The role of D3-type cyclins is related to cytokinin and the bHLH transcription factor SPATULA in Arabidopsis gynoecium development

Vol.:(0123456789) Planta (2024) 260:48 https://doi.org/10.1007/s00425-024-04481-4 ORIGINAL ARTICLE The role ofD3‑type cyclins isrelated tocytokinin andthebHLH transcription factor SPATULA inArabidopsis gynoecium development VincentE.Cerbantez‑Bueno1,2· JoannaSerwatowska1,3· CarolinaRodríguez‑Ramos1· J.ErikCruz‑Valderrama1,4· StefandeFolter1 Received: 21 February 2024 / Accepted: 2 July 2024 / Published online: 9 July 2024 © The Author(s) 2024 Abstract Main conclusion We studied the D3-type cyclin function during gynoecium development in Arabidopsis and how they are related to the hormone cytokinin and the transcription factor SPATULA. Abstract Growth throughout the life of plants is sustained by cell division and differentiation processes in meristematic tissues. In Arabidopsis, gynoecium development implies a multiphasic process where the tissues required for pollination, fertilization, and seed development form. The Carpel Margin Meristem (CMM) is a mass of undifferentiated cells that gives rise to the gynoecium internal tissues, such as septum, ovules, placenta, funiculus, transmitting tract, style, and stigma. Different genetic and hormonal factors, including cytokinin, control the CMM function. Cytokinin regulates the cell cycle transitions through the activation of cell cycle regulators as cyclin genes. D3-type cyclins are expressed in proliferative tissues, favoring the mitotic cell cycle over the endoreduplication. Though the role of cytokinin in CMM and gynoecium development is highly studied, its specific role in regulating the cell cycle in this tissue remains unclear. Additionally, despite extensive research on the relationship between CYCD3 genes and cytokinin, the regulatory mechanism that connects them remains elusive. Here, we found that D3-type cyclins are expressed in proliferative medial and lateral tissues. Conversely, the depletion of the three CYCD3 genes showed that they are not essential for gynoecium development. However, the addition of exogenous cytokinin showed that they could control the division/differentiation balance in gynoecium internal tissues and outgrowths. Finally, we found that SPATULA can be a mechanistic link between cytokinin and the D3-type cyclins. The data suggest that the role of D3-type cyclins in gynoecium development is related to the cytokinin response, and they might be activated by the transcription factor SPATULA. Keywords Carpel margin meristem (CMM)· Cell cycle· Cell division· CYCD3· Cyclins· Cytokinin· Differentiation· Gynoecium· SPATULA Abbreviations AbA Aureobasidin A CMM Carpel margin meristem SPT SPATULA Communicated by Dorothea Bartels. Vincent E. Cerbantez-Bueno and Joanna Serwatowska have contributed equally to this work. * Stefan de Folter [email protected] 1 Unidad de Genómica Avanzada (UGA-Langebio), Centro de Investigación y de Estudios Avanzados del Instituto Politécnico Nacional, 36824Irapuato, México 2 Present Address: Department ofBotany andPlant Sciences, University ofCalifornia Riverside, Riverside, CA92521, USA 3 Present Address: Departamento de Ingeniería Genética, Unidad Irapuato, Centro de Investigación y de Estudios Avanzados del Instituto Politécnico Nacional, 36824Irapuato, México 4 Present Address: Departamento de Biología Molecular de Plantas, Instituto de Biotecnología, Universidad Nacional Autónoma de México, Avenida Universidad 2001, Colonia Chamilpa, 62210Cuernavaca, Morelos, México Planta (2024) 260:4848 Page 2 of 16 Introduction The development of plant organs relies on cell division and differentiation in meristematic tissues (Gaillochet and Lohmann 2015). In Arabidopsis, the development of the gynoecium starts at the center of the floral meristem with the establishment of the gynoecium primordium at stage 6 of floral development (Smyth etal. 1990; Roeder and Yanofsky 2006; Alvarez-Buylla etal. 2010; Denay etal. 2017; Herrera-Ubaldo and de Folter 2022). By stage 7, the gynoecium primordium undergoes growth and division, giving rise to two distinct domains, referred to as medial and lateral domain (Bowman etal. 1999; Reyes-Olalde etal. 2013; Zúñiga-Mayo etal. 2019; Herrera-Ubaldo and de Folter 2022). These medial ridges subsequently merge to form a meristematic tissue known as the Carpel Margin Meristem (CMM). In the subsequent stages of development, CMM plays a pivotal role in generating various components, including the placenta, ovules, septum, transmitting tract, style, and stigma (Alvarez and Smyth 2002; Reyes-Olalde etal. 2013; Reyes-Olalde and de Folter 2019). The meristematic activity within the CMM has been characterized by the expression of specific genes and hormone activity, which regulate cell division and differentiation, akin to their functions in other meristematic tissues (Reyes-Olalde etal. 2013; Reyes-Olalde and de Folter 2019; Herrera-Ubaldo and de Folter 2022). In contrast to the medial region, the lateral region consists of 5–6 cell layers that will generate the valves (Bowman etal. 1999). The correct patterning and growth in the gynoecium axes are controlled by different genetic and hormonal factors (Moubayidin and Østergaard 2017; Reyes-Olalde etal. 2017, 2019; Herrera-Ubaldo and de Folter 2022). Cytokinin, a hormone derived from adenine, plays diverse roles in plant development (El-Showk etal. 2013; Mok and Mok 2001; Kieber and Schaller 2018; MárquezLópez etal. 2019; Wybouw and De Rybel 2019). Within the context of the gynoecium, cytokinin has been documented as a key regulator of gynoecium initiation and patterning (Marsch-Martinez etal. 2012; Zuñiga-Mayo etal. 2014, 2018; Marsch-Martínez and de Folter 2016; Müller etal. 2017; Reyes-Olalde etal. 2017; Gomez-Felipe etal. 2021; Carabelli etal. 2021; Herrera-Ubaldo and de Folter 2022). Moreover, cytokinin has been directly associated with genes involved in the development of various gynoecium tissues, including the CMM (Bartrina etal. 2011; Durán-Medina etal. 2017; Reyes-Olalde etal. 2017; Cuccinotta etal. 2020; Di Marzo etal. 2020; CerbantezBueno etal. 2020; Herrera-Ubaldo and de Folter 2022). For instance, the transcription factor SPATULA (SPT) has been shown to function as a positive regulator of stem cell proliferation through the stimulation of cytokinin signaling (Schuster etal. 2015; Gaillochet etal. 2017). In the context of the CMM, SPT plays a critical role in facilitating cytokinin signaling and fostering the development of associated structures, such as the septum and transmitting tract (Heisler etal. 2001; Alvarez and Smyth 2002; ReyesOlalde etal. 2017). The role of cytokinin in meristematic tissues has also been related to the regulation of the cell cycle, promoting cell division in some tissues (shoot) and differentiation in others (root) (Schaller etal. 2014). The cell cycle in plants is a highly conserved and regulated four-step process that briefly consists of the duplication of the genome (S phase) and the production of two daughter cells (M phase), separated by G1 and G2 phases, respectively (Harashima etal. 2013; Gutierrez 2016; Sablowski and Gutierrez 2022). These transitions of the cell cycle in plants, as in other eukaryotes, are coordinated by the cyclin-dependent kinase proteins (CDKs), and depend on cyclin subunits, which change at different stages of the cell cycle (Sablowski and Gutierrez 2022). It has been found that cytokinin regulates both G1/S and G2/M transitions (Schaller etal. 2014). The regulation of the G1/S transition occurs through the induction of D3-type cyclins (Riou-Khamlichi etal. 1999; Scofield etal. 2013; Schaller etal. 2014). However, despite substantial evidence supporting the role of cytokinin in regulating the G2/M transition, the exact mechanism remains unclear (Hare and van Staden 1997; Laureys etal. 1998; Lipavská etal. 2011; Francis 2011; Schaller etal. 2014). An increase in DNA content can occur through multiple rounds of duplication without intervening mitosis, a phenomenon known as endoreduplication (Breuer etal. 2010, 2014; Edgar etal. 2014). This process typically leads to an increase in cell size and eventual differentiation into specific cell types (Sugimoto-Shirasu and Roberts 2003; Dewitte etal. 2007; Lang and Schnittger 2020). Therefore, the role of cytokinin in controlling G1/S and G2/M transitions is crucial in determining the commitment to mitosis or endoreduplication processes. This differential regulation can influence whether cell division or differentiation predominates in different organs (Schaller etal. 2014). Some functions of D-type cyclins in plants have already been studied (Dewitte and Murray 2003; Dewitte etal. 2007). Among these, D3-type cyclins regulate the G1/S transition in the cell cycle by forming associations with CDK proteins to phosphorylate the retinoblastoma-related protein (RBR). This phosphorylation relieves the inhibition of the E2F transcription factor, thereby promoting entry into the S-cell cycle phase (de Jager etal. 2005; Schaller etal. 2014; Gutierrez and Sablowski 2022). In Arabidopsis, the D3-type cyclin group consists of three members: CYCD3;1, CYCD3;2, and CYCD3;3 (Vandepoele etal. 2002). Their expression has been found in proliferative tissues and they appear to control the type of cell cycle, favoring the mitotic Planta (2024) 260:48 Page 3 of 16 48 over the endoreduplication (Dewitte and Murray 2003; Dewitte etal. 2007; Scofield etal. 2013). The expression of D3-type cyclins is induced by endogenous signals, including plant hormones such as cytokinin (Riou-Khamlichi etal. 1999; Meijer and Murray 2000; Dewitte etal. 2007). Their depletion diminishes the ability of cytokinin to guide shoot formation. However, the