Aluminum-Immobilizing Rhizobacteria Modulate Root Exudation and Nutrient Uptake and Increase Aluminum Tolerance of Pea Mutant E107 (brz)

It is well known that plant-growth-promoting rhizobacteria (PGPRs) increase the tolerance of plants to abiotic stresses; however, the counteraction of Al toxicity has received little attention. The effects of specially selected Al-tolerant and Al-immobilizing microorganisms were investigated using pea cultivar Sparkle and its Al-sensitive mutant E107 (brz). The strain Cupriavidus sp. D39 was the most-efficient in the growth promotion of hydroponically grown peas treated with 80 µM AlCl3, increasing the plant biomass of Sparkle by 20% and of E107 (brz) by two-times. This strain immobilized Al in the nutrient solution and decreased its concentration in E107 (brz) roots. The mutant showed upregulated exudation of organic acids, amino acids, and sugars in the absence or presence of Al as compared with Sparkle, and in most cases, the Al treatment stimulated exudation. Bacteria utilized root exudates and more actively colonized the root surface of E107 (brz). The exudation of tryptophan and the production of IAA by Cupriavidus sp. D39 in the root zone of the Al-treated mutant were observed. Aluminum disturbed the concentrations of nutrients in plants, but inoculation with Cupriavidus sp. D39 partially restored such negative effects. Thus, the E107 (brz) mutant is a useful tool for studying the mechanisms of plant–microbe interactions, and PGPR plays an important role in protecting plants against Al toxicity.


Introduction
Aluminum (Al) is a major negative factor in acid soils for the responsible growth inhibition of agricultural crops due to the disturbances of plant metabolism, the induction of oxidative stress, and a decrease in root development, nutrient uptake, and photosynthesis [1,2]. Plants counteract Al toxicity through increasing the rhizosphere pH, complexation by root exudates, production of mucilage and siderophores, detoxification within root tissues, and efflux from roots [3][4][5]. Pea (Pisum sativum L.), being a very important agricultural crop, is a relatively Al-sensitive species [6][7][8]. A number of negative effects of Al on pea have been described, namely: inhibition of seed germination [9] and root elongation [7,10,11], induction of oxidative stress [9][10][11][12], a decrease in root and/or shoot biomass [6,8,9,13], and diminution of chlorophyll and nitrogen content in leaves [9]. However, pea varieties significantly differ in Al tolerance; thus, more-tolerant or -sensitive genotypes have been identified [7,8,13,14] and used to understand the mechanisms of Al tolerance. In particular, the counteraction of Al-induced oxidative stress occurs due to the high activity of antioxidative enzymes and the

Plant Growth
To select the most-efficient strain, two preliminary hydroponic experiments were performed where all the plants were grown in the presence of 80 µM AlCl 3 only. All the studied strains were applied in the first experiment for the comparison of their effect on plant growth (Table 1). Strain Cupriavidus sp. D39 increased the root and shoot biomass of Sparkle and the root biomass of the E107 (brz) mutant ( Table 1). The root growth of Sparkle improved after inoculation with the yeast Rhodotorula sp. AL1. Strains Paraburkholderia sp. A35 and Paraburkholderia sediminicola C06 increased shoot biomass, whereas the effect of Herbaspirillum sp. E52 was negative. In the second experiment, three of the more efficient strains were compared (Table 1). A significant increase of the root and shoot biomass of both pea genotypes inoculated with Cupriavidus sp. D39 was observed. Inoculation with P. sediminicola C06 or Paraburkholderia sp. A35 increased the root biomass of Sparkle or E107 (brz), respectively. Based on these results, strain Cupriavidus sp. D39 was chosen for further experiments. All plants were treated with 80 µM AlCl 3 . Data are means ± SE. Asterisks show significant differences between uninoculated controls and the relative treatments for each experiment and pea genotype (least-significant difference test, p < 0.05, n =10).
Next, three experiments comparing plant growth response to both treatment with Al and inoculation with Cupriavidus sp. D39 were performed. The root and shoot biomass of Al-treated E107 (brz) decreased, whereas Sparkle was tolerant to this Al concentration ( Figure 1     Accumulation of Al by E107 (brz) was lower in the roots (Figure 3a), but similar in the shoots (Figure 3b), as compared to Sparkle. Inoculation decreased the total amount of Al in the solution by two-times (Figure 3c). Based on these results, the calculation showed that the total Al in the residue was greater in pots with E107 (brz) and increased with Cupriavidus sp. D39 (Figure 3d). Accumulation of Al by E107 (brz) was lower in the roots (Figure 3a), but similar in the shoots (Figure 3b), as compared to Sparkle. Inoculation decreased the total amount of Al in the solution by two-times (Figure 3c). Based on these results, the calculation showed that the total Al in the residue was greater in pots with E107 (brz) and increased with Cupriavidus sp. D39 (Figure 3d). Accumulation of Al by E107 (brz) was lower in the roots (Figure 3a), but similar in the shoots (Figure 3b), as compared to Sparkle. Inoculation decreased the total amount of Al in the solution by two-times (Figure 3c). Based on these results, the calculation showed that the total Al in the residue was greater in pots with E107 (brz) and increased with Cupriavidus sp. D39 (Figure 3d).  The mutant Al E107 (brz) acidified the nutrient solution, particularly when Al was supplemented, but Cupriavidus sp. D39 increased the pH of the Al-treated mutant plants ( Figure 2d).

