OUP user menu

Members of the genus Arthrobacter grow anaerobically using nitrate ammonification and fermentative processes: anaerobic adaptation of aerobic bacteria abundant in soil

Martin Eschbach, Henrik Möbitz, Alexandra Rompf, Dieter Jahn
DOI: http://dx.doi.org/10.1016/S0378-1097(03)00383-5 227-230 First published online: 1 June 2003


Members of the genus Arthrobacter are usually regarded as obligate aerobic bacteria. The anaerobic growth and energy metabolism of two Arthrobacter species were investigated. Arthrobacter globiformis utilized both nitrate ammonification and lactate, acetate and ethanol producing fermentation processes for anaerobic growth. Only nitrate supported anaerobic growth of Arthrobacter nicotianae. Anaerobically induced respiratory nitrate reductase activity was detected in both strains. Neither of the tested strains used the alternative electron acceptors fumarate, dimethylsulfoxide or trimethylamine-N-oxide.

  • Anaerobic metabolism
  • Nitrate reductase
  • Nitrate respiration
  • Fermentation
  • Arthrobacter

1 Introduction

The type species Arthrobacter globiformis of the genus Arthrobacter was initially described by Conn in 1928 as bacteria numerous in various soils with the unusual morphological change from Gram-negative rods in young cultures to Gram-positive cocci in older cultures [1]. On the basis of 16S rRNA cataloging the genus Arthrobacter was identified as a member of the GC rich ‘actinomycete’ branch of Gram-positive bacteria, phylogenetically closely related to other coryneform genera such as Aureobacterium, Cellulomonas, Curtobacterium and Microbacterium [2,3]. The two major ‘species groups’A. globiformis/Arthrobacter citreus and Arthrobacter nicotianae differ in their peptidoglycan structure, teichoic acid content and lipid composition [4]. As originally reported by Conn [1], many groups have shown that the genus Arthrobacter represents a numerically important fraction of the indigenous bacterial flora of soils from all parts of the world. Often they are the most numerous single bacterial group in aerobic plate counts [58]. Arthrobacters can use a wide and diverse range of organic substances as carbon and energy sources including nicotine, nucleic acids and various herbicides and pesticides [4,9]. Like some Bacilli and Rhodococci some Arthrobacter species belong to the group of Gram-positive organic solvent-tolerant bacterial strains [10]. Recently, some Arthrobacter strains with clinical relevance have been isolated [11,12]. Members of the genus Arthrobacter are usually regarded as obligate aerobes with a pure respiratory, never a fermentative, mode of metabolism [4].

Changes in oxygen tension occur frequently in the upper layers of soil, the habitat of the genus Arthrobacter. In order to survive periods of oxygen limitation some Arthrobacter species have developed alternative, oxygen-independent growth strategies. Here, we describe a first initial investigation of the anaerobic growth behavior and its biochemical basis for two members of the genus Arthrobacter.

2 Materials and methods

2.1 Bacterial strains and growth conditions

The type strains A. globiformis (DSM 20124) and A. nicotianae (DSM 20123) were obtained from the Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH, Braunschweig, Germany. All strains were grown at 30°C in Corynebacteria medium containing 1% (w/v) casein hydrolysate (Gibco, Karlsruhe, Germany), 0.5% (w/v) yeast extract (Difco, Detroit, MI, USA) and 0.5% (w/v) NaCl. 25 mM glucose and 10 mM pyruvate were added where indicated. Anaerobic growth was performed in completely filled flasks (<1% gas phase) with rubber stoppers and shaking at 100 rpm to minimize aggregation of bacteria. The media were supplemented with nitrate, fumarate, dimethylsulfoxide (DMSO) and trimethylamine-N-oxide (TMAO) in a final concentration of 10 mM where indicated. Nitrite was added at concentrations of 1 mM or 5 mM where indicated. The media were not completely anaerobic in order to allow an initial and vital adaptation phase. After approximately 20 min of incubation all oxygen was consumed as determined by oxygen electrode measurements and followed by redox indicators. At that point strictly anaerobic growth conditions were achieved [13].

