Microbial Arsenic Metabolism:New Twists on an Old PoisonDuring the early anoxic phase on Earth, some microbes depended onarsenic to respire

John F. Stolz, Partha Basu, and Ronald S. Oremland

Imagine a planet with an atmospherelacking oxygen, its landscape dottedwith volcanic craters, caustic oceans,and basins of brine. Yet, amazingly,these oceans and brines teem with life,

albeit very different from our own—microor-ganisms that breathe arsenic. That descriptionmay describe the Earth during the early Archeaneon.

Although breathing arsenic may sound alien,microorganisms that use arsenic oxyanions foranaerobic respiration are found in environmen-tal niches ranging from garden soil and subsur-face aquifers to more extreme landscapes suchas Mono Lake and Searles Lake in California.Diverse microorganisms metabolize both inor-ganic and organic forms of arsenic, and theiractivities are part of a robust biogeochemicalcycle. What began more than 15 years ago asour quest to learn how bacteria use arsenate as a

terminal electron acceptor led us to discovernovel respiratory and photosynthetic pathways.

The Biogeochemical Cycle of Arsenic

Arsenic has both toxic and therapeutic proper-ties. For instance, the ancient Greek Hip-pocrates treated ulcers with realgar, an arsenicsulfide mineral, while the 1908 Nobel Prize wasawarded to Paul Ehrlich for his development ofsalvarsan, an organoarsenical, as a treatment forsyphilis. Currently, arsenic trioxide is used fortreating acute promyelocytic leukemia.

However, arsenic is perhaps better known asa poison in life and in fiction. Its toxicity de-pends on both its chemical form, whether inor-ganic or organic, and its oxidation state. Arsenicoccurs in four oxidation states, arsenate [As(V)],arsenite [As(III)], elemental [As(0)], and ar-

senide [As(-III)], with As(V) and As(III)more abundant in nature. These latter twoforms are readily soluble in water, withH2AsO4

- and HAsO4� typically found in

oxidized environments, whereas H3AsO30

and H2AsO3- are more typical of anoxic

environments.Because arsenate is a phosphate ana-

logue, it can enter bacterial cells via phos-phate transport systems (Pst and Pit). Onceinside, arsenate uncouples oxidative phos-phorylation from energy production andalso interferes with glycolysis by forming1-arseno-3-phosphoglycerate in place of1,3-bisphosphoglycerate. Arsenite enterscells at neutral pH via aqua-glyceroporins,which ordinarily transport glycerol mole-cules into cells, and binds to sulfhydryl

Summary

• Phylogenetically diverse microorganisms me-tabolize arsenic despite its toxicity and are partof its robust biogeochemical cycle.

• Respiratory arsenate reductase is a reversibleenzyme, functioning in some microbes as anarsenate reductase but in others as an arseniteoxidase.

• As(III) can serve as an electron donor for anoxy-genic photolithoautotrophy and chemolithoau-totrophy.

• Organoarsenicals, such as the feed additive rox-arsone, can be used as a source of energy, releas-ing inorganic arsenic.

John F. Stolz is aProfessor in theDepartment of Bio-logical SciencesDuquesne Univer-sity, Partha Basu isan Associate Pro-fessor in the De-partment of Chem-istry andBiochemistry, Du-quesne University,and Ronald S.Oremland is a Se-nior Scientist at theU.S. Geological Sur-vey, Menlo Park

Volume 5, Number 2, 2010 / Microbe Y 53

groups of cysteine residues, inactivating the pro-teins that it modifies. Meanwhile, arsine gasbinds to red blood cells, damaging membranesand causing hemolysis. High doses can causerespiratory failure and death, whereas lowchronic exposures cause cardiovascular stress,liver disease, and cancers.

Arsenic in the drinking water which resultedin widespread poisoning in Bangladesh andWest Bengal led to greater awareness of thiselement in natural systems. Arsenic-contami-nated water sources have since been uncoveredin Vietnam, Cambodia, Ghana, Mexico, Chile,Argentina, and Eastern Europe. In the UnitedStates, arsenic in drinking water is a concern forabout 3,000 municipalities, as the concentrationsurpasses the maximum allowable concentra-tion (10 ppb).

