Medical Device Link.

Massive gene-sequencing efforts have culminated in the recent completion of a rough draft of the sequence of the human genome, a major milestone that represents the beginning of a new genomic era.^{2, 3} Over the long term, this information will have a profound effect not only on healthcare but also on the quality of life in general.

As more genetic information is becoming available, those within the pharmaceutical and diagnostics industries are experiencing an ever-increasing need for rapid and accurate methods to detect and quantitate nucleic acids. Extensive libraries of potential drug compounds have been compiled using combinatorial chemistry. A high-throughput gene-expression-analysis product is needed to screen these libraries against gene targets associated with certain disease conditions. Nucleic acid–based diagnostic assays also have broad potential in the clinical setting for cancer detection, immune-status testing, infectious-disease diagnosis, and measurement of drug resistance or susceptibility.

In gene-expression analysis, the amount of a specific messenger RNA (mRNA) produced in drug-treated or diseased cells is measured and compared with the amount found in corresponding untreated healthy cells. Observed changes in the level of expression of genes of interest that occur in diseased cells, and their response to drug treatment, can then be used to guide the selection of drug candidates for further development, to elucidate the mechanisms of drug action, and ultimately to assist in the selection of patients for inclusion in preclinical or clinical trials.

Traditionally, gene-expression analysis has been performed using such methods as Northern blots, ribonuclease protection assays (RPA), or reverse transcription polymerase chain reaction (RT-PCR). However, these methods are not well suited for testing large numbers of samples because they are time-consuming and laborious, and in some instances lack adequate sensitivity.

Several new products for detection of specific nucleic acid sequences are either on the market or under development.⁴ The majority are designed to detect the presence or absence of a specific nucleic acid molecule of interest through its hybridization to a complementary sequence (probe nucleic acid) immobilized on a solid phase that is interfaced with a detection device.

To detect hybridization, these methods require the attachment of a fluorescent, radioactive, enzymatic, or chemiluminescent label to the target nucleic acid. The procedures used to label the nucleic acid and the instrumentation needed to detect some types of labels can make these products complex and cumbersome to use.

In addition, amplification of the target nucleic acid using PCR or a related technique is generally required to detect small quantities of a specific nucleic acid using currently available detection methods. This extra step increases the time required to perform the assay and introduces the possibility of sample contamination and other errors during the amplification procedure.

Electrochemical detection of nucleic acids has generated considerable interest because electrochemical methods offer high sensitivity while using relatively inexpensive instruments and simple protocols. DNA hybridization has been successfully detected electrochemically using direct oxidation of DNA over an electrode and indirectly through detection of redox-active hybridization indicators that bind more strongly to duplex DNA than to single-stranded DNA.^{5, 6}

However, the signal obtained from DNA using these methods is small and therefore significantly limits the sensitivity of these methods. This problem can be overcome by using an electrochemical mediator to extract electrons from guanine residues in DNA or RNA and carry them to the electrode, where the mediator is regenerated and can participate in additional electron-transfer events.⁷ This approach is the basis of the technology used at Xanthon Inc. (Research Triangle Park, NC), and is the topic of this article.

Mediated oxidation of guanine residues is a rapid and sensitive method for directly detecting nucleic acids. To implement this approach, nucleic acid probes are covalently bound to a tin-doped indium oxide (ITO) electrode. The electrodes are then exposed to a sample containing target DNA or RNA. The complementary nucleic acid hybridizes to the nucleic acid probe, and noncomplementary nucleic acid sequences are subsequently removed by washing. The nucleic acid hybrid is then detected using a redox-active mediator with an appropriate redox potential, such as tris (2,2'-bipyridyl) ruthenium (II) (Ru(bpy)^₃2+).

In the first step of this catalytic guanine oxidation, the mediator is oxidized to the +3 form (Ru(bpy)^₃3+) by the high positive potential of the electrode:

Ru(bpy)^₃2+ Æ Ru(bpy)^₃3+ + e^–

The oxidized form of the mediator then abstracts an electron from guanine in the hybridized nucleic acid target to form a radical cation that can undergo further reactions:

The regenerated reduced mediator is again oxidized at the electrode, completing a catalytic cycle. The current produced during each oxidation of Ru(bpy)^₃2+ is measured and reflects the amount of guanine present in the nucleic acid target hybridized at the electrode. In contrast to direct oxidation of nucleic acid at an electrode, electron-transfer reactions between Ru(bpy)^₃2+ and guanine, as well as between Ru(bpy)^₃2+and the ITO electrode, are very fast. The major reason for the rapid electron transfer between Ru(bpy)^₃2+ and guanine (the second-order rate constant has been measured to be roughly 1 ¥ 10⁶ M^–1 s^–1) is that the oxidation potential of guanine is almost identical to that of Ru(bpy)^₃2+.⁸

The catalytic nature of the mediated electron transfer yields current values that are much larger than those obtained using nonmediated electron-transfer reactions involving nucleic acids. This makes possible the detection of low copy numbers without amplification. Because detection is based on the oxidation of guanine residues, RNA can be detected in the same manner as DNA, without using reverse transcription to form cDNA. Eliminating these extra preparative steps simplifies the assay procedure and can improve the accuracy and reliability of the detection method.

