Home About Us Laboratory Services Forensic Science Communications Back Issues January 2008 Research Research and Technology - Forensic Science...
Info
This is archived material from the Federal Bureau of Investigation (FBI) website. It may contain outdated information and links may no longer function.

Research and Technology - Forensic Science Communications - January 2008

Research and Technology - Forensic Science Communications - January 2008
fsc_logo_top.jpg
fsc_logo_left.jpg

January 2008 - Volume 10 - Number 1

Contents

Research and Technology

Rapid Electronic Detection of DNA and Nonnatural DNA Analogs for Molecular Marking Applications

Arica A. Lubin
Graduate Student
Department of Chemistry and Biochemistry
University of California at Santa Barbara
Santa Barbara, California

Chunhai Fan
Professor
Shanghai Institute of Applied Physics
Chinese Academy of Sciences
Shanghai, China

Institute for Polymers and Organic Solids
University of California at Santa Barbara
Santa Barbara, California

Morgan Schafer
Graduate Student
Department of Chemistry and Biochemistry
University of California at Santa Barbara
Santa Barbara, California

Catherine Taylor Clelland
Assistant Professor
Department of Structural and Chemical Biology
Mount Sinai School of Medicine
New York, New York

Department of Pathology
Taub Institute for Alzheimer’s Disease Research
Columbia University Medical Center 
New York, New York

Carter Bancroft
Professor
Department of Structural and Chemical Biology
Mount Sinai School of Medicine
New York, New York

Alan J. Heeger
Professor
Department of Chemistry and Biochemistry
Institute for Polymers and Organic Solids
Department of Physics/Materials Department
University of California at Santa Barbara
Santa Barbara, California

Kevin W. Plaxco
Professor
Department of Chemistry and Biochemistry
Biomolecular Science and Engineering Program
Institute for Polymers and Organic Solids
University of California at Santa Barbara
Santa Barbara, California

Abstract | Introduction | Materials and Methods | Results | Discussion | Acknowledgments | References

Abstract

The use of molecular markers as physical labels has seen widespread application in tasks as diverse as the forensic identification of explosives, the identification of counterfeit merchandise, and the tracing of ground water. One such marking method has been to exploit the unique coding abilities inherent in natural and nonnatural nucleic acids, such as DNA. To date, however, the use of these markers has been limited by the relatively cumbersome methods required for the sequence-specific detection of small quantities of DNA embedded in complex, contaminant-ridden samples and the marker’s relative instability against biological degradation. In this study we demonstrate the feasibility of using a reagentless, electronic method of detecting hybridization, termed E-DNA, to detect identifying DNA and locked nucleic acid (LNA, a more stable, more secure nonnatural DNA analog) markers embedded in documents, pharmaceuticals, consumer goods, and tissue samples. We find that the approach is rapid (results obtained within minutes) and sensitive (detection of marker levels as low as parts per billion) and that the employed molecular markers are stable to long-term storage under ambient conditions. These attributes suggest that the E-DNA-based detection of nucleic acid molecular markers will prove useful in a wide range of forensic applications.

Introduction

The use of molecular markers as physical labels has seen widespread application in recent years (Commission on Physical Sciences, Mathematics, and Applications [CPSMA] 1998; Farmer 2006; Leutwyler 2001; Vergano 2000; Wu 1996). Potential uses of such technology include the addition of intrinsic molecular labels for the post-facto identification of explosives (CPSMA 1998; Kapur et al. 2005; Wu 1996) and the creation of secure internal markers for identifying tissue-banking samples, securing the forensic chain of custody, and tracing ground-water flow (Sabir et al. 1999). Growing concerns regarding the piracy, counterfeiting, and diversion of materials ranging from pharmaceuticals (Associated Press 2003; Fishman 2005; Grady 2003; Kramer 2006; Petersen 2001; Saul 2005) and infant formula (Associated Press 1995; Burros 1995) to car and airplane parts (Hopkins et al. 2003; Levin 2006; Shepardson 2006; Wald 1995) and medical devices (Baker and Thomas 2004; Baker and Thomas 2005; Groubert 2004; Nighswonger 2003; Schultz 2004), whether the various items are sold online (Hafner 2006), abroad (Blume 2006; Buckley 2004; Fishman 2005; Galbraith 2006; Kramer 2006), or on our streets (Confessore 2006; Fifield 2002), have similarly motivated calls to develop unforgeable methods for the forensic identification and authentication of physical goods (e.g., Deisingh 2005; Schultz 2004). The ability to intrinsically label such items with stable, uncrackable “codes” that could be conveniently, reliably, and inexpensively detected would prove useful in today’s increasingly security-conscious atmosphere.

