CRISPR Library Screening Made Simple | Part 2: Performing Screens
Keywords: CRISPR library screen, in vitro and in vivo screens, antibiotic kill curve, lentivirus functional titration
CRISPR screening is a powerful technique for identifying key factors involved in biological pathways or cellular processes. In Part 1 of this CRISPR screening blog series, we discussed important considerations for library design and construction, including selecting an appropriate biological system, designing gRNA libraries, and library preparation.
In Part 2, we will discuss key considerations for setting up a successful screen and introduce different types of CRISPR screens, using real-world examples to demonstrate their capabilities and effectiveness. This article will set the stage for upcoming posts, which will cover post-screening data analysis and interpretation.
Key Considerations Before Screening
-
Establish the antibiotic kill curve
When performing a CRISPR screen in vitro, an antibiotic resistance marker is often included in the library to select for successfully transduced cells. However, different cell types can have varying intrinsic resistance to antibiotics. In order to ensure an effective screen, it is therefore critical to determine the optimal concentration of the selection agent by performing a kill curve assay (Figure 1).
Figure 1. Representative kill curve (mock data).
-
Determine the functional titer of the library
As discussed in the previous blog post, several approaches can be used to deliver gRNA libraries, but lentivirus remains the most popular choice for in vitro screens. Following lentivirus library construction, it is important to assess the functional titer in the specific target cells, because transduction efficiency may vary significantly across different cell types. There are different ways to determine the functional titer, and a solid understanding of the library structure is essential to choose the right method. For instance, VectorBuilder’s Whole-Genome Dual-gRNA Library is compatible with multiple approaches for measuring the functional titer, such as using the fluorescent protein, drug-resistant gene, or qPCR. Establishing the functional titer is crucial for selecting the proper MOI for screening, which directly affects the amount of material and reagents required and, more importantly, the quality of hit identification.
-
Calculating the required material to achieve a robust screen
Once the functional titer is determined, the number of cells and viral particles required for an in vitro screen can be calculated based on multiple factors, including the number of replicates, the desired library coverage, and the chosen MOI. All of these factors depend on screen design, downstream data analysis, and available resources. For example, while maintaining a single perturbation per cell is important, an MOI that is too low can significantly increase the number of cells and viral particles required, which may make the screen impractical to perform.
For indirect in vivo screens, unlike an in vitro screen where the required cell numbers can be directly calculated, effective coverage is strongly influenced by cell survival during engraftment, which can vary between models. Therefore, to estimate the number of animals required to achieve reliable library coverage, the cell engraftment rate must first be estimated.
In vivo delivery of a CRISPR library offers the advantage of perturbing cells within their native environment; however, direct delivery of a CRISPR library to animals can be technically challenging. Additionally, both the approach and efficiency of delivery depend heavily on the target organ and cell type. Unlike in vitro or indirect in vivo screens, where a low MOI can be used to achieve primarily single perturbations per cell, maintaining this level of control in vivo is difficult. Allowing multiple perturbations per cell, in contrast, can reduce the amount of material required.
Uncovering Biological Players with CRISPR Screening
After understanding the fundamentals of CRISPR screening, including selecting the appropriate biological system, designing and preparing the library, and carefully optimizing the experimental setup, such as optimizing antibiotic concentration, determining the functional titer of the library, and calculating the required starting material, the truly remarkable aspect of CRISPR screens lies in the biological questions they can uncover. Over the past decade, researchers have used this approach to interrogate gene function, regulatory pathways, and complex molecular networks. Below, we will discuss a few case studies from published work that demonstrate CRISPR screening in action.
-
In vitro screening
One of the early landmark studies demonstrating the power of CRISPR screening was reported by Ophir Shalem and colleagues in 2014. In this study, a genome-scale CRISPR–Cas9 knockout (GeCKO) library was delivered to both cancer and pluripotent cells in a pooled lentiviral format, enabling flexibility with both negative and positive selection screens. The GeCKO library encodes Cas9, a gRNA, and a puromycin resistance marker in the same vector, eliminating the need to first generate a Cas9-stable cell line (Figure 2A). The library targets constitutive 5′ exons, with an average of 3–4 gRNAs per gene.
In the negative selection screen, two human cell lines were transduced with the GeCKO library at a low MOI (0.3) to ensure that most cells receive a single perturbation. After puromycin selection to enrich transduced cells, the cell populations were cultured and analyzed by deep sequencing 14 days after transduction (Figure 2B). They found that gRNAs targeting essential genes became depleted, which allows for the identification of genes required for cell viability through gRNA depletion analysis.
Figure 2. GeCKO library and application for genome-scale negative selection screening. (A) Vector design of the library. (B) Workflow of the in vitro screening using the GeCKO library. Adapted and modified from Ophir Shalem et al., 2014, Science.
The same library was also used in a positive selection screen to identify genes associated with resistance to the BRAF inhibitor Vemurafenib in melanoma cells. After transduction and puromycin selection, cells were treated with the drug, and surviving cells were collected at multiple time points. Deep sequencing revealed gRNAs that were enriched following treatment, indicating genes whose loss promotes drug resistance. Two of the top hits had been previously reported, providing proof of concept, while the others represented potential new targets for modulation to increase successful treatment.
-
In vivo screening
Lentiviral libraries have been widely used for in vitro CRISPR screens because they enable scalable and reliable delivery of pooled libraries. However, this approach cannot be easily applied in vivo. To address this, Chunlong Xu and colleagues developed a strategy using the PB transposon system to deliver a genome-scale gRNA library into the mouse liver through hydrodynamic tail vein injection, facilitating stable integration of gRNA constructs into the genome.
The authors first optimized the library backbone containing a minimal gRNA expression cassette that facilitates insertion of multiple copies of the construct into the genome. Delivery efficiency was evaluated by co-injecting a PB transposon vector expressing EGFP. Following hydrodynamic injection, strong EGFP fluorescence was observed across the liver, demonstrating efficient delivery.
After establishing the system, the researchers injected a genome-scale CRISPR gRNA library together with a vector expressing the transposase, as well as additional vectors that sensitize the genetic background to accelerate tumor formation (Figure 3). 45 days after injection, liver tumors developed, and gRNAs present in each tumor were amplified and analyzed by deep sequencing. Some enriched gRNAs targeted genes that were well-known tumor suppressors, while others had not previously been implicated in mouse liver cancer, such as Cdkn2b. Further validation experiments demonstrated that loss of Cdkn2b strongly promotes liver tumorigenesis.
Figure 3. Schematic of the in vivo CRISPR screen using the PB transposon system. (A) Delivery of the PB-CRISPR library to the mouse liver. (B) Representative liver tumors obtained from the screen. Adapted and modified from Chunlong Xu et al., 2016, PNAS.
This study demonstrated that CRISPR screening can be extended from cell culture to live animals, allowing gene function to be interrogated directly in vivo. It highlights the power of in vivo CRISPR screens to uncover disease-relevant genes in their native physiological context.
From pooled screens in cultured cells to direct perturbation of large gene sets in animals, CRISPR screens allow scientists to systematically uncover genes involved in diverse cellular processes. While this approach has proven invaluable, care must be taken in order to collect reliable, reproducible data. By combining carefully designed libraries, optimized delivery methods, robust readouts, and thoughtful experimental setup, CRISPR screening provides an unbiased way to investigate biological functions and identify novel targets. In the next post of our series, we will discuss how to interpret screening data and maximize the insights obtained from a CRISPR screen.
References
