MAP Kinase Pathways in Brain Endothelial Cells and Crosstalk with Pericytes and Astrocytes Mediate Contrast-Induced Blood–Brain Barrier Disruption

Neurointervention with contrast media (CM) has rapidly increased, but the impact of CM extravasation and the related side effects remain controversial. This study investigated the effect of CM on blood–brain barrier (BBB) integrity. We established in vitro BBB models using primary cultures of rat BBB-related cells. To assess the effects of CM on BBB functions, we evaluated transendothelial electrical resistance, permeability, and tight junction (TJ) protein expression using immunohistochemistry (IHC) and Western blotting. To investigate the mechanism of iopamidol-induced barrier dysfunction, the role of mitogen-activated protein (MAP) kinases in brain endothelial cells was examined. We assessed the effect of conditioned medium derived from astrocytes and pericytes under iopamidol treatment. Short-term iopamidol exposure on the luminal side induced transient, while on the abluminal side caused persistent BBB dysfunction. IHC and immunoblotting revealed CM decreased the expression of TJ proteins. Iopamidol-induced barrier dysfunction was improved via the regulation of MAP kinase pathways. Conditioned medium from CM-exposed pericytes or astrocytes lacks the ability to enhance barrier function. CM may cause BBB dysfunction. MAP kinase pathways in brain endothelial cells and the interactions of astrocytes and pericytes mediate iopamidol-induced barrier dysfunction. CM extravasation may have negative effects on clinical outcomes in patients.


Introduction
Intravascular injections of iodine contrast media (CM) are widely used in the field of radiology for diagnostic imaging, and CM has recently been utilized in endovascular interventions. In clinical practice, the intravascular administration concentration of iodine CM is often about 150-300 mgI/mL, and the CM is usually rapidly excreted in urine by renal glomerular filtration. Despite its widespread and increasing use in clinical settings, CM

Materials and Reagents
All reagents were purchased from Sigma-Aldrich (St. Louis, MO, USA) unless indicated otherwise. Iopamidol was purchased from FUJIFILM Wako Pure Chemical Corporation (Osaka, Japan). The stock solution of iopamidol (612.4 mg/mL, iodine concentration of 300 mgI/mL) was prepared by dissolving it in an RBEC medium, as indicated below. The stock solution was diluted in RBEC medium to prepare the working solutions indicated in each experiment. Mannitol (62 mM) was used as an osmolarity control. The osmotic pressure of 62 mM mannitol is equivalent to that of 30 mgI/mL iopamidol.
Rat cerebral astrocytes were obtained from neonatal Wister rats. Meninges were removed, and cortical pieces were mechanically dissociated in an astrocyte culture medium (DMEM supplemented with 10% FBS). Dissociated cells were seeded in cell culture flasks. To obtain type 1 astrocytes, flasks with confluent cultures were shaken at 37 • C overnight.
Rat cerebral pericytes were obtained via a prolonged, 2-week-long culture of isolated brain microvessel fragments consisting of pericytes and endothelial cells. The same preparations yielded primary RBECs after puromycin treatment. Pericyte survival and proliferation were favored by selective culture conditions using uncoated dishes and DMEM supplemented with 10% FBS and antibiotics. The culture medium was changed every 3 days. Astrocytes and pericytes were frozen in cryo-medium LABO Banker 2 (TOSC Japan Ltd., Tokyo, Japan), and stored at −80 • C until use.
The cell types used in the triple co-culture BBB model have been characterized in detail in our previous publications [15,18,20]. Brain endothelial cell cultures expressed endothelial characteristics [15] as well as BBB markers both at the gene and protein levels [15,21]. More than 90% of the cells in the astrocyte cultures stained positively for glial fibrillary acidic protein [15,18,20]. Pericyte cultures expressed NG-2 and alpha-smooth muscle actin [15,18,20]. The cell types have been regularly tested for these markers by immunocytochemistry.

