Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Evolution and expression of LEAFY genes in ferns and lycophytes

Evolution and expression of LEAFY genes in ferns and lycophytes Background: The LEAFY (LFY) transcription factors are present in algae and across land plants. The available expres‑ sion and functional data of these genes in embryophytes suggest that LFY genes control a plethora of processes including the first zygotic cell division in bryophytes, shoot cell divisions of the gametophyte and sporophyte in ferns, cone differentiation in gymnosperms and floral meristem identity in flowering plants. However, their putative plesio ‑ morphic role in plant reproductive transition in vascular plants remains untested. Results: We perform Maximum Likelihood (ML) phylogenetic analyses for the LFY gene lineage in embryophytes with expanded sampling in lycophytes and ferns. We recover the previously identified seed plant duplication that results in LEAFY and NEEDLY paralogs. In addition, we recover multiple species‑specific duplications in ferns and lycophytes and large‑scale duplications possibly correlated with the occurrence of whole genome duplication ( WGD) events in Equisetales and Salviniales. To test putative roles in diverse ferns and lycophytes we perform LFY expression analyses in Adiantum raddianum, Equisetum giganteum and Selaginella moellendorffii. Our results show that LFY genes are active in vegetative and reproductive tissues, with higher expression in early fertile developmental stages and dur‑ ing sporangia differentiation. Conclusions: Our data point to previously unrecognized roles of LFY genes in sporangia differentiation in lycophytes and ferns and suggests that functions linked to reproductive structure development are not exclusive to seed plant LFY homologs. Keywords: LEAFY, Ferns, Lycophytes, Reproductive transition, Sporogenesis Background clades NEEDLY (NDLY) and LFY. Both copies have been The LEAFY (LFY) gene lineage is a plant specific tran - retained in gymnosperms but NDLY was lost in angio- scription factor family [1]. LFY is best known for its role sperms [5, 6]. In addition, some species-specific duplica - in the flowering plant model species Arabidopsis thaliana tions have been identified in many lineages of land plants (Arabidopsis), where it is a key integrator for flowering [5, 7]. transition, controlling both the floral meristem identity LFY genes encode a ca. 220–350 amino acid pro- and the activation of the floral organ identity genes [1–4]. tein that acts as a homodimer and it has two recogniz- LFY genes are primarily retained as a single copy gene able domains: the N-terminal (also known as LFY_SAM) in algae and across land plants, with only a large-scale and the C-terminal (also known as C_LFY_FLO) DNA duplication occurring in seed plants resulting in two binding domains [8]. The LFY_SAM domain is key for dimerization, DNA binding and DNA accessibility, while the C_LFY_FLO domain has been proposed to medi- *Correspondence: lucia.pabon@udea.edu.co ate DNA binding specificity that has evolved across land Facultad de Ciencias Exactas y Naturales, Instituto de Biología, plants [5, 9]. Based on the dominant amino acids, three Universidad de Antioquia, Medellín, Colombia different motif types have been identified in the DNA Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Rodríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 2 of 16 binding domain in different streptophyte lineages: liver - and circinate fertile leaves, both characterized by active worts + tracheophytes have the type I motif, mosses have cell division [7, 39]. In addition, heterologous expression the type II motif, and algae possess the type III motif [5]. of fern LFY homologs does not fully recover wild type Finally, hornworts have been reported to have a pro- phenotypes in lfy Arabidopsis mutants [40]. Conversely, miscuous motif with versatile DNA binding capabilities endogenous loss-of-function double lfy mutants in C. binding all three types of DNA motifs [5]. richardii exhibit defects in gametophyte development Functional studies of LFY homologs in other flowering by cell division arrest of the apical cell, as well as trun- plants, such as Antirrhinum majus (snapdragon), showed cate sporophyte development accompanied by abnormal that floricaula (flo) mutants, do not transition from veg - leaf development [39, 40]. However, because lfy mutants etative to reproductive stages. In these mutants, inflores - result in interrupted shoot development, it is unclear if cence meristems do not form floral meristems on their LFY homologs play roles during the reproductive transi- flanks, resulting in branched stems forming more leaves tion in ferns. instead of producing flowers, similar to the Arabidopsis Studies in lycophytes are restricted to Isoetes, where lfy mutants [1, 10–12]. Conversely, the overexpression of two LFY homologs were identified. The Isoetes LFY LFY homologs turns the indeterminate inflorescence into homologs are also expressed in the vegetative and repro- a determinate meristem forming solitary flowers [13]. In ductive tissues of the sporophyte with higher expression Arabidopsis, LFY is upregulated by flowering time genes in juvenile tissues and fertile leaves bearing megaspo- including AGAMOUS-LIKE 24 (AGL24), SHORT VEG- rangia and microsporangia [41]. Importantly, heterolo- ETATIVE PHASE (SVP) and SUPRESSOR OF OVEREX- gous expression of Isoetes LFY genes in Arabidopsis did PRESSION OF CONSTANS (SOC1) [14–17]. Once LFY is not rescue lfy mutant phenotypes [41]. Finally, the role active in the floral meristem, it stimulates cytokinin sign - of LFY homologs has also been evaluated in bryophytes, aling and activates floral fate acquisition together with in the model Physcomitrium patens. This species has two APETALA1 (AP1) (a MADS-box homolog) responsible paralogs, PpLFY1 and PpLFY2 regulate the first division for sepal and petal identity [18–22]. LFY can also activate of the zygote and in turn controls proper cell division of other floral organ identity MADS-box genes including the sporophyte as seen in the double mutant with devel- SEPALLATA (SEP), PISTILLATA (PI), APETALA3 (AP3) opmentally arrested zygotes [42]. and AGAMOUS (AG) [23]. The data available for land plant LFY homologs was Based on functional data from Arabidopsis and snap- summarized by Plackett et  al. [39] who concluded that dragon, LFY has been characterized as a critical fac- plesiomorphic roles of LFY genes include the control tor controlling floral meristem identity and floral fate. of cell divisions in the zygote to form the sporophyte in These roles are conserved in other angiosperms, such as bryophytes, and that LFY homologs were subsequently apple, papaya and eucalyptus [24–26]. In addition, other co-opted to maintain indeterminate cell fate in both the roles have been reported for LFY members across angio- gametophyte and the sporophyte in ferns. These authors sperms. For instance, LFY homologs have been recruited also hypothesized that, only in seed plants, LFY genes in the formation of compound leaves in legumes [27, 28], were active during cone development in gymnosperms in shoot apical meristem (SAM) development in cucum- and controlled floral meristem identity in angiosperms. ber, impatiens, and tomato [29–32], in inflorescence Interestingly, due to their early critical roles in shoot branch patterning in rice [33], and in floral merism and development it is still unclear if fern LFY homologs play spiral arrangement of floral parts in columbines [34]. any function in the formation of fertile leaves and their Studies of gene homologs in gymnosperms show that specialization for spore production in ferns. If so, the both LFY and NDLY are expressed constitutively in LFY gene function in reproductive transition and spo- reproductive and vegetative structures, and spatial–tem- rangia development is not exclusive to seed plants. In this poral expression differences between paralogs have been context, we aim to reconstruct the evolution of the LFY linked to male and female cone differentiation [35–38]. In gene lineage with ample sampling from lycophytes and addition, heterologous expression of gymnosperm LFY ferns, and to assess expression patterns of LFY genes in homologs can rescue the wild type phenotype from lfy lycophytes and ferns during their reproductive transition. mutants in Arabidopsis [37], which suggest functional We make use of public databases as well as our own tran- conservation in the transition to reproduction across scriptomes from four fern species with different degrees seed plants. of leaf dimorphism (i.e., morphological differences Expression and function of two LFY paralogs have also between fertile and sterile leaves): the monomorphic (i.e., been studied in the fern Ceratopteris richardii. RT-PCR sterile and fertile leaves with same morphology) Adian- assays and in  situ hybridization showed high expres- tum raddianum, the hemidimorphic (i.e., sterile and fer- sion of the two C. richardii LFY copies in the shoot tips tile leaves with slightly different morphologies having for R odríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 3 of 16 instance differentiated fertile pinnae) Anemia villosa, and The evolution of LFY gene homologs within ferns and two holodimorphic (i.e., sterile and fertile leaves with dif- lycophytes, closely match the phylogeny of those lineages ferent morphology) species, Equisetum giganteum and [43]. However, our analysis recovered large-scale duplica- E. bogotense. The expression of LFY copies was assessed tions of LFY in Equisetales, in Salviniales, within Pteri- by RT-PCR in selected ferns with the extreme changes dineae in the Vittarioideae subfamily, and in Eupolypods in leaf dimorphism (Adiantum raddianum and Equise- II (sensu, [43]). However, the exact timing of the latter tum giganteum), and by in situ hybridization (ISH) in the two duplications cannot be placed with certainty. Unlike lycophyte Selaginella moellendoriffi . Our results allow their expected phylogenetic placements, our analyses us to propose a putative role for lycophyte and fern LFY recovered Cibotium glaucum LFY homolog nested within homologs in reproductive tissue formation (i.e., sporan- the Eupolypod I sequences and the Polypodium plecto- gia development), indicating that this function may have lens homolog within Eupolypod II sequences. In the spe- been present in the LFY gene lineage prior to the diversi- cies phylogenies these species belong to Cyatheales and fication of seed plants. Eupolypoids I, respectively [43]. Finally, our analysis reveals species-specific duplica - Results tions in the lycophyte Huperzia selago, a well-known LEAFY gene family evolution polyploid, and in a number of ferns, most of which have We present here the most comprehensive phylogenetic been identified as tetraploids or recent hybrids, includ - analysis of LFY genes to date, consisting of 228 sequences ing Anemia villosa, Asplenium platyneuron, Botrypus that include 95 from Sayou et  al. [5] and 133 sequences virginianus, Ceratopteris richardii, Cystopteris fragilis, from an expanded sampling targeting ferns and lyco- Dennstaedtia sp., Equisetum arvense, E. bogotense, Leu- phytes (Additional file  1: Fig. S1). The complete aligned costegia immersa, Lindsaea microphylla, Ophioglossum data set consists of 1338 characters from which 833 are vulgatum, Polypodium hesperium, P. plectolens, Psilo- informative. tum nudum, Pityrogramma trifoliata and Sceptridium Similar to previous reports, our analysis recovered LFY dissectum. genes predominantly as single copy genes in streptophyte algae and land plants, with a single exception in seed LFY protein domains and motifs plants. In turn, the gene tree mostly recovers the phylo- LFY proteins are characterized by two conserved genetic relationships recorded for major plant lineages. domains [5, 6, 9], the LFY_SAM (in the N terminus), The bryophyte LFY genes are sister to LFY homologs which in our analysis corresponds to motifs 1, 2 and 3 in vascular plants (Bootstrap Support, BS = 72) (Addi- and the C_LFY_FLO DNA binding domain (in the C ter- tional file  1: Fig. S1). However, within vascular plants, minus) which in our analysis corresponds to motifs 4 and sequences from ferns and lycophytes are recovered as 6 [6] (Fig. 2). We do not find outstanding variation at the sister to each other (BS = 61) and the lycophyte + fern protein level between divergent plant lineages (i.e., lyco- LFY clade is sister to the seed plant LFY homologs with phytes, ferns, gymnosperms and angiosperms), but we low support (Fig.  1). This is inconsistent with the vascu - were able to identify motifs 9 and 10 as exclusive to fern lar plant phylogeny accepted to date, where lycophytes protein sequences (Fig. 2). are sister to euphyllophytes (the clade formed by ferns In addition, the protein alignment does allow the iden- and seed plants) [43]. In addition, our topology recovers tification of some conserved amino acids in fern LFY the LFY and NDLY duplication event prior to the diver- proteins in the most variable region between the SAM_ sification of seed plants with a loss of NDLY homologs LFY and the C_LFY_FLO DNA binding domain (Addi- in angiosperms [5, 38, 40]. Moreover, we found multiple tional file  2: Fig. S2). Finally, following the key amino fern and lycophyte species-specific duplications some of acids positions for DNA binding, our analyses recovered which could explain the multiple homologs reported pre- the amino acids HRH, matching the type I described for viously in the fern Ceratopteris richardii and in several tracheophytes [5]. They correspond to positions 332, 370 Isoetes spp. in the lycophytes [39, 41]. and 413 in our alignment (Fig. 3). (See figure on next page.) Fig. 1 ML analysis of the LFY transcription factor family a Summary tree including sequences from algae, bryophytes and tracheophytes with Boostrap values (BS) as numbers on top of branches for major clades. b Detail of lycophyte and fern LFY clades. Yellow stars indicate large scale (major) duplication events. Number on each node indicate the BS. Black arrowheads point to sequences isolated in this study. The colors correspond to the conventions on the bottom left. Scale: 0.3 Rodríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 4 of 16 Fig. 1 (See legend on previous page.) R odríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 5 of 16 Fig. 2 a Conserved motifs of LFY proteins identified by a MEME analysis, their motifs, numbers assigned and their respective sequence. b Map of the motif positions in land plant representative proteins RT‑PCR expression of LFY homologs in selected ferns linked to gametophyte-to-sporophyte transitions already To hypothesize the roles of LEAFY genes during the reported [39], we evaluated the expression of all LFY cop- reproductive transition in ferns, in addition to those ies in the ferns Adiantum raddianum and Equisetum Rodríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 6 of 16 Fig. 3 Detail of the LFY Protein domains alignment for 34 representative land plant sequences belonging to: Arabidopsis thaliana, Ceratopteris richardii, Azolla filiculoides, Salvinia cucullata, Equisetum giganteum, E. bogotense, Adiantum raddianum, Anemia villosa, Selaginella moellendorffi, Physcomitrium patens and several algae species. The two characteristic domains of LFY proteins reported by Sayou et al. [5, 9] are boxed. Blue arrowheads point to the key positions for DNA binding reported by Sayou et al. [5] R odríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 7 of 16 Fig. 4 Developmental stages of the fern species Adiantum raddianum and Equisetum giganteum. (a–e) A. raddianum stages: Two sterile stages include Fd: fiddlehead and S: sterile