The Phytotoxicity of Meta-Tyrosine Is Associated With Altered Phenylalanine Metabolism and Misincorporation of This Non-Proteinogenic Phe-Analog to the Plant's Proteome

Plants produce a myriad of specialized (secondary) metabolites that are highly diverse chemically, and exhibit distinct biological functions. Here, we focus on meta-tyrosine (m-tyrosine), a non-proteinogenic byproduct that is often formed by a direct oxidation of phenylalanine (Phe). Some plant species (e.g., Euphorbia myrsinites and Festuca rubra) produce and accumulate high levels of m-tyrosine in their root-tips via enzymatic pathways. Upon its release to soil, the Phe-analog, m-tyrosine, affects early post-germination development (i.e., altered root development, cotyledon or leaf chlorosis, and retarded growth) of nearby plant life. However, the molecular basis of m-tyrosine-mediated (phyto)toxicity remains, to date, insufficiently understood and are still awaiting their functional characterization. It is anticipated that upon its uptake, m-tyrosine impairs key metabolic processes, or affects essential cellular activities in the plant. Here, we provide evidences that the phytotoxic effects of m-tyrosine involve two distinct molecular pathways. These include reduced steady state levels of several amino acids, and in particularly altered biosynthesis of the phenylalanine (Phe), an essential α-amino acid, which is also required for the folding and activities of proteins. In addition, proteomic studies indicate that m-tyrosine is misincorporated in place of Phe, mainly into the plant organellar proteomes. These data are supported by analyses of adt mutants, which are affected in Phe-metabolism, as well as of var2 mutants, which lack FtsH2, a major component of the chloroplast FtsH proteolytic machinery, which show higher sensitivity to m-tyrosine. Plants treated with m-tyrosine show organellar biogenesis defects, reduced respiration and photosynthetic activities and growth and developmental defect phenotypes.


INTRODUCTION
The chemical diversity of terrestrial plants is truly exceptional. Plants are estimated to produce hundreds of thousands of different metabolites, probably the largest number among all other species (see e.g., Moghe and Last, 2015;Wurtzel and Kutchan, 2016;Obata, 2019). Such an exceptional diversity may be a consequence of the large diversification and rapid evolution of specialized metabolic pathways in plants since they occupied the terrestrial environment, about 500 million years ago (Morris et al., 2018). Phytochemicals produced by land-plants are also of great economic and ecological importance, as herbicidal compounds to control weeds.
Phytochemicals, as other natural compounds, are classified into two main categories. Primary metabolites that are associated with essential cellular functions, e.g., nucleotides, amino acids, fatty acids, sugars, and organic acids, which are typically present in all organisms and cells. In addition, organisms also produce numerous specialized (secondary) metabolites that are not essential to basic physiological functions, but otherwise play important roles during specific growth and developmental stages, and aid in adapting and adjusting the developmental need of the plant with specific physiological or environmental signals (Witzany, 2006;Tissier et al., 2014;Moghe and Last, 2015;Wurtzel and Kutchan, 2016;Li et al., 2018). Specialized compounds produced in plants were shown to have key roles in defense mechanisms, in inter-or intracellular signaling, coloring, regulation of primary metabolism, as well as in allelochemisrty (i.e., biomolecules produced by one organism that have a physiological effect on another species when released to the environment) (Rizvi and Rizvi, 1992).
Plants have complex relationships with other organisms, which involve physical and chemical interactions with their surrounding (reviewed by e.g., van Loon, 2016;Bennett and Klironomos, 2018;Mhlongo et al., 2018). These are indicated, for example, by complex interactions between different plants and microorganisms in the soil. Growing roots also need to compete with their neighboring plants, and at the same time to attract beneficial microorganisms to supply them with minerals and nutrients, e.g., root nodules in legumes, where bacterial symbionts fix atmospheric nitrogen (van Loon, 2016;Bennett and Klironomos, 2018;Mhlongo et al., 2018). Accordingly, many chemical signals are exchanged between plants and their neighboring organisms in the soil (Mhlongo et al., 2018). Our study focuses on meta-tyrosine (m-tyrosine, or m-Tyr), an oxidized byproduct of the aromatic amino acid phenylalanine, which inhibits early post-germination and seedling growth (Bertin et al., 2007;Bertin et al., 2009).
At various conditions and especially under stresses, the production of tyrosine isomers para-(i.e., the native amino acid), meta-, and ortho-tyrosine, can spontaneously occur by the oxidation of the benzyl ring of phenylalanine (Mager and Berends, 1974). A few plant species, including Euphorbia myrsinites (donkey-tail spurge) and Festuca rubra (red fescue) synthesize enzymatically the isomer meta-tyrosine (m-tyrosine) in their root tip tissues, and then secrete it to the soil to inhibit nearby plant-life (Mothes et al., 1964;Bertin et al., 2007). In E. myrsinites, m-tyrosine is produced via a transamination of m-hydroxyphenylpyruvate (Mothes et al., 1964), whereas the biosynthesis of m-tyrosine in F. rubra is mediated directly through the hydroxylation of phenylalanine (Bertin et al., 2007;Huang et al., 2012). These data indicate that different plants utilize m-tyrosine as a phytotoxic allelochemical using distinct metabolic pathways, activities which likely arose independently during the evolution of land plants (Huang et al., 2012). While m-tyrosine is highly toxic to plants, its structural related isomers, o-and p-tyrosine, have no or only little effects on the germination, growth, or development of land plants (Bertin et al., 2007).
The molecular basis for m-tyrosine toxicity has not been resolved yet. We show, by biochemical and genetic approaches, that m-tyrosine affects early post-germination development of Arabidopsis plantlets by altering the biosynthesis of various amino acids, and in particular the aromatic amino acid Phenylalanine (Phe). Proteomic analyses of young Arabidopsis seedlings, grown in the absence or presence of m-tyrosine, strongly support that m-tyrosine is misincorporated instead of Phe to various organellar (i.e., mitochondrial and plastidial) proteins, an activity which is likely mediated by a duallylocalized organellar phenylalanine-transfer RNA (tRNA) synthetase (PheRS) enzyme (Duchêne et al., 2005).

Microscopic Analyses
For the analysis of plant morphology, plant tissues (i.e., leaves and roots) where obtained from 5-day-old Arabidopsis plants grown on MS-plates in the presence or absence of 10 mM m-tyrosine. The morphologies of mitochondria and plastids were established by transmission electron microscopy (TEM) of ultrathin plant sections, using Tecnai 12 TEM 100 kV (Phillips, Eindhoven, the Netherlands) microscope equipped with MegaView II CCD camera and Analysis ® version 3.0 software (SoftImaging System GmbH, Münstar, Germany), at the Bio-Imaging unit of the Institute of Life Sciences (The Hebrew University of Jerusalem). The relative densities (i.e., pixel intensities) of thylakoid grana membrane stacks and the average surface area of mitochondria have been manually evaluated from TEM images of ultrathin sections of 5-day-old plantlets grown in the absence or presence of m-tyrosine, using the ImageJ software (Version 1.52a) (Jensen, 2013). Student's t-test was performed to determine significant differences (P ≤ 0.05).

