Brassinosteroids regulate organ boundary formation in the ... - PNAS

Brassinosteroids regulate organ boundary formation in the shoot apical meristem of Arabidopsis Joshua M. Gendrona,1, Jiang-Shu Liua,1, Min Fana, Ming-Yi Baia, Stephan Wenkela, Patricia S. Springerb, M. Kathryn Bartona, and Zhi-Yong Wanga,2 a Department of Plant Biology, Carnegie Institution for Science, Stanford, CA 94305; and bDepartment of Botany and Plant Sciences, Center for Plant Cell Biology, University of California, Riverside, CA 92521

Edited by Sarah Hake, University of California, Berkeley, CA, and approved November 7, 2012 (received for review June 25, 2012)

Spatiotemporal control of the formation of organ primordia and organ boundaries from the stem cell niche in the shoot apical meristem (SAM) determines the patterning and architecture of plants, but the underlying signaling mechanisms remain poorly understood. Here we show that brassinosteroids (BRs) play a key role in organ boundary formation by repressing organ boundary identity genes. BR-hypersensitive mutants display organ-fusion phenotypes, whereas BR-insensitive mutants show enhanced organ boundaries. The BR-activated transcription factor BZR1 directly represses the CUP-SHAPED COTYLEDON (CUC) family of organ boundary identity genes. In WT plants, BZR1 accumulates at high levels in the nuclei of central meristem and organ primordia but at a low level in organ boundary cells to allow CUC gene expression. Activation of BR signaling represses CUC gene expression and causes organ fusion phenotypes. This study uncovers a role for BR in the spatiotemporal control of organ boundary formation and morphogenesis in the SAM. hormone

| steroid

T

he shoot apical meristem (SAM) maintains a dynamic population of stem cells, from which aerial organs differentiate continuously as the plant grows (1, 2). Cell division at the center of the meristem maintains the stem cell population, whereas cells at the periphery of the meristem are recruited into lateral organ primordia, which undergo accelerated cell division and expansion to form lateral organs such as leaves and flowers. In contrast, cells surrounding the organ primordia become growth-arrested and enter a quiescent state, forming organ boundaries that separate organs from the central meristem and from adjacent organs (3). Temporal and spatial control of organ primordium initiation and boundary formation determines developmental patterns and is critical for normal plant architecture (3–5). Defects in organ boundary formation cause organ fusions. Arabidopsis mutants with organ-fusion phenotypes have helped to identify the genetic components required for organ boundary formation. These include the CUP-SHAPED COTYLEDON genes (CUC1, CUC2, and CUC3) (6, 7) and LATERAL ORGAN FUSION1 (LOF1) (4) genes, which encode boundary cell-specific transcription factors. In addition, the initial establishment of organ primordia and boundaries requires orientated transport of the phytohormone auxin; alternating auxin-maximum and minimum domains specify the organ primordia and boundaries, respectively (8, 9). Defects in auxin transport or signaling disrupt the expression patterns of the CUC genes (8, 10). Brassinosteroid (BR) is a major growth-promoting steroidal hormone that plays important roles in a wide range of developmental processes (11). BR regulates gene expression and plant development through a well-characterized receptor kinase–mediated signaling pathway (12). BR binds to and activates the receptor kinase BRI1 (13, 14), which initiates a signaling cascade that activates the members of the BZR1 family of transcription factors through antagonistic actions of the GSK3-like kinase BIN2 and protein phosphatase 2A (PP2A) (12, 15–17). BZR1 binds to DNA to regulate more than 1,000 BR target genes, leading to a wide range of developmental 21152–21157 | PNAS | December 18, 2012 | vol. 109 | no. 51

and physiological responses (18, 19). BR-deficient and -insensitive mutants show several developmental defects, including dwarfism, decreased apical dominance, photomorphogenesis in the dark, stomatal clustering, and male sterility (11, 20, 21). BR has also been implicated in development of root meristems and vascular bundles (22–25). Here we present evidence for an important role for BR signaling in shoot-derived organ development. We show that genetic disruption of the BR pathway alters plant architecture in Arabidopsis by affecting organ boundary formation. BR inhibits the expression of CUC genes in organ boundaries through transcriptional repression by BZR1, and a reduced level of BZR1 in the boundary cells is required for normal CUC gene expression and boundary formation. Our study links BR signaling to the organ boundary development pathway and demonstrates that spatial regulation of the BR pathway is necessary for proper organ development in the SAM. Results BR Homeostasis Is Necessary for Proper Organ Boundary Formation.

The dominant BR-hypersensitive bzr1-1D mutant of Arabidopsis, caused by enhanced PP2A dephosphorylation of BZR1 (17), showed architectural defects (15), including bending at stembranch junctions and at the base of siliques (Fig. 1 A, B, D, F, and I). WT Arabidopsis plants had a straight main inflorescence stem (Fig. 1A), whereas all bzr1-1D mutant plants showed stem bending toward the axillary branch and cauline leaf (Fig. 1B). In contrast, the stem of BR-deficient mutant det2-1 was bent away from the axillary branch (Fig. 1C). There was a general trend that mutants with reduced BR signaling, such as bri1-5 and bin2-1, showed bending away from the axillary branch like det2-1 (Fig. 1D), and plants with increased BZR1 activity, such as bzr1-1D or transgenic plants expressing hyperactive forms of bzr1-1D-CFP or BZR1S173A, showed stem bending toward lateral organs (Fig. 1 B and D) (26). These observations reveal a role for BR in controlling plant architecture through BZR1. Examination of mutants using light- and scanning-electron microscopy (SEM) revealed that the stem bending of bzr1-1D was caused by fusion of the cauline leaf and axillary branch to the main stem, whereas these structures were well separated in WT (Fig. 1 E and F; Fig. S1 A–F). In contrast, the axillary junctions of det2-1 showed deeper clefts between the main stem and axillary branch (Fig. 1G; Fig. S1 G and H). The BR-insensitive bin2-1 and bri1-5 mutants showed similar phenotypes as det2-1 (Fig. S1 I and J),