overexpression of CYCD3;1 prompts shoot formation from calli independently of cytokinin, suggesting an autonomy from cytokinin signaling (Riou-Khamlichi etal. 1999). Although the role of CYCD3 genes in cell cycle progression has been reported as dispensable (Dewitte etal. 2007), they influence cell number, contribute to alternative cellular production and expansion processes, and mediate the effects of cytokinin. Despite extensive research on the relationship between CYCD3 genes and cytokinin, the regulatory mechanism that connects them remains elusive. Furthermore, the specific function of CYCD3 genes in the cell cycle within the context of gynoecium tissue development remains unexplored. This study focuses on elucidating the role of CYCD3 genes and their relationship with cytokinin in gynoecium development while attempting to unveil the mechanistic link that bridges cytokinin and D3-type cyclins. We found that the transcription factor SPATULA (SPT) in part is involved. Materials andmethods Plant lines andgrowth conditions All reporter lines (CYCD3;1::GUS, CYCD3;2::GUS, and CYCD3;3::GUS) and mutants (cycd3;1, cycd3;2, cycd3;3 single mutants, and cycd3;1–3 triple mutant) were obtained from James Murray and have been described in Dewitte etal. (2007). The spt-12 and 35S::SPT seeds were provided by Karen Hallyday. The lines are in the Col-0 background, except CYCD3;1::GUS, which is in the Ler background. All Arabidopsis and Nicotiana plants were grown in soil at ~ 23°C under long-day conditions (16h light/8h dark) in a greenhouse or a growth chamber. GUS analysis Gynoecia of different developmental stages were dissected and pre-fixed with cold acetone for 20min, then rinsed and transferred into GUS substrate solution: 50mM sodium phosphate pH 7, 5mM, K3/K4 FeCN, 0.1% (v/v) Triton X-100, and 2mM X-Gluc (Gold Biotechnology Inc.). After application of vacuum for 20min, the samples were incubated at 37°C for different times: ~ 1week for CYCD3;1::GUS, ~ 22h for CYCD3;2::GUS, and ~ 8h for CYCD3;3::GUS. Histological analysis Tissues were fixed in FAE (3.7% formaldehyde, 5% glacial acetic acid, and 50% ethanol, by vol.) with a vacuum for 15min at 4°C and then incubated for 60min at room temperature. The material was rinsed with 70% ethanol and incubated overnight at 4°C, followed by dehydration in a series of alcohol solutions (70, 85, 95, and 100% ethanol) for 60min each, and embedded in Technovit as previously described (Marsch-Martínez etal. 2014). Then, 8µm tissue sections were cut using low profile blades (Leica) on a Leica RM2035 microtome (Leica). Pictures were taken using a Leica DM600B microscope (Leica) with a DFC420C camera (Leica). Tissue staining Tissues were treated as described in histological analysis. Then, transmitting tract staining was performed as previously described (Zúñiga-Mayo etal. 2012). In summary, tissue sections were stained with a solution of 0.5% Alcian blue (pH 3.1; Sigma-Aldrich) for 25min and counterstained with a solution of 0.5% neutral red (Sigma-Aldrich) for 5min. Slides were rinsed in water, air dried, mounted, and observed in a Leica DM600B microscope. Photos were taken with a DFC420C camera (Leica). Cytokinin treatment Seeds were germinated and grown in soil under greenhouse conditions. One week after bolting, drops of BAP or mock solution were placed on the inflorescences once a day for 10days for phenotype analyses. In the case of gene expression analysis (RT-qPCR), drops were placed only once, and the tissue was collected after 2h. BAP solution contained 100µM 6-benzylaminopurine; 0.01% Silwet L-77 (Lehle Seeds), and the mock solution contained only 0.01% Silwet L-77 in distilled water. The treated and control plants were grown under the same conditions. Gene expression analysis For RT-qPCR analysis, floral buds in different stages were collected and total RNA was extracted using TRIzol (Invitrogen). The RNA was analyzed using an Open qPCR system (Chai Inc., Santa Clara, CA, USA) with qPCRBIO SyGreen 1-Step Go Hi-ROX (PCR Biosystems, Wayne, PA, USA) according to the manufacturer´s instructions. Three biological replicates and two technical replicates were done for each assay. Two or three replicates were analyzed using the 2−ΔΔCT method. Target gene expression levels were normalized to ACTIN2/7. Statistical analyses were made using t tests or one-way ANOVA and Tukey’s or Dunnet’s test as Planta (2024) 260:4848 Page 4 of 16 a post hoc for multiple comparisons. Data were analyzed and plotted using Prism (GraphPad). Primer sequences are listed in TableS1. Phenotype analyses Gynoecia and fruits from different plants of the wild type and mutant cycd3;1–3 lines were collected after the 10days of 6-benzyl amino purine (BAP) treatment. The gynoecia and fruits were evaluated in different phenotypical aspects: crests growth (ectopic tissue proliferation), apical–basal