CPI-1189 protects neuronal cells from oxygen glucose deprivation/re-oxygenation-induced oxidative injury and cell death

Oxygen glucose deprivation (OGD)/re-oxygenation (OGDR) induces profound oxidative injury and neuronal cell death. It mimics ischemia-reperfusion neuronal injury. CPI-1189 is a novel tumor necrosis factor alpha-inhibiting compound with potential neuroprotective function. Here in SH-SY5Y neuronal cells and primary murine cortical neurons, CPI-1189 pretreatment potently inhibited OGDR-induced viability reduction and cell death. In OGDR-stimulated neuronal cells, p38 phosphorylation was blocked by CPI-1189. In addition, CPI-1189 alleviated OGDR-induced reactive oxygen species production, lipid peroxidation, and glutathione consumption. OGDR-induced neuronal cell apoptosis was also inhibited by CPI-1189 pretreatment. Furthermore, in SH-SY5Y cells and cortical neurons, CPI-1189 alleviated OGDR-induced programmed necrosis by inhibiting mitochondrial p53-cyclophilin D-adenine nucleotide translocase 1 association, mitochondrial depolarization, and lactate dehydrogenase release to the medium. In summary, CPI-1189 potently inhibited OGDR-induced oxidative injury and neuronal cell death.


INTRODUCTION
Ischemic stroke is a major cause of human morbidities and mortalities around the world [1,2]. The prevalence of this disease is rising in recent years [1,2]. It is therefore important to further understand the pathological mechanisms of neuronal cell injury in ischemic stroke [3,4], and to develop novel therapy strategies [2,5,6].

AGING
Whether CPI-1189 can protect neuronal cells from OGDR-induced oxidative injury remains unknown.

Figure 1. CPI-1189 protects neuronal cells from OGDR-induced cell death. SH-SY5Y neuronal cells (A-C) or primary murine cortical
neurons (D-F) were pretreated for 1h with CPI-1189 (at applied concentrations) and subjected to OGDR procedure, cells were cultured for applied time periods, cell viability, cell death and p38 activation were tested by CCK-8 (A-D), Trypan blue staining (B-E) and Western blotting (C-F) assays, respectively. "Mock" stands for neuronal cells placed in norm-oxygenated regular medium containing glucose (same for all Figures). Quantified values were mean ± standard deviation (SD, n=5). * P < 0.05 vs. "Mock" cells. # P < 0.05 vs. cells with OGDR stimulation but "DMSO (0.1%)" pretreatment. Experiments were repeated three times, with similar results obtained.

DISCUSSION
CPI-1189 is a compound clinically evaluated as a potential therapy for AIDS patients with dementia [13,17]. Recent in vitro and in vivo studies have proposed the potential neuroprotective property of this compound [14][15][16]. Supporting its potential function in protecting neurons, we found that CPI-1189 pretreatment, at only nM concentrations, potently inhibited OGDR-induced viability reduction and death in SH-SY5Y cells and murine cortical neurons. CPI-1189 blocked OGDRinduced p38 activation. This compound was also neuroprotective in p38α-KO SH-SY5Y cells and p38inhibited cortical neurons, indicating the possibility of p38-independent mechanisms.
In the present study, we show that CPI-1189-induced anti-OGDR activity was associated with ROS scavenging. CPI-1189 largely inhibited OGDR-induced ROS production, lipid peroxidation, and GSH consumption in SH-SY5Y cells and cortical neurons. Importantly, OGDR-induced neuronal cell apoptosis, the consequence of oxidative injury [21,22,25], was inhibited by CPI-1189 as well. Therefore, ROS scavenging should be one important mechanism of CPI-1189 protecting against OGDR.

AGING
Besides apoptosis OGDR can also provoke programmed necrosis in neuronal cells. Wang et al., found that OGDR induced NKILA (NF-κB Interacting LncRNA) upregulation to promote neuronal cell necrosis [25]. SphK1 activation by K6PC-5 inhibited OGDR-induced programmed necrosis in neuronal cells [22]. Here in SH-SY5Y cells and murine cortical neurons, CPI-1189 suppressed OGDR-induced programmed necrosis by inhibiting mitochondrial p53-CyPD-ANT1 association, mitochondrial depolarization, and LDH release to the medium. These results suggested a novel mechanism (inhibition of programmed necrosis) of anti-OGDR by CPI-1189. Future studies with concurrent inhibition of OGDRinduced cell necrosis and apoptosis should explain the superior neuroprotective activity by CPI-1189.

Cell culture
SH-SY5Y neuronal cells were provided by Dr. Di [32] and were cultured as described [32]. SH-SY5Y cells were differentiated by the incubation in BDNF plus glutamine medium (serum free) [32]. The primary murine cortical neurons were also provided by Dr. Di, and were cultured using previously described protocols [32]. At day-10 (DIV-10), over 95% of cells were cortical neurons. The protocols of using primary murine cells were approved by the Ethics Committee and IACUC of authors' institution.

