Single compounds, including embelin, loureirin A, loureirin B, pterostilbene, resveratrol, and syringic acid (≥98% purity, determined by HPLC) were purchased from Chengdu Biopurify Phytochemicals Ltd. (Chengdu, China). Native mouse nerve growth factor (NGF) was obtained from Alomone Labs (Jerusalem, Israel). Dexamethasone, dimethyl sulfoxide (DMSO), lipopolysaccharide (LPS), and 3-(4,5-dimethylthiazol)-2,5-diphenyl-tetrazolium bromide (MTT) were purchased from Sigma-Aldrich (St. Louis, MO).

Nine botanical drugs, composed of Ardisia elliptica fruit, AEF [Primulaceae; Ardisia elliptica Thunb.], Baliospermum solanifolium root, BSR [Euphorbiaceae; Baliospermum solanifolium (Burm.) Suresh], Dracaena cochinchinensis stemwood, DCS [Asparagaceae; Dracaena cochinchinensis (Lour.) S.C. Chen], Mesua ferrea flower, MFF [Calophyllaceae; Mesua ferrea L.], Phyllanthus emblica fruit, PEF [Phyllanthaceae; Phyllanthus emblica L.], Terminalia bellirica fruit, TBF [Combretaceae; Terminalia bellirica (Gaertn.) Roxb.], Terminalia chebula fruit, TCF [Combretaceae; Terminalia chebula Retz.], Tarenna hoaensis stemwood, THS [Rubiaceae; Tarenna hoaensis Pit.], and Zingiber officinale root, ZOR [Zingiberaceae; Zingiber officinale Roscoe] were obtained from Tha Phra Chan Herb Co., Ltd. (Bangkok, Thailand) in 2020 and authenticated by Dr. Tina Ting-Xia Dong. The voucher specimens, including AEF#2020-11-25, BSR#2020-11-25, DCS#2020-11-25, MFF#2020-11-25, PEF#2020-11-25, TBF#2020-11-25, TCF#2020-11-25, THS#2020-11-25, and ZOR#2020-11-25, were deposited at the Center for Chinese Medicine, The Hong Kong University of Science and Technology.

A traditional multi-herbal formulation is composed of an equal amount of nine botanical drugs, according to The Royal Medicinal Book of King Rama V. The aqueous extracts from BHP and individual botanical drugs were prepared as follows: the coarse powder of each botanical drug (5 g) was dissolved in 200 mL boiling water and then soaked for 2 h. The solution was centrifuged at 20,000 rpm, 4°C for 10 min. The supernatant was freeze-dried. The ethanol extracts from BHP of DCS and AEF were prepared as follows: the total coarse powder of DCS:AEF (10 g), combined in nine different ratios, including 1:9, 2:8, 3:7, 4:6, 5:5, 6:4, 7:3, 8:2, and 9:1 w/w, was extracted in 90% ethanol (400 mL), and sonicated for 20 min. BHP of DCS and AEF were extracted twice. The solution was concentrated using a rotary evaporator, and then freeze-dried. In parallel, the ethanol extract from individual botanical drugs was conducted in the same manner as the BHP of DCS and AEF at a specific ratio of 1:9 w/w. The coarse powder of DCS (1 g) or AEF (9 g) was extracted in 90% ethanol (400 mL) and then sonicated for 20 min twice. The extracting yields were approximately 7.08% for DCS:AEF (1:9), 20.58% for DCS, and 6.17% for AEF (Supplementary Table S1).

The botanical drug extracts (at 10 mg/mL) were analyzed by HPLC-DAD using a TC-C18 column (4.6 × 250 mm, 5 μm) (Agilent, Santa Clara, CA). The mobile phase comprised 0.2% formic acid (solvent A) and acetonitrile (solvent B). The gradient program was employed as follows: 10%-15% (B) at 0-10 min; 15%-25% (B) at 10-30 min; 25%-35% (B) at 30-55 min; 35%-40% (B) at 55-65 min; 40%-60% (B) at 65-90 min; 60%-80% (B) at 90-95 min, and 80%-100% (B) at 95-120 min. The flow rate was 1.0 mL/min. The injection volume was 5 μL. The detection wavelength was performed at 295 nm. Reference compounds, including loureirin A, loureirin B, pterostilbene, and resveratrol, were used to represent the constituents of DCS. Similarly, syringic acid and embelin were employed as reference compounds for AEF.

