Growth suppression, altered stomatal responses, and augmented induction of heat shock proteins in cytosolic ascorbate peroxidase (Apx1)-deficient Arabidopsis plants

Authors


* For correspondence (fax +1 515 294 1337; e-mail rmittler@iastate.edu).

Summary

The accumulation of hydrogen peroxide (H2O2) in plants is typically associated with biotic or abiotic stresses. However, H2O2 is continuously produced in cells during normal metabolism. Yet, little is known about how H2O2 accumulation will affect plant metabolism in the absence of pathogens or abiotic stress. Here, we report that a deficiency in the H2O2-scavenging enzyme, cytosolic ascorbate peroxidase (APX1), results in the accumulation of H2O2 in Arabidopsis plants grown under optimal conditions. Knockout-Apx1 plants were characterized by suppressed growth and development, altered stomatal responses, and augmented induction of heat shock proteins during light stress. The inactivation of Apx1 resulted in the induction of several transcripts encoding signal transduction proteins. These were not previously linked to H2O2 signaling during stress and may belong to a signal transduction pathway specifically involved in H2O2 sensing during normal metabolism. Surprisingly, the expression of transcripts encoding H2O2 scavenging enzymes, such as catalase or glutathione peroxidase, was not elevated in knockout-Apx1 plants. The expression of catalase, two typical plant peroxidases, and several different heat shock proteins was however elevated in knockout-Apx1 plants during light stress. Our results demonstrate that in planta accumulation of H2O2 can suppress plant growth and development, interfere with different physiological processes, and enhance the response of plants to abiotic stress conditions. Our findings also suggest that at least part of the induction of heat shock proteins during light stress in Arabidopsis is mediated by H2O2 that is scavenged by APX1.

Introduction

Hydrogen peroxide (H2O2) is a signaling molecule involved in the control of key biological processes, such as programmed cell death (PCD), abiotic stress responses, hormonal signaling, and pathogen defense (Dat et al., 2000; Hirt, 2000; Kovtun et al., 2000; Mittler, 2002; Mullineaux and Karpinski, 2002; Pei et al., 2000). It is formed in cells by the direct transfer of two electrons to O2, mediated by enzymes such as glycolate or glucose oxidase, or by the dismutation of superoxide (O2) to H2O2, catalyzed by superoxide dismutases (SODs). Under controlled growth conditions, the production of O2 by photosynthetic cells is estimated at a constant rate of 250 µmol mg−1 chl h−1, and the steady-state level of H2O2, produced by SODs and other cellular sources, at 0.5 µmol g−1 fresh weight (Asada and Takahashi, 1987; Polle, 2001). These rates reflect the background level of H2O2 production in photosynthetic tissues in the light. In the dark, or in non-photosynthetic tissues, H2O2 is mainly formed by the dismutation of O2, produced by leakage of electrons from electron transfer carriers in the mitochondria or microsomes, or by fatty acid oxidation (Corpas et al., 2001; Davidson and Schiestl, 2001).

In response to pathogen infection, the rate of H2O2 production is dramatically enhanced in cells as a result of the activation of O2-producing enzymes, such as NADPH oxidases, and the dismutation of O2 to H2O2 by SODs (Hammond-Kosack and Jones, 1996; Mittler et al., 1999). H2O2 is also formed during pathogen infection by the enzymatic activity of amine oxidases and cell wall-bound peroxidases (Allan and Fluhr, 1997; Vranova et al., 2002). Under these conditions, H2O2 activates different signal transduction pathways essential for programmed cell death and pathogen defense (Hammond-Kosack and Jones, 1996). During abiotic stresses, H2O2 is formed by NADPH oxidases (Cazale et al., 1999; Knight and Knight, 2001; Pastori and Foyer, 2002), by specific cellular pathways associated with stress metabolism, such as the photorespiratory pathway (Corpas et al. 2001), and by the uncoupling of metabolic reactions that result in the leakage of electrons from different electron carriers to the reduction of O2 (Asada, 1999; Dat et al., 2000; Mittler, 2002). The enhanced production of H2O2 via these routes is also thought to act as a signal that activates defense mechanisms and mediates the acclimation or hardening of plants to extreme environments (Bowler and Fluhr, 2000).

Because H2O2 is toxic and yet participates in key signaling events, plant cells require different cellular mechanisms that regulate their intracellular level of H2O2. These include ascorbate peroxidase (APX; Asada, 1999), catalase (CAT; Willekens et al., 1997), and glutathione peroxidase (GPX; Roxas et al., 1997). In addition, the balance between SODs and APX, catalase, or glutathione peroxidase activities is considered to be crucial for determining the steady-state level of O2 and H2O2. This balance, together with sequestering of metal ions such as iron and copper, is thought to be important to prevent the formation of the highly toxic HO˙ via the metal-dependent Haber–Weiss or the Fenton reactions (Asada and Takahashi, 1987; Bowler et al., 1991; Haber and Weiss, 1934). Antioxidants such as ascorbic acid and glutathione, found at very high concentrations in chloroplasts and other cellular compartments, are also crucial for the defense of plants against oxidative stress (Noctor and Foyer, 1998). Consequently, mutants with suppressed ascorbic acid levels (Conklin et al., 1996) and transgenic plants with suppressed H2O2-scavenging enzymes (Mittler et al., 1999; Orvar and Ellis, 1997; Willekens et al., 1997) are hypersensitive to abiotic stress conditions and pathogen attack. In addition, overexpression of O2- and H2O2-scavenging enzymes was found to increase the tolerance of plants to abiotic stress conditions (Allen, 1995).

Recent studies identified a number of signal transduction components involved in the detection of H2O2 and the activation of defense mechanisms in plants. These include: a two-component histidine kinase, a receptor-like protein kinase, the MAPKKK, AtANP1 (also the NtNPK1), and the MAPKs, AtMPK3/6 and Ntp46MAPK (Czernic et al., 1999; Desikan et al., 2001; Kovtun et al., 2000; Samuel et al., 2000; Vranova et al., 2002). In addition, calmodulin has been implicated in H2O2 signaling (Desikan et al., 2001; Harding et al., 1997). However, our current knowledge of the H2O2 signal transduction pathway of plants is very limited (Mittler, 2002; Pastori and Foyer, 2002). Moreover, it is solely based upon the studies in which stress conditions or H2O2 was externally applied to plants to activate the H2O2 signal transduction pathway (Desikan et al., 2001; Kovtun et al., 2000). Because stress treatments or external H2O2 application may activate additional signal transduction pathways such as pathogen-response pathways or general stress-response pathways, these treatments may complicate the analysis of the H2O2 signal transduction pathway of plants.

To study the response of plants to elevated in planta levels of H2O2 in the absence of abiotic stress, pathogens, or oxidants, we introduced a lesion in the H2O2 scavenging machinery of plants and studied plants growing under optimal conditions. Under these conditions, no external stress is imposed on plants; instead, H2O2 levels within cells are specifically elevated because the expression of an H2O2-metabolizing enzyme is disrupted. For our study, we chose to inactivate the gene encoding the major cytosolic isoform of ascorbate peroxidase (Apx1; Mittler and Zilinskas, 1992). In contrast to all other isoforms of APX, APX1 is highly responsive to various biotic and abiotic stresses, and is considered to play an important role in H2O2 scavenging in plants (Asada, 1999; Mittler, 2002; Shigeoka et al., 2002).

