siRNA screening basics: Setting up a successful arrayed siRNA screen

Arrayed siRNA screens are a powerful tool to look at very precise perturbations in cellular pathways. By applying this type of screening, researchers know which gene transcripts are targeted in each well and phenotypic changes can be detected in relatively short time periods, usually between 24 and 122 hours.

Pre-requirements
Pilot experiments should be considered before starting a robust screen.

Library size: confirm there is the capacity to store and handle the library (outlined in the Guidelines and Recommendations for Dharmacon™ siRNA Libraries), to ensure library integrity. The creation of daughter plates from the library should be considered to avoid multiple freeze- thaw cycles, especially when working with automated screening systems.
siRNA format and siRNA type: select individual and/or SMARTpool siRNAs, as well as suitable siRNA modifications (ON-TARGETplus or Accell).
Controls: control siRNAs should be included on every library plate to increase the robustness of the results. Use multiple positive and negative controls to distinguish sequence-specific silencing from non-specific effects. Additional assay-specific controls are recommended to capture moderate and strong phenotypes.¹
Transfection conditions: use gene-targeting controls to evaluate delivery conditions by measuring knockdown efficiency via RT-qPCR in the same plate format intended for the screening. These preliminary experiments will help to determine a suitable cell density, siRNA concentrations, and DharmaFECT formulation and volume (note that Dharmacon's self-delivering Accell siRNAs do not require transfection reagents). This also serves to evaluate the qPCR efficiency.
Plate layout: control reagents can be added to empty wells in catalog libraries. Alternatively, custom plate layouts are feasible via our Cherry-Pick Library tool.
When transfection conditions have been established for the screen, a pilot study testing the conditions in the screening facility can help to identify other variants that could affect the results (e.g.: issues with incubators, compatibility of plates and liquid handling robots, etc.).

Screening
During the screening phase, the most important factor is keeping track of the samples on the plate, especially if the screening is being done with the help of a liquid handling robot. For this, use a barcode system to ensure sample tracking is correct. Depending on how large the screen is, keeping the conditions stable (temperature, CO₂ level, volume of cell culture media, position of siRNA in the plate) could be challenging. By ensuring stable culture conditions, screeners will avoid bias in the result caused by cell death.

The primary screening phase will allow identification of hits or gene candidates for follow up studies; the process of hit identification and validation is described in detail below.

Hit identification
Common readouts of the primary screen include cell viability, protein phosphorylation, cell morphology, and protein detection assays done either by FACS or Western Blot analysis. Importantly, assays must be fine-tuned according to the study objectives with the goal of identifying cellular phenotypic changes. An example is the use of a mutant Ubiquitin-GFP U2OS cell line which allows investigation of the Ubiquitin pathway. In an siRNA screen, if the down regulation of gene transcripts affects the function of the ubiquitin pathway, this causes accumulation of GFP protein. GFP signals can be measured in a high-throughput manner (Figure 1).

Figure 1: Example of cell line model for high-throughput detection of phenotypic changes

After selection of the assay type, the next step is to determine the background signal in the wells, which helps to discriminate between artifacts and true hits.

Artifacts can be caused by factors like plate- or edge effects, evaporation, or simply by the location of the reagent in the well. Since artifacts can lead to misinterpretation of the results, the use of controls is fundamental.

siRNA controls are also used to calculate the Z’ factor as a quality measurement of an arrayed siRNA screen and thereby providing a level of reliability in the results. This helps to identify true specific siRNA hits. The Z´ factor is calculated by comparing the signals (standard derivations and means) obtained from the wells of the targeting- and non-targeting siRNA treated cells in the screen^2,3.

The next step to identify hits in the screen is to calculate the Z score for every siRNA tested. The Z score (based on the number of standard deviations from the mean) provides information on the strength of each siRNA relative to the rest of the sample distribution3. The desired Z score can be calculated depending on the level of confidence in the study.

A common parameter researchers establish is a level of confidence of 95%. In that case, the Z score value is 1.96.

The number of hits identified can be followed up for hit validation and confirmation in a secondary screen⁴.
Hit validation and stratification
As soon as the data has the desired quality, primary hits require further appropriate hit validation strategies.

