SEARCH

SEARCH BY CITATION

Keywords:

  • Oryza sativa;
  • SIZ/PIAS-type SUMO E3-ligase;
  • stress response;
  • SUMO;
  • sumoylation

ABSTRACT

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

Sumoylation is a post-translational regulatory process in diverse cellular processes in eukaryotes, involving conjugation/deconjugation of small ubiquitin-like modifier (SUMO) proteins to other proteins thus modifying their function. The PIAS [protein inhibitor of activated signal transducers and activators of transcription (STAT)] and SAP (scaffold attachment factor A/B/acinus/PIAS)/MIZ (SIZ) proteins exhibit SUMO E3-ligase activity that facilitates the conjugation of SUMO proteins to target substrates. Here, we report the isolation and molecular characterization of Oryza sativa SIZ1 (OsSIZ1) and SIZ2 (OsSIZ2), rice homologs of Arabidopsis SIZ1. The rice SIZ proteins are localized to the nucleus and showed sumoylation activities in a tobacco system. Our analysis showed increased amounts of SUMO conjugates associated with environmental stresses such as high and low temperature, NaCl and abscisic acid (ABA) in rice plants. The expression of OsSIZ1 and OsSIZ2 in siz1-2 Arabidopsis plants partially complemented the morphological mutant phenotype and enhanced levels of SUMO conjugates under heat shock conditions. In addition, ABA-hypersensitivity of siz1-2 seed germination was partially suppressed by OsSIZ1 and OsSIZ2. The results suggest that rice SIZ1 and SIZ2 are able to functionally complement Arabidopsis SIZ1 in the SUMO conjugation pathway. Their effects on the Arabidopsis mutant suggest a function for these genes related to stress responses and stress adaptation.


INTRODUCTION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

Many eukaryotic proteins are regulated by post-translational modifications, such as the reversible covalent attachment of ubiquitin (Ub) and ubiquitin-like (Ubl) proteins (Novatchkova et al. 2004; Kerscher, Kerscher & Hoshstrasser 2006). These post-translational modifications critically alter biological activities, subcellular localization and stability of their target proteins (Dohmen 2004). The small ubiquitin-related modifier (SUMO) is highly similar to ubiquitin with respect to its three-dimensional structure. However, SUMO has only about 20% similarity to Ub in its primary sequence and includes approximately 15 additional N-terminal amino acid residues (Hay 2005).

SUMO conjugation (Sumoylation) to target proteins involves the sequential actions of activation (E1), conjugation (E2) and ligation (E3) enzymes as in ubiquitination. As a contrasting feature different from ubiquitination, sumoylation is guided by short consensus sequences, Ψ-K-X-E/D (Ψ, large hydrophobic amino acid; K, acceptor lysine; X, any amino acid; E/D, glutamate or aspartate) (Schmidt & Müller 2003). Ubiquitination is generally associated with proteasomal degradation of target proteins, whereas sumoylation is linked to roles in transcription regulation, signal transduction in the cell cycle, DNA repair, nuclear import and sub-nuclear compartmentalization (Johnson 2004; Hay 2005). The process is best understood in yeast and animal cells.

In the Arabidopsis genome, all components required for sumoylation have been characterized biochemically and genetically (Novatchkova et al. 2004; Colby et al. 2006). These components are essential for plant development and adaptation during abiotic and biotic stresses (Saracco et al. 2007; Miura et al. 2009). Similarly, the sumoylation machinery is also encoded in the rice genome, although studies on the molecular, biochemical and genetic levels are presently lacking (Miura, Jin & Hasegawa 2007a; Suzuki et al. 2009).

SUMO E3-ligases confer substrate specificity transferring SUMO from E2 conjugating enzymes to substrates. They are essential components in the sumoylation pathway in vivo (Johnson 2004). Therefore, it is reasonable to assume that changes in SUMO E3-ligase activity would have a significant impact on processes that are regulated by sumoylation. Several SUMO E3-ligases, including scaffold attachment factor A/B/acinus/PIAS/MIZ (SIZ)/protein inhibitor of activated signal transducer and activator of transcription (PIAS), RanBP2, Pc2 and HDAC4, have been characterized in yeasts and mammals. However, to date, only two SUMO E3-ligases, SIZ1 and HPY2, have been identified in Arabidopsis, and these proteins contain the SP-RING structural domain known to confer SUMO E3-ligase activity. The Arabidopsis, SIZ1 is a prototype mammalian protein inhibitor of activated signal transducer and activator of transcription (PIAS), and is similar to the yeast Siz [scaffold attachment factors A/B/acinus/PIAS (SAP) and MIZ] family of SUMO E3-ligases (Miura et al. 2005). In contrast, HPY2 constitutes a MMS21-type enzyme, an imperfect PIAS protein, which contains the SP-RING domain alone (Huang et al. 2009; Ishida et al. 2009). Based on genetic and biochemical analyses, stress-responsive SUMO conjugation in Arabidopsis is mediated mainly by the SIZ1 SUMO E3-ligase. Stresses in which sumoylation has been shown to be involved include phosphate starvation, abscisic acid (ABA) treatment, salicylic acid (SA)-dependent pathogen defence, flowering time regulation, basal thermotolerance, and drought and cold tolerance. However, only four specific target proteins of sumoylation have been identified (Yoo et al. 2006; Catala et al. 2007; Lee et al. 2007; Saracco et al. 2007; Miura et al. 2005, 2007b, 2009; Jin et al. 2008).