What are the sources of this arsenic? Al-though only 0.0001% in crustal abundance,arsenic is distributed widely in weathered volca-nic rocks, marine sedimentary rocks, alkalinesoils, fossil fuel deposits, iron hydroxides, and insulfidic minerals such as realgar, orpiment, andarsenopyrite. Anthropogenic sources includemine drainage and tailings, combustion of fossil

fuels, irrigation runoff from farm landstreated with arsenic-containing pesti-cides and herbicides, waste productsfrom the manufacture of glass, pig-ments, and medicinals, and the woodpreservative chromated copper arsenic.

The arsenic in Bangladeshi drinkingwater, for example, derives fromeroded Himalayan sediments. In NewEngland, geologic forces created finger-lings of arsenic-laden granitic rocksthat are part of an arsenic crescent run-ning from northern coastal Mainethrough New Hampshire and centralMassachusetts to Rhode Island.

Several processes contaminate aqui-fers with arsenic, including oxidation ofarsenic-containing pyrites; release ofAs(V) when autochthonous organicmatter, including when it is dissolved inrecharge water, reduces iron oxides; ex-change of adsorbed As(V) with fertilizerphosphates; and recharge with watercontaminated with inorganic arsenicfrom wood preservatives, pesticidessuch as calcium arsenate, and organo-arsenicals such as roxarsone and mono-

methyl arsonic acid.Microbial activities are either directly in-

volved or enhance these processes. For instance,in Mono and Searles Lakes in California, arsenicconcentrations are high enough to support arigorous, microbially driven biogeochemical cy-cle. At the heart of the cycle are oxidation-reduction reactions involving As(V) and As(III)as well as methylated and thioarsenicals (Fig. 1).Because As(V) reduction can be coupled to sul-fide oxidation and As(III) oxidation to nitratereduction, these redox reactions do not dependon oxygen.

Microbial Metabolism of Arsenic

Some microbes not only are resistant to arsenic,but actively metabolize it via methylation, demeth-ylation, oxidation, and reduction reactions, usingsome of these steps to generate energy. Methyl-ation reactions convert As(V) or As(III) into com-pounds such as monomethyl arsonate (MMA(V)),methylarsonite (MMA(III), dimethylarsinate(DMA(V)), dimethylarsinite (DMA(III)), and tri-methylarsine oxide as well as several volatile ars-ines, including monomethylarsine, dimethylars-

F I G U R E 1

The arsenic cycle. As(III) oxidation is driven by light (by phototrophic bacteria, PSBs) orthe reduction of O2, NO3

-, Mn(IV), or Fe(III). As(V) reduction is coupled to organic matter,H2S or H2 oxidation. The background is the view from the northern shore of Mono Lake.

54 Y Microbe / Volume 5, Number 2, 2010

Stolz: Interested in Arsenic from an Old Place

John F. Stolz grew up in a green,that is, environmentally sensitivehousehold before the term wascoined. His parents shunned foodadditives, cultivated an organicgarden, composted kitchen wastes,banned sodas, and promoted recy-cling. “My dad’s favorite sayingwas ‘dump oil,’ and that was in thelate sixties,” Stolz says. Those earlypractices continue to influence himin both his personal and profes-sional activities. At home, he says,“I go to a butcher that has localbeef, buy local produce and freerange chickens and eggs.” In hislab, he wrestles intellectually withthe dangers of arsenic, the com-pound he studies.

“There is a legacy of environ-mental arsenic contamination,” in-cluding from the use of inorganicarsenic compounds as pesticidesand herbicides, as wood preserva-tives, and for raising poultry, Stolzsays. Natural processes also mobi-lize arsenic, helping to account forits presence in drinking water. “Wenow understand that arsenic can bereadily mobilized from rocks andsediments by microbial activity,”he says. “These activities can resultin the contamination of drinkingwater. It has been estimated thatthe number of people affected is inthe tens of millions. These com-pounds are not inert and can bereadily transformed to free arsenic.So what started out as a curiosityhas taken on more socio/politicalimplications.”

