Vol. 58, No. 3APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Mar. 1992, p. 850-8560099-2240/92/030850-07$02.00/0Copyright © 1992, American Society for Microbiology

Reduction of Uranium by Desulfovibrio desulfuricansDEREK R. LOVLEY* AND ELIZABETH J. P. PHILLIPS

430 National Center, Water Resources Division, U.S. Geological Survey, Reston, Virginia 22092

Received 4 November 1991/Accepted 19 December 1991

The possibility that sulfate-reducing microorganisms contribute to U(VI) reduction in sedimentary environ-ments was investigated. U(VI) was reduced to U(IV) when washed cells of sulfate-grown Desulfovibriodesulfuicans were suspended in a bicarbonate buffer with lactate or H2 as the electron donor. There was noU(VI) reduction in the absence of an electron donor or when the cells were killed by heat prior to theincubation. The rates of U(VI) reduction were comparable to those in respiratory Fe(III)-reducing microor-ganisms. Azide or prior exposure of the cells to air did not affect the ability ofD. desulfuricans to reduce U(VI).Attempts to grow D. desulfiricans with U(VI) as the electron acceptor were unsuccessful. U(VI) reductionresulted in the extracellular precipitation of the U(IV) mineral uraninite. The presence of sulfate had no effecton the rate of U(VI) reduction. Sulfate and U(VI) were reduced simultaneously. Enzymatic reduction of U(VI)by D. desulfuricans was much faster than nonenzymatic reduction of U(VI) by sulfide, even when cells of D.desulfuricans were added to provide a potential catalytic surface for the nonenzymatic reaction. The resultsindicate that enzymatic U(VI) reduction by sulfate-reducing microorganisms may be responsible for theaccumulation of U(IV) in sulfidogenic environments. Furthermore, since the reduction of U(VI) to U(IV)precipitates uranium from solution, D. desulfuricans might be a useful organism for recovering uranium fromcontaminated waters and waste streams.

Microbial reduction of soluble U(VI) to insoluble U(IV)may play an important role in the geochemical cycle ofuranium and may also serve as a mechanism for the biore-mediation of uranium-contaminated waters (26). The pre-cipitation of U(IV) as the result of U(VI) reduction inanaerobic marine sediments is the most significant modernglobal sink for dissolved uranium (1, 19, 35). Immobiliza-tion of uranium through U(VI) reduction has led to theformation of many economically important uranium deposits(15, 17, 20, 34). Reductive precipitation of uranium may alsoaccount for the ability of the bottom sediments of algalponds to remove dissolved uranium from uranium minewastewaters (4). Bioreactors containing U(VI)-reducing mi-croorganisms can rapidly remove dissolved uranium fromwater (11).Only two microorganisms, "Geobacter metallireducens"

(previously known as strain GS-15) and Shewanella (previ-ously Alteromonas) putrefaciens, have been previouslyshown to use U(VI) as a terminal electron acceptor (26).Both of these microorganisms are respiratory Fe(III)-reduc-ing microorganisms (25, 27, 28). The finding that respira-tory Fe(III)-reducing microorganisms can reduce U(VI) isconsistent with geochemical evidence indicating that U(VI)is reduced within the Fe(III)-reducing zones of marinesediments (3, 6, 18, 20, 37) and other environments (12,13).However, U(VI) may also be reduced in environments

where Fe(III)-reducing microorganisms may not be abun-dant. For example, the coprecipitation of sulfide and U(IV)minerals from some groundwaters suggests that, in someinstances, U(VI) is reduced in environments in which sulfatereduction is the predominant terminal electron-acceptingprocess (33). A study of the distribution of Fe(III)-reducingmicroorganisms in deep aquifers of the Atlantic CoastalPlain suggested that respiratory Fe(III)-reducing microor-

* Corresponding author.

