Leaf Proteomic Analysis in Seedlings of Two Maize Landraces with Different Tolerance to Boron Toxicity

Boron (B) toxicity is an important stressor that negatively affects maize yield and the quality of the produce. The excessive B content in agricultural lands is a growing problem due to the increase in arid and semi-arid areas because of climate change. Recently, two Peruvian maize landraces, Sama and Pachía, were physiologically characterized based on their tolerance to B toxicity, the former being more tolerant to B excess than Pachía. However, many aspects regarding the molecular mechanisms of these two maize landraces against B toxicity are still unknown. In this study, a leaf proteomic analysis of Sama and Pachía was performed. Out of a total of 2793 proteins identified, only 303 proteins were differentially accumulated. Functional analysis indicated that many of these proteins are involved in transcription and translation processes, amino acid metabolism, photosynthesis, carbohydrate metabolism, protein degradation, and protein stabilization and folding. Compared to Sama, Pachía had a higher number of differentially expressed proteins related to protein degradation, and transcription and translation processes under B toxicity conditions, which might reflect the greater protein damage caused by B toxicity in Pachía. Our results suggest that the higher tolerance to B toxicity of Sama can be attributed to more stable photosynthesis, which can prevent damage caused by stromal over-reduction under this stress condition.


Introduction
Boron (B) is an essential element for plants, and is well known for its structural role in both cell walls and membranes [1][2][3][4][5]. In fact, B establishes diester bonds between the apiose residues of two rhamnogalacturonan-II (RGII) molecules, forming RGII-B complexes that stabilize the pectin network of the cell wall [6][7][8]. Furthermore, B contributes to the preservation of plasmalemma integrity and function [9], likely through the formation of B complexes with membrane components that contain cis-diol groups [10,11]. Thereby, B forms complexes with major constituents of membrane lipid rafts, such as glycosyl inositol phosphoryl ceramides (GIPCs) [12]. Moreover, B participates in the formation of GIPCs-B-RGII complexes, which connect the plasmalemma to the cell wall [13]. Besides these structural roles, B is also involved in plant development, participating in root and shoot elongation, pollen-tube growth, flowering, and fruiting [14][15][16]. In addition, B has been reported to participate in several physiological processes, such as photosynthesis; nucleic acid synthesis; phenolic, nitrogen and polyamine metabolism; protein stabilization and biosynthesis; and gene expression, among others [16][17][18][19][20][21][22].

Results
A total of 2793 proteins were identified in at least one of the biological replicates of a landrace (Sama or Pachía) and a B treatment analyzed (Table S1a). In addition, the number of proteins detected in both Pachía and Sama in each of the B treatments studied was similar at close to 1100 proteins (Table 1). Table S1a shows the dataset of the identified proteins, indicating their gene ontology (GO), biological processes (GOBP), GO molecular functions (GOMF), and GO cellular compartments (GOCC), and Table S1b summarizes the statistical analysis and fold changes in the proteins. To study the differentially accumulated proteins in Pachía and Sama in both B treatments, four comparison groups were established: (1) Sama and Pachía seedlings subjected to the control B condition (S0.05/P0.05), (2) Sama and Pachía treated with 10 mM B (S10/P10), (3) Sama subjected to 10 mM and 0.05 mM B (S10/S0.05), and (4) Pachía treated with 10 mM and 0.05 mM B (P10/P0.05). A total of 303 proteins had statistically Plants 2023, 12, 2322 3 of 20 significant differential expression (p ≤ 0.05) in the above groups (Table S2). The S0.05/P0.05 and S10/P10 groups contained those proteins that were differentially expressed between Sama and Pachía in 0.05 mM or 10 mM B, respectively. In media with 0.05 mM B, more proteins were up-and down-accumulated between Sama and Pachía than in 10 mM B ( Figure 1 and Table 1). In addition, the S10/S0.05 and P10/P0.05 comparison groups included proteins that were differentially expressed in response to B toxicity in Sama or Pachía, respectively. Pachía had a higher number of proteins induced and repressed by B toxicity than Sama; thus, 98 proteins were overexpressed in Pachía in 10 mM B, while only 38 were overexpressed in Sama, and 51 proteins were underexpressed in Pachía under B toxicity versus 28 in Sama ( Figure 1).

Classification into Several Functional Categories of Differentially Accumulated Proteins in Both Maize Landraces and B Treatments
All significant differentially expressed proteins in the four comparison groups described above were functionally classified into 26 categories using several databases (Table S2). The functional categories that included the largest number of differentially accumulated proteins were transcription and translation processes (57), photosynthesis (25), amino acid metabolism (24), protein degradation (23), carbohydrate metabolism (20), and protein stabilization and folding (18) (Figure 2 and Table S2). These main categories together contained more than 50% of the total differentially expressed proteins.

