Interleukin-13 Propagates Prothrombin Kringle-2-Induced Neurotoxicity in Hippocampi In Vivo via Oxidative Stress

The present study investigated expression of endogenous interleukin-13 (IL-13) and its possible function in the hippocampus of prothrombin kringle-2 (pKr-2)-lesioned rats. Here we report that intrahippocampal injection of pKr-2 revealed a significant loss of NeuN-immunopositive (NeuN+) and Nissl+ cells in the hippocampus at 7 days after pKr-2. In parallel, pKr-2 increased IL-13 levels, which reached a peak at 3 days post pKr-2 and sustained up to 7 days post pKr-2. IL-13 immunoreactivity was seen exclusively in activated microglia/macrophages and neutrophils, but not in neurons or astrocytes. In experiments designed to explore the involvement of IL-13 in neurodegeneration, IL-13 neutralizing antibody (IL-13Nab) significantly increased survival of NeuN+ and Nissl+ cells. Accompanying neuroprotection, immunohistochemical analysis indicated that IL-13Nab inhibited pKr-2-induced expression of inducible nitric oxide synthase and myeloperoxidase within activated microglia/macrophages and neutrophils, possibly resulting in attenuation of reactive oxygen species (ROS) generation and oxidative damage of DNA and protein. The current findings suggest that the endogenous IL-13 expressed in pKr-2 activated microglia/macrophages and neutrophils might be harmful to hippocampal neurons via oxidative stress.


Introduction
Reactive oxygen species/reactive nitrogen species (ROS/RNS), such as superoxide anion (O 2 − ), nitric oxide (NO), peroxynitrite (ONOO -) and hydrogen peroxide (H 2 O 2 ), are byproducts during metabolic processes from endogenous sources including mitochondria, peroxisomes, and the endoplasmic reticulum [1][2][3]. Either excess production of ROS/RNS or a deficiency of antioxidants in the system cause oxidative/nitrosative stress, which can potentially cause damage to cellular biomolecules, including lipids and membranes, proteins, and DNA [2,4]. Especially in the central nervous system (CNS), regulating the oxidative/nitrosative stress is essential for brain cell survival and is associated with the initiation and progression of neurodegenerative disease, including Alzheimer's disease (AD), Parkinson's disease (PD), amyotrophic lateral sclerosis (ALS) and stroke [1,[4][5][6]. Accordingly, regulation of molecular mechanisms of oxidative/nitrosative stress is still an unmet need for developing potential therapeutic targets against neurodegeneration.
Microglia, a subset of macrophages in the CNS, play a variety of roles in CNS-related immune response and the maintenance of brain homeostasis [7][8][9]. In pathological conditions, resting microglia develop into an activated state with amoeboid morphology and mediate the inflammatory responses associated with altered diverse cell-surface receptor expression [10,11], increased phagocytic activities [12], and production of proinflam-matory cytokines and oxidative/nitrosative stress [13,14]. Furthermore, microglia are known to participate in ROS/RNS production during neurodegenerative process in AD and dementia [8,15].
Interleukin-13 (IL-13) is a well-known anti-inflammatory cytokine produced by Th2 cells, mast cells, basophils, eosinophils, and B cells. IL-13 modulates the function of mast cells, macrophages, eosinophils, and B cells, and inhibits production of inflammatory cytokines [23,24]. In the CNS, microglia and neurons have been identified as the source of IL-13 [25]. IL-13 has a neuroprotective effect which works by modulating microglial polarity to an anti-inflammatory state [23,26] and promoting death of activated microglia [27,28]. By contrast, IL-13 can induce immune-mediated neurotoxicity through producing proinflammatory cytokines and ROS/RNS stress in activated microglia/macrophages and neutrophils, indicating that IL-13 is harmful to neurons [17,19,29]. Thus, the current study investigated whether activated microglia/macrophages and neutrophils-derived from IL-13 could contribute to pKr-2-induced hippocampal neuronal death by regulating oxidative/nitrosative stress in vivo.

