Introduction

Buckwheat (Fagopyrum spp.) is described as a neglected and underutilised species of cultivated plants; however, it is an attractive crop; it is a good source of, among others, flavonoids, gluten-free proteins, amino acids, dietary fibre, vitamins, minerals (Huda et al. 2021; Li et al. 2019). Of the 23 species of Fagopyrum, only two are F. esculentum and F. tataricum are cultivated (Tomasiak et al. 2022). F. tataricum is characterised by over 100-fold higher rutin content than F. esculentum (Fabjan et al. 2003; Zhang et al. 2021). However, low grain content and strong bitter taste limit its use as food. F. esculentum is more widely distributed than F. tataricum; however the seed yield of this species is often relatively low and unstable (Sytar et al. 2016). One of the most crucial reasons for the low yield is self-incompatibility, as F. esculentum plants are dimorphic with two types of flowers: Thrum (pistils shorter than stamens) and Pin (pistils longer than stamens) (Cawoy et al. 2009). F. tataricum produces only homostylous flowers capable of self-fertilization.

A somatic hybridization technique has been proposed to overcome breeding barriers in buckwheat (Nešković et al. 1987). Among the different cell fusion methods for developing somatic hybrids in plants, PEG-mediated fusion has been the most popular and used so far. Lachmann et al. (1994) obtained the hybrid calli of F. esculentum (+) F. tataricum without further plant regeneration. The other method is protoplast electrofusion, which acquired a preferential position due to its rapidity, efficiency and low cytotoxicity (Rems et al. 2013). Since the first description of the electrofusion process by (Senda et al. 1979), it has been successfully applied to different species of agricultural importance (Mackowska et al. 2023; Sedlak et al. 2022).

In Fagopyrum spp., plant regeneration from protoplasts has been described for F. esculentum from hypocotyl- (Adachi et al. 1989) and callus-derived protoplasts (Zaranek et al. 2023b) while for F. tataricum from hypocotyl- (Lachmann and Adachi 1990) and callus-derived protoplasts (Zaranek et al. 2023a). However, to obtain plant regenerants from protoplast cultures, one of the first and crucial stages is cell wall re-synthesis and deposition, which enables the first cell divisions. The cell wall is a dynamic and highly controlled structure mainly composed of cellulose microfibrils, matrix polysaccharides, polyphenolic compounds, and structural proteins in variable amounts (Showalter 1993). Due to its compounds’ structural and enzymatic rearrangements, the cell walls play an important role in regulating the morphogenetic processes and maintaining the cellular differentiation status of cells (Knox et al. 1991; Potocka et al. 2018). Cellulose, hemicellulose and pectins, arabinogalactan proteins (AGPs) and extensins (EXTs) all participate (to different degrees) in cell expansion or adhesion and as regulatory molecules they are key determinants of the physical properties of the cell wall (Willats et al. 2001).

Cell wall regeneration from protoplasts has been studied in several species using various techniques (Kuki et al. 2020; Parmentier et al. 1995; Wiśniewska and Majewska-Sawka 2007). However, due to the complexity of the cell wall structure, the most informative method is immunocytochemical detection of specific cell wall antigens (Godel-Jędrychowska et al. 2019; Majewska-Sawka and Münster 2003). To date, there is no data about cell wall reconstruction from protoplasts in any of the Fagopyrum species, as well as in their hybrids or the hybrid protoplasts at all, independently of the species. Thus, our goal was to check whether there are differences between those two species and their hybrids (Ft + Fe) during wall regeneration that could point out the potential recalcitrance of hybrid protoplasts. Since xyloglucan, pectins, extensins, and AGPs are essential components of dicotyledonous walls, the specific antibodies that recognise their epitopes were chosen to follow the process of cell wall rebuilding: LM20, LM5 and LM6 recognising pectic epitopes; JIM13 and JIM16 recognising AGPs epitopes; JIM20 recognised extension epitope and LM25 recognising xyloglucan epitope. It should be emphasized that this is the first study that illustrates and compares the regeneration of the wall of Fagopyrum parental protoplasts and their hybrids. Moreover, this is the first record of obtaining hybrids in buckwheat by electrofusion.