Results

Molecular characterization of Apx1-deficient Arabidopsis plants

As shown in Figure 1, we isolated an Arabidopsis line (cv. WS) containing a T-DNA insert in the second exon of Apx1. This line was isolated from the Wisconsin T-DNA collection (Sussman et al., 2000). In contrast to tobacco plants expressing an antisense construct to APX1 (Mittler et al., 1999; Orvar et al., 1997), knockout-Apx1 Arabidopsis plants had a late flowering and delayed development phenotype, suggesting that the lack of Apx1 affected plant development in Arabidopsis (Figure 1b; Table 1). This phenotype was enhanced in plants grown under long days (18 h light cycle) or constant light, and was almost completely abolished in plants grown under short days (8 h light cycle; data not shown). Protein and RNA blots performed on wild-type and knockout-Apx1 plants confirmed that homozygote knockout-Apx1 plants did not contain detectable levels of APX1 protein or RNA (Figure 1c). Activity measurements indicated that total APX activity in knockout-Apx1 plants was suppressed by 70% compared to that of wild types (data not shown).

Figure 1.

Characterization of APX1-deficient plants.

(a) A map showing the site of T-DNA integration into the Apx1 gene.

(b) A photograph of 17-day-old wild-type (WT) and knockout-Apx1 (KO-APX) plants grown under constant light (100 µmol m−2 sec−1).

(c) A Coomassie-stained protein gel (left), a protein gel blot performed with APX1-specific antibodies (middle), and an RNA blot (right), showing that knockout-Apx1 (KO-APX) plants do not contain APX1 mRNA or protein. Methods and experimental protocols are described in Experimental procedures. #2, heterozygous; #8, homozygous.

Table 1.  Developmental time course of wild-type (WT) and knockout-Apx1 plants
Time (days)WTKnockout-Apx1
  1. Eighty wild-type and knockout-Apx1 plants were scored. Unless otherwise stated, more than 95% of the plants are as described. Plants were grown at 21–22°C, constant light 100 µmol m−2 sec−1, and a relative humidity of 70%. This analysis was repeated thrice with similar results.

0End of vernalizationEnd of vernalization
4First two cotyledonsFirst two cotyledons
10First two true leavesFirst two true leaves
17Five true leaves and bolting 2.5 cmSeven true leaves
19Inflorescence stem 6 cmEight true leaves, starting to bolt
23Inflorescence stem 13–15 cm70% Inflorescence stem 0.75 cm
30% Inflorescence stem 5 cm
26Inflorescence stem 20–22 cm, has siliques25% Inflorescence stem 15 cm
12.5% Inflorescence stem 8 cm
12.5% Inflorescence stem 5 cm
50% Inflorescence stem 1 cm
27Inflorescence stem 24 cm, has siliques25% Inflorescence stem 18 cm, has siliques
12.5% Inflorescence stem 11 cm
12.5% Inflorescence stem 8 cm
50% Inflorescence stem 2–4 cm

To avoid complications resulting from developmental differences between plants, we conducted all our comparisons between wild-type and knockout-Apx1 plants with 14–17-day-old plants that were developmentally indistinguishable (all growing under optimal growth conditions, i.e. 21°C, 18 h or constant light cycle, 100 µmol m−2 sec−1, and a relative humidity of 70%). We performed all our assays in triplicates using a minimum of 60 plants per treatment (three replicates of at least 20 plants each) and repeated all experiments at least thrice.

Physiological characterization of knockout-Apx1 plants

For the physiological characterization of knockout-Apx1 plants, we placed wild-type and knockout-Apx1 plants in the dark for 30 min and shifted them back to light. We then measured the photosynthetic activity and stomatal conductance at different intervals using a Li-Cor LI-6400 apparatus. We also sprayed light-grown wild-type and knockout-Apx1 plants with a solution of abscisic acid (ABA; 50 µm) and measured the rate of stomatal closure. As shown in Figure 2, knockout-Apx1 plants had a lower rate of maximal photosynthetic activity (about 60% of wild-type activity, Figure 2a) and altered stomatal responses (Figure 2b). Thus, compared to wild-type plants that opened their stomata upon a shift from dark to light, the stomata of knockout-Apx1 plants were almost non-responsive to this treatment (Figure 2b). The response of knockout-Apx1 plants to abscisic acid application was, however, similar to that of wild types (Figure 2c).

Figure 2.

Suppressed photosynthetic activity and abnormal guard cell responses in knockout-Apx1 (KO-APX) plants.

(a) A graph showing the suppression of photosynthetic activity in KO-APX plants compared to WT plants.

(b) A graph showing the abnormal guard cell response of KO-APX plants upon shift of plants from dark to light. In (a) and (b) plants were placed in the dark for 30 min and then transferred to light (100 µmol m−2 sec−1). Photosynthesis (a) and guard cell responses (b) were recorded simultaneously every minute, starting immediately upon the transfer of plants to light.

(c) A graph showing the closure of stomata in WT and KO-APX plants upon ABA application in the light. Additional methods and experimental protocols are described in Experimental procedures. Data shown is mean and standard deviation of three independent measurements.

DNA array and biochemical analysis of knockout-Apx1 plants

To study changes in gene expression resulting from the lack of APX1 in Arabidopsis plants (grown under optimal conditions), we performed a DNA array analysis using Affymetrix chips (8200 gene chips). Our results expressed as mean and standard error for three different measurements are shown in Tables 2–4. Table 2 shows the expression level of different O2- and H2O2-scavenging enzymes in knockout-Apx1 plants compared to wild-type plants. Surprisingly, we did not detect an increase in the expression level of the other APX isozymes (including APX4 and APX5, not shown in the table) in knockout-Apx1 plants. Moreover, in contrast to tobacco plants expressing an antisense construct to APX1 in which catalase and CuZnSOD expression was elevated (Rizhsky et al., 2002a), the expression of catalase and at least one isozyme of CuZnSOD was suppressed in knockout-Apx1 plants.

Table 2.  Expression levels of different transcripts involved in O2 and H2O2 scavenging in wild-type (WT) and knockout-Apx1 plants, in the absence of stress
TranscriptExpression level compared to WT (% of control ± SD)
WTKnockout-Apx1
  1. Results, presented as percentage of control compared to WT plants, are expressed as average ± SD of three independent measurements each performed with a different chip. RNA was extracted from 120 to 150 WT and knockout-Apx1 plants and hybridized to Affymetrix chips (8200 gene chip) as described in Experimental procedures. Transcripts indicated in bold are significantly different between WT and knockout-Apx1 (P = 0.01) plants. Plants were grown at 21–22°C, constant light 100 µmol m−2 sec−1, and a relative humidity of 70%. Abbreviations used: APX, ascorbate peroxidase; CAT, catalase; GR, glutathione reductase; GPX, glutathione peroxidase; MDAR, monodehydroascorbate reductase; nd, not detected; SD, standard deviation; SOD, superoxide dismutase; str, stroma; tyl, thylakoid; chl, chlorophyll; cyt, cytochrome.

