Orthogonal Validation: A Means To Strengthen Gene Editing and Gene Modulation Research

From RNA interference (RNAi) to CRISPR, there are several methods that researchers can use to manipulate gene function, each with its own strengths and weaknesses. Orthogonal validation – the synergistic use of different methods – makes genetic perturbation studies more robust. Utilizing complementary methods, including RNAi and CRISPR-knockout, -interference and -activation, enables researchers to have confidence in their results.

RNAi, CRISPR and CRISPR interference (CRISPRi)

A widely used approach for studying gene function is to reduce or disrupt its basal expression. Methods employed to knockdown or knockout gene products can answer specific questions about gene function, in addition to helping identify and develop therapeutics. RNA interference (RNAi) was the primary tool for disrupting gene expression (along with zinc finger nucleases (ZFNs) and transcription activator-like effector nuclease (TALENs)) but newer CRISPR-based technologies, including CRISPR knockout (CRISPRko) and CRISPR interference (CRISPRi) are viable options. With an expanding toolkit for gene disruption available to the researcher, it is becoming increasingly important to understand the benefits and limitations of each technology for a given application, and how to validate potential findings using orthogonal approaches. Combining orthogonal loss-of-function (LOF) methods in parallel (orthogonal validation) can help reduce the possibility of spurious results (false negative or positive results). In this manner, orthogonal validation is an essential component of a well-designed and -controlled gene modulation experiment.

As gene-modulating tools, RNAi and CRISPR technologies are often grouped together, despite their disparate modes of action and considerations for effective usage. Both platforms require RNA-based reagents, and can lower gene expression, but they differ in more ways than they are similar.

RNAi involves transferring a double-stranded RNA (dsRNA) into a cell where it is cleaved and processed into smaller RNA fragments (21 nucleotides). When the RNA fragments complex with a complementary (target) mRNA sequence and an endogenous silencing complex, mRNA is fragmented, preventing its translation. Importantly, RNAi-based knockdown is temporary, and effect duration is contingent upon RNA turnover (both for endogenous mRNA and exogenous RNAi modulators). RNAi using dsRNA has been found to evoke a double-stranded RNA immune response in mammalian cells which can affect cell viability. This toxicity can be avoided by switching to short interfering RNAs (siRNAs), which are truncated RNA duplexes that bypass the aforementioned dsRNA cleavage step, retain targeted gene silencing abilities of dsRNA, and do not trigger the deleterious immune response in mammalian cells.

In contrast to RNAi, CRISPR-based gene manipulation requires a guide RNA and a programmable Cas endonuclease protein. The guide RNA complexes with the Cas endonuclease, is transported to the nucleus, and creates a DNA double-strand break (DSB) at a site pre-programmed in the targeting sequence of the guide RNA. Once detected by the cell, blunt-ended DSBs are repaired by the non-homologous end joining (NHEJ) pathway, which is notorious for adding or removing stretches of DNA during repair. From this cleavage and repair process, sequence insertions or deletions (indels) often cause errors in translation, leading to functional gene, and subsequent protein, knockdown.

Achieving a mechanistic and experimental middle ground between CRISPR-mediated gene knockout, whereby gene expression is reduced at the DNA level, and RNAi, in which target knockdown is not permanent, is CRISPRi. CRISPRi features a nuclease-deficient Cas effector, dead Cas9 (dCas9) protein, which has been engineered to prevent formation of DSBs, but retains the ability to form specific complexes with DNA. Most CRISPRi formats involve fusion of dCas9 protein to a subset of transcriptional inhibitors, that when complexed with CRISPRi-specific guide RNA, is programmed to associate with its genomic target, resulting in repression of target genes. Mechanism of gene repression can be the involvement of attached transcriptional silencing modules or by steric hindrance, preventing binding of transcriptional machinery. While CRISPRi gene silencing is transient, it is possible to tailor the silencing modules such that alterations to DNA (e.g., epigenetic modifications) can result in gene knockdown with comparatively greater longevity than RNAi.

Reagent delivery also differs between the three knockout/knockdown technologies. RNAi, comprised of a simple dsRNA, is relatively straightforward and can be transfected into cells using a single-step method with minimal effect on cell viability. Use of an shRNA format enables utilization of RNAi technology, but with an extended effect duration. Viral transduction, and stable genome integration of shRNA cassettes enable continuous transcription of RNA that enters the endogenous RNA silencing complex, described above. Such shRNA cassettes can feature inducible elements and harbor fluorescent and/or drug selection markers, enabling control over the initiation of LOF experiment and integrant selection, respectively.

Compared to RNAi, CRISPR and CRISPRi both involve a Cas protein and guide RNA, and both LOF formats have multiple choices for transferring reagents into cells, which should be informed by desired effect duration and experimental model under study. For example, CRISPRko or CRISPRi machinery can be stably integrated into cell genomes via viral transduction. After a selection process, resulting cells will stably express the Cas, or dCas-effector protein, and the LOF experiment can be initiated upon transfecting guide RNA when gene interruption is desired. Systems where expression of an integrated Cas protein can be initiated by addition of a chemical (drug, ligand, etc.) can add an additional level of control to the LOF experiment. All-in-one stable expression systems can simplify CRISPR experiments by including the Cas protein and sgRNA in a single integration event. Other ways of approaching CRISPR experiments feature the transient introduction of effectors, including transfecting combinations of mRNA-encoding Cas or dCas-effector protein with guide RNA into a population of cells.

