Abstract

   Spinal cord injury (SCI) is a critical condition affecting the central
   nervous system that often has permanent and debilitating consequences,
   including secondary injuries. Oxidative damage and inflammation are
   critical factors in secondary pathological processes. Selenium
   nanoparticles have demonstrated significant antioxidative and
   anti-inflammatory properties via a non-immunosuppressive pathway;
   however, their clinical application has been limited by their
   inadequate stability and functionality to cross the blood-spinal cord
   barrier (BSCB). This study proposed a synthesis method for
   ultra-small-diameter lentinan Se nanoparticles (LNT-UsSeNPs) with
   significantly superior reactive oxygen species (ROS) scavenging
   capabilities compared to conventional lentinan Se nanoparticles
   (LNT-SeNPs). These compounds effectively protected PC-12 cells from
   oxidative stress-induced cytotoxicity, alleviated mitochondrial
   dysfunction, reduced apoptosis. In vivo studies indicated that
   LNT-UsSeNPs efficiently penetrated the BSCB and effectively inhibited
   the apoptosis of spinal neurons. Ultimately, LNT-UsSeNPs directly
   regulated the PI3K-AKT-mTOR and Ras-Raf-MEK-ERK signaling pathways by
   regulating selenoproteins to achieve non-immunosuppressive
   anti-inflammatory therapy. Owing to their ultra-small size, LNT-UsSeNPs
   exhibited strong spinal barrier penetration and potent antioxidative
   and anti-inflammatory effects without compromising immune function.
   These findings suggest that LNT-UsSeNPs are promising candidates for
   further development in nanomedicine for the effective treatment of SCI.

Graphical abstract

   [48]graphic file with name 12951_2024_3054_Figa_HTML.jpg

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s12951-024-03054-7.

   Keywords: Selenium, Spinal cord injury, Antioxidant, Anti-inflammatory,
   Immune protection

Introduction

   Spinal cord injury (SCI) with the high rates of disability and tissue
   damage is a prevalent and serious disorder of the central nervous
   system, ultimately and is associated with considerable incidence and
   mortality. Approximately 700,000 new cases of SCI are reported annually
   worldwide [[49]1]. SCI is typically caused by high-intensity trauma,
   including vehicular falls, accidents, violence, but can also result
   from infections, tumors, degenerative spinal diseases,
   ischemia-reperfusion injury, and vascular abnormalities [[50]2, [51]3].
   SCI effect motor, sensory, and autonomic functions with the temporary
   or permanent impairment, imposing a substantial psychological and
   economic burden on patients and their families. Pathologically, the
   primary injury is an initial contusion or transection caused by
   immediate or sustained mechanical compression or traction on the spinal
   cord [[52]4–[53]8]. Secondary injury is the response of the body and
   cells to primary injury and involves complex changes in the immune
   system, nervous system, and blood circulation. These changes include
   spinal cord ischemia, neuronal death, oxidative stress, inflammatory
   responses, neuronal apoptosis, axonal demyelination, and glial scar
   formation. Secondary injury is a key factor leading to spinal cord
   dysfunction and impedes recovery, with inflammation and oxidative
   stress playing critical roles in the process. Many studies have showed
   that oxidative stress is crucial for the onset and progression of SCI.
   Oxidative stress at the injury site results from a disparity between
   the production and neutralization of ROS, which encompasses reactive
   oxygen molecules like H[2]O[2], O[2]^-, ^1O[2], and O[3] [[54]9,
   [55]10]. Research has confirmed that oxidative stress and inflammation
   mutually promote each other in an inflammatory environment, where
   inflammatory cells release significant amounts of ROS, exacerbating
   oxidative damage [[56]11]. Following SCI, necrotic neurons and other
   cells release high levels of ROS. Owing to high levels of
   polyunsaturated fatty acids, oxidative metabolic activity, and limited
   antioxidant capacity, neurons and glial cells are prone to oxidative
   stress, damaging adjacent surviving neurons. Additionally, local
   ischemia and hypoxia following SCI lead to mitochondrial dysfunction in
   neurons, further increasing ROS production and increasing local
   oxidative stress levels. Multiple studies have demonstrated that
   inhibiting oxidative stress can improve neuronal survival rates,
   restore neurological function, suppress glial cell activation, and
   reduce the expression of inflammatory factors [[57]12–[58]16]. Thus,
   ROS, oxidative stress, and inflammation are closely related to SCI, and
   the inhibition of secondary oxidative stress post-SCI is crucial for
   the recovery of neurological function.

   Currently, clinical approaches primarily use glucocorticoids, such as
   methylprednisolone (MP), to inhibit the production of inflammatory
   factors after SCI, prevent secondary neuronal death, and inhibit lipid
   peroxidation to reduce ROS generation and alleviate edema [[59]17,
   [60]18]. However, high-dose hormone therapy can cause immune system
   disorders and increase the risk of infection and other complications,
   thereby affecting patient recovery. With recent technological
   advancements, nanomaterials have been extensively studied in disease
   treatment applications and have consistently demonstrated significant
   efficacy. Particularly in nanomedicine, nanoparticles often exhibit
   high catalytic effects owing to their unique particle sizes and rich
   surface-active sites. Hence, developing nanomedicines with small size
   and high catalytic activity has become a goal for scientists. Selenium
   (Se), an essential non-metallic trace element in humans and animals,
   plays a vital role in maintaining health. It has high biological safety
   and compatibility. As a key component of the body’s antioxidant system,
   Se is involved in the synthesis of Se-containing proteins, among which
   glutathione peroxidase (GPX) and thioredoxin reductase (TrxR) are the
   most common; these proteins exhibit excellent antioxidant effects,
   regulating immune responses to inhibit harmful inflammatory reactions
   [[61]19–[62]21]. GPX and TrxR family proteins play crucial roles in
   inhibiting oxidative stress following central nervous system injury,
   suppressing oxidative reactions caused by free radicals, reducing body
   damage, and inhibiting inflammation induced by oxidative stress. These
   proteins are crucial for the proper development and ongoing
   functionality of neurons. Studies related to SCI have found that Se can
   inhibit astrocyte activation and inflammatory factor release after SCI,
   significantly down-regulating the expression levels of variety
   oxidative stress markers, such as superoxide dismutase (SOD) and
   malondialdehyde (MDA). Se also reduces neuronal apoptosis through
   pathways involving Bax, Bcl-2, and caspase-3 [[63]22]. Additionally, Se
   supplementation can cause macrophage polarizing to M2 type results in
   the attenuated expression of interleukin 6 (IL-6), inducible nitric
   oxide synthase (iNOS) and tumor necrosis factor α (TNF-α) to suppress
   inflammation, and inhibit local oxidative stress and inflammation
   levels following SCI [[64]23]. Clinical studies have showed that
   analyzing blood from SCI patients who experienced trauma with or
   without neurological recovery revealed significantly higher
   selenoprotein concentrations in the recovery group than in the
   non-recovery group [[65]24]. The result indicates that blood Se levels
   and selenoprotein status reflect the recovery of patients with SCI and
   that Se aids in their neurological recovery. Selenium nanoparticles
   (SeNPs), a novel Se species, possess excellent antioxidant, anti-aging,
   and antitumor activities with low toxicity, high drug-loading capacity,
   superior biocompatibility, and degradability. Previous studies have
   demonstrated their therapeutic effects on SCI [[66]25–[67]30]. However,
   the blood-spinal cord barrier (BSCB) is the main factor influencing
   nanodrug distribution and efficacy in the clinical treatment of SCI.
   Owing to the continuous tight junctions between capillary endothelial
   cells, an intact basement membrane, and the formation of a neuroglial
   membrane by astrocyte foot processes, the BSCB hinders the passage of
   SeNPs approximately 100 nm in size from the bloodstream to the injury
   site. Therefore, there is an urgent clinical need for improving the
   BSCB penetration of SeNPs that do not affect its effect and safety.

