ISCHEMIC TOLERANCE IS ASSOCIATED WITH VEGF-C AND VEGFR-3 SIGNALING IN THE MOUSE HIPPOCAMPUS
M. I. H. BHUIYAN, a,bti J.-C. KIM, a,b S.-N. HWANG, a,b M.-Y. LEE b,c AND S. Y. KIM a,b*
aDepartment of Pharmacology, Cell Death Disease Research Center, College of Medicine, The Catholic University of Korea, 505 Banpo-dong, Socho-gu, Seoul 137-701, South Korea
bCatholic Neuroscience Institute, College of Medicine, The Catholic University of Korea, 505 Banpo-dong, Socho-gu, Seoul 137-701, South Korea
cDepartment of Anatomy, College of Medicine, The Catholic University of Korea, 505 Banpo-dong, Socho-gu, Seoul
137-701, South Korea

Abstract—The functions of vascular endothelial growth fac- tor C (VEGF-C) and the VEGF receptor 3 (VEGFR-3) in the nervous system are not well known. In this study, we exam- ined the role of VEGF-C and VEGFR-3 in ischemic precondi- tioning (IPC)-induced tolerance in the mouse hippocampus. Adult male C57BL/6 mice were subjected to either severe ischemia (SI) induced by 40 min of bilateral common carotid artery occlusion (BCCAO) with or without IPC (5-min BCCAO) or IPC only. Cerebral blood flow was measured dur- ing ischemic periods using laser Doppler flowmetry. Neuro- nal damage was assessed histologically, and VEGF-C and VEGFR-3 expression levels were assessed through immu- nostaining. Fluoro-Jade B-labeled cells were abundant in the CA1 area 7 days after SI without IPC (sham + SI group), whereas cells were rarely labeled in mice subjected to IPC followed by SI (IPC + SI group). Similarly, the number of neuronal nuclei (NeuN)-positive cells in the CA1 area was significantly lower in the sham + SI group than in the IPC + SI group. Interestingly, we found that sublethal IPC treatment induced prominent VEGF-C expression in the CA1 pyramidal neurons and VEGFR-3 expression in the stra- tum radiatum and stratum lacunosum moleculare after 3 days of reperfusion that were sustained for 7 days. More- over, VEGF-C immunoreactivity was also markedly increased, whereas VEGFR-3 expression was sustained in tolerance-acquired CA1 neurons after SI. Application of a VEGFR-3 inhibitor, SAR131675, abolished the IPC-induced neuroprotection in a dose-dependent manner in the mouse
hippocampus. These results suggest that VEGF-C/VEGFR- 3 signaling is associated with IPC-induced hippocampal tolerance to lethal ischemia. ti 2015 IBRO. Published by Elsevier Ltd. All rights reserved.
Key words: cerebral ischemia, ischemic preconditioning, VEGF-C, VEGFR-3, ischemic tolerance.
INTRODUCTION
Ischemic stress induces both harmful and protective mechanisms in the cell (Dirnagl et al., 2003). If the stress is below the damage threshold, protective mechanisms prevail. When another stress is applied at the peak of the stress-controlled period, cells have improved toler- ance. This phenomenon is known as ischemic precondi- tioning (IPC), in which brief episode(s) of ischemia induce an adaptive response (tolerance) to subsequent lethal ischemia (Murry et al., 1986; Kitagawa et al., 1991). This phenomenon has been applied in many inves- tigations to identify the molecular mediators responsible for cardioprotection and neuroprotection (Obrenovitch, 2008; Thompson et al., 2013). It has been reported that heat shock proteins, transcription factors, cytokines and protein growth factors including vascular endothelial growth factor (VEGF) are involved in the mechanisms of ischemic tolerance in the brain (Schmidt-Kastner et al., 2004; Obrenovitch, 2008; Bhuiyan and Kim, 2010). How- ever, the molecular mechanisms underlying IPC-induced neuroprotection in the brain require further elucidation.
The VEGF family of ligands consists of seven different homologous members: VEGF-A, -B, -C, -D, -E, -F, and placental growth factor (Tammela et al., 2005; Yamazaki and Morita, 2006). These ligands signal primar- ily through three high-affinity transmembrane tyrosine kinase receptors, namely, the VEGF receptor 1 (VEG- FR-1), VEGFR-2 and VEGFR-3 (Tammela et al., 2005;

*Correspondence to: S. Y. Kim, Department of Pharmacology, College of Medicine, The Catholic University of Korea, 505 Banpo- dong, Socho-gu, Seoul 137-701, South Korea. Tel: +82-2-2258- 7324; fax: +82-2-536-2485.
E-mail address: [email protected] (S. Y. Kim).
ti Currentt address: Department of Neurology, University of Pittsburgh, PA 15213, USA.
Abbreviations: BCCAO, bilateral common carotid artery occlusion; FJ, Fluoro-Jade B; GFAP, glial fibrillary acidic protein; IPC, ischemic preconditioning; LDF, laser Doppler flowmetry; NeuN, neuronal nuclei; PB, phosphate buffer; PBS, phosphate-buffered saline; rCBF, regional cerebral blood flow; SI, severe ischemia; VEGF-C, vascular endothelial growth factor C; VEGFR-3, VEGF receptor 3.

http://dx.doi.org/10.1016/j.neuroscience.2015.01.025

0306-4522/ti 2015 IBRO. Published by Elsevier Ltd. All rights reserved.
Yamazaki and Morita, 2006). VEGF-A was originally iden- tified as an angiogenic and vascular permeability factor during development and in adult physiology and pathol- ogy, whereas VEGF-C and VEGF-D are mainly lymphan- giogenic factors, inducing the growth of lymphatic vessels during development as well as in the adult (Yamazaki and Morita, 2006; Ny et al., 2008). VEGF-A and VEGF-B are also known to exhibit direct neurotrophic and neuropro- tective effects, particularly against experimentally induced ischemic injury (Greenberg and Jin, 2013). Although VEGF-C has a high degree of sequence and structural
90
homology with VEGF-A and VEGF-B, the role of VEGF-C in the nervous system is not well known. It has been reported that VEGF-C can play a neurotrophic role for neural progenitor cells in the embryonic brain (Le Bras et al., 2006). Recently, our group showed that the expres- sion of VEGF-C and VEGFR-3 is regulated in the adult rat brain after global and focal cerebral ischemia (Shin et al., 2008, 2010a), although the functional implications of this regulation remain unknown. To address whether VEGF-C/VEGFR-3 signaling is involved in the endoge- nous adaptive response of the brain to ischemic injury, we examined the expression of VEGF-C and VEGFR-3 in a mouse model of ischemic tolerance. In addition, using a pharmacological inhibitor of VEGFR-3, we investigated the functional involvement of VEGFR-3 activity in an ischemic tolerance paradigm. Our results show that the IPC-induced VEGF-C mediates ischemic tolerance in the mouse hippocampus, likely through VEGFR-3.

