Stimulates Nitrogen Fixation and Favors Ethanol Production

Preparing to load PDF file. please wait...

0 of 0
100%
Stimulates Nitrogen Fixation and Favors Ethanol Production

Transcript Of Stimulates Nitrogen Fixation and Favors Ethanol Production

International Journal of
Molecular Sciences

Article
The Quorum Sensing Auto-Inducer 2 (AI-2) Stimulates Nitrogen Fixation and Favors Ethanol Production over Biomass Accumulation in Zymomonas mobilis
Valquíria Campos Alencar 1,2, Juliana de Fátima dos Santos Silva 1, Renata Ozelami Vilas Boas 2, Vinícius Manganaro Farnézio 1, Yara N. L. F. de Maria 2, David Aciole Barbosa 2 , Alex Tramontin Almeida 3, Emanuel Maltempi de Souza 3, Marcelo Müller-Santos 3, Daniela L. Jabes 2 , Fabiano B. Menegidio 2, Regina Costa de Oliveira 2, Tiago Rodrigues 1 , Ivarne Luis dos Santos Tersariol 4, Adrian R. Walmsley 5 and Luiz R. Nunes 1,*

Citation: Alencar, V.C.; Silva, J.d.F.d.S.; Vilas Boas, R.O.; Farnézio, V.M.; de Maria, Y.N.L.F.; Aciole Barbosa, D.; Almeida, A.T.; de Souza, E.M.; Müller-Santos, M.; Jabes, D.L.; et al. The Quorum Sensing Auto-Inducer 2 (AI-2) Stimulates Nitrogen Fixation and Favors Ethanol Production over Biomass Accumulation in Zymomonas mobilis. Int. J. Mol. Sci. 2021, 22, 5628. https://doi.org/10.3390/ijms22115628
Academic Editor: Paola Brun
Received: 7 April 2021 Accepted: 27 April 2021 Published: 26 May 2021
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

1 Centro de Ciências Naturais e Humanas, Universidade Federal do ABC (UFABC), Alameda da Universidade, s/n, São Bernardo do Campo 09606-045, SP, Brazil; [email protected] (V.C.A.); [email protected] (J.d.F.d.S.S.); [email protected] (V.M.F.); [email protected] (T.R.)
2 Núcleo Integrado de Biotecnologia, Universidade de Mogi das Cruzes (UMC), Av. Dr. Cândido Xavier de Almeida Souza, 200, Mogi das Cruzes 08780-911, SP, Brazil; [email protected] (R.O.V.B.); [email protected] (Y.N.L.F.d.M.); [email protected] (D.A.B.); [email protected] (D.L.J.); [email protected] (F.B.M.); [email protected] (R.C.d.O.)
3 Setor de Ciências Biológicas-Departamento de Bioquímica e Biologia Molecular, Universidade Federal do Paraná (UFPR), Rua Cel. Francisco H. dos Santos, 100, Curitiba 81531-980, PR, Brazil; [email protected] (A.T.A.); [email protected] (E.M.d.S.); [email protected] (M.M.-S.)
4 Departamento de Bioquímica, Universidade Federal de São Paulo (UNIFESP), Rua Três de Maio, 100, São Paulo 04044-020, SP, Brazil; ivarne.tersariol[email protected]
5 Department of Biosciences, Durham University, South Road, Durham DH1 3LE, UK; [email protected]
* Correspondence: [email protected]; Tel.: +55-11-4996-8371 (ext. 4996-3166)
Abstract: Autoinducer 2 (or AI-2) is one of the molecules used by bacteria to trigger the Quorum Sensing (QS) response, which activates expression of genes involved in a series of alternative mechanisms, when cells reach high population densities (including bioluminescence, motility, biofilm formation, stress resistance, and production of public goods, or pathogenicity factors, among others). Contrary to most autoinducers, AI-2 can induce QS responses in both Gram-negative and Gram-positive bacteria, and has been suggested to constitute a trans-specific system of bacterial communication, capable of affecting even bacteria that cannot produce this autoinducer. In this work, we demonstrate that the ethanologenic Gram-negative bacterium Zymomonas mobilis (a non-AI-2 producer) responds to exogenous AI-2 by modulating expression of genes involved in mechanisms typically associated with QS in other bacteria, such as motility, DNA repair, and nitrogen fixation. Interestingly, the metabolism of AI-2-induced Z. mobilis cells seems to favor ethanol production over biomass accumulation, probably as an adaptation to the high-energy demand of N2 fixation. This opens the possibility of employing AI-2 during the industrial production of second-generation ethanol, as a way to boost N2 fixation by these bacteria, which could reduce costs associated with the use of nitrogen-based fertilizers, without compromising ethanol production in industrial plants.

Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Keywords: Zymomonas mobilis; quorum sensing; AI-2; N2 fixation; ethanol production; transcriptome
1. Introduction Quorum sensing (QS) is a chemical communication process employed by bacteria
to assess the population density in their surrounding environment and synchronize their behavior on a community scale [1,2]. This communication is mediated by small signaling molecules called auto-inducers (AI), which are constitutively produced by the cells,

Int. J. Mol. Sci. 2021, 22, 5628. https://doi.org/10.3390/ijms22115628

https://www.mdpi.com/journal/ijms

Int. J. Mol. Sci. 2021, 22, 5628

2 of 18

during all stages of development. Thus, when the population reaches higher cellular densities, the concentration of AI molecules in the environment becomes critical, triggering a coordinated response, in which all cells reprogram their gene expression patterns simultaneously, activating genes involved with a series of alternative activities, such as bioluminescence, motility, biofilm production and, in many cases, secretion of public goods and/or pathogenicity factors [3]. In this sense, the QS response involves expression of phenotypes that would be too costly and counterproductive, if executed by only a few cells, but result in efficient responses, when expressed in a coordinated manner, by the entire population [3].
Gram-positive bacteria generally use oligopeptides as QS autoinducers, while Gramnegative bacteria use acyl-homoserine-lactone derivatives (AHLs). Collectively, these molecules are called type 1 auto-inducers, or AI-1 [2,3]. In addition, many bacteria can use an alternative QS signaling molecule, known as a type 2 auto-inducer (AI-2) [4]. The basic structure of AI-2 is a 4,5-dihydroxy-2,3-pentanedione (DPD), which cyclizes spontaneously to form a series of interconverting derivatives, which coexist in a dynamic balance [3]. AI-2 is synthesized during the reaction mediated by the enzyme LuxS, as a byproduct of the active methyl cycle, which transfers methyl groups to various substrates, during major biosynthetic pathways of bacterial metabolism [5]. Contrary to the oligopeptide and AHL autoinducers (which display species-specific structure and activity), AI-2 can trigger QS responses in various bacteria (both Gram-negative and Gram-positive), and has been suggested to constitute a trans-specific system of bacterial communication, capable of affecting the composition, and behavior of complex bacterial communities [4–6].
In this manuscript, we show that the Gram-negative bacterium Zymomonas mobilis (a non-AI-2 producer) responds to the presence of this autoinducer with significant alterations in its growth rate and gene expression profile. Such transcriptional reprogramming affects genes associated with various mechanisms typically associated with QS responses in other microorganisms, including motility, DNA repair, and nitrogen fixation, among others. Z. mobilis is a diazotrophic ethanologenic bacterium, widely employed in the production of second-generation ethanol and other industrial bioconversion processes [7]. Interestingly, previous studies have suggested that activation of the bacterial N2-fixing mechanism could result in significant savings for the bioethanol industry, allowing atmospheric N2 gas to be used as a nitrogen source, in substitution to industrial fertilizers [8,9]. Moreover, N2 fixation in Z. mobilis seems to be coupled with reduced glucose-to-biomass and increased glucose-to-ethanol turnover ratios, suggesting that AI-2 may assist in the development of improved mechanisms for bioethanol production.
2. Results 2.1. Z. mobilis Response to Increasing Concentrations of AI-2
To check Z. mobilis response to AI-2, liquid cultures were established in minimal medium (MM), as described in Materials and Methods. When cells reached exponential growth phase, AI-2 was added to each culture, at different concentrations (0, 18, 36, and 45 µM), and cell growth continued to be monitored by OD600 measurements, until all cultures reached stationary phase. As observed in Figure 1A, cells grown in MM showed a long period of adaptive growth (lag phase), reaching exponential phase only after 12–14 h of incubation. In the absence of AI-2, the cultures showed decreased growth rate after 16 h and stationary phase is established after 19 h. However, the decrease in exponential growth rate and the consequent establishment of stationary phase are induced earlier in Z. mobilis cultures (in a concentration-dependent manner) by the presence of AI-2. For example, cultures subjected to 45 µM AI-2 showed a pronounced reduction in growth rate during the first hour of incubation, entering stationary phase at 16 h (Figure 1A).

