The SPL‐family transcription factor MpSPL3 orchestrates the proper regulation of vegetative and reproductive programs in Marchantia polymorpha
Alisha Alisha, Artur Jarmolowski, Zofia Szweykowska‐Kulinska, Izabela Sierocka
TL;DR
This study shows how the MpSPL3 gene controls both vegetative growth and reproductive development in the liverwort Marchantia polymorpha.
Contribution
The paper reveals the first functional insights into MpSPL3's role in regulating developmental programs in Marchantia polymorpha.
Findings
Knockout of MpSPL3 causes growth retardation, disordered thallus morphology, and loss of gametangiophore formation.
Overexpression of the longer MpSPL3.1 isoform delays and reduces gametangiophore production.
Altered gene expression in MpSPL3 mutants affects vegetative development and germ cell specification.
Abstract
SQUAMOSA PROMOTER BINDING PROTEIN‐LIKE (SPL) genes encode plant‐specific transcription factors that are widely distributed across the plant kingdom. In angiosperms, the multimember SPL family regulates various biological processes, including vegetative‐to‐reproductive phase transition, inflorescence architecture, and lateral organ development. In contrast, the liverwort Marchantia polymorpha genome encodes only four SPL genes, with functional studies available only for microRNA‐targeted members, MpSPL1 and MpSPL2. MpSPL1 was shown to control the meristem dormancy to modulate the thallus architecture, whereas MpSPL2 was found to promote the transition from vegetative‐to‐reproductive phase. Here, we investigate the impact of the MpSPL3 gene on M. polymorpha development. We demonstrate that MpSPL3 influences coordination of the vegetative growth and the reproductive phase transition.…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7- —Narodowe Centrum Nauki10.13039/501100004281
- —Uniwersytet Jutra
- —Minigranty Doktoranckie
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsPlant Molecular Biology Research · Plant Reproductive Biology · Plant Taxonomy and Phylogenetics
INTRODUCTION
The successful colonization of the terrestrial environment by green, multicellular plants and their rapid evolution was one of the most significant biological events in the history of life. The body plans of land plants diversified independently in different plant groups, with extant bryophytes representing gametophyte‐dominant forms, while vascular plants represent sporophyte‐dominant forms (Graham et al., 2000; Ligrone et al., 2012; Niklas & Kutschera, 2010; Szövényi et al., 2019). Regardless of the life cycle type, all land plants undergo several developmental phase transitions that need to be precisely regulated to ensure their occurrence under optimal growth conditions. Two major transitions concern the juvenile‐to‐adult phase transition and the vegetative‐to‐reproductive phase changes (Huijser & Schmid, 2011; Ma et al., 2020; Poethig, 2013; Poethig & Fouracre, 2024). Based on genetic studies, it was shown that both these alterations share certain key regulatory factors. Among these, the plant‐specific SQUAMOSA PROMOTER BINDING PROTEIN‐LIKE (SPL) transcription factors (TFs) have been identified as major components of the genetic network that governs plant phase changes. Moreover, members of SPL gene family also affect leaf morphology (Hu et al., 2023), a variety of processes during root, shoot, and flower development (Wang et al., 2008, 2009; Wu & Poethig, 2006; Yamaguchi et al., 2009; Yu et al., 2010, 2015). In addition, SPLs are implicated in the control of various stress responses (Cui et al., 2014; Lan et al., 2019; Li et al., 2022; Morea et al., 2016; Zhao et al., 2022). Although SPL proteins are diverse in their primary protein structure, they share a common characteristic—a SBP (SQUAMOSA‐PROMOTER BINDING PROTEIN) domain—composed of highly conserved 76–78 amino acid residues consisting of two zinc‐binding motifs (Yamasaki et al., 2004). The SBP domain is both necessary and sufficient to recognize and bind the cis‐element TNCGTACAA within target gene promoters, with the GTAC sequence serving as the essential core (Birkenbihl et al., 2005; Cardon et al., 1997, 1999). Furthermore, a bipartite nuclear localization signal (NLS) motif resides at the C‐terminal end of the SBP domain, which overlaps with the second zinc‐binding motif. This NLS is required for the nuclear import of SPL proteins (Yamasaki et al., 2004; Zhang et al., 2019).
Many members of the SPL gene family are post‐transcriptionally regulated by miR156/157, miR529, and miR535 (Li et al., 2022; Tregear et al., 2022; Xu et al., 2016). Out of the16 SPL genes in the model angiosperm Arabidopsis thaliana, 10 are targeted by miR156/157 family members (Preston & Hileman, 2013; Wang et al., 2009; Xu et al., 2016). In contrast, the genome of model liverwort species, Marchantia polymorpha, contains only four MpSPL members, two of which are regulated by miRNAs (Bowman et al., 2017; Streubel et al., 2023; Tsuzuki et al., 2016, 2019). The conserved miR529c‐MpSPL2 module (Mp3g05470, Mp1g10030) controls the transition from vegetative‐to‐reproductive phase of growth in M. polymorpha, resembling a similar mode of action to that observed in angiosperms (Pietrykowska et al., 2023; Tsuzuki et al., 2019). However, in the case of MpSPL1 gene (Mp1g10020), its expression is regulated by the liverwort‐specific miRNA, Mpo‐MR‐13 (also described as MpmiR11671) (Mp6g07990) (Lin et al., 2016; Tsuzuki et al., 2016). The Mpo‐MR‐13‐MpSPL1 module is involved in the control of meristem dormancy in a light‐dependent manner, thereby modulating thallus branching architecture (Pietrykowska et al., 2023; Streubel et al., 2023). Comparative analyses of SPL expression profiles across three bryophyte species and Arabidopsis revealed that miRNA‐targeted SPL genes are typically expressed in a developmentally specific manner or show elevated expression in particular tissues or organs. On the other hand, the Arabidopsis SPL‐family members non‐targeted by miRNA exhibit ubiquitous expression (Alisha et al., 2024). Functional studies have shown that the AtSPL1/12/14/16 genes, orthologs of MpSPL3 (Mp1g13640), and the AtSPL7 gene, the ortholog of MpSPL4 (Mp8g11850), are mainly implicated in different stress responses. For instance, the atspl14 mutant displays a moderately accelerated juvenile‐to‐adult phase transition and delayed flowering, with no major changes in flower morphology. Interestingly, this mutant also exhibits resistance to fumonisin B1 (FB1), a fungal toxin that induces apoptosis via disruption of sphingolipid metabolism (Desai et al., 2002; Stone et al., 2005). These findings highlight a dual role for AtSPL14 in both developmental timing and biotic stress response. In addition, AtSPL1 and AtSPL12 contribute to thermotolerance during reproductive phase. Loss‐of‐function atspl1atspl12 double mutants are hypersensitive to heat stress, whereas overexpression plants exhibit improved fertility under high‐temperature conditions (Chao et al., 2017). Although AtSPL16 function remains less defined, its expression is induced by cold, suggesting a potential role in cold tolerance (Lee et al., 2005; Vogel et al., 2005). In turn, AtSPL7 acts as a central regulator of copper deficiency responses, mediating increased incorporation of exogenous Cu by up‐regulating genes encoding high‐affinity copper transporters (Bernal et al., 2012; Garcia‐Molina et al., 2013; Yamasaki et al., 2009).
In M. polymorpha, the two SPL TFs not targeted by miRNA—MpSPL3 and MpSPL4—have not yet been functionally characterized in terms of their developmental roles. However, transcriptomic profiling identified the MpSPL3 gene, along with two other MpSPL family members, MpSPL1 and MpSPL2, as genes correlated in their expression profile with other genes known to be involved in the reproductive phase transition (Flores‐Sandoval, Eklund, et al., 2018). Given the MpSPL3 expression pattern associated with the reproductive transition, we focused our study on characterizing the biological role of MpSPL3 in M. polymorpha. Taking into account the diverse roles of non‐miRNA‐targeted SPL genes in stress responses and developmental regulation in Arabidopsis, the functional characterization of their bryophyte orthologs represents an important step toward understanding the evolutionary origins of SPL gene function.
In this paper, we show that MpSPL3 is important in maintaining proper thallus morphology and vegetative and generative reproduction. Interestingly, we show that MpSPL3 generates two mRNA isoforms: overexpression of the longer isoform MpSPL3.1, but not the shorter MpSPL3.2, significantly alters the timing and efficiency of gametangiophore formation. Gene expression analyses further revealed that MpSPL3 deficiency affects the expression of a set of genes associated with both the vegetative and generative phases of growth. Collectively, these findings indicate that MpSPL3 functions as one of the central regulatory factors coordinating the developmental programs in M. polymorpha.
