研究背景及目的 非酒精脂肪肝病(Nonalcoholic fatty liver disease，NAFLD)为慢性代谢性疾病，主要病理特征为肝细胞发生脂质沉积及空泡变性。在我国，NAFLD已经超越慢性乙型肝炎成为第一大慢性肝脏疾病，严重影响国民健康。随着肝细胞脂质沉积加重及炎症反应发生，约20％NAFLD患者可进展为非酒精性脂肪性肝炎(non-alcoholic steatohepatitis，NASH)。其中，部分NASH患者逐渐发展为肝细胞纤维化，甚至肝细胞肝癌(hepatocellular carcinoma，HCC)。NAFLD发病机制复杂，目前公认的主流学说为多重打击学说：在NAFLD早期肝细胞开始出现脂质沉积；随着脂质沉积增加肝细胞脂毒性破坏加重，继而诱发氧化应激，导致细胞线粒体受损、大量活性氧(reactive oxygen species，ROS)产生并触发多种炎症因子释放并加剧肝细胞损伤。众多文献证实，ROS及由其触发的炎性反应是促进NASH进展的主要因素之一。因此，寻找靶向干预ROS介导的炎症反应的关键调控分子，对于NASH临床治疗具有重要意义。 肝脏内富含多种固有免疫细胞，固有免疫细胞及相关分子参与多种肝脏疾病致病过程。越来越多的研究表明，固有免疫细胞尤其是肝脏巨噬细胞在NASH疾病进展中发挥关键作用。有报道称，NASH肝内往往伴有肝巨噬细胞(枯否细胞，kuppfer cell，KC)的浸润及炎性活化。KCs活化后可产生多种介质，如ROS和促炎细胞因子等，加剧NASH所诱导的肝损伤。Baeck C等证实，靶向干预趋化因子MCP-1，可有效抑制肝巨噬细胞浸润，减轻胆碱-蛋氨酸缺乏饮食(methionine-choline deficient diet，MCD)诱导的NASH小鼠肝损伤。所以，靶向肝脏巨噬细胞，尤其是肝巨噬细胞来源的ROS及其介导的炎症反应，有望为NASH治疗提供新策略。 Tim-3(T cell immunoglobulin and mucin-domain contining-3，Tim-3)最初被定义为T细胞免疫调节分子。但越来越多的证据表明，多种固有免疫细胞如NK细胞、巨噬细胞均表达Tim-3。单核细胞及巨噬细胞均组成性表达高水平的Tim-3，不同环境因素调节Tim-3表达，进而影响单核／巨噬细胞功能，参与多种疾病进展。肿瘤微环境中的TGFβ促进肿瘤相关巨噬细胞Tim-3表达，升高的Tim-3参与影响巨噬细胞极化、促进HCC发生发展。另有研究称，慢性乙型肝炎(Chronic Hepatitis B，CHB)患者外周Tim-3⁺CD14⁺单核细胞数目与患者血清谷氨酸氨基转移酶(alanine aminotransferase，ALT)呈正相关。然而，Tim-3是否通过调控肝脏巨噬细胞功能参与NASH进展，迄今尚无文献报道。在NASH诱导的肝损伤过程中，Tim-3对巨噬细胞活化及活性氧产生是否具有调控作用，仍有待证实。本研究拟利用MCD建立NASH小鼠模型，探讨Tim-3对肝巨噬细胞的调控及其在NASH中的作用。 方法与结果 1.NASH小鼠肝组织Tim-3表达显著升高，敲除Tim-3明显促进NASH进展 为探讨Tim-3在NASH中作用及可能的机制，我们首先利用MCD饲料、高脂饲料(High-Fat Diet，HFD)饲喂C57BL／6小鼠建立NASH小鼠模型。收集MCD饮食2W、HFD饮食24W小鼠肝组织，采用免疫组化及western blot分析肝内蛋白水平的Tim-3表达。结果表明，与正常饮食(normal diet，ND)组相比较， MCD及HFD诱导的NASH小鼠肝组织中Tim-3表达均呈现显著增加，提示Tim-3可能参与影响NASH。 为进一步明确Tim-3在NASH诱导的肝损伤中的作用，利用TALEN技术构建Tim-3敲除(Tim-3KO)小鼠。分离肝脏单个核细胞，利用流式分析肝内CD45⁺细胞Tim-3表达，证实肝内免疫细胞来源的Tim-3敲除成功。分别对野生型(wild type，WT)及Tim-3KO小鼠进行MCD饮食，诱导NASH模型，比较两组小鼠肝脏损伤及脂质沉积情况。结果表明，与WT小鼠相比较：Tim-3KO小鼠MCD处理后血清ALT显著上调，肝组织甘油三酯(Triglyceride，TG)增加，肝细胞空泡化加剧，肝组织纤维化明显加重。以上结果证实，Tim-3减缓MCD诱导的NASH小鼠的肝脏炎性损伤。 2.Tim-3在NASH小鼠肝内不同巨噬细胞亚群的表达均显著增加 肝脏内的主要固有免疫细胞包括巨噬细胞、NK及NKT等，均表达较高水平的Tim-3，上述免疫细胞在NASH进展中发挥重要作用。为明确NASH小鼠肝内高水平Tim-3的主要固有免疫细胞来源，本实验收集MCD处理2W小鼠肝组织，首先采用40％percol对肝单个核进行体外分离，然后利用流式细胞术分析肝内巨噬细胞、NK及NKT细胞的比例。流式结果表明，MCD诱导的NASH模型肝脏内NKT比例显著降低，而NK及巨噬细胞比例显著升高。进一步对比两组小鼠肝内NK及巨噬细胞Tim-3表达情况，发现相比较ND组，MCD肝内NK细胞Tim-3无显著变化，而肝脏贴壁巨噬细胞Tim-3表达呈现显著上调。肝巨噬细胞为异质性细胞群，包括F4／80⁺CD11b⁺、F4／80⁺CD68⁺、F4／80⁺CD169⁺等巨噬细胞亚群。流式结果表明，MCD诱导的NASH小鼠肝脏内上述三群巨噬细胞亚群中的Tim-3表达均见显著上调。上述流式结果在HFD小鼠模型中也得到了进一步验证，HFD 24W小鼠肝内F4／80⁺CD11b⁺以及F4／80⁺CD68⁺巨噬细胞中Tim-3表达显著升高。 为进一步验证人发生脂肪变的肝组织中巨噬细胞Tim-3是否高表达，收集肝癌患者癌旁组织，免疫荧光技术检测CD68⁺细胞Tim-3表达，结果发现，与对照相比较，发生空泡化(提示可能发生肝细胞脂质沉积)肝组织内CD68⁺巨噬细胞Tim-3表达明显升高。 综上，Tim-3在NASH动物模型肝脏不同巨噬细胞亚群及人发生空泡化的肝组织内巨噬细胞中的表达均见显著增加，提示肝巨噬细胞为NASH肝内高水平Tim-3的主要细胞来源，在NASH中发挥作用。 3.MCD肝匀浆及OA诱导巨噬细胞Tim-3高表达 为明确NASH小鼠肝脏巨噬细胞表达高水平Tim-3的原因，寻找诱导Tim-3升高的组织微环境因素，制备ND及MCD饮食2周小鼠肝匀浆并分别刺激腹腔来源的巨噬细胞，体外模拟NASH肝脏内微环境对巨噬细胞刺激作用。qPCR结果显示，与ND肝匀浆刺激组相比较，MCD肝匀浆刺激后巨噬细胞Tim-3表达显著上调。与此相一致，不饱和脂肪酸油酸(oil acid，OA)刺激巨噬细胞，导致巨噬细胞发生脂质沉积，并诱导巨噬细胞Tim-3表达升高。上述结果提示OA及NASH小鼠肝脏微环境促进巨噬细胞Tim-3高表达。 4.Tim-3明显抑制巨噬细胞ROS及下游炎性因子产生 肝脏巨噬细胞来源的ROS在NASH中发挥重要作用，为研究Tim-3对巨噬细胞来源ROS的具体作用，本实验分别收集MCD、HFD小鼠肝脏贴壁巨噬细胞，流式检测其ROS水平，并比较WT及Tim-3KO小鼠肝组织中NADPH酶2 (NOX2)的表达差异。