Administration of all-trans retinoic acid after experimental traumatic brain injury is brain protective

Abstract

Background and purpose: All-trans retinoic acid (ATRA) is a vitamin A metabolite, important in the developing and mature brain. Pre-injury ATRA administration ameliorates ischaemic brain insults in rodents. This study examined the effects of post-traumatic ATRA treatment in experimental traumatic brain injury (TBI).

Experimental approach: Male adult mice were subjected to the controlled cortical impact model of TBI or sham procedure and killed at 7 or 30 days post-injury (dpi). ATRA (10 mg kg-1, i.p.) was given immediately after the injury and 1, 2 and 3 dpi. Neurological function and sensorimotor coordination were evaluated. Brains were processed for (immuno-) histological, mRNA and protein analyses (qPCR and western blot).

Key results: ATRA treatment reduced brain lesion size, reactive astrogliosis and axonal injury at 7 dpi, and hippocampal granule cell layer (GCL) integrity was protected at 7 and 30 dpi, independent of cell proliferation in neurogenic niches and blood-brain barrier damage. Neurological and motor deficits over time and the brain tissue loss at 30 dpi were not affected by ATRA treatment. ATRA decreased gene expression of markers for damage-associated molecular pattern (HMGB1), apoptosis (caspase-3 and Bax), activated microglia (TSPO), and reactive astrogliosis (GFAP, SerpinA3N) at 7 dpi and a subset of markers at 30 dpi (TSPO, GFAP).

Conclusion and implications: In experimental TBI, post-traumatic ATRA administration exerted brain protective effects, including long-term protection of GCL integrity, but did not affect neurological and motor deficits. Further investigations are required to optimize treatment regimens to enhance ATRA's brain protective effects and improve outcomes.

Keywords: all-trans retinoic acid; apoptosis; astrogliosis; axonal injury; hippocampus; neuroinflammation; traumatic brain injury.

FIGURE 1

Experimental protocols and time‐course. Neurological Severity Score (NSS) and rotarod (RR) performance were examined at 1 day before CCI, at 1 dpi and at 7 dpi (cohort 1), or at 1, 7, and 28 dpi (cohort 2). Animals received vehicle or ATRA (10 mg kg⁻¹ body weight, i.p.) shortly after craniotomy and CCI (or sham procedure) and at 1, 2, and 3 dpi. After a survival time of 7 days (cohort 1) or 30 days (cohort 2), brains were processed for histology, mRNA, and protein analyses

FIGURE 2

ATRA administration does not ameliorate neurological deficits or sensorimotor coordination after TBI. Evaluation of neurological function and sensorimotor coordination by (a) Neurological Severity Score (NSS; 0 = no impairment, 12 = maximal impairment) and (b) rotarod, latency to fall (seconds). Animals were tested at the day before surgery (pre‐OP), at 1 dpi and at 7 dpi (two cohorts), and at 28 dpi (one cohort). Values from individual animals and mean ± SD are shown, 7 dpi: CCI vehicle n = 12, CCI ATRA n = 10, sham n = 8 each; 30 dpi: CCI n = 12 each, sham n = 8 each. *P<0.05, significantly different as indicated (CCI and sham animals); Kruskal–Wallis test for each time point separately followed by post‐hoc Dunn correction for multiple comparison

FIGURE 3

ATRA administration has protective effects on brain tissue damage after TBI. Cresyl‐violet stained coronal brain cryosections from vehicle‐treated and ATRA‐treated mice at 7 dpi (a, b) and 30 dpi (e, f). Boxed regions are shown in higher magnification with detail enlargement of the hippocampal GCL and arrows point to the ipsi‐lesional suprapyramidal blade. Scale bars: 1 mm (a) and 500 μm (a, enlargement). (c) Lesion volume at 7 dpi and (g) brain tissue loss at 30 dpi. Data are expressed as lesion volume in % of the ipsi‐lesional hemisphere at 7 dpi (c) and as ipsi‐lesional tissue loss in % of the contralateral hemisphere at 30 dpi (g). (d, h) GCL thickness in ipsi‐lesional and contra‐lesional hemispheres and sham mice at 7 dpi (d) and 30 dpi (h). Values from individual animals and mean ± SD are shown, 7 dpi: CCI vehicle n = 12, CCI ATRA n = 10, sham n = 8 each; 30 dpi: CCI n = 12 each, sham n = 8 each. *P<0.05, significantly different as indicated (vehicle and ATRA treatment) or (CCI ipsi‐lesional and CCI contra‐lesional or sham); Student's t test (c, g) or one‐way ANOVA with post‐hoc Holm–Sidak correction (d, h)