   Some literatures reported the modification of transmembrane peptides,
   the improvement of fat solubility and the controlled particles size
   both advance the barrier penetration of nanodrugs. Thus, ultra-small Se
   particles have been prepared to better penetrate the BSCB, accumulate
   effectively at the injury site, and significantly improve SCI treatment
   efficacy. Although the ultra-small size allows Se nanoparticles to have
   better BSCB penetration, it introduces new problems, such as a short
   half-life and poor stability. Lentinan polysaccharides (LNT) exhibit
   excellent biocompatibility and possess a molecular structure rich in
   hydroxyl and other functional groups. These groups interact with the
   surfaces of nanomedicines to form stable chemical bonds or physical
   adsorption layers. This interaction prevents the aggregation of
   nanomedicines in aqueous solutions or physiological environments,
   enhancing their dispersibility and stability. Herein, we prepared an
   ultra-small Se-based nanosystem with LNT modification (LNT-UsSeNPs) to
   achieve high BSCB penetration and suitable biostability. Interestingly,
   its nanoscale size offers advantages when crossing the BSCB,
   effectively minimizing the impact of the barrier on therapeutic
   efficacy. Moreover, zero-valence Se-based drugs with strong reducing
   ability can neutralize ROS and inhibit the inflammatory response
   triggered by secondary oxidative stress injury, thereby reducing
   neuronal apoptosis. Finally, they provide protective benefits to an
   animal’s immune system by regulating selenoproteins, thus avoiding the
   immunosuppressive effects caused by glucocorticoids. Therefore, the
   ultra-small Se-based nanosystem, which possesses antioxidant,
   anti-inflammatory, and immune-protective functions, represents a
   promising non-immunosuppressive therapeutic strategy for SCI.

Results and discussion

Design and synthesis of LNT-UsSeNPs

   Firstly, we successfully synthesized two types of Se nanoparticles:
   LNT-SeNPs and LNT-UsSeNPs. Uniformly dispersed LNT-SeNPs with an
   average size of ~ 100 nm were revealed by transmission electron
   microscopy (TEM) images. Conversely, good monodispersity and a
   spherical shape were showed by LNT-UsSeNPs, which were notably smaller,
   averaging 10 nm in diameter (Fig. [68]1A and B). Furthermore, the
   results of hydrated particle size and zeta potential analysis showed
   that LNT-UsSeNPs have an average particle diameter of 13.00 ± 1.00 nm,
   considerably smaller than LNT-SeNPs, which have an average size of
   123.25 ± 1.06 nm. The zeta potential of LNT-SeNPs was − 10.21 ± 0.35
   mV, whereas LNT-UsSeNPs exhibited a zeta potential of -4.26 ± 0.39 mV
   (Fig. [69]1C and D). Next, we assessed the stability of the
   nanoparticles by monitoring changes in their zeta potential in various
   solutions. The zeta potential of LNT-SeNPs remained stable between − 3
   and − 5 mV for up to 96 h, after which an increase in negative
   potential indicated increase in stability. This pattern was
   particularly evident in a saline solution (pH 5.0). Meanwhile, the zeta
   potential of LNT-UsSeNPs remained stable between − 5 and − 10 mV for up
   to 96 h, although an increase in negative values was observed at 96 h
   in pH 5.0, 6.8, and Dulbecco’s modified Eagle medium (DMEM). However,
   in a pH 7.4 buffer saline solution, the zeta potential of LNT-UsSeNPs
   remained stable between − 6 and − 8 mV, indicating enhanced stability
   under physiological conditions and highlighting potential biomedical
   applications (Fig. [70]1H, S1, Supporting Information). Moreover, EDS
   analysis indicated that the main structural component of both LNT-SeNPs
   and LNT-UsSeNPs was Se (Fig. [71]1E). The chemical structures of
   LNT-SeNPs and LNT-UsSeNPs were characterized using Fourier-transform
   infrared spectroscopy (FT-IR) and ultraviolet-visible (UV-Vis)
   absorption spectroscopy. UV-Vis absorption spectroscopy showed that
   both LNT-SeNPs and LNT-UsSeNPs had high absorption peaks at 250 nm and
   350 nm, respectively. Additionally, the stretching vibrations were
   observed at 3420 cm⁻¹ for O-H, 2920 cm⁻¹ for C-H, and 1380 cm⁻¹ for
   carboxyl groups, which appeared in the spectra of both LNT-SeNPs and
   LNT-UsSeNPs, confirmed that LNT was successfully modified on the
   surface of UsSeNPs and SeNPs (Fig. [72]1F and G). Moreover, LNT-UsSeNPs
   were degraded after a 3 h incubation in a 1% tert-butyl hydroperoxide
   (TBHP) environment, suggesting that LNT-UsSeNPs can react with TBHP.
   These results indicated that LNT-UsSeNPs, owing to their smaller
   particle size and nanoscale dimensions, exhibited superior antioxidant
   properties and could quickly remove TBHP (Fig. [73]1I).

Fig. 1.

   [74]Fig. 1
   [75]Open in a new tab

   Characterization of LNT-SeNPs and LNT-UsSeNPs. (A, B) TEM images of
   LNT-SeNPs and LNT-UsSeNPs. (C, D) Size distribution and zeta potential
   of LNT-SeNPs and LNT-UsSeNPs in aqueous solution. (E) EDS mapping of
   LNT-UsSeNPs. (F) FT-IR spectra and UV-Vis spectra (G) of LNT-SeNPs and
   LNT-UsSeNPs. (H) Surface zeta potential of LNT-UsSe NPs over time in
   different solutions. (I) TEM images of LNT-SeNPs and LNT-UsSeNPs after
   incubating with 1% TBHP solution for 3 h. (J): The mechanism and
   process by which LNT-UsSeNPs scavenge free radicals in vitro

Assessment of the ROS scavenging capacity of LNT-UsSeNPs

   Selenium-based drugs and materials are commonly recognized as effective
   antioxidants that aid in reversing SCI caused by harmful ROS. Thus, an
   ABTS scavenging assay was used to assess the extracellular ROS
   scavenging capacity of LNT-SeNPs and LNT-UsSeNPs. The results indicated
   that LNT-UsSeNPs exhibited a higher ABTS scavenging rate than
   LNT-SeNPs, with a clear concentration-dependent effect (Fig. [76]2A).
   With a concentration of 80 µg/mL and an incubation period of 5 min,
   LNT-UsSeNPs achieved an ABTS scavenging rate exceeding 80%,
   significantly higher than the 14% observed with LNT-SeNPs.

Fig. 2.

   [77]Fig. 2
   [78]Open in a new tab

   LNT-UsSeNPs reversed the damage of TBHP on PC-12 cell. (A) The in vitro
   antioxidant activity of LNT-SeNPs and LNT-UsSeNPs was determined using
   the ABTS radical scavenging assay. (B) EPR spectral analysis of
   LNT-SeNPs and LNT-UsSeNPs for scavenging •O[2]^−. •O[2]^− was produced
   via the Fenton reaction using the Fe^2+/TBHP system and was detected
   using DMPO. (C) UV-Vis spectra of salicylic acid after reacting with
   •OH radicals generated by the Fenton reaction using the Fe²⁺/TBHP
   system for 10 min. (D) The changes in total ROS levels in PC-12 cells
   were detected using DCFH-DA staining. (E) Uptake of LNT-SeNPs and
   LNT-UsSeNPs (30 µM) by PC-12 cells at different times as detected via
   ICP-MS. (F) PC-12 cells were incubated with different concentrations of
   LNT-SeNPs and LNT-UsSeNPs for 24 h, and then incubated with TBHP (10
   µM) for another 24 h. (G) The ΔΨm of PC-12 cells was detected via JC-1
   staining. (H) PC-12 cells in JC-1 fluorescence image. (I, J) Flow
   cytometry was performed to analyze the effect of LNT-SeNPs and
   LNT-UsSeNPs on the proportion distribution of the cell cycle in PC-12
   cells treated with TBHP. (K, L) Annexin V-FITC/PI double staining was
   employed to assess the impact of LNT-SeNPs and LNT-UsSeNPs on the
   apoptosis of PC-12 cells following TBHP treatment. (M) Mechanism
   underlying the inhibitory activity of LNT-UsSeNPs against oxidative
   damage in PC-12 cells. Dates were expressed as the mean ± SD from three
   independent experiments. Bars with distinct features represent
   statistically significant differences at the P < 0.05 level

   The hydroxyl radical (•OH)-scavenging effects of LNT-SeNPs and
   LNT-UsSeNPs were further investigated using electron paramagnetic
   resonance (EPR) spectroscopy (Fig. [79]2B). The Fe^2+/TBHP system
   generated •OH via the Fenton reaction, which was detected using 5,5′
   -dimethyl-1-pyrroline N-oxide (DMPO). The EPR spectrum revealed the
   characteristic signal of the DMPO-OH adduct, indicating the successful
   generation of •OH. Upon the addition of LNT-SeNPs and LNT-UsSeNPs to
   the Fe^2+/TBHP system, the signal intensity decreased sharply,
   particularly in the system with added LNT-UsSeNPs. The •OH scavenging
   activity of the nanoparticles was also measured using UV-Vis
   spectrophotometry. As showed in Fig. [80]2C, A characteristic peak at
   520 nm in the UV-Vis spectrum was observed, attributed to the reaction
   of salicylic acid with •OH radicals produced through the Fenton
   reaction. As expected, LNT-UsSeNPs effectively scavenged •OH radicals,
   as indicated by their reduced absorbance at 520 nm. These findings
   highlight the strong extracellular free radical scavenging ability of
   LNT-UsSeNPs, suggesting that the nanosystem may effectively reduce
   oxidative stress and minimize oxidative damage in vivo by efficiently
   eliminating free radicals.