Int. J. Mol. Sci. 2021, 22, 5628

3 of 18

Figure 1. Effects of AI-2 on Z. mobilis growth rate and viability. (A) An initial Z. mobilis culture was established, in 5 mL of liquid MM, from an isolated colony, grown in solid MM. This culture was incubated at 30 ◦C (without shaking) for 24 h, when aliquots were taken and used to inoculate a flask containing 80 mL of liquid MM, to an OD600 = 0.05. This flask was then incubated under the same conditions and cell growth was monitored hourly by OD600 readings. When this culture reached mid-exponential growth phase (at t = 14 h), it was subdivided into four vials of 20 mL each. AI-2 was then added to these cultures, at different concentrations (0, 18, 36, and 45 µM). Next, the cultures were re-incubated, as described above, and the OD600 continued to be monitored, until all reached stationary phase. OD600 readings shown in the graph represent the averages ± SEM from three independent experiments. Aliquots (2 mL) were taken for RNA extraction at the times indicated by the arrows, namely: (i) t = 14 h (before addition of AI-2); (ii) t = 16 h + AI-2 (treated with 45 µM AI2); (iii) t = 20 h + AI-2 (treated with 45 µM AI- 2); and (iv) t = 20 h (not treated with AI-2). See text for further details. (B) Representative dot plots showing SYTO 9 and PI fluorescence obtained by flow cytometry in cells grown in the absence of AI-2 (control), treated with 70% isopropyl alcohol (positive control), and treated with 45 µM AI-2, at the same incubation times highlighted in panel (A). (C) Quantification of percentages of live and dead cells. The results are presented as the mean ± SEM of three independent experiments and * represents statistically significant differences in relation to the respective controls (AI-2 untreated cultures), after a two-tailed t-test, using p < 0.05 as a threshold; ns represent statistically non-significant differences.

