The Social Life of Aeromonas through Biofilm and Quorum Sensing Systems

Abstract

Bacteria of the genus Aeromonas display multicellular behaviors herein referred to as "social life". Since the 1990s, interest has grown in cell-to-cell communication through quorum sensing signals and biofilm formation. As they are interconnected, these two self-organizing systems deserve to be considered together for a fresh perspective on the natural history and lifestyles of aeromonads. In this review, we focus on the multicellular behaviors of Aeromonas, i.e., its social life. First, we review and discuss the available knowledge at the molecular and cellular levels for biofilm and quorum sensing. We then discuss the complex, subtle, and nested interconnections between the two systems. Finally, we focus on the aeromonad multicellular coordinated behaviors involved in heterotrophy and virulence that represent technological opportunities and applied research challenges.

Keywords: bacterial communities; biofilm; cooperation; coordination; multicellularity; quorum sensing; social life; virulence.

Figure 1

Effectors involved in different phases of biofilm development in aeromonads. Planktonic aeromonads initiate the formation of biofilm on surface under influence of environmental conditions. Several bacterial factors are involved in the attachment step, including flagella and other external structures, chemotaxis system, and cytoskeleton. After division, bacteria that were well-aggregated, attached to the surface to form a microcolony. Biofilm acquires its mechanical stability by the production of an EPS matrix encompassing proteins, polysaccharides, extracellular DNA, and lipids. The AI-1 quorum sensing system enhances the maturation of biofilm, which is likely related to the second messenger c-di-GMP involved in the bacterial transition from planktonic to sessile lifestyle. When the conditions of life in biofilm deteriorate (e.g., nutrient limitation), a dispersion phase occurs and aeromonads escape from biofilm and return to the planktonic lifestyle. In another case, the biofilm can be detached by external stress (e.g., shear forces). AI-1, Autoinducer-1 quorum sensing system; AI-2, Autoinducer-2 quorum sensing system; AI-3, Autoinducer-3 quorum sensing system; EAL, protein domains harboring phosphodiesterase activity involved in the c-di-GMP degradation; EPS, extracellular polymeric substances; GGDEF, protein domains harboring guanylate synthase activity involved in the c-di-GMP synthesis; LPS, lipopolysaccharides.

Figure 2

Schematic representation of AI-1 quorum sensing system in Aeromonas. From an in vitro model of autoinducer 1 (AI-1) quorum sensing system of A. hydrophila, Garde et al. (2010) have distinguished two phases since the complex AI-1/receptor (AhyR) activates the quorum sensing loop of the initial AI-1 producer bacterial cell during exponential growth (A) or of other bacterial cells during the stationary phase (B) due to slow decay of the complex AI-1/receptor (AhyR). (A) Autoinduction occurs during exponential growth phase (Garde et al., 2010). In this phase, the enzyme (E) AhyI synthetizes AI-1 signal molecules of acyl-homoserine lactones (AHL) from S-adenosyl-methionine (SAM) and acyl–acyl carrier proteins (acyl) (Swift et al., ; Parsek et al., 1999). The protein AhyR is the sensor (S) of the AI-1 system and is activated by AHL molecules (Swift et al., 1997). Once activated, AhyR is a transcriptional regulator for the ahyRI locus encompassing AhyI and AhyR encoding genes, and participates in the auto-amplification loop (Kirke et al., ; Garde et al., 2010). The transcription of ahyRI locus is also likely enhanced (discontinuous traits) by the second messenger c-di-GMP and by AI-2 synthase LuxS or by AI-3 transcriptional regulatory protein QseB (Kozlova et al., 2012). The AHL molecules are freely diffusible across bacterial membranes and accumulate in the extracellular environment (Garde et al., 2010). (B) Intercellular activation occurs over an AHL concentration threshold corresponding to high cell density occurring at the stationary phase (Garde et al., 2010). Once activated by AHL molecules, AhyR is a transcriptional regulator for several genes associated to virulence and biofilm formation. In contrast to its action during the autoinduction phase, activated-AhyR negatively regulates the transcription of the ahyRI locus (Kirke et al., 2004). This AI-1 system is inhibited by exogenous long chain AHL or furanones (Swift et al., ; Ponnusamy et al., 2010) that may act as competitive inhibitors of AHL for AhyR binding. The AI-1 quorum sensing system negatively regulates the transcription of luxS and qseBC loci, encoding AI-2 synthase and AI-3 two components system, respectively (Kozlova et al., 2011, 2012).