Classification into Several Functional Categories of Differentially Accumulated Proteins in both Maize Landraces and B Treatments
All significant differentially expressed proteins in the four comparison groups described above were functionally classified into 26 categories using several databases (Table  S2). The functional categories that included the largest number of differentially accumulated proteins were transcription and translation processes (57), photosynthesis (25), amino acid metabolism (24), protein degradation (23), carbohydrate metabolism (20), and protein stabilization and folding (18) (Figure 2 and Table S2). These main categories together contained more than 50% of the total differentially expressed proteins.  Figure 1 and Table S2. Seedlings of Sama and Pachía landraces were subjected to 0.05 and 10 mM B for 10 days. Results were obtained through the addition of the DAPs in the four comparisons. For more details, see Section 4. Aa: amino acid metabolism; CA: carbon assimilation and Calvin cycle; CM: carbohydrate metabolism; CDe: cell death; CDi: cell division; CW: cell wall; DNA: DNA and chromatin organization and DNA repair; LM: lipid metabolism; NM: nitrogen metabolism; NPM: nucleotide, purine, and pyrimidine metabolism; OT: others; ORP: oxidation and reduction processes; PLR: photosynthetic light reactions; PB: pigment biosynthesis; PD: protein degradation; PSF: protein stabilization and folding; ROS: reactive oxygen species scavenging pathways/response to oxidative stress; R: respiration metabolism (glycolysis, TCA cycle, and mitochondrial electron transfer); RB: ribosome biogenesis; RBP: RNA binding and processing; SM: secondary metabolism; SG: signaling; ST: stress; TT: transcription and translation processes; TP: transporters and transport processes; NWK: not well-known proteins.

Differentially Expressed Proteins in Sama and Pachía in Response to B Toxicity
Considering that the major aim of this work was to analyze the changes induced by B toxicity in protein expression in Pachía and Sama, we will now focus on the proteins that were differentially expressed due to B toxicity in these landraces. Thus, 66 and 149 proteins were differentially expressed in response to B toxicity in Sama and Pachía, respectively ( Table 1). The main functional categories containing the highest number of differentially expressed proteins under B toxicity in both Sama and Pachía were transcription and translation, photosynthesis, amino acid metabolism, protein degradation, protein stabilization and folding, and reactive oxygen species (ROS) (Figures 3 and 4). Interestingly, most of the proteins belonging to the transcription and translation category were induced in response to B toxicity in both Sama and Pachía, with the number of differentially induced proteins being remarkably higher in Pachía (Figures 3 and 4). However, almost all proteins included in the photosynthesis category were repressed in 10 mM B, with the number of down-accumulated proteins also being higher in Pachía than in Sama ( Figure  4 and Table S2). Regarding protein degradation, and protein stabilization and folding,  Figure 1 and Table S2. Seedlings of Sama and Pachía landraces were subjected to 0.05 and 10 mM B for 10 days. Results were obtained through the addition of the DAPs in the four comparisons. For more details, see Section 4. Aa: amino acid metabolism; CA: carbon assimilation and Calvin cycle; CM: carbohydrate metabolism; CDe: cell death; CDi: cell division; CW: cell wall; DNA: DNA and chromatin organization and DNA repair; LM: lipid metabolism; NM: nitrogen metabolism; NPM: nucleotide, purine, and pyrimidine metabolism; OT: others; ORP: oxidation and reduction processes; PLR: photosynthetic light reactions; PB: pigment biosynthesis; PD: protein degradation; PSF: protein stabilization and folding; ROS: reactive oxygen species scavenging pathways/response to oxidative stress; R: respiration metabolism (glycolysis, TCA cycle, and mitochondrial electron transfer); RB: ribosome biogenesis; RBP: RNA binding and processing; SM: secondary metabolism; SG: signaling; ST: stress; TT: transcription and translation processes; TP: transporters and transport processes; NWK: not well-known proteins.

