Atorvastatin combined with dexamethasone promote hematoma absorption in an optimized rat model of chronic subdural hematoma

Dynamic changes in the hematoma volume in the optimized CSDH model

The design of the experiment was presented in Figure 1. We punctured the dura under a microscope (Figure 1A). The cell-matrix glue mixture was injected in the subdural space (Figure 1B–1D).

We tested the volume of the hematoma immediately, 3 d, 5 d, 7 d, and 14 d by MRI spectrometry (GRE) after CSDH injury. We observed dynamic trends in the hematoma volume at different timepoints after modeling. The volume of the hematoma increased and peaked at 5d post-injury (263.8±52.85 μl, p<0.05). After the peak time, the hematoma was gradually absorbed (Figure 2A–2E). At 14 d, most of the hematoma had been absorbed (47.63±21.96 μl, p>0.05). Based on these results we demonstrated that this model successfully mimicked the chronic process of CSDH rather than the acute process.

Expression changes in inflammatory and angiogenic factors in the capsule of the hematoma at different timepoints post-injury

We performed PCR and ELISAs to test the changes in inflammatory and angiogenic factor expression in the neomembrane and capsule of the hematoma at different timepoints after CSDH injury. The mRNA expression of MMP-9 and VEGF was increased and peaked at 5 d. The expression of these mRNAs was decreased but still higher than that of the control (Figure 3A, 3B), which was consistent with the trend we detected by ELISAs. We also used ELISAs to test the levels of IL-6, IL-8 and IL-10 in the capsule of the hematoma. Similar trends in the expression of the MMP-9 and VEGF proteins were found (Figure 3C–3F). The expression of IL-10 was significantly increased and peaked at 7 d. After the peak time, the expression of IL-10 was reduced at 14 d post-injury but remained higher than that of the control (Figure 3G).

KLF-2 expression after CSDH injury

Western blotting and PCR were performed to detect the expression of KLF-2 in the neomembrane and capsule of the hematoma at different timepoints post-injury. The results showed that KLF-2 expression was inhibited with increasing hematoma volume and peaked at 5 d, reaching nearly half of the control expression, and then gradually increased at 7 d and 14 d but was not higher than that of the control group. More significantly, the trend of KLF-2 was opposite to that of the factors we tested previously (Figure 4A, 4B).

The combined treatment could much more effectively reduce the hematoma volume after CSDH formation

Based on these data, we found the greatest hematoma volume and the largest changes in related protein expression at 5 d post-injury. We chose this timepoint for the following study.

MRI spectrometry (GRE) was performed to detect the changes in hematoma volume and histopathological characterization in different treatment groups post-injury (Figure 5A–5D). We further calculated the total hematoma volume at 5 d in different treatment groups (Figure 5E). The hematoma volume decreased to different degrees in both the monotherapy and combined treatment groups compared with the control group. Moreover, the results showed that in the combined treatment group, the hematoma volume was decreased by nearly half compared with that in the injury group. We observed large and thick hematoma neomembranes in the hematoma group. Robust fresh red blood cells and inflammatory cells infiltrated the hematoma cavity (Figure 5F–5M). Compared with that in the hematoma group, the hematoma cavity in the single atorvastatin and dexamethasone treatment groups had a smaller hematoma volume and lower inflammatory phagocytosis. Notably, the hematoma volume was significantly reduced, the number of fresh red blood cells in the hematoma cavity decreased, and the hematoma capsule was obviously thickened in the combined treatment group. These results suggested that compared with the single treatment, the combined treatment strongly improved the absorption of the hematoma at 5 d post-injury.

The combined treatment decreased vascular permeability significantly after CSDH formation

We detected high permeability in the capsule of the hematoma due to vascular leakage. Therefore, we measured vascular permeability by the evans blue test after establishing the CSDH model and administering different treatments (Figure 6).

We found that the combination treatment reduced the evans blue concentration in the hematoma cavity (Figure 6), suggesting that the leakage of new blood vessels was effectively reduced in the capsule of the hematoma.

The combined treatment more effectively regulated the expression of inflammatory factors and angiogenic factors after CSDH formation

We performed PCR and ELISAs to test the changes in inflammatory factor and angiogenesis factor expression in the neomembranes of hematomas in different treatment groups post-injury. The results suggested that the ratio of Ang-1 and Ang-2 mRNA expression was increased in all treatment groups, indicating more robust mature vessels in the neomembrane (Figure 7A). Besides, we performed ELISAs to evaluate the expression of MMP-9 and VEGF in the hematoma capsules. We found that the levels of both MMP-9 and VEGF were decreased after treatment. Notably, this effect was more obvious in the combined treatment group than in the other groups (Figure 7B, 7C). Then, the levels of IL-6, IL-8 and IL-10 in the capsule of the hematoma were further measured to evaluate the inflammatory response. The MMP-9 and VEGF proteins showed similar trends, which were also consistent with those of IL-6 and IL-8 (Figure 7D, 7E). However, the expression of IL-10 was significantly increased after treatment of the injured rats (Figure 7F).

