA TTG GAT TGG AGC TG-5′

a

Sequences of the targets, chimeric RNA−DNA MBs (chMBs), and DNA MBs. Tm0 is the perfect complementary target; Tm1 is the oligonucleotide target with 1 mismatch, and Tm12 is the target with 12 mismatches. Tm17 is a completely noncomplementary target to MB-A; TGG is the perfect complementary target to MB-B, and Trandom is the perfect complementary target to MB-C. The underlined bases represent the stem of the MB and the bold bases represent the mismatched bases to either chMB or DNA MB-A. “FAM” denotes the 6-carboxyfluorescein fluorophore; “DAB” denotes the dabcyl quencher, and “r” denotes a ribonucleotide. 4322

dx.doi.org/10.1021/ac301669y | Anal. Chem. 2013, 85, 4321−4327

Analytical Chemistry

Article

subsequent absorption and fluorescence measurements of the irradiated solutions mixed with the chMB and DNA MB probes were the same as described previously26,27 and are described in more detail in the Supporting Information.

higher temperatures, the hybrid melts and the chMB reforms the hairpin, leading to a decrease in fluorescence. At temperatures higher than the stem melting temperature (∼48 °C), the chMB again forms a random coil and the fluorescence shows the same trend as the chMB alone. Besides design, various other experimental factors were studied for optimizing the fluorescence of the chMB and are discussed in more detail in the Supporting Information. Nucleic Acid Detection by the chMB. Fluorescence measurements were performed to test the sensitivity of the chMB to detect DNA and to compare it to the DNA MB. The DNA MBs (Table 1) used in this study was designed to the same sequence as the chMBs, with 2′-deoxyribonucleotides comprising the whole loop and stem of the MB. Each MB was hybridized with the perfectly complementary oligonucleotide target (Table 1). Experiments show that 4 h are required for the hybridization step to be completed (data not shown). Figure 2 shows the fluorescence of chMB-A and DNA MB-A



RESULTS AND DISCUSSION The chMBs were carefully designed to maximize their performance as sensitive and specific probes for SNPs and UV-induced nucleic acid damage. This design ensures that the chMB can selectively discriminate single damage sites or mismatches in oligonucleotides. Table 1 demonstrates the structure of the chMBs used in this study. The loop region is composed of ribonucleotides and the stem region is composed of six base pairs, of which three nucleotides are ribonucleotides while the rest are 2′-deoxyribonucleotides. The MBs are designed such that the oligonucleotide target is complementary to the loop region and three nucleotides in the stem region, to avoid sticky end pairing.47 The design of the MB maximizes discrimination of damaged versus undamaged targets, due to the Tm’s of the stem and hybrid; designing the MB to have a Tm for the stem 5−10 °C higher than the Tm of the hybrid ensures maximum sensitivity.26 The fluorescence of the chMB alone and of the hybrid with the perfectly complementary target were measured as a function of temperature to obtain the thermal denaturation profiles (Figure 1). We measured the fluorescence at 520 nm of

Figure 2. Calibration curve for the detection of undamaged poly dT17 target by 2 μM chMB-A (●) and 2 μM DNA MB-A (□). Inset shows the fit to the linear portions of the calibration curves and has the same axes labels as the main graph. Fit parameters are shown in Table 2. Each data point is an average of three replicate measurements, and the error bars correspond to the standard deviation of each measurement. “c.p.s.” denotes counts per second.

