Structural evolution of quinine derivatives for enhanced antifungal potency: key modifications and quantitative structure–activity relationship (QSAR) analysis

Abstract

BACKGROUND

The escalating resistance of phytopathogenic fungi to conventional fungicides, combined with increasing environmental concerns, necessitates the development of novel bioactive agents.

RESULTS

In this study, utilizing the natural product quinine as a lead scaffold, a series of novel 2‐aminoquinine derivatives were designed and synthesized via PyBroP‐mediated amination. The antifungal activities of these compounds against seven plant pathogens, including Rhizoctonia solani (R. solani), Cytospora chrysosperma (C. chrysosperma), Fusarium oxysporum (F. oxysporum), Magnaporthe oryzae (M. oryzae), Fusarium graminearum (F. graminearum), Sphaeropsis sapinea (S. sapinea), and Botrytis cinerea (B. cinerea), were evaluated using the mycelial growth rate method. Among the synthesized derivatives, compound Qed‐3 exhibited broad‐spectrum efficacy, particularly against S. sapinea with a half‐maximal effective concentration (EC₅₀) value of 0.46 μg mL⁻¹. In vivo assays further confirmed its superior curative potential against pine shoot blight compared to the commercial fungicide boscalid. Mechanistic investigations, utilizing scanning electron microscopy (SEM) and transmission electron microscopy (TEM), revealed that compound Qed‐3 disrupts the integrity of hyphal cell walls and membranes, leading to electrolyte leakage and the accumulation of reactive oxygen species (ROS).

CONCLUSIONS

Molecular docking and quantitative structure–activity relationship (QSAR) analyses provided insights into the binding modes and structural determinants essential for antifungal potency. These findings highlight the potential of 2‐arylaminoquinine as a promising and eco‐friendly candidate for novel fungicide development. © 2026 Society of Chemical Industry.

Novel aminoquinine derivatives were rationally designed via structural evolution. Lead compound Qed‐3 exhibits high antifungal potency, supported by quantitative structure–activity relationship (QSAR) modeling and molecular docking, through disrupting cell wall integrity and reactive oxygen species (ROS)‐mediated apoptosis.

1. INTRODUCTION

Plant fungal diseases have emerged as a persistent and escalating threat to global food security, forestry resources, and ecological stability. ¹ , ² , ³ For instance, in agriculture, pathogenic fungi exemplified by Fusarium graminearum ⁴ and Rhizoctonia solani ⁵ significantly impair crop productivity and agricultural economic growth. Meanwhile, in forestry, destructive species such as Sphaeropsis sapinea ⁶ and Cytospora chrysosperma ⁷ directly compromise the health and sustainability of shelterbelt ecosystems in northern China. The strategic application of fungicides is essential for mitigating fungal disease outbreaks and ensuring the stable supply of vital agricultural and forestry commodities. ⁸ Over recent decades, the widespread misuse of various fungicides has dramatically hastened the development and worldwide spread of resistance mechanisms in microbial pathogens. The prolonged dependence on conventional fungicides has intensified pathogen resistance, as evidenced by documented resistance cases to quinone outside inhibitor (QoI), Methyl Benzimidazole Carbamates (MBC), demethylation inhibitor (DMI), succinate dehydrogenase inhibitor (SDHI), and other key fungicide categories. ⁹ The development of novel eco‐friendly fungicides featuring innovative action targets, enhanced efficacy, negligible toxicity, and excellent environmental compatibility constitutes a promising strategy to counteract pathogen resistance. ¹⁰

Cytochrome P450 enzymes (CYP), a class of heme‐bound monooxygenases, perform critical functions in sterol/steroid metabolism, xenobiotic detoxification/drug activation, and diverse secondary metabolite production. ¹¹ Among these, sterol 14α‐demethylase (CYP51) serves as a phylogenetically conserved enzyme present across all biological kingdoms, catalyzing the removal of the 14α‐methyl group from sterol precursor molecules through an evolutionarily conserved three‐step monooxygenation reaction. ¹² The functional impairment of this enzyme results in the abnormal accumulation of sterol precursors and severe depletion of ergosterol within fungal cell membranes. These biochemical alterations induce significant morphological changes in eukaryotic cellular membranes, ultimately causing membrane permeability dysfunction that disrupts microbial proliferation and induces fungal cell death. ¹³ Therefore, the CYP51 enzyme emerges as one of the pivotal targets for developing antifungal agents against phytopathogenic fungi. ¹⁴ , ¹⁵ , ¹⁶ Fungicides possessing the imidazole and triazole moiety are the most representative class of CYP51 inhibitors. However, excessive reliance on these structurally similar azole compounds has been directly linked to escalating drug resistance profiles and adverse toxicological effects. ⁹ Recent studies over the past 3 years have highlighted the rapid emergence of DMI resistance in various agricultural pathogenic fungi. This resistance is predominantly driven by target‐site modifications, such as specific amino acid substitutions in the CYP51 enzyme and the functional divergence of paralogous CYP51 genes. ¹⁷ , ¹⁸ , ¹⁹ Consequently, the design of innovative CYP51 inhibitors featuring optimized molecular architectures represents an urgent priority for managing recalcitrant fungal infections. Naturally derived bioactive compounds in plants (phytochemicals) provide a sustainable alternative to conventional synthetic fungicides in modern agriculture and forestry. ²⁰ , ²¹ , ²² Structural modification of these natural scaffolds continues to serve as a highly effective approach for developing innovative antifungal agents. ²³ Quinine, a prototypical cinchona alkaloid, is a naturally occurring bioactive compound derived from the bark of cinchona trees indigenous to South American rainforests. ²⁴ This plant‐derived alkaloid has served as the primary chemoprophylactic agent against malaria for nearly four centuries. ²⁵ , ²⁶ Moreover, extensive literature review reveals that quinine and its derivatives exhibit remarkable multifunctional properties, including significant antibacterial, antiproliferative, antiviral, antiepileptic, and antifeedant activities. ²⁷ , ²⁸ , ²⁹ Despite these well‐documented pharmacological achievements, research on the application of quinine against agricultural fungal pathogens remains underdeveloped. ³⁰ In practice, quinine exhibits limited antifungal efficacy against phytopathogenic fungi, while the development of agrochemicals utilizing the quinoline scaffold structure has led to great progress. ³¹ , ³² , ³³ , ³⁴ Several commercially available quinoline‐derived fungicides including Quinofumelin, Quinoxyfen, and Ipflufenoquin successfully replace ecotoxic traditional pesticides (Fig. 1). ³⁴ , ³⁵

Figure 1.

Representative commercial fungicides containing quinoline scaffolds.