Author contributions: J.M.G., J.-S.L., M.F., M.-Y.B., S.W., M.K.B., and Z.-Y.W. designed research; J.M.G., J.-S.L., M.F., and M.-Y.B. performed research; J.M.G., J.-S.L., M.F., M.-Y.B., S.W., P.S.S., M.K.B., and Z.-Y.W. contributed new reagents/analytic tools; J.M.G., J.-S.L., M.F., M.-Y.B., S.W., P.S.S., M.K.B., and Z.-Y.W. analyzed data; and J.M.G. and Z.-Y.W. wrote the paper. The authors declare no conflict of interest. This article is a PNAS Direct Submission. 1

J.M.G. and J.-S.L. contributed equally to the work.

2

To whom correspondence should be addressed. E-mail: [email protected].

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. 1073/pnas.1210799110/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1210799110

-20 -40

b

K

s

c

L

G b

b

s

c

c

M

I

O

N

P

Fig. 1. BRs affect plant architecture. (A–D) BR mutants show altered stem bending. A WT plant (A) has straight main stem, the stem of the bzr1-1D mutant (B) bends toward the axillary branches and cauline leaves, and that of the det2-1 mutant (C) bends away from the lateral branch. (D) Measurement of stem bending toward (positive angle) or away from (negative angle) the lateral branch in WT and various BR mutants. (E–G) SEM image of the junction between stem (s), branch (b), and cauline leaf (c) in WT (E), bzr1-1D (F), and det2-1 (G) plants. (H and I) Light and SEM images of straight silique of WT (H) and bending silique of bzr1-1D (I) with a stamen fused to it (arrow). (J) SEM image showing fusion between a stamen and carpel in a mature flower of bzr1-1D. Lower panel is enlarged portion of upper panel. (K and L) SEM images of stamens in WT (K) and bzr1-1D (L). Arrows show the fused region between stamens in bzr11D. (M and N) All WT seedlings (M) have two separate cotyledons, and a small percentage of bzr1-1D seedlings (N) have fused cotyledons. (O and P) SEM images of siliques from WT (O) and bri1-5 (P). Green lines mark the replums (boundaries between valves).

whereas the bzr1-1D;bri1-116 double mutant showed stem bending and leaf-stem fusion similar to the bzr1-1D single mutant, consistent with BZR1 acting downstream of BRI1 (Fig. S1K). Microscopic analyses of sections of the first axillary branch junctions showed that the WT stem contained elongated cells in organized files, and the cells of the axil were small, unelongated, and restricted to a small area at the base of the axillary branch (Fig. S1B). In the bzr1-1D mutant, however, cell files were disrupted, and the unelongated cells were bigger and occupied a larger area at the base of the axillary stem (Fig. S1E). In WT plants, the boundary between the cauline leaf and stem was clearly visible in the SEM image, and the corresponding area in the section showed small boundary cells (Fig. S1 B and C). In contrast, the bzr1-1D mutant showed no visible boundary or distinguishable small boundary cells at the junction of the cauline leaf and stem (Fig. S1 E and F). Opposite to bzr1-1D, the det2-1 mutant plant showed a deeper cleft between the lateral branch, and the main stem and cells in the organ boundaries were smaller than in WT (Fig. S1H). These observations suggest that the organ-fusion defects in bzr11D are caused by alteration of the organ boundaries. The bzr1-1D mutant also showed bending of siliques at the silique-pedicel junction (Fig. 1 H and I; Fig. S1L), and microscopic examination revealed that such bending was due to stamen-carpel fusion (Fig. 1 I and J). It seems that the differential elongation between the stem and the fused leaf caused stem bending toward the branch side, and the fusion to a nonelongating mature stamen restricted carpel elongation and caused silique bending. Many

bzr1-1D mutant flowers also contained fused stamens (Fig. 1 K and L; Table1). Whereas WT had no fused stamens, 32.75% of bzr1-1D flowers had at least one pair of fused stamen, a higher frequency than mutants known to have stamen fusion defects, such as cuc2-3 and cuc3-105. Further analysis of bzr1-1D revealed a low but significant frequency (31/∼10,000 for bzr1-1D compared with 1/∼10,000 for WT) of fusion between the two cotyledons in young seedlings (Fig. 1 M and N). These observations suggest that bzr1-1D has a general defect in various organ boundaries of the SAM and the floral meristem. To further confirm that the organ-boundary defects of bzr1-1D reflect functions of BR, we examined effects of overexpression of the BR biosynthetic gene DWF4. DWF4 overexpression caused organ fusion in a WT background and enhanced the organ fusion phenotype of bzr1-1D (Fig. S1 M–O). Furthermore, BR treatment of WT primary inflorescences induced stamen-to-stamen fusion (Fig. S1Q). These results show that the organ fusion phenotypes of the bzr1-1D mutant result from an increase in BR signaling, causing inappropriate cell expansion and overgrowth in the organ boundary. The stem bending away from the branch in the BR-deficient or -insensitive mutants is likely caused by a reduction in growth of the boundary region, leading to early or more effective separation of lateral organ primordia from the central meristem. Indeed, high frequencies of extrafloral organs resulting from ectopic boundary formation were observed in the bri1-5 and det21 mutants (48% and 3.8%, respectively) (Fig. 1 O and P; Table S1). These observations suggest that inactivation of BR signaling

