The production of organic acids and soluble phosphorus generation was performed using a liquid fermentation process. The fermenters used were 250 ml Erlenmeyer flasks under ambient conditions of light, temperature and relative humidity, and constant agitation of 120 rpm. Pikovskaya (PVK) was peformed as the culture medium, varying the phosphorus source among P-Ca, P-Al and P-Fe, (Cisneros & Prager, 2015). 99 experimental units were performed from these treatment combinations: 3 sources P, 10 microbial strains, plus 3 uninoculated, controls with 3 replications/treatment. Microorganisms are inoculated at 3 x 10⁸ cfu.ml-1 bacteria and 1x10⁶ spores.ml-1 in fungi. Bacteria were identified B1 to B5 and H1 to H5 fungi. The fermentations were carried out for 7 days. 10 ml of sample in each treatments on days 1, 3, 5 and 7, and the analysis were performed as follows:

Was estimated by plating and serial dilution. A qualitative and quantitative identification of organic acids, were performed, respectively. High Efficiency Liquid Chromatography (HPLC) under the conditions used by Mardad et al., (2013), was used as the analytical method.

Soluble phosphorus generated by fungi and bacteria. The blue molybdovanadate was performed as the spectrophotometric technique (Mardad et al., 2013)

It was performed using molecular techniques. In fungi, primers ITS1 and ITS4, were performed, and the FD1 and RD1 in bacteria (Sasso et al., 2012).

The arrangement of the experiments correspond to a completely randomized design, and the response variables were as follows: concentration of soluble phosphorus, organic acids quantified and total acidity generated in the medium by each microorganism. Analysis of variance with a 95% significance was established, and in the required cases, comparison of means by Duncan test. Pearson correlation was analyzed to establish relation among found organic acids and soluble phosphorus. The coefficient of variation (CV) was determined for each of the estimated acids for each source of P used. All analyzes were performed using SAS statistical package version 9.1.3. (r).

In general, both bacterial and fungal strains produced D-malic, D-lactic, L-malic, L-lactic, acetic, citric and gluconic acids. Separate analysis of cultivated strains show that in all five bacteria, predominated gluconic acid secretion with concentrations between 500 and 3000 ppm, especially P-Ca, in which B5, showed the highest value. In the other two P sources, the contents of these organic acids were below 50 ppm and in some cases invaluable. Regarding fungi citric acid predominated with concentrations between 100 and 300 ppm, mainly P-Ca. In P-Al H3 and H5, only citric acid metabolised in the order of 100 and 170 ppm, respectively. P-Fe lower concentrations were recorded (0 to 50 ppm).

When performing the appropriate statistical analysis, CV were found in the evaluated organic acids, fluctuated considerably, presenting equal quantities or greater than 80% in each of the P sources (Table 1).

Table 1

In contrast, Ghaly & Kamal (2004), and Rathi et al. (2002), who in biological systems and fermentation processes found CV between 3.9% and 39.05%. Low and/or negative Peason correlations are also evidence, no significantly response among the estimated acids and soluble phosphorus (Table 2).

Table 2

Conventions: The Pearson correlation coefficients is accompanied by its significance, which it corresponds to the value found in a smaller size. Ac. = Acidity.

However, citric acid, showed the highest correlation in the three sources. These findings raise doubts in terms of the normal distribution in the fermentative processes.

The search for an explanation to this situation have allowed establishing one of the possible causes, and is the way of the interpretation results. Jones et al. (2003), analyzed several studies which underestimated the concentration of organic acids in the rhizosphere due to the dynamics that occur in soil. Based on the above, the organic acid monitoring scheme during the bioprocess was carried out in this research (Figure 1).

Conventions: Ac

As can be shown, from the culture medium and microbial inoculation (Figure 1 A), the organic acids (B), are produced by part of them, in addition, participate in the solubilization reaction (C) and the remaining, is soluble in the solution (D), the latter, which is determined by HPLC and analyzed statistically, produce the unfavorable findings above mentioned.

In addition, the solubilized phosphorus should be related to the total acids produced by each microorganism, not individualities, as mentioned by Oburger et al. (2009), the phosphorus release was higher when citric, malic, oxalic and malonic acids were combined. Mardad et al. (2013), take similar approaches and relate total acids with soluble phosphorus, finding positive, highly significat correlations (p <0.01, r = 0.93). Therefore, a decision was established to determine the total acidity, from converting the units in which express concentrations each estimated acid by HPLC, transforming figures from mg.L^-1 to meq.L^-1, since this last unit, allows concentrations that are equivalent and can be added. In turn, the amount of soluble phosphorus in this unit (meq.L^-1) is the amount of acids, which were involved in the solubilization process (Figure 1 E). Therefore, total acids produced by each microorganism correspond to the sum of the acids determined by HPLC plus the acids involved in the solubilization of phosphate (Table 3).

Table 3

Table 3 continuación

Table 3 continuación

With this methodological adjustment, the statistical results changed considerably since the correlation among total acidity and soluble phosphorus (Table 2), and the minor variation coefficient (P-Al: 39.53%, P-Ca: 18.53%, P-Fe: 38.76%).