Although SIZ1 has been well characterized in Arabidopsis, the involvement of SUMO E3-ligases in sumoylation in other plants is poorly understood. Here, we provide evidence that sumoylation regulates responses to various stresses in rice. To explore sumoylation reactions in rice, we have functionally characterized the putative SUMO E3-ligases, OsSIZ1 and OsSIZ2. We show that OsSIZ1/-2 have SUMO E3 ligase activity and are localized to the nucleus. Transgenic plants expressing OsSIZ1 or OsSIZ2 under the control of the CaMV35S promoter in a SIZ1-knockout Arabidopsis line (atsiz1-2) complemented functions of SIZ1. In addition, SUMO conjugates were enhanced in the transgenic Arabidopsis by heat stress. Our results suggest that SUMO modification by OsSIZ1 or OsSIZ2 are involved in protection, fostering adaptation to environmental stresses in rice.

MATERIALS AND METHODS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

Plant materials and growth conditions

Arabidopsis thaliana Columbia-0 (Col-0) was used. Seeds were surface sterilized, stratified for 3 d at 4 °C and then sown on to a medium in Petri plates containing 1 × Murashige and Skoog basal salt mixture, 3% sucrose, 2.5 mm MES, pH 5.7 and 0.8% agar or soil. Plants were grown under a 16/8 h light/dark at 22 °C, light intensity of 100 µmol m−2 s−1 and relative humidity of 57–80%. Rice (Oryza sativa L. cv. Dongjin) was used for protoplast isolation and suspension cell cultures. Rice suspension culture cells were incubated at 30 °C in continuous light with shaking at 130 rpm.

RNA isolation and RT-PCR

Total RNAs were isolated from Arabidopsis or rice seedlings using TRIZOL reagents (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's protocol and were treated with DNase I (deoxyribonuclease I, Sigma) at room temperature for 15 min and at 70 °C for 10 min. For RT-PCR analyses, the first-strand cDNA was synthesized with ReverTra Ace reverse transcriptase (100 unit µL−1, TOYOBO) in 25 µL of reaction mixture containing 2 µg total RNA and an oligo (dT)20 primer, according to the manufacturer's instructions. RT-PCR analyses were performed as previously described (Lee & An 2009) with EF-Taq DNA polymerase (2.5 unit µL−1, Solgent, Daejeon, Korea) with primers outlined in Supporting Information Table S1. The PCR conditions were as follows: preheating at 95 °C for 2 min, and then 22 or 25 cycles of denaturation at 95 °C for 30 s, annealing at 57 °C for 30 s and extension at 72 °C for 30 s, followed by a final extension for 10 min. Aliquots of individual PCR products were separated on an agarose gel and visualized with ethidium bromide under UV light. Arabidopsis tubulin or rice actin genes were used as internal positive controls.

Plasmid constructs and plant transformation

To generate Arabidopsis transgenic plants over-expressing OsSIZ1 or OsSIZ2 in siz1-2 (SALK_065379) plants, the protein coding regions were amplified by PCR with primer sets outlined in Supporting Information Table S1, using cDNA clone AK105290 for OsSIZ1 (2628 bp) and AK065127 for OsSIZ2 (2442 bp) from RIKEN institute (http://www.riken.jp/engn/index.html). The amplified fragments were cloned into the pGEM-T Easy vector (Promega, Madison, WI, USA), sequenced to verify the correct DNA sequence and then sub-cloned into the pCAMBIA1300-multi vector using the appropriate restriction enzymes. Agrobacterium tumefaciens GV3101 strains harboured each construct over-expressing OsSIZ1 or OsSIZ2 were grown in LB liquid culture with 50 mg/L gentamycin, 50 mg L−1 rifampicin, and 50 mg L−1 kanamycin at 30 °C. The siz 1-2 plants were transformed by the floral deep method as previously described (Clough & Bent 1998; Park et al. 2009). Hygromycin-resistant transgenic plants were selected on 1 × MS medium containing 30 mg L−1 hygromycin.

Analysis of cellular localization

To determine the cellular localization of the OsSIZ1 and OsSIZ2 proteins, respectively, the green fluorescent protein (GFP) sequence was fused to the C-termini of OsSIZ1 and OsSIZ2 using appropriate restriction enzyme sites. Rice protoplasts (4 × 106 mL−1) were prepared as described (Kyozuka & Shimamoto 1991). Protoplasts were mixed with 10 µg of DNA of each construct, and transfected using polyethylene glycol (PEG)-mediated transient transformation. After 48 to 72 h of incubation, GFP and red fluorescent protein (RFP) fluorescence were visualized using a fluorescence microscope (Olympus AX-70, DP-70, Tokyo, Japan).

Stress treatments and in vivo analysis of SUMO conjugates

Wild-type rice seeds were initially surface sterilized with three separate 5 min washes with 96% ethyl alcohol. The seeds were subsequently treated with 3% sodium hypochlorite for 2 h and washed with 10 volume of sterilized water five times. After thorough washing in water, seeds were sprouted in plastic boxes (25 × 18 × 9 cm3) containing ½x MS medium and covered with the same boxes. The sample boxes were incubated in a growth chamber under a 16/8 h light/dark at 25 °C. For stress treatments, we used 10-day-old seedlings grown on ½x MS medium. Abiotic stress conditions were applied as follows: 45 °C for heat stress, 4 °C for cold stress and 250 mm NaCl. In addition, 100 µm ABA was used as a treatment.

