Molecular insight into extreme copper resistance in the extremophilic archaeon 'Ferroplasma acidarmanus' Fer1

Molecular insight into extreme copper resistance inthe extremophilic archaeon ‘Ferroplasmaacidarmanus’ Fer1

Craig Baker-Austin,1 Mark Dopson,13 Margaret Wexler,1 R. Gary Sawers3

and Philip L. Bond1,2

Correspondence

Philip L. Bond

[email protected]

1,2School of Biological Sciences1 and Centre for Ecology, Evolution and Conservation2, Universityof East Anglia, Norwich NR4 7TJ, UK

3Department of Molecular Microbiology, John Innes Centre, Norwich NR4 7UH, UK

Received 29 March 2005

Revised 18 May 2005

Accepted 30 May 2005

‘Ferroplasma acidarmanus’ strain Fer1 is an extremely acidophilic archaeon involved in the

genesis of acid mine drainage, and was isolated from copper-contaminated mine solutions at

Iron Mountain, CA, USA. Here, the initial proteomic and molecular investigation of Cu2+ resistance

in this archaeon is presented. Analysis of Cu2+ toxicity via batch growth experiments and

inhibition of oxygen uptake in the presence of ferrous iron demonstrated that Fer1 can grow and

respire in the presence of 20 g Cu2+ l”1. The Fer1 copper resistance (cop) loci [originally detected

by Ettema, T. J. G., Huynen, M. A., de Vos, W. M. & van der Oost, J. Trends Biochem Sci 28,

170–173 (2003)] include genes encoding a putative transcriptional regulator (copY), a putative

metal-binding chaperone (copZ) and a putative copper-transporting P-type ATPase (copB).

Transcription analyses demonstrated that copZ and copB are co-transcribed, and transcript levels

were increased significantly in response to exposure to high levels of Cu2+, suggesting that

the transport system is operating for copper efflux. Proteomic analysis of Fer1 cells exposed to

Cu2+ revealed the induction of stress proteins associated with protein folding and DNA repair

(including RadA, thermosome and DnaK homologues), suggesting that ‘Ferroplasma acidarmanus’

Fer1 uses multiple mechanisms for resistance to high levels of copper.

INTRODUCTION

Copper is an important co-factor for many enzymesinvolved in diverse cellular processes, such as radicaldetoxification, oxidative phosphorylation and iron meta-bolism (Tsivkovskii et al., 2003). The two oxidation statesof copper (Cu+ and Cu2+) and its relatively broad redoxpotential when bound to protein (200–800 mV) make itimportant to metalloenzymes in many redox-driven reac-tions (Solioz & Stoyanov, 2003). While trace amounts ofcopper are essential for life, copper also catalyses thesynthesis of reactive oxygen species, leading to severedamage of cytoplasmic constituents through the oxidationof proteins, cleavage of DNA and RNA, and lipidperoxidation (Garcia et al., 2002; Halliwell & Gutteridge,1984). Copper also binds with high affinity to histidine,cysteine and methionine, resulting in the inactivation ofproteins (Camakaris et al., 1999). Consequently, tightly

controlled homeostatic systems need to permit delivery ofthe metal to specific target enzymes, while maintaininglow intracellular concentrations of free ionic Cu+. It is likelythat most, if not all, organisms have mechanisms for copperhomeostasis (Rensing et al., 2000), and this has been studiedin certain bacteria that include Enterococcus hirae (Solioz &Stoyanov, 2003), Escherichia coli (Rensing & Grass, 2003)and cyanobacteria (Cavet et al., 2003).

Micro-organisms use various means to control intracellularcopper levels; this includes various influx and efflux mecha-nisms, and copper complexation by cellular components.The cop operon of Ent. hirae encodes a metal-dependentrepressor, an internal Cu+ chaperone, and two membrane-bound P-type ATPases for Cu+ uptake and efflux (Solioz &Stoyanov, 2003). E. coli employs additional mechanisms thatinclude a CBA-type Cu+ efflux Cus system that spans acrossthe inner and outer cell membranes, as well as the plasmid-encoded pco system, by which PcoA is thought to carry Cu+

across the inner membrane, after which it is oxidized in theperiplasm (Rensing & Grass, 2003).

‘Ferroplasma acidarmanus’ Fer1 is a mesophilic, extremeacidophile that was isolated from a disused ironmine in Iron

3Present address: Department of Molecular Biology, Umea University,S-901 87 Umea, Sweden.

Abbreviations: 2D-PAGE, two-dimensional polyacrylamide gel electro-phoresis; MALDI-TOF, matrix-assisted laser desorption ionization time-of-flight.

0002-8076 G 2005 SGM Printed in Great Britain 2637

Microbiology (2005), 151, 2637–2646 DOI 10.1099/mic.0.28076-0

Mountain, CA, USA (Edwards et al., 2000). Environmentscontaining high levels of dissolved metals include bothactive and disusedmines, where the production of acid minedrainage and acid rock drainage is catalysed by the action ofmicro-organisms (Nordstrom & Southam, 1997). Solutionsthat flow from the Iron Mountain mine include some ofthe most acidic and metal-rich mine drainage studied todate (Nordstrom & Alpers, 1999). The concentrations ofiron, zinc, copper and cadmium in Iron Mountain solu-tions colonized by Fer1 typically exceed 28, 2?5, 0?38 and0?25 g l21, respectively (Edwards et al., 1998; Nordstrom &Alpers, 1999). Acidophiles such as archaeal Ferroplasmaspecies can dominate biofilms that are present in extrememetal contamination (Bond et al., 2000a), and in theseenvironments metal resistance would be of primary impor-tance. The levels of copper at which metabolic activity hasbeen recorded in various micro-organisms (Table 1) indicatethe high metal-resistance capabilities of acidophiles. At thepresent time, copper homeostasis is being studied extensivelyin various neutrophilic bacterial species (see references above).Most archaeal genomes sequenced to date contain sequenceswith high similarity to copper transport P-type ATPases.However, investigation of the biochemical and genetic metalhomeostasis and resistance mechanisms of acidophilic micro-organisms, particularly of the archaea, is virtually in itsinfancy. These archaea represent challenging yet environmen-tally relevant systems for an increased understanding of metalresistance and homeostasis (Dopson et al., 2003).

