Root exudates from citrus plants subjected to abiotic stress conditions have a positive effect on rhizobacteria

Plants are constantly releasing root exudates to the rhizosphere. These compounds are responsible of different (positive or negative) interactions with other organisms, including plants, fungi or bacteria. In this work, the effect of root exudates obtained from in vitro cultured citrus plants on two rhizobacteria (Pseudomonas putida KT2440 and Novosphingobium sp. HR1a) was evaluated. Root exudates were obtained from two citrus genotypes differing in their sensitivity to salt and heat stress and affected differentially the growth of both rhizobacteria. Root exudates from salt-stressed plants of C. macrophylla (salt tolerant) induced an increase in bacterial growth higher than that obtained from Carrizo citrange exudates (salt sensitive). Root exudates from heat-stressed plants also had a positive effect on bacterial growth, which was more evident in the heatsensitive C. macrophylla. These results reveal that the growth of these rhizobacteria can be modulated through citrus root exudates, and it can change depending on the stress conditions and the genotype. Biosensors P. putida KT2442 (pMIS5) and Novosphingobium sp. HR1a (pPAH) were used to test the presence of proline and salicylates in root exudates by measuring β-galactosidase activity. This activity increased in presence of root exudates obtained from stressed plants, and in a higher extent in the case of exudates obtained from the genotype resistant to each particular stress, indicating that those root exudates contain larger quantities of proline and salicylates, as it has been described previously. Our data reveals that both P. putida KT2442 (pMIS5) and Novosphingobium sp. HR1a (pPAH), could be used as biosensors of plant stress.