expanded pinnae (a, b). Three fertile stages include: F0 leaves with unrolling pinnae and an indusium of ca. 250 µm (c), F1 Nearly fully expanded leaf and an indusium of ca. 800 µm (d), and F2 Fully expanded leaves and an indusium of ca. 2 mm. (e). Fi = False indusia, S: Sterile stage, F: Fertile stage. (f–o) Developmental stages of E. giganteum. SAMs producing photosynthetic leaves only in as thin and green at sterile stage S0 (f) and thick and brownish‑ green at sterile stage S1 (g) that vary in meristem size (k, l). Stage F0 has fully differentiated peltate fertile leaves (sporangiophores), that are covered and protected by the sterile leaf sheath and the spore mother cells have not undergone meiosis (h, m); Stage F1 is characterized by the exposure of half of the strobili above the sterile leaves, of ca. 1 cm and carry mature spores (i, n). Stage F2 has a fully exposed strobili with sterile leaves only at its base of ca 4 mm and mature spores and expanded elaters (j, o). Scale bars for c–e = 250 µm; f–j = 1 cm; Scale bars k–o = 2 mm. Scale bars in magnifications = 50 µm. Black arrowheads indicate the SAM, black arrowhead in insets of n, o indicate the elaters. Sl = Sterile leaves, Pl = Peltate spore bearing leaves, Sti = sporogenous tissue, S = Spores giganteum. These two genes represent extremes in the tissue protecting the sori) or sori (i.e., the clusters of leaf dimorphism gradient, as A. raddianum is monomor- sporangia) (Fig.  4a, b). In addition, three fertile devel- phic (i.e., there are no gross morphological differences in opmental stages were established, namely, Stage Fer- sterile and fertile leaves) and E. giganteum is holodimor- tile F0, F1 and F2, which considered the degree of leaf phic (i.e., sterile and fertile leaves vary dramatically in expansion and the indusia size (Fig.  4c–e). F0 is char- size, shape and position (Fig. 4). acterized by leaves with expanding pinnae and an indu- In A. raddianum sterile leaves can be separated into sium width of ca. 250  µm (Fig.  4c). F1 is characterized fiddleheads (Fd) and young sterile leaves (S), lack - by almost entirely extended leaves and an indusium ing any sign of indusia (i.e., the membranous covering width of ca. 800  µm (Fig.  4d). F2 is characterized by Rodríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 8 of 16 fully extended leaves and an indusium width of 2  mm. level of expression at the F2 stage. On the other hand, (Fig. 4e). Sporangia developmental stages were not spe- in the holodimorphic E. giganteum we found three cifically described for each stage as sporangia develop - LF Y paralogs with different expression patterns during ment is asynchronous, meaning that different sporangia reproductive transition. EqgiLF Y2 was undetected in are in different developmental stages in a single sori in the stages and tissues studied. EqgiLF Y1 is more highly any given time (Fig. 4a–e). expressed in fertile than sterile stages with two peaks In E. giganteum, we identified two different develop - of expression, one at F0 and the second at F2 (Fig.  5b). mental stages for sterile stems (Fig.  4f, g) and three of Finally, EqgiLF Y3 is detected in sterile and until early reproductive meristems (Fig.  4h–j). Sterile stages cor- fertile stages at F0, with very low levels of expression respond to shoot apical meristems (SAMs) producing detected in the two latter fertile (F1 and F2) stages. leaves only (Fig.  4k, l). These were found in two forms, thin and green (Fig.  4f) or thick and brownish-green (Fig.  4g). Although, anatomical sections revealed dif- In situ hybridization expression of LFY homologs ferences in meristem width, the two forms exclusively in the lycophyte Selaginella moellendorffii produce identical leaves (Fig.  4k, l). The identification To determine the detailed spatio-temporal expression of developmental stages in the fertile stems relied on patterns of the only LF Y gene identified in the het - changes of strobilus size, as well as sporangia and spore/ erosporous lycophyte S. moellendorffii, we performed elater development (Fig.  4h–j, m–o). Stage Fertile 0 (F0) in  situ hybridization in vegetative and reproductive is characterized by fully differentiated peltate leaves (i.e., shoots following developmental stages identified by sporangiophores), but still covered and protected by the Ambrose et  al. [44]. SeMoLF Y is expressed in sterile sterile leaf sheath (Fig.  4h, m). In F0, the spore mother shoot apical meristems (Fig.  6a), and in reproductive cells have not undergone meiosis (Fig.  4m). The Fertile shoots (Fig.  6b). SeMoLF Y expression within the stro- 1 (F1) stage is characterized by the exposure of half of bili is detected in the flanks of the meristem, microphyll the strobilus above the leaf sheath and by mature spores primordia, sporangia primordia as well as in sporan- (Fig.  4i, n). The Fertile 2 (F2) stage is characterized by gial tissue and young microphylls (Fig.  6b). When the fully exposed strobilus with leaves only at their base and sporangium wall and stalk are differentiated, SeMoLF Y by mature spores (Fig.  4j, o). In both stages, F1 and F2 expression is broad in all the cells but later in devel- elaters (i.e., sterile elongated cells that uncoil in response opment it starts to be preferentially expressed in the to changes in humidity, assisting in the dispersal of the sporangial wall and the sporogenous tissue and turned spores) can be easily identified (Fig. 4n, o). off in the stalk (Fig.  5c, d). During sporangia matura- Expression of LF Y homologs was evaluated in all tion, when the tapetum is differentiated, the expres - stages described above. We found two LF Y copies in the sion of SeMoLF Y can be detected in the tapetum itself monomorphic A. raddianum (Fig.  1), both genes have and in the spores before meiosis (Fig.  6e–-f). Dur- identical expression patterns with higher expression in ing the formation of tetrads, expression of SeMoLF Y fertile stages when compared to fiddleheads and ster - becomes restricted to the spores and the sporangial ile leaves (Fig. 5a; Additional file  7: Fig S4) with a lower Fig. 5 Semi‑ quantitative expression of LFY genes by RT‑PCR in selected fern species. a Expression of the monomorphic Adiantum raddianum (Pteridaceae) paralogs b Expression of the holodimorphic E. giganteum (Equisetaceae) copies. Fd: Fiddlehead, S: Sterile stage, F: Fertile stage. ‑ C: negative control. Actin was use as the positive control R odríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 9 of 16 Fig. 6 Expression of Selaginella moellendorffii LFY genes by in situ hybridization. a Vegetative shoot apical meristem (SAM). b Reproductive SAM with arising sporophylls and sporangia primordia. c–d Consecutive developing stages of the sporangium with a developing stalk and a forming sporangial wall surrounding the sporogenous tissue. e Two sporangial wall layers enclosing the sporogenous tissue prior to meiosis. f Two sporangial wall layers and tapetum. g Sporangium after meiosis h Mature sporangium with collapsed wall layers and tapetum and fully developed spores. li = ligule, m = microphyll, sti = sporangia, sw = sporangial wall, sy = sporophyll, t = tapetum, ts = tetrads, s = spores. Black arrowheads indicate the SAM. Scale bars = 100 µm Rodríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 10 of 16 wall (Fig.  6g), and finally expression is only detected in angiosperms. Our analysis does not recover the ancient the spores (Fig. 6h). type I and type II duplications in embryophytes, perhaps because an extensive sampling in bryophytes was not tar- Discussion geted in our study. We were also unable to retrieve the Unlike several other transcription factor families impor- previous reported two Osmunda sequences claimed to tant for plant development, the LFY gene lineage is usu- have promiscuous DNA binding types [6]. In our analy- ally retained primarily as a single copy in streptophytes sis, we were only able to identify DNA binding type I and across land plants [5]. A single large scale duplica- motifs in lycophyte and fern sequences, the same motif tion has been identified in this gene lineage linked to the type retained throughout tracheophytes. Our phyloge- diversification of seed plants resulting in the NDLY and netic analysis also recovers several local duplications LFY paralogs, followed by the loss of NDLY in angio- in lycophytes and ferns, some of which had been previ- sperms while retained in gymnosperms [35, 38, 45]. In ously identified [5, 39], and some are new in Isoetaceae this scenario, all other land plants retained single copy in lycophytes and in Equisetales in ferns. Many of the pre-duplication LFY genes with few species specific duplications occurring in these lineages could be possibly duplication events [5, 7, 39, 40, 46]. Interestingly, despite linked to reported whole genome duplication events [49]. this unusual strong selection acting upon LFY single copy However, most of the duplications found correspond to genes, their roles are divergent in different plant lineages species-specific paralogs and may be the result of other [5, 40]. Functional data points to plesiomorphic roles of mechanisms such as tandem repeat replication or retro- LFY genes in the control of cell divisions in the zygote transposition as copy numbers can be very different in to form the sporophyte in bryophytes [42]. Added evo- closely related species. lutionary roles include the maintenance of indeterminate Understanding the evolution and the putative roles of cell fate in both the gametophyte and the sporophyte in LFY genes in lycophytes and ferns is important as the ferns [39]. New key functions in reproductive transition data will help to clarify the roles of LFY homologs present are part of the functional repertoire of LFY homologs in in tracheophytes that are absent in non-vascular plants, seed plants. Namely, both LFY and NDLY are recruited and the plesiomorphic functions present in tracheo- in male and female cone development in gymnosperms phytes prior to the emergence of NDLY and LFY paral- [35–38, 45] and LFY homologs control floral meristem ogs in seed plants. The most comprehensive analysis to identity in angiosperms [1, 47]. Finally, LFY homologs date of fern LFY genes is that of Plackett et al. [39], which have also been recruited in maintaining the indetermi- identified two recent LFY paralogs in Ceratopteris richar- nate lateral inflorescence meristems in monocots and in dii that act partially redundantly in maintaining the inde- forming compound leaves in several legumes [27, 28, 33]. terminacy of the shoot apex, both in the gametophyte These data challenge the notion of gene duplication as a and the sporophyte. However, this same study showed major driver of gene functional diversification. Instead, that expression of at least one LFY paralog, CrLFY1 in it points to changes in protein sequences, specifically in C. richardii expands beyond the shoot apices, specifi - motifs responsible for DNA binding specificity, as key cally to newly emerged leaves as well as in young fertile features in the evolution of new functions [5]. To date, leaves. This coincides with previous reports identifying the changes in the interaction capabilities of LFY pro- higher expression of C. richardii LFY genes in shoot tips teins in different plant lineages have been directly linked and circinate reproductive leaves [7]. Similarly, the active to sequence changes of LFY genes [5, 40]. expression of LFY paralogs in the lycophyte Isoetes sinen- Contrary to this idea, recent comprehensive phylo- sis in microsporangia and megasporangia points to puta- genetic analyses have identified LFY duplications pre- tive roles in sporogenesis [41], and what has been broadly dating land plant diversification, or at least mosses and considered as the reproductive tissue formation by the liverworts, resulting in ancient duplicates that were lost formation of specialized sporangia with spore mother in tracheophytes [6, 48]. The hypothesis of the existence cells undergoing meiosis. Our data here suggest that of ancient duplicates predating land plant diversification several LFY homologs in lycophytes (S. moellendoriffi ) is supported with the finding of a few moss LFY genes and ferns (Adiantum raddianum and Equisetum gigan- with type I motifs and a few liverwort homologs with teum) are highly, and preferentially expressed in fertile type II motifs [6]. Less conflicting arguments have been stages and the expression is found to be correlated with posed around the evolution of LFY genes in tracheo- reproductive tissues (Figs. 5 and 6). If taken together, the phytes, as most phylogenetic analyses to date recover expression data available, including that presented here, single copy genes with type I motifs in lycophytes and raises the possibility that LFY genes play roles in the ferns and the already mentioned NDLY/LFY duplication formation of sporangia and, if so, in the specialization in seed plants with the retention of only LFY copies in of reproductive leaf derived tissues. If so, this function R odríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 11 of 16 is not new or exclusive to seed plant LFY homologs but 3) DNA binding capabilities have been exclusively was already present in the ancestor of vascular plants, tested with flowering plant partners, such as AG and because it is also present in lycophytes and ferns. Inter- AP1 [9, 40, 51]. Most of the key domains of LFY for estingly, expression of the P. patens PpLFY1 and PpLFY2 floral function have been determined. LFY proteins copies was detected in the sporophytes, beyond the early in gymnosperms, ferns, lycophytes or bryophytes zygote divisions and until the mature sporophyte [42]. A have conserved and easy to identify, oligomerization role of LFY homologs in sporophyte development was and DNA binding domains, but exhibit some non- in fact postulated originally in P. patens [42], but a role neutral changes, sometimes affecting key positions in the induction of the reproductive phase was ruled out [6, 41]. For instance, the C. richardii LFY proteins based on the lack of complementation of Arabidopsis lfy changes in amino acid positions within the N-ter- mutants using PpLFY genes [40]. minal and C-terminal domains have proven to be There are three major aspects to consider in the critical for the recovery of lfy mutants in Arabidopsis assessment of the LFY functional capabilities outside of [40]. angiosperms: Our alignment, as well as those of Yang et  al. [41] 1) Most of the functional data comes from com- and Gao et  al. [6], show that these changes are com- plementation analyses of Arabidopsis lfy mutants mon across lycophytes and ferns. In parallel, we have [37, 40, 41, 45, 50]). The observations showed that also identified motifs 9 in the DNA binding domain and sequences closely related to Arabidopsis frequently motif 10 in the SAM domain as fern-specific. In addition, recovered the wild type phenotype, and that, the res- studies on the putative protein–protein interactions and cue of lfy mutants by gymnosperm, fern and bryo- DNA-binding properties outside angiosperms are lack- phyte sequences decreased in efficiency. Although ing and in turn, LFY partners in lycophytes and ferns informative, complementation experiments only remain unknown making it difficult to assess if protein allow us to address similarities of the functional shifts have changed dimerization and/or DNA binding capabilities of these genes when compared to the processes endogenously. One critical aspect here is that Arabidopsis canonical LFY gene. Therefore, these several genes that are turned on upstream of LFY, such as experiments are very limited to reach conclusions FLOWERING LOCUS T/TERMINAL FLOWER 1 (FT/ regarding the genes endogenous