The novel Nrf2 activator CDDO‐EA attenuates cerebral ischemic injury by promoting microglia/macrophage polarization toward M2 phenotype in mice

Abstract The aim of present study was to explore whether 2‐cyano‐3, 12‐dioxooleana‐1, 9‐dien‐28‐oic acid (CDDO)‐ethylamide (CDDO‐EA) attenuates cerebral ischemic injury and its possible mechanisms using a middle cerebral artery occlusion (MCAO) model in C57BL/6 mice. Our results showed that intraperitoneal injection (i.p.) of CDDO‐EA (2 and 4 mg/kg) augmented NFE2‐related factor 2 (Nrf2) and heme oxygenase‐1 (HO‐1) expression in ischemic cortex after MCAO. Moreover, CDDO‐EA (2 mg/kg, i.p.) significantly enhanced Nrf2 nuclear accumulation, associated with increased cytosolic HO‐1 expression, reduced neurological deficit and infarct volume as well as neural apoptosis, and shifted polarization of microglia/macrophages toward an antiinflammatory M2 phenotype in ischemic cortex after MCAO. Using an in vitro model, we confirmed that CDDO‐EA (100 μg/mL) increased HO‐1 expression and primed microglial polarization toward M2 phenotype under inflammatory stimulation in BV2 microglial cells. These findings suggest that a novel Nrf2 activator CDDO‐EA confers neuroprotection against ischemic injury.

microglia/macrophages are rapidly activated, mobilize to the injury site, and initiate the release of effectors and recruitment of peripheral inflammatory cells. 3,4 Microglia/macrophages have high plasticity that can assume diametrically opposed functional phenotypes when responding to micro-environmental triggers. One phenotype is the "classically activated" M1 that release destructive proinflammatory mediators. The polar extreme phenotype is "alternatively activated" antiinflammatory M2 phenotype that has been associated with neuroprotective effects. Recent studies suggest that modulation of microglia/macrophage polarization toward M2 phenotype may be harnessed as an important treatment strategy for brain repair in ischemic stroke. 5,6 Recent research indicates that activation of the redox transcription factor NFE2 related factor 2 (Nrf2) contributes to the antiinflammatory nature of the M2 microglia/macrophage phenotype. 7,8 Activated Nrf2 translocates to the nucleus and binds to the antioxidative response element (ARE) in cells exposed to oxidative stress, which triggers the transcription of antioxidant and antiinflammatory genes. 9 Nrf2 and its downstream transcriptional target heme oxygenases-1 (HO-1) have neuroprotective effects against some models of ischemic injury in the central nervous system. [10][11][12][13] 2-Cyano-3,12-Dioxooleana-1,9-Dien-28-Oic acid (CDDO) and its analogues originate from oleanic acid and can activate Nrf2 signaling in both cell culture and animal models, exhibiting antiinflammatory and antioxidant activities. 14-16 CDDO-ethyl amide (CDDO-EA, which chemical structure is shown as Figure 1A) has better bio-availability and can effectively penetrate the blood-brain barrier in mice. 17,18 However, it is not known whether CDDO-EA confers neuroprotection against ischemic injury or its impact on microglia/macrophage polarization in the ischemic context.
In this study, we aim to explore the influence of CDDO-EA on cerebral ischemic injury and microglia/macrophage polarization using the mice model of transient focal ischemia.

