Similarities and Differences in the GFP Movement in the Zygotic and Somatic Embryos of Arabidopsis

Intercellular signaling during embryo patterning is not well understood and the role of symplasmic communication has been poorly considered. The correlation between the symplasmic domains and the development of the embryo organs/tissues during zygotic embryogenesis has only been described for a few examples, including Arabidopsis. How this process occurs during the development of somatic embryos (SEs) is still unknown. The aim of these studies was to answer the question: do SEs have a restriction in symplasmic transport depending on the developmental stage that is similar to their zygotic counterparts? The studies included an analysis of the GFP distribution pattern as expressed under diverse promoters in zygotic embryos (ZEs) and SEs. The results of the GFP distribution in the ZEs and SEs showed that 1/the symplasmic domains between the embryo organs and tissues in the SEs was similar to those in the ZEs and 2/the restriction in symplasmic transport in the SEs was correlated with the developmental stage and was similar to the one in their zygotic counterparts, however, with the spatio-temporal differences and different PDs SEL value between these two types of embryos.


INTRODUCTION
Intercellular communication and the spatio-temporal regulation of gene expression are global mechanisms that control development. Plants have developed a unique structure, plasmodesmata (PDs), for intercellular communication in which each plant cell can form direct conduits to its neighbors, thus creating domains of cells that share common components. PDs are active channels that control the movement of the factors that regulate plant development (Heinlein, 2002;Sevilem et al., 2015;Otero et al., 2016;Sager and Lee, 2018 and literature therein).
The presence/absence and permeability of PDs lead to the formation of symplasmic domains, e.g., specialized groups of cells that become isolated either due to the absence of PDs or the downregulation of the cytoplasmic flux on the border of the domain (Bayer and Salmon, 2013;Kitagawa and Jackson, 2017 and literature therein). Such transient symplasmic domains may participate in the coordination of plant growth and development (Sager and Lee, 2014).
Why a symplasmic communication survey during plant development is important? What makes PDs an element of the supracellular information exchange system? By identifying which cells and tissues communicate through PDs, it is possible to determine when and where the signaling is related to the developmental processes. Signaling molecules, transcription factors and mRNA in embryogenesis requires understanding intercellular communication, a good knowledge of the establishment of the symplasmic domain in embryos of different origins is required. Therefore, the aim of the presented study was to analyze symplasmic communication in the SEs in order to determine whether the symplasmic domains that form in SEs correspond to the developing tissue and organs that is similar to their zygotic counterparts.

Plant Material and Culture Conditions
The STM:ER-GFP, STM:1XsGFP, STM:2XsGFP, and STM:3XsGFP transgenic lines were described in Kim et al. (2005a). The AtGL2:tmGFP9, AtGL2:GFP, AtSUC3:tmGFP9, and AtSUC3:GFP *Indicates that the exchange of the GFP occurred only between the hypocotyl tissues; 1 only in the provascular tissue; 2 in the protodermis and 3 in the ground promeristem.
Frontiers in Plant Science | www.frontiersin.org % -the percentage of embryos that enable GFP movement to the number of embryos tested (according to Kim et al., 2002;Wróbel-Marek et al., 2017). A large% value (bold%) indicates that movement was possible in this stage of embryo.
