Repurposing Erythrocytes as a “Photoactivatable Bomb”: A General Strategy for Site-Specific Drug Release in Blood Vessels
将红细胞重新用作“光激活炸弹”：血管中特定位点药物释放的一般策略
首次发布：2021 年 7 月 14 日 https://doi.org/10.1002/smll.202100753 引用：11
Abstract 抽象的
Tumor vasculature has long been considered as an extremely valuable therapeutic target for cancer therapy, but how to realize controlled and site-specific drug release in tumor blood vessels remains a huge challenge. Despite the widespread use of nanomaterials in constructing drug delivery systems, they are suboptimal in principle for meeting this demand due to their easy blood cell adsorption/internalization and short lifetime in the systemic circulation. Here, natural red blood cells (RBCs) are repurposed as a remote-controllable drug vehicle, which retains RBC's morphology and vessel-specific biodistribution pattern, by installing photoactivatable molecular triggers on the RBC membrane via covalent conjugation with a finely tuned modification density. The molecular triggers can burst the RBC vehicle under short and mild laser irradiation, leading to a complete and site-specific release of its payloads. This cell-based vehicle is generalized by loading different therapeutic agents including macromolecular thrombin, a blood clotting-inducing enzyme, and a small-molecule hypoxia-activatable chemodrug, tirapazamine. In vivo results demonstrate that the repurposed “anticancer RBCs” exhibit long-term stability in systemic circulation but, when tumors receive laser irradiation, precisely releases their cargoes in tumor vessels for thrombosis-induced starvation therapy and local deoxygenation-enhanced chemotherapy. This study proposes a general strategy for blood vessel-specific drug delivery.
肿瘤血管系统长期以来被认为是癌症治疗极其有价值的治疗靶点，但如何在肿瘤血管中实现受控和位点特异性药物释放仍然是一个巨大的挑战。尽管纳米材料在构建药物输送系统中得到广泛应用，但由于其易于血细胞吸附/内化且在体循环中的寿命较短，因此原则上它们不能满足这一需求。在这里，天然红细胞 (RBC) 被重新用作远程控制的药物载体，通过以精细调整的修饰密度共价结合在红细胞膜上安装光激活分子触发器，保留红细胞的形态和血管特异性生物分布模式。分子触发器可以在短而温和的激光照射下爆裂红细胞载体，从而导致其有效载荷在特定位置完全释放。这种基于细胞的载体通过加载不同的治疗剂而得到推广，包括大分子凝血酶、凝血诱导酶和小分子缺氧激活化疗药物替拉扎明。体内结果表明，重新利用的“抗癌红细胞”在体循环中表现出长期稳定性，但当肿瘤接受激光照射时，它们会精确地释放肿瘤血管中的货物，用于血栓引起的饥饿治疗和局部脱氧增强化疗。这项研究提出了血管特异性药物输送的一般策略。
1 Introduction 1 简介
Since Folkman first proposed the therapeutic implications of anti-angiogenesis in 1971,[1] tumor vasculature has been considered as an extremely valuable target for cancer therapy,[2-12] because intact vessels support the rapid proliferation of tumor cells by delivering oxygen and nutrients, and provide a major pathway for metastatic spreading.[13] Currently, developed tumor devascularization strategies can primarily be categorized into two major classes. The first class employs angiogenesis inhibitors (e.g., bevacizumab and pegaptinib)[14, 15] or vascular disrupting agents[13] to choke off the local blood supply for tumor starvation, but this method is often criticized by some long-term side effects and the easy occurrence of drug resistance.[16, 17] The other class involves the use of various embolic agents, such as gelatin sponge and polyvinyl alcohol particles, to directly block tumor blood vessels and therefore starve tumor cells.[18] Compared with the first strategy, tumor-specific vessel embolization has the potential to minimize systemic toxicity and the possibility of developing drug resistance. Embolization therapy in clinical settings relies on an implanted microcatheter to deliver embolic agents to the tumor-feeding artery under the guidance of imaging equipment.[19-22] Precise as it is, this technique requires additional surgical operation and is only limited to specific types of tumors such as hepatocellular carcinoma. Therefore, how to achieve controlled and site-specific release of embolic agents in a non-invasive manner remains a tremendous technical challenge.
自 1971 年 Folkman 首次提出抗血管生成的治疗意义以来， 1 肿瘤血管系统已被认为是癌症治疗极其有价值的靶点， 2-12 因为完整的血管支持肿瘤血管的快速增殖。肿瘤细胞通过输送氧气和营养物质，并提供转移扩散的主要途径。 13 目前，已开发的肿瘤断血管策略主要可分为两大类。第一类采用血管生成抑制剂（如贝伐珠单抗、培加替尼） 14, 15 或血管破坏剂 13 来阻断局部血液供应，使肿瘤饥饿，但这种方法经常受到一些人的批评长期副作用且易产生耐药性。 16, 17 另一类是利用各种栓塞剂，如明胶海绵、聚乙烯醇颗粒等，直接阻塞肿瘤血管，从而饿死肿瘤细胞。 18 与第一种策略相比，肿瘤特异性血管栓塞有可能最大限度地减少全身毒性和产生耐药性的可能性。临床环境中的栓塞治疗依靠植入的微导管在成像设备的引导下将栓塞剂输送到肿瘤供血动脉。 19-22 虽然很精确，但该技术需要额外的外科手术，并且仅限于特定类型的肿瘤，例如肝细胞癌。因此，如何以非侵入的方式实现栓塞剂的受控、定点释放仍然是一个巨大的技术挑战。
In these years, researchers have sought to tackle this issue by constructing nanoscale drug delivery systems (DDSs) with good tumor selectivity and/or stimulus responsiveness.[23-29] These systems are mostly endowed with the capability to load embolic agents and selectively release them upon reaching tumor regions. For example, Li et al. fabricated a structure-switchable DNA nanorobot for tumor vessel-targeted delivery of thrombin, a serine protease that can efficiently induce blood clotting by regulating platelet (PLT) aggregation and converting circulating fibrinogen into fibrin.[26] In addition, Mg2Si nanoparticles were reported to scavenge blood oxygen in response to the mildly acidic tumor microenvironment and generate SiO2 aggregates in situ to block tumor capillaries for cancer starvation therapy.[27] However, nanomaterials commonly suffer from short circulation time and low tumor accumulation efficiency, mainly due to the rapid clearance by the mononuclear phagocyte system. Even though a small quantity of nanomaterials reach the tumor site, they tend to extravasate from the leaky vasculature and retain inside the tumor tissue, a phenomenon known as enhanced permeability and retention (EPR) effect, which will severely impair the effectiveness of embolic agents and lead to insufficient vascular blockage. With these concerns in mind, nanomaterials may not serve as an optimal choice for constructing DDSs that need to ensure a specific and effective release of embolic agents in tumor vessels.
近年来，研究人员试图通过构建具有良好肿瘤选择性和/或刺激响应性的纳米级药物递送系统（DDS）来解决这个问题。 23-29 这些系统大多具有装载栓塞剂并在到达肿瘤区域后选择性释放它们的能力。例如，李等人。制造了一种结构可切换的 DNA 纳米机器人，用于肿瘤血管靶向递送凝血酶，凝血酶是一种丝氨酸蛋白酶，可以通过调节血小板 (PLT) 聚集并将循环纤维蛋白原转化为纤维蛋白来有效诱导血液凝固。 26 此外，Mg 2 Si 纳米颗粒据报道可响应弱酸性肿瘤微环境而清除血氧，并原位生成 SiO 2 聚集体以阻断肿瘤用于癌症饥饿疗法的毛细血管。 27 然而，纳米材料普遍存在循环时间短、肿瘤积累效率低的问题，这主要是由于单核吞噬细胞系统的快速清除所致。即使少量纳米材料到达肿瘤部位，它们也倾向于从渗漏的脉管系统中渗出并保留在肿瘤组织内，这种现象被称为增强渗透性和保留（EPR）效应，这将严重损害栓塞剂和栓塞剂的有效性。导致血管堵塞不充分。考虑到这些问题，纳米材料可能不是构建需要确保栓塞剂在肿瘤血管中特异性且有效释放的 DDS 的最佳选择。
Recently, natural cells as a valuable reservoir of biomaterials are gaining increasing attention from researchers. Compared with synthetic materials, natural cell-based DDSs are blessed with abilities to accomplish more complex tasks and adapt to native biological entities.[30-37] To date, a variety of mammalian cells (e.g., red blood cell (RBC),[38-41] PLT,[42-48] leucocyte,[49-53] cancer cell,[54, 55] stem cell,[56] and β-cell[57, 58]) and bacterial cells[59-61] have been artificially engineered as “smart” therapeutic carriers with the help of genetic, chemical, and physical modification technologies. Among them, RBCs are the most abundant cells in blood and can be easily manipulated for efficient drug loading through multiple well-established techniques.[38-41] RBC-based DDSs resemble native RBCs in many aspects, including size, morphology, surface properties, and most importantly, the long-term stability in the systemic circulation, which makes them an ideal class of materials for biomedical applications.[62] Unfortunately, how to engineer RBCs into a controllable drug vehicle for the site-specific delivery of embolic agents is still rarely explored.