Rat pheochromocytoma PC12 cells were obtained from American Type Culture Collection (ATCC 1721, Manassas, VA). PC12 cells were cultured in DMEM supplemented with 6% fetal bovine serum (FBS), 6% horse serum, and 1% penicillin-streptomycin (10,000 U/mL). The cells were under starvation in DMEM condition in 1% FBS and 1% horse serum for 24 h before the treatment. Mouse microglial BV2 cells were purchased from National Infrastructure of Cell Line Resource (NICR, Beijing, China) with Cat# 4201MOU-CCTCC00311. BV2 cells were grown in DMEM containing 10% heat-inactivated FBS and 1% penicillin-streptomycin (10,000 U/mL). Both PC12 cells and BV2 cells were cultured until passage 20. Cell viability was determined using an MTT assay. The cultures were seeded in a 96-well plate for 24 h. BV2 cells, 2 × 10⁴ cells/well, and PC12 cells, 1 × 10⁴ cells/well, were treated with botanical drug extracts or single compounds for 24 h and 48 h, respectively. Following the treatment, MTT reagent was added to each well at a final concentration of 0.5 mg/mL. The absorbance of the formazan was measured at 570 nm.

Cultured PC12 cells were transfected with DNA, which constructed the reporter gene of pNF68-Luc or pNF200-Luc, using jetPRIME (Polyplus Transfection, New York), according to the manufacturer’s protocol. The cells were treated with botanical drug extracts (0.5-20 μg/mL) or single compounds (0.5-10 µM), with or without a low dose of NGF (1.5 ng/mL), for 24 h. The cells were lysed by passive lysis buffer (Promega, Madison, WI). Cell lysates were performed luciferase activity using a kit (Promega, Madison, WI), and normalized by protein concentration.

Cell differentiation was examined by the extension of neurites. PC12 cells, 3 × 10⁴ cells/well, were seeded in a 12-well plate. The cultures were treated with botanical drug extracts, with or without a low dose of NGF (1.5 ng/mL), for 48 h. The cells treated with NGF at 50 ng/mL were performed as a positive control. After the treatment, the images of cell morphology, including neurite length, were captured randomly under the phase-contrast microscope with ×10 objective lens (Carl Zeiss, Oberkochen, Germany) using SPOT imaging software (Diagnostic Instruments, MI). The length of neurites of 100 cells was measured using ImageJ software. The neurite outgrowth was characterized as follows: neurite length was less than 15 μm, 15–30 μm, and more than 30 μm.

The cultured PC12 cells, 1.5 × 10⁵ cells/well in a 12-well plate, were treated with botanical drug extracts (0.5-20 μg/mL), with or without NGF (1.5 ng/mL), for 24 h. NGF at 50 ng/mL was used as the positive control for stimulating neurofilament expression. The cultured BV2 cells, 1.6 × 10⁵ cells/well in a 12-well plate, were pretreated with botanical drug extracts (0.1-10 μg/mL) or single compounds (0.1-5 μM) for 4 h, and then exposed to LPS (100 ng/mL) for 20 h. Dexamethasone at 10 μM was employed as the positive control for suppressing pro-inflammatory molecule mRNA levels. Total RNA was extracted by RNAzol^®RT (Molecular Research Center, OH), according to the manufacturer’s instruction, and converted to cDNA by PrimeScript RT Master Mix (Takara Bio, Japan). Real-time qPCR was conducted using Roche SYBR Green Master mix (Roche, Switzerland) under Roche 480 multiplex quantitative PCR system. The relative mRNA levels of target genes were quantified by the 2^−ΔΔCT method and normalized by GAPDH mRNA. The sequence of primers was presented in Supplementary Table S2.

