Fig 2: LPS administration acutely reduced postsynaptic density protein 95 (PSD)-95 and synaptophysin (SYP) in different brain regions. (A, B) Rats, PSD95 (cortex:0.98 ± 0.33 vs 2.05 ± 0.08 (control) vs 2.08 ± 0.23 (sham), N = 6, hippocampus: 1.52 ± 0.41 vs 3.63 ± 1.09 (control) vs 3.12 ± 0.46 (sham), N = 6) and SYP (cortex:1.05 ± 0.33 vs 2.81 ± 1.20 (control) vs 2.03 ± 0.61 (sham), N = 6, hippocampus: 0.71 ± 0.29 vs 1.77 ± 0.21 (control) vs 1.75 ± 0.33 (sham), N = 6) were lower in the hippocampus and cortex in LPS treated group than the control (P 0.05 or P 0.01). with 2 weeks of prior XYW administration (LPS + XYW [1.86 g·kg-1]), exhibited a significantly higher level of PSD-95 and SYP in the cortex and hippocampus. 0.93 g·kg-1 Xiaoyao Pills also increased PSD-95 protein in the cortex (PSD-95: cortex:0.98 ± 0.33 vs 1.96 ± 0.35 (0.93 g/kg) vs 2.07 ± 0.27 (1.86 g/kg), N = 6, hippocampus: 1.52 ± 0.41 vs 2.23 ± 0.89 (0.93 g/kg) vs 2.20 ± 0.27 (1.86 g/kg), N = 6) (SYP: cortex: 1.05 ± 0.33 vs 1.52 ± 0.71 (0.93 g/kg) vs 2.04 ± 0.73 (1.86 g/kg), N = 6, hippocampus: 0.71 ± 0.29 vs 0.95 ± 0.11 (0.93 g/kg) vs 1.67 ± 0.12 (1.86 g/kg), N = 6) (P 0.05 or P 0.01) (A1: Mann-Whitney test, A2, B1-2: t-test). Compared with the LPS group, *P < 0.05,**P < 0.01.

Fig 3: The upregulation of TGFß1/IGF-1/BDNF expression in the brain of AD mice by PBMT extended to promote adult hippocampal neurogenesis. A, B Western blotting analysis (A) and quantification (B) of TGFß1/IGF-1/BDNF expression in APP/PS1 and 3xTg-AD mouse brain after PBMT, (n = 3–4 per group). C The concentration of TGFß1/IGF-1/BDNF in brain tissues were measured by enzyme linked immunosorbent assay (ELISA), (n = 3–6 per group). D Representative images of Nestin+ (neural stem cell staining) and neuronal class-III ß-tubulin (Tuj1)+ (newborn neurons staining) expression cells in APP/PS1 and 3xTg-AD mouse hippocampal dentate gyrus (DG) at the end of PBMT. Scale bars, 50 µm. E Quantitative analyses of Nestin+ and Tuj1+ area in the hippocampal DG of each group, (n = 4 per group). F Quantitative analyses of the Nestin and Tuj1 mean fluorescence (MFI) in the brain tissues of each group after PBMT. The Nestin and Tuj1 MFI were detected by flow cytometer, (n = 5–6 per group). G Tuj1 antibody was used to staining the newborn neurons, and then, the expression of a-amino-3-hydroxy-5-methyl-4-isoxazole-propionic acid receptors (AMPAR) and postsynaptic density protein 95 (PSD95) on Tuj1+ neurons were detected by flow cytometer. All quantifications are presented as mean ± SEM and were analyzed by one-way ANOVA test; ***p < 0.001, **p < 0.01, *p < 0.05 versus WT group; ###p < 0.001, ##p < 0.01, #p < 0.05 versus indicated group

Fig 4: L-DL regulates cytoskeleton- and synapse-related gene expressions via modulating their alternative splicing. a Cellular component terms for genes with changed alternative splicing in the hippocampus of L-DL knockdown (KD) mice in GO analysis (n = 1,631, P < 0.001). b Diagram for cellular localization of synapse-, dendrite- and cytoskeleton-related proteins, with their genes undergoing aberrant alternative splicing in neurons. c, d Alternative splicing products of CADM1, GIT2, CHL1, CASK, AGRIN, GIT1, CTTN, PSD95, ADGRL1, SYNJ1, EPB41L3 and MAPT gene, determined by RT-PCR in the hippocampus of control and L-DL KD mice (n = 3). * indicates spliced variants, F and R denote primers used for RT-PCR. 24 cycles or 30 cycles were used to detect higher abundant and lower abundant splicing products, respectively. e, f Alternative splicing products of CAMKV determined by RT-PCR (e) and densitometric analyses (f) in the hippocampus of control and L-DL KD mice (n = 3). I4, intron 4 retention; ?I4, intron 4 exclusion. g, h Protein levels of CAMKV, SNAP25, PSD95, NR2B in the hippocampus of control and L-DL KD mice, measured by immunoblotting (g) and densitometric analyses (h). GAPDH was included as an input control. Statistical analysis was performed using two-tailed Student’s t-test; * P < 0.05, ** P < 0.01, *** P < 0.001; error bars denote the SEM

Fig 5: Ultrastructure of neurons and synapses. (A) Diagrammatic sketch of synapse. (B) In the ND group, the neurons were intact with elliptic nuclei, evenly distributed chromatin, and intact nuclear membrane. Mitochondria, rough endoplasmic reticulum, and ribosomes were intact and clear. (C) In the HSD group, the neuronal cell bodies were slightly wrinkled with aggregated chromatin in the nuclei and intact nuclear membranes. Electron density in the cytoplasm was increased, and a few mitochondria showed mild swelling (orange arrow). In addition, a small amount of autophagy (red arrow) and vacuoles (yellow arrow) was observed. N: nucleus, Mi: mitochondria, RER: rough endoplasmic reticulum, GB: Golgi body. (D, E, H) Compared with the ND group, the HSD group showed a significantly lower density of synapse (red circle) per 170 µm2 by electron microscopy (n=3/group). (F, G, I & J) Western blotting was used to detect the expression of PSD95, BDNF, Syn, p62, Atg5, and LC3. Relative expression of proteins normalized to ß-actin and GAPDH (n=3/group, two-sample t-test). Data are expressed as the mean ± SEM; *p < 0.05, **p < 0.01. Scale bar = 1 µm. Abbreviations: BDNF: brain-derived neurotrophic factor; HSD: high-salt diet; ND: normal diet; PSD95: postsynaptic density protein 95; Syn: synaptophysin.