Planta

, Volume 229, Issue 4, pp 1009–1014

Biotic and abiotic stress down-regulate miR398 expression in Arabidopsis

Authors

  • Guru Jagadeeswaran
    • Department of Biochemistry and Molecular BiologyOklahoma State University
  • Ajay Saini
    • Department of Biochemistry and Molecular BiologyOklahoma State University
    • Department of Biochemistry and Molecular BiologyOklahoma State University
Rapid Communication

DOI: 10.1007/s00425-009-0889-3

Cite this article as:
Jagadeeswaran, G., Saini, A. & Sunkar, R. Planta (2009) 229: 1009. doi:10.1007/s00425-009-0889-3

Abstract

MicroRNA398 targets two Cu/Zn superoxide dismutases (CSD1 and CSD2) in higher plants. Previous investigations revealed both decreased miR398 expression during high Cu2+ or paraquat stress and increased expression under low Cu2+ or high sucrose in the growth medium. Here, we show that additional abiotic stresses such as ozone and salinity also affect miR398 levels. Ozone fumigation decreased miR398 levels that were gradually restored to normal levels after relieved from the stress. Furthermore, miR398 levels decreased in Arabidopsis leaves infiltrated with avirulent strains of Pseudomonas syringae pv. tomato, Pst DC3000 (avrRpm1 or avrRpt2) but not the virulent strain Pst DC3000. To our knowledge, miR398 is the first miRNA shown to be down-regulated in response to biotic stress (P. syringae). CSD1, but not CSD2, mRNA levels were negatively correlated with miR398 levels during ozone, salinity and biotic stress, suggesting that CSD2 regulation is not strictly under miR398 control during diverse stresses. Overall, this study further establishes a link between oxidative stress and miR398 in Arabidopsis.

Keywords

Abiotic stressArabidopsisBiotic stressMicroRNA398Superoxide dismutases

Abbreviations

CSD1

Cu/Zn Superoxide dismutase 1

CSD2

Cu/Zn Superoxide dismutase 2

miRNA

MicroRNA

MS-medium

Murashige and Skoog medium

PR1

Pathogenesis related 1

ROS

Reactive oxygen species

SOD

Superoxide dismutase

Introduction

A secondary consequence of several abiotic stresses in plants is the rapid accumulation of reactive oxygen species (ROS) such as superoxide (O2), hydrogen peroxide (H2O2) and hydroxyl radicals (OH) (Bartels and Sunkar 2005). Biotic stress caused by many bacterial and fungal pathogens similarly causes rapid accumulation of ROS (O2 and H2O2) at the onset of the hypersensitive response; termed the “oxidative burst” (Lamb and Dixon 1997). To counteract oxidative stress, plants employ enzymatic [superoxide dismutases (SODs), catalases and peroxidaes etc.] and non-enzymatic scavenging mechanisms (carotenoids, xanthophylls, glutathione, tocopherol, ascorbate etc.). Understanding how these enzymes are regulated at the transcriptional, post-transcriptional and post-translational levels is of fundamental importance.

MicroRNA (miRNA)-guided post-transcriptional gene regulation is essential for normal growth and development and adaptation to stress conditions (Phillips et al. 2007; Sunkar et al. 2007; Zhou et al. 2007). Characterization of miRNAs involved in plant stress responses is an active area of research (Phillips et al. 2007; Sunkar et al. 2007; Zhou et al. 2007). Among the ~20 conserved miRNA families in plants, miR398 can be directly linked with stress regulatory networks, because it targets two Cu/Zn Superoxide dismutases (cytosolic CSD1 and chloroplastic CSD2). CSD1 and CSD2 mRNA levels are often found to be elevated under diverse abiotic stress conditions (Sharma and Davis 1997). Recent studies indicated that the induction of CSD1 and CSD2 mRNA levels in Arabidopsis seedlings exposed to high Cu2+ or high Fe3+ can be attributed to the decreased miR398 levels (Sunkar et al. 2006). Conversely, decreased CSD1 and CSD2 transcript abundance in response to low Cu2+ or high sugar levels has been shown to correlate with the induced miR398 expression (Abdel-Ghany and Pilon 2008; Dugas and Bartel 2008). These observations indicated that modulation of CSD1 and CSD2 mRNA levels under stress seems to be directly dependent on the nature of miR398 response.

To extend our previous report that miR398 levels are regulated by abiotic stress, we used ozone fumigation and salt stress that were not tested previously. Furthermore, we hypothesized a role for miR398 during bacterial infections, because Pseudomonas infection results in an oxidative burst (Dempsey et al. 1999). We report here that miR398 levels were down-regulated in response to ozone fumigation and salt stress as well as in bacterial infiltrations, avirulent strains of Pst DC3000 (avrRpm1) or Pst DC3000 (avrRpt2). Additionally, increased CSD1, but not CSD2, mRNA levels negatively correlated with the decreased miR398 levels under ozone, salinity and biotic stress conditions. These results further confirm that miR398 is associated with abiotic and biotic induced oxidative stress in Arabidopsis.

