Effects of Halophyte Root Exudates and Their Components on Chemotaxis, Biofilm Formation and Colonization of the Halophilic Bacterium Halomonas Anticariensis FP35T.

Increase in soil salinity poses an enormous problem for agriculture and highlights the need for sustainable crop production solutions. Plant growth-promoting bacteria can be used to boost the growth of halophytes in saline soils. Salicornia is considered to be a promising salt-accumulating halophyte for capturing large amounts of carbon from the atmosphere. In addition, colonization and chemotaxis could play an important role in Salicornia-microbe interactions. In this study, the role of chemotaxis in the colonization of the halophilic siredophore-producing bacteria, Halomonas anticariensis FP35T, on Salicornia hispanica plants was investigated. The chemotactic response of FP35T to Salicornia root exudates showed optimum dependence at a salt concentration of 5 % NaCl (w/v). Oleanolic acid, the predominant compound in the exudates detected by HPLC and identified by UPLC-HRMS Q-TOF, acts as a chemoattractant. In vitro experiments demonstrated the enhanced positive effects of wild-type H. anticariensis strain FP35T on root length, shoot length, germination and the vigour index of S. hispanica. Furthermore, these positive effects partially depend on an active chemotaxis system, as the chemotaxis mutant H. anticariensis FP35 ΔcheA showed reduced plant growth promotion for all the parameters tested. Overall, our results suggest that chemotaxis responses to root exudates play an important role in interactions between Salicornia and halophilic bacteria, enhance their colonization and boost plant growth promotion. Preliminary results also indicate that root exudates have a positive impact on H. anticariensis FP35T biofilm formation under saline conditions, an effect which totally depends on the presence of the cheA gene.


Swimming Motility Assay
The swimming motility of the bacterial strains was tested using a plate-based assay described by Ha et al. [47]. Briefly, overnight cultures of bacterial strains were diluted to OD 600 of 0.2 and, using toothpick bacteria, were point-inoculated onto M8 plates containing 0.3% agar. Motility was observed after 16 h of incubation at 32 • C.

Identification of Chemosensory Systems in H. anticariensis FP35 T
The draft genome sequence of H. anticariensis FP35 T was analysed using the RAST Annotation Server [48] and the Integrated Microbial Genomes Database [49]. All the genes found in the H. anticariensis FP35 T genome that coded putative chemosensory related proteins were compared to the well described E. coli chemotaxis proteins (CheA, CheW, CheR, CheB, CheY, CheZ and MCPs) using NCBI available sequences.

Cloning of CheA ORF in a Multicopy Plasmid
The chromosomal cheA gene of H. anticariensis FP35 T was disrupted in a single step by Campbell-type integration of the entire plasmid by homologous recombination using the suicide plasmid pVIK112 as previously described by our research group [25]. For the complementation analysis of mutant FP35 ∆cheA, molecular biology experiments were performed according to standard protocols and the manufacturer's instructions. A DNA fragment flanking a cheA region of the 1922 bp gene was amplified from total DNA previously extracted from H. anticariensis FP35 T using the primers mcheA-EcoRI-Fw3 (5 -GGAAAGAATTCGGATGAAGCCTCCTATTCG-3 ) and mcheA-XbaI-Rev2 (5 -GCGTCTAGACTAGAACTCACTCCAGTCGTC-3 ), cloned into pCR-XL-TOPO (Thermo Fisher Scientific, Waltham, MA, USA) and then sequenced. A correct EcoRI/XbaI insert was isolated and subcloned into the multicopy plasmid pJN105 [50] which was previously digested with the same restriction enzymes. OmniMax E. coli electro-competent cells (Thermo Fisher Scientific, Waltham, MA, USA) were transformed by electroporation according to the manufacturer's instructions, and the transformants were selected in LB agar plates supplemented with gentamicin (20 µg mL −1 ). A correct pJN105-cheA clone and its corresponding empty vector control pJN105 were transferred by biparental mating to the mutant strain H. anticariensis FP35 ∆cheA using the E. coli β2163 as donor strain [51] as described elsewhere [52].

Extraction of Root Exudates from Salicornia Plants
Briefly, six-month-old plant roots were rinsed thoroughly with deionised water and were then placed in a black plastic wrapped beaker containing 200 mL Milli-Q water at 25-20 • C (day-night) during a 10-14 h light-dark period of illumination. After 24 h, the solution containing the released root exudates in the beaker was filtered through a 0.22 µm membrane and then stored at −20 • C [53].

