Secondary Logo

Journal Logo

Review Articles

Interrogating genome function using CRISPR tools: a narrative review

Ren, Xingjiea; Takagi, Maya Asamia; Shen, Yina,b,∗

Author Information
doi: 10.1097/JBR.0000000000000066
  • Open



For decades, genetic approaches including random mutagenesis, gene trapping, and gene targeting are widely used to uncover critical genes and molecular pathways in development and disease. However, the precise genome editing approach, gene targeting, through homologous recombination in mammalian cells is an extremely inefficient process.[1] Therefore, it was not feasible to perform large-scale genetic screening at the endogenous DNA sequence level in mammalian cells with precision. In recent years, the developments of next-generations genome editing techniques including zinc-finger nuclease,[2] transcription activator-like effector nuclease,[3,4] and clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated protein 9 (Cas9) system[5] have drastically increased the efficiency of precise genome editing by engineering a site-specific double strand DNA break (DSB). DSB induced DNA repair allows for the deletion or insertion of a DNA sequence to precisely edit the genome. Among these new methods, CRISPR/Cas9 system is arguably the most widely used genome engineering tool due to its simplicity in experimental design, high efficiency in editing, and capacity for multiplexing.

The CRISPR systems are bacterial adaptive immune systems that protect the cell against viral infection. These systems are classified into two classes: class 1 with multi-subunit effector complexes and class 2 with single-protein effector modules. There are six main types and cumulatively about 30 different subtype CRISPR systems between the two classes.[6] The first CRISPR-based gene-editing tool was developed from the class 2 type II CRISPR system using Streptococcus pyogenes Cas9.[5] Since then, the original CRISPR/Cas9 system has been simplified to two key components to achieve genome editing in mammalian cells: a single-guide RNA (sgRNA) and a Cas9 endonuclease. The sgRNA is a chimeric RNA with the CRISPR RNA (crRNA) and trans-activating crRNA duplexes.[7] The Cas9 protein recognizes and binds to the trans-activating crRNA, and the crRNA functions as a guide so that Cas9 can cut the DNA target sequence to generate DSBs.[5] The first 20 bp of the crRNA sequence can be programmed to direct Cas9 to specific genomic regions with a Protospacer Adjacent Motif (PAM) sequence.[5] The PAM sequence allows Cas9 to load and initiate sgRNA to bind and search for the correct target site in the genome. The Cas9 proteins from different organisms have sequence specificity for PAM sequences and can be leveraged to target a variety of genomic regions[8–14] (Table 1). The versatile CRISPR/Cas9 system and its derivatives have been utilized in many different genetics applications, and in many cell types and organisms.[5,15,16] Here, we will discuss the general applications of CRISPR/Cas9-based genetic tools for genome and epigenome editing in mammalian cells, followed up by how to use these tools for high-throughput genetic screening.

Table 1
Table 1:
Summary of Cas9 PAM sequences

Database search strategy

We performed a series of electronic searches in the PubMed database to prepare this review. For the content related to CRISPR/Cas9 based genome and epigenome editing, we used the following key terms: CRISPR, Cas9, CRISPR interference, CRISPR activation, base editing, prime editing, and CRISPR screen. We focused on publications from January 2012 to December 2019. For additional publications cited in the review, we searched with each specific keyword and did not filter for specific publication dates. The searched results were first screened by the title, abstract, and keywords. Finally, we went through the full text and extracted the relevant information from each publication.

CRISPR/Cas9-based genome and epigenome editing tools and their application in mammalian cells

CRISPR/Cas9-based genome editing results in irreversible changes in DNA sequences in two ways, non-homologous end joining and homology-directed repair (HDR), during DSB repair (Fig. 1A and Table 2). The non-homologous end joining pathway creates indel mutations or small deletions, which can cause a frame shift and disrupt the function of the gene.[17] In contrast, the HDR pathway leads to a specific sequence change in the presence of a donor DNA with homologous sequences flanking the DSB site.[18] CRISPR/Cas9-based genome editing has been leveraged to edit the mammalian genome, achieving over 90% indel frequency[19] and about 30% HDR frequency.[20]

Figure 1
Figure 1:
CRISPR/Cas9 system based genetic tools. (A) CRISPR nuclease. Wild type Cas9 induces double strand breaks at target sites. These double strand breaks can produce either precise editing through homology directed repair or indels through the non-homologous end joining pathway. (B) CRISPR base editing. (i) Fusing a cytidine deaminase protein to dCas9 enables a specific base change without inducing a double strand break. Cytosine is converted into a uridine and the C-G pair changes to a T-A pair through the BER pathway. (ii) Fusing an adenosine deaminase protein to dCas9 converts a specific adenosine base to inosine. The cell recognizes the inosine as a guanosine during the BER pathway. (C) CRISPR prime editing. Fusing Cas9 nickase to an engineered reverse transcriptase programmed with a prime editing guide RNA both targets a specific site and encodes for the wanted edit. (D) CRISPR interference. (i) The sgRNA can be designed to the transcription start site or the gene exonic sequence to sterically repress transcription initiation or block transcriptional elongation, respectively, to repress transcription. (ii) Fusing a repressor domain to dCas9 represses transcription by inducing heterochromatization. (E) CRISPR activation. (i) Fusing an activation domain to dCas9 recruits important cofactors to indirectly increase transcription. (ii) In contrast, fusing p300 catalyzes the acetylation of H3K27 to directly increase transcription. BER = base excision repair, Cas9 = clustered regularly interspaced short palindromic repeat associated protein 9, CRISPR = clustered regularly interspaced short palindromic repeat, dCas9 = dead clustered regularly interspaced short palindromic repeat associated protein 9, DNMT3a = DNA (cytosine-5)-methyltransferase 3A, DSB = double strand DNA break, H3K27 = methylation of histone H3 on lysine 27, HDR = homology-directed repair, KRAB = Krüppel-associated box, MeCP2 = methyl CpG binding protein 2, NHEJ = non-homologous end joining, PAM = Protospacer Adjacent Motif, pegRNA = prime editing guide RNA, sgRNA = single-guide RNA, TF = transcriptional factor.
Table 2
Table 2:
Summary and comparison of CRISPR/Cas9 system based genetic tools

