Kaempferia parviflora Extracts Protect Neural Stem Cells from Amyloid Peptide-Mediated Inflammation in Co-Culture Model with Microglia

The existence of neuroinflammation and oxidative stress surrounding amyloid beta (Aβ) plaques, a hallmark of Alzheimer’s disease (AD), has been demonstrated and may result in the activation of neuronal death and inhibition of neurogenesis. Therefore, dysregulation of neuroinflammation and oxidative stress is one possible therapeutic target for AD. Kaempferia parviflora Wall. ex Baker (KP), a member of the Zingiberaceae family, possesses health-promoting benefits including anti-oxidative stress and anti-inflammation in vitro and in vivo with a high level of safety; however, the role of KP in suppressing Aβ-mediated neuroinflammation and neuronal differentiation has not yet been investigated. The neuroprotective effects of KP extract against Aβ42 have been examined in both monoculture and co-culture systems of mouse neuroectodermal (NE-4C) stem cells and BV-2 microglia cells. Our results showed that fractions of KP extract containing 5,7-dimethoxyflavone, 5,7,4′-trimethoxyflavone, and 3,5,7,3′,4′-pentamethoxyflavone protected neural stem cells (both undifferentiated and differentiated) and microglia activation from Aβ42-induced neuroinflammation and oxidative stress in both monoculture and co-culture system of microglia and neuronal stem cells. Interestingly, KP extracts also prevented Aβ42-suppressed neurogenesis, possibly due to the contained methoxyflavone derivatives. Our data indicated the promising role of KP in treating AD through the suppression of neuroinflammation and oxidative stress induced by Aβ peptides.


Introduction
Dementia is a type of neurodegenerative disease comprising frontotemporal dementia, Lewy body dementia, vascular dementia, and Alzheimer's disease (AD), with AD accounting for 60-70% of all dementia cases [1]. Dementia, including AD, is characterized by neuronal loss, leading to a decline in memory and cognitive ability that negatively impact individuals and their families [1]. AD etiology is multifaceted, with several variables including genetic factors, reduced neurogenesis, mitochondrial dysfunction, oxidative stress accumulation, neuroinflammation, and accumulation of hyperphosphorylated tau and amyloid beta (Aβ) peptides. These variables interact and generate a vicious cycle that accelerates neuronal damage and death, leading to neuron loss and brain atrophy [2,3]. The neuroinflammatory responses surrounding Aβ plaques in AD suggest that inflammation BV2 cells, a mouse microglial cell line, were purchased from Interlab Cell Line Collection (ATL033001, Genova, Italy). The cells were cultured and seeded according to the previous reports [29,30].

Differentiation of NE-4C Neural Stem Cells into Neurons
The differentiation of NE-4C stem cells into neurons was performed according to the previous protocol [31]. Briefly, NE-4C cells were dissociated with 0.05% trypsin-EDTA and plated onto PLL-precoated 24-well plates at 1.5 × 10 6 cells/well. The cells were cultured in supplemented MEM for 12 h before neuronal differentiation induction. Differentiation was induced by replacing the MEM with a neurobasal medium supplemented with 100 units/mL penicillin, 100 µg/mL streptomycin, and B-27. The supplemented neurobasal medium was replaced every 48 h. Six days after induction, aggregates with clusters of neural stem cells were mechanically dissociated, seeded onto PLL-coated 24-well plates, and grown until day twelve.

Co-Culture System between NE-4C Stem Cells and BV-2 Microglial Cells
Co-culture of NE-4C stem cells and microglial cells was performed using 24-well insert plates (Millipore, Burlington, MA, USA). NE-4C cells were cultured in PLL-coated 24-well plates at a density of 1.5 × 10 6 cells/well in supplemented MEM at the bottom of the well, while microglia were cultured in supplemented RPMI at a density of 1 × 10 5 cells/well in the insert for 24 h.
Induction of neuronal differentiation in the co-culture system between NE-4C stem cells and BV-2 microglial cells was performed by replacing the media with the supplemented neurobasal medium. The BV-2 cells were loaded for co-culture with the plated NE-4C cells and shared media of the supplemented neurobasal medium. After six days, neurospheres were collected and dissociated, seeded onto PLL-coated 24-well plates without co-cultured microglial cells, and grown until day twelve.