Table 1. Percentage of flowers that contain stamen-to-stamen fusions Genotype

Col

bzr1-1D

cuc2-3

cuc3-105

cuc2; cuc3

cuc2; bzr1-1D

cuc3; bzr1-1D

cuc2;cuc3; bzr1-1D

% Fusion

1.00 ± 1.61

32.75 ± 12.66

3.32 ± 4.16

12.87 ± 5.54

37.40 ± 14.00

48.65 ± 10.32

44.60 ± 12.58

76.36 ± 14.71

Gendron et al.

PNAS | December 18, 2012 | vol. 109 | no. 51 | 21153

PLANT BIOLOGY

F s

BZR1-CFP

J

H

E

bzr1-CFP!

bin2-1 bzr1-1D

0

det2

20

bri1-5

40 WT

D Angle off axis

C

B

A

leads to ectopic boundary formation causing extra floral organs, and activation of BR signaling leads to diminished boundary formation causing organ fusion. BZR1 Directly Controls the Expression of Organ Boundary Genes. To understand the molecular mechanism through which BR inhibits organ boundary formation, we examined the expression of known organ boundary-specific genes in the axil junctions of BR mutants using quantitative RT-PCR (qRT-PCR) (Fig. 2A). The results showed that the expression of the CUC1, CUC2, CUC3, and LOF1 genes were repressed in both the bzr1-1D mutant and the bzr1-1DCFP transgenic plants but were activated in the det2-1 mutant. These findings mirror the expression pattern of the known BZR1repressed DWF4 gene (Fig. 2A). In contrast, the expression level of LOB, a boundary-specific gene that acts in a pathway distinct from the CUC and LOF genes (4), was slightly up-regulated in both bzr11D and bzr1-1D-CFP plants and repressed in det2-1, opposite to the expression changes of the CUC and LOF genes (Fig. 2A). The CUC1::beta-glucuronidase (GUS), CUC2::GUS, and CUC3::GUS

A

E

Relative expression

Col

control

+BL

Col

bzr1-1D

bzr1-1D-CFP

+ BRZ

CUC1

B

promoter-reporters are expressed at the junctions between the rosette leaves and cotyledons in WT seedlings (6, 27). When the seedlings were grown on medium containing brassinolide, the most active form of BR, the expression levels of these reporter genes were significantly reduced (Fig. 2B). In contrast, expression was increased when seedlings were grown on media containing the BR biosynthetic inhibitor brassinazole (Fig. 2B). In qRT-PCR and reporter gene assays, CUC3 showed more dramatic suppression by bzr1-1D and BR treatment than CUC1 and CUC2. Similarly, BR repressed the expression of the LOF1::GUS reporter in a dosedependent manner (Fig. 2C). Consistent with increased LOB expression detected by qRT-PCR, LOB::GUS expression was observed along the stem-leaf fusion in bzr1-1D (Fig. 2D). LOB expression defines the boundary cells between the leaf and the stem (28), and the expanded expression of this marker suggests that the fusion is caused by an overgrowth of the boundary region. When the CUC3::GUS reporter line was crossed into the bzr11D mutant or bzr1-1D-CFP transgenic background, the CUC3:: GUS expression in the organ boundary regions was reduced (Fig.

CUC3

Relative BZR1 binding

CUC2

F

C LOF1::GUS

G

H

WT

0

1

cuc2; cuc3

100

I LOB::GUS

D

WT

bzr1-1D

WT

bzr1-1D

Relative expression

BL (nM)

cuc2; cuc3

LOF1

Col

cuc2

cuc3

cuc2cuc3

Fig. 2. BR inhibits organ boundary formation through BZR1-mediated repression of CUC and LOF genes. (A) Quantitative RT-PCR analysis of expression levels of boundary identity genes (CUC1, CUC2, CUC3, LOF1, and LOB) in the first and second axillary branch junctions of WT (Col), bzr1-1D and det2 mutants, and a bzr11D-CFP transgenic line with a stronger phenotype than bzr1-1D. (B) The CUC genes are down-regulated by BL. Promoter-GUS transgenic plants were grown on medium containing 10 nM epi-brassinolide (+BL) or 1 μM brassinazole (BRZ) for 9 d and stained for 6 h. (C) LOF1::GUS seedling were grown on 1/2 MS medium with indicated concentration of BL for 10 d in the light before GUS staining. (D) LOB::GUS expression in WT and bzr1-1D. (E) CUC3::GUS expression in the leaf junctions of 2-wk-old seedlings (Top), lateral branch junction (Middle), and inflorescence (Bottom) of WT, bzr1-1D, and bzr1-1D-CFP transgenic line. (F) ChIPqPCR analysis of BZR1 binding to the promoter regions of CUC1, CUC2, and CUC3 (see promoter maps in Fig. S2). The known BZR1 target gene DWF4 and nontarget CNX5 are used as positive and negative controls. (G and H) bzr1-1D has similar stamen-pistil (G) and stamen-stamen (H) fusion phenotypes to the cuc2; cuc3 double mutant. For each panel from left to right: WT, bzr1-1D, and cuc2;cuc3. (I) Quantitative RT-PCR analysis of the expression levels of LOF1 in WT, cuc2, cuc3, and cuc2;cuc3 mutants grown on 1/2 MS medium for 5 d. All data of relative expression were measured as the ratio to the UBC30 gene. Error bars are SEs.

