Synthesis, Bioactivity and Molecular Docking of Nereistoxin Derivatives Containing Phosphonate

Abstract

Novel nereistoxin derivatives containing phosphonate were synthesized and characterized via ³¹P, ¹H and ¹³C NMR and HRMS. The anticholinesterase activity of the synthesized compounds was evaluated on human acetylcholinesterase (AChE) using the in vitro Ellman method. Most of the compounds exhibited good inhibition of acetylcholinesterase. All of these compounds were selected to assess their insecticidal activity (in vivo) against Mythimna separata Walker, Myzus persicae Sulzer and Rhopalosiphum padi. Most of the tested compounds displayed potent insecticidal activity against these three species. Compound 7f displayed good activity against all three insect species, showing LC₅₀ values of 136.86 μg/mL for M. separata, 138.37 μg/mL for M. persicae and 131.64 μg/mL for R. padi. Compound 7b had the highest activity against M. persicae and R. padi, with LC₅₀ values of 42.93 μg/mL and 58.19 μg/mL, respectively. Docking studies were performed to speculate the possible binding sites of the compounds and explain the reasons for the activity of the compounds. The results showed that the compounds had lower binding energies with AChE than with the acetylcholine receptor (AchR), suggesting that compounds are more easily bound with AChE.

1. Introduction

Lepidoptera and Hemiptera are destructive pests of many crops in the world. The larvae of many lepidopteran species feed on leaves, and can eat all the leaves of crops when they occur in large areas, causing serious losses of crops [1,2]. Hemipteran pests can also damage crops by causing abnormal plant growth, mold growth and the transmission of plant viruses [2,3,4]. For example, the oriental armyworm (Mythimna separate Walker), a typical lepidopteran pest, and aphids, the most difficult hemipteran pests to control in many crops, are widely distributed in areas where rice, wheat and corn are grown. Although various insecticides have been discovered and registered to control lepidopteran and hemipteran pests, the excessive and repeated application of these agricultural chemicals has led to the development of resistance in lepidopteran and hemipteran pest populations and an increase in the burden they place on the environment [5,6].

Natural products are the components or secondary metabolites of plants, animals or micro-organisms in nature, which ordinarily are harmless to the environment. An important field of new pesticide research involves discovering and developing new compounds with insecticidal activity from nature or using them as lead compounds for structural modification [7]. Nereistoxin (NTX, Figure 1), a natural small-molecule toxin with insecticidal activity, was isolated from Lumbriconereis heteropoda, which lives in marine sediments [8]. It was the first example of an animal toxin that was successfully used as a basis for analog synthesis. Nereistoxin insecticides are effective for lepidopteran, hemipteran and coleopteran pest control. Konishi synthesized and studied the insecticidal activity of many nereistoxin derivatives and established the structure-activity relationship (SAR) of nereistoxin derivatives [9,10,11,12]. However, there are only five insecticide products (Figure 1) currently on the market that use nereistoxin as their active ingredient. The long-term and excessive use of these insecticides will inevitably cause pest resistance. It is necessary to develop an effective way to synthesize new nereistoxin insecticides to extend the application of such insecticides.

Organophosphorus (OP) compounds are an important chemical class with many applications. OPs are used in agricultural chemicals, in medicines, and in the chemical industry and as nerve agents [13,14,15]. Phosphorothioates and dithioates are important classes of OPs that have applications in medicines and pesticides due to their biological properties [16,17,18]. An important characteristic of thiophosphate compounds is that they have low environmental stability in plants, animals and soil micro-organisms and can be rapidly decomposed into less toxic metabolites. This makes them safer for humans, domestic animals and the environment [19].

Zhang et al. reported the synthesis of four phosphorothioate and dithioate nereistoxin derivatives [20,21,22,23]. Their report shows that the dithiotic compounds have excellent insecticidal activity against Plutella xylostella (Linnaeus) and wheat aphid. In this work, we developed a series of phosphorothioate nereistoxin derivatives with disulfide bonds opened and modified by phosphate esters in a simple, mild and fast way. The derivatives were characterized via ¹H, ¹³C and ³¹P nuclear magnetic resonance (NMR) and high-resolution mass spectrometry (HRMS). We studied the insecticidal activity of the derivatives against pre-third-instar larvae of the oriental armyworm (M. separata) and aphids (M. persicae, R. padi) in vivo, as well as their inhibition properties on acetylcholinesterase (AChE) in vitro. Molecular docking was explored to identify the binding site and the reasons for the different potencies of the compounds.

2. Results and Discussion

2.1. Chemistry

The general synthetic route of compounds 7a–7h is shown in Scheme 1. The key intermediate 3 was conveniently synthesized in two steps from 2-amino-1,3-propanediol 1. The amino group of 1 was dimethylated via the Eschweiler–Clarke reaction, and then, hydroxyl was chlorinated with SOCl₂ to obtain 1,3-dichloro-N,N-dimethyl-2-propanamine hydrochloride 3. Other crucial intermediates, such as S-hydrogen phosphorothioates 6, were obtained from diphosphite 5. Some of them were commercially available and others were obtained via reaction of the appropriate alcohol and PCl₃, with pyridine as an acid-binding agent. Diphosphite 5 was reacted with S₈ and triethylamine to obtain 6. The end-products of 7 were synthesized via a substitution reaction between intermediates 3 and various substituted S-hydrogen phosphorothioates 6. The target compounds 7a–7h were characterized via ¹H NMR, ¹³C NMR, ³¹P NMR and HRMS.

