Physiology of Geobacter Metallireducens Under Excess and Limitation Of

Systematic and Applied Microbiology 37 (2014) 287–295

Contents lists available at ScienceDirect

Systematic and Applied Microbiology

j ournal homepage: www.elsevier.de/syapm

Physiology of Geobacter metallireducens under excess and limitation of

electron donors. Part II. Mimicking environmental conditions during

cultivation in retentostats

a b c,d e

Sviatlana Marozava , Wilfred F.M. Röling , Jana Seifert , Robert Küffner ,

c,f,g a,∗

Martin von Bergen , Rainer U. Meckenstock

Institute of Groundwater Ecology, Helmholtz Zentrum München, Ingolstädter Landstraße 1, 85764 Neuherberg, Germany

Department of Molecular Cell Physiology, Faculty of Earth and Life Sciences, VU University Amsterdam, De Boelelaan 1085, 1081 HV Amsterdam, The Netherlands

Department Proteomics, Helmholtz Center for Environmental Research – UFZ, Permoserstraße 15, 04318 Leipzig, Germany

Institute of Animal Nutrition, University of Hohenheim, Emil-Wolff-Str. 10, 70599 Stuttgart, Germany

Teaching and Research Unit Bioinformatic, Institut für Informatik, Ludwig-Maximilians-Universität München, Amalienstr. 17, 80333 Munich, Germany

Department Metabolomics, Helmholtz Center for Environmental Research – UFZ, Permoserstraße 15, 04318 Leipzig, Germany

Department of Biotechnology, Chemistry and Environmental Engineering, Aalborg University, Sohngaardsholmsvej 49, DK-9000 Aalborg, Denmark

a r t i c l e i n f o a b s t r a c t

Keywords: The strict anaerobe Geobacter metallireducens was cultivated in retentostats under acetate and acetate

Geobacter metallireducens

plus benzoate limitation in the presence of Fe(III) citrate in order to investigate its physiology under

Retentostats −1

close to natural conditions. Growth rates below 0.003 h were achieved in the course of cultivation. A

Label-free proteomics

nano-liquid chromatography–tandem mass spectrometry-based proteomic approach (nano-LC–MS/MS)

In situ physiology

with subsequent label-free quantiﬁcation was performed on proteins extracted from cells sampled

−1

at different time points during retentostat cultivation. Proteins detected at low (0.002 h ) and high

−1

(0.06 h ) growth rates were compared between corresponding growth conditions (acetate or acetate

plus benzoate). Carbon limitation signiﬁcantly increased the abundances of several catabolic proteins

involved in the degradation of substrates not present in the medium (ethanol, butyrate, fatty acids,

and aromatic compounds). Growth rate-speciﬁc physiology was reﬂected in the changed abundances of

energy-, chemotaxis-, oxidative stress-, and transport-related proteins. Mimicking natural conditions by

extremely slow bacterial growth allowed to show how G. metallireducens optimized its physiology in order

to survive in its natural habitats, since it was prepared to consume several carbon sources simultaneously

and to withstand various environmental stresses.

BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

Introduction extreme conditions, such as the deep biosphere, microorganisms

are exposed to a very low supply of organic carbon but bacteria are

The natural environment provides microorganisms with differ- still able to survive under such oligotrophic conditions [13].

ent conditions compared to well-established laboratory cultivation Therefore, the question arises of how microorganisms with slow

systems. In microbial habitats, carbon is typically present at very bacterial growth utilize carbon under environmental conditions of

low concentrations and as a mixture of several substrates [17,20]. low substrate concentrations. One strategy would be to increase the

For example, the content of dissolved organic carbon in ground- amounts of transporters of limiting substrates [17]. Additionally,

␮

water can range from 75 to 180 M [7,20]. Additionally, under to be able to react quickly to changing conditions, microorganisms

may express various catabolic proteins that are not involved in the

degradation of available substrates [18]. Various substrate trans-

port systems not present in growth medium have been found to

∗

Corresponding author at: Institute of Groundwater Ecology, Helmholtz Zentrum

be expressed in carbon-limited chemostats, not only by the model

München, GmbH, Ingolstädter Landstraße 1, 85764 Neuherberg, Germany.

microorganism Escherichia coli [39] but also by the environmentally

Tel.: +49 89 3187 2560; fax: +49 89 3187 3361.

relevant toluene-degrading bacterium Aromatoleum aromaticum

E-mail address: [email protected] (R.U. Mecken-

stock). EbN1 [36].

http://dx.doi.org/10.1016/j.syapm.2014.02.005

288 S. Marozava et al. / Systematic and Applied Microbiology 37 (2014) 287–295

Chemostat experiments performed to date have been conducted Cell counting and dry weight

with fast-growing microorganisms at relatively high growth rates

−1 TM ®

(between 0.02 and 0.05 h [36,39]) where energy ﬂux was suf- Cells were enumerated with a Multisizer 3 Coulter Counter

ﬁcient to sustain the expression of many catabolic pathways. In (Beckman Coulter, CA, USA).

contrast, continuous cultures with biomass retention (retentostats)

can be considered as good candidates for mimicking environmental Proteomic analyses

conditions, since they allow the investigation of bacterial phys-

iology at extremely low growth rates. Substrate inﬂow is kept Cells were sampled several times during the course of retento-

constant while biomass increases, causing the amount of sub- stat cultivation (Supplementary Figs. S1 and S2). The procedures for

strate available per bacterial cell to decrease over time leading to protein analysis and identiﬁcation were performed as described in

extremely low growth rates with doubling times of up to one year Ref. [24].

