Synergistic Inhibition of Acetylcholinesterase by Alkaloids Derived from Stephaniae Tetrandrae Radix, Coptidis Rhizoma and Phellodendri Chinensis Cortex

Alkaloids having acetylcholinesterase (AChE) inhibitory activity are commonly found in traditional Chinese medicine (TCM); for example, berberine from Coptis chinensis, galantamine from Lycoris radiata, and huperzine A from Huperzia serrata. In practice of TCM, Stephaniae Tetrandrae Radix (STR) is often combined with Coptidis Rhizoma (CR) or Phellodendri Chinensis Cortex (PCC) as paired herbs during clinical application. Fangchinoline from STR and coptisine and/or berberine from CR and/or PCC are active alkaloids in inhibiting AChE. The traditional usage of paired herbs suggests the synergistic effect of fangchinoline–coptisine or fangchinoline–berberine pairing in AChE inhibition. HPLC was applied to identify the main components in herbal extracts of STR, CR, and PCC, and the AChE inhibition of their main components was determined by Ellman assay. The synergism of herb combination and active component combination was calculated by median-effect principle. Molecular docking was applied to investigate the underlying binding mechanisms of the active components with the AChE protein. It was found that fangchinoline showed AChE inhibitory potency; furthermore, fangchinoline–coptisine/berberine pairs (at ratios of 1:5, 1:2, 1:1, and 2:1) synergistically inhibited AChE; the combination index (CI) at different ratios was less than one when Fa = 0.5, suggesting synergistic inhibition of AChE. Furthermore, the molecular docking simulation supported this enzymatic inhibition. Therefore, fangchinoline–coptisine/berberine pairs, or their parental herbal mixtures, may potentially be developed as a possible therapeutic strategy for Alzheimer’s patients.


Introduction
Acetylcholinesterase (AChE) is an enzyme that hydrolyzes acetylcholine in the nervous system [1], and it is a target enzyme for drug development in neurodegenerative disease, including Alzheimer's disease (AD) [2]. AChE inhibitors, e.g., tacrine, donepezil, and rivastigmine, are used to improve cholinergic transmission and to restrain cognitive impairment progress, and they are the common drugs for treatment of AD. Because of the complexity of AD, AChE inhibitors only offer limited symptom relief and produce side effects in patients [3,4]. There is no specific cure for AD; therefore, it is important The linearity curves were constructed by plotting the peak area versus the concentration of each analyte. Each regression equation was derived from six data points (n = 6).

Herbal Combination Synergistically Inhibits AChE
To test the synergistic inhibition of STR-CR and STR-PCC, their combinations (concentration ratio for STR:CR was 85:1, and for STR:PCC was 4.7:1) were tested for AChE inhibition. These ratios were found to have excellent synergy (data not shown). The extracts of STR, CR, and PCC inhibited AChE in dose-dependent manners, with IC 50 values of 570.11 ± 10.41, 4.11 ± 0.15, and 22.19 ± 3.12 µg/mL, respectively, as seen in Figure 3A,B, left. In addition, the STR-CR and STR-PCC combinations showed better inhibition in a dose-dependent manner, and the doses of fangchinoline and coptisine were markedly reduced at the same degree of AChE inhibition. The synergistic effects of STR-CR and STR-PCC combinations in AChE inhibition were further evaluated by the median-effect principle. As shown from the synergism analysis by the median-effect principle in Figure 3A right, when F a < 0.4, the CI value of the STR-CR combination is less than one, suggesting that its inhibitory effect is synergistic at low concentrations. Similarly, the STR-PCC combination showed better synergy in terms of AChE inhibition, as seen in Figure 3B right, CI < 1, when F a < 0.8. STR-PCC combinations in AChE inhibition were further evaluated by the median-effect principle. As shown from the synergism analysis by the median-effect principle in Figure 3A right, when Fa < 0.4, the CI value of the STR-CR combination is less than one, suggesting that its inhibitory effect is synergistic at low concentrations. Similarly, the STR-PCC combination showed better synergy in terms of AChE inhibition, as seen in Figure 3B right, CI < 1, when Fa < 0.8.

