SEARCH

SEARCH BY CITATION

Keywords:

  • auxin;
  • lateral root development;
  • Aux/IAAs;
  • auxin response factors;
  • SLR/IAA14;
  • pericycle

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Auxin is important for lateral root (LR) initiation and subsequent LR primordium development. However, the roles of tissue-specific auxin signaling in these processes are poorly understood. We analyzed transgenic Arabidopsis plants expressing the stabilized mutant INDOLE-3 ACETIC ACID 14 (IAA14)/SOLITARY-ROOT (mIAA14) protein as a repressor of the auxin response factors (ARFs), under the control of tissue-specific promoters. We showed that plants expressing the mIAA14-glucocorticoid receptor (GR) fusion protein under the control of the native IAA14 promoter had the solitary-root/iaa14 mutant phenotypes, including the lack of LR formation under dexamethasone (Dex) treatment, indicating that mIAA14-GR is functional in the presence of Dex. We then demonstrated that expression of mIAA14-GR under the control of the stele-specific SHORT-ROOT promoter suppressed LR formation, and showed that mIAA14-GR expression in the protoxylem-adjacent pericycle also blocked LR formation, indicating that the normal auxin response mediated by auxin/indole-3 acetic acid (Aux/IAA) signaling in the protoxylem pericycle is necessary for LR formation. In addition, we demonstrated that expression of mIAA14-GR under either the ARF7 or the ARF19 promoter also suppressed LR formation as in the arf7 arf19 double mutants, and that IAA14 interacted with ARF7 and ARF19 in yeasts. These results strongly suggest that mIAA14-GR directly inactivates ARF7/ARF19 functions, thereby blocking LR formation. Post-embryonic expression of mIAA14-GR under the SCARECROW promoter, which is expressed in the specific cell lineage during LR primordium formation, caused disorganized LR development. This indicates that normal auxin signaling in LR primordia, which involves the unknown ARFs and Aux/IAAs, is necessary for the establishment of LR primordium organization. Thus, our data show that tissue-specific expression of a stabilized Aux/IAA protein allows analysis of tissue-specific auxin responses in LR development by inactivating ARF functions.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Auxin regulates many aspects of plant growth and development, including embryogenesis, tropic responses, lateral root (LR) formation, vascular development, and apical dominance (Davies, 1995). In addition, auxin regulates the transcription of many genes (Abel and Theologis, 1996; Guilfoyle et al., 1998). Recent studies have shown that two families of transcriptional regulator proteins, the auxin response factors (ARFs) and Aux/IAAs, play an important role in auxin-mediated growth and development by controlling auxin-responsive transcription (Berleth et al., 2004; Hagen and Guilfoyle, 2002; Liscum and Reed, 2002).

There are 23 members of the ARF family in the Arabidopsis genome (Hagen and Guilfoyle, 2002; Liscum and Reed, 2002; Remington et al., 2004). The N-terminus of ARF proteins contains a DNA-binding domain that binds to auxin-responsive elements (AuxREs), which are found in the promoters of many auxin-responsive genes (Guilfoyle et al., 1998; Ulmasov et al., 1997a,b, 1999a,b). The middle region of the ARF determines whether it functions as an activator or repressor. ARFs with a Gln-Leu-Ser-rich middle region (ARF5–ARF8 and ARF19) act as transcriptional activators, whereas those with a Pro-Ser-Thr-rich middle region (ARF1–ARF4 and ARF9) act as transcriptional repressors (Tiwari et al., 2003; Ulmasov et al., 1999a). The C-terminus domain (CTD), with the exception of ARF3 and ARF17, is responsible for homodimerization and heterodimerization with other ARFs, and also for heterodimerization with Aux/IAA proteins (Hardtke et al., 2004; Kim et al., 1997; Tatematsu et al., 2004; Ulmasov et al., 1999b).

Arabidopsis Aux/IAA genes encode 29 short-lived nuclear proteins (IAA1 to IAA20, and IAA26 to IAA34) (Abel et al., 1994, 1995; Kim et al., 1997; Liscum and Reed, 2002; Remington et al., 2004). Most of the Aux/IAAs have four highly conserved domains (I, II, III and IV) (Liscum and Reed, 2002; Reed, 2001). Domains III and IV are similar to the CTDs of ARFs, and are also responsible for homodimerization or heterodimerization with other Aux/IAAs, and heterodimerization with ARF proteins (Hardtke et al., 2004; Kim et al., 1997; Ouellet et al., 2001; Tatematsu et al., 2004; Ulmasov et al., 1999b). Domain I, which is similar to the conserved domain of the ethylene response factor (ERF) transcriptional repressors (Hiratsu et al., 2003; Ohta et al., 2001), can inactivate ARF function, thereby repressing auxin-responsive transcription (Tiwari et al., 2004). Domain II is important for the instability of the protein (Colón-Carmona et al., 2000; Ouellet et al., 2001; Tiwari et al., 2001; Worley et al., 2000).

Auxin stimulates the interaction between Aux/IAA proteins and the Skp1-Cullin-F-box/Transport Inhibitor Response 1 (SCFTIR1) complex, which promotes the ubiquitination and degradation of Aux/IAAs by the 26S proteasome (Dharmasiri and Estelle, 2004; Gray et al., 2001). This allows the ARFs to function in auxin-responsive transcription (Tiwari et al., 2004). However, gain-of-function mutations in domain II block interactions between Aux/IAA proteins and the SCFTIR1 complex, thus increasing the stability of the Aux/IAA proteins (Colón-Carmona et al., 2000; Gray et al., 2001; Ouellet et al., 2001; Worley et al., 2000), resulting in constitutive inactivation of ARF functions in transfection assays with Daucus carota (carrot) protoplasts (Tiwari et al., 2004).

