Plant Molecular Biology

, Volume 78, Issue 3, pp 199–209

AINTEGUMENTA-LIKE6 regulates cellular differentiation in flowers

Authors

    • Department of Biological SciencesUniversity of South Carolina
  • Marcie Eaddy
    • Department of Biological SciencesUniversity of South Carolina
Article

DOI: 10.1007/s11103-011-9844-3

Cite this article as:
Krizek, B.A. & Eaddy, M. Plant Mol Biol (2012) 78: 199. doi:10.1007/s11103-011-9844-3

Abstract

During flower development, pluripotent stem cells within the floral meristem give rise to proliferative precursor cells whose progeny eventually acquire specialized functions within each floral organ. The regulatory mechanisms by which plant cells transition from a proliferating state to a differentiated state are not well characterized. Several members of the AINTEGUMENTA-LIKE/PLETHORA (AIL/PLT) transcription factor family, including AINTEGUMENTA (ANT) and AIL6/PLT3, are important regulators of cell proliferation in flowers. To further investigate the role of AIL6 during flower development, we have characterized transgenic plants in which the coding region of AIL6 was expressed under the control of the constitutive 35S promoter (35S:cAIL6). These plants display changes in floral organ size and morphology that are associated with alterations in the pattern and duration of cell divisions within developing organs. In addition, we find that very high levels of AIL6 expression inhibit cellular differentiation. In contrast, ant ail6 double mutants display premature differentiation of floral meristem cells. These results indicate that these two transcription factors regulate both proliferation and differentiation in flowers.

Keywords

ArabidopsisFlower developmentDifferentiationAP2/ERFAIL/PLTOrganogenesis

Introduction

During reproductive development in Arabidopsis thaliana, pluripotent stem cells within the floral meristem give rise to highly proliferative progeny cells which transition through various stages of differentiation before becoming fully differentiated and ceasing cell division. The exact mechanisms controlling the transition of cells within developing floral organ primordia from a proliferating state to a differentiated state and how they are integrated within the programs of floral organ identity specification are not well understood. Floral organ primordia adopt distinct fates in response to the activities of unique combinations of floral organ identity proteins [reviewed in (Irish 2010)]. These proteins appear to regulate the entire process of floral organogenesis, from the progressive elaboration of organ morphology to cell and tissue-type patterning [reviewed in (Sablowski 2010; Ito 2011)].

According to the ABCE model, four different classes of floral organ identity proteins act in different combinations to specify sepal, petal, stamen and carpel identity in whorls one to four of the flower, respectively [reviewed in (Krizek and Fletcher 2005)]. The combination of A and E gene activities specifies sepal identity, the combination of A, B and E activities specifies petal identity, the combination of B, C and E activities specifies stamen identity and the combination of C and E activities specifies carpel identity. With the exception of the class A protein APETALA2 (AP2), the remaining floral organ identity proteins [the class A protein APETALA1 (AP1), the class B proteins APETALA3 (AP3) and PISTILLATA (PI), the class C protein AGAMOUS (AG), and the class E SEPALATA (SEP) proteins] are MADS domain transcription factors. According to the quartet model, these MADS domain proteins form higher order protein complexes, likely tetramers, that bind to CC(A/T)6GG (CArG boxes) sequences in the promoters of target genes (Theissen 2001; Melzer and Theissen 2009; Melzer et al. 2009). The floral organ identity proteins regulate distinct genes at different times in flower development with identified targets including genes involved in hormonal pathways, growth control and patterning (Kaufmann et al. 2009, 2010).

A number of genes regulating floral organ growth in Arabidopsis have been identified with some genes primarily controlling cell proliferation and others cell elongation [reviewed in (Breuninger and Lenhard 2010)]. Promoters of organ growth that regulate cell division include AINTEGUMENTA (ANT) and AINTEGUMENTA-LIKE6/PLETHORA3 (AIL6/PLT3), two members of the AIL/PLT gene family that encode AP2/ERF type transcription factors. ant mutants produce smaller leaves and flowers than wild type and these organ growth defects are enhanced in ant ail6 double mutants (Elliott et al. 1996; Klucher et al. 1996; Krizek 2009). Both genes are expressed throughout the floral meristem and newly initiated floral organ primordia, but ANT expression persists longer than that of AIL6 in developing floral organs with little or no expression in mature organs (Elliott et al. 1996; Nole-Wilson et al. 2005). ANT has been proposed to regulate organ size by controlling the length of time in which cells within developing organs are competent to undergo proliferation (Mizukami and Fischer 2000). Constitutive expression of ANT under the 35S promoter resulted in persistent expression of the G1 cyclin CycD3 in maturing leaves, a prolonged period of cell division and larger organs (Mizukami and Fischer 2000).