Analysis of Salicornia Root Exudates Using UPL-HRMS Q-TOF
The root exudates were collected, freeze-dried, dissolved in 10% (w/v) methanol and analyzed using an ultra high-pressure liquid chromatograph (UPLC) (Acquity UPLC ® , Waters Corporation, Milford, MA, USA) coupled to a SYNAPT G2 Q-TOF high resolution mass spectrometer (HRMS; Waters Corporation, Milford, MA, USA). Mass spectrometry was carried out in positive ionization electrospray mode (ESI+). The data obtained were processed by MassLynx ™ software (Waters Corp., Milford, MA, USA). The growth of H. anticariensis FP35 T in presence or absence of the main component of root exudates was tested using the assay previously described in Section 2.2.

Qualitative Capillary Chemotaxis Assay
The qualitative capillary chemotaxis assay was carried out as described elsewhere [54]. Briefly, the washed cells were suspended in chemotaxis buffer (CB) to reach a final optical density of 0.1 (OD 600 ) and then placed in a chamber containing a U-shaped tube, microscope slide and cover slip. Microcapillary tubes (1 µL) sealed at one end were filled with the Salicornia root exudates extracted as previously described in Section 2.5 and 2% (w/v) low-melting temperature agarose and then inserted into the pool of bacterial cells. Negative controls (CB) were included in all experiments. The chemotactic response at the mouth of the capillary tube was visualized at 0 and 5 min using dark-field microscopy (Olympus IX73) with an Olympus TH4-100 halogen illuminator and was photographed using an Olympus DP73 CCD camera (version 1.8. software). Dark-field illumination was generated using a condenser with a contrast phase ring (Ph2) at an NA of 0.55 with an UPlanFLN 4x NA 0.13 objective. The images were processed by adding contrast and brightness using Adobe Photoshop Lightroom software. The data were normalized to time 0 using the Matlab R2013a program to obtain a heat map of bacterial chemotactic responses.

Quantitative Capillary Assay
A modified version of the capillary assay described by Adler [55] was used to quantify the chemotactic response of bacteria to a chemical gradient. Exponentially grown cells were harvested by centrifugation at 13,000 rpm for 5 min, washed and suspended in chemotaxis buffer (CB) containing 40 mM K 2 HPO 4 /KH 2 PO 4 , 0.05% (w/v) glycerol, 10 mM EDTA, pH 7.0 at 2, 5, 7.5 and 12.5% (w/v) sea salt solution to reach a final optical density of 0.1 (OD 600 ). Microcapillary tubes (1 µL) sealed at one end were filled with the Salicornia exudates extracted as previously described in Section 2.5 or high purity commercially available oleanolic acid from Sigma Aldrich and immersed in the cell suspension for 1 h. The tubes were then removed and their exterior was rinsed with sterile water. The contents of each tube were transferred to tubes containing CB by centrifugation (13,000 rpm), diluted (10 −3 and 10 −4 ) and then spread on appropriate MY agar plates to quantify the chemotactic response (CFU per capillary tube). Negative controls were performed in all experiments. The results were expressed as the increase in CFU for each compound tested and were normalized using CB. Data were expressed as the mean ± SEM of at least three independent experiments, each of which had three technical replicates.

Biofilm Formation Assay
The biofilm formation of H. anticariensis FP35 T and the mutant FP35 ∆cheA in the presence of root exudates was measured as described elsewhere [56]. Briefly, overnight cultures of each bacterium were grown at 32 • C up to OD 600 1.0. The cells were then diluted 1/100, and 96-well microtiter plates were filled with 100 µL of the dilution. Root exudates (10% and 25% (v/v)) or oleanolic acid (0.1 and 1 mM) were added to the medium in each well. Each treatment was replicated four times. After static incubation at 32 • C for 48 h, the biomass of biofilm formed by each bacterium was determined by a crystal violet assay at 0.1% (w/v) according to O´Toole et al. [56]. Biofilm formation was quantified by measuring the OD 590 for each well using a Tecan Sunrise microplate reader and XFluor V4.51 software.