RESULTS
MpSPL3
generates two mRNA isoforms with different expression patterns during M. polymorpha development
Recent phylogenetic studies revealed that SPL genes from land plants can be categorized into four phylogenetic groups (Group 1–4), and the SPL‐family members in M. polymorpha have been classified into each phylogenetic group sequentially. MpSPL3 and MpSPL4 proteins are more closely related to each other and form a sister cluster to MpSPL1 and MpSPL2 (Alisha et al., 2024; Bowman et al., 2017). This characteristic division is reflected in the gene expression profile, with MpSPL3 and MpSPL4 showing a broad range of expression during the vegetative and reproductive phases of M. polymorpha life cycle (Figure S1C,D), while MpSPL1 and MpSPL2 exhibit differential expression, with the highest levels observed in reproductive organs and young sporophytes (Figure S1A,B). In contrast to MpSPL4, the MpSPL3 gene encodes two mRNA isoforms, MpSPL3.1 and MpSPL3.2, that differ in the coding sequence length (Figure 1A). We focused our investigation on characterizing the role of MpSPL3.1 and MpSPL3.2 in M. polymorpha development. The MpSPL3.1 isoform is 183 nucleotides longer than MpSPL3.2, which is a consequence of an in‐frame alternative splicing event that results in intron 3 retention. Consequently, from the MpSPL3 locus, two protein isoforms can be produced: MpSPL3.1 with a length of 1219 aa and MpSPL3.2 with a length of 1158 aa. To determine the expression levels of MpSPL3 mRNA isoforms, reverse transcriptase‐quantitative polymerase chain reaction (RT‐qPCR) was performed. As shown in Figure 1B, the MpSPL3.1 transcript is the dominant isoform during the vegetative and reproductive phases of M. polymorpha life cycle, with its highest expression in archegoniophores and the lowest in 1‐week‐old plants. Moreover, a similar ratio of MpSPL3 mRNA isoforms expression was observed in the sporophyte, the diploid generation. Importantly, far‐red light supplementation is necessary to induce the sexual reproduction program in the meristematic region of the M. polymorpha thallus. To check whether the far‐red light regime influences the expression levels of MpSPL3 mRNA isoforms, RT‐qPCR analysis was performed on RNA isolated from dissected apical notches in comparison to RNA isolated from the remaining part of thalli after apical notches dissection (Figure S2A). The plant material was collected from two experimental setups: in the first, after 2 weeks of white light, and in the second, after 2 weeks of white light followed by 10 days of far‐red light irradiation. As a result, we observed the upregulation of both MpSPL3 isoforms after far‐red supplementation (Figure S2B). However, the MpSPL3.2 isoform level was similar in the apical notch and the thallus in the far‐red light growth conditions, while in the case of MpSPL3.1 isoform, its expression was slightly higher in the thallus compared to the apical notch. Thus, MpSPL3.1 shows a defined pattern of far‐red–induced upregulation, with a weaker induction in the apical notch than in other thallus regions.
*Gene structure and expression profile of the MpSPL3 gene.(A) Schematic diagram of the MpSPL3 gene structure and transcripts produced by alternative splicing. 5′/3′ UTRs are marked as white boxes, coding sequences as black boxes, and spliced introns are indicated by diagonal lines; the SBP domain‐coding region is marked as a blue box; gray triangles above the mRNAs point to primer pairs localization specific for each isoform; in the scheme of isoform 2, the dashed line indicates the location of the exon 3‐exon 4 junction after intron 3 splicing.(B) RT‐qPCR analysis of MpSPL3 gene transcripts expression level: MpSPL3.1 and MpSPL3.2. The analysis was performed with two biological replicates. ANTH, antheridiophores; ARCH, archegoniophores; SP, sporophyte (fully mature sporophytes were collected); T1‐1w, Tak‐1_1 week old; T1‐3w, Tak‐1_3 week old; T1‐G, Tak‐1_gemmae; T2‐1w, Tak‐2_1 week old; T2‐3w, Tak‐2_3 week old; T2‐G, Tak‐2_gemmae. P‐values to determine the statistical significance in each analysis were calculated using Student's t‐test (*P < 0.05, and **P < 0.001).(C–I) Promoter activity of the MpSPL3 gene: (C) 2‐week‐old vegetative thallus with apical notches marked with black arrowheads (left panel) with its cross section (middle and right panels), (D) 3‐week‐old vegetative thallus with young gemma cups marked with white arrowheads, (E) antheridiophore, (F) antheridia, (G) archegoniophore dorsal view, (H) archegoniophore ventral side, and (I) archegonia. The scale below represents: 1 mm in C (left panel), D, E, and G; 250 μm in C (middle and left panels) and H; 50 μm in F and I.(J) Multiple sequence alignment of the 61 aa fragment specific for the MpSPL3.1 protein with peptides from SPL3 orthologs found in other liverworts from clade Marchantiopsida (species in brown) and clade Jungermanniopsida (species in green). Highly conserved amino acid residues are marked in black, conserved substitutions of amino acid residues are indicated in gray, and white marks non‐conserved amino acid residues. Hyphens mark deletions of amino acids. The GRAVY (Grand Average of Hydropathy) value indicates the average hydropathy value of a peptide.
Next, to explore the spatiotemporal MpSPL3 promoter activity, plants expressing a β‐glucuronidase (GUS) reporter gene were prepared, the expression of which was under the control of the MpSPL3 genomic fragment containing 5.1‐kb upstream of the ATG start codon. In proMpSPL3:GUS transgenic lines, widespread histochemical GUS staining was observed in the vegetative thalli after 2 (Figure 1C) and 3 weeks of growth (Figure 1D). High GUS activity was observed in the antheridia‐bearing region of the antheridiophore receptacle (Figure 1E) and during different stages of antheridia development (Figure 1F), while weaker activity was observed in the antheridiophore stalk (Figure 1E). No obvious GUS staining was observed in the archegoniophore stalk, while strong GUS staining was detected at the dorsal side of the archegoniophore receptacle (Figure 1G), with finger‐like lobes stained less at the ventral side (Figure 1H), and strong GUS activity in unfertilized archegonia (Figure 1I). The detailed examination of the MpSPL3 expression profile may hint at its widespread role across different developmental processes during M. polymorpha life cycle.
Regardless of the land plant lineage, the SBP domain is the only common motif for all SPL proteins. We previously have shown that SPL proteins from the same phylogenetic group tend to have similar combinations of additional protein motifs (Alisha et al., 2024). This feature is especially visible in phylogenetic Group 3, to which MpSPL3 belongs. However, the 61 aa fragment specific for the MpSPL3.1 isoform was not identified among these conserved motifs. Therefore, we performed a BLAST search against the available land plant genomes and the 1KP database (Carpenter et al., 2019; One Thousand Plant Transcriptomes Initiative, 2019) to investigate the phylogenetic history of the 61 aa fragment specific for the MpSPL3.1 isoform. As a result, we found that similar peptides are present only in MpSPL3 orthologs from the liverwort lineage, with lengths ranging from 52 to 61 aa. According to the analysis of the amino acid conservation, higher sequence similarity to the MpSPL3.1 peptide was found for peptides of SPL3 orthologs from complex thalloid liverworts (Marchantiopsida) than from simple thalloid liverworts (Jungermanniopsida) (Figure 1J). A characteristic feature of all these identified peptides is the enrichment in serine residues, ranging from 10 to 12 in the majority of the cases. This suggests that serine residues may undergo phosphorylation/dephosphorylation events to regulate protein function (Betts & Russell, 2003). To represent the average hydropathy value of the identified peptides, we calculated the GRAVY (Grand Average of Hydropathy) values (Kyte & Doolittle, 1982). From the hydrophobicity analysis, we found that, except for Ptilidium pulcherrimum, all identified peptides have negative GRAVY values, which means that these peptides are hydrophilic and therefore more likely to be exposed on the protein surface.
Mpspl3
ko male plants show abnormal vegetative thallus, produce a low number of gemma cups, and do not develop antheridiophores
To understand the function of MpSPL3 in M. polymorpha life cycle, the CRISPR/Cas9 system was utilized to prepare Mpspl3 mutated alleles. We did not get any female transgenic plants having the inactivated MpSPL3 gene. However, two independent male transgenic lines were obtained with the edited MpSPL3 locus by gRNA designed to act 11 nucleotides downstream of the start codon (Figure 2A). The Mpspl3‐1.1 allele revealed a single nucleotide substitution together with a 20‐nucleotide insertion, while the Mpspl3‐1.3 allele revealed a seven‐nucleotide deletion. In both lines, the introduced indels caused frameshift mutations, introducing premature stop codons resulting in 126 and 81 nt coding sequences, respectively. The truncated Mpspl3 mutated alleles could generate predicted proteins of only 42 and 27 amino acids in length, respectively, lacking almost the whole wild‐type (WT) protein, including the SBP domain (Figure 2A). Moreover, subsequent RT‐qPCR analysis using primers amplifying both mRNA isoforms (Figure 2A, black arrowheads) revealed nearly undetectable MpSPL3 expression in both lines, indicating a complete loss of function of MpSPL3 (Figure 2B). Both Mpspl3 ^ ko ^ mutants exhibited significant thallus alterations in comparison to M. polymorpha Tak‐1 plants when observed from the 10th day of growth up to 8 weeks under white light conditions (Figure 2C). These included severe growth retardation, manifested by decreased thallus surface area, characteristic dense thallus morphology with no (Mpspl3‐1.3 ^ ko ^) or hardly visible (Mpspl3‐1.1 ^ ko ^) thallus branching, and disrupted production of gemma cups (white arrows). The Mpspl3‐1.3 ^ ko ^ plants produced a single gemma cup every 3–4 months; in turn, the Mpspl3‐1.1 ^ ko ^ plants displayed reduced efficiency of gemma cup production with only 1–3 cups at the eighth week of growth, while at the same time Tak‐1 plants produced an average of about 60 cups per plant (Figure 2C,D). Although lower in number at the fifth week of growth, gemma cups of Mpspl3‐1.1 ^ ko ^ plants had twice the diameter of WT plants (Figure 2E; Figure S3). However, after 8 weeks of growth, the diameter of the gemma cups of both plants reached similar sizes.