[9,21].

In this study, the proteome of the anaerobic iron-reducing,

Statistical analysis

aromatic hydrocarbon-degrading Geobacter metallireducens was

proﬁled at low growth rates in order to examine how different car- Normalization

bon sources would be utilized under conditions close to a natural

Abundances of proteins detected in retentostat and batch exper-

state, which is important for understanding their role and func-

iments were normalized according to the procedure described in

tion during the remediation of contaminated sites [6,14,30]. Under

Ref. [24].

these conditions, G. metallireducens was expected to enter into an

energy-saving mode that would prevent the bacterium from full

Differentially expressed proteins

de-repression of catabolic pathways in contrast to E. coli cultivated

Identiﬁcation of differentially expressed proteins was carried

in chemostats [16].

out as reported in [24]. In order to identify differences speciﬁcally

caused by a certain growth rate, proteins expressed in batch cul-

Materials and methods ture and retentostats were compared to each other for a speciﬁc

substrate condition (acetate or acetate plus benzoate) via one-way

Cultivation of G. metallireducens in retentostats ANOVA with subsequent application of the t-test. False discovery

rates (FDR) were calculated as described in [24]. Two technical

G. metallireducens strain GS-15 (DSM 7210) was purchased from replicates were taken from the retentostat experiments. In order

the Deutsche Sammlung von Mikroorganismen und Zellkulturen to gain sufﬁcient power to recognize differentially abundant pro-

GmbH (DSMZ), Germany. Microorganisms were cultivated in DSMZ teins, the randomization of the protein abundances was applied

−1

Geobacter medium 579 with 1 mL L DSMZ trace element solu- 100 times within each technical sample and not between the con-

−1

tion (SL10) and 0.5 mL L DSMZ 7 vitamins solution. Cultivation of ditions. The FDR threshold was set to 2%.

G. metallireducens in retentostats was undertaken in three repli-

cate runs with 2.5 mM acetate plus 0.7 mM benzoate as mixed Determination of growth rate and biomass production rate

carbon sources, and in two runs with 5 mM acetate as a single car-

bon source. The retentostat equipment was designed and built by The growth rate was estimated according to the following equa-

the electronics and mechanics workshop of the Faculty of Earth tion:

and Life Sciences, VU University Amsterdam, the Netherlands, and r t

x( )

details of the design can be found in [21,32]. The fermenter ves- (1)

x(t)

sel had a maximum capacity of 2 L with a working volume of 1.5 L,

−1

and it was operated under anoxic conditions. A N2/CO2 mixture where is the growth rate (h ), rx(t) is the biomass production

−1

(90%:10%) was passed through a Titanium(III) citrate solution in rate (biomass unit h ) and x(t) is the biomass as a function of time.

order to remove traces of oxygen, prior to entering the reactor rx(t) and x(t) were calculated according to the following equa-

at a constant rate of 2 L per hour. The fermenter was stirred at tions [21]:

200 rpm, and medium was pumped out through a retention unit r

s −msYxsmt

r t = m Y − x e

incorporating a 0.22 ␮m pore size ﬁlter to retain biomass. The pH x( )fit s xsm 0 (2)

of the culture was maintained at 6.8 by the addition of 2 M HCl or

◦ rs rs

2 M NaOH, and the temperature was maintained at 30 C. The gas −msYxsmt x t = − − x e

x( )fit m m 0 (3)

outlet was connected to a bottle ﬁlled with water, which kept the s s

fermenter at slight overpressure in order to avoid oxygen leakage

where t is time (h), x(t) is the biomass as a function of time t, ms

into the system. The fermenter and medium reservoir were kept −1

is the maintenance energy (mmol (biomass unit h) ), Yxsm is the

dark by wrapping in aluminium foil. The fermenter was inoculated

maximum growth yield (biomass units per mmol of a limiting sub-

at 10% v/v with an exponentially growing pre-culture, and culti-

strate), r is the supply rate of the growth-limiting substrates in the

2+ s

vated under batch conditions until the Fe concentration reached −1

current study (mmol h ) that was equal to 0.390 C-mmol substrate

30–40 mM, which indicated complete carbon source consumption. −1 −1

h in acetate-limited retentostats and 0.495 C-mmol substrate h

Then, the peristaltic feed pump was switched on (medium supply

in acetate plus benzoate-limited retentostats, and x is the mea-

−1 0

rate of 50 mL h ) and the fermenter was operated in retentostat

sured biomass at time 0. Values for ms and Yxsm were taken from

mode.

a previous study on the physiology of G. metallireducens in reten-

Cultivation of G. metallireducens in batch experiments is

tostats under acetate limitation in the presence of humic acid as an

described in an accompanying paper [24]. −1

electron acceptor [21], and were −0.016 mmol (biomass unit h)

and 0.053 g biomass per mol substrate, respectively.