Materials and methods

Protoplast isolation

Embryogenic calli of F. esculentum and morphogenic calli of F. tataricum were used as sources of protoplasts. The callus lines were obtained from immature embryos of both species and maintained in the dark on RX medium (Table S1) at 26 ± 1 °C as Betekhtin et al. (2017) described. The subcultures were carried out every two weeks under the same conditions.

The isolation was performed following the protocol of Zaranek et al. (2023b) with minor modifications. Two grams of 10-day-old callus of F. tataricum and 1 g of 12-day-old callus of F. esculentum were incubated for 1 h in 10 ml of PSII/F plasmolysis solution (Table S1) in the dark at room temperature (RT). Then, the solution was removed, and 10 ml of E1 and E2 enzyme solutions (Table S1) were added to F. tataricum and F. esculentum, respectively. The material was incubated for 16 h with gentle shaking (50 rpm, RT). The suspensions were filtered through a 100 μm pore mesh and centrifuged for 5 min (1000 rpm, RT). The pellet was resuspended in 8 ml of Suc/MES solution (Table S1). Next, 2 ml of W5 solution (Table S1) was carefully overlaid to create a gradient and centrifuged for 10 min (1200 rpm, RT). The ring of viable protoplasts between the two phases of the gradient was collected, resuspended in W5 solution up to 10 ml and centrifuged for 5 min (1000 rpm, RT). The pellet was resuspended in 1.5 ml of mannitol solution (Table S1), and the protoplast concentration was adjusted to 8 × 105 cells ml− 1. The isolation was performed twice for both species, and one set was kept as control and as nurse culture.

Electrofusion

The electrofusion methodology was adapted from the protocol proposed by Mackowska et al. (2023). One ml of the F. esculentum and F. tataricum protoplast suspensions were added to the tubes with Rhodamine B (Rhd) and 3,3’-dihexyloxacarbocyanine iodide (DiOC6), respectively and incubated for 10 min in the dark at RT. The preparation of dyes is listed in the Supplementary Information: Table S2.The following washing steps took place on ice. The stained protoplasts were centrifuged for 5 min (1000 rpm, 4 °C), and the pellet was resuspended in 10 ml of ice-cold mannitol solution. These steps were repeated two times more. The pellets were resuspended in 0.9 ml of ice-cold mannitol. The protoplast concentration was adjusted again to 8 × 105 cells ml− 1, adding ice-cold mannitol solution if needed. For parental protoplasts analysis, 1 ml of the protoplast solution of each species was stained with 100 µl of calcofluor white (Table S2) and kept in the dark for 10 min.

Respectively, 0.4 ml of protoplasts stained with Rhd and 0.4 ml stained with DiOC6 were carefully mixed to combine F. tataricum (+) F. esculentum (Fe + Ft, heterokaryon), as shown in Fig. 1. The electrofusion was performed in the Super Electro Cell Fusion Generator 21 (NEPAGEN, Japan) using parameters optimised by our team, standardised for Fagopyrum protoplasts. The 0.8 ml CUY497P2 electrode was used with 5 V of alternating current (AC) for 40 s and 12 V of direct current (DC) for 50 µs. The fused protoplasts were transferred into a sterile tube covered with foil and kept on ice.

Fig. 1
figure 1

General scheme of the protoplast electrofusion and selection technique. Once the protoplasts are isolated, they are stained with two contrasting fluorochrome solutions (Rhd– red; DiOC6– green) and fused using periodic electrical pulses. Electrofusion is a random process, so the hybrid protoplasts (yellow) must be selected, one by one, under the microscope. In the case of the parental protoplasts, calcofluor staining allows the subsequent selection of those from which the cell wall was completely removed (no calcofluor signal– grey). The culture of the hybrid and parental protoplasts must be developed after embedding them in agarose beads, and they must be accompanied by a nurse culture that contributes to their development. AC– alternating current; DC– direct current; DiOC6–3,3’-dihexyloxacarbocyanine iodide; Rhd– rhodamine B. This figure was created with BioRender.com. BioRender certificate confirming the publication rights is available upon request from the authors.