Co-Culture System between Differentiated Neurons Derived from NE-4C Cells and BV-2 Microglial Cells
NE-4C stem cells were differentiated into neuronal cells and cultured in supplemented neurobasal medium for 12 days to promote neuronal maturation in the bottom of the well. Consequently, the BV-2 cells in the insert were loaded as an upper chamber for co-culture with the matured neurons and the shared media of neuronal cells.

Protective Effects of KP Extracts in Both Mono and Co-Culture
The protective effects of KP extract on Aβ 42 -induced cytotoxicity were performed for both monocultures (BV-2 cells, undifferentiated NE-4C cells and differentiated neurons) and co-cultured (BV-2 cells cultured with undifferentiated NE-4C cells, and BV-2 cells cultured with differentiated NE-4C cells) as described. Aβ 42 at 5 µM, which reduced 50% of cellular ATP level, was cotreated with KP extracts at nontoxic concentrations (0.5, 1, 2, 4, and 8 µg/mL) and incubated for 24 h. After exposure, the XTT reduction assay and ATP level were utilized to evaluate cytotoxicity.

Anti-Inflammatory Activities of KP Extracts on Aβ 42 -Induced Inflammation in Monoculture
The BV-2 cells were seed in a 96-well plate at 1 × 10 4 cells/well and co-incubated with 1 µM Aβ 42 with or without KP extract at nontoxic concentrations (0.5, 1, 2, 4 and 8 µg/mL) for 24 h. BV-2 cells were harvested to determine the intracellular reactive oxygen species (ROS) using the DCF fluorescence assay, while the culture supernatant was collected and determined for IL-6 and nitrite levels.

Anti-Inflammatory Effects of KP Extracts on Aβ 42 -Induced Inflammation in Co-Culture
Two co-culture models as matured neuronal cells derived from NE-4C cells and BV-2 microglial cells, and undifferentiated NE-4C neural stem cells and BV-2 microglial cells prepared as mentioned above were used to investigate whether KP extracts could suppress Aβ 42 -induced inflammation via microglial activation. Aβ 42 (1 µM) and extracts including KP1, KP2, and KP3 (0.5, 1, 2, 4, and 8 µg/mL) were added to the culture system medium for 24 h. The culture supernatant was collected and measured for IL-6 and nitrite levels. BV-2 cells were collected to evaluate the expression of IL-6 and iNOS mRNA and determine intracellular ROS. Differentiated neuronal cells and undifferentiated NE-4C neural stem cells were collected to determine cytotoxicity and intracellular ROS.

Effects of KP Extracts on Neurogenesis of Aβ 42 -Induced NE-4C Cells in Monoculture
NE-4C cells were induced for differentiation in the supplemented neurobasal medium for 24 h and then exposed to Aβ 42 (0.25 µM) with or without KP extracts at nontoxic concentrations (0.5, 1, 2, 4, and 8 µg/mL) for six days. The differentiation medium was replaced every 48 h. At day six, cells were dissociated, seeded onto 24-well culture plates and grown until day twelve. Differentiated neuronal cells were collected for expression of βIII-tubulin and MAP-2 analysis.

Effects of KP Extracts on Neurogenesis in Aβ 42 -Induced NE-4C Cells in Co-Culture with BV-2 Cells
The BV-2 cells were loaded for co-culture with plated NE-4C cells for one day in the supplemented neurobasal medium. Then, Aβ 42 (0.25 µM) with or without extracts including KP1, KP2, and KP3 (0.5, 1, 2, 4, and 8 µg/mL) was added, with differentiation continued for five days. Culture media were collected for IL-6 and nitrite level determination. BV-2 cells were collected for quantitative PCR analysis of IL-6 and iNOS expression and to determine the intracellular ROS. NE-4C cells were dissociated, seeded onto 24-well culture plates, and grown until day twelve without Aβ 42 or KP extracts. Differentiated neuronal cells were collected for βIII-tubulin and MAP-2 mRNA analysis.

Sodium 3 -[1-(Phenylaminocarbonyl)-3,4-tetrazolium]-bis (4-Methoxy-nitro) Benzene Sulfonic Acid Hydrate (XTT) Reduction Assay
The cytotoxicity using an XTT-based assay was performed as previously detailed [29,30]. After the exposure times, the media was discarded. Then, 0.3 mg/mL XTT and 125 mM PMS were added. After incubation at 37 • C for 2 h, the absorbance was measured at 450 nm (SpectraMax ® iD3, Molecular Devices, San Jose, CA, USA). Results are expressed as a percentage of cells that were untreated as a negative control.