patterning, septum, transmitting tract, and ovule–funiculus phenotype. The identified effects were scored as the percentage of structures presenting mid-to-severe effects or slight-to-no effects in the analyzed tissues. Data analysis and plots were made using Prism (GraphPad). Y1H Two different fragments of the CYCD3;3 promoter (CYCD3;3 I, 1,540bp, and CYCD3;3 II, 1,629bp; see TableS1 for oligo sequences) were cloned in the pENTR/ D-TOPO vector (Invitrogen), verified by sequencing, and introduced into the CZN1018 vector (pAbAi + Gateway site; Danisman etal. 2012) by Gateway LR recombination. The SPT cDNA fused to the Gal4 activation domain in pDEST22, was previously described (Herrera-Ubaldo etal. 2023). Yeast transformations were performed as previously described (de Folter and Immink 2011) using the strains PJ69A and PJ69α for SPT, and CYCD3;3 promoters, respectively. Transformants were selected on SD-Gluc medium without tryptophan or uracil (SD-TRP or SD-URA). Mating was performed by droplets of each transformed yeast on YPAD medium, and the diploid selection was made on SD-TRP-URA plates. The final assay was done on SD-TRPURA complemented with 150ng/mL of Aureobasidin A (AbA; concentration obtained by a previous autoactivation assay). Yeast was grown at different dilutions at 30°C, and after 5days, interactions were scored. Haploid yeasts containing CYCD3;3 fragments were grown in medium with and without AbA as negative and positive controls. Luciferase assay For the luciferase assay constructs, a modified version of the pGreen II—0800—Luc was created. A 50bp 35S minimum promoter was created annealing two 50bp oligos that generated NotI 5′ and BamHI 3′ sticky ends (see TableS1 for sequences). Then, this 35S minimum promoter was cloned into the pGreen II—0800—Luc NotI 5′ and BamHI 3′ sites, resulting in the mini35SpGreenII—0800—Luc vector (see Fig.S1 for map). Each of the previously generated CYCD3;3 fragments in the pENTR/DTOPO (Invitrogen) vector were subcloned into the ApaI 5′ and EcoRv 3′ sites of the mini35SpGreenII—0800—Luc vector. The transient Luciferase expression assays were performed by the transient transformation of N. benthamiana leaves by Agrobacterium infiltration, performed as previously described (Espley etal. 2009) with minor modifications. In summary, an overnight culture of Agrobacterium was used to prepare an infiltration solution with 10mM MgCl2, 10mM MES, and 150µM acetosyringone (pH 5.6). Each Luciferase vector was co-infiltrated with the previously described 35S::SPT vector (Reyes-Olalde etal. 2017) into young and healthy N. benthamiana leaves. After 3days, small discs were cut out of the infiltrated leaves (3 discs of ~ 6mm diameter per sample) and homogenized in a 1X PBS solution. The liquid part of every sample was then combined in equal parts with the 2X TMCA solution to a final concentration of 100mM Tris HCl (pH 7.8), 5mM MgCl2, 250µM CoA, and 150µg/mL Luciferin-K (Gold BioTechnology Inc., St. Louis, MO, USA) in a 96-well plate. The plate was read in a Luminometer LmaxII384 (Molecular Devices) with each experiment in triplicate and using spaces without samples as blank references. At least three leaves at the same developmental stage were used for each combination, and the experiments were repeated at least two times. The results were analyzed and plotted using Prisma (GraphPad). Results D3‑type cyclin genes are differentially expressed duringgynoecium development The meristematic activity within the CMM has been inferred primarily from the expression of genes and hormone activity linked to cell division and differentiation (Reyes-Olalde etal. 2013; Reyes-Olalde and de Folter 2019; HerreraUbaldo and de Folter 2022). However, specific cell cyclerelated markers have not been definitively identified in this tissue. In the context of Arabidopsis, the D-type cyclins have been recognized for their role in governing entry into the S phase (Schaller etal. 2014), and their expression appears to be associated with actively proliferating tissues (Dewitte and Murray 2003; Dewitte etal. 2007; Menges etal. 2006). Among this cyclin group, the D3-type cyclins have been observed to be expressed in reproductive tissues (Collins etal. 2012; Dewitte and Murray 2003), and data available in databases confirm the gene expression of D3-type cyclins at various gynoecium developmental stages (Fig. S2). Nevertheless, the precise localization of these cyclins within this tissue has not been extensively characterized. Here, we studied the expression of the three Arabidopsis D3-type cyclins during gynoecium development using transcriptional Planta (2024) 260:48 Page 5 of 16 48 promoter GUS fusions (Dewitte etal. 2007). In general, while we obtained the expression patterns for CYCD3;2 and CYCD3;3 (Fig.1), we could not detect any GUS signal in the CYCD3;1::GUS gynoecia analysed (Fig.1a–f) due to the