Figure 3

Schematic representation of AI-2 quorum sensing system in Aeromonas. Aeromonads are able to produce the AI-2 synthase enzyme (E) LuxS, and AI-2 (autoinducer 2) quorum sensing system has been detected in the genus (Kozlova et al., 2008). In the bacterial AI-2 quorum sensing systems, LuxS catalyzes the cleavage of S-ribosyl-homocysteine (SRH) derived from S-adenosyl-methionine (SAM) in homocysteine and 4,5-dihydroxy-2,3-pentanedione (DPD) (Xavier and Bassler, 2003). DPD spontaneously cyclizes to form a furanone, which can possibly react with borate (-B) depending on the bacterial species (discontinuous traits), and leading to AI-2 molecule formation (Chen et al., ; Miller et al., 2004). Based on studies in Vibrio, it has been shown that in absence of AI-2, LuxQ generates a phosphorylation cascade (-P) via LuxU and ultimately LuxO. LuxO is the response regulator that represses the master regulatory protein HapR (V. cholerae). At high cell density, AI-2 freely diffusible molecules reach a threshold and bind the LuxP periplasmic receptors. The autoinducer signal is transduced by the LuxP/AI-2 complex, inactivating the transmembrane sensor kinase LuxQ and subsequently leading to LuxO inactivation, which lifts repression of HapR and influences gene expression (Bassler et al., ; Henke and Bassler, 2004). However, the AI-2-internalization step of aeromonads is not yet known (discontinuous traits) and no luxP homolog were detected into their genomes (Kozlova et al., 2008). Signal transduction may involve the proteins LuxQ, LuxU, LuxO and subsequently the transcriptional regulator LitR (homolog of HapR), but the level of proof is so far only genetic (Kozlova et al., 2011). Overall, the AI-2 activation system in Aeromonas is associated with inhibition of biofilm maturation, enhancement of swimming and a decrease in virulence (Kozlova et al., 2008). The transcription of luxS locus is likely inhibited (discontinuous traits) by AI-1 quorum sensing system (Kozlova et al., 2011).

Figure 4

Schematic representation of AI-3 quorum sensing system in Aeromonas. Although, the two-component system QseB/QseC was characterized in Aeromonas (Khajanchi et al., 2012), the synthesis of autoinducer 3 (AI-3) signals is not yet known in this genus (discontinuous traits). According to the Escherichia coli model, the transmembrane protein QseC is a sensor (S) that can bind at its periplasmic domain: (i) signal molecules of AI-3 from other cells of a bacterial clone or from other bacterial species, or (ii) catecholamines (epinephrine, Epi, or norepinephrine, NE) from a eukaryotic host (Sperandio et al., ; Clarke et al., 2006). QseC then undergoes autophosphorylation (-P) at its cytoplasmic domain. The signal is then transmitted by phosphorylation (-P) to the transcriptional regulatory protein QseB. Subsequently, activated QseB-P binds to the transcription regulator domains of virulence-associated genes (flagella, shiga-like toxin, type III secretion system components, and effectors) and autoregulates its own operon qseBC (Sperandio et al., ; Clarke et al., 2006). Overall, the AI-3 system activation in Aeromonas is associated with inhibition of biofilm maturation, enhancement of swimming and swarming and an increase in virulence (Khajanchi et al., 2012). The transcription of qseBC locus is likely inhibited (discontinuous traits) by AI-1 quorum sensing system (Kozlova et al., 2012).