The genome of ‘Ferroplasma acidarmanus’ Fer1 (hereafternamed Fer1) was recently sequenced to 97% coverage(http://genome.ornl.gov/microbial/faci/), and anotherFerroplasma strain (uncultured), termed Type II strain,detected from an Iron Mountain biofilm has a nearcomplete genome sequence obtained via random shotgunsequencing (Tyson et al., 2004). Both sets of sequence datacontain a putative copper transport gene cluster (cop),which, based on gene homology, encodes a transcriptionalregulator, a copper chaperone and a copper-translocatingP-type ATPase (Ettema et al., 2003). Presently, the cop locimake up the only copper-specific homeostatic systemevident on the Ferroplasma genomes, and the putativecopper transporter (copB) has high similarity to both copperuptake and efflux P-type ATPases.

In this study, we present initial biochemical, proteomic andmolecular characterization of copper resistance in Fer1. Ourresults show that Fer1 is the most copper-resistant archaeonstudied to date, and that transcription of its cop gene clusteris responsive to Cu2+, suggesting an intrinsic role in copperefflux. These findings constitute the first investigative reportof an archaeal copper resistance system.

METHODS

Strain and growth conditions. ‘F. acidarmanus’ Fer1 was grownchemomixotrophically in a continuous-culture vessel at 37 uC usingmineral salts medium (MSM, acidified to pH 1?2 with H2SO4)supplemented with trace elements, ferrous iron and yeast extract.MSM comprised (in g l21): (NH4)2SO4 (3?0), Na2SO4.10H2O (3?2),KCl (0?1), K2HPO4.3H2O (0?06), MgSO4.7H2O (0?5) andCa(NO3)2.4H2O (0?014). MSM was supplemented with the follow-ing trace metals (in mg l21): FeCl3.6H2O (11?0), CuSO4.5H2O(0?5), HBO3 (2?0), MnSO4.4H2O (2?5), Na2MoO4.2H2O (0?8),CoCl2.6H2O (0?6), ZnSO4.7H2O (0?9) and Na2SeO4 (0?1). As energyand carbon sources, ferrous iron (FeSO4.7H2O) and yeast extractwere added at 2?0% and 0?04% (w/v), respectively. The continuousculture vessel operating at dilution rate D=0?01 h21 provided asteady-state source of cells at a mean protein level of 6?9±1?5 mg l21 for all batch experiments. For examination of coppertoxicity, proteomic response and RNA expression, cells were grownchemomixotrophically in MSM under batch-culture conditions inthe presence and absence of various copper or metal additions.Unless otherwise stated, each 100 ml batch culture was inoculatedwith Fer1 cells equivalent to 10 mg of protein (~109 cells) addedfrom the steady-state continuous-culture vessel, and were incubatedwith shaking (200 r.p.m.) in flasks at 37 uC.

Growth of Fer1 in the presence of copper. Growth was mea-sured after 72 h incubation by detecting the increase in protein(Bradford assay, Bio-Rad) compared to initial protein levels (t=0)in 100 ml batch cultures containing 0, 1, 5, 10 or 20 g Cu2+ l21

(added as CuSO4.5H2O, in addition to the trace element Cu2+ con-centration of 0?13 mg l21). All batch-culture toxicity experimentswere conducted in triplicate, and the mean and standard deviation(SD) were calculated. Fer1 cells used as inoculum, equivalent to10 mg of protein, were previously batch cultured in the presence orabsence of 1 g Cu2+ l21.

Fe2+-dependent oxygen consumption. Fer1 cells were harvestedfrom the continuous culture vessel by centrifugation (10 000 gfor 20 min) and washed three times in MSM. Resting cells equiva-lent to 100 mg of protein were resuspended in MSM at 37 uC and

Table 1. Upper-level copper concentrations at whichgrowth has been recorded for acidophilic and neutrophilicprokaryotes

Values of metal toxicity are often determined differently.

Therefore, the upper-level concentration is defined as the highest

concentration at which growth has been detected.

Organism Upper-level

concentration

(mM)

Reference

Acidophiles

‘Ferroplasma acidarmanus’ 312 This study

Acidithiobacillus ferrooxidans 800 Harvey & Crundwell

(1996)

Acidocella aminolytica 20 Ghosh et al. (1997)

‘Acidiphilium symbioticum’

KM2

20 Mahapatra & Banerjee

(1996)

Acidocella strain GS19h 15 Ghosh et al. (1997)

Sulfolobus acidocaldarius 1 Miller et al. (1992)

Neutrophiles

Enterococcus hirae 8 Solioz & Stoyanov

(2003)

Pseudomonas aeruginosa 6 Teitzel & Parsek

(2003)

Escherichia coli 1 Nies (1999)

Saccharomyces cerevisiae 0?2 Soares et al. (2003)

2638 Microbiology 151

C. Baker-Austin and others

equilibrated for 5 min in the presence of various levels of addedCu2+ (0, 1?0, 10, 20, 40 or 80 g l21). Following the addition ofFeSO4.7H2O at 2% (w/v), Fe2+-dependent O2 consumption wasassayed (1 ml final volume) using a Clark-type oxygen electrode at40 uC (Hallberg et al., 1996).

Proteomic analysis of cytoplasmic protein. For protein expres-sion analyses, replicate one-litre Fer1 cultures were grown to lateexponential phase [between 72 and 96 h incubation to a density of6?1±0?5 (mg protein) l21] in MSM with iron and yeast extract (asdescribed above) in the presence and absence of added Cu2+ at1 g l21. Cells were harvested by centrifugation (10 000 g for 20 min)at 4 uC and immediately frozen at 220 uC. When required, frozencells were thawed on ice in 1 ml MSM, centrifuged, resuspended inlysis buffer, and disrupted by sonication (563 s bursts at amplitude7?5 mm). Cell debris was removed by centrifugation (10 000 g for10 min) and the protein extract was stored in frozen aliquots. Two-dimensional polyacrylamide gel electrophoresis (2D-PAGE) wascarried out as previously described (Dopson et al., 2004a).