2.2. Biological Activity

In this study, in accordance with the guidelines for antipesticide laboratory biological activity tests, the insecticidal activity of all synthesized derivatives 7a–7h against common agricultural pests (M. separata, M. persicae and R. padi) were evaluated in the laboratory. The representative commercial insecticides chlorpyrifos, thiamethoxam and flunicotamid were used as positive controls, respectively. The target compounds 7a–7h were also evaluated for acetylcholinesterase inhibitory activity in vitro.

2.2.1. Acetylcholinesterase (AChE) Assay (in vivo)

Organophosphate compounds are inhibitors of AChE [24], and nereistoxin is also proven to be a weak AChE inhibitor [25]. Thus, the cholinesterase inhibitory activity of the synthesized derivatives 7a–7h was evaluated spectrophotometrically against AChE (human) in vitro using the modified Ellman’s method [26,27]. The results are summarized in

Table 1. Average inhibition rates of NTX and compounds 7a–7h on AChE at 2 mmol/L (n = 3).

Compound	R	AChE Inhibition Rate (%)
		5 min	20 min	30 min
7a	Me	90.78	96.10	96.26
7b	Et	91.90	96.82	97.11
7c	nPr	92.71	97.13	97.39
7d	iPr	92.73	97.10	97.43
7e	nBu	92.87	97.21	97.48
7f	tBu	89.56	94.58	94.51
7g	Ph	72.84	66.73	60.64
7h	Bn	92.59	97.08	97.35
Nereistoxin	/	66.02	53.98	41.27

.

The inhibition rates of the compounds exceeded 90% within five minutes. The inhibition was maintained for a while and increased to more than 94% within 30 min, except for 7g. The nereistoxin inhibition of AChE was only 66.02% at 5 min, while the inhibitory effect weakened as time elapsed. The inhibition rate was reduced to 41.27% after 30 min, confirming that nereistoxin is a weak inhibitor of AChE. Compounds 7a–7g exhibited good inhibitory effects on AChE. Based on the average inhibition rates of the compounds listed in Table 1, 7a–7h were chosen for further bioassays. The IC₅₀ and K_i values of the compounds for AChE were calculated (

Table 2. Experimental IC₅₀ and K_i values of all synthesized compounds pertinent to AChE.

Compound	R	IC50 (μM)	Final Concentration (μM) a	Ki (mM−1⋅min−1)
7a	Me	3.332	3.125	115.86 ± 64.99
7b	Et	3.422	3.125	83.83 ± 3.12
7c	nPr	10.28	12.5	33.95 ± 5.62
7d	iPr	37.83	50	7.92 ± 3.73
7e	nBu	179.3	200	0.94 ± 0.21
7f	tBu	— b	—	—
7g	Ph	— c	—	—
7h	Bn	61.4	50	3.67 ± 0.14
Nereistoxin	/	— d	—	—

^a The concentration of the compounds after adding them to the PBS buffer solution containing enzyme, substrate and chromogenic agent DTNB; ^b hydrolyzed easily; ^c poor solubility; ^d the inhibition rate of 2 mM NTX on AChE was 41.27% (30 min), and the IC₅₀ value could not be calculated.

).

The AChE inhibition IC₅₀ values ranged from 3.332 to 179.3 µM for the synthesized compounds, except for the unstable compounds 7f and 7g with poor solubility (Table 2). Compounds 7a and 7b produced high inhibition, with IC₅₀ values of 3.3 μM and 3.4 μM, respectively. The IC₅₀ values gradually increased from 7a to 7e, which suggests that the AChE inhibitory activity decreased as the substituted alkyl chain of phosphorothioate increased. The steric hindrance of various phosphate groups had a bigger influence on the inhibitory activity of AChE, and the lower the steric hindrance of the group, the stronger the inhibitory activity of AChE, according to a thorough analysis of the IC₅₀ values for 7a–7h.

2.2.2. Insecticidal Activity against M. separata

The insecticidal activity of NTX and its derivatives 7a–7h against the pre-third-instar larvae of M. separata was tested via the topical application method at a concentration of 2 mg/mL. Chlorpyrifos was used as the positive control at a concentration of 50 μg/mL. Most compounds showed excellent insecticidal bioactivity against M. separata at 24 h and 48 h (

Table 3. Corrected mortality rate of compounds 7a–7h, nereistoxin and positive controls (chlorpyrifos) against M. separata at 2 mg/mL.

Compound	R	Corrected Mortality Rate (%)
		24 h	48 h
7a	Me	100	100
7b	Et	100	100
7c	nPr	100	100
7d	iPr	70	62.5
7e	nBu	70	62.5
7f	tBu	100	100
7g	Ph	100	100
7h	Bn	40	25
Nereistoxin	/	100	100
Chlorpyrifos	/	100	100

). The mortality values at 24 h and 48 h produced by 7a, 7b, 7c, 7f and 7g at concentrations of 2 mg/mL were all 100%. It is notable that the insecticidal activity of nereistoxin against M. separate was also 100%. This indicates that M. separate can be effectively controlled by synthetic chemicals and nereistoxin in a short amount of time.

