- Top of page
- Materials and Methods
- Supporting Information
Drought is one of the major limitations to food production worldwide. The development of drought-resistant cultivars and water-use-efficient plants is therefore of global concern. In their habitats, plants are not single organisms but are surrounded by dense populations of diverse microorganisms with which they probably interact. Some of these plant–microorganism interactions are beneficial for plant growth and allow plants to better cope with biotic and abiotic stresses (Yang et al., 2009).
Drought periods lead to large physiological and developmental alterations in plants. Water deprivation decreases above-ground vegetative biomass accumulation and therefore reduces plant performance (Boyer & Westgate, 2004; Hummel et al., 2010; Tardieu et al., 2011; Vile et al., 2012). Vegetative growth and the total production of dry matter are closely related to key developmental switches such as reproductive transition (Jung & Muller, 2009). Specifically, flowering time can be delayed (McMaster et al., 2009; Tisne et al., 2010) or hastened (Verslues & Juenger, 2011) in response to drought, most likely depending upon plant species and the occurrence, duration and severity of the stress.
In order to minimize the negative effects of water deficit and complete their life cycle under unfavourable conditions, plants exhibit a variety of strategies (for reviews, see Farooq et al., 2009; Verslues & Juenger, 2011). Physiological changes occur rapidly after the onset of water deficit in order to maintain high tissue water potential. One of the swiftest responses is a reduction of transpiration through reduced leaf conductance following stomata closure. This response is often associated with an accumulation of ABA or enhancement of sensitivity to this hormone in the leaf cells, leading to the induction of related signalling genes (Harb et al., 2010). Rapid osmotic adjustment through active accumulation of solutes also helps in maintaining cell turgor and increases the driving force of water influx into the cell (Yoshiba et al., 1997). In the long term, increased root-to-shoot ratio, through reduced above-ground growth and/or increased root growth, participates in reducing evaporative area and increasing water absorption capacities from the soil (Boyer, 1985), together contributing to increased water-use efficiency (WUE). WUE reflects the tradeoff between CO2 acquisition for growth and water losses and is therefore an important indicator of how plants manage water stress (Blum, 2005; Tardieu, 2012).
Soil microorganisms may interact with plant-specific mechanisms related to drought resistance. Some naturally occurring free-living soil bacteria, namely plant growth-promoting rhizobacteria (PGPR), colonize the root system and maintain mutualistic interactions that lead to plant growth improvement and plant protection against multiple stresses, including drought, salt, heavy metals or pathogens (Dimkpa et al., 2009; Lugtenberg & Kamilova, 2009; Yang et al., 2009). PGPR such as Azospirillum, Azotobacter and Pseudomonas fluorescens are well known for their plant growth-promoting effects and are notably used for improving crop yields (Kloepper et al., 1989; Lucy et al., 2004). PGPR effects involve multiple changes in plant metabolism and signalling networks (Lugtenberg & Kamilova, 2009; Friesen et al., 2011). Modifications in phytohormone content and/or signalling have been reported (see, for review, Dodd et al., 2010), such as decreased ethylene production via bacterial ACC deaminase activity (Glick et al., 1998; Belimov et al., 2009), changes in cytokinin–ABA balance (Figueiredo et al., 2008; Cohen et al., 2009) or changes in auxin signalling (Persello-Cartieaux et al., 2003; Contesto et al., 2010). These effects on hormone pathways are likely to interfere with plant tolerance to drought stress. Some PGPR strains improve plant enzyme activity, such as catalase or superoxide dismutase, which alleviates the oxidative damage induced by drought (Kohler et al., 2008; Wang et al., 2012). Finally, PGPR have been shown to increase drought-response transcript abundances (Wang et al., 2005, 2012).
Despite strong evidence that PGPR influence overall plant performance, their detailed effects on development, growth and physiology under drought have been less well explored. Therefore, integrative studies to explain how PGPR can improve drought tolerance are lacking. Among the specific PGPR-mediated mechanisms identified is the enhancement of wheat growth by Azospirillum sp. strains under various drought intensities, which was associated with better maintenance of plant water status as a result of increased cell wall elasticity (Creus et al., 2004). An increase of photosynthetic capacity has also been shown in Pinus halepensis inoculated with P. fluorescens (Rincon et al., 2008) or in Azospirillum-inoculated rice (Ruiz-Sanchez et al., 2011). Although these physiological studies have detailed measurements of plant water relations, most failed to report drought effect on the dynamics of plant development. Moreover, most studies focus on a single time point, generally at flowering or seed maturity, and reports on plant growth throughout the whole plant cycle are very scarce.
