Bacterial Production of Organic Acids Enhances H2O2-DependentIodide OxidationHsiu-Ping Li,*,† Chris M. Yeager,‡,∥ Robin Brinkmeyer,† Saijin Zhang,† Yi-Fang Ho,† Chen Xu,†

Whitney L. Jones,‡ Kathleen A. Schwehr,† Shigeyoshi Otosaka,§ Kimberly A. Roberts,‡ Daniel I. Kaplan,‡

and Peter H. Santschi†

†Department of Marine Sciences, Texas A&M University, Galveston, Texas 77551, United States‡Savannah River National Laboratory, Aiken, South Carolina 29808, United States§Research Group for Environmental Science, Japan Atomic Energy Agency, Tokai-Mura, Ibaraki 319-1195, Japan

ABSTRACT: To develop an understanding of the role thatmicroorganisms play in the transport of 129I in soil−watersystems, bacteria isolated from subsurface sediments wereassessed for iodide oxidizing activity. Spent liquid mediumfrom 27/84 bacterial cultures enhanced iodide oxidation 2−10fold in the presence of H2O2. Organic acids secreted by thebacteria were found to enhance iodide oxidation by (1)lowering the pH of the spent medium, and (2) reacting withH2O2 to form peroxy carboxylic acids, which are extremelystrong oxidizing agents. H2O2-dependent iodide oxidationincreased exponentially from 8.4 to 825.9 μM with decreasingpH from 9 to 4. Organic acids with ≥2 carboxy groups enhanced H2O2-dependent iodide oxidation (1.5−15-fold) as a functionof increasing pH above pH 6.0, but had no effect at pH ≤ 5.0. The results indicate that as pH decreases (≤5.0), increasing H2O2hydrolysis is the driving force behind iodide oxidation. However, at pH ≥ 6.0, spontaneous decomposition of peroxy carboxylicacids, generated from H2O2 and organic acids, contributes significantly to iodide oxidation. The results reveal an indirectmicrobial mechanism, organic acid secretion coupled to H2O2 production, that could enhance iodide oxidation and organo-iodine formation in soils and sediments.

■ INTRODUCTIONWith the development of nuclear power, large inventories offission products have been produced and, in some cases,released into the environment. 129I, one of the major fissionproducts, has been given relatively little attention despite itspotential toxicity to man, high perceived mobility, radioactivelongevity (t1/2 = 1.6 × 107 yrs), and growing inventories inengineered disposal facilities and in the environment. Forexample, 146 Ci of 129I was deposited in soils at two U.S.Department of Energy (DOE) sites, the Hanford Site andSavannah River Site (SRS), and has been detected in mixedwaste plumes at these sites well above (10−1000×) thefederally regulated Maximum Contaminant Level (MCL) of 1pCi L−1.1,2 As a consequence, 129I has been identified as a keyrisk driver at both sites. It is therefore critical to understand theenvironmental behavior of 129I in order to rigorously assess itsstorage and disposal options.The mobility of 129I in water−soil systems is affected by its

chemical speciation and interactions with soil constituents,including minerals and organic matter. Major iodine species inwater−soil systems include iodide (I−), iodate (IO3

−), andorgano-iodine (OI). Iodide has a lower soil affinity, with adistribution coefficient (Kd) of ∼1 L3 kg−1, than that of iodate(Kd ∼ 1000 L3 kg−1) under oxic conditions.3−7 Both iodide

oxidation and iodate reduction can result in the production ofseveral reactive intermediates, including iodine (I2), hypoiodousacid (HOI), and triiodide (I3

−). These intermediates in turncan be taken up by soil organic matter to form OI throughiodination of aromatic moieties,8−10 adsorbed onto mineralsurfaces,1,6 and volatilized into the atmosphere.11 All theseprocesses would influence the mobility of 129I. To date, severalstudies have demonstrated that organic-rich soils have thetendency to serve as a 129I sink, i.e. immobilize 129I, mainlythrough the iodination process.12−14 In addition, the molecularweight of OI in soils can play a major role in the mobility ofiodine, whereby the higher molecular weight OI would tend tobecome a sink for 129I, whereas lower molecular weightcolloidal or truly dissolved OI species can leach out and migratein groundwater.4,7,13,15,16

In theory, iodide is expected to be the dominant species inmost water−soil systems due to its stability over the typicalrange of Eh and pH found in these environments.17 Oxidationof iodide via a single-step electron transfer is thermodynami-

Received: October 17, 2011Revised: March 6, 2012Accepted: March 28, 2012Published: March 28, 2012

Article

pubs.acs.org/est

© 2012 American Chemical Society 4837 dx.doi.org/10.1021/es203683v | Environ. Sci. Technol. 2012, 46, 4837−4844

cally unfavorable, unless a strong oxidant is available.18 Abioticoxidants (e.g., MnO2, Fe2O3) have been demonstrated tofacilitate iodide oxidation; however, the importance of theirreactions is limited under acidic conditions (pH ≤ 5.0).15,19−21

Microbial activity has long been linked to iodide oxidation andsoil OI formation.11,12,22 These studies have demonstrated thatsoils or sediments treated with heat, γ-irradiation, orprokaryotic antibiotics exhibit a considerable reduction iniodine binding capacity onto soils,11,22 whereas inoculation ofsterilized soil with fresh soil can restore the iodine-organicmatter binding potential.12 Mechanisms of microbial activitiesthat could immobilize 129I include bacterial accumulation andextracellular oxidation, which would also lead to iodination ofsoil organic matter. Indeed, a growing body of literature hasimplicated microbial oxidases, perhydrolases, and peroxidases inthe formation of halogenated organic matter in soils.11,12,22−26

Both iodide oxidation and accumulation activity have beenidentified in marine and brine water bacteria; however, usingsimilar approaches, efforts to demonstrate iodine immobiliza-tion by cellular uptake or oxidation by terrestrial bacteria werenot successful.23,24