EXPERIMENTAL PROCEDURES
Animals
All animal procedures were approved by the Ethics Committee of the Catholic University of Korea and were performed in accordance with the National Institutes of Health’s Guide for the Care and Use of Laboratory Animals (NIH Publication No. 80–23). Male C57BL/6 mice (Koatec, Kyunggi-do, Korea), weighing 20–25 g (9–11 weeks old), were kept in cages under light- controlled conditions (lights on from 08:00 to 20:00) with free access to food and water. Total of 160 mice were used in this study.

Induction of global forebrain ischemia and measurement of regional cerebral blood flow (rCBF)
Forebrain ischemia was induced through bilateral common carotid artery occlusion (BCCAO), according to our previously established method (Cho et al., 2006, 2007) with minor modifications. Briefly, anesthesia was induced with 2% isoflurane and maintained with 1% isoflu- rane in a 70% nitrous oxide and 30% oxygen mixture using a face-mask. To measure rCBF, laser Doppler flow- metry (LDF; PF5010, Perimed, Ja¨rfa¨lla, Sweden) was used from anesthetic induction through 10 min after reperfusion. For LDF, two flexible probes (407) were attached to the intact skull, 3.5 mm right and 3.5 mm left of the bregma with cyanoacrylate adhesives, and rCBF was continuously recorded using a computer-based data acquisition system (Perisoft, Perimed). Both common car- otid arteries were exposed through a midline incision of the neck, loosely encircled with 5-0 surgical thread and temporarily clamped with micro-serrefine. In groups with sham-operations, the common carotid arteries were not occluded but were isolated from the adjacent vagus nerves. The change in rCBF was calculated as a percent- age of the value measured for one min immediately after BCCAO over the baseline value. After reperfusion, the wounds were closed, and the animals were placed in a warm incubator (32–33 tiC) until they were euthanized. During the surgical procedure, the rectal temperatures

of the mice were maintained at 37.0 ± 0.5 ti C with a heat- ing pad. Nine groups were studied, as shown in Fig. 1A: sham-operation without 5-min BCCAO (sham group; n = 15); IPC with 5-min BCCAO (IPC group; n = 24); sham operation + severe ischemia (SI) with 40-min BCCAO (sham + SI group; n = 10); IPC followed by SI (IPC + SI group; n = 13); sham-IPC + SI group treated with vehicle (Veh) (sham-IPC + SI + Veh group; n = 11); IPC + SI group treated with vehicle (IPC + SI + Veh group; n = 16); IPC + SI groups treated with SAR131675 of a low-dose (IPC + SI + SARL group; n = 7) or a high-dose (IPC + SI + SARH group; n = 9); sham-operation treated with SAR131675 of a high-dose (sham + SARH group; n = 5).

Drug administration
The VEGFR-3 inhibitor SAR131675 (Selleck Chemicals, Munich, Germany) was dissolved in DMSO and then diluted in normal saline. On the first day, SAR131675 was intraperitoneally administered at a dose of 25 mg/kg for the IPC + SI + SARL group or 50 mg/kg for the IPC + SI + SARH group as an initial dose 30 min prior to the induction of SI. Beginning the next day, the mice were administered SAR131675 of 12.5 mg/kg/day for the IPC + SI + SARL group or 25 mg/kg/day for the IPC + SI + SARH group for 6 days. For the vehicle- treated group, the same concentration of DMSO in normal saline was administered in the same manner.

Histology
Seven days after SI, the mice were perfused intracardially under anesthesia with normal saline, followed by 4% paraformaldehyde in 0.1 M phosphate buffer (PB) for 30 min under anesthesia. The brains were rapidly removed and further fixed with the same solution at 4 tiC overnight. Post-fixed brains were immersed in 30% sucrose until they settled to the bottom of the container. Using Tissue-Tek (Sakura Finetechnical, Tokyo, Japan), the brains were quickly deep-frozen and then cut into 25-lm thick slices using a cryostat. Fluoro-Jade B (FJ) staining was used to label degenerating cells (Lee et al., 2008; Liu et al., 2009). Briefly, coronal sections were wet mounted and then air-dried on microscope slides. The slides were immersed in 70% ethanol, washed with distilled water, and treated with 0.06% potassium perman- ganate solution for 10 min. Next, the sections were stained with 0.001% FJ (Chemicon, Jefferson, AR, USA) in 0.1% acetic acid for 30 min at room temperature with gentle shaking and then washed with distilled water and dried for 1 h in the dark. The slides were cleared in xylene and coverslipped with DPX mounting medium for histology (Sigma Aldrich, St. Louis, MO, USA).

Immunofluorescence analysis
Immunofluorescence assays were conducted according to our previously published methods (Cho et al., 2007; Jeong et al., 2011) with minor modifications. In brief, brain sections were washed with phosphate-buffered saline (PBS, pH 7.4), permeabilized with 0.15% Triton X-100

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
Fig. 1. A summary of the ischemic preconditioning (IPC) and ischemic tolerance experimental protocols in the C57BL/6 mouse (A). IPC was induced through a 5-min BCCAO. Three days later, severe ischemia was induced by a 40-min BCCAO, and 7 days later, outcome was assessed. (B) Representative immunofluorescence images showing that a 5-min BCCAO induces the prominent expression of heat shock protein 70 in the CA1 pyramidal neurons after 3 days of reperfusion. Scale bar = 20 lm.