CuZnSOD (chl)100 ± 6.248 ± 1.9
CuZnSOD (cyt)100 ± 16128 ± 16
FeSOD100 ± 1195 ± 18
APX1100 ± 4.22 ± 0.01
APX2ndnd
APX3100 ± 887 ± 2.3
APX (tyl)100 ± 6.972 ± 8.8
APX (str)100 ± 1671 ± 14
MDAR100 ± 7.260 ± 9
GR1100 ± 768 ± 11
GR2100 ± 16117 ± 12
CAT1100 ± 4.167 ± 1.6
CAT100 ± 7.274 ± 2
GPX1100 ± 7118 ± 13
GPX2100 ± 5102 ± 4.4
GPX (phospholipid)100 ± 1.5105 ± 3
Table 3.  Transcripts elevated in knockout-Apx1 (KO-APX) plants compared to wild-type (WT) plants in the absence of stress or pathogen infection
 Average log2 foldSE1 h light stress signal log2 foldOther stressesGenBank numberDescription
  1. Results (mean and SE) are presented as fold induction (log2) over WT expression. Two accession numbers are given to each transcript, Affymetrix (left) and Genebank (right). The known or putative function of each transcript is also given (right column). The induction/suppression of each transcript following a 1 h light stress, expressed as fold (log2) over WT, is indicated in a separate column (+, induction; −, suppression; nc, no change). The induction of each transcript by different stressful conditions (C, cold; D, drought; H, hydrogen peroxide; Hs, heat shock; I, excess iron; LO, low oxygen; O, high oxygen; P, pathogen; S, salt; W, wounding) is indicated in the other stresses column. This information was collected from various sources as indicated in the Supplementary Material. RNA preparation and analysis by Affymetrix chips (Arabidopsis 8200 gene chip) are described in Experimental procedures and as Supplementary Material.

Signal transduction transcripts elevated in KO-APX plants
 15616_s_at1.800.16(+) 1.5 L04999Serine threonine kinase (pro25)
 12497_at1.670.21(+) 1.3WF20M17.8Putative receptor-like kinase
 17917_s_at1.630.05nc T3K9.14Calcium-binding protein (CaBP-22)
 13177_at1.630.25ncPT20K18.70Growth factor like protein
 12353_at1.430.05(+) 0.5 F13M22.21Receptor-like kinase
 18003_at1.300.08nc AF188334Toll/interleukin-1 receptor-like
 17499_s_at1.270.19(+) 0.8 AF107726Cyclic nucleotide gated channel
 19848_s_at1.230.12nc T3K9.13Calmodulin-like protein
 19433_at1.170.26(+) 1.1WF21P8.160Serine/threonine kinase
 13022_at1.100.16ncC/D/SF21H2.4Protein phosphatase type 2C
 16952_s_at0.900.08ncWF13H10.4Calmodulin-like protein
 17114_s_at0.670.09nc D21840MAP kinase (MPK4)
 17991_g_at0.530.05ncWAF178075Calmodulin 9 (CAM9)
Transcription factors transcripts elevated in KO-APX plants
 15779_g_at1.600.29ncWX98676ZAT7, zinc finger protein
 20382_s_at1.500.50ncST9D9.6WRKY-type DNA-binding
 13271_g_at1.330.47ncHs/WU68561Heat shock transcription factor 21
 13015_s_at1.230.21(−) 0.9P/H/WX98673ZAT12, zinc finger protein
 18217 _g_at0.870.26(− 1)SX95573Salt-tolerance zinc finger protein
 13273_at0.870.24ncHs/WU68017Heat shock transcription factor 4
 15203_s_at0.700.08ncS(−)/WAB013887RAV2, novel DNA-binding protein
 16064_s_at0.700.16(−) 0.7HAB008106Ethylene responsive factor 4
 18600_at0.630.17ncS/WL76926Putative zinc finger protein
 20032_at0.530.05nc F17L21.4Putative squamosa promoter-binding
Defense and detoxification transcripts elevated in KO-APX plants
 19178_at2.670.12ncP/HY18227Blue copper binding-like protein
 14635_s_at2.500.36(−) 1.6P/HM90508PR-1-like
 13212_s_at2.270.53(−) 0.5P/H/WM90509Beta-1,3-glucanase
 16031_at1.770.05(+) 1.3D/C/IX94248Ferritin
 14636_s_at1.670.25(−) 1P/H/WM90510Thaumatin-like
 16439_at1.500.08ncPX72022CXc750 pathogen-inducible
 14111_s_at1.430.63nc F18A5.290Putative disease resistance protein
 18604_at1.370.12ncHAF069298100aa heat, auxin, ethylene induced
 19181_s_at1.370.12(+) 0.4D/C/OAF053065Late embryogenesis abundant (21)
 12764_f_at1.370.05ncWT17M13.10Putative glutathione S-transferase
 16054_s_at1.130.09(−) 0.5WY14251Glutathine-S-transferase (GST11)
 13695_at1.100.22ncP/WAF021346Disease resistance protein (NDR1)
 15118_s_at1.000.28ncWL11601Glutathione S-transferase
 17119_s_at0.830.09ncWAF132212OPDA-reductase
 16981_s_at0.700.14(−) 1.2 U35829Thioredoxin h (TRX5)
 13186_g_at0.700.08ncDZ35475Thioredoxin
 12802_at0.700.00ncD/C/W/HT19K4.170DnaJ-like protein
 19426_s_at0.670.21ncWAF098964Disease resistance protein RPP1
 13189_s_at0.630.05(−) 1.1 Z35476(TOUL) thioredoxin
 16009_s_at0.630.05(+) 0.5P/H/WF17A22.12Glutathione S-transferase (GST6)
 16060_at0.530.05ncDD89051ERD6 early dehydration-induced
 16001_at0.530.05nc AF035385SEN5, senescence-associated
 16081_s_at0.500.08ncD/CAF141659AtHVA22a, ABA/stress induced
 13785_at0.500.08(−) 0.6C/WF14N22.20Cold-regulated, cor15b
 19665_at0.470.05(−) 0.4 X95585DAD-1, apoptosis suppressor
Cellular organization, metabolism and biogenesis transcripts elevated in KO-APX plants
 14704_s_at3.030.05(+) 0.9 F7H1.2Putat. retroelement pol polyprotein
 18567_at2.631.32nc F14M4.4Putat. alcohol dehydrogenase
 17008_at2.600.57nc F27C12.23Putat. tyrosine aminotransferase
 19121_at1.430.37(+) 1.1 AF055847AIR1 auxin-inducible
 17832_at1.430.21ncLOU94998Non-symbiotic hemoglobin (AHB1)
 19118_s_at1.400.49(+) 1.7 AF098630Cell wall-PM disconn. protein
 16702_at1.400.45nc T16F16.15Putative phloem-specific lectin
 20369_s_at1.100.08(+) 0.6 A44314Ammonium transport (AMT1)
 13617_at1.070.24nc F14M13.10Mito. dicarboxylate carrier
 19961_at1.030.17ncWF20M13.240Glycine-rich 2 (GRP2)
 15085_at0.830.12nc F7H19.200UDP-galactose transporter
 20446_s_at0.800.14nc T25N20.21Putative glucan synthase
 13091_r_at0.770.26(−) 0.3 T14P1.12Putative transport SEC61 beta
 16014_at0.770.05nc U11766GAST1, gibberellin-regulated
 17002_at0.670.12ncD/P/HAJ238804Lipid transfer protein
 19450_at0.670.12ncWX75365SUC1, sucrose-proton symporter
 15115_f_at0.670.05nc AF104330Glycine-rich protein (GRP3S)
 14116_at0.670.09nc AF077407Hexose transporter-like
 16950_s_at0.630.25nc Z26753Sec61 beta-subunit
 18673_at0.600.08nc D79218Non-coding RNA, cyt-repressed
 17882_at0.570.12nc F13M23.60SEC61 GAMMA
 19999_s_at0.570.09nc AB017977cAMP phosphodiesterase
 15452_at0.570.09(−) 0.3 Z49859Copper transporter
 16940_g_at0.570.12nc F25I18.15Putative synaptobrevin
 14567_at0.530.05nc F16M14.6Putative acetyltransferase
 20190_at0.530.12ncD/CT24I21.7Nodulin-like protein
 13093_at0.500.14nc T24C20.20B12D-like, seed development
 19969_at0.500.08nc F9L1.5Ubiquinol-cyt-c-reductase
Table 4.  Transcripts suppressed in knockout-Apx1 (KO-APX) plants compared to wild-type (WT) plants in the absence of stress or pathogen infection
 Average signal
log2 fold
SE1 h light stress
signal log2 ratio
Other stressesGenBank numberDescription
  1. Results (mean and SE) are presented as fold induction (log2) over WT expression. Two accession numbers are given to each transcript, Affymetrix (left) and GenBank (right). The known or putative function of each transcript is also given (right column). The induction/suppression of each transcript following a 1 h light stress, expressed as fold (log2) over WT, is indicated in a separate column (+, induction; −, suppression; nc, no change). The induction of each transcript by different stressful conditions (C, cold; D, drought; H, hydrogen peroxide; Hs, heat shock; I, excess iron; LO, low oxygen; O, high oxygen; P, pathogen; S, salt; W, wounding) is indicated in the other stresses column. This information was collected from various sources as indicated as Supplementary Material. RNA preparation and analysis by Affymetrix chips (Arabidopsis 8200 gene chip) are described in Experimental procedures and as Supplementary Material.