Fangchinoline and Coptisine Synergistically Inhibit AChE
Fangchinoline is a strong AChE inhibitor, and it is an abundant alkaloid in STR. Similarly, coptisine is a strong AChE inhibitor and an abundant alkaloid in CR, as seen in Figure 4A. To test synergistic inhibition by fangchinoline and coptisine, different ratios of their combinations (concentration ratio: 1:5, 1:2, 1:1, 2:1) were tested for AChE inhibition. As shown in Figure 4B, different ratios of fangchinoline and coptisine combinations showed good AChE inhibition in a dosedependent manner. In addition, the AChE inhibitory potency of fangchinoline and coptisine combinations (1:5, 1:2, and 1:1) increased, with IC50 values (expressed as concentration of

Fangchinoline and Coptisine Synergistically Inhibit AChE
Fangchinoline is a strong AChE inhibitor, and it is an abundant alkaloid in STR. Similarly, coptisine is a strong AChE inhibitor and an abundant alkaloid in CR, as seen in Figure 4A. To test synergistic inhibition by fangchinoline and coptisine, different ratios of their combinations (concentration ratio: 1:5, 1:2, 1:1, 2:1) were tested for AChE inhibition. As shown in Figure 4B, different ratios of fangchinoline and coptisine combinations showed good AChE inhibition in a dose-dependent manner. In addition, the AChE inhibitory potency of fangchinoline and coptisine combinations (1:5, 1:2, and 1:1) increased, with IC 50 values (expressed as concentration of fangchinoline) of 1.18 ± 0.11, 1.71 ± 0.40, and 1.79 ± 0.44 µM at different ratios, respectively, which are much lower than when using either alkaloid alone.
The synergistic effect of fangchinoline and coptisine combinations on AChE inhibition was further evaluated by the median-effect principle. As shown from the Fa-CI plots in Figure 4C, when F a = 0.5, the CI values of fangchinoline and coptisine combinations (at ratios of 1:5, 1:2, 1:1, and 2:1) were 0.66, 0.88, 0.76, and 0.88, respectively, suggesting synergy between fangchinoline and coptisine at different concentration ratios. Except at the ratio of 2:1, the combinations of fangchinoline and coptisine showed better AChE inhibitory potency at low concentration when F a < 0.7. However, when F a > 0.8, the CI values of their combinations were close to or greater than one, indicating that the inhibitory effect of their combinations might be addictive or antagonistic.