For the analysis of SUMO conjugates, total proteins were extracted from 10-day-old seedlings grown on medium under the conditions described previously. Plant tissues (0.2 g) were extracted with a mortar and pestle in grinding buffer [100 mm Na-MOPS, pH 7.5, 10 mm NaCl, 1 mm ethylenediaminetetraacetic acid (EDTA), pH 8.0, 10% sucrose, 5% β-mercaptoethanol and 4% sodium dodecyl sulphate (SDS)] at room temperature. Protein concentration was measured using a Bio-Rad Protein Assay Kit (No. 500-0006, Bio-Rad, Hercules, CA, USA), and the proteins were separated by sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS–PAGE), transferred to a polyvinylidene difluoride membrane (No. 162-0177, Bio-Rad), probed with an anti-SUMO1 antibody (ab5316, Abcam Ltd, Cambridge, UK), and detected using the ECL Western blot detection system (Amersham Biosciences, Buckinghamshire, UK).

Agrobacterium-mediated transient SUMOylation assay in tobacco plants

The combinations of A. tumefaciens (strain GV3101) bacteria containing indicated constructs (35S:flag-AtSUMO1, 35S:AtSIZ1, 35S:OsSIZ1 and 35S:OsSIZ2) were infiltrated into leaves of tobacco (N. benthamiana) plants. At 3 DAI (day after inoculation), infiltrated leaves were collected and proteins were prepared in protein extraction buffer [50 mm Tris-HCl (pH 7.5), 150 mm NaCl, 0.5% NP40, 1 mm EDTA, 3 mm DTT, 1 mm PMSF, 5 µg mL−1 of leupeptin, 1 µg mL−1 of aprotinin, 1 µg mL−1 pepstain, 5 µg mL−1 of antipain, 5 µg mL−1 of chymostatin, 2 mm Na2VO4, 2 mM NaF, 50 µm MG132]. Approximately 50 µg proteins were subjected to immunoblot analysis using an anti flag-antibody that recognizes flag-AtSUMO1 conjugates in tobacco leaves.

ABA germination assays

For measuring ABA sensitivity, Arabidopsis seeds were germinated on ½x MS medium containing various concentration (0, 0.2, 5, 0.75, 1 µm) of ABA. To investigate germination rate by ABA, 7-day-old seedlings with emerged green cotyledon were counted and quantified as percentage (±SE). Ten-day-old seedlings were photographed. All data were analysed for significant differences by student's t-test.

RESULTS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

Primary structure and OsSIZ1 and OsSIZ2 expression patterns

AtSIZ1 represents a prototype SUMO E3-ligase of the PIAS type that is involved in regulating plant responses to hormones, in pathogen defence and in the control of flowering (Catala et al. 2007; Miura et al. 2007a; 2009). To begin to investigate biological roles of SUMO E3-ligases in rice, we searched for cDNA homologs of AtSIZ1 using the Rice Genome Annotation Project BLAST Search (http://rice.plantbiology.msu.edu/blast.shtml) database. This led to the identification of two genes and cDNA sequences of putatively expressed genes, OsSIZ1 (LOC_Os05g03430) and OsSIZ2 (LOC_Os03g50980). The two SIZ gene candidates in rice encode proteins of 875 and 813 amino acids in length, with calculated molecular masses of 97 and 90 kDa, respectively (Suzuki et al. 2009; Fig. 1b). The OsSIZ deduced amino acid sequences contained four predicted domains that are conserved in the AtSIZ1 gene. In addition, a Pro-Ile-Asn-Ile-Thr (PINIT) core motif similar to this sequence in Arabidopsis could be identified (Miura et al. 2007a; Suzuki et al. 2009). The five predicted domain structures include a SAP domain that is implicated in DNA binding, a plant homeodomain-finger (PHD) domain that is only present in plant Siz/PIAS proteins, a PINIT domain that is essential for nuclear retention, a SIZ/PIAS-RING (SP-RING) domain that is required for SUMO E3-ligase activity, and an SXS motif that promotes SUMO binding (Miura et al. 2007a; Fig. 1a). The comparison of OsSIZ1 and OsSIZ2 with the AtSIZ1 amino acid sequence showed 51% and 43% sequence identity, respectively (Fig. 1b).

image

Figure 1. Structural features and amino acid sequences of rice SIZ1 and SIZ2. (a) Structural domains of SIZs. The domains include: an N-terminal SAP (I, scaffold attachment factor A/B/acinus/PIAS); the PHD (II, plant homeodomain); the putative PINIT core domain (III); the SP-RING (IV, SIZ/PIAS-RING) domain; and the SXS (V, Ser-X-Ser) domain. (b) Amino acid sequences of SIZ proteins from Arabidopsis thaliana (At) and O. sativa (Os). Sequence identities (black boxes) and similarities (gray boxes) of amino acids were identified by Clustal W.

