Abstract Posttraumatic neuroinflammation is a key driver of secondary injury after traumatic brain injury (TBI). Pyroptosis, a proinflammatory form of programmed cell death, considerably activates strong neuroinflammation and amplifies the inflammatory response by releasing inflammatory contents. Therefore, treatments targeting pyroptosis may have beneficial effects on the treatment of secondary brain damage after TBI. Here, a cysteine-alanine-glutamine-lysine peptide–modified β-lactoglobulin (β-LG) nanoparticle was constructed to deliver disulfiram (DSF), C-β-LG/DSF, to inhibit pyroptosis and decrease neuroinflammation, thereby preventing TBI-induced secondary injury. In the post-TBI mice model, C-β-LG/DSF selectively targets the injured brain, increases DSF accumulation, and extends the time of the systemic circulation of DSF. C-β-LG/DSF can alleviate brain edema and inflammatory response, inhibit secondary brain injury, promote learning, and improve memory recovery in mice after trauma. Therefore, this study likely provided a potential approach for reducing the secondary spread of TBI. __________________________________________________________________ C-β-LG/DSF inhibits pyroptosis and limits the release of inflammatory cytokines in treating secondary brain damage. INTRODUCTION Traumatic brain injury (TBI) is a series of injuries caused by external trauma to the head, such as intracranial hematoma, which can lead to disability or even death ([36]1). With the development of the economy and technology, the incidence, disability, and mortality of TBI are increasing during transportation, construction, sports injuries, and war ([37]2). There are more than 50 million patients with TBI worldwide yearly, and the incidence is increasing by 4.67% annually ([38]3). In addition to high mortality, patients with severe TBI exhibit varying degrees of cognitive function and behavioral dysfunction, leading to lifelong disability, bringing severe economic and health burdens to families and society ([39]2, [40]4–[41]6). The main reason for the aforementioned losses is secondary brain injury; however, there is no effective pharmacotherapy available to prevent the secondary spread of damage. TBI is categorized into primary and secondary brain injuries. Secondary brain injury occurs over a period of hours or days after the primary injury and involves changes in the brain’s cells, chemicals, tissues, and blood vessels, which can further damage brain tissue ([42]7–[43]9). TBI-induced neuroinflammation is the leading cause of secondary brain injury. Recent studies have highlighted that pyroptosis, a type of programmed cell death, is a key driver of TBI-induced inflammation and plays an important role in the pathogenesis of TBI injury ([44]10–[45]12). This process causes cells to swell and eventually burst, releasing inflammatory cytokines [such as interleukin-18 (IL-18) and IL-1β] and triggering a strong inflammatory response ([46]13–[47]15). Many studies have speculated that inhibiting TBI-induced pyroptosis could be an effective way of treating TBI ([48]16–[49]18). However, effective clinical therapeutic strategies for pyroptosis remain scarce. To date, there are scant data regarding the regulation of pyroptosis during the acute phase of TBI. Disulfiram (DSF), a Food and Drug Administration (FDA)–approved drug for the treatment of alcohol addiction, has been found to inhibit pyroptosis by blocking gasdermin D (GSDMD) pore formation ([50]19). However, insufficient drug targeting and retention in the brain injury area have led to the failure of drugs used for TBI therapies. Therefore, there is an urgent need to develop a delivery platform that can target and enhance drug accumulation in the brain injury area. Here, we developed a cysteine-alanine-glutamine-lysine (CAQK) peptide–modified β-lactoglobulin (β-LG) nanoparticle to deliver DSF (C-β-LG/DSF) for TBI therapy ([51]Fig. 1). C-β-LG/DSF represents a potential approach to inhibit pyroptosis and limit the release of inflammatory cytokines in treating secondary brain damage. C-β-LG/DSF consists of two components with different functions: (i) C-β-LG, a modified CAQK peptide, can penetrate the blood–brain barrier (BBB) to accumulate in the brain injury area, which increases the therapeutic concentration of DSF. (ii) DSF inhibits pyroptosis to block the release of proinflammatory factors by eliminating the formation of GSDMD-N plasma membrane pores. Therefore, C-β-LG/DSF effectively inhibits pyroptosis and exerts a neuroprotective effect on the TBI mice model. Our findings open an exciting angle to inhibit pyroptosis by blocking the release of inflammatory cytokines as a potential therapeutic strategy for TBI. Fig. 1. Schematic of C-β-LG/DSF for TBI treatment. [52]Fig. 1. [53]Open in a new tab (A) Schematic representation of the preparation process of C-β-LG/DSF. (B) After TBI, without C-β-LG/DSF treatment, damage-related molecular patterns (DAMPs) increase, which activate the inflammasome by binding to pattern recognition receptors (PRRs) to trigger pyroptosis, leading to the release of inflammatory cytokines (such as IL-18 and IL-1β), which triggers severe neuroinflammation and exacerbates secondary injury. (C) After TBI, C-β-LG/DSF treatment is given. DSF inhibits pyroptosis by blocking the formation of GSDMD-N pores on the cell membrane. Furthermore, it can also inhibit the release of inflammatory cytokines, thereby reducing neuroinflammation and alleviating secondary brain injury. RESULTS Preparation and characterization of C-β-LG/DSF The β-LG/DSF was prepared using the reverse solvent precipitation method. The CAQK peptide selectively binds to a proteoglycan complex that is up-regulated during brain injuries and coupled with drug molecules and nanoparticles to increase their accumulation at the injury site ([54]20, [55]21). Thus, the CAQK peptide was further modified into β-LG/DSF to form C-β-LG/DSF and increase the DSF concentration at the injury site. The diameters of C-β-LG/DSF and β-LG/DSF were 156.54 ± 4.52 nm and 144.91 ± 2.21 nm, respectively, which was consistent with the particle size shown by transmission electron microscopy (TEM) ([56]Fig. 2, A and B). The zeta potentials of C-β-LG/DSF and β-LG/DSF were −28.15 ± 7.93 mV and −19.18 ± 9.13 mV, respectively ([57]Fig. 2C). Scanning electron microscopy (SEM) showed that C-β-LG/DSF ([58]Fig. 2D) and β-LG/DSF ([59]Fig. 2E) were spherical with a uniform size distribution. The DSF-loading capacity and encapsulation efficiency of C-β-LG/DSF were 44.80 ± 1.14% and 90.25 ± 2.28%, respectively. The stability of C-β-LG/DSF in the blood circulation was assessed by conducting serum stability tests. The results showed that C-β-LG/DSF and β-LG/DSF remained stable in 10% fetal bovine serum–containing phosphate-buffered saline (PBS) for 7 days without obvious changes in hydrodynamic diameter ([60]Fig. 2F). β-LG has been widely used in the development of drug delivery systems, due to its excellent biocompatibility ([61]22, [62]23). Nanocarriers using β-LG as a shell have desirable attributes of safety, stability, and biodegradability ([63]24). To simulate a realistic traumatic environment, cerebrospinal fluid (CSF) samples were collected from patients with brain trauma for drug release experiments. The release of DSF from C-β-LG/DSF and β-LG/DSF was evaluated after incubation in TBI-CSF and PBS (pH 7.4), respectively. The results indicated that 29.16 ± 3.54% and 73.89 ± 2.03% of DSF were released from C-β-LG/DSF in PBS and TBI-CSF after 24 hours, respectively ([64]Fig. 2G). These results demonstrated that C-β-LG/DSF was responsive to the post-TBI environment and could release DSF rapidly. The hydrodynamic diameters C-β-LG/DSF and β-LG/DSF under TBI-CSF conditions were observed to increase ([65]Fig. 2H). These results demonstrated that C-β-LG/DSF was successfully prepared and was degraded in TBI-CSF to release DSF. Fig. 2. Preparation and characterization of C-β-LG/DSF. [66]Fig. 2. [67]Open in a new tab (A) Size distribution and TEM image of C-β-LG/DSF. (B) Size distribution and TEM image of β-LG/DSF. (C) The zeta potential of C-β-LG/DSF and β-LG/DSF. (D) SEM image of C-β-LG/DSF. Scale bar, 200 nm. (E) SEM image of β-LG/DSF. Scale bar, 200 nm. (F) Serum stability of C-β-LG/DSF and β-LG/DSF. (G) In vitro drug release profiles under TBI-CSF or pH 7.4 PBS conditions. (H) Size distribution of C-β-LG/DSF and β-LG/DSF at TBI-CSF at 12 hours. Data were expressed as means ± SD, (n = 3). Targeted delivery of C-β-LG/DSF to the TBI site To validate the selectively targeted ability of CAQK-modified β-LG/DSF, fluorescein isothiocyanate (FITC)–labeled β-LG was synthesized to prepare C-β-LG/DSF/FITC nanoparticles. Neuron-ICR (Institute of Cancer Research) cells, used in cell experiments, are primary neurons extracted from the frontal cortex tissue of the brain of ICR mouse embryos with a gestational age of 18 to 19 days ([68]Fig. 3A). First, the neuron-ICR cell uptake and the intracellular fate of C-β-LG/DSF/FITC were evaluated. Endocytosis inhibitors [chlorpromazine hydrochloride, 5-(n-ethyl-n-isopropyl) amiloride (EIPA), genistein, methyl-β-cyclodextrin (mβCD), and dynasore] were incubated with neurons ([69]25). Chlorpromazine hydrochloride was used to inhibit clathrin-mediated endocytosis ([70]26). mβCD and genistein inhibit caveolae-mediated endocytosis by disrupting specific