Jeffrey Perkel has been a scientific writer and editor since 2000. He holds a PhD in Cell and Molecular Biology from the University of Pennsylvania, and did postdoctoral work at the University of Pennsylvania and at Harvard Medical School.
When Biocompare last considered microRNA microarrays in 2011, we pronounced the field “smokin’ hot.” If PubMed is any guide, the heat is still on. A search for “microRNA” pulled up nearly 21,000 hits on the reference database, of which more than a quarter went to press in 2012 and half were published since 2011.
So just what is the fuss all about? To start with, microRNAs (miRNAs) are short (20- to 23-nucleotide) eukaryotic, noncoding regulatory RNAs that are enzymatically generated from longer pre-miRNA transcripts. The mature molecules inhibit translation of messenger RNAs by complementary base pairing between the miRNA and the mRNA, and numerous regulatory and disease processes, including some cancers, have been linked to their expression.
A decade ago, researchers were barely aware these molecules existed. Now, thousands have been identified: Version 19 of the miRBase database, released in August 2012, lists 21,264 identified, mature miRNA sequences from 193 species. That’s an increase of more than 3,100 entries vs. version 18, and the field shows no sign of slowing.
Researchers interested in exploring miRNA regulation in their particular systems will find plenty of tools to help. Whether your application is miRNA expression profiling, biomarker validation, or even target identification, you’re sure to find a product to suit your needs.
Expression analysis tools
As with mRNAs, researchers studying microRNA expression have three options: quantitative PCR (qPCR), microarrays and next-generation DNA sequencing. And, as with mRNAs, the choice of which platform to use depends on multiple factors, including time, cost and application.
According to Corinna Nunn, product manager for gene expression microarrays at Agilent Technologies, the three platforms are not so much competitive as complementary.
Generally, qPCR is used for focused or validation studies because it is fast—most researchers already have access to a qPCR machine—and relatively inexpensive. In addition, qPCR can provide both relative and absolute quantification of expression (though the latter requires a standard curve) and is best suited to measuring a relatively small number of miRNAs (96 or fewer) in a large number of samples.
“Microarrays are the most cost-effective method for profiling all microRNAs in many samples,” Nunn says. But they can really only quantify microRNAs that already have been identified; they cannot be used to discover new (that is, unknown) microRNAs. Microarrays are more comprehensive than qPCR assays but also are more expensive, take longer to run and provide relative quantification of expression levels, rather than absolute numbers.
The strength of next-gen DNA sequencing is its ability to discover new miRNAs in, say, novel or poorly studied species. It also can be used as a quantification method, a process called “digital counting,” but takes far longer than qPCR (most sequencing platform runs take days to complete, not including sample and library preparation steps) and costs considerably more. And of course, next-gen sequencing data must be processed and interpreted, a process that requires extensive back-end bioinformatics work.
At LC Sciences, a microRNA service provider based in Houston, customers mostly request sequencing and array services, says Christoph Eicken, head of microarray technical services—the former for discovery work and the latter for differential expression analysis. “Our customers don’t really use qPCR for profiling, they use it for validating leads they discover in sequencing or arrays,” he explains.
LC Sciences charges $495 per sample for popular catalog microarrays (human, mouse or rat) and $636 for all other species and custom designs (the company uses a proprietary microfluidic array platform). Turnaround time is as fast as five days. Sequencing costs $1,100 per sequencer lane, with up to three samples per lane, plus $400 per sample for library preparation and another $400 per sample for bioinformatics analysis. For customers running three samples, that works out to $3,500 total, or $1,167 per sample, with turnaround time of about a month.
“So the array is a little quicker and a lot less expensive,” says Chris Hebel, the company’s director of business development.
Expression analysis providers
MicroRNA microarrays are available from Affymetrix, Agilent Technologies, Exiqon and Life Technologies, among others.
Affymetrix’s GeneChip® miRNA 3.0 microarray (for use with the company’s GeneChip Scanner systems) and miRNA 3.1 array strip (for the company’s GeneAtlas® system) both include content from miRBase version 17, with nearly 20,000 total probe sets covering 153 organisms, as well as probes for pre-miRNAs and other small noncoding transcripts called snoRNAs and scaRNAs.
Affymetrix miRNA microarrays employ around 200,00025-mer probes in total, with nine replicates per miRNA, says Gianfranco De Feo, the company’s head of North American regional marketing. The resulting arrays, he says, offer advantages in terms of specificity, sensitivity, content and precision. De Feo claims the company’s arrays can detect as little as 0.5 attomolar levels of some miRNAs, as well as “even small changes” in expression levels. “We are unparalleled,” he says.
Agilent Technologies uses an 8x60K array format for its microRNA products, which derive their content from miRBase version 19. Human, mouse and rat designs are available off the shelf, and custom designs are available, as long as they use miRNAs from miRBase.
Agilent’s design uses 60-mer probes including a “stem-loop structure” to ensure high sensitivity and specificity, says Nunn. “The sequences of microRNAs are highly homologous,” she explains. “The stem-loop helps to differentiate mature miRNAs from longer RNAs and stabilizes the specific interaction between probe and miRNA.”
