Analytical Biochemistry 448 (2014) 58–64
Contents lists available at ScienceDirect
Analytical Biochemistry journal homepage: www.elsevier.com/locate/yabio
Kras gene codon 12 mutation detection enabled by gold nanoparticles conducted in a nanobioarray chip Abootaleb Sedighi, Paul C.H. Li ⇑ Department of Chemistry, Simon Fraser University, Burnaby, British Columbia V5A 1S6, Canada
a r t i c l e
i n f o
Article history: Received 20 September 2013 Received in revised form 13 November 2013 Accepted 16 November 2013 Available online 26 November 2013 Keywords: Single nucleotide polymorphism Kras gene Gold nanoparticle Target DNA Microarray Nanobioarray chip
a b s t r a c t This study employs a nanobioarray (NBA) chip for multiple biodetection of single base pair mutations at the Kras gene codon 12. To distinguish between the mutant and wild-type target DNAs, current bioarray methods use high-temperature hybridization of the targets to the allele-specific probes. However, these techniques need prior temperature optimization and become harder to implement in the case of the detection of multiple mutations. We aimed to detect these mutations at a single temperature (room temperature), enabled by the use of gold nanoparticles (AuNPs) on the bioarray created within nanofluidic channels. In this method, a low amount of target oligonucleotides (5 fmol) and polymerase chain reaction (PCR) products (300 pg) were first loaded on the AuNP surface, and then these AuNP-bound targets were introduced into the channels of a polydimethylsiloxane (PDMS) glass chip. The targets hybridized to their complementary probes at the intersection of the target channels to the pre-printed oligonucleotide probe lines on the glass surface, creating a bioarray. Using this technique, fast and high-throughput multiple discrimination of the Kras gene codon 12 were achieved at room temperature using the NBA chip, and the specificity of the method was proved to be as high as that with the temperature stringency method. Ó 2013 Elsevier Inc. All rights reserved.
Single nucleotide polymorphisms (SNPs)1 are the most abundant genetic variation and are responsible for many genetic disorders and for certain cancers such as retinoblastoma, colorectal carcinoma, and adenocarcinoma of the pancreas [1,2]. SNP variations in the Kras gene codon 12 have been frequently observed in genetic disorders, and their detection is critical for the selection of the appropriate type of treatment [3,4]. DNA sequencing has been the most widely used method for detection of SNPs; however, the conventional separation-based sequencing method requires large reagent consumption and has a low throughput [5]. Although these issues have been effectively addressed in various newly emerged next-generation sequencing (NGS) methods, they still have technical limitations such as labor-intensive sample preparation and data interpretation. Real-time polymerase chain reaction (PCR) is another popular SNP detection approach, but it has nonlinear amplification of target DNAs and is limited in multiplex analysis [6]. Due to their high-throughput nature, DNA bioarrays have been used in a variety of applications such as gene expression profiling, ⇑ Corresponding author. Fax: +1 778 782 3765. E-mail address: [email protected] (P.C.H. Li). Abbreviations used: SNP, single nucleotide polymorphism; NGS, next-generation sequencing; PCR, polymerase chain reaction; NBA, nanobioarray; AuNP, gold nanoparticle; ssDNA, single-stranded DNA; dsDNA, double-stranded DNA; PM, perfect match; MM, mismatch; PDMS, polydimethylsiloxane; SDS, sodium dodecyl sulfate; APTES, 3-aminopropyltriethoxysilane; PBS, phosphate-buffered saline; DTA, DNA-toAuNP. 1
0003-2697/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ab.2013.11.019
comparative genomic hybridization, genotyping, and nucleic acid diagnostic applications as well as protein–DNA interaction studies [7,8]. Although recent advances in the NGS technologies have affected the popularity of DNA bioarrays in some clinical diagnostic applications, the convenience and efficiency of low-density arrays give them great advantages over the complex and still expensive NGS assays, especially when particular sites of the genome, instead of the whole genome, are targeted [9]. Here, we combined the lowdensity DNA bioarray with nanofluidics to produce the nanobioarray (NBA) chip in order to make the SNP assay fast and simple with low sample and reagent consumption. But the SNP assays using this method remain an issue because the hybridization temperature of the assays needs to be optimized to achieve single base pair discrimination [10–12]. Gold nanoparticles (AuNPs) have been used extensively in biosensors [13–18]. In 2004, Li and Rothberg [19] reported that unmodified single-stranded DNA (ssDNA), but not double-stranded DNA (dsDNA), could be adsorbed on the citrate-capped AuNP surfaces. This selective adsorption of ssDNA on AuNPs has been used as the basis of several DNA analysis devices [20–22]. This phenomenon was also used by our group, but in a novel manner, to enable single base pair discrimination of fungal pathogen PCR amplicons at room temperature [23], as opposed to the high-temperature method [24]. In this technique, target strands were first loaded on AuNPs and then were introduced to the oligonucleotide probes immobilized on the surface of a glass chip. Our recent kinetic and
Kras gene codon 12 mutation detection / A. Sedighi, P.C.H. Li / Anal. Biochem. 448 (2014) 58–64
thermodynamic studies showed that nonspecific binding of target DNA bases with the AuNP surface alters the kinetic pathways [25] altered hybridization and dehybridization processes of the target strands to the surface-immobilized probes lead to the unexpectedly enhanced discrimination power of AuNP-loaded targets (AuNP–targets), in contrast to the random-coiled targets (free– targets). Although the earlier work by our group provided an interesting finding for microarray research [23], only a single probe was used to distinguish two targets. In the common format of microarray experiments, discrimination of the target molecule between the perfect match (PM) probe and mismatch (MM) probe is normally used as the basis of SNP detection [24]. In the current study, we used four probes (one PM probe and three MM probes) in the NBA chip for multiple analyses of four SNPs at the codon 12 site of the Kras gene, which is clinically relevant in colorectal carcinoma and adenocarcinoma of the pancreas [26]. Here, we demonstrate that AuNP-loaded oligonucleotides or PCR amplicons discriminate between PM probes and MM probes immobilized on the NBA chip channel surface (Fig. 1). In the PM region, the targets are unloaded from the AuNPs and hybridize to the immobilized probes. In the MM region, the targets cannot hybridize to the immobilized probes and remain bound on the AuNPs and are washed downstream. We believe that this method, with further developments and evaluations with various sequences, can potentially replace the conventional temperature stringency method, which requires a tedious temperature optimization procedure and probe reductancy. In addition, our method allows multiplex analysis of SNPs by hybridization at the same temperature and offers the convenience of using the widely used polydimethylsiloxane (PDMS)-based chips to conduct DNA hybridizations because the room temperature operation alleviates the problem of evaporation through the gas-permeable polymer at high temperatures. Materials and methods Materials AuNPs with 5-, 10-, and 20-nm diameters (stabilized with citrate and tannic acid) were purchased from Sigma Life Science, and 12-nm AuNPs (stabilized with citrate) were obtained from NanoComposix (San Diego, CA, USA). Sodium dodecyl sulfate (SDS), 3-aminopropyltriethoxysilane (APTES), 25% glutaraldehyde, and Triton X-100 were purchased from Sigma–Aldrich. Negative
Fig.1. Schematic diagram of single base pair discrimination of AuNP-conjugated targets in PDMS glass microfluidic channel (A), channel intersection with the perfect match (PM) probe line (B), and channel intersection with the mismatch (MM) probe line (C). Both probe spots are 200 200 lm. The channel is not drawn to scale.