Point mutations in the Cas9 endonuclease domains, HNH and RuvC, can inactivate Cas9 endonuclease activity and change the performance of CRISPR/Cas9-based genome editing. For instance, Cas9 nickase consists of a single inactive endonuclease domain and creates a single strand break instead of a DSB. Moreover, catalytically dead Cas9 (dCas9) consists of inactivity in both endonuclease domains and loses its cutting ability, but maintains programmed targeting.

CRISPR/Cas9 mediated HDR allows us to precisely edit the mammalian genome. However, unintended indels cause lower precise editing efficiency and leave a scar at the target site. To achieve precise genome editing at the base pair resolution, CRISPR base editing (CRISPRb) was developed to convert one nucleic acid to another without the DSB and donor template (Table 2). There are two types of base editors: cytosine base editors or adenosine base editors. These systems target specific bases by fusing different DNA modification enzymes to dCas9 or Cas9 nickase. Cytidine deaminase, such as rat cytidine deaminase APOBEC1, activation-induced cytidine deaminase, and the sea lamprey PmCDA1 gene, converts cytosine bases into uridines, and results in the conversion of a C-G pairing into a T-A[21,22] (Fig. 1Bi). This cytosine base editor, on average, reaches efficiencies of 35% in mammalian cell cultures, and an upwards of 75% depending on the sequence.[21] DNA adenosine deaminase, an evolved Escherichia coli transfer RNA adenosine deaminase TadA, converts adenosine into inosine which is treated like guanosine by the cell (Fig. 1Bii). Therefore, this enzyme can convert an A-T pairing into a G-C.[23] The efficiency of an adenosine deaminase base editor is about 50% in human cells.[23] Additional proteins added into base editors can increase editing efficiency and specificity. For example, bacteriophage Mu-derived Gam protein, a nuclease inhibitor, limits indel formation.[24] Additionally, uracil-DNA glycosylase inhibitor subverts the cellular uracil base excision repair pathway to improve base-editing efficiency and reduce indel formation.[21] Taking advantage of base pair resolution, CRISPRb has been used in broad applications. For example, CRISPRb identifies known and novel mutations in directed gene evolution to create and improve protein function via changing the DNA codon for an amino acid.[22] CRISPRb can also create or correct early stop codons to inactivate or correct target genes.[25]

However, the CRISPRb tool is only able to convert four transition mutations (C→T, G→A, A→G, T→C) and, thus, cannot generate the remaining eight transversion mutations (C→A, C→G, G→C, G→T, A→C, A→T, T→A, T→G). The recently developed prime editing technology fills this gap. Prime editing uses a reverse transcriptase fused to Cas9 nickase and a prime editing guide RNA containing genetic modification information to copy genetic information from the prime editing guide RNA into the target site without generating DSBs (Fig. 1C and Table 2).[26] The prime editing efficiency can be improved with an engineered reverse transcriptase and nicking of the non-editing strand. Compared with base editing, prime editing can directly correct pathogenic transversions. For example, prime editing directly reverted the β-globin coding gene HBB mutation (AT-to-TA transversion, the primary genetic causes of sickle cell disease) to the wild type HBB with 26–52% efficiency in HEK293T cells. Furthermore, the prime editor can generate targeted insertions and deletions without DSBs and donor DNA templates.

In addition to modifying DNA sequences, the CRISPR/Cas9 system can be used to modify the epigenome. Epigenetic status can be directly modified with dCas9 fused epigenetic modifiers, such as histone demethylation by LSD1, cytosine methylation by DNMT3A or MQ1, and cytosine demethylation by Tet1.[27,28] Furthermore, CRISPR interference (CRISPRi) and CRISPR activation (CRISPRa) approaches can modify the chromatin status to directly or indirectly modulate transcription (Table 2).

CRISPRi can be carried out either with or without a repressor domain. Without an effector domain, the dCas9 itself shows the ability of decreasing or inhibiting transcription (Fig. 1Di).[29] When the dCas9-sgRNA complex binds to the promoter region of the target gene, the complex blocks RNA polymerase binding and, therefore, interferes with transcription initiation. Alternatively, when the dCas9-sgRNA complex binds downstream of the transcription start site of the target gene, the complex blocks RNA polymerase from moving forward and prevents transcription elongation. CRISPRi also refers to when dCas9 is fused with a repressor domain, such as the Krüppel-associated box (KRAB) domain (Fig. 1Dii). The KRAB domain recruits the KAP1/TIF1β corepressor complex. Subsequently, KAP1 acts as a scaffold to further recruit histone deacetylases, histone methyltransferases, and heterochromatin protein 1 isoforms to induce heterochromatinization at the target region for gene silencing.[30,31] The dCas9-KRAB fusion inhibits gene expression about 2 to 5 fold greater than dCas9 alone.[32] Other effector domains, such as MeCP2, SIN3A, HDT1, MBD2B, NIPP1, and HP1A, which are associated with transcriptional regulation and gene silencing, can be fused to dCas9 to achieve stronger repression (up to 8-fold more than dCas9 alone).[33] Bipartite fusion domains increase repression further (between 5-fold to 60-fold more than dCas9 alone).[33] For example, the dCas9-KRAB-MeCP2 outperformed other bipartite and tripartite repressors when tested in various endogenous target genes.[33]

