The mature leaves of seed plants are usually flat with distinct cell types making up their dorsal (upper) and ventral (lower) tissues, which are derived from the adaxial (side closest to meristem) and abaxial (side farthest from meristem) primordium tissues. A fundamental question in plant development is how this tissue patterning is first specified. Recently, we reported evidence in support of a previously proposed hypothesis that dorsoventral leaf patterning is derived from a pre-pattern present in the shoot apical meristem (SAM), where leaves arise (Hagemann and Gleissberg, 1996; Husbands et al., 2009; Kerstetter et al., 2001; Koch and Meinhardt, 1994). We found that transcription factors that promote dorsal and ventral leaf cell types are expressed in the SAM in a pre-pattern that determines the orientation of the leaves that subsequently form (Caggiano et al., 2017). Additionally, we found that the plant hormone auxin promotes dorsal cell fate by maintaining expression of the Class III HD-ZIP transcription factor REVOLUTA (REV) and repressing KANADI1 (KAN1) expression in the adaxial cells of organ primordia (Caggiano et al., 2017). These findings however contrast with that of another study which concluded that leaf polarity is dependent on an asymmetry in auxin levels within leaf primordia, with relatively low levels of auxin in adaxial cells compared to abaxial cells being critical to maintain dorsal identity (Qi et al., 2014). This conclusion was based in part on the finding that exogenous auxin application to tomato leaf primordia resulted in the formation of radialized leaves that appeared ventralized. Also, it was found that an auxin sensor, the DII (Brunoud et al., 2012; Vernoux et al., 2011) indicates low levels of auxin in adaxial leaf tissues compared to abaxial tissues at leaf initiation (Qi et al., 2014). Hence asymmetries in auxin concentrations between the adaxial and abaxial tissues in leaf primordia, as a result of PIN1 mediated auxin transport, were proposed to help establish leaf dorsoventral cell type patterning (Qi et al., 2014), in contrast to the proposed pre-pattern mechanism (Caggiano et al., 2017). Building further on this conclusion, a more recent study proposed that low levels of auxin in adaxial primordium tissues are necessary to restrict the expression of the WOX1 and PRS genes to the middle domain, since auxin promotes their expression (Guan et al., 2017). Finally, the reported asymmetry in auxin has also been linked to asymmetries in the mechanical properties of leaf tissues and their morphogenesis (Qi et al., 2017).

In-order to investigate the proposed auxin asymmetry proposal in more detail, we decided to examine the distribution of auxin within initiating leaf primordia using the ratio-metric R2D2 reporter, which acts as a proxy for the cellular sensing of auxin (Liao et al., 2015). A high DII/mDII ratio indicates relatively low levels of auxin sensing while a high mDII/DII ratio indicates relatively high levels of auxin sensing (Liao et al., 2015). We first focused on the incipient stages when PIN1 convergence patterns are established, by looking at later forming primordia (third, fourth or fifth) in seedlings 4-5DAS (days after stratification). In general, during these early stages DII/mDII ratios were extremely low in cells coinciding with high levels of PIN1-GFP expression while on either side of these cells, ratios were higher but to a similar degree (Figure 1). Previously, by examining PIN1-CFP together with pREV::REV-2xYPet and pKAN1::KAN1-2xGFP in similarly staged primordia we found that REV expression can coincide with PIN1 at later stages when it expands peripherally but that KAN1 expression is invariably complementary to cells with high PIN1 expression (Caggiano et al., 2017). Overall our observations therefore indicate that KAN1 expression is associated with low auxin sensing cells while REV expression is associated with different levels of auxin sensing depending on the stage of development. We then examined DII/mDII ratios in the leaves after initiation. For this we imaged the first two leaves when they are approximately 4–5 cells long along their proximo-distal axis (3DAS), as well as a day later as they begin to elongate (6–9 cells along the proximo-distal axis; 4DAS) and again included the PIN1-GFP marker in our analysis to correlate cellular auxin sensing with PIN1 expression and polarity. At 3DAS, the expression patterns of REV and KAN1 are already polar within such primordia, although the FILAMENTOUS FLOWER (FIL) expression domain at this stage of development is still being refined (Caggiano et al., 2017). At this stage, PIN1 is polarized towards the distal tip of leaf primordia but has reversed polarity away from the primordia, towards the meristem, in cells adjacent to primordia on the adaxial side. According to the ratio-metric auxin sensor R2D2, DII/mDII ratios varied from moderate to high in adaxial cells of the primordia but also varied similarly in abaxial regions - indicating no obvious asymmetry between the two tissue types (Figure 2A to C and M-N; Figure 2—figure supplements 1 and 2). A low DII/mDII ratio was only found consistently in more distal regions towards the tip of the primordia and in the provasculature, matching the overall pattern of signal from PIN1-GFP. By carefully monitoring DII/mDII ratios from the distal to proximal end of the initiating leaves, the variability in DII/mDII ratios within adaxial and abaxial tissues was clearly apparent (Figure 2D–L and O–S, Figure 2—figure supplements 1 and 2). This same overall pattern of signal was found in 28 out of 28 leaves that were examined each for 3DAS and 4DAS (14 seedlings each). Lastly, by inverting the ratio and visualizing the mDII/DII (high ratio intensity corresponding to high cellular auxin sensing), we could further confirm that high levels of auxin sensing were only found in the more distal regions towards the tip of the primordia and within the provasculature, where PIN1-GFP was expressed (Figure 2—figure supplement 3).