21154 | www.pnas.org/cgi/doi/10.1073/pnas.1210799110

Gendron et al.

BZR1 Accumulates in Distinct Locations in the Meristem. The specific effects of bzr1-1D on CUC genes and organ boundaries suggest that BZR1 activity must be spatially controlled in the meristem to allow proper organ and boundary formation. The BZR1-YFP protein exhibited a discrete pattern of accumulation in the floral meristem and sepal primordia but was absent in the boundary regions, whereas the dominant form of BZR1 (bzr1-1D-CFP), which causes organ fusions, was localized uniformly throughout the floral buds (Fig. S3 B and C). We further crossed the pBZR1::BZR1-YFP and pBZR1::bzr1-1D-CFP transgenic lines to compare the accumulation of BZR1 and bzr1-1D in the same cells. In the F1 hybrid plants expressing both BZR1-YFP and bzr1-1D-CFP, BZR1-YFP accumulated at high levels in the cell nucleus in the floral meristem and

sepal organ primordia, but at a much lower level in the boundary cells. This was in contrast to the uniform distribution of bzr1-1DCFP (Fig. 3 A–D; Movie S1). Similarly, in the apical meristem region, BZR1-YFP accumulated at high levels in the shoot apical meristem and floral primordia but at low levels in the boundary cells. This pattern was again contrasted by bzr1-1D-CFP, which accumulates uniformly at high levels throughout the meristem, primordial, and the boundary region (Fig. 3 E–H; Movie S2). BZR1 accumulation is tightly regulated by BR-dependent phosphorylation/dephosphorylation, and the bzr1-1D mutation increases BZR1 binding to PP2A and causes BR-independent dephosphorylation and nuclear accumulation of BZR1 (12, 17). Therefore, the uniform accumulation of bzr1-1D indicates that the BZR1 gene is transcribed uniformly in the SAM regions, but the WT BZR1 protein is inactivated in the boundary cells because of reduced levels of BR signaling. Consistent with the pattern of bzr11D accumulation, a BZR1-promoter-GUS reporter gene showed uniform expression (Fig. S3A). These results indicate that inactivation of BR signaling in boundary cells is critical for expression of organ boundary identity genes and proper formation of the organ boundaries. These results provide the evidence that BR plays a role as a positional cue or morphogen in plant development. Discussion Our work provides evidence for a unique role for BR signaling in plant shoot architecture. BR signaling is repressed in the organ boundary cells, either because of a reduced BR level or BR sensitivity, to allow expression of organ boundary identity genes CUCs and LOFs, which are required for growth arrest and organ boundary formation. BRs do not undergo long-distance transport and thus are thought to promote growth by acting locally to increase cell elongation/expansion (29). Unlike auxin, whose elaborate transport and distribution is required for lateral organ initiation (22), BRs are not required for organ formation and have not been recognized to play a role in patterning (21, 30). However, recent studies have provided evidence for BR quantitatively modulating specific developmental processes, such as shoot regeneration (22), root meristem size (23, 24), and stomata formation (31). Here we demonstrate a specific role of BR in temporal control of organ boundary formation with impact on the plant architecture. First, the opposite effects of BR hyperresponse and hyporesponse mutants on organ boundaries

bzr1-1D-CFP

BZR1-YFP A

B

Merged C

SEM D

FM FM B B

SP

SP

E

F FP

G

H FP B

B SAM

SAM

Fig. 3. Differential accumulation of BZR1 in the nucleus of primordium and boundary cells is required for proper boundary formation. BZR1-YFP transgenic line of Arabidopsis was crossed with a bzr1-1D-CFP line, and the distributions of BZR1-YFP (A and E) and bzr1-1D-CFP (B and F) in a floral bud (A–C) or apical meristem (E–G) of the F1 plants were analyzed by confocal microscopy. C and G are merges of panels A with B and E with F. (D and G) SEM images of a floral bud (D) or apical meristem (G) show boundary areas (B) between the sepal primordium (SP) and floral meristem (FM) (A–D) or between shoot apical meristem (SAM) and floral primordium (FP) (E–H).

Gendron et al.


