Effect of Burkholderia tropica and Herbaspirillum frisingense strains on sorghum growth is plant genotype dependent

Introduction

The plant sorghum (Sorghum bicolor) originated on the African continent and is cultivated worldwide (Rao et al., 2014). Sorghum has a short growth period, and is therefore a preferred cereal in arid and semi-arid regions (Farre & Faci, 2006; Wu et al., 2010; Funnell-Harris, Sattler & Pedersen, 2013). In Africa, sorghum is mainly cultivated by small farmers as staple food and for beverage production (Haiyambo, Reinhold-Hurek & Chimwamurombe, 2015). By contrast, sorghum is mostly used for the feed market in North and Central America and for animal feedstock, ethanol production, and soil coverage in South America (Dutra et al., 2013; Perazzo et al., 2013; Damasceno, Schaffert & Dweikat, 2014; Rao et al., 2014). Sorghum is currently the fifth most cultivated cereal worldwide (Ramu et al., 2013) and in 2014, approximately 71 million tons of sorghum grains were produced around the world (FAO, 2018).

Sorghum producers often face yield problems due to soil nutrient deficits, limited access to chemical fertilizers, and the frequent need to combat plant pathogens (Haiyambo, Reinhold-Hurek & Chimwamurombe, 2015). Although conventional agricultural methods, such as chemical fertilization and pesticide application, can be used to overcome these limitations, the environmental side effects of these practices may be unsustainable As an alternative, the use of plant growth-promoting bacteria (PGPB) as biofertilizers not only enhances plant biomass and nutrient uptake but also improves pathogen control (Bhattacharyya & Jha, 2012; Dawwam et al., 2013). PGPB can alter the root architecture and promote plant growth by directly facilitating nutrient acquisition or modulating plant hormone levels or by indirectly inhibiting pathogenic organisms (Bhattacharyya & Jha, 2012; Glick, 2012). The best-known processes of plant nutrient acquisition mediated by PGPB are nitrogen fixation, phosphate (P) solubilization and iron sequestration (Lucy, Reed & Glick, 2004). Different groups of bacteria produce plant growth regulators, such as cytokinins, gibberellins, indole-3-acetic acid (IAA), and ethylene, that may also affect the plant’s hormonal balance (Amara, Khalid & Hayat, 2015). Moreover, PGPB can promote plant growth by fixing N₂ and inhibiting plant pathogens by producing antibiotics or lytic enzymes or competing for resources, which can limit disease incidence and severity (Glick, 2012; Da Silveira et al., 2016). In addition to the bacterial modification of plant metabolism, plant exudates have the potential to modify rhizosphere microbial community assembly and interactions (Haichar et al., 2014; Vurukonda et al., 2016). Recent studies suggest that the plant hormone strigolactone (SL) plays an important role in plant rhizosphere bacterial community composition (Funnell-Harris, Pedersen & Marx, 2008; Schlemper et al., 2017). Furthermore, Peláez-Vico et al. (2016) showed that the PGPB Sinorhizobium meliloti reduces orobanchol and orobanchyl acetate levels in nodulated alfalfa plants under P starvation, suggesting a role of SL in rhizobial–legume interactions. In a in vitro assay, Peláez-Vico et al. (2016) demonstrated that swarming motility of Sinorhizobium meliloti is triggered by the synthetic SL analogue GR24.

Many aspects of the interaction between PGPB and plants have been addressed for a wide range of plant species. Specifically in sorghum, some studies have focused on the interaction of PGPB strains isolated from third-party host species as inoculants for the sorghum rhizosphere (Matiru & Dakora, 2004; Dos Santos et al., 2017), whereas other works have reported the inoculation in other plant species of bacterial strains isolated from sorghum.

Matching beneficial bacteria with their preferred crops might optimize root colonization and biocontrol (Raaijmakers & Weller, 2001), especially when different plants are cropped in soils with the same bacterial composition. In this context, it is extremely important to identify bacterial candidates that have similar growth effects on plants that share the same soil. In Brazil, sorghum have been planted during the sugarcane off season as well as in former sugarcane fields (May et al., 2013), and therefore sorghum and sugarcane are frequently exposed to the same soil. Endophytic bacteria with plant growth-promoting traits isolated from sugarcane have been shown to increase biomass and plant N content when inoculated in plantlets of sugarcane. For example, Govindarajan et al. (2006) observed an increase in sugarcane yield of 20%, while Sevilla et al. (2001) observed increases of 31% in plant dry matter, 43% in N accumulation, and 25% in productivity in two sugarcane varieties. We hypothesized that different bacterial species isolated from sugarcane will trigger different growth plant responses in different sorghum cultivars. Thus, to determine if endophytic strains characterized as PGPB in sugarcane can act as non-host-specific PGPB benefiting sorghum performance, we tested the effect of five bacterial strains on the plant biomass and root architecture of four Sorghum bicolor cultivars with different uses and characteristics: SRN-39, an African grain cultivar that produces high amounts of orobanchol; Shanqui-Red (SQR), a Chinese cultivar that produces high amount of 5-deoxystrigol; BRS330, a hybrid grain cultivar from Brazil and BRS509, a hybrid saccharin cultivar from Brazil that produces both orobanchol and sorgomol (Schlemper et al., 2017).

