肝血窦内皮细胞（Liver sinusoidal endothelial cell, LSEC）是肝内特异的内皮细胞，在维持肝脏稳态中具有重要作用。在正常生理情况下，LSEC缺少基底膜，表面布满贯穿的孔洞结构，称为窗孔（Fenestrae）。窗孔允许乳糜颗粒、药物、脂蛋白等小分子物质通过，构成了肝细胞与血液进行物质交换的选择性屏障。此外，窗孔为T细胞提供直接接触肝细胞的机会，在抗原识别和肝脏免疫耐受方面发挥重要作用。在慢性肝损伤引起的肝纤维化早期，LSEC的窗孔结构丢失，发生去窗孔化（Defenestration）或毛细血管化，这是肝纤维化形成的必需初始步骤。去窗孔化的LSEC表现出促炎、促血管生成的连续内皮细胞表型，通过诱导肝星形细胞（Hepatic stellate cell, HSC）的激活，使其分泌大量细胞外基质（Extracellular matrix, ECM），促进肝细胞损伤并引起持续的炎症反应，加速肝纤维化的进程。在此期间，由于ECM过度沉积引起基质刚度的增加，导致了肝脏组织刚度上调，这是肝脏力学微环境中最显著的变化。已有研究报道了硬基底可以促进LSEC窗孔减少的现象，但是基质刚度通过何种力学转导机制调控LSEC发生去窗孔化目前仍不清楚。此外，在代谢相关脂肪性肝病（Metabolic dysfunction-associated steatotic liver disease, MASLD）的早期阶段，尽管肝脏组织刚度的增加和纤维化尚未发生，LSEC已经出现了去窗孔化；推测过量的脂质可能会在LSEC细胞膜中积累，导致胆固醇等脂质成分及脂筏结构增加，从而改变细胞膜张力，对膜变形相关的生物学过程产生影响。窗孔是由质膜围成的封闭膜结构，其形成及稳定与膜变形及融合过程密切相关，膜张力对窗孔的直接调控作用及其在MASLD早期LSEC去窗孔化过程中的潜在作用需要进一步探究。因此，本文以肝纤维化和MASLD为背景，选取基质刚度和细胞膜张力为典型力学因素，分别探究力学微环境和细胞自身力学特征对LSEC窗孔的调控作用，主要研究内容如下：

1. 基质刚度的增加可通过FAK-p38力学转导通路促进LSEC去窗孔化，FAK-p38可作为肝纤维化治疗的潜在靶点。本文构建了弹性模量为2 kPa和75 kPa的软、硬两种水凝胶，用于体外模拟健康和肝硬化肝脏组织刚度。在水凝胶基底上培养原代小鼠LSEC后发现，软基底上培养的LSEC可以维持较多的窗孔和完整的筛板结构，而硬基底上的LSEC窗孔数量和细胞孔隙率显著减少，LSEC的分化表型和部分功能丧失，表明硬基底促进LSEC去窗孔化。通过转录组测序、蛋白组学及磷酸化组学分析，并结合STED超分辨成像及小分子抑制剂的使用，对硬基底促进LSEC去窗孔化的力学转导机制进行了详细的探究。本文发现LSEC通过局部粘着斑激酶（Focal adhesion kinase, FAK）感知基质刚度的增加，激活的FAK将胞外力学信号传递至胞内，诱导p38及其下游分子丝裂原激活蛋白激酶激活蛋白激酶2（Mitogen activated protein kinase activated protein kinase 2, MK2）的激活。磷酸化的MK2进一步激活下游LIM激酶1（LIM Kinase 1, LIMK1）和丝切蛋白（Cofilin），介导细胞骨架重组，导致应力纤维增多，最终促进LSEC窗孔结构的丢失和分化表型及部分功能的丧失；分别抑制FAK和p38-MK2的激活或F-actin的聚合后，硬基底上培养的LSEC均可恢复窗孔结构。为了进一步验证上述力学转导通路在肝纤维化中的重要性，本文构建了CCl₄诱导的肝纤维化小鼠模型。结果发现给药1周后，肝脏组织刚度增加至对照组的2.5倍，证明肝脏组织刚度的升高发生于纤维化早期，是LSEC去窗孔化的促进因素。值得注意的是，分离不同阶段肝纤维化小鼠内的LSEC并在培养过程中对关键力学转导分子FAK或p38-MK2进行抑制，发现已经去窗孔化的LSEC又恢复了窗孔结构及部分分化表型和功能。综合以上结果，增加的基质刚度通过FAK-p38-MK2-LIMK1-Cofilin-F-actin力学传递链促进LSEC去窗孔化，为肝纤维化的治疗提供了新的干预靶点。
2. 细胞膜张力负调控窗孔，主要由膜成分或脂筏结构所决定。本文首先发展了基于原子力显微镜（Atomic force microscopy, AFM）和受激发射损耗（Stimulated emission depletion, STED）方法的活细胞成像技术，可以对窗孔的动态演化进行快速长时程超分辨成像，为后续研究提供方法学基础。随后使用小分子化合物改变LSEC质膜成分或结构，结果发现破坏脂筏结构或降低质膜中的胆固醇含量后，LSEC的细胞膜张力减小，窗孔数量增多，而增加质膜中的胆固醇会引起LSEC膜张力的增加和窗孔的减少，表明改变质膜成分或结构可以调节细胞膜张力并影响窗孔。进一步使用不同渗透压的溶液直接调节细胞膜张力，使用Flipper-TR膜张力荧光探针和AFM对LSEC膜张力及窗孔进行了动态检测和观察。结果显示由高渗溶液处理引起的膜张力降低可以介导LSEC表面迅速形成大量的窗孔，并随时间持续增加，表现出时间依赖性。而低渗溶液处理引起的膜张力增加会促进LSEC胞体破裂，发生去窗孔化并形成间隔样（gap）结构，窗孔数量维持在较低水平基本不变。LSEC的细胞膜张力和窗孔演化对高、低渗环境的不对称响应具有一致性。此外，抑制F-actin的聚合也会造成LSEC细胞膜张力的减小，提示基质刚度的增加可能通过诱导细胞骨架重组增大膜张力，进而促进LSEC去窗孔化。以上结果表明改变细胞膜成分或脂筏结构可以调节细胞膜张力及窗孔数量，细胞膜张力对窗孔具有负调控作用。

