The 2020 Nobel Prize in Chemistry was awarded to Emmanuelle Charpentier and Jennifer Doudna. They received the prestigious award for their groundbreaking description of the clustered regularly interspaced short palindromic repeats (CRISPR)–CRISPR-associated (Cas) system as a genome editing tool in 2012.¹ Since then, what was originally known as the adaptive antiviral immune system of bacteria, these “molecular scissors” have been exploited as a tool for genome editing by biomedical researchers working in fields as diverse as hematology, neuroscience and immunology, with several clinical trials planned or already ongoing for the treatment of diverse inborn pathologies^{2, 3} (also see NCT03872479, NCT04191148 and NCT04601051 at https://clinicaltrials.gov).

CRISPR-mediated genome editing has been garnering interest in cancer therapy, particularly as a tool to generate “off-the-shelf” T-cell products that do not express major histocompatibility complex molecules, or to render T cells refractory to inhibitory signaling.⁴ The employment of T cells with transgenic T-cell receptors (TCRs) or chimeric antigen receptors (CARs) specific to tumor antigens in combination with CRISPR-mediated genome editing is particularly promising. This has resulted in stabilized transgene expression, enhanced T-cell effector function and/or increased efficacy in murine models of both blood cancers and solid tumors.⁴ Nonetheless, the generation of T cells with both knockouts and knock-ins is a laborious process. Furthermore, for targeting specific T-cell subsets, enrichment or cell sorting is required prior to genetic engineering. However, recent work published in Cell Reports by the aforementioned Nobel laureate Jennifer Doudna⁵ has started to pave the way for change.

Electroporation is currently the dominant strategy for delivering Cas9 into cells. Usually, Cas9 is precoupled to the guide RNA and trans-activating CRISPR RNA that are used to target the Cas9 protein to and activate the Cas9 protein at the genomic locus of interest. The combined Cas9–RNA complex is referred to as a Cas9 ribonuclear protein (RNP), and Cas9 RNP electroporation was shown to be an effective method for the genetic modification of human T cells.^{6, 7} However, for large amounts of cells and/or targets, electroporation is impractical. Hamilton et al.⁵ report on a system that shuttles Streptococcus pyogenes Cas9 RNPs into cells utilizing retrovirus-like particles (VLPs; Figure 1a). To produce the Cas9 RNP-containing VLPs, the authors fused Cas9 to the Gag structural viral protein to ensure uptake into VLPs during viral assembly. While the use of VLPs as a vector to introduce Cas9 proteins into cells is not novel, the authors were able to produce VLPs that contain Cas9 RNPs. In contrast to previous reports using VLPs as a Cas9 carrier,⁸ this approach results in a VLP that contains both the Cas9 machinery and required RNAs (i.e. the guide RNA and trans-activating CRISPR RNA), and thus does not require the separate introduction of targeting guide RNAs. Furthermore, this new technique also allows for the simultaneous introduction of a transgene of interest, such as a tumor-specific TCR or CAR. Of note, by tagging the transgene via a self-cleaving peptide with a fluorescent marker, cells expressing the transgene can be easily enriched by, for example, fluorescent cell sorting. Lastly, what also makes this method unique is the ability to target specific cell subsets in a mixed population, at least in vitro. By making use of VLPs that have been pseudotyped with the envelope (Env) protein of the CD4-tropic HIV-1, they were able to produce VLPs that efficiently infect CD4⁺ T cells at a high specificity: approximately 50% of CD4⁺ T cells expressed a fluorescent reporter gene, compared with only about 2.5% of CD8⁺ T cells from the same well.

Currently, T cells equipped with CARs are being employed for the treatment of several blood cancers.⁹ While this type of therapy is very efficient for blood cancers, CAR T cells are less effective for the treatment of solid tumors. Therefore, many groups have tried to enhance T-cell effector function by making use of CRISPR/Cas9-mediated genome editing in order to maximize T-cell killing potential. Ranging from direct augmentation of cytokine production by modulating TCR or costimulatory signaling to fine-tuning post-transcriptional regulation, or by simply knocking out inhibitory receptors such as PD-1,⁴ many of these approaches utilize a two-step process and required sequential genome editing and introduction of tumor antigen-specific CARs or TCRs. Instead, the novel method from the Doudna laboratory is able to do both simultaneously.⁵ Recently, the first-in-human clinical trial with CRISPR/Cas9-generated PD-1 knockout NY-ESO-1 TCR transgenic T cells was performed,¹⁰ where indeed subsequent electroporation and lentiviral transduction steps were required to introduce the PD-1 knockout and transgenic TCR, respectively. The VLP-mediated delivery of both the Cas9 machinery and transgene of interest could have significantly streamlined and optimized the production of the therapeutic T-cell product. Lastly, by targeting β2 microglobulin and/or T-cell receptor alpha chain (TRAC), as exemplified by Hamilton et al.,⁵ VLPs could also be used to produce “off-the-shelf” (CAR) T-cell products for adoptive therapy.⁴

Significantly, the approach of Hamilton et al. results in only transient presence of genome editing machinery components, limiting the risk for prolonged off-target effects as a result of sustained expression upon Cas9-transgene insertion. This is a huge benefit compared with lentiviral or retroviral delivery of Cas9 genes and guide RNAs, as the constitutive expression of genome editing machinery can result in off-target effects over time. While this can be mitigated with on/off switchable or cell-specific promoters, transient Cas9 presence is preferred. Furthermore, it has been shown that a majority of healthy individuals possess antibodies directed against Cas9 proteins,¹¹ further highlighting the importance of transient Cas9 expression and/or presence, especially when employing the end product in therapeutic settings.

