Effects of the Metalloid Oxyanion Tellurite (TeO32-) on Growth Characteristics of the Phototrophic Bacterium Rhodobacter capsulatus

This work examines the effects of potassium tellurite (K₂TeO₃)on the cell viability of the facultative phototroph Rhodobactercapsulatus. There was a growth mode-dependent response in whichcultures anaerobically grown in the light tolerate the presenceof up to 250 to 300 µg of tellurite (TeO₃^2–) perml, while dark-grown aerobic cells were inhibited at telluritelevels as low as 2 µg/ml. The tellurite sensitivity ofaerobic cultures was evident only for growth on minimal saltmedium, whereas it was not seen during growth on complex medium.Notably, through the use of flow cytometry, we show that thecell membrane integrity was strongly affected by tellurite duringthe early growth phase ( $≤$ 50% viable cells); however, at the endof the growth period and in parallel with massive telluriteintracellular accumulation as elemental Te⁰ crystallites, recoveryof cytoplasmic membrane integrity was apparent ( $≥$ 90% viable cells),which was supported by the development of a significant membranepotential ( ${Delta}$ ${psi}$ = 120 mV). These data are taken as evidence thatin anaerobic aquatic habitats, the facultative phototroph R.capsulatus might act as a natural scavenger of the highly solubleand toxic oxyanion tellurite.

INTRODUCTION

Top

Introduction

Potassium tellurite (K₂TeO₃) has long been recognized as toxicto eukaryotic and prokaryotic cells; furthermore, because ofits antimicrobial properties, it has been used in selectivemedia for the isolation of a number of naturally tellurite-resistantbacterial species (19). Over the past few decades, telluriumhas also found wide use in areas such metallurgy and the electronicsindustry. As a consequence, the toxic effects associated withthe accumulation of tellurium compounds, particularly the water-solubleoxyanion tellurite (TeO₃^2–), has become a concern (19).

Some gram-positive bacteria show an intrinsic low-level resistanceto TeO₃^2– (50 to 120 µg/ml) (18), while high-levelresistance (500 to 2500 µg/ml) has been determined incertain gram-negative obligate aerobic photosynthetic species(14, 28). As a reference, Escherichia coli growth is inhibitedat a tellurite concentration as low as 1 µg/ml. Growthin the presence of tellurite is often associated with the reductionof the oxyanion to elemental tellurium (Te⁰), which leads toblackening of the cells due to either internal or periplasmicaccumulation of Te (19, 22). In some instances the capacityto grow at higher tellurite concentrations has been shown todepend on the presence of genetic determinants carried on IncHI,IncHII, and IncP plasmids (25). In addition, chromosomal genesimportant for growth in the presence of K₂TeO₃ have been identifiedin a few species, but their role has not been clearly determined(7, 20, 21).

The purple nonsulfur bacteria include several species that showintrinsic high-level resistance to potassium tellurite rangingfrom 50 to 1,000 µg/ml (9). Rhodobacter sphaeroides ischaracterized by growth mode dependence on the tellurite resistancelevel: complex media confer much lower resistance (10 to 80µg/ml) than minimal media (100 to 1,000 µg/ml),and aerobic cultures are more resistant (800 to 1,000 µg/ml)than photosynthetic cultures (400 to 700 µg/ml). In R.sphaeroides, two loci involved in determining high-level resistanceto tellurite oxyanions have been identified (11). The closerelative Rhodobacter capsulatus is also able to withstand highconcentrations of tellurite and to internally accumulate a considerableamount of elemental tellurium, as demonstrated by transmissionelectron microscopy and X-ray microanalysis (2). It has recentlybeen shown that the uptake of tellurite by R. capsulatus dependson the ${Delta}$ pH component of the transmembrane proton motive force(3). In this facultative phototroph, it was also reported that,during photosynthetic growth, the presence of potassium telluritecaused changes in the structure of the branched respiratorychain leading to a decrease in the content of c-type cytochromes,paralleled by a low cytochrome c oxidase activity (2). An analogouseffect has been seen in Pseudomonas pseudoalcaligenes KF707(4).