Determination of ATP Levels
The total intracellular ATP level was determined using the CellTiter-Glo ® Luminescence assay kit. Following cell treatment, the assay reagent containing substrate was added per each well and mixed for 30 min under light protection. The luminescence at 550 nm was measured (SpectraMax ® iD3, Molecular Devices, San Jose, CA, USA), and the luminescence signal was expressed as a percentage of control. After the treatment, the culture medium was harvested and centrifuged at 2000× g for 10 min at 4 • C and kept at −20 • C until analysis. The culture supernatant was submitted to an IL-6 quantitative sandwich ELISA kit. The absorbance was measured at 450 nm using a microplate reader. The concentration of IL-6 in the samples was calculated in the comparison with the standard, curve (ranging from 10 to 500 pg/mL of IL-6).

Nitric Oxide (NO) Measurement
Griess reagent was used to NO levels. The culture medium (50 µL) was mixed with sulfanilamide solution (50 µL) for 10 min before being incubated with napthylethylenediamine dihydrochloride solution (50 µL) for 10 min. The absorbance at 520 nm was determined by a microplate reader.

Intercellular Reactive Oxygen Species (ROS) Measurement
The treated cells were washed with PBS, followed by preincubation with 20 µM 2,7-dichlorofluorescein diacetate (DCFDA) in a prewarmed culture medium for 30 min at 37 • C in the dark. The supernatant was removed, and the cells were washed with PBS, followed by the addition of 200 µL of cell lysis buffer (90% dimethyl sulfoxide/10% PBS). The mixtures were incubated for 5 min. The mixture (150 µL) was then transferred to a black 96-well plate, and the fluorescence was quantitated using a fluorometric plate reader (SpectraMax ® iD3, Molecular Devices, San Jose, CA, USA) at 480 nm/530 nm excitation/emission wavelengths [36].

Reverse Transcription-Quantitative Polymerase Chain Reaction (RT-qPCR)
Total RNA was extracted from the treated cells using Tri-RNA Reagent. The RNA quality and quantity were analyzed using Nanodrop ND-1000 spectrophotometry (Nan-oDrop Technologies, Wilmington, DE, USA). One hundred nanograms of total RNA was added and prepared in the RT-qPCR reaction mixture, qPCRBIO SyGreen 1-step Lo-ROX. Quantitative polymerase chain reaction (qPCR) was performed using qTOWER3 Real-Time PCR Systems (Analytik Jena, Langewiesen, Germany) and then analyzed using qPCRsoft 3.4 software (Analytik Jena, Langewiesen, Germany). Relative levels of target gene expression were normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) mRNA using the 2 −∆∆CT method [37]. Each experiment was replicated at least three times, and the primer sequences are listed in the Supplementary Table S1.

Statistical Analysis
Data from at least three separate experiments are reported as mean ± standard deviation (SD). GraphPad Prism 9.0 (GraphPad Software, Boston, MA, USA) was used to statistically analyze the data. A one-way ANOVA followed by Tukey's test was used to determine the statistical significance of differences between groups. A probability of 0.05 or less (p ≤ 0.05) was considered statistically significant.

Cytotoxicity Determination of Aβ 42 and K. parviflora Extracts
To investigate the protective effects of K. parviflora extract against Aβ 42 peptides, we first determined the cytotoxic effects of Aβ 42 peptides and the five fractions of K. parviflora extracts to establish the ranges of appropriate concentrations for the whole study, using XTT reduction and ATP levels as two independent cytotoxicity assays. The three cell lines, BV-2, undifferentiated NE-4C and differentiated NE-4C were treated with various ranges of Aβ 42 and the five fractions of K. parviflora extracts (KP1-KP5) for 24 h. Figure 1A,B show that after 24 h of treatment, Aβ 42 at ≤1 µM was not cytotoxic to all tested cells, while Aβ 42 at ≥5 µM showed significant cytotoxicity toward all tested cells in our condition.
respectively. KP3 extracted from chloroform contained major K. parviflora components, suggesting the predominance of nonpolar phytochemicals. HPLC analysis detected caffeic acid and rutin in KP4, whereas only catechin was found in KP5 (Table 1 and Supplementary Figure S1). The flavone contents of KP have been well-documented and methoxyflavone derivatives were also determined, as shown in Table 1 and Supplementary Figure S2. DMF, PMF, and TMF were predominant in KP1 and KP3, with minor amounts in KP2 and KP4. TMF was the most abundant flavone derivative found in KP1 to KP4 compared to DMF and TMF. Interestingly, these three derivatives were not detectable in the KP5 fraction. The value is represented as mean ± SD. ND.: non-detectable. DMF: 5,7-dimethoxyflavone, PMF: 3,5,7,3′,4′-pentamethoxyflavone, and TMF: 5,7,4′-trimethoxyflavone.

