A Shotgun Proteomic Approach Reveals That Fe Deficiency Causes

Subscriber access provided by University of Sussex Library

Article

A shotgun proteomic approach reveals that Fe deficiency causes marked changes in the protein profiles of plasma membrane and detergent resistant microdomain preparations from Beta vulgaris roots Elain Gutierrez-Carbonell, Daisuke Takahashi, Sabine Lüthje, Jose A Gonzalez-Reyes, Sebastien Mongrand, Bruno Contreras-Moreira, Anunciación Abadía, Matsuo Uemura, Javier Abadía, and Ana-Flor López-Millán J. Proteome Res., Just Accepted Manuscript • DOI: 10.1021/acs.jproteome.6b00026 • Publication Date (Web): 20 Jun 2016 Downloaded from http://pubs.acs.org on June 23, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Proteome Research is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

A shotgun proteomic approach reveals that Fe deficiency causes marked changes in the protein profiles of plasma membrane and detergent resistant microdomain preparations from Beta vulgaris roots Elain Gutierrez-Carbonell1, Daisuke Takahashi2,3, Sabine Lüthje4, José Antonio GonzálezReyes5, Sébastien Mongrand6, Bruno Contreras-Moreira 7,8, Anunciación Abadía1, Matsuo Uemura3, Javier Abadía1, Ana Flor López-Millán9* 1

Plant Stress Physiology Group, Plant Nutrition Department, Aula Dei Experimental Station, CSIC, Apdo. 13034, 50080 Zaragoza, España; 2United Graduate School of Agricultural Sciences, Iwate University, Morioka 020-8550, Japan; 3Cryobiofrontier Research Center,

Faculty of Agriculture, Iwate University, Morioka 020-8550, Japan; 4University of Hamburg, Biocenter Klein Flottbek Ohnhorststrasse 18, 22609 Hamburg, Germany; 5Departamento de Biología Celular, Fisiología e Inmunología, Universidad de Córdoba, Campus de Rabanales. Edificio Severo Ochoa; Córdoba, Spain; 6Laboratoire de Biogenèse Membranaire, UMR 5200 CNRS-Université Bordeaux Segalen, Bâtiment A3, INRA Bordeaux Aquitaine, 71 Avenue Edouard Borlaux, CS 20032, F-33140 Villenave d’Ornon; 7Laboratory of Computational and Structural Biology, Aula Dei Experimental Station, CSIC, Apdo. 13034, 50080 Zaragoza, España; 8Fundación ARAID, calle María de Luna 11, 50018 Zaragoza, España; 9USDA-ARS Chindren’s Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, 1100 Bates st, Houston, TX 77030, USA. *Corresponding author: Ana Flor López Millán. USDA-ARS Childrens Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, 1100 Bates Street, Houston, TX 77030, USA. Tel: 713-798-7086; FAX: 713-798-7098, email: [email protected]

Running tittle: Shotgun proteomics of Fe-deficient root plasma membrane 1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract In the present study we have examined, using label-free shotgun proteomic analysis, the effects of Fe deficiency on the protein profiles of highly pure sugar beet root PM preparations and detergent resistant membranes (DRMs), the latter as an approach to study microdomains. Altogether, 545 proteins were detected, with 52 and 68 of them changing significantly with Fe deficiency in PM and DRM, respectively. Functional categorization of these proteins showed that signaling, general and vesicle-related transport accounted for approximately 50% of the differences in both PM and DRM, indicating that from a qualitative point of view, changes induced by Fe deficiency are similar in both preparations. Results indicate that Fe deficiency has an impact in phosphorylation processes at the PM level and highlights the involvement of signaling proteins, especially those from the 14-3-3 family. Lipid profiling revealed Fedeficiency induced decreases in phosphatidic acid derivatives, which may impair vesicle formation, in agreement with the decreases measured in proteins related to intracellular trafficking and secretion. The modifications induced by Fe deficiency in the relative enrichment of proteins in DRMs revealed the existence of a group of cytoplasmic proteins that appears to be more attached to the PM in conditions of Fe-deficiency.

Keywords: Detergent resistant microdomain; Iron deficiency; Lipidomics; Plasma membrane; Proteomics; Sugar beet.

2 ACS Paragon Plus Environment

Page 2 of 45

Page 3 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Introduction The plasma membrane (PM) is a biological structure that surrounds each cell, acting as an active interface between the cytoplasm and the environment. The PM consists of a phospholipid bilayer housing integral and peripheral proteins needed for the functionality of this cell compartment. The initial definition of PM as a “homogeneous and dynamic fluid mosaic” has been revised and amended along the years, to include characteristics such as asymmetry, limited mobility of proteins by interactions with intra- and or extracellular components such as the cytoskeleton or cell wall, and changes in protein density at the membrane surface level.1 The concept of homogeneity has also been debated and there is evidence supporting that membrane microdomains may exist, whose composition differs from the PM continuum.1 The PM is essential for basic cell functions such as cellular differentiation and proliferation, exchange of metabolites and information and cell wall biosynthesis, and it is a primary target for responses to biotic and abiotic stresses.2 In plants, PM proteins play a key role in Fe acquisition by roots. Plants are broadly classified into two groups depending on the strategy displayed for root Fe uptake at the PM level, with monocotyledonous plants using a mechanism that includes Fe(III) chelation by root-secreted phytosiderophores followed by specific uptake of the Fe(III)-complexes, whereas dicotyledonous species need to reduce Fe(III) prior to its uptake.3,4 Roots from dicotyledonous plants undergo both morphological and physiological modifications when Fe is scarce in the soil solution, and these changes increase the root Fe-uptake capacity.5 At the root PM level, major responses of dicotyledonous plants include an induction of the Fe-reduction based uptake machinery consisting on the Fe(III)-reductase enzyme (FRO, Ferric Reductase Oxigenase) and an Fe(II)transporter (IRT, Iron Regulated Transporter)6,7 and an enhanced proton extrusion capacity



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

mediated by a PM H+-ATPase.8 Other Fe deficiency-induced response where root PM plays an important role is the secretion of different organic compounds, including carboxylates, phenolics and flavins.9-13 The PM transporter involved in the excretion of phenolic compounds, ABCG37, has been identified only recently in Arabidopsis,14-15 whereas the transporters responsible for the secretion of carboxylates and flavins upon Fe deficiency still remain largely unknown. The PM is also important for Fe unloading into the vascular system and studies with mutants lacking the PM citrate transporter FRD have shown that FRD-mediated citrate efflux is required for root to shoot Fe delivery.16,17 While the essential components for the root Fe uptake have been almost deciphered, knowledge about other possible modifications induced by Fe-deficiency at the PM level is still very limited. Proteomic approaches are useful to provide information on global changes upon nutritional stresses such as Fe deficiency18-20, and indeed effects of Fe deficiency on the proteomes of different plant organs and subcellular compartments have been reported, mostly using 2-DE approaches.21 However, membrane proteins pose an important challenge for 2-DE proteomic analyses for several reasons, including their high hydrophobicity, low solubility, low abundance and chemical heterogeneity. In the present study we have examined the effects of Fe deficiency on the PM proteome by means of label-free shotgun analyses, a technique that is better suited for membrane proteins than the 2-DE gel-based approaches. Furthermore, we have also studied changes induced by Fedeficiency in detergent resistant membranes (DRMs) as an approach to study changes in the composition of PM microdomains. However, it should be noted that a relationship between membrane proteins identified in DRMs and their localization in specific membrane microdomains has yet to be found. A better understanding of the changes induced by Fe 4 ACS Paragon Plus Environment

Page 4 of 45

Page 5 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


deficiency at the PM, the point of Fe entry into the plant, may help develop stress tolerant crops and minimize the losses in crop yield and quality associated to this stress. For instance, Fe deficiency is a major agronomical problem in areas with calcareous soils such as the Mediterranean basin and the Northern part of the Midwest in the US, where it causes economical losses in fruit tree and soybean production, respectively.22-25

Experimental Procedures Plant Material and Experimental design Sugar beet (Beta vulgaris L. cv. Orbis) was grown in a growth chamber with a photosynthetic photon flux density (PPFD) of 400 µmol m-2 s-1 PAR, 80% relative humidity and a photoperiod of 16 h, 23º C/8 h, 18º C day/night regime. Seeds were germinated and grown in vermiculite for two weeks. Seedlings were grown for an additional two-week period in half-strength Hoagland nutrient solution 26 with 45 µM Fe(III)-EDTA and then transplanted to 20 L plastic buckets (four plants per bucket) containing half-strength Hoagland nutrient solution with either 45 µM Fe(III)EDTA as a control or 0 µM Fe(III)-EDTA, pH 7.5 and 1 g L-1 CaCO3 as the Fe-deficient treatment. Roots were harvested 9-10 days after imposing treatments. Label free nLC-ESIMS/MS analysis was made using six independent biological replicates for PM (each one obtained using a PM preparation per treatment from a different batch of plants) and four biological and two technical replicates for DRM (each one obtained using a PM preparation per treatment from a different batch of plants). Polar lipid profiling was performed using five PM preparation replicates of each treatment. Glycerolipid, sterol and sphingolipid distribution was assessed using five PM preparations replicates of each treatment.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Plasma Membrane Preparation Plasma membrane preparations were obtained from a root microsomal fraction by differential centrifugation in an aqueous-polymer two-phase system.27,28 Roots were homogenized using a mixer (Oster, Madrid, Spain; three times of 10 s at maximum speed) in a medium containing 250 mM Sucrose, 2 mM DTT, 4 mM EDTA, 5 mM ascorbate, 1 mM PMSF, 5% PVP, 0.2% BSA, and 0.2% casein in 50 mM Tris-HCl, pH 8.0. A ratio of 1 g of root fresh weight to 4 mL of homogenization medium was used. The homogenate was filtered through two layers of Miracloth (Calbiochem, Darmstadt, Germany), and the filtrate centrifuged at 10,000 g for 10 min. The pellet was discarded and the supernatant centrifuged at 50,000 g for 60 min to pellet the microsomal fraction. The microsomal fraction was resuspended in 9 mL of a medium containing 200 mM sucrose, 5 mM ascorbate, 3.5 mM KCl and 5 mM phosphate buffer, pH 7.8. Then, the solution was layered over an aqueous-polymer two-phase system, to yield a 36 g system with a final concentration of 6.5% dextran T500, 6.5% PEG 3350, 3.5 mM KCl, 5 mM Na ascorbate, 200 mM sucrose, and 5 mM phosphate buffer, pH 7.8. The partition was carried out three times at 1,500 g for 10 min.28 The final upper PEG phase was diluted 6-fold with PM suspension medium containing 50 mM HEPES, pH 7.0, 250 mM sucrose, and 5 mM Na ascorbate, and centrifuged at 50,000 g for 60 min. The resulting pellet was resuspended in the same buffer and volume. After a second centrifugation at 50,000 g for 60 min, the pellet, containing the PM preparation, was finally suspended in PM suspension medium and stored in aliquots at -80°C. Protein concentration was determined by the Bradford method, using BSA (Bio-Rad, Munich, Germany) as a standard.