Presence of Cupriavidus sp. D39 in Hydroponic System
The number of Cupriavidus sp. D39 in the nutrient solution with the E107 (brz) mutant was about 2.5-times higher than that of Sparkle and even increased in the presence of Al (Figure 4a). A tendency for a 49% increased bacteria number in the solution of Al-treated Sparkle (t = 2.7; p = 0.013; n = 12) as compared to the control plants was observed. The roots of both pea genotypes were actively colonized by Cupriavidus sp. D39, and the bacteria number increased by about four-times in the presence of Al (Figure 4b). The mutant Al E107 (brz) acidified the nutrient solution, particularly when Al was supplemented, but Cupriavidus sp. D39 increased the pH of the Al-treated mutant plants (Figure 2d).

Presence of Cupriavidus sp. D39 in Hydroponic System
The number of Cupriavidus sp. D39 in the nutrient solution with the E107 (brz) mutant was about 2.5-times higher than that of Sparkle and even increased in the presence of Al (Figure 4a). A tendency for a 49% increased bacteria number in the solution of Al-treated Sparkle (t = 2.7; p = 0.013; n = 12) as compared to the control plants was observed. The roots of both pea genotypes were actively colonized by Cupriavidus sp. D39, and the bacteria number increased by about four-times in the presence of Al (Figure 4b).
Plants 2023, 12, x FOR PEER REVIEW 6 of 22 The mutant Al E107 (brz) acidified the nutrient solution, particularly when Al was supplemented, but Cupriavidus sp. D39 increased the pH of the Al-treated mutant plants ( Figure 2d).

Presence of Cupriavidus sp. D39 in Hydroponic System
The number of Cupriavidus sp. D39 in the nutrient solution with the E107 (brz) mutant was about 2.5-times higher than that of Sparkle and even increased in the presence of Al (Figure 4a). A tendency for a 49% increased bacteria number in the solution of Al-treated Sparkle (t = 2.7; p = 0.013; n = 12) as compared to the control plants was observed. The roots of both pea genotypes were actively colonized by Cupriavidus sp. D39, and the bacteria number increased by about four-times in the presence of Al (Figure 4b).  The maximal colonization was observed on the Al-treated roots of E107 (brz). Single cells and small micro-colonies were observed on the root surface of both pea genotypes grown in the absence of Al in the solution (Figure 5a,b). Relatively large colonies were found on the Al-treated roots of Sparkle (Figure 5c), and particularly on the roots of E107 (brz) mutant (Figure 5d). (least-significant difference test, p < 0.05, n = 12). CFU stands for colony-forming units.
The maximal colonization was observed on the Al-treated roots of E107 (brz). Single cells and small micro-colonies were observed on the root surface of both pea genotypes grown in the absence of Al in the solution (Figure 5a,b). Relatively large colonies were found on the Al-treated roots of Sparkle (Figure 5c), and particularly on the roots of E107 (brz) mutant ( Figure 5d).