*Molecular and phenotypic analysis of MpSPL3 knockout mutants generated by CRISPR/Cas9.(A) Upper panel: A schematic representation of a gRNA position to target the MpSPL3 gene region. The position of gRNA is marked by the red arrow; 5′ UTR and 3′ UTR are marked as white boxes; the SBP domain‐coding region is marked as a blue box; coding sequences are marked as black boxes, and introns as black lines. Black triangles point to primer pair localization for amplification of both mRNA isoforms. Lower panel: Alignment of genomic DNA (gray box) and amino acid sequences (orange box) between the MpSPL3 wild‐type gene and two obtained CRISPR/Cas9 KO mutant lines (Mpspl3‐1.1
ko and Mpspl3‐1.3
ko ). gRNA is shown in bold, the start codon is highlighted in gray, the premature stop codon is in red, and identical amino acids are in black.(B) Relative expression level of MpSPL3 in the 3‐week‐old wild‐type Tak‐1 and Mpspl3
ko plants, grown under white light conditions, measured by RT‐qPCR. mRNA levels were normalized to that of MpEF1. The analysis was performed with two biological repetitions. P‐values to determine the statistical significance in each analysis were calculated using Student's t‐test (*P < 0.05, **P < 0.01, and ***P < 0.001).(C) Time course phenotypic analysis of Mpspl3
ko plants in comparison to Tak‐1 plants from 10‐day to 5‐week‐old stage of growth in in vitro conditions. The gemma cups are marked by arrows.(D) Efficiency of gemma cup formation in time course analyzed for Mpspl3
ko plants in comparison to Tak‐1 plants (n = 6).(E) Measurement of gemma cup diameter from the wild‐type Tak‐1 and Mpspl3‐1.1
ko plants after 5 weeks (n = 15 for Tak‐1, n = 6 for Mpspl3‐1.1) and 8 weeks (n = 15 for Tak‐1, n = 9 Mpspl3‐1.1) of growth.(F) The efficiency of antheridiophore formation by Mpspl3‐1.1
ko and Mpspl3‐1.3
ko plants over the time course. Data are mean ± SD; n = 10 plants.(G) Phenotypes of Mpspl3
ko mutant plants as compared to Tak‐1 plants grown for 7 weeks under far‐red irradiation. All images were taken with a Leica M60 microscope. The scale below in (C) represents 1 mm, in (G) it represents 5 mm.*
In the case of the Mpspl3‐1.1 ^ ko ^ line, a small fraction of plants showed less dense thallus morphology, which enabled observation of the branching patterning. In M. polymorpha, the repeatable developmental interval between successive thallus branching events refers to the plastochron (Solly et al., 2017). After 31 days of culture under white light conditions, Tak‐1 plants showed characteristic growth patterning with regular meristem duplication, allowing the thallus to spread radially and to reach the fourth plastochron (Figure S4). In contrast, the Mpspl3‐1.1 ^ ko ^ plants resembled an asymmetric pattern of bifurcation progression, reaching only the first or second plastochron, with 9 out of 20 plants missing half of the thallus. The observed aberrations of the thallus morphology and asymmetry in the Mpspl3 ^ ko ^ plants prompted us to investigate in detail the gemma morphology. WT gemma has a disc‐shaped structure with a uniformly curved edge in which two symmetrically oriented apical notches can be recognized, together with rhizoid precursor cells and a trace of stalk. Only rarely, slight variations from the symmetry of the apical notches arrangement are observed (Figure 3A). By contrast, gemmae of the complete loss‐of‐function mutant, Mpspl3‐1.1 ^ ko ^, can be divided into three types of phenotypes based on their morphology changes: WT‐like, medium, and severe (Figure 3A). Only 16.6% (13 out of 78) of Mpspl3‐1.1 ^ ko ^ gemmae showed the WT‐like phenotype, with the only difference being that the margin was not as transparent as in the Tak‐1 gemmae. Next, 30% (24 out of 78) of Mpspl3‐1.1 ^ ko ^ gemmae indicated medium phenotype changes as follows: a single apical notch or two but asymmetrically localized, rhizoid precursor cells and the stalk more difficult to identify, changed oval shape. Finally, the largest group, comprising as much as 52% (41 out of 78), were gemmae with hard phenotype changes that lacked recognizable notches, rhizoid precursor cells and the stalk. Interestingly, the Mpspl3‐1.1 ^ ko ^ gemmae area was significantly bigger than in Tak‐1 (Figure 3B). After 24 h of growth, nearly all Tak‐1 gemmae started to produce rhizoids (43 out of 44), while in the case of Mpspl3‐1.1 ^ ko ^, rhizoids formation was noted only in 8 out of 60 observed gemmae (Figure S5). Although the Mpspl3‐1.1 ^ ko ^ gemmae were bigger than Tak‐1 at the 0th day of growth (Figure 3A,B), at the fourth day of culture a significant difference in the gemmaling size was observed, with Tak‐1 gemmalings being up to three times bigger than Mpspl3‐1.1 ^ ko ^ gemmalings (Figure 3C,D). Already at the fourth day, all Mpspl3‐1.1 ^ ko ^ gemmalings produced rhizoids; however, they were reduced in number (Figure 3E). We also noticed a significant difference in the shape of the apical notch in Mpspl3‐1.1 ^ ko ^ gemmalings (Figure 3C) which appeared as being more wide‐opened than in Tak‐1 plants (Figure 3F; Figure S6). These observations together may indicate that MpSPL3 is necessary for proper apical notch formation and proper timing of rhizoids development, which in turn require proper expression of genetic factors influencing these processes.
*The Mpspl3 knockout mutation causes changes in gemma morphology.(A) Images of the Tak‐1 and Mpspl3‐1.1
ko 0‐day‐old gemma classified according to the observed changes of morphology. A black arrowhead indicates an apical meristematic notch; a red arrowhead indicates a putative apical meristematic notch; a white arrow indicates a trace of stalk; red arrow indicates a putative trace of stalk; and a white asterisk indicates a rhizoid precursor cell/s. The images were taken with VHX‐7000 Keyence Digital microscope. The scale below represents 50 μm.(B) Quantitative analysis of gemma size shown in (A). Values are mean ± SEM; n = 17. The P‐value to determine the statistical significance was calculated using Student's t‐test (*P < 0.05).(C) Images of the Tak‐1 and Mpspl3‐1.1
ko 4‐day‐old gemmalings. The scale below represents 300 μm.(D– F) Quantitative analysis of the 4‐day‐old gemmalings size (D), the number of rhizoids protruding from the 4‐day‐old gemmaling (E), and the apical notch width presented as the degree of the notch breadth (F). Values are mean ± SEM; n = 17. The P‐value to determine the statistical significance was calculated using Student's t‐test (**P < 0.001). A detailed characterization of the width measurement methodology is shown in Figure S5.
Further on, when the Mpspl3 ^ ko ^ plants were shifted to far‐red light conditions, neither of the Mpspl3 ^ ko ^ mutant plants produced any antheridiophores, even after prolonged exposure for 7 weeks (Figure 2F,G). In conclusion, the lack of MpSPL3 affects proper vegetative thallus growth and abolishes reproductive gametangiophores development in M. polymorpha.
Mpspl3
kd plants show moderate defects of thallus morphology, produce a reduced number of gemma cups, and do not develop gametangiophores
Due to the severe abnormalities and growth retardation observed in the obtained Mpspl3 ^ko^ plants and a lack of a female knockout line to study the MpSPL3 function in female individuals, an alternative strategy employing artificial miRNA (amiR) was implemented. This strategy was used to investigate the effects of MpSPL3 depletion on M. polymorpha development. Following a protocol outlined by Flores‐Sandoval et al. (2016), an amiR targeting both transcripts of the MpSPL3 gene within the coding sequence of exon 1 was designed (Figure 4A). Four knockdown (kd) mutant lines were obtained, three in a male background, amiR‐MpSPL3‐4B ^ kd ^, 6 ^ kd ^ 2C ^ kd ^, and also, one in a female background, amiR‐MpSPL3‐2B ^ kd ^. In Mpspl3 ^ kd ^ plants, significantly elevated levels of pri*‐amiR‐MpSPL3* ^ MIR160 ^ (Figure 4B) and the reduced levels of MpSPL3 transcripts (Figure 4C) were observed, indicating effective knockdown of MpSPL3 expression by artificial miRNA.