Based on these preliminary results (Table 3), the compounds with superior bioactivity were chosen for further bioassays against M. separata at different concentrations (The different concentrations of 7a, 7b and 7c were 125, 250, 500, 1000 and 2000 μg/mL; those of 7f and 7g were 31.25, 62.5, 125, 250 and 500 μg/mL. Because of the high concentrations used, all the tested larvae died, making it impossible to calculate LC₅₀ values.) The results of these bioassays are presented as half of the lethal concentration (LC₅₀; μg/mL) in

Table 4. LC₅₀ values of potent compounds, nereistoxin and chlorpyrifos against M. separata in the laboratory.

Compound	y = ax + b	LC50 (μg/mL)	R2	CI (95%)
7a	y = 1.0818x + 1.8613	796.91	0.9796	336.36–1888.06
7b	y = 1.3741x + 0.9845	836.34	0.9583	408.44–1712.54
7c	y = 1.6902x + 0.0770	817.84	0.9653	448.10–1492.67
7f	y = 1.3767x + 2.0590	136.86	0.9864	73.56–254.45
7g	y = 1.7743x + 1.0230	174.33	0.9979	102.15–297.51
Nereistoxin	y = 1.8644x + 0.9391	150.71	0.9784	92.25–246.21
Chlorpyrifos	y = 1.6690x + 2.9764	16.31	0.9829	9.44–28.20

. The compounds used in the advanced bioassays were 7a, 7b, 7c, 7f, 7g and nereistoxin. These compounds displayed substantial biological activity in the laboratory against M. separata, with LC₅₀ values ranging from 136.86 to 836.34 μg/mL (Table 4). From the 95% confidence interval (CI), compound 7f showed the highest biological activity, with an LC₅₀ value of 136.86 μg/mL, followed by NTX, with an LC₅₀ value of 150.71 μg/mL. NTX was slightly less active than 7f but more active than other synthesized derivatives. The LC₅₀ value of the positive control chlorpyrifos was only 16.31 μg/mL, indicating that the insecticidal activity of the synthesized derivative compounds against M. separata was much lower than that of chlorpyrifos. The LC₅₀ values of 7f and 7g, however, were smaller than those of 7a, 7b and 7c, demonstrating that compounds 7f and 7g have superior insecticidal activity against M. separata compared to the other compounds. Additionally, the steric hindrance of groups on 7f and 7g phosphate esters was relatively high, suggesting that enhancing the steric hindrance of groups on phosphate esters can enhance the compound’s insecticidal activity against M. separata.

2.2.3. Insecticidal Activity against M. persicae Sulzer and R. padi

The insecticidal activity of NTX derivatives 7a–7h against M. persicae and R. padi were tested using the slide-dip method and leaf-dipping method at a concentration of 0.2 mg/mL. Thiamethoxam and flunicotamid were used as positive controls at concentrations of 40 μg/mL. The corresponding mortality rates caused by these compounds at 48 h post-treatment were generally higher than those at 24 h (

Table 5. Corrected mortality rates of compounds 7a–7h, nereistoxin and positive controls (thiamethoxam and flunicotamid) against M. persicae and R. padi at 0.2 mg/mL.

Compound	R	Corrected Mortality Rate (%)
		M. persicae		R. padi
		24 h	48 h	24 h	48 h
7a	Me	21.71	32.71	20.86	44.56
7b	Et	20.65	57.38	30.20	83.57
7c	nPr	23.16	33.45	21.32	42.43
7d	iPr	15.51	17.68	24.63	27.50
7e	nBu	42.93	40.25	43.01	43.12
7f	tBu	13.03	80.03	19.92	89.97
7g	Ph	3.96	3.61	7.15	25.45
7h	Bn	−0.63	23.27	0.48	32.23
Nereistoxin	\	14.25	34.19	12.71	40.22
Thiamethoxam	\	70.01	94.12	—	—
Flunicotamid	\	—	—		93.23

). For example, the corresponding mortality rates of 7b against M. persicae and R. padi after 24 h were 20.65% and 30.20%, respectively. However, the corresponding mortality rates were 57.38% and 80.57% after 48 h, which were nearly 2–3 times greater than those after 24 h. Compound 7f possessed the greatest bioactivity of the tested compounds. The corresponding mortality rates of 7f against M. persicae and R. padi at 24 h were 13.03% and 19.92%, respectively. However, after 48 h, the corresponding mortality rates of 7f sharply increased to 80.03% and 89.97%, which were almost 4.5–6 times higher than those after 24 h. These data suggest that the insecticidal effects of the compounds on both aphids can be effective in the short term. In contrast, the corresponding mortality rates against M. persicae and R. padi produced by NTX were at a moderate level among the tested compounds, with the corresponding mortality rates at 48 h being 34.19% and 40.22%, respectively.

Based on these preliminary results (Table 5), 7b and 7f were used, at different concentrations, for further bioassays against M. persicae and R. padi. The concentrations tested were 12.5, 25, 50, 100 and 200 μg/mL, and the results (LC₅₀; μg/mL) are shown in

Table 6. LC₅₀ values of potent compounds and thiamethoxam against M.persicae in the laboratory.

Compound	M. persicae
	y = ax + b	LC50 (μg/mL)	R2	CI (95%)
7b	y = 0.0014x + 0.4399	42.93	0.9540	41.37–67.58
7f	y = 0.0019x + 0.2371	138.37	0.9662	20.08–57.26
Thiamethoxam	y = 0.0194x + 0.2725	11.73	0.9684	21.80–77.11

and

Table 7. LC₅₀ values of potent compounds and flunicotamid against R. padi in the laboratory.