(Table 1) with varying concentrations of the perfectly complementary oligonucleotide target. Note that the response of the DNA MB-A at all target concentrations is lower than that from the chMB-A. However, at zero target concentration, the emission of the chMB is slightly lower than that of the DNA MB, increasing the sensitivity of the chMB to detect the oligonucleotide target from the decreased fluorescence background. This can be attributed to the three RNA nucleotides present in one side of the chMB-A stem (Table 1), resulting in an increase of the chMB-A stem stability compared to the stem of DNA MB-A. Because of this higher stability, the quenching is enhanced between the fluorescein fluorophore and the dabcyl quencher. When titrated with target oligonucleotide, both the chMB-A and the DNA MB-A hybridize with the target and their fluorescence intensities increase with increasing target concentration. After 1.0 equivalent of the target was added to the chMB-A and the DNA MB-A solutions (i.e., at 2 μM target concentration), higher target concentrations produce no change in the fluorescence intensity (Figure 2). It should be noted that the calibration curve of the MB-target hybrid can be

Figure 1. Thermal denaturation profiles for 200 nM chMB-A alone (■), in the presence of a 10-fold excess of perfectly complementary oligonucleotide target sequence (□) and in the presence of a 10-fold excess of the UV-irradiated oligonucleotide target sequence for 1 (▲) and 10 (△) min. These profiles are scaled to the fluorescence of the MB−Tm0 hybrid at 20 °C. The lines are guides for the eye.

200 nM MB alone and in the presence of a 10-fold excess of perfectly complementary oligonucleotide target at temperatures between 20−72 °C (Figure 1), both of which show the expected profile. At low temperatures, the MB alone acquires the hairpin structure, the fluorophore and the quencher are in close proximity, and the fluorescence is quenched. At the melting temperature of the stem, the fluorescence gradually increases with temperature until reaching a plateau, indicating that the distance between the fluorophore and the quencher is constant in the random coil structure.22 For the chMB-target hybrid, the fluorophore and the quencher are far apart and maximum fluorescence is observed at low temperatures. At 4323

dx.doi.org/10.1021/ac301669y | Anal. Chem. 2013, 85, 4321−4327

Analytical Chemistry

Article

Table 2. Analytical Parameters for the Quantification of Different Oligonucleotides with chMB-A and DNA MB-Aa mismatched oligonucleotidec

undamaged oligonucleotides parameter

b

linear dynamic range (μM) R2 sensitivity (cps M‑1) LOD (nM) LOQ (nM)

damaged oligonucleotidesd

DNA MB

chMB

DNA MB

chMB

DNA MB

chMB

0.40−2.0 0.995 5.19 × 1011 76.0 253

0.00−2.0 0.998 9.43 × 1011 13.5 45.0

1.1−2.0 0.973 8.62 × 1011 45.8 153

0.00−2.0 0.981 1.22 × 1012 10.4 34.6

2.6−4.5 0.995 9.54 × 1011 41.4 138

0.00−2.5 0.996 1.47 × 1012 8.64 28.8

a

For the determination of the blank standard deviation, 20 solutions of the MB hairpin alone were used. The standard deviations of these measurements were 0.4 × 104 and 1.3 × 104 cps for the chMB and the DNA MB, respectively. bIn this table, linear dynamic range is the concentration range corresponding to the linear region in the calibration curve; R2 is the linear regression coefficient squared, and sensitivity is the slope of the calibration curve. LOD is the limit of detection and is 3 times the standard deviation of the blank divided by the sensitivity, and LOQ is the limit of quantification and is 3.3 times the LOD. cThe mismatched oligonucleotides are oligonucleotides with one mismatch. dThe damaged oligonucleotides are oligonucleotides damaged with UVC light.