*Molecular and phenotypic analysis of MpSPL3 knockdown mutants generated using artificial miRNA.(A) A scheme presenting the location of the artificial miRNA target site in MpSPL3 gene transcripts marked by a red arrow.(B) Relative expression level of pri‐amiR in four independent pro
EF1:amiR‐MpARF1
MpMIR160 lines in comparison to the pri‐miR160 in wild‐type Tak‐1 plants as measured by RT‐qPCR. Primers amplifying pri‐mirR160 backbone (Table S1) were used.(C) Relative MpSPL3 transcripts level in four independent pro
EF1:amiR‐MpSPL3
MpMIR160 lines in comparison to wild‐type Tak‐1 plants as measured by RT‐qPCR. RT‐qPCR analysis (B) and (C) was performed with two biological repetitions using 3‐week‐old plants grown under white light conditions. P‐values to determine the statistical significance in each analysis were calculated using Student's t‐test (*P < 0.05, **P < 0.01, and ***P < 0.001).(D) Phenotypes of Mpspl3
kd male and female mutant plants as compared to Tak‐1 and Tak‐2 plants, respectively, grown in in vitro culture. The images were taken after 3 weeks of growth in constant white light conditions, and then after an additional 7 weeks of growth under far‐red irradiation. All the images were taken with a Leica M60 microscope. The scale below represents 5 mm.(E) Efficiency of gemma cup formation by 3‐week‐old plants in Mpspl3
kd mutant plants compared to Tak‐1 and Tak‐2 shown in (D) (n = 4).(F) Measurement of gemma cup diameter from the 3‐week‐old Mpspl3
kd mutant plants and compared to Tak‐1 and Tak‐2 shown in (D); (n = 8 for Tak‐1, Tak‐2, amiR‐MpSPL3‐4B and 6; n = 5 for amiR‐MpSPL3‐2C and 2B).(G) The efficiency of gametangiophores formation by male Mpspl3‐6
kd and female Mpspl3‐2B
kd mutant plants after 7 weeks of growth under far‐red light conditions. Data are mean ± SD; n = 10 plants.*
In comparison to the Mpspl3 ^ ko ^ plants, MpSPL3 gene knockdown conferred a less severe effect on the vegetative growth of M. polymorpha, but it reflected the same direction of changes as in the knockout plants, both in males and females. Namely, the 3‐week‐old thalli of all four amiR‐MpSPL3 ^ kd ^ mutants showed developmental alterations in plant phyllotaxy and thallus morphology consistent across these lines. These alterations included impaired thallus bifurcation, smaller thallus size (Figure 4D), and reduced number of gemma cups (Figure 4D,E) but, in contrast to knockout plants, the gemma cup diameter was not affected (Figure 4F). In the case of reproductive growth, similarly to knockout mutants, even after 7 weeks of far‐red induction, the production of gametangiophores in all four lines was completely abolished, irrespective of sex (Figure 4D,G). Hence, downregulation of MpSPL3 expression is unfavorable for proper vegetative development, but most importantly, even a 40% reduction of MpSPL3 expression level suppresses gametangiophores development.
Overexpression of MpSPL3
.cds1 but not MpSPL3 .cds2 affects the timing and efficiency of gametangiophores production
To investigate the impact of MpSPL3 protein overexpression on M. polymorpha development, the coding sequences (cds) of both MpSPL3 mRNA isoforms, MpSPL3.cds1 and MpSPL3.cds2, respectively, were constitutively expressed under the CaMV35S or MpEF1 gene promoter. Numerous mutant plants were obtained following transformation with MpSPL3.cds1 and MpSPL3.cds2 (Figure 5A). RT‐qPCR analysis revealed elevated levels of MpSPL3.1 (Figure 5B) and MpSPL3.2 (Figure 5C) transcripts in selected MpSPL3.cds1 and MpSPL3.cds2 overexpression mutant plants, respectively, compared to WT plants. During the vegetative phase of the life cycle, the overexpression plants showed morphology comparable to that of WT plants (Figure S7). However, upon exposure to far‐red light, distinct effects were observed between the MpSPL3.cds1 and MpSPL3.cds2 overexpression. Namely, MpSPL3.cds2 overexpression did not cause any changes in the rate and morphology of gametangiophores production (Figure 5D, lower panel). On the other hand, the male plants overexpressing MpSPL3.cds1 (lines 35S:#1 and EF:#6) formed fewer antheridiophores than WT Tak‐1 plants. Most strikingly, the female plants overexpressing MpSPL3.cds1 (EF:#2 and 35S:#2) exhibited a strong delay in archegoniophores formation, with, additionally, a high reduction of archegoniophore number in comparison to WT Tak‐2 plants (Figure 5D, upper panel). However, the morphology of antheridia and archegonia was not affected (Figure S8). As described above (Figure 1), the two MpSPL3 protein versions differ by only 61 aa; however, only the overexpression of MpSPL3.1, possessing the serine‐enriched peptide, has an impact on the vegetative‐to‐reproductive phase transition. Taken together, our results indicate that a balanced level of the MpSPL3.1 isoform is important to maintain proper timing and efficiency of gametangiophores formation.
*Molecular and phenotypic analysis of plants overexpressing MpSPL3 isoforms 1 and 2.(A) Phenotypic analysis of 8‐week‐old plants grown from gemmae under white light for 2 weeks, then under white light with far‐red light for up to 6 weeks. Arrowheads indicate antheridiophores and archegoniophores visible by the naked eye, respectively; bars = 10 mm.(B, C) RT‐qPCR detection of MpSPL3 transcripts levels in MpSPL3cds1
ox (B) and MpSPL3cds2
ox (C) lines compared with wild‐type Tak‐1 and Tak‐2. The analysis was performed with two biological replicates using 3‐week‐old plants grown under white light conditions. P‐values to determine the statistical significance were calculated using a paired t‐test (*P < 0.05, **P < 0.01, and **P < 0.001).(D) The efficiency of gametangiophores formation in MpSPL3 overexpression plants in a time course. Plants were cultured in vitro under far‐red light conditions as described in (A); data are mean ± SD; n = 10 plants; significant differences to wild‐type plants were calculated by the Kruskal–Wallis test followed by the post hoc Mann–Whitney U multiple comparison test (Table S3).
Lack of MpSPL3
affects the expression of genes responsible for proper vegetative and generative development in M. polymorpha plants
Expression of selected genes responsible for vegetative development
We demonstrated that MpSPL3 affects many developmental processes in M. polymorpha, which was manifested by multiple morphological changes in the Mpspl3 ^ ko ^ plants. To have insights into the genetic implications of the lack of MpSPL3, we aimed at the identification of genes that could potentially work downstream of the MpSPL3 gene. Based on available functional data concerning M. polymorpha vegetative growth, we paid attention to four genes, MpANT (MpAINTEGUMENTA, also described as MpPLT—MpPLETHORA) (Mp8g11450), MpROP (MpRHO OF PLANTS) (Mp7g17540), MpKAI2A (MpKARRIKIN INSENSITIVE2A) (Mp2g11710), and MpGLK (MpGOLDEN2‐LIKE) (Mp7g09740), which, when inactivated, exhibited similar direction in phenotype changes as Mpspl3 ^ ko ^ plants; namely, strong (Mpant, Mprop) or moderate (Mpkai2a, Mpglk) reduction of thallus size, delay in gemma cup formation (Mprop, Mpglk, Mpkai2a), and significantly lower rhizoid initial cell number (Mprop) (Fu et al., 2024; Hernández‐Muñoz et al., 2024; Komatsu et al., 2023; Liu et al., 2024; Mizuno et al., 2021; Mulvey & Dolan, 2023; Rong et al., 2022). We selected 3‐week‐old Mpspl3 ^ ko ^ plants grown in white light to check these genes' expression level. In all four cases, downregulation of mRNA levels was observed, with the strongest downregulation of MpANT, which is known as TF essential for meristem development and maintenance (Figure 6A). Our results together indicate that MpSPL3 plays a potential role in the expression regulation of genes involved in various aspects of thallus growth and development during the vegetative phase.
