Chemosphere 134 (2015) 1–6

Contents lists available at ScienceDirect

Chemosphere

journal homepage: www.elsevier .com/locate /chemosphere

Phosphorus solubilization and plant growth enhancementby arsenic-resistant bacteria

http://dx.doi.org/10.1016/j.chemosphere.2015.03.0480045-6535/Published by Elsevier Ltd.

⇑ Corresponding author at: State Key Laboratory of Pollution Control andResource Reuse, School of the Environment, Nanjing University, Jiangsu 210046,China.

E-mail address: lqma@ufl.edu (L.Q. Ma).

Piyasa Ghosh a,c, Bala Rathinasabapathi b, Lena Q. Ma a,c,⇑a State Key Laboratory of Pollution Control and Resource Reuse, School of the Environment, Nanjing University, Jiangsu 210046, Chinab Horticultural Sciences Department, University of Florida, Gainesville, FL 32611, United Statesc Soil and Water Science Department, University of Florida, Gainesville, FL 32611, United States

h i g h l i g h t s

� Arsenic-resistant bacteria were fromrhizosphere of arsenic-hyperaccumulator Pteris vittata.� They produced siderophores and

phytase to solubilize FePO4 andphytate and release P and Fe.� They enhanced tomato Fe and P

uptake and biomass accumulation.

g r a p h i c a l a b s t r a c t

Solubilize FePO4 /phytate Plant growth enhancementBacterial siderophores

a r t i c l e i n f o

Article history:Received 5 February 2015Received in revised form 21 March 2015Accepted 23 March 2015

Handling Editor: J. de Boer

Keywords:Arsenic-resistant bacteriaFePO4 solubilizationPhytate solubilizationTomato growth promotion

a b s t r a c t

Phosphorus is an essential nutrient, which is limited in most soils. The P solubilization and growthenhancement ability of seven arsenic-resistant bacteria (ARB), which were isolated from arsenic hyper-accumulator Pteris vittata, was investigated. Siderophore-producing ARB (PG4, 5, 6, 9, 10, 12 and 16) wereeffective in solubilizing P from inorganic minerals FePO4 and phosphate rock, and organic phytate. Toreduce bacterial P uptake we used filter-sterilized Hoagland medium containing siderophores or phytaseproduced by PG12 or PG6 to grow tomato plants supplied with FePO4 or phytate. To confirm that side-rophores were responsible for P release, we compared the mutants of siderophore-producing bacteriumPseudomonas fluorescens Pf5 (PchA) impaired in siderophore production with the wild type and teststrains. After 7 d of growth, mutant PchA solubilized 10-times less P than strain PG12, which increasedtomato root biomass by 1.7 times. For phytate solubilization by PG6, tomato shoot biomass increasedby 44% than control bacterium Pseudomonas chlororaphis. P solubilization by ARB from P. vittata maybe useful in enhancing plant growth and nutrition in other crop plants.

Published by Elsevier Ltd.

1. Introduction

Phosphorus (P) is an essential nutrient required for optimumgrowth and development in living organisms including plantsand bacteria (Huang et al., 2004). Although natural soil containslarge amount of P (400–1200 mg kg�1), water-soluble P is only�1 mg kg�1 (Rodriguez and Fraga, 1999). Large reserve of insoluble

P accumulates in soils from application of P fertilizers. In acidicsoils, P is immobilized by Fe/Al minerals while in alkaline soils itis trapped by Ca/Mg carbonates. Organic P reserve in soil rangesfrom 5% to 95% of the total soil P, of which phytate (inositol phos-phate) constitutes up to 50% (Rodriguez and Fraga, 1999). In soils,microbes can solubilize both inorganic and organic P by producingorganic exudates such as organic acids, siderophore and phosphatesolubilizing enzymes like phytase (Hayes et al., 2000; Li et al.,2006). It has been shown that plant growth promoting bacteria likePseudomonas, Bacillus, and Burkholderia have the ability to sol-ubilize P and make it more available to plants (Glick, 1995;Rodriguez and Fraga, 1999).

