The multicellular haploid stage of land plants develops from a single haploid cell produced by meiosis – the spore. Starting from a non-polar state, these spores develop polarity, divide asymmetrically and establish the first axis of symmetry. Here, we show that the nucleus migrates from the cell centroid to the basal pole during polarisation of the Marchantia polymorpha spore cell. A microtubule organising centre on the leading edge of the nucleus initiates a microtubule array between the nuclear surface and the cortex at the basal pole. Simultaneously, cortical microtubules disappear from the apical hemisphere but persist in the basal hemisphere. This is accompanied by the formation a dense network of fine actin filaments between the nucleus and the basal pole cortex. Experimental depolymerisation of either microtubules or actin filaments disrupts cellular asymmetry. These data demonstrate that the cytoskeleton reorganises during spore polarisation and controls the directed migration of the nucleus to the basal pole. The presence of the nucleus at the basal pole provides the cellular asymmetry for the asymmetric cell division that establishes the apical-basal axis of the plant.

Polarisation – defined as the development of asymmetry across a cell – is an essential process in the development of multicellular organisms. There are two multicellular stages in the life cycle of land plants. The multicellular diploid phase is derived from the zygote produced by fertilisation of the egg cell. The multicellular haploid phase is derived from the spore cell produced by meiosis. The polarity of each cell type – zygote and spore – defines the orientation of the apical-basal body axis of the plants. Zygote polarity is inherited from the egg cell (Faure et al., 2002). By contrast, there is no evidence that spores are initially polarised (Roeder et al., 2022). Instead, spores start in a non-polar state and polarity develops de novo within days of germination. Here, we test whether microtubules and actin filaments are involved in the establishment of polarity in spores.

Microtubules and actin filaments reorganise during the polarisation of animal zygotes. In Drosophila melanogaster, the orientation of microtubules directs the movement of the nucleus to one cell pole, polarising distinct activities to opposite ends of the cell (Bernard et al., 2018; Huynh and St Johnston, 2004). In Caenorhabditis elegans, the site of sperm entry sets in motion events that reorganise the actin cortex (Munro and Bowerman, 2009). Subsequently, on actin-myosin contraction, cortical flows are generated which direct polarity determinants to the cell poles (Munro et al., 2004). In both examples, this polarity defines the future anterior-posterior axis of the animal. In the zygote of Arabidopsis thaliana, the future apical-basal body axis is defined by microtubules and actin filaments which direct cell elongation and nuclear migration, respectively (Kimata et al., 2016, 2019). Given that microtubules and actin filaments play key roles in the polarisation of fly, worm and angiosperm zygotes, we hypothesised that the cytoskeleton would also function in spore polarisation.

The multicellular haploid phase (gametophyte) of non-seed land plants – bryophytes, lycophytes and monilophytes – begins as a spore; a single haploid cell produced by meiosis and encased in a sporopollenin wall. The spore of Marchantia polymorpha divides asymmetrically to form a large apical cell that proliferates and a small basal cell that differentiates into a rhizoid (O'Hanlon, 1926; Shimamura, 2016). As the spore's first division is asymmetric, this indicates that apical-basal polarity is established before cell division. Very little is known of the mechanisms driving this polarity, except that two microtubule organising centres (MTOCs) – known as polar organisers – and their associated microtubule arrays migrate to one cell pole before cell division (Sakai et al., 2022). We hypothesised that both microtubules and actin filaments are required to polarise the spore of M. polymorpha.

Nuclear migration, accompanied by cytoskeleton reorganisation, occurs before asymmetric cell division in many cell types. For example, during stomatal development, the nucleus moves away from a BASL/BRX polarity domain located on one cell surface (Muroyama et al., 2020). This BASL/BRX domain also locally destabilises cortical microtubule arrays, generating microtubule asymmetry across the cell (Muroyama et al., 2023). Together, these events act to position the asymmetric division plane at the opposite side of the cell. In A. thaliana zygotes, actin cables move the nucleus to one cell pole before cell division, resulting in an asymmetric cell division (Kimata et al., 2016, 2019). Drug-induced actin filament depolymerisation blocks nuclear migration and the zygote divides symmetrically. In the spores of the fern Onoclea sensibilis, disruption of microtubules prevents nuclear migration and results in symmetric cell divisions compared with asymmetrical divisions in the control (Vogelmann et al., 1981). In summary, the cytoskeleton integrates polarity signals and cell shape to position the pre-mitotic nucleus and asymmetric cell division plane in plant cells.

Here, we show that the nucleus moves from the spore cell centroid to the basal pole to establish the asymmetry of the first division. A polar organiser on the leading edge of the nucleus initiates a dense astral microtubule array that polymerises from the nuclear surface towards the cortex at the basal pole. A network of fine actin filaments also forms between the nucleus and the basal cell cortex. Simultaneous with nuclear movement, cortical microtubules disappear from the apical hemisphere but remain in the basal hemisphere, generating microtubule asymmetry in the cortex. Drug-induced depolymerisation of either microtubules or actin filaments disrupts the asymmetry of the first division. These data indicate that microtubules and actin filaments together coordinate nuclear migration and the establishment of polarity in the germinating spore of M. polymorpha.

M. polymorpha spores divide asymmetrically between 24 and 32 h after plating on nutrient media

The multicellular, haploid phase of the M. polymorpha life cycle is derived from a spore. To define the stages of spore development, germinating spores resulting from a cross between wild-type Takaragaike-1 (Tak-1) and Takaragaike-2 (Tak-2) were imaged at 24, 32, 48 and 72 h after plating (Fig. 1A; Fig. S1A,B). At 24 h, all spores were single celled with chloroplasts. By 32 h, most cells in the population had divided asymmetrically to form a large chloroplast-filled apical cell and a small basal cell with relatively fewer chloroplasts. By 48 h, the basal cell had elongated to form a short rhizoid and the apical cell had divided. By 72 h, the apical cell had undergone multiple rounds of cell division, while the basal rhizoid cell had elongated further but had not divided. In conclusion, spore populations develop highly asynchronously, with most spores dividing between 24 h and 32 h after plating.

Fig. 1.

Marchantia polymorpha spores divide asymmetrically between 24 and 32 h after plating and microtubule organisation progressively changes. (A) Development of wild-type spores at 24, 32, 48 and 72 h after plating. Each spore is a different individual from the same population. Scale bars: 20 μm. (B) Position of the nucleus and division plane in spores expressing pMpROP:mScarletI-N7 pMpUBE2:mScarletI-AtLTI6b at 30 h after plating. Presented are central slices in xy of four spores. Scale bars: 10 μm. (C) Microtubule organisation in spores expressing pMpEF1α:GFP-MpTUB1 at 29 h after plating. Presented are full z-projections (1, 4, 5) and z-projections of central slices (2, 3) of five spores. Scale bars: 10 μm. (D) Percentage of the spore population with each type of microtubule array at 24, 28 and 32 h after plating. N is the number of spores analysed within the population at each time point. (E) Organisation of chloroplasts (red) and microtubules (cyan) in spores expressing pMpEF1α:GFP-MpTUB1 at 29 h after plating. Presented are z-projections (E1-E6) and central slices in xy (E1′-E6′) of six spores. PO, polar organisers. Scale bars: 10 μm.

Fig. 1.

Marchantia polymorpha spores divide asymmetrically between 24 and 32 h after plating and microtubule organisation progressively changes. (A) Development of wild-type spores at 24, 32, 48 and 72 h after plating. Each spore is a different individual from the same population. Scale bars: 20 μm. (B) Position of the nucleus and division plane in spores expressing pMpROP:mScarletI-N7 pMpUBE2:mScarletI-AtLTI6b at 30 h after plating. Presented are central slices in xy of four spores. Scale bars: 10 μm. (C) Microtubule organisation in spores expressing pMpEF1α:GFP-MpTUB1 at 29 h after plating. Presented are full z-projections (1, 4, 5) and z-projections of central slices (2, 3) of five spores. Scale bars: 10 μm. (D) Percentage of the spore population with each type of microtubule array at 24, 28 and 32 h after plating. N is the number of spores analysed within the population at each time point. (E) Organisation of chloroplasts (red) and microtubules (cyan) in spores expressing pMpEF1α:GFP-MpTUB1 at 29 h after plating. Presented are z-projections (E1-E6) and central slices in xy (E1′-E6′) of six spores. PO, polar organisers. Scale bars: 10 μm.

As the first division of the spore is asymmetric, we hypothesised that the nucleus would be positioned asymmetrically within the cell before mitosis. To visualise the position of the nucleus in spores, we imaged spores expressing a nuclear and plasma membrane double reporter, pMpROP:mScarletI-N7 pMpUBE2:mScarletI-AtLTI6b (mS-N7 mS-AtLTI6b) (Mulvey and Dolan, 2023; Sauret-Güeto et al., 2020). In single-celled spores at 29 h after plating, the nucleus was either located near the geometric centre of the spore (the centroid) or near one pole of the cell (Fig. 1B1,B2). In two-celled spores the division plane was positioned highly asymmetrically, and each daughter cell had a central nucleus (Fig. 1B3,B4). These data confirm the asymmetry of the first division and suggest that this asymmetry results from the nucleus moving from the cell centroid to one pole of the cell.

