Synthesis, Fungicidal Activity, and Structure Activity Relationship of β-Acylaminocycloalkylsulfonamides against Botrytis cinerea

In order to discover new antifungal agrochemicals that could have highly active and novel motifs, thirty-six new 2-acylaminocycloalkylsulfonamides (IV) were synthesized. Their structures were characterized and confirmed by 1H NMR, 13C NMR, IR, MS, elemental analysis and X-ray single crystal diffraction. In vitro and in vivo activities against various Botrytis cinerea strains were evaluated. Bioassay results revealed that most of the title compounds exhibited excellent in vitro fungicidal activity, in which compound IV-26 showed the highest activity against sensitive, low-resistant, moderate-resistant and high-resistant strains of B. cinerea compared with the positive fungicide procymidone. Meanwhile in vivo fungicidal activity of compound IV-31 was better than the commercial fungicides procymidone and chesulfamide in greenhouse trial. The structure activity relationship (SAR) was also discussed and the results were of importance to the structural optimization and development of more potent sulfonamides antifungal agents.

which was considered to be responsible for precise codon recognition and the absence of these derivatives led to mitochondrial encephalomyopathic diseases. In addition, taurine deoxyadenosine monophosphates (Tau-dAMP, D, Fig. 1) were recently developed as potential substrates for the HIV-1 reverse transcriptase 20 .
While 2-aminocycloalkylsulfonic acid and its derivatives were rarely reported so far 24,25 , and there are no reports on the synthesis of 2-acylaminocycloalkyl-sulfonamides. Although its application in field of medicine was primarily reported, in agricultural research was still poorly applied. Recently, our group reported a series of 2-oxycycloalkylsulfonamides (H-N, Figs 2 and 3), which possessed highly fungicidal activity [26][27][28][29] , of which compound L (chesulfamide, Fig. 3) could be great promise and a lead compound in fungicide research and development. Based on the lead structure of compound L, compounds M and N (Fig. 3) were designed and synthesized with much higher fungicidal activity [30][31][32] .
These findings encouraged us to further extend the structural modification of compound L with the aim to find more potent antifungal agents. In this paper, 2-acylaminocycloalkylsulfonamides (IV, Fig. 3) were constructed by reaction of reductive amination and acylation (Fig. 4). The single-crystal structure of the title  compounds IV-3 and IV-31 were analyzed. The fungicidal activity of the title compounds against various B. cinerea strains was evaluated. According to their fungicidal activities, structure-activity relationship (SAR) was also discussed.

Results and Discussion
Synthesis and Structure Elucidation. The synthetic route of title compounds IV-1 to IV-36 was outlined in Fig. 4 using 2-oxycycloalkylsulfonamides as a starting material. Reductive amination method in ref. 30 was applied to the treatment of ketones with ammonia in ethanol and titanium (IV) isopropoxide, followed by in situ sodium borohydride reduction. In our experiments, however, the method is improved. For the synthesis of compounds II from compounds I, the ethanol solution of ammonia was replaced directly by continuous passing of ammonia gas. The reaction was completed in a short time by monitoring ammonia gas pressure upto 20 mmHg. It was easy to operate and, and the yield of compounds II were from 42% to 96%. In addition, the title compounds IV were easily obtained by the reaction of compounds II with acyl chloride. Yields of title compounds IV were generally high (over 90%).