Here, we investigated the growth and physiological responses of Arabidopsis thaliana inoculated with a free-living PGPR, Phyllobacterium brassicacearum strain STM196 under long-term water deficit. A. thaliana is a useful organism to study plant interactions with PGPR (Ryu et al., 2005; Desbrosses et al., 2009), but the effects of PGPR on the development or physiology of this model species under water stress have been little investigated. STM196 belongs to the Phyllobacteriaceae family in the Rhizobiales order of α-Proteobacteria (Mantelin et al., 2006). This strain was the most efficient PGPR isolated from the rhizoplan of field-grown Brassica napus roots (Bertrand et al., 2001; Larcher et al., 2003).
Previous in vitro studies showed that STM196 enhances shoot and root growth of A. thaliana, and modifies its root architecture and hormonal signalling (Mantelin et al., 2006; Contesto et al., 2010; Galland et al., 2012; Kechid et al., 2013). However, we lack information on plant–bacteria interactions under soil conditions and no study has investigated the effects of this particular strain on plant response to drought. We used the high-throughput plant phenotyping platform PHENOPSIS (Granier et al., 2006) to decipher the effects of STM196 on multiple plant traits related to growth dynamics, development and physiology under well-defined soil water availability. First, we show that the plant growth-promoting effect of STM196 is related to a delay in reproductive transition in two A. thaliana accessions with contrasting flowering phenology. Then, we show that STM196 induces a suite of physiological and developmental changes that lead to enhanced WUE and to a better plant tolerance to water deficit.
- Top of page
- Materials and Methods
- Supporting Information
Plant growth-promoting rhizobacteria can enhance plant performance and plant tolerance to environmental stresses by a large variety of mechanisms (for reviews, see Lugtenberg & Kamilova, 2009; Friesen et al., 2011). These mechanisms have to be elucidated to design strategies for PGPR application in agriculture (for review, see Lucy et al., 2004). Here, we show that Phyllobacterium brassicacearum strain STM196, a PGPR isolated from the rhizosphere of oilseed rape B. napus (Bertrand et al., 2001; Larcher et al., 2003), enhances plant tolerance to drought in two accessions of A. thaliana with contrasting flowering phenology. We highlight a new means by which bacteria can enhance plant performance under both well-watered and drought soil conditions. Specifically, our results show for the first time a PGPR-induced delay in the transition from vegetative to reproductive development. Inoculated plants accumulated more biomass before reproduction and exhibited a better WUE.
STM196 induces a delay in reproductive timing that leads to increased biomass accumulation
The switch from vegetative to reproductive development is highly critical for wild and crop species. Indeed, the timing of flowering is a key event that determines the production of plant biomass and therefore yield (Jung & Muller, 2009). Floral transition can be influenced by abiotic changes in the environment, such as day length, ambient temperature and water availability (Bernier & Perilleux, 2005), and by endogenous stimuli (Huijser & Schmid, 2011). Multiple pieces of evidence support the fact that plant growth rate and the duration of the growth phases depend on flowering time. Alteration of flowering time by genetic modifications or photoperiod resulted in accelerated or decelerated progress towards the vegetative phases of Arabidopsis (Steynen et al., 2001) and others species (Salehi et al., 2005). In addition, close relationships among leaf production, individual leaf growth and flowering time have also been reported (Cookson et al., 2007). Here, we showed that Arabidopsis plants grown in soil inoculated by STM196 exhibit contrasting growth dynamics and phenology. Developmental changes induced by STM196 appeared very early during plant development. For instance, the phyllochron was increased in inoculated plants as early as the emergence of the two first leaves until the reproductive phase. The most noticeable phenological change was a significant delay in flowering time in inoculated plants. This delay coincided with a prolonged adult vegetative phase, as indicated by the postponed morphological transition between adult and cauline leaves. Moreover, inoculated plants exhibited a prolonged production of adult vegetative and cauline leaves that resulted in a higher number of both types of leaves. Previous studies reported that delaying flowering time as a result of shortening photoperiod coincided with a longer duration of leaf production and growth (Koornneef et al., 1998; Cookson et al., 2007). Here, the prolonged vegetative growth and the delayed flowering of PGPR-inoculated plants led to a greater production of vegetative and reproductive biomass. Strikingly, the direction of changes was similar in the two watering regimes, but the intensity of changes was more pronounced under drought and led to better plant tolerance to drought. PGPR effects on the timing of flowering are not common and we found no study reporting a PGPR-induced delay in flowering time. Growth promotion by rhizobacteria is often shown at a given date after germination or inoculation (Ryu et al., 2003; Jaleel et al., 2007; Zahir et al., 2008) and we lack a precise study of their effects on growth dynamics and development. Recently, it has been shown that A. thaliana plants inoculated with a naturally associated rhizobacterium, Pseudomonas sp., exhibited a faster rate of development – plants reached the floral transition earlier – and were bigger (Schwachtje et al., 2011), indicating that various PGPR strains mediate different plant responses.