Download figure to PowerPoint

To determine whether OsSIZ1 and OsSIZ2 show tissue-specific expression, we analyzed their expression levels in various rice tissues. As shown in Fig. 2, transcripts of OsSIZ1 and OsSIZ2 were detected in all tissues of rice, but to a lesser extent in root. Next, we investigated whether the expression of OsSIZ genes could be induced in response to abiotic, biotic, and hormone stimuli. However, none of these treatments induced transcription of the OsSIZ genes (Supporting Information Fig. S1).

image

Figure 2. Expression of OsSIZ1 and OsSIZ2 by RT-PCR analysis. Total RNA was extracted from roots, the shoot apical meristem (SAM), reproductive stage nodes, leaf sheath, leaf blade, and flower. Rice actin (OsActin2) was served as an internal control.

Download figure to PowerPoint

OsSIZ1 and OsSIZ2 are localized to the nucleus

Most sumoylation processes occur in the nucleus in animals and yeasts, where they affect diverse cellular processes (Matunis, Coutavas & Blobel 1996; Johnson & Blobel 1999; Hardeland et al. 2002; Geiss-Friedlander & Melchior 2007; Martin et al. 2007). It had previously been shown that Arabidopsis SIZ1 localizes to the nucleus and is compartmentalized in nuclear speckles, as it is known for PIAS proteins (Miura et al. 2005; Cheong et al. 2009). To identify the subcellular localization of OsSIZ1 and OsSIZ2 in rice, we used GFP and RFP reporter gene fusion constructs in rice protoplasts. The GFP sequence was fused to the C-termini of the OsSIZ1 and OsSIZ2 cDNAs under the control of the CaMV35S promoter (Fig. 3a). Each chimeric OsSIZ1/-2::GFP construct was introduced into rice protoplasts along with an nuclear localization signal (NLS)::RFP construct that represented an authentic nuclear localization marker (Heikal et al. 2000). To further test and monitor potential monocot/dicot differences in localization masked by the heterologous system used, we expressed the GFP fusion constructs in tobacco epidermis cells (Supporting Information Fig. S2) and in Arabidopsis protoplasts (data not shown). Taken together, the subcellular distributions of each OsSIZ1/-2::GFP signal clearly overlapped with those of the NLS::RFP red fluorescence signals, indicating that OsSIZ1/2 are localized to the nucleus (Fig. 3b).

image

Figure 3. Subcellular localization of OsSIZ1 and OsSIZ2. (a) Schematic representation of the constructs for transient expression in rice protoplasts. CaMV35S, cauliflower mosaic virus 35S promoter; sGFP, green fluorescent protein; RFP, red fluorescent protein; NLS, nuclear localization signal; Nos T, nopaline synthase gene terminator are indicated. (b) Co-expression of GFP fused SIZs and NLS::RFP. Two constructs, OsSIZ1::GFP and NLS::RFP, or OsSIZ2::GFP and NLS::RFP were transformed into rice protoplasts, respectively. Signals were observed using a fluorescent microscope. The yellow colour (Merge) reveals co-localization in the nucleus. Scale bars indicate 20 µm.

Download figure to PowerPoint

Stress-induced accumulation of SUMO conjugates in rice

A polyclonal AtSUMO1 antibody cross-reacted with AtSUMO2 protein, which shows 87% amino acid sequence identity with AtSUMO1 (Abcam5316; Kurepa et al. 2003). Considering the amino acid sequence conservation of OsSUMO1 and OsSUMO2 relative to AtSUMO1 with 89% and 84% identity, respectively, it appeared possible that the AtSUMO1 antibody would cross-react with the rice proteins. When anti-AtSUMO1 antibody was used to detect free SUMO in rice and Arabidopsis crude extracts, the anti-AtSUMO1 antibody recognized an abundant 14 kDa protein in both rice and Arabidopsis (Supporting Information Fig. S3a). These proteins likely represent free OsSUMO1 and OsSUMO2 in rice. In addition, anti-AtSUMO1 antibodies recognized recombinant AtSUMO1 and OsSUMO1 proteins expressed in Escherichia coli and tobacco leaves (Supporting Information Fig. S3b & c). These results showed that the anti-AtSUMO1 antibody cross-reacted with OsSUMO1 in rice, suggesting that anti-AtSUMO1 antibody appear to be specific to OsSUMO1.

To study possible changes in sumoylation profiles during abiotic stress treatments, total proteins were extracted from seedlings treated with high temperature (45 °C), cold (4 °C), salt (NaCl) and ABA. The immunoblot analysis using the AtSUMO1 antibody showed formations of increased conjugates after stress treatment (Fig. 4a). A similar sumoylation pattern after heat shock treatment has been reported in Arabidopsis (Kurepa et al. 2003). We subsequently used expression markers for various stress treatments in rice (Fig. 4b; Gutha & Reddy 2008; Zou et al. 2009). SUMO conjugates were strongly induced in rice seedlings treated with heat or ABA, but induced only weakly by cold, or NaCl treatments. However, when rice seedlings were treated with a series of increasing concentrations of H2O2 for one hour, SUMO conjugates changed only marginally (data not shown). This behaviour is unlike in Arabidopsis, possibly suggesting a different time-dependent response pattern or the presence of different SUMO proteins for sumoylation in rice.

image

Figure 4. Identification of SUMO conjugates induced by various stresses in rice. (a) Induced SUMO conjugates by various stresses. Each lane was loaded with 20 µg of total protein extracted from ten-day-old rice seedlings that had been exposed to heat (45 °C), cold (4 °C), 250 mm NaCl, and 100 µm ABA. Rice seedlings were collected at various time intervals after the start of the stress treatment (t = 0). The resulting crude extracts were subjected to SDS-PAGE, and immunoblot analysis was performed with an antibody against AtSUMO-1, which detects both OsSUMO1 and OsSUMO2. (b) Expression profiles after stress treatments of inducible marker genes in rice. Rice actin (OsActin2) was included as a control for constant expression in the assays.