Besides its role in regulating organ growth in the shoot, AIL6/PLT3 acts redundantly with PLT1 and PLT2 in root development (Galinha et al. 2007). plt1plt2 ail6 seedlings have a rootless phenotype with fully differentiated cells present in the root pole of these seedlings at 3 days post germination (Galinha et al. 2007). Four AIL/PLT proteins (PLT1, PLT2, AIL6 and BBM) are detected in gradients within the root with the highest levels present in the stem cell niche and lower levels in the proliferating cells above the meristem (Galinha et al. 2007). Low levels of PLT2 and PLT3 are also detected in cells of the elongation zone. This gradient of PLT activity has been proposed to be instructive for distinct cellular behaviors with high PLT activity specifying stem cell identity, intermediate levels specifying mitotic activity and low levels required for cellular differentiation (Galinha et al. 2007).

To further investigate the role of AIL6 during flower development, we expressed the coding region of AIL6 under the control of the constitutive 35S promoter (35S:cAIL6). These plants exhibit alterations in floral organ morphology that are associated with persistent cell divisions within some developing floral organs. The severity of these phenotypic alterations is correlated with AIL6 mRNA levels. In the highest expressing 35S:cAIL6 line characterized, we also observed negative effects on cellular differentiation. The characteristic surface morphologies of sepal, petal, stamen and carpel epidermal cells were absent in this 35S:cAIL6 line suggesting that cells did not undergo complete differentiation. Correspondingly, we observe that floral meristem cells undergo premature differentiation in ant ail6 flowers. These findings suggest that ANT and AIL6 regulate the transition of cells between proliferation and differentiation.

Materials and methods

Plant materials and growth conditions

Arabidopsis thaliana ecotype Landsberg erecta (Ler) was used as the wild type. Plants were grown on a soil mixture of Metro-Mix 360:perlite:vermiculite (5:1:1) under continuous light or in 16 h days (100–150 μmol/m2/s) at a temperature of 22°C.

Plasmid construction and Agrobacterium transformation

The coding region of AIL6 (cAIL6), corresponding to the most upstream in-frame ATG (Feng et al. 2005), was PCR amplified from cDNA and cloned into the KpnI/BamHI sites of pART7. The resulting 35S:cAIL6 construct was digested with NotI and cloned into pMLBart. The 35S:cAIL6 plasmid was transformed into Agrobacterium tumefaciens strain ASE by electroporation. Arabidopsis Ler plants were transformed with this Agrobacterium strain by vacuum infiltration (Bechtold et al. 1993). Transformants containing the 35S:cAIL6 transgene were selected using basta.

RNA extraction and real time RT-PCR

RNA was extracted from inflorescences (Verwoerd et al. 1989) and treated with Turbo DNase (Applied Biosystems) following the manufacturer’s instructions. First strand cDNA synthesis was performed using Quanta qScript cDNA Super Mix (Quanta BioSciences) following the manufacturer’s instructions. The real time PCR reactions were performed on an iCycler (Biorad) using B-R SYBR Green SuperMix for iQ (Quanta BioSciences) and the following primers for AIL6: RTAIL6-4: 5′-cgagttgctgggaacaaag-3′ and RTAIL6-5: 5′-tcatacgcttcagctgcttc-3′, which amplified a 71 bp fragment. Melt curve analyses were performed at the end of each experiment to check the specificity of the reactions. Experiments were performed twice using two or three biological tissue replicates and reactions in triplicate. The raw fluorescence data was analyzed using Miner, a software program that calculates Ct (threshold cycle) and amplification efficiencies (Zhao and Fernald 2005). Normalization was performed using two reference genes (At5g15710, At5g12240) that are expressed stably and at lower levels than traditional reference genes, since AIL6 expression is low (Czechowski et al. 2005). The fold difference in AIL6 expression was calculated using the Pfaffl method (Pfaffl 2001). Data reported in the figures correspond to those normalized to At5g12240 but similar values were obtained when normalizing to At5g15710.

