Plant and Cell Physiology Advance Access originally published online on February 9, 2007
Plant and Cell Physiology 2007 48(3):523-539; doi:10.1093/pcp/pcm022
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Overexpression of a Type-A Response Regulator Alters Rice Morphology and Cytokinin Metabolism
RIKEN Plant Science Center, 1-7-22, Suehiro, Tsurumi, Yokohama, 230-0045 Japan
*Corresponding author: E-mail, sakaki{at}riken.jp; Fax, +81-45-503-9609
 |
Abstract |
|---|
 Top Abstract IntroductionResults and DiscussionMaterials and MethodsAcknowledgmentsReferences |
|---|
Genome-wide analyses of rice (Oryza sativa L.) cytokinin (CK)-responsive genes using the Affymetrix GeneChip® rice genome array were conducted to define the spectrum of genes subject to regulation by CK in monocotyledonous plants. Application of trans-zeatin modulated the expression of a wide variety of genes including those involved in hormone signaling and metabolism, transcriptional regulation, macronutrient transport and protein synthesis. To understand further the function of CK in rice plants, we examined the effects of in planta manipulation of a putative CK signaling factor on morphology, CK metabolism and expression of CK-responsive genes. Overexpression of the CK-inducible type-A response regulator OsRR6 abolished shoot regeneration, suggesting that OsRR6 acts as a negative regulator of CK signaling. Transgenic lines overexpressing OsRR6 (OsRR6-ox) had dwarf phenotypes with poorly developed root systems and panicles. Increased content of trans-zeatin-type CKs in OsRR6-ox lines indicates that homeostatic control of CK levels is regulated by OsRR6 signaling. Expression of genes encoding CK oxidase/dehydrogenase decreased in OsRR6-ox plants, possibly accounting for elevated CK levels in transgenic lines. Expression of a number of stress response genes was also altered in OsRR6-ox plants.
Keywords: Cytokinin - Gene chip - Microarray - Oryza sativa - Response regulator
Abbreviations: ACC, 1-aminocyclopropane-1-carboxylate; AMT, ammonium transporter; AP2/EREBP, APETALA2/ethylene-responsive element binding protein; ARF, auxin response factor; AUX/IAA, auxin/indole-3-acetic acid; bHLH, basic helix–loop–helix; CaMV, cauliflower mosaic virus; CK, cytokinin; CKX, CK oxidase/dehydrogenase; DMSO, dimethylsulfoxide; GA, gibberellin; GFP, green fluorescent protein; HK, His protein kinase; HP, His-containing phosphotransfer protein; IPT, adenosine phosphate-isopentenyltransferase; PHT, phosphate transporter; PUP, purine permease; RR, response regulator; RT–PCR, reverse transcription–PCR; SULTR, sulfate transporter; tZ, trans-zeatin; tZR, tZ riboside; tZRMP, tZR 5'-monophosphate
|
Introduction |
|---|
|
TopAbstractIntroductionResults and DiscussionMaterials and MethodsAcknowledgmentsReferences |
|---|
Cytokinins (CKs) are a class of plant hormones that regulate many aspects of growth and development, including cell division, apical dominance, leaf senescence, nutrient signaling and shoot differentiation (Mok 1994
, Sakakibara 2006). In the past decade, substantial progress has been made in understanding CK biosynthesis, metabolism and signal transduction. Much of this knowledge is based on research in Arabidopsis, a dicotyledonous model plant. This has been achieved by identifying a series of genes encoding key enzymes, including CK-synthesizing adenosine phosphate-isopentenyltransferase (IPT), CK-degrading CK oxidase/dehydrogenase (CKX) and CK-inactivating CK glycosyltransferase (for reviews, see Mok and Mok 2001, Sakakibara 2006, Werner et al. 2006).
The current model of the CK signaling pathway is a multistep His–Asp phosphorelay system comprised of a sensor His protein kinase (HK), His-containing phosphotransfer protein (HP) and response regulator (RR; for recent reviews, see Aoyama and Oka 2003, Heyl and Schmülling 2003, Kakimoto 2003, Grefen and Harter 2004, Mizuno 2004, Ferreira and Kieber 2005, Hwang and Sakakibara 2006, and references therein). Early CK signaling events have been characterized primarily in Arabidopsis, in which three genes encoding HKs that function as CK receptors (AHK2, AHK3 and AHK4/CRE1/WOL; Inoue et al. 2001, Suzuki et al. 2001, Ueguchi et al. 2001, Yamada et al. 2001), five encoding HPs (AHP1–AHP5; Miyata et al. 1998, Suzuki et al. 1998) and an analog (AHP6; Mähönen et al. 2006), and 23 encoding RRs (ARR1–ARR22 and ARR24; Brandstatter and Kieber 1998, Imamura et al. 1998, Sakai et al. 1998, Imamura et al. 1999, D’Agostino et al. 2000, Mason et al. 2004, Mizuno 2004) have been identified. The ARRs have been generally classified into type-A or type-B based on structural features (Imamura et al. 1999, Mizuno 2004). Type-A ARRs contain a receiver domain with a short C-terminal extension, while type-B ARRs have a receiver domain and a long C-terminal extension that contains transcription-regulating domains serving as transcriptional regulators for certain target genes, including the type-A ARR genes (Sakai et al. 2000, Sakai et al. 2001, Mason et al. 2005, Yokoyama et al. 2007). Expression is rapidly induced in response to exogenous CK in the type-A ARR genes but not in the type-B genes (Brandstatter and Kieber 1998, Taniguchi et al. 1998, Imamura et al. 1999, Kiba et al. 1999, D’Agostino et al. 2000). Overexpression of type-A ARR genes (ARR4, ARR5, ARR6 and ARR7) in Arabidopsis protoplasts represses the ability of CK to induce ARR6 promoter activity, indicating that type-A ARRs are transcriptional repressors (Hwang and Sheen 2001). Overexpression of ARR8 represses shoot regeneration and greening of calli, suggesting that ARR8 plays a role as a negative regulator in CK signaling (Osakabe et al. 2002). Overexpression of ARR15 in Arabidopsis plants resulted in reduced sensitivity to CK (Kiba et al. 2003). Analyses of multiple insertional mutants of type-A ARR genes suggested that type-A ARRs perform partially redundant functions, such as negative regulation of CK signaling (To et al. 2004). Recent genome-wide expression profiling indicates that ARR7 is a negative regulator of early CK-regulated genes and acts as a transcriptional repressor (Lee et al. 2007). In contrast, overexpression of ARR4 in the cultured stems of transgenic plants promotes shoot regeneration in the presence of CKs, suggesting that ARR4 is a positive regulator in CK signaling (Osakabe et al. 2002). However, this positive effect remains uncertain, as a loss-of-function analysis did not support it (To et al. 2004).
