Inhibition of Aspergillus niger Phosphate Solubilization by Fluoride ...

Inhibition of Aspergillus niger Phosphate Solubilization by Fluoride Released from Rock Phosphate Gilberto de Oliveira Mendes,a Nikolay Bojkov Vassilev,b Victor Hugo Araújo Bonduki,a Ivo Ribeiro da Silva,c,e José Ivo Ribeiro, Jr.,d Maurício Dutra Costaa,e Department of Microbiology, Federal University of Viçosa, Viçosa, Brazila; Department of Chemical Engineering, Faculty of Sciences, University of Granada, Granada, Spainb; Department of Soil Science, Federal University of Viçosa, Viçosa, Brazilc; Department of Statistics, Federal University of Viçosa, Viçosa, Brazild; National Council for Scientific and Technological Development (CNPq), Brasilia, Brazile

The simultaneous release of various chemical elements with inhibitory potential for phosphate solubilization from rock phosphate (RP) was studied in this work. Al, B, Ba, Ca, F, Fe, Mn, Mo, Na, Ni, Pb, Rb, Si, Sr, V, Zn, and Zr were released concomitantly with P during the solubilization of Araxá RP (Brazil), but only F showed inhibitory effects on the process at the concentrations detected in the growth medium. Besides P solubilization, fluoride decreased fungal growth, citric acid production, and medium acidification by Aspergillus niger. At the maximum concentration found during Araxá RP solubilization (22.9 mg Fⴚ per liter), fluoride decreased P solubilization by 55%. These findings show that fluoride negatively affects RP solubilization by A. niger through its inhibitory action on the fungal metabolism. Given that fluoride is a common component of RPs, the data presented here suggest that most of the microbial RP solubilization systems studied so far were probably operated under suboptimal conditions.

T

he use of phosphate-solubilizing microorganisms (PSM) is emerging as a biotechnological alternative for producing soluble P fertilizers from rock phosphate (RP) (1). The ability of PSM to mobilize P from sparingly soluble sources can be a useful tool in P fertilization management. Some studies have shown that the product obtained from the treatment of RP with PSM (2) or even the direct application of PSM to soil (3) can improve plant growth and P uptake. This alternative is becoming increasingly important against a backdrop of depletion of high-grade RP reserves. Despite the uncertainties of forecasts about the depletion of these reserves, ranging between 30 and 300 years, there is a consensus that the accessibility and quality of RPs are decreasing and, consequently, production costs of P fertilizers are rising (4). Therefore, efficient processes, including microbially mediated ones, able to exploit lower-grade RPs and/or alternative P sources (5) at low cost should be developed in the near future. Rock phosphates differ in chemical and mineralogical characteristics depending on the location where they are collected. The basic unit is apatite [Ca10(PO4)6(Z)2], which is classified as fluoroapatite, chloroapatite, or hydroxyapatite when Z is F, Cl, or OH (6). In addition to apatite, the RPs contain significant amounts of numerous other chemical elements (7). In some RPs, the concentrations of these accompanying elements can be quite high and include some toxic elements, e.g., uranium, cadmium, and a number of other heavy metals (4, 7). Reyes et al. (8) suggested that the presence of toxic elements in RP could inhibit fungal growth and, consequently, P solubilization. However, to exert any effect, these elements first have to be mobilized, but so far, no reports of which elements are actually released during microbial RP solubilization have been published. Some of these accompanying elements are presumably released together with P during RP solubilization and could inhibit the process. This fact could explain the lower solubilization rate of RPs compared to that of pure synthetic apatites (9). The main mechanisms of microbial P solubilization include the production of organic acids, which have the ability to form stable

4906

aem.asm.org

complexes with cations that form poorly soluble compounds with P (10, 11), and, to a lesser extent, the release of protons (H⫹) into the medium (12). Some elements that may be released during RP solubilization could affect these mechanisms by promoting changes in microbial metabolism (13). Schneider et al. (9) observed lower production of citric and gluconic acids by Aspergillus niger in comparisons of the solubilization of RPs to that of pure synthetic apatite. Elements like Cu, Fe, Mn, and Zn, even at low concentrations, inhibit the production of organic acids by fungi (14, 15) and could be involved in the lower production observed by Schneider et al. (9). Furthermore, Illmer and Schinner (12) proposed that P solubilization by some microbial species is based on the release of H⫹ resulting from processes related to biomass production, such as respiration or NH4⫹ assimilation. Thus, the inhibition of microbial growth could result in a decreased release of H⫹ into the medium and, consequently, diminished P solubilization. Past studies with PSM have overlooked the potential inhibitory effect of elements released during microbial RP solubilization. A better understanding of the P solubilization process can lead to new perspectives on strategies to improve its efficiency. Thus, the objective of this work was to determine which chemical elements are released during fungal RP solubilization and to evaluate the effects of these elements on the P solubilization by A. niger.