transgenic lines were described in Stadler et al. (2005) and PDBG2OE [PD-located beta 1,3 glucanases that is tagged internally with mCitrine was described in Benitez-Alfonso et al. (2013)]. The seeds of all of the lines were sown into pots with garden soil and vermiculite mixed at a 1:1 volume ratio. The plants were grown under controlled conditions at a temperature of 20-23 • C under a 16 h photoperiod with a light intensity of 40 µmol/m −2 s −1 and relative humidity of 60-70%. After 6-8 weeks, siliques with immature zygotic embryos (IZEs) were collected (Gaj, 2001), surface-sterilized for 20 min in a 20% sodium hypochlorite solution and rinsed three times in sterile water. The IZEs were isolated from the seeds in sterile dishes in water using preparation needles under a stereomicroscope. 10-15 IZEs (explants) were grown on a Phytagel solidified (Sigma, Poland; 3.6 g L-1) E5 medium (Sigma, Poland; Gamborg et al., 1968), which had been supplemented with 1.1 mg/ml 2,4dichlorophenoxyacetic acid (2,4-D, Sigma-Aldrich) and 20 g L −1 sucrose (pH 5.8). The embryo culture was conducted at 21 • C under a 16h photoperiod at a light intensity of 20 µmol/m −2 s −1 for up to 21 days. Next, SEs at various stages of development were collected. The analyses were repeated three times. The pictures on the plates show the figures that illustrate the representative results for each variant/replication. For the analyses of the ZEs, 45-71 embryos were tested for each line and the number of examined embryos ranged from 18 in the heart stage to 28 in the torpedo stage per one repetition. For the SEs, the total number of embryos that was analyzed was 65 on average and for each developmental stage, it was about 20 per one repetition. The data in the tables are from the documented and collected images that were taken during the study (a range of "n" = the number of embryos per line/stage; Tables 1-4).

Histochemical Staining
For the histological analyses, the samples were fixed in a solution of 2.5% (w/v) glutaraldehyde (GA) in a phosphate buffer (pH = 7.0) for 12 h at 4 • C. Then, they were embedded in Steedman's wax as was described in Sala et al. (2019). The sections (5-7 µm thick) were cut using a HYRAX M40 rotary microtome (Zeiss, Oberkochen, Germany) and collected on microscopic slides that were covered with Haupt's adhesive (according to Barlow and Kurczyńska, 2007). The sections were stained using the periodic acid-Schiff (PAS) reactions and toluidine blue (TBO, Sigma-Aldrich) staining (0.1% water solution of TBO for 5 min).
(Nikon, Tokyo, Japan). The GFP was excited at a maximum wavelength of 490 nm [Nikon Plan Fluor 10x objective lens (0.30 numerical aperture); 20x (0.5 numerical aperture); and 40x (0.75 numerical aperture)]. The histological images were acquired with a Nikon Eclipse Ni-U microscope equipped with a Nikon Digital DS-Fi1-U3 camera and software (Nikon, Tokyo, Japan).

Classification of the SE Stages
Spherical-shape embryos with an easily distinguishable protodermis and a diameter of about 100 µm were called globular. Heart-shaped, rod-like, or triangular shaped embryos with the cotyledon primordia and size (long axis) between 150 and 250 µm were called the heart. Embryos 250-400 µm long with distinguishable cotyledons were classified as torpedo. The length of the embryos in the cotyledonary stage was greater than 400 µm with a maximum of about 1000 µm (Table 1).

Symplasmic Communication Between the Embryo Organs
To compare the symplasmic communication between the embryo organs (along the apical-basal axis) in the Arabidopsis thaliana SEs with ZEs, GFP variants of different molecular sizes that were under the control of the SHOOT MERISTEMLESS (STM) gene promoter were used. In order to trace the mobility of the molecules of 27 kDa (1XsGFP), 54 kDa (2XsGFP), and 81 kDa (3XsGFP), they were compared with the GFP that had been retained in the endoplasmic reticulum (ER-GFP). The analyses concerned: 1/determining the promoter activity sites and 2/determining the distribution of the 1Xs, 2Xs, and 3Xs mobile GFP molecules (sGFP) at various stages of the ZEs and SEs embryo development. The STM gene promoter in the ZEs was active in the globular stage ( Figure 1A). In the heart ( Figure 1B) stage, the gene promoter activity was detected in the shoot apical meristem (SAM) and cells in nearest vicinity. At the torpedo stage ( Figure 1C), the area of promoter activity was detected in the cotyledon node and the ectopic expression of promoter activity was observed in some of the cells of the hypocotyl (to facilitate the description of the individual areas of embryos, especially SEs, the following terms were used: apicalcomprising the SAM, cotyledon node and cotyledons; centralcomprising the hypocotyl and basal -comprising the root pole). In the cotyledonary stage, the STM promoter activity was observed in the SAM and in the basal part of the hypocotyl (Figure 1D and inset).