complementary to T m0, chMB-B and DNA MB-B are complementary to TGG, and chMB-C and DNA MB-C are complementary to Trandom) (Table 1). Tm0 was mainly used here because of its well-known photochemistry. In this target, the CPD is the major photoproduct formed after UV irradiation, with smaller amounts of the [6−4] pyrimidine pyrimidinone and the Dewar pyrimidinone photoproducts.2 Furthermore, it has been reported that a tripyrimidine stretch represents a hot spot for UV-induced DNA damage.62 The 260 nm absorption band of Tm0 gradually bleaches with increasing irradiation time, indicating photoproduct formation and loss of the C5C6, yielding an independent spectroscopic marker for DNA damage. Thus, the loss of the 260 nm absorption band can be directly related to the concentrations of the photoproducts. Targets TGG and Trandom were chosen to compare the response of the chMB to different DNA sequences damaged with UV light in order to obtain a general idea about the selectivity and sensitivity of chMB compared to DNA MB. In order to optimize the hybridization temperature as a step to enhance the selectivity and sensitivity of the MB for UVinduced DNA damage, we measured the thermal denaturation profile of the hybrids of the chMB-A and the oligonucleotide target subjected to UVC light for 1 min. The hybrid between the chMB-A and the UV-damaged oligonucleotide target Tm0 (Figure 1) shows lower fluorescence intensity than that of the hybrid between the chMB-A and the undamaged target. For the chMB-A hybrid of the target exposed to UV light for 10 min, the thermal denaturation profile is indistinguishable from that of the chMB-A alone (Figure 1). This can be attributed to the fact that most of the oligonucleotide targets are damaged with this exposure time, and most of the chMBs are in the hairpin structure. As shown in Figure 1, the maximum discrimination in the fluorescence between the undamaged and damaged targets hybridized with the chMB occurs at 20 °C. Therefore, we have chosen this hybridization temperature for detecting the formation of the UV-induced photoproducts. The thermal denaturation profiles of the chMBs and chMB-hybrids with TGG and Trandom targets were measured (Figure S5 of the Supporting Information). The results show the same expected trend for MB and MB-hybrid melting curves as explained above. This result demonstrates that changing the ribonucleotide sequence in the loop region in the chMB did not affect the hairpin secondary structure of the chMB. In order to further examine the selectivity of the chMB and compare it to the DNA MB, T m0 , T GG , and T random oligonucleotide targets were irradiated separately at constant temperature. The MB fluorescence was measured (Figure 3)

scaled over a variety of concentration ranges (from < nanomolar to > millimolar), depending on the concentration of the MB probe used for detection. Table 2 shows the analytical parameters for the quantification of oligonucleotide target by chMB-A and DNA MB-A. The table shows that the chMB has a larger linear dynamic range. The sensitivity of the chMB is 1.8 times higher than that of the DNA MB, and the LOD and LOQ for the chMB are 5.6 times lower than those for the DNA MB. These results suggest that the chMB may be a more sensitive and selective probe of DNA damage. To examine the selectivity of the chMB to SNPs, chMB-A was hybridized with the perfect complement (Tm0), an oligonucleotide target containing 1 mismatch (Tm1), one containing 12 mismatches (Tm12), and one that is completely noncomplementary to the chMB-A (Tm17, Table 1). The DNA MB-A was also hybridized with the same targets in order to compare its selectivity to the chMB-A. The results (Figures S2 and S3 of the Supporting Information) are discussed in the Supporting Information and show that the chMBs have superior selectivity over the DNA MBs to SNPs. Also, in order to check the sensitivity of the chMB and DNA MB to detect single base mismatch in the presence of perfectly complementary targets, we measured the fluorescence intensity as a function of the Tm1:Tm0 concentration ratio at a total target concentration equal to either the chMB-A or the DNA MB-A concentration (Figure S4 of the Supporting Information). This experiment measures the ability of the probes to detect a mismatched strand within a solution of perfectly complementary strands. The results are discussed in the Supporting Information, and the parameters for the quantification of the single base mismatch from Figure S4 of the Supporting Information are shown in Table 2. The calibration curve for the chMB shows a larger linear dynamic range than the DNA MB (Figure S4 of the Supporting Information). The sensitivity of detection also is larger by a factor of ∼1.4 for the chMB, leading to a lower limit of detection (LOD) and limit of quantification (LOQ) by 4.4 times. Using the experimentally measured values in Table 2 and Figure S4 of the Supporting Information, we calculate that a single base mismatch within 3300 perfectly matched bases can be detected by the chMB. Detection of UV-Induced Photoproducts with the chMB. In characterizing the chMB for the detection of DNA damage, targets Tm0, TGG, and Trandom were irradiated with UVC light to induce UV damage. The damage is detected with a chMB and DNA MB that are perfectly complementary to the undamaged target (i.e., chMB-A and DNA MB-A are 4324