The expression of neurofilaments was determined using Western blotting. After the treatment of PC12 cells with botanical drug extracts (1-20 μg/mL), with or without NGF (1.5 ng/mL), for 48 h, the cultures were lysed with low salt lysis buffer containing protease inhibitors, including aprotonin, leupeptin, benzamidine and pepstatin A. Cell lysates were centrifuged at 12,000 rpm for 15 min at 4°C. The supernatants were collected, and the amounts of protein were measured using the Bradford assay. Proteins were separated in 8% SDS-polyacrylamide gel electrophoresis and then transferred to nitrocellulose membranes. The membranes were blocked with 5% bovine serum albumin and probed with primary antibodies directed against neurofilament 160 (NF160) and α-tubulin (Cell Signaling Technology, Danvers, MA) at 4°C overnight. Subsequently, the membranes were incubated with secondary antibodies and exposed to an enhanced chemiluminescence (ECL) Western blotting detection kit (Thermo Fisher Scientific, Waltham, MA). The protein bands were visualized using the ChemiDoc Touch Imaging System (Bio-Rad). The intensity of protein bands was analyzed by ImageJ.

The morphology of BV2 cells was illustrated by using immunofluorescence staining. BV2 cells, 1 × 10⁵ cells/dish, were cultured in a 35-mm² confocal dish for 24 h. The cells were pretreated with botanical drug extracts at 10 μg/mL for 4 h, before treatment of 100 ng/mL LPS for 20 h. Dexamethasone at 10 μM was employed as a positive control to suppress LPS-induced inflammation. Subsequently, the cells were fixed with 4% paraformaldehyde for 20 min and blocked with 5% bovine serum albumin in PBS containing 0.2% Triton X-100 for 1 h. The cells were then incubated overnight at 4°C with an anti-Iba1 antibody (ab178846, Abcam) at a dilution of 1:500. Alexa Fluor 647-conjugated anti-rabbit antibody was used as the secondary antibody at a dilution of 1:200 and incubated for 2 h. The nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI, 1:500). The cells were visualized under a Leica SP8 confocal microscope (Wetzlar, Germany) using a ×40 oil immersion objective.

Purified synthetic Aβ_1-42 (GL Biochem, Shanghai, China) was dissolved in 100% hexafluoroisopropanol and sonicated for 20 min at room temperature. The Aβ_1-42 solution was dried out overnight in the fume hood. Aβ_1-42 fibrils were prepared by re-suspending the dried Aβ_1-42 peptide film (0.5 mg) with 20 μL DMSO and diluting with 10 mM HCl to a final concentration of 100 µM. The Aβ_1-42 solution was vortexed for 1 min and incubated at 37°C for 6 days.

Aβ_1-42 fibrils were determined using thioflavin T (ThT) fluorescence dye. The morphology of Aβ₁₋₄₂ fibril or aggregate forms was visualized using atomic force microscopy (AFM). To investigate the inhibition of Aβ_1-42 fibril formation, Aβ_1-42 monomers (10 µM) were incubated with botanical drug extracts (0.5-20 μg/mL) or single compounds (0.5-10 μM) at 37°C for 6 days. To examine the disassembly of Aβ aggregation, Aβ_1-42 fibrils (10 µM) were aggregated with botanical drug extracts (0.5-20 μg/mL) at 37°C for 5 days. Curcumin (30 µM) was performed as a positive control. ThT fluorescence dye was added with a final concentration of 20 µM. The intensity of ThT fluorescence was measured at an excitation wavelength of 435 nm, and emission wavelength of 488 nm, which was conducted in a 96-well black plate. The AFM images were visualized after applying the samples (30 µL) onto the mica sheets (1 × 1 cm²) using tapping mode.

The synergistic effects of botanical drug combination were performed by the Chou-Talalay method (Chou, 2006) using CompuSyn software 1.0 (ComboSyn, Paramus, NJ). The combination index (CI) was identified as synergistic, additive, and antagonistic effects, where CI < 1, CI = 1, and CI > 1, respectively.

