ACCERBATIN, a small molecule at the intersection of auxin and reactive oxygen species homeostasis with herbicidal properties

ACCERBATIN is an ethylene-mimicking small molecule that affects auxin homeostasis and ROS accumulation in etiolated seedlings. In light-grown plants, it exhibits auxin-like herbicidal properties.


A B
Supplementary  All chromatograms were processed, integrated and aligned as published before (Morreel et al., 2014). In total, this yielded 5692 m/z features that could be putatively assigned to 822 compounds following the "peak grouping" algorithm previously described (Morreel et al., 2014). Besides the AEX compound itself, only one other "peak group" (called AEXfrg) was observed to be solely present in those samples that were fed with the AEX compound. Following MS data for AEX and AEXfrg were obtained (relative abundance versus the base peak is given between parentheses for

Supplementary Protocols S2. AEX stability in vitro determined by Nuclear magnetic resonance (NMR)
Given the structure of the component of interest (AEX) with brutoformula C 23 H 19 BrF 2 N 2 O 2 , it was noticed that after heating a certain amount of product degraded to C 19 H 17 O 2 NBr, which corresponds to a loss of a C 4 H 2 NF 2 -fragment. This suggests that at least a part of the fluorinated aromatic ring is fragmented during the heating step. The aim of the NMR measurements discussed here is to explore whether the same fragmentation event can be reproduced and in the case this happens if the new compound after fragmentation can be identified.
Regarding the spectra shown in the this paragraph, a remark has to be made: considering the fact that the original samples were already dissolved in 53 µl of protonated methanol, the baseline of several spectra will be distorted due to the intensity of the two methanol signals. In addition, overlap of the previously mentioned signals with some signals of the molecule of interest cannot be excluded: as can be seen in Figure 2, all the resonances can be assigned except for proton n° 14, which is expected to be located around 4.7 ppm (chemical shift prediction ChemDraw Ultra 13, numbering corresponds with the numbering in Figure 1) and is believed to be overlapping with the second very intense methanol signal residing at 4.80 ppm.
The assignment of the molecule itself is fairly straightforward and can be performed almost solely on the basis of chemical shift and integral values, except for the amine n°10 and amide n°8 where none of the two can be assigned unambiguously. In addition, as can be seen in Figure 2, the integral values are in agreement with the number of protons assigned to each 1 H signal. It has to be noted that when the signal is situated closer to one of the methanol signals, the integral will start to deviate from the correct value due to the partial overlap of the large background of these solvent signals with the signal of interest (e.g. n°13 should correspond with 2 protons, where the integral corresponds with 2.8). In addition, it can be noted that some small impurities are present as well.
Regarding the first comparison of the three different samples in Figure 3, measured at room temperature under identical conditions, it can be noted that there are no significant differences in the signals of interest: both sets of aromatic signals are present and no new signals compared to signals of the AEX component are visible.
During the temperature study, several 1D 1 H measurements were performed at regular intervals of ±30min. Figure 4 shows three spectra, one at the start of the temperature study and two spectra measured after 6 hours and 12 hours of heating at 50°C. As can be seen clearly, no changes whatsoever can be noted throughout the experiments. In special interest, highlighted in the three spectra, the aromatic signals remain identical.
As a last type of experiment, two samples were subjected to a pH study: both the reference sample which wasn't heated as the sample prior heated at 80°C, were both measured at pH ±5 and ±4. As can be seen from Figure 5 and 6, no notable differences concerning the signals of interest as any new signals can be observed.
Considering both the temperature as the pH study don't show any notable differences before and after heating/pH adjustment, one has to conclude the AEX compound is both thermal and pH stable. Another possibility could be that the fragmentation is indeed happening but the resulting fragment and changes in the signals of interest are below the NMR detection limit.

Supplementary Protocols S3. Kinematic and genetic analysis of the effect of AEX on hook development
The exaggerated apical hook curvature is one of the key features of AEX (Fig. 1b, Supplementary   Fig. S4A). To know when AEX starts to act in hook development, a kinematic analysis was performed. The apical hook of etiolated Col-0 seedlings displayed three constitutive phases of development, consistent with previous results: the formation (0~36 hours), maintenance (36~48 hours) and opening phase (48~144 hours) (Supporting Information Fig.S3B) Smet et al., 2014). ACC-treated wild-type seedlings exhibited a significant hook exaggeration (~235° during maintenance phase), and were characterized by an extended formation phase (0~48 hours) and prolonged opening phase (252 hours after germination), while the rate of opening was similar to control. AEX-treated Col-0 seedlings also showed three distinct phases of hook development, but exhibited a formation phase lasting 3.5 days and a prolonged maintenance phase (84~132 hours; 258°) compared to control seedlings. The hook opening rate was slower, and not even completed after 360 hours post germination. Finally, when AEX and ACC were combined, effects on all three phases of apical hook development were even more pronounced.
Hence, the combined treatment led to a significantly more exaggerated curvature than that of AEXor ACC-treated seedlings.
Hook development in Arabidopsis is strongly controlled by the HOOKLESS 1 (HLS1) gene (An et al., 2012). Kinetic analysis of hls1-1 hook development revealed that both control and ACCtreated seedlings immediately entered the opening phase and reached an angle of 0ᵒ at 48 hours after germination (Supplementary Fig. S4A and C). Upon AEX treatment however, hls1-1 seedlings formed a conspicuous hook structure (until 132° at 24 hours after germination), and subsequently started opening, which was completed at 90 hours after germination. Thus, AEX is likely acting downstream of HLS1.