Root Exudation
Nine organic acids were detected in the root exudates of both pea genotypes ( Figure  6).

Root Exudation
Nine organic acids were detected in the root exudates of both pea genotypes ( Figure 6).   (Table S2). As a rule, genotypic differences were retained in the presence of Al in the nutrient solution ( Figure 8, Table  S2).  (Table S2). As a rule, genotypic differences were retained in the presence of Al in the nutrient solution ( Figure 8, Table S2).
The treatment of Sparkle with Al decreased the exudation of nine out of fifteen amino acids, as shown in Figure 8, and of alanine and aspartic acid (Table S2) (Table S2) (Table S2) increased with the Al treatment. In the absence of Al, inoculation with Cupriavidus sp. D39 decreased the amount of many amino acids, except threonine (Figure 8m), glutamic acid, and ornithine (Table S2)   The treatment of Sparkle with Al decreased the exudation of nine out of fifteen amino acids, as shown in Figure 8, and of alanine and aspartic acid (Table S2)   with Al decreased the total exudation of amino acids by Sparkle (Figure 7b). The inoculation with Cupriavidus sp. D39 decreased the amount of amino acids exuded by both genotypes in the absence of Al and of those exuded only by E107 (brz) in the presence of Al (Figure 7b).
Only E107 (brz) roots exuded glucose (Figure 9c), and the exudation of ribose ( Figure  9d) and sucrose (Figure 9e  The total amount of exuded sugars was about two-times greater for E107 (brz) as compared with Sparkle, whereas inoculation with Cupriavidus sp. D39 dramatically decreased the total amount of sugars in pea genotypes (Figure 7c).
Phytohormone IAA was detected in the solution where the inoculated E107 (brz) mutant was cultivated with and without the addition of Al (Figure 10a). A relatively low amount of IAA was also found in the solution of Al-treated Sparkle inoculated with Cu- The total amount of exuded sugars was about two-times greater for E107 (brz) as compared with Sparkle, whereas inoculation with Cupriavidus sp. D39 dramatically decreased the total amount of sugars in pea genotypes (Figure 7c).
Phytohormone IAA was detected in the solution where the inoculated E107 (brz) mutant was cultivated with and without the addition of Al (Figure 10a). A relatively low amount of IAA was also found in the solution of Al-treated Sparkle inoculated with Cupriavidus sp. D39 (Figure 10a). Mutant E107 (brz) exuded three-times more SA, and treatment with Al significantly inhibited the exudation of this phytohormone (Figure 10b). The inoculated treatments of both pea genotypes showed trace amounts of SA only.

Nutrient Uptake by Plants
The treatment with Al increased the concentration of Fe in E107 (brz) roots and the

Nutrient Uptake by Plants
The treatment with Al increased the concentration of Fe in E107 (brz) roots and the concentration of P in roots of both uninoculated pea genotypes (Table 2). Decreased K, Mg, and S concentrations were found in uninoculated Al-treated E107 (brz) roots, whereas the root Mg concentration of E107 (brz) also decreased in the presence of Cupriavidus sp. D39. The inoculation with Cupriavidus sp. D39 increased the root Mg and Mn concentrations in both pea genotypes and the K and S concentrations of the inoculated E107 (brz) ( Table 2). Control mutant roots had increased Mn, P, and Zn concentrations.
Data are means ± standard errors. Different lowercase letters show significant differences between treatments (least-significant difference test, p < 0.05, n = 3). DW stands for dry weight.
Al-treated plants had decreased concentrations of Ca and Mg in the shoots of the inoculated Sparkle and uninoculated E107 (brz), respectively (Table 3). Inoculation with Cupriavidus sp. D39 increased the Ca concentration in the shoots of both pea genotypes, as well as the Fe and Mg concentrations in the Al-treated Sparkle and E107 (brz), respectively (Table 3). Data are means ± standard errors. Different lowercase letters show significant differences between treatments (least-significant difference test, p < 0.05, n = 3). DW stands for dry weight.