In order to minimize background signal in the mediated electron-transfer system, it is preferable that the nucleic acid probes not contain guanine. When a probe sequence for a specific target nucleic acid that contains only adenine, cytosine, and thymine is not available, guanine bases in the probe can be substituted with hypoxanthine, because it is less reactive than guanine in the electron-transfer reaction with Ru(bpy)^₃2+. Since hypoxanthine can form only two of the three hydrogen bonds in a Watson-Crick base pair, other guanine derivatives such as 7-deazaguanine may also be viable and attractive alternatives.⁹

For development purposes, cyclic voltammetry is the electrochemical method of choice for the nucleic acid–detection system described here. This method provides information on the mechanism and kinetics of the electrochemical reaction of interest, as well as quantitative information.

In cyclic voltammetry, a potentiostat is used to apply a potential to the working electrode with respect to the reference electrode, whose potential is well defined and constant during the experiment. The potential of the working electrode is increased linearly with time to a specified value at a constant rate. When the specified value is reached, the potential is then reversed to the starting point at the same rate. This change in potential is the electrochemical driving force that causes oxidation or reduction of the analyte. The current resulting from these events is measured and recorded in a voltammogram, which is a plot of current as a function of the potential applied to the working electrode.

Figure 1 shows cyclic voltam-mograms obtained at ITO electrodes modified with an inosine-substituted oligonucleotide probe with varying amounts of hybridized target nucleic acid as determined radiochemically. Hybridization of the complementary target to the probe results in a significant current enhancement over background. As expected, the current observed increases with increasing amounts of guanine at the electrode surface.

Figure 1. Detection of hybridized nucleic acid with cyclic voltammetry. Guanine in the capture probe was substituted by hypoxanthine. The amount of the target nucleic acid varied between 108 and 593 fmol and was determined radiochemically. Scan rate of 20 V/sec was used for interrogation. Electrode size was 0.28 cm2.

The current response is a linear function of the amount of the target nucleic acid over the range examined (1–1000 fmol) at 6-mm-diam ITO electrodes. The amount of signal generated by noncomplementary target nonspecifically bound in the system is minimal because of careful design of the solid phase, lysis and hybridization solution, and wash reagents.

In the Xanthon system, low nonspecific binding of sample components and extensive posthybridization washes enable the detection of target nucleic acid from whole-cell lysate without purification. The minimal sample preparation required is one of the main advantages of the Xanthon technology, since it offers substantial time savings and reduces the experimental error introduced during complex processing steps.

Another desirable attribute of electrochemical biosensors for nucleic acid diagnostics is that they are well suited for miniaturization. Besides allowing for smaller sample volumes, microelectrodes offer improved sensitivity over macroelectrodes because of their ability to detect low currents with better discrimination against the charging current. Charging current results from the reorganization of the electrode-solution interface when potential is applied, and can be a major source of background in electrochemical experiments.¹⁰

ITO electrodes to be used in the Xanthon system have a 200-µm diam, reduced from the original 6-mm diam, and are produced using microfabrication and photolithographic procedures. Representative cyclic voltammograms at these electrodes showing the oxidation current obtained for attomole levels of guanine, as well as the background current, are given in Figure 2. The peak current is linearly proportional to the amount of guanine at the electrode surface. As can be seen in Figure 2, 43 amol of strand (215 amol of guanine) are easily distinguished from background. Thus, whereas detection limits of the mediated electrochemical detection method on macroelectrodes (electrode area = 0.28 cm²) are at the low femtomole level, they decrease to low attomole levels as the assay is scaled down to a smaller electrode size (electrode area = 3.14 ¥ 10^–4 cm²).

Data such as those shown in Figure 2 are generated using cyclic voltammetry where the amount of signal obtained is a function of scan rate. Thus, the time spent at the peak potential is limited and does not allow for complete oxidation of guanine residues present at the electrode surface. The Xanthon system will use chronoamperometry versus cyclic voltammetry as a detection method, because it allows for collection of the maximum signal out of each guanine molecule and it involves only a single potential step, making it easier to implement.

Figure 2. Cyclic voltammograms illustrating a dose response for varying amounts of guanine at a 200-µm ITO electrode. The probe used was a 21-mer with five guanines. The amount of the attached probe was determined radiochemically. Scan rate of 20 V/sec was used for interrogation.

