Alkyl 4-Aryl-6-amino-7- phenyl-3-(phenylimino)-4,7-dihydro- 3H-[1,2]dithiolo[3,4-b]pyridine-5-carboxylates: Synthesis and Agrochemical Studies

The reaction between dithiomalondianilide (N,N’-diphenyldithiomalondiamide) and alkyl 3-aryl-2-cyanoacrylates in the presence of morpholine in the air atmosphere leads to the formation of alkyl 6-amino-4-aryl-7-phenyl-3-(phenylimino)-4,7-dihydro-3H-[1,2]dithiolo[3,4-b]- pyridine-5-carboxylates in 37–72% yields. The same compounds were prepared in 23–65% yields by ternary condensation of aromatic aldehydes, ethyl(methyl) cyanoacetate and dithiomalondianilide. The reaction mechanism is discussed. The structure of ethyl 6-amino-4-(4-methoxyphenyl)-7-phenyl-3-(phenylimino)-4,7-dihydro-3H-[1,2]dithiolo[3,4-b]pyridine-5-carboxylate was confirmed by X-ray crystallography. Two of the prepared compounds showed a moderate growth-stimulating effect on sunflower seedlings. Three of the new compounds were recognized as strong herbicide safeners with respect to herbicide 2,4-D in the laboratory and field experiments on sunflower.

This work presents the results of our studies of the reactions of N,N'-diphenyldithiomalondiamide 12 with 2-cyanoacrylates derived from cyanoacetic esters.

Synthesis
Earlier, we proposed a cascade method for the preparation of [1,2]dithiolo [3,4-b] pyridines 13 (Scheme 1) from N,N'-diphenyldithiomalondiamide 12 based on the morpholinecatalyzed Michael addition with arylmethylene malononitriles followed by heterocyclization and further air oxidation of 3-thiocarbamoylpyridine-2-thiolate intermediate [16]. As the continuation of this research, we have decided to look into the possibility to using other electron-deficient alkenes as Michael acceptors to prepare dithiolopyridines and we focused our attention on 3-aryl-2-cyanoacrylates 14. The precursors 14 were prepared by Knoevenagel condensation of cyanoacetic esters with aromatic aldehydes in the presence of piperidine or morpholine.
This work presents the results of our studies of the reactions of N,N'-diphenyldithiomalondiamide 12 with 2-cyanoacrylates derived from cyanoacetic esters.

Synthesis
Earlier, we proposed a cascade method for the preparation of [1,2]dithiolo [3,4-b]pyridines 13 (Scheme 1) from N,N'-diphenyldithiomalondiamide 12 based on the morpholine-catalyzed Michael addition with arylmethylene malononitriles followed by heterocyclization and further air oxidation of 3-thiocarbamoylpyridine-2-thiolate intermediate [16]. As the continuation of this research, we have decided to look into the possibility to using other electron-deficient alkenes as Michael acceptors to prepare dithiolopyridines and we focused our attention on 3-aryl-2-cyanoacrylates 14. The precursors 14 were prepared by Knoevenagel condensation of cyanoacetic esters with aromatic aldehydes in the presence of piperidine or morpholine.
At this point, some efforts to optimize the reaction conditions and examine the scope and limitation of the reaction were made. The reaction proceeded smoothly in MeOH or EtOH; other tested solvents (i-PrOH, acetone, DMF) gave rather unsatisfactory results. The nature of amine (morpholine, piperidine, triethylamine) does not significantly affect the yields of products. Along with Michael addition of 12 to other 2-cyanoacrylates 14a-g (Method A, Scheme 4), we also investigated the three-component reaction of dithiomalondianilide 12, cyanoacetic esters and aldehydes in the presence of morpholine (Method B, Scheme 4). However, Method B gives somewhat lower yields of dithiolopyridines 15. First, when a mixture of N,N'-diphenyldithiomalondiamide 12 and (E)-ethy ano-3-(4-methoxyphenyl)acrylate 14a in EtOH were treated with excessive morpho yellow crystal of ethyl 6-amino-4-(4-methoxyphenyl)-7-phenyl-3-(phenylimino) hydro-3H- [1,2]dithiolo [3,4-b]pyridine-5-carboxylate 15a was isolated in 72% (Scheme 3). The structure of 15a was unambiguously confirmed by single crysta diffraction analysis (CCDC # 2219352, Figure 1).   At this point, some efforts to optimize the reaction conditions and examine the scope and limitation of the reaction were made. The reaction proceeded smoothly in MeOH or EtOH; other tested solvents (i-PrOH, acetone, DMF) gave rather unsatisfactory results. The nature of amine (morpholine, piperidine, triethylamine) does not significantly affect the yields of products. Along with Michael addition of 12 to other 2-cyanoacrylates 14a-g (Method A, Scheme 4), we also investigated the three-component reaction of dithiomalondianilide 12, cyanoacetic esters and aldehydes in the presence of morpholine (Method B, Scheme 4). However, Method B gives somewhat lower yields of dithiolopyridines 15.
oxidizing agents (hydrogen peroxide or iodine) leads to resinification of the reaction mixture. Aldehydes with both electron-donor and electron-withdrawing substituents are reacted well. However, we failed to obtain any products in the case of furfural or 2-cyano-3-(furan-2-yl)acrylates even if less nucleophilic and milder base triethylamine was taken instead of morpholine. This is probably due to the strong tendency of furan-ring-bearing electron-withdrawing substituents to undergo a nucleophilic attack at the C-5 position to form a complex mixture of furan-ring-cleavage/recyclization products.  Extended refluxing of the reaction mixture does not favor the formation of dithiolopyridines 15. Thus, when a mixture of 14a and 12 in ethanol was heated under reflux in the presence of morpholine for as long as 15 h, a contaminated deep-brown material was obtained from which dithiolopyridine 15a was isolated by recrystallization in only 13% yield. Less prolonged heating (1.5 h) was also accompanied by side processes and led to decreased yields; in this case, compound 15a was obtained in 30% yield.

