Introduction
Seed germination and post-germinative growth are subject to transcriptional and post-translational control (Nonogaki et al., Reference Nonogaki, Liu, Hewitt and Martin2007). For instance, genes associated with the biosynthesis and deactivation of plant hormones important for seed germination control, such as gibberellin (GA) and abscisic acid (ABA), are mainly regulated at the level of transcription (Yamaguchi et al., Reference Yamaguchi, Kamiya, Nambara, Bradford and Nonogaki2007; Seo et al., Reference Seo, Nambara, Choi and Yamaguchi2009). In contrast, GA and ABA signal transduction is mainly mediated by changes at the protein level. Components important for hormone signal transduction are subject to proteolysis or modification by the 26S proteasome pathway (Steber, Reference Steber, Bradford and Nonogaki2007; Ariizumi et al., Reference Ariizumi, Murase, Sun and Steber2008). While transcriptional and post-translational control of seed germination and seedling establishment is well understood, the mechanisms concerning post-transcriptional regulation of genes associated with seed germination and seedling growth have not been fully investigated. Destabilization of mRNA, especially by microRNA (miRNA), has not been well characterized for seed germination and post-germination growth. miRNAs are expressed during seed imbibition and seedling growth (Martin et al., Reference Martin, Liu and Nonogaki2005, Reference Martin, Liu and Nonogaki2006). Although a few reports have demonstrated physiological roles for miRNAs and miRNA target genes during seed germination and post-germination (Liu et al., Reference Liu, Montgomery, Fahlgren, Kasschau, Nonogaki and Carrington2007), our understanding of miRNA function at these stages is still limited. In this study, we characterized the function of SQUAMOSA PROMOTER-BINDING PROTEIN LIKE13 (SPL13) [see also the accompanying paper (Martin et al., Reference Martin, Asahina, Liu, Kristof, Coppersmith, Pluskota, Bassel, Goloviznina, Nguyen, Martínez-Andújar, Kumar, Pupel and Nonogaki2010) for general functions of SPLs in seeds and plants], which is targeted by miR156/miR157 (the sequences of miR156 and miR157 are nearly identical and the term ‘miR156’ will be used hereafter). The physiological roles of the SPL13 and its regulation by miRNA at post-germinative stages are discussed.
Materials and methods
Generation of miR156-resistant SPL13 mutants
The SPL13 gene (At5g50570) including the 1.3 kb upstream regulatory region was amplified from genomic DNA of Arabidopsis thaliana ecotype Col-0 using SPL13 forward (SPL13 F: 5′-ACCTACTCCTGCCAACACAATGTTCTTACA-3′) and reverse (SPL13 R: 5′-ATCCTACAAGATGGCTCATCTCAACAAGGT-3′) primers. The intact gene was used to generate non-mutated SPL13 transgenic plants. Mutations at the miR156 target site were generated by site overlap extension polymerase chain reaction (PCR) mutagenesis (Ho et al., Reference Ho, Hunt, Horton, Pullen and Pease1989) using SPL13 mutant forward (SPL13mutF: 5′-CTGATTGTGCTCTCTCACTACTATCTTCCT-3′) and reverse (SPL13mutR: 5′-AGGAAGATAGTAGTGAGAGAGCACAATCAG-3′) primers. PCR products were cloned into pCAMBIA1301. Arabidopsis thaliana ecotype Col-0 plants were transformed by floral dip (Clough and Bent, Reference Clough and Bent1998) using Agrobacterium tumefaciens carrying pCAMBIA1301 with intact SPL13 (SPL13) or mutant SPL13 (mSPL13) constructs. Wild-type and transgenic plants were grown at 22°C under 12-h light/12-h dark conditions until rosette stages, and then flowers were induced by transferring plants to 16-h light/8-h dark conditions.
mRNA and small RNA extraction
High molecular weight (HMW) RNA for mRNA expression analysis was extracted using a standard phenol–SDS extraction protocol. Briefly, 100 Arabidopsis seedlings were homogenized in 2 ml RNA extraction buffer [45.5% (v/v) phenol, 9% (v/v) chloroform, 0.45% (w/v) SDS, 41 mM LiCl, 2 mM EDTA, 5.9 mM β-mercaptoethanol, 82 mM Tris–HCl, pH 8.2] with a mortar and pestle. The extract was centrifuged at 10,000 g for 2 min. The supernatant was extracted with one volume of phenol–chloroform–isoamyl alcohol [25:24:1 (v/v/v)] solution and then with one volume of chloroform. LiCl was added to the supernatant (2 M final concentration) and the sample was mixed thoroughly and kept at − 20°C overnight. The sample was thawed, mixed and centrifuged at 10,000 g for 5 min. The pellet was washed with 1 ml 80% (v/v) ethanol, dried, dissolved in water and used for mRNA expression analysis. To obtain low molecular weight (LMW) RNA for microRNA detection, the supernatant from the 2 M LiCl precipitation step in the total RNA isolation was fractionated by isopropanol (Martin et al., Reference Martin, Liu and Nonogaki2005). The pellet from the 35–50% (v/v) isopropanol fraction was washed with 1 ml 80% (v/v) ethanol, dried, dissolved in water and used for miRNA expression analysis.