Deconvolution
If hits in the primary screen were detected using SMARTpool siRNAs, the individual siRNAs in the SMARTpool can be used to deconvolute the results. Alternatively, a second SMARTpool from a different siRNA collection, comprising of different siRNA sequences, can be used for further hit validation. As the number of siRNAs can be very large in siRNA screens, researchers often set up thresholds in the deconvolution process to increase the likelihood of identifying true hits instead of false positives. The accompanying poster "siRNA screening: development of hit stratification strategies” offers deeper guidance.
Rescue Experiments
Rescue experiments are a standard technique to confirm specific hits, although using this method makes larger hit lists an extensive endeavor, due to potential redesigns of UTR-targeting siRNAs and the co-delivery with ORF constructs. Further, the overexpression of a protein from a plasmid bears the risk of complications, such as the accumulation of protein which may be disrupt normal cellular function when present in non-physiological amounts. In conclusion, rescue experiments are generally chosen for small hit lists (in depth validation). For longer hit lists, deconvolution is generally the preferred method.
Seed based controls (seed siblings)
Depending on the outcome and the defined threshold set for the deconvolution step, low confidence hits can be further investigated by seed sibling controls. A seed sibling is an siRNA that has no perfect match to transcripts in a species of interest, but whose seed region (nucleotides 2-7 on the antisense siRNA strand) is a perfect match to an experimental siRNA that gives a phenotype. Seed siblings help to identify if an observed phenotype, only detected with one SMARTpool and one individual siRNA, for example, is a seed-mediated off-target effect caused by microRNA-like silencing or a true hit. Seed sibling controls can be acquired via Dharmacon´s Cherry-Pick Custom Library Tool.
C911 controls
Further experimental methods for hit validation are C911 controls. The C911 control has the same sequence as the experimental siRNA used in the original screen, except the bases in positions 9, 10, and 11 are changed to their complement. The mismatches are predicted to disrupt on-target binding but preserve seed sequence-mediated off-targets, since the seed sequence remains intact. In contrast to seed sibling controls, only one C911 control can be designed for each original targeting siRNA. C911 controls can be created by the C911 Calculator designed by Buehler et al⁵and ordered via our custom siRNA tool.

Orthogonal validation
Strong siRNA data sets are supported by a broad spectrum of orthogonal validation strategies. We encourage customers to consider independent gene modulation and editing tools to confirm screen results by interrupting different cellular machineries. RNAi techniques, like siRNAs or shRNAs, degrade endogenous mRNA and thereby prevent translation. CRISPRmod CRISPRi reagents targeting specific endogenous locations in the transcription start site to block transcription without cutting the DNA. CRISPR knockout causes cuts at specific target locations on the DNA level, resulting in heritable genomic changes indel which lead to gene knockout. The combined usage of different validation strategies leads to high confidence datasets.

References

Frontiers in RNAi; A. Tripp, R., M. Karpilow, J., Eds.; BENTHAM SCIENCE PUBLISHERS, 2014. https://doi.org/10.2174/97816080594091140101.
Zhang, J.-H.; Chung, T. D. Y.; Oldenburg, K. R. A Simple Statistical Parameter for Use in Evaluation and Validation of High Throughput Screening Assays. SLAS Discovery 1999, 4 (2), 67–73. https://doi.org/10.1177/108705719900400206.
Birmingham A, Selfors LM, Forster T, Wrobel D, Kennedy CJ, Shanks E, Santoyo-Lopez J, Dunican DJ, Long A, Kelleher D, Smith Q, Beijersbergen RL, Ghazal P, Shamu CE. Statistical methods for analysis of high-throughput RNA interference screens. Nat Methods. 2009 Aug;6(8):569-75. doi: 10.1038/nmeth.1351.
Sfikas, A.; Banks, P.; Su, L.-I.; Schlossmacher, G.; Perkins, N. D.; Yemm, A. I. Parallel SiRNA Screens to Identify Kinase and Phosphatase Modulators of NF-ΚB Activity Following DNA Damage; preprint; Cell Biology, 2019. https://doi.org/10.1101/866061.
Buehler, E.; Chen, Y.-C.; Martin, S. C911: A Bench-Level Control for Sequence Specific SiRNA Off-Target Effects. PLoS ONE 2012, 7 (12), e51942. https://doi.org/10.1371/journal.pone.0051942.

siRNA screening basics: Setting up a successful arrayed siRNA screen

Pre-requirements
Pilot experiments should be considered before starting a robust screen.

References

Recommended reading (bioinformatic methods for identifying off-target effects)

siRNA screening basics: Setting up a successful arrayed siRNA screen

Pre-requirements Pilot experiments should be considered before starting a robust screen.

References

Recommended reading (bioinformatic methods for identifying off-target effects)

Pre-requirements
Pilot experiments should be considered before starting a robust screen.