Two (or Three) sgRNAs Are Better than One

CRISPR-Cas9 technology has made high-throughput, genome-wide loss-of-function studies to determine the relationship between genotype and phenotype significantly more straightforward. Pooled CRISPR screens using single guide RNAs (sgRNAs) to target individual genetic loci are now commonly used to determine the genes or sequences that play a role in a particular biological function or processes. However, one limitation of this approach is that cellular phenotypes usually involve multiple genes, and screens using only a single sgRNA per cell could therefore be limited in biological relevance.

To better mimic the complex interactions involved, more studies are now employing multiplexed strategies, where multiple loci are simultaneously targeted for editing or transcriptional regulation in a single cell. This article looks at the development of multiplexed screening strategies and the potential application of this exciting addition to the CRISPR screening toolkit.

CRISPR-Cas9—a revolution in functional genomic screening

The premise behind a genotype to phenotype functional screen is simple—change the expression of a gene and observe what happens to the phenotype. The simplicity and efficiency of CRISPR technology makes it easier than ever to introduce targeted changes to genes for use in genome-wide experiments. All you need is an sgRNA to direct the Cas9 endonuclease to the gene of interest, where it will elicit a double-strand break at the target site, activating the non-homologous end joining cellular DNA repair pathway (NHEJ). If the DNA repair results in a frameshift mutation, this can knockout the target gene. As well as gene knockout (CRISPR-KO), CRISPR technology can be used to alter gene expression—with a modified version of Cas9, transcription can be inhibited (CRISPRi) or activated (CRISPRa), particularly useful when looking at drug sensitivities where a full knockout could hide the phenotype of interest.

Genome-wide screening experiments involve a library of sgRNA delivered by lentiviral vector at a titer that will result in no more than a single infection (and so a single sgRNA) per cell. In this way, a cell should only ever contain a single edit or modulation in a single gene. The random integration of the lentiviral cassette also provides each cell with a barcode, enabling the sgRNA responsible for the resultant phenotype to be easily identified by massively parallel sequencing and the ability to perform screens in a pooled format.

Multiplexed sgRNA—an evolution in high-throughput screening

In recent years, a plethora of studies have employed pooled library screening with CRISPR to identify genes that play a role in drug resistance or sensitivity, components of cellular pathways, or those that contribute to a particular disease physiology. But generally, phenotypes are not governed by a single gene—disease states or drug interactions are often complex, involving interconnected genetic networks. There is now a growing trend for a multiplexed approach for CRISPR screening—utilizing multiple guides to edit or transcriptionally regulate gene expression at numerous loci in parallel, enabling combinatorial screening. Pairs of sgRNA targeting different genes within a particular cellular pathway will be used, with each sgRNA pair labeled with a unique barcode. By sequencing the barcodes within each cell in the pooled screen, the targeted gene pair can be identified. By measuring the frequencies of sgRNA barcodes in response to different treatment conditions, the gene combinations involved in the phenotype of interest can be determined.

Here are some examples of how multiplexed approach has been used in high-throughput pooled screens:

Combinatorial screening for cancer and disease

Wong et al utilized this approach to develop Combi-GEM, a system in which barcoded combinatorial genetic libraries could be rapidly assembled and tracked with next-generation sequencing.¹ Using this approach, they created a library of 23,409 barcoded sgRNA combinations targeting 50 genes based on ovarian cancer drug targets to look at gene interactions involved in ovarian cancer cell growth. The study was able to identify those gene pairs whose dual knockout inhibited cancer cell growth—and subsequent drugging of those same pairs with small molecule inhibitors was found to exert synergistic antiproliferative effects, highlighting the potential value for target identification. The method has since been developed further to allow for the creation of three-way multiplexed sgRNA libraries. In a 2020 paper,² Combi-GEN v2.0 was used to identify double- and triple- drug combinations that inhibit cancer cell growth and protect against Parkinson’s disease associated toxicity, providing effective gene combinations to take forward into secondary screens.

Identifying regulatory regions with multiplexed CRISPR screening

There are over 1 million regulatory regions within the human genome, but many targets and relationships are unknown. A combinatorial sgRNA library using CRISPRi was used to develop Enhancer-i, enabling deactivation of multiple enhancers in parallel in order to identify which combinations of estrogen receptor alpha bound enhancers are required to regulate expression of estradiol-responsive genes.³ Gasperini et al went one step further to map enhancer combinations—by increasing the viral titer, each individual cell acquired a unique combination of perturbations, approximately 28 per cell. The effect of the combinatorial transcriptional repression was then profiled by RNA single-cell sequencing.⁴

