The experiment was conducted at the Federal University of Viçosa (UFV) in Viçosa, Minas Gerais, Brazil (20˚45'14''S, 42˚52'53''W, 648.74 m). The resistant genotype BGH-2127 access [7] , conserved in the Gene Bank of Federal University of Viçosa, Minas Gerais State, Brazil, and susceptible cultivar Santa Clara [5] , both belonging to the S. lycopersicum species, were evaluated. The plants were grown in a greenhouse in a completely randomized design with three replicates per genotype. Each replicate consisted of one set of three plants. Each plant was grown in one vessel of 10-L volume for soil.

The collection, preparation, and inoculation of isolates of P. infestans was performed according to a methodology proposed elsewhere [5] . After 45 days of transplanting, the plants were inoculated with a mixture of sporangia from P. infestans isolates, which are pathogenic to tomato, collected in fields of production of tomato from different cities of Minas Gerais State, Brazil, including Cajuri, Coimbra, Ervália and Viçosa. For inoculum multiplication, infected leaflets were kept in plastic trays previously lined with paper towels moistened in distilled water and kept at 18˚C for 48 h. Subsequently, a sporangium suspension was prepared for each isolate, and the hemacytometer concentration was adjusted to 1 × 10³ mL⁻¹ of sporangia. The inoculation was performed by spraying 10-mL of suspension per plant by using the manual costal spray. A day after the inoculation, in order to guarantee high humidity to the environment, the plants were irrigated by micro-sprinkler.

Leaf sampling times were established based on the search for rapid plant responses and the time required for the pathogen to penetrate and colonize the plant tissues [18] . After the inoculation of each plant, the leaves were collected at every 0, 2, and 48 h; the leaves of uninoculated (control) genotype were also collected at the same time. Time 0 (zero) corresponds to the collection of leaves immediately after inoculation. Soon after the collections, the leaves were frozen under liquid nitrogen (at −80˚C) until requirement for the preparation of plant extracts for molecular analysis.

Trypsinization was performed according to the method described earlier [23] , with the following modifications. Initially, differentially abundant spots were removed from the gels. For dye removal, pieces of gel containing the proteins were transferred to 500 μL tubes containing 50% acetonitrile solution in 25 mM (v/v) ammonium bicarbonate solution for the first wash (one time) for 12 h. Then, a second wash in the same solution was performed for 1 h and another wash with 50% methanol in 25 mM (v/v) ammonium bicarbonate solution) for another hour. The decolorization solution was removed, and the gel pieces were dehydrated with acetonitrile for 5 min for two-fold drying in the Speed Vac Concentrator Plus (Eppenddorf) for 10 min. The proteins were reduced with 25 mM DTT in 100 mM ammonium bicarbonate for 30 min at 56°C in a water bath and alkylated with 75 mM iodoacetamide in 100 mM ammonium bicarbonate for 30 min at the room temperature. The gel pieces were washed with 100 mM ammonium bicarbonate for 10 min for two-fold drying and then dehydrated in acetonitrile for 5 min by drying in the SpeedVac for 10 min.

For enzymatic digestion, the gels were rehydrated with a solution containing trypsin (20 μg/mL) in 40 mM ammonium bicarbonate solution (pH 8.0) and 10% acetonitrile. The enzyme solution (15 μL) (Trypsin Gold, Mass Spectrometry grade, Promega V5280) was added cold (at 4˚C), and the samples were kept on ice for 45 min to penetrate the gel. The samples were added with 50 μL of 40 mM ammonium bicarbonate solution in 10% acetonitrile and incubated at 37˚C for 22 h in a water bath.

The gel pieces were sonicated for 10 min, shaken at 1500 g for 2 min under ambient temperature, and the solution was placed into fresh tubes. To the remaining pieces of gel, 40 μL of the solution of 5% formic acid in 50% (v/v) acetonitrile was added in two sequential steps to recover the largest amount of tryptic fragments. They were agitated again at 1500 g for 2 min under ambient temperature and the solution was removed and added to the new tube.

After these steps, the entire solution was completely evaporated in the Speed Vac and the tryptic peptides were re-suspended in 2.5 μL of 0.1% trifluoroacetic acid (TFA) solution. All samples were submitted to desalination on a C18 hydrophobicity column (ZipTip-Eppendorf), according to the recommendations of the manufacturer.

