Camptothecin regulates microglia polarization and exerts neuroprotective effects via the AKT/Nrf2/HO-1-NF-κB signal axis in vivo and in vitro

Parkinson's disease (PD), the second largest neurodegenerative disease seriously affects human health. Microglia, the main immune cells in the brain participate in the innate immune response in the central nervous system (CNS). Studies have shown that microglia can be polarized into pro-inammatory M1 and anti-inammatory M2 phenotypes. Accumulated evidences suggest that over-activated M1 microglia release pro-inammatory mediators that damage neurons and lead to Parkinson's disease (PD). In contrast, M2 microglia release neuroprotective factors and exert the effects of neuroprotection. Camptothecin (CPT), an extract of the plant Camptotheca acuminate, has been reported to have anti-inammation and antitumor effects. However the effect of CPT on microglia polarization and microglia-mediated inammation responses has not been reported. Therefore, we aim to explore the effect of CPT on microglia polarization and its underlying mechanism on neuroinammation.


Introduction
Parkinson's disease (PD), the most common neurodegenerative disease seriously affects human health. PD is characterized by the loss of dopaminergic neurons in the substantia nigra (SN) of midbrain, and its clinical manifestations are motor dysfunction. At present, it is generally believed that environmental, genetic, physiological aging and other factors can cause PD [1,2]. The speci c pathogenesis is not yet clear. Accumulated evidences have shown that neuroin ammation participated in the occurrence and development of PD [3,4]. Neuroin ammation has a two-way regulation effect. On the one hand, when the central nervous system is damaged by various pathogenic factors, immune cells are activated, and the resulting in ammatory response can kill harmful pathogens, remove abnormally accumulated proteins and cell fragments, and maintain damaged neurons, play a role in protecting neurons. On the other hand, when the in ammatory reaction continues, various harmful factors will be released and accumulate in the brain, such as in ammatory chemokines, reactive oxygen species, excitatory amino acid ions, etc., which cause damage to peripheral neurons and lead to PD further exacerbated [5][6][7]. Therefore, proper regulation of neuroin ammation is of great signi cance for alleviating PD.
Microglia, a type of immune cells that exist in the central nervous system (CNS), have the functions of immune monitoring, phagocytosis of debris, secretion of various factors and antigen presentation [8].
Microglia have three phenotypes: M0, M1 and M2. Under normal conditions, microglia are in the M0 phenotype. Once activated, microglia are polarized into M1 or M2 phenotype. M1, classic macrophage, produce pro-in ammatory mediators and chemokines to promote tissue defense and pathogen destruction. However, persistent M1 microglia continue to participate in the pathogenesis of PD and amplify neuronal damage caused by pathological stimuli and toxins, which in turn lead to more extensive damage to neighboring neurons. M2, replacement macrophages, can release CD206, Arg-1, Ym-1 and other molecules to promote wound healing and tissue repair. Contrary to M1, M2 is an anti-in ammatory phenotype [9][10][11]. Therefore, inhibiting M1 polarization in microglia and promoting M2 polarization are of great signi cance for inhibiting neuroin ammation and alleviating PD.
Camptothecin (CPT), a plant drug extracted from Camptotheca acuminate, has rich pharmacological effects such as antivirus, antitumor and changes in the keratinization process of the skin epidermis [12,13]. CPT has a wide range of sources and is easy to extract. Most previous studies have devoted to the therapeutic effects of CPT on cancer. Recent some studies have found that CPT also plays a role in in ammation [14,15]. In vivo sepsis models, CPT can inhibit the expression of several in ammatory cytokines and rescue mortality caused by in ammation [16]. Studies showed that CPT at low concentration can inhibit the expression of topoisomerase 1 and the inhibition of topoisomerase 1 has been found to inhibit the expression of in ammatory genes [17,18]. It is known that NF-κB is closely related to the polarization of macrophages. Studies have also reported that CPT affects the activity of NF-κB [19,20]. However the effect of CPT on microglia polarization and microglia-mediated in ammation responses has not been reported. In the experiment, we aim to explore the effect of CPT on microglia polarization and its underlying mechanism on neuroin ammation.