*The expression of selected genes responsible for proper thallus development and phase transition is downregulated upon MpSPL3 gene knockout.(A) Relative MpANT, MpROP, MpKAI2A, and MpGLK transcript levels were measured by RT‐qPCR in Mpspl3
ko plants in comparison to wild‐type Tak‐1 grown for 3 weeks under white light conditions.(B) Relative MpSPL1, MpSPL2, MpSPL4, and MpARF3 transcript levels were measured by RT‐qPCR in Mpspl3
ko plants in comparison to wild‐type Tak‐1 grown for 3 weeks under white light conditions and in comparison to wild‐type Tak‐1 antheridiophores.(C) Relative MpSPL1, MpSPL2, MpSPL4, and MpARF3 transcript levels were measured by RT‐qPCR in Mpspl3
ko plants in comparison to wild‐type Tak‐1 grown for 2 weeks under white light conditions, followed by 1 (2wWL+1wFR) or 2 weeks (2wWL+2wFR) of far‐red light supplementation. mRNA levels in (A), (B), and (C) were normalized to that of MpEF1. The analysis was performed with three biological repetitions. The Tukey's honestly significant difference test was used for multiple comparisons (P < 0.05, one‐way anova, post hoc Tukey), and means sharing superscripts are not significantly different from each other.*
Expression of MpSPL1
, MpSPL2 , MpSPL4 , and MpARF3
genes
A comparative analysis of gene expression across different developmental stages of M. polymorpha facilitated the identification of TFs upregulated in both sexes as compared to vegetative thalli, and therefore, they were defined as reproductive transition genes. Among these genes, three MpSPLs were identified: MpSPL1, MpSPL2, and MpSPL3. Moreover, it was hypothesized that MpARF3 (AUXIN RESPONSE FACTOR3) (Mp1g07070) inhibits the reproductive transition in M. polymorpha via activation of the Mpo‐MR‐13 and MpMIR529c expression, both targeting and downregulating the expression of MpSPL1 and MpSPL2 transcripts, respectively (Flores‐Sandoval, Eklund, et al., 2018). To investigate whether the lack of MpSPL3 affects the other two MpSPL genes' expression and their putative repressor, MpARF3, we quantified their mRNA levels in 3‐week‐old Mpspl3 ^ ko ^ plants in comparison to 3‐week‐old Tak‐1 grown in white light, and to Tak‐1 antheridiophores in the first experimental setup. In the second setup, after 2 weeks of growth in white light, the culture was supplemented with far‐red light for an additional 1 or 2 weeks for reproductive program induction. In our analysis, we also included the remaining MpSPL4 gene to have a representative of the SPL TFs family that is not recognized as a reproductive transition gene. Surprisingly, in the 3‐week‐old vegetative thallus, not only were MpSPL1 and MpSPL2 transcript levels significantly decreased in the Mpspl3 ^ ko ^ mutants, but also MpSPL4, although to a lesser extent. The MpARF3 transcript level was not changed in the Mpspl3 ^ ko ^ mutants (Figure 6B), and the level of Mp‐miR‐13 precursor remained unchanged (Figure S9A). In our approach, however, we could not detect the miR529c precursor. Importantly, upon the far‐red light regime, MpSPL1 and MpSPL2 transcript levels were still significantly decreased in the Mpspl3 ^ ko ^ mutants when compared to the WT plants (Figure 6C). However, after 2 weeks of growth under far‐red light conditions, the expression profiles of both genes in the Mpspl3 ^ ko ^ mutant resembled those observed in the WT plants, although their overall expression levels remained lower. A different situation was observed for the MpSPL4 and MpARF3 genes, as their expression levels were either unchanged or only slightly altered compared to WT plants after far‐red light supplementation (Figure 6C). Additionally, the level of the Mpo‐miR‐13 precursor remained unchanged under far‐red light conditions in the knockout plants compared to WT plants (Figure S9B). These findings support the hypothesis that the downregulation of MpSPL1 and MpSPL2 in Mpspl3 ^ ko ^ plants is not caused by increased MpARF3 activity, but rather results from the loss of MpSPL3 function. Moreover, our observations indicate that far‐red light signaling can partially compensate for the absence of MpSPL3 in regulating the expression levels of MpSPL1 and MpSPL2. In contrast, MpSPL4 expression does not appear to be significantly affected by the loss of MpSPL3, suggesting that its regulation is largely independent of MpSPL3 under far‐red light conditions.
Expression of selected genes crucial for germ cell specification
One of the main factors necessary for the reproductive phase transition in M. polymorpha is MpBNB (MpBONOBO) (Mp3g23300), whose expression is induced upon far‐red light irradiation (Figure S2C) (Yamaoka et al., 2018). To check whether factors connected to germ cell specification are also misregulated in the Mpspl3 ^ ko ^ plants, we examined the expression levels of MpBNB and additionally MpBZR3 (MpBRASSINAZOLE‐RESISTANT 3) (Mp2g23000), MpLRL (MpLOTUS JAPONICUS ROOTHAIRLESS LIKE) (Mp2g20695), and MpCKI1 (MpCYTOKININ‐INDEPENDENT 1) (Mp2g02750). MpBNB and MpBZR3 expression is tightly connected with environmental conditions, like far‐red light dosage, to induce gametangia initial cell specification (Furuya et al., 2024; Saito et al., 2023; Yamaoka et al., 2018). In the case of MpLRL and MpCKI1, although both genes are broadly expressed during the vegetative phase of M. polymorpha life cycle, they are also crucial for successful germ cell specification (Bao et al., 2024; Breuninger et al., 2016; Saito et al., 2023). MpLRL stabilizes MpBNB by forming a heterodimer, which enables the differentiation of gametangia initial cells and the formation of gametangiophore primordium (Saito et al., 2023). MpCKI1, on the other hand, is required for proper antheridiophore morphology and restricts the MpBNB accumulation to the female germ cell lineage, thereby establishing its identity (Bao et al., 2024). The expression of MpLRL and MpCKI1 was tested in two experimental setups. First, plants were cultured for 3 weeks under white light conditions for vegetative growth. Second, after 2 weeks of growth in white light, the culture was supplemented with far‐red light for an additional 1 or 2 weeks for gametangiophores induction. The expression of MpBNB and MpBZR3 was tested only in the second experimental setup. As a result, we observed that both MpLRL and MpCKI1 were significantly downregulated in the Mpspl3 ^ ko ^ mutants, regardless of the growth conditions (Figure 7A,B), although the effect was stronger in the vegetative phase (Figure 7A). Nevertheless, significant deregulation of MpLRL and MpCKI1 expression was also noted upon 2 weeks of far‐red light (Figure 7B). The levels of both gene mRNAs remained the same after 2 weeks of far‐red light in the Mpspl3‐1.1 ^ ko ^ plants, while in the WT plants their expression was induced by half when comparing a 1‐week far‐red light treatment to a 2‐week far‐red light treatment. Importantly, after 1 and 2 weeks of growth in gametangiophore‐inductive conditions, MpBNB and MpBZR3 levels were strongly downregulated in the Mpspl3‐1.1 ^ ko ^ plants in comparison to WT control, in which the opposite effect was observed—strong expression induction (Figure 7B). Therefore, the most probable reason for the complete abolishment of antheridiophore production in Mpspl3‐1.1 ^ ko ^ plants was the lack of expression induction of all four genes responsible for germ cell specification.
*The expression of selected genes responsible for germ cell specification is downregulated upon MpSPL3 gene knockout.(A) Relative MpLRL and MpCKI1 transcript levels were measured by RT‐qPCR in Mpspl3
ko plants in comparison to wild‐type Tak‐1 grown for 3 weeks under white light conditions and in comparison to wild‐type Tak‐1 antheridiophores.(B) Relative MpLRL, MpCKI1, MpBNB, and MpBRZ3 transcript levels were measured by RT‐qPCR in Mpspl3
ko plants in comparison to wild‐type Tak‐1 grown for 2 weeks under white light conditions followed by 1 (2wWL+1wFR) or 2 weeks (2wWL+2wFR) of far‐red light supplementation. mRNA levels in (A) and (B) were normalized to that of MpEF1. The analysis was performed with three biological repetitions. The Tukey's honestly significant difference test was used for multiple comparisons (P < 0.05, one‐way anova, post hoc Tukey), and means sharing superscripts are not significantly different from each other.*
MpSPL3
isoform 1 overexpression exhibits the strongest effect on MpBNB and MpBZR3 gene expression
A similar effect of deregulation of reproductive transition was observed in the case of MpSPL3 isoform 1 overexpression (MpSPL3.cds1), which raised the question of why both loss‐ and gain‐of‐function mutants exhibited a similar defect. To address this, we examined the expression patterns of other MpSPL genes, together with the four genes involved in germ cell specification, in the MpSPL3.cds1 overexpression plants cultured for 2 weeks in white light conditions, followed by 1 week of far‐red light supplementation. Although a stronger effect of delay in gametangiophores production was observed in female plants, neither MpSPL1 nor MpSPL2 showed changes in expression levels in the female plants overexpressing MpSPL3.cds1 (35S:cds1#2). A similar pattern was observed for the MpSPL4 gene. In contrast, MpSPL1 and MpSPL2 were mildly downregulated, while MpSPL4 was slightly upregulated in male plants overexpressing MpSPL3.cds1 (EF:cds1#6) (Figure S10A). Also, minor changes in MpCKI1 transcript level were noticed in both male and female MpSPL3.cds1 overexpression plants, whereas the MpLRL mRNA level was slightly upregulated only in female MpSPL3.cds1 overexpression plants (Figure S10B). The strongest effect of MpSPL3.cds1 overexpression was observed in the case of the MpBNB and MpBZR3 genes. Both genes were significantly downregulated in male as well as in female overexpression plants, although not to the same extent as in the Mpspl3 ^ ko ^ plants (Figure 7B). To explain the lack of gametangiophores production in Mpspl3 ^ ko ^ plants and the strong delay in their production in MpSPL3.cds1 overexpression plants, we suggest that the substantial reduction in MpBNB and MpBZR3 expression may play a critical role.
DISCUSSION
In this study, we demonstrated by combined analysis of knockout, knockdown, and overexpression approaches that the MpSPL3 gene is an important TF for the correct execution of vegetative and reproductive programs during the M. polymorpha life cycle. The potential importance of MpSPL3 could already be anticipated based on the broad range of MpSPL3 expression (Figure 1). We managed to obtain two male Mpspl3 ^ ko ^ mutant lines, and despite many attempts, no female Mpspl3 ^ ko ^ plant was identified. Nevertheless, the female Mpspl3 ^ kd ^ plants exhibited comparable phenotype changes (Figure 4) to the male Mpspl3 ^ ko ^ (Figure 2) and Mpspl3 ^ kd ^ plants (Figure 4): smaller thallus size, lower efficiency of gemma cup production, and abolished gametangiophores development. Moreover, male and female plants overexpressing MpSPL3 isoform 1 had a similar direction of developmental alterations, although with varying degrees of severity (Figure 5). Consequently, our observations indicate that MpSPL3 is implicated in the regulation of similar developmental processes in both male and female individuals.
