Shoot apical meristem (SAM) growth is critical for stem architecture. Plant hormones gibberellins (GAs) play key roles in coordinating plant growth, but their role in the SAM remains poorly understood. Here, we developed a ratiometric biosensor of GA signaling by engineering the DELLA protein to suppress its essential regulatory function in the GA transcriptional response while preserving its degradation upon GA recognition. We demonstrate that this degradation-based biosensor accurately records changes in GA levels and cellular sensing during development. We used this biosensor to map GA signaling activity in the SAM. We show that high GA signals are present predominantly in cells located between organ primordia, which are precursors to internode cells. Using gain- and loss-of-function approaches, we further demonstrate that GA regulates the orientation of the cell division plane, establishing the canonical cellular organization of internodes, thereby promoting internode specification in the SAM.
The shoot apical meristem (SAM), located at the shoot apex, contains a niche of stem cells whose activity generates lateral organs and stem nodes in a modular and iterative manner throughout the life of the plant. Each of these repeating units, or plant nodes, includes internodes and lateral organs at the nodes, and axillary meristems in the leaf axils1. The growth and organization of plant nodes changes during development. In Arabidopsis, internodal growth is suppressed during the vegetative stage, and axillary meristems remain dormant in the axils of rosette leaves. During the transition to the floral phase, the SAM becomes the inflorescence meristem, generating elongated internodes and axillary buds, branchlets in the axils of cauline leaves, and later, leafless flowers2. Although we have made significant progress in understanding the mechanisms that control the initiation of leaves, flowers, and branches, relatively little is known about how internodes arise.
Understanding the spatiotemporal distribution of GAs will help to better understand the functions of these hormones in different tissues and at different developmental stages. Visualization of the degradation of RGA-GFP fusion expressed under the action of its own promoter provides important information on the regulation of total GA levels in roots15,16. However, RGA expression varies across tissues17 and is regulated by GA18. Thus, differential expression of the RGA promoter may result in the fluorescence pattern observed with RGA-GFP and thus this method is not quantitative. More recently, bioactive fluorescein (Fl)-labeled GA19,20 revealed the accumulation of GA in the root endocortex and the regulation of its cellular levels by GA transport. Recently, the GA FRET sensor nlsGPS1 showed that GA levels correlate with cell elongation in roots, filaments, and dark-grown hypocotyls21. However, as we have seen, GA concentration is not the only parameter controlling GA signaling activity, as it depends on complex sensing processes. Here, building on our understanding of the DELLA and GA signaling pathways, we report the development and characterization of a degradation-based ratiometric biosensor for GA signaling. To develop this quantitative biosensor, we used a mutant GA-sensitive RGA that was fused to a fluorescent protein and ubiquitously expressed in tissues, as well as a GA-insensitive fluorescent protein. We show that the mutant RGA protein fusions do not interfere with endogenous GA signaling when ubiquitously expressed, and that this biosensor can quantify signaling activity resulting from both GA input and GA signal processing by the sensing apparatus with high spatiotemporal resolution. We used this biosensor to map the spatiotemporal distribution of GA signaling activity and quantify how GA regulates cellular behavior in the SAM epidermis. We demonstrate that GA regulates the orientation of the division plane of SAM cells located between organ primordia, thereby defining the canonical cellular organization of the internode.
Finally, we asked whether qmRGA could report changes in endogenous GA levels using growing hypocotyls. We previously showed that nitrate stimulates growth by increasing GA synthesis and, in turn, DELLA34 degradation. Accordingly, we observed that hypocotyl length in pUBQ10::qmRGA seedlings grown under abundant nitrate supply (10 mM NO3−) was significantly longer than that in seedlings grown under nitrate-deficient conditions (Supplementary Fig. 6a). Consistent with the growth response, GA signals were higher in hypocotyls of seedlings grown under 10 mM NO3− conditions than in seedlings grown in the absence of nitrate (Supplementary Fig. 6b, c). Thus, qmRGA also enables monitoring of changes in GA signaling induced by endogenous changes in GA concentration.