Preparation of Crude Organellar Membrane Extracts From Arabidopsis Seedlings
Crude organellar proteins were prepared essentially as described previously (Pineau et al., 2008;Shevtsov et al., 2018). In brief, organellar membrane extracts were obtained from 200 mg Arabidopsis seedlings (5 days-old), grown in the presence or absence of 10 mM m-tyrosine supplemented to the growth media. The seedlings were then homogenized in 2 ml of 75 mM MOPS-KOH, pH 7.6, 0.6 M sucrose, 4 mM ethylenediaminetetraacetic acid (EDTA), 0.2% polyvinylpyrrolidone-40, 8 mM cysteine, 0.2% bovine serum albumin (BSA), and protease inhibitor cocktail (Roche Diagnostics GmbH, Mannheim, Germany). Protein concentration was determined by the Bradford method (Bio-Rad, Catalog no. 5000201), according to the manufacturer's protocol. For immunoassays, crude membrane fraction were suspended in sample loading buffer (Laemmli, 1970) and subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) (at a constant 100 V). Following electrophoresis, the proteins were transferred to a polyvinylidene difluoride (PVDF) membrane (Bio-Rad, Catalog no. 1620177), essentially as described previously , and incubated overnight at 4°C with various antibodies (Table S1). Detection was carried out by chemiluminescence assay after incubation with an appropriate horseradish peroxidase (HRP)conjugated secondary antibody.

Blue Native Gel Electrophoresis for Isolation of Native Organellar Complexes
Blue native (BN)-PAGE of organellar membranous complexes was performed according to the methods described previously (Pineau et al., 2008;Shevtsov et al., 2018). Crude organellar membranes were solubilized with n-dodecyl-ß-maltoside [DDM; 1.5% (w/v)] and loaded onto a native 4 to 16% linear gradient gel. For immunoblotting of non-denaturing PAGE, the proteins were transferred from the gel onto a PVDF membrane (Bio-Rad, Catalog no. 1620177). The membranes were then incubated with specific primary antibodies (Supplemental Table S1), and detection was carried out by chemiluminescence assay after incubation with horseradish peroxidase (HRP)-conjugated "secondary" antibodies.

Proteomic Analyses
Following the extraction of total protein from 5-day-old Arabidopsis seedlings and crude organellar preparations (Pineau et al., 2008;Shevtsov et al., 2018), total proteins were obtained by the borate/ ammonium acetate method (Maayan et al., 2008). For this purpose, plant tissues were homogenized in the presence of polyvinylpolypyrrolidone (PVPP). The homogenate was added to microfuge tubes containing 400 ml ice-cold protein extraction buffer [50 mM Na-borate, 50 mM ascorbic acid, 1.25% (w/v) sodium dodecyl sulfate (SDS), 12.5 mM b-mercaptoethanol, pH 9.0] and the protease inhibitor cocktail "complete Mini" from Roche Diagnostics GmbH (Mannheim, Germany). Proteins were recovered by centrifugation (25,000 g) in the presence of three volumes of icecold 0.1 M ammonium acetate in methanol buffer (NH 4 -OAc-MeOH), following (80% v/v) acetone precipitation. The protein pellet was resuspended with 25 mM Tris-HCl pH 8.0, 10 mM dithiothreitol (DTT), 2% SDS buffer solution. Protein concentration was determined according to the Bradford method, with BSA used as a standard. Twenty-five micrograms of protein was alkylated with 55 mM iodoacetamide (Sigma Chem. Corp. St. Louis, MO) for 30 min at room-temperature in the dark. Removal of SDS followed by digestion with trypsin (Promega Corp., Madison, WS) were performed using the S-Trap microspin column kit (ProtiFi, LLC, Huntington, NY), according to the manufacturer protocol. The tryptic peptides were then desalted, as previously described (Rappsilber et al., 2007). A total of 1.5 µg of peptides from each sample were injected into the mass spectrometer (Q Exactive Plus mass spectrometer, Thermo Fisher Scientific, USA). The data were analyzed using the PEAKS proteomics software, version 8 (Bioinformatics Solutions Inc., Waterloo, ON, Canada), with specified Phe to Tyr replacements (i.e., the mass of m-tyrosine is equal to Tyr) (see Table S3).

Amino Acids Analysis
The extraction of Arabidopsis plant free amino acids (and various other metabolites) was performed according to a previously described procedure with some modifications by (Roessner-Tunali et al., 2003). In general, 50 mg Arabidopsis seedlings (5day-old) were frozen in liquid nitrogen and homogenized in 700 µl of MeOH, in the presence of 30 µl ribitol (0.2 mg/ml in DDW) as an internal standard. Extraction, derivatization, standard addition, and sample injection by gas chromatography-mass spectrometry (GC-MS) were performed essentially as described previously (Lisec et al., 2006). The GC-MS system comprised of the an Agilent 7693 Autosampler, Agilent J&W DB-35ms column, Agilent 7200B gas chromatograph, quadrupole timeof-flight MS with removable electron ionization source. Amino acids were identified in comparison to commercial standard compounds purchased from Sigma. Chromatograms and mass spectra were evaluated using MassHunter Data Analysis by Agilent, Quantitative and Qualitative Analysis (TOF). Calculations of the relative free amino acid levels were evaluated by calculating the relative pick area of each amino acid in untreated versus m-tyrosine grown seedlings. For detailed parameters see Supplemental Dataset 1 (https://figshare.com/ articles/Frontiers_in_Plant_Sciences_MS_data/11627211).

m-Tyrosine Affects Arabidopsis Early Seedlings Establishment and Root Development
Specialized phytochemical compounds are highly diverged chemically, and are expected to arise in plants to aid with key biological functions, such as in surviving and communicating with other organisms in complex environmental niches. F. rubra, a grass species that is widespread across the northern hemisphere, releases to the soil biochemicals that inhibit the growth of nearby plant-life. NMR-spectroscopy led to the identification of m-tyrosine as the key phytotoxic compound within root exudates of F. rubra plants (Bertin et al., 2007). Plants treated with exogenous m-tyrosine are characterized by shortened roots and reduced biomass phenotypes [Bertin et al. (2007) and Supplemental Figure S1]. The phytotoxic effects of m-tyrosine seem to be mediated by mechanisms that are not related to auxin (Bertin et al., 2007). Our analyses further indicate that Arabidopsis seedlings germinated in the presence of m-tyrosine (i.e., 0 to 320 mM), show post-germination developmental defects, and strong effects on root growth (calculated IC 50 value of 2.365 mM; i.e., concentrations required to achieve 50% reduction of Arabidopsis root growth) (Supplemental Figure S1 and Table 1).

