Highly complex immunoassays that identify and quantify many different antigens simultaneously need high-resolution imaging capability. A simple, low-cost technique could be music to our ears.
The capabilities of immunoassays — tests that use the binding of ‘capture’ antibodies to antigens in order to identify the latter — have been advancing in leaps and bounds over the past three decades. Methods of enzyme amplification have increased assay sensitivity. Diode lasers have reduced the size and cost of instrumentation. Microfluidics has enabled both the analysis of very small sample volumes and the parallel processing of multiple samples. And microarrays of capture antibodies, or other recognition molecules, attached to surfaces have made possible the simultaneous testing of a sample for large numbers of target molecules.
This last point harbours a problem. As microarray elements become smaller and smaller for larger and larger numbers of simultaneous tests, the necessary imaging resolution cannot readily be achieved with the standard, off-the-shelf CMOS or CCD technologies. Confocal scanners using microscope lenses can provide the required resolution, but their cost, size and the geometrical alignment of their optics pose other practical problems. Writing in Angewandte Chemie International Edition, Sebastian A. Lange, Günter Roth and colleagues find a way out of this seeming impasse1. They demonstrate an immunoassay readout of high efficiency and sensitivity — using, in a slightly modified form, a pick-up head of a compact-disc player (Fig. 1).
The authors' system is both elegant and simple. They used a so-called sandwich immunoassay, in which the antigen to be measured is bound between two different antibodies. First, a capture antibody is stamped in a 25-micrometre square pattern onto a solid substrate; the desired antigen binds to this antibody. Second, a detector antibody binds to this antigen to make it visible to the CD pick-up head. Here, the authors cleverly employed an antibody carrying a gold nanoparticle. This gold nanoparticle catalyses the deposition of silver grains onto the substrate to which the capture antibody, antigen and detector antibody are all now attached. The reflections from the silver grains could be read by the CD pick-up head with a resolution of around 50 nanometres — considerably smaller than the diameter of most of the silver nano-particles, which is of the order of several hundred nanometres. Using their sandwich system, the authors demonstrated that they could detect antigen in serum at concentrations from 1 microgram per millilitre down to 100 picograms per millilitre.
The authors also performed a carefully controlled study in which they spaced antigens, each attached to a capture molecule, well apart from each other on the substrate1. They suggest that, by correlating the density of the silver precipitate with the coverage of the antigen, it might be possible to detect single molecules with their system. But measuring a single molecule carefully tied to a uniform surface is a far cry from pulling a single molecule out of a solution, capturing it at a surface and measuring it against a background signal generated from any of several variable sources. The potential for using the CD pick-up head with silver staining for single-molecule detection is therefore less convincing than the suggestion that the technology could be widely useful for low-cost, high-sensitivity readout systems for many different types of assay.
Significant issues remain to be addressed. If the CD pick-up head is to be used to detect many different molecules simultaneously, multiple capture molecules will have to be immobilized on surfaces suitable for use with the CD pick-up heads. This is almost sure to require a more flexible technology than the stamping method used by Lange et al. Optical alignment issues will also need to be addressed. In Lange and colleagues' study, the CD pick-up head is only one part of the optics: it is combined with a microscope stage for coarse adjustment of the magnification, as well as a lateral translation stage to adjust the position of the line of sight over the substrate. Although it may be possible to use built-in hardware in a commercially available CD player to perform these functions, this has yet to be demonstrated.
One noteworthy advantage of the CD-pick-up approach, alongside small size, low cost and high resolution, is not emphasized by the authors. Although an assay signal has never actually been generated using a CD as a sensing surface, biochemical manipulations have already been performed directly on a CD. Externally readable fluorescence signals have even been generated in channels within the disc2,3,4,5 — the challenge here being the use of centrifugal force to move the fluids through the processing steps on the surface of the CD to the readout position. The combination of such fluidic approaches with in situ signal generation as demonstrated by Lange et al.1 could potentially lead to a sea change in medical diagnostics. Imagine in the future buying a ‘respiratory pathogen CD’ from the local pharmacy when you catch a cold, inserting your self-test swab and placing it in your portable player to find out whether you should take antibiotics or stick to the chicken soup.