 
for 15 min and washed with PBS (5 min ti 3). After block- ing with 10% normal goat serum (Vector Laboratories, Burlingame, CA, USA) for 1 h at room temperature, the sections were incubated with the appropriate primary anti- bodies glial fibrillary acidic protein (GFAP, 1:400, Millipore; Iba-1, 1:250, Wako; neuronal nuclei (NeuN), 1:300, Millipore; OX-42, 1:150, Millipore; VEGF-C, 1:200, Santa Cruz; VEGFR-3, 1:150, Abnova) overnight at 4 ti C. After several rinses in PBS, the tissues were
incubated with the appropriate secondary antibodies for
2h at room temperature. The primary and secondary antibodies were diluted in 3% normal goat serum. To assess nuclear morphology, the tissues were incubated with 0.5 lM of 40 ,6-diamidino-2-phenylindole (DAPI; Molecular Probes, Eugene, OR, USA) in PB for 10 min. After a final wash in PB (5 min ti 3), glass coverslips were fixed on slides with ProLongti Gold fluorescent mounting medium (Molecular Probes).
Immunohistochemistry
Immunohistochemistry assays were conducted according to our previously published methods (Cho et al., 2007; Jeong et al., 2011) with minor modifications. In brief, brain sections were washed with PBS and incubated with 3% H2O2 and 10% methanol in PBS to destroy endogenous peroxidase activity. After blocking with 10% normal goat serum for 1 h at room temperature, the sections were incubated with anti-VEGF-C primary antibody overnight at 4 ti C. After several rinses in PBS, the tissues were incu- bated with biotinylated secondary antibodies for 2 h at room temperature. The sections were then washed in PBS and incubated in an Avidin–Biotin complex solution (Vectastain, Vector Labs; Burlingame, CA, USA). After washing in PBS, the sections were developed using 0.1% diaminobenzidine tetrahydrochloride and 0.02% H2O2. Finally, the tissues were mounted on slides, air- dried and coverslipped using Canada balsam mounting solution.

Cell counting and quantitative analysis
Cell counting was performed according to our previously established method (Jeong et al., 2011) with minor mod- ifications. Quantifications of VEGF-C and VEGFR-3 expression were determined using ImageJ software (NIH). To analyze NeuN-, VEGF-C-, VEGFR-3- and FJ-positive cells in the hippocampus, three coronal sections, sliced at 100-lm intervals, were obtained from the regions between ti 1.46 and ti1.94 mm from the bregma (Franklin and Paxinos, 2008). For NeuN- and FJ-labeled cells, the medial CA1 region of each section was captured at 400ti magnification using a fluorescence microscope (AxioImager, Carl Zeiss, Jena, Germany). For VEGFR- 3-positive cell counting in the hippocampus, an area cov- ering the stratum radiatum and the lacunosum of the CA1 subfield together with the molecular layer of the dentate gyrus (shown in the box of Fig. 4B) was captured at 100x magnification using an LSM 510 META laser-scan- ning confocal microscope (Carl Zeiss). The total number of labeled cells in each section was counted using the LSM Image Examiner software (Carl Zeiss). The average value for the labeled cells in three sections was calculated for each animal before the means and standard errors were determined for all animals in each group.

Statistical analyses
Data are expressed as the mean ± SEM. All statistical analyses were performed using SPSS 13.0 (SPSS Inc., Chicago, IL, USA). Statistical significance was assessed with a one-way analysis of variance for multiple group comparisons followed by Tukey’s post hoc test. A p-value <0.05 was considered significant.

RESULTS
In vivo IPC model
Because the hippocampal CA1 region is vulnerable to ischemia, we examined this region for different outcome measures (Lee et al., 2008). In previous studies, we found

that 35–40 min of BCCAO consistently produces CA1 neuronal cell death when the rCBF decreases to less than 15.0% of the baseline (Cho et al., 2006, 2007; Lee et al., 2008). Therefore, as a strategy to reduce the variation in neuronal damage, we included only those mice with rCBF decreases to less than 12.0% of the baseline immediately after BCCAO, and we defined a 40-min BCCAO as SI. With this criterion, totally 29 mice (18.1%) were excluded due to insufficient CBF reduction. In addition, although there are many different kinds of protocols for IPC, we found that a 5-min BCCAO and 3 days of reperfusion prominently induced the expression of heat shock protein 70 in hippocampal CA1 pyramidal neurons (Fig. 1B) with- out observable neuronal injury (Liu et al., 1993; Munoz et al., 2003; Wu et al., 2003; Cho et al., 2005; Yoshioka et al., 2011). Therefore, a 5-min BCCAO was chosen as the IPC stimulus in this study. No animals died in the sham-operation group or in the 5-min BCCAO group within 7 days of these procedures in this study. The mor- tality rates within 7 days after BCCAO in the sham + SI and IPC + SI groups were 37.5% and 13.3%, respectively.
To assess IPC-induced neuroprotection in the CA1 region 7 days after SI, FJ staining and NeuN (a neuron- specific marker) immunofluorescence were performed. We found no FJ-positive staining for the pyramidal neurons in the CA1 region of the sham and IPC groups (Fig. 2A). However, extensive FJ-positive degenerating neurons were observed in the sham + SI group. Interestingly, IPC performed 3 days before SI inhibited FJ staining in the CA1 region. A quantitative analysis showed that the number of FJ-positive cells in the sham + SI group was 94.8 ± 12.6 (per field), whereas the number was only 28.3 ± 7.2 (per field) in the IPC + SI group, which was significantly lower than that in the sham + SI group (Fig. 2B). An analysis of the NeuN immunofluorescence showed that CA1 pyramidal neurons in sham-operated and IPC-treated animals formed a layer of 3–5 cells wide (Fig. 2C), with no reduction in the level of NeuN-immunoreactivity compared with non-treated mice (data not shown). A 40-min BCCAO caused a marked loss of NeuN immunoreactivity in the CA1 region assessed after 7 days of reperfusion. Interestingly, IPC performed
3days before SI clearly prevented the loss of NeuN immunoreactivity in the CA1 pyramidal neurons (Fig. 2C). A quantitative analysis showed 31.7 ± 7.8 NeuN-positive cells (per field) in the sham + SI group, whereas the number of NeuN-positive cells in the IPC + SI group was significantly higher at 82.8 ± 12.3 (per field; Fig. 2D).