Materials And Methods
Page 4/43

Animals and models
Sixty male C57BL/6 mice (25-30 g) were purchased from Liaoning Changsheng Technology (Liaoning, China). Our experiment was recognized by the Institutional Animal Care and Use Committee of Jilin University (Changchun, China) (Permit Number: 2015047). During the experiment, we have done our best to reduce the suffering of animals. The PD mouse model was established. The experiment includes four groups: control group (which was injected PBS into the right SN), treated with CPT (0.5 mg/kg) group, LPS-injected group, LPS-injected treated with CPT (0.5 mg/kg) group. The mice were injected with LPS (0.2 mg/kg) or equal volume PBS into the right SN after anesthesia (stereotactic coordinates: dorsoventral (DV) = -2.5 mm, mediolateral (ML) = -0.8 mm and anterior-posterior (AP) = + 0.5 mm.). CPT was dissolved in PBS and given by gavage every two days. In addition, mice received CPT 3 days before surgery and a total of 24 days. Before the operation and on the 28th day after the operation, we measured the body weight of the mice. On the 24th day after the operation, we performed an open eld test.

Open-Field Test
On the 24th day after LPS injection, the open eld assay was performed to test the effect of the CPT on motor activity of mice. Mice were placed in a quiet, low-light box and allowed to adapt to the environment for 5 min. The bottom area of the box was cleaned up prior to test to avoid the effect of previous test. The experimental mice were placed in the open eld for 5 min. The trajectory, total traveled distance and time in the center square within 5 min were recorded by the computer and software (Any-maze, Stoelting Co).

Cell culture
Microglia cell line BV-2, the neuroblastoma SH-SY5Y and dopaminergic neuron MN9D cell were obtained from the Cell Culture Center at the Institute of Basic Medical Sciences (Peking, China). The cells were cultured in DMEM medium which contained 10% FBS (Gibco, Grand Island, NY, USA). The medium was changed every day and the cells were passaged when the con uence of the cells reached 70-80%. In the experiment, cells were treated with CPT for 1 h, stimulated with LPS (1 µg/mL)for different times, and then collected for measurement.

Reverse transcription quantitative real-time polymerase chain reaction (RT-PCR)
Total RNA of the cells and mouse midbrains were collected with the RNA extraction kit (Takara Biotechnology, Ltd., Kyoto, Japan). After detection of RNA concentration, 3 µg RNA was reversetranscribed into cDNA with the PrimerScript® 1st Strand cDNA Synthesis Kit (Sigma-Aldrich, St. Louis, MO, USA). After that, the mRNA levels of M1 markers (IL-6, TNF-α, COX-2 and iNOS) and M2 markers (CD206, Ym-1 and Arg-1) were examined with the SYBR Green Kit (South San Francisco, CA, USA) and the results were analyzed relative to β-actin using the 2 −ΔΔCt means. The sequences of the factors were shown in Table 1, which referred to previous studies [22]. Table 1 The primer sequences for RT-PCR.

Cytotoxicity assay
The effects of CPT on cell viability were tested using CCK-8 method. After cultured for 24 h, the cells were incubated with DMSO and different concentrations CPT (0-2 µM) for 20 h, and then CCK-8 diluent was added to culture medium and cultured for 3 h. After that, the absorbance was examined with a microplate reader at 450 nm.

ELISA
The cells were cultured in the 24-well plates. After incubation with CPT for 1 h and stimulated with LPS (1 µg/mL) for 24 h, the protein expression of TNF-α and IL-6 in the supernatant was examined with the ELISA kits (R&D Systems, Abingdon, UK).

Immuno uorescence staining
Immuno uorescence staining was performed to determine the nuclear translocation of the NF-kB p65 in LPS-exposed BV-2 cells. The nuclear location of the NF-kB p65 is marked by anti-NF-kB p65 antibodies (1:1000) (Abcam, Cambridge, UK).