signaling pathways and lipid composition, respectively ([71]27). EIPA is an inhibitor of micropinocytosis by inhibiting the function of the Na^+/H^+ exchanger ([72]28). Dynasore is an effective inhibitor that targets dynamin-dependent endocytic pathways ([73]29). As shown in [74]Fig. 3B, for C-β-LG/DSF, ~75% reduction in the uptake level was observed at mβCD. Genistein (~65%), EIPA (~27%), and dynasore (~14%) inhibited the neuron cell uptake process. Furthermore, the cell uptake level was decreased in the presence of mβCD and dynasore, which inhibited both the clathrin- and caveolar-mediated pathways. Collectively, the data indicated that C-β-LG/DSF entered the neuron-ICR cell mainly via caveolae-mediated endocytosis, clathrin-mediated endocytosis, and micropinocytosis. Fig. 3. The C-β-LG/DSF targeted delivery DSF into TBI site. [75]Fig. 3. [76]Open in a new tab (A) Scheme of primary neuron extraction. (B) Flow cytometry was used to assess the levels of C-β-LG/DSF uptake by neuron-ICR cells after treatment with various inhibitors. (C) The CLSM images were performed to test the ability of C-β-LG/DSF/FITC to penetrate BBB and target neurons. Red: Dil-labeled BBB layer cell membrane; blue: DAPI-labeled nuclei; green: C-β-LG/DSF/FITC (FITC signal). Scale bar, 25 μm. (D) Penetration of β-LG/DSF/FITC and C-β-LG/DSF/FITC through the cell monolayer. (E) Fluorescent image of β-LG/DSF/Cy5.5 and C-β-LG/DSF/Cy5.5 brain distribution in vivo was observed with small animal imaging system at post-injection 6, 12, 24, and 48 hours. (F) Statistical analysis of fluorescence content of C-β-LG/DSF in mice brains with imaging in vivo. (G) Statistical analysis of the FITC fluorescence intensity of C-β-LG/DSF/FITC in mice brains with a microplate reader. (H) Statistical analysis of the DSF concentration of C-β-LG/DSF/FITC in mice brains with UPLC-MS/MS. (I) Blood retention kinetics of free DSF and C-β-LG/DSF (DSF concentration of 10 mg/kg) in ICR mice. Data were expressed as means ± SD (n = 3). For (B), statistical analysis was calculated via one-way ANOVA test. **P < 0.01, **** P < 0.0001, and ns P > 0.05. For (D), (F), (G), and (H), statistical analysis was calculated via Student’s t test. *P < 0.05, **P < 0.01, and ***P < 0.001. The BBB is a complex structure that comprises not only the tight junctions between endothelial cells but also the presence of other cells that form the neurovascular unit and contribute to the tight regulation of the BBB. The effective penetration of drugs through the BBB is crucial for the treatment of TBI. An in vitro BBB model was established using a Transwell system to verify that C-β-LG/DSF could penetrate the BBB and precisely target traumatized neurons. Human brain microvascular endothelial cells were cultured at the bottom of the upper chamber, and neuron-ICR cells were cultured at the bottom of the lower chamber. When the transendothelial electrical resistance reached 150 ohms•cm^2, the in vitro BBB model was successfully prepared. A neuron-ICR cell TBI model was prepared in the lower chamber of the Transwell system through scratch-induced injury. C-β-LG/DSF/FITC and β-LG/DSF/FITC were separately added to the upper chamber of the Transwell system, and their cellular uptake in the upper and lower chambers was observed using confocal laser scanning microscopy (CLSM). The intensity of FITC fluorescence was detected using an enzyme-linked immunosorbent assay (ELISA) reader after 4 hours of incubation. The mean fluorescence intensity of FITC in C-β-LG/DSF/FITC–treated neuron-ICR cells was significantly higher than β-LG/DSF/FITC–treated cells, with a penetration rate of 69.26 ± 5.08% versus 17.06 ± 1.59%, suggesting that CAQK modification effectively improves the BBB penetration and targeting ability of β-LG/DSF ([77]Fig. 3, C and D). According to previous literature research, CAQK is a tetrapeptide that targets the chondroitin sulfate proteoglycan at the injury site following TBI and carry drugs to the brain injury site ([78]20, [79]21). However, the in vitro BBB model has limitations and cannot fully replicate the BBB’s complexity and functionality in real time, particularly in cases where the BBB has been disrupted after TBI. Therefore, to assess the targeting efficacy, cyanine5.5 (Cy5.5)–conjugated C-β-LG/DSF was injected into TBI-bearing mice via the tail vein. The C-β-LG/DSF/Cy5.5 group yielded a higher intensity of Cy5.5 fluorescence at the brain injury site compared with that yielded by the β-LG/DSF/Cy5.5 group at 6, 12, 24, and 48 hours after injection owing to the targeting effect of CAQK and the enhanced permeability and retention effect ([80]Fig. 3, E and F). Notably, at 6 hours after injection, notable accumulation of C-β-LG/DSF/Cy5.5 at the brain injury site was observed. To further assess the targeting efficacy, we prepared FITC-conjugated C-β-LG/DSF (C-β-LG/DSF/FITC) and measured the FITC intensity and DSF concentration in the brain following injection of C-β-LG/DSF/FITC and β-LG/DSF/FITC injection. Six hours after injection, the FITC intensity and DSF concentration were observed to be higher compared to 12, 24, and 48 hours after injection ([81]Fig. 3, G and H). Furthermore, the distribution of C-β-LG/DSF/Cy5.5 in the heart, liver, spleen, lung, and kidneys was examined at 6, 12, 24, and 48 hours after injection (figs. S1 to S3). Both C-β-LG/DSF/Cy5.5 and β-LG/DSF/Cy5.5 accumulated mostly in the liver, lungs, and kidneys at 6 and 12 hours after injection. After 24 hours of injection, both C-β-LG/DSF/Cy5.5 and β-LG/DSF/Cy5.5 demonstrated high accumulation in the liver and kidneys, and at 48 hours, only few C-β-LG/DSF/Cy5.5 and β-LG/DSF/Cy5.5 were found in the liver and kidneys (fig. S1, A and B). The quantification of FITC content further confirmed the predominant accumulation of C-β-LG/DSF/FITC and β-LG/DSF/FITC in the liver and kidneys after 24 hours of injection (fig. S2). In addition, we extracted DSF from the tissues after C-β-LG/DSF/FITC and β-LG/DSF/FITC injection and found that DSF was also concentrated in the liver and kidneys after 24 hours of injection, consistent with the Cy5.5 and FITC fluorescence data (fig. S3). Together, these results suggested that the C-β-LG/DSF nanoparticles are cleared by the liver and kidneys. The DSF plasma levels showed rapid clearance from the blood, with a half-life (t[1/2]) of ~4.83 hours ([82]Fig. 3I). The half-life (t[1/2]) of C-β-LG/DSF/Cy5.5 was ~8.28 hours. Therefore, C-β-LG/DSF prolonged the retention time of DSF in the blood, providing a greater chance of accumulation at the injury site. Inhibitory effect of C-β-LG/DSF on pyroptosis after TBI TBI-induced pyroptosis is a type of cell death triggered by proinflammatory signals and associated with inflammation, which occurs after TBI. We established an in vitro TBI model using neuron-ICR cells through scratch-induced injury to investigate the inhibitory effect of DSF on pyroptosis caused by TBI. First, the toxicity of DSF to glial cells and neuron-ICR cells was tested through the Cell Counting Kit-8 assays. As shown in figs. S4 and S5, there was no apparent toxicity within the 0.1 to 0.5 μM range of DSF in free or nanoparticle states. Drugs are trapped in endosomal vesicles and degraded in the lysosomal lumen, which leads to failure in the treatment of diseases. Therefore, the endosomal escape ability of C-β-LG/DSF/FITC was tested. As shown in fig. S6 (A to C), the separation between the green and red fluorescence spots was more pronounced, and the colocalization ratio of labeled endosomes/lysosome and C-β-LG/DSF/FITC was 40.67 ± 5.69% at 8 hours, indicating that C-β-LG/DSF/FITC could effectively escape from the endosomes for ensuring the therapeutic efficacy of DSF. Caspase cleavage of GSDMD, the executioner of pyroptosis, liberates an N-terminal p30 fragment (GSDMD-N), which oligomerizes and forms pores in the plasma membrane. These pores serve as a conduit for the release of IL-1β and IL-18 and, ultimately, the demise of the cell ([83]30). However, DSF can prevent pyroptosis by inhibiting the formation of GSDMD N-terminal pores on the cell membrane surface, which stops the release of inflammatory cytokines ([84]Fig. 4A). As shown in [85]Fig. 4B, the results demonstrated that the expression of GSDMD, GSDMD-N, and caspase-1 p10 (key indicators of cell pyroptosis) in neuron-ICR cells significantly increased after scratch-induced injury compared to the sham group, suggesting that neuron-ICR cells undergo pyroptosis. DSF and C-β-LG/DSF substantially inhibited the expression of GSDMD, GSDMD-N, caspase-1, caspase-1p20, and caspase-1 p10 (fig. S7, A to E, and [86]Fig. 4B). The extracellular release of IL-1β, IL-18, and lactate dehydrogenase (LDH) increased in neuron-ICR cells after scratch-induced injury while decreasing after DSF and C-β-LG/DSF treatment, suggesting that DSF effectively inhibited the occurrence of pyroptosis after injury ([87]Fig. 4, C to E). The immunohistochemical test results of GSDMD in TBI mice treated with C-β-LG/DSF further confirmed that GSDMD expression was obviously decreased compared with that in other treatment groups ([88]Fig. 4, F and G). When pyroptosis occurs, pores form in the cell membrane and cell lysis occurs, releasing a large amount of cytoplasmic content, triggering an inflammatory response, and further exacerbating secondary damage. Therefore, to verify whether C-β-LG/DSF can inhibit