Neuroprotective effects of tannic acid in the postischemic brain via direct chelation of Zn2+

ABSTRACT Tannic acid (TA) is a polyphenolic compound that exerts protective effects under pathological conditions. The diverse mechanisms of TA can exert beneficial anti-oxidative, anti-inflammatory, and anti-cancer effects. Herein, we reported that TA affords robust neuroprotection in an animal model of stroke (transient middle cerebral artery occlusion; tMCAO) and exhibits Zn2+-chelating and anti-oxidative effects in primary cortical neurons. Following tMCAO induction, intravenous administration of TA (5 mg/kg) suppressed infarct formation by 32.9 ± 16.2% when compared with tMCAO control animals, improving neurological deficits and motor function. We compared the chelation activity under several ionic conditions and observed that TA showed better Zn2+ chelation than Cu2+. Furthermore, TA markedly decreased lactate dehydrogenase release following acute Zn2+ treatment and subsequently reduced the expression of p67 (a cytosolic component of NADPH oxidase), indicating the potential mechanism underlying TA-mediated Zn2+ chelation and anti-oxidative effects in primary cortical neurons. These findings suggest that anti-Zn2+ toxicity and anti-oxidative effects participate in the TA-mediated neuroprotective effects in the postischemic brain.


Introduction
Zn 2+ , an important trace element essential for human nutrition, acts as an intracellular secondary messenger that regulates several transcription factors and enzymes under various physiological and pathological conditions (Frederickson et al. 2005). Zn 2+ has been detected in multiple brain regions, including the hippocampus, cerebral cortex, and hypothalamus (Donaldson et al. 1973). Zn 2+ released from synaptic vesicles modulates receptors and ion channels that regulate synaptic transmission and neuronal excitability, thereby maintaining brain function. Acute exposure to Zn 2+ increases neuronal cell death by inhibiting glycolysis and ATP production and elevating levels of reactive oxygen species (ROS) by activating NADPH oxidase in neurons (Noh and Koh 2000;Sheline et al. 2000). Moreover, excessive Zn 2+ intake activates poly ADP ribosyl polymerase-1 (PARP-1) and induces neuronal cell death via NAD depletion (Ha and Snyder 1999;Kim et al. 2002;Sheline et al. 2003). The membrane-permeable zinc chelator N,N,N',N'-tetrakis (2-pyridylmethyl) ethylenediamine (TPEN) was found to reduce infarct formation, improve neurological deficits, and decrease Zn 2+ accumulation, further suppressing PARP-1 activation (Zhao et al. 2014) Tannins, categorized into hydrolyzable and nonhydrolyzable (condensed) tannins, are widespread in fruits, cereals, legumes, herbs, green tea, red wine, and vegetables (Chung et al. 1998). Proanthocyanidin (condensed tannin) reportedly exerts cardioprotective effects against ischemia/reperfusion injury by scavenging ROS and inhibiting pro-apoptotic transcription factors and genes JNK-1 and c-Jun (Sato et al. 1999). Punicalagin (hydrolyzable tannin) can suppress cerebral ischemic injury by inducing anti-inflammatory effects (Yaidikar and Thakur 2015). Furthermore, geraniin (hydrolyzable tannin) can upregulate nuclear factor-E2 related factor 2 (Nrf-2), which contributes to enhanced anti-oxidative enzyme expression . Hydrolyzable tannins such as ellagitannins and gallotannins, comprising a general structural motif of galloyl groups, exhibit anti-oxidative and neuroprotective effects (Chung et al. 1998). Moreover, hydrolyzable tannins act as chelators to regulate enzyme activity (Chen et al. 2019). Tannic acid (TA) is a water-soluble polyphenol compound and commercial tannin. TA reportedly exerts anti-carcinogenic, anti-oxidative, and anti-inflammatory effects. TA can attenuate skin cancer induced by 7,12-dimethylbenz[a]anthracene (DMBA) and croton oil by suppressing oxidative and inflammatory responses (Majed et al. 2015). Pretreatment with TA was shown to suppress infarct formation and improve neurological deficits in the middle cerebral artery occlusion (MCAO) stroke model. Depletion of antioxidant enzyme activity was markedly ameliorated during infarct progression in the TA-pretreated MCAO group (Ashafaq et al. 2016). Intraperitoneal administration of TA can upregulate Nrf2, which activates antioxidant enzymes, including heme oxygenase-1 (HO-1) and superoxide dismutase (SOD), in animal models of traumatic brain injury and nephrotoxicity Salman et al. 2020). In addition, TA can attenuate ultraviolet (UV)-induced inflammation, which induces interleukin (IL)−6 production and phosphorylates STAT3 in the epithelial cells (Chou et al. 2012). Moreover, TA could inhibit NF-κB signaling in atopic dermatitis (AD) NC/Nga mice (Karuppagounder et al. 2015). Oral administration of TA afforded marked protection against oxidative stress, given that TA acts as a scavenger and chelator in iron-dextran-induced hepatotoxicity (Basu et al. 2018).
In the present study, we revealed that TA exerted robust neuroprotection against ischemic stroke, markedly suppressing infarct formation and improving behavioral deficits. In addition, we found that TA directly chelated Zn 2+ and inhibited Zn 2+ -mediated ROS production, thereby suppressing neuronal cell death. Therefore, TA provides profound neuroprotection in the postischemic brain, and this neuroprotective effect is partially attributed to its role as a chelator.

