From molecular diagnostics to personalized testing - Future Medicine

R EVIEW

For reprint orders, please contact: [email protected]

From molecular diagnostics to personalized testing Jon E Finan1 & Richard Y Zhao1,2,3† †Author

for correspondence of Maryland School of Medicine, Department of Pathology, Baltimore, MD 21201, USA Tel.: +1 410 706 6301 Fax: +1 410 706 6303; E-mail: rzhao@ som.umaryland.edu 2University of Maryland School of Medicine, Department of MicrobiologyImmunology, Baltimore, MD 21201, USA 3University of Maryland School of Medicine, Institute of Human Virology, Baltimore, MD 21201, USA 1University

Keywords: automation, gene-based amplification, genomics, molecular diagnostics, nanotechnology, personalized medicine, proteomics, single-molecule detection part of

Gene-based molecular diagnostics is changing the practice of medicine and will continue to do so for the foreseeable future. The major underlying principle of these diagnostic tests is the use of specific nucleic acid sequences as surrogates; amplification of the surrogate markers enables the detection of pathogens or disease-related gene mutations. Gene targets can be amplified by target-, probe- or signal-based methods. Combined use of nucleic acid amplification and enzyme-linked immunosorbent assay with methods such as immuno-polymerase-chain reaction allows us to detect protein at femtogram (10–15 g) levels. A variety of choices are available for the detection of amplified amplicons with the fluorophore-linked nanoparticles as the most sensitive markers. The unique advantages of using covalently-linked nanoparticles include the detection of single molecules, the ability to enrich molecules of interest with unprecedented detection sensitivity (up to zeptogram levels, 10–21 g) and the flexibility of multiple functionalization. Automation appears to be the current trend for high-volume molecular testing of infectious diseases. Molecular profiling of various diseases using genomic or proteomic approaches opens up a molecule wonderland with promise and emergence of new molecular testing that will likely impact the practice of medicine to a greater degree in the future. The future of molecular-based testing and the journey toward personalized testing will be discussed.

Molecular diagnostics can be broadly defined as a major division of laboratory medicine where the advances in molecular biological techniques are applied to the clinical laboratory. Currently, the techniques in use are principally for the detection of nucleic acids (DNA or RNA) associated with pathogens, cancer and inherited diseases. As the field itself has widened, so too has the number of technologies that are being applied to the detection of these disease-specific genes. Detection of proteins in conjunction with nucleic acid-based technology could also be included in future molecular diagnostics. The current technologies, along with future technologies, in various stages of development, will certainly change the practice of medicine as we know it. The goals of this review are to briefly summarize many of the gene-based molecular technologies currently used in the molecular diagnostic laboratory, preview some of the techniques on the horizon and provide perspectives on the future of molecular testing and the potential applications in personalized medicine. The invention of polymerase chain reaction (PCR) and recombinant DNA technology revolutionized and began the modern era of molecular biology. The application of restriction fragment length polymorphism analysis for the

10.2217/14622416.8.1.85 © 2007 Future Medicine Ltd ISSN 1462-2416

diagnosis of sickle cell anemia and thalasemia [1] was the genesis of molecular diagnostics as a discipline. From that beginning, multiple new and exciting gene-based methodologies have emerged. These methods have broad applications in medicine, but have predominately been applied to infectious diseases, hematology, oncology and genetic diseases. Gene-based molecular diagnostics, utilizing detection and amplification of disease-specific nucleic acids, has numerous advantages over conventional testing. The speed and specificity that comes with the use of nucleic acids as biomarkers is among the most prominent advantages. The technologies also allow for a large dynamic range of detection, making most of the assays both qualitative and quantitative. Gene-based assays currently have the largest number of clinical applications within infectious diseases. There are multiple reasons for this, but chief among these is the difficulty and hazards associated with cultivating pathogens. Molecular techniques have largely eliminated the need for culturing certain pathogens. In addition, pathogens that cannot be cultured can also be detected by the use of pathogen-specific gene sequences. Within virology, the time needed for diagnosis has been greatly decreased. For example, the Pharmacogenomics (2007) 8(1), 85–99

