Introduction

Salicylic acid (SA) is a key plant defense hormone required for immunity against pathogen infection1. Pathogens often induce SA accumulation both locally and systemically. This pathogen-induced SA is synthesized through the isochorismate pathway comprised of ISOCHORISMATE SYNTHASE 1 (ICS1), ENHANCED DISEASE SUSCEPTIBILITY 5 (EDS5), and avrPphB SUSCEPTIBLE 3 (PBS3)2. A recent study has revealed that ICS1, EDS5 and PBS3 are necessary for pathogen-induced SA biosynthesis in plants2. SA accumulation leads to the activation of plant defense responses, including programmed cell death and systemic acquired resistance. High levels of SA accumulation result in changes in transcriptional activities of genes with antimicrobial activity such as the PATHOGENESIS-RELATED (PR) genes3. In Arabidopsis thaliana, SA-dependent transcriptional reprogramming requires NONEXPRESSER OF PR GENES 1 (NPR1), which was shown to be the bona fide SA receptor and also functions as a transcriptional co-activator4. Many studies have reported SA-induced defense responses upon infection by leaf-colonizing pathogens5, while little is known about the involvement of SA-induced defense responses in roots.

Several studies have shown that SA metabolism is also targeted by phytoparasitic root-knot nematodes (RKNs; Meloidogyne spp.) to ensure successful infection6. RKNs cause severe yield losses of many crop species in multiple regions of the world7. During RKN infections, juveniles invade plant roots and move toward the vascular cylinder. Upon reaching the vascular cylinder, RKNs induce the formation of galls, each containing several giant cells (GCs) which serve as a source of nutrients for the nematode until maturity. Since GCs result from nuclear divisions without cytokinesis (endoreduplication), it is thought that RKN is capable of manipulating the host cell-cycle machinery6. Biotrophic pathogens such as RKNs must keep the host cells alive while also suppressing host defense mechanisms. Repression of genes of the SA pathway was indeed observed in RKN-induced galls8,9,10. In addition, overexpression of NPR1 leads to the decrease in the number of galls and RKN fecundity during RKN infection11. On the other hand, overexpression of SA-degrading salicylate hydroxylase leads to reduced SA accumulation and greater susceptibility to RKN12.

SA accumulation contributes to plant defense responses, while excessive SA often leads to growth inhibition. Therefore, it is generally believed that host plants must balance growth and defense during pathogen infection. How this is accomplished is currently unknown. The Arabidopsis atypical transcription factor DP-E2F-like 1 (DEL1)/E2Fe, a transcriptional repressor known to promote the onset of endoreduplication, has been shown to balance plant growth and defense via the SA response in leaves during pathogenic fungi infection13. Here, we report that DEL1 represses excessive SA accumulation and root growth inhibition of host plants upon RKN infection. Furthermore, lignin deposition was also shown to be elevated in galls of the del1-1 knockout mutant. Our results suggest that DEL1 plays a role in balancing growth and defense in roots as previously reported in leaves13.

Results

SA biosynthesis is enhanced in del1-1 galls

Chandran et al. showed that DEL1 acts as a transcriptional repressor of EDS513. The enhanced resistance to fungi and the small stature phenotypes of del1-1 were shown to be SA-dependent. DEL1-mediated SA accumulation has been demonstrated in leaves13, but it is unclear whether plants also balance growth–defense tradeoffs in the roots. To address this question, we subjected del1-1 to the RKN infection assay. We performed reverse transcription quantitative polymerase chain reaction (RT-qPCR) on genes related to SA biosynthesis in wild-type (WT) and del1-1 galls. EDS5 and PBS3 were up-regulated in the galls of del1-1 compared to the WT, while ICS1 did not show significant difference between transcript levels in del1-1 and WT galls (Fig. 1A). We next tested whether SA accumulation in del1-1 galls was higher than that in WT galls. Indeed, total SA accumulation in del1-1 galls was significantly higher than that in WT galls (Fig. 1B). On the other hand, total SA level in un-inoculated roots of del1-1 did not change significantly compared to that of WT (Fig. 1B). These results showed that DEL1 represses SA biosynthesis in RKN-induced galls.

Figure 1
figure 1

DEL1 regulates SA biosynthesis during RKN infection. (A) RT-qPCR analysis of SA biosynthesis-related genes in WT and del1-1 galls at 7dpi. Values are normalized to the expression levels in un-inoculated roots of WT. The experiment was repeated three times with similar results. (B) Quantification of free and total SA in the WT and the del1-1 mutant. The experiment was repeated three times with similar results. Means ± SD are shown. Different letters denote significant differences after Tukey’s multiple test at P < 0.05.

