Initial pH of medium affects organic acids production but do not affect phosphate solubilization

The pH of the culture medium directly influences the growth of microorganisms and the chemical processes that they perform. The aim of this study was to assess the influence of the initial pH of the culture medium on the production of 11 low-molecular-weight organic acids and on the solubilization of calcium phosphate by bacteria in growth medium (NBRIP). The following strains isolated from cowpea nodules were studied: UFLA03-08 (Rhizobium tropici), UFLA03-09 (Acinetobacter sp.), UFLA03-10 (Paenibacillus kribbensis), UFLA03-106 (Paenibacillus kribbensis) and UFLA03-116 (Paenibacillus sp.). The strains UFLA03-08, UFLA03-09, UFLA03-10 and UFLA03-106 solubilized Ca3(PO4)2 in liquid medium regardless of the initial pH, although without a significant difference between the treatments. The production of organic acids by these strains was assessed for all of the initial pH values investigated, and differences between the treatments were observed. Strains UFLA03-09 and UFLA03-10 produced the same acids at different initial pH values in the culture medium. There was no correlation between phosphorus solubilized from Ca3(PO4)2 in NBRIP liquid medium and the concentration of total organic acids at the different initial pH values. Therefore, the initial pH of the culture medium influences the production of organic acids by the strains UFLA03-08, UFLA03-09, UFLA03-10 and UFLA03-106 but it does not affect calcium phosphate solubilization.


Introduction
Acidity has a direct effect on the activity of the soil microorganisms involved in a variety of processes, including organic matter decomposition, mineralization, immobilization, ammonification, nitrification, volatilization, biological nitrogen fixation and insoluble inorganic phosphate solubilization. Therefore, acidity is a chemical property of soils that plays a central role in agriculture.
Biological, chemical and physical factors may interfere with the ability of soil microorganisms to solubilize insoluble inorganic phosphates. In many cases, acidification is the main mechanism involved in phosphate solubilization. A significant negative correlation between the pH of the culture medium and phosphate solubilization by several genera and species of microorganisms was demonstrated by diverse authors [(Illmer and Schinner, 1992) -r = -0.49, (Chen et al., 2006) -r = -0.80, and(Marra et al., 2011) -r = -0.89]. For instance, Arthrobacter sp. solubilized 519.7 mg P L -1 when the pH of the culture medium decreased from 6.8 to 4.9 (Illmer and Schinner, 1992). In contrast, several studies have shown phosphate solubilization without a significant negative correlation with culture medium pH. For example, Pseudomonas sp. solubilized 31.0 mg P L -1 with no alteration of the culture medium, which was at an initial pH of 6.0 (Hariprasad and Niranjana, 2009). Narsian et al. (1995) also reported lack of correlation between pH and Ca 3 (PO 4 ) 2 solubilization by Aspergillus aculeatus after a 7 day incubation.
Phosphate solubilization depends not only on the decrease of the culture medium pH but also on other factors, such as exopolysaccharide production secreted by microorganisms. Under the same culture conditions, Arthrobacter sp. solubilized 111.7 mg P L -1 as the culture medium pH was lowered from 7.0 to 4.5, whereas Enterobacter sp. solubilized 632.6 mg P L -1 when the culture medium pH decreased from 7.0 to 4.3 which was due to the larger amount of exopolysaccharide produced by Enterobacter sp. Authors suggested that EPS with ability of phosphorus -holding may be a novel important factor in the microbial dissolution of tricalcium phosphate acting synergistically with organic acid (Yi et al., 2008).
Besides promoting a decrease in the pH of the medium, low-molecular-weight organic acids also chelate metals in solution, which increases the phosphorus available to plants. The degree of chelation depends on the type of organic acid involved, the number and proximity of carboxyl groups, the type of metal and the pH of the solution (Jones, 1998).