Protein identification was performed using the MALDI-TOF/TOF Ultraflex III (BruckerDaltonics)-type mass spectrometer, available in the Nucleus of Biomolecules, UFV. About 1 μL of each sample containing the peptides was applied to the steel plate of the mass spectrometer together with α-cyano-4-hydroxy cinnamic acid matrix (Sigma) in a 1:1 ratio. The samples were analyzed in a positive mode with the reflector activated.

The standard list of proteins was obtained from the Lycopersicum Protein Database (download on 03/21/16, with 33,950 entries), Solanaceae family (download on 03/14/2016, with 114,203 entries), and the Viridiplantae group (download on 04/27/15, with 2,864,275 entries), all obtained from UNIPROT, through the use of the MASCOT application, version 2.4.0 (Matrix Science, London, UK). The parameters used for research included enzymatic digestion by trypsin with a lost cleavage, allowing a tolerance error of 0.2 Da for the parental ion and of 0.5 Da for the fragments, carbamidomethylation of cysteine as a fixed modification, and the oxidation of methionine as a variable modification.

Proteins identified as “Uncharacterized” by MASCOT were processed by the BLAST algorithm [24] . Through this algorithm, it was possible to identify the proteins present in the database deposited in the Phytozome, whose sequences showed a higher level of identity with the sequences of the “Uncharacterized” proteins.

The result obtained by MASCOT was validated by the SCAFFOLD application version 3.6.4 (Proteome Software Inc., Portland, OR). For validation of the identified peptides, the Peptide Prophet Algorithm was applied [25] and the Protein Prophet Algorithm [26] was used as the acceptance criterion parameters were adopted with a minimum of 90% probability of identification for both peptides and proteins, with the presence of at least one single peptide for each identified protein.

The identified proteins were functionally categorized using the Mapman program [27] .

Once the differentially abundant proteins for BGH-2127 and Santa Clara genotypes were listed, the construction of a protein-protein interaction network was performed using the STRING v software version 10.0 [28] .

The plant leaf samples were macerated in liquid nitrogen and total RNA was extracted with TRIZOL reagent (Invitrogen), according to the manufacturer’s protocol. In approximately 100 mg of the macerated material, 1 mL of the TRIZOL reagent was added and incubated for 5 min. Subsequently, 200 μL of chloroform was added to the tube and the mixture was homogenized for 3 min. The samples were centrifuged at 12,000 ×g for 15 min at 4˚C. The supernatant was transferred to a new tube, and 500 μL of isopropyl alcohol was added. After 30 min of centrifugation at 12,000 ×g for 10 min at 4˚C, the supernatant was removed and the pellet formed were washed with 75% ethanol. Next, the pellet was centrifuged at 9500 ×g for 5 min at 4˚C, excess ethanol was removed, and the pellet was re-suspended with free water from DNases and RNases. RNA was quantified SpectraMax M5 microplate/cuvette reader (Molecular Devices) and analyzed on 1.5% (w/v) agarose gel stained with 0.1 μg/mL ethidium bromide.

To eliminate any contamination with genomic DNA, total RNA extracted was treated with RQ1 RNase-Free DNase (Promega). To that end, 3 μL of DNase (1 U/μL), 1 μL of RQ1 RNase-Free DNase 10X Reaction Buffer, and DEPC water were added to the total RNA (3 μg) to complete the final volume of 10 μL. The samples were incubated at 37˚C for 30 min. The enzyme was inactivated by the addition of 1μL of RQ1 DNase Stop Solution and incubation for 10 min at 65˚C. For reverse transcription, 500 ng of DNase-treated RNA was used in the reaction and 1 μL of oligodT (stock at 500 μg/mL) was added.

The RNA was denatured at 70˚C for 5 min (followed by incubation on ice for 1 min) and 4 μL of 5× First-Strand Buffer, 2 μL of 0.1M DTT, 1 μL of dNTP mix (dGTP, 1 μL of 40 U/μL RNaseOUT Recombinant Ribonuclease Inhibitor, 1 μL of the enzyme M-MLV RT (200 U/μL; Promega) and DEPC water to complete the final volume of 12 μL. The samples were incubated at 42˚C for 60 min. The obtained cDNA was quantified by the SpectraMax M5 microplate/cuvette reader (Molecular Devices).