Full size image

Selection of protoplasts

These steps were performed entirely in the dark. After electrofusion, 0.25 ml of the protoplast suspension was resuspended in 4.5 ml of ice-cold mannitol and carefully mixed. In a sterile Ibidi µ-Dish 35 mm, low, 0.3 ml of this suspension was placed and allowed to settle for 5 min in the inverted microscope. The protoplasts that simultaneously showed fluorescence for Rhd and DiOC6 were collected using a stripper tip MXL3-75 (ORIGIO Inc, USA) and micromanipulator TransferMan® 4r (Eppendorf, Germany). About 200 hybrid cells were collected and kept on ice until the next steps were performed. The selection process was applied for the parental protoplasts, F. tataricum and F. esculentum. Approximately 200 protoplasts that did not show fluorescence for calcofluor white were collected.

Culture of selected protoplasts

The parental and hybrid protoplasts selected into Eppendorf tubes were resuspended in 9 µl of LMPA (Table S1), and about 10 µl beads were made on polystyrene Petri dishes (Ø6 cm × 1.2 cm) at the rate of five beads per dish (Fig. 2). Non-stained Fe and Ft protoplasts were used as nurse culture during parental and Fe (+) Ft hybrid cells culture. For this, equal volumes of LMPA and the protoplast solution with a concentration of 8 × 105 cells ml− 1 were mixed and placed in three 50 µl beads around the parental and hybrid cells as shown in Fig. 1. Then 4 ml of BM medium supplemented with 4 µl of PSK and 2 µl of Timentin (Table S1) were added to each Petri dish and incubated for 4, 12, 24, 48 and 72 h at 26 ± 1 °C in the dark.

Fig. 2
figure 2

General scheme of the protoplasts immunostaining process. This figure was created with http://BioRender.com. BioRender certificate confirming the publication rights is available upon request from the authors.

Full size image

Protoplasts immunostaining

After each time point, the beads with the F. tataricum and F. esculentum control and hybrid cells were transferred to a 2 ml Eppendorf and fixed with 4% paraformaldehyde (PFA) and 1% glutaraldehyde (GA) in phosphate-buffered saline (PBS, pH = 7.2) overnight at 4 °C (Table S3; Fig. 2). For the 0 h time point; the beads were fixed immediately after they were made. Next, samples were washed with PBS and incubated with blocking buffer (BB) for 30 min (Table S3) and with primary antibodies (Table 1) for 1.5 h.

Negative controls were performed by incubation in BB instead of the primary antibodies. After several rinses with the BB, a secondary antibody conjugated with Alexa Fluor 488 goat anti-rat IgG antibody was applied for 1.5 h in the dark. Next, the material was washed with PBS and counterstained with 0.01% Fluorescent Brightener 28 (calcofluor white) in PBS solution for 20 min.

Table 1 Antibodies used for immunostaining of protoplasts and their respective epitopes in the cell wall
Full size table

Observations

The protoplast beads were individually placed on a glass slide and flattened by the coverslip. The observations and photographic documentation were carried out using an Olympus FV-1000 confocal system (Olympus, Hamburg, Germany) equipped with an Olympus IX81 inverted microscope, a 405-nm diode laser, and a multi-line argon ion laser (excitations 457/488/515 nm; Melles Griot BV, Didam, Netherlands). A series of two-dimensional images of z-stacks (the optical sections through the cells) were taken using two separate photomultipliers. 2D image processing was performed using ImageJ version 1.53s software (Wayne Rasband, National Institutes of Health, USA), and 3D images were performed with Imaris 9.5 software (Bitplane, Zürich, Switzerland). Moreover, 50 cells per antibody were also visualised for statistical analysis with an epifluorescence microscope Nikon Eclipse Ni-U microscope equipped with a Nikon Digital DS-Fi1-U3 camera with the corresponding software (Nikon, Tokyo, Japan). Cells have been scanned in two channels for excitation of Alexa 488 fluorochrome; the 488 nm laser was used, and for FB28, the 405 nm laser. To facilitate the interpretation of the obtained results, the percentage of signal presence from specific epitopes was calculated (number of cells containing signal per total number of cells). Thus, three ranges were specified, where 0–30% means the absence or presence of a signal in a small number of cells; 31–65% occurrence of a signal in the average number of cells; 66–100% occurrence of a signal in most or all cells.