近年来，天然细胞作为生物材料的宝贵储库越来越受到研究人员的关注。与合成材料相比，基于天然细胞的 DDS 具有完成更复杂任务和适应天然生物实体的能力。 30-37 迄今为止，多种哺乳动物细胞（例如红细胞 (RBC)、 38-41 PLT、 42-48 白细胞、 49-53 癌细胞、 54, 55 干细胞、 56 和β细胞 57, 58 ）和细菌细胞 59-61 已被人工改造为“智能”借助遗传、化学和物理修饰技术的治疗载体。其中，红细胞是血液中最丰富的细胞，可以通过多种成熟的技术轻松操作以实现有效的药物装载。 38-41 基于红细胞的 DDS 在许多方面类似于天然红细胞，包括尺寸、形态、表面性质，最重要的是，在体循环中的长期稳定性，这使得它们成为生物医学的理想材料应用程序。 62 不幸的是，如何将红细胞设计成可控的药物载体以用于栓塞剂的特定位点递送仍然很少被探索。
In this work, we repurpose RBCs as a general and remote-controllable DDS capable of releasing its cargoes, including macromolecular enzymes and small-molecule drugs, into blood vessels in a site-specific manner (Scheme 1). To this end, we first decorate the surface of RBCs with a photoactivatable molecular trigger, 2-(1-hexyloxyethyl)-2-devinyl pyropheophorbide-α (HPPH), via covalent modification. The chemically anchored HPPH is demonstrated to burst the modified RBCs (termed HRBCs) under short and mild laser irradiation by generating oxidative 1O2, leading to a complete and fast release of their intracellular contents. This appealing property inspires us to replace the substances (e.g., hemoglobin (HGB)) inside HRBCs with two representative therapeutic agents, that is, thrombin (Th) and tirapazamine (TPZ), through hypotonic/hypertonic treatments. The repurposed RBCs (termed Th/TPZ@HRBCs), acting as a “photoactivatable bomb”, inherit the long circulation time and excellent hemocompatibility from natural RBCs, but when the tumor region receives laser irradiation, precisely release their loaded drugs in tumor vessels. In this scenario, thrombin can rapidly induce the specific occlusion of tumor vessels for tumor deoxygenation, which then activates TPZ to kill surrounding tumor cells in a synergistic manner. We believe that this design based on natural RBCs represents a general and facile strategy for precisely delivering therapeutics in blood vessels in a highly controllable way.
在这项工作中，我们将红细胞重新利用为一种通用且远程可控的 DDS，能够以特定位点的方式将其货物（包括大分子酶和小分子药物）释放到血管中（方案 1）。为此，我们首先通过共价修饰用可光激活的分子触发器2-(1-己氧基乙基)-2-乙烯基焦脱镁叶绿酸-α (HPPH) 装饰红细胞表面。化学锚定的 HPPH 被证明在短而温和的激光照射下通过产生氧化 1 O 2 来破裂修饰的红细胞（称为 HRBC），从而导致其细胞内的完全快速释放内容。这种吸引人的特性启发我们通过低渗/高渗治疗，用凝血酶（Th）和替拉扎明（TPZ）这两种代表性治疗剂来替代HRBC内的物质（例如血红蛋白（HGB））。重新利用的红细胞（称为Th/TPZ@HRBCs）作为“光激活炸弹”，继承了天然红细胞的长循环时间和优异的血液相容性，但当肿瘤区域接受激光照射时，可以在肿瘤血管中精确释放其负载的药物。在这种情况下，凝血酶可以快速诱导肿瘤血管特异性闭塞，使肿瘤脱氧，然后激活TPZ以协同方式杀死周围的肿瘤细胞。我们相信，这种基于天然红细胞的设计代表了一种通用且简便的策略，能够以高度可控的方式在血管中精确输送治疗药物。
该方案说明了 Th/TPZ@HRBC 的制造及其在激光触发肿瘤血管阻塞和缺氧激活化疗中的应用。
2 Results and Discussion
2 结果与讨论
2.1 Characterization of HRBCs
2.1 HRBC 的表征
HRBCs were prepared according to the procedures described in the Supporting Information. The fluorescence intensities of HPPH on HRBCs displayed an increasing trend when we varied the feeding concentrations of HPPH from 2 to 20 µg mL−1, and then reached a plateau at 50 µg mL−1 (Figure 1a and Figure S1, Supporting Information), which was selected as the optimal concentration for RBC modification. The conjugation efficiency of HPPH under this condition was calculated to be 88.9%. Next, we demonstrated that HPPH is superior to some other carboxyl group-containing photosensitizers, including protoporphyrin IX (PpIX), chlorin e6 (Ce6), and rose bengal (RB), for the surface modification of RBCs, because HRBCs presented the strongest surface fluorescence signals and the lowest hemolysis rate compared with the other photosensitizer-modified RBCs (Figure S2, Supporting Information). We have also confirmed that the conjugated photosensitizers could disrupt RBC membranes in a light-controlled manner, and HRBCs exhibited nearly 100% hemolysis at 50 µg mL−1, the optimized modification concentration (Figure 1b and Figure S2b, Supporting Information), suggesting the potential of HRBCs for light-triggered drug release.
HRBC 是根据支持信息中描述的程序制备的。当我们将 HPPH 的补料浓度从 2 改变到 20 µg mL −1 时，HRBC 上 HPPH 的荧光强度呈现增加趋势，然后在 50 µg mL −1 处达到稳定水平（图 1a 和图 S1，支持信息），被选为 RBC 修饰的最佳浓度。该条件下HPPH的结合效率经计算为88.9%。接下来，我们证明 HPPH 对于红细胞的表面修饰优于其他一些含羧基光敏剂，包括原卟啉 IX (PpIX)、二氢卟酚 e6 (Ce6) 和玫瑰红 (RB)，因为 HRBC 具有最强的表面与其他光敏剂修饰的红细胞相比，荧光信号和最低的溶血率（图 S2，支持信息）。我们还证实，缀合光敏剂可以以光控方式破坏红细胞膜，并且 HRBC 在优化修饰浓度 50 µg mL −1 下表现出近 100% 溶血（图 1b 和图 S2b，支持信息），表明 HRBC 具有光触发药物释放的潜力。
HRBC 的表征和 Th@HRBC 的激光触发血栓形成。 a) 用不同浓度的 HPPH（2、5、10、20、50 和 100 µg mL −1 ）修饰后红细胞的共焦荧光图像。比例尺 = 10 µm。 b) 用不同浓度的 HPPH（2、5、10、20、50 和 100 µg mL −1 ）修饰后红细胞的光触发溶血率。光照射条件：671 nm激光，20 mW cm −2 ，60 s。 c) 重建 HRBC 的 3D 共焦荧光图像。 d) 激光照射（671 nm，20 mW cm −2 ，60 s）后红细胞、HRBC 和 HRBC 的代表性 SEM 图像。比例尺 = 5 µm。 e) RBC和HRBC的SDS-PAGE蛋白分析结果。 f) Th@HRBCs 激光触发血栓形成的工作机制示意图。 g) 经过或不经过 671 nm 激光照射（20 mW cm −2 ，60 s）预处理的 Th@HRBC 的时间依赖性凝血酶释放曲线。 h) 在裸凝血酶、Th@HRBC 或经 671 nm 激光照射（20 mW cm −2 ，60 s）。含凝血酶样品中凝血酶的剂量为0.5 U。绿色荧光聚集体表示形成的纤维蛋白-FITC。比例尺 = 20 µm。 j) 用裸凝血酶、Th@HRBCs、“Th@HRBCs + 激光（671 nm，20 mW cm −2 ，60 s）”或“Th@HRBCs + Triton X”处理后的血浆照片-100”。 使用Triton X-100作为膜破坏剂设置阳性对照组。
We then carefully characterized the surface properties of HPPH-modified RBCs. The reconstructed 3D confocal image and scanning electron microscopy (SEM) results confirmed that most HRBCs maintained the normal biconcave shape without evident morphological change (Figure 1c,d). Besides, the protein profiles of HRBCs and natural RBCs were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Figure 1e revealed that the two protein bands were similar, indicating that the native proteins on the RBC membrane were completely reserved after HPPH modification. Apart from the above hemolysis results, the membrane disruption of HRBCs after laser irradiation was further confirmed through SEM (Figure 1d). Strikingly, HRBCs still retained over 80% HPPH molecules after being stored at 4 °C for a month (Figure S3a, Supporting Information), due to the better modification stability of covalent conjugation than that of physical adsorption (Figure S3b, Supporting Information), and maintained the hemolysis levels as low as natural RBCs during the observation period (Figure S3c, Supporting Information). Despite the partial detachment of HPPH from HRBCs during the period of storage, the remaining HPPH molecules were still sufficient to induce complete HRBC hemolysis upon laser irradiation (Figure S3d, Supporting Information). These results highlight the advantage of using chemical conjugation to decorate RBC membranes with HPPH.