Gels were stained with EZBlue (Sigma) or silver nitrate (Blum et al.,1987) and scanned using the proEXPRESS proteomic imaging system(PerkinElmer). Images of silver-stained gels from the control andCu2+-exposed cells were analysed using ProteomWeaver version 1.2(Definiens). Protein spot matching was initially carried out using theautomated matching function and then manually improved by spot‘land marking’. Composite gel images were produced from thereplicates for each condition (n=2–4). Protein spot expression wasnormalized according to the entire protein complement expressed onthe gels (ProteomWeaver), alleviating the need for internal proteinstandards. After detection of differentially expressed proteins in silver-stained polyacrylamide gels, proteins that could be identified fromcorresponding EZBlue stained gels were excised using a sterilemicropipette tip. Samples were treated by trypsin digestion andanalysed using matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectrometry (Bruker Reflex III; Heskethet al., 2002). MALDI-TOF peptide mass fingerprint data were matchedagainst the ‘F. acidarmanus’ Fer1 genome sequence data (http://genome.ornl.gov/microbial/faci/) using MASCOT (http://www.matrixscience.com) and results were only included when peptidefragment matching was statistically significant (MOWSE score greaterthan 74).

In vitro DNA manipulations. Chromosomal DNA was extractedas previously described (Bond et al., 2000b) from 400 ml of Fer1cells originating from the continuous-culture vessel. Oligonucleo-tides were synthesized by MWG Biotech (Table 2). The promoter

regions of copY and copZB (dashed boxes, Fig. 1) were amplifiedusing primer sets copYPfor and copYPrev and copYZBfor andcopYZBrev, respectively, by PCR. The resulting PCR productswere cloned into pCR4-TOPO using a TOPO TA cloning kit(Invitrogen) according to the manufacturer’s instructions. Plasmidscontaining inserts were transformed into E. coli TOP10. Plasmidscontaining putative promoter regions were verified by automatedDNA sequencing.

RT-PCR and primer extension. For RT-PCR, batch chemomixo-trophic cells were grown to late exponential phase (as describedabove) separately in the presence or absence of 1 g l21 Cu2+, Zn2+,Hg2+, Cd2+, Mg2+, As3+, As5+ or 0?64 g l21 Cu+. RNA wasextracted from the Fer1 cells using the RNAwiz (Ambion) reagentaccording to the manufacturer’s instructions. RT-PCR amplificationswere carried out using the Access RT-PCR System (Promega) on 1 mgFer1 RNA and primers annealing to regions 1, 2, 3 and 4 (Fig. 1,Table 2) separately. For primer extension reactions, RNA wasextracted from Fer1 late-log batch cells cultured in the presence and

Table 2. Primer sequences for PCR and primer extensionwith corresponding annealing regions

Target regions correspond to those indicated in Fig. 1.

Primer name Sequence (5§–3§) Target

region

Cloning primers

copYZBfor ATTACCATTGGCCATCTTATA b

copYZBrev ATCAGGCATACTCAAACAGC b

copYPfor GCTTAGTTATAATTTTTG a

copYPrev GCGCTGTTTGAGTATGCCTGAT a

RT-PCR primers

RNAcopZfor ATGTGGAATGAAAGGAAAA 3

RNAcopBfor ATGGCAGTAGACCCTGTTTG 4

RNAcopBrev CATGGCAATTTTCGCTCCA 1, 3 and 4

RNAcopYfor ATGCATAGCAATATGAATAG 1 and 2

RNAcopYrev ACCCCTTATTAATAATACAA 2

Primer extension

primers

copYpe GAAGCAGCCTGTCAGATATT a

copBpe CTGGTTCCATCACGATCAGAG b

Fig. 1. Arrangement of the ORFs for theFer1 cop loci. The suggested genes copY,copZ, and copB code for a putative tran-scriptional regulator, a putative chaperoneand a putative copper-translocating ATPase,respectively. Regions amplified by RT-PCRare shown as horizontal lines 1, 2, 3 and 4.Dashed boxes a and b are regions amplifiedby PCR for cloning. Black and dashed verti-cal lines are putative copper-binding(TRASH) and DNA-binding motifs, respec-tively. Possible coding regions ORFX andORFY flank the cop locus and have similarityto a Sec-independent secretion protein anda phospholipid-binding protein, respectively.

http://mic.sgmjournals.org 2639

Analysis of Ferroplasma copper resistance

absence of 5 g Cu2+ l21, as described above. Primer extensions wereperformed essentially as described by Sawers & Bock (1989) using pri-mers copYpe and copBpe (Table 2), which are complementary to copYand copB mRNA approximately 70 bases downstream from theirrespective 59 ends. Plasmids containing the copY and the copZB pro-moter regions (Fig. 1) were used as template DNA for the corre-sponding manual DNA sequencing reactions. Comparative levels ofRNA expression from control and Cu2+-exposed cells on the sequen-cing gel were determined using a Fujix BAS1000 Phosphorimager.

Computational analysis of DNA and protein sequences. Thepartial genome sequences for Fer1 and Ferroplasma Type II strainsare available from the Oak Ridge National Laboratory (http://genome.ornl.gov/microbial/faci/) and the National Center forBiotechnology Information (http://www.ncbi.nih.gov), respectively.Selected sequences were analysed for protein folding regions,regulator elements and transmembrane spanning regions, using 3DPSSM (http://www.sbg.bio.ic.ac.uk/~3dpssm/; Kelley et al., 2000),NPS@ (http://npsa-pbil.ibcp.fr/cgibin/npsa_automat.pl?page=/NPSA/npsa_hth.html; Dodd & Egan, 1990), Palindrome (EMBOSS, http://emboss.sourceforge.net/) and HMMTOP (www.enzim.hu/hmmtop/index.html; Tusnady & Simon, 2001).

RESULTS

Growth of Fer1 in the presence of Cu2+

Fer1 cells grew at all levels of Cu2+ used (Fig. 2), and growthwas highest without added Cu2+. However, even in thepresence of 20 g Cu2+ l21, the final protein concentrationwas 30% that of cells without added Cu2+. Cells grown with20 g Cu2+ l21 were subsequently harvested and reinocu-lated into fresh medium (without Cu2+), and high levels ofgrowth were observed after 92 h (a mean increase of 6?0 mgof protein l21), confirming the viability of this culture. Fer1cells were harvested from 1 g Cu2+ l2l and subsequentlyrecultured using the same toxicity concentration range(0–20 g Cu2+ l21). Following 72 h incubation, the proteinlevels of the copper-preincubated cells were slightly higherat all data points than the non-preincubated cells; how-ever, the differences were not statistically valid (Fig. 2).This suggests that pre-culturing Fer1 cells in the presence ofcopper did not result in a significant increase in observedCu2+ resistance.

Vanadate was added (5 mM) to batch cultures of Fer1.Vanadate is a P-type ATPase inhibitor and may be expectedto inhibit growth in the presence of copper. However, nosignificant increase in the sensitivity of Fer1 to copper wasobserved (data not shown). Possibly this archaeal P-typeATPase is resistant to vanadate, or vanadate simply did notenter the cells at the concentrations used.