Compound	R. padi
	y = ax + b	LC50 (μg/mL)	R2	CI (95%)
7b	y = 1.6584x + 2.0732	58.19	0.9839	50.62–66.88
7f	y = 1.245x + 2.3602	131.64	0.9858	101.15–171.32
Flunicotamid	y = 2.2140x + 2.7094	10.83	0.9726	6.67–12.12

.

The LC₅₀ values of 7b against M. persicae and R. padi were 42.93 μg/mL and 58.19 μg/mL, respectively, while the 7f values were 138.37 μg/mL and 131.64 μg/mL (Table 6 and Table 7). These data indicate that 7b was more toxic to the two aphids than 7f, although Table 5 shows that the corresponding mortality rate of 7f against M. persicae at 48 h was much higher than 7b. The positive control LC₅₀ value of thiamethoxam against M. persicae was 11.73 μg/mL, and the LC₅₀ of flunicotamid against R. padi was 10.83 μg/mL. These values were both less than the LC₅₀ values corresponding to 7f. It is suggested that the insecticidal activity of these derivatives against M. persicae and R. padi is lower than that of thiamethoxam and flunicotamid. The LC₅₀ values of 7b against both aphids were lower than those of 7f, indicating that 7b has stronger insecticidal activity against both aphids than 7f. The steric hindrance of the substituent group on phosphate ester of compound 7b was smaller than that of 7f, suggesting that the compound was more active with less steric hindrance in the substituent group, which is consistent with the compounds’ inhibition of AChE. In other words, the compound’s anti-aphid effect is directly tied to its inhibition of AChE.

2.3. Structural Analysis of Docking

Since thiophosphates exert insecticidal effects by inhibiting the activity of acetylcholinesterase, while nereistoxin insecticides bind mainly to acetylcholine receptors (AchR), docking simulations were performed to describe the selective behavior of the compounds at the binding site using acetylcholinesterase and Alpha7 nicotinic acetylcholine receptor (α7-nAChR) as target proteins. After optimizing the structures of 7a–7h, they were used as small-molecule ligands, and two protein targets were used as acceptors. The interaction mode of the compounds and target proteins was analyzed to obtain the interactions between the compounds and the protein residues. The interactions could involve hydrogen bonding, π-π interaction and hydrophobic interactions. We referred to the docking score of the compounds with two acceptors to evaluate whether the synthesized compounds had biological activity. The molecular docking results are shown in

Table 8. Binding energies of compounds 7a–7h to active sites.

Compound	R	AChE (kcal/mol)		α7-nAchR (kcal/mol)
		AChE-1	AChE-2	α7-nAchR-1	α7-nAchR-2
7a	Me	−6.2	−6.1	−5.4	−4.8
7b	Et	−6.6	−6.4	−5.7	−4.9
7c	nPr	−7.0	−7.0	−5.9	−5.0
7d	iPr	−7.4	−6.9	−6.0	−5.5
7e	nBu	−7.2	−5.0	−5.9	−5.0
7f	tBu	−6.7	−5.4	−6.2	−5.6
7g	Ph	−9.2	−9.1	−8.1	−7.4
7h	Bn	−9.0	−6.2	−8.5	−6.5

.

The compounds were molecularly docked with two target proteins, respectively. Most compounds had a good binding effect with the two target proteins and had a high degree of matching (binding energy less than −6 kcal/mol). The binding energies of the compounds to the AChE site were lower than that of α7-nAchR, indicating that the compounds were more easily bound with acetylcholinesterase (Table 8). The complexes formed between the docking compounds and protein were visualized using Pymol 2.1 software (Schrödinger, New York, NY, USA)(the compound with the most negative binding energy was selected for each target), and the binding mode between the compound and protein was obtained. According to the binding mode, the binding amino acid residues of the compound and protein pockets were seen. Based on the results of compound substituents and acetylcholinesterase inhibitory activity, we selected 7b (alkyl substituents) and 7g (aryl substituents) for further analysis.

Regarding compound 7b, a strong hydrophobic interaction was formed between the O atoms of 7b and the amino acids (PRO-103, PHE-126) at the active site of α7-nAchR (Figure 2). This interaction plays an important role in stabilizing small molecules in the protein cavity. Also, strong hydrogen bond interactions were formed between the O atoms of 7b and the amino acids (ASN-129, LYS-109) of α7-nAchR. Weak hydrogen bonds also existed between 7b and amino acids (TYR-173, LEU-40, TRP-108). Furthermore, TRP-108 was also able to form a cation-π conjugation with 7b. These interactions make an important contribution to anchoring small molecules in the protein cavity.

The amino acids of α7-nAchR that can interact with 7g mainly include PRO-39, GLY-105, PRO-103, LEU-40, THR-128 and ASP-123 (Figure 3). Compound 7g was well matched with the active site of α7-nAchR. It was able to form three types of interaction with the amino acids of α7-nAchR: a strong hydrogen bond between the O atoms of 7g and LYS-109, hydrophobic interactions in PRO-39, PRO-103, LEU-40 and 7g, and conjugate interactions of GLY-105, ASP-123 and THR-128 with 7g. It is these interactions that effectively promote the formation of stable complexes between small molecules and α7-nAchR (Figure 3).