pyroptosis-mediated inflammation, ELISA was performed to detect the release of IL-1β, IL-18, and LDH from brain tissue on the third day after TBI ([89]Fig. 4, H to J). The results revealed that compared with the sham group, the release of IL-1β, IL-18, and LDH in the brain of the TBI group significantly increased, further indicating that cell pyroptosis occurs in brain tissue following TBI and released a large number of inflammatory factors, leading to secondary damage to brain tissue. Compared with the TBI and DSF groups, the release of these inflammatory factors was remarkably reduced in the C-β-LG/DSF groups. Thus, collectively, C-β-LG/DSF can substantially inhibit pyroptosis occurrence following TBI. To further confirm pyroptosis production in mice following TBI, brain tissue from the trauma site was collected for Western blot analysis. Compared with the sham group, the expression of GSDMD, GSDMD-N, caspase-1 p20, and caspase-1 p10 in the brain of TBI mice significantly increased, confirming that cell pyroptosis occurs in brain tissue following TBI ([90]Fig. 4K). Moreover, compared with the TBI and DSF groups, the expression levels of the GSDMD, GSDMD-N, caspase-1 p20, and caspase-1 p10 proteins were significantly reduced in the C-β-LG/DSF groups (fig. S8, A to E), indicating that DSF inhibited pyroptosis following TBI. Fig. 4. Inhibitory effect of C-β-LG/DSF on pyroptosis after TBI. [91]Fig. 4. [92]Open in a new tab (A) The schematic diagram illustrates the inhibition of pyroptosis by DSF. After the uptake of C-β-LG/DSF by cells, DSF inhibits the aggregation of GSDMD N-terminal pores on the cell membrane surface. Furthermore, experimental results demonstrated that the expression of GSDMD, GSDMD-N, caspase-1, and other related proteins was reduced after treatment with C-β-LG/DSF. (B) Immunoblotting of GSDMD, GSDMD-N, and caspase-1, caspase-1 p10, caspase-1 p20 in neuron-ICR cells with sham, TBI, DSF, and C-β-LG/DSF for 24 hours. (C to E) Levels of (C) IL-1β, (D) IL-18, and (E) LDH in injured neuron-ICR as detected by ELISA 12 hours after TBI. (F) Immunohistochemical staining was used to observe the expression of GSDMD in injured tissues of TBI model mice 3 days after different drugs treatment. Scale bars, 50 μm. (G) Quantitative analysis of GSDMD-positive cells. (H to J) Levels of (H) IL-1β, (I) IL-18, and (J) LDH in injured tissues as detected by ELISA 3 days after TBI. (K) Immunoblotting of GSDMD, GSDMD-N, caspase-1, caspase-1 p20, and caspase-1 p10 in injured tissue with sham, TBI, DSF, β-LG/DSF, and C-β-LG/DSF for 24 hours. Data were expressed as means ± SD (n = 5). For (C) to (E) and (G) to (J), statistical analysis was calculated via one-way ANOVA test. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. Effects of C-β-LG/DSF on BBB permeability and cerebral edema after TBI After TBI, the BBB is damaged, resulting in increased permeability and severe brain edema ([93]31–[94]33). Previous studies have demonstrated that the posttraumatic inflammation cascade promotes TBI-induced BBB disruption ([95]10). Therefore, we performed several tests to determine whether C-β-LG/DSF can effectively inhibit BBB disruption. The diagram and schedule are shown in [96]Fig. 5A. C-β-LG/DSF (6.64 ± 1.13 μg/g) significantly reduced the Evans blue content per gram of brain tissue compared with other treatments (DSF: 23.04 ± 1.51 μg/g; β-LG/DSF: 12.61 ± 0.98 μg/g), suggesting that inhibiting pyroptosis in brain tissue following TBI can effectively alleviate brain edema ([97]Fig. 5, B and C). The brain tissue water content assay further confirmed that C-β-LG/DSF-treated TBI mice exhibited the lowest water content ([98]Fig. 5D). To further investigate the ability of C-β-LG/DSF to improve secondary injuries following TBI in mice, brain tissue edema was examined using magnetic resonance imaging (MRI) with T2-weighted images on the third day after TBI ([99]Fig. 5E, a). Brain tissue edema was evident in the right hemisphere surrounding the TBI and DSF groups, with volumes of 15.56 ± 1.22 mm^3 and 14.92 ± 0.96 mm^3, respectively. However, the brain tissue edema area was noticeably reduced in the β-LG/DSF (10.92 ± 1.33 mm^3) and C-β-LG/DSF groups (7.32 ± 1.06 mm^3). Moreover, TBI mice treated with C-β-LG/DSF exhibited significantly less brain edema than in mice treated with other treatments, with a decreased volume of 7.32 ± 1.06 mm^3 ([100]Fig. 5F). The apparent diffusion coefficient (ADC) value obtained from diffusion-weighted imaging (DWI) was further used to evaluate the dispersion of water molecules on the surface of the cerebral cortex to determine the type and trend of brain tissue edema ([101]Fig. 5E, b). Lower ADC values are generally associated with cytotoxic brain edema, which reflects the gradual occurrence and intensification of cytotoxic edema following the release of neurotoxic substances. The ADC value of DWI in the TBI group was 0.29 ± 0.05 × 10^−3 mm^2/s, which was considerably lower than that in the sham group (0.90 ± 0.06 × 10^−3 mm^2/s), indicating that cytotoxic brain edema was dominant on day 3 following TBI in mice. The ADC values in the β-LG/DSF and C-β-LG/DSF groups were significantly higher than those in the TBI group, with ADC values of 0.64 ± 0.05 × 10^−3 and 0.77 ± 0.05 × 10^−3 mm^2/s, respectively ([102]Fig. 5G), indicating a specific therapeutic effect of β-LG/DSF and C-β-LG/DSF in alleviating cytotoxic brain edema in mice, consistent with the trend of brain edema volume shown in the T2-weighted images. Fig. 5. Effects of C-β-LG/DSF on BBB permeability and cerebral edema after TBI. [103]Fig. 5. [104]Open in a new tab (A) Schematic diagram of TBI model experiments. (B) Images of Evans blue leakage from brain capillary vessels in the right brain of mice for each group 24 hours after TBI. (C) Quantitative analysis of Evans blue leakage. (D) Water content in brain tissue of each group. (E) Representative (a) T2-weighted and (b) DWI MRI images in each group 3 days after TBI. (F) Quantification of the hyperintensity volume around the injured tissue 3 days after TBI. (G) Quantification of the ADC signal from the injured tissue 3 days after TBI. For (B) to (D), data were expressed as means ± SD (n = 5). For (E) to (G), data were expressed as means ± SD (n = 6). For (C), (D), (F), and (G), statistical analysis was calculated via one-way ANOVA test. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. Anti-inflammatory effect of C-β-LG/DSF Pyroptosis can activate strong inflammation and immune responses. To further validate the neuroprotective effect of C-β-LG/DSF, we assessed neuronal necrosis, apoptosis, and brain tissue damage in mice on day 7 after TBI. Hematoxylin and eosin (H&E) staining revealed significantly reduced brain tissue damage in the C-β-LG/DSF group compared with the other treatments ([105]Fig. 6, A and B). Furthermore, Nissl staining revealed an obvious decrease in neuronal survival rate in mice following TBI compared with the sham group, with positive staining cell numbers of 121.8 ± 11.32 and 259.8 ± 34.57, respectively. Compared with the other treatments, C-β-LG/DSF treatment significantly increased neuronal survival with a positive staining cell number of 215.8 ± 22.9 ([106]Fig. 6, C to E). To assess the level of neuroinflammation in brain tissue after TBI, we assessed the expression of glial fibrillary acidic protein (GFAP) and ionized calcium-binding adaptor-1 (Iba-1), which are indices of astrocyte and microglia activation, respectively. After brain tissue damage, microglia and astrocytes are rapidly activated with swollen and branched cells. Microglia are resident immune cells in the central nervous system with sensing, caretaking, and defensive functions. In neurodegenerative diseases, microglia dysfunction exacerbates neuronal damage, while numerous inflammatory factors are released in activated astrocytes. Therefore, we examined the expression of cells through immunohistochemistry to determine the level of neuroinflammation in brain tissue after TBI ([107]Fig. 6, F and I). Furthermore, the hippocampus is closely related to learning and spatial memory, while its degeneration is closely related to TBI prognosis. We found that GFAP- and Iba-1–positive cells were substantially reduced in the cortex surrounding the brain injury area ([108]Fig. 6, G and J) and the hippocampus ([109]Fig. 6, H and K) in the C-β-LG/DSF groups on day 7 after TBI. These results suggest that C-β-LG/DSF reduces inflammation by inhibiting pyroptosis. Fig. 6. Anti-inflammatory effect of C-β-LG/DSF. [110]Fig. 6. [111]Open in a new tab (A) Representative H&E staining of brain tissue 7 days after TBI. (B) Quantitative analysis of injury area ratio of H&E staining. (C) Representative images of Nissl-positive cells 7 days after TBI. (D) Enlarged image of Nissl staining, Scale bar, 50 μm. (E) Quantitative analysis of Nissl-positive cells. (F) Immunohistochemical staining was used to observe the expression of GFAP in the damaged cortex area of each group of TBI model mice 7 days after different drug treatment. Scale bars, 50 μm. (G) Quantitative analysis of GFAP-positive cells in the damaged cortex area of each group of TBI model mice 7 days after different drug treatment. (H) Quantitative analysis of GFAP-positive cells in the hippocampus ipsilateral of each group of TBI model mice 7 days after different drug treatment. (I) Immunohistochemical staining was used to observe the expression of Iba-1 in the damaged cortex area of each group of TBI model mice 7 days after different drug treatment. Scale bar, 50 μm. (J) Quantitative analysis of Iba-1–positive cells in the damaged cortex area of each group of TBI model mice 7 days after different drug treatment. (K) Quantitative analysis of Iba-1–positive cells in the hippocampus ipsilateral of each group of TBI model mice 7 days after different drug treatment. Data were expressed as means ± SD (n = 5). For (B), (E), (G), (H), (J), and (K), statistical analysis was calculated via one-way ANOVA test. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. RNA-seq analysis of inflammasome-mediated inflammation changes following C-β-LG/DSF treatment TBI is a critical public health and socioeconomic concern worldwide. Inflammation-induced secondary injury is a vital pathogenic parameter of TBI. The molecular signaling cascades of pyroptosis, a specific type of cellular necrosis, are key drivers of TBI-induced inflammation. RNA-sequencing (RNA-seq) analysis revealed significant up-regulation of GSDMD, CASP1, and other inflammasome-related genes in the TBI group compared with the sham group ([112]Fig. 7A). However, following C-β-LG/DSF treatment, compared with the TBI group, the expression of pyroptosis-related genes, such as GSDMD, CASP1, and NOD-like receptor thermal protein domain associated protein 3 (NLRP3), was significantly down-regulated ([113]Fig. 7B). We also mapped a gene heatmap of 17 genes closely related to pyroptosis in brain tissue after TBI, and the results were consistent with those of the volcanic map ([114]Fig. 7C). The expression levels of CASP1, nucleotide-binding oligomerization domain (NOD), NLRP3, and GSDMD increased after TBI but decreased after C-β-LG/DSF treatment. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis revealed notable activation of apoptosis, necroptosis, NOD-like receptor, and Toll-like receptor signaling pathways after TBI. Among them, the NOD-like and Toll-like receptor signaling pathways were closely related to pyroptosis and involved in its occurrence. The significant activation of the inflammatory pathway through leukocyte transendothelial migration suggests that concurrent with pyroptosis, inflammatory cells infiltrate the damaged brain tissue, leading to further secondary damage ([115]Fig. 7D) (P < 0.05). In addition, Gene Ontology (GO) analysis revealed that pore formation on the cell membrane and the expression of IL-1, IL-1β, and NLRP3 inflammasome complex, which are closely related to pyroptosis, were obviously up-regulated alongside inflammation-related immune receptors after TBI ([116]Fig. 7E) (P < 0.05). However, compared with the TBI group, pore formation on the cell membrane, pyroptosis-related cell death, leukocyte migration and proliferation, and inflammasome complex were significantly down-regulated following C-β-LG/DSF treatment ([117]Fig. 7F) (P < 0.05). Collectively, these results suggest that C-β-LG/DSF can regulate inflammatory responses. Fig. 7. RNA-seq analysis reveals that C-β-LG/DSF inhibits inflammasome-mediated inflammation following TBI. [118]Fig. 7. [119]Open in a new tab (A) Analysis of differences in expression of pyroptosis-related genes between the TBI group and the sham group (n = 3). (B) Analysis of differences in expression of pyroptosis-related genes between the TBI group and the C-β-LG/DSF group (n = 3). (C) Sham, TBI, and C-β-LG/DSF groups’ gene heatmaps (n = 3). (D) Enrichment analysis of KEGG pathway in the TBI group versus the sham group (n = 3). (E) GO analysis in the TBI group versus the sham group (n = 3). (F) GO analysis in the C-β-LG/DSF group versus the TBI group (n = 3). Effect of C-β-LG/DSF on neural function recovery after TBI in mice We conducted a Morris water maze (MWM) test on mice to evaluate their learning and spatial memory abilities after TBI. The test was conducted from day 21 to day 29 after injury, with the recovery of learning ability assessed during the first 7 days and the recovery of spatial memory tested on day 8 and day 9. A schematic timeline of experiments is displayed in [120]Fig. 8A. Mice treated with C-β-LG/DSF exhibited significantly better recovery of learning and spatial memory abilities than those treated with the other treatments ([121]Fig. 8B). During the first 4 days, the average swimming distance of the sham group mice was shorter than that of the TBI and other experimental group mice ([122]Fig. 8B). Compared with the C-β-LG/DSF group, the TBI and DSF groups demonstrated significantly longer latency period and swimming distances before reaching the platform ([123]Fig. 8C) (P < 0.05). On day 28 after TBI, the 8th day of the water maze test, the memory of the mice was evaluated by removing the platform and measuring the time spent swimming in the quadrant where the platform used to be located as well as determining the number of crossings over the platform area. Compared with the TBI group, mice treated with C-β-LG/DSF spent significantly more time and traveled longer distances within the target quadrant ([124]Fig. 8, D and E), exhibited increased frequency of platform crossings ([125]Fig. 8F), and stayed for a longer duration in the target quadrant ([126]Fig. 8G) (P < 0.05), indicating more purposeful searching for the platform. These results suggest that C-β-LG/DSF treatment promotes spatial memory and learning ability recovery in mice following TBI. We evaluated the recovery of motor function in mice treated with C-β-LG/DSF using a wire suspension test ([127]Fig. 8H). Compared with the sham group, mice in all the other groups exhibited significantly longer hanging times. In addition, compared with the TBI group, mice treated with C-β-LG/DSF demonstrated obviously longer hanging times on the wire (P < 0.05), indicating improved motor function recovery. To assess the recovery of neurological function in mice following TBI, we used the modified neurological severity score (mNSS) to evaluate mice on days 3, 7, and 14 after the injuries using a blinded scoring method. Mice treated with C-β-LG/DSF exhibited significantly improved neurological function compared with those in the TBI group, whereas mice treated with DSF did not show significant improvement ([128]Fig. 8I). These results further demonstrate that C-β-LG/DSF promotes neurological function recovery in mice following TBI by inhibiting neuronal pyroptosis, preventing further neuroinflammation, and notably alleviating secondary brain injury. Fig. 8. Effect of C-β-LG/DSF on neural function recovery after TBI in mice. [129]Fig. 8. [130]Open in a new tab (A) mNSS motor function score, MWM, and suspension test analysis schedule. (B) Computer printouts of the swimming trajectories during the learning phase and memory phase. (C) Swimming distance to the platform and (D) searching time for the platform at the last trial for each of the 7 days of training. During the 2-day memory experiment, (E) the duration and (F) distance of swimming in the platform area of each group of mice. (G) Comparison of crossing platform times among different groups in the memory stage. a.u., arbitrary units. (H) Comparison of the duration of stay in the target quadrant among different groups in the memory stage. (I) Motor function was evaluated by the hanging wire grip test 14 days after TBI. (J) The mNSS scores were calculated 3, 7, and 14 days after TBI. Data were expressed as means ± SD (n = 5). For (E) to (I), statistical analysis was calculated via one-way ANOVA test. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. In vivo compatibility of C-β-LG/DSF To verify whether C-β-LG/DSF had toxic side effects in mice, we administered consecutive doses to mice for 4 days and monitored their body weight every 2 days after TBI. The weight curve of the mice did not exhibit any substantial decline in body weight (fig. S9). In addition, the mice were euthanized in each group 21 days after TBI and administration and the major organs were taken out for tissue paraffin sectioning and H&E staining. No obvious tissue damage or morphological changes were observed in the heart, liver, spleen, lungs, and kidneys tissue sections of the β-LG/DSF and C-β-LG/DSF groups compared with the sham group, indicating good biocompatibility and high safety in mice (fig. S10). The routine hematological examination and biochemical tests were also performed by taking blood samples via the tail vein on the 7th day after administration. The results showed no significant difference in the hematology markers and serum alkaline phosphatase, alanine aminotransferase, aspartate transaminase, creatinine, and blood urea nitrogen levels between the C-β-LG/DSF groups and the sham group (P > 0.05). Therefore, C-β-LG/DSF is not harmful to the liver and kidneys of mice and is biologically safe (fig. S11). DISCUSSION With increasing use of transportation and rapid development of the construction industry owing to socioeconomic growth, TBI has become a frequent occurrence, leading to a significant economic burden for patients’ families and society ([131]2, [132]4–[133]6). TBI is one of the leading causes of disability and death among young adults worldwide, with long-term effects on the survivors’ quality of life ([134]34–[135]36). The pathological mechanisms underlying TBI are complex and often accompanied by secondary brain