MpSPL3
orchestrates different aspects of vegetative growth and development
The lack of MpSPL3 causes morphological aberrations already during gemma development, in a small number of cases leading to mild effects like changing the gemma rim transparency, but in the majority of cases, causing strong gemma deformations reflected in globular shape and lack of characteristic apical notches (Figure 3). Further on, during the first 4 days of gemma development, defects associated with a delay in rhizoid formation, decreased gemmaling size, and increased apical notch width were observed. After a few weeks of culture in white light conditions, loss of MpSPL3 activity resulted in a strong delay of thallus growth and gemma cup production, and also hardly ever recognizable thallus branching (Figure 2). All the observed thallus alterations may be the consequence of the mis‐formed gemmae exhibiting moderate and hard changes of morphology (Figure 3), as the majority of Mpspl3 ^ ko ^ plants grew with the characteristic dense thallus morphology (Figure 2). Only thalli derived from Mpspl3 ^ ko ^ gemmae with mild aberrations (Figure 3) exhibited to some extent the branching pattern (Figure S4). The aberrations caused during the vegetative growth might be a consequence of the downregulation of expression of several genes crucial for proper vegetative development, with the MpANT gene being downregulated the most (Figure 6A). In M. polymorpha, the MpANT gene is essential for meristem maintenance, as its lack results in a drastic reduction of thallus size caused by abnormal meristem morphology. Additionally, the Mpant mutation also causes fewer rhizoids to be produced by developing gemmae (Fu et al., 2024; Liu et al., 2024). Similar aberrations were observed in the case of the MpROP gene knockout, which is also downregulated in the Mpspl3 ^ ko ^ mutants (Figure 6A). The Mprop ^ ko ^ plants exhibited an overall reduction in plant size, occasionally produced gemma cups with gemmae lacking recognizable meristem notches, and showed a decreased number of rhizoid initial cells (Mulvey & Dolan, 2023; Rong et al., 2022). An important role in the proper axis formation in gemma development, but also in thallus growth and bifurcations, is played by auxins (Eklund et al., 2015; Solly et al., 2017; Streubel et al., 2023). Although MpANT is a TF and MpROP belongs to Rho‐like GTPases, both these proteins are implicated in auxin‐related pathways. Auxin maxima formed at the apical notch are crucial for proper MpANT expression in the meristematic region, indicating that MpANT acts downstream of auxin signaling (Melissa et al., 2025). While in the case of MpROP, it has been reported that Mprop mutants show significantly decreased expression of the auxin biosynthesis gene MpYUCCA2 (Mp8g08780), pointing to a link between ROP and auxin signaling in the meristem notch development (Rong et al., 2022). As both MpANT and MpROP are downregulated in the Mpspl3 background, an interesting direction for future studies will be to investigate genes involved in auxin synthesis and transport in the Mpspl3 ^ ko ^ plants.
Among the genes important for vegetative development, MpKAI2A and MpGLK expression is downregulated in Mpspl3 ^ ko ^ mutants (Figure 6A). MpKAI2A is a receptor protein in the KARRIKIN INSENSITIVE2‐dependent signaling pathway that in M. polymorpha positively controls gemma cup formation and gemma initiation (Komatsu et al., 2023). Moreover, it also influences thallus growth, as the size of Mpkai2a loss‐of‐function plants is significantly smaller than that of WT plants (Mizuno et al., 2021). To play its role, MpKAI2A perceives, as yet unidentified, endogenous ligand(s), which, upon binding to MpKAI2A, triggers MpMORE AXILLARY GROWTH2 (MpMAX2; Mp1g02580)‐dependent proteolysis of MpSUPPRESSOR of MORE AXILLARY GROWTH2 1‐LIKE (MpSMXL; Mp3g06310), enabling KAI2‐dependent signal transduction. Interestingly, the MpGLK, in addition to its role of TF critical for chloroplast differentiation, directly activates MpMAX2 expression to regulate the timing of gemma cup formation (Hernández‐Muñoz et al., 2024). Since both of these genes participate in the activation of the KARRIKIN INSENSITIVE2‐dependent signaling pathway to facilitate gemma cup development, the reduced efficiency of gemma cup production observed in the Mpspl3 ^ ko ^ plants might be the result of the combined negative effects of both MpKAI2A and MpGLK downregulation. Overall, our findings indicate that MpSPL3 is engaged in the regulation of diverse processes during vegetative growth in M. polymorpha by modulating the expression of genes orchestrating developmental programs and signaling pathways.
MpSPL3
modulates the expression level of other MpSPLs
Co‐expression data place MpSPL3 together with the reproductive transition genes, MpSPL1 and MpSPL2 (Flores‐Sandoval, Eklund, et al., 2018). A miR529c‐resistant MpSPL2 induces ectopic gametangia, while Mpspl2 knockout only delays gametangiophore formation (Tsuzuki et al., 2019). Although MpSPL1 shows expression similarity to MpSPL2, functional studies link it mainly to meristem regulation: under simulated shade, repression of Mpo‐MR‐13 allows MpSPL1 accumulation, promoting meristem dormancy and suppressing branch formation (Streubel et al., 2023). MpSPL1 may therefore act both in shade‐induced dormancy and later in the transition to reproduction, potentially in coordination with MpSPL2 (Pietrykowska et al., 2023). Both miRNA–SPL modules are regulated by MpARF3 and its repressor MpmiR160 (Mp1g26670): Mpo‐MR‐13 and miR529c are ARF3‐activated, and MpSPL1/2 show ARF3‐dependent post‐transcriptional regulation (Flores‐Sandoval, Eklund, et al., 2018). Consistently, Mparf3 mutants overproduce gametangiophores, whereas Mpmir160 ^ ko ^ mutants fail to form them under far‐red light (Flores‐Sandoval, Romani, & Bowman, 2018). In the case of Mpspl3 ^ ko ^ mutants grown under white light conditions, a strong downregulation of both MpSPL1 and MpSPL2 was observed, with stable MpARF3 and pri‐Mpo‐MR‐13 levels at the same time (Figure 6B; Figure S9A). Interestingly, upon far‐red light, the level of MpSPL1 and MpSPL2 was not as strongly affected as in white light, while the MpARF3 expression resembled that observed in WT plants (Figure 6C). Thus, it is rather unlikely that MpARF3 controls the levels of MpSPL1 and MpSPL2 in Mpspl3 ^ ko ^ mutant plants. Based on our results, we propose that MpSPL3 is involved in MpSPL1 and MpSPL2 gene expression regulation and may work in cooperation with the far‐red light signaling pathway. Intriguingly, the expression of the MpSPL4 gene is dependent directly or indirectly on MpSPL3 only in white light conditions (Figure 6B), as the supplementation with far‐red light was sufficient to restore the MpSPL4 expression to the WT level in the Mpspl3 ^ ko ^ background (Figure 6C). Therefore, we can conclude that MpSPL3 is implicated in maintaining proper expression levels of all other M. polymorpha SPL‐family members in a light‐dependent manner.
MpSPL3
is indispensable for the proper expression of genes crucial for germ cell specification
The most profound effect of the MpSPL3 loss‐of‐function mutation was the abolishment of gametangiophores production, indicating that MpSPL3 is a factor implicated in the control of reproductive induction in M. polymorpha. Two TFs, MpBNB and MpLRL, form heterodimers to function as key regulators of germline fate initiation and gametangia formation (Saito et al., 2023). We showed that the expression levels of both these genes were downregulated in Mpspl3 ^ ko ^ plants grown in the gametangiophore‐inducing conditions (Figure 7B). The downregulation of expression level was also observed for the MpBZR3 gene encoding a TF necessary for the successful development of both antheridia and archegonia, as well as for the MpCKI1 gene encoding a histidine kinase that regulates proper antheridiophore morphology and the female germline specification (Bao et al., 2024; Furuya et al., 2024). Therefore, we propose that the significant reduction of mRNA levels of these master regulators of sexual organ development may be the main reason for gametangiophores formation failure in the Mpspl3 ^ ko ^ plants.
As in the case of MpBNB, in WT plants the MpBZR3 expression is induced in response to far‐red light supplementation, while the MpLRL and MpCKI1 genes are expressed during both the vegetative and reproductive phases of growth (Bao et al., 2024; Breuninger et al., 2016; Furuya et al., 2024; Yamaoka et al., 2018). Interestingly, in the Mpspl3 ^ ko ^ plants, the MpLRL and MpCKI1 genes also exhibited reduced expression levels when plants were cultured in white light for vegetative growth (Figure 7A). This may also contribute to the strong aberration of thallus morphology observed in the Mpspl3 ^ ko ^ plants. Notably, previous studies reported that strong downregulation of MpLRL via amiRNA led to gemma growth defects, including the absence of rhizoids in severe cases (Breuninger et al., 2016). Similarly, mutation in MpCKI1 resulted in abnormal thallus morphology and the absence of gemma cups (Bao et al., 2024). These phenotypic characteristics partially resemble those observed in Mpspl3 ^ ko ^ mutants, supporting the conclusion that MpSPL3 is involved in both vegetative development and reproductive induction.