然后，我们仔细表征了 HPPH 修饰的红细胞的表面特性。重建的 3D 共焦图像和扫描电子显微镜 (SEM) 结果证实，大多数 HRBC 保持正常的双凹形状，没有明显的形态变化（图 1c、d）。此外，通过十二烷基硫酸钠-聚丙烯酰胺凝胶电泳（SDS-PAGE）分析HRBC和天然RBC的蛋白质谱。图1e显示两条蛋白条带相似，表明HPPH修饰后红细胞膜上的天然蛋白被完全保留。除了上述溶血结果外，通过SEM进一步证实了激光照射后HRBC的膜破裂（图1d）。引人注目的是，HRBC 在 4 °C 保存一个月后仍然保留了超过 80% 的 HPPH 分子（图 S3a，支持信息），因为共价结合的修饰稳定性比物理吸附更好（图 S3b，支持信息），并在观察期间将溶血水平维持在与自然红细胞一样低的水平（图 S3c，支持信息）。尽管在储存期间 HPPH 从 HRBC 上部分脱离，但剩余的 HPPH 分子仍然足以在激光照射下诱导完全 HRBC 溶血（图 S3d，支持信息）。这些结果凸显了使用 HPPH 化学缀合修饰红细胞膜的优势。
2.2 Light-Triggered Drug Release from HRBCs
2.2 HRBC 光触发药物释放
Next, we physically encapsulated thrombin into the inner cavity of the HRBCs via a hypotonic/hypertonic method as previously reported.[62-65] The successful drug loading was verified by confocal fluorescence imaging (Figure S4, Supporting Information), and the thrombin encapsulation efficiency was calculated to be 15.3%. In our design, the thrombin-loaded HRBCs (Th@HRBCs) should be stable in the dark but efficiently release the loaded thrombin under laser irradiation to trigger the thrombosis-related reaction (Figure 1f). As expected, the drug release dynamics indicated that the leakage of thrombin from the HRBCs was considerably slow (Figure 1g), but underwent a sharp increase to over 90% after a short-term laser irradiation treatment (671 nm, 20 mW cm−2, 60 s). The above results agree well with the real-time confocal imaging results (Figure S5, Supporting Information). To demonstrate that the released thrombin still maintained its enzymatic activity, we used a thrombin-specific chromogenic substrate S2238, which can be hydrolyzed by thrombin and then show a characteristic absorption peak at around 380 nm. The results revealed that the Th@HRBCs exhibited no enzymatic activity in the dark, but efficiently catalyzed the hydrolysis of S2238 after light exposure with a kinetic curve close to that of naked thrombin (Figure 1h). Confocal imaging results further validated that the released thrombin from Th@HRBCs upon laser irradiation retained the ability to convert soluble fibrinogen (labeled by fluorescein isothiocyanate, FITC) into insoluble fibrin, a major mechanism involved in blood clotting, as evidenced by the formation of green fluorescent aggregates (Figure 1i). A similar conclusion was also drawn from a test-tube assay in which the original fluid human plasma became gelatinous in the presence of laser-pretreated Th-HRBCs (Figure 1j), which was strongly demonstrated by the quantitative kinetic curves of plasma coagulation (Figure S6, Supporting Information). Collectively, these data coincide with our hypothesis that the HRBCs can stably encapsulate thrombin in their inner cavities and efficiently trigger the cargo release by laser irradiation for blood coagulation reaction.
接下来，我们通过先前报道的低渗/高渗方法将凝血酶物理封装到 HRBC 的内腔中。 62-65 通过共焦荧光成像验证了成功的载药（图S4，支持信息），并且计算出凝血酶封装效率为15.3%。在我们的设计中，加载凝血酶的HRBC（Th@HRBCs）应该在黑暗中稳定，但在激光照射下有效释放加载的凝血酶以触发血栓相关反应（图1f）。正如预期的那样，药物释放动力学表明，HRBC 中凝血酶的渗漏相当缓慢（图 1g），但在短期激光照射治疗（671 nm，20 mW cm ，60 秒）。上述结果与实时共焦成像结果非常吻合（图S5，支持信息）。为了证明释放的凝血酶仍保持其酶活性，我们使用了凝血酶特异性显色底物S2238，它可以被凝血酶水解，然后在380 nm左右显示特征吸收峰。结果表明，Th@HRBCs在黑暗中没有表现出酶活性，但在光照后有效催化S2238的水解，动力学曲线接近裸凝血酶的动力学曲线（图1h）。共聚焦成像结果进一步证实，激光照射后 Th@HRBC 释放的凝血酶保留了将可溶性纤维蛋白原（由异硫氰酸荧光素，FITC 标记）转化为不溶性纤维蛋白的能力，这是参与血液凝固的主要机制，绿色的形成证明了这一点。荧光聚集体（图1i）。 从试管测定中也得出了类似的结论，其中原始液体人血浆在激光预处理的 Th-HRBC 存在下变成凝胶状（图 1j），血浆凝固的定量动力学曲线有力地证明了这一点（图 1j）。 S6，支持信息）。总的来说，这些数据与我们的假设相符，即 HRBC 可以将凝血酶稳定地封装在其内腔中，并通过激光照射有效触发货物释放以进行凝血反应。
TPZ is an anticancer prodrug that can be converted into toxic TPZ radical under hypoxic tumor microenvironments.[66-72] To prepare Th/TPZ@HRBCs, TPZ and thrombin were simultaneously encapsulated into the HRBCs via the aforementioned hypotonic/hypertonic method. The TPZ encapsulation efficiency was calculated to be approximately 4.6%. Similar to what we observed in Figure 1g, the unintentional release of TPZ from the HRBC carriers in the dark was considerably low, but a burst release of TPZ was efficiently triggered by laser irradiation (Figure S7, Supporting Information). SEM images confirmed that the morphology of HRBCs did not change after being loaded with thrombin and TPZ, while the intact structure of Th/TPZ@HRBCs was completely disrupted after laser irradiation (Figure S8, Supporting Information).