2 P. Ghosh et al. / Chemosphere 134 (2015) 1–6

Seven As-resistant bacteria (ARB) have been isolated from therhizosphere of As-hyperaccumulator Pteris vittata from the fieldin North Central Florida (Ghosh et al., 2011). P. vittata prefers togrow in alkaline soils, which are characterized by low P and Feavailability and/or high concentrations of As (Lessl and Ma,2013). Phosphorus and As compete for uptake into bacterial cells,so bacteria have developed an effective system to increase P solu-bility from soils. Bacterial siderophores effectively release As frominsoluble minerals like FeAsO4 and AlAsO4 (Ghosh et al., 2011). Inaddition to inorganic P, a major source of organic P in plant rhizo-sphere is phytate so the studied bacteria may also produce phytaseto extract P from phytate.

Siderophores are low molecular weight organic ligands withhigh affinity and specificity for binding Fe. The stability constantsof Fe3+-siderophore complexes range between 1023 to 1052

(Albrecht-Gary and Crumbliss, 1998; Ams et al., 2002). Bacterialability to produce siderophores with high affinity for Fe3+ givesthem selective advantage under Fe-limited conditions such as alka-line soils (Kraemer, 2004; Huang et al., 2004). Most siderophoresbind strongly to trivalent ions like Fe3+ and Al3+, but their abilityin binding divalent cations like Ca2+ is weak, with stability constantof only 102.64 (Kraemer, 2004).

In this study we tested the ability of 7 ARB from the rhizosphereof P. vittata growing in As-rich soils in solubilizing P from insolubleminerals like FePO4 and phosphate rock [Ca10(PO4)6F2], and fromorganic P like phytate. These bacteria were selected because theywere isolated from arsenic-rich soil with low Fe and P availability.The major objectives of this research were to assess the ability ofsiderophores and phytase produced by arsenic-resistant bacteriain: (1) solubilizing P from inorganic minerals like FePO4 and phos-phate rock, and organic P like phytate, and (2) enhancing growth intomato by increasing P and Fe uptake from FePO4 and phytate.

2. Materials and methods

2.1. Arsenic-resistant bacteria and control bacteria

The 7 arsenic-resistant bacteria (ARB; PG4, 5, 6, 9, 10, 12 and 16)that were used for this study had different siderophore-producingability ranging from 9.47 to 115 lM DFOM (deferroxaminemesylate) equiv./OD of cells (Ghosh et al., 2015). All bacteria wereidentified as Pseudomonas sp. by 16S rRNA sequencing. We alsoselected a control bacterium from the same genera so the resultscould be compared with the test strains. All bacteria were grownin modified LB media (pH 7) because the isolated bacteria werecharacterized in this media (Ghosh et al., 2015). We used 7 ARBfor phosphate rock, FePO4 and phytate solubilization experimentsand two bacteria (PG12 and PG6) with different siderophore andphytase producing ability for tomato growth experiment.

For comparison, soil bacterium Pseudomonas chlororaphis 63-28(PC), which was isolated from the rhizosphere of canola grown innon-contaminated soil (Paulitz et al., 2000), was also included. Inaddition to PC, we used wild type and two mutants ofPseudomonas fluorescens Pf-5 (PF) (PchA and PchC) to comparebetween siderophore-producing and non-producing mutants ofthe same bacteria. Mutant PchA is completely impaired in sidero-phore production whereas PchC is partially impaired (Hartneyet al., 2011).

2.2. Bacterial solubilization of FePO4, phosphate rock, and phytate

For the P solubilization experiments we used three insoluble Pcompounds: FePO4, phosphate rock, and phytate. Ground concen-trate phosphate rock [Ca10(PO4)6F2] was obtained from Whitesprings, Florida (Lessl and Ma, 2013). FePO4 as FePO4�H2O and

phytate as phytic acid sodium salt hydrate from rice(C6H18Na12O24P6) were purchased from Sigma–Aldrich.

Extracellularly-produced siderophores from 7 ARB and a controlbacterium PC were collected from their spent growth media.Specifically, the bacteria were grown in modified LB medium for24 h at 30 �C under 200 rpm shaking condition (Ghosh et al.,2011). Then the spent medium was filtered through 0.2 lM filterto remove bacteria. This filtered fraction of the spent media wasagain divided into two parts. One half was boiled for 5 min to inac-tivate enzyme activity and the other half was left intact.