Free Amino Acid Analysis of Arabidopsis Seedlings Grown on m-Tyrosine
The molecular basis for m-tyrosine mediated phytotoxicity is largely unclear in plants. Given its chemical properties, we speculate that m-tyrosine might be incorporated to the plant proteome and/or interferes with cellular metabolism. To address this, we analyzed the relative accumulation of free amino acids in 5-day-old Arabidopsis thaliana seedlings grown in the absence (control, −m-Tyr) or presence of 10 mM m-tyrosine (+m-Tyr) by GC-MS analysis (see Figure 1, and Supplemental Table S2 and S3). The accumulation of various amino acids was evaluated relatively to the control (non-treated) plants grown under the same growth conditions (see Table S3).
Under "normal" growth conditions (see Materials and Methods), the steady-state levels of various amino acids, including the nonpolar (hydrophobic) Ala, Ile, Leu, Met, and Val, the uncharged Ser and Thr, positively charged Arg, His, and Lys, the negatively charged Asp, as well as the aromatic residues Trp and Tyr, were all reduced to some extent (i.e., from 7 to 26%) in the m-tyrosine germinated seedlings ( Figure 1, and Table S2). However, only Phe showed a statistically significant reduction (43.1 ± 6.3%) in the treated plants (see Figure 1, and Table S2). The nonpolar amino acid Pro, which is known to accumulate under different stress conditions (Szabados and Savouré, 2010), was found to be somewhat higher (110.9 ± 18.2%) in plants germinated in the presence of m-tyrosine, whereas the steadystate levels of the negative charged Glu (1.01 ± 0.14%) and the uncharged Gln (103.0 ± 2.9%) amino acids were not significantly affected by the addition of m-tyrosine to the growth media ( Figure 1, Table S2).
Growing plants in the presence of exogenous amino acids was previously shown to partially restore the growth and developmental defect phenotypes of Arabidopsis seedlings treated with m-tyrosine (Bertin et al., 2007). Root growth inhibition by 3 mM m-tyrosine was counteracted by 14 out of 21 different amino acids given at 40 mM concentrations (Bertin et al., 2007). However, when the amino acids were administered to the growth media together with a higher concentration (i.e., 10 mM) of m-tyrosine, only Phe showed a notable "rescue" effect on Arabidopsis root growth inhibition ( Figure 2).

The Effects of m-Tyrosine on Mutants Affected in Arogenate Dehydratase Activity
The biosynthesis of Phe in various organisms involves a series of enzymatic reactions, which convert prephanate (a product of the shikimate pathway) into Phe, from either phenylpyruvate or arogenate (the later serves as the primarily route for Phe synthesis in angiosperms) (Cho et al., 2007;Tzin and Galili, 2010). In Arabidopsis, arogenate dehydratase (ADT) catalyzes the final step in Phe biosynthesis (i.e., decarboxylation of arogenate), a reaction that takes place within the plastids (Tohge et al., 2013). Arabidopsis contains six closely related ADT genes, which differ in their organ locations and contribution to Phe accumulation (Höhner et al., 2018). A r a b i d o p s i s A D T -d e fi c i e n t m u t a n t s s h o w a l t e r e d photosynthetic rates, where stronger phenotypes are seen in lines that had deficiencies in multiple ADT isoforms (Höhner et al., 2018). These data are in accordance with the genetic and biochemical analyses, as well as with the expression patterns of different ADT genes in Arabidopsis, which indicate that different ADT isoforms in land plants are redundant in Phe-biosynthesis (Chen et al., 2016;Höhner et al., 2018). Arabidopsis mutants that accumulate lower levels of free-Phe are expected to show higher sensitivity to m-tyrosine toxicity, due to altered Phe metabolism ( Figure 1 and Table S2) and/or increased misincorporation of the Phe-analog into the plant proteome. Analysis of single, double, triple, and quadruple knockout lines (Chen et al., 2016; TABLE 1 | Effects of m-tyrosine on root length and leaf number of 5-day-old Arabidopsis thaliana (col-0) seedlings.
FIGURE 1 | Relative change in free amino acid levels in 5-day-old Arabidopsis thaliana seedlings grown in the presence or absence of 10 mM m-tyrosine. Arabidopsis seeds were germinated in Murashige and Skoog (MS)-agar plates in the absence or presence of 10 mM m-tyrosine (m-Tyr). Amino acids, extracted from 50 mg of 5-day-old Arabidopsis seedlings, were identified in comparison to commercial standard compounds from each treatment (i.e., control and m-Tyr-grown seedlings), in the presence of ribitol (0.2 mg/ml) as an internal standard. Extraction, derivatization, standard addition, and sample injection were performed essentially as described previously (Lisec et al., 2006). Calculations of the relative free amino acid levels were evaluated from the relative pick area of each amino acid in untreated versus m-tyrosine-treated seedlings. The values are means of four biological replicates, while error bars indicate one standard deviation (see Table S2).
Color boxes indicate to functional groups: in green hydrophobic side chains, blue for electrically charged side chains, purple for polar (uncharged) side chains, red for aromatic residues, and yellow indicates special side chains. Höhner et al., 2018) (generously provided by Prof. Jirong Huang, Shanghai Normal University) indicated that triple adt3/4/5 and quadruple adt3/4/5/6 mutants are more sensitive to m-tyrosine, further supporting a functional redundancy between different ADT paralogs in Arabidopsis thaliana plants (Figure 3 and S2).
In the absence of m-tyrosine, the root sizes were similar in 5-day-old wild-type plants and adt mutant-lines (i.e., about 3 cm in length; Figures 3A, B). Yet, when the seeds were germinated in the presence of 2.5 or 5 mM m-tyrosine, root development was more strongly affected in the triple adt3/4/5 (81% inhibition at 2.5 mM and 89% at 5 mM m-Tyr) or quadruple adt3/4/5/6 mutants (90% inhibition at 2.5 mM and 89% at 5 mM m-Tyr) than in wild-type plants or single adt mutant lines (i.e., 61~75% inhibition at 2.5 mM and 82~84% at 5 mM m-Tyr) ( Figures 3B,  D). Accordingly, the m-tyrosine concentration required to achieve 50% reduction of root growth of adt3/4/5/6 mutantline was notably lower than that of wild-type plants (i.e., IC 50 values 0.761 and 2.365 mM, respectively) ( Figures 3C and S1). Next, we examined the effects of different Phe concentrations (i.e., 20, 40, and 80 mM) on root growth of Arabidopsis thaliana wild-type (Col-0) and adt3/4/5/6 mutants grown in the absence or presence of 10 mM m-tyrosine ( Figure S2A). Increased concentrations of Phe partially rescued the early seedlings establishment and root developmental defect phenotypes by mtyrosine ( Figure S2A). At concentrations above 40 mM phenylalanine, the m-tyrosine mediated altered postgermination development and inhibited root development were restored in both the wild-type and adt3/4/5/6 mutant ( Figure  S2A). To a lesser degree, the effect of m-tyrosine on root growth was also counteracted by the addition of Phe in other adt mutant-lines ( Figure S2B).

