|Home | About | Journals | Submit | Contact Us | Français|
Sarcomas include some of the most aggressive tumors and typically respond poorly to chemotherapy. In recent years, specific gene fusion/mutations and gene over-expression/activation have been shown to drive sarcoma pathogenesis and development. These emerging genomic alterations may provide targets for novel therapeutic strategies and have the potential to transform sarcoma patient care. The RNA-guided nuclease CRISPR-Cas9 (Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR)-associated protein-9 nuclease) is a convenient and versatile platform for site-specific genome editing and epigenome targeted modulation. Given that sarcoma is believed to develop as a result of genetic alterations in mesenchymal progenitor/stem cells, CRISPR-Cas9 genome editing technologies hold extensive application potentials in sarcoma models and therapies. We review the development and mechanisms of the CRISPR-Cas9 system in genome editing and introduce its application in sarcoma research and potential therapy in clinic. Additionally, we propose future directions and discuss the challenges faced with these applications, providing concise and enlightening information for readers interested in this area.
Sarcoma is a rare and heterogeneous group of tumors with 50 different subtypes that exhibit a wide range of differing behaviors and underlying molecular mechanisms [6, 28]. Sarcoma is believed to arise in mesenchymal tissues and often displays highly aggressive behavior with proclivity towards early hematogenous metastasis [15, 62]. Surgery and chemotherapy can significantly improve survival and can be curative when applied early in some types of sarcoma [15, 62]. However, a tendency towards relapse is high and the prognosis of chemo-resistant and disseminated sarcomas remain poor despite multimodal therapeutic approaches . Therefore, further improvements and new approaches for the treatment of sarcoma are essential.
Sarcoma can be classified based on the genetic alterations involved in its development: oncogenic somatic mutations (e.g. KIT and/or platelet-derived growth factor receptor-α (PDGFR-α) mutations in gastrointestinal stromal tumors (GIST)), DNA copy number alterations (e.g. JUN gene amplification in dedifferentiated liposarcoma and myocardin (MYOCD) gene amplification in leiomyosarcoma), and recurrent chromosomal translocations resulting in abnormal fusion proteins (e.g. SYT-SSX gene fusion in synovial sarcoma and Ewing sarcoma breakpoint region 1 and Friend leukemia virus integration 1 (EWSR-FLI1) gene fusion gene in Ewing sarcoma) [15, 52, 72]. More commonly, sarcoma pathogenesis is the result of complex chromosomal abnormalities, as in the case of osteosarcomas and high grade undifferentiated pleomorphic sarcomas (neurofibromin 1 (NF1) gene deletions, point mutations and indels such as P53 and RB) [15, 52, 72]. In addition, a variety of protein kinases including receptor tyrosine kinases (RTKs) are overexpressed or constitutively activated in sarcoma, both in translocation associated sarcomas, such as GIST discussed above, and in karyotypically complex tumors, for instance osteosarcoma [17, 85]. Because RTKs, c-KIT, ATK, EGFR, mTOR, and IGF-1R, for example, and their ligand growth factors are so frequently overexpressed/activated in sarcomas, they represent some of the most attractive therapeutic targets in sarcoma [17, 85].