Surgical procedures for MCA occlusion
In the present study, animal comfort and pain minimization were carefully considered. The animal research protocol was reviewed and approved by the Inha University-Institutional Animal Care and Use Committee prior to study initiation at Inha University. This study was performed in accordance with the Guide for the Care and Food and water were provided ad libitum. MCAO was performed as previously described (Yu et al. 2021). Briefly, SD rats (250-300 g) were anesthetized with isoflurane (2% induction, 1.5% maintenance) using an oxygen/nitrous oxide (30/70%) mixture. MCAO was performed for 60 min using nylon suture (4-0; AILEE, Busan, South Korea), followed by reperfusion. A laser Doppler flowmeter (Periflux System 5000; Perimed, Jarfalla, Sweden) was used for real-time monitoring of regional cerebral blood flow. During surgery, the rectal temperature of the rat was maintained within 37.0 ± 0.5°C using a heat pad.

Infarct volume assessment
To measure brain infarct volume, brains were dissected coronally into 2 mm slices using a metallic brain matrix (RBM-40,000, ASI, Springville, UT, USA) 2 days post-MCAO. Brain slices were quickly incubated in saline containing 2, 3, 5-triphenyl tetrazolium chloride (TTC, 2%) at 37°C for 15 min and fixed with 4% paraformaldehyde (PFA). The infarcted brain tissue areas were analyzed using the Scion Image program (Scion Image program, Frederick, MD, USA). To correct for edema and shrinkage, the infarct volume was calculated using the following formula: (contralateral hemisphere volume × measured injury volume/ipsilateral hemisphere volume).

Evaluation of modified neurological severity scores
Modified neurological severity scores (mNSS) were used to evaluate neurological deficits 2 days post-MCAO. The total mNSS score was 18 points (normal, 0; maximal deficit, 18) and consisted of sensory, motor, reflex, and balance tests (Chen et al. 2001).

Wire hanging test
The wire hanging test, which measures forelimb strength and grasping ability, was performed as previously described (Rakhunde et al. 2014). Briefly, SD rats were suspended by their forelimbs on a steel wire (50 cm long, 2 mm diameter), and the latency to fall was measured using a stopwatch until a cut-off time of 60 s was reached.

Rotarod test
One day before MCAO, the SD rats were trained on a rotarod apparatus (Daejon Instruments, Seoul, Korea) at a speed of 3 rpm until they could remain on the rotating spindle for 180 s. Two days post-MCAO, the latency times on the spindle were measured at spindle speeds of 5 and 10 rpm, with a 1 h rest period after testing at 5 rpm.