Comparison of Al-Immobilizing Strains for Plant Growth Promotion
In the first stage of this research, five recently isolated Al-immobilizing strains [35] were compared for their ability to promote the growth of the pea Sparkle and the E107 (brz) mutant in the presence of a toxic Al concentration. The effect of bacteria and yeasts varied from slightly negative to positive depending on the strain and pea genotype ( Table 1). The maximum plant biomass gains were obtained after inoculations with Cupriavidus sp. D39 (Table 1), providing the reason for choosing this strain for further detailed study.
Probably, apart from the immobilization of Al in the root zone, these strains could affect plant growth by other mechanisms. All the tested bacterial strains produced IAA and utilized ACC as a nitrogen source, suggesting the presence of ACC deaminase activity [35]. It was previously suggested that these traits could play an important role in plant growth promotion under stress conditions caused by Al toxicity. In particular, Ps. fluorescens SPB2137 produced auxins, had ACC deaminase activity, and increased Al tolerance of pea [29]. Al-tolerant strain Viridibacillus arenosi IHBB7171 promoted the growth of pea plants, probably due to producing auxins and the ACC deaminase activity [26]. The increased biomass of ryegrass grown in a volcanic soil rich in Al was obtained after inoculation with IAA-producing and ACC-utilizing Klebsiella sp., Stenotrophomonas sp., Serratia sp., and Enterobacter sp. [40]. A decreased Al toxicity on the growth of mung bean inoculated with Bacillus megaterium CAM12 and Pantoea agglomerans CAH6 [41] or lettuce inoculated with Curtobacterium herbarum CAH5 [42] and Rhodotorula mucilaginosa CAM4 [43] has also been described. In addition, Cupriavidus sp. D39 can fix atmospheric nitrogen [35], having a set of fix genes in its genome [44].

Plant Biomass and Al Uptake
The pronounced decrease in the biomass of the Al-treated E107 (brz) mutant as compared to Sparkle was an expected result, since this mutant was initially characterized as a Al-sensitive genotype due its ability to acidify the rhizosphere and accumulate Al [38]. The root and shoot growth promotion caused by Cupriavidus sp. D39 on Al-treated plants was significantly higher for E107 (brz) (Figure 1). This result is in line with our previous finding, showing higher growth promotion caused by Ps. fluorescens SPB2137 on Al-sensitive pea varieties as compared to Al-tolerant varieties [29]. The effect observed in this study presents original indirect evidence of the bacterial contribution to plant growth promotion by reducing the availability of Al for inoculated plants, since the genetic backgrounds of Sparkle and E107 (brz) are the same. This was also confirmed by the decrease in the concentration of Al in the nutrient solution of both pea genotypes and in the roots of the Al-accumulating E107 mutant (brz) only (Figure 1), as well as by the increased Al accumulation in the residue (Figure 3d). The ability of Cupriavidus sp. D39 to alkalize the nutrient solution could facilitate the precipitation of Al and alleviate its toxic effect on the roots.
There have been several reports demonstrating reduced Al uptake by plants inoculated with PGPRs [33][34][35]. However, the mechanisms of such an effect have not been studied enough. A decreased Al concentration in the batch culture of C. herbarum CAH5 was associated with the accumulation of P in the residue, suggesting the formation of insoluble phosphates [42]. PGPR strain Burkholderia sp. increased the root length of maize and decreased Al accumulation in roots as a result of the binding of Al ions by soil phosphates [27]. The binding toxic Al with phosphates solubilized from soil by several PGPR strains has been pointed out as a possible mechanism for the growth promotion of ryegrass [40]. However, we found here that strain Cupriavidus sp. D39 did not solubilize phosphates in vitro and even decreased the water-soluble P concentration in soil, accompanied by its alkalization [35]. It has also been proposed that PGPRs produce siderophores, leading to the formation of Al 3+ -siderophore complexes, which are hardly available for plants [40]. This property is unlikely to apply to Cupriavidus sp. D39, since no siderophores were found in its culture fluid.
The presence of high Al concentrations in the growth medium of B. megaterium CAM12 and P. agglomerans CAH6 induced the production of exopolysaccharides, contributing to biofilm formation [41]. Strain Ps. fluorescens ATCC 13525 responded to Al toxicity by the biosynthesis of phospholipids, leading to the formation of residue containing Al [45,46]. The treatment with Al of Ps. fluorescens SPB2137 in batch culture induced the development of clumps considered as biofilm-like structures containing cells, exopolysaccharides, phospholipids, and Al phosphates [29]. The present report demonstrated the formation of characteristic micro-colonies of Cupriavidus sp. D39 on Al-treated pea roots (Figure 5c,d), which looked like the clumps previously observed in Ps. fluorescens SPB2137 [29]. We propose that such clumps could be involved in the protection of roots from Al toxicity by Cupriavidus sp. D39.
Another important observation was that Cupriavidus sp. D39, having a high Al tolerance, very actively colonized the pea roots, particularly in the presence of Al (Figure 4b). In contrast, pea root colonization by Ps. fluorescens SPB2137 was inhibited by the Al treatment, and the bacteria number on the roots was about 100-times less [29] as compared with that observed here for Cupriavidus sp. D39 (Figure 4b).