However, extensive structural analyses have confirmed that among the 177 resolved crystal structures of human cytochrome P450 families 1, 2, and 3, merely two structures demonstrate complexes with antimalarial agents: CYP2D6‐quinine (PDB: 4WNV) and CYP2D6‐quinidine (PDB: 4WNU). ³⁶ These antimalarial drugs function as CYP2D6 inhibitors and are not naturally metabolized by this enzyme. ³⁷ Quinidine represents the most potent documented inhibitor of CYP2D6 and is routinely employed as a reference inhibitor for drug interaction studies. ³⁸ Quinine, being the stereoisomeric counterpart of quinidine, exhibits approximately 100‐fold reduced inhibitory potency against CYP2D6, highlighting the critical influence of stereochemical configuration on enzyme binding and inhibition. ³⁷ Although quinine and quinidine demonstrate limited antifungal activity against plant pathogenic fungi, they exhibit potent inhibitory effects on human cytochrome P450 2D6 (CYP2D6). ³⁶ , ³⁷ , ³⁸ Given the large evolutionary distance between human CYP2D6 and fungal CYP51s (Fig. 2), we explicitly frame our drug design strategy not as a direct extrapolation of bioactivity, but as a structure‐guided hypothesis. We hypothesize that the quinoline scaffold possesses an intrinsic, ‘privileged’ affinity for the highly conserved P450 catalytic fold. Consequently, we postulate that through rational structural optimization, it is possible to actively shift the target selectivity of these economical natural products‐diminishing their affinity for human CYP2D6 while successfully tailoring them to specifically fit the analogous active site of fungal CYP51. The present study was designed to systematically test this hypothesis to develop botanical fungicides. (Fig. 3).

Figure 2.

Maximum‐likelihood phylogenetic tree of the CYP51 family and human P450 enzymes.

Figure 3.

Molecular design and derivation of the target compounds.

Thus, a series of quinine derivatives were designed and synthesized via PyBroP‐mediated amination (Fig. 4). Subsequently, in vitro activity assessments of the target compounds were conducted against various agricultural and forestry fungal pathogens, with quantitative structure–activity relationships (QSARs) systematically established. ³⁹ , ⁴⁰ The compound Qed‐3 demonstrated exceptional in vitro antifungal efficacy, prompting in vivo efficacy trials against S. sapinea and preliminary mechanistic investigations. Molecular docking analyses were conducted between compound Qed‐3 and potential target proteins CYP51, thereby providing a theoretical basis for further elucidating the mechanism of action.

Figure 4.

Synthetic route of target quinine derivatives. Reagents and conditions: (i) m‐CPBA (2.5 equiv.), CHCl₃, 0 °C to room temperature (r.t.); (ii) NaHSO₃ (2.5 equiv.), HCl (2.0 equiv.), acetone, r.t.; (iii) PyBroP (1.3 equiv.), iPr₂EtN (3.75 equiv.), DCM, HNR¹R² (1.26 equiv.), r.t.

2. MATERIALS AND METHODS

2.1. Instruments and reagents

Proton (¹H) and carbon‐13 (¹³C) nuclear magnetic resonance (NMR) spectra were acquired on a Bruker spectrometer (Bruker, Karlsruhe, Germany) at frequencies of 500 and 126 MHz, respectively, with tetramethylsilane (TMS) as the internal standard (solvents deuterated chloroform (CDCl₃) or deuterated dimethyl sulfoxide (DMSO‐d ₆)). High‐resolution mass spectrometry (HRMS) spectra were recorded on a Bruker BIO TOF Q instrument. Microscopic morphology of the mycelia was examined with a JSM‐7500F field‐emission scanning electron microscope and an HT‐7800 transmission electron microscope. Confocal micrographs were acquired with a Zeiss LSM 900 laser scanning confocal microscope (Zeiss, Oberkochen, Germany).

All reactions were conducted using commercially available reagents without further purification. All amines were purchased from commercial sources and used directly without special treatment. Azoxystrobin, thifluzamide, and boscalid were employed as the positive controls in the in vitro antifungal assays against the tested phytopathogens.

2.2. Biomaterials

The phytopathogenic fungi employed in this investigation comprised R. solani, C. chrysosperma, Fusarium oxysporum, Magnaporthe oryzae, F. graminearum, S. sapinea, and Botrytis cinerea. All fungal isolates were supplied by the College of Forestry at Northeast Forestry University, China, with the exception of F. oxysporum and F. graminearum strains, which were commercially acquired from the Shanghai Microorganism Preservation Center, China.

2.3. Phylogenetic analysis

The amino acid sequences of 11 representative cytochrome P450 enzymes, including human CYP2D6, CYP3A4, human and plant CYP51s, and CYP51s from pathogenic fungi, were retrieved from the UniProt database. The evolutionary history was inferred by using the maximum likelihood (ML) method and the Jones–Taylor–Thornton (JTT) matrix‐based model. ⁴¹ Initial tree(s) for the heuristic search were obtained automatically by applying Neighbor‐Join and BioNJ algorithms to a matrix of pairwise distances estimated using the JTT model, and then selecting the topology with superior log likelihood value. The tree with the highest log‐likelihood (−12 398.72) is presented. The reliability of the internal branches was evaluated using the bootstrap method with 1000 replicates, and the percentage of trees in which the associated taxa clustered together is shown next to the branches. The final dataset involved 11 amino acid sequences with a total of 608 positions. All evolutionary analyses were performed using MEGA11 software. ⁴²

2.4. Synthetic procedures for the target quinine derivatives

2.4.1. Synthesis of intermediates 1 and 2

Quinine (3.26 g, 10.0 mmol, 1.0 equiv.) was dissolved in chloroform (45 mL). Then, meta‐chloroperoxybenzoic acid (m‐CPBA, 85 wt%, 5.08 g, 25.0 mmol, 2.5 equiv.) was added at 0 °C, and the mixture was stirred for 30 min. Subsequently, the reaction was allowed to warm to room temperature and stirred for another 3 h. The reaction progress was monitored by thin‐layer chromatography (TLC). Upon completion, the mixture was adjusted to pH 10 with a 10% sodium hydroxide (NaOH) solution. The aqueous phase was extracted with a mixed solvent of chloroform/methanol (CHCl₃/MeOH, 15:1, v/v, six times). The combined organic layers were washed with brine, dried over anhydrous sodium sulfate (Na₂SO₄), filtered, and concentrated under reduced pressure to afford the crude intermediate 1 as an orange‐red viscous liquid. ⁴³

The crude intermediate 1 was used directly in the next step. Intermediate 1 was dissolved in acetone (150 mL) to form a clear pale‐yellow solution, which was then cooled in an ice bath. A solution of sodium bisulfite (NaHSO₃, 2.60 g, 25.0 mmol) in 1 m hydrochloric acid (HCl, 20 mL) was added dropwise via a constant‐pressure addition funnel. The mixture was allowed to warm to room temperature and stirred overnight. Upon completion (monitored by TLC), the pH was adjusted to pH 10 using aqueous ammonia. The acetone was removed under reduced pressure, and the aqueous residue was extracted with CHCl₃/MeOH (15:1, v/v, three times). The combined organic layers were washed with brine, dried over anhydrous Na₂SO₄, filtered, and concentrated in vacuo. The crude residue was purified by silica gel column chromatography (dichloromethane/methanol/triethylamine (DCM/MeOH /Et₃N), 25:1:0.01, v/v/v) to afford intermediate 2 as a yellow solid (3.40 g, quantitative yield).