apparent low expression or absence of CYCD3;1 in the analysed tissues (Dewitte etal. 2007; Fig. S2). To discard any putative silencing of the CYCD3;1::GUS line, we performed GUS staining on seedlings and found GUS expression in the leaf tips, as reported before (Fig. S2; Dewitte etal. 2007). The signal of CYCD3;2::GUS was generally observed throughout gynoecium development, being detected in the vascular bundles, ovules, and valves (Fig.1g–l). There were slight variations in expression within different tissues across various developmental stages. Notably, the expression in vascular bundles persisted from the early to mature stages of gynoecium development. However, the GUS activity in the valves and ovules exhibited stage-specific differences. At the early stage (stage 8), high GUS activity was observed in the abaxial area of the valves and the medial vasculature bundles (Fig.1g). Subsequently, at stage 9, the activity in the valves and vascular bundles decreased, but it remained prominent in the abaxial area (Fig.1h). By stage 10, the GUS activity further decreased in the valves, but increased in the vascular bundles, and became noticeable in developing ovules (Fig.1i). CYCD3;2 activity continued to be present in the valves at stage 11, encompassing both the inner and outer layers (Fig.1j). This pattern persisted through stage 12, with a slight increase at this later stage (Fig.1k). In stages 11 and 12, the activity of CYCD3;2 was visible in ovules and vascular bundles too (Fig.1j, k). At stage 13, the expression in the inner cells of the valves was maintained, but the activity decreased in the outer layer of cells; at this stage, the activity in ovules vanished (Fig.1l). The expression of CYCD3;3 was observed in certain stages of gynoecium development. In general, its activity was concentrated in the tissues of the medial domain such as in the CMM, developing septum, funiculi, and the base of the ovule (Fig.1m–r). Additionally, some activity was detected in the valves during specific developmental stages. At stage 9, perhaps even at stage 8, faint GUS Fig. 1 Expression profile of the D3-type cyclin genes during gynoecium development. a–r Transverse sections of the transcriptional fusion reporters CYCD3;1::GUS (a–f), CYCD3;2::GUS (g–l), and CYCD3;3::GUS (m–r) in stages 8 to 13 (left to right) of gynoecium development. Arrowheads represent vasculature in development (g–l) and medial tissue in development (m–o). Graphical representation of gynoecium development and its tissues (s). The position of the transverse cuts is represented by arrowheads in (s). Scale bars = 50µm Planta (2024) 260:4848 Page 6 of 16 signal was visible in the CMM (Fig.1m, n). At stage 10, CYCD3;3::GUS activity was more clearly observed in the CMM/developing septa and in ovule primordia, with slight activity in the valves too (Fig.1o). By stage 11, CYCD3;3 expression intensified and remained prominent in the central regions of the medial tissues, in the funiculi, and certain layers of the septum. During this stage, there was also a mild presence of expression in the valves (Fig.1p). In stage 12, CYCD3;3 expression was prominently localized in the transmitting tract and funiculi, with a subtle presence in the septum and valves (Fig.1q). At stage 13, the CYCD3;3 expression was almost absent, only some faint expression in the funiculi (Fig.1r). In summary, distinct expression patterns were observed for each of the three D3-type cyclins, which generally align with information available in the database (Fig. S2). Specifically, CYCD3;2 was predominantly expressed in lateral tissues, whereas CYCD3;3 was more found in medial tissues. These different expression patterns suggest potential differences in the spatiotemporal functions of D3-type cyclins in the development of gynoecium tissues. D3‑type cyclin genes are induced bycytokinin inthegynoecium Cytokinin has been implicated in the regulation of the cell cycle, particularly in governing the G1/S and G2/M transitions (Schaller etal. 2014). Specifically, cytokinin-induced expression of the D3-type cyclin genes has been described in Arabidopsis seedlings as a mechanism controlling the G1/S transition (Riou-Khamlichi etal. 1999). To investigate whether D3-type cyclins are similarly regulated by cytokinin in the context of the gynoecium, we conducted experiments involving the application of BAP (6-benzyl amino purine) to various Arabidopsis lines. First, we isolated RNA from wild-type inflorescence tissue 2h after BAP application and examined its impact on the expression of the three D3-type cyclins by RT-qPCR. Notably, we observed a significant increase in the gene expression of CYCD3;1 (1.45 ± 0.12 Log2FC), CYCD3;2 (0.63 ± 0.18 Log2FC), and CYCD3;3 (0.73 ± 0.12 Log2FC), as illustrated in Fig.2a. This suggests that the regulation of these three genes by cytokinin occurs in floral tissues and