dx.doi.org/10.1021/ac301669y | Anal. Chem. 2013, 85, 4321−4327

Analytical Chemistry

Article

Figure 3. Fluorescence intensity as a function of irradiation time for the oligonucleotide targets Tm0, TGG, Trandom (□) and unirradiated target control (■) detected by the chMBs (A, B, and C) and DNA MBs (D, E, and F). Only the 2 μM target was irradiated. Aliquots of the irradiated targets were mixed with 200 nM MBs (10 mM Tris, 3 mM MgCl2, 20 mM NaCl, 1 mM EDTA, pH 7.5), and the fluorescence at 520 nm of these mixtures were then measured. The solid line through the irradiated sample points (□) is a single exponential fit to the equation IF = IF,0 + Ae−t/τ . The damage constants (τ), offsets (IF,0) and amplitudes (A) for each of the 6 plots are recorded in Table 3. The control points (■) are fit to straight lines of zero slope by eye.

after aliquots of both the irradiated and unirradiated samples of these solutions were incubated with the complementary chMBs and DNA MBs separately. It should be noted that the MBs were not irradiated, they were only incubated with aliquots of either the irradiated oligonucleotides or unirradiated controls. As shown in Figure 3, the MB fluorescence decreases with target UV irradiation dose and continues to decrease with increasing dose until it reaches a plateau. This indicates that the entire target is damaged, and the entire MB is in the hairpin structure exhibiting minimum fluorescence. The fluorescence signal as a function of irradiation time for the Tm0, TGG, and Trandom oligonucleotide targets (Figure 3) were fit to a single exponential function. The decrease in the fluorescence intensity represents the decreased stability of the damaged target-MB hybrid. Therefore, the faster the fluorescence intensity decreases, the more selective the MB is at detecting UV-induced oligonucleotide damage under identical irradiation conditions for the same target. The damage constants obtained by fitting these fluorescence curves are shown in Table 3 for Tm0, TGG, and Trandom detected by both the chMBs and DNA MBs. It is clear from Table 3 that the damage constant of the Tm0 target is ∼1.1 times faster than that of the TGG target and ∼2.6 times faster than that of the Trandom target. This indicates that the Tm0 target is more damaged under UV irradiation than both TGG and Trandom targets. Comparing the number of photoreactive damage sites in each target reveals that the Tm0 target has 8 TT sites that are most prone to UV damage. However, the TGG has 7 TT sites and Trandom target has two TT sites, one TC site, and one CT site. CPDs are produced preferentially at TT and TC sites, whereas CC and CT dipyrimidine sites are poorly photoreactive.63 So the Trandom target contains a total of four dipyrimidine sites, in which one of them is less photoreactive than the TT site. Thus, it should exhibit slower photodamage kinetics, consistent with Figure 3. This result confirms that we can get information on the amount of UV damage accumulated in different targets by comparing the damage constant values. Table 3 also shows that

Table 3. Damage Constants of the Different DNA Damage Assay Methods method chMB fluorescence

DNA MB fluorescence

absorbance

damage constant (τ) (min)a τTm0 = 0.62 ± 0.02 (IF,0 = 0.16 ± 0.02 × 106, and A = 3.5 ± 0.1 × 106) τTGG = 0.71 ± 0.03 (IF,0 = 0.13 ± 0.03 × 106, and A = 3.8 ± 0.1 × 106) τTrandom = 1.6 ± 0.05 (IF,0 = 0.05 ± 0.04 × 106, and A = 4.1 ± 0.1 × 106) τTm0 = 3.6 ± 0.2 (IF,0 = 0.12 ± 0.04 × 106, and A = 2.5 ± 0.1 × 106) τTGG = 4.3 ± 0.3 (IF,0 = 0.11 ± 0.05 × 106, and A = 2.4 ± 0.1 × 106) τTrandom = 12 ± 0.7 (IF,0 = 0.30 ± 04 × 106, and A = 2.3 ± 0.1 × 106) τ1 = 6.1 ± 0.07 (Ao= 0.21 ± 0.02, and A1 = 0.52 ± 0.01) τ2 = 92 ± 2.00 (A2 = 0.66 ± 0.02)