Arabidopsis Plants Treated With m-Tyrosine Display Chloroplasts and Mitochondria Biogenesis Defect Phenotypes
Arabidopsis seedlings germinated in the presence of m-tyrosine (i.e., above 20~40 mM) had yellowish to white cotyledons ( Figure  S1), with reduced chlorophyll content ( Table 2), thus suggesting that the seedlings are defective in chloroplast development. We observed that the m-tyrosine-associated phenotypes can be partially restored by the addition of sucrose to the growth media ( Figure S3, + sucrose), further suggesting that m-tyrosine affects organellar biogenesis during early seedlings establishment and root development. Microscopic analysis of young (i.e., 5-day-old) Arabidopsis plantlets treated with 10 mM m-tyrosine showed altered chloroplast morphologies, with less grana lamella ( Figure 4). The relative densities of the thylakoid grana stacks in m-tyrosine treated plants were found to be notably lower (0.56 ± 0.07 times) than those of Col-0 plants grown under the same conditions in the absence of mtyrosine (1.00 ± 0.07, Table 2). We further noticed the appearance of many plastoglobuli (PG) in the m-tyrosine treated plants ( Figures  4D-F). These lipoprotein particles, which are commonly observed in colored plastids (i.e., chromoplasts), seem particularly prominent in plants affected in chloroplast development (Austin et al., 2006).
In addition to altered chloroplast biogenesis defects, we also noticed mild alterations in mitochondria structure in the presence of m-tyrosine (i.e., 10 mM). While the electron micrographs of mitochondria of the non-treated plantlets showed characteristic internal cristae formation, as dense folds of the inner-membrane sections ( Figures 4A-C), many mitochondria in m-tyrosine grown seedlings seemed larger (about 3.6x in organellar area, Table 2), with reduced inner mitochondrial membrane electron density and less cristae organization ( Figures 4D-F). Similar organellar morphologies were also noticed in various Arabidopsis mutants affected in mitochondria gene expression (see e.g., Keren et al., 2009;Keren et al., 2012;Zmudjak et al., 2013;Cohen et al., 2014).

m-Tyrosine Treated Plants Display Altered Photosynthesis and Respiration Activities
To determine whether the photosynthetic and respiratory activities were altered in plants grown in the presence of mtyrosine, we monitored the O 2 -evolution rates of 5-day-old seedlings in the dark, using a Clark-type electrode ( Table 2 and Figure S4). In the dark, the average O 2 -uptake rate of untreated plants was 127.81 ± 7.72 nmol O 2 ·min −1 ·gFW −1 . Arabidopsis seedlings grown in the presence of 10 mM m-tyrosine showed lower respiratory activities (i.e., 80.83 ± 3.51 nmol O 2 ·min −1 ·gFW −1 ) than those of the control (i.e., untreated) plants ( Table 2 and Figure S4). The photosynthetic activities of control and m-tyrosine treated plants were examined by monitoring the O 2 -evolution rates at different light intensities (0-1,000 mmol photons · m −2 ·s −2 ). The light compensation point was similar in *1-The data shows the relative root lengths, number of leaves, total chlorophyll, total protein, respiration, and photosynthetic activities in untreated versus m-tyrosine-treated seedlings.
The values are means of 4-7 biological replicates ± SE. Numbers in bold indicate significant differences from Arabidopsis plantlets grown in the absence (0 mM) of m-tyrosine, as determined by Student's t-test (P < 0.05). *2-Measured by a Clark-type electrode. *3-Calculated from measurements at different light intensities ( Figure S1). *4-The relative membrane densities (i.e., pixel intensities) of grana stacks have been assessed from transmission electron microscopy (TEM) images of ultrathin plant sections (Figure 4), using the ImageJ software (Jensen, 2013). Error bars indicate SE (n = 15). Numbers in bold show significant differences from the control (i.e., Arabidopsis plantlets grown in the absence of m-tyrosine), as determined by Student's t-test (P < 0.05). *5-The average area of mitochondria have been assessed from transmission electron microscopy (TEM) images of ultrathin plant sections (Figure 4), using the ImageJ software (Jensen, 2013). Error bars indicate SE (n = 5). Numbers in bold show significant differences from the control (i.e., Arabidopsis plantlets grown in the absence of m-tyrosine), as determined by Student's t-test (P < 0.05). *6-n.d., not determined.
Comparative measurements of light saturation curves of photosynthesis (i.e., O 2 -evolution rates, using a Clark-type electrode) were performed with untreated ( Figure S4, open boxes) and 5-day-old seedlings grown in the presence of 10 mM m-tyrosine ( Figure S4, filled boxes). The characteristics parameters of the photosynthetic activities were evaluated by fitting the net photosynthesis data to the equation Pmax (MacIntyre et al., 2002). The maximal photosynthetic yield value (P max ) of the control plants was 428.5 nmol O 2 ·min −1 ·gFW −1 , while the P max value of m-tyrosine treated plants was notably lower (i.e., 318.4 nmol O 2 ·min −1 ·gFW −1 ). Yet, a lower light intensity was required for the saturation of photosynthesis in the m-tyrosine grown plants (i.e., 50.9 µmol ·m −2 ·s −1 , compare with 102.3 µmol ·m −2 ·s −1 in the control plants). The data also suggested that m-tyrosine treated seedlings experienced photoinhibition under lower light intensities, as seen by a decline in O 2 -evolution rates in light intensities higher thañ 600 µmol photons · m −2 · s −1 ( Figure S4), while the control plants did not show reduced photosynthetic activities in light intensities above~1,000 µmol photons ·m −2 ·s −1 . Photoinhibition in the mtyrosine treated plants was further apparent by their higher b values (i.e., 0.05198), in comparison to those evaluated for the untreated plants (i.e., b values = 0.009976). These data are indicating that the photosynthetic and respiration activities were reduced in Arabidopsis plantlets treated with m-tyrosine.