Determining the correlation among the total acidity and soluble phosphorus have allowed understanding the solubilization process as a biotechnological dynamics induced by microorganisms (Figure 2).

Conventions: The

From which, the phosphate source used, is critical and solubilized, can stimulate or inhibit the organic acids production of microbial origin in this case. Both, the source of P, as the type of acid metabolised and its concentration, influence the generation and amount of available phosphorus. (Figure 3).

Conventions: meq

The phosphate source of microorganisms, showed higher production of organic acids (P-Ca> P-Al> P-Fe), which coincides with the greatest amount of soluble phosphorus.

Normally, when the P solubilization process is analyzed, the result emphasizes in the availability of this element. However, the biotechnological dynamics for solubilizing phosphates "in vitro" (Figure 2), can display other important components. The release of Ca, Al and Fe as organic chelates and/or free ions, depending on the used source, which introduce additional changes that will affect, for example, the growth of microbial populations in environmental conditions that may be positive or negative for the whole process, among others.

Both fungi and bacteria, showed the highest growth population (Figures 4a and c) when the source used was P-Ca, coincident with the production of acids (Figure 3). Papagianni (2004), argues that one of the determining factors in the generation of secondary metabolites (in this case, organic acids) is the development of microorganisms. In the P-Ca medium, fungi remained in exponential phase, while bacteria achieved twice the population obtained by fungi between days 3 and 5. However, 4 of the 5 bacteria showed a population decrease in day 7, except the B3 which took almost 5 days to begin growth and reached only the exponential phase (Figure 4a and c). On the contrary, fungi with smaller populations remain longer, which is a sustained ability to produce acids that are more organic and therefore, solubilize more phosphorus.

Figure 4

In concordance to Papagianni (2004), the initiation of growth population is matched by the release of the first traces of acids, which react with the phosphate source to start P solubilization, vital for cell development. Besides this soluble phosphorus, other nutrients remaining availables (Figure 2), become in food sources for the involved microorganisms. Naveena (2011), established that the available Ca induces fungal growth.

The P-Al source influence negatively the production of organic acids in fungi and bacteria mainly (Figure 3). The toxic effect of solubilized Al is initially invisible and appreciated only through bacterial incubation time, to the extent that their concentration increases in the culture medium (Figure 4b). Similar results have been obtained by Illmer & Erlebach (2003), who found the aluminum concentration exceeded 30 uM, which is a harmful effect is seen on bacteria of the genus Arthrobacter sp. The fungi in P-Al had a similar growth as the one presented in P-Ca (Figure 4c). This behavior could be explained by an internal tolerance mechanism in the cell wall, which has the ability to adsorb the Al³⁺ by functional groups such as hydroxyls, amines and carboxyls present in the polysaccharide. Although the increase in biomass remains stable, acid secretion is decreased (Chao et al. 2013).

The P-Fe, negatively affected the production of organic acids in both microbial groups (Figure 3), with emphasis on bacteria, Chamnongpol et al. (2002), argue that the Fe³⁺ as part of FePO₄ can permeabilize the bacterial membrane from the outside, resulting in lysis and death of the organism, explaining the few or no populations found in this research (Figure 4b).

In this P-Fe medium, fungi grew even less than in P-Ca and the evaluated concentration of organic acids was drastically affected. Philpott (2006), argues that the iron in ferric form is not absorbed, due to the fungi cell wall using a reducing system which involves in two stages: the first ferric iron (Fe³⁺⁾ is reduced to ferrous (Fe²⁺⁾ in the plasma membrane by proteins designated FRE reductases, and then is transported into the cell by specialized proteins called FIT.

When the intracellular concentration of Fe²⁺ is increased, it becomes toxic as it reacts with oxygen and generates free radicals that attack macromolecules such as DNA leading to cell death (de Freitas & Meneghini, 2001). Apparently, this explains that the H1, H2, H3 and H4 fungi would reduce their population (Figure 4c) among the fifth and seventh day of fermentation. The only fungus that did not present this decline was H5, considering that it possibly has a greater number of heme proteins that store intracellular iron (Philpott, 2006), counteracting the toxic effect and having increased tolerance to high concentrations of iron.

In general, the ten evaluated microorganisms were able to grow in greater and/or lesser degree, in the presence of P-Ca, P-Al and P-Fe, while solubilizing these sources. The H5 and H3 fungi, showed the greatest solubilizing capacity, although to a different extent, on the three sources used in this study. H3, was identified as Penicillium ochrochloron, widely recognized as phosphate solubilizing by producing citric and malic acid (Coutinho et al., 2012). Conversely, bacteria solubilized especially P-Ca, the most efficient B5, identified as Kocuria sp, which is registered in the group of plangt growth promoters by different mechanisms, including the phosphates solubilization of (Goswami et al., 2014).

These results, indicates a very important biotechnological potential for the use of microorganisms whicht solubilize P and this can lead to reduce the economic costs coupled with social and ecosystem effects, which is involved in meeting the P needs in crops with environmental friendly alternatives.