与前期报道一致，MCD、HFD小鼠肝贴壁巨噬细胞ROS水平显著高于ND组，说明巨噬细胞来源ROS在NASH中具有重要作用。进一步收集MCD饮食2周的WT、Tim-3KO小鼠肝组织，western blot结果显示，与对照组相比，Tim-3KO小鼠肝组织NOX2表达显著增加，提示Tim-3对ROS发挥调控作用。 为证实以上推论，利用Tim-3中和抗体(anti-Tim-3)阻断骨髓来源巨噬细胞(bone marrow-derived macrophages，BMDM)Tim-3通路后进行体外刺激，流式检测ROS水平。结果表明，LPS、PA刺激经anti-Tim-3预处理的BMDM后ROS水平较IgG组显著升高。上述趋势在siTim-3预处理的腹腔巨噬细胞(peritoneal macrophage，PM)中得到进一步验证。进一步利用Tim-3KO来源的BMDM及PM，同样观察到Tim-3对巨噬细胞ROS及线粒体活性氧的抑制作用。此外，相比较WT对照组，Tim-3KO小鼠来源的肝脏贴壁巨噬细胞ROS产生也明显增加。以上研究结果说明：Tim-3抑制巨噬细胞ROS产生。 有文献称，ROS可诱导炎性小体活化及相关炎性因子产生。为研究Tim-3是否调控ROS介导的炎性小体活化及下游炎性分子分泌，我们对MCD处理后WT及Tim-3KO小鼠肝内炎性小体活化及下游相关分子表达进行分析，结果发现：与MCD处理的WT小鼠相比较，MCD饲喂的Tim-3KO小鼠肝内caspase-1 p20、IL-1β、IL-18表达均呈显著升高，但TNF-a表达无显著差异。同样，Western blot、RT-PCR及ELISA结果表明，anti-Tim-3预处理的BMDM中caspase-1 p20表达升高、IL-1β分泌增加，提示Tim-3抑制巨噬细胞炎性小体活化及相关炎性因子产生。进一步利用ROS抑制剂NAC(N-acetylcysteine，NAC)预处理WT、 Tim-3KO小鼠来源PM，流式结果显示：NAC处理后Tim-3KO对IL-1β的上调作用消失。综上结果提示，Tim-3通过负向调控ROS抑制巨噬细胞炎性小体活化及相关炎性因子释放。 5.ROS抑制剂显著减缓Tim-3KO小鼠中MCD介导的肝损伤 为进一步明确Tim-3抑制ROS在MCD介导肝损伤中的作用，设计NAC体内干预实验。分别给予WT及Tim-3KO小鼠150㎎／㎏NAC隔日腹腔注射处理，MCD饮食饲喂2周后：采集外周血分离血清检测ALT；制备肝匀浆检测TG、IL-1β以及IL-18水平；肝组织提取蛋白检测caspase-1 p20；肝石蜡切片后进行HE染色观察肝细胞空泡化。结果显示，NAC及MCD处理后WT与Tim-3KO小鼠上述指标均无显著差异，提示NAC处理可以显著改善Tim-3KO导致的NASH肝损伤。 结论： NASH肝脏微环境促进肝内多种巨噬细胞亚群高表达Tim-3，Tim-3显著抑制巨噬细胞ROS及下游炎性因子分泌，保护NASH引起的肝损伤。 创新性和意义： 本研究首次报道了NASH个体中肝脏不同巨噬细胞显著高表达Tim-3，发现Tim-3抑制巨噬细胞ROS及其下游炎性反应，发挥减缓非酒精性脂肪肝介导肝损伤的作用。上述研究结果不仅有助于阐明免疫检查点Tim-3在NASH肝固有免疫中的作用，并且为NASH治疗提供了新靶点。 关键词：巨噬细胞；非酒精脂肪肝炎；活性氧；Tim-3

 Background As a multisystem disease, Nonalcoholic fatty liver disease (NAFLD) severely affects the health of human beings. In China, NAFLD has become a most common chronic liver disease besides HBV. While most NAFLD patients remain asymptomatic, roughly 20% of them develop into a severer condition which is evidenced by steatosis and inflammation and known as non-alcoholic steatohepatitis (NASH). Some NASH patients can progress to cirrhosis or even hepatocellular carcinoma (HCC). So far, multiple hits theory is the dominating pathogenesis explanation of NAFLD. Hepatic lipid deposition appears in the early stage of NAFLD and massive lipid accumulation leads to severer mitochondria damage and more Reactive oxygen species (ROS) production, resulting in inflammatory cytokines release and hepatotoxicity. More and more evidence approves that accumulated ROS is one of the key risk factors for NASH development. Therefore, to identify a critical regulator interfering with ROS-mediated hepatic inflammation in NASH mice is strongly suggested. In liver, there exist numerous innate immune cells participating in the pathogenesis of multiple liver diseases. Accumulated evidence proves that key innate immune cells in liver, especially kupffer cells, play important roles in NASH development. According to the reports, amounts of activated macrophages (kupffer cells, KCs) appear in NASH liver tissue. Activated KCs aggravate liver injury by producing mediators such as ROS and inflammatory cytokines in MCD(methionine- and cystine deficient diet) induced NASH model. According to Baeck C, blocking MCP-1 can efficiently