Effects of IPC on VEGF-C and VEGFR-3 expression in the hippocampus
To assess whether the expression of VEGF-C and VEGFR-3 is regulated in the IPC paradigm, we examined immunofluorescence for VEGF-C and its receptor VEGFR-3 in the mouse hippocampus 1–14 days after IPC. We found that IPC treatment induced VEGF-C immunofluorescence in the CA1 region after 3 and 7 days of reperfusion (Fig. 3A).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
Fig. 2. Histopathological analysis of neurodegeneration following an ischemic tolerance paradigm. (A) Representative images of Fluoro-Jade B (FJ)-stained brain sections in the sham-operation group (i), IPC group (ii), sham + SI group (iii) and IPC + SI group (iv). Note the numerous degenerated neurons in the sham + SI group, with few degenerated neurons in the IPC + SI group. Scale bar = 50 lm. (B) The quantitative analysis shows the numbers of FJ-positive degenerated neurons in the CA1 subregion in the different treatment groups. Data are presented as the mean ± SEM; ⁄p < 0.05 compared with the sham + SI group. (C) Representative images of NeuN immunofluorescence in the sham-operation group (i), IPC group (ii), sham + SI group (iii) and IPC + SI group (iv). Note that the NeuN immunofluorescence is lost in the hippocampal CA1 neurons of the sham + SI group but preserved in neurons of the IPC + SI group. Scale bar = 20 lm. (D) The quantitative analysis shows the numbers of NeuN-positive intact pyramidal neurons in the CA1 subregion in the different treatment groups. Data are presented as the mean ± SEM; ⁄p < 0.05 compared with the sham + SI group.

 

Further analysis using double-labeling techniques showed that cells overexpressing VEGF-C were NeuN- expressing neurons (Fig. 3C). IPC treatment did not alter the levels of VEGFR-3 immunofluorescence in the CA1 pyramidal neurons (Fig. 3B). However, IPC
treatment resulted in the increased VEGFR-3 expression in some cells located ventral to the CA1 region, primarily in the stratum radiatum and stratum lacunosum moleculare after 3 and 7 days of reperfusion (Fig. 4). A quantitative analysis showed that 3 and
7 days after IPC, VEGFR-3-positive cells increased 3.3- and 3.6-fold, respectively, compared with sham. In addition, analysis of the double immunofluorescence revealed that the VEGFR-3-expressing cells were colabeled with NeuN (a neuronal marker; Fig. 5A) but not with GFAP (an astrocyte marker; Fig. 5B), OX-42 (a microglial marker; Fig. 5C) or CD31 (an endothelial cell marker; data not shown). Taken together, these results showed that IPC stimulation regulated VEGF-C and VEGFR-3 expression in the hippocampal CA1 region.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
Fig. 3. The effect of IPC on VEGF-C and VEGFR-3 expression in the hippocampal CA1 pyramidal cell layer. (A) Representative images of VEGF-C immunofluorescence in the sham-operation and IPC-treated groups at 1, 3, 7 and 14 days of reperfusion. Note that the VEGF-C immunoreactivity in the CA1 cell layer increased in the IPC groups after 3 and 7 days of reperfusion compared with the sham-operation group and returned to basal level after 14 days. Scale bar = 50 lm. The quantitative analysis shows the relative expression of VEGF-C in the CA1 subregion in the different treatment groups. Data are presented as the mean ± SEM; ⁄p < 0.05 compared with the sham-operation group. (B) Representative images of VEGFR-3 immunofluorescence in the sham-operation and IPC-treated groups at 1, 3, 7 and 14 days of reperfusion. Note that the VEGFR-3 immunoreactivity in the CA1 cell layer remained unchanged in the IPC groups after 1, 3, 7 and 14 days of reperfusion compared with the sham- operation group. Scale bar = 50 lm. The quantitative analysis shows the relative expression of VEGFR-3 in the CA1 subregion in the different treatment groups. (C) Representative images of immunofluorescence showing that the VEGF-C-overexpressing cells are colabeled with NeuN in the CA1 region of the IPC group. Scale bar = 50 lm.
Effects of IPC on VEGF-C and VEGFR-3 expression in the hippocampus after SI

To investigate the possible involvement of VEGF-C and its receptor VEGFR-3 in IPC-induced neuroprotection, we examined their expression levels 7 days after SI. Interestingly, VEGF-C immunoreactivity was decreased in the CA1 pyramidal cell layer in the sham + SI group compared with the sham-operation group (Fig. 5A). In contrast, there was a clear induction of VEGF-C
expression in the IPC + SI group compared with both the sham-operation and the sham + SI groups (Fig. 6A). Immunofluorescence analysis showed that cells with an increased VEGF-C expression in the tolerant group were positive for NeuN (Fig. 6B). In addition, the VEGFR-3 immunoreactivity in the CA1 neurons was decreased in the sham + SI group compared with the sham-operation group, whereas the VEGFR-3 immunoreactivity in CA1 neurons in the IPC + SI group was sustained (Fig. 6C). The sustained

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
Fig. 4. The effect of IPC on VEGFR-3 expression in the hippocampus. (A) Representative images of VEGFR-3 immunofluorescence from the sham-operation and IPC-treated groups at 1, 3 and 7 days of reperfusion. Scale bar = 20 lm. (C) The quantitative analysis shows the number of VEGFR-3-positive cells counted in the area indicated (Fig. 4B, adapted from The Mouse Brain in Stereotaxic Coordinates, Compact 3rd Edition, Keith B.J. Franklin and George Paxinos, The horizontal plates and diagrams, Copyright Elsevier, 2008). Data are presented as the mean ± SEM; ⁄p < 0.05 compared with the sham-operation group.

 

VEGFR-3 expression induced by IPC and the overexpression of VEGF-C in the tolerant neurons suggest that VEGF-C/VEGFR-3 signaling may be associated with the IPC-induced neuroprotection.