Type-A RR genes have also been isolated and characterized from the monocots maize (Sakakibara et al. 1998, Sakakibara et al. 1999, Asakura et al. 2003, Giulini et al. 2004) and rice (Ito and Kurata. 2006, Jain et al. 2006). Treating detached maize tissues with CK induces the accumulation of transcripts of type-A RR genes (ZmRR1–ZmRR7; Sakakibara et al. 1998, Sakakibara et al. 1999, Asakura et al. 2003, Giulini et al. 2004). A single loss-of-function mutation in a type-A RR (ZmRR3/ABPH1) in maize alters leaf phyllotaxy, which is probably the result of enlargement of the shoot apical meristem (Giulini et al. 2004). Thirteen putative type-A RR genes in rice (OsRR) were recently reported to be expressed in an organ-specific manner, and also in response to light conditions (Ito and Kurata 2006, Jain et al. 2006). Application of CK to rice seedlings induced the expression of most of the OsRRs, and OsRR6 expression was also enhanced in response to salinity, dehydration and low temperature stress (Jain et al. 2006). Although the isolation of RR genes involved in CK signaling has been reported in the Gramineae, knowledge of the in planta role of type-A RRs in monocotyledonous species is very limited compared with Arabidopsis.
The search for genes whose expression is modified by CK has yielded a number of valuable tools that have been used to understand CK signaling and the complex developmental processes under control of this hormone. Genome-wide analyses have identified a variety of CK-responsive genes in Arabidopsis (Che et al. 2002, Hoth et al. 2003, Rashotte et al. 2003, Kiba et al. 2004, Brenner et al. 2005, Kiba et al. 2005, Lee et al. 2007). Completion of the rice genome sequence has made the comprehensive identification of CK-responsive genes in this model monocot plant possible. In this study, we identified rice genes regulated by CK using an Affymetrix rice genome array and examined transgenic rice plants that overexpress OsRR6 (referred to as OsRR6-ox). Our study showed that OsRR6-ox plants displayed altered morphologies and changes in CK metabolism, probably due to changes in the gene regulatory network.
|
Results and Discussion |
|---|
|
TopAbstractIntroductionResults and DiscussionMaterials and MethodsAcknowledgmentsReferences |
|---|
Identification of CK-regulated genes in roots and leaves
One aspect of understanding the action of a regulatory signal depends on knowing which genes are induced or repressed in response to it. To gain insight into the mode of action of CK in rice seedlings, excised whole roots and detached leaf blades were treated with 5 µM trans-zeatin (tZ) for 30 or 120 min. Total RNA from these organs was subjected to microarray analysis using an Affymetrix GeneChip® rice genome array, which contains probes to query approximately 48,564 japonica and 1,269 indica transcripts. Three independent replicate treatments were performed for each organ per time point. To increase the chance of detecting differentially expressed genes, transcripts with a ratio of change >2.0 and an Affymetrix flag call with a ‘present’ classification for the higher expression value were classified as regulated.
A total of 677 root and 641 leaf genes met the criteria for up-regulation, and 591 root and 962 leaf genes met the criteria for down-regulation at both the 30 and 120 min time points after CK treatment (Fig. 1A and Supplementary Table S1). The up-regulated genes included three type-A OsRR genes (Table 1). The up-regulation of OsRR4 and OsRR6 was confirmed by semi-quantitative reverse transcription–PCR (RT–PCR) using the same RNA samples used in the microarray analyses (Fig. 1B). Induction of these genes confirms a previous report (Jain et al. 2006) and validates the experimental system used in this study. Jain et al. (2006) showed that rice has at least seven RR genes that are induced by CK, yet this study found only three. This discrepancy might be due to the expression of RR genes below detectable levels in the tissues used in this study. Although probes for all RR genes are present in the gene chip used, the fact that flag calls for seven RR genes other than OsRR2, OsRR4 and OsRR6 are ‘absent’ supports this possibility (Supplementary Table S2). A large proportion of the genes identified in this study as CK regulated were also reported as CK responsive in Arabidopsis (Che et al. 2002, Hoth et al. 2003, Rashotte et al. 2003, Kiba et al. 2004, Brenner et al. 2005, Kiba et al. 2005).
|
|
Hormone metabolism and signaling
Most of the identified CK-responsive genes associated with hormone metabolism and signaling were related to CK itself, ethylene, gibberellin or auxin (Table 1 and Supplementary Table S1), as is the case with Arabidopsis (Brenner et al. 2005
). Seven of the eight genes for the precursor enzyme IPT (OsIPT1–OsIPT8) are present on the gene chip used; OsIPT2 is missing (Supplementary Table S2). These seven IPT genes were weakly expressed and flag calls for only two or three of them were ‘present’ (Supplementary Table S2). No significant CK response from IPT genes was observed under our experimental conditions, whereas a longer period of CK application repressed the accumulation of transcripts of some OsIPTs (Sakamoto et al. 2006). On first inspection, OsIPT4 appears to be induced by tZ (Supplementary Table S2). However, RT–PCR assays could not confirm its induction. This discrepancy may be due to low expression of this gene (see Supplementary Table S2, raw values). Moreover, expression levels of two genes encoding putative CK trans-hydroxylases were also low, making it difficult to evaluate their CK responsiveness (Supplementary Table S2). Up-regulation of OsCKX11 expression was observed with CK application (Table 1 and Supplementary Table S1), which is consistent with a subset of Arabidopsis CKX genes (AtCKX3, AtCKX4, AtCKX5 and AtCKX6; Hoth et al. 2003, Rashotte et al. 2003, Zimmermann et al. 2004, Brenner et al. 2005, Kiba et al. 2005, Werner et al. 2006) and a maize CKX (Ckx1; Brugière et al. 2003). The expression of OsCKX2, which is relevant to the regulation of rice grain production (Ashikari et al. 2005), was markedly induced by tZ, although expression of this gene did not meet the criteria for microarray analyses (see Materials and Methods, Fig. 1B and Supplementary Table S2). Transcript abundance of a putative gene encoding CK O-glucosyltransferase (accession No. AK068176
[GenBank]
) also increased in response to tZ application. Cytokinin O-glucosyltransferase catalyzes the inactivation of zeatin-type CKs by O-glucosylation. Up-regulation of these putative CK-inactivating genes could contribute to homeostatic control of CK activity. Two (accession Nos. AK072660
[GenBank]
and AK071680
[GenBank]
) and three (AK119483
[GenBank]
, AK060241
[GenBank]
and AK071680
[GenBank]
) purine permease (PUP) genes were up-regulated by tZ in root and leaf tissues, respectively (Supplementary Table S1). On the other hand, accumulation of transcripts of one PUP gene (accession No. AK059161
[GenBank]
) was down-regulated in response to tZ in both organs (Supplementary Table S1). In Arabidopsis, some members of the PUP family may be involved in CK transport (Bürkle et al. 2003). Transcript abundance of PUP4 was down-regulated by CK (Brenner et al. 2005). In this context, CK-responsive rice PUP might play a defined role in CK transport. Genes encoding putative HK, HP and type-B RR were not CK responsive (Supplementary Table S2), while AHK4/CRE1/WOL was up-regulated by CK (Rashotte et al. 2003, Kiba et al. 2004, Kiba et al. 2005).
