Surface plasmon field-enhanced fluorescence spectroscopy in PCR produc
http://www.100md.com
《核酸研究医学期刊》
1 Max-Planck-Institute for Polymer Research, Ackermannweg 10, 55128, Mainz, Germany, 2 Max-Planck-Institute of Biochemistry, Am Klopferspitz 18A, 82152, Munich, Germany and 3 Department of Biochemistry B, The Panum Institute, IMBG, Blegdamsvej 3c, DK-2200 Copenhagen N, Denmark
* To whom correspondence should be addressed. Tel: +49 6131 379 160; Fax: +49 6131 379 360; Email: knoll@mpip-mainz.mpg.de
ABSTRACT
Surface plasmon field-enhanced fluorescence spectroscopy (SPFS) was recently developed for PCR product analysis, which allowed for real-time monitoring of hybridization processes and for the detection of trace amounts of PCR products, with a detection limit of 100 fmol on the peptide nucleic acid (PNA) probe surface, and 500 fmol on the DNA probe surface. By selectively labeling the strands of PCR-amplified DNA, it was shown that the heat denaturation process in combination with the application of low-salt condition substantially reduced the interference from the antisense strands and thus simplified the surface hybridization. Furthermore, SPFS was demonstrated to be capable of quantitatively discriminating the difference induced by single nucleotide substitution, even within one minute of contact time.
INTRODUCTION
The recent development of DNA biosensors with the merits of high sensitivity and selectivity has been accelerated greatly by the achievements in the Human Genome Project (HGP). A number of biosensors, based on surface plasmons (1,2), acoustic waves (3), electrochemistry (4) or fiber optics (5), have demonstrated excellent performances in oligonucleotide analysis by enabling real-time monitoring of hybridization kinetics and mismatch discrimination. However, great challenges still exist for detecting PCR-amplified DNA or genomic DNA. As analytes, PCR products differ from oligonucleotides mainly in two aspects. First, oligonucleotide targets used in most of the biosensing studies are usually <30 bases in length, while PCR products vary from 100 bp to several kb in length. The vast amount of bases is accompanied by a significantly enhanced sequence complexity and imposes tremendous difficulties in PCR product analysis. For instance, the non-selective fragments of PCR products may potentially interfere with the specific hybridization by partially matching with the surface-attached probes or by physical adherence, which usually results in high background signals in DNA biosensors (6) and genechips (7). Steric hindrance arising from the solid support and neighboring bound species is also expected to be enhanced in the presence of non-selective fragments, which considerably decreases the binding efficiency of PCR products. These length-dependent effects are presumed to account for the difficulties in biosensor applications for PCR product detection. Second, the double-stranded nature of PCR products is another major problem, because it renders the recognition unit of the sense strand inaccessible to the surface-attached probes. As a consequence, the sensitivities reported for PCR product detection are so far rather poor (mostly in the nanomolar range) (8,9), which is generally several orders of magnitude lower than that reported for oligonucleotides (in the pico-, or femtomolar range) (10). At present, only end-point results can be offered for most of the reported techniques (11–14). These hinder an in-depth investigation of the hybridization mechanism and quantitative detection of single nucleotide polymorphisms (SNPs).
In this paper, surface plasmon field-enhanced fluorescence spectroscopy (SPFS) is employed for investigating the interfacial hybridization of PCR products, in combination with the application of peptide nucleic acids (PNAs) probes. Several advantages are expected in this study. First, combining the enormous surface plasmon field enhancement with the sensitive fluorescence technique, SPFS could be a powerful tool in detecting SNPs and investigating hybridization kinetics. Second, as mentioned earlier, the presence of the antisense strand may interfere with the hybridization of the sense strand with the surface-attached probes. By selectively labeling the sense or/and antisense DNA strand, the influence of the antisense strand can be possibly explored. Third, decorated at the N-terminus by nine ethylene glycol units and a biotin moiety, the PNA probes are immobilized on a streptavidin (SA) modified gold surface via biotin/SA interaction. According to the molecular dimensions of the SA molecule (4.5 x 4.5 x 5.3 nm3) (15), the bound PNA molecules are laterally spaced. Thus, this surface is desirable to alleviate the steric hindrance induced by neighboring sites to some extent. Finally, the uncharged pseudopeptide backbone of PNA allows for the application of low salt condition (16), in which both the inter- and intra-molecular structure of PCR products can be largely eliminated after the pretreatment of heat denaturation process. In view of these factors, our work will be focused on the quantitative detection of SNPs, elucidating the hybridization phenomenon, and systematic study of the limit-of-detection (LOD) of PCR products by employing SPFS.
MATERIALS AND METHODS
Surface functional multilayer assembly
Figure 1 describes the multilayer architecture employed for the PNA probe assembly. The biotinylated thiols and OH-terminated thiols were synthesized in our group and purchased from Sigma, respectively. A mixed thiol solution was prepared in absolute ethanol with a total concentration of 0.5 mM (0.45 mM OH-terminated thiol and 0.05 mM biotinylated thiol). The gold substrates were immersed in the thiol solution for >6 h at room temperature. They were then subjected to repeated rinsing in ethanol. After being dried with a stream of nitrogen gas, the substrates can be stored in nitrogen at 4°C for several days.
Figure 1. (A) Chemical structures of biotinylated thiol and OH-terminated thiol. (B) A schematic representation of the sensor surface architecture: a binary SAM composed of the biotinylated thiol and OH-terminated thiol is formed on the gold surface. A monolayer of SA tetramer is attached by the interaction with the biotin moiety, exposing the binding sites for biotinylated PNA or DNA probe assembly.