PLANT BIOLOGY

2E). To test whether BZR1 directly regulates CUC3 expression, we performed ChIP assays using a pBZR1::BZR1-myc transgenic line. As shown in Fig. 2F and Fig. S2, BZR1 binds strongly to the CUC3 promoter region and weakly to the CUC1 and CUC2 promoter regions compared with BZR1 binding to DWF4, a known BZR1-target gene (19). These results demonstrate that BR represses the expression of CUC1, CUC2, CUC3, and LOF1 genes, and at least for CUC3, the regulation appears to be caused by direct binding of BZR1 to the CUC3 promoter. Consistent with BZR1 repression of CUC gene expression, bzr1-1D’s organ fusion phenotypes are very similar to those of the cuc mutants, particularly the cuc2;cuc3 double mutant (6) (Fig. 2 G and H; Table1). The bzr1-1D stamen fusion phenotype was only slightly enhanced in the bzr1-1D;cuc3 double (Table 1), consistent with bzr1-1D repressing CUC3 expression strongly (Fig. 2A). The bzr1-1D;cuc2 and bzr1-1D;cuc2;cuc3 mutants had enhanced fusion phenotypes compared with bzr1-1D, cuc2, or cuc2;cuc3 mutants alone (Table 1), which is consistent with the result that BZR1 regulates CUC2 more weakly than CUC3 (Fig. 2A), and that BZR1 regulates additional boundary determinants, such as CUC1 or other unknown factors. To test whether direct repression of CUC genes could lead to repression of LOF1, we analyzed LOF1 expression in the cuc2, cuc3, and cuc2;cuc3 mutants using qRT-PCR. Fig. 2I shows that the level of LOF1 RNA was reduced in both cuc2 and cuc3 mutants and was more dramatically reduced in the cuc2;cuc3 double mutant, indicating that CUC2 and CUC3 positively regulate LOF1 expression. Thus, BZR1 repression of CUC3 may contribute to the reduced expression of LOF1.

provide strong genetic evidence for such a role. Second, direct repression of organ boundary identity genes by BZR1 demonstrates a specific mechanism for BR regulation of the boundary developmental program. Finally, the pattern of BZR1 accumulation in the SAM is consistent with its inhibitory role in boundary formation and its positive role in promoting cell growth. Nuclear accumulation of BZR1 is tightly controlled by upstream BR signaling (12, 32), whereas transcription of the BZR1 gene seems ubiquitous as indicated by both pBZR1::bzr1-1D-CFP and pBZR1::GUS. Thus, the discrete pattern of BZR1 accumulation in the SAM may reflect a distribution pattern of either BR itself or a key upstream BR signaling component. Notably this pattern is overall similar to the distribution of auxin (33). Auxin and BR have long been known to act synergistically and to elicit largely overlapping transcriptome changes (34). It was shown recently that auxin increases the expression of a BR biosynthetic gene (35). It seems therefore possible that the pattern of auxin in the SAM, established by self-regulated transport (33, 36), evokes a similar pattern of BR distribution, leading to accumulation of BZR1 in organ primordia but its absence in organ boundary cells. In addition, boundary cell-specific factors may negatively regulate BR level or BR signaling. For example, the LOB transcription factor is reported to activate BAS1, which encodes an enzyme that inactivates BRs (37). The pattern of BR action in the SAM is likely established by both hormone interactions and developmental program. Enhanced BR signaling represses organ boundary genes through BZR1-mediated transcriptional repression, leading to organ fusion and architectural alteration (Fig. 1 B, F, and I). Our results show that BZR1 directly represses CUC genes and possibly reduces LOF gene expression indirectly, as CUCs are required for LOF expression. Whereas this regulatory scheme is likely to contribute to BR regulation of boundary formation, the strong organ-fusion phenotypes of the cuc2,cuc3,bzr1-1D triple mutant suggest that additional BZR1-target genes might be involved. For example, several members of the LOB-domain (LBD) gene family were identified as BZR1 target genes (19). On the other hand, the expression level of LOB itself is not affected by BR or bzr1-1D, but its expression domain is expanded in bzr1-1D, which is consistent with the finding that LOB is upstream of the BR pathway (37). Together the evidence indicates that BR is a key link in the regulatory network specifying boundary cell function. Organ-fusion phenotypes have not been reported in BR mutants of other species. However, overexpression of BRI1 in tomato reduced leaf serration (38), a trait that is regulated by CUC2 activity in Arabidopsis (39) and by a related no apical meristem (NAM) factor in tomato (40). Thus, BR’s role in regulating organ boundary development is likely conserved in dicots. Interestingly, BR also has strong effects on plant architecture in rice (41), a monocot, but through a distinct mechanism. BR promotes rice leaf bending by increasing the expansion of adaxial cells of the laminar joint, through BZR1 regulation of a pair of helix loop helix/basic helix loop helix (HLH/bHLH) factors (42). Such independent mechanisms for a common function support an adaptive value for BR regulation of plant architecture, perhaps to optimally position photosynthetic organs according to environmental or endogenous conditions in addition to coregulation of other photosynthetic aspects such as expression of photosynthetic genes and stomata formation (31, 43–45). Detailed analyses of the architecture of BR mutants in other species will advance our understanding of the evolution of BR’s role in plant architecture. Conversely, the recent 1. Sablowski R (2004) Plant and animal stem cells: conceptually similar, molecularly distinct? Trends Cell Biol 14(11):605–611. 2. Sablowski R (2007) The dynamic plant stem cell niches. Curr Opin Plant Biol 10(6):639–644. 3. Aida M, Tasaka M (2006) Genetic control of shoot organ boundaries. Curr Opin Plant Biol 9(1):72–77. 4. Lee DK, Geisler M, Springer PS (2009) LATERAL ORGAN FUSION1 and LATERAL ORGAN FUSION2 function in lateral organ separation and axillary meristem formation in Arabidopsis. Development 136(14):2423–2432.

