A total of 60 male Sprague–Dawley rats (300 ± 20 g, 9–10 weeks old) were purchased from Vital River Animal Technology. All animals were housed in a standard environment at a temperature of 24°C and humidity of 52%–60%, and subjected to a 12-h/12-h light/dark cycle. All rats were provided free access to food and water. All animal experiments were approved by Wuhan Myhalic Biotechnology Co., Ltd (HLK-202311169) and were conducted according to the Guide for the Care and Use of Laboratory Animals.

Endovascular perforation was conducted following a previously published protocol to mimic the rupture of intracranial aneurysms [19-21]. Rats were anesthetized with 3% isoflurane through tracheal intubation using a rodent ventilator (Shanghai Ranger Apparatus). Next, a sharpened 4-0 nylon suture was inserted into the left internal carotid artery from the cut of the external carotid artery stump. The suture was put out after perforating the bifurcation of the left anterior and middle cerebral artery, followed by suture of the incision. Sham-operated rats were subjected to a similar surgery except artery perforation.

Rats were randomly assigned into the following 4 groups: (a) sham + dimethyl sulfoxide (DMSO) group (n = 10): sham-operated rats received equal volumes of phosphate-buffered saline (PBS) + 1% DMSO; (b) sham + shikonin group (n = 10): rats underwent sham operation and received 25 mg/kg shikonin; (c) SAH + DMSO group (n = 20, 5 rats were excluded due to death): rats were subjected to SAH and administered equal volumes of PBS + 1% DMSO; (d) sham + shikonin group (n = 20, 2 rats were excluded due to death): rats were subjected to SAH and administered 25 mg/kg shikonin. Shikonin was obtained from Santa Cruz Biotechnology (517-89-5) and dissolved in DMSO. Shikonin concentration used in this study was determined based on that reported previously [22]. Rats were intraperitoneally administered 25 mg/kg shikonin after 1 h of SAH.

SAH severity was evaluated by 2 performers based on the SAH grading scale [20]. Rats were anesthetized and sacrificed after 24 h of SAH induction. The ventral side of the brain was divided into 6 areas. The degree of SAH in each area was scored from 0–3 points based on the volume of blood clots on the basal surface. The specific explanation of each grade was described as reported previously [23]. Rats were scored on 6 segments, and those with an average score of < 7 were excluded from subsequent experiments.

The neurological function of rats was assessed using a beam balance test after 24 h of SAH. Briefly, rats were placed on a beam (590 cm) and allowed to walk. Next, the distance that each rat could walk within 1 min was measured and scored. The scoring criteria were as follows:

(a) 4: rats could walk more than 20 cm; (b) 3: rats could walk a distance less than 20 cm; (c) 2: rats could walk but fell from the beam; (d) 1: rats could not walk but could stay on the beam; (e) 0: rats could not walk and fell from the beam. The assay was performed in triplicate.

The Garcia scoring system was also used to measure rats’ neurological functions after 24 h of SAH. Rats were scored from 6 aspects, and the score for each aspect was graded on 0–3 points. The 6 aspects were spontaneous activity, autonomous moving of each limb, forelimb outstretching, climbing, proprioception of tentacles, and body proprioception.

Brain edema in rats was determined by measuring water content in the brain. Rats were anesthetized and sacrificed 24 h after SAH, and their brains were collected. After removal of clotted blood, the initial weight of the wet brains was obtained. Next, the brains were dried in an oven at 100°C for 24 h, and the dry weight was obtained. The brain water content was defined as (wet weight − dry weight)/wet weight × 100%.

The concentrations of MDA and antioxidant enzymes (SOD and GSH) in the brain cortex were measured using the corresponding assay kits following the manufacturers’ recommendations. The kits included a lipid peroxidation (MDA) assay kit (ab118970, Abcam), SOD assay kit (ab65354, Abcam), and GSH assay kit (ab239727, Abcam). A microplate reader (Thermo Fisher Scientific) was used to determine MDA, SOD, and GSH concentrations at a wavelength of 535 nm, 450 nm, and 420 nm, respectively.

An ROS assay kit (Jiancheng Bioengineering Institute) was used to determine ROS production in brain tissues following the manufacturer’s recommendations. ROS content was determined by measuring fluorescence intensity/mg protein.

Evans blue extravasation was measured as previously described [24] to determine BBB permeability. After 24 h of SAH, the left jugular veins of rats were injected with Evans blue dye (2% in saline, 4 ml/kg; Sigma-Aldrich). The procedure lasted for more than 2 min. The dye was allowed to circulate for 1 h. To clear the intravascular dye, the rats were anesthetized and a transcardial perfusion with PBS was performed. The brains were homogenized in PBS and centrifuged at 14,000 ×g for 30 min at 4°C. Next, the supernatant was mixed with the same volume of trichloroacetic acid, incubated for 1 h at 25°C, and centrifuged at 14,000 ×g for 30 min at 4°C. The absorbance of the supernatant at 620 nm was determined using a spectrophotometer (Thermo Fisher Scientific), and BBB permeability was analyzed.

