Metabolomics reveals the effects of hydroxysafflor yellow A on neurogenesis and axon regeneration after experimental traumatic brain injury

Abstract

Context: Hydroxysafflor yellow A (HSYA) is the main bioactive ingredient of safflower (Carthamus tinctorius L., [Asteraceae]) for traumatic brain injury (TBI) treatment.

Objective: To explore the therapeutic effects and underlying mechanisms of HSYA on post-TBI neurogenesis and axon regeneration.

Materials and methods: Male Sprague-Dawley rats were randomly assigned into Sham, controlled cortex impact (CCI), and HSYA groups. Firstly, the modified Neurologic Severity Score (mNSS), foot fault test, hematoxylin-eosin staining, Nissl's staining, and immunofluorescence of Tau1 and doublecortin (DCX) were used to evaluate the effects of HSYA on TBI at the 14th day. Next, the effectors of HSYA on post-TBI neurogenesis and axon regeneration were screened out by pathology-specialized network pharmacology and untargeted metabolomics. Then, the core effectors were validated by immunofluorescence.

Results: HSYA alleviated mNSS, foot fault rate, inflammatory cell infiltration, and Nissl's body loss. Moreover, HSYA increased not only hippocampal DCX but also cortical Tau1 and DCX following TBI. Metabolomics demonstrated that HSYA significantly regulated hippocampal and cortical metabolites enriched in 'arginine metabolism' and 'phenylalanine, tyrosine and tryptophan metabolism' including l-phenylalanine, ornithine, l-(+)-citrulline and argininosuccinic acid. Network pharmacology suggested that neurotrophic factor (BDNF) and signal transducer and activator of transcription 3 (STAT3) were the core nodes in the HSYA-TBI-neurogenesis and axon regeneration network. In addition, BDNF and growth-associated protein 43 (GAP43) were significantly elevated following HSYA treatment in the cortex and hippocampus.

Discussion and conclusions: HSYA may promote TBI recovery by facilitating neurogenesis and axon regeneration through regulating cortical and hippocampal metabolism, BDNF and STAT3/GAP43 axis.

Keywords: Network pharmacology; brain-derived neurotrophic factor; cortex; growth-associated protein 43; hippocampus; signal transducer and activator of transcription 3; traditional Chinese medicine.

Figure 1.

HSYA promotes TBI recovery. (A) The chemical structure of HSYA. (B) HSYA reduces mNSS after TBI. (C) HSYA decreases foot fault rate after TBI in the subacute stage. (D) HSYA tends to improve weight loss after TBI. (E) Representative images of H&E indicate that HSYA normalizes neuron damage and disorganization (black arrow) and decreases inflammatory cells (white arrows) in the cortex and hippocampus. (F) Representative images of Nissl’s staining indicate that HSYA reduces the number of injured neurons in the cortex and the hippocampus. CCI: controlled cortical impact; CTX: cortex; HP: hippocampus; data are expressed as mean ± standard deviation; n = 15 (A and C), n = 5 (B, D and E); P is calculated by one-way ANOVA followed by Dunnett’s t-test; *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 2.

HSYA facilitates neurogenesis and axon regeneration. (A) Immunofluorescent staining of DCX (red) and PCNA (green) indicates more DCX⁺ immature neurons but not DCX⁺/PCNA⁺ proliferating neurons in the HSYA group in the cortex. (B) The statistical graph shows an increased density of DCX in the cortex after HSYA treatment. (C) The statistical graph displays no significance between CCI and HSYA groups on the number of DCX⁺/PCNA⁺ proliferating neurons. (D) Immunofluorescent indicates more immature neurons proliferating neurons in the HSYA group in the hippocampus. (E) HSYA increases the density of DCX in the hippocampus. (F) HSYA increases the number of proliferating neurons in the hippocampus. (G) Immunofluorescent staining of Tau1 (green) indicates more axons in the HSYA group in the cortex around the wound. (H) The statistical graph shows the increased integrative density of Tau1 in the cortex after HSYA treatment. (I) The statistical graph displays an elevated positive area of Tau1 in the HSYA group than in the CCI one. (J) Immunofluorescence shows more axons in the HSYA group in the hippocampus after TBI. (K) HSYA increases the integrative density of Tau1 in the hippocampus. (L) HSYA elevates the positive area of Tau1 in the hippocampus. Data are expressed as mean ± standard deviation; n = 5 (A–I); P is calculated by one-way ANOVA followed by Dunnett’s t-test; *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 3.

HSYA regulates the metabolic pattern of the post-TBI cortex and hippocampus. (A) PLS-DA plot shows distinct metabolic differences among Sham, CCI and HSYA groups in the cortex. (B) The cross-validation graph shows favourable reliability and predictability with high accuracy (1), R2 (0.9898) and Q2 (0.91921) in the cortex. (C) The 100-times permutation test indicates appreciable discriminability and adaptability of the PLS-DA model in the cortex with p < 0.05. (E) PLS-DA plot shows distinct metabolic differences among Sham, CCI and HSYA groups in the hippocampus. (F) The cross-validation graph shows favourable reliability and predictability with high accuracy (1), R2 (0 0.99135) and Q2 (0.9292) in the hippocampus. (G) The 100-times permutation test indicates appreciable discriminability and adaptability of the PLS-DA model in the hippocampus with p < 0.05. (D) Venn graph shows HSYA regulates 19 metabolites in the post-TBI cortex. (H) Venn graph suggests HSYA regulates nine metabolites in the post-TBI hippocampus. (I) Heatmap displays the expression levels of significantly changed metabolites in the cortex. (J) The expression levels HSYA responsive metabolites in the hippocampus. (K) The related metabolic pathway in the cortex and the hippocampus. N = 10 (A–K).

Figure 4.

Differential metabolites-directed target exploration of HSYA. (A) Metabolite-gene network shows 24 HSYA-responsive metabolites related to 766 genes; (B) Venn graph indicates 43 genes are shared in the metabolite-related gene, neurogenesis and axon regeneration-related gene, and HSYA-related gene. (C) Protein-protein network indicates a central role of Bdnf and Stat3 in the common genes. The deeper red and larger the ellipse, the closer relevance to neurogenesis and axon regeneration.

Figure 5.

HSYA Increases BDNF and GAP43 expression. (A) Immunofluorescent staining of BDNF (red) indicates more BDNF protein in the HSYA group in the cortex. (B) The statistical graph shows an increased density of BDNF in the cortex after HSYA treatment. (C) The statistical graph shows an increased positive area of BDNF in the cortex after HSYA treatment. (D) Immunofluorescent indicates more BDNF protein in the HSYA group in the hippocampus. (E) HSYA increases the density of BDNF in the hippocampus. (F) HSYA increases the positive area of BDNF in the hippocampus. (G) Immunofluorescent GAP43 (red) staining indicates more GAP43 protein in the HSYA group in the cortex around the wound. (H) The statistical graph shows the increased integrative density of GAP43 in the cortex after HSYA treatment. (I) The statistical graph displays an elevated positive area of GAP43 in the HSYA group than in the CCI one. (J) Immunofluorescence shows more GAP43 in the HSYA group in the hippocampus after TBI. (K) HSYA increases the integrative density of GAP43 in the hippocampus. (L) HSYA elevates the positive area of GAP43 in the hippocampus. Data are expressed as mean ± standard deviation; n = 5 (A–I); P is calculated by one-way ANOVA followed by Dunnett’s t-test; *p < 0.05; **p < 0.01; ***p < 0.001.