2.4.2. Synthesis of the target quinine derivatives

Intermediate 2 (1.0 mmol), the corresponding amine (1.26 mmol), N,N‐diisopropylethylamine (iPr₂EtN, 3.75 mmol), and PyBroP (1.3 mmol) were dissolved in DCM (3 mL). ⁴⁴ The mixture was stirred at room temperature for 24 h. Upon completion of the reaction (monitored by TLC), saturated aqueous sodium bicarbonate (NaHCO₃) solution (15 mL) was added. The mixture was extracted with CHCl₃/MeOH (15:1, v/v) three times. The combined organic phases were washed with brine, dried over anhydrous Na₂SO₄, filtered, and concentrated under reduced pressure. The crude residue was purified by silica gel column chromatography (eluent: DCM/MeOH/Et₃N, 25:1:0.01, v/v/v) to afford the target compounds Qa‐1–Qed‐9 as white, yellow, or brown solids.

2.5. Antifungal activity assay in vitro

The antifungal activities of quinine and the synthesized quinine derivatives (compounds Qa‐1–Qed‐9) were initially assessed against seven phytopathogenic fungi at 50 μg mL⁻¹ using the mycelial growth inhibition assay. ⁴⁵ The commercial fungicides azoxystrobin, thifluzamide, and boscalid were employed as positive controls. Test compounds were dissolved in dimethyl sulfoxide (DMSO) and incorporated into potato dextrose agar (PDA) medium to achieve the desired concentration. Mycelial plugs (7 mm diameter) were inoculated and incubated in the dark at 28 °C, after which the inhibition rates were calculated. For lead compounds exhibiting high antifungal activity, the half‐maximal effective concentration (EC₅₀) values were determined using the probit regression method. ⁴⁵ All bioassays were performed with three independent replicates, with experimental data presented as mean ± standard deviation. Detailed data are provided in the Supporting Information Data S1.

2.6. In vivo antifungal activity test

Based on in vitro antifungal screening data, compound Qed‐3 was selected for its superior activity to evaluate the efficacy against S. sapinea infections on Pinus sylvestris needles with the commercial fungicide boscalid as the positive control. Concentration‐dependent in vivo protective efficacy was investigated through standardized foliar inoculation bioassays. ⁴⁶ All trials were performed with three independent replicates, with detailed protocols available in the Supporting Information Data S2.

2.7. Morphological characterization via scanning electron microscopy and transmission electron microscopy

Scanning electron microscopy (SEM) ⁴⁷ and transmission electron microscopy (TEM) ⁴⁸ were employed as complementary techniques to investigate the effects of compound Qed‐3 (0.5 μg mL⁻¹) on S. sapinea cellular architecture. These methods allowed for the visualization of hyphal alterations in terms of surface morphology and internal ultrastructure, respectively. Detailed experimental procedures are provided in the Supporting Information Data S3 and Data S4.

2.8. Cell membrane integrity

The S. sapinea mycelial samples treated with compound Qed‐3 were subjected to propidium iodide (PI) staining according to the previously established protocol. ⁴⁹ A laser scanning confocal microscope (Zeiss LSM900; Zeiss) was utilized to detect membrane‐compromised hyphae, which manifested distinct red fluorescence emission. Detailed experimental procedures are provided in the Supporting Information Data S5.

2.9. Reactive oxygen species detection

The S. sapinea mycelia treated with compound Qed‐3 were stained with 2′,7′‐dichlorodihydrofluorescein diacetate (DCFH‐DA; Biosharp, Hefei, China) following a reported method. ⁴⁵ Reactive oxygen species (ROS) accumulation was quantitatively assessed by monitoring the intracellular green fluorescence intensity using a laser scanning confocal microscope (Zeiss LSM900; Zeiss). Detailed experimental procedures are provided in the Supporting Information Data S6.

2.10. Quantitative structure–activity relationship modeling

The experimental dataset comprised 32 structurally defined quinine derivatives, with their antifungal potencies against S. sapinea quantitatively determined through standardized mycelial growth inhibition assays. The derived logarithmic half‐maximal effective concentration (log(EC₅₀)) values were employed as the response variable in subsequent QSAR modeling. GaussView 6.0, ⁵⁰ PaDEL‐Descriptors 2.21, ⁵¹ and QSARINS ⁵² software were employed sequentially for structure optimization, molecular descriptor extraction, and model construction, respectively. ⁴⁵ Detailed computational procedures are provided in the Supporting Information Data S7.

2.11. Molecular docking

To investigate potential protein targets of compound Qed‐3, we retrieved structural data of F. graminearum CYP51 (UniProt ID: I1RBR4) from https://www.uniprot.org/. AutoDock Tools was utilized for the pre‐processing of ligands and proteins, which included hydrogen addition, charge calculation, and file format conversion. AutoDock Vina ⁵³ was subsequently implemented to conduct molecular docking simulations, applying an iterated local search global optimization algorithm to determine the most favorable binding pose. The docking results were visualized and analyzed using PyMOL ⁵⁴ and Discovery Studio 2025 Client. ⁵⁵

3. RESULTS AND DISCUSSION

3.1. Chemical synthesis

As illustrated in Fig. 4 and Fig. S1 (Supporting Information), the synthetic pathway initiated with the m‐CPBA‐mediated oxidation of quinine to yield the bis‐N‐oxide intermediate 1. ⁴³ Subsequent chemoselective reduction with NaHSO₃ under controlled acidic conditions generated the pivotal mono‐N‐oxide intermediate 2. PyBroP‐activated coupling reactions with iPr₂EtN as the base then enabled direct C2‐position nucleophilic substitution on the quinoline scaffold using structurally diverse amines, including aliphatic amines, unsaturated aliphatic amines, alicyclic amines, and aromatic amines. ⁴⁴ This versatile approach established a robust and mild synthetic protocol, successfully affording 32 novel amino‐functionalized quinine derivatives (Qa‐1–Qed‐9), with all compounds rigorously characterized by ¹H NMR, ¹³C NMR, and HRMS. Detailed physicochemical properties and spectroscopic data are provided in the Supporting Information Data S8 and Data S9.