Toxicity of Cu2+ towards Fe2+-dependent oxygenconsumption

Previously it was shown that Fer1 grows chemomixotro-phically, oxidizing Fe2+ in the presence of organic carbon(yeast extract), and the chemomixotrophic growth rates arefour times those of chemo-organotrophic growth (oxidizingorganic carbon) (Dopson et al., 2004b). Experiments werecarried out to determine the effect of Cu2+ on Fe2+

oxidation by Fer1 cells. The abiotic Fe2+ oxidation rate was1?5 nmol O2 min21, compared to 364±34 nmol O2 min21

(mg Fer1 protein)21 (n=3) in the absence of Cu2+,increasing to 514±40, 456±94 and 441±68 nmol O2

min21 (mg protein)21 (n=3) in the presence of 1, 10 and20 g Cu2+ l21, respectively. No significant inhibition of theFe2+ oxidation rate was observed until the addition of 40 gCu2+ l21, and at 80 g Cu2+ l21 the rate was 266 nmol O2

min21 (mg protein)21 (73% of the rate in the absence ofCu2+).

Proteomic analysis of Fer1 cells exposed toCu2+

Fer1 was grown in batch cultures in the presence andabsence of 1 g Cu2+ l21 (corresponding to 16 mM) andprotein profiles were analysed to identify proteins upregu-lated in response to Cu2+ (Fig. 3). Higher concentrations ofcopper significantly affect Fer1 batch-culture growth(Fig. 2), and therefore would invalidate direct comparisonsbetween copper-exposed and non-exposed cell cultures.None the less, the exposure concentration used (16 mM)would be toxic to many acidophiles and to mostneutrophilic bacteria (Table 1), and is two- to threefoldgreater than the typical concentrations detected in the IronMountain mine solutions. Representative comparisons ofprotein separations are shown from cells grown in thepresence and absence of copper (Fig. 3B, C). A total of 21proteins were identified whose synthesis was induced more

Fig. 2. Growth of Fer1 cells measured in the presence ofvarious concentrations of Cu2+. Growth was determined asprotein increase compared to no-growth controls after 72 hincubation at 37 6C. Cells used as the standard inoculum(10 mg of protein) were previously cultured in the absence (&)or presence (#) of 1 g Cu2+ l”1. Data points are mean±SD

(n=3).

2640 Microbiology 151

C. Baker-Austin and others

than 1?5-fold compared to cells grown without additionalCu2+ (Table 3, Fig. 3). Proteins identified included mol-ecular chaperones, DNA repair enzymes, enzymes fromcentral metabolic and biosynthetic pathways, and proteinsassociated with transcription and translation. Certain pro-teins were detected at multiple locations on our 2D gels(Table 3, Fig. 3A), including those with homology to DnaK(spots 20, 36 and 37) and thermosome (spots 27 and 28).High levels of these proteins, detected in the copper-exposedcondition (Fig. 3B, 3C), may have led to detection ofdegradation products or post-translational modifications.The state of a phosphorylation site, such as that on the E. coliDnaK protein (McCarty & Walker, 1991), could result indiffering migration spots for the Fer1 DnaK homologue.

Of the 21 up-regulated proteins detected in the Fer1response to copper, nearly half showed homology to thoseinvolved in protein and DNA repair (Table 3). Oneprotein had high similarity to FKBP transisomerase(copper spot 32; Table 3, Fig. 3A). In E. coli, this istermed the trigger factor, which binds partially foldedprotein intermediates to enhance their folding (Fink, 1999).Proteins with homologies to DnaK (multiple spots 36, 37

and 20), prefoldin (spot 22) and the HSP60 family chape-ronin thermosome (spots 18, 27 and 28), were highlyexpressed in response to copper (Table 3, Fig. 3). Inarchaea, the protein chaperone systems are thought toutilize DnaK, prefoldin and the thermosome to bind un-folded proteins and mediate their folding (Macario et al.,1999). The DNA repair enzymes more highly expressed inresponse to copper were RecA/RadA and endonuclease IV(spots 12 and 17, respectively; Table 3). The characterizedactivities of RadA and endonuclease IV are to repair double-strand DNA breaks and DNA lesions, respectively (Krokanet al., 1997). In E. coli, endonuclease IV is upregulated inresponse to superoxide anion radicals (Chan & Weiss,1987).

Genomic evidence of Fer1 copper transport

A cluster of genes with homology to copper transportsystems was previously detected on the annotated genomesequence of Fer1 (Ettema et al., 2003). The putativeregulator (designated CopY; Fig. 1) was previously depictedas being divergently transcribed from the copper chaperone(designated CopZ) and the cation-translocating P-type

Fig. 3. (A) Representative 2D-PAGE separation of protein prepared from Fer1 cells cultured in the presence of 1 g Cu2+ l”1.Proteins excised for analyses (MALDI-TOF mass spectrometry) are shown as circled spots (spot numbers correspond tothose in Table 3). Dashed boxes from (A) are shown as enlarged areas (B, C) that include corresponding gel separationsections from cells cultured in the absence (upper region, above the dashed line) and presence (lower region) of 1 g Cu2+

l”1. Images for comparison of protein expression (B, C) are taken directly from individual gels. The apparent spot intensitydifferences observed here may not precisely match those listed in Table 3 (stimulation), as the latter were derived bycomparison of composite images (not shown) prepared from replicate 2D-PAGE gels.

http://mic.sgmjournals.org 2641

Analysis of Ferroplasma copper resistance

ATPase (CopB) (Ettema et al., 2003). However, it appearsthat copY is transcribed in the same direction as copZ andcopB (Fig. 1). As previously reported by Ettema et al. (2003),CopY and CopZ contain Cu-binding domains, known asTRASH (Fig. 1). CopY has sequence identity (29%) with acharacterized transcriptional regulator of the LrpA familyfrom Pyrococcus furiosus (Brinkman et al., 2000). This familyof regulators is reportedly involved in either positive ornegative regulation. copB encodes a putative P-typeATPase Cu transporter (Fig. 1), and this protein containspredicted sequences for metal binding, ATP binding,phosphorylation and phosphatase sites. Hydropathy profilesrevealed numerous transmembrane spanning domains inthe P-type ATPase, as expected for a membrane-boundprotein. However, primary sequence information is not

useful to determine whether the role of this putativetransporter is for Cu efflux or uptake.