Figure 1

Download asset Open asset

Figure 2 with 3 supplements see all

Download asset Open asset

Although the detection of variable R2D2 ratios in adaxial and abaxial cells already argues against a critical role for the distribution of auxin in patterning leaf dorsoventrality, we decided to further investigate the spatial patterns of R2D2 quantitatively. Using PIN1-GFP marked vasculature as a dividing line to demarcate adaxial and abaxial tissues, we manually cropped adaxial and abaxial confocal volumes from our confocal image stacks of 4DAS old leaves expressing R2D2 using ImageJ (Fiji; https://fiji.sc) (Figure 3A and B). We then processed these data using a pipeline that included deconvolution of the image stacks, segmentation of nuclear volumes and calculations of R2D2 ratios averaged over the segmented nuclear volumes (Figure 3C–E) (Materials and methods). The aggregated ratios from 20 leaves (10 seedlings) illustrated as separate violin plots for adaxial and abaxial leaf tissues reveals high variability between individual nuclei throughout the leaves and very similar ratio distributions between adaxial and abaxial nuclei (Figure 3F and G). According to the data, abaxial tissues have a slightly larger proportion of cells with high DII/mDII ratios (low auxin sensing) with a mean of 0.29; standard deviation, SD (0.24); standard error of mean, SEM (0.01) compared to adaxial tissues, which have a mean ratio of 0.24, SD (0.20), SEM (0.004) (Figure 3G). Similar distributions are found for individual leaves (Figure 3F), although there is leaf to leaf variability, which is also apparent in the spatial distribution of ratios within and between leaves (Figure 3—figure supplement 1). Given that the data are not normally distributed we analysed the similarity of the distributions using a plot of the ranked ratios from aggregated adaxial and abaxial nuclei. The Area Under the Curve (AUC) measure of similarity was found to be 0.556, which again indicates that the distributions are overall very similar (0.5 would indicate they are identical) (Figure 3H). All together our results demonstrate that the relative levels of auxin sensing in adaxial and abaxial tissues in young leaf primordia, are overall, very similar. The small difference we identify is that a larger proportion of cells in abaxial tissues display higher DII to mDII ratios compared to adaxial tissues, indicating slightly higher levels of cellular auxin sensing in adaxial tissues, which is the opposite conclusion to that reported earlier (Guan et al., 2017; Qi et al., 2014). However, we again would like to emphasize that overall, the ratio distributions are very similar.