VEGFR-3 inhibitor SAR131675 blocks IPC-induced ischemic tolerance
To determine whether VEGF-C/VEGFR-3 signaling in tolerant neurons is functionally involved in mediating neuroprotection against SI, the IPC-treated mice were administered SAR131675, a potent inhibitor of VEGFR-3 (Alam et al., 2012; Espagnolle et al., 2014) for 7 days start- ing 30 min before SI. The histological analysis showed that treating sham-operated mice with SAR131675 of a high- dose did not show any effect on CA1 pyramidal neurons in terms of FJ and NeuN stainings (Fig. 7). However, in
the IPC + SI + SARH (50 mg/kg + 25 mg/kg daily there- after) group, there was an increase in FJ staining as well as a decrease in NeuN immunofluorescence in the hippo- campal CA1 region compared with the IPC + SI + Veh group (Fig. 7). The quantitative analysis showed that the numbers of FJ-positive neurons in the IPC + SI + SARL (25 mg/kg + 12.5 mg/kg daily thereafter) and the
IPC + SI + SARH groups were 52.8 ± 12.6 and 63.9 ± 11.8 (per field), respectively, which were signifi- cantly higher than that in the vehicle-treated IPC + SI group (22.5 ± 6.8; Fig. 7B). Similarly, the numbers of
NeuN-positive neurons were 50.2 ± 10.1 and 39.7 ± 9.0 in the IPC + SI + SARL and IPC + SI + SARH groups, respectively, which were significantly lower than in the vehicle-treated IPC + SI group (81.2 ± 7.0; Fig. 7D), suggesting that the SAR131675 treatment atten- uated IPC-induced tolerance in a dose-dependent

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
Fig. 5. The characterization of VEGFR-3-positive cells in the mouse hippocampus (area shown in the box of Fig. 4B) after 3 days of reperfusion following IPC. Note that the VEGFR-3 signal colabeled with the NeuN (A) but not with GFAP (B) or OX-42 (C) immunofluorescence. Scale bar = 20 lm.
manner. Thus, blockade of the IPC-induced neuroprotec- tion in the mouse hippocampus with a VEGFR-3 inhibitor confirmed that VEGFR-3 activity is functionally involved in the IPC-induced ischemic tolerance.
DISCUSSION
Our study presents evidence that VEGF-C/VEGFR-3 signaling is involved in IPC-induced neuroprotection in
mouse hippocampus. We found that a brief episode of ischemia resulted in a substantial reduction in CA1 neuronal damage in the hippocampus after subsequent SI. The neuroprotection was observed 3 days after the IPC episode, at a time when IPC-induced overexpression of VEGF-C and VEGFR-3 occurred in the hippocampus. We also found that IPC increased VEGF-C expression and preserved a sustained VEGFR- 3 expression in the tolerant neurons after SI.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
Fig. 6. The effect of IPC on VEGF-C and VEGFR-3 expression in the CA1 pyramidal neurons after SI. (A) Representative images of VEGF-C immunohistochemistry in the sham-operation group, sham + SI group and IPC + SI group. Note that there was induced expression of VEGF-C in the IPC + SI group compared with both the sham-operation and sham + SI groups. Scale bar = 50 lm. The quantitative analysis shows the relative expression of VEGF-C in the CA1 subregion in the different treatment groups. Data are presented as the mean ± SEM; ⁄p < 0.05 compared with the sham-operation group and #p < 0.05 compared with the sham + SI group. (B) Representative immunofluorescence images showing that the VEGF-C signal is colabeled with the NeuN signal in the CA1 region of the IPC + SI group. Scale bar = 50 lm. (C) Representative images of VEGFR-3 immunofluorescence in the sham-operation, sham + SI and IPC + SI groups. Note that the VEGFR-3 immunoreactivity in the CA1 neurons was reduced in the sham + SI group compared with the sham-operation group, whereas the VEGFR-3 immunoreactivity in the IPC + SI group was sustained. Scale bar = 50 lm.
Furthermore, we observed that application of the VEGFR- 3 inhibitor SAR131675 abolished IPC-induced neuroprotection in a dose-dependent manner in the
mouse hippocampus, confirming the functional involvement of VEGFR-3 activity in ischemic tolerance in the mouse hippocampus.
VEGF-C and VEGFR-3 are expressed in lymphatic and normal endothelial cells and play a major role in the formation of lymphatic vessels during embryonic development and in adults. In addition to their specific function in lymphangiogenesis, VEGF-C and VEGFR-3 have also been detected in the mouse brain from embryonic day 15.5, particularly in the olfactory bulb and the optic nerve region, acting as a selective growth factor for neural progenitor cells (Le Bras et al., 2006). Interestingly, our group showed that VEGF-C and VEG- FR-3 are expressed in the cortical and hippocampal neu- rons in the adult rat brain and that their expression is
regulated in response to cerebral ischemia (Shin et al., 2008; Hou et al., 2011). In addition, our previous studies showed that VEGF-C mRNA and VEGFR-3 protein expressions were induced in macrophages and reactive astrocytes in the rat brain after middle cerebral artery occlusion- and 4-vessel occlusion-induced lethal ische- mia (Shin et al., 2008, 2010b). In contrast, the results of this study showed that the expression of VEGF-C and VEGFR-3 was induced in CA1 pyramidal neurons and neuronal cells in stratum radiatum/stratum lacunosum moleculare, respectively, after sublethal IPC treatment in mice. Moreover, there was no reactive astrogliosis observed in the mouse hippocampus after sublethal IPC treatment (data not shown). Although the reasons for the differential expression of VEGFR-3 in different kinds of brain cells in our present and previous studies are unclear, those may arise from the difference in animal models adopted. Main differences between animal

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig. 7. Blockade of IPC-induced ischemic tolerance by the VEGFR-3 inhibitor SAR131675. (A) Representative images of FJ-stained brain sections from sham + SARH (i), sham-IPC + SI + Veh (ii), IPC + SI + Veh (iii) and IPC + SI + SARH (iv) groups. Note that there were more degenerated neurons in the IPC + SI + SARH group receiving SAR131675 of 50 mg/kg as an initial dose than in the IPC + SI + Veh group. Scale bar = 50 lm. (B) The quantitative analysis shows the number of FJ-positive degenerated neurons in the CA1 subregion of the different treatment groups including the IPC + SI + SARL group receiving SAR131675 of 25 mg/kg as an initial dose. Data are presented as the mean ± SEM; #p < 0.05 compared with the sham-IPC + SI + Veh group; ⁄p < 0.05 compared with the IPC + SI + Veh group. (C) Representative images of NeuN immunofluorescence in the sham + SARH (i), sham-IPC + SI + Veh (ii), IPC + SI + Veh (iii) and IPC + SI + SARH (iv) groups. Note that NeuN-immunofluorescence in the IPC + SI + SARH group was reduced compared with the IPC + SI + Veh group. Scale bar = 50 lm. (D) The quantitative analysis shows the number of NeuN-positive intact pyramidal neurons in the CA1 subregion for the different treatment groups. Data are presented as the mean ± SEM; #p < 0.05 compared with the sham-IPC + SI + Veh group; ⁄p < 0.05 compared with the IPC + SI + Veh group.