In the case of CRISPRa, two types of domains are fused to dCas9 to achieve genome activation. Type I domains scaffold transcriptional factors and chromatin remodeling complexes associated with active transcription to indirectly activate transcription (Fig. 1Ei).[34] For example, VP64 indirectly activate transcription by recruiting PC4,[35] CBP/p300,[36] and the SWI/SNF complex.[37] Type II domains are usually a chromatin remodeling or modification enzyme that activate transcription. For example, the core p300 domain catalyzes acetylation of histone H3 lysine 27 at its target sites, which leads to robust transcriptional activation of target genes or distal enhancer regions (Fig. 1Eii).[38] In general, multiple domain fusions have higher transcriptional activation efficiency than single fusions. For example, dCas9-VP64-p65-Rta achieved a 22-fold to 320-fold greater activation compared to dCas9-VP64.[39] Therefore, several strategies have been developed to incorporate multiple activation domains including fusing a series of activation domains directly to dCas9, fusing a protein scaffold to dCas9, such as Suntag, to recruit activators, and fusing an activation domain to dCas9 and an RNA scaffold to sgRNA to recruit activation domains, as illustrated in the synergistic activation mediator system.[40,41]

CRISPR/Cas9 system mediated high throughput genetic screening

Genetics screens have led to the discovery of many genes related to biological processes. Previously, transposon mediated insertional mutagenesis and RNA interference mediated gene knockdowns were two major methods for high throughput functional screens in mammalian cells. However, transposon insertion mediated screens are limited to the transposon recognition sites,[42] and the RNA interference mediated approach can only lead to knockdowns with the disadvantage of extensive off-target effects.[43] CRISPR/Cas9 system mediated high throughput screens can address these limitations, and various screen methods have been developed including knock-out,[44] knock-in,[45] transcriptional repression,[46] transcriptional activation,[41,47] and base editing.[48] Both CRISPR mediated genome editing and epigenome editing can be applied to high throughput screening. The genome editing allows for the change or deletion of the DNA sequence in the genome to knock-out protein coding genes, noncoding RNAs (ncRNAs), or regulatory elements. In contrast, the epigenome editing is able to repress or activate transcription without changing the DNA sequence in mammalian cells. In addition, the base editing system pushes the screening resolution to the single base pair level. All these screening strategies can be carried out in two ways: pooled screening and single cell screening. These approaches have been applied to identify genes essential for cell viability,[44,49,50] cell fate specification and reprogramming,[47] drug resistance,[41] key components in biological pathways in response to drug treatment,[44,49] regulatory sequences in controlling gene regulation,[51] and critical amino acids for protein function.[48] Here, we will describe the general workflow for CRISPR/Cas9-mediated high throughput screening.

sgRNA library design and cloning

The first step in performing a CRISPR-based high throughput screening is the generation of a sgRNA library. Many bioinformatics CRISPR tools have been developed for designing sgRNA for various purposes and different species (Table 3). The quality of sgRNAs is typically indicated by a score, which are determined by two criteria: efficiency and specificity. Efficiency shows the likelihood of sgRNA mediated cutting, which positively associates with sgRNA GC content (the ideal GC range is 40%–80%),[49] sgRNA stability,[52] and DNA accessibility of the target site.[52] Specificity accounts for the off-target sites binding of sgRNA, which is controlled by sequence similarity between the on-target site and the off-target sites. Genome-wide profiling of off-target cleavage suggests that almost all predicted mismatches of four or more in off-target sites are not cleaved by Cas9.[53] Therefore, these kinds of predicted off-target sites do not have to be considered in sgRNA library design. In addition, two different kinds of controls, positive and negative, should be added into the sgRNA screening library in order to minimize the false positive and false negative rates of screening during the analysis. The positive control sgRNAs, typically targeting genomic regions, can affect phenotypes related to the screen such as sgRNAs against reporter genes and previously validated sgRNAs targeting essential genes linked to the screening phenotype. In contrast, negative controls can be designed to target either other genomes or genomic regions that are not related to the screening phenotype. There are published sgRNA libraries available online for researchers, which include genome-wide targeting libraries or libraries for specific purposes ( These existing libraries can be used directly for screening, but the drawbacks are that no additional modifications can be made to them directly and they may not be fit for the researcher's purpose. Usually, the libraries are constructed using pooled oligos containing designed sgRNA sequences (Fig. 2A). sgRNA libraries can be cloned into plasmids expressing either single sgRNA or dual sgRNA. For a single sgRNA library, the sgRNA pool is directly inserted into vectors via Gibson assembly or ligation. Usually, single sgRNA libraries are used to create indel mutations, or activate or inhibit transcription. In the case of a dual sgRNA library, sgRNA pairs with linkers are cloned into a sgRNA expression vector. Then, another sgRNA scaffold and U6 promoter are inserted between the sgRNAs pair to replace the linkers. Dual sgRNA expression libraries have been used for deletion screens.[51,54] Additionally, a library of sgRNA expression cassettes can be first cloned into the vector, then a library of the second sgRNA expression cassettes inserts into the vectors which already have the first sgRNA expressing cassettes. This method generates random combinations of each designed sgRNA[55] for combinatorial screens for analyzing genetic interactions.

