Cytokinin regulates the activity of reproductive meristems, flower organ size, ovule formation, and thus seed yield in Arabidopsis thaliana

Abstract

The size and activity of the shoot apical meristem is regulated by transcription factors and low molecular mass signals, including the plant hormone cytokinin. The cytokinin status of the meristem depends on different factors, including metabolic degradation of the hormone, which is catalyzed by cytokinin oxidase/dehydrogenase (CKX) enzymes. Here, we show that CKX3 and CKX5 regulate the activity of the reproductive meristems of Arabidopsis thaliana. CKX3 is expressed in the central WUSCHEL (WUS) domain, while CKX5 shows a broader meristematic expression. ckx3 ckx5 double mutants form larger inflorescence and floral meristems. An increased size of the WUS domain and enhanced primordia formation indicate a dual function for cytokinin in defining the stem cell niche and delaying cellular differentiation. Consistent with this, mutation of a negative regulator gene of cytokinin signaling, ARABIDOPSIS HISTIDINE PHOSPHOTRANSFER PROTEIN 6, which is expressed at the meristem flanks, caused a further delay of differentiation. Terminal cellular differentiation was also retarded in ckx3 ckx5 flowers, which formed more cells and became larger, corroborating the role of cytokinin in regulating flower organ size. Furthermore, higher activity of the ckx3 ckx5 placenta tissue established supernumerary ovules leading to an increased seed set per silique. Together, the results underpin the important role of cytokinin in reproductive development. The increased cytokinin content caused an ~55% increase in seed yield, highlighting the relevance of sink strength as a yield factor.

Figure 1.

Characterization of ckx T-DNA Insertion Alleles. (A) Positions of T-DNA insertions in the ckx mutants. The length of the genomic CKX gene sequences are given in base pairs. The insertional mutants were identified by PCR screening and the site of insertion determined by DNA sequencing of the border fragments. Black boxes represent exons, white boxes represent introns, and triangles indicate T-DNA insertion sites. (B) CKX gene expression in the wild type (WT) and insertional mutants. RNA from 10-d-old seedlings was used as template for the RT-PCR. Actin2 was included as a control.

Figure 2.

The Formation of Flower Primordia Is Increased in the ckx3 ckx5 Mutant. (A) and (B) Inflorescences of the wild type (WT) (A) and ckx3 ckx5 (B). (C) Alignment of flowers from wild-type and ckx3 ckx5 inflorescences, approximately stage 13-16 (Smyth et al., 1990). (D) to (H) Scanning electron micrographs of the main flowering apex of the wild type (D), ckx3 (E), ckx5 (F), ckx3 ckx5 (G), and 35S:CKX1 (H). The arrow in (H) indicates the inflorescence meristem. Stage 2-6 floral buds are numbered from the youngest to oldest. (I) and (J) Close-up pictures of wild-type (I) and ckx3 ckx5 (J) inflorescence meristems. Bars = 100 μm in (D) to (H) and 10 μm in (I) and (J).

Figure 3.

Lateral Organ Initiation and Phyllotaxis of ckx Mutants. (A) The number of siliques on the main stem during one life cycle is shown (n = 15). Wild-type (WT) plants formed 54.7 siliques (100%). (B) Number of flowers and siliques (stage 13-18) of 5-week-old plants. The plants had started to flower at the same time. Staging of floral meristems was according to Smyth et al. (1990) (n = 5). (C) Irregular pattern of siliques on the main stem of the ckx3 ckx5 mutant (right) indicated by arrows compared with the wild type (left). Data represents mean value ± sd. *P < 0.01, calculated by Student’s t test. [See online article for color version of this figure.]

Figure 4.

Phenotype of the Inflorescence Stem of ckx3 ckx5 Mutants Compared with the Wild Type. (A) The inflorescence stem of ckx3 ckx5 mutants (right) is thicker than in the wild type (WT). The insets show a close-up of the inflorescence stems. (B) and (C) Hand-cut transverse sections of the inflorescence stem of the wild type (B) and the ckx3 ckx5 mutant (C) stained with toluidine blue. Sections were made at the base of the stem of 5-week-old plants.

Figure 5.

