Excavating Precursors from the Traditional Chinese Pair Herbs Polygala tenuifolia and Gastrodia elata: Synthesis, Anticonvulsant Activity Evaluation of 3,4,5-Trimethoxycinnamic Acid (TMCA) Peptide Analogs

TL;DR

This study creates and tests new compounds from traditional Chinese herbs to find potential treatments for epilepsy.

Contribution

The paper introduces novel TMCA peptide analogs derived from TCM herbs with anticonvulsant properties.

Findings

Compounds 11, 19, 22, and 23 showed strong anticonvulsant activity in seizure models.

Compounds 19 and 23 modulated GABA-related proteins and affected EEG patterns in epilepsy models.

Abstract

Background: Epilepsy comprises a range of disorders affecting the central nervous system (CNS) characterized by recurrent seizures, impacting approximately 60 million individuals globally. In this study, we designed and derived a series of peptide analogs 1–30 of 3,4,5-trimethoxycinnamic acid (TMCA) from the herbal combinations of Polygala tenuifolia and Gastrodia elata, recognized in Traditional Chinese Medicine (TCM). Methods: All the analogs were synthesized, and their anticonvulsant efficacy was subsequently assessed. The anticonvulsant activity of all TMCA analogs was evaluated using two acute seizure models in mice: the maximal electroshock (MES) and the sc-pentylenetetrazole (PTZ) models. Furthermore, we explored the electroencephalogram (EEG) and double-labeling immunofluorescence experiments were carried out as well. Results: Our findings indicated that compounds 11, 19, 22, and 23 demonstrated favorable anticonvulsant activities during the initial assessment. Compounds 19 and 23 also played roles in controlling the onset of epilepsy in EEG, modulating levels of GABA aminotransferase (GABA-AT)/gamma-aminobutyric acid type A receptor (GABAAR), interacting with active sites of GABA-AT and GABAAR obtained from docking simulation methods. Conclusions: Novel derivatives in this study provide new cores for further design and optimization inspired by TCM herb pair drugs P. tenuifolia and G. elata, with the aim of exploring new anticonvulsant agents.

Figures

Tables

Table 2. Further screening for selected potential TMCA derivatives.

Com.	MES ED50	Sc-PTZ	TOX TD50	PI a(MES)
		Latent Time (Seconds in Dosage of 100 mg/kg)
NS	---	150.8 ± 16.3	---	---
CBZ b	8.8 (5.5–14.1)	---	71.6 (45.9–135)	8.1
11	85.6 (60.3–113.7)	169.4 ± 17.9	313.7 (240.8–398.5)	3.7
19	27.3 (14.6–39.1)	270.3 ± 11.2 **	325.4 (267.9–378.5)	11.9
22	53.4 (49.7–68.3)	243.6 ± 19.7 **	387.1 (273.9–413.5)	7.2
23	42.2 (31.5–59.7)	235.8 ± 10.5 **	896.4 (822.5–981.7)	21.2

Full text

1. Introduction

Epilepsy is a persistent and crippling neurological disorder that impacts around 0.5–1% of the worldwide population [1,2,3]. Although extensive research has been conducted over the decades, anti-seizure medications (ASMs) continue to serve as the fundamental approach in the clinical management of this disorder. Natural products have offered a fresh viewpoint in this area. Traditional Chinese Medicine (TCM) represents a significant natural avenue that meets the criteria for safety and efficacy while mitigating the adverse effects associated with epilepsy treatments [4]. According to prior studies [5,6,7,8], we have identified the anticonvulsant effects of precursors, such as TMCA derived from P. tenuifolia and vanillyl alcohol sourced from G. elata, providing evidence of the anticonvulsant efficacy of TMCA derivatives (Figure 1).

In this article, we design a cluster of TMCA peptide analogs (Scheme 1, 1–30). Among the analogs, the cinnamamide moiety is substituted with TMCA or 2-Cl cinnamic acid, in which TMCA is an important promising CNS agent targeting the GABAA/BZ receptor [9], 5-HT (5-hydroxytryptamine) [10], AChE (acetylcholine) [11] and EP2 [12], including as a lead compound, and the introduction of 2-Cl cinnamic acid in order to investigate the influence of an electron-withdrawing group (EWG) on the anticonvulsant potency. Peptides have been disclosed to show anticonvulsant potency [13], herein, the marked anticonvulsant drug pregabalin [14] and gabapentin [15] are used to prepare novel derivatives. As for the residual groups for the formation of peptide bond or ester bond, we choose the six-membered heterocyclic ring analogs, including piperidine (S2), morpholine (S3), pyrrolidine (S4), and piperazine (S5), which are structural motifs frequently present in the CNS agents [16,17,18,19], and isopentenyl vanillyl alcohol (S6), a moiety derived from vanillyl alcohol in G. elata, has been proven to be a bioactive fragment by our previous work [5,20,21]. Based on our previous research, the isopentenyl vanillyl alcohol group could provide good activity and weak toxicity, as for the substituted cinnamic acid derivatives, we introduced an electron-withdrawing 2-Cl cinnamic acid moiety with favorable anticonvulsant activity to enrich the SAR. All the synthetic derivatives have been characterized and evaluated for anticonvulsant potency using the classic models, with the information about neurotoxicity, EEG, double-labeling immunofluorescence, molecular docking analysis, and SAR study detailed in the article.

2. Results

2.1. Chemistry

All titled compounds (1–30) were successfully prepared through the synthetic protocols presented in Scheme 1, and the structurally simple intermediates 1, 2, 3, 9, 10, 16, 17, 18, 24, and 25 were prepared and directly used to give subsequent derivatives. Characterization spectra are listed in the Supplemental Materials. From the results, we can see that most of the amide derivatives can be successfully prepared with the DCM/MeOH system, and the ester derivatives can be separated with the Hexane/EtOAc system, indicating appropriate polarity. It can be generalized from the NMR spectra that the cis-trans isomerism of TMCA derivatives was not changed (the gathering pair of double peaks nearby δ 7.5 and δ 6.5 in ^1^H NMR, for instance, of compound 4, the coupling constant was higher than 12, suggesting the trans-conformation was preserved in Supplementary Materials), which was consistent of the mechanism of acylation reaction.

(E)-N-((1-(2-oxo-2-(piperidin-1-yl)ethyl)cyclohexyl)methyl)-3-(3,4,5-trimethoxyphenyl)acrylamide (4): yellow solid; yield: 65%; R_f_ = 0.6 (DCM/MeOH = 10:1, v/v); ^1^H NMR (400 MHz, CDCl_3_) δ 7.53 (d, J = 15.4 Hz, 1H), 6.76 (d, J = 15.4 Hz, 2H), 6.71 (s, 2H), 3.87 (s, 5H), 3.84 (s, 3H), 3.67–3.54 (m, 4H), 3.11 (d, J = 0.9 Hz, 2H), 2.14 (s, 2H), 1.71–1.56 (m, 8H), 1.53–1.40 (m, 8H); ^13^C NMR (101 MHz, CDCl_3_) δ 178.05, 165.46, 153.50, 142.42, 139.52, 131.21, 117.10, 105.04, 61.09, 56.34, 53.74, 47.18, 43.54, 43.15, 39.62, 36.96, 26.92, 25.76, 25.71, 24.77, 22.99; HRMS (ESI+) calcd for C_26_H_38_N_2_O_5_ (m/z[M+H]^+^): 459.2859; found: 459.2833.