MpSPL3 generates two mRNA isoforms (MpSPL3.1 and MpSPL3.2). Overexpression of MpSPL3.cds2, the shorter isoform, has no morphological effect, while overexpression of MpSPL3.cds1, the longer isoform, causes deregulation of timing and efficiency of gametangiophores production. The MpSPL3.1 isoform differs from MpSPL3.2 by encoding an additional 61‐aa peptide, which is serine‐enriched (12 residues) (Figure 1J). The observed phenotypic differences between MpSPL3.cds1 and MpSPL3.cds2 overexpression may underline the importance of the 61‐aa specific peptide and the requirement of a balanced level of MpSPL3.1 in maintaining a proper activity of MpSPL3 during the phase transition in M. polymorpha. Especially, that in WT Tak‐1 plants, MpSPL3.1 isoform shows a pattern of far‐red–induced upregulation, with a weaker induction in the apical notch than in other thallus regions (Figure S2B). Moreover, the serine‐enriched peptide is present only in the SPL3 orthologs from liverworts, with higher sequence conservation in representatives of complex thalloid liverworts (Marchantiopsida). Therefore, the preservation of this specific sequence in the course of evolution may indicate that it might be important for controlling some lineage‐specific processes.
As mentioned above, Mpspl3 ^ ko ^ and Mpspl3 ^ kd ^ mutants do not produce gametangiophores. In MpSPL3.cds1 overexpressing mutants, we do observe gametangiophores production; however, the induction is delayed, and the number of gametangiophores is reduced. A potential understanding of these results may be provided by the expression pattern of two genes crucial for germ cell lineage specification, MpBNB and MpBZR3, in the Mpspl3 ^ ko ^ and MpSPL3.1 overexpressing plants. What both genotypes have in common is the downregulation of MpBNB and MpBZR3 expression levels, with MpBNB being strongly downregulated in knockout and overexpression plants, while MpBZR3 is strongly downregulated in knockout and moderately in overexpression plants (Figure 7B; Figure S10B). Functional studies have shown that MpBZR3 expression under gametangiophore‐inductive growth conditions was abolished in Mpbnb ^ ko ^ plants, indicating that MpBZR3 expression might be activated via MpBNB. In the same paper, however, it was reported that inducible MpBZR3 overexpression in the Mpbnb ^ ko ^ background led to the formation of gametangia‐like structures, pointing to a partial MpBNB‐independence in the MpBZR3 function (Furuya et al., 2024). Because the effect of MpBZR3 downregulation is less profound in the MpSPL3.cds1 overexpression plants in comparison to the Mpspl3 ^ ko ^ plants and additionally, MpLRL and MpCKI1 are not reduced in their expression levels in far‐red light conditions in the MpSPL3.cds1 overexpression plants, this might be indicative of a compensatory mechanism, potentially mediated by alternative pathways of transcriptional regulation that can partially substitute for MpBNB and/or MpBZR3 activity when MpSPL3.cds1 is overexpressed. Conversely, in Mpspl3 ^ ko ^ mutants, the simultaneous downregulation of MpBNB, MpBZR3, MpLRL, and MpCKI1 may lead to a cumulative disruption of key regulatory modules, resulting in a more severe phenotype. Future studies should aim to identify potential downstream targets or interacting partners of MpSPL3 that could contribute to this hypothesized compensatory response and to determine whether MpSPL3 exerts control over these genes. Unraveling these mechanisms will be crucial for a more comprehensive understanding of the regulatory network underlying reproductive phase transition in M. polymorpha.
The regulatory networks involving M. polymorpha miRNA‐targeted SPL genes concern the control of specific developmental processes (Streubel et al., 2023; Tsuzuki et al., 2019). In contrast, our results indicate that the MpSPL3 gene is an important player in orchestrating multiple developmental programs during vegetative growth and in regulating vegetative‐to‐reproductive phase transition. Taken together, by affecting various aspects of vegetative development on one site (thallus branching, gemma cup formation, vegetative reproduction) and gametangiophores production on the other site, MpSPL3 appears to be one of the key regulators of the M. polymorpha life cycle.
MATERIALS AND METHODS
Plant material and growth conditions
Male accession Takaragaike‐1 (Tak‐1), female accession Takaragaike‐2 (Tak‐2) were used as WT M. polymorpha subsp. ruderalis. Plants were cultured in vitro on ½ Gamborg's B5 medium (Sigma‐Aldrich, Merck KGaA, Darmstadt, Germany), pH 5.6, supplemented with 1% (w/v) sucrose and 0.8% (w/v) agar in Petri dishes and Duchefa boxes. Plants were grown at 23°C ± 1°C with a 16‐h light/8‐h dark cycle, and the light was provided at an intensity of 50–60 μmol m^−2^ sec^−1^ (LED Neonica Growy #TSG300; Neonica Poland, Lodz, Poland). To induce gametangiophore development, the 14‐day‐old plants were supplemented with far‐red light provided by diodes (LED Engin; OSRAM GmbH, Munich, Germany) emitting ~30 μmol m^−2^ sec^−1^ deep red (~660 nm), ~30 μmol m^−2^ sec^−1^ high‐power blue (1–5 W; ~470 nm), and 20–30 μmol m^−2^ sec^−1^ FR light (735–740 nm).
Cloning and plasmid construction for plant transformation
For the in planta promoter activity analysis, the 5.1‐kb genomic region immediately upstream of the MpSPL3 gene start codon was amplified from Tak‐1 genomic DNA using CloneAmp HiFi PCR Premix (Takara Bio USA, Inc. ‐ formerly Clontech, Mountain View, CA, USA) and the primers listed in Table S1. The PCR product was directionally cloned into the pENTR‐dTOPO vector (Thermo Fisher Scientific Inc., Waltham, MA, USA), followed by an LR reaction to transfer it to the destination vector pMpGWB104 (Ishizaki et al., 2015), generating a fusion to the GUS gene.
For the CRISPR/Cas9 genome editing, guide RNAs (gRNA) to target the MpSPL3 locus were designed using the CRISPRdirect tool (Naito et al., 2015) and CRISPOR tool (Concordet & Haeussler, 2018). Single gRNA sequences were inserted into BsaI‐digested pMpGE_En03 using Ligation High (TOYOBO Co., Ltd., Osaka, Japan) and were then subcloned into the binary vector pMpGE010 (Sugano et al., 2018) by an LR reaction.
For the artificial miRNA (amiR) approach, the protocol outlined by Flores‐Sandoval et al. (2016) was followed. Briefly, in the backbone of pre‐miR160, the miR160 sequence was substituted by the designed 21‐nucleotide sequence of the amiR‐MpSPL3 ^ MpMIR160 ^, which was designed using the Web MicroRNA Designer, WMD3, online tool (Ossowski et al., 2008; Schwab et al., 2006). Off‐target analysis was conducted using BLAST in MarpolBase (Kawamura et al., 2022). amiR was synthesized by GenScript Biotech (Rijswijk, Netherlands) and designed to incorporate EcoRI/HindII sites at the 5′ and 3′ ends of artificial pre‐miRNA, respectively, within the puC57 vector. The amiR was PCR‐amplified using primers ME537 + ME538 (Lagercrantz et al., 2021). The PCR products were then cloned into pENTR‐dTOPO and subsequently transferred to the destination vector, pMpGWB103 (Ishizaki et al., 2015), by an LR reaction.
For the in planta overexpression analysis, the MpSPL3.cds1 and MpSPL3.cds2 without the stop codon were amplified on cDNA derived from RNA isolated from 3‐week‐old male thalli using CloneAmp HiFi PCR Premix (Takara Bio USA, Inc. ‐ formerly Clontech) and the primers listed in Table S1. The PCR products were directionally cloned into the pENTR‐dTOPO vector (Thermo Fisher Scientific Inc). Subsequently, each CDS in pENTR/D‐TOPO was recombined with pMpGWB310 or pMpGWB311 (Ishizaki et al., 2015) by Gateway LR reaction.
The vectors used in this study are listed in Table S2.
Plant transformation and mutant selection
The destination vectors were transformed into Agrobacterium strain GV3101, and the Agrobacterium‐mediated sporeling transformation protocol was followed (Ishizaki et al., 2008). Plants were subjected to two rounds of selection on ½ Gamborg's B5 plates supplemented with 100 μg ml^−1^ cefotaxime and the appropriate antibiotic.
For genotyping and molecular determination of the sex, fragments of plant tissue were crushed in 100 μl of buffer (100 mM Tris–HCl pH 9.5, 1 m KCl, and 10 mM EDTA), diluted with 400 μl of sterilized water, and used as templates for PCR with KAPA3G Plant PCR Kit (Sigma‐Aldrich) with primers listed in Table S1. PCR products were used for Sanger sequencing. All subsequent phenotypic analyses were conducted on the G2 generation.
Microscopy and histochemical analysis
The observation of M. polymorpha vegetative and reproductive structures was conducted using either the Leica M60 microscope or the VHX‐7000 Keyence Digital microscope. The images were captured using Leica Application Suite v4.5 software or VHX‐7000 software, respectively.
Transgenic plants containing proMpSPL3:GUS gene were utilized alongside WT plants as controls for GUS staining, following the protocol by Ishizaki et al. (2012) with minor adjustments. The plant material was arranged in 12 well plate, Scientific Inc., Waltham, MA, USA, and to each well, GUS staining solution was added. Subsequently, the plants underwent vacuum infiltration three times inside Vacutherm™ Oven (Thermo Fisher Scientific Inc). They were then placed in an incubator set at 37°C overnight. The following day, the plants were washed three times with 70% ethanol. Finally, the plants were visualized under a Leica M60 microscope.
Phenotype quantification
The gemma cup diameter was measured using ImageJ or Keyence VHX measurement 7000 software. For scoring the 0‐ and 4‐day‐old gemma area, the appearance of rhizoids in 1‐day‐old gemmae, and the apical notch width assessment Keyence VHX measurement 7000 software was used.