Signal transduction transcripts suppressed in KO-APX plants
 13587_at0.630.26nc F21M11.2Acid phosphatase
 16124_s_at0.570.17nc AF053366Blue light photoreceptor PHR2
 15135_s_at0.600.22(+) 0.6 U01955Laminin receptor-like protein
Transcription factors transcripts suppressed in KO-APX plants
 19887_s_at0.600.00(+) 0.4 U75599Leucine zipper protein
 15676_at0.530.34nc AF138743Zinc finger protein 1 (zfn1)
 14723_f_at0.500.08(−) 0.6 AF003096AP2 domain protein RAP2.3
Defense and detoxification transcripts suppressed in KO-APX plants
 16429_at3.730.19(−) 5.6H/O/WU63815Ascorbate peroxidase (APX1)
 18755_at2.300.80nc F14M19.60Pathogenesis-related protein
 15581_s_at1.070.09ncO(−)AF061519Plastid Cu/Zn SOD
 15606_at1.200.16nc AF061517Plastid Cu/Zn SOD chaperone (CCS1)
 15776_at0.900.14ncD/WD10703Dehydration-induced RD22
 17196_at0.830.12(−) 0.7 F16A16.110Proline-rich APG-like protein
 16629_at0.770.05(−) 1WAF087932Hydroperoxide lyase (HPL1)
 17018_s_at0.630.05(+) 0.4W(−)U18929Cytochrome P450 monooxygenase
 12752_s_at0.600.16(+) 0.3 T8O5.170Peroxidase prxr1
 13154_s_at0.570.12(−) 1WF18O19.30Putative endochitinase
 14856_at1.030.76ncWF13P17.32Putative cytochrome P450
 20442_i_at0.500.16(+) 0.5WF3O9.21Putative cytochrome P450
 13218_s_at0.570.12(+) 0.5H/W(−)AF021937Catalase 3 (CAT3)
Cellular organization, metabolism and biogenesis transcripts suppressed in KO-APX plants
 16575_s_at2.270.26nc L40954Oleosin
 18859_at1.630.75nc F8A5.29Putative clathrin coat assembly protein
 20227_s_at1.370.21(−) 2.2 AB027252f-AtMBP myrosinase binding protein
 16991_at1.030.21(−) 0.4 L73G19.10Fibrillarin-like protein
 18215_at0.930.25nc Z97335Selenium-binding protein like
 17572_s_at0.900.22nc AF083036Ammonium transporter
 12381_at0.870.17nc AB003522Beta subunit of coupling factor one
 15186_s_at0.870.25ncHAF023167Adenosine-5-phosphosulfate red. (APSR)
 13120_at0.830.05nc T9A14.50Extensin-like protein
 17517_at0.770.09nc L41245Thionin (Thi2.2)
 12412_at0.770.05(+) 0.4 T1J8.6S-adenosylmethionine synthetase
 20640_s_at0.770.12(+) 0.3 S45911Glyceraldehyde-3-pho.-dehyd. (GapB)
 18696_s_at0.730.19ncHU96045APS reductase (PRH)
 13099_s_at0.730.19(+) 0.9 T22J18.12Sucrose-proton symporter (SUC2)
 12844_s_at0.730.33(−) 0.7 X1607718S rRNA gene
 15141_s_at0.730.21(−) 1.4 D85191Vegetative storage protein
 15182_at0.700.16nc D85339Hydroxypyruvate reductase
 13588_at0.700.08(+) 0.4 F28A23.40Phosphoglycerate dehydrogenase
 16504_at0.700.08nc Z97335Hydroxymethyltransferase
 15990_at0.670.09nc S74719Sedoheptulose-1,7-bisphosphatase
 15992_s_at0.670.09(+) 0.3 X16432Elongation factor 1-alpha
 13682_s_at0.630.33nc F14P13.17Lycopene beta cyclase
 15627_at0.630.33(+) 0.4 U80186Pyruvate dehydrogenase E1 beta
 16431_at0.630.09(−) 0.5 T6A23.27Nonspecific lipid transfer protein
 18439_s_at0.630.09(+) 0.4 X97484Putative phosphate transporter
 16994_at0.630.05nc F6E13.2560S ribosomal protein L7
 17394_s_at0.600.08nc M64115Glyceraldehyde-3-pho. dehyd. (B)
 18484_at0.600.08(+) 0.4 T28I19.40Squalene epoxidase-like
 15642_at0.570.12nc AF129511Very long chain fatty acid condensing
 14645_at0.570.05(+) 0.5D/SD13043Thiol protease
 16926_at0.570.05nc Z97343Ribosomal protein
 16997_at0.570.17nc X75162BBC1 protein, cell division
 15145_s_at0.570.05nc D64155Possible aldehyde decarbonylase
 17374_at0.570.12nc F17L21.20Putative 60S ribosomal protein L17
 16106_at0.530.05nc U77381WD-40 repeat protein (AtArcA)
 16508_at0.530.05nc X94626AATP2-plastidic ATP/ADP transporter
 18683_s_at0.530.12ncCL27158Omega-3 fatty acid desaturase
 15837_at0.500.08(+) 0.4 T27A16.27Putative thiamin biosynthesis protein
 17386_at0.500.08ncWF26H11.10Putative proline-rich protein
 20117_at0.500.08(+) 0.5 Z97341Putative oligopeptide transporter

In Tables 3 and 4, we have included transcripts with a known or putative function, elevated (Table 3) or suppressed (Table 4), 0.5 (log2)-fold or higher, in knockout-Apx1 plants compared to wild-type plants. The different transcripts have been grouped on the basis of their putative function into: signal transduction components, transcription factors, defense and detoxification transcripts, and cellular organization, metabolism and biogenesis transcripts. In a separate column, we also indicate whether these transcripts are also induced by other stresses.

