Abstract
Introduction
The ability to modulate gene expression with reverse genetics methods is of great importance for both basic and translational research. Gene silencing allows identifying players involved in cellular processes such as signaling, metabolism, and differentiation, among others. The discovery of RNA interference (RNAi) 20 years ago suddenly allowed the downregulation of genes in mammalian cultured cells. Since then, other gene silencing methods based on genome editing have been developed such as zinc finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs), and, more recently, the clustered regularly interspaced short palindromic repeats/Cas (CRISPR/Cas) systems. Thus, many forward genetic methods now exist to downregulate gene expression in mammalian cells, each having its advantages and drawbacks.
RNA Interference
RNAi was discovered in the round worm Caenorhabditis elegans by Andrew Fire and Craig Mello in 1998. 1 This natural biological process entails the production of small double-stranded RNA oligonucleotides of 20 to 25 nucleotides in length called small interfering RNAs (siRNAs) that trigger the destruction of messenger RNAs (mRNAs) with complementary sequence via the RISC complex. RNAi is a universal phenomenon and can be found in many species.
It was quickly recognized that this natural process could be harnessed using synthetic siRNAs or plasmid-encoded short hairpin RNAs (shRNA) to downregulate any gene. Whole-genome siRNA libraries were synthesized and commercialized in the early 2000s, leading to the development of automated RNAi transfection protocols on high-throughput liquid handling robots in high-density microtiter plates. These developments enabled the systematic investigation of the role of every gene in the genome in any biological process.
For the pharmaceutical industry, these technologies allowed investigating genes for their potential as targets for small (chemical compounds) and large molecules (antibodies, nucleic acids). Before engaging on a costly drug development program, targets could be derisked or discovered using RNAi screens in the in vitro model systems of diseases. A second focus of interest for the pharmaceutical industry was the potential use of RNAi as a therapeutic agent in its own right. This led to an intensive search of reagents for delivering RNAi to tissues in vivo. In summary, it was the heyday of RNAi, with massive investments from both academia and the pharmaceutical industry.
As is often the case with high expectations, RNAi turned out to be trickier than anticipated. The initial enthusiasm for the technology was first dampened by reports of off-target effects in 2003 by Jackson et al. 2 Further investigations revealed that imperfect base pair matching of siRNA to mRNAs leads to microRNA-like translational repression. Then in 2008, a comparison of the hit lists of three RNAi genome-wide screens against human immunodeficiency virus (HIV) replication showed an overlap between all three screens of only 3 genes for a total of 834 genes identified as hits.3 –6 Just as the enthusiasm was great at the beginning, the disappointment was bitter; many abandoned RNAi as too unspecific, and pharmaceutical companies closed groups working on RNAi for target validation or therapeutic usage. The king was dead.
Since this phase of low popularity, much has been learned about the pitfalls of RNAi and how to either reduce or detect off-target effects: (a) powerful algorithms have been developed to design efficient RNAi molecules with reduced potential for off-target, (b) chemical modifications have further enhanced the specificity of synthetic siRNA libraries, and (c) bioinformatics tools for detecting likely off-target effects based on the homology of the seed region have been developed. These advances have led to a revival of interest in RNAi, and more laboratories are using the technology again.
Also, the application of RNAi for therapy has been revived in recent years thanks to breakthroughs in stabilizing RNAi in serum and delivering to tissues in vivo. Alnylam has successfully concluded a phase II clinical trial for its lipid nanoparticle (LNP)–delivered siRNA for treating transthyretin-mediated amyloidosis. Thus, RNAi is also gaining again a renewed interest for therapeutic use, although organs other than the liver remain a challenge to target. Long live the king!
Due to the renaissance of the field, we felt it was time to assemble a special issue of the Journal of Biomolecular Screening (JBS) dedicated to RNAi screening.
This special issue highlights some of the advances that have helped rehabilitate RNAi. For instance, in recent years, cell-based screening in sophisticated cellular systems has been gaining popularity. Carrying out RNAi screens in these systems used to be very challenging, but technological and methodological advances have somewhat alleviated those issues. Here several articles report methods to either efficiently optimize transfection protocol or access difficult cell types. Xiao et al. 7 present a multifactorial protocol to rapidly optimize lipid-based transfection conditions using acoustic dispensing. Using their protocol, most cell types should be amenable to RNA silencing in high throughput. For more recalcitrant cell types, Luft and Ketteler 8 review high-throughput electroporation methods to deliver reagents in virtually any cell type. Freeley et al. 9 present their work with self-delivering siRNA in primary human T cells, a notoriously difficult-to-transfect cell type. Last, virus-delivered shRNA allows accessing many transfection-resistant cell types. The performance of virally delivered pooled shRNA screens depends on many factors that are experimentally difficult and costly to assess. In this issue, Stombaugh et al. 10 present a software tool that allows estimating with high accuracy the performance of a screen and to design an appropriate screening protocol to maximize the performance. This collection of articles should help readers to set up their screens in any cellular systems they wish.
The major concern of RNAi is off-target effects, and a number of strategies have been developed to address this issue. Here, Christel et al. 11 describe a postscreening strategy using overexpression of candidate RNAi hits to verify the on-target phenotype of the shRNA screen. Their method allows validating top candidates stemming from screens by quickly designing efficient shRNA vectors to replicate the phenotype and then revert it by expression of a silencing resistant version of the gene.
Another method to evaluate off-target effects is to compare the phenotype caused by several RNAis against a gene with high-content analysis of images. Many parameters can be extracted from images to describe cellular morphology with high accuracy so that on-target effect can be detected as the most frequent phenotype. 12 In this edition, contributions by Rameseder et al. 13 and Zhong et al. 14 offer computational tools to analyze high-content imaging RNAi screens allowing laboratories lacking computer scientists to extract the maximum value out of their images. Last, RNAi is popular to modulate systematically other RNA populations such as microRNAs and long noncoding RNAs (lncRNA). In this issue, Eulalio and Mano 15 discuss the challenges of identifying the crucial genes regulated by microRNAs that have been identified in RNAi screens. Theis et al. 16 present a novel endoribonuclease-prepared small interfering RNA (esiRNA) library targeting lncRNA and present a compendium of lncRNA expression in 11 cell lines and a method for localization using in situ fluorescence hybridization.
With the advent of genome editing technologies, a discussion about RNAi cannot omit mentioning this alternative method to silence genes. A report by Wade 17 and another one by Woodcock and Taylor 18 explore the potential of CRISPR technologies and compare them with RNAi technologies.
Summary
The king is dead, long live the king! It is clear that RNAi is enjoying a revival of popularity and that it is increasingly being applied to discover and validate targets and being investigated for therapeutic use. We hope this JBS special issue will show you the progress the technology has undergone and help you to better plan your next RNAi screen.