85

REVIEW – Finan & Zhao

diagnosis of HIV infection in a newborn takes months based on serologic methods and is complicated by the presence of maternal antibodies. HIV-specific reverse transcription (RT)-PCR allows the diagnosis to be made on the first day of life. Other examples include the diagnosis of herpes simplex virus and cytomegalovirus (CMV)-induced meningitis, which would normally require the virus to be cultured from cerebrospinal fluid, but can now be made by detection of virus-specific nucleic acids directly within cerebrospinal fluid. With molecular testing, the beginnings of personalized medicine can be seen in oncology care with respect to cancer risk, detection and treatment monitoring. Patients with appropriate family history of breast and/or ovarian cancer can be tested for breast cancer (BRCA)1/2 gene mutations [2]. The presence of such mutations would obviously lead to different screening, counseling and medical care for the individual. The detection of malignancy is aided by the use of assays such as T-cell gene rearrangement. This molecular test can greatly aid a pathologist in distinguishing a malignancy from a benign proliferation of lymphocytes. The monitoring of gene products specific to a patient’s malignancy, such as breakpoint cluster region-Abelson (BCR-ABL) in chronic myelogenous leukemia, can assist in assessing treatment response and continued presence of disease. Finally, we are now able to predict with great confidence recurrence of certain cancers, such as breast cancer by using a multiple gene panel in the Oncotype DX™ test that is derived from genome-wide evaluation of global gene transcription profiles. Overall, the applications for gene-based technologies are growing rapidly and will undoubtedly affect all aspects of medicine. In this review, we summarize principles and give examples of various gene-based amplification technologies and automated platforms using some of these technologies; the combination of gene-based technology with conventional enzyme-based electroimmunoassay method for highly sensitive detection and quantification of proteins is described and applications of nanotechnology and genomic/proteomic approaches in the future of molecular testing are postulated. Finally, gene-based molecular testing that is currently commercially available and aimed to provide diagnostic or prognostic analysis of individual susceptibility to various diseases is discussed. 86

Pharmacogenomics (2007) 8(1)

Gene-based molecular diagnostics Gene-based molecular diagnostics involves the following three major steps: primer–template hybridization, synthesis and amplification. It is the method of amplification that most easily delineates the different assays in use. The amplification systems can be divided into target-, probe- or signal-based methods (for specific examples of tests in each category, see Table 1). As their names imply, target-based systems amplify the direct target gene sequence, probe-based systems rely on the amplification of the annealed probes, whereas signal-based systems amplify the detection signal itself. The amplification systems in use today can also be divided into PCR-based as opposed to non-PCR-based methodologies. PCR-based amplification requires the use of a thermocycler to carry out artificial ‘cycles’ of amplification at varying temperatures and in the presence of a heat-stable DNA polymerase. The non-PCRbased methods of amplification are mostly based on natural nucleic acid amplification and are generally performed at a single temperature. This difference eliminates the requirement of thermocylcing and simplifies the performance of the assays. Regardless of the amplification method in use, it is theoretically possible to detect a single molecule of DNA or RNA and amplify it to 109–1012 molecules within a few hours’ time. Due to the impurities and the complex nature of clinical samples, this translates into a real world ability to detect 10–150 copies of a target gene sequence with a high degree of probability. Target-based amplification systems

PCR is the prototypal example of target-based amplification technology. The basic premise when applied to the diagnostic molecular laboratory is for the detection of a disease- or pathogen- specific nucleic acid sequence. For each pathogen, a unique sequence of DNA or RNA is identified and used for amplification. The presence of amplified product, also known as amplicon, in its most simple form indicates the existence of the pathogen of interest within the tested sample. Protocols for a multitude of microorganisms exist throughout the literature. These range from bacteria to viruses and from fungi to parasites. These ‘home-brew’ systems require significant validation and the following of rigorous quality assurance/quality control standards prior to use in the clinical laboratory setting. In addition, future science group

From molecular diagnostics to personalized testing – REVIEW

Table 1. Nucleic acid-based amplification methods. Methods

Amplification targets

Mimicking natural process of

Key required features

Ref.