Establishment of the In Vitro BBB Model
Two BBB models were used in this study: brain endothelial cells alone (E00) and endothelial cells in triple culture with astrocytes and pericytes (EPA). The E00 model was constructed using Millicell ® Hanging Cell Culture Inserts (0.4-µm pore size) in 24well culture plates. RBECs (1.0 × 10 5 cells/insert) were seeded on the upper side of the collagen-and fibronectin-coated membrane of the insert. To construct the EPA model, astrocytes (5.0 × 10 4 cells/well) were seeded on the bottom of the collagen-coated 24-well culture plate, and pericytes (1.0 × 10 4 cells/insert) were seeded on the bottom side of the collagen-coated membrane of the insert. Pericytes were allowed to adhere firmly overnight; then, the next day RBECs (1.0 × 10 5 cells/insert) were seeded on the inserts, as described previously [15]. The BBB models were maintained in RBEC medium supplemented with 500 nM hydrocortisone [22].

Transendothelial Electrical Resistance (TEER)
To evaluate the barrier integrity of cultured brain endothelial cells, TEER was measured using an EVOM resistance meter (World Precision Instruments, Sarasota, FL, USA) and RBECs cultured on Millicell ® inserts in 24-well plates. The resistance values of blank filters without cells (background resistance) were subtracted from those of filters with cells. The values are presented as Ω × cm 2 , and data indicate the change in TEER measured before and after treatment as compared to the control group.

Paracellular Permeability of Sodium Fluorescein (Na-F)
The flux of Na-F through the BBB models was determined, as previously described [15,[23][24][25][26]. Cell culture inserts were transferred to 24-well plates containing 0.9 mL of assay buffer (composition: 136 mM NaCl, 0.9 mM CaCl 2 , 0.5 mM MgCl 2 , 2.7 mM KCl, 1.5 mM KH 2 PO 4 , 10 mM NaH 2 PO 4 , 25 mM glucose, and 10 mM HEPES, in distilled water, sterile filtered, pH 7.4) in the basolateral compartment. In the inserts, the culture medium was replaced with 0.2 mL of buffer containing 10 µg/mL Na-F. Inserts were removed from the abluminal chamber after 30 min. The concentration of Na-F was determined using a Pharmaceutics 2021, 13, 1272 4 of 16 fluorescence multi-well plate reader (Wallac 1420 ARVO Multilabel Counter, MA, USA) with emission at 535 nm and excitation at 485 nm. The flux of Na-F through the BBB models was used as an index of paracellular transport [16,26]. The apparent permeability coefficient (cm/s) was calculated as previously described [27].

CM Treatment in the In Vitro BBB Models
To examine the effect of short-term treatment with CM, iopamidol (300 mgI/mL) was added to the luminal (upper) side of the E00 model. Following 30 min of incubation, CM was replaced with RBEC medium. Then, barrier function was monitored by measuring TEER for 24 h. Permeability of Na-F through the E00 models was measured at 6 h after replacing treatment with RBEC medium. To examine the dose-and time-dependent effects of iopamidol on the barrier function, iopamidol (0, 3, 15, and 30 mgI/mL) was added to the abluminal (lower) side of the E00 model. As the hyperosmolarity of Iopamidol may affect BBB function, mannitol (62 mM), which has the same osmolality as iopamidol (30 mgI/mL), was used as an osmotic control. Then, barrier function was monitored by measuring TEER for up to 72 h. To examine the effect of fibroblast growth factor-2 (FGF-2, ReproCELL Inc., city, Japan) on barrier dysfunction induced by iopamidol, iopamidol (30 mgI/mL) and/or FGF-2 (10 ng/mL) for 24 h was added to the luminal side of the E00 model. The effects of iopamidol on pericytes and astrocytes were examined by using the E00 and EPA models. Iopamidol (30 mgI/mL) was added to the abluminal side of these models.

Collection of Conditioned Medium and Treatment in the In Vitro BBB Models
To examine the effects of iopamidol on astrocytes and pericytes, the conditioned medium from iopamidol-exposed cells was collected. Astrocytes or pericytes were cultured in 6 cm culture dishes in the RBEC medium. A cell-free dish was used as the negative control. After reaching 90% confluency, cells were incubated with iopamidol (15 mg/mL) for 36 h. Following centrifugation (300× g, 5 min), the conditioned medium collected from the supernatant and fresh RBEC medium were mixed in a 1:1 ratio (50% conditioned medium) and stored at −20 • C until use. The conditioned medium was administrated to the luminal and abluminal sides of the E00 model.

