Vol. 57, No. 1APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Jan. 1991, p. 207-2110099-2240/91/010207-05$02.00/0Copyright (© 1991, American Society for Microbiology

Ferric Iron Reduction by Acidophilic Heterotrophic BacteriaD. BARRIE JOHNSON* AND STEPHEN McGINNESS

School of Biological Sciences, University College of North Wales,Bangor, Gwynedd LL57 2UW, United Kingdom

Received 13 August 1990/Accepted 1 November 1990

Fifty mesophilic and five moderately thermophilic strains of acidophilic heterotrophic bacteria were testedfor the ability to reduce ferric iron in liquid and solid media under aerobic conditions; about 40% of themesophiles (but none of the moderate thermophiles) displayed at least some capacity to reduce iron. Both ratesand extents of ferric iron reduction were highly strain dependent. No acidophilic heterotroph reduced nitrateor sulfate, and (limited) reduction of manganese(IV) was noted in only one strain (Acidiphilium facilis), an

acidophile which did not reduce iron. Insoluble forms of ferric iron, both amorphous and crystalline, were

reduced, as well as soluble iron. There was evidence that, in at least some acidophilic heterotrophs, ironreduction was enzymically mediated and that ferric iron could act as a terminal electron acceptor. Inanaerobically incubated cultures, bacterial biomass increased with increasing concentrations of ferric but notferrous iron. Mixed cultures of Thiobacillus ferrooxidans or LeptospiriUum ferrooxidans and an acidophilicheterotroph (SJH) produced sequences of iron cycling in ferrous iron-glucose media.

The ability to reduce ferric iron (Fe3") to ferrous iron(Fe2") is widespread among bacteria, such as members ofthe genera Pseudomonas, Bacillus, Bacteroides, and Des-ulfovibrio (9, 13). Some fungi also have this capacity (14).Although reduction is favored by anaerobic conditions, itmay also occur in the presence of oxygen, although sponta-neous chemical reoxidation may occur rapidly in nonacidic,aerobic environments (13). There have been a number ofproposals regarding the mechanism of bacterial ferric ironreduction (8). For example, when the physicochemical na-ture of the environment (pH, redox potential) is changed,bacteria may induce spontaneous chemical reduction (e.g.,by Vibrio spp.). Alternatively, bacteria may use ferric iron asa hydrogen sink by using a mechanism that is not clearlydefined (the ferrireductase system), in which the enzymefunctions either via an electron transport chain or substrate-level phosphorylation. A third possibility is that bacteriareduce ferric iron by way of an electron transport system,which may or may not involve nitrate reductase. While manyferric iron reducers are facultative anaerobes which alsopossess nitrate reductases, correlation between the two isnot absolute (9).

Ferric iron reduction has been most commonly observedin neutrophilic, heterotrophic bacteria, but it has also beennoted in some acidophilic chemolithotrophs, including meso-philic thiobacilli and the extreme thermophile Sulfolobusacidocaldarius (2). Thiobacillus thiooxidans can reduce fer-ric iron when growing aerobically on elemental sulfur, and T.ferrooxidans can do so either anaerobically on sulfur at pHsbelow ca. 1.8 or aerobically on sulfur at extremely low(<1.3) pH (15). T. ferrooxidans can also reduce tetravalentmanganese and copper(II) when oxidizing elemental sulfur(16, 17).

Highly acidic environments (pH 3.0 or less) are alsoknown to contain obligately acidophilic heterotrophic bacte-ria (3, 4, 18). Relatively few of these bacteria have been fullycharacterized, and those which have appear to belongmainly to the genus Acidiphilium, such as Acidiphiliumcryptum (3); A. angustum, A. facilis, and A. rubrum (19);

* Corresponding author.

207

and A. organovorum (11). These are gram-negative, obli-gately acidophilic, rod-shaped bacteria, and they have alsobeen reported to be obligate aerobes. The environments inwhich they are characteristically found are invariably rich inferrous and ferric iron. Indeed, some have been isolateddirectly from supposedly pure cultures of T. ferrooxidansgrowing on ferrous sulfate (e.g., see reference 4). Thesolubility of ferric iron is highly pH dependent, and acidmine drainage waters tend to contain high concentrations ofthis oxidized species (e.g., see reference 5). There is, there-fore, a major contrast between neutrophiles, which live inenvironments in which ferric iron is found predominantly ininsoluble amorphous or crystalline form, and acidophiles,which are often exposed to high concentrations of solubleferric iron. The potential for bacterially mediated iron reduc-tion might thus appear to be somewhat greater for acido-philic heterotrophs.