Analytical measurements

Monitoring of culture purity

Fe , acetate and benzoate concentrations were determined as Microscopy, culturing and molecular analysis were routinely

described in [24]. applied. Samples taken from retentostats were plated on solid agar

S. Marozava et al. / Systematic and Applied Microbiology 37 (2014) 287–295 289

medium (1.5%, w/v) employing the same medium as used in reten- were found in various COG categories, including those of cell

tostat cultivation, with acetate as a carbon source. Plates were motility, signal transduction, intracellular trafﬁcking, energy con-

incubated under oxic or anoxic conditions. The FastDNA SPIN servation, and carbohydrate transport and metabolism.

Kit for Soil (MP Biomedicals, USA) was used for DNA extraction

from 1 mL of the cell culture. The V3 region of the 16S rRNA gene Carbon metabolism

sequence was ampliﬁed using bacterial 16S rRNA primers (forward

primer F357 with a 40 bp GC clamp attached at the 5 end and Abundances of all proteins in the ethanol utilization pathway

reverse primer R518) [28]. Denaturing gradient gel electrophore- were signiﬁcantly increased in response to low growth rates (Sup-

sis (DGGE) was performed with a Bio-Rad DCode (Hercules, CA) plementary Fig. S6 and Fig. 4). In contrast, only a few proteins from

detection system. The PCR products were loaded onto 1 mm thick other catabolic pathways, such as acetate-, pyruvate-, butyrate-

polyacrylamide gels containing a 30–55% linear denaturant gradi- , benzoate-, phenol-, p-cresol-, and p-hydroxybutyrate-degrading

ent. Electrophoresis was performed at a constant voltage of 200 V pathways, signiﬁcantly increased in abundance under carbon lim-

◦

for 200 min in 1× TAE running buffer at 60 C. After electrophoresis, itation (Fig. 4). The highest-fold change among catabolic proteins

the gels were stained with ethidium bromide buffer and photo- (up to 1470) at low compared to high growth rates was observed

graphed under UV light using a Kodak EDAS 290 camera. for the two proteins of ethanol degradation: the iron-containing

ethanol dehydrogenase (Gmet 1046) and the tungsten-containing

aldehyde ferredoxin oxidoreductase (Gmet 1045) (Supplementary

Results

Tables S2 and S3).

Low growth rates were also characterized by the induction of

Growth of G. metallireducens in carbon-limited retentostats

multiple pathways for acetate and pyruvate metabolism, indicat-

ing metabolic redundancy [33,35]. For example, proteins initiating

Growth of G. metallireducens in acetate-limited retentostats was

acetate consumption, acetate kinase (Gmet 1034) and the ATP-

characterized by acetate concentrations below the detection limit

consuming acetate-CoA ligase (Gmet 2340) were more abundant

(residual acetate of <0.1 mM; Fig. 1 and Supplementary Fig. S1). The

−1 at low growth rates in the acetate-limited retentostats compared to

lowest growth rates obtained (0.0021–0.0024 h ) were 30 times

−1 batch culture (Fig. 4). Increased demand for pyruvate at low growth

lower than in batch culture (0.063 h ) [24].

rates was reﬂected in expression of phosphoenolpyruvate synthase

In acetate plus benzoate-limited retentostats, acetate (2.5 mM)

(Gmet 0770) and the acetyl-CoA synthesizing component of pyru-

and benzoate (0.7 mM) were supplied in approximately equal car-

vate dehydrogenase (Gmet 2510), as well as increased abundance

bon content (5.0 and 4.9 mmol C, respectively). In contrast to batch

of the ATP-consuming pyruvate carboxylase (Gmet 0816) (Fig. 4).

experiments [24], acetate and benzoate were consumed simulta-

During growth in the retentostats with acetate plus benzoate,

neously. Two replicates (run 1 and run 2) had residual benzoate

the benzoyl-CoA pathway was the only aromatic compound-

and one replicate (run 3) had residual acetate in the outﬂow (Sup-

degrading pathway expressed. In contrast to high growth rates,

plementary Fig. S2). The total residual C-substrate concentrations

benzoyl-CoA ligase (Gmet 2143) was not detected in retentostats,

were, however, approximately the same between the replicates

−1 while succinyl:benzoate coenzyme A transferase (Gmet 2054) was

(0.7–1.0 carbon mmol L ) (Supplementary Fig. S2). The lowest

highly abundant (Supplementary Table S1). Proteins degrading

growth rates estimated for acetate plus benzoate-limited reten-

toluene were only detected at high growth rates (Supplementary

tostats were similar to the lowest growth rates estimated for

−1 Table S1 and Fig. 4).

acetate-limited retentostats: 0.0021–0.0025 h (Fig. 1 and Sup-

Proteins from fatty acid metabolism detected solely at low

plementary Fig. S2).

growth rates were related to the catabolic processes (Gmet 0144,

Microscopic examination of cells cultivated in the acetate plus

Gmet 3302, Gmet 0790), while most of the proteins detected only

benzoate-limited retentostats revealed different morphologies at

at high growth rates were related to the anabolic pathway of fatty

the beginning and end of cultivation, suggesting that bacterial cells

acid biosynthesis (Gmet 0115, Gmet 2892, Gmet 0256) (Fig. 4).

did not divide completely at extremely low growth rates (Sup-

Several proteins of the central TCA cycle had insigniﬁcantly

plementary Fig. S3). Examination of culture purity by plating and

decreased abundances at low growth rates (Supplementary Table

molecular analysis (Supplementary Fig. S4) did not reveal contam-

S1). Phosphoenolpyruvate carboxykinase (Gmet 3169) and several

ination during the course of retentostat cultivation.

other proteins of gluconeogenesis had increased abundances at low

growth rates (Fig. 4).