Crystal structures of compounds IV-3 and IV-31 were analyzed by X-ray single crystal diffraction. Their structures were shown in Fig. 5, and their crystal data were shown in Table S1 to Table S3. Compound IV-3 was typical chair conformation, in which the chiralities of the 9 th and the 14 th carbon atoms on cyclohexane were R and S respectively. In addition, the bulky sulfonamide group was on equatorial bond and the smaller amide group was on axial bond. The spatial configuration was presented as cis-1, 2-disubstituent. Two benzene rings were far apart, which avoided the steric hindrance effect. For compound IV-31, the chiralities of the 3 th and the 9 th carbon atoms on cycloheptane were S and R respectively. Similarly, that of the two groups was also on the same side of the ring plane and its space conformation was cis-1, 2-disubstituent. Specific optical rotation of compounds IV-3 and IV-31 were tested as − 39.2° and − 0.67° respectively. Compound IV-3 possessed specific optical rotation value of − 39.2° due to the better stability of cyclohexane, while compound IV-31 was unstable in methanol solution, of which two conformations were mutually transformed to a raceme, resulting in the optical activity disappeared. The structures of the synthesized compounds were confirmed by 1 H NMR, 13 C NMR, IR, LC-MS and elemental analysis. Due to the structural similarity, all the compounds showed similar spectroscopic characteristics. In 1 H NMR spectra of compounds II and IV, the protons on the benzene ring appeared in low field in the range of δ H 7.0 to 8.0 ppm, while cycloalkyl group gave signals in the range of δ H 0 to 5.0 ppm appointed to the protons of CH 3 -, CH 2 -and CH-. In addition, active hydrogen atoms of -NH 2 and SO 2 NH-in compounds II appeared around 8.3 ppm, and these two types of hydrogen signals were combined together to represent a broad singlet. The reason may be that active hydrogen of SO 2 NH-is transferred to -NH 2 , forming a structure of -NH 3+ . While that of O = C-NH and SO 2 NH-in compound IV appeared around 8.3 ppm and 9.3 ppm respectively.
Coupling splitting of protons on CH-SO 2 , is very characteristic. Generally, the proton of CH-SO 2 showed doublet of doublet of doublets (ddd), such as compounds IV-5, IV-7, IV-14, IV-21 and IV-23, and corresponding splitting of proton on CH-N was triplet of doublets (td) (Fig. 6). While in the spectra of some compounds, such as  compound IV-3, the proton of CH-SO 2 showed doublet of doublets (dd), instead of ddd, and corresponding splitting of proton on CH-N was doublet of triplets (dt) (Fig. 7). However, there were some compounds which coupling splitting of protons on CH-SO 2 were special because the conformation was dynamic, such as compounds IV-13, IV-15, IV-16, IV-30 and IV-32 and so on, the protons of CH-SO 2 and CH-N showed different signal peak type. This phenomenon was very interesting, but the reason was still unknown. We choose the dominant conformation to explain the normal splitting characteristic of CH-SO 2 .
Taking the 1 H NMR spectra of compound IV-3 as an example (Fig. 7), according to the crystal structure, the conformation diagram (Fig. 8) of compound IV-3 was drawn to explain the reasons for this splitting characteristic. As shown in Fig. 8, the proton of C 14 linked -SO 2 was located in axial bond (C 14 -H a ) due to coupling splitting  effects of equatorial bond (C9-H e ) on C 9 , equatorial bond (C 13 -H e ) and axial bond (C 13 -H a ) on C 13 . Normally the proton of C 14 linked -SO 2 showed ddd signal due to the magnetic non-equivalence of these three protons (C 9 -H e , C 13 -H e and C 13 -H a ), appeared as dd signal. Its spatial conformation was shown in Fig. 8(a), which can be determined from the single crystal structure in Fig. 5. It showed a strong coupling splitting effect due to C 14 -H a , C 9 -H e and C 13 -H e lying on one side of cyclohexane plane and at a close distance, while it showed a weak coupling splitting effect due to C 13 -H a and C 14 -Ha lying on both sides of cyclohexane plane and at a distant position, which led to its split signal invisible in the spectrum. Therefore, in 1 H NMR spectra C 14 -H a showed double doublets.
Correspondingly, the proton of C 9 -H e linked -NH showed double triplets. The reason can be explained by its spatial conformation. As shown in Fig. 8(b), the difference of magnetic non-equivalence of two protons between C 10 -H e and C 10 -H a is small due to the protons adjacent C 10 -H e and C 10 -H a close to proton of C 9 -H e . So proton of C 9 -H e affected by protons of C 10 -H e and C 10 -H a , which signal showed coupling splitting of triplets, and protons of C 9 -H e affected by protons of C 14 -H e , which signal showed coupling splitting of doublets. Therefore, in 1 H NMR spectra double triplets were assigned to the proton of C9-H e .
In 13 C NMR spectra (see supplementary information), compounds I, II and IV revealed signals of carbon in the range of δ C 0 to 70 ppm assigned to methyl, methylene and methane on naphthene, and carbon signals of benzene ring and trifluoromethyl in the range of δ C 115 to 140 ppm in low field. Compounds I and IV gave carbon signals around 202 ppm and 166 ppm respectively assigned to C = O.