Data were analyzed in the form of concentration-response curve of BHP compared to individual botanical drugs and performed at least three independent experiments (n ≥ 3), which were presented in mean ± SEM. The statistical analysis was conducted using one-way analysis of variance (ANOVA) with Tukey’s post hoc test by GraphPad Prism 9.0.0. The statistical significance was defined by (*) p < 0.05, (**) p < 0.01, and (***) p < 0.001.

Thai traditional medicines have been utilized for their therapeutic effects since ancient times. However, several traditional remedies still require identifying and verifying their therapeutic functions. In this study, the aqueous extract from a traditional multi-herbal formulation, documented in the Royal Medical Book of King Rama V, volume II, in the section of Sabba-Lakkhana Sabba-Guna Scripture, on page 376, lines 11–13, was investigated for its potential activities in promoting neuronal differentiation and reducing Aβ_1-42 fibril formation. Neuronal differentiation is characterized by the expression of neurofilaments that play a vital role in cell differentiation and maintenance (Yuan and Nixon, 2021). Cultured PC12 cells were transfected with pNF68-Luc or pNF200-Luc, and then treated with non-toxic concentrations, 10-100 μg/mL of aqueous extracts from BHP of nine botanical drugs and individual botanical drugs (Supplementary Figure S1A). Treatment with aqueous extracts of the BHP or single botanical drugs resulted in a dose-dependent increase in luciferase activities of the reporter genes (Figures 1A, B). Among the botanical drug extracts, DCS showed a robust increase in the activities of pNF68-Luc and pNF200-Luc. The BHP and other individual botanical drugs did not show strong responses. Besides, the formation of Aβ_1-42 fibrils and aggregates was assessed using ThT fluorescence dye (Xue et al., 2017). Co-incubation of 10 μM Aβ_1-42 monomers with aqueous extracts from the BHP or individual botanical drugs led to a dose-dependent reduction in ThT fluorescence intensity (Figure 1C). Notably, the aqueous extracts from AEF, MFF, and TBF showed greater ability to decrease ThT fluorescence intensity, as compared to other extracts (Figure 1C). Taken together, these findings suggested that the BHP of nine botanical drugs exhibited non-synergistic effects and limited potential for neuroprotection when compared to other botanical drugs.

To investigate potential synergistic effects, the aqueous extracts from BHP of two botanical drugs were tested at a fixed combination of 1:1 w/w ratio. Previous studies have reported that DCS alone could promote neuronal differentiation and inhibit Aβ_1-42 aggregation (Ospondpant et al., 2022). Therefore, DCS was selected as the main component to determine potential activity in BHP. The aqueous extract from BHP of two botanical drugs, including DCS:AEF, DCS:BSR, and DCS:THS at 100 μg/mL, could increase the luciferase activities of NF-68 and NF-200 promoters by 150%-200% (Figure 1D). In addition, BHP of DCS and AEF, or MFF, resulted in approximately 80% decrease in ThT fluorescence intensity (Figure 1E). Among these BHP, DCS:AEF exhibited the most promising effect, and which therefore was selected for further analysis of synergism.

Previous findings have demonstrated that the ethanol extract of DCS exhibited a higher yield and greater pharmacological activities, as compared to the water extract (Ospondpant et al., 2022; Ospondpant et al., 2023). To develop a promising BHP of DCS and AEF, various ratios of DCS and AEF ethanol extracts were conducted to determine the optimal level of synergism. The BHP of DCS:AEF at the specific ratio of 1:9 w/w showed higher luciferase activities of neurofilament promoters, approximately 100%-120% increase, as compared to other ratios (Figure 2A). Consequently, the ethanol extract from BHP of DCS and AEF at 1:9 w/w ratio, denoted as DCS:AEF (1:9), was chosen for further investigation of its synergistic interactions in promoting neuronal differentiation, inhibiting Aβ aggregation, and suppressing neuroinflammation. First, HPLC analysis was performed to characterize the ethanol extracts of DCS:AEF (1:9), DCS, and AEF, and the HPLC fingerprints of these extracts were illustrated in Figure 2B. Pterostilbene and embelin were identified as the predominant peaks in DCS and AEF extracts, respectively. The contents of identified compounds in DCS:AEF (1:9), DCS, and AEF were determined using the HPLC method, as shown in Supplementary Table S1. The extract of DCS:AEF (1:9) contained approximately 5.56 μg/mg of pterostilbene and 73.08 μg/mg of embelin. The cytotoxicity assessment on BV2 cells and PC12 cells revealed that these botanical drug extracts had non-toxic concentrations of up to 10 μg/mL and 20 μg/mL, respectively (Supplementary Figures S1B, C), which were employed for subsequent analysis.

