DNALI1 Promotes Neurodegeneration after Traumatic Brain Injury via Inhibition of Autophagosome‐Lysosome Fusion
DNALI1 通过抑制自噬体-溶酶体融合促进脑外伤后神经退行性变

2.1. DNALI1 is a Critical Gene to Predict Cognitive Impairment Post‐TBI
2.1. DNALI1是预测 TBI 后认知障碍的关键基因

We obtained clinical and genetic data from 107 individuals participating in the Aging, Dementia, and Traumatic Brain Injury Study.^{[ 12 ]} The study's demographics and characteristics are shown in Table S1 (Supporting Information). We categorized participants based on TBI diagnosis, dementia pathology, and dementia clinical diagnosis defined by the original study^{[ 12 ]} (Figure  1a). Utilizing sequencing data from four brain regions (temporal cortex [TCx], parietal cortex [PCx], frontal white matter [FWM], and hippocampus [HIP]) we conducted a preliminary analysis employing principal component analysis, and revealing distinct gene expression patterns in the hippocampus compared to other brain regions (Figure S1a, Supporting Information). Subsequently, we performed an extensive analysis of AT8 levels (mapping tau phosphorylation at Ser202 or Thr205; Figure 1b; Figure S1b–d, Supporting Information) and the brain ptau181/tau ratio (Figure 1c; Figure S1e–g, Supporting Information) across the brain regions described above, comparing “clinical dementia” and “non‐demented” cohorts in TBI and pathology groups. Both AT8 and ptau181/tau ratio exhibited significant differences between “clinical dementia” and “non‐demented” groups exclusively within the hippocampus (Figure 1b,c), suggesting that this brain region may be the primary area affected by TBI. We have also examined changes related to Aβ pathology, detecting no significant differences among groups (Figure S2, Supporting Information), thus indicating that the pathology in the study may be consistent with CTE.
我们获得了参与衰老、痴呆和创伤性脑损伤研究的 107 名个体的临床和遗传数据。 ^{[ 12 ]}该研究的人口统计和特征如表所示 S1 （支持信息）。我们根据原始研究定义的 TBI 诊断、痴呆病理学和痴呆临床诊断对参与者进行分类^{。 12 ]} （数字 1a ）。利用来自四个大脑区域（颞叶皮层 [TCx]、顶叶皮层 [PCx]、额叶白质 [FWM] 和海马体 [HIP]）的测序数据，我们采用主成分分析进行了初步分析，并揭示了不同脑区中不同的基因表达模式。海马体与其他大脑区域的比较（图 S1a ，支持信息）。随后，我们对 AT8 水平进行了广泛的分析（映射 Ser202 或 Thr205 处的 tau 磷酸化；图 1b ;数字 S1b–d ，支持信息）和大脑ptau181/tau比值（图 1c ;数字 S1e–g ，支持信息）跨越上述大脑区域，比较 TBI 和病理组中的“临床痴呆”和“非痴呆”队列。 AT8 和 ptau181/tau 比率在“临床痴呆”组和“非痴呆”组之间仅在海马内表现出显着差异（图 1b,c ），表明该大脑区域可能是受 TBI 影响的主要区域。我们还检查了与 Aβ 病理学相关的变化，发现各组之间没有显着差异（图 S2 ，支持信息），从而表明该研究中的病理学可能与 CTE 一致。

We next examined the transcriptome changes in the cohort, employing bioinformatics analysis specifically within participants originally described as “diagnosed with TBI” and with “dementia pathology”. We then compared “clinical dementia” and “non‐demented” groups (Figure 1d–f; Figure S3, Supporting Information). Through the enrichment of differentially expressed genes (Figure 1d; Figure S3a, Supporting Information), we identified microtubule‐, axoneme‐, and cilium‐related pathways as the most upregulated (Figure 1e), consistent with the results from Gene Set Enrichment Analysis (GSEA) (Figure S3b, Supporting Information). Furthermore, all genes were categorized into modules according to their expression levels and correlation coefficients. We calculated the p‐values for gene modules associated with dementia diagnosis and Braak. The module most relevant to these phenotypes (Figure S3c, Supporting Information, Diagnose: r = 0.58, p = 0.003; Braak: r = 0.54, p = 0.006) underwent further Gene Ontology (GO) analysis (String, version 11.5) (Figure S3d, Supporting Information). The results mirrored those observed in the pathways shown in Figure 1e, underscoring the importance of these pathways.
接下来，我们检查了该队列中转录组的变化，特别是在最初被描述为“诊断患有 TBI”和“痴呆病理学”的参与者中进行了生物信息学分析。然后我们比较了“临床痴呆”和“非痴呆”组（图 1d–f ;数字 S3 ，支持信息）。通过差异表达基因的富集（图 1d ;数字 S3a ，支持信息），我们确定微管、轴丝和纤毛相关途径是上调最多的途径（图 1e ），与基因集富集分析（GSEA）的结果一致（图 S3b ，支持信息）。此外，所有基因根据其表达水平和相关系数被分类为模块。我们计算了与痴呆诊断和 Braak 相关的基因模块的p值。与这些表型最相关的模块（图 S3c ，支持信息，诊断：r = 0.58， p = 0.003； Braak：r = 0.54， p = 0.006）进行了进一步的基因本体（GO）分析（字符串，版本 11.5）（图 S3d ，支持信息）。结果反映了在图所示的路径中观察到的结果 1e ，强调了这些途径的重要性。

Through the analysis of degree rank results in the protein‐protein interaction (PPI) network, we further identified Axonemal dynein light intermediate polypeptide 1 (DNALI1) as the most critical gene in the gene module (Figure 1f, Degree = 42). Indeed, DNALI1 expression was significantly elevated in the “clinical dementia” group (TBI diagnosis and dementia pathology) (Figure 1g, p = 0.0067). When comparing DNALI1 expression at different durations of loss of consciousness, we observed a significant difference between the “clinical dementia” and “non‐demented” groups, particularly when the duration of loss of consciousness exceeded 10 min (Figure 1h, p = 0.047). Importantly, DNALI1 expression was significantly correlated with AT8 (Figure S4a, Supporting Information, r = 0.53, p = 0.011) and the ratio of ptau181/tau (Figure S4b, Supporting Information, r = 0.68, p = 0.002) in participants with TBI‐“clinical dementia”, whereas such correlations were absent in “non‐demented” participants. Furthermore, the expression of DNALI1 in TBI participants exhibited great predictive ability for dementia, which was even better than that of the Braak stage score (Figure 1i, AUC = 0.81). Collectively, these findings indicate that DNALI1 is a critical gene for the development of neurodegeneration after TBI.
通过对蛋白质-蛋白质相互作用（PPI）网络中的程度排序结果的分析，我们进一步确定了Axonemal dynein轻中间多肽1（DNALI1）是该基因模块中最关键的基因（图 1f ，度= 42）。事实上， DNALI1表达在“临床痴呆”组中显着升高（TBI 诊断和痴呆病理）（图 1g ， p = 0.0067）。当比较不同意识丧失持续时间的DNALI1表达时，我们观察到“临床痴呆”组和“非痴呆”组之间存在显着差异，特别是当意识丧失持续时间超过 10 分钟时（图 1h ， p = 0.047）。重要的是， DNALI1表达与 AT8 显着相关（图 S4a ，支持信息，r = 0.53， p = 0.011）和 ptau181/tau 的比率（图 S4b ，支持信息，r = 0.68， p = 0.002）在患有 TBI“临床痴呆”的参与者中，而在“非痴呆”参与者中则不存在这种相关性。此外，TBI参与者中DNALI1的表达表现出对痴呆症的强大预测能力，甚至优于Braak阶段评分（图 1i ，AUC = 0.81）。总的来说，这些发现表明DNALI1是 TBI 后神经退行性疾病发生的关键基因。