Cytotoxicity Determination of Aβ42 and K. parviflora Extracts
To investigate the protective effects of K. parviflora extract against Aβ42 peptides, we first determined the cytotoxic effects of Aβ42 peptides and the five fractions of K. parviflora extracts to establish the ranges of appropriate concentrations for the whole study, using XTT reduction and ATP levels as two independent cytotoxicity assays. The three cell lines, BV-2, undifferentiated NE-4C and differentiated NE-4C were treated with various ranges of Aβ42 and the five fractions of K. parviflora extracts (KP1-KP5) for 24 h. Figure 1A,B show that after 24 h of treatment, Aβ42 at ≤1 μM was not cytotoxic to all tested cells, while Aβ42 at ≥5 μM showed significant cytotoxicity toward all tested cells in our condition. Results from the XTT assay also revealed that all five fractions of K. parviflora extracts (KP1-KP5) displayed different cytotoxic effects. In BV-2 cells, KP1 markedly showed toxicity starting from 40 µg/mL, while KP2-KP3 showed toxicity starting from 200 µg/mL and KP4-KP5 exhibited clear toxicity starting from 1000 µg/mL (Figure 2A-E). In undifferentiated NE-4C, KP1-KP3 showed toxicity starting from 200 µg/mL, and KP4-KP5 showed toxicity starting from 1000 µg/mL. Moreover, in differentiated NE-4C, significant toxicity was obtained from 200 µg/mL of KP1-KP5. Thus, 0.5 to 8 µg/mL of KP1-KP5, which were sub-cytotoxic concentrations, were selected for further investigation. icity starting from 40 µ g/mL, while KP2-KP3 showed toxicity starting from 200 µ g/mL and KP4-KP5 exhibited clear toxicity starting from 1000 µ g/mL (Figure 2A-E). In undifferentiated NE-4C, KP1-KP3 showed toxicity starting from 200 µ g/mL, and KP4-KP5 showed toxicity starting from 1000 µ g/mL. Moreover, in differentiated NE-4C, significant toxicity was obtained from 200 µ g/mL of KP1-KP5. Thus, 0.5 to 8 µ g/mL of KP1-KP5, which were sub-cytotoxic concentrations, were selected for further investigation.

Protective Effects of K. parviflora Extracts on Aβ 42 -Mediated Neurotoxicity
The protective effects of all five K. parviflora extract against Aβ 42 -mediated neurotoxicity in both NE-4C monoculture and co-culture between NE-4C and BV-2 cells were evaluated. Aβ 42 at 5 µM was selected to induce neuronal death as this mirrored the IC 50 value, using subtoxic doses of K. parviflora extracts (0.5-8 µg/mL). The NE-4C monoculture or co-culture between NE-4C and BV-2 cells was treated with Aβ 42 and K. parviflora extracts. After 24 h of treatment, cytotoxicity was tested using the XTT assay. Figure 3A,B show that low concentrations of all five K. parviflora fractions (0.5-1 µg/mL) had no protective effects against Aβ 42 , while only KP1, KP2, and KP3 at 2-8 µg/mL effectively normalized Aβ 42 -mediated neurotoxicity in both monoculture and co-culture of BV-2 cells. Interestingly, KP1 (at least 4 µg/mL), KP2 (at least 2 µg/mL), and KP3 (at least 4 µg/mL), but not KP4 and KP5 significantly suppressed Aβ 42 toxicity in undifferentiated and differentiated NE-4C cells (both monoculture and co-culture). Therefore, KP1, KP2, and KP3 fractions exhibited protective effects against Aβ 42 -induced neuronal death in both monoculture and co-culture models.
After 24 h of treatment, cytotoxicity was tested using the XTT assay. Figure 3A,B show that low concentrations of all five K. parviflora fractions (0.5-1 µ g/mL) had no protective effects against Aβ42, while only KP1, KP2, and KP3 at 2-8 µ g/mL effectively normalized Aβ42-mediated neurotoxicity in both monoculture and co-culture of BV-2 cells. Interestingly, KP1 (at least 4 µ g/mL), KP2 (at least 2 µ g/mL), and KP3 (at least 4 µ g/mL), but not KP4 and KP5 significantly suppressed Aβ42 toxicity in undifferentiated and differentiated NE-4C cells (both monoculture and co-culture). Therefore, KP1, KP2, and KP3 fractions exhibited protective effects against Aβ42-induced neuronal death in both monoculture and co-culture models. Values are mean ± SD of three independent experiments. The ### indicates a significant difference between the control and cells exposed to Aβ42 peptides without KP extract at p < 0.001 using one-way ANOVA followed by Tukey's test. The * indicates a significant difference between cells exposed to Aβ42 peptides without KP extract compared to cells exposed to Aβ42 peptides Values are mean ± SD of three independent experiments. The ### indicates a significant difference between the control and cells exposed to Aβ 42 peptides without KP extract at p < 0.001 using one-way ANOVA followed by Tukey's test. The * indicates a significant difference between cells exposed to Aβ 42 peptides without KP extract compared to cells exposed to Aβ 42 peptides and various concentrations of KP extract (0.5-8 µg/mL) using one-way ANOVA followed by Tukey's test. *, p < 0.05, **, p < 0.01, ***, p < 0.001.