21156 | www.pnas.org/cgi/doi/10.1073/pnas.1210799110

findings of positive effects of ovarian steroid hormones on stem cell proliferation in mammals (46, 47) suggest that BR promotion of boundary cell growth may reflect an ancient function of steroid hormones in promoting stem cell proliferation. Materials and Methods Materials and Growth Conditions. WT, various mutants, and transgenic Arabidopsis thaliana plants were in the Columbia-0 ecotype, except that bri1-5 is in the WS ecotype background. The mutants and transgenic lines have been reported previously for cuc2-3 and cuc3-105 (6), CUC1::GUS, CUC2:: GUS, CUC3::GUS (48), and pBZR1::bzr1-1D-CFP (15). To generate the pBZR1:: BZR1-YFP transgenic line, A 1,091-bp 5′ intergenic region and the full-length BZR1 coding sequence without stop codon were cloned into the GATEWAY entry vector pENTRY/SD/D-TOPO (Invitrogen) and then recombined into the destination vector pEarlyGate 311, and the construct was transformed into Arabidopsis via Agrobacterium tumefaciens GV3101. Seeds were sterilized with 70% (vol/vol) ethanol containing 0.1% Triton X-100 for 10 min and then washed two times with 95% (vol/vol) ethanol. After air drying on filter paper, the seeds were grown on 0.7% phytagel plates containing half-strength Murashige-Skoog (MS) nutrient and 1% sucrose. The imbibed seeds were treated in 4 °C for 2 d and grown in a growth chamber with 16-h light/8-h dark cycles. Seven days later, the plantlets were transferred into soil and grown in a green house supplemented to 16 h light/d. Total RNA Isolation and qRT-PCR Analysis. Total RNA was isolated from the first and second stem-cauling leaf junction of bolting plants or 7-d-old young seedlings using the TRIzol RNA extraction Kit (Invitrogen). First-strand cDNA was synthesized using AMV reverse transcriptase (Fermentas), and qRT-PCR was performed using SYBRRGreen reagent (BioRad) according to the manufacturer’s instructions. Data were from three biologically repeats and normalize to the UBQ30 gene, which was used as an internal reference. GUS Staining. For GUS staining, fresh tissue was excised from plants, placed directly in GUS substrate solution (0.1 M NaPO4, 10 mM EDTA, 0.5 mM Kferricyanide, 0.5 mM K-ferrocyanide, 1 mM X-glucuronide, and 0.1% Triton X-100), and vacuum infiltrated until the tissue sank to the bottom of the solution. Samples were placed at 37 °C until strong histochemical staining was visible in some samples. Photos were taken under a light microscope. SEM and Critical Point Drying. Critical point drying was performed after fixation and dehydration in a DCP-1 critical point drying apparatus (Denton Vacuum) to replace ethanol with liquid CO2 and remove CO2. Dried or fresh tissues were visualized using a Quanta 200 (FEI) scanning electron microscope. Gold coating and other preparatory techniques are not needed with this microscope. Confocal Microscopy. Confocal microscopy was performed on live tissue mounted on slides. The tip region of inflorescence was placed down with the meristem region facing the slide. The tissue was compressed with a coverslip so that young flowers and inflorescence meristems were touching the slide. The samples were imaged on an inverted confocal microscope (Leica SP5), and optical sections were reconstructed using ImageJ and falsely colored in Photoshop. ChIP. ChIP was performed as described previously (19). 35S::BZR1-myc transgenic plants (17) and WT control plants were grown together on 1/2 Murashige and Skoog salt (MS) medium containing 1% sucrose under 16-h light/8-h dark in a growth chamber for 5 d. ChIP was performed using the anti-myc antibody (Sigma). ChIP DNA was analyzed by qPCR, and BZR1 binding was calculated as the ratio between the BZR1-myc sample and WT control. Data are normalized to the CNX5 control gene. ACKNOWLEDGMENTS. We thank Dr. Dongmei Cao for performing tissue sectioning, Dr. Doris Wagner for providing the CUC1, CUC2, and CUC3::GUS reporter lines; and Ulrich Kutschera and Jia-Ying Zhu for comments on the manuscript. This research was supported by National Institutes of Health Grant R01 GM066258 (to Z-y.W).

5. Aida M, Tasaka M (2006) Morphogenesis and patterning at the organ boundaries in the higher plant shoot apex. Plant Mol Biol 60(6):915–928. 6. Hibara K, et al. (2006) Arabidopsis CUP-SHAPED COTYLEDON3 regulates postembryonic shoot meristem and organ boundary formation. Plant Cell 18(11): 2946–2957. 7. Aida M, Ishida T, Fukaki H, Fujisawa H, Tasaka M (1997) Genes involved in organ separation in Arabidopsis: An analysis of the cup-shaped cotyledon mutant. Plant Cell 9(6):841–857.

Gendron et al.

Gendron et al.

