Full title: Label free quantitative proteomics of Qualea grandiflora Mart. (Vochysiaceae) roots indicates an aluminium requirement for growth and development

Aluminium in acid soils is a hindrance to crop growth. Nonetheless, Brazilian Cerrado possesses many species such as Qualea grandiflora, which are adapted to acid soils with large amounts of Al and accumulate this metal in its tissues and organs. Nonetheless, the mechanisms involved in these processes are poorly understood, mainly at molecular level. Thus, a root proteomic analysis was accomplished to identify Al-responsive proteins in Q. grandiflora plants. Concomitantly, a root growth analysis of plants grown with and without Al supplementation was conducted to determine the effects of Al on the whole plant. Subsequently, proteins from both treatments were identified and quantified by LC-MS/MS in a label-free fashion. From the 2,520 identified proteins, 410 were differentially abundant between the two treatments, which were associated with carbohydrate metabolism, redox activity, stress response and catabolism of organic compounds. Furthermore, Al was crucial for the growth and development of Q. grandiflora. In fact, this species may have an Al-dependent metabolism. Moreover, it was possible to correlate plant growth to Al-upregulated proteins that were directly involved in cell wall synthesis, oxidative phosphorylation, genetic information processing, and amino acid metabolism. Additionally, this work provides an extensive dataset of Al-regulated protein in Q. grandiflora, which will be crucial to understanding the functions of Al in this species.


60
Aluminium (Al) is highly abundant in soils worldwide where it is found in different 61 chemical species. In acidic conditions (pH < 5.5), these compounds are found as Al +3 , which 62 constitutes one of the most toxic form of Al, which is limiting for crop production. 63 Nevertheless, there are plants that do not resent the presence of Al +3 , and some of which are 64 commonly found in the Brazilian Cerrado. The Cerrado biome is a neotropical savannah whose Furthermore, proteomics has been used to identify Al-responsive proteins. For example, 109 Wang et al. (2014) found 106 differentially abundant proteins from Al-treated rice roots and 110 observed that the glycolysis/gluconeogenesis processes were upregulated by Al. In addition, 111 all glycolysis/gluconeogenesis related genes were more expressed in Al-tolerant rice cultivar 112 than in the sensitive [13]. This fact suggests energy availability is crucial for Al-tolerance. 113 To date and to the best of our knowledge, data on genes and proteins involved in Al   125 A flowchart was elaborated indicating the main steps carried out in the growth and 126 proteomic analyses of roots of Q. grandiflora in response to Al (Fig 1).   All growth analysis data was statistically analysed by two-way ANOVA and the 161 differences among the means were tested by the Student's t-test (p< 0.05). Approximately 100 mg of each root sample were ground in liquid nitrogen using a mortar 177 and pestle. Then, the powder was added to a solution of 10% (w/v) trichloroacetic acid and 178 0.07% (v/v) β-mercaptoethanol in cold acetone, and the resulting suspension was thoroughly 179 mixed by vortexing, and then incubated for 3 h at 4 °C.

180
After incubation, samples were centrifuged at 10,000 g for 20 min at 4 °C. Then, the 181 supernatant was removed, and the remaining pellet was washed five times with 10% (w/v) 182 trichloroacetic acid in acetone until the total disappearance of pigments. The pellet was dried 183 using a Speed-Vac concentrator and resuspended in rehydration buffer (7 M urea, 2 M thiourea, 184 250 mM TEAB, pH 8.5). Protein concentration was determined by using Qubit ® 2.0 assay 185 (Invitrogen, Carlsbad, USA) and extracted protein quality was assessed in 10% SDS-PAGE. 188 The extracted proteins (200 μg) were reduced with 10 mM dithiothreitol (DTT) for 60 min 189 at 56 °C and alkylated with 100 mM iodoacetamide (IAA) for 60 min at 37 °C in the dark. 190 Then, the samples were diluted in 100 mM NH 4 HCO 3 (ammonium bicarbonate), pH 8. 1. 191 Subsequently, the alkylated proteins were digested with trypsin (1:50 v/v -Promega, Madison,192 USA) at 37 °C for 16 h. After digestion, the resulting peptide solution was acidified with 0.1% 193 trifluoroacetic acid and centrifuged at 10,000 g for 10 min. Then, the supernatant was desalted   The files obtained from the mass spectrometer were analysed using the software Progenesis 218 IQ for alignment of the MS1 peaks found in the chromatograms, extracted ion chromatogram (XIC)-based quantification and normalization. A first statistical analysis was performed before 220 the identification of the MS1 features, to filter for identification only those presenting ANOVA 221 p-values < 0.05.