Although the development of adequate transgenic models has been elusive, sarcoma modeling via genetic engineering remains an important approach for sarcoma research. Conventional transgenic techniques often involve the sophisticated and time-consuming processes of germline manipulation and labor intensive animal cross-breeding, and can model the complexity and multistep nature of sarcoma mutations. RNA interference (RNAi) provides an alternative method for genetic engineering, however, it can usually achieve only temporary and partial knockdown . RNAi is also restricted to expressed genes and has pervasive off-target effects [1, 45]. Cre (cyclization recombinase)-loxP (locus of recombination in P1) recombination consists of a single enzyme, Cre recombinase. Cre recognizes a 34-bp site on the P1 genome called loxP and catalyzes reciprocal conservative DNA recombination between pairs of loxP sites . Much of the power of the Cre-loxP system derives from the potential to generate conditional mutants [47, 79]. However, conditional deletion of every gene in every cell type may not be realistic [47, 79]. Conditional deletions are only as good as the promoters that regulate Cre expression, and transgenic Cre driver lines have proven to be somewhat problematic, exhibiting expression outside the target tissue as well as inefficient cleavage leading to mosaicism [47, 79]. Recently, clustered regularly interspaced short palindromic repeats (CRISPR)-associated Cas9 protein, a novel genome editing tool has been implemented in a multitude of model organisms and cell types [4, 13]. Compared with RNAi and Cre-LoxP, CRISPR-Cas9 is an exogenous system that does not compete with endogenous processes and functions at the DNA level targeting transcripts, which results in knockdown or complete elimination of gene function. Furthermore, CRISPR-Cas9 provides a larger targetable sequence space in which promoters of the gene may also be targeted. CRISPR-Cas9 genome editing technology holds many advantages and it has already started to supplant incumbent genome editing technologies, such as transcription activator-like effector nucleases (TALENs) and zinc finger nucleases (ZFNs) (Table 1) . We review the development and potential applications of the CRISPR-Cas9 system in genome editing and introduce these applications in sarcoma modeling and therapy. We also discuss the challenges of CRISPR-Cas9 and future avenues for innovation.
The CRISPR-Cas9 system is currently used for RNA-guided endonuclease gene editing. The core components of this system are a nuclease Cas9 and a single guided RNA (sgRNA) [4, 7, 13, 45, 48, 59, 64, 77, 80, 84]. The nuclease Cas9 consists of two catalytic active domains: HNH and RuvC [4, 7, 13, 45, 48, 59, 64, 77, 80, 84]. The HNH domain is a single nuclease domain, whereas the RuvC domain contains three subdomains across the linear protein sequence [4, 7, 13, 45, 48, 59, 64, 77, 80, 84]. RuvC I near the N-terminal region of Cas9 and RuvC II/III flank the HNH domain is near the middle of the protein. The HNH and RuvC nuclease domains are properly positioned for cleavage of the complementary and non-complementary strands of the target DNA, respectively [4, 7, 13, 45, 48, 59, 64, 77, 80, 84]. The sgRNA, which has an invariant scaffold region and a spacer region, is derived from CRISPR RNA (crRNA) and trans-activating crRNA (tracr RNA). The sgRNA binds to Cas9 and directs it to the locus of interest by a 20-nt guide sequence via base pairing to the genomic target [4, 7, 13, 45, 48, 59, 64, 77, 80, 84] (Fig. 1).
The target sequence in the genomic DNA paired to the sgRNA sequence is immediately followed by a NGG sequence called the protospacer adjacent motif (PAM). The PAM sequence is located on the immediate 3′ end of the sgRNA recognition sequence, but it is not a part of the 20-nt guide sequence within the sgRNA [4, 7, 13, 45, 48, 59, 64, 77, 80, 84] (Fig. 1). The CRISPR-Cas9 system uses Cas9, which complexes with the sgRNA, to cleave DNA 3–4 base pairs upstream of PAM and generates double-strand breaks (DSBs) in a sequence-specific manner [4, 7, 13, 45, 48, 59, 64, 77, 80, 84]. The DSBs are then repaired either by non-homologous end joining (NHEJ)-mediated error-prone DNA repair or homologous directed repair (HDR)-mediated error-free DNA repair (Fig. 2). The former repair can rapidly ligate the DSB, but generates small insertion and deletion mutations at the target sites [20, 55, 64, 65]. These mutations could disrupt and abolish the function of target genes or genomic elements [20, 55, 64, 65]. For instance, an sgRNA that targets a protein-coding region can produce loss of function frame-shifting indels through NHEJ-mediated DNA repair. HDR-mediated error-free DNA repair requires a homology-containing donor DNA sequence as repair template. Through co-transfection with Cas9, KRAS sgRNA and an oligonucleotide template into wild-type human intestinal organoids, the KRASG12D (GGT>GAT) mutation was successful achieved [20, 55, 64, 65].