Hamilton et al.¹² exploited the cell tropism of HIV-1, but this can be expanded upon by pseudotyping VLPs with other viral envelope proteins, for example, the envelope protein of the rabies virus for targeting neural cells, or influenza for targeting airway epithelial cells. Nonetheless, the specific targeting of CD4⁺ T cells with CRISPR VLPs has broad applications (Figure 1b), ranging from the (in vivo) production of CAR T cells, where Cas9-mediated knockout of TRAC genes can stabilize transgene expression,⁴ to rendering CD4⁺ T cells of HIV-1-infected individuals refractive to infection by knocking out, for example, CCR5 and/or replacing it with the CCR5Δ mutant that does not allow cellular entry,¹³ to the correction of inborn errors affecting T-cell effector function¹⁴ by knocking out defective molecules and replacing them with functional copies.

Of note, the HIV-1-pseudotyped version of the Cas9-VLP resulted in lower genome editing frequencies (approximately 14–28%) in primary human T cells compared with conventional methods such as electroporation.^{6, 7} This can potentially be increased by making use of recently described more efficient Cas9 molecules,¹⁵ but this remains to be determined. Furthermore, CD4-tropic HIV-1 strains have also been shown to infect macrophages, which should be taken into careful consideration, especially before implementing this tool for in vivo gene editing. Future research in humanized mice could shed light on in vivo safety and feasibility of Cas9-VLP-mediated genome editing. Nonetheless, the study by Hamilton et al. provides a novel angle for the therapeutic use of CRISPR-mediated genome editing and provides a solid base for future research.

2020 年诺贝尔化学奖授予 Emmanuelle Charpentier 和 Jennifer Doudna。2012年，他们凭借对成簇规则间隔短回文重复序列 (CRISPR)-CRISPR 相关 (Cas) 系统作为基因组编辑工具的开创性描述而获得了著名的奖项。¹从那时起，最初称为适应性抗病毒免疫系统在细菌中，这些“分子剪刀”已被血液学、神经科学和免疫学等多个领域的生物医学研究人员用作基因组编辑的工具，并计划或正在进行多项临床试验，以治疗各种先天性疾病^{2, 3}（另见 https://clinicaltrials.gov 上的 NCT03872479、NCT04191148 和 NCT04601051）。

CRISPR 介导的基因组编辑在癌症治疗中引起了人们的兴趣，特别是作为一种工具来产生不表达主要组织相容性复合分子的“现成”T 细胞产品，或使 T 细胞对抑制性信号具有抵抗力。⁴ T 细胞与肿瘤抗原特异性的转基因 T 细胞受体 (TCR) 或嵌合抗原受体 (CAR) 结合 CRISPR 介导的基因组编辑特别有前景。这导致稳定的转基因表达、增强的 T 细胞效应器功能和/或增加的血癌和实体瘤小鼠模型的功效。⁴尽管如此，同时敲除和敲入的 T 细胞的产生是一个费力的过程。此外，为了针对特定的 T 细胞亚群，需要在基因工程之前进行富集或细胞分选。然而，上述诺贝尔奖获得者詹妮弗·杜德娜 (Jennifer Doudna) ⁵最近在《细胞报告》 ( Cell Reports) 上发表的工作已经开始为变革铺平道路。