In the SEs in the early globular stage, no STM promoter activity was observed ( Figure 1E) and this activity appeared in the embryos in the late globular stage (Figure 1E inset). In the heart stage, promoter activity was detected in the SAM and in the hypocotyl ( Figure 1F). In the torpedo stage ( Figure 1G), the activity of the STM promoter was observed in the cells of the emerging SAM (cotyledon node), the hypocotyl and the basal part of the embryo. In the embryos in the cotyledonary stage, a double distribution pattern of the promoter activity was observed: in the SAM cells (Figure 1H inset) and in the basal part of the hypocotyl ( Figure 1H). To summarize, the STM promoter activity in the heart stage SEs was not only found in the SAM but also in the hypocotyl cells and from the torpedo stage was similar to that described for the ZEs.
The distribution pattern of 3XsGFP, which is expressed under the STM promoter in different developmental stages of ZEs and SEs, was also compared. In the globular ZEs, 3XsGFP was distributed almost uniformly in the entire embryo ( Figure 1I). In the heart stage embryos, the GFP did not move from the sites of its expression or only moved into the cells in its nearest vicinity ( Figure 1J). In the torpedo stage, the GFP distribution pattern in the SAM was similar to the one that was observed for the heart stage, but additionally, the GFP was detected in the basal part of the hypocotyl (Figure 1K). In the cotyledonary stage, the 3XsGFP was detected in the SAM and the basal part of the hypocotyl ( Figure 1L). In the globular stage of the SEs, no fluorescence of the 3XsGFP was detected ( Figure 1M). In the heart stage, the GFP was detected in the hypocotyl and basal part of the embryo corresponding to novel subdomain( Figure 1N and Table 2). In the torpedo ( Figure 1O) and cotyledonary ( Figure 1P) stages, the 3XsGFP was present only in the embryo areas that corresponded to the sites of promoter activity. The results suggest that for molecules up to 81 kDa in the ZEs and SEs, three symplasmic domains were present from the torpedo stage ( Table 2).
The distribution pattern of the 1XsGFP and 2XsGFP was analyzed in both embryo types (Figure 2). The distribution of the 1XsGFP at different stages of the ZEs development showed that all of the domain boundaries permitted the passage of the 1XsGFP to spread from the STM expression site (Figures 2A-D).
In the SEs in the globular stage, the GFP was detected in the entire embryo ( Figure 2E). In the heart stage, the 1XsGFP was present in the entire embryo except for several layers of the cells at the distal parts of the cotyledons and the basal part of the embryo ( Figure 2F). This restricted movement of the 1XsGFP in the SEs ( Figure 2F) might indicate that novel subdomain boundaries must be established for the movement of the 1XsGFP from the STM expression site in the direction toward the distal part of the cotyledons. In the torpedo stage, the 1XsGFP was observed in the hypocotyl and the cotyledon node ( Figure 2G). The cotyledonary stage was characterized by the presence of the 1XsGFP only in the SAM, the basal part of hypocotyl and the root (Figure 2H and inset). These results indicate that in SEs, restrictions in symplasmic transport for molecules up to 27 kDa began in the heart stage of embryo development and from the torpedo stage led to the formation of the three symplasmic domains (apical, central and basal, that corresponded to the somatic embryo organs (cotyledon, hypocotyl and root). To summarize: (1) the distribution pattern of the GFP in the ZEs indicates that all of the domain boundaries permitted the passage of molecules up to 27 kDa in all of the developmental stages; for SEs, the distribution pattern of the 1XsGFP indicates the presence of the symplasmic domains and subdomains from the heart stage, and therefore, the domain boundaries had been established earlier than in ZEs; (2) a globular SEs and ZE are a single symplasmic domain in which the SEL of the PDs is at least 27 kDa; (3) in the heart stage SE, the SEL of the PDs between the symplasmic domains is equal to 27 kDa; (4) in the torpedo stage SE, there are three symplasmic domains: a cotyledon and a root meristem domain with the SEL of the PDs equal to or less than 27 kDa and a hypocotyl domain with the PDs SEL on the boundaries that are equal to or more than 27 kDa and (5) in the cotyledonary stage, three symplasmic domain are present ( Table 2).