Int. J. Mol. Sci. 2021, 22, 5628

4 of 18

To verify whether AI-2 exhibits toxicity against Z. mobilis cells, a flow cytometry assay, employing the LIVE/DEAD® BacLigh® Bacterial Viability Kit, was conducted at the same experimental conditions (see Materials and Methods, for details). This assay uses two fluorochromes that bind to nucleic acids, SYTO 9, and propidium iodide (PI). SYTO 9 is a cell-permeant green fluorescent dye that labels live and dead bacteria, whereas PI, as a cell-impermeant red fluorescent dye, only enters into bacterial cells with altered membrane permeability (damaged/dead). The binding of PI to nucleic acids results in an apparent diminished green fluorescence of SYTO 9, due to quantum yielding, when both fluorochromes are inside the cells, allowing visual separation of viable cells (containing intact membranes), from membrane-compromised/dead bacteria [10]. In this assay, isopropyl alcohol was used to promote a positive control for cell damage/death, as indicated in the manufacturer’s instructions (see Materials and Methods). Based on the position of cell populations in both untreated controls and alcohol-treated experiments, a gate-based strategy was used to distinguish between live and damaged/dead cells. The representative dot plots presented in Figure 1B show that AI-2 did not alter the viability of Z. mobilis cells, when AI-2-treated cells were compared to the negative control (absence of AI-2). The quantification of the percentages of live and dead/damaged cells is presented in Figure 1C, considering a total of three independent replicas. Thus, these data show that the decreased growth rate induced by AI-2 does not seem to be due to AI-2 toxicity.
Global transcriptome analyses were next conducted to characterize the main metabolic traits of Z. mobilis that were affected by the presence of AI-2, using RNA-seq libraries (see Materials and Methods for details). Thus, aliquots were taken from cultures treated with 45 µM AI-2 for periods of 2 and 6 h (t = 16 h + AI-2 and t = 20 h + AI-2, respectively, as shown in Figure 1). The transcriptomes of cells harvested in these two moments were then compared to the transcriptome of cells not treated with AI-2, in exponential growth phase, harvested at time t = 14 h (immediately before the addition of AI-2). However, since cells involved in these analyses were at different growth phases (exponential phase at t = 14h, or stationary phase at t = 16 h + AI-2 and t = 20 + AI-2), cells were also harvested from a culture not treated with AI-2, at time t = 20 h. These cells also showed a growth pattern compatible with stationary phase, allowing us to differentiate gene expression changes involved with entry into stationary phase from those derived from the presence of AI-2, when comparing all available transcriptomes (t = 20 h, t = 16 h + AI-2 and t = 20 h + AI-2) against the same reference (t = 14 h).
2.2. Characterization of Gene Modulation Patterns in Z. mobilis, in Response to Cell Growth and the Presence of AI-2
Total RNA was extracted from the bacteria harvested at the abovementioned timepoints and used to construct RNA-seq libraries. All transcriptome comparisons were made using the transcriptome of untreated cells, at mid-exponential growth phase (t = 14 h) as a common reference, resulting in three pairs of comparisons: (i) t = 20 h × t = 14 h; (ii) t = 16 h + AI-2 × t = 14 h and (iii) t = 20 h + AI-2 × t = 14 h. As highlighted in Materials and Methods, all analyses were conducted with results obtained from three independent cultures, obtained for each experimental condition. The complete dataset derived from the RNA-seq analyses can be found at the Open Science Framework (OSF) repository (https://osf.io/rs8pu/, accessed on 25 April 2021). Differentially expressed genes were identified by ANOVA, followed by a Benjamini–Hochberg (BH) post-test, using q ≤ 0.01 as the limit for statistical reliability. Additionally, only genes displaying absolute log2 modulation ≥ 0.6, in at least one experimental timepoint, were considered for further analyses (this threshold is equivalent to >50% up, or down-regulation, in comparisons with gene expression values at t = 14 h). These conditions led to the identification of 724 genes, whose modulation patterns are shown in Figure 2 (details concerning gene identities are provided in the Supplementary Materials File S1). To verify the overall reliability of the RNA-seq results, a set of 20 genes had their relative expression ratios confirmed by qPCR and a qualitative comparison of the results obtained by these two methodologies indicate

Int. J. Mol. Sci. 2021, 22, 5628

5 of 18

that ~80% of them were equally identified as either induced, or repressed, by these two independent approaches (see Supplementary Materials File S2).

Figure 2. Gene expression patterns of Z. mobilis cultures, in response to the presence of AI-2 and different growth phases. The gene expression patterns shown above were obtained through RNA-seq experiments, employing samples obtained from the different timepoints depicted in Figure 1. RNA obtained from cells grown for t = 14 h (exponential growth phase) was used as a common reference for transcriptome comparisons with cells grown for t = 20 h (stationary phase), as well as cells induced to early stationary phase, due to the presence of 45 µM AI-2 (t = 16 h + AI-2 and t = 20 h + AI-2). The figure shows the average values for each of the 724 genes identified as modulated during three relative comparisons [log2(t = 20 h /t = 14 h), Log2(t = 16 h + AI-2/t = 14 h) and log2(t = 20 h + AI-2/t = 14 h)] (see Materials and Methods for details). The genes and experimental conditions were subjected to a hierarchical clustering algorithm and visualized with the aid of TMEV [11]. Genes highlighted in Cluster 1 show an expression pattern that differentiates cells treated with AI-2 from those that spontaneously entered stationary phase (see text for details).