Cortical primary neuron culture
Primary cortical cells were obtained from rat cortices (E15.5) and cultured as previously described (Kim et al. 2020). Cortical cells (4×10 5 cells per well) were plated on poly-d-lysine (100 μg/ml)-and laminin (100 μg/ml)coated plates. The cultured cortical cells were maintained without antibiotics in modified Eagle's medium (MEM) containing 5% fetal bovine serum (FBS), 5% horse serum, 21 mM glucose, and 2 mM glutamine. On day in vitro (DIV) 7, cytosine arabinofuranoside (10 μΜ) was added and maintained for two days to halt microglial growth. FBS and glutamine were not supplemented from DIV 7, and the medium was replaced every other day after DIV 7. Cultures were used on DIV 12-14. Primary cortical cells were treated with 300 μM Zn 2+ for 15 min in HCSS (HEPES-controlled salt solution) and then incubated with 21 mM glucose-containing MEM.

Statistical analysis
Statistical analysis was performed using analysis of variance followed by the Newman-Keuls test. The analysis was performed using the GraphPad PRISM software 5.0 (GraphPad Software, Inc., La Jolla, CA, USA). Results are presented as the mean ± standard error of the mean (SEM), and statistical significance was set at α = 0.05.

TA suppressed zn 2+ -induced neuronal cell death in primary cortical culture
To explore the neuroprotective mechanisms of TA, we examined whether TA suppresses lactate dehydrogenase (LDH) release during Zn 2+ -induced neuronal cell death. An LDH cytotoxicity assay was conducted to examine the cytotoxic potential of Zn 2+ on primary cortical neurons (Fig. S1). To determine the characteristics of TA, primary cortical neurons underwent pre-, co-, and post-treatment with TA. Following acute incubation with Zn 2+ , co-treatment with TA (100 μM) or EDTA Figure 1 . Neuroprotective effects of tannic acid in the postischemic brain. (A) TA (1, 2.5, or 5 mg/kg, i.v.) was administered 1 h after MCAO, and mean infarct volume was measured by TTC staining. Representative images of infarctions in coronal brain slices are shown (A), and quantitative results are presented as the mean ± standard error of the mean (SEM) (n = 6-7) (B). (C-F) TA (5 mg/kg, i.v.) was administered 1 h after MCAO. (C) Neurological deficits were evaluated using mNSS 2 days after MCAO. (D) Wire hanging test was performed to assess motor impairment 2 days after MCAO. (E, F). The rotarod test was performed 2 days after MCAO. Latency time was recorded at a spindle speed of 5 and 10 rpm. Results are presented as the mean ± SEM (n = 6-7). * p < 0.05, ** p < 0.01 vs. MCAO group. MCAO, middle cerebral artery occlusion; mNSS, modified neurological severity scores; i.v., intravenously; TA, tannic acid.