Real-time PCR assays were conducted according to the Applied Biosystems manuals. Specific primers were designed using online software Primer 3 (http://frodo.wi.mit.edu/primer3/). The cDNA samples were diluted to a final concentration of 100 ng/μL. Amplification of the target fragments was performed using StepOne™ Real-Time PCR System (Applied Biosystems). The total reaction volume per sample was 10 μL, which was composed of specific oligonucleotides encoding the 2-cis peroxyredoxin BAS1 (FW: 5'-CCTTCTGCTTTTAATGGACTTCGT-3'; RV: 5'-TGAGCAACCCGAGTATTGATTG-3'), and thaumatin-like (FW: 5'-GCGGTGGTCGACGTCTTG-3'; RV: 5'-ACCCCATATACGTGCCATCTTAGT-3'), inoculated and uninoculated (control) BGH-2127 plant cDNAs with P. infestans, and the SYBR Green PCR Master Mix Kit (Applied Biosystems). The reactions were performed in duplicate on a 48-well plate. The amplification conditions followed were 95˚C for 10 min, 40 cycles of 94˚C for 15 s, and 60˚C for 1 min. For the quantification of gene expression, the REST software was used. The primer specific actin (FW: 5'-CCTTCAACGTTCCAGCTATG-3'; RV: 5'-TCACCAGAGTCCAACACAATAC-3') [29] was used as an endogenous control for the normalization of qRT-PCR. For the quantification of gene expression, the REST software was used (https://www.qiagen.com/) [30] . Statistical differences are significant when p < 0.05 was calculated by comparing the cycle thresholds of the target gene with that of the housekeeping gene.

The two-dimensional gel proteomic analysis of tomato genotypes BGH-2127 and Santa Clara after 0, 2, and 48 h of inoculation with P. infestans, a total of 311 spots were detected, 94 of which were considered to be differentially (p < 0.05). From this result, 56 spots were identified by mass spectrometry, of which 17 were associated to Santa Clara cultivar (Figure 1(a)) and 39 to BGH 2127 (Figure 1(b)). Only 6 differentially abundant spots (19, 28, 122, 132, 188, and 303) were common to the two genotypes, evidencing that different mechanisms are used by these genotypes against late blight.

The 17 proteins identified from the Santa Clara genotype were categorized into five functional groups, 23% were related to metabolism and energy (4 proteins), 47% to photosynthesis (8 proteins), 18% to stress and defense (3 proteins), 6% to transcription (1 protein) and 6% to non-characterized group (1 protein). Of these, 8 genotypes were down-regulated and 9 were upregulated (Table 1). An increase in the relative abundance of proteins involved in the stress and defense process and for uncharacterized protein were observed. Proteins related to metabolism and energy as well as those related to photosynthesis showed changes in their balanced abundance levels. For protein related to transcription, a reduction in the relative abundance was observed (Figure 2).

A total of six functional groups of proteins related to BGH-2127 genotype were formed, 15% of which were associated with energy and metabolism (6 proteins), 46% to photosynthesis (18 proteins), 23% to stress and defense (8 proteins), 8% to transcription (3 proteins), 3% to other proteins (1 protein), and 5% to non-characterized groups (2 proteins). Among the 39 proteins identified

^aID represents the identification number of the proteins identified in the two-dimensional gels. (T0), (T2), and (T48) at the time when the spot was differentially abundant. ^bNumber of access in UNIPROT. ^cIsoelectrical (pI) points and experimental and theoretical molecular masses (MM). ^dLevel of abundance (%) of spot volume: signals (+) and (−) indicate higher (upregulated) and lower (downregulated) protein abundance, respectively. ^ePercentage of coverage. ^fScore Mascot. ^gPeptídes identified and confirmed with more than 90% certainty by Scaffold software. *BLAST performed from the sequence obtained by UNIPROT in Phytozome.

genotype, 15 were downregulated and 24 were upregulated (Table 2). After inoculation, an increase was noted in the relative abundance of most proteins involved in energy and metabolism processes, photosynthesis, stress, and defense, in addition to uncharacterized proteins. Most proteins related to transcription and even other proteins had reduced relative abundance (Figure 3).