Sarcoma pathogenesis is a multistep process that involves many genetic alterations and epigenetic changes. Genome sequencing studies have identified a large collection of genetic and epigenetic alterations that occur in different types of human sarcomas [22, 65]. Experimental strategies to manipulate the genomes of either normal cells or tumor cells are crucial for understanding the pathogeneses of sarcoma and for discovering potential therapeutic targets. Although several genetic engineering technologies, such as Cre-loxP, ZFNs, and TALENs, have been applied to different tumor model systems, they are generally labor intensive, time consuming, and have low efficiency for gene targeting [26, 47]. Thus, the need for easier, rapid, and high efficiency genetic modeling approaches remains. The CRISPR-Cas9 genome editing technology has revolutionized the field of genetic engineering and may overcome many of the limitations of earlier techniques including Cre-loxP, ZFNs, and TALENS, on carrying out deletions, insertions, translocations, and inversions at specific sites in the DNA of cells [4, 7, 13, 45, 48, 59, 64, 77, 80, 84]. Although it is capable of generating transgenic mice via embryo manipulations similar to those in Cre-lox techniques, CRISPR-Cas9 has been shown to be easy to design and use, and its multiplexing nature streamlines the generation of animal and cellular cancer models enabling rapid functional interrogation of cancer-associated genes (Table 2) [4, 7, 13, 45, 48, 59, 64, 77, 80, 84].
Genome sequencing studies have identified multiple driver gene mutations in human sarcomas. CRISPR-Cas9 is capable of inducing loss of function (LOF), and gain of function (GOF) mutations in vitro and in vivo. So far, CRISPR-Cas9 has been used in some malignant tumor models, including lung cancer, colorectal cancer, and myeloid leukemia [22, 27, 28, 30, 31]. By delivering the combination of sgRNA and Cas9 with a lentiviral vector, five genes (TET2, DNMT3A, RUNX1, NF1, and EZH2) demonstrated LOF in a single hematopoietic stem cell, leading to clonal outgrowth and myeloid malignancy . Hematopoietic stem cells were also derived from mice with knock-in of the FLT3 internal tandem duplication (FLT3-ITD) mutation . Using cooperating LOF mutations in genes encoding epigenetic modifiers, transcription factors, and mediators of cytokine signaling (such as TET2, DNMT3A, RUNX1, NF1, SMC3, P53, ASXL1, and EZH2), researchers were able to generate models of acute myeloid leukemia (AML). These models recapitulated the combinations of mutations observed in patients . A recent study has identified that deregulation of APC, P53, KRAS, and SMAD4 is sufficient for transformation of cultured mouse colon cells into colorectal cancer . CRISPR-Cas9 was utilized to induce LOF mutations in APC, P53, and SMAD4, and GOF mutation in KRAS in cultured human intestinal stem cells . To introduce the GOF mutation, an oligonucleotide with the KRAS mutation and two silent mutations was designed to serve as a template for HDR. Mutant organoids (KRASG12D/APCKO, KRASG12D/APCKO/P53KO, and KRASG12D/APCKO/P53KO/SMADKO) were selected by removing individual growth factors from the culture medium. Upon xenotransplantation into mice, quadruple mutants grew as tumors with features of invasive carcinoma . Another colorectal cancer model using CRISPR-Cas9-mediated engineering of human intestinal organoids has been established by Matano et al. . By modulating the culture conditions to mimic the intestinal environment, LOF mutations of APC, P53, SMAD4, and GOF mutations of KRAS and PIK3CA were introduced into organoids derived from normal human intestinal epithelium. To knock in the KRASG12V mutation, researchers electroporated an sgRNA targeting a sequence adjacent to KRAS exon 2 together with a donor vector containing homologous arms but encoding the KRASG12V mutation. They then selected the isogenic organoids that carried mutations in the tumor suppressor genes APC, P53, and SMAD4, and the oncogenes KRAS and/or PIK3CA . Organoids engineered to express all five mutations grew independently of niche factors in vitro, and an adenocarcinoma sequence model was formed after implantation under the kidney subcapsule in mice .