Microtubule organisation changes during the development of spore asymmetry

We hypothesised that a cytoskeleton-based mechanism would be involved in directing the nucleus from the cell centroid to one pole of the cell. First, we characterised the organisation of microtubules during early spore development by labelling live microtubules with the pMpEF1α:GFP-MpTUB1 (GFP-MpTUB1) reporter (Buschmann et al., 2016). Five distinct microtubule arrays have previously been described in the epidermal cells of M. polymorpha thalli (Attrill and Dolan, 2024); an interphase cortical array located next to the cell membrane (Fig. S1C1); microtubule foci that form on the nucleus surface in preprophase (Fig. S1C2); perinuclear and astral arrays nucleated from two polar organisers located on opposite sides of the preprophase nucleus (Fig. S1C3); mitotic spindle arrays (Fig. S1C4); and phragmoplast arrays for cytokinesis (Fig. S1C5). These five microtubule arrays were also observed among populations of unsynchronised spores at 29 h after plating (Fig. 1C). Microtubule foci were often centrally positioned in spores, whereas the mitotic spindle and phragmoplast were consistently positioned near one pole of the cell. Chloroplasts were densely populated in the hemisphere opposite to where the mitotic spindle was positioned (Fig. 1E). The position of polar organisers varied between the cell centroid and one pole. In summary, the same microtubule arrays are present in spores and epidermal cells, but mitotic and cytokinetic arrays are polar localised in spores.

To determine the timing of each microtubule reorganisation event, spores were imaged at 24, 28 and 32 h after plating and their microtubule organisations quantified (Fig. 1D; Fig. S1D). Cortical arrays were present in all spores at 24 h. Microtubule foci and polar organisers first appeared in the population at 24 h, and by 28 h they were present in 61% of spores. The first cell divisions occurred at 28 h, and by 32 h almost 60% of the population were either in mitosis, cytokinesis or had divided. These data indicate that the microtubule organisation progressively changes – from cortical microtubule arrays to preprophase arrays to mitotic spindle and phragmoplast arrays – in the spore population between 24 h and 32 h after plating as the population develops asynchronously.

Polar organisers and the nucleus move from the cell centroid to the basal pole before cell division

To determine the exact order of microtubule reorganisation events in spores, we analysed the microtubule dynamics in individual spores by timelapse microscopy. First, we investigated polar organiser formation. Microtubule foci and polar organisers were present in a small percentage of spores at 24 h. By 28 h, almost half the spore population formed polar organisers (Fig. 1D). Polar organisers are therefore not inherited from the sporocyte (spore mother cell) but are formed de novo after spore germination. Timelapse imaging of single spores at 29 h after plating demonstrated that multiple low intensity foci coalesce to form the two brighter foci – polar organisers – on opposite sides of the nucleus (Fig. 2A). This is consistent with two polar organisers forming through aggregation of multiple smaller foci, as previously observed in M. polymorpha tissue cells (Buschmann et al., 2016).

Fig. 2.

Migration of the nucleus from the cell centroid to the basal pole is led by a polar organiser and dense astral array. (A-F) Timelapses of microtubule reorganisation in spores expressing pMpEF1α:GFP-MpTUB1 starting at 29 h after plating. Presented are deconvolved central slices in xy. (A) Formation of two polar organisers (arrowheads at 60 min) from three foci (arrowheads at 0 min). Observed in n=8 spores. (B) Major microtubule reorganisation events during spore development: polar organiser migration (0-60 min), mitosis (90 min) and cytokinesis (105-165 min). Observed full process (0-165 min) in n=16 spores, and part of process (60-165 min) in n=8 spores. (C) Co-migration of the nucleus (magenta) and polar organisers/microtubules (cyan) from the cell centroid to the basal pole followed by mitosis. Nuclear position was captured by the expression of pMpROP:mScarletI-N7 pMpUBE2:mScarletI-AtLTI6b. Observed in n=7 spores. (D) A dense astral array spans between the basal polar organiser (filled arrowhead) and basal cortex (D1). This array is magnified in D2. There is no such array between the apical polar organiser (empty arrowhead) and apical cortex. (E) Subtle rotation of the polar organiser axis (dotted line) before/during migration to the basal pole. Observed in n=14 spores. (F) Large 90° rotation of the polar organiser axis (dotted line) before/during migration to the basal pole. Observed 45-90° rotation in n=5 spores. Scale bars: 10 μm.

Fig. 2.

Migration of the nucleus from the cell centroid to the basal pole is led by a polar organiser and dense astral array. (A-F) Timelapses of microtubule reorganisation in spores expressing pMpEF1α:GFP-MpTUB1 starting at 29 h after plating. Presented are deconvolved central slices in xy. (A) Formation of two polar organisers (arrowheads at 60 min) from three foci (arrowheads at 0 min). Observed in n=8 spores. (B) Major microtubule reorganisation events during spore development: polar organiser migration (0-60 min), mitosis (90 min) and cytokinesis (105-165 min). Observed full process (0-165 min) in n=16 spores, and part of process (60-165 min) in n=8 spores. (C) Co-migration of the nucleus (magenta) and polar organisers/microtubules (cyan) from the cell centroid to the basal pole followed by mitosis. Nuclear position was captured by the expression of pMpROP:mScarletI-N7 pMpUBE2:mScarletI-AtLTI6b. Observed in n=7 spores. (D) A dense astral array spans between the basal polar organiser (filled arrowhead) and basal cortex (D1). This array is magnified in D2. There is no such array between the apical polar organiser (empty arrowhead) and apical cortex. (E) Subtle rotation of the polar organiser axis (dotted line) before/during migration to the basal pole. Observed in n=14 spores. (F) Large 90° rotation of the polar organiser axis (dotted line) before/during migration to the basal pole. Observed 45-90° rotation in n=5 spores. Scale bars: 10 μm.

To investigate the events after polar organiser formation, further timelapses of individual spores were captured at 29 h after plating. Polar organisers and perinuclear arrays were initially located near the cell centroid, but 60 min later were positioned near one cell pole (Fig. 2B). By 90 min this structure had disassembled and was replaced by the mitotic spindle. By 105 min the mitotic spindle was replaced by a phragmoplast aligned with the spindle equator. Cytokinesis was completed at this plane between 105 and 165 min. In summary, the polar organisers and associated perinuclear arrays moved from the cell centroid to one cell pole, where the future mitotic spindle and phragmoplast formed. This aligns with previous observations made by Sakai et al. (2022). We define this pole as the ‘basal pole’ – as this domain will form the basal cell on division – and the opposite pole as the ‘apical pole’.

As the nucleus and polar organisers both localise to the basal pole before cell division, we predicted that these structures would move together. To define the position of the nucleus and polar organisers simultaneously, spores expressing the double nuclear and plasma membrane reporter, mS-N7 mS-AtLTI6b, and the microtubule reporter, GFP-MpTUB1, were imaged at 29 h after plating. At the first time point, the nucleus localised between the two polar organisers in the cell centroid (Fig. 2C). As the polar organisers migrated from the cell centroid to the basal pole, the nucleus also migrated. The nuclear signal disappeared at 60 min, indicating disintegration of nuclear envelope, followed by appearance of the mitotic spindle. During cytokinesis, two daughter nuclei were reformed (Fig. S2A,B). In summary, the nucleus and polar organisers migrate together to the basal pole before the first cell division.

Next, we investigated the dynamics of the two polar organisers and the associated astral arrays during nuclear migration. In timelapses, one polar organiser was consistently positioned closer to the cell cortex and appeared to lead the migration (Fig. 2D1). A dense array of astral microtubules always extended from this leading polar organiser to the cell cortex (Fig. 2D2). This cortical region at the basal pole is defined as the ‘basal cortex’ and the leading polar organiser as the ‘basal polar organiser’. During migration, the opposing ‘apical polar organiser’ initiated few astral microtubules which extended out to the ‘apical cortex’. These data indicate that the formation of a dense astral array – initiated from the basal polar organiser and polymerising towards the basal cortex – is correlated with nuclear migration.

Pairs of polar organisers on opposite sides of the nucleus often rotated before and during their migration to the basal cortex. In timelapses, the polar organiser axis – the straight line joining the two polar organisers – rotated around the cell centroid (0-30 min in Fig. 2E,F). The degree of rotation varied between spores; often the shift was subtle (Fig. 2E) but occasionally larger rotations of up to 90° from the starting orientation were observed (Fig. 2F; Fig. S2C). This initial rotation roughly aligned the polar organiser axis parallel to the future mitotic spindle axis. On migration the polar organiser axis continued to rotate, by a smaller degree, to fully align with the future apical basal axis. This rotation is consistent with the hypothesis that polar organisers orient and migrate towards a specific, pre-determined cortical region.