Table 3
Table 3:
Bioinformatic tools for CRISPR/Cas9 system sgRNA design
Figure 2
Figure 2:
Experiment overview for performing high throughput screening with CRISPR/Cas9 system. (A) sgRNA library design, synthesis, and cloning. Libraries can target various regions and can be either a single or dual sgRNA library. The sgRNA oligos are synthesized in pool. A single sgRNA library can be cloned via ligation or Gibson assembly. A dual sgRNA library can be cloned through two different approaches. First, the sgRNA library is inserted into a cloning vector to build the “intermediate step” sgRNA library. Then, the second sgRNA expression cassette is inserted into the “intermediate step” library to achieve the final random combinational dual sgRNA library. Second, sgRNA pairs with removable linkers are cloned into a sgRNA expression vector. Then, the linkers between sgRNA pairs are replaced with another sgRNA scaffold and a U6 promoter. (B) Library delivery. The libraries can be delivered efficiently to cells through viral delivery or transfection delivery. (C) Screening. The transduced cells are analyzed either via positive screening or negative screening, which rely on the enrichment or depletion of sgRNA respectively. Cas9 = clustered regularly interspaced short palindromic repeat associated protein 9, CRISPR = clustered regularly interspaced short palindromic repeat, sgRNA = single-guide RNA.

sgRNA library delivery

Once the designed sgRNA oligos are cloned into a desired vector, the next step is to deliver the CRISPR/Cas9 system into cells (Fig. 2B). The most common system for CRISPR/Cas9 delivery is viral delivery systems, including adeno-associated virus, lentivirus (LV) and adenovirus.[56] Adeno-associated virus is significantly less immunogenic than other viruses, but the packaging capacity is only 4.7 kb.[56] Therefore, Cas9 and sgRNA have to be delivered separately. In contrast, LV and adenovirus allow up to 8 kb and 35 kb packaging capacity, respectively, allowing Cas9 and sgRNA to be included within one vector.[56] All of these viruses infect both dividing and non-dividing cells, but only LV integrates into the host genome efficiently. Adenovirus does not integrate into the host genome, and adeno-associated virus is mainly episomal.[56] In comparison, LV can integrate into the host genome and result in stable and long-term expression of Cas9 and sgRNA.[56] Thus, the readout of insertional sgRNA sequences in the infected cells can be detected by sequencing in later steps, making LV the most popular delivery method for high-throughput screening.

In addition to these viral delivery systems, transfection is another method for CRISPR/Cas9 system delivery. The transfection method can deliver the sgRNA library into the cell without integration into the genome.[57,58] For example, a Cas9/sgRNA library and the HDR library were co-transfected in a multiplex HDR screening. Coupling CRISPR/Cas9 mediated DSBs with multiplex HDR demonstrated a saturation of editing in genomic regions.[45] Additionally, the transfection method can insert a sgRNA library into the genome by coupling it with another integration system, such as the transposon system. piggyBac or Sleeping Beauty transposon vectors have a greater capacity for larger insertions and have the ability of integration into the host genome. Co-transfection of transposon based sgRNA vectors and transposes were used to deliver CRISPR/Cas9 system into the cell.[57,58] One potential drawback of the transposon system is that the efficiency of transfection is lower than transduction in certain cell types.[59]

High throughput screening

In pooled library screens, cells are infected at a very low multiplicity of infection to ensure each cell receives no more than one viral particle. To avoid dropouts during the screening, the screening library should infect cells at a high coverage (eg, 1000×). The screening can be divided into two categories, positive screening and negative screening (Fig. 2C). Positive screens are used to identify the cells which pass the selection criteria, such as cell survival or expression of reporter genes. In contrast, negative screens identify the cells which do not pass the selection criteria. After the delivery of screening library, a small portion of the infected cells are set aside to serve as a non-selected control, while the majority of the cells continue to be used for selection. In the end, the sgRNAs in these two sets of cells are polymerase chain reaction amplified and sequenced via next generation sequencing to determine which sgRNAs are enriched or depleted during selection. The enrichment or depletion of sgRNAs can be used to infer which genomic loci are essential for selection conditions. Positive screens have been used to identify genes involved in drug resistance,[44] reveal host genes essential for Zika virus replication,[60] and discover cis-regulatory elements.[51] Meanwhile, negative screenings can identify essential genes for cell survival and regulatory elements which contribute to cell proliferation.[44,61]