Stolz, 54, directs the DuquesneUniversity Center for Environmen-tal Research and Education and is aprofessor of environmental micro-biology. He describes himself as ageomicrobiologist interested in mi-crobes, minerals, and metals. Hisresearch, primarily funded by theNational Aeronautics and SpaceAgency (NASA) and the NationalScience Foundation (NSF), focuseson microbial metabolism of ar-senic. “We have been investigatingmicrobial arsenic cycling and thepossibility of arsenic-based ecosys-tems on other planetary systems,”

he says. “We have been isolatingand characterizing microbes thatare capable of growing on arsenic.Although it does seem a bit far-fetched, you could base an entireecosystem on arsenic cycling.”

Stolz was raised in Syosset, N.Y.,on Long Island, and is one of fourchildren. His father, who diednearly five years ago, was a profes-sional pianist and organist. Hismother lives in Pittsburgh. “Itseems I always wanted to be a sci-entist,” he says. “I was a nerd inhigh school and did a lot of reading,mostly science fiction, Heinlein,Arthur C. Clarke, C. S. Lewis. I gota telescope in my early teens andthat got me excited about space.Growing up during the heyday ofthe space program and the moonmissions helped fan the flames.

“I was a big fan of Carl Sagan inhigh school,” Stolz continues.“There was a chapter in the bookon early life on Earth, and I likedthe idea of studying past life onEarth to gain insight into what wemight find on other planetary sys-tems.” But, he adds, “my highschool biology teacher, Mrs. JillMatulewich—who looked a lot likeKim Novak—turned me on to biol-ogy.” Stolz earned his B.S. in 1977from Fordham University, and hisPh.D. in 1984 from Boston Univer-sity. He calls his Ph.D. mentor,Lynn Margulis, his greatest influ-ence. “She taught me the value ofcollaboration and free thinking,”he says. “As students, we read Gen-esis of a Scientific Fact by LudwigFleck, and watched endless hoursof movies of termite hindgut mi-crobes. I still marvel at the oppor-tunities I had as a graduate student,doing field work in Baja California,Mexico, attending internationalmeetings, participating in NASAand MBL [Marine Biological Lab-oratory in Woods Hole, Mass.]summer courses, and meeting sci-entists.

“My first plane flight was a tripto the [San Francisco, California]Bay area to attend the second post-Viking symposium at NASA,

Ames,” he says. “Lynn lent me herbicycle, and I commuted to NASAfrom the Stanford Universitydorms. For a kid who grew upreading about space exploration, itwas really exciting to actually be atNASA.”

Stolz spent 1984–86 in his firstpostdoctoral fellowship at theNASA Jet Propulsion Lab in Pasa-dena, Calif., where he worked withJoe Kirschvink in the department ofgeology and planetary sciences.“Joe’s lab was populated by a re-markable group of students work-ing on bio- and paleo-magnetics,”he says. “I worked on magnetotac-tic bacteria and magnetofossils.”From 1987 to 1990, Stolz was apostdoctoral fellow in plant biol-ogy at the University of Massachu-setts at Amherst, in the lab of ClintFuller, where he studied photosyn-thesis in green sulfur bacteria. “Ilearned about purifying proteinsand cloning genes, and to always beopen to new research opportuni-ties,” he says.

His wife, Donna, is an associateprofessor at the University of Pitts-burgh cell biology department andassociate director of its center forbiological imaging. She studies liverregeneration. Their daughter is afreshman at the University of Day-ton, majoring in environmental bi-ology. Their son, a sophomore atthe Pittsburgh High School for Cre-ative and Performing Arts, is study-ing tuba, piano, and cello.

In his spare time, Stolz enjoyscooking, playing darts, kayaking,and shooting hoops with his son.The two—father and son—also aretaking piano lessons. “My dadtaught me as a kid, but I hadn’tplayed in decades,” he says. “Be-fore he passed away, he started giv-ing my son and me lessons. After hedied, it took a while, but we found areally wonderful piano teacher. It’sbeen a great challenge, but I reallyenjoy tickling the ivories on theSteinway upright.”