Inhibition of hypocotyl elongation by CK in etiolated Arabidopsis seedlings is due to CK-induced ethylene production (Cary et al. 1995), suggesting a connection between CK and ethylene action. Post-transcriptional stabilization of at least two 1-aminocyclopropane-1-carboxylate (ACC) synthase enzymes (ACS5 and ACS9) contributes to CK-induced ethylene biosynthesis (Vogel et al. 1998, Chae et al. 2003). The ACS5 gene is up-regulated at the transcriptional level by application of CK (Vogel et al. 1998). Although the mode of interaction between CK and ethylene in rice remains unknown, the abundance of transcripts for the ethylene synthesis pathway enzymes ACC oxidase and ACC synthase increased in response to CK in both roots and leaves (Table 1 and Supplementary S1). Moreover, several genes involved in ethylene signaling were also modulated, including two putative ethylene receptors, CTR-like protein kinase, ETO1-like protein 1 and ethylene-insensitive-3-like protein (Table 1 and Supplementary S1). The regulatory interactions between CK and ethylene in rice may be at least in part mediated by the regulation of these genes.
Several genes involved in gibberellin signaling and metabolism were modulated by CK (Table 1 and Supplementary Table S1). The expression of gibberellin-inducible OsTMK, which encodes a leucine-rich repeat receptor protein kinase in deepwater rice, is down-regulated by CK. The expression of OsTMK is particularly high in regions undergoing cell division and elongation, and low in the non-growing region of the internode (van der Knaap et al. 1999), implying that CK affects rice growth regulation mediated by OsTMK. In addition, transcript abundance of the OsWRKY71 gene, encoding a transcriptional repressor of gibberellin signaling in aleurone cells (Zhang et al. 2004), was also decreased by CK (Table 1 and Supplementary Table S3), suggesting antagonism between CK and gibberellin signaling.
Positive regulation of the metabolic pathway for active gibberellin species might occur (i.e. up-regulation of GA20ox4 and down-regulation of GA2ox3) in CK-treated leaves (Table 1). This type of regulation apparently does not suit the proposed model of CK and gibberellin interactions in shoot apical meristems for maintenance of the meristem function in Arabidopsis (Jasinski et al. 2005, Yanai et al. 2005) or rice (Sakamoto et al. 2006). Although there is no robust explanation at present, there could be different modes of interaction between these hormones in different organs.
Several genes encoding auxin-regulated auxin/IAA (AUX/IAA) and auxin response factor (ARF) family transcription factors, auxin-responsive proteins and putative auxin transporters were modulated (Table 1, and Supplementary Tables S1, S3). Also, a putative IAA-amino acid hydrolase homolog was up-regulated and an auxin biosynthesizing nitrilase gene was down-regulated. Down-regulation of a nitrilase gene by CK was also reported in Arabidopsis (Brenner et al. 2005). Regulation of these genes could contribute to synergy and antagonism between CK and auxin (Coenen and Lomax 1997, Rashotte et al. 2005), but there is as yet too little known about these interactions and the genes that regulate them to suggest a model.
Transcriptional regulators
The rice genome probably encodes >2,000 transcription factors or transcriptional regulators (Gao et al. 2006; http://drtf.cbi.pku.edu.cn./). Application of CK altered the expression of a large number of transcriptional regulators, including APETALA2/ethylene-responsive element-binding protein (AP2/EREBP), basic helix–loop–helix (bHLH), WRKY, bZIP, MYB, NAC and MADS family transcription factors (Supplementary Table S3). In Arabidopsis, some CK-inducible AP2 genes, designated CYTOKININ RESPONSE FACTORS (CRFs), mediate CK responses in concert with the His–Asp phosphorelay system (Rashotte et al. 2006). Phylogenetic analyses found several CK-up-regulated rice AP2 genes that fall into a clade near the CRF genes (Supplementary Figs. S1A, S1B), indicating that the rice AP2 and CRF gene products may play a similar role. The expression of most of the affected WRKY genes was lowered in both roots and leaves (17 of 18 in roots, 14 of 17 in leaves) including OsWRKY71. This trend was also observed in Arabidopsis (Brenner et al. 2005). It is noteworthy that overexpression of OsMADS18, which showed CK inductiveness in roots (Supplementary Table S3), induced early flowering in rice, and that the formation of axillary shoot meristems was accelerated (Fornara et al. 2004).