ganisms may not be present in some sulfidogenic subsurfaceenvironments (22). Therefore, a mechanism other than theactivity of Fe(III)-reducing microorganisms may be requiredto explain U(VI) reduction in such environments. Further-more, a recent study has suggested that in some marinesediments, U(VI) might be reduced within the sulfate-reduc-ing rather than the Fe(III)-reducing zone (19).Nonenzymatic reduction of U(VI) by sulfide has been the

traditional explanation for U(VI) reduction in sulfidogenicenvironments (15, 17, 20, 30, 32, 33, 36). However, ourstudies (26) have indicated that sulfide is a poor U(VI)reductant. An alternative possibility is that microorganismsliving within the sulfidogenic environment enzymaticallyreduce U(VI). U(VI) was reduced in sulfate-reducing cul-tures of Desulfovibrio desulfuricans (32, 36). Although thiswas attributed to a nonenzymatic reduction of U(VI) by thesulfide produced during sulfate reduction (32, 36), the resultsdid not preclude direct enzymatic reduction of U(VI) by D.desulffuricans.The possibility of enzymatic U(VI) reduction by D. des-

ulfjuricans is supported by another study, which stated thatcell extracts of D. desulfuricans could reduce U(VI) (38).However, no data supporting this statement were given.Furthermore, the physiological significance of such metalreductions in cell extracts to whole-cell metabolism is ques-tionable (9).The purpose of the study reported here was to investigate

the possibility that in addition to Fe(III)-reducing microor-ganisms, sulfate-reducing microorganisms are able to useU(VI) as a terminal electron acceptor. The results indicatethat D. desulfuricans can enzymatically reduce U(VI). Sincethis uranium reduction can proceed simultaneously withsulfate reduction, this metabolism may explain the simulta-neous reduction of sulfate and uranium that has been ob-served in some uranium deposits. The results also suggestthat D. desulfuricans may be a useful organism in efforts tobioremediate uranium-contaminated waters.

850

on May 10, 2021 by guest

http://aem.asm

.org/D

ownloaded from

REDUCTION OF U(VI) BY DESULFOVIBRIO DESULFURICANS 851

MATERIALS AND METHODS

Source of the organism. The type strain of D. desulfuricans(ATCC 29577) was obtained from the American Type Cul-ture Collection Rockville, Md.

Culturing. Standard anaerobic techniques (2, 16, 31) wereused throughout the study. Gases were passed through acolumn of hot reduced copper filings to remove traces ofoxygen. The growth medium was a modification of themedium previously used to cultivate "G. metallireducens"(25). In addition to vitamins and trace minerals, the constit-uents (per liter of the basic medium) were as follows:NaH2CO3, 2.5 g; NH4Cl, 1.5 g; NaH2PO4, 0.6 g; KCl, 0.1 g;60% lactate syrup, 20 ml; MgSO4, 1.5 g; and Na2SO4, 1.5 g.This medium was dispensed in 100- or 10-ml volumes into160-ml glass serum bottles or anaerobic pressure tubes. Themedium was flushed with N2-C02 (80:20) to remove dis-solved oxygen. After the medium had been autoclaved, thefollowing were added from sterile anaerobic stocks to pro-vide the stated final concentrations per liter: yeast extract, 2g; proteose peptone, 5 g; L-cysteine-HCl, 0.25 g; and ferrousammonium sulfate, 0.1 g. D. desulfuricans produced a blackferrous sulfide precipitate in this medium. Therefore, tocollect cells without interference from the ferrous sulfideprecipitate, the ferrous ammonium sulfate was left out of themedium for at least two transfers (10% inoculum) prior to thecollection of cells for experiments. Cells were grown at 35°C.

"G. metallireducens" and S. putrefaciens were culturedat 30°C with acetate ("G. metallireducens") or lactate (S.putrefaciens) as the electron donor and Fe(III) citrate as theelectron acceptor as described previously (25, 27).

Cell suspensions. Under N2-C02, cells were collected bycentrifugation, resuspended in anaerobic bicarbonate buffer(2.5 g of sodium bicarbonate per liter), centrifuged, and thenresuspended in bicarbonate buffer (10 ml) in 25-ml serumbottles to provide ca. 1 to 5 mg of cell protein. The pH was6.8. For studies in which sulfate reduction was also investi-gated, the buffer also contained 0.25 g of L-cysteine-HCI perliter.U(VI) (1 mM unless otherwise noted) was provided as