Marlene Cimons

Marlene Cimons lives and writes inMaryland.

Volume 5, Number 2, 2010 / Microbe Y 55

ine, and trimethylarsine. Although many of theenzymes involved remain unknown, a methyltransferase, ArsM, from Rhodobacter sphaeroidesconfers resistance to arsenic and can generate tri-methylarsine.

As(V) can be reduced through two distinctmechanisms. The ArsC system, that confers re-sistance, reduces As(V) but does not generateenergy. Although the resistance genes were orig-inally discovered on plasmids, they have sincebeen found on the chromosomes of a diversegroup of organisms, including Archaea, Bacte-ria, yeasts, and protoctists. Three systems, ex-

emplified by Escherichia coli, Staphylo-coccus, and yeast, evolved throughconvergent evolution. Each system hasthree components: the arsenate reduc-tase ArsC (or ACR2), a low-molecular-mass protein (13–15 kDa) related totyrosine phosphate phosphatases; ArsB(or ACR3), the arsenite-specific effluxpump; and a source of reducing equiv-alents provided by either reduced thi-oredoxin or glutaredoxin. Additionalcomponents include an ATPase (ArsA),which forms an ArsAB complex, andregulatory elements (ArsR, ArsD).

The second mechanism, involvingdissimilatory reduction of arsenate, ispart of a respiratory pathway. Discov-ered in Sulfurospirillum arsenophilummore than 15 years ago, it was subse-quently identified in several Crenarch-eota and more than 30 species of Bacte-ria. In each case, As(V) serves as aterminal electron acceptor (Fig. 2). Typ-ically these organisms are metabolicallyversatile, using organic (e.g., acetate,lactate) and inorganic (e.g., H2, H2S)electron donors as well as other elec-tron acceptors (e.g., oxygen, nitrate,Fe(III), sulfate, and thiosulfate).

The respiratory arsenate reductase(Arr) is a heterodimer with a catalyticsubunit, ArrA, and a smaller electrontransfer protein, ArrB. ArrA containsthe molybdopterin center and a [3Fe-4S] cluster while the small subunit con-tains three, possibly four, [4Fe-4S] clus-ters. The core enzyme, ArrAB, is highlyconserved. However, the arr operondiffers from organism to organism innumber of genes. For instance, Sh-

ewanella sp. strain ANA has only the genes forthe core enzyme (arrAB), while Desulfitobacte-rium hafniense has a gene encoding a putativemembrane-anchoring peptide as well as a mul-ticomponent regulatory system (arrSKRCAB).

The microbial oxidation of arsenite was firstdemonstrated in a bacillus in 1918. Establishedas a mechanism for detoxification, it has onlyrecently been linked to energy generation. Alsophylogenetically widespread, arsenite oxidationoccurs in more than 30 strains representing atleast 9 genera, including members of the Cren-archeaota, Aquificales, and Thermus, as well as

F I G U R E 2

Phylogenetic diversity of representative arsenic metabolizing Archaea and Bacteria.Dissimilatory arsenate respirers are indicated by red circles. Arsenite oxidizers areindicated by blue squares. The presence of both a square and a circle indicates anorganisms found able to both respire As(V) and oxidize As(III). Chemoautotrophic speciesare denoted by an asterix.

56 Y Microbe / Volume 5, Number 2, 2010

�-, �-, and �- proteobacteria (Fig. 2). Inmost cases, the organisms are aerobicheterotrophs or chemolithautotrophs,using oxygen as the electron acceptorfor As(III) oxidation.