Several gain-of-function mutants in the Aux/IAA genes have also indicated the importance of Aux/IAAs in growth and development in planta (Berleth et al., 2004; Liscum and Reed, 2002; Reed, 2001). Gain-of-function iaa mutants in domain II have been identified in at least 10 of the 29 Aux/IAA genes (iaa1/axr5, iaa3/shy2, iaa6/shy1, iaa7/axr2, iaa12/bdl, iaa14/slr, iaa17/axr3, iaa18, iaa19/msg2 and iaa28) (Fukaki et al., 2002; Hamann et al., 2002; Nagpal et al., 2000; Reed, 2001; Rogg et al., 2001; Rouse et al., 1998; Tatematsu et al., 2004; Tian and Reed, 1999; Yang et al., 2004). Most of these iaa mutants have pleiotropic phenotypes in auxin-related growth and development, and altered gene expression in response to auxin, but they have different or opposite phenotypes in optical dominance, lateral root formation, and root hair formation, indicating that Aux/IAA genes have overlapping and specific functions (Knox et al., 2003; Liscum and Reed, 2002; Reed, 2001). Forward and reverse genetic analyses have identified the functions of some ARFs in growth and development, including embryo patterning (monopteros (MP)/ARF5; Hardtke and Berleth, 1998), apical hook formation and senescence (HLS1 suppressor (HSS1)/ARF2; Li et al., 2004; Okushima et al., 2005a), floral development (ETTIN/ARF3; Sessions et al., 1997), tropic responses (nonphototropic hypocotyl 4 (NPH4)/massugu 1 (MSG1)/TIR5/ARF7; Harper et al., 2000), hypocotyl elongation (ARF2; Okushima et al., 2005a; ARF8; Tian et al., 2004), and LR formation (NPH4/MSG1/TIR5/ARF7 and ARF19; Okushima et al., 2005b; Wilmoth et al., 2005). However, the roles of other ARFs are still unknown because single mutations in them confer no distinct phenotype (Okushima et al., 2005b). Considering the strong phenotypes caused by gain-of-function iaa mutations, stabilized Aux/IAA protein may inactivate the function of multiple ARFs. Thus, it is not fully understood which ARF members regulate specific auxin-mediated processes.

Lateral root formation is one of the auxin-regulated developmental processes (Laskowski et al., 1995). In Arabidopsis, LRs are initiated from the anticlinal cell divisions in the pericycle adjacent to the two protoxylem poles (protoxylem pericycle), and this initiation is dependent on auxin (Beeckman et al., 2001; Casimiro et al., 2001, 2003; Himanen et al., 2002). A large number of mutants defective in auxin biosynthesis, auxin transport, or auxin signaling affect LR formation (Boerjan et al., 1995; Celenza et al., 1995; Hobbie and Estelle, 1995; King et al., 1995; Marchant et al., 2002; Ruegger et al., 1998). The slr mutant, a gain-of-function mutation in domain II of IAA14, has no LRs and exhibits the other auxin-related phenotypes, including limited root hair formation and reduced root gravitropism (Fukaki et al., 2002). The slr mutation blocks auxin-induced pericycle cell divisions for LR initiation, indicating that auxin-responsive transcription mediated by SLR/IAA14 is important for LR formation (Fukaki et al., 2002). However, arf7 arf19 double mutants also have few LRs, while single arf7 or arf19 mutants have a weak or subtle phenotype of LR formation, indicating that there are overlapping functions of ARF7 and ARF19 in LR formation (Okushima et al., 2005b; Wilmoth et al., 2005). ARF7and ARF19 genes are expressed in root stele tissues, including the pericycle, where IAA14 is expressed (Fukaki et al., 2002; Okushima et al., 2005b; Wilmoth et al., 2005), strongly suggesting that, in the slr mutant, stabilized mutant IAA14 (mIAA14) protein inactivates ARF7/ARF19 functions in the stele, thereby blocking LR formation. However, as IAA14 is expressed not only in the stele but also in the epidermis and lateral root cap at the root tip (Fukaki et al., 2002), it has not been determined which tissues expressing mIAA14 directly contribute to blocking of LR formation in the slr mutant.

In this paper, we examine the effects of tissue-specific expression of the stabilized mIAA14 protein on LR development to elucidate the role of tissue-specific auxin responses mediated by Aux/IAA–ARF signaling. Using an inducible system with the mIAA14-glucocorticoid receptor (GR) fusion protein (Lloyd et al., 1994; Schena et al., 1991), we present molecular genetic evidence that a normal auxin response in the protoxylem pericycle is necessary for LR initiation, and that a normal auxin response in the LR primordium is important for the establishment of LR patterning. Our data show that tissue-specific expression of a stabilized Aux/IAA protein enables a definitive examination of tissue-specific auxin responses in growth and development by inactivating ARF functions.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Establishment of an inducible system using the stabilized mutant IAA14-GR fusion protein

We previously showed that expression of mIAA14 cDNA under the native IAA14 promoter (2.0 kb) resulted in expression of the slr phenotype in roots, including the absence of LR formation, very limited root hair formation, and a reduced gravitropic response in roots (Fukaki et al., 2002). To control the timing of mIAA14 expression during development, we used an inducible system with the GR (Lloyd et al., 1994; Schena et al., 1991) and produced transgenic plants expressing mIAA14-GR fusion cDNA under the control of the native IAA14 promoter in a wild-type (Col) background. In the absence of dexamethasone (Dex), 10-day-old pIAA14::mIAA14-GR seedlings showed the normal LRs, root hairs, and gravitropism characteristic of the wild-type phenotype (Figure 1a). By contrast, 10-day-old pIAA14::mIAA14-GR seedlings grown on the 1 μm Dex-containing medium displayed almost no LR formation, almost no root hair formation (Figure 1b,c), and a reduced primary root gravitropic response (data not shown). These phenotypes are nearly identical to those of the slr mutant (Fukaki et al., 2002), indicating that the mIAA14-GR protein is functional in the presence of Dex. We also observed expression of the pCyclinB1;1::CyclinB1;1(NT)-GUS reporter which allowed us to identify LR initiation sites (Colón-Carmona et al., 1999) in the pIAA14::mIAA14-GR line. Ten-day-old Dex-treated pIAA14::mIAA14-GR/pCyclinB1;1::CyclinB1;1(NT)-GUS seedlings contained almost no LR initiation sites with GUS activity along the primary roots (Figure 1e), indicating that the mIAA14-GR protein blocks pericycle cell divisions, much as the mIAA14 protein does in the slr mutant. In addition, Dex-treated pIAA14::mIAA14-GR seedlings were less sensitive to exogenous auxin in the LR induction assay (Figure 2), i.e. Dex-treated pIAA14::mIAA14-GR seedlings produced fewer LRs in response to exogenous IAA, whereas untreated seedlings produced as many LRs as wild-type plants, indicating that mIAA14-GR could also block auxin-induced LR formation in the presence of Dex.

image

Figure 1. pIAA14::mIAA14-GR seedling phenotypes. (a, b) Dexamethasone (Dex)-untreated (a) and Dex-treated (b) 10-day-old pIAA14::mIAA14-GR seedlings. Insets show root hair formation for each treatment. Dex treatment causes solitary-root-like phenotypes, including no LR formation and few root hairs. Bar, 1 cm. (c) Number of lateral roots (LRs) per primary root in 10-day-old wild-type Columbia (Col) and pIAA14::mIAA14-GR seedlings in the presence or absence of Dex. Error bars indicate standard error (n > 15). (d, e) pCyclinB1;1::CyclinB1;1(NT)-GUS expression in Dex-untreated (d) and Dex-treated (e) 10-day-old pIAA14::mIAA14-GR seedlings. GUS activity is observed at the LR initiation site in (d) but not in (e). Bar, 100 μm. (f–i) Dex-untreated (f, h) and Dex-treated (g, i) pIAA14::GFP-mIAA14-GR roots. GFP expression at the mature root region (f, g) and at the root tip (h, i) is shown. Bar, 100 μm.