Study coding regions with CRISPR/Cas9-mediated high throughput screening

The first genome-wide coding gene CRISPR/Cas9 screening in human cells targeted 18,080 protein coding genes and identified genes necessary for cell resistance to the treatment of vemurafenib, a therapeutic RAF inhibitor for melanoma.[44] The reproducibility of identifying same genes with independent sgRNA targeting demonstrate the feasibility of performing genome-wide screening using CRISPR/Cas9 system.[44] In addition to directly disrupting coding genes with random indels by sgRNA, CRISPRb enables the identification of functional domains in protein at a single amino acid resolution. Specifically, base mutations using CRISPRb can create targeted missense mutations and, thus, allow for the identification of crucial amino acids of the proteins of interest. For instance, a CRISPRb-based screening was carried out in mice for Dnd1, a critical gene for primordial germ cell development. The screen identified four amino acids that are essential for Dnd1 protein function. Missense mutations in these four amino acids can result in Dnd1 instability and thus abolish protein-protein interaction, and leads to the complete loss of Dnd1 in primordial germ cells.[48] In parallel, epigenome screening can be used to identify genetic components important for biological processes by gene activation (CRISPRa) or repression (CRISPRi). For example, a CRISPRa screening targeting 2428 transcriptional factors and DNA binding proteins systematically identified 74 regulatory factors that positively contribute to neuronal cell fate specification.[47] In another example, a genome wide CRISPRi screening targeting 15,977 human protein-coding genes identified genes essential for cell growth in K562 cells. The top hits from the screening were the genes involved in essential cellular processes, such as translation, transcription, and DNA replication.[62]

Study noncoding regions with CRISPR/Cas9-mediated high throughput screening

98% of the human genome is noncoding sequences including regulatory elements, repeated elements, and ncRNAs. More than 90% of diseases or trait-associated variants identified by the genome-wide association studies are concentrated in noncoding regions.[63,64] However, the targets of noncoding regulatory elements and the precise function of ncRNAs are, for the most part, still unknown. CRISPR/Cas9 system based genetic tools have proven to be a valuable tool for annotating the noncoding region.


An ncRNA is an RNA molecule that is transcribed from genomic DNA, but not translated into a protein. Unlike protein coding genes, point mutation, small deletion, or insertion are less likely to disrupt the function of ncRNA, especially long ncRNAs (lncRNAs). For functional ablation of lncRNA, CRISPRa, CRISPRi, and CRISPR-mediated deletions have been used to study the biological function of lncRNAs with high throughput screening.[50,65,66] In addition, exon skipping or intron retention generated by targeting splice donor or splice acceptor sites with CRISPR/Cas9 system is another approach to interfere lncRNA function. A genome-scale splicing-targeted screen covering 10,996 lncRNAs was performed in chronic myeloid leukemia K562 cells and identified 230 lncRNAs that are essential for cellular growth of K562 cells.[67] The same sgRNA library was applied in the lymphoblastoid cell line GM12878 and HeLa cell line identified 220 and 115 lncRNAs, respectively, affecting cell growth and proliferation.[67] These screens illustrate that the approach of targeting splicing works in multiple cell lines and can be utilized to discover the diverse functions of lncRNAs.

Regulatory elements

Regulatory elements control spatial and temporal gene expression, which are crucial for organism growth, development and disease. Multiple functional assays, such as massively parallel reporter assays[68] and self-transcribing active regulatory region sequencing,[69] have been developed to test putative regulatory elements. However, these reporter based assays use heterogeneous promoters and cannot test DNA sequences function in its native chromatin content. To address these limitations, CRISPR/Cas9 approaches can also be applied for characterizing regulatory elements in a high throughput fashion. For example, single sgRNA libraries can be used for testing putative regulatory elements mapped by biochemical signatures with high resolution, and can even identify motifs and transcriptional factors binding sites.[70–72] On the other hand, dual-sgRNA libraries can be used to tile and delete large genomic intervals with high coverage. One example of this application is called CREST-seq, which can screen for genomic regions over megabase pairs in an unbiased manner.[51] In addition, CRISPRi can also be applied for the discovery of regulatory elements.[61,73,74]

CRISPR/Cas9-mediated high throughput single cell screening

More recently, methods have developed to combine pooled CRISPR screenings with single-cell RNA sequencing technologies, which enable reading pooled CRISPR screens at the single-cell transcriptome resolution. Methods, including CROP-seq,[75] Perturb-seq,[76,77] CRISP-seq,[78] Mosaic-seq,[73] and crisprQTL,[74] can facilitate functional dissection of complex biological mechanisms in heterogeneous cell populations. Typically, sgRNAs are transcribed from a human or mouse U6 promoter by RNA polymerase III. Therefore, the sgRNA transcripts lacking a polyadenylated tail cannot be detected by polyadenylated enrichment during single-cell RNA sequencing. The CROP-seq construct makes the sgRNA part of the puromycin polyadenylated mRNA transcript by inserting a hU6-sgRNA cassette into a 3′ long terminal repeat so that it can be detectable by the polyadenylated enrichment RNA-seq approach. Perturb-seq and CRISP-seq use guide barcodes to detect sgRNA within individual cells. Compared to different sgRNA barcoding strategies, CROP-seq is only suitable for single perturbation in individual cells, whereas Perturb-seq and CRISP-seq allow inductions of multiple perturbations within the same cell. Therefore, Perturb-seq and CRISP-seq are useful for systematic identification of complex genetic interactions with single-cell resolution. In terms of testing non-coding regulatory elements, Mosaic-seq and crisprQTL can facilitate large-scale examination of regulatory element activity by introducing random combinations of CRISPRi mediated perturbations in each cell, followed by single cell RNA-seq.[73,74]

Future direction

The CRISPR/Cas9 genome editing tools have significantly improved our ability to manipulate the genome and have revolutionized both basic and translational research. The combination of single cell technology and CRISPR system-mediated high-throughput screening in primary cells or tissue will provide an efficient approach to decoding genetic mechanisms and to develop new treatment for human diseases. In addition, since many players are involved in biological processes and work in a network with one another, expanding the current analysis to combinational screens for genetics interactions between different loci will reveal novel biological insights in human health and diseases.

