Research ArticlePLANT SCIENCES

An epidermis-driven mechanism positions and scales stem cell niches in plants

See allHide authors and affiliations

Science Advances  29 Jan 2016:
Vol. 2, no. 1, e1500989
DOI: 10.1126/sciadv.1500989
  • Fig. 1 The epidermis-driven model reproduces the wild-type gene expression patterns.

    (A) Expression domains of cytokinin receptors (AHK2, AHK3, AHK4), and cytokinin-activating enzymes (LOG4, LOG7) in green (fluorescent protein markers, Supplementary Materials) together with plasma membranes in red (FM4-64 dye). Scale bars, 20 μm. (B) Representation of the model. Solid lines indicate interactions supported by molecular evidence, whereas dashed lines show interactions supported by indirect evidence. (C) Optimized domains of WUS, CLV3, and KAN1 in two-dimensional (2D) (left) and 3D (right) abstract representations of the tissue. Gene expression intensity is color-coded. (D) Optimized domains of CLV3 and WUS in a segmented tissue obtained from confocal microscopy. Gene expression domains from the simulation are shown in green, and plasma membranes are shown in red (FM4-64 dye). The diffusive signal concentrations are displayed with the color map of (C) (minimum, blue, to maximum, red). Model equations are defined in the Supplementary Materials, and model parameter values are given in table S1.

  • Fig. 2 WUS and CLV3 expression domains respond to the geometry of the meristem.

    (A) From left to right: In situ hybridizations of WUS and CLV3 in wild-type meristems and their simulated counterparts, clavata phenotype simulated in a wild-type 2D meristem and in a geometry mimicking the enlarged tissue observed in mutant plants, and in situ hybridizations of WUS and CLV3 in clv1 loss of function meristems. Scale bars, 50 μm. (B) Expression domains of WUS and CLV3 for various meristem sizes. From top to bottom: WUS expression experimentally observed (lateral section and top projection) and simulated, and CLV3 expression experimentally observed and simulated. LD indicates plants grown under long-day conditions; SD-LD indicates growth in short days followed by long days. Scale bars, 20 μm. (B) shows representative examples of meristems presented in (C). (C) Correlation between meristem sizes and gene expression domain sizes: left, experimental data (from top projections); right, simulations (domain size is the number of expressing cells). Values obtained from linear regression are given for the experimental data (slope, interception, and P value; see the Supplementary Materials). Data for individual conditions are provided in figs. S5 and S6.

  • Fig. 3 WUS expression domain is elongated for flatter meristems.

    (A) Meristem and expression domain shapes. The x axis is the value of the a parameter of the parabola outlying the meristem; the higher the value, the flatter the meristem (see the Supplementary Materials). The y axis is the ratio between the longest vertical axis of pWUS >> GFP expression domain and its longest horizontal axis; the higher the value, the more horizontally elongated the domain is. (B) Recorded meristem shapes plotted as parabolas. The horizontal span of a plain parabola corresponds to the width of the treated image. (C) Modeled meristem and expression domain shapes. The x axis shows the different tissue templates displayed in fig. S12. For the parameter sets optimized for the 2D template, the y axis shows WUS expression domain elongation (longest vertical axis over longest horizontal axis). (D) Examples of processed images. The automatically defined parabolas and expression domains axes are plotted in blue. Scale bar, 20 μm.

  • Fig. 4 A localized tissue deformation is sufficient to explain the formation of new stem cell domains in flower primordia.

    (A) Model equilibria in a set of different tissues mimicking the growth of a flower primordium. As the deformation expands, new WUS and CLV3 domains appear in the expanding organ. (B) Top-view projections of WUS and CLV3 expression domains in a SAM surrounded by flower primordia at different stages. Older (larger) primordia express the constructs whereas the younger ones do not. Scale bars, 50 μm.

  • Fig. 5 Scaling mechanism: Two types of epidermal signals form an incoherent feed-forward motif.

    (A) Ratios between diffusion and degradation rates for the L1 signals obtained after multiple optimizations of the 2D template (left). Dashed lines mark the shape of the resulting gradients (right). The bottom two examples represent gradients from short-range signals, whereas the top one represents a gradient from a long-range signal. (B) The incoherent feed-forward network motif activating WUS. (C) Illustration of the mechanism within a simplified model [shown in (B)] simulated on a 1D tissue of varying size (Supplementary Materials). Lines show signal concentrations (red, cytokinin; green, AHK-), and dashed lines are activity thresholds (kcyt, kahk); the point where the concentration of a signal meets its threshold is represented by a large dot. WUS expression is shown in blue. WUS is activated to the right of the red dot and repressed to the right of the green dot. When the tissue size varies, the distance between the red and the green dots changes, causing the WUS expression domain to scale with the tissue (left). As the tissue expands, a de novo WUS expression domain is formed when the red dot passes the green dot (right). Note that the regulations are not step functions and some “leakage” of expression appears outside the activity thresholds.