222
After the peptide peaks were quantified and grouped, the identification of proteins was 223 performed using Peaks 7.0 software, which deduced sequences from the fragmentation  The identified proteins were filtered at a rate of 1% for false discovery rate (FDR), and a 232 minimum of one unique peptide per protein was required for identification.

233
The protein identification information was imported into the Progenesis IQ software, which 234 combined them with previously generated quantitative data. Multivariate PCA analysis was 235 performed in Progenesis to evaluate the grouping of replicates and conditions, as well as to 236 cluster the abundance profiles. In this study, a protein was considered differentially abundant 237 when it presented fold-change ≥ 1.5 with p ≤ 0.05 after the ANOVA test at the protein level.    The means followed by the same letters within same column do not differ statistically by the student's 275 t test (p < 0.05), n = 5. SD: standard deviation.  285 grandiflora plants grown with or without Al. It is noteworthy that the chloroplasts from each 286 treatment had a distinct structure (Fig 3). Note that the chloroplast from leaves of Al-287 supplemented plants had a standard structure, with typical shape and regular internal membrane 288 system with grana and thylakoids (Fig 3 A). Differently, the chloroplasts from non-treated 289 plants showed abnormal structure. Moreover, even the chloroplasts from green leaves already 290 had a peculiar internal membrane system (Fig 3 B). Note that in these chloroplasts the lumen 291 of thylakoids appeared dilated, and no typical assembly of granum (Fig 3 B). Furthermore, in 292 yellowish leaves, the thylakoid dilation was very pronounced, and the stroma appeared to be 293 exceedingly large compared with chloroplasts from green leaves (Fig 3 B).

302
A growth analysis of Q. grandiflora shoots and roots was performed to investigate whether 303 the morphological differences observed between the treatments were associated with the lack 304 of Al supplementation. The results indicated that Al was critical for growth and development 305 of Q. grandiflora plants. Note that leaves from Al-supplemented plants were greener than those 306 from non-treated plants. Moreover, it is noteworthy that shoots and roots from Al-treated plants 307 were considerably higher and the roots were longer and more branched (Fig 2, Table 2). 308 Furthermore, after 120 days of cultivation the average length of shoot of Al-treated plants was 309 about 14% higher than shoots from non-treated plants. In roots this difference was even greater 310 and reached 24% in favour of plants grown with Al (Table 2). 311 Besides, root biomass accumulation was determined by measuring root fresh and dry 312 weights (Table 3).  Numbers followed by the same letters, within the same column and, are statistically similar as calculated by the 320 Student's t-test (p < 0.05), n=20. Numeric values the means ± standard error.

Root biomass accumulation 323 324
Consistent with root length data, Al-treated plants had significantly higher root biomass 325 compared with those not treated ( Table 3). The results showed that after 120 days of cultivation 326 root fresh and dry weights of roots grown with Al were respectively 24% and 40% higher than 327 those that did not receive Al. This fact indicates that the root system of Q. grandiflora was 328 stimulated by Al.  Gel-free and label-free approach were used to determine the proteomic changes in Q. 338 grandiflora roots grown with or without Al-supplementation. A total of 2,520 distinct proteins 339 were identified in both treatments by using a transcriptome sequence database and PEAKS and 340 PepExplorer software. For the identification of differentially abundant proteins, it has been 341 considered a fold-change ≥ 1.5 and a p-value ≤ 0.05 (Supplementary Table 1). Thus, in Q.  The most enriched BP categories in Q. grandiflora roots were metabolic processes whose 366 proteins were associated with nitrogen compound metabolism, biosynthetic process, organic 367 substance metabolism, catabolic process, regulation of cellular process, single organism 368 metabolism, and primary metabolism. In fact, within these categories the primary metabolism 369 had 112 upregulated proteins in roots from Al-treated plants (Supplementary Table 2).