One other promising application of CRISPR is therapeutic editing in disease-relevant somatic cells. There are already several preliminary examples of therapeutic interventions in model organisms and cell culture systems. For example, haploinsufficiency induced obesity can be rescued by CRISPRa mediated up-regulation of Sim1 functional allele in a Sim1 and Mc4r heterozygous mouse model.[79] In another study, dCas9-Tet1 mediated demethylation of CGG repeats upstream of FMR1 promoter in fragile X syndrome iPSCs and neurons induced local chromatin state change and restored FMR1 expression.[80] These examples are just the beginning towards leveraging CRISPR for developing therapeutic approaches for human diseases.



Author contributions

XR and MAT wrote the manuscript and prepared the figures. YS revised the manuscript. All authors reviewed and approved the final version of the manuscript.

Financial support


Conflicts of interest

The authors declare that they have no conflicts of interest.


[1]. Capecchi MR. Gene targeting in mice: functional analysis of the mammalian genome for the twenty-first century. Nat Rev Genet 2005;6:507–512.
[2]. Bibikova M, Golic M, Golic KG, et al. Targeted chromosomal cleavage and mutagenesis in Drosophila using zinc-finger nucleases. Genetics 2002;161:1169–1175.
[3]. Hockemeyer D, Wang H, Kiani S, et al. Genetic engineering of human pluripotent cells using TALE nucleases. Nat Biotechnol 2011;29:731–734.
[4]. Miller JC, Tan S, Qiao G, et al. A TALE nuclease architecture for efficient genome editing. Nat Biotechnol 2011;29:143–148.
[5]. Cong L, Ran FA, Cox D, et al. Multiplex genome engineering using CRISPR/Cas systems. Science 2013;339:819–823.
[6]. Koonin EV, Makarova KS, Zhang F. Diversity, classification and evolution of CRISPR-Cas systems. Curr Opin Microbiol 2017;37:67–78.
[7]. Jiang F, Doudna JA. CRISPR-Cas9 structures and mechanisms. Annu Rev Biophys 2017;46:505–529.
[8]. Hsu PD, Scott DA, Weinstein JA, et al. DNA targeting specificity of RNA-guided Cas9 nucleases. Nat Biotechnol 2013;31:827–832.
[9]. Kleinstiver BP, Prew MS, Tsai SQ, et al. Engineered CRISPR-Cas9 nucleases with altered PAM specificities. Nature 2015;523:481–485.
[10]. Ran FA, Cong L, Yan WX, et al. In vivo genome editing using Staphylococcus aureus Cas9. Nature 2015;520:186–191.
[11]. Amrani N, Gao XD, Liu P, et al. NmeCas9 is an intrinsically high-fidelity genome-editing platform. Genome Biol 2018;19:214.
[12]. Kim E, Koo T, Park SW, et al. In vivo genome editing with a small Cas9 orthologue derived from Campylobacter jejuni. Nat Commun 2017;8:14500.
[13]. Müller M, Lee CM, Gasiunas G, et al. Streptococcus thermophilus CRISPR-Cas9 systems enable specific editing of the human genome. Mol Ther 2016;24:636–644.
[14]. Acharya S, Mishra A, Paul D, et al. Francisella novicida Cas9 interrogates genomic DNA with very high specificity and can be used for mammalian genome editing. Proc Natl Acad Sci U S A 2019;116:20959–20968.
[15]. Hruscha A, Krawitz P, Rechenberg A, et al. Efficient CRISPR/Cas9 genome editing with low off-target effects in zebrafish. Development 2013;140:4982–4987.
[16]. Ren X, Sun J, Housden BE, et al. Optimized gene editing technology for Drosophila melanogaster using germ line-specific Cas9. Proc Natl Acad Sci U S A 2013;110:19012–19017.
[17]. Chiruvella KK, Liang Z, Wilson TE. Repair of double-strand breaks by end joining. Cold Spring Harb Perspect Biol 2013;5:a012757.
[18]. Jasin M, Rothstein R. Repair of strand breaks by homologous recombination. Cold Spring Harb Perspect Biol 2013;5:a012740.
[19]. Hendel A, Bak RO, Clark JT, et al. Chemically modified guide RNAs enhance CRISPR-Cas genome editing in human primary cells. Nat Biotechnol 2015;33:985–989.
[20]. Yu C, Liu Y, Ma T, et al. Small molecules enhance CRISPR genome editing in pluripotent stem cells. Cell Stem Cell 2015;16:142–147.
[21]. Komor AC, Kim YB, Packer MS, et al. Programmable editing of a target base in genomic DNA without double-stranded DNA cleavage. Nature 2016;533:420–424.
[22]. Ma Y, Zhang J, Yin W, et al. Targeted AID-mediated mutagenesis (TAM) enables efficient genomic diversification in mammalian cells. Nat Methods 2016;13:1029–1035.
[23]. Gaudelli NM, Komor AC, Rees HA, et al. Programmable base editing of A•T to G•C in genomic DNA without DNA cleavage. Nature 2017;551:464–471.
[24]. Komor AC, Zhao KT, Packer MS, et al. Improved base excision repair inhibition and bacteriophage Mu Gam protein yields C:G-to-T:A base editors with higher efficiency and product purity. Science Adv 2017;3:eaao4774.