The treatment of PC12 cells with DCS:AEF (1:9) or DCS showed a significant increase in pNF68-Luc and pNF200-Luc activities (Figures 2C, D), as well as an upregulation of neurofilament medium chain (Nefm) mRNA level (Figure 2E), in a dose-dependent manner. In parallel, AEF induced a slight increase in luciferase activities and mRNA expression (Figures 2C–E). The combination of DCS and AEF, DCS:AEF (1:9) at 0.5-20 μg/mL, exhibited a synergistic effect in enhancing pNF200-Luc activity, as indicated by the CI value less than 1 (Figure 2D). However, only DCS:AEF (1:9) at 0.5 μg/mL demonstrated synergistic activities in increasing luciferase activity of NF68 promoter and mRNA expression of Nefm, as indicated by the CI value less than 1 (Figures 2C–E).

Neurotrophic factors, such as nerve growth factor (NGF), play a crucial role in enhancing neuronal survival, development, and differentiation. The level of NGF was decreased during aging and neurodegenerative diseases (Connor and Dragunow, 1998). Here, the combined effect of a low dose of NGF (1.5 ng/mL) and botanical drug extracts on neuronal differentiation, using PC12 cells as a model system, was determined. Treatment with DCS:AEF (1:9), DCS, AEF, or 1.5 ng/mL NGF alone did not significantly increase the length of neurites (Figure 3A). However, DCS:AEF (1:9) at 20 μg/mL combined with NGF at 1.5 ng/mL showed a significant extension in neurite length (Figure 3A). In addition, the combination of DCS:AEF (1:9) and 1.5 ng/mL NGF led to an increase in luciferase activities of NF68 and NF200 promoters in dose-dependent manners (Figures 3B, C). The co-treatment of DCS:AEF (1:9) and 1.5 ng/mL NGF also increased Nefm mRNA level (Figure 3D). The protein expression of neurofilament 160 (NF160) was determined using Western blot analysis. The co-treatment of DCS:AEF (1:9), 5-20 μg/mL, and 1.5 ng/mL NGF significantly upregulated NF160 protein level by approximately 2-fold (Figure 3E).

To determine the inhibitory effect of botanical drug extracts on Aβ_1-42 fibril formation, DCS:AEF (1:9), DCS, or AEF, was co-incubated with 10 μM Aβ_1-42 monomers for 6 days. DCS:AEF (1:9) significantly reduced the intensity of ThT fluorescence by 20%-80%, in a concentration-dependent response (Figure 4A). Similarly, DCS and AEF individually showed significant decreases in ThT fluorescence intensities by 30%-60% and 40%-70%, respectively (Figure 4A). Notably, BHP of DCS and AEF exhibited a stronger effect in blocking Aβ_1-42 fibril formation, as indicated by the CI value less than 1 (Figure 4B). The morphology of Aβ_1-42 fibrils was observed under AFM with a scan size of 10 μm. After co-incubation with botanical drug extracts, the formation of Aβ_1-42 fibrils was visibly reduced (Figure 4C).

To examine the disassembly of Aβ_1-42 aggregates, the botanical drug extracts were co-incubated with 10 μM Aβ_1-42 fibrils for 5 days. DCS:AEF (1:9), DCS, and AEF showed dose-dependent reductions in ThT fluorescence intensity, indicating the disruption of Aβ_1-42 aggregation (Figure 5A). DCS:AEF (1:9) significantly decreased ThT fluorescence intensity by 30%-70%, while DCS and AEF reduced ThT fluorescence intensity by 30%-40%, and 25%-60%, respectively (Figure 5A). The synergistic activity of DCS:AEF (1:9) in disrupting Aβ_1-42 aggregation was observed, as indicated by the CI value less than 1 (Figure 5B). AFM analysis further confirmed the reduction of Aβ_1-42 aggregates upon co-application of the botanical drug extracts (Figure 5C).

