Systemic defense priming by Pseudomonas putida KT2440 in maize depends on benzoxazinoid exudation from the roots

Abstract

Exudation of benzoxazinoid metabolites from roots of young maize seedlings recruits the rhizobacterial strain Pseudomonas putida KT2440 from the soil to the rhizosphere. In this study, we have investigated whether these rhizobacteria are beneficial for maize by eliciting systemic defense priming. Root colonization of the maize hybrid cultivar Delprim by P. putida primed wound- and jasmonic acid (JA)-inducible emission of aromatic and terpenoid volatiles, but not the emission of the green leaf volatile (Z)-3-hexenyl acetate. Furthermore, root colonization by P. putida primed stress-inducible transcription of the JA-dependent gene SerPIN, whereas JA-dependent induction of the MPI gene was unaffected. Systemic priming of SerPIN by P. putida only occurred in benzoxazinoid-producing plants, and was absent in benzoxazinoid-deficient plants. The results from this study suggest that root colonization by P. putida primes a selection of JA-dependent defenses in Maize, which is reliant on benzoxazinoid exudation from the roots.

The relationship between cereal plants and soil bacteria has evolved over 60 million years, the time at which the first firm evidence for grass pollens exists.¹ This long period of co-evolution has resulted in a wide-range of cereal-biotic interactions, ranging from beneficial to pathogenic. Selected beneficial strains of non-pathogenic soil bacteria can promote plant growth. This interaction has attracted a great deal of interest as it provides opportunities for exploitation in sustainable food production by cereals. Pseudomonas spp., particularly P. fluorescens² and P. putida,^3,4 have been extensively studied for their ability to promote plant growth. Plant growth-promotion by rhizobacteria has been ascribed to various mechanisms, including nitrogen fixation, solubilization of essential plant nutrients, production of plant-like growth hormones, inhibition of growth-repressing ethylene production and direct antagonism of growth-suppressing plant pathogens in the rhizosphere.⁵ Recently, ref. 6 proposed an alternative mechanism: rhizobacteria induce growth promotion in Arabidopsis by inducing a starvation-like response. The authors proposed that the resulting increase in soluble carbohydrates in the plant not only benefits bacteria on the rhizoplane, but may also contribute to growth promotion.

In addition to the mechanisms noted above, some growth-promoting bacteria are capable of improving plant health via eliciting an induced systemic resistance (ISR) response. In this case, colonization by rhizobacteria results in long-lasting resistance against a broad range of pathogens.⁷ The plant signaling mechanisms mediating ISR have been studied extensively in Arabidopsis following root colonization by P. fluorescens WCS417r.^8,9 In this model system, ISR is based on systemic priming of the plant immune system, resulting in a quicker and more potent accumulation of ethylene- and JA-dependent gene transcripts and callose-rich papillae after pathogen attack.^10-12 Pseudomonas-elicited ISR has also been reported in a variety of crop-pathogen partnerships, including Cotton-Fusarium oxysporum,¹³ Cucumber-Colletotrichum orbiculare¹⁴ and Rice-Magnoporthe oryzae.¹⁵ In the latter, ISR is associated with an augmented capacity for pathogen-induced callose deposition and functional responsiveness to the plant hormone JA.

We recently reported attraction of the soil bacterium P. putida KT2440 cells to the Maize rhizoplane in response to exudation of the benzoxazinoid metabolite 2,4-dihydroxy-7-methoxy-2H-1,4-benzoxazin-3(4H)-one (DIMBOA).¹⁶ DIMBOA is exuded in relatively high concentrations from the roots of young Maize plants,¹⁶ and is known for its insecticidal and phytotoxic activities.^17,18 In addition, DIMBOA plays a signaling role in above-ground defenses against aphids and fungi, where it functions as an apoplastic signal for induction of cell wall defense.¹⁹ Despite these defense activities, roots of benzoxazinoid-producing Maize lines are subject to higher levels of P. putida KT2440 root colonization than benzoxazinoid-deficient Maize lines carrying a mutation in the ZmBX1 gene. In vitro experiments revealed that this difference is based on enhanced tolerance of P. putida KT2440 to high concentrations of DIMBOA, combined with a positive chemotactic response to the compound. However, it remained untested whether Maize plants benefit from root colonization by P. putida and to what extent such beneficial host effects rely on benzoxazinoid-dependent recruitment of bacteria. In this study, we have investigated whether P. putida bacteria prime JA-dependent defense mechanisms in Maize, and whether these responses rely on the host plant’s ability to produce benzoxazinoids.

Results and Discussion

Analysis of VOC emissions from intact Maize plants indicated that root colonization by P. putida has no direct effects on emission of the majority of volatiles tested (Fig. 1). The only volatile showing increased emission rates in P. putida-colonized plants was the monoterpene (±)-Linalool. Linalool has been shown to affect insect behavior.^26-28 Ref. 29 reported that VOC blends with higher levels of (-)-isomer deterred oviposition on Datura wrightii by Manduca sexta moths. More recently, ref. 30 reported contradictory data, and showed that increased emission of (+)-Linalool from transgenic tobacco plants deters oviposition by the moth Helicoverpa armigera, but has no effect on larval development or feeding.³⁰ Since we did not investigate the chirality of linalool identified in our experiments it is not possible to predict the likely effect of increased emission upon insect behavior.