PCR/RT-PCR

DNA/RNA

n/a

Thermocycling, DNA polymerase/RT

NASBA

RNA

Reverse transcription

RT, RNA polymerase, RNase H

[12]

TMA

RNA

Reverse transcription

RT with RNase activity; RNA polymerase

[14]

Target-based [1]

SDA

DNA

Excision DNA repair

Endonuclease

[17]

RCA

Circular DNA

Plasmid replication

DNA polymerase

[51]

HDA

DNA

DNA unwinding

DNA helicase

[52]

LCR

DNA

DNA ligase

DNA ligase

[16]

Qβ-replicase

RNA

Bacteriophage replication

Qβ-replicase, RNaseIII

[53]

bDNA

RNA/DNA

n/a

Multimer amplifier

[19]

BCA

DNA

n/a

Magnetic enrichment; bio-barcodes detection

[54]

Probe-based

Signal-based

BCA: Bio-barcode assay; bDNA: Branched DNA; HAD: Helicase-dependent isothermal DNA amplification; LCR: Ligase chain reaction; n/a: Non-applicable. NASBA: Nucleic acid sequence-based amplification; PCR: Polymerase-chain reaction; RCA: Rolling circle amplification; RNase: Ribonuclease; RT: Reverse transcription; SDA: Strand displacement assay, TMA: Transcription-mediated assay.

provisions for proper specimen collection are also needed. Such provisions must include blood collection into acid citrate dextrose (yellow top) or ethylene diamine tetra-acetic acid (purple top) containing tubes instead of heparin (green top) that inhibits the PCR reaction. Additional concerns that must be addressed when using PCR in the clinical setting include those centering on cross-contamination. Due to the large number of amplicons generated by each reaction, the laboratory must have good laboratory practices in place to assure postamplified product does not contaminate preamplified specimens and thereby cause false-positive reactions. Such practices as separate, dedicated areas of the laboratory for pre- and post-amplification, as well as unidirectional operations, can help minimize this concern. Commercial Amplicor® PCR-based kits (Roche Diagnostics, Basel, Switzerland) use uracil-N-glycosylase (AmpErase®) to break down preamplified and uracil-incorporated amplicons that effectively minimize crosscontamination. These kits are available for a number of applications including HIV-1, hepatitis B and C viruses (HBV/HCV), human papilloma virus, Chlamydia trachomatis, Neisseria gonorrhea and Mycobacterium tuberculosis. Conventional PCR is generally used for qualitative studies, such as for the presence of a given pathogen. It can, however, be converted to a quantitative assay by incorporating proper internal controls to the assays [3]. Nevertheless, due to the exponential nature of PCR amplification, future science group

www.futuremedicine.com

it has several notable shortcomings. For example, small differences in the efficiency of early amplifications cycles can translate into huge variations in the final quantity of the amplified product. This results in large interassay variability that must be corrected for. The methods for such correction are time consuming and cumbersome. In addition, even when these limitations are corrected for, the assay still has a fairly narrow dynamic range of approximately three logs. These shortcomings in conventional PCR are overcome in the real-time PCR assay. In realtime PCR, the accumulation of product is monitored in ‘real time’, that is, on a cycle-by-cycle basis [4]. The use of fluorescent probes that hybridize to a sequence within the amplified product allows not only for monitoring of the amplification process, but also increases the specificity of the assay. These probes alter their fluorescence based on whether they are bound to a specific nucleic acid sequence or free in solution (see detailed description on how the probes work below). The changes in light emission are monitored throughout the PCR reaction. The DNA copy number is based on the threshold cycle, which is the PCR cycle when the amount of product formation is in the most exponential phase of amplification. The ability to identify the exponential phase of amplification results in a much more accurate DNA copy number. This is largely due to the fact that as the PCR reaction reaches completion, the amplification is no 87