(E)-N-((1-(2-morpholino-2-oxoethyl)cyclohexyl)methyl)-3-(3,4,5-trimethoxyphenyl)acrylamide (5): yellow solid; yield: 60%; R_f_ = 0.6 (DCM/MeOH = 10:1, v/v); ^1^H NMR (400 MHz, CDCl_3_) δ 7.64–7.57 (m, 1H), 6.73 (d, J = 3.0 Hz, 2H), 6.69 (s, 1H), 3.89–3.85 (m, 9H), 3.75–3.70 (m, 5H), 3.46–2.81 (m, 1H), 2.21–1.65 (m, 2H), 1.54–1.02 (m, 10H); ^13^C NMR (101 MHz, CDCl_3_) δ 165.68, 153.57, 139.83, 130.84, 115.88, 105.17, 67.03, 61.13, 56.37, 51.22, 40.78, 39.70, 36.99, 29.86, 25.79, 23.01; HRMS (ESI+) calcd for C_25_H_36_N_2_O_6_ (m/z[M+H]^+^): 461.2652; found: 461.2641.

(E)-N-((1-(2-oxo-2-(pyrrolidin-1-yl)ethyl)cyclohexyl)methyl)-3-(3,4,5-trimethoxyphenyl)acrylamide (6): brown solid; yield: 41%; R_f_ = 0.5 (DCM/MeOH = 10:1, v/v); ^1^H NMR (600 MHz, CDCl_3_) δ 8.09–7.96 (m, 1H), 7.67 (d, J = 8.4 Hz, 1H), 6.74 (s, 2H), 6.60 (d, J = 15.4 Hz, 1H), 3.93–3.83 (m, 9H), 3.59 (t, J = 7.2 Hz, 3H), 3.45–3.29 (m, 3H), 2.05–1.99 (m, 5H), 1.91 (q, J = 6.9 Hz, 3H), 0.86 (dd, J = 18.3, 11.3 Hz, 11H); ^13^C NMR (151 MHz, CDCl_3_) δ 163.74, 153.58, 142.04, 130.32, 127.05, 125.24, 118.24, 105.30, 56.40, 46.30, 29.89, 26.35, 24.55, 22.88, 18.61; HRMS (ESI+) calcd for C_25_H_36_N_2_O_5_ (m/z[M+H]^+^): 445.2702; found: 445.2687.

(E)-N-((1-(2-(4-methylpiperazin-1-yl)-2-oxoethyl)cyclohexyl)methyl)-3-(3,4,5-trimethoxyphenyl)acrylamide (7): brown solid; yield: 48%; R_f_ = 0.5 (DCM/MeOH = 10:1, v/v); ^1^H NMR (600 MHz, CDCl_3_) δ 7.57 (d, J = 15.3 Hz, 1H), 6.74 (d, J = 15.8 Hz, 3H), 3.87 (d, J = 13.3 Hz, 9H), 3.80–3.60 (m, 9H), 2.44 (t, J = 5.0 Hz, 4H), 2.32 (s, 4H), 1.23 (t, J = 7.1 Hz, 8H); ^13^C NMR (151 MHz, CDCl_3_) δ 165.60, 153.61, 148.77, 143.10, 139.18, 131.04, 116.50, 105.22, 61.16, 58.61, 56.42, 46.22, 42.32, 24.84, 18.64; HRMS (ESI+) calcd for C_26_H_39_N_3_O_5_ (m/z[M+H]^+^): 474.2968; found: 474.2923.

3-methoxy-4-((3-methylbut-2-en-1-yl)oxy)benzyl (E)-2-(1-((3-(3,4,5-trimethoxyphenyl)acrylamido)methyl)cyclohexyl)acetate (8): white solid; yield: 46%; R_f_ = 0.7 (Hexane/EtOAc = 3:1, v/v); ^1^H NMR (600 MHz, CDCl_3_) δ 7.62 (d, J = 15.9 Hz, 1H), 6.96–6.92 (m, 2H), 6.86 (d, J = 8.0 Hz, 1H), 6.73 (s, 2H), 6.37 (d, J = 15.9 Hz, 1H), 5.50 (dt, J = 5.4, 2.5 Hz, 1H), 5.16 (s, 2H), 4.58 (d, J = 6.8 Hz, 2H), 3.88 (s, 4H), 3.86 (d, J = 1.4 Hz, 10H), 1.76 (s, 4H), 1.72 (s, 4H), 1.56 (s, 2H), 1.21–0.52 (m, 7H); ^13^C NMR (151 MHz, CDCl_3_) δ 166.99, 153.57, 149.64, 148.62, 145.20, 140.31, 137.88, 130.02, 128.58, 121.46, 120.00, 117.37, 112.31, 105.41, 66.75, 65.94, 61.10, 58.53, 56.29, 56.10, 32.08, 29.86, 25.97, 22.85, 18.56, 18.38, 14.27; HRMS (ESI+) calcd for C_34_H_45_NO_8_ (m/z[M+Na]^+^): 618.3043; found:618.3097.

(E)-3-(2-chlorophenyl)-N-((1-(2-oxo-2-(piperidin-1-yl)ethyl)cyclohexyl)methyl)acrylamide (11): brown solid; yield: 54%; R_f_ = 0.6 (DCM/MeOH = 10:1, v/v); ^1^H NMR (600 MHz, CDCl_3_) δ 7.90 (dd, J = 15.6, 3.6 Hz, 1H), 7.54 (dd, J = 6.2, 3.2 Hz, 1H), 7.36–7.34 (m, 2H), 7.25–7.20 (m, 2H), 6.82 (dd, J = 15.6, 3.7 Hz, 1H), 3.66 (s, 3H), 3.53 (s, 2H), 1.64 (d, J = 8.4 Hz, 2H), 1.19 (d, J = 6.9 Hz, 17H); ^13^C NMR (151 MHz, CDCl_3_) δ 165.27, 138.06, 134.73, 134.17, 130.31, 127.79, 127.07, 121.28, 47.37, 43.55, 43.53, 38.77, 37.32, 32.15, 30.25, 29.92, 29.58, 24.84, 22.88, 14.34; HRMS (ESI+) calcd for C_23_H_31_ClN_2_O_2_ (m/z[M+H]^+^): 403.2152; found: 403.2130.

(E)-3-(2-chlorophenyl)-N-((1-(2-morpholino-2-oxoethyl)cyclohexyl)methyl)acrylamide (12): yellow solid; yield: 50%; R_f_ = 0.6 (DCM/MeOH = 10:1, v/v); ^1^H NMR (600 MHz, CDCl_3_) δ 7.98 (dd, J = 15.5, 2.6 Hz, 1H), 7.55 (d, J = 7.1 Hz, 1H), 7.40–7.34 (m, 1H), 7.26–7.22 (m, 2H), 6.80 (d, J = 15.6 Hz, 1H), 5.26 (s, 1H), 3.69 (s, 9H), 2.86 (d, J = 5.0 Hz, 3H), 2.80–2.49 (m, 3H), 2.46 (s, 3H), 2.39–1.79 (m, 1H), 1.64–0.75 (m, 4H); ^13^C NMR (151 MHz, CDCl_3_) δ 165.26, 139.05, 133.54, 130.56, 127.67, 127.04, 119.78, 67.87, 67.07, 53.59, 52.05, 46.30, 25.95, 22.79, 21.80, 18.52; HRMS (ESI+) calcd for C_23_H_29_ClN_2_O_3_ (m/z[M+H]^+^): 405.1945; found: 405.1951.