Only one study (Chaiharn and Lumyong, 2009) has assessed the influence of the initial pH of the growth medium on phosphate solubilization; however, these authors did not assess the production of organic acids. Marra et al. (2012) studied the solubilization ability of 82 strains in three types of phosphates (P, Al and Fe) both in solid (82 strains) and liquid (5 strains) media and verified that Ca phosphates are the most solubilized. They also found that there was a significant negative correlation between the pH of the medium and the amount of soluble phosphorus for CaHPO4 (r = -0.51**) and FePO 4 .2H 2 O (r = -0.28**), a relationship that was not observed for Al(H 2 PO 4 ) 3 . The highest solubilization of Ca-phosphate is due to its weaker chemical bound. Besides, Ca is a nutrient required in higher amounts than Fe. Al phosphate has the strongest bound and it is not a nutrient for microorganisms. Therefore, in this study, we sought to assess the influence of the initial pH of the culture medium on the production of low-molecular-weight organic acids and on the solubilization of calcium phosphate.
medium with different pH values and incubated at 28°C for 10 days. At the end of this period, the presence or absence of growth was assessed. The study design was completely randomized and included three replicates.

Solubilization of calcium phosphate and production of organic acids
Two experiments were performed to verify the ability of the above strains to solubilize insoluble inorganic phosphate from calcium phosphate in National Botanical Research Institute's solid and liquid growth media (NBRIP) (Nautiyal, 1999) containing 10 g L -1 glucose, 5 g L -1 MgCl 2 .6H 2 O, 0.25 g L -1 MgSO 4 .7H 2 O, 0.2 g L -1 KCl and 0.1 g L -1 (NH 4 ) 2 SO 4 . NBRIP medium was supplemented with Ca 3 (PO 4 ) 2 to a final concentration of 1000 mg of phosphorus per L in the solid medium and 100 mg of phosphorus per L in the liquid medium; different initial pH values were adjusted: 5.0, 6.0 and 7.0.
To produce and standardize inocula, the strains were inoculated into liquid medium 79 (Fred and Waksman, 1928) containing 0.5 g L -1 K 2 HPO 4 , 0.2 g L -1 MgSO 4 .7H 2 O, 0.1 g L -1 NaCl, 10.0 g L -1 mannitol and 0.4 g L -1 yeast extract, pH 6.8. Strains were incubated with shaking (110 rpm) at room temperature under aerobic conditions. Readings were performed periodically on a spectrophotometer at a wavelength of 560 nm until an optical density (OD) of 0.5 was reached, which was equal to approximately 10 8 cells per mL. A 0.85% saline solution was used to adjust cells to the desired density when the OD exceeded 0.5.
For assessment in solid NBRIP medium, Petri dishes containing NBRIP medium at each initial pH condition were inoculated in quadruplicate with 20 mL aliquots of each culture (strain) at an OD of 0.5. The control treatment consisted of non-inoculated NBRIP medium. The culture dishes were incubated at 28°C, the diameter of the solubilization halo (translucent area surrounding colonies) was measured at the beginning of solubilization, i.e., at the 3 rd day and, after the 15 th day incubation, using a digital paquimeter, and the Solubilization Index (SI) expressed as halo diameter (mm) / colony diameter (mm) was calculated (Akintokun et al., 2007). The investigated strains were classified based on their SI as demonstrating low (SI < 2.00), intermediate (2.00 < SI < 4.00) and high (SI > 4.0) solubilization capacities.
For assessment in liquid NBRIP medium, a 1 mL aliquot of culture medium 79 with an OD of 0.5 at 560 nm was inoculated into a 125 mL Erlenmeyer flask containing 50 mL of NBRIP medium at different initial pH values. The flasks were incubated at 28°C with shaking at 130 rpm for 10 days. Subsequently, the samples were centrifuged (19,187 g for 5 min), and the pH was measured, as well as the amount of soluble phosphorus in the supernatant using the phosphomolybdate method (Murphy and Riley, 1962). In addition, the organic acids produced in the medium were quantified. For each initial pH value, a non-inoculated control was assessed. The ability of each strain to solubilize phosphate was calculated as the difference between the concentration of soluble phosphorus in the culture medium of samples that had been inoculated with bacterial strains and that of the non-inoculated control treatment.
