A Thousand Points of Light

Not so long ago, researchers had somewhat limited choices for locating and following a particular piece of DNA. A probe could be labeled using radioactivity, by kinasing an end or nick-translating the whole piece. A fragment of interest could be visualized (along with all other DNA and RNA species in the preparation) using ethidium bromide. With sufficient skill and patience an investigator could obtain from these rather crude techniques fairly impressive information, such as the precise 5' end of an RNA transcript or the presence of a previously characterized mutation. Today, however, the questions are more demanding. In addition to determining whether a particular piece of DNA is present, researchers usually want to know the precise sequence of the target and how many copies are present. Frequently this precise identity and quantity question is asked of an enormous number of sequences at once in an array format, or "on the fly" during an analytical reaction, or in a location-specific fashion in situ. This new sophistication has driven--or has been driven by--a diverse collection of fluorescent labels and new techniques for exploiting them. There are plenty of systems on the market, such as vectors explicitly designed to generate probes, that represent improvements to conventional approaches. This article focuses on some of the innovative and novel strategies that are gaining acceptance. Before getting into specific assays, however, it's worth taking a closer look at the physics of fluorescence and some related phenomena such as resonance energy transfer and quenching. Some horrible equations derived from quantum theory describe how bonding electrons in complex molecules are constrained to a variety of allowed electronic, vibrational, and rotational states, each of which has a very specific associated energy. When a molecule is exposed to light, electrons or bonds can shift between different electronic, vibrational, or rotational states--provided the energy of the incoming photon matches exactly the energy difference between the two states. This process is called excitation. From this excited state, several things can happen. In most molecules a return to the ground state happens via vibrational or rotational processes on the order of 10-12 seconds; however, in some molecules this vibrational or rotational "damping" is prohibited by quantum rules and the decay to the ground state is slower and involves the emission of a new photon. In most biological systems, fluorescence excitation occurs in some region of the ultraviolet-visible spectrum (uv is185-400 nm, visible is 400-700 nm, and infrared is 700-15,000 nm). Excitations in the uv-vis region usually involve electronic, rather than vibrational or rotational, transitions, and as these excited states tend to be unaffected by rotational or vibrational "damping," the return to the ground state is often accompanied by photon fluorescent emission. In practice a large number of closely related transitions results from illumination with a continuum of photon energies. This results in both the excitation and emission profiles of a molecule following roughly Gaussian profiles around defined maxima. Most instrumentation takes advantage of this broader swath of excitation and emission to obtain greater numbers of excitation and emission photons. Signal detection is usually improved by positioning the detector system at right angles to the excitation system; this minimizes the effects of Rayleigh--and Raman--scattering. This is easy to accomplish in a spectrometer, but it is harder to engineer in a system that examines a flat surface such as a microarray or in situ experiment. In these situations, where confocal excitation and detection is often used, filtering becomes a significant issue, and the use of filters to reduce the bandwidth of the excitation light and to reject that light in the emission path is critical. The excitation source is also an important consideration: Insufficient intensity will mean too few fluorophores will be excited, leading to low signal-to-noise and a requirement for long averaging times; too much intensity will mean the sample may be photobleached, leading to far-from-linear responses between the amount of target and the measured signal. Pass the Parcel (Photon Style) Many of the new DNA tags and labels depend on two phenomena that are extensions of fluorescence: quenching and energy transfer. In general, anything that reduces the lifetime of the excited state decreases the quantum yield of the fluorophore; anything that decreases the quantum yield is called quenching. There are three main mechanisms for determining these phenomena: collisional, in which the excited state of the fluorophore loses its energy by bumping into a nonfluorescent molecule; static, in which the excited state reacts with the quencher, forming a nonfluorescent complex; and energy transfer, which involves the nonradiative transfer of energy from a donor to an acceptor. For fluorescence resonant energy transfer (FRET) to occur, there must be a precise overlap in quantum energy levels between