The combined treatment can remarkably attenuate local inflammation and endothelial injury

Then, we detected the expression of NF-κb and ZO-1 in the different treatment groups post-injury by Western blots and PCR analyses. We found that the combined treatment attenuated the expression of NF-κb and enhanced the expression of KLF-2 and ZO-1 compared with the single treatments (Figure 8A), resulting in reducing the leakage of new blood vessels in the capsule of the hematoma. The same phenomenon was also shown in the PCR assay (Figure 8B).

The combined treatment significantly improved the neurological functional outcome after CSDH injury

We further evaluated the neurological functional outcome by corner turning test scores and forelimb-use asymmetry scores in different treatment groups at different timepoints post-injury. Firstly, we calculated the number of rats died during the experiment timepoint in the different treatment group. We found that a total of 10 (36%) rats died in the CSDH group, which was much higher than the monotherapy group and combined treatment (atorvastatin group: n=6 (33%); dexamethasone group: n=4 (28%); combined group: n=3 (17%)). As shown in Figure 9, the rats in the hematoma group exhibited lower scores at all timepoints post-injury for the corner turning test scores, which indicated poor outcomes, than the rats in the treatment groups. In the forelimb-use asymmetry score, higher scores indicate more serious neurological dysfunction. The scores were strongly increased at 3 d and 5 d. However, the result indicated that the scores decreased in the treatment group, especially in the combined treatment group, compared with the injury group at 3 d and 5 d. All of the results indicated that the rats with CSDH have severe neurological functional deficits, but the combined treatment could more effectively rescue the injured rats and improve their neurological outcomes than the single treatments.

Animals

We obtained adult Sprague Dawley (SD) rats (350-400g,) from Beijing HFK Bioscience (Beijing, China). The rats were housed in the facilities of Tianjin Medical General Hospital. All procedures were approved by the Ethics Committee of Tianjin Medical University General Hospital (Tianjin, China).

Cell culture and CSDH model

BEnd.3 cells were purchased from the ATCC (ATCC, Manassas, VA, USA) and cultured in Dulbecco’s modified Eagle’s medium (DMEM, Corning, Tewksbury, MA, USA) with 10% fetal bovine serum (FBS, 26140079, Gibco, Waltham, MA, USA). We changed the medium every other day.

We performed the CSDH model as our previous study [10]. Matrigel E6909 (Sigma-Aldrich, St. Louis, MO, USA) was placed at 4° C overnight (the colloid liquid state was maintained) before use. We suspended the cells in sterile PBS with a concentration of 4 × 10⁶ cells/ml, and then mixed them with Matrigel at a ratio of 4:1. The final concentration of the cell-Matrigel mix was 10⁶ cells/ml. We placed the mix on ice to maintain the Matrigel liquid state before use.

We generally anesthetized the rats by intraperitoneal injection of 10% chloral hydrate (3 ml/kg). Then, the anesthetized rats were placed in a stereotaxic frame (RWD Life Science, Shenzhen, China). We made a sphenoid burr hole over the right coronal suture with dural integrity. Then, under a microscope (RWD Life Science), the dura was punctured by a 0.2 mm diameter needle. A tapered tip (20G, BD Venflon^TM, Helsingborg, Sweden) was fitted tightly into the hole without contacting the dura. The details of the method are described in our previous studies [8, 36, 38]. A 150 μl mixture or Matrigel alone was injected into the subdural space by a stereotaxic injector (RWD Life Science) at a rate of 25 μl/min. We removed the needle in 10 min until the gel solidified after finishing the injection.

Magnetic resonance imaging (MRI)

We used 3.0T MRI (MR750, GE Healthcare, Chicago, IL,USA) to identify whether the CSDH model was successfully established and to test the hematoma volume. Briefly, immediately and 3, 5, 7, and 14 d after surgery (n=30), anesthetized rats were positioned appropriately with the coil in the MR scanner. MRI images were acquired with the GRE sequence. We calculated the volume of the hematoma at different timepoints by the GE workstation.