In this work we analyzed the effects of potassium telluriteon the viability of R. capsulatus cells. We report that photosyntheticanaerobic cultures are able to grow in the presence of highconcentrations of tellurite (250 to 300 µg of TeO₃^2–per ml); conversely, dark-grown aerobic cultures are highlysensitive to low tellurite concentrations (2 µg/ml) onlyfor growth in minimal salt medium. We also show by flow cytometricanalysis that the high proportion of damaged and dead cellsseen at the early growth phase is drastically reduced at theend of the growth curve in parallel with a consistent intracellularaccumulation of elemental tellurium. These results suggest thatfacultative photosynthetic bacteria might play a significantenvironmental role in intracellularly precipitating the toxicoxyanion tellurite (TeO₃^2–) into a form, Te⁰, that isless toxic and less available to other organisms.

MATERIALS AND METHODS

Top

Materials and Methods

Bacterial strain and growth conditions.
R. capsulatus B100 was grown under chemotrophic conditions at30°C in Erlenmeyer flasks shaken at 150 rpm and under phototrophicconditions in completely filled screw-cap tubes and bottlesat 30°C with an incident light intensity of 200 W m^–2.The cells were cultivated in both rich (YPS) and minimal (RCV)media (26). YPS medium contained yeast extract, 0.3%; peptone,0.3%; CaCl₂, 2 mM; and MgSO₄, 2 mM, in deionized water, pH 6.8.RCV medium contained DL-malic acid, 0.1%; (NH₄)₂SO₄, 0.4%; EDTA,0.002%; MgSO₄ · 7H₂O, 0.02%; trace elements, 1 ml/liter;CaCl₂ · 2H₂O, 0.0075%; FeSO₄ · 7H₂O, 0,0012%;thiamine hydrocloride, 10^–4%; and KPO₄ buffer, 10 mM,in deionized water, pH adjusted to 6.8 before autoclaving. Traceelements solution contained, per 250 ml of deionized water,MnSO₄ · H₂O, 397.5 mg; H₃BO₃, 700 mg; Cu(NO₃)₂ ·3H₂O, · 10 mg; ZnSO₄ · 7H₂O, 60 mg; and NaMoO₄· 2H₂O, 187.5 mg. Anaerobic growth in the dark was achievedby adding fructose (20 mM) and dimethyl sulfoxide (60 mM) tofilled screw-cap tubes containing either YPS or RCV medium andincubating at 30°C in the dark. Potassium tellurite (K₂TeO₃)was added to the growing cultures at the concentrations specifiedfor each experiment in the Results section.

MIC.
The MIC of potassium tellurite was determined on agar plateswith either rich or minimal medium prepared with increasingconcentrations of K₂TeO₃; 50-µl drops of one- to fivefolddilutions of fully grown cultures were laid on the surface ofthe plates, which were then incubated at 30°C under eitherrespiratory, photosynthetic, or anaerobic-dark conditions for3 to 14 days, depending on the growth mode. Anaerobic incubationswere performed in closed anaerobic jars in the presence of theAnaerocult A anaerobic system (Merck). The blackening of thearea corresponding to the culture drop, as a result of telluritereduction to elemental tellurium by the cells, was scored aspositive for growth.

Biochemical methods.
The protein content of whole cells was determined by the methodof Lowry et al. (8) after a 1-min incubation with 0.1 N NaOHin boiling water. Crystalline bovine serum albumin (Sigma) wasused as the protein standard. The quantitative determinationof potassium tellurite in liquid media was done with the reagentdiethyldithiocarbamate (Sigma) as described by Turner et al.(23). The dissolved oxygen concentration in the growth mediaand respiratory activities were determined through the use ofa Clark-type oxygen electrode (YSI 53; Yellow Springs InstrumentsInc., Yellow Springs, Ohio).