Materials and methods

Plants of Arabidopsis ecotype Columbia-0 (seeds from Lehle Seeds, Round Rock, TX, USA) were grown in a controlled growth chamber at 21 ± 1°C with a 10 h light/14 h dark photoperiod.

Abiotic stress treatments

For ozone treatment, 3-week-old Col-0 plants were exposed to 300 nL L−1 ozone for 6 h from a UV-based ozone generator connected to an oxygen tank by Tygon tubing. Ozone levels were monitored with an ozone monitor. Rosette leaves were harvested during ozone fumigation as well as during recovery from stress. For NaCl and copper stress 15-day-old seedlings grown on MS-medium supplemented with (1%) sucrose were used. For imposing salt stress, seedlings were transferred to 200 mM NaCl solution containing dishes and harvested at 0, 3, 6, 12, 24, 48 and 72 h after salt stress. For Cu2+ stress, seedlings grown on MS-medium were sprayed with 100 μM Cu2+ and harvested 8 and 24 h later.

Biotic stress treatments

Four-week-old Arabidopsis leaves were infiltrated with Pst DC3000 carrying either the vector only or vector containing the avr genes avrRpt2 or avrRpm1 at 5 × 107 colony-forming units (cfu) per milliliter. Rosette leaves were harvested at 6, 12, 24 and 48 h after infiltration. Uninfected leaves (0 h inoculation) served as controls. The leaves and tissues in all treatments were frozen in liquid nitrogen and stored at −80°C.

Northern-blot analysis and real-time PCR

Total RNA was extracted from seedlings or rosette leaves with TRIzol reagent (Invitrogen). miR398 and CSD1, CSD2 expression analysis was determined as reported previously (Sunkar et al. 2006). Real-Time PCR was carried out using the same RNA samples that were used for northern analysis. Total RNA (2 μg) was treated with DNAse I and reverse transcribed using oligo-dT primer, reverse transcriptase and deoxy-nucleotides. Real-Time PCR analysis was carried out using MaximaTM SYBR Green qPCR Master Mix in a 7500 Real-Time PCR System using 100 ng cDNA and 7.5 picomole of each gene specific primer. The analysis was performed using two independent cDNA preparations and triplicate PCR. The relative expression ratio was calculated using \( 2^{- \Updelta \Updelta {{\text{Ct}}}} \) method using actin as reference gene.

Results and discussion

Ozone and salt stress down-regulate miR398 levels

It was found in a previous study that miR398 levels were down-regulated in response to the stress imposed by methyl viologen, high light and high Cu2+ (Sunkar et al. 2006). Here we have characterized the response of miR398 to additional abiotic stresses such as ozone and salinity stress.

Exposure of plants to ozone results in the accumulation of ROS in the apoplast (Sharma and Davis 1997). To examine the response of miR398 to ozone stress, 3-week-old Arabidopsis plants grown on soil under controlled conditions were transferred to an ozone chamber and exposed to ozone fumigation for 6 h. The rosette leaves were harvested after 1, 3 and 6 h of ozone stress. The plants were then transferred to a growth chamber for recovery. The rosette leaves were again harvested after 6 and 18 h of recovery time. Figure 1 shows that miR398 levels were progressively decreased in rosette leaves during ozone stress. The levels of miR398 continued to decrease following 6 h of recovery, but were gradually restored to normal levels by the 18 h of recovery from the ozone stress. These results confirm that miR398 is intimately linked with the oxidative stress.
https://static-content.springer.com/image/art%3A10.1007%2Fs00425-009-0889-3/MediaObjects/425_2009_889_Fig1_HTML.gif
Fig. 1

Effect of ozone fumigation and NaCl salinity on miR398, CSD1 and CSD2 levels. a Time course analysis of miR398, CSD1, and CSD2 levels in response to ozone fumigation and recovery from the ozone stress. b Relative expression ratios of CSD1 and CSD2 transcripts at 6 and 18 h of recovery from ozone fumigation compared to untreated control (0 h) as determined by real-time PCR analyses. c Time course analysis of miR398, CSD1, and CSD2 expression in response to 200 mM NaCl treatment. d Relative expression ratios of CSD1 and CSD2 transcripts at 48 and 72 h of NaCl salinity compared to untreated control (0 h) as determined by real-time PCR analyses. Standard deviation is indicated by error bars

To determine the effect of altered miR398 levels on its target genes during ozone fumigation and recovery, CSD1 and CSD2 mRNA levels were analyzed using Northern blot and real-time PCR in the same RNA samples in which miR398 levels were measured. CSD1 mRNA levels were up-regulated, whereas CSD2 mRNA levels were down-regulated during ozone fumigation (Fig. 1a, b). The differential response of CSD1 and CSD2 accumulation during ozone fumigation is consistent with a previous report (Sharma and Davis 1997). Our data show additional differential response patterns between CSD1 and CSD2 at the 18 h recovery time point. At 18 h, only CSD1 mRNA levels were decreased demonstrating an opposite correlation with miR398 levels (Fig. 1a, b).