As first proposed by Lebacq (1992), the inherent sequence complexity of nucleic acids suggests that DNA and other oligonucleotides might prove useful as unique tags for molecular marking applications. The approach works by labeling a document or other physical good with a “coded message” encrypted in an oligonucleotide that is physically affixed to, or mixed within, the item. This marker can subsequently be detected, historically via enzymatic methods such as the polymerase chain reaction (PCR) and Sanger sequencing (e.g., Cox 2001a; Cox 2001b; Sabir et al. 1999). The extreme specificity and sensitivity of these methods render them suitable for the detection of even exceptionally low levels of added markers, further increasing the utility of this approach. Moreover, concealment of the marker oligonucleotide within a vast excess of similar but noncognate “masking” DNA yields a “DNA steganography” approach that can be made arbitrarily difficult to counterfeit (Clelland et al. 1999).

Despite much promise, however, the widespread use of oligonucleotide-based molecular markers has been hampered by three potentially significant drawbacks. The first is the cumbersome, time- and reagent-intensive methods currently employed in order to confirm the presence of low concentrations of a specific DNA sequence; existing DNA-based authentication methods (Cox 2001a; Cox 2001b; Sabir et al. 1999) typically require a well-equipped laboratory and take hours to days to return confirmation. Second, cloning and/or PCR can be used to amplify and “decode” DNA markers, making them prone to illicit duplication unless precautions, such as the addition of significant amounts of masking DNA, are taken (Clelland et al. 1999). Finally, the relative instability of DNA renders it unsuitable for marking some materials, such as nuclease-containing tissue samples employed in tissue banking or being used as criminal evidence (where such markers could be employed to strengthen the chain of evidence). A technology that could rapidly and inexpensively detect both DNA ­and, perhaps more important, more stable, nonnatural DNA analogs­ in complex, multicomponent samples should thus prove of significant forensic utility. Here we demonstrate that the previously described electronic DNA (E-DNA) sensor (Fan et al. 2003; Lubin et al. 2006) provides a solution to these problems, and we demonstrate its use in detecting markers based on both DNA and more stable and copy-resistant nonnatural analogs.

The E-DNA-sensing strategy provides a reagentless, reusable electronic means of detecting DNA and DNA analogs (Figure 1). Notably, E-DNA signaling is based on a target-induced conformational change in the sensing probe (as opposed to simple adsorption to the sensing surface). The approach is particularly insensitive to the presence of contaminants, rendering it especially convenient for applications involving complex samples (Lubin et al. 2006). The E-DNA sensor is also rapid (results obtained within minutes), fully reusable, and highly specific (Fan et al. 2003; Lubin et al. 2006). These attributes suggest that the E-DNA-sensing strategy is well suited for the detection of oligonucleotide-based molecular markers.

Figure 1: The E-DNA sensor comprises an immobilized stem-loop oligonucleotide possessing terminal thiol and redox (here methylene blue [MB]) groups (Fan et al. 2003; Lubin et al. 2006). In the absence of a target, the stem-loop structure holds the redox moiety in close proximity to the electrode, thus ensuring rapid electron transfer (eT) and a large, readily detectable current. Upon target hybridization with either naturally occurring oligonucleotides or nonnatural DNA analogs, electron transfer is impeded, reducing the signaling current.

Materials and Methods

Reagents

The E-DNA probe sequence, 5′-HS-(CH2)6-GCGAGGTAAAACGACGGCCAGTCTCGC-(CH2)7-MB-3′ was synthesized (BioSource International, Foster City, California) by conjugating the terminal amino group with MB N-hydroxysuccinimide ester (EMP Biotech, Berlin, Germany) (Hermanson 1996). The modified oligo was purified via C18 reverse-phase high-performance liquid chromatography (HPLC) and polyacrylamide gel electrophoresis (PAGE) and confirmed by mass spectrometry. The DNA molecular marker, 5′-ACTGGCCGTCGTTTTAC-3′, was purchased from Sigma-Genosys (The Woodlands, Texas). A mixed DNA /LNA (locked nucleic acid) molecular marker, 5′-*A*C*T*GGCC*GT*CGTT*T*T*A*C-3′ (where the * in front of a base indicates it is LNA), was obtained from Integrated DNA Technologies (Coralville, Iowa). The 6-mercaptohexanol (Sigma-Aldrich, St. Louis, Missouri), tris(2-carboxyethyl)phosphine hydrochloride (Molecular Probes [Invitrogen], Carlsbad, California), guanidine hydrochloride (Pierce Biotechnology [Thermo Scientific], Rockford, Illinois), fetal calf serum and calf-thymus single-stranded DNA (Sigma-Aldrich), Xerox paper, Lipitor, Neupogen, Paul Mitchell shampoo, Similac baby formula, and Chanel N° 5 perfume were off-the-shelf commercial formulations used as received.

