Pyruvate relieves the necessity of high induction levels of catalase and enables Campylobacter jejuni to grow under fully aerobic conditions


Rab de Jonge, Laboratory for Zoonoses and Environmental Microbiology, RIVM, P.O. Box 1, 3720 BA Bilthoven, The Netherlands.


Aims:  Several cases of campylobacteriosis reported worldwide seemingly conflict with the strict growth requirements and sensitivity to environmental stress of Campylobacter jejuni. In this study, the need for a micro-aerobic environment [dissolved oxygen tension (DOT): 0·1–90%; 100% air saturation)] and the adaptive responses to oxygen stress were studied.

Methods and Results:  The growth of C. jejuni in continuous culture was assessed under different DOT in the presence or absence of pyruvate. In a medium without pyruvate, continuous cultures of C. jejuni showed typically micro-aerobic behaviour and cells were unable to grow under fully aerobic conditions. However in the presence of pyruvate (25 mmol l−1), continuous cultures of C. jejuni were able to grow in a broad DOT range, varying from 0·1% to at least 90%, and the catalase activity was decreased.

Conclusions:  Addition of pyruvate results in the decrease in the concentration of hydrogen peroxide, which enables C. jejuni to grow aerobically.

Significance and Impact of the Study:  New information on the oxidative physiology of C. jejuni and its ability to grow aerobically in media supplemented with pyruvate is presented.


Campylobacter jejuni is the leading cause of bacterial food-borne gastroenteritis throughout the world (Friedman et al. 2000; Oberhelman and Taylor 2000). Several cases of campylobacteriosis reported worldwide seemingly conflict with the strict growth requirements and the sensitivity to environmental stress of C. jejuni as described in literature (Butzler and Oosterom 1991; Altekruse et al. 1999; Murphy et al. 2006).

Currently C. jejuni is seen as an obligate micro-aerophile, which requires oxygen concentrations of 3–15% and carbon dioxide (CO2) concentrations of 3–5% for satisfactory growth, using amino acids as carbon and energy sources (Hoffman et al. 1979a; Smibert 1984; Kelly 2001; Park 2002; Sellars et al. 2002). However, it has been suggested that Campylobacter spp. can adapt to aerobic growth (Jones et al. 1993; Chynoweth et al. 1998) or can grow aerobically in 10% CO2 in moist air (Fraser et al. 1992). Campylobacter jejuni has a respiratory metabolism with a branched electron transport chain based on the use of oxygen as a terminal electron acceptor; but alternative terminal electron acceptors can also be used (Mendz et al. 1997; Sellars et al. 2002). The partial reduction of oxygen to water during microbial respiration results in toxic reactive oxygen intermediates (ROI). ROI, such as the superoxide anion (O2) and hydrogen peroxide (H2O2) are implicated in lethal damage to nucleic acids, proteins and membranes (Fridovich 1978; Imlay and Linn 1988).

For the defense against the damaging effects of oxidative stress C. jejuni can make use of a single iron co-factored superoxide dismutase (SOD) (Pesci et al. 1994; Purdy and Park 1994; Purdy et al. 1999), and of catalase (KatA) (Grant and Park 1995). SOD catalyses the conversion of superoxide to H2O2 and oxygen, and the accumulation of H2O2 is prevented by the action of KatA. The sensitivity of C. jejuni to high oxygen concentrations might be caused by high oxidative stress, whereby the capacity of the defense systems is exceeded.

Next to protective enzymes, the addition of protective components, such as blood, charcoal, FeSO4, metabisulfite and pyruvate can protect C. jejuni against oxygen stress and even enable C. jejuni to grow fully aerobically (George et al. 1978; Hoffman et al. 1979b; Bolton et al. 1984; Fraser et al. 1992; Hodge and Krieg 1994). These compounds are thought to act as quenchers of ROI, but the exact mechanism is not yet clear.

