Josh P. Roberts has an M.A. in the history and philosophy of science, and he also went through the Ph.D. program in molecular, cellular, developmental biology, and genetics at the University of Minnesota, with dissertation research in ocular immunology.
Research into the presence and roles of microRNA (miRNA) continues to advance with blazing speed. Yet for many, miRNA discovery and investigation are more a means to an end—such as a deeper understanding of the mechanisms underlying HER2+ breast cancer—than the focus of their research. Similarly, a lab may be less interested in the intricacies of microarray or next-generation sequencing library preparation than in simply knowing which miRNAs are upregulated following a particular drug treatment.
For such labs, outsourcing to a contract research organization (CRO) skilled in miRNA experimental design, execution and interpretation may be the best option. Outsourcing may even save valuable resources by assuring that time and money are not squandered on unnecessary or botched work and wasted samples.
Why outsource?
Beyond the sheer convenience of getting someone else to do the work, researchers might look to outsource because miRNA expression analysis is not the major focus of their lab and they lack the wherewithal to undertake the necessary tasks. Core facilities may be equipped to do qPCR, run Affymetrix chips or even set up massively-parallel next-generation DNA sequencing (NGS), yet they tend to be generalists and so may not have the expertise necessary to design and interpret miRNA experiments.
“Generally our customers would use our service because they don’t have the experience looking at these types of data and doing these types of experiments, employing these types of methods,” says Chris Hebel, vice president of business development at LC Sciences. “There is a certain skill level and understanding that needs to be there to be able to get reliable data.”
Indeed, the value of a CRO very often is practical experience. “This is our business,” explains Patrick Hurban, global head of genomic research and development at Expression Analysis, a Quintiles company. “We have a great deal of understanding [about] how these assays work, and how they don’t. All of the things you would have to do as an investigator to make your way up that learning curve, we’re already done all that for you.”
That learning curve includes handling the data that come out of an experiment. Many, if not most, miRNA-focused CROs have a strong bioinformatics team, and a certain amount of data analysis is built into the cost of the service. Publication-ready tables and figures often are delivered to the researcher along with the raw data.
Tools and choices
Service providers are equipped with several expression-analysis tools in their toolboxes, the most common being some version of a high-density microarray, qPCR and NGS. They can help researchers understand the pros and cons of each of these approaches and guide them toward the most appropriate for their particular needs.
Perhaps the first consideration is whether you’re looking to profile known miRNAs in your samples or discover new ones. For well-studied species like human, mouse or rat, an array may be a good solution. “Just about all the miRNAs that are going to be found have been found” in those species and can be put onto a single chip, notes Hebel. But, he cautions, pay attention to whether the latest version of miRBase (currently Release 20, from June 2013) is being used to synthesize the chips, as you don’t want to be relying on outdated data.
An array is robust, has a simple workflow and the data are easily interpreted. And “it’s still less expensive and quicker than sequencing,” Hebel adds. “We only take about two weeks to do a full array service, including the data analysis.”
On the other hand, for cases where maybe not very many miRNAs have been found—in agricultural genomics, for example—or when looking for novel isoforms or splice variants, sequencing may be the better option.
CRO Expression Analysis generally tries to steer its customers directly to sequencing, because of what it considers to be “the superior technical merit of the platform.” (The exception is for labs already in the midst of a microarray-based research program: “They may wish to stay with microarrays simply to ensure data continuity,” Hurban says.) The open-ended NGS platform allows for a wide dynamic range of miRNAs to be queried without prior knowledge of their sequence, meaning even previously unknown molecules can be identified and counted. The amount and complexity of data generated may be daunting, but that’s what the CRO is for.
And, because miRNAs are all approximately the same length, sequencing miRNA can be more quantitative than it is for mRNA, says Marie-Louise Lunn, director of product management at Exiqon, which launched a miRNA NGS service in January.
A third option is qPCR, which is more sensitive and more quantitative, but requires less sample than NGS or microarrays, Lunn notes. qPCR assays do not cover the entirety of miRBase, so researchers typically choose a focused panel consisting of individual assays, either curated by the vendor or chosen by the researchers. Such assays often are used to validate microarray findings.
As a riff on this, Arraystar’s miRStar™ Human Cancer Focus miRNA & Target mRNA PCR Array enables a researcher to simultaneously quantify both miRNAs and their targets, “so instead of using [the] theory that a particular miRNA degrades or down-regulates an mRNA, you have the experimental data,” explains Marc Rogers, assistant sales and marketing manager with the company.
More to think about
miRNA analysis requires RNA, but how does the CRO get it? For a fee, many providers will perform an RNA-extraction service, from frozen cells or formalin-fixed paraffin-embedded tissues, for example. If you’re sending your own purified RNA—often the best choice, especially if samples need to clear Customs—be sure to use a kit or protocol designed to retain small RNA.
Services are typically billed on a per-sample basis (if there are cost savings to be had by running multiple samples together, these are often passed along). But don’t skimp on biological replicates and controls if you’re looking to get statistically meaningful data. Of course, “some people are just doing a pilot study to see what shows up,” says Rogers. “If they see something they’re interested in, then they follow up with a more in-depth study.”
Be clear about what services you’re contracting for and what you expect as deliverables. Some providers, for example, will include novel miRNA discovery as part of an NGS package, while others charge extra for it.
Also, do your homework. Some CROs make lofty claims about their capabilities, but don’t take their word for it. Check them out. “Many of our customers feel it’s important to deal with someone who knows miRNA,” notes Lunn.
And finally, look at the whole package. Prices and turnaround times tend to be somewhat competitive across the industry, so customers often choose not on the basis of price but rather on the “perceived value of the service that we provide overall—not only what we can do in the lab but their interactions at project-management level as well as bioinformatics capabilities that we have,” says Hurban.
With such details out of the way, you can get back to the science you do best. After all, isn’t that the point of a CRO?