Download figure to PowerPoint

OsSIZ1 and OsSIZ2 function as SUMO E3-ligases

In order to assess the ability of OsSIZ1 and OsSIZ2 to act as SUMO E3 ligase, constructs for OsSIZ1 or OsSIZ2 in combination with flag-tagged AtSUMO1 were co-infiltrated into tobacco (N. benthamiana) leaves for Agrobacterium-mediated transient transformation and expression (Voinnet et al. 2003). The rice SIZs were expressed under control of the 35S CaMV promoter. AtSIZ1, the SUMO E3 ligase in Arabidopsis, with flag-AtSUMO1 also was co-infiltrated as a control to compare sumoylation activity. Total protein extracts were prepared from tobacco leave, separated by SDS-PAGE, and subjected to immunoblot analysis with anti-flag antibody recognizing flag-AtSUMO1 protein. As shown in Fig. 5, when AtSUMO1, AtSIZ1, OsSIZ1 or OsSIZ2 alone were expressed, the extent of sumoylation was substantially lower in the high molecular weight range. However, co-expression of AtSIZ1, OsSIZ1 or OsSIZ2 together with AtSUMO1 dramatically increased sumoylation in the high molecular weight range (Fig. 5), suggesting that both OsSIZ1 and OsSIZ2 could act as a SUMO E3 ligase. Interestingly, the sumoylation activity of OsSIZ1 proved to be stronger than that of OsSIZ2 or AtSIZ1. The result indicates that OsSIZ1 may act as the major SUMO E3 ligase in rice.

image

Figure 5. Accumulation of AtSUMO1-protein conjugates induced by OsSIZ1 and OsSIZ2. Tobacco (N. benthamiana) leaves were co-infiltrated with AtSIZ1, OsSIZ1 or OsSIZ2 and with (+) and without (−) flag tagged AtSUMO1. At 3 d after inoculation, total proteins were extracted and immunoblot analysis was carried out with anti-flag antibody to detect AtSUMO1-protein conjugates.

Download figure to PowerPoint

To confirm that OsSIZ1/-2 functions as a SUMO E3-ligase in rice, we isolated rice T-DNA mutant allele of OsSIZ1 (Ossiz1). But, there is no available mutant allele known for Ossiz2. Although heat shock increases SUMO conjugates in wild-type Arabidopsis, the extent of heat shock-induced sumoylation is much less in Arabidopsis siz1-2 mutant seedlings (Kurepa et al. 2003; Yoo et al. 2006). As shown in Fig. S4, heat shock-induced accumulation of SUMO conjugates in Ossiz1 mutant is significantly lower than that in the wild-type rice, suggesting that OsSIZ1 functions as a SUMO E3-ligase in rice.

OsSIZ1 and OsSIZ2 partially complement siz1-2 mutants

To further understand the function of OsSIZ1 and OsSIZ2, we generated transgenic Arabidopsis siz1-2 plants harbouring OsSIZ1 and OsSIZ2 genes under the control of the CaMV35S promoter. When the OsSIZ proteins were expressed in siz1-2 plants, the morphological phenotype of siz1-2 was partially complemented by OsSIZ1 and OsSIZ2 in 5-week-old soil-grown plants (Fig. 6a & b). As shown in Fig. 6c, the siz1-2 phenotype of dwarfism with leaf curl and short leaf length was partially alleviated in OsSIZ1 (#97-3) and OsSIZ2 (#41-7) transgenic Arabidopsis plants, compared to siz1-2 and vector control plants. In addition, leaf length of the OsSIZ1/-2 transgenic Arabidopsis plants was increased, in a range from 15% to 27% larger, based on sizes observed in the largest leaves of each plant (Fig. 6d). It has been also reported that AtSIZ1 negatively regulated ABA signalling during seed germination and seedling growth (Miura et al. 2009). To determine whether expression of OsSIZ1 and OsSIZ2 functionally complement an ABA hypersensitive phenotype of siz1-2, we tested the response of OsSIZ1 (#97-3) and OsSIZ2 (#41-7) transgenic Arabidopsis to exogenous application of ABA. As shown in Fig. 7, for plants grown on media without ABA, germination and seedling growth of siz1-2 and OsSIZ1/-2 transgenic plants did not differ. However, on media supplemented with ABA, OsSIZ1/-2 seedlings were less sensitive to ABA than siz1-2 seedlings, although they were more sensitive than wild type (Fig. 7a & b). The observation is consistent with the partial complementation of morphological phenotypes shown by siz1-2 by OsSIZ1 and OsSIZ2. Taken together, the results imply that OsSIZ1/-2 proteins can at least partially complement functions of SIZ1 in Arabidopsis.

image

Figure 6. Complementation of OsSIZ1 and OsSIZ2 and their corresponding expression levels in siz1 mutant plants. (a) Phenotypes and expression of atsiz1-2 plants expressing OsSIZ1. (b) Phenotypes and expression of atsiz1-2 plants expressing OsSIZ2. Total RNA was extracted from five-week-old plants, and RT-PCR was performed using gene-specific primers. Tubulin was used as an internal control to normalize the amount of cDNA templates. (c) Leaf morphology of plants expressing OsSIZ1 or OsSIZ2. The leaves were taken and arranged in each row, from the left. Plants expressing OsSIZ1 and OsSIZ2 in atsiz1-2 mutant Arabidopsis were compared with plants transformed with Col-0, siz1-2, and the vector control. (d) Statistical analysis of leaf length for each of the lines illustrated in (c). The largest leaves of each line were used for leaf length measurement. The data are the means of three different experiments (n = 30) and indicate the percentage (±SE) of leaf length in each transgenic plant line. Significant differences from atsiz1-2 (asterisks) at P < 0.05 are indicated.