Preparation and Characterization of E-DNA Sensors

The E-DNA sensor was constructed by assembling the MB-modified DNA stem-loop at the surface of a gold electrode (Figure 1). The polycrystalline gold disk electrodes (1.6 mm diameter, BASi, West Lafayette, Indiana) used in this study were prepared by polishing with diamond and alumina (BASi), sonicating in water, and electrochemically cleaning (a series of oxidation and reduction cycles in 0.5 M H2SO4, 0.01 M KCl/0.1 M H2SO4, and 0.05 M H2SO4) before modification with E-DNA. The probe DNA was immobilized onto the gold surface by incubating the clean electrode in 0.1 µM DNA/1 µM TCEP [tris(2-carboxyethyl)phosphine hydrochloride] in 1 M NaCl/10 mM potassium phosphate pH 7 buffer for 14–16 hours. The surface was then rinsed with water and subsequently backfilled with 1 mM 6-mercaptohexanol in 1 M NaCl/10 mM potassium phosphate buffer, pH 7, for 4 hours. The electrodes were then rinsed with deionized water before electrochemical analysis using alternating current voltammetry (ACV) at 10 Hz frequency, 25 mV amplitude, over a potential range of −0.1 to −0.4 V, using a CHI 603 potentiostat (CH Instruments, Austin, Texas) in a standard cell with a platinum counter electrode and Ag/AgCl (3 M NaCl) reference electrode (BASi, West Lafayette, Indiana). All experiments were conducted in 1 M NaCl/10 mM potassium phosphate buffer, pH 7 (“buffered saline”). After use, the sensing electrodes were rinsed with room-temperature Millipore water (Millipore, Billerica, Massachusetts) or, in the case of blood serum samples, 8 M guanidine hydrochloride, for 30 seconds in order to regenerate the sensor before it was reemployed to test the next sample.

Marking and Detection

Each sample was “labeled” by mixing a small aliquot with either 200 ng (39 pmol) of a DNA molecular marker (with the exception of the paper sample in Figure 3, for which 20 ng (3.9 pmol) of DNA was employed) or 20 ng (3.7 pmol) of an LNA molecular marker suspended in 0.5 µL of buffered saline. To mimic the marking of a document, the molecular marker solution (with or without 1000X excess of calf-thymus “masking DNA” inserted to render copying by cloning or amplification more difficult) was suspended in 0.5 µL of buffered saline and spotted onto a 5- by 5-mm piece of paper and allowed to dry before being extracted at room temperature by submerging the paper in 200 µL of buffered saline for 10 minutes. For the solid samples, 25 mg of a ground Lipitor tablet (~1/8 of a tablet) or 10 mg of powdered Similac formula, respectively, was mixed with the appropriate molecular marker in 0.5 µL of buffered saline and allowed to dry before being brought to 200 µL with buffered saline plus 5% glycerol for analysis, to yield final marker levels of 8 (or 0.8 for LNA) and 20 (or 2 for LNA) parts per million (ppm, mass-to-mass ratio), respectively. The Paul Mitchell shampoo and Chanel N° 5 perfume were prepared by adding the molecular marker in 0.5 µL of buffered saline to 100 µL of sample before being diluted twofold with buffered saline plus 5% glycerol before analysis, to yield a final marker concentration of 2 (or 0.2 for LNA) ppm (mass-to-volume ratio). The Neupogen was labeled by adding the molecular marker in 0.5 µL of buffered saline to 200 µL of the liquid pharmaceutical and analyzed without further dilution, giving marker levels of 1 ppm for the DNA and 0.1 ppm (mass-to-volume ratio) for the LNA. All samples were either tested immediately upon labeling or stored for two months in the dark at room temperature (with the exception of the Neupogen samples, which were stored at 4 °C in accordance with the manufacturer’s recommendations) in air-filled microfuge tubes or Petri dishes sealed with Parafilm (Pechiney Plastic Packaging, Menasha, Wisconsin). The labeled serum samples were prepared by adding 2 µL of a 0.1 mM solution (~1 µg) of the appropriate molecular marker to 500 µL of fetal calf serum before storage in dark, Parafilm-sealed microfuge tubes at 4 °C and at room temperature. After storage, the samples were diluted twofold with buffered saline to give a final marker concentration of 200 nM.