Results

In the presented studies, a spatio-temporal analysis of cell wall regeneration was carried out on protoplasts derived from the callus of two cultivated species of buckwheat (F. tataricum and F. esculentum) and on their hybrid (heterokaryon: Ft + Fe). Protoplasts were analysed at various time points: 0 h, 4 h, 12 h, 24 h, 48 h, and 72 h. The wall components that were studied included cellulose (calcofluor staining), pectins (LM5, LM6, LM20), arabinogalactan proteins (JIM13, JIM16), xyloglucan (LM25) and extensin (JIM20). A comparison of protoplasts from two species and hybrids revealed similarities and differences in the cell wall composition at different time points (Table S4).

Arabinogalactan proteins

AGP epitope recognised by the JIM13 antibody occurred in all analysed types of protoplasts (F. tataricum, F. esculentum and hybrids) and was present at each of the analysed time points, from 0 h to 72 h (Fig. 3A– F, G– L, M– R; Table S4; Supplementary videos S1– S6). This epitope was located within the cell and cell membrane/cell wall at later stages. It was also noticed that the signal from the JIM13 antibody, from 12 h to 72 h, correlated with the arrangement of cellulose microfibrils that were visualised with calcofluor staining (Fig. 3 compare D and D’, E and E’, F and F’, K and K’, L and L’, P and P, Q and Q’). The presence of JIM13 was also observed outside the cells (Fig. 3, e.g. E, I, J, O). The character of the signal was mainly continuous.

Fig. 3
figure 3

The presence of JIM13 AGPs epitope during cell wall regeneration in F. tataricum (A-F), F. esculentum (G-L), and Ft + Fe hybrid (M-R) with the corresponding signal from calcofluor (A’-R’)

Full size image

JIM16 antibody, directed against the AGP glycan component, was also observed at every stage of cell wall development for each type of analysed cell (Fig. 4A– F, G– L, M– R; Supplementary videos S7– S12). However, a reduced number of cells (average range) with the presence of JIM16 was observed for F. tataricum at 72 h, F. esculentum at 24 h and heterokaryons at 48 h (Fig. 4; Table S4). This AGP epitope outside the cell was also observed at each stage of development examined (Fig. 4, e.g., J, K, O, P). Moreover, a difference in the JIM16 distribution was observed. For F. tataricum, from 0 h to 24 h, the JIM16 signal was continuous with few areas of higher intensity (Fig. 4A– D). At 48 h and 72 h, the JIM16 epitope did not occur on the entire cell wall but was only present in some regions (Fig. 4E, F). In the case of F. esculentum, a continuous signal was observed at 0 h, 4 h, 12 h and 48 h, where areas with increased signal intensity were also observed (Fig. 4G, H, I, K). However, at 24 h and 72 h, the signal was present only in some areas of the cell wall (Fig. 4J, L). For the heterokaryons, the continuous signal was present at 0 h, 12 h, 24 h and 72 h (Fig. 4M, O, P, R); however, for 4 h and 48 h, JIM16 was localised within the walls in a patch-like manner (Fig. 4N, Q).

Fig. 4
figure 4

The distribution of JIM16 AGPs epitope during cell wall reconstruction in F. tataricum  (A-F), F. esculentum (G-L), and Ft + Fe hybrid (M-R) with the corresponding signal from calcofluor (A’-R’)

Full size image
Fig. 5
figure 5

JIM20 extensin epitope occurrence during wall reconstruction in F. tataricum (A-F), F. esculentum (G-L), and Ft + Fe hybrid (M-R) with corresponding signal from calcofluor (A’-R’)

Full size image

In summary, the analysed AGP epitopes were present in all cell types, from removing the cell wall (time 0 h) until its reconstitution (time 72 h; Table S4), and were present in the membrane and/or wall. Moreover, the signal outside the cell was also observed. The observed differences concerned the signal’s character, especially for JIM16.

Extensin

In F. tataricum protoplasts, the JIM20 antibody, which detects extensin epitopes, was present in several cells at the initial stages of wall regeneration (0 h, 4 h; Table S4). In the cells where the signal was detected, it was present in the intracellular compartments (Fig. 5A, B; Supplementary videos S13– S18). An increase in the number of cells with the occurrence of extensins was observed at 12 h and 24 h, followed by a decrease at 48 and 72 h (Fig. 5 compare C– F). If present, the signal from this antibody was continuous except for 24 h, where it had a patch-like character (Fig. 5D). Moreover, it was noticed that at 12 h, 48 h and 72 h in cells where JIM20 was noticed, the signal from the antibody corresponded to the signal from calcofluor staining (Fig. 5 compare, e.g. E and E’).