TPZ是一种抗癌前药，在缺氧的肿瘤微环境下可转化为有毒的TPZ自由基。 66-72 为了制备Th/TPZ@HRBCs，通过上述低渗/高渗方法将TPZ和凝血酶同时封装到HRBCs中。计算得出TPZ封装效率约为4.6%。与我们在图1g中观察到的类似，在黑暗中HRBC载体中TPZ的无意释放相当低，但激光照射有效地触发了TPZ的爆发释放（图S7，支持信息）。 SEM图像证实HRBCs的形态在加载凝血酶和TPZ后没有改变，而Th/TPZ@HRBCs的完整结构在激光照射后被完全破坏（图S8，支持信息）。
2.3 Hypoxia-Enhanced Chemotherapy
2.3 缺氧强化化疗
We then evaluated the in vitro anticancer efficacy of Th/TPZ@HRBCs. As TPZ is a hypoxia-activated drug, we first investigated the cytotoxicity of TPZ toward murine breast cancer cells (4T1 cells) under normoxic and hypoxic conditions. As expected, TPZ exhibited significantly enhanced dose-dependent cytotoxicity under hypoxic condition (Figure 2a). In our design, the Th/TPZ@HRBCs are tasked with releasing their payloads in tumor vessels under laser irradiation, and then the TPZ penetrates through the vessel wall into tumor tissues for chemotherapy. To mimic this process in vitro, we established a transwell model where the upper chamber was seeded with a monolayer of human umbilical vein endothelial cells (HUVEC) and the lower chamber was cultured with 4T1 cells (Figure 2b). We first confirmed that upon laser irradiation, the cytoskeletons of HUVEC were severely damaged and the cells shrank to a certain extent, resulting in the appearance of many gaps between neighboring cells (Figure S9, Supporting Information). As a result, the released TPZ molecules may easily cross the HUVEC layer due to its enhanced permeability. Flow cytometric data confirmed that without laser irradiation, the Th/TPZ@HRBCs in the upper chamber exhibited negligible toxicity to 4T1 cells under both normoxic and hypoxic conditions; while upon laser irradiation, the TPZ molecules could be efficiently released and kill the 4T1 cells beneath the HUVEC monolayer in a hypoxia-enhanced manner (Figure 2c). Given that the released thrombin can trigger the blockage of tumor blood vessels and thus cut off the local oxygen supply, the exacerbated tumor hypoxia is anticipated to further sensitize the chemotherapeutic effect of TPZ.
然后我们评估了 Th/TPZ@HRBCs 的体外抗癌功效。由于TPZ是一种缺氧激活药物，我们首先研究了常氧和缺氧条件下TPZ对小鼠乳腺癌细胞（4T1细胞）的细胞毒性。正如预期的那样，TPZ 在缺氧条件下表现出显着增强的剂量依赖性细胞毒性（图 2a）。在我们的设计中，Th/TPZ@HRBCs的任务是在激光照射下在肿瘤血管中释放其有效负载，然后TPZ穿透血管壁进入肿瘤组织进行化疗。为了在体外模拟这一过程，我们建立了一个 Transwell 模型，其中上室接种单层人脐静脉内皮细胞 (HUVEC)，下室培养 4T1 细胞（图 2b）。我们首先证实，在激光照射下，HUVEC的细胞骨架受到严重破坏，细胞出现一定程度的萎缩，导致相邻细胞之间出现许多间隙（图S9，支持信息）。因此，释放的 TPZ 分子由于其渗透性增强，可以轻松穿过 HUVEC 层。流式细胞术数据证实，在没有激光照射的情况下，上室的Th/TPZ@HRBCs在常氧和缺氧条件下对4T1细胞的毒性可以忽略不计；而在激光照射下，TPZ分子可以有效释放并以缺氧增强的方式杀死HUVEC单层下方的4T1细胞（图2c）。鉴于释放的凝血酶可引发肿瘤血管阻塞，从而切断局部氧气供应，预计肿瘤缺氧加剧将进一步增强TPZ的化疗效果。
常氧或低氧条件下Th/TPZ@HRBCs的体外细胞毒性评价。 a) 分别用不同浓度的TPZ（1、2、5、10、15和20 µg mL −1 ）处理的4T1细胞在常氧和缺氧下的细胞活力。数据以平均值±标准差（n = 3）表示，两组之间的差异通过学生检验进行分析（***p < 0.001）。 b) Transwell 测定示意图，测试释放的 TPZ（来自 Th/TPZ HRBC）穿透上室 HUVEC 单层并在常氧或缺氧环境中杀死下室 4T1 细胞的能力。 c)从下室收集的4T1细胞经不同处理后的细胞凋亡/坏死分析结果。
2.4 Light-Triggered Tumor Vascular Blockage
2.4 光触发肿瘤血管阻塞
To investigate whether the Th/TPZ@HRBCs would realize on-demand drug release and induce the specific tumor vessel occlusion in vivo, we then tracked the biodistribution of Th/TPZ@HRBCs in 4T1 tumor-bearing BALB/c nude mice. After intravenous (i.v.) injection, the Th/TPZ@HRBCs showed no specific tumor accumulation (Figure 3a). Surprisingly, when the tumor area received laser irradiation (671 nm, 20 min) at 6 h postinjection, the “Laser (+)” group displayed sharply increased tumor fluorescence intensities, as suggested by the quantitative data in Figure 3b. Moreover, the enhanced fluorescence signals at the tumor regions were still observable even at 7 d postinjection (Figure S10, Supporting Information). These results implied that the external laser-irradiation induced the burst release of thrombin and local blood clotting in tumor vessels. According to the blood clotting theory,[73] we propose that the in situ formed blood clots might trap some circulating Th/TPZ@HRBCs and cause the retention of Th/TPZ@HRBCs in tumor regions. To directly demonstrate the light-triggered occlusion of tumor vessels, we next measured the blood flow in the tumor microcirculation using a full-field laser perfusion imager. The dark colors in Figure 3c reflect lower blood flow and vice versa. After laser irradiation, the blood perfusion in the tumors (especially their central regions) from the mice treated with Th/TPZ@HRBCs was severely blocked, and this situation lasted for at least 7 days without substantial recovery. In comparison, the tumor blood perfusion in the PBS and Th/TPZ@HRBCs (without laser irradiation) groups was appreciably higher. Quantitative results in Figure 3d confirmed that the treatment of Th/TPZ@HRBCs plus laser irradiation caused a large decrease in blood flow compared with the PBS treatment at day 1, and the difference between the two groups was still significant at day 7. Together, these data suggested the induction of local blood coagulation in the tumor vessels of Th/TPZ@HRBC-treated mice in a light-controlled manner.