Detection of the markers was carried out at room temperature by applying 5 µL of the appropriate buffer-diluted sample onto the electrode and incubating for 5 minutes (paper sample) or 30 minutes (all other samples). Control experiments with the relevant markers in buffered saline indicate that complete sensor equilibration is achieved within this time frame (Figure 2). Once hybridized, all electrodes were rinsed with buffered saline before being interrogated in buffered saline using ACV as described above. This ex situ detection approach, in which detection is conducted in buffered saline after the sensor is challenged with and then removed from the sample, was employed in order to protect the silver chloride reference electrode from contamination; previous studies indicate, however, that the E-DNA sensor itself is robust to such conditions and can be deployed, for example, directly in blood serum (Lubin et al. 2006). Reported values represent the mean and standard deviations of measurements conducted using at least three independent electrodes.

Figure 2: The response of E-DNA over time to 200 nM of DNA and 20 nM of the mixed polymer of DNA and the nonnatural DNA analog, LNA. E-DNA-based detection of DNA and LNA molecular markers occurs within minutes for both markers and is most sensitive to LNA detection. At 200 nM (the concentration of DNA we have employed for marking), the sensor is fully equilibrated in less than 10 minutes. Even at the lower concentration of LNA we have employed (20 nM), we observe a robust signal change in less than 10 minutes and complete sensor equilibration within 40 minutes.

Results

The E-DNA sensor appears well suited for the detection of DNA-based markers. To show this, we have marked and “authenticated” a number of physical goods, including office paper (as a proxy for documents); the prescription drugs Lipitor and Neupogen, both of which have been the subject of recent counterfeiting efforts (Fishman 2005; Grady 2003; Petersen 2001; Saul 2005); the toiletries Paul Mitchell shampoo and Chanel N° 5 perfume and Similac brand infant formula, all of which have reportedly been the subject of significant counterfeiting or product “diversions” (Associated Press 1995; Burros 1995; Strugatch 2001); and, finally, fetal calf serum (a safe and convenient proxy for tissue samples). We spiked each of these materials with 200 ng (39 pmol, corresponding to a final concentration of 1–20 ppm or 200 nM) of a 17-base, single-stranded synthetic DNA oligonucleotide and stored the marked goods at room temperature in air (in the dark) for varying lengths of time. After storage, we either extracted the molecular marker with a 10-minute saline buffer soak (for the paper samples); suspended the sample in 200 µL of buffered saline (Lipitor, Similac); diluted the sample 50% with buffered saline (shampoo, perfume, blood serum); or used the sample directly without further processing (Neupogen). Of note, the E-DNA is not harmed by deployment in these materials; after testing the sensor against one sample, the same sensor was simply rinsed with water to regenerate and was challenged again with the next sample. Our results follow.

Authenticating Documents

To test the feasibility of E-DNA-based authentication of documents (such as letters and product packaging), we deposited the 17-base, single-stranded DNA molecular marker embedded in a 1000-fold excess of single-stranded calf-thymus masking DNA (introduced to thwart copying by cloning and sequencing [Clelland et al. 1999]) on a piece of standard office paper. After recovering the molecular marker by soaking in room-temperature buffered saline and incubating the E-DNA sensor in the extract for 5 minutes, we observed a 30.6% reduction in signal. In contrast, the E-DNA signal remained essentially unchanged when a DNA spot containing only 20 µg of single-stranded masking DNA was similarly tested (Figure 3), demonstrating that the sensor is unaffected by the masking DNA (Lubin et al. 2006).

Figure 3: The E-DNA sensor readily detects DNA-based molecular markers embedded in paper documents. Shown are AC voltammograms from an E-DNA sensor before and after incubation with the saline elutant from paper containing either the masking DNA plus 20 ng (3.9 pmol) of a sequence-specific DNA marker or 20 µg of a random masking DNA inserted to thwart copying of the molecular marker by sequencing or cloning. E-DNA provides a rapid (these data were collected after only 5 minutes of hybridization to the sensor), sensitive, and highly specific means of detecting DNA-based molecular markers.

Molecular Marker Detection in Complex Matrices

In addition to detecting molecular markers from materials (such as paper) that are relatively free of contaminants, the selectivity of the E-DNA sensor (Lubin et al. 2006) suggests that it could also serve as a convenient platform for the detection of oligonucleotide markers mixed within the matrices present in pharmaceuticals, consumer goods, and foodstuffs. Using case studies for such applications, we explored the application of E-DNA for the detection of molecular markers contained within two widely used pharmaceuticals: Lipitor tablets, a cholesterol-lowering drug produced by Pfizer, and Neupogen, an injectable neutropenia therapy produced by Amgen, both of which have been the targets of recent counterfeiting efforts (Fishman 2005; Grady 2003; Petersen 2001; Saul 2005). We spiked small samples of Lipitor (ground) and Neupogen (liquid) with a DNA molecular marker at concentrations of 8 and 1 ppm, respectively, before recovering the molecular marker by suspension (Lipitor) or directly testing the drug (Neupogen). We find that, whereas in samples without added markers we do not observe any significant signal change (0.6 ± 2.2 and 0.6 ± 0.7%, respectively), samples containing the markers produced 39.6 ± 4.9 and 33.6 ± 3.8% suppression of the sensor signal, respectively (Figure 4, Table 1).