JIM20 in F. esculentum protoplasts at 0 h and 4 h was observed in less than half of the analysed cells. However, when the signal was present, it was localised in the intracellular compartments (Fig. 5G, H). Interestingly, at 12 h there was an increase in cells with the JIM20 signal (Fig. 5), followed by a significant decrease at 24 h, and the signal, if present, was dotted (Fig. 5I, J). At 48 h and 72 h, the extensin epitope was observed in some cells, and its character was dotted and patch-like, respectively (Fig. 5K, L).

During wall reconstruction in hybrid protoplasts, the JIM20 epitope was present in most cells at 12 h and 24 h. However, at 0 h and 72 h, the JIM20 signal was observed only in some cells, while at 4 h it was present in a small number of cells, and at 48 h it was absent. At the initial stages of culture, at 0 h and 4 h, the signal, if present, was detected in intracellular compartments (Fig. 5M, N). At 12 h, 24 h, and 72 h, the epitope was present in areas corresponding to those where the calcofluor signal was observed (Fig. 5 compare O and O’, P and P’, R and R’). In addition, a punctate signal was observed around the cells at all stages (Fig. 5, e.g. O and P).

In summary, the extensin epitope recognised by the JIM20 antibody showed variable occurrence at different time points in all analysed variants (Fig. 5; Table S4). It can be noted that for all cell types at 12 h, the epitope was detected in most cells. At the remaining time points, differences in the number of cells with a positive JIM20 signal were observed between the hybrid and the two Fagopyrum species.

Xyloglucan

At 0 h, the epitope recognised by the LM25 antibody was absent from a minor part of the protoplasts in most of the protoplasts (Fig. 6A– F, G– L, M– R; Supplementary videos S19– S24); however, the fluorescence signal was weak or of moderate intensity (Fig. 6A, G, M), exhibiting a mesh-like pattern. In the next time point, LM25 was detected predominantly in intracellular compartments (Fig. 6B, H, N). From the 12 h time point, the wall was rebuilt by protoplasts (Fig. 6C’– F’, I’– L’, O’– R’) until the 72 h cell wall was restored. The LM25 epitope was present abundantly in all the protoplasts, regardless of genotype, during these time points. The fluorescence signal was observed in the wall matrix as continuous strands that corresponded with the course of cellulose microfibrils (Fig. 6C– F, I– L, O– R).

Fig. 6
figure 6

LM25 xyloglucan epitope distribution during regeneration of cell wall in F. tataricum (A-F), F. esculentum (G-L), and Ft + Fe hybrid (M-R) with the corresponding signal from calcofluor (A’-R’)

Full size image

In summary, at 0 and 4 h, there were differences in the number of protoplasts with detected fluorescence signal (Fig. 6; Table S4), with F. esculentum in average rate at 0 h and F. esculentum and hybrid protoplasts in average rate at 4 h. The LM25 epitope was abundantly observed in most cells during the later stages of wall regeneration regardless of the genotype (Fig. 6; Table S4). As mentioned above, no differences in the distribution pattern within the newly arisen wall matrix were observed.

Pectins

Methyl-esterified homogalacturonan (pectins)

The epitope recognised by the LM20 antibody (Fig. 7A– F, G– L, M– R; Supplementary videos S25– S30) at 0 h occurred predominantly in hybrid protoplasts compared to F. tataricum and F. esculentum protoplasts (Fig. 7A, G, M). At 0 h, the punctate fluorescence signal was detected within cytoplasmic compartments, and in the further time points (4 h, 12 h), this occurrence became more progressed, if present (Fig. 7 compare A– C, G– I, M– O).

Fig. 7
figure 7

The presence of LM20 pectic epitope during cell wall rebuilding in F. tataricum (A-F), F. esculentum (G-L), and Ft + Fe hybrid (M-R) with the corresponding signal from calcofluor (A’-R’)

Full size image

In F. tataricum protoplasts, the LM20 epitope was observed abundantly in intracellular compartments in the following 12, 24 and 48 h time points (Fig. 7C, D and E). In 72 h, epitope occurred in the wall matrix and was detected in a punctate manner (Fig. 7F; Supplementary video S26). In F. esculentum protoplasts, from 24 h and further, a decrease in LM20 epitope occurrence could be seen; namely, epitope was detected but only as a punctate signal within the matrix (Fig. 7J– L).