Macronutrient acquisition and distribution
Subsets of genes for transporters of macronutrients such as nitrate, ammonium, phosphate, potassium and sulfate were repressed by the application of CK (Table 2 and Supplementary Table S1). Similar regulation was reported in Arabidopsis (Brenner et al. 2005, Kiba et al. 2005, Sakakibara et al. 2006), suggesting a common mechanism in monocots and dicots for controlling macronutrient acquisition and distribution. Regulation of these genes has been well studied in Arabidopsis. Some nitrate transport genes are induced by nitrate (Orsel et al. 2002), whereas others, such as ammonium transporter (AMT; Gazzarrini et al. 1999), as well as transporters of phosphate (PHT; Muchhal et al. 1996), potassium (HAK; Rodríguez-Navarro and Rubio 2006) and sulfate (SULTR; Grossman and Takahashi 2001), are induced by substrate starvation. In addition, CK-dependent repression of SULTR1;1 and SULTR1;2 is mediated by AHK4/CRE1/WOL (Maruyama-Nakashita et al. 2004), and induction of some phosphate starvation-responsive genes, including PHT1;1, is mediated by AHK4/CRE1/WOL and AHK3 (Franco-Zorrilla et al. 2002, Franco-Zorrilla et al. 2005). To our knowledge, there is little information about the correlation between macronutrient uptake and distribution and CK content in rice at present. Rice grown in paddy fields predominantly utilizes ammonium rather than nitrate during most of the growing period, since ammonium is the major form of inorganic nitrogen in hypoxic and anaerobic soils (Sasakawa and Yamamoto 1978). Thus, the form of available inorganic nitrogen is different between rice and most other agricultural plants. IPT3 is selectively and rapidly regulated by nitrate in Arabidopsis (Miyawaki et al. 2004, Takei et al. 2004). It would be interesting to know how rice IPT genes are regulated by nitrogen and by other macronutrient responses to understand the adaptive mechanisms to differing nutritional habitats and the role of CK as a regulator of macronutrient acquisition and distribution.
|
Protein synthesis
Previous studies have pointed out the up-regulation of genes coding for ribosomal proteins by CK in soybean (Crowell et al. 1990
) and Arabidopsis (Brenner et al. 2005, Kiba et al. 2005). In this study, the expression of 120 ribosomal protein genes was up-regulated in leaves, while only one ribosomal protein gene was up-regulated in roots (Supplementary Table S4). This contrasting result is a strong indicator of organ-specific CK regulation of genes encoding ribosomal proteins.
Overexpression of OsRR6 alters plant morphology and CK metabolism
A list of rice CK-responsive genes is now available. It became possible to examine the effects of in planta manipulation of type-A RRs on CK-responsive genes by overexpression of an RR gene. Thirteen putative type-A RR genes in rice (OsRR) were identified (Ito and Kurata 2006, Jain et al. 2006, Supplementary Table S5). We selected OsRR6 for examining the role of type-A RRs in detail, because its pattern of expression is well studied (Jain et al. 2006; Fig. 1B; Table 1 and Supplementary Table S1). To determine the intracellular localization of OsRR6, we employed an OsRR6::GFP (green fluorescent protein) fusion. The results of a bombardment expression assay with epidermal cells of rice leaf sheaths showed that OsRR6::GFP displayed a localization profile very similar to that of GFP, suggesting that OsRR6::GFP was localized mainly in the cytoplasmic space (Fig. 1C).
We generated transgenic rice plants that overexpress OsRR6 (OsRR6-ox) by fusing its coding sequence to the rice actin1 (Act1) promoter, because this promoter generally produces much higher levels of constitutive expression in rice than the cauliflower mosaic virus (CaMV) 35S promoter (Zhang et al. 1991, Sentoku et al. 2000). Callus transformed with Act1::OsRR6 showed severe retardation of shoot regeneration compared with callus transformed with a control vector (Fig. 2). Addition of N6-benzyladenine, a CK, to the regeneration medium increased the number of shoots that differentiated from wild-type rice calli (Scarpella et al. 2003). Moreover, rice calli transformed with the ipt-type MAT vector regenerated in the absence of exogenous CK (Endo et al. 2002). These observations suggest that CK promotes shoot regeneration in rice. Thus, OsRR6 probably acts as a negative regulator of CK signaling.
|
Hygromycin B-resistant plants (T0 generation) derived from transformed callus were individually transferred onto MS-agar plates containing hygromycin B for 7 d and then hydroponically grown for 75 d. Eight of 11 independent transgenic lines carrying Act1::OsRR6 (OsRR6-ox) had a dwarf phenotype with poorly developed root systems, and small panicles with a reduced branching pattern and a reduction in the number of spikelets (Fig. 3A, B). We confirmed that these transgenic plants accumulated large amounts of OsRR6 transcripts by RT–PCR (Fig. 3C). Two additional experiments using independently transformed Act1::OsRR6 plants gave essentially the same results (data not shown), and confirmed that overexpression of OsRR6 results in aberrant development.
|
Using an alternative genetic approach to determine OsRR6 function, a missense mutant of OsRR6, OsRR6D103E, was made in which the putative phosphoryl-accepting aspartate residue (Asp103) was replaced with glutamate. Because the aspartate residue is a putative phosphorylation site, the predicted gene product should lose its phosphor-accepting ability. In some experiments, aspartate to glutamate substitutions in RR proteins mimic the active and phosphorylated state (Hass et al. 2004
, Leibfried et al. 2005). Thus, the D103E mutation should activate OsRR6. However, callus transformed with Act1::OsRR6D103E showed no retardation of shoot regeneration (Fig. 2). Moreover, each of 20 OsRR6D103E-ox independent lines was indistinguishable from plants transformed with a control vector (control plants; Fig. 3A, B). These results support the hypothesis that growth defects associated with OsRR6 overexpression are due to a requirement for phosphorylation of OsRR6. The change to glutamate could have failed to mimic the activated form of the RR, or this substituted protein simply was not stable. The latter possibility is not likely because protein gel blot analyses using anti-cMyc antibody confirmed that sufficient amounts of 5x cMyc::OsRR6D103E were accumulated (data not shown). It should be noted that overexpression of a cytosolic RR, ARR22, which is quite different from authentic type-A ARRs, caused severe growth defects, particularly in the development of roots (Kiba et al. 2004). Overexpression of a mutated RR, which is changed at a putative phosphoryl-accepting aspartate to glutamate in ARR22 (Asp74), conferred no severe phenotype. Mutations of the conserved aspartate residues to asparagine in ARR4 (Asp95) and ARR6 (Asp86) did not alter their ability to reduce ARR6 promoter activity (Hwang and Sheen 2001). Moreover, overexpression of a mutated, nuclear-localized ARR7, which had been changed from Asp85 to glutamate, severely affected the function of shoot apical meristems (Leibfried et al. 2005). These lines of evidence imply that the biological function of type-A RRs is not simply to act as a negative regulator of CK signaling in all conditions, but that their phosphorylation state, as dictated by primary protein structure and/or subcellular location, affects the regulation of downstream developmental events.