Arabidopsis Plants Treated With m-Tyrosine Show Mild Reductions in the Accumulation of Various Organellar Complexes
The accumulation of organellar proteins was analyzed in 5-dayold Arabidopsis seedlings treated with 10 or 20 mM m-tyrosine. Total protein, extracted from control (−m-Tyr) and m-tyrosine treated plantlets, was analyzed by (12%) SDS-PAGE, following immunoblotting with antibodies raised against various organellar proteins ( Figure 5A and Table S1). The steady-state levels of the mitochondrial NAD9, CA2, and COX2 (i.e., between 30 up to 55% reductions), the plastidial PsbC protein (about 64% lower), and to a lesser extent the plastidial PsaA subunit (~25% lower), were all found to be reduced in seedlings germinated in the presence of m-tyrosine ( Figures 5B, C). The accumulations of AtpA, RISP, VDAC, and PsbD subunits were not significantly affected by the addition of m-tyrosine to the growth media ( Figures 5B, C). The rotenone-insensitive NADH dehydrogenases (NDBs) and alternative oxidases (AOXs) are induced in a coordinated manner under respiratory chain dysfunction and oxidative stress conditions (Clifton et al., 2005;Garmier et al., 2008;Rasmusson et al., 2008;Yoshida and Noguchi, 2009;Keren et al., 2012). The signals of AOX and NDB2 were both found to be somewhat higher (about 2.5 and 7.2 times higher, respectively) in the 5-day-old m-tyrosine treated Arabidopsis plantlets (Figures 5B, C).
To investigate the effects of m-tyrosine on the accumulation of native organellar complexes, crude membranous fractions were obtained from 5-day-old seedlings germinated in the absence or presence of 10 mM m-tyrosine. The membranous proteins were separated under native conditions by blue nativepolyacrylamide gel electrophoresis (BN-PAGE), and then subjected to immunoblot analyses with various antibodies (Table S1), as indicated in Figure 6. Arrows indicate to the native CI, CIII, VIV, CV, and PSI complexes. Under the gel electrophoretic conditions (Pineau et al., 2008;Shevtsov et al., 2018), we noticed to reduced steady-state levels of the respiratory complexes CI (47.5%), CIII (36.7%), CIV (59.9%), and CV (26.4%) upon m-tyrosine treatment ( Figure 6). Reduced CI activity was also apparent by "in-gel" activity assays ( Figure 6A). Similarly, immunoblots with PsaA indicated to reduced accumulation in photosystem I (PI) in seedlings grown in the presence of 10 mM m-tyrosine ( Figure  6B). The immunoassays with anti-CA2, COX2, and anti-PsaA antibodies further indicated the presence of additional protein bands ( Figure 6B, marked with asterisks). We speculate that the lower molecular weight bands in the CA2 and COX2 blots (i.e., about 800 and 150 kDa, respectively) may correspond to a partially assembled sub-CI and CIV particles, respectively (Meyer et al., 2011), whereas the upper-band (~500 kDa) in the PsaA blot may correspond to an oligomeric state of PSI (Salomon and Keren, 2011).

Analysis of m-Tyrosine Incorporation Into the Plant Proteome
The Phe-analog, m-tyrosine, affects early seedlings development and radicle elongation of plants ( Figures S1 and S3, and Bertin et al., 2007). Growing Arabidopsis seedlings in the presence of mtyrosine results with a reduction in the level of free-Phe ( Figure  1). Accordingly, adt mutants are more severely affected by mtyrosine (Figures 3 and S2), whereas the adt2-1D mutant plantline that accumulates higher levels of free-Phe (Cho et al., 2007) seems less susceptible to m-tyrosine toxicity . Similarly, the addition of exogenous Phe can partially restore the growth and developmental defect phenotypes associated with mtyrosine ( Figure 2, and Bertin et al., 2007). Here, we further show that m-tyrosine alters the biogenesis and activities of both the mitochondria and chloroplasts during early developmental stages (Figures 4, 6, S1, and S3, Table 2). It remains possible, therefore, that the developmental defect phenotypes cause by mtyrosine may relate to its misincorporation to the mitochondrial and plastidial proteomes.
To analyze whether m-tyrosine is incorporated into Arabidopsis proteins, we performed MS-based proteomic analyses of 5-day-old plantlets grown in the absence or presence of m-tyrosine (Supplemental Table S3, doi: 10.6084/ m9.figshare.11627211). The proteomic analyses indicated many Phe ! m-Tyr/Tyr exchanges (i.e., due to their identical masses it is not possible to distinguish between the native amino acid Tyr and its structural related isomers, o-and p-and m-tyrosine in the MS data) in various proteins of Arabidopsis seedlings grown in the presence of m-tyrosine, which were not observed in plants grown in the absence of the analog (i.e., under the same growth conditions). Table 3 summarizes the MS-based data of four independent proteomic analyses of 5-day-old seedlings grown in the absence (control) or presence of 10 mM and 20 mM mtyrosine. The frequencies of Phe mistranslation is dosedependent, as the misincorporation frequencies increase about two folds when plants were treated with 20 mM instead of 10 mM m-tyrosine (Table 3). In accordance with previous observations suggesting that organellar PheRSs might be more prone to FIGURE 5 | Relative accumulation of organellar proteins in 5-day-old Arabidopsis plants grown in the absence or presence of m-tyrosine. Immunoblot analyses of 5-day-old Arabidopsis thaliana (Col-0) plants grown in the absence or presence of m-tyrosine. For the quantification of the relative abundances of organellar proteins in the plants, different amounts of total proteins extracted from control (untreated) and m-tyrosine-grown seedlings were loaded and separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). After electrophoresis, the gel was stained with Coomassie Blue (A), or transferred to a PVDF membrane for immunobloting (B). The blots were probed with polyclonal antibodies raised to different organellar proteins (see Table S1), as indicated in the right panel. Detection was carried out by chemiluminescence assays after incubation with horseradish peroxidase (HRP)-conjugated secondary antibody. These include the mitochondrial proteins, ATP-synthase subunit 1 (AtpA), subunit CA2 of complex I (g-type carbonic anhydrase-like subunit 2) (Perales et al., 2005), and NADH dehydrogenase subunit 9 (NAD9) (Lamattina et al., 1993), the Rieske iron-sulfur protein (RISP) of complex III (cytochrome c reductase) (Carrie et al., 2010), cytochrome oxidase subunit 2 (COX2) of complex IV (C-IV), alternative oxidase (AOX1a) (Elthon et al., 1989), the external NADH-dehydrogenase subunit 2 (NDB2) (Carrie et al., 2008), and the voltage-dependent anion channel (VDAC, or Porin), and the plastidial Photosystem I P700 chlorophyll a apoprotein A1 (PsaA), photosystem II CP43 reaction center protein (PsbC), and photosystem II D2 protein (PsbD). Arrows in the blots indicate to AtpA (55 kDa), CA2 (28 kDa), NAD9 (23 kDa), COX2 (30 kDa), RISP (23 kDa), VDAC (29 kDa), AOX1/2 (36 kDa), NDB2 (50 kDa), PsaA (52 kDa), PsbC (43 kDa), PsbD (39 kDa). Hybridization signals were analyzed by chemiluminescence assays after incubation with HRP-conjugated secondary antibody. The intensities of protein signals in panel B were evaluated using the ImageJ software (Jensen, 2013). The data shows the relative accumulation of each protein in untreated versus m-tyrosine-treated seedlings (panel C).
The values are means of three biological replicates ± SE. FIGURE 6 | Relative accumulation of native organellar complexes in 5-day-old Arabidopsis plants grown in the absence or presence of m-tyrosine. Blue nativepolyacrylamide gel electrophoresis (BN-PAGE) of crude organellar preparations was performed according to the method described in Pineau et al. (2008). Equal amounts of crude membranous fractions (equivalent to 50 mg fresh-weight seedlings), were obtained from 5-day-old Arabidopsis seedlings, solubilized with 1.5% n-dodecyl-b-d-maltoside (DDM), and the organellar complexes were resolved by BN-PAGE. (Panel A) shows stained PAGE, followed by in-gel activity assay of complex I, according to the method described in Eubel et al. (2005). For immunodetections (panel B), proteins were transferred from the native gels onto a polyvinylidene difluoride (PVDF) membrane (see Materials and Methods) and then probed with specific antibodies (see Table S1), as indicated below each blot. Arrows indicate to the native mitochondrial complexes I (~1,000 kDa), III (dimer,~500 kDa), IV (~220 kDa), V (~600 kDa), and the plastidial PSI complex (~250 kDa). Asterisks in the COX2 and PsaA panels indicate the presence of lower or higher (respectively) molecular weight particles of unknown origins. Hybridization signals were analyzed by chemiluminescence assays after incubation with horseradish peroxidase (HRP)-conjugated secondary antibody. The intensities of protein signals in (panel B) were evaluated using the ImageJ software (Jensen, 2013). The data shows the relative accumulation of each protein in untreated versus m-tyrosine-treated seedlings. The data shows the relative accumulation of each protein in untreated versus m-tyrosine-treated seedlings (panel C). The values are means of three biological replicates ± SE. misacylation of m-tyrosine (Klipcan et al., 2009;Klipcan et al., 2010), the proteomic data indicated that Phe ! m-Tyr/Tyr exchanges occur more frequently during the synthesis of organellar proteins, and in particularly of those translated within the chloroplasts (Table 3 and Supplementary Data File 1).