ROS scavenging effects and protective effects of LNT-UsSeNPs on oxidative
stress injury in PC-12 cells

   PC-12 cells, a rat-derived pheochromocytoma cell line, are widely used
   as models in studies on neurotoxicity and neurodegenerative diseases.
   To investigate the protective influences of LNT-SeNPs and LNT-UsSeNPs
   on PC-12 cells, we established an in vitro SCI model by using TBHP, a
   well-known inducer of oxidative stress. SeNPs have garnered significant
   attention because of their low cytotoxicity [[81]31]. First, the
   ROS-scavenging effects of LNT-SeNP and LNT-UsSeNPs on PC-12 cells were
   examined. Intracellular ROS levels were evaluated using the
   ROS-sensitive fluorescent probe 2′, 7′ -dichlorodihydrofluorescein
   diacetate (DCFH-DA) [[82]32]. As illustrated in Fig. [83]2D, incubation
   with TBHP for 10 min increased ROS levels in PC-12 cells by 347.18% and
   remained elevated at 236.00% after 120 min compared with the control
   group, which was set at 100%. Notably, LNT-UsSeNPs significantly
   inhibited the increase in intracellular ROS, reducing it to 106.48% at
   120 min, which was lower than the 142.71% observed with LNT-SeNPs. This
   demonstrated that LNT-UsSeNPs have a superior ability to scavenge
   intracellular ROS. Next, we investigated the cell-protective effects of
   LNT-UsSeNPs after ROS scavenging. Initially, the safe dosage ranges of
   LNT-SeNPs and LNT-UsSeNPs in PC-12 cells were determined. The results
   indicated a lack of significant cytotoxicity within the concentration
   range of 0 to 5 µM (Figure S2, Supporting Information).

   We also analyzed the Se content in PC-12 cells at different time points
   post-administration using inductively coupled plasma mass spectrometry
   (ICP-MS). As illustrated in Fig. [84]2E, the intracellular Se
   concentration of LNT-SeNPs and LNT-UsSeNPs groups reached a higher
   level after incubation for 2 h, indicating that LNT-SeNPs and
   LNT-UsSeNPs had better potential to protect PC-12 cells after
   pre-administration for 2 h. Thus, the CCK-8 assay demonstrated that
   exposure to 100 µM TBHP reduced PC-12 cell viability to 65–68%.
   Treatment with both LNT-SeNPs and LNT-UsSeNPs counteracted this effect.
   Specifically, at a Se concentration of 40 nM, LNT-UsSeNPs increased the
   PC-12 cell survival rate from 65 to 83%, which was significantly higher
   than the survival rate of 75% achieved using LNT-SeNPs (Fig. [85]2F).
   In summary, the size effect gives LNT-UsSeNPs better antioxidant
   capacity, improving their ability to mitigate oxidative stress injury.

   Cell cycle arrest and apoptosis were also assessed to explore the
   possible protective mechanisms of LNT-SeNPs and LNT-UsSeNPs on PC-12
   cells, which are the primary mechanisms by which oxidative stress
   impairs cell growth and induces cell death [[86]33]. Mitochondrial
   membrane potential (ΔΨm), an early marker in the mitochondrial-mediated
   apoptotic pathway, was evaluated using JC-1 dye in flow cytometry. As
   showed in Fig. [87]2G, treatment with 4 µM LNT-UsSeNPs reduced the
   proportion of cells with mitochondrial depolarization from 34.98 to
   10.17%, while LNT-SeNPs reduced it to 6.09%. Thus, LNT-UsSeNPs are more
   effective at reducing mitochondrial damage.

   Fluorescence images confirmed these findings, showing that LNT-UsSeNPs
   reversed the TBHP-induced reduction in the red-to-green fluorescence
   ratio (Fig. [88]2H). Thus, LNT-UsSeNPs effectively downregulated
   TBHP-induced cell damage. Additionally, cell cycle analysis
   (Fig. [89]2I and J) indicated that TBHP treatment significantly
   increased S-phase cell cycle arrest in PC-12 cells. The Sub-G1 phase of
   cells was increased from 2.35 to 9.78%, whereas the proportion of cells
   in the G2/M phase was rose from 21.04 to 33.52%. LNT-UsSeNPs treatment
   effectively mitigated the TBHP-induced increasing of Sub-G1 phase (from
   9.78 to 0.91%) and normalized the G2/M phase cell percentage to
   approximately 23.03%. Apoptosis was evaluated using Annexin V-FITC and
   propidium iodide (PI) staining, followed by flow cytometry analysis. As
   showed in Fig. [90]2K and L, the early apoptotic cell percentage was
   significantly increased from 0.55 to 24.40% after TBHP incubating.
   Conversely, the early apoptotic cell count of LNT-UsSeNPs treatment was
   substantially decreased to 2.95%, that highlighted the strong
   antiapoptotic properties of LNT-UsSeNPs. These results strongly
   indicate that LNT-UsSeNPs exert protective effects on mitochondria and
   normal cell cycle progression by scavenging TBHP-induced ROS in PC-12
   cells, thereby inhibiting apoptosis (Fig. [91]2M).

Metabolic and biosafety evaluation of LNT-SeNPs and LNT-UsSeNPs

   To investigate the in vivo metabolism and biosafety of LNT-SeNPs and
   LNT-UsSeNPs, we used ICP-MS to analyze their pharmacokinetics in rats.
   The half-life of LNT-UsSeNPs in the bloodstream of rats was determined
   to be 65.3 h, longer than the 45.6 h observed for LNT-SeNPs
   (Fig. [92]3A and B). This indicated that LNT-UsSeNPs maintained high
   plasma drug concentrations for prolonged periods, potentially enhancing
   drug delivery to SCI sites. Furthermore, ICP-MS analysis revealed that
   the kidneys showed the highest accumulation of these nanoparticles,
   suggesting renal-based metabolism for both LNT-SeNPs and LNT-UsSeNPs.
   Notably, the Se concentration of LNT-UsSeNPs group was up-regulated in
   the spinal cord, suggesting improved absorption by spinal tissues,
   potentially beneficial for treating SCI. ICP-MS analysis of various
   organs reveals that selenium concentrations in the LNT-SeNPs group,
   with the exception of the spinal cord, are higher compared to those in
   the LNT-UsSeNPs group. This observation suggests that, when
   administered at equivalent concentrations, LNT-UsSeNPs exhibit reduced
   absorption by non-SCI-related organs, thus maintaining elevated levels
   in systemic circulation over prolonged durations. Such extended
   circulation increases the likelihood of LNT-UsSeNPs crossing BSCB and
   reaching the targeted site of SCI. (Figure [93]3C and H).

Fig. 3.

   [94]Fig. 3
   [95]Open in a new tab

   In vivo analysis of biosafety and pharmacokinetics of LNT-SeNPs and
   LNT-UsSeNPs. (A) Pharmacokinetic analysis of LNT-SeNPs and LNT-UsSeNPs
   (B) in SD rats. (C–H) Selenium concentrations in various organs of the
   rats at 24 and 48 h post-injection of LNT-SeNPs and LNT-UsSeNPs. (I)
   H&E staining analysis of the major organs of rats after treatment with
   LNT-SeNPs and LNT-UsSeNPs. (J) In vivo fluorescence imaging of
   LNT-SeNPs and LNT-UsSeNPs at the final time point

   To assess the in vivo toxicity of the nanoparticles, we conducted
   histological examinations using hematoxylin and eosin (H&E) staining of
   sections of major organs post-injection. The results in Fig. [96]3I
   indicated no significant tissue damage to the main organs, confirming
   the adequate biosafety of LNT-SeNPs and LNT-UsSeNPs within the
   administered dose range.