3.2. In vitro antifungal activity for the target compounds

The in vitro antifungal efficacy of all synthesized quinine derivatives (compounds Qa‐1–Qed‐9) was systematically assessed at a standardized concentration of 50 μg mL⁻¹ against seven phytopathogenic fungal strains, with thifluzamide, azoxystrobin, and boscalid serving as reference positive controls. ⁴⁵ As summarized in Table 1, the majority of compounds displayed broad‐spectrum antifungal properties, wherein the halogenated aromatic amine‐substituted quinine derivatives (Qed series) exhibited particularly pronounced activity against S. sapinea and C. chrysosperma, demonstrating substantially enhanced potency compared to their aliphatic and alicyclic amine‐substituted analogs. The Qed series achieved complete mycelial growth inhibition (100%) against S. sapinea (Qed‐1, Qed‐2, Qed‐4, Qed‐8) and C. chrysosperma (Qed‐1, Qed‐3), while concurrently maintaining robust inhibitory effects against the agriculturally significant F. graminearum (Qed‐1: 92.14%; Qed‐5: 89.57%; Qed‐3: 83.97%), with all values significantly exceeding the activity of unmodified quinine. Conversely, R. solani and M. oryzae exhibited consistently low susceptibility to all tested compounds, whereas F. oxysporum and B. cinerea demonstrated variable response patterns. Based on the consistently high sensitivity of S. sapinea, C. chrysosperma, and F. graminearum to the Qed series, these three target pathogens were selected for subsequent EC₅₀ determinations.

Table 1.

In vitro antifungal activities of quinine derivatives at 50 μg mL⁻¹

Compound	Inhibitory rate ± standard deviation (%)†
	R. s	C. c	F. o	S. s	M. o	F. g	B. c
Quinine	22.20 ± 2.64	53.8 ± 0.65	22.10 ± 2.20	70.46 ± 0.34	0.00 ± 0.00	27.07 ± 1.73	19.68 ± 4.89
Qa‐1	21.90 ± 1.36	66.2 ± 1.43	29.80 ± 0.75	64.06 ± 1.85	0.00 ± 0.00	18.67 ± 0.52	0.00 ± 0.00
Qa‐2	28.30 ± 1.33	69.91 ± 2.45	0.00 ± 0.00	84.44 ± 0.29	33.76 ± 1.85	43.91 ± 0.54	0.00 ± 0.00
Qa‐3	19.23 ± 1.46	71.15 ± 2.53	13.55 ± 2.47	88.86 ± 0.80	22.96 ± 1.20	80.79 ± 2.06	25.60 ± 0.75
Qb‐1	30.78 ± 1.27	88.46 ± 1.27	14.64 ± 2.04	88.11 ± 1.94	0.00 ± 0.00	71.38 ± 0.57	40.58 ± 2.51
Qc‐1	31.82 ± 1.53	91.12 ± 0.27	28.22 ± 0.91	90.31 ± 1.24	46.84 ± 1.26	71.04 ± 1.32	0.00 ± 0.00
Qc‐2	27.40 ± 0.49	86.63 ± 1.64	40.62 ± 1.72	96.11 ± 3.16	49.24 ± 2.28	78.86 ± 0.68	47.51 ± 1.12
Qc‐3	34.30 ± 0.62	92.16 ± 1.38	19.79 ± 0.86	85.40 ± 1.64	3.18 ± 1.18	87.06 ± 0.53	42.50 ± 0.39
Qd‐1	31.36 ± 1.12	93.28 ± 0.35	22.13 ± 1.35	95.14 ± 1.18	24.64 ± 1.27	86.96 ± 1.13	48.50 ± 0.58
Qd‐2	46.08 ± 2.43	96.93 ± 1.52	51.35 ± 2.09	100.00 ± 0.00	50.56 ± 2.95	89.54 ± 0.79	70.34 ± 1.25
Qd‐3	19.80 ± 2.38	67.9 ± 1.33	17.78 ± 2.65	85.32 ± 2.13	0.00 ± 0.00	63.75 ± 1.71	0.00 ± 0.00
Qea‐1	59.60 ± 0.43	97.02 ± 0.29	21.01 ± 0.47	95.21 ± 2.44	43.27 ± 0.96	90.42 ± 0.79	64.63 ± 1.52
Qea‐2	56.84 ± 2.38	93.64 ± 1.68	32.75 ± 1.57	94.90 ± 1.41	42.77 ± 3.21	88.14 ± 1.82	73.80 ± 1.51
Qea‐3	57.14 ± 0.27	81.67 ± 1.48	45.81 ± 0.82	82.84 ± 1.09	52.54 ± 1.17	70.13 ± 2.61	77.29 ± 1.54
Qeb‐1	53.23 ± 2.38	95.37 ± 0.38	31.06 ± 1.93	94.85 ± 0.68	43.59 ± 0.73	76.92 ± 1.13	41.63 ± 1.45
Qeb‐2	41.63 ± 0.76	93.28 ± 0.75	55.15 ± 2.04	88.33 ± 1.08	48.26 ± 2.09	69.58 ± 1.02	45.03 ± 1.50
Qec‐1	31.82 ± 2.45	94.58 ± 0.83	19.58 ± 0.09	98.24 ± 2.49	32.32 ± 1.95	84.11 ± 0.96	49.11 ± 0.89
Qec‐2	16.24 ± 1.48	93.33 ± 0.75	35.08 ± 2.15	100.00 ± 0.00	0.00 ± 0.00	93.53 ± 0.19	68.43 ± 3.03
Qec‐3	37.80 ± 2.73	95.74 ± 0.78	17.80 ± 2.56	87.87 ± 2.55	23.14 ± 1.42	78.4 ± 1.53	47.97 ± 0.77
Qec‐4	29.77 ± 0.38	91.48 ± 0.12	23.39 ± 1.67	90.31 ± 1.43	50.00 ± 1.91	85.77 ± 1.70	46.33 ± 4.82
Qec‐5	46.52 ± 2.48	92.11 ± 1.54	25.03 ± 2.63	92.94 ± 1.59	0.00 ± 0.00	84.22 ± 1.15	60.48 ± 0.97
Qec‐6	36.74 ± 0.33	92.83 ± 0.56	39.36 ± 1.65	84.62 ± 2.86	32.88 ± 1.48	76.79 ± 0.83	67.96 ± 2.48
Qec‐7	32.17 ± 2.14	93.71 ± 0.37	24.75 ± 1.98	92.59 ± 0.41	0.00 ± 0.00	85.64 ± 1.93	43.24 ± 0.79
Qec‐8	21.81 ± 2.13	85.62 ± 1.82	18.48 ± 2.62	87.00 ± 2.32	44.33 ± 1.14	58.08 ± 1.09	0.00 ± 0.00
Qed‐1	53.83 ± 2.35	100.00 ± 0.00	24.60 ± 0.88	100.00 ± 0.00	50.00 ± 3.32	92.14 ± 0.35	73.80 ± 0.58
Qed‐2	54.09 ± 0.45	98.72 ± 1.41	58.70 ± 2.19	100.00 ± 0.00	45.26 ± 1.26	82.42 ± 3.01	80.16 ± 1.80
Qed‐3	53.46 ± 1.83	100.00 ± 0.00	51.80 ± 1.96	96.81 ± 0.95	44.59 ± 1.65	83.97 ± 0.48	78.95 ± 1.84
Qed‐4	54.90 ± 0.43	96.56 ± 0.64	64.86 ± 0.87	100.00 ± 0.00	47.51 ± 1.54	70.49 ± 0.74	83.91 ± 0.82
Qed‐5	49.95 ± 0.43	94.79 ± 0.43	53.48 ± 1.42	97.59 ± 1.02	36.90 ± 2.03	89.57 ± 0.23	79.64 ± 1.42
Qed‐6	41.10 ± 2.45	84.62 ± 1.80	32.68 ± 0.32	89.81 ± 1.15	39.10 ± 1.58	67.94 ± 2.49	36.66 ± 2.25
Qed‐7	54.32 ± 2.45	94.71 ± 1.04	54.21 ± 1.13	90.26 ± 1.26	49.20 ± 1.70	66.21 ± 1.67	80.04 ± 1.96
Qed‐8	54.29 ± 1.42	93.46 ± 0.69	57.01 ± 1.70	100.00 ± 0.00	52.19 ± 0.85	73.22 ± 2.84	72.78 ± 2.51
Qed‐9	59.58 ± 2.13	86.79 ± 0.43	57.91 ± 0.84	94.81 ± 0.42	54.83 ± 0.67	61.21 ± 0.41	81.41 ± 1.48
Thifluzamide	98.63 ± 1.04	98.86 ± 0.13	62.51 ± 0.76	98.64 ± 0.49	77.61 ± 1.51	83.59 ± 0.44	51.98 ± 3.23
Azoxystrobin	68.33 ± 2.43	67.8 ± 1.71	46.48 ± 1.27	93.54 ± 1.42	28.75 ± 2.41	50.63 ± 1.91	69.26 ± 1.94
Boscalid	90.5 ± 2.54	21.23 ± 1.86	40.97 ± 0.77	51.78 ± 1.17	30.67 ± 1.42	36.25 ± 1.31	96.36 ± 0.93