Applied and Environmental Microbiology

Received 6 May 2013 Accepted 4 June 2013 Published ahead of print 14 June 2013 Address correspondence to Maurício Dutra Costa, [email protected]. Supplemental material for this article may be found at http://dx.doi.org/10.1128 /AEM.01487-13. Copyright © 2013, American Society for Microbiology. All Rights Reserved. doi:10.1128/AEM.01487-13

p. 4906 – 4913

August 2013 Volume 79 Number 16

Inhibition of Fungal P Solubilization by Fluoride

MATERIALS AND METHODS

TABLE 1 Chemical composition of Araxá rock phosphate

Microorganism and cultivation conditions. The isolate Aspergillus niger FS1 was obtained from the Collection of Phosphate Solubilizing Fungi, Institute of Biotechnology Applied to Agriculture (BIOAGRO), Federal University of Viçosa, Viçosa, MG, Brazil. Batch fermentations were conducted in 125-ml flasks containing 50 ml of the National Botanical Research Institute’s phosphate growth medium (NBRIP medium) (16) [10 g glucose, 5 g MgCl2 · 6H2O, 0.25 g MgSO4 · 7H2O, 0.2 g KCl, 0.1 g (NH4)2SO4, 1 liter deionized water]. The P source in the NBRIP growth medium used in the present experiments was either Araxá RP or K2HPO4. The medium pH was adjusted to 7.0 before the application of the P source. The flasks were inoculated with 106 conidia from a conidial suspension prepared in 0.1% (vol/vol) Tween 80. All flask cultures were incubated on an orbital shaker at 160 rpm and 32°C. Rock phosphate characterization. RP from Araxá (Brazil) was used in the solubilization studies. Chemical analyses (see listing in Table 1) were done after the digestion of an RP sample with aqua regia acid solution (3 HCl:1 HNO3) or lithium metaborate (17). The concentration of the chemical elements was determined by inductively coupled plasma optical emission spectrometry (ICP-OES) and inductively coupled plasma mass spectrometry (ICP-MS), except for that of P, which was determined by an ascorbic acid method (18), and those of Cl⫺ and F⫺, which were determined with specific ion electrodes. The mineralogical composition was determined by powder X-ray diffraction (XRD) in a multifunctional Panalytical X’Pert Pro PW 3040/60 diffractometer equipped with a 1,800-W, 60-kV cobalt tube (Co-K␣ radiation, ␭ ⫽ 1.790269 Å) operated at 40 kV and 30 mA. Powder samples were mounted in holders in order to minimize preferred orientation, and the scans were performed in a step-by-step mode from 4o to 80o 2␪ with 0.05o increments every 2 s. Kinetics of rock phosphate solubilization. Some elements were prioritized based on their biological significance in order to simplify further analyses. Among the 63 elements detected in Araxá RP, 10 (Ag, Au, Bi, Cs, In, Sb, Sn, Te, Tl, and W) were excluded for being below the detection limit when 3 g liter⫺1 of RP was added to the medium. Of the 53 remaining elements, besides P, 28 (Al, As, B, Ba, Be, Ca, Cd, Co, Cr, Cu, F, Fe, Ga, K, Li, Mg, Mn, Mo, Na, Ni, Pb, Rb, S, Si, Sr, V, Zn, and Zr) were selected for analysis. The kinetics study was conducted for 10 days in NBRIP medium supplemented with 3 g liter⫺1 of RP. In all, 66 flask cultures were set up and, every 12 h, three flasks were removed from the shaker for analyses. The spent medium was passed through 0.45-␮m-pore-size membranes by vacuum filtration and analyzed for pH and the chemical elements released. The fungal biomass retained on the membranes was collected, dried in an oven at 70°C to a constant weight, and incinerated at 500°C for 6 h. Biomass yield was determined by subtracting from the weight of dried fungal mycelium the weight of the residue left after its incineration. This method avoids overestimation due to the adherence of phosphate particles to the mycelium (19). Uninoculated flasks from the beginning and end of the experiment were used as controls in the determination of the solubilized elements. Screening of chemical elements affecting rock phosphate solubilization. The chemical elements released during the kinetics study were combined in a factorial experiment in order to screen which ones affect the RP solubilization. Due to the large number of elements, a Plackett-Burman design (PBD) was chosen for forming the combinations of elements in the treatments. The PBD is a fractional factorial design where n factors are combined at two levels in at least n ⫹ 1 treatments for estimating the main effect (linear coefficient) of each factor, excluding the interactions between them (20). Thus, each one of the 17 elements was added at two coded levels: ⫺1 (absence of the element) and 1 (maximal concentration of the element achieved over the entire kinetics study) (Table 2). The combinations of elements in the treatments (see Table S1 in the supplemental material) were created using the DOE (design of experiments) option of Minitab 16 statistical software. A central point (mean between