To test bacterial ability in solubilizing P, 1 mL of spent mediumfrom the 7 ARB was incubated with 0.1 g of phosphate rock (PR),0.01 g of FePO4 (FP) or 0.04 g phytate (PA) for 24 h at 30 �C under200 rpm shaking condition. The amounts of insoluble P were usedto ensure sufficient surface exposure with the spent growth med-ium for effective P solubilization. To test whether P was solubilizedby siderophores via Fe chelation, we used 1 mL of the filter-steri-lized spent medium of wild type and mutants PchA and PchC(Hartney et al., 2011) and incubated it with 0.01 g FePO4 for 24 hat 30 �C. The water soluble P was measured using modified molyb-denum blue method (Ghosh et al., 2011).

2.3. Bacterial enhancement of tomato growth

One-month old tomato seedlings were transferred to0.2-strength Hoagland nutrient solution (HNS) at pH 7. The 0.2�HNS contained 6.2 mg L�1 P and 0.2 mg L�1 Fe. For the FePO4

experiment, the 0.2� HNS was provided with only 1 mg L�1 Pand no Fe (low P � Fe). For the phytate experiment, the 0.2�HNS was provided with 1 mg L�1 P and 0.2 mg L�1 Fe (lowP + Fe). The tomato seedlings were acclimated in 0.2� HNS with(1) low P � Fe or (2) low P + Fe for 35 d. The pH was checked regu-larly and the volume was maintained to 1 L in aerated containersduring the experiment.

For the FePO4 experiment, we selected two bacterial strains,PG12 (siderophore-producing ARB) and mutant PchA impaired insiderophore-production. The control was selected so that we cansee the effect of siderophore in the bacterial spent media. For thephytate experiment, we selected two bacterial strains, PG6 (high-est phytate solubilization among 7 ARB) and control strain P.chlororaphis (low phytate solubilization). Tomato plants acclimatedin 0.2� HNS with low P and 0.2 mg L�1 Fe received 20 mL of spentmedium from PG12 or PchA mutant and 0.2 g of phytate. Thetomato plants acclimated in 0.2� HNS with low P and no Fereceived 20 mL of spent medium from PG6 or PC and 0.2 g ofFePO4. The plants were then grown for 1 week with 3 replicates.

After 7-d of growth, plant biomass was determined after dryingin oven at 80 �C for 2 d. To determine the role of P in enhancingplant growth, the dried plant samples were digested usingH2SO4/H2O2 and was analyzed for P by modified molybdenum bluemethod (Ghosh et al., 2011). Total Fe concentration in the plantbiomass was measured in the same extracts using inductively-cou-pled plasma emission spectroscopy (PerkinElmer 5300DV,Waltham, MA) via EPA Method 2007.

2.4. Statistical analysis

All plant and bacterial experiments were conducted with threereplicates for each treatment and every bacterial experiment wasrepeated twice. The analysis of variance (ANOVA) and Tukey’smean grouping were used to determine significance differencebetween the treatment means. All statistical analyses were per-formed using SAS statistical software (SAS Inst., Cary, NorthCarolina, USA).

P. Ghosh et al. / Chemosphere 134 (2015) 1–6 3

3. Results and discussion

3.1. Bacterial solubilization of phosphate rock and FePO4

The spent medium was collected after growing bacteria for24 h, which contained bacterial exudates without bacteria includ-ing organic acids, siderophore and phytase. Since those bacteriawere isolated from an arsenic-rich soil with low Fe and P availabil-ity, it was expected that those bacteria have better ability tosolubilize Fe and P compared to typical bacteria. To determine bac-terial ability in solubilizing insoluble P mineral phosphate rock, wemeasured the P concentration in the spent media with and withoutaddition of P mineral. The spent media without phosphate rockcontained 0.08–0.74 mg L�1 P (data not shown). Among the 7ARB tested, only 5 strains (PG4, 5, 9, 10, and 12) solubilized moreP than the control bacterium PC (Table 1). PC was used a controlbecause its P solubilization would be similar to other rhizospherepseudomonad. The strains PG5 and PG12 were the most efficientin solubilizing P from phosphate rock (1.36–2.40 mg L�1) whilePG16 and PG6 were unable to solubilize P (Table 1).