Data analyses
Results were presented with mean ± SD, and analyzed with SPSS 13.0 software package (SPSS Inc., Chicago, USA). The differences were evaluated with the one-way analysis of variance (ANOVA) method. The p < 0.05 was considered to be signi cant and p < 0.01 was considered to be markedly signi cant.

CPT treatment alleviates the weight loss and behavioral disorder of LPS-injected PD mice
Unilateral injection of LPS into the SN can cause weight loss and motor dysfunction in mice. In order to prove the protective effect of CPT, we studied the effect of CPT treatment on weight loss and motor behavior disorders in the LPS-injected PD mouse model. The experimental process is executed as shown in Fig. 1A. On the 28th day after LPS injection, the weight change of the mice was measured. The results showed that LPS injection caused weight loss in mice, and CPT treatment could signi cantly improve LPS-induced weight loss in the LPS-injected PD mouse model (Fig. 1B). On the 24th day after LPS injection, we tested the athletic ability of mice through an open eld experiment. The results showed that LPS injection caused the distance moving in the open eld and time in the center less, and CPT treatment could improve this phenomenon ( Fig. 1C-E). These results illustrated that CPT treatment alleviates the weight loss and behavioral disorder of LPS-injected PD mouse model.

CPT treatment decreases dopaminergic neurons degeneration in LPS-injected PD mice
The main pathological feature of PD is the degeneration of dopaminergic neurons. To explore protective effect of CPT, we studied the effect of CPT on the dopaminergic neurons in LPS-injected PD mice. On the 28th day after LPS injection, we obtained the midbrain of mice and examined the number of TH-positive cells using immunohistochemistry method. Results showed that LPS injection caused a signi cant decrease of midbrain dopaminergic neurons in LPS-injected PD mice, and CPT treatment could protect the neurons from the damage caused by LPS ( Fig. 2A, B). The protein level of TH was examined using western blot. The results also con rmed the protective effect of CPT from a protein perspective (Fig. 2C). These results illustrated that CPT treatment decreases dopaminergic neurons degeneration in LPSinjected PD mice.

CPT treatment inhibits in ammatory response and regulates microglia polarization in LPS-injected PD mice
To further con rm neuroprotection of CPT and its mechanism, we studied the effect of CPT on in ammatory response and microglia polarization in LPS-injected PD mouse model. On the 28th day after LPS injection, we obtained the midbrain of mice. AKT/Nrf2/HO-1-NF-κB signal axis was detected by western blot. The results showed that CPT activated AKT/Nrf2/HO-1 and inhibited NF-κB pathways ( Fig. 3A-C). Then we detected the mRNA levels of M1 markers (IL-6, TNF-α, iNOS and COX-2) and M2 markers (Ym-1, CD206 and Arg-1) in the midbrain of mice by RT-PCR. Results showed that CPT inhibited release of M1 markers and promoted release of M2 markers ( Fig. 3D-J). These results illustrated that CPT treatment inhibits in ammatory response and regulates microglia polarization via AKT/Nrf2/HO-1-NF-κB signal axis in LPS-injected PD mouse model. Results are shown as means ± SD (n = 5). ## p < 0.01 vs. the no-treatment (NT) group; **p < 0.01 vs. the LPS-exposed group.
3.5 CPT inhibits activation of NF-κB pathway in LPSexposed BV-2 cells NF-κB pathway, a key pathway of in ammation, affects the production of many pro-in ammatory mediators. To investigate the mechanism which CPT regulates the polarization of M1 and M2 microglia, we detected the effect of CPT on the activation of NF-κB pathway. BV-2 cells were pretreated with CPT for 1 h and stimulated with LPS for another 1 h. Then phosphorylation of IκB, NF-κB p65 and degradation of IκB were detected using western blot. After that, we also examined the nuclear positioning of NF-κB p65 by immuno uorescence. Results showed that CPT inhibited phosphorylation of NF-κB p65 (Fig. 5A, B), IκB (Fig. 5A, C) and degradation of IκB (Fig. 5A, D) and nuclear translocation of NF-κB p65 (Fig. 5E, F). These results illustrated that CPT inhibits activation of NF-κB in LPS-exposed BV-2 cells. Results are shown as means ± SD (n = 5). ## p < 0.01 vs. the untreated group (NT); **p < 0.01 vs. the LPS-exposed group.
3.6 CPT promotes phosphorylation of AKT, activation of Nrf2 and up-regulates the expression of HO-1 in