Concurring with our working hypothesis that the disruption of Apx1 will result in an internal oxidative stress to plants (growing under controlled conditions), the expression of a number of genes that were previously associated with oxidative stress was elevated in knockout-Apx1 plants (Table 3). These included the pathogenesis-related protein-1 (PR-1), blue-copper-binding protein, glutathione-S-transferase (GST6), ferritin, and the zinc finger protein ZAT12 (Desikan et al., 2001). The expression of a number of transcripts encoding signal transduction components and transcription factors was also elevated in knockout-Apx1 plants compared to wild-type plants. Some of these were previously associated with other stresses whereas others were not (to the best of our searching ability). Some of these transcription factors and signaling transcripts, elevated in knockout-Apx1 plants, may belong to the H2O2 signal transduction pathway of plants. In addition to the signal transduction and transcription factor transcripts, the expression of a number of defense, cellular organization, and metabolic transcripts was altered in knockout-Apx1 plants. These may serve as putative downstream transcripts regulated by the H2O2 signal transduction pathway in response to the absence of APX1.

To confirm the results obtained by the DNA chip analysis, we performed RNA blots with RNA from wild-type and knockout-Apx1 plants and probed these with some of the transcripts shown in Table 3. As shown in Figure 3(a), the results from the RNA blots were in good agreement with our DNA chip analysis.

Figure 3.

Enhanced expression of transcripts associated with oxidative stress and elevated levels of reduced glutathione (GSH) and H2O2 in knockout-Apx1 (KO-APX) plants.

(a) RNA gel blots showing the enhanced expression of different transcripts associated with oxidative stress in KO-APX plants.

(b) A graph showing the elevated levels of GSH and H2O2 in KO-APX plants compared to WT plants. Methods for RNA gel blots and biochemical analysis are described in Experimental procedures.

Data shown in (b) is mean and standard deviation of three independent measurements. HSF, heat shock transcription factor 21.

We also measured the level of reduced glutathione (GSH) and H2O2 in leaf extracts from wild-type and knockout-Apx1 plants. The level of these compounds can serve as a good measure of the degree of oxidative stress encountered by plants (Mittler, 2002; Noctor and Foyer, 1998). As shown in Figure 3(b), knockout-Apx1 plants had higher levels of reduced glutathione and H2O2 compared to wild-type plants.

Table 4 summarizes the transcripts suppressed in knockout-Apx1 plants grown under optimal growth conditions. In addition to chloroplastic CuZnSOD and catalase, a number of transcripts encoding signal transduction enzymes and transcription factors were suppressed in knockout-Apx1 plants. The suppression of these may be linked to the H2O2 signal transduction pathway. Additional tables showing the expression level of transcripts with unknown function, elevated or suppressed in knockout-Apx1 plants, can be found as Supplementary Material to this article.

Light stress of knockout-Apx1 plants

To further characterize knockout-Apx1 plants and to examine their response to a mild abiotic stress, we subjected wild-type and knockout-Apx1 plants to a moderate light stress of 425 µmol m−2 sec−1. As shown in Figure 4, this light stress resulted in the induction of APX1 and ELIP (early light inducible protein) in wild-type plants, and the induction of ELIP in knockout-Apx1. Interestingly, the induction of transcripts encoding the 70 kDa heat shock protein (HSP70), 1 h following the application of light stress, was much higher in knockout-Apx1 plants compared to that in wild-type plants (Figure 4). The conditions for light stress used in our experiment did not result in any apparent cell death in wild-type or knockout-Apx1 plants (not shown), or the induction of APX2, previously reported to be induced during high light stress (2000 µmol m−2 sec−1; Karpinski et al., 1999; Figure 4; also measured by the DNA chips and by RT-PCR, not shown; see Figure 6 for a positive control for APX2 induction). The induction of APX1 in wild-type plants during light stress (425 µmol m−2 sec−1; Figure 4) suggested that this level of light stress enhanced the production of H2O2 in Arabidopsis.

Figure 4.

Enhanced expression of heat shock protein HSP70 in knockout-Apx1 (KO-APX) plants during light stress.

RNA gel blots were performed with RNA obtained from control and light stress (425 µmol m−2 sec−1)-treated WT and KO-APX plants. The induction of HSP70 is shown to be augmented in KO-APX plants 1 h following light stress application compared to WT plants. 18S rRNA was used to control for RNA loading. Methods and experimental protocols are described in Experimental procedures. ELIP, early light inducible protein; HSP, heat shock protein; LL, low light (100 µmol m−2 sec−1); SL, strong light (425 µmol m−2 sec−1).

To extend our analysis of gene expression during light stress in knockout-Apx1 plants, we performed an Affymetrix chip analysis comparing wild-type and knockout-Apx1 plants subjected to a 1 h light stress (8200 gene chip; Table 5, 1 h light stress column in Table 3, Figure 4). Our analysis revealed some interesting differences between the response of wild-type and knockout-Apx1 plants to this stress. Thus, the induction of at least 16 transcripts encoding different heat shock proteins and two transcripts encoding putative heat shock transcription factors (HSFs) by light stress was much higher in knockout-Apx1 plants compared to that in wild-type plants (indicated in bold in Table 5; see also Figure 4).

Table 5.  Transcripts differentially induced in knockout-Apx1 (KO-APX) plants 1 h following light stress compared to wild-type (WT) plants subjected to the same treatment
 Fold log2Gene accession numberDescription
  1. Results are presented as fold induction (log2) over WT expression. Two accession numbers are given to each transcript, Affymetrix (left) and Genebank (right). The known or putative function of each transcript is also given (right column). Transcripts associated with the heat shock response are indicated in bold. Transcripts associated with H2O2 removal are indicated by an (*). RNA preparation and analysis by Affymetrix chips (Arabidopsis 8200 gene chip) are described in Experimental procedures and as Supplementary Material.

