Summary: Secondary metabolism is characterized by an impressive structural diversity. Here, we have addressed the mechanisms underlying structural diversification upon damage‐induced activation of glucosinolates, a group of thioglucosides found in the Brassicales. The classical pathway of glucosinolate activation involves myrosinase‐catalyzed hydrolysis and rearrangement of the aglucone to an isothiocyanate. Plants of the Brassicaceae possess specifier proteins, i.e. non‐heme iron proteins that promote the formation of alternative products by interfering with this reaction through unknown mechanisms. We have used structural information available for the thiocyanate‐forming protein from Thlaspi arvense (TaTFP), to test the impact of loops protruding at one side of its β‐propeller structure on product formation using the allylglucosinolate aglucone as substrate. In silico loop structure sampling and semiempirical quantum mechanical calculations identified a 3L2 loop conformation that enabled the Fe²⁺ cofactor to interact with the double bond of the allyl side chain. Only this arrangement enabled the formation of allylthiocyanate, a specific product of TaTFP. Simulation of 3,4‐epithiobutane nitrile formation, the second known product of TaTFP, required an alternative substrate docking arrangement in which Fe²⁺ interacts with the aglucone thiolate. In agreement with these results, substitution of 3L2 amino acid residues involved in the conformational change as well as exchange of critical amino acid residues of neighboring loops affected the allylthiocyanate versus epithionitrile proportion obtained upon myrosinase‐catalyzed allylglucosinolate hydrolysis in the presence of TaTFP in vitro. Based on these insights, we propose that specifier proteins are catalysts that might be classified as Fe²⁺‐dependent lyases.

Access to inorganic phosphate (Pi), a principal intermediate of energy and nucleotide metabolism, profoundly affects cellular activities and plant performance. In most soils, antagonistic Pi-metal interactions restrict Pi bioavailability, which guides local root development to maximize Pi interception. Growing root tips scout the essential but immobile mineral nutrient; however, the mechanisms monitoring external Pi status are unknown. Here, we show that Arabidopsis LOW PHOSPHATE ROOT 1 (LPR1), one key determinant of Fe-dependent Pi sensing in root meristems, encodes a novel ferroxidase of high substrate specificity and affinity (apparent K_M ∼ 2 μM Fe²⁺). LPR1 typifies an ancient, Fe-oxidizing multicopper protein family that evolved early upon bacterial land colonization. The ancestor of streptophyte algae and embryophytes (land plants) acquired LPR1-type ferroxidase from soil bacteria via horizontal gene transfer, a hypothesis supported by phylogenomics, homology modeling, and biochemistry. Our molecular and kinetic data on LPR1 regulation indicate that Pi-dependent Fe substrate availability determines LPR1 activity and function. Guided by the metabolic lifestyle of extant sister bacterial genera, we propose that Arabidopsis LPR1 monitors subtle concentration differentials of external Fe availability as a Pi-dependent cue to adjust root meristem maintenance via Fe redox signaling and cell wall modification. We further hypothesize that the acquisition of bacterial LPR1-type ferroxidase by embryophyte progenitors facilitated the evolution of local Pi sensing and acquisition during plant terrestrialization.

Abiotic and biotic stress can have a detrimental impact on plant growth and productivity. Hence there is a substantial demand for key factors of stress responses to improve yield stability of crops. Members of the poly(ADP-ribose)polymerase (PARP) protein family, which post-translationally modify (PARylate) nuclear proteins, have been suggested as such universal determinants of plant stress responses. A role under abiotic stress has been inferred from studies in which a genetic or, more commonly, pharmacological inhibition of PARP activity improved the performance of stressed plants. To further elucidate the role of PARP proteins under stress, T-DNA knockout mutants for the three Arabidopsis thaliana PARP genes were subjected to drought, osmotic, salt, and oxidative stress. To exclude a functional redundancy, which was indicated by a transcriptional upregulation of the remaining parp genes, a parp triple mutant was generated. Surprisingly, parp mutant plants did not differ from wild type plants in any of these stress experiments, independent from the number of PARP genes mutated. The parp triple mutant was also analyzed for callose formation in response to the pathogen-associated molecular pattern flg22. Unexpectedly, callose formation was unaltered in the mutant, albeit pharmacological PARP inhibition robustly blocked this immune response, confirming previous reports. Evidently, pharmacological inhibition appears to be more robust than the abolition of all PARP genes, indicating the presence of so-far undescribed proteins with PARP activity. This was supported by the finding that protein PARylation was not absent, but even increased in the parp triple mutant. Candidates for novel PARP-inhibitor targets may be found in the SRO protein family. These proteins harbor a catalytic PARP-like domain and are centrally involved in stress responses. Molecular modeling analyses, employing animal PARPs as templates, indeed indicated a capability of the SRO proteins RCD1 and SRO1 to bind nicotinamide-derived inhibitors. Collectively, the results of our study suggest that the stress-related phenotypes of parp mutants are highly conditional, and they call for a reconsideration of PARP inhibitor studies. In the context of this study, we also propose a unifying nomenclature of PARP genes and mutants, which is currently highly inconsistent and redundant.

Significance: The continued high rates of using antibiotics in healthcare and livestock, without sufficient new compounds reaching the market, has led to a dramatic increase in antimicrobial resistance, with multidrug-resistant bacteria emerging as a major public health threat worldwide. Because the vast majority of antiinfectives are natural products or have originated from them, we assessed the predictive power of plant molecular phylogenies and species distribution modeling in the search for clades and areas which promise to provide a higher probability of delivering new antiinfective compound leads. Our approach enables taxonomically and spatially targeted bioprospecting and supports the battle against the global antimicrobial crisis.