Salicornia Seed Bacterization and Germination Assay
Salicornia seeds (provided by Alsa Garden, Niederhaslach, France) were surface-disinfected with 10% w/v NaClO solution for 10 min and with 70% v/v ethanol for a further 5 min. They were then washed three times with sterile distilled water. To check the efficiency of the sterilisation process, the seeds were placed on plates containing MY medium and incubated for 4 days. The seeds were bacterized according to the method described by Jha et al. with slight modifications [57]. Overnight cultures of H. anticariensis FP35 T and H. anticariensis FP35 ∆cheA were centrifuged at 10,000 rpm for 5 min, and each pellet was washed with PBS (pH 7.2). The pellets were re-suspended in PBS to reach a final optical density of 0.6 (OD 600 ). For bacterization, the sterilised seeds were immersed for 24 h in bacterial suspension containing 0.1% (w/v) carboxymethyl cellulose. As a control, seeds were suspended in PBS containing 0.1% (w/v) carboxymethyl cellulose without bacterial culture. Germination tests were performed on 30 bacterized seeds in sterilised petri dishes moistened with NaCl solution (0 and 0.25 mol L −1 ). All treatments were incubated at 27 ± 0.5 • C and 35 ± 1% humidity and carried out in triplicate. The final percentage of germination was determined after 30 days by measuring root and shoot length. The vigour index was calculated according to the formula described by Abdul Baki and Anderson [58]: vigour index = mean root length + mean shoot length × germination (%).
In order to quantify root bacterial colonization, one-month-old plant roots were rinsed thoroughly with deionised water, and 0.1 g of mixed root tips was ground with 1 mL sterilized distilled water using a mortar. The suspensions were diluted, plated on MY medium using 5% sea salt solution (w/v) and then counted after incubating the plates for 2 days at 32 • C.

Statistical Analyses
The data obtained were subjected to ANOVA analysis, and multiple pair-wise comparisons were made using the Tukey test.

Assessment of the Chemotactic Response of H. anticariensis FP35 T to Salicornia Root Exudates
Bacterial motility was assayed in order to select halophilic motile bacteria for further chemotaxis study. Twelve halophilic bacteria: Halomonas anticariensis FP35 T , H. rifensis HK31 T , H. elongata 1H9 T , H. almeriensis M8 T , H. cerina SP4 T , H. eurihalina FP6 T , H. ventosae Al-12 T , H. stenophila N12 T , H. ramblicola RS16 T , H. salina F8-11 T , H. organivorans G16.1 T and Pseudomonas halophila DSM 3051 were tested for motility. Seven of the bacteria tested showed some degree of swimming motility, and H. anticariensis FP35 T , which had the highest motility, was selected (data not shown).
In this study, we have analyzed the draft genome sequence of FP35 T and we have found one cluster of chemotaxis-related genes and 21 chemoreceptors. The gene cluster seems to govern the chemotactic behavior of FP35 T and it is very similar to the cluster from Escherichia coli ( Figure S1). To investigate the chemotactic response of the moderately halophilic bacterium H. anticariensis to Salicornia root exudates, chemotaxis experiments were performed using the wild-type strain and the mutant FP35 ∆cheA. The cheA gene of FP35 T coding for MCPs, which plays an important role in the transduction module of the chemotaxis pathway due to its homology with E. coli, was disrupted in a previous study [25]. The chemotactic response of H. anticariensis FP35 T and its mutant FP35 ∆cheA to Salicornia root exudates was first qualitatively analyzed in salt concentrations ranging from 2 to 12.5% (w/v) in order to determine the optimal saline concentration for bacterial chemotaxis. Of the different sea salt solution concentrations tested, the results indicate that the optimal concentration is 5% (w/v) for the chemo-attraction of wild-type cells to root exudates ( Figure 1). As expected, the FP35 ∆cheA mutant was observed to lack a chemotaxis response under all the experimental conditions analysed. Figure S2 shows the absence of a detectable chemotaxis response of FP35 T chemotaxis buffer (negative control) at the optimal 5% (w/v) concentration for bacterial chemoattraction. To confirm the role of cheA observed in chemotactic qualitative experiments, quantitative assays were carried out in the same range of salt concentrations ( Figure 2). Of the different sea salt solution concentrations tested, the results confirm that the optimal concentration is 5% (w/v) for the accumulation of wild-type cells in capillary tubes containing root exudates ( Figure 2). By contrast, the FP35 ΔcheA mutant showed an accumulation of below or close to zero under all the experimental conditions analysed. The negative CFU values at concentrations of 2 and 7.5% (w/v) reflect negative chemotaxis. The cheA required for positive chemotaxis was confirmed by the complementation of the FP35 ΔcheA mutant with a copy of the cheA gene in trans (pJN105-cheA; Figure 2). To confirm the role of cheA observed in chemotactic qualitative experiments, quantitative assays were carried out in the same range of salt concentrations ( Figure 2). Of the different sea salt solution concentrations tested, the results confirm that the optimal concentration is 5% (w/v) for the accumulation of wild-type cells in capillary tubes containing root exudates ( Figure 2). By contrast, the FP35 ∆cheA mutant showed an accumulation of below or close to zero under all the experimental conditions analysed. The negative CFU values at concentrations of 2 and 7.5% (w/v) reflect negative chemotaxis. The cheA required for positive chemotaxis was confirmed by the complementation of the FP35 ∆cheA mutant with a copy of the cheA gene in trans (pJN105-cheA; Figure 2).