In soils and sediments of the SRS F-area, based on theobserved ranges of pH (3.2−6.8) and Eh (360−654 mV),iodide is expected to be the predominant iodine species.2

However, ≥50% of iodine species is present in an oxidized form(i.e., IO3

− and OI) in the groundwater of the SRS F-area,indicating that iodide oxidation occurred.2,7,16 We previouslydemonstrated that bacteria isolated from SRS F-area subsurfacesediments irreversibly bound iodide, but only a small fraction(0.2−2%) was accumulated at ambient iodine concentrations(0.1 μM).27 It was concluded that bacterial iodide accumulationwas not responsible for the high fraction of OI detected in F-area groundwater. Interestingly, enhanced iodide oxidation wasobserved in the spent medium from a number of SRS F-areabacterial isolates in the presence of H2O2. In water−soilsystems, H2O2 can originate from common metabolic processesof fungi and bacteria under aerobic conditions and freely diffuseacross the cell membrane into the surrounding milieu wherethe concentration can reach up to 2−15 μM.28,29 Moreover,studies have demonstrated that H2O2 produced by photo-chemical or biological reactions in surface soils could betransported down-gradient to the subsurface environment.30,31

This result leads us to believe that a number of terrestrialbacteria are capable of influencing the chemical behavior of 129Ivia extracellular H2O2-dependent reactions. In this study, wereport that organic acids secreted by SRS F-area bacteria canenhance H2O2-dependent iodide oxidation.

■ MATERIALS AND METHODSBacterial Culture Conditions and Preparation of

Spent Medium and Crude Cell Extracts. Bacteria isolatedfrom soils in the F-area of the SRS27 were cultured aerobicallyby shaking (150 rpm) in liquid 1X M9GY medium (M9minimal salt medium enriched with 0.4% glucose and 0.4%yeast extract, pH 7.4 ± 0.2) at 30 °C in the dark. When theoptical density (OD600) of the cultures reached ∼1.2 (lateexponential phase), bacterial cells were harvested bycentrifugation (3200g, 20 °C, 15 min), then washed andsuspended in fresh 1X M9G medium. After 24 h incubation at30 °C in the dark with shaking (150 rpm), spent M9G mediumand cell pellets were separated by centrifugation (3200g, 20 °C,15 min). The spent medium was then directly used for theassays detailed below. To obtain crude cell extracts, cell pellets

were suspended with fresh 0.1X M9G medium and thendisrupted by 5 cycles of 20 s sonication (Misonix XL2000,100W) at 20 kHz and 40 s cooling in an ice bath. Completelysis of bacterial cells was confirmed through microscopy.Crude cell extracts were centrifuged to remove cell debris.

Determination of H2O2-Dependent Iodide OxidizingActivity. A method for monitoring triiodide (I3

−) formation24

was applied to determine iodide oxidation rates. Reactionmixtures for the determination of I3

− formation were preparedin 96-well plates (polystyrene, Greiner UV-Star) and consistedof the following: 150 μL of spent media or crude cell extract,100 mM phosphate buffer (pH 7.2), 10 mM KI, and 5 mMH2O2 in 300 μL total volume. Controls (i.e., reactions withoutKI or spent medium/crude cell extract) were also carried out todifferentiate abiotic from biotic reactions. Formation of I3

− wasmonitored spectrophotometrically at 353 nm after incubationat 30 °C for 20 min. The increase in absorbance at 353 nm wasconverted to the concentration of I3

− by applying Beer−Lambert’s law with a molar extinction coefficient (ε) of 25.5mM−1 cm−1.24 In this study, the amount of I3

− formed, in μM,from 150 μL of spent medium or crude cell extract was used toexpress the iodide oxidizing capacity.

16S rRNA Gene Sequencing and PhylogeneticAnalysis. For phylogenetic analysis of SRS bacteria capableof oxidizing iodide extracellularly, 16S rRNA gene wassequenced as previously described.27 Briefly, genomic DNAwas extracted by Tris-EDTA buffer (pH 7.5). Amplification bypolymerase chain reaction (PCR) of the 16S rRNA wasperformed by using bacterial 27f primer (5′-GAGTTT-GATCMTGGCTCAG-3′) and the 1492r primer (5′-GGTTACCTTGTTACGACTT-3′) and followed with apurification procedure (MinElute PCR purification kit,Qiagen). Sequencing of purified PCR products was performedby the Georgia Sequencing Facility at the University of Georgia,Athens. Phylogenetic analysis of 16S rRNA gene sequences wasperformed using BLAST searches (http://www.ncbi.nlm.nih.gov/Genbank/index.html), the RDP classifier,32 and maximumlikelihood reconstruction of phylogenetic trees with ARB.33

Correlation of Peroxidase and Iodide OxidationActivities. To evaluate the correlation of H2O2-dependentiodide oxidation and peroxidase activities, 1−5 mL of spentmedium was fractionated by centrifugal ultrafiltration withmolecular weight cut-offs of 3, 10, 30, 50, and 100 kDa(AmiconUltra, Millipore). Polysulfonate standards (40 and 100kDa, Sigma) were used as ultrafiltration controls. The retentatewas washed with 300 μL of fresh 1X M9G at least 3 times andsuspended to 1 mL with 1X M9G. The retentates and filtratesof each fraction were analyzed to determine H2O2-dependentiodide oxidation rates using the I3

− formation method. Thefraction with the highest iodide oxidizing activity was alsoanalyzed for general peroxidase activity using 2,2′-azino-bis-3-ethylbenzothiazoline-6-sulfonic acid (ABTS, Sigma) as thesubstrate. The assay mixture contained 1.3 mM ABTS, 150 μLretentate or filtrate sample, and 5 mM H2O2 in 100 mMphosphate buffer (pH 7.2). Oxidized ABTS was monitoredspectrophotometrically at 420 nm at 30 °C over 30 minintervals for 2 h. The increase in absorbance at 420 nm wasconverted to the concentration of oxidized ABTS by applyingBeer−Lambert’s law with ε of 36 mM−1 cm−1.34