Page 6 of 45

Page 7 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The purity of PM preparations was assessed using enzymatic activities characteristic of PM, tonoplast and mitochondria and by means of electron microscopy (see below). The activity of the PM-associated, vanadate-sensitive and Mg2+-dependent ATPase was assayed by measuring the release of phosphate from ATP at pH 6.5, using the formation of a colored phosphomolybdic complex as previously described.29 The activities of nitrate-sensitive ATPase 30 and cytochrome c oxidase 31 were used as markers for tonoplast and mitochondria contamination, respectively.

Detergent Resistant Membranes (DRM) Preparation Preparation of DRM fractions was carried out according to the method of Peskan et al.32 as described previously.33 Briefly, 800 µg of PM preparations were pelleted by centrifugation at 231,000 g for 35 min and resuspended in 1.08 mL of 10 mM MOPS-KOH (pH 7.3), 2 mM EGTA, 250 mM sucrose. An aliquot (120 µL) of 10% (w/v) Triton X-100 in TED buﬀer consisting in 50 mM Tris-HCl (pH 7.4), 3 mM EGTA, 1 mM DTT was added to the PM suspension (1:15, protein:detergent ratio) and the PM-TX-100 mixture was incubated on ice for 30 min. Following this incubation, 4.8 mL of 65% (w/w) sucrose in TED buffer were immediately added and mixed well. To obtain the DRM fractions, 48, 35, 30, and 5% (w/w) sucrose-TED solutions were subsequently overlaid on the PM-TX-100 mixture and the whole was centrifuged at 141,000 g and 4°C for 20 h, using a swing-type rotor P28S (Hitachi Koki, Tokyo, Japan). After centrifugation, the white band at the interface of the 35 and 48% sucrose layers (Fig. S1) was collected, diluted with the PM suspension medium, and centrifuged at 231,000 g for 50 min. This fraction was designated as DRM. The protein concentration of the DRM preparations was measured as described above for the PM preparations.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Electron Microscopy and Morphometry On-section staining with phosphotungstic acid (PTA) was used as an additional purity check for PM.34 Samples of microsomal fractions and purified PM preparations were fixed in a mixture of 2.5% glutaraldehyde and 2% paraformaldehyde in sodium cacodylate buffer (0.1 M, pH 6.8) for 12 h and postfixed in 1% osmium tetroxide in the same buffer. After dehydration in an ascendant ethanol series, pieces were transferred to propylene oxide and sequentially infiltrated in EMbed812 resin (EMS; USA). We used the sequence propylene oxide-resin 2:1, 1:1 and 1:2 throughout a 24 h period. Afterwards samples were transferred to pure resin for 24 hour. Blocks were formed in fresh resin and then allowed to polymerize for 48 hour at 65°C. Ultrathin (40-60 nm width) sections were obtained using an Ultracut Reichert ultramicrotome (Reichert, Vienna, Austria), placed on Ni grids, incubated for 30 min in 1% periodic acid and then stained with 1% PTA in 10% chromic acid for 10 min. Sections were viewed and photographed in a JEOL JEM 1400 electron microscope (JEOL, Tokyo, Japan). At least 12 photographs were taken from this material for each treatment and the purity of the PM was estimated by planimetry as described previously.35 Label free nano-liquid chromatography-tandem mass spectrometry (nLC-ESI-MS/MS) Label free nLC-ESI-MS/MS analysis was made using six independent biological replicates for PM (each one obtained using a PM preparation per treatment from a different batch of plants) and four biological and two technical replicates for DRM (each one obtained using a PM preparation per treatment from a different batch of plants). Sample preparation for label free nLC-ESI-MS/MS was carried out according to previously described methods.33,36 Briefly, 4.5 µg 8 ACS Paragon Plus Environment

Page 8 of 45

Page 9 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


of purified PM and DRM preparations were subjected to 1-DE to remove non-protein compounds. The resulting gel bands were cut into four pieces, proteins were in gel digested with trypsin, and peptides were subsequently extracted. Peptide solutions were concentrated in a trap column (Lcolumn Micro 0.3 x 5 mm; CERI, Japan) using an ADVANCE UHPLC system (Michrom Bioresources, Aubum, CA, U.S.A.). Elution was carried out with 0.1% (v/v) formic acid in acetonitrile (ACN) and concentrated peptides were separated in a Magic C18 AQ nano column (0.1 x 150 mm; Michrom Bioresources) using a linear gradient of ACN (from 5% to 45%) and a flow rate of 500 nL min-1. Peptide ionization was carried out with a spray voltage of 2 kV using an ADVANCE spray source (Michrom Bioresources). Mass spectrometry analysis was carried out on an LTQ Orbitrap XL (Thermo Fisher Scientific, Waltham, MA, U.S.A.) equipped with Xcalibur software (version 2.0.7, Thermo Fisher Scientific). Normalized collision energy and isolation width were set to 35% and 2 m/z, respectively. Dynamic exclusion was enabled with the following parameters: repeat count, 1; repeat duration, 30 s; exclusion list size, 500; exclusion duration, 60 s; and exclusion mass width, ±0.01 m/z. Under the data-dependent scanning mode, full scan mass spectra were obtained in the range of 400 to 1800 m/z with a resolution of 30,000. Collisioninduced fragmentation was applied to the ten most intense ions with charge states 2+ and 3+ at a threshold above 500. Mass data analysis was performed according to Takahashi et al.36 For semi-quantitative analysis of PM and DRM proteins, experimental raw MS/MS data files were analyzed with Progenesis LC-MS software (version 4.0, Nonlinear Dynamics, Newcastle, U.K.) according to the software’s instructions and a list of peptides was obtained. In this process, a reference run was selected and retention times of each run were aligned. Each feature was normalized based on 9 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 45

the quantitative abundance ratio. Peptide abundances were compared using ANOVA (p < 0.05). Proteomes of five sequenced beet accessions (RefBeet, KDHBv, YMoBv, UMSBv and YTiBv) were downloaded from http://bvseq.molgen.mpg.de/Genome/Download, corresponding to gene models annotated as of February 2013. In addition, all B. vulgaris protein sequences annotated in Uniprot (www.uniprot.org) were retrieved and added to the set, which was subsequently filtered to remove redundant sequences with software CD-HIT 37 with cutoff -c 1.0 and otherwise default parameters. The final non-redundant set contained 82,368 protein sequences. Protein identification was carried out using the full peptide list with the Mascot search engine (version 2.3.02, Matrix Science, London, U.K.) and the non-redundant B. vulgaris 20140811 database (82,368 sequences; 28,127,547 residues). Search parameters were: peptide mass tolerance ±5 ppm, MS/MS tolerance ±0.6 Da, one allowed missed cleavage, allowed fixed modification carbamidomethylation (Cys) and variable modification oxidation (Met). Positive protein identification was assigned with at least two unique top-ranking peptides with scores above the threshold level (p < 0.05). Protein information was exported from Mascot xml format and imported to Progenesis software, which then associates peptide and protein information. Full information about peptides identified and quantified is listed in the Table S1. All detected proteins in the PM and DRM fractions were classified according to their GO: BP annotation and are listed in Table S2. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the Pride partner repository with the data set identifier PXD002981. To assess the effect of Fe deficiency on the PM and DRM protein profiles, we calculated the ratio of normalized abundance between Fe-deficient and Fe-sufficient both for the PM


Page 11 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


preparations and DRM fractions. Only changes with a p ≤ 0.05 (ANOVA) and a –Fe/+Fe ratio ≥ 2 or ≤ 0.5 were considered as statistically significant and biologically relevant, respectively. To reveal the enrichment of proteins in DRM, the DRM/PM ratio was calculated for all detected proteins in each treatment using the normalized abundances. Proteins were then separated into three different classes: DRM-enriched (DRM/PM ≥ 2.0, ANOVA p ≤ 0.05), DRM-depleted proteins (DRM/PM ≤ 0.5, ANOVA p ≤ 0.05), and non-preferentially partitioned (including those with DRM/PM ratios between 2.0 and 0.5 and ANOVA p ≤ 0.05 and those with ANOVA p ≥ 0.05). Lipid Profiling of the Plasma Membranes Polar lipid extraction and ESI-MS/MS analysis was performed in the Kansas Lipidomic Research Center Analytical Laboratory as described elsewhere.38 Briefly, PM preparations (100 µg protein) were extracted with chloroform/methanol (1:1) three times. Lipid samples were analyzed on a triple quadrupole ESI-MS/MS. Polar lipids in each class were quantified in comparison to two internal standards within the class. Five replicates of each treatment for PM preparations were analyzed. The Q-test was performed on the total amount of polar lipids in each head group class and data from discordant samples were removed.39 Paired values were subjected to the t test to determine statistical significance. Full information about polar lipids identified and quantified is listed in the Table S3. In an independent lipid class analysis, the sphingolipid and sterol composition of plasma membrane preparations was analyzed by GC-MS as described elsewhere.40 Full information about specific lipids identified and quantified is listed in the Table S4. Results and Discussion Preparation and Characterization of PM and DRM Fractions 11 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The activities of marker enzymes used to assess the purity of PM indicated that final PM fractions were highly enriched in PM vesicles. First, PM-specific vanadate-sensitive ATPase activity was higher in PM than in microsomal fractions from control (2.8-fold) and Fe-deficient (2.4-fold) preparations (Table S5). On the other hand, contamination of the PM fraction by mitochondrial membranes, assessed by the activity of the cytochrome c oxidase, was not detectable, whereas contamination by tonoplast estimated by the activity of the nitrate sensitive ATPase was always lower than 5.7% (Table S5). Furthermore, electron microscopy of the isolated fractions indicated that PM membranes accounted for 94.4±1.7 and 93.9±2.0% of the total membranes in preparations from control and Fe-deficient plants, respectively (Fig. S2). All these data indicate a high degree of purity in the PM preparations, with most (94%) of the membranes in preparations deriving from PM, although the detection of tonoplast proteins could also be expected. Recovery of DRM from PM fractions (based on protein amount) was 13.8 and 10.7% for control and Fe-deficient samples, respectively. These values are within the range of those described previously for the DRM/PM yield in other plant species.36, 41

Iron deficiency causes marked changes on the protein profile of the PM Altogether, 545 proteins were detected in PM fractions, with 135 of them changing significantly as a result of Fe-deficiency. Among these, 76 proteins were identified and quantified with ≥ 2 peptides and 52 of them presented fold changes above the threshold level (≥ 2-fold or ≤ 0.5fold). In this last group, eight of them increased and 44 decreased in abundance as a result of Fedeficiency (Fig. 1A and Table 1). Functional categorization of the 52 proteins indicated that 12 ACS Paragon Plus Environment