Fangchinoline and Berberine Synergistically Inhibit AChE
Berberine is a well-known AChE inhibitor existed in many herbal medicines, particularly in CR and PCC. Fangchinoline and berberine inhibited AChE in a dose-dependent manner, as seen in Figure 5A. To test the synergistic inhibition of fangchinoline and berberine, different ratios of their combinations (1:5, 1:2, 1:1, 2:1) were tested for AChE inhibition. The different ratios of fangchinoline and berberine combinations showed AChE inhibition in a dose-dependent manner, as seen in Figure 5B. In addition, the AChE inhibitory potency of fangchinoline and berberine combinations (1:5, 1:2, and 2:1) increased largely, with IC 50 values (expressed as concentration of fangchinoline) of 0.19 ± 0.02, 0.55 ± 0.17, and 0.79 ± 0.47 µM, respectively, and the efficiency doses of fangchinoline and coptisine were markedly reduced, as compared to either compound alone. and PCC. Fangchinoline and berberine inhibited AChE in a dose-dependent manner, as seen in Figure 5A. To test the synergistic inhibition of fangchinoline and berberine, different ratios of their combinations (1:5, 1:2, 1:1, 2:1) were tested for AChE inhibition. The different ratios of fangchinoline and berberine combinations showed AChE inhibition in a dose-dependent manner, as seen in Figure  5B. In addition, the AChE inhibitory potency of fangchinoline and berberine combinations (1:5, 1:2, and 2:1) increased largely, with IC50 values (expressed as concentration of fangchinoline) of 0.19 ± 0.02, 0.55 ± 0.17, and 0.79 ± 0.47 μM, respectively, and the efficiency doses of fangchinoline and coptisine were markedly reduced, as compared to either compound alone.  The synergistic effect of fangchinoline and berberine on AChE inhibition was further evaluated by the median-effect principle. When F a = 0.5, the CI values of fangchinoline and berberine combinations (at ratios of 1:5, 1:2, 1:1, and 2:1) were 0.46, 0.73, 0.84, and 0.59, respectively, as seen in Figure 5C. When F a < 0.7, the CI values of fangchinoline and berberine combination were less than one at ratios of 1:5, 1:2, and 2:1, which indicated a synergistic effect between fangchinoline and berberine at different concentration ratios; similarly, when F a < 0.5, the CI value of their combination was less than one at the ratio of 1:1, indicating synergistic action. However, when F a > 0.8, the CI values of fangchinoline-berberine were above one, indicating that the inhibitory effect of their combination might be antagonistic.

AChE Binding Sites Analysis of Fangchinoline, Coptisine and Berberine
According to the docking simulation results, the binding modes of fangchinoline, coptisine, berberine, and donepezil with AChE (PDB code: 1eve) were different in terms of their interactions with the AChE binding pockets. As shown in Table 3, the fitting scores of fangchinoline, coptisine, and berberine were −7.11 ± 0.32, −6.76 ± 0.02, and −7.00 ± 0.01, respectively, and it was consistent with their AChE inhibitory activities. As shown in Figure 6A, there is an H-donor interaction (2.59 Å) between hydroxyl oxygen atoms in the benzene ring of fangchinoline and TYR 334 (peripheral anionic site, PAS active site) of the AChE protein, and a pi-H interaction (4.15 Å) between a pi bond of the aromatic ring in fangchinoline and PHE 288 (the acyl pocket active site) of AChE. As for coptisine and berberine, as seen in Figure 6B,C, there are the same H-π interactions (3.99 and 4.03 Å) between hydrogen atoms in the nitrogen heterocyclic aromatic ring and PHE 330 (gated flexible residues) of AChE. Similarly, there are other H-π interactions (3.90 and 3.67 Å) between hydrogen atoms in the methene group of five oxygen heterocyclic rings and TRP 279 (the PAS active site) of AChE. However, there is an H-π interaction (3.57 Å) between hydrogen atoms in the methene group of berberine's six-membered heterocyclic nitrogen ring and TYR 334 (PAS active site) of AChE. Therefore, coptisine and berberine have a similar binding model with AChE's active pocket, and berberine has more intermolecular interaction with AChE via active binding. Coptisine and berberine can combine with the gated flexible residue PHE 330 of AChE, while fangchinoline has no such binding site. Donepezil is a commonly used therapeutic drug in AD treatment [19]. As shown in Table 3, donepezil can bind with the anionic subsite (TRP 84 ) via H-π interaction (4.06 Å), PAS active site (TYR 334 and TRP 279 ) by H-π interactions (4.13, 3.45/4.03 Å), and the gated flexible active site (PHE 330 ) by π-π interaction (3.54 Å). Interestingly, the binding of fangchinoline, coptisine, or berberine with AChE's active site is largely consistent with that of donepezil. Hence, it is speculated that the combination of fangchinoline, coptisine, or berberine may have a synergistic effect on AChE inhibition. 6-ring TRP84 (A) H-π 4.06 6-ring 6-ring PHE330 (A) π-π 3.54 1 The fitting score is expressed as Mean ± SEM, n = 3, and was negatively correlated with fitting effect. 2 Donepezil served as the positive control in docking.