can be inferred to also take place in the gynoecium. To further investigate this, we examined the promoter::GUS lines of the three D3-type cyclins after BAP application. As anticipated, following a 10-day BAP application, we observed the emergence of ectopic tissues originating from the replum, a phenomenon previously documented (Marsch-Martínez etal. 2012; Zúñiga-Mayo etal. 2014; Cerbantez-Bueno etal. 2020). Subsequent to the GUS staining and histological procedures, the CYCD3;1::GUS line maintained showing no signal at the gynoecium stage 12 or other stages analysed (Figs.2b, e and S3). This outcome suggests that the cytokinin-mediated induction observed in the RT-qPCR analysis of CYCD3;1 might be taking place in other tissues used for this experiment such as the meristem (Dewitte etal. 2007). In the case of CYCD3;2 and CYCD3;3, we did observe an increased signal in gynoecia, in line with the obtained RT-qPCR results (Fig.2c, d, f, g). Besides increased GUS signal intensity for CYCD3;2 in the valves and vascular bundles at stage 12, we observed seemingly additional vascular bundles marked by CYCD3;2 expression (Fig.2f). This effect was also present in other developmental stages (Fig. S3). These additional vascular bundles in the repla region corresponded with the locations of the ectopic outgrowths, where CYCD3;2 was similarly expressed in later fruit developmental stages (Fig. S4). The GUS signal of CYCD3;3 was also increased after exogenous cytokinin treatment, maintaining the same pattern in the medial domain at stage 12 (Fig.2g), and other stages analysed (Fig. S3). Nevertheless, the GUS signal was not observed in the ectopic outgrowths nor in the valves (Fig. S4). D3‑type cyclins role ingynoecium development The D-type cyclin group comprises ten genes classified into seven groups (Vandepoele etal. 2002). Their function has been associated with cell proliferation during the G1 phase (Meijer and Murray 2000). Among the D-type cyclins, the D1 and D3 types have been notably linked to reproductive development (Soni etal. 1995; Meijer and Murray 2000; Gaudin etal. 2000). Although D1-type cyclins were also observed to be expressed in floral tissues, though to a lesser extent (Fig. S2), here we focus on the role of D3-type cyclins in gynoecium development. To investigate this, we examined the phenotypes of the three CYCD3 single mutants and the cycd3;1–3 triple mutant. The cycd3;1, cycd3;2, and cycd3;3 single and cycd3;1–3 triple mutants have been previously described and recognized as null alleles (Dewitte etal. 2007). These mutants did not exhibit clear differences in inflorescence, flower, fruit, or external and internal gynoecium morphology in comparison to the wild type (Figs. S5 and S6). However, concerning the appearance of the cells in the gynoecium in the triple cycd3;1–3 mutant, they appear to be bigger, and less in number, at early stages of gynoecium development; clearly visible in the medial and lateral domains at stage 8 gynoecium (Figs.3c and S5). This phenotype of bigger cells without affecting organ size has already been described for other organs in this triple mutant (Dewitte etal. 2007) and supports the idea of yet a specific role for the D3-type cyclins in the gynoecium of Arabidopsis. This observation suggests that the D3-type cyclins might have redundant or partial redundant roles during gynoecium development, and support the notion that these D3-type Planta (2024) 260:48 Page 7 of 16 48 cyclins are not essential for the progression of the cell cycle, as previously proposed (Dewitte etal. 2007). D3‑type cyclins role related tocytokinin duringgynoecium development To obtain more insight into the function of the D3-type cyclins in relation to cytokinin in the gynoecium, we examined the effects of applying exogenous cytokinin in the triple D3-type cyclin mutant. Following a 10-day BAP application in wild-type and cycd3;1–3 triple mutant plants, we observed a phenotype in the inflorescence that exhibit overgrown gynoecia (larger) both in wild-type and cycd3;1–3 triple mutant background (Fig. S7). In addition, most of the wild-type gynoecia showed crest growth (Fig.3a, f), or so-called ectopic proliferative tissue from the repla, as previously described (Marsch-Martínez etal. 2012). However, in the cycd3;1–3 triple mutant, we observed no tissue proliferation in most of the gynoecia analysed (Fig.3b, f). Subsequently, we made transverse Fig. 2 D3-type genes are positively regulated by exogenous cytokinins. a RT-qPCR fold change values for the three D3-type genes after 2h of BAP treatment in inflorescences with respect to mock values. CYCD3;1 (1.45 ± 0.12SD Log2FC, n = 3), CYCD3;2 (0.63 ± 0.18SD Log2FC, n = 3), and CYCD3;3 (0.73 ± 0.12SD Log2FC, n = 3). Significant differences were determined by a t test considering P < 0.05. e–g Transverse sections of the transcriptional fusion reporters