Download figure to PowerPoint

image

Figure 2. Lateral root (LR) induction assays using plants expressing the stabilized mutant IAA14/SOLITARY-ROOT–glucocorticoid receptor fusion protein (mIAA14-GR) under control of tissue-specific promoters. Four-day-old dexamethasone (Dex)-treated or Dex-untreated seedlings [Columbia (Col), solitary-root-1 (slr), pIAA14::mIAA14-GR, pSHR::mIAA14-GR, and pSCR::mIAA14-GR] were transferred onto Dex-containing or Dex-free medium supplemented with or without 0.1 μm indole-3 acetic acid (IAA). After 3 days of incubation, the number of LRs and primary root lengths were measured. The numbers of LRs per 1 mm primary root are shown. Error bars indicate the standard error (SE) (n > 10). The values of the slr-1 mutant for each treatment are all 0.0 ± 0.0 (mean ± SE).

Download figure to PowerPoint

To confirm the cellular localization of the mIAA14-GR protein, we produced plants expressing the GFP-tagged mIAA14-GR fusion protein under the control of the IAA14 promoter (pIAA14::GFP-mIAA14-GR) in a wild-type background. In the absence of Dex, the pIAA14::GFP-mIAA14-GR seedlings had the wild-type phenotype (data not shown) and had very weak GFP signals in the cytoplasm and nuclei of stele cells along the root, and in the epidermis at the root tip (Figure 1f,h). By contrast, the pIAA14::GFP-mIAA14-GR seedlings grown on Dex-containing media exhibited the slr phenotype (no LRs, few root hairs and reduced gravitropism; data not shown) and had strong GFP signals in the nuclei of stele cells along the root and in the epidermis and lateral root cap at the root tip (Figure 1g,i). The tissue specificity of GFP-mIAA14-GR expression under the control of the IAA14 promoter was the same as the pIAA14::GUS and pIAA14::mIAA14-GFP patterns (Fukaki et al., 2002; data not shown). These results indicate that the GFP-mIAA14-GR protein is functional with Dex treatment.

In previous work, the 2.0-kb IAA14 promoter was shown to be insufficient for expression in aerial shoots and hypocotyls (Fukaki et al., 2002). Consistent with previous findings, Dex-treated pIAA14::mIAA14-GR seedlings displayed no obvious phenotype in the aerial shoots and hypocotyls (data not shown). Taken together, these results indicate that the mIAA14-GR protein is functional in the presence of Dex, and acts as a repressor to block LR formation when it is expressed by the native IAA14 promoter.

LR formation is blocked in the root region where mutant IAA14-GR protein is activated

We investigated whether mIAA14-GR blocked LR formation specifically in the region where it was activated during root growth. When 4-day-old pIAA14::mIAA14-GR seedlings from media that contained no Dex, and which had no visible LRs, were transferred onto Dex-containing media and incubated for an additional 4 days, there was no production of LRs on newly grown primary roots (Figure 3b), whereas primary roots formed on Dex-free media produced LRs normally (Figure 3a). These results indicate that activation of mIAA14-GR from 4 days after germination can block subsequent LR formation. The pIAA14::mIAA14-GR seedlings grown in the absence of Dex formed several LRs in the older, proximal root zone, even after transfer onto Dex-containing media (Figure 3b), suggesting that these LRs were initiated on the Dex-free medium. By contrast, the distal (younger) root region of the Dex-untreated pIAA14::mIAA14-GR seedlings did not produce any LRs after transfer to Dex-containing media (Figure 3b), suggesting that activated mIAA14-GR could block LR formation in this region. However, when 4-day-old Dex-treated pIAA14::mIAA14-GR seedlings with no LRs were transferred onto Dex-free media, the new growth began to form normal LRs (Figure 3d), whereas the seedlings transferred onto Dex-containing media did not produce LRs (Figure 3c). These results indicate that the removal of Dex allows LR formation during subsequent root growth. In this treatment, several LRs developed from the region pre-treated with Dex after transfer onto Dex-free media, especially in the distal (younger) root region (Figure 3d), indicating that these LRs were initiated after Dex depletion, probably as a result of the inactivation of mIAA14-GR. Together, these Dex-shift experiments indicate that LR formation in pIAA14::mIAA14-GR seedlings is blocked in the root region where the mIAA14-GR protein is activated.

image

Figure 3. Effects of dexamethasone (Dex)-shift treatments on lateral root (LR) formation in pIAA14::mIAA14-GR seedlings. Four-day-old Dex-treated or Dex-untreated pIAA14::mIAA14-GR seedlings were transferred onto Dex-containing or Dex-free media, and incubated for an additional 4 days. The proximal root region above the line was pre-treated with or without Dex before transfer to the new medium. The distal root region below the line is the new growth after transfer to the new media. Dex-containing media contained 1 μm Dex. (a) Seedlings transferred from Dex-free to Dex-free medium. (b) Seedlings transferred from Dex-free to Dex-containing medium. (c) Seedlings transferred from Dex-containing to Dex-containing medium. (d) Seedlings transferred from Dex-containing to Dex-free medium.

Download figure to PowerPoint

Expression of mutant IAA14-GR protein in the stele blocks LR formation

Because IAA14 promoter activity has been observed at high levels not only in the stele but also in the epidermis at the root tip (Fukaki et al., 2002), it has not yet been determined which tissue expressing mIAA14 contributes to blocking LR formation in slr. To determine whether activated mIAA14-GR expression in the stele blocked LR formation, we used the stele-specific promoter of the SHORT-ROOT (SHR) gene in Arabidopsis (Helariutta et al., 2000). The 2.5-kb SHR promoter is activated in the initial and daughter cells of the stele, including the pericycle (Helariutta et al., 2000). When plants expressing mIAA14-GR under the control of the SHR promoter (pSHR::mIAA14-GR) were germinated on Dex media, they developed almost no LRs, whereas pSHR::mIAA14-GR seedlings not treated with Dex had normal LR development (Figure 4a–c). As shown in Figure 2, LR induction assays also showed that the Dex-treated pSHR::mIAA14-GR seedlings were less sensitive to exogenous auxin: Dex-treated pSHR::mIAA14-GR seedlings produced fewer LRs in response to exogenous auxin, indole-3 acetic acid (IAA), than the seedlings that were not treated with Dex. In contrast to the pIAA14::mIAA14-GR seedlings, Dex-treated pSHR::mIAA14-GR seedlings appeared to have normal root hairs and root gravitropism (Figure 4a,b), suggesting that the auxin response in the stele does not contribute to root hair formation and root gravitropism. These results indicate that a normal auxin response mediated by Aux/IAA signaling in the stele is necessary for LR formation.

image

Figure 4. Phenotypes of pSHR::mIAA14-GR seedlings. (a, b) Dexamethasone (Dex)-untreated (a) and Dex-treated (b) 10-day-old pSHR::mIAA14-GR seedlings. Insets show root hair formation in each treatment. Bar, 1 cm. (c) Number of LRs per primary root in 10-day-old pSHR::mIAA14-GR seedlings in the absence or presence of Dex. Error bars indicate standard error (n > 31).