Proteome of G. metallireducens at low (retentostat) and high

(batch) growth rates Electron transport

The total number of proteins detected in the cells grown in Many proteins related to electron transport were detected

retentostats and in batch culture was similar (Supplementary Table at low growth rates only or had increased abundances com-

S1). The impact of growth rate (retentostat versus batch) was pared to high growth rates (Fig. 5). More abundant were, for

examined by one-way ANOVA analysis of protein abundances cor- example, proteins related to nitrate reduction (e.g. Gmet 1020-

responding to a certain growth condition (growth with a single 1021), nickel-dependent hydrogenase (Gmet 3330-3331), and

carbon source (acetate) or growth with two carbon sources (acetate various cytochromes (e.g. Gmet 1094, Gmet 0581, Gmet 0557,

plus benzoate). This analysis revealed that 129 and 118 proteins Gmet 3091) (Fig. 5). Some electron-transferring proteins highly

were differentially abundant on acetate (Supplementary Table S2) abundant at low growth rates were also related to protection

and acetate plus benzoate (Supplementary Table S3), respectively. against oxidative stress (e.g. Gmet 3330 [38], Gmet 1930 [27], etc.)

A total of 39 proteins were detected as differentially abundant in (Fig. 5).

both growth conditions.

The distribution of proteins over clusters of orthologous groups Other changes in physiology in response to low growth rates

(COGs) protein categories was highly similar at high and low

growth rates (Supplementary Fig. S5). Proteins with signiﬁcantly At low growth rates, the number of signiﬁcantly abundant

higher abundances during retentostat cultivation compared to proteins related to cell motility, intracellular trafﬁcking, secretion

batch culture (Fig. 2), or detected only at low growth rates (Fig. 3), and vesicular transport, as well as defense mechanisms, was

290 S. Marozava et al. / Systematic and Applied Microbiology 37 (2014) 287–295

Fig. 1. Growth of G. metallireducens in anoxic, acetate- (A1, A2) and acetate plus benzoate- (B1, B2) limited retentostats with Fe(III) citrate as an electron acceptor. ()

Measured cell numbers in retentostats, ( ) fitted biomass x(t)fit, ( ) Fe , (᭹) acetate and () benzoate concentrations in the filtrate. (- - -) indicates the sampling

points for proteomic analysis and the respective doubling times (td(h)) at the time of sampling. Data on all retentostat replicates can be found in the supplementary material

(Figs. S1 and S2).

increased compared to those of high growth rates (Figs. 2 and 3). The majority of low growth rate-speciﬁc transporters were

The majority of signal transduction proteins detected only at low related to the transport of metals (e.g. Gmet 1547-1548,

growth rates belonged to histidine kinases. Two proteins that Gmet 0134) and unknown substrates (Fig. 5). In contrast, proteins

might be related to the production of cyclic-di-GMP and bioﬁlm detected only at high growth rates were annotated as involved

formation (Gmet 0700 and Gmet 0742) [22], as well as a protein in the transport of phosphate (e.g. Gmet 2701 and Gmet 2704),

related to adhesion to Fe(III) oxide (Gmet 0556) [34], were also vitamin B12 (Gmet 2735), potassium (Gmet 0063), and lipids

detected only at low growth rates (Fig. 5). (Gmet 2349) (Fig. 5).

Fig. 2. Distribution of differentially expressed proteins (Distribution is indicated for all differentially expressed proteins identiﬁed by ANOVA (see Supplementary Tables S1

and S2). They have a ratio of average protein abundances detected in retentostats at sampling times t0, t1, t2, or t3 compared to average protein abundances detected in batch

culture with corresponding substrates (acetate or acetate plus benzoate) of >5 (positive values) or <0.2 (negative values)) over COG categories. (A) Acetate plus benzoate- and

−1 −1

(B) acetate-limited retentostats. Samples were taken at t0, the beginning of retentostat cultivation, at t1, growth rates of 0.0043–0.0058 h and 0.0036–0.0043 h in acetate-

−1 −1

and acetate plus benzoate-limited retentostats, respectively, at t2, growth rates of 0.0018–0.0022 h and 0.0028–0.0031 h in acetate- and acetate plus benzoate-limited

−1

retentostats, respectively, and at t3, growth rates of 0.0019–0.0020 h in acetate plus benzoate-limited retentostats.

S. Marozava et al. / Systematic and Applied Microbiology 37 (2014) 287–295 291

Fig. 3. Distribution of growth-speciﬁc proteins (Growth-speciﬁc proteins were considered to be the proteins detected only at low (positive values) or high (negative values)

growth rates) over COG categories. (A) Acetate plus benzoate- and (B) acetate-limited retentostats. Samples were taken at t0, the beginning of retentostat cultivation, at t1,

−1 −1 −1

growth rates of 0.0043–0.0058 h and 0.0036–0.0043 h in acetate- and acetate plus benzoate-limited retentostats, respectively, at t2, growth rates of 0.0018–0.0022 h

−1 −1

and 0.0028–0.0031 h in acetate- and acetate plus benzoate-limited retentostats, respectively, and at t3, growth rates of 0.0019–0.0020 h in acetate plus benzoate-limited

retentostats.

