Expression of Thrombin-Cleaved Osteopontin and Integrin α9 and β1 Signaling Pathway Molecules in Chronic Subdural Hematomas

Chronic subdural hematoma (CSDH) is considered to be an inammatory and angiogenic disease. Osteopontin is an extracellular matrix protein. Osteopontin is cleaved by thrombin, resulting in N-half osteopontin, which is more prominent in integrin signal transduction. We examined the expression of N-half osteopontin in the CSDH uid and the expression of integrin α9 and β1 and the downstream components of the angiogenic signaling pathways in the outer membrane of CSDHs. Twenty samples of CSDH uid and 8 samples of CSDH outer membrane were included. The concentrations of N-half osteopontin in the CSDH uid were measured using ELISA kits. The expression of integrin α9 and β1, vinculin, talin-1, focal adhesion kinase (FAK), paxillin, α-actin, Src and β-actin was examined by western blot analysis. The expression of integrin α9 and β1, FAK and paxillin was also examined by immunohistochemistry. We investigated whether CSDH uid could activate FAK in cultured endothelial cells in vitro. The concentration of N-half osteopontin in CSDH uid was signicantly higher than that in the serum. Western blot analysis revealed above-mentioned molecules. In addition, integrin α9 and β1, FAK and paxillin were localized in the endothelial cells of vessels within the CSDH outer membrane. FAK was signicantly phosphorylated immediately after treatment with CSDH uid. Our data suggest that N-half osteopontin in CSDH uid promotes neovascularization in endothelial cells through integrins α9 and β1. The N-half osteopontin and integrin signaling pathway might be a useful therapeutic target for treating the growth of refractory CSDH. kinase;


Introduction
Chronic subdural hematoma (CSDH) is a neovascularized and in ammatory disease. High concentrations of growth factors and in ammatory mediators in CSDH uids have been reported and are involved in angiogenesis within the outer membrane. Vascular endothelial growth factor (VEGF) and angiopoietin are some of these growth factors [1,2], while in ammatory mediators include interleukin-6 and high-mobility group box 1 (HMGB1) [3,4]. These factors in CSDH uid activate the nuclear factor κB (NF-κB) [5] and Ras/MEK/ERK pathways [6], phosphoinositide 3-kinase (PI3-kinase)/protein kinase B (Akt) pathway [7] and Janus kinase (JAK)/signal transducer and activator of transcription 3 (STAT3) signaling pathway [8] in endothelial cells, resulting in angiogenesis and inducing the growth of the CSDH outer membrane.
Osteopontin is an extracellular matrix (ECM) protein that was rst identi ed in osteoblasts. Osteopontin has been implicated as an important factor in bone remodeling. Osteopontin plays critical roles in physiological and pathological processes, including in ammation through integrin receptors. After cerebral ischemia, osteopontin plays a role in matrix remodeling, which renovates new matrix-cell interactions [9]. Osteopontin extends astrocyte process and repairs blood-brain barrier dysfunction after ischemic stroke [10]. Thrombin-cleaved osteopontin (N-half osteopontin) plays a more effective protection than full-length osteopontin after focal cerebral ischemia in mice [11] and is useful as a blood indicator of acute atherothrombotic cerebral ischemia [12]. These previous data suggested that N-half osteopontin is also closely involved in in ammatory central nervous diseases. Therefore, we explored whether N-half osteopontin exists in CSDH uid and investigated the expression of integrin and subsequent signaling pathway molecules in the outer membrane of CSDH using western blotting and immunohistological analyses.

Material And Methods
All chemicals were purchased from Sigma Chemical (St. Louis, MO) unless otherwise speci ed.

Patients
This study included twenty patients who underwent surgical trepanation surgery for CSDH at Aichi Medical University Hospital. These patients, thirteen men and seven women, ranged in age from 41 to 83 years (mean age of 65 years). The Certi ed Clinical Research Review Board of Aichi Medical University Hospital approved this study (17-H047). Informed consent was obtained from each patient or the patient's family.

Analysis of thrombin-cleaved osteopontin in CSDH uid
Fluids from 20 consecutive CSDHs were obtained during trepanation surgery. As a control, serum samples were obtained from 5 patients suffering from CSDH. After collection, all samples were immediately centrifuged, and the supernatant was stored at -80°C until analysis.
We measured the concentration of N-half osteopontin using enzyme-linked immunosorbent assay (ELISA) kits (IBL, Gunma, Japan) according to the manufacturer's instructions. The mean minimum detectable dose of these assays was 8.3 pg/mL for N-half osteopontin.
The homogenates were centrifuged at 12,000×g for 10 min at 4°C. The protein concentrations of the supernatants were separated using 7.5% sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis.
After washing away unbound antibodies, the membranes were incubated in secondary antibodies with horseradish peroxidase (Sigma) at a 1:3000 dilution for 30 min at room temperature. The reactions were developed with ECL Plus (GE Healthcare, Buckinghamshire, UK). Positive controls were RAW264.7 cell lysate (Cell Signaling Technology), rat liver lysate (BD Bioscience, Franklin Lakes, NJ) and A431 cell lysate (Santa Cruz Biotechnology, Dallas, TX).

