This study is focused on the production of DNA-silver nanoparticle conjugates for applications in e.g. plasmonic biosensors and molecular electronics. Most research in DNA-nanoparticle conjugates is focused on the use of gold nanoparticles as these are more stable than silver nanoparticles. However, silver has a higher extinction coefficient and a sharper localized surface plasmon resonance than gold, which may provide higher sensitivity in plasmonic sensors. Additionally, the conductivity of silver is higher than that of gold, and this makes it interesting from a molecular electronics point of view.
Here we report a method for the construction of DNA-silver nanoparticle conjugates in only 2 hours. The binding of DNA to Ag nanoparticles is based on the specific interaction between phosphorothioate groups on the DNA and silver. In this respect, phosphorothioated single stranded DNA and hairpin structured DNA modified with phosphorothioate groups in the loop region is used. Single stranded random sequence DNA is used as a reference in order to be able to distinguish phosphorothioate binding from non-specific binding. Furthermore, the binding of phosphorothioated ATP, with ATP used as a reference, is examined in order to determine its stabilizing effect.
Absorbance measurements and gel electrophoresis experiments showed that phosphorothioated single stranded DNA provided the largest stabilizing effect for the silver nanoparticles. Hairpin structured DNA with a phosphorothioated loop region provided the second highest stabilization, and single stranded DNA without phosphorothioate groups provided the smallest stabilizing effect. Hereby the specific interaction between phosphorothioate groups and silver nanoparticles was confirmed. In contrary to what was observed with DNA, the presence of a phosphorothioate group on ATP did not improve the stabilizing ability of the molecule, since aggregation was induced instead.
As a second part of this study we compare the sensitivity of surface plasmon polariton based surface plasmon resonance biosensors. The sensors involve the DNA binding peptide indolicidin-4 and either double stranded DNA linked to a gold film or double stranded DNA-silver nanoparticle conjugates linked to a gold film. The presence of the nanoparticles is envisaged to enhance the sensor sensitivity. It is expected that the binding of indolicidin-4 to the DNA will result in a change of the DNA conformation, and that this will change the nanoparticle-film distance when the indolicidin-4 is bound to the conjugates. The distance change is expected to be visible as scattering of the gold film surface plasmon polariton by the nanoparticle is distance dependent. The linking of double stranded DNA and the conjugates to the gold film was confirmed by surface plasmon resonance spectroscopy, and furthermore the linking of the conjugates was confirmed by atomic force microscopy. With the two sensors a larger shift of the resonance angle was observed at indolicidin-4 binding to double stranded DNA compared to binding to double stranded DNA-silver nanoparticle conjugates. Furthermore, the expected change in surface plasmon polariton scattering at binding to the conjugates was not clearly observed. However, association and dissociation rate constants of 432M-1s-1 and 0.00002s-1 for the binding of indolicidin-4 to double stranded DNA were obtained, and to our knowledge this is the first time kinetic constants for this interaction has been reported.