SA was purchased from Sigma. Exposing the binary thiol modified gold substrates to a 1 μM SA solution, a monolayer of the tetrameric protein was attached within minutes via the interaction with the biotin moiety. The biotinylated PNA P2, PNA-11 (synthesized in our group), or DNA P2 (MWG-Biotech) can then be assembled onto the SA layer (cf. Table 1). The buffer condition used in this work is 10 mM phosphate buffer (PB, pH 7.4), containing 0.005% Tween-20 and 2 mM EDTA. High sodium concentrations were obtained by adding sodium chloride.
Table 1. The sequences of all of the PNA and DNA samples used in this work
PCR amplification
The PCR protocol was optimized on a thermocycler (Biometra) by designing the primer sequences and adjusting the annealing temperature and concentration of each component. Recombinant plasmids (4 kb) were used as templates for the PCR amplification. The forward and reverse primers were purchased from MWG-Biotech. As listed in Table 1, both are 30-base oligonucleotides with both calculated melting temperatures being close to each other. Unlabeled or labeled by Cy5 at the 5' end, primers were employed as needed. Unless otherwise stated, the Cy5-labeled forward primer and unlabeled reverse primer were used routinely for amplifying the PCR products with only the sense strand labeled.
All PCRs were carried out under the same conditions. Reagents for each 50 μl reaction volume included: 5 U of Taq polymerase (Amersham Bioscicences), 1x PCR buffer (Amersham Biosciences), 60 pmol of the forward primer and 80 pmol of reverse primer, 0.2 mM dNTPs (Fermentas) and 100 ng of the plasmid DNA. After an initial denaturation step at 96°C for 1 min, each of the 30 cycles of amplification consisted of 30 s of template denaturation at 96°C, 30 s of primer annealing at 50°C and 30 s of primer extension at 72°C.
Preparation of the analytes
The resulting 196 bp PCR products (T1, T2 and T3, cf. Table 1) were identified by agarose electrophoresis (2%). As shown in Table 1, the three PCR products contain a 15-base recognition sequence present in the middle of the sense strand approximately, which fully matches or exhibits a one-base/two-base mismatch with PNA P2, respectively. Ethanol precipitation was performed afterwards in order to remove the salts, dNTPs, excess primers and enzyme from the PCR sample. For 50 μl of PCR sample, 125 μl ethanol (Sigma) and 5 μl of sodium acetate (3 M) were added. The mixture was shaken, and then kept at –20°C for >6 h to induce precipitation. The PCR products were collected by centrifugation at 16 000 g for 0.5 h. The ethanol precipitation was empirically examined to effectively purify the PCR products, even at a large sample scale.
Since all the PCR products used in this paper are end-labeled by Cy5 in the sense or/and antisense strand, their concentrations can be determined by recording the adsorption of Cy5 at 650 nm by the following equation:
(1)
assuming a molar extinction coefficient of = 250 000 M–1 cm–1, as provided by the manufacturer (Amersham Biosciences).
The PCR products underwent the following denaturation process prior to the biosensing studies, unless otherwise stated: the heat denaturation is composed of three steps, heating the DNA sample (in 10 mM PB buffer) at 96°C for 10 min, quickly quenching it to 0°C in an ice–water mixture and returning to room temperature.
Instrumental
The SPFS set-up was constructed based on a conventional surface plasmon spectrometer, in combination with the photon counting technique (17). The emitted fluorescence light from the sample surface was collected through a quartz flow cell. A lens (f = 40 mm; Owis) focused the light through an interference filter ( = 670 nm, T = 69%, LOT) onto a photomultiplier tube (Hamamatsu), which fed the signal to a photon counter (Hewlett Packard). This fluorescence detection unit was mounted on the sample arm, such that its position was fixed relative to the sample. The kinetic measurements were realized by setting the incident angle at the resonant angle and recording the fluorescence intensity as a function of time.
RESULTS AND DISCUSSION
Role of the antisense strand in surface hybridization
In order to elucidate the role of the antisense strand in hybridization, the three PCR products were prepared, i.e. the sense strand, antisense strand and both strands labeled PCR T2. They were obtained by having the forward primer, the reverse primer or both primers labeled, respectively. After going through the heat denaturation process, the three PCR T2 (each concentration c0 =100 nM) were applied for hybridization with the surface-attached PNA P2 at 10 mM sodium concentration, monitored by SPFS. As shown in Figure 2 , the hybridization signal obtained by both strands labeled PCR T2 is slightly higher than that obtained with the sense strand labeled, which is in accordance with the weak fluorescence signal induced by the antisense strand .
Figure 2. Hybridizations of PCR T2 (100 nM) with both strands labeled (A), only the sense strand labeled (B) or only the antisense strand labeled (C) on the PNA P2 surface at 10 mM sodium concentration. An attenuator (with an attenuation factor of 14) was applied to reduce the fluorescence intensity.
As known to us, the heat denaturation process consists of three temperature states, heating at 96°C, quickly quenching to 0°C and returning to room temperature. Empirically, the double strands melt to give single-stranded DNA if heated at 96°C for several minutes. The subsequent quenching step is expected to preserve the single-stranded structure to the largest extent, by quickly running through the temperature range favorite for re-annealing (20 to 70°C). Nevertheless, the double strands tend to re-associate with the temperature returning to room temperature, and the re-association is largely dependent on the bulk salt concentration. Obviously, a lower salt concentration is desirable to inhibit the re-annealing effect by virtue of the enhanced electrostatic repulsion between DNA/DNA strands. As shown in Figure 2 (C), only a minority of antisense strands were observed to gradually attach to the surface during the hybridization process, presumably by physical adsorption, or by partially hybridizing to the bound DNA sense strand. This provides a physically clear picture of PCR product hybridization, in which the interference from the antisense strand is minimized and a relatively constant concentration of free sense strand can thus be maintained.