SEM

Tissue for SEM was fixed, dried, dissected and coated as described previously (Krizek 1999). SEM analyses were performed on a FEI Quanta 200 ESEM.

Paraffin and epoxy sectioning

Tissue for paraffin sectioning was prepared and sectioned as described below for in situ hybridization. After sectioning, slides were placed in 0.05% Toluidine Blue O for 10 min, rinsed with water and allowed to air dry (Sakai 1973). The paraffin was removed with xylenes and slides mounted with permount. For embedding in epoxy, tissue was fixed in 0.2 M sodium cacodylate buffer for 2 h at room temperature, rinsed with 0.1 M cacodylate buffer, postfixed in 0.1 M cacodylate buffered 1% osmium tetroxide for 1 h at 4°C, rinsed with 0.1 M cacodylate buffer, dehydrated in ethanol and acetone, and incubated overnight in a 1:1 mixture of EMBed-812 (Electron Microscopy Sciences) and acetone. The next day the tissue was incubated for 5 h in a 3:1 mixture of EMBed-812 and acetone and then incubated overnight in 100% EMBed-812. The samples were embedded the next day in 100% EMBed-812 and cured for approximately 24 h. Samples were sectioned on an ultramicrotome and stained with Toluidine Blue O (Epoxy tissue stain; Electron Microscopy Sciences).

In situ hybridization

Inflorescences were fixed, embedded, sectioned, hybridized and washed as described previously (Krizek 1999). Digooxigenin-labeled antisense RNA probes (AIL6, histone H4) were synthesized as described previously (Krizek 1999; Nole-Wilson et al. 2005).

Results

35S:cAIL6 plants exhibit alterations in floral organ development

To better understand the role of AIL6 in plant development, the coding region of AIL6 was expressed under the control of the constitutive 35S promoter from the cauliflower mosaic virus. The majority of T1 plants (75 lines) exhibited alterations in flower development while 20 lines had a wild-type appearance. The alterations in flower development included changes in floral organ size and morphology (Fig. 1a). Most T1 lines were infertile, but we were able to obtain seeds from several lines. We chose two lines with phenotypes, lines 22 and 31, for further phenotypic characterization. These lines had partially overlapping but somewhat distinct phenotypes.
https://static-content.springer.com/image/art%3A10.1007%2Fs11103-011-9844-3/MediaObjects/11103_2011_9844_Fig1_HTML.gif
Fig. 1

AIL6 misexpression alters flower and seed development. a Flowers from wild type (Ler) and 35S:cAIL6 lines 31, 34, 30 and 22 showing phenotypic variation from primarily effects on organ size to dramatic changes in organ morphology (left to right). b Front (replum up) and side (replum on side) views of mature Ler and 35S:cAIL6 line 22 carpels. c Sections of paraffin-embedded tissue showing the apical region of mature Ler and 35S:cAIL6 line 22 carpels. Arrows point to the stylar region, which undergoes increased proliferation in 35S:cAIL6 line 22. Seeds from Ler (d), 35S:cAIL6 line 22 (e) and 35S:cAIL6 line 31 (f). AIL6 mRNA levels in Ler and 35S:cAIL6 (lines 1, 22, and 31) inflorescences. The expression level in Ler is set to one and error bars show standard deviation