   Furthermore, we explored the distribution efficiency and processing of
   LNT-UsSeNPs in vivo. After the injection of fluorescently labeled Se
   nanoparticles, the fluorescence intensity at the SCI site in the
   LNT-UsSeNPs group was significantly higher at various time points than
   in the LNT-SeNPs group, as observed in the fluorescence images of the
   mouse viscera. Furthermore, the images of LNT-UsSeNP group exhibited
   significantly higher fluorescence intensity at the SCI site than those
   of LNT-SeNPs group, with relatively high intensities also observed in
   the liver, spleen, and kidneys (Fig. [97]3J). These results indicate
   that LNT-UsSeNPs can more effectively cross the BSCB through blood
   circulation to reach the SCI site in mice, achieving a therapeutically
   effective local drug concentration that may facilitate subsequent
   treatment of SCI.

Enhancement of motor function and neuronal survival in mice with SCI upon
treatment with LNT-SeNPs and LNT-UsSeNPs

   To further evaluate the neuroprotective effects of LNT-SeNPs and
   LNT-UsSeNPs in vivo, we evaluated the recovery of hindlimb motor
   function in mice following the surgical procedure. Using the Basso
   Mouse Scale (BMS) and inclined plane tests, we recorded performance on
   days 1, 3, 5, 7, 10, 14, 21, and 28. The initial results, displayed in
   Fig. [98]4A, presented hind limb paralysis in all groups except for the
   sham surgery group, with a BMS score of 0 on day one. Over time, we
   observed gradual improvements in the BMS scores across all treatment
   groups, with the LNT-UsSeNPs group demonstrating significant
   enhancement by day 10. This group also showed improved climbing angles
   compared to the sham group (Fig. [99]4B). Gait analysis indicated that
   the SCI group displayed uncoordinated movement and significant
   resistance to hindlimb motion. Notably, the LNT-UsSeNPs group exhibited
   a substantial recovery of hind limb gait and motor coordination, as
   illustrated in Fig. [100]4C.

Fig. 4.

   [101]Fig. 4
   [102]Open in a new tab

   Improvement effect of LNT-UsSeNPs on motor recovery and neuronal
   survival in SCI mice. (A) BMS scores and slope test (B) results of mice
   at different times. (C) Footprint analysis was conducted for each group
   of mice to assess gait and locomotor function. (D) A snapshot from the
   recorded video shows the sequence of hind limb movements for each group
   of mice during walking. Points and lines are used to represent the
   iliac crest, knee joint, and ankle joint. The direction of hind limb
   movement is indicated by an arrow. (E) H&E staining and (F) Nissl
   staining was performed on spinal cord tissues from each group, with
   black arrows indicating the presence of Nissl bodies. (G)
   Immunofluorescence staining of the spinal cord tissue in each group
   (Neun, β-tubulin, myelin basic, and GFAP). Dates were expressed as the
   mean ± SD from three independent experiments. Bars with distinct
   features represent statistically significant differences at the
   P < 0.05 level

   Further investigation of the motion of the mouse hind limbs during
   walking broke down the movements into a sequence of phases–stance,
   push-off, lift, swing, and stance–completing one cycle, as depicted in
   Fig. [103]4D. The SCI group consistently demonstrated hind limb
   paralysis and dragging. In contrast, mice treated with LNT-UsSeNPs
   showed notable improvements, such as restored standing posture (stance
   ability), enhanced weight support, indicated by an increase in iliac
   crest height, and an expanded range of joint motion, facilitating
   effective push and swing movements.

Effects of LNT-SeNPs and LNT-UsSeNPs on spinal cord tissue pathology and
neuronal survival

   To evaluate the effect of LNT-SeNPs and LNT-UsSeNPs treatments on
   spinal cord tissue pathology, the experiments of H&E, Nissl, and dual
   immunofluorescence staining were conducted. H&E staining showed that
   the spinal cord structures in the sham surgery group remained intact,
   displaying well-organized neurons with clearly defined nucleoli. In
   contrast, the SCI group displayed disorganized gray matter, necrotic
   cavities, irregular neuronal shapes, nuclear pyknosis, and motor neuron
   loss. Treatment with LNT-UsSeNPs significantly alleviated these
   pathological changes, indicating a protective effect (Fig. [104]4E).
   Nissl staining, which is used to assess neuronal integrity [[105]34],
   revealed a marked decrease in Nissl bodies in the SCI group, whereas
   LNT-UsSeNPs treatment facilitated the restoration of Nissl bodies
   (Fig. [106]4F). (Neuron count: Sham: 9; SCI: 4; LNT-SeNPs: 7;
   LNT-UsSeNPs: 8) Dual immunofluorescence staining assessed the
   expression of neuronal nuclei protein (Neun), which is important for
   synaptic generation and neural circuit balance, and β-tubulin [[107]35,
   [108]36], which is essential for axon guidance and maturation. We
   observed increased Neun expression and a reduced number of staining
   cavities in the LNT-UsSeNPs group, suggesting decreased neuronal
   apoptosis and improved survival (Fig. [109]4G).

   Myelin integrity in the spinal cord was assessed using myelin basic
   protein staining [[110]37]. In the SCI group, extensive demyelination
   and axonal breaks were prominent. However, the LNT-UsSeNPs group
   exhibited significantly smaller cavities and less axonal damage,
   indicating preserved axonal integrity and myelination. Additionally, we
   evaluated glial fibrillary acidic protein (GFAP) expression to analyze
   astrocytic responses post-injury [[111]38]. SCI typically induces
   astrocytes to produce GFAP, leading to glial scar formation that
   impedes neuronal regeneration. GFAP staining revealed that GFAP
   expression was substantially lower in the LNT-UsSeNPs group, suggesting
   reduced oxidative stress, inflammation, and less glial scar formation.
   Collectively, these results confirm that LNT-UsSeNPs provide
   neuroprotective effects, inhibit axonal demyelination, and reduce glial
   scar formation in vivo, thereby offering significant therapeutic
   benefits for SCI treatment.

LNT-UsSeNPs regulate selenoenzymes to mitigate oxidative stress and enhance
immune protection following SCI

   MDA, a byproduct of lipid peroxidation, serves as a biomarker for
   oxidative stress and is prevalent in the lipid-rich spinal cord, where
   lipid peroxidation is likely to occur post-injury [[112]39]. Thus, we
   evaluated the MDA content in the spinal cord, finding it notably
   decreased in the LNT-UsSeNPs group compared to the SCI group
   (Fig. [113]5A). These results demonstrate that LNT-UsSeNPs exhibit
   superior antioxidant capacity compared to LNT-SeNPs, consistent with
   previous in vitro antioxidant studies.

Fig. 5.