To examine whether salinity stress imposed by NaCl affects miR398 expression, 15-day-old Arabidopsis seedlings were exposed to salt stress (200 mM NaCl), and the seedlings were harvested after 3, 6, 12, 24, 48 and 72 h of stress. A moderate reduction in miR398 expression was evident in seedlings exposed to NaCl for 12 h and onwards (Fig. 1c). Conversely, CSD1 and CSD2 transcript levels were moderately elevated as the duration of exposure increased. CSD1 showed a comparatively early induction that corresponded with miR398 suppression. CSD2 induction was only noticeable at 48 h and onwards during salt stress (Fig. 1c, d).

Pseudomonas syringae infection decreases miR398 levels

ROS levels accumulate in leaves challenged with the Psuedomonas syringae (Dempsey et al. 1999). Under similar conditions, Cu/Zn SOD protein levels were found to be elevated. These SOD elevations are likely to detoxify increased ROS, particularly superoxide radicals (Kliebenstein et al. 1999). These observations suggest a role for miR398 during pathogen infection also, because under diverse conditions CSD1 and CSD2 mRNA levels were found to be under miR398 regulation (Sunkar et al. 2006). To examine role of miR398 during biotic stress, miR398 expression was analyzed in Arabidopsis leaves infiltrated with avirulent strains of Pst (avrRpm1) or Pst (avrRpt2), or virulent strain Pst DC3000. Figure 2a and c shows that miR398 levels were decreased in leaves infiltrated with both avirulent pathogens. The Pst (avrRpm1) infiltration decreased miR398 abundance by 12 h and continued until 24 h post-inoculation (Fig. 2a). With Pst (avrRpt2) infiltration a similar trend was noticed, although suppression of miR398 was not as strong (Fig. 2c). Virulent PstDC3000 had very minor effects on miR398 levels although a small decrease was observed 24 h post-infection (Fig. 2e). In Arabidopsis, the miR398 family is represented by three loci: MIR398a, MIR398b and MIR398c (Sunkar et al. 2006). Semi-quantitative RT-PCR was performed with primers specific for each locus and the results confirmed that the down-regulation of all three loci (MIR398a, MIR398b and MIR398c) occurs in leaves challenged with avirulent but not virulent strains of Pseudomonas (data not shown).
https://static-content.springer.com/image/art%3A10.1007%2Fs00425-009-0889-3/MediaObjects/425_2009_889_Fig2_HTML.gif
Fig. 2

Effect of virulent and avirulent strains of bacterial pathogen Pseudomonas syringae pv. Tomato, on miR398, CSD1, CSD2 and PR1 levels. a Time course analysis of miR398, CSD1, CSD2 and PR1 levels in response to avirulent P. syringae pv. tomato Pst DC3000 (avrRpm1). b Relative expression ratio of CSD1 and CSD2 transcripts as compared to untreated control (0 h) in response to avirulent Pseudomonas syringe pv. tomato (Pst) (avrRpm1) as determined by real-time PCR analyses. c Time course analysis of miR398, CSD1, CSD2, and PR1 levels in response to avirulent Pst DC3000 (avrRpt2). d Relative expression ratio of CSD1 and CSD2 transcripts as compared to untreated control (0 h) in response to avirulent Pst DC3000 (avrRpt2) as determined by real-time PCR analyses. e Time course analysis of miR398, CSD1, CSD2 and PR1 levels in response to virulent Pst DC3000. f Relative expression ratio of CSD1 and CSD2 transcripts as compared to untreated control (0 h) in response to virulent PstDC3000 as determined by real-time PCR analyses. Standard deviation is indicated by error bars

During pathogen infection, ROS temporal kinetics vary depending on the pathogen strain involved. The oxidative burst induced by avirulent pathogens frequently adopts a biphasic generation of ROS, with the second phase accompanied by local cell death. In case of interactions with the virulent strains the oxidative burst was either absent or only the early phase was stimulated (Scheel 2002). Our data show that infection with avirulent pathogens avrRpm1 or avrRpt2 that cause hypersensitive responses, generate an oxidative burst and down-regulate miR398 abundance in Arabidopsis.