LOD assessments
The detection limit of PCR products was assessed on the PNA probe layer. Figure 3 presents the hybridization of PCR T2 to a PNA P2 surface over a wide range of concentrations. Each injection of the sample solution was followed by a certain period of time for interaction and then by a surface regeneration step by rinsing the cell with a 10 mM NaOH solution for 1 min. The probe surface demonstrated excellent robustness, sustaining repetitive regeneration for at least 30 times. At such low target concentrations, the initial binding process is limited by the target diffusion from the bulk to the surface, displaying a linear signal increase with time (at a constant sample solution flow). The slope of the fluorescence signal (R) increase is proportional to the analyte concentration c0, given by the following equation:
(2)
where km is the mass-transport rate constant. A calibration curve is hence obtained by plotting the binding slopes versus the analyte concentrations. The LOD is reached if the calibration curve intersects with the horizontal line, given by the baseline stability. The baseline stability is defined as three times the standard deviation of measurements of the time-dependent fluorescence intensity at c0 = 0, which is typically 10 cps min–1 for our present SPFS system. The calibration curve intersects with the baseline deviation level and gives the detection limits of PCR T2 at 100 fmol. This value is quite comparable with that of oligonucleotide target (data not shown), although the PCR product deserves a slower mass-transport rate and hence a poorer detection limit. The unexpected low LOD of the PCR product probably benefits from the largely reduced surface quenching effect. As both theoretically (18) and experimentally (19) demonstrated, the dyes carried by oligonucleotide targets are still subjected to metal-induced surface quenching upon hybridization, as a result of the short sequence length and the limited separation distance (5–6 nm) imposed by the biotin-SA multi-layer structure. In contrast, labeled at the end of the 196-base sense strand of PCR T2, the dye can be spaced away from the metal surface with a considerably larger distance, and thereby deserves a higher fluorescence yield because of less quenching.
Figure 3. (A) Hybridizations of PCR T2 to the PNA P2 surface at 10 mM sodium concentration, over the concentration range of 1–100 pM; (B) calibration curve yielded from (A).
As a comparison, the LOD of PCR products was also assessed on the DNA probe layer. As presented in Figure 4, a sodium concentration of 450 mM was applied in the hybridization of PCR products on the DNA P2 surface in order to overcome the electrostatic barrier of DNA/DNA strands. The extrapolated LOD was 500 fmol, higher than that obtained on the PNA P2 surface (100 fmol). We attribute it to the promoted re-annealing effect in the high salt condition. The re-annealing of PCR products depletes the effective sense DNA strands for surface hybridization and accounts for the ascended LOD concentration. The re-annealing effect was also reflected by the kinetic curves in Figure 4. The fluorescence signals were shown to be equilibrated after the linear increase in a short period of time, as a result of the reduced concentration of the free sense strand. All these again revealed the importance of low salt condition in PCR product hybridization.
Figure 4. (A) Hybridizations of PCR T2 to the DNA P2 surface at 450 mM sodium concentration, over the concentration range of 10–1000 pM; (B) calibration curve yielded from (A).
SNP discrimination
Figure 5 presents the hybridization of PCR T2 and T1 to the PNA P2 surface at 10 mM sodium concentration, respectively. Upon equilibrium, the fluorescence signal obtained by the fully complementary sequence is higher than that with one-base mismatch (A/C mismatch) by a factor of 4.5 (7 by subtracting the signal from the non-selective binding). The binding kinetics suggests that the SNP can be possibly distinguished even after only 1 min of contact time, with the signals differing by a factor of 3.5 with the non-selective binding subtracted (cf. inset of Figure 5). The signal change induced by the one-base substitution occurring in the recognition unit is rather pronounced and reproducibly detected.
Figure 5. Hybridization of T2 (A), T1 (B) or T3 (C) at a concentration of 100 nM to the PNA P2 surface at 10 mM sodium concentration; (D) non-specific binding of PCR P2 on the PNA-11 surface. An attenuator (with an attenuation factor of 14) was applied.
The two-base mismatched situation was checked by the hybridization of PCR T3 with PNA P2, shown in Figure 5C. The resulting hybridization signal was found to be slightly higher than that obtained by the control experiment (cf. Figure 5D), in which PCR T2 was allowed to hybridize with non-complementary PNA-11. This is in accordance with that observed in oligonucleotide hybridization (17), i.e. the affinity of the recognition sequence was substantially reduced by the two-base mismatch and even close to that of the non-specific binding. The non-specific binding of PCR products arises not only from physical adhesion, but also possibly from the partial hybridization of non-selective DNA fragments with the surface probes. It has the tendency to be enhanced with the increase in sequence length and complexity, which, in most cases, can be largely eliminated by performing the hybridization reactions at a higher temperature (30–40°C) (16).
CONCLUSIONS
In this work, SPFS fulfilled the requirements for real-time and in situ analysis of surface hybridization of PCR products with an excellent sensitivity. Working under mass-transport controlled hybridization conditions, the detection limits of PCR products were assessed to be 100 fmol for PNA probes and 500 fmol for DNA probes, respectively. Moreover, SPFS demonstrates rapid and quantitative identification of SNPs, which holds the potential for the development of a new generation of DNA biosensors.
