Electrostatic readout of DNA microarrays with charged microspheres - Nature Biotechnology

Summary:
Paper describing a method for label-free electrostatic readout of DNA or RNA hybridization on microarrays which is based on the electrostatic properties of microarrays. Changes in surface charge density as a result of specific hybridization can be detected and are measured from the position and motion of charged microspheres randomly dispersed over the surface (in 100uM NaCl). Interactions between the microspheres and substrate can be imaged by a variety of optical methods that provide a rapid indicator of DNA hybridization. The naked eye is sufficient to read out the hybridization, which may facilitate broad application of multiplexed assays. Practical diagnostics require rapid and simple quantitative readouts that do not use dedicated instrumentation or intensive image processing. (Published: 29/06/08)

Notes:

• microarray assays
  
• typically rely on fluorescence detection, which requires time-consuming chemical labeling, reverse transcription, high-power excitation sources and sophisticated instrumentation for scanning
• consequently, microarray assays tend to be performed by dedicated centers rather than by individual laboratories, and not by clinics in developing countries
• alternative label-free DNA detection techniques
• e.g. surface plasmon resonance, electrochemical sensing, fluorescent polymers,
  atomic force microscopy, microcantilevers and electronic depletion of a field effect transistor
• none of these have gained widespread use because each requires
  
• either complex device fabrication
  
• or sophisticated instrumentation for readout
• additionally, none are compatible with conventional DNA microarrays where up to 10^6 sequences can be interrogated in a single experiment
  
• electrostatic-based DNA or RNA detection method
• Complementary oligonucleotide binding strongly affects the electrostatic charge of the surface due to the negatively charged DNA phosphate backbone
• hybridization is measured electrostatically using charged microspheres that are highly responsive to changes in charge density on the microarrayed surface
• Interactions between the microspheres and substrate can be imaged by a variety of optical methods that provide a rapid indicator of DNA hybridization
• role of each silica microsphere is analogous to that of an electrostatic force microscope (EFM) tip where the vertical deflection of the tip is used to report local electrostatic surface properties
• EFM, however, is a serial technique that is practically limited to a field of view of 100 um2
• particle-based technique described here is capable of parallel sampling of a microarray surface over centimeter-length scales
• glass support is positively biased using an aminosilane modification
• balances the negative charge contributed by both the glass surface and the printed single strand (ss)DNA molecules
• necessary because the charged microparticles are responsive to a limited range of surface charge densities
• typical experiment
• a prepared substrate is mounted in a well chamber and hybridized
• unlabeled and negatively charged 5.6 mm–diameter silica microparticles are then added and allowed to sediment above (or otherwise interact with) the array over a period of 20 min.
  
• Microspheres uniformly distribute across the entire surface and adsorb to the positively charged background.
  
• However, over sufficiently negatively charged areas, they adopt an equilibrium height that is dictated by a balance between gravitational and local electrostatic forces.
  
• To determine the precise heights and positions of the population of levitated microspheres, we then acquired a collection of dualwavelength reflection interference contrast microscopy (RICM) images covering the entire array area
• Image acquisition was automated using a motorized translation stage, and a softwaredriven autofocus routine.
  
• At each stage position, 20 images were acquired (0.4 fps) yielding 20,000 images/mm2.
  
• On average, there were 20 microspheres per field of view (30 x 30 um) resulting in 400,000 microparticle observations/mm2.
  
• The interference images corresponding to individual microspheres were used to determine their position with 1-nm vertical resolution and 16-nm lateral resolution.
  
• Although this is an optical technique, the resolution is not diffraction limited
• it is determined by the particle position resolution and the density of particle observations.
• A quantitative spatial map of the surface charge density is generated by compiling the set of three-dimensional (3D) particle position measurements.
• practical diagnostics require rapid and simple quantitative readouts that do not use dedicated instrumentation or intensive image processing
• To develop such a readout strategy we take advantage of the fact that silica microspheres respond to the surface charge in an easily observable manner.
  
• If the surface is negatively charged, microparticles remain laterally mobile
  
• can be easily visualized by monitoring the intensity variance in a time series of brightfield images.
  
• Alternatively, positively charged areas can be identified by the presence of electrostatically adhered microspheres.
  
• This can be rapidly imaged using darkfield or brightfield microscopy.
  
• Therefore, adhesion of charged particles provides a simple test to map the sign of the surface charge.
• Direct comparison between fluorescence and electrostatic detection on the same substrates, under identical conditions reveals comparable figures of merit, indicating that sensitivity is primarily limited by hybridization, not the readout
• This particle-based electrostatic detection method offers multiple advantages over existing microarray detection methods.
  
• First, expression profiling and SNP detection using primary mRNA can now be performed without reverse transcription and fluorescent tagging.
  
• Second, electrostatic detection is compatible with conventional microarrays as well as unconventional arrays fabricated on injection-molded plastic or embedded within microfluidic architectures
• Third, imaging by pixel-by-pixel statistical averaging of colloidal particle positions is not diffraction limited, which suggests a strategy to take advantage of DNA nanoarrays that can hold 10^4 more features per unit area than conventional microarrays.
• Finally, microarrays consisting of proteins, small molecules, polymers and heterogeneous catalysts are rapidly coming online and might benefit from the electrostatic readout platform we describe herein.