Immunostaining
To observe changes in interendothelial junctions, cells were stained with primary antibodies against ZO-1 junctional associated molecule (33-9100, Invitrogen, Carlsbad, CA, USA) and claudin-5 tight junction protein (35-2500, Invitrogen). All primary antibodies were used at a dilution of 1:50. As a secondary antibody, Alexa Fluor 488-conjugated donkey anti-mouse immunoglobulin (Invitrogen) was used at a dilution of 1:200. After 24 h of treatment with or without iopamidol (30 mgI/mL) or mannitol (62 mM), cultured cells were washed in phosphate-buffered saline (PBS) and fixed in 3% paraformaldehyde in PBS for 10 min. Following permeabilization with 0.1% Triton-X 100 for 10 min, cells were blocked with 3% bovine serum albumin-PBS for 30 min. Then, cells were incubated with anti-ZO-1 and anti-claudin-5 antibodies overnight at 4 • C. The cells were rinsed with PBS and incubated with the secondary antibody for 45 min at 37 • C. Preparations were mounted in Gel Mount (Biomeda, Foster City, CA, USA), and staining was examined under an EVOS ® FL Cell Imaging System (Thermo Fisher Scientific, Eugene, OR, USA).

Statistical Analysis
All data are expressed as the mean ± standard error of the mean. Student's t-test was employed to perform two-group comparisons, whereas analysis of variance, followed by Dunnett's or Tukey-Kramer's post hoc analysis was used for multiple-group comparisons (GraphPad Prism 8.4; GraphPad Software, San Diego, CA, USA). A p-value of less than 0.05 was considered statistically significant. All experiments were repeated at least twice, and the number of parallel samples in each experiment was 3-4.