   [114]Fig. 5
   [115]Open in a new tab

   Immunoprotective effect of LNT-UsSeNPs by regulating selenoenzyme
   expression. (A) MDA levels were detected to determine the levels of
   lipid peroxidation (mM) in the spinal cord tissue after injury in the
   mice of different groups (A: Sham, B: SCI, C: SCI + LNT-SeNPs, D:
   SCI + LNT-UsSeNPs). (B) A GSH-Px assay was used to detect the activity
   of GSH-Px in different groups of spinal cord tissue after injury. (C)
   qPCR detection results of anti-inflammatory and antioxidant-related
   Se-containing proteins among the different groups. (A: Sham, B: SCI, C:
   SCI + LNT-SeNPs, D: SCI + LNT-UsSeNPs) (D) LNT-UsSeNPs upregulate the
   expression of selenase in neurons. (E–J) Flow cytometry analysis of the
   proportion of splenic immune cells in the different groups of SCI mice
   (1: Sham, 2: SCI, 3: SCI + MPSS, 4: SCI + LNT-UsSeNPs). (K–R)
   Classification and statistics of white blood cells in SCI mice (1:
   Sham, 2: SCI, 3: SCI + MPSS, 4: SCI + LNT-UsSeNPs). (S) Effect of body
   immunity on repair of SCI. Dates were expressed as the mean ± SD from
   three independent experiments. Bars with distinct features represent
   statistically significant differences at the P < 0.05 level

   The human body contains at least 25 Se-containing proteins, several of
   which function as antioxidant enzymes that contribute to mitigating
   oxidative stress-induced damage [[116]40, [117]41]. Thus, we assessed
   the mRNA expression of the antioxidative selenoprotein as as
   glutathione peroxidase (GSH-Px), TrxR, selenoprotein K (SelK), and
   selenoprotein T (SelT). As illustrated in Fig. [118]5B, treatment with
   both LNT-SeNPs and LNT-UsSeNPs significantly enhanced GSH-Px activity,
   with LNT-UsSeNPs showing notably higher activity than LNT-SeNPs.
   Furthermore, we investigated the effect of LNT-UsSeNPs on the
   expression of antioxidant selenoproteins using qPCR and western blot
   analyses. The qPCR results (Fig. [119]5C) showed that treatment with
   LNT-UsSeNPs significantly increased the mRNA expression of GPX1, GPX2,
   SelW, and TrxR1 compared to the SCI group. Subsequent western blot
   analysis validated that LNT-UsSeNPs upregulated the expression of GPX1,
   GPX2, and TrxR1 in spinal cord tissues. These findings suggest that
   LNT-UsSeNPs mitigate oxidative stress-induced SCI by regulating
   antioxidant selenoproteins, thereby exerting neuroprotective effects in
   mice with SCI. The western blot results further verified the qPCR
   results (Fig. [120]5D).

   The common clinical treatment for SCI involves high-dose corticosteroid
   therapy, specifically with methylprednisolone sodium succinate (MPSS),
   which suppresses immune function despite its anti-inflammatory and
   antioxidative benefits. This suppression can lead to complications such
   as bedsores and urinary and respiratory infections, significantly
   hindering patient recovery and increasing both treatment duration and
   cost [[121]18, [122]42]. Therefore, we investigated whether
   LNT-UsSeNPs, which also exhibit anti-inflammatory and antioxidant
   properties, affect immune function using MPSS as a clinical comparator.
   Flow cytometric analysis of splenic cells revealed significant changes
   in immune cell populations. After MPSS treatment, there was a notable
   decrease in the proportion of B lymphocytes (CD19+) compared to the
   sham groups, while the LNT-UsSeNPs group showed an increase in the
   proportion of B lymphocytes. No significant differences existed among
   the groups in the NK cell proportion (Fig. [123]5E, S4). T cell
   profiling showed that the proportion of CD4 + cells decreased following
   MPSS treatment, whereas that of CD8 + cells increased in the MPSS
   group. Conversely, the LNT-UsSeNPs group showed an increased proportion
   of CD4 + cells, with no significant changes in CD8 + cells. The
   CD4+/CD8 + ratio, a key indicator of immune function, generally
   decreases when cellular immunity is suppressed and immune function is
   compromised [[124]43, [125]44]. The ratio was considerably lower for
   the MPSS group compared to the sham group, suggesting the presence of
   immunosuppression. However, the CD4+/CD8 + ratio in the LNT-UsSeNPs
   group increased, suggesting that LNT-UsSeNPs do not suppress immune
   function but exert antioxidative and anti-inflammatory effects
   (Fig. [126]5F and J, S5). Complete blood count results further
   supported these findings (Fig. [127]5K and R, Table [128]S1). The MPSS
   group showed a decrease in the proportion of lymphocytes among white
   blood cells compared to the sham and control groups, whereas the
   LNT-UsSeNPs group showed an increase in the lymphocyte proportion,
   surpassing that of the sham group. In addition, we found that the
   number and proportion of neutrophils in the MPSS group were
   significantly higher than those in the other groups, suggesting that
   mice injected with a high dose of MPSS experienced a certain degree of
   bacterial infection (Fig. [129]5K and O). This reflects a disruption in
   their immune system, leading to the occurrence of bacterial infection.
   However, no such phenomenon was observed in the LNT-UsSeNPs group.
   Therefore, safeguarding the immune system and minimizing immune-related
   complications during SCI treatment are crucial for patient recovery
   (Fig. [130]5S).

LNT-UsSeNPs exert antioxidant and anti-inflammatory effects to inhibit
neuronal apoptosis via the PI3K-AKT-mTOR-eIF4EBP1 and Ras-Raf-MEK-WIPI2
pathways

   To explore whether LNT-UsSeNPs regulate signaling pathways to inhibit
   neuronal apoptosis, we employed Tandem mass tag (TMT) labeling for
   quantitative proteomic analysis to identify differences in protein
   expression among experimental groups. Comprehensive enrichment analysis
   of the TMT proteomic data identified differentially expressed proteins
   between the LNT-UsSeNPs and SCI groups that had significant biological
   relevance (Fig. [131]6A). Subcellular localization analysis indicated
   that the majority of these differentially expressed proteins were
   primarily found in the cytoplasm, extracellular matrix and nucleus,
   suggesting that they play critical roles in different cellular
   compartments (Fig. [132]6B). Volcano plot analysis identified a number
   of proteins with significant differential expression (Fig. [133]6C),
   shedding light on the potential molecular mechanisms underlying the
   effects of LNT-UsSeNPs. Utilizing the Kyoto Encyclopedia of Genes and
   Genomes (KEGG) pathway enrichment analysis, we annotated the functions
   of the proteins that exhibited variable expression and conducted a
   comparative analysis to distinguish between those that were upregulated
   and those that were downregulated (Fig. [134]6D and F). The results
   demonstrated that LNT-UsSeNPs significantly upregulated key proteins in
   the PI3K-AKT-mTOR and Ras-Raf-MEK-ERK signaling pathways, promoting
   cell growth, proliferation, and differentiation. After SCI, oxidative
   stress and inflammation contribute significantly to secondary damage,
   neuronal injury, and functional loss. The PI3K-AKT-mTOR and
   Ras-Raf-MEK-ERK pathways are crucial in mitigating oxidative stress and
   inflammation. The PI3K-AKT-mTOR pathway promotes neural repair by
   regulating cell survival and inhibiting oxidative stress and
   inflammation. AKT activation reduces free radicals by upregulating
   antioxidant enzymes like SOD, while also suppressing the release of
   inflammatory factors such as TNF-α and IL-6. The mTOR pathway further
   supports neuroprotection by promoting protein synthesis and cell
   regeneration, reducing tissue destruction. The Ras-Raf-MEK-ERK pathway,
   essential for cell proliferation and survival, aids in tissue repair by
   enhancing neuronal survival and glial proliferation. ERK activation
   reduces oxidative stress by increasing antioxidant gene expression and
   limiting free radical accumulation, while also inhibiting
   pro-inflammatory factor production, thereby preventing additional
   neuronal damage and promoting neural repair. Therefore, we hypothesized
   that LNT-UsSeNPs could effectively enhance the survival of neurons
   following SCI by activating these signaling pathways, thereby
   protecting them from apoptosis induced by secondary oxidative stress
   and inflammation. This ultimately leads to an improved neuronal
   survival rate and partial recovery of motor function. Further analysis
   of cellular markers indicated that neuron-related proteins dominated
   the differential proteins (Fig. [135]6G), which was consistent with our
   initial hypothesis and further validated the neuron-specific effects of
   LNT-UsSeNPs. Additionally, KEGG pathway analysis suggested that
   LNT-UsSeNPs may further potentiate their neuroprotective effects
   through modulation of proteins expression associated with antioxidant,
   anti-inflammatory, and apoptotic processes linked to the PI3K-AKT-mTOR
   and Ras-Raf-MEK-ERK signaling pathways (Fig. [136]6I). These findings
   offer new perspectives on the potential mechanisms of action of
   LNT-UsSeNPs in the treatment of SCI and lay the foundation for further
   in-depth research.

Fig. 6.