Page 12 of 45

Page 13 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


signaling, general transport and vesicle-related transport account for 52% of the total differential proteins (23, 15 and 14%, respectively; Fig. 2A). Proteins in the PM increasing with Fe deficiency Among the eight PM protein increasing in relative abundance with Fe deficiency, the highest increases (4.2- and 3.6-fold) were found for two proteins identified as HIPL1 and a (-) germacrene D synthase, respectively (Table 1). HIPL1 is a glycosylphosphatidylinositol (GPI)anchored protein (APs), found mostly in the outer leaflet of the PM, which belongs to the glucose dehydrogenase superfamily. This protein putatively reduces quinones using a CH-OH group as hydrogen or electron donor (InterPro domain annotation). GPI-AP proteins have been described to participate in various physiological processes in roots and specifically in cell wall; however their specific function is not yet known.42 The specific role this protein has in the Fe deficiency response would deserve further studies. The (-) germacrene D synthase (EC 4.2.3.75) is involved in the synthesis of the sesquiterpenoid D germacrene. This enzyme has not been shown to be associated to the PM before; however, we could speculate that an increase in the sesquiterpenoid pool may point towards an increase in lipidation post-translational modifications (PMTs) or changes in the PM lipid composition induced by Fe deficiency. The third largest increase (2.9 fold) was measured for phosphoenolpyruvate carboxylase (PEPC; Table 1). This cytosolic enzyme plays a role in anaplerotic C fixation and increases in enzymatic activity and transcript abundance in roots have been widely reported as a response to Fe deficiency in several plant species including sugar beet.43,44 The presence of PEPC in highly pure PM preparations might indicate that it may also have an additional function requiring an association with a still unknown PM target.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Other proteins increasing in relative abundance as a result of Fe deficiency included two scaffold proteins, CASP-like protein Ni6 and tetraspanin-9 (Table 1). CASP-like proteins are four membrane span proteins that mediate deposition of casparian strips by recruiting the lignin polymerization machinery and therefore regulate membrane-cell wall junctions and localized cell wall deposition.45 Iron deficiency has recently been reported to induce changes in cell wall,21,46,47 and from our data we could speculate that these include modifications in the casparian band structure that might ultimately affect lateral diffusion in roots. On the other hand, tetraspanins facilitate assembly of signaling complexes by tethering proteins to integrin heterodimers; these complexes are usually partitioned into specific microdomains at the PM.48 Two-fold increases were also measured for a small GTPase (Rab2BV), a respiratory burst oxidase (RBOHB), and a cytosolic fatty acid binding protein possibly homologous to lipocalins (Table 1). Rab2BV belongs to the Rab family involved in vesicular protein traffic and contains two geranylgeranyl prenylation motifs that allow association to the PM (UniProt entry). RBOHB is a Ca binding flavo-oxidoreductase that produces superoxide radicals in the apoplast and may be involved in the typical tip growth of root hairs characteristic of Fe-deficient roots,49-52 probably mediated by auxin signaling.53 The lipocalin-like protein TIL1 in Arabidopsis is peripherally associated with the PM and is likely to act protecting lipids from peroxidation during oxidative stress elicited by Fe deficiency.54,55 It should be noted that B. vulgaris orthologs of known components of Fe uptake mechanism at the PM level such as the Fe(III)-reductase [AtFRO2 7] and the phloem Fe transporter [AtOPT3 56

] were also found increased in this study (≥ 2-fold increase), although they were among the

proteins only identified or quantified with 1 peptide. Proteins in the PM decreasing with Fe deficiency 14 ACS Paragon Plus Environment

Page 14 of 45

Page 15 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Among the 44 protein decreasing in relative abundance with Fe deficiency, 12 were involved in signaling, eight in transport, six in vesicle-mediated transport, eight were distributed among different categories, and ten had an unknown function (Table 1). Proteins involved in phosphorylation processes in the PM decreased with Fe deficiency. The signaling related category contained five protein kinases depending on several elicitors such as Ca, lectin, chitin and one (pto-interacting protein 1) mediating hypersensitive-induced response (HIR), as well as seven signal transducers (three proteins related to the HIR, 14-3-3 protein 10, calmodulin-like protein 13, remorin and a proton pump interactor; Table 1). All these decreases suggest that Fe-deficiency strongly impacts phosphorylation related processes at the PM level and also indicate that several signaling cascades are affected, including the HIR-mediated cascade (four HIR-related proteins) and cascades having Ca as a second messenger (a Cadependent kinase and calmodulin-like protein 13). Decreases in HIR1 had been detected in Fedeficient Arabidopsis roots 57 and furthermore a phosphoproteomic study of Fe deficient Arabidopsis roots also revealed that Ca-dependent kinases play critical functions in Fe homeostasis.58 Our data support that the decrease in HIR1 associated to Fe deficiency occurs in the PM. Iron deficiency also caused strong decreases in the relative abundances of five N and three K transporters at the PM level. Proteins involved in NO3- (NTR1 -two proteins- and NTR2.5) and NH4+ (AMT1.2 and AMT3.1) transport showed large decreases in relative abundance (from 50 to 97%) as a result of Fe deficiency (Table 1). Proteomic studies of Arabidopsis roots have also described reductions in AtNTR1 and phosphorylated AtNRT2.1 protein abundances upon Fe starvation in Arabidopsis roots.57,58 The high affinity transporters NTR1 and NTR2.5 are involved in root to shoot NO3- translocation and in the transfer of NO3- from stored pools to the 15 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

cytoplasm, respectively,59 whereas the AMT transporters participate in NH4+ uptake from the soil.60 Their decreases suggest that Fe deficiency modifies both NH4+ uptake and NO3distribution and highlight the already described interplay between Fe and N nutrition.11,61 On the other hand, since NTR1 can promote auxin transport out of young lateral roots when NO3availability is low, therefore repressing lateral root growth,62 the decreased accumulation of NTR1 in Fe deficiency would be in line with the lateral root formation commonly observed in Fe-deficient roots. Decreases were also measured in proteins involved in K transport, including the K transporter 5 (POT5), the cation/H+ exchanger 20 (CHX20), and the beta subunit of a voltage-gated K channel (Table 1). These decreases are in line with the slight decrease in K concentration reported in Fe-deficient sugar beet roots.63 Since K is one of the major osmolytes, a decreased accumulation of K transporters may be related to the decreased water fluxes associated to the low photosynthetic rates of Fe deficient plants. Six proteins participating in vesicular trafficking and secretion in the PM also decreased with Fe deficiency. These included: i) subunits Sec3A and Sec8 of the exocyst complex involved in exocytosis at the PM of different compounds, including polarized pectin delivery for the formation of new primary cell wal,64,65 ii) the clathrin heavy chain involved in endocytosis and polar distribution of PIN auxin transporters,66 and iii) the SNARE protein SYP121, involved in specific vesicle fusion with the PM (Table 1)2. Interestingly, SYP121 participates in the anchoring of K channels within microdomains of the PM,67 and its decrease is in line with the decrease observed in the voltage gated K channel in our study. Two more proteins, RABD2a (a small GTPase protein) and an EH domain containing protein 1 involved in ER-Golgi membrane trafficking and endocytosis also decreased. All these changes highlight that Fe deficiency leads to a reduced vesicular transport at the PM level. Some of these processes may include a reduced


Page 16 of 45

Page 17 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


exocytosis of pectin mediated by the exocyst complex and decreases in the allocation of auxin and K transporters to the PM mediated by the clathrin and SNARE pathways, respectively. In line with this hypothesis, a clustering analysis of Fe responsive proteins obtained from an iTRAQ study of Fe deficient Arabidopsis roots also suggested a link between Fe deficiency and altered vesicle transport.57 Our data support that this alteration associated to Fe deficiency occurs at the PM level. Finally, other decreases worth mentioning included those of an α-1,4-glucan-protein synthase with a possible role in the synthesis of cell wall polysaccharides 68 that highlights the effect of Fe deficiency in cell wall formation, a fasciclin-like arabinogalactan protein 6 precursor which may be a cell surface adhesion protein 69 and two disulfide isomerase precursors participating in the folding of proteins that are usually associated to ER that may arise from ER cross-contamination or ER-PM-contact sites.70

Iron deficiency causes changes in the lipid/protein ratio of the PM Iron deficiency caused a 20% decrease in the amount of PM-associated polar lipids, when expressed on a protein basis (Table 2). This decrease was mainly associated to decreases in three phospholipid classes: phosphatidylglycerol (PG, 30%), phosphatidylethanolamine (PE, 20%) and phosphatidic acid (PA, 36%) derivatives (Table 2). Some phospholipid species belonging to the phosphatidylcholine (PC) and phosphatidylserine (PS) classes also changed significantly, although the total amount of these classes was not affected by the treatment. The decreases in specific phospholipid species are listed in Table S3. Iron deficiency did not change the polar lipid unsaturation degree of the PM or the proportion of polar head groups in each lipid class (Fig. S3). In addition, the percentages of glycerolipids, sphingolipids, and sterols in the PM 17 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

were approximately 60, 10, 30%, respectively, and they did not change with Fe deficiency (Fig. 3). Therefore, in the absence of other changes the decrease in lipid packing in the PM may cause a reduced PM fluidity, which in turn is likely to affect other PM-related processes. Interestingly, PA derivatives were the lipid class showing the largest decrease (36%, with nine out 12 PA species identified in PM decreasing; Table S3). PAs have a small highly charged head very close to the glycerol backbone and increases in PA have been described to promote MP curvature during vesicle formation.71 Therefore, the PA decrease measured in PM preparations from Fedeficient plants supports the impairment of vesicle formation, in agreement with the decreases measured in proteins assigned to intracellular trafficking and secretion (Table 1). On the other hand, phospholipids such as PA are also involved in cell signaling as membrane-derived second messengers (either by themselves or playing a role as precursors) and therefore changes in their relative composition may also affect signaling processes in the cell. Iron deficiency causes marked changes on the protein profile of the DRM, some of them not detected at the PM level Altogether, 316 of the 545 detected proteins in this study changed significantly in DRM preparations as a result of Fe-deficiency; among these, 175 were identified and quantified with ≥ 2 peptides and 68 presented fold changes above the threshold level (≥ 2-fold or ≤ 0.5-fold), with 12 of them increasing and 56 decreasing in abundance (Fig. 1B and Tables 1 and 2). The functional categorization of these proteins was similar to that observed for proteins changing in PM fractions, with signaling, transport, and membrane trafficking accounting for 51% of the total differential proteins (26, 15 and 10%, respectively; Fig. 2B and Tables 1 and 2). Therefore, from a qualitative point of view, changes induced by Fe deficiency in PM and DRMs were similar. However, a larger number of proteins was found to be significantly affected by Fe18 ACS Paragon Plus Environment