59
photoresist (SU-8 50) and its developer were purchased from MicroChem (Newton, MA, USA). The glass microscope slides of 3 2 inches were obtained from Fisher Scientific (Ottawa, Ontario, Canada). All oligonucleotides and the primers were synthesized and modified by Integrated DNA Technologies (Coralville, IA, USA). The 60-mer target oligonucleotides with the sequence of Kras gene were modified by a biotin molecule at the 50 end (Table 1). The sequences were designed in such a way that the codon 12 sequence was located at the center of the oligonucleotides. Four different 20mer oligonucleotide probes, each of them complementary to one of the targets (60-mer), were designed. The probes were modified with an amine group and a C12 spacer at the 50 end. All oligonucleotides are listed in Table 1. NBA chip fabrication Fabrication of a PDMS slab (2 2 inches) with 16 parallel channels has been described elsewhere [27]. The width of straight channels was 200 lm, and the height was 35 lm. A flat-end syringe needle (15 gauge) was used to punch out the solution reservoirs (1.5 mm in diameter) at both ends of the channels on the 2-mmthick PDMS. The glass slides were aldehyde functionalized using an established procedure [28]. The PDMS slab was sealed reversibly to the modified glass slide to create the NBA chip. Probe immobilization The probes were printed using a procedure similar to that described in previous reports [26,28]. Briefly, 0.5 ll of 25 lM probe solution (in 1.0 M NaCl + 0.15 M NaHCO3) was added to the inlet reservoir of each of the 16 horizontal channels of the NBA chip, and the solution was then filled in the channel using suction applied at the outlet reservoir. After 30 min of incubation at room temperature, the probe solution was pumped out of the channel. Following washing of the channel by the bicarbonate washing solution (0.15 M NaHCO3 + 0.2% SDS), the PDMS slab was peeled off, leaving behind 16 probe lines on the glass slide. The glass slide was rinsed by water and dried. Amplification of genomic DNA Two genomic DNA samples, containing different allele compositions of the Kras gene codon 12, were purchased from Horizon Diagnostics (Cambridge, UK). One of the samples contained the pure wild-type allele, and the other contained a mixture of 50% wild type and 50% G12D mutant (the genotypes were characterized by the vendor using Sanger sequencing). A pair of specific primers (forward: 50 -biotin-TGACTGAATATAAACTTGTGGTAGTTGGAG-30 ; reverse: 50 -ATGATTCTGAATTAGCTGTATCGTCAAGGC-30 ) was used in order to obtain the 80-bp PCR products. Asymmetric PCR was performed by varying the concentration of the reverse primer (0.05, 0.1, 0.2, 0.3, 0.4, and 0.5 lM) in the presence of the forward primer with a constant concentration of 0.5 lM. A custom PCR protocol on a Cetus thermocycler (PerkinElmer) was used for the amplification. The thermocycling was initiated by 3 min of denaturation, followed by 30 thermal cycles of 95 °C for 40 s (denaturation), 55 °C for 30 s (annealing), and 72 °C for 60 s (extension), and then was terminated by 10 min of final extension at 72 °C. The amplified products were purified using a nucleotide removal kit (Qiagen). DNA hybridization A second PDMS slab with 16 channels was sealed against the glass slide pre-printed with the probe lines. The straight channels
60
Kras gene codon 12 mutation detection / A. Sedighi, P.C.H. Li / Anal. Biochem. 448 (2014) 58–64
Table 1 Sequences of probes and target oligonucleotides. Targets
Probes
WT (W60) G12Aa (A60) G12Db (D60) G12Vc (V60) WT (W) G12A (A) G12D (D) G12V (V)
50 -/Biotin/GAA TAT AAA CT T GTG GTA GTT GGA 50 -/Biotin/GAA TAT AAA CT T GTG GTA GTT GGA 50 -/Biotin/GAA TAT AAA CT T GTG GTA GTT GGA 50 -/Biotin/GAA TAT AAA CT T GTG GTA GTT GGA 50 -/C12amine/CC TAC GCC ACC AGC TCC AAC-30 50 -/C12amine/CC TAC GCC AGC AGC TCC AAC-30 50 -/C12amine/CC TAC GCC ATC AGC TCC AAC-30 50 -/C12amine/CC TAC GCC AAC AGC TCC AAC-30
GCT GCT GCT GCT
GGT GGC GTA GGC AAG AGT GCC TTG ACG ATA CAG-30 GCT GGC GTA GGC AAG AGT GCC TTG ACG ATA CAG-30 GAT GGC GTA GGC AAG AGT GCC TTG ACG ATA CAG-30 GTT GGC GTA GGC AAG AGT GCC TTG ACG ATA CAG-30
Note: WT, wild type. a G12A represents Gly12Ala amino acid substitution. b G12D represents Gly12Asp amino acid substitution. c G12V represents Gly12Val amino acid substitution.
were orthogonal to the probe lines on the glass slide surface. Before hybridization to the probe, target oligonucleotides were conjugated noncovalently to the surface of AuNPs. To achieve this, AuNP solution was added to the target solution and the mix was incubated at 95 °C for 15 min. Free–target solutions (control) were not mixed with the AuNP solutions. The target solutions (free and AuNP-conjugated) were then diluted in the hybridization buffer (1 SSC [sodium chloride and sodium citrate] + 0.2% SDS) to a final concentration of 10 nM for oligonucleotides and 0.6 ng/ll for PCR products. The AuNP–target solution (0.5 ll) was added to the inlet reservoir, filled in the channel using vacuum suction, and then incubated for 20 min at 22 °C (unless noted otherwise). The hybridization of the targets to the complementary probes occurred at the intersection of target channels with the probe lines (resulting in the hybridization detection area of 200 200 lm2). After incubation, the target solutions were pumped out from the channels. High-temperature experiments (55 °C) were achieved using a Peltier device. Immediately after hybridization, the target solutions were pumped out and the channels were washed by 2 ll of the hybridization buffer. To detect the duplex oligonucleotides formed at the hybridization patches, streptavidin–Cy5 solution (50 lg/ml in 1 PBS [phosphate-buffered saline]) was added to the channels. After incubation for 15 min, the channel was rinsed using a wash solution (1 PBS and 0.1% Tween 20), and then the PDMS slab was peeled off from the glass slide. Detection Following rinsing and drying, the glass slide was scanned on a confocal laser fluorescent scanner (Typhoon 9410, GE Healthcare) at a 10-lm resolution, as described previously [29,30]. The excitation and emission wavelengths were 633 and 670 nm, respectively. The photomultiplier tube voltage was set to 600 V. The scanned image was analyzed by ImageQuant 5.2 software. The average fluorescent signals were measured in relative fluorescent units. Results and discussion The discrimination between the wild-type and G12D mutant, which is the most frequent one in the Kras gene codon 12 mutations in adenocarcinomas [31], was chosen for the current study. The scanned image in Fig. 2A shows the spots resulting from hybridization of two 60-mer targets, W60 and D60, to their corresponding PM and MM probes (W and D), respectively, performed in duplicate. The histogram (Fig. 2B) was created based on measured signal intensities of the spots along the vertical target channel line. The discrimination ratio, calculated based on the ratio of the mean signal intensity of PM to that of MM, is shown above the bars in Fig. 2B. We observe that when the targets are free, there is no significant difference between the signal
Fig.2. (A) Scanned image with the spots resulting from hybridization of 10 nM 60mer targets (W60 and D60) with their corresponding PM and MM probes (W and D) at 22 °C for 20 min in the NBA chip. The targets were either free (free–targets) or previously conjugated with 10-nm AuNPs (AuNP–targets). The boxed regions are the expected true positive binding regions. (B) Histogram showing the signal intensities of the spots obtained along the vertical target channels, with the true positive binding signals represented by the hatched bars. Error bars show the standard deviations of two measurements. The number above each column shows the discrimination ratio, which is determined by dividing the intensity of the PM spots (W60–W and D60–D) by that of the MM spots (W60–D and D60–W).
intensities of the PM and MM spots, or the discrimination ratio is close to 1.0, showing why a high temperature such as 55 °C would have been required to achieve sufficient discrimination [32,33]. An improvement in the discrimination ratio is observed in the case of AuNP–targets, where the PM/MM discrimination ratios are 1.6 ± 0.2 and 1.7 ± 0.2 for W60 and D60 targets, respectively.
Kras gene codon 12 mutation detection / A. Sedighi, P.C.H. Li / Anal. Biochem. 448 (2014) 58–64
61
Fig.3. Scanned images and the corresponding histograms (below each image) resulting from free–target hybridization at 22 °C (A), AuNP–target hybridization at 22 °C (B), and free–target hybridization at 55 °C (C). The 60-mer target DNAs were conjugated with 5-nm AuNPs at a DTA of 1:1. Error bars show the standard deviations of three measurements. The number above each column shows the discrimination ratio, which is determined by dividing the intensity of the PM spots by the average of the three MM spots. For other conditions, see Fig. 2.
Fig.4. Original scanned image showing half of the 16 16 NBA chip (A), the inset inside the blue box (B), and the resulting histogram (C) of free–target (60-mer) and 5-nm AuNP–target based on 60 min of hybridization at 22 °C. Error bars are standard deviations from three replicates. For other conditions, see Fig. 2.