MATERIALS AND METHODS

Bacteria. Pedigreed strains of A. cryptum, A. angustum,A. facilis, A. rubrum and A. organovorum and the mix-otrophic acidophile T. acidophilus were obtained eitherdirectly from national culture collections or as gifts fromcolleagues. Other mesophilic heterotrophic bacteria wereisolated from cultures of T. ferrooxidans (4) from a disusedpyrite mine in North Wales (5) or from the Blackbird CobaltMine (Noranda Mining Co., Cobalt, Idaho). A total of 50mesophilic heterotrophs were screened; these were domi-nantly Acidiphilium-like bacteria and were differentiatedfrom each other in terms of morphological and behavioralcharacteristics. In addition, five moderately thermophilicheterotrophic bacteria were isolated from an acidic hotspring (pH 2.5, 50°C), Frying Pan Hot Spring in YellowstoneNational Park, Wyoming. Bacterial isolates were obtainedeither by direct plating of environmental samples ontoFeTSB solid medium (6) or following enrichment in liquidmedium. FeTSB consisted of 25 mM ferrous sulfate and0.025% (wt/vol) tryptone soya broth in phosphate-free basalsalts gelled with agarose; the liquid medium had the samecomposition except that agarose was omitted. A more de-tailed account is given elsewhere (7).

on May 16, 2018 by guest

http://aem.asm

.org/D

ownloaded from

208 JOHNSON ANDMcGINNESSAPLENRO.McBI.

Screening for ferric iron reduction. All acidophilic hetero-

trophs were screened for the ability to reduce ferric iron in

the presence of oxygen. This involved streaking solid media

containing ferric iron and glucose with actively growingbacteria. The medium was based on the FeTSB medium

described by Johnson et al. (6) but with the amendments that

ferric sulfate was substituted for ferrous sulfate (final con-

centration, 25 mM; supplied from a 500 mM stock solution,

pH 1.7, sterilized by membrane filtration) and that glucosewas added (final concentration, 10 mM) from a separately

sterilized, unacidified stock solution. Inoculated plates were

incubated aerobically for up to 20 days at 280C (450C for

thermophiles) and checked periodically for signs of ferric

reduction (estimated visually by observing decoloration of

the plates). The fact that decoloration was attributable to

ferric reduction rather than a change in medium pH was

confirmed in liquid media (described below). Bacteria were

categorized into four groups, depending on their abilities to

reduce iron under these conditions. Some isolates were

unable to grow on the medium described, either because

ferric iron was toxic to them at 25 mM or because they were

inhibited by 10 mM glucose at the pH of the medium (it was

noted that glucose at this concentration became increasinglytoxic to many acidophilic heterotrophs as the medium pHfell below ca. 2.5). Ferric iron reduction was also tested on

solid medium, as described, by using glycerol (final concen-

tration, 10 mM) rather than glucose as the principal carbon

source.:Ferric iron reduction in liquid media. Media containing 10

to 50 mM ferric sulfate, 10 mM glucose, and 0.025% tryptone

soya broth-basal salts, pH 2.0, were prepared. Inoculated

cultures were incubated aerobically (50 ml of medium in

100-ml Erlenmeyer flasks, shaken [100 rpm] or unshaken) or

anaerobically (in universal tubes completely filled with ni-

trogen-sparged media) at 28 or 450C. Ferric iron reduction

was monitored by estimating ferrous iron concentrations

(titrating with 1 mM KMnO4 in 1 M H2S04). Microbial

biomass was estimated by measurement of optical density at

550 nm and by direct cell counts (acridine orange-stainedcells adsorbed onto Nuclepore 0.2-p.m-pore-size mem-

branes).