roles. TFL1) and MADS-box proteins acting as flowering inte - 2) Fern LFY genes have pleiotropic early roles in the grators have lycophyte and fern homologs (Article under gametophyte and early sporophyte development, review and [44]). For example, the type II classic MADS- preventing the observation and analyses of their box genes from Selaginella moellendoriffi , SmMADS1 functions during the reproductive transition. The and SmMADS6, are expressed from the earliest stages comparative expression data available, does not rule of sporangia development to mature spores and, there- out the possibility that LFY genes play roles in the fore, overlap with SmLFY expression patterns [44]. More reproductive transition or at least in sporangia for- importantly, these MADS-box transcription factors seem mation in lycophytes and ferns, like those recorded to be expressed in an overlapping manner to LFY in the in seed plants. If so, an early acquisition of repro- sporophytes [52–56]. Future studies could aim at assess- ductive roles for LFY homologs could be traced back ing spatial temporal expression of MADS-box genes as at least to the common ancestor of tracheophytes. well as assessing protein partners in lycophytes and ferns Also unclear is the putative contribution of species- to assess what interactions were in place during sporo- specific paralogs to gene functional diversification. phyte growth and are not exclusive to angiosperms, simi- Our data and that of Yang et  al. [41] and Plackett lar to evaluations already done in gymnosperms [51, 57]. et  al. [39] points to some overlap between copies Understanding the rewiring of genetic regulatory net- resulting in redundancy but also recover important works for these critical developmental control genes will expression differences between paralogs suggesting allow an understanding of the apparent neofunctionaliza- some specialization among paralogs. Only the eval- tion of LFY that emerges in different plant lineages. uation of the endogenous gene function in ferns and lycophytes, will allow an examination of the con- Conclusions tribution of local duplicates to reproductive transi- Large-scale and species-specific LFY duplications have tion and in turn, assess the functional evolution of been identified in lycophytes and ferns. Some of these the LFY gene lineage, one of the core developmental may be the result of ancient WGD as is the case for Equi- genes in the evolution of land plants. setales, while the remaining species-specific copies are probably a result of recent polyploidy or hybridization Rodríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 12 of 16 events. We recovered little protein variation among The tissue was ground using liquid nitrogen and total LFY proteins across land plants. The type I motif that RNA extraction was carried out using PureLink Plant is shared by all tracheophytes is recovered in the DNA RNA Reagent (Invitrogen, USA). The total RNA was binding domain of ferns and lycophytes as expected; but quantified using a Nanodrop (Thermo, USA) and visu - the new motifs 9 and 10 are found to be fern-specific ally inspected in a 1% agarose gel stained with ethidium and located in LFY_SAM and the DNA Binding domain, bromide. Quality was assessed based on the presence respectively. As both domains are key for DNA binding, and integrity of ribosomal RNA bands. RNA extrac- further analyses are needed to elucidate downstream tions with concentrations over 200  ng/ul were sent to factors, and partner proteins that interact during sporo- the sequencing facility (Macrogen, South Korea). RNA- phyte growth in lycophytes and ferns. LFY genes in the seq experiments were conducted using a TruSeq mRNA lycophyte S. moellendoriffi and the two fern species A. library construction kit (Illumina) and sequenced in a raddianum and E. giganteum show a wide expression HiSeq2000 instrument producing 100 base paired-end among tissues. However, these expression patterns were reads. The transcriptomes were assembled de novo with found to be relatively higher in differentiating young fer - Trinity V2 at the Centro Nacional de Secuenciación tile stages and early sporangia developmental stages that Genómica (CNSG), following default settings [58]. Read then decreases once mature stages are reached. In addi- cleaning was performed with prinseqlite v0.20.4 with a tion, functional diversification among paralogs is found quality threshold of Q35 [59]. Contig metrics for each for E. giganteum that matches with that reported for C. species can be find on Additional file 3: Table S1. richardii LFY copies. Roles in reproduction have been reported in gymnosperms and angiosperms, but as the Gene phylogenies assessment of reproductive function across LFY genes To isolate LEAFY genes, sequences from Arabidopsis have been based on the ability of homologs to comple- thaliana, Selaginella moellendoriffi , Physcomitrium pat- ment Arabidopsis lfy mutants, it has not been possible to ens and Ceratopteris richardii LFY homologs were used clarify the role of LFY outside seed plants. Therefore, the as queries in BLASTN searches in all available ferns evaluation of endogenous gene function is key to know and lycophyte transcriptomes and genomes. BLAST the contribution of these genes and their duplicates in searches were made in Phytozome (http:// www. phyto reproductive transition. Our results allow us to propose zome. net/), NCBI, (https:// blast. ncbi. nlm. nih. gov/ Blast. a putative role for lycophyte and fern LFY homologs in cgi), and Fernbase (https:// www. fernb ase. org/), as well reproductive tissue diffenretiation, indicating that this as in transcriptome database OneKP (http:// www. bioin function may be present in the LFY gene lineage prior fodata. org/ Blast 4OneKP/). Most hits were retrieved as to the diversification of seed plants. Even though further complete coding sequence (CDS) with some partial CDS investigations are needed, our report represents a step derived from transcriptomic data. Partial sequences forward in the assessment of LFY genes in a reproductive were included only if they had 50% of the ca. 420 amino context outside seed plants. acids reported for the gene [1], and at least one of the two LFY domains [40]. Using the same strategy, we iso- Methods lated sequences from our own generated transcriptomes Transcriptome generation from Adiantum raddianum, Anemia villosa, Equise- Four species representing leaf monomorphism (Adi- tum giganteum and E. bogotense. These sequences were antum raddianum), hemidimorphism (Anemia vil- deposited in GenBank and can be found under the fol- losa) and holodimorphism (Equisetum bogotense and E. lowing accession numbers (MW219613, MW219614, giganteum) were chosen based on material availability MW219618–MW219620, MW219623–MW219625, of growing populations close to Medellin (Colombia). MW820861–MW820863). The species vouchers were deposited in the herbarium A comprehensive LEAFY matrix was built primar- of University of Antioquia (HUA) (C. Rodríguez-Pelayo ily using that of Sayou et  al. [5] but integrating all new 1–5). The material was collected in the field and was flash homologs identified here. All sequences were compiled in frozen in liquid nitrogen. For A. raddianum and A. vil- Bioedit (http:// www. mbio. ncsu. edu/ bioed it/ bioed it. html) losa the plant parts collected included fiddleheads as well and manually edited to exclusively keep the CDS for all as fertile and sterile pinnae at different developmental transcripts. Nucleotide sequences were subsequently stages. For E. giganteum and E. bogotense fertile and ster- aligned using the TranslatorX online platform (http:// ile shoot apical portions at different developmental stages trans latorx. co. uk/) implementing the online MAFFT were collected. A detailed morphological description of alignment algorithm [60] with default settings. The align - the stages included for each species can be found below ment was refined manually considering the SAM_LFY in the RT-PCR expression section. and DNA Binding domains reported as conserved in R odríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 13 of 16 LFY genes and can be found in Additional file  4: Fig. S3. Thermal cycling profiles followed an initial denatura - Maximum likelihood (ML) phylogenetic analyses using tion step (94 °C for 30 s), an annealing step (50–62 °C for the nucleotide sequences were performed in RaxML- 30  s) and an extension step with polymerase (72  °C for HPC2 BlackBox [61] through the CIPRES Science Gate- up to 1 min) repeated for 30–40 amplification cycles. As way [62]. Bootstrapping parameters were set for 1000 endogenous controls we tested Actin1 [65] as a control Bootstrap (BS) iterations. The tree was rooted with for all samples. Finally, the PCR products were run on a Algae LFY genes. Trees were observed and edited using 1.0% agarose gel stained with ethidium bromide and digi- FigTree v1.4.4 [63]. All the retrieved fern and lycophytes tally photographed using a Whatman Biometra BioDoc sequences can be found in Additional file 5: Table S2. Analyzer. To provide a more “quantitative” analysis of band brightness we converted our raw image data using Identification of protein domains and motifs imageJ. We counted the number of pixels and compared To detect both reported and new conserved motifs in it to the ACTIN band. Visualization of replicates is found LEAFY protein sequences across land plants we selected in Additional file 7: Fig. S4. a total of 34 sequences representing all subclades within the gene lineage. Sequences were permanently translated Developmental series of Equisetum giganteum and uploaded as amino acids to the online Multiple Em Sterile and reproductive shoots were collected in the field for Motif Elicitation (MEME) server (http:// meme- suite. and immediately fixed in formaldehyde–acetic acid– org/) and run to find up to 10 motifs with the default ethanol (FAA; 3.7% formaldehyde: 5% glacial acetic acid: options [64]. The motifs retrieved by MEME are reported 50% ethanol). For light microscopy, fixed material was according to their statistical significance. The MEME manually dehydrated through an alcohol–histochoice suite finds, in the given sequences, the most statistically series and embedded in Paraplast X-tra (Fisher Health- significant (low E value) motifs first. The E value of a care, Houston, Texas, USA). The samples were sec - motif is based on its log likelihood ratio, width, sites, and tioned at 10–20  μm with an RM2125 RTS (USA) rotary the size of the set. We numbered the motifs following the microtome. Sections were stained with Johansen’s safra- statistical significance resulting in the analyses. nin and 0.5% Astra Blue [66] and mounted in Permount (Fisher Scientific, Pittsburgh, Pennsylvania, USA). Sec - Expression analyses by Reverse Transcriptase PCR (RT‑PCR) tions were viewed and digitally photographed with a To examine and compare the expression patterns of LFY Zeiss Axioplan compound microscope and Zeiss stereo genes in monomorphic, hemidimorphic and holodimor- microscope equipped with an AxioCamERc5s digital phic ferns we dissected different plant parts and targeted camera with ZEN software. a range of developmental stages for Adiantum rad- dianum, and Equisetum giganteum, as described in the In situ hybridization results section of RT-PCR expression. In situ hybridization experiments were performed for We dissected and flash froze meristems, fiddleheads, the Selaginella moellendori L ffi FY homolog. Thus, we and fertile and sterile pinnae at the developmental stages used meristematic vegetative portions and strobili and described previously for the two ferns species. Total RNA followed sporangium development stages established by was prepared from dissected tissues using the same pro- previous descriptions in other Selaginella species [12, 44, tocol described above. RNA samples were treated with 67, 68]. The material was fixed for 3  h under vacuum in DNAseI (Roche, Basel, Switzerland) to remove DNA freshly prepared FAA (50% ethanol, 3.7% formaldehyde contamination and later quantified with a NanoDrop and 5% glacial acetic acid). The material was dehydrated 2000 (Thermo Scientific, Waltham, MA, USA). Three through an alcohol‐histochoice series and embedded in micrograms of RNA were used as a template for cDNA Paraplast X‐tra (Fisher, Waltham, MA, USA) with opti- synthesis (SuperScript III RT, Invitrogen) using OligodT mizations made by A. Vasco (i.e., fixation time 3 h; infil - primers. The cDNA was used undiluted for amplification tration solution changes every 3  h and hybridization reactions by RT-PCR. To ensure specificity for ampli - temperature fixed at 55  °C). Samples embedded were fication of each copy, the primers were designed in the maintained at 4 °C until use. The tissue was sectioned at regions outside of the conserved domains (Additional 8–10 μm on a Microm HM3555 rotary microtome. file 6: Table S3). Each amplification reaction incorporated DNA template for the S. moellendori S ffi eMoLFY probe 9 μL of EconoTaq (Lucigen, Middleton, WI, USA), 6 μL of was obtained by PCR amplification of a 500 bp fragment. nuclease-free water, 1 μL of BSA (bovine serum albumin) The fragment was cleaned using the QIAquick PCR puri - (5 μg/mL), 1 μL of Q solution (betaine 5 μg/μL), 1 μL of fication kit (Qiagen, Valencia, CA, USA). Digoxigenin forward primer (10 mm), 1 μL of reverse primer (10 mm), labeled RNA probe was prepared using T7polymerase and 1 μL of template cDNA, for a total reaction of 20 μL. (Roche, Switzerland), murine RNAse inhibitor (New Rodríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 14 of 16 Authors’ contributions England Biolabs, Ipswich, MA, USA), and RNA labeling CRP, NPM, BAA and AVG planned and designed the research. All authors mix (Roche, Switzerland) according to the manufacturer performed experiments, analyzed the data and wrote the final version of the protocols. In  situ hybridization following the protocol manuscript. All authors read and approved the final manuscript. that has been described previously optimizing incuba- Funding tion temperatures for each probe and tissue [5, 69, 70]. This study was funded by the Convocatoria de Sostenibilidad 2018–2019 and The sections were hybridized overnight at 55  °C. In  situ the Fundación para la Promoción de la Investigación y la Tecnología from Banco de la Republica (Project Number 4413). hybridized sections were subsequently dehydrated and permanently mounted in Permount (Fisher, Waltham, Availability of data and materials MA, USA). All sections were digitally photographed Sequences generated in this work were deposited in GenBank and can be found under the following Accession Numbers: MW219613, MW219614, using a Zeiss Axioplan microscope equipped with a MW219618–MW219620, MW219623–MW219625, MW820861–MW820863. Nikon DXM1200C digital camera. Sense probe controls are shown in Additional file 8: Fig. S5). Declarations Ethical approval and consent to participate Abbreviations The work was done with plants, mostly commonly cultivated and not under LFY: LEAFY; NDLY: NEEDLY; FLO: FLORICAULA; AGL24: AGAMOUS‑LIKE 24; any threat, so animal ethical approval and consent to participate are not appli‑ SVP: SHORT VEGETATIVE PHASE; SOC1: SUPRESSOR OF OVEREXPRESSION OF cable. As some plant species are cultivated and the ones collected are under CONSTANS; AP1: APETALA1; SP: SEPALLATA ; PI: PISTILLATA ; AP3: APETALA3; AG: collecting permission “Resolución 0524 del 27 de mayo del 2014”, Autoridad AGAMOUS; BSA: Bovine serum albumin; Q: Betaine; ML: Maximum likelihood; Nacional de Licencias Ambientales (ANLA). BS: Bootstrap; MEME: Multiple Em for Motif Elicitation; qRT‑PCR: Quantitative polymerase chain reaction; RT‑PCR: Reverse transcription polymerase chain Consent for publication reaction; SAM: Shoot apical meristem; WGD: Whole genome duplication. Authors declare consent for publication. Competing interests Supplementary Information Authors declare no competing interests on this research. The online version contains supplementary material available at https:// doi. org/ 10. 