Several proteins were identified in more than one spot on the same gel. The 17 spots identified in the Santa Clara genotype corresponded to 14 proteins (Table 1), while 39 spots being related to BGH-2127, which corresponds to 25 proteins (Table 2). Three spots were identified as different forms of the protein fructose bisphosphate aldolase (90, 94, and 96) in Santa Clara (Table 1). Two spots associated to the BGH-2127 genotype were identified as ATP synthase alpha subunit (122 and 125), three as ATP synthase beta subunit (24, 129, and 132), four as subunit II of the reaction center of photosystem I (37, 140, 268, and 271), two as proteins involved in oxygen enrichment 1 (185 and 188), and two as proteins involved in oxygen enrichment 2 (43 and 47) (Table 2).

An important point observed from among the results was that the same protein represented by different spots had opposite levels of abundance. This was observed in spots 37, 140, 268, and 271, identified as subunit II of the reaction center of photosystem I for BGH-2127. Spots 37 and 268 were upregulated, while 271 was downregulated. It should be noted that spot 140, differentially abundant in the three times evaluated, was upregulated at 0 and 48 h of inoculation and downregulated at 2 h of inoculation (Table 2).

The gene transcription levels (mRNA) encoding the 2-Cys peroxiredoxin BAS1 (spot 44) and thaumatin-like (spot 45) proteins evaluated for BGH-2127 did not

^aID represents the identification number of the proteins identified in the two-dimensional gels. (T0), (T2), and (T48) at the time when the spot was differentially abundant. ^bNumber of access in UNIPROT. ^cIsoelectrical (pI) points and experimental and theoretical molecular masses (MM). ^dLevel of abundance (%) of spot volume: signals (+) and (−) indicate higher (upregulated) and lower (downregulated) protein abundance, respectively. ^ePercentage of coverage. ^fScore Mascot. ^gPeptídes identified and confirmed with more than 90% certainty by Scaffold software. *BLAST performed from the sequence obtained by UNIPROT in Phytozome.

correlate with the abundance levels of the respective proteins. Expression of the gene encoding 2-Cys peroxiredoxin BAS1 significantly decreased at time 0 (zero) for the inoculated treatment, while its corresponding protein obtained greater abundance. For thaumatin-like (spot 45), no difference was noted in the expression level of the gene coding for this protein at 48 h of inoculation; however, the abundance of the respective protein was higher in the inoculated treatment. The relative abundance of proteins may or may not be proportional to the mRNA levels since, after mRNA synthesis, different regulatory processes occur, such as post-transcriptional, post-translational, and protein degradation, which act to control protein abundance.

The observed interactions among proteins identified to be associated with Santa Clara genotype occurred among those related to the photosynthetic process as well as between the metabolic and energy proteins. The proteins thioredoxin peroxidase (spot 48), N2-acetylornithine (spot 116), and β-1,3 endo glucan glucosidase B (spot 303) have no functional relationship to each other and with other proteins (Figure 4).

In the protein interaction network of the BGH-2127 genotype, for stress-defense-related leucine-rich N-terminal domain (spot 84), interaction was observed with the ribulose 3-phosphate epimerase protein (spot 53), which in turn interacted with ribulose bisphosphate carboxylase chain small chain (spot 5) and ribulose bisphosphate carboxylase small chain 3B (spot 275). Two other networks of observed interactions occurred between the proteins related to the photosynthetic process and also between energy proteins and metabolism. For the other proteins related to stress and defense, transcription and other proteins, no functional relationship was observed (Figure 5).

Based on the variations in the differential abundance of proteins identified at different times (0, 2, and 48 h) before and after inoculation, their relationships with different biological processes were considered, aiming to demonstrate the possible contributions to the resistance to late blight from BGH-2127. Infection by the late blight induced a greater change in the number of proteins with differential abundance in the genotype BGH-2127 in relation to the Santa Clara genotype.