Differentially Expressed Proteins in Sama and Pachía in Response to B Toxicity
Considering that the major aim of this work was to analyze the changes induced by B toxicity in protein expression in Pachía and Sama, we will now focus on the proteins that were differentially expressed due to B toxicity in these landraces. Thus, 66 and 149 proteins were differentially expressed in response to B toxicity in Sama and Pachía, respectively ( Table 1). The main functional categories containing the highest number of differentially expressed proteins under B toxicity in both Sama and Pachía were transcription and translation, photosynthesis, amino acid metabolism, protein degradation, protein stabilization and folding, and reactive oxygen species (ROS) (Figures 3 and 4). Interestingly, most of the proteins belonging to the transcription and translation category were induced in response to B toxicity in both Sama and Pachía, with the number of differentially induced proteins being remarkably higher in Pachía (Figures 3 and 4). However, almost all proteins included in the photosynthesis category were repressed in 10 mM B, with the number of down-accumulated proteins also being higher in Pachía than in Sama ( Figure 4 and Table S2). Regarding protein degradation, and protein stabilization and folding, most of the differentially expressed proteins in 10 mM B were found in Pachía, suggesting that B toxicity would alter the structure and folding of proteins in this landrace. In addition, many of the proteins in the ROS category were induced by B toxicity in both landraces ( Figure 4 and Table S2). Although the groups of carbon assimilation and metabolism, lipid metabolism, and respiration included a smaller number of proteins than those mentioned above, a larger number of differentially expressed proteins were found in Pachía under B toxicity ( Figure 4 and Table S2). Other interesting categories were cell death, cell division, cell wall, ribosome biogenesis, and RNA binding and processing, which, despite having a very small number of proteins regulated by B toxicity, had an interesting distribution in both landraces and B treatments. In fact, in the cell death and cell wall categories, only proteins whose expressions were induced by B toxicity were found in Pachía; however, the cell division, ribosome biogenesis, and RNA binding and processing categories also contained proteins with higher accumulation in 10 mM B in both landraces ( Figure 4 and Table S2).
Plants 2023, 12, x FOR PEER REVIEW 5 of 23 most of the differentially expressed proteins in 10 mM B were found in Pachía, suggesting that B toxicity would alter the structure and folding of proteins in this landrace. In addition, many of the proteins in the ROS category were induced by B toxicity in both landraces ( Figure 4 and Table S2). Although the groups of carbon assimilation and metabolism, lipid metabolism, and respiration included a smaller number of proteins than those mentioned above, a larger number of differentially expressed proteins were found in Pachía under B toxicity ( Figure 4 and Table S2). Other interesting categories were cell death, cell division, cell wall, ribosome biogenesis, and RNA binding and processing, which, despite having a very small number of proteins regulated by B toxicity, had an interesting distribution in both landraces and B treatments. In fact, in the cell death and cell wall categories, only proteins whose expressions were induced by B toxicity were found in Pachía; however, the cell division, ribosome biogenesis, and RNA binding and processing categories also contained proteins with higher accumulation in 10 mM B in both landraces ( Figure 4 and Table S2).   races in response to B toxicity, with the amino acid metabolism and photosynthesis categories having the highest number of proteins ( Table 2). All proteins of the amino acid metabolism group were up-accumulated under B toxicity conditions, with these inductions being slightly greater in Pachía than in Sama. Interestingly, however, all commonly expressed proteins from the photosynthesis category were repressed by B toxicity, with these repressions being remarkably higher in Pachía than in Sama (Table 2).    A total of 18 proteins were commonly expressed (repressed or induced) in both landraces in response to B toxicity, with the amino acid metabolism and photosynthesis categories having the highest number of proteins ( Table 2). All proteins of the amino acid metabolism group were up-accumulated under B toxicity conditions, with these inductions being slightly greater in Pachía than in Sama. Interestingly, however, all commonly expressed proteins from the photosynthesis category were repressed by B toxicity, with these repressions being remarkably higher in Pachía than in Sama (Table 2).
Tables S3 and S4 list the most strongly differentially expressed proteins that were upor down-regulated more than twofold by B toxicity in Pachía and Sama, respectively. In Pachía, 105 proteins had strong differential expression under B toxicity, while only 27 were found in Sama. Photosynthesis was the functional category containing the highest number of proteins whose expressions were strongly down-accumulated in response to B toxicity in both Pachía and Sama (Tables S3 and S4); however, interestingly, a minor number of both repressed and very strongly repressed (FC < 0.33) proteins were observed in Sama (Table 3,  Table 4, Tables S3 and S4). Different subunits of the NDH complex (NDHS, B1, B2, J, and H) were strongly repressed by B toxicity in Pachía but not in Sama (Tables S3 and S4). In addition, only in Pachía did we detect proteins related to protein degradation processes whose expressions were mainly induced by B toxicity, suggesting that enhanced damage would be induced by 10 mM B in Pachía proteins (Table 3 and Table S3). Furthermore, B toxicity markedly induced a larger number (15) of proteins in Pachía belonging to the transcription and translation category (Table S3). ATPase-coupled transmembrane transporter activity 1 Protein ID: protein identification (ID) number in the UniProt database; 2 Gene Name: name or ID number of the corresponding gene of the differentially expressed protein as searched in the Maize Genetics and Genomics Database (MaizeGDB; https://www.maizegdb.org/, accessed between 6 June 2022 and 24 January 2023). 3 Fold Change is expressed as the ratio of LFQ intensities (on a logarithmic scale) of proteins between 10 and 0.05 mM B treatments; Induced proteins are highlighted with light green rows and repressed proteins with light red rows. 4 p-value: statistical level (using Student's t-test) ≤ 0.05, at which differential protein expression was accepted as significant; 5 FCSA/FCPA, is the ratio between fold change of Sama and Pachía. 6 Function/Biological process: annotated biological functions or biological process based on different databases. Induced proteins are highlighted with light green rows and repressed proteins with light red rows. For more details, see Section 4. Results were obtained from 3-4 separate plants of each landrace.    Table 5 shows the proteins that were strongly up-or down-accumulated when the protein expressions of Sama were compared to those of Pachía in media with 10 mM B. Sama had remarkable up-accumulation of four proteins involved in photosynthesis (ZmPIFI and OEE2-1), chlorophyll biosynthesis (ChlH1), and secondary metabolism (PAO1) ( Table 5), with this last protein also being very strongly induced in response to B toxicity (Table 4). However, in Pachía, several proteins were detected to exhibit strong accumulation in 10 mM B when compared with Sama (shown in Table 5 as strongly down-accumulated proteins in Sama) highlighting, among them, histone H1 and ribosomal protein S7, which were very strongly induced by B toxicity (Table 3).  Finally, in both Pachía and Sama, proteins exclusively detected in one of these landraces were found; among them were Nfc103a and eIF3a, which were only identified in Pachía in 10 mM B (Table 6).