To distinguish between enzymatic and siderophore-mediated Psolubilization, the spent medium was boiled for 5 min to deacti-vate enzyme activity because siderophores are relatively tolerantto temperature (Kraemer, 2004). The 5 strains lost 58–84% of itsP solubilization ability after boiling, indicating the importance ofenzymes in P solubilization (Table 1). However, they all solubilizedP after boiling, but it was less than unboiled extracts. The resultsindicated that both enzymatic and non-enzymatic processes con-tributed to P solubilization from phosphate rock by bacterial spentmedia. Similar results were also found in a study by Altomare et al.(1999) on a plant growth promoting fungi Trichoderma harzianumRifai 1295-22. The cell-free extracts from the fungi solubilized 5times more P from phosphate rock than the filtrate with fungalcells, which is attributed to chelating substances present in thespent media. The reduction in P solubilization in the filtrate withfungal cells may be due to the P uptake by live fungal cells afterP solubilization. Based on previous studies, the primary mode ofP solubilization by microbes is by production of organic acids(Vassilev et al., 2006; Chen et al., 2006), which is supported bypH reduction in the solution. Chen et al. (2006) also found aninverse relationship between the pH and P solubilization by bacter-ial strains Arthrobacter sp., Bacillus megaterium, and Serratiamarcescens. The decrease in pH (6.8–7 to 4.9–6) was due to theexudation of organic acids like citric, gluconic, succinic and lacticacid. Another study by Vassilev et al. (2006) also proved that a fun-gal strain Aspergillus niger efficiently solubilized P (597 mg kg�1)from phosphate rock. In this case the pH was also lowered by theexudates produced by the fungus from 7 to 2.9. However, we did

Table 1Siderophore production and phosphate solubilized in 1 mL spent growth media by 7 arseni1 d of incubation. The values represent mean ± SE of three replicates.

Phosphate compounds Arsenic-resistant bacteria

PC PG16 PG1

Siderophore (lMol DFOM equiv./OD value of cells) 18.6 ± 0.03 15.8 ± 0.02 9.47mg L�1

Phosphate rock 0.46 ± 0.03 00 1.05Enzymatic (%) 43 – 71FePO4 3.30 ± 0.01 3.90 ± 0.14 3.38Enzymatic (%) 40 17 12Phytate 0.19 ± 0.01 0.76 ± 0.03 0.53Enzymatic (%) 11 80 45

Total Pb 3.58 ± 0.03A 4.66 ± 0.2B 4.91

a Not determined as the bacteria failed to grow in the CAS assay medium.b Total = Sum of P solubilized by each bacteria from all three insoluble forms, phosph

not find significant pH change in our experiment, indicating thatthe solubilization of phosphate rock was probably not attributedto organic acid exudates. Overall, the amount of P solubilized fromphosphate rock was limited partially due to the low chelationcapacity of siderophores with Ca (Kraemer, 2004), so phosphaterock was not included in the further experiments.

In addition to phosphate rock, we also determined the ability ofbacterial spent media in solubilizing P from FePO4 (Table 1). Unlikephosphate rock, spent media from all 7 bacteria solubilized P fromFePO4. Bacterial strain PG10 solubilized the least amount of P at3.38 mg L�1 and PG12 the most P at 9.05 mg L�1 (Table 1). Forexample, PG12 solubilized 3.8-fold more P from FePO4 than phos-phate rock (Table 1).

We hypothesized that bacterial siderophores were partiallyresponsible for P solubilization from FePO4. This hypothesiswas supported by the strong correlation (r2 = 0.94) between bac-terial siderophore-production and P solubilization from FePO4

(Table 1). After boiling, the spent media of PG4 did not lose its abil-ity whereas other 6 strains lost 7.2–30% of their ability in P sol-ubilization (Table 1). Though PG 5 and 12 were most efficient inP solubilization from both phosphate rock (65–84% enzymatic)and FePO4 (27–30% enzymatic), siderophore probably played amore important role in solubilizing FePO4 (Table 1). However, thedata indicated that, besides siderophores, enzymatic process alsoplayed a role in P solubilization. Extracellular enzymes exudedby rhizobacteria have been found to improve plant growth and Puptake (Rodriguez et al., 2006).