Effects of Short-Term Exposure to CM on Barrier Function
We examined the effect of an iodinated CM. Iopamidol is characterized as a highosmolar type and contains one benzene ring in the structure; therefore, it is called a monomer. In clinical settings, iopamidol is injected at concentrations of 200-370 mgI/mL. To examine whether short-term exposure of CM affects the barrier integrity, in vitro BBB model (E00 model) was treated with iopamidol (300 mgI/mL) for 30 min. After replacing the CM with a fresh culture medium, TEER was measured. Short-term exposure to iopamidol decreased TEER, compared to the control value at 3, 6, and 12 h time points. This reduction of TEER induced by iopamidol exposure was recovered at 24 h ( Figure 1a). Iopamidol also increased the paracellular flux of Na-F, a small water-soluble marker, across the RBEC monolayer at 6 h ( Figure 1b). These findings suggested that short-term exposure to iopamidol might increase endothelial permeability and lead to the leakage of CM into the brain parenchyma.
Pharmaceutics 2021, 13, x 6 of 16 Figure 1. Effects of short-term exposure to contrast media on barrier functions using in vitro bloodbrain barrier (BBB) models. Monolayer BBB models were exposed to iopamidol (300 mgI/mL) for 30 min. After replacing contrast media with RBEC medium, recovery of barrier function was evaluated by measuring transendothelial resistance (TEER) and the permeability of the BBB to sodium fluorescein (Na-F): (a) iopamidol decreased TEER at 3, 6, and 12 h. The reduction of TEER induced by iopamidol exposure was recovered at 24 h; (b) the permeability of the BBB to Na-F was examined at 6 h. Iopamidol increased the permeability of Na-F (n = 4, unpaired t-test, * p < 0.05, *** p < 0.001).
Although intravascularly injected CM is rapidly eliminated from the body, CM can affect BBB function for a prolonged period if it leaks into the brain parenchyma for an extended period of time. Therefore, we examined the dose and time dependency of the effects of iopamidol on barrier function. Iopamidol (0, 3, 15, 30 mgI/mL) was administered to the abluminal (lower) side of the monolayer BBB model. Iopamidol (15, 30 mgI/mL) induced a significant decrease of TEER at 6 h, which was sustained for up to 72 h ( Figure  2). Mannitol as osmotic control decreased TEER in the early phase (at 6 h), but the effect did not last for a long time ( Figure 2). These findings indicate that iopamidol might induce barrier dysfunction independently of osmolarity. Dose and time dependency of the effects of iopamidol treatment on barrier function in the in vitro blood-brain barrier (BBB) models. Monolayer BBB models were exposed to iopamidol (0, 3, 15, 30 mgI/mL) abluminally for up to 72 h. Mannitol (62 mM) was used as an osmolarity control for iopamidol (30 mgI/mL). Barrier integrity was assessed by transendothelial resistance (TEER) measurement. Iopamidol (15 and 30 mgI/mL) significantly decreased TEER at 6 h, and the effect was TEER (% of control) Figure 1. Effects of short-term exposure to contrast media on barrier functions using in vitro bloodbrain barrier (BBB) models. Monolayer BBB models were exposed to iopamidol (300 mgI/mL) for 30 min. After replacing contrast media with RBEC medium, recovery of barrier function was evaluated by measuring transendothelial resistance (TEER) and the permeability of the BBB to sodium fluorescein (Na-F): (a) iopamidol decreased TEER at 3, 6, and 12 h. The reduction of TEER induced by iopamidol exposure was recovered at 24 h; (b) the permeability of the BBB to Na-F was examined at 6 h. Iopamidol increased the permeability of Na-F (n = 4, unpaired t-test, * p < 0.05, *** p < 0.001).
Although intravascularly injected CM is rapidly eliminated from the body, CM can affect BBB function for a prolonged period if it leaks into the brain parenchyma for an extended period of time. Therefore, we examined the dose and time dependency of the effects of iopamidol on barrier function. Iopamidol (0, 3, 15, 30 mgI/mL) was administered to the abluminal (lower) side of the monolayer BBB model. Iopamidol (15, 30 mgI/mL) induced a significant decrease of TEER at 6 h, which was sustained for up to 72 h ( Figure 2). Mannitol as osmotic control decreased TEER in the early phase (at 6 h), but the effect did not last for a long time ( Figure 2). These findings indicate that iopamidol might induce barrier dysfunction independently of osmolarity.
Pharmaceutics 2021, 13, x 6 of 16 Figure 1. Effects of short-term exposure to contrast media on barrier functions using in vitro bloodbrain barrier (BBB) models. Monolayer BBB models were exposed to iopamidol (300 mgI/mL) for 30 min. After replacing contrast media with RBEC medium, recovery of barrier function was evaluated by measuring transendothelial resistance (TEER) and the permeability of the BBB to sodium fluorescein (Na-F): (a) iopamidol decreased TEER at 3, 6, and 12 h. The reduction of TEER induced by iopamidol exposure was recovered at 24 h; (b) the permeability of the BBB to Na-F was examined at 6 h. Iopamidol increased the permeability of Na-F (n = 4, unpaired t-test, * p < 0.05, *** p < 0.001).
Although intravascularly injected CM is rapidly eliminated from the body, CM can affect BBB function for a prolonged period if it leaks into the brain parenchyma for an extended period of time. Therefore, we examined the dose and time dependency of the effects of iopamidol on barrier function. Iopamidol (0, 3, 15, 30 mgI/mL) was administered to the abluminal (lower) side of the monolayer BBB model. Iopamidol (15, 30 mgI/mL) induced a significant decrease of TEER at 6 h, which was sustained for up to 72 h ( Figure  2). Mannitol as osmotic control decreased TEER in the early phase (at 6 h), but the effect did not last for a long time ( Figure 2). These findings indicate that iopamidol might induce barrier dysfunction independently of osmolarity. Dose and time dependency of the effects of iopamidol treatment on barrier function in the in vitro blood-brain barrier (BBB) models. Monolayer BBB models were exposed to iopamidol (0, 3, 15, 30 mgI/mL) abluminally for up to 72 h. Mannitol (62 mM) was used as an osmolarity control for iopamidol (30 mgI/mL). Barrier integrity was assessed by transendothelial resistance (TEER) measurement. Iopamidol (15 and 30 mgI/mL) significantly decreased TEER at 6 h, and the effect was TEER (% of control) Figure 2. Dose and time dependency of the effects of iopamidol treatment on barrier function in the in vitro blood-brain barrier (BBB) models. Monolayer BBB models were exposed to iopamidol (0, 3, 15, 30 mgI/mL) abluminally for up to 72 h. Mannitol (62 mM) was used as an osmolarity control for iopamidol (30 mgI/mL). Barrier integrity was assessed by transendothelial resistance (TEER) measurement. Iopamidol (15 and 30 mgI/mL) significantly decreased TEER at 6 h, and the effect was sustained for up to 72 h. TEER is expressed as a percent of the control value at each time point (n = 4, one-way ANOVA, p-values were calculated from Tukey's multiple comparison test, * p < 0.05, *** p < 0.001). CM, contrast media.