Quantification of Organic Acid Concentration. Spentmedium was passed through a 0.22-μm syringe filter (celluloseacetate, VWR) and injected (50 μL) into an ion chromatographequipped with an IonPac AS11-HC column set (analytical

Environmental Science & Technology Article

dx.doi.org/10.1021/es203683v | Environ. Sci. Technol. 2012, 46, 4837−48444838

column, 4 × 250 mm; guard column, 4 × 50 mm) and a

conductivity detector from Dionex. The anionic components inthe samples were eluted with a gradient program of NaOH

flowing at 1.5 mL min−1. The gradient program of NaOH wasset as follows: 1 mM NaOH for 6 min, NaOH to 15 mM over 7

min, NaOH to 30 mM over 15 min, and then NaOH wasmaintained at 30 mM for 7 min. The eluted anionic

components were quantified based on the standard curve of

the following organic acids: lactic, pyruvic, succinic, maleic,oxalic, and citric acids.

■ RESULTS AND DISCUSSION

Bacterial Enhancement of H2O2-Dependent IodideOxidation. Iodide can be oxidized directly by reacting withH2O2 (eq 1).

+ + → +− + −3I (aq) H O (aq) 2H I (aq) 2H O2 2 3 2 (1)

Table 1. Enhancement of H2O2-Dependent Iodide Oxidation by Spent Medium and Crude Cell Extracts of Various BacterialCultures and the pH Values and Carboxylic Acid Concentrations in the Spent Medium

phylogenic affiliation I3− formation (μM)a carboxylic anions (mM)b

bacteriaaccession

no.most closely related organism

in GenBank database%

similarityspent

mediumcrude cellextract pH Lac/Ace Pyr Suc Mal Oxa Cit

M9Gc 38 ± 0 38 ± 0 7.45 >5.55 ND ND ND ND NDFA-86 JQ765450 Paenibacillus sp. HGF7 83 250 ± 2 <DL 6.24 >5.55 0.18 1.68 ND 0.06 NDFA-121 JQ765444 Bacillus cereus Rock3−44 99 180 ± 6 <DL 6.43 >5.55 0.11 2.13 ND 0.00 NDFA-130 JQ765447 Bacillus anthracis CNEVA-9066 95 342 ± 4 <DL 6.04 >5.55 0.11 1.81 ND 0.34 NDFA-132 JQ765446 Bacillus thuringiensis IBL 4222 91 269 ± 11 <DL 6.28 >5.55 0.08 1.85 ND 0.11 NDFA-139 JQ765445 Burkholderia phytof irmans PsJN 98 196 ± 2 <DL 6.42 >5.55 0.14 1.40 ND 0.06 NDFA-149 JQ765451 Bacillus cereus G9842 83 247 ± 5 <DL 6.28 >5.55 0.07 1.72 ND 0.25 NDFA-151 JQ765449 Brevibacillus brevis NBRC

10059994 337 ± 27 <DL 6.11 >5.55 0.08 2.34 ND 0.46 ND

FA-152 JQ765448 Bacillus cereus 03BB108 83 306 ± 8 <DL 6.13 >5.55 0.09 2.12 ND 0.17 NDFA-183B

JQ765442 Bacillus thuringiensis IBL 4222 98 173 ± 5 <DL 6.60 >5.55 0.29 1.25 ND 0.06 ND

FA-2CB*

JQ765443 Bacillus cereus Rock3−44 98 143 ± 13 <DL 6.50 >5.55 0.42 1.32 ND 0.04 ND

aThe standard deviation is derived from triplicate assays. <DL, below detection limit. bAbbreviations of organic acids: Lac/Ace, lactate/acetate; Pyr,pyruvate; Suc, succinate; Mal, maleate; Oxa, oxalate; Cit, citrate. ND, not detected. cM9G medium without cells.

Figure 1. (A) Relationship between pH and H2O2-dependent iodide oxidation in M9G medium. (B) Correlation between the measured rate ofH2O2-dependent iodide oxidation and that estimated from pH effects alone in spent medium from 10 SRS bacterial isolates. (C) Difference in themeasured rate (solid circles) of succinic acid enhanced H2O2-dependent iodide oxidation and the rate estimated to be due to pH effects alone(hollow circles). The pH values of the succinic acid-amended M9G medium are indicated with stars. (D) Correlation between the measured rate ofH2O2-dependent iodide oxidation and the rate estimated from the combined action of pH (equation from Figure 1A) and organic acids (equationsfrom Figure 2). Error bars in each figure represent one standard deviation (n = 3).

Environmental Science & Technology Article

dx.doi.org/10.1021/es203683v | Environ. Sci. Technol. 2012, 46, 4837−48444839

In this study, increased H2O2-dependent iodide oxidation wasobserved in liquid cultures of numerous bacterial strainsisolated from F-area of the SRS. Analysis of spent mediumand crude cell extracts localized the activity to the extracellularmedium (Table 1). It should be noted that components of cellextracts could hamper H2O2-dependent iodide oxidation (e.g.,intracellular catalases and/or peroxidases could rapidly depleteH2O2). Spent medium from 84 SRS bacteria was subsequentlyscreened for iodide oxidizing activity in the presence of H2O2,and enhancement occurred with 27 of the strains. These 27positive strains were members of Actinobacteria, Bacteriodetes,Firmicutes, and Proteobacteria phyla. No obvious trends relatingthe phylogenic affiliation of these bacteria to their capacity forH2O2-dependent iodide oxidation were observed (data notshown). In M9G medium without cells, 38 μM I3

− was formedfrom 10 mM iodide and 5 mM H2O2 over a 20 min incubationperiod. In contrast, 342 ± 4 μM I3