To understand whether the GA signaling activity detected by qmRGA depends on GA concentration and GA perception, as expected based on the sensor design, we analyzed the expression of the three GID1 receptors in vegetative and reproductive tissues. In seedlings, the GID1-GUS reporter line showed that GID1a and c were highly expressed in cotyledons (Fig. 3a–c). In addition, all three receptors were expressed in leaves, lateral root primordia, root tips (except for the root cap of GID1b), and the vascular system (Fig. 3a–c). In the inflorescence SAM, we detected GUS signals only for GID1b and 1c (Supplementary Fig. 7a–c). In situ hybridization confirmed these expression patterns and further demonstrated that GID1c was uniformly expressed at low levels in the SAM, whereas GID1b showed higher expression at the periphery of the SAM (Supplementary Fig. 7d–l). The pGID1b::2xmTQ2-GID1b translational fusion also revealed a graded range of GID1b expression, from low or no expression in the center of the SAM to high expression at the organ borders (Supplementary Fig. 7m). Thus, GID1 receptors are not uniformly distributed across and within tissues. In subsequent experiments, we also observed that overexpression of GID1 (pUBQ10::GID1a-mCherry) increased the sensitivity of qmRGA in hypocotyls to external GA application (Fig. 3d, e). In contrast, fluorescence measured by qd17mRGA in the hypocotyl was insensitive to GA3 treatment (Fig. 3f, g). For both assays, seedlings were treated with high concentrations of GA (100 μM GA3) to assess the rapid behavior of the sensor, where the ability to bind to the GID1 receptor was enhanced or lost. Together, these results confirm that the qmRGA biosensor serves a combined function as a GA and GA sensor, and suggest that differential expression of the GID1 receptor can significantly modulate the emissivity of the sensor.
To date, the distribution of GA signals in the SAM remains unclear. Therefore, we used qmRGA-expressing plants and the pCLV3::mCherry-NLS stem cell reporter35 to calculate high-resolution quantitative maps of GA signaling activity, focusing on the L1 layer (epidermis; Fig. 4a, b, see Methods and Supplementary Methods), since L1 plays a key role in controlling SAM growth36. Here, pCLV3::mCherry-NLS expression provided a fixed geometric reference point for analyzing the spatiotemporal distribution of GA signaling activity37. Although GA is considered essential for lateral organ development4, we observed that GA signals were low in the floral primordium (P) starting from the P3 stage (Fig. 4a, b), whereas young P1 and P2 primordiums had moderate activity similar to that in the central region (Fig. 4a, b). Higher GA signaling activity was detected at the organ primordium boundaries, starting at P1/P2 (at the sides of the boundary) and peaking at P4, as well as in all cells of the peripheral region located between the primordia (Fig. 4a, b and Supplementary Fig. 8a, b). This higher GA signaling activity was observed not only in the epidermis but also in the L2 and upper L3 layers (Supplementary Fig. 8b). The pattern of GA signals detected in the SAM using qmRGA also remained unchanged over time (Supplementary Fig. 8c–f, k). Although the qd17mRGA construct was systematically downregulated in the SAM of T3 plants from five independent lines that we characterized in detail, we were able to analyze the fluorescence patterns obtained with the pRPS5a::VENUS-2A-TagBFP construct (Supplementary Fig. 8g–j, l). In this control line, only minor changes in the fluorescence ratio were detected in the SAM, but in the SAM center we observed a clear and unexpected decrease in VENUS associated with TagBFP. This confirms that the signaling pattern observed by qmRGA reflects GA-dependent degradation of mRGA-VENUS, but also demonstrates that qmRGA may overestimate GA signaling activity in the meristem center. In summary, our results reveal a GA signaling pattern that primarily reflects the distribution of primordia. This distribution of the inter-primordial region (IPR) is due to the gradual establishment of high GA signaling activity between the developing primordium and the central region, while at the same time GA signaling activity in the primordium decreases (Fig. 4c, d).