Full size image

Gall formation is compromised in del1-1

To determine whether the elevated SA levels activate the defense response, we analyzed the expression of the SA-dependent genes PR1, PR2, PR5 and NPR1 in galls. PR1 transcripts were not detected from the galls of neither WT nor del1-1 (Fig. S1), suggesting that PR1 is likely not involved in RKN infection. On the other hand, PR2 and PR5 were up-regulated in the galls of del1-1 compared to the WT, whereas NPR1 expression level did not change (Fig. S1). These results suggest transcript level of PR2 and PR5 may be modified by SA accumulation in the galls of del1-1. Previous studies have shown that SA signaling confers increased host plant resistance to nematode infection11,12,14,15. To examine the functions of DEL1 in RKN-induced gall formation, we assayed gall formation frequencies and GC formation in del1-1. Gall numbers were reduced in the del1-1 mutant compared to that in the WT (Fig. 2A). Furthermore, histological analysis showed that GC areas were significantly reduced in del1-1 at 14 days post inoculation (dpi) by approximately one third of the WT value (Figs. 2B and S2), which is consistent with the results from a previous study16. These results suggest that excessive SA accumulation from the loss-of-function of del1-1 mutation renders the host plant more resistant to RKN infection. Therefore it is possible that the endocycle machinery itself is involved in the RKN infection together with SA signaling, as de Almeida Engler et al. (2019) has shown that del1-1 mutant with defect of the endocycle machinery produced malformed giant cells16.

Figure 2
figure 2

Gall formation is compromised in del1-1 mutants. (A) WT-normalized gall numbers in WT and del1-1 plants at 7 dpi (n = 15, WT; n = 16, del1-1). Sample sizes denote the number of petri dishes were tested. (B) Areas of GCs of 14 dpi galls in WT and del1-1 plants (n = 6). Sample sizes denote the number of galls. Means ± SD are shown. P values were analyzed by Student’s t-test.

Full size image

DEL1 is involved in lignin deposition during RKN infection

In Arabidopsis roots, cell wall lignification is restricted to the xylem and the Casparian strip during normal development17. Lignin contributes to the protection of plants from physical and chemical stresses, including pathogen challenges. Several studies have revealed that lignification can affect parasitic nematode infection rates18,19,20,21. Galls during early stages were subjected to phloroglucinol-HCl staining to detect lignin, which was weakly stained in 3 dpi galls but became more prominent in 5 dpi galls in WT (Fig. 3A). In contrast, lignin was strongly stained in both 3 dpi and 5 dpi galls of del1-1 (Fig. 3A). In fact, 5 dpi galls of del1-1 showed stronger lignin staining than WT galls at 5 dpi (Fig. 3A). Non-galling parts of RKN-infected seedlings in both WT and del1-1 plants did not show lignin staining except for the vascular tissues (Fig. 3A). Next, we analyzed the transcript levels of lignin biosynthesis-related genes in galls by RT-qPCR. The expression levels of 4-coumarate: CoA ligase 1 (4CL1), 4CL2, alcohol dehydrogenase 5 (CAD5), phenylalanine ammonia-lyase 1 (PAL1), PAL2, and cinnamate 4-hydroxylase (C4H) were significantly up-regulated in del1-1 galls compared to WT galls (Fig. 3B). These results indicate that DEL1 negatively regulates local lignin deposition during RKN infection.

Figure 3
figure 3

Lignin accumulates specifically in del1-1 plants after RKN infection. (A) Phloroglucinol-HCl staining in non-infected regions of RKN-infected roots (control), 3 dpi and 5 dpi galls (5 individuals were observed with similar results). Top: WT. Bottom: del1-1. Scale bars = 100 µm. (B) RT-qPCR analysis of lignin synthesis-related genes in WT and del1-1 7 dpi galls. Values are normalized to expression levels in the WT. 4CL: 4-coumarate-CoA ligase, CAD: cinnamyl alcohol dehydrogenase, PAL: phenylalanine ammonia-lyase, C4H: cinnamate 4-hydroxylase. The experiment was repeated three times with similar results. Means ± SD are shown. ***P < 0.001 by Student’s t-test.