BV-2 cells
To further clarify the mechanism of CPT anti-in ammatory, we studied the effect of CPT on AKT, Nrf2 and HO-1 in ammation pathways. After the cells were treated with CPT (0, 0.25, 0.5 and 1 µM) for 3 h, the protein levels of phos-AKT, HO-1, nuclear-nrf2 and Cytoplasm-nrf2 were detected by western blot. Results showed that CPT treatment promoted phosphorylation of AKT, activation of Nrf2 and up-regulated protein level of HO-1 (Fig. 6A-D).Then we pretreated cells with MK2206 (an AKT inhibitor, 10 µM) for 4 h and studied the effect of CPT on activation of Nrf2 using western blot. Results showed that MK2206 inhibited the promotion effect of CPT on the protein level of nuclear-nrf2 ( Fig. 6E-F). We pretreated cells with RA (a nrf2 inhibitor, 5 µM) for 4 h and studied the effect of CPT on protein level of HO-1. Results showed that RA inhibited the effect of CPT on up-regulation of HO-1 (Fig. 6G-H). These results illustrated that CPT promotes AKT/Nrf2 /HO-1 signaling pathways in BV-2 cells. 3.7 CPT regulates microglial polarization via AKT/ Nrf2/HO-1-NF-κB signal axis We pretreated cells with SnPP IX (a HO-1 inhibitor, 40 µM) for 3 h and studied the effect of CPT on NF-κB pathway activation. Results showed that SnPP IX can reverse the effect of CPT on inhibition of NF-κB pathway activation (Fig. 7A, B). Then we pretreated BV-2 cells with different inhibitors (MK2206 for 4 h, RA for 4 h or SnPP IX for 3 h) and studied the effect of CPT on microglial polarization. Results showed that the inhibitors (MK2206, RA and SnPP IX) can reverse the regulatory effect of CPT on the release of MI markers (IL-6, TNF-α, COX-2 and iNOS) (Fig. 7C-F) and M2 markers (CD206, Ym-1 and Arg-1) (Fig. 7G-I). The above results illustrated that CPT regulates microglial polarization via AKT/Nrf2/HO-1-NF-κB signal axis.  4. Discussion PD, the second largest neurodegenerative disease seriously affects the physical and mental health of elderly people. Clinically, PD patients have severe motor impairments and massive loss of dopaminergic neurons in the SN of midbrain [23]. The etiology of PD is still unclear, and the accumulated evidences demonstrate that neuroin ammation plays an important role in the occurrence and development of PD.
When neuroin ammation occurs, the immune cells, mainly microglia are over-activated to release proin ammatory mediators, resulting in the degeneration of peripheral neurons [24][25][26]. Therefore, inhibition of neuro-in ammation is also considered to be a target for the treatment of PD. LPS, a component of the cell wall of Gram-negative bacteria can induce in ammation response. Studies have demonstrated that the injection of LPS into the SN of rats can induce PD symptoms [27,28]. In the experiment, we found that mice injected with LPS in the SN had PD symptoms such as weight loss and motor dysfunction compared with control mice. Further research found that dopaminergic neurons decreased in the SN of LPS-injected mice. Our research also found that CPT could relieve PD symptoms and improve the damage of neurons of LPS-injected mice. The results prompted that CPT has a neuroprotective effect.
Microglia, immune cells in the CNS, is the main participants of neuro-in ammatory response. Studies have reported that there are numerous microglia activated abnormally in the midbrain of PD patients and PD model animals [29,30]. BV-2 cells, microglial cell lines, have been widely used to study the function of microglia due to the similarity to microglia [31,32]. To further study the neuroprotective effect of CPT and its mechanism, we studied the effect of CPT on microglia in ammation using BV-2 cells. In the experiment, we found that CPT treatment can inhibit the release of M1 markers and promote the release of M2 markers. We also found that CPT exhibited this effect in LPS-injected mice. The results prompted that CPT inhibits the polarization of microglia to the pro-in ammatory M1 and promotes to the antiin ammatory M2, thereby exerting an anti-in ammatory effect.