Suppression of Aβ42-Induced Inflammation and Oxidative Stress by K. parviflora Extracts in Co-Culture between Differentiated NE-4C and BV-2 Cells
Inflamed microglia eventually lead to neuronal dysfunction and death. Figures 3 and  4 show the protective effects of KP1, KP2, and KP3 against neuronal death, inflammation, and oxidative stress in microglial BV-2 cell monoculture. Thus, the anti-inflammation and antioxidative stress of KP1, KP2, and KP3 fractions were further studied in co-culture between differentiated NE-4C and BV-2 cells. IL-6 and inducible nitric oxide synthase (iNOS) mRNA levels and cellular ROS levels in BV2-cells were significantly induced after Aβ42 treatment, confirming the neurotoxic effect of Aβ42 peptides ( Figure 5A-C). However, these markers were reduced when KP1 and KP2 at 4-8 µ g/mL and KP3 at 8 µ g/mL were applied. Values are mean ± SD of three independent experiments. The # shows a significant difference between the control and cells exposed to Aβ 42 peptides without KP extract using one-way ANOVA followed by Tukey's test. # , p < 0.05, ## , p < 0.01, ### , p < 0.001. The * shows a significant difference between cells exposed to Aβ 42 peptides without KP extract compared to cells exposed to Aβ 42 peptides and various concentrations of KP extract (0.5-8 µg/mL) using one-way ANOVA followed by Tukey's test. *, p < 0.05, **, p < 0.01, ***, p < 0.001.

Suppression of Aβ 42 -Induced Inflammation and Oxidative Stress by K. parviflora Extracts in Co-Culture between Differentiated NE-4C and BV-2 Cells
Inflamed microglia eventually lead to neuronal dysfunction and death. Figures 3 and 4 show the protective effects of KP1, KP2, and KP3 against neuronal death, inflammation, and oxidative stress in microglial BV-2 cell monoculture. Thus, the anti-inflammation and antioxidative stress of KP1, KP2, and KP3 fractions were further studied in co-culture between differentiated NE-4C and BV-2 cells. IL-6 and inducible nitric oxide synthase (iNOS) mRNA levels and cellular ROS levels in BV2-cells were significantly induced after Aβ 42 treatment, confirming the neurotoxic effect of Aβ 42 peptides ( Figure 5A-C). However, these markers were reduced when KP1 and KP2 at 4-8 µg/mL and KP3 at 8 µg/mL were applied.
The effects of Aβ42 peptides on differentiated NE-4C cells were also investigated. In the co-culture, Aβ42 peptides markedly decreased cell viability and caused oxidative stress in differentiated NE-4C cells compared to the untreated control ( Figure 5D-F), suggesting the occurrence of inflamed microglia-mediated neuronal death. KP1 and KP2 (2, 4, and 8 µ g/mL) and KP3 (4 and 8 µ g/mL) prevented neuronal death and oxidative stress in differentiated NE-4C cells. Figure 5G,H show that IL-6 and nitrite levels in the cell culture medium were dramatically induced by Aβ42 peptides, whereas levels were significantly reduced by KP1, KP2, and KP3 treatments. Therefore, Aβ42 peptides caused inflammation and oxidative stress in co-culture between microglia and differentiated NE-4C cells leading to neuronal death. Interestingly, KP fractions KP1, KP2, and KP3 significantly decreased inflamed microglia and improved neuronal viability under Aβ42 induction. Values are mean ± SD of three independent experiments. The ### shows significant difference between the control and cells exposed to Aβ 42 peptides without KP extract at p < 0.001 using one-way ANOVA followed by Tukey's test. The * shows significant difference between cells exposed to Aβ 42 peptides without KP extract compared to cells exposed to Aβ 42 peptides and various concentrations of KP extract (1-8 µg/mL) using one-way ANOVA followed by Tukey's test. *, p < 0.05, **, p < 0.01, ***, p < 0.001. The effects of Aβ 42 peptides on differentiated NE-4C cells were also investigated. In the co-culture, Aβ 42 peptides markedly decreased cell viability and caused oxidative stress in differentiated NE-4C cells compared to the untreated control ( Figure 5D-F), suggesting the occurrence of inflamed microglia-mediated neuronal death. KP1 and KP2 (2, 4, and 8 µg/mL) and KP3 (4 and 8 µg/mL) prevented neuronal death and oxidative stress in differentiated NE-4C cells. Figure 5G,H show that IL-6 and nitrite levels in the cell culture medium were dramatically induced by Aβ 42 peptides, whereas levels were significantly reduced by KP1, KP2, and KP3 treatments. Therefore, Aβ 42 peptides caused inflammation and oxidative stress in co-culture between microglia and differentiated NE-4C cells leading to neuronal death. Interestingly, KP fractions KP1, KP2, and KP3 significantly decreased inflamed microglia and improved neuronal viability under Aβ 42 induction.