The first aim of this study was to investigate the effect of different oxygen tensions on growth yield and KatA and SOD activities of C. jejuni. More insight into these activities might help to further elucidate the toxic effect of oxygen on C. jejuni and determine why C. jejuni cannot grow in fully aerobic environments without protective components. The second aim of this study was to examine the effect of adding pyruvate on growth yield, and KatA and SOD activities of C. jejuni grown under different oxygen tensions. Mendz et al. (1997) showed that C. jejuni can use pyruvate as electron acceptor. It was examined whether pyruvate acts as a quencher of ROI as suggested in literature (Hodge and Krieg 1994), or that the protective effect of pyruvate results from its use as an alternative electron acceptor for fermentation, as a switch in the metabolism from respiration to fermentation would subsequently result in the formation of less ROI.

In most studies on the effect of oxygen, the cultivation conditions were described as the percentage oxygen in the gaseous phase. However, the dissolved oxygen tension (DOT) better reflects the availability of oxygen and the conditions at which the cells are exposed, as the DOT is not only dependent on the percentage of oxygen in the input gas, but also on the geometry of the system, the biomass concentration and its specific rate of oxygen consumption (Alexeeva et al. 2002). In this study, continuous cultures were used in which the DOT was controlled.

Materials and methods

Growth conditions

Campylobacter jejuni NCTC 11168 was obtained from the National Collection of Type Cultures (Colindale, UK). Laboratory stocks of cells were stored at −70°C in a mixture of 80% brain heart infusion broth (BHI; Difco, Franklin Lakes, New Jersey, USA) and 20% glycerol. Inocula were prepared by thawing one cryovial of which 0·5 ml was used to inoculate 100 ml of BHI in a wide-necked Erlenmeyer flask. After 24 h of culturing at 37°C in a micro-aerobic atmosphere (10% O2, 5% CO2, 85% N2) with shaking (100 rev min−1), 5 ml was used to inoculate a Bioflo fermentor (New Brunswick Scientific, Edison, New Jersey, USA; culture volume is 1 l) or a 1-l vessel controlled by the ADI 1030 controller (Applikon, Schiedam, The Netherlands). Cultures were grown at a constant dilution rate (growth rate) of 0·1 h−1 (medium supply per vessel volume) in BHI without glucose (20 g l−1; Difco) with or without 25 mmol l−1 of pyruvate (Acros Organics, Geel, Belgium). Pyruvate was sterilized by filtration (0·2 μm) and added aseptically to a medium which had been sterilized for 15 min at 121°C. A pH of 7·0 was maintained automatically using 1 mol l−1 of HCl and the temperature was maintained at 37°C. The DOT (100% air saturation) was measured using an Ingold O2 sensor (Mettler Toledo, Tiel, The Netherlands) and regulated by adjusting the agitation speed (200–600 rev min−1) and the in-going percentage of oxygen in the gas supply. The in-going gas consisted of compressed air, nitrogen, and CO2 (5%, except at high DOT conditions). Baffles were installed in the fermentor vessels to improve the oxygen transfer from the gaseous to the liquid phase. Anti-foam A (0·1 g l−1; Sigma-Aldrich, Zwijndrecht, The Netherlands) was added to the culture to prevent excessive foaming.


Campylobacter jejuni was grown in continuous culture under different DOT (0·1, 1, 10, 25, 50, 75 and 90%; 90% is maximal reachable DOT during aeration with pure compressed air as part of the oxygen is consumed by the growing bacteria) in the presence or absence of pyruvate. To reach steady-state conditions, cells were sampled 2 days after setting the DOT. The growth yield, activities of oxygen stress-protective enzymes (KatA and SOD), activities of fermentation enzymes (lactate dehydrogenase and alcohol dehydrogenase) and the metabolites used and formed were determined for all the conditions. All conditions were sampled independently in triplicate.

Purity and growth yield

The purity of samples was verified microscopically and by streak plating, using Columbia agar plates supplemented with 5% lysed horse blood (CAB; Oxoid, Basingstoke, UK), incubated for 48 h at 37°C under micro-aerobic conditions (10% O2, 5% CO2, 85% N2).

The growth yield was monitored by measuring the dry weight. A tube was dried for 48 h at 105°C and weighed. In the tube 15 ml of the culture was centrifuged (12 min, 4500 g, 4°C); the supernatant was then discarded and the pellet was resuspended in water, which was repeated twice. Following the final wash the tube with the pellet was dried for 48 h at 105°C and re-weighed. The dry weight (of 15-ml culture) was calculated as the weight of the tube with the pellet minus the weight of the tube at the start.