Root Exudation
For the first time, root exudation of organic compounds by the E107 (brz) mutant was described, and an increased exudation of organic acids, amino acids, and sugars as compared to the wild-type Sparkle was revealed. Moreover, the treatment with Al stimulated exudation, particularly by the mutant plants. The E107 (brz) mutant not only had an increased release of protons [38,47], but also an abnormal efflux of various lowmolecular-weight organic compounds. This makes the E107 (brz) mutant a unique plant for studying plant physiology and plant-microbe interactions, taking into account that these compounds play an important role in plant metabolism and nutrition, as well as serve as a major nutrient source for symbiotic microorganisms.
The observed upregulated efflux of various plant metabolites and photosynthates by the E107 (brz) roots could be responsible for its impaired growth reported earlier [38] and is consistent with our results (Figure 1). The increased exudation of organic acids by the roots of the E107 (brz) mutant did not help counteract the Al toxicity. It is known that the exuded organic acids are involved in the complexation of Al in the root zone, and this phenomenon is considered an important mechanism of Al tolerance in many plant species [4]. However, there was no correlation between the intensity of organic acid exudation and the growth response to Al toxicity of 11 pea genotypes [13]. The obtained results suggested that, although Al toxicity induced the exudation of organic acids by pea roots, their protective role in the Al tolerance of this species is ambiguous.
Previously, it was shown that treatment with SA reduced oxidative stress, the accumulation of Al in shoots, and stimulated the growth of Al-treated rice [48], alfalfa [49], soybean [50], and Cassia tora [51]. Treatment with Al increased the endogenous SA concentration in the roots of soybean [52] and induced the expression of genes related to SA biosynthesis in Vitis quinquangularis [53]. Here, the treatment with Al inhibited the exudation of SA by both pea genotypes, whereas Cupriavidus sp. D39 utilized it in vitro (Table S1) and most probably in the nutrient solution ( Figure 10). An intriguing question arises: Can the decrease in SA exudation under the action of Al be a protective response of plants due to its preservation inside root tissues and/or translocation into shoots? The results also suggested that it is difficult to explain the positive effects of Cupriavidus sp. D39 on pea growth and Al accumulation due to SA's modulation.
Aluminum affects biosynthesis and disturbs the balance of endogenous amino acids in plants [3,54,55]. The roots of wheat seedlings responded to Al toxicity by increased exudation of many proteinogenic amino acids [56]. The exudation of proteinogenic amino acids by four pea genotypes was also increased in Al-treated plants [29]. At the same time, a decreased exudation of asparagine, isoleucine, and glutamic acid was observed in Al-treated trifoliate orange [54]. In the present study, the Al treatment inhibited amino acid exudation by the cultivar Sparkle (Figures 7b and 8). The effect of Al on the E107 (brz) mutant varied from positive to negative depending on the amino acid (Figure 8), resulting in a similar exudation of the total amount of amino acids of both pea genotypes (Figure 7b). The results suggested high genotypic variability in amino acid exudation and disturbance of this trait in the E107 (brz) mutant.
The protective action of amino acids against Al toxicity has been repeatedly reported: arginine is involved in the biosynthesis of putrescine, leading to decreased Al retention in the cell walls of wheat [57]; cysteine increased the activity of glutathione [58]; GABA induced the antioxidant system in barley [59] and Arabidopsis thaliana [60]; glycine played a stress alleviating role as a component of glycine betaine in rice [61] and glycine-rich proteins in A. thaliana [62]; phenylalanine was involved in the biosynthesis of phenylpropanoids and flavonoids with defense functions in alfalfa [63,64]; proline protected root cells as an osmolyte [55,61,65]; threonine activated H + -ATPase in soybean [66]. Four of these amino acids (Arg, Cys, GABA, and Gly) were exuded more by the E107 (brz) mutant treated with Al (Figure 8a-d). Therefore, it is possible that the disturbance of amino acid exudation and the loss of these substances from the roots can negatively affect the adaptation of E107 (brz) to the toxic Al.
The decreased amount of amino acids in the nutrient solution inoculated with Cupriavidus sp. D39 was probably due to their utilization as a nitrogen or carbon source, since these compounds were nutrients for the bacteria in vitro (Table S1). It is important to pay attention to the increased exudation of tryptophan by E107 (brz), especially in the presence of Al, and its absence in the inoculated solution (Figure 8o). The increased exudation of tryptophan by pea roots treated with Al was observed previously [29], and the present results confirmed this observation. PGPRs use tryptophan for the biosynthesis of auxins, which induces the expression of genes involved in the plant's Al tolerance [67], increased root exudation of organic acids [68,69], and promoted the growth of Al-treated plants [70][71][72][73][74]. However, the overproduction of auxins due to Al toxicity in A. thaliana roots inhibited root elongation [75]. On the other hand, the treatment of alfalfa with Al decreased the IAA concentration in apical buds and root tips, resulting in root growth inhibition [73]. This means that the optimal auxin concentration is needed to counteract the toxic effects of Al on plant growth. It could be proposed that the auxin-producing Cupriavidus sp. D39 [35] could be used to explore the intensive exudation of tryptophan by the E107 (brz) mutant (Figure 8o) for IAA production (Figure 10a), leading to growth promotion ( Figure 1). Recently, we suggested such a mechanism for pea growth promotion by IAA-producing Ps. fluorescens SPB2137 [29]. The growth promotion of Al-treated maize seedlings by Ps. fluorescens 002 was proposed due to its auxin production [76]. Here, this hypothesis was confirmed by the disappearance of tryptophan and the presence of IAA in the nutrient solution of the inoculated E107 (brz) mutant.
The influence of Al stress on the exudation of sugars by plant roots has received little attention. The increased exudation of glucose by wheat [56] and fructose, glucose, and ribose by pea [29] was previously described for Al-treated seedlings. Intra-species variation in sugar exudation was also described for pea [29] and soybean [77]. Here, quantitative and qualitative differences in the exudation of sugars between the E107 (brz) mutant and its parental cultivar Sparkle in the presence and absence of toxic Al were revealed ( Figure 9). As with organic acids and amino acids, excessive exudation of photosynthates in the form of sugars by the E107 (brz) mutant could be negatively manifested in the adaptation to Al stress. On the other hand, this trait could support active root colonization by Cupriavidus sp. D39 (Figure 4b), since the strain used these substances as nutrients (Table S1) and decreased their amount in the solution (Figure 9), enhancing the plant-growth-promoting effects, particularly on Al-treated plants.