Reduction of insoluble ferric iron. Amorphous insoluble

ferric hydroxysulfate was prepared by adding ferric sulfate

(final concentration, 25 mM) from the stock solution to a 10

mM glucose-tryptone soya broth medium, raising the pH to

2.5 (NaOH), and boiling it for 10 s to induce precipitationand flocculation. Crystalline ferric sulfate, in the form of

natrojarosite, was prepared microbiologically by growing T.

ferrooxidans in 100 mM ferrous sulfate-200 mM sodium

sulfate-basal salts medium at pH 2.5, discarding the spent

liquor, and replacing it with 10 mM glucose-tryptone soya

broth medium. The characteristic yellow precipitate which

adheres strongly to the sides of the flask under these

conditions has been shown to be primarily natrojarosite (10).Cultures were inoculated with the acidophilic isolate SJH

and incubated unshaken at 280C.

Iron cycling in mixed cultures. Mixed communities of

iron-oxidizing and heterotrophic acidophiles were grown in

batch cultures containing T. ferrooxidans (NCIB 8455) or

Leptospirillum ferrooxidans (Cae Coch isolate SJ1) with

heterotrophic isolate SJH. Cultures (100 ml in 250-ml Erlen-

meyer flasks) were grown in 25 mM ferrous sulfate medium

supplemented with 10 mM glucose and 0.025% tryptone soya

broth and incubated, shaken or unshaken, at 280C.

Nitrate and manganese reduction by acidophilic hetero-

trophs. Anaerobic cultures containing 10 mM glucose and

TABLE 1. Classification of acidophilic heterotrophic bacteria interms of aerobic ferric iron reduction potential'

Class Total no. of Organismsmesophiles

1 9 A. organovorum NCIB 11745,NCIB 11746, NCIB 12826 (SJH)

2 13 T. acidophilus, A. cryptum3b ~~22 A. facilis, A. angustum, A. rubrum

NCIB 118224 6 UnclassifiedI Based on visual assessment of decoloration of ferric iron-glucose plates as

described in Results.b All five of the thermophiles tested were in class 3.

0.025% tryptone soya broth were supplemented with 25 mMsodium nitrate or 0.1% (wt/vol) finely ground (<63-p.mparticle diameter) manganese dioxide (analytical grade;BDH Ltd.), and the medium pH was adjusted to 2.5.Cultures were incubated for up to 4 weeks at 280C. Thepresence of nitrite was tested for by addition of sulfanilicacid, followed by 5-amino-2-naphthylenesulfonic acid (1).Manganese reduction was estimated visually (noting thedisappearance or otherwise of black, insoluble MnO2) andby analysis of soluble manganous ions by using atomicabsorption spectrophotometry.

RESULTS

A subjective classification of acidophilic heterotrophicbacteria based on their abilities to reduce ferric iron in thepresence of oxygen with glucose as the principal energysource is shown in Table 1. The criteria used for classifica-tion were based on the extent of decoloration of plates,estimated visually, when examined after 10 and 20 days ofincubation (freshly prepared plates were bright orange).Class 1 isolates showed partial decoloration after 10 daysand complete or nearly complete decoloration after 20 days,class 2 isolates showed little or no decoloration after 10 daysand only partial decoloration (restricted to the near vicinityof colonies) after 20 days, class 3 isolates showed nodiscernible decoloration after 20 days, and class 4 isolatesfailed to grow on the solid medium used. A single acidophilicunidentified yeast isolate that was tested was in class 3. Allof the moderately thermophilic isolates also failed to reduceferric iron in liquid or solid media. The distributions ofmesophilic heterotrophic isolates among the four classeswere very similar for bacteria isolated from either the CaeCoch or Cobalt mine; for example, 42% of 27 isolates fromthe Cae Coch mine and 40% of 15 isolates from the Cobaltmine were in class 1 or 2. The patterns of ferric ironreduction when acidophilic heterotrophs were grown onglycerol-ferric iron plates were very similar to those dis-played on glucose-ferric iron medium. Four class 4 isolates(on glucose plates) grew without reducing ferric iron onglycerol plates, and three class 3 isolates reduced ferric ironwhen grown on glycerol medium. In other cases, the cate-gorizations of acidophilic heterotrophs were the same onboth media.

