March 15, 2007 (Vol. 27, No. 6)
Focused Protein Profiling Strategies and Reverse-phase Arrays Drive Growth
Multiplexed protein measurements using protein microarrays have several promising applications in drug discovery and clinical diagnostics. These include delineating and interrogating signaling pathways and protein networks in health and disease states, enabling biomarker discovery and validation and use of biomarkers as surrogate outcomes in clinical studies, performing time-course studies to monitor fluctuating protein expression levels, and developing sensitive, robust, and reproducible diagnostic tests. The future lies in multiplexed protein microarrays and in producing a broader spectrum of high-quality antibodies.
Dominic Eisinger, Ph.D., president of Multiplex Biosciences (www.multiplexbiosciences.com), which merged with Rules-Based Medicine (RBM; www.rbmmaps.com) in October 2006, identified two critical challenges facing multi-analyte profiling at present: the first relates to assay development and the complexity of designing large protein panels and using antibodies that can work in concert and avoid cross-reactivity; the second is the need for high-quality antibodies optimized for microarray applications.
Despite having adequate miniaturization and speed to meet the demands of clinical diagnostic applications, protein microarrays have not yet made the leap to the lucrative diagnostics market, mainly because the results generated with this emerging technology “do not yet provide the appropriate therapeutic relevance,” in the view of Thomas Joos, Ph.D., head of the biochemistry department at the Natural and Medical Sciences Institute (NMI) at the University of Tubingen, Germany. In clinical diagnostics, “when dealing with individual patients, you need a clear yes or no answer on which to base therapeutic decisions.”
The hurdles the microarray industry must overcome are both scientific and economic. With increasing knowledge about the different types and states of disease, more personalized treatment approaches can be envisioned. “Protein microarrays have the potential to be the preferred platform for such assays when more than five parameters need to be measured simultaneously,” said Dr. Joos.
NMI utilizes a variety of protein array technologies, including interaction arrays and protein profiling arrays. Examples of the latter include µELISA in a microarray format and reverse-phase protein arrays, which are being applied for tumor profiling and biomarker identification.
Identifying several trends in the protein microarray arena, Dr. Joos stated that, “the industry has moved forward into focused protein-profiling approaches.” Rather than trying to fit the entire proteome on an array, companies are looking for highly relevant antibody populations against which to screen their targets. Dr. Joos does not envision that miniaturized, parallelized, label-free approaches now being introduced into the research setting will be able to achieve high enough sensitivity to be applicable in a diagnostic environment, where ultrasensitive detection will be the major demand. “However, there are dozens of research and monitoring applications for label-free detection systems,” he added.
“The whole field of protein microarray technology has to be driven today by applications, using the existing technologies,” Dr. Joos concluded. You can only apply robust, reliable technology capable of throughput that allows screening of large numbers of samples in an affordable manner. As an example, he points to RBM, which is routinely screening patient samples for more than 180 parameters from minimal amounts of sample.
Protein Profiling
In January, researchers gathered at CHI’s “Peptalk” meeting to present their views on the current status and future promise of protein microarrays in clinical diagnostics and drug discovery.
Dr. Eisinger described RBM as a multi-analyte profiling service based on the use of quantitative immunoassay panels that can analyze 190 human analytes in a 100-microliter sample of blood. RBM incorporates both a rodent multi-analyte profile (MAP™) and a human MAP that includes cancer antigens, cytokines and chemokines, hormones, and growth factors, for example, which can serve as either direct biomarkers or surrogates for drug safety and efficacy, disease, inflammatory changes, acute phase reactions, or other indicators of pathology.
“Our plan is to build the human panel to 500 analytes over the next 18 months,” said Dr. Eisinger, “and out of that we will develop focused panels. We are not trying to make a 500-plex. Our MAPs are broken into sub-multiplexes of five to 40 analytes, based on how the antibodies work together and the respective concentrations of the analytes in serum.”
Citing an example of how one can use a multiplexed microarray to identify surrogate markers of acute coronary syndrome (ACS), Dr. Eisinger presented data that clearly differentiated age-matched controls from those subsequently diagnosed with ACS based on levels of eight other biomarkers performed on blood drawn on admission to the emergency room.