Suppression of Aβ 42 -Induced Inflammation and Oxidative Stress by K. parviflora Extracts in
Co-Culture between Undifferentiated NE-4C and BV-2 Cells Figure 5 shows the effects of inflamed microglia on differentiated NE-4C death; however, the effects of inflammatory microglia on neural stem cells that eventually develop into functional neurons remain unclear. Hence, the results of Aβ 42 -mediated inflamed microglia on undifferentiated NE-4C cells using the same co-culture strategy were further elucidated. Figure 5A-C illustrate that Aβ 42 peptides activated IL6 and iNOS mRNA expressions and oxidative stress (DCF fluorescence), while all inflammatory and oxidative stress markers were quenched by KP1, KP2, and KP3 extracts. Interestingly, inflamed BV-2 cells also decreased cell viability ( Figure 6A,B), and caused oxidative stress ( Figure 6C) and inflammation ( Figure 6D,E) of undifferentiated neural stem cells in the same manner as differentiated neurons ( Figure 5D-F). Treatment with KP extracts (KP1-KP3) protected against neuronal death and decreased oxidative stress ( Figure 6A-E). These data implied that (i) Aβ 42 -mediated inflamed microglia led to the cytotoxicity of differentiated neurons and also undifferentiated neural stem cells, and (ii) KP1-KP3 decreased inflamed microglia, which subsequently resulted in improved viability of undifferentiated neural stem cells.

Protective Effects of K. parviflora Extracts on Neurogenesis of Aβ 42 -Treated NE-4 Cells in Monoculture
Data in Figure 6 show that Aβ 42 peptides reduced the viability of neural stem cells. Protective effects of KP extracts on neurogenesis-inhibiting Aβ 42 peptide-treated cells were further elucidated by studying the expressions of two neuron-specific proteins, namely class III β tubulin (beta-III tubulin) and microtubule-associated protein 2 (MAP-2) in monoculture of differentiated NE-4 cells. Reductions in both beta-III tubulin and MAP2 mRNA levels were obtained when differentiated NE-4 cells were exposed to Aβ 42 peptides ( Figure 7A,B), suggesting that Aβ 42 peptides directly inhibit neurogenesis, even at a nontoxic dose (0.25 µM, Figure 1A,B). Cells receiving both Aβ 42 and various concentrations of KP fractions (KP2 and KP3 starting at 2 µg/mL, while KP1 starting at 4 µg/mL) showed significant recovery of beta-III tubulin and MAP-2 mRNA expressions, albeit at different potency. KP1, KP2, and KP3 fractions showed promising protective effects against neurogenesis-inhibiting Aβ 42 peptides in the monoculture of differentiated NE-4 cells ( Figure 7A,B).