Element

Concn (mg kg⫺1)

Ag Al As Au B Ba Be Bi Ca Cd Ce Cl Co Cr Cs Cu Dy Er Eu F Fe Ga Gd Ge Hf Hg Ho In K La Li Lu Mg Mn Mo Na Nb Nd Ni P Pb Pr Rb Re S Sb Sc Se Si Sm Sn Sr Ta Tb Te Th Ti Tl Tm U V W Y Yb Zn Zr

0.66 3,032 13.5 0.10 374.5 20,704 12.7 0.06 302,450 0.89 3,468 ⬍20 17.5 136.8 0.35 31.0 113 36.4 70.0 15,931 59,700 17.3 193.9 3.7 30.5 0.0 17.2 0.07 600 1,580 1.5 2.0 2,700 1,750 6.0 2,967 1,290 1,689 55.5 139,700 27.4 435.3 4.5 ⬍0.1 1,050 0.12 38.3 6.5 12,080 281.1 10.8 7,622 27.1 22.5 0.15 240.4 21,600 0.60 3.5 70.4 131.5 7.1 315.5 17.0 190.5 1,564


levels ⫺1 and 1, coded as 0) was added to the experiment and replicated five times in order to determine the experimental error but was not included in the adjusted model. The main effects of the elements on P solubilization were estimated through regression analysis by the least-

aem.asm.org 4907

Mendes et al.

olvation gas flow 500 liter h⫺1, collision gas flow 0.14 ml min⫺1, mode positive.

TABLE 2 Maximum concentration of elements achieved during solubilization of Araxá rock phosphate by Aspergillus nigera Concn (mg liter⫺1) Al B

Ba

Ca

F

Fe

Mn Mo Na

Ni

Pb

Rb

Si

Sr

V

Zn

Zr

2.8 0.8 1.87 122.3 22.9 6.47 2.03 0.03 9.25 0.09 0.07 0.05 6.69 17.33 0.09 0.13 0.4 a

The study was done using 50 ml of NBRIP medium supplemented with 3 g of rock phosphate (particle size ⬍ 75 ␮m in diameter) per liter. The flasks were inoculated with 106 conidia from a conidial suspension prepared in 0.1% (vol/vol) Tween 80 and incubated on an orbital shaker at 160 rpm and 32°C.

squares method. The DOE option of the Minitab software was used to analyze the data, and the results were interpreted based on the significance (P ⬍ 0.05) of the regression coefficient of each element in the fitted equation. The model adopted was as follows: yijm ⫽ ␤0 ⫹

n

兺 ␤jxm ⫹ ␧ijm j⫽1

(1)

where yijm is the value of solubilized P observed in run i with the combination of n chemical elements at level xm, ␤0 is the regression constant, ␤j is the regression coefficient of the linear effect of each factor (n chemical elements), m is the coded level of each element (⫺1 or 1), and εijm is the experimental error associated with the observation yijm. The experiment was conducted for 60 h in NBRIP medium supplemented with 3 g liter⫺1 of RP. For each element studied, a solution was prepared with the appropriate concentration for adding 1 ml to flasks at level 1. The following chemicals were used: AlCl3, H3BO3, BaCl2 · 2H2O, CaCl2 · 2H2O, KF · 2H2O, FeCl3, MnCl2 · 4H2O, (NH4)6Mo7O24 · 4H2O, NaCl, NiCl2, Pb(NO3)2, RbCl, K2SiO3, SrCl2 · 6H2O, V standard solution (Spectrum), ZnCl2, and Zr standard solution (Vetec). The solutions were prepared in ultrapure water, and all glassware was washed in 2% HCl before use. Given the low concentration of most elements, changes in the medium composition by the accompanying ions were negligible. The flasks were filled with 25 ml of a double-concentrated NBRIP medium and, after the addition of the corresponding element solutions, were made up to 50 ml with ultrapure water. At the end of the experiment, the spent medium was filtered through quantitative filter paper (phosphorus free, 15- to 17-␮m pore size), and the solubilized P in the filtrate was determined as described above. Effect of fluoride on RP solubilization and metabolism of A. niger. The effect of fluoride on the solubilization process was studied under two sets of cultivation conditions by adjusting the P source. The experiment was conducted for 60 h in NBRIP medium supplemented with 3 g RP per liter or 1 g K2HPO4 per liter. NaF was added to the medium at concentrations ranging from 0 to 50 mg of fluoride per liter, with increments of 5 mg liter⫺1. At the end of the experiment, solubilized P and the pH were determined in the treatments with RP. The P/biomass yield (YP/X) was calculated from the ratio of solubilized P (mg) to fungal biomass produced (g). The fungal biomass and the production of organic acids were determined in the treatments that received K2HPO4 as a P source. The experiment was conducted using an entirely randomized design with three replicates at the central point (25 mg F per liter) followed by regression analysis. Organic acids were determined by ultraperformance liquid chromatography-tandem MS (UPLC/MS/MS) using a UPLC Acquity system coupled to a Xevo TQS mass spectrometer (Waters, Milford, MA). Based on the previous characterization of the isolate A. niger FS1 (21), the analysis was focused on citric, gluconic, and oxalic acids. Chromatographic separations were performed using an Acquity UPLC BEH C18 column (1.7 ␮m, 2.1 mm by 100 mm) under the following conditions: mobile phase, 0.1% phosphoric acid; flow, 0.2 ml min⫺1; sample injection volume, 10 ␮l; and analysis time, 3.5 min. Mass spectrometry was performed under the following conditions: source electrospray (ESI), source temperature 150°C, desolvation temperature 300°C, cone gas flow 150 liter h⫺1, des-