Flower Phenotype and Seed Yield of ckx Mutants. (A) Stage 13 flowers. From left to right: the wild type, ckx3, ckx5, and ckx3 ckx5. (B) Petal surface of ckx mutants, stage 14 flowers, 39 DAG (n = 30). WT, wild type. (C) Cell number per surface area in wild-type and ckx3 ckx5 mutant petals (n = 11). (D) The corresponding gynoecia of flowers shown in (A). From left to right: the wild type, ckx3, ckx5, and ckx3 ckx5. (E) Floral organ number per flower in the wild type, ckx3, ckx5, and ckx3 ckx5 (n = 50). (F) Young ovules of the wild type and ckx3 ckx5. Staging of ovules is according to Schneitz et al. (1995). Bars = 10 μm. (G) Number of ovules per gynoeceum (n = 12). (H) Seed yield of the wild type and ckx3 ckx5 under growth chamber conditions (n = 30). Data represents mean value ± sd. *P < 0.01, calculated by Student’s t test.

Figure 6.

CKX3 and CKX5 mRNA Expression Pattern in Inflorescence and Flower Tissues. RNA localization by in situ hybridization with CKX3 ([A] to [D]), WUS ([E] and [F]), and CKX5 ([G] to [O]) antisense probes hybridized to wild-type tissues. (A) to (G) and (I) to (K) show longitudinal sections through inflorescence meristems, floral meristems, and flowers. Staging of floral meristems is according to Smyth et al. (1990). Bars = 25 μm. (A) and (B) CKX3 is expressed in the center of the inflorescence meristem (A) and in the center of the floral meristem at stage 2 (B). (C) In stage 4 flowers, CKX3 is expressed in a broader central area of the floral meristem. (D) In stage 5 flowers, the CKX3 signal appears between long stamen primordia and gynoecia primordia. (E) and (F) WUS is expressed in the organizing center of the inflorescence (E) and floral meristems (F) during different developmental stages (a stage 4 flower is shown here). (G) to (I) CKX5 is expressed in the procambium of inflorescence stems ([G] and [H]) and flowers (I). (H) shows a transverse section through an inflorescence stem below the meristem. (J) CKX5 is expressed in stage 6 flowers between long stamen primordia and gynoecia primordia (arrowheads). (K) CKX5 is expressed in stage 7 flowers. (L) to (O) Transverse section through ovaries. CKX5 transcripts are detected in the developing placentas of stage 7 (L), stage 8 (M), stage 9 (N), and stage 11 (O) flowers.

Figure 7.

Detection of WUS mRNA Expression in the Inflorescence Meristem. (A) The wild type. (B) The ckx3 ckx5 mutant. Transcripts were identified by in situ hybridization. Bars = 25 μm.

Figure 8.

Enhanced Cytokinin Signaling Leads to Even More Active Inflorescence Meristems. (A) to (C) mRNA localization by in situ hybridization with AHP6 antisense probes hybridized to wild-type tissues. AHP6 is expressed in the inflorescence meristem at the position where the next flower primordia will arise and the youngest primordia (A). AHP6 transcripts can also be detected in floral organ primordia of stage 3 flowers (B) and in stage 6 flowers at the distal end of the developing gynoecium (C). (D) to (F) Scanning electron micrographs of inflorescence meristems of 4-week-old wild-type (D), ckx3 ckx5 (E), ahp6 (F), ckx3 ahp6 (G), ckx5 ahp6 (H), and ckx3 ckx5 ahp6 (I) plants. Floral buds are numbered from the first primordium of stage 2 to the first primordium stage 3. (J) ckx3 ckx5 ahp6 mutants form more siliques on the main stem than do ckx3 ckx5 mutants (n = 20). (K) The gynoecia of ckx3 ckx5 ahp6 triple mutants do not form more ovules than ckx3 ckx5 mutants do (n = 18). Bars = 20 μm ([A] to [C]) and 50 μm ([D] to [F]). Data represents mean value ± sd. Values of mutant lines were compared with the wild type. *P < 0.01, calculated by Student’s t test.

Figure 9.

Model of CKX and AHP6 Gene Function in Regulating Inflorescence Meristem Size and Activity. The model shows the action of CKX and AHP6 genes and cytokinin in concert with other known factors regulating shoot meristem size in Arabidopsis. The model integrates results of this article and other work described in the text. The genes studied in this work are shown in gray boxes.