Transmission electron microscopy.
Bacterial cells grown in the presence of tellurite (25 to 50µg/ml) were harvested after the cultures became blackand processed for electron microscopic analysis. Cell pelletswere first washed in 0.05 M cacodylate buffer (pH 7.2) and thenfixed for 2 h in 0.05 M cacodylate-1.5% (wt/vol) glutaraldehyde(pH 7.2). The same buffer was then used for overnight washingof the sample, followed by 2 h of fixation with 2% (wt/vol)osmium tetroxide and dehydration with ethanol. Finally, thesamples were embedded in Durcopan, and thin sections preparedwith an LKB Ultratome Nova were double stained with uranyl acetateand lead citrate (13). Specimens were examined with a PhilipsCM-100 transmission electron microscope.

Flow cytometry.
Flow cytometry experiments were done according to Ziglio etal. (29). Briefly, cell pellets were diluted with 0.22-µm-pore-size-filteredTricine buffer (0.1 M Tricine, 10 mM MgCl₂ at pH 7.4) and stainedwith SYBR Green I (SYBR I) and propidium iodide (Molecular ProbesInc., Eugene, Oreg.) by adding 10 µl of both fluorochromesper ml of sample containing about 10⁶ to 10⁷ cells. The double-stainingequilibrium was assumed to be reached within 15 min of incubationat room temperature in the dark. Samples were analyzed witha Bryte-HS flow cytometer (Bio-Rad, Hercules, Calif.) with ahigh light-scattering sensitivity (16) and equipped with a short-arcxenon lamp (Hamamatsu). Samples were excited at 470 to 490 nm,and fluorescence emission was detected at 515 to 565 nm forSYBR I ( ${lambda}$ _ex = 488 nm, ${lambda}$ _em = 525 nm) and at 590 to 720 nm for propidiumiodide ( ${lambda}$ _ex = 530 nm, ${lambda}$ _em = 620 nm). Samples were analyzed for3 to 4 min at a flow rate of 1.5 µl min^–1 for cellenumeration. Data acquisition was triggered to reduce interferenceby nonfluorescent particles (debris), while photomultipliergains were set in the logarithmic mode, and data were recordedas list mode files by WinBryte software (Bio-Rad).

TPP⁺ uptake measurements.
A polyvinyl chloride membrane selectively permeable to tetraphenylphosphonium(TPP⁺) was constructed as described (5), and TPP⁺ uptake measurementswere performed as reported by Rugolo and Zannoni (14). An internalcell volume of 102 µl per µmol of bacteriochlorophyllwas assumed, according to Kell et al. (6).

RESULTS

Top

Results

MIC determination under different growth conditions.
Rhodobacter capsulatus, a member of the family Rhodobacteraceae,is metabolically highly versatile and, as a consequence, ableto grow under a wide variety of conditions. The MICs of TeO₃^2–were determined for cells growing by aerobic respiration, photosynthesis,and anaerobic respiration (with dimethyl sulfoxide as the exogenousoxidant) on different media. It is immediately apparent fromTable 1 that aerobic conditions greatly influenced the responseof the cells grown on minimal RCV medium to the presence oftellurite. Under these conditions R. capsulatus was highly resistantto tellurite under photosynthetic-anaerobic conditions (250µg/ml) but very sensitive (2 µg/ml) in the presenceof oxygen. Anaerobic nonphotosynthetic conditions gave a mediumlevel of resistance (20 µg/ml).

View this table:
[in this window]
[in a new window]
TABLE 1. MICs of K₂TeO₃ for R. capsulatus grown with different media and growth conditions

The resistance profile was different for cells grown on complexYPS medium. Under all growth conditions, there was a high levelof resistance. Aerobiosis did not appear to have any evidentrole in influencing the MIC of tellurite on this medium. Underaerobic conditions, the addition to minimal RCV medium of thecomponents of complex YPS medium (yeast extract and peptone)did not restore the high MIC typical of these cultural conditions,even when added together and at the same concentrations usedin YPS medium (Table 1). When these components were added ata low concentration (0.06%), the MIC increased from 2 to 10µg/ml, but if they were added at the concentration typicalof complex medium (0.3%), the MIC increased to 50 µg/ml.The same effect on the MIC was seen if the two components wereadded separately. The opposite effect was seen if YPS mediumplates were amended with malate under aerobic conditions. Underthese conditions the cells became more sensitive to telluriterelative to growth on normal rich medium, suggesting that malateis a key element in determining the level of tolerance of R.capsulatus cells to potassium tellurite.

