恶性肿瘤是一种复杂的系统性疾病，严重危害人类的生命健康。免疫治疗作为21世纪肿瘤治疗领域的重大突破，彻底改变了传统治疗的范式。在这场医学革命中，过继细胞治疗（ACT）通过基因工程手段赋予免疫细胞更强的抗肿瘤能力，开启了精准靶向治疗的新纪元。其中，以嵌合抗原受体T细胞（CAR-T）治疗为主要着力点的过继细胞疗法已经开发到第五代，全球已有12款CAR-T产品获批用于治疗血液系统恶性肿瘤，并获得显著的临床疗效。与此同时，CAR-T疗法在实体瘤治疗中的应用也成为一个新的研究方向，但实体瘤复杂的免疫抑制性微环境和周围致密的细胞外基质屏障，限制了CAR-T细胞运输和浸润到肿瘤内部的能力，细胞易耗竭进而降低其效应功能，使得CAR-T疗法在此领域的进展相对滞后。因而，亟需在固有免疫细胞中开发新的工程化抗肿瘤免疫细胞，以弥补CAR-T细胞在这些方面的不足。巨噬细胞是肿瘤微环境（TME）中重要的固有免疫细胞，在实体瘤中的浸润比例最高可达50%，会被持续招募至TME中长期存活，并且具有高度的可塑性和强大的吞噬能力，可将肿瘤抗原提呈给T细胞，促进其活化并启动适应性免疫应答，从而提供持久的抗肿瘤作用。巨噬细胞的这些优势，使其成为理想的能够负载CAR分子的固有免疫细胞，目前已有许多研究报道将CAR基因编辑到巨噬细胞中以提升安全性，突破实体瘤免疫抑制屏障。

巨噬细胞具有高度的异质性，极易受TME的影响发生极化，在TME中的巨噬细胞大部分为促进肿瘤进展的M2型巨噬细胞（M1为抗肿瘤亚型），因而目前关于CAR-M的设计理念大多集中于通过CAR的信号转导使促肿瘤巨噬细胞重编程为抗肿瘤巨噬细胞，以增强其抗肿瘤效应功能，而靶向巨噬细胞自身的免疫抑制基因进行基因编辑的研究较少。蛋白酶3（PRTN3）是丝氨酸蛋白酶家族的一员，主要储存在中性粒细胞的嗜天青颗粒中，在免疫防御中发挥关键作用。我们课题组前期研究发现，在肿瘤相关巨噬细胞中PRTN3是维持其M2表型的关键分子，通过免疫抑制作用，阻止其抗肿瘤免疫反应，推动肿瘤的进展。基于PRTN3对巨噬细胞的免疫调控作用，本研究利用敲降PRTN3的巨噬细胞作为效应细胞构建CAR-M，以增强CAR-M的抗肿瘤能力，并进一步阐明其中的作用机制。

首先，本研究成功构建了稳定敲降PRTN3的小鼠单核巨噬细胞系RAW264.7（RAW264.7-shPRTN3），在shPRTN3细胞中M1表型相关基因的表达显著上调，M2表型相关基因的表达显著下调，这表明敲降PRTN3后巨噬细胞向M1表型极化。在被肿瘤细胞激活后shPRTN3细胞中M1表型相关分子CD80的表达上升，M2表型相关分子CD206的表达下降，同时，在活化状态下促炎基因Il-6、Il-1β和Tnf-α的表达也显著上升，说明下调PRTN3可以促进巨噬细胞向促炎表型转变。在体外吞噬和杀伤实验中，shPRTN3细胞对肿瘤细胞也表现出更强的吞噬和杀伤效果。在小鼠乳腺癌原位移植模型中，输注RAW264.7-shPRTN3细胞治疗可以显著延缓小鼠乳腺癌的生长和转移。

接下来，本研究在shNC和shPRTN3细胞的基础上成功构建了靶向HER2阳性肿瘤细胞的嵌合抗原受体巨噬细胞（CAR-M），以下称为CAR-M-shNC和CAR-M-shPRTN3细胞。通过流式细胞术确认CAR分子在细胞膜成功表达后，进一步证明了CAR-M-shNC和CAR-M-shPRTN3可以被HER2抗原特异性激活。随后，本研究也检测了CAR-M的吞噬能力，相较于Ctrl-shNC和Ctrl-shPRTN3，CAR-M-shNC、CAR-M-shPRTN3均能有效吞噬HER2阳性的肿瘤细胞，且CAR-M-shPRTN3表现出更强的吞噬能力。然而在体外杀伤实验中，CAR-M-shNC的靶向杀伤效果更强。在与4T1-HER2细胞共培养后，CAR-M中促炎标志物基因的表达显著上升，CAR-M-shPRTN3细胞在激活状态下与抗肿瘤免疫应答、细胞因子的产生及炎症相关信号通路富集。上述结果表明，CAR-M-shPRTN3在体外可通过直接的吞噬作用、分泌促炎因子有效杀伤肿瘤细胞。

