Investigating the kinetics of DNA–DNA and PNA–DNA interactions using surface plasmon resonance-enhanced fluorescence spectroscopy

https://doi.org/10.1016/S0956-5663(01)00239-1Get rights and content

Abstract

Plasmon surface polaritons, resonantly excited in the Kretschmann format, are used to enhance the fluorescence emission of chromophore-labeled oligonucleotides (15mers) binding to surface-attached (via biotin–streptavidin linkages) complement catcher probes. A detailed analysis of the association and dissociation kinetics as well as the affinity constants is given for a mismatch 1 hybrid, emphasizing, in particular, the experimental conditions that are required to allow for an artifact-free determination of rate constants. A first comparison between DNA- and peptide nucleic acid (PNA-) probes shows similar affinities, however, significant deviations from single-exponential kinetics predicted by a simple Langmuir model for the PNA case are found.

Introduction

With the introduction of the DNA-chip (Service, 1988) the interest in surface hybridization reactions or, more precisely, the interest in the detection of association and dissociation processes between surface-attached single-stranded catcher probe oligonucleotides and complement target strands from solution, has steeply increased. Two key issues have to be addressed and solved, (1) what is the best strategy to design and build the interfacial architecture of the probe oligonucleotide layer allowing for a sensitive monitoring of the hybridization events and (2) what is the best detection method. ‘Best surface layer’ can mean several things: the surface functionalization of the sensor surface by probe oligonucleotides should lead to a rapid hybridization reaction with high efficiency while minimizing non-specific adsorption (Brockman et al., 1999); the reaction should be highly sequence specific, because in the extreme case, a single mismatch needs to be detected (Kelley et al., 1999), and the fabrication of the active surface has to be compatible with (massive) parallel read-out schemes by allowing for the preparation of large arrays of different sensor pads, each with its own specific functionality (Fodor et al., 1993, Zizlsperger and Knoll, 1998). ‘Best detection method’, in general, would be a label-free technique monitoring the hybridization directly, e.g. by the corresponding mass deposition (Okahata et al., 1998) to or change in optical thickness at the sensor surface (Piscevic et al., 1995). However, the current formats, to a large extent, lack the required sensitivity and/or yield only qualitative data.

Given the tremendous interest in this technology and the resulting large number of papers that deal with different aspects of surface hybridization reactions, it is rather surprising that reliable quantitative data describing the interfacial association and dissociation steps by a kinetic model with the corresponding rate constants are barely available. Moreover, in cases where rate constants were measured it was not clear as to what extent the mere presence of the interface, as well as, the chemical nature of the interfacial binding layer, e.g. the chemistry used to attach the catcher probes, or the general supramolecular architecture of the binding matrix, influenced those rate constants (Gotoh et al., 1995). Currently, there exist no models and kinetic parameters of surface hybridization reactions that would allow for a detailed comparison with the corresponding bulk hybridization data.

We started a project that aims at filling this gap by providing kinetic and equilibrium (affinity) data that (1) should lead to a better understanding of the basic mechanisms underlying surface hybridization reactions; that (2) will allow us to develop theoretical models describing quantitatively the observed experimental findings; and (3) should give us the predictive power to optimize different catcher architectures for specific sensor formats and detection purposes.

In a first account, we reported on the kinetics and affinity constants of hybridization reactions between surface-attached oligonucleotides, 15mers of a specific sequence coupled to a streptavidin monolayer at the sensor surface via a biotinylated 15mer thymine spacer, and complement strands from solution (Liebermann et al., 2000). The detection method that we used was a recently developed combination of surface plasmon optics and fluorescence spectroscopy (Liebermann and Knoll, 2000). The target sequences are labeled with fluorescent probes the emission of which is observed by exciting the chromophores with the highly enhanced optical field of a resonantly coupled surface plasmon mode. We demonstrated that the observed data with an excellent signal-to-noise ratio allowed for a quantitative determination of the kinetic parameters of a simple Langmuir adsorption model, i.e. the rate constants kon and koff, and presented evidence for the very high discrimination against single or double base-mismatches in the target 15mer oligonucleotides.