Page 18 of 45

Page 19 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


deficiency in DRMs than in PM (68 vs. 52). This is probably associated to the nature of DRM preparations, which are relatively simpler in composition, making easier to quantify changes in low abundance proteins. When proteins changing as a result of Fe deficiency in PM and DRM were compared (Fig. 1C), 42 of them were common to PM and DRM (34 decreasing and 8 increasing; Table 1). Differences included ten proteins that decreased with Fe deficiency only in PM (Table 1) and 26 (22 decreasing and 4 increasing, Table 3) that changed only in DRM. However, it should be noted that nine of the ten proteins decreasing significantly only in PM also decreased in DRM, but the threshold for the fold decrease (≤ 0.5) was not met. The four proteins increasing with Fe deficiency only in DRM included two unknowns and two related to general metabolism, an adenosylhomocysteinase and alcohol dehydrogenase 2 (Table 3). Surprisingly, none of these two proteins are membrane proteins or contain motifs for post-translational modifications (PMTs) that would allow anchoring to the PM. Their increase in DRM and not in PM as a result of Fe deficiency may indicate that their functions (regulation of methylation and NADH production, respectively) require an association to proteins in specific areas of the PM. Among the 22 proteins decreasing in abundance only in DRM, nine were related to signaling, two to vesicle trafficking, two were hexose and Ca transporters, three were unknown and six presented miscellaneous functions (Table 3). Interestingly, six of the nine signaling related proteins decreasing only in DRMs were 14-3-3-like proteins (thought to be loosely attached to the PM), suggesting that Fe deficiency has an effect on their interaction with the PM. Indeed, 143-3 proteins bind to phosphorylated target proteins and the Fe deficiency-induced decrease in DRM would be in line with the decreases in relative abundance measured in an important number of kinases (5 in PM and 2 in DRM), which suggests a reduced number of phosphorylated 19 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

proteins in the root PM from Fe deficient plants. Furthermore, 14-3-3 proteins are mainly thought to be regulators of C and N metabolism, with nitrate reductase, PEPC and several TCA enzymes being targets of their action.72,73 Several studies at the transcriptomic, proteomic and physiological levels have demonstrated that Fe deficiency causes an alteration in the C and N metabolic pathways, especially in PEPC/TCA and NO3- assimilation.11,44,61 Our results indicate that PM may play a role in the signaling cascades leading to these changes via interactions with 14-3-3 proteins. Although the specific targets for these 14-3-3 proteins in Fe deficiency would require detailed studies, it is tempting to speculate that Fe deficiency causes a decrease in DRMs of a network comprising 14-3-3 proteins, CDPKs, and Ca transporters, which possibly regulate target genes in TCA and N metabolic pathways. This hypothesis would be in line with the substantial changes in the abundance of phosphoproteins involved in primary carbohydrate metabolism, and the possible involvement of CDPKs, described in a phosphoproteomic study of Fe-deficient Arabidopsis roots.58 On the other hand, since another target of 14-3-3 proteins is a K channel (KAT1),74 the decreases measured here in both K transporters and 14-3-3 proteins may also be related. With regard to the two vesicle mediated transport proteins decreasing only in DRM (Table 3), SEC15B seems to have a similar function in pectin delivery to those units changing in PM and the second one, SEC5A, is specifically involved in autophagy-related, Golgi-independent membrane traffic to the vacuole,75 highlighting the effect of Fe-deficiency in vesicle mediated transport systems in specific membrane domains.

Iron deficiency affects the partitioning of 30 proteins between the PM and DRM


Page 20 of 45

Page 21 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


To assess whether there were quantitative changes in the partitioning of each PM protein into DMRs as a result of Fe-deficiency, we analyzed changes in the DRM/PM ratios in both Fesufficient and Fe-deficient conditions. In this analysis, 402 of the 545 detected proteins presented significant changes (p ≤ 0.05) in the DRM/PM ratio in at least one of the treatments (Fig. 4A). Among them, 218 were identified and quantified with ≥ two peptides and 30 of them had fold changes [(DRM/PM) in –Fe vs. (DRM/PM) in +Fe] within the fold-change cut-offs considered biologically relevant (≥ 2 or ≤ 0.5; Fig. 4A and Table 4). For ease of viewing, the log2DRM/PM ratios of Fe-deficient samples were plotted against those of Fe-sufficient samples (Fig. 4B). This plot revealed that i) most of the proteins were not partitioned differentially as a consequence of Fe deficiency (black dots within the two diagonal lines depicting the threshold values) and ii) that among the 30 proteins changing significantly and above the threshold levels (Table 3 and marked in red in Fig. 4B), 11 were relatively more abundant in the DRM of Fe-deficient plants (red dots above the upper diagonal), whereas 19 proteins were relatively less abundant (red dots below the lower diagonal). Among the 11 more abundant proteins in DRM of Fe-deficient plants, two (the NRT1 transporter and a Ca-dependent kinase, red dots in the upper right quadrant above the upper diagonal) were always enriched in the DRM but relatively more in Fe-deficient samples than in the controls, indicating a relative increase in their abundances. On the other hand, nine proteins (red dots in the lower left quadrant above the upper diagonal) were always depleted (washed) in the DRM but less in Fe-deficient samples than in the controls, suggesting that their association with DRMs was stronger in Fe-deficient conditions (Table 4). Within these nine “less washed” proteins, we confirmed the existence of a group of four cytoplasmic proteins related to metabolism that appears to be somehow more attached to PM in Fe-deficient conditions. These 21 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

included alcohol dehydrogenase 2 (ADH2), a 6-phosphogluconate dehydrogenase (6PGD), an adenosylhomocysteinase (SAHH), and an aminocyclase (Acy 1). ADH2 and 6PGD produce NADH and NADPH, respectively, and their relative increase may help supply the increased demand of reducing power of the Fe uptake system, therefore establishing a link between the source of reducing power (cytoplasm) and the consumption place (PM). Interestingly, in Chlamydomonas, the chloroplast counterparts of these two proteins increased under Fe deficiency and hierarchical network analyses of proteomic results indicated that they are also clustered together .76 The other two proteins in this group, SAHH and ACY 1, generate free homocysteine and aliphatic amino acids, and their relative increase may be related to the N recycling processes occurring in Fe-deficient roots.11,61,77 On the other hand, S-adenosyl methionine, as the substrate for NA and ethylene biosynthesis, plays a central role of in the metabolism of Fe deficient plants 57 and the relative increase in SAHH in Fe-deficient DRMs might also be related to these processes. This analysis also revealed new players among the “less washed” proteins in Fe-deficient DRMs, including three proteins related to defense against oxidative stress (two heme binding peroxidases located in the extracellular region and the monodehydroascorbate reductase) and two chaperones (HSP80 and calreticulin). The fact that these proteins are more strongly attached to the PM in Fe-deficient conditions suggests a role for them in the Fe deficiency response. Among the 19 proteins relatively less abundant in Fe-deficient DMRs (red dots below the lower diagonal), two unannotated sequences (unknown) were always enriched in the DRMs but relatively more in the controls than in the Fe-deficient samples (red dots in the upper right quadrant below the lower diagonal), whereas four were enriched in DRMs of control samples but depleted in those of Fe-deficient samples (red dots in the lower right quadrant below the lower


Page 22 of 45

Page 23 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


diagonal) and 13 were always depleted in DRMs but less in the controls than in the Fe-deficient samples (red dots in the lower left quadrant below the lower diagonal) (Fig. 4B, Table 4). Among the latter 17 proteins, we confirmed the Fe deficiency-induced relative DRM decreases of six 14-3-3 proteins, as well as in the K transporter 5 (POT5) and the cation/H+ exchanger 20 (CHX20), highlighting their possible implication in the regulation of the Fe-deficiency response as commented above. The role of 14-3-3 proteins in the signaling cascades of N, P and S nutrient deficiencies has already been postulated 73 and our results clearly indicate that they also participate in the Fe-deficiency signaling cascade. In addition, three ribosomal proteins (60S L15, 60S L12-1 and 40S S13) and an elongation factor (EF-1-delta) showed a relative decrease in Fe-deficient DRMs (Table 4), suggesting that there are less membrane bound ribosomes in Fe-deficient samples, although their presence may also indicate some ER cross-contamination. Fe shortage also induced a decrease in abundance in several constituents of the ribosome in A. thaliana roots, and authors hypothesized that these modifications in the translation machinery could favor the synthesis of specific groups of proteins.57,58 Relative decreases in DRM were also measured in two proteins, pyruvate kinase and enolase, that yield phosphoenolpyruvate as a product, in contrast with the relative increase induced by Fe deficiency in the abundance of PEPC in PM. These decreases highlight the effects of Fe-deficiency on C metabolism and its complex regulation. Finally, relative decreases in Fe-deficient DRMs were also measured in a tubulin, an unknown protein, and in the long chain acyl-CoA synthetase 8. This latter decrease may be related with the decrease measured in PM lipids. It is also worth mentioning that when the 218 identified proteins showing differential DRM partitioning in any of the treatments (104 DRM-enriched, 114 DRM-non enriched) were functionally classified, the DRM-enriched subproteome displayed, when compared to the



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

depleted DMR fraction, a larger number of proteins known to be detergent resistant, such as proteins assigned to transport (20%), membrane trafficking (18%) and signaling kinases, (28%) (Fig. S4). On the other hand, the number of proteins known to be detergent-labile, such as those involved in protein synthesis (4%) and general metabolism (5%) was lower in the DRM-enriched subproteome, whereas they accounted for almost 50% of the DRM-non enriched subproteome (Fig. S4).

Conslusion The shotgun proteomic analysis of the PM and DRMs reveals that changes of proteins related to Fe uptake (which indeed remain hidden in the study due to limitations in their identification) are only the tip of the Fe-deficiency response iceberg. This study reveals that Fe deficiency causes changes in the abundance of many PM proteins that were previously not known to occur, as well as decreases in the lipid/protein ratio that may lead to decreased membrane fluidity. Changes found mainly reflect modifications in phosphorylation, intracellular trafficking and secretion, and in several signaling cascades, as well as in the associations of cytoplasmic proteins with the PM. The latter changes appear to be closely linked with metabolic responses commonly reported to occur in Fe-deficient roots, such as increases in PEPC/TCA and changes in N assimilation. Results also highlight the role of the 14-3-3 protein family in the Fe-deficiency signaling cascade and point towards a relatively small number of candidates (up to six) as preferential targets for further studies. Acknowledgements Supported by the Spanish Ministry of Economy and Competitivity (MINECO; projects AGL2012-31988, and AGL2013-42175-R, co-financed with FEDER), and the Aragón 24 ACS Paragon Plus Environment

Page 24 of 45

Page 25 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Government (Group A03). Research conducted in Iwate University was in part supported by Grants-in-Aid for Scientific Research (#24-7373, #22120003, and #24370018) from the Japan Society for the Promotion of Science. E.G.-C. was supported by JAE Pre-CSIC contract. Authors declare no conflict of interest.