The Fer1 genome (http://genome.ornl.gov/microbial/faci/)was screened (TBLASTN) for other possible copper trans-porters using protein sequences from various organisms.This included sequences of ATP7A (human), CopA (Ent.hirae, Helicobacter pylori, Bacillus subtilis and E. coli),CopB (Ent. hirae and Archaeoglobus fulgidus), Ccc2(Saccharomyces), PacS and CtaA (Synechococcus elongates).One Fer1 gene (number 1580, contig 169) had approxi-mately 25% sequence identity to these. However, this genealso matched with much higher similarity to known K+

transport systems. Genes encoding other copper-resistancemechanisms, such as the E. coliCus (transport), Pco (copper

Table 3. Proteins greater than 1?5-fold upregulated in cells grown in the presence of 1 g Cu2+ l”1 compared to cells in theabsence of added Cu2+, as calculated from composite spot intensities from replicate 2D-PAGE gel sets

Number* GeneD Putative protein

identificationd

Percentage

identityd

Stimulation§

Metabolic and electron transport proteins

33 159.694 Dihydroorotate dehydrogenase 80 NP||

26 164.1020 Pyridoxine synthase 72 4?5

30 166.1268 Transaldolase 38 3?4

7 169.1648 Glutamate dehydrogenase 76 2?1

Biosynthetic proteins

29 166.1244 Carbamoylphosphate synthase 66 6?6

9 155.459 Ornithine cyclodeaminase 56 2?7

24 166.1244 Carbomylphosphate synthase 66 2

13 152.338 Isochorismatase 64 1?6

Transcription and translation components

38 168.1427 30S ribosomal protein (S2) 77 NP

8 150.283 mRNA 39-end processing factor 76 2?6

14 166.1227 c-Glutamyltransferase 58 2?4

Heat shock, protein chaperones and

protein protection

18 165.1085 Thermosome subunit (HSP60 family) 82 NP

36 169.1708 DnaK (HSP70 family) 79 NP

37 169.1708 DnaK (HSP70 family) 79 NP

22 147.213 Prefoldin subunit 46 NP

27 157.567 Thermosome subunit (HSP60 family) 28 11?9

20 169.1708 DnaK (HSP70 family) 79 4?9

32 167.1360 FKBP transisomerase 71 4?6

28 157.567 Thermosome subunit (HSP60 family) 28 3?4

DNA repair proteins

12 160.749 RecA/RadA recombinase 55 7?5

17 158.618 Endonuclease IV 70 6?4

*The number refers to the numbered protein spots in Fig. 3.

DThe number indicates the contig and gene numbers for the identified protein, as listed on the ‘Ferroplasma acidarmanus’ Fer1 draft genome

analysis web page (http://genome.ornl.gov/microbial/faci/).

dPutative identification and Percentage identity are based on BLAST2 similarity searches against the UniProt database (Apweiler et al., 2004).

§Stimulation is the increase in protein expression determined by comparison of gel sets using ProteomWeaver (see Methods).

||NP, Spot not present in the corresponding gel set.

2642 Microbiology 151

C. Baker-Austin and others

oxidation) and Cut (metal sequestration) systems were notdetected on the Fer1 genome.

RT-PCR and primer extension analysis of thecop genes

To investigate regulation of the cop genes in response toCu2+ and other metals, RT-PCR was performed for ampli-fication of regions 1 (copYZB), 2 (copY), 3 (copZB) and 4(copB; Fig. 1). No RT-PCR products were obtained for anyof the primer sets in the absence of Cu2+. In the presence of1 g Cu2+ l21, no RT-PCR product was obtained for region1. However, a 2?2 kb RT-PCR product was obtained forregion 3, which indicated that the copZ and copB genes areco-transcribed in Fer1, and that copY is transcribed sepa-rately. Exposure of Fer1 to other metals (Zn2+, Hg2+,Cd2+, Mg2+, As3+, As5+ and Cu+) did not generate a copBor copZB RT-PCR product, suggesting that these metals didnot stimulate transcription under these conditions.

To locate the transcriptional start sites of copZB and copY,primer extension reactions were performed. Two possibletranscriptional start sites were detected at 41 and 62 bpupstream from the copY translational start, and a slightincrease of copY expression was detected when cells wereexposed to Cu2+ (Fig. 4A). Densitometry analysis of thecopZB cDNA product revealed a 15-fold increase of thecopZB transcript when cells were exposed to 5 g Cu2+ l21

(Fig. 4B). In these putative promoter regions, possibleTATA boxes are indicated (Fig. 4), and these elementsare 25, 22 and 21 bases upstream from suggested tran-scription start sites of copZB, copY1 and copY2, respectively.Putative BRE (transcription factor B recognition element)regions are also indicated (Fig. 4), and those for copZBand copY2 are relatively purine rich. Different palindromicsequences 10 bp in size were detected in the TATA-boxregion of copY2 and 9 bases upstream from the copZBtranscription start, possibly indicating protein-bindingelements.

DISCUSSION

Our work has shown that Fer1 is capable of growing in 20 gCu2+ l21, and the level of copper resistance demonstratedby this isolate is amongst the highest recorded to date (Dewet al., 1999). The copper concentrations are greater thantenfold higher than those that most other prokaryotes cantolerate (Table 1). Thus, determining and characterizingcopper transport and resistance mechanisms in Ferroplasmaspecies is of special interest with regard to microbial copperhomeostasis and relevant to the ecology of acidic bioleach-ing environments.

