Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • ADVERTISEMENT FEATURE Advertiser retains sole responsibility for the content of this article

A probe-free rRNA depletion method expands RNA-seq use

Generating an accurate view of the RNA transcriptome is easier said than done, particularly for microbes and lesser studied subjects.Credit: Shutterstock

On December 20, 2018, MIT microbiologist Michael Laub tweeted: “Anybody also suffering from the discontinuation of RiboZero kits for bacterial RNA-seq? Has anyone got a work-around or alternative?”

Dozens responded to Laub’s post, many concerned about their ongoing experiments. Such an outcry is rare for a discontinued kit, but RiboZero occupied a valuable niche. The kit helped researchers eliminate ribosomal RNA (rRNA), which is essential for transcriptional profiling using RNA sequencing (RNA-seq). rRNA can make up more than 90% of a sample. If not removed, its sequencing reads will vastly outnumber reads from other RNA species, making datasets less informative.

Without recourse, some researchers suggested reverse-engineering the RiboZero kit, but that too presented a challenge. “The reagents used in depletion kits are often not well described,” says Harris Wang, a researcher at Columbia University in New York.

The only solution was to find a new method.

Getting rid of rRNA

RNA-seq uses next-generation sequencing technology to profile RNA in samples. Seeking insights into gene expression and transcription, researchers using RNA-seq attempt to measure mRNA and other regulatory RNAs, while reducing the detection of overly abundant species of rRNA, such as 16S and 28S.

Researchers usually have two solutions: directly capture RNAs of interest or eliminate uninteresting ones. To capture RNAs directly, researchers frequently employ poly-A+ selection. Here, primers with long stretches of thymine (T) nucleotides are used to bind RNAs with poly-A tails.

The approach is effective in mammals, but many organisms, such as bacteria, lack RNAs with poly-A tails. The approach also doesn’t work to isolate regulatory or noncoding RNAs, even in mammals. Researchers must instead eliminate rRNA, most often using probes that target rRNAs for isolation and capture.

While poly-A+ selection methods have not changed much in recent years, rRNA depletion methods are developing rapidly.

Finding the right probe

Claus Kuhn from the University of Bayreuth, in Germany, is one of a growing number of researchers developing their own rRNA depletion techniques. Kuhn’s lab studies Piwi-interacting RNAs (piRNAs) and Piwi proteins, regulatory proteins involved in stem cell differentiation in planarian flatworms.

“The deeper motivation for us to study piRNAs and Piwi proteins in planarians is our quest to uncover [this] potential link between piRNA function and regeneration,” Kuhn says. Planarians possess unique regenerative capabilities, thought to depend on non-coding RNA. To look for a link with piRNAs, Kuhn and his team needed more information on piRNA expression, which meant RNA-seq experiments, but here Kuhn hit a hurdle.

Most probe-based rRNA depletion strategies target conserved sections or motifs of rRNA, so they work with different organisms. Unfortunately for Kuhn, planarian rRNAs are very A-U rich, making multispecies probes unusable. Instead, Kuhn and his team synthesized custom, planarian-specific probes with biotin tags for easy pulldown1.

“We achieved amazing depletion rates, with only 2% rRNA left,” Kuhn notes. He also retained much of his target RNA. Because species-specific probes are fully complementary to rRNAs, he says, adjusting hybridization conditions can increase target specificity while decreasing the potential for off-target depletion.

Kuhn’s work is ongoing, but the custom probe approach allowed his team, for the first time, to study piRNA expression and function across the entire transcriptome.

Typical probe-based rRNA depletion strategies are effective, but often require custom probes and a number of steps, which risk losing material.Credit: Nature Research Custom Media

Scaling for the future

With poly-A+ selection, large-scale mammalian transcriptome analysis using RNA-seq is now fairly common. That’s not true for bacterial transcriptomes—an inequity that Columbia’s Wang cannot abide. “We’d really like to move microbial transcriptome analysis to the level of mammalian transcriptome studies,” he says.

Wang and his colleagues recently developed a new, scalable method for rRNA depletion2. Their approach uses oligonucleotides or amplicon-based single-stranded DNAs as probes to bind rRNA species as duplexes. In the early stages, the process largely mirrors other hybridization-based rRNA depletion strategies, but then it diverges. Rather than capture rRNAs using biotin or beads and magnets, Wang adds a ribonuclease that degrades the duplexes. In one stroke, Wang eliminates the rRNA and probes while retaining other RNA species.

“We can do this depletion in a single reaction, without [the need for] multiple clean-up steps where you lose material,” Wang explains. He says his team has examined around 1,500 bacterial transcriptomes so far, a number not feasible using probe capture strategies.

Probe-less depletion

In December 2019, the biotechnology company, Zymo Research, debuted a new rRNA depletion kit that removed the need for probes altogether. The Zymo-Seq RiboFree Total RNA Library Kit was based on work that had been developing since 2004. At that time, researchers at the Shemyakin and Ovchinnikov Institute of Bioorganic Chemistry in Russia published a paper describing how to normalize cDNA libraries using a double-stranded DNA nuclease (DSN)3.

DSN cleaves both double-stranded DNA and DNA in DNA-RNA hybrids. Because those hybrids form earlier, and with the most abundant RNA species, it was possible to cleave the cDNA from the highly abundant ribosomal RNAs highly abundant RNAs by timing the addition of DSN, opening the door to probe-free rRNA depletion methods.

A probe-free rRNA depletion strategy could speed RNA-seq workflows and enable rapid analysis of more organisms.Credit: Nature Research Custom Media

In 2011, Hana Yi and colleagues at Seoul National University applied this approach to deplete rRNA4. However, their DSN technique proved overly complicated; it involved extra experimental steps and longer RNA-seq library preparation times.

Inspired by this work, the Next-Generation Sequencing team at Zymo Research decided to develop their own novel method. Their RiboFree Universal Depletion approach does not require probes or capture steps, making it suitable for any species. That could benefit researchers who, like Kuhn and Wang, wish to apply RNA-seq to non-mammalian species. It could also simplify RNA-seq for scientists who study less-researched organisms, such as those used in agriculture, ecology, and marine biology.

In their kit, Zymo Research optimized the necessary reagents into a premixed format and significantly shortened the process time. Researchers can move from library prep to sequence in a single day, which could blunt the blow of reduced bench time due to COVID-19 restrictions. The approach’s simplicity also lends itself to high-throughput automation, making it useful for researchers plowing through increasingly ambitious transcriptome studies.

The impacts of Zymo Research’s kit remain to be seen. It is new, and almost certainly the evolution of rRNA depletion strategies will continue to advance. But where once researchers were panicked at the discontinuation of a tried-and-true kit, a new and more versatile one has now emerged.

To learn more about the Zymo-Seq RiboFree Total RNA Library Kit, the company has set up a dedicated source, complete with studies, on its site.

References

  1. 1.

    Kim et al. BMC Genomics 20:909 (2019)

    Google Scholar 

  2. 2.

    Huang et al. Nucleic Acids Res. 48(4):e20 (2020)

    Google Scholar 

  3. 3.

    Zhulidov et al. Nucleic Acids Res. 32(3):e37 (2004)

    Google Scholar 

  4. 4.

    Yi et al. Nucleic Acids Res. 39(20):e140 (2011)

    Google Scholar 

Download references

Search

Quick links