In this method, the potential of the working electrode is stepped from a value where no electron-transfer reaction occurs (for example, 0 mV) to a potential where the mediator is oxidized (1100 mV) and therefore able to extract electrons from guanine. The working electrode is held at the step potential for a specified period, and the current measured as a function of time. The signal associated with target nucleic acid is obtained by integrating the current measured during a specified period, resulting in charge passed as a result of guanine oxidation, which is linearly proportional to the amount of guanine at the electrode surface.

One of the major features of the Xanthon technology is that it can readily be adapted to a variety of assay formats, including biochips for genomic re-search, handheld devices for point-of-care medical diagnostics, and microplates for drug discovery.

The initial product is designed for gene-expression analysis. In order to minimize the need for changes in existing instrumentation and to ensure easy integration into the pharmaceutical drug-discovery process, the first product will be presented in a 96-well microplate format. A schematic of the Xanthon Xpression Analysis Plate is shown in Figure 3. Each well of the plate contains seven 200-µm working electrodes, six derivatized with immobilized probes selected by the individual customers including one probe to serve as a positive control; and one electrode to act as negative control. Inclusion of both a positive and a negative control in each well will permit adjustment for cell-culture variables, thus helping to ensure the integrity of the data.

Figure 3. Schematic of the Xanthon Xpression Analysis Plate.

The interrogation period is extremely short and allows for 5-minute processing time per plate, enabling users to collect approximately 138,000 data points per day, plus 55,000 controls. This product is suited for the lead optimization phase of the drug-discovery process because it allows customers to evaluate the impact of a large number of test compounds on the expression levels of a selected set of target genes.

The Xanthon product is designed for analysis of cells cultured in 96-well microplates (typically 50,000 cells per well). Following exposure to test compounds, cells are washed and lysed using a proprietary lysis and hybridization solution. Then 50 µl of the lysate is transferred to a well of a Xanthon Xpression Analysis Plate, where hybridization of the target nucleic acid to the probe immobilized at the electrode surface occurs. After hybridization, the wells are washed to remove unhybridized materials, the mediator solution is added, and electrochemical interrogation is performed.

The Xanthon Xpression Analysis System is capable of multiplexing because the signal generated is localized within each electrode. In the present electrode layout, there is no cross talk between the electrodes, and the number of electrodes per well could be increased significantly. Figure 4 illustrates interrogation of one well of the 96-well microplate using the Xanthon HT96/7 instrument and software. In this experiment, electrodes 1–4 had a probe complementary to the target nucleic acid, while the probe immobilized on electrodes 5–7 was noncomplementary to the target. It is clear that hybridization occurred only at electrodes with the complementary probe, resulting in a significant increase in the charge obtained at these electrodes over background.

Figure 4. Interrogation of one well of a Xanthon Xpression Analysis Plate.

In the Xanthon instrument, each electrode will be interrogated three times. During the first interrogation, all of the guanine at the electrode surface is oxidized, while the second and third interrogation provide background information for each electrode. Thus, each electrode provides its own internal control, resulting in a high degree of reliability and accuracy.

The Xanthon HT96/7 is a compact instrument that will be a common platform for all of the company's products incorporating the 96-well microplate format (see Figure 5). The instrument consists of a robotic arm for plate handling and the plate-reader circuitry. It provides automated reading of the Xanthon Xpression Analysis Plate plus extensive data analysis capability, including an option to export the data to existing laboratory information systems as an ASCII file.

Figure 5. The Xanthon HT96/7 is a compact instrument that will be a common platform for all of the company's products using the 96-well microplate format.

Instrument cost is low relative to comparable instruments required for use with optical-based detection methods. Prior to market introduction, the integrated Xanthon system will be independently tested at five different sites to validate its performance characteristics when compared with other gene-expression analysis methods, such as RT-PCR, RPA, or Northern blots.

Mediated electrochemical detection of nucleic acid provides distinct advantages over existing methods for expression analysis. Multiple targets can be analyzed from a single sample using a highly automated system that requires minimal sample handling and uses direct detection of the target nucleic acid. In addition, the system described here is adaptable for the simultaneous detection of nucleic acids (expression analysis) and proteins using mediated electrochemistry on the same instrument platform. The ability to detect levels of expressed mRNA and corresponding protein in the same sample using the same detection technique would provide a wealth of information inaccessible by other methods.

The age of genomics is here, with advances in science opening the door to new approaches to drug discovery, as well as more powerful and efficient nucleic acid–based diagnostic tools. Genomic-based drug discovery will enable the development of therapeutic agents with high specificity for a given disease target. Availability of these therapeutics will, in turn, drive the need for genomic-based diagnostics that can specifically identify these disease targets.

Because it is a simple and rapid method applicable to the detection of unlabeled DNA and RNA, as well as proteins in the same sample, the detection system described here has broad applications and will have a significant impact on the way drug discovery is conducted, diagnostic testing is performed, and test results are used to select specific therapeutic approaches.