Note: Rhizoctonia solani (R. s), Cytospora chrysosperma (C. c), Fusarium oxysporum (F. o), Magnaporthe oryzae (M. o), Fusarium graminearum (F. g), Sphaeropsis sapinea (S. s), and Botrytis cinerea (B. c).

^† All values are the mean of three replicates.

To quantitatively characterize the antifungal potency, we established the dose–response relationships for the lead compound, Qed‐3, as a representative example. As illustrated in Fig. 5(A), Qed‐3 exhibited a typical concentration‐dependent inhibitory effect against these three pathogens. The sigmoidal curves intuitively reflect the higher sensitivity of compound Qed‐3 towards S. sapinea and C. chrysosperma compared to F. graminearum, which is highly consistent with the calculated EC₅₀ values presented in Fig. 5(B). Following this representative analysis, a comprehensive evaluation of the EC₅₀ data for all derivatives was conducted to elucidate the structure–activity relationships. The EC₅₀ values presented in Table S1 (Supporting Information) further demonstrate that aromatic amine functionalization at the C2 position of quinine affords derivatives with substantially enhanced antifungal efficacy compared to their aliphatic or alicyclic amine‐substituted analogs. The aromatic/heteroaromatic amine‐modified quinine derivatives (Qd, Qe series) exhibited 5‐ to 20‐fold reductions in EC₅₀ values against F. graminearum relative to the Qa series. From an electronic perspective, strong electron‐withdrawing substituents‐particularly halogen atoms (chlorine (Cl), fluorine (F), bromine (Br)) on the aromatic amine ring‐were identified as pivotal determinants of high potency. This establishes the Qed series as the most efficacious subclass. The introduction of these halogens significantly alters the local electron density, perfectly aligning with the QSAR model's identification of polarizability (AATSC5p) and electronegativity (ETA_Epsilon) as key drivers of bioactivity (which will be discussed in detail in the subsequent section). Conversely, electron‐donating groups, such as the methoxy group (OCH₃), displayed significantly diminished efficacy against S. sapinea, likely due to unfavorable electronic repulsion or weakened electrostatic interactions within the target binding site. Furthermore, moderate electron‐withdrawing or weak electron‐donating groups like cyano (CN) and methyl (CH₃) only conferred moderate activity. Regarding steric hindrance and the positional effects (i.e., ortho‐, meta‐, and para‐substitutions) of halogen substituents, the location of the halogen on the aromatic ring exerted profound and pathogen‐specific regulatory effects. For instance, para‐substitution (e.g., the para‐chloro Qed‐3) achieved optimal activity against S. sapinea (EC₅₀ = 0.46 μg mL⁻¹), dramatically exceeding the performance of its meta‐chloro analog (Qed‐4, EC₅₀ = 12.85 μg mL⁻¹). This implies that for S. sapinea, the para‐position introduces minimal steric clash within the primary binding pocket, allowing deeper penetration of the molecule. In contrast, ortho‐substitution (e.g., the ortho‐chloro Qed‐5) maintained high‐efficiency inhibition against both C. chrysosperma (0.82 μg mL⁻¹) and F. graminearum (3.38 μg mL⁻¹). Chemically, the bulky ortho‐halogen may cause significant steric hindrance, restricting the free rotation of the C–N linker and rigidifying the structure to better fit the spatial requirements of the target enzymes in the pathogen. Collectively, the enhanced potency of structurally optimized derivatives (e.g., Qed‐3, Qed‐5, Qeb‐1) markedly exceeded that of quinine, underscoring their developmental potential as novel antifungal agents.