Download figure to PowerPoint

image

Figure 7. Suppression of ABA-hypersensitivity of siz1-2 by complementation of OsSIZ1 and OsSIZ2. Seeds of wild-type (col-0), siz1-2 (atsiz1-2), and each of the complemented lines (OsSIZ1 #97-3 and OsSIZ2 #41-7) were sown on MS medium without or supplemented with ABA. (a) Cotyledon expansion of 10-day-old seedling in the absence or presence of 0.75 µm ABA. (b) Germination rates of seedlings seven days after sowing. Seedlings with emerged green cotyledons were counted as germinated. The data are averages of three independent experiments. Indicated are the percentages (±SE) of seedlings with green cotyledon in each genotype. Significant differences from atsiz1-2 (asterisks) at P < 0.05 are indicated.

Download figure to PowerPoint

Heat shock-induced SUMO conjugation in OsSIZ1 or OsSIZ2 expressed siz1-2 plants

Previous studies had shown that heat shock-induced sumoylation in Arabidopsis and, to a lesser extent, in Arabidopsis siz1-2 seedlings (Kurepa et al. 2003; Yoo et al. 2006). In our study, SUMO conjugates were strongly induced in heat-treated rice seedlings (Fig. 4a). To determine whether OsSIZ1 and OsSIZ2 exhibit SUMO E3-ligase activities inducible by heat stress, we performed immunoblot analysis using anti-SUMO1 antibodies on control and heat-treated OsSIZ1/-2 transgenic Arabidopsis plants. As shown in Fig. 8, the SUMO conjugates increased in OsSIZ1/-2 seedlings when compared to vector and siz1-2 seedlings after heat treatment, although the extent of SUMO conjugation in OsSIZ1/-2 was lower than in col-0. Transcript amounts of both OsSIZs were not altered after heat shock treatment. Irrespectively, SUMO conjugation products were correlated with transcripts and accumulated in the high expression lines (#97-3 and #41-7) of both OsSIZs, while much less accumulation was observed in low expression lines (#1-11 and #22-3) after heat treatment (Fig. 8b). The result suggests control of activity of OsSIZ1/-2 not by changes of transcription after environmental stresses but by control at the protein level. Taken together, the results indicate that OsSIZ1 and OsSIZ2 encompass SUMO E3-ligase activity in planta.

image

Figure 8. Heat shock-induced SUMO conjugation in siz1-2 seedlings requires OsSIZ1 and OsSIZ2, respectively. Ten-day-old seedlings of Col-0 (wild-type), siz1-2 (atsiz1-2), vector, and each of the complemented lines (#1-11, #8-2, and #97-3 for OsSIZ1 and #22-3, #35-10, #41-7 for OsSIZ2) were exposed to a 30 min-heat shock at 39 °C in the dark. Total protein was extracted from untreated (a) or heat shock-treated (b) seedlings, and 20 µg of proteins was loaded for SDS-PAGE, and the immunoblot was probed with anti-AtSUMO1 antibody. The Arabidopsis tubulin gene (Tubulin) was included as a control to show expression of OsSIZ1 and OsSIZ2 in the transgenic plants.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

Sumoylation is associated with stress responses, transcriptional regulation, as well as genome maintenance functions in eukaryotes. However, while extensive studies have been conducted in animals and yeast cells, in plants only Arabidopsis has been studied in any detail. Investigation and characterization of the biological functions of sumoylation in crop species such as, for example, rice are lacking. We have now isolated two O. sativa SIZ genes, OsSIZ1 and OsSIZ2, based on their homology to Arabidopsis SIZ1. These proteins encode SIZ/PIAS-type SUMO E3-ligases that contain five predicted domains (SAP, PHD, PINIT, SP-RING, and SXS domains) conserved in SIZ/PIAS SUMO E3-ligases (Fig. 1a). Also, these putative rice SUMO E3-ligases include the PINIT core motif, required for nuclear retention, that is present in other SIZ/PIAS proteins (Fig. 1a & b; Duval et al. 2003). Given the evolutionarily deep anchoring of SIZ and the sumoylation pathway, it appeared possible that the rice proteins, like AtSIZ1, would respond to environmental stimuli, but the presence of a duplicated SIZ homolog required confirmation of functional assignments.