Fluorescence and absorption curves fitting parameters are shown for all the oligonucleotide sequences used in this work. The fluorescence is fit to a single exponential function (IF = IF,0 + Ae−t/τ), while the absorption is fit to a double-exponential function (A = A0 + A1et/τ1+ A2et/τ2). A (cps), A1, and A2 are the amplitudes; and IF,0 (cps) and A0 are the y offsets (shown in the brackets); τ, τ1, and τ2 are the time constants obtained from fitting the experimental data points in Figure 3 and Figure S6 of the Supporting Information. a

the damage constants of Tm0, TGG, and Trandom targets detected by the chMB are ∼6 times faster than those detected by the DNA MB. Thus, the chMB has superior selectivity for the detection of UV damage in nucleic acids compared to the DNA MB. UV absorbance measurements as a function of Tm0 target irradiation time (Figure S6 of the Supporting Information) were used to quantify the amount of UV damage. This curve was fit to a double-exponential function and the damage constants are listed in Table 3. The results show that the damage constant of the chMB is 9−10 times faster than the 4325

dx.doi.org/10.1021/ac301669y | Anal. Chem. 2013, 85, 4321−4327

Analytical Chemistry

Article

It is worth mentioning that values recorded in Table 2 for the LOD and LOQ for DNA MB detection of DNA damage are obtained using the standard deviation of the blank measurements and the sensitivity of the method, while the LOD and LOQ will be practically limited to ∼2.6 μM (Figure 4B) due to the unvarying fluorescence signal of the DNA MB at low photoproduct concentrations. From this data, we calculate that the chMB can detect one damage site in the presence of ∼4000 undamaged sites.

fastest absorption damage constant. This confirms that the fluorescent chMB has superior selectivity for detection of UV damage in nucleic acids over the absorption method.25,27 This decay in absorbance with increasing irradiation time (Figure S6 of the Supporting Information) represents the formation of thymine photoproducts containing saturated C5− C6 bonds and can be correlated to the concentration of the damage products. In order to check the sensitivity of the chMB for damage detection, we used the UV absorbance measurements as a function of UV irradiation time to quantify the amount of UV damage and to develop calibration curves of the UV-induced photoproducts detected by the chMB and DNA MB. The procedure and calculation of the photoproducts concentration from the absorbance measurements of the irradiated solutions have been explained previously.64 Figure 4 shows the calibration curve obtained upon plotting the



CONCLUSION In summary, the assay reported here uses RNA nucleotides in the design of a chimeric RNA−DNA MB as a fluorescence sensor for the detection and quantification of nucleic acids, single base mismatches, and UV-induced nucleic acid damage by taking advantage of the increased stability of the RNA− DNA hybrids over the DNA−DNA hybrids. This method proves to have good sensitivity and selectivity for UVC-induced nucleic acid damage and is superior to the DNA MB.



ASSOCIATED CONTENT

* Supporting Information S

This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: (780) 492-9704. Fax: (780) 492-8231. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The authors acknowledge financial support for this work from the Canadian Natural Sciences and Engineering Research Council (NSERC) Discovery Grants-in-aid.

Figure 4. Calibration curve of DNA photodamage formed upon UV irradiation of the Tm0 target for (A) chMB-A and (B) DNA MB-A. The inset shows the fit to the linear portions of the calibration curves. Each data point is an average of three replicate measurements, and the error bars correspond to the standard deviation of the measurements.