injury, which can result in more severe consequences such as inflammation and neural damage. However, currently, there is no effective treatment for secondary brain injury in clinical practice ([136]37–[137]39). Pyroptosis is a specific type of cellular necrosis that is a key driver of inflammation in TBI ([138]40–[139]42) and can be activated by damage-associated molecular patterns and pathogen-associated molecular patterns recognized by Toll-like receptors and NOD-like receptors on the membranes of neurons and glial cells, which activates inflammasomes, cleaves GSDMD proteins, forms pores on the cell membrane surface, promotes the release of cell contents into the intracellular environment, and causes severe inflammation and cytotoxic edema, thereby exacerbating the inflammatory response and brain tissue edema of neurons and other brain tissue cells ([140]43–[141]46). Our study showed that the expression of GSDMD, GSDMD-N, caspase-1 p20, and caspase-1 p10 increased and the levels of IL-1β, IL-18, and LDH improved in the brain of TBI mice, suggesting that cell pyroptosis occurred in brain tissue following TBI ([142]Fig. 4). Furthermore, RNA-seq analysis revealed notable up-regulation of GSDMD, caspase-1, IL-1β, IL-18, and other inflammasome-related genes in TBI mice ([143]Fig. 7A). The outcomes of our study suggest that pyroptosis increases in the damaged brain following TBI, consistent with the results of numerous studies ([144]30, [145]47, [146]48). Therefore, inhibiting pyroptosis in neurons and glial cells following TBI is a potential target for treating secondary brain injury. However, effective clinical therapeutic strategies for pyroptosis remain scarce. In addition, there is debate over the optimal time window for inflammation treatment following TBI. The therapeutic time window is complex, as TBI is both acute and chronic ([147]49). For acute brain injuries, the optimal treatment time window is typically within a few hours after the injury. For subacute and chronic brain injuries, the optimal treatment time window may be longer, depending on the type and severity of the injury in specific cases. It is generally recommended to initiate treatment within 24 hours of the trauma to minimize subsequent damage and promote recovery. Early intervention can help control the inflammatory response caused by the injury, thereby reducing secondary damage ([148]50). However, initiating treatment too early may interfere with the body’s own stress response and repair processes ([149]51). DSF is an FDA-approved drug for alcohol addiction and can inhibit pyroptosis by blocking GSDMD pore formation. However, DSF is rapidly metabolized, hydrophobic, and less stable, which limits its systemic application. Therefore, there is an imperative demand to develop an effective delivery system of DSF for treating TBI. Here, we fabricated CAQK peptide–modified β-LG to deliver DSF (C-β-LG/DSF) to enhance DSF accumulation in the TBI area and inhibit pyroptosis from blocking the release of proinflammatory cytokines. β-LG is an edible whey protein and CAQK is a tetrapeptide (cysteine-alanine-glutamine-lysine) with targeting capability for traumatic brain tissue, which are all safe materials. C-β-LG/DSF could effectively penetrate the BBB and target injured brain tissue in vitro and in vivo ([150]Fig. 3). C-β-LG/DSF with a half-life (t[1/2]) of 8.28 hours prolonged the retention time of DSF in the blood and increased its chances of accumulating at the injury site. Furthermore, we observed obvious accumulation of C-β-LG/DSF at the TBI site 6 hours after injection, and this high concentration was maintained even at 24 hours after injection. C-β-LG/DSF was administered three times every other day. Upon entering the TBI site, 73.89 ± 2.03% of DSF was released within 24 hours, whereas only 29.16 ± 3.54% of DSF was released from C-β-LG/DSF in PBS. These results suggested that C-β-LG/DSF is responsive to the post-TBI environment and meets a specific time window for anti-inflammatory treatment of TBI. Here, we investigated the efficacy of C-β-LG/DSF to inhibit pyroptosis by blocking the release of proinflammatory cytokine following TBI. Results from an in vitro traumatic model experiment on neuron-ICR primary neurons from fetal mice revealed that C-β-LG/DSF inhibited cell pyroptosis as well as the extracellular release of IL-1β, IL-18, and LDH. In vivo TBI mouse model studies further supported that C-β-LG/DSF effectively blocks the extracellular release of IL-1β, IL-18, and LDH, and down-regulated the expression of GSDMD, GSDMD-N, caspase-1 p20, and caspase-1 p10 in the brain of TBI mice. Neuroinflammation and BBB disruption following TBI are interrelated ([151]52–[152]54). The posttraumatic dysfunction of the BBB causes BBB permeability and brain edema. C-β-LG/DSF effectively protected BBB and reduced brain edema, as demonstrated by the Evans blue extravasation experiment, brain water content measure, and MRI assay ([153]Fig. 5). Diffusion-weighted images and ADC coefficients showed that C-β-LG/DSF significantly reduced cytotoxic edema in brain tissue following TBI. These findings indicate that C-β-LG/DSF can effectively reduce cerebral edema, possibly through inhibiting cell pyroptosis after TBI. In addition, Nissl staining revealed that C-β-LG/DSF significantly reduced neuronal damage following TBI. Chronic nonresolving inflammation activated by microglia is a typical feature of secondary injury following TBI. Early-activated microglia can clear cell debris, which is beneficial for TBI recovery ([154]55, [155]56). However, microglia still maintain an activated phenotype during the chronic stage of TBI, which is usually considered harmful. To verify the regulatory effect of C-β-LG/DSF on neuroinflammation, we detected the expression of GFAP and Iba-1, activation markers for astrocytes and microglia, respectively, on day 7 following injury to elucidate the strength of the neuroinflammatory responses among the different groups. The immunohistochemical quantification analysis of GFAP and Iba-1 in the C-β-LG/DSF group revealed a significant reduction in positive cells, indicating that C-β-LG/DSF can effectively inhibit neuroinflammation. In a behavioral experiment, mNSS was used to evaluate the motor, sensory, reflex, and balance abilities of mice following TBI. C-β-LG/DSF exhibited neurological function closer to that of normal mice. The MWM test was used to assess learning and memory in TBI mice. The results revealed that C-β-LG/DSF exhibited significantly improved learning ability and could find the platform faster. Memory testing 2 days after removing the platform showed that C-β-LG/DSF exhibited significantly improved spatial memory following TBI. The above results suggest that C-β-LG/DSF significantly enhances DSF accumulation in brain injury areas and inhibits cell pyroptosis and inflammatory cytokine release in neurons and mouse brain tissue following TBI to alleviate neuroinflammation, resulting in an effective therapeutic effect on neurological function. Pyroptosis is a relatively rare therapeutic target for TBI due to a lack of clear understanding of the mechanisms of pyroptosis in TBI. In addition, the duration and intensity of inflammatory responses may vary among individuals, making the optimal treatment time window different for each person. Personalized treatment plans may need to take into account factors such as the type and severity of the trauma, individual patient characteristics, and other relevant factors. Further research is needed to determine the optimal timing for treatment and develop personalized treatment strategies. We developed a CAQK peptide–modified β-LG nanoparticle for the delivery of DSF to inhibit pyroptosis and block the release of inflammatory cytokines, thereby alleviating neuroinflammation and preventing secondary damage spread. C-β-LG/DSF effectively accumulates in the target TBI wound cavity and enhances DSF concentration. C-β-LG/DSF significantly inhibits pyroptosis and relieves neuroinflammation in vivo. Furthermore, C-β-LG/DSF effectively protected the BBB and alleviated brain edema. Last, TBI mice exhibited behavioral function recovery following C-β-LG/DSF treatment. Therefore, C-β-LG/DSF is a potential therapeutic candidate as it inhibits the occurrence of pyroptosis following TBI, thereby reducing secondary injury. MATERIALS AND METHODS Materials The following were the materials used in this study: DSF (Sigma-Aldrich, PHR1690, USA), β-LG (Sigma-Aldrich, L3908, USA), CAQK (GL Biochem, Shanghai, China), succinimidyl-4-(N-maleimidomethyl cyclohexane)-1-carboxylate) (SMCC) (Aladdin, M123456, Shanghai, China), FITC (Sigma-Aldrich, 46950, USA), Cy5.5 N-hydroxysuccinimide (NHS) ester (Cy5.5 NHS) (Ruixi, Xi’an, China), Dil (Beyotime, C1036, Shanghai, China), Evans blue (TCI, E0197, Japan), anti–caspase-1 p20 (Santa Cruz Biotechnology, sc-398715, USA), anti-cleaved GSDMD (Cell Signaling Technology, #10137S, USA), anti–caspase-1 p10 (GeneTex, GTX134551, USA), anti–glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (Proteintech, 60004-1-Ig, Wuhan, China), anti-GFAP (Cell Signaling Technology, #45946, USA), anti–Iba-1 (Sigma-Aldrich, SAB2702364, USA), Dulbecco’s modified Eagle’s medium (DMEM; WISENT, 319-015-CL, Canada), Neurobasal (Gibco, 21103-049, USA), extracellular matrix (ScienCell, 1001, USA), 4′,6-diamidino-2-phenylindole (DAPI; Beyotime, C1005, China), Cell Counting Kit-8 (Meilunbio MA0218, Dalian, China), poly-l-lysine hydrobromide (Sigma-Aldrich, P6282, USA), LysoTraker (Meilunbio MB604-1, Dalian, China), B-27 (Gibco, 17504-044, USA), epidermal growth factor (Peprotech, AF-100-15, USA), fibroblast growth factor (Peprotech, AF-100-188, USA), IL-1β ELISA Kit (Jonln, JL18442, Shanghai, China), IL-18 ELISA Kit (Jonln, JL20253, Shanghai, China), LDH ELISA Kit (Solarbio, BC0685, Beijing, China), and sesame oil (Macklin, S905724, Shanghai, China). Preparation and characterization of C-β-LG/DSF C-β-LG/DSFs were prepared via an antisolvent precipitation method. A β-LG solution (1 mg/ml) was prepared by adding 10 mg of β-LG to 10 ml of distilled water, stirring the mixture at room temperature for 1 hour, adjusting pH to 7.0 with 0.1 M HCl or NaOH, and heating the protein solution to 90°C or above for 30 min. A DSF solution was prepared by dissolving 15 mg of DSF in 200 μl of acetone, which was then slowly added to the β-LG solution (under ice bath conditions) and subjected to ultrasonic shock for 10 min. The resulting β-LG/DSF solution was vaporized under pressure at 25°C for 30 min to remove residual acetone. One milligram of CAQK and 1 mg of SMCC were dissolved in 1 ml of water to obtain CAQK solution (1 mg/ml) and SMCC solution (1 mg/ml), respectively. Add 1 ml of water to a 2-ml centrifuge tube and add 100 μl of CAQK solution, then stir vigorously. At the same time, CAQK-SMCC solution was prepared by adding 100 μl of SMCC solution into the above solution and stirring for 3 hours under ice bath conditions at 1500 rpm. Absorb 100 μl of CAQK-SMCC solution and slowly add it to β-LG/DSF solution and stir it violently for 2 hours under ice bath conditions to prepare C-β-LG/DSF. The particle size, zeta potential, and its stability in serum (10%) were measured by a laser particle size analyzer (Zetasizer NanoZS90, Malvern, UK). The morphology of the delivery system was observed by TEM Tecnai G2 Spirit Twin (FEI, USA). The samples dispersed in PBS were dropped onto 200-mesh C-coated Cu grid before recording the TEM images. The drug loading (DL) and encapsulation efficiency (EE) of DSF were analyzed by ultraperformance liquid chromatography (UPLC)–tandem mass spectroscopy (MS/MS). The column was an advanced Waters Acquity UPLC BEH C18 100 × 2.1 mm, and the mobile phase (1 ml/min) consisted of water with 0.1% methanoic acid:acetonitrile with 0.1% methanoic acid (1:1). Mass spectrometer parameters were as follows: capillary, 0.5 kV; cone voltage, 18 V; collision energy, 10 V; ion source temperature, 150°C; desolvation temperature, 500°C; cone gas flow, 800 liters/hour; cone gas flow, 20 liters/hour; mass-charge ratio: 297 to 116.0633. The prepared C-β-LG/DSF was freeze-dried to obtain dry-powder C-β-LG/DSF nanoparticles. The assay was repeated three times by dissolving 1 mg of lyophilized nanoprodrug powder in 1 ml of water containing 0.1% methanoic acid and mixing an equal volume of acetonitrile with 0.1% methanoic acid, 1 μl per injection. The experiment was repeated three times. A standard curve of DSF (0 to 100 ng ml^−1) in aqueous solution was tested to determine the concentration. Y = 437.38 × X - 86.5239, R^2 = 0.9916. [MATH: EE (%)=(weight of drug loaded)/(weight of drug in feed)×100% :MATH] [MATH: DL (%)=(weight of drug loaded)/(weight of nanoparticles)×100% :MATH] [MATH: EE (%) of DSF=90.25±2.28% :MATH] [MATH: DL (%) of DSF=44.80±1.14% :MATH] Drug release C-β-LG/DSF was prepared as mentioned in the “Preparation and characterization of C-β-LG/DSF” section. C-β-LG/DSF nanoparticles (1 ml) were transferred into dialysis bags with 3500 molecular weights, and then immersed in 100 ml of PBS (pH 7.4) and CSF of TBI patients, respectively. Drug release was detected at 37°C. Samples of 200 μl were taken at different time periods and stored at 4°C. At the same time, the corresponding volume of PBS or CSF of TBI patients was added for volume supplement. Released DSF was detected by UPLC-MS/MS. The detection methods and chromatographic and MS conditions were the same as those in the “Preparation and characterization of C-β-LG/DSF” section. This study was performed after informing patients and their families with fully informed consent, and it was approved by the Ethics Committee of the Affiliated Hospital of Xuzhou Medical University (ethics number: XYFY2020-KL208-01). Cell culture The human brain microvascular endothelial cells (HBMECs) were obtained from the Shanghai Cell Bank, Type Culture Collection Committee, Chinese Academy of Sciences, and were routinely cultured in ECM endothelial cell medium containing 5% fetal bovine serum and 0.1% Endothelial Cell Growth Supplement. Neuron-ICR was extracted from the frontal cortex of embryonic mice aged 18 to 19 days during pregnancy. Pregnant ICR mice were anesthetized with isoflurane gas and euthanized by decapitation, and their abdomen was disinfected with alcohol. The abdominal cavity was opened with scissors and forceps, and embryos were removed from the uteri. The brains of the embryos were dissected and transferred to prechilled Hanks’ balanced salt solution (HBSS) without phenol red and kept on ice. The frontal cortex tissues were isolated from the brain tissues using ophthalmic scissors and forceps under ice bath conditions. The isolated frontal cortex tissues were transferred to HBSS with phenol red, minced with ophthalmic scissors, and transferred to a sterile 15-ml Eppendorf tube for centrifugation at 1500 rpm for 5 min. The supernatant was removed, and 2 ml of 0.25% trypsin-EDTA was added. The sample was pipetted up and down, placed in a cell culture incubator for 15 min, and then pipetted in 5 ml of DMEM+++ (10% fetal bovine serum, 1% glutamine, and 1% penicillin-streptomycin solution) to remove the trypsin. After mixing, the cells were washed three times, and the supernatant was discarded after centrifugation. Subsequently, 5 ml of DMEM+++ was added and vigorously pipetted 20 to 30 times. The processed cell suspension was then passed through a cell strainer, counted, and seeded in 10-cm dishes with 2 million cells in 7 ml of DMEM+++. For six-well plates, 300,000 cells were seeded in 2 ml of DMEM+++, while for 96-well plates, 6000 cells were seeded in 200 μl of DMEM+++. The plates were kept in a cell culture incubator for 6 hours and observed under a microscope. If the cells were adherent, the medium was replaced with Neurobasal+++ (2% B27, 1% glutamine, and 1% penicillin-streptomycin solution). The medium was replaced every 7 days, and the cells were discarded after 21 days ([156]57). Cell Counting Kit-8 Neuron-ICR cells were plated with poly-l-lysine (Sigma-Aldrich) in 96-well plates. HBMEC cells were plated in 96-well plates. After 24 hours, the cells were treated with different concentrations of DSF, β-LG/DSF, and C-β-LG/DSF, and the culture was continued for 24 hours. Last, 10 μl of Cell Counting Kit-8 solution was added to each well and incubated for 30 min. The absorbance value was detected at 450-nm wavelength. TBI cell model Neuron-ICR cells cultured in 10-cm dishes were divided into four experimental groups: sham group (no scratch injury model), TBI group (scratch injury without drug treatment), DSF group (scratch injury with free DSF drug treatment), and C-β-LG/DSF group (scratch injury with β-LG–encapsulated DSF targeted peptide–modified nanomedicine treatment). The cells were placed in a laminar flow hood, and sterile pipette tips were used to create vertical and horizontal scratches (except for the sham group) on the bottom of the dish, with five scratches in each direction and a distance of 0.5 cm between each scratch. After 1 hour, the medium was replaced with Neurobasal+++ medium containing the respective group drugs (except for the TBI group), with a DSF concentration of 0.5 μM. Cellular uptake of C-β-LG/DSF Neuron-ICR cells were seeded in a six-well plate under cell culture conditions. After 48 hours of cell growth, the cells were treated at 4° or 37°C for 1 hour with 25 μM chlorpromazine hydrochloride, 100 μM EIPA, 5 mM mβCD, 740 μM genistein, and 80 μM dynasore, respectively ([157]25–[158]29). Subsequently, FITC-labeled C-β-LG/DSF nanoparticles were added to the wells and incubated for 2 hours. The cells were then washed twice with PBS, and the collected cells were subjected to flow cytometry. Endosomal escape of C-β-LG/DSF Neuron-ICR cells were seeded in a six-well plate under cell culture conditions. FITC-labeled C-β-LG/DSF nanoparticles were added to the wells and incubated for 1, 4, and 8 hours, respectively. The cells were then washed twice with PBS and incubated with Neurobasal+++ medium containing Lysotracker for 30 min, and the collected cells were subjected to CLSM. ImageJ (version 2.9.0) software was used for fluorescent colocalization analysis. Detection of C-β-LG/DSF to cross BBB The TBI cell model was prepared as mentioned in the “TBI cell model” section. After preparing the injury model for 12 hours, neuron-ICR cells were cultured in the lower compartment of the culture plate. HBMEC cells were cultured in Transwell. When the trans-endothelial electrical resistance reached 150 ohms•cm^2, BBB formation was successful. C-β-LG/DSF/FITC and β-LG/DSF/FITC were added to the upper chamber of Transwell. The fluorescence intensity of FITC (green) in C-β-LG/DSF/FITC–incubated HBMEC and neuron-ICR