Flowers of line 22 sometimes produce floral organs larger than wild type but more often display dramatic changes in organ morphology (Fig. 1a). The severity of these morphological changes varies somewhat between plants and between flowers on a single plant with the most variation observed in the sepal and petal phenotypes. In the most severe cases, the sepals are bent backwards, cupped and bumpy with a fringe-like appearance at their distal end (Fig. 2a, b). Because of their altered morphology, they do not fully enclose the floral bud. Petals are cupped and sometimes fringed at their distal end (Fig. 2c). The anthers of stamens are somewhat misshapen with less clearly defined locules, many of which do not release pollen (Fig. 2d). The floral organs that differ most dramatically from wild type are the carpels, which have large outgrowths in the stylar region of the organ (Fig. 1b). These carpel outgrowths are most dramatic in the medial domain (corresponding to the region of fusion of the two carpels) but smaller bumpy protrusions are also detected in the lateral region (Figs. 1b, 2e–h). Sections through the gynoecium show increased numbers of cells in the stylar region of the 35S:cAIL6 line 22 carpels compared to wild-type carpels (Fig. 1c). Flowers of line 31 produce larger floral organs that are generally similar in overall appearance to those of wild type (Fig. 1a). However, the sepals of these flowers are more cupped in shape than in wild type and the carpels have similar but somewhat less dramatic protrusions as described for line 22.
https://static-content.springer.com/image/art%3A10.1007%2Fs11103-011-9844-3/MediaObjects/11103_2011_9844_Fig2_HTML.jpg
Fig. 2

SEM analyses of 35S:cAIL6 line 22 floral organs. a Ler flower. b35S:cAIL6 sepal with a bumpy appearance. Arrow points to fringe present at the distal end of the sepal. c35S:cAIL6 petal with fringe at the distal end (arrow). d Ler (left) and 35S:cAIL6 (right) stamen anthers. e Front view (replum up) of Ler carpel. f Side view (replum to side) of Ler carpel. g Front view of 35S:cAIL6 carpel showing small bumpy protrusions below the stigma. h Side view of 35S:cAIL6 carpel showing large extensions beneath the stigma. i Ler flower of stage 10. Arrow points to the top of the carpel. j35S:cAIL6 flower of stage 10. Arrow points to the top of the carpel that has a normal appearance at this stage of development. k Ler flower of stage 12. Arrow points to style. l35S:cAIL6 flower of stage 12. Style (arrow) is not as distinctly separated from the ovary valves as in wild type. m Closeup view of side view of Ler carpel. np Developmental series showing the progressive enlargement of the carpel outgrowths of 35S:cAIL6 flowers. SE sepal, PE petal, ST stamen, CA carpel. Size bars are 100 μM (b), 200 μM in (c, d, i, j, mp), 400 μM in (a, eh) and 500 μM (k, l)

Some later-arising flowers in both 35S:cAIL6 line 22 and 31 have a more normal appearance and are fertile, allowing us to obtain some seeds from these lines. The seeds obtained are sometimes larger than normal with light brown or greenish seed coats that never take on the brown color of wild-type mature seeds (Fig. 1d–f). The seed phenotype, like the flower phenotype, is more severe in line 22 compared to line 31.

To determine whether the phenotypic differences between lines 22 and 31 might be a consequence of different steady-state levels of AIL6 mRNA, we characterized AIL6 expression in three 35S:cAIL6 lines (1, 22, 31) and wild type inflorescences by real-time RT-PCR. 35S:cAIL6 line 1 produces flowers with a wild-type appearance and accumulates similar amounts of AIL6 mRNA as wild-type inflorescences (Fig. 1g). In contrast, AIL6 mRNA levels were much higher in lines 22 and 31 as compared with wild type. AIL6 mRNA levels were approximately 55 fold higher in line 22 and approximately 30 fold higher in line 31 (Fig. 1g). These results suggest that the stronger floral and seed phenotypes of line 22 are a consequence of higher levels of AIL6 mRNA.

To better characterize the altered morphology of 35S:cAIL6 line 22 carpels, we examined these organs throughout flower development by scanning electron microscopy. In stage 11 and younger flowers, 35S:cAIL6 carpels appear normal in morphology (Fig. 2i, j). During stage 12 of wild-type flower development, stigmatic papillae appear at the apex of the carpel and the stylar region become clearly distinct from the stigma above and the ovary valves below (Fig. 2k). In 35S:cAIL6 flowers, the stylar region is tapered but not as well-defined and distinct from the ovary valve tissue (Fig. 2l). This continues to be true through opening of the floral bud in stage 13 flowers. Bulging from the stylar region begins to occur in 35S:cAIL6 line 22 carpels during stage 13 and continues with the bulges developing into larger outgrowths (Fig. 2m–p).