为了研究Th/TPZ@HRBCs是否能够实现按需药物释放并诱导体内特定的肿瘤血管闭塞，我们随后追踪了Th/TPZ@HRBCs在4T1荷瘤BALB/c裸鼠中的生物分布。静脉 (i.v.) 注射后，Th/TPZ@HRBCs 显示没有特定的肿瘤积累（图 3a）。令人惊讶的是，当肿瘤区域在注射后 6 小时接受激光照射（671 nm，20 分钟）时，“激光（+）”组的肿瘤荧光强度急剧增加，如图 3b 中的定量数据所示。此外，即使在注射后 7 天，仍然可以观察到肿瘤区域增强的荧光信号（图 S10，支持信息）。这些结果表明，外部激光照射诱导了凝血酶的爆发释放和肿瘤血管中的局部血液凝固。根据血液凝固理论， 73 我们提出，原位形成的血凝块可能会捕获一些循环中的Th/TPZ@HRBCs，并导致Th/TPZ@HRBCs滞留在肿瘤区域。为了直接证明光触发的肿瘤血管闭塞，我们接下来使用全场激光灌注成像仪测量了肿瘤微循环中的血流量。图 3c 中的深色反映了较低的血流量，反之亦然。激光照射后，Th/TPZ@HRBCs治疗的小鼠肿瘤（特别是中心区域）的血液灌注被严重阻断，这种情况持续至少7天而没有实质性恢复。相比之下，PBS 和 Th/TPZ@HRBCs（无激光照射）组的肿瘤血液灌注明显较高。 图3d中的定量结果证实，与PBS处理相比，Th/TPZ@HRBCs加激光照射的治疗在第1天导致血流量大幅下降，并且在第7天两组之间的差异仍然显着。这些数据表明，Th/TPZ@HRBC 处理小鼠的肿瘤血管以光控方式诱导局部血液凝固。
Th/TPZ@HRBC 的体内生物分布和激光触发肿瘤血管阻塞。 a) 荷瘤小鼠在静脉注射后不同时间点拍摄的体内荧光图像。注射Th/TPZ@HRBCs（凝血酶：500 U kg −1 ，TPZ：3 mg kg −1 ）。对于“激光（+）”组，在注射后6小时进行激光照射（671 nm，30 mW cm −2 ，20分钟）。绿色虚线圆圈表示肿瘤区域。 b) 所示不同治疗的小鼠肿瘤区域的荧光强度变化。所有实验条件与（a）中描述的条件相同。 c) 不同组小鼠肿瘤部位的代表性彩色编码激光散斑图像。荷瘤小鼠静脉注射PBS（对照）或Th/TPZ@HRBCs（凝血酶：500 U kg −1 ，TPZ：3 mg kg −1 ），然后在分别为注射后 1、3 和 7 天。对于“激光（+）”组，在注射后6小时进行激光照射（671 nm，30 mW cm −2 ，20分钟）。红色和蓝色分别表示较高和较低的血流量。红色虚线圆圈表示肿瘤区域。 d) 与(c)相对应的局部血流的定量结果。数据以平均值±标准差（n = 5）表示，并通过单向方差分析（ANOVA）进行分析（*p < 0.05，**p < 0.01，***p < 0.001）。 “n.s.”代表无意义。 e)所示不同组的肿瘤切片的免疫荧光图像。绿色免疫荧光信号表示肿瘤组织中CD41或HIF-1α的表达，蓝色荧光信号表示Hoechst 33342染色的细胞核。比例尺 = 100 µm。 f) 反映静脉注射TPZ@HRBCs或Th/TPZ@HRBCs的4T1荷瘤小鼠肿瘤区域血氧饱和度的PAI数据（凝血酶：500 U kg −1 ，TPZ：3毫克·千克 −1 ）。注射后 6 小时进行激光照射（671 nm，30 mW cm −2 ，20 分钟）。图像是在激光照射后 24 小时拍摄的。将未处理的小鼠设为对照组。绿色虚线圆圈表示肿瘤区域。 g) 指定治疗后不同时间点小鼠肿瘤区域的定量氧饱和度水平。统计数据以平均值±标准差（n = 5）表示，两组之间的差异通过学生检验进行分析（*p < 0.05，**p < 0.01，***p < 0.001）。
2.5 Intratumoral Deoxygenation Effect
2.5 瘤内脱氧作用
Generally, thrombin-induced blood clotting involves the activation of PLTs with upregulated expression of CD41 on their surfaces. Immunofluorescence assays manifested that evident PLT plugs, that is, the aggregates of activated PLTs, occurred in the tumor slices from the “Th@HRBCs + laser” and “Th/TPZ@HRBCs + laser” groups (Figure 3e), indicating the formation of thrombi in these tumor vessels. Besides, we also observed that the levels of hypoxia-inducible factor 1α (HIF-1α), a biomarker associated with cellular hypoxia, were concomitantly upregulated in the above two groups (Figure 3e). These results strongly implied the correlation between the exacerbated tumor hypoxia and the light-triggered tumor vasculature blockage. Afterward, we further employed photoacoustic imaging (PAI) to dynamically monitor the intratumoral blood oxygen levels. As expected, after tumor-localized laser irradiation, the blood oxygen saturation level of tumor vasculatures in the mice treated with Th/TPZ@HRBCs underwent a significant decline; in particular, the level plummeted to its lowest value (≈19%) at 24 h, suggesting the presence of significant tumor deoxygenation (Figure 3f,g and Figure S11, Supporting Information). We noticed that the oxygen saturation level in the “Th/TPZ@HRBCs + laser” group gradually recovered after 24 h. This phenomenon was possibly caused by the permeation of oxygen from surrounding healthy tissues. Besides, due to the chemotherapeutic effect of TPZ, a reduction of surviving tumor cells could also decrease oxygen consumption to a certain extent. Collectively, the above data demonstrated the ability of Th/TPZ@HRBCs to deplete the intratumoral oxygen and exacerbate tumor hypoxia.
一般来说，凝血酶诱导的血液凝固涉及 PLT 的激活，其表面 CD41 的表达上调。免疫荧光分析表明，“Th@HRBCs + 激光”组和“Th/TPZ@HRBCs + 激光”组的肿瘤切片中出现了明显的 PLT 栓塞，即激活的 PLT 聚集体（图 3e），表明形成了 PLT 栓塞。这些肿瘤血管中的血栓。此外，我们还观察到缺氧诱导因子1α（HIF-1α）（一种与细胞缺氧相关的生物标志物）的水平在上述两组中同时上调（图3e）。这些结果强烈暗示了加剧的肿瘤缺氧与光触发的肿瘤脉管系统阻塞之间的相关性。随后，我们进一步采用光声成像（PAI）来动态监测瘤内血氧水平。正如预期的那样，在肿瘤局部激光照射后，接受 Th/TPZ@HRBCs 治疗的小鼠肿瘤脉管系统的血氧饱和度水平显着下降；特别是，该水平在 24 小时时骤降至最低值（约 19%），表明存在显着的肿瘤脱氧（图 3f、g 和图 S11，支持信息）。我们注意到“Th/TPZ@HRBCs +激光”组的氧饱和度水平在24小时后逐渐恢复。这种现象可能是由于周围健康组织的氧气渗透造成的。此外，由于TPZ的化疗作用，存活肿瘤细胞的减少也可以在一定程度上降低耗氧量。总的来说，上述数据证明了 Th/TPZ@HRBCs 消耗瘤内氧气并加剧肿瘤缺氧的能力。
2.6 In Vivo Antitumor Therapy
2.6 体内抗肿瘤治疗
In our design, the deoxygenation effect could not only induce severe damage to the tumor cells, but also ensure the hypoxia-activated conversion of TPZ molecules into the cytotoxic TPZ radicals (Figure 4a). To this end, we evaluated the in vivo therapeutic potential of Th/TPZ@HRBCs in 4T1 tumor-bearing BALB/c nude mice. We found that the mice treated with TPZ or “TPZ@HRBCs + laser” displayed a slower tumor growth trend (Figure 4b,c), due to the chemotherapeutic effect of TPZ. The “Th@HRBCs + laser” group also presented an evident tumor inhibition outcome, which confirmed the promising therapeutic effectiveness of cancer starvation therapy by blocking tumor-associated vasculatures. Excitingly, the “Th/TPZ@HRBCs + laser” treatment exhibited a superior antitumor effect and no tumor relapse occurred within 14 days. The tumor weights were also measured on the 14th day and the results presented in Figure S12, Supporting Information, indicated a similar conclusion. Hematoxylin and eosin (H&E) staining results verified that the tumor tissues in the “Th/TPZ@HRBCs + laser” group were severely damaged, as evidenced by the condensed cell nuclei and substantial necrosis (Figure 4d). Using terminal deoxynucleotidyl transferase deoxyuridine triphosphate (dUTP) nick-end labeling (TUNEL) assay, we confirmed that “Th/TPZ@HRBCs + laser” also caused extensive tumor cell apoptosis with severe DNA fragmentation (Figure 4e). These results collectively proved the promising antitumor effectiveness of Th/TPZ@HRBCs (plus laser) in vivo through tumor starvation therapy and hypoxia-activated chemotherapy.