In hybrid protoplasts at 24 and 48 h, LM20 epitope was detected in the wall matrix; however, at 72 h, only a low number of protoplasts displayed fluorescence signal, with most of the cells that were devoid of LM20 epitope (Fig. 7R; Supplementary video S30). In some of the analysed protoplasts, the fluorescence signal correlated with cellulose microfibril arrangement (Fig. 7 compare F and F’, K and K’, O and O’, P and P’, Q and Q’). The LM20 epitope mainly occurred in a low or average number of cells (Table S4). At 0 and 4 h, hybrid protoplasts exhibited fluorescence signals more frequently than F. tataricum or F. esculentum protoplasts. From 12 h, the LM20 epitope was detected in F. tataricum more often than in F. esculentum or hybrid protoplasts. However, unlike other analysed epitopes, the LM20 epitope was not detected abundantly within the wall matrix.

Galactan (RGI, pectins)

The epitope recognised by the LM5 antibody (Fig. 8A– F, G– L, M– R; Supplementary videos S31– S36) was detected in F. tataricum protoplasts in intracellular compartments at 0–12 h (Fig. 8A– C), intracellular compartments/wall matrix at 24 h (Fig. 8D) and in wall matrix in the following 48 and 72 h (Fig. 8E, F; Supplementary video S32). A similar pattern of occurrence was observed in F. esculentum and hybrid protoplasts; however, the wall-bound epitope appeared earlier, from 12 h time point for F. esculentum (Fig. 8I– L) and even 4 h for heterokaryons (Fig. 8N– R). The alignment of cellulose microfibrils corresponded with the fluorescence signal, which was depicted in LM5 epitope distribution in different time points regardless of genotype (Fig. 8 compare E and E’, F and F’, I and I’, K and K’, L and L’, P and P’, Q and Q’, R and R’).

Fig. 8
figure 8

LM5 pectic epitope distribution during wall reconstruction in F. tataricum (A-F), F. esculentum (G-L), and Ft + Fe hybrid (M-R) with corresponding signal from calcofluor (A’-R’)

Full size image

In summary, the fluorescence signal was detected in significant part of the protoplasts (Table S4) with some differences at 0 h (F. tataricum and F. esculentum belonging to the average range, in hybrid protoplasts– expressed in more prominent number of cells), 24 h (average range for F. tataricum in comparison to F. esculentum and hybrid) and 48 h (average range for F. esculentum in comparison to F. tataricum and hybrid).

Arabinan (RGI, pectins)

The epitope recognised by the LM6 antibody (Fig. 9A– F, G– L, M– R; Supplementary videos S37– S42) was detected abundantly in intracellular compartments of F. tataricum protoplasts up to 24 h time point (Fig. 9A– D). At 48 and 72 h, the fluorescence signal was localised within the wall matrix (Fig. 9E, E’, F, F’) and corresponded with the course of some cellulose microfibrils (Fig. 9F, F’).

Fig. 9
figure 9

The occurrence of LM6 pectic epitope during cell wall reconstruction in F. tataricum (A-F), F. esculentum (G-L), and Ft + Fe hybrid (M-R) with the corresponding signal from calcofluor (A’-R’)

Full size image

In F. esculentum and hybrid protoplasts, the LM6 epitope in the wall matrix was exhibited earlier. Beginning from the 4 h time point, the LM6 epitope appeared at the protoplast surface even before apparent cellulose incorporation (Fig. 9H and H’), and during further stages, the fluorescence signal corresponded with cellulose microfibril deposition (Fig. 9I– L).

In hybrid protoplasts, the cytoplasmic occurrence was noticed only at 0 h, with a significant decrease in the 4 h time point (Fig. 9; Table S4). From the subsequent time points, the occurrence of LM6 was observed in the wall matrix, where the fluorescence signal corresponded with the deposited cellulose component (Fig. 9O– R).

In summary, the occurrence of LM6 epitope was widespread in protoplasts that regenerated cell walls, except 4 h time point where significant differences were noticed. Namely, only an average number of F. tataricum contained this compound, most of F. esculentum and a few hybrid protoplasts (Table S4). Also, at 24 h, in F. tataricum, a fluorescence signal was detected in an average number of protoplasts compared to the predominant distribution in F. esculentum and hybrid protoplasts.