Download figure to PowerPoint

Expression of mIAA14-GR in the protoxylem-adjacent pericycle blocks LR formation

Arabidopsis LRs are initiated from the protoxylem-adjacent pericycle (protoxylem pericycle) in the stele (Casimiro et al., 2001). To determine whether protoxylem pericycle-specific expression of activated mIAA14-GR blocked LR formation, we used the targeted mis-expression system with the GAL4-VP16-based enhancer trap line J0121, in which GAL4-VP16 was expressed in the protoxylem pericycle of the root. In this line, the activity of GAL4-VP16 can be monitored by GFP expression under the control of the Upstream Activating Sequence (UAS) (http://www.plantsci.cam.ac.uk/Haseloff/geneControl/catalogues/Jlines/record/record_0.html; Figure 5a,b). As shown in Figure 5(c), J0121 seedlings transformed with the 5xUAS::mIAA14-GR gene produced almost no LRs under Dex treatment, unlike the parental J0121 line. Almost no LR initiation sites were observed in Dex-treated 5xUAS::mIAA14-GR/J0121 seedlings (data not shown), indicating that expression of activated mIAA14-GR in the protoxylem pericycle is sufficient to block LR initiation. These results indicate that the normal auxin response mediated by Aux/IAA signaling in the protoxylem pericycle is crucial for LR formation, and particularly for LR initiation.

image

Figure 5. Expression of the stabilized mutant IAA14/SOLITARY-ROOT–glucocorticoid receptor fusion protein (mIAA14-GR) in the protoxylem pericycle blocks lateral root (LR) formation. (a) J0121 parental line expressing GFP in the protoxylem pericycle. (b) Scheme of the Arabidopsis root in cross-section. ep, epidermis; co, cortex; en, endodermis. (c) Dexamethasone (Dex)-treated 10-day-old 5xUAS::mIAA14-GR/J0121 seedlings which have no lateral roots (LRs) (right) and parental J0121 seedlings with many LRs (left). Numbers of LRs (mean ± standard error) for each genotypes are shown (n > 19). Bar, 1 cm.

Download figure to PowerPoint

Expression of mIAA14-GR under control of either the ARF7 or ARF19 promoter blocks LR formation, probably by inactivating ARF7/ARF19 functions

It was recently shown that ARF7 and ARF19 are key regulators necessary for the formation of LRs (Okushima et al., 2005b; Wilmoth et al., 2005). As both the 2.5-kb ARF7 promoter::GUS and 2.5-kb ARF19 promoter::GUS reporters are expressed in the stele where IAA14 is expressed (Fukaki et al., 2002; Okushima et al., 2005b), it is thought that the mIAA14 protein of the slr mutant may repress the activity of ARF7 and ARF19 through interaction between domains III and IV. To determine whether mIAA14 expression in either the ARF7 or ARF19 expression domain blocked LR formation, we produced plants expressing mIAA14-GR under the control of either the 2.5-kb ARF7 or the 2.5-kb ARF19 promoter (pARF7::mIAA14-GR, pARF19::mIAA14-GR). When plants expressing mIAA14-GR under the control of either the ARF7 or the ARF19 promoter were germinated on Dex media, they developed no LRs, as was observed in the arf7 arf19 double mutants (Figure 6b,d), whereas pARF7::mIAA14-GR and pARF19::mIAA14-GR seedlings without Dex had normal LR development (Figure 6a,c). We confirmed that IAA14 interacts with ARF7 and ARF19 in the yeast two-hybrid system (Figure 6e), supporting our hypothesis that IAA14 represses the activities of ARF7 and ARF19. These results strongly suggest that the mIAA14-GR protein expressed in the pARF7::mIAA14-GR and pARF19::mIAA14-GR transgenic plants directly inactivates ARF7/ARF19 functions where ARF7 and ARF19 are co-expressed, thereby blocking LR formation.

image

Figure 6. Phenotypes of pARF7::mIAA14-GR and pARF19::mIAA14-GR seedlings and interactions between SOLITARY-ROOT/IAA 14 and auxin response factors (ARFs) in yeast. (a–d) Dexamethasone (Dex)-untreated (a, c) and Dex-treated (b, d) 10-day-old pARF7::mIAA14-GR seedlings (a, b) and pARF19::mIAA14-GR seedlings (c, d). Bar, 1 cm. Numbers of lateral roots (LRs) (mean ± standard error) for each genotype are shown (n > 13). Bar, 1 cm. (e) Interactions between SLR/IAA14 and ARFs in the yeast two-hybrid system. Yeast cells transformed with the indicated plasmids were analyzed for reporter gene expression (β-galactosidase activity). C-terminal regions of ARF5, ARF7 and ARF19 were fused to the GAL4 DNA-binding domain (BD) and full-length SLR/IAA14 was fused to the GAL4 activation domain (AD). Plasmids with BD or AD alone (−) were used as negative controls. Data represent the mean ± standard deviation of three to six independent colonies.

Download figure to PowerPoint

Expression of mIAA14-GR in the LR primordium under control of the SCR promoter alters LR development