Signal transduction transcripts elevated in KO-APX plants following a 1 h light stress
 17553_at2.2AF084570KBP12 interacting protein (FIP37)
 18178_s_at1.5U95973Phosphatidylinositol-4-phosphate 5-kinase
 17445_at1.1F9D16.40Phosphatase like protein
 18847_at1F16F14.7Putative purple acid phosphatase precursor
Transcription factors transcripts elevated in KO-APX plants following a 1 h light stress
 12431_at1.5T19L18.4Putative heat shock transcription factor
 20342_at1.3F4F15.20Putative heat shock transcription factor
 20659_at1.2U90439Putative CCCH-type zinc finger protein
 14479_at1.1TPA14.90EF-Hand containing protein-like
 13533_at1F8K7.13Similar to SWI/SNF complex regulator
Defense and detoxification transcripts elevated in KO-APX plants following a 1 h light stress
 13278_f_at4.7Y14070Heat shock protein 17.6 A
 12434_at4.2F9C22.6Cyclophilin-like protein
 13275_f_at3.9X1729317.4 kDa heat shock protein
 13282_s_at3.3U72958HSP23.6- mitochondrial
 13279_at3.2X63443HSP17.6-II
 15954_at2.6U72155Beta-glucosidase (psr3.2)
 20323_at2.2F16P2.12Putative small heat shock protein
 13285_at2.1M62984Heat shock protein 83
 13284_at2.1AJ002551Heat shock protein 70
 15404_at1.8F15I1.13HSP20/alpha crystallin family
 15172_s_at1.6D84414Luminal binding protein (BiP)
 16916_s_at1.5X77199Heat shock cognate 70–2
 17815_at1.5Z97342Disease resistance RPP5 like
 15985_at1.4X98808(*) peroxidase ATP3a
 16466_s_at1.4Y08903HSC70-G7
 13287_at1.3Z70314Heat shock protein
 17942_s_at1.3X98322(*) peroxidase, prxr10
 13558_s_at1.2T1J8.9Putative ABC transporter
 13641_at1.2F17M5.60Putative NBS/LRR disease resistance
 18497_at1.2F22D22.1370 kDa heat shock protein
 13274_at1.1+ U13949Heat shock protein AtHSP101
 13552_at1.1F18O19.28Putative endochitinase
 16905_s_at1.1M4E13.140(*) catalase
 13283_at1Z11547Mitochondrial chaperonin HSP60
Cellular organization, metabolism and biogenesis transcripts elevated in KO-APX plants following a 1 h light stress
 13291_at3.4U18413IAA11 gene (auxin induced)
 15609_s_at3.2D34630Acetyl-CoA carboxylase
 14390_at2.8F8K7.5Similar to POS5 from Saccharomyces cerevisiae
 18698_s_at2.2X17528Mitochondrial citrate synthetase.
 13656_at2T7B11.13Unknown, similar to bacterial tolB proteins
 16489_at1.9X67421extA extensin gene
 12846_s_at1.5X52631DNA for rRNA intergenic region
 15169_s_at1.5AF061286Gamma-adaptin 1
 14057_at1.5F10M23.310Putative aconitase
 17627_at1.5F23M19.8Similar to FAB1 protein from S. cerevisiae
 17674_at1.5T32G6.2Putative U4/U6 small nuclear ribonucleoprotein
 16541_s_at1.3AB023423AST91 mRNA for sulfate transporter
 15793_at1.3T19F6.22FtsH protease
 15192_at1.1D17582Putative sugar transport protein, ERD1
 17581_g_at1.1AF105064GIGANTEA (GI)
 15015_at1.1T14P1.16Putat. mitoch. translation elongation factor G
 17473_at1.1Z18242Calnexin homolog.
 20354_s_at1.1T12M4.19Putative SF2/ASF splicing modulator
 12975_at1.1F24H14.14Putative spliceosome-associated protein
 18760_at1.1F21M12.1Putative leucyl-tRNA synthetase
 18823_s_at1.1F17M5.20WD-repeat-like protein
 18826_at1.1F22D22.29Putative glucan synthase
 13205_at1U37587Cell division cycle protein (CDC48)
 15689_at1U40341Carbamoyl phosphate synthetase large chain
 17091_s_at1U40269Origin recognition complex largest subunit
 17570_g_at1AF066080Sihydrolipoamide S-acetyltransferase
 13126_g_at1T27E13.15Ubiquitin activating enzyme 1 (UBA1)
 17857_at1T3P4.5Cytoplasmic aconitate hydratase
 12606_at1AF075598Putative fibrillin

Analysis of gene expression performed on wild-type and knockout-Apx1 plants at 0, 1, and 48 h of light stress (425 µmol m−2 sec−1), using DNA chips, revealed that the induction of the two putative heat shock transcription factors and the different heat shock proteins was transient and did not continue after plants acclimated to the light stress treatment (Figure 5). This result was in agreement with the RNA blots for HSP70 shown in Figure 4.

Figure 5.

Analysis of gene expression during light stress in knockout-Apx1 (KO-APX) plants.

Wild-type (WT) and KO-APX plants were subjected to light stress (425 µmol m−2 sec−1) for 0, 1, and 48 h, and changes in transcript levels were assayed by DNA arrays (chips). (a) and (b) show the changes in expression of the two putative heat shock transcription factors (HSFs) and heat shock proteins (HSPs), and (c) shows the changes in expression of catalase (CAT), two peroxidases, per10 (peroxidase prxr10) and perATP3 (peroxidase ATP3a), involved in H2O2 removal, and cyclophilin. The fold change between the induction of these transcripts in WT and KO-APX plants is also shown in Table 5. Protocols for RNA isolation and DNA chip analysis are described in Experimental procedures and as Supplementary Material.

Light stress in knockout-Apx1 plants resulted in the induction of transcripts encoding two typical plant peroxidases and a catalase (indicated by an asterisk (*) in Table 5), suggesting that these might be involved in the removal of H2O2 during light stress in knockout-Apx1 plants. In contrast to the transient induction of the heat shock-associated transcripts at 1 h light stress (Figures 4 and 5), the induction of these transcripts was elevated in knockout-Apx1 plants at early and late time points, suggesting that the removal of H2O2 by peroxidases and catalase was critical in these plants during all stages of plant acclimation to light stress (Figure 5). Analysis of the expression pattern of cyclophilin indicated that this transcript had a very different expression profile between wild-type and knockout-Apx1 plants during light stress (Figure 5c).

Heat shock of knockout-Apx1 plants

To further test the involvement of APX1 in the induction of heat shock proteins in Arabidopsis, we subjected wild-type and knockout-Apx1 plants to a heat shock treatment (37°C, 5 h; 100 µmol m−2 sec−1). Apx1 contains a functional heat shock factor-binding element in its promoter (Storozhenko et al., 1998) and is induced by heat shock (Mittler and Zilinskas, 1992; Rizhsky et al., 2002b; see also Figure 6a). However, in contrast to the differences observed in the induction of heat shock proteins between wild-type and knockout-Apx1 plants during light stress (Table 5; Figures 4 and 5), we could not find a similar difference in the induction of heat shock proteins between wild-type and knockout-Apx1 plants during heat shock.

Figure 6.

The induction of heat shock proteins (HSPs) during heat shock is not affected in knockout-Apx1 (KO-APX) plants.

(a) RNA gel blots performed with RNA obtained from control and heat shock (37°C)-treated WT and KO-APX plants showing that the induction of heat shock proteins in knockout-Apx1 plants during heat shock is not altered.

(b) A control RNA gel blot performed with RNA obtained from WT and KO-APX plants subjected to high light stress (2000 µmol m−2 sec−1) showing the induction of APX2 under these conditions in WT and KO-APX plants. Methods and experimental protocols are described in Experimental procedures. KO, KO-APX; sHSP, small HSP; VSL, very strong light (2000 µmol m−2 sec−1). The probe used for the detection of HSF was a non-specific HSF probe.

Under our experimental conditions, APX2 was not induced during light stress (425 µmol m−2 sec−1; Figure 4; also tested by RT-PCR and DNA chips; not shown) or heat shock (Figure 6a; also tested by RT-PCR; not shown). This finding was in contrast to a previous report on the induction of APX2 during heat shock (Panchuk et al., 2002). As shown in Figure 6(b), APX2 was however induced during high light stress, i.e. 2000 µmol m−2 sec−1 (see also Karpinski et al., 1999), suggesting that this isoform of APX is very specialized.

