Synthesis and Biological Activity of 2′,3′-iso-Aryl-abscisic Acid Analogs

2′,3′-iso-Benzoabscisic acid (iso-PhABA), an excellent selective abscisic acid (ABA) analog, was developed in our previous work. In order to find its more structure-activity information, some structural modifications were completed in this paper, including the substitution of phenyl ring and replacing the ring with heterocycles. Thus, 16 novel analogs of iso-PhABA were synthesized and screened with three bioassays, Arabidopsis and lettuce seed germination and rice seedling elongation. Some of them, i.e., 2′,3′-iso-pyridoabscisic acid (iso-PyABA) and 2′,3′-iso-franoabscisic acid (iso-FrABA), displayed good bioactivities that closed to iso-PhABA and natural (+)-ABA. Some others, for instance, substituted-iso-PhABA, exhibited certain selectivity to different physiological process when compared to iso-PhABA or (+)-ABA. These analogs not only provided new candidates of ABA-like synthetic plant growth regulators (PGRs) for practical application, but also new potential selective agonist/antagonist for probing the specific function of ABA receptors.


Introduction
The crucial role of abscisic acid (S-(+)-ABA 1, Figure 1) as a phytohormone in a wide variety of physiological processes of plants, including inhibiting or promoting growth, maintaining bud and seed dormancy, affecting flowering and sex differentiation, as well as conducting environmental stresses response [1][2][3][4][5][6][7][8], motivates numerous scientific investigations into the agricultural application of ABA. However, the widely using of abscisic acid (ABA) has been hindered by its rapid metabolism in plants and light isomerization in vitro [9][10][11][12][13][14][15][16][17]. Focusing on the development of metabolism resistant analogs, significant efforts were devoted by different researchers. For example, methoxyl [18], alkynyl [19], or fluorine atoms [20,21] at the 8 or 9 -positionm and halogen atoms [22,23] at the 3 -position had been introduced to avoid 8 -hydroxylation and Michael addition, and the corresponding analogs displayed excellent ABA-like activities. Nonetheless, they are too expensive to practical use in field due to their complicated synthetic routes.
A successful case of 8 -hydroxylation resistant analog was 2 ,3 -benzoabscisic acid 2 [24,25] (Figure 1), which had excellent bioactivity and relatively efficient synthetic route. Inspired by this compound and iso-ABA 3 [26], in our previous works, we had developed 2 ,3 -iso-Benzoabscisic acid (iso-PhABA) 4a [27], which is an excellent and easier preparation ABA analog. The deeply investigations of bioactivity and agonist-receptor interactions for 4a suggested that it is a state-of-art ABA-like regulator and a selective ABA agonist, i.e., PYL10 has the highest inhibitory effect on the phosphatase activity of HAB1 in the presence of (+)-iso-PhABA. Meanwhile, the analysis on

Chemistry
As shown in Scheme 1, to obtain target compounds substituted-iso-PhABA (4b-4d) and heterocyclic analogs 2',3'-iso-heterocylicABA (iso-HetABA, 5a-5d, 6), four different pathways (a-d in Scheme 1) for the preparation of dicarbonyl intermediates 9b-9d, 13a-13d, and 17 were established, respectively. Firstly, the preparation of intermediates 9b-9d (Scheme 1a) were followed the similar pathway of the synthesis of 2′,3′-iso-PhABA 4 [25,27], comprising the vicinal methylation of the carbonyl of subtituted-1-tetralones (step i in Scheme 1a) to obtain 8b-8d and the oxidation of the benzyl carbon with Co(acac)2/t-BuOOH system [28] (step ii in Scheme 1a). Then, for the preparations of pyridine analogs 13a-13d, two different methods were applied for the intermediates 12a-12d. The reaction of  multiple hydrophobic interactions. Thus, it provides a new and robust precursor for the design of ABA receptor agonists/antagonists. Consequently, in this work, we aimed at the further studies of the analogs of 2′,3′-iso-PhABA 4a to obtain more compounds with good ABA-like bioactivity and selective agonist/antagonists-receptor interactions. Here, we reported the synthesis and biological activity of substituted-iso-PhABA analogs 4b-4d and heterocyclic analogs 5 and 6 (see Figure 2). Their structure-activity relationship (SAR) was also discussed.

Structure-Activity Relationship
The Arabidopsis, lettuce seed germination and the rice seedling elongation inhibiting tests in vivo were conducted to gain insights into the bioactivity of these iso-PhABA analogs. The (±)-iso-PhABA 4a and (+)-ABA 1 were used as the control agents. The corresponding data of bioactivities were summarized in Table 1. For the comparison between the bioactivities of substituted-iso-PhABA analogs, the overall bioactivity of 4b-4d were very close to those of each other, indicating that the effects of electron-donating and withdrawing groups, as well as their substituted-positions (7 and 6 ) were not crucial for their bioactivity. However, the bioactivities of all the substituted-iso-PhABA 4b-4d were worse than those of the control agents 4a and 1, suggesting that a substituent in the benzene ring of iso-PhABA had a significant impact on the bioactivity. Furthermore, there were some preferences of substituted-iso-PhABA to different physiologic process. Compound 4b-4d observed the highest IC 50 values for the rice seedlings elongation, which was the lowest values for both the two control agents 4a and 1, indicating their different preferential nature of bioactivity compared to iso-PhABA. Table 1. Inhibitory bioactivities of title compounds.
To study the effect of the size of the substituted group on the bioactivity, compound 5b-5d were designed and synthesized for the bioassays, too. Obviously, the IC 50 values of these compounds are decreasing with the increasing size of the substituted, i.e., -Me 5b to -n-Pr 5c to -Ph 5d. Therefore, this evidence validated that a big group in the aryl ring is not preferential for the bioactivity.
In order to acquire more analogs for the discussion of structure-activity relationship, the acetylenic acids 24, 25, and 26 were also synthesized and tested with three bioassays. As shown in Table 1, their bioactivities were significantly worse than those of their corresponding alkene acids 4, 5, and 6, suggesting the importance of the double-bond for the bioactivity. Besides, by the comparison between the acetylenic acids, acetylenic iso-PhABA 24a had the best activity for Arabidopsis seed germination (IC 50 = 1.38 µM), acetylenic 2 -Me-iso-PyABA 25b had the best activity for lettuce seed germination (IC 50 = 5.97 µM), and acetylenic iso-FrABA 26 had the lowest IC 50 value (4.32 µM) for rice seedlings elongation. Meanwhile, like the substituted-iso-PhABA analogs, there were obvious preferences of some acetylenic acids to different physiological process. For example, acetylenic 7 -OMe-iso-PhABA 24b showed a weakest inhibitor on Arabidopsis seed germination (IC 50 > 10 µM) in three bioassays; acetylenic 6 -Br-iso-PhABA 24d, on the other hand, had a lowest bioactivity on rice seedlings elongation (IC 50 > 10 µM). Hence, these results implied that there might be some difference in ABA signaling transduction between different acetylenic acids, as well as the acetylenic acids and alkene acids.