62
Kras gene codon 12 mutation detection / A. Sedighi, P.C.H. Li / Anal. Biochem. 448 (2014) 58–64
Method optimization Because the improvement in the discrimination ratio using the AuNP method at room temperature is marginal, this method needs to be further optimized. To improve the discrimination ratio, we use a single factor analysis approach to investigate potentially effective factors such as AuNP size and DNA-to-AuNP ratio (DTA). The AuNP size was the first factor whose potential impact on the discrimination ratios was tested. This was performed by loading the targets on AuNPs of various sizes (5, 10, 12, and 20 nm) and then comparing the discrimination ratios with the ones obtained from free–targets. As shown in Fig. S2 of the online supplementary material, there is not much improvement in discrimination ratio as the AuNP sizes get smaller from 20 to 10 nm. But a great increase in the ratios (2.2 or 2.7) is obvious at 5 nm AuNPs. This higher discrimination can be explained by the enhanced dehybridization rate constants in the presence of small AuNPs (5 nm) due to the greater curvature of the smaller particles [25]. As we know, dehybridization starts from binding of the AuNPs to the ssDNA parts in the duplex, but this was hindered by the repulsion between the negatively charged surface of large AuNPs and the neighboring folded parts of the duplex, where the negative charge density is twice as much as the ssDNA parts. Owing to the higher degree of curvature in small AuNPs (5 nm), this repulsion is less of an issue, leading to the enhanced dehybridization when 5-nm particles are used. The DTA was another factor whose influence on the discrimination ratios was investigated. As concluded from Fig. S2, as DTA decreases from 3 to 1, meaning that the number of AuNPs per target strand increases from 0.33 to 1, the discrimination ratio increases (up to 2.7). Therefore, DTA of 1 was chosen as the optimum, which corroborated with our previous results [23]. However, DTAs lower than 1 could not be applied because AuNPs aggregate in those solutions.
It is noted that although the discrimination ratio is improved, the signal intensity decreases in both the AuNP method and the high-temperature method. These lower signal intensities result from the lower hybridization rate constants of AuNP–targets as compared with free–targets [25]. This issue can be resolved by using long hybridization times given that 12 h is usually needed to achieve high sensitivities and specificities in conventional nucleic acid bioarray experiments [34]. But we did not need 12 h; we hybridized for 60 min instead of 20 min, with the results shown in Fig. 4. Although the signal intensities are generally higher, more important, the PM spots of AuNP–targets show comparable signal intensities to the free–target spots. Furthermore, the greater decrease in the intensities of the MM spots resulted in much higher discrimination ratios ranging from 6.8 to 19.7. This additional incubation time dramatically increases the discrimination ratios, whereas the method completed in 1 h is still faster than the conventional method [32,33]. Detection of genomic DNA Having demonstrated a satisfactory performance on the SNP detection of the single-stranded oligonucleotide targets, the AuNP-assisted technique was evaluated using the PCR amplicons
Multiple mutation detection Following the optimization of AuNP-enabled method using wild type and one of its mutant (G12D) of Kras sequences, the method was fully evaluated using all three codon 12 mutants plus the wild type. Fig. 3A and B show the image and the corresponding histogram of room temperature hybridization of free–targets and AuNP-conjugated targets. The number above each group of PM or MM column is the discrimination ratio of each target, which is the signal of the PM spot divided by the average of the signals of the other three MM spots (Eq. (1)):
Discrimination ratio ¼
PM 1=3ðMM1 þ MM2 þ MM3 Þ
ð1Þ
Whereas the discrimination of the AuNP D60 target remains low (2.4), the ratios resulting from the AuNP–targets range from 5.9 for the V60 target to 7.1 for the A60 target. As a comparison, a similar experiment using the conventional temperature stringency method at 55 °C was performed, and this hybridization temperature is commonly used for Kras sequence discrimination in bioarray analysis [32]. As shown in Fig. 3C, free–targets at 55 °C created spots with discrimination ratios ranging from 1.4 to 6.6. These discrimination ratios are also in the same range as those obtained previously using bioarray approaches [32,33]. The hightemperature values are comparable to, if not worse than, the ones resulting from AuNP–targets at room temperature hybridization (ratios of 2.4–7.1). These results demonstrate that using the AuNP-enabled method, the targets can effectively discriminate between their PM probes and the MM probes at room temperature at a short hybridization time of 20 min.
Fig.5. Original scanned image and the corresponding histogram resulting from hybridization of PCR products amplified from genomic DNAs with 100% wild-type Kras alleles (W0 ) at 22 °C for 60 min on the NBA chip. Asymmetric PCR was performed using different concentrations of the reverse primer (0.05, 0.1, 0.2, 0.3, 0.4, and 0.5 lM) with a constant forward primer concentration of 0.5 lM. All of the targets were previously loaded on the 5-nm AuNPs. In the histogram, the column bars show the averages of signal intensities of the spots, measured at the intersection of horizontal probe lines and the vertical target lines, and the true positive binding signals of W0 are represented by the hatched bars. Error bars show the standard deviations of three measurements. The number above each column shows the discrimination ratio, which is determined by dividing the intensity of the PM spots (W0 to W) by that of the MM spots (W0 to A, V, and D).
Kras gene codon 12 mutation detection / A. Sedighi, P.C.H. Li / Anal. Biochem. 448 (2014) 58–64
of genomic DNA (double-stranded). Because the PCR amplicons are longer (80 bp) and are double-stranded, as compared with oligonucleotides, we further optimized the hybridization conditions. First, although DTAs lower than 1 cause AuNP aggregation in oligonucleotides, AuNP PCR products with DTAs down to 0.6 can be obtained without AuNP aggregation (data not shown). The higher stability of AuNPs in PCR amplicons, as compared with oligonucleotides, is probably obtained because the PCR amplicons (80 bp) are longer than the target oligonucleotides (60-mer) used here. Second, we used asymmetric PCR in order to reduce the influence of complementary strands, which will not only compete with the binding target strands for the immobilized probes but also occupy the AuNPs with nonbinding strands. In asymmetric PCR, decreasing the concentration of the unlabeled reverse primer leads to preferential amplification of the labeled strand at higher concentrations [34]. We compared the hybridization signals of the 100% W amplicons using different concentrations of reverse primer with a constant forward primer concentration of 0.5 lM. As shown in Fig. 5, no significant discrimination resulted if the concentration of the reverse primer was equal to or higher than 0.3 lM. We believe that the higher concentration of the reverse primer resulted in a higher concentration of complementary strands, inducing the unloading of the target strands from the AuNPs by hybridization. These unloaded target strands will behave like the free–targets; therefore, the overall specificity will be affected. A sharp rise in the discrimination ratio is obvious on reducing the concentration of reverse primer from 0.2 to 0.05 lM (10% of concentration of forward primer). The performance of the AuNP-assisted technique in the detection of the PCR amplicons from the genomic samples containing mixed alleles was also investigated. A genomic sample containing a mixture of wild type and G12D (50% each), together with another sample containing 100% wild type, was amplified and detected on
63
the NBA chip. As shown in Fig. 6, the mixed allele sample showed a distinct positive hybridization signal on probes W and D in the presence of other probes. The discrimination ratio is satisfactory, and the signals are as high as those in the free–targets. Considering that the targets W and D show lower discrimination ratios than the targets A and V (Fig. 3 and 4), we believe that the PCR amplicons from A and V mutants will have higher performance. So, this experiment demonstrates the flexibility and power of the AuNP-assisted technique in SNP detection of genomic samples. Conclusion In this study, we have developed a fast and multiplex method for multiple SNP detection at room temperature. The AuNP-assisted method enables the DNA bioarrays to perform detection of the wild type and three mutants of the Kras gene codon 12. The method proved to be able to discriminate between the wild-type sequence and different oligonucleotide mutants as effectively as the high-temperature method. The various factors were investigated and optimized for effective single base pair discriminations. It was confirmed that a higher discrimination ratio is achievable when a longer hybridization time is used without loss of signal intensity. The method was tested on PCR amplicons of the wildtype allele (W) and of one mutant allele (D). Further developments are required in order to extend the method to detect other clinically relevant Kras gene mutations, such as those at codons 13 and 61, as well as the mutations in the other sequences. Acknowledgments We gratefully acknowledge financial support from the Natural Sciences and Engineering Research Council of Canada for a Discovery Grant. We thank John Brunstein (BC Children’s Hospital) for useful suggestions in detection of the Kras mutation, Dipankar Sen (Department of Molecular Biology and Biochemistry, Simon Fraser University) for useful advice and technical support on PCR amplification, and Byron Gates and Idah Pekcevic (Department of Chemistry, Simon Fraser University) for AuNP characterization. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ab.2013.11.019. References
Fig.6. Original scanned image (A) and the corresponding histogram (B) resulting from hybridization of PCR products amplified from genomic DNAs with different Kras allele compositions (one sample is 100% wild type, and the other is 50% wild type plus 50% G12D mutant). The targets are either free or loaded 5-nm AuNPs. Error bars are from three replicates. For other conditions, see Fig. 5.