Subsequently, the impact of Aβ_1-42 fibrils on cell viability was investigated in cultured PC12 cells using two different testing designs. First, cultured PC12 cells were pretreated with botanical drug extracts before exposure to Aβ_1-42 fibrils. Treatment with 10 μM Aβ_1-42 fibrils alone resulted in approximately 40% loss of cell viability (Figure 6A). Notably, DCS:AEF (1:9) and DCS significantly decreased the Aβ_1-42 fibril-induced cell death by 20%-60%, and 30%-50%, respectively, in dose-dependent responses (Figures 6A, B). However, AEF did not show a significant increase in cell viability in the presence of Aβ_1-42 fibrils. The BHP of DCS and AEF at 0.5-20 μg/mL exhibited a strong effect in decreasing Aβ_1-42 fibril-induced cell death, as indicated by the CI value less than 1 (Figure 6C). In the second testing method, co-incubation of 10 μM Aβ_1-42 monomers with DCS:AEF (1:9), DCS, or AEF for 6 days before cell treatment significantly reduced Aβ_1-42 fibril-induced cell death in dose-dependent manner, as compared to treatment with Aβ_1-42 fibrils alone (Figures 6D, E). This finding suggested that the inhibitory effect of the botanical drug extracts on Aβ_1-42 fibril formation was associated with the prevention of Aβ_1-42 fibril-induced loss of cell viability. DCS:AEF (1:9) at 10-20 μg/mL completely restored cell viability in the presence of Aβ_1-42 (Figure 6D). The synergistic effect of DCS:AEF (1:9) in reducing Aβ_1-42 fibril-induced cell toxicity was indicated by the CI value less than 1 (Figure 6F).

The synergistic effect of DCS:AEF (1:9) in suppressing inflammation was determined by analyzing the mRNA expressions of pro-inflammatory markers, including TNF-α, IL-1β, and iNOS, using RT-qPCR. Treatment of BV2 cells with DCS:AEF (1:9), DCS, or AEF significantly decreased the LPS-induced mRNA expressions of TNF-α, IL-1β, and iNOS in dose-dependent manners (Figures 7A–C). DCS:AEF (1:9) showed a significant reduction of TNF-α, IL-1β, and iNOS mRNA levels by 25%-60%, 30%-85%, and 30%-80%, respectively. The suppression of LPS-induced TNF-α, IL-1β, and iNOS mRNA expressions demonstrated the synergistic interaction between the BHP of DCS and AEF at 5-10 μg/mL, as indicated by the CI value less than 1 (Figures 7B, C). In addition, the morphological changes, induced by inflammation, were examined in BV2 microglial cells. LPS-treated BV2 cells displayed roundish or amoeboid-like shapes with fewer and shorter branches (Figure 7D). However, the treatments with DCS:AEF (1:9), DCS, or AEF at 10 μg/mL, as well as dexamethasone (DEX, 10 μM), resulted in cells presenting spindle shapes, with long thin processes extending from the cell body, that was resembling the morphology of untreated cells (Figure 7D).

The toxicity of six compounds presented in the BHP of DCS and AEF, namely, loureirin A, loureirin B, resveratrol, pterostilbene, embelin, and syringic acid, was determined in cultured BV2 cells and PC12 cells. Non-toxic concentrations of these compounds, up to 5 μM for BV2 cells, and 10 μM for PC12 cells (Supplementary Figures S1D, E), were employed to determine their activities in promoting neuronal differentiation, inhibiting Aβ_1-42 fibril formation, preventing Aβ_1-42 fibril-induced cell toxicity, and suppressing LPS-induced pro-inflammatory molecule expression. Treatment of PC12 cells with loureirin A, loureirin B, resveratrol, and pterostilbene showed a significant increase in luciferase activities of NF68 and NF200 promoters, in dose-dependent responses (Figures 8A, C). In parallel, when these compounds were combined with a low dose of NGF at 1.5 ng/mL, the induction of pNF68-Luc and pNF200-Luc activities was higher by 50%-100% (Figures 8B, D). Treatment of cells with embelin and syringic acid, with or without 1.5 ng/mL NGF, did not show a significant increase in pNF68-Luc and pNF200-Luc activities.