The outstanding sensitivity of the SPFS technique gives new insights into the molecular details of the hybridization behavior of PCR products, which is of significance in fundamental research and largely unexplored so far. In contrast to DNA probes, the neutral backbone of PNA probes allow for the application of a relatively low salt condition. By labeling the sense and/or the antisense strand, it was revealed that only a minority of the antisense strand participated in the surface hybridization, with the low salt condition applied (10 mM Na+). This provides us with a clear hybridization profile, and holds the promise for quantitative interpretation of the corresponding hybridization kinetics in the future.
ACKNOWLEDGEMENTS
This work was supported by an EU grant (QLKI-2000-31658, DNA-track) and by the Deutsche Forschungsgemeinschaft (KN 224/13-1).
REFERENCES
Yu,F., Yao,D. and Knoll,W. ( (2004) ) Oligonucleotide hybridization studied by a surface plasmon diffraction sensor (SPDS). Nucleic Acids Res., , 32, , e75.
Tawa,K. and Knoll,W. ( (2004) ) Mismatching base-pair dependence of the kinetics of DNA–DNA hybridization studied by surface plasmon fluorescence spectroscopy. Nucleic Acids Res., , 32, , 2372–2377.
Wang,J., Nielsen,P.E., Jiang,M., Cai,X., Fernandes,J.R., Grant,D.H., Ozsoz,M., Beglieter,A. and Mowat,M. ( (1997) ) Mismatch sensitive hybridization detection by peptide nucleic acids immobilized on a quartz crystal microbalance. Anal. Chem., , 69, , 5200–5202.
Wang,J. ( (2002) ) Electrochemical nucleic acid biosensors. Anal. Chim. Acta, , 469, , 63–71.
Wolfbeis,O.S. ( (2002) ) Fiber-optic chemical sensors and biosensors. Anal. Chem., , 74, , 2663–2677.
Almadidy,A., Watterson,J., Piunno,P.A.E., Raha,S., Foulds,I.V., Horgen,P.A., Castle,A. and Krull,U. ( (2002) ) Direct selective detection of genomic DNA from coliform using a fiber optic biosensor. Anal. Chim. Acta, , 461, , 37–47.
Weiler,J., Gausepohl,H., Hauser,N., Jensen,O.N. and Hoheisel,J.D. ( (1997) ) Hybridisation based DNA screening on peptide nucleic acid (PNA) oligomer arrays. Nucleic Acids Res., , 25, , 2792–2799.
Peter,C., Meusel,M., Grawe,F., Katerkamp,A., Cammann,K. and B?rchers,T. ( (2001) ) Optical DNA-sensor chip for real-time detection of hybridization events. Fresen. J. Anal. Chem., , 371, , 120–127.
Sawata,S., Kai,E., Ikebukuro,K., Iida,T., Honda,T. and Karube,I. ( (1999) ) Application of peptide nucleic acid to the direct detection of deoxyribonucleic acid amplified by polymerase chain reaction. Biosens. Bioelectron., , 14, , 397–404.
Abel,A.P., Weller,M.G., Duveneck,G.L., Ehrat,M. and Widmer,H.M. ( (1996) ) Fiber-optic evanescent wave biosensor for the detection of oligonucleotides. Anal. Chem., , 68, , 2905–2912.
Ross,P.L., Lee,K. and Belgrader,P. ( (1997) ) Discrimination of single-nucleotide polymorphisms in human DNA using peptide nucleic acid probes detected by MALDI-TOF mass spectrometry. Anal. Chem., , 69, , 4197–4202.
Kai,E., Sawata,S., Ikebukuro,K., Lida,T., Honda,T. and Karube,I. ( (1999) ) Detection of PCR products in solution using surface plasmon resonance. Anal. Chem., , 71, , 796–800.
Tombelli,S., Mascini,M., Braccini,L., Anichini,M. and Turner,A.P.F. ( (2000) ) Coupling of a DNA piezoelectric biosensor and polymerase chain reaction to detect apolipoprotein E polymorphisms. Biosens. Bioelectron., , 15, , 363–370.
Marrazza,G., Tombelli,S., Mascini,M. and Manzoni,A. ( (2001) ) Detection of human apolipoprotein E genotypes by DNA biosensors coupled with PCR. Clin. Chim. Acta, , 307, , 241–248.
Hendrickson,W.A., Pahler,A., Smith,J.L., Satow,Y., Merritt,E.A. and Phizackerley,R.P. ( (1989) ) Crystal-structure of core streptavidin determined from multiwavelength anomalous diffraction of synchrotron radiation. Proc. Natl Acad. Sci. USA, , 85, , 2190–2194.
Weiler,J., Gausepohl,H., Hauser,N., Jensen,O.N. and Hoheisel,J.D. ( (1997) ) Hybridisation based DNA screening on peptide nucleic acid (PNA) oligomer arrays. Nucleic Acids Res., , 25, , 2792–2799.
Liebermann,T., Knoll,W., Sluka,P. and Herrmann,R. ( (2000) ) Complement hybridization from solution to surface-attached probe-oligonucleotides observed by surface-plasmon-field-enhanced fluorescence spectroscopy. Colloids Surf. A, , 169, , 337–350.
Barnes,W.L. ( (1998) ) Fluorescence near interfaces: the role of photonic mode density. J. Mod. Optic., , 45, , 661–699.