To further test our hypothesis that P was solubilized by bacter-ial siderophores, we tested wild type of P. fluorescens and twomutants PchA and PchC. While mutant PchA was completelyimpaired in siderophore production mutant PchC was partiallyimpaired. The wild type (4.4 mg L�1) solubilized 2-fold and 4-foldmore P than mutants PchC and PchA, which was consistent withtheir ability in siderophore production (Fig. 1). However, the factthat PchA solubilized 1.0 mg L�1 P indicated that other processeswere also responsible for P solubilization from FePO4. Boilingreduced their ability of P solubilization by 22–33%, again implyinglimited enzyme-mediated P solubilization. The highest amount ofenzymatic solubilization occurred in the filtered spent media ofPG12 at 84% for phosphate rock and PG5 at 30% for FePO4

(Table 1). Apparently, different bacteria showed different enzy-matic ability in solubilizing organic and inorganic P compounds.

Fe-siderophore complex is more stable compared to that of Ca-siderophore (Albrecht-Gary and Crumbliss, 1998; Ams et al., 2002).For example, Fe3+-complex of trihydroxamate siderophore desfer-rioxamine-B (DFO-B) has a 1:1 stability constant of 1031 while thatof Ca complex is only k = 102.6 (Martell et al., 2001). This was con-sistent with the lower P solubilized by bacterial spent media from

c-resistant bacteria from 0.01 g FePO4, 0.1 g phosphate rock and 0.04 g of phytate after

0 PG9 PG6 PG4 PG5 PG12

± 0.02 10.8 ± 0.04 22.7 ± 0.01 NDa 73.2 ± 0.01 115 ± 0.02

± 0.03 1.05 ± 0.15 00 0.51 ± 0.06 1.36 ± 0.03 2.41 ± 0.1577 – 58 65 84

± 0.12 4.12 ± 0.09 4.88 ± 0.06 4.87 ± 0.36 5.77 ± 0.34 9.05 ± 0.367.2 15 0 30 27

± 0.01 0.78 ± 0.06 0.79 ± 0.04 0.77 ± 0.04 0.43 ± 0.06 0.41 ± 0.0158 47 63 28 26

± 0.2B 6.03 ± 0.3C 5.58 ± 0.16C 6.15 ± 0.46D 7.56 ± 0.43E 11.9 ± 0.56E

ate rock, FePO4 and phytate.

0

1

2

3

4

5

WT PchC PchA

PO4

conc

. (m

gL-1

)

SGM+FePO4Boiled SGM+FePO4

a

Bb

C c

A

Fig. 1. P solubilization by two mutants of Pseudomonas fluorescens Pf-5 in 1 mLspent growth LB medium (SGM) before and after boiling. The bacteria were grownfor 24 h then filtered through 0.2 lM membrane filter and added with 0.01 g ofsterilized FePO4. The bars represent mean ± SE of three replicates. Treatmentsfollowed by different letters are not significantly different at a = 0.05.

4 P. Ghosh et al. / Chemosphere 134 (2015) 1–6

phosphate rock than FePO4 (Table 1 and Fig. 2). The data also sug-gested that siderophore played a more important role than organicacids in solubilizing FePO4 than phosphate rock.

Arsenate (AsV) is taken up into all living cells by phosphatetransporters. Hence, AsV competes with P for entering into bacteriathrough P transporters. Arsenate forms insoluble compounds withCa/Mg and Fe/Al (Silver and Phung, 2005) so it is mostly unavail-able. Despite its well-established toxicity to life, microbes growingin As-rich environment are tolerant to arsenic. The presence of AsVin bacteria may induce P deficiency, making As-resistant bacteriamore competitive in P solubilization from soil as well as in Puptake to survive under low P environment. This was

0

0.1

0.2

0.3

0.4

0.5

0.6

Shoot Root

Bio

mas

s (g

dw-1

/pla

nt) Control (Low P+Fe)

PC

PG6

0

2

4

6

8

10

Shoot Root

P co

ncen

trat

ion

(mg

g-1dw

)

ab

c

C

B

A

A

c

b

a

B

AA

B

Fig. 2. The effect of phytase-producing bacterium PG6 and control bacterium PC onplant growth (A) and P concentrations in tomato (B) after growing for 7 d in 0.2-strength Hoagland solution with low P at 1 mg L�1 with and without 0.2 g ofphytate. The bars represent mean ± SE of three replicates. Treatments followed bydifferent letters are not significantly different at a = 0.05.

demonstrated by their higher ability in solubilizing P from phos-phate rock and FePO4 than the control bacterium PC.