Full size image

del1-1 exhibits root growth inhibition after RKN infection

Chandran et al. (2014) showed that DEL1 controls rosette leaf size upon pathogen infection, while SA accumulation and lignin deposition are often associated with inhibition of growth in Arabidopsis13,22,23,24,25,26,27. We therefore questioned whether DEL1 controls the balance between root growth and immunity after RKN infection. 10-day-old un-inoculated del1-1 seedlings did not show significant differences in root growth compared to WT (Fig. 4A,B), whereas the total root length of RKN-inoculated del1-1 seedlings at 7 dpi was significantly reduced compared to WT (Fig. 4A,C). These results indicate that DEL1 plays an important role in the balancing growth–defense tradeoffs in root during RKN-infection (Figs. 2 and 4).

Figure 4
figure 4

Root growth is inhibited in del1-1 mutants after RKN infection. (A) Representative images of WT and del1-1 seedlings with or without RKN inoculation. Top: 10-days-old non-inoculated seedlings. Bottom: RKN-inoculated seedlings at 7 dpi. (B) Total root length (primary root + lateral roots) of non-inoculated WT and del1-1 plants (n = 12, WT; n = 11, del1-1). (C) Total root length of RKN-inoculated WT and del1-1 plants (n = 18). Means ± SD are shown. Sample sizes denote the number of plants. The experiment was repeated three times with similar results. P values were analyzed by Student’s t-test.

Full size image

Discussion

In plants, the activation of immunity is often inversely correlated with growth. Thus, the tradeoff between immunity and growth is thought to be important for survival of plants under biotic stresses. We demonstrated that knocking out DEL1/E2Fe leads to enhanced resistance and growth inhibition during RKN infection, likely due to excessive lignification and/or SA accumulation in RKN-induced galls (Fig. 5). Our RT-qPCR analysis showed that PR2 and PR5 were up-regulated in del1-1 galls, but PR1 was not (Fig. S1). This suggests a portion of the SA metabolism genes respond to RKN-induced SA accumulation in del1-1 galls. Since Fu et al. (2012) has shown that NPR1 protein level is up-regulated post-transcriptionally after SA treatment28, NPR1 proteins thus are likely to be highly accumulated in del1-1 galls (Fig. S1). A number of studies have revealed that alterations in lignin biosynthesis result in changes in both growth and defense, while excessive lignification inhibits plant growth25,26,27,29. In addition, transgenic and mutant plants with elevated basal SA levels also exhibit growth reduction13,22,23,24. Moreover, recent studies have shown that SA accumulation in response to pathogen attack positively correlates with lignin deposition and acquisition of immunity30,31,32,33. In light of these lines of evidence, DEL1 may mediate the balance between defense and growth by limiting SA accumulation during RKN infection in root. Chandran et al. (2014) have reported that del1-1 plants exhibit small rosette leaves and SA accumulation higher than that of WT regardless of fungal infection13. On the other hand, our results showed that del1-1 exhibited root growth inhibition only in the presence of RKN infection (Fig. 4). SA accumulation in roots of del1-1 in the absence of RKN was not significantly different compared to that of WT (Fig. 1B), suggesting this difference in response between in leaves and in roots may come from differences of SA level in the absence of pathogen infection.

Figure 5
figure 5

Schematic model depicting the role of DEL1 during RKN infection. Proposed role for DEL1 as a mediator plant growth and defense in the RKN infection in Arabidopsis roots. In WT plants, DEL1 controls basal SA revels and lignification in galls after RKN attack to balance plant growth and defense (green panel). Excessive SA accumulation and lignification due to misexpression of DEL1 contribute to plant defense after RKN attack, whereas the root growth is inhibited (orange panel).

Full size image

Although reduced parasitic nematode fecundity on del1-1 has been attributed to the deficiency in cell division during syncytia/gall formation16, our results point to the additional possibility that the up-regulation of SA accumulation and lignification may also be contributing factors in gall formation efficiency. Further detailed analysis of the function of DEL1 in SA accumulation/lignification during RKN infection may provide insights into the mechanisms of balancing defense and growth in plants.

Methods

Plant materials and growth conditions

All Arabidopsis thaliana plants used in this study are of the Col-0 ecotype background. The mutant line, del1-1, was obtained from the Arabidopsis Biological Resource Center (ABRC). Arabidopsis seeds were stratified for 2 days in 4 °C in the dark, then allowed to germinate and grown for 5 days on 0.25 × Murashige and Skoog (MS) salt mixture (Sigma), 0.5% (w/v) sucrose, and 0.6% (w/v) gellan gum at pH 6.4 under continuous light at 23 °C.