In IR spectra of compounds I and IV, the absorption peak of carbonyl stretching vibration appeared around 1700 cm −1 and 1650 cm −1 , respectively. While the absorption peak of imino group stretching vibration appeared around 3300 cm −1 . In addition, the stretching vibration absorption (-NH 2 and -SO 2 NH) of compounds II appeared around 3500 cm −1 and 3150 cm −1 .
In LC-MS (ES + mode) spectrum of IV-1 (Fig. 9), the quasi-molecular ion peak was 491 [M + H] + , which accorded with the nitrogen rule. Firstly, sulfonamide bond was broken into a characteristic ion peak at m/z 296, and then fragment ion peak of m/z 135 was obtained by amide bond fracture. Finally, fragment ion peaks of m/z 92 and m/z 77 were obtained via McLafferty rearrangement on the benzene ring and after losing a methylene, respectively. According to the above analysis, fragment missing was reasonable.
Bioassay of Fungicidal Activities. B. cinerea strains showed multiple physiological characteristics because of the different living environment and fungicide application level. As a result, sensitivities of strains from different areas are also disparate to new compounds.
Fungicidal activity and structure-activity relationship of compounds IV-1~IV- 29. In order to screen out active compounds correctly and quickly, firstly the title compounds (IV-1~IV-29) were tested against two B cinerea strains (Dd-15 and Sy-10), which inhibition rates were shown in Fig. 10. Two-factor analysis of variance between strains and compounds was conducted by SPSS20.0. Analytical results showed that there were  sensitivity differences in the twenty-nine new compounds against the two strains. For example, the activity of the title compounds against Dd-15 was generally high, and average inhibition rate was about 76.0%. While the activity was relatively low against Sy-10, which average inhibition rate was about 58.4%. According to the analysis of bioactivity against two strains, there were twenty-one compounds, which fungicidal activities were higher than that of the positive control procymidone.
The preliminary structure-activity relationship can be summarized in four points. First, for substituent benzoyl chloride (IV-1~IV-17), fungicidal activity was mediocre on the benzene ring containing two substituents, and substituted phenyl groups at ortho-and para-position with methoxyl group and fluorine atom showed excellent activity. However, fungicidal activity was higher at meta-substituted methyl group and meta-substituted chlorine atom. When trifluoromethyl group was on the benzene ring such as compounds IV-15 and IV-16, fungicidal activity was the highest. Second, for alkylacyl chloride (IV-18~IV-23), fungicidal activity of compounds showed a rising trend with the increase of carbon number in the alkyl group, for example, that of which containing n-hexanoyl chloride (IV-22) or n-heptanoyl chloride (IV-23) was the highest. Third, for halogenated acetyl chloride (IV-24~IV-27), the bioactivity increased with the increase of chlorine atom number, and activity of chloro-substituted compounds was higher than that of the bromo-substituted ones. Finally, for 2-alkoxyl acetyl chloride (IV-28 and IV-29), the activity of 2-ethoxyl acetyl chloride was higher than that of the 2-methoxyl acetyl chloride. As a result, eleven highly active compounds were chosen as candidates in the second round screening.
As shown in Fig. 11, eleven compounds were screened out to determine fungicidal activity against other six different B. cinerea strains. Two-factor analysis of variance results showed that there remained significant differences in sensitivities of the six B. cinerea strains to the title compounds. For example, the average inhibition rates of eleven compounds against As-12, Cy-07, Dd-04, Dl-17, Fs-06 and Hld-16 were 51.97%, 29.17%, 62.82%, 55.84%, 35.70% and 47.74% respectively. The activities of eleven compounds against the six B. cinerea strains could be divided into eight subsets (a-h), in which those of compounds IV-23, IV-24, IV-26 and IV-29 were higher than the positive control procymidone. These four compounds were selected to do the further study and their EC 50 values were evaluated and shown in Table 1.
Fungicidal activities of all the four title compounds were higher than that of the positive fungicide procymidone. Overall, the fungicidal activities of compound IV-26 against six strains (As-12, Cy-07, Dd-04, Dl-17, Fs-06, and Hld-16), IV-29 against four strains (As-12, Cy-07, Dd-04, and Fs-06), IV-23 and IV-24 against three strains (As-12, Cy-07, and Fs-06) were higher than those of procymidone. Activities of the four title compounds against the six B. cinerea strains were different, for example, EC 50 values of compound IV-26 against the six strains were 0.37~7.56 mg/L, while those of procymidone were 2.49~75.84 mg/L. Referring to resistant grading standards to procymidone 33,34 , Dd-04 and Dl-17 were low-resistant strains; As-12, Cy-07 and Hld-16 were moderate-resistant strains; Fs-06 was high-resistant strain. The results displayed that the in vitro activities of compound IV-26 against all the resistant strains were excellent.