2.2. Altered DNALI1 Protein in In Vitro and In Vivo TBI Models
2.2.体外和体内 TBI 模型中 DNALI1 蛋白的改变

DNALI1 is a component of axonemal dynein, responsible for transporting cargo along the axoneme of eukaryotic cilia and flagella in conjunction with other components.^{[ 13 ]} Despite limited studies connecting DNALI1 to brain diseases, we explored whether DNALI1 expression is altered under conditions mimicking TBI. Serum deprivation has been previously used to simulate neuronal stress after TBI,^{[ 14 ]} and our immunofluorescence assay revealed a significant increase in DNALI1‐positive staining (Figure  2a, p = 0.012), which was further confirmed by western blot analysis. Prolonged serum deprivation further amplified DNALI1 expression (Figure 2b), indicating its responsiveness to neuronal stress. Considering the previous correlation between AT8 and DNALI1 in human patients, we examined tau expression and phosphorylation. The expression of tau was unchanged (Figure 2c, p = 0.99), while the AT8/tau ratio increased (Figure 2c, p = 0.0094), aligning with previous findings (Figure S4a, Supporting Information).
DNALI1 是轴丝动力蛋白的一个组成部分，负责与其他组件一起沿着真核纤毛和鞭毛的轴丝运输货物。 ^{[ 13} 尽管将 DNALI1 与脑部疾病联系起来的研究有限，^但我们探索了 DNALI1 表达在模拟 TBI 的条件下是否会发生改变。血清剥夺以前曾被用来模拟 TBI 后的神经元应激， ^{[ 14 ]}并且我们的免疫荧光检测显示 DNALI1 阳性染色显着增加（图 2a ， p = 0.012），通过蛋白质印迹分析进一步证实。长时间的血清剥夺进一步放大了 DNALI1 的表达（图 2b ），表明其对神经元应激的反应。考虑到之前人类患者中 AT8 和 DNALI1 之间的相关性，我们检查了 tau 表达和磷酸化。 tau的表达没有变化（图 2c ， p = 0.99），而 AT8/tau 比率增加（图 2c ， p = 0.0094），与之前的发现一致（图 S4a ，支持信息）。

Our analysis highlights the predominance of tau pathology in the human study, with Abeta pathologies remaining unaltered. To further investigate, we established an animal model of repetitive traumatic brain injuries (rTBIs) mimicking CTE. We employed closed‐skull, two mild replicate hit models with specific hit parameters (3.0 m s⁻¹ strike velocity, 1.0 mm strike depth, and 500 ms) and an experiment time after injury of 30 days, (Figure 2d) based on a literature survey^{[ 15 ]} and pre‐experimentation (Figure S5, Supporting Information). We monitored blood flow using laser speckle imaging and assessed changes in cognitive performance in the model. Blood flow in the brains of mice significantly decreased after injury (Figure 2e), confirming the success of the model. Under the parameters used in this study, the rTBIs model did not induce motor dysfunction, as evidenced by the unaltered latency to fall in the rotarod test compared to controls (Figure 2f, p = 0.62). Conversely, mice exhibited significant cognitive impairment, as evidenced by significantly reduced alternation in the Y‐maze (Figure 2g, p < 0.001), reduced discrimination index in the novel object recognition (NOR) test (Figure 2h, p = 0.018), and prolonged escape latency in the Morris water maze (MWM) (Figure 2i). Subsequent to modeling, we found significant increases in DNALI1 expression in both the cortex and hippocampus (Figure 2j, cortex, p = 0.012; hippocampus, p = 0.011), along with an increased AT8/tau ratio (Figure 2k, p < 0.001).
我们的分析强调了 tau 病理学在人类研究中的主导地位，而 Abeta 病理学保持不变。为了进一步研究，我们建立了模仿 CTE 的重复性创伤性脑损伤 (rTBIs) 动物模型。我们采用了闭合颅骨、两个具有特定击打参数（3.0 ms ^-1击打速度、1.0 mm 击打深度和 500 ms）的轻度重复击打模型，以及受伤后 30 天的实验时间（图 2d ）基于文献调查^{[ 15 ]}和预实验（图 S5 ，支持信息）。我们使用激光散斑成像监测血流并评估模型中认知能力的变化。小鼠受伤后脑部血流量显着减少（图 2e ），证实了模型的成功。在本研究中使用的参数下，rTBIs 模型不会诱发运动功能障碍，这一点可以通过与对照组相比在旋转杆测试中跌倒的潜伏期不变来证明（图 2f ， p = 0.62）。相反，小鼠表现出明显的认知障碍，Y 迷宫中的交替显着减少就证明了这一点（图 2g ， p < 0.001），新物体识别（NOR）测试中的辨别指数降低（图 2h ， p = 0.018），并且莫里斯水迷宫（MWM）中的逃避潜伏期延长（图 2i ）。建模后，我们发现皮质和海马中 DNALI1 表达显着增加（图 2j ，皮质， p = 0.012；海马体， p = 0.011），同时 AT8/tau 比率增加（图 2k ， p< 0.001)。

2.3. DNALI1 Knockdown Prevents Neurodegeneration after Repeated Mild Closed‐Head Injury
2.3. DNALI1敲低可预防反复轻度闭合性头部损伤后的神经退行性变