Inoculation of the cultivars SRN-39 and BRS330 with Burkholderia tropica or Herbaspirillum frisingense strains resulted in significant increases in plant biomass. Moreover, cultivar BRS330 exhibited significant increases in average root diameter when inoculated with either strain. This study shows that Burkholderia tropica strain IAC/BECa 135 and H. frisingense strain IAC/BECa 152 are promising PGPB strains for use as inocula for sustainable sorghum cultivation.

Materials and Methods

Mesocosm experiment

The experiment was performed in a greenhouse of the Netherlands Institute of Ecology (NIOO-KNAW), located in Wageningen, the Netherlands. The experiment was carried out from September to October 2016, with a total duration of 30 days. A complete random design was used. The treatments consisted of four sorghum cultivars, each one inoculated independently with five bacterial isolates, with a total of nine replicates per treatment. A non-inoculated treatment under phosphate starvation conditions was established as a control. Seeds were disinfected as described by Liu et al. (2013). Briefly, seeds were soaked in 70% ethanol for 3 min, transferred to a new tube containing 2.5% sodium hypochlorite solution, and shaken for 5 min. The seeds were then washed with 70% ethanol solution for 30 s. Finally, the seeds were rinsed with sterile water four times. After the last washing step, 20 μl of the remaining water was plated on Petri dishes with Luria-Bertani (LB) medium to confirm the success of disinfection. After disinfection, the seeds were placed in Petri dishes containing 1% water agar medium, and the plates were incubated at 25 °C for 2 days in the dark for seed germination. The experiment is illustrated in Fig. 1. When radicle emerged from the seed coat, the seedlings were transplanted from the Petri dishes to 11 × 11 × 12 cm plastic pots filled with autoclaved silver sand as substrate. The pots containing one plant each were maintained under greenhouse conditions for 4 weeks. During the first week, the pots were watered with 1/2 Hoagland 10% P nutrient solution, followed by P starvation. To create P starvation conditions, the substrate with plants was first flushed with 500 mL of 1/2-strength Hoagland nutrient solution without phosphate to remove any remaining phosphate in the substrate by drainage through the pot. After 2 days, to simulate field conditions, two g (125 μM) of insoluble tricalcium phosphate (Ca₃(PO₄)₂), which can be solubilized by microorganisms but not taken up directly by the plant (Estrada et al., 2013), was diluted in Hoagland nutrient solution and applied to the pots. The watering regime was maintained by applying 25 ml of nutrient solution every 2 days.

Results

Discussion

This work aimed to evaluate the effect of five bacterial strains isolated from sugarcane on the plant growth and root architecture of four sorghum cultivars. Cultivars SRN-39 and BRS330 inoculated with Burkholderia tropica strain IAC/BECa 135 or H. frisingense strain IAC/BECa 152 and cultivar SRN-39 inoculated with E. asburiae strain IAC/BECa 128 or H. frisingense strain IAC/BECa 152 exhibited significant increases in sorghum root and shoot biomass, respectively, compared with the control. Although the number of replicates in our study was small (6) our results corroborate those of Chiarini et al. (1998), who found that isolates belonging to the genera Burkholderia and Enterobacter co-inoculated in the sorghum rhizosphere promoted a significant increase in root growth compared to non-inoculated plants. Furthermore, species belonging to the genera Burkholderia and Herbaspirillum promote the growth of sugarcane and maize (Pereira et al., 2014; Da Silva et al., 2016), which, like sorghum, are C4 grass species. H. frisingense strain IAC/BECa 152 produces siderophores and IAA, whereas Burkholderia tropica strain IAC/BECa 135 does not. The strain IAC/BECa 152 might possess a set of mechanisms that improve plant nutrient uptake either by increasing nutrient availability in the rhizosphere or influencing the biochemical mechanisms underlying nutritional processes (Pii et al., 2015). Such mechanisms include changes in the root system architecture and shoot-to-root biomass ratio, increases in proton efflux by modulating H⁺ATPase activities, indirect effects of IAA produced by PGPB, or acidification of the rhizosphere to enhance nutrient solubility (Pii et al., 2015). In addition to growth regulators, siderophores can be produced and are known to assist Fe acquisition by roots (Saravanan, Madhaiyan & Thangaraju, 2007; Mehnaz et al., 2013).