An asymmetric distribution of cortical microtubules develops progressively as the polar organisers migrate towards the basal pole

Cortical microtubules form a network throughout the cell cortex that reorganises during spore polarisation. At 24 h after plating, before polar organiser formation, a sparse random network of microtubules covered the entire spore cortex (Fig. 3A). Most cortical microtubules were short and highly dynamic, treadmilling across the spore cortex (Fig. S2D). A random cortical network was also present in spores at 29 h after plating, when microtubule foci or polar organisers were positioned near the cell centroid (Fig. 3B,C). By contrast, when the basal polar organiser was positioned at the basal cortex, cortical microtubules were consistently dense in the basal hemisphere but absent from the apical hemisphere (Fig. 3D). In summary, cortical microtubules are randomly organised before and during polar organiser formation. Subsequently, after polar organiser migration, they disappear from the apical hemisphere and accumulate near the basal pole.

Fig. 3.

Cortical microtubules deplete from the apical hemisphere during polar organiser migration to the basal pole. (A-D) Organisation of cortical microtubules in spores expressing pMpEF1α:GFP-MpTUB1 at 24 h (A) or 29 h (B-D) after plating. Presented are z-projections (A-D) and central planes in xy (A′-C′). Cortical arrays were randomly arranged in spores with no cytosolic arrays (A), microtubule foci (B) and polar organisers near the cell centroid (C). Cortical arrays were basally localised in spores with polar organisers at the basal pole (D1-D5). Scale bars: 10 μm. (E) Timelapse of a spore expressing pMpEF1α:GFP-MpTUB1 starting at 29 h after plating. Presented are z-projections (E) showing the depletion of cortical microtubules from the apical cortex as the polar organiser, seen in the central planes in xy (E′), migrates to the basal pole. Observed in n=17 spores. Scale bars: 10 μm. (F) Plot of the cortical microtubule density in spore E over 120 min. (G) Plot of the cortical microtubule bundling in spore E over 120 min. (H,I) Tracking of microtubule plus ends in spores expressing pMpEF1α:GFP-AtEB1a at 29 h after plating. Presented are colour-coded temporal projections of GFP-AtEB1 in the cortex (H) and near the basal polar organiser (I) in two spores each. Timelapses were captured at 1.2 s intervals. Images were deconvolved and z-projected before temporal projection. Scale bars: 2 μm.

Fig. 3.

Cortical microtubules deplete from the apical hemisphere during polar organiser migration to the basal pole. (A-D) Organisation of cortical microtubules in spores expressing pMpEF1α:GFP-MpTUB1 at 24 h (A) or 29 h (B-D) after plating. Presented are z-projections (A-D) and central planes in xy (A′-C′). Cortical arrays were randomly arranged in spores with no cytosolic arrays (A), microtubule foci (B) and polar organisers near the cell centroid (C). Cortical arrays were basally localised in spores with polar organisers at the basal pole (D1-D5). Scale bars: 10 μm. (E) Timelapse of a spore expressing pMpEF1α:GFP-MpTUB1 starting at 29 h after plating. Presented are z-projections (E) showing the depletion of cortical microtubules from the apical cortex as the polar organiser, seen in the central planes in xy (E′), migrates to the basal pole. Observed in n=17 spores. Scale bars: 10 μm. (F) Plot of the cortical microtubule density in spore E over 120 min. (G) Plot of the cortical microtubule bundling in spore E over 120 min. (H,I) Tracking of microtubule plus ends in spores expressing pMpEF1α:GFP-AtEB1a at 29 h after plating. Presented are colour-coded temporal projections of GFP-AtEB1 in the cortex (H) and near the basal polar organiser (I) in two spores each. Timelapses were captured at 1.2 s intervals. Images were deconvolved and z-projected before temporal projection. Scale bars: 2 μm.

We correlated cortical microtubule reorganisation to polar organiser migration through timelapse microscopy of spores at 29 h after plating. As the polar organiser moved to, and remained at, the basal pole, the cortical microtubule network progressively decreased in density (20-50 min in Fig. 3E, 60-90 min in Fig. S3A,B). This depletion started at the apical pole and proceeded towards the basal pole. Eventually only a few short, scattered microtubules remained in the basal hemisphere (60 min in Fig. 3E). Quantification showed that cortical microtubule density progressively decreased over time (Fig. 3F; Fig. S3C,D). By contrast, microtubules became increasingly bundled over time, peaking once the polar organiser had docked at the basal pole (60 min in Fig. 3G, 105 min in Fig. S3E,F). Both measures then dropped when the remaining cortical microtubules disappeared at mitosis (120 min in Fig. 3F,G). In conclusion, there is a gradual change from a random to an asymmetric distribution of cortical arrays as polar organisers migrate to the basal pole.

We hypothesised that the asymmetry of the cortical arrays could be linked to microtubule depolymerisation in the apical hemisphere and/or increased initiation of microtubules from the basal polar organiser which feed into the basal hemisphere. To determine whether cortical microtubules can be initiated from polar organisers, polymerising microtubules were tracked using the END BINDING 1 (EB1) protein which binds to microtubule plus ends (Chan et al., 2003). Timelapses of spores expressing the pMpEF1α:GFP-AtEB1a reporter (GFP-AtEB1) were captured at 1.2 s intervals starting at 29 h after plating (Buschmann et al., 2016). Z-projections of each time point were colour-coded and overlaid to generate temporal projections. At the cortex, EB1 comets moved in distinct tracks, confirming that GFP-AtEB1 labelled polymerising cortical microtubule ends in spores (Fig. 3H; Movie 1). In some spores, a high density of EB1 comets rapidly moved away from a single point (a polar organiser) (Fig. 3I; Movie 2). This reflects consistent growth of microtubule plus ends away from the basal polar organiser. Tracking of EB1 comets into the basal cortex indicated that some microtubule plus ends continued to grow into the cell cortex. This is consistent with the hypothesis that the cortical microtubule population in the basal hemisphere is, at least in part, maintained by microtubules initiated from the basal polar organiser.

The final position of the basal polar organiser and nucleus defines the division asymmetry

We hypothesised that polarity would be fixed once the polar organisers had migrated to the basal pole. The polar organiser axis frequently rotated before migration (Fig. 2E,F). However, after migration and docking of the basal polar organiser at the basal pole, no further rotation was ever observed. Instead, the polar organiser remained in this docked position for a considerable time before disintegration, as indicated by the change in fluorescence signal from distinct points to broad flat areas around the perinuclear arrays (30-40 min in Fig. 4A). In summary, docked polar organisers do not move or reorient. This is consistent with the hypothesis that apical-basal polarity is fixed once the nucleus is at the basal pole.

Fig. 4.

The position of the basal polar organiser defines the first asymmetric division plane. (A-C) Timelapses of microtubule reorganisation before and during cell division in spores expressing pMpEF1α:GFP-MpTUB1 starting at 29 h after plating. Timelapses show the persistence of the polar organiser at the basal pole and its breakdown (A), mitosis (B) and phragmoplast expansion (C). Presented are z-projections (A,B,C) and central planes in xy (A′,B′). Images in B and C were deconvolved. Events in A were observed in n=13 spores, events in B in n=2 spores (with 5 min intervals) and in n=26 spores (with 10-15 min intervals), and events in C in n=32 spores. Scale bars: 10 μm. (D,E) Microtubule organisation in multi-celled sporelings expressing pMpEF1α:GFP-MpTUB1 at 50 h after plating. Images show a two-celled sporeling with random cortical arrays and polar organisers in the apical cell (D) and two three-celled sporelings with distinct cortical arrays in the apical and basal cells (E). Presented are z-projections (D,E) and central planes in xy (D′). Arrowheads indicate the division planes. Scale bars: 10 μm.

Fig. 4.

The position of the basal polar organiser defines the first asymmetric division plane. (A-C) Timelapses of microtubule reorganisation before and during cell division in spores expressing pMpEF1α:GFP-MpTUB1 starting at 29 h after plating. Timelapses show the persistence of the polar organiser at the basal pole and its breakdown (A), mitosis (B) and phragmoplast expansion (C). Presented are z-projections (A,B,C) and central planes in xy (A′,B′). Images in B and C were deconvolved. Events in A were observed in n=13 spores, events in B in n=2 spores (with 5 min intervals) and in n=26 spores (with 10-15 min intervals), and events in C in n=32 spores. Scale bars: 10 μm. (D,E) Microtubule organisation in multi-celled sporelings expressing pMpEF1α:GFP-MpTUB1 at 50 h after plating. Images show a two-celled sporeling with random cortical arrays and polar organisers in the apical cell (D) and two three-celled sporelings with distinct cortical arrays in the apical and basal cells (E). Presented are z-projections (D,E) and central planes in xy (D′). Arrowheads indicate the division planes. Scale bars: 10 μm.