Nutrient Uptake by Plants
Mutant E107 (brz) possessed low tolerance, but increased accumulation of Fe [39] and Al [38] due to the acidification of the rhizosphere. The previously hydroponically cultivated E107 (brz) had elevated shoot Mn and Zn concentrations when growing in Fedeficient solution, but the supplement with 10 µM Fe-citrate further increased the root Fe and the shoot Fe, Mn, Zn, and Cu concentrations [39,47]. Experiments with vermiculite and soil cultures revealed the ability of the E107 (brz) mutant to actively accumulate Ca, Fe, K, Mg, and Mn in the leaves [47]. In this study, genotypic differences between Sparkle and the E107 (brz) mutant were also found in the root concentrations of Mn, P, and Zn (Table 2) and in the shoot concentration of Ca (Table 3)  This new information about the nutrient uptake by the E107 (brz) mutant is important for its detailed characterization.
The effects of Cupriavidus sp. D39 on nutrient uptake were directed towards increasing concentrations of elements in plants and were more pronounced in the Al-treated mutant plants (Tables 2 and 3). The latter observation was probably due to the large number of bacteria in this variant of the experiment. However, the effect of Al on the uptake of nutrients was, for the most part, negative. In particular, Cupriavidus sp. D39 eliminated the negative effects of Al on the concentrations of K, Mg, and S in the mutant roots and of Mg in the mutant shoots. It was reported that the toxic Al inhibits nutrient uptake in various crops [1,2] including pea [78], and the ability of counteracting Al-induced disturbances of nutrient uptake was shown as a principal mechanism for the Al tolerance of this plant species [13]. Moreover, the inoculation of pea with a microbial consortium containing PGPR Ps. fluorescens SPB2137 [79] improved the uptake of nutrients from Al-contaminated soil [79]. We propose that Cupriavidus sp. D39 contributes to the adaptation of plants to the Al toxicity via the enhancement of the nutrient uptake by the roots and that such a positive effect is more pronounced and more important for Al-sensitive plant genotypes.