To determine whether DNALI1 contributes to neurodegeneration induced by repeated mild closed‐head injuries, we knocked‐down DNALI1 in the brain by employing stereotaxic injection of adeno‐associated virus (AAV8)‐EF‐Cas9 + AAV8‐mDNALI1‐sp.g3 (1.0 × 10^12 GC mL⁻¹) or AAV8 empty vectors as the control (Figure  3a). This injection was administered 14 days before inducing head injury. Immunofluorescence analysis confirmed the efficacy of AAV8‐mDNALI1‐sp.g3 in attenuating the increase in DNALI1 protein expression after head injury (Figure 3b, p < 0.0001). Subsequent motor and cognitive performance assessments revealed that the knockdown of DNALI1 did not alter motor function (Figure 3c, p < 0.95) but successfully rescued the cognitive impairment resulting from repeated mild closed head injury (Figure 3d–f). Consistent with this finding, western blot analysis of total tau and its phosphorylation revealed that the AT8/tau ratio significantly decreased after AAV injection (Figure 3g, p = 0.0090), consistent with the rescue of cognitive impairment. Taken together, these results suggest that targeting DNALI1 after a head injury may offer a promising avenue to prevent further neurodegeneration.
为了确定DNALI1是否会导致反复轻度闭合性头部损伤引起的神经变性，我们通过立体定位注射腺相关病毒 (AAV8)-EF-Cas9 + AAV8-mDNALI1-sp.g3 (1.0 × 10^12 GC mL ⁻¹ )或AAV8空载体作为对照(图 3a ）。这种注射是在引起头部损伤前 14 天进行的。免疫荧光分析证实了 AAV8-mDNALI1-sp.g3 在减弱头部损伤后 DNALI1 蛋白表达增加方面的功效（图 3b ， p< 0.0001)。随后的运动和认知表现评估显示， DNALI1的敲低并没有改变运动功能（图 3c ， p < 0.95），但成功挽救了重复轻度闭合性头部损伤导致的认知障碍（图 3d–f ）。与这一发现一致的是，总 tau 及其磷酸化的蛋白质印迹分析表明，注射 AAV 后 AT8/tau 比率显着下降（图 3g ， p = 0.0090），与认知障碍的挽救一致。总而言之，这些结果表明，头部受伤后靶向DNALI1可能为预防进一步的神经退行性疾病提供一个有希望的途径。

2.4. DNALI1 Affects the Autophagosome‐Lysosome Fusion and Autophagic Flux in TBI
2.4. DNALI1 影响 TBI 中自噬体-溶酶体融合和自噬通量

Given the primary involvement of DNALI1 in axonal transport, particularly in powering the beating of cilia through axonemal dynein,^{[ 16 ]} we examined differences in intraflagellar transport (IFT)‐related genes between “clinical dementia” and “non‐demented” in the Aging, Dementia and Traumatic Brain Injury Study. Notably, IFT protein 46 (IFT46), a component of the IFT complex, was significantly increased in participants with “clinical dementia” (Figure  4a, p = 0.020), accompanied by a significant positive correlation with DNALI1 (Figure 4b, r = 0.46, p = 0.0019). This correlation was not observed in non‐dementia participants, suggesting that DNALI1 may regulate IFT46 in the context of the disease.
鉴于 DNALI1 主要参与轴突运输，特别是通过轴突动力蛋白为纤毛的跳动提供动力， ^{[ 16 ]}我们在衰老、痴呆和创伤性脑损伤研究中检查了“临床痴呆”和“非痴呆”之间鞭毛内运输（IFT）相关基因的差异。值得注意的是，IFT 蛋白 46（ IFT46 ）（IFT 复合物的一个组成部分）在患有“临床痴呆”的参与者中显着增加（图 4a ， p = 0.020），并与DNALI1呈显着正相关（图 4b ，r = 0.46， p = 0.0019）。在非痴呆症参与者中没有观察到这种相关性，这表明DNALI1可能在疾病背景下调节IFT46 。

IFT46, recognized as a core component of the intraflagellar transport machinery, plays a role in ciliogenesis. It is crucial for the IFT complex's movement along the microtubule‐based axoneme of cilia and flagella, transporting proteins and other molecules necessary for the assembly and maintenance of these structures during ciliogenesis.^{[ 17 ]} Indeed, the presence of detectable primary cilia, indicative of changes in ciliogenesis, increased by 400% 72 h after serum removal (Figure 4c, p = 0.0011).
IFT46 被认为是鞭毛内运输机制的核心组成部分，在纤毛发生中发挥作用。它对于 IFT 复合物沿着纤毛和鞭毛的微管轴丝运动至关重要，在纤毛发生过程中运输组装和维持这些结构所需的蛋白质和其他分子。 ^{[ 17 ]}事实上，在去除血清后 72 小时，可检测到的初级纤毛的存在增加了 400%，表明纤毛发生的变化（图 4c ， p = 0.0011）。

Previous reports have indicated that ciliogenesis regulates autophagy via the Hedgehog pathway, inducing autophagy by directly acting on essential autophagy‐related proteins strategically located at the cilium's base through ciliary trafficking proteins.^{[ 17a ]} Therefore, we hypothesized that DNALI1 participated in autophagy as part of the ciliogenesis process. We identified markers of autophagy after serum deprivation, including an elevated LC3II/LC3I ratio, decreased p62 expression (Figure S6a, Supporting Information), an increased number of autophagic vesicles (Figure S6b, Supporting Information, p < 0.001), and a notably elevated fluorescence intensity of LC3 (Figure S6c, Supporting Information). Meanwhile, we found that DNALI1 expression can be reduced by pretreatment with the autophagy inhibitor 3‐MA (1 µm, 6 h) during serum deprivation (Figure 4d, p = 0.021). This suggests that DNALI1 may be functional within the autophagy process.
先前的报告表明，纤毛发生通过 Hedgehog 途径调节自噬，通过纤毛运输蛋白直接作用于战略性位于纤毛基部的重要自噬相关蛋白，从而诱导自噬。 ^{[ 17a ]}因此，我们假设 DNALI1 参与自噬作为纤毛发生过程的一部分。我们鉴定了血清剥夺后自噬的标志物，包括 LC3II/LC3I 比率升高、p62 表达降低（图 S6a ，支持信息），自噬囊泡数量增加（图 S6b ，支持信息， p < 0.001），并且 LC3 的荧光强度显着升高（图 S6c ，支持信息）。同时，我们发现在血清剥夺期间用自噬抑制剂3-MA（ 1μm ，6小时）预处理可以降低DNALI1的表达（图 4d ， p = 0.021）。这表明 DNALI1 可能在自噬过程中发挥作用。

To further clarify the role of DNALI1 in autophagy, we employed lentiviruses to overexpress DNALI1 (Lv DNALI1) (Figure 4e, p = 0.012). This aggressive overexpression of DNALI1 led to a significant increase in ciliogenesis (Figure 4f, p = 0.001), potentially impacting autophagic flux. The characteristics of autophagic flux (LC3II/LC3I ratio and p62 expression) exhibited no significant changes between the control and DNALI1 overexpression cells when serum was added, with or without 6 h pretreatment of protease inhibitors (20 mm NH₄Cl and 100 µm Leupeptin) (Figure 4g). Protease inhibitors block the degradation of autophagic cargo, promoting autophagic flux (changes in LC3‐II content after blocking lysosomal degradation).^{[ 18 ]} However, we observed a significant reduction in autophagic flux upon serum removal in the DNALI1 overexpression cells (Figure 4h), indicating that the impact of elevated DNALI1 was specific to pathological conditions, such as nutrient deprivation.
为了进一步阐明DNALI1在自噬中的作用，我们采用慢病毒过表达DNALI1 （Lv DNALI1）（图 4e ， p = 0.012）。 DNALI1的这种积极过度表达导致纤毛发生显着增加（图 4f ， p = 0.001），可能影响自噬通量。当添加血清时，无论有或没有蛋白酶抑制剂（20 m m NH ₄ Cl和100 µm ）预处理6小时，自噬流特征（LC3II/LC3I比值和p62表达）在对照和DNALI1过表达细胞之间没有表现出显着变化。亮肽素）（图 4g ）。蛋白酶抑制剂阻止自噬货物的降解，促进自噬通量（阻止溶酶体降解后 LC3-II 含量的变化）。 ^{[ 18 ]}然而，我们观察到 DNALI1 过表达细胞中血清去除后自噬通量显着减少（图 4h ），表明 DNALI1 升高的影响特定于病理条件，例如营养缺乏。