4908

aem.asm.org

RESULTS

Rock phosphate characterization. The chemical analyses revealed a complex constitution of the Araxá RP (Table 1). Among 66 elements investigated, only Cl, Hg, and Re were not detected. The RP contained 13.97% P, with a molar Ca:P ratio of 1.67. This value is consistent with the theoretical molar ratio of apatite [Ca5(PO4)3], showing that P was predominantly linked to Ca. However, only 4% of the total P was soluble in 2% citric acid. The chemical analyses also showed considerable concentrations of rare earth elements (Ce, Dy, Er, Eu, Gd, Ho, La, Lu, Nd, Pr, Sc, Sm, Tb, Tm, Y, and Yb) that probably became part of the apatite structure during rock crystallization (7). Based on XRD and chemical analysis, the RP was classified as a fluorapatite with some mixture of hydroxyapatite, with the theoretical formula Ca5(PO4)3(F,OH). Kinetics of rock phosphate solubilization. High growth rates and a quick drop in the pH of the medium from 5.5 to 3.2 occurred in the first 36 h of incubation (Fig. 1). The concentration of solubilized P increased rapidly in the first 60 h, reaching approximately 80 mg liter⫺1. Afterwards, the pH dropped slightly, and the P concentrations increased and decreased to different extents at irregular intervals. The biomass continued to grow at a lower rate after the first 60 h. Fungal biomass increased during four successive intervals (h 0 to 36, 48 to 120, 132 to 180, and 192 to 240) over the course of the experiment (Fig. 1). The first hours of the second (h 48 to 120) and the third (h 132 to 180) growth intervals coincided with the decreases in solubilized P in the medium. During RP solubilization, 17 chemical elements were released (Table 2) and most presented a pattern similar to that of P (Fig. 1 and 2). The correlations between the concentrations of these elements and that of P were higher than 0.7 (P ⬍ 0.01) (see Table S2 in the supplemental material). Low concentrations of B, Mo, Ni, Pb, Rb, V, Zn, and Zr were detected in the medium during incubation (see Table S3 in the supplemental material), reflecting their low concentrations in the Araxá RP (Table 1). In fact, at an initial RP quantity of 3 g liter⫺1, these elements, except for B and Zr, were expected to be released into the medium at concentrations lower than 1 mg liter⫺1. As, Be, Cd, Co, Cr, Cu, Ga, and Li were not released in detectable quantities. The concentrations of K, Mg, and S in the medium after RP dissolution were also determined, but the data are not presented here because these elements are also constituents of the NBRIP medium and showed practically no variation during incubation. Screening of chemical elements affecting rock phosphate solubilization. Due to the higher solubilization rate observed during the first 60 h of incubation and the fluctuations in solubilized P observed thereafter (Fig. 1), the study of the effects of the released elements on RP solubilization was limited to that time interval. For this, a stressful condition was established by adding each element to the medium at the highest concentration recorded during RP solubilization (Table 2). Given that the concentration of each element rose from 0 to a maximum value concomitantly with the increases in P concentrations, the amount of each element initially added to the medium for the screening was higher than the actual concentration that would be found at a particular time and without supplementation. This overestimation, however, was intentionally introduced to facilitate the identification of potentially inhibitory elements that might exert their effects at some point



FIG 1 Solubilized phosphorus, changes in pH, and biomass accumulation during Araxá rock phosphate solubilization by Aspergillus niger. Values are the means of the results of three replicates. Error bars denote the standard deviations.