High-performance liquid chromatography (HPLC) (Agilent HP Series 1100) was used to identify and quantify organic acids. Samples were collected, filtered through a 0.45 mm cellulose membrane and injected into a Supelcogel C-610H 9 mm chromatographic column measuring 30 cm x 7.8 mm. The eleven Merck® pro-analysis organic acids that have been reported in the literature as being involved in solubilization were used as analytical standards. The mobile phase was 0.1% H 3 PO 4 (pH 1.81) with a 0.5 mL min -1 flow rate and a 100 mL injection per sample. The method was according manufacturer (SUPELCO/SIGMA ALDRICH) of the column Supelcogel. The acquisition time of the chromatograms was estimated to be 30 min with 30 min intervals between runs. Detection was performed by UV at 210 nm with a diode array detector (DAD). The molecules identified and their typical retention times were as follows: oxalic acid (10.10 min), 2-ketogluconic acid (12.10 min), citric acid (12.40 min), gluconic acid (13.04 min), maleic acid (13.33 min), tartaric acid (13.45 min), malic acid (14.85 min), malonic acid (15.23 min), lactic acid (17.89 min), succinic acid (17.91 min) and propionic acid (25.08 min). The quantification of acids was performed using calibration curves of the standards.
The experiment with NBRIP liquid medium was performed in independent assays for each initial pH value with a completely randomized design and two replicates. The results were evaluated by variance analysis using Sisvar (version 4.6) (Ferreira, 2008), and means were compared using the Scott-Knott test at 5%.

Results
All strains grew in the culture medium 79 at all of the initial pH values studied. In solid NBRIP medium, the strain UFLA 03-116 did not solubilize Ca 3 (PO 4 ) 2 at any of the initial pH values studied. The other strains did solubilize Ca 3 (PO 4 ) 2 at all pH values and exhibited low SI after a 15-day incubation. The only exception was the strain UFLA 03-08 (R. tropici), which demonstrated an intermediate SI at all of the investigated initial pH values (Table 1).
In liquid NBRIP medium, the strain UFLA 03-116 (Paenibacillus sp.) exhibited the same behavior as in the solid medium and did not solubilize Ca 3 (PO 4 ) 2 at any initial pH value, and the pH of the medium did not change from its initial value (Figure 1). The strains UFLA 03-08 (R. tropici), UFLA 03-09 (Acinetobacter sp.), UFLA 03-10 (P. kribbensis) and UFLA 03-106 (P. kribbensis) solubilized Ca 3 (PO 4 ) 2 at all initial pH values in liquid NBRIP medium, with no difference being observed in the amount pH and organic acids in P solubilization 369  (  of soluble phosphorus among the different initial pH values of the medium. It is worth noting that more than 60% (as much as 77% in some cases) of insoluble inorganic phosphate (Ca 3 (PO 4 ) 2 ) was solubilized by these strains.
For most strains, the initial pH of 5.0 did not change after 10-day incubation nor it differed from the control (Figure 1); the only exception was the strain UFLA 03-09 (Acinetobacter sp.) in which the pH decreased to less than 4.0. At initial pH values of 6.0 and 7.0, the pH decreased after a10-day incubation among all the strains where solubilization occurred. The greatest difference was found at the initial pH of 7.0, which decreased to less than 4.0 for the UFLA 03-09 strain.