Protective Effects of K. parviflora Extracts Neurogenesis in Aβ 42 -Induced Differentiated NE-4C in Co-Culture with BV-2 Cells
In addition to monoculture (Figure 7), we further investigated the protective effects of K. parviflora extracts against neurogenesis-inhibiting Aβ 42 peptides in co-culture between differentiated NE-4C and BV-2 cells. Figure 8A-C show that without KP extracts, Aβ 42 peptides induced inflammation and oxidative stress in BV-2 cells, similar to previous data ( Figure 5A-C). Reduced inflammation and oxidative stress were also observed when each KP extract was added. Consistent with results obtained from undifferentiated cells shown in Figure 7, differentiated NE-4C cells exposed to Aβ 42 peptides showed a reduction of both beta-III tubulin and MAP-2 mRNA expressions in co-culture ( Figure 8D,E), suggesting that Aβ 42 peptides not only inhibited neurogenesis directly but also reduced neurogenesis indirectly through inflamed microglia. Values are mean ± SD of three independent experiments. The ### shows a significant difference between the control and cells exposed to Aβ42 peptides without KP extract at p < 0.001 using one-way ANOVA followed by Tukey's test. The * shows a significant difference between cells exposed to Aβ42 peptides without KP extract compared to cells exposed to Aβ42 peptides and various concentrations of KP Values are mean ± SD of three independent experiments. The ### shows a significant difference between the control and cells exposed to Aβ 42 peptides without KP extract at p < 0.001 using one-way ANOVA followed by Tukey's test. The * shows a significant difference between cells exposed to Aβ 42 peptides without KP extract compared to cells exposed to Aβ 42 peptides and various concentrations of KP extract (1-8 µg/mL) using one-way ANOVA followed by Tukey's test. *, p < 0.05, **, p < 0.01, ***, p < 0.001.
showed significant recovery of beta-III tubulin and MAP-2 mRNA expressions, albeit at different potency. KP1, KP2, and KP3 fractions showed promising protective effects against neurogenesis-inhibiting Aβ42 peptides in the monoculture of differentiated NE-4 cells ( Figure 7A,B). The ### shows a significantly different between control and cells exposed to Aβ42 peptides without KP extract at p < 0.001 using one-way ANOVA followed by Tukey's test. The * shows a signifi- The ### shows a significantly different between control and cells exposed to Aβ 42 peptides without KP extract at p < 0.001 using one-way ANOVA followed by Tukey's test. The * shows a significantly different between cells exposed to Aβ 42 peptides without KP extract compared to cells exposed to Aβ 42 peptides and various concentrations of KP extract (0.5-8 µg/mL) using one-way ANOVA followed by Tukey's test. *, p < 0.05, **, p < 0.01, ***, p < 0.001.
Protective effects of the promising KP extracts KP1, KP2, and KP3 against neurogenesisinhibiting Aβ 42 peptides were also investigated. Intriguingly, protection indicated by the induction of beta-III tubulin and MAP-2 mRNA expressions was observed in all three KP fractions at 4 and 8 µg/mL ( Figure 8D,E). These data indicated that KP1, KP2, and KP3 improved neurogenesis against Aβ 42 peptide treatment. The values are mean ± SD of three independent experiments. The ### shows a significantly different between control and cells exposed to Aβ42 peptides without KP extract at p < 0.001 using one-way ANOVA followed by Tukey's test. The * shows a significantly different between cells exposed to Aβ42 peptides without KP extract compared to cells exposed to Aβ42 peptides and various concentrations of The values are mean ± SD of three independent experiments. The ### shows a significantly different between control and cells exposed to Aβ 42 peptides without KP extract at p < 0.001 using one-way ANOVA followed by Tukey's test. The * shows a significantly different between cells exposed to Aβ 42 peptides without KP extract compared to cells exposed to Aβ 42 peptides and various concentrations of KP extract (1-8 µg/mL) using one-way ANOVA followed by Tukey's test. *, p < 0.05, **, p < 0.01, ***, p < 0.001.