To further verify this, we transfected cells with a pH‐sensitive reporter (mCherry–GFP–LC3; a fusion of mCherry, green fluorescent protein, and LC3) that highlighted autophagosomes as yellow puncta and autophagolysosomes (post‐lysosomal fusion) as red puncta. We found that the basal levels of autophagic vacuoles were comparable in control and DNALI1 overexpression cells, consistent with previous western blotting results. However, upon serum removal, cells overexpressing DNALI1 exhibited a significant reduction in autophagolysosome content (Figure 4i, p < 0.001). The impact of DNALI1 on autophagosome‐lysosome fusion and autophagic flux also affected cell viability following serum deprivation, with DNALI1 overexpression rendering cells more susceptible (Figure 4j, p < 0.001).
为了进一步验证这一点，我们用 pH 敏感报告基因（mCherry-GFP-LC3；mCherry、绿色荧光蛋白和 LC3 的融合体）转染细胞，该报告基因将自噬体突出显示为黄色点，将自噬溶酶体（溶酶体融合后）突出显示为红色点。我们发现对照细胞和 DNALI1 过表达细胞中自噬泡的基础水平相当，这与之前的蛋白质印迹结果一致。然而，去除血清后，过表达 DNALI1 的细胞表现出自噬溶酶体含量显着减少（图 4i ， p< 0.001)。 DNALI1 对自噬体-溶酶体融合和自噬通量的影响也会影响血清剥夺后的细胞活力，DNALI1 过度表达使细胞更容易受到影响（图 4j ， p< 0.001)。

Autophagy is also associated with several cell death pathways, including apoptosis,^{[ 19 ]} necroptosis,^{[ 20 ]} and ferroptosis.^{[ 21 ]} Inhibition of autophagy can lead to the accumulation of damaged organelles, prompting autophagic vacuolization and ultimately triggering cell death pathways.^{[ 22 ]} Here, we found that DNALI1 overexpression also affected cell susceptibility to these cell death pathways under conditions of serum deprivation (Figure 4k), while such effects were not observed in normal serum conditions (Figure S7a, Supporting Information). Additionally, we analyzed the influence of DNALI1 overexpression on autophagy regulators. However, we only observed significant differences when VPS34‐IN1 was used (Figure 4l), likely due to its potent inhibitory effect on both VPS34 and DNALI1. Importantly, the normal serum status did not alter the serum levels (Figure S7b, Supporting Information).
自噬还与多种细胞死亡途径相关，包括细胞凋亡， ^{[ 19 ]}坏死性凋亡， ^{[ 20 ]}和铁死亡。 ^{[ 21 ]}自噬的抑制会导致受损细胞器的积累，促进自噬空泡化并最终触发细胞死亡途径。 ^{[ 22 ]}在这里，我们发现 DNALI1 过表达也会影响细胞在血清剥夺条件下对这些细胞死亡途径的敏感性（图 4k ），而在正常血清条件下没有观察到这种效应（图 S7a ，支持信息）。此外，我们还分析了 DNALI1 过表达对自噬调节因子的影响。然而，我们仅在使用 VPS34-IN1 时观察到显着差异（图 4l ），可能是由于其对 VPS34 和 DNALI1 都有有效的抑制作用。重要的是，正常的血清状态不会改变血清水平（图 S7b ，支持信息）。

To further validate the effect of DNALI1 on autophagy, we employed siRNA to inhibit DNALI1 expression (Si DNALI1) (Figure  5a, p = 0.0055). Autophagic flux, as indicated by the LC3II/LC3I ratio and p62 expression, significantly increased during serum deprivation, whereas no such effect was observed under normal serum conditions (Figure 5b). The effects of DNALI1 on autophagic flux also affected cell viability following serum deprivation, with cells exhibiting reduced DNALI1 displaying increased resistance (Figure 5c, p = 0.033). Moreover, we also found that DNALI1 reduction affected cell resistance to cell death and autophagy regulators under serum‐deprived conditions (Figure 5d,e). In contrast, only autophagy regulators were affected in normal serum (Figure S8, Supporting Information). Additionally, experiments conducted using a previously constructed AAV8‐Dnali1 KD animal model showed that LC3 and p62 immunofluorescence results indicated a significant increase in autophagic flux upon Dnali1 KD (Figure 5f,g).
为了进一步验证DNALI1对自噬的影响，我们采用siRNA抑制DNALI1的表达（Si DNALI1）（图 5a ， p = 0.0055）。如 LC3II/LC3I 比率和 p62 表达所示，自噬通量在血清剥夺期间显着增加，而在正常血清条件下未观察到这种效应（图 5b ）。 DNALI1 对自噬流的影响也会影响血清剥夺后的细胞活力，DNALI1 减少的细胞表现出抵抗力增加（图 5c ， p = 0.033）。此外，我们还发现DNALI1的减少影响了血清剥夺条件下细胞对细胞死亡和自噬调节因子的抵抗力（图 5d,e ）。相比之下，正常血清中只有自噬调节因子受到影响（图 S8 ，支持信息）。此外，使用先前构建的 AAV8- Dnali1 KD 动物模型进行的实验表明，LC3 和 p62 免疫荧光结果表明Dnali1 KD 后自噬通量显着增加（图 5f,g ）。

Reagents  试剂

Reagents were purchased from Sigma‐Aldrich unless otherwise specified.
除非另有说明，试剂均购自 Sigma-Aldrich。

Data Acquisition and Study Design
数据采集​​和研究设计

Publicly available datasets were obtained from the Aging, Dementia, and Traumatic Brain Injury Study (http://aging.brain‐map.org/),^{[ 12 ]} which provides a systematic and extensive dataset encompassing specimen metadata, histology, immunohistochemistry, in situ hybridization, RNA sequencing, protein quantification, and isoprostane quantification for 107 individuals. Participants were categorized into eight groups based on TBI diagnosis, dementia pathology (CERAD>1), and clinical dementia diagnosis (using the NINCDS/ADRDA criteria).
公开可用的数据集来自衰老、痴呆和创伤性脑损伤研究（ http://aging.brain‐map.org/ ), ^{[ 12 ]}它提供了系统且广泛的数据集，涵盖 107 名个体的标本元数据、组织学、免疫组织化学、原位杂交、RNA 测序、蛋白质定量和异前列烷定量。根据 TBI 诊断、痴呆病理学 (CERAD>1) 和临床痴呆诊断（使用 NINCDS/ADRDA 标准）将参与者分为八组。

