Reading cancer-specific signatures

Assays that detect disease-specific biomarkers are rapidly emerging as important clinical tools that reveal disease-induced biochemical changes and allow for early diagnosis of a wide range of conditions. Biomarkers are molecules (proteins, nucleic acids, and so on) that when detected at unusual levels indicate a specific disease state. For example, high levels of antibodies against a particular type of bacterium present in the bloodstream accurately point to infection. Alternatively, different types of cancerous cells overproduce certain molecules, thereby allowing different cancers to be identified based on their unique molecular ‘signatures.’ Time is a critical factor in cancer diagnosis and treatment, and early disease detection generally leads to an improved prognosis. As such, accurate reading of these signatures both allows for early diagnosis and gives clinicians a wealth of information concerning disease progression, thus enabling carefully tailored treatment.

Biomarkers are often present in very small quantities in biological fluids (e.g., blood and saliva) and may be difficult to detect among the many other components in a given sample. As an increasing number of cancer-specific biomarkers are being discovered, biosensor technology must also advance to achieve the high sensitivity and specificity required to measure these molecules. An important performance measure for the sensitivity of a biosensor is the limit of detection (LOD), which is the lowest concentration of a biomarker at which the biosensor is capable of distinguishing the true signal from the background noise.

Figure 1. Detection volume comparison of (a) fluorescent and (b) confocal microscopy. Human interleukin-8 (IL-8) antibodies bind to the sensor surface via a capture probe that itself is bound to the surface through the interaction of streptavidin and biotin (not to scale). Human IL-8 protein is detected using a fluorescently labeled (Alexa Fluor 488) antibody. (c) A typical optical intensity distribution normal to surface. The location of peak light intensity represents the best signal-to-noise measurement.

In surface-based sensing techniques, a blocking protein is traditionally used to reduce the noise produced by nonspecific binding. Here, we present additional noise-reducing strategies. We have developed two ultrasensitive biosensor techniques for detecting oral cancer biomarkers from saliva that demonstrate dramatic LOD improvements. One is a fluorescence-based protein sensor that uses confocal optics to dramatically reduce background noise.¹ The other is an electrochemical DNA sensor^{2, 3} that exploits a steric hindrance scheme to suppress nonspecific signals.

Figure 2. Illustration of the specific signal amplification using hairpin-probe electrochemical detection. Upon binding with the complementary target messenger RNA, the hairpin opens up, and the HRP (horseradish peroxidase) complex is formed. TMB (3, 3′, 5, 5′-tetramethylbenzidine) regenerates the reactive HRP, thus amplifying the current signal.

The interleukin-8 (IL-8) protein is present at higher concentrations in the saliva of patients with oropharyngeal squamous cell carcinoma relative to healthy patients.⁴ However, accurately measuring IL-8 protein levels in saliva samples requires the reduction of competing signals. This is achieved by minimizing the fluorescence detection volume using a confocal microscope (see Figure 1). While recording the fluorescent signal, the pinhole in the confocal microscope prevents light originating from any plane other than the focal plane from reaching the detector. By rejecting unfocused light, only a 1μm-thick region around the focal plane is measured by the photon detector, thereby excluding optical noise and improving the signal-to-noise ratio. With this method, we achieved an LOD of 4fM,¹ which is 275 times greater than traditional techniques that demonstrate a sensitivity of only 1.1pM.

The messenger RNA of IL-8 present in saliva is also a biomarker for oral cancer.⁴ However, the LOD of electrochemical sensors used for detecting nucleic acids can limit their application. To reduce the background noise, we have engineered the conformation of the probe to generate considerable steric hindrance,³ which suppresses nonspecific signals and generates a signal-on amplification process for target detection (see Figure 2).⁵ The stem-loop configuration brings the reporter end of the probe in close proximity to the surface and makes it unavailable for binding with the mediator. Target binding opens the hairpin structure of the probe, and the mediator can then bind to the accessible reporter. This signal-on process is characterized by a low basal signal, a strong positive readout, and a large dynamic range.

In addition to the hairpin probe design, target binding (hybridization) can also be controlled via changes in the electrical field. Since oligonucleotides are negatively charged, a positive electrical field will increase target binding, whereas a negative electrical field will denature the hybridized duplex. While applying a cyclic square-wave electrical field during the hybridization process, the short positive pulse improves complementary target binding, whereas the longer negative pulse eliminates nonspecific interactions. By combining electric field control with hairpin probes, the limit of detection for RNA is about 0.4fM, which is 10,000-fold more sensitive than conventional linear probes.^{1, 3}

By extending the sensitivity of these techniques, without the need for enzymatic amplification steps, we have expanded the range of detection for these biosensors while maintaining their simplicity. The resultant methods are well suited for point-of-care detection of specific biomarkers within complex bodily fluids such as saliva and are a significant advance toward routine molecular diagnostics in the clinical setting. There are many biomarkers in body fluids that have very low concentrations. With femtomolar sensitivity and without enzymatic amplification, we can further extend the range of detection of molecular diagnostic techniques for different biomarkers. Future efforts could potentially include integrating the detection method with a microfluidic system to create a lab-on-a-chip product.

This work was supported by funds from the National Institutes of Health/National Institute of Dental and Craniofacial Research (UO1DE 017790, UO1DE015018, and RO1DE017593) and the National Aeronautics and Space Administration/National Space Biomedical Research Institute (TD00406).