the donor and the acceptor, the energy being transferred by dipolar coupling rather than emission and reabsorption of a photon. FRET has been used very productively to create dyes for DNA sequencing, where a common donor eliminates the need for multiple excitation wavelengths but instead transfers its energy to four separate dyes that have easily discernable emission spectra.1 FRET and fluorescence quenching are very distance dependant, allowing their exploitation in several novel assays that alter donor-acceptor geometries. Many of the methods described in this article depend on a variety of modified oligonucleotides. Many fluorescent dyes are available as dye-phosphoramidites (or as dye-CPG derivatives), which are compatible with automated oligonucleotide synthesis methods. Using this approach, dyes can be incorporated at the 5' or 3' end or at any internal position during routine synthesis. Similarly, amino-modified bases can be incorporated into an oligo at any position, enabling a wider variety of labeling, because many additional dyes are available in an NHS-ester form that can be conjugated to an amino-modified oligonucleotide after synthesis. Different applications call for different modifications, including such esoterica as variable-length spacers, universal bases, and branched backbones. An excellent online description of many variants and their synthesis can be found at www.interactiva.de/knowledge/nucleicchem/ modifiedoligos.html. It used to be standard procedure to purify oligonucleotides, generally using a trityl-on synthesis and an oligonucleotide purification cartridge. Now, many techniques work well with nonpurified materials; desalting is all that is needed for routine PCR. However, labeled oligos are a different story. Reverse-phase HPLC (RP-HPLC) purification is generally less expensive than Polyacrylamide gel electrophoresis, and is often sufficient, especially for shorter oligos. Some dyes are incorporated from phosphoramidites more efficiently than others. Fam and Tet, for example, are readily incorporated and simple desalting is often adequate; Cy3 and Cy5, by contrast, are less efficiently incorporated using many protocols, and HPLC purification to eliminate unlabeled oligos from the preparation will usually improve the performance of these materials. (Oligos incorporating dyes postsynthesis often require two rounds of HPLC purification, so lower yields should be expected when ordering these materials.) Genisphere's 3DNA Submicro Detection System A goal of many modern detection systems is to monitor the concentration of specific nucleotide sequences without having to isolate the probe/target hybrid. To achieve this, the probe must produce a detectable signal only in the presence of the target sequence. One approach is to engineer the signal mechanism into oligonucleotides and one of the most elegant systems for this is the molecular beacon.2 The molecular beacon is a hairpin-shaped oligo with a loop sequence complementary to part of the target sequence and flanked by two arms that anneal to form a short (5-7 base pair) stem. At the end of one arm is a fluorophore and at the other a quencher that prevents fluorescence when the stem is intact. However, with careful consideration given to the relative stability of the stem versus that of the beacon-target hybrid, the oligo is designed to remain folded in free solution but to readily hybridize to any available target; once hybridized, the quencher is moved away from the fluorophore, which then fluoresces to signal that target is present. Molecular beacons thus can be used to monitor real-time PCR by using a target sequence in the middle of the amplicon and measuring fluorescence during the annealing step of PCR. Molecular beacons are now available from several sources. For example, Stratagene of La Jolla, Calif., offers molecular beacons for real-time monitoring of PCR reactions. Among other criteria, molecular beacons must be designed to dissociate from the target at PCR extension temperatures so that they won't interfere with polymerization. The Amplifluor(tm) system from Intergen Co. of Purchase, N.Y. is an approach distinctly different from molecular beacons that directly labels the product of a PCR reaction (amplicon), avoiding the problems of dissociation encountered with molecular beacons. Furthermore, the fluorescent labeling of the amplicon itself provides a more suitable analytical result, avoiding false negative data if the dissociation fails. Thus, the PCR products are directly detected by incorporation of a hairpin oligonucleotide into the PCR reaction product. When incorporation occurs, the fluorophore label is separated from a quencher molecule, and intense fluorescence directly proportional to the amplicon created is observed. The Amplifluor system allows researchers to design hairpin primers for both direct target detection and universal application to a variety of targets presented in a variety of applications. In addition to real-time PCR quantitation, molecular beacons and similar systems can also be used to detect RNA in vivo, to detect viruses3 and other pathogens, and