The levels of target genes (CSD1 and CSD2) were also monitored in the same RNA samples in which miR398 levels were analyzed.CSD1 levels were up-regulated and this correlated with down-regulated miR398 levels in leaves inoculated with both Pst (avrRpm1) and Pst (avrRpt2) (Fig. 2a–d). In contrast to CSD1 transcript levels, CSD2 levels were not induced by any of the bacterial pathogen infiltrations, either virulent or avirulent. Indeed, CSD2 levels were down-regulated (Fig. 2a, c, e). The pathogenesis related 1 (PR1) expression was monitored in the same blots and this served as a control for the pathogen treatment. As anticipated, an early and strong induction of PR1 was noticed in leaves infiltrated with avirulent pathogens (carrying avrRpm1or avrRpt2), and a late and weak induction was observed in leaves infected with the virulent pathogen, Pst (DC3000) (Fig. 2a, c, e).

The gene expression programs in response to diverse biotic and abiotic stimuli, utilize a complex array of transcriptional, post-transcriptional and post-translational events. Several recent studies found negative correlations between miR398 and it’s target mRNAs (CSD1 and CSD2), using high light, paraquat, high Cu2+, high Fe3+ stress or high sucrose levels or low Cu2+ levels in the growth medium (Sunkar et al. 2006; Yamasaki et al. 2007; Dugas and Bartel 2008). Interestingly, in this study we found that CSD2 mRNA abundance is not strictly under the control of miR398 during stress imposed by ozone (abiotic) or P. syringae (biotic). In contrast, CSD1 mRNA levels are negatively correlated with miR398 levels during diverse stress conditions (Figs. 1, 2). The differential responses of CSD2 mRNA levels to diverse stress conditions (increased during high Cu2+ stress but decreased during pathogen and ozone fumigation) suggests the existence of regulatory mechanisms dependent and independent of miR398-guided post-transcriptional regulation in Arabidopsis. Consistent with this evidence, it has been recently shown that aconitase binds to the 5′ UTR of CSD2 mRNA and regulates CSD2 mRNA stability in Arabidopsis and tobacco (Moeder et al. 2007). Furthermore, CSD2 transcript levels were higher in aconitase knockout plants relative to wild-type plants, whereas the level of CSD1 was unaffected (Moeder et al. 2007).

Previously, we showed that decreased miR398 levels cause up-regulation of CSD1 and CSD2 mRNA levels in Arabidopsis seedlings exposed to oxidative stress (high light, high Cu2+ or high Fe3+) (Sunkar et al. 2006). Here we show that down-regulation of miR398 is common response to oxidative stress caused by abiotic and biotic stress. Furthermore, the involvement of miR398 in oxidative stress responses is well supported by the decreased miR398 levels during ozone fumigation but its gradual restoration to normal levels after relieved from the stress. Because miR398 is highly conserved in higher plants, its regulation and function are also likely to be conserved in other plants (Zhang et al. 2006, 2008; Sunkar and Jagadeeswaran 2008). Interestingly, unlike the previous reports (Sunkar et al. 2006; Yamasaki et al. 2007; Dugas and Bartel 2008), we found that CSD2 mRNA levels are not always negatively correlated with miR398 levels during diverse stress conditions causing oxidative stress. Plant miRNAs primarily affect the target mRNA levels by directing mRNA cleavage (Jones-Rhoades et al. 2006). However recent data show that miRNA in plants affect protein translation to a greater extent than previously thought (Brodersen et al. 2008). Studying protein levels of CSD2 under diverse stress conditions could clarify if miR398 is affecting the CSD2 protein levels during the stress conditions examined in this study.

Involvement of miRNAs in plant pathogen interactions has been reported recently. miR393 levels were up-regulated in Arabidopsis leaves infiltrated with bacterial flagellin or P. syringae pv. Tomato (Fahlgren et al. 2007; Navarro et al. 2006). In this study, we found that miRNAs are also negatively regulated by bacterial infections. To our knowledge, miR398 is the first miRNA found to be down-regulated in response to biotic stress (P. syringae). The mechanism for this response is under further study, however it is possible that altered miRNA (up- or down-regulated) levels regulate their target gene(s), which modulate adaptation to pathogen resistance.

Acknowledgments

This work was supported by the Oklahoma Agricultural Experiment Station and USDA-(NRI-2007-02019) to R. Sunkar. Ajay Saini is supported by the BOYSCAST fellowship from the Department of Science and Technology, India. We sincerely thank Dr. Jeff Dangl and Dr. Ramesh Raina for providing the pathogen strains, Drs. N. Jambunathan for the initial help with the experiments and R. Mahalingam for permitting us to use the ozone chamber.

Copyright information

© Springer-Verlag 2009