Dosage dependent effects of AIL6 misexpression on cellular differentiation in floral organs

Each of the four floral organs exhibits distinct and characteristic epidermal cell morphologies at maturity (Fig. 3a–d). The abaxial surface of sepals contains cells of variable size with characteristic cuticular thickenings (Fig. 3a). The adaxial petal surface consists of conical shaped cells with cuticular markings running down from the cell apex (Fig. 3b). Anther cells are interdigitated and also display characteristic cuticular thickenings (Fig. 3c) while ovary valve epidermal cells have specks of epicuticular wax on their surface (Fig. 3d). Epidermal cells present on 35S:cAIL6 line 22 floral organs lack these characteristic features of differentiated floral organs. The epidermal cells present on sepals, petals and stamens are smooth in appearance, lacking the cuticular thickenings (Fig. 3e–g), and epicuticular wax was not present on ovary valve cells (Fig. 3h). In addition cell size and shape was sometimes altered. No giant cells were present on the sepal abaxial epidermis (Fig. 3e). Some adaxial petal epidermal cells were more elongated (Fig. 3f) and anther cells were less interdigitated (Fig. 3g). The altered cell shape and epidermal characteristics were somewhat variable; ranging from some organs with normal epidermal characteristics to those with a few normal cells in a background of mostly cells with smooth appearances to organs that only contained organs with smooth cells. These cellular phenotypes were correlated with overall organ morphology; organs most dramatically affected in overall appearance displayed the most dramatic differences at the cellular level. No alterations in epidermal cell surface morphology or cell shape were ever observed in 35S:cAIL6 line 31 floral organs (Fig. 3i–l).
https://static-content.springer.com/image/art%3A10.1007%2Fs11103-011-9844-3/MediaObjects/11103_2011_9844_Fig3_HTML.jpg
Fig. 3

SEM analyses of epidermal surface of 35S:cAIL6 floral organs. Epidermal surface of Ler sepal (a), petal (b), anther (c) and ovary valve (d). Epidermal surface 35S:cAIL6 line 22 sepal (e), petal (f), anther (g) and ovary valve (h). These cells lack the characteristic cuticular markings present on the corresponding Ler cells and sometimes have an altered shape. Epidermal surface 35S:cAIL6 line 31 sepal (i), petal (j), anther (k) and carpel valve (l). These cells possess the characteristic cuticular markings present on the corresponding Ler cells. Size bars are 50 μM (al)

To further investigate whether AIL6 regulates cell differentiation, we examined epoxy-embedded sections of ant ail6 inflorescences since ail6 mutants have a wild-type appearance. The sections were stained with Toluidine Blue O, a polychromatic stain that differentially stains different cellular components (O’Brien et al. 1964; Sakai 1973). The inflorescence meristem and stage 1–3 flowers of Ler and ant ail6 inflorescences are similar in appearance with both inflorescence and floral meristems composed of small and densely staining cells filled with cytoplasm (Fig. 4a, d). Floral meristem cells in stage 5 wild-type flowers also exhibit dense staining (Fig. 4b), while cells in the floral meristem region of similar stage ant ail6 flowers are less densely staining with larger vacuoles (Fig. 4e). While young stamen and carpel primordia in wild-type flowers stain darkly (Fig. 4c), this is not true of primordia that arise in the center of ant ail6 flowers (Fig. 4f). Thus in the absence of ANT and AIL6 function, floral meristem cells undergo premature differentiation, which could be at least partly responsible for the reduced size of ant ail6 floral organs (Krizek 2009).
https://static-content.springer.com/image/art%3A10.1007%2Fs11103-011-9844-3/MediaObjects/11103_2011_9844_Fig4_HTML.jpg
Fig. 4

ant ail6 flowers show premature differentiation of floral meristem cells. Thin epoxy sections of Ler inflorescence (a), stage 5 flower (b) and stage 7 flower (c). The floral meristem (FM) consists of small densely stained cells. In c, stamen (st) and carpel (ca) primordia consist primarily of small densely stained cells. Thin sections of ant-4 ail6-2 inflorescence (d), stage 5 flower (e) and stage 6 flower (f). Cells in the floral meristem (e) or within organ primordia (f) are larger, less densely stained with large vacuoles compared to those in Ler. IM inflorescence meristem, FM floral meristem, ST stamen, CA carpel. Size bars correspond to 50 μM (af)