1186/ s13227‑ 021‑ 00188‑9. Author details Facultad de Ciencias Exactas y Naturales, Instituto de Biología, Universidad de Antioquia, Medellín, Colombia. The New York Botanical Garden, Bronx, Additional file 1: Figure S1. ML analysis of the LFY gene family. a. Sum‑ 3 4 NY, USA. Botanical Research Institute of Texas, Fort Worth, TX, USA. Centro mary tree including sequences from algae, bryophytes and tracheophytes. Nacional de Secuenciación Genómica, Facultad de Medicina, Sede de Investi‑ b. ML analysis of the LFY family including algae and land plant sequences. gación Universitaria, Universidad de Antioquia, Medellín, Colombia. Yellow stars indicate large duplication events. Number on each node indicate the bootstrap value (BS). Black arrowheads point to sequences Received: 23 September 2021 Accepted: 18 December 2021 isolated in this study. The colors correspond to the conventions on the bottom left. Scale: 0.3 Additional file 2: Figure S2. Protein sequences of the LFY family. For a total of 34 representative land plant sequences belonging to: Arabidop‑ sis thaliana, Ceratopteris richardii, Azolla filiculoides, Salvinia cucullata, References Equisetum giganteum, E. bogotense, Adiantum raddianum, Anemia 1. Weigel D, Alvarez J, Smyth DR, Yanofsky MF, Meyerowitz EM. LEAFY con‑ villosa, Selaginella moellendorffii, Physcomitrium patens and several trols floral meristem identity in Arabidopsis. Cell. 1992;69:843–59. algae species. The two characteristic domains of LFY proteins reported by 2. Blazquez MA, Soowal LN, Lee I, Weigel D. LEAFY expression and flower Sayou et al. [5, 9] are boxed. Blue arrowheads point to the key positions for initiation in Arabidopsis. Development. 1997;124:3835–44. DNA binding reported by Sayou et al. [5] .Black arrowheads point to the 3. Hempel FD, Weigel D, Mandel MA, Ditta G, Zambryski PC, Feldman LJ, conserved section among fern sequences. et al. Floral determination and expression of floral regulatory genes in Arabidopsis. Development. 1997;124:3845–53. Additional file 3: Table S1. Contig metrics. 4. Siriwardana NS, Lamb RS. The poetry of reproduction: the role of LEAFY in Additional file 4: Figure S3. Complete alignment in fasta for LFY Arabidopsis thaliana flower formation. Int J Dev Biol. 2012;56:207–21. homologs. 5. Sayou C, Monniaux M, Nanao MH, Moyroud E, Brockington SF, Thévenon Additional file 5: Table S2. LFY sequences complete data set. E, et al. A promiscuous intermediate underlies the evolution of LEAFY DNA binding specificity. Science. 2014;343:645–8. Additional file 6: Table S3. Primer sequences for fern and lycophyte LFY 6. Gao B, Chen M, Li X, Zhang J. Ancient duplications and grass‑specific genes used here for RT‑PCR and ISH. transposition influenced the evolution of LEAFY transcription factor Additional file 7: Figure S4. Quantification of LFY homologs expression genes. Commun Biol. 2019;2(1):10. using Image J onto gel images with three biological replicates. 7. Himi S, Sano R, Nishiyama T, Tanahashi T, Kato M, Ueda K, et al. Evolu‑ tion of MADS‑box gene induction by FLO/LFY genes. J Mol Evol. Additional file 8: Figure S5. Sense experiments for ISH in the meristem 2001;53:387–93. (SAM) and in reproductive tissue (See also Zumajo‑ Cardona et al. [44]). 8. Lu S, Wang J, Chitsaz F, Derbyshire MK, Geer RC, Gonzales NR, et al. CDD/ SPARCLE: the conserved domain database in 2020. Nucleic Acids Res. 2020;48:D265–8. Acknowledgements 9. Sayou C, Nanao MH, Jamin M, Posé D, Thévenon E, Grégoire L, et al. A We thank the staff at The New York Botanical Garden especially Cecilia SAM oligomerization domain shapes the genomic binding landscape of Zumajo‑ Cardona and Tynisha Smalls for help in the standardization of lab the LEAFY transcription factor. Nat Commun. 2016;7:11222. protocols. We thank the OneKP repository staff for facilitating access to the 10. Carpenter R, Coen ES. Floral homeotic mutations produced by transpo‑ online database. We thank members of the Plant Evo Devo lab for help in the son‑mutagenesis in Antirrhinum majus. Genes Dev. 1990;4:1483–93. laboratory and helpful discussions. R odríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 15 of 16 11. Coen ES, Romero J, Doyle S, Elliott R, Murphy G, Carpenter R. floricaula: a containing vegetative shoot and undifferentiated male cone primordia. homeotic gene required for flower development in Antirrhinum majus. Planta. 1998;206:619–29. Cell. 1990;63:1311–22. 36. Shindo S, Sakakibara K, Sano R, Ueda K, Hasebe M. Characterization of a 12. Schultz EA, Haughn GW. LEAFY, a homeotic gene that regulates inflores‑ FLORICAULA/LEAFY homologue of Gnetum parvifolium and its implica‑ cence development in Arabidopsis. Plant Cell. 1991;3:771–81. tions for the evolution of reproductive organs in seed plants. Int J Plant 13. Moyroud E, Kusters E, Monniaux M, Koes R, Parcy F. LEAFY blossoms. Sci. 2001;162:1199–209. Trends Plant Sci. 2010;15:346–52. 37. Moyroud E, Monniaux M, Thévenon E, Dumas R, Scutt CP, Frolich M, Parcy 14. Parcy F, Nilsson O, Busch MA, Lee I, Weigel D. A genetic framework for F. A link between LEAFY and B‑ gene homologues in Welwitschia mirabilis floral patterning. Nature. 1998;395:561–6. sheds light on ancestral mechanisms prefiguring floral development. 15. Lee J, Oh M, Park H, Lee I. SOC1 translocated to the nucleus by interaction New Phytol. 2007;216:469–81. with AGL24 directly regulates LEAFY. Plant J. 2008;55:832–43. 38. Vázquez‑Lobo A, Carlsbecker A, Vergara‑Silva F, Alvarez‑Buylla ER, Piñero 16. Liu C, Xi W, Shen L, Tan C, Yu H. Regulation of floral patterning by flower ‑ D, Engström P. Characterization of the expression patterns of LEAFY/ ing time genes. Dev cell. 2009;16:711–22. FLORICAULA and NEEDLY orthologs in female and male cones of the 17. Gregis V, Andrés F, Sessa A, Guerra RF, Simonini S, Mateos JL, et al. Identifi‑ conifer genera Picea, Podocarpus, and Taxus: implications for current evo‑ cation of pathways directly regulated by SHORT VEGETATIVE PHASE dur‑ devo hypotheses for gymnosperms. Evol Dev. 2007;9:446–59. ing vegetative and reproductive development in Arabidopsis. Genome 39. Plackett AR, Conway SJ, Hewett Hazelton KD, Rabbinowitsch EH, Lang‑ Biol. 2013;14:R56. dale JA, Di Stilio VS. LEAFY maintains apical stem cell activity during shoot 18. Kaufmann K, Wellmer F, Muiño JM, Ferrier T, Wuest SE, Kumar V, et al. development in the fern Ceratopteris richardii. Elife. 2018;7: e39625. Orchestration of Floral Initiation by APETALA1. Science. 2010;328:85–9. 40. Maizel A, Busch MA, Tanahashi T, Perkovic J, Kato M, Hasebe M, et al. 19. Moyroud E, Minguet EG, Ott F, Yant L, Posé D, Monniaux M, et al. The floral regulator LEAFY evolves by substitutions in the DNA binding Prediction of regulatory interactions from genome sequences using a domain. Science. 2005;308:260–3. biophysical model for the Arabidopsis LEAFY transcription factor. Plant 41. Yang T, Du M, Guo Y, Liu X. Two LEAFY homologs ILFY1 and ILFY2 Cell. 2011;23:1293–306. control reproductive and vegetative developments in Isoetes L. Sci Rep. 20. Winter CM, Austin RS, Blanvillain‑Baufumé S, Reback MA, Monniaux M, Wu 2017;7:225. M‑F, et al. LEAFY target genes reveal floral regulatory logic, cis motifs, and 42. Tanahashi T, Sumikawa N, Kato M, Hasebe M. Diversification of gene a link to biotic stimulus response. Dev cell. 2011;20:430–43. function: homologs of the floral regulator FLO/LFY control the first 21. Han Y, Zhang C, Yang H, Jiao Y. Cytokinin pathway mediates APETALA1 zygotic cell division in the moss Physcomitrella patens. Development. function in the establishment of determinate floral meristems in Arabi‑ 2005;132:1727–36. dopsis. Proc Natl Acad Sci USA. 2014;111:6840–5. 43. PPGI. A community‑ derived classification for extant lycophytes and ferns. 22. Winter CM, Yamaguchi N, Wu M‑F, Wagner D. Transcriptional programs J Syst Evol. 2016;54:563–603. regulated by both LEAFY and APETALA1 at the time of flower formation. 44. Ambrose BA, Smalls TL, Zumajo‑ Cardona C. All type II classic MADS‑ Physiol Plant. 2015;155:55–73. box genes in the lycophyte Selaginella moellendorffii are broadly yet 23. William DA, Su Y, Smith MR, Lu M, Baldwin DA, Wagner D. Genomic discretely expressed in vegetative and reproductive tissues. Evol Dev. identification of direct target genes of LEAFY. Proc Natl Acad Sci USA. 2021;23:215–30. 2004;101:1775–80. 45. Mouradov A, Glassick T, Hamdorf B, Murphy L, Fowler B, Marla S, et al. 24. Southerton SG, Strauss SH, Olive MR, Harcourt RL, Decroocq V, Zhu X, NEEDLY, a Pinus radiata ortholog of FLORICAULA/LEAFY genes, expressed et al. Eucalyptus has a functional equivalent of the Arabidopsis floral in both reproductive and vegetative meristems. Proc Natl Acad Sci USA. meristem identity gene LEAFY. Plant Mol Biol. 1998;37:897–910. 1998;95:6537–42. 25. Wada M, Cao Q, Kotoda N, Soejima J, Masuda T. Apple has two ortho‑ 46. Frohlich MW, Parker DS. the mostly male theory of flower evolutionary logues of FLORICAULA/LEAFY involved in flowering. Plant Mol Biol. origins: from genes to fossils. Syst Bot. 2000;25:155–70. 2002;49:567–77. 47. Nilsson O, Weigel D. Modulating the timing of flowering. Curr Opin 26. Yu Q, Moore PH, Albert HH, Roader AHK, Ming R. Cloning and characteri‑ Biotechnol. 1997;8:195–9. zation of a FLORICAULA/LEAFY ortholog, PFL, in polygamous papaya. Cell 48. Brunkard JO, Runkel AM, Zambryski PC. Comment on “A promiscuous Res. 2005;15:576–84. intermediate underlies the evolution of LEAFY DNA binding specificity.” 27. Hofer J, Turner L, Hellens R, Ambrose M, Matthews P, Michael A, et al. Science. 2015;347:621–621. UNIFOLIATA regulates leaf and flower morphogenesis in pea. Curr Biol. 49. Huang C‑H, Qi X, Chen D, Qi J, Ma H. Recurrent genome duplication 1997;7:581–7. events likely contributed to both the ancient and recent rise of ferns. J 28. Wang H, Chen J, Wen J, Tadege M, Li G, Liu Y, et al. Control of compound Integr Plant Biol. 2019;62:433–55. leaf development by FLORICAULA/LEAFY ortholog SINGLE LEAFLET1 in 50. Shiokawa T, Yamada S, Futamura N, Osanai K, Murasugi D, Shinohara K, Medicago truncatula. Plant Physiol. 2008;146:1759–72. et al. Isolation and functional analysis of the CjNdly gene, a homolog 29. Zhao W, Chen Z, Liu X, Che G, Gu R, Zhao J, Wang Z, Hou Y, Zhang X. in Cryptomeria japonica of FLORICAULA/LEAFY genes. Tree Physiol. CsLFY is required for shoot meristem maintenance via interaction with 2008;28:21–8. WUSCHEL in cucumber (Cucumis sativus). New Phyt. 2017;218:344–56. 51. Silva CS, Puranik S, Round A, Brennich M, Jourdain A, Parcy F, et al. Evolu‑ 30. Pouteau S, Nicholls D, Tooke F, Coen E, Battey N. The induction and main‑ tion of the plant reproduction master regulators LFY and the MADS tenance of flowering in Impatiens. Development. 1997;124:3343–51. transcription factors: the role of protein structure in the evolutionary 31. Molinero‑Rosales N, Jamilena M, Zurita S, Gómez P, Capel J, Lozano R. development of the flower. Front Plant Sci. 2016;6. https:// doi. org/ 10. FALSIFLORA, the tomato orthologue of FLORICAULA and LEAFY, controls 3389/ fpls. 2015. 01193/ full. Accessed 29 Dec 2020. flowering time and floral meristem identity. Plant J. 1999;20:685–93. 52. Münster T, Pahnke J, Rosa AD, Kim JT, Martin W, Saedler H, et al. Floral 32. Ahearn KP, Johnson HA, Weigel D, Wagner DR. NFL1, a Nicotiana tabacum homeotic genes were recruited from homologous MADS‑box genes LEAFY‑like gene, controls meristem initiation and floral structure. Plant preexisting in the common ancestor of ferns and seed plants. Proc Natl Cell Physiol. 2001;42:1130–9. Acad Sci USA. 1997;94:2415–20. 33. Rao NN, Prasad K, Kumar PR, Vijayraghavan U. Distinct regulatory role 53. Svensson ME, Johannesson H, Engström P. The LAMB1 gene from the for RFL, the rice LFY homolog, in determining flowering time and plant clubmoss, Lycopodium annotinum, is a divergent MADS‑box gene, architecture. Proc Natl Acad Sci USA. 2008;105:3646–51. expressed specifically in sporogenic structures. Gene. 2000;253:31–43. 34. Sharma B, Meaders C, Wolfe D, Holappa L, Walcher‑ Chevillet C, Kramer 54. Tanabe Y, Uchida M, Hasebe M, Ito M. Characterization of the Selaginella EM. Homologs of LEAFY and UNUSUAL FLORAL ORGANS promote the remotifolia MADS‑box gene. J Plant Res. 2003;116:69–73. transition from inflorescence to floral meristem identity in the Cymose 55. Huang Q, Li W, Fan R, Chang Y. New MADS‑box gene in fern: cloning Aquilegia coerulea. Front Plant Sci. 2019. https:// doi. org/ 10. 3389/ fpls. 2019. and expression analysis of DfMADS1 from Dryopteris fragrans. PLoS ONE. 01218/ full. Accessed 6 Nov 2020. 2014;9: e86349. 35. Mellerowicz EJ, Horgan K, Walden A, Coker A, Walter C. PRFLL–a Pinus radiata homologue of FLORICAULA and LEAFY is expressed in buds Rodríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 16 of 16 56. Ruiz‑Estévez M, Bakkali M, Martín‑Blázquez R, Garrido ‑Ramos MA. Dif‑ ferential expression patterns of MIKCC‑type MADS‑box genes in the endangered fern Vandenboschia speciosa. Plant Gene. 2017;12:50–6. 57. Wang Y‑ Q, Melzer R, Theißen G. Molecular interactions of orthologues of floral homeotic proteins from the gymnosperm Gnetum gnemon provide a clue to the evolutionary origin of ‘floral quartets.’ Plant J. 2010;64:177–90. 58. Grabherr MG, Haas BJ, Yassour M, Levin JZ, Thompson DA, Amit I, et al. Full‑length transcriptome assembly from RNA‑Seq data without a refer ‑ ence genome. Nat Biotechnol. 2011;29:644–52. 59. Schmieder R, Edwards R. Quality control and preprocessing of metagen‑ omic datasets. Bioinformatics. 2011;27:863–4. 60. Katoh K, Misawa K, Kuma K, Miyata T. MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Res. 2002;30:3059–66. 61. Stamatakis A, Hoover P, Rougemont J. A rapid bootstrap algorithm for the RAxML web servers. Syst Biol. 2008;57:758–71. 62. Miller MA, Pfeiffer W, Schwartz T. Creating the CIPRES science gateway for inference of large phylogenetic trees. Gateway Computing Environments Workshop (GCE), 2010: 1–8. 63. Rambaut A. FigTree v.1.4.4: Tree Figure Drawing Tool. 2018. http:// tree. bio. ed. ac. uk/ softw are/ figtr ee/. Accessed 29 Dec 2021. 64. Bailey TL, Williams N, Misleh C, Li WW. MEME: discovering and analyzing DNA and protein sequence motifs. Nucleic Acids Res. 2006;34:W369–73. 65. Brouwer P, Bräutigam A, Külahoglu C, Tazelaar AOE, Kurz S, Nierop KGJ, et al. Azolla domestication towards a biobased economy? New Phytol. 2014;202:1069–82. 66. Kraus JE, de Sousa HC, Rezende MH, Castro NM, Vecchi C, Luque R. Astra blue and basic fuchsin double staining of plant materials. Biotech Histo‑ chem. 1998;73:235–43. 67. Lyon FM. A study of the sporangia and gametophytes of Selaginella apus and Selaginella rupestris. Bot gaz. 1901;32:124–41. 68. Schulz C, Little DP, Stevenson DW, Bauer D, Moloney C, Stützel T. An overview of the morphology, anatomy, and life cycle of a new model species: the lycophyte Selaginella apoda (L.) spring. Int J Plant Sci. 2010;171:693–712. 69. Ambrose BA, Lerner DR, Ciceri P, Padilla CM, Yanofsky MF, Schmidt RJ. Molecular and genetic analyses of the Silky1 gene reveal conservation in floral organ specification between eudicots and monocots. Mol Cell. 2000;5:569–79. 70. Vasco A, Ambrose BA. Simple and divided leaves in ferns: exploring the genetic basis for leaf morphology differences in the genus Elaphoglos - sum (Dryopteridaceae). Int J Pl Sci. 2020;21:5180. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png EvoDevo Springer Journals

Evolution and expression of LEAFY genes in ferns and lycophytes

Download PDF
16 pages

Loading next page...