pKr-2 Induces Degeneration of Hippocampal Neurons and Activation of Microglia/Macrophages and Neutrophils in the Hippocampus In Vivo
pKr-2 (48 µg) or PBS as a control was unilaterally injected into the CA1 layer of the hippocampus of rats. Seven days later, the brains were removed, and sections were processed for immunostaining for neuronal nuclei (NeuN) to detect general neurons or Nissl staining. There was a considerable loss of NeuN-immunopositive (NeuN + ) ( Figure 1E) and Nissl-stained cells ( Figure 1F) at 7 days in the pKr-2-injected hippocampus compared with PBS-injected controls ( Figure 1A,B). In parallel, pKr-2-activated microglia/macrophages and neutrophils, visualized by immunohistochemical staining using antibody against the complement receptor type 3 (OX-42) ( Figure 1G) and the major histocompatibility complex class II antigens (OX-6) ( Figure 1H), were also observed in the CA1 layer of the hippocampus, where degeneration of hippocampal neurons was found compared with PBS-injected controls ( Figure 1C,D). pKr-2 triggered profound activation of microglia/macrophages and neutrophils in the hippocampus, with enhanced staining and activated morphology (large cell bodies with short, thick, or no processes) in OX-42 + cells ( Figure 1G) compared with PBS-injected control ( Figure 1C) showing the resting state of OX-42 + cells (small cell bodies with ramified processes). These data also showed that pKr-2 treatment resulted in many OX-6 + cells in the CA1 layer of the hippocampus ( Figure 1H), whereas few of OX-6 + cells were seen in PBS-treated controls ( Figure 1D).

pKr-2-Induced IL-13 Expression Is Localized within Activated Microglia/Macrophages and Neutrophils in the Hippocampus In Vivo
We examined whether intrahippocampal injection of pKr-2 induced expression of interleukin 13 (IL-13) in the hippocampus. Analysis by immunohistochemical staining demonstrated that pKr-2 induced IL-13 expression as early as 1 day after injection, with maximal levels reached 3 days after injection and sustained up to 7 days after injection ( Figure 2B-E). In contrast to pKr-2-injected hippocampus, IL-13 expression was inconsiderable in PBS-injected hippocampus as control (1 day; Figure 2A,E). To identify the cell type expressing endogenous IL-13, double immunofluorescence staining was performed in combination with IL-13 with tomato lectin (TL) + microglia/ macrophages and neutrophils, NeuN + neurons, or glial fibrillary acidic protein (GFAP) + astrocytes at 12 h after pKr-2 injection. Endogenous IL-13 was mainly expressed in TL + microglia/macrophages and neutrophils ( Figure 2F-H), but neither in NeuN + neurons ( Figure 2I-K) nor GFAP + astrocytes ( Figure 2L-N) in the CA1 layer of hippocampus in vivo.

Neurotoxic Action of IL-13 on Degeneration of Hippocampal Neurons via iNOS and MPO in the CA1 Layer of Hippocampus In Vivo
To elucidate the physiological functions of IL-13 mainly expressed in microglia/ macrophages and neutrophils after pKr-2 injection, we examined whether pKr-2-induced degeneration of hippocampal neurons could be affected by treatment of IL-13 neutralizing antibody (IL-13Nab) for blocking the function of IL-13. For this purpose, IL-13Nab was unilaterally co-injected with pKr-2 into the CA1 layer of hippocampus ( Figure 3E,F). Seven days later, NeuN immunostaining (A,C,E) and Nissl (B,D,F) staining demonstrated protective effects of IL-13Nab on hippocampal neurons in vivo. When NeuN + cells on the ipsilateral side were quantified, it was found that IL-13Nab significantly increased the number of NeuN + cells in the CA1 layer of the hippocampus compared with pKr-2 only ( Figure 3I). As controls, IL-13Nab alone ( Figure 3I) and nonspecific goat IgG in the absence ( Figure 3G,H) or presence of pKr-2 [16] (data not shown) did not affect neuronal survival [16,19,29,30], similar to that observed with PBS or pKr-2 only.
Activated microglia/macrophages and neutrophils produce inducible nitric oxide synthase (iNOS) and myeloperoxidase (MPO), which induces oxidative/nitrosative stress, resulting in neurodegeneration [4,16]. Thus, we examined whether intrahippocampal injection of pKr-2 induced expression of iNOS and MPO in activated microglia/macrophages and neutrophils. pKr-2 (48 µg) or PBS as a control was unilaterally injected into the CA1 layer of the hippocampus of rats. Three days later, immunohistochemical analysis showed that pKr-2 significantly increased expression of iNOS ( Figure 4B,G) and MPO ( Figure 4E,G) within OX-42 + microglia/macrophages and neutrophils, respectively, compared to PBS-injected control ( Figure 4A,D,G). To verify involvement of iNOS and MPO with IL-13-induced neurotoxicity, IL-13Nab was unilaterally co-injected with pKr-2 into the CA1 layer of hippocampus. Three days later, immunohistochemical analysis demonstrated that treatment of IL-13Nab significantly reduced pKr-2-triggered expression of iNOS ( Figure 4C,G) and MPO ( Figure 4F,G).