Neither Arabidopsis hextuple type-A arr mutants (arr3, 4, 5, 6, 8 and 9) nor plants overexpressing ARR5, ARR6, ARR7 or ARR15 have severe phenotypes when grown under normal conditions, indicating that there are highly overlapping functions among members of this large gene family (Kiba et al. 2003, To et al. 2004, Leibfried et al. 2005). In contrast, maize mutants defective for ABPH1 (abph 1) have defects in phyllotaxis and meristem size regulation (Giulini et al. 2004). Considering what is known from Arabidopsis and maize, along with our present results, which show that simple overexpression of OsRR6 causes growth defects, type-A RR genetic redundancy may be lower in grass species.
OsRR6-ox did not set normal flowers, resulting in sterility. Very recently, it has been reported that the rice LONELY GUY (LOG) gene, which encodes a CK-activating enzyme, is required to maintain meristem activity (Kurakawa et al. 2007). The shoot meristem of log mutants flattens and prematurely stops floral organ differentiation, suggesting that CK plays a key role in proper maintenance of shoot meristem activity in rice. In this context, the sterility of OsRR6-ox might be caused by negative regulation of CK signaling in the meristem. To explore whether overexpression of OsRR6 influences endogenous CK content, we measured CK concentrations in shoots of two independently grown 75-day-old OsRR6-ox T0 generation and control plants. OsRR6-ox plants had an increased content of tZ-type CKs (Fig. 4 and Supplementary Table S6). The concentrations of tZ, the biologically most active CK, was increased by >8-fold, and the tZ derivatives tZ riboside (tZR) and tZR 5'-monophosphate (tZRMP) were increased by >4-fold. The concentrations of cis-zeatin-type, isopentenyladenine-type and dihydrozeatin-type CKs were not significantly affected. Although the increase in CK content may not be sufficient to compensate for the growth defects of OsRR6-ox, the evidence of an increase is an indicator of a homeostatic control mechanism. Interestingly, in Arabidopsis, reduced CK signaling also led to an increase in CK content, in particular when AHK3 was mutated (Riefler et al. 2006).
|
Mutual regulation between CK signaling and CK metabolism
The balance between CK biosynthesis and degradation determines endogenous CK levels. To gain some insight into the mechanism of increased CK content in OsRR6-ox, we carried out a genome-wide expression profile of OsRR6-ox. Total RNA from the same extractions as those used for quantification of CK was subjected to microarray analysis. There were changes of at least 3-fold in 441 genes in OsRR6-ox plants as compared with control plants (Table 3 and Supplementary Table S7). Interestingly, the expression of OsCKX4 and OsCKX5 was substantially reduced in OsRR6-ox transformants (Table 3 and Supplementary Table S7). In our study, none of the six detected rice IPT genes had altered transcript levels even if the threshold was set at a low stringency of 2.0 (data not shown). Down-regulation of OsCKX4 and OsCKX5 may thus partly contribute to the increased CK content of OsRR6-ox plants. A putative gene for CK-O-glucosyltransferase (accession No. AK104985 [GenBank] ) is up-regulated (Table 3 and Supplementary Table S7), which is in agreement with the observed increase of tZ-O-glucosides in OsRR6-ox plants (Supplementary Table S6). The microarray data were verified for a few selected genes by semi-quantitative RT–PCR (Fig. 5A).
|
|
Overexpression of OsRR6 also affected the expression of CK-responsive genes. Of the 441 genes modulated in OsRR6-ox, 165 were CK responsive (Fig. 5B and Supplementary Table S8). Most of the up- or down-regulated genes in OsRR6-ox were up-regulated (17 of 21; 81%) or down-regulated (126 of 144; 88%) by CK, suggesting that the increased cellular CK primarily modulates CK-responsive genes, especially CK-repressive ones, in OsRR6-ox plants. It is likely then that a mutual control mechanism links CK signaling and metabolism, which contributes to a fine-tuning of the CK response. Eighteen CK-inducible genes were down-regulated in OsRR6-ox (Fig. 5B and Supplementary Table S8). The expression of several CK-inducible genes was reduced by the over-expression of a negative regulator of CK signaling in Arabidopsis, ARR8, suggesting that CK-inducible genes could be the targets of ARR8 (Osakabe et al. 2002
). Considering the fact that OsRR6 acts as a negative regulator of CK signaling (Fig. 2), these 18 genes could be the targets of OsRR6 signaling. Although there are a few exceptions [e.g. an auxin-induced protein-related-like protein (CB667155
[GenBank]
) and a potassium transporter (AK066919
[GenBank]
), Tables 2, 3], expression of CK-responsive genes associated with hormone metabolism and signaling, and macronutrient transport was not significantly altered in OsRR6-ox (Tables 1, 2), suggesting that these categories of genes are not under the control of OsRR6 signaling. It may seem counterintuitive that type-A RRs are not included in the list of genes modulated in OsRR6-ox transformants. One possible explanation is that OsRR6 participates in the repression of transcription of some type-A RRs, as is the case with Arabidopsis, in which activation of the type-A RRs provides a negative feedback mechanism (Hwang and Sheen 2001).
Genes involved in stress responses are altered in OsRR6-ox plants
A major fraction of the genes modulated in OsRR6-ox play a particular role in the plant's response to hormones or are related to pathogen defense or responses to abiotic stress with parallel up- and down-regulation within the same category (Table 3 and Supplementary Table S7). These results imply a certain, defined role for OsRR6 in the integration of plant signal transduction. The response of OsRR6 to both CK and abiotic stresses such as salinity, dehydration and cold supports this idea (Jain et al. 2006). The sterility of transgenic rice plants that overexpress OsRR6 prevented us from experimentally examining their phenotypes under various stresses. Future analyses of knockout or RNA interference mutants of OsRR6 will enable us to define further its possible participation in stress responses. The expression of some type-A ARR genes is induced by different environmental stresses such as drought, salinity and low temperature (Urao et al. 1998). Moreover, it has been reported that the type-A ARR family participates in light signaling in a complex manner (Sweere et al. 2001, To et al. 2004, Salomé et al. 2006). Taken together, type-A RRs may not only play a role in CK signaling, but may also form a complex network that is predominantly responsible for integration, fine-tuning and cross-talk of many plant signaling pathways.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionResults and DiscussionMaterials and MethodsAcknowledgmentsReferences |
|---|
Plant materials and growth conditions
For the identification of CK-responsive genes by microarray analysis, rice (Oryza sativa L. cv Nipponbare) seeds were sterilized with 70% ethanol for 1 min and then with 2% (w/v) sodium hypochlorite for 15 min, soaked in distilled water at 30°C for 1 d, and 30 germinated seeds were transferred to a nylon net floating on tap water that had been adjusted to pH 5.5. Seedlings were grown for 2 weeks in an environment-controlled greenhouse with a photoperiod of 12 h light (30°C)/12 h dark (25°C). Whole roots and third leaves, which were cut at the lamina joint with a razor blade in water, were excised from the 2-week-old seedlings and immediately dipped in distilled water containing either 5 µM tZ in dimethylsulfoxide [DMSO; 0.1% (v/v)] or an equal volume of DMSO as a control. Each excised organ was incubated at 30°C for 30 or 120 min, harvested and stored at –80°C until the RNA was extracted. Three independent biological replicates were prepared for each experiment.
