NPR1 protein regulates pathogenic and symbiotic interactions between Rhizobium and legumes and non-legumes

Abstract

Background: Legumes are unique in their ability to establish symbiotic interaction with rhizobacteria from Rhizobium genus, which provide them with available nitrogen. Nodulation factors (NFs) produced by Rhizobium initiate legume root hair deformation and curling that entrap the bacteria, and allow it to grow inside the plant. In contrast, legumes and non-legumes activate defense responses when inoculated with pathogenic bacteria. One major defense pathway is mediated by salicylic acid (SA). SA is sensed and transduced to downstream defense components by a redox-regulated protein called NPR1.

Methodology/principal findings: We used Arabidopsis mutants in SA defense pathway to test the role of NPR1 in symbiotic interactions. Inoculation of Sinorhizobium meliloti or purified NF on Medicago truncatula or nim1/npr1 A. thaliana mutants induced root hair deformation and transcription of early and late nodulins. Application of S. meliloti or NF on M. truncatula or A. thaliana roots also induced a strong oxidative burst that lasted much longer than in plants inoculated with pathogenic or mutualistic bacteria. Transient overexpression of NPR1 in M. truncatula suppressed root hair curling, while inhibition of NPR1 expression by RNAi accelerated curling.

Conclusions/significance: We show that, while NPR1 has a positive effect on pathogen resistance, it has a negative effect on symbiotic interactions, by inhibiting root hair deformation and nodulin expression. Our results also show that basic plant responses to Rhizobium inoculation are conserved in legumes and non-legumes.

Figure 1. Induction of root hair deformation and attachment of S. meliloti to A. thaliana hairs.

(A–C) Wild-type (A), nim1/npr1 mutant (B) and NahG transformed (C) A. thaliana seedlings were grown on nitrogen poor medium (1/60 strength MS) for eight days. Plants were left untreated (A–C, cont), inoculated with S. meliloti in zone 1 region (lower-mid root) (S.m), inoculated with P. putida (P.p), inoculated with S. meliloti nodA or nodH mutants (not shown) or treated with purified nod factors (NF). Roots were photographed 4 days after inoculation under bright light. Bar = 25 µm. Arrows point to deformed root tips. (D) Quantization of root hair deformation response in A. thaliana plants inoculated with S. meliloti producing intact nod factor (grey) or nodA ⁻ mutant (black), P. putida (black stipes), or left uninoculated (white). One hundred root hairs from 10 seedlings of wild-type and nim1/npr1 mutants in zone 1 were scored in each treatment. Error bars represent SD. The experiment was repeated at least 3 times with similar results.

Figure 2. Attachment of Rhizobium to A. thaliana root hairs.

(A) Wild-type nim1/npr1 and dnd1 mutants were inoculated with S. meliloti. Four days after inoculation plant roots were rinsed with phosphate buffer, as described by , followed by buffer supplemented with 100, 200, or 500 mM NaCl. S. meliloti in 2500 µm² area in the vicinity of A. thaliana root hairs were vewed at 600X magnification with Olympus IX70 microscope and quantified from six roots of each treatment. Wild-type (white), nim1/npr1 (grey) or dnd1 (black). The ndr1 mutants showed similar results to dnd1, and therefore were not included. Error bars represent SD. (B) Confocal image of GFP-expressing S. meliloti bacteria in the vicinity of wild-type and nim1/npr1 root hairs after washing with 150 mM NaCl as described in (A).

Figure 3. Gene expression analysis of early and late M. truncatula nodulin gene orthologs in A. thaliana roots.

(A) Roots of ten days old wild-type or nim1/npr1 seedlings grown on nitrogen poor medium were inoculated in zone 1 with either S. meliloti or treated with nod factor (NF). The expression of Arabidopsis ENOD20 homolog (AtENOD20, At5g57920) was tested two days after inoculation, using quantitative Real-time RT-PCR. The results show the mean of 3 independent repeats for each treatment. Error bars represent SD. (B) Roots of ten days old wild-type and nim1/npr1 seedlings grown on nitrogen poor medium were inoculated with S. meliloti or with P.putida. Late nodulin, (AtMtN21, At5g07050) was assayed after 7 days. Plant roots were frozen in liquid nitrogen and gene expression was analyzed by semiquantitative RT-PCR. The RNA samples were normalized according to actin-2 gene expression. Experiments were repeated 3 times with very similar results.

Figure 4. Accumulation of reactive oxygen species (ROS) in roots of Medicago and of wild-type and nim1/npr1 Arabidopsis seedlings, inoculated with S. meliloti, P. putida, or P. syringae.