Hydrodynamic injection is an established method of delivering plasmids selectively to the liver in animal models . Tail vein injection is the most common procedure in rats and mice . DNA plasmids encoding Cas9 and sgRNAs can be delivered to the liver through hydrodynamic tail vein injection to target the tumor suppressor genes PTEN and P53 alone or in combination . As a result, when PTEN was mutated by CRISPR, elevated AKT phosphorylation and lipid accumulation in hepatocytes were observed, while simultaneous mutation of PTEN and P53 induced liver tumors . Furthermore, the feasibility of inducing GOF mutations by CRISPR-Cas9 in the liver was also determined by co-injection of Cas9-sgRNA plasmids targeting β-catenin gene and a single-stranded DNA oligonucleotide donor carrying activating point mutations, which led to the generation of hepatocytes with nuclear localization of β-catenin . To broadly enable the application of Cas9 in vivo, a Cre-dependent Rosa26 Cas9 knock in mouse was established . Using this type of mouse, the top three significantly mutated genes in lung adenocarcinoma (KRAS, P53, and LKB1) were simultaneously modeled, leading to macroscopic tumor of lung cancer pathology .
Recently, CRISPR-Cas9 has been used to knock out tumor suppressor genes cyclin-dependent kinase inhibitor 2A (CDKN2A), P53, and PTEN in mice to establish three murine sarcoma models . First, scientists developed a multiple lentiviral expression (MuLE) viral system which allows multiple sgRNAs to be expressed together with hCas9 from a single viral construct expression . To model human sarcoma, a series of MuLE vectors were designed to systematically investigate the single and combinatorial effects of gain of H-RAS function and loss of CDKN2A, P53, and PTEN functions . To investigate the single effects of the driven gene, researchers generated tricistronic MuLE vectors expressing sgRNAs targeting exon 7 or exon 8 of the P53 locus or exon 2 of the CDKN2A locus together with expression of hCas9 and puromycin resistance . To demonstrate that cooperative genetic tumor modeling can be achieved using CRISPR-Cas9, they generated tetracistronic MuLE vectors designed to express either scrambled sgRNA or sgRNA targeting P53 exon 7 or exon 8, as well as H-RASG12V, hCas9, and puromycin resistance. Strikingly, several types of soft tissue sarcomas were developed in mice when their muscles were injected with CDKN2A+H-RAS, P53+H-RAS, and P53+PTEN+H-RAS combinatorial viruses. Histological analysis of these sarcomas revealed that they were undifferentiated sarcoma with pleomorphic and rhabdoid features . These studies demonstrate that CRISPR-Cas9 is a powerful genome editing technique that is able to simultaneously target genetic mutations to multiple loci and allows for the rapid and systematic generation of genetically complex .
Abnormalities in chromosome number and structure are frequently observed in sarcoma cells but have been difficult to generate in a highly specific manner for function analysis . CRISPR-Cas9 may be able to mediate complex manipulations of gene structure, such as chromosomal rearrangements, which give rise to oncogenic fusion genes, oncogene amplifications, or oncosuppressor deletions [24, 45]. In lung cancer, three types of chromosomal rearrangements were successfully generated by CRISPR-Cas9, including the CD74-ROS1 translocation event and the EML4-ALK and KIF5B-RET inversion events [5, 12, 46]. Authors designed sgRNAs targeting intron 6 of CD74 and intron 33 of ROS1, which were then co-expressed with Cas9 in HEK293T cells. Translocations were detected in cells expressing both CD74 and ROS1 sgRNAs [5, 12, 46]. Similarly, for both EML4-ALK and KIF5B-RET, the expected inversions were only detected in cells expression Cas9 along with the appropriate pair of sgRNAs [5, 12, 46]. In addition to modeling chromosomal rearrangements in cell lines, mouse models of EML4-ALK gene rearrangement by CRISPR-Cas9 in non-small-cell lung cancers (NSCLCs) were generated . First, sgRNAs targeting intron 14 of EML4 and intron 19 of ALK were cloned into the Cas9-expreesing plasmid PX330, and the resulting constructs were co-transfected into NIH/3T3 cells . The desired EML4-ALK inversion was detected in cells. Next, to deliver Cas9 and sgRNAs to the lungs of adult mice, the dual sgRNA-Cas9 cassette was transferred into an adenoviral shuttle vector and recombinant adenoviruses (Ad-EA) were produced . Then, a cohort of adult mice was infected by intratracheal instillation of Ad-EA. The EML4-ALK rearranged NSCLCs developed in mice two months after inoculation . While various types of sarcomas show characteristic translocations, gene fusions generated from these translocations are the initiating events of many sarcomas and are likely essential in some subtypes of these tumors .
