mass spectrometry untangles plant membrane protein...

Trends in Plant Science

TRPLSC 1957 No. of Pages 15

Review

Mass Spectrometry Untangles PlantMembrane Protein Signaling Networks

Yanmei Chen1,* and Wolfram Weckwerth2,3,*

HighlightsMembrane receptors, kinases, andtransporters communicate with intracel-lular processes through protein interac-tion networks. Characterization of plantPM proteins – especially hydrophobicproteins – remains challenging despiteadvances in separation and analysistechniques.

Rapid advances in MS instrumentationand data analysis have enabled markedprogress in deciphering the membraneproteome and mapping protein interac-tion partners, leading to a better under-

Plasmamembranes (PMs) act as primary cellular checkpoints for sensing signalsand controlling solute transport. Membrane proteins communicate withintracellular processes through protein interaction networks. Deciphering thesesignaling networks provides crucial information for elucidating in vivo cellularregulation. Large-scale proteomics enables system-wide characterization ofthe membrane proteome, identification of ligand–receptor pairs, and elucidationof signals originating at membranes. In this review we assess recent progress inthe development of mass spectrometry (MS)-based proteomic pipelines fordetermining membrane signaling pathways. We focus in particular on currenttechniques for the analysis of membrane protein phosphorylation and interac-tion, and how these proteins may be connected to downstream changes ingene expression, metabolism, and physiology.

standing of PM protein complexes.

Analysis ofmembraneprotein phosphor-ylation under specific cellular conditionsis crucial for elucidating the molecularmechanisms underlying signal sensing,transport, and metabolic processes.Phosphoproteomics allows unbiasedlocalization and site-specific quantifica-tion of in vivo protein phosphorylation,thus facilitating the dissection of mem-brane signaling networks.

1State Key Laboratory of PlantPhysiology and Biochemistry, College ofBiological Sciences, China AgriculturalUniversity, Beijing 100193, China2Department of Functional andEvolutionary Ecology, MolecularSystems Biology (MOSYS), University ofVienna, Vienna, 1090, Austria3Vienna Metabolomics Center (VIME),University of Vienna, Vienna, 1090,Austria

*Correspondence:[email protected] (Y. Chen) [email protected](W. Weckwerth).

Simple Beginning: The Significance of PM ProteinsPMs play a crucial role in the cellular structure by providing a physical barrier between cells andtheir environment. They constitute a sensor for modifications to the cellular environment and aplatform for intricate orchestration of signal transduction, allowing translation of external signalsinto finely tuned adaptive responses. Plant PMs are structurally highly complex and are very dif-ferent from those of animals because they are surrounded by cell walls, and this diversity allowsthem to have distinct regions that can fulfill different roles, despite their fluid nature.

Membrane proteins that are integrated into – or attached to – the lipid bilayer are the main medi-ators of plant membrane functions. For instance, transmembrane proteins carry out selectivetransport of small molecules, thus establishing controlled exchange between the inside and theoutside of the cell. Such transport processes are mediated by dozens of membrane-localizedtransporters [1,2]. Analyses of multiple complete genome sequences estimate that 20–30% ofthe cellular proteome consists of transmembrane proteins [1,3]. Membrane proteins are associ-ated with PMs and other membrane-containing organelles such as plastids, chloroplasts andmitochondria. We focus here mainly on PM proteins not only because they represent a largegroup of plant membrane proteins, suggesting that they have special importance in plant signaling,but also because recent research has uncovered interesting new aspects of biological regulation.

Cellular sensing, signaling, and transporter regulation mediated through interactions with mem-brane proteins are of utmost importance to plants. Plant PMs contain many kinases, receptors,and enzymes, which are key regulators of these crucial cellular processes. Membrane-localizedreceptor-like kinases (RLKs) such as BRI1 and FLAGELLIN INSENSITIVE 2 function in signalingpathways [4,5]. Arabidopsis thaliana RLKs comprise a large family with N610 members – N75%of these contain transmembrane domains (TMDs). Several leucine-rich repeat (LRR)-RLKs haveproven functional roles in regulating plant growth, morphogenesis, and responses to the environ-ment [6–9]; however, the functions of most RLK family members remain unknown.

Trends in Plant Science, Month 2020, Vol. xx, No. xx https://doi.org/10.1016/j.tplants.2020.03.013 1© 2020 Elsevier Ltd. All rights reserved.

https://doi.org/10.1016/j.tplants.2020.03.013https://doi.org/10.1016/j.tplants.2020.03.013https://doi.org/10.1016/j.tplants.2020.03.013https://doi.org/10.1016/j.tplants.2020.03.013

GlossaryAffinity purification MS (AP-MS): astrategy for examining a subset ofproteins in a complex sample bypurifying protein interaction partnersusing an affinity tag coupled to qualitativeand quantitative mass spectrometry(MS) workflows.Bottom-up proteomics: a proteomicapproach that proceeds through apeptide-to-protein pipeline. Proteinmixtures first undergo tryptic digestioninto smaller peptides. Their sequencesare then identified through fragmentationof peptides in a mass spectrometerfollowed by measurement of thesefragment ions.Detergent-resistant membrane(DRM) fraction: the non-solubilizedfraction of plasma membranes (PMs)extracted by mild detergent underdefined conditions. Membrane proteinsidentified within this fraction arefrequently termed lipid raft proteins.Interactome: the complete signalingcomponents of protein–proteininteractions found within a biologicalsample.Membrane-associated proteins:proteins attached or anchored tomembrane surfaces but not traversingthe bilayer.Membrane microdomain: a specificsubdomain in the plane of a membranebilayer that exhibits a composition andbiological function distinct from those ofthe surrounding membrane.Microdomains are highly dynamic andcomprise specific proteins and lipids.Membrane protein interaction:protein interactions among membraneproteins, and interactions of membraneproteins with soluble proteins.Metal oxide affinitychromatography (MOAC): a peptideor protein purification technique thatrelies on the strong affinity of thephosphate group for specific metalscovalently bound in metal oxidecoordinators.15N-metabolic labeling: the use ofheavy nitrogen (15N)-labeled biologicalmaterials as nitrogen tracers for isotoperatio MS in quantitative proteomics andmetabolomics. A quantitative techniquethat is based on isotopic difference in themixture of labeled and unlabeledsamples.Phosphoproteome: the complete setof phosphopeptides andphosphorylation sites found within abiological sample.


Owing to the relatively low abundance of membrane proteins and the short timescale of thesedynamic signaling events, it is notoriously challenging to analyze membrane protein function[10]. Although proteomic technologies for the analysis of soluble proteins (see Glossary) haveprogressed rapidly in recent years, membrane proteins have lagged behind and are typicallyunder-represented in datasets. Gel-based proteomic approaches are limited in their ability toresolve basic or hydrophobic proteins as well as those with greater than three transmembraneregions. In addition, limited dynamic range and difficulty in identifying low-abundance proteinsmake lack of sensitivity a key issue.

With advances in high-throughput MS, shotgun proteomics and shotgun phosphoproteomicshave increasingly become powerful alternatives to gels for analyzing complex protein samples[11–16]. Compared with intact protein analysis in top-down proteomics, peptides are easierto work with because they are readily solubilized and separated before MS, and are easily disso-ciated to produce useful fragmentation ions for identification. The breakthrough in efficientproteome-wide analysis of membrane proteins came with development of suitable membraneisolation methods, and extensive experimental studies of membrane proteomes have beenreported in Homo sapiens, Mus musculus, and yeast [17–19]. In plants, large-scale analysisof several membrane proteomes is advancing [20–22] and detailed views of membranephosphoproteomes have been generated [16,23,24]. These studies represent a fundamentallydifferent approach to studying cell signaling. However, membrane proteomics remains amultifaceted, rapidly developing, and open-ended endeavor. Despite tremendous recentsuccess, membrane proteomics still faces significant technical challenges.

In this review we focus on MS-based bottom-up proteomics methodologies for identifyingplant membrane proteins, phosphorylation, and protein–protein interactions (Figure 1). Specifi-cally, we briefly cover advances in MS for determining membrane phosphorylation and addressthe need for future experimental strategies to better understand membrane signaling networkson a proteome-wide scale.