Rhizobacteria often induce modifications in phytohormone signalling (for a review, see Yang et al., 2009), which may mediate effects on meristem activity and identity (Hayat et al., 2010). Our results showed that ABA was increased in STM196-inoculated plants. By contrast with other PGPR strains, STM196 is not a high auxin producer (Contesto et al., 2010) and, thus, cannot supply plant roots with extra auxin. However, it has been shown that inoculation with STM196 changed auxin distribution within Arabidopsis roots towards apices, which probably explains the positive effect of STM196 on lateral root development (Contesto et al., 2010). It is worth mentioning these effects of STM196 on auxin distribution because this hormone also plays a role in the regulation of leaf and floral initiation and of the position of lateral organs (Reinhardt et al., 2000). In addition, other hormonal pathways are modified by STM196, including ethylene, which participates in root hair elongation in vitro (Contesto et al., 2008; Galland et al., 2012). Further investigations are required to disentangle the interactions between signalling pathways that might explain the developmental changes following STM196 inoculation (Achard et al., 2006).
These results are novel in the context of plant–microorganism interactions and are promising for agronomy. Reducing or eliminating flowering by altering the endogenous mechanisms involved in the flowering pathway is one of the strategies to increase the yield of biomass crops (Jung & Muller, 2009). For instance, overexpression of the Arabidopsis floral repressor gene, FLOWERING LOCUS C (FLC), in tobacco resulted in a significant delay in flowering time and a concomitant increase in the biomass yield (Salehi et al., 2005). In vegetative crops such as cabbage (Brassica oleracea), early bolting and flowering limit the potential for yield increases (Jung & Muller, 2009). Therefore, manipulation of flowering time through rhizospheric flora can have important applications in stressed conditions, but underlying regulatory genes remain to be investigated.
The growth slowdown of STM196-inoculated plants, superimposed on that of water deficit, contributes to lifetime water economy and to increased drought resistance
Multiple combinations of traits can participate in plant strategies for dealing with drought, including those that allow drought escape or drought resistance (Verslues & Juenger, 2011). In addition, several soil microorganisms, including PGPR, can represent an added value to these strategies. For instance, some rhizobacteria help plants to maintain a favourable water status under water deficit (Creus et al., 2004), by enhancing the development of the root system (Marulanda et al., 2009). Here, we quantified numerous morphophysiological traits related to plant growth and development in order to decipher the added value of the PGPR STM196 to the drought response strategies of Arabidopsis. The automated phenotyping platform PHENOPSIS allowed the water limitation in the soil to be precisely controlled and maintained from as early as germination and up to the reproductive phase. Steady-state drought as applied here during the whole-plant cycle is highly relevant for the study of plant acclimation to drought (Verslues & Juenger, 2011). Acclimation processes during steady-state drought may reinforce plant resistance to this stress. Here, the soil water deficit was strong enough to cause an 80% decrease in above-ground FM of noninoculated plants at bolting. This biomass reduction is comparable to previous reports using a similar experimental procedure (daily irrigation to a steady-state soil water content) and similar intensities of drought applied to Col-0 (Hummel et al., 2010; Vile et al., 2012) and other accessions (Tisne et al., 2010).