cells was detected by CLSM after 4 hours. Cell membranes of HBMEC were labeled with Dil fluorescent dye (red). Neuron-ICR nuclei were labeled with DAPI (blue). The fluorescence intensity of the upper and lower chamber of each group was detected by a microplate reader after a 4-hour incubation. When red and green were double-labeled in images, red was replaced by purple pseudo-color. The fraction of C-β-LG/DSF/FITC and β-LG/DSF/FITC penetrating the BBB was determined by calculating the ratio of the lower chamber fluorescence intensity to the sum of the upper and lower chamber fluorescence intensity in each group. The calculation formula is as follows: Penetration rate (%) = Upper chamber fluorescence intensity/(Upper chamber fluorescence intensity + Lower chamber fluorescence intensity) × 100%. Animal Six-week-old male ICR mice were purchased from Zhejiang Weitonglihua Company (Zhejiang, China), with certificate number 20220826Abzz0619000492. They were housed in individually ventilated cages at a controlled environment temperature of approximately 24°C and provided with sufficient food and water. After 1 to 2 weeks of adaptation, TBI mouse model experiments were performed. All animal experimental procedures were carried out in accordance with the relevant regulations of the “Jiangsu Provincial Experimental Animal Management Measures” and have been approved by the Animal Use Committee of Xuzhou Medical University (no. 202208S110). The TBI mouse model was randomly divided into five groups: sham group (sham operation group, only the right parietal bone was opened without inducing injury), TBI group (control group, without treatment after injury), DSF group (intraperitoneal administration of free DSF drug after injury), β-LG/DSF group (intravenous administration of β-LG–encapsulated DSF nanomedicine after injury), and C-β-LG/DSF group (intravenous administration of CAQK-modified β-LG–encapsulated DSF nanomedicine after injury), with five mice in each group. TBI animal model ICR mice were randomly divided into sham, TBI, DSF, β-LG/DSF, and C-β-LG/DSF groups. We modified the mature technique of Feeney’s TBI model. ICR mice were anesthetized with isoflurane gas (induction concentration: 3%, maintenance concentration: 2%, and gas flow rate: 400 ml/min). After complete anesthesia, the ICR mice were fixed on a stereotaxic instrument with the nose close to the gas anesthesia machine mask. The scalp hair on the cranial vertex was trimmed with scissors, followed by routine disinfection with iodine and deiodination with alcohol. The scalp was incised along the midline of the frontal vertex of the ICR mouse for approximately 1.5 cm to expose the skull and remove the periosteum. A hole was drilled on the right side of the sagittal suture of the skull using a handheld skull drill, with a diameter of approximately 0.5 cm (careful operation to ensure that the dura mater was not injured). A mouse TBI impactor (Zhenghua Biological Instrument Co. Ltd., Anhui) was used to induce TBI, with a hammer mass of 40 g and a fall height of 7.5 cm. After successful preparation of the TBI model, the injured brain tissue and dura mater were promptly handled. In case of brain bleeding, sterile cotton balls were used to compress and stop bleeding. The scalp was carefully sutured and disinfected. PBS, DSF (DSF dissolved in sterile edible sesame oil and injected intraperitoneally), β-LG/DSF, and C-β-LG/DSF were administered at a dose based on 10 mg/kg of DSF. The sham group received no treatment. All groups were administered once a day for three consecutive days and transferred to cages with group labels affixed, with five mice in each group. After 1 hour of TBI, mice status was observed. Mice were carefully monitored 2 hours after TBI surgery in the cages after their waking up. If a mouse died after TBI administration but before the end of the 2-hour observation period (2 hours after TBI administration), the data were not included for analysis. Pharmacokinetic detection The TBI animal model was prepared as mentioned in the “Detection of C-β-LG/DSF to cross BBB” section. DSF (intraperitoneal injection), β-LG/DSF, and C-β-LG/DSF were injected through the caudal vein at the dosage of 10 mg/kg DSF. One hundred microliters of blood was collected from tail vein at different time points (5 min, 10 min, 20 min, 30 min, 1 hour, 2 hours, 4 hours, 8 hours, 12 hours, and 24 hours), and transferred to a medical anticoagulant tube. Ten microliters of 10% phosphoric acid was added and centrifuged at 15,000 rpm. Fifty microliters of plasma was carefully absorbed and transferred to a new 1.5-ml centrifuge tube. Fifty microliters of protein precipitator was added (acetonitrile:10% trichloroacetic acid = 2:3). The plasma was thoroughly mixed and centrifuged at 15,000 rpm for 10 min. The supernatant was absorbed with pipetting gun, filtered, and 1 μl was injected. The concentration of DSF in each tissue was determined by the UPLC-MS/MS method. The experimental conditions and parameters of UPLC-MS/MS were the same as those in the “Preparation and characterization of C-β-LG/DSF” section, and the experiment was repeated three times. In vivo tissue targeting distribution analysis The TBI animal model was prepared as mentioned in the “Detection of C-β-LG/DSF to cross BBB” section. C-β-LG/DSF and β-LG/DSF were labeled with Cy5.5 NHS and injected into mice. After 6, 12, 24, and 48 hours after injection, the brain, heart, liver, spleen, lungs, and kidneys of mice were dissected out and imaged by a small animal imaging system. C-β-LG/DSF and β-LG/DSF were labeled with FITC and injected into mice. After 6, 12, 24, and 48 hours after injection, the brain, heart, liver, spleen, lung, and kidneys of mice were dissected out, homogenized, and weighed to detect the intensity of FITC fluorescence and DSF concentration in each organ by a microplate reader. One milliliter of tissue homogenate was collected and transferred to a centrifuge tube. One hundred microliters of 10% phosphoric acid was added and centrifuged at 15,000 rpm. Five hundred microliters of supernatant fluid was carefully absorbed and transferred to a new centrifuge tube. Five hundred microliters of protein precipitator was added (acetonitrile:10% trichloroacetic acid = 2:3). The tissue homogenate was thoroughly mixed and centrifuged at 15,000 rpm for 10 min. The supernatant was absorbed with pipetting gun, filtered, and 1 μl was injected. The concentration of DSF in each tissue was determined by the UPLC-MS/MS method. The experimental conditions and parameters of UPLC-MS/MS were the same as those in the “Preparation and characterization of C-β-LG/DSF” section, and the experiment was repeated three times. Enzyme-linked immunosorbent assay The TBI cell model was prepared as mentioned in the “TBI cell model” section. After preparing the TBI cell model for 12 hours, the culture medium of each cell group was transferred to sterile centrifuge tubes and centrifuged at 3000 rpm, 4°C for 20 min, and then the supernatant of each group was collected. ELISA experiments were performed to detect the cellular release levels of IL-1β, IL-18, and LDH. Twenty-four hours after the preparation of the TBI animal model, the mice were injected intraperitoneally with 1% pentobarbital sodium and were deeply anesthetized. The mice were killed by decapitating, and the brain tissue was removed. The brain was separated along the longitudinal cerebral fissure, and the quality of the brain tissue on the striking side was accurately weighed and recorded. The tissue grinder was used for full grinding (ice water bath), and after full homogenization, the grinding liquid was transferred to a centrifuge tube with a pipette. After centrifugation (4°C, 12,000 rpm, 20 min), the supernatant was taken to detect the release of the above three cytokines in the brain tissue. The specific experimental procedures were strictly carried out according to the instructions of the IL-1β and IL-18 ELISA assay kits (Jonln, JL20253, Shanghai, China) and the LDH ELISA assay kit (Solarbio, BC0685, Beijing, China). Immunoblot assay The TBI cell model was prepared as mentioned in the “TBI cell model” section. After preparing the TBI cell model for 12 hours, neuron-ICR cells were collected. For immunoblot assay, collected cells were lysed in the radioimmunoprecipitation assay lysis buffer containing 50 mM tris-HCl (pH 7.4), 150 mM NaCl, 5 mM EDTA, 10% glycerol, 1% NP-40, 0.1% SDS, 1 mM sodium fluoride, and 1 mM sodium orthovanadate. Twenty-four hours after the preparation of the TBI animal model, the mice were injected intraperitoneally with 1% pentobarbital sodium and were deeply anesthetized. The mice were killed by decapitating, and the brain tissue was removed. The brain was separated along the longitudinal cerebral fissure, and the quality of the brain tissue on the striking side was accurately weighed and recorded. The proper amount of tissue cell lysate was added and fully ground with a tissue grinder (on ice), the grinding liquid was transferred to a centrifuge tube with pipette after full homogenization and centrifuged (4°C, 10,000 rpm, 10 min), and the supernatant, that is, total protein, was taken. Protein sample concentration was determined and stored at −80°C. Proteins were boiled in the loading buffer (Beyotime, P0015L, Shanghai, China). Primary antibodies used for immunoblot were listed