Figure 1. Emission of volatile organic compounds (VOCs) from maize leaves upon root-colonization by Psedomonas putida KT2440 and subsequent defense elicitation by leaf wounding and JA application. Shown are average emission rates (± SEM) from two independent experiments over a 24 h collection period. Asterisks indicate statistically significant difference between bacterized and non-bacterized control groups (Student’s t-test; p < 0.05).

In contrast to undamaged plants, stress treatment by leaf wounding and JA application in P. putida-colonized plants resulted in augmented emissions of nearly all volatiles tested, except for the green leaf volatile (Z)-3-hexenal (Fig. 1). These priming effects were statistically significant in at least one of the two experiments performed. Hence, root colonization by P. putida KT2440 appears to prime emission of stress-inducible aromatic and terpenoid volatiles from shoots. Why stress-induced emission of (Z)-3-hexenyl acetate was not primed by P. putida remains unclear. However, Maize plants primed by caterpillar herbivory show a similar pattern of aromatic and terpenoid volatile potentiation with no effect upon green leaf volatiles. This suggests a similar mechanism between P. putida-induced defense priming and priming following exposure to herbivore-induced volatiles from neighboring plants.

To characterize defense priming in Maize in relation to benzoxazinoid-dependent root colonization by P. putida, we quantified stress-inducible gene transcription in control- and P. putida-treated plants of the benzoxazinoid-producing BX1 igl line and the BX-deficient bx1 igl line. Similar to stress-induced emission of aromatic and terpenoid volatiles (Fig. 1), basal levels of transcription of JA-dependent MPI and SerPIN were not directly influenced by the presence of P. putida (Fig. 2). Leaf wounding in combination with JA application resulted in transcriptional induction of both MPI and SerPIN, and was of similar intensity in both genotypes tested, suggesting that benzoxazinoids do not play a direct role in transcriptional activation of JA-dependent genes. Furthermore, stress-inducible MPI expression in both Maize genotypes was not influenced by P. putida root colonization. Hence, systemic defense priming by P. putida has no influence on the transcriptional responsiveness of MPI gene. This also resembles the response to herbivore-induced volatiles in Maize, where the MPI gene remained unresponsive to priming treatment.²¹ In contrast, stress-inducible transcription of the SerPIN gene was strongly augmented in P. putida treated BX1 igl plants, while there was no evidence for such transcriptional gene priming in bx1 igl plants. Hence, the host plant’s ability to synthesize benzoxazinoids determines P. putida-induced defense priming in the leaves.

Figure 2. Transcription of JA-responsive genes in maize leaves of benzoxazinoid-producing (BX1 igl) and benzoxazinoid-deficient (bx1 igl) maize after root-colonization by P. putida KT2440 and subsequent leaf wounding and JA application. Shown are means of relative gene expression (± SEM; n = 6) at 8 h after wounding and JA application. Values are normalized to the average expression value of intact, control-treated BX1 igl plants. The asterisk indicates a statistically significant difference between the bacterized and non-bacterized group (Student’s t-test; p < 0.05).

Considering that root exudation of benzoxazinoids recruits P. putida to the rhizosphere,¹⁶ our results suggest that benzoxazinoid-dependent root colonization by P. putida is important for aboveground defense priming in the host plant. It is, however, also possible that root-exuded benzoxazinoids exert additional influence on P. putida physiology than simply stimulating chemotaxis and root colonization. For instance, benzoxazinoids may induce bacterial production of ISR-eliciting determinants in the rhizobacteria. A third explanation for benzoxazinoid-dependent defense priming by P. putida could arise from differences in defense physiology between benzoxazinoid-producing and benzoxaiznoid-deficient host plants. We have previously demonstrated that apoplastic accumulation of DIMBOA in Maize during initial stages of aphid feeding and fungal infection boosts callose deposition.¹⁹ Although benzoxazinoid-producing and benzoxaiznoid-deficient Maize lines do not show differences in JA-dependent defense gene expression in the absence of P. putida, we cannot exclude that the development of a primed defense state upon root colonization by P. putida requires functional benzoxazinoid metabolism in the leaves. However, considering that belowground exudation of DIMBOA promotes root colonization by P. putida,¹⁶ we propose that the lack of defense priming in benzoxazinoid-deficient Maize plants relies on activity of the bacterial partner.