AIL6 overexpression does not affect leaf development

In addition to their altered appearance, many 35S:cAIL6 seeds exhibited delayed germination such that seedling size was quite variable 7 days after transfer of seeds on plates into the growth room (Fig. 5a–c). This phenotype was stronger in line 22 compared with line 31. AIL6 overexpression does not affect vegetative development as 35S:cAIL6 plants are normal in appearance approximately 17 days after germination (Fig. 5d–f) and bolt at the same time as wild type (Fig. 5g). Because of the normal appearance of these plants, we measured steady-state AIL6 mRNA levels in 20 day old shoots. Both 35S:cAIL6 lines 22 and 31 had greatly increased levels of AIL6 transcript with line 22 exhibiting almost 70 fold and line 31 approximately 50 fold more AIL6 mRNA than wild type (Fig. 5h). We also examined the surface morphology of rosette leaves and found no differences in the epidermal cellular characteristics of 35S:cAIL6 leaves line 22 or 31 as compared with Ler leaves (Fig. 5i–k). These results indicate that high levels of AIL6 mRNA do not affect vegetative development or the switch to reproductive development.
https://static-content.springer.com/image/art%3A10.1007%2Fs11103-011-9844-3/MediaObjects/11103_2011_9844_Fig5_HTML.gif
Fig. 5

AIL6 misexpression does not affect leaf development. Ler (a), 35S:cAIL6 line 22 (b) and 35S:cAIL6 line 31(c) seedlings 7 days after transfer of seeds on plates to growth room. 17 day old Ler (d), 35S:cAIL6 line 22 (e), and 35S:cAIL6 line 31(f) plants. The 35S:cAIL6 and Ler plants are of similar size at this stage of development. g 30 day old Ler, 35S:cAIL6 line 22, 35S:cAIL6 line 31 plants. The 35S:cAIL6 and Ler plants bolt at approximately the same time. hAIL6 mRNA levels in 20 day old Ler and 35S:cAIL6 (lines 22, and 31) plants. The expression level in Ler is set to one and error bars show standard deviation. Scanning electron micrographs of the surface of mature Ler (i), 35S:cAIL6 line 22 (j) and 35S:cAIL6 line 31 (k) leaves. Leaf epidermal cell morphology is similar in the 35S:cAIL6 lines and Ler. Size bars correspond to 50 μM (ik)

Persistent cell division is associated with the altered development of 35S:cAIL6 floral organs

To determine whether the altered size and/or morphology of 35S:cAIL6 floral organs were associated with changes in cell division, we examined the expression pattern of histone H4 by in situ hybridization. Several obvious differences in histone H4 expression were observed in 35S:cAIL6 lines. The number of histone H4 expressing cells was increased in 35S:cAIL6 line 31 stage 10 petals as compared with Ler petals of the same age indicating that more cells are dividing in the 35S:cAIL6 petals at this time in development (Fig. 6a, b). In stage 13 and older flowers, an increased number of histone H4 expressing cells were detected in the carpel valves and stylar region of 35S:cAIL6 line 22 flowers as compared with Ler (Fig. 6c–f). These results indicate that ectopic expression of AIL6 during later stages of flower development promotes cell division in some floral tissues.
https://static-content.springer.com/image/art%3A10.1007%2Fs11103-011-9844-3/MediaObjects/11103_2011_9844_Fig6_HTML.jpg
Fig. 6

Extended histone H4 expression in 35S:cAIL6 floral organs. aHistone H4 expression in a transverse section of a stage 12 Ler flower. The white arrow points to the gynoecium and the black arrow points to a petal. b Transverse section showing increased numbers of histone H4 expressing cells in the gynoecium and petals of a 35S:cAIL6 line 31 stage 12 flower. The white arrow points to the gynoecium and the black arrow points to a petal. c Transverse section of a stage 13 Ler flower. Few cells in the ovary valves (arrow) express histone H4 at this stage of development. d Transverse section of a stage 13 35S:cAIL6 line 22 flower showing histone H4 expression in many ovary valves cells (arrow). e Longitudinal section of a stage 13 Ler flower. No cells in the stylar region (arrow) express histone H4 at this stage of development. f Longitudinal section of a stage 13 35S:cAIL6 line 22 flower showing histone H4 expression in the stylar region (black arrow) and ovary valves (white arrow). Size bars correspond to 50 μM (af)