Mutants Affected in FtsH2 Show Hypersensitivity to m-Tyrosine
Our results (Table 3) indicate that m-tyrosine can be delivered into the proteomes of plants, an abnormal cellular condition that can be associated with increased protein turnover (Rodgers et al., 2002). FtsH is a membrane-bound ATP-dependent zinc metalloprotease complex, which has been characterized in bacteria and organelles of eukaryotic cells (Kato and Sakamoto, 2018 and references therein). While Escherichia coli harbors a single FtsH gene, Arabidopsis plants encode 12 related FtsHs that play key roles in protein quality control, as removal of damaged and abnormal proteins (Ostersetzer and Adam, 1997;Adam and Ostersetzer, 2001;Zaltsman et al., 2005;Sakamoto, 2006). Nine FtsHs reside in the plastids, whereas the other three paralogs are located within the mitochondria (Kato and Sakamoto, 2018). In chloroplasts, FtsH is made of different heterohexamer complexes (Moldavski et al., 2012). FtsH2 (At2g30950, also denoted as VAR2) is the most abundant isoform within the plastids. var2 mutants show aberrant leaf development with severe variegation phenotypes, which are related to altered chloroplast development (Chen et al., 2000;Takechi et al., 2000;Sakamoto et al., 2003).
Based on these data we speculated that var2 mutants might show higher sensitivity to m-tyrosine, due to the accumulation of abnormal and damaged proteins within the plastids. To examine this hypothesis, we analyzed the root development and early seedling establishment of Arabidopsis wild-type (Col-0) and var2 mutants (Takechi et al., 2000;Sakamoto et al., 2003;Zaltsman et al., 2005), germinated in the absence or presence of various concentrations of m-tyrosine. The data indicated that var2 mutants seem to be more susceptible to inhibition by mtyrosine (Figure 7). In particular, the Phe-analog, m-tyrosine, had a strong effect on chloroplast biogenesis of var2 plants, as evident by the cotyledon chlorosis ( Figure 7A) and reduced chlorophyll content ( Figure 7B). Reduced leaf pigmentation in var2 was already observed at m-tyrosine concentrations ≥ 1 mM, while no chlorophyll was detected when the mutants were grown in m-tyrosine concentrations above 5.0 mM (Figure 7).

Mutants Affected in Autophagy Show Higher Sensitivity to m-Tyrosine Toxicity
Autophagy is an essential process in eukaryotic cells that involves a control degradation of large substrates including protein aggregates, pathogens, and organelles (reviewed by e.g., Avin-Wittenberg et al., 2018). Forward genetics led to the identification of many autophagy-related loci in Arabidopsis, denoted as ATG genes (Young et al., 2019). Among these genes are ATG5 (At5g17290) and ATG7 (At5g45900), which their mutants show a full inhibition of autophagy (Barros et al., 2017 and references herein). When Arabidopsis wild-type plants (Col-0) and atg mutants (Hofius et al., 2009;Yoshimoto et al., 2009;Avin-Wittenberg et al., 2015) were germinated in the absence or presence of various concentrations of m-tyrosine ( Figure S5A), the atg mutant-lines were found to be more susceptible to high concentrations of m-tyrosine (i.e., > 10 mM). Yet, the phenotypes of the autophagy mutants were undistinguishable from those of wild-type (Col-0) plants at lower concentrations of the Phe-analog ( Figure S5B). These results seem to be correlated with the organellar biogenesis defects of Arabidopsis plantlets germinated in the presence of m-tyrosine (Figures 4, S1, and S3, Table 2).