| Experimental design
Using an in vivo model, we first evaluated the optimal dose of CDDO-EA (1, 2 and 4 mg/kg, i.p.) to induce Nrf2 and HO-1 protein expression and subcellular compartmentalization in MCAO F I G U R E 1 CDDO-EA increases Nrf2 and HO-1 expression in cortex after transient MCAO. A, Chemical structure of CDDO-EA. B, Scheme for the experimental design. C, Nrf2 and HO-1 protein expression in the ischemic cortex was detected by Western blot 48 h after MCAO. Mice were injected with CDDO-EA (0, 1, 2, and 4 mg/kg, i.p.) 30 min after MCAO surgery. Cortexes were collected at 48 h after MCAO, and homogenates were blotted with anti-Nrf2, anti-HO-1, and anti-β-actin. Quantification of optical density was normalized to sham controls. D, Mice were injected with CDDO-EA (2 mg/kg, i.p.) followed by MCAO. Ischemic cortexes were collected at 48 h after MCAO and subjected to subcellular fractionation. Nuclear Nrf2 and cytosolic HO-1 were detected by Western blot, and Histone H3 and β-actin were used as loading controls for nuclear and total protein, respectively. Data are mean ± SD. (ns denotes not significant, * P < 0.05, n = 3) mice. Thirty mice were divided into five groups (n = 6/group): sham, MCAO + vehicle, MCAO + CDDO-EA (1 mg/kg, 2 mg/kg, or 4 mg/ kg, i.p.). CDDO-EA was delivered 30 minutes after the end of the ischemic period (ie, after reperfusion).
Expression levels of HO-1, CD16, and CD11b, and microglial phagocytotic activity were determined by Western blot analysis, real-time quantitative PCR, and fluorescent microsphere uptake in LPS-activated BV2 microglial cells.
All animals were randomly assigned to sham, MCAO + vehicle, or MCAO + CDDO-EA groups using a lottery-drawing box. All experiments were performed by an investigator blinded to experimental group assignments.

| In vivo model of transient focal ischemia
All animal experiments were approved by the Ethics Committee of the Shandong First Medical University and performed in accordance with guidelines of the Use of Experimental Animals of National Institutes of Health. Male C57BL/6 mice (10-to 12-week old, 22-30 g) were purchased from Pengyue Laboratory Animal Breeding Co., Ltd.
Mice were deeply anesthetized with 3% isoflurane and maintained with 1.5% isoflurane in oxygen/nitrous oxide (30%:70%) by a rodent ventilator (RWD, China), and focal ischemia was induced by intraluminal occlusion of left middle cerebral artery (MCAO) as in our previous study. 19 Briefly, a midline cervical skin incision was made under the surgical microscope; then, the underlying muscular attachment was separated to expose the left common carotid artery (CCA), the internal carotid artery (ICA) and its pterygopalatine artery branch, and the external carotid artery (ECA). The superior thyroid and distal ECA were permanently coagulated, but the other arteries were only temporarily ligated. A 6-0 nylon suture was inserted into the left ICA through a dissected ECA, which was advanced 9 mm to arrive the bifurcation of the anterior cerebral artery. The animals underwent MCAO for 60 minutes after which the suture was removed, allowing for reperfusion. To confirm the occurrence of MCAO, changes in local cerebral blood flow (CBF) were measured using a laser Doppler blood flow imager (MoorLDI2). Animals that died or failed to show a CBF reduction of at least 75% were excluded from further experimentation.
In sham-operated groups, mice were anesthetized, the ECA branches were dissected, and the wound was sutured. Thirty min after reperfusion, CDDO-EA was administered i.p. to treatment groups.

| Western blot
Western blot analysis was performed according to previous reports. 9,20 The peri-infarct region of the cortex from the left hemisphere was harvested 48 hours after ischemia. Whole-cell lysates were prepared using an RIPA Lysis Buffer (P0013, Beyotime). Cytoplasmic and nu-

| Neurological score
After MCAO mice were awake for 24 and 48 hours, neurological deficit was scored using the modified Longa method (Table 1) according to a previous report. 13 The mice were randomly assigned to MCAO + vehicle (Vehicle) and MCAO + CDDO-EA (CDDO-EA) groups using a lottery-drawing box. All of the outcome assessments were performed by investigators blinded to the group assignments.

| Assessment of cerebral infarct
2, 3, 5-triphenyltetrazolium chloride (TTC) staining was used to assess the infarct size according to a previous study. 19 The forebrains of mice were removed at 48 hours after MCAO under anesthesia with 3% isoflurane. The infarct volume was measured by 2% TTC (Sigma-Aldrich) staining on 2-mm-thick coronal fresh brain sections. Normal brain area was stained red, while the infarct area not stained (white). The infarct volume is quantified from the stained sections by using ImageJ software.