12
When the reaction of 12 with 14a was conducted in inert atmosphere under nitrogen stream, dithiolopyridine 15a was not isolated. This fact proves the crucial role of air oxygen for the final step of dithiolopyridine system formation. Nevertheless, the addition of oxidizing agents (hydrogen peroxide or iodine) leads to resinification of the reaction mixture. Aldehydes with both electron-donor and electron-withdrawing substituents are reacted well. However, we failed to obtain any products in the case of furfural or 2-cyano-3-(furan-2-yl)acrylates even if less nucleophilic and milder base triethylamine was taken instead of morpholine. This is probably due to the strong tendency of furan-ring-bearing electron-withdrawing substituents to undergo a nucleophilic attack at the C-5 position to form a complex mixture of furan-ring-cleavage/recyclization products.
In the IR spectra of compounds 15, the absorption bands corresponding to NH 2 valence vibrations (ν 3371-3474 cm −1 and ν 3263-3290 cm −1 ), RO(C=O) group bands (ν 1651-1661 cm −1 ) and imino group C=N-Ph bands at ν 1622-1638 cm −1 were observed while the bands corresponding to C≡N groups were absent. 1 H NMR spectra of 15 revealed characteristic singlets attributed to H-4 protons at δ 5.02-5.94 ppm and a very broadened peak of NH 2 protons at δ 7.00-7.19 ppm. It is interesting that in some cases the signal of OCH 2 protons is detected not as a typical quartet but as a complex multiplet, probably due to the hindered rotation caused by the intramolecular hydrogen bond C=O...H-NH. Alternatively, the observed splitting of OCH 2 signal can be caused by the shielding effect of the aromatic ring at C-4, which affects one of OCH 2 protons. 13