To examine the expression patterns of the OsSIZ1/-2 genes in rice, we performed RT-PCR analysis with RNA purified from various tissues (Fig. 2). The OsSIZ1/-2 genes were expressed in all tissues, to a lesser extent in roots. In addition, we investigated the expression levels of the OsSIZ1/-2 genes in response to abiotic, biotic and hormonal stimuli. There were no significant transcript inductions in response to these treatments (Supporting Information Fig. S1). However, SUMO conjugates were induced in response to abiotic stresses such as heat shock, cold, NaCl, and ABA (Fig. 4). Thus, it is likely that the activities of OsSIZ1/-2 might be controlled by post-translational modification, rather than the transcriptional level in response to stresses (Catala et al. 2007). Although OsSIZ1 and OsSIZ2 included all conserved domains present in the AtSIZ1 protein, it had to be established whether OsSIZ1 and OsSIZ2 could function as SUMO E3-ligases like AtSIZ1. Our sumoylation assay demonstrated not only that OsSIZ1 and OsSIZ2 indeed facilitated sumoylation in the tobacco transient expression system (Fig. 5), but also that especially, OsSIZ1 functions as a SUMO E3-ligase in rice (Supporting Information Fig. S4). The results identified OsSIZ1 and OsSIZ2 as functional SUMO E3-ligases in rice. In further support of such a function, nuclear localization of OsSIZ1/-2 was determined using rice protoplasts and tobacco epidermal cells (Fig. 3 & Supporting Information Fig. S2). This is significant because many transcription factors are direct targets of SUMO conjugation mediated by SIZ/PIAS proteins (Gill 2005). SUMO conjugation affects transcription factor function through activation, repression, or protein stabilization. In Arabidopsis, the transcription factors PHR1, ICE1, ABI5 and FLD have been identified as targets of SIZ1. They regulate gene expression changes under low-phosphate and low-temperature conditions, in ABA responses, and the control of flowering time (Miura et al. 2005, 2007b, 2009; Jin et al. 2008). Our results indicate that the localization of OsSIZ1 and OsSIZ2 in the rice nucleus through SIZ-dependent SUMO conjugation may be involved, as are the Arabidopsis homologs, in the regulation of transcription factors in response to diverse environmental conditions.

Rice, one of the world's major crops, may frequently become exposed to environmental stresses such as nutrient starvation, heat, cold, drought, submersion or pathogens. The dramatic increase of SUMO conjugates in response to high temperature, H2O2, or ethanol has been demonstrated in Arabidopsis (Kurepa et al. 2003). In addition, SIZ1 is one of the sumoylation components important in phosphate starvation responses, cold tolerance, basal thermotolerance, SA-dependent pathogen defence, and flowering time regulation by loss of function studies (Yoo et al. 2006; Lee et al. 2007; Miura et al. 2005, 2007b; Jin et al. 2008). Therefore, we performed an analysis of in vivo pattern of SUMO conjugates with anti-SUMO1 antibody following various stress treatments, including heat, cold, NaCl and ABA in rice seedlings (Fig. 3a). Although SUMO conjugates were showed little response after 1 h of H2O2 treatment at concentrations ranging up to 50 mm (data not shown), the conjugate amount and pattern structure strongly increased in rice seedlings after treatments with high temperature and/or ABA, while responding less strongly to cold or NaCl treatment. The results suggested that sumoylation in rice is a part of processes responsible for adaptations to adverse environmental conditions, and that it plays an important role in the reaction to environmental stresses.

SIZ1 is the only Arabidopsis protein that appears to contain all of the prototypical domains found in SIZ/PIAS-type proteins, including the SAP, PINIT, SP-RING and SXS domains (Kotaja et al. 2002). The siz1 knockout mutant plants show a dwarf phenotype with leaf curling and short petioles (Miura et al. 2005; Catala et al. 2007; Lee et al. 2007). Therefore, we generated OsSIZ1/-2 constructs, under the control of the CaMV35S promoter, and transformed the constructs into siz1-2 mutant plants to study the SUMO E3-ligase phenotypic functions of OsSIZ1/-2 in Arabidopsis (Figs 6 & 7). As the results show, the morphological phenotypes and ABA sensitivity of atsiz1-2 were partially complemented by both OsSIZ1 and OsSIZ2. Additional evidence supporting the SUMO E3-ligase function of OsSIZ1/-2 in vivo derives from the analysis of induced SUMO conjugates using an anti-SUMO1 antibody on untreated and heat-treated OsSIZ1/-2 transgenic Arabidopsis plants. Interestingly, atsiz1-2 transgenic plants over-expressing OsSIZ1 or OsSIZ2 under control of the CaMV 35S promoter showed increased levels of SUMO-conjugates, and the increases were correlated with the transcript levels observed in individual lines of OsSIZ1 and OsSIZ2 plants in response to heat-shock treatment (Fig. 8). In addition, the transcript levels of OsSce1, the SUMO E2 enzyme in rice, were also increased by heat-shock stress (Nigam et al. 2008).

To summarize, we have isolated and characterized biological functions of the OsSIZ1 and OsSIZ2 SUMO E3-ligases in rice. Our studies documented the partial complementation of the siz1-2 phenotype by OsSIZ1 and OsSIZ2 over-expression in Arabidopsis. Both genes are expressed at nearly identical levels in all tissues investigated and show SUMO E3-ligase activity that is modulated by several stress conditions. Conceivably, OsSIZ1/-2 could play an important role in the protection and adaptation of rice against environmental stresses. Our report demonstrated the biological function of both SIZ/PIAS-type SUMO E3-ligases in rice and indicated substantial evolutionary conservation of function crossing the dicot/monocot divide. In order to achieve a greater understanding of the functions of OsSIZ1/-2 in rice, we have begun to analyze OsSIZ1/-2 using T-DNA tagged mutant lines. Studies on these mutant lines will pinpoint the regulatory function of OsSIZ1 and OsSIZ2 in hormonal and environmental responses in rice.