An analysis of the 2XsGFP distribution in the ZEs showed that up to the heart stage, the 2XsGFP was observed throughout the entire embryo (Figures 2I,J). In the torpedo stage, the presence of the 2XsGFP was observed only in the hypocotyl (Figure 2K). This indicates a restriction in the GFP movement into the cotyledons and the basal part of embryos at this stage of development. In the cotyledonary stage, the 2XsGFP was detected in the basal part of the hypocotyl and the SAM (Figure 2L). In the globular stage SEs, no fluorescence of the 2XsGFP was detected ( Figure 2M). In the heart stage, the 2XsGFP was observed in groups of irregularly distributed cells in the hypocotyl (Figure 2N). In the SEs in the torpedo stage, the 2XsGFP was detected only in the SAM and root pole (Figure 2O). In the the cotyledonary stage SEs, the 2XsGFP was detected in the cotyledon node cells and in the basal part of the embryo (Figure 2P). To summarize: (1) restrictions in the movement of molecules up to 54 kDa began to occur in the torpedo stage of the ZEs; (2) the 2XsGFP did not move within the SEs to the same extent as it did in the ZEs; (3) the globular ZEs, which comprise one domain as the distribution pattern of 3XsGFP compared with that of ER-GFP, indicate that all of the PD can traffic molecules of at least 81 kDa; (4) in the heart stage, the 2XsGFP appeared to spread into the SE cotyledons from the hypocotyl expression zone (see the ER-GFP pattern in SE); (5) the distribution pattern of the 2XsGFP in the SEs seemed to be more restricted than in the 3XsGFP; (6) the distribution of the 2XsGFP in the torpedo SEs was observed only within the hypocotyl and there was little to no expression in the root pole and (7) in the SE, the 2XsGFP and 3XsGFP did not spread from the location of their expression, unlike in the ZEs.

Symplasmic Communication Between the Embryo Tissues
The sites of the AtGL2 promoter activity (Arabidopsis thaliana GLABRA 2) were analyzed using the transgenic lines AtGL2:tmGFP. The distribution pattern of the GFP molecule between the protodermis and underlying tissues was determined using the AtGL2:GFP transgenic line (AtGL2 promoter/GFP; in the AtGL2:tmGFP transgenic line, the GFP was fused to the C-terminus of the transmembrane helicase of the AtSTP9 monosaccharide transporter; Stadler et al., 2005).
The tmGFP expression site in the ZEs and SEs indicated that the AtGL2 promoter was inactive in the globular stage (not shown). It was activated in the heart stage and was expressed in the protodermal cells of the hypocotyl (Figure 3A). These sites of promoter activity persisted in the successive stages of the development of the ZEs and, in some cases, the fluorescence of tmGFP also occurred in the proximal part of the cotyledons  Figure 3B). There was a characteristic pattern in the distribution of the fluorescence of the tmGFP in the hypocotyl in the torpedo stage (Figure 3B inset), which was quite pronounced in the cotyledonary stage (Figure 3C and inset). The protodermal cells in which the tmGFP was expressed formed files along the long embryo axis and alternated with the cells that did not express the tmGFP. The AtGL2 promoter was only active in the protodermal cells (Figures 3B,C and insets).
In the SEs, promoter activity was mainly observed in the heart stage embryo's basal parts (Figure 3D). In the torpedo stage, the fluorescence of the tmGFP was mainly observed in the hypocotyl and also in a punctate pattern within the protodermal cells of the cotyledons (Figure 3E). Within the hypocotyl, the tmGFPexpressing cells formed irregular files along the organ's long axis, which alternated with the tmGFP-negative cell files (Figure 3E and inset). The expression of the tmGFP in the cotyledonary SEs was detected in the hypocotyl and cotyledons (Figure 3F). Similar to the ZEs, the AtGL2 promoter was active only in the protodermal cells of the SEs (Figures 3E inset,F inset). The results indicate that there are similarities in the sites of the AtGL2 promoter activity in the SEs and ZEs, but that in the SEs, the expression pattern of the tmGFP in the hypocotyl was quite irregular and was also visible in the cotyledons ( Table 3).