Experimental group

Rats with CSDH were randomly separated into four groups: the control group (saline, n=18), single atorvastatin group (3 mg/kg/day, Pfizer, USA, n=18), single dexamethasone group (0.405 mg/kg, Tianjin Pacific Pharmaceutical Co., Ltd., China; n=18), and atorvastatin and dexamethasone combined treatment group (atorvastatin:3 mg/kg/day, dexamethasone: 0.405 mg/kg; n=18). We treated the rats at 1d post-injury by gavage administration, and the period of the treatment was 5 d.

Tissue preparation

In the model group, we sacrificed the rats immediately (n=6) and at 3 d (n=6), 5 d (n=6), 7 d (n=6), and 14 d (n=6) to test the expression of related proteins and mRNAs by enzyme-linked immunosorbent assays (ELISA) and quantitative real-time polymerase chain reaction (qRT-PCR). Besides, western blot was performed with 18 rats (immediately: n=6; 3 d: n=6; 5 d: n=6; 7 d: n=6; 14 d:n=6) to evaluate the expression of proteins. There were 72 rats in the treatment group at 5 d (control: n=18; atorvastatin: n=18; dexamethasone: n=18; combined group: n=18) for H&E staining, Evans blue analysis, ELISAs, and PCR. The tissues were collected and fixed with 4% paraformaldehyde (PFA). Thereafter, the capsules of hematoma in different groups were collected and stored at -80° C for Western blotting, ELISAs, and PCR analysis at 5 d. For Evans blue injection and detection, the rats were infected with dye (2%, 1 ml/kg) through the right femoral vein and circulated for 30 min, and then euthanized under anesthesia. The neomembrane of the hematoma was dissected and fixed with 4% PFA. For Evans blue analysis, the tissues were incubated with formamide (60° C, overnight). The absorbance values of the extract at 620 nm were measured.

H&E staining

We performed as previous protocols [16]. At 5 days, the rats were killed and fixed with 4% paraformaldehyde. Brain tissue was collected and immediately frozen in - 80° C refrigerator for 10 minutes. Histologically, the part containing hematoma and dura was examined for 10 mm per 0.5 mm thickness.

qRT-PCR

qRT-PCR was performed as described in our previously published protocol [16]. The primers we used were as follows:

MMP-9: forward 5′-TTCAAGGACGGTCGGTATT-3′

reverse 5′-CTCTGAGCCTAGACCCAACTTA-3′;

VEGF: forward 5′-GAGCAGGAGCCGAAGCC-3′

reverse 5′-GAGCCCAGAAGTTGGACGA-3′;

KLF-2: forward 5′-TGCCGTCCTTTGCCACTTTCG-3′

reverse 5′- GCACGCTGTTTAGGTCCTCATCC-3′;

NF-κb: forward 5′- TGTGGTGGAGGACTTGCTGAGG -3′

reverse 5′- AGTGCTGCCTTGCTGTTCTTGAG-3′;

ACTB: forward 5′-GAAGTACCCCATTGAACACGG-3′

reverse 5′-TGGGTCATCTTTTCACGGTTG-3′

ZO-1: forward 5′-ATAAAGAGAAAGGTGAAACACTGCT-3′

reverse 5′-TCACAGTGTGGTAAGCGCAG-3′

ELISA and western blotting

We tested the expression of MMP-9, VEGF, IL-6, IL-8, and IL-10 by ELISA kits (R&D Systems, Minneapolis, MN, USA) according to the manufacturer’s instructions. We measured the absorbance at 450 nm via a plate reader (Molecular Device, Sunnyvale, CA, USA). Western blotting was performed as described in previous studies [39]. Anti-KLF-2 (1:1000, Abcam, ab17008), anti-NF-κb (1:1 000, Cell Signaling Technology, 8242s, Danvers, MA, USA) and anti-ZO-1 (1:1000, Cell Signaling Technology), were used as the primary antibodies. Thereafter, horseradish peroxidase (HRP)-conjugated secondary IgG was incubated with the membranes for 1 h at room temperature. An enhanced chemiluminescence (ECL) system (Millipore, Billerica, MA, USA) was used to expose the membranes. We tested the gray value by ImageJ software.

Corner turning score and forelimb-use asymmetry score

The corner turning score and forelimb-use asymmetry score were calculated before induction of the CSDH model and repeated on days 3, 5 in the different treatment groups post-injury. We experimented with the steps in a previous study [8]. We used a blinded method to evaluate the experimental treatment.

Statistical analysis

GraphPad Prism (IBM, USA) was used to analyze the results and they are presented as the mean ± SD. We performed one-way ANOVA to analyze data. P<0.05 was defined as significant.