After the above test, structure-activity relationship between acyl chloride and fungicidal activity was confirmed. It was to be sure that trichloroacetyl chloride had the greatest contribution to the fungicidal activity for the twenty-nine acyl chlorides. Therefore, trichloroacetyl chloride was marked as the active group in the later structural modification of cycloalkyl group (IV-30~IV-36). Moreover, compared to the fungicidal activity screened for one strain, it was more reliable for different strains from different areas to choose as the test targets.  Table 2, after the structural modification of cycloalkyl group, compounds IV-30~IV-36 also had very high fungicidal activity against five other B. cinerea strains. Referring to resistant grading standards to procymidone 33,34 , As-11 was sensitive strain; Dl-11 was low-resistant strain; Cy-09 was moderate-resistant strain; Fs-11 and Hld-15 were high-resistant strains. The statistical results of SPSS showed that compounds IV-26, IV-30, IV-31, IV-32, IV-33 and IV-34 exhibited excellent activity. It was found that the size of cycloalkyl group was important factor to determine the fungicidal activity compared with that of IV-26. For example, the EC 50 values of compounds IV-26, IV-30 and IV-31 were 0.15~3.64 mg/L, 0.66~11.68 mg/L and 0.82~9.49 mg/L, respectively. The activities of compound IV-26 containing 6-membered ring were better than those of compounds IV-30 and IV-31 respectively containing 5-and 7-membered ring. In addition, it was found that the types of substituent alkyl group on the cyclohexane had a significant effect on the fungicidal activity by comparing activities of compounds IV-32~IV-36. The activity decreased with the increase of alkyl carbon number. Moreover, the position of alkyl group also had effect on the activity. For example, compared with compounds IV-32, IV-33 and IV-34, their fungicidal activity was para-methyl > ortho-methyl > meta-methyl in the order.
In vivo fungicidal activity against B. cinerea on leaves of cucumber (mycelium inoculation method). Six compounds (IV-25, IV-26, IV-30, IV-31, IV-32, and IV-33) were tested for their in vivo fungicidal activity on leaves of cucumber, and the leading compound chesulfamide (L, Fig. 3) was used as the positive control. The bioassay results in Table 3 showed that the control efficiency of compound IV-31 was significantly higher than that of the positive control chesulfamide. Fungicidal activity of compounds IV-26, IV-30, IV-32 and IV-33 was equivalent to the chesulfamide.
Compared with the previous work, the structure of the title compounds was modified and new. Meanwhile, their fungicidal activity had greater improvement than that of the lead compound. From the point of view of chemical synthesis, novel key intermediates 2-aminocycloalkylsulfonamides (II-1~II-8) were obtained, which had the vital significance for obtaining the title molecules with structural diversity. In addition, the effective improvements of synthesis method for compounds II were made, which greatly increased the yield and the reaction progress. On the other hand, structural characterization of title compounds IV was described in detail. In particular, the NMR spectra are very characteristic. The single crystal structure was obtained, which provided the  basis for accurately structural analysis. According to the single crystal structure, the computer-aided design could be simulated, which is helpful to further molecular design and structural optimization. Preliminary mechnism study indicated that cyclohexyl alkyl sulfonamides might inhibit the growth of gray mould by affecting the synthesis of the internal substance 35 . The elucidation of the mode of action of these new compound is worth research, which will be studied in detail in the future.

Conclusion
In conclusion, we reported the synthesis of a new series of 2-acylaminocycloalkylsulfonamides and their in vitro and in vivo fungicidal activities against various B. cinerea strains were evaluated. Some title compounds showed notable activity, especially compound IV-31 was of great potential to be developed as new antifungal agents for plant protection. Moreover, single crystal structure of compound IV-31 was determined to assist the further molecular design and structural modification. In addition, the SAR results indicated that structure of acylchloride and naphthenic scaffold had significant effects on the activity. Thus, the present results were of great promise for the design and development of novel sulfonamides antifungal agents. Further research was necessary on the more extensive structural modification and the broad determination of the fungicidal spectra.