Discussion
The results of this work were obtained with seedlings of two maize landraces, Sama and Pachía, since at this stage of development, the plants are more sensitive to B excess. Increasing work is being performed to elucidate the molecular mechanisms that allow crop varieties to be tolerant to B toxicity [25,27,[36][37][38]. The identification of proteins that contribute to B tolerance could be a useful tool to breed higher-yielding and higher-quality maize varieties.
Although 2793 proteins were detected in this proteomic analysis, only 303 proteins were differentially accumulated (Tables S1a and S2), which were classified into 26 functional categories. Functional analysis indicated that pathways involved in the transcription and translation processes, amino acid metabolism, photosynthesis, carbohydrate metabolism, protein degradation, and protein stabilization and folding were highly enriched categories in both landraces ( Figure 2). Remarkably, the expression levels of proteins related to these enriched processes were significantly different between Pachía and Sama.

Several Proteases and Translation-Related Proteins Allow Pachía to Survive in Media with B Excess
Pachía is a B-sensitive maize cultivar described by Mamani-Huarcaya et al. [27]. Interestingly, the highest number of differentially accumulated proteins (DAPs) was found in the comparison group P10/P0.05 ( Figure 1) suggesting that the B toxicity damage caused in Pachía could be partially relieved by these proteins. A remarkable number of these DAPs included in the categories of protein degradation (11), and transcription and translation (15), were strongly overexpressed in Pachía (Table S3). However, only five proteins of the transcription and translation group were markedly induced by 10 mM B in Sama (Table S4). The B-sensitive Citrus grandis had a higher number of proteins involved in protein degradation and was also overexpressed under B toxicity conditions in comparison with B-tolerant Citrus sinensis [39]. These authors concluded that B toxicity caused greater protein damage and proteolysis in C. grandis. Therefore, the high number of protein degradation-related proteins that were overexpressed in Pachía in 10 mM B would suggest that B toxicity would cause greater damage in Pachía proteins than in those of Sama, leading to increased proteolysis in B-sensitive Pachía. Proteins related to protein degradation that were strongly overexpressed in Pachía included, among others, cysteine protease14 and four serine proteases (Table S3). Proteases have been implicated in plant acclimation to abiotic stress, playing a major role in the degradation of damaged and misfolded proteins, thus contributing to cell survival. In fact, cysteine and serine proteases are involved in the degradation of misfolded proteins and protection against abiotic stresses [40][41][42][43]. Hence, these five proteases could have a main role in the degradation of damaged and misfolded proteins in Pachía under excess B, contributing to maintaining the correct conformation of Pachía proteins and, therefore, to the survival of this landrace under this stressful condition. In addition, a noteworthy number of proteins involved in transcription and translation processes were overexpressed at 10 mM B in Pachía (30, in contrast to only 9 in Sama) (Table S2). Proteomic analysis performed with dehydration, salt, and temperature stresses in cereals also displayed alterations in the levels of translation-related proteins, such as initiation factors and the ribosome constituent proteins ( [44] and references therein). Furthermore, it has been suggested that a B excess induces the inhibition of RNA-dependent processes, such as transcription and translation, owing to the ability of B to form complexes with ribose molecules [45]. In this regard, Tanaka et al. [46] have suggested that B or boric acid acts on the translation machinery, likely forming complexes with cis-diol groups of rRNA and tRNA. In addition, it has recently been proposed that high-B stress enhances ribosome frequency on stop codons, leading to global ribosome stalling [47]. Consequently, the high content of leaf-soluble B in Pachía seedlings subjected to 10 mM B, reported by Mamani-Huarcaya et al. [27], would generate an increased formation of B complexes with cis-diol groups of RNA, which would damage ribosomes, leading to a drop in protein synthesis, likely through global ribosome stalling. The strong overexpression of several ribosomal proteins would maintain the Pachía ribosome stability in B toxicity (Table S3). These results are consistent with those reported for rice, where several ribosomal protein large subunit genes were upregulated under temperature stress; this suggests that their encoded proteins might be involved in stress amelioration, likely maintaining the proper functioning of ribosomes [44]. Interestingly, the eukaryotic translation initiation factor 3 subunit A (eIF3a) was exclusively detected in B toxicity in Pachía (Table 6). These factors are one of the most significant components involved in plant protein synthesis and, specifically, rice eIF3A has been proposed to play an important role in different stresses [48]. Therefore, eIF3a would also help to alleviate the drop in protein synthesis in Pachía. Thereby, Pachía would partly ameliorate injuries caused by B toxicity in protein synthesis and ribosomes by overexpressing a high number of transcription-and translation-related proteins, abolishing the non-viable reduction in transcription and translation processes.