uranyl acetate from a 10 mM stock solution in bicarbonatebuffer. Lactate (10 mM) was included as an electron donor inall experiments unless otherwise stated. Sulfate (2 mM) wasadded in the studies on the concurrent reduction of U(VI)and sulfate. Lactate and sulfate were added from concen-trated anaerobic stocks of the sodium salts in deionizedwater. H2 was provided by injecting 10 ml of H2 into theheadspace.To heat kill cells, we heated the buffer to 80°C in a water

bath and maintained it at 80°C for an additional 15 min afterthe addition of the cells and prior to the addition of U(VI) orsulfate. We exposed cells to air by bubbling air through thesuspension for 10 min. The suspensions were then bubbledwith N2-C02 for 5 min before U(VI) was added. Studies onthe effect of azide were conducted by preincubating the cellsuspension in the presence of 1 mM sodium azide for 15 minprior to the addition of U(VI). Sulfide (1 mM) was added tocell suspensions, when noted, from a freshly prepared an-aerobic stock in bicarbonate buffer. Cell suspensions wereincubated at 35°C, which was optimal for U(VI) reduction byall three organisms.

Analytical techniques. The U(VI) concentration was mea-sured under anaerobic conditions as previously described(11, 26) with a Kinetic Phosphorescence analyzer (KPA-10;Chemchek Instruments), which uses a pulsed nitrogen dyelaser and a complexing agent to measure U(VI) levels in

solution. To measure the total uranium content, a 1-mlsample was brought out into the air, acidified with 0.5 ml ofconcentrated nitric acid, and diluted with deionized waterunder aerobic conditions. This treatment oxidized U(IV) toU(VI). The U(IV) concentration was calculated as the dif-ference between the U(VI) and total uranium determina-tions. Samples for sulfate determination were filtered (Gel-man Acrodisc; pore diameter, 0.2 ,um) and analyzed on aDionex ion chromatograph. Protein was determined by themethod of Lowry et al. (29) with bovine serum albumin as astandard.For X-ray diffraction analysis, the precipitate resulting

from U(VI) reduction in cell suspensions was collected onWhatman no. 5 filter paper, dried under N2, and ground to afine powder with a mortar and pestle. The randomly orientedpowder was mounted on a glass slide with amyl acetate. TheX-ray diffractometer was equipped with a graphite mono-chrometer and a nickel filter and used CuK5 radiation with awavelength of 1.5418 A (0.15418 nm). The scan rate was 102(theta)/min, with a soller slit of 10 and a receiving slit of0.25°.

Electron microscopy. A U(VI)-reducing cell suspension ofD. desulfunicans was examined by transmission electronmicroscopy by using a Philips EM400T electron microscopeoperating at 100 kV. The elemental composition of theprecipitate was determined by energy-dispersive X-ray anal-ysis by using a Link EDS system (spot size, 10 nm) attachedto a Link light element detector.

RESULTS

U(VI) reduction in cell suspensions. When washed cellsuspensions of D. desulfuricans were suspended in bicar-bonate buffer containing lactate as the potential electrondonor, there was a rapid loss of U(VI) over time and acorresponding increase in the amount of U(IV) (Fig. 1A).Hydrogen served equally as well as lactate as an electrondonor for U(VI) reduction (Fig. 1B). There was no U(VI)reduction in the absence of an added electron donor or in thepresence of an electron donor (lactate) with heat-killed cells(Fig. 1B).On the basis of cell protein measurement, the U(VI)-

reducing ability of D. desulfuncans compared favorably withthe rates of U(VI) reduction by GS-15 and S. putrefaciens,the two organisms previously shown to reduce U(VI) (Fig.2). D. desulfuricans reduced U(VI) slightly faster than GS-15did and slightly more slowly than S. putrefaciens did.Azide (1 mM) or prior exposure of the cells to air did not

inhibit U(VI) reduction by D. desulfunicans (Fig. 3). Similarresults have been observed with "G. metallireducens" andS. putrefaciens (10).Attempts to grow D. desulfuricans with U(VI) as the sole

electron acceptor were unsuccessful. There was no growthor U(VI) reduction when D. desulfuricans was inoculatedinto the previously described medium (26) that supported thegrowth of "G. metallireducens" and S. putrefaciens onU(VI). When U(VI) (2 to 8 mM) was added to a sulfate-grown culture of D. desulfuricans when the sulfate wasdepleted, U(VI) was reduced but there was no cell growth.Addition of as much as 5 mM uranium to the sulfate-containing medium routinely used to grow D. desulfuricansdid not prevent growth, indicating that the lack of growth onU(VI) alone was not the result of U(VI) toxicity.