Challenges in the Preparation of Membrane ProteinsCellular membranes consist of a phospholipid bilayer and associated proteins which are catego-rized according to how they associate with the membrane: integral or intrinsic proteins traversethe membrane, whereas peripheral or extrinsic proteins are associated with components residingin the membrane but do not themselves reside within the membrane [25]. Integral membraneproteins are fundamental to many physiological processes and constitute 20–30% of the cellularproteome [26]. Most integral membrane proteins have TMDs; these proteins are often studied byMS because they act as cell-surface markers, receptors, or adhesion factors, and their cytoplas-mic domains operate in cellular signaling pathways [27–29]. Some membrane-anchoredproteins are tethered to the membrane bilayer by a lipid anchor. Such proteins have hydrophiliccharacteristics of soluble proteins and can be dissociated upon cleavage of the anchor usingphospholipases [26].

Integral membrane proteins containing multiple TMDs tend to aggregate and precipitate uponremoval from the lipid bilayer, and this appears to be exacerbated by the presence of reactivethiolates of reduced cysteinyl residues. Unfortunately, TMDs are poorly represented in bottom-up proteomics because their lack of charged residues leads to very poor peptide recovery;however, they can be accessed using top-down proteomics [30]. In addition, varying sizes ofcytoplasmic and extracellular loops, as well as post-translational modifications (PTMs), conferdiverse physicochemical properties to integral membrane proteins, resulting in sample losseswith some sample preparation methods. Therefore, efficient extraction and digestion methods

2 Trends in Plant Science, Month 2020, Vol. xx, No. xx

Phosphoproteomics: the scientificstudy of all phosphoproteins foundwithin a biological sample. Specifically,phosphoproteomics is the ‘systematicstudy of the unique phosphorylationnetworks that specific cellular processesleave behind’.Soluble proteins: cytoplasmic proteinsthat are soluble in aqueous solutions.Strong cation exchange (SCX):chromatography using the strong cationelectrostatic interactions betweenionized groups of the stationary phaseand peptide counterions in the sample ata given pH. Elution of target analytes isachieved by modulating the strength ofinteractions using salts or pH buffers.Tandem affinity purification (TAP):the use of multiple affinity tags permitstwo consecutive or tandem affinitypurifications for protein–proteininteraction analysis.TandemMOAC: a two-stepprocedurecombining Al(OH)3-basedphosphoprotein enrichment and TiO2-based phosphopeptide enrichmentfollowing phosphoprotein trypticdigestion.Top-down proteomics: a proteomicapproach that identifies intact proteins incomplexmixtureswithout prior digestioninto their corresponding peptides.Diverse sources of intramolecularcomplexity are thus preserved.


based on soluble peptides from loop regions can increase integral membrane protein cover-age [31].

Although peripheral or othermembrane-associated proteins bound to the membrane surfacecan be readily solubilized by treatment with high-pH or high-salt buffers, they do not generallypresent the same analytical challenge as integral membrane proteins [32]. When consideringmembrane protein interaction partners, it is clear that proteins associate with the PM in manycomplex arrangements. Indeed, traditional biochemical approaches for characterizingmembraneproteins, such as antibody purification and native PAGE, involve isolating these hydrophobic pro-teins, which is laborious and technically challenging [33]. Plant membrane proteins are more dif-ficult to isolate than those from mammalian cells because of the complexities and variation ofplants and their tissues, and especially their wide dynamic range. Therefore, removal of highlyabundant soluble proteins such as the RuBisCO large subunit and chlorophyll a/b bindingproteins can improve membrane protein identification.

MS to Determine the Complete Membrane ProteomeSeveral recent technological advances, including improved sample preparation, instrumentation,and better liquid chromatography (LC) performance, have led to a substantial increase in PM pro-tein representation in large datasets [34,35]. However, membrane preparation and subsequentsteps are decisive for improving membrane protein coverage. Purification of membrane proteinsoften involves cell lysis and sequential centrifugation steps to remove cell debris and isolate themembrane from the soluble fraction. Further purification can be accomplished by ultracentrifuga-tion followed by density gradient centrifugation (e.g., using sucrose, sorbitol, or Percoll) [36] orfilter-assisted digestion [37]. Accordingly, highly purified microsomes can be prepared byan aqueous–polymer two-phase partitioning step in which membrane capture is followed by'shaving' (peptides are cleaved from the intact microsome surface) [16].

Typically, detergents are used to promote membrane protein solubility, for example a low con-centration of SDS or Brij-58 is used to solubilize the membrane-enriched fraction. However,these reagents require multiple processing steps to remove them before LC/MS analysis,which can result in sample loss and quantitative variability. Alternatively, membrane proteinscan also be solubilized using organic solvents in a methanol-bicarbonate buffer [27]. InArabidopsis this approach resulted in a twofold increase in the detection of membrane proteinssuch as transporters and RLKs [27]. Although membrane proteins can be easily enriched bycentrifugation, this lacks the resolution to provide highly purified PM fractions without substantialcontamination by membranes from other subcellular organelles [38]. PMs share similar physio-chemical properties with other membrane components and tend to exist as multiple structures.Fractionation procedures that reduce the protein complexity of themembrane proteome increasethe chances of identifying proteins of low abundance.

To improve MS detection capabilities, isolated proteins or peptides can be fractionated usingmultidimensional chromatography, strong cation exchange (SCX) LC has proved to beeffective in gel-free multidimensional analyses of complex peptide mixtures [19]. This stepfacilitates an increase in proteome coverage by partitioning peptides into fractions of reducedcomplexity that can be more comprehensively sampled by the mass spectrometer. The exploita-tion of more than one property of a peptide is a powerful chromatographic technique for proteinidentification from complex samples [27,39].

Optimization of peptide chromatography and processing can also facilitate analysis of low-abundance proteins in complex mixtures, for example by the use of very long monolithic columns

Trends in Plant Science, Month 2020, Vol. xx, No. xx 3

Membrane samplepreparation

HPLC

ESI

Ser624DSer624AWT Mutant OE

Phos-RLK

CBB-RLK

-log1

0 P

valu

e

Enrichment factor

Transport

Condition

Kin

ase

activ

ity

GQLFS(p)PSGFR

Protein purification

Protein digestion

TrendsTrends inin PlantPlant ScienceScience

Figure 1. Membrane Proteomics Workflow. (i) Membrane proteins are extracted from a sample of interest. Enrichment by affinity purification may be performed.Proteins are digested into peptides that are optionally enriched for modifications. (ii) After separation by liquid chromatography, peptides are ionized and injected into ahigh-resolution mass spectrometer for fragmentation and detection. (iii) Raw data for the proteome, phosphoproteome, or interactome are collected for further analysis.The highest scoring spectrum match is taken as a candidate for the identity of the sequence of the peptide, as well as the mass and locus of the modification. A targetdecoy search is used to control false positive rates. Causative proteins can be identified through network analysis and validated by in vivo and in vitro functionalexperiments. Abbreviations: CBB, Coomassie brilliant blue staining; ESI, electrospray ionization; LC-MS, liquid chromatography with mass spectrometry; OE, overexpression; phos, phosphorylated; RLK, receptor-like kinase; WT, wild type.


with very high separation capacities [40]. This has helped in deep probing of the membraneproteome and represents an active area of research that holds great promise [19,41]. Inspectionof the prefractionation screen chromatogram allows fine adjustment of the sample elution gradi-ent and intelligent pooling of peptide fractions based on peak distributions and intensities [23]. A


Image of Figure 1


major advance in improving membrane proteome coverage is the use of less-hydrophobic resins,such as C4 columns instead of the commonly used C18 material, and this has been valuable forthe separation of very hydrophobic peptides [42]. The application of heat to reverse-phase LCcan also enhance the identification of extremely hydrophobic proteins and peptides [43]. Inaddition to improving the performance of affinity chromatography, use of alternative enzymesto trypsin or Lys-C for protein digestion as well as modifications to the digestion protocol canincrease the number of peptides available for MS analysis [44,45].