 
/lp/springer-journals/evolution-and-expression-of-leafy-genes-in-ferns-and-lycophytes-GQJEOd0yFy

References (79)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2022
eISSN
2041-9139
DOI
10.1186/s13227-021-00188-9
Publisher site
See Article on Publisher Site

Abstract

Background: The LEAFY (LFY) transcription factors are present in algae and across land plants. The available expres‑ sion and functional data of these genes in embryophytes suggest that LFY genes control a plethora of processes including the first zygotic cell division in bryophytes, shoot cell divisions of the gametophyte and sporophyte in ferns, cone differentiation in gymnosperms and floral meristem identity in flowering plants. However, their putative plesio ‑ morphic role in plant reproductive transition in vascular plants remains untested. Results: We perform Maximum Likelihood (ML) phylogenetic analyses for the LFY gene lineage in embryophytes with expanded sampling in lycophytes and ferns. We recover the previously identified seed plant duplication that results in LEAFY and NEEDLY paralogs. In addition, we recover multiple species‑specific duplications in ferns and lycophytes and large‑scale duplications possibly correlated with the occurrence of whole genome duplication ( WGD) events in Equisetales and Salviniales. To test putative roles in diverse ferns and lycophytes we perform LFY expression analyses in Adiantum raddianum, Equisetum giganteum and Selaginella moellendorffii. Our results show that LFY genes are active in vegetative and reproductive tissues, with higher expression in early fertile developmental stages and dur‑ ing sporangia differentiation. Conclusions: Our data point to previously unrecognized roles of LFY genes in sporangia differentiation in lycophytes and ferns and suggests that functions linked to reproductive structure development are not exclusive to seed plant LFY homologs. Keywords: LEAFY, Ferns, Lycophytes, Reproductive transition, Sporogenesis Background clades NEEDLY (NDLY) and LFY. Both copies have been The LEAFY (LFY) gene lineage is a plant specific tran - retained in gymnosperms but NDLY was lost in angio- scription factor family [1]. LFY is best known for its role sperms [5, 6]. In addition, some species-specific duplica - in the flowering plant model species Arabidopsis thaliana tions have been identified in many lineages of land plants (Arabidopsis), where it is a key integrator for flowering [5, 7]. transition, controlling both the floral meristem identity LFY genes encode a ca. 220–350 amino acid pro- and the activation of the floral organ identity genes [1–4]. tein that acts as a homodimer and it has two recogniz- LFY genes are primarily retained as a single copy gene able domains: the N-terminal (also known as LFY_SAM) in algae and across land plants, with only a large-scale and the C-terminal (also known as C_LFY_FLO) DNA duplication occurring in seed plants resulting in two binding domains [8]. The LFY_SAM domain is key for dimerization, DNA binding and DNA accessibility, while the C_LFY_FLO domain has been proposed to medi- *Correspondence: lucia.pabon@udea.edu.co ate DNA binding specificity that has evolved across land Facultad de Ciencias Exactas y Naturales, Instituto de Biología, plants [5, 9]. Based on the dominant amino acids, three Universidad de Antioquia, Medellín, Colombia different motif types have been identified in the DNA Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Rodríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 2 of 16 binding domain in different streptophyte lineages: liver - and circinate fertile leaves, both characterized by active worts + tracheophytes have the type I motif, mosses have cell division [7, 39]. In addition, heterologous expression the type II motif, and algae possess the type III motif [5]. of fern LFY homologs does not fully recover wild type Finally, hornworts have been reported to have a pro- phenotypes in lfy Arabidopsis mutants [40]. Conversely, miscuous motif with versatile DNA binding capabilities endogenous loss-of-function double lfy mutants in C. binding all three types of DNA motifs [5]. richardii exhibit defects in gametophyte development Functional studies of LFY homologs in other flowering by cell division arrest of the apical cell, as well as trun- plants, such as Antirrhinum majus (snapdragon), showed cate sporophyte development accompanied by abnormal that floricaula (flo) mutants, do not transition from veg - leaf development [39, 40]. However, because lfy mutants etative to reproductive stages. In these mutants, inflores - result in interrupted shoot development, it is unclear if cence meristems do not form floral meristems on their LFY homologs play roles during the reproductive transi- flanks, resulting in branched stems forming more leaves tion in ferns. instead of producing flowers, similar to the Arabidopsis Studies in lycophytes are restricted to Isoetes, where lfy mutants [1, 10–12]. Conversely, the overexpression of two LFY homologs were identified. The Isoetes LFY LFY homologs turns the indeterminate inflorescence into homologs are also expressed in the vegetative and repro- a determinate meristem forming solitary flowers [13]. In ductive tissues of the sporophyte with higher expression Arabidopsis, LFY is upregulated by flowering time genes in juvenile tissues and fertile leaves bearing megaspo- including AGAMOUS-LIKE 24 (AGL24), SHORT VEG- rangia and microsporangia [41]. Importantly, heterolo- ETATIVE PHASE (SVP) and SUPRESSOR OF OVEREX- gous expression of Isoetes LFY genes in Arabidopsis did PRESSION OF CONSTANS (SOC1) [14–17]. Once LFY is not rescue lfy mutant phenotypes [41]. Finally, the role active in the floral meristem, it stimulates cytokinin sign - of LFY homologs has also been evaluated in bryophytes, aling and activates floral fate acquisition together with in the model Physcomitrium patens. This species has two APETALA1 (AP1) (a MADS-box homolog) responsible paralogs, PpLFY1 and PpLFY2 regulate the first division for sepal and petal identity [18–22]. LFY can also activate of the zygote and in turn controls proper cell division of other floral organ identity MADS-box genes including the sporophyte as seen in the double mutant with devel- SEPALLATA (SEP), PISTILLATA (PI), APETALA3 (AP3) opmentally arrested zygotes [42]. and AGAMOUS (AG) [23]. The data available for land plant LFY homologs was Based on functional data from Arabidopsis and snap- summarized by Plackett et  al. [39] who concluded that dragon, LFY has been characterized as a critical fac- plesiomorphic roles of LFY genes include the control tor controlling floral meristem identity and floral fate. of cell divisions in the zygote to form the sporophyte in These roles are conserved in other angiosperms, such as bryophytes, and that LFY homologs were subsequently apple, papaya and eucalyptus [24–26]. In addition, other co-opted to maintain indeterminate cell fate in both the roles have been reported for LFY members across angio- gametophyte and the sporophyte in ferns. These authors sperms. For instance, LFY homologs have been recruited also hypothesized that, only in seed plants, LFY genes in the formation of compound leaves in legumes [27, 28], were active during cone development in gymnosperms in shoot apical meristem (SAM) development in cucum- and controlled floral meristem identity in angiosperms. ber, impatiens, and tomato [29–32], in inflorescence Interestingly, due to their early critical roles in shoot branch patterning in rice [33], and in floral merism and development it is still unclear if fern LFY homologs play spiral arrangement of floral parts in columbines [34]. any function in the formation of fertile leaves and their Studies of gene homologs in gymnosperms show that specialization for spore production in ferns. If so, the both LFY and NDLY are expressed constitutively in LFY gene function in reproductive transition and spo- reproductive and vegetative structures, and spatial–tem- rangia development is not exclusive to seed plants. In this poral expression differences between paralogs have been context, we aim to reconstruct the evolution of the LFY linked to male and female cone differentiation [35–38]. In gene lineage with ample sampling from lycophytes and addition, heterologous expression of gymnosperm LFY ferns, and to assess expression patterns of LFY genes in homologs can rescue the wild type phenotype from lfy lycophytes and ferns during their reproductive transition. mutants in Arabidopsis [37], which suggest functional We make use of public databases as well as our own tran- conservation in the transition to reproduction across scriptomes from four fern species with different degrees seed plants. of leaf dimorphism (i.e., morphological differences Expression and function of two LFY paralogs have also between fertile and sterile leaves): the monomorphic (i.e., been studied in the fern Ceratopteris richardii. RT-PCR sterile and fertile leaves with same morphology) Adian- assays and in  situ hybridization showed high expres- tum raddianum, the hemidimorphic (i.e., sterile and fer- sion of the two C. richardii LFY copies in the shoot tips tile leaves with slightly different morphologies having for R odríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 3 of 16 instance differentiated fertile pinnae) Anemia villosa, and The evolution of LFY gene homologs within ferns and two holodimorphic (i.e., sterile and fertile leaves with dif- lycophytes, closely match the phylogeny of those lineages ferent morphology) species, Equisetum giganteum and [43]. However, our analysis recovered large-scale duplica- E. bogotense. The expression of LFY copies was assessed tions of LFY in Equisetales, in Salviniales, within Pteri- by RT-PCR in selected ferns with the extreme changes dineae in the Vittarioideae subfamily, and in Eupolypods in leaf dimorphism (Adiantum raddianum and Equise- II (sensu, [43]). However, the exact timing of the latter tum giganteum), and by in situ hybridization (ISH) in the two duplications cannot be placed with certainty. Unlike lycophyte Selaginella moellendoriffi . Our results allow their expected phylogenetic placements, our analyses us to propose a putative role for lycophyte and fern LFY recovered Cibotium glaucum LFY homolog nested within homologs in reproductive tissue formation (i.e., sporan- the Eupolypod I sequences and the Polypodium plecto- gia development), indicating that this function may have lens homolog within Eupolypod II sequences. In the spe- been present in the LFY gene lineage prior to the diversi- cies phylogenies these species belong to Cyatheales and fication of seed plants. Eupolypoids I, respectively [43]. Finally, our analysis reveals species-specific duplica - Results tions in the lycophyte Huperzia selago, a well-known LEAFY gene family evolution polyploid, and in a number of ferns, most of which have We present here the most comprehensive phylogenetic been identified as tetraploids or recent hybrids, includ - analysis of LFY genes to date, consisting of 228 sequences ing Anemia villosa, Asplenium platyneuron, Botrypus that include 95 from Sayou et  al. [5] and 133 sequences virginianus, Ceratopteris richardii, Cystopteris fragilis, from an expanded sampling targeting ferns and lyco- Dennstaedtia sp., Equisetum arvense, E. bogotense, Leu- phytes (Additional file  1: Fig. S1). The complete aligned costegia immersa, Lindsaea microphylla, Ophioglossum data set consists of 1338 characters from which 833 are vulgatum, Polypodium hesperium, P. plectolens, Psilo- informative. tum nudum, Pityrogramma trifoliata and Sceptridium Similar to previous reports, our analysis recovered LFY dissectum. genes predominantly as single copy genes in streptophyte algae and land plants, with a single exception in seed LFY protein domains and motifs plants. In turn, the gene tree mostly recovers the phylo- LFY proteins are characterized by two conserved genetic relationships recorded for major plant lineages. domains [5, 6, 9], the LFY_SAM (in the N terminus), The bryophyte LFY genes are sister to LFY homologs which in our analysis corresponds to motifs 1, 2 and 3 in vascular plants (Bootstrap Support, BS = 72) (Addi- and the C_LFY_FLO DNA binding domain (in the C ter- tional file  1: Fig. S1). However, within vascular plants, minus) which in our analysis corresponds to motifs 4 and sequences from ferns and lycophytes are recovered as 6 [6] (Fig. 2). We do not find outstanding variation at the sister to each other (BS = 61) and the lycophyte + fern protein level between divergent plant lineages (i.e., lyco- LFY clade is sister to the seed plant LFY homologs with phytes, ferns, gymnosperms and angiosperms), but we low support (Fig.  1). This is inconsistent with the vascu - were able to identify motifs 9 and 10 as exclusive to fern lar plant phylogeny accepted to date, where lycophytes protein sequences (Fig. 2). are sister to euphyllophytes (the clade formed by ferns In addition, the protein alignment does allow the iden- and seed plants) [43]. In addition, our topology recovers tification of some conserved amino acids in fern LFY the LFY and NDLY duplication event prior to the diver- proteins in the most variable region between the SAM_ sification of seed plants with a loss of NDLY homologs LFY and the C_LFY_FLO DNA binding domain (Addi- in angiosperms [5, 38, 40]. Moreover, we found multiple tional file  2: Fig. S2). Finally, following the key amino fern and lycophyte species-specific duplications some of acids positions for DNA binding, our analyses recovered which could explain the multiple homologs reported pre- the amino acids HRH, matching the type I described for viously in the fern Ceratopteris richardii and in several tracheophytes [5]. They correspond to positions 332, 370 Isoetes spp. in the lycophytes [39, 41]. and 413 in our alignment (Fig. 3). (See figure on next page.) Fig. 1 ML analysis of the LFY transcription factor family a Summary tree including sequences from algae, bryophytes and tracheophytes with Boostrap values (BS) as numbers on top of branches for major clades. b Detail of lycophyte and fern LFY clades. Yellow stars indicate large scale (major) duplication events. Number on each node indicate the BS. Black arrowheads point to sequences isolated in this study. The colors correspond to the conventions on the bottom left. Scale: 0.3 Rodríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 4 of 16 Fig. 1 (See legend on previous page.) R odríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 5 of 16 Fig. 2 a Conserved motifs of LFY proteins identified by a MEME analysis, their motifs, numbers assigned and their respective sequence. b Map of the motif positions in land plant representative proteins RT‑PCR expression of LFY homologs in selected ferns linked to gametophyte-to-sporophyte transitions already To hypothesize the roles of LEAFY genes during the reported [39], we evaluated the expression of all LFY cop- reproductive transition in ferns, in addition to those ies in the ferns Adiantum raddianum and Equisetum Rodríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 6 of 16 Fig. 3 Detail of the LFY Protein domains alignment for 34 representative land plant sequences belonging to: Arabidopsis thaliana, Ceratopteris richardii, Azolla filiculoides, Salvinia cucullata, Equisetum giganteum, E. bogotense, Adiantum raddianum, Anemia villosa, Selaginella moellendorffi, Physcomitrium patens and several algae species. The two characteristic domains of LFY proteins reported by Sayou et al. [5, 9] are boxed. Blue arrowheads point to the key positions for DNA binding reported by Sayou et al. [5] R odríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 7 of 16 Fig. 4 Developmental stages of the fern species Adiantum raddianum and Equisetum giganteum. (a–e) A. raddianum