综上所述，本文从基质刚度和细胞膜张力出发，阐明了基质刚度对LSEC去窗孔化的生物力学调控机制，探究了细胞膜张力对窗孔动态演化的负调控作用，为肝纤维化的治疗提供了新的治疗策略，也为MASLD早期LSEC去窗孔化的潜在机制提供了新思路。

Liver sinusoidal endothelial cells (LSECs), a specialized type of endothelial cells, play an essential role in the maintenance of liver homeostasis. In normal physiological conditions, LSECs lack an organized basement membrane and are covered with pore structures called fenestrae. Fenestrae act as a selective sieving barrier to control the exchange of small molecules such as chylomicron remnants, drugs, and lipoproteins, between the blood circulation and the hepatic parenchyma. In addition, the fenestrae provide T cells to directly contact with hepatocytes and play an important role in antigen recognition and hepatic immune tolerance. In the early stages of chronic liver injury, the fenestrae of LSECs tend to disappear, called the defenestration or capillarization, which is a necessary preliminary step in the formation of liver fibrosis. The capillarized LSECs express pro-inflammatory and pro-angiogenic phenotypes, and accelerate the progression of liver fibrosis by inducing the activation of hepatic stellate cells (HSCs) that secrete a large amount of extracellular matrix (ECM), promoting hepatocellular injuries and inducing persistent inflammatory responses. During this period, the increase of matrix stiffness induced by excessive deposition of ECM leads to the upregulation of liver tissue stiffness, which is the most significant change in the liver mechanical microenvironment. It has been reported that stiff substrates can promote the decrease of fenestrae, but the mechanotransduction mechanism in the impact of matrix stiffening on LSEC defenestration is still unclear. Furthermore, in the early stages of metabolic dysfunction associated steatotic liver disease (MASLD), LSECs have developed into defenestration even though increased liver tissue stiffness and fibrosis have not yet occurred. In this study, it is hypothesized that excess lipids may accumulate in LSEC cell membranes, leading to an increase in both lipid components, such as cholesterol, and lipid raft structures, which alters cell membrane tension and yields an impact on biological processes related to membrane deformation. Indeed, fenestrae are closed membrane structures surrounded by plasma membrane, and their formation and stability are closely related to membrane deformation and fusion processes. The direct regulatory role of cell membrane tension on fenestrae and its potential role in LSEC defenestration during the early stages of MASLD need to be further investigated. Thus, starting from the biological significances of liver fibrosis and MASLD, this dissertation focuses on investigating the roles of typical mechanical factors of matrix stiffness and cell membrane tension on the regulation of their fenestrae, respectively. The main works are divided into the following two parts:

