OUP user menu

Detection of quorum sensing signals in the haloalkaliphilic archaeon Natronococcus occultus

Roberto A. Paggi, Celina B. Martone, Clay Fuqua, Rosana E. De Castro
DOI: http://dx.doi.org/10.1016/S0378-1097(03)00174-5 49-52 First published online: 1 April 2003


Bacteria communicate at high cell density through quorum sensing, however, there are no reports about this mechanism in archaea. The archaeon Natronococcus occultus produces an extracellular protease at the end of growth. Early production of protease activity was observed when a low density culture was incubated with late exponential conditioned medium suggesting the presence of factor(s) inducing this activity. Conditioned medium and ethyl acetate extracts corresponding to the transition from exponential to stationary phase showed a positive signal in Agrobacterium biosensor. We report the detection of potential autoinducer molecules of the acylated homoserine lactone type in the archaeon N. occultus. These molecules may be responsible for the production/activation of extracellular protease.

  • Quorum sensing
  • Autoinducer
  • Archaeon
  • Natronococcus occultus
  • Protease

1 Introduction

The coordinated regulation of changes in gene expression that allows the bacterial population to follow synchronized adaptive responses, or at least to modify their behavior depending on density, cellular proximity or stage of growth, has been described for Gram-negative and Gram-positive bacteria [13]. It involves the bacterial production, release and sensing of an autoinducer signal or pheromone and has been named quorum sensing [4]. Similar signaling mechanisms have not been demonstrated in archaea.

The haloalkaliphilic archaeon Natronococcus occultus produces an extracellular protease that has been purified and characterized at the biochemical level in our laboratory [5,6]. This proteolytic activity is detected in the medium in the late exponential and stationary growth phases [5] and during starvation [7]. In Western blots using antibodies raised against the extracellular protease, we have detected the precursor and active forms of the enzyme and we have observed that the total amount of protease per cell increases as the cells enter the late exponential growth phase [8]. The aim of this study was to investigate the presence of quorum sensing signals in the haloalkaliphilic archaeon N. occultus and to evaluate their potential role in extracellular protease production/activation.

2 Materials and methods

2.1 Bacterial strains and culture conditions

N. occultus NCIBM 2192 cells (National Collection of Industrial, Food and Marine Bacteria, Aberdeen, UK) were grown at 37°C in a rotary shaker at 150 rpm using a modification of the haloalkaliphilic medium (pH 10) described by Tindall et al. [9]. Cell growth was monitored by the increase in optical density (OD) at 600 nm (when necessary, samples were diluted 1/10 in culture medium). To obtain a conditioned medium, N. occultus cells were grown until late exponential phase (OD600=1.0), harvested by centrifugation at 12 000×g for 10 min and then the supernatant was ultrafiltrated (YM 30, Amicon) to eliminate the extracellular protease that may be present in the medium. Then, a low density culture (OD600=0.2–0.3) of N. occultus was centrifuged at 12 000×g and the cells were suspended in the same volume of either fresh medium (control) or late exponential conditioned medium. Both cultures were incubated at 37°C with agitation and samples were withdrawn at various times during growth. Then cells were harvested as described before and the cell-free culture media were stored at 4°C for determination of proteolytic activity and autoinducer detection in diffusion plate assays.

The reporter strain is the Ti plasmidless Agrobacterium tumefaciens NTL4 harboring plasmids pCF218, expressing the acylated homoserine lactone (AHL)-responsive regulator TraR, and pCF372 carrying a TraR/AHL-dependent lacZ fusion. TraR responds to a wide range of AHLs [10]. The AHL-positive control is A. tumefaciens KYC6 (traM::Tn5gusA7) that overproduces N-3-oxo-octanoyl-homoserine lactone [4]. These strains were grown in minimal medium plus glucose [11] supplemented with the corresponding antibiotics [12] and incubated at 25°C.