inhibit hepatic macrophages infiltration and improve MCD induced NASH liver injury. Thus, targeting liver macrophages and ROS mediated inflammation has a constructive meaning for NASH therapy. Tim-3 (T cell immunoglobulin and mucin-domain containing-3) was first recognized as T cell immune regulator. However, in recent years, more and more evidence confirm that Tim-3 as well is expressed on NK, NKT and macrophages. Tim-3 is highly constitutively expressed on monocytes/macrophages and is regulated by different microenvironment. By affecting the function of monocytes/macrophages, Tim-3 participates in the process of diverse disease. In tumor microenvironment, elevated TGFβ plays roles in enhancing Tim-3 expression on tumor associated macrophages and increased Tim-3 can then promote HCC development by regulating macrophage polarization. According to another report, Tim-3⁺CD14⁺macrophages has a positive correlation with serum ALT in CHB (Chronic hepatitis B) patients. However, there are no reports yet to show whether Tim-3 participates in NASH by regulating hepatic macrophages. Also, whether Tim-3 is involved in macrophage activation or ROS mediated inflammation in NASH still need further verification. Herein, we applied MCD induced NASH model to discuss the potential regulatory roles of Tim-3 on macrophages in NASH. Methods and Results 1.Tim-3 is increased significantly in NASH liver and Tim-3KO promotes NASH progress To discuss the potential roles and possible mechanism of Tim-3 in NASH mice, we fed WT C57BL/6 mice with MCD or HFD (High-Fat Diet) diet to establish NASH animal models. Liver tissues were collected from MCD 2 weeks and HFD 24 weeks mice to analyze Tim-3 expression by IHC and western blot techniques. The data showed that, compared to ND (normal diet) group, Tim-3 was enhanced significantly in livers of MCD and HFD mice, indicating that Tim-3 may be involved in the pathogenesis of NASH. To furtherly determine whether Tim-3 affects NASH induced liver injury, we successfully constructed Tim-3KO mice model by TALEN technique. Firstly, we separated liver monocytes and analyzed the knockdown efficiency of Tim-3 in hepatic CD45⁺ monocytes by performing FCM. Then, we used WT and Tim-3KO mice to establish MCD-induced NASH model and evaluated the exact roles of Tim-3 in NASH mice. In vivo results displayed that, compared to WT MCD mice: serum ALT was elevated, liver TG was increased, hepatic vacuolization was enlarged and liver fibrosis was severer in Tim-3KO mice. The above data suggested Tim-3 as a protector in MCD-induced NASH mice. 2.Tim-3 on different hepatic macrophage subpopulations is enhanced in NASH liver Livers macrophages, NK and NKT cells occupy the majority of liver innate immune cells, with high Tim-3 surface expression. The mentioned hepatic innate immune cells play key roles in NASH progress. To check the exact cell source of increased Tim-3 expression in NASH liver, we collected liver tissues from MCD 2 weeks mice and separated liver monocytes by 40% percoll. And we