Agrochemical Studies
The new compounds were tested as herbicide safeners with respect to 2,4-dichlorophenoxyacetic acid (2,4-D) and as plant growth regulators. 2,4-D is an herbicide that is widely used for plant protection and was reported to show no significant toxicity to humans [38]. However, the use of 2,4-D has negative side effects, including its inhibition effect on the crops themselves that gives a decrease in yield by~15-60%. To eliminate such negative effects and to raise crop yields, herbicide safeners (also called herbicide antidotes or detoxifiers) are successfully used. Herbicide safeners [39][40][41] can be defined as agrochemicals that are able to neutralize phytotoxins in plants, thus protecting crop plants from herbicide injury. Safeners are harmless to crop plants (or even have a growth-stimulating effect), but do not affect the activity of herbicides against weeds.
It is known that 3-aminothieno[2,3-b]pyridines, which can be considered as structural analogs of the prepared 3-imino-3H- [1,2]dithiolo [3,4-b]pyridines 15, are reported to be effective herbicide safeners [42,43] and plant growth regulators [44]. We studied the efficiency of new 3-imino-3H- [1,2]dithiolo [3,4-b]pyridines as 2,4-D antidotes using sunflower seedlings using the reported procedure [42] (see also Materials and Methods). The antidote effect A was determined as a ratio of the hypocotyl (or root) length of sunflower seedlings in the "herbicide + antidote" experiments to the length in the reference group (where the seedlings were treated with 2,4-D only) (Equation (1)): where L exp is an organ length (mm) in the group of seedlings treated with herbicide and antidote, and L ref is an organ length (mm) in the reference group of sunflower seedlings. We found that three of the new compounds, dithiolopyridines 15a,c,f, exhibited a strong 2,4-D antidote effect in the laboratory experiments ( Table 1).
The antidote activity of 3H- [1,2]dithiolo [3,4-b]pyridines 15a,c,f was also studied in field experiments on sunflower in the experimental field of the Federal Scientific Center for Biological Protection of Plants (Krasnodar, Russia). Sunflower plants of cv. Master in the phase of 10-16 leaves were treated with an aqueous solution of 2,4-dichlorophenoxyacetic acid at a dose of 18 g/ha and 3 days later a safener solution was applied at a dose of 100 g/ha with the working fluid rate of 300 L/ha. (where the seedlings were treated with 2,4-D only) (Equation (1)): where Lexp is an organ length (mm) in the group of seedlings treated with herbicide and antidote, and Lref is an organ length (mm) in the reference group of sunflower seedlings. We found that three of the new compounds, dithiolopyridines 15a,c,f, exhibited a strong 2,4-D antidote effect in the laboratory experiments (Table 1). As we can see from the Table 1 Experiments were conducted in plots of 2.8 m 2 with five-fold repetition. Sunflower harvesting was performed at the time of full seed maturity. The field antidote effect AF was determined by the absolute value of the crop yield increase to the herbicide reference by the Equation (2): where AF is antidote effect, %; Y1 is crop yield in "herbicide + antidote" group; and Y2 is crop yield in "herbicide" (reference) group. The obtained data were statistically processed using Student's t-test. The field test results are presented in Table 2. As it can be seen, the use of compounds 15a,c,f as herbicide safeners under field conditions provides an antidote effect in the range of 41.4-51.4%. As we can see from the Table 1 Experiments were conducted in plots of 2.8 m 2 with five-fold repetition. Sunflower harvesting was performed at the time of full seed maturity. The field antidote effect AF was determined by the absolute value of the crop yield increase to the herbicide reference by the Equation (2): where AF is antidote effect, %; Y1 is crop yield in "herbicide + antidote" group; and Y2 is crop yield in "herbicide" (reference) group. The obtained data were statistically processed using Student's t-test. The field test results are presented in Table 2. As it can be seen, the use of compounds 15a,c,f as herbicide safeners under field conditions provides an antidote effect in the range of 41.4-51.4%. The field tests included the following variants: -Control group-untreated plants; -"Herbicide" (reference) group-plants treated with herbicide 2,4-D only; -"Herbicide + antidote" group-plants treated with herbicide 2,4-D and an antidote.
Experiments were conducted in plots of 2.8 m 2 with five-fold repetition. Sunflower harvesting was performed at the time of full seed maturity. The field antidote effect A F was determined by the absolute value of the crop yield increase to the herbicide reference by the Equation (2): where A F is antidote effect, %; Y 1 is crop yield in "herbicide + antidote" group; and Y 2 is crop yield in "herbicide" (reference) group. The obtained data were statistically processed using Student's t-test. The field test results are presented in Table 2. As it can be seen, the use of compounds 15a,c,f as herbicide safeners under field conditions provides an antidote effect in the range of 41.4-51.4%.
The growth-stimulating activity of compounds 15a,c,f was evaluated in laboratory experiments using the known procedure [45] on cv. Master sunflower seedlings ( Table 3). The effect was evaluated by the elongation of stems and roots of treated seedlings in comparison to control (untreated seeds) (Equation (3)): where E is growth-stimulating effect, %; L treated is the length (mm) of stems/roots in the treated group of seedlings; and L control is the length (mm) of stems/roots in the control (untreated) group of seedlings. As we can see from the Table 3, compounds 15a and 15f are more active than 15c and favored stem elongation by 12-20% relative to control and stimulated root growth by 13-22%, depending on the concentration. In general, the growth-stimulating effect of compounds 15a and 15f can be considered as moderate.
The study of the acute toxicity of the new compounds is in progress. According to preliminary data, the compounds do not possess obvious phytotoxicity; this is indirectly indicated by the observed plant growth-stimulating effect of the tested samples (Table 3).
where AF is antidote effect, %; Y1 is crop yield in "herbicide + antidote" group; and Y2 is crop yield in "herbicide" (reference) group. The obtained data were statistically processed using Student's t-test. The field test results are presented in Table 2. As it can be seen, the use of compounds 15a,c,f as herbicide safeners under field conditions provides an antidote effect in the range of 41.4-51.4%. The growth-stimulating activity of compounds 15a,c,f was evaluated in laboratory experiments using the known procedure [45] on cv. Master sunflower seedlings ( Table 3). The effect was evaluated by the elongation of stems and roots of treated seedlings in comparison to control (untreated seeds) (Equation (3)): where E is growth-stimulating effect, %; Ltreated is the length (mm) of stems/roots in the treated group of seedlings; and Lcontrol is the length (mm) of stems/roots in the control (untreated) group of seedlings. As we can see from the Table 3, compounds 15a and 15f are more active than 15c and favored stem elongation by 12-20% relative to control and stimulated root growth by 13-22%, depending on the concentration. In general, the growth-stimulating effect of compounds 15a and 15f can be considered as moderate. The growth-stimulating activity of compounds 15a,c,f was evaluated in laboratory experiments using the known procedure [45] on cv. Master sunflower seedlings ( Table 3). The effect was evaluated by the elongation of stems and roots of treated seedlings in comparison to control (untreated seeds) (Equation (3)): where E is growth-stimulating effect, %; Ltreated is the length (mm) of stems/roots in the treated group of seedlings; and Lcontrol is the length (mm) of stems/roots in the control (untreated) group of seedlings. As we can see from the Table 3, compounds 15a and 15f are more active than 15c and favored stem elongation by 12-20% relative to control and stimulated root growth by 13-22%, depending on the concentration. In general, the growth-stimulating effect of compounds 15a and 15f can be considered as moderate.  The growth-stimulating activity of compounds 15a,c,f was evaluated in laboratory experiments using the known procedure [45] on cv. Master sunflower seedlings ( Table 3). The effect was evaluated by the elongation of stems and roots of treated seedlings in comparison to control (untreated seeds) (Equation (3)): where E is growth-stimulating effect, %; Ltreated is the length (mm) of stems/roots in the treated group of seedlings; and Lcontrol is the length (mm) of stems/roots in the control (untreated) group of seedlings. As we can see from the Table 3, compounds 15a and 15f are more active than 15c and favored stem elongation by 12-20% relative to control and stimulated root growth by 13-22%, depending on the concentration. In general, the growth-stimulating effect of compounds 15a and 15f can be considered as moderate. The study of the acute toxicity of the new compounds is in progress. According to preliminary data, the compounds do not possess obvious phytotoxicity; this is indirectly indicated by the observed plant growth-stimulating effect of the tested samples (Table 3).