mRNA gel blot
An equal amount (2 μg) of HMW RNA was separated on a 1.3% (w/v) agarose gel containing 7% (v/v) formaldehyde, transferred to a positively charged Hybond-N+nylon membrane (GE Healthcare Bio-Sciences Corp./Amersham, Piscataway, New Jersey, USA), and UV-cross-linked. To make the RNA probe, the SPL13 cDNA was cloned into the pCR-TOPO4.0 vector (Invitrogen, Carlsbad, California, USA) and transcribed using a digoxigenin (DIG)-labelled nucleoside triphosphate (NTP) mixture (Roche Applied Science, Indianapolis, Indiana, USA) and T7 RNA polymerase (Applied Biosystems/Ambion, Austin, Texas, USA). After a 15-min prehybridization at 60°C with a hybridization buffer containing 50% (v/v) deionized formamide, 4% (w/v) blocking reagent (Roche Applied Science), 0.2% (w/v) SDS, and 5 × SSC (sodium chloride/sodium citrate), hybridization was carried out in the same buffer containing approximately 100 ng ml− 1 RNA probe for 16–18 h at 60°C. Membranes were washed once with 2 × SSC, 0.1% (w/v) SDS at 75°C for 25 min and then twice with 0.2 × SSC, 0.1% (w/v) SDS for 25 min at 75°C. Membranes were blocked for 30 min with 5% (w/v) non-fat milk in 0.1 M maleic acid buffer, pH 7.5, containing 0.15 M NaCl, and 0.3% (v/v) Tween 20 (buffer A) and were then incubated with alkaline phosphatase conjugated anti-DIG antibody for 1 h at 25°C. After washing with buffer A, the membranes were subjected to chemiluminescence detection. The signal was detected on X-ray film after exposure.
Promoter analysis
The 5′ upstream region of SPL13 ( − 1126 to+185) was amplified using forward (SPL13-Pro-F: 5′-TCACCGAGTATAGCCCAACAAG-3′) and reverse (SPL13-Pro-R: 5′-AAGTCGACACACCTACCAGTAGAGTC-3′) primers and cloned into the pCR-Blunt II-TOPO vector (Invitrogen). The region ( − 958/+185) was cloned into the SacI and SalI sites of a shuttle vector (Grebenok et al., Reference Grebenok, Pierson, Lambert, Gong, Afonso, Haldeman-Cahill, Carrington and Galbraith1997) which contained the uidA (GUS: β-glucuronidase) gene. The promoter–GUS construct (4.4 kb) in pRJG23 was excised with SacI and SpeI and subcloned into the SacI and XbaI sites in the pGPTV-KAN binary vector (Becker et al., Reference Becker, Kemper, Schell and Masterson1992). GUS staining of seedlings was done as previously described (Weigel and Glazebrook, Reference Weigel and Glazebrook2002) using 100 mM sodium phosphate buffer (pH 7.0) containing 0.1% (v/v) Triton X-100 and 2 mM X-Gluc (RPI Co., Mount Prospect, Ilinois, USA).
Results
Over-accumulation of miR156-resistant mutant form of SPL13 (mSPL13)
The miR156 target site (5′-GUGCUCUCUCUCUUCUGUCA-3′) in SPL13 mRNA contains one mismatched base and is located at a position corresponding to the amino acid sequence CALSLLS (Fig. 1A). SPL13 was amplified from the Arabidopsis genome and four silent mutations were introduced at the miRNA target site (Fig. 1A). Constructs with the native promoter and the mutated SPL13 (mSPL13) or intact gene (SPL13) were used to transform Arabidopsis. mSPL13 plants over-accumulated SPL13 mRNA, wild-type plants had low levels of SPL13 and control transgenic plants with intact SPL13 had intermediate levels of SPL13 mRNA. An additional RNA band, which is most likely the product of RNA cleavage, was detected in control plants but was not detected in mSPL13 (Fig. 1B). These results validated the deregulation of mSPL13 mRNA from miR156.