Bioinformatics Analysis  生物信息学分析

Differential expression was assessed using a linear model and the Bioconductor limma package^{[ 40 ]} in R, version 3.6.3 (R Foundation) based on a |fold change| ≥ 0.5 and p‐value < 0.05. The results were input into the Metascape database^{[ 41 ]} (https://metascape.org/gp/index.html) to explore Gene Ontology (GO) pathways. Heatmaps were generated using the Pheatmap package^{[ 42 ]} in R, version 3.6.3 (R Foundation), with z‐scores calculated for each gene row using the mean expression of biological replicates. Volcano plots, illustrating differentially expressed genes, were generated using the Enhanced Volcano package^{[ 43 ]} in R, version 3.6.3 (R Foundation). The network diagram for PPI was calculated and visualized using Cytoscape (https://www.cytoscape.org/), and hub genes were ranked using cytoHubba.^{[ 44 ]} Gene Set Enrichment Analysis (GSEA) (version 4.0.3, https://www.broadinstitute.org/gsea/)^{[ 45 ]} was employed to identify overrepresented GO pathways (MSigDB, version 7.1) for upregulated or downregulated genes. Co‐expression networks for RNA‐seq data and their correlation values with phenotypic data were constructed and calculated using the weighted gene co‐expression network analysis (WGCNA) package^{[ 46 ]} in R, version 3.6.3 (R Foundation).
使用线性模型和 Bioconductor limma 包评估差异表达^{[ 40 ]}在 R 版本 3.6.3 (R Foundation) 中基于 |折叠更改| ≥ 0.5 且p值 < 0.05。结果被输入Metascape数据库^{[ 41 ]} ( https://metascape.org/gp/index.html ）探索基因本体（GO）途径。热图是使用 Pheatmap 包生成的^{[ 42 ]}在 R 版本 3.6.3（R Foundation）中，使用生物重复的平均表达为每个基因行计算 z 分数。使用增强型火山包生成火山图，说明差异表达的基因^{。 43 ]}在 R 版本 3.6.3（R 基金会）中。使用 Cytoscape 计算并可视化 PPI 的网络图（ https://www.cytoscape.org/ ），并使用 cytoHubba 对 hub 基因进行排序。 ^{[ 44 ]}基因集富集分析 (GSEA)（版本 4.0.3， https://www.broadinstitute.org/gsea/ ） ^{[ 45 ]}用于识别上调或下调基因的过度表达的 GO 途径（MSigDB，版本 7.1）。使用加权基因共表达网络分析（WGCNA）包构建并计算了RNA-seq数据的共表达网络及其与表型数据的相关值^{[ 46 ]}在 R 版本 3.6.3（R 基金会）中。

Cell Culture  细胞培养

The neuronal cell line STHdhQ7/Q7 (a gift from Dr. Boxun Lu, Fudan University) was cultured in Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 10% fetal calf serum and 100 Units/ml penicillin/streptomycin (all purchased from Thermo Fisher Scientific, USA).
神经元细胞系STHdhQ7/Q7（复旦大学陆伯勋博士赠品）在添加有10％胎牛血清和100单位/ml青霉素/链霉素的Dulbecco's Modified Eagle's Medium（DMEM）中培养（均购自Thermo Fisher）科学，美国）。

In Vitro TBI Model  体外 TBI 模型

The serum deprivation model was employed to replicate neuronal stress after TBI, as previously described.^{[ 14 ]} Briefly, cells were seeded in 6/96‐well plates (5×10⁴ cells/mL). After plating, the cell culture medium was removed, washed three times with DPBS (Thermo Fisher Scientific, USA), and then replaced with a serum‐free medium. The duration of serum deprivation was adjusted according to experimental requirements, as indicated in the figure legends.
如前所述，血清剥夺模型用于复制 TBI 后的神经元应激。 ^{[ 14 ]}简而言之，将细胞接种到6/96孔板中（5×10 ^{4 个}细胞/mL）。铺板后，除去细胞培养基，用DPBS（Thermo Fisher Scientific，美国）洗涤3次，然后更换为无血清培养基。根据实验要求调整血清剥夺的持续时间，如图图例所示。

Cell Viability Assay  细胞活力测定

Cells were seeded into 96‐well plates (5×10⁴ cells/mL) and treated with the selected compounds [Rapamycin (Cat No. S1039, Selleck Chemicals), 3‐Methyladenine (Cat No. S2767, Selleck Chemicals), VPS34‐IN1 (Cat No. S7980, Selleck Chemicals), RSL3 (Cat No. S8155, Selleck Chemicals), Staurosporine (Cat No. S1421, Selleck Chemicals), (20S)‐Protopanaxadiol (Cat No. S4746, Selleck Chemicals)] after plating. Cell viability was assessed at different time points after treatment (24 h unless otherwise specified) using the Cell Counting Kit‐8 (CCK‐8) cytotoxicity assay (Bimake, B34304), as previously described.^{[ 47 ]}
将细胞接种到 96 孔板（5×10 ^{4 个}细胞/mL）中，并用选定的化合物[雷帕霉素（货号 S1039，Selleck Chemicals）、3-甲基腺嘌呤（货号 S2767，Selleck Chemicals）、VPS34- 进行处理。 IN1（货号 S7980，Selleck Chemicals），RSL3（货号 S8155， Selleck Chemicals）、星形孢菌素（货号 S1421，Selleck Chemicals）、（20S）-原人参二醇（货号 S4746，Selleck Chemicals）] 电镀后。使用细胞计数试剂盒-8（CCK-8）细胞毒性测定（Bimake， B34304 ），如前所述。 ^{[ 47 ]}

Transmission Electron Microscopy
透射电子显微镜

After treatment, cells were collected into a 1.5 mL EP tube, and a low‐speed centrifugation step was employed to settle the cells at the bottom of the EP tube. Subsequently, the supernatants were removed. The collected cell pellets were fixed with 2.5% glutaraldehyde for 4 h at 4 °C and post‐fixed in 1% osmium tetroxide for 2 h at 20 °C. Then, they were dehydrated using a gradient of ethanol (50–100%) and acetone, embedded in epoxy resin, and polymerized for 48 h at 60 °C. Ultrathin sections (80 nm) were cut and stained with uranyl acetate and lead citrate prior to transmission electron microscopy (HT7700, HITACHI). Images were captured using a SlowScan CCD camera and the iTEM software (Ver 01.07, Olympus Soft Imaging Solutions). The quantification of autophagic bodies followed established procedures as previously described.^{[ 48 ]}
处理后，将细胞收集至1.5 mL EP管中，并采用低速离心步骤使细胞沉降在EP管底部。随后，除去上清液。收集的细胞沉淀用2.5％戊二醛在4℃下固定4小时，并在1％四氧化锇中在20℃下固定2小时。然后，使用乙醇（50-100%）和丙酮的梯度对它们进行脱水，包埋在环氧树脂中，并在60°C下聚合48小时。在透射电子显微镜（HT7700，HITACHI）之前切割超薄切片（80 nm）并用乙酸双氧铀和柠檬酸铅染色。使用 SlowScan CCD 相机和 iTEM 软件（版本 01.07，Olympus Soft Imaging Solutions）捕获图像。自噬体的定量遵循先前描述的既定程序。 ^{[ 48 ]}