Uraninite formation. U(VI) reduction in cell suspensionsof D. desulfuricans resulted in the formation of a blackprecipitate which, according to X-ray diffraction analysis,

VOL. 58, 1992

on May 10, 2021 by guest

http://aem.asm

.org/D

ownloaded from

852 LOVLEY AND PHILLIPS

E 0.75I-I

1-

D 0.50cr0

>1.0.25

1-

>_,

0 1 2 3 4 0 1 2 3 4HOURS HOURS

FIG. 1. (A) Loss of U(VI) and corresponding increase in U(IV) concentration over time when D. desulfuricans was suspended inbicarbonate buffer with lactate as the electron donor. (B) U(VI) concentrations over time with various electron donors and live cells of D.desulfuricans or with lactate as the electron donor and heat-killed cells.

was made up of the U(IV)-containing mineral uraninite(U02) (Fig. 4). Examination of the cell suspensions bytransmission electron microscopy indicated that the uraniumprecipitation was extracellular (Fig. 5). This is only thesecond report of biogenic uraninite. "G. metallireducens"was also recently found to form uraninite during U(VI)reduction (11).

Concurrent reduction of U(VI) and sulfate. Sulfate wouldtypically be present in many sedimentary environments inwhich U(VI) is reduced. Therefore, the influence of sulfateon U(VI) reduction was investigated. There was little or nosulfate reduction in bicarbonate buffer alone (Fig. 6A; datanot shown). The addition of cysteine to the wash and

1.0

E 0.6E

>

0.4

0.2 F

0

0 1 2HOURS

suspension buffer resulted in higher rates of sulfate reduc-tion, but had no effect on the rate of U(VI) reduction (Fig.6A and B). Therefore, studies of the potential for U(VI)reduction in the presence of sulfate were conducted by usingbicarbonate buffer amended with cysteine. Sulfate had nosignificant effect on U(VI) reduction, and U(VI) additionsdid not influence the rate of sulfate reduction; the two werereduced simultaneously (Fig. 6C).U(VI) reduction during sulfate reduction by D. desulfuri-

cans has previously been suggested to be the result ofnonenzymatic reduction of the U(VI) by sulfide generatedfrom sulfate reduction (32, 36). To investigate this possibil-ity, cells were suspended in bicarbonate buffer containing 1mM sulfide and incubated at various temperatures for 1 h.The temperature optimum for U(VI) reduction was consis-tent with an enzymatic rather than a nonenzymatic reduction

0.9

0.8

0.71

E 0.60-.

1- 0.5

0.41

0.313 4

FIG. 2. U(VI) reduction over time when equal amounts of cellprotein (2.1 mg) of strain GS-15 ("G. metallireducens"), D. desulfu-ricans, and S. putrefaciens were suspended in bicarbonate bufferwith acetate ("G. metallireducens") or lactate (D. desulfuricans andS. putrefaciens) as the electron donor.

0.2L_0

FIG. 3. Effects of priorreduction.

1 2HOURS

3 4

air exposure and of azide on U(VI)

GS-15

is

Shewanella |putrefaciens

-lA LIVEOAIR EXPOSEEAZIDE

APPL. ENVIRON. MICROBIOL.

:-D

q

r

on May 10, 2021 by guest

http://aem.asm

.org/D

ownloaded from

REDUCTION OF U(VI) BY DESULFOVIBRIO DESULFURICANS 853

FIG. 4. Results of X-ray diffraction analysis of the black precipitate formed during U(VI) reduction by D. desulfunicans.

of U(VI) by sulfide (Fig. 7). There was no reduction of U(VI)by sulfide at 30°C in buffer containing no cells (26).