The distribution of GID1b and GID1c receptors (see above) suggests that differential expression of GA receptors helps shape the pattern of GA signaling activity in the SAM. We wondered whether differential accumulation of GA might be involved. To investigate this possibility, we used the nlsGPS1 GA FRET sensor21. Increased activation frequency was detected in the SAM of nlsGPS1 treated with 10 μM GA4+7 for 100 min (Supplementary Fig. 9a–e), indicating that nlsGPS1 responds to changes in GA concentration in the SAM, as it does in roots21. Spatial distribution of nlsGPS1 activation frequency revealed relatively low GA levels in the outer layers of the SAM, but showed that they were elevated in the center and at the borders of the SAM (Fig. 4e and Supplementary Fig. 9a,c). This suggests that GA is also distributed in the SAM with a spatial pattern comparable to that revealed by qmRGA. As a complementary approach, we also treated the SAM with fluorescent GA (GA3-, GA4-, GA7-Fl) or Fl alone as a negative control. The Fl signal was distributed throughout the SAM, including the central region and primordium, albeit at a lower intensity (Fig. 4j and Supplementary Fig. 10d). In contrast, all three GA-Fl accumulated specifically within the primordium borders and to varying degrees in the rest of the IPR, with GA7-Fl accumulating in the largest domain in the IPR (Fig. 4k and Supplementary Fig. 10a,b). Quantification of fluorescence intensity revealed that the IPR to non-IPR intensity ratio was higher in GA-Fl-treated SAM compared to Fl-treated SAM (Fig. 4l and Supplementary Fig. 10c). Together, these results suggest that GA is present at higher concentrations in IPR cells that are located closest to the organ border. This suggests that the pattern of SAM GA signaling activity results from both differential expression of GA receptors and differential accumulation of GA in IPR cells near organ borders. Thus, our analysis revealed an unexpected spatiotemporal pattern of GA signaling, with lower activity in the center and primordium of the SAM and higher activity in the IPR in the peripheral region.
To understand the role of differential GA signaling activity in the SAM, we analyzed the correlation between GA signaling activity, cell expansion, and cell division using real-time time-lapse imaging of the SAM qmRGA pCLV3::mCherry-NLS. Given the role of GA in growth regulation, a positive correlation with cell expansion parameters was expected. Therefore, we first compared GA signaling activity maps with maps of cell surface growth rate (as a proxy for the strength of cell expansion for a given cell and for daughter cells at division) and with maps of growth anisotropy, which measures the directionality of cell expansion (also used here for a given cell and for daughter cells at division; Fig. 5a,b, see Methods and Supplementary Methods). Our maps of SAM cell surface growth rate are consistent with previous observations38,39, with minimal growth rates at the border and maximal growth rates in developing flowers (Fig. 5a). Principal component analysis (PCA) showed that GA signaling activity was negatively correlated with cell surface growth intensity (Figure 5c). We also showed that the main axes of variation, including GA signaling input and growth intensity, were orthogonal to the direction determined by high CLV3 expression, confirming the exclusion of cells from the SAM center in the remaining analyses. Spearman correlation analysis confirmed the PCA results (Figure 5d), indicating that higher GA signals in the IPR did not result in higher cell expansion. However, correlation analysis revealed a slight positive correlation between GA signaling activity and growth anisotropy (Figure 5c, d), suggesting that higher GA signaling in the IPR influences the direction of cell growth and possibly the position of the cell division plane.