Discussion
AD is an intractable neurodegenerative disease, and it is characterized by cognitive disorder and brain atrophy with histopathologic changes of senile plaques (SP), neurofibrillary tangles (NFT), cholinergic neurons loss, and gliosis with chronic inflammation [20][21][22][23]. The pathogenesis of AD is complex. The hypotheses of cholinergic defects, Aβ oligomerization, and τ-protein hyperphosphorylation have been proposed [21,24,25]. According to the cholinergic hypothesis, the cholinergic function of the hippocampus and basal ganglia in the forebrain is closely related to

Discussion
AD is an intractable neurodegenerative disease, and it is characterized by cognitive disorder and brain atrophy with histopathologic changes of senile plaques (SP), neurofibrillary tangles (NFT), cholinergic neurons loss, and gliosis with chronic inflammation [20][21][22][23]. The pathogenesis of AD is complex. The hypotheses of cholinergic defects, Aβ oligomerization, and τ-protein hyperphosphorylation have been proposed [21,24,25]. According to the cholinergic hypothesis, the cholinergic function of the hippocampus and basal ganglia in the forebrain is closely related to memory and cognition. Cholinergic deficiency, i.e., cholinergic nerve cell destruction, AChE activity elevation, and acetylcholine level reduction in synaptic cleft, has been identified in AD patients [26]. Hence, it is speculated that increased levels of neurotransmitters could enhance synaptic transmission, ameliorate impaired memory, and restrain the progress of cognitive impairment in AD patients [27]. AChE is an enzyme that hydrolyzes acetylcholine in the nervous system, which has two isoforms in the brain, i.e., a monomer (G1) and a tetramer (G4). The G4 isoform represents the majority of total AChE, and the change of the G4 isoform is closely related to cognition [28][29][30]. Interestingly, AChE can accelerate and induce Aβ oligomerization by forming stable complexes with molecular partners [31][32][33][34]. AChE plays an important role in the cholinergic anti-inflammatory pathway (CAP) by modulating the relationship of ACh and α7nAChR [35,36], which can participate in regulating immunity in the brain. Different kinds of AChE inhibitors have been developed for treatment of AD. However, they only offer limited symptom treatment, due to drug toxicity and the complexity of AD.
In clinical practice, the treatment of AD with TCM has shown good efficiency [37][38][39]. TCMs have been used in treating mental-related problems for thousands of years in China, and many of them show good effect in tranquilizing the mind, relieving anxiety, and enhancing learning and memory. According to the theory of TCM, mental state, emotional activities, especially high-level intellectual activities (e.g., planning and making decisions), are closely related to the functions of the liver and kidney. TCM herbal decoctions could have overall and comprehensive therapeutic effects by matching the corresponding herbs under the guidance of syndrome differentiation [38,39], e.g., the "liver-fire" excess symptom (showing forgetfulness, muttering, flushing, bitter taste, irritability) of AD patients could be relieved by therapies of clearing "liver-fire" and relieving depression when paired with CR and Gardeniae Fructus (as in Huanglian Jiedu Tang), and the sea of medulla insufficiency with "damp-heat" of downward flow symptoms (showing forgetfulness, lumbar debility, deep-colored urine, or difficult, painful urination) could be relieved by therapies of nourishing yin and reducing pathogenic fire of downward flow when paired with PCC and Rehmanniae Radix (Zhibai Dihuang Tang). In current clinical application, a herbal mixture of STR and Rehmanniae Radix (Fangji Dihuang Tang) is used for treatment of vascular dementia by therapies of nourishing yin and promoting diuresis. Among these therapies, herbal mixtures of CR-Gardeniae Fructus, PCC-Rehmanniae Radix, and STR-Rehmanniae Radix are effective paired herbs in AD treatment [17,37]. Here, we propose another herb pair: STR + CR/PCC can be tailored for AD treatment, as the contained alkaloids show synergistic effects towards AChE inhibition.