The co-incubation of Aβ_1-42 monomers with resveratrol or embelin, at 0.5-10 μM, significantly reduced ThT fluorescence intensity by 20%-60%, and 25%-90%, respectively, in dose-dependent manners (Figure 8E). In contrast, loureirin A, loureirin B, pterostilbene and syringic acid showed non-significant decrease in ThT fluorescence intensities. Treatment of PC12 cells with loureirin A, resveratrol and pterostilbene significantly decreased Aβ_1-42 fibril-induced cell toxicity by 40%, 40%-60%, and 30%-60%, respectively, in dose-dependent manners (Figure 8F). However, loureirin B, embelin and syringic acid did not significantly restore cell viability in the presence of Aβ_1-42 fibrils (Figure 8F).

In cultured BV2 cells treated with 100 ng/mL LPS, loureirin A, loureirin B, pterostilbene, embelin, and syringic acid, at 5 μM, showed a significant reduction in TNF-α mRNA levels by 25%-45%. These compounds, at 0.5-5 μM, significantly decreased the LPS-induced IL-1β mRNA levels by 25%-70%. Notably, syringic acid exhibited the most potent effect in reducing IL-1β mRNA expressions. Treatment with these compounds, at 5 μM, also significantly decreased iNOS mRNA expressions by 30%-40% (Figure 8G). Resveratrol showed slight inhibitory effects on the suppression of LPS-induced mRNA expression of pro-inflammatory molecules (Figure 8G).

To determine the effects of PHP mimicking the composition of DCS:AEF (1:9), assays were conducted following a similar approach as with botanical drug extracts. PHP was composed of six phytochemicals, including loureirin A, loureirin B, resveratrol, pterostilbene, embelin, and syringic acid, denoted as PHP-6. The content of PHP-6, accounted for in the BHP of DCS and AEF at 20 μg/mL, contained 0.16 µM loureirin A, 0.18 µM loureirin B, 0.16 µM resveratrol, 0.42 µM pterostilbene, 4.96 µM embelin, and 0.39 µM syringic acid, were determined using HPLC analysis (Supplementary Table S1). In PC12 cells transfected with pNF68-Luc or pNF200-Luc, the treatments of DCS:AEF (1:9), PHP-6, and pterostilbene exhibited an enhancement of luciferase activities (Figures 9A, B). However, the other phytochemicals did not induce luciferase activities, with or without a low dose of NGF. These results suggested that pterostilbene, at least partly, might mediate the activation of neurofilaments. In addition, the co-incubation of 10 μM Aβ_1-42 monomers with embelin, PHP-6, or DCS:AEF (1:9) significantly decreased ThT fluorescence intensity (Figure 9C). Embelin and PHP-6 exhibited similar activities in reducing ThT fluorescence intensity by approximately 65%, suggesting that the presence of embelin might predominantly contribute to the effect observed in PHP-6. However, DCS:AEF (1:9) showed greater inhibitory activity compared to PHP-6 (Figure 9C). DCS:AEF (1:9) significantly reduced cell loss induced by 10 μM Aβ_1-42 fibrils, while individual chemicals or the PHP-6 did not show a significant increase in cell survival (Figure 9D). Next, to examine the anti-inflammatory properties of PHP-6, a combination of 0.04 µM loureirin A, 0.045 µM loureirin B, 0.04 µM resveratrol, 0.11 µM pterostilbene, 1.24 µM embelin and 0.98 µM syringic acid, was replicated in DCS:AEF (1:9) at 5 μg/mL. In the presence of LPS, the treatment with DCS:AEF (1:9) showed a greater decrease in mRNA expressions of TNF-α, IL-1β, and iNOS compared to PHP-6 (Figure 9E).