Supporting information available: Table S-1: List of peptides identified and quantified by shotgun proteomics and Progenesis LC-MS analyses; Table S-2: List of proteins detected in the shotgun proteomic analyses; Table S-3: List of lipid compounds detected in PM preparations of Fe-sufficient and Fe-deficient plants; Table S-4: Concentration of sterols, C16/C18 fatty acids and very long chain fatty acids (VLCFA) in PM preparations of Fe-sufficient and Fe-deficient plants; Table S-5. Activities (in nmol mg protein-1 min-1) of enzymes selected as purity markers for PM fractions; Figure S-1: Picture depicting the white band enriched in DRM in the interface of the 35 and 48% sucrose layers; Figure S-2: Electron microscopy of control and Fe-deficient microsomal fractions and PM preparations stained with PTA; Figure S-3: Degree of lipid unsaturation and proportion of polar head groups in each lipid class in Fe-deficient and Fe-sufficient PM preparations; Figure S-4. Functional classification of proteins showing changes in DRM partitioning.

References (1)

(2)

Malinsky, J.; Opekarova, M.; Grossmann, G.; Tanner, W., Membrane microdomains, rafts, and detergent-resistant membranes in plants and fungi. Annu. Rev. Plant Biol. 2013, 64, 501-29. Marmagne, A.; Ferro, M.; Meinnel, T.; Bruley, C.; Kuhn, L.; Garin, J.; Barbier-Brygoo, H.; Ephritikhine, G., A high content in lipid-modified peripheral proteins and integral receptor 25 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(3) (4) (5) (6)

(7) (8) (9) (10)

(11)

(12)

(13)

(14)

(15)

(16)

(17)

kinases features in the arabidopsis plasma membrane proteome. Mol. Cell. Proteomics 2007, 6 (11), 1980-96. Romheld, V.; Marschner, H., Mechanism of iron uptake by peanut plants : I. Fe reduction, chelate splitting, and release of phenolics. Plant Physiol. 1983, 71 (4), 949-54. Romheld, V.; Marschner, H., Evidence for a specific uptake system for iron phytosiderophores in roots of grasses. Plant Physiol. 1986, 80 (1), 175-80. Marschner, H., Mineral nutrition of higher plants. 2nd ed.; 1995. Eide, D.; Broderius, M.; Fett, J.; Guerinot, M. L., A novel iron-regulated metal transporter from plants identified by functional expression in yeast. Proc. Natl. Acad. Sci. U. S. A. 1996, 93 (11), 5624-28. Robinson, N. J.; Procter, C. M.; Connolly, E. L.; Guerinot, M. L., A ferric-chelate reductase for iron uptake from soils. Nature 1999, 397 (6721), 694-97. Santi, S.; Schmidt, W., Dissecting iron deficiency-induced proton extrusion in Arabidopsis roots. New Phytol. 2009, 183 (4), 1072-84. Abadía, J.; López-Millán, A. F.; Rombolà, A.; Abadía, A., Organic acids and Fe deficiency: a review. Plant and Soil 2002, 241 (1), 75-86. Cesco, S.; Neumann, G.; Tomasi, N.; Pinton, R.; Weisskopf, L., Release of plant-borne flavonoids into the rhizosphere and their role in plant nutrition. Plant and Soil 2010, 329 (1), 1-25. Rodríguez-Celma, J.; Lattanzio, G.; Grusak, M. A.; Abadía, A.; Abadía, J.; López-Millán, A. F., Root responses of Medicago truncatula plants grown in two different iron deficiency conditions: changes in root protein profile and riboflavin biosynthesis. J. Proteome Res. 2011, 10 (5), 2590-601. Rodríguez-Celma, J.; Lin, W. D.; Fu, G. M.; Abadía, J.; López-Millán, A. F.; Schmidt, W., Mutually exclusive alterations in secondary metabolism are critical for the uptake of insoluble iron compounds by Arabidopsis and Medicago truncatula. Plant Physiol. 2013, 162 (3), 1473-85. Rodríguez-Celma, J.; Vazquez-Reina, S.; Orduna, J.; Abadía, A.; Abadía, J.; ÁlvarezFernández, A.; López-Millán, A. F., Characterization of flavins in roots of Fe-deficient strategy I plants, with a focus on Medicago truncatula. Plant Cell. Physiol. 2011, 52 (12), 2173-89. Fourcroy, P.; Siso-Terraza, P.; Sudre, D.; Saviron, M.; Reyt, G.; Gaymard, F.; Abadía, A.; Abadía, J.; Álvarez-Fernández, A.; Briat, J. F., Involvement of the ABCG37 transporter in secretion of scopoletin and derivatives by Arabidopsis roots in response to iron deficiency. New Phytol. 2014, 201 (1), 155-67. Schmid, N. B.; Giehl, R. F.; Doll, S.; Mock, H. P.; Strehmel, N.; Scheel, D.; Kong, X.; Hider, R. C.; von Wiren, N., Feruloyl-CoA 6'-Hydroxylase1-dependent coumarins mediate iron acquisition from alkaline substrates in Arabidopsis. Plant Physiol. 2014, 164 (1), 16072. Durrett, T. P.; Gassmann, W.; Rogers, E. E., The FRD3-mediated efflux of citrate into the root vasculature is necessary for efficient iron translocation. Plant Physiol. 2007, 177, 197205. Roschzttardtz, H.; Seguela-Arnaud, M.; Briat, J. F.; Vert, G.; Curie, C., The FRD3 citrate effluxer promotes iron nutrition between symplastically disconnected tissues throughout Arabidopsis development. Plant Cell 2011, 23 (7), 2725-37.


Page 26 of 45

Page 27 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(18) Hossain, Z.; Komatsu, S., Contribution of proteomic studies towards understanding plant heavy metal stress response. Front. Plant Sci. 2012, 3, 310. (19) Jorrin-Novo, J. V., Plant proteomics. J. Proteomics 2009, 72 (3), 283-4. (20) Liang, C.; Tian, J.; Liao, H., Proteomics dissection of plant responses to mineral nutrient deficiency. Proteomics 2013, 13 (3-4), 624-36. (21) López-Millán, A. F.; Grusak, M. A.; Abadía, A.; Abadía, J., Iron deficiency in plants: an insight from proteomic approaches. Front. Plant Sci. 2013, 4, 254. (22) Coulombe, B. A.; Chaney, R. L.; Wiebold, W. J., Bicarbonate directly induces iron chlorosis in susceptible soybean cultivars. Soil Sci. Soc. Am. J. 1984, 48, 1297-301. (23) Froechlich, D. M.; Fehr, W. R., Agronomic Performance of Soybeans with Differing Levels of Iron Deficiency Chlorosis on Calcareous Soil. Crop Sci. 1981, 21 (3), 438-41. (24) Abadía, J.; Vazquez, S.; Rellán-Álvarez, R.; El-Jendoubi, H.; Abadía, A.; ÁlvarezFernández, A.; López-Millán, A. F., Towards a knowledge-based correction of iron chlorosis. Plant Physiol. Biochem. 2011, 49 (5), 471-82. (25) Rombolà, A.; Tagliavini, M., Iron nutrition of fruit tree crops. In Iron Nutrition in Plants and Rhizospheric Microorganisms, Barton, L. L.; Abadia, J., Eds. Springer: Dordrecht, Netherlands, 2006; pp 61-83. (26) Terry, N., Limiting factors in photosynthesis .1. Use of iron stress to control photochemical capacity in vivo. Plant Physiol. 1980, 65 (1), 114-20. (27) Brüggemann, W.; Moog, P. R., NADH-dependent Fe3+EDTA and oxygen reduction by plasma membrane vesicles from barley roots. Physiol. Plant. 1989, 75 (2), 245-54. (28) Larsson, C., Plasma membranes. In Modern methods of plant analysis, Linskens, H. F.; Jackson, J. F., Eds. Springer: Berlin, Heidelberg, New York, Tokyo, 1985; Vol. 1, pp 85104. (29) Widell, S.; Larsson, C., A critical evaluation of markers used in plasma membrane purification. In The plant plasma membranes: structure, function, and molecular biology, Larsson, C.; Mdler, I. M., Eds. Springer-Verlag: Berlin, 1990; pp 16-43. (30) O'Neill S, D.; Bennett, A. B.; Spanswick, R. M., Characterization of a NO3- sensitive HATPase from corn roots. Plant Physiol. 1983, 72 (3), 837-46. (31) Hodges, T. K.; Leonard, R. T., Purification of a plasma membrane-bound adenosine triphosphatase from plant roots. Methods Enzymol. 1974, 32, 392-406. (32) Peskan, T.; Westermann, M.; Oelmuller, R., Identification of low-density Triton X-100insoluble plasma membrane microdomains in higher plants. Eur. J. Biochem. 2000, 267 (24), 6989-95. (33) Takahashi, D.; Li, B.; Nakayama, T.; Kawamura, Y.; Uemura M., Shotgun proteomics of plant plasma membrane and microdomain proteins using nano-LC-MS/MS. In Plant Proteomics: Methods and Protocols, Novo, J. V. J.; Komatsu, S.; Wechwerth, W.; Wienkoopeds, S., Eds. Springer Science+Media, LLC: New York, 2014; Vol. 1072, pp 481–98. (34) Susín, S.; Abadía, A.; González-Reyes, J. A.; Lucena, J. J.; Abadía, J., The pH requirement for in vivo activity of the iron-deficiency-induced "Turbo" ferric chelate reductase (A comparison of the iron-deficiency-induced iron reductase activities of intact plants and isolated plasma membrane fractions in sugar beet). Plant Physiol. 1996, 110 (1), 111-23. (35) Serrano, A.; Córdoba, F.; González-Reyes, J. A.; Navas, P.; Villalba, J. M., Purification and characterization of two distinct NAD(P)H dehydrogenases from onion (Allium cepa L.) root plasma membrane. Plant Physiol. 1994, 106 (1), 87-96. 27 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(36) Takahashi, D.; Kawamura, Y.; Uemura, M., Changes of detergent-resistant plasma membrane proteins in oat and rye during cold acclimation: association with differential freezing tolerance. J. Proteome Res. 2013, 12 (11), 4998-5011. (37) Fu, L.; Niu, B.; Zhu, Z.; Wu, S.; Li, W., CD-HIT: accelerated for clustering the nextgeneration sequencing data. Bioinformatics 2012, 28 (23), 3150-2. (38) Brugger, B.; Erben, G.; Sandhoff, R.; Wieland, F. T.; Lehmann, W. D., Quantitative analysis of biological membrane lipids at the low picomole level by nano-electrospray ionization tandem mass spectrometry. Proc. Natl. Acad. Sci. U. S. A. 1997, 94 (6), 2339-44. (39) Shoemaker, J. P.; Garland, C. W.; Steinfeld, J. I., Experiments in Physical Chemistry. McGraw-Hill, Inc.: New York, 1974; p 34-39. (40) Cacas, J.-L.; Buré, C.; Grosjean, K.; Gerbeau-Pissot, P.; Lherminier, J.; Rombouts, Y.; Maes, E.; Bossard, C.; Gronnier, J.; Furt, F.; Fouillen, L.; Germain, V.; Bayer, E.; Cluzet, S.; Robert, F.; Schmitter, J.-M.; Deleu, M.; Lins, L.; Simon-Plas, F.; Mongrand, S. Revisiting plant plasma membrane lipids in tobacco: A focus on sphingolipids. Plant Physyol. 2016, 170, 367-384. (41) Takahashi, D.; Kawamura, Y.; Yamashita, T.; Uemura, M., Detergent-resistant plasma membrane proteome in oat and rye: similarities and dissimilarities between two monocotyledonous plants. J. Proteome Res. 2012, 11 (3), 1654-65. (42) Borner, G. H.; Lilley, K. S.; Stevens, T. J.; Dupree, P., Identification of glycosylphosphatidylinositol-anchored proteins in Arabidopsis. A proteomic and genomic analysis. Plant Physiol 2003, 132 (2), 568-77. (43) López-Millán, A. F.; Morales, F.; Andaluz, S.; Gogorcena, Y.; Abadía, A.; De Las Rivas, J.; Abadía, J., Responses of sugar beet roots to iron deficiency. Changes in carbon assimilation and oxygen use. Plant Physiol. 2000, 124 (2), 885-97. (44) Zocchi, G., Metabolic changes in iron-stressed dicotyledoneus plants. In Iron nutrition in plants and rhizospheric microorganisms, Barton, L. L.; Abadía, J., Eds. Springer: Dordrecht, 2006; pp 359-70. (45) Roppolo, D.; Boeckmann, B.; Pfister, A.; Boutet, E.; Rubio, M. C.; Denervaud-Tendon, V.; Vermeer, J. E.; Gheyselinck, J.; Xenarios, I.; Geldner, N., Functional and evolutionary analysis of the casparian strip membrane domain protein family. Plant Physiol. 2014, 165 (4), 1709-22. (46) Donnini, S.; Castagna, A.; Ranieri, A.; Zocchi, G., Differential responses in pear and quince genotypes induced by Fe deficiency and bicarbonate. J. Plant Physiol. 2009, 166 (11), 1181-93. (47) Donnini, S.; Dell'Orto, M.; Zocchi, G., Oxidative stress responses and root lignification induced by Fe deficiency conditions in pear and quince genotypes. Tree Physiol. 2011, 31 (1), 102-13. (48) Berditchevski, F., Complexes of tetraspanins with integrins: more than meets the eye. J. Cell Sci. 2001, 114 (Pt 23), 4143-51. (49) Arthikala, M. K.; Sanchez-Lopez, R.; Nava, N.; Santana, O.; Cardenas, L.; Quinto, C., RbohB, a Phaseolus vulgaris NADPH oxidase gene, enhances symbiosome number, bacteroid size, and nitrogen fixation in nodules and impairs mycorrhizal colonization. New Phytol. 2014, 202 (3), 886-900. (50) Foreman, J.; Demidchik, V.; Bothwell, J. H.; Mylona, P.; Miedema, H.; Torres, M. A.; Linstead, P.; Costa, S.; Brownlee, C.; Jones, J. D.; Davies, J. M.; Dolan, L., Reactive