[1] F.S. Collins, Medical and societal consequences of the Human Genome Project, N. Engl. J. Med. 341 (1999) 28–37. [2] S. White, M. Kalf, Q. Liu, M. Villerius, D. Engelsma, M. Kriek, E. Vollebregt, B. Bakker, G.J.B. van Ommen, M.H. Breuning, J.T. den Dunnen, Comprehensive detection of genomic duplications and deletions in the DMD gene, by use of multiplex amplifiable probe hybridization, Am. J. Hum. Genet. 71 (2002) 365– 374. [3] S. Ye, S. Dhillon, X. Ke, A.S. Collins, Y.N.M. Day, An efficient procedure for genotyping single nucleotide polymorphisms, Nucleic Acids Res. 29 (2001) 7773–7782. [4] C.J. Allegra, J.M. Jessup, M.R. Somerfield, S.R. Hamilton, E.H. Hammond, D.F. Hayes, P.K. McAllister, R.F. Morton, R.L. Schilsky, American Society of Clinical Oncology provisional clinical opinion: testing for KRAS gene mutations in patients with metastatic colorectal carcinoma to predict response to antiepidermal growth factor receptor monoclonal antibody therapy, J. Clin. Oncol. 27 (2009) 2091–2096. [5] K. Sato, M. Onoguchi, Y. Sato, K. Hosokawa, M. Maeda, Non-cross-linking gold nanoparticle aggregation for sensitive detection of single-nucleotide polymorphisms: optimization of the particle diameter, Anal. Biochem. 350 (2006) 162–164. [6] N.N. Iscove, M. Barbara, M. Gu, M. Gibson, C. Modi, N. Winegarden, Representation is faithfully preserved in global cDNA amplified exponentially from sub-picogram quantities of mRNA, Nat. Biotechnol. 20 (2002) 940–943.
64
Kras gene codon 12 mutation detection / A. Sedighi, P.C.H. Li / Anal. Biochem. 448 (2014) 58–64
[7] L. Wang, P.C.H. Li, Microfluidic DNA microarray analysis: a review, Anal. Chim. Acta 687 (2011) 12–27. [8] S. Russell, L.A. Meadows, Microarray Technology in Practice, Academic Press/ Elsevier, Amsterdam, 2009. [9] A. Sedighi, P.C.H. Li, Challenges and future trends in DNA microarray analysis, in: C. Simó, A. Cifuentes, V. García-Cañas (Eds.), Fundamentals of Advanced Omic Technologies: From Genes to Metabolites, Elsevier Science, San Diego, 2013. [10] C. Situma, Y. Wang, M. Hupert, F. Barany, R.L. McCarley, S.A. Soper, Fabrication of DNA microarrays onto poly(methyl methacrylate) with ultraviolet patterning and microfluidics for the detection of low-abundant point mutations, Anal. Biochem. 340 (2005) 123–135. [11] A. Sedighi, L. Wand, P.C.H. Li, 2D nanofluidic bioarray for nucleic acid analysis, in: K. Iniewski, S. Selimovic (Eds.), Nanopatterning and Nanoscale Devices for Biological Applications, Taylor & Francis/CRC Press, Boca Raton, FL, 2013. [12] O.Y.F. Henry, C.K. O_Sullivan, Rapid DNA hybridization in microfluidics, Trends Anal. Chem. 33 (2012) 9–22. [13] S. Link, M.A. El-Sayed, Spectral properties and relaxation dynamics of surface plasmon electronic oscillations in gold and silver nanodots and nanorods, J. Phys. Chem. B 103 (1999) 8410–8426. [14] R. Elghanian, J.J. Storhoff, R.C. Mucic, R.L. Letsinger, C.A. Mirkin, Selective colorimetric detection of polynucleotides based on the distance-dependent optical properties of gold nanoparticles, Science 277 (1997) 1078–1081. [15] K. Sato, K. Hosokawa, M. Maeda, Rapid aggregation of gold nanoparticles induced by non-cross-linking DNA hybridization, J. Am. Chem. Soc. 125 (2003) 8102–8103. [16] K. Sato, K. Hosokawa, M. Maeda, Non-cross-linking gold nanoparticle aggregation as a detection method for single-base substitutions, Nucleic Acids Res. 33 (2005) e4. [17] C. Wang, Y. Chen, T. Wang, Z. Ma, Z. Su, Biorecognition-driven self-assembly of gold nanorods: a rapid and sensitive approach toward antibody sensing, Chem. Mater. 19 (2007) 5809–5811. [18] C.C. Huang, Y.F. Huang, Z. Cao, W. Tan, H.T. Chang, Aptamer-modified gold nanoparticles for colorimetric determination of platelet-derived growth factors and their receptors, Anal. Chem. 77 (2005) 5735–5741. [19] H.X. Li, L. Rothberg, Colorimetric detection of DNA sequences based on electrostatic interactions with unmodified gold nanoparticles, Proc. Natl. Acad. Sci. USA 101 (2004) 14036–14039. [20] H.X. Li, L.J. Rothberg, Label-free colorimetric detection of specific sequences in genomic DNA amplified by the polymerase chain reaction, J. Am. Chem. Soc. 126 (2004) 10958–10961.
[21] H. Lee, S.W. Joob, S.Y. Lee, C.H. Lee, K.A. Yoon, K. Lee, Colorimetric genotyping of single nucleotide polymorphism based on selective aggregation of unmodified gold nanoparticles, Biosens. Bioelectron. 26 (2010) 730–735. [22] P.C. Ray, Diagnostics of single base-mismatch DNA hybridization on gold nanoparticles by using the hyper-Rayleigh scattering technique, Angew. Chem. Int. Ed. 45 (2006) 1151–1154. [23] L. Wang, P.C.H. Li, Gold nanoparticle-assisted single base-pair mismatch discrimination on a microfluidic microarray device, Biomicrofluidics 4 (2010) 032209. [24] L. Wang, P.C.H. Li, Optimization of a microfluidic microarray device for the fast discrimination of fungal pathogenic DNA, Anal. Biochem. 400 (2010) 282–288. [25] A. Sedighi, P.C.H. Li, Kinetic and thermodynamic analyses of DNA hybridization reveal the mechanism of gold nanoparticle-assisted single base-pair discrimination in the Nanobioarray chip, mTAS (2013) 865–867. [26] L. Yanez, J. Groffen, D.M. Valenzuela, C-K-ras mutations in human carcinomas occur preferentially in codon 12, Oncogene 1 (1987) 315–318. [27] L. Wang, P.C.H. Li, Flexible microarray construction and fast DNA hybridization conducted on a microfluidic chip for greenhouse plant fungal pathogen detection, J. Agric. Food Chem. 55 (2007) 10509–10516. [28] H. Urakawa, S. El Fantroussi, H. Smidt, J.C. Smoot, E.H. Tribou, J.J. Kelly, P.A. Noble, D.A. Stahl, Optimization of single-base-pair mismatch discrimination in oligonucleotide microarrays, Appl. Environ. Microbiol. 69 (2003) 2848–2856. [29] H. Chen, L. Wang, P.C.H. Li, Nucleic acid microarrays created in the doublespiral format on a circular microfluidic disk, Lab Chip 8 (2008) 826–829. [30] X.Y. Peng, P.C.H. Li, H.Z. Yu, M. Parameswaran, W.L. Chou, Spiral microchannels on a CD for DNA hybridizations, Sens. Actuat. B 128 (2007) 64–69. [31] R. Zinsky, S. Bolukbas, H. Bartsch, J. Schirren, A. Fisseler-Eckhoff, Analysis of KRAS mutations of exon 2 codons 12 and 13 by SNaPshot analysis in comparison to common DNA sequencing, Gastroenterol. Res. Pract. 2010 (2010) 789363. [32] M. Maekawa, T. Nagaoka, T. Taniguchi, H. Higashi, H. Sugimura, K. Sugano, H. Yonekawa, T. Satoh, T. Horii, N. Shirai, A. Takeshita, T. Kanno, Threedimensional microarray compared with PCR–single-strand conformation polymorphism analysis/DNA sequencing for mutation analysis of K-ras codons 12 and 13, Clin. Chem. 50 (2004) 1322–1327. [33] J.H. Park, I.J. Kim, H.C. Kang, Y. Shin, H.W. Park, S.G. Jang, J.L. Ku, S.B. Lim, S.Y. Jeong, J.G. Park, Oligonucleotide microarray-based mutation detection of the K-ras gene in colorectal cancers with use of competitive DNA hybridization, Clin. Chem. 50 (2004) 1688–1691. [34] P.C. McCabe, Production of single-stranded DNA by asymmetric PCR, in: M.A. Innis, D.H. Gelfand, J.J. Sninsky, T.J. White (Eds.), PCR Protocols, Academic Press, San Diego, 1990.