Membrane proteins are highly dynamic and vary between different cell types. Comparison of theproteome with transcriptome profiles is a good way to determine whether a MS analysis iscomprehensive. Although transcript and protein levels might not correspond entirely, they shouldprovide a useful qualitative benchmark of the portion of the protein-encoding genome that couldpotentially be identified in an MS experiment [46]. For accurate peptide identification, the highest-scoring peptide spectrum match (PSM) is taken as a candidate for the identity of the peptide.Most workflows control the false discovery rate using a target decoy approach [47]. Fragmenta-tion spectra are searched not only against the target database but also against a decoy databasedesigned to produce false positive PSMs.

MS-based Proteomics to Analyze Membrane MicrodomainsMS is also effective for analyzing proteins ofmembranemicrodomains named lipid rafts, whichrepresent a very active area of membrane research. These specialized membrane domains havebeen studied by preparing the detergent-resistant membrane (DRM) fraction from purifiedPM fractions [48]. However, plant DRMs have been understudied relative to their animal counter-parts. Proteomic analysis in several species has demonstrated that such DRMs contain 5–10%oftotal PM proteins in plants [49–51]. All these proteome data point to the importance of membranemicrodomain proteins and their dynamics during various signaling processes. These investiga-tions suggest that the DRM protein profile is distinct from the whole PM proteome. However,many proteins that are clearly not related to the PM or its microdomains copurify with DRMs.One way to tackle this problem is through quantitative proteomics in combination with comple-mentary biochemical approaches. Quantitative methods including label-free quantification, stableisotope labeling, and isotope dilution have been applied to DRM proteome analysis, showing thatmany DRM proteins localize to distinct microdomains of plant cells [52,53]. Microdomain-specificlocalization of DRM proteins functions in the establishment of cell polarity, membrane trafficking,and specialization of host membranes upon interaction with microbes [54].

Current proteomic technology allows identification of N1000 DRM proteins and the relationshipwith their localization in specific membrane microdomains [53]. Other evidence from quantitativeproteomic analysis indicates that rapid and discrete modifications of proteins associated withDRMs take place upon stimulation with pathogen-associated molecular patterns [55]. Consistentproteomic data were obtained for Arabidopsis cells following flg22 stimulation, tobacco cellstreated with the elicitor cryptogein [56], and rice challenged with chitin for 5–15 min [57]. Thesimilarity of the timing of such modifications (5–15 min) deserves a mention, as does the factthat different complementary approaches further confirm the involvement of most identifiedDRM proteins in the early steps of immune signaling as well as their association with PMmicrodomains [4,58,59]. These quantitative proteomic analyses have revealed a restricted listof proteins that are specifically enriched in DRMs that define specialized functional structures.

MS-based Methods to Determine Membrane Protein Interaction NetworksSystematic large-scale analyses of protein–protein interactions among membrane proteins areclearly under-represented despite the utmost importance of membrane protein complexes in



transport processes as well as in signal perception and transduction. In-depth characterization ofindividual PM protein complexes in plants suggests a highly modular and dynamic system inwhich protein–protein interactions are adjusted according to cellular conditions [60].

MS is one of the most powerful tools for identifying and quantifying proteins, determiningPTMs, and for exploring protein–protein interactions (Figure 2A,B). Plant membrane proteininteractions can be complex and are a crucial determinant of function that can be effectivelyinvestigated using MS. Analysis of protein complexes through immunopurification (IP) followedby MS is widely used because of its high throughput and sensitivity. Suitable controls and quan-titation are required for distinguishing bona fide binding proteins from background contaminants.


Figure 2. Large-Scale Mass Spectrometry (MS)-Based Approaches for Unraveling Phosphorylation-DrivenMembrane Signaling Networks. (A) Targetedor untargeted interactome strategy for unraveling protein interaction partners. Protein interaction partners are isolated using one-step or tandem tag affinity purification, theeluate is subjected to MS for proteomic analysis. (B) Global membrane phosphoproteome strategy for discovering putative substrates of protein kinases andphosphatases. Experimental data ultimately need to be combined with systems biology analysis which translates the separate, large-scale datasets into signalingnetworks. Predicted connections within and between signaling cascades need to be experimentally verified by, for instance, analysis of protein complexes and analysisof kinase or substrate mutant plants. Abbreviation: LC-MS, liquid chromatography with mass spectrometry.


Image of Figure 2


IP-MS has recently been applied in plant aquaporin interactome studies [61], enabling thedetection of more than pairwise interactions, although the technique is limited by the number ofbait proteins tested. Owing to the high dynamic range, membrane protein–protein interaction isdifficult to study using traditional methods such as IP using specific antibodies that recognizethe bait proteins. Moreover, specific antibodies are often not available; in contrast to the animaland human fields, plant protein antibodies are relatively scarce, hindering high-throughputanalysis of the plant protein interactome.

Recent affinity purification MS (AP-MS) methods have used epitope tags on target proteinsdirectly in the genome as affinity capture probes for identifying interacting proteins, withoutrequiring specific antibodies for each new bait protein (Figure 2A) [62]. The principle of AP-MSis to isolate the bait from the plant sample by affinity purification under near-physiologicalconditions; binding partners of the bait can then be recovered from the sample and identifiedby MS. One key attribute of MS-based approaches for such analysis is that they can potentiallybe applied to any membrane protein in vivo given that this technique requires no specificantibodies. Despite the high numbers of membrane protein interaction partners identified usinga split-ubiquitin interaction screen [1], this method investigates interactions ectopically in yeast,whereas AP-MS is closer to the endogenous system of plants. The resulting datasets also differfrom those of other approaches that deliver pairwise interactions, and AP-MS yields co-complexdata (both direct and indirect interactions). Therefore, different techniques tend to deliver comple-mentary datasets.

Initial studies of membrane protein interactions by MS were small in scale and focused on well-characterized proteins of interest, such as analysis of the Arabidopsis cell-cycle interactome[63]. This is because the short, single-step purification used in AP-MS experiments leaves behindconsiderable amounts of background contaminants. Compared with single-step methods,tandem affinity purification (TAP) detects membrane protein–protein interactions with a highersignal-to-noise ratio and also allows purification of protein complexes [64]. The TAP tag originallydeveloped in yeast was improved for plants, yielding the TAPi tag [62]. This was further boostedthrough implementation of the GS tag (a fusion between protein G and streptavidin-bindingpeptide) which is superior to the original TAP tag in both specificity and complex yield [65].Applying this method to Arabidopsis cells, Pauwels and coauthors revealed that TOPLESS-related proteins affect multiple jasmonate signaling pathways through interaction with specificadaptor proteins [66].

A prior fractionation step for baits present in specific subcellular fractions can increasemembrane protein interactome coverage [67]. The development of easy purification proto-cols and ultra-sensitive MS has allowed widespread use of TAP-MS for screening the mem-brane interactome of plants. Moreover, studying protein complexes in their developmentalcontext through the isolation of PMs has now become feasible. TAP has been successfullyadopted for screening membrane partners of the circadian multiprotein complex and theprotein kinase interactome [68–72]. These investigations have uncovered novel signalingevents involved in the interaction networks of target proteins such as cyclin-dependentkinase, phytochrome, and target of rapamycin (TOR) kinase. Another interactomic analysisusing affinity enrichment effectively identified interactive proteins of membrane-bound recep-tor complexes such as LRR kinases [73]. The two-step TAP system reduces backgroundcontaminants and identifies stable interactions, but unlike single-step AP often does notretain transient interactions. These approaches are complementary to genetic methodssuch as yeast two-hybrid screens in which proteins are overexpressed and must be ableto interact in the yeast nucleus. Combined with quantitative approaches, TAP-MS can



identify condition-specific protein interactions and also allows a dynamic view of the mem-brane protein interactome.

Quantitative approaches permit the use of mild IP protocols and allow specific membrane-binding proteins to stand out by their quantitative signature, even among a very large backgroundof proteins. Using high mass accuracy MS, a newly reported integrative analysis combiningphosphoproteomics and interactomics has identified dozens of novel signaling components ofTOR kinase networks [70]. Overall, these MS-based approaches could prove valuable for kinaseinteraction network studies given that they enable the proteome-wide discovery of both up- anddownstream network components. Because relying on the identification of proteins by pull-downleads to many false positives, it is crucial to distinguish background binders from significantlyenriched bona fide interactors. Statistical tests can identify true interactors but require a controlfor comparison [74,75]. Owing to its quantitative nature, AP-MS can probe not only steady-state interactions but also dynamic rewiring upon stimulation by internal or external stimuli. Inaddition, because this approach isolates and enriches proteins, it can identify PTMs, either withor without further PTM enrichment, enabling in-depth assessment of the role of these modifica-tions in regulating interactions. An AP-MS analysis of Arabidopsis ubiquitylation enabled directdetection of 950 ubiquitylated proteins, N100 of which had previously escaped detection [76].In this regard, MS has the unique ability to identify and functionally profile previously unknowninteractions from a complex biological state.