2.2 Assay for extracellular protease activity

Azocasein prepared in our laboratory as described by Kirtley and Koshland [13] was used as substrate. The reaction mixture containing 0.5% azocasein, 0.1 M sodium borate buffer (pH 8), 2 M NaCl and 0.1 ml of cell-free culture medium in a final volume of 0.5 ml was incubated at 50°C for 5 and 16 h. Incubations were stopped by adding 1 volume of cold 10% trichloroacetic acid and the assay tubes were left at 4°C for 30 min and then centrifuged at 3000×g for 10 min. Acid-soluble products were detected in the supernatant by measuring OD335. One unit of activity (U) was defined as the amount of enzyme that produced an increase of 1 in OD335 per hour under the conditions described above.

2.3 Preparation of ethyl acetate extracts

The haloalkaliphilic medium was autoclaved, split into two parts, one used as negative control (uninoculated) and the other inoculated with N. occultus cells. Samples (10 ml) were withdrawn throughout the growth curve and centrifuged at 12 000×g for 15 min. The supernatants from the samples and negative control were acidified to pH 5.5 by addition of glacial acetic acid, and then they were extracted three times with 20 ml of acidic ethyl acetate (0.1 ml glacial acetic acid per liter). The organic solvent was evaporated to dryness under a continuous flow of N2 gas. The residues were dissolved in 0.1 ml of the same solvent and stored at −80°C for autoinducer detection [14].

2.4 Diffusion plate assays

To overcome the problem of high salt concentration present in N. occultus culture media, a modified technique for autoinducer detection was developed for testing the presence of autoinducer in conditioned media. A culture of the reporter strain was suspended (1:5) in molten minimal medium (final concentration 0.4% agar) containing X-gal (40 µg ml−1) and poured over the surface of 1.5% minimal agar plates. After the top agar solidified, a 4-mm well was made with a hot glass rod tip in the middle of the plate and 10 µl of sample (control or conditioned medium) was loaded in individual plate wells. When ethyl acetate extractions were tested, A. tumefaciens NTL4 culture was streaked on 3.5-cm minimal medium agar plates containing X-gal and a well was made as described before. Two parallel streaks were performed in order to better visualize diffusion of the autoinducer. These plates were incubated at 25°C until cell growth was evident, and then 10 µl sample was added, air-dried and further incubated. In both assays, the plates were incubated at 25°C for 6 h and subsequently examined for β-galactosidase activity. The presence of autoinducer activity within a sample was evidenced by development of a blue color and it was quantified based on the intensity of the response. The arbitrary scale used was as follows: colorless, 0; light blue, 1; blue, 2; dark blue, 3. Ethyl acetate extract (2 µl) of A. tumefaciens KYC6, an overproducer of N-3-oxo-octanoyl-homoserine lactone, was used as positive control whereas plates without additions or 10 µl of uninoculated haloalkaliphilic medium (fresh or ethyl acetate-extracted) were used as negative controls.

3 Results and discussion

As mentioned, extracellular protease production in N. occultus started in the late exponential growth phase and reached a maximum in the stationary phase [5,8]. These observations led us to propose that protease production may be induced in response to high cell density which may be indicative of quorum sensing signaling. To examine this possibility, N. occultus low density cultures were suspended in fresh medium (control medium) or in a cell-free culture medium filtrate obtained from cells grown to late exponential phase (conditioned medium OD600=1.0) and the time course of extracellular protease activity was examined. As shown in Fig. 1, cell growth was very similar in both cultures (g=9 h), however, the onset of extracellular protease activity was shifted from late exponential to early exponential phase in the culture containing conditioned medium as compared to the control. The detection of protease activity in the culture containing conditioned medium was 16 h earlier than in control medium, and both cultures reached similar maximum activity levels (0.45 U ml−1). When the same experiment was carried out by adding conditioned media harvested at OD600=1.6 and 2.4, detection of protease activity was also evident 16 h earlier than with control medium. In both cases, the generation time was about 9 h and the maximum level of activity attained was 0.3 U ml−1. The outcome of these experiments indicated that extracellular protease production/activation in N. occultus may be related to factor(s) present in late exponential phase conditioned media.

Figure 1

Extracellular protease activity profile of N. occultus grown in fresh and conditioned media. N. occultus low density culture (OD600=0.3) was centrifuged and the cells were suspended in the same volume of fresh haloalkaliphilic medium (□ cell density; ■ protease activity) or a cell-free culture filtrate of N. occultus grown to late exponential phase containing fresh medium (9:1) (▵ cell density; ▲ protease activity). Cell growth was monitored by the increase in optical density at 600 nm and the proteolytic activity was measured by the determination of azocasein digestion by cell-free culture medium samples (see Section 2).