Discussion
The cytokine IL-13 is secreted protein and can play both beneficial and detrimental roles in the CNS [23,26]. A study in humans with multiple sclerosis found that high levels of IL-13 in the cerebral spinal fluid might exert a neuroprotective effect by enhancing gamma aminobutyric acid over glutamate transmission [31]. Intracerebral injection of IL-13 reduced amyloid deposition and improved spatial learning and memory in an AD transgenic mouse model, indicating neuroprotection [32]. By marked contrast, IL-13 contributed to degeneration of cortical neurons in pKr-2-treated rat cortex [17] and nigral dopamine neurons in mice lacking IL-13 receptor alpha 1 [33], suggesting a neurotoxic action of IL-13. In addition, intrahippocampal injection of amyloid beta 1-42 [19] and thrombin [29] produced IL-13-mediated neurodegeneration in CA1 layer of rat hippocampus. In the current study, blocking action of IL-13 by IL-13Nab results in survival of hippocampal neurons against pKr-2-induced neurotoxicity. These data suggest that endogenous IL-13 is an important determinant in pKr-2-induced neurotoxicity in the CA1 layer of hippocampus in vivo.
Microglia, resident immune cells of the CNS, are associated with a variety of neuropathologies [13,34,35]. Upon activation, microglia contribute to oxidative/nitrosative stress by producing considerable ROS/RNS, which cause the death of microglia as well as neurons [5,30,[36][37][38]. Activated microglia can induce increased levels of iNOS, one of the major sources for excessive production of ROS/RNS and resultant oxidative damage to proteins, lipids, and DNA in neurodegenerative diseases [3,4,6]. Many in vivo studies, including ours, showed iNOS-induced generation of ROS/RNS in the cerebral cortex, substantia nigra, and hippocampus treated with thrombin, pKr-2, LPS, and betaamyloid   [17,19,39,40]. Accompanying the attenuation of neurotoxicity, treatment with IL-13Nab attenuated expression of iNOS, ROS/RNS, and oxidative damages in DNA and proteins in pKr-2-treated hippocampus in vivo, indicative of IL-13-induced oxidative/nitrosative damage in neurons. This is consistent with our findings that neutralization of IL-13 by IL-13Nab significantly reduced the expression of iNOS and/or ROS/RNS from activated microglia in pKr-2-treated cortex [17] and in hippocampus treated with thrombin- [29] or beta-amyloid 1-42 [19] and the resultant survival of neurons in vivo.
MPO, a critical inflammatory enzyme, is present in infiltrated neutrophils, monocytes/macrophages, and neurons as well as activated microglia and astrocytes [5,[41][42][43]. Activation of MPO can catalyze the reaction of chloride and H 2 O 2 to produce hypochlorous acid (HOCl), which is a crucial cytotoxic factor and has oxidative activity able to react with DNA, proteins, and lipids [44,45]. MPO mediates oxidative stress by promoting the production of ROS/RNS, which can modulate the inflammation-related signaling pathways in microglia and cause oxidative damages to DNA, proteins and lipids, resulting in neurodegeneration in vivo [39,[46][47][48][49][50][51]. Many reports, including ours, have shown that various antioxidants, such as compounds from medicinal herbs, capsaicin, and ethyl pyruvate, inhibited expression of MPO-derived ROS/RNS, reduced oxidative damage, and prevented neurodegeneration [42,52,53]. Our data demonstrated that treatment with IL-13Nab significantly mitigated pKr-2-induced MPO expression with attenuation of ROS/RNS levels and oxidative damages in the hippocampus in vivo, resulting in survival of neurons. This indicates that like the iNOS, IL-13 regulates the production of ROS/RNS derived from MPO.
In summary, neuroinflammation is associated with glial activation and glia-derived oxidative stress and inflammatory molecules, resulting in pathology of neurodegenerative diseases. Among them, oxidative/nitrosative stress (the imbalance between production and destruction of ROS/RNS) has been considered as a major risk factor of neurodegeneration. Here, we show that microglia/macrophages and neutrophils-derived from IL-13 plays a critical role on pKr-2-induced neurodegeneration of hippocampal CA1 areas in vivo. IL-13 neutralization attenuates levels of iNOS and MPO expression on microglia/macrophages and neutrophils, ROS production, and oxidative damages, resulting in neuronal survival. This suggests that endogenous IL-13 might be associated with neuropathological processes in which neuroinflammation and/or oxidative stress are involved.

Animals
Animal care, handling, and all experiments were performed according to the guideline set by Committee on Animal Research of Kyung Hee University (KHU-ASP-20-235; 30 June 2020). Female Sprague-Dawley rats were kept in temperature-(21-23 • C) and humiditycontrolled conditions under a 12-h light/dark cycle. Throughout animal housing and the experiment, all animals had free access to food and water. All experiments were carried out to minimize animal suffering and the minimal number of animals necessary to produce significant scientific data were used.