Introduction
Plants release between 5 and 25% of net fixed carbon into the rhizosphere in the form of compounds ranging from simple organic anions to complex polymer mucilages (Bais et al., 2006). Those specialized metabolites, produced and secreted by plants, play a critical role in the interaction of plants with soil organisms at the root vicinity, generally increasing the total quantity and activity of microbes around plant roots. These microorganisms use the compounds released by the plants through their roots as major nutrient sources, mainly carbon and nitrogen, constituents of sugars and amino acids, which are the compounds released to the rhizosphere in higher quantities (Lugtenberg et al., 1999;Moe, 2013). This is the case of the plant growth promoting rhizobacteria (PGPR) Pseudomonas putida, which growth is positively affected by proline, one of the amino acids largely released to the rhizosphere (Vílchez et al., 2000b). In addition, flavonoids, anthocyanins or salicylates, present in root exudates in lower quantities can affect rhizosphere biotic composition (Badri and Vivanco, 2009;Cesco et al., 2012;).
Root exudation is highly influenced by various biotic and abiotic factors in the surrounding environment, which can lead to a significant shift in the rhizosphere microbiota (Kawasaki et al., 2016). Several studies carried out with herbaceous species, reveal that PGPR and arbuscular mycorrhizal fungi (AMF) alleviate the damage induced by abiotic stresses conditions such as drought, salinity, flooding, nutrient deprivation, heavy metal or high temperatures in plants (Ali et al., 2011). The association with the bacteria Scytonema hofmanni alleviates the adverse effects of salt stress in rice plants (Rodríguez et al., 2006). Meanwhile, Burkholderia phytofirmans PsJN improves shoot and root growth and tuberization in potato plants cultured, either in vitro or ex vitro, under heat stress conditions (Bensalim et al., 1998). Wheat plants inoculated with P. putida AKMP7 are less damaged by high temperatures than non-inoculated plants, showing higher levels of chlorophylls, sugars, proline, starch, proteins and amino acids, and a reduction in superoxide dismutase, ascorbate peroxidase and catalase activities (Ali et al., 2011). Among the mechanisms responsible of this protection against abiotic stresses are the bacterial production of indole acetic acid or nitric oxide, which stimulate root growth and development and facilitate nutrient fixation; the decrease of plant endogenous content of ethylene by the bacterial 1-aminocyclopropane-1-carboxylic acid (ACC) deaminase activity or by the induction of changes in the root cell wall or membrane, such as the production of biofilms (Dimkpa et al., 2009). Therefore, understanding the interactions among living organisms in the rhizosphere will have a great value in improving plant resistance to both biotic and abiotic stresses, which will ultimately result in an increase in crop productivity.
There is little information about the colonization of PGPRs in citrus plant rhizosphere.
Isolates belonging to genera Burkholderia, Pantoea, Pseudomonas, Bacilli, Painibacillus, and Serratia were found to be associated with citrus roots, being the soil bacteria population different in plants affected by huanglongbing (HLB) (Trivedi et al., 2011). The beneficial effect of P. putida in absence of any stress condition has been also described (Chiquito-Contreras et al., 2012). On the other hand, it has been reported that the AMF mitigates some biotic and abiotic stresses, including phytophthora infection (Watanarojanaporn et al., 2011), drought (Wu and Zou 2009), high salinity (Navarro et al. 2014;Satir et al. 2016;Zhang et al. 2017) or chilling (Wu and Zou 2010).
Among citrus genotypes there is a wide variability concerning to their tolerance to different abiotic or biotic stress conditions. Carrizo citrange is salt sensitive and heat stress tolerant, whereas Citrus macrophylla is salt tolerant and heat stress sensitive (Vives-Peris et al., 2017). In addition, our previous work demonstrates that citrus plants are capable to exude proline and salicylates as cinnamic acid or salicylic acid (SA) by their roots. These and other metabolites present in citrus root exudates could affect the growth of different bacteria present in the rhizosphere (Vílchez et al., 2000a;Vives-Peris et al., 2017;. In this work, the effect of root exudates of two citrus genotypes Carrizo citrange and Citrus macrophylla with contrasting tolerance to salt and heat stress on the bacteria P. putida KT2440 and Novosphingobium sp. HR1a has been studied. Although P. putida KT2440 has been previously reported as a PGPR (Planchamp et al., 2015), the role of Novosphingobium sp. HR1a as a PGPR is not clear, being limited to other strains of this genus (Zhang et al., 2016).
Seeds of both genotypes were peeled and disinfected for 10 min in a 0.5% (vol/vol) sodium hypochlorite solution containing 0.1% (vol/vol) Tween-20 wetting agent and rinsed three times with sterile distilled water. Seeds were sown individually in 25x150 mm culture tubes with 20 mL of germination medium (MS) consisting of Murashige and Skoog salt solution (Murashige and Skoog, 1962) and 3% of sucrose as carbon source.
The pH was set at 5.7 ± 0.1 with 0.1 N NaOH before autoclaving. The medium was solidified by the addition of 0.9 % agar (Conda, Madrid, Spain). The cultures were maintained at 25 ºC, first in darkness for two weeks and two more weeks with a photoperiod of 16 hours and illumination of 150 mmol m -2 s -1 . After that, 30 plants per treatment were transferred to liquid MS medium and roots were pruned in order to favor the development of new roots.
Twenty days after the transference to MS liquid media, plants with a well-developed root system were transferred to the exudation media, composed by sterile deionized water in control and heat-stressed plants. For salt stress treatments, 60 or 90 mM NaCl was added to the exudation medium. For heat stress, plants were cultured at 30 or 40 ºC with a 16 hours photoperiod. Control and salt-stressed plants were maintained at 25 ºC with the same photoperiod. Medium was collected after ten days of exudation, frozen with liquid nitrogen and stored at -80 ºC. The absence of contaminations in root exudates was tested by culturing a 20 µL aliquot in potato dextrose agar medium (Conda, Madrid, Spain).

