II. Production de données omiques et méthodologies d’analyses
II.5 L’analyse du protéome pariétal des rosettes d’A thaliana
Le protéome pariétal de rosette d’A. thaliana est peu connu avec seulement 148 protéines pariétales identifiées sur les 700 prédites comme étant sécrétées (WallProtDB,
http://www.polebio.scsv.ups-tlse.fr/WallProtDB/). Les études précédemment réalisées sur cette
organe ont été réalisées avec une méthode non destructive d’infiltration sous vide (Boudart et
al. 2005; Haslam et al. 2003; Trentin et al. 2015).
A l’occasion de la mise en œuvre de notre projet, deux méthodes d’analyse des protéomes ont été utilisées :
i) une méthodologie destructive d’extraction des protéines pariétales suivie d’une séparation par une electrophorèse-1D courte. Chaque piste a ensuite été découpée en trois bandes pour être analysée en spectrométrie de masse (MS). Cette séparation permettait de diviser par trois le nombre de protéines analysées à chaque passage en spectrométrie de masse, augmentant ainsi la couverture et la finesse de l’analyse ;
ii) une méthodologie reposant sur la même méthode destructive d’extraction de protéines pariétales, mais cette fois suivie d’une stratégie de type shotgun pour l’analyse en MS. Cette technique a notamment permis de réduire considérablement le temps d’analyse.
Il était important de comparer les différentes méthodologies et techniques d’analyse avant de poursuivre par une exploitation quantitative. Cette étude comparative est d’ailleurs discutée dans une publication parue dans la revue Proteomics, reprise intégralement ci-après.
Proteomics 2016, 16, 3183–3187 DOI 10.1002/pmic.201600290 3183
DATASETBRIEF
An enlarged cell wall proteome of Arabidopsis thaliana
rosettes
Vincent Herv ´e1,2∗, Harold Durufl ´e1∗, H ´el `ene San Clemente1, C ´ecile Albenne1, Thierry Balliau3,4, Michel Zivy3,4, Christophe Dunand1 and Elisabeth Jamet1
1Laboratoire de Recherche en Sciences V ´eg ´etales, Universit ´e de Toulouse, CNRS, UPS, Castanet Tolosan, France 2INRS—Institut Armand Frappier, Laval, Canada
3CNRS, PAPPSO, UMR 0320/UMR 8120 G ´en ´etique V ´eg ´etale, Gif sur Yvette, France 4INRA, PAPPSO, UMR 0320/UMR 8120 G ´en ´etique V ´eg ´etale, Gif sur Yvette, France
Received: July 7, 2016 Revised: September 28, 2016 Accepted: October 21, 2016
Plant cells are surrounded by cell walls playing many roles during development and in response to environmental constraints. Cell walls are mainly composed of polysaccharides (cellulose, hemicelluloses and pectins), but they also contain proteins which are critical players in cell wall remodeling processes. Today, the cell wall proteome of Arabidopsis thaliana, a major dicot model plant, comprises more than 700 proteins predicted to be secreted (cell wall proteins—CWPs) identified in different organs or in cell suspension cultures. However, the cell wall proteome of rosettes is poorly represented with only 148 CWPs identified after extraction by vacuum in- filtration. This new study allows enlarging its coverage. A destructive method starting with the purification of cell walls has been performed and two experiments have been compared. They differ by the presence/absence of protein separation by a short 1D-electrophoresis run prior to tryptic digestion and different gradient programs for peptide separation before mass spectrom- etry analysis. Altogether, the rosette cell wall proteome has been significantly enlarged to 361 CWPs, among which 213 newly identified in rosettes and 57 newly described. The identified CWPs fall in four major functional classes: 26.1% proteins acting on polysaccharides, 11.1% oxido-reductases, 14.7% proteases and 11.7% proteins possibly related to lipid metabolism.