35S:cAIL6 lines exhibit uneven AIL6 expression

Some of the phenotypic effects of AIL6 overexpression, such as bumps on the surface of sepals and fringe on the distal end of sepals and petals, suggest uneven growth within developing floral organs. To determine whether this might be due to different effects of AIL6 activity in different cells or unequal accumulation of AIL6 mRNA, we examined AIL6 expression in 35:cAIL6 lines by in situ hybridization. In wild type inflorescences AIL6 mRNA is detected on the periphery of the inflorescence meristem and in young flowers (Fig. 7a). AIL6 mRNA is not detected in flowers older than stage 6 (Fig. 7d, h). In 35S:cAIL6 lines, we found that AIL6 mRNA did not accumulate to similar levels in all tissues even though it is expressed under the control of a constitutive promoter. Although AIL6 mRNA accumulates unevenly within developing flowers, these expression patterns were consistent among flowers of the same stage. AIL6 mRNA levels are high in the sepals and pedicels of stage 4–6 35S:cAIL6 flowers (Fig. 7b, c). Starting about stage 8, AIL6 mRNA was detected in petals and stamens in addition to the sepals and pedicels of 35S:cAIL6 flowers (Fig. 7e, f). Starting about stage 11, AIL6 mRNA was detected in sepals, petals, stamens and carpels in 35S:cAIL6 flowers and remained especially high in the ovary valves (Fig. 7g) and stylar region of mature 35S:cAIL6 line 22 carpels (Fig. 7i, j). AIL6 expression was generally stronger in stage 13 flowers of line 22 as compared with line 31. The one tissue that we observed somewhat variable AIL6 expression was sepals of stage 13 flowers. Stage 13 sepals with high levels of AIL6 mRNA were smaller in size than those with lower AIL6 expression.
https://static-content.springer.com/image/art%3A10.1007%2Fs11103-011-9844-3/MediaObjects/11103_2011_9844_Fig7_HTML.jpg
Fig. 7

AIL6 mRNA does not accumulate evenly in 35S:cAIL6 flowers. aAIL6 mRNA is detected in incipient and young floral primordia (arrows) in Ler inflorescences. AIL6 mRNA accumulates to high levels in some tissues of 35S:cAIL6 line 31 (b) and 22 (c) inflorescences. dAIL6 is expressed weakly in petal primordia (arrow) in stage 7 Ler flowers but is not expressed in later stages of flower development. eAIL6 mRNA accumulates to high levels in the sepals (arrows) of these stage six and eight 35S:cAIL6 line 31 flowers and in the stamens of the stage 8 flower. fAIL6 mRNA in sepals, petals (arrow) and stamen filaments of a stage ten 35S:cAIL6 line 22 flower. Note that the sepals are reduced in size and do not enclose the floral bud. g In stage 13 35S:cAIL6 line 22 flowers, AIL6 mRNA is detected in the sepals, petals and in the inner cell layers of the ovary valves. hAIL6 mRNA is not detected in stage 13 Ler carpels. i, j Strong AIL6 expression is detected in the ovary valves (black arrow) and stylar region (white arrow) of stage 13 35S:cAIL6 line 22 carpels. IM inflorescence meristem, SE sepal, PE petal, ST stamen, CA carpel. Size bars correspond to 50 μM (aj)

Discussion

AIL6 expression in developing floral organs affects organ size and morphology

Misexpression of AIL6 in flowers can result in increase growth of floral organs similar to what has been observed in plants in which ANT or AIL5 are constitutively expressed (Krizek 1999; Mizukami and Fischer 2000; Nole-Wilson et al. 2005). However in the case of 35S:cAIL6 plants, organ morphology is also affected, something not observed in 35S:ANT or 35S:AIL5 flowers. These changes in organ morphology include bumps or protrusions on sepals and carpels and fringe on the distal end of sepals and petals. The uneven growth and bulging observed in some 35S:cAIL6 floral organs may be a consequence of the uneven accumulation of AIL6 mRNA within developing floral organs. We observed overlap between cells exhibiting persistent histone H4 expression and AIL6 misexpression, suggesting that AIL6 activity can promote cell proliferation. Patchiness in the distribution of AIL6 mRNA could lead to uncoordinated cell division within an organ and consequently affect organ morphology.

