Many of the best methods available for monitoring biological binding events can't be used in a diverse range of clinical samples. An ultrasensitive assay based on magnetic signals overcomes this problem.
The Matrix and its two sequel films explore the complex reality of an eponymous cyber-generated world that allows machines to dominate humans. This science-fiction trilogy was famous for its distinctive visual effects, which were generated using innovative combinations of previously existing photographic techniques. In Nature Medicine, Gaster et al.1 report how they have used a similar strategy — the combination of established techniques — to devise an ultra-sensitive assay that detects biomarker proteins associated with disease or metabolic states. Their pioneering approach uses magnetic signals to overcome the effects of the biological matrix, the host of compounds found in all biological samples that cause interference in assays. The sensitivity of the authors' technique is 1,000 times better than the current gold-standard method, the enzyme-linked immunosorbent assay (ELISA).
Interference by biological matrices is a real problem in immunoassays, which rely on the binding of an antibody to its target antigen. Such interference has been defined as “the sum of the effects of all of the components [in a sample], qualitative and quantitative, with the exception of the analyte to be measured”2. It occurs ubiquitously, both at the stage of specific antigen–antibody binding and during the detection phase, when the amount of antigen–antibody coupling is quantitatively translated into a measurable signal (such as light absorption or fluorescence). The intrinsic, non-zero light absorption and/or fluorescence of biological milieux hinder accurate measurements of analytes (antigens), especially at lower concentrations.
Gaster and colleagues' approach1 is based on the fundamental observation that even the most complex biological matrices lack a detectable magnetic signal, and would therefore not interfere with a magnetic-field-based detection method. Their assay uses magnetic nanoparticles, bound with high affinity specifically to the biomarkers of interest, as the basis of its detection system. To generate an electronic readout of the assay, they adapted the magnetic-sensing capabilities of giant magnetoresistive (GMR) sensors, devices that were originally developed for use in the read heads of computer hard drives.
So how exactly does Gaster and colleagues' assay work? The authors first attach specific antibodies for a target biomarker to the surface of a GMR sensor, and expose the surface to a fluid sample containing that biomarker, whereupon the target molecules bind to the immobilized antibodies (Fig. 1). The authors then wash the sensor with a second set of antibodies that have been labelled with a compound called biotin; these antibodies also bind specifically to the trapped biomarkers. The next step is to treat the sensor with an aqueous suspension of magnetic nanoparticles — tiny spheres containing iron oxide — to which the protein avidin has been attached. Avidin binds with high affinity to biotin, and so the magnetic nanoparticles become strongly linked to the biomarker–antibody complexes on the surface of the sensor. Finally, the authors activate the sensor by exposing it to a magnetic field. The resulting electronic readout is proportional to the extent of nanoparticle binding, and thus provides a quantitative measure of the amount of biomarker bound to the sensor's surface.
Bioassays that use GMR sensors to detect molecules sandwiched between a pair of antibodies (one immobilized and the other introduced in solution) have been reported previously3,4. But there are two factors that distinguish Gaster and colleagues' results1 from the others. First, they have meticulously characterized their approach for several candidate analytes to prove its generality and its superiority to ELISA. And second, they demonstrate that their nanosensor can be used for multiplexing, so that as many as 64 assays can be performed on the same device. This capability is possible because of the specificity and sensitivity of their design, and because of the lack of biological-matrix interference.
The authors showed that their nanosensor assay system works in all biological fluids studied, including blood, urine, saliva and cell lysates — although the signal strength in saliva is less than that in other media, perhaps because the viscosity of saliva affects the binding kinetics of the assay. The authors also demonstrated that real-time readouts of binding are possible with their system, and that it can be used in vivo to follow the earliest stages of tumour progression by monitoring appropriate biomarkers. Turbidity and sample pH do not significantly affect the assay, but Gaster et al. found that the output signal is affected by temperature, so that hot or cold samples create undesirable spikes in the baseline of the electronic readout before they equilibrate to room temperature. The authors were able to correct for this, however, by processing the data using a mathematical algorithm. No other assay system, including ELISA, has such a combination of broad applicability, high sensitivity and low background 'noise' caused by biological-matrix interference.
It should be noted that Gaster and colleagues' approach does not actually prevent the biological matrix from physically interfering with antigen–antibody interactions. Nevertheless, such interference can be minimized by carefully screening antibodies to find those that don't interact with components of the biological matrix2. The assay could also be adversely affected by inadvertent exposure to strong magnetic fields, such as those present in nuclear magnetic resonance imaging scanners, but this can be prevented by appropriate shielding.
Gaster et al. speculate that their assay will be useful for several applications, such as studying protein–protein interactions and screening compounds for biological activity in drug-discovery programmes. Furthermore, the sensitivity and rapid responsiveness of the system permit biomarker monitoring with both high spatial and temporal resolution. This might open up exciting medical applications — for example, by tracking appropriate biomarkers, tumour responses to therapy could be anticipated before any effect becomes apparent. That could reduce the risk of untoward drug effects, and allow adjustments to be made to medication in a more timely way than is currently possible.