a, b Heat maps of mean surface growth (a) and growth anisotropy (b) in SAM averaged over seven independent plants (used as proxies for the strength and direction of cell expansion, respectively). c PCA analysis included the following variables: GA signal, surface growth intensity, surface growth anisotropy, and CLV3 expression. PCA component 1 was mainly negatively correlated with surface growth intensity and positively correlated with GA signal. PCA component 2 was mainly positively correlated with surface growth anisotropy and negatively correlated with CLV3 expression. Percentages represent the variation explained by each component. d Spearman correlation analysis between GA signal, surface growth intensity, and surface growth anisotropy at the tissue scale excluding CZ. The number on the right is the Spearman rho value between two variables. Asterisks indicate cases where the correlation/negative correlation is highly significant. e 3D visualization of Col-0 SAM L1 cells by confocal microscopy. New cell walls formed in the SAM (but not the primordium) at 10 h are colored according to their angle values. The color bar is shown in the lower right corner. The inset shows the corresponding 3D image at 0 h. The experiment was repeated twice with similar results. f Box plots display cell division rates in IPR and non-IPR Col-0 SAM (n = 10 independent plants). The center line shows the median, and the box boundaries indicate the 25th and 75th percentiles. Whiskers indicate the minimum and maximum values determined with R software. P values were obtained with Welch’s two-tailed t-test. g, h Schematic diagram showing (g) how to measure the angle of the new cell wall (magenta) with respect to the radial direction from the center of the SAM (white dotted line) (only acute angle values, i.e., 0–90°, are considered), and (h) the circumferential/lateral and radial directions within the meristem. i Frequency histograms of cell division plane orientation across the SAM (dark blue), IPR (medium blue), and non-IPR (light blue), respectively. P values were obtained by a two-tailed Kolmogorov-Smirnov test. The experiment was repeated twice with similar results. j Frequency histograms of cell division plane orientation of the IPR around P3 (light green), P4 (medium green), and P5 (dark green), respectively. P values were obtained by a two-tailed Kolmogorov-Smirnov test. The experiment was repeated twice with similar results.
Therefore, we next investigated the correlation between GA signaling and cell division activity by identifying newly formed cell walls during the assay (Fig. 5e). This approach allowed us to measure the frequency and direction of cell division. Surprisingly, we found that the frequency of cell divisions in the IPR and the rest of the SAM (non-IPR, Fig. 5f) was similar, indicating that differences in GA signaling between IPR and non-IPR cells do not significantly affect cell division. This, and the positive correlation between GA signaling and growth anisotropy, prompted us to consider whether GA signaling activity could influence the orientation of the cell division plane. We measured the orientation of the new cell wall as an acute angle relative to the radial axis connecting the meristem center and the center of the new cell wall (Fig. 5e-i) and observed a clear tendency for cells to divide at angles close to 90° relative to the radial axis, with the highest frequencies observed at 70–80° (23.28%) and 80–90° (22.62%) (Fig. 5e,i), corresponding to cell divisions in the circumferential/transverse direction (Fig. 5h). To examine the contribution of GA signaling to this cell division behavior, we analyzed cell division parameters in the IPR and non-IPR separately (Fig. 5i). We observed that the division angle distribution in IPR cells differed from that in non-IPR cells or in cells in the entire SAM, with IPR cells exhibiting a higher proportion of lateral/circular cell divisions, i.e., 70–80° and 80–90° (33.86% and 30.71%, respectively, corresponding proportions) (Fig. 5i). Thus, our observations revealed an association between high GA signaling and a cell division plane orientation close to the circumferential direction, similar to the correlation between GA signaling activity and growth anisotropy (Fig. 5c, d). To further establish the spatial conservation of this association, we measured the division plane orientation in IPR cells surrounding the primordium starting from P3, since the highest GA signaling activity was detected in this region starting from P4 (Fig. 4). The division angles of the IPR around P3 and P4 showed no statistically significant differences, although an increased frequency of lateral cell divisions was observed in the IPR around P4 (Fig. 5j). However, in the IPR cells around P5, the difference in the orientation of the cell division plane became statistically significant, with a sharp increase in the frequency of transverse cell divisions (Fig. 5j). Together, these results suggest that GA signaling can control the orientation of cell divisions in the SAM, which is consistent with previous reports40,41 that high GA signaling can induce lateral orientation of cell divisions in the IPR.
It is predicted that cells in the IPR will not be incorporated into primordia but rather into internodes2,42,43. The transverse orientation of cell divisions in the IPR may result in the typical organization of parallel longitudinal rows of epidermal cells in internodes. Our observations described above suggest that GA signaling likely plays a role in this process by regulating the direction of cell division.