Vasilev,K., Knoll,W. and Kreiter,M.J. ( (2004) ) Fluorescence intensities of chromophores in front of a thin metal film. Chem. Phys., , 120, , 3439–3445.(Danfeng Yao1, Fang Yu1, Junyoung Kim1, J)
* To whom correspondence should be addressed. Tel: +49 6131 379 160; Fax: +49 6131 379 360; Email: knoll@mpip-mainz.mpg.de
ABSTRACT
Surface plasmon field-enhanced fluorescence spectroscopy (SPFS) was recently developed for PCR product analysis, which allowed for real-time monitoring of hybridization processes and for the detection of trace amounts of PCR products, with a detection limit of 100 fmol on the peptide nucleic acid (PNA) probe surface, and 500 fmol on the DNA probe surface. By selectively labeling the strands of PCR-amplified DNA, it was shown that the heat denaturation process in combination with the application of low-salt condition substantially reduced the interference from the antisense strands and thus simplified the surface hybridization. Furthermore, SPFS was demonstrated to be capable of quantitatively discriminating the difference induced by single nucleotide substitution, even within one minute of contact time.
INTRODUCTION
The recent development of DNA biosensors with the merits of high sensitivity and selectivity has been accelerated greatly by the achievements in the Human Genome Project (HGP). A number of biosensors, based on surface plasmons (1,2), acoustic waves (3), electrochemistry (4) or fiber optics (5), have demonstrated excellent performances in oligonucleotide analysis by enabling real-time monitoring of hybridization kinetics and mismatch discrimination. However, great challenges still exist for detecting PCR-amplified DNA or genomic DNA. As analytes, PCR products differ from oligonucleotides mainly in two aspects. First, oligonucleotide targets used in most of the biosensing studies are usually <30 bases in length, while PCR products vary from 100 bp to several kb in length. The vast amount of bases is accompanied by a significantly enhanced sequence complexity and imposes tremendous difficulties in PCR product analysis. For instance, the non-selective fragments of PCR products may potentially interfere with the specific hybridization by partially matching with the surface-attached probes or by physical adherence, which usually results in high background signals in DNA biosensors (6) and genechips (7). Steric hindrance arising from the solid support and neighboring bound species is also expected to be enhanced in the presence of non-selective fragments, which considerably decreases the binding efficiency of PCR products. These length-dependent effects are presumed to account for the difficulties in biosensor applications for PCR product detection. Second, the double-stranded nature of PCR products is another major problem, because it renders the recognition unit of the sense strand inaccessible to the surface-attached probes. As a consequence, the sensitivities reported for PCR product detection are so far rather poor (mostly in the nanomolar range) (8,9), which is generally several orders of magnitude lower than that reported for oligonucleotides (in the pico-, or femtomolar range) (10). At present, only end-point results can be offered for most of the reported techniques (11–14). These hinder an in-depth investigation of the hybridization mechanism and quantitative detection of single nucleotide polymorphisms (SNPs).
In this paper, surface plasmon field-enhanced fluorescence spectroscopy (SPFS) is employed for investigating the interfacial hybridization of PCR products, in combination with the application of peptide nucleic acids (PNAs) probes. Several advantages are expected in this study. First, combining the enormous surface plasmon field enhancement with the sensitive fluorescence technique, SPFS could be a powerful tool in detecting SNPs and investigating hybridization kinetics. Second, as mentioned earlier, the presence of the antisense strand may interfere with the hybridization of the sense strand with the surface-attached probes. By selectively labeling the sense or/and antisense DNA strand, the influence of the antisense strand can be possibly explored. Third, decorated at the N-terminus by nine ethylene glycol units and a biotin moiety, the PNA probes are immobilized on a streptavidin (SA) modified gold surface via biotin/SA interaction. According to the molecular dimensions of the SA molecule (4.5 x 4.5 x 5.3 nm3) (15), the bound PNA molecules are laterally spaced. Thus, this surface is desirable to alleviate the steric hindrance induced by neighboring sites to some extent. Finally, the uncharged pseudopeptide backbone of PNA allows for the application of low salt condition (16), in which both the inter- and intra-molecular structure of PCR products can be largely eliminated after the pretreatment of heat denaturation process. In view of these factors, our work will be focused on the quantitative detection of SNPs, elucidating the hybridization phenomenon, and systematic study of the limit-of-detection (LOD) of PCR products by employing SPFS.
MATERIALS AND METHODS
Surface functional multilayer assembly
Figure 1 describes the multilayer architecture employed for the PNA probe assembly. The biotinylated thiols and OH-terminated thiols were synthesized in our group and purchased from Sigma, respectively. A mixed thiol solution was prepared in absolute ethanol with a total concentration of 0.5 mM (0.45 mM OH-terminated thiol and 0.05 mM biotinylated thiol). The gold substrates were immersed in the thiol solution for >6 h at room temperature. They were then subjected to repeated rinsing in ethanol. After being dried with a stream of nitrogen gas, the substrates can be stored in nitrogen at 4°C for several days.
Figure 1. (A) Chemical structures of biotinylated thiol and OH-terminated thiol. (B) A schematic representation of the sensor surface architecture: a binary SAM composed of the biotinylated thiol and OH-terminated thiol is formed on the gold surface. A monolayer of SA tetramer is attached by the interaction with the biotin moiety, exposing the binding sites for biotinylated PNA or DNA probe assembly.