Alveolar rhabdomyosarcoma (A-RMS), the third most common soft-tissue sarcoma in children, is typified by a translocation that fuses the PAX3 gene on chromosome 2 to the FOXO1 gene on chromosome 13 [15, 62, 75]. It is complicated to mimic this translocation in a mouse, because the PAX3 and FOXO1 genes in the mice are in an opposite orientation on their respective chromosomes . To circumvent this limitation, Lagutina and colleagues took a two-step approach . First, they created a mouse model by chromosomal engineering via Cre recombinase-mediated genetic alterations . In this model, the orientation of a 4.9Mb syntenic fragment on chromosome 3, encompassing FOXO1, is inverted. Then, they used Cas9-sgRNAs targeting chromosome 1 in intron 7 of PAX3 and chromosome 3 in intron 1 of FOXO1 to induce DNA DSBs in PAX3 and FOXO1 . Using the above method, a CRISPR-Cas9-mediated PAX3-FOXO1 fusion gene was successfully generated in mice . The study showed that myoblasts isolated from fore and hind limbs kept their PAX3-expressing identity and co-localization of PAX3-FOXO1 had a higher frequency in fore limb myoblasts than hind limb myoblasts . This mouse model will be a valuable tool for studying mechanisms underlying the initial stage of the A-RMS implicated chromosome translocations.
To generate human chromosomal translocations using CRISPR-Cas9, Ewing sarcoma (ES) was chosen as the test model as it is defined by the occurrence of a chromosomal translocation. ES is a rare aggressive malignant neoplasm that occurs primarily in teenagers and young adults and that typically arises in tissues of mesenchymal origin. It is characterized by the t(11;22)(q24;q12) chromosomal translocation, which leads to the generation of the EWSR1-FLI1 fusion gene . To target the fusion gene, Torres et al. designed four sgRNAs to target intron 4 (F1, F2) of FLI1 and intron 7 (E1, E2) of EWSR1 . First, the ES hallmark t(11;22)/EWSR1-FLI1 chromosomal translocation was induced and characterized in HEK293 cells and in human primary mesenchymal stem cells (hMSCs) . Then, DNA DSBs in the target loci were generated by transfection of HEK293 cells with a plasmid expressing Cas9 and sgRNAs E1, E2, F1, and F2 . Quantification in metaphase spreads and interphase nuclei revealed a reciprocal translocation rate of 1.76±0.2% or the E1F2 pair-wise sgRNA combination, whereas none of the three other combinations scored positive for illegitimate events in 280–290 nuclei examined . Next, the group addressed whether the chromosomal translocation driven by CRISPR-Cas9 would be able to replicate the synthesis of the functional EWSR1-FLI1 fusion gene . The functionality of the fusion protein in the pool of hygromycin-selected t(11;22) positive populations was examined. qRT-PCR confirmed the upregulation of six well-known target genes (EZH2, HMGA2, NKX2.2, ID2, NR0B1, and SOX2) of the EWSR1-FLI1 fusion protein, which suggested that the fusion protein expressed after induction of a chromosomal translocation by the CRISPR-Cas9 system has a similar activity to that expressed in primary ES cells .