Variations in the rates of ferric iron reduction by some ofthe acidophilic heterotrophs tested are shown in Fig. 1.Trends very similar to those observed on solid media wereseen; for example, A. facilis (a class 3 strain) displayed noferric iron reduction and A. cryptum and T. acidophilus(class 2 strains) only partially reduced the available ferric

APPL. ENVIRON. MICROBIOL.

on May 16, 2018 by guest

http://aem.asm

.org/D

ownloaded from

FERRIC IRON REDUCTION BY ACIDOPHILIC HETEROTROPHS

TABLE 2. Reduction of iron, manganese, and nitrate by someacidophilic heterotrophic bacteria

Reduction off:Organism

Fe3+ Mn4+

A. cryptum (+)A. facilis (+)A. organovorum +T. acidophilus (+))NCIB 12826 (SJH) +

a Symbols: +, complete reduction; (+), partial reduction; -, no reduction.None of the organisms tested reduced NO3-.

0 25 50 75 100

Time (hr)

FIG. 1. Ferric iron reduction by some acidophilic heterotrophicbacteria grown aerobically in glucose-tryptone soya broth liquidmedium. Ferric iron reduction was monitored by estimating ferrousiron concentrations (titration with potassium permanganate). Sym-bols: A, A. cryptum; *, A. facilis; *, T. acidophilus; V, NCIB11745; *, NCIB 12826 (SJH).

iron over the 100-h incubation period; strains SJH (NCIB12826) and NCIB 11745 completely reduced the availableferric iron, although there were differences in the rates ofreduction (SJH was faster); both were categorized as class 1

isolates on solid media. A. organovorum reduced iron at a

rate similar to that of SJH (data not shown). Also, in culturesof class 1 strain SJH, ferric iron reduction correlated withvegetative growth, but this was not the case with A. cryptum(or T. acidophilus), whose iron reduction was confined to thelate logarithmic and stationary growth phases (Fig. 2).None of the five mesophilic acidophiles tested had the

ability to reduce nitrate, and only one (A. facilis) reducedmanganese(IV), and even in this case the extent of reduction

>10 J H ~ -1

E

'̂..! ~~~~~~~~~~10770

L)j

10631U1)

0 35 70 105 140

Time (hr)FIG. 2. Correlation between growth and ferric iron reduction in

cultures of A. cryptum and isolate SJH. Cultures were grown inglucose-tryptone soya broth liquid media; cell numbers were esti-mated by direct counts, and ferric iron reduction was determined byestimating ferrous iron concentrations (titration with potassiumpermanganate). Symbols: A, cell number (A. cryptum); A, ferriciron reduction (A. cryptum); 0, cell number (SJH); 0, ferric ironreduction (SJH).

was not great and was apparent only in cultures that under-went long-term (14 to 28 days) incubation (Table 2). Therewas no evidence of any sulfate reduction (to sulfide or

elemental sulfur) in any culture.Isolate SJH, which had proved to be one of the most

powerful reducers of ferric iron, was used exclusively inmany subsequent experiments. Reduction of ferric ironoccurred in shaken (100 rpm) and unshaken cultures, as wellas on solid media (all incubated aerobically). SJH had theability to reduce insoluble, as well as soluble, forms ofiron(III). Amorphous ferric hydroxysulfate was reduced(and solubilized) within 4 days, and natrojarosite was re-

duced within 11 days, in glucose cultures at 28°C. In thelatter cultures, the strongly adhering ferric deposits typicallyfound in oxidized cultures of T. ferrooxidans were com-

pletely removed, producing turbid off-white cultures in cleanflasks. Munch and Ottow (12) had earlier noted that amor-phous ferric iron was reduced more rapidly by neutrophilicbacteria than was crystalline iron oxide. The effect of surfacearea of insoluble ferric compounds was not investigated,although it may be anticipated that this would greatly affectthe rate of ferric iron reduction. Ferric iron reduction by SJHalso occurred when substrates other than glucose were used,e.g., xylose, glycine, arginine, and sodium citrate (all at 10mM), but iron reduction was much slower when SJH was

grown on sodium malonate. Also, it was noted that growth,but not iron reduction, took place in glucose cultures incu-bated at 45°C.Use of ferric iron as a direct electron acceptor or as a

general environmental electron sink might be expected topromote bacterial growth in the absence of oxygen. This wasthe case with isolate SJH (Fig. 3). Numbers increased withincreasing concentrations of ferric but not ferrous iron; thesecultures were grown in strictly anoxic conditions, and theyield of bacterial biomass was equivalent to 720 mg (dryweight) per mol of ferric iron reduced. Inference of enzymicaction in ferric iron reduction by SJH also came from theobservations that when bacteria were removed (by mem-

brane filtration or centrifugation) from actively reducingcultures reduction did not continue and that heating culturesto 70°C for 15 min halted iron reduction. No organic acidswere detected (by using a high-pressure liquid chromato-graph fitted with an Aminex ion-exclusion column; Bio-RadInc.) in spent media of SJH grown anaerobically withoutadded ferric iron.Mixed cultures of either T. ferrooxidans or L. ferrooxi-

dans with heterotroph SJH produced sequences of ironcycling between the ferrous and ferric ionic states (Fig. 4).With T. ferrooxidans, ferrous iron oxidation in mixed cul-tures began to slow down (relative to the pure cultures)approximately midway through the first phase of oxidation