1. Increased matrix stiffness promotes LSEC defenestration via the FAK-p38 mechanotransduction pathway, which may serve as a potential target for liver fibrosis therapy. Here two sets of PA hydrogels with elastic modulus of 2 (soft) or 75 (stiff) kPa were fabricated to mimic tissue stiffness in healthy and cirrhotic livers, respectively. Collected primary mouse LSECs cultured on hydrogel substrates revealed that intact fenestrae and sieve plates were maintained on soft substrates, while the fenestra number and cellular porosity were significantly reduced on stiff substrates, with loss of specific differentiated phenotypes and functions, indicating that stiff substrates promote LSEC defenestration. The mechanotransductive mechanisms of stiff substrates that promotes LSEC defenestration was investigated in detail by RNA sequencing, proteomic and phospho-proteomic analyses, and combined with STED super-resolution imaging and the application of typical inhibitors. Results indicated that LSECs sensed the increase in matrix stiffness via focal adhesion kinase (FAK), and activated FAK transmitted extracellular mechanical signals to intracellular ones, inducing the activation of p38 and its downstream molecules mitogen activated protein kinase activated protein kinase 2 (MK2). Phosphorylated MK2 further activated LIMK1 and Cofilin. This signaling cascade triggered the reorganization of the actin cytoskeleton and the accumulation of stress fibers, and ultimately facilitated LSEC defenestration. After inhibiting activation of FAK and p38-MK2 or polymerization of F-actin, respectively, LSECs cultured on stiff substrates all restored fenestrae structure. To further demonstrate the importance of the above mechanotransduction pathway in liver fibrosis, a mouse model of CCl₄-induced liver fibrosis was constructed. It was found that the liver tissue stiffness increased to 2.5-fold of the control group after 1 week of treatment, suggesting that matrix stiffening occurs at the early stages of liver fibrosis and serves as a promoter of defenestration. It is worth noting that the LSECs, which were isolated from fibrotic livers and exhibited loss of fenestrae, presented a remarkable recovery in fenestrae following various inhibition of the key mechanotransduction molecules FAK or p38-MK2. In conclusion, matrix stiffness governs the progress of LSEC defenestration through FAK-p38-MAPKAPK2-LIMK1-Cofilin-F-actin mechanotransductive axis, which provides a new target in the treatment of liver fibrosis.
2. Cell membrane tension negatively regulates LSEC fenestrae, which is highly associated with membrane lipid components and lipid rafts. First, two live cell imaging techniques of atomic force microscopy (AFM) and stimulated emission depletion (STED) was developed, allowing fast and long-term super-resolution imaging of fenestra dynamic evolution. Alteration of LSEC plasma membrane composition or structure using small molecule compounds revealed that disruption of the lipid raft structure or reduction of cholesterol content in the plasma membrane resulted in a decrease in cell membrane tension and an increase in fenestra number in LSECs, whereas increasing cholesterol in the plasma membrane induced an increase in membrane tension and a decrease in fenestra number, suggesting that altering plasma membrane composition or structure can modulate cell membrane tension and fenestrae. Next, cell membrane tension was directly modulated by solutions with different osmotic pressures. Time courses of LSEC membrane tension and fenestrae evolution were detected and observed using the Flipper-TR membrane tension fluorescent probe and AFM. Results showed that the low membrane tension induced by hypertonic shock mediated a rapid formation of fenestrae on the surface of LSECs, and the fenestra number continued to increase with time in a duration-dependent manner. However, high membrane tension induced by hypotontic shock promoted the rupture of LSECs and the loss of fenestrae and formation of gap, and the fenestra number was essentially unchanged ad maintained at a lower level. Consistent responses for the asymmetry of cell membrane tension and fenestrae were observed in relation to hypertonic and hypotonic shocks. In addition, inhibition of F-actin polymerization caused a decrease in LSEC cell membrane tension, implying that increased matrix stiffness may increase membrane tension by inducing cytoskeletal reorganization, which in turn promotes LSEC defenestration. Collectively, changing the cell membrane composition or lipid raft structure can regulate the cell membrane tension and fenestra number, and the cell membrane tension has a negative regulatory effect on fenestrae.

In summary, this dissertation elucidates the biomechanical regulatory mechanism of matrix stiffness on LSEC defenestration, and explores the negative regulatory roles of cell membrane tension on the fenestra dynamic evolution. This work provides a new therapeutic strategy for liver fibrosis treatment, and also proposes a new idea about the potential mechanism of LSEC defenestration in early stage of MASLD.