Contents lists available at ScienceDirect
Analytical Biochemistry journal homepage: www.elsevier.com/locate/yabio
Kras gene codon 12 mutation detection enabled by gold nanoparticles conducted in a nanobioarray chip Abootaleb Sedighi, Paul C.H. Li ⇑ Department of Chemistry, Simon Fraser University, Burnaby, British Columbia V5A 1S6, Canada
a r t i c l e
i n f o
Article history: Received 20 September 2013 Received in revised form 13 November 2013 Accepted 16 November 2013 Available online 26 November 2013 Keywords: Single nucleotide polymorphism Kras gene Gold nanoparticle Target DNA Microarray Nanobioarray chip
a b s t r a c t This study employs a nanobioarray (NBA) chip for multiple biodetection of single base pair mutations at the Kras gene codon 12. To distinguish between the mutant and wild-type target DNAs, current bioarray methods use high-temperature hybridization of the targets to the allele-specific probes. However, these techniques need prior temperature optimization and become harder to implement in the case of the detection of multiple mutations. We aimed to detect these mutations at a single temperature (room temperature), enabled by the use of gold nanoparticles (AuNPs) on the bioarray created within nanofluidic channels. In this method, a low amount of target oligonucleotides (5 fmol) and polymerase chain reaction (PCR) products (300 pg) were first loaded on the AuNP surface, and then these AuNP-bound targets were introduced into the channels of a polydimethylsiloxane (PDMS) glass chip. The targets hybridized to their complementary probes at the intersection of the target channels to the pre-printed oligonucleotide probe lines on the glass surface, creating a bioarray. Using this technique, fast and high-throughput multiple discrimination of the Kras gene codon 12 were achieved at room temperature using the NBA chip, and the specificity of the method was proved to be as high as that with the temperature stringency method. Ó 2013 Elsevier Inc. All rights reserved.
Single nucleotide polymorphisms (SNPs)1 are the most abundant genetic variation and are responsible for many genetic disorders and for certain cancers such as retinoblastoma, colorectal carcinoma, and adenocarcinoma of the pancreas [1,2]. SNP variations in the Kras gene codon 12 have been frequently observed in genetic disorders, and their detection is critical for the selection of the appropriate type of treatment [3,4]. DNA sequencing has been the most widely used method for detection of SNPs; however, the conventional separation-based sequencing method requires large reagent consumption and has a low throughput [5]. Although these issues have been effectively addressed in various newly emerged next-generation sequencing (NGS) methods, they still have technical limitations such as labor-intensive sample preparation and data interpretation. Real-time polymerase chain reaction (PCR) is another popular SNP detection approach, but it has nonlinear amplification of target DNAs and is limited in multiplex analysis [6]. Due to their high-throughput nature, DNA bioarrays have been used in a variety of applications such as gene expression profiling, ⇑ Corresponding author. Fax: +1 778 782 3765. E-mail address: [email protected] (P.C.H. Li). Abbreviations used: SNP, single nucleotide polymorphism; NGS, next-generation sequencing; PCR, polymerase chain reaction; NBA, nanobioarray; AuNP, gold nanoparticle; ssDNA, single-stranded DNA; dsDNA, double-stranded DNA; PM, perfect match; MM, mismatch; PDMS, polydimethylsiloxane; SDS, sodium dodecyl sulfate; APTES, 3-aminopropyltriethoxysilane; PBS, phosphate-buffered saline; DTA, DNA-toAuNP. 1
0003-2697/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ab.2013.11.019
comparative genomic hybridization, genotyping, and nucleic acid diagnostic applications as well as protein–DNA interaction studies [7,8]. Although recent advances in the NGS technologies have affected the popularity of DNA bioarrays in some clinical diagnostic applications, the convenience and efficiency of low-density arrays give them great advantages over the complex and still expensive NGS assays, especially when particular sites of the genome, instead of the whole genome, are targeted [9]. Here, we combined the lowdensity DNA bioarray with nanofluidics to produce the nanobioarray (NBA) chip in order to make the SNP assay fast and simple with low sample and reagent consumption. But the SNP assays using this method remain an issue because the hybridization temperature of the assays needs to be optimized to achieve single base pair discrimination [10–12]. Gold nanoparticles (AuNPs) have been used extensively in biosensors [13–18]. In 2004, Li and Rothberg [19] reported that unmodified single-stranded DNA (ssDNA), but not double-stranded DNA (dsDNA), could be adsorbed on the citrate-capped AuNP surfaces. This selective adsorption of ssDNA on AuNPs has been used as the basis of several DNA analysis devices [20–22]. This phenomenon was also used by our group, but in a novel manner, to enable single base pair discrimination of fungal pathogen PCR amplicons at room temperature [23], as opposed to the high-temperature method [24]. In this technique, target strands were first loaded on AuNPs and then were introduced to the oligonucleotide probes immobilized on the surface of a glass chip. Our recent kinetic and
Kras gene codon 12 mutation detection / A. Sedighi, P.C.H. Li / Anal. Biochem. 448 (2014) 58–64
thermodynamic studies showed that nonspecific binding of target DNA bases with the AuNP surface alters the kinetic pathways [25] altered hybridization and dehybridization processes of the target strands to the surface-immobilized probes lead to the unexpectedly enhanced discrimination power of AuNP-loaded targets (AuNP–targets), in contrast to the random-coiled targets (free– targets). Although the earlier work by our group provided an interesting finding for microarray research [23], only a single probe was used to distinguish two targets. In the common format of microarray experiments, discrimination of the target molecule between the perfect match (PM) probe and mismatch (MM) probe is normally used as the basis of SNP detection [24]. In the current study, we used four probes (one PM probe and three MM probes) in the NBA chip for multiple analyses of four SNPs at the codon 12 site of the Kras gene, which is clinically relevant in colorectal carcinoma and adenocarcinoma of the pancreas [26]. Here, we demonstrate that AuNP-loaded oligonucleotides or PCR amplicons discriminate between PM probes and MM probes immobilized on the NBA chip channel surface (Fig. 1). In the PM region, the targets are unloaded from the AuNPs and hybridize to the immobilized probes. In the MM region, the targets cannot hybridize to the immobilized probes and remain bound on the AuNPs and are washed downstream. We believe that this method, with further developments and evaluations with various sequences, can potentially replace the conventional temperature stringency method, which requires a tedious temperature optimization procedure and probe reductancy. In addition, our method allows multiplex analysis of SNPs by hybridization at the same temperature and offers the convenience of using the widely used polydimethylsiloxane (PDMS)-based chips to conduct DNA hybridizations because the room temperature operation alleviates the problem of evaporation through the gas-permeable polymer at high temperatures. Materials and methods Materials AuNPs with 5-, 10-, and 20-nm diameters (stabilized with citrate and tannic acid) were purchased from Sigma Life Science, and 12-nm AuNPs (stabilized with citrate) were obtained from NanoComposix (San Diego, CA, USA). Sodium dodecyl sulfate (SDS), 3-aminopropyltriethoxysilane (APTES), 25% glutaraldehyde, and Triton X-100 were purchased from Sigma–Aldrich. Negative
Fig.1. Schematic diagram of single base pair discrimination of AuNP-conjugated targets in PDMS glass microfluidic channel (A), channel intersection with the perfect match (PM) probe line (B), and channel intersection with the mismatch (MM) probe line (C). Both probe spots are 200 200 lm. The channel is not drawn to scale.