Effects of Iopamidol on Cell Viability and Expression of TJ Proteins in RBECs
Iopamidol dose-dependently decreased the viability of RBECs measured by CCK-8 assay, whereas mannitol had no effect (Figure 3a). We also investigated whether barrier dysfunction induced by iopamidol is related to the expression change of TJ proteins with immunostaining and Western blotting. Immunostaining revealed that iopamidol (30 mgI/mL) treatment disrupted the continuous, pericellular belt-like pattern of junctional proteins ZO-1 and claudin-5 (Figure 3b). In addition, iopamidol (30 mgI/mL) treatment decreased the expression of the claudin-5 protein, while the expression of occludin, ZO-1, and VE-cadherin TJ proteins remained unchanged (Figure 3c).
Pharmaceutics 2021, 13, x 7 of 16 sustained for up to 72 h. TEER is expressed as a percent of the control value at each time point (n = 4, one-way ANOVA, p-values were calculated from Tukey's multiple comparison test, * p < 0.05, *** p < 0.001). CM, contrast media.

Effects of Iopamidol on Cell Viability and Expression of TJ Proteins in RBECs
Iopamidol dose-dependently decreased the viability of RBECs measured by CCK-8 assay, whereas mannitol had no effect (Figure 3a). We also investigated whether barrier dysfunction induced by iopamidol is related to the expression change of TJ proteins with immunostaining and Western blotting. Immunostaining revealed that iopamidol (30 mgI/mL) treatment disrupted the continuous, pericellular belt-like pattern of junctional proteins ZO-1 and claudin-5 (Figure 3b). In addition, iopamidol (30 mgI/mL) treatment decreased the expression of the claudin-5 protein, while the expression of occludin, ZO-1, and VE-cadherin TJ proteins remained unchanged (Figure 3c).

Influence of Iopamidol on the Phosphorylation of MAP Kinases in RBECs
The MAP kinase pathway is activated by various stimuli, such as growth factors, cytokines, oxidative stress, and hyperosmotic stress. Activated MAP kinase affects cellular processes including gene expression, proliferation, and apoptosis. Previous reports indicated that iopamidol influenced the MAP kinase pathway in renal cells and tissues [28,29]. To investigate the possible mechanism of iopamidol-induced barrier dysfunction, expression levels of the total-and phosphorylated-MAP kinase were examined. Following treatment with iopamidol (30 mgI/mL) for 0, 3, and 6 h, phosphorylated MAP components (extracellular signal-regulated kinase 1 and 2 (ERK), c-Jun N-terminal kinase (JNK), p38 mitogen-activated protein kinase (p38)) were detected by Western blot. Iopamidol decreased phosphorylation of ERK and increased p38 phosphorylation in this study (Figure 4a), but there were no notable changes of MAP kinase expression in RBECs treated with mannitol having the same osmolarity (Figure 4b).