20

E

~0-0a)w-

VOL. 57, 1991 209

on May 16, 2018 by guest

http://aem.asm

.org/D

ownloaded from

210 JOHNSON AND McGINNESS

3C

2C

2015

1C

FIG.acidop}soya brto inocand celferroustures.

and wasthere wferrousreducticmixed eabout 4oxidaticnet reduThe resdifferenwas toxpure cuferrooxi(almost

E

-oU)

+

a)LL

-ic

FIG.and hete

ferrous

28°C. Fe

potassiuarose froT. ferrocidans; A

one phase of oxidation; the negative values of ferrous ironoxidized also indicate reduction of ferric iron in the inoculumby heterotroph SJH. In mixed cultures not amended withglucose (data not shown), there were no phases of ironreduction; oxidation rates were similar in pure and mixed T.ferrooxidans cultures, but oxidation proceeded at a fasterrate in L. ferrooxidans-SJH cultures than in pure cultures ofthe chemolithotroph, a phenomenon that has also beenobserved with pyrite cultures (7). Shaken cultures of T.ferrooxidans also showed sequences of iron oxidation andreduction (data not shown), although the cycling patternswere shifted more in the direction of ferric iron than in theless well-aerated, unshaken cultures. Changing the concen-tration of glucose in T. ferrooxidans-SJH mixed cultures

0o 5 10 15 20 25 30 modified the cycling patterns; cycling was observed in 1mMglucose cultures (with less pronounced reduction), and 100

Iron concentration (mM) mM glucose cultures showed three distinct ferrous-ferriciron cycling phases. The extent of ferric iron reduction by

3. Effects of ferrous and ferric iron on cell numbers of heterotroph SJH in mixed cultures was therefore limited byhilic bacterium SJH grown anaerobically in glucose-tryptone the availability of glucose. The fact that reduction ceasedroth medium. Ferrous or ferric sulfate was added aseptically when glucose was depleted was confirmed by addition ofulated media, cultures were incubated at 28°C for 14 days11 numbers were estimated by direct counts. Symbols: * glucose (final concentration, 10 mM) to mixed cultures aftersulfate-amended cultures; 0, ferric sulfate amended cul: approximately 600 h of incubation; a further cycle of iron

reduction which was complete within 50 h was observed.

DISCUSSIONlUI1lWJVU Uy I1IrL IIII C.AIJII, aLLI"VW% In many ways, it is not surprising that many acidophilic

as no detectable ferric iron (the negative values of Inemany bacteIa potsss theabty torceiferriciron oxidized shown in Fig. 2 came about from heterotrophic bacteria possess the ability to reduce ferric

rn of the ferric iron included in the inoculum). The iron, since their natural environments (such as acid mine

cultures remained poised at 100% ferrous iron for drainage) are invariably iron rich. It is apparent, however,

days and then went through a second phase of that this ability is to be found in all such bacteria. Over

n, this time to completion; a second, minor phase of 50% of the isolates tested did not reduce ferric iron when

iction was observed as the culture aged to ca. 500 h grown either glucose or glycerol solid media or when

iponse of the L- ferrooxidans strain used was quite grown under strictly anoxic conditions in liquid cultures.,pin that the concentration of glucose used (10 mM) Even among bacteria which have the ability, it may beit,cinthatpue culturesn(gratio occuref iglucose -feudeveloped to a greater or lesser extent; indeed, it has not

ic to pure cultures (growth occurred in glucose-free been ascertained whether spontaneous chemical reduction inLltures; data not shown). Mixed cultures of this L.

epnet hne nrdxptnilmgtb epnilidans strain and heterotroph SJH took a long time response to changes in redox potential might be responsible