ACKNOWLEDGMENTS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

This work was supported by grants from the Biogreen 21 Program (20080401034023) of the Rural Development Administration, the World Class University Program (R32-10148) and the Environmental Biotechnology National Core Research Center (Grant#: R15-2003-012-01002-00) funded by the Ministry of Education, Science and Technology in Korea.

REFERENCES

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

Supporting Information

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

Figure S1. Expression patterns of OsSIZ1 and OsSIZ2 genes in response to various stresses and hormone treatments. (a) Expression patterns of OsSIZ1 and OsSIZ2 genes in response to various abiotic stresses. Total RNA was isolated from 10-day-old seedlings treated with cold (12 h or 24 h at 4 °C), 0.1 mM ABA (4 h), 400 mM NaCl (4 h), or drought (no drain), respectively. RT-PCRs were performed with OsSIZ1 or OsSIZ2 gene specific primers using first strand cDNA as a template. RT-PCR of OsActin2 transcript was employed as a quantitative control. (b) Expression patterns of OsSIZ1 and OsSIZ2 genes in response to rice blast fungus infection (biotic stress). Rice suspension cells were inoculated with rice blast of either an incompatible (KJ401) or compatible (KJ101) race at 1 × 105 sopres mL−1, respectively. The samples were harvested and total RNA was extracted. (c) Expression patterns of OsSIZ1 and OsSIZ2 genes by various hormone stimuli. Total RNA was isolated from 10-day-old seedlings treated with 1% DMSO (dimethyl sulfoxide), 1 μ M NAA (naphthaleneacetic acid), 1 μ M BR (brassinosteroid), 1 μ M kinetin, 10 μ M GA (gibberellin), 10 μ M ACC (1-aminocyclopropane-1-carboxylic acid), 0.1 mM ABA (abscisic acid), 50 mM JA (jasmonic acid), and 50 mM SA (salicylic acid), respectively.

Figure S2. Subcellular localization of OsSIZ1 and OsSIZ2 in tobacco epidermal cells. (a) Schematic representation of the constructs used for transient expression. The SIZ cDNAs were transiently expressed in tobacco epidermis cells under control of the CaMV35S promoter. Symbols are: CaMV35S, cauliflower mosaic virus 35S promoter; smGFP, green fluorescent protein; RFP, red fluorescent protein; NLS, nuclear localization signal; TNos, nopaline synthase gene terminator; HPT, hygromycin phosphotransferase gene. (b) Subcellular localization of OsSIZ1 and OsSIZ2 proteins. Tobacco epidermis cells were infiltrated with Agrobacterium harbouring two constructs: OsSIZ1::GFP plus NLS::RFP, or OsSIZ2::GFP plus NLS::RFP, respectively. The transiently transformed tobacco epidermal cells were examined under a fluorescent microscope after transformation. The bright-field image (Bright) was obtained using the bright-field filter. Green and red images are GFP and RFP signals, respectively. Co-localized GFP and RFP (Merge) appear as yellow. At least three independent transformation experiments were performed with each fusion construct.

Figure S3. Cross reactivity of anti-SUMO1 antibody to AtSUMO1 and OsSUMO1. (a) Detection of free SUMO in rice and Arabidopsis with anti-SUMO1 antibody. Arabidopsis (col-0 and atsiz1-2) and rice (wild-type and Ossiz1) seedlings were grown for 2 weeks. Rice seedlings were subjected to heat shock at 45 °C for 1 h. The resulting crude extracts were subjected to SDS-PAGE and immunoblot analysis was performed with anti-SUMO1 antibody. (b) Recognition of recombinant AtSUMO1 and OsSUMO1 with anti-SUMO1 antibody. Recombinant GST or His fusion AtSUMO1 and OsSUMO1 proteins were expressed in E. coli and were separated by SDS–PAGE. Immunoblot analysis was carried out with anti-SUMO1 and anti-GST antibodies, respectively. (c) Recognition of AtSUMO1 and OsSUMO1 in tobacco transient expression system with anti-SUMO1 antibody. Tobacco leaves were infiltrated with flag tagged AtSUMO1 and OsSUMO1. At 3 d after inoculation, total proteins were extracted and subjected to SDS-PAGE and immunoblot analyses were carried out with anti-flag and anti-SUMO1 antibodies, respectively.

Figure S4. Heat shock-induced SUMO1/2 conjugation pattern in wild type and Ossiz1 rice plants. Two-week-old rice wild-type and Ossiz1 seedlings were exposed to a 1 h heat shock treatment (45 °C, dark). Total protein was extracted from untreated or heat shock-treated seedlings. Twenty μg of protein was loaded onto an SDS-PAGE, and the immunoblot was probed with anti-SUMO1antibody.

Table S1. Gene-specific primer sequences used to detect genes by RT-PCR.

FilenameFormatSizeDescription
PCE_2195_sm_fS1.pdf484KSupporting info item
PCE_2195_sm_fS2.pdf303KSupporting info item
PCE_2195_sm_fS3.pdf434KSupporting info item
PCE_2195_sm_fS4.pdf135KSupporting info item
PCE_2195_sm_tS1.doc31KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.