− accumulated in spentmedium from strain FA-130 (phyla Firmicutes) containing thesame concentrations of iodide and H2O2 (Table 1). Theseresults indicate that a component(s) commonly secreted bySRS bacteria can accelerate iodide oxidation by reacting withH2O2 or using H2O2 as an electron acceptor. To characterizethese reactive component(s), 10 SRS strains exhibitingenhanced H2O2-dependent iodide oxidation capacity, 142−342 μM I3

− produced from 10 mM iodide and 5 mM H2O2over 20 min, were selected for further experiments (Table 1).Initially, we hypothesized that the reactive component(s)

responsible for enhanced iodide oxidation was a haloperoxidaseor other, nonspecific peroxidase. Peroxidases are commonlyfound in soil bacteria,34 and haloperoxidases have beenimplicated in halogenation of natural organic matter.25,26

However, using molecular-size fractionation, the reactivecomponent(s) secreted by SRS bacteria was found to besmaller than 3 kDa (data not shown), and peroxidase-likeactivity (2, 2′-azino-bis-3-ethylbenzothiazoline-6-sulfonic acidoxidation assay34) was not detected (data not shown).Furthermore, the spent medium of three different strainsexhibited the same level of enhanced triiodide formation beforeand after boiling (15 min). These results indicated thatenzymatic activity was not directly responsible for the enhancedH2O2-dependent iodide oxidation detected in spent mediumfrom SRS bacteria.Influence of pH on H2O2-Dependent Iodide Oxida-

tion. Changes in pH can also influence H2O2-dependent iodideoxidation according to the reaction shown in eq 1. To evaluatethe effect of pH on H2O2-dependent iodide oxidation in M9Gmedium, I3

− formation was measured as a function of pH. Thedegree of I3

− formation exponentially increased from 8 ± 2 to826 ± 73 μM with decreasing pH from 9 to 4 (Figure 1A). ThepH values of the spent M9G medium from the SRS bacterialcultures ranged from 6.04 to 6.60, which is substantially lowerthan the pH (7.45) of fresh M9G medium (Table 1). Atheoretical iodide oxidation rate, factoring in the effect of pH,was estimated for each SRS spent medium listed in Table 1using the equation shown in Figure 1A. These calculationsindicate that the bacterial-mediated decrease in pH value wouldhave provided an appreciable enhancement of H2O2-dependentiodide oxidation (Figure 1B). However, the estimated rates ofiodide oxidation were still significantly lower (31−58%) thanthe experimentally determined values (Figure 1B), indicatingadditional factors were involved.Organic Acids Enhance H2O2-Dependent Iodide

Oxidation. Four anions of organic acids, including lactate,

pyruvate, succinate, and oxalate anions, were commonlydetected in the <3 kDa fraction of the spent medium fromSRS cultures (Table 1). In each case, high concentrations oflactate (>5.55 mM) were observed, which could contributesubstantially to the pH decrease of the bacterial culturemedium. The high concentration of lactate interfered with thequantification of acetate; thus, lactate and acetate concen-trations are shown as a single value using lactate as thestandard.Organic acids can acidify aqueous solutions by dissociation of

the hydrogen ion from the carboxyl group. To differentiate pH-related effects from other mechanisms that could be associatedwith the observed enhancement of iodide oxidation, rates of I3

−

formation were measured in M9G medium containing puresuccinic acid (the predominant dicarboxylic acid detected in thebacterial cultures; Table 1) ranging in concentration from 0 to15 mM, then compared to the expected rates based on pHchange alone using the equation shown in Figure 1A. As shownin Figure 1C, when the succinic acid concentration increasedfrom 0 to 15 mM, the pH value of the M9G medium decreasedfrom 7.45 to 6.65. This discrepancy in pH (0.8 units) wasestimated to enhance I3

− accumulation from 38 μM in M9Gmedium to 61 μM, an increase of <2-fold over the time courseof the experiment. Yet, I3

− concentration in the presence of 15mM succinic acid was measured as 323 ± 19 μM, an increase of∼10 fold (Figure 1C). These results further demonstrate thatorganic acid enhanced H2O2-dependent iodide oxidation is notsimply a function of pH change.Next, rates of I3

− formation were measured in reactionmixtures buffered at a constant pH (pH 6.0) containing M9Gmedium and 0−20 mM of select organic acids (Figure 2).Enhanced H2O2-dependent iodide oxidation was detected inthe reaction mixtures containing citric, maleic, oxalic, orsuccinic acid, but not acetic, lactic, or pyruvic acid (Figure 2).Moreover, the H2O2-dependent iodide oxidation capacity

Figure 2. Correlation between I3− formation and organic acid

concentration. Reactions were buffered at pH 6.0 and contained citric(solid rectangles), succinic (solid diamonds), oxalic (solid circles),maleic (solid triangles), acetic (hollow diamonds), lactic (hollowrectangles), or pyruvic (hollow circles) acid in M9G medium with 5mM H2O2. Each line expressed a second order correlation betweentriiodide formation and organic acid concentration with R2 ≥ 0.995(citric acid, Y = 95.1X2 + 205.6X + 7.3; succinic acid, Y = 7.0X2 +50.0X + 15.8; maleic acid, Y = 7.8X2 + 25.0X + 2.6, oxalic acid, Y =4.2X2 + 37.5X + 8.9). Error bars represent one standard deviation (n =3).

Environmental Science & Technology Article

dx.doi.org/10.1021/es203683v | Environ. Sci. Technol. 2012, 46, 4837−48444840

exhibited a “second order correlation” with respect to theconcentration of citric, succinic, maleic, or oxalic acid (R2 ≥0.995). The most plausible mechanism that can explain theenhanced capacity of iodide oxidation by organic acids in thepresence of H2O2 is the formation of peroxy carboxylic acid (eq2).