28. Shuai B, Reynaga-Peña CG, Springer PS (2002) The lateral organ boundaries gene defines a novel, plant-specific gene family. Plant Physiol 129(2):747–761. 29. Symons GM, Reid JB (2004) Brassinosteroids do not undergo long-distance transport in pea. Implications for the regulation of endogenous brassinosteroid levels. Plant Physiol 135(4):2196–2206. 30. Nakamura A, et al. (2006) The role of OsBRI1 and its homologous genes, OsBRL1 and OsBRL3, in rice. Plant Physiol 140(2):580–590. 31. Kim TW, Michniewicz M, Bergmann DC, Wang ZY (2012) Brassinosteroid regulates stomatal development by GSK3-mediated inhibition of a MAPK pathway. Nature 482(7385):419–422. 32. Gendron JM, Wang ZY (2007) Multiple mechanisms modulate brassinosteroid signaling. Curr Opin Plant Biol 10(5):436–441. 33. Rast MI, Simon R (2008) The meristem-to-organ boundary: More than an extremity of anything. Curr Opin Genet Dev 18(4):287–294. 34. Vert G, Nemhauser JL, Geldner N, Hong F, Chory J (2005) Molecular mechanisms of steroid hormone signaling in plants. Annu Rev Cell Dev Biol 21:177–201. 35. Chung Y, et al. (2011) Auxin stimulates DWARF4 expression and brassinosteroid biosynthesis in Arabidopsis. Plant J 66(4):564–578. 36. Grunewald W, Friml J (2010) The march of the PINs: Developmental plasticity by dynamic polar targeting in plant cells. EMBO J 29(16):2700–2714. 37. Bell EM, et al. (2012) Arabidopsis LATERAL ORGAN BOUNDARIES negatively regulates brassinosteroid accumulation to limit growth in organ boundaries. Proc Natl Acad Sci USA 109:21146–21151. 38. Holton N, et al. (2007) Tomato BRASSINOSTEROID INSENSITIVE1 is required for systemin-induced root elongation in Solanum pimpinellifolium but is not essential for wound signaling. Plant Cell 19(5):1709–1717. 39. Nikovics K, et al. (2006) The balance between the MIR164A and CUC2 genes controls leaf margin serration in Arabidopsis. Plant Cell 18(11):2929–2945. 40. Berger Y, et al. (2009) The NAC-domain transcription factor GOBLET specifies leaflet boundaries in compound tomato leaves. Development 136(5):823–832. 41. Yamamuro C, et al. (2000) Loss of function of a rice brassinosteroid insensitive1 homolog prevents internode elongation and bending of the lamina joint. Plant Cell 12(9):1591–1606. 42. Zhang LY, et al. (2009) Antagonistic HLH/bHLH transcription factors mediate brassinosteroid regulation of cell elongation and plant development in rice and Arabidopsis. Plant Cell 21(12):3767–3780. 43. Luo X-M, et al. (2010) Integration of light- and brassinosteroid-signaling pathways by a GATA transcription factor in Arabidopsis. Dev Cell 19(6):872–883. 44. Oh E, Zhu JY, Wang ZY (2012) Interaction between BZR1 and PIF4 integrates brassinosteroid and environmental responses. Nat Cell Biol 14(8):802–809. 45. Bai MY, et al. (2012) Brassinosteroid, gibberellin and phytochrome impinge on a common transcription module in Arabidopsis. Nat Cell Biol 14(8):810–817. 46. Joshi PA, et al. (2010) Progesterone induces adult mammary stem cell expansion. Nature 465(7299):803–807. 47. Asselin-Labat ML, et al. (2010) Control of mammary stem cell function by steroid hormone signalling. Nature 465(7299):798–802. 48. Kwon CS, et al. (2006) A role for chromatin remodeling in regulation of CUC gene expression in the Arabidopsis cotyledon boundary. Development 133(16):3223–3230.