Chemoattraction to Salicornia Root Exudates in a Variety of Halophilic Bacteria
After confirming the positive effect of Salicornia root exudates on the chemotactic response of H. anticariensis FP35 T , we hypothesized that these root exudates could also act as chemoattractants to other halophilic bacteria. A qualitative assessment of swimming motility showed an increase in halo size in the other four bacteria, H. ventosae Al-12 T , H. elongata 1H9 T , H. salina F8-11 T and Pseudomonas salina DSM 3051, all grown at an optimal 5% (w/v) salt concentration previously observed for H. anticariensis FP35 T . Only two of these bacteria, P. halophila DSM 3051 and H. salina F8-11 T , exhibited chemoattraction to Salicornia root exudates ( Figure 3). Nevertheless, Salicornia exudates elicited a lower chemotaxis response to these two strains as compared to that to H. anticariensis FP35 T .

Chemoattraction to Salicornia Root Exudates in a Variety of Halophilic Bacteria
After confirming the positive effect of Salicornia root exudates on the chemotactic response of H. anticariensis FP35 T , we hypothesized that these root exudates could also act as chemoattractants to other halophilic bacteria. A qualitative assessment of swimming motility showed an increase in halo size in the other four bacteria, H. ventosae Al-12 T , H. elongata 1H9 T , H. salina F8-11 T and Pseudomonas salina DSM 3051, all grown at an optimal 5% (w/v) salt concentration previously observed for H. anticariensis FP35 T . Only two of these bacteria, P. halophila DSM 3051 and H. salina F8-11 T , exhibited chemoattraction to Salicornia root exudates ( Figure 3). Nevertheless, Salicornia exudates elicited a lower chemotaxis response to these two strains as compared to that to H. anticariensis FP35 T .

Figure 3.
Qualitative capillary chemotaxis assays comparing responses of four halophilic bacteria to root exudates grown at a 5% (w/v) concentration of sea salt solution. All photographs were taken after 10 min. First column: dark-field images of cells gathered at the mouth of capillaries containing exudates; second column: heat map of normalized images (Matlab R2013a).

Identification of Salicornia Exudate Composition Using UPLC -HRMS Q-TOF
Electrospray quadrupole time-of-flight mass spectrometry (Q-TOFS MS) was used to identify metabolites produced by the exudates tested. Figure 4 shows the total ion chromatogram (TIC) spectrum of the Salicornia root exudates. Profiles of the root exudates from Salicornia plants demonstrate the presence of different peaks, the most abundant of which according to area was detected at a retention time of 14.16 min. This peak, corresponding to [M + H] + at m/z 455.3525, afforded the molecular formula C30H48O3 and was identified by LC-HRMS to correspond to oleanolic acid ( Figure 5).

Halomonas salina F8-11 T
Halomonas elongata 1H9 T Figure 3. Qualitative capillary chemotaxis assays comparing responses of four halophilic bacteria to root exudates grown at a 5% (w/v) concentration of sea salt solution. All photographs were taken after 10 min. First column: dark-field images of cells gathered at the mouth of capillaries containing exudates; second column: heat map of normalized images (Matlab R2013a).

Identification of Salicornia Exudate Composition Using UPLC -HRMS Q-TOF
Electrospray quadrupole time-of-flight mass spectrometry (Q-TOFS MS) was used to identify metabolites produced by the exudates tested. Figure 4        The concentration of this compound in the Salicornia root exudates sampled was 90 µM. An optimum growth of the bacterium FP35 T in presence of oleanolic acid (at concentration of 0.1 mM) was observed ( Figure S3). The bacterium growth pattern after 14 h revealed similar values in absence or in presence of this compound (OD 600 control: 1.006 ± 0.09 vs. OD 600 oleanolic acid, 0.1 mM: 1.16 ± 0.14).