− + → − +R COOH H O R COOOH H O2 2 2 (2)

Peroxy carboxylic acids are powerful oxidants that could readilyoxidize I−. Similar to H2O2, their instability in aqueous solutionleads to a spontaneous decomposition, whereby 2 electrons areconsumed by each peroxy carboxyl group (−COOOH) (eq3).35−37

− + +→ − + +

− +

−R COOOH 3I 2H

R COOH H O I2 3 (3)

The data also revealed a correlation between the degree ofenhanced H2O2-dependent iodide oxidization and the numberof carboxyl groups (−COOH) on a given organic acid (Figure2). The highest iodide oxidizing capacity (479 ± 16 μM I3

−

produced) was observed in reaction mixtures with citric acid,which contains 3 carboxyl groups. Reactions containingsuccinic, maleic, or oxalic acid, all dicarboxylic acids, produced224, 151, and 139 μM I3

−, respectively. H2O2-dependent iodideoxidation was not enhanced in reactions containing monovalentorganic acids (acetic, lactic, or pyruvic acid; I3

− formation inM9G medium, with or without these monocarboxylic acids, was38 μM). These results imply that the H2O2-dependent iodideoxidation enhancement potential of a given organic acid isdependent on the number of available sites for peroxy carboxylgroup formation. The lack of H2O2-dependent iodide oxidationenhancement by monovalent organic acids, which could onlyproduce one peroxy carboxyl group, does not imply that noiodide oxidation occurred. It is more likely related to the factthat the spontaneous decomposition of one peroxy carboxylgroup consumed 2 electrons, the same electrons required forthe hydrolysis of one H2O2.As discussed earlier, 31−58% of the H2O2-dependent iodide

oxidation capacity measured in the spent medium of SRSbacterial cultures was not accounted for when pH change wasthe only variable considered (Figure 1B). To determine if thisdifferential could be explained by the action of peroxycarboxylic acids, the estimated H2O2-dependent iodideoxidation capacities were recalculated by summing (i) theinfluence of peroxy carboxylic acids, using the regressionsderived from Figure 2, and (ii) the influence of pH change,using the equation from Figure 1A. For example, the estimatedI3− formation for strain FA-86, which secreted 1.68 mM

succinate and 0.06 mM oxalate and lowered the pH value from7.45 to 6.24 (Table 1), was 248 μM using the followingequations:

(1)

× + × +

=

7.00 (1.68) 49.99 (1.68) 15.79

119.87 mM

2

from succinate (Figure 2).(2)

× + × −

=

4.19 (0.06) 37.51 (0.06) 8.93

2.34 mM

2

from oxalate (Figure 2).

(3)

× =− ×45469 e 125.74mM( 0.94 6.24)

from the decrease in pH value (Figure 1A).

After performing this normalization, the ratio of the estimatedH2O2-dependent iodide oxidation capacities to the measuredvalues approximated 1 (Figure 1D). This result demonstratesthat the enhanced H2O2-depenedent iodide oxidation from SRSbacterial strains could be fully accounted for by considering pHand the concentration of dicarboxylic acids in the mixture.

Relevance of Organic Acids Influencing 129I Mobilityin Natural Water−Soil Systems. Several reasons lead us tobelieve that iodide oxidation by peroxy carboxylic acids may berelevant in the F-area subsurface. First, although organic acidconcentrations in SRS F-area groundwater or sediments havenot been measured, evidence from other studies suggests thatorganic acids are ubiquitous and abundant in soil−watersystems. Numerous field observations place organic acidconcentrations in pore waters and stream waters in the μMrange.38−41 Organic acids in soils are mainly derived from bioticprocesses including microbial decomposition and exudationfrom plant roots, bacteria, and fungi, and appreciable quantities(μM−mM) of organic acids can be found in the surroundingmilieu of these organisms.42,43 Although the range of organicacid concentrations evaluated in our experiments (400 μM to10 mM) exceeded that found in the bulk phase of typical soilporewater, it was well within the range of concentrationsexpected for microenvironments surrounding bacterial biofilms,plant roots, and fungal hyphae.43,44 Finally, much of the lowlevel radioactive waste inventoried in soils and sediments atDOE sites, including F-area, contain abundant chelating agentsthat were used in the decontamination processes,2 of whichorganic acids, including citric and oxalic acid, were oftenprincipal components.45,46

In addition to organic acids, enhancement of iodide oxidationvia the formation of peroxy carboxylic acids described in thispaper requires H2O2. As addressed earlier, H2O2 in water−soilsystem mainly originates from metabolic processes of fungi andbacteria whereby porewater concentration can reach up to 2−15 μM and possibly much higher at micro- or nanoscale nearthe microorganisms.28,29 To determine if environmentallyrelevant concentrations of H2O2 could interact with organicacids to enhance iodide oxidation, I3

− formation was measuredin the presence of citric, succinic, or oxalic acid as a function ofH2O2 concentration (Figure 3). Iodide oxidation was enhanced

Figure 3. I3− formation as a function of H2O2 concentration with 3

mM citric acid (hollow triangles), succinic acid (hollow squares), andoxalic acid (hollow circles), and without organic acids (solid circles).Organic acid standards were prepared in M9G medium (pH 7.45).

Environmental Science & Technology Article

dx.doi.org/10.1021/es203683v | Environ. Sci. Technol. 2012, 46, 4837−48444841

by organic acids in the presence of H2O2 at all concentrationstested. However, the degree of enhancement (I3

− formationwith organic acid versus I3

− formation without organic acid)decreased from 2.5−5.7× enhancement to 1.5−3.0× enhance-ment as H2O2 concentrations decreased from 5 mM to 5 μM.Our results implicate groundwater pH as a primary control

on H2O2-dependent iodide oxidation (Figure 1A). In the SRSsubsurface, bacterial activity is not likely to affect the pH of bulksoils or porewater, because the alkalinity of soil or porewater isin the mM range. However, microniches surrounding bacterialbiofilms, fungi, and plant roots are often more acidic than thesurrounding bulk phase due to respiratory activity and organicacid secretion.43,44 Coupled to H2O2 production, these acidicmicroniches could serve as hotspots for iodide oxidation inwater−soil systems.Theoretically, the production rate of peroxy carboxylic acids

is directly related to the availability of protons.35,36,47,48 The pHof the contaminated groundwater within F-area of SRS rangesfrom 3.0 to 9.0 due to the acidic nature of the mixed wasteplume and ongoing base injection remediation efforts.2 Todetermine if environmentally relevant pH ranges couldinfluence the organic acid enhancement of H2O2-dependentiodide oxidation, I3

− formation was measured in the presence ofcitric, succinic, or oxalic acid as a function of pH. Overall, therate of H2O2-dependent iodide oxidation decreased withincreasing pH from 3.0 to 8.0 (Figure 4A), similar to the

pattern illustrated in Figure 1A. However, as presented inFigure 4B, the contribution of peroxy carboxylic acids to H2O2-dependent iodide oxidation, as calculated in eq 4, prevails onlyat pH ≥ 6.0.