3.2. Solubilization of phytate by bacteria spent media

The fact that boiling of spent growth media significantlyreduced the ability of P solubilization from phosphate rock andFePO4 clearly indicated the importance of enzymatic processes insolubilizing P. Phytate accounts for 20–80% of the total organic Pin soil (Richardson, 2001), so we tested the ability of spent bacter-ial media in solubilizing P from phytate. Similar to FePO4, spentmedia from all bacteria solubilized P from phytate, indicating theirability in producing phytase, the only enzyme known to releaseinorganic P from phytate (Table 1). However, our effort to deter-mine phytase in the spent media was unsuccessful due to theinterference of components in the media (data not shown). In addi-tion, they were all more efficient than the control bacterium PC,with PG4, 6, 9 and 16 being more effective. PG12 and PG5, the mostefficient in solubilizing P from phosphate rock and FePO4, wereleast efficient in solubilizing P from phytate (Table 1), whereasPG 6, the least effective in solubilizing P from phosphate rock(Table 1), was most effective in solubilizing P from phytate(0.89 mg L�1) (Table 1). The data clearly indicated that the 7 ARBhad different ability in solubilizing phosphate rock, FePO4, andphytate via either enzymatic and/or non-enzymatic mechanisms.

According to total P solubilized from the three P compounds,strains PG12 (11.9 mg L�1), PG5 (7.56 mg L�1) and PG4(6.15 mg L�1) were more effective in extracting P (Table 1). Otherstrains PG9 (6.03 mg L�1), PG6 (5.58 mg L�1), and PG10(4.91 mg L�1) showed lower ability of producing siderophores(9.50–22.7) also solubilized less P from insoluble P compounds(Table 1). Amongst the 7 bacteria, PG6 and PG16 failed to solubilizeP from phosphate rock.

The amount of P solubilized from phytate by spent bacterialmedia was more than that from phosphate rock but less than thatfrom FePO4 (Table 1). Upon boiling, the amount of P solubilizedfrom phytate was reduced by 26–80% (Table 1), indicatingenzyme-controlled P solubilization. Among the bacteria, reductiondue to boiling in PG16 was the highest (80%) whereas reduction inthe control strain PC was the least at only 11%, which was consis-tent with their ability in P solubilization from phytate. The fact thatafter boiling still substantial amount of P was released from phy-tate indicated that boiling for 5 min did not completely deactivatebacterial phytase. Phytase is the only known enzyme to solubilize Pfrom phytate (Lessl et al., 2013). It is a phosphomonoesteraseenzyme, which aids in stepwise breakdown of phytate(Hariprasada and Niranjana, 2009). It is possible that temperaturehad limited effect on phytase.

3.3. Plant growth and nutrient uptake in tomato plants

As PG6 was most effective in solubilizing P from phytate at0.89 mg L�1 P (Table 1), spent media from PG6 was added to 0.2-strength HNS containing low P at 1 mg L�1, with or without 0.2 gof phytate to grow tomato plants for 7 d. The addition of spentmedia from PG6 and control PC enhanced tomato growth thanthe control, with PG 6 being more effective than PC. The shoot bio-mass increased by 65% to 0.56 g per plant with PG6 addition com-pared to the control and 44% compared to control PC (Fig. 2A). Wehypothesized that the increase in plant biomass was probablyassociated with the increased P uptake by tomato, which was sup-ported by P data in tomato plant biomass (Fig. 2B). The P concen-trations in tomato roots and shoots in PG6 treatment was 1.8–2.2fold higher (8.63–9.20 mg g�1) than that in the control PC (4.1–7.6 mg g�1).