Cytoscape has emerged as the de facto standard for general-purpose network analysis [77].Through its plugin infrastructure such as Stich, it provides a wealth of analysis and visualizationtools, often integrating quantitative proteomic technologies with interaction networks. Indepen-dently of the biological question, the true value of MS in analyzing membrane protein interactionslies in its complementarity with biochemistry, molecular biology, and optical approaches.Quantitative MS approaches and continuous developments in bioinformatics can provide aprecise definition of temporal changes in membrane protein interactions and are expected tocontinue expanding the range of AP-MS applications to plant biology studies.

Proteomics to Understand Protein Phosphorylation Signaling Networks AcrossPlasma MembranesTechniques to Improve Phosphoprotein IdentificationProtein phosphorylation is one of the most important PTMs in plant cells, and is involved in manybiological functions including signal transduction, differentiation, hormone perception, transfor-mation, and metabolism. All these signal transductions lead from receptor proteins at the PMto transcription factor proteins in the nucleus (Figure 3). More than 5.5% of the Arabidopsisgenome comprises genes encoding kinases, and more than half of these gene products arelocated at membranes, suggesting a complex phosphorylation signaling network in plants [1].The regulation of membrane proteins through phosphorylation has been studied extensivelyusing purified proteins and by site-directed mutagenesis of target phosphorylation sites.

MS analysis of membrane protein signaling has gained attention in recent years and is revealingsignaling pathways at unprecedented levels [20,28,78]. Membrane phosphoprotein analysis byMS is nevertheless more challenging because modified peptides may be present at lowabundance, and the modification may be labile during fragmentation. In the context of signaltransduction, this challenge is exacerbated by the generally low copy number of many proteinswith pivotal roles in signaling cascades [1]. Fortunately, low abundance can often be alleviatedby phosphopeptide purification – this purification captures all modified peptides of interest andno others (Figure 2B) [15,79–83]. In practice, phosphopeptides can be enriched more than



Figure 3. Membrane Protein-Mediated Signaling Pathways. Signal processing and amplification by plasma membrane-proximal events are followed bycommunication of the signal to different cellular compartments. This process involves activation of multiple signal cascades by receptors, differential proteinphosphorylation, and crosstalk between signaling pathways and feedback loops to ensure optimal signaling output. (A) Binding of receptor-like kinases (GPCRs orother receptors) to their cognate ligands at the cell surface results in receptor dimerization and autophosphorylation. Phosphorylated RLKs then recruit andphosphorylate signaling mediators such as membrane channels or pumps. Phosphorylation of membrane channels may activate their transport activities, potentiallyleading to increased concentrations of cytosolic ions such as Ca2+ or K+. Increased Ca2+ as a second messenger further leads to activation of membrane-associatedkinases such as CDPKs, CIPKs, SnRKs, and TOR kinases. Assembly of these signaling complexes subsequently activates multiple signaling cascades involving MEKs,MAPKKKs, PP2Cs, 14-3-3 proteins, cytosolic enzymes, and transcription factors [28,108,109]. (B) The SNARE protein interacts with membrane syntaxin and can bephosphorylated by protein kinases within a few minutes in response to flagellin elicitation, subsequently triggering multiple defense responses [110]. Abbreviations:GPCR, G protein-coupled receptor-like kinase; CBL, calcineurin B-like proteins; CIPK, CBL-interacting protein kinase; MEK, mitogen-activated protein kinase kinase;MAPKKK, mitogen-activated protein kinase kinase kinase; P, phosphorylation; PP2C, protein phosphatase type 2C; RLK, receptor-like kinase; SNARE, solubleN-ethylmaleimide-sensitive factor attachment protein receptor; TF, transcription factor.


100-fold by resins that chemically coordinate the phosphogroups, for example by the use of metaloxide coordinators such as alumina and titanium dioxide [82–84]. The term metal oxide affinitychromatography (MOAC) was coined by us [83]. Furthermore, combined subsequent enrichmentof phosphorylated proteins and their corresponding phosphorylated tryptic peptides – tandemMOAC – increases the coverage of low-abundance phosphoproteins [15,80,81,85]. Novel signalingtargets for mitogen-activated protein (MAP) kinase and SnRK1-TOR kinase were successfully de-tected using tandemMOAC [15,81]. In plantmembrane phosphoproteomics, MOAChas been suc-cessfully applied to phosphopeptide enrichment, yielding thousands of phosphoproteins [16,23,86].However, generating large-scale datasets for membrane phosphoproteins usually requires fraction-ation of samples from a larger amount of plant material. A major breakthrough in the detection ofphosphorylated membrane proteins was recently reported from Arabidopsis roots, where N2000phosphorylation sites were identified from 1 g of plant material using a one-step membrane fractionenrichment strategy [87]. This method is promising for the identification of high-quality membranephosphopeptides in large-scale proteomics and is applicable to low amounts of plant materials.


Image of Figure 3


Quantitative PhosphoproteomicsGlobal membrane phosphoproteomic analysis follows the same generic shotgun proteomicsworkflow as phosphorylation-specific analyses, with an enrichment step inserted either at themembrane protein level or more typically at the peptide level. Even for phosphopeptide-enriched samples, confident identification of phosphopeptides and unambiguous localization ofphosphorylation sites is more difficult than the identification of unmodified peptides becausesearch engines need to consider additional possibilities. Rigorously determined false discoveryrates for phosphorylation determination are important for accurate confirmation, especiallywhen phosphorylation studies are intended to serve as community resources [88,89]. Highmass accuracy in fragmentation spectra is required for unambiguous localization of phosphory-lation sites, and this increases the confidence of both peptide identification and phosphorylationsite localization. However, the goal of phosphorylation studies is usually not only to catalog manysites but also to determine how they change during cell signaling. This necessitates quantitativemapping or possibly time-resolved studies of phosphoproteomics [6,16,23,90]. To investigatedynamic phosphorylation signaling events in early plant hormone stimulation, we combinedMOAC [83] with a novel data processing and mining technique called mass accuracy precursoralignment (MAPA) [91], which enables the alignment of MS peaks without dependence on data-base identification, and we applied this to the analysis of membrane phosphoproteins involved inphytohormone signaling [16,91]. In addition to label-free quantitative approaches, stable isotopelabeling such as 15N-metabolic labeling-based phosphoproteomics allows accurate quantifi-cation of transient phosphorylation changes of membrane proteins [23]. These approachescoupled with genetic analysis were used to reveal several kinase–substrate networks involvedin ethylene signaling relay pathways [92,93]. Using a similar strategy, Wu and colleagues charac-terized the in vivo function of a transmembrane sucrose-induced receptor kinase (SIRK1) in thecontext of sucrose resupply signaling in Arabidopsis seedlings [94]. In combination with singleion-reactionsmonitoring (SRM) on a triplequadmass spectrometer, target sites for specific phos-phoproteins could be efficiently analyzed across a wide range of samples [95]. A very importantand currently underestimated problem is to reference changes in quantitative protein phosphor-ylationmeasured by phosphoproteomic techniques to quantitative changes in the correspondingunphosphorylated proteins. This can be achieved by comparing phosphoproteomic and proteo-mic data from the same sample, as in our recent global quantitative phosphoproteomic studies[15,96,97]. These investigations indicate that quantitative membrane phosphoproteomics hasenabled a deeper characterization of protein kinase signaling networks.