Assessment of Chemotactic Response of H. anticariensis FP35 T to Oleanolic Acid
We evaluated the chemotactic responses of FP35 T to the predominant compound, oleanolic acid (Sigma-Aldrich, Missouri, USA), in Salicornia exudates. We carried out experiments at a 5% (w/v) concentration of a sea salt solution which generates higher background accumulation of wild-type cells in capillaries containing root exudates. The concentrations of oleanolic acid (0.1 and 1 mM) tested in this study were selected according to the positive chemotaxis response previously observed for Pseudomonas and Bacillus to other compounds detected in root exudates in this range of concentrations [18,59].
The wild-type strain FP35 T showed positive chemotaxis for oleanolic acid tested at concentrations of 0.1 mM and 1 mM (Figure 6), while the FP35 ∆cheA mutant showed negative CFU values, corresponding to negative chemotaxis. As expected, positive responses to oleanolic acid were restored in the complemented mutant FP35 ∆cheA pJN105-cheA ( Figure 6). Background accumulation of cells, at close to 1 × 10 3 , found in capillaries containing 1mM of oleanolic acid was similar to that in the capillaries containing Salicornia exudates. Negative chemotaxis response was observed for ethanol, the solvent used to dissolve oleanolic acid (data not shown). The concentration of this compound in the Salicornia root exudates sampled was 90 µM. An optimum growth of the bacterium FP35 T in presence of oleanolic acid (at concentration of 0.1 mM) was observed ( Figure S3). The bacterium growth pattern after 14 h revealed similar values in absence or in presence of this compound (OD600 control: 1.006 ± 0.09 vs. OD600 oleanolic acid, 0.1 mM: 1.16 ± 0.14).

Assessment of Chemotactic Response of H. anticariensis FP35 T to Oleanolic Acid
We evaluated the chemotactic responses of FP35 T to the predominant compound, oleanolic acid (Sigma-Aldrich, Missouri, USA), in Salicornia exudates. We carried out experiments at a 5% (w/v) concentration of a sea salt solution which generates higher background accumulation of wild-type cells in capillaries containing root exudates. The concentrations of oleanolic acid (0.1 and 1 mM) tested in this study were selected according to the positive chemotaxis response previously observed for Pseudomonas and Bacillus to other compounds detected in root exudates in this range of concentrations [18,59].
The wild-type strain FP35 T showed positive chemotaxis for oleanolic acid tested at concentrations of 0.1 mM and 1 mM (Figure 6), while the FP35 ΔcheA mutant showed negative CFU values, corresponding to negative chemotaxis. As expected, positive responses to oleanolic acid were restored in the complemented mutant FP35 ΔcheA pJN105-cheA ( Figure 6). Background accumulation of cells, at close to 1 × 10 3 , found in capillaries containing 1mM of oleanolic acid was similar to that in the capillaries containing Salicornia exudates. Negative chemotaxis response was observed for ethanol, the solvent used to dissolve oleanolic acid (data not shown).

Evaluation of H. anticariensis FP35 T Biofilm Formation in Response to Salicornia Root Exudates and Oleanolic Acid
To ascertain their impact on the biofilm formation of H. anticariensis FP35 T , 10 and 25% (v/v) concentrations of Salicornia root exudates were added to the cultures grown in 5% (w/v) sea salt solution. The results indicate that both these concentrations of root exudates significantly increased the biofilm biomass of FP35 T (Figure 7). However, no statistically significant differences were observed in the chemotaxis mutant FP35 ∆cheA with respect to the two exudate root concentrations tested. This suggests that an active chemotaxis system is required in FP35 T order to promote biofilm formation by Salicornia root exudates. Because oleanolic acid is the predominant compound in Salicornia exudates, its impact in the biofilm formation of FP35 T was also evaluated. Changes in biofilm in this case were not observed because no significant differences were found between the biofilm formation in the presence of this compound and in the presence of ethanol (the compound used to dissolve this synthetic compound with limited solubility) (OD 540 /OD 600 control ethanol, 0.1 mM or 1 mM: 1.29 ± 0.65; 1.23 ± 0.10 vs. OD 540 /OD 600 oleanolic acid, 0.1 mM or 1 mM: 0.57 ± 0.14; 1.11 ± 0.44).