Alkaloids are a group of naturally occurring compounds in medicinal herbs. There are different varieties of alkaloids in TCM, e.g., galantamine from Lycoris radiata, berberine from Coptis chinensis, and huperzine A from Huperzia serrata. Many of the alkaloids in TCMs have AChE inhibitory activity and show anti-inflammatory activity [40][41][42]. Out of the 150 alkaloids that we tested, 60 showed AChE inhibition. Plant alkaloids are known to have toxicity; however, the alkaloids from TCM could provide more information on their pharmacodynamic effect and toxicity, as their originated herbal medicines have been used in clinics for many years. Here, STR, CR, and PCC are three commonly used TCM herbs in Asia/China. According to the theory of TCM, STR is commonly used in promoting diuresis and clearing damp heat, and CR and PCC are commonly used in clearing damp heat of middle and downward flow. They have been used in dementia treatment for many years at dosages of 6-24 (STR), 9-12 (CR) and 6-12 (PCC) grams [43][44][45]. Fangchinoline is a bisbenzylisoquinoline alkaloid in STR, while coptisine and berberine are isoquinoline alkaloids in CR and PCC. Here, our results supported the clinical usage of STR-CR/PCC as paired herbs. More important, fangchinoline-coptisine/berberine combinations (at ratio of 1:5) could produce similar levels of effect at much lower dosage than that of either alkaloid alone, which suggests that the usage of fangchinoline-coptisine/berberine (at ratio of 1:5) could be much better in terms of drug safety. Our results showed that the required dosage of STR was higher than CR and PCC in STR-CR/PCC paired herbs in terms of AChE inhibition, while the dosage of fangchinoline was lower than coptisine and berberine in fangchinoline-coptisine/berberine combinations (at ratio of 1:5). This could be closely related to the different contents of fangchinoline, coptisine, and berberine in the herbal extracts.
Computer-aided drug screening is a convenient and efficient method for lead compound discovery [46]. Here, molecular docking was used to study the AChE inhibition by alkaloids from STR, CR, and PCC. It is known that the active site of AChE [19] is a deep and narrow gorge with a catalytic site, a PAS active site, and an anionic subsite active site. AChE contains a gated flexible residue active site and acylation active sites. The gated flexible residue (PHE 330 ) is responsible for substrate trafficking down the gorge, and the spatial orientations of PHE 330 are different when complexed with different substrates or inhibitors. As seen from the docking results, both coptisine and berberine could bind with the gated flexible active site (PHE 330 ) and PAS active site (TRP 279 ), while fangchinoline could interact with the acyl active sites (PHE 288 ) and PAS active site (TYR 334 ). It is speculated that fangchinoline and coptisine/berberine could form a ternary complex with AChE by simultaneously binding the PAS active site and the acyl active site. In addition, the synergistic inhibition of AChE by using combinations of alkaloids is closely correlated with enhancement of the affinities between ligands and active sites. Meanwhile, donepezil could bind with the anionic subsite (TRP 84 ), PAS active site (TYR 334 and TRP 279 ), and gated flexible active site (PHE 330 ) by π-π interaction (3.54 Å). Hence, it could also be suggested that the fangchinoline-coptisine/berberine combination has a synergistic effect on AChE inhibition, since the binding of fangchinoline, coptisine, and berberine with AChE's active site is similar to that of donepezil.
AChE can also have noncholinergic effect, i.e., inducing Aβ oligomerization via the PAS active site (TRP 279 ), or regulating inflammation by CAP. The main alkaloids in STR, CR, and PCC could alleviate Aβ-induced injury or have an antineuroinflammatory effect [12,47,48]. The alkaloids in STR, CR, and PCC can affect other factors related with AD: fangchinoline has a protective effect on cyanide-induced neurotoxicity [49,50]; coptisine ameliorates cognitive impairment by inhibiting indoleamine 2,3-dioxygenase [51]; and berberine shows multiple activities to relieve symptoms of AD [52]. In summary, the combination of fangchinoline-coptisine/berberine could provide a potential therapeutic strategy for AD treatment; however, it needs to be studied further.