To determine whether expression of activated mIAA14-GR in the LR primordium affected LR primordium development, we used the promoter of the SCARECROW (SCR) gene in Arabidopsis (Di Laurenzio et al., 1996; Malamy and Benfey, 1997). The SCR gene, which is expressed in the endodermis, cortex/endodermal initials, and quiescent center (QC) in the root apical meristem, is also expressed during LR primordium development (Di Laurenzio et al., 1996; Malamy and Benfey, 1997). SCR expression starts at stage II, in the outer daughter cells after periclinal cell divisions of the pericycle, and is present in the cell lineage that makes the endodermis, cortex/endodermal initials, and QC in future LRs (Malamy and Benfey, 1997). We produced plants expressing mIAA14-GR under the control of the SCR promoter (pSCR::mIAA14-GR). Whereas 10-day-old pSCR::mIAA14-GR seedlings not treated with Dex exhibited the wild-type phenotype (Figure 7a), Dex-treated pSCR::mIAA14-GR seedlings were severely impaired in LR formation; there were no visible LRs under a dissecting microscope (Figure 7b). There were 13.4 ± 0.5 [mean ± standard error (SE)] LRs per primary root in pSCR::mIAA14-GR seedlings not treated with Dex, and 0.0 ± 0.0 LRs per primary root in Dex-treated pSCR::mIAA14-GR seedlings (n > 32). Dex treatment also caused reduced root growth, with a mean (±SE) primary root length of 56.7 (±0.7) mm for pSCR::mIAA14-GR seedlings not treated with Dex and 40.2 (±1.6) mm for Dex-treated pSCR::mIAA14-GR plants (n > 13), but did not affect root hair formation. In addition, Dex-treated pSCR::mIAA14-GR seedlings were less sensitive to exogenous IAA in the LR induction assay (Figure 2). However, we observed that 10-day-old Dex-treated pSCR::mIAA14-GR primary roots had an increased number of early LR primordia at stage II (two layers of cells; Figure 7c), and at stages III and IV (three and four cell layers of LR primordium). As shown in Table 1, whereas 10-day-old pSCR::mIAA14-GR primary roots not treated with Dex had a mean (±SE) of only 5.3 (±0.8) LR primordia per primary root during stages I–IV (20.9% of total initiation sites), 10-day-old Dex-treated pSCR::mIAA14-GR primary roots had 29.3 (±2.2) LR primordia per root during stages I–IV (97.7% of total initiation sites). The total number of initiation sites per primary root in the pSCR::mIAA14-GR seedlings was not significantly changed by Dex treatment (Table 1). These observations suggest that LRs were normally initiated in Dex-treated pSCR::mIAA14-GR primary roots but that these LR primordia might be arrested during early stages of LR development. An additional notable observation was that the 2-week-older Dex-treated pSCR::mIAA14-GR seedlings frequently developed short aberrant LRs (Figure 7e,f,h–j). These LRs had a ball-like shape and appeared not to have a functional root apical meristem (Figure 7e,f,h–j), unlike the normal LR (Figure 7d,g). Starch staining showed that these disorganized LRs had almost no differentiated columella cells with amyloplasts (Figure 7h–j), whereas normal columella cells with amyloplasts were differentiated in pSCR:mIAA14-GR LRs not treated with Dex (Figure 7g). Furthermore, in the disorganized LRs, xylem vessels were differentiated toward the tip (Figure 7i,j), and root hairs were differentiated close to the root tip (Figure 7f,j), suggesting that meristematic cells (stem cells) were not maintained in these aberrant LRs. These results suggest that expression of mIAA14-GR in the SCR domain during LR primordium formation, after stage II, caused aberrant LR patterning. Although the possibility that mIAA14 expression in the endodermis of the primary root might somehow affect LR development in a non-tissue-autonomous manner has not been excluded, these results indicate that a normal auxin response mediated by Aux/IAA signaling in the LR primordium is necessary for the establishment of LR patterning.

image

Figure 7. Phenotypes of pSCR::mIAA14-GR seedlings. (a, b) Dexamethasone (Dex)-untreated (a) and Dex-treated (b) 10-day-old pSCR::mIAA14-GR seedlings with lateral roots (LRs). Bar, 1 cm. (c) LR initiation site at stage II in the Dex-treated pSCR::mIAA14-GR primary root. The bracket indicates two layers of cells after periclinal divisions. Bar, 100 μm. (d–f) Confocal images of LRs in Dex-untreated (d) and Dex-treated (e, f) pSCR::mIAA14-GR seedlings. (e) Abnormal LR with a ball-like shape. (f) Abnormal LR with meristematic cells. Cells at the tip are expanded and differentiated. Bar, 100 μm. (g–j) Starch staining in the Dex-untreated (g) and Dex-treated (h–j) pSCR::mIAA14-GR LRs. (g) The root cap columella with amyloplasts is differentiated in the Dex-untreated LR (arrow). (h) The abnormal LR with weak starch staining (arrow). (i) The abnormal LR has no root cap columella with amyloplasts. (j) The abnormal LR in which the xylem vessels (arrowhead) are differentiated toward the root tip and some root hairs (arrow) are differentiated around the tip. Bar, 100 μm.

Download figure to PowerPoint

Table 1.  Lateral root (LR) formation in pSCR::mIAA14-GR seedlings
 Total number of initiation sites per primary rootNumber of LR primordia at stages I–IVNumber of LR primordia at stages I–IV as a percentage of total initiation sitesNumber of LR primordia at stages I–IV per cm primary root
  1. Primary roots of 10-day-old Dex-untreated and -treated pSCR::mIAA14-GR seedlings (n = 12) were used for each measurement. The values for the mean ± standard error are indicated. NS, no statistically significant difference at P > 0.05 (t-test).

  2. *Statistically significant differences at P < 0.01 (t-test).

Dex (−) pSCR::mIAA14-GR25.3 ± 1.05.3 ± 0.820.90.8 ± 0.1
Dex (+) pSCR::mIAA14-GR30.0 ± 2.3 NS29.3 ± 2.2*97.75.1 ± 0.4*

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Tissue-specific expression of the stabilized Aux/IAA protein enables identification of the auxin-responsive tissues important for LR development

Auxin regulates many growth and developmental processes in higher plants, but some of these processes are probably facilitated either by cross-talk between auxin responses in several tissues or by the auxin response in one tissue. Although mutational analyses using auxin-related mutants provide some understanding of auxin responses in which tissues regulate each process, a more complete understanding is elusive because most of the auxin-related mutants have pleiotropic phenotypes. Although LRs in Arabidopsis are known to be initiated from the protoxylem pericycle in the stele, to date no study has been performed showing the tissue-specific roles for LR development. Park et al. (2002) have shown that ectopic expression of the stabilized mutant IAA1-GR protein under the control of the Cauliflower Mosaic Virus35S promoter reduces LR formation with Dex treatment, but it was not determined which tissues expressing mIAA1-GR contribute to reduce LR formation. In this paper, we demonstrate that expression of the stabilized mutant SLR/IAA14, a gain-of-function repressor for the ARFs, alters LR development under the control of several tissue-specific promoters. We also show that the mIAA14-GR fusion protein is functional in a Dex-dependent manner, and that it blocks LR initiation or normal LR primordium development, depending on the tissues in which it is expressed. GFP-tagged mIAA14-GR protein is localized in the nucleus of cells where the pIAA14 promoter has transcriptional activity, indicating that mIAA14-GR acts in a tissue-autonomous manner. Therefore, we conclude that tissue-specific expression of the stabilized Aux/IAA protein allows us to dissect the roles of tissue-specific auxin signaling and to gain insights into the unknown Aux/IAA–ARF functions in plant growth and development.