during the solubilization of Araxá RP. If a given element was not inhibitory at the maximum concentration used in the screening, it would not be inhibitory at lower concentrations. Among the 17 elements screened, only F and Sr exerted significant effects on the level of solubilized P (Fig. 3). A positive regression coefficient was observed for Sr, indicating that it stimulates RP solubilization. On the other hand, the regression coefficient for F was negative and presented a high value, resulting in a decrease of 81% in the mean level of solubilized P in the treatments where fluoride was added at level 1 (Fig. 3). This strong inhibitory effect of F is evident in comparisons of the values of solubilized P in the treatments with and without this element (see Table S1 in the supplemental material). Effect of fluoride on RP solubilization and metabolism of A. niger. Given the observations on the inhibitory effect of fluoride, another set of experiments was performed to evaluate the effects of different fluoride doses on RP solubilization. Besides Araxá RP, a soluble P source, K2HPO4, was used to evaluate the effects of fluoride on the metabolic processes involved in RP solubilization in a less complex medium. The results clearly reflected decreases in medium acidification and RP solubilization when fluoride doses were increased (Fig. 4a). Additionally, NaF and KF were compared in order to rule out a possible effect of the counter-ion in the fluoride source, but no difference in RP solubilization was detected between the two salts (data not shown). At the dose corresponding to the maximal fluoride concentration detected during the solubilization process (22.9 mg fluoride per liter; Table 2), a decrease of about 55% in solubilized P was estimated from the adjusted regression equation (Fig. 4a). At this fluoride concentration, a decrease of about 75% in fungal growth was found (Fig. 4b). Furthermore, increasing fluoride concentrations resulted in a lower yield of solubilized P per unit of biomass (YP/X) (Fig. 4b). The data show that the production of citric acid was almost completely inhibited at fluoride concentrations higher than 20 mg


liter⫺1 whereas the production of gluconic and oxalic acids was stimulated by concentrations of up to 35 mg liter⫺1 (Fig. 4c). DISCUSSION

The release of chemical elements concurrently with P release during microbial RP solubilization has received little attention in past studies. In this work, it has been clearly demonstrated that various chemical elements are released together with P. Among them, F significantly lowered the levels of solubilized P. Fluoride is toxic to microbial cells, affecting a series of physiological processes (22). Fluoride toxicity to bacteria and fungi results most likely from blocking the functions of enzymes (23) such as enolase (24), peroxidase (25), heme oxidases (22), ATPases (26), phosphatases (22), and copper enzymes such as polyphenol oxidases (27). The inhibition results from direct HF/F⫺ binding or by metal-F complex binding (22). Under the conditions reported here, fluoride affects metabolic processes directly involved in RP solubilization. As reported for other fungal species (23, 28), the increase in fluoride concentrations resulted in less growth of A. niger. Furthermore, at high fluoride doses, the biomass became less effective at RP solubilization, given that the amount of P solubilized per unit of biomass was less (Fig. 4b). These observations are presumably related to the lower medium acidification and lower production of citric acid detected at higher fluoride doses. As these changes affected the solubilization agents, namely, H⫹ and citric acid levels, it is reasonable to conclude that the concurrent mobilization of fluoride during the solubilization of Araxá RP decreased the efficiency of the process. Different responses were observed for the production of organic acids by the isolate A. niger FS1 when exposed to fluoride. The production of citric acid, one of the most effective agents for the release of P from RP (29), was almost completely inhibited at the highest fluoride concentration (22.9 mg liter⫺1) detected dur-

aem.asm.org 4909

Mendes et al.

FIG 2 Release of chemical elements during the solubilization of Araxá rock phosphate by Aspergillus niger.

4910

aem.asm.org



FIG 3 Effects of chemical elements on the solubilization of Araxá rock phosphate (RP) by Aspergillus niger. Data represent the means of solubilized P data for each element at the levels ⫺1 (absence of the element) and 1 (maximum concentration of the element achieved during RP solubilization; Table 2). The linear regression coefficient of the element is presented at the top of each graph (R2 of regression: 0.88). *, significant as determined by t test (P ⬍ 0.05).