 

models of our present and previous studies include ische- mic severity (sublethal versus lethal) and species (mouse versus rat). Indeed, we have found that SI (40-min BCCAO) in mice resulted in huge astrogliosis with thick- ened glial processes and hypoertrophoid cell body and, also, induced the expression of VEGFR-3 in glial cells in the hippocampus (data not shown), suggesting that ische- mic severity is one of the reasons for differential expres- sion of VEGFR-3 in different kinds of brain cells. Therefore, apart from playing a neurotrophic role in
normal physiology, VEGF-C/VEGFR-3 signaling may have additional functions under pathological conditions in the brain (Shin et al., 2008, 2010a). Indeed, one recent study showed that VEGF-C protects dopaminergic neu- rons in vitro and provides an additive survival effect with glial-derived neurotrophic factor on tyrosine hydroxy- lase-positive cells in the rat unilateral 6-hydroxydopamine model of Parkinson’s disease, indicating a neuroprotec- tive function of VEGF-C (Piltonen et al., 2011). Given that VEGF-C has neurotrophic and neuroprotective effects (Le
Bras et al., 2006; Piltonen et al., 2011), the IPC-induced expression of VEGF-C and sustained expression of VEG- FR-3 in neurons after SI observed in our study could rea- sonably reflect a pro-survival response of neurons against ischemic death. Indeed, Wang et al. showed that VEGFR- 3 is activated in endothelial cells under oxidative stress and serves as a cell survival modulator, which supports our hypothesis that VEGF-C/VEGFR-3 signaling is involved in the endogenous adaptive response of the brain to ischemic stress (Wang et al., 2004). To date, no report has shown a neuroprotective function for VEGF-C in cerebral ischemia or ischemic tolerance. Based on previous reports examining the neuroprotective function of VEGF-A and VEGF-B, the functional implica- tions of our findings can be inferred according to the fol- lowing. VEGF-A and VEGF-B are neuroprotective against glutamate-induced excitotoxicity and in vitro ischemia-induced cell death as well as in animal models of stroke (Matsuzaki et al., 2001; Svensson et al., 2002; Sun et al., 2003; Li et al., 2008). Moreover, the involve- ment of VEGF-A in IPC-induced neuroprotection in cul- tured neurons and neonatal rat and piglet brains also supports our hypothesis (Lee et al., 2009; Ara et al., 2013). Importantly, the neuroprotective effects of VEGF- A and VEGF-B have been ascribed to the activation of VEGFR-1 or VEGFR-2 (Wick et al., 2002; Lee et al., 2009; Ara et al., 2013). In the present study, we used a pharmacological inhibitor of VEGFR-3 to show that VEGF-C mediated its neuroprotective function via VEGFR-3. However, blocking VEGFR-3 activity did not completely attenuate the IPC-induced neuroprotection. Therefore, VEGFR-2 signaling may contribute to some of the neuroprotective effects of VEGF-C we observed, as VEGF-C can bind to and mediate some of its actions via VEGFR-2 (Yamazaki and Morita, 2006). Further, it has been reported that both VEGF-A and VEGF-C can induce formation of receptors heterodimers (VEGFR-2/
VEGFR-3) in addition to homodimers and can exert their biological responses (Dixelius et al., 2003; Nilsson et al., 2010). Moreover, the bioavailability of SAR131675 in the brain is also not known, although it has been reported that transient forebrain ischemia can open the blood–brain barrier in the hippocampus which facilitates entering of small molecule drugs into the brain (Won et al., 2011). Therefore, future investigations are needed to examine the bioavailability of VEGFR-3 inhibitors in the brain and the possible effect of VEGFR-3 inhibitors on VEGF-A sig- naling in the ischemic tolerance paradigm as well. Although the molecular mechanism for the VEGF-C/VEG- FR-3 mediated neuroprotection was not examined in our study, our results raise, for the first time, the possibility of a neuroprotective function mediated through the VEGF-C/VEGFR-3 axis in the mouse hippocampus.
Several cellular signaling pathways including the phosphatidylinositol 3-kinase (PI3K/Akt) and mitogen- activated protein kinase pathways are involved in the mechanisms of ischemic tolerance (Wick et al., 2002; Choi et al., 2007; Bhuiyan et al., 2011). The multifaceted protein kinase Akt inhibits programed cell death and enhances cell survival via regulating its many cytoplasmic targets such as Bcl-2-associated death promoter,

glycogen synthase kinase-3b, procaspase-9 and cAMP response element-binding protein. Interestingly, the VEGF-C/VEGFR-3 axis is also known to induce the PI3K/Akt pathway, which inhibits serum deprivation- induced apoptosis in endothelial cells (Makinen et al., 2001) and protects against chemotherapy-induced apop- tosis in leukemia cells (Dias et al., 2002) via Bcl-2 induc- tion, suggesting that the cell survival function of VEGF-C is associated with the activation of the VEGFR-3/PI3K/Akt pathway and inhibition of apoptosis. In line with the above notion, it has been reported that VEGF-A, a close homo- log of VEGF-C, mediates IPC-induced ischemic tolerance via activation of VEGFR-2 and the PI3K/Akt pathway in cultured neurons (Wick et al., 2002). Therefore, it is pos- sible that VEGF-C/VEGFR-3 signaling also mediates ischemic tolerance, likely via the PI3K/Akt pathway, although further investigation is required to validate this hypothesis.

CONCLUSION
We showed that the VEGF-C/VEGFR-3 system underlies the protective effect of IPC against forebrain ischemia in the mouse hippocampus. Future studies examining the mechanism for the neuroprotection mediated by the VEGF-C/VEGFR-3 axis may help to fully elucidate the mechanism of ischemic tolerance. Our study demonstrated the importance of the development of VEGF-C and its agonists as potential IPC mimetics as well as neuroprotective agents against cerebral ischemia.