Histological examinations
For the analysis of the cellular expression of integrin α9, integrin β1, FAK and paxillin, immunohistochemical staining was performed on samples from three patients at room temperature using the avidin-biotinylated peroxidase complex (ABC) technique. To preserve the outer membrane of the CSDH samples, the membranes were incubated in 10 mL of ice-cold 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) for 3 h and then embedded in para n.
In this study, 10-μm-thick sections were prepared on a microtome and mounted onto MAS-coated glass slides (Matsunami Glass, Kishiwada Japan). The sections were depara nized in xylene, immersed in decreasing concentrations of ethanol, and rehydrated in water. Endogenous peroxidase activity was blocked with 0.3% H 2 O 2 in 100% methanol for 20 min. All sections for immunostaining were processed for microwave-enhanced antigen retrieval. Slide-mounted sections immersed in 0.01 M sodium citrate buffer (pH 6.0) were placed for 15 min in a 700-W microwave oven at maximum power.
Nonspeci c immunoreactivity was blocked by incubation with goat or donkey serum for 30 min, depending on the primary antibody. The samples were treated with primary antibodies against integrin α9 (R&D Systems) at a dilution of 1:150, integrin β1 (Cell Signaling Technology) at a dilution of 1:100, paxillin (Gene Tex) at a dilution of 1:500 and FAK (Cell Signaling Technology) at a dilution of 1:500 over 2 nights at 4°C. After several rinses in PBS, the samples were incubated with secondary biotinylated antibodies (anti-goat IgG 1:200, anti-rabbit IgG 1:200; Santa Cruz Biotechnology) at room temperature for 2 h. After several more rinses in PBS, they were incubated with Vectastain ABC reagent (Vectastain ABC Kit; Vector Laboratories, Burlingame, CA) for 1 h. After several more rinses in PBS, the bound peroxidase was visualized by incubating the sections with a solution containing 0.05% 3,3'-diaminobenzidine tetrahydrochloride (Sigma Aldrich) and 0.01% H2O2 in 0.05 M Tris-HCl (pH 7.4) for 10 min. After several rinses in water, the immunostained sections were dehydrated and cover-slipped with Entellan new (Merck, Kenilworth, NJ).

Cultured vascular endothelial cells
Endothelial cells of mouse brain (b.End3) were obtained from the HPA Cultured Collection (London, United Kingdom). The endothelial cells were cultured in Dulbecco's modi ed Eagle's medium (Nissui, Tokyo, Japan) that contained 10% fetal bovine serum under the circumstances at 37°C and 5% CO 2 .

Effect of CSDH uid on FAK
We centrifuged CSDH uids to remove debris and collected together. The b.End3 cells were incubated with serum-containing media that contained CSDH uid obtained during trepanation surgery. The volumes of the media and CSDH uid were 7.5 mL and 2.5 mL per culture dish, respectively. Protein lysates were prepared from the harvested cells at 5 min, 15 min, and 60 min (n = 3 per group). We used b.End3 cells treated with media alone as the control (n = 3). Total cell lysates were subjected to western blotting analysis using antibodies against p-FAK at Tyr 397 (Thermo Fisher Scienti c, Tokyo, Japan), FAK (Cell Signaling Technology) and β-actin (Sigma) as discussed above. All antibodies were used at a 1:750 dilution.The band intensities were quantitated using densitometry with the ImageQuant software (GE Healthcare).

Statistical analysis
Data are expressed as the mean ± standard error. The Mann-Whitney U test was used for the analysis between two groups. Statistical analyses were performed by one-way analysis of variance (ANOVA) followed by the Fisher's post hoc test, as appropriate. Signi cance was indicated when p < 0.05.

Concentration of N-half osteopontin in CSDH uid
The concentration of N-half osteopontin in CSDH uid (29,451.5 ± 8,146.5 pg/ml) was signi cantly higher than that in serum (365.8 ± 73.2 pmol/L) according to the Mann-Whitney U test (p < 0.01, Fig. 1).
Western blotting analysis of integrins and the angiogenic signaling pathway however, in some cases, the signals were weak. Positive controls revealed that these molecules had been correctly detected. and H). In the negative controls examined without primary antibodies, the endothelial cells were consistently negative for the markers listed above (Fig. 3I).