Discussion

Isolated protoplasts can regenerate new walls developed enough in structure and composition to permit cell division and differentiation into specific cellular types, tissues, and even whole plants. The protocols for successful protoplast isolation and plant regeneration from protoplast-derived cells have been established for F. tataricum and F. esculentum (Zaranek et al. 2023a, b).

Structural proteins

The involvement of different AGP epitopes in wall formation during protoplast culture is well documented. In sugar beet, JIM8, JIM13 and LM2 epitopes were detected abundantly in incipient of cell walls and labelling increased as the rebuilding of the walls progressed (Butowt et al. 1999; Majewska-Sawka and Münster 2003). Similar observations with JIM8 and JIM13 epitopes were reported for Daucus protoplast cultures (Godel-Jędrychowska et al. 2019; Mock et al. 1990). Another study that involved guard cell- and mesophyll-derived protoplasts from sugar beet and tobacco implied some species-specific differences (Wiśniewska and Majewska-Sawka 2008). Namely, LM2, MAC207, and JIM13 were detected abundantly in sugar beet protoplasts in contrast to JIM4, JIM8, and JIM15 epitopes, while in tobacco, only JIM13 epitope from all epitopes mentioned was present in abundance (Wiśniewska and Majewska-Sawka 2008). Our results are consistent with these findings– both JIM13 and JIM16 epitopes are present during all stages of wall rebuilding in Fagopyrum protoplasts, regardless of protoplast source. AGPs have the properties of binding to β-glycans; thus, these plasma-membrane-associated glycoproteins may act as cell surface attachment sites for cell wall matrix polysaccharides (Pennell et al. 1989). Our findings are also in accordance with the reported extracellular localization of AGPs around the regenerating protoplast (Mock et al. 1990), as we also observed fluorescence signals at the surface and around protoplasts. Fagopyrum protoplasts are another system where the significant involvement of AGPs in wall formation is postulated (Cassab 1998; Sadava and Chrispeels 1973; Ye and Varner 1991). Unlike AGPs, which are easy to extract from the wall and are considered “mobile”, extensins are glycoproteins incredibly resistant to extraction. After being secreted into the wall, they are immediately immobilised by covalent binding with other extensin molecules (Cooper and Varner 1984; Lamport 1986) or wall polymers, presumably pectins (RG-I, Qi et al. 1995). Recent studies, however, indicate that extensins may also be correlated with the increase in cell size (Moore et al. 2014) or the initiation of growth (Cannon et al. 2008).

It was shown that in Nicotiana sylvestris protoplasts, the genes encoding the extensins are successively transcribed in a few hours after the isolation procedure (Parmentier et al. 1995). In carrot’s protoplast-derived cells, a JIM12 extensin epitope was detected only between 10 and 20 days of culture in the already differentiated cells (Godel-Jędrychowska et al. 2019). Our study shows the occurrence of another extensin epitope, JIM20, in intracellular compartments at the early stages (0 h, 4 h) of wall regeneration and the wall-bound localization during the further time points (12–72 h). Despite differences in the number of protoplasts exhibiting the fluorescence signal, a 12 h time frame is common for all genotypes and points to the most abundant JIM20 occurrence. Thus, it can be stated that JIM20 is involved in the wall regeneration process in Fagopyrum species.