59
photoresist (SU-8 50) and its developer were purchased from MicroChem (Newton, MA, USA). The glass microscope slides of 3 2 inches were obtained from Fisher Scientific (Ottawa, Ontario, Canada). All oligonucleotides and the primers were synthesized and modified by Integrated DNA Technologies (Coralville, IA, USA). The 60-mer target oligonucleotides with the sequence of Kras gene were modified by a biotin molecule at the 50 end (Table 1). The sequences were designed in such a way that the codon 12 sequence was located at the center of the oligonucleotides. Four different 20mer oligonucleotide probes, each of them complementary to one of the targets (60-mer), were designed. The probes were modified with an amine group and a C12 spacer at the 50 end. All oligonucleotides are listed in Table 1. NBA chip fabrication Fabrication of a PDMS slab (2 2 inches) with 16 parallel channels has been described elsewhere [27]. The width of straight channels was 200 lm, and the height was 35 lm. A flat-end syringe needle (15 gauge) was used to punch out the solution reservoirs (1.5 mm in diameter) at both ends of the channels on the 2-mmthick PDMS. The glass slides were aldehyde functionalized using an established procedure [28]. The PDMS slab was sealed reversibly to the modified glass slide to create the NBA chip. Probe immobilization The probes were printed using a procedure similar to that described in previous reports [26,28]. Briefly, 0.5 ll of 25 lM probe solution (in 1.0 M NaCl + 0.15 M NaHCO3) was added to the inlet reservoir of each of the 16 horizontal channels of the NBA chip, and the solution was then filled in the channel using suction applied at the outlet reservoir. After 30 min of incubation at room temperature, the probe solution was pumped out of the channel. Following washing of the channel by the bicarbonate washing solution (0.15 M NaHCO3 + 0.2% SDS), the PDMS slab was peeled off, leaving behind 16 probe lines on the glass slide. The glass slide was rinsed by water and dried. Amplification of genomic DNA Two genomic DNA samples, containing different allele compositions of the Kras gene codon 12, were purchased from Horizon Diagnostics (Cambridge, UK). One of the samples contained the pure wild-type allele, and the other contained a mixture of 50% wild type and 50% G12D mutant (the genotypes were characterized by the vendor using Sanger sequencing). A pair of specific primers (forward: 50 -biotin-TGACTGAATATAAACTTGTGGTAGTTGGAG-30 ; reverse: 50 -ATGATTCTGAATTAGCTGTATCGTCAAGGC-30 ) was used in order to obtain the 80-bp PCR products. Asymmetric PCR was performed by varying the concentration of the reverse primer (0.05, 0.1, 0.2, 0.3, 0.4, and 0.5 lM) in the presence of the forward primer with a constant concentration of 0.5 lM. A custom PCR protocol on a Cetus thermocycler (PerkinElmer) was used for the amplification. The thermocycling was initiated by 3 min of denaturation, followed by 30 thermal cycles of 95 °C for 40 s (denaturation), 55 °C for 30 s (annealing), and 72 °C for 60 s (extension), and then was terminated by 10 min of final extension at 72 °C. The amplified products were purified using a nucleotide removal kit (Qiagen). DNA hybridization A second PDMS slab with 16 channels was sealed against the glass slide pre-printed with the probe lines. The straight channels
60
Kras gene codon 12 mutation detection / A. Sedighi, P.C.H. Li / Anal. Biochem. 448 (2014) 58–64
Table 1 Sequences of probes and target oligonucleotides. Targets
Probes
WT (W60) G12Aa (A60) G12Db (D60) G12Vc (V60) WT (W) G12A (A) G12D (D) G12V (V)
50 -/Biotin/GAA TAT AAA CT T GTG GTA GTT GGA 50 -/Biotin/GAA TAT AAA CT T GTG GTA GTT GGA 50 -/Biotin/GAA TAT AAA CT T GTG GTA GTT GGA 50 -/Biotin/GAA TAT AAA CT T GTG GTA GTT GGA 50 -/C12amine/CC TAC GCC ACC AGC TCC AAC-30 50 -/C12amine/CC TAC GCC AGC AGC TCC AAC-30 50 -/C12amine/CC TAC GCC ATC AGC TCC AAC-30 50 -/C12amine/CC TAC GCC AAC AGC TCC AAC-30
GCT GCT GCT GCT
GGT GGC GTA GGC AAG AGT GCC TTG ACG ATA CAG-30 GCT GGC GTA GGC AAG AGT GCC TTG ACG ATA CAG-30 GAT GGC GTA GGC AAG AGT GCC TTG ACG ATA CAG-30 GTT GGC GTA GGC AAG AGT GCC TTG ACG ATA CAG-30
Note: WT, wild type. a G12A represents Gly12Ala amino acid substitution. b G12D represents Gly12Asp amino acid substitution. c G12V represents Gly12Val amino acid substitution.
were orthogonal to the probe lines on the glass slide surface. Before hybridization to the probe, target oligonucleotides were conjugated noncovalently to the surface of AuNPs. To achieve this, AuNP solution was added to the target solution and the mix was incubated at 95 °C for 15 min. Free–target solutions (control) were not mixed with the AuNP solutions. The target solutions (free and AuNP-conjugated) were then diluted in the hybridization buffer (1 SSC [sodium chloride and sodium citrate] + 0.2% SDS) to a final concentration of 10 nM for oligonucleotides and 0.6 ng/ll for PCR products. The AuNP–target solution (0.5 ll) was added to the inlet reservoir, filled in the channel using vacuum suction, and then incubated for 20 min at 22 °C (unless noted otherwise). The hybridization of the targets to the complementary probes occurred at the intersection of target channels with the probe lines (resulting in the hybridization detection area of 200 200 lm2). After incubation, the target solutions were pumped out from the channels. High-temperature experiments (55 °C) were achieved using a Peltier device. Immediately after hybridization, the target solutions were pumped out and the channels were washed by 2 ll of the hybridization buffer. To detect the duplex oligonucleotides formed at the hybridization patches, streptavidin–Cy5 solution (50 lg/ml in 1 PBS [phosphate-buffered saline]) was added to the channels. After incubation for 15 min, the channel was rinsed using a wash solution (1 PBS and 0.1% Tween 20), and then the PDMS slab was peeled off from the glass slide. Detection Following rinsing and drying, the glass slide was scanned on a confocal laser fluorescent scanner (Typhoon 9410, GE Healthcare) at a 10-lm resolution, as described previously [29,30]. The excitation and emission wavelengths were 633 and 670 nm, respectively. The photomultiplier tube voltage was set to 600 V. The scanned image was analyzed by ImageQuant 5.2 software. The average fluorescent signals were measured in relative fluorescent units. Results and discussion The discrimination between the wild-type and G12D mutant, which is the most frequent one in the Kras gene codon 12 mutations in adenocarcinomas [31], was chosen for the current study. The scanned image in Fig. 2A shows the spots resulting from hybridization of two 60-mer targets, W60 and D60, to their corresponding PM and MM probes (W and D), respectively, performed in duplicate. The histogram (Fig. 2B) was created based on measured signal intensities of the spots along the vertical target channel line. The discrimination ratio, calculated based on the ratio of the mean signal intensity of PM to that of MM, is shown above the bars in Fig. 2B. We observe that when the targets are free, there is no significant difference between the signal
Fig.2. (A) Scanned image with the spots resulting from hybridization of 10 nM 60mer targets (W60 and D60) with their corresponding PM and MM probes (W and D) at 22 °C for 20 min in the NBA chip. The targets were either free (free–targets) or previously conjugated with 10-nm AuNPs (AuNP–targets). The boxed regions are the expected true positive binding regions. (B) Histogram showing the signal intensities of the spots obtained along the vertical target channels, with the true positive binding signals represented by the hatched bars. Error bars show the standard deviations of two measurements. The number above each column shows the discrimination ratio, which is determined by dividing the intensity of the PM spots (W60–W and D60–D) by that of the MM spots (W60–D and D60–W).
intensities of the PM and MM spots, or the discrimination ratio is close to 1.0, showing why a high temperature such as 55 °C would have been required to achieve sufficient discrimination [32,33]. An improvement in the discrimination ratio is observed in the case of AuNP–targets, where the PM/MM discrimination ratios are 1.6 ± 0.2 and 1.7 ± 0.2 for W60 and D60 targets, respectively.
Kras gene codon 12 mutation detection / A. Sedighi, P.C.H. Li / Anal. Biochem. 448 (2014) 58–64
61
Fig.3. Scanned images and the corresponding histograms (below each image) resulting from free–target hybridization at 22 °C (A), AuNP–target hybridization at 22 °C (B), and free–target hybridization at 55 °C (C). The 60-mer target DNAs were conjugated with 5-nm AuNPs at a DTA of 1:1. Error bars show the standard deviations of three measurements. The number above each column shows the discrimination ratio, which is determined by dividing the intensity of the PM spots by the average of the three MM spots. For other conditions, see Fig. 2.
Fig.4. Original scanned image showing half of the 16 16 NBA chip (A), the inset inside the blue box (B), and the resulting histogram (C) of free–target (60-mer) and 5-nm AuNP–target based on 60 min of hybridization at 22 °C. Error bars are standard deviations from three replicates. For other conditions, see Fig. 2.