=−

−

−

I ratioI produced from peroxy carboxylic acids and H O

I produced from H O hydrolysis only

3

3 2 2

3 2 2

(4)

It is quite possible that iodide oxidation at pH <6.0 can becarried out concomitantly via H2O2 hydrolysis and spontaneousdecomposition of peroxy carboxylic acids. However, our resultsindicate that at low pH values, particularly ≤ pH 5.0, increasedH2O2 hydrolysis is the driving force behind iodide oxidation;whereas, above pH 5.0, spontaneous decomposition of peroxycarboxylic acids, generated from H2O2 and organic acids, cancontribute significantly to iodide oxidation. Indeed, thedissociation constants of citric (pKa3 = 6.40), succinic (pKa2= 5.60), and oxalic acids (pKa2 = 4.14) are such that at higherpH values (i.e., pH >7.0) full dissociation of the carboxylgroups is more likely, thus increasing the number of reactivesites to form peroxy carboxylic groups.35

In summary, organic acids produced by terrestrial bacteriacan enhance iodide oxidation in the presence of H2O2. Asdemonstrated in this study, the magnitude of this effect can becontrolled by several environmental factors such as pH, thecomposition and redox properties of the organic acids, and theconcentration of H2O2. In terms of the long-term fate andtransport of 129I, H2O2- and organic acid-dependent iodinationreactions could be a relevant mechanism for immobilizing 129Iin the terrestrial environment.

■ AUTHOR INFORMATION

Corresponding Author*E-mail: original@tamu.edu.

Present Address∥Bioscience Division, Los Alamos National Laboratory, LosAlamos, New Mexico 87545, USA.

NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTS

This work was funded by the U.S. Department of Energy’sSubsurface Biogeochemical Research Program within the Officeof Science (DE-SC0006823).

■ REFERENCES(1) Kaplan, D. I.; Roberts, K. A.; Schwehr, K. A.; Lilley, M. S.;Brinkmeyer, R.; Denham, M. E.; Diprete, D.; Li, H.-P.; Powell, B. A.;Xu, C.; Yeager, C. M.; Zhang, S. J.; Santschi, P. H. Evaluation of aRadioiodine Plume Increasing in Concentration at the Savannah RiverSite. Environ. Sci. Technol. 2011, 45 (2), 489−495.(2) Otosaka, S.; Schwehr, K. A.; Kaplan, D. I.; Roberts, K. A.; Zhang,S.; Xu, C.; Li, H.-P.; Ho, Y.-F.; Brinkmeyer, R.; Yeager, C. M.;Santschi, P. H. Factors controlling mobility of 127I and 129I species inan acidic groundwater plume at the Savannah River Site. Sci. TotalEnviron. 2011, 409, 3857−3865.(3) Ashworth, D. J.; Shaw, G. A comparison of the soil migration andplant uptake of radioactive chlorine and iodine from contaminatedgroundwater. J. Environ. Radioact. 2006, 89 (1), 61−80.

Figure 4. (A) I3− formation as a function of pH with 3 mM citric acid

(hollow triangles), succinic acid (hollow squares), and oxalic acid(hollow circles), and without organic acids (solid circles). (B) RelativeI3− formation (i.e., ratio of I3

− formed with the presence of 3 mMorganic acid to I3

− formed without the presence of organic acid) as afunction of pH. Citric acid (hollow triangles), succinic acid (hollowsquares), and oxalic acid (hollow circles) were prepared in the pH-adjusted M9G medium individually.