after 10-day BAP treatment: CYCD3;1::GUS (e), CYCD3;2::GUS (f), and CYCD3;3::GUS (g). b–d Transverse sections of the CYCD3 transcriptional fusion reporters without BAP treatment, corresponding with e, k, q in Fig.1. The position of the transverse cuts for stage 12 is represented by arrowheads in Fig.1s. Scale bars = 50µm Planta (2024) 260:4848 Page 8 of 16 sections of the treated gynoecia and fruits to gain deeper insights into the inner morphology. In general, the application of BAP induced the proliferation of ectopic tissues within the gynoecium, in the medial region and appeared to impact the balance between proliferation and differentiation in the cycd3;1–3 triple mutant. The cycd3;1–3 triple mutant gynoecia exhibited affected septum development, resulting in fruits with greatly altered septum structures (Fig.3e). Though we did observe mild defects in septum development in gynoecia and fruit in the wild-type (e.g., wider septum, bigger replum and in some cases a third septum primordium formation, ~ 65%, n = 60, Fig.3c, d), the effects in septum development were more severe and frequent in the cycd3;1–3 triple mutant line (Fig.3f, ~ 90%, n = 60). The impact on the transmitting tract led to a bifurcated structure in the gynoecium and a consequent aberrant arrangement of this tissue in the fruits of most of the cycd3;1–3 samples analysed (Fig.3c, f, 97%, n = 60), whereas an effect was observed in only some of the wild-type gynoecia and fruit samples (Fig.3c, f, 43%, n = 60). The development of the ovule and funiculus was also affected; a shorter ovule-funiculus primordia in mid-stages of gynoecium development Fig. 3 The role of the D3-type cyclins is related to cytokinin during gynoecium development. Phenotypes of gynoecia of WT (a) and cycd3;1–3 triple mutant (b) after 10-day treatment of BAP. c Transverse sections of the WT (top) and cycd3;1–3 triple mutant (bottom) during early (stage 8), mid (stage 10), and mature (stage 12) stages of gynoecium development after 10-day BAP or mock treatment. The position of the transverse cuts for each stage is represented by arrowheads in Fig.1s. Transverse sections of fruits (stage 15) of WT (d) and cycd3;1–3 triple mutant (e) after 10-day BAP treatment. f Percentage of different phenotypes found in the WT and cycd3;1–3 triple mutant after 10-day BAP treatment. Phenotypes were classified artificially in two groups: slight-to-no phenotype (Slight-No), and mid-to-severe phenotype (mid–severe). n = 60 for all lines. Scale bars = 1mm (a, b), 50µm (c–e) Planta (2024) 260:48 Page 9 of 16 48 marked a delay in the development of these structures in both the wild-type and the cycd3;1–3 mutant line after BAP application (Fig.3c), which led to a malformation of ovules and funiculi in the fruits (Fig.3d, e). Intriguingly, the cycd3;1–3 mutant displayed this phenotype more frequently and severely (Fig.3f, ~ 75%, n = 60) than the wildtype gynoecia and fruits analysed (Fig.3f, ~ 30%, n = 60). In summary, the cycd3;1–3 triple mutant resulted hypersensitive to cytokinin application. This suggests that the D3-type cyclins have a correlation with the cytokinin response. This could be associated with the correct patterning of the gynoecium’s internal tissues, which is simultaneously regulated by cytokinin (Reyes-Olalde and Folter 2019). Other genes related to this cytokinin response process, such as SPATULA (SPT) and B-type ARABIDOPSIS RESPONSE REGULATORS (ARRs), have been reported to alter inner tissue patterning when not functional (Reyes-Olalde etal. 2017). This suggests that the D3-type cyclins could have a relationship with cytokinin, and some other genetic factors involved in the cytokinin response. The transcription factor SPATULA regulates D3‑type cyclin gene expression SPATULA (SPT) is a bHLH transcription factor known to play a role in various aspects of plant development (Groszmann etal. 2010). Within the gynoecium, SPT is expressed in the medial tissues (Heisler etal. 2001; Groszmann etal. 2010). The function of SPT has been directly associated with cytokinin, as it facilitates the hormonal response in CMM development by activating the B-type ARR1 gene (Reyes-Olalde etal. 2017). The spt single mutant displays impairments in the development of medial tissues, such as the septum and transmitting tract (Alvarez and Smyth 1999; Heisler etal. 2001). In addition, the spt mutant, when treated with BAP, does not exhibit the characteristic cell proliferation originating from the replum, as is typically observed in the wild-type (Reyes-Olalde etal. 2017). A similar lack of response to BAP was observed in the cycd3;1–3 triple mutant (Fig.3). Furthermore, a portion of the expression of SPT overlaps with the expression of the CYCD3;3 gene in the CMM and its derived tissues (Fig.1). These areas are known to have cytokinin activity reported by a TCS::GFP line (Marsch-Martínez etal. 