siRNA Transfection  siRNA转染

Cells were seeded onto 6/96‐well plates (5×10⁴ cells/mL) and transfected with siRNA. The target sequences for the siRNA used in this study are shown in Table S2 (Supporting Information). The final siRNA concentration used was 10 nm, and the transfection was carried out using RNAiMax (Invitrogen, Cat No. 13778150) following the manufacturer's specifications. Follow‐up experiments were continued 24 h after transfection, as described previously.^{[ 49 ]}
将细胞接种到6/96孔板（5×10 ^{4 个}细胞/mL）并用siRNA转染。本研究中使用的 siRNA 的靶序列如表所示 S2 （支持信息）。使用的最终siRNA浓度为10 nm ，并使用RNAiMax（Invitrogen，目录号13778150）按照制造商的说明书进行转染。如前所述，转染后 24 小时继续进行后续实验。 ^{[ 49 ]}

Animals  动物

Eight‐week‐old male C57BL/6 mice were obtained from Beijing HFK Bioscience Co., Ltd. and housed in a specific pathogen‐free facility at the State Key Laboratory of Biotherapy (Sichuan University, China). The mice underwent an adaptive feeding period of one week prior to the commencement of the experiments. All mouse‐related procedures were conducted in accordance with the Institutional Guidelines of the Animal Care and Use Committee (K2018071, Sichuan University, China).
8周龄雄性C57BL/6小鼠获自北京华富康生物科技有限公司，饲养在生物治疗国家重点实验室（中国四川大学）的无特定病原体设施中。实验开始前，小鼠经历了一周的适应性喂养期。所有小鼠相关程序均按照动物护理和使用委员会的机构指南（K2018071，四川大学，中国）进行。

Repetitive Traumatic Brain Injuries
重复性脑外伤

A mouse model involving closed skulls and repetitive mild head injuries was used, as described previously.^{[ 15} , ^{50 ]} Briefly, mice were placed in a stereotaxic frame following anesthesia with isoflurane, 4% for induction and 1–2% for maintenance. The injury was induced using a 3 mm blunt metal impactor tip positioned at 1.8 mm caudal to bregma and 2.0 mm left of midline. The injury was triggered by an electromagnetically controlled cortical impact device (Custom Design & Fabrication, Inc, USA) with 3.0 m s⁻¹ for strike velocity, 1.0 mm for strike depth, and 500 ms for dwell time to the exposed skull. Post‐impact, the skin was sutured, disinfected with iodophor, and the mice were allowed to recover from anesthesia on a warming pad before being returned to their home cages. A second identical injury procedure was performed 24 h later. Sham injuries followed the same procedures and anesthesia, excluding the delivery of an impact.
如前所述，使用了包含闭合头骨和重复性轻度头部损伤的小鼠模型。 ^{[ 15} , ^{50 ]}简而言之，小鼠在异氟烷麻醉后被放置在立体定位架中，4% 用于诱导，1-2% 用于维持。使用位于前囟尾部 1.8 毫米、中线左侧 2.0 毫米的 3 毫米钝金属冲击器尖端诱发损伤。伤害是由电磁控制的皮质冲击装置（美国定制设计与制造公司）触发的，冲击速度为 3.0 m s ^-1 ，冲击深度为 1.0 mm，暴露头骨的停留时间为 500 ms。撞击后，缝合皮肤，用碘伏消毒，让小鼠在温暖的垫子上从麻醉中恢复，然后再返回它们的笼子。 24 小时后进行第二次相同的损伤手术。假手术伤害遵循相同的程序和麻醉，但不包括撞击。

Y‐Maze Spontaneous Alternation Test
Y 迷宫自发交替测试

The Y‐maze spontaneous alternation test was used to assess spatial working memory, as described previously.^{[ 51 ]} Briefly, one month after model establishment, each mouse was placed naive to the Y maze (39.5×8.5×13 cm, Sansbio co. ltd, China), at the same end of one arm and allowed to move freely through the maze during an 8‐minute session. The entire session was video‐recorded (Basler acA640‐120gm, Basler Vision Technology), and the number of arm entries for each mouse was subsequently calculated.
如前所述，Y 迷宫自发交替测试用于评估空间工作记忆。 ^{[ 51 ]}简而言之，模型建立一个月后，将每只小鼠置于 Y 迷宫（39.5×8.5×13 cm，Sansbio co. ltd，中国），在一只手臂的同一端，并允许其在迷宫中自由移动。 8 分钟的会议。整个过程被视频记录（Basler acA640-120gm，Basler Vision Technology），随后计算每只小鼠的手臂进入次数。

Morris Water Maze Test  莫里斯水迷宫测试

The Morris water maze (MWM) test was conducted to assess spatial learning, as previously described.^{[ 52 ]} Briefly, a circular water tank (diameter 80 cm) was filled with milk powder‐stained water, housing a concealed round platform (diameter 7 cm) positioned 1 cm below the water surface at the center of a specific quadrant. The test consisted of a place navigation test (five days) and a spatial probe test (one day). During the place navigation test stage, mice started from one of the four quadrants facing the pool wall and ended upon reaching the platform. If mice failed to locate the platform within 120 s, they were guided to it. In the subsequent spatial probe test, the platform was removed, and task performance was recorded for 120 s. Mouse movements in the water pool were recorded by a video camera (Basler acA640‐120gm, Basler Vision Technology), and task‐related metrics, including swimming paths, speed, and time spent in each quadrant, were recorded using WMT‐100 software (Chengdu Techman Software Co. Ltd, China).
如前所述，进行莫里斯水迷宫（MWM）测试是为了评估空间学习。 ^{[ 52 ]}简而言之，一个圆形水箱（直径80厘米）充满了奶粉染色的水，里面有一个隐藏的圆形平台（直径7厘米），位于特定象限中心水面以下1厘米处。该测试包括地点导航测试（五天）和空间探测测试（一天）。在位置导航测试阶段，小鼠从面向池壁的四个象限之一开始，并在到达平台时结束。如果小鼠未能在 120 秒内找到平台，它们就会被引导到平台。在随后的空间探测测试中，平台被移除，并记录任务表现120秒。通过摄像机（Basler acA640-120gm，Basler Vision Technology）记录小鼠在水池中的运动，并使用 WMT-100 软件记录与任务相关的指标，包括游泳路径、速度和在每个象限中花费的时间（中国成都达明软件有限公司）。