With respect to the identification, for the strain UFLA 03-08 (Rhizobium tropici), the highest total acid concentration (malic acid, 18.90 mmol L -1 ) was found at an initial pH of 7.0; for this strain, acid production varied with the initial pH of the culture medium. Organic acids identified at pH 5.0 and 6.0 were respectively: 2-ketogluconic (1.28 mmol L -1 ) and lactic/succinic acid (0.37 mmol L -1 ). Other peaks were not identified in all pH values (Figure 4 -Chromatograms G, H and I).
For the strain UFLA 03-09 (Acinetobacter sp.), the only acid detected was the gluconic acid, at pH values of 5.0 (34.25 mmol L -1 ), 6.0 (30.64 mmol L -1 ) and 7.0 (37.34 mmol L -1 ), which indicates consistency in the production of acids independent of the initial pH of the growth medium.
The UFLA 03-116 (Paenibacillus sp.) was the only strain that did not produce any organic acids under any cul-pH and organic acids in P solubilization 371 ture condition and, therefore, it exhibited the same behavior as the control treatment; the chromatograms show a peak at a retention time of 8.91 min, which differs from all the retention times exhibited by the acids in this study (Figure 4). It is worth noting that this peak was also present in all of the other treatments and does not interfere with the identification and quantification of the acids studied. Under these treatment conditions, citric acid, oxalic acid, maleic acid and malonic acid were not found.

Discussion
The pH of the culture medium directly influences the growth of microorganisms and the biochemical processes they perform. In many cases, acidification is the main mechanism involved in phosphate solubilization (Halder et al., 1990;Jha et al., 2009;Marra et al., 2011;Marra et al., 2012;Whitelaw, 2000). However, several studies have shown a lack of correlation between solubilized phosphorus and pH of the medium (Chaiharn and Lumyong, 2009;Xie, 2009). Therefore, a better understanding of the behavior of phosphate-solubilizing bacteria inoculated into culture media at different initial pH values may contribute to the production and management of inoculants that improve crop production.
Our results showed that in both solid and liquid NBRIP medium, the initial pH did not affect the solubilizing activity of strain UFLA 03-116 (Paenibacillus sp.) 372 Marra et al.  because it was not able to solubilize Ca 3 (PO 4 ) 2 under these conditions. These results reveal that the inability to solubilize phosphate under these conditions is intrinsic to this strain, because it grew on solid medium, which was visible in Petri dishes, and liquid medium, as was verified by the presence of bacterial biomass during the centrifugation process. Studies (Marra et al., 2012) performed with this same strain of Paenibacillus sp. in solid and liquid GELP medium (Sylvester-Bradley et al., 1982) at an initial pH of 7.0 also demonstrated its inability to solubilize CaHPO 4 , Al(H 2 PO 4 ) 3 and FePO 4 .2H 2 O. The only exception was for FePO 4 .2H 2 O in liquid medium; for this phosphate, more than 20% of the phosphorus was solubilized (Marra et al., 2012). An initial pH of 7.0 in GELP medium may contribute to the solubilization of FePO 4 .2H 2 O by strain UFLA 03-116 (Paenibacillus sp.). With respect to the strains that solubilized Ca 3 (PO 4 ) 2 , UFLA 03-08 (R. tropici) was the only one to have an intermediate SI, which occurred at all of the initial pH values studied, after a 15-day incubation at 28°C. These SI values were higher than those reported for Rhizobium species obtained from Crotalaria retusa and Crotalaria verrucosa inoculated onto solid Pikovskaya culture medium (Pikovskaya, 1948) at an initial pH of 7.0 (Sridevi et al., 2007), thereby demonstrating that pH does not interfere with solubilization by this strain.
Conversely, the low SI exhibited by the strains UFLA 03-09, UFLA 03-10 and UFLA 03-106 on solid medium contrasts with that of liquid medium in which these strains solubilized significant quantities of phosphates.