longer exponential due to the limiting of reagents, among other factors. In real-time PCR the original copy number can be calculated at a point where the original copy number is directly proportional to the threshold cycle and therefore allows for a much greater dynamic range. In addition to the advantages of quantization real-time PCR provides, it can also be used for mutational and genotyping analyses. If the internal probe is constructed in a sequence with some degree of polymorphism, the melting temperature of the template–probe complex will be slightly altered due to mismatches between the two sequences. These mismatches will alter the melting temperatures of the template–probe complex [5,6]. The differences between the wildtype and mutant genes can be detected through melting curve analysis by measuring the change in fluorescence over a varying temperature range. Currently real-time PCR probes can be divided based on the type of probe in use. The major kinds are hydrolysis probes used in realtime PCR for quantification, and hybridization probes for the detection of single nucleotide polymorphisms (SNPs) and mutations. TaqMan™ probes, molecular beacons probes and scorpion primer probes are all examples of hydrolysis probes. TaqMan probes are linear oligonucleotides with a fluorescent reporter dye at one end of the molecule and a quencher at the opposing end [7]. No signal is detected when the fluorescent reporter and the quencher are in close proximity (as when they are attached to the same oligonucleotide), since the fluorescence from the reporter is absorbed by the quencher. During amplification, the probe binds and is incorporated into the product and the 5´ nuclease activity of the polymerase cleaves the fluorescent reporter due to its position at the 5´ end of the molecule. This cleavage produces free reporter molecule within solution that is removed from its quencher and therefore detectable. Molecular beacon probes work on a similar principle [8]. Like the TaqMan probes, these probes are labeled with a fluorescent reporter at one end and a quencher at the opposite end. However, the unbound probe forms a stem loop structure that brings the two ends in close proximity to one another, thus causing the fluorescent signal to be quenched. Once the probes binds to the template the stem loop structure is eliminated and the quencher is far enough from the reporter that the fluorescence can be detected. The scorpion primer-probes are similar conceptually, yet they incorporate the probe into the primer [9]. This 88


allows unbound primer to be readily distinguished from amplified product. A different and partially double-stranded probe is introduced to the Abbott’s LCx® RealTime™ PCR assays [10]. Instead of labeling the probe with both the reporter and quencher molecules on a single strand of oligonucleotide, the reporter molecule is labeled at the 5´ end of the gene-specific probe; the quencher molecule is labeled to an oligonucleotide complementary to the 5´ end of the probe. In this configuration, signal is not created by probe hydrolysis but by separation of the probe from quencher oligonucleotide. The rationale is that the uncoupling of the probe hydrolysis from polymerase extension will give rise to clean background and hybridization of the longer probe oligonucleotide to the gene target at relative low temperature (56°C) will yield tolerance to mismatches, thus allowing less stringent hybridization and broad coverage of detection such as HIV-1 subtypes [11]. Hybridization probes are used for both the detection of mutations and for SNPs along with melting curve analysis and sequence-specific fluorescence resonance energy transfer (FRET) [5,6]. In this system, two probes are used, which are complimentary to a contiguous segment of the sequence of interest. One probe is labeled with fluorescein on its 3´ end and the second probe is labeled with LightCycler® Red 640-N-hydroxysuccinimide ester (LC Red640) on its 5´ end. Upon binding to the sequence, excitation of the florescein causes some of its emission energy to be transferred to the adjacent LC Red640 and its subsequent emission can be detected. Following this detection phase, a melting curve analysis is performed. Melting curves are based on the principle that even a single change of nucleotide (polymorphisms or mutations) will result in a change of the melting temperature of the probetemplate duplex due to mismatches between the two. A mutation will cause a lower melting temperature because less energy is needed to disrupt the probe–template duplex. With one probe spanning the area of interest (mutation probe) and the second probe on the adjacent sequence (the anchor probe) and using FRET, the melting temperature can be determined. A template with a mutation will melt at a lower temperature and consequently result in less emission from LC Red640 at that lower temperature. An example of a currently-used application of this technology is the detection of Factor V Leiden mutation detection with the LightCycler from Roche Applied Sciences (IN, USA). future science group