Growth in the presence of tellurite.
In accord with the MICs determined on agar plates, liquid culturesof R. capsulatus with 50 µg of potassium tellurite perml developed only in the absence of oxygen or on complex medium,with no growth occurring on minimal medium under aerobic conditions.An unexpected observation was that even photosynthetic culturesinoculated in completely filled screw-cap tubes containing RCVminimal medium started to grow and to reduce tellurite onlyafter a lag period of approximately 120 h. This finding is insharp contrast to the observation that photosynthetic growthand blackening of the cells on agar plates amended with telluritewere clearly visible within 48 h of inoculation.

To reconcile these contrasting findings, it was considered thattrue anaerobiosis was normally reached only for agar platesanaerobically incubated in jars where all the gaseous O₂ wasrapidly (30 min) consumed by the anaerobiosis-generating system,while in freshly inoculated screw-cap tubes there was enoughdissolved O₂ (about 190 µM, compared to a concentrationof 250 µM for O₂-saturated water, as measured by a Clark-typeoxygen electrode) to prevent growth in the presence of telluriteas seen under aerobic conditions. Accordingly, in newly inoculatedtubes kept in the dark at 30°C for 20 h to allow the metabolicactivity of the cells to consume the dissolved oxygen, no measurable( $≤$ 20 µM by polarography) dissolved oxygen in the growthmedium was present. As a consequence of this treatment, thelag period seen before the onset of growth decreased to about5 h (Table 2). The same dark incubation was performed in thepresence of 2 µg of the redox indicator resazurin perml, which turned colorless, confirming that reducing-anaerobicconditions had been reached (not shown).

View this table:
[in this window]
[in a new window]
TABLE 2. Lag periods before the onset of growth and dissolved oxygen concentrations of R. capsulatus photosynthetic cultures grown on minimal medium in the presence of potassium tellurite^a

As a control for the oxygen effect, hydrogen peroxide (H₂O₂)was added after the dark incubation period at a final concentrationof 0.73 mM in the presence of 50 µg/ml tellurite, andthe cultures were put in the light. Under these conditions,there was a catalase-dependent production of about 240 µMoxygen and the initial 120-h lag period was restored (Table2). Control cultures with no tellurite added showed that thesame concentration of H₂O₂ had no measurable effect on photosyntheticgrowth. In order to better define the effect of the oxyanionon cellular metabolism, the dark incubation was performed inthe presence of tellurite. At the end of the incubation period,no oxygen had been consumed by the cells, and the cultures stillshowed the long lag period before the onset of photosyntheticgrowth (Table 2). These findings point to tellurite inhibitionof oxidative metabolism.

Tellurium uptake by cells.
R. capsulatus cultures growing in the presence of telluritetake up the oxyanion from the medium and accumulate it insidethe cells as black deposits of elemental tellurium (Te⁰) (2).This activity requires an energized cytoplasmic membrane andis a ${Delta}$ pH-dependent process (3). Transmission electron microscopyphotographs (Fig. 1) clearly showed that Te⁰ accumulated inthe cytoplasm in shapes and sizes that were dependent on thegrowth conditions. Aerobically grown cells showed rod-like andbarrel-like particles that were 40 to 50 nm in size (Fig. 1B),whereas cells grown under photosynthetic reducing conditionsshowed black deposits of splinterlike shape, varying in lengthfrom 75 to more than 350 nm (Fig. 1C and D).