ing the experiment on RP solubilization kinetics (Fig. 4c). Agrawal et al. (30) reported similar results and suggested that the decrease in citric acid production by NaF resulted from the inhibition of enolase activity, which is involved in the conversion of 2-phosphoglycerate (2-PG) to phosphoenolpyruvate (PEP), thus disrupting the supply of precursors for citric acid production. Interestingly, the production of gluconic and oxalic acids was stimulated by increasing the fluoride concentration to 35 mg liter⫺1 (Fig. 4c). The synthesis of gluconic acid is catalyzed by the extracellular enzyme glucose oxidase (GOD), which converts glucose into gluconic acid (31). Thus, the positive effect of fluoride on the production of gluconic acid may result from the surplus of glucose in the medium due to the inhibition of fungal growth. However, the reason for the stimulatory effect of fluoride on oxalic acid production remains unclear. The synthesis of this organic acid is catalyzed by the enzyme oxaloacetase, which converts oxaloacetate into oxalate and acetate. The reaction takes place in the cytoplasm and does not involve the tricarboxylic acid (TCA) cycle, since A. niger is capable of forming oxaloacetate from pyruvate and CO2 through a cytoplasmic, constitutive pyruvate carboxylase (32). Given the putative partial inhibition of glycolysis and, consequently, of the TCA cycle, the supply of oxaloacetate precursors must come from alternative sources, a hypothetical one being the gluconic acid produced. Aspergillus niger possesses a modified (nonphosphorylating) Entner-Doudoroff pathway in which gluconate is converted to glyceraldehyde and pyruvate in two steps (33). Pyruvate could be subsequently converted into oxaloacetate by pyruvate carboxylase and, then, oxaloacetate


cleaved into oxalate and acetate by oxaloacetase. Müller (34) demonstrated that A. niger can use gluconic acid as a C source and accumulate oxalate as an end product. However, that author failed to detect the enzymes of the Entner-Doudoroff pathway and of the nonphosphorylating Entner-Doudoroff system in cell extracts of his A. niger strain (35). In contrast to Elzainy et al. (33), Müller (35) added gluconic acid to the medium after fungal growth on glucose, which explains the absence of the enzymes that were shown to be inducible by gluconate (33). The conversion of glucose into gluconic acid and its subsequent use through the nonphosphorylating Entner-Doudoroff system could be an alternative means for A. niger to overcome the inhibition of glycolysis caused by fluoride. Further studies are necessary to confirm this hypothesis. In contrast to fluoride, Sr had a positive effect on RP solubilization. This element has no apparent biological function, being nonspecifically accumulated in the biomass of filamentous fungi and yeasts (36). It can act as a Ca analogue in some situations (37) and can mitigate the inhibitory effects of Na on fungal growth (38). However, the Na concentration found in the medium (Table 2) was not high enough to inhibit fungal growth (38). The effect of Sr was probably more closely related to the chemical equilibrium in the medium. Since Sr stimulated RP solubilization, albeit at a low level, an exhaustive exploration of this issue was not a concern in the present work. The dynamic variations of the medium conditions due to changes in the A. niger metabolism and in the chemical equilibria are probably the reasons for the variations in the solubilized P

aem.asm.org 4911

Mendes et al.

FIG 4 Effect of fluoride on the solubilization of Araxá rock phosphate and the metabolism of Aspergillus niger. (a) Solubilized P and medium pH after 60 h of cultivation in NBRIP medium supplemented with 3 g liter⫺1 RP. (b) Fungal biomass and P/biomass yield (YP/X ⫽ mg solubilized P per g of biomass). (c) Organic acids produced after 60 h of cultivation in NBRIP medium supplemented with K2HPO4. All regression coefficients are significant as determined by t test (P ⬍ 0.01).

concentrations observed throughout the incubation (Fig. 1). Vassilev et al. (39) observed that decreases in soluble P in the fermentation medium were accompanied by decreases in titratable acidity and suggested that this resulted from the consumption of organic acid by the fungus under conditions of C depletion. The data obtained in our work support this hypothesis since the decreases in soluble P levels apparently occurred in response to the beginning of a new growth cycle, when the fungus may have used part of the organic acids in its metabolism. Organic acids affect P solubility by forming complexes with metal cations in solution, thereby avoiding the precipitation of metal phosphates (11). Furthermore, Illmer and Schinner (12) showed that changes in the medium conditions during the solubilization of P-Ca minerals