for genotyping.4 The latter can be achieved by using multiple fluorescent oligos with different colors for each allele. When fully optimized, molecular beacons and their relatives make for efficient detection systems, but occasionally some pitfalls are encountered. False positives or low signal-to-background can result from impure preparations that contain free fluorophores or from oligos with a fluorophore but no quencher, or from design problems such as a stem that is too strong at low temperatures. Care must be taken with design as well as with the necessary control experiments to ensure that molecular beacons operate as intended,5 and many suppliers offer services that include helping with oligonucleotide design. Companies that Sell Fluorescently Tagged Probes Alpha DNA (800) 511-3654 www.alphadna.com Amersham Pharmacia Biotech Inc. (800) 526-3593 www.apbiotech.com Annovis Inc. (610) 579-1200 www.annovis.com Applied Biosystems (800) 345-5224 www.appliedbiosystems.com Biosearch Technologies Inc. (800) GENOME1 www.biosearchtech.com Caltag Laboratories Inc. (800) 874-4007 www.caltag.com CyberSyn Inc. (800) DNA-FAST www.cybersyn.com emp Biotech GmbH (49) 30-9489-2201 www.empbiotech.com EY Laboratories Inc. (800) 821-0044 www.eylabs.com Gemini Biotech (800) 561-3454 www.geminibio.com Genisphere (877) 888-3DNA www.genisphere.com Genpak Inc. (800) 385-9248 www.genpakdna.com Genset Oligos (800) 995-0308 www.gensetoligos.com Glen Research (703) 437-6191 www.glenres.com ID Labs Inc. (800) 463-4782 www.idlabs.com IT BioChem Integrated DNA Technologies Inc. (800) 328-2661 www.idtdna.com INTERACTIVA (49) 731-93579-290 www.interactiva.de Intergen Co. (800) 431-4505 www.intergenco.com Life Technologies Inc. (800) 338-5772 www.lifetech.com Lofstrand Labs Ltd. (800) 541-0362 www.lofstrand.com Macromolecular Resources (800) 491-0424 mmr.bmb.colostate.edu Midland Certified Reagent Co. (800) 247-8766 www.mcrc.com Molecula Research Laboratories (800) 381-0856 www.molecula.com MWG-Biotech (877) 694-2832 www.mwgbiotech.com Operon Technologies Inc. (800) 668-2248 www.operon.com Paragon BioServices Inc. (800) 545-6569 www.paragonbioservices.com Research Genetics Inc. (800) 533-4363 www.resgen.com Retrogen Inc. (800) 738-7643 www.retrogen.com Roche Molecular Biochemicals (800) 428-5433 http://biochem.roche.com Sequitur Inc. (508) 650-1459 www.sequiturinc.com SeqWright Inc. (800) 720-4363 www.seqwright.com Stratagene (800) 424-5444 www.stratagene.com Synthetic Genetics (800) 562-5544 www.syntheticgenetics.com Tetra Link InternationaI Inc. (800) 747-5170 www.tetra-link.com TriLink BioTechnologies Inc. (800) 546-0004 www.trilinkbiotech.com Vector Laboratories Inc. (800) 227-6666 www.vectorlabs.com Vysis Inc. (800) 553-7042 www.vysis.com WebOligos.com (888) 4-OLIGOS www.weboligos.com Outta' the Way A cousin of the molecular beacon is the TaqMan probe from Applied Biosystems of Foster City, Calif. This system exploits the 5' exonuclease activity of Taq DNA polymerase. During the PCR extension an annealed oligonucleotide that has a reporter fluorophore at the 5' exonuclease and a quencher at the 3' exonuclease is chewed up by a polymerase 5'-3' exonuclease activity, releasing the fluorophore from its quencher (the presence of the TaqMan probe doesn't significantly inhibit PCR product synthesis). The resulting fluorescence is proportional to the amount of PCR product. In addition to monitoring PCR product concentration, TaqMan has been applied to assays for detecting pathogenic Escherichia Coli strains in food,6 detection of nucleotide polymorphisms,7 and as a reporter system for transgenic screening. An alternative to the fluorophore-quencher system is a dual fluorophore approach that exploits FRET. This is the principle behind the LightCycler hybridization probes from Roche Molecular Biosystems of Indianapolis. Two oligo probes, rather than TaqMan's one, anneal to the amplicon; one carries a fluorescein label (the FRET donor) at its 3' end and the second is labeled with LC red 640 (the FRET acceptor) at its 5' end. The oligos are designed to hybridize in a head-to-tail orientation with the fluorophores separated at a distance that is compatible with efficient energy transfer. Simple and rapid detection of a target, however, is often only part of a research project. What if you need to localize mutations at the cellular level? In situ studies, especially in vivo, are inherently more challenging. An example is the localization of tumor cells with specific, perhaps single nucleotide, mutations. One system that can be applied at the cellular level is the padlock probe: an oligonucleotide with target complementarity at both ends. Upon hybridization to the target sequence, the 5' and 3' ends of the probe become a substrate for DNA ligase, which seals the nick and leaves the resultant single-stranded circular DNA catenated to the target. Though the catenated probes can be detected directly, a cunning twist involves using this circular molecule as a substrate for rolling circle replication, leading to 103-fold amplification. Other strategies also leave this replication product covalently or