Materials and Methods
1 H and 13 C DEPTQ NMR spectra and 2D NMR experiments were recorded in solutions of DMSO-d6 on a Bruker AVANCE-III HD instrument (Bruker BioSpin AG, Fällanden, Switzerland) (at 400.40 or 100.61 MHz, respectively). Residual solvent signals were used as internal standards in DMSO-d6-2.49 ppm for 1 H, and 39.50 ppm for 13 C nuclei. Single crystal X-ray diffraction analysis of compound 15a was performed on an The growth-stimulating activity of compounds 15a,c,f was evaluated in laboratory experiments using the known procedure [45] on cv. Master sunflower seedlings ( Table 3). The effect was evaluated by the elongation of stems and roots of treated seedlings in comparison to control (untreated seeds) (Equation (3)): where E is growth-stimulating effect, %; Ltreated is the length (mm) of stems/roots in the treated group of seedlings; and Lcontrol is the length (mm) of stems/roots in the control (untreated) group of seedlings. As we can see from the Table 3, compounds 15a and 15f are more active than 15c and favored stem elongation by 12-20% relative to control and stimulated root growth by 13-22%, depending on the concentration. In general, the growth-stimulating effect of compounds 15a and 15f can be considered as moderate. The study of the acute toxicity of the new compounds is in progress. According to preliminary data, the compounds do not possess obvious phytotoxicity; this is indirectly indicated by the observed plant growth-stimulating effect of the tested samples (Table 3). The growth-stimulating activity of compounds 15a,c,f was evaluated in laboratory experiments using the known procedure [45] on cv. Master sunflower seedlings ( Table 3). The effect was evaluated by the elongation of stems and roots of treated seedlings in comparison to control (untreated seeds) (Equation (3)):

Materials and Methods
where E is growth-stimulating effect, %; Ltreated is the length (mm) of stems/roots in the treated group of seedlings; and Lcontrol is the length (mm) of stems/roots in the control (untreated) group of seedlings. As we can see from the Table 3, compounds 15a and 15f are more active than 15c and favored stem elongation by 12-20% relative to control and stimulated root growth by 13-22%, depending on the concentration. In general, the growth-stimulating effect of compounds 15a and 15f can be considered as moderate. The study of the acute toxicity of the new compounds is in progress. According to preliminary data, the compounds do not possess obvious phytotoxicity; this is indirectly indicated by the observed plant growth-stimulating effect of the tested samples (Table 3).