Figure 3 with 1 supplement see all

Download asset Open asset

As our results using the R2D2 auxin sensor indicate a different cellular auxin sensing pattern compared to that reported previously using the DII marker (Qi et al., 2014), we decided to re-examined the pattern of DII auxin sensor expression at the same early developmental stages. In contrast to the R2D2 pattern, the DII pattern showed an asymmetry of expression in leaf primordia at 3 DAS, indicating relatively low auxin levels in adaxial primordium cells, as found previously. DII signal appeared strongest in the adaxial epidermis but was also stronger in the adaxial sub-epidermal cell layer compared to abaxial epidermal and sub-epidermal cell layers (Figure 4- A, C, E and Figure 4—figure supplement 1) (n = 18/18 leaves, nine seedlings). One day later, DII signal started to show increased relative expression in the abaxial epidermal and sub-epidermal layers compared to earlier stages (n = 14/14 leaves, seven seedlings) (Figure 4G). At least for very young primordia, aged 3DAS, our results are similar to those obtained previously and consistent with the proposal that there are low levels of auxin in the adaxial regions of leaf primordia, in contrast to our results using the R2D2 sensor. Given this discrepancy, we next examined the expression of the mDII sensor which is driven by the same 35S promoter as the DII sensor but is not auxin sensitive. Surprisingly we found that, like the DII results, expression of the mDII marker was also higher in the adaxial cells of leaf primordia at 3DAS (n = 16/16 leaves, eight seedlings) (Figure 4-B,D,F and Figure 4—figure supplement 2). The pattern appeared almost identical to the pattern found using the DII marker except that the mDII marker also exhibited high levels of expression in the shoot meristem whereas the DII sensor did not (compare Figure 4-A,B; Figure 4—figure supplement 1A–G and Figure 4—figure supplement 2A–F). The similarity of expression between DII and mDII was also apparent at 4DAS when both markers started to show expression on the abaxial side as well with no obvious asymmetry in their expression (n = 20/20 leaves, 10 seedlings) (Figure 4H-Figure 4—figure supplement 3). To verify the auxin sensitivity of the sensors used we imaged seedlings before and after treatment with 5 mM NAA and found a strong decrease in DII expression compared to mDII and an increase in the mDII-tdTom/DII-V ratio for the R2D2 sensor, consistent with an increase in auxin levels (Figure 4—figure supplement 4).

Figure 4 with 4 supplements see all

Download asset Open asset

The finding that both the mDII and DII markers, when driven by the 35S promoter, are expressed in an asymmetric pattern during leaf initiation suggests that one reason our conclusions may differ from those reported earlier (Qi et al., 2014) is that previously, the DII marker was used to assess relative auxin levels without direct comparison to the mDII marker. We note that although a single section showing expression of the control mDII sensor in older leaves was cited by Guan et al., 2017 (Wang et al., 2014), this information was not adequate for properly assessing similarities and differences, emphasizing the importance of using ratio-metric reporters and quantitative analyses. More recently, these authors have posted a pre-print reporting data for both R2D2 and mDII markers in leaf primordia and concluding that the 35S promoter driven mDII marker is not asymmetric at early leaf stages and that R2D2 ratios again indicate lower auxin levels in adaxial leaf tissues (Guan et al., 2018). However for the R2D2 data, the analysis is again based on arbitrarily chosen sections taken from confocal volumes. Regarding the 35S mDII data, the leaves shown have developed to a later stage compared to the leaves we analyze at both 3DAS and 4DAS (4–6 cells long in this study vs 8–9 cells at 2DAS and 6–9 cells in this study vs 9–11 at 3DAS). As our time-lapse imaging shows, the pattern changes rapidly over 12 hr (Figure 4—figure supplement 3) and so a staging discrepancy can explain this difference.

In summary, our findings strongly argue against the proposal that asymmetries in the distribution of auxin within young leaf primordia have a strong influence on leaf polarity (Qi et al., 2014) or regulate tissue mechanics (Qi et al., 2017). Might low levels of auxin in adaxial tissues still limit the pattern WOX1 or PRS gene expression (Guan et al., 2017)? In this study, we find high levels of auxin sensing in presumptive provascular cells that also express high levels of PIN1-GFP. Is this where WOX1 and PRS are normally expressed? In fact simultaneous imaging of PIN1-GFP together with PRS-GFP in young leaf primordia indicates that PRS expression is not restricted to cells with high PIN1-GFP expression, although they are both in the middle domain (reconstructed transverse optical section; Figure 4E of (Caggiano et al., 2017)). More significantly, it was shown that if auxin is applied exogenously, WOX1 and PRS do not become expressed ectopically (Caggiano et al., 2017). Overall then, our data argue against a role for auxin in limiting WOX1 and PRS expression to the middle domain. Rather, they support previous results indicating that, like the ventrally expressed KANADI and Auxin Response Factor genes (ARF2, ARF3 and ARF4), the adaxially expressed HD-ZIPIII genes restrict auxin response, thereby limiting WOX1 and PRS expression to the middle domain whether the level of auxin in adaxial cells is high or not (Caggiano et al., 2017). Our results do however leave other observations unexplained. In particular, to understand the apparent ventralisation of leaf primordia in tomato in response to exogenous auxin (Qi et al., 2014) will require further work in assessing how auxin distribution patterns change in response to exogenous treatments and how auxin is distributed during regular development in tomato, preferentially using a ratio-metric auxin sensor such as R2D2.

Source data files used for all figures are available via the BioStudies database (https://www.ebi.ac.uk/biostudies/studies/S-BSST223). All scripts and software for nuclear segmentation and signal quantification are available via the Sainsbury Laboratory gitlab server (https://gitlab.com/slcu/teamHJ/publications/bhatia_et_al_2019).