Wall polysaccharides

Xyloglucan is the dominant hemicellulose of the primary cell walls of dicotyledonous plants (Brett and Waldron 1996). The network created between cellulose microfibrils and xyloglucan stabilises the spatial structure of the entire cell wall (Park and Cosgrove 2012; Peña et al. 2004; Ryden et al. 2003). However, it was suggested that the formation of a cellulose network during wall rebuilding is independent of xyloglucan as this compound was not essential during the early stages of cellulose assembly and, during the early stages of pea protoplast cultures, it was shown that the binding between cellulose and xyloglucan was not as strong as in the intact pea cell walls (Hayashi et al. 1986). Nevertheless, our results indicate the co-localization of xyloglucan LM25 epitope and cellulose microfibrils. In Fagopyrum protoplasts, xyloglucan synthesis and incorporation into the wall matrix accompany the cellulose network formation. Methyl-esterified homogalacturonan (HG) recognized by LM20 was present in intracellular compartments, which indicates the ongoing synthesis. In F. tataricum this cytoplasmic distribution was detected even up to 48 h, in contrast to F. esculentum and hybrid protoplasts, in which the wall-bound occurrence was observed earlier. However, our results showed a relatively moderate distribution within the wall matrix in the examined time frames. Other results indicate that the amount of esterified HG increases with time during the wall regeneration. For example, in carrot protoplast culture, methyl-esterified pectins were abundantly present after 4 days of culture, and the amount increased along with the culture age (Godel-Jędrychowska et al. 2019). Similarly, in flax hypocotyl-derived protoplasts at 6th day of culture, the amount of methyl-esterified pectins increased compared to analysis from 3rd day (David et al. 1995). Fresh protoplasts of sugar beet showed trace content of methyl-esterified pectin, in contrast to mesophyll protoplast-derived cells (Majewska-Sawka and Münster 2003; Wiśniewska and Majewska-Sawka 2008).

The side chains of RG-I, arabinan and galactan influence the mechanical properties and porosity of walls (Verhertbruggen et al. 2009; Willats et al. 2001) as well as cell adhesion (Iwai et al. 2001; Leboeuf et al. 2004; Orfila et al. 2001; Peña and Carpita 2004). Arabinan and galactan often show different distributions within cells or tissues (Bush et al. 2001; Orfila and Knox 2000) and may have different functions depending on the plant species or cell type (McCartney et al. 2000; Orfila and Knox 2000; Willats et al. 1999). (McCartney et al. 2003; Willats et al. 1999) The occurrence of the LM5 epitope in the walls is correlated with their strengthening (McCartney and Knox 2002; McCartney et al. 2000), in contrast to arabinan (LM6 epitope), the presence of which ensures the elasticity of cell walls (Jones et al. 2003, 2005; Moore et al. 2008, 2013).

Surprisingly, both epitopes were present during wall regeneration in Fagopyrum protoplasts. However, it is noteworthy that in hybrid protoplasts, the LM6 occurrence was significantly reduced at a 4 h time point compared to F. tataricum and F. esculentum. Similarly to the HG, LM5 and LM6 were detected in intracellular compartments and subsequently within the wall, which suggests increased synthesis followed by their incorporation into the wall matrix. So far, LM5 and LM6 have been detected in guard and mesophyll cell protoplasts of tobacco, in contrast to sugar beet protoplasts, in which LM5 was detected in trace amounts, and LM6 occurred only in protoplast-derived callus (Majewska-Sawka and Münster 2003; Wiśniewska and Majewska-Sawka 2008). Thus, the occurrence of galactan and arabinan during wall regeneration seems to be species-specific. Since regenerating walls have to be competent to contain the expansion of the protoplast, withstand the osmotic pressure and, consequently, allow for cell division (Burgess and Linstead 1976), they have to be elastic and strengthened. It could explain the concurrent presence of both LM5 and LM6 epitopes during wall re-establishment in Fagopyrum protoplasts.

Final remarks

We want to highlight that the regenerated walls of protoplasts should not be considered equivalent to the walls of cells from which protoplasts were isolated. The sugar composition of cell walls from protoplasts and walls of their corresponding parent tissue differs, and the type of regeneration media (solid or liquid) affects polysaccharide content (David et al. 1995; Takeuchi and Komamine 1978, 1982). Protoplast walls could be compared, in terms of composition, to a very young cell wall formed during cytokinesis (Franz and Blaschek 1985). Indeed, xyloglucan, pectins, AGPs, and extensins participate in cell plate formation (Shibaya and Sugawara 2009; Sinclair et al. 2022). Additionally, during the early stages of wall formation, wall compounds disintegrate from the surface of the cells and are lost into the medium (Franz and Blaschek 1985) which explains the extracellular occurrence of some analysed epitopes in the current study.

Conclusion

In the presented study, we focused on the process of wall rebuilding for Fagopyrum spp., a new yet subsequent system introduced in this basic research area. Surprisingly, the early stages of wall formation in hybrid protoplasts were not impaired compared to those observed in parental protoplasts of F. tataricum or F. esculentum. These results can contribute to further research on somatic hybrids, as understanding the development patterns may impact the successful modulation of yield productivity and yield stability for future buckwheat breeding programmes.