Next, we determined whether the basal position of the polar organisers was retained by the mitotic spindle and guided the plane of cytokinesis. At mitosis, roughly 30 h after plating, the polar organiser and perinuclear arrays disappeared (<5 min) and the mitotic spindle formed (5 min in Fig. 4B). The spore then passed through metaphase, anaphase and telophase in the next 20-25 min. Importantly, the axis between the mitotic spindle poles was parallel to the axis formed by the polar organisers. During cytokinesis, roughly 31-32 h after plating, a phragmoplast formed of loosely packed, short parallel microtubules expanded centrifugally until it touched the cell cortex (0-40 min in Fig. 4C). The orientation of expansion was perpendicular to the mitotic spindle axis, dividing the cell asymmetrically. As the phragmoplast dismantled, cortical microtubules polymerised at distinct points across the cortex until the surface was covered by a random network (80 min in Fig. 4C). Overall, the location of the docked polar organiser defines the basal pole, determines the site of mitosis and the plane of cell division.

To investigate the microtubule organisation in the daughter cells derived from division of the spore, multi-celled sporelings were imaged at 50 h after plating. Two-celled sporelings comprised of a large apical cell and small basal cell in which microtubules were randomly arranged (Fig. 4D). Polar organisers were occasionally observed near the centroid of the apical cell. Three-celled sporelings comprised two apical cells, in which cortical microtubules were randomly organised, and one basal cell that had initiated a tip-growing projection (rhizoid) (Fig. 4E). In this basal cell, microtubules were often arranged parallel to the long axis of the projection as observed in other tip-growing plant cells. Together, these data indicate that microtubules re-establish after the first division to form different arrays in the apical and basal cells, reflecting their different cell identities.

An actin filament array forms between the nucleus and the basal cortex

We hypothesised that actin filaments would also reorganise and function in the polarisation of spores. Live actin filament dynamics were visualised in spores using the pMpWDL:GFP-LifeAct (GFP-LifeAct) reporter (Fig. S4A-C). This contains a short peptide sequence, LifeAct, that labels plant actin filaments in vivo (Era et al., 2009; Riedl et al., 2008; Vidali et al., 2009). In spores expressing GFP-LifeAct, a multi-layered, dense ‘mesh’ of filaments was present in the spore cell cortex at 29 h after plating (Fig. S4D,E). These filaments were randomly organised, often formed bundles and varied in length. Short interval timelapses captured the constant polymerisation and highly dynamic nature of these filaments (Fig. S4F). Taken together, GFP-LifeAct successfully labels actin filaments in spores.

To test whether actin filaments reorganise during nuclear migration in spores, we simultaneously imaged actin filaments (GFP-LifeAct), the nucleus and the plasma membrane (mS-N7 mS-AtLTI6b) in spores at 29 h after plating. Actin filament organisation was then compared between spores with a nucleus located at the centroid (central nucleus) and with a nucleus located near the basal pole (basal nucleus). In spores with a central nucleus, bundles of actin filaments were uniformly distributed through the cortex with no actin filament arrays localised around the nucleus (Fig. 5A). By contrast, in spores with a basal nucleus, a dense array of fine actin filaments accumulated between the nucleus and the basal pole – a ‘basal actin filament network’ (Fig. 5B). Furthermore, there was a high density of actin filaments at the cell cortex below the nucleus at the basal pole – a ‘basal actin filament patch’. In timelapses, the basal actin filament network and patch were present when the nucleus was near the basal cell cortex and during early mitosis (Fig. 5C). In summary, a basal actin filament network and patch form when the nucleus nears the basal pole.

Fig. 5.

An actin filament network forms between the nucleus and basal cortex. (A,B) The organisation of actin filaments (green) differs between spores with a nucleus (magenta) in the cell centre (A) or at the basal pole (B). The white arrowhead indicates the presence of an actin filament network between the basal-positioned nucleus and the basal cortex. Spores expressing pMpWDL:GFP-LifeAct and pMpROP:mScarletI-N7 pMpUBE2:mScarletI-AtLTI6b were imaged at 29 h after plating. Presented are z-projections of GFP (A1,B1) and mScarletI (A2,B2) and central xy slices of GFP (A1′,B1′) and mScarletI (A2′,B2′). A1′, B1′ and A2′, B2′ are merged in A3, B3, respectively. These spores were analysed and the images are repeated in Fig. S5A,G. Scale bars: 10 μm. (C) Timelapse of actin filament (green) reorganisation and the nucleus (magenta) position in a spore expressing pMpWDL:GFP-LifeAct and pMpROP:mScarletI-N7 pMpUBE2:mScarletI-AtLTI6b at 29 h after plating. Presented are merged channels as z-projections (C) or central slices in xy (C′). A white arrowhead indicates the actin network between the nucleus and basal cortex. Observed in n=2 spores. Scale bars: 10 μm. (D,E) Comparison of the RFP (plasma membrane labelled by pMpUBE2:mScarletI-AtLTI6b) and GFP (actin filaments labelled by pMpWDL:GFP-LifeAct) signal intensities in spores with a central nucleus (D) and basal nucleus (E) at 29 h after plating. Presented are sum-of-slices projections of a central 2.6 μm section in RFP (D1,E1) and GFP (D2,E2) used for analysis. Plots (D3,E3) present the normalised intensity values of GFP (green) and RFP (magenta) around the spore perimeter (μm). Arrowheads indicate the start point and direction that intensity was measured. False coloured region indicates a GFP peak at the basal pole. Scale bars: 10 μm. (F) Actin filament organisation during cytokinesis of spores expressing pMpWDL:GFP-LifeAct at 29 h after plating. Presented are z-projections of two spores. White arrowheads indicate the phragmoplast. Observed in n=8 spores. Scale bars: 10 μm. (G,H) Actin filament organisation in two-celled spores expressing pMpWDL:GFP-LifeAct (green) and pMpUBE2:mScarletI-AtLTI6b pMpROP:mScarletI N7 (magenta) captured at 29 h (G) or 50 h (H) after plating. Presented are z-projections of the GFP (G1,H1), RFP (G2,H2) and merged channels (G3,H3). The white arrowhead indicates the actin filament network at the rhizoid cell tip. Scale bars: 10 μm.

Fig. 5.

An actin filament network forms between the nucleus and basal cortex. (A,B) The organisation of actin filaments (green) differs between spores with a nucleus (magenta) in the cell centre (A) or at the basal pole (B). The white arrowhead indicates the presence of an actin filament network between the basal-positioned nucleus and the basal cortex. Spores expressing pMpWDL:GFP-LifeAct and pMpROP:mScarletI-N7 pMpUBE2:mScarletI-AtLTI6b were imaged at 29 h after plating. Presented are z-projections of GFP (A1,B1) and mScarletI (A2,B2) and central xy slices of GFP (A1′,B1′) and mScarletI (A2′,B2′). A1′, B1′ and A2′, B2′ are merged in A3, B3, respectively. These spores were analysed and the images are repeated in Fig. S5A,G. Scale bars: 10 μm. (C) Timelapse of actin filament (green) reorganisation and the nucleus (magenta) position in a spore expressing pMpWDL:GFP-LifeAct and pMpROP:mScarletI-N7 pMpUBE2:mScarletI-AtLTI6b at 29 h after plating. Presented are merged channels as z-projections (C) or central slices in xy (C′). A white arrowhead indicates the actin network between the nucleus and basal cortex. Observed in n=2 spores. Scale bars: 10 μm. (D,E) Comparison of the RFP (plasma membrane labelled by pMpUBE2:mScarletI-AtLTI6b) and GFP (actin filaments labelled by pMpWDL:GFP-LifeAct) signal intensities in spores with a central nucleus (D) and basal nucleus (E) at 29 h after plating. Presented are sum-of-slices projections of a central 2.6 μm section in RFP (D1,E1) and GFP (D2,E2) used for analysis. Plots (D3,E3) present the normalised intensity values of GFP (green) and RFP (magenta) around the spore perimeter (μm). Arrowheads indicate the start point and direction that intensity was measured. False coloured region indicates a GFP peak at the basal pole. Scale bars: 10 μm. (F) Actin filament organisation during cytokinesis of spores expressing pMpWDL:GFP-LifeAct at 29 h after plating. Presented are z-projections of two spores. White arrowheads indicate the phragmoplast. Observed in n=8 spores. Scale bars: 10 μm. (G,H) Actin filament organisation in two-celled spores expressing pMpWDL:GFP-LifeAct (green) and pMpUBE2:mScarletI-AtLTI6b pMpROP:mScarletI N7 (magenta) captured at 29 h (G) or 50 h (H) after plating. Presented are z-projections of the GFP (G1,H1), RFP (G2,H2) and merged channels (G3,H3). The white arrowhead indicates the actin filament network at the rhizoid cell tip. Scale bars: 10 μm.