4912

aem.asm.org

(brushite and apatite) can lead to P reprecipitation. The consumption of organic acids probably triggers the reprecipitation of metal phosphates and, as a consequence, leads to a decrease in soluble P. The absence of variation in the pH concomitant with these reactions reinforces the hypothesis that the changes in solubilized P depend mainly on the complexation of metal cations by organic acids. The fluctuations in the concentrations of some of the elements released from Araxá RP followed a pattern similar to that of P (Fig. 1 and 2). Al, Ba, Ca, Fe, and Sr can form complexes of low solubility with P (40) and could be involved in P precipitation during the periods of organic acid consumption discussed above. In the case of F, the reasons for the decreases in its concentrations are not clear. As F is found predominantly as fluoride anions (F⫺) in solution, one possible explanation is that it could form complexes of low solubility with Ca2⫹ (CaF2) or Al3⫹ (AlF3) (41), which presumably would be released when organic acids are consumed. Thus, during the dissolution of Araxá RP, cycles of solubilization and precipitation of some ion pairs probably occur in accordance with their solubility. However, the extensive exploration of chemical equilibria in the medium was beyond the scope of this work. This is the first report showing the inhibitory effect of fluoride on RP solubilization. It contributes to an understanding of the pronounced decrease in fungal solubilization of RPs compared to that of pure synthetic apatite (9). The release of fluoride during microbial RP solubilization has been ignored so far, even though most RPs contain large amounts of fluoride. In fact, RP constitutes the main natural reserves of F (41). These findings open new avenues for improving RP solubilization efficiency through strategies for removing fluoride during microbial solubilization or by selecting more fluoride-tolerant strains. Conclusions. Among the various chemical elements mobilized during the solubilization of Araxá RP by A. niger, only fluoride significantly lowered solubilization efficiency. Fluoride decreased fungal growth, citric acid production, medium acidification, and P solubilization. The data from this study show that fluoride limits the solubilization of Araxá RP by A. niger by negatively affecting metabolic processes involved in phosphate solubilization. Given the ubiquitous distribution of fluoride in RPs, most microbial RP solubilization systems studied so far have probably been operated under suboptimal conditions. ACKNOWLEDGMENTS We are grateful to Maurício P. F. Fontes for his assistance in XRD analysis. We are also thankful to the National Council for Scientific and Technological Development (CNPq) for financing this work and providing scholarships to Gilberto de Oliveira Mendes and Maurício Dutra Costa. Financial support for this study was also provided by Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG), project CAGAPQ-00712-12, and the Spanish projects CTM2011-027797 and P09RNM-5196.

REFERENCES 1. Vassileva M, Serrano M, Bravo V, Jurado E, Nikolaeva I, Martos V, Vassilev N. 2010. Multifunctional properties of phosphate-solubilizing microorganisms grown on agro-industrial wastes in fermentation and soil conditions. Appl. Microbiol. Biotechnol. 85:1287–1299. 2. Vassilev N, Franco I, Vassileva M, Azcon R. 1996. Improved plant growth with rock phosphate solubilized by Aspergillus niger grown on sugar-beet waste. Bioresour. Technol. 55:237–241. 3. Jain R, Saxena J, Sharma V. 2010. The evaluation of free and encapsulated Aspergillus awamori for phosphate solubilization in fermentation and soil-plant system. Appl. Soil Ecol. 46:90 –94.



4. Cordell D, White S. 2011. Peak phosphorus: clarifying the key issues of a vigorous debate about long-term phosphorus security. Sustainability 3:2027–2049. 5. Vassilev N, Martos E, Mendes G, Martos V, Vassileva M. 2013. Biochar of animal origin: a sustainable solution to the global problem of highgrade rock phosphate scarcity? J. Sci. Food Agric. 93:1799 –1804. 6. Smith EA, Mayfield CI, Wong PTS. 1977. Physical and chemical characterization of selected natural apatites in synthetic and natural aqueous solutions. Water Air Soil Pollut. 8:401– 415. 7. Nash WP. 1972. Apatite chemistry and phosphorus fugacity in a differentiated igneous intrusion. Am. Mineral. 57:877– 886. 8. Reyes I, Baziramakenga R, Bernier L, Antoun H. 2001. Solubilization of phosphate rocks and minerals by a wild-type strain and two UV-induced mutants of Penicillium rugulosum. Soil Biol. Biochem. 33:1741–1747. 9. Schneider KD, van Straaten P, de Orduna RM, Glasauer S, Trevors J, Fallow D, Smith PS. 2010. Comparing phosphorus mobilization strategies using Aspergillus niger for the mineral dissolution of three phosphate rocks. J. Appl. Microbiol. 108:366 –374. 10. Banik S, Dey B. 1982. Available phosphate content of an alluvial soil as influenced by inoculation of some isolated phosphate-solubilizing microorganisms. Plant Soil 69:353–364. 11. Fox TR, Comerford NB, McFee WW. 1990. Kinetics of phosphorus release from spodosols: effects of oxalate and formate. Soil Sci. Soc. Am. J. 54:1441–1447. 12. Illmer P, Schinner F. 1995. Solubilization of inorganic calcium phosphates—solubilization mechanisms. Soil Biol. Biochem. 27:257–263. 13. Gadd GM. 1993. Interactions of fungi with toxic metals. New Phytol. 124:25– 60. 14. Gadd GM. 1999. Fungal production of citric and oxalic acid: importance in metal speciation, physiology and biogeochemical processes. Adv. Microb. Physiol. 41:47–92. 15. Shu P, Johnson MJ. 1948. The interdependence of medium constituents in citric acid production by submerged fermentation. J. Bacteriol. 56:577– 585. 16. Nautiyal CS. 1999. An efficient microbiological growth medium for screening phosphate solubilizing microorganisms. FEMS Microbiol. Lett. 170:265–270. 17. Burman JO, Ponter C, Bostrom K. 1978. Metaborate digestion procedure for inductively coupled plasma-optical emission spectrometry. Anal. Chem. 50:679 – 680. 18. Braga JM, Defelipo BV. 1974. Determinação espectrofotométrica de fósforo em extratos de solo e material vegetal. R. Ceres 21:73– 85. 19. Reyes I, Bernier L, Simard RR, Antoun H. 1999. Effect of nitrogen source on the solubilization of different inorganic phosphates by an isolate of Penicillium rugulosum and two UV-induced mutants. FEMS Microbiol. Ecol. 28:281–290. 20. Plackett RL, Burman JP. 1946. The design of optimum multifactorial experiments. Biometrika 33:305–325. 21. Mendes GO, Freitas ALM, Pereira OL, Silva IR, Vassilev NB, Costa MD. 19 May 2013, posting date. Mechanisms of phosphate solubilization by fungal isolates when exposed to different P sources. Ann. Microbiol. doi: 10.1007/s13213-013-0656-3. 22. Marquis RE, Clock SA, Mota-Meira M. 2003. Fluoride and organic weak