Proteins That Can Confer More B Toxicity Tolerance to Sama
Polyamine oxidase 1 (PAO1) is an interesting protein that was clearly up-accumulated in Sama when compared to Pachía at 10 mM B, and was also very strongly induced in Sama by B toxicity (Tables 4 and 5). This enzyme catalyzes the back-conversion of spermine (Spm) to spermidine (Spd), and Spd to putrescine (Put) [49]. Maize polyamines play a crucial role in the abiotic stress response [33]. In fact, it has been reported that Put protects the plant's photosynthetic apparatus against several abiotic stresses [50]. Moreover, the conjugation of Put to PSII proteins may lead to the structural and functional stability of PSII [49,51]. Therefore, the over-accumulation of PAO1 in Sama plants subjected to B toxicity would generate an increase in Put levels that would protect their photosynthetic apparatus, resulting in the higher P N observed in Sama under this stress, as described by Mamani-Huarcaya et al. [27]. This finding is consistent with the results reported for Karoon, a drought-tolerant maize cultivar. Pakdel et al. [49] proposed that higher expression of PAO genes and enzymatic polyamine oxidation activity can protect the photosynthetic apparatus of Karoon under water stress.

Lower Repression of Photosynthesis-Related Proteins Can Enhance the B Toxicity Tolerance of Sama
Photosynthesis is one of the essential physiological processes affected by B toxicity [2,22]. Photosynthetic efficiency can be achieved in Sama under B toxicity conditions by increasing the synthesis of photosynthetic pigments, since chlorophyll content is a major limiting component of photosynthetic efficiency [52]. Interestingly, Sama had a strong over-accumulation of magnesium-chelatase subunit H1 chloroplastic (ChlH1) at 10 mM B in comparison with Pachía (Table 5). ChlH binds to porphyrin and catalyzes the insertion of Mg 2+ into protoporphyrin IX [53]. Accordingly, the over-accumulation of ChlH1 in Sama would explain its higher content of chlorophyll a in B toxicity and the higher P N described by Mamani-Huarcaya et al. [27].
In this study, 25 proteins related to photosynthetic light reactions were differentially accumulated, with most of them involved in electron transport, light harvesting, and oxygen-evolving processes ( Figure 2 and Table S2). Pachía and Sama presented several photosynthesis-related proteins that were repressed by B toxicity when their expressions were compared with those of Pachía and Sama, respectively, in media with 0.05 mM B (Table S2). However, the number of these DAPs was lower in Sama than in Pachía (11 versus 16, respectively; Table S2); moreover, those proteins commonly downaccumulated in both landraces had a weaker decrease in Sama (Table 2). In addition, only two photosynthetic proteins were very strongly underexpressed (3-fold or more (corresponding to FC ≤ 0.33)) by B toxicity in Sama in contrast to ten proteins found in Pachía (Tables 3 and 4). This decreased accumulation of photosynthesis related-proteins may cause lower photosynthetic performance in B toxicity-treated Pachía plants than in Sama plants, as described by Mamani-Huarcaya et al. [27]. Therefore, Sama can retain sufficient levels of photosynthesis-related proteins in 10 mM B, which allows it to maintain photo-synthetic parameters at similar levels to those of the control conditions, as reported by Mamani-Huarcaya et al. [27]. Furthermore, three photosynthesis-related proteins were upaccumulated in Sama when their expressions were compared with those of Pachía in 10 mM B, namely, oxygen-evolving enhancer protein 2-1 chloroplastic (OEE2-1), post-illumination chlorophyll fluorescence increase (ZmPIFI), and NAD(P)H-quinone oxidoreductase subunit S chloroplastic (NDHS) ( Table 5 and Table S2). OEE2-1 is likely an extrinsic protein of the oxygen-evolving complex (OEC) (UniProt; https://www.uniprot.org/, accessed between 6 June 2022 and 24 January 2023). The OEC is stabilized and protected by extrinsic polypeptides [54]. The strong OEE2-1 over-accumulation in 10 mM B in Sama could facilitate