Figure 4: Samples of various consumer products labeled with DNA-based molecular markers at concentrations ranging from 1 to 20 ppm and detected using the E-DNA sensor. As shown, robust and specific E-DNA signals are observed from all samples even after two months’ storage under ambient conditions (under air at room temperature, except for Neupogen, which was stored at 4 ËšC in accordance with the manufacturer’s recommendations). The samples were tested via either room-temperature aqueous extraction (for example, via a 10-minute soak in buffered saline for the paper sample), via suspension or dilution with buffered saline (for the consumer goods and Lipitor), or via direct application to the sensor (for the liquid drug Neupogen). See Table 1 for data.

Table 1: Detection of DNA and LNA Molecular Markers in Various Consumer Goods

In addition to documents and pharmaceuticals, a wide range of other merchandise—including brand-protected products such as baby formula, shampoos, and perfumes—has been the subject of counterfeiting efforts (Associated Press 1995; Burros 1995; Hopkins et al. 2003; Strugatch 2001). Thus motivated, we have tested the feasibility of using E-DNA to detect DNA-based markers stored in these materials at concentrations of either 2 ppm (shampoo, perfume) or 20 ppm (baby formula). Incubating the sensor in either buffer suspensions (formula) or twofold dilutions (shampoo, perfume) of these materials produces 29.8 ± 7.0, 27.9 ± 4.9, and 23.7 ± 5.3% drops in signal, respectively, demonstrating again that these items can be authenticated conveniently using this method (Figure 4, Table 1).

DNA Marker Stability Under Storage Conditions

Many applications of molecular markers require that the marker be able to withstand long-term storage prior to detection. Thus motivated, we stored spiked samples of each of the six above-described materials for two months at room temperature (with the exception of Neupogen, which was stored at 4 °C following the manufacturer’s guidelines) prior to detection. The observed signal suppression arising from the stored samples was effectively indistinguishable from that observed for freshly made samples (the slight drop in observed signal suppression for the samples stored for 60 days versus those measured on day zero may be due to the current limitations in sensor fabrication; similar signal variations are commonly observed between different batches of sensors [Lubin et al. 2006]), thus indicating the suitability of DNA as a long-term molecular marker for many applications (Figure 4, Table 1).

LNA as a Molecular Marker

As demonstrated by the above results, DNA appears to be an effective molecular marker for documents, pharmaceuticals, and consumer goods. In contrast, however, DNA is not suitable for marking nuclease-contaminated materials such as tissue samples, such as those employed forensically (e.g., where molecular markers might serve to improve the security of chains of evidence), or for tissue banking (where molecular markers can serve as secure labels) (Paoli 2005). Consistent with this concern, we find that we cannot detect DNA-based markers after 10 days of incubation at 4 ËšC in blood serum (Figure 5). The E-DNA-sensing technology, however, does not require the action of enzymes and thus allows us to overcome this limitation by employing enzyme-resistant, nonnatural DNA analogs for marking purposes. We have done so using molecular markers composed of a mixed polymer of DNAs and LNAs, an analog of DNA that is similar in its structure but contains a 2′-O, 4′-C methylene bridge that “locks” the ribose in a C3′-endo conformation. Importantly, while mixed LNA/DNA oligonucleotides support Watson-Crick hybridization, they are extremely resistant to enzymatic degradation (Frieden et al. 2003; Koch 2003; Wengel et al. 2003). Consistent with this, we find that LNA-based markers can be stored in blood serum (at a concentration of 40 nM) for more than 3 weeks at room temperature and more than 16 weeks at 4 °C without any significant decrease in observed E-DNA signal (Figure 5).