More recently, species of anaerobicchemolithoautotrophic bacteria that cou-ple As(III) oxidation to NO3� reductionhave been found. Arsenite oxidation iscarried out by arsenite oxidase (Aox), aheterodimer with a catalytic subunit,AoxB, and a smaller beta subunit, AoxA(Fig. 3). AoxB contains the molybdopt-erin center and a [3Fe-4S] cluster, whileAoxA contains a Rieske-type [2Fe-2S]cluster containing subunit. Although Aoxand Arr are both mononuclear molyb-doenzymes, they are distinct enzymes. Aswith the arr operons, there is heterogene-ity in the aox operon. While the orderaoxAB is conserved, other elements aregenerally not. Further, homologs of ar-senite oxidase are found in the genomesof the Crenarcheota Aeropyrum pernixand Sulfolobus tokodaii, as well as greenanoxyphototrophic bacteria (Chlo-roflexus aurantiacus, Chlorobium limi-cola, Chlorobium phaeobacteroides), Ni-trobacter hamburgensis, Rhodoferaxferrireducens, and Burkholderia species(Fig. 4).

Respiratory Arsenate Reductase

Is a Reversible Enzyme

The efforts of one of us (RSO) to dissect thebiogeochemical cycle of arsenic in Mono Lakeresulted in the isolation and characterizationof a number of unique haloalkaliphiles, in-cluding As(V)-reducing heterotrophs (Bacillusselenitireducens and B. arseniciselenatis),As(V)-reducing chemolithoautotrophs (strainMLMS-1), photoautotrophic As(III) oxidizers(strains PHS-1 and MLP-2), and chemolitho-trophic As(III) oxidizers (Alkalilimnicola ehr-lichii).

One microbe in particular, A. ehrlichiistrain MLHE-1T, can express two completelydifferent physiologies. As an aerobe, it growsheterotrophically, with acetate as the electrondonor and carbon source. As an anaerobe, it isa chemolithoautotroph, coupling the oxida-

tion of As(III) to the reduction of nitrate tonitrite. Analysis of its genome revealed severalstriking features, including operons for nitricoxide reductase (norDQBC), nitrous oxidereductase (nosLYDZR), and nitrate reductase(narLXK2GHJI). However, the genes for re-spiratory nitrite reductase (nirK, nirS) are ab-sent, explaining the organism’s inability todenitrify.

More surprising was the failure to find aoxgenes. Using a combination of proteomics andactivity assays, we identified an enzyme witharsenate oxidase activity that is a homolog ofArr (Fig. 4). The enzyme with arsenate reductaseactivity couples the oxidation of As(III) to thereduction of DCIP, benzyl viologen, and methylviologen. It functions as arsenite oxidase in A.ehrlichii, as was confirmed through gene disrup-tion experiments by Kamrun Zargar and ChadW. Saltikov of the University of California,Santa Cruz.

The active Arr is a heterodimer, with a 91-kDa catalytic subunit containing the motif for a

F I G U R E 3

Structure of arsenite oxidase from Alcaligenes faecalis (left) and the respiratory arsenatereductase from Alkalilimnicola ehrlichii (right) that functions as an arsenite oxidase. Thelarge, catalytic subunits containing the molybdenum co-factor (AroB and ArrA) are shownin blue. The smaller Rieske subunit of AoxA and the iron-sulfur cluster subunit of ArrB areshown in red. The iron-sulfur clusters are shown in yellow (S) and gold (Fe). Courtesy ofCourtney Sparacino-Watkins.

Volume 5, Number 2, 2010 / Microbe Y 57

[4Fe-4S] cluster (C-X2-C-X3-C-X27-C) ratherthan the [3Fe-4S] cluster of AoxB (Fig. 3). Thebeta subunit, ArrB, is 28 kDa and is predicted tohave four [4Fe-4S] clusters (Fig. 3). The arroperon from A. ehrlichii has additional genesthat appear to encode a 45-kDa [4Fe-4S] cluster-containing subunit (ArrB2), a 44-kDa mem-brane-anchoring subunit (ArrC), and a 33-kDachaperone TorD (ArrD). Homologs of thesegenes are found in arr operons from other arsen-ate-respiring bacteria but not aox operons.