Compound 7b was well matched with AChE. Strong hydrogen bonds were formed in the O atoms of 7b and the amino acids (TYR-337, TYR-124) of AChE, which assisted in the stabilization of small molecules in protein pockets (Figure 4). Hydrophobic interactions also existed between 7b and amino acids, such as TRP-86, TYR-72 and TRP-286. Furthermore, conjugated interactions of the amino acids PHE-338, PHE-297 and TYR-341 with 7b were created, which are also helpful for stabilizing small molecules.

Concerning compound 7g, it is well matched with AChE. There were hydrogen bonds formed between 7g and the amino acids THR-238 and ASN-233. Hydrophobic interactions were also observed between the phenyl rings of compound 7g and the amino acids PRO-537, PRO-410 and PRO-235. Conjugated interactions were created in 7g with the amino acids ARG-247 and GLU-313 (Figure 5). All of these interactions can effectively promote the formation of a stable complex between 7g and AChE.

Docking analysis revealed that the compounds were more likely to bind with AChE compared with AchR. Increasing hydrogen bonds and hydrophobic interactions with the active site of AChE could easily cause more potent inhibition in AChE. Hydrogen bonds were mainly formed between the O atom of the phosphate esters and the amino acid residues of AChE. Both alkyl and aryl groups can have hydrophobic interactions with the amino acid residues. Therefore, subsequent compounds can be designed to enrich the hydrogen binding sites, such as modifying phosphate esters with -COOH, -OH, -NH₂.

3. Experimental Section

3.1. Chemistry

3.1.1. General Procedures

All solvents and reagents were purchased from commercial sources and used without further purification. TLC analysis was performed on pre-coated, glass-backed silica gel plates and visualized using UV light and KMnO₄. High-resolution mass spectra (HRMS) were obtained via Agilent 1290/6545 UHPLC-QTOF/MS. Tetramethylsilane was used as the internal standard. A Bruker 300M spectrometer was used to record ¹H, ¹³C and ³¹P nuclear magnetic resonance (NMR) spectra in chloroform (CDCl₃) and deuteroxide (D₂O). Melting points were determined using METTLER TOLEDO MP90 Melting Point System.

3.1.2. General Procedure for the Preparation of 2

Compound 2 was synthesized via N-methylation of the Eschweiler–Clarke reaction. Formic acid (46.97 g, 98%, 1 mol) and 2-amino-1,3-propanediol (9.1 g, 0.1 mol) were added to a 250 mL round-bottom flask under ice bath conditions, and then, formaldehyde solution (aq., 37%, 18.4 g, 0.22 mol) was added dropwise to the above system. The reaction was stirred at 60 °C for 3 h until no additional bubbles were generated; then, the temperature was increased to 90 °C to continue the reaction for 6 h. After the reaction was completed, the excess formic acid and water were evaporated under reduced pressure to obtain crude 2-dimethylamino-1,3-propanediol 2 (10.35 g), which was a light-yellow oil.

3.1.3. General Procedure for the Preparation of 3

To a 25 mL round-bottom flask, crude compound 2 (2 g) was added, and SOCl₂ (6.1 mL, 0.1 mol) was dropped slowly into an ice bath, and then, stirred under reflux for 0.5 h. Excess SOCl₂ was removed and 1,3-dichloro-N,N-dimethyl-2-propanamine hydrochloride 3 (2.26 g) was obtained via recrystallization from chloroform.

Data for 3. White solid, m.p.: 93.5 °C, yield 70%. ¹H NMR (300 MHz, D₂O) δ 4.03 (dt, J = 12.4, 5.4 Hz, 5H), 3.00 (s, 6H).¹³C NMR (75 MHz, CDCl₃) δ 64.95 (s), 41.86 (s), 39.01 (s). MS (HRMS-ESI): Calcd for C₅H₁₁Cl₂N, [M+H]⁺: 156.0341, found: 156.0327.

3.1.4. General Procedure for the Preparation of 5

Some phosphite diesters were obtained from commercial sources and others were synthesized according to previously reported procedures [28]. To a solution of the corresponding alcohol (0.3 mol) and pyridine (15.82 g, 0.2 mol) in Et₂O (50 mL) at 0 °C was added PCl₃ (13.74 g, 0.1 mol) over the course of 1 h. After complete addition, the reaction mixture was allowed to slowly warm to ambient temperature, and it was stirred for 16 h. The white suspension was then filtered under suction, and the residual pyridinium chloride was washed twice with Et₂O (50 mL). The filtrates were concentrated under reduced pressure and dried under reduced pressure to yield the desired phosphonates, which were colorless liquids.

3.1.5. General Procedure for the Preparation of 6

Compounds 6a–6h were synthesized based on information from the literature [28]. To a suspension of the appropriate phosphonate (20 mmol) and S₈ (0.704 g, 22 mmol) in Et₂O in a round-bottom flask was slowly added NEt₃ (2.23 g, 22 mmol) in an ice bath. After the full conversion of the phosphonate, as monitored via ³¹P NMR spectroscopy, the suspension was diluted with Et₂O to 100 mL, and then, washed with aqueous HCl (100 mL, 1 M), dried over MgSO₄ and concentrated under reduced pressure. The resulting suspension was filtered to yield S-hydrogen phosphorothioates.