Total protein in brain samples was extracted using a protein extraction kit (Beyotime). A bicinchoninic acid kit (Beyotime) was used to determine protein concentration. Next, the proteins (50 μg) were isolated using 10%–12% sodium dodecyl sulfate–polyacrylamide gel electrophoresis and loaded on a nitrocellulose membrane (Millipore). The membrane was blocked with 5% nonfat milk and incubated overnight at 4°C with primary antibodies against zonula occludens-1 (ZO-1; sc-33725, 1:1,000, Santa Cruz Biotechnology), Occludin (ab167161, 1:1,000, Abcam), Claudin-5 (35-2500, 1:1,000, Thermo Fisher Scientific), Sirt1 (ab110304, 1:1,000), Nrf2 (ab313825, 1:500), HO-1 (ab305290, 1:1,000), and GAPDH (loading control, ab9485, 1:2,500, Abcam). Next, the membranes were washed and incubated with secondary antibodies. Enhanced chemiluminescence reagent (Millipore) was used to visualize the protein bands, and the intensity of the bands was quantified using ImageJ software (National Institutes of Health). The uncropped images of bands are provided in Supplementary Fig. 1, 2.

Data were processed using GraphPad Prism 8 (GraphPad Software) and are presented as mean ± standard deviation. Shapiro–Wilk test was used to assess whether the means of the variables were the same or different. One-way analysis of variance followed by Tukey’s post-hoc analysis were used for comparisons among the 4 groups. p < 0.05 was considered statistically significant. The detailed numerical results of the statistical analysis are provided in Supplementary Table 1.

The chemical structure of shikonin is presented in Fig. 1A and the design idea is shown in Fig. 1B. Sprague–Dawley rats were used to establish an animal model of SAH and were primarily assigned into the following 2 groups: a sham-operated group and an SAH-induction group. After the establishment of the rat model of SAH, DMSO or shikonin was administered to rats in the sham + DMSO, sham + shikonin, SAH + DMSO, and SAH + shikonin groups. SAH grade and neurological score were determined after 24 h of SAH induction. Next, western blotting was performed and levels of the biochemical markers were determined.

SAH can cause neurologic complications, including brain edema and neurological deficits. Therefore, behavior tests were conducted to evaluate the neurological functions of rats. In the beam balance test, rats in the SAH + DMSO group had a low score of 1 or even 0 with significantly reduced neurological behavior scores compared with those in the sham + DMSO group (Fig. 2A). According to the modified Garcia scoring system, model rats showed obvious neurological deficits versus those in the sham-operated group (Fig. 2B). The scores of rats in the sham + shikonin group were similar to those in the sham + DMSO group, indicating that shikonin did not cause brain damage to rats (Fig. 2A, B). Shikonin administration prominently alleviated the neurological functions of model rats, as evidenced by the increased scores of rats in the SAH + shikonin group compared with those in the SAH + DMSO group (Fig. 2A, B). Additionally, the brain water content increased markedly after SAH modeling, which was effectively alleviated after shikonin treatment (Fig. 2C). Sixty rats were used for this study; 7 rats were excluded as they died within 24 h of inducing SAH. The mortality rate of rats with SAH was 17.5% (7/40), and none of the rats died in the sham groups (0/20) (Table 1). The brain samples of rats from the 4 groups are shown in Fig. 2D. Subarachnoid blood clots can be seen on the ipsilateral side, within the surface of the brain stem and around the circle of Willis after SAH modeling (Fig. 2D). These changes induced by SAH were mitigated after shikonin treatment (Fig. 2D). The SAH grade showed a significant increase in the SAH + DMSO group compared with that in the sham + DMSO group (Fig. 2E). There was no notable difference in SAH grade between the SAH + DMSO group and the SAH + shikonin group (Fig. 2E). Overall, shikonin could alleviate neurological deficits in rats with SAH but hardly change the extent of SAH.

Changes in the levels of oxidative stress markers in the brain tissues of rats were measured to determine the effect of shikonin on oxidative stress after SAH. Shikonin treatment led to a minimal change in MDA and GSH levels and SOD production in rats in the sham group (Fig. 3A-C). On the other hand, SAH modeling led to a marked elevation in MDA content while decreasing GSH content and SOD production. Shikonin treatment exerted a rescuing effect by reversing the increase in MDA content and the reduction of GSH and SOD levels (Fig. 3A-C). Furthermore, ROS content was markedly elevated in response to SAH induction, and this trend was markedly counteracted after shikonin administration (Fig. 3D). These findings collectively indicated that shikonin could attenuate oxidative stress after SAH by improving the dysfunction of oxidative indicators and reducing ROS levels.

The effect of shikonin on BBB permeability after SAH was determined based on Evans blue extravasation. Extravasation was significantly promoted after 24 h of SAH versus that in the sham + DMSO group (9.78 ± 1.1 versus 3.95 ± 0.49) and the alteration in rats with SAH was reversed after shikonin treatment (6.22 ± 0.71) (Fig. 4A). The protein levels of the tight junction proteins of the BBB (ZO-1, Occludin, and Claudin5) in brain tissues were quantified using western blotting. ZO-1, Occludin, and Claudin-5 levels were noticeably decreased in the SAH + DMSO group, whereas shikonin treatment reversed this change (Fig. 4B, C). These results suggested that shikonin attenuated BBB permeability after SAH by reducing the loss of tight junctions.

Western blotting was performed to measure the protein levels of Sirt1, Nrf2, and HO-1 in the cerebral cortex of rats to determine the mechanism related to the protective role of shikonin against SAHinduced brain injury. SAH induction prominently reduced Sirt1, Nrf2, and HO-1 levels compared with those in sham-operated rats (p < 0.001, Fig. 5A-C). Shikonin treatment significantly rescued the reduction in Sirt1, Nrf2, and HO-1 levels in the brain tissues of rats with SAH (p < 0.001, Fig. 5A-C). These findings indicated that shikonin activated Sirt1/Nrf2/HO-1 signaling in rats with SAH.