Expression of mIAA14-GR in the stele inactivates ARF7/ARF19 functions

A recent finding that ARF7 and ARF19 are key regulators necessary for the formation of LRs strongly suggested that the mIAA14 protein of the slr mutant might repress the activity of ARF7 and ARF19 through interaction between domains III and IV (Okushima et al., 2005b; Wilmoth et al., 2005). The ARF7 and ARF19 genes are expressed in the stele tissues where IAA14 is expressed (Fukaki et al., 2002; Okushima et al., 2005b; Wilmoth et al., 2005), suggesting that IAA14 interacts with these ARFs. In this study, we demonstrated that IAA14 interacts with ARF7 and ARF19 in the yeast two-hybrid system (Figure 6e). Our data showing that Dex-treated pIAA14::mIAA14-GR and pSHR::mIAA14-GR plants were very similar to the arf7 arf19 double mutant with respect to LR formation (Figures 1 and 4) suggested that the mIAA14-GR protein expressed under the control of either the IAA14 or the SHR promoter interacted with ARF7 and ARF19, thereby inactivating their functions. In addition, mIAA14-GR expression under the control of either the ARF7 or the ARF19 promoter blocked LR formation with Dex treatment (Figure 6). These results support the hypothesis that mIAA14-GR inactivates ARF (probably ARF7/ARF19) functions where ARF7 and ARF19 are co-expressed.

Auxin response in the protoxylem pericycle is important for LR formation

We also demonstrated that expression of activated mIAA14-GR in the protoxylem pericycle blocks LR initiation (Figure 5), indicating that the normal auxin response mediated by Aux/IAA signaling in the protoxylem-adjacent pericycle is crucial for LR initiation. When LRs are initiated, protoxylem pericycle cells undergo anticlinal cell divisions (Casimiro et al., 2001; Himanen et al., 2002, 2004). Auxin transport inhibitors suppress these anticlinal cell divisions in the protoxylem pericycle, but exogenous auxin induces anticlinal cell divisions in the protoxylem pericycle (Himanen et al., 2002), a phenomenon that also indicates the importance of auxin for anticlinal cell divisions. Kurup et al. (2005) has recently shown that most of the cells in LR primordia are derived from the central file of the three cell files of the protoxylem pericycle adjacent to the xylem pole. Thus, we conclude that the normal auxin response mediated by Aux/IAA signaling is necessary for the anticlinal cell divisions in the protoxylem pericycle for LR initiation, and that this step is required for subsequent LR primordium development. However, it is possible that the auxin response in other stele tissues, including the protophloem pericycle, and xylem- and phloem-associated cells, may also be involved in LR formation. As mIAA14-GR is expressed in the protoxylem pericycle, in which native mIAA14 is genuinely expressed in the slr mutant, mIAA14-GR is thought to inactivate the same ARFs (probably ARF7/ARF19) as mIAA14 inactivates in the slr mutant.

Auxin response in LR primordia regulates the establishment of LR patterning

Our analysis of pSCR::mIAA14-GR seedlings suggests that the auxin response mediated by Aux/IAA signaling in LR primordia is also important for the establishment of LR patterning (Figure 7). The observation that Dex-treated pSCR::mIAA14-GR seedlings had more LR initiation sites during stages I–IV suggests that mIAA14-GR expression in the SCR domain in LR primordia (in the outer daughter cells after periclinal cell divisions of the pericycle, and their cell lineage; Malamy and Benfey, 1997) suppresses and/or alters subsequent LR primordium development. This phenotype is different from that of the slr mutant, in which periclinal cell division of the pericycle never occurs (Fukaki et al., 2002). Dex-treated pSCR::mIAA14-GR seedlings frequently developed aberrant, disorganized LRs (Figure 7). Such phenotypes in pSCR::mIAA14-GR LRs are novel and have never been reported in gain-of-function iaa or loss-of-function arf mutants. In pSCR::mIAA14-GR plants, mIAA14-GR protein is ectopically expressed in the tissue in which native mIAA14 is not expressed in the slr mutant. Although the specific ARFs and Aux/IAAs functioning in the SCR expression domain are not known, our data indicate that the unknown ARF(s), which could be negatively regulated by the mIAA14-GR protein in the SCR expression domain, may regulate the establishment of LR patterning in wild-type plants. Recent genetic studies using multiple mutations in the PIN-FORMED (PIN) genes encoding the auxin efflux facilitators, and the weak allele of gnom mutants defective in the proper localization of PIN proteins, have shown that polar auxin transport is important both for root meristem organization and for the establishment of LR primordium development after the anticlinal pericycle cell divisions (Benkova et al., 2003; Blilou et al., 2005; Geldner et al., 2004). The disorganized LRs in pSCR::mIAA14-GR seedlings may be a result of altered auxin signaling in which expression of the essential components for LR meristem establishment (including PIN genes) is altered. In addition, the phenotypes in the abnormal LRs of pSCR::mIAA14-GR seedlings, including the lack of normal columella root cap cells, and the existence of differentiated vascular tissues and root hairs toward the root tip, are similar to the defects observed in the plethora1 (plt1) plt2 double mutants, in which the root stem cells are not maintained, thereby losing normal root distal patterning (Aida et al., 2004). These observations suggest that the PLETHORA-mediated stem cell niche, which regulates root distal patterning, is altered in the disorganized LRs. Because SCR is also expressed in the QC lineage in LR primordium formation (Malamy and Benfey, 1997), expression of mIAA14-GR driven by the SCR promoter, probably in the QC lineage, might disrupt normal root patterning. Although there remains a possibility that endodermal expression of mIAA14 in the primary root might somehow affect LR development, our observation of the disorganized LRs in pSCR::mIAA14-GR seedlings provides an entry point to study the mechanisms of the establishment of LR patterning.

The role of Aux/IAAs and ARFs in auxin-mediated growth and development processes

Root hair formation was blocked by the expression of mIAA14-GR under the control of the native IAA14 promoter, which is expressed in the epidermis at the differentiation zone (Figure 1), whereas mIAA14-GR expression limited to the stele (pSHR) or the endodermis (pSCR) did not affect root hair formation (Figures 4 and 7). These results indicate that root hair formation is not primarily regulated by auxin signaling in the stele or endodermis, but is dependent on SLR/IAA14-mediated auxin signaling in the epidermis. Consistent with this, most of regulators of root hair formation [GLABRA2 (GL2), CAPRICE (CPC), and WERWOLF (WER)] are specifically expressed in the root epidermis (Lee and Schiefelbein, 1999; Wada et al., 2002). In addition to the slr/iaa14 mutant, gain-of-function mutations in AXR2/IAA7 and AXR3/IAA17 also block root hair formation (Knox et al., 2003; Leyser et al., 1996; Wilson et al., 1990). Therefore, ARFs responsible for root hair formation should be present in the epidermis, which could be negatively regulated by IAA7, IAA17, and IAA14 proteins.