3.1.6. General Procedure for the Preparation of 7a–7h

A mixture of S-hydrogen phosphorothioates 6a–6h (3 mmol) and NaH (60% in mineral oil, 0.12 g, 3 mmol) in dry acetonitrile (10 mL) was stirred at room temperature. A quantity of 1,3-dichloro-N,N-dimethyl-2-propanamine hydrochloride 3 (0.193 g, 1 mmol) was added after 10 min, and the reaction was stirred at 50 °C, as monitored via TLC. After completion, the reaction mixture was filtered and the organic phase concentrated in vacuo. The crude product was purified via silica gel column chromatography to give compounds 7a–7h in 20–78% yields (see Supplementary Materials).

Data for 7a. Yellow oil, yield 26% (silica gel chromatography: ethyl acetate/methanol = 20:1, R_f = 0.16). ³¹P NMR (121 MHz, CDCl₃) δ 31.10 (s). ¹H NMR (300 MHz, CDCl₃) δ 3.79 (d, J = 12.6 Hz, 12H), 3.12–2.83 (m, 5H), 2.31 (s, 6H). ¹³C NMR (75 MHz, CDCl₃) δ 64.68 (t, J = 5.3 Hz), 54.01 (d, J = 6.1 Hz), 40.20 (s), 30.17 (d, J = 3.4 Hz). MS (HRMS-ESI): Calcd for C₉H₂₃NO₆P₂S₂, [M+H]⁺: 368.0515, found: 368.0536.

Data for 7b. Yellow oil, yield 54% (silica gel chromatography: petroleum ether/ethyl acetate = 1:2, R_f = 0.20). ³¹P NMR (121 MHz, CDCl₃) δ 27.66 (s). ¹H NMR (300 MHz, CDCl₃) δ = 4.34–4.04 (m, 8H), 3.27–2.71 (m, 5H), 2.34 (s, 6H), 1.52–1.28 (m, 12H). ¹³C NMR (75 MHz, CDCl₃) δ 64.61 (t, J = 5.7 Hz), 63.72 (d, J = 6.2 Hz), 40.25 (s), 30.24 (d, J = 3.4 Hz), 16.13 (d, J = 7.2 Hz). MS (HRMS-ESI): Calcd for C₁₃H₃₁NO₆P₂S₂, [M+H]⁺: 424.1141, found: 424.1177.

Data for 7c. Yellow oil, yield 74% (silica gel chromatography: petroleum ether/ethyl acetate = 1:6, R_f = 0.18). ¹P NMR (121 MHz, CDCl₃) δ 27.82 (s). ¹H NMR (300 MHz, CDCl₃) δ 4.20–3.94 (m, 8H), 3.18–2.86 (m, 5H), 2.34 (s, 6H), 1.88–1.62 (m, 8H), 0.98 (t, J = 7.4 Hz, 12H). ¹³C NMR (75 MHz, CDCl₃) δ 69.16 (d, J = 5.9 Hz), 64.54 (t, J = 5.9 Hz), 40.27 (s), 30.19 (d, J = 3.3 Hz), 23.58 (d, J = 7.3 Hz), 10.12 (s). MS (HRMS-ESI): Calcd for C₁₇H₃₉NO₆P₂S₂, [M+H]⁺: 480.1767, found: 480.1791.

Data for 7d. Yellow oil, yield 61% (silica gel chromatography: petroleum ether/ethyl acetate = 3:2, R_f = 0.22). ³¹P NMR (121 MHz, CDCl₃) δ 25.23 (s). ¹H NMR (300 MHz, CDCl₃) δ 4.85–4.63 (m, 4H), 3.16–2.85 (m, 5H), 2.34 (s, 6H), 1.37 (t, J = 5.8 Hz, 24H). ¹³C NMR (75 MHz, CDCl₃) δ 72.71 (d, J = 6.5 Hz), 64.37 (t, J = 6.4 Hz), 40.22 (s), 30.41 (d, J = 3.3 Hz), 24.08–23.55 (m). MS (HRMS-ESI): Calcd for C₁₇H₃₉NO₆P₂S₂, [M+H]⁺: 480.1767, found: 480.1796.

Data for 7e. Yellow oil, yield 78% (silica gel chromatography: petroleum ether/ethyl acetate = 1:3, R_f = 0.25). ³¹P NMR (121 MHz, CDCl₃) δ 27.80 (s). ¹H NMR (300 MHz, CDCl₃) δ 4.23–3.95 (m, 8H), 3.21–2.79 (m, 5H), 2.31 (s, 6H), 1.78–1.57 (m, 8H), 1.50–1.31 (m, 8H), 0.92 (t, J = 7.4 Hz, 12H). ¹³C NMR (75 MHz, CDCl₃) δ 67.42 (dd, J = 6.5, 0.9 Hz), 64.54 (t, J = 5.9 Hz), 40.28 (s), 32.19 (d, J = 7.2 Hz), 30.21 (d, J = 3.3 Hz), 18.77 (s), 13.63 (s). MS (HRMS-ESI): Calcd for C₂₁H₄₇NO₆P₂S₂, [M+H]⁺: 536.2393, found: 536.2429.