Activation of FAK in endothelial cells by CSDH uid
To investigate the role of CSDH uid, we examined phosphorylation of FAK in endothelial cells following treatment with subdural hematoma uid (Fig. 4). The addition of CSDH uid to cultured vascular endothelial cells resulted in signi cantly higher phosphorylated FAK levels compared with the controls 5 min after treatment (p < 0.05). In contrast, the FAK and b-actin levels were constant.

Discussion
The expression of N-half osteopontin in CSDH uid was signi cantly higher than that in serum.
Molecules of integrins and the subsequent angiogenic pathway intermediates were detected in the outer membrane of CSDH. Integrin α9 and β1, FAK and paxillin were expressed in the endothelium of vessels of the CSDH outer membrane. CSDH uid activated FAK in the endothelial cells immediately after treatment.
The extracellular matrix protein osteopontin is a glycoprotein and is involved in physiological and pathological events during in ammatory processes. The concentrations of thrombin-cleaved osteopontin in synovial uid correlate well with the severity of knee osteoarthritis [13]. Compared with full-length osteopontin, N-half osteopontin induced markedly greater cell attachment through integrin receptors [14]. Disease activity in lupus nephritis correlates with urine N-half osteopontin instead of full-length osteopontin, suggesting that N-half osteopontin is an indicator of in ammation of the kidney [15]. A previous study revealed that excessive coagulation, generation of thrombin and increased brinolysis occur within CSDH uids [16]. Considering previous studies and our data together, osteopontin is cleaved by thrombin within CSDH uids, and N-half osteopontin plays a role in in ammatory reactions in CSDH outer membranes. To the best of our knowledge, this study is the rst to demonstrate the existence of Nhalf osteopontin in CSDH uids.
Integrins are α/β heterodimeric cell surface receptors that mediate cell-cell and cell-ECM interactions and orchestrate cell attachment, movement, growth, differentiation and survival. Integrin α9 is widely expressed in a variety of cell types, including epithelium, in vivo [17]. Integrin β1 is the main β subunit for α9 in these cells [17]. The β1 class of integrins takes part in many aspects of vascular biology, especially angiogenesis [18]. β1 integrins play an important role in endothelial cell adhesion, migration, survival during angiogenesis and vascular remodeling [19,20]. A de cit of endothelial β1 integrins prohibited endothelial cell maturation, migration and sprouting and induced endothelial cell apoptosis [21]. Thrombin-cleaved osteopontin can attach to the integrin α9 and β1 via the sequence SVVYGLR, which is located between the arginine-glycine-aspartic acid (RGD) sequence and the thrombin cleavage site [22]. From our data, there might be a possibility that thrombin-cleaved osteopontin, i.e., N-half osteopontin, induces angiogenesis through this integrin α9 and β1 in the endothelium of the outer membrane.
After these integrin receptors combine with extracellular matrix, the formation of complex multiprotein structures occurs. FAK is a regulator of signals from the ECM to the cytoplasmic actin cytoskeleton (Fig. 5) [23]. Angiogenesis is mandatory for tumor development. FAK participates in endothelial cell proliferation, which has been revealed to control tumor angiogenesis in many cancers [24] and promotes angiogenesis in overexpressed transgenic mice [25]. FAK induces tumor angiogenesis in a dosedependent manner [26]. Both FAK and Src form a dual kinase complex, which play an important role in promoting VEGF-associated tumor angiogenesis [27]. α-Actin is a highly conserved protein and a member of the actin cross-linking protein family. α-Actin is phosphorylated on tyrosine residue by FAK and binds to actin [28]. These molecules regulate the ow of signals from the extracellular matrix to the actin cytoskeleton and induce angiogenesis (Fig. 5).
Paxillin is an adaptor protein located at the interface between the actin cytoskeleton and the plasma membrane [29] and is one of the key components of integrin signaling (Fig. 5). The FAK/Src complex phosphorylates tyrosine and serine residues of paxillin and promotes cell migration and regulates adhesion turnover at the cell front through paxillin [30]. Netrin-1 is a laminin-like secreted protein that is thought to be an axon guidance molecule during neural development. Netrin-1 activates the FAK/Src/paxillin pathway and modulates angiogenesis, which is accompanied by the upregulation of VEGF [31]. Both talin and vinculin also play an important role in cell growth, morphogenesis, and cell migration during the development. Marked defects in focal adhesions and embryonic death occur in case of loss of either talin or vinculin in mice [32,33]. Talin is also the key regulator of the link between the cytoskeleton and integrins, having multiple interaction sites for other adhesome components (Fig. 5) [34]. Talin-1 is essential for endothelial proliferation and postnatal angiogenesis [35]. Furthermore, vinculin is a key regulator of cell adhesion by direct interactions with talin and actin (Fig. 5) [36, 37]. Our data revealed that all these molecules were con rmed by western blot analysis and were located in the endothelium of CSDH outer membranes by immunohistochemistry. Moreover, FAK in the endothelial cells was activated by CSDH uid. N-half osteopontin activated by thrombin signals through integrin α9 and β1 located on the cell surface to the actin cytoskeleton and induces angiogenesis within the CSDH outer membrane.
Angiogenesis is a complex process regulated by numerous receptors, growth factors, ECM-cell interactions and so on. The concentration of VEGF in hematoma uid has been reported to be a highly important mechanistic factor in the pathophysiological progression and growth of CSDH and angiogenesis [2,38]. The angiogenic effect of VEGF depends on the presence of integrinβ1 [39]. FAK and Src coordinate together for angiogenesis induced by VEGF in vitro [40]. We have revealed that activation of mitogen-activated protein kinases (MAPKs) occurs in CSDH outer membranes by VEGF and plays a critical role in the angiogenesis of CSDHs [41,6]. Phosphorylated c-Jun N-terminal kinase (JNK) is expressed in the vascular endothelium of the CSDH outer membrane. FAK activates JNK through an extraordinary mechanism involving the recruitment of paxillin to the plasma membrane [42]. Taken together, we have to consider targeted therapy against not only growth factors but also the osteopontin/integrin pathway (Fig. 5).
In the present study, we have to remark several limitations. First, from our limited number of patients, we could not detect a correlation between the concentrations of N-half osteopontin in CSDH uids, the data from western blot analyses and the growth stage of CSDH. Further studies, including more patients, will be necessary to clarify this point. Second, we only found out the presence of integrin α9 and β1 and subsequent angiogenic signaling molecules in the outer membrane of CSDHs. We have to nd out these signaling molecules are activated during the development of CSDH.
In the present study, we detected the expression of N-half osteopontin in CSDH uids and integrin α9 and β1, FAK, paxillin, vinculin and the subsequent angiogenic signaling pathway in the CSDH outer membrane for the rst time. Signi cantly high concentrations of N-half osteopontin in CSDH uid might play an important role in angiogenesis and in ammation in CSDH, resulting in the growth of the hematoma. This angiogenic signaling pathway through integrin α9 and β1 might be a therapeutically alternative target for the treatment of refractory CSDH.