370
In addition, the main cellular components associated with Al response in Q. grandiflora 371 roots were located at plasma membrane, cell wall, plastid, mitochondrion, and ribosomal 372 subunits, as well as the extracellular compartment ( Fig 5). to 117 different metabolic pathways. Also, the results showed that differentially abundant 397 proteins were mainly involved in genetic information processing, followed by carbohydrate 398 metabolism, cellular processes, amino acid metabolism, energy metabolism and lipids (Fig 7).   Therefore, the enrichment analysis revealed 24 metabolic routes that were statistically 427 significant (FDR<0.05) in Q. grandiflora roots in response to Al (Table 4). Moreover, the most 428 relevant metabolic pathways indicated that proteins with differential abundance were mainly  STRING analysis was performed with a high level of stringency (confidence score = 0.900).

439
A network of interaction between ribosomal family proteins involving 21 proteins whose 440 abundance was significantly increased with FDR 1.36 e-11 (Fig 8A) was observed. The

441
STRING analysis also indicated that the differential proteins with significantly decreased 442 abundance and a strong interaction in response to stresses were present. This interaction 443 network involved 31 proteins, mostly HSPs (heat shock proteins) with FDR 1.07 e-05 (Fig 8B).  biomass. Additionally, as it is discussed below, these morphological and growth data are Besides, proteins related to cellular components such as ribosomes, cell membranes, 493 mitochondria, and plastids had their abundance increased in Al-treated plants (Fig 5).

538
For instance, germins and germin-like proteins (GLPs) are ubiquitously found in plants and 539 have been associated with various developmental and biological processes including cell wall The comparative analysis of differentially abundant proteins in response to Al in Q. 594 grandiflora roots revealed 20 proteins involved in lipid metabolism whose abundances 595 increased in response to Al. A KEGG analysis showed that fatty acid degradation pathways 596 were significantly enriched (Table 3) Al-treated Q. grandiflora roots had better growth rate and accumulated more biomass. It is noteworthy that the KEGG analysis indicated that about 30% of the differentially 636 abundant proteins were related to genetic information processing, which is the highest among 637 all categories catalogued (Fig 7). That is not a coincidence. Concomitantly, elevated levels of indicating that ribosomal proteins, amino acid metabolism, and biosynthesis, pyrimidine, and 642 purine metabolism were responsive to Al in roots of Q. grandiflora (Table 3).  membrane components as well as proteins associated with mitochondria and plastids (Fig 5).
Furthermore, the most active molecular functions of proteins were related to catalytic and 700 binding activities (Fig 6). Therefore, a few aspects that involve cellular components and protein 701 molecular functions that have not yet been addressed and are crucial for plant growth and 702 development will be further discussed.

831
 Carbohydrates components of pectins. These pectin components are synthesized highly methylated, otherwise they would prevent cell 832 expansion due to crosslinking of de-esterified pectin polymers by Ca. Pectin de-esterification is a highly controlled process that 833 involves cell-wall based enzymes named pectinesterases, and it is directly associated with cell wall adhesion/loosening.

834
 SAMS is believed to be crucial for genetic information processing, as it may provide methyl groups through SAM synthesis for DNA 835 methylation, which is one epigenetic mechanism that often modify gene expression. Besides, SAM is also associated with RNA 836 capping, RNA splicing, RNA transportation, mRNA stability, and translation initiation. Without SAMS, the flow of genetic 837 information would be seriously impaired. Acyl-CoA oxidase (ACOX) -catalyses the first step of beta-oxidation of fatty acids, likely in peroxisomes, which may form acetyl-CoA, an important 855 compound of TCA cycle, and thereby, crucial for ATP production.

859
The present work indicates that Al is a required for the growth and development of