2012; Reyes-Olalde etal. 2017). In the following experiments, we studied a possible relationship between SPT and the CYCD3 genes. First, we performed expression analysis by RT-qPCR in perturbation experiments of SPT. For this, we extracted RNA from inflorescence tissue of SPT overexpression (SPTOE) and spt-12 mutant lines, followed by RT-qPCR. Our findings showed that the overexpression of SPT triggered an increase in the expression of two out of the three D3-type cyclins: CYCD3;1 (0.86 ± 0.007 Log2FC) and CYCD3;3 (0.79 ± 0.14 Log2FC) (Fig.4a). On the other hand, in the spt-12 mutant, the expression of two out of three D3-type cyclins was reduced: CYCD3;2 (−0.55 ± 0.16 Log2FC) and CYCD3;3 (−0.74 ± 0.09 Log2FC), relative to wild-type (Fig.4a). In essence, SPT positively regulated the expression of CYCD3;1, but this gene remained unaffected in the absence of SPT. We have mentioned that CYCD3;1::GUS was not found in the gynoecium, suggesting a regulatory role of SPT on this gene in meristematic tissues. Conversely, the scenario was different for CYCD3;2 expression, where SPT overexpression had no impact on CYCD3;2 expression, but the spt-12 mutant led to a decrease. For CYCD3;3, this gene exhibited both positive and negative responses in the SPTOE and spt-12 backgrounds, respectively. All three D3-type cyclins were upregulated by cytokinin (Fig.2a). To investigate whether SPT is essential for cytokinin-mediated regulation of the three different D3-type cyclins, we applied BAP to the spt-12 mutant and conducted RT-qPCR. After 2h of BAP application, the relative expression level of CYCD3;1 remained unchanged, compared to the values observed in both spt-12 and wild-type samples (Fig.4b). In contrast, the expression levels of CYCD3;2 and CYCD3;3 in the spt-12 + BAP sample exhibited alterations and reached levels similar to those in the wild-type (Fig.4b). This implies that CYCD3;1 relies on SPT for its regulation by cytokinin, but in the case of CYCD3;2 and CYCD3;3, SPT function can potentially be replaced by other factors in their regulation following BAP application. Thus, it appears that SPT plays distinct regulatory roles for each of the D3-type cyclins. Transcription factors often exert direct control over gene expression by binding to their promoters and thereby regulating transcription. Conventionally, SPT has been known to predominantly bind to specific motifs featuring a G-box sequence (CAC GTG ) (Girin etal. 2011; Reymond etal. 2012). We conducted an analysis of the upstream regions starting from the ATG of the CYCD3;1, CYCD3;2, and CYCD3;3 genes. In the regulatory regions examined for the CYCD3;1 and CYCD3;2 genes, we did not identify any G-boxes. However, in the regulatory region of the CYCD3;3 gene, we located two G-boxes at positions −443 and −912bp (Fig.4c). Despite the fact that SPT may not be indispensable for the regulation of CYCD3;3 after cytokinin application, there exists a strong connection: they exhibit co-expression, and changes in SPT expression consistently influence CYCD3;3. To assess whether the regulation of CYCD3;3 by SPT is a direct process, we carried out protein-DNA interaction assays. We amplified an approximately 1,500bp fragment encompassing the SPT binding motifs of the CYCD3;3 gene (CYCD3;3 I; Fig.5c) and employed it to test the interaction via Y1H and Luciferase transactivation assays. We amplified another 1,500bp fragment upstream of the other fragment Planta (2024) 260:4848 Page 16 of 16 exploring and analyzing large-scale biological data sets. PLoS ONE 2(8):e718. https:// doi. org/ 10. 1371/ journ al. phone. 00007 18 Wybouw B, De Rybel B (2019) Cytokinin—a developing story. Trends Plant Sci 24(2):177–185. https:// doi. org/ 10. 1016/j. tplan ts. 2018. 10. 012 Zúñiga-Mayo VM, Reyes-Olalde JI, Marsch-Martinez N, de Folter S (2014) Cytokinin treatments affect the apical–basal patterning of the Arabidopsis gynoecium and resemble the effects of polar auxin transport inhibition. Front Plant Sci 5:191. https:// doi. org/ 10. 3389/ fpls. 2014. 00191 Zuñiga-Mayo VM, Baños-Bayardo CR, Díaz-Ramírez D, MarschMartínez N, de Folter S (2018) Conserved and novel responses to cytokinin treatments during flower and fruit development in Brassica napus and Arabidopsis thaliana. Sci Rep 8(1):6836. https:// doi. org/ 10. 1038/ S4159801825017-3 Zúñiga-Mayo VM, Gómez-Felipe A, Herrera-Ubaldo H, de Folter S (2019) Gynoecium development: networks in Arabidopsis and beyond. J Exp Bot 70(5):1447–1460. https:// doi. org/ 10. 1093/ jxb/ erz026 Zúñiga-Mayo VM, Marsch-Martínez N, de Folter S (2012) JAIBA, a class-II HD-ZIP transcription factor involved in theregulation of meristematic activity, and important for correct gynoecium and fruit development in Arabidopsis. PlantJ 71(2):314–326. https:// doi. org/ 10. 1111/J. 1365313X. 2012. 04990.X Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.