1. Software

   1. Åhl H
   2. Green M
   3. Krupinski P
   4. Melke P
   5. Sahlin P
   6. Jönsson H

   (2018) Costanza, version 15332d51

   Gitlab.

   https://gitlab.com/slcu/teamHJ/Costanza
2. 1. Bhatia N
   2. Bozorg B
   3. Larsson A
   4. Ohno C
   5. Jönsson H
   6. Heisler MG

   (2016) Auxin acts through monopteros to regulate plant cell polarity and pattern phyllotaxis

   Current Biology 26:3202–3208.

   https://doi.org/10.1016/j.cub.2016.09.044

   • PubMed
   • Google Scholar
3. Software

   1. Bhatia N
   2. Åhl H
   3. Jönsson H
   4. Heisler M

   (2018) Bhatia_et_al_2019, version 4011c8db

   Gitlab.

   https://gitlab.com/slcu/teamHJ/publications/bhatia_et_al_2019
4. 1. Brunoud G
   2. Wells DM
   3. Oliva M
   4. Larrieu A
   5. Mirabet V
   6. Burrow AH
   7. Beeckman T
   8. Kepinski S
   9. Traas J
   10. Bennett MJ
   11. Vernoux T

   (2012) A novel sensor to map auxin response and distribution at high spatio-temporal resolution

   Nature 482:103–106.

   https://doi.org/10.1038/nature10791

   • PubMed
   • Google Scholar
5. 1. Caggiano MP
   2. Yu X
   3. Bhatia N
   4. Larsson A
   5. Ram H
   6. Ohno CK
   7. Sappl P
   8. Meyerowitz EM
   9. Jönsson H
   10. Heisler MG

   (2017) Cell type boundaries organize plant development

   eLife 6:e27421.

   https://doi.org/10.7554/eLife.27421

   • PubMed
   • Google Scholar
6. 1. Guan C
   2. Wu B
   3. Yu T
   4. Wang Q
   5. Krogan NT
   6. Liu X
   7. Jiao Y

   (2017) Spatial Auxin Signaling Controls Leaf Flattening in Arabidopsis

   Current Biology 27:2940–2950.

   https://doi.org/10.1016/j.cub.2017.08.042

   • PubMed
   • Google Scholar
7. Preprint

   1. Guan C
   2. Du F
   3. Jiao Y

   (2018) Uniform distribution of 35S promoter-driven mDII auxin control sensor in leaf primordia

   bioRxiv.

   https://doi.org/10.1101/293753

   • Google Scholar
8. 1. Hagemann W
   2. Gleissberg S

   (1996) Organogenetic capacity of leaves: The significance of marginal blastozones in angiosperms

   Plant Systematics and Evolution 199:121–152.

   https://doi.org/10.1007/BF00984901

   • Google Scholar
9. 1. Husbands AY
   2. Chitwood DH
   3. Plavskin Y
   4. Timmermans MC

   (2009) Signals and prepatterns: new insights into organ polarity in plants

   Genes & Development 23:1986–1997.

   https://doi.org/10.1101/gad.1819909

   • PubMed
   • Google Scholar
10. Software

    1. Jones E
    2. Oliphant T
    3. Peterson P

    (2001) SciPy: Open Source Scientific Tools for Python

    scipy.

    http://www.scipy.org/
11. 1. Kerstetter RA
    2. Bollman K
    3. Taylor RA
    4. Bomblies K
    5. Poethig RS

    (2001) KANADI regulates organ polarity in Arabidopsis

    Nature 411:706–709.

    https://doi.org/10.1038/35079629

    • PubMed
    • Google Scholar
12. 1. Koch AJ
    2. Meinhardt H

    (1994) Biological pattern formation: from basic mechanisms to complex structures

    Reviews of Modern Physics 66:1481–1507.

    https://doi.org/10.1103/RevModPhys.66.1481

    • Google Scholar
13. 1. Liao CY
    2. Smet W
    3. Brunoud G
    4. Yoshida S
    5. Vernoux T
    6. Weijers D

    (2015) Reporters for sensitive and quantitative measurement of auxin response

    Nature Methods 12:207–210.

    https://doi.org/10.1038/nmeth.3279

    • PubMed
    • Google Scholar
14. 1. Lucy LB

    (1974) An iterative technique for the rectification of observed distributions

    The Astronomical Journal 79:745.