View larger version (147K):
[in this window]
[in a new window]
FIG. 1. Electron micrographs of R. capsulatus B100 cells grown in the absence (A) or in the presence (B, C, D) of 50 µg of K₂TeO₃ per ml. (A) Photosynthetic-anaerobic growth. (B) Aerobic growth. (C) Photosynthetic-anaerobic growth on minimal medium. (D) Photosynthetic-anaerobic growth on rich medium.

The uptake of potassium tellurite by the cells was measuredby monitoring the disappearance of the oxyanion from the growthmedium in both aerobic and photosynthetic cultures. At an initialtellurite concentration of 25 µg/ml, the oxyanion completelydisappeared from the medium in approximately 30 h under aerobicconditions on rich medium and after 22 to 24 h under photosynthetic-anaerobicconditions (Fig. 2B). Similar results were obtained at an initialconcentration of 50 µg of tellurite per ml, with the oxyaniondisappearing from the medium in approximately 34 h (not shown).The presence of tellurite in the medium negatively affectedthe growth rate of the culture to a degree that was influencedby the growth mode (Fig. 2A). At 25 µg of K₂TeO₃ per ml,photosynthetic cultures showed a more than twofold increasein doubling time, 2.7 for cells grown on rich medium and 2.3for cells grown on minimal medium, versus a 3.8-fold increasefor aerobic cultures. Growing the cultures at 50 µg oftellurite per ml further increased the measured doubling times,as expected (data not shown).

View larger version (20K):
[in this window]
[in a new window]
FIG. 2. (A) Growth curves of cells grown in the presence of 25 µg of K₂TeO₃ per ml. Solid lines, K₂TeO₃-grown cultures; dashed lines, control cultures. (B) Variation of tellurite concentration in the growth medium. The same symbols represent the same growth conditions in the two panels. Symbols: ${blacksquare}$ , aerobic conditions on rich medium; •, photosynthetic conditions on minimal medium; ${blacktriangleup}$ , photosynthetic conditions on rich medium.

Cell membrane integrity as determined by flow cytometry.
In this study, the determination of viable and dead R. capsulatuscells grown in the presence and absence of potassium telluritewas done on the basis of their membrane integrity as determinedby staining with SYBR I and propidium iodide, two high-affinitynucleic acid dyes that give bright staining of cells (1, 27).Although no specific information is available on the structureand chemical properties of SYBR I or on its mode of bindingto DNA (10), it has widely been shown to be capable of stainingall cells, living or dead. Conversely, the polarity of propidiumiodide allows it to penetrate only inactive cell membranes,which are characteristic of dead or damaged cells. In dead cells,the simultaneous presence of SYBR I and propidium iodide activatesenergy transfer between SYBR I and propidium iodide so thatthe fluorescence emission of SYBR I is no longer visible. Inthis way it is possible to distinguish viable cells (green fluorescent)from dead cells (red fluorescent).

In the cytograms shown in Fig. 3 two regions are defined, inthe three-dimensional representation, by peaks of fluorescencesignals which are indicative of different bacterial physiologicalstates, dead and damaged cells and viable cells. R. capsulatuscells harvested from a control culture in the exponential growthphase (Fig. 3A) showed multiple peaks in a region of high-intensitygreen fluorescence and low-intensity red fluorescence. A minutepeak corresponding to dead cells was visible in the low-green-fluorescencearea. Cells grown in the presence of potassium tellurite (Fig.3B) showed, on the contrary, a tall peak in the correspondingdead cells area. The data from flow cytometry experiments arecollected in Table 3. Here it is shown that at the beginningof the growth phase (4 h), approximately 94% of the cells inthe control cultures and 51% of the cells in tellurite-growncultures were viable, suggesting significant damage of the cytoplasmicmembrane by the toxic oxyanion at 50 µg/ml, correspondingto a tellurite-cell ratio of 1 ng/5,000. The percentage of damagedand dead cells decreased drastically after 24 h of growth, parallelto a complete disappearance of tellurite from the growth mediumover the same period (Table 3; Fig. 2). At the end of the growthperiod (48 h), with no tellurite present in the medium, a strongrecovery of the cytoplasmic membrane integrity in tellurite-growncells was apparent, with more than 90% of cells viable, similarto the control cultures (Table 3).