Figure 5.

(A) Dose–response curves of compound Qed‐3 against Sphaeropsis sapinea, Cytospora chrysosperma, and Fusarium graminearum. Data points represent the mean of independent replicates (± standard deviation). (B) The EC₅₀ values of compound Qed‐3 against the three representative pathogens.

3.3. In vivo antifungal bioactivity of Qed‐3 against S. sapinea

To substantiate the agricultural applicability of the lead compound, in vivo antifungal assays were then performed on S. sapinea‐infected Pinus needles. ⁴⁶ The conventional fungicide boscalid was employed as the benchmark reference to assess the relative performance of Qed‐3. As depicted in Fig. 6, the untreated control group (CK) exhibited extensive necrotic lesions, averaging 86.30 ± 3.93 mm in length. Remarkably, Qed‐3 treatment exerted dose‐responsive suppression of pathogen progression, with its inhibitory effects consistently surpassing those of boscalid across all evaluated concentrations (5, 10, 20, and 40 μg mL ⁻¹ ). Specifically, at the reduced concentration of 10 μg mL ⁻¹ , Qed‐3 demonstrated exceptional disease suppression, limiting lesion development to 14.67 ± 1.12 mm (83.0% control efficacy). This performance markedly exceeded that of the reference standard boscalid, which produced substantially longer lesions of 54.68 ± 2.31 mm (36.6% efficacy) at equivalent concentration. Notably, at 40 μg mL⁻¹, Qed‐3 achieved near‐complete protection with 90% efficacy (8.65 ± 0.52 mm lesions), confirming its robust therapeutic potential against S. sapinea pathogenesis. These in vivo findings exhibit strong concordance with prior in vitro activity profiles, conclusively positioning Qed‐3 as a preeminent lead compound for managing S. sapinea‐associated arboreal pathologies.

Figure 6.

In vivo protective efficacy of compound Qed‐3 against Sphaeropsis sapinea infection on Pinus needles. Comparative analysis of lesion progression and protective efficiency versus boscalid standard. Results expressed as mean ± standard deviation.

3.4. QSAR model and statistical analysis

Given the target compounds' pronounced efficacy against S. sapinea, this pathogen was selected as the model strain for QSAR investigation.

3.4.1. Computational chemistry analysis

Following molecular geometry optimization performed with GaussView 6, a comprehensive set of 1444 molecular descriptors were systematically computed utilizing PaDEL‐Descriptor software. These calculated descriptors encompassed: (a) zero‐dimensional (0D) constitutional descriptors; (b) one‐dimensional (1D) descriptors, including functional group enumerations and atom‐centric fragment features; (c) two‐dimensional (2D) descriptors, specifically topological indices, molecular walk/path enumerations, connectivity metrics, information‐theoretic indices, 2D spatial autocorrelations, edge adjacency descriptors, Burden matrix eigenvalues, topological charge distributions, and spectral graph invariants; (d) electronic charge distribution descriptors; and (e) intrinsic molecular property parameters.

3.4.2. QSAR modeling protocol

The experimental dataset was strategically partitioned into a training set comprising 25 compounds and an independent test set of seven compounds through application of the Kennard–Stone selection algorithm. ⁵⁶ To discern molecular descriptors exhibiting maximal correlation with EC₅₀ bioactivity values, all 1444 structural descriptors computed by QSARINS computational platform were systematically screened for model development. The optimal descriptor subset dimensionality was established when incremental variable incorporation ceased to produce statistically meaningful enhancements in predictive performance. Model robustness was rigorously assessed via leave‐one‐out (LOO) cross‐validation methodology. The resultant four‐descriptor QSAR model (Eqn (1)) yielded determination coefficients (R ² = 0.7655; Q ² _LOO = 0.6533) as illustrated in Table 2. While these metrics demonstrate satisfactory predictive capacity, approximately 25% of the variance remains (S: standard error) unexplained. The relatively high standard error mainly arises from experimental variability, structural diversity of the dataset, and the inherent approximation of QSAR statistical models. Despite such unavoidable prediction deviations, the model still demonstrates reliable structure–activity trends and reasonable predictive power for compound optimization, providing valuable guidance for further rational design. Comprehensive statistical evaluation confirmed the final model's stability, robustness, and reliable predictive power. The antilogarithm‐transformed EC₅₀ predictions generated by the optimized model are shown in Table 3, with corresponding regression diagnostics visualized in Fig. 7(A).

$$\mathrm{Log}\left({\mathrm{EC}}_{50}\right)=\begin{array}{l}-41.5757\text{AATSC}5\mathrm{p}–0.5744\mathrm{C}3\text{SP2}–0.4285\text{nHBint}10–98.9303\mathrm{ETA}\_\text{Epsilon}+46.9974\end{array}$$ (1)

Table 2.

Statistical parameters of the optimal quantitative structure–activity relationship (QSAR) model

Training set (n = 25)		Prediction set (n = 7)
R 2	0.7655	Q 2 LOO	0.6533
R 2 adj	0.7186	R 2 − Q 2 LOO	0.1122
R 2 − R 2 adj	0.0469	RMSEcv	0.2666
LOF	0.104	MAEcv	0.2359
K xx	0.1765	PRESScv	1.7775
∆K	0.0517	CCCcv	0.8045
RMSEtr	0.2193	CCCtr	0.8672
MAEtr	0.1902	S	0.2452
RSStr	1.2022	F	16.3212

Note: n, number of investigated compounds in set; R ², squared regression coefficient; LOF, friedman's lack of fit; K _xx, global correlation among descriptors; ∆K, difference in correlation; RMSE, root‐mean‐square error; MAE, mean absolute error; RSS, residual sum of squares; CCC, concordance correlation coefficient; S, standard error; F, variance ratio, Fisher coefficient; Q ² _LOO, weighted correlation coefficient by leave‐one‐out method. PRESS, predicted residual error sum of squares.

Table 3.