then performed FCM to detect the proportion of hepatic macrophages, NK and NKT cells. The results showed that the percentage of NKT was decreased significantly, while NK and macrophages were increased in NASH livers. We furtherly detected Tim-3 expression in hepatic NK cells and adherent liver macrophages. We found that Tim-3 expression on macrophages was increased but showed no significant difference in hepatic NK. Hepatic macrophages are heterogeneous, consisting of F4/80⁺CD11b⁺ liver motile macrophages, F4/80⁺CD68⁺ and F4/80⁺CD169⁺ residential liver macrophages. FCM(Flow Cytometry) results displayed that Tim-3 expression was enhanced on all detected liver macrophages in MCD mice. Meanwhile, FMO (Fluorescence Minus One) test strongly approved the specificity of FCM results. In HFD 24W mice, the same tendency was discovered with significantly enhanced Tim-3 expression on F4/80⁺CD11b⁺and F4/80⁺CD68⁺ hepatic macrophages. To analyze Tim-3 level in human fatty livers, we collected paracancerous tissues with or without vacuolation from livers of HCC patients and conducted fluorescent immunohistochemistry. Compared to groups without hepatic vacuolation (as normal controls), Tim-3 expression on CD68⁺ hepatic macrophages was increased in vacuolated liver tissues (as a reminder of fatty liver). Taken together, Tim-3 expression on diverse macrophages subgroups from different NASH models and human vacuolated livers was remarkably increased. The above results strongly suggested that hepatic macrophages were the main cell source of increased Tim-3 and played important roles in NASH development. 3.MCD liver homogenate and OA contribute to Tim-3 overexpression in macrophages To identify the possible inducer or microenvironment factor which leads to Tim-3 overexpression in NASH liver, we employed liver homogenate from ND and MCD mice to stimulate PM(peritoneal macrophages) and analyzed Tim-3 levels by qPCR technique. Compared to ND group, Tim-3 expression was significantly boosted by MCD liver homogenate. Next, we detected massive lipid deposition and high Tim-3 expression in PM after unsaturated fatty acid OA (oil acid, OA) stimulation. Our results indicated that OA and NASH hepatic microenvironment could be potential inducers for Tim-3 overexpression in macrophages. 4.Tim-3 inhibits ROS and downstream cytokines release in macrophages ROS from macrophages plays important roles in NASH. Firstly, we collected adherent hepatic macrophages from MCD and HFD mice and detect ROS levels. Our result was coincident with previous report, showing higher ROS levels in adherent hepatic macrophages from NASH livers and pointing macrophage derived ROS as a key player in NASH. Furtherly, to determine whether Tim-3 regulated macrophages derived ROS, we collected livers of WT and Tim-3KO MCD 2W mice and detected NOX2 expression by western blot. Compared to WT group, NOX2 expression was much higher in Tim-3KO MCD liver, suggesting Tim-3 act as a potential regulator upon ROS production. To certify the above hypothesis, we