Loss of function of several DELLA genes results in a constitutive GA response, and della mutants can be used to test this hypothesis44. We first analyzed the expression patterns of five DELLA genes in the SAM. Transcriptional fusion of the GUS line45 revealed that GAI, RGA, RGL1, and RGL2 (to a much lesser extent) were expressed in the SAM (Supplementary Fig. 11a–d). In situ hybridization further demonstrated that GAI mRNA accumulates specifically in primordia and developing flowers (Supplementary Fig. 11e). RGL1 and RGL3 mRNA were detected throughout the SAM canopy and in older flowers, whereas RGL2 mRNA was more abundant in the border region (Supplementary Fig. 11f–h). Confocal imaging of pRGL3::RGL3-GFP SAM confirmed the expression observed by in situ hybridization and showed that RGL3 protein accumulates in the central part of the SAM (Supplementary Fig. 11i). Using the pRGA::GFP-RGA line, we also found that RGA protein accumulates in the SAM, but its abundance decreases at the border starting from P4 (Supplementary Fig. 11j). Notably, the expression patterns of RGL3 and RGA are consistent with higher GA signaling activity in the IPR, as detected by qmRGA (Fig. 4). Moreover, these data indicate that all DELLAs are expressed in the SAM and that their expression collectively spans the entire SAM.
We next analyzed the cell division parameters in the wild-type SAM (Ler, control) and the gai-t6 rga-t2 rgl1-1 rgl2-1 rgl3-4 della quintuple (global) mutants (Fig. 6a, b). Interestingly, we observed a statistically significant shift in the distribution of cell division angle frequencies in the della global mutant SAM compared to the wild type (Fig. 6c). This change in the della global mutant was due to an increase in the frequency of 80–90° angles (34.71% vs. 24.55%) and, to a lesser extent, 70–80° angles (23.78% vs. 20.18%), i.e., corresponding to transverse cell divisions (Fig. 6c). The frequency of non-transverse divisions (0–60°) was also lower in the della global mutant (Fig. 6c). The frequency of transverse cell divisions was significantly increased in the SAM of the della global mutant (Fig. 6b). The frequency of transverse cell divisions in the IPR was also higher in the della global mutant compared to the wild type (Fig. 6d). Outside the IPR region, the wild type had a more uniform distribution of cell division angles, whereas the della global mutant preferred tangential divisions like the IPR (Fig. 6e). We also quantified the orientation of cell divisions in the SAM of ga2 oxidase (ga2ox) quintuple mutants (ga2ox1-1, ga2ox2-1, ga2ox3-1, ga2ox4-1, and ga2ox6-2), a GA-inactive mutant background in which GA accumulates. Consistent with the increase in GA levels, the SAM of the quintuple ga2ox mutant inflorescence was larger than that of Col-0 (Supplementary Fig. 12a, b), and compared to Col-0, the quintuple ga2ox SAM showed a distinctly different distribution of cell division angles, with the angle frequency increasing from 50° to 90°, i.e. again favoring tangential divisions (Supplementary Fig. 12a–c). Thus, we show that constitutive activation of GA signaling and GA accumulation induce lateral cell divisions in the IPR and the rest of the SAM.
a, b 3D visualization of the L1 layer of PI-stained Ler (a) and global della mutant (b) SAM using confocal microscopy. New cell walls formed in the SAM (but not the primordium) over a 10-h period are shown and colored according to their angle values. The inset shows the SAM at 0 h. The color bar is displayed in the lower right corner. The arrow in (b) points to an example of aligned cell files in the global della mutant. The experiment was repeated twice with similar results. ce comparison of the frequency distribution of cell division plane orientations in the whole SAM (d), IPR (e), and non-IPR (f) between Ler and global della. P values were obtained using a two-tailed Kolmogorov-Smirnov test. f, g 3D visualization of confocal images of PI-stained SAM of Col-0 (i) and pCUC2::gai-1-VENUS (j) transgenic plants. Panels (a, b) show new cell walls (but not primordia) formed in the SAM within 10 h. The experiment was repeated twice with similar results. h–j Comparison of the frequency distribution of cell division plane orientations located in the entire SAM (h), IPR (i) and non-IPR (j) between Col-0 and pCUC2::gai-1-VENUS plants. P values were obtained using a two-tailed Kolmogorov–Smirnov test.