Bacterial strains and plasmids
To study the effect of root exudates on rhizobacterial growth, P. putida strain KT2440 and Novosphingobium sp. strain HR1a were used. For the detection of proline and salicylates, β-galactosidase assays using P. putida KT2442 (pMIS5) and Novosphingobium sp. HR1a (pPAH) respectively were performed. In pMIS5 (Vílchez et al., 2000a) the PputA promoter was cloned before the 'lacZ promoterless gene coding for β-galactosidase, easily measurable, in plasmid pMP220 (Spaink et al., 1987) and in pPAH the PpahA promoter was controlling the expression of the 'lacZ gene (Segura et al., 2017).
Chloramphenicol (Cm), rifampicin (Rif) and tetracycline were added to the bacterial cultures at final concentrations of 30, 10 and 10 µg/mL respectively when it was necessary (Table 1).

Bacterial growth assays
The initial bacterial cultures were maintained overnight in 25 mL flasks containing 10 mL of lysogeny broth (LB) liquid medium (Bertani, 1951) at 30ºC in an orbital shaker at 200 rpm. These cultures were washed with sterile distilled water twice and adjusted to an initial OD660nm of the cultures of 0.1.
To perform the different experimental approaches, bacteria were cultured in glass tubes with 2 mL of liquid M8 minimal medium (Kohler et al., 2000) supplemented with 20 mM succinate as carbon source and root exudates at the original concentrations. For this, 15 mL of root exudates were freeze-dried and resuspended in 1.5 mL of sterile deionized water, being 10-fold diluted when applied to the bacterial medium. Mocks with bacteria growing in M8 minimal medium containing 20 mM succinate, and supplemented with 60 or 90 mM NaCl were used to consider the salt effect on the bacteria. Bacterial cultures with root exudates were incubated at 30 ºC in an orbital shaker at 200 rpm during 48 h.
Quantification of bacterial growth was performed both, by assessing OD660nm and by counting the colony forming units (CFU). OD at a wavelength of 660 nm was recorded with a 96 well microplate spectrophotometer (Sunrise, Tecan, Männedorf, Switzerland).
Results were expressed as the variation of OD660nm in comparison with the respective mock and referred to the root fresh weight (Doornbos et al., 2011;Neal et al., 2012).
CFUs were determined by plating appropriate bacterial culture dilutions on LB medium supplemented with Cm and Rif for P. putida KT2440 and Novosphingobium sp. HR1a cultures respectively. CFUs were counted after 24 h of incubation at 30°C (Goldman and Green, 2008).

Induction of the PputA and PpahA promoters by different metabolites
P. putida KT2442 (pMIS5) and Novosphingobium sp. HR1a (pPAH) were cultured in M9 minimal media plus 20 mM succinate as carbon source in the presence of different chemicals related to proline and salicylate. Specifically, tubes containing 2 mL of M9 medium plus 20 mM succinate supplemented with L-proline or hydroxyl-L-proline (0.1 and 1 mM) were inoculated with P. putida KT2442 (pMIS5) to an initial OD660nm of 0.1.
Tubes with 2 mL of M9 medium plus 20 mM of succinate supplemented with 0.1 or 1mM of different salicylic acid biosynthesis and conjugation pathways compounds (methyl salicylate, sodium salicylate, L-phenylalanine, sodium benzoate, p-coumaric acid and tcinnamic acid; Fig. Sup. 1) were inoculated with Novosphingobium sp. HR1a (pPAH) to an initial OD660nm of 0.1. Tubes were cultured at 30 °C in an orbital shaker and β-Galactosidase activity was measured after 7 and 24 hours as in Miller (1972).

Induction of the PputA and PpahA promoters by root exudates
β-Galactosidase activity was measured in P. putida KT2442 (pMIS5) and Novosphingobium sp. HR1a (pPAH) bacteria cultured in M8 minimal medium supplemented with 20 mM succinate and root exudates. β-Galactosidase activity was determined as described in Miller (1972). Finally, results of β-galactosidase activity were calculated in Miller units, and the results were normalized according to the root fresh weight.