View larger version (47K):
[in this window]
[in a new window]
FIG. 3. Three-dimensional fluorescence cytograms of R. capsulatus cells grown on minimal medium and stained with SYBR I and propidium iodide. (A) Cells grown in the absence of potassium tellurite; total count, 2.8 x 10⁸ cells/ml. (B) Cells grown in the presence of 50 µg of potassium tellurite per ml; total count, 3.3 x 10⁸ cells/ml.

View this table:
[in this window]
[in a new window]
TABLE 3. Cell viability in photosynthetic cultures grown on minimal medium in the presence of 50 µg of potassium tellurite per ml

Cell membrane permeability as determined by TPP⁺ cation uptake.
To test the capacity of the cell membrane to generate and maintaina consistent electric potential ( ${Delta}$ ${psi}$ ) produced by proton extrusioncoupled to either respiratory or photosynthetic electron transport,the distribution of the lipophilic cation TPP⁺ was analyzed.Figure 4 (continuous trace) shows that after the addition ofR. capsulatus cells grown for 24 h in the absence of tellurite(control), a rapid upward deflection of the trace, indicativeof TPP⁺ uptake (development of a negative potential inside thecells), was seen. Notably, the initial level of membrane potential( ${Delta}$ ${psi}$ ) cannot be maintained by respiration, possibly due to rate-limitingoxygen diffusion through the external membrane or wall structureof R. capsulatus. Under steady-state respiratory conditions(reached after approximately 5 min), the estimated ${Delta}$ ${psi}$ was approximately114 ± 3 mV, while under both respiration and continuousillumination (light on), the ${Delta}$ ${psi}$ went up to 136 ± 3 mV. ${Delta}$ ${psi}$ = 0 was defined by the TPP⁺ level seen in the dark after additionof a 1.5 µM concentration of the protonophore FCCP (carbonylcyanide p-trifluoromethoxyphenylhydrazone).

View larger version (13K):
[in this window]
[in a new window]
FIG. 4. Light-induced and oxygen-dependent uptake of TPP⁺ ions by R. capsulatus cells grown in the absence (continuous trace) or in the presence (interrupted trace) of potassium tellurite. Following the calibration addition of TPP⁺ (1 µM final concentration), cells (2.8 and 3.3 mg of protein for control and tellurite-grown cells, respectively) were added to 2 ml of air-saturated medium in 50 mM TES buffer, pH 7.5-10 mM KCl at 28°C. TPP⁺, tetraphenylphosphonium cation; FCCP, carbonyl cyanide p-trifluoromethoxyphenylhydrazone; ON, light on; OFF, light off.

The interrupted trace in Fig. 4 shows TPP⁺ uptake by R. capsulatuscells grown (24 h) in the presence of tellurite. Apparently,while the general TPP⁺ uptake pattern is qualitatively similarto that seen in cells grown in the absence of tellurite (continuousline), the ${Delta}$ ${psi}$ values reached by respiration (104 ± 3 mV)and respiration plus photosynthesis (120 ± 3 mV) werelower than those seen with control cells. It is noteworthy thatthe ${Delta}$ ${psi}$ values calculated for tellurite-grown cells were alreadycorrected for a decreased total internal volume (V_i) due toan 11% increase in damaged cells and a 30% increase in bacteriochlorophyllcontent (3). Respiratory measurements performed in parallelwith TPP⁺ determinations indicated that the endogenous respirationby cells of R. capsulatus grown in the presence of telluritewas inhibited 40% by light, whereas it was severely inhibited(80%) in control cells (not shown). This suggests, in line withearly reports (14), that the light-generated ${Delta}$ ${psi}$ exhibits strongcontrol over respiration under normal growth conditions, whereasthis phenomenon is much less evident in tellurite-grown cells.The reduced capacity of tellurite-grown cells to inhibit respirationduring continuous illumination might be due to a restrictedlight-dependent electron flow caused by a lower cytochrome ccontent (2) (see Discussion).