Data for 7f. Yellow oil, yield 30% (silica gel chromatography: petroleum ether/ethyl acetate = 1:6, R_f = 0.22). ³¹P NMR (121 MHz, CDCl₃) δ 17.31 (s). ¹H NMR (300 MHz, CDCl₃) δ 3.15–2.81 (m, 5H), 2.34 (s, 6H), 1.55 (s, 36H). ¹³C NMR (75 MHz, CDCl₃) δ 84.60 (dd, J = 9.5, 1.5 Hz), 63.90 (t, J = 6.5 Hz), 40.43 (s), 30.85 (d, J = 3.7 Hz), 30.30 (d, J = 4.3 Hz). MS (HRMS-ESI): Calcd for C₂₁H₄₇NO₆P₂S₂, [M+H]⁺: 536.2393, found: 536.2424.

Data for 7g. White oil, yield 26% (silica gel chromatography: petroleum ether/ethyl acetate = 3:1, R_f = 0.32). ³¹P NMR (121 MHz, CDCl₃) δ 21.40 (s). ¹H NMR (300 MHz, CDCl₃) δ 7.43–7.10 (m, 20H), 3.15–2.74 (m, 4H), 2.68–2.44 (m, 1H), 2.13 (s, 6H). ¹³C NMR (75 MHz, CDCl₃) δ 150.25–149.92 (m), 129.92 (d, J = 1.0 Hz), 125.77 (s), 120.76 (dd, J = 7.9, 4.9 Hz), 64.19 (t, J = 5.8 Hz), 40.01 (s), 31.10 (d, J = 3.6 Hz). MS (HRMS-ESI): Calcd for C₂₉H₃₁NO₆P₂S₂, [M+H]⁺: 616.1141, found: 616.1153.

Data for 7h. Yellow oil, yield 20% (silica gel chromatography: petroleum ether/ethyl acetate = 1:2, R_f = 0.15). ³¹P NMR (121 MHz, CDCl₃) δ 28.76 (s). ¹H NMR (300 MHz, CDCl₃) δ 7.35 (d, J = 5.9 Hz, 20H), 5.24–4.99 (m, 8H), 3.05–2.63 (m, 5H), 2.14 (s, 6H). ¹³C NMR (75 MHz, CDCl₃) δ 135.46 (d, J = 7.4 Hz), 128.67 (s), 128.18 (s), 69.08 (d, J = 6.0 Hz), 64.07 (t, J = 6.0 Hz), 40.09 (s), 30.25 (d, J = 3.4 Hz). MS (HRMS-ESI): Calcd for C₃₃H₃₉NO₆P₂S₂, [M+H]⁺: 672.1767, found: 672.1733.

3.2. Biological Tests

3.2.1. Bioassay Methods

All of the bioassays were performed on representative test insects reared in the laboratory. Bioassays were repeated in triplicate. Insecticidal assessments were made on a dead/alive basis, and mortality rates were corrected using Abbott’s formula.

3.2.2. Inhibitory Activity of Acetylcholinesterase (AChE)

In vitro AChE inhibitory activity of the synthesized derivatives was determined spectrophotometrically against AChE using the modified Ellman’s method. The solutions and concentrations involved in the assay were as follows: phosphate buffer (0.1 M, Na₂HPO₄/NaH₂PO₄, pH = 7.4), the enzyme (0.4 μg/mL in phosphate buffer, pH = 7.4), DTNB (5,5-dithiobis (2-nitrobenzoic acid)) (1.5 mmol/L concentration) and ATCh (acetylthiocholine iodide) (2.25 mmol/L concentration). The compounds were dissolved in dimethyl sulfoxide (DMSO, 99%, Macklin, Shanghai, China) and configured as a stock solution at a concentration of 200 mmol/L. The solutions were then added to the buffer. The maximum concentration of DMSO in the assays was 2%, a concentration that had no effect on inhibitory activity in independent experiments without the inhibitor. The absorbance at 412 nm was determined at 37 °C. It was continuously measured for 30 min, and each measurement interval was 1 min. The linear reaction part was taken for the kinetic calculation. The same amount of PB buffer solution replaced the test sample as the blank control. It was considered that the absorbance change was proportional to the enzyme concentrations in principle. The IC₅₀ values were determined via probit analysis using GraphPad Prism 8.3 software (GraphPad Software, San Diego, CA, USA).

The formulas involved in the experiment were as follows:

Inhibitory mortality rate (%) = (B₄₁₂ − T₄₁₂) × 100/B₄₁₂

where B₄₁₂ is the absorbance at 412 nm of the blank control, and T₄₁₂ is the absorbance at 412 nm of the treated group.

K_i = ln[v₀/v_t]/[IN]t

where [IN] is the initial concentration of the tested compound, and v₀ and v_t are the reaction rates at time zero and time t, respectively [29].

3.2.3. Insecticidal Activity against M. separata

The insecticidal activity of the synthesized compounds was evaluated in the laboratory using a topical application method [30]. Fresh, tender corn leaves were cut into squares with a side length of about 0.5 cm. Two drops of the sample solution were added to the leaves with a 1 μL micro dropper, and the operation was repeated for different concentrations of the sample solution. The treated leaves were air dried and then placed into the insect-rearing box. Ten pre-third-instar larvae of M. separata were raised in the insect box, and ten leaves were placed in each box. The test was repeated 4 times, with acetone as the blank control and 97% chlorpyrifos as the positive control. The experiment was carried out at 25 ± 2 °C and a relative humidity (RH) of 65−80%, and over a 12 h/12 h (light/dark) photoperiod. The number of survivors and deaths were recorded after 24 h and 48 h. The insecticidal activity of the tested compounds against the pre-third-instar larvae of M. separata was calculated using the formula:

Corrected mortality rate (%) = (T − C) × 100/(1 − C)

where T is the mortality rate in the treated group expressed as a percentage, and C is the mortality rate in the untreated group expressed as a percentage.