We next tested the effect of inhibiting GA signaling specifically in the IPR. To this end, we used the cotyledon cup 2 (CUC2) promoter to drive expression of a dominant negative gai-1 protein fused to VENUS (in the pCUC2::gai-1-VENUS line). In the wild-type SAM, the CUC2 promoter drives expression of most IPRs in the SAM, including border cells, from P4 onwards, and similar specific expression was observed in pCUC2::gai-1-VENUS plants (see below). The distribution of cell division angles across the SAM or IPR of pCUC2::gai-1-VENUS plants was not significantly different from that of the wild type, although unexpectedly we found that cells without an IPR in these plants divided at a higher frequency of 80–90° (Fig. 6f–j).
It has been suggested that the direction of cell division depends on the geometry of the SAM, in particular the tensile stress generated by the tissue curvature46. We therefore asked whether the shape of the SAM was altered in the della global mutant and pCUC2::gai-1-VENUS plants. As reported previously12, the size of the della global mutant SAM was larger than that of the wild type (Supplementary Fig. 13a, b, d). In situ hybridization of CLV3 and STM RNA confirmed the meristem expansion in della mutants and further showed the lateral expansion of the stem cell niche (Supplementary Fig. 13e, f, h, i). However, the SAM curvature was similar in both genotypes (Supplementary Fig. 13k, m, n, p). We observed a similar increase in size in the gai-t6 rga-t2 rgl1-1 rgl2-1 della quadruple mutant without a change in curvature compared to the wild type (Supplementary Fig. 13c, d, g, j, l, o, p). The frequency of cell division orientation was also affected in the della quadruple mutant, but to a lesser extent than in the della monolithic mutant (Supplementary Fig. 12d–f). This dosage effect, along with the lack of an effect on curvature, suggests that residual RGL3 activity in the Della quadruple mutant limits changes in cell division orientation caused by loss of DELLA activity and that changes in lateral cell divisions occur in response to changes in GA signaling activity rather than changes in SAM geometry. As described above, the CUC2 promoter drives IPR expression in the SAM starting at P4 (Supplementary Fig. 14a, b), and in contrast, the pCUC2::gai-1-VENUS SAM had a reduced size but higher curvature (Supplementary Fig. 14c–h). This change in pCUC2::gai-1-VENUS SAM morphology may result in a different distribution of mechanical stresses compared to the wild type, in which high circumferential stresses start at a shorter distance from the SAM center47. Alternatively, the changes in pCUC2::gai-1-VENUS SAM morphology may result from changes in regional mechanical properties induced by transgene expression48. In both cases, this could partially offset the effects of changes in GA signaling by increasing the likelihood that cells will divide in the circumferential/transverse orientation, explaining our observations.
Taken together, our data confirm that higher GA signaling plays an active role in the lateral orientation of the cell division plane in the IPR. They also show that meristem curvature also influences the orientation of the cell division plane in the IPR.
The transverse orientation of the division plane in the IPR, due to high GA signaling activity, suggests that GA pre-organizes a radial cell file in the epidermis within the SAM to define the cellular organization that will later be found in the epidermal internode. Indeed, such cell files were frequently visible in SAM images of della global mutants (Fig. 6b). Thus, to further explore the developmental function of the spatial pattern of GA signaling in the SAM, we used time-lapse imaging to analyze the spatial organization of cells in the IPR in wild-type (Ler and Col-0), della global mutants, and pCUC2::gai-1-VENUS transgenic plants.
