Team Members:

Mr. Kuosheng Ma
Mr. Joshua Kim

Research summary:

1. DNA array background:

The wealth of available DNA sequence data makes it possible to identify many diseases as well as biological threats such as the presence of an infectious agent in the environment. Improving the sensitivity, selectivity, speed, simplicity and reducing the cost of such assays are important goals that will significantly affect the administration of health care at locations ranging from the patient’s bedside to the battlefield.

Alterations in gene expression have profound effects on biological functions. These variations in gene expression are at the core of altered physiologic and pathologic processes. DNA array technologies provide the most effective means of identifying gene expression and genetic variations.

DNA are maybe prepared from a wide variety of samples such as tissue, bacteria, saliva, etc. For genotyping analysis, the sample is genomic DNA. For expression analysis, the sample is cDNA, DNA copies of RNA. The DNA samples are tagged with a radioactive or fluorescent label and applied to the array. Single stranded DNA will bind to a complementary strand of DNA. At positions on the array where the immobilized DNA recognizes a complementary DNA in the sample, binding or hybridization occurs. The labeled sample DNA marks the exact positions on the array where binding occurs, allowing automatic detection. The output consists of series of hybridization events, indicating the presence or the relative abundance of specific DNA sequences that are present in the sample.

2. Electronic detection of DNA hybridization:

Conventional assays often rely on the detection of fluorescence from a molecular fluorophore. Electronic detection of hybridization is expected to require less complicated instrumentation and feature similar detection limits compared to the traditional optical methods.

When a conductor is placed in an electrolyte solution, a potential is generated due to an unequal distribution of charges across the interface (Figure 1(a)). Two oppositely charged layers, one on the electrode surface and one inside the electrolyte form a “double layer,” which behaves as a parallel plate capacitor. When an additional layer is present at the electrode, such as an oxide or a self-assembled monolayer (SAM), an additional capacitance and/or resistance associated with that layer is added to the circuit. In principle, the introduction of molecules, such as DNA, to that interfacial region, will affect the measured complex impedance through changes of the local geometry, the dielectric constant, and the amount of charges at the electrode/electrolyte interface. A robust sensor must be able to measure the impedance change with a signal to noise ratio (S/N) that yields the desired sensitivity.

(a)                                          (b)

(c)

Figure 1. Interface phenomena and Randles’ equivalent circuit. (a). Electrical double layer capacitance Cdl. (b). Charge leaking across the double layer represented by ‘charge transfer’ resistance Rct and diffusion of ions to the interface represented by Warburg impedance Zw. (c). Randles’ equivalent circuit.

A perfect biomolecule recognition layer, for example a self-assembled mono-layer of single stranded DNA (a SAM of ssDNA), would cover the electrode completely and the selective binding of complementary biomolecules to that layer (i.e., hybridization in the given example) would be the only contributing impedance element. As shown in Figure 1(b), the general electronic equivalent scheme for the electrode/electrolyte includes (1) a ‘charge transfer’ resistance Rct, corresponding to charge leakage across the double layer (this could involve electrons and/or ions), which for small applied signal amplitudes is given by

,Where R is the universal gas constant, T is the absolute temperature, n is the number of electrons involved in the electrode reaction (in the case of electron exchange at the interface, for ion exchange n corresponds to the charge of the ion), F is Faraday’s constant, i0 is the exchange current density (electrons and/or ions); and (2) the Warburg impedance Zw, corresponding to the diffusion of ions to the interface from the bulk of the electrolyte. It is given by

Where w is the angular frequency (s-1) and s is the Warburg coefficient (Os-1/2) .The equivalent circuit of the interface impedance is represented in Figure 1(c), and is known as Randles’ equivalent circuit. To maximize the device’s ability to detect bio-recognition processes selectively, the impedance variation must be dominated by capacitive processes and not by the electrolyte resistance, charge transfer or diffusion effects of ions. To optimize the sensors’ sensitivity one consequently works in a frequency regime where the capacitive contributions associated with the binding of biomolecules dominates. For example, in the case of a Au electrode covered with a SAM of ssDNA, at too high a measuring frequency, the impedance data are almost entirely controlled by the electrolyte resistance Re, i.e., the resistance between the sensor and the counter electrode. At too low a frequency the impedance of the sensor exhibits phase and quadrature values that are proportional to, indicative of Warburg impedance control. Only measurements conducted at intermediate frequencies are dominated by the binding /sensing phenomena occurring at the sensor surface.

Impedance based sensors are also confounded by non-selective binding. In order to remove the non-specific signal an experimental set-up including a reference sensor may be used. This sensor, functionalized in exactly the same way as the working sensor (except with non complementary DNA or a non-binding protein), is expected to have the same behavior as the working sensor towards non-specific interactions, and therefore to allow a specific differential measurement.