as follows: anti–caspase-1 p20 (Santa Cruz Biotechnology, sc-398715; 1:1000), anti–cleaved GSDMD (Cell Signaling Technology, #10137S; 1:1000), anti–caspase-1 p10 (GeneTex, GTX134551; 1:1000), and anti-GAPDH (Proteintech, 60004-1-Ig; 1:1000). Representative uncropped immunoblots were presented in the Supplementary Materials. All immunoblot results were independently repeated for at least three times. Brain water content assay The water content of mouse brain tissue was determined by measuring wet weight and dry weight. On the third day of the preparation of the TBI mouse model (as mentioned in the “TBI cell model” section), 1% pentobarbital sodium was injected intraperitoneally, the mice were deeply anesthetized, and the mice were sacrificed by decapitation. After the brain tissue was removed, the cerebellum and brainstem were removed. The brain was separated along the longitudinal cerebral fissure, the healthy side brain tissue was discarded, and the wet weight of the affected side brain tissue was accurately weighed. Then, the brain tissue of the striking side was placed in a 90°C oven for 3 days, and its mass was accurately weighed as the brain tissue dry weight. The experiment was repeated five times. The formula for calculating water content is as follows: Water content of brain tissue (%) = (Wet weight − Dry weight)/Wet weight × 100%. Evans blue tests The TBI mouse model was prepared using the same method mentioned in the “Detection of C-β-LG/DSF to cross BBB” section). On day 3, Evans blue dye was injected into the tail vein (dissolved normal saline, administered at 2%, at a dose of 2 ml/kg). After 3 hours, the Evans blue dye is fully circulated in the blood system. Mice were deeply anesthetized by intraperitoneal injection of 1% pentobarbital sodium. The mouse chest was cut open by tissue scissors, and 20 ml of normal saline and 20 ml of 4% paraformaldehyde were injected into the apex of the mouse heart (left ventricle) through a catheter. The brain tissue was carefully peeled off, so that it does not get damaged. The brain was separated along the longitudinal cerebral fissure, and the quality of the brain tissue on the striking side was accurately weighed and recorded. The brain tissue was transferred to a centrifuge tube containing 1 ml of PBS, and the tissue grinder was fully homogenized for centrifugation at 4°C at a rotational speed of 10,000 rpm for 10 min. Tissue homogenate supernatant (300 μl) was added to 700 μl of acetone, incubated at room temperature for 24 hours, and centrifuged at 4°C, at a rotational speed of 10,000 rpm, for 10 min. The absorbance value was measured at 620 nm by enzyme label. The corresponding Evans blue content was calculated by plotting a standard curve with known different concentration gradients. MRI test After 3 days of preparation of the mouse TBI model (using the same method as that mentioned in the “Detection of C-β-LG/DSF to cross BBB” section), T2-weighted sequences and DWI-weighted sequences were detected by a small-animal MRI apparatus (Bruker, Bio Spin MRI Pharma Scan 7.0 T). Isoflurane gas was inhaled into anesthetized mice (induction concentration: 3%, maintenance concentration: 2%, and gas flow: 400 ml/min). After induced anesthesia, mice were fixed on an MRI scanner operating bed. The inhalation flow of isoflurane gas was adjusted according to the data during the scan so that the mice would not die from overanesthesia. If a mouse dies during the scan, it will be excluded. T2 parameters were as follows: repetition time: 2500 ms, echo time: 36 ms, layer thickness: 1 mm, field of view: 20 × 20 mm, and image matrix: 256 × 256. DWI parameters were as follows: repeat time: 3000 ms, echo time: 20 ms, section thickness: 1 mm, and field of view: 100 × 110 mm. All MRIs were evaluated by a skilled MRI technician; ImageJ software was used to calculate the volume of cerebral edema around the injury site in T2-weighted images. The ADC value was automatically calculated using Paravision 6.1 software of an MRI scanner based on diffusion-weighted images. The diffusion-weighted images were processed by Paravision 6.1 software with false color. Section staining and immunohistochemistry The brain tissue of mice was taken out after anesthesia and dealt with 10% formalin and ethanol dehydration and then embedded in paraffin. Later sections were stained with Nissl, H&E, GSDMD, GFAP, and Iba-1. Then, sections were observed with a fluorescence microscope (Leica, Germany), and pictures were analyzed with Image-Pro Plus 16.0. RNA-seq analysis At 48 hours after TBI, cortical tissue samples proximal to or located at the injury site were sent to Haiteng Century Biotechnology Co. Ltd. (Beijing, China) for RNA-seq library preparation. After clustering, we performed transcriptome sequencing on the DNBSEQ-T7 platform that generated the raw reads. Data quality was assessed by the FastQC tool after removal of adaptor sequences, ambiguous “N” nucleotides (proportion of “N” > 5%), and low-quality sequences (quality score < 10). The ggplotR package was used to identify StringTie genes. Difference was statistically significant threshold for |log[2] fold change| ≥ 1 or P < 0.05. GO enrichment, KEGG pathway analysis of genes and genomes, and heatmap and volcanic figure analysis were carried out at [159]https://www.omicshare.com (Gene Denovo, Guangzhou, China). Morris water maze On the 21st day after the preparation of the TBI animal model (using the same method as that mentioned in the “Detection of C-β-LG/DSF to cross BBB” section), the MWM experiment was conducted, and the water temperature was set at 24°C. In the first 7 days, mice were randomly placed into water from a certain quadrant. These mice could swim freely in the swimming pool until they found a resting platform, which was hidden 1 cm below the water surface. The time (t) and swimming distance (s) were recorded. The swimming time of each mouse was 120 s. The mice that could not find the platform by the end of time were guided to the platform with the guide stick and stayed for 30 s. On the 8th day, the platform was removed, and quadrants were randomly selected to allow mice to find the original location of the platform. Each mouse swam for 120 s, and the time in the original location and quadrant of the platform as well as the crossing times of the platform were recorded. All data during the test period were recorded and analyzed by a video-tracking system (Anhui Zhenghua Biological Instrument Equipment Co. Ltd., Huaibei, China). Swimming tests were performed on each mouse before the experiment, and those mice that could not swim were excluded. There were five mice in each group. mNSS score On the 3rd, 7th, and 14th day after the preparation of the TBI animal model (using the same method as that mentioned in the “Detection of C-β-LG/DSF to cross BBB” section), the mNSS was performed on the trauma model mice by the double-blind method, and the score was recorded. mNSS scores range from 0 (normal) to 18 (most severe) for movement, sensation, balance, and reflexes. Hanging wire grip test To test the strength, balance, and endurance of forelimbs after brain trauma in rats, on the 14th day after the preparation of the TBI animal model (using the same method as that mentioned in the “Detection of C-β-LG/DSF to cross BBB” section), the hanging wire grip test was conducted on the TBI model mice by the double-blind method. A piece of wire 50 cm above the table top was hung. The hind legs of the mice were wrapped with tape (to prevent the mice from climbing), and the front legs were placed on the wire. The hanging time of mice was recorded, five mice in each group. Toxicity test After 14 days of TBI animal model preparation (using the same method as that mentioned in the “Detection of C-β-LG/DSF to cross BBB” section), mice were deeply anesthetized by intraperitoneal injection of 1% pentobarbital sodium. Twenty milliliters of normal saline and 20 ml of 4% paraformaldehyde were injected into the apex of the heart (left ventricle) through the catheter. The heart, liver, spleen, lungs, and kidneys were dissected and fixed in a centrifuge tube filled with 4% paraformaldehyde solution. Paraffin blocks were prepared, sliced, dewaxed with xylene, dehydrated with alcohol, and stained with H&E staining. The experiment was repeated three times. Blood biochemical examination On day 7 of TBI animal model preparation (using the same method as that mentioned in the “Detection of C-β-LG/DSF to cross BBB” section), mice were lightly anesthetized by intraperitoneal injection of 1% pentobarbital sodium. The blood routine and blood biochemical indexes of each treatment group were detected, and the physiological indexes and liver and kidney function were evaluated by the blood extraction method of inner canthus, which was transferred to the anticoagulant tube and the coagulant promoting tube, respectively. Statistical analyses Each experiment was independently repeated more than three times. All data were included for statistical analyses using GraphPad Prism 9.0. Unpaired Student’s t test (two-tailed) was used for the comparison between unpaired two-groups, and one-way analysis of variance (ANOVA) test was applied for multigroup data comparison. Bar graphs were expressed as means ± SD, with statistical significance at *P < 0.05, **P < 0.01, ***P < 0.001, or ****P < 0.0001. Acknowledgments