Evaluation of H. anticariensis FP35 T Biofilm Formation in Response to Salicornia Root Exudates and Oleanolic Acid
To ascertain their impact on the biofilm formation of H. anticariensis FP35 T , 10 and 25% (v/v) concentrations of Salicornia root exudates were added to the cultures grown in 5% (w/v) sea salt solution. The results indicate that both these concentrations of root exudates significantly increased the biofilm biomass of FP35 T (Figure 7). However, no statistically significant differences were observed in the chemotaxis mutant FP35 ΔcheA with respect to the two exudate root concentrations tested. This suggests that an active chemotaxis system is required in FP35 T order to promote biofilm formation by Salicornia root exudates. Because oleanolic acid is the predominant compound in Salicornia exudates, its impact in the biofilm formation of FP35 T was also evaluated. Changes in biofilm in this case were not observed because no significant differences were found between the biofilm formation in the presence of this compound and in the presence of ethanol (the compound used to dissolve this synthetic compound with limited solubility) (OD540/OD600 control ethanol, 0.1 mM or 1 mM: 1.29 ± 0.65; 1.23 ± 0.10 vs. OD540/OD600 oleanolic acid, 0.1 mM or 1 mM: 0.57 ± 0.14; 1.11 ± 0.44).

Salicornia Plant Reactions to Inoculation with H. anticariensis strain FP35 T and its Mutant Strain FP35ΔcheA
The effect of the wild-type H. anticariensis strain FP35 T and its mutant H. anticariensis strain FP35 ΔcheA on the root length, shoot length, germination and vigour index of Salicornia was tested. An increase in the shoot and root length of Salicornia seeds was observed following inoculation with H. anticariensis strain FP35 T as compared to the non-inoculated control ( Table 1). The bacterization of seeds by the halophilic bacteria also significantly increased the percentage of germination. In stark contrast, seeds bacterized by H. anticariensis strain FP35 ΔcheA only boosted root growth, while shoot length and the percentage of germination were similar to those for the non-inoculated control (Table 1). Furthermore, the 153% increase in the vigour index caused by inoculation with the wild-type strain exceeded the 79% increase in that for the chemotaxis mutant. These results highlight the important role of chemotaxis in the PGPR activity of H. anticariensis strain FP35 T in Salicornia. Root colonization of the bacterial strains under seed bacterization conditions was tested, which showed a 2.8-fold increase in the H. anticariensis FP35 T colonization of Salicornia roots (6.3 × 10 5 ± 0.42)

Salicornia Plant Reactions to Inoculation with H. anticariensis strain FP35 T and its Mutant Strain FP35∆cheA
The effect of the wild-type H. anticariensis strain FP35 T and its mutant H. anticariensis strain FP35 ∆cheA on the root length, shoot length, germination and vigour index of Salicornia was tested. An increase in the shoot and root length of Salicornia seeds was observed following inoculation with H. anticariensis strain FP35 T as compared to the non-inoculated control ( Table 1). The bacterization of seeds by the halophilic bacteria also significantly increased the percentage of germination. In stark contrast, seeds bacterized by H. anticariensis strain FP35 ∆cheA only boosted root growth, while shoot length and the percentage of germination were similar to those for the non-inoculated control (Table 1). Furthermore, the 153% increase in the vigour index caused by inoculation with the wild-type strain exceeded the 79% increase in that for the chemotaxis mutant. These results highlight the important role of chemotaxis in the PGPR activity of H. anticariensis strain FP35 T in Salicornia. Root colonization of the bacterial strains under seed bacterization conditions was tested, which showed a 2.8-fold increase in the H. anticariensis FP35 T colonization of Salicornia roots (6.3 × 10 5 ± 0.42) CFUs g −1 dry root) as compared to colonization by the chemotaxis mutant (2.3 × 10 5 ± 0.29) CFUs g −1 dry root).