(E)-3-(2-chlorophenyl)-N-((1-(2-oxo-2-(pyrrolidin-1-yl)ethyl)cyclohexyl)methyl)acrylamide (13): pale yellow solid; yield: 51%; R_f_ = 0.6 (DCM/MeOH = 10:1, v/v); ^1^H NMR (600 MHz, CDCl_3_) δ 7.98 (d, J = 15.7 Hz, 1H), 7.54 (s, 1H), 7.33 (s, 1H), 7.21–7.19 (m, 3H), 6.67 (d, J = 15.6 Hz, 1H), 3.54 (dt, J = 19.6, 6.9 Hz, 5H), 2.77 (d, J = 17.1 Hz, 1H), 1.94 (p, J = 6.8 Hz, 2H), 1.83 (p, J = 6.9 Hz, 2H), 1.79–1.64 (m, 1H), 1.56–0.36 (m, 3H); ^13^C NMR (151 MHz, CDCl_3_) δ 164.26, 137.62, 134.67, 133.76, 127.72, 126.94, 121.98, 46.70, 46.12, 31.94, 29.71, 26.18, 24.35, 23.83, 22.72; HRMS (ESI+) calcd for C_22_H_29_ClN_2_O_2_ (m/z[M+H]^+^): 389.1996; found: 389.1987.

(E)-3-(2-chlorophenyl)-N-((1-(2-morpholino-2-oxoethyl)cyclohexyl)methyl)acrylamide (14): pale yellow solid; yield: 63%; R_f_ = 0.6 (DCM/MeOH = 10:1, v/v); ^1^H NMR (600 MHz, CDCl_3_) δ 7.98 (d, J = 15.5 Hz, 1H), 7.58 (dd, J = 6.7, 2.7 Hz, 1H), 7.42–7.39 (m, 1H), 7.27 (d, J = 5.3 Hz, 2H), 6.85 (d, J = 15.5 Hz, 1H), 5.30 (s, 1H), 3.78–3.69 (m, 4H), 2.45 (t, J = 5.0 Hz, 4H), 2.33 (s, 3H), 2.10–1.35 (m, 3H), 1.31–0.76 (m, 10H); ^13^C NMR (151 MHz, CDCl_3_) δ 174.36, 165.33, 138.77, 134.78, 133.90, 130.50, 130.33, 127.81, 127.10, 120.52, 58.53, 55.44, 54.83, 46.17, 42.26, 29.89, 22.88, 18.63; HRMS (ESI+) calcd for C_23_H_32_ClN_3_O_2_ (m/z[M+H]^+^): 418.2261; found: 418.2273.

3-methoxy-4-((3-methylbut-2-en-1-yl)oxy)benzyl (E)-2-(1-((3-(2-chlorophenyl)acrylamido)methyl)cyclohexyl)acetate (15): white solid; yield: 48%; R_f_ = 0.7 (Hexane/EtOAc = 4:1, v/v); ^1^H NMR (600 MHz, CDCl_3_) δ 8.13 (d, J = 16.0 Hz, 1H), 7.60 (dd, J = 7.7, 1.8 Hz, 1H), 7.52 (dd, J = 6.5, 3.0 Hz, 1H), 7.38–7.37 (m, 1H), 6.97–6.92 (m, 2H), 6.88–6.85 (m, 2H), 6.50–6.44 (m, 1H), 5.53–5.49 (m, 1H), 5.19 (s, 2H), 4.58 (d, J = 7.3 Hz, 3H), 3.88 (s, 3H), 3.07 (d, J = 2.8 Hz, 1H), 2.63 (s, 1H), 1.76 (s, 5H), 1.72 (s, 4H), 1.07 (d, J = 8.9 Hz, 2H), 0.88–0.83 (m, 7H); ^13^C NMR (151 MHz, CDCl_3_) δ 188.30, 163.96, 144.81, 141.07, 133.87, 131.28, 130.38, 129.07, 128.24, 127.84, 127.28, 127.27, 121.40, 120.84, 120.17, 119.51, 113.05, 112.21, 110.90, 66.87, 58.64, 56.06, 37.31, 32.13, 29.91, 29.57, 26.02, 22.90, 18.62, 18.42, 14.32; HRMS (ESI+) calcd for C_31_H_38_ClNO_5_ (m/z[M+Na]^+^): 562.2336; found: 562.2329.

(R,E)-N-(4-methyl-2-(2-oxo-2-(piperidin-1-yl)ethyl)pentyl)-3-(3,4,5-trimethoxyphenyl)acrylamide (19): brown solid; yield: 63%; R_f_ = 0.4 (DCM/MeOH = 10:1, v/v); ^1^H NMR (600 MHz, CDCl_3_) δ 7.54 (d, J = 15.4 Hz, 1H), 7.47 (dd, J = 15.6, 2.8 Hz, 1H), 6.72 (s, 3H), 6.31 (dd, J = 15.6, 2.3 Hz, 1H), 3.88 (d, J = 2.8 Hz, 9H), 3.65 (t, J = 5.3 Hz, 3H), 3.60–3.55 (m, 3H), 3.43–3.35 (m, 3H), 2.50–2.40 (m, 1H), 2.27–2.18 (m, 2H), 1.60 (q, J = 5.7 Hz, 8H), 1.13 (dt, J = 13.4, 6.4 Hz, 1H), 0.93–0.87 (m, 6H); ^13^C NMR (151 MHz, CDCl_3_) δ 171.60, 165.39, 153.47, 153.45, 142.42, 140.25, 131.21, 120.86, 117.06, 104.90, 104.84, 61.12, 56.31, 56.26, 47.18, 44.68, 43.09, 38.05, 32.82, 26.66, 25.73, 25.41, 24.79, 24.55, 22.80; HRMS (ESI+) calcd for C_25_H_38_N_2_O_5_ (m/z[M+H]^+^): 447.2859; found: 447.2863.

(R,E)-N-(4-methyl-2-(2-morpholino-2-oxoethyl)pentyl)-3-(3,4,5-trimethoxyphenyl)acrylamide (20): brown solid; yield: 54%; R_f_ = 0.5 (DCM/MeOH = 10:1, v/v); ^1^H NMR (600 MHz, CDCl_3_) δ 7.60 (d, J = 15.3 Hz, 1H), 7.47 (d, J = 15.6 Hz, 1H), 6.72 (s, 2H), 6.28 (dd, J = 15.6, 1.6 Hz, 1H), 3.87 (s, 9H), 3.72 (s, 5H), 3.67 (d, J = 1.5 Hz, 3H), 2.41–2.23 (m, 3H), 1.66 (hept, J = 6.7 Hz, 1H), 1.29–1.20 (m, 2H), 1.18–1.12 (m, 1H), 0.90 (ddd, J = 12.4, 6.5, 1.8 Hz, 6H); ^13^C NMR (151 MHz, CDCl_3_) δ 171.95, 165.63, 153.53, 153.50, 143.50, 140.61, 130.81, 120.55, 115.81, 105.04, 104.91, 68.28, 67.18, 61.14, 56.33, 52.14, 37.45, 32.88, 25.45, 22.80; HRMS (ESI+) calcd for C_24_H_36_N_2_O_6_ (m/z[M+H]^+^): 449.2652; found: 449.2648.