    https://doi.org/10.1086/111605

    • Google Scholar
15. 1. Otsu N

    (1979) A Threshold Selection Method from Gray-Level Histograms

    IEEE Transactions on Systems, Man, and Cybernetics 9:62–66.

    https://doi.org/10.1109/TSMC.1979.4310076

    • Google Scholar
16. 1. Qi J
    2. Wang Y
    3. Yu T
    4. Cunha A
    5. Wu B
    6. Vernoux T
    7. Meyerowitz E
    8. Jiao Y

    (2014) Auxin depletion from leaf primordia contributes to organ patterning

    PNAS 111:18769–18774.

    https://doi.org/10.1073/pnas.1421878112

    • PubMed
    • Google Scholar
17. 1. Qi J
    2. Wu B
    3. Feng S
    4. Lu S
    5. Guan C
    6. Zhang X
    7. Qiu D
    8. Hu Y
    9. Zhou Y
    10. Li C
    11. Long M
    12. Jiao Y

    (2017) Mechanical regulation of organ asymmetry in leaves

    Nature Plants 3:724–733.

    https://doi.org/10.1038/s41477-017-0008-6

    • PubMed
    • Google Scholar
18. 1. Richardson WH

    (1972) Bayesian-based iterative method of image restoration*

    Journal of the Optical Society of America 62:55.

    https://doi.org/10.1364/JOSA.62.000055

    • Google Scholar
19. 1. van der Walt S
    2. Schönberger JL
    3. Nunez-Iglesias J
    4. Boulogne F
    5. Warner JD
    6. Yager N
    7. Gouillart E
    8. Yu T
    9. scikit-image contributors

    (2014) scikit-image: image processing in Python

    PeerJ 2:e453.

    https://doi.org/10.7717/peerj.453

    • PubMed
    • Google Scholar
20. 1. Vernoux T
    2. Brunoud G
    3. Farcot E
    4. Morin V
    5. Van den Daele H
    6. Legrand J
    7. Oliva M
    8. Das P
    9. Larrieu A
    10. Wells D
    11. Guédon Y
    12. Armitage L
    13. Picard F
    14. Guyomarc'h S
    15. Cellier C
    16. Parry G
    17. Koumproglou R
    18. Doonan JH
    19. Estelle M
    20. Godin C
    21. Kepinski S
    22. Bennett M
    23. De Veylder L
    24. Traas J

    (2011) The auxin signalling network translates dynamic input into robust patterning at the shoot apex

    Molecular Systems Biology 7:508.

    https://doi.org/10.1038/msb.2011.39

    • PubMed
    • Google Scholar
21. 1. Wang Y
    2. Wang J
    3. Shi B
    4. Yu T
    5. Qi J
    6. Meyerowitz EM
    7. Jiao Y

    (2014) The Stem Cell Niche in Leaf Axils Is Established by Auxin and Cytokinin in Arabidopsis

    The Plant Cell 26:2055–2067.

    https://doi.org/10.1105/tpc.114.123083

    • PubMed
    • Google Scholar
22. Book

    1. Wiener N

    (1949)

    Extrapolation, Interpolation, and Smoothing of Stationary Time Series: With Engineering Applications

    Cambridge, Mass: MIT Press.

    • Google Scholar

Author details

1. Neha Bhatia

   School of Life and Environmental Sciences, University of Sydney, Sydney, Australia

   Contribution

   Conceptualization, Investigation, Visualization, Methodology, Writing—original draft, Project administration, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-2165-5183

2. Henrik Åhl

   1. Sainsbury Laboratory, University of Cambridge, Cambridge, United Kingdom
   2. Department of Applied Mathematics and Theoretical Physics, University of Cambridge, Cambridge, United Kingdom

   Contribution

   Data curation, Software, Formal analysis, Validation, Investigation, Visualization, Methodology

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-0655-806X

3. Henrik Jönsson

   1. Sainsbury Laboratory, University of Cambridge, Cambridge, United Kingdom
   2. Department of Applied Mathematics and Theoretical Physics, University of Cambridge, Cambridge, United Kingdom
   3. Department of Astronomy and Theoretical Physics, Lund University, Lund, Sweden

   Contribution

   Supervision, Funding acquisition, Methodology, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-2340-588X

4. Marcus G Heisler

   School of Life and Environmental Sciences, University of Sydney, Sydney, Australia

   Contribution

   Conceptualization, Supervision, Funding acquisition, Visualization, Methodology, Writing—original draft, Project administration, Writing—review and editing

   For correspondence

   marcus.heisler@sydney.edu.au

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-5644-8398

• 3,704

  views

• 535

  downloads

• 22

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.