Plants
Seed samples of the pea (Pisum sativum L.) cultivar Sparkle and its mutant E107 (brz) were kindly provided by Prof. Frederique C. Guinel. The seeds were propagated under the same soil and climate conditions in summer under natural light, air moisture, and temperature (experimental greenhouse of ARRIAM, Saint Petersburg, Russia).

Microorganisms
Strains of the Al-immobilizing bacteria Cupriavidus sp. D39, Paraburkholderia sp. A35, Paraburkholderia sediminicola C06, and Herbaspirillum sp. E52 and yeast Rhodotorula sp. AL1 [28] were obtained from the Russian Collection of Agricultural Microorganisms (RCAM) (St.-Petersburg, Russian Federation; http://www.arriam.ru/kollekciya-kul-tur1/; accessed on 7 July 2021) and maintained on potato dextrose agar (PDA) medium (HiMedia Laboratories Pvt. Ltd., Mumbai, India). The strain Cupriavidus sp. D39 was marked with the gene encoding green fluorescent protein (GFP) using the mini-transposon suicide delivery system pAG408 [80] as previously described [81]. The utilization of organic acids, amino acids, sugars, and indole-3-acetic acid (IAA) by Cupriavidus sp. D39 was assessed in vitro via the cultivation of bacteria in a mineral liquid medium containing these substances as a sole carbon or nitrogen source [77]. Siderophore production by Cupriavidus sp. D39 was determined using a chrome azurol S (CAS) shuttle solution, as described by Schwyn and Neilands [82]. The assay was calibrated by generating a standard curve for samples containing 1 to 100 µM deferoxamine mesylate. The capability of Cupriavidus sp. D39 to solubilize phosphates was tested in vitro by detecting the clearing zones developed around the bacterial colonies grown for 5 days on an agar medium supplemented with 500 mg L −1 of Ca 3 (PO 4 ) 2 , Ca-phytate, Al 3 PO 4 , or Fe 3 PO 4 [83].

Presence of Cupriavidus sp. D39 in Hydroponic System
At the end of experiments, the number of Cupriavidus sp. D39 in the nutrient solution was determined by the serial dilution method using PDA medium as described previously [29]. Samples of the roots were picked, and images were taken to estimate the presence of GFP-tagged Cupriavidus sp. D39 using a fluorescent microscope (Axio Imager A2, Carl Zeiss, Oberkochen, Germany).