DISCUSSION

The results demonstrate that D. desulfuricans can rapidlyreduce U(VI) to U(IV). This extends the range of microor-ganisms known to enzymatically reduce U(VI) beyond dis-similatory Fe(III)-reducing microorganisms. As discussed indetail below, this observation provides a potential mecha-nism for the precipitation of uranium that is observed insulfidogenic (sulfate-reducing) environments. Furthermore,it suggests another organism that could be used in theremediation of uranium-contaminated waters.

Enzymatic uranium reduction. The results demonstratethat D. desulfuricans reduces U(VI) directly rather than.indirectly through sulfide production with the subsequentnonenzymatic reduction of U(VI) by sulfide. D. desulfuni-cans readily reduced U(VI) in cell suspensions that had beenwashed to remove sulfide. Furthermore, even in the pres-ence of as much as 1 mM sulfide, D. desulfuricans had atemperature optimum for U(VI) reduction that was consis-tent with a direct, enzymatic reduction of U(VI). Previousstudies have demonstrated that under the assay conditionsused here, sulfide does not reduce U(VI) (26).A previous detailed study on the removal of U(VI) from

solution in cultures of D. desulfiuincans concluded that theU(VI) loss was the result of the combined effect of U(VI)adsorption onto cell walls and the nonenzymatic reduction ofthe adsorbed U(VI) by sulfide (32). However, the possibilitythat D. desulfuricans could directly reduce U(VI) was notinvestigated. There were no control experiments with met-abolically inactive cells or studies without sulfide present.Therefore, the results of that previous study are also consis-tent with the enzymatic reduction of U(VI) that was ob-served in the studies reported here.

In contrast to the respiratory Fe(III)-reducing microorgan-isms which can conserve energy from U(VI) reduction tosupport growth (26), attempts to grow D. desulfuricans withU(VI) as the electron acceptor were unsuccessful, eventhough the added U(VI) did not appear to be toxic. In asimilar manner, other strains of D. desulfuricans have beenfound to use 02 as a terminal electron acceptor, but couldnot be grown with 02 as the sole electron acceptor (7).

However, aerobic respiration did result in ATP formation(7). The potential for ATP formation during U(VI) reductionhas not yet been investigated.The finding that the U(VI)-reducing system in D. desulfu-

ricans is not irreversibly inhibited by exposure to atmo-spheric oxygen suggests that the purification of the compo-nents involved in U(VI) reduction may be approached bystandard biochemical techniques. Such studies should yieldfurther information on the potential for energy conservationthrough U(VI) reduction.

Geological significance. Many of the economically impor-tant uranium ores in the western United States are fromuranium accumulations known as sandstone or roll-typedeposits. Geological evidence suggests that these depositsare formed when aerobic, neutral to alkaline groundwaters,carrying dissolved U(VI)-carbonate complexes and oxidizedsulfur compounds, enter anaerobic zones (15, 17, 33). In theanaerobic zones, U(VI) is reduced to U(IV) and sulfate isreduced to sulfide with the formation of U(IV) and sulfideprecipitates. This has generally been attributed to sulfideproduction by sulfate reducers with the subsequent nonen-zymatic reduction of U(VI) by sulfide (15, 17, 20, 30, 32, 33,36).However, the results presented here suggest that the

concurrent precipitation of U(IV) and sulfide minerals inroll-type deposits might be directly attributed to enzymaticreactions carried out by D. desulfuricans and possibly othersulfate-reducing organisms. D. desulfunicans can enzymati-cally reduce U(VI) and sulfate simultaneously. Furthermore,at the circumneutral pH and low temperatures characteristicof the conditions during the formation of roll-type deposits,direct enzymatic reduction of U(VI) is a more likely mech-anism than production of sulfide with the subsequent non-enzymatic reduction of U(VI).

Dissimilatory Fe(III)-reducing microorganisms might alsobe responsible for some of the U(VI) reduction in roll-typedeposits (26). However, there is evidence in many instancesthat the sediments making up the zone of U(IV) depositionhad already undergone extensive sulfate reduction prior tothe introduction of U(VI)-bearing waters (17, 33). In theseinstances, it is unlikely that there were large active popula-tions of Fe(III)-reducing microorganisms during the U(VI)reduction because microbially reducible Fe(III) oxidesshould have been depleted from the sediments. Respiratory

VOL. 58, 1992

on May 10, 2021 by guest

http://aem.asm

.org/D

ownloaded from

854 LOVLEY AND PHILLIPS

0 00

cz Wo

_E

0.