23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41.

acids as modulators of microbial physiology. FEMS Microbiol. Rev. 26: 493–510. Li S, Breaker RR. 2012. Fluoride enhances the activity of fungicides that destabilize cell membranes. Bioorg. Med. Chem. Lett. 22:3317–3322. Kustrzeba-Wójcicka I, Golczak M. 2000. Enolase from Candida albicans—purification and characterization. Comp. Biochem. Physiol. B Biochem. Mol. Biol. 126:109 –120. Thibodeau EA, Bowen WH, Marquis RE. 1985. pH-dependent fluoride inhibition of peroxidase activity. J. Dent. Res. 64:1211–1213. Sutton SV, Bender GR, Marquis RE. 1987. Fluoride inhibition of protontranslocating ATPases of oral bacteria. Infect. Immun. 55:2597–2603. Fernández EVA, Sanchez-Amat A, Solano F. 1999. Location and catalytic characteristics of a multipotent bacterial polyphenol oxidase. Pigm. Cell Res. 12:331–339. Treshow M. 1965. Responses of some pathogenic fungi to sodium fluoride. Mycologia 57:216 –221. Kpomblekou-A K, Tabatabai MA. 1994. Effect of organic acids on release of phosphorus from phosphate rocks. Soil Sci. 158:442– 453. Agrawal PK, Bhatt CS, Viswanathan L. 1983. Effect of some metabolic inhibitors on citric acid production by Aspergillus niger. Enzyme Microb. Technol. 5:373–376. Mischak H, Kubicek CP, Röhr M. 1985. Formation and location of glucose oxidase in citric acid producing mycelia of Aspergillus niger. Appl. Microbiol. Biotechnol. 21:27–31. Kubicek CP, Schreferl-Kunar G, Wohrer W, Rohr M. 1988. Evidence for a cytoplasmic pathway of oxalate biosynthesis in Aspergillus niger. Appl. Environ. Microbiol. 54:633– 637. Elzainy TA, Hassan MM, Allam AM. 1973. New pathway for nonphosphorylated degradation of gluconate by Aspergillus niger. J. Bacteriol. 114: 457– 459. Müller HM. 1986. Utilization of gluconate by Aspergillus niger. II. Enzymes of degradation pathways and main end products. Zentralbl. Mikrobiol. 141:461– 469. Müller HM. 1985. Utilization of gluconate by Aspergillus niger. I. Enzymes of phosphorylating and nonphosphorylating pathways. Zentralbl. Mikrobiol. 140:475– 484. Rome L, Gadd G. 1991. Use of pelleted and immobilized yeast and fungal biomass for heavy metal and radionuclide recovery. J. Ind. Microbiol. 7:97–104. Botton B. 1978. Influence of calcium on the differentiation and growth of aggregated organs in Sphaerostilbe repens. Can. J. Microbiol. 24:1039 – 1047. Jones EBG, Jennings DH. 1965. The effect of cations on the growth of fungi. New Phytol. 64:86 –100. Vassilev N, Baca MT, Vassileva M, Franco I, Azcon R. 1995. Rock phosphate solubilization by Aspergillus niger grown on sugar-beet waste medium. Appl. Microbiol. Biotechnol. 44:546 –549. Lide DR. 2004. CRC Handbook of chemistry and physics, 85th ed. CRC Press, Boca Raton, FL. Aigueperse J, Mollard P, Devilliers D, Chemla M, Faron R, Romano R, Cuer JP. 2000. Fluorine compounds, inorganic, p 397– 441. Ullmann’s encyclopedia of industrial chemistry, vol 15. Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany.

aem.asm.org 4913