(A) Roots of 6 day-old M. truncatula seedlings were inoculated with wild-type or nodA mutant S. meliloti, or with P. putida. ROS production was assayed 5, 24 and 48 hours after inoculation by epi-fluorescent microscopy with 2′,7′-dichlorodihydrofluorescein diacetate and narrow-band GFP filter (Ex 485±10 nm/Em 525±10 nm). ROS production in uninoculated roots was below detection level (left bar space in each group). P. syringae inoculation produced similar result as P. putida (not shown). Error bars indicate standard deviation of the mean (N = 12). (B) Roots of nine day-old wild-type or nim1/npr1 A. thaliana seedlings were inoculated with wild-type or nodA, or nodH (not shown) mutant S. meliloti, P. putida, or P. syringae. ROS production was assayed 24 hours after inoculation by epi-fluorescent microscopy with 2′,7′-dichlorodihydrofluorescein diacetate and a narrow-band GFP filter (Ex 485±10 nm/Em 525±10 nm). All samples were analyzed using identical exposure conditions. Fluorescence from npr1 plants inoculated with S. meliloti nodH⁻ mutant was even below uninoculated control (data not shown). Shown are representative A. thaliana roots images from four similar experiments 24 hours after inoculation. Bar = 125 µm. (C) Quantitation of ROS production in A. thaliana roots, 24 hours after inoculation with wild-type S. meliloti (S.mel), nodA mutants (nodA⁻), P. syringae (P.syr) or P. putida (P. put). ROS were quantified using ImagePro Plus software package. Error bars indicate standard deviation of the mean (N = 12).

Figure 5. Accumulation of ROS in wild-type, nim1/npr1, and NahG A. thaliana roots, treated with nod factor.

Roots of nine day-old wild-type, nim1/npr1, or NahG transformants A. thaliana seedlings were either left intact or treated with nod factor (NF) and 24 hours later were assayed for ROS production by epi-fluorescent microscopy using 2′,7′-dichlorodihydrofluorescein diacetate and a narrow-band GFP filter (Ex 485±10 nm/Em 525±10 nm). All samples were analyzed using identical exposure conditions. Bars = 50 µm.

Figure 6. The effect of diphenyleneiodonium on S.meliloti-induced ENOD expression.

Roots of ten days old wild-type or nim1/npr1 seedlings were transferred to plates supplemented with 8 µM diphenyleneiodonium (DPI) or replanted on the same (nitrogen poor) medium. After four hours the roots of all plants were inoculated in zone 1 with S. meliloti. The expression of Arabidopsis ENOD20 homolog (At5g57920) was tested two days after inoculation, using quantitative Real-time RT-PCR. The RNA samples were normalized according to actin-2 gene expression. The results show the mean of 3 independent repeats for each treatment. Error bars represent SD.

Figure 7. The effect of NPR1 overexpression or silencing by RNAi on the root hair deformation/curling.

(A, B) Roots of seven days-old M. truncatula were bombarded with A. tumefaciens containing an empty vector, RNAi silencing (NPR1-RNAi), or with NPR1 overexpressing (NPR1-Overexp) vectors. After transformation the seedlings were transferred to new plates and either left intact (A), or inoculated with S. meliloti after one hour (B). Seedlings were observed for root hair deformation or curling by bright light microscopy two days after inoculation. Roots of six seedlings of each treatment were analyzed. Note the root hair curling already 2 days after S. meliloti inoculation in the NPR1-RNAi transformed roots, while the empty vector transformed roots (control) show only moderate root hair deformation. Bars = 50 µm. (C) NPR1 gene expression following transformation, normalized according to EF1a gene expression. (D) Quantification of the root hair deformation and curling in M. truncatula seedlings overexpressing the NPR1 gene (NPR1-OE) or transformed with RNAi construct (NPR1-RNAi) to suppress NPR1 gene expression.

Figure 8. Semiquantitative RT-PCR analysis of pathogenesis-associated gene expression in M. truncatula.

M. truncatula seeds were germinated on an N-free medium and after 6 days inoculated with 10⁷ cells of S. meliloti. Total RNA was extracted 24 hours after Rhizobium inoculation. The amount of RNA in the samples was normalized according to EF1a gene expression. Primers to the NPR1-dependent genes in M. truncatula: LRK (TC107689), ARP (TC104767) and WAK (TC109689) were identified by BLAST analysis, using the M. truncatula homologues to A. thaliana gene sequences . M. truncatula seedlings were germinated on N-free medium and inoculated with S. meliloti three days later. Total RNA was extracted 0, 4, 9, 24, and 72 hours post inoculation (h.p.i.) and analyzed by semiquantitative RT-PCR, using the M. truncatula EF1a expression to normalize the amounts of RNA. All experiments were repeated at least three times, independently, with very similar results.