在我们的设计中，脱氧效应不仅可以对肿瘤细胞造成严重损伤，还可以确保缺氧激活TPZ分子转化为细胞毒性TPZ自由基（图4a）。为此，我们评估了 Th/TPZ@HRBCs 在 4T1 荷瘤 BALB/c 裸鼠中的体内治疗潜力。我们发现，由于TPZ的化疗作用，用TPZ或“TPZ@HRBCs +激光”治疗的小鼠表现出较慢的肿瘤生长趋势（图4b，c）。 “Th@HRBCs + 激光”组还呈现出明显的肿瘤抑制结果，证实了通过阻断肿瘤相关脉管系统的癌症饥饿疗法的有希望的治疗效果。令人兴奋的是，“Th/TPZ@HRBCs+激光”治疗表现出优越的抗肿瘤效果，14天内没有出现肿瘤复发。第 14 天还测量了肿瘤重量，图 S12（支持信息）中显示的结果表明了类似的结论。苏木精和伊红（H＆E）染色结果证实，“Th/TPZ@HRBCs +激光”组的肿瘤组织严重受损，细胞核固缩和大量坏死（图4d）。使用末端脱氧核苷酸转移酶脱氧尿苷三磷酸（dUTP）缺口末端标记（TUNEL）测定，我们证实“Th/TPZ@HRBCs +激光”也引起广泛的肿瘤细胞凋亡和严重的DNA碎片（图4e）。这些结果共同证明了 Th/TPZ@HRBCs（加激光）通过肿瘤饥饿疗法和缺氧激活化疗在体内具有良好的抗肿瘤功效。
Th/TPZ@HRBCs 的体内抗肿瘤作用。 a) 示意图说明了我们系统中肿瘤内脱氧的两种途径以及 TPZ 缺氧反应性转化为 TPZ 自由基。 b) 肿瘤生长曲线和 c) 所示的各种治疗后小鼠的代表性照片。注射后 6 小时进行激光照射（671 nm，30 mW cm −2 ，20 分钟）。数据以平均值±标准差（n = 5）表示，并通过单向方差分析（**p < 0.01，***p < 0.001）进行分析。 d) H&E 染色和 e) TUNEL 染色的肿瘤组织切片，取自小鼠，在所示的各种处理后第 14 天处死小鼠。比例尺 = 100 µm。
2.7 Biosafety Evaluations of Th/TPZ@HRBCs
2.7 Th/TPZ@HRBCs的生物安全性评价
Finally, we comprehensively assessed the in vivo biosafety of the Th/TPZ@HRBC system. To investigate whether the light-triggered thrombin release in tumor vasculatures would cause the non-specific blood vessel occlusion in other parts of the mouse body, we monitored the whole-body perfusion of the mice treated with Th/TPZ@HRBC plus laser irradiation at 1, 3, 7, and 14 d postinjection, respectively. Notably, as shown in Figure 5a, compared to the control group, no evident ischemia could be observed throughout the above Th/TPZ@HRBC-treated mouse bodies, especially in the cardiac and pulmonary regions, suggesting the absence of obvious thrombosis in normal tissues. As the positive control group, the direct i.v. administration of naked thrombin at the same injected dose (500 U kg−1) caused severe ischemia in a large area of the mouse body, due to the extensive thrombosis in systemic circulation. Another noteworthy issue is that it is difficult to ensure the activation of all the injected Th/TPZ@HRBCs during the laser irradiation period, which may leave a proportion of intact Th/TPZ@HRBCs in the vascular system. Therefore, we next checked whether the long-term retention of Th/TPZ@HRBCs in systemic circulation would lead to potential side effects. Some key coagulation parameters related to thrombotic risk, viz. prothrombin time (PT), activated partial thromboplastin time (APTT), D-dimer, and fibrinogen, were monitored in the Th/TPZ@HRBC-treated mice (thrombin: 500 U kg−1) at 3, 7, and 14 d postinjection. Excitingly, there was no significant difference between healthy mice (injected with PBS) and Th/TPZ@HRBC-treated mice with respect to the values of these indices (Figure 5b), implying the absence of thrombotic risk or excessive blood clotting. By marked contrast, intravenous administration of the same dose of naked thrombin induced prolonged PT and an increased level of D-dimer in mouse bodies at 30 min postinjection. In addition, no histological abnormalities were found in the H&E-stained slices of hearts, livers, spleens, lungs, and kidneys from the mice that were sacrificed at 14 d after the administration of Th/TPZ@HRBCs (thrombin doses: 100 and 500 U kg−1) without the laser irradiation treatment (Figure 5c). In addition, hematological and biochemical analyses revealed that the Th/TPZ@HRBC treatment had no significant influence on the levels of relevant parameters (Figure 5d), including white blood cell (WBC), haematocrit (HCT), HGB, mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), PLT, plateletcrit (PCT), alanine aminotransferase (ALT), aspartate aminotransferase (AST), blood urea nitrogen (BUN), and creatinine (CRE), which implied the good hemocompatibility and low hepatotoxicity of Th/TPZ@HRBCs. The body weight curves of the mice in different groups also proved that these treatments did not cause observable loss of the mouse weights (Figure S13, Supporting Information).
最后，我们全面评估了Th/TPZ@HRBC系统的体内生物安全性。为了研究肿瘤脉管系统中光触发的凝血酶释放是否会导致小鼠身体其他部位的非特异性血管闭塞，我们监测了接受 Th/TPZ@HRBC 加激光照射治疗的小鼠的全身灌注。分别为注射后 1、3、7 和 14 天。值得注意的是，如图5a所示，与对照组相比，上述Th/TPZ@HRBC处理的小鼠体内没有观察到明显的缺血，特别是在心脏和肺部区域，表明正常组织中没有明显的血栓形成。作为阳性对照组，直接静脉注射。以相同注射剂量（500 U kg −1 ）给予裸凝血酶，由于体循环中广泛血栓形成，导致小鼠身体大面积严重缺血。另一个值得注意的问题是，很难确保在激光照射期间所有注射的Th/TPZ@HRBCs都被激活，这可能会在血管系统中留下一部分完整的Th/TPZ@HRBCs。因此，我们接下来检查了 Th/TPZ@HRBCs 在体循环中的长期保留是否会导致潜在的副作用。一些与血栓风险相关的关键凝血参数，即。在 Th/TPZ@HRBC 治疗的小鼠（凝血酶：500 U kg −1 ）中于 3、注射后7天和14天。令人兴奋的是，健康小鼠（注射 PBS）和 Th/TPZ@HRBC 治疗小鼠之间的这些指数值没有显着差异（图 5b），这意味着不存在血栓形成风险或过度凝血。 与之形成鲜明对比的是，注射后30分钟，静脉注射相同剂量的裸凝血酶可诱导小鼠体内PT延长和D-二聚体水平增加。此外，给予Th/TPZ@HRBCs（凝血酶剂量：100和500）后14天处死的小鼠的心脏、肝脏、脾脏、肺和肾脏的H&E染色切片未发现组织学异常。 U kg −1 ），无需激光照射处理（图 5c）。此外，血液学和生化分析显示，Th/TPZ@HRBC治疗对相关参数的水平没有显着影响（图5d），包括白细胞（WBC）、血细胞比容（HCT）、HGB、平均红细胞体积（ MCV)、平均红细胞血红蛋白(MCH)、平均红细胞血红蛋白浓度(MCHC)、PLT、血小板比容(PCT)、丙氨酸转氨酶(ALT)、天冬氨酸转氨酶(AST)、血尿素氮(BUN)和肌酐(CRE)，这表明Th/TPZ@HRBCs具有良好的血液相容性和较低的肝毒性。不同组小鼠的体重曲线也证明这些治疗不会导致小鼠体重明显下降（图S13，支持信息）。
Th/TPZ@HRBCs 的体内生物安全性评估。 a) 代表性的颜色编码激光散斑图像，显示小鼠的全身血流，这些图像是在静脉注射后不同时间点成像的。注射Th/TPZ@HRBCs（凝血酶：500 U kg −1 ，TPZ：3 mg kg −1 ）。注射后6小时进行激光照射（671 nm，30 mW cm −2 ，20分钟）。注射PBS和裸凝血酶(500 U kg −1 )的小鼠分别设置为阴性对照组和阳性对照组。 b) 静脉注射Th/TPZ@HRBCs（凝血酶：500 U kg −1 ，TPZ：3）的小鼠采血的凝血参数（PT、APTT、D-二聚体和纤维蛋白原） mg kg −1 ）在注射后 3、7 和 14 天。注射PBS和裸凝血酶(500 U kg −1 )的小鼠分别设置为阴性对照组和阳性对照组。 c) 从静脉注射后第 14 天处死的小鼠身上切下的主要器官（心脏、肝脏、脾脏、肺和肾）的 H&E 染色组织切片。注射PBS或Th/TPZ@HRBCs（凝血酶：100或500 U kg −1 ）。比例尺 = 100 µm。 d) BALB/c小鼠(未接种肿瘤)在静脉注射后不同时间点采集的血液分析和生化分析结果。注射Th/TPZ@HRBCs（凝血酶：500 U kg −1 ）。注射PBS和裸凝血酶(500 U kg −1 )的小鼠分别设置为阴性对照组和阳性对照组。统计数据以平均值±标准差（n = 5）表示，并通过单向方差分析（*p < 0.05，**p < 0.01，***p < 0.001）进行分析。 “n.s.”代表无意义。
3 Conclusions 3 结论
The most challenging part in the systemic delivery of thrombin is how to completely rein in its catalytic activity during the blood circulation and efficiently reactivate it inside tumor vessels (but not in tumor stroma or tumor cells). In this system, we fully take advantage of the intrinsic properties of RBCs to address this issue. First, mature RBCs lack cell nuclei and most organelles, and thus possess maximal intracellular spaces that can be utilized for accommodating thrombin as well as other therapeutics. Second, intact RBC membranes segregate the loaded thrombin from exterior blood components, particularly the fibrinogen, which effectively avoids non-specific blood clotting. Last but not least, the use of whole RBCs ensures that the encapsulated thrombin can be exclusively released in vessels because of the intrinsic blood circulation property of natural RBCs. Herein, we developed light-controllable semi-artificial RBCs capable of precisely delivering therapeutic drugs (thrombin and TPZ) to tumor-associated blood vessels for cancer starvation therapy and hypoxia-activated chemotherapy. The artificially engineered Th/TPZ@HRBCs acting as “RBC disguisers” inherited the same size, shape, and membrane proteins from natural RBCs, which endowed them with good hemocompatibility and immunological inertness during systemic circulation. We carefully demonstrated that the external laser irradiation of the tumor region could trigger the burst release of the therapeutic drugs (thrombin and TPZ) from Th/TPZ@HRBCs to elicit highly efficient blockage of tumor blood supply, resulting in the exacerbated intratumoral hypoxia and the generation of toxic TPZ radicals. In 4T1 tumor-bearing mouse models, the i.v. administration of Th/TPZ@HRBCs plus single laser irradiation treatment could completely inhibit the tumor growth for at least 2 weeks without noticeable off-target toxicity to the vascular system or major organs. Compared with the traditional anti-angiogenesis therapy, this light-activated local vasculature occlusion technique not only offers a promising anticancer strategy for starving tumors, but also advances the future development of personalized and precision medicine.