Int. J. Mol. Sci. 2021, 22, 5628

6 of 18

As seen in Figure 2, significant gene modulation can be verified in Z. mobilis cells after 2 h of treatment with AI-2 (t = 16 h + AI-2 × t = 14 h) and such pattern remains practically unchanged, even after 6 h of induction (t = 20 h + AI-2 × t = 14 h). In fact, most of these modulations are likely associated with bacterial entry into stationary phase (which seems to be early induced by AI-2), as they can also be verified in control cells (not treated with the autoinducer), which have spontaneously entered stationary phase, at t = 20 h (t = 20 h × t = 14 h). However, when these three pairs of comparisons are viewed with the aid of a hierarchical clustering algorithm, it is possible to identify a series of genes that are specifically modulated only in AI-2-treated cells, confirming that this signaling molecule induces a significant number of transcriptional changes in Z. mobilis, which are not related with entry into stationary phase (see cluster 1 in Figure 2).
2.3. Metabolic and Structural Alterations Verified in Z. mobilis in Response to Cell Growth and the Presence of AI-2
A prospective evaluation of the Z. mobilis functional/structural response to AI-2, at different growth conditions, was obtained by comparing the expression profile of the 724 genes observed in Figure 2 with the OMICS Dashboard tool, available through the Pathway Tools software (https://arxiv.org/abs/1510.03964, accessed on 20 June 2020). Thus, genes were initially distributed into their respective Gene Ontology (GO) categories [12] and grouped in specific structural/functional systems/subsystems, according to the Z. mobilis ZM4 Pathway/Genome Database (PGDB), available at the BioCyc repository (https://biocyc.org/, accessed on 20 June 2020). Expression ratios for each categorized gene are shown in Figure 3, along with the overall modulation of each system/subsystem, which is represented by the sum of relative expression ratios of their individual genes (detailed information regarding individual genes present in each system/subsystem can be found in the Supplementary Materials File S3).
These analyses confirmed that many Z. mobilis structural/functional systems/subsystems display similar modulation patterns in all three transcriptome comparisons, suggesting that they are likely associated with entry into stationary phase, rather than AI-2-specific responses. For example, genes involved in processes associated with protein synthesis and metabolism are always negatively modulated, regardless of the presence of AI-2 (Figure 3A and Supplementary Materials File S3A). A similar situation can be verified with genes involved in general biosynthetic processes, with emphasis on amino acid and nucleotide biosynthesis (Figure 3B and Supplementary Materials File S3B), as well as in catabolic processes involved in the degradation of such biomolecules (Figure 3C and Supplementary Materials File S3C). However, when the transcriptome profiles of AI-2 treated cells (at t = 16 h + AI-2 and/or at t = 20 h + AI-2) are compared with untreated controls (at t = 20 h), significant differences can be found in the expression profile of genes involved in: (i) cell membrane composition (Figure 3D and Supplementary Materials File S3D); (ii) locomotion (Figure 3E and Supplementary Materials File S3E); (iii) DNA repair (Figure 3F and Supplementary Materials File S3F); (iv) Entner–Doudoroff glycolytic pathway (Figure 3G and Supplementary Materials File S3G) and (v) inorganic nutrient metabolism, with emphasis on nitrogen (N2) fixation (Figure 3H and Supplementary Materials File S3H), a metabolic process that has been proposed to bear significant biotechnological potential for the bioethanol industry [8,9].