Discussion
Alzheimer's disease (AD) induces gradual cognitive impairment, with major pathological features such as the deposition of extracellular amyloid beta (Aβ) plaques and intercellular neurofibrillary tangles of tau protein. Microglial activation and neuroinflammation by Aβ plaques have been proposed as underlying and connecting components in the development of AD [2][3][4]40]. The involvement of stem cells in adult neurogenesis is critical for memory and cognitive performance [41]. In AD cases, Aβ plaque impairs the proliferation and differentiation of both neural stem cells and neurons [5][6][7] due to excessive induction of neuronal death and oxidative stress [6,7]. Therefore, limiting this vicious cycle by suppressing Aβ toxicity and promoting differentiation and/or proliferation of neural stem cells by safety agents shows promise as a potential therapeutic strategy to delay AD progression. K. parviflora has shown promising neuroprotective activity in several previous studies [42][43][44]. An ethanolic extract of K. parviflora protected glutamate-induced cell injury in mouse hippocampal neuronal cells [43], while K. parviflora protected oxidative stress-related brain damage and memory deficit induced by focal cerebral ischemia in rats [44]. The anti-AD effects targeting AChE and Aβ formation of K. parviflora have been previously studied [45,46] but the protective impacts of K. parviflora on Aβ-induced neurotoxicity, neuroinflammation, and neurogenesis have not yet been examined. In this study, NE-4C neural stem cells (undifferentiated), BV-2 microglia, and NE-4C-derived neurons (differentiated) were used in both monoculture and co-culture of NE-4C neural stem cells and BV-2 microglia, and co-cultures of NE-4C derived neurons and BV-2 microglia, to assess the impact of K. parviflora fractions on neuroprotective and anti-neuroinflammatory Aβ 42 challenges, and on neurogenesis processes compromised by Aβ 42 .
To study the impact of KP extracts on Aβ 42 -disrupted neurogenesis, a low concentration of Aβ 42 (0.25 µM) together with KP extract was applied in a monoculture of NE-4C cells for six days during 12 days of differentiation (Figure 7). The stated exposure period aimed was to assess how Aβ 42 and the extracts impacted early differentiation before neuronal maturation [56]. The three KP fractions (KP1, KP2, and KP3) inhibited the action of Aβ 42 and reversed the expression of specific neuronal mRNA markers, βIII-tubulin and MAP-2. The extracts reduced Aβ 42 -driven IL-6 and iNOS mRNA levels as well as intracellular ROS in microglial cells, coupled with lower IL-6 and NO levels in the coculture medium ( Figure 8). Concerning differentiated neurons from NE-4C cells cocultured with microglia cells, the extracts also restored impaired neuronal development from Aβ 42 in a similar way to the NE-4C cell monoculture. No data exist to demonstrate the influence of K. parviflora on neural stem cell differentiation but K. parviflora protected the rat brain against valproic acid-induced impairments in spatial memory and neural stem cell proliferation in the dentate gyrus [20]. These findings showed that K. parviflora extracts prevented the influence of Aβ 42 on neurogenesis both directly on the functioning of NE-4C cells and indirectly by suppressing microglia-induced neuroinflammation. Apart from antioxidative and anti-inflammatory activities, neurogenesis signal transduction pathways involve ERK1/2, protein kinase A (PKA), Akt, WNT/β-catenin, protein kinase C (PKC), and BDNF [10,12]. Further research into molecular targets and the many biological neurogenetic activities of K. parviflora should concentrate on the ERK1/2, PKA, Akt, and BDNF signaling pathways [15,43,57].
The biological and toxic properties of methoxyflavones have generally focused on DMF. This study examined and compared the anti-inflammation and cytotoxicity of DMF, TMF and PMF in human dermal fibroblasts. TMF exhibited the highest cytotoxicity, while PMF did not influence on cell viability. TMF was the most effective polymethoxyflavone in decreasing TNF-α induced fibroblast damage through the MAPK and NF-κB pathways [58], while PMF was nontoxic and protective against oxidative DNA damage in RAW 264.7 macrophages [59]. Limited evidence exists on the anti-neuroinflammation, neuroprotection, and neurogenesis of DMF, TMF, and PMF individually. Current research is unable to distinguish their activity in each KP fraction or compare their efficacy. Further research is required to determine which methoxyflavone contributes to neuroinflammation and neurogenesis. Moreover, whether the neuroprotective mechanism of the KP extract and its active compound(s) is related to a decrease in ROS or other off-target effects should be investigated further.

Conclusions
Our results revealed that K. parviflora extracts protected against Aβ 42 induced cellular damage of NE-4C neural stem cells, BV-2 microglia, and NE-4C derived neurons and also reduced microglia-induced cellular damage in NE-4C neural stem cells and neurons by Aβ 42 . Moreover, K. parviflora extracts also inhibited the inhibitory effects of Aβ 42 on neuronal differentiation. These findings provide clues about the role of K. parviflora extracts in neuroprotection and neurogenesis and curing Aβ 42 -dependent AD. However, further detailed investigations are required to explore the molecular pathways governing neuroprotective actions against Aβ 42 and the stimulatory effects in the neurogenesis of K. parviflora extracts.