SA was purchased from Sigma. Exposing the binary thiol modified gold substrates to a 1 μM SA solution, a monolayer of the tetrameric protein was attached within minutes via the interaction with the biotin moiety. The biotinylated PNA P2, PNA-11 (synthesized in our group), or DNA P2 (MWG-Biotech) can then be assembled onto the SA layer (cf. Table 1). The buffer condition used in this work is 10 mM phosphate buffer (PB, pH 7.4), containing 0.005% Tween-20 and 2 mM EDTA. High sodium concentrations were obtained by adding sodium chloride.
Table 1. The sequences of all of the PNA and DNA samples used in this work
PCR amplification
The PCR protocol was optimized on a thermocycler (Biometra) by designing the primer sequences and adjusting the annealing temperature and concentration of each component. Recombinant plasmids (4 kb) were used as templates for the PCR amplification. The forward and reverse primers were purchased from MWG-Biotech. As listed in Table 1, both are 30-base oligonucleotides with both calculated melting temperatures being close to each other. Unlabeled or labeled by Cy5 at the 5' end, primers were employed as needed. Unless otherwise stated, the Cy5-labeled forward primer and unlabeled reverse primer were used routinely for amplifying the PCR products with only the sense strand labeled.
All PCRs were carried out under the same conditions. Reagents for each 50 μl reaction volume included: 5 U of Taq polymerase (Amersham Bioscicences), 1x PCR buffer (Amersham Biosciences), 60 pmol of the forward primer and 80 pmol of reverse primer, 0.2 mM dNTPs (Fermentas) and 100 ng of the plasmid DNA. After an initial denaturation step at 96°C for 1 min, each of the 30 cycles of amplification consisted of 30 s of template denaturation at 96°C, 30 s of primer annealing at 50°C and 30 s of primer extension at 72°C.
Preparation of the analytes
The resulting 196 bp PCR products (T1, T2 and T3, cf. Table 1) were identified by agarose electrophoresis (2%). As shown in Table 1, the three PCR products contain a 15-base recognition sequence present in the middle of the sense strand approximately, which fully matches or exhibits a one-base/two-base mismatch with PNA P2, respectively. Ethanol precipitation was performed afterwards in order to remove the salts, dNTPs, excess primers and enzyme from the PCR sample. For 50 μl of PCR sample, 125 μl ethanol (Sigma) and 5 μl of sodium acetate (3 M) were added. The mixture was shaken, and then kept at –20°C for >6 h to induce precipitation. The PCR products were collected by centrifugation at 16 000 g for 0.5 h. The ethanol precipitation was empirically examined to effectively purify the PCR products, even at a large sample scale.
Since all the PCR products used in this paper are end-labeled by Cy5 in the sense or/and antisense strand, their concentrations can be determined by recording the adsorption of Cy5 at 650 nm by the following equation:
(1)
assuming a molar extinction coefficient of = 250 000 M–1 cm–1, as provided by the manufacturer (Amersham Biosciences).
The PCR products underwent the following denaturation process prior to the biosensing studies, unless otherwise stated: the heat denaturation is composed of three steps, heating the DNA sample (in 10 mM PB buffer) at 96°C for 10 min, quickly quenching it to 0°C in an ice–water mixture and returning to room temperature.
Instrumental
The SPFS set-up was constructed based on a conventional surface plasmon spectrometer, in combination with the photon counting technique (17). The emitted fluorescence light from the sample surface was collected through a quartz flow cell. A lens (f = 40 mm; Owis) focused the light through an interference filter ( = 670 nm, T = 69%, LOT) onto a photomultiplier tube (Hamamatsu), which fed the signal to a photon counter (Hewlett Packard). This fluorescence detection unit was mounted on the sample arm, such that its position was fixed relative to the sample. The kinetic measurements were realized by setting the incident angle at the resonant angle and recording the fluorescence intensity as a function of time.
RESULTS AND DISCUSSION
Role of the antisense strand in surface hybridization
In order to elucidate the role of the antisense strand in hybridization, the three PCR products were prepared, i.e. the sense strand, antisense strand and both strands labeled PCR T2. They were obtained by having the forward primer, the reverse primer or both primers labeled, respectively. After going through the heat denaturation process, the three PCR T2 (each concentration c0 =100 nM) were applied for hybridization with the surface-attached PNA P2 at 10 mM sodium concentration, monitored by SPFS. As shown in Figure 2 , the hybridization signal obtained by both strands labeled PCR T2 is slightly higher than that obtained with the sense strand labeled, which is in accordance with the weak fluorescence signal induced by the antisense strand .
Figure 2. Hybridizations of PCR T2 (100 nM) with both strands labeled (A), only the sense strand labeled (B) or only the antisense strand labeled (C) on the PNA P2 surface at 10 mM sodium concentration. An attenuator (with an attenuation factor of 14) was applied to reduce the fluorescence intensity.
As known to us, the heat denaturation process consists of three temperature states, heating at 96°C, quickly quenching to 0°C and returning to room temperature. Empirically, the double strands melt to give single-stranded DNA if heated at 96°C for several minutes. The subsequent quenching step is expected to preserve the single-stranded structure to the largest extent, by quickly running through the temperature range favorite for re-annealing (20 to 70°C). Nevertheless, the double strands tend to re-associate with the temperature returning to room temperature, and the re-association is largely dependent on the bulk salt concentration. Obviously, a lower salt concentration is desirable to inhibit the re-annealing effect by virtue of the enhanced electrostatic repulsion between DNA/DNA strands. As shown in Figure 2 (C), only a minority of antisense strands were observed to gradually attach to the surface during the hybridization process, presumably by physical adsorption, or by partially hybridizing to the bound DNA sense strand. This provides a physically clear picture of PCR product hybridization, in which the interference from the antisense strand is minimized and a relatively constant concentration of free sense strand can thus be maintained.