CZii M

CZ

0-C 0

;E CZ

Ut C;0

U 3

0.

o 0

_

.Q .%

4-. CZ

0.

0o

CZ

oo

=o

-

S0.

EI

APPL. ENVIRON. MICROBIOL.

on May 10, 2021 by guest

http://aem.asm

.org/D

ownloaded from

REDUCTION OF U(VI) BY DESULFOVIBRIO DESULFURICANS 855

2 4HOURS

0 200MINUTES

FIG. 6. Reduction of sulfate (A) and U(VI) (B) over time in the presence and absence of cysteine (cys). (C) Reduction of sulfate and U(VI)over time in cell suspensions with sulfate alone, U(VI) alone, or sulfate and U(VI) as potential electron acceptors.

Fe(III) reducers do not appear to persist for long periods insubsurface habitats in the absence of Fe(III) reduction (22).The finding that D. desulfuricans can enzymatically re-

duce U(VI) also has implications for U(VI) reduction inmarine sediments. In oligotrophic marine sediments much ofthe U(VI) reduction appears to take place in the zone inwhich Fe(III) is the predominant electron acceptor fororganic matter oxidation (3, 5, 6, 14, 18, 37). In theseenvironments, Fe(III) reducers may be responsible for mostof the U(VI) reduction (26) since it could be expected thatFe(III) reducers are metabolically dominant over sulfate

^:Jo) 50-

0

a40-

D

30-

D 20-

z

w

o 10

10 20 35 50 80TEMPERATURE (OC)

FIG. 7. Extent of U(VI) reduction after 1 h at the stated temper-atures in cell suspensions amended with 1 mM sulfide. The initialU(VI) concentration was 0.35 mM.

reducers within the Fe(III) reduction zone (21, 24). How-ever, sulfate reducers could play an important role in U(VI)reduction in more eutrophic marine sediments in whichmicrobially reducible Fe(III) oxides are rapidly depleted,and therefore there is often not a distinct segregation be-tween the Fe(III)- and sulfate-reducing zones.The relative contribution of Fe(III) and sulfate reducers to

U(VI) reduction in sediments is unlikely to be affectedgreatly by the ability of Fe(III) reducers to conserve energyfrom U(VI) reduction and the apparent inability of D.desulfuricans to obtain energy for growth from this reaction.The concentrations of U(VI) in marine waters are ca. 12 nM(8). Therefore, even when energy can be conserved fromU(VI) reduction, this process can be expected to provide, atbest, a minimal amount of cell growth in marine sediments.The more important consideration controlling the relativeextent of U(VI) reduction by Fe(III) and sulfate reducers islikely to be the availability of Fe(III) oxides to support anFe(III)-reducing population.

Potential advantages of D. desulfuricans over Fe(III) reduc-ers for remediation of U(VI)-contaminated waters. The pre-cipitation of uranium from water as a result of microbialreduction of U(VI) to U(IV) is a potential mechanism for theremoval of uranium from contaminated waters that hasseveral possible advantages over present methods of ura-nium removal (11, 26). Although Fe(III)-reducing microor-ganisms were considered in previous evaluations of thistechnique, the use of D. desulfunicans might be more prac-tical in some applications. For example, a bioreactor forremoving uranium from solution could be established withD. desulfuricans if low concentrations of sulfate were in-cluded so that sulfate reduction could provide energy tosupport the growth of the U(VI)-reducing biomass. U(IV) isstable in the presence of sulfate. In contrast, there are

2]

E

0

0

0.6

E 0.4

>:

n

B_

Jcysk/Il

no cys-

2.5

_ 2.0

E

LLU-Dcn

_ 1.0

- 0.5300

VOL. 58, 1992

U.ZII

Uz )

on May 10, 2021 by guest

http://aem.asm

.org/D

ownloaded from

856 LOVLEY AND PHILLIPS

limitations to growing Fe(III) reducers on Fe(III) in a urani-um-removing bioreactor because soluble Fe(III) forms mayoxidize U(IV) to U(VI) and solid Fe(III) oxides may adsorbU(VI), forming large volumes of uranium-containing solidwaste (23). The relative merits of Fe(III) reducers and D.desulfuricans for U(VI) removal from contaminated watersare currently being evaluated.