In addition to PCR, other target-based amplification systems include Nucleic Acid Sequence Based Amplification (NASBA) and Transcription Mediated Amplification (TMA). Unlike PCR, NASBA and TMA are based on a natural replication system, namely that of retroviral RNA replication [12,13]. With NASBA the basic reaction includes T7 RNA polymerase, RNase H, avian myeloblastosis virus, reverse transcriptase, nucleotide triphosphates, two specific primers and appropriate buffers. NASBA uses the three above enzymes to produce rapid RNA amplification through a cDNA intermediate. TMA is similar but uses just two enzymes, a reverse transcriptase with RNase H activity and a T7 RNA polymerase. Detailed descriptions of gene amplification by NASBA or TMA can be found elsewhere [12,14,15]. NASBA and TMA have two major advantages involving sample requirements and assay conditions. The assay can be performed on any type of specimen including blood, serum, cerebrospinal fluid, plasma, tissue or genital secretions. In addition, there is no specific requirement for anticoagulant, unlike PCR. The second advantage is the isothermal nature of NASBA and TMA, thus eliminating the need for a thermocycler. The major disadvantage is it can only be used for detecting RNA targets such as HCV and HIV. Also, a two-step process involving RNA extraction is necessary. This, along with the extreme amplification strength of the assay, increases the possibility of cross-contamination. Commercial NASBA applications for detection of CMV pp67, respiratory syncytial virus (A and B), enterovirus and HIV are marketed under the trade name NucliSens™ through bioMerieux Inc. (Marcy l'Etoile, France). Commercial applications using TMA include multiple GenProbe® AMPLIFIED™ products, including assays for C. trachomatis, N. gonnorhoeae and M. tuberculosis. Probe-based amplification technologies

Probe-based amplification technologies are designed to amplify the probe homologous to the target as opposed to the target sequence itself (Table 1). These technologies include ligase chain reaction (LCR) [16] and strand displacement amplification (SDA) [17]. Although these assays have not been as broadly commercialized, they still represent a few commercial and possible future applications. LCR takes advantage of the principle that ligation of DNA molecules is most efficient when the molecules are aligned in a future science group


head-to-tail fashion [18]. Probes are designed to be complementary to a contiguous sequence that serves as a mechanism to bring the two molecules together. In the presence of DNA ligase the molecules are joined and the new ligated molecule can now also serve as a template. This continued process will result in the logarithmic accumulation of the ligation product consisting of two probe molecules. Currently, Abbott Laboratories (IL, USA) markets an HIV-1 test based on LCR, under the trade name HIV-1 LCx. SDA is based on the nature process of DNA excision repair [17]. In DNA excision and repair, a single strand nick in a double-stranded molecule results in the displacement of that strand by a newly synthesized strand. By introducing a singlestrand nick in the same site repeated, copies of that nicked strand can be produced repeatedly. In order to accomplish this, an endonuclease restriction enzyme HincII or BsoBI is used to introduce single-strand nicks. This enzyme only nicks the DNA when the strand opposite its recognition site is hemiphosphorothiolated (dCTPαS instead of dCTP). Through primers designed with the recognition sites and adjacent primers it is possible to generate a template with displaced strands that are capable of serving as template in subsequent reactions. This sets up a repetitive generation of nicking and displacement that is capable a 108-fold amplification in 2 h. Based on this technology, assays for the detection of Legionella pneumophilia, C. trachomatis and N. gonorrhea are available from Becton, Dickinson and Company (NJ, USA). Signal-based amplification methods

Branched DNA (bDNA) is the most common form of signal-based amplification technology in use [19]. This technique utilizes several hybridization steps to result in significant amplification of a detection signal as opposed to a template or a probe. Branched DNA first uses a probe that is bound to a solid surface. Once bound, an extended probe is added that binds to an adjacent sequence on the capture probe. From this point, bDNA molecules, known as amplification molecules, are added. These bDNA molecules contain multiple (3000 up to 22,380) branch sites per target molecule for alkaline phosphatase labeled probes to adhere to. After exposure to a chemiluminscent substrate, a signal can be detected. This assay eliminates the risks of contamination that is present in other amplification assays. In addition, the assay is simple and therefore has a low degree of 89


Figure 1. The principles of of immuno-polymerase-chain reaction and enzyme-mediated immuno-polymerase-chain reaction. B.

A.

S

P

B St E

B

P

P

ELISA

iPCR

C.