Determination of Root Exudation
The nutrient solution was centrifuged for 15 min at 9000× g and 4 • C. The supernatant was vacuum filtered through 0.45 µm filters (Corning, Germany) and concentrated at 45 • C using a rotary vacuum evaporator (BUCHI R-200 (BUCHI, Switzerland)). The concentrate was analyzed for root exudates (organic acids, sugars, amino acids, IAA, and SA) using the UPLC system Waters ACQUITY H-Class (Waters, Milford, MA, USA) as described previously [77,84].

Elemental Analysis
Aliquots of the supernatants were taken for the determination of the Al concentration in nutrient solutions. The roots and shoots were dried at 50 • C, combined into one sample for each pot, and ground to a powder. All samples were digested in a mixture of concentrated HNO 3 and 38% H 2 O 2 at 70 • C using the digestion system DigiBlock (LabTech, Sorisole, Italy). The content of Al and the nutrient elements (Ca, Fe, K, Mg, Mn, P, S, and Zn) in the digested samples was determined using an inductively coupled plasma emission spectrometer (ICPE-9000, Shimadzu, Tokyo, Japan), according to the manufacturer's instructions. At the end of the experiments, a precipitate (residue) formed in the pots, which in previous studies with peas and PGPR was collected and analyzed and contained precipitated Al [36]. In this work, the amount of Al in this residue was calculated as the difference between the amount of Al introduced into the pot and its total amount in the plants and the solution.

Statistical Analysis
The statistical analysis of the data was performed using the software STATISTICA Version 10 (TIBCO Software Inc., Palo Alto, CA, USA). MANOVA analysis with Fisher's LSD test and Student's t test were used to evaluate differences between the means.

Conclusions
Al-tolerant and Al-immobilizing bacteria recently selected from the rhizosphere of pea grown in acid soils showed plant-growth-promoting activity for pea variety Sparkle and, particularly, for its Al-sensitive and Al-accumulating mutant, E107 (brz). The most-efficient strain, Cupriavidus sp. D39, increased the root and shoot biomass of the Al-treated E107 (brz) by about two-times. The growth-promoting effect was accompanied by the immobilization of Al and the decrease in the Al concentration in the roots. The E107 (brz) mutant showed upregulated exudation of many low-molecular-weight organic compounds (organic acids, amino acids, and sugars) in the absence or presence of a toxic Al concentration. In most cases, the treatment with Al increased exudation, particularly by the mutant plants. On the one hand, the abundant presence of exudates (potential nutrients for bacteria) in the root zone increased the proliferation, root colonization, and probably, metabolic activities of the bacteria, leading to a more intensive growth promotion of E107 (brz). This could be confirmed by the exudation of tryptophan and the formation of IAA by Cupriavidus sp. D39 in the root zone of the Al-treated mutant. On the other hand, we hypothesized that excessive loss of carbon, nitrogen, and probably, other elements originating from photosynthates and seeds could impair the metabolism and defense responses of E107 (brz) to Al stress. The treatment with Al disturbed the concentration of nutrients in the plants, particularly in the mutant roots, but inoculation with Cupriavidus sp. D39 partially restored such negative effects. Thus, a pair of cultivar Sparkle and its E107 (brz) mutant was a very suitable genetic model for studying the mechanisms of the bacterial effects on plants in the presence of Al toxicity. The Al-immobilizing PGPRs like Cupriavidus sp. D39 offer a promise for increasing the tolerance of plants to Al toxicity under acid soil conditions. Supplementary Materials: The following Supporting Information can be downloaded at: https: //www.mdpi.com/article/10.3390/plants12122334/s1, Table S1: Utilization of organic acids, amino acids, and sugars by Cupriavidus sp. D39 as a sole source of carbon or nitrogen in batch culture; Table S2: The amount of amino acids (µg g −1 root dry weight) exuded by roots of cultivar Sparkle and E107 (brz) mutant inoculated with Cupriavidus sp. DG39 and treated with 80 µM AlCl 3 .

Data Availability Statement:
The data presented in this study are available on request from the corresponding author. The data are not publicly available due to the rules of the publication activity of the ARRIAM.