The LC₅₀ values were determined via probit analysis using IBM SPSS Statistics software (IBM, New York, NY, USA).

3.2.4. Insecticidal Activity against M. persicae

The insecticidal activity of the synthesized compounds against M. persicae was evaluated in the laboratory using the slide-dip method [31]. The compounds were dissolved in acetone, and then, diluted in 0.1% Tween-80 aqueous solution to final concentrations ranging from 0.125 to 2 mg/mL; the diluent (0.1% Tween-80) alone served as a blank control, and 98% thiamethoxam was the positive control. Adult aphids were fixed on double-sided tape on slides using a small brush. One end of the slide with the aphids was immersed in the test solutions for 5 s. The excess liquid was removed using absorbent paper, and the mortality rates were calculated at 24 h and 48 h after treatment. Each test was repeated three times with 30 adult aphids each time.

3.2.5. Insecticidal Activity against R. padi

The insecticidal activity of the synthesized compounds against R. padi was evaluated in the laboratory using the leaf-dipping method [32]. The compounds were dissolved in acetone, and then, diluted in 0.1% Tween-80 aqueous solution to final concentrations ranging from 12.5 to 200 μg/mL; the diluent (0.1% Tween-80) alone served as a blank control, and 96% flunicotamid was the positive control. Wheat leaves were cut into 2–5 cm sections (3–5 leaves for each concentration), immersed into different concentrations of sample solution for 3–5 s, and then, taken out. The excess liquid on the leaves was removed using filter paper, and the leaves were air dried. A total of 20–50 aphids were raised on each leaf at room temperature. The test was repeated 4 times for each concentration. The mortality and survival rates at 24 h and 48 h were recorded.

3.3. Molecular Docking

3.3.1. Preparation of Target Proteins and Small-Molecular Structures

The crystal structures of AChE (PDB ID: 6WVO) and a7-nAchR (PDB ID: 7EKP) target proteins came from the protein database (https://www.rcsb.org/ (accessed on 31 May 2023)). After the irrelevant small molecules in the protein molecule were deleted by Pymol 2.1 software (Schrödinger, New York, NY, USA), the protein molecule was guided to AutoDock Tools 1.5.6 software (Scripps Institute, San Diego, CA, USA) to delete water molecules, add hydrogen atoms and set the atomic type. It was saved as a pdbqt file.

The structures of 7a–7h were constructed using ChemDraw (CambridgeSoft, Cambridge, MA, USA). The energies of the downloaded compounds were minimized using Chem3D and converted into mol² format. The small-molecule compounds were imported into AutoDock Tools 1.5.6 software (Scripps Institute, San Diego, CA, USA), their atomic charges were added and their atomic types were assigned. All the flexible keys were rotated by default, and it was finally saved as a pdbqt file.

3.3.2. Analysis of Molecular Docking

The treated compounds were used as small-molecule ligands and two protein targets were used as acceptors. The center position and length, width and height of the Grid Box were determined based on the interaction between small molecules, and the targets were set to 40 Å × 40 Å × 40 Å. AutoDock Vina was used to conduct batch molecular docking, and the results of molecular docking were analyzed. Pymol 2.1 software (Schrödinger, New York, NY, USA) was used to visualize interactions between the compounds and proteins. The final docking structure was evaluated according to the binding free energy.

4. Conclusions

In this study, a class of ring-opened nereistoxin derivatives modified with phosphate esters was synthesized based on S-hydrogen phosphorothioates and 2-dimethylamino-1,3-dichloropropane, and characterized via ¹H, ¹³C and ³¹P NMR and HRMS. The biological activity of the compounds was assessed based on the inhibition of AChE and insecticidal activity of armyworms and aphids. The majority of substances demonstrated potent AChE inhibition in a short time (within 5−30 min). The enzyme inhibitory effect of compounds was closely related to the steric hindrance of the substituted groups on phosphate. The smaller the steric hindrance, the stronger the inhibitory activity of AChE. That is to say, 7a had the strongest enzyme inhibitory effect, with a minimum IC₅₀ value of 3.332 μM. Compounds 7a, 7b, 7c, 7f and 7g showed the most effective insecticidal activity against armyworm among these derivatives, with a 24 h final mortality rate of 100%. Compounds 7b and 7f showed greater insecticidal efficacy against M. persicae and R. padi, with a maximum mortality of 90% after 48 h. These findings imply that pests of the hemiptera and lepidopteran families are both susceptible to the insecticidal effects of phosphonate-modified nereistoxin derivatives. The results of this study may help to expand the use of nereistoxin insecticides and contribute to the development of new insecticide formulations. Additionally, molecular docking was performed to analyze the selectivity of compounds for binding to AchR or AChE. The results demonstrated that the compounds had a higher propensity to bind with AChE. The interactions between the chemicals and AChE were made more effective by hydrogen bonds and hydrophobic interactions.