STEDy Improvement

Just as nanofluidics has stolen some of the limelight from micro-fluidics, advances in nanoscopy are beginning to make standard microscopy seem a bit passe. One of the first techniques described for optical imaging at the nanoscale is stimulated emission depletion microscopy, or STED, which couples excitation light with a quenching beam in a surrounding ring. The result is a smaller fluorescing spot, leading to so-called super-resolution optical microscopy. While the technology is impressive, the tools leave a bit to be desired. Unlike the array of fluorophores available for traditional confocal microscopy, the options for STED microscopy have generally been limited to far-red organic dyes. Recently, a HaloTag-based labeling approach has been described that is compatible with STED microscopy, but the fluorescent ligand is cell impermeant and thus restricted in its potential uses. Another class of fluorophore-binding factors are the fluorogen activating proteins (FAPs), which bind (and thereby render fluorescent) the far-red dye malachite green. In an article in Bioconjugate Chemistry, Fitzpatrick et al. show that an engineered variant of this system is ideally suited to live-cell imaging using STED microscopy. Resolution is similar to that obtained when using organic dyes with STED, but the genetically encoded FAP enables the types of protein-labeling studies commonly performed in traditional fluorescence microscopy. Together with a new far-red variant of DsRed, which is described in a recently published Biochemistry article, the FAP–malachite green labeling method should accelerate the use of STED nanoscopy for dynamic imaging of subcellular components in living cells.

Actin as visualized by confocal microscopy (left) and STED nanoscopy (right). Reprinted with permission © 2009 ACS. (Click to enlarge)

Fitzpatrick et al. 2009. STED nanoscopy in living cells using fluorogen activating proteins. Bioconjugate Chem. 20:1843–1847.

Diversity Matters

There is seemingly endless variety in protein detection platforms these days, but the options contract sharply when it comes to acceptable sample types. Biomolecules that autofluoresce or absorb at inconvenient wavelengths aren't welcome in many ELISAs or protein microarrays. By the same token, if a sample strays outside narrow pH, ionic strength, or temperature constraints, electrochemical or microcantilever-based techniques are off limits. In a technical report appearing in Nature Medicine, Gaster et al. describe a detection scheme intended to be as open as possible to diverse sample conditions. The assay mirrors a sandwich ELISA, but instead of a colorimetric or other matrix-sensitive detection method, readout is by magnetoresistive sensors. The components of the assay include a capture antibody immobilized on a magneto-nano sensor chip, a biotinylated detection antibody, and a streptavidin-conjugated magnetic nanoparticle tag. Magnetic fields induced in the latter can be detected in real time at the sensor surface. Because biological materials do not generate detectable magnetic fields, signal generation and capture are insensitive to matrix-mediated interference. In addition, the sensor does not waver even in the face of vast pH (4–10) and temperature shifts (−20–48°C), and the optical properties of the sample have no effect on detection. Side-by-side comparison with traditional ELISA showed a compelling advantage in linear dynamic range (6 versus 2 orders of magnitude) and a 1000-fold boost in sensitivity. The lack of matrix interference was demonstrated by equal detection sensitivity whether test biomarkers were spiked into simple buffers or complex solutions such as serum and urine. Signals in saliva were noticeably lower, which may be a consequence of protease degradation or sample viscosity. Data from ongoing monitoring of serum samples for tumor markers in a mouse model of colorectal cancer are shown as one application, but the technology is as appropriate for basic research as for clinical diagnostics.

Gaster et al. 2009. Matrix-insensitive protein assays push the limits of biosensors in medicine. Nat Med. 15:1327–1332.

Expressing Our Synbodies

With the growing interest in the field of proteomics, there is a need to generate more protein affinity reagents than is possible using standard antibody approaches, which tend to be costly and time-consuming due in large measure to the need for animal immunization. While alternative binders based on engineered immunoglobulin domains, protein scaffolds, and aptamers can be easier to obtain, they still require multiple rounds of time- and labor-intensive in vitro selection and amplification, steps that are not amenable to high-throughput methodologies. In addition, because alternative binding reagents based on small molecule ligands are commonly selected from combinatorial libraries, they typically have only moderate binding affinities for a target protein when compared to antibodies. Increased binding affinity can be achieved by attaching two or more ligands to a tethering molecule to create a higher affinity, multivalent binding reagent. In a recent issue of the Journal of the American Chemical Society, Williams et al. describe a novel strategy for the scalable production of such synthetic antibodies—which they term synbodies—by tethering two peptide ligands into a single high-affinity bivalent protein binding reagent using DNA. The authors first used peptide microarrays to isolate moderate affinity peptide binders to a specific target protein. Two peptides that bind to different sites on the target protein were then tethered to a short DNA oligonucleotide by amine coupling to modified bases at various positions, with each peptide on a separate strand. The optimal distance and orientation of the two peptides for generating the strongest binding synbody was determined using surface plasmon resonance (SPR) analysis where all possible pairs (both hetero- and homo-) of peptides attached to the DNA at a variety of locations were screened on a chip. Using the yeast Gal80 regulatory protein and human transferrin as targets, pairs of peptides with individual binding affinities in the low micro-molar range were transformed into synbodies with ~1000-fold higher binding affinities, levels of affinity comparable to monoclonal antibodies. These synbodies were shown to work in ELISA and protein pull-down assays, further demonstrating their effectiveness as substitutes for antibodies.