Novel Object Recognition Test
新物体识别测试

The novel object recognition (NOR) test was performed to evaluate memory retention, as described previously.^{[ 52 ]} Briefly, 24  h before the test, mice underwent a 5‐minute habituation period in the arenas (50 cm × 50 cm plastic container, Sansbio co. ltd, China) without objects. On the subsequent day, the mice re‐entered the arena from the same starting point (facing the bottom‐left corner) for the training and testing stages. In the first stage of the test, animals were confronted with two identical objects for 10 min. In the second stage, 1 h after the familiarization period, the animals were exposed to two dissimilar objects in the same open field for 5 min: one familiar object used in the first phase and another novel object. The time spent exploring each object in stage two was detected using Supermaze (ver 3.3, Shanghai XinRuan Information Technology co. ltd, China). The discrimination index (DI) was calculated using the following equation: DI = TN/ (TN + TF), where TN represents the exploration time devoted to the novel object and TF is the exploration time for the familiar object.
如前所述，进行新物体识别（NOR）测试来评估记忆保留情况。 ^{[ 52 ]}简而言之，测试前24小时，小鼠在没有物体的竞技场（50 cm×50 cm塑料容器，Sansbio co. ltd，中国）中经历5分钟的习惯期。第二天，小鼠从同一起点（面向左下角）重新进入竞技场进行训练和测试阶段。在测试的第一阶段，动物面对两个相同的物体 10 分钟。在第二阶段，熟悉期后1小时，将动物在同一开放场地中暴露于两个不同的物体5分钟：一个用于第一阶段的熟悉物体和另一个新物体。使用 Supermaze（3.3 版，上海欣软信息技术有限公司，中国）检测第二阶段探索每个对象所花费的时间。使用以下等式计算辨别指数（DI）：DI = TN/（TN + TF），其中TN表示用于新物体的探索时间，TF是用于熟悉物体的探索时间。

Rotarod Treadmill Test  旋转跑步机测试

The rotarod treadmill test was conducted to assess motor coordination, as described previously.^{[ 53 ]} One month after establishing the model, animals were placed on a rotarod treadmill (RWD, Shenzhen, China) in the accelerating rotor mode (10 speeds ranging from 4 to 40 rpm for 5 min). The interval from the moment the animal mounted the rod to when it fell off was recorded as the retention time, and mice that remained on the accelerating rotating rod for 300 s were recorded as survivors. Animals underwent training for two days, with three trials per day, before model establishment, and the mean duration on the rod was recorded to obtain stable baseline values. Performance on the rotarod test was measured three times a month after repeated mild closed‐head injuries.
如前所述，进行旋转跑步机测试是为了评估运动协调性。 ^{[ 53 ]}模型建立一个月后，将动物置于旋转跑步机（RWD，深圳，中国）上，以加速转子模式（从 4 到 40 rpm 的 10 种速度，持续 5 分钟）。从动物登上杆到脱落的时间记录为停留时间，在加速旋转杆上保持300秒的小鼠记录为幸存者。模型建立前，动物接受为期两天的训练，每天进行3次试验，并记录在杆上的平均持续时间以获得稳定的基线值。在重复轻度闭合性头部损伤后，每月测量三次旋转杆测试的表现。

Immunofluorescence  免疫荧光

For immunofluorescence of brain tissue sections, mice were anesthetized with pentobarbitone and perfused with ice‐cold PBS (0.01 m, pH 7.4). Brains were isolated, fixed in 4% paraformaldehyde in PBS overnight, and then sectioned into 4 µm paraffin sections. After staining with primary antibodies (Anti‐DNALI1, Abcam, ab155490, 1:500; Anti‐Acetylated Tubulin, Proteintech, 66200‐1‐Ig, 1:250) and fluorescent‐tagged secondary antibodies (Alexa Fluor 488 Goat Anti‐Mouse IgG, Jackson ImmunoResearch, 115‐005, 1:500; Cy3 Goat Anti‐Mouse IgG, Jackson ImmunoResearch, 115–165, 1:500), the nuclei were counterstained with 4,6‐diamidino‐2‐phenylindole (DAPI), and coverslips were applied. The labeled sections were captured using a confocal laser‐scanning microscope (Nikon ECLIPSE Ti‐S). Brightness and contrast adjustments were uniformly applied to the captured images using CaseViewer software, version 2.4 (3DHIESTECH Ltd).
对于脑组织切片的免疫荧光，用戊巴比妥麻醉小鼠并用冰冷的 PBS（0.01 m ，pH 7.4）灌注。分离大脑，在 PBS 中的 4% 多聚甲醛中固定过夜，然后切成 4 µm 石蜡切片。用一抗（抗 DNALI1，Abcam，ab155490，1:500；抗乙酰化微管蛋白，Proteintech，66200-1-Ig，1:250）和荧光标记二抗（Alexa Fluor 488 山羊抗小鼠 IgG）染色后，杰克逊免疫研究，115‐005，1:500； IgG，杰克逊免疫研究，115-165，1:500），细胞核用 4,6-二脒基-2-苯基吲哚 (DAPI) 复染，并盖上盖玻片。使用共焦激光扫描显微镜（Nikon ECLIPSE Ti-S）捕获标记的切片。使用 CaseViewer 软件 2.4 版（3DHIESTECH Ltd）对捕获的图像统一进行亮度和对比度调整。

For immunofluorescence, cells were seeded on cell‐climbing slides. After treatment, cells were fixed with 4% paraformaldehyde at room temperature for 30 min. Subsequently, cells were blocked for 1 h at room temperature with 10% normal goat serum (Solarbio, SL038) dissolved in 0.2% Triton X‐100 PBS and incubated with the primary antibody (DNALI1, Abcam, ab155490, 1:500) overnight at 4 °C in a humidified chamber. The following day, the secondary antibody (Alexa Fluor 488 AffiniPure Alpaca Anti‐Rabbit IgG (H+L), Jackson ImmunoResearch, 611‐545‐215, 1:500) and DAPI were used for imaging. Images were captured under a fluorescence microscope (OLYMPUS VS200, JPN) and quantified using ImageJ (, version 1.5.1 (NIH, USA).
对于免疫荧光，将细胞接种在细胞攀爬载玻片上。处理后，用4%多聚甲醛在室温下固定细胞30分钟。随后，用溶解在 0.2% Triton X-100 PBS 中的 10% 正常山羊血清 (Solarbio, SL038) 在室温下封闭细胞 1 小时，并与一抗 (DNALI1, Abcam, ab155490, 1:500) 一起在 1:500 下孵育过夜。 4°C 加湿室中。第二天，使用二抗（Alexa Fluor 488 AffiniPure Alpaca Anti-Rabbit IgG (H+L), Jackson Nutrition, 611-545-215, 1:500）和 DAPI 进行成像。在荧光显微镜（OLYMPUS VS200，日本）下捕获图像并使用 ImageJ（1.5.1 版（NIH，美国）进行定量）。