Keywords:
Arabidopsis thaliana / Cell wall / Leaf / Plant proteomics / Rosette
Additional supporting information may be found in the online version of this article at the publisher’s web-sitePlant cells are surrounded by cell walls playing many roles during development and in response to environmental con- straints. Cell walls are mainly composed of complex polysac- charidic networks of cellulose, hemicelluloses and pectins [1]. They contain many proteins which diversity has been re- vealed by dedicated proteomics studies [2–4]. These proteins also called cell wall proteins (CWPs) are critical players in cell wall remodeling processes and in signaling [5, 6]. Our
Correspondence: Dr. E. Jamet, UMR 5546 UPS/CNRS, Laboratoire
de Recherche en Sciences V ´eg ´etales, BP 42617, F-31326 Castanet- Tolosan, France
E-mail: jamet@lrsv.ups-tlse.fr
Fax:+33(0)534 32 38 02
Abbreviation: CWP, cell wall protein
knowledge of plant cell wall proteomes is quickly enlarging since the beginning of the development of mass spectrome- try (MS) technologies applied to protein analysis in the early 2000’s. Today, the cell wall proteome of Arabidopsis thaliana, a major dicot model plant, is the best described. It comprises more than 700 proteins predicted to be secreted which have been identified in different organs or in cell suspension cul- tures, which is between one-third and one-half of its expected size [7]. However, the cell wall proteome of rosettes is poorly represented with only 148 CWPs identified after extraction of proteins from the rosette apoplast by vacuum infiltration [8–10], described as a non-destructive method [3]. This new study aims at enlarging the coverage of this proteome using
∗Co-first authors
Colour Online: See the article online to view Figs. 1 and 2 in colour.
C
3184 V. Herv ´e et al. Proteomics 2016, 16, 3183–3187
Figure 1. 1DE patterns of A. thaliana proteins extracted from pu-
rified rosette cell walls by CaCl2 0.2 M and LiCl 2 M solutions.
Fortyg of proteins of each sample (1, 2, 3: biological repeats 1,
2 and 3 of experiment 1, respectively) have been separated by regular 1DE and stained with Coomassie blue. Molecular mass (MM) markers are indicated in kDa.
a destructive method for protein extraction starting with the purification of cell walls [11, 12]. In addition, two different techniques of protein preparation prior to tryptic digestion and MS analysis have been compared. They differ at two levels: (i) one protein extract has been separated by a short 1D-electrophoresis (1DE) run (experiment 1), whereas the second has been subjected to shotgun analysis (experiment 2) and (ii) the separation of tryptic peptides by liquid chro- matography (LC) has been performed with a shorter column (150 mm vs. 300 mm) and a gradient program steeper in experiment 1 than in experiment 2 (Supporting Information Fig. S1).
A. thaliana plants have been cultivated in growth chambers
at 22⬚C with a photoperiod of 16 h light/8 h dark. Rosettes have been collected after 4 weeks. Twenty plants have been used per biological replicate and three biological replicates have been performed for each experiment. Briefly, cell walls have been purified as described [12]. Proteins have been ex- tracted from lyophilized cell walls in four steps using a 5 mM acetate buffer pH 4.6 complemented with 0.2 M CaCl2
(two successive extractions) or 2 M LiCl (two successive ex- tractions) [11]. The four salt extracts were pooled to get the protein samples to be analyzed. Typically, 35 mg dry cell walls/g fresh material and 0.57 mg proteins/g fresh material were obtained. The quality of the protein extracts has been checked by regular 1DE. The presence of thin well-separated bands after Coomassie blue staining showed the absence of protein degradation. In addition, all three biological repeats for each experiment looked similar (see Fig. 1 showing the samples of experiment 1).
Experiment 1 included a short separation (5 mm run) of proteins by 1DE, followed by a cutting of the gel in three pieces prior to tryptic digestion of proteins in gelo [13]. Experiment 2 was a shotgun protein analysis directly after their extraction from purified cell walls. Proteins were digested in solution by trypsin as described [14]. For the two experiments, LC-MS/MS analyses were performed with a Q-exactive mass spectrometer (Thermo Fisher Scientific, Villebon-sur-Yvette) as described in Supporting Information Fig. S1. The X!Tandem software was used for protein identification [15] and the X!Tandem pipeline for MS data processing http://www.thegpm.org). The detailed procedure for LC-MS/MS analysis and the parameters used for peptide and protein identification are detailed in Supporting Informa- tion Tables S1–S2 (sheets “parameters”). The false discovery rates (FDRs) used for peptides and proteins are the following ones: 0.079% and 0%, respectively, for experiment 1; 0.076% and 0%, respectively, for experiment 2. All the MS/MS data have been made publicly available in the PROTICdb database (/http://moulon.inra.fr/protic/cell_wall_athaliana_rosettes) [16]. Those regarding the CWPs can be found in
WallProtDB as well as their functional annotations
(http://www.polebio.lrsv.ups-tlse.fr/WallProtDB) [17]. The identification of a given protein has been validated in one experiment when at least two proteotypic peptides of this protein were present in at least two biological replicates. Proteins have been annotated using the ProtAnnDB pipeline to predict sub-cellular localization and functional domains (http://www.polebio.lrsv.ups-tlse.fr/ProtAnnDB) [18]. Only the proteins having a predicted signal peptide, no endo- plasmic reticulum retention signal and no more than one trans-membrane domain have been retained as CWPs. These predictions are done with several bioinformatics program and literature data are used when available to refine this analysis.