62
Kras gene codon 12 mutation detection / A. Sedighi, P.C.H. Li / Anal. Biochem. 448 (2014) 58–64
Method optimization Because the improvement in the discrimination ratio using the AuNP method at room temperature is marginal, this method needs to be further optimized. To improve the discrimination ratio, we use a single factor analysis approach to investigate potentially effective factors such as AuNP size and DNA-to-AuNP ratio (DTA). The AuNP size was the first factor whose potential impact on the discrimination ratios was tested. This was performed by loading the targets on AuNPs of various sizes (5, 10, 12, and 20 nm) and then comparing the discrimination ratios with the ones obtained from free–targets. As shown in Fig. S2 of the online supplementary material, there is not much improvement in discrimination ratio as the AuNP sizes get smaller from 20 to 10 nm. But a great increase in the ratios (2.2 or 2.7) is obvious at 5 nm AuNPs. This higher discrimination can be explained by the enhanced dehybridization rate constants in the presence of small AuNPs (5 nm) due to the greater curvature of the smaller particles [25]. As we know, dehybridization starts from binding of the AuNPs to the ssDNA parts in the duplex, but this was hindered by the repulsion between the negatively charged surface of large AuNPs and the neighboring folded parts of the duplex, where the negative charge density is twice as much as the ssDNA parts. Owing to the higher degree of curvature in small AuNPs (5 nm), this repulsion is less of an issue, leading to the enhanced dehybridization when 5-nm particles are used. The DTA was another factor whose influence on the discrimination ratios was investigated. As concluded from Fig. S2, as DTA decreases from 3 to 1, meaning that the number of AuNPs per target strand increases from 0.33 to 1, the discrimination ratio increases (up to 2.7). Therefore, DTA of 1 was chosen as the optimum, which corroborated with our previous results [23]. However, DTAs lower than 1 could not be applied because AuNPs aggregate in those solutions.
It is noted that although the discrimination ratio is improved, the signal intensity decreases in both the AuNP method and the high-temperature method. These lower signal intensities result from the lower hybridization rate constants of AuNP–targets as compared with free–targets [25]. This issue can be resolved by using long hybridization times given that 12 h is usually needed to achieve high sensitivities and specificities in conventional nucleic acid bioarray experiments [34]. But we did not need 12 h; we hybridized for 60 min instead of 20 min, with the results shown in Fig. 4. Although the signal intensities are generally higher, more important, the PM spots of AuNP–targets show comparable signal intensities to the free–target spots. Furthermore, the greater decrease in the intensities of the MM spots resulted in much higher discrimination ratios ranging from 6.8 to 19.7. This additional incubation time dramatically increases the discrimination ratios, whereas the method completed in 1 h is still faster than the conventional method [32,33]. Detection of genomic DNA Having demonstrated a satisfactory performance on the SNP detection of the single-stranded oligonucleotide targets, the AuNP-assisted technique was evaluated using the PCR amplicons
Multiple mutation detection Following the optimization of AuNP-enabled method using wild type and one of its mutant (G12D) of Kras sequences, the method was fully evaluated using all three codon 12 mutants plus the wild type. Fig. 3A and B show the image and the corresponding histogram of room temperature hybridization of free–targets and AuNP-conjugated targets. The number above each group of PM or MM column is the discrimination ratio of each target, which is the signal of the PM spot divided by the average of the signals of the other three MM spots (Eq. (1)):
Discrimination ratio ¼
PM 1=3ðMM1 þ MM2 þ MM3 Þ
ð1Þ
Whereas the discrimination of the AuNP D60 target remains low (2.4), the ratios resulting from the AuNP–targets range from 5.9 for the V60 target to 7.1 for the A60 target. As a comparison, a similar experiment using the conventional temperature stringency method at 55 °C was performed, and this hybridization temperature is commonly used for Kras sequence discrimination in bioarray analysis [32]. As shown in Fig. 3C, free–targets at 55 °C created spots with discrimination ratios ranging from 1.4 to 6.6. These discrimination ratios are also in the same range as those obtained previously using bioarray approaches [32,33]. The hightemperature values are comparable to, if not worse than, the ones resulting from AuNP–targets at room temperature hybridization (ratios of 2.4–7.1). These results demonstrate that using the AuNP-enabled method, the targets can effectively discriminate between their PM probes and the MM probes at room temperature at a short hybridization time of 20 min.
Fig.5. Original scanned image and the corresponding histogram resulting from hybridization of PCR products amplified from genomic DNAs with 100% wild-type Kras alleles (W0 ) at 22 °C for 60 min on the NBA chip. Asymmetric PCR was performed using different concentrations of the reverse primer (0.05, 0.1, 0.2, 0.3, 0.4, and 0.5 lM) with a constant forward primer concentration of 0.5 lM. All of the targets were previously loaded on the 5-nm AuNPs. In the histogram, the column bars show the averages of signal intensities of the spots, measured at the intersection of horizontal probe lines and the vertical target lines, and the true positive binding signals of W0 are represented by the hatched bars. Error bars show the standard deviations of three measurements. The number above each column shows the discrimination ratio, which is determined by dividing the intensity of the PM spots (W0 to W) by that of the MM spots (W0 to A, V, and D).
Kras gene codon 12 mutation detection / A. Sedighi, P.C.H. Li / Anal. Biochem. 448 (2014) 58–64
of genomic DNA (double-stranded). Because the PCR amplicons are longer (80 bp) and are double-stranded, as compared with oligonucleotides, we further optimized the hybridization conditions. First, although DTAs lower than 1 cause AuNP aggregation in oligonucleotides, AuNP PCR products with DTAs down to 0.6 can be obtained without AuNP aggregation (data not shown). The higher stability of AuNPs in PCR amplicons, as compared with oligonucleotides, is probably obtained because the PCR amplicons (80 bp) are longer than the target oligonucleotides (60-mer) used here. Second, we used asymmetric PCR in order to reduce the influence of complementary strands, which will not only compete with the binding target strands for the immobilized probes but also occupy the AuNPs with nonbinding strands. In asymmetric PCR, decreasing the concentration of the unlabeled reverse primer leads to preferential amplification of the labeled strand at higher concentrations [34]. We compared the hybridization signals of the 100% W amplicons using different concentrations of reverse primer with a constant forward primer concentration of 0.5 lM. As shown in Fig. 5, no significant discrimination resulted if the concentration of the reverse primer was equal to or higher than 0.3 lM. We believe that the higher concentration of the reverse primer resulted in a higher concentration of complementary strands, inducing the unloading of the target strands from the AuNPs by hybridization. These unloaded target strands will behave like the free–targets; therefore, the overall specificity will be affected. A sharp rise in the discrimination ratio is obvious on reducing the concentration of reverse primer from 0.2 to 0.05 lM (10% of concentration of forward primer). The performance of the AuNP-assisted technique in the detection of the PCR amplicons from the genomic samples containing mixed alleles was also investigated. A genomic sample containing a mixture of wild type and G12D (50% each), together with another sample containing 100% wild type, was amplified and detected on
63
the NBA chip. As shown in Fig. 6, the mixed allele sample showed a distinct positive hybridization signal on probes W and D in the presence of other probes. The discrimination ratio is satisfactory, and the signals are as high as those in the free–targets. Considering that the targets W and D show lower discrimination ratios than the targets A and V (Fig. 3 and 4), we believe that the PCR amplicons from A and V mutants will have higher performance. So, this experiment demonstrates the flexibility and power of the AuNP-assisted technique in SNP detection of genomic samples. Conclusion In this study, we have developed a fast and multiplex method for multiple SNP detection at room temperature. The AuNP-assisted method enables the DNA bioarrays to perform detection of the wild type and three mutants of the Kras gene codon 12. The method proved to be able to discriminate between the wild-type sequence and different oligonucleotide mutants as effectively as the high-temperature method. The various factors were investigated and optimized for effective single base pair discriminations. It was confirmed that a higher discrimination ratio is achievable when a longer hybridization time is used without loss of signal intensity. The method was tested on PCR amplicons of the wildtype allele (W) and of one mutant allele (D). Further developments are required in order to extend the method to detect other clinically relevant Kras gene mutations, such as those at codons 13 and 61, as well as the mutations in the other sequences. Acknowledgments We gratefully acknowledge financial support from the Natural Sciences and Engineering Research Council of Canada for a Discovery Grant. We thank John Brunstein (BC Children’s Hospital) for useful suggestions in detection of the Kras mutation, Dipankar Sen (Department of Molecular Biology and Biochemistry, Simon Fraser University) for useful advice and technical support on PCR amplification, and Byron Gates and Idah Pekcevic (Department of Chemistry, Simon Fraser University) for AuNP characterization. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ab.2013.11.019. References
Fig.6. Original scanned image (A) and the corresponding histogram (B) resulting from hybridization of PCR products amplified from genomic DNAs with different Kras allele compositions (one sample is 100% wild type, and the other is 50% wild type plus 50% G12D mutant). The targets are either free or loaded 5-nm AuNPs. Error bars are from three replicates. For other conditions, see Fig. 5.