Genome-wide deletion screening of long non-coding RNAs using paired sgRNA

Using CRISPR-based knockout technology to screen non-coding regions proves challenging as the introduction of an indel into these regions is unlikely to produce a loss-of-function phenotype. However, a multiplexed approach has been used to develop a deletion-based screening method for long non-coding RNAs (lncRNAs). Using paired sgRNA (pgRNAs), Zhu et al were able to generate a large fragment deletion in the non-coding region and so perform high-throughput screening of approximately 700 human lncRNAs, identifying those that play a role in cancer cell growth.⁵

Combinatorial screening with CRISPR-Cas12a

The vast majority of CRISPR experiments utilize Cas9, an endonuclease that is highly efficient for gene knockout. However, the nature of Cas9 sgRNAs, as well as the size of Cas9, makes for more complicated cloning strategies, potentially introducing bias during library development, which can decrease the efficiency of editing when used in a multiplex strategy. Another option is to leverage the intrinsic RNA processing capabilities of the Cas12a enzyme (formally known as Cpf-1), in order to generate multiple transcripts from a single array. But the editing efficiency of Cas9-12a has so far been sufficient for a positive screen—where the outcome is to identify those sgRNA that have been enriched following selection. A negative, or drop out, screen requires high editing efficiency due to the inherent noise in order to identify hits. In a 2020 Nature Communications paper, Gier et al⁶ optimized the Cas12a system to achieve sufficient gene editing efficiency for use in a dropout genetic screen and then used a multiplexed strategy to identify synthetic sick and lethal interactions in Mll-Af9 leukemia cells.

Higher throughput still—moving from cloning to synthesis

Multiplexed CRISPR-based screening equips researchers with a raft of additional capabilities for functional genomics—including the ability to excise promoters, lncRNAs and other non-coding elements, or to systematically look at genetic interactions. However, one barrier to entry is that incorporating multiple guides into a single vector using standard cloning methods is a labor-intensive process. Improvements in DNA synthesis have the potential to unblock this step, provided oligos can be synthesized at sufficient length without compromising sequence quality. One company working to address this is Twist Bioscience, a manufacturer of synthetic DNA. “We are now able to extend the length of our oligos available in our Oligo Pools to 300 nucleotides,” says Emily Leproust, CEO at Twist Bioscience, “Scientists can now easily fit two sgRNAs or even a Cas12a crRNA array into this expanded design space—meaning high order combinatorial screens can been designed programmatically without having to worry about downstream cloning issues.”

Intricate genetic networks often drive human disease and so new therapeutic strategies will need to target multiple pathways to increase treatment efficacy. CRISPR-based technologies have revolutionized high-throughput genomic screening, and the development of multiplexed strategies will offer an additional powerful tool in the understanding of these complex cellular pathways and the identification of new combinatorial drug treatments.

Key Takeaways

• CRISPR-based technologies are now an increasingly popular way to conduct high-throughput genome-wide genotype to phenotype screens
• Genome-wide sgRNA libraries enable efficient, specific, and robust changes to gene expression at high-throughput scale, from gene knockout to transcriptional regulation
• Targeting multiple sgRNA simultaneously with a multiplex approach enables combinatorial screening, where researchers can better mimic the complex genetic interactions that govern biological relationships
• Multiplexed CRISPR-based screens can be used to identify those gene interactions that drive human disease, identify synthetic sick or lethal interactions for drug discovery, or better understand the role non-coding elements play in cellular processes

References

1. Wong, Alan S L et al. "Multiplexed barcoded CRISPR-Cas9 screening enabled by CombiGEM". Proceedings of the National Academy of Sciences of the United States of America vol. 113,9 (2016): 2544-9. doi:10.1073/pnas.1517883113

2. Zhou, Peng et al. "A Three-Way Combinatorial CRISPR Screen for Analyzing Interactions among Druggable Targets". Cell reports vol. 32,6 (2020): 108020. doi:10.1016/j.celrep.2020.108020

3. Carleton, Julia B et al. "Multiplex Enhancer Interference Reveals Collaborative Control of Gene Regulation by Estrogen Receptor α-Bound Enhancers". Cell systems vol. 5,4 (2017): 333-344.e5. doi:10.1016/j.cels.2017.08.011

4. Gasperini, Molly et al. "A Genome-wide Framework for Mapping Gene Regulation via Cellular Genetic Screens". Cell vol. 176,1-2 (2019): 377-390.e19. doi:10.1016/j.cell.2018.11.029

5. Zhu, Shiyou et al. "Genome-scale deletion screening of human long non-coding RNAs using a paired-guide RNA CRISPR-Cas9 library". Nature biotechnology vol. 34,12 (2016): 1279-1286. doi:10.1038/nbt.3715

6. Gier, Rodrigo A et al. "https://www.nature.com/articles/s41467-020-17209-1". Nature communications vol. 11,1 3455. 13 Jul. 2020, doi:10.1038/s41467-020-17209-1