Discussion
Increase in soil salinity, which constitutes an enormous challenge for agriculture, needs to be tackled in order to come up with solutions to achieve sustainable crop production. One of the principal strategies for improving crop yields in saline soils is the use of PGPR bacteria for the promotion of halophyte growth in this type of soil. Rhizosphere colonization by PGPR bacteria is associated with root exudates [60] which play an important role in the rhizosphere and act as signals that mediate root-microorganism interactions [61]. Chemotaxis is a key player in initiating crosstalk between plant roots and PGPR bacteria [62]. Though extensively studied in model organisms such as Escherichia coli and Pseudomonas spp. [19], little is known about these chemotactic bacterial systems in halophilic bacteria such as those of the genus Halomonas [25][26][27], which have, however, been studied in relation to Salicornia root colonization [63]. We assayed the motility of bacteria belonging to the genus Halomonas that were isolated from the halophyte rhizosphere. H. anticariensis strain FP35 T , which had the highest motility of all the bacteria tested, was selected.
The halophilic bacterium H. anticariensis FP35 T has plant growth-promoting properties such as siderophore production, one of the principal advantages of this type of bacterium, which excludes other microorganisms and reduces competition for constituents of root exudates [64]. Strain FP35 T produces 2,3-butanediol during glucose fermentation, as well as nitrogenase and nitrate and nitrite reductase, and hydrolyses gelatine [15]. FP35 T also produces sulphated exopolysaccharides with many biotechnological applications [17], as well as quorum-sensing signalling molecules [16].
In a previous study, an identification of the draft genome sequence of H. anticariensis FP35 T was performed [65]. Analysing this genome sequence, we have found one cluster of chemotaxis-related genes and twenty-one chemoreceptors. This gene cluster seems to govern the chemotactic behavior of FP35 T and it is very similar to the cluster from Escherichia coli including the cheA gene. Similar results were previously described for this strain and other species from the genus Halomonas [27]. Given its importance in the chemotaxis pathway of other bacteria, the cheA gene, through its disruption in strain FP35 T , was found to play a crucial role in chemotactic responses to environmental hydrocarbon pollutants [25]. We carried out chemotaxis and biofilm formation experiments to investigate how Salicornia root exudates mediate the enhanced root colonization of H. anticariensis strain FP35 T . The chemotactic response of the halophilic bacterium FP35 T to Salicornia root exudates was analysed using qualitative and quantitative chemotaxis experiments. In addition, a complemented chemotaxis mutant strain FP35 ∆cheA pJN105-cheA was constructed in order to corroborate the chemotaxis results. Salicornia root exudates were observed to act as chemoattractants for strain FP35 T grown in concentrations of sea salt solution ranging from 2 to 12.5% (w/v), with the strongest chemotactic response found at 5% (w/v). These results were confirmed by the cheA mutant's lack of chemotaxis response and the chemoattraction of the complemented mutant strain FP35∆cheA pJN105-cheA. Interestingly, in a previous study, strain FP35 T showed the strongest chemoattraction to environmental pollutant compounds such as phenol and naphthalene when grown in a sea salt solution concentration of 7.5% (w/v) [25]. These findings point to the significance of salt concentrations in the chemotaxis system and of the chemotaxis responses of H. anticariensis FP35 T to different substances found in the soil. To the best of our knowledge, only few studies analyse the chemotactic behaviour of Halomonas strains using swimming assays with times of incubation higher than 24 h [27,66].
The chemotaxis response of FP35 T to root exudates found differed from that observed in this study for other species of the genus Halomonas spp. This indicates that the genus Halomonas spp. is comprised of bacteria with chemosensory signalling mechanisms of varying complexity, a conclusion previously reached in relation to species of the genus Pseudomonas spp. For example, although tetrachloroethylene acts as a chemoattractant for P. putida F1 and P. stutzeri OX1 [67,68], it has been found to be a chemorepellent for P. aeruginosa PAO1 [69,70].
Chemoattraction and biofilm formation by root exudates are associated with the colonization process [71]. Salicornia exudates act as chemoattractants for FP35 T and as stimulators of biofilm formation at sea salt concentrations of 5% (w/v), the optimum concentration for chemoattraction.
Previous studies have indicated that tomato and Arabidopsis root exudates increase biofilm formation in different species of the genus Bacillus [24,72]. Furthermore, a variety of exudate components such as citric acid, malic acid and fumaric acid have been reported to stimulate the biofilm formation of bacteria [59,71]. However, to our knowledge, little is known about the impact of halophyte exudates on Halomonas biofilm. As previous studies of other bacteria of the genus Bacillus have suggested [59,72], exudates could act as a carbon source and the growth of FP35 T on Salicornia exudates may alter the metabolism of this bacterium in a manner favourable to biofilm formation.