Statistical analyses
Statistical analyses were assessed with the Statgraphics Plus v.5.1. Software (Statistical Graphics Corp., Herndon, VA, USA). Data are means of three independent replications and were subjected to one-or two-way analysis of variance (ANOVA) and a Tukey posthoc test (p ≤ 0.05) when statistical significant differences were detected.

Bacterial tolerance to saline conditions
The addition of NaCl had a negative effect on the turbidity of cultures of both bacterial strains. After 24 hours, P. putida KT2440 showed to be more sensitive to saline stress, being unable to reach the OD recorded in absence of NaCl at the used NaCl concentrations (a decrease about the 35% with respect to the control without NaCl at 60 or 90 mM was observed). In the case of Novosphingobium sp. HR1a, the severity of the imposed stress (60 or 90 mM NaCl) conditioned the turbidity in comparison with cultures grown in absence of NaCl. The diminution of the OD660nm in cultures with 90 mM NaCl was lower than in the case of P. putida KT2440 being in Novosphingobium sp. HR1a the decrease only of 19.07% after 24 hours. Moreover, the turbidity of Novosphingobium sp. HR1a cultures was only affected after 24 hours in presence of 90 mM NaCl, whereas the turbidity of the cultures with 60 mM NaCl was similar to the observed in control conditions (Fig. 1A).
Similar results were observed in the number of CFU after 6 hours of culture, when a decrease of this parameter was observed in cultures of P. putida KT2440, with a diminution of CFU values about 52 and 99% in bacteria grown with 60 and 90 mM, respectively, in comparison to control cultures, while no differences were observed in cultures of Novosphingobium sp. HR1a in presence of NaCl (Fig. 1B).

Bacterial growth in presence of root exudates from salt-stressed citrus plants
Root exudates obtained from Carrizo citrange did not significantly promoted the growth of P. putida KT2440 or Novosphingobium sp. HR1a (Fig 2A and 2C); furthermore, the number of P. putida KT2440 CFUs decreased with exudates from Carrizo plants stressed with 90 mM NaCl (Fig 3).
Interestingly, when cultures of the two strains were supplemented with root exudates obtained from salt-stressed Macrophylla plants, a significant increase in turbidity was observed when compared those obtained from non-stressed plants ( Fig. 2B and 2D). The increase in turbidity was higher with exudates from plants treated with 90mM NaCl than with those from plants treated with 60 mM NaCl, at least during the first 10 hours. The increase in turbidity between the cultures with root exudates from control plants and the culture with root exudates from plants treated with 90 mM NaCl 6 h after inoculation ( Fig. 2B) correlates with an increase of two orders of magnitude in the number of P. putida KT2440 CFUs (Fig. 3). However, an increase of turbidity of 0.5 OD660nm per gram of root in Novosphingobium sp. HR1a (t = 6h) the increase in the number of CFU was only of one order of magnitude (Fig. 3).

Bacterial growth in presence of root exudates from heat-stressed citrus plants
Exudates obtained from plants stressed at 40ºC were able to support growth of both strains (Fig 4), except exudates from Carrizo plants with Novosphingobium sp. HR1a (Fig. 4C).
The effect of the exudates from heat-stressed plants was more evident after 10 hours of growth. Accordingly with the increase in turbidity (Fig. 4), no differences in the number of P. putida KT2440 CFUs were detected with exudates from Carrizo plants at 24 hours ( Fig 5); while the number of CFUs increase more than one order of magnitude with exudates from Macrophylla plants stressed at 40ºC. At 24 h the increases in the turbidity observed in Novosphingobium sp. HR1a cultures with exudates from both treatments were not high enough as to reflect any increases in the number of CFUs.