The strains UFLA 03-08 (R. tropici), UFLA 03-09 (Acinetobacter sp.), UFLA 03-10 (P. kribbensis) and UFLA 03-106 (P. kribbensis) solubilized Ca 3 (PO 4 ) 2 to a similar degree (more than 60%) at all of the initial pH values of NBRIP medium studied. UFLA 03-09 (Acinetobacter sp.) was the only strain that decreased the pH during all treatments, which may be related to its production of gluconic acid. Chaiharn and Lumyong (2009) found that after a 5-day incubation of Acinetobacter sp. in nutrient broth with an initial pH of 7.0 or 9.0, the pH of the medium decreased, but the pH increased to 6.17 in medium with an initial pH of 5.0; nevertheless, solubilization of calcium occurred. However, these authors did not assess the production of organic acids.
Several authors have suggested that a decrease in pH due to the production of organic acids and the release of protons is a basic principle of phosphate solubilization, (Chen et al., 2006;Sperber, 1958;Whitelaw, 2000). However, the strains UFLA 03-08, UFLA 03-10 and UFLA 03-106 did not decrease the initial pH of 5.0 after 10-day incubation, thereby demonstrating that acidification is not the mechanism used to promote solubilization at this initial pH, even when organic acids are produced in different concentrations. In this case, the acids may be present in anionic forms and therefore do not function in medium acidifica-tion but, rather, in Ca 2+ chelation (Jones, 1998;Whitelaw, 2000). Moreover, the strain UFLA 03-08 at an initial pH of 6.0 exhibited efficient solubilization, with a decrease of pH but producing a low concentration of lactic/succinic acid (0.37 mmol L -1 ). This result indicates that other solubilization mechanisms are involved and that this strain utilizes different mechanisms when the pH of the medium varies. These mechanisms can be: proton exclusion (via cellular respiration and ammonium absorption as N source) (Illmer P and Schinner F, 1992), siderophores (Hamdali et al., 2008) and exopolisaccharide (EPS) production (Yi et al., 2008). The first two mechanisms were not evaluated in this paper, however, all these three strains produce large amounts of EPS that could act synergistically with acid production as suggested by Yi et al. (2008). Non-solubilizer strain UFLA 3-116 also produces large amounts of EPS however it did not produces organic acids.
Conversely, medium acidification occurred at initial pH of 6.0 and 7.0, after a 10-day incubation, followed by the production of lactic/succinic and malic acids by the strain UFLA 03-08, 2-ketogluconic, tartaric and propionic acid by the strain UFLA 03-10, and tartaric and propionic acids by the UFLA 03-106. Propionic acid was produced to the greatest degree by the latter two strains which belong to the same species.
The acids produced in larger amounts (propionic > gluconic > tartaric > malic) have pka varying from 4.79 (propionic) to 3.07 (tartaric) without showing any relationship with phosphate solubilization. On the other hand, strain UFLA03-09 (Acinetobacter sp.) which decreased the pH to lower level, only produced gluconic acid which has an intermediate pka (3.65).
The strains UFLA 03-08, UFLA 03-09, UFLA 03-10 and UFLA 03-106 solubilized Ca 3 (PO 4 ) 2 at all three initial pH values studied. Brazilian soils usually exhibit acidic pH values, often varying between 5.0 and 6.5, and the practice of liming aims to reach pH 5.5-6.5. Therefore, these strains may increase and maintain the availability of phosphorus to plants across a wide variety of soil management practices. Besides contributing to solubilization, the production of certain acids by these strains may also serve as a readily accessible source of carbon for these microorganisms (Jones, 1998).
The results show that: the initial pH of the culture medium influences the production of organic acids by the strains UFLA 03-08, UFLA 03-09, UFLA 03-10 and UFLA 03-106 but they do not promote the solubilization of calcium phosphate; thus medium acidification is not the mechanism by which the strains UFLA 03-08, UFLA 03-10 and UFLA 03-106 solubilize calcium phosphate when the initial pH at medium is 5.0 and that strains UFLA 03-09 and UFLA 03-10 produced the same acids when the culture medium exhibited different initial pH values. 374 Marra et al.