stages: Two sterile stages include Fd: fiddlehead and S: sterile expanded pinnae (a, b). Three fertile stages include: F0 leaves with unrolling pinnae and an indusium of ca. 250 µm (c), F1 Nearly fully expanded leaf and an indusium of ca. 800 µm (d), and F2 Fully expanded leaves and an indusium of ca. 2 mm. (e). Fi = False indusia, S: Sterile stage, F: Fertile stage. (f–o) Developmental stages of E. giganteum. SAMs producing photosynthetic leaves only in as thin and green at sterile stage S0 (f) and thick and brownish‑ green at sterile stage S1 (g) that vary in meristem size (k, l). Stage F0 has fully differentiated peltate fertile leaves (sporangiophores), that are covered and protected by the sterile leaf sheath and the spore mother cells have not undergone meiosis (h, m); Stage F1 is characterized by the exposure of half of the strobili above the sterile leaves, of ca. 1 cm and carry mature spores (i, n). Stage F2 has a fully exposed strobili with sterile leaves only at its base of ca 4 mm and mature spores and expanded elaters (j, o). Scale bars for c–e = 250 µm; f–j = 1 cm; Scale bars k–o = 2 mm. Scale bars in magnifications = 50 µm. Black arrowheads indicate the SAM, black arrowhead in insets of n, o indicate the elaters. Sl = Sterile leaves, Pl = Peltate spore bearing leaves, Sti = sporogenous tissue, S = Spores giganteum. These two genes represent extremes in the tissue protecting the sori) or sori (i.e., the clusters of leaf dimorphism gradient, as A. raddianum is monomor- sporangia) (Fig.  4a, b). In addition, three fertile devel- phic (i.e., there are no gross morphological differences in opmental stages were established, namely, Stage Fer- sterile and fertile leaves) and E. giganteum is holodimor- tile F0, F1 and F2, which considered the degree of leaf phic (i.e., sterile and fertile leaves vary dramatically in expansion and the indusia size (Fig.  4c–e). F0 is char- size, shape and position (Fig. 4). acterized by leaves with expanding pinnae and an indu- In A. raddianum sterile leaves can be separated into sium width of ca. 250  µm (Fig.  4c). F1 is characterized fiddleheads (Fd) and young sterile leaves (S), lack - by almost entirely extended leaves and an indusium ing any sign of indusia (i.e., the membranous covering width of ca. 800  µm (Fig.  4d). F2 is characterized by Rodríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 8 of 16 fully extended leaves and an indusium width of 2  mm. level of expression at the F2 stage. On the other hand, (Fig. 4e). Sporangia developmental stages were not spe- in the holodimorphic E. giganteum we found three cifically described for each stage as sporangia develop - LF Y paralogs with different expression patterns during ment is asynchronous, meaning that different sporangia reproductive transition. EqgiLF Y2 was undetected in are in different developmental stages in a single sori in the stages and tissues studied. EqgiLF Y1 is more highly any given time (Fig. 4a–e). expressed in fertile than sterile stages with two peaks In E. giganteum, we identified two different develop - of expression, one at F0 and the second at F2 (Fig.  5b). mental stages for sterile stems (Fig.  4f, g) and three of Finally, EqgiLF Y3 is detected in sterile and until early reproductive meristems (Fig.  4h–j). Sterile stages cor- fertile stages at F0, with very low levels of expression respond to shoot apical meristems (SAMs) producing detected in the two latter fertile (F1 and F2) stages. leaves only (Fig.  4k, l). These were found in two forms, thin and green (Fig.  4f) or thick and brownish-green (Fig.  4g). Although, anatomical sections revealed dif- In situ hybridization expression of LFY homologs ferences in meristem width, the two forms exclusively in the lycophyte Selaginella moellendorffii produce identical leaves (Fig.  4k, l). The identification To determine the detailed spatio-temporal expression of developmental stages in the fertile stems relied on patterns of the only LF Y gene identified in the het - changes of strobilus size, as well as sporangia and spore/ erosporous lycophyte S. moellendorffii, we performed elater development (Fig.  4h–j, m–o). Stage Fertile 0 (F0) in  situ hybridization in vegetative and reproductive is characterized by fully differentiated peltate leaves (i.e., shoots following developmental stages identified by sporangiophores), but still covered and protected by the Ambrose et  al. [44]. SeMoLF Y is expressed in sterile sterile leaf sheath (Fig.  4h, m). In F0, the spore mother shoot apical meristems (Fig.  6a), and in reproductive cells have not undergone meiosis (Fig.  4m). The Fertile shoots (Fig.  6b). SeMoLF Y expression within the stro- 1 (F1) stage is characterized by the exposure of half of bili is detected in the flanks of the meristem, microphyll the strobilus above the leaf sheath and by mature spores primordia, sporangia primordia as well as in sporan- (Fig.  4i, n). The Fertile 2 (F2) stage is characterized by gial tissue and young microphylls (Fig.  6b). When the fully exposed strobilus with leaves only at their base and sporangium wall and stalk are differentiated, SeMoLF Y by mature spores (Fig.  4j, o). In both stages, F1 and F2 expression is broad in all the cells but later in devel- elaters (i.e., sterile elongated cells that uncoil in response opment it starts to be preferentially expressed in the to changes in humidity, assisting in the dispersal of the sporangial wall and the sporogenous tissue and turned spores) can be easily identified (Fig. 4n, o). off in the stalk (Fig.  5c, d). During sporangia matura- Expression of LF Y homologs was evaluated in all tion, when the tapetum is differentiated, the expres - stages described above. We found two LF Y copies in the sion of SeMoLF Y can be detected in the tapetum itself monomorphic A. raddianum (Fig.  1), both genes have and in the spores before meiosis (Fig.  6e–-f). Dur- identical expression patterns with higher expression in ing the formation of tetrads, expression of SeMoLF Y fertile stages when compared to fiddleheads and ster - becomes restricted to the spores and the sporangial ile leaves (Fig. 5a; Additional file  7: Fig S4) with a lower Fig. 5 Semi‑ quantitative expression of LFY genes by RT‑PCR in selected fern species. a Expression of the monomorphic Adiantum raddianum (Pteridaceae) paralogs b Expression of the holodimorphic E. giganteum (Equisetaceae) copies. Fd: Fiddlehead, S: Sterile stage, F: Fertile stage. ‑ C: negative control. Actin was use as the positive control R odríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 9 of 16 Fig. 6 Expression of Selaginella moellendorffii LFY genes by in situ hybridization. a Vegetative shoot apical meristem (SAM). b Reproductive SAM with arising sporophylls and sporangia primordia. c–d Consecutive developing stages of the sporangium with a developing stalk and a forming sporangial wall surrounding the sporogenous tissue. e Two sporangial wall layers enclosing the sporogenous tissue prior to meiosis. f Two sporangial wall layers and tapetum. g Sporangium after meiosis h Mature sporangium with collapsed wall layers and tapetum and fully developed spores. li = ligule, m = microphyll, sti = sporangia, sw = sporangial wall, sy = sporophyll, t = tapetum, ts = tetrads, s = spores. Black arrowheads indicate the SAM. Scale bars = 100 µm Rodríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 10 of 16 wall (Fig.  6g), and finally expression is only detected in angiosperms. Our analysis does not recover the ancient the spores (Fig. 6h). type I and type II duplications in embryophytes, perhaps because an extensive sampling in bryophytes was not tar- Discussion geted in our study. We were also unable to retrieve the Unlike several other transcription factor families impor- previous reported two Osmunda sequences claimed to tant for plant development, the LFY gene lineage is usu- have promiscuous DNA binding types [6]. In our analy- ally retained primarily as a single copy in streptophytes sis, we were only able to identify DNA binding type I and across land plants [5]. A single large scale duplica- motifs in lycophyte and fern sequences, the same motif tion has been identified in this gene lineage linked to the type retained throughout tracheophytes. Our phyloge- diversification of seed plants resulting in the NDLY and netic analysis also recovers several local duplications LFY paralogs, followed by the loss of NDLY in angio- in lycophytes and ferns, some of which had been previ- sperms while retained in gymnosperms [35, 38, 45]. In ously identified [5, 39], and some are new in Isoetaceae this scenario, all other land plants retained single copy in lycophytes and in Equisetales in ferns. Many of the pre-duplication LFY genes with few species specific duplications occurring in these lineages could be possibly duplication events [5, 7, 39, 40, 46]. Interestingly, despite linked to reported whole genome duplication events [49]. this unusual strong selection acting upon LFY single copy However, most of the duplications found correspond to genes, their roles are divergent in different plant lineages species-specific paralogs and may be the result of other [5, 40]. Functional data points to plesiomorphic roles of mechanisms such as tandem repeat replication or retro- LFY genes in the control of cell divisions in the zygote transposition as copy numbers can be very different in to form the sporophyte in bryophytes [42]. Added evo- closely related species. lutionary roles include the maintenance of indeterminate Understanding the evolution and the putative roles of cell fate in both the gametophyte and the sporophyte in LFY genes in lycophytes and ferns is important as the ferns [39]. New key functions in reproductive transition data will help to clarify the roles of LFY homologs present are part of the functional repertoire of LFY homologs in in tracheophytes that are absent in non-vascular plants, seed plants. Namely, both LFY and NDLY are recruited and the plesiomorphic functions present in tracheo- in male and female cone development in gymnosperms phytes prior to the emergence of NDLY and LFY paral- [35–38, 45] and LFY homologs control floral meristem ogs in seed plants. The most comprehensive analysis to identity in angiosperms [1, 47]. Finally, LFY homologs date of fern LFY genes is that of Plackett et al. [39], which have also been recruited in maintaining the indetermi- identified two recent LFY paralogs in Ceratopteris richar- nate lateral inflorescence meristems in monocots and in dii that act partially redundantly in maintaining the inde- forming compound leaves in several legumes [27, 28, 33]. terminacy of the shoot apex, both in the gametophyte These data challenge the notion of gene duplication as a and the sporophyte. However, this same study showed major driver of gene functional diversification. Instead, that expression of at least one LFY paralog, CrLFY1 in it points to changes in protein sequences, specifically in C. richardii expands beyond the shoot apices, specifi - motifs responsible for DNA binding specificity, as key cally to newly emerged leaves as well as in young fertile features in the evolution of new functions [5]. To date, leaves. This coincides with previous reports identifying the changes in the interaction capabilities of LFY pro- higher expression of C. richardii LFY genes in shoot tips teins in different plant lineages have been directly linked and circinate reproductive leaves [7]. Similarly, the active to sequence changes of LFY genes [5, 40]. expression of LFY paralogs in the lycophyte Isoetes sinen- Contrary to this idea, recent comprehensive phylo- sis in microsporangia and megasporangia points to puta- genetic analyses have identified LFY duplications pre- tive roles in sporogenesis [41], and what has been broadly dating land plant diversification, or at least mosses and considered as the reproductive tissue formation by the liverworts, resulting in ancient duplicates that were lost formation of specialized sporangia with spore mother in tracheophytes [6, 48]. The hypothesis of the existence cells undergoing meiosis. Our data here suggest that of ancient duplicates predating land plant diversification several LFY homologs in lycophytes (S. moellendoriffi ) is supported with the finding of a few moss LFY genes and ferns (Adiantum raddianum and Equisetum gigan- with type I motifs and a few liverwort homologs with teum) are highly, and preferentially expressed in fertile type II motifs [6]. Less conflicting arguments have been stages and the expression is found to be correlated with posed around the evolution of LFY genes in tracheo- reproductive tissues (Figs. 5 and 6). If taken together, the phytes, as most phylogenetic analyses to date recover expression data available, including that presented here, single copy genes with type I motifs in lycophytes and raises the possibility that LFY genes play roles in the ferns and the already mentioned NDLY/LFY duplication formation of sporangia and, if so, in the specialization in seed plants with the retention of only LFY copies in of reproductive leaf derived tissues. If so, this function R odríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 11 of 16 is not new or exclusive to seed plant LFY homologs but 3) DNA binding capabilities have been exclusively was already present in the ancestor of vascular plants, tested with flowering plant partners, such as AG and because it is also present in lycophytes and ferns. Inter- AP1 [9, 40, 51]. Most of the key domains of LFY for estingly, expression of the P. patens PpLFY1 and PpLFY2 floral function have been determined. LFY proteins copies was detected in the sporophytes, beyond the early in gymnosperms, ferns, lycophytes or bryophytes zygote divisions and until the mature sporophyte [42]. A have conserved and easy to identify, oligomerization role of LFY homologs in sporophyte development was and DNA binding domains, but exhibit some non- in fact postulated originally in P. patens [42], but a role neutral changes, sometimes affecting key positions in the induction of the reproductive phase was ruled out [6, 41]. For instance, the C. richardii LFY proteins based on the lack of complementation of Arabidopsis lfy changes in amino acid positions within the N-ter- mutants using PpLFY genes [40]. minal and C-terminal domains have proven to be There are three major aspects to consider in the critical for the recovery of lfy mutants in Arabidopsis assessment of the LFY functional capabilities outside of [40]. angiosperms: Our alignment, as well as those of Yang et  al. [41] 1) Most of the functional data comes from com- and Gao et  al. [6], show that these changes are com- plementation analyses of Arabidopsis lfy mutants mon across lycophytes and ferns. In parallel, we have [37, 40, 41, 45, 50]). The observations showed that also identified motifs 9 in the DNA binding domain and sequences closely related to Arabidopsis frequently motif 10 in the SAM domain as fern-specific. In addition, recovered the wild type phenotype, and that, the res- studies on the putative protein–protein interactions and cue of lfy mutants by gymnosperm, fern and bryo- DNA-binding properties outside angiosperms are lack- phyte sequences decreased in efficiency. Although ing and in turn, LFY partners in lycophytes and ferns informative, complementation experiments only remain unknown making it difficult to assess if protein allow us to address similarities of the functional shifts have changed dimerization and/or DNA binding capabilities of these genes when compared to the processes endogenously. One critical aspect here is that Arabidopsis canonical LFY gene. Therefore, these several genes that are turned on upstream of LFY, such as experiments are very limited to reach conclusions FLOWERING LOCUS T/TERMINAL FLOWER 1 (FT/ regarding