RBPJ

这项研究揭示了肿瘤内CTLs分化和功能的多样性及关键的调控转录因子，并为整合CTLs细胞命运基因调控网络和改善肿瘤免疫反应的潜在靶点提供了一个系统框架。 美国圣裘德儿童研究医院迟洪波团队在 Nature 期刊发表了题为：Single-cell CRISPR screens in vivo map T cell fate regulomes in cancer 的研究论文。 该研究使用单细胞CRIPSR筛选，在体内绘制了肿瘤T细胞命运调控网络。揭示了促使耗竭T细胞前体/祖细胞（Tpex）细胞退出静息状态以及增强终末耗竭T细胞（Tex）细胞增殖状态的关键调控转录因子，并用以改善肿瘤免疫疗法。 在迟洪波教授的指导下，周培培、史豪、黄宏龄等博士在CTLs中利用单细胞CRIPSR筛选（scRNA-seq）系统性绘制了CTLs基因调控网络并发现了其分化的检查点及关键的转录调控因子（IKAROS–TCF-1、ETS1–BATF和RBPJ–IRF1）。 首先，他们发现Tpex细胞退出静息状态并连续分化为中间型Tex1细胞（有较强的功能和增殖能力的一群Tex细胞），这一过程由IKAROS和ETS1差异调控。IKAROS促进Tpex1（有较弱增殖能力的一群Tpex细胞）的代谢激活及其分化为Tpex2细胞（有较强增殖能力的一群Tpex细胞）。 敲除IKAROS抑制了CTLs的功能，增加了肿瘤内CTLs的干细胞性的同时降低了细胞代谢以及mTORC1信号通路的活性，暗示IKAROS缺陷可能使细胞滞留在过度静止的状态。并且，敲除IKAROS导致的CTL细胞积累无法单独或响应ICB改善抗肿瘤免疫反应。 相反，敲除ETS1通过增强mTORC1信号通路活性和代谢重编程来调控Tpex2到Tex1分化过程，并且在多个肿瘤模型中增强了过继细胞疗法（ACT）和免疫检查点阻断疗法（ICB）抗肿瘤效果，并且ETS1基因表达量与癌症患者对ICB的响应呈负相关。 在机制上，他们揭示了TCF-1和BATF可以分别被IKAROS和ETS1负向调控，是IKAROS和ETS1在CTL细胞中的靶点。 肿瘤内CTLs的体内scCRISPR筛选揭示协同功能模块和基因程序的连通性 其次，他们发现敲除RBPJ阻断了中间型Tex1向终末Tex2细胞（有较弱的功能和增殖能力的一群Tex细胞）分化，同时通过增加Tex细胞的增殖能力来扩大Tex1细胞累积。并且，敲除RBPJ在多个肿瘤模型中改善了ACT和ICB的抗肿瘤效果。另外，RBPJ基因的表达量与来自癌症患者CTLs的终末Tex分化以及癌症患者对免疫检查点阻滞疗法的低反应性正相关。 在机制上， RBPJ通过抑制IRF1转录因子的活性来促进了中间型Tex1到终末型Tex2细胞的分化。因此，敲除RBPJ通过增强有较强功能和增殖能力的Tex1细胞的分化来改善ACT和ICB的抗肿瘤效果。 总的来说，这项研究揭示了肿瘤内CTLs分化和功能的多样性及关键的调控转录因子，并为整合CTLs细胞命运基因调控网络和改善肿瘤免疫反应的潜在靶点提供了一个系统框架。

Scientists mapped the gene regulatory networks responsible for progressive differentiation states of tumor-infiltrating T cells, using a technology known as single-cell CRISPR screening to knock out or genetically perturb multiple genes potentially involved. T cells are the key immune system component involved in killing cancer. Tumors produce signals that turn off these T cells, partially by forcing them to progressively differentiate (mature) into a hypofunctional state known as exhaustion. Scientists from St. Jude Children's Research Hospital comprehensively examined the transcription factors involved in T-cell differentiation states in cancer. They then used this information to enhance anticancer activity in preclinical models by promoting or blocking T-cell differentiation. The findings, which have implications for cancer immunotherapy, were published today in Nature. T cells are used in adoptive cell therapy (ACT) to target and kill cancer cells. Among ACTs, chimeric antigen receptor (CAR) T cells have shown clinical efficacy in blood cancers but have not been as effective in solid tumors. This variance in efficacy is partly because tumors promote T-cell exhaustion, in which the cells are less effective at actively killing cancer. The scientists showed they could precisely interrupt the flow of the differentiation process, thereby enhancing antitumor efficacy. "T cells are the cornerstone of tumor immunotherapy, and we found a new way to reprogram T cells to make them more effective," said Hongbo Chi, Ph.D., St. Jude Department of Immunology. "We can push them towards a specialized state so that they can be more functional tumor-killing cells." Modifying differentiation to flow toward cancer-killing T cells T-cell differentiation is conceptually similar to a river running down a mountain. The top of the mountain is the precursor cell, and the bottom an exhausted cell. The desired state is between these two extremes -- like having a lake in the middle of the mountain -- that is differentiated enough, but not too much, to have high proliferation and effective anticancer activity. The lake receives water from above but also has a stream constantly draining it further down the slope. Uninterrupted, all water in the lake will eventually drain through the river leading down below -- all the T cells will become exhausted. The