Higher-throughput, label-free, real-time molecular interaction analysis

The burgeoning number of commercial biosensors geared toward higher throughput analysis clearly demonstrates that the market for this technology is expanding. Beyond throughput, the second most important issue is sensitivity. At this time, we have tried to steer clear of comparing systems based on sensitivity, principally because it is difficult to do so without a benchmark study using the same molecular system to evaluate every instrument. Suffice it to say that there is likely to be a trade-off between resolution and throughput. For example, the sensitivity of the two-dimensional array-based systems that we discussed likely is not high enough at this stage to allow direct detection of the binding of small molecules to immobilized targets. So, in many ways, more could in fact mean less. It is, however, important to remember that the primary factor that determines the quality of any biosensor data set is the quality of the target itself and how it is immobilized. None of these sensors can yet take a denatured and precipitated protein and turn it into a fully active target on the sensor surface. (If there was such a sensor, we would have highlighted it.) In contrast, low-percentage activity often is not a hindrance for enzymatic-, radioligand-, and fluorescent-based assays. Because these assays can be 1000 times more sensitive than optical biosensors, they often are used to analyze targets that are not especially pure or highly active (whether this is actually a good idea is debatable). It is also important to recognize that most optical biosensor assays are performed as direct binding assays. For small molecule work, for instance, we can use the sensor to determine whether a compound actually interacts with a particular target. These direct binding assays, on their own, do not tell us whether the compound actually inhibits or activates a particular target. For this reason, most pharmaceutical companies will continue to use activity-based assays as a primary screening tool. In this setting, biosensors are ideal tools to be used for secondary screening to confirm whether the hit is real and, if so, to provide dynamic information about its interaction with the target. Information on how fast something binds and the half-life of the complex is becoming an essential part of the optimization process for both small molecules and biopharmaceuticals. We applaud these biosensor manufacturers for developing tools that will increase the speed of interaction research, ultimately leading to a better understanding of both basic biology and improved health care. © 2006 Elsevier Inc. All rights reserved.