An analysis of the GFP distribution in the ZEs of the AtGL2:GFP line showed that in the heart stage, the GFP was detected throughout the entire embryo in both the protodermis and in the underlying cell layers (Figure 3G). A similar distribution of the GFP was observed for the   Figure 1). A-C,C insets,D,G,H,J -images from the epifluorescence microscope; E,E inset,F,I,K,L -images from the CLSM. Scale bars: A,A inset G -50 µm; B,C,C inset 1,D-F,H,I-K,K inset,L,L inset = 100 µm embryos in the torpedo stage ( Figure 3H). In the cotyledonary stage, the GFP fluorescence was only observed in the protodermal cells (Figure 3I and inset). To summarize, the GFP moves until (including) torpedo stage, within the entire embryo from the protodermis to the underlying tissues, thus indicating that the movement of molecules of 27 kDa through the PDs in centripetal direction was possible. In the cotyledonary stage, the protodermis is a symplasmic domain for molecules up to 27 kDa.
An analysis of the GFP distribution in the SEs of the AtGL2:GFP line, the GFP was not detected in the globular stage (not shown). In the heart stage, the GFP fluorescence was visible throughout the embryo's hypocotyl and embryo basal parts ( Figure 3J). In the torpedo stage, the GFP fluorescence was observed in the hypocotyl protodermis and in the underlying cell layers (Figure 3K inset) as well as in the apical part of the embryo (Figure 3K). In the embryos in the cotyledonary stage, the GFP fluorescence was observed in the hypocotyl and in the proximal and middle parts of the cotyledons. The GFP was not detected in the underlying cell layers ( Figure 3L and Table 3). The results indicate that there were restrictions in the symplasmic movement of the GFP between the protodermal cells in the cotyledonary stage in both the SEs and ZEs.
The transgenic lines AtSUC3:tmGFP and AtSUC3:GFP (Arabidopsis Suc-transporter3 AtSUC3 gene promoter) were used to design these constructs (Stadler et al., 2005). Using the AtSUC3:tmGFP line, the sites of the AtSUC3 promoter activity in the ZEs were examined first. The analysis showed that in the early stages of embryogenesis, the AtSUC3 promoter was active in the suspensor and the hypophysis (Figure 4A). In the torpedo stage, the promoter activity was observed in all of the cells in the basal part of the embryo ( Figure 4B) and in the cotyledonary stage, it was visible in the columella cells and the root cap peripheral cells ( Figure 4C, inset 1) as well as in the cotyledon provascular tissue ( Figure 4C, inset 2).
In the SEs, the AtSUC3 promoter was inactive in the globular stage (not shown). In the heart stage, the promoter activity was observed in the cells of the middle (hypocotyl) and basal (the root pole) parts of the embryo (Figure 4D). The area of the tmGFP-derived fluorescence at this stage covered a significant part of the embryo. In the torpedo stage, the tmGFP was expressed only in the cells in the basal part of the embryo (Figure 4E and inset). The expression of the tmGFP in the SEs in the cotyledonary stage was detected in the cells of the root and in the basal part of the hypocotyl (Figure 4F). The results indicate that the sites of the AtSUC3 promoter activity in the SEs and ZEs are similar and include the embryonic root surface cells; however, in the SEs, especially in the heart stage, the number of cells expressing the tmGFP was greater.