电穿孔是目前将 Cas9 递送到细胞中的主要策略。通常，Cas9 与引导 RNA 和反式激活 CRISPR RNA 预偶联，这些 CRISPR RNA 用于将 Cas9 蛋白靶向并激活感兴趣的基因组位点处的 Cas9 蛋白。结合的 Cas9-RNA 复合物被称为 Cas9 核糖核蛋白 (RNP)，Cas9 RNP 电穿孔被证明是人类 T 细胞遗传修饰的有效方法。^{6, 7}然而，对于大量细胞和/或目标，电穿孔是不切实际的。汉密尔顿等人。⁵报告化脓性链球菌穿梭系统利用逆转录病毒样颗粒（VLP；图 1a）将 Cas9 RNP 导入细胞。为了生产含有 Cas9 RNP 的 VLP，作者将 Cas9 与 Gag 结构病毒蛋白融合，以确保在病毒组装过程中摄入 VLP。虽然使用 VLP 作为载体将 Cas9 蛋白引入细胞并不是什么新鲜事，但作者能够生产包含 Cas9 RNP 的 VLP。与之前使用 VLP 作为 Cas9 载体的报告相比，⁸这种方法导致 VLP 包含 Cas9 机制和所需的 RNA（即引导 RNA 和反式激活 CRISPR RNA），因此不需要单独引入靶向引导 RNA。此外，这项新技术还允许同时引入感兴趣的转基因，例如肿瘤特异性 TCR 或 CAR。值得注意的是，通过带有荧光标记的自切割肽标记转基因，表达转基因的细胞可以很容易地通过例如荧光细胞分选来富集。最后，该方法的独特之处还在于能够靶向混合群体中的特定细胞亚群，至少在体外. 通过利用已被 CD4 嗜性 HIV-1 的包膜 (Env) 蛋白假型化的 VLP，他们能够生产 VLP，以高特异性有效感染 CD4 ⁺ T 细胞：大约 50% 的 CD4 ⁺ T细胞表达荧光报告基因，而来自同一孔的 CD8 ⁺ T 细胞仅表达约 2.5% 。

目前，配备有 CAR 的 T 细胞正被用于治疗多种血癌。⁹虽然这种类型的疗法对血癌非常有效，但 CAR T 细胞对实体瘤的治疗效果较差。因此，许多团体试图通过利用 CRISPR/Cas9 介导的基因组编辑来增强 T 细胞效应器功能，以最大限度地发挥 T 细胞杀伤潜力。从通过调节 TCR 或共刺激信号直接增加细胞因子产生到微调转录后调节，或通过简单地敲除抑制性受体，如 PD- ^1、4其中许多方法采用两步过程，需要顺序基因组编辑和肿瘤抗原特异性 CAR 或 TCR 的引入。相反，来自 Doudna 实验室的新方法能够同时进行。⁵最近，进行了 CRISPR/Cas9 生成的 PD-1 敲除 NY-ESO-1 TCR 转基因 T 细胞的首次人体临床试验，¹⁰确实需要随后的电穿孔和慢病毒转导步骤来分别引入 PD-1 敲除和转基因 TCR。VLP 介导的 Cas9 机制和感兴趣的转基因的传递可以显着简化和优化治疗性 T 细胞产品的生产。最后，通过靶向 β2 微球蛋白和/或 T 细胞受体 α 链 (TRAC)，如 Hamilton等人的例子，⁵ VLP 也可用于生产“现成”(CAR) T 细胞产品，用于过继疗法。⁴

值得注意的是，汉密尔顿等人的方法。导致基因组编辑机制组件仅短暂存在，从而限制了由于 Cas9 转基因插入时持续表达而导致长期脱靶效应的风险。与 Cas9 基因和引导 RNA 的慢病毒或逆转录病毒递送相比，这是一个巨大的好处，因为基因组编辑机制的组成型表达会随着时间的推移导致脱靶效应。虽然这可以通过开/关可切换或特定于细胞的启动子来缓解，但首选瞬时 Cas9 存在。此外，已经表明大多数健康个体拥有针对 Cas9 蛋白的抗体，¹¹ 进一步强调瞬时 Cas9 表达和/或存在的重要性，尤其是在治疗环境中使用最终产品时。

汉密尔顿等人。¹²利用了 HIV-1 的细胞趋向性，但这可以通过用其他病毒包膜蛋白假型 VLP 来扩展，例如，用于靶向神经细胞的狂犬病病毒的包膜蛋白，或用于靶向气道上皮细胞的流感病毒。尽管如此，CRISPR VLPs对 CD4 ⁺ T 细胞的特异性靶向具有广泛的应用（图 1b），从CAR T 细胞的（体内）生产，其中 Cas9 介导的 TRAC 基因敲除可以稳定转基因表达，⁴到渲染CD4 ⁺HIV-1 感染者的 T 细胞通过敲除 CCR5 和/或用不允许细胞进入的 CCR5Δ 突变体替换它来抵抗感染¹³，以纠正影响 T 细胞效应器功能的先天错误¹⁴通过敲除有缺陷的分子并用功能性副本替换它们。

值得注意的是，与电穿孔等传统方法相比，Cas9-VLP 的 HIV-1 伪型版本导致原代人类 T 细胞的基因组编辑频率较低（约 14-28%）。^{6, 7}通过使用最近描述的更有效的 Cas9 分子，这可能会增加，¹⁵但这仍有待确定。此外，CD4 嗜性 HIV-1 毒株也已被证明会感染巨噬细胞，应仔细考虑这一点，尤其是在实施该工具进行体内基因编辑之前。未来对人源化小鼠的研究可以阐明Cas9-VLP 介导的基因组编辑的体内安全性和可行性。尽管如此，汉密尔顿等人的研究. 为 CRISPR 介导的基因组编辑的治疗用途提供了一个新的角度，并为未来的研究提供了坚实的基础。