scientists found a way to increase the water flow from above and dam the river draining the lake, resulting in more water accumulation in the lake or, more aptly, more T cells that effectively kill tumors. For example, when the researchers deleted the transcription factor ETS proto-oncogene 1 (ETS1), the amount of cancer-killing T cells was increased by enhancing differentiation to the intermediate state (increasing flow into the lake). On the other hand, knocking out the transcription factor recombination signal binding protein for immunoglobulin kappa J region (RBPJ) increased the accumulation of intermediate state T cells by blocking terminal differentiation (damming the river at the lake). Each approach showed markedly increased antitumor activity. Even more promising, when the researchers combined their modified T cells with a type of immunotherapy called immune checkpoint blockade, they further increased antitumor activity. "We've provided new potential strategies to enhance immunotherapy," said co-first author Peipei Zhou, Ph.D., St. Jude Department of Immunology. "We increased the amount of cancer-killing T cells by differentiating them from upstream precursors, and we could also block terminal differentiation to increase accumulation of the functional cells. Both approaches improved antitumor efficacy in our immune checkpoint blockade models." While the work was done in mouse models, the researchers used bioinformatics data from human clinical samples to verify that these same genes are important in these processes in humans. Mapping the differentiation journey of exhausted T cells T cells infiltrate tumors to go about their work of destroying them. The researchers mapped the transcription factors expressed in the T cells infiltrating tumors. Their work showed which individual transcription factors and their networks are responsible for each T cell differentiation state, providing targets for improving immunotherapy. "We created a map for the field of T-cell differentiation and exhaustion in tumors based on the loss-of-function genetic screening at high dimension," said co-first author Hao Shi, Ph.D., St. Jude Department of Immunology. "This map will provide a guide for future researchers to refer to and use to identify ways to improve T cell-based immunotherapies." Where past attempts to understand T-cell exhaustion focused on an individual molecule or transcription factor, this screen was more comprehensive. This allowed the researchers to document the entirety of the gene regulatory networks involved. "We've provided the most complete picture of the transcription factors in T cells inside a tumor ever created," Zhou said. "Such a causal transcriptional network provides new insights into lineage differentiation, a fundamental process in biology," added Shi. The group performed this feat using single-cell CRISPR-Cas9 screening, a gene editing technology that analyzes gene expression profiling of individual cells after selectively removing transcription factors in a comprehensive screen. Transcription factors are proteins that directly regulate gene expression. By examining an individual T cell's gene expression pattern, the researchers could compare them to see which knocked-out transcription factors most affected T-cell differentiation and anticancer activity. "The same approach may be more broadly applicable to increase our knowledge in many biological contexts beyond T cells and immunology," Chi said. "We showed this new technology, single-cell CRISPR screening in vivo, can drive novel biological discovery. I think we're at the advent of something truly exciting."