Int. J. Mol. Sci. 2021, 22, 5628

7 of 18

Figure 3. Functional distribution of differentially modulated genes. The OMICS Dashboard tool, available through the Pathway Tools package, was used to distribute the 724 genes shown in Figure 2 into the 8 main functional/structural systems/subsystems defined in the ZM4 Pathway Genome Database (PGDB), available at BioCyc (panels A to H) (see text, for details). Total modulation of the different categories is shown by the sum of relative expression (Log2 ratios) for all genes present in each system/subsystem. Details regarding each gene present in each functional/structural category can be found in Supplementary Materials File S3.
2.4. The Presence of AI-2 Increases N2 Fixation by Z. mobilis The Z. mobilis genome contains all genes necessary to perform N2-fixation, including
an RpoN-like sigma factor (Sigma 54) [13]. All additional genes involved in this process are located in a single chromosomal region [13], which includes the gene encoding the NifA regulator, the operon nif HDKENX-fdxB-nif Q (which encodes the nitrogenase enzyme), two operons involved in nitrogenase maturation (nifB-fdxN and iscN-nif USVW-modD), and the RNF operon (rnf ABCDGEH), which encodes members of the RNF electron-transport complex, responsible for donating electrons to nitrogenase, during N2 fixation. As observed in Figure 4, most of these elements were identified as positively modulated, in response to the presence of AI-2.

Int. J. Mol. Sci. 2021, 22, 5628

8 of 18

Figure 4. Transcriptional profile of genes responsible for N2 fixation in Z. mobilis, in response to AI-2. The upper panel (A) shows the N2 fixation locus present in the Z. mobilis chromosome. This locus consists of four operons, which contain genes directly involved in the production/maturation of nitrogenase, as well as those encoding electron donors, involved in the N2 fixation process. The bottom panel (B) shows the expression pattern of the genes contained in this locus, in the presence or absence of AI-2, according to our RNA-seq data. The top element in panel B shows the expression profile of gene ZMO_RS01195, which is located outside the N2 fixation locus shown in panel A. This element encodes an RpoN-like sigma factor (Sigma 54), supposedly involved in transcription of N2-fixation genes, which is also upregulated in response to AI-2.

Int. J. Mol. Sci. 2021, 22, 5628

9 of 18

To verify if the AI-2-induced modulation verified in the abovementioned genes really affects nitrogen fixation in Z. mobilis, cultures of this bacterium were subjected to an acetylene reduction assay (ARA), the conventional methodology used to measure nitrogenase activity in bacterial cells [14]. In this assay, bacterial nitrogenase activity reduces acetylene (C2H2) to ethylene (C2H4), which is then measured by gas chromatography (see Materials and Methods for details). As observed in Figure 5, Z. mobilis cells incubated in the presence of 45 µM AI-2 showed a significant increase (~4–8×) in nitrogenase activity, when compared to cells incubated in the absence of this auto-inducer. Interestingly, N2 fixation appears to be stimulated by AI-2 both in the presence, or absence, of fixed nitrogen (NH4+) in the culture medium, although total enzyme activity is reduced in the first condition, as previously observed in other diazotrophic bacteria [15].