m-Tyrosine Is a Strong Allelochemical That Affects Plants Early Growth and Development
m-Tyrosine is a non-proteinogenic analog of the aromatic amino acids phenylalanine and tyrosine, which acts as a natural herbicide, affecting the post-germination development and early establishment of nearby plant life (Bertin et al., 2007). Published data indicates that m-tyrosine is significantly more toxic to plants than its structural related isomers, o-and p-tyrosine (Bertin et al., 2007). Our data provide further insights into m-tyrosine-mediated phytotoxicity. Under different conditions, and more particularly under oxidative stress, m-tyrosine is accumulating to higher levels in the cell (Ipson and Fisher, 2016). m-tyrosine is directly generated by hydroxyl radical attack of phenylalanine when levels of reactive oxygen species are elevated (Rodgers and Shiozawa, 2008), or might be produced enzymatically, as indicated in a few plant species (Mothes et al., 1964;Bertin et al., 2007;Huang et al., 2012). In F. rubra, m-tyrosine biosynthesis occurs through the hydroxylation of phenylalanine in the root tips. The allelochemical is then released to the biosphere, were it inhibits the post-germination development of nearby plants (Huang et al., 2012). Arabidopsis seedlings germinated in the presence of mtyrosine show slow seedlings establishment, altered root elongation, and reduced biomass (Figures 3, S1 -S3, and S5, and Bertin et al., 2007). The severity of these phenotypes are correlated with the levels of m-tyrosine in the growth media, with calculated IC 50 value of 2.365 mM ( Figure S1). At m-tyrosine concentrations higher than 40~80 mM, the plants show postgermination arrest of embryo development ( Figure S1). Apparently, while m-tyrosine had no obvious effect on the photosynthetic activity of lettuce seedlings (Bertin et al., 2007), at concentrations above 40 mM Arabidopsis seedlings showed signs of leaf (i.e., cotyledon) chlorosis ( Figure S1). Analyses of 5 day-old seedlings grown in the absence or presence of 10 mM mtyrosine indicated to reduced chlorophyll production ( Table 2), altered organellar biogenesis ( Figure 4) and reduced photosynthetic activities (Table 2 and Figure S4). Our data also indicate to a reduction in oxygen consumption when the plants are grown in the presence of m-tyrosine ( Table 2). Respiration in some CI-deficient Arabidopsis mutants can be upregulated, a phenomenon that has been associated with the induction of the alternative respiratory pathway, mediated by non-phosphorylating NAD(P)H dehydrogenases (NDs) and alternative oxidases (AOXs) (see e.g., Kuhn et al., 2015). We speculate that although AOX and ND2 are upregulated in mtyrosine treated plants, the high energy demand (i.e., ATP production) during germination and early seedling metabolism necessitates the electron transport via the canonical respirasome mediated pathway. Accordingly, in different plant species, the affinity of AOX is significantly lower compared to that of COX to O 2 (see e.g., van Dongen et al., 2011). The m-tyrosine-mediated defect phenotypes can be partially rescued by the addition of sucrose to the growth media ( Figure S3), in a similar manner to the effects of sugar on Arabidopsis mutants affected in chloroplast or mitochondria biogenesis (see e.g., Keren et al., 2012;Van Dingenen et al., 2016).
Based on the experimental results and published data we consider two different cellular pathways in which m-tyrosine inhibits the early establishment of Arabidopsis plants. Together, the analyses of the steady-state levels of free amino acids in control and m-tyrosine-grown seedlings ( Figure 1 and Table  S2), the effects of various amino acids added to the growth media on m-tyrosine toxicity (Figure 2), the higher-sensitivity of adt mutants to m-tyrosine (Figures 3 and S2) and lower sensitivity of adt2-1D mutant which accumulates higher levels of Phe (Cho et al., 2007), strongly suggest that m-tyrosine affects the biosynthesis of phenylalanine in early-developing Arabidopsis seedlings. Another molecular mechanism by which m-tyrosine can affect Arabidopsis vitality involves the misincorporation of this aromatic non-proteinogenic analog into the plant proteome instead of Phe, and in particular to various organellar proteins. These are evident by proteomic data (Table 3 and Supplemental  Table S3, doi: 10.6084/m9.figshare.11627211), altered organellar biogenesis and functions ( Table 2, Figures 4-6 and S4), the hypersensitivity of var2 mutants to m-tyrosine (Figure 7), as well as previous data indicating that tRNA Phe can be charged with mtyrosine, in vitro (Gurer-Orhan et al., 2006;Klipcan et al., 2009).

Phenylalanine Biosynthesis Is Affected in Arabidopsis Plants Treated With m-Tyrosine
Published data Höhner et al., 2018) and our own results (Figures 2 and S2) strongly suggest that m-tyrosine toxicity is affected by the (bio-) availability of Phe. Accordingly, exogenous Phe can partially rescue the growth defect phenotypes mediated by m-tyrosine (Figure 2). Seedlings grown in the presence of the amino acid methionine (Met) are also less susceptible to inhibition by m-tyrosine ( Figure 2, and Bertin et al., 2007). Therefore, it remains possible that Met plays a nonredundant role through indirectly affecting Phe metabolism, or by scavenging reactive oxygen species (ROS) that may accumulate to higher levels in the m-tyrosine treated plants. Methionine may also act as catalytic antioxidants, protecting protein where they reside, as well as other molecules within the cells (Luo and Levine, 2009). Accordingly, Arabidopsis seedlings germinated in the presence of m-tyrosine show altered organellar activities (Table 2 and Figure S4) and accumulate high levels of AOX ( Figure 5).
An effect of m-tyrosine on Phe biosynthesis is further supported by genetic analyses. Arogenate dehydratase (ADT) is a key enzyme in Phe biosynthesis, converting arogenate into Phe. ADT activity in plants is positively regulated by Tyr and negatively regulated by Phe (Jung et al., 1986;Siehl and Conn, 1988). While Arabidopsis adt mutants show a hipper sensitivity to m-tyrosine (Figures 3 and S2), the adt2-1D mutant-line that accumulates higher levels of Phe, most likely due to a disruption in the feedback inhibition (i.e., allosteric effects) of the ADT enzyme (Cho et al., 2007), are less susceptible to m-tyrosine toxicity . Accordingly, analyses of the steady-state levels of free amino-acids indicate that the phytotoxic effects of m-tyrosine correlate with altered amino acid metabolism, in particularly phenylalanine (Figure 1 and Table S2). Taken together, these data suggest that m-tyrosine reduces the biosynthesis of Phe, probably by directly affecting the activity of ADT. However, these assumptions need to be validated experimentally.