Genome sequencing studies demonstrate that human sarcoma is a complex process that involves a large collection of genetic alterations; the determination of which mutations are casually related to tumorigenesis remains a major challenge . CRISPR-Cas9 has been used for rapid functional investigation of candidate genes in well-established models of cancer. Using a KRASG12D-driven lung cancer model, CRISPR-Cas9 was adopted to perform functional characterization on a panel of tumor suppressor genes (NKX2-1, PTEN, and APC), with known LOF alterations in human lung cancer . Cre-dependent somatic activation of oncogenic KrasG12D combined with CRISPR-Cas9-mediated genome editing of NKX2-1, PTEN, and APC genes resulted in lung cancer with distinct histopathological and molecular features. Through the introduction of constitutive Cas9 and inducible P53 sgRNAs into hematopoietic stem/progenitor cells (HSPCs), which were then transplanted into myelo-ablated recipient mice, a hematopoietic-cell-restricted P53-knockout mouse model was developed . Interestingly, this study generated mutations that not only caused loss of P53 protein but also novel mutant P53 proteins that could promote lymphoma development . In addition, a genome-wide Cas9 knockout screen was used to develop a mouse model of tumor evolution . In the study, the initial lung cancer cell line had little capacity to form metastases; however, after being mutagenized with mouse genome-scale CRISPR knockout library A (mGeCKOa), the cell population formed highly metastatic tumors . This study provides a roadmap for in vivo Cas9 screens and makes genome-scale CRISPR screening feasible using a transplant model with virtually any cell line or genetic background (e.g., mutations in KRAS, CDKN2A, P53, PTEN, etc.).
Recently, mouse models of osteosarcoma, ES, A-RMS, and synovial sarcoma have shown that inactivation of the P53 pathway is present in the vast majority of human sarcomas (Table 3) [8, 14, 15, 18, 19, 27, 29, 34, 37, 38, 41, 43, 44, 54, 58, 61–63, 68, 70, 75, 76, 81]. For example, mutant KRAS expression and P53 loss cooperate in the development of undifferentiated pleomorphic sarcomas [15, 75]. Therefore, the CRISPR-Cas9 system could be useful for functional genomic studies in these established sarcoma models.
CRISPR-Cas9 can not only build various sarcoma models, but can also be used to explore drug treatment and resistance. In the NSCLC mouse model mentioned above, CRISPR-Cas9 mediated EML4-ALK rearrangement. The resulting tumors harbored the EML4-ALK inversion, expressed the fusion EML4-ALK gene, and responded to treatment with the ALK inhibitor crizptinib [46, 51]. In a new mouse model of ARF−/− Eμ-myc B-cell lymphoma, CRISPR-Cas9-mediated disruption of p53 conferred cells with resistance to doxorubicin treatment [49, 51]. These mouse models provide powerful tools for studying the mechanisms of drug resistance and testing novel therapies [49, 51].
Several studies have revealed that CDK11 is essential in cancer, including in sarcoma cell growth and survival [21, 23, 33, 71, 86]. Our previous study identified that CDK11 is critical for the growth and proliferation of liposarcoma cells, suggesting that CDK11 may be a promising therapeutic target for treatment of liposarcoma patients . Recently, CRISPR-Cas9 was applied to determine the effect of targeting endogenous CDK11 at the DNA level in osteosarcoma cell lines . Furthermore, the migration and invasion activity was markedly reduced by CDK11 knockout, indicating that CDK11 maybe a novel therapeutic target for osteosarcoma .
The development of multidrug resistance (MDR) is one of the major obstacles in the chemotherapy treatment of osteosarcoma . Multidrug resistance gene 1 (MDR1), which encodes the membrane efflux pump P-glycoprotein (P-gp), plays an important role in the drug resistance process . Overexpression of MDR1 results in an active efflux of anti-cancer agents from cells, thus lowering intracellular drug concentrations and inducing cancer cell resistance to chemotherapeutic drugs, such as doxorubicin and paclitaxel . Similar to targeting CDK11 as described above, the CRISPR-Cas9 system can also be applied to knockout MDR1 in drug resistant sarcoma cells to reverse drug resistance. Our group has designed an sgRNA targeting exon 5 of MDR1 gene (Fig. 3). When plasmids with Cas9 and sgRNA were transfected into osteosarcoma MDR cell lines (KHOSR2 and U2OSR2), MDR1 was efficiently knocked out (Fig. 4).