Exiqon’s miRCURY LNA™ microRNA arrays and qPCR panels also draw content from miRBase v19. The microarray includes 3,100 probes covering human, mouse and rat sequences, plus some viral miRNAs and select sequences not included in miRBase.
The probes in these assays use locked nucleic acid (LNA) insertions (designed by a proprietary algorithm) to equilibrate melting temperate and hybridization conditions across the array, says Kim Bundvig Barken, Exiqon’s global product manager for qPCR and sample preparation products.
“When working with microRNAs, the GC content can vary from 20% to 90%, which means the melting temperature varies quite significantly,” Barken explains. By adding LNA bases, Exiqon scientists can normalize those values so all probes on the array hybridize with equivalent efficiency. “Using LNAs, we can adjust the Tm of all assays so they perform equally well.”
Exiqon also offers a miRNome PCR panel including some 750-odd LNA-containing primer sets in two 384-well plates, as well as more focused panels for stem cell, cancer and toxicology research. Also available is a “Pick-&-Mix” panel option, with which customers can design their own PCR panels.
Life Technologies’ offerings, NCode™ Human miRNA Microarray Kit V3 and NCode Multi-Species miRNA Microarray Kit V2, cover miRBase versions 10 and 9, respectively. The company also offers the complete set of TaqMan® microRNA assays from miRBase v19, available in a number of assay formats (e.g., individual reactions, 384-well TaqMan Array Cards, etc.).
MicroRNA mimics and inhibitors
Also available are microRNA mimics and inhibitors, which are used to test the impact of exogenous miRNA expression and miRNA loss of function, respectively.
Such products are available from Life Technologies (mirVana™ miRNA mimics and inhibitors), Sigma-Aldrich (MISSION miRNA Human Mimics and MISSION Synthetic and Lentiviral miRNA Inhibitors), Thermo Scientific (miRIDIAN microRNA mimics, miRIDIAN microRNA hairpin inhibitors and miRIDIAN shMIMIC Lentiviral microRNA) and Qiagen (miScript miRNA inhibitors and miScript miRNA mimics).
Target identification and other tools
According to Eicken, the maturation of the miRNA field means that, at least for well-studied organisms, most microRNAs have been identified. Now, he says, researchers have moved on to target identification: “which microRNAs are regulating which genes.”
LC Sciences supports that application for agricultural researchers specifically with an application it calls “degradome sequencing.” In plants, Hebel explains, miRNA functions to degrade mRNA targets (whereas in eukaryotes, it functions by blocking translation).
In degradome sequencing, Hebel says, the company prepares a library that captures the degraded mRNAs in a sample. Fragments that are overrepresented can be identified through bioinformatics as specific targets of specific miRNAs. “That’s definitely one area where we’re seeing significant growth,” he says.
Another option for target site analysis is Affymetrix’s Axiom® miRNA Target Site Genotyping Array Plate (for the GeneTitan® platform). According to De Feo, this is a “large-format array” representing some 250,000 known miRNA target sites throughout the genome, which enables researchers to determine whether a particular sample harbors mutations or variants in those sites in the genome.
“The researcher can ask [if there are] any mutations in those sequences that might alter how a microRNA functions,” De Feo says.
Qiagen’s miScript Target Protectors enable researchers to specifically block miRNA binding to a specific target, thereby allowing them to probe the impact of a particular miRNA on that one mRNA. In one example on its website, the company illustrates how researchers can combine these tools to probe microRNA biology. The example involves DNMT3A, which encodes a DNA methyltransferase. The online miRNA target prediction site TargetScan identified a candidate binding site for miR-29a within the gene’s 3’ untranslated region.
“Transfection of a miR-29a miScript miRNA Mimic led to downregulation of DNMT3A as shown by real-time RT-PCR, indicating that miR-29a downregulates DNMT3A expression at the transcriptional level. Transfection of a miScript Target Protector designed for the miR-29a binding site predicted by TargetScan resulted in increase in DNMT3A expression. These experiments provided convincing evidence that DNMT3A is negatively regulated by miR-29a via the binding site predicted by TargetScan.”
For researchers interested in visualizing miRNAs in situ, Affymetrix offers the QuantiGene® ViewRNA miRNA ISH Cell Assay. The product, De Feo says, enables researchers to determine, for instance, where in a cell a particular pre-miRNA is processed, or if the mature molecule tends to be localized in a particular cell type or subcellular location. “That can be a huge boon in research,” he says. “It’s not in vitro, it’s in situ, in tissue.”
Rounding out the miRNA toolbox are some more fundamental necessities, such as miRNA isolation kits, cDNA preparation kits and tools for next-gen sequencing library preparation, among others.
Bottom line: If you’re tempted to probe what role, if any, miRNAs play in your particular research model, you can be sure the tools are there to make those studies possible.
The image at the top of the page is from Agilent Technologies.