The arsenate reductase of A. ehrlichii is an en-zyme that also runs in reverse, reducing As(V)when coupled to oxidation of benzyl viologen ormethyl viologen. With these results in mind, wedetermined that Arr from two arsenate-respiringbacteria, Alkaliphilus oremlandii and Shewanellasp. strain ANA-3, could also oxidize arsenite.These results reveal that Arr is an oxido/reductasecapable of functioning as an arsenate reductase orarsenite oxidase. Physiologically, however, the en-zyme works only as an oxidase or reductase. Whatdetermines how it functions could depend on elec-tron potentials of metal centers, additional sub-

units, and how the electron transfer chainis organized in a particular organism.

Arsenite Oxidation Linked

to Photosynthesis

Finding arsenite oxidation in periphytoncommunities and identifying arsenite ox-idase homologs in the genomes of Chlo-roflexus aurantiacus and two Chloro-bium species (C. limnicola and C.phaeobacteroides) raises the possibilitythat As(III) is involved in photosynthesis.In the summer of 2007, members of theU.S. Geological Survey (USGS) team ledby one of us (RSO) visited hot springs onPaoha Island in Mono Lake, where pig-mented microbial biofilm communitiesthrive in arsenic-rich waters. Anoxygenicphototrophic bacteria dominate the redbiofilms, while the green biofilms arepopulated with an Oscillatoria-like cya-nobacterium.

Both biofilm types oxidize As(III) un-der anaerobic conditions in the light,suggesting that a phototroph growingunder anoxygenic conditions was usingAs(III) as an electron donor in photo-lithoautotrophy. To investigate this

possibility, members of the USGS research teamisolated strain PHS-1, a photosynthetic bacte-rium closely related to Ectothiorhodospira sha-poshnikovii. They found that it indeed grewanaerobically in the light with As(III) and as inA. ehrlichii, the Arr functioned as the arseniteoxidase.

As(III)-linked anoxygenic photosynthesis andthe reversibility of Arr provide insights into theorigins of microbial arsenic metabolism. As(III)was probably the predominant form of arsenicin the early Archaean period. Dissimilatory ar-senate reduction could arise only after sufficientAs(V) became available, presumably derivedfrom aerobic arsenite oxidation. The use ofAs(III) as an electron donor in anoxygenic pho-totrophy, however, provides a mechanism forgenerating As(V) in the absence of atmosphericoxygen. The widespread occurrence of Arr andAox in the bacterial and archaeal domains alsosuggests that both appeared before those twodomains diverged. The ability of Arr to act as anoxido/reductase, however, implies a more an-cient origin for this enzyme. Regardless of which

F I G U R E 4

Phylogenetic tree (Neighbor Joining) showing relatedness of the catalytic subunit ofrespiratory arsenate reductase (ArrA) and arsenite oxidase (AoxB) with members of theDMSO reductase family of molybdoproteins. The Arr homolog from A. erhlichii forms aseparate branch with Arr homologs from PHS-1, H. halophilus, and M. magnetotacticum.BisC, biotin sulfoxide reductase (Escherichia coli); DorA, DMSO reductase (E. coli); FdhG,formate dehydrogenase (E. coli); NapA, periplasmic nitrate reductase (E. coli); NarG,respiratory nitrate reductase (E. coli); NasA, assimilatory nitrate reductase (Klebsiellapneumonae); PsrA, polysulfide reductase (Salmonella typhimurium); SerA, respiratoryselenate reductase (Thauera selenatis); TorA, trimethylamine oxide reductase (E. coli);TtrA, tetrathionate reductase (Wolinella succinogenes). The putative arsenate reductasefrom Pyrobaculum aerophilum is also shown.

58 Y Microbe / Volume 5, Number 2, 2010

protein came first, arsenic was an importantfactor during evolution.

Microbial Transformation

of Organoarsenicals

Although inorganic arsenic pesticides and her-bicides were banned during the 1970s, or-ganoarsenicals still are being widely used. Forinstance, about 2,500 tons of monosodiummethanearsenate (MSMA) and dimethyl ars-inic acid (DMA) are used annually for weedcontrol on cotton fields, citrus groves, and golfcourses.