(R,E)-N-(4-methyl-2-(2-oxo-2-(pyrrolidin-1-yl)ethyl)pentyl)-3-(3,4,5-trimethoxyphenyl)acrylamide (21): brown solid; yield: 59%; R_f_ = 0.5 (DCM/MeOH = 10:1, v/v); ^1^H NMR (600 MHz, CDCl_3_) δ 7.54 (dd, J = 15.4, 3.0 Hz, 1H), 6.70 (d, J = 3.2 Hz, 3H), 6.56 (dd, J = 15.3, 3.0 Hz, 1H), 3.82 (d, J = 15.7 Hz, 9H), 3.53 (t, J = 7.1 Hz, 3H), 3.31 (s, 2H), 3.20 (d, J = 14.2 Hz, 2H), 3.01 (s, 1H), 2.31–2.14 (m, 1H), 1.98 (d, J = 3.0 Hz, 4H), 1.86 (dt, J = 14.0, 7.9 Hz, 4H), 0.88–0.78 (m, 7H); ^13^C NMR (151 MHz, CDCl_3_) δ 171.34, 164.98, 142.14, 139.66, 130.89, 117.93, 105.23, 63.17, 56.26, 55.09, 46.81, 46.25, 29.74, 26.16, 24.39, 15.28, 14.22; HRMS (ESI+) calcd for C_24_H_36_N_2_O_5_ (m/z[M+H]^+^): 433.2702; found: 433.2711.

(R,E)-N-(4-methyl-2-(2-(4-methylpiperazin-1-yl)-2-oxoethyl)pentyl)-3-(3,4,5-trimethoxyphenyl)acrylamide (22): brown solid; yield: 51%; R_f_ = 0.5 (DCM/MeOH = 10:1, v/v); ^1^H NMR (600 MHz, CDCl_3_) δ 7.53 (d, J = 15.3 Hz, 1H), 7.43 (d, J = 15.6 Hz, 1H), 6.69 (s, 3H), 6.28 (d, J = 15.6 Hz, 1H), 3.84 (s, 6H), 3.81 (s, 4H), 3.66 (d, J = 41.2 Hz, 4H), 3.47–3.21 (m, 3H), 2.39 (s, 4H), 2.27 (s, 3H), 2.21 (s, 2H), 1.64 (dt, J = 13.7, 6.8 Hz, 1H), 1.16–1.03 (m, 1H), 0.94–0.72 (m, 6H); ^13^C NMR (151 MHz, CDCl_3_) δ 171.60, 165.39, 153.47, 153.45, 142.42, 140.25, 131.21, 120.86, 117.06, 104.90, 104.84, 61.12, 56.31, 56.26, 47.18, 44.68, 43.09, 38.05, 32.82, 26.66, 25.73, 25.41, 24.79, 24.55, 22.80; HRMS (ESI+) calcd for C_25_H_39_N_3_O_5_ (m/z[M+H]^+^): 462.2968; found: 462.2960.

3-methoxy-4-((3-methylbut-2-en-1-yl)oxy)benzyl (R,E)-5-methyl-3-((3-(3,4,5-trimethoxyphenyl)acrylamido)methyl)hexanoate (23): white solid; yield: 51%; R_f_ = 0.7 (Hexane/EtOAc = 4:1, v/v); ^1^H NMR (600 MHz, CDCl_3_) δ 7.61 (d, J = 15.9 Hz, 1H), 6.95–6.91 (m, 2H), 6.86 (d, J = 8.1 Hz, 1H), 6.73 (s, 2H), 6.37 (d, J = 15.8 Hz, 1H), 5.51–5.48 (m, 1H), 5.15 (s, 2H), 4.57 (d, J = 6.9 Hz, 2H), 3.87 (s, 3H), 3.85 (s, 9H), 2.98 (s, 1H), 1.88 (s, 8H), 1.76–1.75 (m, 4H), 1.71 (s, 4H), 0.92–0.82 (m, 6H). ^13^C NMR (151 MHz, CDCl_3_) δ 169.77, 167.04, 153.57, 149.63, 148.62, 145.23, 137.90, 130.02, 128.58, 121.47, 119.99, 117.36, 113.04, 112.33, 105.42, 66.77, 65.94, 61.10, 58.46, 56.29, 37.26, 32.09, 29.86, 29.52, 25.96, 22.85, 18.52, 14.27; HRMS (ESI+) calcd for C_33_H_45_NO_8_ (m/z[M+Na]^+^): 606.3043; found: 606.3043.

(R,E)-3-(2-chlorophenyl)-N-(4-methyl-2-(2-oxo-2-(piperidin-1-yl)ethyl)pentyl)acrylamide (26): white solid; yield: 50%; R_f_ = 0.5 (DCM/MeOH = 10:1, v/v); ^1^H NMR (600 MHz, CDCl_3_) δ 7.93 (d, J = 15.6 Hz, 1H), 7.57 (q, J = 4.7 Hz, 1H), 7.38 (d, J = 8.7 Hz, 1H), 7.25–7.22 (m, 3H), 6.86 (d, J = 15.5 Hz, 1H), 5.28 (s, 1H), 3.74–3.64 (m, 4H), 3.61–3.45 (m, 3H), 2.25 (t, J = 13.1 Hz, 1H), 1.69–1.66 (m, 2H), 1.50–1.39 (m, 3H), 1.23 (dt, J = 14.2, 7.1 Hz, 7H), 1.18 (t, J = 7.0 Hz, 2H), 0.93–0.81 (m, 7H); ^13^C NMR (151 MHz, CDCl_3_) δ 171.68, 165.73, 138.05, 136.23, 130.27, 127.76, 127.69, 127.05, 124.52, 121.22, 50.96, 45.01, 43.24, 38.13, 32.82, 26.24, 25.51, 24.41, 22.83, 18.61; HRMS (ESI+) calcd for C_21_H_31_ClN_2_O_2_ (m/z[M+H]^+^): 391.2125; found: 391.2123.