topologically attached to the target. Padlock probes have a wide array of possible applications; their adaptation to fluorescence in situ hybridization (FISH) probably being the most valuable.8 One potential problem with padlock probes is that they may slip along the DNA, away from the target. This is unlikely to be a problem at the chromosomal scale, but it could result in the probe's falling off the end of a short linear target. Development of the smaller "earring probe," in which the circular oligo is more tightly entwined with the target DNA sequence, eliminates any possibility of slippage; molecular earrings may also find applications in DNA-based diagnostics and other areas.9 Although they provide many powerful analytical tools, dual-labeled oligos have their problems. Several groups are working on strategies that also will allow real-time fluorescence monitoring of PCR amplifications without the need for reagents that are difficult or expensive to prepare or purify. One promising approach (although not yet commercially available) is a novel fluorogenic technology from Life Technologies of Rockville, Md. An oligonucleotide primer is labeled with a unique fluorophore so that it is relatively nonfluorescent as unincorporated primer but exhibits a significant increase in fluorescence when hybridized to target or incorporated into PCR product. Accumulation of amplification product during PCR therefore leads to a proportional increase in fluorescence.10 The quest always seems to be one of "more for less": More signal from less material. Most approaches run into signal-to-noise limitations, necessitating long signal averaging times, but several new methods are incorporating different levels of amplification as they move toward reliable single-molecule detection. One commercial application is the 3DNA Submicro detection system from Genisphere of Montvale, N.J., intended for labeling material used to probe expression arrays. A cDNA is generated using oligos that have an extension (the "capture sequence") that is complementary to the 3DNA molecule. After the cDNA is generated, it is hybridized to the 3DNA molecule: a 3-dimensional nucleic acid matrix--often called a dendrimer--that contains an average of 250 fluorescent dye molecules (Cy3 or Cy5). The 3DNA/cDNA complex is then hybridized to the array. As no dyes are directly incorporated into the cDNA, synthesis and hybridization is very efficient, and as the labeling is targeted against the capture sequence, it is independent of the cDNA base composition. Researchers can use less RNA for probe synthesis and can obtain much higher signal-to-noise using 3DNA dendrimer technology.11 Further, because individual dendrimers can be detected, appropriate calibration should allow determination of the actual numbers of captured molecules. Another route to increasing sensitivity involves labeling with expressible DNA fragments.12,13 The technology is quite complex: A capture probe that has sequence complementarity to the target molecule is immobilized in a microtiter well using digoxigenin/antidigoxygenin or biotin/streptavidin. The target DNA is simultaneously hybridized to this capture probe and to the expression probe. The expression probe not only has sequence complementarity to the target sequence, but it contains a T7 promoter driving a detection molecule. Using T7-driven in vitro transcription and translation of firefly luciferase, luciferin luminescence detected 20.5 atto mole target DNA and was linearly related to target DNA in the 5-5,000 atto mole range. Using Ca2+ triggered bioluminescence of aequorin, synthesized in a similar manner, 25 amol target could be detected. Brightest Bang for the Buck People are often curious about the "brightness" of a fluorescent dye, which depends on many parameters. The parameters can be divided between the physical and chemical properties of the dyes and the excitation system. The important physical properties of the dyes are quantum yield and extinction coefficient. The quantum yield is an expression of the number of photons emitted divided by the number of photons absorbed. A quantum yield of 0 indicates a nonfluorescent molecule, and a quantum yield of 1 indicates that 100 percent of the excitation photons result in lower-wavelength emitted photons. The extinction coefficient is an expression of the probability that a photon of a given wavelength will be absorbed by the fluorophore. A high extinction coefficient combined with a high quantum yield generally leads to a "bright" fluorophore; fluorescein, for example, is a relatively "bright" dye, having an extinction coefficient of about 80,000 at its absorption maximum and a quantum yield of ~0.9. For stable dyes, a more intense photon source will yield more fluorescence photons. However, the intrinsic chemical properties of most dyes prevent infinite absorption events due to a decomposition mechanism called photo-bleaching. In general, rhodamine dyes are resistant to photo-bleaching, fluorescein dyes have intermediate stability to photo-bleaching, and cyanine dyes are prone to photo-bleaching. Dye brightness is dependent on the