the stability and protection of the OEC, leading to the higher photosynthetic electron transporter rate (ETR) observed in this landrace [27]. Regarding ZmPIFI, it is homologous to the PIFI protein of Arabidopsis thaliana (AtPIFI), an essential component of the NAD(P)H dehydrogenase (NDH) complex involved in chlororespiratory electron transport around PSI [55]. The Atpifi mutant had lower nonphotochemical quenching (NPQ) than the wild type under high light irradiances, suggesting that AtPIFI can protect plants from photooxidative stress triggered by excessive light [55]. Consequently, both ZmPIFI over-accumulation and the higher NPQ values that Sama showed in 10 mM B, unlike those from Pachía (Table 5; [27]), suggest that ZmPIFI is also a component of the maize NDH complex, playing a role in the oxidative photoprotection of this landrace under B toxicity conditions. Furthermore, unlike Sama, several subunits of the NDH complex were markedly repressed in Pachía by B toxicity (Tables S2-S4). The NDH complex mediates cyclic electron transport around PSI, playing a crucial role in C 4 photosynthesis [56,57]. NDH-mediated cycle electron transport (NDH-CET) performs two functions: (1) maintaining photosynthetic redox balance in electron transfer, avoiding stromal overreduction and functioning as a safety valve for excess electrons under stress, and (2) supplying ATP for efficient carbon assimilation, especially under stressful conditions [56][57][58][59]. The finding that none of the above components of the NDH complex was significantly repressed by B toxicity in Sama suggests that its NDH-CET can prevent stromal overreduction and protect against photooxidation. This fact would explain the high values of net photosynthetic CO 2 assimilation (P N ), maximum photochemical efficiency (Fv /Fm ), and quantum yield efficiency of PSII electron transport (Φ PSII ) reported in Sama at 10 mM B, which were similar to those of the control conditions [27]. Consistent with our data, Zhu et al. [59] have suggested that an increased abundance of NDH subunits in salt-stressed wheat would enhance NDH-CET, alleviating the accumulation of excess electrons and maintaining energy homeostasis. Moreover, the subunit S of the NDH complex was over-accumulated in Sama under B toxicity when compared to that in Pachía, likely leading to a higher amount of NDH complex, which would provide extra ATP to achieve better P N and growth in this landrace in medium with 10 mM B as, in fact, was observed by Mamani-Huarcaya et al. [27]. In addition, a higher supply of ATP was obtained in Sama in comparison to Pachía under B toxicity due to a weaker decrease in the αand β-chloroplastic subunits of ATP synthase in Sama (Table 2). Although B excess causes photosynthetic damage [2,22], plants have evolved mechanisms to repair these injuries that require a high amount of ATP from chloroplastic ATP synthase [60,61]. In Sama, B toxicity barely affected the photosynthetic parameters [27]. This finding indicates that this landrace possesses mechanisms to repair its photosynthetic machinery. It is likely that one of these mechanisms is to provide greater ATP availability, which can be achieved by maintaining sufficient levels of NDH and ATP synthase complexes to synthesize the amounts of ATP needed to repair its photosynthetic machinery and, therefore, to maintain its photosynthetic values at levels similar to those of control conditions. Our data provide valuable information for future research to breed improved maize varieties against this abiotic stress. Nevertheless, it is known that results obtained under laboratory conditions are not always applicable to field conditions. Therefore, it would be interesting to evaluate the productivity of Sama and Pachía in soils with B excess. However, the strong evidence of higher photosynthesis-related protein content in Sama (Table 2 and Table S2), which is supported by the higher values of photosynthetic parameters [27], suggests that field results may be very similar to those obtained under laboratory conditions in this work.