(R,E)-3-(2-chlorophenyl)-N-(4-methyl-2-(2-morpholino-2-oxoethyl)pentyl)acrylamide (27): brown solid; yield: 51%; R_f_ = 0.5 (DCM/MeOH = 10:1, v/v); ^1^H NMR (600 MHz, CDCl_3_) δ 8.00 (d, J = 15.5 Hz, 1H), 7.61–7.51 (m, 1H), 7.38 (t, J = 8.5 Hz, 1H), 7.28–7.22 (m, 2H), 6.81 (d, J = 15.5 Hz, 1H), 3.71 (s, 8H), 2.87 (d, J = 5.1 Hz, 5H), 2.39–2.21 (m, 2H), 1.35–1.11 (m, 3H), 0.89 (dd, J = 13.4, 6.5 Hz, 4H); ^13^C NMR (151 MHz, CDCl_3_) δ 171.96, 165.34, 139.15, 136.40, 133.64, 130.31, 127.75, 127.08, 124.20, 119.86, 77.23, 67.16, 52.13, 46.45, 42.92, 42.21, 37.50, 32.69, 25.45, 23.00, 22.76; HRMS (ESI+) calcd for C_21_H_29_ClN_2_O_3_ (m/z[M+H]^+^): 393.1945; found: 393.1938.

(R,E)-3-(2-chlorophenyl)-N-(4-methyl-2-(2-oxo-2-(pyrrolidin-1-yl)ethyl)pentyl)acrylamide (28): pale yellow solid; yield: 50%; R_f_ = 0.5 (DCM/MeOH = 10:1, v/v); ^1^H NMR (600 MHz, CDCl_3_) δ 8.06–7.95 (m, 1H), 7.56 (q, J = 4.1 Hz, 1H), 7.35 (d, J = 5.5 Hz, 1H), 7.23 (dt, J = 7.8, 3.2 Hz, 2H), 6.76–6.66 (m, 1H), 6.19 (d, J = 3.9 Hz, 0H), 3.54 (dd, J = 11.5, 5.5 Hz, 2H), 3.27 (s, 4H), 2.71 (d, J = 94.1 Hz, 1H), 2.37 (d, J = 6.0 Hz, 1H), 2.00–1.95 (m, 5H), 1.80–1.60 (m, 2H), 0.88–0.76 (m, 7H); ^13^C NMR (151 MHz, CDCl_3_) δ 174.27, 164.41, 137.82, 130.40, 130.22, 127.82, 127.02, 122.04, 52.26, 49.81, 46.83, 46.23, 45.01, 29.80, 26.27, 24.45, 24.28; HRMS (ESI+) calcd for C_21_H_29_ClN_2_O_2_ (m/z[M+H]^+^): 377.1996; found: 377.1981.

(R,E)-3-(2-chlorophenyl)-N-(4-methyl-2-(2-(4-methylpiperazin-1-yl)-2-oxoethyl)pentyl)acrylamide (29): pale yellow solid; yield: 50%; R_f_ = 0.5 (DCM/MeOH = 10:1, v/v); ^1^H NMR (600 MHz, CDCl_3_) δ 7.94 (d, J = 15.5 Hz, 1H), 7.58–7.51 (m, 1H), 7.36 (d, J = 7.5 Hz, 1H), 7.24–7.21 (m, 2H), 6.82 (d, J = 15.5 Hz, 1H), 6.38 (d, J = 15.7 Hz, 1H), 3.67 (d, J = 56.6 Hz, 5H), 3.46–3.10 (m, 2H), 2.57 (s, 5H), 2.40 (t, J = 5.0 Hz, 4H), 2.28 (s, 3H), 2.23 (d, J = 16.6 Hz, 1H), 1.79–1.59 (m, 1H), 0.92–0.77 (m, 5H); ^13^C NMR (151 MHz, CDCl_3_) δ 171.66, 165.19, 138.60, 134.65, 133.78, 130.42, 130.21, 127.71, 127.03, 120.45, 55.36, 54.71, 46.10, 42.19, 41.10, 37.71, 32.80, 29.79, 25.43, 22.98, 22.72; HRMS (ESI+) calcd for C_22_H_32_ClN_3_O_2_ (m/z[M+H]^+^): 406.2261; found: 406.2259.

3-methoxy-4-((3-methylbut-2-en-1-yl)oxy)benzyl (R,E)-3-((3-(2-chlorophenyl)acrylamido)methyl)-5-methylhexanoate (30): white solid; yield: 49%; R_f_ = 0.7 (Hexane/EtOAc = 4:1, v/v); ^1^H NMR (600 MHz, CDCl_3_) δ 8.10 (d, J = 15.9 Hz, 1H), 7.58 (dd, J = 7.7, 1.9 Hz, 1H), 7.38 (dd, J = 7.9, 1.5 Hz, 1H), 7.25 (dtd, J = 22.0, 7.6, 1.6 Hz, 2H), 6.93 (s, 2H), 6.85 (d, J = 8.5 Hz, 1H), 6.44 (d, J = 16.0 Hz, 1H), 5.50–5.48 (m, 1H), 5.17 (s, 2H), 4.56 (d, J = 6.8 Hz, 3H), 3.86 (s, 3H), 2.66 (s, 5H), 1.74 (s, 4H), 1.70 (s, 4H), 0.87–0.79 (m, 8H); ^13^C NMR (151 MHz, CDCl_3_) δ 177.29, 166.57, 149.55, 148.49, 141.00, 137.83, 135.04, 130.25, 127.19, 121.32, 120.66, 119.92, 113.00, 112.15, 66.79, 65.88, 58.18, 55.99, 32.02, 29.79, 25.88, 22.78, 18.35, 14.19; HRMS (ESI+) calcd for C_30_H_38_ClNO_5_ (m/z[M+Na]^+^): 550.2336; found: 550.2343.

2.2. Pharmacology

2.2.1. Maximal Electroshock (MES) Test

In the initial determination of the MES, all derivatives demonstrated protection in the tested mice after 0.5 h, with the exception of morpholine-substituted derivatives 5, 12, 20, and 27 (Table 1), suggesting their ability to inhibit the spread of epileptic seizures. Notably, analogs such as 11, 19, 22, and 23 were prominent at 100 mg/kg after 0.5 h in the MES assessment, indicating their favorable anticonvulsant properties. Among the synthetic analogs, only derivatives 22 and 23 were active at the same dosage after 3 h, which indicated that derivatives 22 and 23 demonstrated a rapid onset and prolonged action duration. Moreover, the protective effect of the analogs at lower doses was determined as well. The ED_50_ values of potential derivatives 11, 19, 22, and 23 were 85.6, 27.3, 53.4, and 42.2 mg/kg, respectively (Table 2). The anticonvulsant potency of the most promising derivatives 19 and 23 was close to carbamazepine (CBZ, ED_50_: 8.8 mg/kg, Table 2), used as a positive control. Compounds 19 and 23 exhibited higher protective index (PI) values than CBZ, indicating more potential in safety and effectiveness in general, while compounds 11 and 22 exerted lower PI values than CBZ. Compound 23 exerted the highest value of PI among the determined derivatives (>30), indicating significant anticonvulsant activity and safety of the derivatives.

2.2.2. Pentylenetetrazole (PTZ)-Induced Seizure Test

The anticonvulsant activity of promising derivatives 11, 19, 22, and 23 was determined for scPTZ-induced epileptic seizure in mice (Table 2). In the latent time test, at the dosage of 100 mg/kg, the mentioned test derivatives protected model animals from the scPTZ-induced epileptic seizure in latent time (p < 0.01), and the start of tonic epileptiform behavior of test mice was significantly postponed according to the observation of ethology [22]. Generally, derivatives with heterocyclic rings and pregabalin structural motifs were more potent in this model.