最后，本研究构建了小鼠乳腺癌原位荷瘤模型以评估CAR-M-shNC和CAR-M-shPRTN3的治疗效果，结果表明，Ctrl-shPRTN3、CAR-M-shNC和CAR-M-shPRTN3细胞均显著抑制肿瘤生长，其中CAR-M-shPRTN3治疗组抑制肿瘤进展的趋势最明显。同时，相较于对照组，CAR-M组中肿瘤的肺转移显著减少，且CAR-M-shPRTN3治疗组中几乎没有肺转移灶的出现，说明CAR-M中PRTN3介导的免疫调控可以有效控制肿瘤的远处转移。最后，本研究利用相同的小鼠模型评估了CAR-M-shNC、CAR-M-shPRTN3分别与抗PD-1抗体联合治疗的效果，与输注PBS治疗的荷瘤小鼠相比，抗PD-1抗体联合CAR-M-shPRTN3治疗可以发挥最强的肿瘤抑制效果。

综上所述，本研究探索了PRTN3对巨噬细胞表型极化和抗肿瘤功能的影响，并在此基础上构建了靶向HER2阳性肿瘤细胞的新型CAR-M，分析了PRTN3对CAR-M表型、吞噬、杀伤及体内抗肿瘤功能的调控作用，阐明了靶向抑制PRTN3有望成为基于巨噬细胞抗肿瘤免疫治疗的有效策略，为工程化改造巨噬细胞提供有价值的实验基础，为实体瘤免疫治疗提供新的策略和思路。

Malignant tumors are complex systemic diseases that seriously endanger human life and health. Immunotherapy, as a major breakthrough in the field of oncotherapy in the 21st century, has completely changed the paradigm of traditional treatment. In this medical revolution, adoptive cell therapy (ACT) has opened a new era of precision targeted therapy by endowing immune cells with stronger anti-tumor ability through genetic engineering means. Among them, chimeric antigen receptor-T cell (CAR-T) therapy has been developed into the fifth generation, and 12 CAR-T products have been approved for the treatment of hematologic malignancies worldwide, and have achieved remarkable clinical efficacy. Meanwhile, the application of CAR-T therapy in the treatment of solid tumors has also become a new research direction. However, the complex immunosuppressive microenvironment and the surrounding dense extracellular matrix barrier of solid tumors limit the ability of CAR-T cells to transport and infiltrate into the tumor, and the cells are easy to be depleted and thus reduce their effector function, which makes the progress of CAR-T therapy in this field relatively lagging behind. Thus, there is an urgent need to develop new engineered anti-tumor immune cells in innate immune cells to compensate for the deficiencies of CAR-T cells in these areas. Macrophages are important innate immune cells in the tumor microenvironment (TME), with an infiltration ratio of up to 50% in solid tumors. They are continuously recruited into the TME for long-term survival, and have high plasticity and strong phagocytosis, which can present tumor antigens to T cells, promote their activation and initiate adaptive immune responses, thus providing a long-lasting anti-tumor effect. These advantages of macrophages make them ideal innate immune cells capable of loading CAR molecules. At present, many studies have been reported on the editing of CAR genes into macrophages to enhance safety and break through the immunosuppressive barrier in solid tumors.

Macrophages are highly heterogeneous and highly susceptible to polarization under the influence of the TME. Most of the macrophages in TME are M2-type macrophages that promote tumor progression (M1 is the anti-tumor subtype). Therefore, most of the current design concepts of CAR-M focus on reprogramming tumor-promoting macrophages into anti-tumor macrophages through CAR signal transduction to enhance their anti-tumor effector function, while there are fewer studies targeting the autoimmune suppressive genes of macrophage for gene editing. Protease 3 (PRTN3), a member of the serine protease family, is mainly stored in the azurophilic granules of neutrophils and plays a key role in immune defense. Our group's previous study found that PRTN3 is a key molecule in tumor-associated macrophages to maintain their M2 phenotype, which drives tumor progression by preventing their anti-tumor immune response through immunosuppressive effects. Based on the immunoregulatory effect of PRTN3 on macrophages, this study utilized macrophages with PRTN3-knockdown as effector cells to construct CAR-M in order to enhance the anti-tumor ability of CAR-M and further elucidate the mechanism of action.