Off-targeting effects, or the unintended genetic modulation (in this case, loss-of-function) at untargeted genomic sites, has been reported for RNAi, CRISPRko and CRISPRi technologies. In most cases, off-targeting occurs when non-target sequences in DNA or RNA exhibit a similarity to the targeted sequence. In practice, the risk of off-target effects can be minimized by taking proper precautions in reagent design. In RNAi, miRNA-like off-targeting can affect mRNA targeting specificity, causing downregulation of unintended gene targets. Such off-targeting can be prevented in siRNA experiments with position-specific modifications to the seed region, the use of modified nucleotide bases and differential modifications to the passenger RNA strand. Off-targeting in CRISPRko experiments can occur when the guide RNA causes nonspecific binding of the Cas9 nuclease to unintended genomic locations, resulting in permanent and heritable edits to locations outside of the experimental intention.

Undesired off-target edits by the CRISPR system can be minimized with rigorous design of the guide RNA, minimizing flaws and defects when bound to the genomic target. CRISPRi off-targeting can occur when guide RNA causes nonspecific binding of dCas9 complex to unintended transcriptional start sites, and can also be ameliorated with rigorous guide design. In addition, certain transcriptional start sites can be part of a bidirectional promoter, where dCas9-tethered transcriptional activators can initiate transcription of multiple gene products. Similar to guide RNA design for CRISPR knockout experiments, predictive algorithms and extensive guide design can help circumvent CRISPRi off-targeting due to bidirectional promoters.

Since CRISPRko, CRISPRi and RNAi are similar in effect, but dissimilar in permanence and delivery, there are several considerations relevant to the gene target in question, cell type in the study and experimental design. A non-exhaustive list of factors that can affect gene modulation experiments is shown in Table 1.

Table 1: Comparison between RNAi, CRISPRko and CRISPRi

It is important to note that none of the modulation technologies discussed herein “check all of the boxes,” and, in many cases, exhibit a comparatively differential balance of positive and negative attributes. When designing gene interference experiments, it is of paramount importance to consider the experimental design and hypothesis, such as the cell type(s) being studied, experimental hypothesis, cellular reactivity, nature of gene target, and the desire to study effects of gene target recovery after interference. CRISPR knockdown, CRISPRi and RNAi each have underlying strengths and weaknesses, and, as such, it is not easily possible for one gene interference technology to optimally address all of these considerations. Each platform presents with a known and inherent risk that a particular resultant phenotype could have, either in whole or in part, arisen from a technical limitation or an uncontrolled experimental artifact.

Benefits of orthogonal validation

The risk of results being associated with technical artifacts or variables can be reduced or eliminated by using additional independent interference technologies, known as the process of orthogonal validation, thus increasing confidence in downstream findings. Orthogonal validation can be thought of as using independent assays to eliminate the possibility of false negative or positive findings. In this manner, orthogonal validation may be an essential component of a well-designed and -controlled experiment. An additional benefit of introducing an orthogonal method is the possibility of revealing additional properties of the target gene. For example, by comparing results from RNAi and CRISPRko experiments, one might find that the gene product is necessary for cell viability or exhibits dosage effects. Two such studies ^1,2 present an empirical rationale for pursuing a strategy based on orthogonally validating results from either of the three gene interference methods. These studies illustrate the risk of attributing primary experimental findings to purported effects of the gene target under examination. In each case, the authors were able to eliminate data that were (likely) attributable to nonspecific effects of the experimental setup, underscoring the importance of orthogonal validation in gene interference experiments.

Gene interference is, due to the invasive properties of the method (forcing reactive nucleases and or RNA into cells), a complicated experiment, with several moving parts. However, when properly controlled and validated, resultant experimental data can be of reproducibly high quality. In many cases, orthogonal, or independent, validation is a requisite addition to the gene editing toolbox and can ensure that consequent findings are arrived at in the most reliable and efficient manner.

 Credit: Revvity.

About the author:

Dr. Ziemba joined the Revvity team in 2021. Brian obtained his PhD in Pharmacology and Structural Biology where he explored properties of alcohol- and odorant-binding proteins. Thereafter, he studied signaling arrays that govern chemotaxis as a postdoctoral researcher and managed the biochemistry arm of the Integrative Physiology of Aging Laboratory in Boulder, Colorado. He is currently fascinated with the field of genome engineering and enjoys family, cats, music, building things and the outdoors.

1. Smith I, Greenside PG, Natoli T, et al. Evaluation of RNAi and CRISPR technologies by large-scale gene expression profiling in the Connectivity Map. PLOS Biology. 2017;15(11):e2003213. doi: 10.1371/journal.pbio.2003213

2. Stojic L, Lun ATL, Mangei J, et al. Specificity of RNAi, LNA and CRISPRi as loss-of-function methods in transcriptional analysis. Nucleic Acids Res. 2018;46(12):5950-5966. doi: 10.1093/nar/gky437