凝血酶全身输送中最具挑战性的部分是如何完全控制其在血液循环过程中的催化活性并有效地在肿瘤血管内（但不在肿瘤基质或肿瘤细胞中）重新激活它。在这个系统中，我们充分利用红细胞的固有特性来解决这个问题。首先，成熟的红细胞缺乏细胞核和大多数细胞器，因此拥有最大的细胞内空间，可用于容纳凝血酶以及其他治疗剂。其次，完整的红细胞膜将负载的凝血酶与外部血液成分（特别是纤维蛋白原）隔离，从而有效避免非特异性血液凝固。最后但并非最不重要的一点是，由于天然红细胞固有的血液循环特性，使用全红细胞可确保封装的凝血酶能够在血管中专门释放。在此，我们开发了光控半人工红细胞，能够将治疗药物（凝血酶和TPZ）精确输送到肿瘤相关血管，用于癌症饥饿疗法和缺氧激活化疗。人工改造的Th/TPZ@HRBC作为“红细胞伪装者”继承了天然红细胞相同的大小、形状和膜蛋白，这赋予了它们在体循环过程中良好的血液相容性和免疫惰性。我们仔细证明，肿瘤区域的外部激光照射可以触发Th/TPZ@HRBCs中治疗药物（凝血酶和TPZ）的爆发释放，从而引起肿瘤血液供应的高效阻断，导致瘤内缺氧加剧，产生有毒的TPZ自由基。在 4T1 荷瘤小鼠模型中，静脉注射 Th/TPZ@HRBCs联合单次激光照射治疗可以完全抑制肿瘤生长至少2周，并且对血管系统或主要器官没有明显的脱靶毒性。与传统的抗血管生成疗法相比，这种光激活局部血管闭塞技术不仅为饥饿肿瘤提供了一种有前景的抗癌策略，而且推动了个性化和精准医疗的未来发展。
On the other hand, the artificially engineered drug carriers on the basis of real RBCs, such as the HRBCs we introduced here with light-activatable property, may open a new door for researchers aiming to achieve spatiotemporally controllable drug release in blood vessels. Unlike nanoscale drug carriers that are easily cleared by the immune system or escape from vessels to enter the tumor stroma via the EPR effect, HRBCs are stable in systemic circulation and are thus extremely suitable to deliver thrombin as well as other therapeutic agents that take effect only in blood vessels. In the near future, we believe that the current HRBC-based strategy may be further developed as a universal “drug reservoir” that enables the precise drug release in blood under remote control.
另一方面，基于真实红细胞的人工工程药物载体，例如我们在此介绍的具有光激活特性的HRBC，可能为研究人员实现血管中时空可控药物释放打开一扇新的大门。与容易被免疫系统清除或通过 EPR 效应从血管逃逸进入肿瘤基质的纳米级药物载体不同，HRBC 在体循环中稳定，因此非常适合输送凝血酶以及其他仅起效的治疗药物在血管中。在不久的将来，我们相信目前基于HRBC的策略可能会进一步发展为通用的“药物储存库”，能够在远程控制下实现血液中药物的精确释放。
4 Experimental Section 4 实验部分
Materials 材料
Thrombin was obtained from Shanghai Yingxin Laboratory Equipment Co., Ltd. (Shanghai, China). HPPH was purchased from Chemleader Biomedical Co., Ltd. (Shanghai, China). PpIX, bovine serum albumin (BSA), FITC, RB, N-hydroxysulfosuccinimide sodium salt (sulfo-NHS), 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide hydrochloride (EDC HCl), and reduced glutathione (GSH) were ordered from Aladdin Chemistry Co., Ltd. (Shanghai, China). Ce6 was purchased from J&K Scientific Co., Ltd. (Beijing, China). Dimethyl sulfoxide (DMSO) and N,N-dimethylformamide (DMF) were bought from Shanghai Lingfeng Reagent Chemical Co., Ltd. (Shanghai, China). TPZ and Triton X-100 were purchased from Sigma-Aldrich (Shanghai, China). Adenosine triphosphate (ATP) was obtained from Maclin Biochemical Co., Ltd. (Shanghai, China). S2238 was purchased from Adhoc International Technologies Co., Ltd. (Beijing, China). Hoechst 33342 and Coomassie brilliant blue R-250 were bought from Beyotime Institute Biotechnology (Shanghai, China). ActinGreen was purchased from KeyGen Biotech (Jiangsu, China). Alsever's solution was purchased from Beijing Solarbio Science & Technology Co., Ltd. (Beijing, China). FITC-labeled CD41a monoclonal antibody was purchased from Thermo Fisher Scientific, Inc. (Waltham, MA, USA). FITC-labeled goat anti-rabbit IgG antibody was obtained from Bioss Antibodies (Beijing, China). HIF-1α antibody was obtained from Cell Signaling Technology (Beverly, MA, USA). Annexin V-FITC/propidium iodide (PI) apoptosis detection kit was obtained from US EVERBRIGHT Inc. (Suzhou, China). Dialysis membranes (molecular weight cut-off (MWCO): 1 and 8 kDa) were ordered from Spectrumlabs, Inc. (Rancho Dominguez, CA, USA). Deionized water (18.2 MΩ cm) was obtained from a Milli-Q system (Millipore, Billerica, MA, USA).
凝血酶购自上海迎新实验室设备有限公司（中国上海）。 HPPH 购自凯立德生物医药有限公司（中国上海）。 PpIX、牛血清白蛋白 (BSA)、FITC、RB、N-羟基磺基琥珀酰亚胺钠盐 (sulfo-NHS)、1-乙基-3-(3-二甲基氨基丙基)-碳二亚胺盐酸盐 (EDC HCl) 和还原型谷胱甘肽 (GSH)购自阿拉丁化学有限公司（中国上海）。 Ce6 购自 J&K Scientific Co., Ltd.（中国北京）。二甲亚砜（DMSO）和N,N-二甲基甲酰胺（DMF）购自上海凌峰试剂化学有限公司（中国上海）。 TPZ 和 Triton X-100 购自 Sigma-Aldrich（中国上海）。三磷酸腺苷（ATP）购自麦克林生化有限公司（中国上海）。 S2238购自Adhoc国际技术有限公司（中国北京）。 Hoechst 33342和考马斯亮蓝R-250购自碧云天生物技术研究院（中国上海）。 ActinGreen 购自 KeyGen Biotech（中国江苏）。 Alsever 的解决方案购自北京索拉宝科技有限公司（中国北京）。 FITC 标记的 CD41a 单克隆抗体购自 Thermo Fisher Scientific, Inc.（美国马萨诸塞州沃尔瑟姆）。 FITC 标记的羊抗兔 IgG 抗体购自 Bioss Antibodies（中国北京）。 HIF-1α 抗体获自 Cell Signaling Technology（Beverly，MA，USA）。 Annexin V-FITC/碘化丙啶（PI）凋亡检测试剂盒购自美国EVERBRIGHT Inc.（中国苏州）。透析膜（截留分子量 (MWCO)：1 和 8 kDa）购自 Spectrumlabs, Inc.（美国加利福尼亚州兰乔多明格斯）。去离子水 (18.2 MΩ cm）是从 Milli-Q 系统（Millipore，Billerica，MA，USA）获得的。
Isolation of RBCs 红细胞的分离
The whole blood was collected from healthy BALB/c mice and centrifuged at 1500 rpm for 5 min to remove the plasma. The resulting RBCs were then washed with cold PBS for 3 times and stored in Alsever's solution at 4 °C.