Similarly, roxarsone, which is 3-nitro-4-hy-droxybenzene arsonic acid, was used widely as afeed additive by the poultry industry until re-cently. It prevents coccidiosis, and also acceler-ates chicken weight gain and improves pigmen-tation. Although little of the roxarsoneaccumulates in birds, the compound ends up inlitter, in which it degrades rapidly. Specifically,

clostridial species, including our lab strain Al-caliphilus oremlandii, reduce the nitro group,producing 3-amino-4-hydroxybenzene arsonicacid as an end product. We (JFS and PB) findthat when A. oremlandii is grown with lactateand roxarsone, cell yields are 10-fold greaterthan if cells are grown on lactate alone. Cells ofA. oremlandii can also respire arsenate and thio-sulfate, suggesting that this species also gener-ates ATP through oxidative phosphorylationlinked to roxarsone reduction.

Whether the source is natural or anthropo-genic, it is remarkable how microorganismsadapt to grow on compounds containing ar-senic. Their metabolic activities affect its chem-ical state, mobility, and thus its toxicity—insome cases, poisoning water supplies. Thus, theongoing biogeochemical cycling of arsenic, pre-sumably more prominent during the early an-oxic phase of life on Earth, continues to have animportant impact on biology in general andhuman health in particular.

ACKNOWLEDGMENTS

We thank the current and former members of our respective labs, as well as our colleagues including D. K. Newman, C. W.Saltikov, J. M. Santini, B. D. Lanoil, M. Schreiber, A. Barchowsky, M. Madigan, and J. T. Hollibaugh. This work has beensupported in part by grants from the National Aeronautics and Space Administration, the National Science Foundation, andthe U.S. Geological Survey.

SUGGESTED READING

Hoeft, S. E., J. Switzer Blum, J. F. Stolz, F. R. Tabita, B. Witte, G. M. King, J. M. Santini, and R. S. Oremland. 2007.Alkalilimnicola ehrlichii, sp. nov. a novel aresnite-oxidizing halophylic gamma proteobacterium capable of chemoautotrophicor heterotrophic growth with nitrate or oxygen as the electron acceptor. Int. J. Sys. Evol. Microbiol. 57:504–512.Kulp, T. R., S. E. Hoeft, M. Madigan, J. T. Hollibaugh, J. Fischer, J. F. Stolz, C. W. Culbertson, L. G. Miller, and R. S.Oremland. 2008. Arsenic(III) fuels anoxygenic photosynthesis in hot spring biofilms from Mono Lake, California. Science321:967–970.Oremland, R. S., and J. F. Stolz. 2003. The ecology of arsenic. Science 300:939–944.Oremland, R. S., T. R. Kulp, J. Switzer Blum, S. E. Hoeft, S. Baesman, L. G. Miller, and J. F. Stolz. 2005. A microbial arseniccycle in a salt-saturated, extreme environment: Searles Lake, California. Science. 308:1305–1308Richey, C., P. Chovanec, S. E. Hoeft, R. S. Oremland, P. Basu, and J. F. Stolz. 2009. Respiratory arsenate reductase as abidirectional enzyme. 382:298–302.Rosen, B. 2002. Biochemistry of arsenic detoxification. FEBS Letters 529:86–92.Saltikov, C. W., and D. K. Newman. 2003. Genetic identification of a respiratory arsenate reductase. Proc. Natl. Acad. Sci.USA 100:10983–10988.Silver, S., and L. T. Phung. 2005. Genes and enzymes involved in bacterial oxidation and reduction of inorganic arsenic. Appl.Environ. Microbiol. 71:599–608.Stolz, J. F., E. Perera, B. Kilonzo, B. Kail, B. Crable, E. Fisher, M. Ranganathan, L. Wormer, and P. Basu. 2007.Biotransformation of 3-nitro-4-hydroxybenzene arsonic acid and release of inorganic arsenic by Clostridium species. Environ.Sci. Tech. 41:818–823.

Volume 5, Number 2, 2010 / Microbe Y 59