Discussion increase in the uptake capacity of limiting substrates, and expres-

sion of enzymes degrading alternative carbon substrates (not

Under carbon-limiting conditions in retentostats, microbial currently present in the medium) [11], while the response to

physiology is a reﬂection of both carbon limitation and slow slow growth would be independent from the source of limitation

growth. The response to carbon limitation is expected to be char- (e.g. decrease in protein and DNA synthesis and increase in signal

acterized by a relief from carbon catabolite repression (CCR), transduction).

Fig. 4. Catabolic proteins with differential abundances at low vs. high growth rates. Differentially expressed proteins were identiﬁed by ANOVA (false discovery rates <2%)

and also comprised proteins detected at low or high growth rates only. The protein annotation can be found in the text and in the supplementary material (Tables S2 and S3).

292 S. Marozava et al. / Systematic and Applied Microbiology 37 (2014) 287–295

S. Marozava et al. / Systematic and Applied Microbiology 37 (2014) 287–295 293

Carbon limitation-speciﬁc physiology production of acetyl-CoA from pyruvate might be expressed at low

growth rates in G. metallireducens.

Relief from CCR

In G. metallireducens, the anaerobic degradation of acetate and

Growth rate-speciﬁc physiology

benzoate proceeds through distinct peripheral catabolic pathways

that might lead to sequential utilization of these substrates. As

In bacteria, reduced growth rates induce broad resistance to var-

already shown [35], benzoate gives higher biomass yield per mol

ious environmental stresses [12,19]. Similarly, our study showed

of substrate consumed and could be of primary interest for the

that besides speciﬁc metabolic responses to carbon limitation, low

organism. However, in an accompanying paper [24], we have

growth rates triggered a general stress response in G. metallire-

shown that acetate is preferred over benzoate during growth under

ducens. This was reﬂected by increased abundances of enzymes

carbon excess. In contrast, at low growth rates in retentostats,

related to the utilization of alternative electron acceptors, signal

G. metallireducens was able to co-utilize acetate and benzoate.

transduction and protection against oxidative stress (Fig. 5). In G.

Simultaneous consumption of aromatic compounds and easily

sulfurreducens, some of these homologous proteins were also found

degradable substrates (e.g. acetate) in carbon-limited chemostats

to belong to the regulon of sigma RpoS, a master regulator of the

has also been reported for the aerobic microorganism Ralsto-

general stress response [29]. Such physiology is advantageous for

nia pickettii PKO1 [4] and Pseudomonas sp. [5,31]. Therefore, in

Geobactereaceae to survive in natural habitats with energy limita-

carbon-limited natural environments with traces of aromatic con-

tion, heavy metals, or oxidative stress.

taminants, microorganisms may generally use several substrates

simultaneously.

Readiness to utilize alternative electron acceptors

Induced expression of enzymes related to reduction of nitrate,

Fe(III), Mn(IV), and humic acids (Fig. 5) in response to low growth

Readiness to utilize alternative carbon substrates

rates agrees with previous studies on Geobactereaceae cultivated

In contrast to previous studies on microbial physiology under

in AQDS- or carbon-limited retentostats [21], insoluble Fe(III)

carbon limitation, where chemostats were run at growth rates

−

1 oxides in batch culture [1] or sediments [15]. In addition, at low

of 0.01 h and above [2,3,16,36,39], the current study inves-

growth rates, G. metallireducens seemed to increase its iron reduc-

tigated the physiology of G. metallireducens at growth rates

−

1 tion capabilities by increasing the abundance of ﬂagella and pili

below 0.003 h , which might be more relevant for carbon-

biogenesis-related proteins, which are involved in iron reduction

limited environments [21]. It was expected that such low growth

[23,37].

rates might induce an energy-saving mode in G. metallireducens

and prevent expression of many alternative catabolic path-

ways when their substrates are not readily available. Instead,

Readiness to withstand oxidative stress

G. metallireducens expressed key proteins related to utiliza-

It has been shown previously that Geobacter species can be

tion of nearly 50% of the carbon sources (Fig. 4) it could

oxidatively stressed in batch incubations with insoluble oxides of

use [23]. During carbon limitation, G. metallireducens seemed

Fe(III) and Mn(IV) [1], in chemostats with soluble Fe(III) citrate

to employ a strategy where some preferred metabolic path-

[25,26], and in natural sediments containing Fe(III) oxides [15].

ways were de-repressed while others needed speciﬁc induction

The current study also revealed an oxidative stress response of G.

(e.g. the toluene-degrading pathway). Thus, G. metallireducens

metallireducens in retentostats with soluble Fe(III) citrate. The pro-

appears to be adapted to typical anoxic habitats where pri-

teomic data suggested that, at low growth rates, G. metallireducens

mary fermenting microorganisms deplete sugars and proteins

was prepared for exposure to superoxide radicals (e.g. increased

[10], and release alcohols and fatty acids as fermentation prod-

abundances of cytochrome c (Gmet 0115) and cytochrome bd

ucts.

(Gmet 1930), as well as ferredoxin (Gmet 0148)) [26], and hydro-

gen peroxide (increase in abundance of cytochrome c catalase

(Gmet 1094) [1]).