applied LPS or PA to stimulate anti-Tim-3 pretreated BMDM (bone marrow-derived macrophages). Then, we detected ROS levels by FCM analysis. Our data showed that Tim-3 negatively regulated ROS generation in BMDM. Consistently, we got the same conclusion in siTim-3 pretreated PM. Consistent with the above results, in Tim-3KO derived BMDM and PM, we observed the inhibitory function of Tim-3 on ROS and Mitochondrial ROS production. Additionally, higher ROS levels were observed in Tim-3KO hepatic macrophages. Taken together, Tim-3 acts as an inhibitor upon ROS generation in macrophages. According to the report, ROS could trigger inflammasome activation and enhance related inflammatory cytokines release. Hence, we furtherly studied whether Tim-3 had the capability to inhibit ROS triggered inflammasome activation. Firstly, we compared levels of inflammasome and related cytokines in livers of WT and Tim-3KO MCD mice. Compared to WT group, significantly elevated caspase-1 p20, IL-1β and IL-18 was observed in Tim-3KO MCD livers, however, TNF-a level showed no difference in Tim-3KO mice. Also, results from anti-Tim-3 pretreated BMDM is with the same trend. Furtherly, we pretreated WT and Tim-3KO derived PM with NAC(a small molecular inhibitor for ROS) and then stimulated them with HFD liver homogenate. Interestingly, the difference of IL-1β in WT and Tim-3KO group was reversed after NAC treatment. The above data indicated Tim-3 as a potential regulator on ROS/inflammasome axis and downstream cytokine release in macrophages. 5.The aggravated liver injury of Tim-3KO MCD mice is alleviated by inhibiting ROS To furtherly testify the Tim-3 function of regulating ROS generation in NASH livers. We intervened WT and Tim-3KO MCD mice with NAC. 150㎎/㎏ NAC was given to WT and Tim-3KO MCD mice every other day. Serum and liver tissues were obtained after 2 weeks to observe liver inflammation index. Serum was separated for ALT detection; liver homogenate was prepared for TG, IL-1β and IL-18 analysis; liver protein was extracted for caspase-1 p20 detection; while liver paraffin section was obtained for vacuolization analysis. The above data showed no significant difference between WT and Tim-3KO MCD group, suggesting NAC pretreatment could improve liver injury induced by Tim-3KO treatment. Taken together, the aggravating impact of Tim-3KO on MCD mice is ROS dependent. Conclusion: NASH liver microenvironment enhances Tim-3 expression on diverse hepatic macrophages; Tim-3 inhibits ROS and downstream inflammatory cytokines release in macrophage, protecting liver damage induced by NASH. Innovation and significance: For the first time, our study showed that Tim-3 was highly expressed on different macrophages subpopulations in NASH livers and enhanced Tim-3 could inhibit ROS and downstream response, reducing NASH mediated liver injury. Above-mentioned studies not only highlighted the roles of Tim-3 in innate immune response in NASH livers, but also offered new target for NASH treatment. Key words: macrophage; NASH; ROS; Tim-3