First experimental evidence suggests use of glucobrassicin as source of auxin in drought-stressed Arabidopsis thaliana

The synthesis of indole-3-acetonitrile (IAN) from the indolic glucosinolate (iGSL) glucobrassicin (GB) is a unique trait of members of the Brassicales. To assess the contribution of this pathway to indole-3-acetic acid (IAA) synthesis under stress conditions, drought stress (DS) experiments with Arabidopsis thaliana were performed in vitro. Analysis of GSLs in DS plants revealed higher contents of GB in shoots and roots compared to control plants. Deuterium incorporation experiments showed the highest turnover of GB compared to all other GSLs during drought conditions. Evidence suggests the involvement of the thioglucosidase BGLU18 in the degradation of GB. The nitrile specifier proteins NSP1 and NSP5 are known to direct the GSL hydrolysis towards formation of IAN. Nitrilases like NIT2 are able to subsequently synthesize IAA from IAN. Expression of BGLU18, NSP1, NSP5 and NIT2 and contents of GB, IAN and IAA were significantly elevated in DS plants compared to control plants suggesting the increased use of GB as IAA source. Significantly higher contents of reactive oxygen species in DS bglu18 and epithionitrile specifier protein (esp) mutants compared to Col-0 indicate higher stress levels in these mutants highlighting the need for both proteins in DS plants. Furthermore, GB accumulation in leaves was higher in both mutants during DS when compared to Col-0 indicating enhanced synthesis of GB due to a lack of breakdown products. This work provides the first evidence for the breakdown of iGSLs to IAN which seems to be used for synthesis of IAA in DS A. thaliana plants.


Supplementary Figure S2
Indicators of stress analyzed in Arabidopsis thaliana Col-0 subjected to mild or severe drought stress for 7 days. Plants were grown on petri dishes for five weeks prior to application of drought stress. A-B: Total leaf water content; C-D: Content of reactive oxygen species (ROS) analyzed as 2,2′-azobis(2-amidino-propane) equivalents (AAPHE); E-F: Expression of the drought-induced gene P5CS1 relative to the standard gene EF1α; G,I: Control plants, H: Mildly droughtstressed plants subjected to 20% PEG 20,000 for 7 days; J: Severely drought-stressed plants subjected to 40% PEG 20,000 for 7 days. Bars represent means of three biological replicates consisting of four pooled plants each. Different letters indicate significant differences (p<0.05) between control and drought-stressed plants.

Supplementary Figure S3
Repetition of severe drought-stress experiment ( Supplementary Fig. 2). Indicators of stress analyzed in Arabidopsis thaliana Col-0 subjected severe drought stress for seven days. Plants were grown on petri dishes for five weeks prior to application of severe drought stress (40% PEG 20,000). A: Total leaf water content; B: Content of reactive oxygen species (ROS) analyzed as 2,2′-azobis(2-amidino-propane) equivalents (AAPHE); C: Expression of the drought-induced gene P5CS1 relative to the standard gene EF1α. Bars represent means of three biological replicates consisting of four pooled plants each. Different letters indicate significant differences (p<0.05) between control and droughtstressed plants.  Figure S4 Remaining contents of aliphatic (A-D) and indolic glucosinolates (E-F) in leaves of mildly drought-stressed Arabidopsis thaliana Col-0 plants subjected to 20% PEG 20,000 for seven days (Fig. 2). Plants were grown on petri dishes for five weeks prior to application of mild drought stress (20% PEG 20,000). Bars represent means of three biological replicates consisting of four pooled plants each. Different letters indicate significant differences (p<0.05) between control and drought-stressed plants.

Supplementary Figure S5
Remaining contents of aliphatic (A-D) and indolic glucosinolates (E-F) in leaves of severely drought-stressed Arabidopsis thaliana Col-0 plants subjected to 40% PEG 20,000 for seven days (Fig. 4). Plants were grown on petri dishes for five weeks prior to application of severe drought stress (40% PEG 20,000). Bars represent means of three biological replicates consisting of four pooled plants each. Different letters indicate significant differences (p<0.05) between control and drought-stressed plants.  (Fig. 4). Aliphatic (B) and indolic (A, C, D) glucosinolate contents in roots of severely drought-stressed Arabidopsis thaliana Col-0 plants subjected to severe drought stress (40% PEG 20,000) for seven days. Plants were grown on petri dishes for five weeks prior to application of drought stress. Bars represent means of three biological replicates consisting of four pooled plants each. Different letters indicate significant differences (p<0.05) between control and drought-stressed plants.

Supplementary Figure S7
Preceding severe drought-stress experiment with plants grown on soil. Aliphatic (A-E) and indolic (F-H) glucosinolate contents in leaves of severely drought-stressed Arabidopsis thaliana Col-0 plants subjected to 40% soil water content for five days. Plants were grown with optimal water supply for five weeks prior to application of drought stress. Bars represent means of three biological replicates consisting of three pooled plants each. Different letters indicate significant differences (p<0.05) between control and drought-stressed plants. CYP71A13, H) to the reference gene EF1α. Arabidopsis thaliana Col-0 plants were subjected to 40% soil water content for five days. Plants were grown with optimal water supply for five weeks prior to application of drought stress. Bars represent means of three biological replicates consisting of three pooled plants each. Different letters indicate significant differences (p<0.05) between control and drought-stressed plants.

Supplementary Figure S9
Repetition of severe drought-stress experiment (Fig. 3). Aliphatic (A-E) and indolic (F-H) glucosinolate contents in leaves of severely drought-stressed Arabidopsis thaliana Col-0 plants subjected to severe drought stress (40% PEG 20,000) for seven days. Plants were grown on petri dishes for five weeks prior to application of drought stress. Bars represent means of three biological replicates consisting of four pooled plants each. Different letters indicate significant differences (p<0.05) between control and drought-stressed plants.  (Fig. 7). Relative expression of genes involved in glucosinolates breakdown (BGLU18, A; NSP5, B; NSP1, C; ESP, E) and modification (Nit2, D) of breakdown products to the reference gene EF1α. Plants were subjected 40% PEG 20,000 (severe drought stress) for seven days. Plants were grown on petri dishes for five weeks prior to application of drought stress. Bars represent means of three biological replicates consisting of four pooled plants each. Different letters indicate significant differences (p<0.05) between control and droughtstressed plants.
Supplementary Figure S11 Repetition of severe drought-stress experiment (Fig 8). Relative expression of genes involved in transport (GTR1, A) and regulation (WRKY63, B; MYB29, C) of glucosinolates and synthesis of a key enzyme of indole-3-acetic acid synthesis (CYP71A13, D) to the reference gene EF1α. Plants were subjected to 40% PEG 20,000 (severe drought stress) for seven days. Plants were grown on petri dishes for five weeks prior to application of drought stress. Bars represent means of three biological replicates consisting of four pooled plants each. Different letters indicate significant differences (p<0.05) between control and drought-stressed plants.

Supplementary Table S12
List of all mutants analyzed in this study.

Mutant name TAIR code Mutant information nsp1
At3g16400 SALK_072600C nit2-cluster At3g44300/ At3g44310/ At3g44320 RNAi mutant provided by Stephan Pollmann (Lehmann et al., 2017) bglu18 Supplementary Figure S15 Expression of TGG1 (A) and TGG2 (B) to the reference gene EF1α. Plants were subjected to 40% PEG 20,000 (severe drought stress) for seven days. Plants were grown on petri dishes for five weeks prior to application of drought stress. Bars represent means of three biological replicates consisting of four pooled plants each. Different letters indicate significant differences (p<0.05) between control and drought-stressed plants.