Differential induction of the PputA and PpahA promoters involved in proline and salicylates detection
P. putida KT2442 (pMIS5) was previously reported to be able to detect the amino acid proline (Vílchez et al., 2000b) and Novosphingobium sp. HR1a (pPAH) detects salicylate and different polycyclic aromatic hydrocarbons (Segura et al., 2017). As proline, salicylate and related compounds can be found in the rhizosphere, we tested the βgalactosidase activity of these biosensors in response to different commercial chemicals.
In cultures of P. putida KT2442 (pMIS5), β-galactosidase activity was affected in the presence of L-proline and hydroxy-L-proline. L-proline increased the β-galactosidase activity around 7-and 12-fold at 0.1 and 1 mM concentration respectively. Induction with hydroxyl-L-proline was only detectable at 1 mM (almost 3-fold increase) and only after 24 hours (Fig. 6A).
These experiments demonstrated that P. putida KT2442 (pMIS5) and Novosphingobium sp. HR1a (pPAH) are able to detect compounds present in the medium.

β-Galactosidase activity in presence of citrus root exudates
Proline, hydroxy-L-proline, salicylate, methyl-salicylate and benzoate were barely detected in Carrizo exudates in presence of exudates from salt-stressed plants, and significant increases in β-galactosidase activity were not detected after 7 hours (Fig. 7).
However, these compounds were detected in Macrophylla exudates at higher levels in those obtained from 60 or 90 mM NaCl stressed plants, reaching 6 and 7-fold increases in P. putida (pMIS5) and 7 and 1.4-fold increase in the case of Novosphingobium sp.
In presence of exudates from heat-stressed plants, β-galactosidase activity did not show big increases after 24 hours; although there was a tendency toward higher β-galactosidase activity levels in Carrizo exudates from plants stressed at 40ºC but no significant differences in the expression of the respective promoters of the biosensors were observed (Fig. 8).

Discussion
In this work, the effect of root exudates from citrus plants subjected to salt or heat stress on plant rhizobacteria P. putida KT2440 and Novosphingobium sp. HR1a has been studied. The work was performed using root exudates from two genotypes differing in their tolerance to salt and heat stress: while Carrizo is a salt-sensitive but heat-tolerant The presence of 60 and 90 mM NaCl in the culture medium has a detrimental effect in the survival of P. putida KT2440 and Novosphingobium sp. HR1a but especially in P. putida KT2440 (Fig. 1). Novosphingobium sp. HR1a was originally isolated from seaside soils which commonly present high salt levels, and it has been reported that this bacteria is also able to growth on marine artificial media (Segura et al., 2017). Our results sustained that this strain has developed adaptive mechanisms to tolerate high salinity.
These results are in concordance with other works which confirm that size and activity of the soil microbial community is reduced under soil salinity and sodicity induced by irrigation, resulting in a reduction in the soil organic matter decomposition and the mineralization of carbon, nitrogen, sulphur and phosphorus (Rietz and Haynes, 2003).