Preparation of cell-free cell extracts (CFE)

Approximately 40 ml of the culture was harvested by centrifugation (12 min, 4500 g, 4°C). The pellet was washed with 70 mmol l−1 of potassium phosphate-buffered saline (PBS; 9 g l−1 NaCl), pH 7·2 and centrifuged. Thereafter, the pellet was resuspended in 4 ml of PBS. The cells were disrupted by sonication (6 min, 50% duty cycle, 100 W; Branson 450 sonifier, Branson, Danbury, Connecticut, USA). The sonified suspensions were centrifuged for 10 min at 10 000 g at 4°C and the supernatant was used as CFE. KatA activity was measured directly; the remainder of the CFE was stored at −20°C for later use in other determinations.

KatA activity

The KatA activity was measured by monitoring the enzymatic breakdown of H2O2 at 240 nm as described by Beers and Sizer (1952).

SOD activity

The SOD activity of CFE was determined with an SOD Assay Kit – WST (Dojindo, Maryland, USA) as described in the manual. Technically the highly water-soluble tetrazolium salt WST-1, which produces a water-soluble formazan dye upon reduction with an O2, was used as the indicating scavenger for the superoxide generated by the xanthine oxidase reaction. The rate of the reduction with superoxide is linearly related to the xanthine oxidase activity and is inhibited by SOD. One unit of SOD activity was defined as the amount of enzyme that provides 50% of inhibition. The inhibition activity was determined by a colorimetric method.

Lactate dehydrogenase (LDH) activity

The LDH activity of CFE was measured by determining the oxidation of NADH (0·15 mmol l−1) in phosphate buffer (pH = 7·2) using pyruvate (1 mol l−1) as the electron acceptor at 340 nm, as described by Streekstra et al. (1987).

Alcohol dehydrogenase (ADH) activity

The ADH activity of CFE was measured by determining the reduction of NAD (0·15 mmol l−1) in sodium pyrophosphate buffer (12 mmol l−1) using ethanol (350 mmol l−1) as the electron donor at 340 nm, according to Clark and Cronan (1980).

Cellular protein determination

Cellular protein content of CFE was determined according to Lowry et al. (1951). Bovine serum albumin (Sigma) was used as a standard.

Determination of metabolites used and formed

HPLC analysis of supernatants was performed exactly as described by Alexeeva et al. (2003).

Statistical analysis

The KatA activity, SOD activity and dry weight, measured at different DOT, and in the presence or the absence of pyruvate, were compared by the Student’s t-test for statistically significant differences.


Growth in BHI without protecting components

In the complete DOT range of 1–75% the KatA activity increased significantly. In contrast, the SOD activity only increased significantly in the initial DOT range from 1% to 25%; when exposed to higher DOT no statistically significant differences in SOD activity were measured (Fig. 1a,b). The maximum observed KatA activity of 58 mmol H2O2 min−1 mg−1 protein was reached at a DOT of 75% (Fig. 1a). At a DOT of 25% the maximum observed SOD activity of 0·1 U mg−1 protein was reached (Fig. 1b). The growth yield of C. jejuni remained constant when the DOT was 25% or lower; when exposed to higher DOT in the range of 25–50% the growth yield diminished significantly, and at a DOT above 75% the growth rate did not match the dilution rate and consequently C. jejuni was washed out (Fig. 1c).

Figure 1.

Campylobacter jejuni 11168 culture grown in brain heart infusion broth, in the absence (black) or presence of pyruvate (25 mmol l−1; white), at dissolved oxygen tensions (DOT) varying from 0·1% to 90%: (a) the catalase activity (mmol H2O2 min−1 mg−1); (b) the superoxide dismutase (SOD) activity (U mg−1 protein); (c) the dry weight (g l−1). All determinations were performed at least in triplicate and results are means ± standard deviation. n.d., not determined; n.m., not measurable; n.g., no sufficient growth.