Discussion

Compensation for Apx1 deficiency in Arabidopsis

Studying plants that lack H2O2 scavenging genes may result in the identification of alternative enzymes or pathways that compensate for the loss of H2O2 removal activity (Mittler, 2002). Tobacco plants with suppressed APX1 expression contained elevated levels of transcripts encoding cytosolic CuZnSOD, catalase, and glutathione reductase to compensate for the loss of APX1 (Rizhsky et al., 2002a). Interestingly, the response of Arabidopsis plants to Apx1 deficiency was different. Under normal growth conditions, we could not detect an induction of catalase, CuZnSOD, or glutathione reductase (Table 2). Furthermore, we did not detect an increase in the expression of the cytosolic APX isozymes, APX2 (Figures 4 and 5) and APX3 (Table 2), or the chloroplastic APX isozymes (Table 2). In contrast, we found that the expression of chloroplastic CuZnSOD was suppressed in knockout-Apx1 plants (Table 2). This response may result from the relatively high sensitivity of this isozyme to H2O2 (Scioli and Zilinskas, 1988), or may represent an attempt by knockout-Apx1 plants to reduce the level of H2O2 produced in chloroplasts. If the latter is correct, a new question may arise, i.e. how is superoxide formation suppressed in the chloroplasts of knockout-Apx1 plants?

The reason(s) underlining the differences between the response of tobacco and Arabidopsis to Apx1 deficiency is unknown. It is possible that in contrast to Arabidopsis, tobacco with its tetraploid genome is able to compensate for Apx1 deficiency in a more efficient manner. Tobacco was found to have a high degree of plasticity in response to APX1, catalase, or APX1 + catalase deficiency, and was able to compensate for the loss of APX1 in a manner that prevented the accumulation of H2O2 in cells (Rizhsky et al., 2002a). In contrast, Arabidopsis, that did not induce alternative H2O2 scavenging enzymes, was unable to prevent the accumulation of H2O2 (Figure 3) and had a delayed growth and flowering phenotype (Figure 1; Table 1).

An essential component of the response of Arabidopsis to the lack of APX1, possibly resulting from the accumulation of H2O2, appeared to be the enhanced expression of transcripts encoding the iron-binding protein ferritin, and a copper-binding protein (blue-copper-binding protein; Table 3). These may be critical for sequestering of free iron and copper ions and preventing the formation of hydroxyl radicals. Although the expression of catalase and other H2O2-scavenging enzymes was not elevated in knockout-Apx1 plants under low-light conditions (100 µmol m−2 sec−1; Table 2), the treatment of knockout-Apx1 plants with high light (425 µmol m−2 sec−1) resulted in the induction of catalase and at least two different peroxidases (prxr10 and ATP3a; Table 5). Typical plant peroxidases were not considered to play an important role in H2O2 scavenging in plants (Asada and Takahashi, 1987). The finding that at least two typical plant peroxidases are induced in knockout-Apx1 plants during light stress may, however, change this concept, especially because some typical peroxidases can use ascorbic acid as their reducing substrate (Asada and Takahashi, 1987; Mittler and Zilinskas, 1992). Further studies are required to examine the possibility that prxr10 and ATP3a are involved in H2O2 removal in plants.

Signal transduction transcripts elevated in knockout-Apx1 plants

The majority of signal transduction transcripts induced in knockout-Apx1 plants (Tables 3 and 5) were not reported to be involved in the H2O2 signal transduction pathway (Czernic et al., 1999; Desikan et al., 2001; Kovtun et al., 2000; Samuel et al., 2000; Vranova et al., 2002). As was previously suggested (Bowler and Fluhr, 2000; Knight and Knight, 2001), calcium appears to play a central role in H2O2 signaling in plants. Thus, we found that the expression of transcripts encoding at least four different calcium-binding proteins, two calmodulin-like proteins, CaBP22 and calmodulin 9, was elevated in knockout-Apx1 plants. Interestingly, the expression of transcripts encoding a cyclic nucleotide-gated channel (CNGC4), possibly involved in calcium signaling and stomatal responses, was upregulated in knockout-Apx1 plants (Table 3). Because knockout-Apx1 plants were impaired in their stomatal responses (Figure 2b), it is possible that the abnormal expression of this cyclic nucleotide-gated channel is linked to these alterations in knockout-Apx1 plants. Hydrogen peroxide was suggested to play a central role in ABA-mediated stomatal closure by directly activating a calcium channel in guard cells (Pei et al., 2000). Using whole plants deficient in APX1 (Figure 1) and containing higher than normal levels of H2O2 (Figure 3b), we found that H2O2 may also be involved in the opening of guard cells during a shift of plants from dark to light (Figure 2b). Alternatively, the enhanced levels of H2O2 in knockout-Apx1 plants may have prevented the opening of stomata because they induced stomatal closure as proposed by Pei et al. (2000). We are currently studying how Apx1 deficiency affects ion currents in intact guard cells of knockout-Apx1 plants.

A number of putative receptor-like kinases were upregulated in knockout-Apx1 plants (Table 3). We did not, however, detect an enhanced expression of a two-component histidine kinase, previously suggested to play a role in H2O2 sensing (Desikan et al., 2001). It is also not clear whether the receptor-like kinases shown in Table 3 are involved in H2O2 sensing. However, because their expression is elevated in plants containing higher than normal levels of H2O2 (Figure 3b), such a role is possible. The expression of at least two serine/threonine kinases and a protein phosphatase 2C (PP2C) was enhanced in knockout-Apx1 plants, suggesting that different protein phosphorylation reactions are involved in H2O2 sensing. Protein phosphatase 2C was previously linked to different abiotic stress conditions including drought, cold, and salt stresses (Rodriguez, 1998). The induction of this transcript in knockout-Apx1 plants in the absence of any external stresses may suggest that protein phosphatase 2C induction during different environmental stresses is mediated at least in part by H2O2. Interestingly, we did not detect an enhanced expression of MAPK3 or MAPK6, previously linked to H2O2 sensing in plants (Kovtun et al., 2000). Instead, we detected an increase in the expression level of MAPK4 (Table 1).

Many of the transcription factors elevated in knockout-Apx1 plants are also induced during biotic or abiotic stresses (Table 3). Based on our findings and at least one additional report (Desikan et al., 2001), the ZAT zinc finger family of transcription factors may be linked to H2O2 responses. Other transcription factors involved in H2O2 sensing are WRKY, heat shock transcription factors, and ethylene response factors (Table 3; Mittler, 2002). Because the APX1 promoter contains at least one functional heat shock transcription factor (HSF) binding site (Storozhenko et al., 1998), the induction of HSF4 and HSF21 in knockout-Apx1 (Table 3) plants may suggest that these factors are involved in H2O2 signaling in plants.

A link between the heat shock response and H2O2 accumulation during light stress in Arabidopsis

We identified a link between H2O2 and the induction of heat shock proteins during light stress in Arabidopsis (Figures 4 and 5; Table 5). It was previously reported that in Arabidopsis, high light stress results in the induction of the cytosolic APX isozymes, APX1 and APX2 (Mullineaux and Karpinski, 2002; see also Figure 4 for APX1). However, the significance of this induction was not entirely clear because H2O2 is predominantly produced in the chloroplast and peroxisomes during high light stress. Because H2O2 can be transported through aquaporins (Henzier and Steudle, 2000), and because H2O2 was shown to leak from isolated chloroplasts treated with high light (Asada et al., 1974), it was postulated that the induction of APX1 and APX2 protects the cytosol and other cellular compartments during high light stress (Mittler, 2002; Mullineaux and Karpinski, 2002). Here, we show that in the absence of APX1, light stress in Arabidopsis results in the augmented induction of heat shock proteins (Figures 4 and 5; Table 5), suggesting that H2O2 produced during light stress in Arabidopsis diffuses into the cytosol and activates a signal transduction pathway that enhances the expression of heat shock proteins in the different cellular compartments. In the absence of APX1 this induction is much stronger because H2O2 that leaks into the cytosol is not scavenged. This model, shown in Figure 7, suggests that at least part of the induction of heat shock proteins during light stress in Arabidopsis is mediated by H2O2 that is scavenged by APX1. In contrast, the induction of heat shock proteins during heat shock may be mediated by a different pathway that does not involve APX1 (Figure 6).