DISCUSSION

Top

Discussion

The facultative photosynthetic bacterium Rhodobacter capsulatusshows a high level of resistance to potassium tellurite, likeother phototrophic species (9), and accumulates it in the cytoplasm,upon reduction, as elemental tellurium (Fig. 1) (2). We determinedthe MICs of tellurite under a variety of growth conditions andshowed that the increasing level of resistance depends primarilyon the absence of oxygen (Table 1). Aerobic conditions on minimalmedium allowed very low resistance (2 µg/ml), while underphotosynthetic-anaerobic conditions on the same medium, thecells were able to grow in the presence of up to 250 µgof potassium tellurite per ml. Even growth under anaerobic conditionsin the dark, made possible by the addition of dimethyl sulfoxideas an electron sink, which represents an energetically lessfavorable situation than growth by aerobic respiration (15),showed a resistance level of 20 µg/ml.

A second factor influencing the level of resistance to TeO₃^2–was the composition of the medium. YPS rich medium determinedan increase of resistance under all growth conditions, mostlyin aerobiosis (Table 1). This finding was taken as an indicationthat the components of the rich medium may exert a protective,antioxidative action on the cells in the presence of potassiumtellurite, which is believed to have strong oxidant properties(17). Interestingly, when MIC determinations were made on mediaof mixed composition, there was a significant effect under aerobicconditions: the addition of malate to rich medium was sufficientto drastically decrease the level of resistance to tellurite,while the addition of yeast extract and peptone, either togetheror separately, to minimal medium had a protection effect thatonly marginally increased the measured MIC (Table 1). Underphotosynthetic conditions there was no such effect. These datasuggest that malate might have an important role in the mechanismdetermining the susceptibility of R. capsulatus cells to potassiumtellurite.

The effects of growth conditions and medium composition on theresistance pattern to tellurite, as shown here, appear in sharpcontrast to previous observations made for the closely relatedspecies R. sphaeroides (9). In this species, the effect of mediumcomposition on TeO₃^2– resistance, even under aerobic conditions,was indeed the opposite of that seen in R. capsulatus. Thesecontrasting findings may possibly be reconciled by consideringthat species belonging to the genus Rhodobacter show a relativelyhigh degree of metabolic differences and that the method usedfor determining resistance to TeO₃^2–, i.e., MICs measuredon liquid media (9), probably masked the selection of hyperresistantvariants, resulting in an increase in the apparent MICs (11).

The effect of oxic conditions was also evident in liquid cultures.Aerobic incubation on minimal medium in the presence of telluriteprevented the growth of R. capsulatus cells. Even completelyfilled screw-cap tubes, used for phototrophic growth, containedenough dissolved oxygen (188 µM) to inhibit growth inthe presence of the oxyanion. The protective action of the richmedium components was seen in the liquid cultures as for growthon agar plates. The effect of O₂ on growth on minimal mediumin the presence of tellurite was confirmed in experiments inwhich the level of dissolved oxygen was changed, taking advantageof the metabolic activity of the cells, or by addition of H₂O₂to the culture tubes (Table 2).