| Real-Time quantitative PCR
Real-time quantitative PCR was performed as in a previous report. 6 Total RNA (4 μg) was extracted from BV2 cells and reversed transcribed into cDNA using SuperScript III First-Strand System (18080-051, Invitrogen). PCR was performed on a real-time PCR detection system (Bio-Rad) using corresponding primers (

| Statistical analysis
Data were analyzed with GraphPad Prism 6.0 Software and expressed as mean ± SD (standard deviation). Levene test was used for assessing the homogeneity of variance. Student two-tailed t test was used for the comparison of two experimental groups when the data were normally distributed. Wlicoxon rank-sum test was used for data comparisons with non-normal distributions. P value < 0.05 was considered statistically significant.

| CDDO-EA increases Nrf2 and HO-1 in cortex after MCAO
We first wished to determine the optimal dose of CDDO-EA in order to induce Nrf2 protein expression following MCAO. Mice were ad-  Figure 1C and Figure S1A). No significant difference between the two higher doses of CDDO-EA was observed ( Figure 1C, middle panel). Using the 2 mg/kg dose of CDDO-EA, we next performed subcellular fractionation to determine whether Nrf2 is enriched in the nuclear fraction in ischemic brain, which is reflective of the dissociation of Nrf2 with tethering proteins in the cytosol. Neither sham nor vehicle-treated MCAO brain exhibited substantial levels of Nrf2 protein  Figure 1D and Figure S1B). These results indicate that treatment with CDDO-EA following ischemic stroke can both increase total Nrf2 protein expression in brain and increase nuclear translocation of Nrf2 protein, consistent with a possible increase in transactivational activity.
HO-1 is a transcriptional target of nuclear Nrf2 transactivational activity; thus, increased expression of HO-1 is considered consistent with Nrf2 nuclear activity. Following MCAO in vehicle-treated mice, HO-1 protein expression increased significantly compared to sham controls ( Figure 1C), reflecting endogenous processes are active in ischemic brain that increase HO-1 expression 48 hours after MCAO.
Administration of CDDO-EA 30 minutes following MCAO significantly up-regulated HO-1 protein expression, consistent with the up-regulation and nuclear localization of Nrf2 protein ( Figure 1C).
Furthermore, CDDO-EA increased the presence of HO-1 in the cytosol 48 hours after MCAO ( Figure 1D). Together, these data indicate that CDDO-EA stimulates Nrf2 protein expression and activation, evidenced by the increased protein expression levels of the Nrf2 target, HO-1.

| CDDO-EA improves neurological score and reduces neuronal death in MCAO mice
Acute neurological function was assessed in mice 24 and 48 hours after MCAO or sham surgery, using a modified Longa method.
Compared to the sham-operated group, neurological scores

| CDDO-EA primes microglia/macrophages toward an M2 phenotype in MCAO brain
To investigate whether CDDO-EA impacts the phenotypic polarization of microglia/macrophages following MCAO, the representative M1 phenotype marker protein (CD16) and the M2 phenotype

| CDDO-EA increases HO-1 expression and promotes M2 polarization in LPS-stimulated microglia
The  Figure 4E). Together, these data indicate that CDDO-EA directly impacts microglial cultures exposed to an inflammatory stimulus by increasing HO-1 expression, promoting polarization toward an M2 phenotype, and enhancing phagocytic activity, all dependent at least in part on HO-1 activity.  Under physiological conditions, the transcription factor Nrf2 is inactive and binds to Kelch-like erythroid cell-derived protein with CNC homology-associated protein 1 (Keap1) in cytoplasm. 21,22 Upon oxidative stress, Nrf2 is activated and released from Keap1, then translocates to nucleus and binds to ARE, and activates expres-   Further studies are warranted to explore the effect of CDDO-EA on long-term functional recovery after stroke.

CO N FLI C T O F I NTE R E S T
The author(s) declared no potential conflicts of interest for the research, authorship, and publication of this article.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data sharing not applicable-no new data generated.