从健康 BALB/c 小鼠中收集全血，并以 1500 rpm 离心 5 分钟以去除血浆。然后用冷 PBS 洗涤所得红细胞 3 次，并在 4°C 下保存在 Alsever 溶液中。
Preparation of Photosensitizer-Modified RBCs
光敏剂修饰红细胞的制备
To prepare HRBCs, 2 mg HPPH was first dissolved in 20 µL of DMSO and then diluted to 1 mg mL−1 using water. To activate HPPH molecules, the HPPH solution was added with sulfo-NHS and EDC HCl at an HPPH:sulfo-NHS:EDC·HCl molar ratio of 1:5:5, and the pH of the obtained mixture was adjusted to 6.0 using a 2-(N-morpholino)ethanesulfonic acid (MES) buffer solution (pH = 6.0), followed by reaction at room temperature for 1 h. Next, the activated HPPH was harvested by centrifugation (12 000 rpm, 10 min) and washed with MES buffer solutions for at least 3 times. The obtained precipitate was dissolved in 20 µL of DMSO and then diluted to 1 mg mL−1 (quantified by the ultraviolet–visible (UV–vis) spectroscopy using a spectrophotometer (UV-2600, Shimadzu, Japan)) using PBS for immediate use. To react RBCs with HPPH, 1 mL of RBCs was dispersed in 8.5 mL of PBS and mixed with 0.5 mL of the above activated HPPH solution to reach a final HPPH concentration of 50 µg mL−1. The resultant mixture was kept at 25 °C under shaking for 2 h and centrifuged at 1500 rpm for 5 min to obtain the HRBCs. The resulting HRBCs were washed with PBS for 5 times to remove the unconjugated HPPH. The Ce6-, PpIX-, or RB-modified RBCs were also prepared according to the similar procedures as described above with a photosensitizer concentration of 50 µg mL−1. To modify the RBCs with HPPH via physical adsorption, the HPPH was first dissolved in DMSO to 100 mg mL−1 and then dispersed in the RBC suspension at a final HPPH concentration of 50 µg mL−1. The mixture was kept shaking at room temperature for 1 h and washed with PBS for 5 times to remove unbound HPPH molecules.
为了制备 HRBC，首先将 2 mg HPPH 溶解在 20 µL DMSO 中，然后用水稀释至 1 mg mL −1 。为了活化HPPH分子，在HPPH溶液中加入磺基-NHS和EDC HCl，HPPH:磺基-NHS:EDC·HCl摩尔比为1:5:5，并使用pH调节器将所得混合物的pH值调节至6.0。 2-(N-吗啉代)乙磺酸(MES)缓冲溶液(pH=6.0)，然后在室温下反应1小时。接下来，通过离心（12000rpm，10分钟）收获活化的HPPH并用MES缓冲溶液洗涤至少3次。将获得的沉淀物溶解在 20 µL DMSO 中，然后稀释至 1 mg mL −1 （使用分光光度计（UV-2600，Shimadzu，Japan）通过紫外可见 (UV-vis) 光谱进行定量）使用 PBS 立即使用。为了使红细胞与 HPPH 反应，将 1 mL 红细胞分散在 8.5 mL PBS 中，并与 0.5 mL 上述活化的 HPPH 溶液混合，以达到 50 µg mL −1 的最终 HPPH 浓度。将所得混合物在 25°C 下振荡保持 2 小时，并在 1500 rpm 下离心 5 分钟以获得 HRBC。将所得 HRBC 用 PBS 洗涤 5 次以去除未结合的 HPPH。 Ce6-、PpIX-或RB-修饰的红细胞也按照与上述类似的程序制备，光敏剂浓度为50 µg mL −1 。为了通过物理吸附用 HPPH 修饰红细胞，首先将 HPPH 溶解在 DMSO 中至 100 mg mL −1 ，然后分散在 RBC 悬浮液中，HPPH 最终浓度为 50 µg mL −1 .将混合物在室温下保持振荡1小时，并用PBS洗涤5次以去除未结合的HPPH分子。
Preparation of Drug-Loaded HRBCs
载药 HRBC 的制备
To prepare the Th/TPZ@HRBCs, 60 µL of the above-obtained HRBCs was dispersed in 140 µL of PBS and then mixed with 1 mL of PBS solution containing thrombin (100 U mL−1) and TPZ (2 mg mL−1). Next, the mixture was placed inside a dialysis bag (MWCO:1 kDa) and dialyzed against 100 mL of a hypotonic buffer (10 mm NaHCO3, 10 mm NaH2PO4, 20 mm glucose, 2 mm ATP, and 3 mm reduced GSH) at 4 °C for 30 min. Afterwards, the mixture was withdrawn from the dialysis bag and added with a hypertonic solution (100 mm sodium pyruvate, 100 mm inosine, 10 mm glucose, 4 mm MgCl2, 190 mm NaCl, 1666 mm KCl, 33 mm NaH2PO4, and 20 mm ATP) at a volume ratio of hypertonic solution:mixture = 1:9. After incubation at 37 °C for 30 min, the mixture was centrifuged at 3500 rpm for 5 min, and the collected Th/TPZ@HRBCs were washed with cold PBS for 3 times to remove the unloaded drugs. Th@HRBCs and TPZ@HRBCs were also prepared as described above.
为了制备 Th/TPZ@HRBC，将 60 µL 上述获得的 HRBC 分散在 140 µL PBS 中，然后与 1 mL 含有凝血酶（100 U mL −1 ）和 TPZ（ 2 毫克/毫升 −1 ）。接下来，将混合物放入透析袋（MWCO：1 kDa）中，并用 100 mL 低渗缓冲液（10 mm NaHCO 3 、10 mm NaH 2 PO 4 、20 mm 葡萄糖、2 mm ATP 和 3 mm 还原型谷胱甘肽 (GSH)，4°C 30 分钟。然后，从透析袋中取出混合物并加入高渗溶液(100mm丙酮酸钠、100mm肌苷、10mm葡萄糖、4mm MgCl 2 、190mm NaCl、1666mm KCl、33 mm NaH 2 PO 4 和 20 mm ATP），高渗溶液:混合物的体积比 = 1:9。 37℃孵育30分钟后，3500rpm离心5分钟，收集的Th/TPZ@HRBCs用冷PBS洗涤3次以去除未负载的药物。 Th@HRBCs 和 TPZ@HRBCs 也按上述方法制备。
To prepare FITC-labeled thrombin@HRBCs (Th-FITC@HRBCs), Th-FITC was first synthesized as follows: 2 kU of thrombin was first dissolved in 1 mL of water, and then mixed with 17.8 µL of FITC solution (1 mg mL−1 in DMF). The mixture was adjusted to pH 9.5 using a Na2CO3/NaHCO3 buffer solution and kept reaction at 4 °C overnight. After being dialyzed (MWCO = 8 kDa) against water for 2 days at 4 °C, the purified Th-FITC was freeze-dried and stored at −20 °C. The Th-FITC@HRBCs were prepared according to the same procedure as mentioned above.
为了制备 FITC 标记的凝血酶@HRBCs (Th-FITC@HRBCs)，首先如下合成 Th-FITC：首先将 2 kU 凝血酶溶解在 1 mL 水中，然后与 17.8 µL FITC 溶液（1 mg）混合毫升 −1 在 DMF 中）。使用 Na 2 CO 3 /NaHCO 3 缓冲溶液将混合物调节至 pH 9.5，并将反应在 4 °C 下保持过夜。在 4 °C 下对水透析（MWCO = 8 kDa）2 天后，纯化的 Th-FITC 被冷冻干燥并储存在 -20 °C 下。 Th-FITC@HRBCs 按照与上述相同的程序制备。
Acknowledgements 致谢
Y.X.Z. and H.R.J. contributed equally to this work. This work was supported by the National Natural Science Foundation of China (21673037 and 81771980) and the Open Project of Guangxi Key Laboratory of High-Incidence-Tumor Prevention & Treatment and Key Laboratory of High-Incidence-Tumor Prevention & Treatment, Ministry of Education (GKE-KF202001). All research involving animal and human participants was completed with approval from Southeast University's ethics committee.
Y.X.Z。和 H.R.J.对这项工作做出了同样的贡献。该工作得到国家自然科学基金项目（21673037和81771980）以及广西高发肿瘤防治重点实验室和教育部高发肿瘤防治重点实验室开放项目的资助(GKE-KF202001)。所有涉及动物和人类参与者的研究都是在东南大学伦理委员会的批准下完成的。
Conflict of Interest 利益冲突
The authors declare no conflict of interest.
作者声明不存在利益冲突。