Identification of OsRR genes
OsRR genes were identified by BLAST queries in GenBank (www.ncbi.nlm.nih.gov/BLAST/), The Institute for Genomic Research (TIGR rice genome project; http://www.tigr.org/tdb/e2k1/osa1/index.shtml), the DNA Data Bank of Japan (DDBJ; http://www.ddbj.nig.ac.jp/search/blast-j.html) and the rice cDNA database at KOME (http://cdna01.dna.affrc.go.jp/cDNA/) using the OsRR6 amino acid sequence. OsRR11 cDNAs were amplified using RT–PCR with the primers 5'-GAGAGGTTGAGAAGTTGAGATGTC-3' and 5'-AAAAGTCTCCGAAACAAACAATGC-3'. Total RNA was prepared from 3-week-old seedlings with the RNeasy plant Mini Kit (Qiagen, Hilden, Germany) with RNase-free DNase I (Qiagen). cDNA was synthesized using SuperScript II RT (Invitrogen, Carlsbad, CA, USA) with oligo(dT)12–18 primers. The amplified fragment was cloned into pCR-Blunt-TOPO (Invitrogen) and sequenced to confirm the amplification fidelity. Analyses of cDNA and amino acid sequences were carried out using GENETYX-MAC version 11 (Software Development Co., Tokyo, Japan).
Construction and growth of OsRR6-ox transgenic plants
The entire coding sequence of OsRR6 was amplified from a full-length cDNA clone (accession No. AK059734
[GenBank]
) as template using the BamHI and SalI integral primers 5'-AGGATCCATGGCGGCAGCGGCGCAGGCT-3' and 5'-TGTCGACTCATCTGATACGGCTGCAGAG-3'. The resulting PCR product was cloned into the BamHI/SalI site of pGBKT7 (Clontech, Palo Alto, CA, USA), and designated pGBKT7-OsRR6. A site-directed mutant of OsRR6, OsRR6D103E, was made by amplifying pGBKT7-OsRR6 with the substitution-introducing primers 5'-CATCACCGAGTACTGGATGCCCG-3' and 5'-CGGGCATCCAGTACTCGGTGATG-3'. An NdeI/BamHI cassette containing four copies of the c-Myc epitope (4x cMyc) was introduced upstream of the OsRR6 and OsRR6D103E genes to give an N-terminal fusion. The resulting constructs, 5x cMyc-OsRR6, 5x cMyc-OsRR6D103E were amplified by PCR as XbaI/EcoRV fragments and introduced into the pActnos/Hmz binary vector (Sentoku et al. 2000) at an XbaI/SmaI site downstream of the rice actin1 gene promoter. An XbaI–EcoRV 5x cMyc cassette was also introduced into the pActnos/Hmz binary vector and used as a control. Agrobacterium tumefaciens (EHA101) carrying the above constructs was used to transform rice following the method of Hiei et al. (1994). Hygromycin B-resistant plants (defined as transgenic plants of the T0 generation) from callus grown on regeneration agar plates for 6 weeks were transferred to agar plates containing 1x Murashige and Skoog salts, 3% (w/v) sucrose and 50 µg ml–1 hygromycin-B (MS-agar plate) for 7 d, and then hydroponically grown until heading in an environment-controlled greenhouse with a photoperiod of 12 h light (30°C)/12 h dark (25°C) as described by Kamachi et al. (1991).
Particle bombardment
The coding sequence of OsRR6 was fused to the 5' terminus of the GFP gene, which was under the control of the CaMV 35S promoter (35S-sGFP [S65T]). This construct was introduced into rice leaf sheath cells by bombardment with 1 µm gold particles as described in the supplier's protocol (PDS-1000/He, Biorad, Hercules, CA, USA). After overnight incubation, transient expression was observed by confocal laser-scanning fluorescence microscopy (Fluoview IX5, Olympus, Melville, NY, USA).
Semi-quantitative reverse transcription–PCR
Total RNA was prepared using an RNeasy plant Mini Kit (Qiagen) with RNase-free DNase I (Qiagen). Approximately 1 µg of RNA was used as template for first-strand cDNA synthesis, which was performed by SuperScript II RT (Invitrogen, Carlsbad, CA, USA) with oligo(dT)12–18 primers in a reaction volume of 20 µl. A 1 µl aliquot of the reaction mixture was amplified by PCR using ExTaq DNA polymerase (TAKARA SHUZO CO. LTD, Kyoto, Japan) in a reaction volume of 50 µl with the primers 5'-ATGGCTGACGCCGAGGATATCCAG-3' and 5'-TTAGAAGCATTTCCTGTGCACAATG-3' for OsAct1 (actin1; McElroy et al. 1990), 5'-TCAGAGGACGAAGCCAAGATTGTC-3' and 5'-TTCATAGTAGCAGTAGCACTTAGC-3' for OsRR4, 5'-ATGACCGGCTACGAGCTCCTCAAG-3' and 5'ATTCTAACTTGCTAAGGCAGTAAG-3' for OsRR6, 5'-AACAACATGTGCTGCAGCCAGTGG-3' and 5'-GATCCCAAAGCTAGCTAGCTCACC-3' for lectin (accession No. M24504
[GenBank]
), 5'-TCATCGACTGCAAGAAGAACAACG-3' and 5'-ATCATTTGCAATCTCTGACCTGAC-3' for AP2/EREBP family transcription factor (accession No. AK065008
[GenBank]
), 5'-GTCCACGACGGCGAGCTCAA-3' and 5'-TCATGCGAGTGGTGACGTGA-3' for OsCKX2, 5'-ACGATGAACTTCAACCACGATGAG-3' and 5'-CATTGAGGTAGCACAGCTCTTCTC-3' for OsCKX4, and 5'-CTCATCTACCCCATGAACAG-3' and 5'-CAGAATTGTCTACACATCAG-3' for OsCKX5. The number of cycles varied depending on the gene to avoid saturation of the amplification as follows: OsAct1, 22 cycles; OsRR4 and OsRR6, 25 cycles; and lectin, AP2/EREBP family transcription factor, OsCKX2, OsCKX4 and OsCKX5, 28 cycles. A 5 µl aliquot of the reaction was separated on 0.8% (w/v) agarose gels and stained with ethidium bromide.