Determination of Alkaloids in Herbal Extracts
The herb extracts of STR, CR, and PCC were weighed accurately and dissolved in 50% MeOH at concentration of 2.18, 0.40, and 0.40 mg/mL, respectively. The dissolved extract was filtered by a 0.22 µm Millipore filter, and the filtrate was subsequently collected for HPLC determination. The standards of fangchinoline, tetrandrine, coptisine, berberine, palmatine, and epiberberine were weighed accurately and dissolved to stock solution of 1 mg/mL by MeOH. Different volumes of fangchinoline and tetrandrine stocks were mixed to prepare the mixed stock solution of alkaloids in STR, with final concentrations of 100 and 200 µg/mL, respectively. Similarly, the mixed stock solution of alkaloids in CR and PCC was prepared in MeOH, containing epiberberine, coptisine, palmatine, and berberine at 20, 40, 75, 75, and 300 µg/mL, respectively. Then, the mixed stock solutions were diluted to series of working standards with MeOH for HPLC determination. Chromatographic analysis was performed by Waters 2695 HPLC equipped with a UV-VIS photodiode array detector. Sample separation was achieved on an Innoval C 18 column (4.6 × 250 mm, 5 µm) with a constant flow rate of 1.0 mL/min at 25 • C. The mobile phase of STR was composed of MeCN (A) and water (B, 0.1% phosphoric acid), using a gradient elution of 10-30% at 0-20 min, 30-45% at 20-40 min. Detection was performed at 280 nm. The mobile phase of CR and PCC was composed of MeCN and 0.05 M KH 2 PO 4 solution (containing 0.4 g sodium dodecyl sulfate/mL, the pH was adjusted to 4.0 by phosphoric acid). The analyses were detected at 345 nm. The injection volume was set at 10 µL.

Ellman Assay
The mouse brain lysate (0.1 g/mL, containing 10 mM HEPES (pH 7.4), 1 mM EDTA, 1 mM EGTA, 150 mM NaCl, 0.5% triton X-100, 10 µg/mL leupeptin, 10 µg/mL aprotinin, 10 µg/mL pepstatin, and 50 µg/mL benzamidine HCL) was prepared. AChE enzymatic activity was assayed by 96-well microtiter plate with a final volume of 200 µL, according to the Ellman method [53,54]. The assay medium consisted of brain lysate (5 mg/mL) containing different concentrations of testing drugs, 80 mM Na 2 HPO 4 (pH 7.4), 0.1 mM iso-OMPA, 0.625 mM ATCh, and 0.5 mM DTNB. Briefly, the mixture of brain lysate containing the testing drugs, Na 2 HPO 4 buffer, and iso-OMPA solution were incubated at 37 • C for 10 min, and then ATCh and DNTB solutions were added. The reaction solution was incubated at 37 • C for 30 min, and AChE activity was determined by measuring the absorbance at 405 nm. To eliminate the drug-solvent influence on AChE enzymatic activity, the concentration of DMSO was controlled at 0.5% in the 200 µL reaction volume. In order to correct for color interference at 405 nm, a background color contrast group without brain lysate and a drug color contrast group without brain lysate were set, respectively. AChE activity percent of inhibition was calculated by the following equation: AChE activity inhibition (%) = (1 − absorbance with inhibitor/absorbance without inhibitor) × 100%. Tacrine was used as a positive control. Protein concentrations were measured with a kit from Bio-Rad (Hercules, CA, USA).