We found that qmRGA showed that GA signaling activity in the IPR increased from P1/P2 and peaked at P4, and this pattern remained constant over time (Fig. 4a–f and Supplementary Fig. 8c–f, k). To analyze the spatial organization of cells in the IPR with increasing GA signal, we labeled Ler IPR cells above and to the sides of P4 according to their developmental fate analyzed 34 h after first observation, i.e., more than two plastid times, allowing us to follow IPR cells during primordium development from P1/P2 to P4. We used three different colors: yellow for those cells that were integrated into the primordium near P4, green for those that were in the IPR, and purple for those that participated in both processes (Fig. 7a–c). At t0 (0 h), 1–2 layers of IPR cells were visible in front of P4 (Fig. 7a). As expected, when these cells divided, they did so mainly via the transverse division plane (Figs. 7a–c). Similar results were obtained using Col-0 SAM (focusing on P3, whose border folds similarly to P4 in Ler), although in this genotype the fold formed at the floral border hid the IPR cells more quickly (Fig. 7g–i). Thus, the division pattern of IPR cells pre-organizes the cells into radial rows, as in internodes. The organization of radial rows and the localization of IPR cells between successive organs suggest that these cells are internodal progenitors.
Here, we developed a ratiometric GA signaling biosensor, qmRGA, that allows quantitative mapping of GA signaling activity resulting from combined GA and GA receptor concentrations while minimizing interference with endogenous signaling pathways, thereby providing information on GA function at the cellular level. To this end, we constructed a modified DELLA protein, mRGA, that has lost the ability to bind DELLA interaction partners but remains sensitive to GA-induced proteolysis. qmRGA responds to both exogenous and endogenous changes in GA levels, and its dynamic sensing properties enable assessment of spatiotemporal changes in GA signaling activity during development. qmRGA is also a very flexible tool as it can be adapted to different tissues by changing the promoter used for its expression (if necessary), and given the conserved nature of the GA signaling pathway and the PFYRE motif across angiosperms, it is likely to be transferable to other species22. Consistent with this, an equivalent mutation in the rice SLR1 DELLA protein (HYY497AAA) was also shown to suppress the growth repressor activity of SLR1 while only slightly reducing its GA-mediated degradation, similar to mRGA23. Notably, recent studies in Arabidopsis showed that a single amino acid mutation in the PFYRE domain (S474L) altered the transcriptional activity of RGA without affecting its ability to interact with transcription factor partners50. Although this mutation is very close to the 3 amino acid substitutions present in mRGA, our studies show that these two mutations alter distinct characteristics of DELLA. Although most transcription factor partners bind to the LHR1 and SAW domains of DELLA26,51, some conserved amino acids in the PFYRE domain may help stabilize these interactions.
Internode development is a key trait in plant architecture and yield improvement. qmRGA revealed higher GA signaling activity in IPR internode progenitor cells. By combining quantitative imaging and genetics, we showed that GA signaling patterns superimpose circular/transverse cell division planes in the SAM epidermis, shaping the cell division organization required for internode development. Several regulators of cell division plane orientation have been identified during development52,53. Our work provides a clear example of how GA signaling activity regulates this cellular parameter. DELLA can interact with prefolding protein complexes41, so GA signaling may regulate cell division plane orientation by directly influencing cortical microtubule orientation40,41,54,55. We unexpectedly showed that in SAM, the correlate of higher GA signaling activity was not cell elongation or division, but only growth anisotropy, which is consistent with a direct effect of GA on the direction of cell division in the IPR. However, we cannot exclude that this effect could also be indirect, for example mediated by GA-induced cell wall softening56. Changes in cell wall properties induce mechanical stress57,58, which can also influence the orientation of the cell division plane by affecting the orientation of cortical microtubules39,46,59. The combined effects of GA-induced mechanical stress and direct regulation of microtubule orientation by GA may be involved in generating a specific pattern of cell division orientation in the IPR to define internodes, and further studies are needed to test this idea. Similarly, previous studies have highlighted the importance of the DELLA-interacting proteins TCP14 and 15 in the control of internode formation60,61 and these factors may mediate the action of GA together with BREVIPEDICELLUS (BP) and PENNYWISE (PNY), which regulate internode development and have been shown to influence GA signaling2,62. Given that DELLAs interact with brassinosteroid, ethylene, jasmonic acid, and abscisic acid (ABA) signaling pathways63,64 and that these hormones can influence microtubule orientation65, the effects of GA on cell division orientation may also be mediated by other hormones.