REFERENCES

(1) Jordan, D. M.; Ramensky, V. E.; Sunyaev, S. R. Curr. Opin. Str. Biol. 2010, 20, 342−350. (2) Ruzsicska, B. P.; Lemaire, D. G. E. In DNA Photochemistry, CRC Handbook of Organic Photochemistry and Photobiology; Horspool, W. M., Song, P.-S.; Eds., CRC Press: New York, 1995; pp 1289−1317. (3) Marrot, L.; Meunier, J. R. J. Am. Acad. Dermatol. 2008, 58, S139− S148. (4) Lindahl, T. Nature 1993, 362, 709−715. (5) Marnett, L. J.; Burcham, B. C. Chem. Res. Toxicol. 1993, 6, 771− 785. (6) Ames, B. N.; Gold, L. S.; Willett, W. C. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 5258−5265. (7) Sauna, Z. E.; Kimchi-Sarfaty, C. Nat. Rev. Genet. 2011, 12, 683− 691. (8) Rochette, P. J.; Therrien, J. P.; Drouin, R.; Perdiz, D.; Bastien, N.; Drobetsky, E. A.; Sage, E. Nucleic Acids Res. 2003, 31, 2786−2794. (9) Weinfeld, M.; Soderlind, K.-J. M. Biochemistry 1991, 30, 1091− 1097. (10) Bromberg, A.; Jensen, E. C.; Kim, J.; Jung, Y. K.; Mathies, R. A. Anal. Chem. 2012, 84, 963−970. (11) Le, X. C.; Xing, J. Z.; Lee, J.; Leadon, S. A.; Weinfeld, M. Science 1998, 280, 1066−1069. (12) Wang, J.; Rivas, G.; Ozsoz, M.; Grant, D. H.; Cai, X.; Parrado, C. Anal. Chem. 1997, 69, 1457−1460. (13) Kerman, K.; Vestergaard, M.; Nagatani, N.; Takamura, Y.; Tamiya, E. Anal. Chem. 2006, 78, 2182−2189.

chMB-A and DNA MB-A fluorescence intensity as a function of the calculated concentration of the photoproducts. At high damage concentrations, the hybrids formed between the MBs and the damaged strands are completely unstable; the MBs acquire the hairpin structure with quenched fluorescence. Additional formation of photoproducts cannot lead to any more dehybridization, so the fluorescence signal remains constant at a minimum value. Consistent with Figures 2 and 3, the chMB fluorescence decreases immediately as the photoproduct concentration increases, while the DNA MB (Figure 4B) fluorescence signal requires a 2.6 μM photoproduct concentration before decreasing. The DNA MB shows even lower selectivity for the detection of DNA damage compared to detecting mismatches (Figure S4 of the Supporting Information). Table 2 shows the parameters for the quantification of UVinduced DNA damage from Figure 4. The calibration curve for chMB shows a larger linear dynamic range than that of DNA MB (Figure 4). The sensitivity of detection also is larger by a factor of ∼1.5 for the chMB, leading to a lower limit of detection (LOD) and limit of quantification (LOQ) by 5 times. 4326