3´ p

p 3´

St

AP

B

P

em-iPCR

(A) Schematic drawing of ELISA; (B) immuno-PCR; and (C) enzyme-mediated immuno-PCR. Scissors represent λ exonuclease. AP: Alkaline phosphatase; B: Biotin; C: Color-changes; ELISA: Enzyme-linked immunosorbent assay; E: Enzyme; PCR: Polymerase-chain reaction; P: Protein; St: Streptavidin; S: Substrate.

variability. Although the format allows for larger number of tests to be performed at one time, the turn-around time is significantly longer than with PCR-based assays. 90


Protein-based molecular diagnostics

Due to the amplification powers described earlier, nucleic acids can be detected at a sensitivity level far below those for the detection of proteins future science group


by conventional immunological methods, such as the antibody-based enzyme-linked immunosorbent assay (ELISA). This limitation of protein detection can be circumvented by combination of the nucleic acids-based amplification with ELISA (Figure 1). Immuno-PCR

Immuno-PCR (iPCR) combines the versatility of the well-established ELISA methodology with the amplification power of real-time PCR [20,21]. In a conventional ELISA (Figure 1A), protein or antigen is first captured by a primary antibody onto a solid phase such as a microtiter plate. A second antibody conjugated with an enzyme binds to a second site of the protein. The level of protein can be quantified by luminometry through enzyme-mediated cleavage of the enzyme substrate, which results in light emission or color changes. The sensitivity level of ELISA is typically in the range of picograms (10–12 g). The iPCR captures the protein the same way as ELISA (Figure 1B). However, the second antibody is conjugated with biotin that, through strong binding of biotin to streptavidin, links to biotinylated oligonucleotides. In such a way, the protein level can be quantified by amplification of the nucleotides that attach to the second antibody. The sensitivity level of protein detected by iPCR is in the range of femtogram (10–15 g), which is approximately 103-times more sensitive than the conventional ELISA assays. Thus, the iPCR method has great potential for sensitive detection of proteins such as antigens. For example, iPCR has been used for quantification of HIV-1 p24 antigenemia and prion detection [22,23]. Enzyme-mediated iPCR

Although iPCR could in principle become one of the most sensitive diagnostic tools for protein detections, it is technically difficult to create the antibody–DNA conjugates. There are also increasing concerns about the substantial backgrounds it generates due to nonspecific binding of the antibody-DNA conjugates to the solid phases. To overcome these problems, an enzyme-mediated iPCR (em-iPCR) method was developed [24]. In this assay (Figure 1C), the protein or antigen is captured the same way as ELISA; the second antibody is, however, conjugated with alkaline phosphatase (AP). Upon binding of the second antibody to protein, a double-stranded 5´-phosphorylated oligonucleotide substrate for AP is added. Dephosphorylation of the oligonucleotide by AP future science group


renders it resistant to λ exonuclease, which only degrades 5´ phosphorylated DNA, and thus is available for DNA amplification and quantification (Figure 1B). In contrast, if AP-conjugated second antibody fails to bind the protein, AP will be washed away from the reaction. In the absence of AP, λ exonuclease will rapidly degrade the oligonucleotides leaving no DNA template for amplification and detection. The em-iPCR should in principle minimize the intrinsic problems presented in iPCR. However, its actual performance and utility in molecular testing is yet to be tested. No commercial method has been developed. Detection methods Detection of amplicons Once amplified, the target sequence must be detected. This can be accomplished through a number of methods including radioisotopes, enzyme reporter molecules, antigenic substrates, chemiluminescent moieties and fluorescent labels. The original method for detection was the use of radioactive 32phosphorus or 35sulfur probes that could be hybridized to the recently amplified target sequence and visualized by using autoradiography. However, radioactive materials are biohazards and have short life spans. These methods have given way to enzyme reporters, chemiluminescent molecules and fluorescent probes. Chemiluminscent molecules are chemical groups that produce light after exposure to a particular substrate. Examples include acridinium esters that can be incorporated in nucleic acid probes and detected upon exposure to their substrate. These chemical groups sometimes surpass the sensitivity of radioisotope labels. Enzyme reporters such as AP and horseradish peroxidase can be incorporated into oligonucleotide probes that become hybridized to a specific target. Following hybridization, the complex is exposed to the enzymes substrate that has been linked to a colorimetric or chemiluminescent moiety. Cleavage of the enzyme substrate results in light emission that can be detected by luminometry. These systems not only detect the sequence of interest but also amplify the signal as in the case of branched DNA. Although enzymatic or chemiluminescent reporters are commonly used, they typically have limited choice of detection colors and short life span of activity. Increasingly, the detection of amplified products is being accomplished with the use of fluorescent dyes known as fluorophores. These dyes can be excited by a narrow range of light and 91