Altogether, 1093 proteins have been identified among which 328 have been predicted as CWPs. Experiments 1 (1D- E) and 2 (shotgun) have allowed the identification of 286 and 250 CWPs, respectively (Fig. 2A and Supporting Information Tables S1–S3). Even if a great proportion of the identified CWPs are in common between the two experiments, 78 have been specifically identified in experiment 1 and 42 in experi- ment 2. These features may be due to (i) differences in tryptic digestion efficiency, (ii) the splitting of the protein extract in three parts in experiment 1 and/or (iii) biological variability between the two experiments. It seems that CWPs specific to Experiment 1 or 2 were identified with less proteotypic pep- tides than those common to the two experiments, suggesting that they are less abundant proteins (see Supporting Infor- mation Fig. S2). Finally, the shotgun experiment was more convenient to handle, but it has led to the identification of less proteins.
This study has led to a significant enlargement of the rosette cell wall proteome to 361 CWPs, with 57 CWPs newly identified, 213 newly identified in rosettes, and 93 specific to this proteome (Table 1). One-hundred-and-fourteen CWPs
C
Proteomics 2016, 16, 3183–3187 3185
Figure 2. Characteristics of the new rosette cell wall proteome
of A. thaliana. (A) Number of CWPs identified in experiment 1 (1DE) vs. number of CWPs identified in experiment 2 (shotgun). (B) Comparison of the number of CWPs identified in this study (destructive method) and in previous studies (non-destructive method) [8–10]. (C) Distribution (%) of CWPs identified in the overall rosette cell wall proteome (non-destructive and destruc- tive methods) in functional classes according to bioinformatic prediction of functional domains: PAC (proteins acting on carbo- hydrates), OR (oxido-reductases), P (proteases), LM (proteins pos- sibly related to lipid metabolism), ID (proteins with interactions domains with carbohydrates or proteins), S (proteins possibly in- volved in signaling), SP (structural proteins), M (miscellaneous proteins), UF (proteins of yet unknown function).
Table 1. Number of proteins specific to the enlarged rosette
cell wall proteome and common to other cell wall pro- teomes: roots [21–23], stems [24], etiolated hypocotyls [11, 12, 25], cell suspensions culture and culture medium [26–31]
CWPs Number
Newly identified 57
Newly identified in rosettes 213
Specific to rosettes 93
Common to roots 219
Common to stems 64
Common to etiolated hypocotyls 135
Common to cell suspension cultures/liquid medium
109
Data are from WallProtDB (http://www.polebio.lrsv.ups-tlse.fr/ WallProtDB/index.php).
were found to be common between the new rosette cell wall proteome (destructive method) and the previously described ones (non-destructive methods) and 34 specific to the previ- ous ones (Fig. 2B). These differences could be related to the fact that CWPs interact in various ways with cell wall compo- nents. The non-destructive method usually allows recovering proteins with weak interactions with cell wall components. Such proteins may be lost during the cell wall purification pro- cedure used in the destructive method [2]. These two methods are complementary and have allowed increasing the coverage of the rosette cell wall proteome.
The distribution of CWPs in functional classes has revealed four major groups: 26.1% proteins acting on polysaccharides, 11.1% oxido-reductases, 14.7% proteases and 11.7% proteins possibly related to lipid metabolism (Fig. 2C). This distribu- tion differs slightly from that of Solanum tuberosum mature leaves with less proteins acting on polysaccharides (33.6% in S. tuberosum), less proteases (19.8%) and more proteins possibly related to lipid metabolism (5.3%) [19]. It is also different from that of Brachypodium distachyon leaves with less oxido-reductases (13.9% in B. distachyon), less proteases (18.3%) and more proteins with interactions domains with polysaccharides or proteins (4.8%) [20].