   [137]Fig. 6
   [138]Open in a new tab

   LNT-UsSeNPs inhibit neuronal apoptosis by exerting antioxidant and
   anti-inflammatory effects. (A) Principal component analysis score
   interaction diagram of the SCI and LNT-UsSeNPs groups (3D). (B)
   Subcellular localization of differentially expressed proteins between
   the SCI and LNT-UsSeNPs groups. (C–F) Volcano plot illustrating the 27
   differentially expressed genes between the SCI and LNT-UsSeNPs groups
   along with the KEGG pathway enrichment analysis of overlapping genes
   (C: Volcano map; D: Enrichment analysis of the differential genes; E:
   enrichment analysis of the upregulated genes; F: Enrichment analysis of
   downregulated genes). (G) Analysis of cellular markers of the
   differentially expressed genes. (H) The molecular mechanism LNT-UsSeNPs
   inhibit neuronal apoptosis. (I) Western blot analysis of the
   differentially expressed proteins identified using TMT proteomics(1:
   Sham, 2: SCI, 3: SCI + MPSS, 4: SCI + LNT-UsSeNPs)

   Western blotting was then performed to validate the proteomic data
   (Fig. [139]6H). The results revealed that LNT-UsSeNPs increased the
   expression of proteins within the PI3K-AKT-mTOR signaling pathway and
   simultaneously suppressed the expression of the translation initiation
   repressor, eukaryotic translation initiation factor 4E binding protein
   1 (eIF4EBP1). This dual action enhanced mRNA translation and protein
   synthesis in neurons. Our data suggest that the upregulation of
   eIF4EBP1/2 activates eukaryotic Initiation Factor 4E (eIF4E),
   initiating cap-dependent mRNA translation. This process is essential
   for neuronal development and the growth of synaptic connections
   [[140]45]. Additionally, LNT-UsSeNPs downregulated the expression of
   the apoptosis-related protein WD repeat domain
   phosphoinositide-interacting protein 2 (WIPI2), enhancing neuronal cell
   survival and reducing apoptosis. Simultaneously, the upregulation of
   the Ras-Raf-MEK-ERK pathway inhibits the expression of Cullin 2, a
   component of the ubiquitin complex, suppressing neuronal apoptosis.

   Furthermore, LNT-UsSeNPs have been showed to reduce the expression
   levels of pro-inflammatory proteins, such as S100A9, iNOS, and IL-6,
   and concurrently enhance the expression of the anti-inflammatory
   protein IL-10. Regarding antioxidant properties, LNT-UsSeNPs promoted
   the expression of apoptosis protease activating factor-1 interacting
   protein (APIP), mitigating oxidative stress-induced neuronal apoptosis.
   Furthermore, LNT-UsSeNPs enhanced the levels of NADH dehydrogenase
   iron-sulfur protein 3 (NDUFS3), maintaining normal aerobic metabolism
   in neurons and upregulated acyl-CoA oxidase 1 (ACOX1) expression,
   stabilizing intracellular lipid metabolism. LNT-UsSeNPs treatment also
   increased the expression of neural cell adhesion molecule 1 (NCAM1),
   facilitating neural regeneration and transmembrane signal transmission,
   which are beneficial for neural function repair post-SCI.

   Collectively, these results robustly demonstrate that LNT-UsSeNPs
   mitigate neuronal apoptosis by upregulating the PI3K-AKT-mTOR and
   Ras-Raf-MEK-ERK pathways and modulating the expression of related
   antioxidant and anti-inflammatory proteins. These effects reduce local
   oxidative stress and inflammation while enhancing the expression of
   proteins involved in neural repair and promoting functional recovery.
   In general, LNT-UsSeNPs exerted anti-inflammatory and antioxidant
   effects without inhibiting the immune system, as demonstrated by
   immune-related and tissue protein-related experiments. This has a
   significant advantage over glucocorticoid therapy, as it avoids related
   infectious side effects.

Conclusions

   In this study, we synthesized ultra-small LNT-UsSeNPs, which exhibited
   obvious therapeutic effects against SCI owing to their improved
   structural design and surface modifications. LNT was used as a
   modifying agent, producing LNT-UsSeNPs to improve stability and
   cellular uptake. LNT-UsSeNPs possess a larger specific surface area
   than LNT-SeNPs, resulting in relatively enhanced antioxidant capacity,
   which effectively reduces ROS-induced cellular damage. In addition,
   LNT-UsSeNPs effectively counteracted TBHP-induced cell cycle G2/M
   arrest, mitochondrial disruption, and apoptosis. In vivo, LNT-UsSeNPs
   significantly enhanced the hind limb motor function of SCI-affected
   mice. Histological analyses revealed the neuroprotective effects of
   LNT-UsSeNPs in mice with SCI. Proteomic and western blot analyses
   confirmed that these nanoparticles activated the PI3K and MAPK
   pathways, upregulated selenoprotein expression, mitigated oxidative
   damage and inflammation, and protected neurons from further injury.
   Furthermore, they exhibit antioxidant and anti-inflammatory properties
   without impairing the immune system. Overall, these ultra-small
   LNT-UsSeNPs offer several advantages. The smaller particles size of
   LNT-UsSeNPs provides an advantage in crossing the BSCB, effectively
   reducing the impact of the barrier on drug accumulation and enhancing
   therapeutic efficacy. Additionally, the strong reductive properties of
   zero-valent selenium in the nanoparticles can neutralize ROS and
   regulate selenoprotein levels, thereby inhibiting secondary oxidative
   stress and associated inflammatory responses, ultimately reducing
   neuron apoptosis caused by secondary injury. Moreover, LNT-UsSeNPs
   exhibited protective effects on the immune system, avoiding the
   immunosuppressive effects commonly induced by anti-inflammatory drugs
   such as glucocorticoids. In summary, LNT-UsSeNPs have promising
   potential as non-immunosuppressive nanomedicines for treating SCI and
   may be a novel therapeutic strategy for managing oxidative stress and
   inflammation via selenoprotein regulation.

Materials and methods

Preparation of LNT-UsSeNPs

   To synthesize ultra-small selenium nanoparticles (UsSeNPs), we began by
   mixing 20 mg of selenium powder with 10 mL of polyethylene glycol 400
   (PEG 400) at ambient room temperature (20–25 °C). The resulting mixture
   was subjected to heating at 290 °C under continuous stirring at 360 rpm
   for 2 h to ensure thorough dispersion and nanoparticle formation. Upon
   completion of the reaction, the resulting UsSeNPs suspension was
   allowed to cool and subsequently stored at room temperature for later
   use. To create LNT-UsSeNPs, we first dissolved 20 mg of LNT in 8.4 mL
   of ultrapure water to produce a homogeneous LNT solution. Following,
   the UsSeNPs suspension (1.6 mL) was gradually introduced to the LNT
   solution dropwise, maintaining a controlled rate of 1–2 s per drop.
   Stirring was sustained at 300–500 rpm at room temperature (20–25 °C)
   for 12 h to promote stable nanoparticle-ligand interaction and uniform
   coating formation. To confirm successful nanoparticle formation and
   quantify the selenium content within the LNT-UsSeNPs, ICP-MS analysis
   was performed.

Preparation of LNT-SeNPs

   The synthesis of LNT-SeNPs was initiated by dissolving 20 mg of LNT in
   9 mL of ultrapure water, ensuring complete solubilization at room
   temperature (20–25 °C). Following, 0.5 mL of a 100 mM sodium selenite
   (Na₂SeO₃) solution was gradually added to the LNT solution. Next, 0.5
   mL of a 400 mM ascorbic acid (vitamin C) solution was added into the
   mixture. The addition of ascorbic acid was performed gradually to
   ensure a controlled reduction process, promoting the uniform formation
   of selenium nanoparticles within the LNT solution. The mixture was
   stirred continuously at 300–500 rpm for 12 h, allowing sufficient time
   for complete reduction and particle stabilization. To remove any
   unreacted components and by-products, the resulting solution was
   dialyzed against ultrapure water for 48 h.The Se concentration in the
   final product, LNT-SeNPs, was determined using ICP-MS.

Characterization of LNT-SeNPs and LNT-UsSeNPs

   The microstructural properties of LNT-SeNPs and LNT-UsSeNPs were
   examined using TEM, providing a detailed visualization of the
   nanoparticle morphology. The particle size, zeta potential and
   stability of LNT-SeNPs and LNT-UsSeNP were analyzed by Malvern
   Zetasizer Nano ZS. In order to evaluate the stability of LNT-SeNPs and
   LNT-UsSeNP under different physiological conditions, stability
   experiment was carried out in different time periods at pH 5.8, 6.1 and
   7.4, phosphate buffer (PBS) and DMEM (0–168 h). The chemical structures
   of the nanoparticles were characterized by UV-Vis and FT-IR
   spectroscopy. These tests are critical for assessing the potential of
   these nanoparticles for biomedical applications, particularly in terms
   of their stability and integrity in diverse biological environments.

