December 1, 2014 (Vol. 34, No. 21)
Lisa Heiden Ph.D. Director of Business Development MyBioSource
Yes, many are attracted to next-generation sequencing, but loyalty to chip-based phenotyping has its rewards, particularly in screens of large sample sets.
The microarray, the legacy technology for measuring gene expression, will not go gentle into that good night. True, in some respects, the microarray will be eclipsed by next-generation sequencing (NGS), specifically, whole transcriptome shotgun sequencing. But the microarray will remain indispensible for many applications. Beyond that, the microarray will evolve to complement NGS applications, finding reasons to stay relevant besides the
loyalty of its long-time users.
Microarrays are built upon the concept of hybridization, the binding of cDNA, complementary DNA strands derived from sample RNA, to strands of synthetic DNA. In the prototypic microarray, the synthetic strands are attached to a small, solid support. This support—a biochip—can be manufactured with thousands of probes or capture agents.
Microarray technology evolved from Southern blotting, which was invented by Edwin Southern, Ph.D. at Edinburgh University in 1975. In Southern blotting, DNA fragments are blotted onto a membrane and probed with known sequences to identify specific sequences via probe-target hybridization assays.
At Stanford University in 1995, microarrays were first developed as miniaturized cDNA arrays by Mark Schena, Ph.D., a postdoctoral biochemistry fellow who was later proclaimed the “Father of Microarrays” by DDN. Dr. Schena, now president and chief science officer at Arrayit, remarks that over the past two decades, microarrays have become a deeply rooted and highly developed technology.
In fact, according to Todd Martinsky, founder and executive vice president of Arrayit, the ex situ method (Arrayit U.S. Patent 6.101.946) has enabled all types of biomolecules to be put into microarray formats as capture agents. This includes antibodies, antigens, bacterial components, carbohydrates, DNA/cDNA, lipids, oliogonucleotides, peptides, proteins, and other small molecules as well as whole cells and tissue samples.
After all this development, it may be a good time to reflect on the best uses of microarray technology. Some of these uses are already established; others, such as those complementing NGS, are still emerging.
Large Sample Sets
By allowing simultaneous analysis of thousands of parameters within a single experiment, microarrays are ideal for a range of applications with large sample sets. For example, Dr. Schena says “a single newborn screening array can allow 1,000 babies to each be tested for 80 different diseases; that’s 80,000 tests at once!”
As Daniel St. Louis, Ph.D., the senior vice president and general manager of the expression business unitat Affymetrix, says, “Microarrays are perfect for genomic studies requiring processing of large numbers of samples, fast turnaround time, precise data, or cost effectiveness. Clinical applications such as genetic analysis in reproductive health (prenatal, preimplantation, and postnatal), oncology (solid tumor and hematological cancers), and transplantation are all good examples where time, accurate data, and affordability are of the essence.”
Microarrays are increasingly “playing key roles in translational science as new discoveries in the laboratory move to the clinic or field for routine applications,” says Dr. St. Louis. In fact, “over the past several years, we’ve seen a fascinating transformation of the microarray landscape,” indicates Daniel Peiffer, Ph.D., senior marketing manager, Illumina.
Drs. Louis and Peiffer both point to the agricultural–food industry, where large numbers of samples must be processed quickly and cost-effectively if seasonal requirements for marker-assisted breeding programs are to be satisfied. In fact, various microarrays already support crop and livestock genotyping. Dr. Peiffer adds that growth in agrigenomics “has been tremendous,” particularly in livestock applications, “where microarrays are being used for genomic selection by increasing accuracy and improving phenotype prediction.”
Microarray-based techniques to screen for genetically modified organisms (GMOs) in the food supply are an example of technology in the development phases that may one day become routine. However, the topic of labeling products containing GMOs is politically charged, and it remains to be elucidated if the risk of GMOs warrants routine testing.
Diagnostic Test Panels
Microarrays have been used from the get-go to study the genetic factors involved in disease. Their applications in the clinical arena are one of the greatest success stories of translational science.
“Being able to get to a highly reliable answer quickly and without extensive additional cost is a key attribute necessary for any lab,” says Nicole Ellis-Ovadia, senior global product manager, molecular cytogenomics, Agilent Technologies. In this regard, the microarray platform is perfectly suited for inclusion in testing approaches that leverage panels that can, for example, identify key genetic factors involved in a particular phenotype of concern.
“Comparative genomic hybridization (CGH) microarrays are the gold standard for genome wide copy number changes,” continues Ellis-Ovadia. “CGH arrays may be used in combination with karyotyping and phenotyping for understanding genetic abnormalities, particularly for those associated with postnatal developmental delays.”
“Another example of microarrays used in diagnostics,” says Dr. St. Louis, “is the Affymetrix CytoScan® Dx Assay, which is the first-and-only FDA-cleared whole-genome microarray for aiding physicians to diagnose postnatal intellectual disability and developmental delay.”