400 h) to begin iron oxidation and there was only for the reduction observed in cultures of class 2 acidophiles(although why presumably similar redox depressions in class3 cultures did not induce reduction would, in that case, bedifficult to explain). It is also possible that when grown on

glucose class 3 isolates utilized a fermentative pathway,thereby eliminating the need for ferric iron to act as a

terminal electron acceptor or general electron sink, although

nearly all of the isolates categorized in class 3 when grownon glucose also failed to reduce ferric iron when grown on

glycerol (a nonfermentable substrate). Among the acido-philic heterotrophs which were capable of reducing ferriciron, there may be more than one biochemical mechanisminvolved. For example, although no organic acids (whichmay induce iron reduction) were detected in spent media of

SJH, it has not been ascertained whether this is also the casewith other acidophilic heterotrophic bacteria, for example,class 2 isolates.

0 120 240 360 480 600 It is clear that strictly anoxic conditions are not required

Time (h) for ferric iron reduction by acidophilic heterotrophs, as

..rnccigie . . shown by the phenomenon that occurred on aerobically4. Iron cycling In pure and mixed cultures of iron-oxidizing incubated solid media, and reduction in shake flask cultures.

rotrophic acidophiles. Aerobic cultures containing 25 mM On the other hand, these experiments show that many

sulfate and 10 mM glucose were incubated, unshaken, at . hrrous iron concentrations were estimated by titration with acidophlic heterotrophic bacteria are not obligate aerobes,

m permanganate. Negative values for oxidized ferrous iron as was thought earlier (e.g., see references 4 and 19), but can)m reduction of ferric iron added in the inocula. Symbols: , live anaerobically by using ferric iron as an electron accep-

xidans; *, T. ferrooxidans-heterotroph SJH; A, L. ferroox- tor. Although a more limited number of moderately thermo-, L. ferrooxidans-heterotroph SJH. philic heterotrophs were tested, none was capable of reduc-

0x

m-81a)

APPL. ENVIRON. MICROBIOL.

z fnilnixipti hii npt irnn rptiiirtinn nt thi- P-nci nf whirh

on May 16, 2018 by guest

http://aem.asm

.org/D

ownloaded from

FERRIC IRON REDUCTION BY ACIDOPHILIC HETEROTROPHS

ing ferric iron, and isolate SJH (which has a widetemperature spectrum) also failed to reduce iron at 45°C,indicating that bacterial iron reduction is not a common

feature at elevated temperatures, although extreme thermo-philes, such as Sulfolobus acidocaldarius, can reduce iron(2).The lack of ability of any acidophilic heterotroph to reduce

nitrate implies that ferric iron reduction in the bacteria testedis not mediated through nitrate reductases. Nitrate generallyoccurs in very low concentrations in acid mine drainage, andthe dominant form of inorganic nitrogen is often ammonium(e.g., see reference 5). The evidence does suggest stronglythat, with isolate SJH at least, reduction is enzyme medi-ated; ferric iron reduction by T. ferrooxidans has also beenshown to involve an enzyme, sulfur:ferric ion oxidoreduc-tase (16, 17). Interestingly, only one acidophilic heterotro-phic bacterium (A. facilis) was able to reduce manga-

nese(IV), and this microorganism was unable to reduceiron(III); the very slow rate of reduction implies nonenzymicmediation.The environmental significance of ferric iron reduction by

acidophilic heterotrophic bacteria may be considerable. Forexample, in acid mine drainage waters, particularly thosecontaining significant amounts of dissolved organic matter,heterotrophs may continuously reverse the process of ironoxidation by chemolithotrophic acidophiles, and this wouldneed to be borne in mind when calculating, for example, thestoichiometry of pyrite oxidation in situ. The process couldalso be of use in situations in which rapid reversal of ironoxidation may be required, to produce a liquor containingiron only in the ferrous state, and when dissolution andsolubilization of ferric deposits, both crystalline and amor-phous, is desired.

ACKNOWLEDGMENT

Stephen McGinness is grateful to the Natural Environment Re-search Council, United Kingdom, for the provision of a studentship(GT4/87/ALS/41).

REFERENCES1. Baird-Parker, A. C. 1979. Methods for identifying Staphylo-

cocci, p. 201-210. In F. A. Skinner and D. W. Lovelock (ed.),Identification methods for microbiologists. Academic Press,Inc. (London), Ltd., London.