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/2/1/e1500989/DC1

    Fig. S1. Templates.

    Fig. S2. Optimization strategy.

    Fig. S3. The model is able to represent a large collection of perturbations.

    Fig. S4. pCLV3::WUS.

    Fig. S5. WUS domain size variation.

    Fig. S6. CLV3 domain size variation.

    Fig. S7. pCLV3 >> GFP meristem shapes and expression domains.

    Fig. S8. clasp1 pWUS >> GFP meristems, grown in long days.

    Fig. S9. clasp1 pWUS >> GFP meristems, grown in long days followed by short days.

    Fig. S10. Col.0 pWUS >> GFP meristems, grown in long days.

    Fig. S11. Col.0 pWUS >> GFP meristems, grown in long days followed by short days.

    Fig. S12. WS-4 pWUS >> GFP meristems, grown in long days.

    Fig. S13. WS-4 pWUS >> GFP meristems, grown in long days followed by short days.

    Fig. S14. clasp1 pCLV3 >> GFP meristems, grown in long days.

    Fig. S15. clasp1 pCLV3 >> GFP meristems, grown in long days followed by short days.

    Fig. S16. Col.0 pCLV3 >> GFP meristems, grown in long days.

    Fig. S17. Col.0 pCLV3 >> GFP meristems, grown in long days followed by short days.

    Fig. S18. WS-4 pCLV3 >> GFP meristems, grown in long days.

    Fig. S19. WS-4 pCLV3 >> GFP meristems, grown in long days followed by short days.

    Fig. S20. Example of expression domain variations upon tissue shape changes.

    Fig. S21. Meristem size measure versus shape measure.

    Fig. S22. Long-range and short-range signals.

    Fig. S23. WUS sensitivity analysis.

    Fig. S24. CLV3 sensitivity analysis.

    Table S1. Example parameter sets.

    Movie S1. Primordium growth.

  • Supplementary Materials

    This PDF file includes:

    • Fig. S1. Templates.
    • Fig. S2. Optimization strategy.
    • Fig. S3. The model is able to represent a large collection of perturbations.
    • Fig. S4. pCLV3::WUS.
    • Fig. S5. WUS domain size variation.
    • Fig. S6. CLV3 domain size variation.
    • Fig. S7. pCLV3 >> GFP meristem shapes and expression domains.
    • Fig. S8. clasp1 pWUS >> GFP meristems, grown in long days.
    • Fig. S9. clasp1 pWUS >> GFP meristems, grown in long days followed by short days.
    • Fig. S10. Col.0 pWUS >> GFP meristems, grown in long days.
    • Fig. S11. Col.0 pWUS >> GFP meristems, grown in long days followed by short days.
    • Fig. S12. WS-4 pWUS >> GFP meristems, grown in long days.
    • Fig. S13. WS-4 pWUS >> GFP meristems, grown in long days followed by short days.
    • Fig. S14. clasp1 pCLV3 >> GFP meristems, grown in long days.
    • Fig. S15. clasp1 pCLV3 >> GFP meristems, grown in long days followed by short days.
    • Fig. S16. Col.0 pCLV3 >> GFP meristems, grown in long days.
    • Fig. S17. Col.0 pCLV3 >> GFP meristems, grown in long days followed by short days.
    • Fig. S18. WS-4 pCLV3 >> GFP meristems, grown in long days.
    • Fig. S19. WS-4 pCLV3 >> GFP meristems, grown in long days followed by short days.
    • Fig. S20. Example of expression domain variations upon tissue shape changes.
    • Fig. S21. Meristem size measure versus shape measure.
    • Fig. S22. Long-range and short-range signals.
    • Fig. S23. WUS sensitivity analysis.
    • Fig. S24. CLV3 sensitivity analysis.
    • Table S1. Example parameter sets.
    • Legend for movie S1

    Download PDF

    Other Supplementary Material for this manuscript includes the following:

    • Movie S1 (.mp4 format). Primordium growth.

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article