2.2.3. Acute Neurotoxicity Screening

Neurotoxicity screening revealed that heterocyclic analogs 11, 19, and 22 induced motor impairment (TD_50_: 300–400 mg/kg), as evidenced by frequent falls from the rotarod. In contrast, ester 23 showed negligible toxicity (TD_50_: 896.4 mg/kg; Table 2), highlighting its superior safety profile. The PI values of compounds 11 and 22 were lower than the reference standard carbamazepine, and the opposite phenomenon appeared on compounds 19 and 23. Compound 23 was considered to be the most promising candidate from the perspective of both security and effectiveness.

2.2.4. Validation of Anti-Seizure Potential Through Electroencephalogram (EEG)

Based on previous results, compounds 19 and 23 showed strong activity and higher PI values than positive controls in the MES model. Therefore, the mentioned two potential compounds have been selected for further EEG activity testing (Figure 2). The EEG results revealed that compounds 19 and 23 showed a coherent neurophysiological anticonvulsant profile: (1) early spike delay—compounds 19 and 23 extended the latency to the first PTZ-induced epileptic spike; (2) spike density reduction—compounds 19 and 23 significantly reduced spike frequency within 30 s post-initial spike compared to PTZ control. The dual efficacy of compounds 19 and 23 in suppressing both EEG spikes and behavioral convulsions positions them as a high-priority candidate for further development. Within the framework of preclinical epilepsy research, EEG recordings from mouse models have become an indispensable tool for evaluating potential ASMs [24], which provide critical insights into seizure generation, propagation, and suppression at both cellular and network levels. These recordings allow researchers to directly measure abnormal brain electrical activity characteristic of epileptic seizures, assess the impact of experimental compounds on these patterns, and identify potential biomarkers of treatment response [25].

2.2.5. Immunofluorescence Staining

The relationship between gamma-aminobutyric acid transaminase (GABA-AT) and gamma-aminobutyric acid type A receptor (GABA_AR_) represents a critical axis in the regulation of GABAergic neurotransmission and, consequently, neuronal excitability [26,27,28,29]. These two proteins function within an integrated signaling pathway that controls the availability and effectiveness of GABA in the synaptic cleft [30]. The relationship between GABA-AT and GABA_AR_ is characterized by a form of indirect coupling, where the activity of GABA-AT influences the availability of GABA for GABA_AR_. This relationship is distinct from direct protein–protein interactions but represents a functional connection within the broader context of GABAergic signaling. Dysregulation of this integrated pathway has been implicated in the pathophysiology of epilepsy. Enhancing GABAergic inhibition by targeting either GABA-AT or GABA_AR_ can provide effective seizure control [31]. Herein, we present the results of the immunofluorescence double-labeling for GABA-AT (modulated by ABAT) and GABA_AR_ (modulated by GABARA1) (Figure 3). Immunofluorescence revealed 1.8- and 1.6-fold elevations in ABAT and GABRA1 labeling in the hippocampal CA1 region of PTZ-treated mice versus controls (p < 0.01). Similar 1.6-fold increases in ABAT and GABRA1 were observed in the dentate gyrus (DG) region (p < 0.01) [32]. These findings underscore GABA-AT/GABA_AR_ dysregulation in PTZ-induced epilepsy. Immunolabelling intensity of ABAT and GABARA1 in both CBZ, compounds 19 and 23 groups was lower than the values found in the PTZ group (Figure 3), which can be measured in both CA1 and DG regions. Immunofluorescence results provided evidence for potential of 19 and 23 mediating GABA-AT and GABA_AR_ in PTZ-induced epilepsy mice, exerting therapeutic effects of epilepsy through GABA-AT/GABA_AR_ signaling.

2.2.6. Computational Prediction of Psychochemical Parameters

As detailed in Table 3, four kinds of active compounds were predicted for their drug-likeness potential. In general, the selected derivatives confirmed Lipinski’s rule; moreover, a certain extent of transmittance for BBB was predicted as well, which was essential for anticonvulsant activity in the brain. Results of CYP2C19 and P-gp inhibition prediction revealed that most of tested compounds could inhibit the activity of P-gp but none of them could inhibit the CYP2C19 activity.

2.2.7. Molecular Docking Analysis

GABA-AT (EC 2.6.1.19) catalyzes the transamination of GABA to α-ketoglutarate, yielding succinic semialdehyde and glutamate-a pivotal step in the GABA shunt linking GABA metabolism to the TCA cycle [33]. Recent cryo-EM studies further elucidate its active site conservation across species [34]. GABA_AR_s are pentameric ligand-gated ion channels that mediate fast inhibitory neurotransmission in the CNS. The three-dimensional structures of GABA-AT (PDB ID: 1OHW) [33] and GABA_AR_ (PDB ID: 6D6U) [35] have been elucidated through X-ray crystallography; herein, active derivatives were selected to evaluate the interactions between ligands and potential targets GABA-AT and GABA_AR_. According to the bioactivity determination results, the promising derivatives 11, 19, 22, and 23 were screened as ligands. As exhibited in Table 4, total scores and the binding model of compounds were evaluated. In the receptor–ligand complex, all the potential H-bonds were shorter than 3 Å (Figure 4), indicating stable interactions between TMCA peptides and potential targets GABA-AT and GABA_AR_. As for the docking score, the ranking can be summarized as 11 < CBZ < 22 < 23 < 19 for GABA-AT and 22 < 23 < 11< 19 < CBZ for GABA_AR_, which were similar to the trends in the previous determinations. The active site of GABA-AT contains several highly conserved residues, including LYS 329, ILE 50, GLU 270, ARG 192, VAL 241, and PHE 189, which participate in substrate binding and catalysis. The structural studies of GABA-AT have revealed a high degree of conservation across species, particularly within the active site region. This conservation reflects the fundamental importance of GABA metabolism in various organisms and has facilitated the development of inhibitors that target GABA-AT across different species. The complex pentameric structure of GABA_AR_ gives rise to multiple binding pockets located at various subunit interfaces and within individual subunits. These binding sites exhibit distinct pharmacological profiles and mediate the effects of a diverse array of compounds, including agonists, antagonists, and allosteric modulators. As for the flumazenil binding site, HIS 102, THR 142, PHE 100, TYR 160, PHE 77, SER 205, SER 206, and THR 207, TYR 210 were reported to be the active residues playing different roles in interactions. The mentioned active residues can be found in the interactions of TMCA peptides, indicating the reliability of the docking results. From the current study, we could see that the TMCA peptides can interact with GABA-AT and GABA_AR_ to some degree, providing insights for the design of innovative ASMs.