2.2 Preparation of cell-free extracts and enzyme assays

Bacteria from the mid-exponential growth phase (OD578 of 0.2–0.8) were sedimented by centrifugation at 5000×g and washed with buffer, resuspended in 50 mM Tris, pH 7.4, 10 mM MgCl2, 100 mg l−1 DNase, 50 mg l−1 RNase and passed several times through a French press. The crude extract was centrifuged at 15 000×g and the supernatant was used for enzyme assays. All operations were carried out under strict oxygen exclusion.

Nitrate reductase activity in cell-free extracts was measured photometrically by the nitrate-dependent oxidation of reduced benzyl viologen in anoxic cuvettes [13]. One unit of activity corresponds to the reduction of 1 µmol nitrate per minute. Reported values are the average of three independent experiments performed in triplicate. Nitrate, nitrite, ammonia and acetoin concentrations were determined as described before [13].

2.3 High performance liquid chromatography (HPLC) analysis of fermentation products

Fermentation products were determined from the supernatant of the cultures after removal of the bacteria by centrifugation [14].

3 Results and discussion

3.1 The anaerobic energy metabolism of A. globiformis and A. nicotianae

Anaerobic growth experiments in the presence and absence of nitrate were performed using two Arthrobacter strains which represent the two major ‘species groups’A. globiformis/A. citreus and A. nicotianae [4]. The bacteria were first grown aerobically and subsequently shifted to anaerobic growth conditions. Similar to observations made for Bacillus subtilis a transition period with continuously decreasing oxygen tension is required to allow subsequent anaerobic growth [15]. The shift from aerobic to strictly anaerobic conditions resulted in a growth lag phase of approximately 12–24 h for both tested Arthrobacter strains (data not shown) [15]. Oxygen electrode measurements demonstrated that all detectable oxygen had been utilized approximately 20 min after the shift. Nevertheless, A. globiformis and A. nicotianae showed significant anaerobic growth which was stimulated by the presence of nitrate in the medium (Fig. 1a,b). In agreement with these findings complete reduction of nitrate via nitrite to ammonia by both organisms was demonstrated via time-dependent analysis of the growth medium (Fig. 1c,d). A. globiformis converted most of the formed nitrite immediately into ammonia (Fig. 1c) while A. nicotianae accumulated nitrite up to levels of 5 mM prior to further reduction (Fig. 1d). No obvious signs for denitrification processes such as the development of gas or increased pressure in the anaerobic flasks during anaerobic growth with nitrate were observed. Finally, benzyl viologen-dependent nitrate reductase activity was found in cell-free extracts prepared from anaerobically grown A. globiformis and A. nicotianae. Nitrate reductase activities were detected in extracts prepared from A. globiformis grown anaerobically in the presence (107±10 U g−1 protein) and absence (92±10 U g−1 protein) of nitrate. Extracts prepared from aerobically grown A. globiformis in the presence or absence of nitrate did not contain significant nitrate reductase activities (less than 10 U g−1 protein). These results indicate an anaerobic induction of A. globiformis respiratory nitrate reductase activity independent of the presence of the utilized electron acceptor, nitrate. In contrast, significant nitrate reductase activity (45±7 U g−1 protein) was only observed in extracts prepared from A. nicotianae grown anaerobically in the presence of nitrate.

Figure 1

Aerobic growth (▲) of A. globiformis DSM 20124 (a) and A. nicotianae DSM 20123 (b). A. globiformis and A. nicotianae were grown at 30°C anaerobically in Corynebacteria medium as outlined in Section 2 using 25 mM glucose and 10 mM pyruvate as carbon source with no further additions (▵) and with the addition of 10 mM nitrate (□). Growth was followed by OD578 nm measurements in combination with viable cell counts. The nitrate (●), nitrite (■) and ammonia (♦) concentrations in the growth media during anaerobic growth of A. globiformis (c) and A. nicotianae (d) were determined.