To verify the presence and evaluate the variability of the basal actin filament patch, the GFP (GFP-LifeAct) and RFP (mScarletI-AtLTI6b) fluorescence intensities around the cell surface were quantified and their fluorescence intensity profiles were compared. In spores with a central nucleus, GFP and RFP intensities generally varied together along the spore perimeter, with the exception of a few narrow, random spikes in GFP intensities (Fig. 5D; Fig. S5A). In spores with a basal nucleus, a local enrichment in GFP intensity was observed over an ∼10 μm region located half-way around the spore circumference – between 28 and 38 μm distance – which included the basal pole (Fig. 5E). By contrast, such an enrichment in RFP intensity was not observed in the same region. This specific enrichment of GFP intensity in the basal cortex was consistent among all spores quantified with a basal nucleus (Fig. S5B-G). Other very narrow, ‘non-specific’ peaks in GFP intensity were variably positioned outside of this region, similar to observations in spores with a central nucleus. Overall, this verified our observations that actin filaments consistently accumulated at the basal pole when the nucleus was basally positioned. The intensity and appearance of the GFP peak at the basal pole varied between spores, appearing as one wide peak, as two high separate narrower peaks, or as multiple smaller scattered peaks at the basal pole (Fig. S5B-G). This suggests that the basal actin filament network is highly dynamic. In summary, a dense network of fine actin filaments forms at the basal cortex as the nucleus migrates from the cell centroid to basal pole.

A dense actin filament network develops in the basal daughter cell after cell division

Next, we determined whether actin filaments reorganise during or after the first asymmetric cell division. In spores undergoing cytokinesis at 29 h after plating, a short, dense array of actin filaments corresponding to the phragmoplast developed (Fig. 5F). On the apical side of the phragmoplast, actin filaments were organised into bundled arrays, but not on the basal side. After cytokinesis, actin filaments were bundled in the apical cell and relatively unbundled in the basal cell (Fig. 5G). In two-celled spores at 50 h after plating, the basal cell had elongated at the apex and an array of fine actin filaments developed at the site of tip-growth (Fig. 5H). In summary, actin filament organisation differs between the apical and basal hemisphere of the spore during and after cytokinesis.

Inhibition of microtubule polymerisation disrupts formation and positioning of the new cell wall

Nuclear movement is accompanied by the formation of asymmetric microtubule arrays in spores. We hypothesised that microtubules are required for nuclear migration, and therefore inhibition of microtubule polymerisation using oryzalin would decrease division asymmetry. We first identified a suitable oryzalin dose by growing wild-type spores on media containing increasing oryzalin concentrations over 4 days (Fig. S6). On 0.1% DMSO (control), most spores divided and produced a rhizoid cell and apical cell by day 2. On 1 µM oryzalin, spores divided but produced short, branched rhizoids, and at 3.3 µM oryzalin spores rarely divided and instead swelled. Oryzalin therefore disrupts rhizoid outgrowth and cell division. To test whether the asymmetry of divided spores was disrupted by oryzalin, the position of the new cell wall was defined in sporelings expressing a nuclear and plasma membrane reporter (mS-N7 mS-AtLTI6b) grown on 3.3 µM oryzalin or 0.1% DMSO for 48 h. On DMSO, 96-100% of sporelings – in which the division plane could be determined – divided asymmetrically (Fig. 6A,B; Fig. S8A). On 3.3 µM oryzalin, 29-36% divided asymmetrically and 0-5% symmetrically (Fig. 6A; Fig. S8A). The remaining 58-71% had ‘partially divided’ – they developed either an incomplete new cell wall or had two nuclei but no new cell wall (i.e. mitosis occurred without cytokinesis). The position of the incomplete cell wall and daughter nuclei in these cells was often central, thereby predicting a symmetrical division plane (Fig. 6C). These data are consistent with the hypothesis that dynamic microtubules are required for the migration and/or retention of the nucleus at the basal pole to define the asymmetry of the first division.

Fig. 6.

Depolymerisation of the cytoskeleton increases the symmetry of first division. (A) Percentage of scorable spores that divided asymmetrically, symmetrically or partially after 48 h growth on 0.1% DMSO (control), 3.3 μM oryzalin or 0.1 μM LatB. Partial divisions and symmetrical divisions increased on oryzalin and LatB treatments, respectively, compared with the DMSO control. N is the number of divided spores in which the division plane position could be assessed. Presented are three repetitions representing three independent spore populations. (B-D) Position of the new cell wall and daughter nuclei in sporelings expressing pMpROP:mScarletI-N7 pMpUBE2:mScarletI-AtLTI6b when grown on 0.1% DMSO (B), 3.3 μM oryzalin (C) or 0.1 μM LatB (D) for 48 h. Presented are z-projections of multiple independent spores showing partial divisions on oryzalin treatment (C1-C4) and symmetrical divisions on LatB treatment (D1-D4). (E) Model of cytoskeleton reorganisation and nuclear migration during spore polarisation and the first asymmetric cell division. (1) The nucleus is located at the cell centroid and is not attached to actin filaments or microtubules (∼24 h after plating). (2) Microtubule foci form around the central nucleus (∼28 h after plating). (3) Foci aggregate into two polar organisers and nucleate perinuclear arrays and astral arrays. (4) The nucleus and polar organisers migrate towards the basal cell cortex. A dense astral array nucleated from the basal polar organiser polymerises towards the basal cell cortex. Actin filaments accumulate between the basal cortex and nucleus base. (5) The basal polar organiser and nucleus docks at the basal cortex, where the actin filament network is positioned. The nucleus remains here until mitosis. (6) Mitosis with mitotic spindle positioned at the basal pole (∼32 h after plating). (7) Phragmoplast expansion in an asymmetric plane. (8) Two-cell stage with a larger apical cell and smaller basal cell.

Fig. 6.

Depolymerisation of the cytoskeleton increases the symmetry of first division. (A) Percentage of scorable spores that divided asymmetrically, symmetrically or partially after 48 h growth on 0.1% DMSO (control), 3.3 μM oryzalin or 0.1 μM LatB. Partial divisions and symmetrical divisions increased on oryzalin and LatB treatments, respectively, compared with the DMSO control. N is the number of divided spores in which the division plane position could be assessed. Presented are three repetitions representing three independent spore populations. (B-D) Position of the new cell wall and daughter nuclei in sporelings expressing pMpROP:mScarletI-N7 pMpUBE2:mScarletI-AtLTI6b when grown on 0.1% DMSO (B), 3.3 μM oryzalin (C) or 0.1 μM LatB (D) for 48 h. Presented are z-projections of multiple independent spores showing partial divisions on oryzalin treatment (C1-C4) and symmetrical divisions on LatB treatment (D1-D4). (E) Model of cytoskeleton reorganisation and nuclear migration during spore polarisation and the first asymmetric cell division. (1) The nucleus is located at the cell centroid and is not attached to actin filaments or microtubules (∼24 h after plating). (2) Microtubule foci form around the central nucleus (∼28 h after plating). (3) Foci aggregate into two polar organisers and nucleate perinuclear arrays and astral arrays. (4) The nucleus and polar organisers migrate towards the basal cell cortex. A dense astral array nucleated from the basal polar organiser polymerises towards the basal cell cortex. Actin filaments accumulate between the basal cortex and nucleus base. (5) The basal polar organiser and nucleus docks at the basal cortex, where the actin filament network is positioned. The nucleus remains here until mitosis. (6) Mitosis with mitotic spindle positioned at the basal pole (∼32 h after plating). (7) Phragmoplast expansion in an asymmetric plane. (8) Two-cell stage with a larger apical cell and smaller basal cell.

Depolymerisation of actin filaments decreases division asymmetry

Actin filaments form a network between the migrating nucleus and basal cell cortex in spores. To test the hypothesis that actin filaments are required for nuclear migration, actin filaments were depolymerised using latrunculin B (LatB) and the division symmetry quantified. We first identified a suitable LatB dose by growing wild-type spores on media containing increasing LatB concentrations over 4 days (Fig. S7). On 0.1% DMSO (control), most spores divided to produce a rhizoid cell and proliferating apical cell. On 0.03 µM LatB, spores divided but most lacked rhizoids. On 0.1 µM LatB, some spores divided to form two equal sized chloroplast-filled cells. These data indicate that LatB treatment blocked rhizoid outgrowth and were consistent with the hypothesis that depolymerisation of actin filaments resulted in increased symmetric cell division. To further test whether LatB affects the asymmetry of the first cell division, the position of the new cell wall was defined in sporelings expressing a nuclear and plasma membrane reporter (mS-N7 mS-AtLTI6b) grown on 0.1 µM LatB or 0.1% DMSO for 48 h. On DMSO, 96-100% of sporelings – in which the division plane could be determined – divided asymmetrically (Fig. 6A,B; Fig. S8A). By comparison, on 0.1 µM LatB, 66-82% of spores divided asymmetrically and 18-34% divided symmetrically (Fig. 6A,D; Fig. S8A). We confirmed the depolymerisation of actin filaments by 0.1 µM LatB treatment on sporelings expressing an actin filament reporter (Fig. S8B,C). The increase in the percentage of symmetric divisions on LatB treatments compared with the controls indicates that depolymerisation of actin filaments disrupts cell division asymmetry. In the absence of actin filaments, the nucleus remains in the cell centre leading to a symmetric cell division in some spores. These data are consistent with the hypothesis that actin filaments are required for the basal migration of the nucleus to define the asymmetry of the first division.