The proteins identified were classified into different functional groups, confirming that the resistance of plants to pathogens results from several physiological factors and/or biochemical mechanisms [34] . For both the genotypes, the groups of functional classifications were practically the same, being related to the processes of energy and metabolism, photosynthesis, stress and defense, transcription, and uncharacterized features. The exception was the group of other proteins related to BGH-2127 access. However, only six differentially abundant proteins were found to be common to both the genotypes. Among these proteins, Ferredoxin-1 and ATP synthase alpha subunit were downregulated and ATP synthase beta subunit and protein involved in oxygen enrichment 2 were upregulated, whereas the IV subunit of the reaction center of photosystem I and Glucan endo-1,3-beta-glucosidase B obtained opposite patterns in both genotypes. The variability between proteins with differential abundance during infection by late blight evidences different responses among the evaluated genotypes.

Among the proteins found unique to each genotype, some were identified in more than one spot on the same gel. For the Santa Clara genotype, spots 90, 94, and 96 were identified as the protein fructose bisphosphate aldolase. In the genotype BGH-2127, two spots were identified as ATP synthase alpha subunit (122 and 125), three as ATP synthase beta subunit (24, 129, and 132), four as subunit II of the reaction center of photosystem I (37, 140, 268, and 271), two as proteins involved in oxygen enrichment 1 (185 and 188), and two as proteins involved in oxygen enrichment 2 (43 and 47). These results indicated spots with different values of molecular mass (MM) and isoelectric point (pI) for the same protein, which occurred due to changes in the proteome that can be attributed to post-translational modifications or the presence of different forms of the same protein of a given functional family [35] . Post-translational modifications are important processes as they influence the activity and function of proteins by linking them with other functional groups [36] .

Proteins represented by spots 37, 140, 268, and 271 were identified as subunit II of the reaction center of photosystem I with different levels of abundance (some upregulated and other downregulated spots). One hypothesis for this phenomenon is that the different forms of protein play distinct molecular functions and are also responsible for the regulation of complex mechanisms during pathogen infection [35] .

In plants, many proteins are involved in the metabolic and energetic pathways, as they are considered important in response to infection [37] , because they mobilize high metabolic costs to overcome pathogen attack [38] . Significant variations in the energy-related proteins and metabolism between the two treatments (inoculated and control not inoculated with P. infestans were evident. Differences in the regulation of proteins of this functional group can contribute to resistance to the re-chemistry of BGH-2127 in comparison with that of Santa Clara genotype. Approximately 67% of the proteins involved in this process for BGH-2127 had increased relative abundance level after inoculation, whereas, for Santa Clara, only half of these proteins showed an increase in the level. The increase in energy-related protein abundance and metabolism in infected plants may be justified by the fact that these plants have a higher energy demand to overcome the pathogen attack, which usually occurs by increased respiration [39] .

Among the proteins that showed increased levels of abundance after inoculation with the pathogen for BGH-2127, it is highlighted that they form a part of the ATP synthase complex, which consists of several subunits and use the transmembrane proton gradient for cellular ATP biosynthesis [40] . Considering plant resistance to pathogens, cellular ATP plays an important role as a signaling molecule in response to stress, including those caused by pathogens [41] .

Triphosphate isomerase is a key glycolytic pathway protein that catalyzes the interconversion between glyceraldehyde-3-phosphate and dihydroxyacetone phosphate. This protein has often been identified as a potential target of both glutathionylation and nitrosylation in proteomic studies, suggesting that its activity can be regulated under stress conditions [42] . In the stages of development of the phytopathogenicoomycete Phytophthora capsici, the protein triphosphate isomerase was upregulated in the presence of the pathogen hyphae [43] .

Photosynthesis is an important metabolic process that broadly contributes to the overall state of cellular energy for redox balance by providing NADPH, ATP, and carbon skeletons that support plant growth. In addition, they stimulate the initiation and maintenance of responses against external stress [44] , which play the key role in host defense, which is one of the main processes affected [45] . Plants when subjected to some type of stress usually respond with a decrease in the synthesis of cellular proteins and an increase in the specific synthesis of proteins involved in defense mechanisms [46] . This indicates that, during a pathogen attack, biosynthesis of defense-related compounds is a priority for a plant, while other cellular activities are reduced, such as in the case of photosynthesis, until pathogen growth is disrupted [39] [47] .