The study revealed that two proteins, MMP-3 and SGOT (measured by mass rather than enzymatic activity), were diagnostically more accurate in diagnosing a heart attack on admission to the emergency room than CK-MB, troponin, or the two combined. Since MMP-3 is positive in unstable angina as well as heart attack, it is unlikely that it is a marker of unstable plaque in these patients. The company has filed an IP claim as the basis of further exploring the diagnostic potential of these markers in acute coronary syndrome.
In the fabrication of microarrays, consistency of printing or spot deposition is a critical parameter. John Austin, Ph.D., president of Aushon BioSystems(www.aushon.com), described the technology his company has incorporated into its new solid pin printing system.
A key challenge in producing highly consistent arrays is the diverse characteristics of the samples, whether antibodies, proteins, or cell lysates, and the range of buffer types used. “The associated range of fluid viscosities is incompatible with most existing array instruments, many of which were designed for genomic applications,” said Dr. Austin.
Further complicating the process, companies generating protein microarrays may utilize a variety of surfaces and coatings on which to print. Yet another challenge, when working with clinical samples, such as specimens obtained from laser capture microdissection, is the ability to accommodate low sample volumes. Dr. Austin described how Aushon has developed a protein-array printing system that can accommodate a range of sample materials and surface types and provides uniform spot deposition and high-throughput.
Arraying Cell Lysates
“Reverse-phase screening is a hot topic,” said Dr. Joos. Reverse-phase lysate arrays, in which an entire cell lysate is applied to an individual spot on a microarray, can then be probed with specific antibodies to detect a protein of interest. Each slide can accommodate hundreds of samples. Dilution series of each lysate are printed on the array, and the colorimetric readout yields a measure of how much protein is present.
Reverse-phase arrays are being used for quantitative analysis of protein expression in tumor cells; detection of up- or down-regulation of specific proteins may be able to guide therapeutic decision-making.
Paul Kroeger, Ph.D., senior group leader, gene expression analysis, global pharmaceutical R&D at Abbott (www.abbott.com), described reverse-phase lysate protein arrays as being a more sensitive, quantitative, and higher throughput alternative to Western blotting. Abbott is using reverse-phase arrays to explore signaling pathways, assess drug target and biomarker expression, and understand a drug candidate’s mechanism of action. Dr. Kroeger presented examples of screening campaigns. The first used a known target protein in a defined phosphorylation cascade and looked for downstream markers.
“The dilution curves give increased quantitation over Western blots, and we were able to calculate EC50’s for all the compounds screened,” said Dr. Kroeger. “We were able to identify a downstream marker to use as a surrogate for inhibition of the target, calculate the effect of each compound on the pathway, rank-order the compounds, and identify which were the most potent.”
Another application of reverse-phase lysate arrays included a series of cell lines with varying degrees of resistance or sensitivity to a compound. By identifying proteins that correlate with resistance, the researchers hope to be able to use the microarray technology to understand what is causing the resistance.
The third example focused on pathway analysis, determining changes in protein expression in cells in response to treatment with different compounds. The ability to trace, for example, the phosphorylation cascade down one branch of a pathway versus another can guide researchers in developing compounds synergistic with the desired effects.
Dr. Kroeger pointed to some current limitations in leveraging the full value of this technology, including the need for more validated antibodies—human as well as rodent for studying animal models of disease—and improvements in spotting throughput in array production to increase screening sensitivity.
Whatman’s (www.whatman.com) FAST® Slide Microarray Platform is based on nitrocellulose-coated glass microscope slides compatible with standard slide-based processing and detection instrumentation. Whatman has taken three different approaches to designing a quantitative array: FAST Quant™ for performing sandwich immunoassays, reverse-phase protein microarrays, and Combichip™ antigen arrays. All enable multiplexed analysis using fluorescence-based detection and are also compatible with colorimetric or chemiluminescence-based protocols.