Figure 5: Molecular markers composed of a mixed polymer of DNA and the nonnatural DNA analog, LNA, exhibit improved stability in biological materials over pure DNA markers. Whereas DNA-based molecular markers are stable in many applications, they are unsuitable for use in nuclease-containing samples such as tissue; we do not, for example, observe any significant signal arising from 200 nM DNA markers stored for 10 days in blood serum. Molecular markers (at 20 nM) composed of a mixed polymer of DNA and the nonnatural DNA analog, LNA, however, do not exhibit any significant degradation after even 25 days’ storage in blood serum at room temperature or 116 days at 4 °C. Notably, whereas the enzymatic methods traditionally employed for the detection of DNA cannot be used to detect nonnatural analogs, the E-DNA sensor is more sensitive to LNA than to DNA.

In addition to its improved stability, LNA offers two other advantages in terms of its suitability as a molecular marker. First, the modified sugars employed in LNA increase the affinity with which it hybridizes to DNA (Singh and Wengel 1998), and thus the limit of detection for LNA is lower than that of DNA (e.g., Figure 2). Second, because LNA does not interact with enzymes, LNA-based markers are impossible to “crack” via cloning and sequencing. Motivated by these attributes, we have also studied the suitability of LNA as a means of molecularly marking physical goods. We find that we can readily detect mixed LNA/DNA markers at concentrations as low as 200 ppb in the consumer goods and pharmaceutical samples employed above, and we do not observe any detectable degradation after three months’ storage in these materials at room temperature (Figure 6, Table 1).

Figure 6: LNA also can be used as an effective molecular marker for consumer goods. Shown here are the E-DNA responses from materials doped with 0.1 to 2 ppm of LNA and stored at room temperature (4 °C for the Neupogen) for three months. The smaller suppression observed with the Neupogen sample might arise from interaction between the LNA and the drug/components of Neupogen. See Table 1 for data.

Discussion

The E-DNA sensor provides a convenient solution to the problem of detecting molecular markers based on oligonucleotides or nonnatural oligonucleotide analogs. For example, whereas our preliminary demonstration of document “marking” was performed using paper stored for up to two months, previous studies have shown that DNA encapsulated in typical letter paper (Cook and Cox 2003) and in ink (Butland and Baggot 1998) is stable for years at room temperature. This, in addition to the general longevity of DNA and its use in long-term storage of information (Bancroft et al. 2001), suggests that the E-DNA approach is applicable to the widespread authentication of documents and paper-based or other packaging materials. Similarly, the response of the sensor to marked oral and injectable drugs, foodstuffs, and toiletries—even after long-term storage under ambient conditions—indicates that the E-DNA-sensing strategy is also applicable to the task of detecting molecular markers in a variety of complex, contaminant-ridden substances. Finally, under conditions in which DNA-based markers succumb to enzymatic degradation, we can readily detect markers composed of the nonnatural DNA analog, LNA, even, for example, after long-term storage in blood serum.

The E-DNA sensor offers many substantial advantages over existing PCR- (Clelland et al.1999), electrophoretic- (Cook and Cox 2003), or fluorescence-based (Tyagi and Kramer 1996) approaches to the detection of oligonucleotide-based molecular markers. The high specificity of the E-DNA sensor enables the ready detection of these markers, even in the presence of multiple orders of magnitude of excess masking DNA, rendering it effectively impossible to forge the marker via PCR amplification, cloning, or other enzymatic copying methods (Clelland et al. 1999). Additionally, E-DNA detection is rapid (the extraction of the marker and the appearance of the signal occur within minutes and can be performed directly in saline dilutions of complex materials such as blood serum [Lubin et al. 2006]); inexpensive (oligonucleotides of the appropriate length cost a negligible price per microgram, and although DNA/LNA oligos are still more expensive than DNA to synthesize, given the nanogram quantities employed in marking, even the currently higher price of LNA does not appear prohibitive); and convenient (the sensor is reusable [Lubin et al. 2006] and does not require extensive processing or the addition of exogenous reagents). Finally, the E-DNA device itself is an electronic, surface-immobilized detection platform that can be readily converted into a handheld microarray device that is both user-friendly and cost-effective. Although here we use a single pixel labeled with one E-DNA recognition probe, microarrays that would allow monitoring with multiple different E-DNA recognition probes are quite feasible with this system and something we are currently developing. For example, we have differentially labeled a dual-pixel sensor chip (Lai et al. 2006) with two different E-DNA probes in order to simultaneously detect different target sequences (manuscript under submission). Similarly, the cost of fabricating E-DNA sensors is quite reasonable, because they are amenable to mass production. We are currently working to develop a handheld, battery-operated device that would be inexpensive and portable.