Previous work strongly suggests that Fer1 obtains energy byiron oxidation (Dopson et al., 2004b). Growth rates are

Fig. 4. (A) Primer extension analysis of the Fer1 copY putative Cu2+-resistance transcriptional regulator. Lanes 1–4, DNAsequencing reaction (G, A, T and C, respectively); lane 5, primer extension product from control cells; lane 6, primer extensionproduct from 5 g Cu2+ l”1-exposed cells. (B) Primer extension analysis of the Fer1 copZB putative chaperone and P-typeATPase genes. Lanes 1–4, DNA sequencing reaction (G, A, T and C); lane 5, primer extension product from control cells;lane 6, primer extension product from 5 g Cu2+ l”1-exposed cells. Transcription start positions are indicated by bent arrows atthe left of the figures. DNA sequence fragments at the foot of each figure indicate putative TATA-box regions (underlined),putative BRE regions (lower case) and transcriptional start sites (bold).

http://mic.sgmjournals.org 2643

Analysis of Ferroplasma copper resistance

elevated during iron oxidation, and electron transport chaininhibitors decrease both Fe2+ oxidation (azide) andchemomixotrophic growth on Fe2+ (azide or antimycinA; data not shown). Thus, we examined a link between Fer1energetics and copper exposure by measuring Fe2+-dependent oxygen consumption during short-term expo-sure to Cu2+. This activity was detected in the presence ofextremely high concentrations of Cu2+ (40 and 80 g l21).Moreover, in comparison to no Cu2+ addition, the rateof oxygen consumption increased in the presence of 10or 20 g Cu2+ l21. Similar stimulation of substrate-dependent oxygen consumption has been observed forarsenite addition to the acidophile Acidithiobacillus caldus(Hallberg et al., 1996). This increased respiration in thepresence of added copper is not attributed to an increase ingrowth, as the assay period is conducted over a few minutesonly, and we already detect lower Fer1 growth rates in thepresence of added copper. The increased respiration can beexplained as a response to a higher energy demand requiredto alleviate the affects of increased intracellular levels ofcopper, in processes such as cation efflux, DNA repair andprotein folding.

The putative transcriptional regulator copY is transcribedseparately but in the same direction as the putative coppermetallochaperone copZ and a P-type ATPase transportercopB. The bacterial metallochaperone CopZ coordinates andtransports metals to P-type ATPases (Solioz & Stoyanov,2003), and this is its suggested role in Fer1. The CopBprotein was found to encode the functional consensusmotifs necessary to perform as a copper transporter. Ourprimer-extension results suggest that expression of copZBwas strongly upregulated during Cu2+ exposure (Fig. 4),which was confirmed by RT-PCR. As proposed for a numberof other prokaryotic copper-resistance systems, a possiblemechanism for Cu-mediated regulation may be via a con-formational change in CopY upon binding free cytoplasmicCu+, resulting in its release from the promoter. Thus,transcription of copZ and copB can proceed, presumably toperform copper efflux. Alternatively, CopY may act as apositive regulator, and its binding (in the presence ofcopper) may enhance transcription of copZB. Furtherwork, such as footprinting analysis/S1 mapping, is requiredto determine the transcriptional regulation mechanism ofthis locus in response to copper stress. So far, ourattempts to overexpress CopB have proved unsuccessful,and the cloning suggests that multicopy Fer1 copB istoxic to E. coli (C. Baker-Austin, unpublished results).Regardless of this, the increased expression of the putativechaperone and transporter that we detect in response toincreased copper suggests that their purpose is copperefflux.

Effective metal-ion homeostasis requires the balancedactivity of metal-ion-uptake, efflux, sequestration andredox systems (Solioz & Stoyanov, 2003; Rensing & Grass,2003; Cavet et al., 2003). So far, our screening of theFerroplasma genome sequences has failed to reveal the

presence of other copper transport mechanisms. There isthe possibility that in the acidic metal-leaching environmentof strain Fer1, copper-uptake mechanisms are not required.However, other putative cation and multi-drug transportsystems are evident on the genome, and future character-ization of these and completion of the genome sequence arerequired to facilitate a more complete picture of copperhomeostasis.

The presence of copper can stimulate production of reactiveoxygen species, inducing DNA-strand breaks and oxidationof nucleotide bases (Gaetke & Chow, 2003; Kawanishi et al.,1989). The upregulation of proteins associated with DNArepair suggests that Fer1 undergoes free-radical-inducedstress associated with Cu2+ exposure (Table 3). Cu2+ hasalso been shown to be toxic due to its ability to bind tohistidine residues, and is linked to oxidative damage tolipids and proteins (Gaetke & Chow, 2003). The repertoireof the so-called heat-shock proteins (HSPs; i.e. thermosomeand DnaK) upregulated in Cu2+-exposed Fer1 cells may bedirectly related to conformational alterations to proteingroups within the cell (Table 3). The paradigm for HSPs,also termed protein chaperones or stress proteins, is toprevent aggregation and maintain proteins folded in nativeconformations (Fink, 1999). HSPs are present in allorganisms, and are synthesized at increased levels understress conditions, such as during changes to temperature,salt, pH and redox (Hartl & Hayer-Hartl, 2002).Additionally, upregulation of HSP in response to metalexposure has been widely reported in numerous eukarya, forexample, the effects of cadmium on sea urchins (Roccheriet al., 2004) and of copper on nematodes (Kammenga et al.,1998). However, surprisingly, HSP response to metalexposure is not commonly reported in prokaryotes,although it has been linked to the response to Cu2+ inRhizopus nigricans (Cernila et al., 1999), Co2+ exposure inEnterobacter liquefaciens (Marrero et al., 2004) and Cd2+

exposure in E. coli (Ferianc et al., 1998). It is striking in thisinstance that five of the 21 Fer1 proteins upregulated inresponse to Cu2+ are associated with protein stability. Itwill be of interest to investigate the physiological con-sequences of exposure of Fer1 to high levels of metals and tosee if this is a common response to metal exposure in otherprokaryotes.

A number of other proteins differentially synthesized at least1?5-fold were detected using 2D-PAGE (Table 3). Theseincluded metabolic and electron transport proteins, bio-synthetic proteins, and enzymes involved in transcriptionand translation. This is consistent with a proteome-widetoxicological response to metals, as has been demonstratedin other proteomic studies concerned with microbial metalresistance (Noel-Georis et al., 2004; Vido et al., 2001).Proteins CopB and CopZ were not detected by 2D-PAGE,although we show these were upregulated during Cu2+

exposure. This may be explained by the fact thathydrophobic membrane proteins (CopB) and very smallproteins (CopZ) are difficult to detect using this method

2644 Microbiology 151

C. Baker-Austin and others

(Pradet-Balade et al., 2001). The Fer1 response to metalexposure will involve features other than protein refolding,DNA repair and active copper efflux. For example, adefining characteristic of extremely acidophilic archaea isthe highly impermeable nature of their cell membranes (vande Vossenberg et al., 1998), and this is likely an importantfeature of highly metal-resistant organisms.