DNA microarray analysis
Microarray analysis was performed using a GeneChip® rice genome array (Affymetrix, Santa Clara, CA, USA). Preparation of labeled target cRNA, subsequent purification and fragmentation were carried out using One-cycle target labeling and control reagents (Affymetrix). Double-stranded cDNA was prepared from 10 µg of total RNA. Hybridization, washing, staining and scanning were performed as described in the supplier's protocol. A 10 µg aliquot of fragmented cRNA was subjected to hybridization. It should be noted that all the above manipulations were carried out independently for each RNA sample.
Data analysis was performed using GeneChip® Operating Software (GCOS; Affymetrix) and GeneSpring 7 (Agilent Technologies, Palo Alto, CA, USA). Normalization per chip and per gene of the log2 values was performed as recommended by the GeneSpring manual for Affymetrix gene chips. Briefly, per chip normalization with the 50th median of all measurements on the chip was performed. In addition, for CK treatment experiments, each treatment chip was specifically normalized to a control chip treated with DMSO for a similar time to the actual treatment.
Three independent replicates were used for CK treatment experiments. The Affymetrix flag calls (present, P; marginal, M; or absent, A) were computed to give expression classes as follows: a gene was classified as present (P) when the expression call was P for all six gene chips; a gene was classified as ‘turning from present to absent’ (P 
A) when the expression calls were P for three gene chips of the DMSO time point controls and either A or M for three gene chips of the experimental samples; and a gene was classified as ‘turning from absent to present’ (A P) when the expression calls were either A or M for three gene chips of the DMSO controls and P for three replicates of the experimental samples. Genes that did not meet these criteria were classified as ‘absent’ and eliminated from further analysis, and any gene whose mean from all six experiments for each time point was lower than 500 was discarded as being unreliable data. Furthermore, one-way analysis of variance (ANOVA) Welch's t-tests were performed using the Statistical Group Comparison tool in GeneSpring 7 software. Comparisons were made between treatments of tZ and DMSO for individual time points. A P-value cut-off of 0.05 was selected for all tests. Finally, transcripts with a ratio of induction of >2.0 and either a P or an A P classification were regarded as up-regulated. Transcripts with a ratio of induction of <0.5 and either a P or a P A classification were regarded as down-regulated.
Two independent transgenic plants were used for each microarray experiment with OsRR6-ox and control transgenic plants carrying empty vector. Only those transcripts whose flag calls were declared ‘present’ in all four chips were taken into account. Moreover, a raw cut-off level of 150 was assigned to all four-gene chip experiments. Finally, genes with a 3.0-fold change in expression were chosen. Changes in signal intensity were calculated from the mean value of the normalized values obtained from independent gene chip experiments.
Quantification of CK
Extraction and determination of CK contents from 
100 mg of rice tissue was performed using a liquid chromatography–tandem mass chromatography system (model 2695/Quattro Ultima Pt, Waters, Milford, MA, USA) as described previously (Nakagawa et al. 2005).
Supplementary material
Supplementary material mentioned in the article is available to online subscribers at the journal website www.pcp.oxfordjournals.org.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionResults and DiscussionMaterials and MethodsAcknowledgmentsReferences |
|---|
We thank Dr. M. Matsuoka, Nagoya University, for providing us with pActnos/Hmz, and Dr Y. Niwa, Shizuoka Prefectual University, for supplying 35S-sGFP[S65T DNA]. We are grateful to the Rice Genome Project of the National Institute of Agrobiological Sciences and the Rice Genome Resource Center, Japan, for providing us with full-length cDNAs of OsRR6. This work was supported by the Ministry of Education, Culture, Sports, Science and Technology (grant No. 17780052 to N.H.) and by a Grant-in-Aid for Scientific Research on the functional analysis of genes relevant to agriculturally important traits in the rice genome (grant No. IP-3003 to H.S.) from the Ministry of Agriculture, Forestry and Fisheries, Japan.
|
Footnotes |
|---|
Nucleotide sequence data from this article have been deposited in the DDBJ/EMBL/GenBank database under accession numbers AB249653 (OsRR6), AB249654 (OsRR5), AB249655 (OsRR3), AB249656 (OsRR11), AB249657 (OsRR7), AB249658 (OsRR4), AB249659 (OsRR9), AB249660 (OsRR10), AB249661 (OsRR1), AB249662 (OsRR2), AB249663 (OsRR8), AB249664 (OsRR12) and AB249665 (OsRR13). Microarray data have been deposited in The National Center for Biotechnology Information Gene Expression Omnibus (NCBI GEO) database under accession number GSE6737.
|
References |
|---|
|
TopAbstractIntroductionResults and DiscussionMaterials and MethodsAcknowledgmentsReferences |
|---|
Aoyama T and Oka A. (2003) Cytokinin signal transduction in plant cells. J. Plant Res. 116:221–231.[CrossRef][ISI][Medline]
Asakura Y, Hagino T, Ohta Y, Aoki K, Yonekura-Sakakibara K, Deji A, Yamaya T, Sugiyama T, Sakakibara H. (2003) Molecular characterization of His–Asp phosphorelay signalling factors in maize leaves: implications of the signal divergence by cytokinin-inducible response regulators in the cytosol and the nuclei. Plant Mol. Biol. 52:331–341.[CrossRef][ISI][Medline]
Ashikari M, Sakakibara H, Lin S, Yamamoto T, Takashi T, Nishimura A, Angeles ER, Qian Q, Kitano H, Matsuoka M. (2005) Cytokinin oxidase regulates rice grain production. Science 309:741–745.
Brandstatter I and Kieber JJ. (1998) Two genes with similarity to bacterial response regulators are rapidly and specifically induced by cytokinin in Arabidopsis. Plant Cell 10:1009–1019.