Here, we focus on details of how to run the kinetic experiment in order to obtain reliable rate constants, check for reproducibility, give a comprehensive analysis of experiments with a DNA hybrid double strand representing a mismatch 1 (MM1) situation, and give first data with a peptide nucleic acid (PNA) catcher probe (Jensen et al., 1997).

Section snippets

Experimental

The instrument that we used for the experiments is schematically shown in Fig. 1. Since details of this surface plasmon/fluorescence combination set-up were reported recently (Liebermann and Knoll, 2000), we only give a very brief summary of the fundamental principles. A p-polarized HeNe-laser beam is reflected off the Au-coated (50 nm) base of a 90° glass prism (n=1.846@λ=633 nm) and monitored with a photodiode (Knoll, 1998). The angular control of this θ–2θ Kretschmann configuration is

Results and discussion

The first set of investigations was aimed at documenting the reproducibility of the experiments when performed in an identical way but with different sample preparations. To this end, four different catcher probe layers were prepared as described above (cf. Fig. 2a) starting from four different Au-coated substrates. Each sample was then mounted to the spectrometer, and the fluorescence intensity was recorded (@θ=56°) as a function of time after switching from rinsing pure buffer solution

Conclusions

We have demonstrated that the combination of fluorescence detection schemes with the resonant coupling of surface plasmons as the excitation ‘light source’ at a metal/dielectric interface, i.e. of the sensor surface in contact with the analyte solution, allows for a very sensitive monitoring of the hybridization of fluorophore-labeled complement target oligonucleotides with their complementary probe strands at the surface. We have shown that this way, the kinetic rate constants of association

Acknowledgements

Helpful discussions with P. Sluka and R. Herrmann are gratefully acknowledged. This work was partly supported by a EU-Project (BMH4-CT96-1081).

References (14)

  • W. Knoll et al.

    Colloids Surf. A

    (2000)
  • T. Liebermann et al.

    Colloids Surf. A

    (2000)
  • T. Liebermann et al.

    Colloids Surf. A

    (2000)
  • B. Persson et al.

    Anal. Biochem.

    (1997)
  • D. Piscevic et al.

    Appl. Surf. Sci.

    (1995)
  • J.M. Brockman et al.

    J. Am. Chem. Soc.

    (1999)
  • S.P.A. Fodor et al.

    Nature

    (1993)
There are more references available in the full text version of this article.

Cited by (90)

  • Overview of Nanolayers: Formulation and Characterization Methods

    2017, Nanolayer Research: Methodology and Technology for Green Chemistry
  • Fischer carbene mediated covalent grafting of a peptide nucleic acid on gold surfaces and IR optical detection of DNA hybridization with a transition metalcarbonyl label

    2016, Applied Surface Science
    Citation Excerpt :

    Formation of monolayers of PNA strands onto gold surfaces has been previously achieved by introducing a cysteine residue (i.e. a sulfhydryl functionality) at one extremity of the sequence (generally the pseudo 5′ end or N-terminus) during solid-phase synthesis followed by direct chemisorption to the surface [22–36]. Alternatively, PNA strands have been immobilized by affinity via the biotin/avidin system [35,37–40] or by conjugation to carboxyl-terminated SAMs [41,42]. We have also shown that appending four lysine residues to the C-terminus of the homothymine decamer enabled the formation of a dense and stable layer of PNA molecules at the surface of a gold electrode nanostructured with gold nanoparticles by formation of Au-N bonds [43,44] For this work, we devised a more general immobilization strategy making use of the single primary amine function located at the pseudo 5′ terminus (NH2-terminus) of the PNA molecule and of an appropriate bifunctional linker able to connect the PNA t10 to the Au surface [36] (Scheme 1).

  • Real-time PCR detection chemistry

    2015, Clinica Chimica Acta
View all citing articles on Scopus
View full text