P. infestans is a hemibiotrophic pathogen, the infection of plants by this pathogen involves the transition from an asymptomatic initial biotrophic phase to a necrotrophic state, which is characterized by cell death [13] . In this context, the maintenance of photosynthesis would be advantageous for the development of late blight; therefore, reduction of photosynthesis is beneficial to the host and can be an important part in the defense of the plant, because the reduction of photosynthesis may deprive the pathogen of important nutrients [48] . For BGH-2127, after inoculation with P. infestans, an increase in the relative abundance of most proteins involved in the photosynthetic process was noticed; it was upregulated in ten proteins and downregulated in eight. However, it was noted that approximately 88% of the downregulated proteins (spots 5, 19, 28, 53, 140, 226, and 275) were differentially accumulated in time 2 h after inoculation, which corresponds to the biotrophic infection phase described in a previous study [13] . For the Santa Clara genotype, there existed a balance between the relative abundance levels of the proteins identified at 2 h after inoculation, with 50% of these down-regulated (spots 19, 90, and 96). These results also indicate that photosynthesis is affected differently among the genotypes evaluated during blight infection.

The stress and defense responses of an organism, including, in plants, play an important role during cellular development against pathogen infection [35] . Among the proteins identified for the genotypes BGH-2127 and Santa Clara, 20.5% and 17.6%, respectively, belong to the stress and defense group. Only the β-1,3 endoglucan glucosidase B (spot 303) protein differentially abundant 48 h after inoculation was found to be common to both the genotypes. However, opposite proteins abundance levels were observed, with values equal to −6.32 for BGH-2127 and +1.70 for Santa Clara, evidencing that this protein is not the major protein involved in the resistance of BGH-2127 and that different defense mechanisms are used by these genotypes.

Another protein that had reduced levels of relative abundance was the protease inhibitor (spot 210), which differentially accumulated at times 0 and 48 h after inoculation. During the infection process, the pathogen produces enzymes that target host cells, which in turn encode proteins with protease inhibitory activity to suppress the enzymatic activity of the pathogen [49] . Due to the reduction of the levels of this protein against the infection of the late blight, it is evident that the protease inhibitor protein is not the main defense mechanism used by the resistant genotype.

Two proteins related to oxidative stress have relative abundance that increased at time 0 after inoculation for BGH-2127, 2-cis peroxiredoxin BAS1 (spots 44 and 213), and 2-cis peroxiredoxin (spot 51). In plants, the attack of pathogens results in the rapid formation of reactive oxygen species, which can degrade proteins [50] . Reactive oxygen species can be eliminated by a network of enzymatic and non-enzymatic antioxidants like peroxiredoxins [51] , which justifies the increase in the relative abundance of these proteins immediately after inoculation. Peroxiredoxins are a part of subfamilies in which the peroxidases are subdivided [52] . When they act in defense of cells against oxidative explosion, peroxidases consume hydrogen peroxide in the cytosol, vacuole, cell wall, and extracellular space. They are also linked to metabolic processes, lignin, and suberine formation, structuring and stretching of the cell wall, and hypersensitivity reactions, making it difficult for the pathogen to invade the plant [9] [53] .

The differentially abundant spot 45 after 48 h of inoculation was identified with a protein from the thaumatin-like family. Proteins of this family are involved in the plant defense system against biotic and abiotic stresses [54] , including important functions like resistance to pathogens, cold tolerance, salinity, and resistance to drought [55] . The abundance level of thaumatin was +3.45 times greater as compared to the control treatment, indicating the importance of this protein for resistance of BGH-2127, since the presence of thaumatin-likeprotein has been associated with activity against oomycetes [10] . The proteins of the thaumatin-like family belong to group 5 (PR-5) of proteins related to pathogenesis [10] . Thaumatin’s act by inhibiting the growth of hyphae and reduce the germination of spurs, probably by membrane permeabilization and/or by interaction with pathogen receptors [56] . The PR-5 has a strong affinity and hydrolytic activity for 1,3-β-glucan, which extracellularly detect 1,3-β Glucan due to the presence of a kinase domain and owing to a transmembrane potential [10] . The significant ability of thaumatin to degrade 1,3-β-glucan is a strong indication of mechanisms associated with BGH-2127-access resistance, since the cell wall of oomycetes is composed mainly of this polymer [57] . In molecular studies involving the pathosystem Zingiberzerumbetvs Pythium aphaniderm for ZzPR5 protein belonging to the family of thaumatins, it was observed that the constituent amino acids of the protein hydrolyze 1,3-β-glucan, suggesting a potential anti-oomiceto activity [58] . In the specific case of the Phytophthora infestans oomycete, resistance tests on transgenic potatoes elucidated the potential of thaumatin-like protein against the pathogen, as there are no efficient control methods or resistant cultivars. The overexpression of thaumatin-like genes was found to be responsible for the substantial increase in the pathogen resistance [59] .