In conclusion, we have begun characterization of thecopper-resistance mechanisms of ‘Ferroplasma acidarma-nus’ Fer1, one of the most metal-resistant organisms studiedto date. Responses to copper exposure (at environmentallyrelevant levels) were detected; these included: an increase inrespiration rate, we suggest resulting from increased energydemands; an increase in mRNA transcripts encoding aputative metal chaperone and a P-type ATPase transporter,suggesting their role in copper efflux; and the increasedexpression of DNA-repair and stress proteins (proteinchaperones), to an extent not previously reported in pro-karyotes. These findings provide valuable starting pointsfor further investigation of biochemical and molecularmechanisms of extreme metal resistance.

ACKNOWLEDGEMENTS

We thank Professor Jillian Banfield for useful discussions and the initialprovision of the Fer1 strain. We are indebted to Dr Francis Mulholland(Institute of Food Research, Norwich, UK) for advice and the use ofproteomic scanning equipment, and to Professor Andy Johnston forhis useful suggestions regarding our experiments and criticalcomments on the manuscript. We thank the John Innes CentreProteomic Facility for mass spectrometric analysis and Lynda Flegg fortechnical assistance. C. B. A. was supported by a BBSRC studentship.

REFERENCES

Apweiler, R., Bairoch, A., Wu, C. H. & 12 other authors (2004).UNIPROT: the Universal Protein Knowledgebase. Nucleic Acids Res 32,D115–D119.

Blum, H., Beier, H. & Gross, H. J. (1987). Improved silver staining ofplant-proteins, RNA and DNA in polyacrylamide gels. Electrophoresis8, 93–99.

Bond, P. L., Druschel, G. K. & Banfield, J. F. (2000a). Comparisonof acid mine drainage microbial communities in physically andgeochemically distinct ecosystems. Appl Environ Microbiol 66,4962–4971.

Bond, P. L., Smriga, S. P. & Banfield, J. F. (2000b). Phylogeny ofmicroorganisms populating a thick, subaerial, predominantlylithotrophic biofilm at an extreme acid mine drainage site. ApplEnviron Microbiol 66, 3842–3849.

Brinkman, A. B., Dahlke, I., Tuininga, J. E. & 7 other authors (2000).An Lrp-like transcriptional regulator from the archaeon Pyrococcusfuriosus is negatively autoregulated. J Biol Chem 275, 38160–38169.

Camakaris, J., Voskoboinik, I. & Mercer, J. F. (1999). Molecularmechanisms of copper homeostasis. Biochem Biophys Res Commun261, 225–232.

Cavet, J. S., Borrelly, G. P. M. & Robinson, N. J. (2003). Zn, Cu andCo in cyanobacteria: selective control of metal availability. FEMSMicrobiol Rev 27, 165–181.

Cernila, B., Cresnar, B. & Breskvar, K. (1999). Induction of Hsp70

in the fungus Rhizopus nigricans. Biochem Biophys Res Commun 265,

494–498.

Chan, E. & Weiss, B. (1987). Endonuclease-IV of Escherichia coli is

induced by paraquat. Proc Natl Acad Sci U S A 84, 3189–3193.

Dew, D. W., Muhlbauer, R. & van Buuren, C. (1999). Bioleaching of

copper sulphide concentrates with mesophiles and thermophiles. In

Alta Copper 99. Brisbane, Australia.

Dodd, I. B. & Egan, J. B. (1990). Improved detection of Helix-Turn-

Helix DNA-binding motifs in protein sequences. Nucleic Acids Res

18, 5019–5026.

Dopson, M., Baker-Austin, C., Koppineedi, P. R. & Bond, P. L.(2003). Growth in sulfidic mineral environments: metal resistance

mechanisms in acidophilic micro-organisms. Microbiology 149,

1959–1970.

Dopson, M., Baker-Austin, C. & Bond, P. L. (2004a). First use of

two-dimensional polyacrylamide gel electrophoresis to determine

phylogenetic relationships. J Microbiol Methods 58, 297–302.

Dopson, M., Baker-Austin, C., Hind, A., Bowman, J. P. & Bond, P. L.(2004b). Characterization of Ferroplasma isolates and Ferroplasma

acidarmanus sp nov., extreme acidophiles from acid mine drainage

and industrial bioleaching environments. Appl Environ Microbiol 70,

2079–2088.

Edwards, K. J., Schrenk, M. O., Hamers, R. & Banfield, J. F. (1998).Microbial oxidation of pyrite: experiments using microorganisms

from an extreme acidic environment. Am Miner 83, 1444–1453.

Edwards, K. J., Bond, P. L., Gihring, T. M. & Banfield, J. F. (2000). Anarchaeal iron-oxidizing extreme acidophile important in acid mine

drainage. Science 287, 1796–1799.

Ettema, T. J. G., Huynen, M. A., de Vos, W. M. & van der Oost, J.(2003). TRASH: a novel metal-binding domain predicted to be

involved in heavy-metal sensing, trafficking and resistance. Trends

Biochem Sci 28, 170–173.

Ferianc, P., Farewell, A. & Nystrom, T. (1998). The cadmium-stress

stimulon of Escherichia coli K-12. Microbiology 144, 1045–1050.

Fink, A. L. (1999). Chaperone-mediated protein folding. Physiol Rev

79, 425–449.

Gaetke, L. M. & Chow, C. K. (2003). Copper toxicity, oxidative stress,and antioxidant nutrients. Toxicology 189, 147–163.

Garcia, S., Prado, M., Degano, R. & Dominguez, A. (2002). A

copper-responsive transcription factor, CRF1, mediates copper and

cadmium resistance in Yarrowia lipolytica. J Biol Chem 277, 37359–

37368.

Ghosh, S., Mahapatra, N. R. & Banerjee, P. C. (1997). Metal

resistance in Acidocella strains and plasmid-mediated transfer of this

characteristic to Acidiphilium multivorum and Escherichia coli. Appl

Environ Microbiol 63, 4523–4527.

Hallberg, K. B., Dopson, M. & Lindstrom, E. B. (1996). Arsenic

toxicity is not due to a direct effect on the oxidation of reduced

sulfur compounds by Thiobacillus caldus. FEMS Microbiol Lett 145,

409–414.

Halliwell, B. & Gutteridge, J. M. C. (1984). Oxygen-toxicity, oxygenradicals, transition-metals and disease. Biochem J 219, 1–14.

Hartl, F. & Hayer-Hartl, M. (2002). Protein folding – molecular

chaperones in the cytosol: from nascent chain to folded protein.

Science 295, 1852–1858.

Harvey, P. I. & Crundwell, F. K. (1996). The effect of As(III) on the

growth of Thiobacillus ferrooxidans in an electrolytic cell under

controlled redox potential. Min Engin 9, 1059–1068.