the genes endogenous roles. TFL1) and MADS-box proteins acting as flowering inte - 2) Fern LFY genes have pleiotropic early roles in the grators have lycophyte and fern homologs (Article under gametophyte and early sporophyte development, review and [44]). For example, the type II classic MADS- preventing the observation and analyses of their box genes from Selaginella moellendoriffi , SmMADS1 functions during the reproductive transition. The and SmMADS6, are expressed from the earliest stages comparative expression data available, does not rule of sporangia development to mature spores and, there- out the possibility that LFY genes play roles in the fore, overlap with SmLFY expression patterns [44]. More reproductive transition or at least in sporangia for- importantly, these MADS-box transcription factors seem mation in lycophytes and ferns, like those recorded to be expressed in an overlapping manner to LFY in the in seed plants. If so, an early acquisition of repro- sporophytes [52–56]. Future studies could aim at assess- ductive roles for LFY homologs could be traced back ing spatial temporal expression of MADS-box genes as at least to the common ancestor of tracheophytes. well as assessing protein partners in lycophytes and ferns Also unclear is the putative contribution of species- to assess what interactions were in place during sporo- specific paralogs to gene functional diversification. phyte growth and are not exclusive to angiosperms, simi- Our data and that of Yang et  al. [41] and Plackett lar to evaluations already done in gymnosperms [51, 57]. et  al. [39] points to some overlap between copies Understanding the rewiring of genetic regulatory net- resulting in redundancy but also recover important works for these critical developmental control genes will expression differences between paralogs suggesting allow an understanding of the apparent neofunctionaliza- some specialization among paralogs. Only the eval- tion of LFY that emerges in different plant lineages. uation of the endogenous gene function in ferns and lycophytes, will allow an examination of the con- Conclusions tribution of local duplicates to reproductive transi- Large-scale and species-specific LFY duplications have tion and in turn, assess the functional evolution of been identified in lycophytes and ferns. Some of these the LFY gene lineage, one of the core developmental may be the result of ancient WGD as is the case for Equi- genes in the evolution of land plants. setales, while the remaining species-specific copies are probably a result of recent polyploidy or hybridization Rodríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 12 of 16 events. We recovered little protein variation among The tissue was ground using liquid nitrogen and total LFY proteins across land plants. The type I motif that RNA extraction was carried out using PureLink Plant is shared by all tracheophytes is recovered in the DNA RNA Reagent (Invitrogen, USA). The total RNA was binding domain of ferns and lycophytes as expected; but quantified using a Nanodrop (Thermo, USA) and visu - the new motifs 9 and 10 are found to be fern-specific ally inspected in a 1% agarose gel stained with ethidium and located in LFY_SAM and the DNA Binding domain, bromide. Quality was assessed based on the presence respectively. As both domains are key for DNA binding, and integrity of ribosomal RNA bands. RNA extrac- further analyses are needed to elucidate downstream tions with concentrations over 200  ng/ul were sent to factors, and partner proteins that interact during sporo- the sequencing facility (Macrogen, South Korea). RNA- phyte growth in lycophytes and ferns. LFY genes in the seq experiments were conducted using a TruSeq mRNA lycophyte S. moellendoriffi and the two fern species A. library construction kit (Illumina) and sequenced in a raddianum and E. giganteum show a wide expression HiSeq2000 instrument producing 100 base paired-end among tissues. However, these expression patterns were reads. The transcriptomes were assembled de novo with found to be relatively higher in differentiating young fer - Trinity V2 at the Centro Nacional de Secuenciación tile stages and early sporangia developmental stages that Genómica (CNSG), following default settings [58]. Read then decreases once mature stages are reached. In addi- cleaning was performed with prinseqlite v0.20.4 with a tion, functional diversification among paralogs is found quality threshold of Q35 [59]. Contig metrics for each for E. giganteum that matches with that reported for C. species can be find on Additional file 3: Table S1. richardii LFY copies. Roles in reproduction have been reported in gymnosperms and angiosperms, but as the Gene phylogenies assessment of reproductive function across LFY genes To isolate LEAFY genes, sequences from Arabidopsis have been based on the ability of homologs to comple- thaliana, Selaginella moellendoriffi , Physcomitrium pat- ment Arabidopsis lfy mutants, it has not been possible to ens and Ceratopteris richardii LFY homologs were used clarify the role of LFY outside seed plants. Therefore, the as queries in BLASTN searches in all available ferns evaluation of endogenous gene function is key to know and lycophyte transcriptomes and genomes. BLAST the contribution of these genes and their duplicates in searches were made in Phytozome (http:// www. phyto reproductive transition. Our results allow us to propose zome. net/), NCBI, (https:// blast. ncbi. nlm. nih. gov/ Blast. a putative role for lycophyte and fern LFY homologs in cgi), and Fernbase (https:// www. fernb ase. org/), as well reproductive tissue diffenretiation, indicating that this as in transcriptome database OneKP (http:// www. bioin function may be present in the LFY gene lineage prior fodata. org/ Blast 4OneKP/). Most hits were retrieved as to the diversification of seed plants. Even though further complete coding sequence (CDS) with some partial CDS investigations are needed, our report represents a step derived from transcriptomic data. Partial sequences forward in the assessment of LFY genes in a reproductive were included only if they had 50% of the ca. 420 amino context outside seed plants. acids reported for the gene [1], and at least one of the two LFY domains [40]. Using the same strategy, we iso- Methods lated sequences from our own generated transcriptomes Transcriptome generation from Adiantum raddianum, Anemia villosa, Equise- Four species representing leaf monomorphism (Adi- tum giganteum and E. bogotense. These sequences were antum raddianum), hemidimorphism (Anemia vil- deposited in GenBank and can be found under the fol- losa) and holodimorphism (Equisetum bogotense and E. lowing accession numbers (MW219613, MW219614, giganteum) were chosen based on material availability MW219618–MW219620, MW219623–MW219625, of growing populations close to Medellin (Colombia). MW820861–MW820863). The species vouchers were deposited in the herbarium A comprehensive LEAFY matrix was built primar- of University of Antioquia (HUA) (C. Rodríguez-Pelayo ily using that of Sayou et  al. [5] but integrating all new 1–5). The material was collected in the field and was flash homologs identified here. All sequences were compiled in frozen in liquid nitrogen. For A. raddianum and A. vil- Bioedit (http:// www. mbio. ncsu. edu/ bioed it/ bioed it. html) losa the plant parts collected included fiddleheads as well and manually edited to exclusively keep the CDS for all as fertile and sterile pinnae at different developmental transcripts. Nucleotide sequences were subsequently stages. For E. giganteum and E. bogotense fertile and ster- aligned using the TranslatorX online platform (http:// ile shoot apical portions at different developmental stages trans latorx. co. uk/) implementing the online MAFFT were collected. A detailed morphological description of alignment algorithm [60] with default settings. The align - the stages included for each species can be found below ment was refined manually considering the SAM_LFY in the RT-PCR expression section. and DNA Binding domains reported as conserved in R odríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 13 of 16 LFY genes and can be found in Additional file  4: Fig. S3. Thermal cycling profiles followed an initial denatura - Maximum likelihood (ML) phylogenetic analyses using tion step (94 °C for 30 s), an annealing step (50–62 °C for the nucleotide sequences were performed in RaxML- 30  s) and an extension step with polymerase (72  °C for HPC2 BlackBox [61] through the CIPRES Science Gate- up to 1 min) repeated for 30–40 amplification cycles. As way [62]. Bootstrapping parameters were set for 1000 endogenous controls we tested Actin1 [65] as a control Bootstrap (BS) iterations. The tree was rooted with for all samples. Finally, the PCR products were run on a Algae LFY genes. Trees were observed and edited using 1.0% agarose gel stained with ethidium bromide and digi- FigTree v1.4.4 [63]. All the retrieved fern and lycophytes tally photographed using a Whatman Biometra BioDoc sequences can be found in Additional file 5: Table S2. Analyzer. To provide a more “quantitative” analysis of band brightness we converted our raw image data using Identification of protein domains and motifs imageJ. We counted the number of pixels and compared To detect both reported and new conserved motifs in it to the ACTIN band. Visualization of replicates is found LEAFY protein sequences across land plants we selected in Additional file 7: Fig. S4. a total of 34 sequences representing all subclades within the gene lineage. Sequences were permanently translated Developmental series of Equisetum giganteum and uploaded as amino acids to the online Multiple Em Sterile and reproductive shoots were collected in the field for Motif Elicitation (MEME) server (http:// meme- suite. and immediately fixed in formaldehyde–acetic acid– org/) and run to find up to 10 motifs with the default ethanol (FAA; 3.7% formaldehyde: 5% glacial acetic acid: options [64]. The motifs retrieved by MEME are reported 50% ethanol). For light microscopy, fixed material was according to their statistical significance. The MEME manually dehydrated through an alcohol–histochoice suite finds, in the given sequences, the most statistically series and embedded in Paraplast X-tra (Fisher Health- significant (low E value) motifs first. The E value of a care, Houston, Texas, USA). The samples were sec - motif is based on its log likelihood ratio, width, sites, and tioned at 10–20  μm with an RM2125 RTS (USA) rotary the size of the set. We numbered the motifs following the microtome. Sections were stained with Johansen’s safra- statistical significance resulting in the analyses. nin and 0.5% Astra Blue [66] and mounted in Permount (Fisher Scientific, Pittsburgh, Pennsylvania, USA). Sec - Expression analyses by Reverse Transcriptase PCR (RT‑PCR) tions were viewed and digitally photographed with a To examine and compare the expression patterns of LFY Zeiss Axioplan compound microscope and Zeiss stereo genes in monomorphic, hemidimorphic and holodimor- microscope equipped with an AxioCamERc5s digital phic ferns we dissected different plant parts and targeted camera with ZEN software. a range of developmental stages for Adiantum rad- dianum, and Equisetum giganteum, as described in the In situ hybridization results section of RT-PCR expression. In situ hybridization experiments were performed for We dissected and flash froze meristems, fiddleheads, the Selaginella moellendori L ffi FY homolog. Thus, we and fertile and sterile pinnae at the developmental stages used meristematic vegetative portions and strobili and described previously for the two ferns species. Total RNA followed sporangium development stages established by was prepared from dissected tissues using the same pro- previous descriptions in other Selaginella species [12, 44, tocol described above. RNA samples were treated with 67, 68]. The material was fixed for 3  h under vacuum in DNAseI (Roche, Basel, Switzerland) to remove DNA freshly prepared FAA (50% ethanol, 3.7% formaldehyde contamination and later quantified with a NanoDrop and 5% glacial acetic acid). The material was dehydrated 2000 (Thermo Scientific, Waltham, MA, USA). Three through an alcohol‐histochoice series and embedded in micrograms of RNA were used as a template for cDNA Paraplast X‐tra (Fisher, Waltham, MA, USA) with opti- synthesis (SuperScript III RT, Invitrogen) using OligodT mizations made by A. Vasco (i.e., fixation time 3 h; infil - primers. The cDNA was used undiluted for amplification tration solution changes every 3  h and hybridization reactions by RT-PCR. To ensure specificity for ampli - temperature fixed at 55  °C). Samples embedded were fication of each copy, the primers were designed in the maintained at 4 °C until use. The tissue was sectioned at regions outside of the conserved domains (Additional 8–10 μm on a Microm HM3555 rotary microtome. file 6: Table S3). Each amplification reaction incorporated DNA template for the S. moellendori S ffi eMoLFY probe 9 μL of EconoTaq (Lucigen, Middleton, WI, USA), 6 μL of was obtained by PCR amplification of a 500 bp fragment. nuclease-free water, 1 μL of BSA (bovine serum albumin) The fragment was cleaned using the QIAquick PCR puri - (5 μg/mL), 1 μL of Q solution (betaine 5 μg/μL), 1 μL of fication kit (Qiagen, Valencia, CA, USA). Digoxigenin forward primer (10 mm), 1 μL of reverse primer (10 mm), labeled RNA probe was prepared using T7polymerase and 1 μL of template cDNA, for a total reaction of 20 μL. (Roche, Switzerland), murine RNAse inhibitor (New Rodríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 14 of 16 Authors’ contributions England Biolabs, Ipswich, MA, USA), and RNA labeling CRP, NPM, BAA and AVG planned and designed the research. All authors mix (Roche, Switzerland) according to the manufacturer performed experiments, analyzed the data and wrote the final version of the protocols. In  situ hybridization following the protocol manuscript. All authors read and approved the final manuscript. that has been described previously optimizing incuba- Funding tion temperatures for each probe and tissue [5, 69, 70]. This study was funded by the Convocatoria de Sostenibilidad 2018–2019 and The sections were hybridized overnight at 55  °C. In  situ the Fundación para la Promoción de la Investigación y la Tecnología from Banco de la Republica (Project Number 4413). hybridized sections were subsequently dehydrated and permanently mounted in Permount (Fisher, Waltham, Availability of data and materials MA, USA). All sections were digitally photographed Sequences generated in this work were deposited in GenBank and can be found under the following Accession Numbers: MW219613, MW219614, using a Zeiss Axioplan microscope equipped with a MW219618–MW219620, MW219623–MW219625, MW820861–MW820863. Nikon DXM1200C digital camera. Sense probe controls are shown in Additional file 8: Fig. S5). Declarations Ethical approval and consent to participate Abbreviations The work was done with plants, mostly commonly cultivated and not under LFY: LEAFY; NDLY: NEEDLY; FLO: FLORICAULA; AGL24: AGAMOUS‑LIKE 24; any threat, so animal ethical approval and consent to participate are not appli‑ SVP: SHORT VEGETATIVE PHASE; SOC1: SUPRESSOR OF OVEREXPRESSION OF cable. As some plant species are cultivated and the ones collected are under CONSTANS; AP1: APETALA1; SP: SEPALLATA ; PI: PISTILLATA ; AP3: APETALA3; AG: collecting permission “Resolución 0524 del 27 de mayo del 2014”, Autoridad AGAMOUS; BSA: Bovine serum albumin; Q: Betaine; ML: Maximum likelihood; Nacional de Licencias Ambientales (ANLA). BS: Bootstrap; MEME: Multiple Em for Motif Elicitation; qRT‑PCR: Quantitative polymerase chain reaction; RT‑PCR: Reverse transcription polymerase chain Consent for publication reaction; SAM: Shoot apical meristem; WGD: Whole genome duplication. Authors declare consent for publication. Competing interests Supplementary Information Authors declare no competing interests on this research. The online version contains supplementary material available at https:// doi. org/ 10. 