The GFP in the ZEs of the AtSUC3:GFP line in the heart stage was detected in the cells of the entire embryo ( Figure 4G). In the torpedo stage, the GFP fluorescence was observed in the basal and central parts of the embryo, where it was present in the protodermal cells, ground promeristem and provascular tissue ( Figure 4H). In the cotyledonary stage, the GFP fluorescence was observed in the basal part of the embryo and in discontinuous cell files (representing the provascular tissue) within the cotyledon (Figure 4I). The results indicate that in the cotyledonary stage of the ZEs, symplasmic isolation occurs between the embryo root and the other embryo organs and between the cells of the provascular tissue and ground promeristem ( Table 4).
The presence of the GFP in the SEs of AtSUC3:GFP in the heart stage was detected in the central and basal parts of the embryo (Figure 4J). In the torpedo stage (Figure 4K), the GFP fluorescence was seen in the protodermal cells of the entire embryo and in the ground promeristem cells in the basal part of the hypocotyl (Figure 4K and inset). In the cotyledonary stage, the GFP fluorescence was observed throughout the protodermal cells of the hypocotyl and cotyledons but the distribution pattern was patchy ( Figure 4L and Table 4).
The PdBG1OE-mCitrine line (PdBG1 -a Callose-Degrading Enzyme in PDs; Benitez-Alfonso et al., 2013) was used to determine the involvement of callose in the formation of the symplasmic domains during embryogenesis. This enzyme is directly involved in degrading the β-1,3 glucans and indirectly in modifying the callose deposition in the PDs. The PdBG1 tagged with mCitrine shows areas with a higher enzyme activity that corresponds to less callose deposition (Benitez-Alfonso et al., 2013). Present studies were performed on ZEs in the cotyledonary stage and SEs in the torpedo and cotyledonary stages, for which the symplasmic domains were determined and described above.
In the ZEs, the area without the PdBG1 was detected in the basal part of the embryos, which might indicate that there is a higher level of callose compared to the other embryo parts ( Figure 5A). In the SEs in torpedo and cotyledonary stages, the areas without the PDBG1 were localized in the cotyledon node and at the boundary between the hypocotyl and the root pole, that is, in the areas that corresponded to the distinguished symplasmic domains (Figures 5B,C).

Histology of the SE in the Different Developmental Stages
The studies on symplasmic communication in the SEs were accompanied by a histological analysis to define the histology of the SEs. The SE developmental stages were globular (Supplementary Figure 1A and inset), heart (Supplementary Figure 1B), torpedo (Supplementary Figure 1C), and cotyledonary (Supplementary Figure 1D). The embryos had a more or less spherical shape in the globular stage with an easily distinguishable protodermis (Supplementary Figure 1A). Embryos in the heart stage had cotyledon primordia, ground promeristem and provascular tissue (Supplementary Figure 1B and inset). The histological structure of the embryos in the torpedo stage was similar to that of the embryos in the heart stage (Supplementary Figure 1C). The SAM was rarely convex and was usually flat (Supplementary Figure 1C and inset). In the cotyledonary stage, the SEs had well-developed cotyledons, hypocotyl and embryonic root and protodermis, provascular tissue and a ground promeristem (Supplementary  Figures 1D,E). The SEs quite often had more than two cotyledons (not shown), fused hypocotyls and roots (Supplementary  Figures 1F,H) or had a malformed hypocotyl (Supplementary Figure 1G). The provascular tissue ran from the root meristem along the hypocotyl, then branched and passed into the cotyledons (Supplementary Figure 1E). The abnormalities in tissue arrangement and cytological features of the cells that comprised the tissues were distinct from the cotyledonary stage. The most pronounced malformations were detected in the ground promeristem and provascular strands ( Supplementary  Figures 1F-H). The files of the ground promeristem cells were not aligned (Supplementary Figure 1D inset) and in many of the SEs were composed of more cell layers than in their zygotic counterparts (Supplementary Figures 1C-H for comparison, a ZE is shown as an inset in Supplementary Figure 1G). The provascular tissue was well visible and like the ground promeristem was composed of more cell files than their zygotic counterparts (Supplementary Figures 1F,H and Table 4).