In Arabidopsis, there are many Aux/IAA (29) and ARF (23) regulatory protein family members (Liscum and Reed, 2002; Remington et al., 2004). Previous studies have shown that there appears to be no distinct specificity in Aux/IAA and ARF interactions, at least in the yeast two-hybrid system (Hardtke et al., 2004; Kim et al., 1997; Ouellet et al., 2001; Tatematsu et al., 2004; Ulmasov et al., 1999b; this study). However, it is unknown whether this is the case in planta. Phenotypic differences among gain-of-function iaa mutants may be a result of functional differences among Aux/IAA proteins (specificity for interaction with ARFs) and/or differences in their expression (e.g. transcription levels in response to auxin, and organ/tissue specificity). For example, the slr mutation blocks LR formation but does not affect embryonic (primary) root formation (Fukaki et al., 2002). This is because IAA14 is not expressed during embryogenesis (Abel et al., 1995). Concordant with this observation, ectopic expression of mIAA14 during embryogenesis under the control of either the SCR or the SHR promoter, which drives expression in ground tissue or provascular tissue during embryogenesis (Helariutta et al., 2000; Wysocka-Diller et al., 2000), resulted in mp/arf5- and bdl/iaa12-like phenotypes, including the lack of embryonic root and hypocotyl (HF, YN and MT, unpublished data; Hamann et al., 2002; Hardtke and Berleth, 1998). These results suggest that mIAA14 inactivates ARF functions, possibly those of ARF5, during early embryogenesis, as the stabilized bdl/iaa12 protein does, thereby causing the mp/arf5- and bdl/iaa12-like phenotypes. IAA14 could interact with ARF5 in the yeast two-hybrid system (Figure 6e). These results are also in agreement with the observation that phenotypic differences between slr and bdl mutants are a result of differences in their expression profiles. Weijers et al. (2005) also demonstrated, in promoter-swapping experiments in which stabilized bdl/iaa12, iaa13, and shy2/iaa3 proteins were expressed, that specificity in Aux/IAA function is transcriptionally regulated in most cases. Intriguingly, they also showed that there are functional differences among Aux/IAA proteins with respect to their inhibitory effects on ARFs (Weijers et al., 2005). As in the case of the pSCR::mIAA14-GR plants, when stabilized mutant IAA protein is ectopically expressed in the tissues where the IAA protein is not expressed, it is important to interpret the data with due consideration for the native interactions that may occur between Aux/IAAs and ARFs in the tissues, even if some transcriptional or developmental changes occur. Although it has not been conclusively determined whether stabilized mIAA14 can inactivate all ARFs with the CTD, similar experiments using various tissue-specific promoters with gain-of-function Aux/IAA repressors will clarify these issues.

In summary, we have shown that tissue-specific expression of stabilized SLR/IAA14 protein, as a repressor of the ARFs, provides a powerful tool for understanding the tissue-specific auxin responses mediated by Aux/IAA–ARF signaling in plant growth and development, especially in LR development. We present evidence that normal auxin signaling mediated by Aux/IAA–ARF interactions in the protoxylem pericycle is crucial for LR formation, and that normal auxin signaling in LR primordia is important for the establishment of LR patterning. The use of tissue-specific promoters in combination with stabilized Aux/IAA repressors can provide a foundation for understanding the molecular basis of various developmental processes regulated by multiple Aux/IAA–ARF interactions. In addition, controlling the stage- and tissue-specific expression of these repressors will allow us to manipulate plant growth and development.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Plant material and growth conditions

The Arabidopsis ecotypes used in this study were Columbia (Col) and C24. The slr-1 mutant was described previously (Fukaki et al., 2002). The GAL4 enhancer trap line J0121 (http://www.plantsci.cam.ac.uk/Haseloff/geneControl/catalogues/Jlines/record/record_0.html) was obtained from the Arabidopsis Biological Stock Center (Ohio State University, Ohio, USA). Peter Doerner (University of Edinburgh, Edinburgh, UK) kindly supplied the pCycB1;1:: CycB1;1(NT)-GUS seeds. Seeds were surface-sterilized and plated on Murashige and Skoog (MS) medium containing 1.0% sucrose solidified with 0.5% agar as described previously (Fukaki et al., 1996). For Dex treatment, Dex was added to a final concentration of 1 μm in MS medium from a 30 mm stock in ethanol. When plant hormones were added, they were derived from 10 mm stocks in dimethyl sulfoxide (DMSO). Kanamycin was added to a final concentration of 25 μg ml−1 from a 50 mg ml−1 stock in water. Plates and transplanted seedlings were grown at 23°C under constant white light, as described previously (Fukaki et al., 1996).

Phenotypic characterization

Root growth and the number of LRs were measured under a dissecting microscope. The number of early LR primordia was counted using Nomarski differential interference contrast optics (Nikon, Tokyo, Japan). GUS activity was detected by incubating the seedlings containing transgenes at 37°C for 2 h in 50 mm sodium phosphate, pH 7, containing 1 mm potassium ferricyanide, 1 mm potassium ferrocyanide and 0.3 mg ml−1 5-bromo-4-chloro-3-indolyl-glucuronide (X-Gluc). After incubation, samples were rinsed with 70% ethanol and observed under a Nomarski optics microscope (Nikon). GFP fluorescence in the roots was observed using fluorescent microscopy (Nikon). For observation of detailed root anatomy, root samples were counterstained with 10 μg ml−1 propidium iodide (Sigma Chemical Co., St Louis, MO, USA) and imaged using confocal microscopy (LSM510; Carl Zeiss Co., Ltd, Japan) with the rhodamine channel (red: propidium iodide). Starch staining was performed as described previously (Fukaki et al., 1998).