The comparison of the rosette cell wall proteome to those of other A. thaliana organs has shown that 219 CWPs were common with that of roots [21–23], 64 to that of stems [24] and 135 to that of etiolated hypocotyls [11,12,25]. Besides, 109 CWPs were common to the cell wall proteomes of rosettes and cell suspension cultures [26–31] (Table 1). Altogether, the cell wall proteome of A. thaliana remains the best de- scribed with 766 CWPs indexed in WallProtDB. Each new set of data allows completing the CWP atlas and identifying CWPs specific to a given organ or present as housekeepers. This rosette cell wall proteome has been obtained from 4- week-old plants. It would be interesting to look for changes in its composition during plant development or in response to environmental cues. Such an approach has been undertaken in B. distachyon and has revealed changes between the protein profile of young and mature leaves [20]. Finally, the shotgun strategy appears as an interesting tool to get in a few steps a cell wall proteome from small amounts of material. It paves the way for the development of micro-proteomics in order to identify tissue-specific CWPs and associate these results to micro-transcriptomics or -metabolomics data for example.
The authors are thankful to Universit´e Paul Sabatier (Toulouse, France) and CNRS for supporting their research. They also wish to thank Olivier Langella for introducing the LC-MS/MS data in the PROTICdb public repository. HD is supported by a grant of the Midi-Pyr´en´ees region and of the Fed- eral University of Toulouse. LC-MS/MS analyses were performed at the Plateforme d’Analyse Prot´eomique de Paris Sud-Ouest (PAPPSO).
The authors have declared no conflict of interest.
C
3186 V. Herv ´e et al. Proteomics 2016, 16, 3183–3187
References
[1] Carpita, N. C., Gibeaut, D. M., Structural models of primary cell walls in flowering plants, consistency of molecular struc- ture with the physical properties of the walls during growth.
Plant J. 1993, 3, 1–30.
[2] Albenne, C., Canut, H., Hoffmann, L., Jamet, E., Plant cell wall proteins: a large body of data, but what about runaways?
Proteomes 2014, 2, 224–242.
[3] Albenne, C., Canut, H., Jamet, E., Plant cell wall proteomics: the leadership of Arabidopsis thaliana. Front. Plant Sci. 2013,
4, 111.
[4] Rose, J., Bashir, S., Giovannoni, J., Jahn, M., Saravanan, R., Tackling the plant proteome: practical approaches, hurdles and experimental tools. Plant J. 2004, 39, 715–733. [5] Frankov ´a, L., Fry, S. C., Biochemistry and physiological roles
of enzymes that "cut and paste" plant cell-wall polysaccha- rides. J. Exp. Bot. 2013, 64, 3519–3550.
[6] Van der Hoorn, R., Plant proteases: from phenotypes to molecular mechanisms. Annu. Rev. Plant Biol. 2008, 59, 191– 223.
[7] Jamet, E., Canut, H., Boudart, G., Pont-Lezica, R., Cell wall proteins: a new insight through proteomics. Trends Plant
Sci. 2006, 11, 33–39.
[8] Boudart, G., Jamet, E., Rossignol, M., Lafitte, C. et al., Cell wall proteins in apoplastic fluids of Arabidopsis thaliana rosettes: identification by mass spectrometry and bioinfor- matics. Proteomics 2005, 5, 212–221.
[9] Haslam, R. P., Downie, A. L., Raventon, M., Gallardo, K. et al., The assessment of enriched apoplastic extracts us- ing proteomic approaches. Ann. Appl. Biol. 2003, 143, 81–91.
[10] Trentin, A., Pivato, M., Mehdi, S., Barnabas, L. et al., Pro- teome readjustments in the apoplastic space of Arabidop-
sis thaliana ggt1 mutant leaves exposed to UV-B radiation. Front. Plant Sci. 2015, 6, 128.
[11] Irshad, M., Canut, H., Borderies, G., Pont-Lezica, R., Jamet, E., A new picture of cell wall protein dynamics in elongating cells of Arabidopsis thaliana: confirmed actors and newcom- ers. BMC Plant Biol. 2008, 8, 94.
[12] Feiz, L., Irshad, M., Pont-Lezica, R. F., Canut, H., Jamet, E., Evaluation of cell wall preparations for proteomics: a new procedure for purifying cell walls from Arabidopsis hypocotyls. Plant Methods 2006, 2, 10.
[13] Ligat, L., Lauber, E., Albenne, C., San Clemente, H. et al., Analysis of the xylem sap proteome of Brassica oleracea reveals a high content in secreted proteins. Proteomics 2011,
11, 1798–1813.
[14] Blein-Nicolas, M., Albertin, W., da Silva, T., Valot, B. et al., A systems approach to elucidate heterosis of protein abun- dances in yeast. Mol. Cell. Proteomics 2015, 14, 2056– 2071.
[15] Craig, R., Beavis, R., TANDEM: matching proteins with tan- dem mass spectra. Bioinformatics 2004, 20, 1466–1467. [16] Langella, O., Valot, B., Jacob, D., Balliau, T. et al., Man-
agement and dissemination of MS proteomic data with
PROTICdb: example of a quantitative comparison between methods of protein extraction. Proteomics 2013, 13, 1457– 1466.