Experimental and predicted activity logarithmic half‐maximal effective concentration (log(EC₅₀)) by the developed quantitative structure–activity relationship (QSAR) model

Compound	Status	Experimental activity	Predicted activity
Qa‐1	Training	1.49	1.38
Qa‐2	Training	1.20	1.14
Qa‐3	Training	1.32	1.16
Qb‐1	Prediction	0.59	0.41
Qc‐1	Training	0.94	0.76
Qc‐2	Prediction	0.93	0.91
Qc‐3	Training	0.84	0.60
Qd‐1	Training	0.41	0.65
Qd‐2	Training	0.48	0.91
Qd‐3	Training	0.84	0.81
Qea‐1	Training	0.17	0.03
Qea‐2	Training	−0.10	0.03
Qea‐3	Training	0.23	0.46
Qeb‐1	Training	−0.02	−0.12
Qeb‐2	Training	−0.02	−0.23
Qec‐1	Prediction	0.61	0.42
Qec‐2	Training	0.39	0.81
Qec‐3	Training	1.05	0.97
Qec‐4	Training	0.71	0.63
Qec‐5	Training	0.60	0.77
Qec‐6	Prediction	0.94	0.80
Qec‐7	Training	0.56	0.39
Qec‐8	Training	0.62	0.52
Qed‐1	Prediction	0.19	0.20
Qed‐2	Training	0.77	0.62
Qed‐3	Training	−0.34	−0.47
Qed‐4	Prediction	1.11	0.74
Qed‐5	Training	0.31	0.55
Qed‐6	Training	0.88	0.50
Qed‐7	Prediction	1.01	0.74
Qed‐8	Training	0.16	0.30
Qed‐9	Training	1.07	0.87

Figure 7.

QSAR model validation plots for the optimal four‐descriptor model. (A) Experimental versus predicted log(EC₅₀) values for the training set (n = 25) and test set (n = 7). (B) Williams plot of standardized residuals versus leverage values (h* = 0.600) used to define the applicability domain. No outliers were detected within the warning leverage limit.

3.4.3. Model interpretation analysis

The interpretability of the constructed QSAR model is substantiated by providing detailed chemical explanations for the physicochemical significance of its selected molecular descriptors: (i) AATSC5p (Autocorrelation of Topological Structure Codes) ⁵⁷ : Quantifies the weighted spatial autocorrelation at topological distance 5 using atomic polarizability. Chemically, a higher polarizability implies that the molecule's electron cloud can dynamically distort, facilitating optimal van der Waals forces and dipole–induced dipole interactions with non‐polar amino acid residues within the target enzyme's hydrophobic binding pocket. (ii) C3SP2 (Hybridized Carbon Typology): Specifically denotes a sp²‐hybridized carbon center forming three covalent bonds with adjacent carbon atoms. This descriptor directly correlates with the presence of conjugated aromatic rings (such as the quinoline scaffold and substituted aniline moiety). The abundance of these planar sp²‐hybridized systems is critical for stabilizing the drug–target complex through robust π–π stacking and cation–π interactions with aromatic residues at the active site. (iii) nHBint10 (Hydrogen Bond Interaction Potential) ⁵⁸ : Enumerates the electrotopological state values for hydrogen bond donor/acceptor pairs within a ten‐bond topological radius. This highlights that a precise spatial distribution of hydrogen bond donors and acceptors (e.g., the secondary amine linker and the core structural nitrogens) is chemically vital for anchoring the fungicide to specific polar residues in the binding pocket, directly driving binding affinity and target specificity. (iv) ETA_Epsilon (Electronegative Topochemical Atom): ⁵⁹ Computes the relative density of electronegative atomic centers within the molecular framework. Chemically, the positive contribution of electronegativity underscores the importance of introducing highly electronegative atoms (such as Cl, F, Br halogens). These electronegative centers not only modulate the compound's overall lipophilicity for improved penetration through fungal cell membranes but also enhance localized polar interactions and safeguard the molecule against rapid metabolic degradation.

3.4.4. Model validation analysis

In this investigation, the applicability domain (AD) assessment and prediction reliability verification of the established model were systematically conducted through Williams plot analysis employing the leverage methodology, as shown in Fig. 7(B). Critical examination of the plot revealed that all training set compounds consistently remained below the predefined warning leverage threshold (h* = 0.600). Moreover, comprehensive screening confirmed the absence of response outliers across both the training cohort and independent test set. ⁶⁰

3.5. Investigation on the morphology and cell structure of S. sapinea hyphae

To elucidate the antifungal mechanism of compound Qed‐3 against S. sapinea, SEM was utilized to examine hyphal morphological alterations at 0.5 μg mL⁻¹. As demonstrated in Fig. 8(B), blank control hyphae displayed standard developmental patterns, maintaining structural continuity, consistent diameter, and undisturbed surface topology. Conversely, Qed‐3 exposure provoked severe morphological disruptions. Treated hyphae exhibited pronounced deformation, surface corrugation, aberrant branching patterns, and overall structural collapse, implying extensive fungal cellular damage. The impact of compound Qed‐3 (0.5 μg mL⁻¹) on the hyphal ultrastructure of S. sapinea was examined via TEM. As illustrated in Fig. 8(A), untreated control mycelia exhibited normal morphological integrity, featuring continuous cell wall architecture, intact plasma membranes, and uniformly distributed cytoplasmic organelles. In stark contrast, Qed‐3 treatment induced severe ultrastructural alterations: intracellular organelles underwent fragmentation and disorganization, the plasma membrane detached from the cell wall (plasmolysis phenomenon), and cytoplasmic contents leaked extensively. These observations collectively demonstrate that Qed‐3 effectively degrades fungal cell wall structural stability and critically impairs membrane permeability regulation.

Figure 8.

A preliminary exploration of the antifungal mechanism of compound Qed‐3 against Sphaeropsis sapinea. (A) Transmission electron micrographs of S. sapinea mycelia. (i, iii) 0.5% DMSO control; (ii, iv) Qed‐3 treatment (0.5 μg mL⁻¹ in 0.5% DMSO). (B) Scanning electron micrographs of S. sapinea mycelia. (i, iii) 0.5% DMSO control; (ii, iv) Qed‐3 treatment (0.5 μg mL⁻¹ in 0.5% DMSO). (C) Cell membrane integrity visualized by propidium iodide (PI) staining. (i) 0.5% DMSO control; (ii) Qed‐3‐treated (12 μg mL⁻¹ in 0.5% DMSO); (iii) thifluzamide positive control (12 μg mL⁻¹ in 0.5% DMSO). (D) Reactive oxygen species (ROS) accumulation visualized by DCFH‐DA staining. (i) 0.5% DMSO control; (ii) Qed‐3‐treated (12 μg mL⁻¹ in 0.5% DMSO); (iii) thifluzamide positive control (12 μg mL⁻¹ in 0.5% DMSO).