Two hypothetical explanations can be considered to rationalizethese observations. First, oxygen above a certain thresholdlevel induces the production of molecules involved in aerobicmetabolism, which are preferred targets for the oxidative actionof the TeO₃^2– anion and whose inactivation prevents aerobicgrowth. In this scenario, components of the rich medium couldexert a protective, antioxidative action, thereby making aerobicgrowth possible. As a second hypothesis, tellurite inhibitsthe cellular response to oxidative stresses, as would happenduring the subculturing of photosynthetic cultures, in whichcells adapted to photosynthetic-anaerobic conditions must confronta sudden increase of oxygen due to the high level (75 to 80%of saturation) of O₂ normally present in fresh medium. Undernormal conditions, R. capsulatus cells can easily overcome thistype of oxidative stress, while the presence of tellurite wouldprevent the adaptive response. The decreased tellurite resistanceunder oxic conditions might be due to tellurite damage of proteinsand free thiols as well as displacement of other metal ionsfrom biomolecules (25). Indeed, oxidation of free thiol groupson proteins and small redox buffers such as glutathione, cytochromes,and dihydroascorbate can be a source of oxidative stress (24).Thus, proteins acting in defense against the oxidative stresscould be the immediate targets of tellurite oxyanions.

The ability of R. capsulatus cells to reduce tellurite to elementaltellurium and internalize it, as part of their tellurite responsemechanism, is evident in transmission electron microscopic imagesin which deposits of Te⁰ are clearly visible in the cytoplasm(Fig. 1) (2). Apparently differences in the size and internaldistribution of the tellurium deposits are related to the growthmode of the cultures (Fig. 1). Whether these various forms ofelemental tellurium depend on different reduction mechanismsremains to be determined.

Growth in the presence of potassium tellurite increased thedoubling time of the culture (Fig. 2A) to a degree that dependedon the initial concentration of the oxyanion (not shown). Atall initial concentrations and under all growth conditions tested,R. capsulatus cultures were able to completely take up the telluriteoxyanion from the medium and accumulate it in the cytoplasm.Interestingly, as soon as the cells overcame the initial tellurite-inducedstress (tellurite/cell ratio of 1 ng/5,000), presumably uponactivation of a consistent tellurite uptake that leads to completedisappearance of the toxic oxyanion from the medium (Fig. 2B),the membrane integrity, which was severely impaired in the earlygrowth phase, was substantially restored, as indicated by flowcytometric analysis (Table 3; Fig. 3) and by membrane potentialmeasurements (Fig. 4). In this respect, various parameters suchas (i) reproductive growth, (ii) metabolic activity, and (iii)membrane integrity have recently been discussed in the lightof the most suitable markers defining the concept of cell viability(10). Apparently, reproductive growth represents the most stringentproof of viability because it requires both conditions ii andiii. However, cells with an intact membrane are supposed tobe capable of metabolic activity and show reproductive growthunder appropriate conditions. Conversely, dead bacterial cellscontain a permeabilized cytoplasmic membrane, freely exposedto the environment.

Here we have shown by flow cytometry that potassium telluriteaffects the membrane integrity of R. capsulatus cells in theirearly growth phase (Table 3), while the membrane permeabilityof stationary-phase cells is similar to that of control cells.Furthermore, the data in Table 3 show that the photosyntheticgrowth capacity of control cells and cells grown in the presenceof tellurite is similar. These results, along with those recentlyobtained with membrane vesicles (chromatophores) isolated fromlight-grown cells in the presence of tellurite and harvestedat their stationary phase of growth (2) can be taken as evidenceof the presence of an efficient photosynthetic metabolism andcell viability. Herein, we have also shown that tellurite toxicityis strongly enhanced in R. capsulatus by aerobic conditions,particularly in minimal salt medium. Finally, the capacity ofR. capsulatus to massively accumulate the highly soluble telluritein the form of insoluble elemental tellurium crystallites suggeststhat facultative phototrophs might play a significant environmentalrole in subtracting the toxic oxyanion from polluted aquaticsites.

ACKNOWLEDGMENTS

We thank MIUR for supporting this work (PRIN2003).

We are indebted to R. Randi and F. Pennetta for their skillfulassistance with transmission electron microscopy and telluriumuptake determinations, respectively.

FOOTNOTES

* Corresponding author. Mailing address: Dipartimento di Biologia, Università di Bologna, Via Irnerio 42, 40126 Bologna, Italy. Phone: (051) 2091300. Fax: (051) 242576. E-mail: roberto.borghese{at}unibo.it.

REFERENCES

Top