1186/ s13227‑ 021‑ 00188‑9. Author details Facultad de Ciencias Exactas y Naturales, Instituto de Biología, Universidad de Antioquia, Medellín, Colombia. The New York Botanical Garden, Bronx, Additional file 1: Figure S1. ML analysis of the LFY gene family. a. Sum‑ 3 4 NY, USA. Botanical Research Institute of Texas, Fort Worth, TX, USA. Centro mary tree including sequences from algae, bryophytes and tracheophytes. Nacional de Secuenciación Genómica, Facultad de Medicina, Sede de Investi‑ b. ML analysis of the LFY family including algae and land plant sequences. gación Universitaria, Universidad de Antioquia, Medellín, Colombia. Yellow stars indicate large duplication events. Number on each node indicate the bootstrap value (BS). Black arrowheads point to sequences Received: 23 September 2021 Accepted: 18 December 2021 isolated in this study. The colors correspond to the conventions on the bottom left. Scale: 0.3 Additional file 2: Figure S2. Protein sequences of the LFY family. For a total of 34 representative land plant sequences belonging to: Arabidop‑ sis thaliana, Ceratopteris richardii, Azolla filiculoides, Salvinia cucullata, References Equisetum giganteum, E. bogotense, Adiantum raddianum, Anemia 1. Weigel D, Alvarez J, Smyth DR, Yanofsky MF, Meyerowitz EM. LEAFY con‑ villosa, Selaginella moellendorffii, Physcomitrium patens and several trols floral meristem identity in Arabidopsis. Cell. 1992;69:843–59. algae species. The two characteristic domains of LFY proteins reported by 2. Blazquez MA, Soowal LN, Lee I, Weigel D. LEAFY expression and flower Sayou et al. [5, 9] are boxed. Blue arrowheads point to the key positions for initiation in Arabidopsis. Development. 1997;124:3835–44. DNA binding reported by Sayou et al. [5] .Black arrowheads point to the 3. Hempel FD, Weigel D, Mandel MA, Ditta G, Zambryski PC, Feldman LJ, conserved section among fern sequences. et al. Floral determination and expression of floral regulatory genes in Arabidopsis. Development. 1997;124:3845–53. Additional file 3: Table S1. Contig metrics. 4. Siriwardana NS, Lamb RS. The poetry of reproduction: the role of LEAFY in Additional file 4: Figure S3. Complete alignment in fasta for LFY Arabidopsis thaliana flower formation. Int J Dev Biol. 2012;56:207–21. homologs. 5. Sayou C, Monniaux M, Nanao MH, Moyroud E, Brockington SF, Thévenon Additional file 5: Table S2. LFY sequences complete data set. E, et al. A promiscuous intermediate underlies the evolution of LEAFY DNA binding specificity. Science. 2014;343:645–8. Additional file 6: Table S3. Primer sequences for fern and lycophyte LFY 6. Gao B, Chen M, Li X, Zhang J. Ancient duplications and grass‑specific genes used here for RT‑PCR and ISH. transposition influenced the evolution of LEAFY transcription factor Additional file 7: Figure S4. Quantification of LFY homologs expression genes. Commun Biol. 2019;2(1):10. using Image J onto gel images with three biological replicates. 7. Himi S, Sano R, Nishiyama T, Tanahashi T, Kato M, Ueda K, et al. Evolu‑ tion of MADS‑box gene induction by FLO/LFY genes. J Mol Evol. Additional file 8: Figure S5. Sense experiments for ISH in the meristem 2001;53:387–93. (SAM) and in reproductive tissue (See also Zumajo‑ Cardona et al. [44]). 8. Lu S, Wang J, Chitsaz F, Derbyshire MK, Geer RC, Gonzales NR, et al. CDD/ SPARCLE: the conserved domain database in 2020. Nucleic Acids Res. 2020;48:D265–8. Acknowledgements 9. Sayou C, Nanao MH, Jamin M, Posé D, Thévenon E, Grégoire L, et al. A We thank the staff at The New York Botanical Garden especially Cecilia SAM oligomerization domain shapes the genomic binding landscape of Zumajo‑ Cardona and Tynisha Smalls for help in the standardization of lab the LEAFY transcription factor. Nat Commun. 2016;7:11222. protocols. We thank the OneKP repository staff for facilitating access to the 10. Carpenter R, Coen ES. Floral homeotic mutations produced by transpo‑ online database. We thank members of the Plant Evo Devo lab for help in the son‑mutagenesis in Antirrhinum majus. Genes Dev. 1990;4:1483–93. laboratory and helpful discussions. R odríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 15 of 16 11. Coen ES, Romero J, Doyle S, Elliott R, Murphy G, Carpenter R. floricaula: a containing vegetative shoot and undifferentiated male cone primordia. homeotic gene required for flower development in Antirrhinum majus. Planta. 1998;206:619–29. Cell. 1990;63:1311–22. 36. Shindo S, Sakakibara K, Sano R, Ueda K, Hasebe M. Characterization of a 12. Schultz EA, Haughn GW. LEAFY, a homeotic gene that regulates inflores‑ FLORICAULA/LEAFY homologue of Gnetum parvifolium and its implica‑ cence development in Arabidopsis. Plant Cell. 1991;3:771–81. tions for the evolution of reproductive organs in seed plants. Int J Plant 13. Moyroud E, Kusters E, Monniaux M, Koes R, Parcy F. LEAFY blossoms. Sci. 2001;162:1199–209. Trends Plant Sci. 2010;15:346–52. 37. Moyroud E, Monniaux M, Thévenon E, Dumas R, Scutt CP, Frolich M, Parcy 14. Parcy F, Nilsson O, Busch MA, Lee I, Weigel D. A genetic framework for F. A link between LEAFY and B‑ gene homologues in Welwitschia mirabilis floral patterning. Nature. 1998;395:561–6. sheds light on ancestral mechanisms prefiguring floral development. 15. Lee J, Oh M, Park H, Lee I. SOC1 translocated to the nucleus by interaction New Phytol. 2007;216:469–81. with AGL24 directly regulates LEAFY. Plant J. 2008;55:832–43. 38. Vázquez‑Lobo A, Carlsbecker A, Vergara‑Silva F, Alvarez‑Buylla ER, Piñero 16. Liu C, Xi W, Shen L, Tan C, Yu H. Regulation of floral patterning by flower ‑ D, Engström P. Characterization of the expression patterns of LEAFY/ ing time genes. Dev cell. 2009;16:711–22. FLORICAULA and NEEDLY orthologs in female and male cones of the 17. Gregis V, Andrés F, Sessa A, Guerra RF, Simonini S, Mateos JL, et al. Identifi‑ conifer genera Picea, Podocarpus, and Taxus: implications for current evo‑ cation of pathways directly regulated by SHORT VEGETATIVE PHASE dur‑ devo hypotheses for gymnosperms. Evol Dev. 2007;9:446–59. ing vegetative and reproductive development in Arabidopsis. Genome 39. Plackett AR, Conway SJ, Hewett Hazelton KD, Rabbinowitsch EH, Lang‑ Biol. 2013;14:R56. dale JA, Di Stilio VS. LEAFY maintains apical stem cell activity during shoot 18. Kaufmann K, Wellmer F, Muiño JM, Ferrier T, Wuest SE, Kumar V, et al. development in the fern Ceratopteris richardii. Elife. 2018;7: e39625. Orchestration of Floral Initiation by APETALA1. Science. 2010;328:85–9. 40. Maizel A, Busch MA, Tanahashi T, Perkovic J, Kato M, Hasebe M, et al. 19. Moyroud E, Minguet EG, Ott F, Yant L, Posé D, Monniaux M, et al. The floral regulator LEAFY evolves by substitutions in the DNA binding Prediction of regulatory interactions from genome sequences using a domain. Science. 2005;308:260–3. biophysical model for the Arabidopsis LEAFY transcription factor. Plant 41. Yang T, Du M, Guo Y, Liu X. Two LEAFY homologs ILFY1 and ILFY2 Cell. 2011;23:1293–306. control reproductive and vegetative developments in Isoetes L. Sci Rep. 20. Winter CM, Austin RS, Blanvillain‑Baufumé S, Reback MA, Monniaux M, Wu 2017;7:225. M‑F, et al. LEAFY target genes reveal floral regulatory logic, cis motifs, and 42. Tanahashi T, Sumikawa N, Kato M, Hasebe M. Diversification of gene a link to biotic stimulus response. Dev cell. 2011;20:430–43. function: homologs of the floral regulator FLO/LFY control the first 21. Han Y, Zhang C, Yang H, Jiao Y. Cytokinin pathway mediates APETALA1 zygotic cell division in the moss Physcomitrella patens. Development. function in the establishment of determinate floral meristems in Arabi‑ 2005;132:1727–36. dopsis. Proc Natl Acad Sci USA. 2014;111:6840–5. 43. PPGI. A community‑ derived classification for extant lycophytes and ferns. 22. Winter CM, Yamaguchi N, Wu M‑F, Wagner D. Transcriptional programs J Syst Evol. 2016;54:563–603. regulated by both LEAFY and APETALA1 at the time of flower formation. 44. Ambrose BA, Smalls TL, Zumajo‑ Cardona C. All type II classic MADS‑ Physiol Plant. 2015;155:55–73. box genes in the lycophyte Selaginella moellendorffii are broadly yet 23. William DA, Su Y, Smith MR, Lu M, Baldwin DA, Wagner D. Genomic discretely expressed in vegetative and reproductive tissues. Evol Dev. identification of direct target genes of LEAFY. Proc Natl Acad Sci USA. 2021;23:215–30. 2004;101:1775–80. 45. Mouradov A, Glassick T, Hamdorf B, Murphy L, Fowler B, Marla S, et al. 24. Southerton SG, Strauss SH, Olive MR, Harcourt RL, Decroocq V, Zhu X, NEEDLY, a Pinus radiata ortholog of FLORICAULA/LEAFY genes, expressed et al. Eucalyptus has a functional equivalent of the Arabidopsis floral in both reproductive and vegetative meristems. Proc Natl Acad Sci USA. meristem identity gene LEAFY. Plant Mol Biol. 1998;37:897–910. 1998;95:6537–42. 25. Wada M, Cao Q, Kotoda N, Soejima J, Masuda T. Apple has two ortho‑ 46. Frohlich MW, Parker DS. the mostly male theory of flower evolutionary logues of FLORICAULA/LEAFY involved in flowering. Plant Mol Biol. origins: from genes to fossils. Syst Bot. 2000;25:155–70. 2002;49:567–77. 47. Nilsson O, Weigel D. Modulating the timing of flowering. Curr Opin 26. Yu Q, Moore PH, Albert HH, Roader AHK, Ming R. Cloning and characteri‑ Biotechnol. 1997;8:195–9. zation of a FLORICAULA/LEAFY ortholog, PFL, in polygamous papaya. Cell 48. Brunkard JO, Runkel AM, Zambryski PC. Comment on “A promiscuous Res. 2005;15:576–84. intermediate underlies the evolution of LEAFY DNA binding specificity.” 27. Hofer J, Turner L, Hellens R, Ambrose M, Matthews P, Michael A, et al. Science. 2015;347:621–621. UNIFOLIATA regulates leaf and flower morphogenesis in pea. Curr Biol. 49. Huang C‑H, Qi X, Chen D, Qi J, Ma H. Recurrent genome duplication 1997;7:581–7. events likely contributed to both the ancient and recent rise of ferns. J 28. Wang H, Chen J, Wen J, Tadege M, Li G, Liu Y, et al. Control of compound Integr Plant Biol. 2019;62:433–55. leaf development by FLORICAULA/LEAFY ortholog SINGLE LEAFLET1 in 50. Shiokawa T, Yamada S, Futamura N, Osanai K, Murasugi D, Shinohara K, Medicago truncatula. Plant Physiol. 2008;146:1759–72. et al. Isolation and functional analysis of the CjNdly gene, a homolog 29. Zhao W, Chen Z, Liu X, Che G, Gu R, Zhao J, Wang Z, Hou Y, Zhang X. in Cryptomeria japonica of FLORICAULA/LEAFY genes. Tree Physiol. CsLFY is required for shoot meristem maintenance via interaction with 2008;28:21–8. WUSCHEL in cucumber (Cucumis sativus). New Phyt. 2017;218:344–56. 51. Silva CS, Puranik S, Round A, Brennich M, Jourdain A, Parcy F, et al. Evolu‑ 30. Pouteau S, Nicholls D, Tooke F, Coen E, Battey N. The induction and main‑ tion of the plant reproduction master regulators LFY and the MADS tenance of flowering in Impatiens. Development. 1997;124:3343–51. transcription factors: the role of protein structure in the evolutionary 31. Molinero‑Rosales N, Jamilena M, Zurita S, Gómez P, Capel J, Lozano R. development of the flower. Front Plant Sci. 2016;6. https:// doi. org/ 10. FALSIFLORA, the tomato orthologue of FLORICAULA and LEAFY, controls 3389/ fpls. 2015. 01193/ full. Accessed 29 Dec 2020. flowering time and floral meristem identity. Plant J. 1999;20:685–93. 52. Münster T, Pahnke J, Rosa AD, Kim JT, Martin W, Saedler H, et al. Floral 32. Ahearn KP, Johnson HA, Weigel D, Wagner DR. NFL1, a Nicotiana tabacum homeotic genes were recruited from homologous MADS‑box genes LEAFY‑like gene, controls meristem initiation and floral structure. Plant preexisting in the common ancestor of ferns and seed plants. Proc Natl Cell Physiol. 2001;42:1130–9. Acad Sci USA. 1997;94:2415–20. 33. Rao NN, Prasad K, Kumar PR, Vijayraghavan U. Distinct regulatory role 53. Svensson ME, Johannesson H, Engström P. The LAMB1 gene from the for RFL, the rice LFY homolog, in determining flowering time and plant clubmoss, Lycopodium annotinum, is a divergent MADS‑box gene, architecture. Proc Natl Acad Sci USA. 2008;105:3646–51. expressed specifically in sporogenic structures. Gene. 2000;253:31–43. 34. Sharma B, Meaders C, Wolfe D, Holappa L, Walcher‑ Chevillet C, Kramer 54. Tanabe Y, Uchida M, Hasebe M, Ito M. Characterization of the Selaginella EM. Homologs of LEAFY and UNUSUAL FLORAL ORGANS promote the remotifolia MADS‑box gene. J Plant Res. 2003;116:69–73. transition from inflorescence to floral meristem identity in the Cymose 55. Huang Q, Li W, Fan R, Chang Y. New MADS‑box gene in fern: cloning Aquilegia coerulea. Front Plant Sci. 2019. https:// doi. org/ 10. 3389/ fpls. 2019. and expression analysis of DfMADS1 from Dryopteris fragrans. PLoS ONE. 01218/ full. Accessed 6 Nov 2020. 2014;9: e86349. 35. Mellerowicz EJ, Horgan K, Walden A, Coker A, Walter C. PRFLL–a Pinus radiata homologue of FLORICAULA and LEAFY is expressed in buds Rodríguez‑Pelayo et al. EvoDevo (2022) 13:2 Page 16 of 16 56. Ruiz‑Estévez M, Bakkali M, Martín‑Blázquez R, Garrido ‑Ramos MA. Dif‑ ferential expression patterns of MIKCC‑type MADS‑box genes in the endangered fern Vandenboschia speciosa. Plant Gene. 2017;12:50–6. 57. Wang Y‑ Q, Melzer R, Theißen G. Molecular interactions of orthologues of floral homeotic proteins from the gymnosperm Gnetum gnemon provide a clue to the evolutionary origin of ‘floral quartets.’ Plant J. 2010;64:177–90. 58. Grabherr MG, Haas BJ, Yassour M, Levin JZ, Thompson DA, Amit I, et al. Full‑length transcriptome assembly from RNA‑Seq data without a refer ‑ ence genome. Nat Biotechnol. 2011;29:644–52. 59. Schmieder R, Edwards R. Quality control and preprocessing of metagen‑ omic datasets. Bioinformatics. 2011;27:863–4. 60. Katoh K, Misawa K, Kuma K, Miyata T. MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Res. 2002;30:3059–66. 61. Stamatakis A, Hoover P, Rougemont J. A rapid bootstrap algorithm for the RAxML web servers. Syst Biol. 2008;57:758–71. 62. Miller MA, Pfeiffer W, Schwartz T. Creating the CIPRES science gateway for inference of large phylogenetic trees. Gateway Computing Environments Workshop (GCE), 2010: 1–8. 63. Rambaut A. FigTree v.1.4.4: Tree Figure Drawing Tool. 2018. http:// tree. bio. ed. ac. uk/ softw are/ figtr ee/. Accessed 29 Dec 2021. 64. Bailey TL, Williams N, Misleh C, Li WW. MEME: discovering and analyzing DNA and protein sequence motifs. Nucleic Acids Res. 2006;34:W369–73. 65. Brouwer P, Bräutigam A, Külahoglu C, Tazelaar AOE, Kurz S, Nierop KGJ, et al. Azolla domestication towards a biobased economy? New Phytol. 2014;202:1069–82. 66. Kraus JE, de Sousa HC, Rezende MH, Castro NM, Vecchi C, Luque R. Astra blue and basic fuchsin double staining of plant materials. Biotech Histo‑ chem. 1998;73:235–43. 67. Lyon FM. A study of the sporangia and gametophytes of Selaginella apus and Selaginella rupestris. Bot gaz. 1901;32:124–41. 68. Schulz C, Little DP, Stevenson DW, Bauer D, Moloney C, Stützel T. An overview of the morphology, anatomy, and life cycle of a new model species: the lycophyte Selaginella apoda (L.) spring. Int J Plant Sci. 2010;171:693–712. 69. Ambrose BA, Lerner DR, Ciceri P, Padilla CM, Yanofsky MF, Schmidt RJ. Molecular and genetic analyses of the Silky1 gene reveal conservation in floral organ specification between eudicots and monocots. Mol Cell. 2000;5:569–79. 70. Vasco A, Ambrose BA. Simple and divided leaves in ferns: exploring the genetic basis for leaf morphology differences in the genus Elaphoglos - sum (Dryopteridaceae). Int J Pl Sci. 2020;21:5180. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions

Journal

EvoDevo – Springer Journals

Published: Jan 8, 2022

Keywords: LEAFY; Ferns; Lycophytes; Reproductive transition; Sporogenesis

There are no references for this article.