LOD assessments
The detection limit of PCR products was assessed on the PNA probe layer. Figure 3 presents the hybridization of PCR T2 to a PNA P2 surface over a wide range of concentrations. Each injection of the sample solution was followed by a certain period of time for interaction and then by a surface regeneration step by rinsing the cell with a 10 mM NaOH solution for 1 min. The probe surface demonstrated excellent robustness, sustaining repetitive regeneration for at least 30 times. At such low target concentrations, the initial binding process is limited by the target diffusion from the bulk to the surface, displaying a linear signal increase with time (at a constant sample solution flow). The slope of the fluorescence signal (R) increase is proportional to the analyte concentration c0, given by the following equation:
(2)
where km is the mass-transport rate constant. A calibration curve is hence obtained by plotting the binding slopes versus the analyte concentrations. The LOD is reached if the calibration curve intersects with the horizontal line, given by the baseline stability. The baseline stability is defined as three times the standard deviation of measurements of the time-dependent fluorescence intensity at c0 = 0, which is typically 10 cps min–1 for our present SPFS system. The calibration curve intersects with the baseline deviation level and gives the detection limits of PCR T2 at 100 fmol. This value is quite comparable with that of oligonucleotide target (data not shown), although the PCR product deserves a slower mass-transport rate and hence a poorer detection limit. The unexpected low LOD of the PCR product probably benefits from the largely reduced surface quenching effect. As both theoretically (18) and experimentally (19) demonstrated, the dyes carried by oligonucleotide targets are still subjected to metal-induced surface quenching upon hybridization, as a result of the short sequence length and the limited separation distance (5–6 nm) imposed by the biotin-SA multi-layer structure. In contrast, labeled at the end of the 196-base sense strand of PCR T2, the dye can be spaced away from the metal surface with a considerably larger distance, and thereby deserves a higher fluorescence yield because of less quenching.
Figure 3. (A) Hybridizations of PCR T2 to the PNA P2 surface at 10 mM sodium concentration, over the concentration range of 1–100 pM; (B) calibration curve yielded from (A).
As a comparison, the LOD of PCR products was also assessed on the DNA probe layer. As presented in Figure 4, a sodium concentration of 450 mM was applied in the hybridization of PCR products on the DNA P2 surface in order to overcome the electrostatic barrier of DNA/DNA strands. The extrapolated LOD was 500 fmol, higher than that obtained on the PNA P2 surface (100 fmol). We attribute it to the promoted re-annealing effect in the high salt condition. The re-annealing of PCR products depletes the effective sense DNA strands for surface hybridization and accounts for the ascended LOD concentration. The re-annealing effect was also reflected by the kinetic curves in Figure 4. The fluorescence signals were shown to be equilibrated after the linear increase in a short period of time, as a result of the reduced concentration of the free sense strand. All these again revealed the importance of low salt condition in PCR product hybridization.
Figure 4. (A) Hybridizations of PCR T2 to the DNA P2 surface at 450 mM sodium concentration, over the concentration range of 10–1000 pM; (B) calibration curve yielded from (A).
SNP discrimination
Figure 5 presents the hybridization of PCR T2 and T1 to the PNA P2 surface at 10 mM sodium concentration, respectively. Upon equilibrium, the fluorescence signal obtained by the fully complementary sequence is higher than that with one-base mismatch (A/C mismatch) by a factor of 4.5 (7 by subtracting the signal from the non-selective binding). The binding kinetics suggests that the SNP can be possibly distinguished even after only 1 min of contact time, with the signals differing by a factor of 3.5 with the non-selective binding subtracted (cf. inset of Figure 5). The signal change induced by the one-base substitution occurring in the recognition unit is rather pronounced and reproducibly detected.
Figure 5. Hybridization of T2 (A), T1 (B) or T3 (C) at a concentration of 100 nM to the PNA P2 surface at 10 mM sodium concentration; (D) non-specific binding of PCR P2 on the PNA-11 surface. An attenuator (with an attenuation factor of 14) was applied.
The two-base mismatched situation was checked by the hybridization of PCR T3 with PNA P2, shown in Figure 5C. The resulting hybridization signal was found to be slightly higher than that obtained by the control experiment (cf. Figure 5D), in which PCR T2 was allowed to hybridize with non-complementary PNA-11. This is in accordance with that observed in oligonucleotide hybridization (17), i.e. the affinity of the recognition sequence was substantially reduced by the two-base mismatch and even close to that of the non-specific binding. The non-specific binding of PCR products arises not only from physical adhesion, but also possibly from the partial hybridization of non-selective DNA fragments with the surface probes. It has the tendency to be enhanced with the increase in sequence length and complexity, which, in most cases, can be largely eliminated by performing the hybridization reactions at a higher temperature (30–40°C) (16).
CONCLUSIONS
In this work, SPFS fulfilled the requirements for real-time and in situ analysis of surface hybridization of PCR products with an excellent sensitivity. Working under mass-transport controlled hybridization conditions, the detection limits of PCR products were assessed to be 100 fmol for PNA probes and 500 fmol for DNA probes, respectively. Moreover, SPFS demonstrates rapid and quantitative identification of SNPs, which holds the potential for the development of a new generation of DNA biosensors.