Misincorporation of Free m-Tyrosine Into the Proteome of Arabidopsis Plants
Another molecular mechanism, by which m-tyrosine affects plants, may involve the incorporation of unnatural chemical groups (i.e., non-canonical amino acids) into proteins (Smith and Fowden, 1968), which are expected to affect the activities of proteins and enzymes, or to interfere with the functions of native side chains. Proteomic analyses of plants germinated in the absence or presence of m-tyrosine indicate that this analog is misincorporated into the plant proteome ( Table 3, and  Supplemental Table S3, doi: 10.6084/m9.figshare.11627211). These data are also in agreement with earlier reports suggesting that m-tyrosine may be wrongly delivered into proteins in bacteria (Aronson and Wermus, 1965) and animals (Gurer-Orhan et al., 2006). In Arabidopsis, the majority of the Phe ! m-Tyr/Tyr replacements are identified in proteins that are expressed within the organelles, in particular plastidial ones (Table 3). Furthermore, analyses of mutant plants affected in FtsH2, a major component of the chloroplast protein quality control system (Kato and Sakamoto, 2018), resulted with an increased sensitivity to m-tyrosine, an effect that is expected to be associated with the accumulation of damaged and abnormal organellar proteins (Ostersetzer and Adam, 1997;Adam and Ostersetzer, 2001;Sakamoto, 2006). Also, mutant plants affected in autophagy are more severely affected in seedling establishment and show strong cotyledon bleaching at mtyrosine concentrations above 10 mM ( Figure S5), although the effects at high m-tyrosine could rather relate to general stress responses rather than to a specific cellular damage.
How can m-tyrosine be incorporated into the plant proteome? Aminoacyl tRNA synthases (aaRSs) ensure the integrity of the translation of the genetic code, by covalently attaching an appropriate amino acid to the corresponding nucleic acid adaptor tRNA molecule. Although aaRS are known to be highly specific, mistakes in the recognition may still occur, due to stereo-chemical similarities shared by some native amino acids and their nonproteinogenic analogs Chaliotis et al., 2016;Mohler and Ibba, 2017). A key repair mechanism to ensure the acetylation of the correct amino acid involves an intrinsic editing activity, by which misacylated tRNAs can be hydrolyzed. In plants, the attachment of phenylalanine to tRNA Phe is catalyzed by two major forms of PheRSs. These include i) a heterotetrameric (2·ab) form, which is structurally related to the bacterial 2·ab enzymes and acts in the cytosol, and ii) a degenerated monomeric form (contains fragments of a and b), which dually functions in the plastid and mitochondria of plants (Duchêne et al., 2005;Klipcan et al., 2010).
The aromatic amino acids Phe, Tyr, and their cognate nonproteinogenic m-tyrosine analog, are distinguished by a single hydroxyl group at the aromatic ring, and thus discrimination between these molecules may not always be accurate (Roy et al., 2004;Kotik-Kogan et al., 2005;Ling et al., 2007;Klipcan et al., 2009;Moghal et al., 2014;Popp et al., 2015). To maintain accuracy during translation, the PheRS enzyme possess an editing domain that hydrolyzes misacylated tRNA Phe (Roy et al., 2004;Kotik-Kogan et al., 2005;Ling et al., 2007). Structural and biochemical assays suggest that bacterial and both eukaryotic PheRS forms can load mtyrosine into tRNA Phe (Klipcan et al., 2009;Moghal et al., 2016). However, only the bacterial enzyme seems to be able to efficiently hydrolyze the misacylated m-tyrosine (Klipcan et al., 2009). In the cytosolic PheRS form of eukaryotes, discrimination against mtyrosine mainly occurs during the aminoacylation step (Klipcan et al., 2009). Structural and biochemical studies suggest that organellar PheRS is not able to discriminate against m-tyrosine and lacks editing activity (Klipcan et al., 2009;Klipcan et al., 2010). It is, therefore, expected that the dually-localized organellar PheRS enzyme will be more prone to errors, opening up the rout for delivery of misacylated tRNA Phe -m-tyrosine to the organellar ribosomes.
Why Are Plants More Sensitive to m-Tyrosine Than Other Life Forms? Currently, we are not able to provide with a definitive answer. Plants are highly susceptible to m-tyrosine [Figures S1 and S3, and (Bertin et al., 2007)], with IC 50 (i.e., root growth inhibition) at the low micromolar range ( Figure S1), about 100~1,000 times more toxic to plants than to bacterial (Smith et al., 1964;Aronson and Wermus, 1965), yeast (Bullwinkle et al., 2014), or animal cells (Gurer-Orhan et al., 2006). Phe is an essential a-amino acid that is produced through the shikimate pathway, which is present in bacteria, fungi, and plants, as well as in some protozoa, but not in Animalia who need to obtain aromatic amino acids from their diet. While animals cannot produce Phe, it was previously shown that m-tyrosine can be biodegraded in several microorganisms, which can utilize this metabolite as the sole source of carbon, nitrogen, and energy (Khan et al., 2013 and references therein). In addition, the levels of free Phe, seem to be notably higher in animal cells (about 55 to 60 mM in human cells) then in plants (i.e., Arabidopsis,~15 mM) (Kaufman, 1999;Voll et al., 2004).
As indicated above, there are also differences in the way that m-tyrosine can be delivered into the proteomes of different organisms. While prokaryotes encode a single heterotetrameric PheRS that is able to efficiently hydrolyze the misacylated m-tyrosine (Klipcan et al., 2009), eukaryotes possess two PheRS isoforms, one of which resides within the organelles (Klipcan et al., 2009). In plants, the organellar PheRS isoform resides in both the plastids and mitochondria (Duchêne et al., 2005). Plastid biogenesis seems in particularly amenable to m-tyrosine toxicity. This is apparent by altered chloroplast morphologies, reduced grana stacking, the presence of many plastoglobuli ( Figure 4) and lower photosynthetic rates (Table 2 and Figure S4), associated with Arabidopsis plantlets germinated in the presence of mtyrosine, as well as the fact that many polypeptides that undergo Phe ! (m-)Tyr exchanges are identified as plastidial proteins (Table 3). These may relate to the high complexity of the translational apparatus in plant chloroplasts (Zoschke and Bock, 2018). We speculate that the effects of mtyrosine on Phe biosynthesis together with the ability of the organellar PheRS to deliver the non-proteinogenic analog to both mitochondrial and plastidial proteins makes plants highly sensitive to m-tyrosine.
In summary, this work is founded on earlier reports that indicate that m-tyrosine is highly toxic to different plant species. Based on the data we consider two related molecular mechanisms by which m-tyrosine can affect angiosperm's root development, radicle elongation, and early establishment. These involve (a) a direct interference in the metabolism of various amino acids, and in particular Phe, and (b) the misincorporation of the Pheanalogue to some cytosolic and mainly the plant organellar proteins. The correct transfer of information from the genome to proteins is pivotal for the development and physiology of plants, as well as for other organisms. The specificity of aminoacyl-tRNA synthetases (aaRSs) is key for ensuring the proper decoding of the genetic information into proteins. Distinguishing between closely related amino acids and their analogs by aaRSs is not always accurate, leading to errors in tRNA loading and hence for the translation of aberrant polypeptides. When a non-cognate amino acid is activated, some aaRSs employ an editing mechanism leading to the hydrolysis of the misacylated tRNA molecule. Biochemical and structural studies suggest that organellar PheRS forms lack editing activity. The two effects (i.e., altered Phe biosynthesis and increased misincorporation of m-tyrosine) are related to one another, i.e., reduced availability of the natural amino acid Phe would ultimately result an increased misincorporation of its analogous compound, m-tyrosine, in particularly into organellar proteins (Klipcan et al., 2009). The effects of m-tyrosine are in particularly prevalent during early development, this in accordance with mutant plants affected in organellar biogenesis, which also show early growth retardation, inhibition of root development, and altered seedling establishment phenotypes (see e.g., Keren et al., 2012;Van Dingenen et al., 2016).

DATA AVAILABILITY STATEMENT
The datasets generated for this study can be found in the https:// doi.org/10.6084/m9.figshare.11627211.

AUTHOR CONTRIBUTIONS
HZ: Experimental design, plant growth and analysis, analyses of the protein profiles, BN-PAGE assays. NM: Assisted in growth and analyses: HM and TA-W: GC-MS analyses of plant extracts. LK: Assisted with experimental design, co-corresponding author. OO-B: Principal Investigator, MS preparation and corresponding authors.

FUNDING
Research at the TA-W group is funded by the Israel Science Foundation (ISF grant no. 1899/16). This work was supported by grants from the 'Israel Ministry of Agriculture, to LK and OO-B and the 'Israel Science Foundation' (ISF grant no. 741/15) to OO-B.

ACKNOWLEDGMENTS
We thank Mr. Yuval Kolodny for his help with the analyses of photosynthetic measurements, the Arabidopsis biological resource center for providing wild-type Arabidopsis thaliana var Columbia (Col-0) seeds, Prof. Zach Adam for providing Arabidopsis var2 mutant-lines, and Prof. Jirong Huang (Shanghai Normal University) for providing us Arabidopsis adt mutant lines.