Western Blot  蛋白质印迹

Samples were homogenized in cell lysis buffer (P0013, Beyotime) supplemented with the protease inhibitor phenylmethylsulfonyl fluoride (1:100, ST507, Beyotime) and centrifuged at 12000 × g for 30 min. The supernatant was collected, and the total protein concentration was determined using a BCA protein assay kit (P0011, Beyotime). Equal amounts of protein were separated on 4–12% bis‐Tris gels with MOPs running buffer at 140 V for 1.5 h and then transferred to nitrocellulose membranes using a Trans‐Blot system at 100 V for 1 h. Then, the membranes were washed with 1 × TBST for 5 min at room temperature (20 °C), shaken, and blocked with 5% skim milk in 1 × TBST for 1 h at room temperature. The following antibodies were diluted in 1 × TBST as indicated: 1:3000 for anti‐LC3 (14600‐1‐AP, Proteintech), 1:3000 for anti‐P62 (18420‐1‐AP, Proteintech), 1:10 000 for anti‐mouse IgG (A9044, Sigma‐Aldrich), 1:10 000 for anti‐rabbit IgG (A0545, Sigma‐Aldrich), 1:2000 for anti‐DNALI1 (ab155490, Abcam & 17601‐1‐AP, Proteintech); 1:1000 for anti‐Tau5 (ab80579, Abcam); 1:1000 for Anti‐Phospho‐Tau (Ser202, Thr205) (MN1020, Thermo Fisher Scientific). All uncropped images of western blots are shown in Figure S9 (Supporting Information).
将样品在补充有蛋白酶抑制剂苯甲基磺酰氟（1:100，ST507，Beyotime）的细胞裂解缓冲液（P0013，Beyotime）中匀浆，并在 12000 × g 下离心 30 分钟。收集上清液，使用BCA蛋白测定试剂盒（P0011，Beyotime）测定总蛋白浓度。将等量的蛋白质在 4-12% bis-Tris 凝胶上用 MOP 运行缓冲液在 140 V 下分离 1.5 小时，然后使用 Trans-Blot 系统在 100 V 下转移到硝酸纤维素膜上 1 小时。然后用 1×TBST 室温（20℃）洗膜 5 min，振摇，用 5%脱脂牛奶溶于 1×TBST 室温封闭 1 h。以下抗体按所示在 1 × TBST 中稀释：抗 LC3（14600-1-AP，Proteintech）为 1:3000，抗 P62（18420-1-AP，Proteintech）为 1:3000，1:10 000用于抗小鼠 IgG (A9044, Sigma-Aldrich)，1:10 000 用于抗兔 IgG (A0545, Sigma-Aldrich)，1:2000 用于抗 DNALI1 (ab155490, Abcam & 17601-1-AP, Proteintech)； 1:1000 抗 Tau5（ab80579，Abcam）； 1:1000 抗 Phospho-Tau（Ser202、Thr205）（MN1020，Thermo Fisher Scientific）。所有未裁剪的蛋白质印迹图像如图所示 S9 （支持信息）。

Adeno‐Associated Viral Vector Construction and Preparation
腺相关病毒载体的构建和制备

To knockdown murine DNALI1 with a recombinant adeno‐associated virus (rAAV) in vivo, sgRNAs were designed using online tools (https://crispr.mit.edu/ and https://crispr.cos.uni‐heidelberg.de/). Off‐target effects were assessed using resources available at https://asia.ensembl.org/. sgRNA sequences with fewer off‐target sites were selected for further analysis. The target sequences of the sgRNA used in this study are shown in Table S3 (Supporting Information). The sgRNAs for murine DNALI1 knockdown were inserted into the plasmid, which was used to produce mDNALI1‐sp. g3‐rAAV8 (Obio Technology, Shanghai, China) was used to knockdown DNALI1, and the AAV8‐empty vector was used as a control. All rAAV8 vectors were generated using the triple‐plasmid co‐transfection method in human embryonic kidney 293 cells. After 72 h post‐transfection, the rAAV8 vectors were collected, purified through two rounds of CsCl gradient ultracentrifugation, followed by silver staining and genome copy titration. The viral vectors were aliquoted and stored at −80 °C before use.
为了在体内用重组腺相关病毒（rAAV）敲低小鼠 DNALI1，使用在线工具设计了 sgRNA（ https://crispr.mit.edu/ 和 https://crispr.cos.uni‐heidelberg.de/ ）。使用可用资源评估脱靶效应 https://asia.ensembl.org/ 。选择脱靶位点较少的 sgRNA 序列进行进一步分析。本研究中使用的sgRNA的靶序列见表 S3 （支持信息）。用于敲低小鼠 DNALI1 的 sgRNA 被插入质粒中，用于产生 mDNALI1-sp。 g3-rAAV8（Obio Technology，上海，中国）用于敲低DNALI1，AAV8-空载体用作对照。所有 rAAV8 载体均采用三质粒共转染方法在人胚肾 293 细胞中生成。转染72小时后，收集rAAV8载体，通过两轮CsCl梯度超速离心纯化，然后进行银染和基因组拷贝滴定。使用前将病毒载体等分并储存在-80°C。

AAV Injection  腺病毒注射液

For stereotaxic injection, mice were anesthetized by means of intraperitoneal injection of pentobarbital (100 mg k⁻¹g, P11011, Bioreagent) and fixed on a stereotaxic plate (RWD, Shenzhen, China). A hole was drilled into the bone using a hand drill (RWD, Shenzhen, China). Eight microliters of the virus, with a total titer of 1.0 × 10¹² genome copies, were injected into the lateral ventricle (AP = −0.6 mm, ML = −1.0 mm, DV = 2 mm) relative to the bregma. The needle was then held in place for an additional 10 min. After injection, the needle was withdrawn, and the wound was sutured.
对于立体定位注射，通过腹腔注射戊巴比妥（100 mg k ^-1 g，P11011，生物试剂）麻醉小鼠并固定在立体定位板上（RWD，深圳，中国）。使用手钻（RWD，中国深圳）在骨头上钻一个孔。将总效价为 1.0 × 10 ¹²基因组拷贝的 8 微升病毒注射到相对于前囟的侧脑室（AP = -0.6 mm，ML = -1.0 mm，DV = 2 mm）。然后将针继续固定 10 分钟。注射后，拔出针，缝合伤口。

Statistical Analysis  统计分析

Data are presented as individual values. Statistical analysis was conducted using GraphPad Prism 8.0 software (GraphPad Software, Inc., USA), and Student's t‐test was used to analyze statistical differences between two groups. One‐way ANOVA or repeated‐measures two‐way ANOVA with Tukey's post hoc test was used when appropriate. Values are presented as mean ± SEM, and individual data points represent individual samples or animals. Further details of the statistical analysis can be found in the figure legends. p‐value < 0.05 was considered statistically significant.
数据以单独值的形式呈现。使用GraphPad Prism 8.0软件（GraphPad Software，Inc.，美国）进行统计分析，并使用Student's t检验分析两组之间的统计差异。在适当的时候使用单因素方差分析或重复测量双向方差分析和 Tukey 事后检验。值以平均值±SEM表示，各个数据点代表各个样品或动物。统计分析的更多详细信息可以在图形图例中找到。 p值<0.05被认为具有统计学意义。