The outstanding sensitivity of the SPFS technique gives new insights into the molecular details of the hybridization behavior of PCR products, which is of significance in fundamental research and largely unexplored so far. In contrast to DNA probes, the neutral backbone of PNA probes allow for the application of a relatively low salt condition. By labeling the sense and/or the antisense strand, it was revealed that only a minority of the antisense strand participated in the surface hybridization, with the low salt condition applied (10 mM Na+). This provides us with a clear hybridization profile, and holds the promise for quantitative interpretation of the corresponding hybridization kinetics in the future.
ACKNOWLEDGEMENTS
This work was supported by an EU grant (QLKI-2000-31658, DNA-track) and by the Deutsche Forschungsgemeinschaft (KN 224/13-1).
REFERENCES
Yu,F., Yao,D. and Knoll,W. ( (2004) ) Oligonucleotide hybridization studied by a surface plasmon diffraction sensor (SPDS). Nucleic Acids Res., , 32, , e75.
Tawa,K. and Knoll,W. ( (2004) ) Mismatching base-pair dependence of the kinetics of DNA–DNA hybridization studied by surface plasmon fluorescence spectroscopy. Nucleic Acids Res., , 32, , 2372–2377.
Wang,J., Nielsen,P.E., Jiang,M., Cai,X., Fernandes,J.R., Grant,D.H., Ozsoz,M., Beglieter,A. and Mowat,M. ( (1997) ) Mismatch sensitive hybridization detection by peptide nucleic acids immobilized on a quartz crystal microbalance. Anal. Chem., , 69, , 5200–5202.
Wang,J. ( (2002) ) Electrochemical nucleic acid biosensors. Anal. Chim. Acta, , 469, , 63–71.
Wolfbeis,O.S. ( (2002) ) Fiber-optic chemical sensors and biosensors. Anal. Chem., , 74, , 2663–2677.
Almadidy,A., Watterson,J., Piunno,P.A.E., Raha,S., Foulds,I.V., Horgen,P.A., Castle,A. and Krull,U. ( (2002) ) Direct selective detection of genomic DNA from coliform using a fiber optic biosensor. Anal. Chim. Acta, , 461, , 37–47.
Weiler,J., Gausepohl,H., Hauser,N., Jensen,O.N. and Hoheisel,J.D. ( (1997) ) Hybridisation based DNA screening on peptide nucleic acid (PNA) oligomer arrays. Nucleic Acids Res., , 25, , 2792–2799.
Peter,C., Meusel,M., Grawe,F., Katerkamp,A., Cammann,K. and B?rchers,T. ( (2001) ) Optical DNA-sensor chip for real-time detection of hybridization events. Fresen. J. Anal. Chem., , 371, , 120–127.
Sawata,S., Kai,E., Ikebukuro,K., Iida,T., Honda,T. and Karube,I. ( (1999) ) Application of peptide nucleic acid to the direct detection of deoxyribonucleic acid amplified by polymerase chain reaction. Biosens. Bioelectron., , 14, , 397–404.
Abel,A.P., Weller,M.G., Duveneck,G.L., Ehrat,M. and Widmer,H.M. ( (1996) ) Fiber-optic evanescent wave biosensor for the detection of oligonucleotides. Anal. Chem., , 68, , 2905–2912.
Ross,P.L., Lee,K. and Belgrader,P. ( (1997) ) Discrimination of single-nucleotide polymorphisms in human DNA using peptide nucleic acid probes detected by MALDI-TOF mass spectrometry. Anal. Chem., , 69, , 4197–4202.
Kai,E., Sawata,S., Ikebukuro,K., Lida,T., Honda,T. and Karube,I. ( (1999) ) Detection of PCR products in solution using surface plasmon resonance. Anal. Chem., , 71, , 796–800.
Tombelli,S., Mascini,M., Braccini,L., Anichini,M. and Turner,A.P.F. ( (2000) ) Coupling of a DNA piezoelectric biosensor and polymerase chain reaction to detect apolipoprotein E polymorphisms. Biosens. Bioelectron., , 15, , 363–370.
Marrazza,G., Tombelli,S., Mascini,M. and Manzoni,A. ( (2001) ) Detection of human apolipoprotein E genotypes by DNA biosensors coupled with PCR. Clin. Chim. Acta, , 307, , 241–248.
Hendrickson,W.A., Pahler,A., Smith,J.L., Satow,Y., Merritt,E.A. and Phizackerley,R.P. ( (1989) ) Crystal-structure of core streptavidin determined from multiwavelength anomalous diffraction of synchrotron radiation. Proc. Natl Acad. Sci. USA, , 85, , 2190–2194.
Weiler,J., Gausepohl,H., Hauser,N., Jensen,O.N. and Hoheisel,J.D. ( (1997) ) Hybridisation based DNA screening on peptide nucleic acid (PNA) oligomer arrays. Nucleic Acids Res., , 25, , 2792–2799.
Liebermann,T., Knoll,W., Sluka,P. and Herrmann,R. ( (2000) ) Complement hybridization from solution to surface-attached probe-oligonucleotides observed by surface-plasmon-field-enhanced fluorescence spectroscopy. Colloids Surf. A, , 169, , 337–350.
Barnes,W.L. ( (1998) ) Fluorescence near interfaces: the role of photonic mode density. J. Mod. Optic., , 45, , 661–699.
Vasilev,K., Knoll,W. and Kreiter,M.J. ( (2004) ) Fluorescence intensities of chromophores in front of a thin metal film. Chem. Phys., , 120, , 3439–3445.(Danfeng Yao1, Fang Yu1, Junyoung Kim1, J)