Therefore, understanding the interaction among plants and microbes under salt stress
conditions is of great interest to evaluate the soil quality, especially in agricultural soils of the Mediterranean area. We have observed that exudates from Macrophylla saltstressed plants are able to promote the growth of P. putida KT2440 and Novosphingobium sp. HR1a, on the contrary exudates from Carrizo salt-stressed plants did not promote bacterial growth (Figs. 2 and 3). Moreover, in presence of exudates from Macrophylla salt-stressed plants, the promotion of the development in Novosphingobium sp. HR1a was higher than in P. putida KT2440, which could be due to the higher tolerance of this strain to high salinity.
Interestingly, in root exudates from the salt-tolerant genotype Macrophylla the amount of proline and/or hydroxyl-L-proline (detected with the biosensor P. putida KT2442 (pMIS5)) and the amount of salicylates (detected with the biosensor Novosphingobium sp. HR1a (pPAH)) is higher than in exudates of Carrizo; furthermore, the induction of the promoters of both biosensors is higher in exudates from Macrophylla plants treated with salt, suggesting a higher exudation of these compounds under salt stress (Fig. 7). Previous studies reported an increase of proline and salicylates content in citrus root exudates from plants subjected to salt stress (Vives-Peris et al., 2017), which is in concordance with the measurements with these biosensors, suggesting that these biosensors could be effectively used to detect exuded proline and salicylates. These compounds can be used as nutrients by soil microorganisms and therefore, the higher growth supported by Macrophylla exudates could be due to the presence of these nutrients that are not present (or present at lower levels) in Carrizo exudates. Root exudates from Glycine max (soybean) and Phaesolus vulgaris (common bean), have been also described as promoters of Chryseobacterium balustinum growth (Dardanelli et al. 2010(Dardanelli et al. , 2012. However, we cannot exclude the effect of other phytohormones or metabolites present in the exudates of the plant. Previous studies with citrus plants subjected to different abiotic stresses as drought or high temperatures, have demonstrated that tolerant rootstocks have an enhanced antioxidant system (Zandalinas et al., 2017). In addition, tolerant citrus plants subjected to salt stress maintain a higher photosynthetic rate than sensitive plants, being this parameter directly linked with the exudation rate. The exudation of metabolites obtained from the C fixed in the photosynthesis which could be used as nutrient sources by PGPRs would be increased (Bais et al., 2006;López-Climent et al., 2008). Finally, the higher concentration of antioxidant compounds in tissue which could be related to their higher exudation, promoting the growth of PGPRs and playing an important role in the mutualism among citrus and PGPRs, which could contribute to a better colonization by these bacteria and the alleviation of salt stress to the plant, as it has been reported that P. putida and Novosphingobium sp. are able to mitigate the adverse effects of this stress in herbaceous plants (He et al., 2017;Krishnan et al., 2017;Yao et al., 2010).
When root exudates were obtained from heat-stressed plants, those obtained from plants subjected to 30 ºC did not affect the growth of either P. putida KT2440 or Novosphingobium sp. HR1a, but root exudates from plants subjected to 40 ºC stimulate growth of both strains, mainly in presence of root exudates from the heat sensitive Macrophylla (Fig. 4). The lack of differences in the growth of both strains in presence of root exudates from plants subjected to 30 ºC may be due to the fact that this temperature is not high enough to induce a severe stress in plants (considering that citrus is a subtropical crop) and consequently it does not affect root exudation as much as salt stress or higher temperatures such as 40 ºC (Vives-Peris et al. 2017). Moreover, the higher sensitivity to this stress condition could explain an earlier (in a lower threshold temperature) modification of root exudate composition in Macrophylla plants. In addition, the higher growth of both strains in presence of exudates from salt-stressed plants in comparison with those obtained from heat-stressed ones can be due to the different character of each adverse condition: whereas salt stress affects first to the root system, originating direct changes in this organ, heat stress affects mainly to the canopy, having the soil a buffer effect that attenuates the injuries to the root system (Zhang et al., 2005). Furthermore, we have not detected significant changes in proline and salicylates in exudates from heat-stressed plants, suggesting that the growth promotion in this case is mediated by different mechanisms than in the case of exudates obtained from saltstressed Macrophylla plants (Fig. 8). This is also supported by the delay in the growth promotion induced by exudates from heat-stressed plants.

Conclusions
To conclude, this work reveals that citrus root exudates can modulate the growth of the rhizobacteria P. putida KT2440 and Novosphingobium sp. HR1a, and it is affected by the genotype and abiotic stress conditions, promoting generally an increase of their growth when they are obtained from 60 or 90 mM NaCl, or 40 ºC stressed plants. Moreover, root exudates from plants subjected to stress conditions tend to induce the expression of putA and pahA genes in both bacterial strains, indicating that they are able to detect proline and salicylates present in root exudates. The concordance between β-galactosidase activity and proline and salicylates contents quantified by other methods reveals that P. putida KT2442 (pMIS5) and Novosphingobium sp. HR1a (pPAH) can be respectively used for the detection of proline and salicylates in root exudates. On the other hand, this work reveals the importance of root exudates in the growth of PGPRs, which could be used as a fertilizer to improve PGPR colonization and therefore trigger plant benefits, especially under abiotic stress conditions.