The validation of a drug target is an essential step in drug discovery and development [35, 77]. While identification of drug resistance mutations is considered as the gold standard for target confirmation, further validation of drug-target interaction requires the introduction of a resistance mutation into the wild-type background [35, 77]. However, such an approach has not been straight forward in mammalian cells until the advent of the CRISPR-Cas9 technique. Recently, a proof of concept for drug validation in human cells was achieved based on resistance selection and genome sequencing in combination with CRISPR-Cas9 genome editing [35, 77].
Human exportin-1 (XPO1), also known as chromosome region maintenance 1 protein (CRM1), is considered an anticancer target . In recent years, over expression of XPO1 has been observed in osteosarcoma and has been correlated with poor prognosis and resistance to therapy . Selinexor (KPT-330), an inhibitor of XPO1, is currently undergoing phase 2 clinical trials and has demonstrated high response rates as a therapeutic agent in phase 1 trials for heavily pretreated, relapsed, and refractory hematological and solid tumor malignancies in humans [16, 36, 53]. Selinexor could inhibit the function and formation of XPO1, possibly through binding to the cysteine 528 residue of XPO1, which leads to an accumulation of tumor suppressor proteins in the nucleus of treated cells, and thus cell cycle arrest and apoptosis [16, 36, 53]. To validate this drug-target interaction, CRISPR-Cas9 was adopted to introduce a single XPO1 C528S mutation in acute T cell leukemia Jurkat cells. As expected, this mutation prevented Selinexor-meditated functional inhibition of XPO1 by blocking XPO1-Selinexor binding [53, 77]. The study confirmed that XPO1 served as the prime target of Selinexor in cancer cells [53, 77].
Ispinesib, an inhibitor of kinetin spindle protein (KSP/HsEg5), has entered clinical trials as an anticancer drug . The drug demonstrated a high level of in vivo anti-tumor activity against ES  and has been applied to patients with rhabdomyosarcoma and osteosarcoma in a clinical trials . Sequencing studies and bioinformatics analysis suggest that mutations in the Ala133 residue of KSP may be responsible for ispinesib resistance [35, 77]. To confirm this, vectors harboring the Cas9 ‘nickase’ and sgRNAs along with template DNA bearing the desired mutation were transfected into HeLa cells . In the mutant transfectants, mutagenesis of Ala133 was confirmed by using the Surveyor mutation-detection and Sanger sequencing of the genomic locus. The results showed that the A133P substitution conferred >150-fold resistance to ispinesib, which validated previous results from sequencing and bioinformatics studies .
In addition to specific drug-target validation, CRISPR-Cas9 systems can also be used to conduct genome-scale screens for mutations that confer drug resistance . The Braf V600E mutation has been recently detected in a subset of histiocytic tumors, particularly in histiocytic sarcoma and Langerhans cell histiocytosis [10, 32, 56]. Vemurafenib, a therapeutic Braf inhibitor, demonstrated efficacy in some tumors carrying the Braf V600E mutation . A genome-scale CRISPR-Cas9 knockout (GeCKO) library targeting 18,080 genes with 64,751 unique guide sequences was used to identify genes whose loss is involved in resistance to vemurafenib . The identified candidates included previously validated genes (NF1 and MED12) and novel genes such as NF2, CUL3, TADA2b, and TADA1 . Similarly, a library consisting of 70,290 guides targeting all human RefSeq (the Reference Sequence of NCBI) coding isoforms was synthesized to screen for genes that confer resistance to vemurafenib . In this study, a new activation system called synergistic activation mediator (SAM) was used . SAM is an engineered protein complex for the transcriptional activation of endogenous genes . The results demonstrated that the SAM system is robust and specific, and can facilitate genome-scale GOF screening when combined with a compact pooled sgRNA library. The SAM-mediated screens’ results confirmed genes previously shown to yield drug resistance, and novel candidates were validated using individual sgRNA and complementary DNA overexpression . In comparison with RNAi screens using a short hairpin RNA (shRNA) library, genome-scale screening with CRISPR-Cas9 showed a higher reagent consistency and stronger phenotypic effects of individual sgRNAs [66, 77]. In addition, these studies exemplified the significant advantages of the CRISPR-Cas9 system in genome-wide LOF and GOF screens for anticancer drug development .