[1] F.S. Collins, Medical and societal consequences of the Human Genome Project, N. Engl. J. Med. 341 (1999) 28–37. [2] S. White, M. Kalf, Q. Liu, M. Villerius, D. Engelsma, M. Kriek, E. Vollebregt, B. Bakker, G.J.B. van Ommen, M.H. Breuning, J.T. den Dunnen, Comprehensive detection of genomic duplications and deletions in the DMD gene, by use of multiplex amplifiable probe hybridization, Am. J. Hum. Genet. 71 (2002) 365– 374. [3] S. Ye, S. Dhillon, X. Ke, A.S. Collins, Y.N.M. Day, An efficient procedure for genotyping single nucleotide polymorphisms, Nucleic Acids Res. 29 (2001) 7773–7782. [4] C.J. Allegra, J.M. Jessup, M.R. Somerfield, S.R. Hamilton, E.H. Hammond, D.F. Hayes, P.K. McAllister, R.F. Morton, R.L. Schilsky, American Society of Clinical Oncology provisional clinical opinion: testing for KRAS gene mutations in patients with metastatic colorectal carcinoma to predict response to antiepidermal growth factor receptor monoclonal antibody therapy, J. Clin. Oncol. 27 (2009) 2091–2096. [5] K. Sato, M. Onoguchi, Y. Sato, K. Hosokawa, M. Maeda, Non-cross-linking gold nanoparticle aggregation for sensitive detection of single-nucleotide polymorphisms: optimization of the particle diameter, Anal. Biochem. 350 (2006) 162–164. [6] N.N. Iscove, M. Barbara, M. Gu, M. Gibson, C. Modi, N. Winegarden, Representation is faithfully preserved in global cDNA amplified exponentially from sub-picogram quantities of mRNA, Nat. Biotechnol. 20 (2002) 940–943.
64
Kras gene codon 12 mutation detection / A. Sedighi, P.C.H. Li / Anal. Biochem. 448 (2014) 58–64
[7] L. Wang, P.C.H. Li, Microfluidic DNA microarray analysis: a review, Anal. Chim. Acta 687 (2011) 12–27. [8] S. Russell, L.A. Meadows, Microarray Technology in Practice, Academic Press/ Elsevier, Amsterdam, 2009. [9] A. Sedighi, P.C.H. Li, Challenges and future trends in DNA microarray analysis, in: C. Simó, A. Cifuentes, V. García-Cañas (Eds.), Fundamentals of Advanced Omic Technologies: From Genes to Metabolites, Elsevier Science, San Diego, 2013. [10] C. Situma, Y. Wang, M. Hupert, F. Barany, R.L. McCarley, S.A. Soper, Fabrication of DNA microarrays onto poly(methyl methacrylate) with ultraviolet patterning and microfluidics for the detection of low-abundant point mutations, Anal. Biochem. 340 (2005) 123–135. [11] A. Sedighi, L. Wand, P.C.H. Li, 2D nanofluidic bioarray for nucleic acid analysis, in: K. Iniewski, S. Selimovic (Eds.), Nanopatterning and Nanoscale Devices for Biological Applications, Taylor & Francis/CRC Press, Boca Raton, FL, 2013. [12] O.Y.F. Henry, C.K. O_Sullivan, Rapid DNA hybridization in microfluidics, Trends Anal. Chem. 33 (2012) 9–22. [13] S. Link, M.A. El-Sayed, Spectral properties and relaxation dynamics of surface plasmon electronic oscillations in gold and silver nanodots and nanorods, J. Phys. Chem. B 103 (1999) 8410–8426. [14] R. Elghanian, J.J. Storhoff, R.C. Mucic, R.L. Letsinger, C.A. Mirkin, Selective colorimetric detection of polynucleotides based on the distance-dependent optical properties of gold nanoparticles, Science 277 (1997) 1078–1081. [15] K. Sato, K. Hosokawa, M. Maeda, Rapid aggregation of gold nanoparticles induced by non-cross-linking DNA hybridization, J. Am. Chem. Soc. 125 (2003) 8102–8103. [16] K. Sato, K. Hosokawa, M. Maeda, Non-cross-linking gold nanoparticle aggregation as a detection method for single-base substitutions, Nucleic Acids Res. 33 (2005) e4. [17] C. Wang, Y. Chen, T. Wang, Z. Ma, Z. Su, Biorecognition-driven self-assembly of gold nanorods: a rapid and sensitive approach toward antibody sensing, Chem. Mater. 19 (2007) 5809–5811. [18] C.C. Huang, Y.F. Huang, Z. Cao, W. Tan, H.T. Chang, Aptamer-modified gold nanoparticles for colorimetric determination of platelet-derived growth factors and their receptors, Anal. Chem. 77 (2005) 5735–5741. [19] H.X. Li, L. Rothberg, Colorimetric detection of DNA sequences based on electrostatic interactions with unmodified gold nanoparticles, Proc. Natl. Acad. Sci. USA 101 (2004) 14036–14039. [20] H.X. Li, L.J. Rothberg, Label-free colorimetric detection of specific sequences in genomic DNA amplified by the polymerase chain reaction, J. Am. Chem. Soc. 126 (2004) 10958–10961.
[21] H. Lee, S.W. Joob, S.Y. Lee, C.H. Lee, K.A. Yoon, K. Lee, Colorimetric genotyping of single nucleotide polymorphism based on selective aggregation of unmodified gold nanoparticles, Biosens. Bioelectron. 26 (2010) 730–735. [22] P.C. Ray, Diagnostics of single base-mismatch DNA hybridization on gold nanoparticles by using the hyper-Rayleigh scattering technique, Angew. Chem. Int. Ed. 45 (2006) 1151–1154. [23] L. Wang, P.C.H. Li, Gold nanoparticle-assisted single base-pair mismatch discrimination on a microfluidic microarray device, Biomicrofluidics 4 (2010) 032209. [24] L. Wang, P.C.H. Li, Optimization of a microfluidic microarray device for the fast discrimination of fungal pathogenic DNA, Anal. Biochem. 400 (2010) 282–288. [25] A. Sedighi, P.C.H. Li, Kinetic and thermodynamic analyses of DNA hybridization reveal the mechanism of gold nanoparticle-assisted single base-pair discrimination in the Nanobioarray chip, mTAS (2013) 865–867. [26] L. Yanez, J. Groffen, D.M. Valenzuela, C-K-ras mutations in human carcinomas occur preferentially in codon 12, Oncogene 1 (1987) 315–318. [27] L. Wang, P.C.H. Li, Flexible microarray construction and fast DNA hybridization conducted on a microfluidic chip for greenhouse plant fungal pathogen detection, J. Agric. Food Chem. 55 (2007) 10509–10516. [28] H. Urakawa, S. El Fantroussi, H. Smidt, J.C. Smoot, E.H. Tribou, J.J. Kelly, P.A. Noble, D.A. Stahl, Optimization of single-base-pair mismatch discrimination in oligonucleotide microarrays, Appl. Environ. Microbiol. 69 (2003) 2848–2856. [29] H. Chen, L. Wang, P.C.H. Li, Nucleic acid microarrays created in the doublespiral format on a circular microfluidic disk, Lab Chip 8 (2008) 826–829. [30] X.Y. Peng, P.C.H. Li, H.Z. Yu, M. Parameswaran, W.L. Chou, Spiral microchannels on a CD for DNA hybridizations, Sens. Actuat. B 128 (2007) 64–69. [31] R. Zinsky, S. Bolukbas, H. Bartsch, J. Schirren, A. Fisseler-Eckhoff, Analysis of KRAS mutations of exon 2 codons 12 and 13 by SNaPshot analysis in comparison to common DNA sequencing, Gastroenterol. Res. Pract. 2010 (2010) 789363. [32] M. Maekawa, T. Nagaoka, T. Taniguchi, H. Higashi, H. Sugimura, K. Sugano, H. Yonekawa, T. Satoh, T. Horii, N. Shirai, A. Takeshita, T. Kanno, Threedimensional microarray compared with PCR–single-strand conformation polymorphism analysis/DNA sequencing for mutation analysis of K-ras codons 12 and 13, Clin. Chem. 50 (2004) 1322–1327. [33] J.H. Park, I.J. Kim, H.C. Kang, Y. Shin, H.W. Park, S.G. Jang, J.L. Ku, S.B. Lim, S.Y. Jeong, J.G. Park, Oligonucleotide microarray-based mutation detection of the K-ras gene in colorectal cancers with use of competitive DNA hybridization, Clin. Chem. 50 (2004) 1688–1691. [34] P.C. McCabe, Production of single-stranded DNA by asymmetric PCR, in: M.A. Innis, D.H. Gelfand, J.J. Sninsky, T.J. White (Eds.), PCR Protocols, Academic Press, San Diego, 1990.