Early cytological studies showed that both the inner and outer regions of the Arabidopsis SAM are required for internode development2,42. The fact that GA actively regulates cell division in the inner tissues12 supports a dual function of GA in regulating meristem and internode size in the SAM. The pattern of directional cell division is also tightly regulated in the inner SAM tissue, and this regulation is essential for stem growth52. It will be interesting to examine whether GA also plays a role in orienting the cell division plane in the inner SAM organization, thereby synchronizing the specification and development of internodes within the SAM.
Plants were grown in vitro in soil or 1x Murashige-Skoog (MS) medium (Duchefa) supplemented with 1% sucrose and 1% agar (Sigma) under standard conditions (16 h light, 22 °C), except for hypocotyl and root growth experiments in which seedlings were grown on vertical plates under constant light and 22 °C. For nitrate experiments, plants were grown on modified MS medium (bioWORLD plant medium) supplemented with adequate nitrate (0 or 10 mM KNO3), 0.5 mM NH4-succinate, 1% sucrose and 1% A-agar (Sigma) under long-day conditions.
GID1a cDNA inserted into pDONR221 was recombined with pDONR P4-P1R-pUBQ10 and pDONR P2R-P3-mCherry into pB7m34GW to generate pUBQ10::GID1a-mCherry. IDD2 DNA inserted into pDONR221 was recombined into pB7RWG266 to generate p35S:IDD2-RFP. To generate pGID1b::2xmTQ2-GID1b, a 3.9 kb fragment upstream of the GID1b coding region and a 4.7 kb fragment containing the GID1b cDNA (1.3 kb) and terminator (3.4 kb) were first amplified using the primers in Supplementary Table 3 and then inserted into pDONR P4-P1R (Thermo Fisher Scientific) and pDONR P2R-P3 (Thermo Fisher Scientific), respectively, and finally recombined with pDONR221 2xmTQ268 into the pGreen 012567 target vector using Gateway cloning. To generate pCUC2::LSSmOrange, the CUC2 promoter sequence (3229 bp upstream of ATG) followed by the coding sequence of large Stokes-shifted mOrange (LSSmOrange)69 with the N7 nuclear localization signal and the NOS transcriptional terminator were assembled into the pGreen kanamycin targeting vector using the Gateway 3-fragment recombination system (Invitrogen). The plant binary vector was introduced into Agrobacterium tumefaciens strain GV3101 and introduced into Nicotiana benthamiana leaves by Agrobacterium infiltration method and into Arabidopsis thaliana Col-0 by floral dip method, respectively. pUBQ10::qmRGA pUBQ10::GID1a-mCherry and pCLV3::mCherry-NLS qmRGA were isolated from the F3 and F1 progenies of the respective crosses, respectively.
RNA in situ hybridization was performed on approximately 1 cm long shoot tips72, which were collected and immediately fixed in FAA solution (3.7% formaldehyde, 5% acetic acid, 50% ethanol) pre-cooled to 4 °C. After 2 × 15 min vacuum treatments, the fixative was changed and samples were incubated overnight. GID1a, GID1b, GID1c, GAI, RGL1, RGL2, and RGL3 cDNAs and antisense probes to their 3′-UTRs were synthesized using the primers shown in Supplementary Table 3 as described by Rosier et al.73. Digoxigenin-labeled probes were immunodetected using digoxigenin antibodies (3000-fold dilution; Roche, catalog number: 11 093 274 910), and sections were stained with 5-bromo-4-chloro-3-indolyl phosphate (BCIP, 250-fold dilution)/nitroblue tetrazolium (NBT, 200-fold dilution) solution.
Post time: Feb-10-2025