“In the area of cancer research,” adds Ellis-Ovadia, “CGH arrays can help guide treatment based on subtyping of the cancer to help understand the best course of action.”
Archival formalin-fixed, paraffin-embedded (FFPE) tissues, invaluable resources for cancer research, are benefitting from specialized microarray-based assays. For example, as Dr. St. Louis notes, “Affymetrix’ OncoScan® FFPE Assay Kit can perform genome-wide copy number analysis from as little as 80 ng of DNA from banked FFPE tissue samples.”
Advances in basic research are fueling new and innovative personalized medicine approaches involving microarray applications. For example, the ovarian cancer antibody protein microarray Arrayit OvaDx® is built upon the concept that inflammation and tumorigenesis are inextricably intertwined. According to Brian Kinnerk, vice president of operations, Arrayit, “OvaDx profiles approximately 100 serum protein biomarkers associated with ovarian tumor-induced immune system activation. It is used for both presymptomatic diagnostic testing, where inflammatory markers may be identified up to five years prior to overt symptoms, and disease monitoring.”
“Tissue microarrays (TMAs) are perfect tools for investigating biomarkers in FFPE tissue samples,” says Stan Krajewski, M.D., Ph.D., founder and scientific officer of Cellestan ImmunoQuant and research professor at Sanford-Burnham Medical Research Institute.
“However, TMAs still lack routine high throughput and standardized methods. They should be standard tools for clinical trials and FDA drug approval processes, but we just are not there yet,” concedes Dr. Krajewski. “Hopefully, the emerging digital pathology field with pattern recognition and algorithm quantitation tools, such as the Indica Labs and Aperio-Leica platforms, will soon bring readouts from next-generation TMAs to high-throughput efficiency levels comparable to other microarray platforms.”
Xuan Van Le, M.D., director of research pathology at ILSbio, contends that “the automated construction of next-generation TMAs results in better quality and affordability than conventional TMAs.” On that note, Dr Schena adds that “automated, mass-produced next-generation TMAs containing thousands of tissues are just around the corner.”
“A system biology approach that correlates next-generation TMAs with protein, DNA, and RNA microarrays from large patient cohorts is the wave of the future,” asserts Phillip Schwartz, president and CEO, Protein Biotechnologies. “We develop protein microarray content with patient donor tissues (tumor and normal adjacent) procured by our sister company ILSbio through its global tissue collection network and approved institutional review board (IRB) protocols. The reserve frozen and FFPE tissue can be used for processing other microarray types.”
Microarray technology is increasingly being compared and contrasted to NGS. NGS evolved as a group of new, high-throughput DNA sequencing technologies that dramatically reduced sequencing costs. Today’s NGS sequencing platforms include genome sequencing and resequencing, transcriptome profiling (RNA-seq), methylation analysis (methyl-seq), DNA-protein interactions (ChIP-seq), and exome analysis.
In academic publications, news journals, and other public forums, opinion leaders have tried to answer a recurring question: Why not just sequence everything rather than use the more targeted microarray technology? The answer, in many instances, is that sequencing and microarray technologies only appear to be rivals. In fact, the supposed can often work well together.
As Kevin Poon, Ph.D., global product manager, gene regulation, Agilent Technologies, points out: “Whether it’s NGS, a microarray, or some other genomic technology, [each kind of tool has] distinct advantages. NGS is a very powerful tool for discovery, given that you don’t need prior sequence knowledge. In contrast, we shouldn’t be surprised that the bulk of molecular diagnostic tests are actually done using data-driven [polymerase chain reaction] technology in conjunction with microarrays.”
Microarrays are less expensive and get to the point faster and more efficiently than NGS. In addition, microarrays are not burdened by the data analysis challenges inherent in NGS. Specialized computational methods are still evolving for the massive amount of data generated, and, as Dr. Poon says, “NGS analytical pipelines are less developed than those for microarrays.”
“The two technologies, NGS and microarrays, are actually more synergistic or complementary than competitive,” adds Dr. Schena, and “new sequences discovered through NGS become rich content for microarrays.”
Yong Yi, global marketing director, NGS and gene regulation, Agilent Technologies, offers the miRBase microRNA database as a good example: “This database only really started growing at a rapid pace once NGS became widely adopted. As NGS results feed the database, the new transcripts are tapped to generate microRNA microarrays for follow up studies.”
“From bovine arrays, to focused arrays in the cancer and immunology spaces, to new sets of variants,” Dr. Peiffer elaborates, “NGS is constantly informing ways to provide innovative and highly impactful content sets.”
The important thing is to use “the right tools for the right applications,” concludes Dr. St. Louis. “Irrespective of what other technologies may emerge, there will always be the right applications for microarrays.