Influence of Iopamidol on the Phosphorylation of MAP Kinases in RBECs
The MAP kinase pathway is activated by various stimuli, such as growth factors, cytokines, oxidative stress, and hyperosmotic stress. Activated MAP kinase affects cellular processes including gene expression, proliferation, and apoptosis. Previous reports indicated that iopamidol influenced the MAP kinase pathway in renal cells and tissues [28,29]. To investigate the possible mechanism of iopamidol-induced barrier dysfunction, expression levels of the total-and phosphorylated-MAP kinase were examined. Following treatment with iopamidol (30 mgI/mL) for 0, 3, and 6 h, phosphorylated MAP components (extracellular signal-regulated kinase 1 and 2 (ERK), c-Jun N-terminal kinase (JNK), p38 mitogen-activated protein kinase (p38)) were detected by Western blot. Iopamidol decreased phosphorylation of ERK and increased p38 phosphorylation in this study ( Figure  4a), but there were no notable changes of MAP kinase expression in RBECs treated with mannitol having the same osmolarity (Figure 4b).

Effects of FGF-2 on Viability and Barrier Function in Iopamidol-Treated RBECs
Based on the change of MAP kinase activation induced by iopamidol treatment, we next examined the effects of the regulation of MAP kinase activation on BBB function. It is known that several growth factors regulate the MAP kinase pathway. We examined the effect of FGF-2 on ERK and p38 phosphorylation. FGF-2 (10 ng/mL) treatment for 6 h increased ERK phosphorylation and decreased p38 phosphorylation (Figure 5a). As FGF-2 has a counter-regulatory effect on MAP kinase activation as iopamidol, we examined the effect of FGF-2 on iopamidol-induced BBB dysfunction. To examine the effect of FGF-2 on cell viability in iopamidol-treated RBECs, the cells were treated with iopamidol (15, 30 mgI/mL), in combination with FGF-2 (0-10 ng/mL) luminally for 24 h. Cell viability decreased by Iopamidol was dose-dependently enhanced by FGF-2 treatment (Figure 5b). We also examined the effect of FGF-2 on barrier dysfunction induced by iopamidol. RBECs (E00 model) were treated with iopamidol (30 mgI/mL), with or without FGF-2 (10 ng/mL) for 24 h. FGF-2 partially improved the iopamidol-induced barrier disruption (Figure 5c,d). These findings suggested that iopamidol disrupted the barrier integrity of brain endothelial cells through the MAP kinase pathways.

Effects of Iopamidol on NVU Types
The properties of the BBB are induced and maintained by crosstalk between brain capillary endothelial cells and the neighboring cells along the brain capillary, such as astrocytes and pericytes. Extravasated iopamidol may affect these cells and brain capillary endothelial cells. We hypothesized that iopamidol affects astrocytes and/or pericytes leading to abnormal interaction among these three cell types, which may disturb BBB function. To compare the effect of iopamidol on barrier function between the endothelial monolayer model (E00) and the endothelial-pericyte-astrocyte co-culture model (EPA), the abluminal side of these BBB models were treated with iopamidol (30 mgI/mL) for 24 h. Iopamidol decreased TEER (Actual TEER values; E00 control: 255.2 ± 6.62 (Ω × cm 2 ), E00 iopamidol: 123.7 ± 9.92 (Ω × cm 2 ), EPA control: 439.7 ± 29.41 (Ω × cm 2 ), EPA iopamidol: 131.9 ± 9.26 (Ω × cm 2 )) and increased the permeability of the BBB to Na-F in both the E00 and EPA models. However, the size of the effect relative to the control was larger in the EPA model (Figure 6a,b). This finding suggested that iopamidol directly affects the activities of astrocytes and/or pericytes, thereby enhancing iopamidol-induced BBB damage. Next, we investigated the effect of iopamidol on the viability of astrocytes and pericytes. Iopamidol dose-dependently decreased the viability in both cell types (Figure 6c,d).