Plant Materials and Growth Conditions
Sama and Pachía, two Peruvian maize landraces from the Sama valley and the Pachía district (to the east of Tacna), were used in this study. Seeds were surface-sterilized as described by Mamani-Huarcaya et al. [27]. Afterwards, the seeds were placed in seedbeds filled with a perlite/vermiculite mixture (1/1, v/v) and watered with deionized H 2 O. After seven days, seedlings were transplanted to 30 L plastic containers with a nutrient solution (NS) that was identical to the one used by Mamani-Huarcaya et al. [27]. After two days of acclimation to hydroponic medium, the seedlings were divided into groups and transferred to fresh NS supplemented with 10 mM H 3 BO 3 (B toxicity conditions) or 0.05 mM H 3 BO 3 (control conditions). This medium was aerated by air pumps and renewed twice a week. The seedlings were germinated and grown hydroponically in a growth chamber under a 12 h light/12 h dark regime (215 µmol m -2 s -1 of photosynthetically active radiation at plant height), at 22 • C and 50% relative humidity. The plants were randomly harvested 10 days after the onset of the B treatments, and their leaves were quickly separated with a scalpel, frozen in liquid nitrogen and stored at -80 • C until further analysis.
The cleaning of maize protein extract, protein digestion, and mass spectrometry determinations were carried out at the Proteomics Facility for Research Support Central Service (SCAI) of the University of Córdoba (Spain) as follows.
Biological quadruplicate samples were separated and cleaned as described. Leaf protein extracts (50 µg of BSA protein equivalents per sample) were electrophoretically pre-concentrated in a centimeter band of 10% (w/v) SDS-PAGE gel. Protein bands were excised from the gels and, afterwards, the gel pieces were distained in 200 mM ammonium bicarbonate/50% acetonitrile for 15 min, followed by 5 min in 100% acetonitrile. Proteins were reduced by adding 20 mM dithiothreitol in 25 mM ammonium bicarbonate and incubated for 20 min at 55 • C. The mixture was cooled to room temperature, and then, free thiols were alkylated by adding 40 mM iodoacetamide in 25 mM ammonium bicarbonate for 20 min in the dark. Finally, the gel pieces were washed twice in 25 mM ammonium bicarbonate.
Proteolytic digestion was performed by adding trypsin to a final concentration of 12.5 ng/µL in 25 mM ammonium bicarbonate at 37 • C overnight. Protein digestion was stopped by adding trifluoroacetic acid at a final concentration of 1% (v/v). Finally, the digested samples were vacuum-dried and dissolved in a mixture of 2% (v/v) acetonitrile and 0.05% (v/v) trifluoroacetic acid.

Shotgun-DDA-LC-MS/MS Analysis
Peptide separations were performed on a nano-LC using Dionex Ultimate 3000 nano UPLC (Thermo Scientific, San Jose, CA, USA), equipped with a C18 75 µm × 50 cm Acclaim Pepmap column (Thermo Scientific, San Jose, CA, USA), at 40 • C, at a flow rate of 300 nL/min. Peptide mixtures were previously concentrated and cleaned on a 300 µm × 5 mm Acclaim Pepmap precolumn (Thermo Scientific, San Jose, CA, USA) using 2% acetonitrile/0.05% trifluoroacetic acid, at 5 µL/min, for 5 min. Peptides were eluted with a gradient of 60 min ranging from 96% solvent A (0.1% formic acid) to 90% solvent B (80% acetonitrile and 0.1% formic acid), followed by an 8 min wash at 90% solvent B and a 12 min reequilibration at 4% solvent B. Eluted peptides were converted into gas-phase ions via nanoelectrospray ionization and analyzed on a Thermo Orbitrap Fusion mass spectrometer (Thermo Scientific, San Jose, CA, USA) operated in positive mode. Survey scans of peptide precursors were acquired over an m/z range 400-1500 at 120K resolution (at 200 m/z) with a 4 × 10 5 ion count target. Tandem MS was performed via isolation at 1.2 Da with the quadrupole. Monoisotopic precursor ions were fragmented via CID (Chemically Induced Dimerization) in an ion trap, which was set up as follows: automatic gain control, 2 × 10 3 ; maximum injection time, 50 ms; and normalized collision energy, 35%. Only those precursors with charge states of 2-5 were sampled for MS2. A dynamic exclusion time of 15 s and a tolerance of 10 ppm around the selected precursor and its isotopes were used to avoid redundant fragmentations. The instrument was run in top 30 mode with 3 s cycles, meaning the instrument would continuously perform MS2 events until a maximum of 30 non-excluded precursors or 3 s, whichever was shorter.