Immunohistochemistry (IHC) and Nissl Staining
Animals were transcardially perfused with physiological saline containing 0.5% sodium nitrate and heparin (10 U/mL) and fixed by 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). After post-fixing with 30% sucrose solution, rat brains were coronally cut at 40-µm-thick using sliding microtome. Serial coronal sections were processed for immunohistochemical staining as previously described [16,17]. Briefly, brain sections were rinsed with PBS 2 times, quenched in PBS containing 3% H 2 O 2 , rinsed with PBS 3 times, blocked in PBS containing 1% bovine serum albumin (BSA; Millipore Corporation., Kankakee, IL, USA) with 0.2% Triton X-100 (Sigma-Aldrich., St. Louis, MO, USA), and then incubated with primary antibodies at room temperature on a shaker overnight (Table 1). After rinsing in PBS containing 0.5% BSA 2 times, sections were incubated with biotin-conjugated mouse IgG (Seracare., Milford, MA, USA) for 1 h, rinsed with PBS containing 0.5% BSA 2 times, and processed with avidin-biotin complex kit (Vector Laboratories., Burlingame, CA, USA). After that, sections were visualized by the mixture of 0.05% 3,3 -diaminobenzidine tetrahydrochloride (DAB; Sigma-Aldrich., St. Louis, MO, USA) with 0.003% H 2 O 2 . Finally, labeled brain sections were mounted on adhesive microscope slides and dried. A series of sections were viewed under a bright-field microscope (Olympus Optical., Tokyo, Japan). The antibodies used for immunohistochemical staining (IHC) are shown below (Table 1). For Nissl staining, brain sections were mounted on adhesive microscope slides and dried. After that, sections were stained with 0.5% cresyl violet acetate solution (Sigma-Aldrich., St. Louis, MO, USA), dehydrated and viewed under a bright-field microscope.

Immunofluorescence (IF) Staining
Serial coronal sections (40-µm-thick) were processed for immunofluorescence staining as previously described [16,17]. Briefly, brain sections were mounted on adhesive microscope slides and dried. After that, sections were rinsed with PBS 3 times, blocked in PBS containing 1% BSA with 0.2% Triton X-100, and then incubated with primary antibodies at 4 • C overnight (Table 2). After rinsing in PBS containing 0.5% BSA 2 times, sections were incubated with the appropriate fluorescence-conjugated secondary antibodies for 1 h and rinsed with PBS 3 times. Finally, labeled brain sections were covered with the Vectashield mounting medium (Vector Laboratories., Burlingame, CA, USA) for fluorescence with DAPI and analyzed under a confocal microscope (Carl Zeiss., Oberkochen, Germany). The antibodies used for IF are shown below ( Table 2).  [16,17]. DHE freely permeate the cell membrane and react with superoxide anions to form a red fluorescent product (2-hydroxyethidium) [54]. At 3 days after pKr-2 injection, hydroethidine (1 mg/kg, reconstituted in PBS containing 1% dimethyl sulfoxide) (Invitrogen., Carlsbad, CA, USA) was intravenously injected into the rat tail vein. After 45 min, rats were transcardially perfused and fixed. After postfixing, rat brains were coronally cut at 40-µm-thick using sliding microtome and mounted on adhesive microscope slides. The labeled brain sections were covered with the Vectashield mounting medium (Vector Laboratories., Burlingame, CA, USA) and the oxidized hydroethidine product (ethidium) was investigated by confocal microscope (Carl Zeiss., Oberkochen, Germany).

Image J Analysis
Imaging data obtained from bright-field microscope and confocal microscope were analyzed as pixel values using the Image J program (National Institutes of Health, Washington County, MD, USA). Image J was used to quantify the chromogenic signal intensity of images adjusted above the threshold to rule out unspecific background signal. For measurement of IL-13, ROS, 8-OHdG, and nitrotyrosine, images were transformed to 8-bit grayscale and adjusted at the endpoint of threshold histogram. Then the pixel value was quantified and normalized by unstained area. For measurement of iNOS and MPO within OX-42 + microglia/macrophages and neutrophils, the adjusted images of each channel are colocalized using a colocalization plugin, then the overlaid signal are quantified.

Statistical Anaylsis
Statistical analyses were performed using GraphPad Prism 5 (GraphPad Software., San Diego, CA, USA). Results are expressed as mean ± standard error of the mean (SEM). For statistical evaluation between two groups, p value was assessed by Student's unpaired t-test. For comparison of multiple groups, p value was assessed by one-way of variance (ANOVA) with Newman-Keuls analysis. p < 0.05 was considered to indicate statistical significance.

Data Availability Statement:
The data used to support the funding of this study are all provided within the article.

Conflicts of Interest:
The authors declare no conflict of interest.