Description of the experiment

The work was carried out during the 2018 autumn-winter agricultural cycle in northern Sinaloa, Mexico (25° 45’ 49” north latitude, 108° 51’ 41” west longitude). The soil of the region has a clayey-loamy texture (50% clay, 30% silt and 20% sand), low organic matter content (<1%), bulk density of 1.15 g cm^-3 and a volumetric moisture content of 0.155 cm³ cm^-3. The sowing and tillage techniques used were those recommended by the technical guide of INIFAP (^{SIAP, 2013}). Before fertilization, a soil sampling was carried out at a depth of 30 cm to determine physical and chemical characteristics. The sowing was carried out in wet soil and pest management was successfully controlled throughout the cycle.

Treatments

The treatments consisted of four doses of phosphorus (P₂O₅) (T1: 0 kg ha^-1, T2: 25 kg ha^-1, T3: 50 kg ha^-1, T4: 100 kg ha^-1) with and without Bacillus subtilis. The inoculation was performed by impregnating 5 ml (2x10⁸ CFU ml^-1) of the Bacillus subtilis Q11^® strain per each kilogram of Azufrado Higuera seed. The dose of nitrogen (150 kg ha^-1) and phosphorus was applied at pre-sowing using the Blaukorn^® Classic source (12-8-16). For irrigation programming, the Irrimodel platform (^{Sifuentes, 2012}) was used, which estimates the levels of soil moisture abatement in the root zone. Additionally, an irrigation criterion (IC) of 50% of the usable moisture was established and the moisture content was verified using TDR 300 (Time Domain Reflectometry).

Experimental design

The experiment consisted of plots divided into randomized complete blocks with three repetitions. The main plot consisted of the doses of phosphorus and its dimension was 64 m², the sub-plots consisted of the inoculation of Bacillus subtilis Q11 and the control whose dimension was 32 m² (four furrows arranged linearly within the main plot).

Micronutrient absorption

To estimate the nutrient absorption of the crop, vegetative analyses (destructive samplings) were carried out in different phenological stages (third trifoliate leaf, flowering, pod filling and physiological maturity). The sampling method consisted of the removal of the aerial biomass in a location (1 m) in each plot. Subsequently, the samples were taken to the laboratory where all the structures were separated (leaves, stems, pods) and subjected to drying (70 °C) in a forced air furnace for 72 h until reaching constant weight.

The concentration of micronutrients was determined with the filtered material that was obtained from wet digestion for phosphorus and then aliquots were placed for reading in a spectrophotometer following the protocol described in the Official Mexican Standard (^{NOM-021-RECNAT-2000}). The absorption of micronutrients was estimated by multiplying the dry weight by its concentration (kg ha^-1) (equation 1).

1).

DM (kg ha^-1) represents dry matter in different phenological stages. The micronutrients in the seed were estimated with the dry weight of the grain multiplied by its concentration as indicated in equation 2.

2).

Data analysis

Data on nutritional absorption and seed concentration were subject to an appropriate analysis of variance (Anova) (^{Minitab, 2017}). The mean difference in each treatment was performed using the Fisher’s LSD test (p≤ 0.05).

Micronutrient absorption

No interaction effect was found between the evaluated factors, therefore all results are presented separately. The absorption pattern showed that all micronutrients gradually increased from flowering, absorption was maximized in pod filling and decreased at the end of the cycle.

^{Westermann et al. (2011}) mentioned that the highest absorption occurred in a production system without input restrictions, which coincides with the results of this work, where plants under the doses of 50 and 100 kg of P exhibited the highest absorption range. The preference in total absorption was in the following order: Fe> Mn> Zn> B> Cu with 16, 4.3, 2.8, 1.2 and 0.54 kg ha^-1. These data are consistent with what was reported by ^{Chaudhary et al. (2008}) and ^{Bender et al. (2015}), who mention the effect of P and the absorption of micronutrients in soybean (Glycine max L.). Iron absorption increased in response to P fertilization throughout the crop cycle (Figure 1A).

Figure 1 Absorption of micronutrients in phenological stages in bean because of the dose of phosphorus. 

This absorption process was higher in the treatments under the doses of 50 kg and 100 kg of P (10.5 and 9.8 kg ha^-1) respectively. Works by ^{Tofiño-Rivera et al. (2016}) mention that fertilization with P could increase the concentration of Fe, but not that of Zn in plant tissues or grains. It is worth mentioning that in many cases immobilization processes rather than scarcity in the soil affect the supply of Fe for plants. Inoculated plants increased 12.5% in the absorption of Fe over non-inoculated plants (Figure 2A).

Figure 2 Absorption of micronutrients in phenological stages in bean due to the effect of Bacillus subtilis. 

In that sense, ^{Dehner et al. (2010}) showed that inoculation with rhizobacteria increased the formation of siderophores (low molecular weight chelators) thus increasing the absorption of Fe in different cultures, since these act as solubilizers in organic and mineral compounds under limiting conditions of Fe (^{Indiragandhi et al., 2008}).