FAST Slides have a microporous surface and “a large surface binding capacity,” said Michael Harvey, Ph.D., director of development, microarrays, and molecular biology R&D at Whatman. “The capture ability of the surface is particularly important for reverse-phase arrays,” he added, as the aim is to bind all of the proteins present in a cell lysate to an individual spot on the array. Whatman’s multipad FAST Slide configuration allows researchers to probe the same lysate with multiple antibodies.
Dr. Harvey also described the company’s CombiChip AutoAntigen Array, which can be used to detect antibodies in patient samples that react with self antigens, as a means of diagnosing autoimmune disease. Whatman introduced the array in Europe in September.
Label-free Detection
In February, Quadraspec (www.quadraspec.com) delivered the first shipment of its Spinning Disc Interferometry™ (SDI™) label-free immunoassay system. The underlying technology, which came out of Purdue University in 2004, will first be commercially available to a large veterinary reference laboratory as a diagnostic test for canine heartworm. Using the SDI system, the lab will be able to run 260 samples in less than an hour, the company reports.
In parallel development at Quadraspec is the Assay Development Kit, designed for the research and drug development market. It contains a disposable SDI disc on which a researcher can deposit an array of antibodies—onto an existing carpet of proteins—and screen those against an antigen of interest. These discs can replace existing ELISA tests or be the basis for developing multiplexed assays.
The SDI label-free technology is based on the phenomenon of interferometry, in which light shines down on the discs’ reflective surface. The thin surface layer is composed of an upper surface and a lower surface, and the light waves that reflect off of these interact, either attenuating or extinguishing each other. At the outset, the light source and detector are positioned to achieve maximal interferometric light intensity. Placing something on the surface, such as bound antibodies, “detunes” the interferometric intensity, and, as a result, the intensity of the reflected light will decrease.
“That is a sensitive measure of the mass of the object deposited on the surface,” explained Joerg Schreiber, Ph.D., COO of Quadraspec. Bound antigens change the interferometric response again. By measuring the resulting mass change, the system detects binding of the target protein without the need for a second antibody or label. Thus the technology enables label-free detection of the immune reaction.
The SDI disc enables multiplexing, offering the ability to measure up to 100 unique analytes in 260 patient samples. As the disk spins, a laser tracks the position and contents of each 125-micron spot (containing the immobilized antibodies). Although there is some noise in the system, noise-reduction methods account for this by subtracting out the background signal and averaging measurements within a spot and between spots.
Maven Biotechnologies’ (www.mavenbiotech.com) LFIRE™ is a label-free, high-density imaging system that measures molecular binding reactions in a microarray or well-plate format. LFIRE achieves real-time monitoring of reactions through the use of total internal reflection ellipsometry, a technique that measures changes in the polarization of light reflected from the interface between materials, providing optical properties of thin films. The light enters the underside of the microarray and reflects off the solid-liquid interface between the substrate and the sample solution. The presence of bound protein on the array and the resulting mass change due to specific binding of analytes is detected by measuring the changes in the polarization state of the reflected light.
Shane Dultz, Ph.D., CSO at Maven, described “the increased mass from the binding reaction” as an “internal label that permits sensitive measurement of biomarkers in biological fluids.” According to the company, “LFIRE has been validated with gene and protein microarray densities of 2,500 spots/cm2 and is capable of detecting surface attachment of molecules as small as 150 Daltons.” For antibodies that attach to the substrate surface, it can detect roughly 12 molecules/µm2.
The main advantages of LFIRE over surface plasmon resonance (SPR) technologies are that it does not require metal films, has more spatial resolution, and is inherently more sensitive than imaging SPR, explained Dr. Dultz.
“Furthermore LFIRE has been validated against commercial single-plex diagnostic kits (ELISA) and fluorescence multiplexed diagnostics (Luminex) to demonstrate capability in markets that SPR technologies have not been designed to address,” he added. “These commercial immunoassay markets require economical, high-throughput testing for single to multiple tests per patient, with an open-systems approach using disposables compatible with most robotic sample-handling technologies.”
LFIRE technology is in the early stages of commercial development—a commercial instrument is about a year away from launch. The company plans to deliver both slide-based and plate-based systems that would replace labeled assays used in diagnostics, research, and drug discovery and development applications and that would be compatible with disposable formats already on the market.