P. Ghosh et al. / Chemosphere 134 (2015) 1–6 5

Since PG12 was the most effective in solubilizing P from FePO4,spent media from PG12 was added to 0.2-strength HNS containinglow P (1 mg L�1) and no Fe, with and without 0.2 g of FePO4 togrow tomato for 7 d. The root and shoot biomass in PG12 treat-ment were 2.3–4.0 fold (0.04–0.07 g per plant) higher than thatin the control and PchA mutant (Fig. 3A). The increased tomato bio-mass was probably associated with increased P and Fe uptake inthe plant. The shoots and roots with PG12 treatment had 1.5–2.7fold higher P concentration (7.46–11.6 mg g�1) than that in thecontrol and 1.3–1.7 fold higher than that of the mutant PchA(Fig. 3B). The Fe concentrations in tomato plant also increased10-fold (10.8 mg kg�1) in the shoots and 2-fold (270 mg kg�1) inthe roots with PG12 treatment compared to PchA mutant (Fig. 3C).

Previous studies on maize showed that seeds coated with P sol-ubilizing bacteria increased grain yield by 64–85% compared to theseeds that were not coated due to P solubilization from phosphaterock by bacteria (Hameeda et al., 2008). Still the prospect of eval-uating the role of siderophores in releasing insoluble FePO4 andphytase in solubilizing phytate in soil has not been explored exten-sively. Our study is the first of the few demonstrating the ability ofthe filtered spent media of arsenic-resistant bacteria in solubilizingphosphate rock, FePO4 and phytate and improving tomato growth.

0

2

4

6

8

10

12

14

ShootRoot

P co

ncen

tratio

n (m

g g-1

dw)

0

50

100

150

200

250

300

ShootRoot

Fe c

oncn

(mg

Kg

-1 d

w)

0

0.02

0.04

0.06

0.08

0.1

Biom

ass (

g dw

-1/p

lant

) Control (Low P-Fe)PchAPG12

A

cA A

ba

c b a

C

B

C

A

a

B

b

a

C

B

A

B

Shoot Root

Fig. 3. Effect of siderophore-producing bacterium PG12 and mutant PchA (impairedin siderophore production) on plant growth (A), and concentrations of P (B) and Fe(C) in tomato biomass after growing for 7 d in 0.2-strength Hoagland solution withlow P at 1 mg L�1 and no Fe with and without 0.2 g of FePO4. The bars representmean ± SE of three replicates. Treatments followed by different letters are notsignificantly different at a = 0.05.

4. Conclusions

We concluded that the spent growth media of arsenic-resistantbacteria were effective in solubilizing P from both organic (phy-tate) and inorganic (FePO4 and phosphate rock) via both enzymaticand/or non-enzymatic processes. We have also demonstrated thatphytase-producing bacterial strain PG6 solubilized P from phytateand siderophore-producing PG12 solubilized P from FePO4, whichsignificantly improved plant growth and nutrition in tomato seed-lings. These bacterial strains may have a potential to be used in thefield to improve plant P and Fe nutrition and thereby enhance plantgrowth.

Acknowledgements

This project is supported in part by UF-IFAS Innovation Grant.The authors thank Dr. Joyce Loper (Oregon State University) andEd Davis (USDA-ARS HCRL, Corvallis, OR) for providing us withthe wildtype and mutants of Pseudomonas fluorescens Pf-5.

References

Albrecht-Gary, A.M., Crumbliss, A.L., 1998. Coordination chemistry of siderophores:thermodynamics and kinetics of iron chelation and release. Met. Ions Biol. Syst.35, 239–327.

Altomare, C., Norvell, W.A., Björkman, T., Harman, G.E., 1999. Solubilization ofphosphates and micronutrients by the plant-growth-promoting and biocontrolfungus Trichoderma harzianum Rifai 1295-22. Appl. Environ. Microbiol., 2926–2933

Ams, D.A., Maurice, P.A., Hersman, L.E., Forsythe, J.H., 2002. Siderophore productionby an aerobic Pseudomonas mendocina bacterium in the presence of kaolinite.Chem. Geol. 188, 161–170.

Chen, Y.P., Rekha, P.D., Arun, A.B., Shen, F.T., Lai, W.-A., Young, C.C., 2006. Phosphatesolubilizing bacteria from subtropical soil and their tricalcium phosphatesolubilizing abilities. Appl. Soil Ecol. 34, 33–41.