Cell viability
Cell Counting Kit-8 (CCK-8, Dojindo Laboratories, Kumamoto, Japan) was utilized to test cell viability. Neuronal cells were seeded into 96-well plates at 4, 000 cells per well. After treatment, neuronal cells were incubated with CCK-8 reagent for 3h. In each well, CCK-8 optical density (OD) was tested at 450 nm.

Cell death
Neuronal cells were seeded into 96-well plates at 4, 000 cells per well. Following treatment, dead cells were positively stained with Trypan blue. The ratio was recorded by an automatic cell counter (Roche, Shanghai, China).

Lactate dehydrogenase (LDH) assay
Neuronal cells were seeded into six-well plates at 1×10 5 cells per well. With the applied treatment, LDH contents in culture medium were analyzed through a two-step enzymatic reaction LDH assay kit (Takara, Tokyo, Japan), which were then normalized to total LDH contents.

OGD/re-oxygenation
As reported [7], neuronal cells were placed in an airtight chamber with a continuous flux of gas (95% N2/5% CO2). The chamber was sealed and the cells were incubated under OGD for 4h. Cells were then re-oxygenated (OGDR) and cultured in regular medium for applied time period. "Mock" neuronal cells were placed in normoxygenated with regular medium containing glucose.

Lipid peroxidation assay
Following treatment, thiobarbituric acid reactive substances (TBAR) assay was carried out to examine the cellular lipid peroxidation contents. The detailed protocol was reported before [21,33].

Western blotting
Protocols for Western blotting were described previously [34]. In brief, 30 μg protein lysates per treatment were loaded to 10-12% SDS-PAGE gels and transfected to PVDF blots. The blots were then blocked and incubated with the applied primary and secondary antibodies. ECL reagents were utilized to examine the targeted protein band. The ImageJ software (NIH) was utilized to quantify the protein band.

TUNEL (terminal deoxynucleotidyl transferase dUTP nick end labeling) assay
Neuronal cells were seeded into12-well plates (at 5 × 10 4 cells per well). Following treatment, a TUNEL In Situ AGING Cell Death Detection Kit (Roche) was applied to measure apoptotic nuclei. Nuclei were co-stained with TUNEL and DAPI. Cells were then visualized under a confocal microscope (Leica). For each treatment, 1000 cells in five random views (1×100 magnification) were counted to calculate the average TUNEL ratio (% vs. DAPI).

ROS assay
Neuronal cells were seeded into six-well plates. Following treatment, cells were stained with fluorescent dye CellROX (Sigma, 7.5 μM for 1h). CellROX intensity was tested by a fluorescence spectrofluorometer (Molecular Devices, San Jose, CA, USA). Representative CellROX fluorescence images were presented.

DNA breaks
Neuronal cells were seeded into six-well plates. Following treatment, single strand DNA (ssDNA) contents, indicating DNA breaks, were measured through ssDNA ApoStrandTM ELISA kit (BIOMOL International, Plymouth Meeting, PA, USA). ELISA OD was examined at 450 nm.

CRISPR-Cas9-induced p38α knockout (KO)
A CRISPR-Cas9-p38α-KO-GFP-puromycin construct was designed by Genechem (Shanghai, China) and transfected into cultured SH-SY5Y cells (in polybrene medium). GFP-positive SH-SY5Y cells were sorted by FACS and distributed to 96-well plates to achieve single cells. Stable cells were further selected by puromycin. In the stable cells, p38α KO was verified by Western blotting. The target DNA sequence of p38α sgRNA is TGGACGTTTTTACACCTGCA (PAM: AGG).

Mitochondrial depolarization
JC-1 fluorescence dye aggregates into mitochondria to form green monomers in cells with mitochondrial depolarization [36]. Neuronal cells were seeded into 12well plates. Following treatment, cells were stained with JC-1 (15 μg/mL, Sigma), and then washed and examined under a fluorescence spectrofluorometer (F-7000, Hitachi, Japan) at 488 nm (green). The representative JC-1 fluorescence images integrating both green (at 488 nm) and red (at 625 nm) fluorescence channels were presented as well.

Annexin V-FACS
Neuronal cells were seeded into six-well plates at 1×10 5 cells per well. With the applied treatment, cells were costained with Annexin V (15 μg/mL) and Propidium Iodide (PI, 15 μg/mL), and measured under a FACS machine (BD, Shanghai, China). Annexin V-positive cells (apoptotic cells) were gated and its ratio was recorded.

Statistics
Data were expressed as means ± standard deviation (SD). Statistical analyses among different groups were tested by one-way analysis of variance (ANOVA) and Tukey's post hoc multiple comparison tests (SPSS 23.0, SPSS, Chicago, IL, USA). The Student t test (Excel2007) was applied to compare statistical difference between two groups. P< 0.05 was considered as statistically significant.

CONFLICTS OF INTEREST
The authors declare that they have no conflicts of interest.

FUNDING
This work was generously supported by grants from Affiliated Hospital of Yangzhou University.