The different consequences of ANT and AIL6 overexpression suggest that these genes have somewhat distinct functions during flower development. This is consistent with the different expression patterns of these genes; ANT expression is maintained in developing floral organs much longer than AIL6 (Elliott et al. 1996; Nole-Wilson et al. 2005). In addition, previous work has revealed other differences between ANT and AIL6 suggesting that AIL6 does not just act similarly to ANT in a subset of ANT functions within the shoot. For example, ail6 mutants are more sensitive to the effects of the auxin transport inhibitor NPA on floral meristem initiation than ant mutants (Krizek 2011).

High AIL6 mRNA levels are sufficient to inhibit cell differentiation

We observed somewhat distinct phenotypes in two 35S:cAIL6 lines that accumulated high levels of AIL6 mRNA. Floral organs of 35S:cAIL6 line 31, which had 30 fold higher AIL6 mRNA levels as compared with wild type, were increased in size but had fairly normal morphology of sepals, petals and stamens. Cell differentiation occurred normally in this line with epidermal cells of floral organs displaying characteristics of fully differentiated cells. In contrast, floral organs of 35S:cAIL6 line 22, which had 55 fold higher AIL6 mRNA levels as compared with wild type, exhibited changes in organ morphology and incomplete cell differentiation. Our results suggest that high levels of AIL6 activity in developing floral organs promotes cell proliferation while extremely high levels of AIL6 activity can inhibit cell differentiation. These results suggest that AIL6 regulates both cell proliferation and differentiation and that it can affect cell behaviors in flowers in a dose-dependent manner. The observed premature differentiation of floral meristem cells in ant ail6 flowers further supports roles for AIL6 and ANT in inhibition of cellular differentiation during early stages of flower development. These functions are consistent with the expression pattern of these genes. While ANT expression persists longer than AIL6 in developing floral organs, neither is expressed in mature organs (Elliott et al. 1996; Nole-Wilson et al. 2005). We hypothesize that changes in the combined AIL6 and ANT activity level regulate the transitioning of cells from a proliferating state toward complete differentiation.

Although AIL6 mRNA levels in 35S:cAIL6 lines 22 and 31 were higher in leaves as compared with inflorescences, no effect on organ size, morphology or cellular differentiation was observed in the leaves of these plants. This indicates that high AIL6 activity is not sufficient to promote cell proliferation or inhibit cell differentiation in leaves.

Previous work has suggested that four AIL/PLT proteins (PLT1, PLT2, AIL6/PLT3 and BBM) act in a dosage dependent manner in the root with different levels of overall PLT activity specifying distinct cell behaviors along the longitudinal axis of the root (Galinha et al. 2007). Manipulation of this gradient by misexpression of PLT2 under the control of the 35S promoter resulted in a larger meristem due to continued division of cells that normally would have stopped proliferating and started elongating (Galinha et al. 2007). Our work is consistent with AIL/PLT proteins acting in a similar dose-dependent fashion within flowers with different AIL/PLT activity levels associated with distinct cellular behaviors.

AIL6 expression under the 35S promoter led to uneven accumulation of AIL6 mRNA

Surprisingly, we found that AIL6 does not accumulate to high levels in all tissues of 35S:cAIL6 inflorescences even though it is expressed under a constitutive promoter. It is possible that the 35S promoter used in this experiment does not confer constitutive or nearly constitutive expression. Although the pART7 plasmid containing the 35S promoter has been commonly used to misexpress genes, we were unable to find published data showing tissue-specific expression resulting from the use of this promoter. Thus, it is possible that uneven AIL6 accumulation results from uneven expression of the transgene. Alternatively, AIL6 may be subject to post-transcriptional regulation. Further work examining the expression of other genes expressed under the control of this 35S promoter is needed to distinguish between these possibilities.

Acknowledgments

We thank Soumitra Ghoshroy and the Electron Microscopy Center staff for advice on the use of the SEM and the ABRC for the F12B17 BAC clone. This work was supported by National Science Foundation (NSF) grant IOS 0922367.

Copyright information

© Springer Science+Business Media B.V. 2011