When fungal cell wall integrity is compromised, the plasma membrane becomes vulnerable to direct exposure to extracellular conditions. ⁴⁹ PI staining was employed to examine membrane integrity. As evidenced in Fig. 8(C), both the untreated control and thifluzamide‐treated positive control exhibited negligible red fluorescence, confirming preserved membrane integrity. In contrast, hyphae exposed to 12 μg mL⁻¹ Qed‐3 displayed intense and widespread PI fluorescence, demonstrating substantial membrane disruption that enabled PI permeation and subsequent nuclear staining. These findings verify Qed‐3's capacity to destabilize the permeability barrier of S. sapinea plasma membranes. Beyond CYP51 enzymatic dysfunction‐induced structural reorganization of eukaryotic membrane systems leading to permeability dysregulation, ROS induce fungal cell membrane damage through phospholipid peroxidation, which increases membrane permeability and results in cell death. To assess intracellular ROS production, hyphae were stained with the fluorescent probe DCFH‐DA. As illustrated in Fig. 8(D), untreated hyphae exhibited complete absence of green fluorescence signals, while the thifluzamide‐treated group displayed only marginal fluorescence emission. Strikingly, hyphae exposed to 12 μg mL⁻¹ Qed‐3 demonstrated substantially enhanced fluorescence intensity compared to both control groups. These observations indicate that Qed‐3 exposure triggers a pronounced surge in endogenous ROS generation, wherein the consequent oxidative damage serves as one of the fundamental pathways mediating fungal programmed cell death.

3.6. Molecular docking

Fusarium graminearum CYP51B, recognized as the most conserved CYP51 gene across fungal species, encodes the enzyme primarily responsible for sterol 14α‐demethylation, a role essential for ascospore formation. ⁶¹ Given the potent growth inhibitory effects of quinine‐derived compounds Qed‐5 (EC₅₀ = 3.38 μg mL⁻¹) and Qed‐3 (EC₅₀ = 5.06 μg mL⁻¹) against F. graminearum, molecular docking simulations were performed using the F. graminearum CYP51 homology model (UniProt ID: I1RBR4) to elucidate their binding mechanisms with this sterol 14α‐demethylase, employing the natural product quinine as reference ligand (Fig. 9). Compounds Qed‐5 and Qed‐3 exhibited superior binding affinity with a calculated energy of −9.2 kcal mol⁻¹, significantly outperforming quinine.

Figure 9.

Comparative structural docking analysis of quinine (top panel), its optimized derivatives Qed‐5 (middle panel) and Qed‐3 (bottom panel) with Fusarium graminearum CYP51 sterol 14α‐demethylase.

Although Qed‐5 and Qed‐3 exhibit identical calculated binding energies, their distinct in vitro activities (3.38 versus 5.06 μg mL⁻¹) suggest that empirical energy scores alone are insufficient to fully explain the efficacy differences. The distinct substitution positions of the Cl atom alter the local steric bulk, forcing the two highly similar regioisomers to adopt distinct binding poses to accommodate the rigid CYP51 pocket. Therefore, the superior activity of Qed‐5 is likely driven by its specific conformation, which perfectly engages critical functional regions.

Analysis of the CYP51–Qed‐5 complex architecture (middle panel, Fig. 9) revealed the specific molecular interactions responsible for its improved binding affinity. (i) A distinct hydrogen bond forms between Asp227 and the hydroxyl group of Qed‐5 (2.3 Å). (ii) The phenyl ring of aromatic amine introduced at the C2 position of quinine engages in hydrophobic π–alkyl interactions with the alkyl moiety of Lys192 and Ala195. (iii) The above hydrogen bonding and hydrophobic interactions position the methoxy group on Qed‐5's quinoline ring in proximity to Ser314, Ser315, and Thr316, establishing weak but significant interactions. CYP51's crystal structure indicates this region coincides with a critical discontinuity and folding of the I‐helix, a structural feature essential for forming the active site channel. By occupying this channel, Qed‐5 likely obstructs either sterol substrate access to the active site or the exit of 14α‐demethylated metabolites, ultimately compromising membrane integrity and function. ⁶² The results demonstrate that Qed‐5 possesses the capacity to form a thermodynamically stabilized complex within the CYP51 enzymatic active site, exhibiting significantly enhanced inhibitory potency compared to the parent quinine (binding energy difference, ∆G = 0.9 kcal mol⁻¹). Nevertheless, the precise molecular interactions between this lead compound and its target enzyme, particularly its modulatory effects on the conformational dynamics of the I‐helix, require further elucidation.

4. CONCLUSIONS

In this investigation, 32 structurally optimized quinine derivatives were designed and synthesized through rational molecular modification. Antifungal assessment against seven phytopathogenic fungi demonstrated that the majority of synthesized compounds possessed remarkable inhibitory potency. Of particular note, Qed‐3 exhibited broad‐spectrum efficacy, showing exceptional activity against S. sapinea (EC₅₀ = 0.46 μg mL⁻¹), surpassing the performance of the conventional fungicide thifluzamide. SEM/TEM analyses revealed Qed‐3 induces catastrophic cellular deformation, including hyphal collapse, organelle fragmentation, and cytoplasmic shrinkage, unequivocally demonstrating its dual capacity to compromise both cell wall architecture and membrane functionality. Complementary biochemical assays (PI staining and ROS detection) further established Qed‐3's ability to induce permanent membrane perforation and lethal oxidative burst, ultimately precipitating fungal apoptosis.

QSAR modeling generated a predictive computational model (R ² = 0.7655, Q ² = 0.6533) that pinpointed electronegative moieties and hydrogen‐bond donor capacity as crucial pharmacophoric elements. Docking simulations with F. graminearum CYP51 confirmed Qed‐5's superior binding affinity, facilitated by a synergistic combination of hydrogen bonding (Asp227) and hydrophobic complementarity (Lys192/Ala195). 2‐Arylaminoquinine constitutes a structurally optimized lead compound for the development of innovative botanical‐based plant protection fungicides. Subsequent investigations will focus on systematic structural refinement to enhance antifungal efficacy, definitive identification of molecular targets, and elucidation of the biochemical mechanisms underlying its antifungal activity.

CONFLICT OF INTEREST

The authors declare no competing financial interest.

Supporting information

Data S1. Antifungal activity assay in vitro.

Data S2. In vivo antifungal activity test.

Data S3. Morphological characterization via SEM.

Data S4. Morphological characterization via TEM.

Data S5. Evaluation of cell membrane integrity.

Data S6. Reactive oxygen species detection.

Data S7. Quantitative structure–activity relationship modeling.

Figure S1. Synthetic route for quinine derivatives.

Data S8. Synthetic route of key intermediates and target compounds.

Data S9. ¹H NMR, ¹³C NMR and HRMS spectrum of compounds.

Table S1. EC₅₀ of the target compounds against four plant pathogenic fungi.

DATA AVAILABILITY STATEMENT

The data that supports the findings of this study are available in the supplementary material of this article.