For label-less impedance sensing of biomolecule binding both functionalized insulators and metals have been attempted1. The feasibility of using electrochemical impedance measurements on functionalized heterostructures (semiconductor/dielectric/electrolyte) to directly detect hybridization between complementary homo-oligomer DNA strands without labels was demonstrated. When metal electrodes such as Pt and Au are used as the substrate for biolayers they showed stronger signals2,3,4. A typical example is the work by Bergren4 who demonstrated the feasibility of a biosensor for direct detection of DNA hybridization on gold, but even here the specificity was not very high and the reproducibility of the sensors was poor. Janata et al. show how small variations in the electrolyte could affect the impedance signal not because of a Cdl variation but because of Rct instability5. The latter gives us an indication of the typical signal level one can expect when detecting biomolecules using a differential impedance method without the use of any labels. We will show that with signal amplification methods much higher impedance differences (orders of magnitude) can be obtained. Charge transfer effects always confound label–less impedance results and often dominate the total system impedance. To reduce this effect one must attempt to use a well-polarized electrode. There is a lot of skepticism about the ability to ever fabricate a perfectly-polarized membrane6. Recently, with the introduction of different grafting techniques such as self-assembled monolayers or polypyrrole functionalization, it has been demonstrated that it is possible to reach a higher level of electrode polarization, avoiding the Faradic or diffusive current variations to a significant degree. Commercial products remain elusive though as reproducibility and selectivity remain major obstacles. In our approach an impedance amplifying label overcomes all the shortcomings of the label-free impedance measurements described in this section.

3. Recent results of impedence analysis:

In order to increase the impedance changes in DNA hybridization label-less experiments by at least an order of magnitude, we compared different types of Au electrode surfaces. The table below shows the data obtained after extracting the double layer capacitance at a fixed frequency of 100Hz from the real part of the measured impedances and normalizing each of these impedance values with the bare gold impedance.

These data demonstrate that the impedance increases up to 400% after DNA hybridization. Thus, we postulate that, in principle, real time DNA hybridization can be monitored using a single frequency impedance measurement.

Preliminary experiments on electronic detection of DNA hybridization with impedance amplifying labels were also performed in our lab. As a starting point we used a commercially available Enzyme-Labeled Fluorescence (ELF) signal amplification technology (Molecular Probes, Eugene OR). This well-known technology involves a streptavidin alkaline phosphatase reporter in an enzyme mediated detection scenario in which the ELF substrate (ELF-97), a water soluble weak blue fluorescent molecule, is converted to an insoluble, intense green fluorescent molecule. The insoluble product precipitates onto the surface of the solid phase used in the assay. Our hypothesis was that this type of a precipitation on an underlying conductor electrode would enhance/amplify the impedance response significantly. Since the product is fluorescent we could visualize where the precipitate formed and link fluorescence intensity data with our impedance measurements. Using synthesized biotinylated oligos, we carried out post-hybridization labeling with the streptavidin alkaline phosphatase enzyme. This was followed by a vigorous rinse and exposure to the ELF-97 substrate. From fluorescence imaging we found that the precipitate occurred only on or very close to the microelectrodes.

Figure 2. Experimental results with enzyme-linked amplified electrochemical sensing technique.

A typical impedance measurement at 100 Hz, comparing DNA hybridization with and without enzyme amplification is shown in Figure 2. On the left, hybridization of ssDNA without the enzyme label is used as a reference point (normalized to 1), and on the right, with amplification, an impedance change of more than a factor 10 is observed. Clearly the insoluble precipitate from the enzyme reaction does amplify the impedance change after hybridization. These measurements are quite encouraging, especially since they were not optimized in terms of the working pH, electrode potential, flatness of the electrode, the nature of the conductor electrode, etc. However, the experimental results were quite erratic due to the nature of the salt deposit, as such a film might not adhere well, might have pinholes or be deposited unevenly. A self-passivated insulator on a suitable conductor electrode (e.g., Ti or Fe forming a titanium oxide respectively iron oxide) upon enzyme-substrate reaction will lead to a much more reproducible result and orders of magnitude signal amplification

References:

[1] C. Berggren, B. Bjarnason and G. Johansson, “Capacitive Biosensors”. Electroanalysis. 13 (3). 173 (2001).

[2] H. Maupas, A. P. Soldatkin, C. Martelet, N. Jaffrezic-Renault and B. Mandrand, “Direct Immunosensing Using Differential Electrochemical Measurements of Impedimetric Variations”. J. of Electroanal. Chem.421. 165 (1997).

[3] S. Ameur, C. Martelet, N. Jaffrezic-Renault and J. M. Chovelon, “Sensitive Immunodetection Through Impedance Measurements onto Gold Functionalized Electrodes”. Appl. Biochem. and Biotech. 89. 161 (2000).

[4] C. Berggren, P. Stalhandske, J. Brundell and G. Johansson, “A Feasibility Study of a Capacitive Biosensor for Direct Detection of DNA Hybridization”. Electroanal.11(3). 156 (1999).

[5] J. Janata, “Twenty Years of Ion Selective Field-Effect Transisitors. Analyst. 119(11). 2275 (1994).

[6] J. Janata and G. F. Blackburn, “Immunochemical Potentiometric Sensors”. Ann. NY Acad. Sci. 428 (Jun.). 286 (1984)