Central metabolism

The reduced ﬂux of metabolites during slow growth on a

Increase in motility and signal transduction

limiting carbon source might be achieved with lower abundances

Geobacter spp. are considered to possess the highest number of

of the enzymes belonging to central catabolic pathways. Simi-

encoded signal transduction proteins among microorganisms with

lar to A. aromaticum cultivated under carbon limitation [36], G.

completely sequenced genomes [8]. This is related to the versatility

metallireducens had decreased abundances of most of its TCA

proteins. of Geobacter spp. to utilize many electron acceptors in the environ-

ment [8]. Therefore, increased sensing and motility, as reﬂected

Nevertheless, the increase in the abundance of the proteins that

in higher abundances or de novo formation of corresponding pro-

carry out anaplerotic reactions (pyruvate carboxylase (Gmet 0816)

teins at low growth rates (Figs. 2 and 3), might indicate that G.

and a component of pyruvate dehydrogenase (Gmet 2510)) (Fig. 4)

metallireducens is prepared to utilize alternative electron acceptors

suggested a requirement to reﬁll the TCA cycle with oxaloacetate,

simultaneously, as well as search for new habitats.

as well as an increased readiness to utilize pyruvate during carbon

limitation. Segura et al. [33] suggested that pyruvate degradation

via pyruvate dehydrogenase is not functional in G. sulfurreducens Regulation of catabolic pathways in G. metallireducens

during exponential growth under unlimited conditions in batch

culture. However, the detection of a subunit of the putative E1 In conclusion, we propose a model for regulation of catabolic

protein of the pyruvate dehydrogenase complex (Gmet 2510) pathways by G. metallireducens at high and low substrate concen-

at low growth rates suggests that this alternative pathway for trations (Fig. 6).

Fig. 5. Abundances of selected differentially expressed proteins at low (retentostats) and high (batch) growth rates. Growth conditions: A – acetate, AB – acetate plus

benzoate. Average abundances of biological replicates are presented. t0, t1, t2, and t3 correspond to the sampling points during retentostat cultivation (see Supplementary

Tables S2 and S3).

294 S. Marozava et al. / Systematic and Applied Microbiology 37 (2014) 287–295

Fig. 6. Schematic representation of the proposed regulation of catabolic pathways in G. metallireducens at high (A) and low (B) substrate concentrations. Thick blue arrows

represent simpliﬁed degradation pathways. Open arrows indicate the positive (+) or negative (−) effect on expression of degradation pathways by a corresponding substrate.

(For interpretation of the references to color in this ﬁgure legend, the reader is referred to the web version of the article.)

In batch culture, when G. metallireducens is exposed to high for their excellent technical assistance, and Dr. Housna Mouttaki for

substrate concentrations and performs high growth rates, the uti- her valuable comments. Jana Seifert and Martin von Bergen were

lization of aromatic compounds is repressed by acetate and ethanol partially supported by DFG SPP1319.

via CCR (Fig. 6A) [24]. Although aromatic compounds contain more

energy per molecule than short-chain fatty acids and alcohols, Appendix A. Supplementary data

it is more advantageous for G. metallireducens to use acetate or

ethanol as they are degraded faster [24]. Degradation pathways Supplementary data associated with this article can be found,

of aromatics such as toluene, benzoate, phenol, and p-cresol are in the online version, at http://dx.doi.org/10.1016/j.syapm.2014.

co-expressed on toluene because they converge at the level of 02.005.

benzoyl-CoA (Fig. 6A). Butyrate induces expression of the benzoyl-

CoA degradation pathway via an unknown mechanism, which References

could also be related to the joining of these two pathways at the

[1] Aklujkar, M., Coppi, M.V., Leang, C., Kim, B.-C., Chavan, M.A., Per-

level of butyrate-derived metabolic intermediates (e.g. crotonyl-

petua, L.A., Giloteaux, L., Liu, A., Holmes, D. (2013) Proteins involved

CoA that subsequently undergoes fatty acid ␤-oxidation [24]).

in electron transfer to Fe(III) and Mn(IV) oxides by Geobacter sul-

Central metabolic pathways, such as the TCA cycle, are character- furreducens and Geobacter uraniireducens. Microbiology 159, 515–535,

http://dx.doi.org/10.1099/mic.0.064089-0.

ized by regulation through changes in metabolite concentrations

[2] Berney, M., Cook, G.M. (2010) Unique ﬂexibility in energy metabolism

rather than gene expression (Fig. 6A).

allows mycobacteria to combat starvation and hypoxia. PLoS One 5, e8614,

Under environmentally relevant scenarios with acetate- or http://dx.doi.org/10.1371/journal.pone.0008614.

acetate plus benzoate-limited retentostats, G. metallireducens [3] Bobadilla Fazzini, R.A., Bielecka, A., Quintas, A.K.P., Golyshin, P.N., Preto, M.J.,

Timmis, K.N., dos Santos, V.A.P.M. (2009) Bacterial consortium proteomics

de-represses pathways involved in degradation of benzoyl-CoA

under 4-chlorosalicylate carbon-limiting conditions. Proteomics 9, 2273–2285.

and fermentation products such as ethanol, acetate, and butyrate

[4] Bucheli-Witschel, M., Hafner, T., Ruegg, I., Egli, T. (2009) Benzene degradation

(Fig. 6B). Expression of proteins related to activation of aromatic by Ralstonia pickettii PKO1 in the presence of the alternative substrate succinate.