PLANT BIOLOGY

8. Furutani M, et al. (2004) PIN-FORMED1 and PINOID regulate boundary formation and cotyledon development in Arabidopsis embryogenesis. Development 131(20): 5021–5030. 9. Heisler MG, et al. (2005) Patterns of auxin transport and gene expression during primordium development revealed by live imaging of the Arabidopsis inflorescence meristem. Curr Biol 15(21):1899–1911. 10. Aida M, Vernoux T, Furutani M, Traas J, Tasaka M (2002) Roles of PIN-FORMED1 and MONOPTEROS in pattern formation of the apical region of the Arabidopsis embryo. Development 129(17):3965–3974. 11. Clouse SD, Sasse JM (1998) BRASSINOSTEROIDS: Essential regulators of plant growth and development. Annu Rev Plant Physiol Plant Mol Biol 49:427–451. 12. Kim TW, Wang ZY (2010) Brassinosteroid signal transduction from receptor kinases to transcription factors. Annu Rev Plant Biol 61:681–704. 13. Hothorn M, et al. (2011) Structural basis of steroid hormone perception by the receptor kinase BRI1. Nature 474(7352):467–471. 14. She J, et al. (2011) Structural insight into brassinosteroid perception by BRI1. Nature 474(7352):472–476. 15. Wang ZY, et al. (2002) Nuclear-localized BZR1 mediates brassinosteroid-induced growth and feedback suppression of brassinosteroid biosynthesis. Dev Cell 2(4): 505–513. 16. He JX, Gendron JM, Yang Y, Li J, Wang ZY (2002) The GSK3-like kinase BIN2 phosphorylates and destabilizes BZR1, a positive regulator of the brassinosteroid signaling pathway in Arabidopsis. Proc Natl Acad Sci USA 99(15):10185–10190. 17. Tang W, et al. (2011) PP2A activates brassinosteroid-responsive gene expression and plant growth by dephosphorylating BZR1. Nat Cell Biol 13(2):124–131. 18. He J-X, et al. (2005) BZR1 is a transcriptional repressor with dual roles in brassinosteroid homeostasis and growth responses. Science 307(5715):1634–1638. 19. Sun Y, et al. (2010) Integration of brassinosteroid signal transduction with the transcription network for plant growth regulation in Arabidopsis. Dev Cell 19(5):765–777. 20. Li J, Nagpal P, Vitart V, McMorris TC, Chory J (1996) A role for brassinosteroids in lightdependent development of Arabidopsis. Science 272(5260):398–401. 21. Szekeres M, et al. (1996) Brassinosteroids rescue the deficiency of CYP90, a cytochrome P450, controlling cell elongation and de-etiolation in Arabidopsis. Cell 85(2): 171–182. 22. Cheon J, Park SY, Schulz B, Choe S (2010) Arabidopsis brassinosteroid biosynthetic mutant dwarf7-1 exhibits slower rates of cell division and shoot induction. BMC Plant Biol 10:270. 23. González-García MP, et al. (2011) Brassinosteroids control meristem size by promoting cell cycle progression in Arabidopsis roots. Development 138(5):849–859. 24. Hacham Y, et al. (2011) Brassinosteroid perception in the epidermis controls root meristem size. Development 138(5):839–848. 25. Ibañes M, Fàbregas N, Chory J, Caño-Delgado AI (2009) Brassinosteroid signaling and auxin transport are required to establish the periodic pattern of Arabidopsis shoot vascular bundles. Proc Natl Acad Sci USA 106(32):13630–13635. 26. Gampala SS, et al. (2007) An essential role for 14-3-3 proteins in brassinosteroid signal transduction in Arabidopsis. Dev Cell 13(2):177–189. 27. Aida M, Ishida T, Tasaka M (1999) Shoot apical meristem and cotyledon formation during Arabidopsis embryogenesis: interaction among the CUP-SHAPED COTYLEDON and SHOOT MERISTEMLESS genes. Development 126(8):1563–1570.

Supporting Information Gendron et al. 10.1073/pnas.1210799110

Fig. S1. (A–P) Close-up views of cauline leaf axils of WT (Col and WS) and brassinosteroid (BR) mutant or transgenic plants as indicated in each panel. (A, D, G, I, J, K–M, O, and Q) Light image of branch junction areas. (Scale bar, 1 mm.) (B, E, and H) Light microscope images of sections of the axils of WT (Col), bzr1-1D, and det2-1. Red arrows point to the junction between lateral branch and main stem, where the det2-1 plant shows a deeper cleft. The yellow arrows point to the axil cells and boundary cells of different sizes between the samples. The yellow lines separate the cells of main stem organized in files from the area of unelongated axil cells. (C, F, and N) Scanning electron microscopy (SEM) images of branch junctions. (L) Inflorescences of WT and bzr1-1D. Arrows point to bending at silique-pedicel junctions with stamens attached. (M–O) DWF4 overexpression causes fusion between the cauline leaf and lateral branch (yellow arrows). (Q) SEM image of a WT flower treated with 500 nM brassinolide every day for 12 d, showing fusion of two stamens.

Gendron et al. www.pnas.org/cgi/content/short/1210799110

1 of 3

Fig. S2. Diagrams of CUP-SHAPED COTYLEDON (CUC)1, CUC2, and CUC3 promoters, showing putative BZR1 binding sites and regions tested by ChIP-qPCR. White bars show promoter regions of CUC1 (1.5 kb), CUC2 (4 kb), and CUC3 (5 kb) that contain the BR-response element (BRRE; BZR1 binding site), which are marked as solid circles. The black bars marked p1–p4 show the regions analyzed by PCR after ChIP (results shown in Fig. 2F). Numbers show position relative to transcription start (+1).

Fig. S3. BZR1 accumulates in meristem and primordial cells but not in boundary cells. (A) Histochemical staining of 7-μm serial sections of the pBZR1-GUS reporter gene in the inflorescence apex. (B) Confocal images of floral buds of pBZR1::BZR1-YFP transgenic plants show differential BZR1 accumulation in the meristem (M), boundary (B), and sepal primordia (SP) of a young floral bud. (C) Confocal image of a floral bud of pBZR::bzr1-1D-CFP plant shows even accumulation of the dominant bzr1-1D in a young floral bud.

Table S1. Number and percentage of ovaries that have two or more than two valves per ovary in BR mutants Genotype Col bzr1-1D det2-1 bri1-5

Two valves

More than two valves

Percentage with more than two valves

100 100 127 33

0 0 5 31

0 0 3.8 48.4


2 of 3

Movie S1. Three-dimensional view of BZR1-YFP (green) and bzr1-1D-CFP (red), constructed from confocal images of a floral primordium, as shown in Fig. 3 A–C. Movie S1

Movie S2. Three-dimensional view of BZR1-YFP (green) and bzr1-1D-CFP (red), constructed from confocal images of an apical meristem, as shown in Fig. 3 E–G. Movie S2


3 of 3