Acknowledgements—The authors have no conflicts of interest to declare. This research was supported by the Mid-Career Researcher Program through the National Research Foundation of Korea (NRF) grant funded by the MEST (2011-0028319).
REFERENCES

Alam A, Blanc I, Gueguen-Dorbes G, Duclos O, Bonnin J, Barron P, Laplace MC, Morin G, Gaujarengues F, Dol F, Herault JP, Schaeffer P, Savi P, Bono F (2012) SAR131675, a potent and selective VEGFR-3-TK inhibitor with antilymphangiogenic, antitumoral, and antimetastatic activities. Mol Cancer Ther 11: 1637–1649.
Ara J, Shukla P, Frank M (2013) Enhanced expression of the Flt-1 and Flk-1 receptor tyrosine kinases in a newborn piglet model of ischemic tolerance. J Neurochem 124:735–746.
Bhuiyan MI, Kim YJ (2010) Mechanisms and prospects of ischemic tolerance induced by cerebral preconditioning. Int Neurourol J 14:203–212.
Bhuiyan MI, Jung SY, Kim HJ, Lee YS, Jin C (2011) Major role of the PI3K/Akt pathway in ischemic tolerance induced by sublethal oxygen-glucose deprivation in cortical neurons in vitro. Arch Pharm Res 34:1023–1034.
Cho S, Park EM, Zhou P, Frys K, Ross ME, Iadecola C (2005) Obligatory role of inducible nitric oxide synthase in ischemic preconditioning. J Cereb Blood Flow Metab 25:493–501.
Cho KO, Kim SK, Cho YJ, Sung KW, Kim SY (2006) A simple method for predicting hippocampal neurodegeneration in a mouse model of transient global forebrain ischemia. Korean J Physiol Pharmacol 10:167–172.
Cho KO, Kim SK, Cho YJ, Sung KW, Kim SY (2007) Regional differences in the neuroprotective effect of minocycline in a mouse model of global forebrain ischemia. Life Sci 80:2030–2035.
Choi YS, Cho KO, Kim EJ, Sung KW, Kim SY (2007) Ischemic preconditioning in the rat hippocampus increases antioxidant activities but does not affect the level of hydroxyl radicals during subsequent severe ischemia. Exp Mol Med 39:556–563.
Dias S, Choy M, Alitalo K, Rafii S (2002) Vascular endothelial growth factor (VEGF)-C signaling through FLT-4 (VEGFR-3) mediates leukemic cell proliferation, survival, and resistance to chemotherapy. Blood 99:2179–2184.
Dirnagl U, Simon RP, Hallenbeck JM (2003) Ischemic tolerance and endogenous neuroprotection. Trends Neurosci 26:248–254.
Dixelius J, Makinen T, Wirzenius M, Karkkainen MJ, Wernstedt C, Alitalo K, Claesson-Welsh L (2003) Ligand-induced vascular endothelial growth factor receptor-3 (VEGFR-3)
heterodimerization with VEGFR-2 in primary lymphatic endothelial cells regulates tyrosine phosphorylation sites. J Biol Chem 278:40973–40979.
Espagnolle N, Barron P, Mandron M, Blanc I, Bonnin J, Agnel M, Kerbelec E, Herault JP, Savi P, Bono F, Alam A (2014) Specific inhibition of the VEGFR-3 tyrosine kinase by SAR131675 reduces peripheral and tumor associated immunosuppressive myeloid cells. Cancers (Basel) 6:472–490.
Franklin KBJ, Paxinos G (2008) The mouse brain in stereotaxic coordinates. San Diego: Academic Press.
Greenberg DA, Jin K (2013) Vascular endothelial growth factors (VEGFs) and stroke. Cell Mol Life Sci 70:1753–1761.
Hou Y, Shin YJ, Han EJ, Choi JS, Park JM, Cha JH, Choi JY, Lee MY (2011) Distribution of vascular endothelial growth factor receptor- 3/Flt4 mRNA in adult rat central nervous system. J Chem Neuroanat 42:56–64.
Jeong KH, Lee KE, Kim SY, Cho KO (2011) Upregulation of Kruppel- like factor 6 in the mouse hippocampus after pilocarpine-induced status epilepticus. Neuroscience 186:170–178.
Kitagawa K, Matsumoto M, Kuwabara K, Tagaya M, Ohtsuki T, Hata R, Ueda H, Handa N, Kimura K, Kamada T (1991) ‘Ischemic tolerance’ phenomenon detected in various brain regions. Brain Res 561:203–211.
Le Bras B, Barallobre MJ, Homman-Ludiye J, Ny A, Wyns S, Tammela T, Haiko P, Karkkainen MJ, Yuan L, Muriel MP, Chatzopoulou E, Breant C, Zalc B, Carmeliet P, Alitalo K, Eichmann A, Thomas JL (2006) VEGF-C is a trophic factor for neural progenitors in the vertebrate embryonic brain. Nat Neurosci 9:340–348.
Lee KE, Kim SK, Cho KO, Kim SY (2008) Pre-ischemic treatment with ampicillin reduces neuronal damage in the mouse hippocampus and neostriatum after transient forebrain ischemia. Korean J Physiol Pharmacol 12:287–291.
Lee HT, Chang YC, Tu YF, Huang CC (2009) VEGF-A/VEGFR-2 signaling leading to cAMP response element-binding protein phosphorylation is a shared pathway underlying the protective effect of preconditioning on neurons and endothelial cells. J Neurosci 29:4356–4368.
Li Y, Zhang F, Nagai N, Tang Z, Zhang S, Scotney P, Lennartsson J, Zhu C, Qu Y, Fang C, Hua J, Matsuo O, Fong GH, Ding H, Cao Y, Becker KG, Nash A, Heldin CH, Li X (2008) VEGF-B inhibits apoptosis via VEGFR-1-mediated suppression of the expression of BH3-only protein genes in mice and rats. J Clin Invest 118:913–923.
Liu Y, Kato H, Nakata N, Kogure K (1993) Temporal profile of heat shock protein 70 synthesis in ischemic tolerance induced by preconditioning ischemia in rat hippocampus. Neuroscience 56:921–927.
Liu F, Schafer DP, McCullough LD (2009) TTC, fluoro-Jade B and NeuN staining confirm evolving phases of infarction induced by middle cerebral artery occlusion. J Neurosci Methods 179:1–8.
Makinen T, Veikkola T, Mustjoki S, Karpanen T, Catimel B, Nice EC, Wise L, Mercer A, Kowalski H, Kerjaschki D, Stacker SA, Achen MG, Alitalo K (2001) Isolated lymphatic endothelial cells transduce growth, survival and migratory signals via the VEGF- C/D receptor VEGFR-3. EMBO J 20:4762–4773.
Matsuzaki H, Tamatani M, Yamaguchi A, Namikawa K, Kiyama H, Vitek MP, Mitsuda N, Tohyama M (2001) Vascular endothelial