Construction of transgenic plants

To generate the mIAA14-GR translational fusion, a modified GR fragment was inserted in-frame immediately before the stop codon of the mIAA14 coding region in the pGEM T Easy vector (Fukaki et al., 2002). To do this, inverse PCR amplification of the mIAA14 coding region was performed by amplifying the entire pGEM-T Easy vector containing the mIAA14 coding region using two primers, 5′-TGAACAAAAAAAAAAGAGGACAATAT-3′ and 5′-TGATCTGTTCTTGAACTTCTCCATTG-3′. The 0.84-kb GR coding region with a short linker sequence (GlyGlyGly) at both ends was amplified from the 35S::GVG plasmid containing the GAL4-VP16-GR sequence (Aoyama and Chua, 1997) (kindly provided by Takashi Aoyama, Kyoto University, Uji, Japan) using two end primers, 5′-gggggtggcCAGCAAGCCACTGCAGGAGTC-3′ and 5′-gggtccaccTTTTTGATGAAACAGAAGCTT-3′, and cloned into the SmaI site of pUC19 (Takara Bio Inc., Kusatsu, Japan). The 0.84-kb SmaI fragment containing the GR coding region and the whole T-Easy vector fragment containing the mIAA14 coding region at its 3′ end were ligated and the resultant 1.5-kb mIAA14-GR fusion was confirmed by nucleotide sequencing. The 1.5-kb mIAA14-GR fragment was inserted between the 2.0-kb IAA14 promoter and the nopaline synthase (nos) 3′ region by replacing the GUS coding region in the IAA14 promoter–GUS construct (Fukaki et al., 2002). For the construction of the transgenes pSHR::mIAA14-GR and pSCR::mIAA14-GR, each promoter fragment (kindly provided by Philip N. Benfey, Duke University, Durham, NC, USA) was inserted upstream of the mIAA14-GR-nos 3′ by replacing the IAA14 promoter fragment of pIAA14::mIAA14-GR. For the construction of the transgenes pARF7::mIAA14-GR and pARF19::mIAA14-GR, each promoter fragment was amplified from the BAC clone containing the target gene (T1M15 for ARF7 and T29M8 for ARF19) using primer sets (5′-ggcgtcgacAGTACGTAGATTATTTTCCAC-3′ and 5′-ggcgtcgacGATCACTCAACTTTACTTTCTC-3′ for the ARF7 promoter, and 5′-ggcgtcgacAGAGAGTGTGTGTGGTTACGA-3′ and 5′-ggcgtcgacGGTTTATAGAAAGAACGAAAAAATTGG-3′ for the ARF19 promoter) and was cloned into the SalI site of the pBluescript SK+ vector. Each ARF promoter was inserted upstream of mIAA14-GR-nos 3′ by replacing the IAA14 promoter fragment of pIAA14::mIAA14-GR. These constructs were introduced into Agrobacterium tumefaciens MP90 and the strains containing each construct were used to transform wild-type Col plants using the floral dip method (Clough and Bent, 1998). T1 seeds were selected on media containing 25 μg ml−1 kanamycin, and resistant seedlings were transferred to soil and allowed to set T2 seeds. T3 KanR homozygous lines were used in the Dex-inducible experiments.

To generate a GFP-mIAA14-GR translational fusion, the 0.7-kb GFP coding region with a short linker sequence (GlyGlyGly) at both ends (Fukaki et al., 2002) was inserted in-frame at the start codon of the mIAA14-GR coding region cloned in the pBluescript SK+ vector. To do this, inverse PCR amplification of the mIAA14-GR coding region was performed by amplifying the entire pBluescript vector containing the mIAA14-GR coding region using two primers, 5′-ATGAACCTTAAGGAGACGGAG-3′ and 5′-ATCTCTTCTTGCTGTCTATAT-3′. The 0.7-kb SmaI fragment containing the GFP coding region and the whole vector fragment containing the mIAA14-GR coding region at its 3′ end were ligated, and the sequence of the resultant 2.2-kb GFP-mIAA14-GR fusion was confirmed. The 2.2-kb GFP-mIAA14-GR fragment with XbaI sites at both ends was inserted between the 2.0-kb IAA14 promoter and the nopaline synthase 3′ region by replacing the GUS coding region in the IAA14 promoter–GUS construct (Fukaki et al., 2002). This construct was introduced into A. tumefaciens MP90 and strains containing the construct were used to transform wild-type Col plants. T1 seeds were selected on media containing 25 μg ml−1 kanamycin, and resistant seedlings were transferred to soil and allowed to set seeds. T2 KanR seedlings were used in the Dex-inducible experiments.

To generate the 5xUAS::mIAA14-GR construct, the 0.24-kb 5xUAS fragment was amplified from the 5xUAS-GFP vector, kindly provided by Jim Haseloff (University of Cambridge, Cambridge, UK) using two end primers: 5′-gtcgacAAGCTTGCATGCCTGCAGGTC-3′ and 5′-gtcgacCTGCAGGTCGTCCTCTCCAAA-3′. The resultant 0.24-kb PCR fragment, which contained SalI sites at both ends, was cloned into T-vector (Stratagene, La Jolla, CA, USA) and the sequence was confirmed. The 0.24-kb 5xUAS fragment was inserted in the SalI site upstream of the mIAA14-GR-nos 3′ region in a binary vector based on pBI101. These constructs were introduced into A. tumefaciens MP90 and strains containing the construct were used to transform the J0121 (ecotype C24) line. T1 seedlings that were impaired in LR formation were selected on standard MS medium, transferred to soil and allowed to set T2 seeds. These T2 seeds were sown on Dex-containing media to confirm the linkage between the phenotype and presence of the transgene.

Yeast two hybrid experiment

The cDNA fragments encoding the C-terminus of ARF5 (amino acids 778–902), ARF7 (amino acids 1031–1164) and ARF19 (amino acids 952–1086) were amplified from a flower cDNA library using the following primer sets: 5′-gagaattcAATAGTAAAGGCTCATCATGGCAG-3′ and 5′-cagtcgacGTTACATTTATGAAACAGAAGTCTTAAGATCG-3′ for ARF5, 5′-agtcgacaAGCTCAGACTCAGCGAATGCG-3′ and 5′-cagtcgacTCACCGGTTAAACGAAGTGGC-3′ for ARF7, and 5′-gagaattcAATCAGACTCAACGAATGCG-3′ and 5′-cagtcgacCTATCTGTTGAAAGAAGCTGCAGC-3′ for ARF19. The full-length IAA14 open reading frame was amplified using two primers, 5′-cgaattcATGAACCTTAAGGAGACGGAGC-3′ and 5′-tgtcgacTCATGATCTGTTCTTGAACTTCTCC-3′. PCR products were subcloned into pCR-Blunt II TOPO (Invitrogen, Carlsbad, CA, USA) and sequenced before in-frame insertion into pAD-GAL4-2.1 or pBD-GAL4 Cam (Stratagene, CA, USA) via EcoRI/SalI (IAA14, ARF5 and ARF19) or SalI (ARF7) sites. Constructs were introduced into Saccharomyces cerevisiae Y190 cells, and transformants were subjected to assays for β-galactosidase activity as previously described (Kaiser et al., 1994).

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

We wish to thank Philip N. Benfey for the SCR and SHR promoters, Peter Doerner for seeds, Jim Haseloff, Takashi Aoyama and Takehide Kato for other vectors, and ABRC for BAC clones. We also thank Keiko Uno for technical assistance, and Mitsuhiro Aida for helpful comments on the manuscript. This work was supported in part by a Grant-in-Aid to HF for Scientific Research on Priority Areas (Molecular Basis of Axis and Signals in Plant Development, Grant No. 15031218 and No. 17027019) and a Grant-in-Aid to HF for Scientific Research for Young Scientists (Grant No. 15770028) from The Ministry of Education, Culture, Sports, Science and Technology, Japan, and a grant to MT from the ‘Research for the Future’ program of the Japan Society for the Promotion of Science.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References