E2F4 Promotes Neuronal Regeneration and Functional Recovery after Spinal Cord Injury in Zebrafish

Abstract

Mammals exhibit poor recovery after spinal cord injury (SCI), whereas non-mammalian vertebrates exhibit significant spontaneous recovery after SCI. The mechanisms underlying this difference have not been fully elucidated; therefore, the purpose of this study was to investigate these mechanisms. Using comparative transcriptome analysis, we demonstrated that genes related to cell cycle were significantly enriched in the genes specifically dysregulated in zebrafish SCI. Most of the cell cycle-related genes dysregulated in zebrafish SCI were down-regulated, possibly through activation of e2f4. Using a larval zebrafish model of SCI, we demonstrated that the recovery of locomotive function and neuronal regeneration after SCI were significantly inhibited in zebrafish treated with an E2F4 inhibitor. These results suggest that activation of e2f4 after SCI may be responsible, at least in part, for the significant recovery in zebrafish. This provides novel insight into the lack of recovery after SCI in mammals and informs potential therapeutic strategies.

Keywords: DREAM complex; E2F4; comparative transcriptome analysis; spinal cord injury; systems pharmacology; zebrafish.

Figure 1

Venn diagrams of differentially expressed genes in mouse, rat and zebrafish SCI. Transcriptome data of SCI in mouse (GSE47681), rat (GSE45006), and zebrafish (GSE39295) were downloaded from a public database (GEO). The DEGs between SCI and control groups in each model at 1 or 3 dpi were identified using a false discovery rate of 20% as the threshold. The numbers of DEGs in each group and the overlap between groups are shown in the Venn diagrams.

Figure 2

Identification of e2f4/tfdp1/foxm1 and Nfic/Tead4 as the key TFs that regulate DEGs specific for zebrafish or mouse/rat SCI. (A) DEGs specific for zebrafish SCI potentially regulated by e2f4, tfdp1 and/or foxm1 at 1 or 3 dpi are shown. (B) DEGs specific for mouse/rat SCI potentially regulated by Nfic and/or Tead4 at 1 or 3 dpi are shown.

Figure 3

Larval zebrafish SCI model used in this study. At 5 dpf, the spinal cords of zebrafish were injured using a device for microdissection (Movie S1). At 0, 1, 2, and 3 dpi, the locomotive behavior of zebrafish was analyzed (Figure 4, Movies S2, S3). At 3 dpi, in vivo imaging of the spinal cord was performed (Figure 5). A representative SCI in zebrafish is indicated by the red arrow.

Figure 4

Recovery of zebrafish locomotive behavior after SCI was inhibited by treatment with E2F4 inhibitor. (A) Protocol for the assessment of zebrafish locomotive behavior. Zebrafish were placed in each well of a 48-well-plate. The plate was set in a device to monitor the behavior of zebrafish in real-time. In the device, the plate was illuminated for 30 min for acclimatization, followed by five sets of dark (3 min) and light (3 min) periods. Zebrafish with SCI (B–D) or without SCI (E–G) were treated with an E2F4 inhibitor HLM006474 (5 μM) or with vehicle. The distance moved during the base (30 min, B,E), dark periods (15 min, C,F) or light periods (15 min, D,G) are shown. N = 36 for each group with SCI and eight for each group without SCI. ^#p < 0.05 compared to SCI without HLM006474.

Figure 5

Neuronal regeneration after SCI was impaired in zebrafish treated with E2F4 inhibitor. (A) Representative in vivo imaging of Tg(eno2: Cerulean) zebrafish. (B) Representative in vivo imaging of Tg(eno2: Cerulean) zebrafish with SCI at 3 dpi treated with or without HLM006474. (C) The Cerulean area in the ROI was quantified and compared between zebrafish with SCI treated with or without HLM006474. N = 24 and 22 for zebrafish without and with HLM006474, respectively. ^#p < 0.05 compared to SCI without HLM006474.