Materials and Methods
Daltonics, Bremen, Germany) equipped with an electrospray ionization source in positive ion detection mode. The voltage at the ionization source was 3.5 kV, the drying gas flow rate was 8 L/min, the spray gas pressure was 2 bar, the temperature of the ionization source was 250 • C, the mass scanning range (m/z) was 50-1000, the scanning speed was 3 Hz. The data were processed using Bruker Data Analysis 4.1 software. See Supplementary Materials File for NMR, FTIR and HRMS spectral charts and X-ray analysis data.
General procedure for the preparation of dithiolopyridines 15 (Method A). A vial was charged, under air, with dithiomalondianilide 12 (300 mg, 1.047 mmol), 3-aryl-2cyanoacrylate 14 (1.1 mmol) and EtOH (6-8 mL). A mixture was treated with morpholine (0.14 mL, 1.57 mmol) at 25 • C. Complete dissolution of starting materials and formation of a deep-yellow solution occurred for a very short time (a matter of minutes). The solution was stirred for 3 h and left to stand to allow slow evaporation in air at ambient temperature. After evaporating the solvent, the resulting tarry residue was triturated with an appropriate solvent (usually acetone : EtOH (1:1) or n-BuOH were used). The yellow or light-brown crystalline solid was filtered off, washed with EtOH and hexane and recrystallized from acetone (if appropriate) to give pure dithiolopyridines 15. General

Herbicide-Safening Effect Studies
Germinated sunflower seeds (cv. Master) with 2-4 mm long embryo roots were placed in a solution of 2,4-D (10 -3 % by weight) for 1 h to achieve 40-60% inhibition of hypocotyl growth. After treatment, the seedlings were washed with pure water and placed into a solution of the corresponding compound 15a,c,f (concentrations 10 -2 , 10 -3 , 10 -4 or 10 -5 % by weight, "herbicide + antidote" experiments). After 1 h the seedlings were washed with pure water and placed on paper strips (10 × 75 cm, 20 seeds per strip). The strips were rolled and placed into beakers with water (50 cm 3 ). The reference group of seedlings ("herbicide" experiments) was kept in 2,4-D solution (10 -3 %) for 1 h and then in water for 1 h. The "control" seedlings were kept in water for 2 h. The temperature of all solutions was maintained at 28 • C. The seedlings were then thermostated for 3 days at 28 • C. Each experiment was performed in triplicate; 20 seeds were used in each experiment. The results are given in Table 1.

Growth-Stimulating Effect Studies
Sunflower seeds were treated with a solution of a test compound at different concentrations (10 -2 to 10 -5 by weight) for 1 h. After 1 h, the seeds were spread evenly on strips of filter paper, rolled up, placed in beakers with water and thermostated at 28 • C for 3 days. Then stem and root length were measured, and the data were statistically processed using Student's t-test, p = 0.95. Each experiment was carried out in triplicate with 100 seeds. The results are given in Table 3.

Conclusions
Generally, a new preparative method for synthesis of alkyl 6-amino-4-aryl-7-phenyl-3-(phenylimino)-4,7-dihydro-3H- [1,2]dithiolo [3,4-b]pyridine-5-carboxylates 15 based on the base-catalyzed cascade reaction of dithiomalondianilide 12 with alkyl 3-aryl-2-cyanoacrylates was developed. Alternatively, the same compounds can also be prepared in a single step starting from aromatic aldehydes, ethyl (or methyl) cyanoacetate and dithiomalondianilide. Related reactions of dithiomalondianilide 12 with other Michael acceptors are currently underway in our laboratory. Some compounds showed a strong antidote effect with respect to herbicide 2,4-D accompanied by moderate growth-regulating activity in the experiments on sunflower seedlings. Thus, compounds 15a,c,f reduced the negative effect of 2,4-D on sunflower seedling hypocotyls by 34-60% and by 40-55% on sunflower seedling roots in the laboratory experiments and showed an antidote effect of 41.4-51.4% in the field experiments.