Environmental Science & Technology Article

dx.doi.org/10.1021/es203683v | Environ. Sci. Technol. 2012, 46, 4837−48444842

(4) Fukui, M.; Fujikawa, Y.; Satta, N. Factors affecting interaction ofradioiodide and iodate species with soil. J. Environ. Radioact. 1996, 31(2), 199−216.(5) Hu, Q. H.; Zhao, P. H.; Moran, J. E.; Seaman, J. C. Sorption andtransport of iodine species in sediments from the Savannah River andHanford Sites. J. Contam. Hydrol. 2005, 78 (3), 185−205.(6) Kaplan, D. I.; Serene, R. J.; Parker, K. E.; Kutnyokov, I. V. IodideSorption to Subsurface Sediments and Illitic Minerals. Environ. Sci.Technol. 2000, 34, 6.(7) Schwehr, K. A.; Santschi, P. H.; Kaplan, D. I.; Yeager, C. M.;Brinkmeyer, R. Organo-iodine formation in soils and aquifer sedimentsat ambient concentrations. Environ. Sci. Technol. 2009, 43, 7258−7264.(8) Moulin, V.; Reiller, P.; Amekraz, B.; Moulin, C. Directcharacterization of iodine covalently bound to fulvic acids byelectrospray mass spectrometry. Rapid Commun. Mass Spectrom.2001, 15 (9), 2488−2496.(9) Schlegel, M. L.; Reiller, P.; Mercier-Bion, F.; Barre, N.; Moulin, V.Molecular environment of iodine in naturally iodinated humicsubstances: Insight from X-ray absorption spectroscopy. Geochim.Cosmochim. Acta 2006, 70 (22), 5536−5551.(10) Xu, C.; Zhong, J.; Hatcher, P. G.; Zhang, S.; Li, H.-P.; Ho, Y.-F.;Schwehr, K. A.; Kaplan, D. I.; Roberts, K.; Brinkmeyer, R.; Yeager, C.M.; Santschi, P. H. Molecular environment of stable iodine andradioiodine (129I) in natural organic matter: Evidence inferred fromNMR and binding experiments at environmentally relevant concen-trations. Geochim. Cosmochim. Acta 2011.(11) Amachi, S.; Kasahara, M.; Hanada, S.; Kamagata, Y.; Shinoyama,H.; Fujii, T.; Muramatsu, Y. Microbial participation in iodinevolatilization from soils. Environ. Sci. Technol. 2003, 37 (17), 3885−3890.(12) Muramatsu, Y.; Yoshida, S.; Fehn, U.; Amachi, S.; Ohmomo, Y.Studies with natural and anthropogenic iodine isotopes: iodinedistribution and cycling in the global environment. J. Environ. Radioact.2004, 74, 221−232.(13) Xu, C.; Zhang, S.; Ho, Y.-F.; Miller, E. J.; Roberts, K. A.; Li, H.-P.; Schwehr, K. A.; Otosaka, S.; Kaplan, D. I.; Brinkmeyer, R.; Yeager,C. M.; Santschi, P. H. Is soil natural organic matter a sink or source forradioiodine (129I) in a nuclear waste disposal site? Geochim.Cosmochim. Acta 2011, 75, 5716−5735.(14) Yamaguchi, N.; Nakano, M.; Takamatsu, R.; Tanida, H.Inorganic iodine incorporation into soil organic matter: Evidencefrom iodine K-edge X-ray absorption near-edge structure. J. Environ.Radioact. 2010, 101 (6), 451−457.(15) Xu, C.; Miller, E. J.; Zhang, S.; Li, H.-P.; Ho, Y.-F.; Schwehr, K.A.; Kaplan, D. I.; Roberts, K. A.; Otosaka, S.; Brinkmeyer, R.; Yeager,C. M.; Santschi, P. H. Sequestration and re-mobilization of radioiodine(129I) by soil organic matter and possible consequences of the remedialaction at Savannah River Site. Environ. Sci. Technol. 2011, 45 (23),9975−9983.(16) Zhang, S.; Schwehr, K. A.; Ho, Y.-F.; Xu, C.; Roberts, K.;Kaplan, D.; Brinkmeyer, R.; Yeager, C. M.; Santschi, P. H.Determination of 127I and 129I speciation in environmental watersusing a novel gas chromatography-mass spectrometry method. Environ.Sci. Technol. 2010, 44, 9042−9048.(17) Pourbaix, M. Atlas of Electrochemical Equilibria in AqueousSolutions; NACE International: Celelcor, Brussels, 1974.(18) Luther, G., W.; Wu, J.; Cullen, J., B., Redox Chemistry of Iodinein Seawater. In Aquatic Chemistry: Interfacial and Interspecies Processes;Huang, C. P.; O’Melia, C. R.; Morgan, J. J., Eds.; American ChemicalSociety: Washinton, DC, 1995; Vol. 244, pp 135−155.(19) Fox, P. M.; Davis, J. A.; Luther, G. W. The kinetics of iodideoxidation by the manganese oxide mineral birnessite. Geochim.Cosmochim. Acta 2009, 73 (10), 2850−2861.(20) Fox, P. M.; Kent, D. B.; Davis, J. A. Redox Transformations andTransport of Cesium and Iodine (−1,0,+5) in Oxidizing and ReducingZones of a Sand and Gravel Aquifer. Environ. Sci. Technol. 2010, 44(6), 1940−1946.(21) Gallard, H.; Allard, S.; Nicolau, R.; von Gunten, U.; Croue, J. P.Formation of Iodinated Organic Compounds by Oxidation of Iodide-