[25]. Billon P, Bryant EE, Joseph SA, et al. CRISPR-mediated base editing enables efficient disruption of eukaryotic genes through induction of STOP codons. Mol Cell 2017;67:1068–1079.e4.
[26]. Anzalone AV, Randolph PB, Davis JR, et al. Search-and-replace genome editing without double-strand breaks or donor DNA. Nature 2019;576:149–157.
[27]. Xiong X, Chen M, Lim WA, et al. CRISPR/Cas9 for human genome engineering and disease research. Annu Rev Genomics Hum Genet 2016;17:131–154.
[28]. Pickar-Oliver A, Gersbach CA. The next generation of CRISPR-Cas technologies and applications. Nat Rev Mol Cell Biol 2019;20:490–507.
[29]. Larson MH, Gilbert LA, Wang X, et al. CRISPR interference (CRISPRi) for sequence-specific control of gene expression. Nat Protoc 2013;8:2180–2196.
[30]. Urrutia R. KRAB-containing zinc-finger repressor proteins. Genome Biol 2003;4:231.
[31]. Friedman JR, Fredericks WJ, Jensen DE, et al. KAP-1, a novel corepressor for the highly conserved KRAB repression domain. Genes Dev 1996;10:2067–2078.
[32]. Gilbert LA, Larson MH, Morsut L, et al. CRISPR-mediated modular RNA-guided regulation of transcription in eukaryotes. Cell 2013;154:442–451.
[33]. Yeo NC, Chavez A, Lance-Byrne A, et al. An enhanced CRISPR repressor for targeted mammalian gene regulation. Nat Methods 2018;15:611–616.
[34]. Perez-Pinera P, Kocak DD, Vockley CM, et al. RNA-guided gene activation by CRISPR-Cas9-based transcription factors. Nat Methods 2013;10:973–976.
[35]. Kaiser K, Stelzer G, Meisterernst M. The coactivator p15 (PC4) initiates transcriptional activation during TFIIA-TFIID-promoter complex formation. EMBO J 1995;14:3520–3527.
[36]. Ikeda K, Stuehler T, Meisterernst M. The H1 and H2 regions of the activation domain of herpes simplex virion protein 16 stimulate transcription through distinct molecular mechanisms. Genes Cells 2002;7:49–58.
[37]. Neely KE, Hassan AH, Wallberg AE, et al. Activation domain-mediated targeting of the SWI/SNF complex to promoters stimulates transcription from nucleosome arrays. Mol Cell 1999;4:649–655.
[38]. Chen M, Qi LS. Repurposing CRISPR system for transcriptional activation. Adv Exp Med Biol 2017;983:147–157.
[39]. Chavez A, Scheiman J, Vora S, et al. Highly efficient Cas9-mediated transcriptional programming. Nat Methods 2015;12:326–328.
[40]. Tanenbaum ME, Gilbert LA, Qi LS, et al. A protein-tagging system for signal amplification in gene expression and fluorescence imaging. Cell 2014;159:635–646.
[41]. Konermann S, Brigham MD, Trevino AE, et al. Genome-scale transcriptional activation by an engineered CRISPR-Cas9 complex. Nature 2015;517:583–588.
[42]. Kawakami K, Largaespada DA, Ivics Z. Transposons as tools for functional genomics in vertebrate models. Trends Genet 2017;33:784–801.
[43]. McManus MT, Sharp PA. Gene silencing in mammals by small interfering RNAs. Nat Rev Genet 2002;3:737–747.
[44]. Shalem O, Sanjana NE, Hartenian E, et al. Genome-scale CRISPR-Cas9 knockout screening in human cells. Science 2014;343:84–87.
[45]. Findlay GM, Boyle EA, Hause RJ, et al. Saturation editing of genomic regions by multiplex homology-directed repair. Nature 2014;513:120–123.
[46]. Du D, Roguev A, Gordon DE, et al. Genetic interaction mapping in mammalian cells using CRISPR interference. Nat Methods 2017;14:577–580.
[47]. Liu Y, Yu C, Daley TP, et al. CRISPR activation screens systematically identify factors that drive neuronal fate and reprogramming. Cell Stem Cell 2018;23:758–771.e8.
[48]. Li Q, Li Y, Yang S, et al. CRISPR-Cas9-mediated base-editing screening in mice identifies DND1 amino acids that are critical for primordial germ cell development. Nat Cell Biol 2018;20:1315–1325.
[49]. Wang T, Wei JJ, Sabatini DM, et al. Genetic screens in human cells using the CRISPR-Cas9 system. Science 2014;343:80–84.
[50]. Liu SJ, Horlbeck MA, Cho SW, et al. CRISPRi-based genome-scale identification of functional long noncoding RNA loci in human cells. Science 2017;355:aah7111.
[51]. Diao Y, Fang R, Li B, et al. A tiling-deletion-based genetic screen for cis-regulatory element identification in mammalian cells. Nat Methods 2017;14:629–635.
[52]. Moreno-Mateos MA, Vejnar CE, Beaudoin JD, et al. CRISPRscan: designing highly efficient sgRNAs for CRISPR-Cas9 targeting in vivo. Nat Methods 2015;12:982–988.
[53]. Tsai SQ, Zheng Z, Nguyen NT, et al. GUIDE-seq enables genome-wide profiling of off-target cleavage by CRISPR-Cas nucleases. Nat Biotechnol 2015;33:187–197.
