In the mid-1990s, toxicologist Raymond Biagini was looking for a faster way to evaluate pest-control workers for exposure to pesticides. The existing process was slow and labor-intensive: Urine samples were hydrolyzed overnight in acid, extracted, cleaned, and then analyzed by gas chromatography or high-performance liquid chromatography. "And then at the end of all that, you have one number that took you a day and a half to get, minimum," says Biagini, of the Biomonitoring and Health Assessment Branch of the Centers for Disease Control and Prevention, and the National Institute of Occupational Safety and Health in Cincinnati, Ohio. Compounding this problem, most pesticide workers are exposed to four or five pesticides over the course of a week.
Biagini and his teammates found a much faster solution in a new bead-based multiplexing system from Luminex of Austin, Texas. Using raw urine and specially developed beads designed in his lab, four pesticides can be measured simultaneously. "And, bingo, you have four numbers that would take you at least three days to get using classical methods," says Biagini, who has since developed assays for multiplexed detection of exposure to toxic mold, vaccines, and bioterrorism agents.
To many researchers, multiplexing – the ability to monitor multiple assays on one platform in a single run – is synonymous with the bead-based flow cytometry assays epitomized by Luminex's xMAP technology and BD Biosciences' Cytometric Bead Array (CBA). Such assays do dominate the market, but multiplexing is more than just beads. Companies have devised a variety of alternative technologies to help researchers with their higher-throughput protein analysis, genotyping, and gene-expression profiling needs. We review some of these competitors, as well as the leading bead-based products, below.
Similar in format to ELISAs, both the xMAP and CBA systems employ microspheres labeled with different concentrations of fluorescent dyes that serve as a sort of assay fingerprint. The spheres are coated with a probe to a specific analyte, and targets are labeled with a secondary tag for detection. The spheres are then suspended and run through a flow instrument, which detects both the fluorescent bar code on each bead (to identify the assay) and the secondary tag (to quantify each analyte).
In the case of Luminex's technology, as many as 100 color combinations can be associated with the beads, permitting up to 100 separate reactions to be monitored simultaneously, at least for genotyping applications. Company spokesperson Grant Gibson notes that antibody-based assays are limited by cross-reactivity, so most multiplex immunoassays carry less than 30 or 40 simultaneous reactions.
Everything needed for the CBA platform can be purchased directly from BD, which offers a variety of assays, including chemokine and cytokine detection kits that work with any flow cytometer. Luminex, on the other hand, sells uncoupled beads and instrumentation to end users who develop their own assays, or the company licenses its technology to a variety of "partner" vendors who develop custom xMAP assays and, in some cases, new instrumentation and software. Commercially available assays include detection kits for cytokines, phosphoproteins, and endocrine and pituitary hormones, to name a few.
"There are many different technologies that are published and described, but not all of them are robust enough to be successful in the long term," says Dario Vignali, an immunologist at St. Jude Children's Research Hospital, Memphis, Tenn. "The rigidity of the xMAP assay system has been borne out by the fact that many different companies have taken this platform on board, developed it, and now sell a variety of assays that are based around this technology."
Vignali adds that the technology's speed and ease of use makes it superior to standard plate-based ELISAs, which can probe only one analyte at a time. It's also more sensitive: Conventional ELISAs can measure cytokines down to 50 pg/ml, whereas Vignali has reported sensitivity levels of 10 pg/ml for a multiplexed cytokine bead assay.1
The costs for the multiplexed Luminex bead kit vary by company. Qiagen's LiquiChip Human Cytokine Core Kit, for example, lists for $229 (US) for a 96-well assay and can be combined with a LiquiChip Human Detection Kit ($169 per 96-well) to measure up to 15 cytokines simultaneously. Biosource International's Human Cytokine 10-Plex kit has a list price of $1,449 per 100 reactions, whereas Biosource's single-plex CD14 direct-detection ELISA kit costs $465 (US) for 96 reactions. A BD CBA Human Allergy Mediator Kit, which can measure six cytokines, costs $1,350 for 80 tests.
Luminex beads can also be used for genotyping applications. John Landers, chief scientific officer of Worcester, Mass.-based PolyGenyx, which specializes in genotyping and haplotyping, says that the cost of a Luminex-style detection system (less than $60,000, makes it an attractive alternative to some competing products (such as Illumina's BeadArray system, which costs about $250,000). But there are some caveats to the Luminex technology: It takes some "art" to master, says Biagini, and Landers adds that users who develop their own assays will have limited technical support.