2. Brock, T. D., and J. Gustafson. 1976. Ferric iron reduction bysulfur- and iron-oxidizing bacteria. Appl. Environ. Microbiol.32:567-571.

3. Harrison, A. P., Jr. 1981. Acidiphilium cryptum gen. nov., sp.nov., heterotrophic bacterium from acidic mine environments.

Int. J. Syst. Bacteriol. 32:327-332.4. Johnson, D. B., and W. I. Kelso. 1983. Detection of heterotro-

phic contaminants in cultures of Thiobacillus ferrooxidans andtheir elimination by subculturing in media containing coppersulfate. J. Gen. Microbiol. 129:2969-2972.

5. Johnson, D. B., W. I. Kelso, and D. A. Jenkins. 1979. Bacterialstreamer growth in a disused pyrite mine. Environ. Pollut.18:107-118.

6. Johnson, D. B., J. H. M. Macvicar, and S. Rolfe. 1987. A newmedium for the isolation and enumeration of Thiobacillus fer-rooxidans and acidophilic heterotrophic bacteria. J. Microbiol.Methods 7:9-18.

7. Johnson, D. B., M. F. Said, M. A. Ghauri, and S. McGinness.1990. Isolation of novel acidophiles and their potential use inbioleaching operations, p. 403-414. In J. Salley, R. G. L.McCready, and P. L. Wichlacz (ed.), Biohydrometallurgy 1989.Canmet, Ottawa, Ontario, Canada.

8. Jones, J. G. 1986. Iron transformations by freshwater bacteria.Adv. Microb. Ecol. 9:149-185.

9. Jones, J. G., W. Davison, and S. Gardener. 1984. Iron reductionby bacteria: range of organisms involved and metals reduced.FEMS Microbiol. Lett. 21:133-136.

10. Lazaroff, N., W. Sigal, and A. Wasserman. 1982. Iron oxidationand precipitation of ferric hydroxysulfates by resting Thiobacil-lus ferrooxidans cells. Appl. Environ. Microbiol. 43:924-938.

11. Lobos, J., T. E. Chisolm, L. H. Bopp, and D. S. Holmes. 1986.Acidiphilium organovorum sp. nov., an acidophilic heterotrophisolated from a Thiobacillus ferrooxidans culture. Int. J. Syst.Bacteriol. 36:139-144.

12. Munch, J. C., and J. C. G. Ottow. 1982. Einfluss von Zellkon-takt und Eisen(III)-oxidform auf die bakterielle Eisenreduktion.Z. Pflanzenernaehr. Dueng. Bodenkd. 145:66-77.

13. Nealson, K. 1983. The microbial iron cycle, p. 159-190. In W. E.Krumbein (ed.), Microbial geochemistry. Blackwell ScientificPublications, Ltd., Oxford.

14. Ottow, J. C. G., and A. von Klopotek. 1969. Enzymatic reduc-tion of iron oxide by fungi. Appl. Microbiol. 18:41-43.

15. Sand, W. 1989. Ferric iron reduction by Thiobacillus ferrooxi-dans at extremely low pH-values. Biogeochemistry 7:195-201.

16. Sugio, T., Y. Tsujita, K. Hirayama, K. Inagaki, and T. Tano.1988. Mechanism of tetravalent manganese reduction with ele-mental sulfur by Thiobacillus ferrooxidans. Agric. Biol. Chem.52:185-190.

17. Sugio, T., Y. Tsujita, K. Inagaki, and T. Tano. 1990. Reductionof cupric ions with elemental sulfur by Thiobacillus ferrooxi-dans. Appl. Environ. Microbiol. 56:693-696.

18. Wichlacz, P. L., and R. F. Unz. 1981. Acidophilic, heterotrophicbacteria of acidic mine waters. Appi. Environ. Microbiol.41:1254-1261.

19. Wichlacz, P. L., R. F. Unz, and T. A. Langworthy. 1986.Acidiphilium angustum sp. nov., Acidiphilium facilis sp. nov.,and Acidiphilium rubrum sp. nov.: acidophilic heterotrophicbacteria from acidic coal mine drainage. Int. J. Syst. Bacteriol.36:197-201.

VOL. 57, 1991 211

on May 16, 2018 by guest

http://aem.asm

.org/D

ownloaded from