Biodegradation 20, 419–431.

compounds requires the presence of the respective substrates (spe-

[5] Duetz, W.A., Marques, S., de Jong, C., Ramos, J.L., van Andel, J.G. (1994) Inducibil-

ciﬁc induction, Fig. 6B).

ity of the TOL catabolic pathway in Pseudomonas putida (pWW0) growing on

The proposed model suggests optimized consumption of carbon succinate in continuous culture: evidence of carbon catabolite repression con-

trol. J. Bacteriol. 176, 2354–2361.

sources by G. metallireducens in the environment. At high substrate

[6] Fang, Y., Wilkins, M.J., Yabusaki, S.B., Lipton, M.S., Long, P.E. (2012) Evaluation

concentrations, it prefers easily degradable compounds to gain

of a genome-scale in silico metabolic model for Geobacter metallireducens by

energy for fast growth, while under carbon limitation the spectrum using proteomic data from a ﬁeld biostimulation experiment. Appl. Environ.

Microbiol. 78, 8735–8742.

of various de-repressed catabolic enzymes enables quick responses

[7] Fredrickson, J.K., Fletcher, M. 2001 Subsurface microbiology and biogeochem-

to changes in the substrate composition of habitats. Readiness to

istry, Wiley-Liss, Inc., USA, pp. , 39–71.

express the benzoyl-CoA-degrading pathway suggests that under [8] Galperin, M.Y. (2005) A census of membrane-bound and intracellular signal

transduction proteins in bacteria: bacterial IQ, extroverts and introverts. BMC

oligotrophic conditions G. metallireducens co-consumes aromatic

Microbiol. 5, 1–19.

compounds and fermentation products.

[9] Gofﬁn, P., van de Bunt, B., Giovane, M., Leveau, J.H.J., Hoppener-Ogawa, S.,

Teusink, B., Hugenholtz, J. (2010) Understanding the physiology of Lactobacillus

plantarum at zero growth. Mol. Syst. Biol. 6, 413.

Acknowledgments [10] Hallbeck, L., Pedersen, K. (2008) Characterization of microbial processes in deep

aquifers of the Fennoscandian Shield. Appl. Geochem. 23, 1796–1819.

[11] Harder, W., Dijkhuizen, L. (1983) Physiological responses to nutrient limitation.

This work was supported by the EU initial training network

Annu. Rev. Microbiol. 37, 1–23.

(ITN) project “Goodwater” (Grant Agreement Number 212683). We [12] Hengge-Aronis, R. (2002) Recent insights into the general stress response reg-

ulatory network in Escherichia coli. J. Mol. Microbiol. Biotechnol. 4, 341–346.

thank Kathleen Eismann, Christine Schumann and Martin Braster

S. Marozava et al. / Systematic and Applied Microbiology 37 (2014) 287–295 295

[13] Hoehler, T.M., Jorgensen, B.B. (2013) Microbial life under extreme energy lim- [26] Mouser, P.J., Holmes, D.E., Perpetua, L.A., DiDonato, R., Postier, B., Liu, A., Lovley,

itation. Nat. Rev. Microbiol. 11, 83–94. D.R. (2009) Quantifying expression of Geobacter spp. oxidative stress genes in

[14] Holmes, D.E., Giloteaux, L., Barlett, M., Chavan, M.A., Smith, J.A., Williams, pure culture and during in situ uranium bioremediation. ISME J. 3, 454–465.

K.H., Wilkins, M., Long, P., Lovley, D.R. (2013) Molecular analysis of the in situ [27] Muller, M.M., Webster, R.E. (1997) Characterization of the tol-pal and cyd region

growth rates of subsurface Geobacter species. Appl. Environ. Microbiol. 79, of Escherichia coli K-12: Transcript analysis and identiﬁcation of two new pro-

1646–1653. teins encoded by the cyd operon. J. Bacteriol. 179, 2077–2080.

[15] Holmes, D.E., O’Neil, R.A., Chavan, M.A., N’Guessan, L.A., Vrionis, H.A., Perpetua, [28] Muyzer, G., de Waal, E.C., Uitterlinden, A.G. (1993) Proﬁling of complex

L.A., Larrahondo, M.J., DiDonato, R., Liu, A., Lovley, D.R. (2009) Transcriptome of microbial populations by denaturing gradient gel electrophoresis analysis of

Geobacter uraniireducens growing in uranium-contaminated subsurface sedi- polymerase chain reaction-ampliﬁed genes coding for 16S rRNA. Appl. Environ.

ments. ISME J. 3, 216–230. Microbiol. 59, 695–700.

[16] Ihssen, J., Egli, T. (2005) Global physiological analysis of carbon- and energy- [29] Núnez,˜ C., Esteve-Núnez,˜ A., Giometti, C., Tollaksen, S., Khare, T., Lin, W., Lovley,

limited growing Escherichia coli conﬁrms a high degree of catabolic ﬂexibility D.R., Methé, B.A. (2006) DNA microarray and proteomic analyses of the RpoS

and preparedness for mixed substrate utilization. Environ. Microbiol. 7, regulon in Geobacter sulfurreducens. J. Bacteriol. 188, 2792–2800.