The addition of 5 mM nitrite stimulated anaerobic growth of A. globiformis, however, after a lag phase of up to 24 h which is probably due to the toxicity of the electron acceptor (data not shown). Our current data do not allow to discriminate whether the analogs of the in Gram-negative bacteria periplasmic Nap-Nrf or the cytoplasmic NarG-NirB systems [16], or even both, function in A. globiformis. However, as discussed by Potter et al. [17] no Nap homologs have yet been found in Gram-positive bacteria.

A. globiformis also grew anaerobically in the absence of nitrate and nitrite (Fig. 1a). Anaerobic growth experiments using 25 mM glucose and 10 mM pyruvate as carbon sources demonstrated the formation of acetate, lactate and ethanol indicative of a mixed acid fermentation (Table 1). No obvious formate formation was observed. Again, similar to observations made for B. subtilis fermentative growth, only the combination of glucose with pyruvate as carbon sources yielded sufficient anaerobic growth (Fig. 1a) [13,14,18]. Glucose as sole carbon source only allowed very weak anaerobic fermentative growth of A. globiformis (data not shown). In agreement only small quantities of fermentation products were formed (1.6 mM acetate, 0.9 mM lactate and no detectable ethanol) after 48 h of anaerobic cultivation with glucose as sole carbon source. HPLC analysis demonstrated the almost stoichiometrical conversion of the utilized carbon sources into the detected fermentation products (Table 1). Similar to B. subtilis, significant fermentation product formation was also observed under anaerobic growth conditions in the presence of the alternative electron acceptor, nitrate. Nitrate was completely converted by A. globiformis into ammonia after 20 h of anaerobic growth (Fig. 1c). After 48 h all of the pyruvate and about 30% of the added glucose were consumed and converted into 12 mM lactate, 7 mM acetate and 4 mM ethanol. Based on our observations we propose a model for the mixed acid fermentation of A. globiformis which is shown in Fig. 2.

View this table:
Table 1

Fermentation product formation by A. globiformis

Electron acceptor (mM)Consumed carbon sources (mM)Concentrations of grown cells (g (wet weight) l−1)Formed products (mol mol−1 pyruvate) [total (mM)]
107104.10.50 [12]0.29 [7]0.17 [4]
260.90.50 [5]0.40 [4]0.05 [0.5] [0.9]0.53 [1.6]0 [–]
  • A. globiformis was incubated anaerobically for 48 h at 30°C in Corynebacteria medium using 10 mM nitrate as electron acceptor, 25 mM glucose and 10 mM pyruvate as carbon source where indicated. The amounts of glucose and pyruvate consumed and fermentation products formed were quantified using HPLC analysis of the growth media as described in Section 2. Formed CO2 and redox equivalent turnover as outlined in Fig. 2 have to be added to balance the fermentation equation.

  • The amount of pyruvate consumed includes the quantity of externally added pyruvate consumed with pyruvate formed from glucose. Even in experiments where only glucose was used, glucose metabolism proceeds via pyruvate. Therefore, to make the results easy to compare we related the formed products to the single C3-carbon source pyruvate.

  • Compounds utilized.

Figure 2

Proposed initial model for the A. globiformis fermentation processes. Our investigation does not exclude the presence of additional fermentative pathways.

No fermentative growth was observed for A. nicotianae. The alternative electron acceptors TMAO (10 mM), DMSO (10 mM), fumarate (10 mM) and thiosulfate (2 mM, 5 mM) did not increase the anaerobic growth of A. globiformis and A. nicotianae.

3.2 Conclusion

Not all members of the genus Arthrobacter are strict aerobes. Anaerobic cultures of A. nicotianae reduced nitrate to ammonia. A. globiformis was even more flexible since it can use nitrate as terminal electron acceptor and mixed acid fermentation under anaerobic conditions.


We thank R. Brandsch, Freiburg, Germany, for the gift of Arthrobacter strains and helpful discussions. We thank R.K. Thauer, Marburg, Germany, for continuous support. Financial support was provided by the Deutsche Forschungsgemeinschaft and the Fonds der Chemischen Industrie.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
View Abstract