Materials and Methods
General. Nuclear magnetic resonance (NMR) spectra were recorded in CDCl 3 and DMSO-d 6 unless indicated otherwise with a Bruker Avance III 600 MHz spectrometer (Bruker, Fallanden, Switzerland), using tetramethylsilane (TMS) as an internal standard. Infrared (IR) spectra were recorded on a Shimadzu IR Affinity-   Synthesis. The synthetic routes of the key intermediates II and title compounds IV were outlined in Fig. 4.

Synthesis of the key intermediates N -(2-trifluoromethyl-4-chlorophenyl)-2-aminocycloalkylsulfonamides II-1~II-8. The synthetic route of compounds II-1 to II-8 was outlined
in Fig. 4, according to the method given in the ref. 36, under a nitrogen atmosphere, compounds I (30 mmol) and titanium (IV) isopropoxide (17 mL, 60 mmol) in dry ethyl alcohol (150 mL) were stirred, while the ammonia gas passed through the reaction mixture and maintained the pressure of ammonia upto 20 mmHg at room temperature for 6 h, which was monitored by TLC analysis. Then sodium borohydride (1.7 g, 45 mmol) was added slowly to the resulting mixture at room temperature and stirred for 3 h. The reaction was quenched by addition of ammonium hydroxide solution (2 M, 120 mL). The resulting inorganic precipitate was filtered off, and washed with ethyl acetate (150 mL). The filtrate was concentrated under reduced pressure to remove ethyl acetate, and then extracted with ethyl acetate (200 mL). The combined organic extracts were washed with brine (300 mL), dried over anhydrous Na 2 SO 4 , evaporated under reduced pressure, and recrystallized from methanol to afford pure key intermediates II. Their physical and spectra data were shown as follows.  N-(2-trifluoromethyl-4-chlorophenyl)-2-aminocyclopentylsulfonamide (II-2). ( N-(2-trifluoromethyl-4-chlorophenyl)-3-methyl-2-aminocyclohexylsulfonamide (II-4). (

N-(2-trifluoromethyl-4-chlorophenyl)-2-(propionylamino) cyclohexylsulfonamide (IV-19)
. ( Bioassays. In vitro fungicidal activity. In vitro effects of compounds against B. cinerea were evaluated by mycelium growth rate method [30][31][32] . The tested compounds were dissolved in DMSO (dimethyl sulfoxide) and mixed with sterile molten potato dextrose agar (PDA) to a final concentration of 50 mg/L. EC 50 values were estimated using logit analysis. The concentration gradients were 50, 12.5, 3.13, 0.78 mg/L on PDA and a commercial fungicide procymidone was used as the positive control. EXCEL 2010 was used to analyze bioassay data. The variance analysis was carried out by using SPSS 20.0 software for the inhibition rate, EC 50 and control efficiency. The relative inhibition rate of the synthetic compounds compared to blank control was calculated via the following equation (1): In which, I stands for the rate of inhibition (%), C is the diameter of mycelia in the blank control test (in mm), and T is the diameter of mycelia in the presence of tested compounds (in mm).
In vivo antifungal activity. In vivo effects were checked on leaves of cucumber (Cucumis sarivus L.) by mycelium inoculation method with pot cultural test in greenhouse [37][38][39][40] . The cucumber seedlings at 2-3 leaf stages were used to assay the fungicidal activity against B. cinerea. The compounds were confected to 2.5% EC (emulsifiable concentrate) formulation. The formulation was diluted to 500 mg/L with water and sprayed on the surface of the cucumber leaves. After air drying, the surface of the leaves was inoculated with 6 mm plugs of B. cinerea, which was maintained on potato dextrose agar (PDA). This procedure was repeated three times, and nine replicates were performed per treatment. The chesulfamide (L, Fig. 3) was used as the positive control.
The fungicidal activity was assessed when the untreated cucumber plant (blank control) fully developed symptoms. The area of inoculated leaves covered by disease symptoms was evaluated and compared to that of untreated ones to determine the average disease index. The relative control efficacy of compounds compared to the blank assay was calculated via the following equation (2) where I is relative control efficacy, CK is the average disease index during the blank assay and PT is the average disease index after treatment during testing.