Cellular asymmetry develops in the M. polymorpha spore and directs nuclear movement to orient the first asymmetric cell division plane. At 24 h after plating the nucleus is located at the spore centroid, but ∼29 h after plating the nucleus migrates to the cortex at the basal pole over the course of less than an hour. Early in spore polarisation, two polar organisers – liverwort-specific MTOCs – form de novo at opposite sides of the nucleus located in the cell centroid. Together with the nucleus, the polar organisers move to the basal pole where chloroplasts become depleted. At the same time, cortical microtubules disappear from the apical hemisphere but remain in the basal hemisphere where there is active growth from microtubule plus ends. A dense network of fine actin filaments also forms between the nucleus and the basal cortex. As a result of nuclear migration, the mitotic spindle forms near the basal pole and the phragmoplast expands to divide the spore into a relatively large apical cell and small basal cell. During polarisation, the initially non-polar spore undergoes a series of cytoskeletal reorganisations, nuclear migration and chloroplast repositioning to establish an internal apical-basal asymmetry that directs the first division of the spore (Fig. 6E).

Actin filaments often provide mechanical force for nuclear movement in plant cells (Ketelaar et al., 2002; Vidali and Hepler, 2001). In spores, we observed a basal actin filament network form during nuclear migration, and upon depolymerisation of actin filaments we quantified a decrease in cell division asymmetry. These data are consistent with the hypothesis that a dynamic actin filament network moves the nucleus towards the basal pole, setting up the first asymmetric cell division of the spore cell. Such actin-dependent movement would require the activity of myosin motor proteins. Myosin XI-I has been implicated in nuclear positioning after the asymmetric division of stomata progenitor cells (Muroyama et al., 2020). We speculate that myosin XI is also involved in directed nuclear migration in M. polymorpha spores.

During the first 24 h of spore development, a dynamic cortical microtubule array forms in a random organisation, as theoretically expected for a spherical cell where there is no preferential axis of growth. Between roughly 28 and 32 h of spore development, and ∼1 h before nuclear migration, two polar organisers form on opposite sides of the nucleus. Once formed, these polar organisers reorient to align their axis with the future spore apical-basal axis. The polar organiser on the leading edge of the nucleus initiates a dense astral microtubule array that polymerises towards the cortex at the basal pole. We speculate that this astral array anchors to a polar cortical domain to guide the nucleus to this domain. A similar system of MTOC rotation and nuclear alignment occurs in fucoid algal zygotes, where astral arrays from the basal-most centrosome anchor to adhesion sites at the basal cortex (Bisgrove and Kropf, 2001). However, employing acentrosomal MTOCs in cell polarisation is a novel polarity mechanism of M. polymorpha spores not described in other plant species to date.

Cortical microtubules disappear from the apical hemisphere of the spore but remain in the basal hemisphere during nuclear migration. The precise role of this cortical microtubule asymmetry is unclear. Typically, cortical arrays control the direction of plant cell expansion [see for example, Burk and Ye (2002)]. However, spores do not change shape during polarisation but remain spherical, and therefore these cortical microtubules are unlikely to function in directional cell expansion. Microtubules in the basal cortex were often roughly aligned with the equator of the future mitotic spindle. One possibility is that these arrays deposit proteins in a cortical division zone to guide the expanding phragmoplast during spore cytokinesis, similar to the function of the preprophase band in other cell types (Buschmann and Müller, 2019).

To establish cell polarity, the organisation and function of actin filaments and microtubules are often distinct but complementary (Li and Gundersen, 2008). In spores, both a fine filamentous actin network and an astral microtubule array occupy the volume between the migrating nucleus and the basal cortex. We propose a model for their roles in spore polarisation based on our observations (Fig. 6E). First, a cue orients the formation of a polarity domain at the basal pole. Formation of a polarity domain is a common mechanism to define a cell pole, but the identity of polarity domain proteins in spores is unknown (Hartman and Muroyama, 2023; Wallner, 2020). Such a basal polarity domain would then anchor the plus ends of astral microtubules initiated from the closest polar organiser. As progressively more microtubules are initiated from the now ‘basal’ polar organiser, a dense astral array forms, which orients the nucleus towards the basal pole. Simultaneously, actin filaments nucleate and polymerise from the basal cortex, forming a network which attaches to and pulls the nucleus to the basal pole. According to this model, microtubules orient the nucleus while an actin filament network generates the mechanical force for nuclear migration. Ultimately these cytoskeletal rearrangements generate an apical-basal polarity which determines the asymmetry of the first cell division and the formation of the two distinct cell types – a differentiated rhizoid and proliferating apical cell that forms the mature plant body.

Reporter constructs

The pMpEF1α:GFP-MpTUB1 (GFP-MpTUB1) and pMpEF1α:GFP-AtENDBINDING1a (GFP-AtEB1) reporters were generated by Buschmann et al. (2016) to label microtubules and microtubule plus ends, respectively, in GFP. Both are expressed under the constitutive promotor of the M. polymorpha ELONGATION FACTOR 1-α (EF1α) gene.

The pMpROP:mScarletI-N7 pMpUBE2:mScarletI-AtLTI6b (mS-N7 mS-AtLTI6b) reporter labels the nucleus and plasma membrane, respectively, in mScarletI and was generated as a single construct using the OpenPlant Kit (Sauret-Güeto et al., 2020). First, L1_ pMpROP:mScarletI-N7 was generated through a BsaI assembly reaction between L0_PROM5_MpROP (Mulvey and Dolan, 2023), OP-025, OP-040, OP-053 and OP-002. Then, L1_ pMpROP:mScarletI-N7 was combined with OP-062, OP-011, OP-070 (L1_ pMpUBE2:mScarletI-AtLTI6b) and OP-005 in a L2 SapI assembly reaction to generate the expression construct encoding the modified acetolactate synthase gene (conferring chlorsulfuron resistance), pMpROP:mScarletI-N7 and pMpUBE2:mScarletI-AtLTI6b. MpROP and MpUBE2 promoters were selected for their constitutive expression (Mulvey and Dolan, 2023; Sauret-Güeto et al., 2020).

The pMpWDL:GFP-LifeAct (GFP-LifeAct) reporter labels actin filaments and is composed of a LifeAct sequence – ATGGGTGTCGCAGATTTGATCAAGAAATTCGAAAGCATCTCAAAGGAAGAA – fused to GFP expressed under the constitutive promotor of the M. polymorpha WDL gene (Champion et al., 2021).

Amplification of plasmids and transformation into Agrobacterium

Plasmids were transformed into One Shot OmniMAX 2-T1R chemically competent Escherichia coli cells (Thermo Fisher Scientific) by heat-shock at 42°C for 30 s then ice for 2 min. After adding 250 μl S.O.C. medium, the mixture was incubated at 37°C for 1 h shaking at 225 rpm. The mixture was diluted into LB medium (1:50), spread onto solid LB media plates containing 100 μg/ml spectinomycin and incubated overnight at 37°C. Single colonies were grown at 37°C overnight in liquid LB media containing 100 μg/ml spectinomycin. Plasmids were isolated from the transformed E. coli cells using the GeneJet Plasmid MiniPrep kit (Thermo Fisher Scientific).

Amplified plasmids were transformed into Agrobacterium tumefaciens GV3101 strain by electric shock: 1 μl plasmid and 50 μl thawed Agrobacterium were micro-pulsed before adding 1 ml of warm S.O.C. media. Transformed Agrobacterium was grown for 2 days at 28°C on LB media plates containing 50 μg/ml gentamycin, 50 μg/ml rifampicin and 50 μg/ml spectinomycin. Single transgenic colonies were transferred into liquid LB antibiotic media.

Transformation of M. polymorpha

Plasmids were transformed into wild-type M. polymorpha sporelings – derived from crossing wild-type accessions Tak-1 and Tak-2 – using transgenic Agrobacterium following the method developed in Ishizaki et al. (2008) and improved upon in Honkanen et al. (2016).

Plant lines, growth conditions and crossings

The wild-type M. polymorpha accessions used were Tak-1 and Tak-2. Plants were grown on half-strength B5 Gamborg's medium containing 1.5 g/l B5 Gamborg, 0.5 g/l MES hydrate, 1% sucrose, pH adjusted to 5.5, set with 1% agar. Plants were grown at 23°C in continuous white light at 50-60 μmol m²s¹.

To induce reproductive development, plants were potted on soil containing a 1:3 ratio of fine vermiculite and Neuhaus N3 compost within SacO2 Microbox containers and grown at 20°C in long day conditions of 16 h light/8 h dark. White light was set at 50-60 μmol m²s¹ and enhanced with far-red light at 30-40 μmol m²s¹. Male and female plants were crossed to generate spores. Sporangia were harvested and either immediately stored in water at 4°C or dried then frozen at −70°C. Fresh sporangia were sterilised in 1% sodium dichloroisocyanurate (NaDCC) for 3 min before washing and bursting in sterile water. Frozen sporangia were sterilised in 0.1% NaDCC for 1 min and centrifuged at 12,000 rpm (13,700 g) for 2 min before the NaDCC solution was removed and sterile water added.

Stereomicroscope imaging

Spores grown on media plates were imaged using a Keyence VHX-7000 digital microscope equipped with a VHX-7020 camera and VH-ZST lens.