dx.doi.org/10.1021/ac301669y | Anal. Chem. 2013, 85, 4321−4327

Analytical Chemistry

Article

(14) Kaur, H.; Halliwell, B. Biochem. J. 1996, 318, 21−23. (15) Douki, T.; Court, M.; Sauvaigo, S.; Odin, F.; Cadet, J. J. Biol. Chem. 2000, 275, 11678−11685. (16) Douki, T.; Cadet, J. Biochemistry 2001, 40, 2495−2501. (17) Ravanat, J.-L.; Douki, T.; Cadet, J. J. Photochem. Photobiol., B 2001, 63, 88−102. (18) Totsingan, F.; Rossi, S.; Corradini, R.; Tedeschi, T.; Sforza, S.; Juris, A.; Scaravelli, E.; Marchelli, R. Org. Biomol. Chem. 2008, 6, 1232−1237. (19) Kumar, A.; Tyagi, M. B.; Jha, P. N. Biochem. Biophys. Res. Commun. 2004, 318, 1025−1030. (20) Trevithick-Sutton, C. C.; Mikelsons, L.; Filippenko, V.; Scaiano, J. C. Photochem. Photobiol. 2007, 83, 556−562. (21) Xing, J. Z.; Lee, J.; Leadon, S. A.; Weinfeld, M.; Le, X. C. Methods 2000, 22, 157−163. (22) Tyagi, S.; Kramer, F. R. Nat. Biotechnol. 1996, 14, 303−308. (23) Marras, S. A. E.; Kramer, F. R.; Tyagi, S. Genet. Anal.: Biomol. Eng. 1999, 14, 151−156. (24) Kostrikis, L. G.; Tyagi, S.; Mhlanga, M. M.; Ho, D. D.; Kramer, F. R. Science 1998, 279, 1228−1229. (25) Yarasi, S.; McConachie, C.; Loppnow, G. R. Photochem. Photobiol. 2005, 81, 467−473. (26) El-Yazbi, A.; Loppnow, G. R. Can. J. Chem. 2011, 89, 402−408. (27) Oladepo, S. A.; Loppnow, G. R. Anal. Bioanal. Chem. 2010, 397, 2949−2957. (28) Marras, S. A. E.; Kramer, R. F.; Tyagi, S. Nucleic Acids Res. 2002, 30, 122. (29) Martí, A. A.; Li, X.; Jockusch, S.; Stevens, N.; Li, Z.; Raveendra, B.; Kalachikov, S.; Morozova, I.; Russo, J. J.; Akins, D. L.; Ju, J.; Turro, N. J. Tetrahedron 2007, 63, 3591−3600. (30) Liu, C.; Wang, L.; Jiang, W. Talanta 2001, 81, 597−601. (31) Heller, M. J.; Morrison, L. E.; Prevatt, W. D.; Akin, C. Eur. Pat. Appl. 1983, 70, 685. (32) Morrison, L. E.; Stols, L. M. Biochemistry 1993, 32, 3095−3104. (33) Mergny, J.-L.; Boutorine, A. S.; Garestier, T.; Belloc, F.; Rougée, M.; Bulychev, N. V.; Koshkin, A. A.; Bourson, J.; Lebedev, A. V.; Valeur, B.; Thuong, N. T.; Hèléne, C. Nucleic Acids Res. 1994, 22, 920−928. (34) Marras, S. A. E.; Tyagi, S.; Kramer, F. R. Clin. Chim. Acta 2006, 363, 48−60. (35) Peng, X. H.; Cao, Z. H.; Xia, J. T.; Carlson, G. W.; Lewis, M. M.; Wood, W. C.; Yang, L. Cancer Res. 2005, 65, 1909−1917. (36) Sei-Iida, Y.; Koshimoto, H.; Kondo, S.; Tsuji, A. Nucleic Acids Res. 2000, 28, e59. (37) Sandhya, S.; Chen, W.; Mulchandani, A. Anal. Chim. Acta 2008, 614, 208−212. (38) Yang, C. J.; Medley, C. D.; Tan, W. H. Curr. Pharm. Biotechnol. 2005, 6, 445−452. (39) Tang, H. X.; Yang, X. H.; Wang, K. M.; Tan, W. H.; Li, W. Chin. Sci. Bull. 2009, 54, 1507−1514. (40) Mhlanga, M. M.; Vargas, D. Y.; Fung, C. W.; Kramer, F. R.; Tyagi, S. Nucleic Acids Res. 2005, 33, 1902−1912. (41) Molenaar, C.; Marras, S. A.; Slats, J. C. M; Truffert, J. C.; Lemaitre, M.; Raap, A. K.; Dirks, R. W.; Tanke, H. J. Nucleic Acids Res. 2001, 29, 89. (42) Tsourkas, A.; Behlke, M. A.; Bao, G. Nucleic Acids Res. 2002, 30, 5168−5174. (43) Kuhn, H.; Demidov, V. V.; Gildea, B. D.; Fiandaca, M. J.; Coull, J. C.; Frank-Kamenetskii, M. D. Antisense Nucleic Acid Drug Dev. 2001, 11, 265−270. (44) Petersen, K.; Vogel, U.; Rockenbauer, E.; Nielsen, K. V.; Kolvraa, S.; Bolund, L.; Nexo, B. Mol. Cell. Probes 2004, 18, 117−122. (45) Iwase, R.; Murakami, A. J. Synth. Org. Chem., Jpn. 2002, 60, 1179−1189. (46) Wu, J. C.; Zou, Y .; Li, C. Y.; Sicking, W.; Piantanida, I.; Yi, T.; Schmuck, C. J. Am. Chem. Soc. 2012, 134, 1958−1961. (47) Yang, C. J.; Wang, L.; Wu, Y.; Kim, Y.; Medley, C. D.; Lin, H.; Tan, W. H. Nucleic Acids Res. 2007, 35, 4030−4041.