Protein Quantification
Charge state deconvolution and deisotoping were not performed. MS2 spectra were searched using MaxQuant software v. 1.5.7.4 [64]. MS2 spectra were searched with Andromeda engines against a database of Uniprot Zea mays_Jun19. Peptides generated from tryptic digestion were searched by employing the following parameters: up to one missed cleavage, the carbamidomethylation of cysteines as fixed modifications, and the oxidation of methionine as variable modifications. The precursor mass tolerance was 10 ppm and product ions were searched at 0.6 Da tolerances. A target-decoy search strategy was applied, which integrates multiple peptide parameters such as length, charge, number of modifications, and identification score into a single quality that acts as statistical evidence of the quality of each single peptide spectrum match. The identified peptides were grouped into proteins according to the law of parsimony and filtered to a 1% false discovery rate (FDR). Peptide quantification was carried out using MaxQuant software, via a MaxLFQ label-free quantification method [65]. In the MaxLFQ label-free quantification method, a retention time alignment and identification transfer protocol ("match-between runs" feature inMaxQuant) was applied. Proteins identified from only one peptide were not taken into account in this analysis. Peak intensities across the whole set of quantitative data for all peptides in the samples were imported from the LFQ intensities of proteins from the MaxQuant analysis and normalized according to Cox et al. [65]. LFQ-normalized intensity values were transformed to a logarithmic scale with a base of two. Protein quantification and the calculation of statistical significance were carried out using a Student's t-test and error correction (p-value ≤ 0.05). The criteria used to consider a protein as differentially expressed were as follows: (a) the protein was consistently present in at least three biological replicates per condition; (b) it had statistically significant differences (Student's t-test, p ≤ 0.05) between genotypes or B treatments; and (c) it had a fold change ≥1.5 or ≤0.66667. The differentially accumulated proteins were manually categorized by function using different databases (Uniprot, https://www.uniprot.org/; Maize Genetics and Genomics, https://www.maizegdb.org/; ExplorEnz, https://www.enzyme-database.org/; BRENDA, https://www.brenda-enzymes.org/; KEGG: Kyoto Encyclopedia of Genes and Genomes, https://www.genome.jp/kegg/; and PANTHER: Protein ANalysis THrough Evolutionary Relationships, http://pantherdb.org/, accessed between 6 June 2022 and 24 January 2023).

Conclusions
The higher B content in Pachía leaves than in Sama leaves can cause greater damage to their proteins. The overexpression of several proteases, mainly cysteine protease14 and four serine proteases, can increase the degradation of damaged and misfolded proteins in Pachía. Subsequently, the over-accumulation of transcription-and translation-related proteins allows Pachía to: (1) partially replace proteins damaged by B toxicity, and (2) reduce the injury caused to ribosomes and protein synthesis by B excess by abolishing the non-viable decrease in transcription and translation processes, thus allowing Pachía to survive under this stress condition.
In Sama, PAO1 over-accumulation can protect the photosynthetic apparatus. Furthermore, ZmPIFI and NDHS up-accumulation, along with the lower knockdown of several subunits of NDH and ATP synthase complexes under B excess, confers greater B toxicity tolerance to this landrace by: (1) acting as an electron safety valve that prevents stromal overreduction, and thus, decreases photosynthetic damage, and (2) providing an additional supply of ATP that contributes to repairing the photosynthetic system of Sama. On the other hand, OEE2-1 overexpression can stabilize and protect the OEC, leading to a higher photosynthesis rate in this landrace.
Supplementary Materials: The following supporting information can be downloaded at https: //www.mdpi.com/article/10.3390/plants12122322/s1, Table S1a: Dataset of proteins identified via shotgun-DDA analysis; Table S1b: Dataset, statistical analysis, and fold changes of proteins identified via shotgun-DDA analysis; Table S2: Fold change ratios, p-values, and statistical significance of all significantly accumulated proteins, classified by functional category. Table S3: Proteins with higher differential expression in Pachía leaves in response to boron (B) toxicity. This table shows the proteins that were strongly induced or repressed by B toxicity in Pachía by comparing their expressions with those of Pachía in medium with 0.05 mM B. Table S4: Proteins with higher differential expression in Sama leaves in response to boron (B) toxicity. This table shows the proteins that were strongly induced or repressed by B toxicity in Sama by comparing their expressions with those of Sama in medium with 0.05 mM B.