In contrast to the large number of kinases, the Arabidopsis genome encodes about 150 proteinphosphatases [98]. A previous phosphoproteomic study revealed that BRASSINOSTEROIDSIGNALING KINASE 8 is involved in the regulation of SUCROSE PHOSPHATE SYNTHASE(SPS) through activation of a phosphatase, which in turn dephosphorylates and thus activatesSPS [78]. A more recent phosphoproteomic analysis using a PROTEIN PHOSPHATASE 2C(PP2C) mutant found an enrichment of several kinase phosphorylation motifs [99]. These reportssuggest that phosphatases control phosphorylation sites by directly or indirectly modulating theactivities of enzymes or kinases. Overall, quantitative membrane phosphoproteome analysisprovides an informative perspective on cell signaling and will be crucial for modeling signalingnetworks in a systems biology approach.

Functional PhosphoproteomicsMutational analysis of protein kinases or specific phosphorylation sites will provide clues abouttheir involvement in plant signaling. However, a key attraction of MS is that it can quantifykinase-associated changes in the entire signaling network. Thus, quantitative phosphoproteomicscan be applied to the functional analysis of many membrane (or membrane-associated) protein



kinases, such as BRI, BAK1, abscisic acid (ABA) receptors (PYR1/PYLs), CIPKs, SnRKs, andRLKs (Figure 3) [6,15,28,100]. In combination with genetic analysis, phosphoproteomic studieshave revealed several substrates of SnRK2 kinase by quantifying phosphorylation changesbetween Arabidopsis snrk mutants and wild-type materials [101,102]. A recent functionalphosphoproteome study reported that sucrose stimulus activates SIRK1, and that SIRK1 interactswith and phosphorylates another coreceptor membrane kinase named QSK1 to regulate thephosphorylation state of aquaporins [29].

Identification of a range of possible substrates has provided insight into the role of protein kinases;however, more mechanistic approaches are now being adopted that attempt to integrate allobserved changes rather than merely focusing on the most dramatic or most reproducible.High-quality phosphorylation datasets containing thousands of quantified phosphorylation sitesare an attractive resource for system-wide analysis of signaling networks (Figure 3). For example,recent analysis of TOR kinase phosphorylation provided evidence that network modeling canidentify crucial phosphorylation sites that are key control points for membrane-localized ABAreceptors [28]. Similar studies also demonstrated that MAPKs activated by nutrient limitationphosphorylate PM channel boron transporters, which in turn control the polarization of borontransport in Arabidopsis roots [80]. Integration of information about cell type-specific kinaseexpression, cellular localization, substrate phosphorylationmotifs, and other functional interactiondata with global phosphorylation data is useful for predicting kinase–substrate relationships insignaling networks (Figure 3) [6,96,97,103].

Connecting kinasemotif and phosphorylation-dependent binding domainswith in vivo phosphor-ylation data further helps in matching phosphorylation sites to their putative up- or downstreamkinases, and computational analysis of phosphoproteomic data can be used to investigate themolecular evolution of signaling networks [79,86]. In addition to finding partners for full-lengthproteins, membrane proteomics is uniquely suited for exploring phosphorylation-mediated inter-actions in signaling networks. The power of quantitative proteomics to discover phosphorylation-induced protein interactions was recently demonstrated in an example inspired by interactomicanalysis of TOR kinase signaling networks [70]. Proteomics and interactomics reveal comple-mentary subspaces of kinase signaling pathways, enabling system-wide discovery of networkcomponents for deciphering cellular membrane signaling networks (Figure 2A,B).

Phosphoproteomics methods are becoming the methods of choice for decoding membraneprotein functions at a system level. As shown here, membrane phosphoproteomics is nowused for mapping signaling networks in considerable depth, for delineating cellular control points,and, owing to its unbiased nature, for uncovering unexpected biological connections. However,future progress in MS-based membrane proteomics and phosphoproteomics necessitatesfurther significant improvements in instrumentation before we can comprehensively and routinelyanalyze whole membrane proteomes or phosphoproteomes. Furthermore, increasing thedynamic range of phosphoproteomics will be necessary to determine substoichiometric phos-phorylation sites on low-abundance membrane proteins in the presence of high-abundancephosphopeptides.

Other PTMsAlthough MS can in principle analyze all PTMs, phosphorylation is by far the most extensivelystudied PTM of membrane proteins; hence we focus here on explaining methodologies andkey applications of phophoproteomics. Other PTMs such as nitrosylation and glycosylation arealso important, and a range of approaches have been developed to characterize them usingMS; however, high-throughput analysis is only beginning. Previous glycoproteomic analysis


Outstanding QuestionsHow can we increase membraneproteome or phosphoproteomecoverage for MS analysis?

How can we determine if MS analysis ofa given sample is truly comprehensive?

How can we control for false positivesduring peptide identification? Whatare the advantages of gel-free proteo-mics relative to gel-based techniques.

How can peptide or phosphopeptidequantification accuracy be increased?

Can the TAP-MS approach be appliedto any plant species and any biologicalstate interactions?

What determines the probability ofPTMs during the identification ofprotein phosphorylation sites?

Can protein interaction partners bescreened using an MS method?

How do membrane protein kinasesmodulate intercellular signalingnetworks?

Can the molecular function of anindividual phosphorylation site becharacterized using MS methods?

Can networks of kinases and theirsubstrates be deciphered at aproteome-wide level.

What defines the resolution andprecision of large-scale phosphoryla-tion site mapping?


using lectin enrichment of N-linked peptides yielded 318 N-glycosylation sites in proteinsextracted from tomato fruits, but there was less enrichment in membrane proteins [104]. Effortsdirected at O-linked complex glycans are gaining ground. The development of bioinformaticstools has generated databases of plant membrane proteins and their PTMs that are accessibleon the Internet [105,106]. These can be useful for identifying networks of proteins that are highlyconserved among species and for suggesting possible roles for unknown proteins. However, inmany cases these data are predicted by programs or come from in vitro experiments, which oftengenerate false positive identification rates. In addition, subcellular predictions occasionally failbecause some proteins do not have an exclusive destination but are dually targeted betweencellular organelles [107]. Therefore, no attempt has been made to extrapolate 'consensus'targeting information. Experimental data are presently the only reliable information on proteintargeting to subcellular compartments. Stringent parameter setting and manual validation ofMS spectra are required for accurate PTM confirmation. Overall, although the analysis of phos-phorylation by MS is relatively straightforward, comprehensive determination of all PTMs remainsa daunting endeavor.

Concluding RemarksMembrane networks receive and transmit signals that manifest as decisions to modulatephysiological functions in response to environmental changes. The complexity of these signalingnetworks requires the use of high-throughput proteomic approaches to investigate how informa-tion is processed and how protein–interactor relationships are determined. In recent years, MShas evolved from producing mere lists of identified membrane proteins to producing highlysophisticated strategies that are indispensable for the analysis of complex biological systems.These unexpected findings have had a significant effect on the membrane protein field, propellingit in new and exciting directions. As shown here, MS can now be used to comprehensivelyidentify membrane proteins, map membrane protein–protein interactions, and elucidatemembrane signaling networks in considerable depth (see Outstanding Questions). Consider-ing the crucial role played by membrane proteins in cellular function and signaling interactions,membrane proteomics is likely to play a major and indispensable role in the future study ofsignaling modulation.

AcknowledgmentsThis work was supported by grants from the Natural Science Foundation of China (31470345).