Page 28 of 45

Page 29 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(51)

(52) (53)

(54) (55)

(56)

(57)

(58) (59)

(60)

(61) (62)

(63)

(64)

oxygen species produced by NADPH oxidase regulate plant cell growth. Nature 2003, 422 (6930), 442-6. Monshausen, G. B.; Bibikova, T. N.; Messerli, M. A.; Shi, C.; Gilroy, S., Oscillations in extracellular pH and reactive oxygen species modulate tip growth of Arabidopsis root hairs. Proc. Natl. Acad. Sci. U. S. A. 2007, 104 (52), 20996-1001. Montiel, J.; Arthikala, M. K.; Quinto, C., Phaseolus vulgaris RbohB functions in lateral root development. Plant Signal. Behav. 2013, 8 (1), 6. Muller, K.; Linkies, A.; Leubner-Metzger, G.; Kermode, A. R., Role of a respiratory burst oxidase of Lepidium sativum (cress) seedlings in root development and auxin signalling. J. Exp. Bot. 2012, 63 (18), 6325-34. Charron, J. B.; Ouellet, F.; Houde, M.; Sarhan, F., The plant Apolipoprotein D ortholog protects Arabidopsis against oxidative stress. BMC Plant Biol. 2008, 8, 86. Chi, W. T.; Fung, R. W.; Liu, H. C.; Hsu, C. C.; Charng, Y. Y., Temperature-induced lipocalin is required for basal and acquired thermotolerance in Arabidopsis. Plant Cell Environ 2009, 32 (7), 917-27. Zhai, Z.; Gayomba, S. R.; Jung, H. I.; Vimalakumari, N. K.; Pineros, M.; Craft, E.; Rutzke, M. A.; Danku, J.; Lahner, B.; Punshon, T.; Guerinot, M. L.; Salt, D. E.; Kochian, L. V.; Vatamaniuk, O. K., OPT3 is a phloem-specific iron transporter that is essential for systemic iron signaling and redistribution of iron and cadmium in Arabidopsis. Plant Cell 2014, 26 (5), 2249-64. Lan, P.; Li, W.; Wen, T. N.; Shiau, J. Y.; Wu, Y. C.; Lin, W.; Schmidt, W., iTRAQ protein profile analysis of Arabidopsis roots reveals new aspects critical for iron homeostasis. Plant Physiol. 2011, 155 (2), 821-34. Lan, P.; Li, W.; Wen, T. N.; Schmidt, W., Quantitative phosphoproteome profiling of irondeficient Arabidopsis roots. Plant Physiol. 2012, 159 (1), 403-17. Orsel, M.; Eulenburg, K.; Krapp, A.; Daniel-Vedele, F., Disruption of the nitrate transporter genes AtNRT2.1 and AtNRT2.2 restricts growth at low external nitrate concentration. Planta 2004, 219 (4), 714-21. von Wiren, N.; Lauter, F. R.; Ninnemann, O.; Gillissen, B.; Walch-Liu, P.; Engels, C.; Jost, W.; Frommer, W. B., Differential regulation of three functional ammonium transporter genes by nitrogen in root hairs and by light in leaves of tomato. Plant J. 2000, 21 (2), 16775. Borlotti, A.; Vigani, G.; Zocchi, G., Iron deficiency affects nitrogen metabolism in cucumber (Cucumis sativus L.) plants. BMC Plant Biol. 2012, 12, 189. Krouk, G.; Lacombe, B.; Bielach, A.; Perrine-Walker, F.; Malinska, K.; Mounier, E.; Hoyerova, K.; Tillard, P.; Leon, S.; Ljung, K.; Zazimalova, E.; Benkova, E.; Nacry, P.; Gojon, A., Nitrate-regulated auxin transport by NRT1.1 defines a mechanism for nutrient sensing in plants. Dev. Cell. 2010, 18 (6), 927-37. López-Millán, A. F.; Morales, F.; Abadía, A.; Abadía, J., Changes induced by Fe deficiency and Fe resupply in the organic acid metabolism of sugar beet (Beta vulgaris) leaves. Physiol. Plant. 2001, 112 (1), 31-38. Fendrych, M.; Synek, L.; Pecenkova, T.; Toupalova, H.; Cole, R.; Drdova, E.; Nebesarova, J.; Sedinova, M.; Hala, M.; Fowler, J. E.; Zarsky, V., The Arabidopsis exocyst complex is involved in cytokinesis and cell plate maturation. Plant Cell 2010, 22 (9), 3053-65.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(65) Kulich, I.; Cole, R.; Drdova, E.; Cvrckova, F.; Soukup, A.; Fowler, J.; Zarsky, V., Arabidopsis exocyst subunits SEC8 and EXO70A1 and exocyst interactor ROH1 are involved in the localized deposition of seed coat pectin. New Phytol. 2010, 188 (2), 615-25. (66) Kitakura, S.; Vanneste, S.; Robert, S.; Lofke, C.; Teichmann, T.; Tanaka, H.; Friml, J., Clathrin mediates endocytosis and polar distribution of PIN auxin transporters in Arabidopsis. Plant Cell 2011, 23 (5), 1920-31. (67) Amtmann, A.; Blatt, M. R., Regulation of macronutrient transport. New Phytol. 2009, 181 (1), 35-52. (68) Bocca, S. N.; Kissen, R.; Rojas-Beltran, J. A.; Noel, F.; Gebhardt, C.; Moreno, S.; du Jardin, P.; Tandecarz, J. S., Molecular cloning and characterization of the enzyme UDPglucose: protein transglucosylase from potatodaggerdagger. This paper is specially dedicated to the memory of Dr Juana S. Tandecarz, deceased on December 10, 1996. Plant Physiol. Biochem. 1999, 37 (11), 809-19. (69) Johnson, K. L.; Jones, B. J.; Bacic, A.; Schultz, C. J., The fasciclin-like arabinogalactan proteins of Arabidopsis. A multigene family of putative cell adhesion molecules. Plant Physiol. 2003, 133 (4), 1911-25. (70) Larsson, K. E.; Kjellberg, J. M.; Tjellstrom, H.; Sandelius, A. S., LysoPC acyltransferase/PC transacylase activities in plant plasma membrane and plasma membrane-associated endoplasmic reticulum. BMC Plant Biol. 2007, 7, 64. (71) Kooijman, E. E.; Chupin, V.; de Kruijff, B.; Burger, K. N., Modulation of membrane curvature by phosphatidic acid and lysophosphatidic acid. Traffic 2003, 4 (3), 162-74. (72) Diaz, C.; Kusano, M.; Sulpice, R.; Araki, M.; Redestig, H.; Saito, K.; Stitt, M.; Shin, R., Determining novel functions of Arabidopsis 14-3-3 proteins in central metabolic processes. BMC Syst. Biol. 2011, 5 (192), 1752-0509. (73) Shin, R.; Jez, J. M.; Basra, A.; Zhang, B.; Schachtman, D. P., 14-3-3 proteins fine-tune plant nutrient metabolism. FEBS Lett. 2011, 585 (1), 143-7. (74) Sottocornola, B.; Visconti, S.; Orsi, S.; Gazzarrini, S.; Giacometti, S.; Olivari, C.; Camoni, L.; Aducci, P.; Marra, M.; Abenavoli, A.; Thiel, G.; Moroni, A., The potassium channel KAT1 is activated by plant and animal 14-3-3 proteins. J. Biol. Chem. 2006, 281 (47), 35735-41. (75) Kulich, I.; Pecenkova, T.; Sekeres, J.; Smetana, O.; Fendrych, M.; Foissner, I.; Hoftberger, M.; Zarsky, V., Arabidopsis exocyst subcomplex containing subunit EXO70B1 is involved in autophagy-related transport to the vacuole. Traffic 2013, 14 (11), 1155-65. (76) Hohner, R.; Barth, J.; Magneschi, L.; Jaeger, D.; Niehues, A.; Bald, T.; Grossman, A.; Fufezan, C.; Hippler, M., The metabolic status drives acclimation of iron deficiency responses in Chlamydomonas reinhardtii as revealed by proteomics based hierarchical clustering and reverse genetics. Mol. Cell. Proteomics 2013, 12 (10), 2774-90. (77) Donnini, S.; Prinsi, B.; Negri, A. S.; Vigani, G.; Espen, L.; Zocchi, G., Proteomic characterization of iron deficiency responses in Cucumis sativus L. roots. BMC Plant Biol. 2010, 10, 268.