2.2.8. Molecular Dynamics (MD) Simulation

Based on comprehensive MD simulations and MM/GBSA analysis targeting GABA_AR_, CBZ demonstrated superior binding stability and energetic favorability compared to compounds 19 and 23 (Figure 5). Trajectory analysis revealed that CBZ maintained exceptional conformational stability within the receptor’s binding site, as evidenced by consistent ligand RMSD values throughout the simulation period. This structural fidelity substantially exceeded the stability profiles of compounds 19 and 23. CBZ preserved its optimal binding mode through extensive interactions with key residues, including PHE 189, LYS 329, ILE 72, VAL 300, THR 207, TYR 160, PHE 100, TYR 210, and TYR 58. The compound established a robust hydrophobic interaction network with aromatic and aliphatic residues while maintaining electrostatic contacts with charged amino acids. Compound 19 exhibited moderate binding affinity, forming hydrogen bonds with THR 353 and SER 205, complemented by hydrophobic interactions with PHE 351, TYR 348, GLU 270, PHE 77, ALA 79, PHE 100, HIS 102, TYR 160, and TYR 210. Compound 23 displayed the most extensive interaction profile, engaging ARG 192 and LYS 203 through hydrogen bonding while establishing contacts with multiple residues, including LYS 329, ARG 422, TYR 348, and others within the binding pocket. MM/GBSA free energy calculations (Table 5) quantified the thermodynamic basis of these binding preferences. Van der Waals forces dominated stabilization via shape complementarity, while electrostatics (mainly H-bonds) added further affinity, offset by polar solvation. These profiles align with functional assays, with CBZ exhibiting the strongest GABA_AR_ binding. The energetically favorable binding profiles correlated with the compounds’ functional performance in biochemical assays, with CBZ showing the highest binding affinity to GABA_AR_. These computational findings establish the structural basis for GABA_AR_ modulation and inform future structure-based optimization strategies targeting the identified key residues.

2.2.9. Structure–Activity Relationships (SARs)

In general, we acquired hopeful anticonvulsant derivatives after our chemical modification. The SAR study can be summarized as follows (Figure 6): (1) cinnamamide moiety: 3,4,5-trimethoxyl is better than 2-Cl; (2) amino acid moiety: pregabalin is better than gabapentin; (3) amine or alcohol moiety: activity: piperdine > isopentene vanillyl alcohol > piperazine > pyrrolidone > morpholine, while isopentene vanillyl alcohol performed better when the neurotoxicity and in vitro test were taken into consideration.

3. Discussion

In the present study, inspired by the TCM herb pair P. tenuifolia and G. elata, we designed and synthesized a series of TMCA peptide analogs (compounds 1–30). Comprehensive pharmacological evaluation, including the MES and scPTZ acute seizure models in mice, EEG recordings, double-labeling immunofluorescence of GABA-AT and GABAAR, molecular docking, molecular dynamics simulations, and SAR analysis, identified compounds 11, 19, 22, and 23 as the most promising derivatives in the initial screening.

Among them, compounds 19 and 23 exhibited particularly favorable anticonvulsant profiles: potent protection in the MES test, significant delay in seizure onset and reduction in spike frequency in the PTZ-induced EEG model, downregulation of overexpressed GABA-AT and GABA_AR_ in hippocampal CA1 and dentate gyrus regions of PTZ-treated mice, stable binding interactions with key residues of GABA-AT and GABA_AR_, and superior protective indices with low neurotoxicity compared to the reference drug CBZ.

These results align with and extend our previous investigations on TMCA-derived compounds, such as piperazine amide derivatives from P. tenuifolia and A. tatarinowii [5] and ester derivatives from P. tenuifolia and G. elata [6], which collectively underscore TMCA as a privileged scaffold for GABAergic-targeted anticonvulsants. Research on 3,4,5-trimethoxycinnamic acid (TMCA) has spanned over 20 years, initially demonstrating its anti-stress, sedative, and anxiolytic/antidepressant effects in the central nervous system through modulation of noradrenergic systems and prolongation of hexobarbital-induced sleep [36], and later expanding to anticonvulsant activity, hippocampal long-term potentiation enhancement via PKA and calcium-permeable AMPA receptor signaling [37]. The present study further deepens the exploration of TMCA’s neurological potential by shifting focus from anti-anxiety/depression profiles to robust anticonvulsant efficacy via GABAergic mechanisms. The incorporation of peptide motifs (especially pregabalin-derived) and isopentenyl vanillyl alcohol further improved potency, duration, and safety, offering novel structural insights for TCM-inspired drug design.

Future directions include advancing lead candidates (compounds 19 and 23) into chronic models (e.g., kindling or post-status epilepticus) to evaluate disease-modifying effects; performing in-depth mechanistic validation across multiple pathways; assessing ADME properties and long-term safety; and conducting further structural optimization to enhance potency, selectivity, and drug-likeness. Collectively, the novel TMCA peptide analogs, particularly 19 and 23, represent valuable preclinical leads for developing safer, more effective anticonvulsant agents derived from TCM herb pairs.

4. Materials and Methods

4.1. Chemistry

The reagents used for the reactions were purchased from several companies, including Macklin (Shanghai, China) and J&K China (Beijing, China). The conditions of the instruments for the characterization were reported from previous work [20,21].

Synthesis of TMCA Peptide Derivatives

As shown in Scheme 1. Firstly, pregabalin or gabapentin (20 mmol) was dissolved in excess methanol, and the sulfoxide chloride (10 mmol) was added dropwise into the solvent in an ice bath. The methanol was distilled after the complete reaction to obtain the methyl amino acid ester hydrochloride. Then, the methyl amino acid ester hydrochloride (6 mmol) was added with substituted cinnamic acid (8 mmol), hydroxybenzotriazole (HOBt, 4 mmol), and EDCI (8 mmol) in the presence of DCM (30 mL). The reactants were mixed thoroughly at room temperature for a period of 6 to 10 h until the reaction was complete, as demonstrated by TLC analysis. The resulting products were purified through silica gel chromatography with DCM and methanol (in a 20:1 v:v ratio) as the initial eluent to yield substituted cinnamamide methyl amino acid esters. Then, the products were dissolved in the methanol, and the 15% (v:v) NaOH solution was added to remove the methyl ester at a temperature ranging from 0 °C to 10 °C. The solution was concentrated after the complete hydrolysis, and the bare carboxyl derivatives were used to synthesize final products through the acylation reaction, once again similar to the preconditions mentioned above, heterocycle ammonia analogs or isopentenyl vanillyl alcohol (the preparation method can be found in ref. [20]) was added as the substances for the reaction. 1H NMR and 13C NMR spectra were obtained by a Varian Gemini 2000 DMX600 MHz FT NMR (Agilent Technologies, Inc., Santa Clara, CA, USA) using CDCl_3_ as a solvent and TMS as an internal standard. The chemical shifts were expressed in ppm. Mass spectral ESI measurements were executed on Agilent 6520 Accurate-mass Q-TOF LC/MS instruments (Agilent Technologies, Santa Clara, CA, USA).

4.2. Biological Activity

4.2.1. Animals and Experimental Conditions

Male KunMing-strain mice weighing 22–26 g were used for the in vivo tests in this research. Mice were kept under controlled environmental conditions (22 ± 2 °C; 50 ± 20% humidity; 12 h light/dark cycle) with free access to pellet food and water. The test solutions were immediately prepared before use (normal saline containing 0.5% Tween-80). Our experiment was authorized by the Institutional Animal Care and Use Committee of Shaanxi University of Chinese Medicine (animal ethics certificate number: SUCMDL20210309027). According to the ARRIVE guidelines, a total of 320 mice were used. After tail markings were made on the experimental animals, they were randomly divided into groups. All the named authors were aware of allocation, conduct, and outcome assessment for experimental animals.