In context of our previous findings on the role of benzoxazinoids in Maize-biotic interactions,^16,19 our study further justifies the conclusion that these secondary metabolites play an important regulatory role in below- and aboveground defense responses of Maize. The implication that belowground benzoxazinoids recruit bacteria that promote aboveground defense responsiveness has consequences at multiple trophic levels. Further support for this notion comes from ref. 31, who demonstrated that root benzoxazinoids can be exploited by the specialist root herbivore Diabrotica virgifera to localize nutrient-rich crown roots, which, in turn, can alter defense responses aboveground.²³ Further research on the effects of root-exuded benzoxazinoids on communities of plant-associated microbes and arthropods is warranted to fully reveal the importance of benzoxazinoids in cereal-biotic interactions.

Materials and Methods

Biological material

The green fluorescent protein-expressing strain FBC004 was used for all experiments, which is a derivative of Pseudomonas putida KT2440. Bacteria were cultivated as described previously.¹⁶ P. putida-induced priming of wound- and JA-inducible volatile emission was studied in the Maize cultivar Delprim, which is routinely employed to study herbivore-induced VOC emission due to a robust and relatively strong volatile response to wounding.²⁰ To determine the role of benzoxazinoids in P. putida KT2440-induced defense gene priming, benzoxazinoid-producing and benzoxazinoid-deficient mutant lines of Maize were used, derived from a cross between bx1 single-mutant and indole-deficient igl mutant lines, as described by ref. 19. Because the bx1 mutant produces residual amounts of benzoxazinoids due to a functional Indole-3-Glycerol phosphate Lyase gene (IGL),¹⁹ comparisons were made between benzoxazinoid-producing BX1 and benzoxazinoid-deficient bx1 lines in the igl mutant genetic background (i.e., BX1 igl vs. bx1 igl).

Root inoculation with P. putida and plant cultivation

Maize seeds were pre-germinated in wetted Petri-dishes for 3–4 d in the dark. Bacterial root colonization was effected by gently shaking sprouting seeds for 30 min in a suspension of washed P. putida cells from an overnight culture as described by ref. 16. A second set of seedlings were shaken in the same manner in sterile salt solution (3.4 mM NaHPO₄; 2 mM KH₂PO₄; 0.9 mM NaCl; 0.9 mM NH₄Cl) to provide the non-colonized control treatment. Visual observation of the roots using epi-fluorescence microscopy confirmed that roots exposed to GFP-expressing P. putida FBC004 were extensively covered in a bacterial film. Sets of four seedlings were transferred to 80 mL pots containing compost and grown for 10 d under controlled conditions (25°C; 16:8 h light-dark cycles; 150 μE m⁻² s⁻¹).

Collection and quantification of volatile organic compound emission

For each treatment (with or without bacteria), 12 plants in three pots were stimulated for JA-dependent VOC emission by wounding the first three leaves at two separate sites, using 12-inch serrated dressing forceps dipped in a 100 μM jasmonic acid solution (Sigma-Aldrich; J2500). Similar numbers of plants of each treatment remained free of mechanical stress. VOC emission was measured by air-entrainment as described previously.^21,22 Potted plants were placed in air-tight glass vessels and charcoal-purified air was pumped through at a rate of 0.7 L min⁻¹. Air exiting the vessels was passed through a trap containing Porapak™ Q beads. After 24 h, volatile traps were removed and the absorbed VOCs were eluted with three sequential 750 μL washes of redistilled diethyl ether, spiked with 200 ng/mL tridecane as internal standard. VOCs contained in the eluent were then identified using gas chromatography coupled to mass spectrometry (GC-MS) using a capillary gas chromatography column (EC05, 30 min length, 0.25 mm i.d., 0.25 μm film thickness) directly coupled to a mass spectrometer (VG Autospec, Fisons Instruments). Ionization was performed by electron impact (70 eV, 250°C). The oven temperature was maintained at 30°C for 5 min, and then programmed to rise 5°C min⁻¹ up to 250°C. Volatile quantities were estimated on the basis of the internal standard (tridecane). Tentative compound identities were based on comparison of mass spectra with existing databases, and were confirmed by comparison of retention indices and mass spectra of authentic standards.

Gene expression analysis

Plants for gene expression analysis were treated similarly to VOC analysis, but were not kept in glass vessels after treatment by wounding and JA. Four shoots per pot were collected from three pots per treatment at 8 h after induction treatment. RNA extraction and cDNA synthesis were performed as described previously.²³ Quantitative PCR (qPCR) analysis of transcript accumulation of the Maize Proteinase Inhibitor gene (MPI) and the Serine Proteinase Inhibitor gene (SerPIN) was performed using a Corbett Rotor-Gene-6000, using previously described DNA primers.²¹ Two technical replicates of each sample were subjected to the qPCR reaction. PCR efficiency (E) of primer pairs were estimated from data obtained from multiple amplification plots using the equation (1 + E) = 10^slope. Transcript levels were calculated relative to the constitutively expressed Actin-1 and Glycerol phosphate dehydrogenase C (GAPC) genes,²¹ using the 2^-ΔΔCt method.^24,25 Gene expression levels were normalized to average expression levels in control-treated, unwounded BX1 igl plants.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.