Page 30 of 45

Page 31 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48


Table 1- List of proteins showing significant changes in relative abundance in PM above or below the respective fold-change levels (ANOVA, p ≤ 0.05; Fe-deficient sample vs Fe-sufficient treatment fold-changes ≥ 2.00 or ≤ 0.50) identified and quantified with at least two unique peptides. Values in bold indicate p ≤ 0.05. Grey and white backgrounds in the first column indicates proteins changing significantly only in PM, and in both PM and DRM preparations, respectively. Detailed information about identification, BLAST analysis and abundances is given in Table S2. Mean of normalized abundance

Beta vulgaris DB Entry

KDHBv_S06825_245030.t1 KDHBv_S05354_212150.t1 Bv_45170_qnkp.t1 YMoBv_S01250_43530.t1 KDHBv_S13688_53390.t1 KDHBv_S01806_86470.t1 Bv5_108520_qgdi.t1 KDHBv_S02973_141020.t1 UMSBv_S60675_290610.t1 KDHBv_S02849_136140.t1 KDHBv_S06794_244090.t1 KDHBv_S03278_153490.t1 KDHBv_S01382_54610.t1 KDHBv_S02849_136130.t1

PM -Fe

DRM +Fe

DRM -Fe

Ratio PM -Fe vs +Fe

Ratio DRM -Fe vs +Fe

Transport 77235 40399 126814 76921 119521 291642 7763

2646 1503 11347 24511 62279 15635 1403

189815 74504 150652 196311 227972 661212 20657

8818 6597 22550 54868 110793 14463 1257

0.03 0.04 0.09 0.32 0.50 0.05 0.18

0.05 0.09 0.15 0.28 0.49 0.02 0.06

47265

21302

44702

11849

0.45

0.27

Signaling/Regulation 27411 14916

15722

8781

0.50

0.56

PM +Fe

Protein Description

high affinity nitrate transporter 2.5 NRT1/PTR family protein 6.3 NRT1/PTR family protein 6.3 ammonium transporter 1-2 ammonium transporter 3-1 potassium transporter 5 cation/H(+) antiporter 20 Probable voltage-gated K channel subunit β proton pump-interactor 1 hypersensitive-induced response protein 1 calcium-dependent protein kinase remorin 14-3-3 protein 10 hypersensitive-induced response protein 1

17014

9058

22435

13904

0.50

0.62

4346 6246 73541

1144 424 36815

7078 12661 24617

4614 484 4838

0.26 0.07 0.50

0.65 0.04 0.20

59532

12904

138965

37294

0.22

0.27



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

Page 32 of 45

KDHBv_S02849_136120.t1

hypersensitive-induced response protein 1

282493

125904

701267

279186

0.45

0.40

KDHBv_S02425_119360.t1

probable calcium-binding CML13

42953

17164

5680

1919

0.40

0.34

10080

2468

27961

8020

0.24

0.29

27111

5287

39607

11873

0.20

0.30

28254

7035

13308

4676

0.25

0.35

432967

124938

346329

139761

0.29

0.40

29395 115007 41264 16855 60876 26607 112802

20303 60707 15165 7561 28885 61062 44377

0.48 0.50 0.32 0.48 0.44 2.04 0.50

0.69 0.60 0.37 0.45 0.47 2.29 0.39

38535 31353 784

8518 78649 4331

0.29 2.90 4.22

0.22 2.51 5.53

KDHBv_S03242_152320.t1 KDHBv_S01229_39180.t1 KDHBv_S01236_39790.t1 Bv6_144220_edoh.t1

Bv5_106010_yoiz.t1 Bv1u_019780_eiwa.t1 UMSBv_S00090_352590.t1 KDHBv_S00584_222620.t1 KDHBv_S03242_152330.t1 KDHBv_S00809_269470.t1 UMSBv_S14375_92090.t1

G-type lectin S-receptorserine/threonine-protein kinase At1g34300 precursor Pto-interacting protein 1 chitin elicitor receptor kinase 1 precursor probable leucine-rich repeat receptor-like protein kinase At5g49770 precursor exocyst complex unit SEC3A exocyst complex unit SEC8 syntaxin-121 (SYP121) Ras-related protein RABD2a EH domain-containing protein 1 Ras-related protein Rab2BV clathrin heavy chain 1

KDHBv_S03976_176500.t1 carbonic anhydrase 2, chloro. KDHBv_C14254217_08230.t1 PEP carboxylase 2 Bv6_154380_rnxg.t1 HIPL1 protein precursor Bv6_126010_pdxa.t1 KDHBv_S03855_173000.t1 KDHBv_S03825_172370.t1 UMSBv_S02360_158040.t1 KDHBv_S01348_51500.t1

putative disease resistance protein RGA3 protein disulfide-isomerase precursor protein disulfide isomerase-like 1-2 precursor respiratory burst oxidase homolog protein B α-1,4-glucan-protein synthase 2

Vesicle mediated transport 15231 7267 65501 32751 47685 15269 38457 18619 45793 20054 26109 53233 66543 34900 Carbon metabolism 35920 10361 66444 192827 39893 168276 Cell redox homeostasis / defense 509782

277935

752855

403687

0.50

0.60

763073

380768

46400

31804

0.50

0.69

10666

2962

798

297

0.28

0.37

18905

37452

26533

93751

2.00

3.53

Miscellaneous 39857 19298

13177

11925

0.48

0.90


Page 33 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48


KDHBv_C14180820_06830.t1 40S ribosomal protein S17 KDHBv_S04395_187110.t1 acetyl-CoA acetyltransferase, cyt. 1 KDHBv_S25783_125560.t1 (-)-germacrene D synthase fasciclin-like arabinogalactan UMSBv_S02775_179140.t1 protein 6 precursor YTiBv_S04182_189510.t1 CASP-like protein Ni6 cytosolic fatty acid binding Bv8_196790_huzg.t1 proteins-cytosolic lipid transporter Bv6_130090_rgzj.t1 tetraspanin-9 KDHBv_S09457_293900.t1 Bv3_055310_riir.t1 YMoBv_S05240_232570.t1 KDHBv_S04452_188920.t1 KDHBv_S02854_136340.t1 YTiBv_S00119_34990.t1 KDHBv_S08049_268330.t1 KDHBv_S00653_238960.t1 UMSBv_S01691_115330.t1 KDHBv_S02231_110810.t1

59220 42642 14528

29925 20856 52596

44470 30252 19784

24307 8693 51983

0.50 0.49 3.62

0.60 0.29 2.63

427194

214335

11039

3687

0.50

0.33

51139

102366

111869

243834

2.00

2.18

41742

87389

1560

4764

2.09

3.05

16661 Unknown 8371 48773 72056 22974 4656 69979 93770 5586 22903 110580

36457

26351

83854

2.19

3.18

3487 4840 11024 5930 1656 14525 35177 2505 7864 42187

17634 61081 180217 38955 3918 44633 40010 9922 44692 132606

11552 7307 23567 5738 764 9471 11458 3824 18174 63319

0.42 0.10 0.15 0.26 0.36 0.21 0.38 0.45 0.34 0.38

0.66 0.12 0.13 0.15 0.19 0.21 0.29 0.39 0.41 0.48



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

Table 2- Polar lipid concentrations (in pmol mg protein-1) in PM preparations of Fe-sufficient and Fe-deficient samples. Values are means ± SD (n=5). DGDG, digalactosyldiacylglycerol; MGDG, monogalactosyldiacylglycerol; PG, phosphatidylglycerol; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PI, phosphatidylinositol; PS, phosphatidylserine; PA, phosphatidic acid. Lipid degradation, measured as the amount of lyso (PG+PC+PE), accounted for ca. 3% of the total lipids. Detailed information about lipid species quantified is listed in Table S3.

pmol mg protein-1

+Fe

-Fe

-Fe/ +Fe

DGDG

15.3 ± 3.8 a

14.4 ± 6.8 a

0.9

MGDG

5.5 ± 0.4 a

4.4 ± 1.5 a

0.8

PG

6.1 ± 1.2 b

4.4 ± 1.1 a

0.7

PC

125.3 ± 32.6 a

99.3 ± 41.3 a

0.8

PE

114.4 ± 20.1 b

90.6 ± 19.7 a

0.8

PI

10.6 ± 6.6 a

10.0 ± 2.4 a

1.0

PS

4.0 ± 1.2 a

3.3 ± 1.3 a

0.8

PA

35.6 ± 12.3 b

22.7 ± 9.2 a

0.6

Total lipids in PM

327.0 ± 74.4 b

257.8 ± 83.0 a

0.8

Lyso(PG+PC+PE)

10.3 ± 3.0

8.6 ± 4.7


Page 34 of 45

Page 35 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48


Table 3- List of proteins showing significant changes in relative abundance only in DRM above or below the respective fold-change levels (ANOVA, p ≤ 0.05; Fe-deficient sample vs Fe-sufficient treatment fold-changes ≥ 2.00 or ≤ 0.50) identified and quantified with at least two unique peptides. Values in bold indicate p ≤ 0.05. Detailed information about identification, BLAST analysis and abundances is given in Table S2. Mean of normalized abundance

Beta vulgaris DB Entry

PM +Fe

Protein Description

PM -Fe

DRM +Fe

DRM -Fe

Ratio PM -Fe vs +Fe

Ratio DRM -Fe vs +Fe

7695

20797

9382

0.67

0.45

50475

17913

0.46

0.35

77169 253817 134228 22288 124175 5420 65550 32266

26645 88943 48491 9294 55118 2532 15801 9873

0.62 0.71 0.83 0.81 0.85 0.82 1.63 0.45

0.35 0.35 0.36 0.42 0.44 0.47 0.24 0.31

25815

10333

0.27

0.40

69513 53402

31823 25709

0.73 0.61

0.46 0.48

35132 6614

17146 25732

1.17 1.27

0.49 3.89

Transport UMSBv_S15773_105210.t1 KDHBv_S05816_222050.t1 KDHBv_S04127_180150.t1 KDHBv_S12835_44330.t1 KDHBv_S02163_107980.t1 KDHBv_S08502_276450.t1 KDHBv_S18793_91550.t1 KDHBv_S02710_129950.t1 KDHBv_S07778_263100.t1 UMSBv_S06984_314900.t1 KDHBv_S06442_236910.t1