excitation source. If an intense laser is used, the photons are generally restricted to one or a few wavelengths, which may or may not be matched with the absorption maximum of the fluorescent dye. For example, with the use of a 488 nm argon ion laser, a dye with an absorption maximum of 490 nm will look very much brighter than a dye with an absorption maximum of 510 nm, even though their extinction coefficients and quantum yields may be identical. This is because the extinction coefficient of the longer wavelength dye is much lower at 488 nm than at 510 nm. The fluorescence quantum yield (and therefore overall fluorescent brightness) is often strongly dependent on environmental factors such as pH, solvent polarity, and intermolecular interactions with nucleic acids, proteins, and other dye molecules. Extinction coefficients show much less variability in these respects. Most providers can incorporate a variety of fluorophores into an oligo, although many companies do have preferences. Fluorescein derivatives that are available as amidites, Applied Biosystems' rhodamine TAMRA, Amersham Pharmacia Biotech's Cy3 and Cy5 dyes, and Molecular Probes' Alexa Fluor dyes are popular, often combined with DABCYL when a quencher also is needed. A table listing excitation and emission maxima as well as extinction coefficients for many commonly used fluorophores can be found on Integrated DNA Technologies' Web site at www.idtdna.com/technotes_faqs/Fluorescent_Dye_ Labeled_Oligonucleotides.html. Perhaps the next generation of labels that researchers will use in these diverse methods for labeling DNA will be luminescent quantum dots. Described by their creators as "tiny light bulbs" (about a billionth the size of a conventional light bulb) zinc sulfide-capped cadmium selenide quantum dots emit about 20 times as much light and are about 100 times as resistant to photobleaching as most currently used labels.14 Quantum dot properties emerge from the peculiar properties of small (1-5 nm) clusters of metal or semiconductor molecules; one fascinating aspect of this is that the emission can be tuned from blue to red simply by changing the particle size. The emission bandwidth is also narrow compared to most fluorophores, and, interestingly from a molecular biologist's perspective, they can be excited at 514 nm using an Argon-ion laser employed by many current instruments. Chan and Nie's preparations are compatible with aqueous solutions and are conveniently carboxyl-terminated, in theory allowing their attachment to a wide range of substrates, including support matrices, proteins, and nucleic acids. Bob Sinclair (bobsinclair@tech-write-edit.com) is a freelance writer in Salt Lake City. Paul G. Wolf (wolf@biology.usu.edu) is an associate professor of biology at Utah State University References 1. J. Ju et al., "Fluorescent energy transfer dye-labeled primers for DNA sequence analysis," Proceedings of the National Academy of Sciences, 92:4347-51, 1995. 2. S. Tyagi, F.R. Kramer, "Molecular beacons: probes that fluoresce upon hybridization," Nature Biotechnology, 14:303-8, 1996. 3. A.J.-C. Eun, S.-M. Wong, "Molecular beacons: A new approach to plant virus detection," Phytopathology, 90:269-75, March 2000. 4. L.G. Kostrikis et al., "Spectral genotyping of human alleles," Science, 279:1228-19, 1998. 5. G. Bonnet et al., "Thermodynamic basis of the enhanced specificity of structured DNA probes," Proceedings of the National Academy of Sciences, 96:6171-6, 1999. 6. R.D. Oberst et al., "PCR-based DNA amplification and presumptive detection of Escherichia coli O157:H7 with an internal fluorogenic probe and the 5' nuclease (TaqMan) assay," Applied and Environmental Microbiology, 64:3389-96, 1998. 7. I. Tapp et al., "Homogenous scoring of single-nucleotide polymorphisms: comparison of the 5'-nuclease TaqMan assay and Molecular Beacon probes," Biotechniques, 28:732-8, April 2000. 8. M. Nilsson et al., "Padlock probes reveal the in situ distribution of alpha-satellite sequences on chromosomes 13 and 21, differing in a single nucleotide position," Nature Genetics, 16:252-5, 1997. 9. H. Kuhn et al., "An earring for the double helix: assembly of topological links comprising duplex DNA and a circular oligonucleotide," Journal of Biomolecular Structure and Dynamics, 11:221-6, May 2000. 10. D. Schuster, "Novel fluorescent oligonucleotides for homogenous detection and quantitation of nucleic acids," Abstracts from the Cambridge Healthcare Institute's fifth annual conference on Gene Quantification, online at www.healthtech.com/conference/ 00qga/index.html, 2000. 11. R.L. Stears et al., "A novel, sensitive detection system for high- density microarrays using dendrimer technology," Physiological Genomics, 3:93-9, August 2000. 12. N.M. Chiu, T.K. Christopoulos, "Hybridization assays using an expressible DNA fragment encoding firefly luciferase as a label," Analytical Chemistry, 68:2304-8, 1996. 13. S.R. White, T.K. Christopoulos, "Signal amplification system for DNA hybridization assays based on in vitro expression of a DNA label encoding

A Thousand Points of Light

Interested in reading more?

Become a Member of