DISCUSSION
The establishment of the body pattern during embryogenesis, both zygotic and somatic, is under the control of auxin signaling and differential gene expression (Smertenko and Bozkhov, 2014;Horstman et al., 2017;Fehér, 2019;Tian et al., 2020 and literature therein). Increasing evidence had indicated that symplasmic communication is also involved in the control of embryogenesis (Xu et al., 2012;Brunkard et al., 2015;Choudhary et al., 2020;Godel-Jedrychowska et al., 2020 and literature therein) as well as postembryonic development (Sorkin and Nusinow, 2021;Sager et al., 2021 and literature therein). In the present study, the distribution pattern of the GFP within the SEs and ZEs at different developmental stages was studied to determine the spatio-temporal localization of the symplasmic domains that accompany the establishment of the embryo organs and tissues.
The results of the GFP distribution in the ZEs and SEs showed that: (1) in the SEs, the symplasmic domains for molecules up to 27 kDa can be distinguished from the heart stage; (2) in the ZEs, the symplasmic domains were established from the torpedo stage for molecules up to 54 kDa; (3) the symplasmic domains between the embryo tissues in the SEs is similar to the one in the ZEs; (4) a key difference between the ZEs and SEs is that in the SE, there is no expression of the STM in the globular stage, which might indicate that the apical-basal polarity is not established at this stage and (5) a restriction in symplasmic transport in the SEs and ZEs is correlated with the developmental stages ( Figure 6).

Symplasmic Domains and Embryo organ Development
During embryogenesis, along the apical-basal axis, the SAM, cotyledons, hypocotyl and radicula are determined and in the radial direction, the protodermis, ground promeristem and provascular tissues are established (Laux et al., 2004). Achieving such an organization requires cell specification in an integrated manner (Laux et al., 2004;Radoeva et al., 2019 and literature therein). The involvement of symplasmic communication/restriction in embryogenesis was first described for development in Arabidopsis ZEs. Studies using the GFP as a mobile fluorescent protein have shown that the symplasmic domains accompany the development of the embryo organs and are established by the mid-torpedo stage with the PDs SEL of 54 kDa at the organ boundaries (Kim et al., 2005a,b). Similar symplasmic domains were detected in the Arabidopsis SEs, but these subdomains appeared in the heart stage and the PDs SEL at their boundaries was determined to be 27 kDa (Figure 6). Regardless of the identified differences, the results support the hypothesis that restrictions in symplasmic communication was correlated with embryo development and the idea that postulates the participation of the PDs as control "points" for the movement of signals during embryogenesis (Otero et al., 2016;Sager and Lee, 2018 and literature therein). The question then arises of whether the identified differences between the ZEs and SEs are developmentally significant. It seems not because the correlation between the emerging domains and the developing embryo organs is clear, and therefore, from a qualitative point of view, there are no differences between the SEs and ZEs in terms of the correlation between the embryo development and the formation of the symplasmic domains. The reason that the limitations in symplasmic communication appear earlier in the SEs than in the ZEs is unknown. It can be presumed that they arise from the morphological heterogeneity (a greater number of cotyledons, the malformation of the SAM and RAM in SEs in comparisons to the ZEs) between the SEs and ZEs that have been described FIGURE 6 | The symplasmic domains identified in the longitudinal direction (between the embryo organs; dotted lines indicate the embryo parts; apical central; basal). In the radial direction -between the embryo tissues; dark green -protodermis; light green -ground promeristem; olive -provascular tissue; octagon -1XsGFP; triangle -2XsGFP; star -3XsGFP; yellow circle -SAM; greenish circle -GFP expressed under pSUC3; arrows -the movement in the direction indicated by the arrowhead; a crossed arrows indicates no movement in the direction indicated by the arrowhead. in many species (Dodeman et al., 1997;Mordhorst et al., 1998;Pullman et al., 2003;Etienne et al., 2013;Jariteh et al., 2015) and that was observed in the present study. Such heterogeneity may be the result of disturbances in the spatio-temporal establishment of the apical-basal and radial polarity of the SEs. The detected differences could also have resulted from the diverse capacity of the PDs to transport molecules in these two types of embryos. The GFP movement is (generally) a passive, diffusion-driven transport. Such transport is a function (among others) of the area of passage, the length of the PDs, the wall effects and the electrochemical potential differences between adjacent cells (Liarzi and Epel, 2005;Dashevskaya et al., 2008). It cannot be ruled out that these parameters are different in the ZEs and SEs, at least in the early stages of development. The the shape of the PDs can also influence the GFP movement between organs/tissues (Dashevskaya et al., 2008;Amsbury et al., 2018 and literature therein). Thus, in future studies, the cell wall thickness and the the shape of the PDs in the SEs must also be evaluated.