In contrast to the effects of Burkholderia tropica, H. frisingense and E. asburiae on the growth of both grain sorghum cultivars, these strains had no significant effect on BRS 509 (sweet sorghum) and on SQR cultivars compared with the control. Interestingly, in accordance with our results, Dos Santos et al. (2017) observed significant increase in the biomass of grass and grain sorghum inoculated with Burkholderia ssp. or Herbaspirillum ssp. but not sweet sorghum inoculated with the same isolates. Taken together with our results, these findings suggest that the effects of strains of Burkholderia tropica and H. frisingense on plant growth are dependent on sorghum genotype. It is unclear why the effects of these strains were greater in certain sorghum cultivars than others. However, different sorghum genotypes release different SL molecules in different quantities under P starvation (Schlemper et al., 2017). Thus, the high relative abundance of the genus Burkholderia in the rhizosphere of sorghum cultivar SRN-39 could be related to the level of orobanchol, which is 300 and 1,100 times higher in SRN-39 than in the cultivars SQR, BRS330 and BRS509 as suggested by Schlemper et al. (2017).

When inoculated with E. asburiae strain IAC/BECa 128, cultivar BRS330 displayed an increase in SRL and area compared with the control. These finding are in agreement with a study by Kryuchkova et al. (2014), who reported that Enterobacter species can promote increased root length and lateral roots in sunflower. The strain IAC/BECa 128 can solubilize phosphate, which trait might explain the increase in root biomass. The cultivar BRS330 inoculated with strain IAC/BECa 135 (Burkholderia tropica) or IAC/BECa 152 (H. frisingense) exhibited a significant decrease in average root diameter compared with the control but slight increase in root density. Plants under P deficiency conditions may increase root density probably to enhance nutrient acquisition (Kapulnik & Koltai, 2014). Moreover, species belonging to the genus Herbaspirillum can influence plant root architecture and improve signaling pathways of plant hormone production (Straub et al., 2013). No significant effects on sorghum growth or root architecture modification were observed when K. radicincitans strain IAC/BECa-99 or P. fluorescens strain IAC/BECa-141 was inoculated in the rhizosphere of any evaluated sorghum cultivar. Although there are many reports on the effects of K. radicincitans (formerly known as E. radicincitans) on a range of plants, such as Arabidopsis thaliana, radish, and tomato (Berger, Brock & Ruppel, 2013; Brock et al., 2013; Berger et al., 2015), there are no reports on the effects of this bacterial species on sorghum growth or root architecture modification. With respect to P. fluorescens, Marcos et al. (2016) found that the strain IAC/BECa 141, when used as an inoculant applied to two sugarcane varieties, increased chlorophyll a content without changing plant growth. Kumar et al. (2012) studying the effect of seven different fluorescent Pseudomonas spp. strains with single or multiple PGPR traits, in sorghum growth, observed that all strains were able to increase sorghum growth compared to a non-inoculated control. Our results demonstrated that selected bacterial strains characterized as PGPB in sugarcane were able to promote plant growth and root architecture modification in sorghum. Based on the reproducibility of the performance of bacterial strains for different crops, our findings shed light on the identification of bacterial candidate strains for improving the growth and yield of crops that share the same soil bacterial source in intercropping or crop rotation systems. However, since we did not evaluate the bacterial community that actually colonized the root system, it is difficult to confirm that the differences we observed in sorghum biomass were due to the effects of the inoculation or a side effect of the plant due to the simple presence of these bacterial cells in the rhizosphere. Hence, we strongly recommend future studies to recover the bacterial community from the endosphere and rhizosphere compartments as proof of inoculation. We suggest that SL plays a role in the effectiveness of PGPB in promoting the growth of specific sorghum genotypes, although our experimental set-up did not allow us to make a straightforward conclusion. More specific experiments are needed to better address the relationship between plant SL production and plant bacterial infection.

Conclusion

Here, we demonstrated that bacteria strains characterized as PGPB in sugarcane were able to promote plant growth and root architecture modification in sorghum. Our results demonstrated that cultivars SRN-39 and BRS330 inoculated with Burkholderia tropica strain IAC/BECa 135 or H. frisingense frisingense strain IAC/BECa 152 exhibited a significant increase in plant biomass. Moreover, cultivar BRS330 inoculated with either strain displayed a significant decrease in AvD. The results of this study indicate that Burkholderia tropica strain IAC/BECa 135 and H. frisingense strain IAC/BECa 152 are promising PGPB strains for use as inocula for sustainable sorghum cultivation.

Supplemental Information