Figure 7.

A model showing the involvement of APX1 in the induction of heat shock proteins (HSPs) during light stress in Arabidopsis.

Light (hv) is shown to enhance the production of H2O2 in the chloroplast and peroxisomes (red arrows). H2O2 then diffuses into the cytosol where it is scavenged by APX1. H2O2 not scavenged by APX1 activates a signal transduction pathway (dashed arrows) that results in the induction of chloroplastic, mitochondrial, and cytosolic heat shock proteins, HSPs (green arrows). In the absence of APX1 (knockout-Apx1 plants), this response is augmented (Figures 4 and 5; Table 5).

Growth suppression as a result of H2O2 accumulation

The cause of inhibition of plant growth and flowering time in Apx1-deficient Arabidopsis (Figure 1; Table 1) is unknown. It is possible that the enhanced levels of H2O2 in these plants (Figure 3) affected the expression of transcription factors involved in the regulation of plant growth and flowering. Alternatively, it is possible (however, unlikely; Asada and Takahashi, 1987) that APX1, similar to certain typical peroxidases, is involved in the biodegradation of auxin. Because the growth suppression of knockout-Apx1 plants was dependent upon day length and light intensity (not shown), it is reasonable to assume that this effect was directly linked to H2O2 accumulation in cells and not to an effect of APX1 activity on the level of auxin. Furthermore, a similar developmental effect was not observed in tobacco plants expressing an antisense construct to APX1. These plants induced alternative H2O2 scavenging pathways and did not contain elevated levels of H2O2 (Rizhsky et al., 2002a). Although our chip analysis revealed that the expression of GIGANTEA, a gene involved in the determination of flowering time in Arabidopsis (Fowler et al., 1999), was elevated during light stress in knockout-Apx1 plants (Table 5), it is not clear whether this gene is directly involved in the suppression of flowering time in knockout-Apx1 plants. More research is needed to establish a link between the accumulation of H2O2 in cells during normal conditions or stress, and the inhibition of growth and flowering time of plants. The knockout-Apx1 plants described in this report may provide an entry point into these studies because they may mimic the effect of different environmental stresses on plant growth and development through a known substrate, i.e. H2O2.

Experimental procedures

Plant material and growth conditions

Arabidopsis thaliana (cv. WS) plants were grown in growth chambers (Percival E-30HB and Conviron E7-2) under controlled conditions: 21–22°C, 18 h or constant light cycle, 100 µmol m−2 sec−1, and a relative humidity of 70%. Knockout Arabidopsis plants (cv. WS) containing a T-DNA insert in Apx1 were obtained from the Arabidopsis knockout facility at the University of Wisconsin-Madison according to the knockout facility recommended protocols (http://www.biotech.wisc.edu/Arabidopsis/) using the following DNA primers: JL-202 5′-CATTTTATAATAACGCTGCGGACATCTAC-3′ and APXI 5′-TTTTCCCATCTATATACCACCAACCCTAA-3′. The selected knockout-Apx1 plants plants were out-crossed and selfed to check for segregation and to obtain a pure homozygote line as recommended by Sussman et al. (2000). Confirmations of APX1 deficiency and segregation analysis were performed by PCR, genomic DNA blots, and RNA and protein blots.

Stress treatments

Light stress was performed by increasing the light intensity from 100 to 425 µmol m−2 sec−1. Controlled plants were kept at 100 µmol m−2 sec−1. All other growth parameters were maintained constant. At different times (0, 1, 4, 24, and 48 h), plants were sampled for RNA analysis. Heat shock was performed by changing the temperature from 22 to 37°C. Controlled plants were kept at 22°C. All other growth parameters were maintained constant. At different times (0, 0.5, 1, and 5 h), plants were sampled for RNA analysis. All experiments were performed in parallel on wild-type and knockout-Apx1 plants (each in triplicates).

Molecular, physiological, and biochemical analysis

RNA and protein were isolated and analyzed by RNA and protein blots as previously described (Pnueli et al., 2002). A ribosomal 18S rRNA probe or ethidium bromide staining was used to control for RNA loading. Coomassie Blue R-250 staining of protein gels was used to control for protein loading. Photosynthesis, stomatal conductance, and dark respiration were measured with a Li-Cor LI-6400 apparatus as described by Rizhsky et al. (2002b) using the Arabidopsis leaf chamber (Li-Cor, Lincoln, NE, USA). These measurements were performed on plants kept in the dark for 30 min and shifted to light (100 or 200 µmol m−2 sec−1) for 15 min. To induce stomatal closure by ABA, a solution of 50 µm ABA was sprayed on plants in light (100 µmol m−2 sec−1). Following a 1 min incubation, leaves were clamped and the decrease in stomatal conductance was measured with LI-6400 as described above. Reduced glutathione was determined by HPLC as described in Xiang and Oliver (1998), H2O2 was assayed as described by Rizhsky et al. (2002a), and APX activity was measured according to Mittler and Zilinskas (1992).

DNA chip analysis

In three independent experiments, RNA was isolated from 40 to 50 wild-type or knockout-Apx1 plants (a total of 120–150 plants per line), grown under controlled conditions as described above, or subjected to light stress. This RNA was pooled and used to perform the chip analysis. Each of the different pools of wild-type or knockout-Apx1 plants, grown under non-stressful conditions, was assayed by three different chips. Each of the different pools of wild-type or knockout-Apx1 plants, subjected to 0, 1, and 48 h light stress, was assayed by one chip. Affymetrix chip analysis (Arabidopsis 8200 gene chip; Affymetrix, Santa Clara, CA, USA) was performed at the University of Iowa DNA facility (http://dna-9.int-med.uiowa.edu/microarrays.htm). Conditions for RNA isolation, labeling, hybridization, and data analysis are described as Supplementary Material (in accordance with MIAME recommendations). A comparative analysis of samples was performed with the GeneChip mining tool V 5.0 and the Silicon Genetics GeneSpring V 5.1. Some of the comparison results were confirmed by RNA blots.

Acknowledgements

We thank Drs Eve Syrkin-Wurtele, Carol Foster, and Hailong Zhang for their help with Affymetrix data analysis. We also thank Drs David Oliver and Chengbin Xiang for help with glutathione determination. This work was supported by the Israeli Academy of Science, The Biotechnology Council Iowa State University, The Fund for the Promotion of Research at the Technion, and funding from The Plant Sciences Institute.

Supplementary Material

Files showing the expression level of transcripts with unknown function, elevated or suppressed in knock-out Apx1 plants, as well as files with Affymetrix chip data, are available to download from the following website: http://www.blackwellpublishing.com/products/journals/suppmat/TPJ/TPJ1715/TPJ1715sm.htm

Ancillary

Advertisement