Effects of Iopamidol on the Astrocyte-or Pericyte-Induced Enhancement of Barrier Function in RBECs
It is known that astrocyte-and pericyte-derived humoral factors enhance the properties of the BBB [11]. As iopamidol influenced the viability of astrocytes and pericytes, humoral factors derived from iopamidol-exposed cells may be changed compared to untreated cells. Therefore, we investigated the effects of conditioned medium samples collected from iopamidol (15 mgI/mL)-treated or untreated (control) cells. The collected conditioned medium samples were diluted 1:1 with fresh RBEC medium (50% conditioned medium). Conditioned medium derived from astrocytes and pericytes under control conditions increased barrier functions. In contrast, conditioned medium derived from astrocytes and pericytes exposed to iopamidol did not enhance barrier function (Figure 7a,b). Western blotting revealed increased claudin-5 expression in the control groups, which is consistent with enhancement of barrier function as measured by TEER (Figure 7c). Figure 6. Effects of iopamidol on pericytes and astrocytes. Iopamidol (30 mgI/mL) was added to the abluminal side of the monolayer or co-culture blood-brain barrier (BBB) models for 24 h. Barrier function was evaluated by transendothelial electrical resistance (TEER) (a) and the permeability of the BBB to Na-F (b) (n = 3-4, unpaired t-test, * p < 0.05, ** p < 0.01, *** p < 0.001). To examine the effect of iopamidol on cell viability, astrocytes (c) and pericytes (d) cells were treated with iopamidol (0-30 mgI/mL) for 24 h. Iopamidol decreased the viability of astrocytes (n = 8) and pericytes (n = 4-8) (one-way ANOVA, p-values were calculated from Tukey's multiple comparison test, ** p < 0.01, *** p < 0.001).

Effects of Iopamidol on the Astrocyte-or Pericyte-Induced Enhancement of Barrier Function in RBECs
It is known that astrocyte-and pericyte-derived humoral factors enhance the properties of the BBB [11]. As iopamidol influenced the viability of astrocytes and pericytes, humoral factors derived from iopamidol-exposed cells may be changed compared to untreated cells. Therefore, we investigated the effects of conditioned medium samples collected from iopamidol (15 mgI/mL)-treated or untreated (control) cells. The collected conditioned medium samples were diluted 1:1 with fresh RBEC medium (50% conditioned medium). Conditioned medium derived from astrocytes and pericytes under control conditions increased barrier functions. In contrast, conditioned medium derived from astrocytes and pericytes exposed to iopamidol did not enhance barrier function (Figure 7a,b). Western blotting revealed increased claudin-5 expression in the control groups, which is consistent with enhancement of barrier function as measured by TEER (Figure 7c). Figure 7. Effects of iopamidol on the astrocyte-or pericyte-induced enhancement of barrier function in rat brain endothelial cells (RBECs). To collect the conditioned medium, cell-free medium, astrocytes, or pericytes were incubated with iopamidol (15 mg/mL) for 36 h. The collected conditioned medium was diluted by 50% with fresh RBEC medium and used to treat RBECs monolayer model for 24 h. Barrier function was evaluated by transendothelial electrical resistance (TEER) (a) and the permeability of the blood-brain barrier (BBB) model to Na-F (b). The ability of astrocytes or pericytes to strengthen BBB barrier function was diminished by exposure to iopamidol. (c) Effect of conditioned medium on claudin-5 expression as examined by Western blotting. The blots are representative images of the experiment. The graph represents the relative expression of claudin-5 (n = 4, two-way ANOVA, p-values were calculated from Dunnett's multiple comparison test, * p < 0.05, ** p < 0.01, *** p < 0.001, ns = no statistically significant differences). Figure 7. Effects of iopamidol on the astrocyte-or pericyte-induced enhancement of barrier function in rat brain endothelial cells (RBECs). To collect the conditioned medium, cell-free medium, astrocytes, or pericytes were incubated with iopamidol (15 mg/mL) for 36 h. The collected conditioned medium was diluted by 50% with fresh RBEC medium and used to treat RBECs monolayer model for 24 h. Barrier function was evaluated by transendothelial electrical resistance (TEER) (a) and the permeability of the blood-brain barrier (BBB) model to Na-F (b). The ability of astrocytes or pericytes to strengthen BBB barrier function was diminished by exposure to iopamidol. (c) Effect of conditioned medium on claudin-5 expression as examined by Western blotting. The blots are representative images of the experiment. The graph represents the relative expression of claudin-5 (n = 4, two-way ANOVA, p-values were calculated from Dunnett's multiple comparison test, * p < 0.05, ** p < 0.01, *** p < 0.001, ns = no statistically significant differences).