NF-κB, a classic in ammatory pathway, is involved in cellular in ammation and many nervous system diseases. Studies have reported that NF-κB can regulate the transcription of various pro-in ammatory mediators and considered to be a target for in ammatory diseases [33,34]. There are also reports that NF-κB pathway is closely related to the polarization of immune cells [35]. Under normal conditions, NF-κB p65 is present in the cytoplasm and binds to IκB subunit. Once activated, IκB subunit degrades and NF-κB p65 undergoes nuclear translocation and phosphorylation [36]. In the experiment, we found that CPT can inhibit the degradation of IκB and nuclear translocation of NF-κB p65 in LPS-exposed BV-2 cells. The results prompted that CPT inhibits microglia in ammation by repressing the activation of the NF-κB pathway.
AKT pathway, or PI3K-AKT pathway is involved in fundamental cellular processes including protein synthesis, proliferation and survival. AKT signaling has been implicated in various in ammation responds [37,38]. Nrf2 is a key transcription factor regulating oxidative stress. Studies found that activated-nrf2 enter into the nuclear and regulates transcription of many in ammatory factors. HO-1 is a key anti-in ammatory protein downstream of Nrf2 [39,40]. In the experiment, we found that CPT treatment can activate the AKT pathway, promote activation of Nrf2 and up-regulate the expression of HO-1. To further study anti-in ammatory effect of CPT on microglia, we treated BV-2 cells with MK2206 (an AKT inhibitor), RA (a Nrf2 inhibitor) or SnPP IX (a HO-1 inhibitor) respectively. The results showed that after blocking the AKT pathway with MK2206, the effect of CPT on Nrf2 nuclear transcription was suppressed. These indicate that CPT promotes activation of Nrf2 through activating AKT in microglia. At the same time, RA and SnPP IX treatment also demonstrated the regulatory effect of CPT on in ammatory pathways. In addition, we also explored the effect of CPT on the release of M1 and M2 markers in microglia after BV2 cells treated with MK2206, RA or SnPP IX. The results showed that MK2206, RA or SnPP IX treatment reversed the regulatory effect of CPT on M1 and M2 markers to a certain extent. The results prompted that CPT regulates the M1 and M2 polarization of microglia through AKT/Nrf2/HO-1-NF-κB signaling axis. The experiments in vivo also con rmed this.
To explore the neuroprotective effect of CPT, we pretreated BV-2 cells with CPT and stimulated with LPS to induce microglial activation. Then we change the fresh medium for 18 h and obtain conditioned medium. Next we studied the effect of conditioned medium on viability of SHSY5Y and MN9D neuronal cell lines. The results showed that CPT could inhibit microglial activation-mediated neurotoxicity and protect neurons.

Conclusions
In conclusion, our study found for the rst time that CPT inhibits microglial M1 polarization and promotes M2 polarization through the AKT/Nrf2/HO-1-NF-κB signal axis, thereby inhibiting neuroin ammation and exerting the effect of neuroprotection in vivo and in vitro (Fig. 9). This experiment would provide new ideas for the treatment of in ammation-mediated PD. Combining its extensive pharmacological effects and easily available properties, CPT is expected to be further developed and studied. Future research will also focus more on the potential therapeutic role of CPT in in ammatory diseases and assess the possibility of being developed as anti-in ammatory drugs.       Results are shown as means ± SD (n= 5).##p < 0.01 vs. the no-treatment (NT) group; **p< 0.01 vs. the LPS-exposed group. Results are shown as means ± SD (n= 5).##p < 0.01 vs. the no-treatment (NT) group; **p< 0.01 vs. the LPS-exposed group. Results are shown as means ± SD (n= 5).##p < 0.01 vs. the no-treatment (NT) group; **p< 0.01 vs. the LPS-exposed group.