[17] San Clemente, H., Jamet, E., WallProtDB, a database re- source for plant cell wall proteomics. Plant Methods 2015,
11, 2.
[18] San Clemente, H., Pont-Lezica, R., Jamet, E., Bioinformatics as a tool for assessing the quality of sub-cellular proteomic strategies and inferring functions of proteins: plant cell wall proteomics as a test case. Bioinform. Biol. Insights 2009, 3, 15–28.
[19] Lim, S., Chisholm, K., Coffin, R. H., Peters, R. D. et al., Protein profiling in potato (Solanum tuberosum L.) leaf tissues by differential centrifugation. J. Proteome Res. 2012, 11, 2594– 2601.
[20] Douch ´e, T., San Clemente, H., Burlat, V., Roujol, D. et al.,
Brachypodium distachyon as a model plant toward improved
biofuel crops: search for secreted proteins involved in bio- genesis and disassembly of cell wall polymers. Proteomics 2013, 13, 2438–2454.
[21] Basu, U., Francis, J. L., Whittal, R. W., Stephens, J. L. et al., Extracellular proteomes of Arabidopsis thaliana and Bras-
sica napus roots: analysis and comparison by MUdPIT and
LC-MS/MS. Plant Soil 2006, 286, 357–376.
[22] Chen, Y., Ye, D., Held, M. A., Cannon, M. C. et al., Identifi- cation of the abundant hydroxyproline-rich glycoproteins in the root walls of wild-type Arabidopsis, an ext3 mutant line, and its phenotypic revertant. Plants 2015, 4, 85–111. [23] Nguyen-Kim, H., San Clemente, H., Balliau, T., Zivy, M. et al.,
Arabidopsis thaliana root cell wall proteomics: increasing
the proteome coverage using a combinatorial peptide lig- and library and description of unexpected Hyp in perox- idase amino acid sequences. Proteomics 2016, 16, 491– 503.
[24] Minic, Z., Jamet, E., Negroni, L., der Garabedian, P. A. et al., A sub-proteome of Arabidopsis thaliana trapped on Con- canavalin A is enriched in cell wall glycoside hydrolases. J.
Exp. Bot. 2007, 58, 2503–2512.
[25] Zhang, Y., Giboulot, A., Zivy, M., Valot, B. et al., Combin- ing various strategies to increase the coverage of the plant cell wall glycoproteome. Phytochemistry 2011, 72, 1109– 1123.
[26] Bayer, E. M., Bottrill, A. R., Walshaw, J., Vigouroux, M. et al., Arabidopsis cell wall proteome defined using multidimen- sional protein identification technology. Proteomics 2006, 6, 301–311.
[27] Borderies, G., Jamet, E., Lafitte, C., Rossignol, M. et al., Pro- teomics of loosely bound cell wall proteins of Arabidopsis
thaliana cell suspension cultures: a critical analysis. Elec- trophoresis 2003, 24, 3421–3432.
[28] Chivasa, S., Ndimba, B. K., Simon, W. J., Robertson, D. et al., Proteomic analysis of the Arabidopsis thaliana cell wall. Elec-
trophoresis 2002, 23, 1754–1765.
[29] Kwon, H.-K., Yokoyama, R., Nishitani, K., A proteomic ap- proach to apoplastic proteins involved in cell wall regenera- tion in protoplasts of Arabidopsis suspension-cultured cells.
Plant Cell Physiol. 2005, 46, 843–857.
C
Proteomics 2016, 16, 3183–3187 3187
[30] Borner, G. H., Lilley, K. S., Stevens, T. J., Dupree, P., Identifi- cation of glycosylphosphatidylinositol-anchored proteins in Arabidopsis. A proteomic and genomic analysis. Plant Phys-
iol. 2003, 132, 568–577.
[31] Oh, I. S., Park, A. R., Bae, M. S., Kwon, S. J. et al., Secretome analysis reveals an Arabidopsis lipase involved in defense against Alternaria brassicicola. Plant Cell 2005, 17, 2832– 2847.
C
41 Cette étude nous a permis de comparer notre méthodologie à celles suivies par les études précédentes. En tenant compte de la faible quantité de protéines produites et du nombre d’échantillons traités, la stratégie shotgun est apparue très pertinente pour la suite de notre étude. C’est donc avec cette méthodologie que l’analyse des protéomes de tige a été traitée.