Brenner WG, Romanov GA, Köllmer I, Bürkle L, Schmülling T. (2005) Immediate-early and delayed cytokinin response genes of Arabidopsis thaliana identified by genome-wide expression profiling reveal novel cytokinin-sensitive processes and suggest cytokinin action through transcriptional cascades. Plant J. 44:314–333.[CrossRef][ISI][Medline]
Brugière N, Jiao S, Hantke S, Zinselmeier C, Roessler JA, Niu X, Jones RJ, Habben JE. (2003) Cytokinin oxidase gene expression in maize is localized to the vasculature, and is induced by cytokinins, abscisic acid, and abiotic stress. Plant Physiol. 132:1228–1240.
Bürkle L, Cedzich A, Döpke C, Stransky H, Okumoto S, Gillissen B, Kühn C, Frommer WB. (2003) Transport of cytokinins mediated by purine transporters of the PUP family expressed in phloem, hydathodes, and pollen of Arabidopsis. Plant J. 34:12–26.
Cary AJ, Liu W, Howell SH. (1995) Cytokinin action is coupled to ethylene in its effects on the inhibition of root and hypocotyl elongation in Arabidopsis thaliana seedlings. Plant Physiol. 107:1075–1082.[Abstract]
Chae HS, Faure F, Kieber JJ. (2003) The eto1, eto2, and eto3 mutations and cytokinin treatment increase ethylene biosynthesis in Arabidopsis by increasing the stability of ACS protein. Plant Cell 15:545–559.
Che P, Gingerich DJ, Lall S, Howell SH. (2002) Global and hormone-induced gene expression changes during shoot development in Arabidopsis. Plant Cell 14:2771–2785.
Coenen C and Lomax TL. (1997) Auxin–cytokinin interactions in higher plants: old problems and new tools. Trends Plant Sci. 2:351–356.[CrossRef][ISI][Medline]
Crowell DN, Kadlecek AT, John MC, Amasino RM. (1990) Cytokinin-induced mRNAs in cultured soybean cells. Proc. Natl Acad. Sci. USA 87:8815–8819.
D'Agostino IB, Deruère J, Kieber JJ. (2000) Characterization of the response of the Arabidopsis response regulator gene family to cytokinin. Plant Physiol. 124:1706–1717.
Endo S, Sugita K, Sakai M, Tanaka H, Ebinuma H. (2002) Single-step transformation for generating marker-free transgenic rice using the ipt-type MAT vector system. Plant J. 30:115–122.[CrossRef][ISI][Medline]
Ferreira FJ and Kieber JJ. (2005) Cytokinin signaling. Curr. Opin. Plant Biol. 8:518–525.[CrossRef][ISI][Medline]
Fornara F, Parenicova L, Falasca G, Pelucchi N, Masiero S, Ciannamea S, Lopez-Dee Z, Altamura MM, Colombo L, Kater MM. (2004) Functional characterization of OsMADS18, a member of the AP1/SQUA subfamily of MADS box genes. Plant Physiol. 135:2207–2219.
Franco-Zorrilla JM, Martín AC, Leyva A, Paz-Ares J. (2005) Interaction between phosphate-starvation, sugar, and cytokinin signaling in Arabidopsis and the roles of cytokinin receptors CRE1/AHK4 and AHK3. Plant Physiol. 138:847–857.
Franco-Zorrilla JM, Martín AC, Solano R, Rubio V, Leyva A, Paz-Ares J. (2002) Mutations at CRE1 impair cytokinin-induced repression of phosphate starvation responses in Arabidopsis. Plant J. 32:353–360.[CrossRef][ISI][Medline]
Gao G, Zhong Y, Guo A, Zhu Q, Tang W, Zheng W, Gu X, Wei L, Luo J. (2006) DRTF: a database of rice transcription factors. Bioinformatics 22:1286–1287.
Gazzarrini S, Lejay L, Gojon A, Ninnemann O, Frommer WB, Wirén NV. (1999) Three functional transporters for constitutive, diurnally regulated, and starvation-induced uptake of ammonium into Arabidopsis roots. Plant Cell 11:937–947.
Giulini A, Wang J, Jackson D. (2004) Control of phyllotaxy by the cytokinin-inducible response regulator homologue ABPHYL1. Nature 430:1031–1034.[CrossRef][Medline]
Grefen C and Harter K. (2004) Plant two-component systems: principles, functions, complexity and cross talk. Planta 219:733–742.[ISI][Medline]
Grossman A and Takahashi H. (2001) Macronutrient utilization by photosynthetic eukaryotes and the fabric of interactions. Annu. Rev. Plant Physiol. Plant Mol. Biol. 52:163–210.[CrossRef][ISI][Medline]
Hass C, Lohrmann J, Albrecht V, Sweere U, Hummel F, Yoo SD, Hwang I, Zhu T, Schafer E, Kudla J, Harter K. (2004) The response regulator 2 mediates ethylene signalling and hormone signal integration in Arabidopsis. EMBO J. 23:3290–3302.[CrossRef][ISI][Medline]
Heyl A and Schmülling T. (2003) Cytokinin signal perception and transduction. Curr. Opin. Plant Biol. 6:480–488.[CrossRef][ISI][Medline]
Hiei Y, Ohta S, Komari T, Kumashiro T. (1994) Efficient transformation of rice (Oryza sativa L.) mediated by Agrobacterium and sequence analysis of the boundaries of the T-DNA. Plant J. 6:271–282.[CrossRef][ISI][Medline]
Hoth S, Ikeda Y, Morgante M, Wang X, Zuo J, Hanafey MK, Gaasterland T, Tingey SV, Chua NH. (2003) Monitoring genome-wide changes in gene expression in response to endogenous cytokinin reveals targets in Arabidopsis thaliana. FEBS Lett. 554:373–380.[CrossRef][ISI][Medline]
Hwang I and Sakakibara H. (2006) Cytokinin biosynthesis and perception. Physiol. Plant 126:528–538.
Hwang I and Sheen J. (2001) Two-component circuitry in Arabidopsis cytokinin signal transduction. Nature 413:383–389.[CrossRef][Medline]
Imamura A, Hanaki N, Nakamura A, Suzuki T, Taniguchi M, Kiba T, Ueguchi C, Sugiyama T, Mizuno T. (1999) Compilation and characterization of Arabidopsis thaliana response regulators implicated in His–Asp phosphorelay signal transduction. Plant Cell Physiol. 40:733–742.