Over-accumulation of miRNA-resistant mSPL13. (A) The miR156 complementary sequence in SPL13. Silent mutations (no change in amino acid sequence CALSLLS) introduced into the miRNA-resistant form of SPL13 (mSPL13) are in red. (B) RNA gel blot of SPL13. RNA was extracted 3 DAI from seedlings of wild-type (WT), transgenic SPL13 and mSPL13 plants. Ribosomal RNA (rRNA) bands are shown as loading control. Arrowhead and asterisk indicate the positions of intact and cleaved SPL13 mRNA bands, respectively.
Post-germination phenotypes in the mSPL13 mutants
During phenotypic analysis, apparent differences were detected between control and mSPL13 mutant plants following seed germination, while seed germination per se (radicle emergence) was not affected by the mutation. We found that many mSPL13 seedlings (over 75%) had delayed transition from the two- to four-leaf stage (Fig. 2A). We stained normal and delayed seedlings from segregating populations of putative heterozygous mutant lines for transgene marker (GUS) (the transformation vector contained 35S:GUS to identity transformants). The delayed seedlings with two vegetative leaves were GUS positive while the normal seedlings with four vegetative leaves were negative (Fig. 2B), indicating that the delay in seedling development was caused by the transgene, which caused over-accumulation of mSPL13. Three independent homozygous mSPL13 transgenic lines were further characterized for this phenotype. All seedlings in the homozygous lines exhibited delayed seedling development during post-germination stages. Wild-type seedlings and control transgenic seedlings expressing non-mutated SPL13 reached the four-leaf stage 10 days after imbibition (DAI), while the mSPL13 seedlings were still at the two-leaf stage after the same period (Fig. 2C).
Delayed vegetative leaf emergence in mSPL13 mutants at post-germinative stages. (A) A normal 10-DAI seedling with four vegetative leaves (left) and a delayed seedling with only two vegetative leaves found in putative heterozygous mSPL13 lines. (B) GUS (β-glucuronidase, transgene marker) staining of the two- (upper row) and four-leaf seedlings (bottom row) from the heterozygous mSPL13 population. (C) Comparison of 10-DAI seedlings of wild type (WT), control transgenic (SPL13) and homozygous mSPL13 mutants (mSPL13).
Further analysis indicated that the difference between control and mSPL13 mutant seedlings was initiated earlier, at the cotyledon seedling stages. The first two vegetative leaves emerged from the shoot apex of wild-type and SPL13 seedlings 5–6 DAI (Fig. 3A). On days 4 and 5, leaf primordia were visible in the wild-type and SPL13 shoot apical meristem (SAM) under a dissection microscope, while only a trace primordium was observed in mSPL13 seedlings (Fig. 3B). Thus, the over-accumulation of mSPL13 suppressed the development of vegetative leaf primordia 3 DAI during the post-germinative stages.
Delayed leaf primordium development in mSPL13 seedlings. (A) Seedling development in wild-type (WT), control transgenic (SPL13) and mutant (mSPL13) at early stages (3–6 days). Note that there is a delay in vegetative leaf emergence in mSPL13 seedlings. (B) Close-up view of the shoot apex in WT, SPL13 and mSPL13 seedlings.
Localization of SPL13 expression
The accumulation of SPL13 in different plant developmental stages was analysed using the Arabidopsis eFP Browser (http://bar.utoronto.ca/efp/cgi-bin/efpWeb.cgi). SPL13 seems to accumulate in a broad range of tissues and organs during different developmental stages (data not shown). However, the accumulation of mRNA does not necessarily represent the expression of miRNA target genes, because miRNAs eliminate the target mRNAs by cleavage (except in the case of miRNA targets regulated by translational repression). In fact, SPL13 mRNA does not accumulate at very high levels in seedlings, based on the analysis using the Arabidopsis eFP Browser (data not shown), suggesting that SPL13 is normally suppressed by miR156 in seedlings. To see whether SPL13 is actually expressed in cotyledon-stage seedlings, transgenic plants expressing GUS under the control of the SPL13 promoter were generated and the expression of the reporter gene was analysed. Promoter analysis indicated that SPL13 was expressed mainly in the hypocotyl of the seedling 3 DAI (Fig. 4). Weak signals were also detected in the vascular tissues at the basal part of cotyledons. These results indicate that SPL13 is expressed in the hypocotyl of the cotyledon-stage seedlings, but is normally suppressed by miRNA156. The over-accumulation of miR156-resistant mSPL13 in this tissue affected the emergence of vegetative leaves.
SPL13 promoter:GUS expression in 3-DAI seedlings. The 5′ upstream region of SPL13 ( − 958 to+185) was used for the promoter analysis.