Nematode preparation and inoculation

Root-knot nematodes (RKN, Meloidogyne incognita) were prepared aseptically as described previously34. Briefly, 6- to 7-week-old tomato plants were inoculated at 3-day intervals for a total of four inoculations. Approximately 80,000 juveniles were used to inoculate each plant. The inoculated tomato plants were then transferred to a hydroponic system, and after 2 to 4 day intervals infective juveniles were collected from the hydroponic culture media. Six Arabidopsis seeds were sown on a MS plate. Five-days-old Arabidopsis seedlings were inoculated with approximately 80 nematodes per plant and incubated under short-day conditions (8 hours light/16 hours dark) at 25 °C. The roots of the seedlings were covered with black paper to mimic the dark underground environment in nature.

Evaluation of gall formation efficiency

Gall numbers were normalized for each petri dish of six seedlings to evaluate the gall formation efficiency (Data S1). The calculation procedure is as described below: (1) Galls number and germinated seedlings number were counted for each petri dish of six seedlings. (2) Gall number/seedling values were calculated. (3) Average values of galls/seedling in WT were calculated for each independent experiment to define the basal value (highlighted in orange in Data S1). (4) The galls/seedling values from WT and del1-1 were normalized to the basal value for each petri dish to calculate relative galls number. (5) Average values of relative gall numbers were calculated (highlighted in blue in Data S1). These values were used for Fig. 2A.

Phloroglucinol-HCl staining

Lignin was detected by the Wiesner test by immersing roots or galls of RKN-infected seedlings in phloroglucinol reagent [1% (w/v) phloroglucinol in 20% (v/v) HCl] for 5 minutes. Stained samples were then mounted in chloral hydrate solution (8 g chloral hydrate, 2 ml ultrapure water and 1 ml glycerol). Samples were imaged with an Axio Imager M1 microscope (Carl Zeiss) mounted with a DP71 digital camera (Olympus).

Gene expression analysis

Total RNA was extracted from roots or galls (approximately 50 mg fresh weight) using the RNeasy plant mini kit (Qiagen), then treated with Recombinant DNase I (Takara) according to the manufacturer’s instruction. First-strand cDNA was synthesized from 300 ng of total RNA using PrimeScript RT Master Mix (Takara). The transcript level of target genes was assayed using FastStart Essential DNA Green Master (Roche) and the LightCycler 480 system (Roche). The thermal cycler program was 95 °C for 5 min followed by 55 cycles of 95 °C for 10 s, 60 °C for 10 s, and 72 °C for 10 s. GAPDH was used as the internal control and relative expression levels were calculated by the ΔΔCt method. Sequences of primers used are listed in Table S1. Each of the three biological replicates were performed in technical triplicates.

Histological analysis

Galls were dissected and transferred into 2% glutaraldehyde in 20 mM cacodylate buffer, pH 7.4, vacuum-infiltrated for 10 min twice, then incubated in 4 °C overnight. Samples were dehydrated in a graded ethanol series and embedded in Technovit 7100 (Kulzer) according to the manufacturer’s protocol. Sample blocks were sectioned to 5 µm thickness using an ultramicrotome (LEICA RM2255, Leica) and stained with 0.01% (w/v) toluidine blue O (WALDECK) containing 1% (w/v) sodium borate decahydrate (Nacalai) for 2 min. All samples were rinsed in deionized water for 1 min. After drying, sections were mounted in EUKITT (O. Kindler). Samples were imaged with an Axio Imager M1 microscope (Carl Zeiss) mounted with a DP71 digital camera (Olympus). Lengths and areas of the galls were quantified using ImageJ.

Quantification of total SA

Total SA (free SA plus SA glucosides) from Arabidopsis thaliana roots was extracted as previously described35. Non-inoculated whole root systems (primary roots and lateral roots) of 12-days-old seedlings, or 7 dpi galls (more than 150 mg fresh weight) were collected. SA of the root exudates was determined by high performance liquid chromatograph–fluorescence detector (HPLC-FL) (LC-2000 plus and FP-2020 plus, Jasco) according to the previous report35. Briefly, 20 µL of sample was injected in water/methanol (20:80) eluent to be separated from other organic compounds by using a multi-mode column (Scherzo SM-C18, Imtakt, 2 mm × 50 mm). Fluorescence at 407 nm was monitored with 305 nm excitation for SA and o-anisic acid.

Statistical information

All P values were derived from two-sided Student’s t-tests or Tukey’s tests. All statistical tests and n numbers, including sample sizes or biological replications, are described in the figure legends.