References

1. Jones, A.M. et al. (2014) Border control – a membrane-linked

interactome of Arabidopsis. Science 344, 711–7162. Wang, Y. and Wu, W.H. (2013) Potassium transport and

signaling in higher plants. Annu. Rev. Plant Biol. 64,451–476

3. Schwacke, R. and Flugge, U.I. (2018) Identification and char-acterization of plant membrane proteins using ARAMEMNON.Methods Mol. Biol. 1696, 249–259

4. Wang, L. et al. (2015) Spatiotemporal dynamics of the BRI1receptor and its regulation by membrane microdomains inliving Arabidopsis cells. Mol. Plant 8, 1334–1349

5. Li, X. et al. (2019) Characterization of somatic embryogenesisreceptor-like kinase 4 as a negative regulator of leaf senes-cence in Arabidopsis. Cells 8, E50

6. Perraki, A. et al. (2018) Phosphocode-dependent functionaldichotomy of a common co-receptor in plant signalling. Nature561, 248–252

7. Mang, H. et al. (2017) Differential regulation of two-tiered plantimmunity and sexual reproduction by ANXUR receptor-likekinases. Plant Cell 29, 3140–3156

8. Mitra, S.K. et al. (2015) An autophosphorylation site databasefor leucine-rich repeat receptor-like kinases in Arabidopsisthaliana. Plant J. 82, 1042–1060

9. Hohmann, U. et al. (2017) The structural basis of ligand per-ception and signal activation by receptor kinases. Annu. Rev.Plant Biol. 68, 109–137

10. Wang, L. et al. (2018) Exploring the spatiotemporal organiza-tion of membrane proteins in living plant cells. Annu. Rev.Plant Biol. 69, 525–551

11. Glinski, M. and Weckwerth, W. (2006) The role of mass spec-trometry in plant systems biology. Mass Spectrom. Rev. 25,173–214

12. Weckwerth, W. (2011) Green systems biology – from singlegenomes, proteomes and metabolomes to ecosystemsresearch and biotechnology. J. Proteome 75, 284–305

13. Weckwerth, W. et al. (2004) Process for the integrated extrac-tion identification, and quantification of metabolites, proteinsand RNA to reveal their co-regulation in biochemical networks.Proteomics 4, 78–83

14. Wienkoop, S. and Weckwerth, W. (2006) Relative and absolutequantitative shotgun proteomics: targeting low-abundanceproteins in Arabidopsis thaliana. J. Exp. Bot. 57, 1529–1535

15. Nukarinen, E. et al. (2016) Quantitative phosphoproteomicsreveals the role of the AMPK plant ortholog SnRK1 as a meta-bolic master regulator under energy deprivation. Sci. Rep. 6,31697


http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0005http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0005http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0010http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0010http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0010http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0015http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0015http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0015http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0020http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0020http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0020http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0025http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0025http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0025http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0030http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0030http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0030http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0035http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0035http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0035http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0040http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0040http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0040http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0045http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0045http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0045http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0050http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0050http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0050http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0055http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0055http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0055http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0060http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0060http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0060http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0065http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0065http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0065http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0065http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0070http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0070http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0070http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0075http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0075http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0075http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0075


16. Chen, Y. et al. (2010) Comparative analysis of phytohormone-responsive phosphoproteins in Arabidopsis thaliana usingTiO2-phosphopeptide enrichment and mass accuracy precur-sor alignment. Plant J. 63, 1–17

17. Chen, E.I. et al. (2008) Comparisons of mass spectrometrycompatible surfactants for global analysis of the mammalianbrain proteome. Anal. Chem. 80, 8694–8701

18. Moebius, J. et al. (2005) The human platelet membrane prote-ome reveals several new potential membrane proteins. Mol.Cell. Proteomics 4, 1754–1761

19. Washburn, M.P. et al. (2001) Large-scale analysis of the yeastproteome by multidimensional protein identification technology.Nat. Biotechnol. 19, 242–247

20. Abdallah, C. et al. (2014) The membrane proteome ofMedicago truncatula roots displays qualitative and quantitativechanges in response to arbuscular mycorrhizal symbiosis.J. Proteome 108, 354–368

21. Aloui, A. et al. (2018) The plasma membrane proteome ofMedicago truncatula roots as modified by arbuscular mycorrhi-zal symbiosis. Mycorrhiza 28, 1–16

22. Srivastava, V. et al. (2013) Quantitative proteomics reveals thatplasma membrane microdomains from poplar cell suspensioncultures are enriched in markers of signal transduction, molec-ular transport, and callose biosynthesis. Mol. Cell. Proteomics12, 3874–3885

23. Chen, Y. and Hoehenwarter, W. (2015) Changes in thephosphoproteome and metabolome link early signaling eventsto rearrangement of photosynthesis and central metabolism insalinity and oxidative stress response in Arabidopsis. PlantPhysiol. 169, 3021–3033

24. Niittyla, T. et al. (2007) Temporal analysis of sucrose-inducedphosphorylation changes in plasma membrane proteins ofArabidopsis. Mol. Cell. Proteomics 6, 1711–1726

25. Wu, C.C. and Yates III, J.R. (2003) The application of massspectrometry to membrane proteomics. Nat. Biotechnol. 21,262–267

26. Gupta, K. et al. (2017) The role of interfacial lipids in stabilizingmembrane protein oligomers. Nature 541, 421–424

27. Mitra, S.K. et al. (2009) An efficient organic solvent based ex-traction method for the proteomic analysis of Arabidopsisplasma membranes. J. Proteome Res. 8, 2752–2767

28. Wang, P. et al. (2018) Reciprocal regulation of the TOR kinaseand ABA receptor balances plant growth and stress response.Mol. Cell 69, 100–112

29. Wu, X.N. et al. (2019) Sucrose-induced receptor kinase 1 ismodulated by an interacting kinase with short extracellulardomain. Mol. Cell. Proteomics 18, 1556–1571

30. Kar, U.K. et al. (2017) Integral membrane proteins: bottom-up,top-down and structural proteomics. Expert Rev. Proteomics14, 715–723

31. Fischer, F. et al. (2006) Toward the complete membraneproteome: high coverage of integral membrane proteinsthrough transmembrane peptide detection.Mol. Cell. Proteomics5, 444–453

32. Savas, J.N. et al. (2011) Mass spectrometry accelerates mem-brane protein analysis. Trends Biochem. Sci. 36, 388–396

33. Schagger, H. et al. (1994) Analysis of molecular masses andoligomeric states of protein complexes by blue native electro-phoresis and isolation of membrane protein complexes bytwo-dimensional native electrophoresis. Anal. Biochem. 217,220–230

34. Blonder, J. et al. (2006) Identification of membrane proteinsfrom mammalian cell/tissue using methanol-facilitated solubili-zation and tryptic digestion coupled with 2D-LC-MS/MS. Nat.Protoc. 1, 2784–2790

35. Collins, C.A. et al. (2017) Simplified enrichment of plasmamembrane proteins from Arabidopsis thaliana seedlings usingdifferential centrifugation and Brij-58 treatment. Methods Mol.Biol. 1564, 155–168

36. Li, B. et al. (2012) Membrane proteomic analysis comparingsquamous cell lung cancer tissue and tumour-adjacent normaltissue. Cancer Lett. 319, 118–124

37. Wisniewski, J.R. et al. (2009) Combination of FASP and StageTip-based fractionation allows in-depth analysis of the hippocampalmembrane proteome. J. Proteome Res. 8, 5674–5678

38. Griffin, N.M. and Schnitzer, J.E. (2011) Overcoming keytechnological challenges in using mass spectrometry formapping cell surfaces in tissues. Mol. Cell. Proteomics 10R110.000935

39. Mohammed, S. and Heck Jr., A. (2010) Strong cationexchange (SCX) based analytical methods for the targetedanalysis of protein post-translational modifications. Curr.Opin. Biotechnol. 22, 9–16

40. Wienkoop, S. et al. (2004) Linking protein fractionation withmultidimensional monolithic reversed-phase peptide chroma-tography/mass spectrometry enhances protein identificationfrom complex mixtures even in the presence of abundantproteins. Rapid Commun. Mass Spectrom. 18, 643–650

41. Nelson, C.J. et al. (2006) A quantitative analysis of Arabidopsisplasma membrane using trypsin-catalyzed O-18 labeling. Mol.Cell. Proteomics 5, 1382–1395

42. Howard, J.W. et al. (2012) UHPLC for the separation ofproteins and peptides. Bioanalysis 4, 2971–2988

43. Speers, A.E. et al. (2007) Shotgun analysis of integral mem-brane proteins facilitated by elevated temperature. Anal.Chem. 79, 4613–4620

44. Min, L. et al. (2015) Improved protease digestion conditions formembrane protein detection. Electrophoresis 36, 1690–1698

45. Zhang, X. (2015) Less is more: membrane protein digestionbeyond urea-trypsin solution for next-level proteomics. Mol.Cell. Proteomics 14, 2441–2453

46. Schmollinger, S. et al. (2014) Nitrogen-sparing mechanisms inChlamydomonas affect the transcriptome, the proteome, andphotosynthetic metabolism. Plant Cell 26, 1410–1435