Growth in BHI with pyruvate added

In the presence of pyruvate, only at a DOT of 75% and higher the KatA activity was above 10 mmol H2O2 min−1 mg−1 protein, an activity already reached at a DOT of 10% when grown without pyruvate. The highest observed activity in cells grown at 90% DOT (16 mmol H2O2 min−1 mg−1 protein) was much lower than the highest observed KatA activity observed when grown without pyruvate (58 mmol H2O2 min−1 mg−1 protein) (Fig. 1a). In contrast to the large effect on KatA, pyruvate did not significantly affect the SOD activity (Fig. 1b).

Adding pyruvate allowed growth over a wide DOT range from 0·1% to at least 90% (Fig. 1c). Neither LDH, nor ADH activity was found in CFE of C. jejuni (data not shown) and no fermentation products (succinate, lactate, acetate, formate, fumarate and ethanol) were detected (data not shown) by high-performance liquid chromatography (HPLC) analysis.


So far the effect of oxygen on the activities of the oxygen-protective enzymes KatA and SOD of C. jejuni has not been studied. We expected that an increase in DOT, leading to more oxygen stress and higher ROI concentrations, would result in elevated KatA and SOD activities, enzymes which are used to breakdown ROI. The measured KatA activity matched this assumption; the higher the DOT, from 0·1% to 90%, the higher the measured KatA activity. The rise in SOD activity in the DOT ranging from 1% to 25% also suggested that SOD activity regulation as a function of the DOT. However at 25% a maximal induction of SOD was reached, where after the SOD activity stayed the same irrespective of the increase in DOT from 25% to 75%. In the absence of pyruvate, the KatA and SOD activities might not be sufficient at higher DOT to breakdown the ROI formed, resulting in the observed decrease in growth yield at DOT higher than 25% and a lesser growth rate to match the dilution rate at DOT higher than 75%. Besides the limited KatA and SOD activities, the presence of oxygen-sensitive enzymes might also account for the micro-aerobic behaviour of C. jejuni.

Next to the protective enzymes KatA and SOD, the presence of pyruvate was described to protect C. jejuni against the harmful effects of oxygen, enabling C. jejuni to grow aerobically (George et al. 1978; Bolton et al. 1984; Hodge and Krieg 1994). However, these studies were performed on plates or in flasks, in which the cultivation conditions were only given as the percentage of oxygen in the gaseous phase. In this study, the DOT was regulated in a continuous culture, which is a more appropriate measure for the availability of oxygen as it reflects better the DOT to which the bacteria are exposed, than the standard description of cultivation conditions given as percentage of oxygen in the gaseous phase (i.e. 3–21% O2). Our measurements show that C. jejuni no longer acts micro-aerobically when pyruvate is added, but can grow in a broad DOT range from 0·1% to at least 90%, during which the growth yield remains high.

This protective effect of pyruvate might be attributed to a switch in metabolism from aerobic respiration to fermentation when pyruvate is added, as pyruvate is used during fermentation rather than oxygen and no toxic ROI are formed. Campylobacter jejuni is known to possess enzymes that are capable of fermenting pyruvate to various products like lactic acid and alcohol (Mendz et al. 1997). However, our HPLC results did not show the formation of any fermentation products and also the typical fermentation enzymes, LDH and ADH, could not be detected, which excludes a switch from respiration to fermentation as a possible explanation for the protective effect of pyruvate.

Instead, our results confirm that the protective effect of pyruvate might be based on the ability of pyruvate to act as a quencher of ROI, as assumed in literature (Bolton et al. 1984; Hodge and Krieg 1994). The measured decreased KatA activity in the presence of pyruvate in our study supports the suggestion of Giandomenico et al. (1997) that chemical elimination of H2O2 by pyruvate can compete effectively with protective enzymes such as KatA. In other words, addition of pyruvate decreases the concentration of H2O2, which enables C. jejuni to grow aerobically.

In conclusion, while currently C. jejuni is seen as an obligate micro-aerophile, growing under conditions with 3–15% oxygen, this study showed that, in the presence of pyruvate, also aerobic growth of C. jejuni is possible. As amino acids can be converted to pyruvate, further research into environmental pyruvate concentrations is needed to determine if environments with correct pyruvate concentrations for growth of C. jejuni do exist. However, to serve as source of C. jejuni these environments also have to fulfil other strict growth requirements of C. jejuni, e.g. a growth temperature above 30°C.