4.2.2. MES Test

The MES test followed the method reported by Swinyard [38]. In the MES test, a 60 Hz current of 50 mA intensity (instrument: YSD-4G, SN: ZH0056232, Anhui, China) was applied through corneal electrodes for a 0.25 s duration. Protection against the spread of MES-induced seizures was defined as the absence of the tonic hind limb extension. The conditions of dosages and groups were reported from previous work [20]. Carbamazepine (CBZ, 53 mg/kg) was dissolved in an aqueous Tween-80 (1% v/v, 0.9% NaCl) solution as the positive control group. Negative control groups were composed of 1% Tween-80 solution. All the tested compounds were prepared as suspensions in aqueous Tween-80 (1% v/v, 0.9% NaCl), and intraperitoneally injected (i.p.) in a standard volume of 0.1 mL/10 g body weight. Protection against the spread of MES-induced seizures was defined as the absence of the tonic hind limb extension (hind limbs of animals outstretched 180° to the plane of the body). Mice were considered protected if they did not exhibit tonic hind. For the test groups, mice were, respectively, administered intraperitoneally with tested compounds at 100 mg/kg in preliminary evaluation and 75, 50, 25, 12.5 mg/kg in further evaluation to calculate ED_50_. Depending on the specific circumstances, additional subgroups of doses can be added. After administration of the compounds to all the mice for 0.5, 1, 2, 3, and 4 h, they received an electrical stimulation in the preliminary evaluation. Based on the previous research conducted by the research group [39], before the MES experiment, animals that did not respond to the electric shock stimulus would be excluded in advance. In this study, mice that did not show severe lethal reactions after receiving the drug protection and exhibited convulsive seizures at the preset voltage were selected. Before the rotarod test, animals that were unable to climb on the rotating rod under the non-treatment condition were excluded. As for the neurotoxicity test and related protection index calculation, it can be found in a previous study; pre-training was conducted before the experiment. Animals that could climb normally on the rotating rod apparatus were included in the study. Animals that had not been modeled but were unable to climb normally on the rotating rod would be excluded from the research.

4.2.3. PTZ-Induced Seizures

The experiments used 85 mg/kg PTZ, which produced clonic seizures lasting for a period of at least 5 s in 97% of animals tested [40], and PTZ (85 mg/kg) was dissolved in Tween-80 (1% v/v, 0.9% NaCl) solution for subcutaneous injections to mice 30 min after the treatment of compounds. Each animal was observed throughout the infusion period, and the duration of time between the start of infusion and the onset of the seizure was recorded [41]. The mice were considered protected when the compounds were in the absence of the effect of PTZ on seizure threshold. The selected derivatives were injected intraperitoneally into the test animals (100 mg/kg), and the number of protected animals and the latent time were observed.

4.2.4. Acute Neurotoxicity Screening

The acute neurotoxicity of the selected derivatives was determined in mice through the rotarod test [42]. The tested mice were trained to be placed on a 4 cm rod (24 rpm/min). The conditions of dosages and groups were reported from previous work [20].

4.2.5. Electroencephalogram (EEG)

Aseptic craniotomy was performed on urethane-sedated mice, entailing the stereotaxic placement of cortical screw electrodes in alignment with bregma coordinates (anterior–posterior: +3.0 mm; mediolateral: ±1.5 mm) alongside a reference electrode (anterior–posterior: −2.0 mm; mediolateral: −1.5 mm). Electrodes were fastened with dental acrylic, and then the animals had a recuperation phase of 5–7 days prior to the instrumentation. Continuous EEG recordings were obtained using 8-channel bioamplifiers connected to a PowerLab 8/30 system (AD Instruments, Dunedin, New Zealand). Following a 30 min acclimatization phase in the recording rooms, the initial levels of activity were recorded over a span of 15–20 min. The test compounds (19 and 23 100 mg/kg) were administered intraperitoneally. Following a duration of 60 min, pentylenetetrazol (PTZ, 85 mg/kg intraperitoneally) was implemented. initiating a seizure. The collection of EEG signals took place at a 200 Hz frequency, employing a bandpass filter that varied between 0.1 and 60 Hz [43]. Electrographic seizures were marked by spikes exceeding twice the baseline amplitude, defined by the following: (1) the interval preceding the initial spike (seconds post-PTZ); (2) spike frequency (counts 30 s from the initial spike) using recognized peak-detection techniques.

4.2.6. Immunofluorescence Staining

Brain samples were extracted, preserved in 4% paraformaldehyde for a day, and subsequently immersed in a 30% sucrose solution to dehydrate. Post OCT embedding, slices of brain tissue measuring 16 μm in thickness were sliced using a cryostat. Following the permeabilization using 0.5% TritonX-100 and subsequent blocking with 10% goat serum at ambient temperature to remove nonspecific staining, the tissue samples were subjected to an overnight incubation at 4 °C with the primary antibody, GABRA1 (Proteintech, Rosemont, IL, USA, Cat No. 12410-1-AP, Lot: 00026377, rabbit anti-rat 1:200), ABAT (Proteintech, Cat No. 11349-1-AP, Lot: 00070356, rabbit anti-rat 1:200). On the next day, the slides underwent incubation using a secondary antibody. The slides underwent counterstaining using DAPI. Ovarian slices were examined and visualized utilizing the VS200-BU Slide Scanner (Olympus Corporation, Tokyo, Japan).

4.2.7. Statistical Analysis

All data were described as mean ± standard deviation. The results were analyzed by a one-way ANOVA. The level of significance for all tests was set at p < 0.05. The SPSS 19.0 software was used as a statistical analysis tool.

4.2.8. Molecular Modeling

The crystal structure of GABA-AT (PDB ID: 1OHW) [33] and GABA_AR_ (PDB ID: 6D6U) [35] were obtained from the PDB database. Docking studies were processed on reference standard CBZ and potential derivatives 11, 19, 22, and 23 against 1OHW and 6D6U. The ligand–protein interactions obtained by the Amdock tool 4.2.6 [44]. To perform the docking studies, input site sphere dimension values were adjusted as X = 10.09, Y = −0.81, Z = −21.22 (1OHW), and X = 120.16, Y = 167.82, and Z = 153.59 (6D6U). The binding pocket was set based on the coordinates of the original ligands with a radius of 15 Å. The docked complexes were graded with binding affinity (Kcal/mol) value to assess the degree of matching. The hydrogen and hydrophobic bond interaction pattern visualization was performed by Discovery Studio Visualizer v. 2025.

4.2.9. Molecular Dynamics (MD) Simulation

The best binding conformation of compounds CBZ, 19, and 23 among the poses given by the molecular docking program was selected, and the protein-NAM complexes were used as MD simulation starting points in GROMACS 2022.3 [45]. The ligand–target complex was solvated in a similar volume of tip4p water box, energy minimized, equilibrated in NVT and NPT ensembles, respectively, and finally undertaken for 100 ns of production run in NPT ensemble at 300 K with the AMBER force field as the restricted condition [45]. The visualization was carried out with the VMD program. The MM-PBSA methods in GROMACS were used to calculate the binding free energies and energy decomposition.