Editorial Article
Monday April 19, 2010
by Josh P. Roberts
Since the immunoassay was first described more than half a century ago, it has become a mainstay in clinical and research labs alike. Innovations in protocols, instrumentation, reagents and software continually make detecting specific analytes in biological samples easier, faster, and more reliable. Tedious separations and scintillation counts have for the most part given way to same-well (tube) preparations and automated light readings. Yet at its core the immunoassay remains fundamentally a technique in which antibodies are adhered to a target of interest, leading to a signal roughly proportional to the amount of bound ligand in the sample.
Historically speaking, immunohistochemistry
is still the cornerstone of the immunoassay in cancer diagnostics, notes Matthew Landry, VP of sales and marketing for GenWay Biotech: take a biopsy, stain it with a labeled antibody to a tumor marker, and examine the slide under a microscope. As for liquid samples, it's not necessary to be a trained pathologist to interpret an ELISA, and that helped to fuel the rise of plate-based assays and their kin—what most of us think of when we think “immunoassay.”
Getting started with immunoassays doesn't have to be a daunting task. There are kits available to detect many common analytes, and labeled—and unlabeled— antibodies for others, pre-made buffers, sticky plates, ways to amplify faint signals, ways to multiplex, and computer algorithms to interpret even the most convoluted data.
Enzyme Linked?
ELISA stands for Enzyme Linked Immunosorbant Assay. Traditionally, this meant that the signal was generated when an enzyme (such as horseradish peroxidase or alkaline phosphatase) converts a substrate to a colored product. The enzyme could be conjugated to an antibody that directly recognized the analyte, or attached to an antibody that recognizes another antibody. The assay—often in a 96- or 384-well plate—is then fed into a reader that shines light through the plate and analyzes the absorbance at a particular wavelength.
Yet ELISA these days is generally used as a catch-all phrase for any plate-based immunoassay, says Stephen Shiflett, technical product manager for Thermo Scientific Pierce Protein Research. This could mean an enzyme-, chemiluminescence-, or even fluorescence-based assay.
Which system to use will depend, in part, on what the researcher is comfortable with and what equipment is available, says BD Biosciences, Pharmingen Product Manager Trent Colville. A typical newer plate-based reader generally has at least basic fluorescence capabilities, Shiflett notes. A chemiluminescence reader is a more specialized instrument.
A kissing-cousin of the ELISA is the bead assay. Instead of coating a plate with antibodies to capture an analyte, antibodies can be linked to microscopic beads. The beads are then analyzed by flow cytometry, but the assay is otherwise essentially a fluorescence-based ELISA with multiplex capabilities. Several companies offer such systems, including Luminex—which requires a dedicated instrument—and BD, whose proprietary beads can be run on any flow cytometer. The downside is sensitivity: “Currently most bead assays can reliably detect down to about 5-10 pg/ml, [while] a really good ELISA can get you down to 1, maybe 0.5 pg/ml,” explains Colville—something that BD is “trying to address.”
Signal:Noise
Sensitivity is based on a variety of factors.
Although it involves some extra steps, immunoassays frequently make use of both a primary antibody to recognize the antigen, as well as a secondary antibody that recognizes the primary antibody. This amplifies the signal by having multiple secondary antibodies per primary. Another advantage is the ready availability of relatively inexpensive, high-quality secondary antibodies conjugated to an enzyme or fluorophore, meaning that it can be bought off-the-shelf rather than needing to be conjugated (or purchased conjugated) for each analyte under study.
The other half of the signal:noise ratio is background.
In a sandwich ELISA, after a plate is coated with analyte-specific antibodies, the plate is incubated with a blocking solution to reduce non-specific binding. Many researchers use a skim milk-based buffer solution (known as BLOTTO), or one based on fish gelatin or bovine serum albumin. These can be made up in the lab or purchased as ready-made solutions from a variety of vendors. “But blocking buffers can have a tremendous effect on the degree of background in many cases,” Shiflett says. While BLOTTO and BSA are good starting points, optimized protein-based products (such as Thermo Scientific Superblock Buffer), as well as protein-free buffers, may yield cleaner results—especially if biotin is part of the assay. “You need to empirically determine which blocker is going to be the best for your particular system. There's really not any well-defined guideline or criteria,” he adds.
If a buffer comes with a kit, make sure to use it, emphasizes Irwin Libeskind, president of Cell Sciences, a supplier of immunoassay kits and reagents. Similarly, literature accompanying antibodies generally specifies optimized assay conditions, including buffers.
Sample Type
Antibody reagents are typically optimized and validated for the type of sample being assayed as well. “Some kits are made specifically for plasma, or for serum only, or plasma with citrate or EDTA,” Libeskind explains. “Make sure before purchasing that the kit is actually compatible with the type of sample.”
Certain kinds of samples are probably better assayed with a commercial kit, notes Jay Westcott, owner of ELISA Tech, a contract research organization specializing in immunoassays. “Those companies have spent a lot of effort trying to validate: you have precious samples, they require validation because there could be things in your fluid that interfere with the assays.”
Landry elaborates: proteins in solution react. They “might be degraded by proteases, or might be bound to another protein. And so these matrix effects make producing ELISAs difficult.”
Antibodies
Whether you can't find a kit that meets your needs, or you prefer the DIY (Do It Yourself) approach for financial or other reasons, it's crucial to obtain the right antibodies for the job. Obviously, luciferin isn't going to be cleaved by alkaline phosphatase, and a fluorophore isn't going to cleave anything, so it's important to have the right label for the right system.
An antibody designed for western blotting—raised against a peptide or denatured protein—isn't likely to recognize the native form found in serum or supernatant, points out Landry. And similarly, a heavily glycosylated urine or saliva protein may not be recognized by an antibody raised against a recombinant protein grown in E. coli.
Vendors will often sell “matched pairs” of antibodies—sets designed so that the capture and detection antibodies recognize different epitopes (regions) of the analyte and don't interfere with each other, and validated for a particular application.
Another consideration is how the antibodies themselves were purified. Affinity purification will remove “other contaminants that are in an antibody solution” that could interfere with the assay, explains Shiflett. Sometimes, though—especially with secondary antibodies—that may not be enough. He explains: “For example, if you're using an anti-rat and you wanted to make sure there wasn't any cross-reactivity to the other antibody that you're using in that same assay, you might want to select an antibody that's been cross-adsorbed against that other antibody.”
Standards
An oft-overlooked variable in any immunoassay is the standard. In an ELISA you generate a standard curve by assaying known quantities of a reference analyte, notes Westcott, “and what you're using as a standard is crucial in comparing it to your unknown.” Especially when the amounts are in the nanogram range, he says, what's sold as the same quantity by two different vendors can differ by as much as an order of magnitude. In that case, only relative quantification may be possible.
It's also important that the sample dilution is within the sensitivity range of the assay, points out Libeskind. “Kits are typically made against 'normal' ranges of analytes in the bodily fluid, but in a disease state the analyze may be highly elevated or quite a bit lower than expected.”
Its wider dynamic range may be one reason that “people are continuing to migrate to fluorescence,” says Shiflett. “You need to determine the dilution range or concentration of whatever it is you're looking for to know what volume to apply to the plate as a sample. Using something that has a working range of three or four orders of magnitude eliminates a lot of the headache associated with that.”