Evaluation of the ROS scavenging activity

   The antioxidant capabilities of LNT-SeNPs and LNT-UsSeNPs were assessed
   on the basis of their ability to scavenge ROS using the ABTS
   (2,2’-azino-bis[3-ethylbenzothiazoline-6-sulfonic acid]) assay. ABTS
   solution (5 mM) was then activated by adding 10 mg of manganese dioxide
   (MnO[2]) and filtering through filter paper and a membrane filter. The
   activated ABTS solution was stored in the dark at -20 °C for 12 h prior
   to use. In the experimental setup, various concentrations of the
   nanoparticles (2.5–80 µg/mL) were prepared in PBS. An equal volume of
   these nanoparticle solutions (100 µL) was mixed with ABTS (100 µL)
   solution in a microplate format. The mixtures were subsequently
   incubated at 20–25 °C under dark conditions for 25–30 min. Absorbance
   readings at 734 nm were taken at one-minute intervals over a 15-minute
   period. The ROS scavenging rate was determined utilizing the following
   formula:
   graphic file with name M1.gif

Evaluation of •OH and •O[2]

   Firstly, a 1% TBHP was prepared by dissolving 70% TBHP in ultrapure
   water. The TBHP in the solution undergoes decomposition, leading to the
   generation of •O[2]^−. In the TBHP reaction system, the oxygen
   generated from the decomposition of TBHP can combine with electrons to
   form •O[2]^−. Subsequently, •OH were produced via the Fenton reaction
   using a Fe²⁺/TBHP system, with 1.6 mM FeSO₄ and 1% TBHP for 15 min. The
   scavenging capacities of LNT-SeNPs and LNT-UsSeNPs against •OH and
   •O[2]^− were then analyzed using EPR spectroscopy. Briefly, LNT-SeNPs
   and LNT-UsSeNPs (15 µg/mL) were added to solutions containing the
   generated •OH or •O[2]^− radicals and incubated for 30 min. Then, DMPO
   was added and EPR was used to detect the ability of LNT-SeNPs and
   LNT-UsSeNPs to scavenge •OH and •O[2]^− radicals. The •OH scavenging
   capacities of LNT-SeNPs and LNT-UsSeNPs were indirectly measured by
   detecting the formation of 2,3-dihydroxybenzoic acid from salicylic
   acid (2 mM) and •OH radicals, presenting a characteristic absorption at
   510 nm in UV-Vis spectroscopy. A DHE fluorescent probe was used to
   detect the oxygen-scavenging capacity of LNT-SeNPs and LNT-UsSeNPs.

PC-12 cell culture and assessment of cell viability

   PC-12 cells were obtained from the ATCC (Manassas, VA, USA) and
   cultured in DMEM supplemented with fetal bovine serum (10%) and
   penicillin-streptomycin (1%) at 37 °C in an environment containing 5%
   CO₂. 4 × 10^3 PC-12 cells were plated in 96-well plates. After drug
   incubating, 20 µL of the CCK-8 was added to evaluated the cell
   viability. To investigate the cellular uptake dynamics of LNT-SeNPs and
   LNT-UsSeNPs, PC-12 cells were incubated with these nanoparticles under
   controlled conditions. Briefly, PC-12 (1 × 10^5 cells/mL) were seeded
   in 6-cm culture dishes. Both types of nanoparticles were administered
   to the cells, and incubation periods were set at 2, 4, 8, and 24 h to
   capture the time-dependent uptake. After incubation with LNT-SeNPs and
   LNT-UsSeNPs respectively, the cells were harvested, and the cell
   suspensions were subjected to nitric acid digestion to prepare them for
   elemental analysis. The Se concentrations in the cells at each time
   point were quantitatively measured using ICP-MS. This method allowed
   the precise determination of the amount of Se taken up by the cells
   over specified durations, providing insight into the kinetics of
   nanoparticle absorption by neural-like cells. This approach is crucial
   to understanding the bioavailability and potential therapeutic efficacy
   of Se-based nanoparticles in a neurological context. To evaluate the
   safe dose range of LNT-SeNPs and LNT-UsSeNPs for PC-12 cells, the cells
   were exposed to different concentrations of LNT-SeNPs and LNT-UsSeNPs
   for 24 h. Subsequently, we assessed the viability of PC-12 cells after
   24 h using the CCK-8 assay. To determine the impact of LNT-SeNPs and
   LNT-UsSeNPs on reversing TBHP-induced oxidative damage to cell
   viability, PC-12 cells were treated with LNT-SeNPs and LNT-UsSeNPs for
   24 h, followed by drug removal and subsequent exposure to TBHP for
   another 24 h. Next, the viability of PC-12 cells was evaluated 24 h
   post-treatment using the CCK-8 assay.

Flow cytometry analysis of the cell cycle, apoptosis, and ΔΨm

   1.6 × 10^5 PC-12 cells were cultured in 6-well plates. Initially, the
   cells were treated with LNT-SeNPs and LNT-UsSeNPs for a period of 24 h,
   after which they were further treated with TBHP for an additional 24 h.
   Afterward, probe staining and flow cytometry were conducted to analyze
   the cells. In conducting cell cycle analysis, PC-12 cells were first
   fixed using 70% ethanol and stored at -20 °C for 48 h. Subsequently,
   the cells were stained with PI at room temperature, ranging from 20 to
   25 °C, for a duration of 30 min. For the assessment of apoptosis, the
   cells were stained with Annexin V and PI, adhering to the protocol
   specified in the Annexin V-FITC/PI Apoptosis Detection Kit.
   Additionally, the JC-1 probe at a concentration of 10 mg/L was used to
   stain PC-12 cells, which were then incubated at 37 °C for 30 min.
   Fluorescence microscopy was subsequently utilized to capture JC-1
   fluorescence images of the cells, facilitating the analysis of the ΔΨm.

Detection of total intracellular ROS levels

   PC-12 cells were initially seeded into a 96-well plate (10 × 10⁴
   cells/mL). Following a 24 h period, the cells underwent pre-treatment
   with 4 µM concentrations of both LNT-SeNPs and LNT-UsSeNPs for 6 h.
   Subsequently, the cells were stained by incubation with DCF-DA (37 °C,
   30 min), prior to a 5-min exposure to 100 µM TBHP. The resulting
   fluorescence intensity was measured at the wavelengths of 488/525 nm
   using the BioTek Cytation5 imaging reader.

In vivo metabolism and toxicity studies

   6–8 weeks old Sprague–Dawley rats were procured from the Guangdong
   Provincial Laboratory Animal Center and randomly assigned to two
   groups. Each group was intravenously injected with 1 mL of either
   LNT-SeNPs or LNT-UsSeNPs at a concentration of 3 µM. Orbital blood
   samples were collected before injection and at 0.5, 1, 2, 4, 6, 8, 12,
   24, and 48 h post-injection. Blood Se concentrations were measured
   using ICP-MS. At 24 and 48 h post-injection, two rats from each group
   were euthanized, and their spleens, kidneys, lungs, livers, and hearts
   were collected in 4% paraformaldehyde, followed by paraffin embedding
   and sectioning. The H&E staining of tissue sections for microscopic
   examination to assess the acute visceral toxicity of the nanoparticles.