47. Tyanova, S. et al. (2016) The MaxQuant computational plat-form for mass spectrometry-based shotgun proteomics. Nat.Protoc. 11, 2301–2319

48. Takahashi, D. et al. (2014) Shotgun proteomics of plant plasmamembrane and microdomain proteins using nano-LC-MS/MS.Methods Mol. Biol. 1072, 481–498

49. Lefebvre, B. et al. (2007) Characterization of lipid rafts fromMedicago truncatula root plasma membranes: a proteomicstudy reveals the presence of a raft-associated redox system.Plant Physiol. 144, 402–418

50. Peskan, T. et al. (2000) Identification of low-density Triton X-100-insoluble plasma membrane microdomains in higherplants. Eur. J. Biochem. 267, 6989–6995

51. Kierszniowska, S. et al. (2009) Definition of Arabidopsis sterol-rich membrane microdomains by differential treatment withmethyl-beta-cyclodextrin and quantitative proteomics. Mol.Cell. Proteomics 8, 612–623

52. Nakamura, M. and Grebe, M. (2018) Outer, inner and planarpolarity in the Arabidopsis root. Curr. Opin. Plant Biol. 41,46–53

53. Gutierrez-Carbonell, E. et al. (2016) A shotgun proteomic ap-proach reveals that fe deficiency causes marked changes inthe protein profiles of plasma membrane and detergent-resistant microdomain preparations from Beta vulgaris roots.J. Proteome Res. 15, 2510–2524

54. Tapken, W. and Murphy, A.S. (2015) Membrane nanodomainsin plants: capturing form, function, and movement. J. Exp. Bot.66, 1573–1586

55. Keinath, N.F. et al. (2010) PAMP (pathogen-associated molec-ular pattern)-induced changes in plasma membrane compart-mentalization reveal novel components of plant immunity.J. Biol. Chem. 285, 39140–39149

56. Stanislas, T. et al. (2009) Quantitative proteomics reveals adynamic association of proteins to detergent-resistant mem-branes upon elicitor signaling in tobacco.Mol. Cell. Proteomics8, 2186–2198

57. Fujiwara, M. et al. (2009) Proteome analysis of detergent-resistant membranes (DRMs) associated with OsRac1-mediated innate immunity in rice. Plant Cell Physiol. 50,1191–1200

58. Hao, H. et al. (2014) Clathrin and membrane microdomainscooperatively regulate RbohD dynamics and activity inArabidopsis. Plant Cell 26, 1729–1745

59. Lv, X. et al. (2017) Membrane microdomains and the cytoskel-eton constrain AtHIR1 dynamics and facilitate the formation ofan AtHIR1-associated immune complex. Plant J. 90, 3–16


http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0080http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0080http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0080http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0080http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0080http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0085http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0085http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0085http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0090http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0090http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0090http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0095http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0095http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0095http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0100http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0100http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0100http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0100http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0105http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0105http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0105http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0110http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0110http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0110http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0110http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0110http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0115http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0115http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0115http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0115http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0115http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0120http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0120http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0120http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0125http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0125http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0125http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0130http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0130http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0135http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0135http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0135http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0140http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0140http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0140http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0145http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0145http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0145http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0150http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0150http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0150http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0155http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0155http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0155http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0155http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0160http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0160http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0165http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0165http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0165http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0165http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0165http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0170http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0170http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0170http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0170http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0175http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0175http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0175http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0175http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0180http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0180http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0180http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0185http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0185http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0185http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0190http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0190http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0190http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0190http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0195http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0195http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0195http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0195http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0200http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0200http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0200http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0200http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0200http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0205http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0205http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0205http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0210http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0210http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0215http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0215http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0215http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0220http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0220http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0225http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0225http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0225http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0230http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0230http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0230http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0235http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0235http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0235http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0240http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0240http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0240http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0245http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0245http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0245http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0245http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0250http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0250http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0250http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0255http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0255http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0255http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0255http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0260http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0260http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0260http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0265http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0265http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0265http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0265http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0265http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0270http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0270http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0270http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0275http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0275http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0275http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0275http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0280http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0280http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0280http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0280http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0285http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0285http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0285http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0285http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0290http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0290http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0290http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0295http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0295http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0295


60. Haruta, M. et al. (2014) A peptide hormone and its receptorprotein kinase regulate plant cell expansion. Science 343,408–411

61. Bellati, J. et al. (2016) Novel aquaporin regulatory mechanismsrevealed by interactomics. Mol. Cell. Proteomics 15,3473–3487

62. Dedecker, M. et al. (2015) Unravelling plant molecularmachineries through affinity purification coupled to massspectrometry. Curr. Opin. Plant Biol. 24, 1–9

63. Van Leene, J. et al. (2007) A tandem affinity purification-based technology platform to study the cell cycle interac-tome in Arabidopsis thaliana. Mol. Cell. Proteomics 6,1226–1238

64. Van Leene, J. et al. (2008) Boosting tandem affinity purificationof plant protein complexes. Trends Plant Sci. 13, 517–520

65. Burckstummer, T. et al. (2006) An efficient tandem affinity puri-fication procedure for interaction proteomics in mammaliancells. Nat. Methods 3, 1013–1019

66. Pauwels, L. et al. (2010) NINJA connects the co-repressorTOPLESS to jasmonate signalling. Nature 464, 788–791

67. Babu, M. et al. (2012) Interaction landscape of membrane-protein complexes in Saccharomyces cerevisiae. Nature 489,585–589

68. Van Leene, J. et al. (2010) Targeted interactomics reveals acomplex core cell cycle machinery in Arabidopsis thaliana.Mol. Syst. Biol. 6, 397

69. Rohila, J.S. et al. (2009) Protein–protein interactions of tandemaffinity purified protein kinases from rice. PLoS One 4, e6685

70. Van Leene, J. et al. (2019) Capturing the phosphorylation andprotein interaction landscape of the plant TOR kinase. NaturePlants 5, 316–327

71. Huang, H. et al. (2015) Identification of evening complex asso-ciated proteins in Arabidopsis by affinity purification and massspectrometry. Mol. Cell. Proteomics 15, 201–217

72. Gilbert, M. and Schulze, W.X. (2019) Global identification ofprotein complexes within the membrane proteome ofArabidopsis roots Using a SEC-MS approach. J. ProteomeRes. 18, 107–119

73. Roux, M. et al. (2011) The Arabidopsis leucine-rich repeatreceptor-like kinases BAK1/SERK3 and BKK1/SERK4 are re-quired for innate immunity to hemibiotrophic and biotrophicpathogens. Plant Cell 23, 2440–2455

74. Hein, M.Y. et al. (2015) A human interactome in three quantita-tive dimensions organized by stoichiometries and abundances.Cell 163, 712–723

75. Cox, J. and Mann, M. (2011) Quantitative, high-resolution pro-teomics for data-driven systems biology. Annu. Rev. Biochem.80, 273–299

76. Kim, D.Y. et al. (2013) Advanced proteomic analyses yield adeep catalog of ubiquitylation targets in Arabidopsis. PlantCell 25, 1523–1540

77. Shannon, P. et al. (2003) Cytoscape: a software environmentfor integrated models of biomolecular interaction networks.Genome Res. 13, 2498–2504

78. Wu, X. et al. (2014) A kinase-phosphatase signaling modulewith BSK8 and BSL2 involved in regulation of sucrose-phosphate synthase. J. Proteome Res. 13, 3397–3409

79. Wang, K. et al. (2018) Quantitative and functional posttransla-tional modification proteomics reveals that TREPH1 plays arole in plant touch-delayed bolting. Proc. Natl. Acad. Sci.U. S. A. 115, E10265–E10274

80. Chen, Y. and Hoehenwarter, W. (2019) Rapid and reproduciblephosphopeptide enrichment by tandem metal oxide affinitychromatography: application to boron deficiency inducedphosphoproteomics. Plant J. 98, 370–384

81. Hoehenwarter, W. et al. (2013) Identification of novel in vivoMAP kinase substrates in Arabidopsis thaliana through use oftandem metal oxide affinity chromatography. Mol. Cell.Proteomics 12, 369–380