In summary, the results demonstrate that D. desulfuricansis capable of U(VI) reduction, and this suggests that, inaddition to Fe(III) reducers, sulfate reducers may be respon-sible for the reductive precipitation of uranium in some

sedimentary environments. The enzymatic mechanisms forU(VI) reduction in D. desulfuricans are under investigation.

ACKNOWLEDGMENTS

We thank Terry Beveridge and Robert Harris for performing theelectron microscopy, Daniel Webster for X-ray diffraction analyses,Y. Gorby for helpful discussions, and Richard Smith for helpfulcomments on the manuscript.

REFERENCES1. Anderson, R. F., A. P. LeHuray, M. Q. Fleisher, and J. W.

Murray. 1989. Uranium deposition in Saanich Inlet sediments,Vancouver Island. Geochim. Cosmochim. Acta 53:2205-2213.

2. Balch, W. E., G. E. Fox, L. J. Magrum, C. R. Woese, and R. S.Wolfe. 1979. Methanogens: reevaluation of a unique biologicalgroup. Microbiol. Rev. 43:260-296.

3. Bonatti, E., D. E. Fisher, 0. Joensuu, and H. S. Rydell. 1971.Postdepositional mobility of some transition elements, phospho-rus, uranium and thorium in deep sea sediments. Geochim.Cosmochim. Acta 35:189-201.

4. Brierley, J. A., and C. L. Brierley. 1980. Biological methods toremove selected inorganic pollutants from uranium mine waste-water, p. 661-667. In P. A. Trudinger, M. R. Walter, and B. J.Ralph (ed.), Biogeochemistry of ancient and modern environ-ments. Springer-Verlag, New York.

5. Cochran, J. K., A. E. Carey, E. R. Sholkovitz, and L. D.Surprenant. 1986. The geochemistry of uranium and thorium incoastal marine sediments and sediment pore waters. Geochim.Cosmochim. Acta 50:663-680.

6. Colley, S., and J. Thomson. 1985. Recurrent uranium relocationsin distal turbidities emplaced in pelagic conditions. Geochim.Cosmochim. Acta 49:2339-2348.

7. Dilling, W., and H. Cypionka. 1990. Aerobic respiration insulfate-reducing bacteria. FEMS Microbiol. Lett. 71:123-128.

8. Durrance, E. M. 1986. Radioactivity in geology. John Wiley &Sons, Inc., New York.

9. Ghiorse, W. C. 1988. Microbial reduction of manganese andiron, p. 305-331. In A. J. B. Zehnder (ed.), Biology of anaerobicmicroorganisms. John Wiley & Sons, Inc., New York.

10. Gorby, Y. A. 1991. Unpublished data.11. Gorby, Y. A., and D. R. Lovley. 1991. Enzymatic uranium

precipitation. Environ. Sci. Technol. 26:205-207.12. Harrison, R. K. 1975. Concretionary concentrations of the rarer

elements in permo-triassic red beds of South-West England.G.B. Geol. Surv. Bull. 52:1-26.

13. Hoffman, B. A. 1990. Reduction spheroids from northern Swit-zerland: mineralogy, geochemistry and genetic models. Chem.Geol. 81:55-81.

14. Honeyman, B. D., and P. H. Santschi. 1988. Metals in aquaticsystems. Environ. Sci. Technol. 22:862-871.

15. Hostetler, P. B., and R. M. Garrels. 1962. Transportation andprecipitation of uranium and vanadium at low temperatures withspecial reference to sandstone-type uranium. Econ. Geol. 57:137-167.

16. Hungate, R. E. 1969. A roll tube method for cultivation of strictanaerobes. Methods Microbiol. 3B:117-132.

17. Jensen, M. L. 1958. Sulfur isotopes and the origin of sandstone-type uranium deposits. Econ. Geol. 53:598-616.

18. Kadko, D. 1980. A detailed study of some uranium seriesnuclides at an abyssal hill area near the East Pacific rise at8'45'N. Earth Planet. Sci. Lett. 51:115-131.