Discussion
The findings of the present study suggest that TA could reduce infarct formation and improve neurological deficits in patients with postischemic stroke. Herein, TA exerted neuroprotective effects via Zn 2+ chelation and reduced ROS generation. Ashafaq et al. have suggested the neuroprotective effects of TA in the MCAO model; however, detailed mechanisms explaining TA-mediated inhibition of inflammation or oxidative stress are lacking. A protective mechanism of TA by inhibiting NF-κB has been previously suggested in nephrotoxicity and cardiotoxicity Xue et al. 2020); however, the TA mechanism underlying these effects remains unclear. We speculated that the neuroprotective effects of TA were mainly derived from its chelating effect, given that TA is a water-soluble polyphenol with an 8 gallic acid group and acts as a metal chelator (Strlic et al. 2002). Chelation therapy is a medical treatment used to reduce the toxic effects of metals. Chelating agents easily bind to metal ions to form complex structures that inhibit toxic effects by excreting the formed complexes from intracellular or extracellular spaces. Cerebral ischemia induces substantial intracellular Zn 2+ accumulation, contributing to brain damage (Koh et al. 1996). Ca-EDTA, a high-affinity zinc chelator, can reduce neuronal cell death and attenuate the mitochondrial release of cytochrome C and Caspase-3 activation (Calderone et al. 2004). Furthermore, TPEN, a membrane-permeable zinc chelator, was found to reduce Zn 2+ accumulation in cerebral ischemia animal models by inhibiting PARP-1 activation (Zhao et al. 2014). 2, 2'-ipyridyl, a lipophilic iron chelator, can attenuate apoptotic cell death in photothrombosis-induced focal ischemia (Hoecke et al., 2005). In particular, we found that TA exhibited a strong Zn 2+ chelating effect, which contributed to the suppression of infarct formation and improvement of neurological impairment.
Reportedly, TA exerts anti-oxidative effects by regulating NF-κB and Nrf-2 activity Salman et al. 2020). TA suppresses lipid oxidation, regulates anti-oxidative enzymes in aluminum oxidative neurotoxicity (Tüzmen et al. 2015). In the present study, we found that the anti-oxidative effect of TA was related to Zn 2+ chelation ( Figure 3B), which subsequently suppressed ROS production and expression of the NADPH oxidase subunit ( Figure 4D). This TA-induced Zn 2+ chelation is speculated to be, at least partially, responsible for the neuroprotective effects in the postischemic brain. Chelation of intracellular Zn 2+ can suppress brain damage. The membrane-permeable zinc chelator TPEN plays a role in reducing infarct volume, improving motor function, and suppressing cerebral ischemiainduced blood-barrier damage in the postischemic brain (Zhao et al. 2014;Qi et al. 2016). Interestingly, our study revealed that TA selectively chelated Zn 2+ but not Cu 2+ (Figure 3). However, tannin fractions from walnuts exhibit stronger Cu 2+ chelation than Zn 2+ chelation (Karamać 2009). It is speculated that Zincon (a colorimetric indicator for Zn 2+ and Cu 2+ ) does not undergo chelation with TA, given its higher affinity for Cu 2+ than TA ( Figure 3D).
To determine the pharmacological characteristics of TA, we performed pre-, post-, and co-treatment with TA in a Zn 2+ -induced neuronal cell death model ( Figure 2). We noted the characteristic extracellular chelator activity of TA in acute Zn 2+ toxicity, similar to that of EDTA. In chronic Zn 2+ toxicity, pretreatment with TA reduced LDH release; however, pretreatment with EDTA or TPEN failed to exhibit this effect ( Figure 2D). Given that TA can activate the Nrf2 pathway, pretreatment with TA could afford protection against chronic Zn 2+ toxicity. Interestingly, the duration of TA pretreatment and ZnCl 2 treatment are important factors governing the activation of Nrf2 signaling and protection against Zn 2+ toxicity. In the chronic Zn 2+ treatment, pretreatment with TA was performed for 15 min longer than that in acute Zn 2+ treatment ( Figure 2B, D). NMDA receptor activity induces elevated Zn 2+ levels and ROS generation (Bossy-Wetzel et al. 2004). The activation of NMDA receptors leads to superoxide generation, mainly induced by NADPH oxidase (Brennan et al. 2009), and excitotoxicity, which is mediated by PARP-1 (Mandir et al. 2000). NMDA receptor-induced calcium influx produces prolonged elevation of intracellular Ca 2+ concentrations and triggers signaling pathways (Dingledine et al. 1999), wherein TA can chelate Ca 2+ . However, we only demonstrated that TA could inhibit NMDAinduced LDH release and ROS generation in cortical neurons, and further experiments are warranted (Fig.  S2). In addition, we used oxygen/glucose deprivation (OGD), which mimics ex vivo ischemic stroke. The OGDinduced Zn 2+ influx was reduced by co-treatment with TA or TPEN (Fig. S3). TA appears to exert a chelating effect in extracellular and intracellular environments and activates the Nrf2 pathway Salman et al. 2020).
In summary, TA exerts neuroprotective effects in the postischemic brain. Our findings suggest that TAmediated chelation induces anti-oxidative and anti-Zn 2 + toxicity effects.