Reduced plant size and total leaf area are common plant strategies to reduce water consumption and therefore drought injury (Tardieu et al., 2011). Indeed, we recently showed that the inherent size of various Arabidopsis ecotypes was negatively related to drought resistance (Vile et al., 2012), and that mutants that cope better with extreme stress often display a dwarfed stature (see references in Skirycz & Inze, 2010). Our results suggest that the growth slowdown of inoculated plants, superimposed on that of water deficit, has contributed to lifetime water economy and to increased drought resistance. In addition, as found in a previous study performed under similar drought scenarios, the reproductive timing under drought tended to occur earlier in Col-0 and later in An-1 (Vile et al., 2012). This illustrates the variability in drought response strategies in terms of reproductive phenology, and contrasts with the generally held view that drought escape is a common strategy of Arabidopsis (Verslues & Juenger, 2011). However, as in well-watered conditions, inoculated plants of both accessions exhibited delayed reproductive timing under drought. Inoculated plants accumulated twice as much biomass and produced more leaves of a larger area before flowering, had bigger reproductive stems and therefore higher expected reproductive yield. Several lines of evidence point to a higher survival and seed production of later-flowering Arabidopsis accessions (Korves et al., 2007). The timing of flowering often correlates with abiotic and biotic stress avoidance, which is frequently scored as a component of yield, for example in maize (Chardon et al., 2004). In addition, quantitative trait loci for adaptation to drought are often related to flowering time loci (Ducrocq et al., 2008). The advantage of a delay is that there is more time to accumulate more mass that can be invested towards seeds (Metcalf & Mitchell-Olds, 2009). Among the drawbacks of such a strategy, plants have to maintain a favourable use of water during a longer period, especially under drought conditions.
We showed that WUE of inoculated Col-0 plants was significantly improved under water deficit. We did not find any significant change in WUE in response to drought in noninoculated plants, in contrast to previous Arabidopsis studies that reported an increase in WUE (Juenger et al., 2005; Aubert et al., 2010). Interestingly, McKay et al. (2003) reported that higher WUE was genetically correlated with delayed flowering in Arabidopsis. Here, higher WUE of inoculated plants was mainly a result of a significantly lower water loss through daytime and night-time transpiration, which may reflect a better drought avoidance strategy. However, as reported by Westgate & Boyer (1985), a decrease in transpiration (by stomatal closure) can be followed on a longer timescale by a reduced plant growth rate, as was observed here in inoculated plants. We also reported a decline in photosynthesis in inoculated plants, but the sucrose content in leaves was increased regardless of the soil condition. This is in accordance with the literature reporting an increase in sucrose content in leaves even if CO2 diffusion is lowered under water deficit (Quick et al., 1992; Hummel et al., 2010). This could be a result of the uncoupling between photosynthesis and growth under water deficit (Muller et al., 2011). Higher concentrations of ABA in the leaves of inoculated plants can explain the lower transpiration rate resulting from stomatal closure. Some bacteria have the capacity to modulate gas exchanges and ABA metabolism. In A. thaliana, Zhang et al. (2008) interpreted the augmentation of photosynthetic rate in plants inoculated with Bacillus subtilis as being the result of decreased ABA concentrations in planta. In the common bean, Paenibacillus polymyxa and Rhizobium tropici coinoculation has been shown to decrease the ABA content in response to WD (Figueiredo et al., 2008). In addition to acting on the biosynthesis of ABA, some pathogenic bacteria can also modify stomatal opening by acting downstream of ABA biosynthesis. This is the case with Pseudomonas syringae, which exude coronatine, a substance that inhibits ABA signalling and prevents stomatal closure (Melotto et al., 2006). By contrast, Cohen et al. (2009) found a twofold increase in ABA concentrations in Azospirillum brasilense-inoculated Arabidopsis. Bacteria-induced increase of ABA content has been proposed to play a role in alleviation of the drought effect in maize (Cohen et al., 2009). Such an observation would be consistent with our results.
Roots also play a key role in WUE and adaptation to drought. Root biomass was higher in STM196-inoculated plants, and modifications of the root architecture as a result of the presence of the bacteria may have enhanced the water absorption capacity. Indeed, studies performed in gnotobiotic conditions showed that STM196 increased lateral root length (Mantelin et al., 2006; Kechid et al., 2013), and the density and length of root hairs (Galland et al., 2012). Both effects must lead to a greater exchange surface with soil and consequently higher water flux through the whole root system up to the leaves.
Overall, our results show that the PGPR P. brassicacearum STM196 induces a suite of developmental and physiological changes that represent a significant added value to the drought response strategy of Arabidopsis. Developmental and early growth slowed down, but prolonged vegetative growth and reduced transpiration contributed to increasing drought resistance and WUE. Prolonged vegetative growth and delayed flowering induced by PGPR are new in the context of plant–microorganism interactions and may be promising for agronomy. Reducing or eliminating flowering by altering the endogenous mechanisms involved in flowering is one of the strategies for increasing crop yield. Delaying flowering time by rhizobacteria inoculation could represent a valuable strategy for increasing biomass yield.