Ghosh, P., Rathinasabapathi, B., Ma, L.Q., 2011. Arsenic-resistant bacteria solubilizedarsenic in the growth media and increased growth of arsenic hyperaccumulatorPteris vittata L. Bioresour. Technol. 102, 8756–8761.

Ghosh, P., Rathinasabapathi, B., Teplitski, M., Ma, L.Q., 2015. Bacterial ability inAsIII oxidation and AsV reduction: Relation to arsenic tolerance, P uptake,and siderophore production. Chemosphere. http://dx.doi.org/10.1016/j.chemosphere.2014.12.046.

Glick, B.R., 1995. The enhancement of plant growth by free living bacteria. Can. J.Microbiol. 42, 109–117.

Hameeda, B., Harini, G., Rupela, O.P., Wani, S.P., Reddy, G., 2008. Growth promotionof maize by phosphate-solubilizing bacteria isolated from composts andmacrofauna. Microbiol. Res. 163, 234–242.

Hariprasada, P., Niranjana, S.R., 2009. Isolation and characterization of phosphatesolubilizing rhizobacteria in improving plant health of tomato. Plant Soil 316,13–24.

Hartney, S.L., Mazurier, S., Kidarsa, T.A., Quccine, M.C., Lemanceau, P., Loper, J.E.,2011. TonB-dependent outer-membrane proteins and siderophore utilization inPseudomonas fluorescens Pf-5. Biometals 24, 193–213.

Hayes, J.E., Richardson, A.E., Simpson, R.J., 2000. Components of organic phosphorusin soil extracts that are hydrolysed by phytase and acid phosphatase. Biol. Fertil.Soils 32, 279–286.

Huang, X., El-Alawi, Y., Penrose, D.M., Glick, B.R., Greenberg, B.M., 2004. Responsesof three grass species to creosote during phytoremediation. Environ. Pollut. 130,453–463.

Kraemer, S.M., 2004. Iron oxide dissolution and solubility in the presence ofsiderophores. Aquat. Sci. 66, 3–18.

Lessl, J.T., Ma, L.Q., 2013. Sparingly-soluble phosphate rock induced significant plantgrowth and arsenic uptake by Pteris vittata from three contaminated soils.Environ. Sci. Technol. 47, 5311–5318.

Lessl, J.T., Ma, L.Q., Rathinasabapathi, B., Guy, C., 2013. Novel phytase from Pterisvittata resistant to arsenate, high temperature, and soil deactivation. Environ.Sci. Technol. 47, 2204–2211.

Li, H., Smith, S.E., Holloway, R.E., Zhu, Y., Smith, F.A., 2006. Arbuscular mycorrhizalfungi contribute to P uptake by wheat grown in a phosphorus-fixing soil even inthe absence of positive-growth responses. New Phytol. 172, 536–543.

Martell, A.E., Smith, R.M., Motekaitis, R.J., 2001. NIST critically selected stabilityconstants of metal complexes. NIST standard reference database 46, Version 6.0,NIST Gaithersburg.

Paulitz, T., Nowak-Thompson, B., Gamard, P., Tsang, E., Loper, J., 2000. A novelantifungal furanone from Pseudomonas aureofaciens, a biocontrol agent of fungalplant pathogens. J. Chem. Ecol. 26, 1515–1524.

Richardson, A.E., 2001. Prospects for using soil microorganism to improve theacquisition of phosphate by plant. Aust. J. Plant Physiol. 28, 897–906.

6 P. Ghosh et al. / Chemosphere 134 (2015) 1–6

Rodriguez, H., Fraga, R., 1999. Phosphate solubilizing bacteria and their role in plantgrowth promotion. Biotechnol. Adv. 17, 319–339.

Rodriguez, H., Fraga, R., Gonzalez, T., Bashan, Y., 2006. Genetic of phosphatesolubilization and its potential applications for improving plant growth-promoting bacteria. Plant Soil 287, 15–21.

Silver, S., Phung, L.T., 2005. Genes and enzymes involved in bacterial oxidation andreduction of inorganic arsenic. Appl. Environ. Microbiol. 71, 599–608.

Vassilev, N., Medina, A., Azcon, R., Vassileva, M., 2006. Microbial solubilization ofrock phosphate on media containing agro-industrial wastes and effect of theresulting products on plant growth and P uptake. Plant Soil 287, 77–84.