An alternative to flow cytometry-based assays for genomics applications is Hayward, Calif.-based Quantum Dot's recently launched Mosaic Gene Expression Assay System, which currently can probe the expression of 100 user-defined genes simultaneously. Mosaic is similar to xMAP in that it employs color-coded beads, but the beads are coded with quantum dots rather than dyes and are imaged in plates rather than passed through a flow instrument.
According to company spokesperson Stephen Chamberlain, the Mosaic assay offers sensitivity comparable to quantitative PCR but requires 10 times less sample per gene. Mosaic's cost per sample or assay, including the price of sample prep, currently ranges from $80 to $160 ($1 to $2 per gene, depending on the size of the project, Chamberlain says. The price does not take into account the cost of the plate reader itself.
San Diego-based Illumina's BeadArray technology offers an even higher level of multiplexing. Designed for SNP genotyping and RNA profiling, Illumina's technology consists of hexagonally packed bundles of optical fibers arranged onto a 96-bundle Array Matrix or slide-shaped BeadChip containing millions of wells. Glass microbeads derivatized with oligonucleotide probes self-assemble onto the array such that one bead fits into an etched well at the end of each fiber. The beads are assembled randomly into the wells and are located and identified via a decoding process based on sequential hybridization.
Illumina's BeadArray reader scans the chips with 0.8-micron resolution and can process 96 samples in less than 90 minutes. According to company representative Bill Craumer, Illumina's Sentrix Human Ref-8 BeadChip (24,000 transcripts per array) can query eight samples on a single microarray. Running costs per sample (array plus reagents) are less than $150.
The technology has found a natural home in the International HapMap project. Andrei Verner, manager of genotyping services at McGill University and Genome Quebec Innovation Center in Montreal, purchased an Illumina system two years ago because it was the only technology available at the time that could handle the high throughput needed for his group's role in the project. Verner's team has been able to analyze 600,000 genotypes per day with a higher level of accuracy than he had seen in previous experiments. Verner attributes this success rate to the level of redundancy built into each assay. " [Illumina's] array has about 50,000 wells for glass beads, but they use 1,500 probes for SNPs, which lets them have the same type of bead 30 times on the array. And if one or two of them didn't work well, you could still have a good feeling about your call," he says.
Illumina currently sells focused arrays, including human toxicology and human, mouse, and
Despite their dominance, beads aren't actually necessary for multiplexing. Gaithersburg, Md.-based Meso Scale Discovery, for example, offers a plate-based system based on electrochemiluminescence detection. Microtiter plates embedded with electrodes are spotted with four, seven, or 10 distinct probes per well. Probes that bind to analytes are labeled with a ruthenium bipyridyl tag, which emits light in the presence of a current supplied by the electrodes. Coupled with high-speed instruments, MSD's system offers a throughput of more than 15,000 individual analytical assays per minute. These assays typically have sub-picogram detection limits in plasma or serum and have linear ranges of at least three logs.
Company spokesperson Hans Biebuyck says the system provides high sensitivity because the electrochemiluminescent signal is relatively unaffected by changes in pH, external light, or sample complexity. Meso Scale's detection instruments range in price from $25,000 to $190,000. According to Biebuyck, per-well prices for assays start at 10 cents and move up to several dollars, depending on the type and number of assays in the well, the number of reagents included in the kit, and the volume of the order. (The per-plate list price for a 10-plex Th1/Th2 panel in a 96-well format is $1,440 for low-volume orders.)
Tucson, Ariz.-based High Throughput Genomics' ArrayPlate is another beadless multiplexing system. Each of the ArrayPlate's 96 wells contains 16 oligonucleotide probes for DNA, RNA, and protein targets. The probes are universal, says company founder Bruce Seligmann. Programming of the probes is achieved by using custom linkers that bind to the probes and the target of interest.
For gene-expression quantification, High Throughput Genomics offers the quantitative nuclease protection assay, a proprietary technology that requires no extraction, amplification, or biosynthesis steps. According to the company, this makes the system's cost per data point low ($0.45 to $1.15) compared to competing assays.
Selected Multiplex Companies
Kyle Chan, vice president for business development at San Diego-based Celgene, uses the arrays for medium-throughput target validation. "What I really like about it is that the sample prep is simple, it's one step. You don't have any RNA isolation. You just grow your system in a 96-well format and pipette it in this onestep solution and then go from there," he says.