单细胞转录组测序数据显示胆管上皮细胞分化产生的肝细胞相较于肝脏祖细胞上调表达WNT/β-catenin信号通路相关基因，提示WNT/β-catenin信号通路调控胆管上皮细胞向肝细胞的分化。 中国科学院分子细胞科学卓越创新中心（生物化学与细胞生物学研究所）周斌研究组和诺华生物医学研究所Jan S. Tchorz研究组合作，在 Nature 子刊 Nature Genetics 上以封面故事发表了题为：Bipotent transitional liver progenitor cells contribute to liver regeneration 的研究论文。 该研究在人类多种肝脏疾病和小鼠肝损伤模型中鉴定出一群同时表达胆管上皮细胞和肝细胞特征的肝脏祖细胞；发现这群肝脏祖细胞起源于胆管上皮细胞；利用双同源重组示踪技术发现这群肝脏祖细胞具有向胆管上皮细胞和肝细胞分化的能力；通过基因敲除及过表达发现Notch和WNT/β-catenin信号通路分别调控胆管上皮细胞向肝脏祖细胞、肝脏祖细胞向肝细胞的分化。 成体肝稳态中，大部分肝细胞处于静息状态，增殖缓慢；当肝脏发生损伤后，肝细胞重新进入细胞周期，参与损伤修复，维持肝脏功能。近年来，越来越多的实验证明在肝脏损伤中，肝细胞具有显著的再生能力，肝损伤主要通过肝细胞的自我增殖进行损伤修复，而肝祖细胞对肝细胞的贡献几乎可以忽略。然而，在肝细胞中进行β1-integrin基因敲除或者p21基因过表达来抑制肝细胞的增殖，同时利用谱系示踪技术标记胆管上皮细胞，在肝损伤模型中直接观察到胆管上皮细胞转分化为肝细胞，促进肝脏损伤修复。 研究人员通过给予小鼠长期喂养TAA等药物来诱导慢性肝损伤，也观察到在慢性肝损伤过程中胆管上皮细胞可以转分化为肝细胞参与肝脏再生。随后，多个不同研究组独立证明在肝细胞增殖受阻的肝损伤条件下，胆管上皮细胞可以贡献给肝细胞参与肝脏损伤修复。以往的研究中由于诱导胆管上皮细胞向肝细胞转变的效率太低，很难进行细胞和分子机制的深入研究。关于胆管上皮细胞向肝细胞分化是否经历肝脏祖细胞状态这一细胞机制及胆管上皮细胞向肝细胞分化的分子机制尚未明确。 在该研究中，研究人员首先构建了高效稳定的胆管上皮细胞向肝细胞转分化的遗传学模型，为胆管上皮细胞向肝细胞转分化的细胞及分子机制研究奠定基础。研究人员在Fah基因中插入loxp-stop-loxp序列，构建了Fah-loxp-stop-loxp（Fah-LSL）小鼠，阻止Fah基因的正常表达。Fah-LSL小鼠与组织特异性CreER小鼠结合可以在特定类型细胞中发生Cre-loxp同源重组，切除stop序列，使细胞恢复表达Fah基因的能力。对CK19-CreER;R26-tdT;Fah-LSL/Fah-LSL小鼠进行他莫昔芬诱导，可将胆管上皮细胞及其子代细胞标记为tdT+，同时在胆管上皮细胞及其子代细胞中将Fah基因中的stop序列切除，使胆管上皮细胞及其子代细胞恢复表达Fah基因的能力。之后对小鼠停止给予NTBC药物，肝脏呈现严重损伤状态，刺激胆管上皮细胞向肝细胞分化。