A significant advantage of the molecular marking approach described here is that DNA and, even more so, LNA molecular markers can be made arbitrarily difficult to counterfeit. For example, attempts to copy DNA molecular markers by cloning and sequencing (or PCR amplification and sequencing) can be rendered arbitrarily difficult by increasing the amount of masking DNA employed; thus even if one could sequence the few nanograms of DNA employed in marking, one would not know which of the many hundreds of thousands of sequences present was the correct sequence without laboriously synthesizing and testing each one. LNA-based markers would be even harder to crack, because they are entirely recalcitrant to the enzymatic processes employed in sequencing and cloning. Nor can DNA or LNA markers be thwarted via the use of mass-spectrometric sequencing methods (Edwards et al. 2005), because these are ensemble techniques and thus using them to sequence a complex mixture of a molecular marker “buried” under multiple orders of magnitude of excess masking DNA/LNA would return an ambiguous sequence. Given these advantages, we believe that the E-DNA-based detection of oligonucleotide markers may prove useful in a wide range of forensic applications.

Acknowledgments

We thank Roger C. Dunham, M.D., for discussions on the importance of pharmaceutical authentication and for providing the pharmaceutical samples used in this study. We also thank Brook Vander Stoep Hunt and Sophie Hanscom for their assistance with the sample preparation. This research received partial support from the National Science Foundation (NSF-DMR-0099843) and a National Institutes of Health grant (GM 62958-01), as well as support from the Institute for Collaborative Biotechnologies through grant DAAD19-03-D-0004 from the U.S. Army Research Office.

References

Associated Press. Baby formula counterfeited; Man is held, New York Times, February 12, 1995, late ed., section 1, p. 29.

Associated Press. Nineteen indicted in Florida in case of phony drugs, New York Times, July 22, 2003, late ed., p. A14.

Baker, T. and Thomas, R. Firms recall counterfeit surgical mesh product, Biomedical Safety & Standards (2004) 34:33–35.

Baker, T. and Thomas, R. Medical device recalls continue to soar, Biomedical Safety & Standards (2005) 35:113–115.

Bancroft, C., Bowler, T., Bloom, B., and Clelland, C. T. Long-term storage of information in DNA, Science (2001) 293:1763.

Blume, M. It’s true: These museum exhibits are fake, International Herald Tribune [Online]. (February 9, 2006).

Buckley, C. U.S. groups press China over piracy, New York Times, September 17, 2004, p. W1.

Burros, M. Eating well; F.D.A. target: Baby formula, New York Times, September 6, 1995, late ed., p. C1.

Butland, C. L. and Baggot, B. (DNA Technologies). Labeling Technique for Countering Product Diversion and Product Counterfeiting. U.S. Patent 6,030,657, February 29, 1998.

Clelland, C. T., Risca, V., and Bancroft, C. Hiding messages in DNA microdots, Nature (1999) 399:533.

Commission on Physical Sciences, Mathematics, and Applications. Containing the threat from illegal bombings: An integrated national strategy for marketing, tagging, rendering inert, and licensing explosives and their precursors. In: Taggants for Preblast and Postblast Identification of Explosives. National Academy Press, Washington, D.C., 1998, pp. 73–95.

Confessore, N. No-name, brand-name or phony: It’s all here, New York Times, October 9, 2006, late ed., p. B1.

Cook, L. J. and Cox, J. P. L. Methylated DNA labels for marking objects, Biotechnology Letters (2003) 25:89–94.

Cox, J. P. L. Bar coding objects with DNA, The Analyst (2001a) 126:545–547.

Cox, J. P. L. Long-term data storage in DNA, Trends in Biotechnology (2001b) 19:247–250.

Deisingh, A. K. Pharmaceutical counterfeiting, The Analyst (2005) 130:271–279.

Edwards, J. R., Ruparel, H., and Ju, J. Mass-spectrometry DNA sequencing, Mutation Research (2005) 573:3–12.

Fan, C., Plaxco, K. W., and Heeger, A. J. Electrochemical interrogation of conformational changes as a reagentless method for the sequence-specific detection of DNA. In: Proceedings of the National Academy of Sciences (2003) 100:9134–9137.

Farmer, S. NFL makes this a brand new ball game, Los Angeles Times, February 2, 2006, home ed., p. D1.

Fifield, A. The knock-off squad, New York Times, June 23, 2002, final ed., section 14, p. 1.

Fishman, T. C. Manufaketure, New York Times, January 9, 2005, late ed., section 6, p. 40.

Frieden, M., Hansen, H. F., and Koch, T. Nuclease stability of LNA oligonucleotides and LNA-DNA chimeras, Nucleosides, Nucleotides & Nucleic Acids (2003) 22:1041–1043.

Galbraith, R. Made in Italy: Counterfeits that were once mere imports, International Herald Tribune [Online]. (October 3, 2006).

Grady, D. F.D.A. outlines plans to counter growing trade in counterfeit pharmaceuticals, New York Times [Online]. (October 3, 2003).