Spinning disk imaging

Spores were setup in imaging chambers made following the protocol in Kirchhelle and Moore (2017) with a few adjustments. A breathable gum border (Carolina Observation gel) was filled with Gamborg media and layered with cellophane soaked in liquid Gamborg media (half-strength B5 Gamborg's medium without agar). One sterile sporangium was burst into 200 µl water and 40 µl of the solution added to the chamber. Spores were grown within the chamber for 29-50 h before imaging. Alternatively, for imaging gemmalings, gemma were transferred into the chamber, perfluorodecalin was added and gemma grown for 1-2 days.

Imaging used an Olympus IX3 Series (IX83) inverted microscope equipped with a Yokogawa W1 spinning disk, Hamamatsu ORCA-Fusion CMOS camera and a 100×/1.45 NA oil objective or 40×/0.75 NA air objective. For GFP, excitation was set at 488 nm and emission captured at 525 nm. For mScarletI, excitation was set at 561 nm and emission captured at 617 nm. For chlorophyll, excitation was set at 640 nm and emission captured at 685 nm. To capture microtubules and actin filaments, z-stacks of 20-22 µm with 0.26 µm slices were taken at single or multiple time points. To track microtubule plus ends, a Piezo Z stage and Hamamatsu ORCA-Flash 4.0 camera rapidly captured z-stacks of 0.99 µm with 0.33 µm slices at 1.2 s intervals. The laser power, exposure time and time intervals were tested and adjusted to prevent bleaching over the timelapses. A Z-Drift compensation autofocus system stabilised imaging overtime. As spore populations developed asynchronously, and were heterogeneous for the reporter expression, specific spores within each population were selected for imaging.

Image deconvolution and conversion

Image deconvolution used Huygens software (Scientific Volume Imaging). Z-projections and central slices were converted using ImageJ Fiji (Schindelin et al., 2012). For temporal projections, each image slice was deconvolved in Huygens then maximum projections for each time point were generated in ImageJ Fiji before temporal projection in ImageJ Fiji.

Analysis of microtubule organisation

The analysis of cortical microtubule density and bundling used ImageJ Fiji and the ImageJ LPX package published in Higaki et al., 2010. From timelapse z-projection images, the spore perimeter was manually outlined. Microtubules were skeletonised using the LPX Filter2d with the Otsu method and a line extract value of 5. The skeletonised microtubules were masked by the spore outline at each time point and analysed by the LPX script. Microtubule bundling was measured by the skewness in the fluorescence intensity distribution along a microtubule, with higher intensity values indicating higher bundling. Microtubule density was measured as the number of segmented pixels in a cell divided by the cell area (npix/µm²). Graphs were created in Microsoft Excel.

Analysis of actin filament accumulation

The signal intensity profiles of GFP and mScarletI fluorophores around the spore perimeter were generated in ImageJ Fiji. A sum-of-slices projection of a 5.2 µm region surrounding the medial plane was created for each spore. The perimeters were outlined using the segmented line tool set at spline 5, starting at 12 o'clock and moving clockwise. The script then applied background subtraction, scaled the two fluorophore intensities to the mean and normalised the relative signal intensities. The resultant fluorophore intensity profiles, ranging from 0 to 1, were plotted against the distance around cell perimeter (µm) from the starting point. Plots were automatically generated using the ImageJ Fiji script.

Drug treatments

For dose selection, wild-type spores from a cross between Tak-1 and Tak-2 were grown on Gamborg media plates containing oryzalin or LatB dissolved in DMSO. The final oryzalin concentrations were 0.01 µM, 0.03 µM, 0.1 µM, 0.33 µM, 1 µM, 3.3 µM and 10 µM. The final LatB concentrations were 0.01 µM, 0.03 µM, 0.1 µM, 0.33 µM, 1 µM and 5 µM. As a control, 0.1% DMSO was used. Sporelings were grown for 4 days and imaged on days 2, 3 and 4 with the stereomicroscope.

For quantification of division asymmetry, spores from a cross between wild type and plants expressing pMpROP:mScarletI-N7 pMpUBE2:mScarletI-AtLTI6b were grown on a Gamborg media slab containing 0.1% DMSO, 0.1 µM LatB or 3.3 µM oryzalin within imaging chambers. Spores were grown for 48 h then imaged with the spinning disk. Spores were visually categorised into: undivided, divided but plane uncertain (e.g. new cell wall present but oriented such that the exact plane could not be determined), divided in an asymmetric plane, divided in a symmetric plane, and partially divided (e.g. had an incomplete new cell wall or two nuclei without a new cell wall). The latter three categories formed the scorable spore population from which each category percentage was calculated.

We thank Katharina Jandrasits and Magdalena Mosiolek for their great assistance in the lab. We acknowledge the BioOptics Facility and Plant Science Unit at the Vienna BioCenter Core Facilities (VBCF), Austria. In particular, we thank Pawel Pasierbek, Alberto Monerono Cencerrado and Thomas Lendl for their imaging and image analysis expertise. We also thank Maria Gravato-Nobre for her mentoring and great discussions, and Charlotte Kirchhelle for sharing the imaging chamber design and discussing experiments. We are grateful to Henrik Buschmann for insightful discussions and sharing reporter lines. We also acknowledge the valuable comments of three anonymous referees.

Author contributions

Conceptualization: S.T.A., L.D.; Methodology: S.T.A., H.M.; Formal analysis: S.T.A.; Investigation: S.T.A.; Resources: S.T.A., H.M., C.C.; Data curation: S.T.A.; Writing - original draft: S.T.A.; Writing - review & editing: S.T.A., H.M., L.D.; Visualization: S.T.A.; Supervision: L.D.; Project administration: L.D.; Funding acquisition: L.D.

Funding

This work was funded by a Scholarship from MoA Technology Ltd. to S.T.A.; a Biotechnology and Biological Sciences Research Council Doctoral Training Partnership Scholarship (grant no. BB/M011224/1) to H.M.; a Clarendon Scholarship of University of Oxford and the Sidney Perry Foundation to C.C.; and a European Research Council Advanced Grant, DE NOVOP (Project No. 787613) to L.D. Open Access funding provided by Gregor Mendel Institute of Molecular Plant Biology. Deposited in PMC for immediate release.

Data availability

All relevant data can be found within the article and its supplementary information.

The peer review history is available online at https://journals.biologists.com/dev/lookup/doi/10.1242/dev.202823.reviewer-comments.pdf

Special Issue

This article is part of the Special Issue ‘Uncovering developmental diversity’, edited by Cassandra Extavour, Liam Dolan and Karen Sears. See related articles at https://journals.biologists.com/dev/issue/151/20.