Containing Waters with Manganese Dioxide. Environ. Sci. Technol.2009, 43 (18), 7003−7009.(22) Muramatsu, Y.; Yoshida, S.; Uchida, S.; Hasebe, A. Iodinedesorption from rice paddy soil. Water, Air Soil Pollut. 1996, 86 (1−4),359−371.(23) Amachi, S.; Kimura, K.; Muramatsu, Y.; Shinoyama, H.; Fujii, T.Hydrogen peroxide-dependent uptake of iodine by marine Flavobacter-iaceae bacterium strain C-21. Appl. Environ. Microbiol. 2007, 73 (23),7536−7541.(24) Amachi, S.; Muramatsu, Y.; Akiyama, Y.; Miyazaki, K.; Yoshiki,S.; Hanada, S.; Kamagata, Y.; Ban-nai, T.; Shinoyama, H.; Fujii, T.Isolation of iodide-oxidizing bacteria from iodide-rich natural gasbrines and seawaters. Microb. Ecol. 2005, 49 (4), 547−557.(25) Bastviken, D.; Thomsen, F.; Svensson, T.; Karlsson, S.; Sanden,P.; Shaw, G.; Matucha, M.; Oberg, G. Chloride retention in forest soilby microbial uptake and by natural chlorination of organic matter.Geochim. Cosmochim. Acta 2007, 71 (13), 3182−3192.(26) Ortiz-Bermudez, P.; Kolby, C. H.; Srebotnik, E.; Hammel, K. E.Chlorination of lignin by ubiquitous fungi has a likely role in globalorganochlorine production. Proc. Natl. Acad. Sci. U.S.A. 2007, 104(10), 3895−3900.(27) Li, H.-P.; Brinkmeyer, R.; Jones, W. L.; Zhang, S. J.; Xu, C.;Schwehr, K. A.; Santschi, P. H.; Kaplan, D. I.; Yeager, C. M. IodideAccumulation by Aerobic Bacteria Isolated from Subsurface Sedimentsof a I-129-Contaminated Aquifer at the Savannah River Site, SouthCarolina. Appl. Environ. Microbiol. 2011, 77 (6), 2153−2160.(28) Gonzalez-Flecha, B.; Demple, B. Homeostatic Regulation ofIntracellular Hydrogen Peroxide Concentration in AerobicallyGrowing Escherichia coli. J. Bacteriol. 1997, 179 (2), 382−388.(29) Guillen, F.; Martinez, A. T.; Martínez, M. J. Production ofhydrogen peroxide by aryl-alcohol oxidase from the ligninolytic fungusPleurotus eryngii. Appl. Microbiol. Biotechnol. 1990, 32 (4), 465−469.(30) Cooper, W. J.; Lean, D. R. S. Hydrogen peroxide concentrationin a northern lake: photochemical formation and diel variability.Environ. Sci. Technol. 1989, 23 (11), 1425−1428.(31) Watts, R. J.; Finn, D. D.; Cutler, L. M.; Schmidt, J. T.; Teel, A.L. Enhanced stability of hydrogen peroxide in the presence ofsubsurface solids. J. Contam. Hydrol. 2007, 91 (3−4), 312−326.(32) Cole, J. R.; Wang, Q.; Cardenas, E.; Fish, J.; Chai, B.; Farris, R.J.; Kulam-Syed-Mohideen, A. S.; McGarrell, D. M.; Marsh, T.; Garrity,G. M.; Tiedje, J. M. The Ribosomal Database Project: Improvedalignments and new tools for rRNA analysis. Nucleic Acids Res. 2009,37 (suppl_1), D141−145.(33) Wolfgang, L.; Oliver, S.; Ralf, W.; Lothar, R.; Harald, M.;Yadhukumar; Arno, B.; Tina, L.; Susanne, S.; Gangolf, J.; Wolfram, F.;Igor, B.; Stefan, G.; Anton, W. G.; Oliver, G.; Silke, G.; Stefan, H.ARB: A software environment for sequence data. Nucleic Acids Res.2004, 32 (4), 1363.(34) Passardi, F.; Theiler, G.; Zamocky, M.; Cosio, C.; Rouhier, N.;Teixera, F.; Margis-Pinheiro, M.; Ioannidis, V.; Penel, C.; Falquet, L.;Dunand, C. PeroxiBase: The peroxidase database. Phytochemistry 2007,68 (12), 1605−1611.(35) Ball, R. E.; Edwards, J. O.; Haggett, M. L.; Jones, P. A kineticand isotopic study of the decomposition of monoperoxyphthalic acid.J. Am. Chem. Soc. 1967, 89 (10), 2331−2333.(36) Ferdousi, B. N.; Islam, M. M.; Okajima, T.; Ohsaka, T.Electroreduction of peroxycitric acid coexisting with hydrogenperoxide in aqueous solution. Electrochim. Acta 2007, 53 (2), 968−974.(37) Awad, M. I.; Oritani, T.; Ohsaka, T. SimultaneousPotentiometric Determination of Peracetic Acid and HydrogenPeroxide. Anal. Chem. 2003, 75 (11), 2688−2693.(38) Ali, T.; Bylund, D.; Essen, S. A.; Lundstrom, U. S. Liquidextraction of low molecular mass organic acids and hydroxamatesiderophores from boreal forest soil. Soil Biol. Biochem. 2011, 43 (12),2417−2422.(39) Bylund, D.; Norstrom, S. H.; Essen, S. A.; Lundstrom, U. S.Analysis of low molecular mass organic acids in natural waters by ion

Environmental Science & Technology Article

dx.doi.org/10.1021/es203683v | Environ. Sci. Technol. 2012, 46, 4837−48444843

exclusion chromatography tandem mass spectrometry. J. Chromatogr.,A 2007, 1176 (1−2), 89−93.(40) Kehew, A. E. Applied Chemical Hydrogeology; Prentice-Hall, Inc.:Upper Saddle River, NJ, 2001; p 368.(41) Thurman, E. M. Organic Geochemistry of Natural Waters;Martinus Nijhoff/Dr W. Junk Publishers: Boston, MA, 1985; p 497.(42) Jones, D. L. Organic acids in the rhizosphere − A critical review.Plant Soil 1998, 205 (1), 25−44.(43) McMahon, P. B.; Chapelle, F. H. Microbial production oforganic acids in aquitard sediments and its role in aquifergeochemistry. Nature 1991, 349, 233−235.(44) Long, M. H.; McGlathery, K. J.; Zieman, J. C.; Berg, P. The roleof organic acid exudates in liberating phosphorus from seagrass-vegetated carbonate sediments. Limnol. Oceanogr. 2008, 53 (6), 2616−2626.(45) Riley, R. G.; Zachara, M. M. Chemical Contaminants on DOELands and Selection of Contaminants Mixtures for Subsurface ScienceResearch; DOE/ER-0547T; U.S. Department of Energy: WashingtonDC, 1992.(46) Serne, R. J.; Cantrell, K. J.; Lindenmeier, C. W.; Owen, A. T.;Kutnyakov, I. V.; Orr, R. D.; Felmy, A. R. Rodionuclide-Chelating AgentComplexes in Low-Level Radioactive Decontamination Waste; Stability,Adsorbtion and Transport Potential; PNNL-13774; Pacific NorthwestNational Laboratory: Richland, WA, 2002.(47) Dul’neva, L. V.; Moskvin, A. V. Kinetics of formation ofperoxyacetic acid. Russ. J. Gen. Chem. 2005, 75 (7), 1125−1130.(48) Zhao, X. B.; Cheng, K. K.; Hao, J. B.; Liu, D. H. Preparation ofperacetic acid from hydrogen peroxide, part II: Kinetics forspontaneous decomposition of peracetic acid in the liquid phase. J.Mol. Catal., A: Chem. 2008, 284 (1−2), 58−68.

Environmental Science & Technology Article

dx.doi.org/10.1021/es203683v | Environ. Sci. Technol. 2012, 46, 4837−48444844