Hesketh, A. R., Chandra, G., Shaw, A. D., Rowland, J. J., Kell, D. B., Bibb,M. J. & Chater, K. F. (2002). Primary and secondary metabolism, and

http://mic.sgmjournals.org 2645

Analysis of Ferroplasma copper resistance

post-translational protein modifications, as portrayed by proteomicanalysis of Streptomyces coelicolor. Mol Microbiol 46, 917–932.

Kammenga, J. E., Arts, M. S. J. & Oude-Breuil, W. J. M. (1998).Hsp60 as a potential biomarker of toxic stress in the nematodePlectus acuminatus. Arch Environ Contam Toxicol 34, 253–258.

Kawanishi, S., Inoe, S. & Yamamoto, K. (1989). Hydroxyl radicaland singlet oxygen production and DNA damage induced bycarcinogenic metal compounds and hydrogen peroxide. Biol TraceElem Res 21, 367–372.

Kelley, L. A., MacCallum, R. M. & Sternberg, M. J. E. (2000).Enhanced genome annotation using structural profiles in theprogram 3D-PSSM. J Mol Biol 299, 499–520.

Krokan, H. E., Standal, R. & Slupphaug, G. (1997). DNA glycosylasesin the base excision repair of DNA. Biochem J 325, 1–16.

Macario, A. J. L., Lange, M., Ahring, B. K. & Conway de Macario, E.(1999). Stress genes and proteins in the Archaea. Microbiol Mol BiolRev 63, 923–967.

Mahapatra, N. R. & Banerjee, P. C. (1996). Extreme tolerance tocadmium and high resistance to copper, nickel and zinc in differentAcidiphilium strains. Lett Appl Microbiol 23, 393–397.

Marrero, J., Gonzalez, L. J., Sanchez, A., Ayala, M., Paz-Lago, D.,Gonzalez, W., Fallarero, A., Castellanos-Serra, L. & Coto, O. (2004).Effect of high concentration of Co (II) on Enterobacter liquefaciensstrain C-1: a bacterium highly resistant to heavy metals with anunknown genome. Proteomics 4, 1265–1279.

McCarty, J. S. & Walker, G. C. (1991). DnaK as a thermometer:Threonine-199 is site of autophosphorylation and is critical forATPase activity. Proc Natl Acad Sci U S A 88, 9513–9517.

Miller, K. W., Risanico, S. S. & Risatti, J. B. (1992). Differentialtolerance of Sulfolobus strains to transition-metals. FEMS MicrobiolLett 93, 69–73.

Nies, D. H. (1999). Microbial heavy-metal resistance. Appl MicrobiolBiotechnol 51, 730–750.

Noel-Georis, I., Vallaeys, T., Chauvaux, R., Monchy, S., Falmagne,R., Mergeay, M. & Wattiez, R. (2004). Global analysis of the Ralstoniametallidurans proteome: prelude for the large-scale study of heavymetal response. Proteomics 4, 151–179.

Nordstrom, D. K. & Alpers, C. N. (1999). Negative pH, efflorescentmineralogy, and consequences for environmental restoration at theiron mountain superfund site, California. Proc Natl Acad Sci U S A96, 3455–3462.

Nordstrom, D. K. & Southam, G. (1997). Geomicrobiology ofsulfide mineral oxidation. In Geomicrobiology: Interactions BetweenMicrobes and Minerals, vol. 35, pp. 361–390. Edited by J. F. Banfield

& K. H. Nealson. Washington, DC: Mineralogical Society ofAmerica.

Pradet-Balade, B., Boulme, F., Beug, H., Mullner, E. W. & Garcia-Sanz, J. A. (2001). Translation control: bridging the gap betweengenomics and proteomics? Trends Biochem Sci 26, 225–229.

Rensing, C. & Grass, G. (2003). Escherichia coli mechanisms ofcopper homeostasis in a changing environment. FEMS Microbiol Rev27, 197–213.

Rensing, C., Fan, B., Sharma, R., Mitra, B. & Rosen, B. P. (2000).CopA: an Escherichia coli Cu(I)-translocating P-type ATPase. ProcNatl Acad Sci U S A 97, 652–656.

Roccheri, M. C., Agnello, M., Bonaventura, R. & Matranga, R.(2004). Cadmium induces the expression of specific stress pro-teins in sea urchin embryos. Biochem Biophys Res Commun 321,80–87.

Sawers, G. & Bock, A. (1989). Novel transcriptional control of thepyruvate formate-lyase gene – upstream regulatory sequences andmultiple promoters regulate anaerobic expression. J Bacteriol 171,2485–2498.

Soares, E. V., Hebbelinck, K. & Soares, H. (2003). Toxic effectscaused by heavy metals in the yeast Saccharomyces cerevisiae: acomparative study. Can J Microbiol 49, 336–343.

Solioz, M. & Stoyanov, J. V. (2003). Copper homeostasis inEnterococcus hirae. FEMS Microbiol Rev 27, 183–195.

Teitzel, G. M. & Parsek, M. R. (2003). Heavy metal resistance ofbiofilm and planktonic Pseudomonas aeruginosa. Appl EnvironMicrobiol 69, 2313–2320.

Tsivkovskii, R., Efremov, R. G. & Lutsenko, S. (2003). The role ofthe invariant His-1069 in folding and function of the Wilson’sdisease protein, the human copper-transporting ATPase ATP7B.J Biol Chem 278, 13302–13308.

Tusnady, G. E. & Simon, I. (2001). The HMMTOP transmembranetopology prediction server. Bioinformatics 17, 849–850.

Tyson, G. C. J., Hugenholtz, P., Allen, E. E., Ram, R. J., Richardson,P. M., Solovyev, V. V., Rubin, E. M., Rokhsar, D. S. & Banfield, J. F.(2004). Community structure and metabolism through reconstruc-tion of microbial genomes from the environment. Nature 428,25–26.

van de Vossenberg, J., Driessen, A. J. M. & Konings, W. N. (1998).The essence of being extremophilic: the role of the unique archaealmembrane lipids. Extremophiles 2, 163–170.

Vido, K., Spector, D., Lagniel, G., Lopez, S., Toledano, M. B. &Labarre, J. (2001). A proteome analysis of the cadmium response inSaccharomyces cerevisiae. J Biol Chem 276, 8469–8474.

2646 Microbiology 151

C. Baker-Austin and others