(48) Wang, L.; Yang, C. J.; Medley, C. D.; Benner, S. A.; Tan, W. H. J. Am. Chem. Soc. 2005, 127, 15664−15665. (49) Perlette, J.; Tan, W. H. Anal. Chem. 2001, 73, 5544−5550. (50) Braasch, D. A.; Corey, D. R. Chem. Biol. 2001, 8, 1−7. (51) Egholm, M.; Nielsen, P. E.; Buchardt, O.; Berg, R. H. J. Am. Chem. Soc. 1992, 114, 9677−9678. (52) Dueholm, K. L.; Egholm, M.; Behrens, C.; Christensen, L.; Hansen, H. F.; Vulpius, T.; Petersen, K. H.; Berg, R. H.; Nielsen, P. E. J. Org. Chem. 1994, 59, 5767−5773. (53) Bergmann, F.; Bannwarth, W.; Tam, S. Tetrahedron Lett. 1995, 36, 6823−6826. (54) Gildea, B. D.; Casey, S.; MacNeill, J.; Perry-O′Keefe, H.; Sorensen, D.; Coull, J. M. Tetrahedron Lett. 1998, 39, 7255−7258. (55) Shimayama, T.; Nishikawa, F.; Nishikawa, S.; Taira, K. Nucleic Acids Res. 1993, 21, 2605−2611. (56) Rice, M. C.; Czymmek, K.; Kmiec, E. B. Nat. Biotechnol. 2001, 19, 321−326. (57) Gamper, H. B.; Parekh, H.; Rice, M. C.; Bruner, M.; Youkey, H.; Kmiec, E. B. Nucleic Acids Res. 2000, 28, 4332−4339. (58) Coljee, V. W.; Murray, H. L.; Donahue, W. F.; Jarrell, K. A. Nat. Biotechnol. 2000, 18, 789−791. (59) Rizzo, J.; Gifford, L. K.; Zhang, X.; Gewirtz, A. M.; Lu, P. Mol. Cell. Probes 2002, 16, 277−283. (60) Elbashir, S. M.; Harborth, J.; Lendeckel, W.; Yalcin, A.; Weber, K.; Tuschl, T. Nature 2001, 411, 494−498. (61) Nakano, S.; Kanzaki, T.; Sugimoto, N. J. Am. Chem. Soc. 2004, 126, 1088−1095. (62) Weinfeld, M.; Liuzzi, M.; Paterson, M. C. Biol. Chem. 1989, 264, 6364−6370. (63) Douki, T.; Cadet, J. Biochemistry 2001, 40, 2495−2501. (64) El-Yazbi, A. F.; Loppnow, G. R. Anal. Chim. Acta 2012, 726, 44−49.

4327

dx.doi.org/10.1021/ac301669y | Anal. Chem. 2013, 85, 4321−4327