To monitor the progress of gametangiophore formation in transgenic lines, plants were grown from gemmae on ½ Gamborg's B5 plates with 1% sucrose under white light for 14 days. After that, the far‐red light was turned on, and individual plants were observed and counted for visible gametangiophores each day. The number of plants examined was 10 for each line per experiment.
Gene expression analysis by quantitative real‐time PCR (RT‐qPCR)
Total RNA was isolated from: gemmae, 1‐, 2‐, 3‐week‐old plants grown in white light conditions; from 3‐ to 4‐week‐old plants grown for 2 weeks in white light followed by 1 or 2 weeks of far‐red light supplementation, respectively; as well as from antheridiophores, archegoniophores, and sporophytes. To quantify mRNA levels, total RNA was extracted from plant material as indicated above; the genotypes used for RNA isolation are specified in each paragraph of the results chapter and figure captions. RNA isolation was carried out using the Direct‐zol™ RNA kit with slight modifications. TRIzol‐like reagent (1 ml) was added to the plant material, and after a 5‐min incubation at room temperature (RT), the samples were centrifuged three times for 10 min at 4°C and maximum speed each time. The supernatant was then transferred to RNase‐free Eppendorf tubes. Subsequent RNA isolation steps followed the manufacturer's protocol. The obtained RNA was quantified using the DS‐11 Denovix spectrophotometer, and its quality was assessed through 2% agarose gel electrophoresis. DNAse treatment was performed using TURBO DNAse (TURBO DNA‐free kit, Thermo Fisher Scientific Inc), as per the manufacturer's instructions. For routine gene expression analysis, cDNA was synthesized from 1 μg of DNAse‐treated RNA using Oligo (dT)18 primers (Thermo Fisher) and SuperScript III Reverse Transcriptase (Thermo Fisher Scientific Inc), following the manufacturer's protocol.
RT‐qPCR was performed in a 10 μl volume containing Power SYBR™ Green PCR MasterMix or SsoAdvanced™ Universal SYBR™ Green Supermix using QuantStudio 7 & Flex Real‐Time PCR system. The primers used for RT‐qPCR are listed in Table S1. Mean expression levels were calculated using two or three biological replicates with two technical replicates per biological replicate and gene. MpEF1α (MpELONGATION FACTOR 1α, Mp3g23400) was used as the reference gene. Relative expression was calculated using the 2(−ΔCt) method (Figure 1) or the 2(−ΔΔCt) method (Figures 2 and 4, 5, 6, 7; Figures S2, S9, and S10).
Multiple sequence alignment and amino acid composition analysis
Multiple sequence alignment was performed using Clustal Omega program at the EMBL‐EBI web server (Madeira et al., 2024). The conservation of amino acid residues within the alignments was visualized with BOXSHADE online tool https://junli.netlify.app/apps/boxshade/. The grand average of hydropathy value, GRAVY (Kyte & Doolittle, 1982), for the peptide sequences was performed using online GRAVY calculator, https://www.bioinformatics.org/sms2/protein_gravy.html.
Statistical analysis
For RT‐qPCR analysis, statistical significance was assessed using two‐tailed Student's t‐test with three significance levels: *P < 0.05, **P < 0.01, and ***P < 0.001, and for multiple comparisons one‐way anova followed by Tukey's honestly significant difference test was used (P < 0.05). For comparing the gametangiophore formation progresses in overexpression lines, the Kruskal–Wallis test followed by post hoc Mann–Whitney U test with Bonferroni correction was used for multiple comparisons (Table S3).
AUTHOR CONTRIBUTIONS
AA and IS: carried out the experiments and data analysis. IS: designed the experimental plan and supervised the experiments. AA and IS: wrote the manuscript draft. IS, AJ, and ZS‐K: discussed the experiments and their interpretation, reviewed and edited the final version of the manuscript. All authors have approved the final manuscript.
CONFLICT OF INTEREST
The authors declare no conflict of interest.
Supporting information
Figure S1. The RNA‐seq‐based expression profiles of four SPL genes from different developmental stages in Marchantia polymorpha. Figure S2. Expression level of the two MpSPL3 mRNA isoforms and the MpBNB mRNA in the apical notch compared with the remaining thallus. Figure S3. Phenotypic analysis of gemma cup diameter in Mpspl3‐1.1 ^ ko ^ mutants as compared to Tak‐1 plants. Figure S4. Thallus bifurcation is strongly delayed in the Mpspl3‐1.1 ^ ko ^ plants. Figure S5. The Mpspl3‐1.1 ^ ko ^ mutation causes a delay in rhizoid development. Figure S6. Detailed characterization of the apical notch width measurement methodology conducted on Tak‐1 and Mpspl3‐1.1 ^ ko ^ 4‐day‐old gemmalings. Figure S7. No morphological differences were observed in plants overexpressing MpSPL3 isoforms 1 and 2 in comparison to Tak‐1 and Tak‐2 plants during vegetative growth. Figure S8. No morphological differences were observed in gametangia produced by plants overexpressing MpSPL3 isoform 1 in comparison to Tak‐1 and Tak‐2 plants. Figure S9. The expression level of the Mp‐pre‐miR‐13 precursor is not changed upon MpSPL3 gene knockout. Figure S10. Expression levels of other MpSPL family members and genes responsible for germ cells specification in plants overexpressing MpSPL3 isoform 1. Table S1. Oligonucleotides used in this study. Table S2. Vectors used in this study.
Table S3. Statistical differences in gametangiophores production between M. polymorpha wild type and MpSPL3 overexpression plants calculated by Kruskal–Wallis test (in relation to Figure 5D).
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Alisha, A. , Szweykowska‐Kulinska, Z. & Sierocka, I. (2024) Comparative analysis of SPL transcription factors from streptophyte algae and embryophytes reveals evolutionary trajectories of SPL family in streptophytes. Scientific Reports, 14, 1611. Available from: https://www.nature.com/articles/s 41598‐024‐51626‐2 [Accessed May 5, 2025].38238367 10.1038/s 41598-024-51626-2PMC 10796333 · doi ↗ · pubmed ↗
- 2Bao, H. , Sun, R. , Iwano, M. , Yoshitake, Y. , Aki, S.S. , Umeda, M. et al. (2024) Conserved CKI 1‐mediated signaling is required for female germline specification in Marchantia polymorpha . Current Biology, 34, 1324–1332.e 6. Available from: https://linkinghub.elsevier.com/retrieve/pii/S 0960982224000137 [Accessed May 5, 2025].38295795 10.1016/j.cub.2024.01.013 · doi ↗ · pubmed ↗
- 3Bernal, M. , Casero, D. , Singh, V. , Wilson, G.T. , Grande, A. , Yang, H. et al. (2012) Transcriptome sequencing identifies SPL 7‐regulated copper acquisition genes FRO 4/FRO 5 and the copper dependence of iron homeostasis in Arabidopsis . Plant Cell, 24, 738–761. Available from: https://academic.oup.com/plcell/article/24/2/738‐761/6097130 [Accessed October 11, 2025].22374396 10.1105/tpc.111.090431 PMC 3315244 · doi ↗ · pubmed ↗
- 4Betts, M.J. & Russell, R.B. (2003) Amino acid properties and consequences of substitutions. In: Barnes, M.R. & Gray, I.C. (Eds.) Bioinformatics for geneticists. Hoboken, NJ: Wiley, pp. 289–316. Available from: 10.1002/0470867302.ch 14 · doi ↗
- 5Birkenbihl, R.P. , Jach, G. , Saedler, H. & Huijser, P. (2005) Functional dissection of the plant‐specific SBP‐domain: overlap of the DNA‐binding and nuclear localization domains. Journal of Molecular Biology, 352, 585–596. Available from: https://linkinghub.elsevier.com/retrieve/pii/S 0022283605007904 [Accessed May 5, 2025].16095614 10.1016/j.jmb.2005.07.013 · doi ↗ · pubmed ↗
- 6Bowman, J.L. , Kohchi, T. , Yamato, K.T. , Jenkins, J. , Shu, S. , Ishizaki, K. et al. (2017) Insights into land plant evolution garnered from the Marchantia polymorpha genome. Cell, 171, 287–304.e 15. Available from: https://linkinghub.elsevier.com/retrieve/pii/S 0092867417311248 28985561 10.1016/j.cell.2017.09.030 · doi ↗ · pubmed ↗
- 7Breuninger, H. , Thamm, A. , Streubel, S. , Sakayama, H. , Nishiyama, T. & Dolan, L. (2016) Diversification of a transcription factor family led to the evolution of antagonistically acting genetic regulators of root hair growth. Current Biology, 26, 1622–1628. Available from: https://linkinghub.elsevier.com/retrieve/pii/S 0960982216304146 [Accessed October 11, 2025].27265398 10.1016/j.cub.2016.04.060PMC 4920954 · doi ↗ · pubmed ↗
- 8Cardon, G. , Höhmann, S. , Klein, J. , Nettesheim, K. , Saedler, H. & Huijser, P. (1999) Molecular characterisation of the Arabidopsis SBP‐box genes. Gene, 237, 91–104. Available from: https://linkinghub.elsevier.com/retrieve/pii/S 037811199900308 X [Accessed May 5, 2025].10524240 10.1016/s 0378-1119(99)00308-x · doi ↗ · pubmed ↗