[54]. Gasperini M, Findlay GM, McKenna A, et al. CRISPR/Cas9-mediated scanning for regulatory elements required for HPRT1 expression via thousands of large, programmed genomic deletions. Am J Hum Genet 2017;101:192–205.
[55]. Han K, Jeng EE, Hess GT, et al. Synergistic drug combinations for cancer identified in a CRISPR screen for pairwise genetic interactions. Nat Biotechnol 2017;35:463–474.
[56]. Xu CL, Ruan MZC, Mahajan VB, et al. Viral delivery systems for CRISPR. Viruses 2019;11:28.
[57]. Schertzer MD, Thulson E, Braceros KCA, et al. A piggyBac-based toolkit for inducible genome editing in mammalian cells. RNA 2019;25:1047–1058.
[58]. Ishida K, Xu H, Sasakawa N, et al. Site-specific randomization of the endogenous genome by a regulatable CRISPR-Cas9 piggyBac system in human cells. Sci Rep 2018;8:310.
[59]. Djurovic S, Iversen N, Jeansson S, et al. Comparison of nonviral transfection and adeno-associated viral transduction on cardiomyocytes. Mol Biotechnol 2004;28:21–32.
[60]. Li Y, Muffat J, Omer Javed A, et al. Genome-wide CRISPR screen for Zika virus resistance in human neural cells. Proc Natl Acad Sci U S A 2019;116:9527–9532.
[61]. Fulco CP, Munschauer M, Anyoha R, et al. Systematic mapping of functional enhancer-promoter connections with CRISPR interference. Science 2016;354:769–773.
[62]. Gilbert LA, Horlbeck MA, Adamson B, et al. Genome-scale CRISPR-mediated control of gene repression and activation. Cell 2014;159:647–661.
[63]. Hnisz D, Abraham BJ, Lee TI, et al. Super-enhancers in the control of cell identity and disease. Cell 2013;155:934–947.
[64]. Maurano MT, Humbert R, Rynes E, et al. Systematic localization of common disease-associated variation in regulatory DNA. Science 2012;337:1190–1195.
[65]. Zhu S, Li W, Liu J, et al. Genome-scale deletion screening of human long non-coding RNAs using a paired-guide RNA CRISPR-Cas9 library. Nat Biotechnol 2016;34:1279–1286.
[66]. Bester AC, Lee JD, Chavez A, et al. An integrated genome-wide CRISPRa approach to functionalize lncRNAs in drug resistance. Cell 2018;173:649–664.e20.
[67]. Liu Y, Cao Z, Wang Y, et al. Genome-wide screening for functional long noncoding RNAs in human cells by Cas9 targeting of splice sites. Nat Biotechnol 2018;36:1203–1210.
[68]. Inoue F, Ahituv N. Decoding enhancers using massively parallel reporter assays. Genomics 2015;106:159–164.
[69]. Muerdter F, Boryń ŁM, Arnold CD. STARR-seq - principles and applications. Genomics 2015;106:145–150.
[70]. Sanjana NE, Wright J, Zheng K, et al. High-resolution interrogation of functional elements in the noncoding genome. Science 2016;353:1545–1549.
[71]. Korkmaz G, Lopes R, Ugalde AP, et al. Functional genetic screens for enhancer elements in the human genome using CRISPR-Cas9. Nat Biotechnol 2016;34:192–198.
[72]. Diao Y, Li B, Meng Z, et al. A new class of temporarily phenotypic enhancers identified by CRISPR/Cas9-mediated genetic screening. Genome Res 2016;26:397–405.
[73]. Xie S, Duan J, Li B, et al. Multiplexed engineering and analysis of combinatorial enhancer activity in single cells. Mol Cell 2017;66:285–299.e5.
[74]. Gasperini M, Hill AJ, McFaline-Figueroa JL, et al. A genome-wide framework for mapping gene regulation via cellular genetic screens. Cell 2019;176:377–390.e19.
[75]. Datlinger P, Rendeiro AF, Schmidl C, et al. Pooled CRISPR screening with single-cell transcriptome readout. Nat Methods 2017;14:297–301.
[76]. Adamson B, Norman TM, Jost M, et al. A multiplexed single-cell CRISPR screening platform enables systematic dissection of the unfolded protein response. Cell 2016;167:1867–1882.e21.
[77]. Dixit A, Parnas O, Li B, et al. Perturb-Seq: Dissecting molecular circuits with scalable single-cell rna profiling of pooled genetic screens. Cell 2016;167:1853–1866.e17.
[78]. Jaitin DA, Weiner A, Yofe I, et al. Dissecting immune circuits by linking CRISPR-pooled screens with single-cell RNA-Seq. Cell 2016;167:1883–1896.e15.
[79]. Matharu N, Rattanasopha S, Tamura S, et al. CRISPR-mediated activation of a promoter or enhancer rescues obesity caused by haploinsufficiency. Science 2019;363:eaau0629.
[80]. Liu XS, Wu H, Krzisch M, et al. Rescue of fragile X syndrome neurons by DNA methylation editing of the FMR1 gene. Cell 2018;172:979–992.e6.

CRISPR/Cas9; genetic tools; genome function; high-throughput screens; mammalian cells

Copyright © 2020 The Chinese Medical Association, Published by Wolters Kluwer Health, Inc. under the CCBY-NC-ND license.