Firstly, in this study, we successfully constructed a mouse monocyte macrophage cell line RAW264.7 with stable knockdown of PRTN3 (RAW264.7-shPRTN3). The expression of M1 phenotype-related genes was significantly up-regulated and M2 phenotype-related genes were significantly down-regulated in shPRTN3 cells, which indicated that macrophages were polarized towards M1 phenotype after knockdown of PRTN3. The expression of the M1 phenotype-related molecule CD80 increased and the expression of the M2 phenotype-related molecule CD206 decreased in shPRTN3 cells after activation by tumor cells, while the expression of the pro-inflammatory genes Il-6, Il-1β and Tnf-α also increased significantly in shPRTN3 cells in the activated state, suggesting that the down-regulation of PRTN3 promotes the shift of macrophages toward a pro-inflammatory phenotype. In in vitro phagocytosis and killing assays, shPRTN3 cells also demonstrated stronger phagocytosis and killing effects on tumor cells. In an orthotopic transplantation model of breast cancer in mice, infusion of RAW264.7-shPRTN3 cells treatment significantly delayed the breast cancer growth and metastasis.

Next, in this study, chimeric antigen receptor macrophages (CAR-M) targeting HER2-positive tumor cells were successfully constructed on the basis of shNC and shPRTN3 cells, hereinafter referred to as CAR-M-shNC and CAR-M-shPRTN3 cells. After confirming that CAR molecules were successfully expressed at the cell membrane by flow cytometry, we further demonstrated that CAR-M -shNC and CAR-M-shPRTN3 could be specifically activated by HER2 antigen. Subsequently, we also examined the phagocytosis ability of CAR-M, and compared with Ctrl-shNC and Ctrl-shPRTN3, both CAR-M-shNC and CAR-M-shPRTN3 were able to effectively phagocytose HER2-positive tumor cells, and CAR-M-shPRTN3 showed stronger phagocytosis ability. However, in the in vitro killing assay, CAR-M-shNC had a stronger targeted killing effect. After co-culture with 4T1-HER2 cells, the expression of pro-inflammatory marker genes in CAR-M was significantly increased. The anti-tumor immune response, cytokine production, and inflammation-related signaling pathways were enriched in CAR-M-shPRTN3 cells in activated state. The above results suggest that CAR-M-shPRTN3 can effectively kill tumor cells in vitro through direct phagocytosis and secretion of pro-inflammatory factors.

Finally, in this study, we constructed in situ tumor-bearing model of mouse breast cancer to evaluate the therapeutic effects of CAR-M-shNC and CAR-M-shPRTN3. The results showed that Ctrl-shPRTN3, CAR-M-shNC, and CAR-M-shPRTN3 significantly inhibited the tumor growth, and the trend of inhibiting tumor progression in the CAR-M-shPRTN3 treatment group was the most obvious. Meanwhile, compared with the control group, the lung metastasis of the tumor was significantly reduced in the CAR-M groups, and there was almost no lung metastasis in the CAR-M-shPRTN3 group, suggesting that PRTN3-mediated immunomodulation in CAR-M can effectively control the distant metastasis of the tumor. Finally, this study evaluated the effects of CAR-M-shNC and CAR-M-shPRTN3 in combination with anti-PD-1 antibody, respectively using the same mouse model, and the anti-PD-1 antibody combined with CAR-M-shPRTN3 treatment could exert the strongest tumor suppression effect compared with that of the PBS-infused treated tumor bearing mice.

In summary, this study explored the effects of PRTN3 on macrophage phenotypic polarization and anti-tumor function, and constructed a novel CAR-M targeting HER2-positive tumor cells on this basis, analyzed the regulatory effects of PRTN3 on the phenotype, phagocytosis, tumor killing and in vivo anti-tumor function of CAR-M. This study elucidated that the targeting inhibition of PRTN3 is expected to be an effective strategy for anti-tumor immunotherapy based on macrophage, providing a valuable experimental basis for engineering the transformation of macrophages, and providing new strategies and ideas for immunotherapy of solid tumors.