Discussion
The SQUAMOSA PROMOTER-BINDING PROTEIN (SBP) family is one of the plant-specific transcription factor families. SBP genes were first identified and characterized in Antirrhinum majus (Klein et al., Reference Klein, Saedler and Huijser1996) and were found to be widespread in green plants from unicellular algae (Kropat et al., Reference Kropat, Tottey, Birkenbihl, Depege, Huijser and Merchant2005) to angiosperms (Cardon et al., Reference Cardon, Hohmann, Klein, Nettesheim, Saedler and Huijser1999; Xie et al., Reference Xie, Wu and Xiong2006). The SBP proteins regulate SQUAMOSA (SQUA) (Huijser et al., Reference Huijser, Klein, Lonnig, Meijer, Saedler and Sommer1992), a gene associated with floral induction in A. majus, by binding to a 32 bp ( − 295/ − 264) region in the 5′ upstream regulatory sequence of this gene (Cardon et al., Reference Cardon, Hohmann, Nettesheim, Saedler and Huijser1997). SQUA is an orthologue of APETALA1 (AP1), a floral homeotic gene in Arabidopsis thaliana (Huijser et al., Reference Huijser, Klein, Lonnig, Meijer, Saedler and Sommer1992; Mandel et al., Reference Mandel, Gustafson-Brown, Savidge and Yanofsky1992). Seventeen SBP-LIKE genes (SPLs) are found in the Arabidopsis genome (Cardon et al., Reference Cardon, Hohmann, Klein, Nettesheim, Saedler and Huijser1999). SPL3, SBP1 and SBP2 proteins interact in vitro with the AP1 promoter element which has similarity to the SBP-binding motif in SQUA. The AP1 and SQUA motifs contain 19-mer core sequences including an identical 10-bp conserved stretch (GTCCGTACAA) (Cardon et al., Reference Cardon, Hohmann, Nettesheim, Saedler and Huijser1997). Over-expression of SPL3 causes early flowering (Cardon et al., Reference Cardon, Hohmann, Nettesheim, Saedler and Huijser1997) which is similar to the phenotype observed in the AP1 over-expressors (Mandel and Yanofsky, Reference Mandel and Yanofsky1995). This suggests that Arabidopsis SPL3 modulates the timing of flowering through the regulation of AP1.
While SPLs were originally identified in relation to flowering mechanisms, they appear to play multiple roles in plant growth and development. Transposon mutagenized spl8 plants exhibit reduced fertility primarily due to premeiotic abortion of the sporocyte, leading to abnormal development of microsporangia (Unte et al., Reference Unte, Sorensen, Pesaresi, Gandikota, Leister, Saedler and Huijser2003). The SPL8 gain-of-function mutants have nondehiscent anthers (Zhang et al., Reference Zhang, Schwarz, Saedler and Huijser2006). These results suggest the involvement of SPL8 in reproductive development. In addition, constitutive expression of SPL8 resulted in dwarf plants exhibiting dark-green rosettes and compact inflorescences, phenotypes typical of gibberellin (GA)-deficient or GA-insensitive mutants. Seed germination, which is also regulated by GA, was reduced in seeds over-expressing SPL8 (Zhang et al., Reference Zhang, Schwarz, Saedler and Huijser2006), therefore SPL8 may play a role in GA biosynthesis or signal transduction.
In the present study, deregulation of SPL13 from miR156 did not cause phenotypes in seed germination per se. However, transgenic plants expressing the miR156-resistant mSPL13 exhibited apparent phenotypes at post-germinative stages (Figs 2 and 3). As indicated by the promoter analysis, SPL13 is actually expressed in the hypocotyl and in a part of the cotyledons (Fig. 4), but its mRNA does not seem to accumulate in these tissues, based on the output from the eFP browser analysis. These results support the contention that SPL13 is negatively regulated by miR156 at the mRNA level. Consistently, silent mutations in the SPL13 sequence complementary to the miR156 sequence caused over-accumulation of SPL13 mRNA, which caused a delay in seedling development. The function of SPL13 is most likely to suppress the development of leaf primordia. This study demonstrates that post-transcriptional regulation of gene expression plays an important role at post-germinative stages in Arabidopsis. A possible mechanism to explain the consequence of SPL13 over-accumulation is described in the accompanying paper (Martin et al., Reference Martin, Asahina, Liu, Kristof, Coppersmith, Pluskota, Bassel, Goloviznina, Nguyen, Martínez-Andújar, Kumar, Pupel and Nonogaki2010).
Acknowledgements
We are grateful to Tony Chen, Department of Horticulture, Oregon State University, for providing us with the transformation vector. This work was supported by NSF grant IBN-0237562 (to H.N.) and American Seed Research Foundation grant (to H.N. and R.C.M.).