该遗传学模型从基因组水平进行全身性Fah基因敲除，不存在抑制肝细胞增殖的效率问题，因为每个肝细胞都是缺乏Fah基因的，这样能够高效稳定地诱导具有表达Fah基因能力的胆管上皮细胞向肝细胞转分化，为研究转分化的细胞和分子机制奠定基础。 在Fah-LSL/Fah-LSL小鼠诱导的肝损伤中，研究人员对Epcam+细胞进行单细胞转录组测序，结果显示Epcam+细胞群分为不同亚群；一群细胞高表达细胞增殖相关基因（Cdk1），定义为增殖的胆管上皮细胞群（Proliferating BECs）；一群细胞同时表达胆管上皮细胞分子标记（Krt19、Epcam）和肝细胞分子标记（HNF4a、Ttr），定义为肝脏祖细胞群（LPCs）。将上述单细胞转录组测序数据与之前报道的2组DDC诱导的肝损伤中Epcam+细胞的单细胞转录组测序数据进行整合分析，研究人员未在DDC诱导的肝损伤中检测到肝脏祖细胞；也未在正常肝脏中检测到肝脏祖细胞。同时，研究人员通过免疫荧光染色实验在Fah-LSL/Fah-LSL小鼠诱导的肝损伤中检测到CK19+HNF4a+的肝祖细胞，并且这些细胞也表达Epcam、A6、OPN等肝脏祖细胞分子标记；而DDC诱导的肝损伤中未检测到CK19+HNF4a+的肝祖细胞（图1）。此外，研究人员在多种人类肝脏疾病包括非酒精性脂肪性肝炎和病毒性肝炎等中发现肝细胞呈现衰老状态，并检测到CK19+HNF4a+肝祖细胞，发现肝细胞的衰老程度与肝祖细胞的产生呈现正相关联系；并且在这些人类肝脏疾病中，胆管上皮细胞未呈现衰老状态，提示在人的肝脏疾病中，胆管上皮细胞可能作为肝脏祖细胞的来源贡献给肝细胞而发挥重要作用。 图1：小鼠和人类肝脏LPC的鉴定 研究人员对单细胞转录组测序数据进行拟时序分析发现肝脏祖细胞来源于胆管上皮细胞；之后利用CK19-CreER工具小鼠标记并示踪胆管上皮细胞，揭示胆管上皮细胞可以贡献给肝脏祖细胞。为了研究肝脏祖细胞的分化潜能，研究人员利用双同源重组酶介导的谱系示踪技术来标记肝脏祖细胞（图2），示踪结果揭示肝脏祖细胞具有向胆管上皮细胞和肝细胞分化的潜能。值得注意的是，肝脏祖细胞不具有增殖活性，并且单个肝脏祖细胞不能同时产生胆管上皮细胞和肝细胞。 图2：利用双重组酶特异性标记肝脏祖细胞 单细胞转录组测序数据显示肝脏祖细胞群相较于胆管上皮细胞群下调表达Hippo信号通路相关基因和Notch信号通路相关基因，提示Notch信号通路调控胆管上皮细胞向肝脏祖细胞的分化。研究人员在胆管上皮细胞中进行Rbpj基因敲除，发现抑制Notch信号通路能够促进胆管上皮细胞向肝脏祖细胞分化，从而促进胆管上皮细胞向肝细胞分化（图3）。此外，研究人员利用Dibenzazepine药物来抑制Notch信号通路也证实抑制Notch信号通路能够促进胆管上皮细胞向肝脏祖细胞分化，从而促进胆管上皮细胞向肝细胞分化。当在胆管上皮细胞中激活Notch信号通路后，则完全抑制了胆管上皮细胞向肝脏祖细胞的分化，阻碍了肝脏再生。 图3：Notch信号通路抑制促进BEC形成TLPC