19. Klinkhammer, G. P., and M. R. Palmer. 1991. Uranium in theoceans: where it goes and why. Geochim. Cosmochim. Acta55:1799-1806.

20. Langmuir, D. 1978. Uranium solution-mineral equilibria at lowtemperatures with applications to sedimentary ore deposits.Geochim. Cosmochim. Acta 42:547-569.

21. Lovley, D. R. 1991. Dissimilatory Fe(III) and Mn(IV) reduction.Microbiol. Rev. 55:259-287.

22. Lovley, D. R., F. H. Chapelle, and E. J. P. Phillips. 1990.Fe(III)-reducing bacteria in deeply buried sediments of theAtlantic Coastal Plain. Geology 18:954-957.

23. Lovley, D. R., and E. J. P. Phillips. Unpublished data.24. Lovley, D. R., and E. J. P. Phillips. 1987. Competitive mecha-

nisms for inhibition of sulfate reduction and methane productionin the zone of ferric iron reduction in sediments. Appl. Environ.Microbiol. 53:2636-2641.

25. Lovley, D. R., and E. J. P. Phillips. 1988. Novel mode ofmicrobial energy metabolism: organic carbon oxidation coupledto dissimilatory reduction of iron or manganese. AppI. Environ.Microbiol. 54:1472-1480.

26. Lovley, D. R., E. J. P. Phillips, Y. A. Gorby, and E. R. Landa.1991. Microbial reduction of uranium. Nature (London) 350:413-416.

27. Lovley, D. R., E. J. P. Phillips, and D. J. Lonergan. 1989.Hydrogen and formate oxidation coupled to dissimilatory reduc-tion of iron or manganese by Alteromonas putrefaciens. Appl.Environ. Microbiol. 55:700-706.

28. Lovley, D. R., J. F. Stolz, G. L. Nord, and E. J. P. Phillips. 1987.Anaerobic production of magnetite by a dissimilatory iron-reducing microorganism. Nature (London) 330:252-254.

29. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall.1951. Protein measurement with the Folin phenol reagent. J.Biol. Chem. 193:265-275.

30. Maynard, J. B. 1983. Geochemistry of sedimentary ore depos-its. Springer-Verlag, New York.

31. Miller, T. L., and M. J. Wolin. 1974. A serum bottle modifica-tion of the Hungate technique for cultivating obligate anaerobes.Appl. Microbiol. 27:985-987.

32. Mohagheghi, A., D. M. Updegraff, and M. B. Goldhaber. 1985.The role of sulfate-reducing bacteria in the deposition of sedi-mentary uranium ores. Geomicrobiol. J. 4:153-173.

33. Reynolds, R. L., and M. B. Goldhaber. 1983. Iron disulfideminerals and the genesis of roll-type uranium deposits. Econ.Geol. 78:105-120.

34. Taylor, G. H. 1979. Biogeochemistry of uranium minerals, p.485-514. In P. A. Trudinger and D. J. Swaine (ed.), Bio-geochemical cycling of mineral-forming elements. Elsevier Sci-ence Publishing, Inc., New York.

35. Veeh, H. H. 1967. Deposition of uranium from the ocean. EarthPlanet. Sci. Lett. 3:145-150.

36. Viragh, K., and J. Szolnoki. 1970. Bakteriumok szerepe amecseki uranere keletkezeseben es kesobbi athalmozasaban.Foldt. Kozl. 100:43-54.

37. Wallace, H. E., J. Thomson, T. R. S. Wilson, P. P. E. Weaver,N. C. Higgs, and D. J. Hydes. 1988. Active diagenetic formationof metal-rich layers in N.E. Atlantic sediments. Geochim.Cosmochim. Acta 52:1557-1569.

38. Woolfolk, C. A., and H. R. Whiteley. 1962. Reduction ofinorganic compounds with molecular hydrogen by Micrococcuslactilyticus. J. Bacteriol. 84:647-658.

APPL. ENVIRON. MICROBIOL.

on May 10, 2021 by guest

http://aem.asm

.org/D

ownloaded from