Ca-transporting ATPase 2, PMtype sugar transport protein 14 14-3-3-like protein 14-3-3-like protein 14-3-3-like protein 14-3-3-like protein GF14 kappa 14-3-3-like protein annexin D2 14-3-3-like protein D protein LYK5 precursor calcium-dependent protein kinase 10

Bv1_012820_cdiz.t1 Bv7_164300_mszm.t1

exocyst complex unit SEC5A exocyst complex unit SEC15B

KDHBv_S03064_144220.t1 KDHBv_S08615_278650.t1

enolase adenosylhomocysteinase

11499

36902 16791 Signaling/Regulation 112831 69872 296317 210856 111997 93333 48071 38957 147568 125685 10940 8923 166805 271731 22878 10352 17863

4740

Vesicle mediated transport 24884 18185 33881 20759 Carbon metabolism 256470 299742 95968 121892



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

Bv1_011870_sghe.t1

alcohol dehydrogenase 2

YTiBv_S02161_114850.t1

glutathione S-transferase

Bv1u_019010_zfif.t1 KDHBv_S14321_60210.t1

tubulin beta-1 chain 60S ribosomal protein L15 membrane steroid-binding protein 2 cytochrome b5

KDHBv_S03400_158650.t1 KDHBv_S03386_157900.t1 Bv3_066760_yana.t1 YMoBv_S01567_76960.t1 Bv7_169500_ajxr.t1 Bv6_125470_heit.t1 tr|Q6SJP5|Q6SJP5_BETVU

8779 16177 Cell redox homeostasis / defense 212634 150765 Miscellaneous 16906 14785 10170 3898

Page 36 of 45

1053

5336

1.84

5.07

22973

11551

0.71

0.50

20754 964

7828 59

0.87 0.38

0.38 0.06

57371

4676

1946

0.69

0.42

126547 80118 Unknown 7606 4805 16526 14572 7815 5442 38274 58485 219140 329626

5429

2541

0.63

0.47

15349 7331 14439 43629 40646

5002 2816 6929 105516 106516

0.63 0.88 0.70 1.53 1.50

0.33 0.38 0.48 2.42 2.62

83000


Page 37 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48


Table 4- List of proteins showing changes in DRM vs. PM partitioning as a result of Fe-deficiency. The relative protein enrichment in the DRM was calculated for each treatment as the ratio between protein abundances in DRM and PM. The final relative partitioning was calculated as the ratio between the previous two. Only proteins identified and quantified with at least two unique peptides, showing significant changes (ANOVA, p ≤ 0.05 in at least one DRM vs. PM comparison), and with a final relative enrichment ratio ≥ 2 or ≤ 0.5 fold were considered for biological interpretation. Values in bold indicate p ≤ 0.05. Detailed information about identification, BLAST analysis and abundances is given in Table S2.

Beta vulgaris identifier

NCBI identifier

Protein description

Peptide quant.

Ratio DRM vs. PM +Fe

Ratio DRM vs PM -Fe

Ratio (DRM/PM) +Fe vs. (DRM/PM) -Fe

5 3 2

5 3 2

0.03 0.01 0.01

0.06 0.02 0.11

2.01 2.30 18.38

2

2

0.09

0.18

2.00

2 6

2 6

0.12 0.07

0.33 0.21

2.75 3.06

Peptide count

Proteins relatively enriched in -Fe DRM Cell redox homeostasis/Defense KDHBv_S05896_224260.t1 gi|15865451 Bv1_013710_wkrw.t1 gi|514804465 Bv1_013750_hurs.t1 gi|55701011 Carbon metabolism

Monodehydroascorbate reductase Peroxidase 12 precursor Peroxidase 12 precursor

KDHBv_S04244_183660.t1

gi|603221

Bv1_011870_sghe.t1 KDHBv_S08615_278650.t1 Amino acid metabolism Bv9_221970_kynh.t1 Signaling/regulation KDHBv_S04847_198070.t1

gi|306416861 gi|71000473

6-phosphogluconate dehydrogenase, decarboxylating 1 Alcohol dehydrogenase 2 Adenosylhomocysteinase

gi|225428336

Aminoacylase-1

2

2

0.01

0.05

5.41

gi|147770307

Heat shock cognate protein 80

7

6

0.37

0.75

2.02



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

KDHBv_S00519_207880.t1 KDHBv_S06794_244090.t1 Transport KDHBv_S05354_212150.t1 Signaling/regulation KDHBv_S07778_263100.t1 KDHBv_S01382_54610.t1 KDHBv_S08502_276450.t1 KDHBv_S02163_107980.t1 KDHBv_S12835_44330.t1 KDHBv_S18793_91550.t1 Carbon metabolism KDHBv_S03064_144220.t1 UMSBv_S15244_100420.t1 Protein synthesis KDHBv_S14321_60210.t1 KDHBv_S04688_194440.t1 KDHBv_S00113_26550.t1 KDHBv_S05482_215040.t1 Transport Bv5_108520_qgdi.t1 KDHBv_S01806_86470.t1 Lipid metabolism Bv9_210020_rnuz.t1 Cytoskeleton Bv1u_019010_zfif.t1 Unknown YMoBv_S01567_76960.t1 Bv_00790_wmgy.t1 Bv3_066760_yana.t1

gi|11131631 gi|15234435

Calreticulin precursor Calcium-dependent protein kinase

gi|303280251

Page 38 of 45

5 2

0.01 1.63

0.04 4.03

2.79 2.48

Protein NRT1/ PTR FAMILY 6.3 3 Proteins relatively enriched in +Fe DRM

3

1.84

4.39

2.38

gi|350536935 gi|225469784 gi|225469784 gi|462395986 gi|8099063 gi|291293221

14-3-3-like protein D 14-3-3 protein 10 14-3-3-like protein GF14 kappa 14-3-3-like protein 14-3-3-like protein 14-3-3-like protein

5 6 5 9 9 8

3 4 2 4 4 2

0.39 0.33 0.46 1.20 0.86 0.84

0.06 0.13 0.23 0.52 0.42 0.44

0.15 0.39 0.50 0.43 0.49 0.50

gi|533474 gi|217074908

Enolase Pyruvate kinase, cytosolic isozyme

6 4

6 4

0.14 0.54

0.06 0.27

0.42 0.50

gi|2245098 gi|6015063 gi|502160159 gi|3986695

60S ribosomal protein L15 Elongation factor 1-delta 40S ribosomal protein S13 60S ribosomal protein L12-1

2 2 2 2

2 2 2 2

0.09 0.15 0.79 0.56

0.02 0.07 0.40 0.28

0.16 0.47 0.50 0.50

gi|15231867 gi|557546529

Cation/H(+) antiporter 20 Potassium transporter 5

6 11

2 11

2.66 2.27

0.90 0.93

0.34 0.41

gi|57157824

Long chain acyl-CoA synthetase 8

2

2

0.15

0.05

0.31

gi|8928423

Tubulin beta-1 chain

12

2

1.23

0.53

0.43

2 2 2

2 2 2

0.44 4.03 2.02

0.19 1.90 1.03

0.44 0.47 0.50

gi|561018287 gi|215768932 gi|548841463


5 2

Page 39 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure Legends: Fig. 1- Effect of Fe-deficiency on the PM (A) and DRM (B) protein profiles as revealed by labelfree shotgun analyses. The normalized abundance of all detected proteins was calculated using Progenesis LC-MS software and ratios between the Fe-deficient and Fe-sufficient treatments were calculated for PM (A) and DRM (B). Numbers indicate the total number of detected proteins, those showing significant changes (p ≤ 0.05), identified and quantified with at least two peptides and above the threshold level (fold-change ≥ 2 or ≤ 0.5). Venn diagram (C) depicting the number of proteins changing as a result of Fe deficiency (p ≤ 0.05, number of peptides for identification and quantification ≥ 2 and fold-change ≤ 0.5 or ≥ 2) in PM, DRM and both preparations. Fig. 2 - Functional classification according to their GO: Biological Process of proteins changing significantly as a result of Fe-deficiency in PM (A) and DRM (B). Fig. 3- Glycerolipid, sterol and sphingolipid distribution in PM preparations from Fe-deficient and Fe-sufficient sugar beet roots. Data are means ± SD of 3 different PM preparations per treatment. Fig. 4- Effect of Fe deficiency on the relative partitioning of proteins in DRM (A) and Scatter plot of identified proteins (B). The normalized abundance of all detected proteins was calculated using Progenesis LC-MS software. The log2 value of the DRM vs PM ratio of Fe deficient samples was plotted against that of Fe-sufficient samples. Grey circles are proteins not showing significant differences (ANOVA, p ≥ 0.05), grey circles with black border are proteins showing significant differences (p ≤ 0.05) but less than two peptides used for quantification or identification, black circles are significant proteins identified and quantified with at least two



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

unique peptides and red circles are proteins showing significant differences in partitioning outside the fold-change thresholds (fold ≥ 2.0 or fold ≤ 0.5; Table 3).


Page 40 of 45

Page 41 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42


A

545 (Total)

B

545 (Total) 316 (p2)

52 (fold>2)

68 (fold>2)

C PM

Figure 1

DRM

ACS Paragon Plus Environment

DRM


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

A

Page 42 of 45

B Unknown 19%

Transport 15%

Unknown 21%

Signaling 23%

Miscellaneous 15%

Transport 15%

Miscellaneous 15%

Vesicle transport 14%

Signaling 26%


Figure 2


Page 43 of 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42


%

80

PM +Fe

70 PM -Fe

60 50 40 30 20 10 0

Sterols

Glycerolipids

Figure 3


Sphingolipids


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

A

Page 44 of 45

B

log 2(DRM/PM) 17 -Fe p>0.05 p≤0.05, pep2,fold2, fold>2

545 (Total) 402 (p2)

-8

-3 -8

30 (fold>2) -13

(DRM/PM)-Fe vs. (DRM/PM) +Fe

Figure 4


-18

2

12 log 2(DRM/PM) +Fe

TOC Page for 45 of 45only

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42


8

4

17 p>0.05 p≤0.05, pep2,fold2, fold>2 Unknown 19%

2 -18

-8

-3

2

Transport 15%

-18

Transport 15%

12

log2(DRM/PM) +Fe

Signaling 23%

Miscellaneous 15%

-8 -13

Unknown 21%


Miscellaneous 15%

Signaling 26%


ACS Paragon Plus Environment Photographs courtesy of Elain Gutierrez-Carbonell. Copyright 2015.

A Shotgun Proteomic Approach Reveals That Fe Deficiency Causes

Recommend Documents