In many bacteria, the production of extracellular hydrolytic enzymes (including proteases) is regulated by quorum sensing [13,15]. This signaling mechanism is mediated by molecules of autoinducer that accumulate in the medium as the culture reaches a high cell density. To investigate the presence of potential autoinducers in N. occultus culture media, a modification of the diffusion plate assay was used. The results showed that addition of conditioned medium OD600=1 (Fig. 2A, c) elicited a response in the biosensor strain similar to that observed for the positive control (Fig. 2A, d), whereas the plates containing uninoculated haloalkaliphilic medium or without any addition (Fig. 2A, a and b, respectively) showed a negative response. The biosensor used in these experiments, Agrobacterium, is very sensitive to N-3-oxo-octanoyl- and N-3-oxodecanoyl-homoserine lactones, however, it can also detect homoserine lactones with longer side chains [14]. The induction of β-galactosidase activity in the reporter strain by N. occultus conditioned medium suggests the presence of AHL-like molecules in the culture medium of this archaeon. Byers et al. have reported that N-3-oxohexanoyl-l-homoserine lactone molecules are unstable at alkaline pH values [16]. Taking into account this observation, the autoinducer activity detected in N. occultus should have a half-life long enough to satisfy its function or may not be typical AHLs that are inactivated by alkaline conditions, but rather, are base stable.

Figure 2

Detection of autoinducer molecules in N. occultus. A: The detection of autoinducer molecules in N. occultus conditioned media was carried out pouring molten minimal soft agar media plus X-gal and A. tumefaciens NTL4 cells (reporter strain) on solidified agar medium. Additions to the wells were: (a) uninoculated haloalkaliphilic medium, (b) no additions, (c) N. occultus conditioned media, (d) positive control (A. tumefaciens KYC6 ethyl acetate extract). The plates were incubated at 25°C for 6 h and examined for β-galactosidase activity indicated by a blue color development. B: To analyze the time course of autoinducer activity, cell-free media from a culture of N. occultus harvested throughout the growth curve were acidified to pH 5.5, and then they were extracted with acidic ethyl acetate and tested against biosensor streaks. Samples were withdrawn at the following OD600: (a) 0.25; (b) 0.42; (c) 0.60; (d) 1.10; (e) 1.7; (f) 2.10; (g) 3.0. Plates h and i correspond to the negative (uninoculated haloalkaliphilic medium) and positive controls, respectively. The results shown in B were quantified based on the intensity of the response and plotted in the figure shown in C: ▲ OD600; ● autoinducer activity.

The presence of autoinducer molecules was analyzed throughout the N. occultus growth curve (Fig. 2B,C). Samples were withdrawn at various times and cell-free supernatants were extracted with ethyl acetate and tested for autoinducer activity. Fig. 2B shows the results obtained in the diffusion plate assay and Fig. 2C represents an estimation of autoinducer activity (arbitrary units) along the growth curve. As observed in Fig. 2B, β-galactosidase activity was detected in samples corresponding to OD600=1.10, 1.70 and 2.10 (d–f), the latter showing the strongest response. These samples correspond to the transition from exponential to stationary growth phase (Fig. 2C), the stage at which protease activity is detected in the extracellular medium in the control culture (Fig. 1).

At present, there are no reports about the presence of quorum sensing-like mechanisms in archaea. We report for the first time preliminary evidence on the occurrence of potential autoinducer molecules of the homoserine lactone type in the archaeon N. occultus. These molecules may be responsible for the production/activation of extracellular protease activity during the late exponential growth phase. Further work is being carried out in our group to investigate whether this type of molecule is widespread in other archaea. Future work will be aimed towards the isolation and biochemical characterization of these molecules and to investigate their relationship with protease production/activation.


This work was supported by a grant from the Agencia Nacional de Promoción Cientı́fica y Tecnológica — Universidad Nacional de Mar del Plata (PICT2000, no. 01-09763), Argentina.


  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].
View Abstract