Skip to main content
Premium Trial:

Request an Annual Quote

Team Designs Method for Sequencing Small RNA From Single Cells


NEW YORK (GenomeWeb) – A team of researchers from the Ludwig Institute for Cancer Research and the Karolinska Institute in Stockholm, Sweden has designed a method to sequence small RNA from single cells.

The technique has an advantage over standard small-RNA sequencing protocols in that it requires much less starting material, enabling small RNA from single cells to be sequenced, and it is quantitative, first author Omid Faridani told GenomeWeb. The researchers published their method last week in Nature Biotechnology and have filed a patent application on the method, which Faridani said could be applied to study small-RNA expression in complex tissues like tumors and embryos.

Sequencing small RNA, which includes microRNA and RNA fragments derived from tRNAs, typically requires large amounts of sample. As such, sequencing these molecules at the single cell level has not been possible, despite the fact that some studies have shown that they could have tissue, cancer-type, and cell-state-specific expression. In the study, the researchers showed that with their method, decreasing the starting RNA down to 1 nanogram from 1,000 nanograms, they could detect comparable numbers of microRNA molecules. At 10 picograms, which is close to the single-cell level, they detected 40 percent of the microRNAs found at the 1,000-nanogram level.

Faridani is a postdoctoral associate in Rickard Sandberg's lab at the Karolinska Institute, which has developed other single-cell RNA sequencing methods, including Smart-seq2, which has since been commercialized and is exclusively supplied by Clontech Laboratories.

Faridani said that the main problem with typical small-RNA sequencing methods is that the libraries have "huge adapter dimers" that require size selection steps to remove. The size selection, which involves running a gel, results in a lot of material being lost, so researchers must have a large starting amount.

However, Faridani's team got around this gel-based size selection. They constructed libraries by ligating adapters to all RNA molecules that had a 5' phosphate and a 3' hydroxyl group, regardless of their size. Then, they used so-called "masking oligonucleotides" to avoid the abundant rRNAs. An enzymatic digestion reduced the formation of adapter dimers. In addition, barcodes were added to the 5' adapters to enable counting.

Aside from reducing the amount of starting material needed, Faridani said, another important step was improving cell lysis. Small RNAs often bind to protein complexes, and when cells are lysed, they stay bound to those protein complexes. "We have to make sure that all the small RNAs are released," he said. The researchers used a lysis buffer that essentially denatured the protein complexes without harming the RNA molecules.

They first tested their method on human embryonic stem cells and also compared single-cell small-RNA sequencing of an embryonic kidney cell line known as HEK cells to bulk small-RNA sequencing of the HEK cell line. They developed a bioinformatics pipeline to assign reads to the different small-RNA groups, focusing on microRNAs, transfer RNA-derived small RNAs (tsRNA), and a type of RNA fragment derived from small nucleolar RNAs called sdRNA.

The team captured 3,800 microRNA, 3,500 tsRNA, and 600 sdRNA molecules per cell, on average.

When they compared microRNA expression from naïve and primed embryonic stem cells, they found that 60 percent of the molecules were differentially expressed. For example, the miR-302 family, which is associated with cell cycle regulation and apoptosis, was more highly expressed in primed embryonic stem cells, while the miR-371-3 cluster, which is important for pluripotency, was more highly expressed in naïve embryonic stem cells. Looking at cell-to-cell variability, the researchers found that the miR-371-3 cluster varied in expression across primed human embryonic stem cells, but not naïve cells.

To see if they could identify cell types based on microRNA expression, the researchers sequenced small-RNA libraries from single glioblastoma cells, naïve and primed embryonic stem cells, and HEK cells. They found that microRNA expression could separate the different cell types. "Single-cell miRNA profiling may therefore have unrecognized potential to decode cellular heterogeneity within complex tissues," the authors wrote.

Faridani said the technique would have applications in studying tumors and could help better understand tumor heterogeneity. In addition, he said, it could have applications for in vitro fertilization. "We hope that by measuring the miRNAs secreted from human embryos, we can predict the best embryo for IVF," with the goal of increasing the success rate of the procedure.

The researchers have filed for a patent on the method, Faridani said, which would be owned and licensed out by the Ludwig Institute for Cancer Research.