Attrill
,
S.
and
Dolan
,
L.
(
2024
).
KATANIN-mediated microtubule severing is required for MTOC organisation and function in Marchantia polymorpha
.
Development
151
,
dev202672
.
Bernard
,
F.
,
Lepesant
,
J. A.
and
Guichet
,
A.
(
2018
).
Nucleus positioning within Drosophila egg chamber
.
Semin. Cell Dev. Biol.
82
,
25
-
33
.
Bisgrove
,
S. R.
and
Kropf
,
D. L.
(
2001
).
Asymmetric cell division in fucoid algae: a role for cortical adhesions in alignment of the mitotic apparatus
.
J. Cell Sci.
114
,
4319
-
4328
.
Burk
,
D. H.
and
Ye
,
Z. H.
(
2002
).
Alteration of oriented deposition of cellulose microfibrils by mutation of a katanin-like microtubule-severing protein
.
Plant Cell
14
,
2145
-
2160
.
Buschmann
,
H.
and
Müller
,
S.
(
2019
).
Update on plant cytokinesis: rule and divide
.
Curr. Opin. Plant Biol.
52
,
97
-
105
.
Buschmann
,
H.
,
Holtmannspötter
,
M.
,
Borchers
,
A.
,
O'Donoghue
,
M. T.
and
Zachgo
,
S.
(
2016
).
Microtubule dynamics of the centrosome-like polar organizers from the basal land plant Marchantia polymorpha
.
New Phytol.
209
,
999
-
1013
.
Champion
,
C.
,
Lamers
,
J.
,
Shivas Jones
,
V. A.
,
Morieri
,
G.
,
Honkanen
,
S.
and
Dolan
,
L.
(
2021
).
Microtubule associated protein WAVE DAMPENED2-LIKE (WDL) controls microtubule bundling and the stability of the site of tip-growth in Marchantia polymorpha rhizoids
.
PLoS Genet.
17
,
e1009533
.
Chan
,
J.
,
Calder
,
G. M.
,
Doonan
,
J. H.
and
Lloyd
,
C. W.
(
2003
).
EB1 reveals mobile microtubule nucleation sites in Arabidopsis
.
Nat. Cell Biol.
5
,
967
-
971
.
Era
,
A.
,
Tominaga
,
M.
,
Ebine
,
K.
,
Awai
,
C.
,
Saito
,
C.
,
Ishizaki
,
K.
,
Yamato
,
K. T.
,
Kohchi
,
T.
,
Nakano
,
A.
and
Ueda
,
T.
(
2009
).
Application of Lifeact reveals F-actin dynamics in Arabidopsis thaliana and the liverwort, Marchantia polymorpha
.
Plant Cell Physiol.
50
,
1041
-
1048
.
Faure
,
J. E.
,
Rotman
,
N.
,
Fortuné
,
P.
and
Dumas
,
C.
(
2002
).
Fertilization in Arabidopsis thaliana wild type: developmental stages and time course
.
Plant J.
30
,
481
-
488
.
Hartman
,
K. S.
and
Muroyama
,
A.
(
2023
).
Polarizing to the challenge: new insights into polarity-mediated division orientation in plant development
.
Curr. Opin. Plant Biol.
74
,
102383
.
Higaki
,
T.
,
Kutsuna
,
N.
,
Sano
,
T.
,
Kondo
,
N.
and
Hasezawa
,
S.
(
2010
).
Quantification and cluster analysis of actin cytoskeletal structures in plant cells: role of actin bundling in stomatal movement during diurnal cycles in Arabidopsis guard cells
.
Plant J.
61
,
156
-
165
.
Honkanen
,
S.
,
Jones
,
V. A. S.
,
Morieri
,
G.
,
Champion
,
C.
,
Hetherington
,
A. J.
,
Kelly
,
S.
,
Proust
,
H.
,
Saint-Marcoux
,
D.
,
Prescott
,
H.
and
Dolan
,
L.
(
2016
).
The mechanism forming the cell surface of tip-growing rooting cells is conserved among land plants
.
Curr. Biol.
26
,
3238
-
3244
.
Huynh
,
J. R.
and
St Johnston
,
D.
(
2004
).
The origin of asymmetry: Early polarisation of the Drosophila germline cyst and oocyte
.
Curr. Biol.
14
,
438
-
449
.
Ishizaki
,
K.
,
Chiyoda
,
S.
,
Yamato
,
K. T.
and
Kohchi
,
T.
(
2008
).
Agrobacterium-mediated transformation of the haploid liverwort Marchantia polymorpha L., an emerging model for plant biology
.
Plant Cell Physiol.
49
,
1084
-
1091
.
Ketelaar
,
T.
,
Faivre-Moskalenko
,
C.
,
Esseling
,
J. J.
,
De Ruijter
,
N. C. A.
,
Grierson
,
C. S.
,
Dogterom
,
M.
and
Emons
,
A. M. C.
(
2002
).
Positioning of nuclei in Arabidopsis root hairs: An actin-regulated process of tip growth
.
Plant Cell
14
,
2941
-
2955
.
Kimata
,
Y.
,
Higaki
,
T.
,
Kawashima
,
T.
,
Kurihara
,
D.
,
Sato
,
Y.
,
Yamada
,
T.
,
Hasezawa
,
S.
,
Berger
,
F.
,
Higashiyama
,
T.
and
Ueda
,
M.
(
2016
).
Cytoskeleton dynamics control the first asymmetric cell division in Arabidopsis zygote
.
Proc. Natl. Acad. Sci. U. S. A
113
,
14157
-
14162
.
Kimata
,
Y.
,
Kato
,
T.
,
Higaki
,
T.
,
Kurihara
,
D.
,
Yamada
,
T.
,
Segami
,
S.
,
Morita
,
M. T.
,
Maeshima
,
M.
,
Hasezawa
,
S.
,
Higashiyama
,
T.
et al.
(
2019
).
Polar vacuolar distribution is essential for accurate asymmetric division of Arabidopsis zygotes
.
Proc. Natl. Acad. Sci. U. S. A
116
,
2338
-
2343
.
Kirchhelle
,
C.
and
Moore
,
I.
(
2017
).
A simple chamber for long-term confocal imaging of root and hypocotyl development
.
J. Vis. Exp.
55331
.
Li
,
R.
and
Gundersen
,
G. G.
(
2008
).
Beyond polymer polarity: how the cytoskeleton builds a polarized cell
.
Nat. Rev. Mol. Cell Biol.
9
,
860
-
873
.
Mulvey
,
H.
and
Dolan
,
L.
(
2023
).
RHO GTPase of plants regulates polarized cell growth and cell division orientation during morphogenesis
.
Curr. Biol.
33
,
2897
-
2911.e6
.
Munro
,
E.
and
Bowerman
,
B.
(
2009
).
Cellular symmetry breaking during Caenorhabditis elegans development
.
Cold Spring Harb. Perspect. Biol.
1
,
a003400
.
Munro
,
E.
,
Nance
,
J.
and
Priess
,
J. R.
(
2004
).
Cortical flows powered by asymmetrical contraction transport PAR proteins to establish and maintain anterior-posterior polarity in the early C. elegans embryo
.
Dev. Cell
7
,
413
-
424
.
Muroyama
,
A.
,
Gong
,
Y.
and
Bergmann
,
D. C.
(
2020
).
Opposing, polarity-driven nuclear migrations underpin asymmetric divisions to pattern arabidopsis stomata
.
Curr. Biol.
30
,
4467
-
4475
.
Muroyama
,
A.
,
Gong
,
Y.
,
Hartman
,
K. S.
and
Bergmann
,
D. C.
(
2023
).
Cortical polarity ensures its own asymmetric inheritance in the stomatal lineage to pattern the leaf surface
.
Science
381
,
54
-
59
.
O'Hanlon
,
M. E.
(
1926
).
Germination of spores and early stages in development of gametophyte of Marchantia polymorpha
.
Bot. Gaz.
82
,
215
-
222
.
Riedl
,
J.
,
Crevenna
,
A. H.
,
Kessenbrock
,
K.
,
Yu
,
J. H.
,
Neukirchen
,
D.
,
Bista
,
M.
,
Bradke
,
F.
,
Jenne
,
D.
,
Holak
,
T. A.
,
Werb
,
Z.
et al.
(
2008
).
Lifeact: a versatile marker to visualize F-actin
.
Nat. Methods
5
,
605
-
607
.
Roeder
,
A. H. K.
,
Otegui
,
M. S.
,
Dixit
,
R.
,
Anderson
,
C. T.
,
Faulkner
,
C.
,
Zhang
,
Y.
,
Harrison
,
M. J.
,
Kirchhelle
,
C.
,
Goshima
,
G.
,
Coate
,
J. E.
et al.
(
2022
).
Fifteen compelling open questions in plant cell biology
.
Plant Cell
34
,
72
-
102
.
Sakai
,
Y.
,
Higaki
,
T.
,
Ishizaki
,
K.
,
Nishihama
,
R.
,
Kohchi
,
T.
and
Hasezawa
,
S.
(
2022
).
Migration of prospindle before the first asymmetric division in germinating spore of Marchantia polymorpha
.
Plant Biotechnol.
12
,
5
-
12
.
Sauret-Güeto
,
S.
,
Frangedakis
,
E.
,
Silvestri
,
L.
,
Rebmann
,
M.
,
Tomaselli
,
M.
,
Markel
,
K.
,
Delmans
,
M.
,
West
,
A.
,
Patron
,
N. J.
and
Haseloff
,
J.
(
2020
).
Systematic tools for reprogramming plant gene expression in a simple model, Marchantia polymorpha
.
ACS Synth. Biol.
9
,
864
-
882
.
Schindelin
,
J.
,
Arganda-Carreras
,
I.
,
Frise
,
E.
,
Kaynig
,
V.
,
Longair
,
M.
,
Pietzsch
,
T.
,
Preibisch
,
S.
,
Rueden
,
C.
,
Saalfeld
,
S.
,
Schmid
,
B.
et al.
(
2012
).
Fiji: an open-source platform for biological-image analysis
.
Nat. Methods
9
,
676
-
682
.
Shimamura
,
M.
(
2016
).
Marchantia polymorpha: taxonomy, phylogeny and morphology of a model system
.
Plant Cell Physiol.
57
,
230
-
256
.
Vidali
,
L.
and
Hepler
,
P. K.
(
2001
).
Actin and pollen tube growth
.
Protoplasma
215
,
64
-
76
.
Vidali
,
L.
,
Rounds
,
C. M.
,
Hepler
,
P. K.
and
Bezanilla
,
M.
(
2009
).
Lifeact-mEGFP reveals a dynamic apical F-actin network in tip growing plant cells
.
PLoS One
4
,
e5744
.
Vogelmann
,
T. C.
,
Brassel
,
A. R.
and
Miller
,
J. H.
(
1981
).
Effects of microtubule-inhibitors on nuclear migration and rhizoid differentiation in germinating fern spores (Onoclea sensibilis)
.
Protoplasma
109
,
295
-
316
.
Wallner
,
E. S.
(
2020
).
The value of asymmetry: How polarity proteins determine plant growth and morphology
.
J. Exp. Bot.
71
,
5733
-
5739
.

Competing interests

L.D. and C.C. are co-founders of MoA Technology Ltd. L.D. is a non-executive director of MoA Technology Ltd.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed.

Supplementary information