FIGURE 4

Brain protective effects of ATRA treatment are independent of cell proliferation in neurogenic niches and BBB integrity after TBI. (a, b) Double‐immunostaining for Ki‐67 (red) and GFAP (green) in the SVZ at 7 and 30 dpi. Arrows point to Ki‐67 IR cells. Scale bar: 50 μm (a). (c, d) Histograms showing the ratio of Ki‐67 IR cells in ipsi‐lesional vs. contra‐lesional SVZ. (e, f) Double‐immunostaining for Ki‐67 (red) and GFAP (green) in the SGL at 7 and 30 dpi. (g, h) Histograms showing the ratio of Ki‐67 IR cells in ipsi‐lesional to contra‐lesional SGL. (i) Representative IgG protein dots revealed with anti‐mouse IgG. (j, k) Densitometric quantification of IgG dot blots as a proxy of BBB damage. Values represent mean ± SD, 7 dpi: sham n = 8 each, CCI vehicle n = 12, CCI ATRA n = 10; 30 dpi: sham n = 7 each, CCI vehicle n = 9, CCI ATRA n = 11. *P<0.05, significantly different as indicated (CCI and sham animals) or between individual groups; one‐way ANOVA with post‐hoc Holm–Sidak correction

FIGURE 5

Gene expression analyses indicate anti‐inflammatory and anti‐apoptotic effects of ATRA after TBI. Gene expression analyses by qPCR at 7 and 30 dpi of pro‐inflammatory cytokines (a: Il1b; and b: Tnfa), microglia activation marker (c: Tspo), astroglial activation markers (d: Gfap; e: Serpina3n), DAMP (f: Hmgb1), and apoptosis markers (g: Casp3; h: Bax). Values represent mean ± SD, 7 dpi: CCI vehicle n = 12, CCI ATRA n = 10, sham n = 8 each; 30 dpi: CCI n = 12 each, sham n = 8 each. *P<0.05, significantly different as indicated (CCI and sham animals) or between individual groups; one‐way ANOVA with post‐hoc Holm–Sidak correction or Kruskal–Wallis with post‐hoc Dunn correction

FIGURE 6

ATRA does not affect cell death‐associated αII‐spectrin cleavage and attenuates short‐term but not long‐term up‐regulation of the astrogliosis marker protein GFAP after TBI. (a, b) Western blot images showing CCI‐induced generation of spectrin breakdown products (SBDPs) in CCI and sham animals at 7 and 30 dpi. (c–f) Quantification of SBDPs of 145/150 kDa and 120 kDa at 7 and 30 dpi. (g, h) Western blot images showing CCI‐induced GFAP expression including breakdown products (GFAP–BDP). (i, j) Quantification of GFAP protein levels at 7 and 30 dpi. Values represent mean ± SD, 7 dpi: sham n = 8 each, CCI vehicle n = 12, CCI ATRA n = 10; 30 dpi: sham n = 7 each, CCI vehicle n = 9, CCI ATRA n = 11. *P<0.05, significantly different as indicated (CCI and sham animals) or (vehicle and ATRA treatment); one‐way ANOVA with post‐hoc Holm–Sidak correction

FIGURE 7

ATRA reduces short‐term but not long‐term astrogliosis and axonal injury after TBI. (a, b) Immunofluorescence images of cryosections for glial fibrillary acidic protein (GFAP) showing increased perilesional astrocyte activation at 7 and 30 dpi in CCI animals. Scale bar: 50 μm, (c, d) Quantitative assessment of GFAP IR cells at 7 and 30 dpi. (e, f) Immunofluorescence images of cryosections for non‐phosphorylated NF‐H indicating increased axonal injury at 7 dpi in CCI animals. Scale bar: 50 μm, 25 μm (inserts). (g, h) Quantitative assessment of SMI‐32 IR particles at 7 and 30 dpi. Values represent mean ± SD, 7 dpi: CCI vehicle n = 12, CCI ATRA n = 10, sham n = 8 each; 30 dpi: CCI n = 12 each, sham n = 8 each. *P<0.05, significantly different as indicated (CCI and sham animals) or (vehicle and ATRA treatment); one‐way ANOVA with post‐hoc Holm–Sidak correction