SCI Animal model establishment

   Female C57/BL6J mice with 20–22 g weight (6–8 weeks) were procured from
   the Guangdong Jicui Yaokang Animal Center. The animal experiment
   protocol has been reviewed and approved by Laboratory Animal Welfare
   and Ethics Committee of Jinan University (Ethics Approval Number:
   IACUC-20230919-06). Mice were anesthetized with tribromoethanol and
   underwent laminectomy at the T10 level under sterile conditions. A
   Model III-NYU impactor (W.M. Keck/USA) was used to induce SCI with
   parameters set at a 0.6 mm impact depth, 1.2 m/s impact speed, and 80
   ms dwell time. Sham-operated animals underwent laminectomy without
   inducing spinal cord damage. The wounds were then sutured in layers and
   were daily injected with ceftriaxone (0.3 g per mouse) for seven
   consecutive days. Mice were randomly divided into four groups (Sham,
   SCI, LNT-SeNPs, LNT-UsSeNPs). 2 mg/kg of LNT-SeNPs and LNT-UsSeNPs were
   administered intravenously. The sham and SCI groups were administered
   equivalent volumes of saline. All treatments were given once daily for
   seven consecutive days, starting on the day of SCI modeling. Behavioral
   assessments Basso Mouse Scale (BMS) and inclined plane tests were
   conducted regularly. On day 28, gait analysis was performed, followed
   by euthanasia of the mice and extraction of the spinal cord.

Behavioral assessments

   Behavioral assessments were conducted by trained, blinded observers
   using the BMS, inclined plane test, footprint analysis, and hind limb
   locomotor cycle analysis. The BMS is a 9-point scale representing the
   recovery of hind limb function, ranging from 0 (no observed hind limb
   movement) to 9 (normal hind limb movement). The inclined plane test
   measured the maximum angle of the surface at which a mouse could
   maintain its position for five seconds. For footprint analysis, the
   mice were coated with the black dye on their paws and were allowed to
   walk along a 0.5-meter-long narrow corridor lined with white paper to
   record their footprints. Additionally, video recordings of mice walking
   along the corridor were used to analyze the locomotor cycle of the hind
   limbs, which included the phases of contact, propulsion, lift-off,
   swing, and back-to-contact.

Histological staining

   After the necessary pretreatment, the tissue was were stained with H&E,
   Nissl, and dual immunofluorescence (Neun/DAPI, β-tubulin/DAPI, Myelin
   basic/DAPI, or GFAP/DAPI). All antibodies were purchased from Abcam,
   USA.

Analysis of LNT-SeNPs and LNT-UsSeNPs metabolism in mice

   Briefly, 2.4 mL of a 2.5 mM LNT-SeNPs solution and 2.4 mL of 2.5 mM
   LNT-UsSeNPs solution were each mixed with 6 mg of Indocyanine
   Green-Thiol (ICG-SH) and stirred for 24 h to obtain ICG-SH-LNT-SeNPs
   and ICG-SH-LNT-UsSeNPs solutions, respectively. Twelve 6–8 weeks-old
   C57 mice were then selected, anesthetized, and subjected to SCI
   modeling. Six mice were randomly chosen and injected with 300 µL of
   ICG-SH-LNT-SeNPs and ICG-SH-LNT-UsSeNPs (300 µl) by intravenous
   injection. Then, the fluorescence imaging was performed on mice before
   and at 0.1, 0.5, 1, 2, 4, 6, and 8 h after drug injection to analyze
   and record the fluorescence intensity emitted by the drugs within the
   mice. After 8 h, the mice injected with ICG-SH-LNT-SeNPs and
   ICG-SH-LNT-UsSeNPs were euthanized, and the fluorescence imaging of
   major organ, including liver, heart, lungs, spleen, kidneys, and spine
   were measured to analyze and record the fluorescence intensity of the
   drugs. Finally, the liver, heart, lungs, spleen, kidneys, and spine
   were digested with aqua regia, quantified, and analyzed for Se
   concentrations using ICP-MS.

In vivo antioxidant capacity analysis

   The antioxidant properties of LNTSeNPs and LNT–UsSeNPs in vivo were
   assessed through the quantification of GSH-Px activity and MDA levels
   in spinal cord homogenates. Firstly, samples were mixed with RIPA lysis
   buffer, which was fortified with protease inhibitors. Then, the BCA
   Protein Assay Kit was used to measure the protein content within these
   homogenates. GSH-Px activity and The MDA content were determined using
   specific assay kits the respective assay kits (Beyotime, China)
   following standard protocols.

qPCR analysis of selenoprotein expression

   Trizol reagent (Takara Biotechnology, Japan) was utilized to extract
   total RNA from the spinal cord. Subsequently, the reverse transcription
   process was carried out using the PrimeScript RT Master Mix (Takara
   Biotechnology, Japan). The cDNA was then amplified via qPCR with SYBR
   Premix Ex Taq II, another product of Takara Biotechnology, Japan, on a
   CFX Connect Real-Time PCR Detection System, which is manufactured by
   Bio-Rad, USA. After thermal cycling conditions, the relative gene
   expression analysis used β-actin as internal control.

Proteomic analysis

   Three mice were randomly selected from each group three days after SCI
   modeling. The animals were euthanized via cervical dislocation, and
   spinal cord tissues were harvested and stored in cryovials, which were
   then frozen in liquid nitrogen. TMT proteomic analysis was then
   performed on the samples. The experimental steps are as follows:

TMT labeling

   Peptides were redissolved in 100 mM TEAB and labeled with TMT reagents
   (peptide = 5:1). The labeling reaction was performed at 20–25 °C for
   1–2 h.

Reversed-phase high-pressure liquid chromatography (HPLC) fractionation

   Peptides from each sample were combined and injected into a C18 column
   for separation via gradient elution (5–90%) at 0.3 mL/min (elution
   peaks: 214 nm), with ten fractions collected, merged based on
   chromatograms, and freeze-dried.

Quantitative detection by nano-LC-MS/MS

   The dried peptides were reconstituted in 0.1% FA, centrifuged at
   20,000×g (10 min), and loaded onto a self-packing C18 column.
   Separation was performed on a Thermo EASY-nLC™ 1200 system with a flow
   rate of 300 nL/min, using a gradient elution from 8 to 90% solvent B
   over 62 min. Peptides were ionized via nano-ESI and analyzed on an
   Orbitrap Exploris™ 480 mass spectrometer in DDA mode (MS range:
   350-1,500 m/z, resolution: 120,000). Secondary fragmentation was
   carried out using HCD at 36% energy with dynamic exclusion (60 s) and a
   resolution of 45,000.

Statistical analysis

   The differential expression thresholds were set at a 1.2-fold change
   relative to baseline (P < 0.05). Subcellular localization predictions
   were conducted using the WoLFPSORT database, while the associated
   protein functions were identified through UniProt and a review of
   relevant literature. The KEGG database was used to perform enrichment
   analysis on the differentially expressed proteins.

Western blot analysis

   BCA assay was conducted to measure the concentration of protein, and
   total protein (30 µg) was separated on SDS-PAGE gel (10%). The proteins
   were transferred to the membrane of PVDF (Millipore, USA) using a
   microtransblot device (Bio-Rad, USA). The membrane after BSA solution
   (5%) blocking was incubated with the primary antibody overnight at
   4 °C. It was then treated with an enzyme-labeled secondary antibody
   (4 °C). GAPDH and β-actin served as internal controls.

   Primary reactants: GPX1, GPX2, TrxR1, PI3K, P-AKT, mTOR, eIF4EBP1, Ras,
   Raf, MEK, ERK1/2, WIPI2, ACOX1, S100A9, iNOS, IL-6, IL-10, NCAM1, APIP,
   β-actin, and GAPDH (Abcam, USA).

   Secondary antibody: Anti-mouse IgG (Cell Signaling Technology, USA).

In vivo immunological analysis in mice after LNT-UsSeNPs treatment

   The C57/BL6J mice were used for immunological analysis. LNT-UsSeNPs
   were administered intravenously at a Se concentration of 2 mg/kg daily
   for three consecutive days, starting from the day of SCI induction.
   MPSS was administered as a high-dose pulse therapy: 0.5 h after
   post-SCI, 30 mg/kg of MPSS was injected intravenously, followed by a
   second injection after 2 h (15 mg/kg), and a third injection of
   15 mg/kg after 4 h. On the following day, 35 mg/kg MPSS was
   administered every 4 h for a total of three doses. Both sham and SCI
   groups received saline. All the mice were treated once daily. Day
   three, blood was collected via orbital puncture for a complete blood
   count with differential count, and the mice were euthanized via
   cervical dislocation. The spleens were then harvested, homogenized, and
   lysed. Splenic immune cells were harvested via centrifugation, labeled
   with specific antibodies, and assessed using flow cytometry.

Electronic supplementary material

   Below is the link to the electronic supplementary material.
   [141]Supplementary Material 1^ (247.2KB, docx)

Acknowledgements