Declarations Acknowledgements
None.

Authors' Contribution
KO and YO equally designed the study with signi cant instruction from YW and SH. MO, CS, MA and KI contributed to the data collection and elaboration of the article. CS and YW contributed valuable contributions to graphical designs. All authors reviewed and editied the manuscript and are responsible for its content. The corresponding author attests that all listed authors meet authorship criteria and that no others meeting the criteria have been omitted.

Funding
This work was supported in part by a Japanese Grant-in-Aid for Scienti c Research (C), Grant Number 17K10853 (KO).

Data Availability
All data generated during this study are included in this published article.

Ethics Approval
Approval was obtained from the ethics committee of Aichi Medical University (17-H047).

Con ict of interest
The authors have no personal nancial or institutional interest in any of the drugs, materials, or devices described in this article.

Consent to participate
Written informed consent was obtained from the patients or the patient's families. Figure 1 Concentrations of thrombin-cleaved osteopontin (N-half osteopontin) in serum (serum, n = 5) and chronic subdural hematoma (CSDH, n = 20). The concentration of N-half osteopontin in CSDH uid was signi cantly higher than that in serum by the Mann-Whitney U-test. Data represent the median values and 25 th and 75 th percentiles with maximum/minimum whiskers. *p < 0.05 by the Mann-Whitney U-test.

Figure 2
Western blots showing the expression of integrin α9 and β1 and the subsequent angiogenic pathway molecules in the outer membrane of chronic subdural hematomas from eight patients. Integrins β1 and α9, vinculin, talin-1, focal adhesion kinase (FAK), paxillin, a-actinin and Src were detected in almost all cases. Positive controls are shown in the right three lanes and suggest that these molecules were correctly detected. RAW 264.7, murine leukemia macrophage cell line lysate; Rat liver, rat liver whole cell lysate; A431 cell lysate, epidermoid carcinoma cell lysate. Ten-micrometer consecutive slices were immunostained with polyclonal antibodies against integrin α9 (A and B), integrin β1 (C and D) focal adhesion kinase (FAK, E and F) and paxillin (G and H) using the ABC method. The areas within the rectangle, labeled A, C, E and G, are shown at a higher magni cation in panels B, D, F and H, respectively. Note that these molecules were expressed in endothelial cells (B, D, F and H). Slices immunostained without primary antibodies are shown in (I). Scale bars = 100 µm.  The mean ± SE values from the data of 3 series are shown. *p < 0.05 vs the control (1-way ANOVA followed by Fisher's PLSD)