It is also mentioned that siderophores not only complex Fe, but also form stable complexes with other metals such as Al, Cd, Cu, Zn and Pb (^{Neubauer et al., 2000}). The absorption of manganese (Mn) was also increased by the supply of P during the crop cycle.

Plants under doses of 50 kg of P had higher absorption (2.3 kg ha^-1), while plants under doses of 100 kg of P decreased this absorption at maturity (1.7 kg ha^-1) (Figure 1B). As seen in the absorption pattern, Cu extraction was strongly influenced by P doses and increased consistently throughout the cycle. Likewise, the plants treated with the doses of 50 and 100 kg of P had the highest absorption range (0.38 kg ha^-1), which declined in the stage of maturity (0.16 kg ha^-1) (Figure 1C). The inoculation with Bs increased Cu absorption by 20% (Figure 2C).

The total absorption of Zn was lower than that of Mn (0.7 kg ha^-1) (Figure 1D). According to ^{Domínguez-Vivancos (1997}), this process occurs due to high levels of P, which lead to physiological inhibition from the roots to the aerial parts. Findings reported by ^{Astudillo and Blair (2008}); ^{Joshi et al. (2010}); ^{Velu et al. (2014}) mention that when the content of P in the plant increases, that of Zn decreases. Inoculated treatments increased 5% in the absorption of Zn compared to non-inoculated treatments (Figure 2D). The total absorption of boron (B) was 0.3 kg ha^-1, which was slightly lower in plants under doses of 100 kg of P.

It is mentioned that the boron absorption process is reduced in soils with good drainage and low moisture, as well as dry climate conditions (^{Havlin et al., 2005}). In addition, the highest absorption (0.7 kg ha^-1) occurred in plants with high P content in the pod filling stage (Figure 1E). On the other hand, the treatments inoculated with Bs substantially increased the absorption process compared to non-inoculated plants (Figure 2). Inoculated plants increased the absorption of Mn, Cu and B by 12, 25 and 35% compared to non-inoculated plants (Figure 2B, C and E).

Research conducted by ^{Armada et al. (2014}) showed that the absorption of micronutrients such as Zn ²⁺ , Mn²⁺ and Cu²⁺ increased by applying Bacillus megaterium, Bacillus thuringiensis and Bacillus spp., in lavender plants (Lavandula angustifolia L.) under water stress and in sage plants (Salvia divinorum L.). The results suggest that the stimulating effect of plant growth-promoting rhizobacteria on the absorption process can be attributed to their activity on nutrient solubilization (^{Ndakidemi et al., 2011}). The effect of Bs could be related to the production of organic acids that bind metals such as Fe, Zn, Cu and Mn, which at the same time increase the availability for plant absorption (^{Havlin et al., 2005}). Some authors have reported that individual application of inoculants to common bean has promoted greater nutritional absorption and accumulation in plant tissues.

Nutrient concentration in grain

Nutrient removal is a very important parameter in terms of quality for human nutrition. This is determined by the protein and mineral content present (^{Mune et al., 2013}). According to the mineral content in bean, it has fluctuated depending on the genotype, crop management and storage conditions. When considering the concentration of micronutrients in bean grains, this concentration was higher for all treatments (application of P and treatments inoculated with Bs) compared to what was reported by ^{Delfini et al. (2020}). They found a maximum concentration of 115, 23, 17 and 47 mg kg^-1 of Fe, Mn, Cu and Zn in 1512 accessions of common bean.

In this study, the data showed that there was no interaction between the factors evaluated (P × Bs) (Table 1). Except for Fe (p< 0.005), the rest of the micronutrients were not affected by the doses of P (p≥ 0.42). The maximum concentration of Fe was found in the treatments under the doses of 50 kg ha^-1 P, and the lowest concentration was observed in the treatments that were not applied P. The contents of Mn, Cu, Zn and B in the grain were in the order of 139, 93, 315 and 114 mg kg^-1. ^{Astudillo and Blair (2008}) showed that doses of P significantly increased concentrations of Fe and Zn in the grain.

One possible reason that may explain a high absorption of micronutrients is related to the production of nodules promoted by phosphorous fertilization, as mentioned by ^{Amare et al. (2014}); ^{Shanka et al. (2018}). Inoculation with PGPR has been associated with increasing micronutrient bio-fortification in common bean grains (^{Talaat et al., 2015}; ^{Jalal et al., 2021}). Except for Zn, inoculation with Bs increased the concentration of micronutrients (Table 1).

The increase in micronutrients in inoculated plants compared to non-inoculated plants was in the order of 140, 54, 25, 33 mg kg^-1, respectively for Fe, Mn, Cu and B. Although no effects were observed due to the inoculation of Bs on the concentration of Zn in grain, the values were even higher than those presented by ^{Delfini et al. (2020}). In some other crops such as wheat, the use of Bs has significantly increased the concentration of Zn in grain (68%) (^{Mumtaz et al., 2018}).