Figure 5. Acetylene reduction assay (ARA) to measure nitrogenase activity in Z. mobilis, in response
to the presence of AI-2. Z. mobilis cells were inoculated in semi-solid MM, either in the absence or in the presence of 45 µM AI-2. After incubation for 48 h at 30 ◦C, 1 mL of 10% acetylene was injected into the headspace of the flasks, which were incubated for another hour, at 30 ◦C. The ethylene (C2H4) formed inside the bottles (resulting from the action of bacterial nitrogenase) was then detected and
quantified by gas chromatography. The cellular biomass present in each flask was estimated by total
protein quantification and the nitrogenase activity expressed as nmols of C2H4 produced/min/mg of protein. The experiments were carried out in triplicate and the graphs show the mean and SEMs
obtained for each condition. The figure also compares the results obtained in an assay conducted
in MM medium containing 1 mM (NH4)2SO4 (A) and a similar assay, conducted in MM lacking (NH4)2SO4 (B), since the presence of NH4+ is known to inhibit nitrogenase activity [16].
2.5. The Presence of AI-2 Favors Glucose-to-Ethanol Conversion, in Detriment of Biomass Accumulation in Z. mobilis Cells
As mentioned above, previous studies have suggested that activation of the N2fixing mechanism in Z. mobilis could be economically relevant for bioethanol plants, which could use N2 gas as a substitute for industrial fertilizers, during the production of second-generation ethanol [8,9]. Interestingly, the activation of N2 fixation in Z. mobilis seems to favor glucose-to-ethanol conversion (instead of diverting this carbon source to biomass accumulation), resulting in more efficient ethanol/glucose ratios during fermentation [8]. Glucose internalization is mediated by a specific permease, encoded by gene ZMO_RS01265 [17]. Once inside the cell, this monosaccharide is degraded by the Entner–Doudoroff glycolytic pathway (ED), followed by pyruvate decarboxylation and acetaldehyde reduction to ethanol [18]. However, the presence of AI-2 seems to affect the expression of few genes involved in glucose-to-ethanol interconversion and it is difficult

Int. J. Mol. Sci. 2021, 22, 5628

10 of 18

to infer the overall effect of AI-2 on ethanol production based solely on the RNA-seq data (see Supplementary Materials File S4).
Thus, the production of ethanol by cells grown in the presence or absence of AI-2 was tested directly, using aliquots harvested from cultures grown under conditions similar to those shown in Figure 1. These aliquots (harvested at times t = 16 h and 20 h) were centrifuged and the concentration of ethanol in their supernatants was measured with the aid of the EnzyChrom® Ethanol Assay kit (see Materials and Methods for details). The results showed that total ethanol production was not significantly affected in all cultures, regardless of the presence or absence of AI-2 (Figure 6). However, since AI-2 inhibits Z. mobilis growth (thus reducing cellular biomass), the ethanol-to-glucose turnover ratio, per cell, is significantly higher in AI-2-treated cultures, when compared to untreated controls (Figure 6). Overall, these results point to the possible use of this autoinducer in improved mechanisms for bioethanol production, based on boosting N2 fixation (see below).

Figure 6. Ethanol production by Z. mobilis in response to AI-2. Cultures of Z. mobilis were established, as described in Figure 1 and grown for 14 h, until they reached mid exponential growth phase. At this time, AI-2 was added, at 45 µM final concentration, and the cultures were further incubated for additional six hours, until they all reached stationary growth phase. Aliquots (2 mL) were collected at times t = 16 h and t = 20 h for ethanol quantification, using the EnzyChrom® Ethanol Assay kit, and to quantify total protein accumulated in the bacterial biomass, at these different timepoints (see Materials and Methods for details). Blue bars display the absolute ethanol concentration in each sample (left axis), while the red dots display their relative ethanol production, normalized by biomass accumulation (right axis). The graph shows the average results and their respective SEMs, obtained from three independent experiments. * represents statistically significant differences in relation to the respective controls (AI-2 untreated cultures), after a two-tailed t-test, using p < 0.05 as a threshold.
3. Discussion
QS is defined as a chemical communication mechanism employed by bacteria to assess the population density in their surrounding environment and synchronize behavior on a community scale [1,2]. In this sense, QS seems to blur a well-established distinction between prokaryotes and eukaryotes, by allowing bacteria to coordinate their gene expression patterns not as single-cell individuals, but rather as observed in multicellular eukaryotic structures, such as tissues and organs [19]. Although different signaling molecules can be used to trigger this process, the QS response invariably involves behaviors that can be considered unproductive when performed by a few cells acting alone, but become beneficial to the community, when performed simultaneously by a large number of individuals, such as biofilm formation, motility, and toxin production, among others [3]. Moreover, the existence of a trans-specific QS signaling system, involving AI-2, allows the QS response to affect the behavior and composition of even complex bacterial communities, as recently demonstrated by studies involving the intestinal microbiota [20,21].
CellsMaterialsGenesPresencePhase