Discussion
In the present study, we investigated the effects of the contrast agent iopamidol on culture models of the BBB and individual cell types forming the BBB. The results revealed that short-term exposure to clinically relevant concentrations of iopamidol on the luminal side induced BBB dysfunction-a change that was reversible. This temporary BBB dysfunction might allow the leakage of CM into the brain parenchyma. The presence of low concentrations of iopamidol in the brain parenchyma (abluminal side) can lead to permanent BBB disruption. We found that iopamidol worsened BBB function via the MAP kinase pathways (ERK pathway downregulation and p38 pathway activation). In addition, iopamidol promoted BBB injury through direct effects on pericytes and astrocytes.
Iopamidol is a water-soluble compound that rarely crosses the BBB. After an intravascular injection, iopamidol is promptly excreted through the urine via renal function, and the concentration of CM in blood also tends to decrease rapidly. However, there are clinical situations in which brain endothelial cells are exposed to high concentrations of iopamidol because of delayed excretion associated with renal dysfunction, repeated intra-arterial injection (e.g., neurointervention), and cerebral circulatory insufficiency. Renal failure, prior stroke, and heart failure have been reported as risk factors for CIE [9], which is consistent with these situations. CM exposure temporarily impairs BBB function, allowing CM to pass through the BBB into the brain parenchyma. This condition is also clinically observed, and computed tomography after mechanical thrombectomy for acute ischemic stroke can reveal CM leakage that matches the CM injection area [30]. In addition, iopamidol damaged endothelial cells of the rat aorta in vivo and decreased human umbilical vein endothelial cell (HUVEC) and H5V cell viability in vitro measured by CCK8 assay [31]. Although there are some reports of CM-induced cytotoxicity in other organs (e.g., human vascular smooth muscle cells, leukocytes, mast cells [32][33][34]), there are no reports indicating that iopamidol damages brain endothelial cells and causes BBB dysfunction. The present findings suggest that short-term exposure to a high concentration of iopamidol causes temporary BBB injury. Iopamidol that leaked into the abluminal side (representing the side of brain parenchyma), even at a low concentration, decreased endothelial cell viability and TJ protein expression, leading to lasting BBB dysfunction. The effect of osmotic pressure is considered one of the drivers of CM-induced endothelial cell damage. However, our experiments revealed that the decrease in RBEC viability, TJ protein expression, and BBB function was independent of the osmotic pressure alone, although a temporary decrease in TEER was observed by mannitol used at the same osmolarity.
MAP kinases are serine-threonine kinases that mediate intracellular signaling associated with a variety of cellular activities, including cell proliferation, differentiation, survival, death, and transformation [35,36]. The main MAP kinase family consists of ERK, p38, and JNK pathways. The ERK pathway, which is the classical MAP kinase pathway, is activated by growth factors, and it acts mainly on cell proliferation. The p38 and JNK pathways, which are stress response MAP kinase pathways, are induced by environmental stress, osmotic shock, UV irradiation, oxidative stress, protein synthesis inhibitors, and pro-inflammatory cytokines, and they play a central role in inducing apoptosis and controlling immune responses and inflammation [35]. Previous reports indicated that CM-induced renal apoptosis may be mediated by MAP kinase signaling, especially the JNK/p38 MAP kinase pathway [28]. However, the mechanism by which iopamidol affects brain endothelial cells is unclear. It has also been reported that CM reduces proliferation and increases apoptosis in HUVECs [32]. Our results suggested that iopamidol activated the MAPK pathway, suppressed brain endothelial cell proliferation by downregulating the ERK pathway, and induced inflammation and apoptosis by activating the p38 pathway, resulting in decreased BBB function. No prior studies indicated that the p38 pathway mediated the iopamidol-induced BBB dysfunction. The MAP kinase pathway may also be a therapeutic target for contrast-induced BBB dysfunction.
Brain endothelial cells have dynamic interactions with other neighboring cells. Crosstalk between the cells of the NVU is crucial for the formation and maintenance of a functional BBB [11,15]. CM extravasation may affect these cells, as well as brain capillary endothelial cells. No previous studies to date described the relationships of CM with pericytes and/or astrocytes. Our experiments show that iopamidol can dose-dependently decrease the viability of both astrocytes and pericytes, and conditioned medium derived from astrocytes