Carboxypeptidase (spot 93) is a proteolytic protein involved in several cellular processes [60] are responsible for protein metabolism and in hydrolyzing peptides acting at the ends of the protein [61] . The observed increase in the relative abundance of this protein may contribute to the retention of BGH-2127, since carboxypeptidases are involved in the regulation of defense against infection by pathogens and oxidative stress [62] .

Spot 84 was identified as the Leucine repeat rich N-terminal domain protein. Most dominant R genes encode proteins of this class, the structure of which is composed of an N-terminal domain, a nucleotide-linked central domain and a leucine-rich C-terminal domain [63] . This protein functions as an intracellular receptor that detects pathogen-affecting proteins, which activates a set of defense responses associated with programmed cell death [64] . Effector-triggered immunity is a rapid response of the plant to the pathogen’s attack; therefore, a +1.63-fold increase was noted in the relative abundance of the leucine-rich N-terminal domain protein at 0 h after inoculation. In tomato culture, a leucine-rich protein (CC-NBS-LRR) has been reported to confer resistance to P. infestans [65] .

Two transcript-related proteins were differentially accumulated in BGH-2127. The protein-splicing factor 3b, subunit 4 (spot 265), was differentially accumulated at times 0 and 48 h after inoculation, both showing upregulation. Splicing is a process that removes the introns and binds the exons after RNA transcription. In this context, splicing factor proteins are important to maintain efficient processing of RNA under stress conditions. Post-transcriptional mechanisms based on splicing regulate the activation of pre-existing molecules to ensure an immediate response to stress [66] . The second protein identified for transcription was peptidyl-prolyl cis-trans isomerase (spot 270) differentially acclimated in 48 h of inoculation. This protein is considered to be a part of the plant’s immume system that catalyzes the isomerization of peptide bonds in proline residues to specifically regulate protein conformation changes [67] . Peptil-prolylcis-trans isomerases are known to interact with other biological units to facilitate protein folding, regulate developmental processes, and certain aspects related to stress responses and cell death [68] . There is increasing evidence pointing to functions associated with plant defense against pathogens exerted by proteins members of the peptidyl-prolyl-cis-trans isomerase family [69] .

Proteins do not act as individual units in cells, but rather in conjunction with other proteins to form certain networks of interactions [35] . To determine how the late blight interacts with tomato proteins and affects cellular functions, the differentially accumulated proteins were analyzed through protein-protein interaction networks. The different interactions that occur among proteins are at the core of cell processing, and their systematic characterization helps to contextualize these proteins in the biological molecular system [28] . This fact can be verified in the Santa Clara genotype for proteins related to the photosynthetic process as well as between metabolic and energy proteins. In the protein interaction network of BGH-2127, interactions occurred among proteins related to the photosynthetic process and also between energy proteins and metabolism. Another interaction network occurred between the protein N-terminal domain rich in leucine repetition (spot 84) with the protein ribulose 3-phosphate epimerase (spot 53), which in turn interacted with ribulosebiphosphate carboxylase small chain proteins (spot 5) and ribolosyphosphate carboxylase small chain 3B (spot 275). The interactions of the ribulose 3-phosphette epimerase (spot 53) protein with distinct functional groups are of great interest since, in general, each “protein community” represents a distinct cellular process. Thus a protein that is a member of more than one “community” participates in several processes and can be considered to be one of the connections among them [70] . The connectivity of proteins observed in this biological network indicates the occurrence of important physiological changes in the functional regulation of the cellular mechanisms involved in the resistance of BGH-2127 to late blight.