82. Wolschin, F. and Weckwerth, W. (2005) Combining metaloxide affinity chromatography (MOAC) and selective massspectrometry for robust identification of in vivo protein phos-phorylation sites. Plant Methods 1, 9

83. Wolschin, F. et al. (2005) Enrichment of phosphorylated pro-teins and peptides from complex mixtures using metal oxide/

hydroxide affinity chromatography (MOAC). Proteomics 5,4389–4397

84. Larsen, M.R. et al. (2005) Highly selective enrichment of phos-phorylated peptides from peptide mixtures using titanium diox-ide microcolumns. Mol. Cell. Proteomics 4, 873–886

85. Beckers, G.J. et al. (2014) Tandem metal-oxide affinitychromatography for enhanced depth of phosphoproteomeanalysis. Methods Mol. Biol. 1072, 621–632

86. Hsu, C.C. et al. (2018) Universal plant phosphoproteomicsworkflow and its application to tomato signaling in responseto cold stress. Mol. Cell. Proteomics 17, 2068–2080

87. Wu, X.N. et al. (2017) Highly efficient single-step enrichment oflow abundance phosphopeptides from plant membrane prep-arations. Front. Plant Sci. 8, 1673

88. Elias, J.E. and Gygi, S.P. (2007) Target-decoy search strategyfor increased confidence in large-scale protein identificationsby mass spectrometry. Nat. Methods 4, 207–214

89. Sinitcyn, P. et al. (2018) MaxQuant goes Linux. Nat. Methods15, 401

90. Engelsberger, W.R. and Schulze, W.X. (2011) Nitrate and am-monium lead to distinct global dynamic phosphorylation pat-terns when resupplied to nitrogen-starved Arabidopsisseedlings. Plant J. 69, 978–995

91. Hoehenwarter, W. et al. (2008) A rapid approach forphenotype-screening and database independent detection ofcSNP/protein polymorphism using mass accuracy precursoralignment. Proteomics 8, 4214–4225

92. Yang, Z. et al. (2013) Stable isotope metabolic labeling-basedquantitative phosphoproteomic analysis of Arabidopsis mu-tants reveals ethylene-regulated time-dependent phosphopro-teins and putative substrates of constitutive triple response 1kinase. Mol. Cell. Proteomics 12, 3559–3582

93. Qing, D. et al. (2015) Quantitative and functionalphosphoproteomic analysis reveals that ethylene regulateswater transport via the C-terminal phosphorylation of aquapo-rin PIP2;1 in Arabidopsis. Mol. Plant 9, 158–174

94. Wu, X.N. et al. (2013) Sucrose-induced receptor kinase SIRK1regulates a plasma membrane aquaporin in Arabidopsis. Mol.Cell. Proteomics 12, 2856–2873

95. Lehmann, U. et al. (2008) If the antibody fails – a mass westernapproach. Plant J. 55, 1039–1046

96. Roustan, V. et al. (2017) Quantitative in vivophosphoproteomics reveals reversible signaling processesduring nitrogen starvation and recovery in the biofuel model or-ganism Chlamydomonas reinhardtii. Biotechnol. Biofuels 10,280

97. Roustan, V. and Weckwerth, W. (2018) Quantitativephosphoproteomic and system-level analysis of TOR inhibitionunravel distinct organellar acclimation in Chlamydomonasreinhardtii. Front. Plant Sci. 9, 1590

98. Bhaskara, G.B. et al. (2019) The flip side of phospho-signalling:regulation of protein dephosphorylation and the protein phos-phatase 2Cs. Plant Cell Environ. 42, 2913–2930

99. Wong, M.M. et al. (2019) Phosphoproteomics of Arabidopsishighly ABA-induced1 identifies AT-Hook-Like10 phosphoryla-tion required for stress growth regulation. Proc. Natl. Acad.Sci. U. S. A. 116, 2354–2363

100. Kim, T.W. et al. (2009) Brassinosteroid signal transduction fromcell-surface receptor kinases to nuclear transcription factors.Nat. Cell Biol. 11, 1254–1260

101. Wang, P. et al. (2013) Quantitative phosphoproteomics iden-tifies SnRK2 protein kinase substrates and reveals the effectorsof abscisic acid action. Proc. Natl. Acad. Sci. U. S. A. 110,11205–11210

102. Umezawa, T. et al. (2013) Genetics and phosphoproteomicsreveal a protein phosphorylation network in the abscisicacid signaling pathway in Arabidopsis thaliana. Sci. Signal 6,rs8

103. Werth, E.G. et al. (2019) Investigating the effect of target ofrapamycin kinase inhibition on the Chlamydomonas reinhardtiiphosphoproteome: from known homologs to new targets.New Phytol. 221, 247–260

104. Ruiz-May, E. et al. (2014) A comparative study of lectin affinitybased plant N-glycoproteome profiling using tomato fruit as amodel. Mol. Cell. Proteomics 13, 566–579


http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0300http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0300http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0300http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0305http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0305http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0305http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0310http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0310http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0310http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0315http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0315http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0315http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0315http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0320http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0320http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0325http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0325http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0325http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0330http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0330http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0335http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0335http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0335http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0340http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0340http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0340http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0345http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0345http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0350http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0350http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0350http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0355http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0355http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0355http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0360http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0360http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0360http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0360http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0365http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0365http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0365http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0365http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0370http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0370http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0370http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0375http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0375http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0375http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0380http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0380http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0380http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0385http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0385http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0385http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0390http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0390http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0390http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0395http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0395http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0395http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0395http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0400http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0400http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0400http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0400http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0405http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0405http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0405http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0405http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0410http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0410http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0410http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0410http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0415http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0415http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0415http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0415http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0420http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0420http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0420http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0425http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0425http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0425http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0430http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0430http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0430http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0435http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0435http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0435http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0440http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0440http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0440http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0445http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0445http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0450http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0450http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0450http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0450http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0455http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0455http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0455http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0455http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0460http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0460http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0460http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0460http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0460http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0465http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0465http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0465http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0465http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0470http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0470http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0470http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0475http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0475http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0480http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0480http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0480http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0480http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0480http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0485http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0485http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0485http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0485http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0490http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0490http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0490http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0495http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0495http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0495http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0495http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0500http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0500http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0500http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0505http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0505http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0505http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0505http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0510http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0510http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0510http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0510http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0515http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0515http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0515http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0515http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0520http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0520http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0520


105. Schwacke, R. et al. (2003) ARAMEMNON, a novel database forArabidopsis integral membrane proteins. Plant Physiol. 131, 16–26

106. Durek, P. et al. (2010) PhosPhAt: the Arabidopsis thalianaphosphorylation site database. An update. Nucleic Acids Res.38, D828–D834

107. Alli-Balogun, G.O. and Levine, T.P. (2019) Regulation oftargeting determinants in interorganelle communication. Curr.Opin. Cell Biol. 57, 106–114

108. Guo, X. et al. (2017) Cold signal shuttles from membrane tonucleus. Mol. Cell 66, 7–8

109. Ding, Y. et al. (2015) OST1 kinase modulates freezing toleranceby enhancing ICE1 stability in Arabidopsis. Dev. Cell 32,278–289

110. Pajonk, S. et al. (2008) Activity determinants and functionalspecialization of Arabidopsis PEN1 syntaxin in innate immunity.J. Biol. Chem. 283, 26974–26984


http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0525http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0525http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0530http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0530http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0530http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0535http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0535http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0535http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0540http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0540http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0545http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0545http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0545http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0550http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0550http://refhub.elsevier.com/S1360-1385(20)30113-8/rf0550

Mass Spectrometry Untangles Plant Membrane Protein Signaling NetworksSimple Beginning: The Significance of PM ProteinsChallenges in the Preparation of Membrane ProteinsMS to Determine the Complete Membrane ProteomeMS-based Proteomics to Analyze Membrane MicrodomainsMS-based Methods to Determine Membrane Protein Interaction NetworksProteomics to Understand Protein Phosphorylation Signaling Networks Across Plasma MembranesTechniques to Improve Phosphoprotein IdentificationQuantitative PhosphoproteomicsFunctional Phosphoproteomics

Other PTMsConcluding RemarksAcknowledgmentsReferences

mass spectrometry untangles plant membrane protein...

Documents