Synergism Calculation by the Median-Effect Principle
The interaction of herb/alkaloid synergy was evaluated by the median-effect principle [18,55]. The median effect equation (F a /F u = (D/D m ) m ) was used to calculate single or combined dose of herbs/alkaloids required for specific effect according to dose effective curves of single dose and their combination, in which D is the dose of drug, F a is the dose effect (equals to AChE activity inhibition showing with decimal fraction), F u is the unaffected fraction (F u = 1 − F a ), D m is the dose required for 50% effect, and m is the slope of the linear equation (equal to the coefficient of the median effect equation). For convenience of computation, the median effect equation was taken as the logarithm of both sides and transformed to a linear equation: log(F a /F u ) = mlogD − mlogD m . In addition, it was assumed that: y = log(F a /F u ), x = logD, a = −mlogD m ; then, the linear equation (log(F a /F u ) = mlogD − mlogD m ) could be rewritten as a more convenient linear equation: y = mx + a. The linearity curves and equations of herbs/alkaloids were constructed by plotting the dose effect versus the log concentration of each dose effect. Hence, the D m of herbs/alkaloids was calculated according to the equation (D m = (−a/m) 10 ). The combination index (CI) was used to evaluate the synergistic effect of pairing according to the Chou-Talalay equation (CI = (D) 1 /(D x ) 1 + (D x ) 2 /(D x ) 2 ), in which, D 1 and D 2 are the dose of single herb/alkaloid required to produce the x% AChE active inhibition in their combination, respectively, while (D x ) 1 and (D x ) 2 was the dose of single herb/alkaloid in combination required to produce the same effect alone. (D x ) 1 and (D x ) 2 were similarly calculated according to the effect F a (y = log(F a /F u )) by linear equation (y = mx + a), after which, the CI values of herbs/alkaloids pair at different doses effect could be calculated. The values of CI < 1, CI = 1, and CI > 1 refer to a synergistic, additive, and antagonistic effect, respectively.

Molecular Docking
To investigate the underlying binding mechanisms of fangchinoline, coptisine, and berberine with AChE protein, we performed the molecular docking simulation using the MOE Dock module. The AChE 3D structure [19] was obtained from the Protein Data Bank (PDB code: 1eve), and its structure was prepared by correcting the structure issues (such as break bonds, missing loops, etc.), adding hydrogens, and calculating partial charges. The binding pocket was established using the binding site of cocrystalized AChE ligand 1-benzyl-4-[(5,6-dimethoxy-1-indanon-2-yl) methyl]-piperidine. The structures of fangchinoline, coptisine, and berberine were built with MOE Builder module and converted to 3D structures through energy minimization. Classical triangle matching was selected for the placement method, the output docking poses were evaluated by the default London dG score. Then, the rigid receptor method, keeping the ligand-binding groove rigid, was employed in the refinement step. The number of placement and refinement poses was set to 200 and 50, respectively, then the poses were submitted to minimizing using Amber10: EHT forcefield in MOE. The binding mode and the ligand-protein interaction of fangchinoline, coptisine, and berberine were analyzed in MOE after the refinement minimization.

Statistical Analysis
Statistical tests were done by DPS 18.10 software, using the bioassay module [56]. Data was expressed as the Mean ± SEM of three independent experiments, with triplicates of each experiment. The multiple drug effect analysis was used to evaluate the drug interaction according to the median-effect principle, as described in Section 2.5.

Conclusions
Both STR-CR and STR-PCC paired herbs showed synergistic effects on AChE inhibition, and the main alkaloids in the herbal extracts showed good synergistic effects. Fangchinoline-coptisine combinations (at ratios of 1:5, 1:2, and 1:1) and fangchinoline-berberine combinations (at ratios of 1:5, 1:2, and 2:1) showed good synergistic effects on AChE inhibition. Among the different ratios of fangchinoline-coptisine/berberine combinations, the ratio of 1:5 showed a better synergistic effect on AChE inhibition than the other ratios. In addition, the molecular docking results showed that fangchinoline, coptisine, and berberine could bind with the AChE enzyme in the active pocket site, which might result in the synergistic effect on AChE inhibition.