Radial Pattering of Embryo and Symplasmic Domains
The results showed that using two transgenic lines, it was possible to trace the GFP distribution pattern between the embryo tissues during embryogenesis. The epidermis has been shown to become symplasmically isolated from the underlying shoot/embryo tissues for tracer dyes and several transcription factors and the reasons for this have previously been discussed (e.g., Roberts and Oparka, 2003;van Bel, 2018). The symplasmic communication between the protodermis and underlying tissues in the ZEs occurred freely in the heart and torpedo stages, thus indicating that these embryos were a single symplasmic domain (Stadler et al., 2005). The results for the SEs were similar to those that were obtained for their zygotic counterparts. In the cotyledonary stage, the protodermis was a distinct symplasmic domain in the ZE and SE, thus indicating that the protodermis, at least for molecules equal to or greater than 27 kDa, was isolated from the underlying tissues. Why is it important to isolate the protodermis as a separate symplasmic domain? Perhaps, this covering tissue must be specified because in the post-embryonic development, it differentiates into several different cell types, but whether this is the only reason is unknown.
Studies on the ZEs of Arabidopsis showed that the embryos in the heart stage that had been derived from the AtSUC3 promoter/GFP plants were a single symplasmic domain (Stadler et al., 2005). In the torpedo stage, only the hypocotyl was a single domain, but in the fully developed embryos, there was restricted movement between the embryo tissues (Stadler et al., 2005). It appeared that in the SEs, the symplasmic domain occurred earlier in the temporal sense but that it was similar to their zygotic counterparts in qualitative sense (Figure 6).

PD SEL Changes and Embryogenesis
The PDs SEL is regulated during development (Sager and Lee, 2018;Petit et al., 2020 and literature therein) and can be changed by callose deposition (Benitez-Alfonso et al., 2013;Wu et al., 2018;Li et al., 2020 and literature therein). Callose turnover in the PDs plays a key role in different developmental processes (Benitez-Alfonso et al., 2013 and literature therein), including embryogenesis (Du et al., 2020). The reason for the difference in the PDs SEL between the ZEs and SEs is not known. Because the PDs SEL is associated with callose deposition, it seems reasonable to look at this mechanism for an explanation of the detected difference. Results describing the symplasmic communication between the embryonic and nonembryonic areas of an Arabidopsis explant indicated that callose deposition at the PDs is a prerequisite for changing the cell fate (Godel-Jedrychowska et al., 2020). The present results using the PdBG1OE-mCitrine line indicate that callose degradation was lower on the boundaries of the distinguished symplasmic domains along the apical-basal axis. These results support the role of the PDBG1 in callose deposition in the PD and indicate that the establishment of symplasmic domains is important for embryogenesis independent of the origin of an embryo.

CONCLUSION
Despite the detected differences in the the spatio-temporal diversity in the formation of the symplasmic domains, there was a clear correlation between the identified domains and the embryo development independent of origin of an embryo (Figure 6). This may indicate that symplasmic communication, which is based on the restrictions of the symplasmic transport of signals, is a mechanism that is involved in regulating embryogenesis.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

AUTHOR CONTRIBUTIONS
KK-Ł, KG-J, and EK designed the experiments. KK-Ł and KG-J performed the experiments and analyzed the data. EK and KG-J wrote the manuscript. All the authors provided feedback on the manuscript.