growth factor rescues hippocampal neurons from glutamate- induced toxicity: signal transduction cascades. FASEB J 15:1218–1220.
Munoz A, Nakazaki M, Goodman JC, Barrios R, Onetti CG, Bryan J, Aguilar-Bryan L (2003) Ischemic preconditioning in the hippocampus of a knockout mouse lacking SUR1-based K(ATP) channels. Stroke 34:164–170.
Murry CE, Jennings RB, Reimer KA (1986) Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium. Circulation 74:1124–1136.
Nilsson I, Bahram F, Li X, Gualandi L, Koch S, Jarvius M, Soderberg O, Anisimov A, Kholova I, Pytowski B, Baldwin M, Yla-Herttuala S, Alitalo K, Kreuger J, Claesson-Welsh L (2010) VEGF receptor 2/- 3 heterodimers detected in situ by proximity ligation on angiogenic sprouts. EMBO J 29:1377–1388.
Ny A, Koch M, Vandevelde W, Schneider M, Fischer C, Diez-Juan A, Neven E, Geudens I, Maity S, Moons L, Plaisance S, Lambrechts D, Carmeliet P, Dewerchin M (2008) Role of VEGF-D and VEGFR-3 in developmental lymphangiogenesis, a chemicogenetic study in Xenopus tadpoles. Blood 112: 1740–1749.
Obrenovitch TP (2008) Molecular physiology of preconditioning- induced brain tolerance to ischemia. Physiol Rev 88:211–247.
Piltonen M, Planken A, Leskela O, Myohanen TT, Hanninen AL, Auvinen P, Alitalo K, Andressoo JO, Saarma M, Mannisto PT (2011) Vascular endothelial growth factor C acts as a neurotrophic factor for dopamine neurons in vitro and in vivo. Neuroscience 192:550–563.
Schmidt-Kastner R, Aguirre-Chen C, Kietzmann T, Saul I, Busto R, Ginsberg MD (2004) Nuclear localization of the hypoxia-regulated pro-apoptotic protein BNIP3 after global brain ischemia in the rat hippocampus. Brain Res 1001:133–142.
Shin YJ, Choi JS, Lee JY, Choi JY, Cha JH, Chun MH, Lee MY (2008) Differential regulation of vascular endothelial growth factor-C and its receptor in the rat hippocampus following transient forebrain ischemia. Acta Neuropathol 116:517–527.
Shin YJ, Choi JS, Choi JY, Cha JH, Chun MH, Lee MY (2010a) Enhanced expression of vascular endothelial growth factor receptor-3 in the subventricular zone of stroke-lesioned rats. Neurosci Lett 469:194–198.
Shin YJ, Choi JS, Choi JY, Hou Y, Cha JH, Chun MH, Lee MY (2010b) Induction of vascular endothelial growth factor receptor-3 mRNA in glial cells following focal cerebral ischemia in rats. J Neuroimmunol 229:81–90.
Sun Y, Jin K, Xie L, Childs J, Mao XO, Logvinova A, Greenberg DA (2003) VEGF-induced neuroprotection, neurogenesis, and angiogenesis after focal cerebral ischemia. J Clin Invest 111: 1843–1851.
Svensson B, Peters M, Konig HG, Poppe M, Levkau B, Rothermundt M, Arolt V, Kogel D, Prehn JH (2002) Vascular endothelial growth factor protects cultured rat hippocampal neurons against hypoxic injury via an antiexcitotoxic, caspase-independent mechanism. J Cereb Blood Flow Metab 22:1170–1175.
Tammela T, Enholm B, Alitalo K, Paavonen K (2005) The biology of vascular endothelial growth factors. Cardiovasc Res 65:550–563.
Thompson JW, Dave KR, Saul I, Narayanan SV, Perez-Pinzon MA (2013) Epsilon PKC increases brain mitochondrial SIRT1 protein levels via heat shock protein 90 following ischemic preconditioning in rats. PLoS One 8:e75753.
Wang JF, Zhang X, Groopman JE (2004) Activation of vascular endothelial growth factor receptor-3 and its downstream signaling promote cell survival under oxidative stress. J Biol Chem 279:27088–27097.
Wick A, Wick W, Waltenberger J, Weller M, Dichgans J, Schulz JB (2002) Neuroprotection by hypoxic preconditioning requires sequential activation of vascular endothelial growth factor receptor and Akt. J Neurosci 22:6401–6407.
Won SM, Lee JH, Park UJ, Gwag J, Gwag BJ, Lee YB (2011) Iron mediates endothelial cell damage and blood-brain barrier opening in the hippocampus after transient forebrain ischemia in rats. Exp Mol Med 43:121–128.
Wu C, Fujihara H, Yao J, Qi S, Li H, Shimoji K, Baba H (2003) Different expression patterns of Bcl-2, Bcl-xl, and Bax proteins after sublethal forebrain ischemia in C57Black/Crj6 mouse striatum. Stroke 34:1803–1808.
Yamazaki Y, Morita T (2006) Molecular and functional diversity of vascular endothelial growth factors. Mol Divers 10:515–527.

Yoshioka H, Niizuma K, Katsu M, Sakata H, Okami N, Chan PH (2011) Consistent injury to medium spiny neurons and white matter in the mouse striatum after prolonged transient global cerebral ischemia. J Neurotrauma 28:649–660.
(Accepted 16 January 2015)
(Available online 28 January 2015)