References
Ali, S.Z., Sandhya, V., Grover, M., Linga, V.R., Bandi, V., 2011. Effect of inoculation with a thermotolerant plant growth promoting Pseudomonas putida strain AKMP7 on growth of wheat (Triticum spp.) under heat stress. J. Plant Interact. 6, 239-246. Figure 1. Tolerance of P. putida KT2440 and Novosphingobium sp. HR1a to salt conditions. A: Turbidity at OD660nm of cultures of P. putida KT2440 and Novosphingobium sp. HR1a in presence of 0, 60 and 90 mM NaCl in M8 medium plus 20 mM succinate; white circles refer to control; black circles refer to 60 mM NaCl and black triangles refer to 90 mM NaCl treatments. B: Number of CFU of cultures of P. putida KT2440 and Novosphingobium sp. HR1a in presence of 0, 60 and 90 mM NaCl in M8 medium plus 20 mM succinate, after 6 hours; white bars refer to control at 0 hours; light grey, dark grey, and black bars refer to cultures in presence of 0, 60 and 90 mM, respectively. Values indicate the mean of three replicates ± standard error. Different letters to statistically significant differences at P ≤ 0.05.   Novosphingobium sp. HR1a (C and D) in M8 medium with 20 mM succinate. White circles refer to control; black circles and black triangles refer to root exudates from plants subjected to 30 ºC and 40 ºC, respectively. Values indicate the mean of three replicates ± standard error. Figure 5. Effect of root exudates from Carrizo citrange and C. macrophylla heat-stressed plants in the increase of the number of colony forming units of P. putida KT2440 and Novosphingobium sp. HR1a in M8 medium with 20 mM succinate after 24 hours. White bars refer to control; grey bars refer to cultures with root exudates from 30 ºC stressed plants, and black bars refer to cultures with root exudates from 40 ºC stressed plants.

FIGURE AND TABLE CAPTIONS
Values are normalized according to root fresh weight and indicate the mean of three replicates ± standard error. Different letters to statistically significant differences at P ≤ 0.05. Figure 6. β-Galactosidase activity of cultures of P. putida KT2442 (pMIS5) and Novosphingobium sp. HR1a (pPAH) with different commercial standards. A: β-Galactosidase activity of cultures of P. putida KT2442 (pMIS5) supplemented with L-Proline (L-Pro) and L-Hydroxyproline (L-Hyp). B: β-Galactosidase activity of cultures of Novosphingobium sp. HR1a (pPAH) supplemented with methyl salicylate (Me-SA), sodium salicylate (Na-SA), L-phenylalanine (Phe), sodium benzoate (Na-Benz), pcoumaric acid (p-coum) and t-cinnamic acid (t-cin). Commercial standards were added at concentrations of 0.1 mM (white bars) and 1 mM (grey bars), and β-galactosidase activity was measured after 7 (non-lined bars) and 24h (lined bars). Values are referred to mock culture and indicate the mean of three replicates ± standard error. Figure 7. β-Galactosidase activity of P. putida KT2442 (pMIS5) and Novosphingobium sp. HR1a (pPAH) grown with root exudates from citrus plants subjected to salt stress for 10 days. β-Galactosidase activity was measured at 7 hours. White bars refer to control; grey bars refer to 60 mM NaCl, and black bars refer to 90 mM NaCl. Values indicate the mean of three replicates ± standard error. Figure 8. β-Galactosidase activity of P. putida KT2442 (pMIS5) and Novosphingobium sp. HR1a (pPAH) grown with root exudates from citrus plants subjected to heat stress for 10 days. β-Galactosidase activity was measured at 24 hours. White bars refer to control, grey bars refer to 30 ºC, and black bars refer to 40 ºC. Values indicate the mean of three replicates ± standard error. Table 1. Bacteria, plasmids and plants used. Ap r , Cm r , Rif r and Tc r refer to resistance to ampicillin, chloramphenicol, rifampicin and tetracycline respectively.