On the other hand, we characterized Salicornia exudates in order to determine which constituent is related to FP35 T chemoattraction. Plant root exudates are highly complex compound mixtures and vary according to the plant development stage and environmental conditions [73], although some of the constituents of these compound mixtures are unique to certain plants species [74]. Profiles of Salicornia exudates showed a high-intensity peak due to oleanolic acid (OA), a pentacyclic triterpenoid compound widely distributed in the plant kingdom. OA has a variety of biological functions in nature, with several promising pharmacological applications [75], plays an important role in regulating the virulence pathway of pathogens such as Ralstonia solanacearum [76], appears to be the principal component of Salicornia root exudates and acts as a chemoattractant for FP35 T . The levels of bacterial cell accumulation caused by OA were similar to those caused by chemotaxis to exudates, indicating that this organic acid may be the principal mediator of Salicornia plant-bacterium interactions.
Chemotactic reactions to OA were found to be positive at concentrations of 0.1 mM and 1 mM. At these concentrations, a positive chemotaxis response has also been observed to other compounds detected in root exudates such as the gamma-aminobutyric acid (GABA) chemoreceptor in Pseudomonas putida strain KT2440 [77], as well as oxalic and malic acids in Bacillus amyloliquefaciens SQR9 [59] and Pseudomonas fluorescens [18]. An unexpected chemorepellent response of the chemotaxis mutant FP35 ∆cheA to root exudates (at sea salt concentrations of 2 and 7.5%) and OA was observed. Considering that CheA is an essential protein in the chemotaxis signalling cascade in E. coli (bacteria model of chemotaxis signal transduction), the cheA mutant should be non-chemotactic at all. A previous study indicated that a cytoplasmic chemoreceptor, TlpD, in Helicobacter pylori form an autonomous signaling unit and mediate a repellent chemotaxis response to conditions that promote oxidative stress [78]. TlpD is able to localize to the pole and recruits CheW, CheA, and at least two CheV proteins to this location operating independently of other chemoreceptors to organize a chemotaxis signaling complex that mediate this chemorepellent response. The inactivation of cheA abolishes chemotaxis responses but could benefit, under certain conditions, the operating action of cytoplasmic chemoreceptors that could change the role of the principal kinases in the chemotaxis system. Future studies should be performed to increase the knowledge of cytoplasmic chemoreceptors in representatives of the genus Halomonas that could alter the function and the role of chemoreceptor sensor proteins like CheA.
In this study it was difficult to determine the biofilm formation of FP35 T caused by oleanolic acid. The levels of biofilm formation in the presence of this compound were similar to those caused by ethanol (used to dissolve this synthetic compound with limited solubility). It has been previously demonstrated that low concentrations of ethanol stimulate biofilm formation in other bacteria such as Pseudomonas aeruginosa [79] and this effect can be a problem when measuring the real biofilm formed by FP35 T .
The positive relationship between the PGPR H. anticariensis strain FP35 T and Salicornia root exudates could contribute to preferential colonization and stimulation of seed germination. In vitro experiments were carried out to examine the growth-promoting effects of strain FP35 T on Salicornia seeds under sterile conditions. Seeds bacterized with the halophilic bacterium FP35 T showed an increase in germination percentages, as well as shoot and root length, as compared to the non-inoculated control. Similar results were obtained by the inoculation of Salicornia seeds with halotolerant bacteria such as Brachybacterium saurashtrense and Pseudomonas sp. strain JG10 [53]. However, seeds bacterized with the chemotaxis mutant FP35 ∆cheA only showed an increase in root length as compared to the non-inoculated control. Deletion of the gene mcpG coding for a GABA chemoreceptor or the gene cheA3 coding for a histidine kinase in P. putida KT2440 reduced root colonization that requires chemotaxis [80,81].
This study indicates that the cheA gene coding for MCPs could play an important role in root colonization by FP35 T . Furthermore, the increase in biofilm formation, previously observed only in the wild-type strain in the presence of Salicornia exudates, could enable the bacterium to colonize the roots more effectively. Our results suggest that there is a connection between the absence of chemotaxis and the reduction in the vigour index and colonization of halophyte seeds.

Conclusions
This study demonstrates that Salicornia exudates act as chemoattractants for the halophilic bacterium H. anticariensis FP35 T . Our results also indicate that salt concentration influences chemotactic responses, with optimum levels of chemoattraction observed at a sea salt concentration of 5% (w/v). Oleanolic acid, the principal constituent of Salicornia exudates, appears to chemotactically mediate plant-bacterium interactions. The chemotactic responses and increased biofilm formation observed in this study suggest that halophyte exudates have the potential to expand the colonization of these salt-tolerant plants by halophilic bacteria. The enhancement of bacterial recruitment could be a potentially effective strategy for promoting halophyte growth and lead to major benefits for agriculture under saline conditions. Supplementary Materials: The following are available online at http://www.mdpi.com/2076-2607/8/4/575/s1, Figure S1: Chemosensory cluster identified in the draft genome sequence of H. anticariensis FP35 T . Figure S2: Qualitative capillary chemotaxis assays comparing responses of wild-type H. anticariensis FP35 T and the mutant strain FP35 ∆cheA towards chemotaxis buffer grown at a concentration of 5% (w/v) of sea salt solution. Figure S3: H. anticariensis FP35 T growth in M8 plates containing 0.3% agar in absence or presence of oleanolic acid (A and B, respectively).