1568–1581. [30] Rooney-Varga, J.N., Anderson, R.T., Fraga, J.L., Ringelberg, D., Lovley, D.R. (1999)

[17] Konopka, A. (2000) Microbial physiological state at low growth rate in natural Microbial communities associated with anaerobic benzene degradation in a

and engineered ecosystems. Curr. Opin. Microbiol. 3, 244–247. petroleum-contaminated aquifer. Appl. Environ. Microbiol. 65, 3056–3063.

[18] Kovarova-Kovar, K., Egli, T. (1998) Growth kinetics of suspended microbial [31] Rueegg, I., Hafner, T., Bucheli-Witschel, M., Egli, T. (2007) Dynamics of ben-

cells: from single-substrate-controlled growth to mixed-substrate kinetics. zene and toluene degradation in Pseudomonas putida F1 in the presence of the

Microbiol. Mol. Biol. Rev. 62, 646–666. alternative substrate succinate. Eng. Life Sci. 7, 331–342.

[19] Kube, M., Heider, J., Amann, J., Hufnagel, P., Kuhner, S., Beck, A., Reinhardt, R., [32] Schrickx, J.M., Raedts, M.J., Stouthamer, A.H., van Verseveld, H.W. (1995)

Rabus, R. (2004) Genes involved in the anaerobic degradation of toluene in a Organic acid production by Aspergillus niger in recycling culture analyzed by

denitrifying bacterium, strain EbN1. Arch. Microbiol. 181, 182–194. capillary electrophoresis. Anal. Biochem. 231, 175–181.

[20] Langwaldt, J.H., Munster, U., Puhakka, J.A. (2005) Characterization and micro- [33] Segura, D., Mahadevan, R., Juarez, K., Lovley, D.R. (2008) Computational and

bial utilization of dissolved organic carbon in groundwater contaminated with experimental analysis of redundancy in the central metabolism of Geobacter

chlorophenols. Chemosphere 59, 983–996. sulfurreducens. PLoS Comput. Biol. 4, e36.

[21] Lin, B., Westerhoff, H.V., Roling, W.F.M. (2009) How Geobacteraceae may [34] Smith, J.A., Lovley, D.R., Tremblay, P.-L. (2012) Outer cell surface components

dominate subsurface biodegradation: physiology of Geobacter metallire- essential for Fe(III) oxide reduction by Geobacter metallireducens. Appl. Environ.

ducens in slow-growth habitat-simulating retentostats. Environ. Microbiol. 11, Microbiol. 79, 901–907.

2425–2433. [35] Sun, J., Sayyar, B., Butler, J.E., Pharkya, P., Fahland, T.R., Famili, I., Schilling, C.H.,

[22] Liu, N., Xu, Y., Hossain, S., Huang, N., Coursolle, D., Gralnick, J.A., Boon, E.M. Lovley, D.R., Mahadevan, R. (2009) Genome-scale constraint-based modeling

(2012) Nitric oxide regulation of cyclic di-GMP synthesis and hydrolysis in of Geobacter metallireducens. BMC Syst. Biol. 3, 174.

Shewanella woodyi. Biochemistry 51, 2087–2099. [36] Trautwein, K., Lahme, S., Woehlbrand, L., Feenders, C., Mangelsdorf, K., Harder,

[23] Lovley, D.R., Ueki, T., Zhang, T., Malvankar, N.S., Shrestha, P.M., Flanagan, K.A., J., Steinbuechel, A., Blasius, B., Reinhardt, R., Rabus, R. (2012) Physiological and

Aklujkar, M., Butler, J.E., Giloteaux, L., Rotaru, A.-E., Holmes, D.E., Franks, A.E., proteomic adaptation of Aromatoleum aromaticum EbN1 to low growth rates

Orellana, R., Risso, C., Nevin, K.P. (2011) Geobacter The microbe electric’s phys- in benzoate-limited, anoxic chemostats. J. Bacteriol. 194, 2165–2180.

iology, ecology, and practical applications. In: Poole, R.K. (Ed.),. Adv. Microb. [37] Tremblay, P.-L., Aklujkar, M., Leang, C., Nevin, K.P., Lovley, D. (2012) A genetic

Physiol., 1–100. system for Geobacter metallireducens: role of the ﬂagellin and pilin in the reduc-

[24] Marozava, S., Röling, W.F.M., Seifert, J., Köffner, R., von Bergen, M., Meckenstock, tion of Fe(III) oxide. Environ. Microbiol. Rep. 4, 82–88.

R.U. (2014) Physiology of Geobacter metallireducens under excess and limitation [38] Tremblay, P.-L., Lovley, D.R. (2012) Role of the NiFe hydrogenase Hya in oxida-

of electron donors. Part I. Batch cultivation with excess of carbon sources. Syst. tive stress defense in Geobacter sulfurreducens. J. Bacteriol. 194, 2248–2253.

Appl. Microbiol (accepted for publication). [39] Wick, L.M., Quadroni, M., Egli, T. (2001) Short- and long-term changes in pro-

[25] Methe, B.A., Webster, J., Nevin, K., Butler, J., Lovley, D.R. (2005) DNA microarray teome composition and kinetic properties in a culture of Escherichia coli during

analysis of nitrogen ﬁxation and Fe(III) reduction in Geobacter sulfurreducens. transition from glucose-excess to glucose-limited growth conditions in contin-

Appl. Environ. Microbiol. 71, 2530–2538. uous culture and vice versa. Environ. Microbiol. 3, 588–599.