consequently will emit light at a different wavelength. The emission is easily detected by a detector which is built into the device the fluorophore is being used in. The five most common fluorophores in use include FAM™ (SYBR® Green I), JOE (VIC™), TAMRA™ (NED™ or Cy3™), ROX™ (Texas Red®), and Cy5™. Each one of these dyes has a separate pattern of excitation and emission that allows for simultaneous detection. This can provide for detection of five different amplification products at once. The most common use of fluorophores is with the covalent linkage to a primer or probe sequence. Once the primer or probe is incorporated into the product the increase in fluorescent can be detected in real time. This approach allows for quantification of the product during each cycle of real-time PCR. Single molecule detection & enrichment by nanoparticles

Recent technologies for detection of biological molecules include fluorescent semiconductor nanocrystals known as quantum dots (Qdots) [25]. Qdots are unique nanoparticles that are highly stable against photobleaching. Qdots can be excited at a single wavelength but display five different colors that are controlled by its size and composition, thus allowing real-time imaging and quantitative determination of multiple molecule types present in cells. Use of Qdots in molecular diagnostic testing is still at its early stage. However, companies such as Ventana Medical Systems Inc. (AZ, USA) are offering several tests using this technology and is actively marketing it as a multiplex pathology method. Metal nanoparticles such as gold [26] or silver [27] nanoparticles also have great potential in single-molecule detection and molecular testing. Unlike Qdots, molecules are detected by fluorophores that are covalent-linked to the metal particles. There are several unique advantages in using the fluorophore-linked nanoparticles for molecular testing. First, it allows the detection of a single molecule of either nucleotide or protein. Figure 2 depicts detection of an individual oligonucleotide by silver nanoparticles attached with fluorophore-labeled gene-specific oligonucleotide. To prepare for the gene-specific binding nanoparticles, gene-specific oligonucleotides are covalently linked to a fluorophore dye and are then coated onto the surface of the nanoparticles. The diameters of the nanoparticles could range on average from 1.5–200 nm. Presence of a molecule representing a pathogen or a gene of interest can be detected by complementary 92


hybridizations between the target molecule and the probe oligos. It has been shown that fluorescent signals of bound DNA duplexes between the nanoparticles are much higher than unbound free nanoparticles due to enhanced fluorescence. As an alternative design of gene-specific oligo probes to minimize detection background, molecular beacons, as described earlier for real-time PCR, has also been used for detecting single molecules [28]. With such a probe design, the fluorescence is normally highly quenched by the nanoparticle through a distance-dependent shielding. However, as much as thousands-folds increases of fluorescence could be observed when the gene-specific oligos hybridize to a complementary DNA, thus allowing detection of either a pathogen or a specific gene target of interest. By using this technology, single mismatch gene mutation could also be detected [28]. A second unique feature of using the nanoparticle detection method is the captured molecules can be enriched by magnetic collection of the metal nanoparticles. An example of such an assay is the bio-barcode amplification (BCA) [29]. Using a similar principle described in Figure 2, the gene-specific gold nanoparticles capture the gene of interest by sandwiching the DNA between two particles. The ‘particle–DNA–particle’ sandwich is then removed magnetically for final DNA recognition and quantification. Since each nanoparticle is coated with thousands of gene-specific oligonucleotides, along with barcode-like nucleotide sequences for recognition, it thus automatically amplifies each of the DNA molecules by at least thousand-folds. As the DNA molecules are further enriched by magnetic collection, the BCA offers extremely high detection sensitivity with its limit at approximately 500 zM target DNA; that is, less than 500 molecules per milliliter of solution [30]. BCA has been used to detect amyloid-β-derived diffusible ligands (ADDLs), a protein diagnostic marker for the onset of Alzheimer's disease [26]. Normally the ADDL concentration in brain or cerebrospinal fluid is too low to be detected (