Groubert, M. FBI puts squeeze on starlets—& their bootleg implants! Weekly World News [Online]. (May 19, 2004).

Hafner, K. Seeing fakes, angry traders confront eBay, New York Times, January 9, 2006, late ed., section 1, p. 1.

Hermanson, G. T. Bioconjugate Techniques. 1st ed., Academic Press, San Diego, California, 1996.

Hopkins, D. M., Kontnik, L. T., and Turnage, M. T. Counterfeiting Exposed: Protecting Your Brand and Customers. 1st ed., John Wiley & Sons, Hoboken, New Jersey, 2003.

Kapur, B. G., Hutson, R. H., Davis, M. A., and Rice, P. L. The United States twenty-year experience with bombing incidents: Implications for terrorism preparedness and medical response, Journal of Trauma Injury, Infection, and Critical Care (2005) 59:1436–1444.

Koch, T. Locked nucleic acids: A family of high affinity nucleic acid probes, Journal of Physics: Condensed Matter (2003) 15:S1861–S1871.

Kramer, A. E. Drug piracy: A wave of counterfeit medicines washes over Russia, New York Times [Online]. (September 5, 2006).

Lai, R. Y., Lee, S., Soh, H. T., Plaxco, K. W., and Heeger, A. J. Differential labeling of closely spaced biosensor electrodes via electronic lithography, Langmuir (2006) 22:1932–1936.

Lebacq, P. (Bioprobe Systems). Method and Apparatus for High Security Crypto-Marking for Protecting Valuable Objects. U.S. Patent 5,139,812, August 18, 1992.

Leutwyler, K. Fingerprinting Super Bowl footballs, Scientific American [Online]. (January 26, 2001).

Levin, C. Fake auto parts have real consequences, U.S. Senate Press Release [Online]. (October 6, 2006).

Lubin, A. A., Lai, R. Y., Baker, B. R., Heeger, A. J., and Plaxco, K. W. Sequence-specific, electronic detection of oligonucleotides in blood, soil, and foodstuffs with the reagentless, reusable E-DNA sensor, Analytical Chemistry (2006) 78:5671–5677.

May, J. J & J issues warning on counterfeit test strips, The Star Ledger [Newark, New Jersey], October 14, 2006, final ed., business section, p. 28.

Nighswonger, G. Pursuing counterfeit medical devices, Medical Device & Diagnostic Industry [Online]. (January 2003).

Paoli, P. D. Biobanking in microbiology: From sample collection to epidemiology, diagnosis, and research, FEMS Microbiology Review (2005)29:897–910.

Petersen, M. Three fake drugs are found in pharmacies, New York Times [Online]. (June 5, 2001).

Report of the Food and Drug Administration: Combating Counterfeit Drugs. Food and Drug Administration, Rockville, Maryland, 2004, p. 54.

Sabir, I. H., Torgersen, J., Haldorsen, S., and Aleström, P. DNA tracers with information capacity and high detection sensitivity in groundwater detection, Hydrogeology Journal (1999) 7:264–272.

Saul, S. F.D.A. hoping for indictment over fake pills, New York Times [Online]. (July 26, 2005).

Schultz, D. G. F.D.A. public health Web notification: Counterfeit polypropylene mesh, Food and Drug Administration [Online]. (May 7, 2004).

Shepardson, D. Carmakers: Stop flow of phony parts, The Detroit News [Online]. (April 20, 2006). Available: http://www.detnews.com/apps/pbcs.dll/article?AID=20060420.

Singh, S. K., Nielsen, P., Koshkin, A. A., and Wengel, J. LNA (locked nucleic acids): Synthesis and high-affinity nucleic acid recognition, Chemical Communications (1998) 4:455–456.

Strugatch, Warren. L.I. @ Work, New York Times, January 28, 2001, late ed., section 14, p. 6.

Tyagi, S. and Kramer, F. R. Molecular beacons: Probes that fluoresce upon hybridization, Nature Biotechnology (1996) 14:303.

Vergano, D. Dab of DNA helps keep the counterfeiters at bay, USA Today, October 5, 2000, p. D9.

Wald, M. L. F.A.A. to train inspectors to detect counterfeit aircraft parts, New York Times, October 13, 1995, late ed., p. A17.

Wengel, J., Petersen, M., Frieden, M., and Koch, T. Chemistry of locked nucleic acids (LNA): Design, synthesis, and bio-physical properties, Letters in Peptide Science (2003) 10:237–253.

Wu, C. Tagged out: New markers for explosives may lay old questions to rest, Science News (1996) 150:168–171.