CRISPR-Cas9 technology has evolved to create a simple, RNA-programmable method to precisely mediate genome editing in mammalian cells, although some shortcomings still exist. One potential limitation of CRISPR-Cas9 technology is that the approach may create off-target effects, though relatively uncommon as compared with other genome editing tools, such as RNAi, TALEN, or ZFN[11, 74]. The efficiency of editing by CRISPR-Cas9 systems can also be further improved [11, 74]. There are several reports demonstrating that the efficiency of CRISPR-Cas9, albeit sufficient to induce tumors, is relatively low [45, 46, 57, 65, 82]. Various strategies have been reported to reduce off-target effects [11, 45, 74, 87–89]. The choice of unique target sequences is important for avoiding off-target effects . The target sequences should differ from any other sites in the genome by at least two or three nucleotides in a 20-nt sequence . In addition, paired Cas9 nickases are highly specific in human cells, and can generate two single-stand breaks or nicks on different DNA strands . Additionally, more efficient deliveries, more powerful sgRNAs, and more potent Cas9 systems are in development [11, 45, 74, 87, 88]. The SAM system, which has been developed by engineering the sgRNA through appending a minimal hairpin aptamer to the tetraloop and stem loop 2 of sgRNA, is specific with minimal off-target activity when SAM-mediated gene activation [39, 87]. Streptococcus pyogenes Cas9-HF1 (SpCas9-HF1) was described to be a high-fidelity CRISPR-Cas9 nuclease . It retains on-target activities comparable to wild-type SpCas9 with >85% of sgRNAs tested in human cells . Recently, “enhanced specificity” SpCas9 (eSpCas9) was named. It was able to dramatically reduce “off-target editing” to undetectable levels in the specific cases examined by changing three of the approximately 1,400 amino acids that make up SpCas9 .
CRISPR-Cas9 systems have been used in sarcoma modeling and therapy explorations. However, the CRISPR-Cas9 system has an even wider potential for application in sarcoma. For example, CRISPR interference (CRISPRi), which refers to transcriptional suppression of target genes by Cas9 binding, can be used to knock down cancer-associated genes for functional interrogations [25, 45]. Catalytically inactivated Cas9 (deadCas, dCas9) can be repurposed as a RNA-guided DNA-binding domain fused to various effectors, such as green fluorescent proteins, that enable live imaging of genome loci of interest [25, 45]. When the effector is an epigenetic modifier, such as a DNA methyltransferase or histone acetyltransferase, it reaches the target sites directed by sgRNA and regulates epigenetic alterations, a process termed epi-genome editing . In sarcoma, epigenetic abnormalities and genomic mutations are two sides of one coin. They interact with each other and cooperate to drive carcinogenesis. Given this, dCas9-mediated epigenome editing may become a convenient tool in the future for modeling sarcoma-related epigenetic abnormalities and a promising avenue for sarcoma therapy .
In conclusion, CRISPR-Cas9 systems can mediate genome editing, epigenetic regulations, and transcriptome modulations . Though some challenges remain ahead, the application of this technology to several aspects of sarcoma biology, ranging from basic research to clinical and translational applications, offers numerous exciting opportunities for a better understanding and potential treatment of these devastating diseases.
This work was supported, in part, by the Gattegno and Wechsler funds, the Kenneth Stanton Fund, and the Jennifer Hunter Yates Foundation. Dr. Duan is supported, in part, through a grant from Sarcoma Foundation of America (SFA), a grant from National Cancer Institute (NCI)/National Institutes of Health (NIH), UO1, CA 151452-01, a pilot grant from Sarcoma SPORE/NIH, and a grant from an Academic Enrichment Fund of MGH Orthopaedics. Dr. Liu is supported by a scholarship from the Chinese Scholarship Council.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Author contributionsAll authors met the International Committee for Medical Journal Editors criteria for authorship, were fully involved in manuscript development, and assume responsibility for the direction and content. All authors have approved the manuscript for submission.
Conflict of interest